Sika Deer: Biology and Management of Native and Introduced Populations

Sika Deer

Dale R. McCullough • Seiki Takatsuki Koichi Kaji Editors

Sika Deer Biology and Management of Native and Introduced Populations

Editors Dale R. McCullough Professor Emeritus Department of Environmental Science, Policy, and Management and Museum of Vertebrate Zoology University of California, Berkeley California 94720-3114 USA

Koichi Kaji Department of Ecoregion Science Laboratory of Wildlife Conservation Tokyo University of Agriculture and Technology 3-5-8 Saiwaicho Fuchu, Tokyo 183-8509 Japan

Seiki Takatsuki Laboratory of Wildlife Ecology and Conservation Azabu University 1-17-71 Fuchinobe Sagamihara, Kanagawa 229-8501 Japan

Front cover: Top: Male sika deer and (bottom) a group of grazing sika deer. Photos by D. R. McCullough. Back cover: Top left: A male during the rut. Top right: A female grooming her fawn in late winter. Bottom: A female and her fawn in early summer. Photos by S. Takatsuki. All photographs are from Kinkazan Island, northern Japan.

ISBN: 978-4-431-09428-9 Springer Tokyo Berlin Heidelberg New York e-ISBN: 978-4-431-09429-6 DOI: 10.1007/978-4-431-09429-6 Library of Congress Control Number: 2008934594 © Springer 2009 Printed in Japan This work is subject to copyright. All rights are reserved, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in other ways, and storage in data banks. The use of registered names, trademarks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. Printed on acid-free paper Springer is a part of Springer Science+Business Media springer.com

Preface

This book reviews in detail the sika deer of eastern Asia to bring together under one cover a compilation of the growing literature on this fascinating animal of such great ecological and economic importance. The sika deer is important not only in its native range of east Asia, but also in the many parts of the world where it has been introduced, and become a naturalized member of the fauna, for better or worse. The literature on sika deer is widely scattered and in many different languages, including Japanese, Russian, Chinese, and Vietnamese. This literature is difficult for most readers to access and, for practical purposes, the information does not exist, despite the fact that some of the best research on deer in the world is now being done by Japanese scientists and scattered researchers in other parts of the sika deer’s Asiatic range are accumulating information on local populations. Consequently, sika deer remain rather poorly known, even among experienced deer biologists, despite their importance in the affairs of humans in many countries and cultures and this growing knowledge. Our purpose in producing this book is to compile and integrate the immense amount of knowledge of sika deer in one source. It is directed primarily towards biologists, conservationists, and hunters. However, most interested lay readers may profitably gain from a reading. Although the subject matter is science, the presentation is made at a level that any reasonably well-read person can access the essential information. This book emphasizes sika deer as a wild species and/or conservation issue, and lesser attention has been given to the sika as a domestic species despite large numbers being raised in captivity in many countries. It is organized into six parts. Part I introduces the basic biology of the species, its origins and evolution, genetic structure, physiology, nutrition, and reproduction. The next three parts discuss food and habitat relations (Part II), behavior, migration, and breeding systems (Part III), and population dynamics and management (Part IV). Most work on sika deer has been done in Japan, and it is appropriate that this area predominates in the coverage—a simple numerical decision. Part V covers sika deer in the remainder of their native range on the Asian continent in China, North Korea, Vietnam, and Far East Russia, and on the island of Taiwan. The last part (Part VI) covers the many introduced and naturalized populations of sika deer around the world. v

vi

Preface

In seeking authors for the various chapters we have attempted to gather together the most recognized experts on each topic. We have asked independent workers to collaborate in new ways to optimize the expertise and coverage to achieve an integration of all that is known about the sika deer in one book. The organization of the book by biology and distribution themes inevitably results in some overlap of subject matter between parts. For example, the topic of reproduction reoccurs in nearly every part of the book. This should remind us that placing subjects into categories violates the reality that sika deer are whole entities, with completely integrated physical and biological systems connected with real climates, topography, vegetation, competitors, and predators. So, in many ways this book reads more like a symposium volume than an easy overview of the species. To assist the reader through the material we have begun each chapter with an abstract, a brief synopsis to help put the chapter in perspective. At the same time we have tried to take advantage of the multiple overlapping studies to use a comparative approach to advantage. For example, there are many studies now available of sika deer populations on islands or isolated populations in various parts of the range, and we have included many of these to extract the lessons to be learned from comparative work. We have erred on the side of completeness, so that this will be the reference work that serves to give thorough insight into the species, rather than a summary work predigested for the casual reader. Nonetheless, the more general reader can pick and choose to achieve the latter goal. We ask the reader’s indulgence and trust by the end of the book the desired integration, still retaining important nuance, will have been achieved. Dale R. McCullough

Acknowledgements

I, along with my co-editors, Seiki Takatsuki and Koichi Kaji, wish to thank the following people for their indispensable contributions to the completion of the book. First, we thank all of the authors for their giving of their expertise and time to write chapters and appreciate their patience with the process of give and take in shaping the needs of each chapter to the larger goal of an integrated whole. My long-time administrative assistant, Margaret Jaeger, who has suffered through a number of previous book efforts with me, was once again her stalwart self, clarifying meaning, checking details, catching errors, keeping track of things, and generally supporting the effort. Without her assistance and guidance this project could not have been completed. Thanks once again, Margaret, for a job well done. Mayuko Tanigawa helped with early coordination between Japanese authors and we three editors and Margaret. Karen Klitz, illustrator for the Museum of Vertebrate Zoology at Berkeley, checked the illustrations and made changes when necessary for legibility and style. She also drew some new map figures to complete the work. Aiko Hiraguchi, Executive Editor for Life Sciences of Springer Japan was most helpful in bringing the book to publication. Her patience, understanding, and continued support of the project were much appreciated. I thank my co-editors, Seiki Takatsuki and Koichi Kaji, for their indispensable assistance in identifying appropriate Japanese authors, organizing book planning meetings, encouraging authors to write, and helping me negotiate the cultural gaps in ways of doing things between cultures. Their dedication and expertise, and long experience with research on sika deer, were critical to the conception and production of this book. For reading and commenting on all or parts the draft book manuscript we thank George Bubenik, Paul Krausman, and John Kie. We are indebted for financial support. The A. Starker Leopold Endowed Chair at the University of California, Berkeley, of which Dale R. McCullough was the chair-holder, was the main support for preparation of the book. We also thank the Ninth International Mammal Congress planning committee for making additional funds available to help complete the book. To all of the above we give our heartfelt appreciation and we hope that the outcome justifies their contributions. vii

Contents

1

Introduction .............................................................................................. Dale R. McCullough

1

Part I Basic Biology 2

Fossil Record of Sika Deer in Japan ...................................................... Yoshinari Kawamura

3

Two Genetically Distinct Lineages of the Japanese Sika Deer Based on Mitochondrial Control Regions ............................................. Junco Nagata

27

Evolutionary Significance of Admixture and Fragmentation of Sika Deer Populations in Japan ......................................................... Hidetoshi B. Tamate

43

Nutritional Physiology of Wild and Domesticated Japanese Sika Deer .................................................................................. Takayoshi Masuko and Kousaku Souma

61

4

5

6

Endocrinology of Sika Deer .................................................................... Kiyoshi Yamauchi and Yukiko Matsuura

7

Reproduction of Female Sika Deer in Japan, with Special Reference to Kinkazan Island, Northern Japan................................... Nobumasa Ohnishi, Masato Minami, Rie Nishiya, Kotoyo Yamada, Hiroyuki Nishizuka, Hiroshi Higuchi, Azusa Nara, Masatsugu Suzuki, and Seiki Takatsuki

11

83

101

ix

x

Contents

Part II Food and Habitat Relations 8 Food Habits of Sika Deer on Kinkazan Island, Northern Japan with Reference to Local Variations, Size Effects, and Comparison with the Main Island ................................................ Seiki Takatsuki and U. K. G. K. Padmalal

113

9 Plants and Plant Communities on Kinkazan Island, Northern Japan, in Relation to Sika Deer Herbivory ........................ Seiki Takatsuki and Takehiko Y. Ito

125

10 Productivity and Foraging Efficiency of the Short-Grass (Zoysia japonica) Community for Sika Deer ....................................... Takehiko Y. Ito, Mariko Shimoda, and Seiki Takatsuki

145

11 Home Range, Habitat Selection, and Food Habits of the Sika Deer Using the Short-Grass Community on Kinkazan Island, Northern Japan ...................................................................................... Takehiko Y. Ito and Seiki Takatsuki 12 Shift to Litterfall as Year-Round Forage for Sika Deer after a Population Crash ................................................................................ Masami Miyaki and Koichi Kaji 13

The Dynamics of Forest Stands Affected by Sika Deer on Nakanoshima Island—Change of Size Structure Similar to the Thinning Effect............................................................................ Masami Miyaki and Koichi Kaji

159

171

181

14 Biology of Sika Deer in Hyogo: Characteristics of Reproduction, Food Habits, Growth, and Condition ......................... Mayumi Yokoyama

193

15 Bark-Stripping Preference of Sika Deer and Its Seasonality on Mt. Ohdaigahara, Central Japan .............................. Masaki Ando and Ei’ichi Shibata

207

16 North-South Variations in Sika Deer Ecology as a Forest-Dwelling Cervid ................................................................. Seiki Takatsuki

217

17 Geographical Variations in Food Habits of Sika Deer: The Northern Grazer vs. the Southern Browser ................................ Seiki Takatsuki

231

Contents

xi

18 What Is “Natural” Vegetation? A Reconsideration of Herbivory by Wild Ungulates .......................................................... Seiki Takatsuki

239

Part III Behavior: Migration and Breeding Systems 19 Seasonal Migration of Sika Deer on Hokkaido Island, Japan .......... Hiromasa Igota, Mayumi Sakuragi, and Hiroyuki Uno

251

20 Migratory and Sedentary Behavior Patterns of Sika Deer in Honshu and Kyushu, Japan ............................................................ Tsuneaki Yabe and Seiki Takatsuki

273

21 Variation in Mating Behavior of Sika Deer: Mating Behavior of Sika Deer on Nozaki Island .............................................................. Akira Endo

285

22 Reproductive Ecology of Sika Deer on Kinkazan Island, Northern Japan: Reproductive Success of Males and Multi-Mating of Females ............................................................... Masato Minami, Nobumasa Ohnishi, Ayumi Okada, and Seiki Takatsuki

297

23 Life-Time Reproductive Success of Female Sika Deer on Kinkazan Island, Northern Japan .................................................. Masato Minami, Nobumasa Ohnishi, Naoko Higuchi, Ayumi Okada, and Seiki Takatsuki 24

Reproduction of Female Sika Deer in Western Japan ....................... Toru Koizumi, Shin-ichiro Hamasaki, Mayumi Kishimoto, Mayumi Yokoyama, Masato Kobayashi, and Aiko Yasutake

319

327

Color Plates Part IV

Population Dynamics and Management

25 Sika Deer in Nara Park: Unique Human-Wildlife Relations ............ Harumi Torii and Shirow Tatsuzawa 26 A 20-Year History of Sika Deer Management in the Mt. Goyo Area, Northern Honshu............................................. Seiki Takatsuki

347

365

xii

Contents

27 Survival Patterns of Male and Female Sika Deer on Kinkazan Island, Northern Japan .................................................. Masato Minami, Nobumasa Ohnishi, and Seiki Takatsuki

375

28 Sika Deer in an Evergreen Broad-Leaved Forest Zone on the Boso Peninsula, Japan ............................................................... Masahiko Asada and Keiji Ochiai

385

29 Sika Deer Population Irruptions and Their Management on Hokkaido Island, Japan ................................................................... Hiroyuki Uno, Koichi Kaji, and Katsumi Tamada

405

30 Irruptive Behavior of Sika Deer ........................................................... Koichi Kaji, Hiroshi Takahashi, Hideaki Okada, Masao Kohira, and Masami Yamanaka

421

31 The Management of Sika Deer Populations in Hyogo Prefecture, Japan ................................................................... Hiroshi Sakata, Shin-ichiro Hamasaki, and Hiromune Mitsuhashi

437

32 Management Strategy of Sika Deer Based on Sensitivity Analysis ........................................................................... Shingo Miura and Kunihiko Tokida

453

Part V

Sika in Mainland Asia and Taiwan

33 Sika Deer in Russia ................................................................................ Vladimir V. Aramilev 34 Sika Deer Distribution Changes at the Northern Extent of Their Range in the Sikhote-Alin Mountains of the Russian Far East ................................................................................................... Inna V. Voloshina and Alexander I. Myslenkov

475

501

35 Sika Deer in Mainland China ............................................................... Dale R. McCullough, Zhi-Gang Jiang, and Chun-Wang Li

521

36 Sika Deer in Korea and Vietnam ......................................................... Dale R. McCullough

541

37 Sika Deer in Taiwan............................................................................... Dale R. McCullough

549

38

The Present Status of the Re-introduced Sika Deer in Kenting National Park, Southern Taiwan ....................................... Kurtis Jai-Chyi Pei

561

Contents

xiii

Part VI Introduced Sika Deer 39 Sika Deer in Continental Europe ......................................................... Luděk Bartoš

573

40 Sika Deer in the British Isles ................................................................ Graeme M. Swanson and Rory Putman

595

41 Free–Ranging and Confined Sika Deer in North America: Current Status, Biology, and Management ......................................... George A. Feldhamer and Stephen Demarais

615

42 The Sika in New Zealand ...................................................................... D. Bruce Banwell

643

Index ...............................................................................................................

657

Contributors

Ando, Masaki Research Associate, Department of Ecology and Environmental Science, Faculty of Applied Biological Science, Gifu University, 1-1 Yanagido, Gifu 501-1193, Japan Aramilev, Vladimir V. Scientist, Laboratory of Nature Resources Use, Pacific Institute of Geography, Far-Eastern Branch, Russian Academy of Science, Radio Street 7, 690041 Vladivostok, Russia Asada, Masahiko Senior researcher, Natural History Museum and Institute, 955-2 Aoba-cho, Chuo-ku, Chiba 260-8682, Japan Banwell, D. Bruce Author about deer in New Zealand, Na Fiadh, 39 Lane Street, Allenton, Ashburton, Canterbury 7700, New Zealand Bartoš, Luděk Professor of Ethology, Department of Ethology, Institute of Animal Science, Přátelství 815, Praha Uhříněves, Czech Republic Demarais, Stephen Professor of Wildlife Management, Department of Wildlife and Fisheries, Mississippi State University, Mississippi State, Mississippi 39762, USA Endo, Akira Researcher, Department of Applied Biological Sciences, Faculty of Agriculture, Saga University, 1 Hongo, Saga 840-8502, Japan Feldhamer, George A. Professor of Zoology, and Director of the Environmental Studies Program, Department of Zoology MC 6501, Southern Illinois University, Carbondale, Illinois 62901, USA Hamasaki, Shin-ichiro Chief of Kansai Branch, Wildlife Management Office, 4-10-6 Fujiwaradaiminamimachi, Kita-ku, Kobe 651-1303, Japan xv

xvi

Contributors

Higuchi, Hiroshi Kinkazan Deer Research Group, c/o Picchio Wildlife Community Institute, 1549-3-1 Oiwake, Karuizawa, Nagano 389-0115, Japan Higuchi, Naoko Graduate Student, Laboratory of Functional Animal Ecology, Department of Biology and Geosciences, Graduate School of Science, Osaka City University, 3-3-138 Sugimoto, Sumiyoshi-ku, Osaka 558-8585, Japan Igota, Hiromasa Assistant Professor, Department of Biosphere and Environmental Sciences, Faculty of Environmental Systems, Rakuno Gakuen University, 582 Bunkyodai-midorimachi, Ebetsu, Hokkaido 069-8501, Japan Ito, Takehiko Y. Assistant Professor, Arid Land Research Center, Tottori University, 1390 Hamasaka, Tottori 680-0001, Japan Jiang, Zhi-Gang Key Laboratory of Animal Ecology and Conservation Biology, Institute of Zoology, Chinese Academy of Sciences, Beijing, China Kaji, Koichi Professor, Department of Ecoregion Science, Laboratory of Wildlife Conservation, Tokyo University of Agriculture and Technology, 3-5-8 Saiwaicho, Fuchu, Tokyo 183-8509, Japan Kawamura, Yoshinari Professor of Paleontology and Stratigraphy, Department of Earth Sciences, Aichi University of Education, Kariya, Aichi 448-8542, Japan Kishimoto, Mayumi Researcher of Kansai Branch, Wildlife Management Office, 4-10-6 Fujiwaradaiminamimachi, Kita-ku, Kobe 651-1303, Japan Kobayashi, Masato Student at Laboratory of Forest Ecology, Faculty of Agriculture, Kyoto University, Kitashirakawaoiwake-cho, Sakyo-ku, Kyoto 606-8502, Japan Kohira, Masao Head of Conservation and Management Section and Senior Researcher, Shiretoko Nature Foundation, 531 Iwaobetsu, Shari-chou, Hokkaido 099-4356, Japan Koizumi, Toru Director of Department of Wildlife Biology, Department of Wildlife Biology, Forest and Forest Products Research Institute, 1 Matsunosato, Tsukuba, Ibaraki 305-8687, Japan

Contributors

xvii

Li, Chun-Wang Associate Professor, Key Laboratory of Animal Ecology and Conservation Biology, Institute of Zoology, Chinese Academy of Sciences, Beijing 100101, China Masuko, Takayoshi Professor, Faculty of Bioindustry, Tokyo University of Agriculture, Abashiri, Hokkaido 099-2493, Japan Matsuura, Yukiko Field Science Center for Northern Biosphere, Hokkaido University, Kita 9, Nishi 9, Kita-ku, Sapporo, Hokkaido 060-0809, Japan McCullough, Dale R. Professor Emeritus, Department of Environmental Science, Policy, and Management, and Museum of Vertebrate Zoology, University of California, Berkeley, California 94720-3114, USA Minami, Masato Director, Wildlife Community Institute, 1549-3-1 Oiwake, Karuizawa, Nagano 389-0115, Japan Mitsuhashi, Hiromune Assistant Professor, Institute of Natural and Environmental Sciences, University of Hyogo, 6 Yayoigaoka, Sanda, Hyogo 669-1546, Japan Miura, Shingo Professor, School of Human Sciences, Waseda University, 2-579 Migashima, Tokorozawa, Saitama 359-1192, Japan Miyaki, Masami Director of Conservation Department, Hokkaido Institute of Environmental Sciences, Kita 19, Nishi 12, Kita-ku, Sapporo, Hokkaido 060-0819, Japan Myslenkov, Alexander I. Department of Science, Lazovsky State Nature Reserve, Centralnaya, 56, Lazo, Primorsky Krai 692980, Russia Nagata, Junco Forestry and Forest Products Research Institute, Tsukuba, Ibaraki 305-8687, Japan Nara, Azusa Kinkazan Deer Research Group, Wildlife Community Institute, 1549-3-1 Oiwake, Karuizawa, Nagano 389-0111, Japan Nishiya, Rie Wild Bird Observation Center of Mt. Zao, 162-1 Uenohara, Zao-machi, Miyagi 989-0800, Japan

xviii

Contributors

Nishizuka, Hiroyuki Kinkazan Deer Research Group, Wildlife Community Institute, 1549-3-1 Oiwake, Karuizawa, Nagano 389-0111, Japan Ochiai, Keiji Senior researcher, Natural History Museum and Institute, Chiba, 955-2 Aoba-cho, Chuo-ku, Chiba 260-8682, Japan Ohnishi, Nobumasa Senior Researcher, Eco-planning Research Co. Ltd., 2-28-5 Sakaemachi, Higashimurayama, Tokyo 189-0013, Japan Okada, Ayumi Lecturer, Department of Environmental Bioscience, School of Veterinary Medicine, Kitasato University, 23-35-1 Higashi, Towada, Aomori 034-8628, Japan Okada, Hideaki Deputy Director and Senior Researcher, Shiretoko Nature Foundation, 531 Iwaobetsu, Shari-chou, Hokkaido 099-4356, Japan Padmalal, U.K.G.K. Head, Post Graduate Studies, Environmental Sciences Open University of Sri Lanka, Nawala, Negegoda, Sri Lanka Pei, Kurtis Jai-Chyi Professor of Wildlife Ecology and Management, Institute of Wildlife Conservation, National Pingtung University of Science and Technology, Neipu, Pingtung 91201, Taiwan Putman, Rory Professor of Behavioural and Environmental Biology (Emeritus at the Manchester Metropolitan University), Keil House, Ardgour, by Fort William, Inverness-shire PH33 7AH, Scotland Sakata, Hiroshi Associate Professor, Institute of Natural and Environmental Sciences, University of Hyogo, 940 Sawano, Aogaki, Tanba, Hyogo 669-3842, Japan Sakuragi, Mayumi Laboratory of Boreal Forest Conservation, Field Science Center for Northern Biosphere, Hokkaido University, Kita 9, Nishi 9, Kita-ku, Sapporo, Hokkaido 060-8589, Japan Shimoda, Mariko Graduate School of Agriculture and Life Sciences, The University of Tokyo, 1-1-1 Yayoi, Bunkyo-ku, Tokyo 113-8657, Japan Shibata, Ei’ichi Professor of Forestry, Graduate School of Bioagricultural Sciences, Nagoya University, Chikusa-ku, Nagoya 464-8601, Japan

Contributors

xix

Souma, Kousaku Assistant Professor, Faculty of Bioindustry, Tokyo University of Agriculture, Abashiri, Hokkaido 099-2493, Japan Suzuki, Masatsugu Professor, Laboratory of Zoo and Wildlife Medicine, Faculty of Applied Biological Sciences, Gifu University, 1-1 Yanagido, Gifu 501-1193, Japan Swanson, Graeme Christ’s College, Rolleston Avenue, Christchurch, New Zealand Tamada, Katsumi Researcher, Nature Conservation Department, Hokkaido Institute of Environmental Sciences, Kita 19, Nishi 12, Kita-ku, Sapporo, Hokkaido 060-0819, Japan Tamate, Hidetoshi B. Department of Biology, Faculty of Science, Yamagata University, 1-4-12 Kojirakawa, Yamagata 990-8560, Japan Takahashi, Hiroshi Kansai Research Center, Forestry and Forest Products Research Institute, Fushimi-ku, Kyoto 612-0855, Japan Takatsuki, Seiki Professor, Laboratory of Wildlife Ecology and Conservation, Azabu University, 1-17-71 Fuchinobe, Sagamihara, Kanagawa 229-8501, Japan Tatsuzawa, Shirow Assistant Professor, Research Group of Regional Sciences, Graduate School of Letters, Hokkaido University, Kita 10, Nishi 7, Kita-ku, Sapporo, Hokkaido 060-0810, Japan Tokida, Kunihiko Senior Scientist, Japan Wildlife Research Center, 3-10-10 Sitaya, Taitou-ku, Tokyo 110-8676, Japan Torii, Harumi Associate Professor, Education Center for Natural Environment, Nara University of Education, Takabatake, Nara 630-8528, Japan Uno, Hiroyuki Nature Conservation Department, Hokkaido Institute of Environmental Sciences, Kita 19, Nishi 12, Kita-ku, Sapporo, Hokkaido 060-0819, Japan Voloshina, Inna V. Department of Science, Lazovsky State Nature Reserve, Centralnaya, 56 Lazo, Primorsky Krai 692980, Russia

xx

Contributors

Yabe, Tsuneaki Kyushu Research Center, Forestry and Forest Products Research Institute, 4-11-16 Kurokami, Kumamoto 860-0862, Japan Yamada, Kotoyo Kinkazan Deer Research Group, c/o Picchio Wildlife Research Center Co. Ltd. 2148 Nagakura, Karuizawa, Nagano 389-0111, Japan Yamanaka, Masami Director and Chief Researcher, Shiretoko Nature Foundation, 531 Iwaobetsu, Shari-chou, Hokkaido 099-4356, Japan Yamauchi, Kiyoshi Technical Researcher, Natural Environment Laboratory, Department of Earth Science, Research Institute for Environmental Science and Public Health of Iwate Prefecture (I-RIEP), 1-36-1 Iiokashinden, Morioka, Iwate 020-0852, Japan Yasutake, Aiko Researcher of Kansai Branch, Wildlife Management Office, 4-10-6 Fujiwaradaiminamimachi, Kita-ku, Kobe 651-1303, Japan Yokoyama, Mayumi Institute of Natural and Environmental Sciences, Wildlife Management Research Center, University of Hyogo, 940 Sawano, Aogaki, Tanba, Hyogo 669-3842, Japan

Chapter 1

Introduction Dale R. McCullough

Abstract The sika deer (Cervus nippon) is an important member of the native fauna in eastern Asia, and it has been widely introduced into many other parts of the world. It has an interesting paleogeographic history, having reached the Japanese Islands and Taiwan during low sea level periods. It has a long history of close association with humans, both positive and negative, given that its preferred habitat is also prime agricultural and developmental land for humans. In this respect, sika deer are similar to secondary-successional deer species in North America and Europe. However, sika deer have the ability to cause damage to crops and forests, as well as to their own habitats, that far exceed those of comparable deer in other parts of the world. In different parts of their range sika deer are over-abundant, or extinct in the wild. They are commonly raised in captivity for their antlers. Consequently, they present an unusually wide array of evolutionary, ecological, and management issues. We begin with the question, what is the sika deer? Peruse any book of Japanese or Chinese art, and within a few pages you will encounter a landscape painting with a graceful spotted deer surveying its world from beneath the pendant leaves of a luxurious tree. This deer, the sika, is an icon of nature in the far eastern rim of the Asian continent—from Vietnam in the south, through China and Korea to Russia in the north, and on the continental islands of Taiwan and the Japanese Archipelago— where it holds almost a religious status. Indeed, it is the sacred deer of Japan where it occupies the sanctuary grounds surrounding many Shinto shrines and temples. An ancient legend says that a god rode into Nara Temple near Kyoto (probably the most famous temple in Japan, which has numerous sika deer yet today) on the back of a sika deer. The English-speaking world knows this beautiful deer as the “sika deer.” “Sika” is the Japanese word for deer (although pronounced “she-ka”), so this linguistic hybrid results in a redundant deer-deer. Nevertheless, “sika deer” is so ingrained in the western literature that it is rather pointless to urge adoption of the simpler, and more correct, “sika.” The deer’s more evocative and descriptive Chinese name, my hwa lu, means “white-flower deer.” This name refers to the white spots arrayed over the body that are particularly apparent in the summer coat.

D. R. McCullough et al. (eds.), Sika Deer: Biology and Management of Native and Introduced Populations, DOI: 10.1007/978-4-431-09429-6_1, © Springer 2009

1

2

D. R. McCullough

The elegant pattern of white spots against a rusty red background—which seems to be the work of an artist—evolved as concealment in forests where the white spots mimic the splotches of light on the forest floor produced by sunlight filtered through the canopy leaves of a broad-leaved forest. It can be surprisingly cryptic in this environment despite being so vivid in open land habitats. This pattern of light is characteristic of deciduous broad-leaved forests. Although the sika deer occupies some coniferous forests, particularly at the higher elevations and northern edges of its distribution, this animal is essential a denizen of the broad-leaved forest regions of southeastern Asia. This spotted pattern is found in young fawns of many members of the deer family in which the vulnerable newly born avoid detection by predators by lying quietly concealed in the vegetation of the forest floor. In the wintertime, when the leaves of broad-leaved trees are cast, the patterning of light in the forest is less spotty and, predictably, the spots of the sika deer in winter coat are either lost (in the north where conifers or deciduous broad-leaved trees predominate), or become diffuse (in the south where semitropical broad-leaved trees do not cast their leaves in autumn). The retention of juvenile spots by adults in the sika deer contributes to the esthetics of the species—a correlate of the Bambi syndrome. Also important is the slender body form and graceful movements, again adaptations to traversing a forested environment with many tree trunks, woody shrubs, downed logs, and other obstructions of a natural forest. While the sika deer is a beautiful, and seemingly fragile, part of a munificent and peaceful nature, when saturated by male hormones during the fall rut the males are transformed into super-aggressive fighting machines, literally slavering from the mouth. The male deer of all species undergo these changes during rut, but none to the extreme that the sika deer does. Most people think of the sika deer as a deep forest animal, and it is true that this deer is dependent on forests for escape cover. Still, this is a misrepresentation of the preferred habitat of sika. An equally important need is areas where the mature forest has been disturbed. Sika deer are an “edge” species that meets life requirements from different kinds or stages of habitat. Although sika deer are dependent on forest cover to evade enemies, and sometimes climatic extremes, they are equally dependent on disturbed areas where, because sunlight reaches the ground level and plant competition is reduced, the nutritional quality of the forage is appreciably higher. Like with the white-tailed deer (Odocoileus virginianus) in North America, the roe deer in Europe (Capreolus capreolus) and Asia (C. pygargus), some of the muntjacs (Muntiacus), and a few other deer species, the sika deer thrives in forests with a mix of different habitats—either separate vegetation types or different stages of succession within a vegetation type. Their evolutionary history was shaped through opportunities presented by the mosaic in the forest created by natural disasters—fires, wind storms, floods, etc.—which were the original source of forest disturbance and renewal. In such landscape mosaics the sika deer can shift about to satisfy different requirements for survival. Edge species of deer, in response to natural predators, often spend the daylight hours secreted in the forests, and venture out at night to feed in the disturbed areas where during daylight they would be exposed to detection.

1

Introduction

3

Since the ascendance of modern humans with their activities of burning, cultivating, and forest cutting, human impacts have supplemented, if not largely supplanted, the natural processes of disturbance. Humans, too, have benefited from the mosaic of vegetation across the landscape that furnish their separate requirements for food, fuel, and shelter. It is neither accidental, nor surprising, that deer preferring disturbed environments with lots of edges between different kinds of vegetation capitalized on the alterations of humans and became closely associated with them. For the deer, the disturbed areas, particularly enhanced by the presence of human-grown crops, proved highly attractive feeding areas yielding superior nutrition. Sika deer and humans have lived in close proximity in Asia for much longer than edge-occupying deer in other parts of the world. Throughout their range in Asia sika deer and humans have had almost identical habitat preferences, and the best habitats for sika are the best environments for agricultural development. Indeed, agricultural development has come at the expense of, or benefit to sika deer, depending upon the technology of humans to exploit or control the deer. It is notable that in Japan, the wolf (Canis lupus) was considered a sacred animal for most of its history because it protected the crop fields of early humans from the ravages of sika deer (Knight 2003; Walker 2005). Only later—when humans invented firearms and gained the ability to directly control sika deer, rabies was accidentally introduced to Japan (with consequent rabid-wolf attacks on people), and wolves attacked the stock of the fledgling livestock industry—that wolves went from being sacred animals to being big, bad wolves. Thus, for the last century, sika deer and humans have been on a collision course, and this nexus of human civilization and sika deer population density persists because of their need for the same environments, often for the same reasons. Usually it has worked to the detriment of sika deer, and they have been severely reduced or driven to extinction in the wild over much of the non-Russian part of the mainland distribution, and in Taiwan. But even where extirpation has occurred, the tie of sika deer to humans continues; they are commonly raised in captivity, in China for several thousand years, to supply the demand for “panty” (velvet antler) for the Asian medicine market. Perhaps this long history of cohabitation at least partially accounts for the sika deer’s extraordinary capacity to negatively impact its own habitats. Most edge-species of deer show tendencies towards boom-bust population trajectories—depending on the amount and frequency of forest disturbance—with consequent time lags between numbers and food resources (“irruptive” population behavior; see review of McCullough (1997) ). At highs in these population fluctuations it is common that excessive consumption by deer has unsustainable impacts on the vegetation. No other deer, however, can match the sika in its ability to strip the vegetation bare and expose soils to massive erosion—thereby creating a wasteland. Many other species suppress young woody plants (and, consequently, subsequent forest tree composition), and a few strip bark from small, fast-growing trees such as aspen. Only sika deer peel the bark off large, fully mature forest trees, sometimes killing by girdling trees with a diameter of over a meter in size. Many deer follow the sound of

4

D. R. McCullough

chainsaws in winter to feed on the twigs and small stems of felled trees. Only the voracious sika deer peels the bark off of the trunks as well. Most such foods are considered starvation rations, and they are eaten only in the winter when alternative foods are absent. But once again, sika deer stretch the envelope. They sometimes strip bark off of the trunks of mature trees in the summertime when alternative foods of greatest quality are most available. Why? From this feeding behavior and consequent destructive capacity it is apparent that sika deer were “managing” forests long before human beings arrived upon the scene. When mature forests dominated the landscape, and naturally disturbed areas were small patches in the forest matrix, sika deer probably were at low numbers and influenced forest vegetation mainly by delaying the succession of openings through their feeding effects on the vegetation. The earlier “balance-of-nature” views of ecology proposed that predators regulated sika deer and other herbivores within rather narrow limits and maintained balances between plants, herbivores, and carnivores. Such simplistic explanations for nature have been largely dispelled by research experience with ecosystem processes, which are not nearly so neat and precise. Natural systems are much messier, replete with time lags, and don’t conform much to balance-of-nature paradigms. Indeed, the very idea of such a precise balance and order in nature is more a belief of a quasi-religious nature. Even the proposition that such a balance is apparent over only larger landscape scales and time scales is being dispelled by a rapidly growing knowledge of paleobiology—the study of deep time horizons. Such studies are overwhelmingly showing that the only constant in nature is change. Such neat balances between ecosystem components and integrity of natural communities are products of our limited lifetimes, which allow our imaginations to impose on nature an orderliness by how we perceive it, think about it, map it, and catalog it. Although we probably underestimate the degree and kind of disturbance that went on in natural forests, there is no doubt that in the modern world humans have altered forests over a shorter time and to a greater extent than is typical in the paleobiology record. No longer is the “undisturbed” (mature forest) forest the matrix and the disturbed areas the isolated patches; now, the mature forest occurs only as isolated patches in the matrix of disturbed area. In other words, the matrix and patches have been transposed (McCullough 1996). Having been selected strongly for exploiting naturally disturbed patches in mature vegetation, there was little counterselection on sika deer for not overdoing it when humans altered landscapes on a massive scale. Rather than being distributed in relatively rare patches, ideal sika deer habitat is the matrix, and the large sika populations favored (where they are not overexploited by hunting) are probably having effects on the remaining mature forest patches of a magnitude that was rare in the pre-civilization human world. Over most of human history local tribes influenced the abundance of deer by hunting. These hunters were severely limited in their effect on deer populations by their stone-age technology. For the humans in such early landscapes, losing some crops to deer could be balanced by hunting and eating the deer. The cost-benefit ratios on both sides were favorable so long as a reasonable balance—a sort of sharing of the joy and pain—was struck. When things got out of whack—too little or too

1

Introduction

5

much hunting success—either deer or humans suffered disproportionately; but there were consequences for both. Only with the development of modern technology—guns, traps, etc.—has it become routinely possible for humans to overexploit sika deer populations, and in some areas (Vietnam, Taiwan) drive them to extinction in the wild, despite the abundance of ideal habitat created by human disturbances. Just as vegetation mosaics are important to sika deer on a fine spatial and time scale, they also have been important on a broad spatial and time scale. Geographically sika deer inhabit the eastern fringes of the Asian continent, and their evolution has been greatly influenced by geological and climatic processes. On a north-south axis, sika deer occupy a band of land on the eastern edge of the Asian continent. They are found mainly on low-elevation plains and surrounding hills and lower reaches of mountains. Their habitats vary from the tropical and semitropical jungles of Vietnam to broad-leaved deciduous forests of Far East Russia and northern Japan. Climates vary from tropical in the south to cold and snowy winters in the north. Their greatest extension westward to the inland of the continent occurred in China in the Sichuan Basin, which although at much higher elevation, has habitat characteristics similar to more coastal areas. Recent studies of sika deer mitochrondrial DNA (Goodman et al. 2001) show that the species is split broadly into a northern and a southern type, and these differences have an evolutionary and genetic basis. In the south they are relatively small in body size and have more similar summer and winter coats. They are nonmigratory, non-winter adapted, and feed predominantly on evergreen broad-leaved plants (Takatsuki 1991). In the northern parts of their range, conversely, climate, especially snow depth, sets the limit to their northern distribution. They have large body sizes, and the summer and winter coats are substantially different. They use mountainous areas, migrate from low-elevation winter areas to high-elevation summer areas, and are adapted to winter conditions. In winter they feed mainly on grasses (Takatsuki 1991). During the Pleistocene repeated climatic fluctuations shifted the latitudinal distribution northwards and southwards, accordion-like, as cold climates pushed southward and then retreated northward. On an east-west axis sea level changes associated with climate fluctuations similarly shifted the distributional range of sika deer and fractionated the populations. At low sea level periods—when water was stored in large polar ice masses—the sika range extended eastward on the exposed ocean floor to the edge of the continental shelf, beyond which the earth fell off precipitously to great ocean depths. The current islands of Taiwan and the Japanese Archipelago are mountain ranges uplifted by subduction of the Pacific and Philippine plates under the Asian plate. They protruded out of the vast plain marking the eastern edge of the continental shelf. When the glaciers melted and released the stored water, rising sea levels isolated these mountains as islands, and sika deer were trapped. There were two such land bridge periods during the major evolutionary history of the sika deer, one about 150,000 years ago, and a second at the end of the Pleistocene about 12,000 years ago (Cook et al. 1999). Given that fossils of sika deer are known from the islands

6

D. R. McCullough

that trace to the earlier event, the sika deer range was very substantially larger during land bridge periods. Virtually all of the current lands beneath the East China Sea and the Japanese Sea were sika deer habitat at the time. Isolation of populations on islands was a major factor in the evolution of the species as the distributional range morphed north-south and east-west under the force of climate change. These forces that shaped the sika deer over paleontological times raise the question of how we should classify the modern survivors for taxonomic purposes. Although sika deer hybridize with red deer and wapiti (Cervus elaphus), which are their closest relatives, most taxonomists accept that sika constitute a valid species. The more pertinent question is how the species should be divided into subspecies, and on this topic there is, and almost certainly will continue to be, ongoing debate. This is inevitable given that subspecific designations are arbitrary, just as a cake can be cut into many small or fewer large pieces, depending on one’s feeling about appropriate serving sizes. “Lumpers” and “splitters” can debate forever. In this book we treat the problem from the view of evidence—the fossil record, geologic and climate events, genetic structuring, biological variation across the range, etc.— and leave the question of the how best to subdivide for classification to other works. As a practical matter, we use commonly accepted subspecies names for different parts of the range, but our emphasis is on the variation in sika deer by geographic area, climate, habitat selection, and other factors that formed the animal as we see it today. Superimposed on this complex of north-south and east-west gyrations of the sika deer range due to climate and topography is the recent advent of modern, technological human societies. Alteration of habitats and creation of new artificial habitats attractive to sika deer have changed the environmental background to a degree not seen before in history. Given the long association of humans and sika deer these changes, inevitably, have had a major impact on the sika populations—both good and bad. Thus, conservation intersects with ecology and evolution. In the southern end of the range the surviving populations are managed as threatened and endangered. Some populations have gone to extinction in the wild. On Taiwan, free-roaming populations have been reestablished by release of captive animals to the wild. In Vietnam no sika occur in the wild anymore, although they are commonly held in captivity and, hopefully, when the economic conditions allow, wild populations can be reestablished. In China, a once extensive range of the sika deer extended eastward to the Sichuan Basin. Now sika have been heavily exploited, and their range has contracted to a few isolated pockets where their future continues to waver under human pressure. In Primorsky Krai in extreme southeastern Russia sika deer still occur in substantial numbers while on many Japanese islands they are extremely overabundant. Raising crops or regenerating forests is impossible without fencing to exclude the deer. Deer-car collisions there are an increasingly serious hazard on the roadways. Sika deer are invading Japanese cities just as white-tailed deer did several decades earlier in the eastern United States, and black-tailed deer (O. hemionus columbianus) have in recent decades along the Pacific Coast. Management to control overpopulations is a key issue in Japan. As in other parts of the world the large carnivores that

1

Introduction

7

originally served to partially control deer numbers were extirpated (wolves) or reduced (brown and Asiatic black bears). Hunting is widely practiced, but as a sika deer control measure it is made difficult by a small and declining hunter population (due to age; there is almost no recruitment of young hunters). Hunters in other parts of the world, where annual bag limits are typical one deer, are amazed to hear that hunters in some areas of southern Japan can legally kill 180 sika deer per year. High annual bag limits are the rule across the country. Still, due to a lack of hunters, paid cullers have been necessary to achieve the desired control in a number of areas. So, is easy to be fooled by the sika deer’s appearance, delicate build, and liquid dark eyes. Indeed, the sika deer is a study in contradictions, both beauty and beast. Whether sika deer are a blessing or a curse is entirely dependent on the situation. Sika deer are seen as a positive and esthetic element when they are on the fringes of the human enterprise, whereas in high numbers they can destroy the livelihoods of the people. No other deer species reaches the extremes shown by the sika deer. That is part of what makes the sika deer such a fascinating animal. This great intertwining of climate, geology, people, and sika deer created a plethora of experiments over time and space driven by natural and cultural forces. These, in turn, present biologists with an array of research opportunities to study processes that have led to the modern distribution and status of the sika deer. Many researchers have been attracted to the natural experiments and the management problems. Research on the sika deer has burgeoned in recent years, and students of deer biology in other parts of the world are not aware of much of this work, which rivals in quality the best being done anywhere. Knowledge about sika deer has now reached a critical mass and justifies this book. It is a propitious time to bring the results scattered about in journal articles, governmental reports, and researchers’ experiences together and integrated in one source.

Literature Cited Cook, C. E., Y. Wang, and G. Sensabaugh. 1999. A mitochondrial control region and cytochrome b phylogeny of sika deer (Cervus nippon) and report of tandem repeats in the control region. Molecular Phylogenetics and Evolution 12:47–56. Goodman, S. J., H. B. Tamate, R. Wilson, J. Nagata, S. Tatsuzawa, G. M. Swanson, J. M. Pemberton, and D. R. McCullough. 2001. Bottlenecks, drift and differentiation: The population structure and demographic history of sika deer (Cervus nippon) in the Japanese archipelago. Molecular Ecology 10:1357–1370. Knight, J. 2003. Waiting for wolves in Japan: An anthropological study of people-wildlife relations. Oxford University Press, Oxford, United Kingdom. McCullough, D. R. 1996. Metapopulations and wildlife conservation. Island Press, Covelo, California, USA. McCullough, D. R. 1997. Irruptive behavior in ungulates. Pages 69–98 in W. J. McShea, H. B. Underwood, and J. H. Rappole, editors, The science of overabundance: Deer ecology and population management. Smithsonian Institution Press, Washington, DC, USA and London, United Kingdom. Takatsuki, S. 1991. Food habits of sika deer in Japan with reference to dwarf bamboo in Northern Japan. Pages 200–204 in N. Maruyama, editor, Wildlife conservation: Present trends and

8

D. R. McCullough

perspectives for the 21st century. Proceedings of the International Symposium on Wildlife Conservation in Tsukuba and Yokohama, Japan, August 21–25, 1990. Japan Wildlife Research Center, Tokyo, Japan. Walker, B. 2005. The lost wolves of Japan. University of Washington Press, Seattle, Washington, USA.

Chapter 2

Fossil Record of Sika Deer in Japan Yoshinari Kawamura

Abstract Diagnoses for antlers of The sika deer is now presented to distinguish it from other medium-sized deer species based on observation of numerous fossil and living deer antlers from Japan and the adjacent Asian continent as well as examination of published data on the antlers. Well-dated records of fossil antlers from 16 localities were selected from numerous fossil records hitherto known in Japan as reliable occurrences of C. nippon. All the selected localities were situated in the Honshu-Shikoku-Kyushu complex constituting the main part of Japan. Morphological discussions and chronological comments are briefly given for the fossil antlers from these localities. A revision was also made to records of fossil antlers of the medium-sized deer from the adjacent continent which are probably conspecific to C. nippon. The records of C. nippon from the 16 localities are arranged in a chronological framework, which includes climatic records during the Quaternary, to reconstruct the history of C. nippon. C. nippon is inferred to have migrated from China into Honshu-Shikoku-Kyushu through the short-lived land bridge formed in and around the present-day Korea Strait at MIS 12 (about 0.43 Ma) and then to have inhabited Honshu-Shikoku-Kyushu for a long time in association with Cervus kazusensis, a survivor of the medium-sized deer from the Early Pleistocene fauna. After the extinction of C. kazusensis at the terminal stage of the Late Pleistocene, C. nippon became the only medium-sized deer species in Honshu-Shikoku-Kyushu and has survived until the present day.

Introduction The sika deer (Cervus nippon) is now distributed over all the main islands of Japan. From a viewpoint of mammalian biogeography, however, the main islands can be divided into two parts, Hokkaido and the Honshu-Shikoku-Kyushu complex (Fig. 2.1). The present-day fauna of Hokkaido is similar to that of the adjacent continent, while that of Honshu-Shikoku-Kyushu shows higher endemism. Such faunal differences can be explained in relation to the depths of the straits in and around the main islands. The straits between Hokkaido and Sakhalin and the adjaD. R. McCullough et al. (eds.), Sika Deer: Biology and Management of Native and Introduced Populations, DOI: 10.1007/978-4-431-09429-6_2, © Springer 2009

11

12

Y. Kawamura

Fig. 2.1 Location map of the fossil localities discussed here. The numbers of the localities correspond to those in Fig. 2.3.

cent continent are much shallower than the Tsugaru Strait between Hokkaido and Honshu-Shikoku-Kyushu (depth about 130 m). The Korea Strait between Tsushima Island and the Korean Peninsula rivals the Tsugaru Strait in depth and is considered to be the biogeographic boundary between Honshu-Shikoku-Kyushu and the adjacent Asian continent. During the Quaternary, the global climatic oscillation linked to the glacial cycles caused severe sea level changes. The straits sometimes dried up to form land bridges which enabled the immigration of nonflying land mammals from the continent. The sika deer must have immigrated from the continent in some stage of the Quaternary and then spread all over the main islands. Fossils of the sika deer provide direct evidence to clarify its Quaternary history in Japan. Numerous deer fossils have been recorded from the Quaternary of Japan by paleontological studies for more than 100 years. They are easily grouped into fossils of small-, medium-, and large-sized deer, which are exemplified by Hydropotes (water deer), Cervus, and Alces (moose), respectively. Fossils of the medium-sized deer are much more abundant than those of the others. In paleontological studies, especially on the medium-sized deer, antlers have been generally considered as the

2

Fossil Record of Sika Deer in Japan

13

most important fossil records, because they are frequently found in fossil state and are more useful for specific distinction than teeth and bones. But the antlers show a wide range of intraspecific variation, which has often led to over-splitting taxonomy of fossil antlers by previous paleontologists. Thus, reliable taxonomy requires consideration of the variation in fossil antlers. Accurate chronological data for each fossil antler is also required to place it in a historical framework. In this chapter, I provide reliable diagnoses which distinguish antlers of the sika deer from those of other medium-sized deer species. Using the diagnoses and considering the state of preservation, I selected the records of the well-dated antlers assignable to the sika deer from the numerous fossil records in Japan. On the basis of the selected records, I reconstruct the Quaternary history of the sika deer in Japan.

Taxonomy I have examined numerous published data on fossil and living materials of the medium-size deer from Japan and the adjacent continent. I have also observed a great many antler specimens of living and fossil deer stored in many museums, institutes, and universities in Japan and China, as well as those kept as private collections, in order to comprehend intraspecific variation and interspecific difference in deer antlers. My studies have revealed that the Quaternary medium-sized deer fossils from Japan with sufficient morphological and chronological data can be classified into five species, Cervus nippon, C. kazusensis (including C. praenipponicus as a synonym), C. kyushuensis, Axis japonicus, and Elaphurus bifurcatus (including E. shikamai as a synonym) by the following characters of their antlers: 1. 2. 3. 4. 5. 6. 7.

Number of points. Height of the first forking (h in Fig. 2.2). Angle between the brow tine and beam (a in Fig. 2.2). Direction of the brow tine. Brow tine branchless or bifurcated. Curvature of the beam between the first and second forkings. Direction of the second tine.

C. nippon is distinguishable from the other species by the following diagnoses: (1) the antler usually has four points; (2) the first forking is low; (3) the angle between the brow tine and beam is moderate or obtuse; (4) the brow tine extends anterodorsally; (5) the brow tine is branchless; (6) the beam between the first and second forkings curves gently and is not lyre-shaped; and (7) the second tine extends anterodorsally or dorsally. In fossil antlers, the state of preservation strongly affects the taxonomic allocation of each specimen. Complete antlers, although few in fossil state, can be allocated convincingly, while the allocation of antlers in poor preservation is less reliable. Thus it is necessary to categorize the state of preservation to evaluate the reliability of each allocation. Six categories are devised here to express the state of

14

Y. Kawamura

Fig. 2.2 Schematic sketches of antlers of Cervus nippon showing the terminology and categories of preservation (A1–C). Black: preserved part, a: angle between the brow tine and beam, h: height of the first forking.

preservation for antlers of C. nippon (Fig. 2.2). The shaft and burr are, of course, preserved in all the categories. A 1: A2: A3: B1: B2: C:

Complete or almost complete in preservation. All the forkings are preserved. The brow and second tines are also preserved, but the beam above the third forking or the third tine may be damaged. All the forkings are preserved. The brow tine, beam above the third forking, and third tine are damaged. All the forkings are preserved. The brow, second, and third tines as well as the beam above the third forking are damaged. The first and second forkings are preserved, while the third forking is lost. The brow tine is also preserved. The second tine may be damaged. The first and second forkings are preserved, while the third forking is lost. The brow tine is damaged. The second tine may be damaged. The first forking and brow tine are preserved, while the second and third forkings are lost. The beam between the first and second forkings is almost preserved.

Among the antler characters given above (1)–(7), five or more are observable in fossil antlers belonging to these categories (Table 2.1), which were adopted for the present study. However, antlers in poorer preservation than those of category C were omitted. Consequently, antlers from 16 localities with reliable chronological data were selected for the subsequent discussion on the fossil records of C. nippon (Fig. 2.1). All of them were located in Honshu-Shikoku-Kyushu, as were the other four medium-sized deer species recognized herein. Evaluation based on fossil record is, therefore, not possible in Hokkaido. In the following discussion the morphological

2

Fossil Record of Sika Deer in Japan

15

Table 2.1 Relationship between the state of preservation and observable characters in the fossil antlers: + observable or inferable in most specimens, ±inferable in part of specimens, − not observable. State of preservation Character 1 2 3 4 5 6 7

A1

A2

A3

B1

B2

C

+ + + + + + +

± + + + − + +

± + + + − + +

− + + + + + +

− + + + − + +

− + + + + + −

characters of fossil antlers from the 16 localities are considered from a taxonomic point of view and chronological comments are given for the antlers.

Fossil Record in Japan Middle Pleistocene Record Fossil antlers reliably assigned to C. nippon were recorded from only two Middle Pleistocene localities: Yoshinoda, Sodegaura, Chiba Prefecture, and Sahama, Hamamatsu, Shizuoka Prefecture (Fig. 2.1). These can be regarded as the earliest record of C. nippon so far known in Honshu-Shikoku-Kyushu (Fig. 2.3). A left antler from Yoshinoda was described as C. (Sika) nippon by Takakuwa (2006). Judging from his description and figure (his fig. 2), it is referred to A1 in preservation and shows characters (1), (2), and (4)–(7), well coincident with the above-mentioned diagnoses of C. nippon. As regards character (3), the angle is relatively acute in the specimen, but antlers with similar angles are sometimes found in living populations of C. nippon (for example, plates 3 and 4 of Ohtaishi 1976). Thus its original allocation is reliable. This antler occurred from the Kiyokawa Formation in the Shimosa Group. This group is one of the stratotypes of the Middle and Upper Pleistocene in Japan and is accurately correlated with the oxygen isotope fluctuation curve and marine isotope stages (MIS). The horizon yielding the antler is positioned in MIS 7.2–7.4, and dated to about 0.22 Ma (Okazaki et al. 2006). A right antler from Sahama was described and figured by Takahashi et al. (2003), who allocated it to C. nippon. The description and figure (their fig. 4 of plate 1) show its preservation as B2, and all the preserved characters (2)–(4), (6),

16

Y. Kawamura

Fig. 2.3 Chronological distribution of Cervus nippon in Honshu-Shikoku-Kyushu. The dots with numbers on its range bar indicate the fossil localities shown in Fig. 2.1. That of another mediumsized deer species, C. kazusensis, is also shown. The climatic background based on the oxygen isotope fluctuation curve by Shackleton (1995) is shown on the left side. The first appearance of the key proboscidean species (Palaeoloxodon naumanni) in Honshu-Shikoku-Kyushu is indicative of a land bridge, through which land mammals immigrated to Japan from the continent. (The chronological range of the proboscidean species is cited from Yoshikawa et al. 2007.)

and (7) accord with the diagnoses of C. nippon, which justifies its original allocation. This antler was recovered from the Sahama Mud Bed in the Hamamatsu Formation. The formation is reliably correlated with the well-dated standard sequences including the Shimosa Group by using marker tephras intercalated, and the bed is allocated to MIS 7 (Sugiyama 1991).

Late Pleistocene Record Fossil antlers reliably assigned to C. nippon occurred from only three Late Pleistocene localities: Nishikigaoka, Shimoda, Aomori Prefecture, the Hanaizumi site, Hanaizumi, Iwate Prefecture, and the Tategahana site, Shinano, Nagano Prefecture (Fig. 2.1). A fragmented skull with a left and right antler from Nishikigaoka was described by Takakuwa (2004) with five figures of the antlers (his figs. 1–5). He referred it to C. (Sika) nippon. On the basis of his description and figures, the left antler is better preserved and belongs to A1. This antler therefore preserves all the characters (1)– (7), which agree with the diagnoses of C. nippon. This indicates that its original

2

Fossil Record of Sika Deer in Japan

17

allocation is undoubted. The skull was found in marine terrace deposits, which were assigned to MIS 5 (Kuwabara 2004; Takakuwa 2004). A small right antler from the Hanaizumi site was studied by Matsumoto et al. (1959), who identified it as “C. (Sika) natsumei.” Judging from their description and figures (their fig. 3, and fig. 1 of plate 32), this antler probably belongs to a young individual and is referable to B1 in preservation. The preserved characters (2)–(7) coincide with the diagnoses of C. nippon, and thus the antler is reliably assigned to C. nippon. “C. natsumei” proposed by Matsumoto (1938) is considered to be synonymous with C. nippon, but the taxonomic problems with “C. natsumei” will be discussed in a separate paper. The sediments yielding the antler are dated between 16 and 28 ka by the radiocarbon method (Kanto Loam Research Group and Shinshu Loam Research Group 1962; Hanaizumi Site Excavation Research Group 1993). A left antler from the Tategahana site was referred to “Cervus sp. cf. nippon” by Nojiriko Excavation Research Group (1975). According to their description and figure (fig. 11-8), the antler belongs to C in preservation, and shows the characters (2)–(6) coincident with the diagnoses of C. nippon. It is therefore referable to C. nippon. The sediments yielding the antler are determined as T1–T3 units of the Tategahana Sand Member in the Nojiri-ko Formation. These units range from 41 to 43 ka in radiocarbon age (Geology Research Group for Nojiri-ko Excavation 2004).

Holocene Record Fossil antlers of Holocene age are more numerous than those of Middle and Late Pleistocene ages and were obtained from archaeological sites which were accurately dated mainly by the archaeological method using pottery. The eleven sites listed in Table 2.2 yield antlers which show the characters coincident with the diagnoses of C. nippon (at least five characters among the seven, see Table 2.1), and thus they are reliably assigned to C. nippon. In these sites, however, antlers suitable for specific determination are few (mostly one or two). The only exception is the case of the Torihama shell mound site, which yields many well-preserved antlers figured by Torihama Shell Mound Research Group (1983, 1985, 1987a,b) and Shigehara et al. (1991). The antlers figured show a wide range of morphological variation, but their allocation to C. nippon is undoubted, because similar variation is observable in living populations of C. nippon.

Fossil Record in the Adjacent Continent In northern China, well-preserved and abundant fossils of the medium-sized deer have been obtained from cave sediments of many localities at Zhoukoudian (= Choukoutien), a famous place for paleoanthropology and paleontology in Beijing.

Kainohana shell mound site, Matsudo, Chiba Pref. Mawaki site, Noto, Ishikawa Pref. Aoshima shell mound site, Minamikata, Miyagi Pref.

7

9

8

6

Sanya shell mound site, Sodegaura, Chiba Pref. Morinomiya site, Higashi-ku, Osaka

Onji site, Higashi-osaka, Osaka Pref. Kamei site, Hirano-ku, Osaka, and Yao, Osaka Pref. Shirahama shell mound site, Fukue (Goto-shi), Nagasaki Pref. Gionbara shell mound site, Ichihara, Chiba Pref.

5

4

3

2

1

Locality

ca. 2 ka

ca. 2–3 ka

ca. 2.5–4 ka Archaeological dating using pottery

Japanese sika deer Deer

Japanese sika deer Cervus nippon nippon Japanese sika deer

A1 B1, B2 A3 A1, B1

Cervus nippon nippon Cervus (Sika) nippon matsumotoi

B1, B2 A1

Cervus nippon nippon

B2

B2

ca. 2.5– Archaeological dating using 6.5 ka pottery ca. 3–6.5 ka Archaeological dating using pottery

ca. 2.5–5 ka Archaeological dating using pottery, and radiocarbon

ca. 2.5–4 ka Archaeological dating using pottery ca. 2–5 ka Archaeological dating using pottery, and radiocarbon

Archaeological dating using pottery

Archaeological dating using pottery, and radiocarbon Archaeological dating using pottery

ca. 2 ka

Cervus nippon

B2

Dating method

Metric age

State of Original preservation identification

Miyazaki and Hiraguchi (1986) Matsumoto (1930); Kato and Goto (1975)

Taruno and Ishii (1978); Matsuo (1978) Yawata (1973)

Uriudo Site Research Group (1980) Osaka Cultural Property Center (1984) Fukue Municipal Board of Education (1980) Cultural Property Center of Ichihara City (1999) Kaneko et al. (1973)

Literature

Latest to Early Jomon in age Late to Early Jomon in age

Latest to Late Jomon in age Yayoi and Latest to Middle Jomon in age Latest to Middle Jomon in age

Early Yayoi and Latest Jomon in age Latest to Late Jomon in age

Yayoi in age

Yayoi in age

Remarks

Table 2.2 Well-dated localities of Holocene age yielding antlers assignable to Cervus nippon with confidence. The numbers of the localities correspond to those in Figs. 2.1 and 2.3.

Torihama shell mound site, Mikata (Wakasa-cho), Fukui Pref.

Awazu shell midden site, Otsu, Shiga Pref.

10

11

B1, B2 Cervus nippon

A1, A2, A3, Cervus nippon B1, B2, C

ca. 4–10 ka

Archaeological dating using pottery

ca. 5–6.5 ka Archaeological dating using pottery

Torihama Shell Mound Early Jomon Research Group in age (1983, 1985, 1987a, b); Shigehara et al. (1991) Shiga Prefectural Middle to Earliest Board of Education Jomon in age and Shiga Institute for Cultural Heritage Protection (1984, 1997)

20

Y. Kawamura

The fossils of the genus Cervus from the Middle Pleistocene sediments of Locality 1 were assigned to “Pseudaxis grayi” (= Cervus grayi) by Zdansky (1928) and Young (1932). “P. grayi” was also reported from the early Middle Pleistocene sediments of Locality 13 (Teilhard de Chardin and Pei 1941). On the basis of the descriptions and figures in these papers, the antlers of “P. grayi” are large and have broad angles between the brow tine and beam, but they are not basically different from those of C. nippon. “P. grayi” is, therefore, considered to be a synonym of C. nippon. Further discussion on the systematic position of “P. grayi” will be given in a separate paper. Additionally Pei (1936) described an antler from the late Middle Pleistocene sediments of Locality 3 and referred it to “Pseudaxis hortulorum.” This antler shows the characters coincident with the diagnoses of C. nippon. A skull with both antlers was recovered from the Late Pleistocene sediments of Guxiangtun (= Ku-hsiang-tung), Harbin, Heilongjiang Province and was described as the type specimen of a new species named “C. harbinensis” (Tokunaga and Naora 1939). The morphological characters of the antlers described well agree with the diagnoses of C. nippon. Thus “C. harbinensis” is synonymous with C. nippon, but the problems of “C. harbinensis” will be discussed in a separate paper. The Holocene archaeological site of Anyang, Henan Province yielded several antlers of “P. hortulorum,” which were described by Teilhard de Chardin and Young (1936). Their description and figures indicate that the antlers bear the diagnoses of C. nippon, and are assignable to this species. In conclusion, the above-mentioned fossil records indicate that C. nippon has continuously inhabited northern China from the early Middle Pleistocene to Holocene.

Immigration The reliable fossil records of C. nippon from Honshu-Shikoku-Kyushu are arranged chronologically in Fig. 2.3. They indicate that C. nippon ranges from the late Middle Pleistocene (0.22 Ma) to the present day. Taking the above-mentioned chronological distribution in northern China into account, C. nippon probably immigrated from China prior to 0.22 Ma. In Honshu-Shikoku-Kyushu, proboscidean (elephant-related) fossils are successively found in the Quaternary sequences, and the chronological range of each proboscidean species is determined accurately in the sequences. Kawamura (1998) considered that the first appearance of each species indicated its immigration from China through a land bridge. This idea was improved by the more precise range of each species and correlation with the oxygen isotope fluctuation curve (Konishi and Yoshikawa 1999; Kawamura and Taruno 2000; Yoshikawa et al. 2007). The curve linked to global climatic changes has a decisive meaning when we consider the extreme sea-level drops which formed land bridges across the present-day Tsugaru and Korea Straits (Fig. 2.1). The last proboscidean species in Honshu-ShikokuKyushu (Palaeoloxodon naumanni) ranges from about 0.38 to about 0.02 Ma (Fig. 2.3).

2

Fossil Record of Sika Deer in Japan

21

This species is inferred to have immigrated at MIS12, because it is a very cold stage just prior to 0.38 Ma. As regards C. nippon, three cold stages (MIS 8, 10, and 12) are recognized prior to 0.22 Ma. Among them, MIS 12 is most probable as the time of its immigration, because MIS 12 is the coldest and P. naumanni is inferred to have immigrated as mentioned above. At MIS 12, C. nippon possibly immigrated from northern China through the exposed area in and around the present-day Korea Strait in association with P. naumanni. As pointed out by Kawamura (2007), this land bridge is considered to have been short-lived, because the fluctuation curve shows a very sharp peak at MIS 12, as other cold stages. C. nippon may have expanded its distribution into Hokkaido through the Tsugaru Strait when it dried up at MIS 12. There is no clear evidence indicative of mammalian immigration or land bridge formation after MIS 12, excepting the ice bridge formation across the Tsugaru Strait at MIS 2 (Fig. 2.3). The ice bridge was inferred by Kawamura (1985, 1989, 1994), through which a few elements of the mammoth fauna complex of the continent invaded Honshu-Shikoku-Kyushu from Hokkaido. At MIS 2, C. nippon may have interchanged between Honshu-Shikoku-Kyushu and Hokkaido through this ice bridge, but it is unlikely that the continental population of this species immigrated to Honshu-Shikoku-Kyushu through Sakhalin and Hokkaido which were undoubtedly connected by land with the adjacent continent. The reason is that on the continent, the Ussuri region is the northernmost distribution of C. nippon at the present day which is in a very warm stage (MIS 1); the northern limit is inferred to have shifted southward to northern China at MIS 2 and, thus, C. nippon was probably absent from the continent adjacent to Sakhalin at MIS 2.

Changes in the Medium-Sized Deer Fauna Besides Cervus nippon, four medium-sized deer species are recognized here in welldated fossil records of Quaternary age from Honshu-Shikoku-Kyushu, as mentioned before. Among the four, C. kyushuensis, Axis japonicus, and Elaphurus bifurcatus are restricted in their occurrences to the Early Pleistocene. The remaining species, C. kazusensis, has a long chronological range from the Early Pleistocene to the terminal stage of the Late Pleistocene (Fig. 2.3). In spite of the over-lumping taxonomy by Nakaya (1993), C. kazusensis is distinguishable from C. nippon by antler morphology; namely characters (1), (2), and (7) differ between them. In C. kazusensis, the antler has three points, the first forking is high, and the second tine extends posterodorsally and medially. The differences are sufficiently specific. The fossil records suggest that C. kazusensis was a survivor of the more diversified medium-sized deer fauna of the Early Pleistocene and lived alone in HonshuShikoku-Kyushu during the early part of the Middle Pleistocene. After the immigration of C. nippon at MIS 12, C. kazusensis coexisted with C. nippon for a long time, possibly more than 400,000 years. They seem to have been distributed

22

Y. Kawamura

allopatrically (separately) in Honshu-Shikoku-Kyushu, because the two species have never been found together in well-dated sediments in a single locality. In Honshu-Shikoku-Kyushu, several mammalian species including C. kazusensis and Palaeoloxodon naumanni became extinct at the terminal stage of the Late Pleistocene, between 20 and 10 ka (Kawamura 1991, 1994; Fig. 2.3). This extinction event decreased the species diversity of the mammalian fauna and made the medium-sized deer fauna again monospecific, the survivor being C. nippon. This species was well established in the Holocene mammalian fauna of HonshuShikoku-Kyushu and has survived to the present despite increasing hunting pressure by prehistoric and historic people. Acknowledgements This chapter has been written to publish a part of my research on Quaternary deer fossils from Japan and adjacent countries. I wish to express my gratitude to Prof. T. Ozawa (Cyber University), Mr. H. Taruno (Osaka Museum of Natural History), Prof. S. Matsu’ura (Ochanomizu University), Prof. T. Inada (Okayama University), Prof. H. Yamazaki (Tokyo Metropolitan University) for their helpful discussion and encouragement and affording facilities for my observation of specimens or literature collection. I am also indebted to Mr. Y. Abe (Taga Town Museum), Dr. Y. Kondo (Nojiri-ko Museum), Mr. Y. Matsuhashi (Aichi University of Education), Mr. Y. Takakuwa (Gunma Museum of Natural History), and Mr. I. Risho for providing useful information or helpful assistance. Thanks are due to Prof. C. Z. Jin, Dr. Y. Q. Zhang, and Dr. W. Dong (Institute of Vertebrate Paleontology and Paleoanthropology, Chinese Academy of Sciences), and Prof. Z. J. Feng, Mr. Y. G. Xu, and Ms. L. H. Sun (Institute of Zoology, Chinese Academy of Sciences) for giving me convenience to access specimens from China and to obtain useful information concerned. I also acknowledge Dr. D. R. McCullough and Ms. M. Jaeger for critical reading of my manuscript to improve English expression and to adjust it to the purpose of this book; and Prof. S. Takatsuki (Azabu University) for giving me the opportunity to publish it in this book. This study was financially supported by Grants-in-Aid for Scientific Research from the Japan Society for the Promotion of Science (project nos. 13854001 and 14390027).

Literature Cited Cultural Property Center of Ichihara City, editor. 1999. Gionbara shell mound. Ichihara Municipal Board of Education, Chiba Prefecture, Japan. (In Japanese.) Fukue Municipal Board of Education, editor. 1980. Shirahama shell mound. Fukue Municipal Board of Education, Nagasaki Prefecture, Japan. (In Japanese.) Geology Research Group for Nojiri-ko Excavation. 2004. Geology of the excavation site at Lake Nojiri and its surrounding areas, northern part of Nagano Prefecture − On the redefinition of the Nojiri-ko Formation. Bulletin of the Nojiri-ko Museum 12:1–13. (In Japanese with English abstract.) Hanaizumi Site Excavation Research Group. 1993. The Hanaizumi site. Board of Education of Hanaizumi-cho, Iwate Prefecture, Japan. (In Japanese.) Kaneko, H., M. Koyanagi, and Y. Ushizawa. 1973. Vertebrate remains from the Sanya shell mound at Iitomi. Pages 221–229, plates 85–90 in Sanya shell mound, Sodegaura-cho. Tokyo Electric Power Company, Tokyo and Urban Development Public Corporation of Chiba Prefecture, Chiba, Japan. (In Japanese.) Kanto Loam Research Group and Shinshu Loam Research Group. 1962. On the geological age and sedimentary environment of the Hanaizumi Bed. Earth Science (Chikyu Kagaku) 62:1–10; 63:10–18. (In Japanese with English abstract.)

2

Fossil Record of Sika Deer in Japan

23

Kato, T., and K. Goto. 1975. Report on the excavations of the Aoshima shell mound, Minamikatamachi, Tome-gun. Pages 3–145 in Report on the excavations of the Aoshima shell mound, Minamikata-machi, Tome-gun, Miyagi Prefecture: Research on an inland freshwater shell mound. Minamikata-machi, Miyagi Prefecture, Japan. (In Japanese.) Kawamura, Y. 1985. Succession of the mammalian fauna in Japan since the Last Glacial Period. The Earth Monthly 7:349–353. (In Japanese.) Kawamura, Y. 1989. Quaternary rodent faunas in the Japanese Islands (Part 2). The Memoirs of the Faculty of Science, Kyoto University, Series of Geology and Mineralogy 54:1–235. Kawamura, Y. 1991. Quaternary mammalian faunas in the Japanese Islands. The Quaternary Research (Daiyonki-Kenkyu) 30:213–220. Kawamura, Y. 1994. Late Pleistocene to Holocene mammalian faunal succession in the Japanese Islands, with comments on the Late Quaternary extinctions. ArchaeoZoologia 6:7–22. Kawamura, Y. 1998. Immigration of mammals into the Japanese Islands during the Quaternary. The Quaternary Research (Daiyonki-Kenkyu) 37:251–257. (In Japanese with English abstract.) Kawamura, Y. 2007. Recent progress in paleontological studies on the Quaternary mammals of Japan. Honyurui Kagaku (Mammalian Science) 47:107–114. (In Japanese.) Kawamura, Y., and H. Taruno. 2000. Immigration of mammals into Japan during the Quaternary, with comments on land or ice bridge formation enabled human immigration. Acta Anthropologica Sinica, supplement to volume 19:264–269. Konishi, S., and S. Yoshikawa. 1999. Immigration times of the two proboscidean species, Stegodon orientalis and Palaeoloxodon naumanni, into the Japanese Islands and the formation of land bridge. Earth Science (Chikyu Kagaku) 53:125–134. (In Japanese with English abstract.) Kuwabara, T. 2004. Relative sea-level changes and marine-terrace deposits in Kamikita Plain, northern end of Honshu, Japan. The Journal of the Geological Society of Japan 110:93–102. (In Japanese with English abstract.) Matsumoto, H. 1930. Report of the mammalian remains obtained from the sites at Aoshima and Hibiku, Province of Rikuzen. The Science Reports of the Tohoku Imperial University, Second Series (Geology) 13:59–93, plates 30–38. Matsumoto, H. 1938. On some fossil antlers of deers [sic] from the Basal Calabrian at Nagahama, Minato Town, Province of Kazusa, Japan. The Zoological Magazine (Dobutsugaku Zasshi) 50:111–115. (In Japanese with English résumé.) Matsumoto, H., H. Mori, K. Marui, and H. Ozaki. 1959. On the discovery of the Upper Pliocene fossiliferous and culture-bearing bed at Kanamori, Hanaizumi Town, Province of Rikuchu. Bulletin of the National Science Museum, Tokyo 4:287–324, plates 25–48. Matsuo, N. 1978. Conclusion. Pages 171–177 in The Morinomiya site: Report on the third and fourth excavations. Naniwanomiya-ato-kenshokai, Osaka, Japan. (In Japanese.) Miyazaki, N., and T. Hiraguchi. 1986. Animal remains. Pages 346–400 in The Mawaki site in Noto-cho, Ishikawa Prefecture. Board of Education of Noto-cho and Mawaki Site Excavation Research Group, Ishikawa Prefecture, Japan. (In Japanese.) Nakaya, H. 1993. Evolution of Quaternary middle-sized Cervus in Japan and China. Pages 106– 114 in N. Ohtaishi, and H. I. Sheng, editors. Deer of China: Biology and management. Elsevier Science, Amsterdam, The Netherlands. Nojiriko Excavation Research Group. 1975. The Lake Nojiri excavation 1962–1973. Kyoritsu Shuppan Company, Ltd, Tokyo, Japan. (In Japanese with English résumé.) Ohtaishi, N. 1976. Developmental variation of the antlers on Japanese deer at Nara Park (Preliminary). Pages 107–128 in Annual report of Nara Deer Research Association for the fiscal year 1976. Kasugakenshokai, Nara, Japan. (In Japanese with English summary.) Okazaki, H., H. Nakazato, and H. Ikeda. 2006. The flood deposit of the Middle Pleistocene Kiyokawa Formation, Shimosa Group, eastern Japan. The Quaternary Research (DaiyonkiKenkyu) 45:157–167. (In Japanese with English abstract.) Osaka Cultural Property Center. 1984. The Kamei site II. Osaka Cultural Property Center, Osaka, Japan. (In Japanese.) Pei, W. C. 1936. On the mammalian remains from Locality 3 at Choukoutien. Palaeontologia Sinica, Series C 7 (fasc. 5):1–121.

24

Y. Kawamura

Shackleton, N. J. 1995. New data on the evolution of Pliocene climatic variability. Pages 242–248 in E. S. Vrba, G. H. Denton, T. C. Partridge, and L. H. Burckle, editors. Paleoclimate and evolution, with emphasis on human origins. Yale University Press, New Haven, Connecticut, USA. Shiga Prefectural Board of Education and Shiga Institute for Cultural Heritage Protection, editors. 1984. Awazu shell midden: A lake bottom site. Shiga Prefectural Board of Education and Shiga Institute for Cultural Heritage Protection, Otsu, Japan. (In Japanese.) Shiga Prefectural Board of Education and Shiga Institute for Cultural Heritage Protection, editors. 1997. Awazu no.3 shell midden. Shiga Prefectural Board of Education and Shiga Institute for Cultural Heritage Protection, Otsu, Japan. (In Japanese with English summary.) Shigehara, N., H. Hongo, and K. Amitani. 1991. Mammal remains from the 1985 excavation of the Torihama shell mound. Bulletin of the National Museum of Japanese History 29:329–342, plates 1–15. (In Japanese with English abstract.) Sugiyama, Y. 1991. The Middle Pleistocene deposits in the Atsumi Peninsula and along the east coast of Lake Hamana, Tokai district – sedimentary cycles formed by the glacio-eustatic sea-level change and their correlations to the contemporaneous deposits in the Kanto and Kinki districts–. Bulletin of the Geological Survey of Japan 42:75–109. (In Japanese with English abstract.) Takahashi, K., H. Matsuoka, H. Taru, K. Yasui, and Y. Hasegawa. 2003. Vertebrate fossils from the excavation site of the Naumann’s elephant at Sahama. Shizuoka Chigaku 87:15–21. (In Japanese.) Takakuwa, Y. 2004. Sika deer fossils from the Middle Pleistocene of Nishikigaoka, Shimoda, Kamikita-gun, Aomori Prefecture. Report on sika deer fossils from Nishikigaoka, Shimodacho. Board of Education of Shimoda-cho, Aomori Prefecture, Japan. (In Japanese.) Takakuwa, Y. 2006. Cervid fossils from the Kiyokawa Formation of Shimosa Group, Sodegaura, Chiba Prefecture, Japan. The Quaternary Research (Daiyonki-Kenkyu) 45:197–206. (In Japanese with English abstract.) Taruno, H., and M. Ishii. 1978. Animal remains from the Morinomiya site (the third excavation). Pages 160–165 in The Morinomiya site: Report on the third and fourth excavations. Naniwanomiya-ato-kenshokai, Osaka, Japan. (In Japanese.) Teilhard de Chardin, P., and C. C. Young. 1936. On the mammalian remains from the archaeological site of Anyang. Palaeontologia Sinica, Series C 12 (fasc. 1):1–79. Teilhard de Chardin, P., and W. C. Pei. 1941. The fossil mammals from Locality 13 of Choukoutien. Palaeontologia Sinica, New Series C 11:1–119. Tokunaga, S., and N. Naora. 1939. Fossil remains excavated at Ku-hsiang-tung near Harbin, Manchoukuo. Pages 1–229, plates 1–22 in Report of the first scientific expedition to Manchoukuo, Section 2, Part 4. Office of the Scientific Expedition to Manchoukuo, Tokyo, Japan. (In Japanese with English résumé.) Torihama Shell Mound Research Group, editor. 1983. Torihama shell mound: Outline of the excavations and results of the researches in the fiscal years 1981–1982. Fukui Prefectural Board of Education and Wakasa Museum of History and Folklore, Fukui Prefecture, Japan. (In Japanese.) Torihama Shell Mound Research Group, editor. 1985. Torihama shell mound: Outline of the excavations and results of the researches in the fiscal year 1984. Fukui Prefectural Board of Education and Wakasa Museum of History and Folklore, Fukui Prefecture, Japan. (In Japanese.) Torihama Shell Mound Research Group, editor. 1987a. Torihama shell mound: Researches of the fiscal years 1980–1985. Fukui Prefectural Board of Education and Wakasa Museum of History and Folklore, Fukui Prefecture, Japan. (In Japanese.) Torihama Shell Mound Research Group, editor. 1987b. Torihama shell mound: Outline of the excavations and results of the researches in the fiscal year 1985. Fukui Prefectural Board of Education and Wakasa Museum of History and Folklore, Fukui Prefecture, Japan. (In Japanese.) Uriudo Site Research Group, editor. 1980. The Onji site. Uriudo Site Research Group, Higashiosaka, Osaka Prefecture, Japan. (In Japanese.)

2

Fossil Record of Sika Deer in Japan

25

Yawata, I., editor. 1973. The shell mounds of Kainohana. Board of Education of Matsudo City, Chiba Prefecture, Japan. (In Japanese with English summary.) Yoshikawa, S., Y. Kawamura, and H. Taruno. 2007. Land bridge formation and proboscidean immigration into the Japanese Islands during the Quaternary. Journal of Geosciences, Osaka City University 50:1–6. Young, C. C. 1932. On the Artiodactyla from the Sinanthropus site at Chouk’outien. Palaeontologia Sinica, Series C 8 (fasc. 2):1–159. Zdansky, O. 1928. Die Säugetiere der Quartärfauna von Chou-K’ou-Tien. Palaeontologia Sinica, Series C 5 (fasc. 4):1–146, plates 1–16.

Chapter 3

Two Genetically Distinct Lineages of the Japanese Sika Deer Based on Mitochondrial Control Regions Junco Nagata

Abstract To investigate genetic diversity among populations of sika deer, Cervus nippon, in Japan, nucleotide sequences (705–824 bases) of the mitochondrial control regions were determined in 61 animals from localities in the Japanese islands and 13 animals from three localities in China. A phylogenetic tree constructed by the sequences indicated that the Japanese sika deer are separated into two distinct lineages: the Northern Japan group (Hokkaido Island and most of the Honshu mainland) and the Southern Japan group (a part of the southern Honshu mainland, Kyushu Island and small islands around Kyushu Island). All sika deer examined in this study shared four to seven units of repetitive sequences (37–40 bases each) within the control region sequences. The number of tandem repeats was different between the two lineages. Six or seven repeats occurred in the northern group, while four or five repeats occurred in the southern group. Based on these control region data, separation of the two lineages was estimated to have occurred approximately 0.35 million years before present. The divergence of the two groups coincides with the last glacial period during the Pleistocene and suggests that there were at least two invasions from the continent to Japan possibly through the land bridges of the Korean Strait.

Introduction The sika deer has been classified as a member of the family Cervidae in the order Artiodactyla (Corbet and Hill 1991). Sika deer occupy most of the islands of the Japanese Archipelago, from Hokkaido in the north to islands off the coast of Okinawa in the south. The sika deer on the Japanese islands are divided into six subspecies (Ohtaishi 1986), and this classification is widely accepted at present: Cervus nippon yesoensis (Hokkaido Island population), C. n. centralis (Honshu mainland and Tsushima Island populations), C. n. nippon (Kyushu Island, Shikoku Island, and Goto Islands populations), C. n. mageshimae (Mageshima Island and Tanegashima Island populations),

D. R. McCullough et al. (eds.), Sika Deer: Biology and Management of Native and Introduced Populations, DOI: 10.1007/978-4-431-09429-6_3, © Springer 2009

27

28

J. Nagata

C. n. yakushimae (Yakushima Island and Kuchinoerabu Island populations), and C. n. keramae (Ryukyu Islands population). However, there have been some questions on taxonomy of the Tsushima population (Imaizumi 1970; Corbet 1978; Matsumoto et al. 1984; Endo 1996). The sika deer in Japan show a striking variation in body weight from the northern to southern islands: the largest is C. n. yesoensis, while the smallest is C. n. yakushimae with a weight less than half of the former. It is unusual for mammals to show as great a difference as Japanese sika deer show. Morphological variation within Japanese sika is almost equivalent to that seen among individual species of other mammals. Despite remarkable differences in body size and morphology, taxonomy of sika deer subspecies has been controversial. Thus, sika deer in Japan provide an excellent opportunity to study relationships between biogeographic and genetic characters. Several previous studies have reported that there are two mitochondrial (mt) lineages of sika deer in Japan: the Northern Japan group and the Southern Japan group (Nagata et al. 1995; Tamate and Tsuchiya 1995; Tamate et al. 1998; Cook et al. 1999; Nagata et al. 1999). These molecular data did not reflect the previous morphological classification of sika deer subspecies. Fossil records indicated that the sika deer could have colonized the Japanese islands in the Mid- to Late Pleistocene (Kawamura 1982; Kawamura et al. 1989; Kawamura 1991, Kawamura chapter 2). The general opinion is that the animals could have migrated to the islands through land bridges that were repeatedly formed between Japan and the Asian continent. In the Late Pleistocene, as the Japanese islands were isolated by surrounding straits, deer populations became confined to each island. Such isolation would have initiated subspeciation. However, where the Japanese sika deer originally came from and their path of migration and speciation are still unclear. As techniques for genetics improve, theoretical concepts have been confirmed, and it is clear that information from DNA sequences is very useful to understand phylogenetic relationships among animals. Phylogenetic analyses have been used in many fields in biology because convergence is easier to control for in DNA studies than in morphological studies. Genetic analysis of mtDNA sequences have been recently applied in wildlife studies to provide reliable information on the relationships among closely related species and among populations, because of more rapid evolution of mtDNA, maternal inheritance, and nonrecombination (Brown et al. 1979, 1982). In particular, the control region sequence is most variable in mtDNA and its substitution rate is estimated to be five times as much as that of the rest of the sequences in mtDNA (Aquadro and Greenburg 1983). For this reason, my colleagues and I have been studying genetic diversity of sika deer. It is becoming clear that the Japanese sika deer has a dynamic history that can not be inferred from its distribution pattern or variation of morphological characters. In the present study, we determined nucleotide sequences of the mtDNA control region and present genetic characteristics of the Japanese sika deer. Then, we discuss genetic relationships among Japanese sika deer populations to understand their history and evolution.

3

Two Genetically Distinct Lineages of the Japanese Sika Deer

29

Materials and Methods Animal Collection and DNA Extraction A total of 74 sika deer specimens (blood, muscle, or liver tissues) were collected from three Chinese subspecies (n = 13) and all Japanese subspecies (n = 61) of sika deer (Table 3.1, Fig. 3.1). Nucleotide sequences from 31 of the 74 animals have already been reported by Nagata et al. (1999), and the remaining animals (43 samples) were newly analyzed in this study. From the whole blood or other tissues, total DNA was extracted using QIAamp minikits (QIAGEN), following the manufacture’s instructions. Extract without any tissue was used as a negative control in the subsequent polymerase chain reaction (PCR) amplification.

PCR Product-Direct Sequencing of the Control Region To amplify and sequence the control regions of the sika deer, 17 primers were used (Table 3.2, Fig. 3.2). The primer pair of L15926/H597 was used for PCR amplification. In the case that PCR failed with L15926/H597, other pairs L15926/ HD6 and LD3/H597 were used. In order to amplify the control region a PCR reagent kit (TaKaRa) was used according to the manufacturer’s instructions. One microliter of the DNA extract was subjected to PCR amplification in a reaction mixture of 50 µl including 10 mM Tris (pH 8.3), 50 mM KCl, 1.5 mM MgCl2, each dNTP at 0.2 mM, 1.25 U of Taq DNA polymerase, and each primer at 0.5 µM. The step program for PCR amplification was as follows: 94 °C for 1 min, 55 °C for 1 min, and 72 °C for 1 min. The cycle was repeated 30–40 times followed by a reaction completion at 72 °C for 10 min. Sequences were obtained with ABI Dye Terminator Kit (Applied Biosystems) and an ABI 310 automatic sequencer. Cycle sequencing reactions were performed according to the instructions provided by the manufacturer. All PCR primers (except L15926) were used for the direct sequencing.

Sequence Analysis Sequence analysis was performed using a computer software GENETYX-MAC Ver.8.5 (SOFTWARE DEVELOPMENT CO., Tokyo, Japan). Phylogenetic trees were constructed using the neighbor-joining (NJ) method (Saitou and Nei 1987) in the computer software Clustal X 1.81 (Thompson et al. 1997). Numbers of nucleotide substitutions per site were estimated for multiple substitutions using the Kimura’s two-parameter method (Kimura 1980). The 31 sequences from Nagata et al. (1999) were included in

30

J. Nagata

Table 3.1 Taxonomy and haplotypes of sika deer (Cervus nippon) (74 individuals) analyzed in this study. No. in Subspeciesa Fig. 3.1 JAPAN C. n. yesoensis 1 2 C. n. centralis 3 4 5

Haplotypes (N/N of tandem repeats)

Groups b

Shari, Hokkaido 2 Utanobori, Hokkaido 4

Hka (2/7) Hkb (4/7)

N N

Mt. Goyo, Iwate, Honshu Chiba, Honshu

2

Gyo1 (2/6)

N

13

N

2

Kmo1 (1/7)/Kmo2 (3/7)/Ama1 (9/6) Kna1 (1/7)/Kna2 (1/7)

N

1 2

Tga1 (1 /7) Wda1 (1/6)/Wda2 (1/6)

N N

3 6

Yma1 (3/4) Tma1 (6/4)

S S

Gto1 (3/4) Gto1 (1/4) Mya1 (1/5)/Mya2 (1/5)/Mya3 (2/5) Kgo1 (4/4)

S S S

Localities

N examined

8 9

Kanayama,Gifu, Honshu Taga, Shiga, Honshu Wadayama, Hyogo, Honshu Yamaguchi, Honshu Tsuhima Island

10 11 12

Goto Islands Nagasaki, Kyushu Miyazaki, Kyushu

3 1 4

13 C. n. yakushimae 14

Kagoshima, Kyushu

4

Yakushima Island

7

15 C. n. mageshimae 16 C. n. keramae 17

Kuchinoerabu Island

6 7

C. n. nippon

CHINA C. n. hortulorum 18 C. n. sichuanicus 19 C. n. kopchi 20 a

S

3

Yku1 (1/4)/Yku2 (1/4)/ Yku3 (2/4)/Yku4 (2/4)/Yku5 (1/4) Kti1 (3/4)

S

S

Tanegashima Island

3

Tng1 (3/4)

S

Kerama Island

1

Kra1 (1/5)

S

Jilin Province

7

CNCI3 (7/4)

C

Sichuan Province

4

CNCI1 (4/4)

C

Anhui Province

2

CNCI2 (1/4)/CNCI4 (1/4)

C

Ohtaishi (1986) and Ohtaishi and Gao (1990). N, S, or C represent Northern Japan group, Southern Japan group, or Chinese group, respectively (see Figs. 3.2 and 3.3).

b

3

Two Genetically Distinct Lineages of the Japanese Sika Deer

31

Fig. 3.1 Sampling locations of the sika deer, Cervus nippon. Numbers correspond to those in Table 3.1.

Table 3.2 Primers for PCR and sequencing. Name

Sequences (5′–3′)

L strand LD15926 CTAATACACCAGTCTTGTAAACC CervL1 CAACCTTCAAGGAAGAAGCCAT LD5 AAGCCATAGCCCCACTATCAA LD15 TATATGCCCCATGCTTATAAGC CervL3 ACCATGCCGCGTGAAACCAG LD3 CTCTTCTCGCTCCGGGCCCATGAA CervL4 GACTAATGACTAATCAGCCCAT LD7 ACTCAGCAATGGCCGTCTGAGG LD9 ATCATCATTTTTAACACACTTT H strand HD10 TAGGACATAAATGTAAATTGGGTG HD2 CCTGAAAAAAGAACCAGATG HD14 TGGGGATGCTCAAGATGCAG HD8 TTGACTTAATGCGCTATGTA HD6 GTCCTGTGACCATTGACTGC H597 AGGCATTTTCAGTGCCTTGCTTTG CervH1 CAAACCTATGTGTTTATGGAGT CervH3 CCCGGAGCGAGAAGAGGGAT *Note: Primers with asterisks were created for this study.

Length

Reference and DNA data base accession No.

23 22 21 22 20 24 22 22 22

Nagata et al. 1998 AB295410 Nagata et al. 1998 Nabata et al. 2004 AB295414 Unpublished AB295416 * *

24 20 20 20 20 24 22 20

Nabata et al. 2004 Nagata et al. 1998 * Nagata et al. 1998 Nagata et al. 1998 Nagata et al. 1998 AB295418 AB295415

32

J. Nagata Primers for L strand Tandem repeat CervL1

LD15

L 15926 LD5

Thr

tRNA

LD9

CervL3 CervL5 LD3

LD7

1 2345 67

tRNAPro

HD10

tRNA

HD8

HD2

CervH3

HD6 HD14

Phe

CervH1 H597

Primers for H strand

Fig. 3.2 Schematic diagram of the mitochondrial DNA control region (shaded boxes) of the sika deer. Numbers 1 through 7 indicate tandemly repeated units. Bars show the rough positions of primers used for amplification and/or sequencing.

the analysis. The sequence obtained from one red deer, Cervus elaphus xanthopygus, (AF296817) was used as an outgroup to set a root for the phylogenetic tree. Gap sites and tandem repeated regions were deleted for sequence analysis. The bootstrap analysis (Felsenstein 1985) consisted of 1,000 replications for the NJ tree. We defined three sika deer lineages in the NJ tree; the Northern Japan, the Southern Japan, and China. Nucleotide diversity, π (Nei 1987) and percentage differences were calculated for the three lineages. As an approach to investigate the relationships among the three lineages, genetic distances were calculated using AMOVA (Excoffier et al. 1992). In this analysis, the pairwise sequence differences among the haplotypes were used as a measure of molecular divergence. Significance of the F st value was determined by performing randomization tests using 1,000 replications (Excoffier et al. 1992).

Results MtDNA control regions of the 74 sika deer individuals from 17 localities of Japan and three localities in China (Table 3.1, Fig. 3.1) were successfully PCR-amplified with primers L15926/H597, L15926/HD6, and LD3/H597. Using PCR productdirect sequencing with 16 primers (Fig. 3.2, Table 3.2 except L15926), nucleotide sequences from all sika deer samples were determined. Sequence analysis of control region was operated based on the shortest sequence Yku2. The length of the compared region varied from 705 bases to 824 bases due to insertion or deletion and varied unit number of tandem repeats (see below). We found 30 haplotypes in control region sequences (Table 3.1). Gap sites and the repeated domain in sequence alignment were deleted and then 577 bases were used for further

3

Two Genetically Distinct Lineages of the Japanese Sika Deer

33

pairwise comparisons. As a result, Mya1 and Yku1 were identical with Mya2, Yku2 and Yku4 respectively, because the nucleotide differences between them were within their repeated domains. The nucleotide sequences reported in this paper will appear in the DDBJ/EMBL/ GenBank nucleotide sequence databases with the accession numbers AB378335– AB378377.

Molecular Phylogeny of Sika Deer Control Region The control region phylogenetic tree of the 30 haplotypes was reconstructed by the neighbor-joining method (Fig. 3.3). The phylogenetic tree showed that the Japanese sika deer are separated into two lineages: the Northern Japan group and the Southern Japan group (Fig. 3.3, Table 3.1) with high bootstrap values (958 and 891 on Fig. 3.3). Haplotypes in the Northern Japan group were seen in Hokkaido Island and most of the Honshu mainland, while haplotypes of the Southern Japan group were located in Yamaguchi in the southern most Honshu mainland, Kyushu Island, and islands around Kyushu Island (Fig. 3.4).

Genetic Divergence and Relationships among Sika Deer Groups The number of haplotypes seen in the Northern Japan group, the Southern Japan group, and the China group were 11, 15, and 4, respectively (Table 3.3). Among the Northern Japan group, nucleotide diversity, π, was 10.10 ± 0.92, while it was 13.38 ± 0.44 for the Southern Japan group and 13.26 ± 1.53 for the China group (Table 3.3). We performed the AMOVA analysis to assess genetic differentiation among these lineages. The AMOVA analysis partitioned 33% of the total variation among three groups and 67% of the total within groups. The Fst value for three groups was 0.66 (p < 0.001 in the randomization test).

Tandem Repeats in the Sika Deer Control Region All sika deer examined in this study possessed arrays consisting of repeat units of 37–40 bases, each of which included several substitutions with the other units (see Nagata et al. 1999). The composition of the repeat unit types was different among groups due to substitutions (data not shown). As shown in Table 3.1 and Fig. 3.4, specimens from the Northern Japan group (Hokkaido and the Honshu islands, except Yamaguchi) contained six or seven units, while animals from the Southern Japan group (Yamaguchi, Kyushu, and southern islands) shared four or five units. Specially, five units were only seen in Miyazaki and Kerama Island. All animals from China possessed four units.

34

J. Nagata Hkb

998

Hka

337

Ama1 401

Kmo2

921 885

Kmo1

660 Wda1 Wda2 657 Tga1

672 958

Gyo1

Northern Japan

0.005

Kna2 698

Kna1

683

Yma1

556

753

Tma1 Kra1

419

Kgo1

891

582

Tng1 730

Yku5 Yku4

865 853

Southern Japan

504 Mya1 948 Mya2 786 Mya3 313 Gto1

759 Yku2 736 Yku1 Yku3

501

Ktil

China

CNCI3 CNCI4 CNCI2

914 927

CNCI1 Cervus elaphus xanthopygus

Fig. 3.3 NJ tree based on nucleotide sequences (577 bases) of the mitochondrial control regions in sika deer. Numbers near internal branches indicate the support values from 1,000 bootstrap replicates.

Discussion Classification and Phylogeny of the Japanese Sika Deer Many questions about the evolution of the Japanese sika deer are still open. One interesting question is the taxonomic position of the Tsushima population and the Hokkaido population. In traditional taxonomy of sika deer subspecies, the population on the Honshu mainland and that on the Tsushima Island were both classified as Cervus nippon centralis, and that on the Hokkaido Island was considered as C. n. yesoensis (Ohtaishi 1986). This classification has been widely accepted so far.

3

Two Genetically Distinct Lineages of the Japanese Sika Deer

35

Fig. 3.4 Distribution of control region haplotypes. Numbers indicate number of repeat units.

Table 3.3 Number of haplotypes and nucleotide diversity (π) of nucleotide sequences of the mitochondrial control region in three sika deer lineages, Northern Japan, Southern Japan, and China. Nucleotide diversity No. of individuals No. of haplotypes (π × 1,000) Northern Japan Southern Japan China

26 35 13

11 15 4

10.10 ± 0.92 13.38 ± 0.44 13.26 ± 1.53

However, Imaizumi (1970) classified the Tsushima deer as a distinct species C. pulchellus based on peculiar characters such as strikingly narrow constriction of the jugal bone, deep emargination of the anterior border of the nasal bones, and a longer first segment of the antler, and he considered it a primitive and relic sika deer, compared with other Japanese sika deer populations. In addition, he classified the Hokkaido population in the species of the Asian continent C. hortulorum (Imaizumi 1970). The phylogenetic tree of control region sequences (Fig. 3.3) indicated two genetically distinct groups of sika deer (the Northern and Southern Japan groups) showing high bootstrap values. In the Northern Japan group, Hokkaido C. n. yesoensis and Honshu C. n. centralis (except for individuals from Yamaguchi and Tsushima)

36

J. Nagata

were very closely related to each other, and no obvious genetical distance was found between these subspecies. In this study we also analyzed Asian continental subspecies including C. n. hortulorum (Table 3.1). All individuals from C. n. hortulorum shared the same haplotype CNCI4. Our NJ tree (Fig. 3.3) obviously showed that CNCI4 did not have a close relationship with sika deer on Hokkaido Island, which implies that they are relatively closer to the Southern Japan group than to the Northern Japan group. Intrasubspecific differences between Honshu centralis of the northern group and Yamaguchi centralis or Tsushima centralis were much greater than intersubspecific differences between centralis and yesoensis of the northern group. The Southern Japan group including Yamaguchi and Tsushima centralis was clustered with a relatively high bootstrap value (753) (Fig. 3.3). These results are in agreement with data of mitochondrial cytochrome b sequences (Tamate et al. 1998). Additionally although the Shikoku population has been classified as C. n. nippon (Ohtaishi 1986), recently Yamada et al. (2006) found the two Japanese groups co-existing on the island. Thus, the relationships among these subspecies do not reflect the taxonomy previously reported (Imaizumi 1970; Corbet 1978; Matsumoto et al. 1984; Ohtaishi 1986). Tandemly repeated sequences including 37–40 bp were detected in the sika deer control region (Fig. 3.2). Between the Northern group and the Southern group in Japan, there were large differences in the unit number and the constitution of the first to third units. Moreover, the number of repeat units is specific to each population except for a population in Chiba (Table 3.1). Such a replication slippage could have occurred after isolation of populations on the Japanese islands, and the repeat patterns could have been fixed into each population probably due to genetic drift and/or inbreeding. The tandem repeat domains, therefore, are not concordant with the morphological taxonomy. However, it can be a handy genetic marker for identifying the northern lineage from the southern lineage.

Evolution and History of the Japanese Sika Deer From paleontological evidence, the American bison, Bison bison, was estimated to have diverged from the cow approximately one million years (Myr) before present. Based on the divergence time and the control region data (375 bases) reported by Loftus et al. (1994), the substitution rate in the cow control region was estimated to be 10.6%/Myr. In the sika deer control region sequences, percentage difference of the Northern group/Southern group, China/Northern group, and China/Southern group were 3.7%, 3.2%, and 2.9%, respectively. Considering the above substitution rate and the sequence differences obtained in the present study, the two lineages of Japanese sika deer were estimated to have diverged approximately 0.35 Myr before present. Percentage differences between the Northern group and the Southern group showed similar magnitude to that between the Northern group and the Chinese sika deer as well as that between the Southern group and Chinese sika. The large genetic divergence between the Northern group and the Southern group indicates

3

Two Genetically Distinct Lineages of the Japanese Sika Deer

37

that segregation of mtDNA could have occurred prior to their colonization to Japan in the glacial period during the middle Pleistocene. The divergence time 0.35 Myr for the two groups of Japanese sika deer estimated in this chapter is not in discordance with Kawamura’s conclusion on the first sika deer colonization of Japan (0.4 Myr before present, Kawamura chapter 2). Kawamura (chapter 2) concluded that the most possible sika deer colonization route from the continent to Japan could be through the southern route via land bridges of the Korean Strait but could not be through the northern route via Sakhalin and Hokkaido Islands. As mentioned above, the haplotypes of Hokkaido Island, Hka and Hkb, are most genetically diverged from the haplotype CNCI4 which was detected among C. n. hortulorum, the northern most subspecies in China (Fig. 3.3). Thus, the genetic data also indicated that the northern route was considered to be unlikely. Although paleontological data do not show multiple sika deer colonization to Japan (see Kawamura’s chapter), the current peculiar distribution of the two mitochondrial lineage groups (see Fig. 3.4) implies that colonization from the continent to Japan could have occurred at least twice; one could have brought the ancestors of the Northern group, the other could have brought the ancestors of the Southern group. A probable scenario is that the first colonization brought ancestors of the Northern group (about 0.4 Myr). Judging from habitat preferences of extant Japanese sika deer, their ancestors may have established a population in temperate areas after the first colonization, then expanded northwards to suitable habitats during warmer stages. The first colonized group must have extended up to Sahama in Shizuoka Prefecture on the Honshu mainland at the very least (see Fig. 2.3 of Kawamura’s chapter and Takahashi et al. 2003). The second colonization brought ancestors of the Southern group and led secondly to contact between the two lineages in the Japanese archipelago, then may have pushed northwards the first colonized group, the Northern Japan group. Although there is no information on land bridges, according to Fig. 2.3 in Kawamura’s chapter, it is most likely that the second colonization occurred in MIS 6, about 0.15 Myr, which was one of the remarkably cold stages in the middle Pleistocene. A relatively smaller genetic distance between the Southern Japan group and the China group than one between the Northern Japan group and the China group (Table 3.3) would support that the southern group colonized the Japanese archipelago later than the northern group.

Biogeographic Boundaries, Genetic Boundary, and Biological Characteristics In Japan, biogeographic boundaries often lie between islands. Blakiston’s line, which lies in the Tsugaru Strait between Hokkaido Island and the Honshu mainland (Fig. 3.4), is one clear boundary for distribution of some mammals. For example the Asiatic black bear (Ursus thibetanus), Japanese squirrel (Sciurus lis), serow (Capricornus hircus), and Japanese macaque (Macaca fuscata) all occur on the southern side of Blakiston’s line. The Watase line, which lies in the Tokara Strait

38

J. Nagata

between Yakushima Island and Amami Ohshima Island (Fig. 3.4), forms a boundary between neotropical fauna/flora derived from the late Tertiary and the fauna/ flora originating in the Pleistocene. The deer on Kerama Island (no. 17 in Fig. 3.1) had been introduced from Kyushu (Okinawa Prefecture 1996). Our results are consistent with the Watase line. Part of our results suggests that biogeographic boundary (or zone of overlap) of sika deer lineages is not in agreement with Blakiston’s line. The demarcation occurs much further south, near the south end of the Honshu mainland. Kawamoto (2002) and Kawamoto et al. (2007) reported two mitochondrial lineages, the Eastern Japan lineage (corresponding to our Northern Japan) and the Western Japan lineage (corresponding to our Southern Japan), in Macaca fuscata. Interestingly, this species also has a genetic boundary which nearly coincides with the range of the two major groups of Japanese sika deer They estimated that the Western lineage is older than the Eastern lineage, which is consistent with the two major types of fauna/flora in Japan (Kondo 1982; Kamei et al. 1987; Kawamura et al. 1989). Recently, Hosoi et al. (2005) and Yamada et al. (2006) conducted further study on geographic boundaries between the two groups and found they co-exist in Hiroshima and Shimane prefectures, western parts of the Honshu mainland, and Shikoku Island. Interestingly, there are no boundaries between the two groups, and zones of overlap exist in the western Honshu mainland and Shikoku Island. Although the Hokkaido population (C. n. yesoensis) is morphologically close to the northernmost Chinese population (C. n. hortulorum) and was classified into a single species C. horturolum in the past (Imaizumi 1970), the Hokkaido haplotypes were not genetically closely related to CNCI4 which C. n. hortulorum possessed but were closely related to most of the Honshu haplotypes (Fig. 3.3). Kawamura (chapter 2) reported that the colonization into Hokkaido Island may have occurred from the Honshu mainland through land or ice bridge connections between the Honshu mainland and Hokkaido Island in remarkably cold stages of the late Middle and Late Pleistocene, or by human introduction during the early Holocene. The close genetic distance between the Hokkaido population and the Honshu population in this study supports rather recent colonization from the Honshu mainland to Hokkaido Island. Are there any biological differences in the two lineages of Japanese sika deer? We know the feeding style may differ between the two lineages. Takatsuki (1991) reported the percentage of graminoids in the sika deer diet drastically changes from north to south. The northern sika deer diet is mostly graminoids, while southern sika eat broad-leaved plants. In other words, the northern sika is predominately a grazer, and the southern sika is a browser. Significantly, this diet shift zone is concordant with the genetic boundary. There are some other biological differences between the two lineages. For example, in the northern lineage populations in Hokkaido, Iwate, Tochigi, and Kanagawa, sika deer migrate seasonally between their winter and summer habitats (Miura 1974; Maruyama 1981; Ito and Takatsuki 1987; Uno and Kaji 2000). However, Takatsuki (2000) and Igota et al. (2004) revealed that seasonal migratory animals and residents are mixed in Iwate and Hokkaido. On the other hand, southern lineage

3

Two Genetically Distinct Lineages of the Japanese Sika Deer

39

populations, such as Kyushu, do not show seasonal migration (Endo and Doi 1996; Yabe et al. 2001). In other words, the populations of the northern lineage are more adapted to cold regions with strong seasonality. How did the northern lineage in Japan differentiate to the southern lineage type? Studies on floral changes in Japan based on fossil pollen (Tsukada 1984) revealed that boreal conifer forests covered broad areas of the Japanese Archipelago from the southern Hokkaido Island to the western Honshu mainland and, in fact, reflected the cold climate during the last glacial period (25,000–15,000 years ago). When the second sika colonization happened from the continent to Japan (about 0.15 Myr ago) ancestors of the Southern group would have pushed ancestors of the Northern group northwards. This may have allowed the Northern group to adapt to the following cold climate during the last glacial period resulting in it being phenotypically more like northern sika on the continent. This study demonstrates that genetic distances of mtDNA sequences can be used to estimate genetic diversity in Japanese sika deer populations and to reconstruct evolutionary relationships between the populations. These results could provide a reliable insight for reconsidering subspecies taxonomy, biogeographic boundaries, and history of the sika deer. However, it is still unclear how mtDNA genetic difference affects the biological differences between north and south. To understand the evolution history of sika deer, we need to study relationships between those biological differences and genetic difference using other appropriate genes.

Literature Cited Aquadro, C. F., and B. D. Greenberg. 1983. Human mitochondrial DNA variation and evolution: Analysis of nucleotide sequences from seven individuals. Genetics 103:287–312. Brown, W. M., M. George, Jr., and A. C. Wilson. 1979. Rapid evolution of animal mitochondrial DNA. Proceedings of the National Academy of Sciences USA 76:1967–1971. Brown, W. M., E. M. Prager, A. Wang, and A. C. Wilson. 1982. Mitochondrial sequences of primates: Tempo and mode of evolution. Journal of Molecular Evolution 18:225–239. Cook, C. E., Y. Wang, and G. Sensabaugh. 1999. A mitochondrial control region and cytochrome b phylogeny of sika deer (Cervus nippon) and report of tandem repeats in the control region. Molecular Phylogenetics and Evolution 12:47–56. Corbet, G. B. 1978. The mammals of the Palaearctic Region: A taxonomic view. British Museum, London, United Kingdom. Corbet, G. B., and J. E. Hill. 1991. A world list of mammalian species. Oxford University Press, Oxford, United Kingdom. Endo, A., and T. Doi. 1996. Home range of female sika deer Cervus nippon on Nozaki Island, the Goto Archipelago, Japan. Mammal Study 21:27–35. Endo, H. 1996. Scientific and Japanese names of Artiodactyls of Japan. Honyurui Kagaku (Mammalian Science) 35:203–209. (In Japanese with English abstract.) Excoffier, L., P. E. Smouse, and J. M. Quattro.1992. Analysis of molecular variance inferred from metric distances among DNA haplotypes: Application to human mitochondrial DNA restriction data. Genetics 131:479–491. Felsenstein, J. 1985. Confidence limits on phylogenies: An approach using the bootstrap. Evolution 39:783–791.

40

J. Nagata

Hosoi, E., M. Yamada, H. Tado, and S. Ozawa. 2005. Geographic boundaries of the two distinct lineages of the sika deer, Cervus nippon, in Japan. The 9th international Mammalogical Congress, July-August, 2005, Sapporo, Hokkaido, Japan. Igota, H., M. Sakuragi, H. Uno, K. Kaji, M. Kaneko, R. Akamatsu, and K. Maekawa. 2004. Seasonal migration patterns of female sika deer in eastern Hokkaido, Japan. Ecological Research 19:169–178. Imaizumi, Y. 1970. Description of a new species of Cervus from the Tsushima Island, Japan, with a revision of the subgenus sika based on clinal analysis. Bulletin of the National Science Museum, Tokyo 13:185–196. Ito, T., and S. Takatsuki. 1987. Distribution and migration pattern of sika deer in Mt. Goyo. The Bulletin of Yamagata University 11:411–430. Kamei, S., Z. Kawamura, and H. Taruno. 1987. Mammals. Pages 86–89 in The Quantity Society of Japan, editor, Illustrated map of the quantity in Japan. Tokyo University Press, Tokyo, Japan. (In Japanese.) Kawamoto, Y. 2002. Population genetics on establishment of a species of Macaca fuscata. Asian Paleoprimatology 2:55–73. Kawamoto, Y., T. Shotake, K. Nozawa, S. Kawamoto, K. Tomari, S. Kawai, K. Shirai, Y. Morimitsu, N. Takagi, H. Akaza, H. Fujii, K. Hagihara, K. Aizawa, S. Akachi, T. Oi, and S. Hayashi. 2007. Postglacial population expansion of Japanese macaques (Macaca fuscata) inferred from mitochondrial DNA phylogeography. Primates 48:27–40. Kawamura, Y. 1982. Biogeographical aspects of the Quaternary mammals of Japan. Honyurui Kagaku (Mammalian Science) 43–44:99–130. (In Japanese.) Kawamura, Y. 1991. Quaternary mammalian faunas in the Japanese islands. Quaternary Research 30:213–220. Kawamura, Y., T. Kamei, and H. Taruno. 1989. Middle and Late Pleistocene mammalian faunas in Japan. Quaternary Research 28:317–326. (In Japanese with English abstract.) Kimura, M. 1980. A simple method for estimating evolutionary rates of base substitutions through comparative studies of nucleotide sequences. Journal of Molecular Evolution 16:111–120. Kondo, N. 1982. Mammal fauna of Japan, with special reference to the ecological property of species, paleo-environment and the Tsugaru Strait. Honyurui Kagaku (Mammalian Science) 43–44:131–144. (In Japanese.) Loftus, R. T., D. E. MacHugh, D. G. Bradley, P. M. Sharp, and P. Cunningham. 1994. Evidence for two independent domestications of cattle. Proceedings of the National Academy of Sciences USA 91:2757–2761. Maruyama, N. 1981. A study about seasonal migration and gathering of sika deer, Cervus nippon TEMMINCK. Report of the Department of Agriculture, Tokyo University of Agriculture and Technology 23:85. (In Japanese with English summary.) Matsumoto, M., H. Nishinakagawa, and J. Otsuka. 1984. Morphometrical study on the skull of Cervus pulchellus, Cervus nippon mageshimae and Cervus nippon yakushimae. Journal of the Mammalogical Society of Japan 10:41–53. (In Japanese with English abstract.) Miura, S. 1974. Seasonal changes of sika deer habitats in Hinoebora, Tanzawa, Japan. Honyu Dobutugaku Zassi 6:51–66 (In Japanese). Nabata, D., R. Masuda, O. Takahashi, and J. Nagata. 2004. Bottleneck effects on the sika deer Cervus nippon population in Hokkaido, revealed by ancient DNA analysis. Zoological Science 21:473–481. Nagata, J., R. Masuda, and M. C. Yoshida. 1995. Nucleotide sequences of the cytochrome b and 12S rRNA genes in the Japanese sika deer Cervus nippon. Journal of the Mammalogical Society of Japan 20:1–8. Nagata, J., R. Masuda, K. Kaji, M. Kaneko, and M. C. Yoshida. 1998. Genetic variation and population structure of Japanese sika deer (Cervus nippon) in Hokkaido Island, based on mitochondrial D-loop sequences. Molecular Ecology 7:871–877. Nagata, J., R. Masuda, H. B. Tamate, S. Hamasaki, K. Ochiai, M. Asada, S. Tatsuzawa, K. Suda, H. Tado, and M. C. Yoshida. 1999. Two genetically distinct lineages of the sika deer, Cervus nippon,

3

Two Genetically Distinct Lineages of the Japanese Sika Deer

41

in Japanese islands: Comparison of mitochondrial D-loop region sequences. Molecular Phylogenetics and Evolution 13:511–519. Nei, M. 1987. Molecular evolutionary genetics. Columbia University Press, New York, New York, USA. Ohtaishi, N. 1986. Preliminary memorandum of classification, distribution and geographic variation on sika deer. Honyurui Kagaku (Mammalian Science) 53:13–17. (In Japanese.) Ohtaishi, N., and Y. T. Gao. 1990. A review of the distribution of all species of deer (Tragulidae, Moschidae and Cervidae) in China. Mammal Review 20:125–144. Okinawa Prefecture. 1996. A report of conservation and management of sika deer in Kerama Island. Board of Education, Okinawa Prefecture, Okinawa, Japan. (In Japanese.) Saitou, N., and M. Nei. 1987. The neighbor-joining method: A new method for reconstructing phylogenetic trees. Molecular Biology and Evolution 4:406–425. Takahashi, K., H. Matsuoka, H. Taru, K. Yasui, and Y. Hasegawa. 2003. Vertebrate fossils from the excavation site of the Naumann’s elephant (Palaeoloxodon naumanni) at Sahama. Shizuoka Chigaku 87:15–21. (In Japanese.) Takatsuki, S. 1991. Feeding ecology of ungulates with reference to cervids. Pages 119–144 in Asahi M. and Kawamichi K., editors, Modern Mammalogy. Asakura Publishing, Tokyo, Japan. (In Japanese.) Takatsuki, S. 2000 Seasonal elevational movements of sika deer on Mt. Goyo, northern Japan. Mammal Study 25:107–114. Tamate, H. B., and T. Tsuchiya. 1995. Mitochondrial DNA polymorphism in subspecies of the Japanese sika deer, Cervus nippon. Journal of Heredity 86:211–215. Tamate, H. B., S. Tatsuzawa, K. Suda, M. Izawa, T. Doi, K. Sunagawa, F. Miyahira, and H.Tado. 1998. Mitochondrial DNA variations in local populations of the Japanese sika deer, Cervus nippon. Journal of Mammalogy 78:1396–1403. Thompson, J. D., T. J. Gibson, F. Plewniak, F. Jeanmougin, and D. G. Higgins. 1997. The ClustalX windows interface: Flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic Acids Research 24:4876–4882. Tsukada, M. 1984. A vegetation map in the Japanese archipelago approximately 20,000 years B. P. Japanese Journal of Ecology 34:203–208. (In Japanese with English abstract.) Uno, H., and K. Kaji. 2000. Seasonal movements of female sika deer in eastern Hokkaido, Japan. Mammal Study 25:49–57. Yabe, T., T. Koizumi, A. Endo, S. Seki, and Y. Miura. 2001. Home range of sika deer in the central mountains in Kyushu. Kyushu Journal of Forest Research 54:131–132. (In Japanese.) Yamada, M., E. Hosoi, H. B. Tamate, J. Nagata, S. Tatsuzawa, H. Tado, and S. Ozawa. 2006. Distribution of two distinct lineages of sika deer (Cervus nippon) on Shikoku Island revealed by mitochondrial DNA analysis. Mammal Study 31:23–28.

Chapter 4

Evolutionary Significance of Admixture and Fragmentation of Sika Deer Populations in Japan Hidetoshi B. Tamate

Abstract The sika deer is genetically very close to the red deer (Cervus elaphus): nucleotide divergences between the two species are less than 1% in most of the protein-coding sequences. Genetic markers that were developed for red deer and other cervine, ovine, and bovine species are readily applicable to genetic analyses of sika deer. By using such DNA markers, I and my colleagues studied the level of genetic diversity of local populations, past demographic changes of populations, and spatial structures of populations. A phylogenetic tree constructed from microsatellite allele frequencies separates “northern” and “southern” groups in different clusters, showing a similar split pattern between the two lineages as observed in the mtDNA-based phylogenetic tree. However, the level of differentiation between the northern and southern mtDNA groups is lower than that among the populations in Honshu, which suggest that the genetic difference between the two groups has been reduced by the admixture. The genetic differentiation among populations was caused by a loss of genetic variation during past fragmentation of populations and not by the accumulation of novel mutations in each population. Two cases of the fragmentation of sika deer populations at a local scale, which were revealed by DNA analyses, are described in this chapter. Population genetic studies thus provide empirical data for monitoring and predicting long-term changes in demography and population structure of sika deer.

Introduction Recent molecular phylogenetic studies have shown that the Japanese sika deer is genetically subdivided into two major mitochondrial DNA (mtDNA) lineages, the northern and southern haplotype groups, and that a split between the two had occurred in the mid-Pleistocene (Nagata et al. 1999; Nagata chapter 3). Despite the deep divergence time, sympatric distribution of the two groups has been observed in eastern Shikoku Island (Yamada et al. 2006), suggesting secondary contact between genetically differentiated populations in the Japanese Archipelago. Such encounters between populations often lead to the formation of a new population by a genetic D. R. McCullough et al. (eds.), Sika Deer: Biology and Management of Native and Introduced Populations, DOI: 10.1007/978-4-431-09429-6_4, © Springer 2009

43

44

H. B. Tamate

process known as “admixture” if they are not separated reproductively. Admixture sometimes causes the genetic traits of the resultant population to change because more advantageous alleles are likely to be selected during a merger of populations. Since there is no geographical boundary between the two groups of sika deer on Honshu Island, it is likely that admixture caused the genetic variations that might have been selected independently in each lineage to intermingle, and hence it caused the morphological and physiological differences—if they exist—between the northern and southern deer to diminish. As the admixture proceeds, it will reinforce the adaptability of deer in the merged population in accordance with their new environment. Genetic studies of the population structure, therefore, are important to help elucidate the nature and evolutionary significance of the sika deer in different environments in Japan, from subtropical to cold-temperate zones. In this chapter, I will first summarize our current knowledge of the basic genetics of the Japanese sika deer. I then refer to the population history of the sika deer and discuss how local populations have been established in the Japanese Archipelago. Finally, I will return to an old question—the taxonomy and species concept of the sika deer—to which recent genetic studies have provided some clues.

Basic Genetic Characteristics of the Sika Deer Genes To date, about seven hundred nucleotide sequences of the sika deer have been deposited in DNA databases such as GenBank, EMBL Nucleotide Sequence Database (EMBL) and DNA Databank of Japan (DDBJ). Those sequences consist of nuclear DNA, mitochondrial DNA, and complementary DNA (Table 4.1). Homology analysis of the deposited nucleotide sequences has shown that the sika deer is genetically very close to the red deer/wapiti (Cervus elaphus) (wapiti are the somewhat different lineage of Cervus elaphus distributed in East Asia and North America); nucleotide divergences between the sika and red deer are less than 1% in most of the protein-coding sequences (Table 4.2). Our recent study also demonstrated that sequences at the second exon of the major histocompatibility complex (MHC) DRB locus in the sika deer are not separated phylogenetically from those of the red deer (Fig. 4.1), suggesting that some genetic variations are shared by the two species. Therefore, genetic markers that are developed for the red deer are readily applicable to studies of sika deer. Also, bovine and ovine microsatellites are often utilized in genetic analyses of the sika deer as well as other deer species because they are conserved well among Artiodactyla (Slate et al. 1998). A large number of such genetic markers can be found in the linkage map of the red deer (Slate et al. 2002). Those markers allowed us to study the level of genetic diversity of a single population (Tamate et al. 1998),

4

Evolutionary Significance of Admixture and Fragmentation

45

Table 4.1 Number of nucleotide sequences deposited in GenBank in June, 2006. Category Sika deer Red deer Mitochondrial DNA Protein coding rRNA or tRNA coding D-loop Nuclear DNA Protein coding without MHC MHC Microsatellite and/or non-coding cDNA and EST Patented sequencea Total a Features are not described in detail.

148 43 12 93 378 25 330 23 23 0 549

162 51 17 94 220 62 97 61 72 6 460

Fig. 4.1 Neighbor-joining tree based on MHC-DRB sequences (210 bps) among sika deer (CeniDRB) and red deer (CeelDRB). Branching patterns are shown if the bootstrap value at a node is more than 50%. Branch lengths do not reflect genetic distances. Red deer sequences (CeelDRB.13, 21, 26, 37, 46, and CeelDRQ.03 which was used as an outgroup) were retrieved from Genbank. Sika deer sequences (CeniDRB.clone 1, 2, 3, 5, and 9) were determined in the present study.

past demographic changes of populations (Goodman et al. 2001), pedigree structure (Okada and Tamate 2000), rate of gene flow among populations (Goodman 1999), and genetic relatedness between individuals (Okada et al. 2005). Moreover they provide empirical data to help determine management units (Yuasa et al. 2007). These topics of study will be discussed later in this chapter.

46

H. B. Tamate

Despite the increasing number of DNA and protein sequence data in the database, genes that determine the distinctness of the deer are yet to be identified. Morphological characteristics of the sika deer such as body size, the shape of the antlers, and pelage pattern are supposed to be under the control of a group of genes known as the quantitative trait loci (QTL). To date, a large panel of microsatellite markers has been used to map the QTL in the red deer, and this map has begun to show the chromosomal regions that are responsible for various quantitative traits (Slate et al. 2002). It is expected that in future research the genes responsible for the morphological and physiological characteristics of the sika deer will be identified by means of a genome-wide analysis of the red deer and related species.

Chromosomes An odd number of chromosomes in a diploid organism is often a consequence of the hybridization between closely related species that differ in chromosome number. Such a phenomenon has been shown to occur in the Japanese sika deer. The number of chromosomes in the Japanese sika deer varies from 65 to 68 (Ohmura et al. 1983). The mode of the chromosome number in northern sika deer, C. n. yesoensis from Hokkaido, is 68 with a pair of metacentric chromosomes (Ohmura et al. 1983; Yokohama et al. 1994), which is supposed to be the basic chromosome number in Cervidae (Slate et al. 2002). In contrast, C. n. nippon from the Chugoku region, a representative of the southern sika deer, possesses 66 chromosomes including two pairs of metacentric chromosomes (Ohmura et al. 1983), implying that a metacentric fusion of chromosomes took place during the differentiation of the southern sika deer. Because an odd number of chromosomes, 67, was found among deer in Honshu and Hokkaido, Ohmura et al. (1982) suggested that a past hybridization had

Table 4.2 Number of nucleotide substitutions between the sika and red deer in protein-coding genes. Number of Maximum number nucleotides of substitution Protein-coding genes compared (bps) sites (bps) Reference Amelogenin Y Growth hormone precursor Kappa-casein precursor Prion protein

101 273 371 706

0 1 2 3

Sex determining factor (SRY)

166

1

Yamauchi et al. (2000) Chikuni et al. (1994) Cronin et al. (1996) Jeong et al. (accession number DQ234266) Takahashi et al. (1998)

4

Evolutionary Significance of Admixture and Fragmentation

47

occurred between two different karyotype groups of the Japanese sika deer. He then hypothesized that the Japanese sika deer have evolved from those deer that migrated from the south at one period and from the north at another period. After 16 years, his hypothesis gained further support from molecular studies (Tamate et al. 1998; Nagata et al. 1999). Chromosomes are still important markers in the study of the evolution of the sika deer because they are used to detect genome-wide changes such as recombination and translocation, which cannot be easily analyzed by mtDNA and microsatellite markers.

Population History of the Japanese Sika Deer Revealed by Genetic Studies Admixture in the Past MtDNA-based phylogeography of the sika deer indicated clearly that the two genetically differentiated populations—the northern and southern haplotype groups—came in contact with each other on the Japanese Archipelago. The question then arises as to what extent the two populations have been mixed after their encounter. Recently, Hosoi and his colleagues conducted fine-scale analysis of the mtDNA phylogeography over the contact zone of the two haplotype groups in eastern Shikoku and found that southern and northern haplotypes are distributed in a mosaic-like pattern within a hundred-kilometer range (Yamada et al. 2006, 2007). Considering that the rate of nuclear gene flow is generally higher than the rate of MtDNA introgression in animals with female-biased philopatry, it is reasonable to assume that genetic admixture beyond the contact zone is in progress. If the admixture proceeds, genetic differences between the northern and southern groups will be reduced unless disruptive selection, in which individuals with extreme phenotypes have a higher fitness than those with mean values, takes place (Page and Holmes 1998). To determine the level of genetic differentiation among the Japanese sika deer on a regional scale, I and my colleagues conducted population-genetic analyses of samples collected from three main regions, namely, Hokkaido, Honshu, and Kyushu (Fig. 4.2), by using nine microsatellite loci (Goodman et al. 2001). In contrast to mtDNA markers that represent matrilineal lineage only, microsatellite markers on autosomal chromosomes correspond to gene flow mediated by both sexes and are therefore suitable for studying population history under an assumption of the neutral model of molecular evolution. The neutral model is based on a theory that the most of genetic variations within a species are selected by chance (Page and Holmes 1998), which gives a theoretical framework for population-genetic analyses. The microsatellite data demonstrated that some alleles are found exclusively in the southern mtDNA haplotype group; allele 169 at the BMC4107 locus and allele 123 at the OARfcb193 locus, for instance, are only observed among populations from southern Honshu and Kyushu. A phylogenetic tree constructed from microsatellite allele frequencies separates “northern” and “southern” groups in different

48

H. B. Tamate

Fig. 4.2 Distribution of sika deer populations discussed in the present study. Dark-shaded areas show current distribution of sika deer in Japan except the Ryukyu Islands. Localities are indicated by numbers: (1) NW Hokkaido, (2) Akan, (3) Daisetsu, (4) Hidaka, (5) Iwate, (6) Kinkazan Island, (7) Shizuoka, (8) Nara, (9) Shoudoshima Island, (10) Hyogo, (11) Shimane, (12) Yamaguchi, (13) Tsushima Island, (14) Nagasaki, (15) Miyazaki, (16) Tanegashima Island, and (17) Yakushima Island. Populations that belong to the northern and southern mtDNA lineages are indicated by open circles and closed circles, respectively.

clusters, showing a similar split pattern between the two lineages as observed in the mtDNA-based phylogenetic tree (Fig. 4.3). Genetic analysis of population differentiation revealed, however, that the level of differentiation between the northern and southern groups is lower than that among the populations within the region of Honshu (Table 4.3). Genetic differentiation among groups of populations was quantified by calculating θ and ρ, estimators of population differentiation based on Fst (Weir and Cockerham 1984) and Rst (Slatkin 1995), respectively. There is a weak but significant correlation between geographical distance and the magnitude of genetic differentiation (Goodman et al. 2001). This represents a pattern of “isolation by distance” which suggests gene flow over whole populations. We concluded, therefore, that the genetic difference between the northern and southern mtDNA groups in the nuclear genome has been reduced by the admixture, even though the groups are subdivided phylogenetically.

4

Evolutionary Significance of Admixture and Fragmentation

49

Fig. 4.3 Phylogenetic tree based on Cavalli-Sforza Edwards distance calculated from microsatellite data. Populations of sika deer in Scotland (Fife and Peebles) and in England (Dorset) and a population of red deer in Scotland (Argyll) are included in the analysis. Branch lengths reflect genetic distance according to scale, and numbers at nodes show bootstrap values over 50% (from Goodman et al. 2001).

Bottlenecking in the Past According to our microsatellite data, two measures of genetic diversity of a population, average expected heterozygosity and allelic richness (average number of alleles per locus), ranged from 0.10 to 0.65, and from 1.5 to 5.1, respectively,

50

H. B. Tamate

Table 4.3 Population differentiation among sika deer in hierarchical groupings (from Goodman et al. 2001). Populations θ ρ All 0.429 0.577 (0.565–0.613) Japan only 0.415 0.483 (0.475–0.519) Hokkaido-North Honshu-South 0.343 0.323 (0.295–0.375) Honshu-Tsushima-Kyushu-Kyushu Islands Hokkaido-Honshu-Kyushu 0.219 0.245 (0.229–0.275) Northern mtDNA haplotypes-Southern 0.166 0.190 (0.161–0.229) mtDNA haplotypes 0.035 (0.018–0.131)** Hokkaido (Akan-Daisetsu-Hidaka-NW Hokkaido) 0.009* Honshu (Iwate-Kinkazan-Hyogo-Shimane0.296 0.321 (0.304–0.370) Yamaguchi-Tsushima) Kyushu (Nagasaki-Miyazaki0.552 0.543 (0.514–0.605) Tanegashima-Yakushima) UK introduced sika 0.471 0.447 (0.412–0.583) Values in parentheses for ρ are 95% confidence intervals. *P = 0.200, **P = 0.24, all other comparisons P < 0.001, permutation test.

among populations (Fig. 4.4). The genetic diversity is mainly determined by the balance between genetic drift and gene flow. Genetic diversity declines rapidly due to the effect of genetic drift if a population has a small effective population size (Ne) or has experienced a sudden decrease in the size of a population, often called “bottlenecking,” in the past. Gene flow among populations, on the other hand, counteracts the genetic drift and decelerates the reduction of the diversity. Regression analysis showed that the diversity measure positively correlates with the habitat fragment size (Goodman et al. 2001), indicating that current population size is a major factor that determines the level of genetic diversity. There are, however, some cases in which current population size and diversity do not match. Hokkaido populations have a remarkably low level of diversity even though they occupy the largest habitat area among the populations in Japan (Goodman et al. 2001). This can be explained as a consequence of bottlenecking that occurred in the past. Molecular phylogeographic studies revealed that the current Hokkaido populations have recovered from three remnant populations that survived severe bottlenecking in the last century (Nagata et al. 1998, 2004). The microsatellite analysis provides another piece of evidence for the recent bottlenecking in Hokkaido and presents empirical data showing that genetic diversity does not keep up with rapid growth of populations over 100 or so years. In contrast to the Hokkaido populations, the genetic diversity is unexpectedly high in a small population on Kinkazan Island (for a description of the island see Takatsuki and Padmalal chapter 8). The census population size (Nc) of deer on this island, which has been recorded since the 1970s, never exceeded about 550 individuals. An estimate of the effective population size (Ne), on the other hand, has been calculated to be 220 based on the proportion of reproductive individuals (Tamate et al. 2000), and 256 based on the allele frequency data (Goodman et al. 2001). It is expected theoretically that the genetic diversity declines rapidly with such a small

4

Evolutionary Significance of Admixture and Fragmentation

51

Fig. 4.4 Genetic diversity and habitat sizes of sika deer populations in Honshu and Kyushu. Genetic diversities were measured as heterozygosity and an average number of alleles per locus (Allelic Richness). Habitat sizes are shown in square kilometers. Numbers in open circles correspond to sampled populations as shown in Fig. 4.2.

population size. Therefore, it remains an unsolved question why the Kinkazan population retains a relatively high level of diversity despite its small population size. The highest genetic diversity is observed in the Nara population (for a description of this population see Torii and Tatsuzawa chapter 25). Historically, this population has been protected for religious reasons since the Middle Ages, which leads to an assumption that the population has been free from bottlenecking caused by overexploitation. By using microsatellite markers, we carried out two different tests, M ratio (Garza and Williamson 2001) and heterozygosity excess test (Cornuet and Luikart 1996), to determine whether the Nara population has experienced bottlenecking in the past. The results showed no sign of bottlenecking. It is likely, therefore, that the Nara population is most likely to represent the genetic variability of the ancestral populations of sika deer on the Honshu mainland. Goodman et al. (2001) carried out further microsatellite analysis using a coalescent-based genetic approach to test whether the observed diversity data are best explained by either of two alternative demographic hypotheses, “gene flow and drift” or “drift only” models. The result suggested that populations in Kyushu have been under “drift and gene flow” equilibrium—a balanced condition in which genetic variations will be lost due to genetic drift but will be supplemented by gene flow from other populations. It is therefore likely that the sika deer in Kyushu consist

52

H. B. Tamate

of a metapopulation: a group of partially isolated populations among which exchange of individuals is quite infrequent. The population history that led to such a metapopulation was described for tule elk, and the effect of demographic changes on retention of genetic diversity was discussed in McCullough et al. (1996). In contrast, the genetic differences among the Honshu populations are explained by the “drift-only” model, which implies that the populations have been isolated from each other. It is concluded, therefore, that genetic differentiation among current Japanese populations on a regional scale was facilitated not by a long-term evolutionary process but by a short-term stochastic change in the allelic frequency of individual populations.

Fragmentation at Present To evaluate the impact of habitat fragmentation over populations within a region, Yuasa et al. (2007) conducted local-scale genetic analysis of the population structure of the sika deer in southern Kanto, central Japan. The distribution of the deer in this region is discontinuous due to the development of large urban areas and road networks. Genetic diversity indices have suggested that seven local populations in southern Kanto probably experienced population bottlenecking in the recent past. Nested clade analysis of the mtDNA control region variations among local populations indicated restricted gene flow within the region. The sika deer in the southern Kanto region, therefore, are fragmented not only geographically, but genetically into several local populations. These populations are likely to function as a metapopulation, as suggested by the genetic and ecological studies (Yuasa et al. 2007). Another case of fragmented populations was observed on Shoudoshima Island. The population on this island is subdivided into two subpopulations (the northern Shoudo and southern Santo peninsula populations) only about a 10 km distance from each other, separated by a residential area. Microsatellite analysis demonstrated that the two local populations are genetically differentiated (Fig. 4.5). An assignment test indicated that none of the individuals were likely to immigrate between the two populations (data not shown). These results showed clearly that population fragmentation within a small area can be caused by human activities, even for highly mobile animals like the sika deer. In summary, population-genetic studies revealed significant genetic differentiation among populations of the Japanese sika deer. Such genetic differentiation, however, is mainly caused by a loss of genetic variation during population fragmentation, and not by the accumulation of novel mutations in each population. It can be said, therefore, that populations of the Japanese sika constitute a single evolutionary significant unit so far as neutral loci are concerned. Nevertheless, it remains unknown whether the Japanese sika deer populations differ locally in genetic variations that correspond to their adaptability to different habitats.

4

Evolutionary Significance of Admixture and Fragmentation

53

Fig. 4.5 Phylogenetic tree showing genetic relationships among individuals sampled from Shoudoshima population. The tree was constructed by the neighbor-joining method based on Cavalli-Sforza Edwards distance calculated from microsatellite data. Deer in northern Shoudo population (open circles) and southern Santo peninsula population (closed circles) are clustered into separate clades according to the localities except for one individual on the Santo peninsula.

Taxonomy and Species Concept of the Sika Deer Reexamined by Recent Genetic Studies Taxonomy The sika deer was first described as the species “Cervus nippon” by Temminck in 1838, who examined a specimen that was collected in Kyushu. In the classification system in use at the time, the definition of a species was based solely on the morphological species concept; if a group of animals has remarkable morphological characteristics that can be easily distinguished from other groups, it should be classified as a separate taxon. To the eyes of researchers who came from Europe in the nineteenth century, the sika deer was no doubt a distinct species because of its unique pelage pattern, antler shape, and small body size. Even at that time, however, a question was posed regarding the taxonomic status of the sika deer in Hokkaido. Blakiston (1883), a British zoologist who studied mammalian fauna in Japan, wrote that the deer in Hokkaido and those on the mainland were identical. He also noted, however, that Reverend Père Heude, who had examined a head of deer from

54

H. B. Tamate

Hokkaido, informed him that the specimen was rather like C. manchuricus minor or an unknown species. Some taxonomist later classified the sika deer in Hokkaido and northern China as a separate taxon, C. horturolum (Imaizumi 1970). Another debate over the taxonomy of the sika deer has continued since Imaizumi (1970) classified the sika deer from the Tsushima Islands into a separate species, C. pulchellus. He proposed this novel taxon because several measures of the cranial morphology of the deer did not fit into the “cline” of the morphological variations among C. nippon and C. hortulorum. Matsumoto et al. (1984), however, later pointed out that the morphological differences between the deer from the Tsushima Islands and C. nippon are not readily distinguishable. Even morphological variations among local races of sika deer are evident (Kuroda and Okada 1950), a range of morphological variations that do not overlap with those of the red deer/wapiti (Cervus elaphus), a closely related species that is sympatric with sika deer in northeastern China and Far East Russia (see Aramilev chapter 33). For this reason, it is widely accepted at present that in morphology-based taxonomy the sika deer is a single species, and all local races of the sika deer are classified into subspecies according to their natural distribution (Ohtaishi and Gao 1990). Recent molecular phylogenetic studies have provided further support for the classification; both the deer in Hokkaido (C. horturolum sensu Imaizumi) and the deer in the Tsuhima Islands (C. pulchellus sensu Imaizumi 1970) remain within a clade of C. nippon in molecular phylogenetic trees. It has also been shown that nucleotide sequence divergence between C. pulchellus sensu Imaizumi 1970 and C. nippon stays within the level observed within a species (Kuwayama and Ozawa 2000; Li et al. 2003). It has been concluded, therefore, that C. nippon Temminck 1883 is the only valid taxon for the Japanese sika deer at the species level. At the subspecies level, as discussed in a earlier chapter (Nagata chapter 3), molecular studies have demonstrated that the morphology-based taxonomy does not reflect the phylogeny. For example, three subspecies, C. n. keramae, C. n. nippon and C. n. yakushimae, are indistinguishable in the mtDNA cytochrome-b lineage, as they share a common haplotype (Fig. 4.6). The subspecies of the sika deer do not necessarily correspond to evolutionarily significant units within the species; allopatry is the only reliable criterion for defining the subspecies of the sika deer.

Species Concept Revisited Since the 1980s, biochemical and genetic studies have explored “species differences” within the genus Cervus. Based on the biochemical analysis of serum proteins, Harrington (1985) suggested that the sika deer is serologically similar to Asian wapiti and suggested that the sika deer is a very close relative of the red deer. His idea was later confirmed by DNA-based studies (Kuwayama and Ozawa 2000; Pitra et al. 2004). A close relationship between the two species has also been demonstrated by geneticists who studied hybridization between the sika deer and red deer where sika have been introduced in Europe, which in both captive and natural populations often

4

Evolutionary Significance of Admixture and Fragmentation

55

Fig. 4.6 Parsimonious cladogram based on mtDNA cytochrome-b gene sequences (429 bps) among subspecies in Honshu and Kyushu. Large circles indicate the observed haplotypes and small closed circles indicate hypothetical haplotypes inferred from parsimonious analysis. A common haplotype among three subspecies, C. n. keramae, C. n. nippon, and C. n. yakushimae, is shown by a larger circle in the center.

leave fertile offspring (Lowe and Gardiner 1975; Abernethy 1994; Goodman et al. 1999; Swanson and Putman chapter 40). The lack of reproductive isolation was also reported among species of Cervinae, such as in hybrids between red and Père David’s deer (Tate et al. 1997) and between sika and axis deer (Asher et al. 1999). Therefore, sika deer do not match up simply with the criteria of the biological species concept in which “species” is defined as a reproductively isolated group. Although the sika deer and other species are not genetically differentiated enough to establish reproductive isolation, they are separated clearly in molecular phylogeny except for MHC genes. All phylogenetic trees constructed from mtDNA data agree that the sika deer and wapiti form a monophyletic group to which European red deer, i.e., C. elaphus sensu stricto, become a sister group (Polziehn and Strobek 1998; Cook et al. 1999; Kuwayama and Ozawa 2000; Li et al. 2003; Pitra et al. 2004). Fossil records indicate that both C. nippon and C. elaphus appeared in China in the late Pleistocene (Dong 1993), while European red deer first appeared in the middle Pleistocene (Lister 1984). Among the molecular studies, however, the estimated divergence time for the sika + wapiti/European red deer split varies from the late Pliocene to the Pleistocene among the studies (Table 4.4). It is likely that in some molecular studies the rate of nucleotide substitution was underestimated because calibration points were taken at a deep divergence time. Alternatively, the longer estimated time for the split among the species can be explained if the split in the mtDNA lineage occurred much earlier than the onset of morphological differentiation among the species. Within the sika-wapiti lineage, all races of the sika deer constitute a monophyletic group in both the mitochondrial cytochrome-b gene (Cook et al. 1999) and the control region sequences. The divergence time between the sika and wapiti is as deep as those observed between species of Cervidae (Li et al. 2003; Pitra et al. 2004; Gilbert et al. 2006), providing further support for classifying the sika deer and wapiti into separate species.

56

H. B. Tamate

Table 4.4 Estimated divergence time between the sika and related species and among the major clades of the sika deer. Calibration points Estimated rate Estimated time for the molecular of nucleotide Split Literature Gene of the split clock substitution Wapiti/Red Polziehn and Strobek (1998) Mahmut et al. (2002)

D-loop

Pleistocene Red/fallow split (0.5–0.6 Mya) 1.6 Mya

D-loop

Late Pleistocene (0.3–0.4 Ma)

Kuwayama and cytchrome b Pleistocene Ozawa (2000) (0.8 Ma) Li et al. (2003) cytchrome b Middle Pliocene (3.9 Mya)a Pitra et al. (2004) cytchrome b Middle Plioceneb

Sika/Wapiti Kuwayama and cytchrome b Pleistocene Ozawa (2000) (0.57 Ma) Li et al. (2003) cytchrome b Late Pliocene (2.77 Ma)b Pitra et al. (2004) cytchrome b Late Pliocenea

Nd

Red/fallow split 1.6 Ma Muntianinae/ Cervinae split 7.0 Ma Muntianinae/ Cervinae split 7.0 Ma, The oldest known New World Odocoileinae 5.0 Ma Red/fallow split 1.6 Ma Muntianinae/ Cervinae split 7.0 Ma Muntianinae/ Cervinae split 7.0 Ma, The oldest known New World Odocoileinae 5.0 Ma

Based on human D-loop data (Stoneking et al. 1992) Based on human D-loop data (Stoneking et al. 1992) 0.035 substitutions/ site/Mya 0.016 substitutions/ site/Mya 0.0257 substitutions/ site/Mya

0.035 substitutions/ site/Mya 0.016 substitutions/ site/Mya 0.0257 substitutions/ site/Mya

Northern/Southern Sika Nagata et al. D-loop (1999)

Late Pleistocene (0.3 Ma)

Northern/southern, Based on bovine northern/China, D-loop data southern/China (Loftus et al. were 3.7, 3.2 1994) and 2.9%.

Among Chinese Sika Wu et al. (2004) D-loop

Not determined

Cluster I and II 2.9%

a

Recalculated from data in Table 4.2 from Li et al. (2002). Described in Fig. 4.3 from Pitra et al. (2004).

b

Based on bovine D-loop data (Loftus et al. 1994)

4

Evolutionary Significance of Admixture and Fragmentation

57

Conclusion Genetic studies have unveiled the unique evolutionary history of the sika deer: speciation, admixture, and fragmentation. Ancestors of the Japanese sika deer split into at least two lineages as vicariants in the Asian mainland. They would have taken a different course of adaptation, but they met again in the Japanese Archipelago. Since then admixture has been reducing the genetic differences between the two lineages, the northern and southern deer. Before completing panmixia, however, the admixing process was interrupted by population fragmentation, possibly caused by anthropologic activities, which resulted in a loss of the genetic diversity of the sika deer. Nevertheless, a relatively large number of genetic variations remain in some populations that have not suffered from extensive overexploitation in the past. The genetic study by Yuasa et al. (2007) suggested that fragmented populations of the sika deer are likely to form a metapopulation. One of the important issues to be studied in future genetic research is the stability of local populations under the scheme of a metapopulation; strong hunting pressure due to culling conducted by local governments may alter the dispersal pattern of deer herds and, hence, modify the pattern of gene flow among subpopulations. Further genetic research on the Japanese sika deer will provide information not only about the past of the species— the phylogeny and phylogeography—but also about the status of populations in the present and future. Population-genetic analyses will provide empirical data that will be useful for monitoring and predicting long-term changes in demography and population structure.

Literature Cited Abernethy, K. 1994. The establishment of a hybrid zone between red and sika deer (genus Cervus). Molecular Ecology 3:551–562. Asher, G. W., D. S. Gallagher, M. L. Tate, and C. Tedford. 1999. Hybridization between sika deer (Cervus nippon) and axis deer (Axis axis). Journal of Heredity 90:236–240. Blakiston, T. W. 1883. Zoological indications of ancient connection of the Japan islands with the continent. Transactions of the Asiatic Society of Japan 11:126–140. Chikuni, K., T. Tabata, M. Monma, and M. Saito. 1994. Direct sequencing of the promoter region of growth hormone gene from Artiodactyla. Journal of Animal Science and Technology 65:120–124. Cook, C. E., Y. Wang, and G. Sensabaugh. 1999. A mitochondrial control region and cytochrome b phylogeny of sika deer (Cervus nippon) and report of tandem repeats in the control region. Molecular Phylogenetics and Evolution 12:47–56. Cornuet, J.-M., and G. Luikart. 1996. Description and power analysis of two tests for detecting recent population bottlenecks from allele frequency data. Genetics 144:2001–2014. Cronin, M. A., R. Stuart, B. J. Pierson, and J. C. Patton. 1996. K-casein gene phylogeny of higher ruminants (Pecora, Artiodactyla). Molecular Phylogenetics and Evolution 6:295–311. Dong, W. 1993. A morphological analysis of cheek teeth of Eurasian Pliocene cervids. Pages 65–72 in N. Ohtaishi, and H. L. Sheng, editors, Deer of China. Elsevier Science, Amsterdam, the Netherlands. Garza, J. C., and E. G. Williamson. 2001. Detection of reduction in population size using data from microsatellite loci. Molecular Ecology 10:305–318.

58

H. B. Tamate

Gilbert, C., A. Ropiquet, and A. Hassanin. 2006. Mitochondrial and nuclear phylogenies of Cervidae (Mammalia, Ruminantia): Systematics, morphology, and biogeography. Molecular Phylogenetics and Evolution 40: 101–117. Goodman, S. J., N. H. Barton, G. Swanson, K. Abernethy, and J. M. Pemberton. 1999. Introgression through rare hybridization: A genetic study of a hybrid zone between red and sika deer (genus Cervus) in Argyll, Scotland. Genetics 152:355–371. Goodman, S. J, H. B. Tamate, R. Wilson, J. Nagata, S. Tatsuzawa, G. M. Swanson, J. M. Pemberton, and D. R. McCullough. 2001. Bottlenecks, drift and differentiation: The population structure and demographic history of sika deer (Cervus nippon) in the Japanese archipelago. Molecular Ecology 10:1357–1370. Harrington, R. 1985. Evolution and distribution of the Cervidae. Pages 3–11 in P. F. Fennessy and K. R. Drew, editors, Biology of deer reproduction. Bulletin 22 of the Royal Society of New Zealand, Wellington, New Zealand. Imaizumi, Y. 1970. Description of a new species of Cervus from the Tsushima Island, Japan, with a revision of the subgenus Sika based on clinal analysis. Bulletin of the Natural Science Museum, Tokyo 13:185–194. Kuroda, N., and Y. Okada. 1950. On two new races of Cervus nippon from the southern islands of Kyushu, Japan. Annotationes Zoologicae Japonenses 24:59–64. Kuwayama, R., and T. Ozawa. 2000. Phylogenetic relationships among European red deer, wapiti, and sika deer inferred from mitochondrial DNA sequences. Molecular Phylogenetics and Evolution 15:115–123. Li, M., H. B. Tamate, F. W. Wei, X. Wang, R. Masuda, H. L. Sheng, and N. Ohtaishi. 2003. Phylogenetic relationships among deer in China derived from mitochondrial DNA cytochrome b sequences. Acta Theriologica 48:207–219. Lister, A. M. 1984. Evolutionary and ecological origins of British deer. Proceedings of the Royal Society of Edinburgh 82b:205–229. Loftus, R. T., D. E. MacHugh, D. G. Bradley, P. M. Sharp, and P. Cunningham. 1994. Evidence for two independent domestications of cattle. Proceedings of the National Academy of Sciences of the United States of America 91:2757–2761. Lowe, V. P. W., and A. S. Gardiner. 1975. Hybridization between red deer (Cervus elaphus) and sika deer (Cervus nippon) with particular reference to stocks in N. W. England. Journal of Zoology 177:553–566. Mahmut, H., R. Masuda, M. Onuma, M. Takahashi. J. Nagata, M. Suzuki, and N. Ohtaishi. 2002. Molecular phylogeography of the red deer (Cervus elaphus) populations in Xinjiang of China: Comparison with other Asian, European, and North American populations. Zoological Science 19:485–495. Matsumoto, M., H. Nishinakagawa, and J. Otsuka. 1984. Morphometrical study on the skull of Cervus pulchellus, Cervus nippon mageshimae and Cervus nippon yakushimae. Journal of the Mammalogy Society of Japan 10:41–53. McCullough, D. R., J. K. Fischer, and J. D. Ballou. 1996. From bottleneck to metapopulation: Recovery of the tule elk in California. Pages 375–403 in D. R. McCullough, editor, Metapopulations and Wildlife Conservation. Island Press, Washington, DC, USA. Nabata, D., R. Masuda, O. Takahashi, and J. Nagata. 2004. Bottleneck effects on the sika deer Cervus nippon population in Hokkaido, revealed by ancient DNA analysis. Zoological Science 21:473–481. Nagata, J., R. Masuda, K. Kaji, M. Kaneko, and M. C. Yoshida. 1998. Genetic variation and population structure of the Japanese sika deer (Cervus nippon) in Hokkaido Island, based on mitochondrial D-loop sequences. Molecular Ecology 7:871–877. Nagata, J., R. Masuda, H. B. Tamate, S. Hamasaki, K. Ochiai, M. Asada, S. Tatsuzawa, K. Suda, H. Tado, and Y. C. Yoshida. 1999. Two genetically distinct lineages of the sika deer, Cervus nippon, in Japanese islands: Comparison of mitochondrial D-loop region sequences. Molecular Phylogenetics and Evolution 13:511–519. Ohmura, Y., Y. Fukumoto, and K. Ohtaki. 1983. Chromosome polymorphism in Japanese sika, Cervus (Sika) nippon. Japan Journal of Veterinary Science 45:23–30. Ohtaishi, N., and Y. Gao. 1990. A review of the distribution of all species of deer (Tragulidae, Moschidae and Cervidae) in China. Mammalian Review 20:125–144.

4

Evolutionary Significance of Admixture and Fragmentation

59

Okada, A., and H. B. Tamate. 2000. Pedigree analysis of the sika deer (Cervus nippon) using microsatellite markers. Zoological Science 17:335–340. Okada, A., H. B. Tamate, M. Minami, N. Ohnishi, and S. Takatsuki. 2005. Use of microsatellite markers to assess the spatial genetic structure of a population of sika deer Cervus nippon on Kinkazan Island, Japan. Acta Theriologica 50:227–240. Page, R. D. M., and E. C. Holmes. 1998. Molecular evolution: A phylogenetic approach. Blackwell Science, Oxford, United Kingdom. Pitra, C., J. Fickel, E. Meijaard, and P. C. Groves. 2004. Evolution and phylogeny of old world deer. Molecular Phylogenetics and Evolution 33:880–895. Polziehn, R. O., and C. Strobeck. 1998. Phylogeny of wapiti, red deer, sika deer, and other North American cervids as determined from mitochondrial DNA. Molecular Phylogenetics and Evolution 10:249–258. Slate, J., C. W. Coltman, S. J. Goodman, I. MacLean, J. M. Pemberton, and J. L. Williams. 1998. Bovine microsatellite loci are highly conserved in red deer (Cervus elaphus), sika deer (Cervus nippon) and Soay sheep (Ovis aries). Animal Genetics 29:307–15. Slate, J., T. C. Van Stijin, R. M. Anderson, K. M. McEwan, N. J. Maqbool, H. C. Mathias, M. J. Bixley, D. R. Stevens, A. J. Molenaar, J. E. Beever, S. M. Galloway, and M. L. Tate. 2002. A deer (subfamily Cervinae) genetic linkage map and the evolution of ruminant genomes. Genetics 160:1587–1597. Slatkin, M. 1995. A measure of population subdivision based on microsatellite allele frequencies. Genetics 139:457–462. Stoneking, M., S. T. Sherry, A. J. Redd, and L. Vigilant. 1992. New approaches to dating suggest a recent age for the human mtDNA ancestor. Philosophical Transactions of the Royal Society B 337:34–37. Takahashi, M., R. Masuda, H. Uno, M. Yokoyama, M. Suzuki, M. C. Yoshida, and N. Ohtaishi. 1998. Sexing carcass remains of the sika deer (Cervus nippon) using PCR amplification of the Sry gene. Journal of Veterinary Medical Science 60:713–716. Tamate, H. B., S. Tatsuzawa, T. Suda, M. Izawa, T. Doi, K. Sunagawa, F. Miyahira, and H. Tado. 1998. Mitochondrial DNA variations in local populations of the Japanese sika deer, Cervus nippon. Journal of Mammalogy 79:1396–1403. Tamate, H. B., A. Okada, M. Minami, N. Ohnishi, H. Higuchi, and S. Takatsuki. 2000. Genetic variations revealed by microsatellite markers in a small population of the sika deer (Cervus nippon) on Kinkazan Island, northern Japan. Zoological Science 17:47–53. Tate, M. L., G. J. Goosen, H. Patene, A. J. Pearse, K. M. McEwan, and P. F. Fennessy. 1997. Genetic analysis of Pere-David x red deer interspecies hybrids. Journal of Heredity 88:361–365. Weir, B. S., and C. C. Cockerham. 1984. Estimating F-statistics for the Analysis of population structure. Evolution 38:1358–1370. Wu, H., Q. H. Wan, and S. G. Fang. 2004. Two genetically distinct units of the Chinese sika deer (Cervus nippon): Analyses of mitochondrial DNA variation. Biological Conservation 119:183–190. Yamada, M., E. Hosoi, H. B. Tamate, J. Nagata, S. Tatsuzawa, H. Tado, and S. Ozawa. 2006. Distribution of two distinct lineages of sika deer (Cervus nippon) on Shikoku Island revealed by mitochondrial DNA analysis. Mammal Study 31:23–28. Yamada, M., E. Hosoi, J. Nagata, H. B. Tamate and H. Tado. 2007. Phylogenetic relationship of the southern Japan lineages of the sika deer (Cervus nippon) in Shikoku and Kyushu islands, Japan. Mammal Study 32:121–127. Yamauchi, K., S. Mamasaki, K. Miyazaki, T. Kukusui, Y. Takeuchi, and Y. Mori. 2000. Sex determination based on fecal DNA analysis of the amelogenin gene in sika deer. Journal of Veterinary Medical Science 62:669–671. Yokohama, M., Y. Hoshi, H. Nishi, H. Sumiyoshi, and Y. Ishijima. 1994. Karyotype analysis of the Yeso sika (Cervus nippon yesoensis) and its related species. Journal of Agricultural Science 39:170–176. Yuasa, T., J. Nagata, S. Hamasaki, H. Tsuruga, and K. Furubayashi. 2007. The impact of habitat fragmentation on genetic structure of the Japanese sika deer (Cervus nippon) in southern Kantoh, revealed by mitochondrial D-loop sequences. Ecological Research 22:97–106.

Chapter 5

Nutritional Physiology of Wild and Domesticated Japanese Sika Deer Takayoshi Masuko and Kousaku Souma

Abstract The nutritional physiology of Yeso sika deer (Cervus nippon yesoensis, Hokkaido Island) and Honshu sika deer (C. n. centralis, Honshu Island) is reviewed and compared to results from domestic ruminants. Wild sika deer grazed on various types of plants, and the fiber content in these plants was low. The tastes of Yeso sika deer for existing feeds for ruminant livestock resembled those of sheep. Though the digestibility of these feeds in Yeso sika deer was slightly lower than that in sheep, the nutritive values of digestible crude protein (DCP) and total digestible nutrients (TDN) were similar between the two species, suggesting that feed for sheep can be utilized. Therefore, in small-scale farming of Yeso sika deer, the feeding amount in feeding planning can be determined using the feeding standards for sheep. However, when concentrates are fed, correction of TDN is necessary. In large-scale native pasturage, the nutritional intake in summer is adequate because Yeso sika deer graze on various types of wild plants or herbage. In early winter, they mainly graze on sasa (Sasa senanensis), and supplementary food may be necessary to supply TDN. Thus, since Yeso sika deer graze on many types of wild plants, existing feeds for ruminant livestock can be used. In addition, plant biomasses except concentrates that do not cause competition with existing livestock may be effectively utilized in Yeso sika deer, suggesting their importance as animal resources. Many problems must be evaluated before the deer farming industry can grow. In addition to administrative support, research results that enhance deer farming technology must be accumulated as quickly as possible. On the basis of the above research results on the nutritional physiology of Japanese sika deer, analysis of factors that affect fattening and meat quality of deer is necessary.

Introduction Deer farming for antler velvet or venison of sika deer (Cervus nippon), red deer (Cervus elaphus), rusa deer (Cervus timorensis), sambar (Cervus unicolor), and fallow deer (Dama dama) occurs in Oceania, Europe, and Asia (Drew et al. 1989; Fennessy and Taylor 1989; Fletcher 1989). At present, the number of farmed D. R. McCullough et al. (eds.), Sika Deer: Biology and Management of Native and Introduced Populations, DOI: 10.1007/978-4-431-09429-6_5, © Springer 2009

61

62

T. Masuko, K. Souma

sika deer is the highest in China, and that of red deer is the highest in New Zealand. In Japan, about three tons of velvet is annually imported from China and Taiwan and about 300 tons of venison from New Zealand (Yokohama et al. 1991). With such a background, deer farming has attracted attention in Japan where six subspecies of wild sika deer occur and deer farming was initiated in the mid-1980s. Since capture of wild deer is legally restricted, sika deer that have been maintained in zoos and parks in Japan, those imported from China and Taiwan, and red deer are farmed at present. The number of domesticated deer has increased gradually since 1990, and 4,600 deer were being raised at 144 farms in Japan in 1998. However, deer farming became stagnant thereafter, and the number of domesticated deer has not increased. Many fundamental problems have been suggested as causes, including lack of deer farming skills, inadequate slaughterhouse facilities, and poor development of the distribution system for venison, and more scientific and administrative support has been demanded. Approaches from nutritional physiology, management, and breeding science are needed to establish effective deer farming technology. However, the history of studies on nutritional physiology of sika deer in Japan is short (Miyazaki et al. 1984), and data that can be used as reference are scarce. In other countries, there have been studies on digestion and seasonal changes in feed intake in red deer and fallow deer under farming conditions (Henke et al. 1988; Ramanzin et al. 1997). In New Zealand, an advanced deer farming country, Barry and colleagues (Barry and Wilson 1990; Barry et al. 1991) have been leading research; they recently evaluated the association between feed intake and the endocrine system and reported results of studies on feed intake control, for instance, for prevention of a decrease in feed intake in winter. Recently, data concerning the nutritional physiology of Yeso (Hokkaido Island) sika deer (C. n. yesoensis) and Honshu sika deer (C. n. centralis) have been collected by the Masuko Group of the Faculty of Bioindustry, Tokyo University of Agriculture, and the Ikeda Group of Miyagi Agricultural College, respectively. In this chapter, the nutritional physiology of Japanese sika deer is discussed on the basis of these data.

Nutritional Composition of Plants Grazed on by Wild Japanese Sika Deer Yeso Sika Deer Souma et al. (1996) collected 13 types of herbaceous plants and 17 types of twigs, bark, endodermis, and fallen leaves of arboraceous plants that are frequently grazed on by wild Yeso sika deer and analyzed their compositions for proximate analysis (Horowitz 1980) (Table 5.1). When the component contents of herbaceous plants were compared with those of the heading stages of orchardgrass (Dactylis glomerata)

5 Nutritional Physiology of Wild and Domesticated Japanese Sika Deer

63

Table 5.1 Chemical composition of herbaceous plants and arboraceous plants. Moisture (% FMd) Herbaceous plants Aralia cordata 90.9 Aster glehni 80.4 Polygonatum odo- 84.3 ratum var. maximowiczii Lilium cordatumu 88.3 var. glehnii Heracleum 89.4 lanatum Trillium 86.0 kamtschaticum Patasites 91.2 japonicus var. giganteus Filipendula 76.3 kamtschatica Impatiens 90.7 noli-tangere Picris hieracioides 78.9 var. glabrescens 91.1 Anemonne flaccida Trifolium pratence 76.8 66.8 Sasa senanensis Arboraceous plants 43.8 Ulmus davisiana var. japonica 37.2 Alnus hirsula 30.7 Acer mono Betula platyphylla 29.7 var. japonica 39.6 Ulmus laciniata Fraxinus mandshu- 40.1 rica var. japonica 60.0 Taxus cuspidate 25.8 Tilia japonica 85.9 Euonymus sieboldianus 67.9 Aralia elata a EE: Ether extract b NFE: Nitrogen-free extract c NDF: Neutral detergent fiber d FM: Fresh matter e DM: Dry matter

Crude protein (% DMe)

EEa (% DM)

NFEb (% DM)

NDFc (% DM)

Gross energy (Mcal/ DMkg)

26.4 15.3 22.2

4.7 5.7 5.8

41.3 45.5 41.1

39.2 48.7 30.3

– 4.6 4.7

9.4

2.9

67.3

55.1

4.3

24.0

4.3

44.1

37.6

4.5

16.7

4.3

52.2

29.7

4.6

19.7

3.4

43.2

34.1

4.1

12.5

1.8

55.2

68.8

4.3

17.2

3.9

44.3

57.9

4.3

10.2

7.2

44.2

51.7

4.6

21.7

4.9

41.9

–

–

17.7 13.9

2.2 2.5

55.2 34.3

41.7 65.4

4.6 4.2

6.6

2.6

46.0

67.1

4.4

7.1 6.7 3.5

9.8 2.7 10.2

50.7 38.5 40.5

58.1 71.8 72.2

5.2 4.9 5.7

7.2 4.1

2.4 2.9

38.4 44.7

63.9 69.1

4.4 4.6

6.6 4.3 14.9

3.0 5.1 18.4

52.9 50.8 34.8

53.6 67.9 32.3

4.7 4.7 5.0

5.1

2.4

56.7

48.7

4.6

64

T. Masuko, K. Souma

and timothy (Phleum pratense), general grass types for dairy cows, (Secretariat of Agriculture, Forestry, and Fisheries Research Council, Ministry of Agriculture, Forestry, and Fisheries 1995), the nitrogen-free extract (NFE) content was similar, but the crude protein content in the herbaceous plants was higher, being only slightly lower than that in alfalfa hay. Comparison of component contents among the twigs, bark, and withered leaves showed a higher crude protein content in the twigs and withered leaves than in the bark but a higher crude fat content in the bark and withered leaves than in the twigs. On the other hand, when the component contents in the twigs, bark, and withered leaves were compared with those in orchardgrass and timothy, the acid detergent fiber (ADF) and neutral detergent fiber (NDF) contents in the twigs and bark were higher than those in these grasses, corresponding to the values in the blooming stage of each grass. The crude protein content in the withered leaves was lower than that in the grasses. In grazing tests, most wild plants were grazed on. However, when tested by feeding 67 types of herbaceous and arboraceous plants, some plants were not grazed on, and the grazing degree of some differed among individual deer (Souma et al. 1996). Among plants in which grazing marks by wild Yeso sika deer had been confirmed, plants such as Anemone flaccida (windflower) and Lilium glehni (Liliaceae) were not grazed. Why anemones were not selected may be that while young sprouts and young leaves are palatable (Takatsuki 1992), other parts contain a weakly toxic alkaloid substance (Hashimoto 1986; Nakai 1988). For Lilium glehni, the reason may be that parts other than the floral axis generally selected by wild Yeso sika deer were fed in the test. These results suggest that caution should used when plants containing toxic substances occur in pasturage using natural geographical features, but wild plants growing in deer farm pastures can be utilized. Masuko et al. (2001) collected twigs, bark, and withered leaves of 36 woody plants and measured chemical composition and in vitro dry matter digestibility (IVMD). The crude protein content and IVMD of withered leaves were higher than those of twigs and bark. In withered leaves, the crude protein content of Alnus hirsute (Manchurian alder) was highest (15.7% on a dry matter basis), and IVMD of Ulmus laciniata (Manchurian elm) was highest (69.4%). The crude fiber, ADF, and NDF contents of twigs and bark were higher than those of withered leaves. The fibrous contents of twigs and bark differed greatly among species of woody plants. The crude fiber, ADF, and NDF contents of Acer mono (painted maple) and Betula platyphylla var. japonica (Japanese white birch) were higher than those of other species. Concerning the composition of dwarf bamboo, Sasa senanensis, the contents of dry matter, crude fat, and crude ash increased while the contents of fiber components such as crude fiber, ADF, NDF, and hemicellulose decreased from leaf-bud formation to withering, but changes from October to June next year were slight (Souma et al. 1999) (Table 5.2). In general, the composition of grass fed to ruminant livestock varies markedly according to the growth stage (Morimoto 1989). In orchardgrass and timothy, the contents of dry matter, crude fat, crude fibers, ADF, and NDF increase with the growth period, but the content of the crude ash is nearly constant (Secretariat of Agriculture, Forestry, and Fisheries Research Council, Ministry of Agriculture, Forestry, and Fisheries 1995). When composition changes

5 Nutritional Physiology of Wild and Domesticated Japanese Sika Deer

65

Table 5.2 Changes in the chemical composition of Sasa senanensis with growth stage.

Date of collection

Dry matter (%FMd)

14 June 1995 25.7 5 August 1995 37.0 26 September 1995 47.5 27 October 1995 51.6 3 November 1995 47.1 31 December 1995 43.6 3 April 1996 49.3 16 May 1996 41.3 20 June 1996 48.0 14 July 1996 50.9 12 August 1996 51.6 14 September 1996 54.2 15 October 1996 61.7 a EE: Ether extract b NFE: Nitrogen-free extract c NDF: Neutral detergent fiber d FM: Fresh matter e DM: Dry matter

Crude protein (%DMe)

EEa (%DM)

NFEb (%DM)

NDFc (%DM)

Crude ash (%DM)

24.1 14.5 16.2 15.2 14.5 12.6 13.9 15.8 14.7 13.3 11.3 11.6 11.1

2.0 2.4 3.0 3.8 3.5 2.0 2.5 3.4 3.9 4.0 4.2 4.0 5.2

35.7 42.5 39.1 40.3 41.2 44.3 42.5 40.0 38.9 37.9 39.3 38.0 37.8

71.5 77.6 74.1 69.5 68.5 64.6 65.2 66.4 64.2 61.4 60.7 58.7 60.5

8.0 8.2 11.9 12.1 12.8 14.4 14.5 13.9 16.3 18.4 19.7 20.9 21.6

with growth are compared between Sasa senanensis and grass, the patterns of changes in the contents of crude fibers, ADF, and NDF differed, showing reverse changes (Souma et al. 1999). The slight composition changes from October to June next year may be due to inhibition of growth metabolism caused by low ground temperature and occasional snow cover that occurs during this period. In vitro dry matter digestibility decreased with the growth stage and changed only slightly from October to the following June, showing a pattern similar to that of composition changes (Souma et al. 1999) (Table 5.3). In general, digestibility also decreases with growth in grass, mainly due to an increase in the lignin content (Nakamura 1977; Morimoto 1989). Lignin is only slightly digestible, and an increase in lignin reduces degradation by enzymes, resulting in a decrease in digestibility. However, in Sasa senanensis, the contents of fiber components decreased with the growth stage, and it is unlikely that an increased lignin content reduced in vitro dry matter digestibility. Concerning the components of Sasa senanensis other than fiber, the crude ash content markedly increased while the organic matter content decreased with the growth stage (Table 5.2). This may have reduced in vitro dry matter digestibility. From autumn to spring, the composition and dry matter digestibility of Sasa senanensis are relatively stable. The dry matter intake during this period is also high (Masuko et al. 1999a, b). This suggests that a certain nutrient intake from Sasa senanensis is possible in Yeso sika deer.

66

T. Masuko, K. Souma Table 5.3 Changes in in vitro dry matter digestibility of Sasa senanensis leaves with growth stage. In vitro dry matter Date of collection digestibility (%) 14 June 1995 5 August 1995 26 September 1995 27 October 1995 3 November 1995 31 December 1995 3 April 1996 16 May 1996 20 June 1996 14 July 1996 12 August 1996 14 September 1996 15 October 1996

72.2 61.1 54.2 57.4 58.2 59.9 58.1 59.8 58.5 55.0 53.9 55.2 55.9

Honshu Sika Deer Ikeda et al. (1999) studied seasonal changes in the contents of crude protein and crude fiber in the leaves and edible parts of branches and stalks of 18 plant species (six deciduous trees, three deciduous shrubs, three evergreens, two forbs, and four grasses) for suitability of feeding by Honshu sika deer and serow (Capricornis crispis) in the Sendai area, which lies in the temperate/snow forest climate zone. The crude protein content in the leaves of growing deciduous trees varies widely among species and ranged from about 16% in Morus australis (Korean mulberry) to 10% in Clethra barvinervis (pepperbush). Among deciduous shrubs, it was high at about 17% in Lespedeza bicolor (bush clover) and about 10% in Viburnum dilatatum (linden arrowwood); it decreased to about 10% immediately before shedding of leaves in October to November and further to about 5% after shedding of leaves. In the leaves of evergreens, the crude protein content ranged from about 13% in Neolitsea sericea (Lauraceae) to about 9% in Pinus densiflora (Japanese red pine), but seasonal changes in each species were small. It remained high at about 15% throughout the year in Sasaella ramosa (a groundcover bamboo) and Sasa nipponica (a dwarf bamboo). In Zoysia japonica (lawn grass) and Miscanthus sinensis (silver grass), it was about 12% in summer but decreased to about 8% in autumn and thereafter. Seasonal changes in the crude protein content were smaller in the branches and stalks than in the leaves. In the growing period, the crude protein content of the leaves was 10% or higher in most plants and reached 15% or higher in some plants. In consideration of these nutritional contents and the abundant vegetation, the feeding environment of deer can be considered rich in the growing period of plants, but the crude protein content in the leaves of most plants decreased to about 5% in winter. Bamboo grass maintained a high crude protein content and was abundant throughout the year, so it is

5 Nutritional Physiology of Wild and Domesticated Japanese Sika Deer

67

an excellent feed for deer. Although Zoysia japonica is highly productive during summer, its value as a feed for deer during winter is small because of its limited actual bulk. Miscanthus sinensis has a large actual bulk, but it has a low nutritional value, because the crude protein content in its leaves decreased to about 3% in winter. The branches and withered leaves of woody plants retained a protein content of about 5%, and their supply was large even in winter if they were not buried under snow. The crude fiber content was high at 25–35% in leaves of Acuba japonica (Japanese laurel), Pinus densiflora, Quercus serrata (sawthorn oak), Morus australis, Viburnum dilatatum, Rubus palmatus var. coptophyllus (Rosaceae), and grasses. It was even higher in the branches and stalks, being 35% or higher in many species. As a feeding environment for Honshu sika deer and serow, the area of the Tohoku District along the Pacific coast seems to be adequate during summer but to be markedly deteriorated during winter.

The Fluid and Solid Contents of the Rumen of Sika Deer Composition of the Fluid and Solid Contents of the Rumen of Wild Yeso Sika Deer The fluid content of the rumen of wild Yeso sika deer (collected as harmful animals in Abashiri City by the Abashiri branch of the Hokkaido Government) showed a pH of 5.54, total volatile fatty acid (VFA) concentration of 22.7 mmol/dl, ammonia concentration of 48.6 mg/dl, acetic acid molar ratio of 48.2%, propionic acid molar ratio of 35.4%, and protozoa count of 8.9 × 105/ml (Table 5.4). The total VFA concentration, ammonia concentration, and the propionic acid molar ratio in Yeso sika deer were higher than those in cattle and sheep, but the pH and acetic acid molar ratio were lower. The number of protozoa in Yeso sika deer was similar to that in sheep and goats.

Table 5.4 Composition of the fluid rumen contents of wild Yeso sika deera (mean ± standard error, N = 11). pH 5.45 ± 0.06 Ammonia-N (mg/dl)b 48.6 ± 6.6 VFA (mmol)c 22.7 ± 2.4 Acetic acid (molar ratio %) 48.2 ± 1.1 Propionic acid (molar ratio %) 35.4 ± 1.0 Butyric acid (molar ratio %) 12.3 ± 0.4 Protozoa count (´105/ml) 8.9 ± 2.2 a Wild Yeso sika deer collected as harmful animals b N: Nitrogen c VFA: Volatile fatty acid

68

T. Masuko, K. Souma

Seasonal differences were observed in the rumen contents. Arboreal plants such as Sasa senanensis leaves, twigs, and bark were found in addition to herbaceous plants in spring, but mainly herbaceous plants were found in summer. In autumn and winter, root vegetables, nuts, and leaves (including dry leaves) of Sasa senanensis and other species were observed. These contents were similar to those of wild Yeso sika deer captured for scientific purposes at Ashiyoro-cho and Onbetsu-cho in Hokkaido (Hokkaido Institute of Environmental Sciences 1995). These findings suggested that Yeso sika deer graze on a wide variety of foods and like to graze on field crops and fallen leaves. The fermentation pattern of these rumen contents was similar to that observed in cattle or sheep grazing on plants with a low content of structural carbohydrates (cellulose, hemicellulose, and lignin) and a high protein content (Masuko et al. 1992). Rumen samples were also collected from 36 deer killed in non-agricultural areas for 11 months from March through the following January to study the rumen contents of wild Yeso sika deer. Major components of the contents were bamboo grasses in March, grasses in April through June, leguminous plants in July and August, and grasses and bamboo grasses in October and thereafter. The pH of fluid content of the rumen was highest at 7.34 in March, decreased to 5.40 in May, and increased gradually to 6.53 in January. The ammonia concentration was lowest at 10.4 mg/dl in March, reached a peak at 69.6 mg/dl in May, remained above 20.9 mg/dl until October, and decreased further in and after November (Fig. 5.1). The major minerals sodium (Na), phosphorus (P), potassium (K), and calcium (Ca) and the trace elements iron (Fe), copper (Cu), and zinc (Zn) were analyzed in the fluid and solid contents of the rumen. Only the mineral concentrations in the solid contents are discussed here. The Ca and P contents showed large peaks between June and October and were low in the months before and after this period (Fig. 5.2). Seasonal changes in the Na and K contents were small. The Fe content showed a large peak between March and October, and the Cu and Zn contents were low throughout the year (Fig. 5.3). In domesticated deer, mineral requirements can be provided by feeding, but wild sika deer must consume minerals selectively in the diet. The sources of minerals

Fig. 5.1 The ammonia concentration of the fluid contents of the rumen of wild Yeso sika deer.

5 Nutritional Physiology of Wild and Domesticated Japanese Sika Deer

69

Trace minerals (ppm DM)

Fig. 5.2 The major minerals contents of the solid contents of the rumen of wild Yeso sika deer.

1400 1200

Fe

1000 Cu

800

Zn

600 400 200 0 3

4

5

6

7

8

10

11

12

1

Months of year Fig. 5.3 The trace mineral contents of the solid contents of the rumen of wild Yeso sika deer.

could not be determined in this study, but the seasonal changes observed in the mineral concentrations suggest a relationship of the mineral intake with the route of migration of wild sika deer. Studies of this relationship are needed for the future. The dry matter ratio in the contents of the rumen-abomasum was much higher than that reported in sheep (Tsuda 1990), suggesting a high concentration of gastric contents in Yeso sika deer (Masuko et al. 1996). The ratio of the total content weight (fresh matter weight) in the rumen and reticulum to body weight was 6.1% (Table 5.5). This ratio has been reported to be 4.1% in roe deer, 9.0% in red deer, and 4.3% in fallow deer (Nagy and Regelin 1975). The ratio of the rumen content

70

T. Masuko, K. Souma

Table 5.5 Ratio of the total content weight (mean ± standard error; N = 9) in the rumen, reticulum, omasum, and abomasum.

Rumen Reticulum Omasum Abomasum a b

Fresh matter weight (g)

Dry matter weight (g)

FMWa/body weight (g/kg)

DMWb/body weight (g/kg)

4,598.6 ± 678.0 190.0 ± 42.3 49.2 ± 15.3 152.9 ± 20.4

743.9 ± 102.3 29.1 ± 6.8 8.9 ± 3.0 23.0 ± 3.4

58.44 ± 5.35 2.38 ± 0.42 0.70 ± 0.20 2.03 ± 0.29

9.48 ± 0.88 0.36 ± 0.06 0.12 ± 0.04 0.30 ± 0.04

FMW: Fresh matter weight DMW: Dry matter weight

weight (dry matter weight) to body weight was nearly 1.0% (Masuko et al. 1996). Considering that the daily feed intake per body weight was 2.3–2.8% in Japanese sika deer allowed free access to alfalfa pellets and beet pulp (Tsuda et al. 1987), this ratio (1.0%) corresponded to 50% of the daily dry matter intake, suggesting adequate feed intake in wild Yeso sika deer.

Composition of the Fluid Contents of the Rumen of Domesticated Honshu Sika Deer The results of analysis of the fluid contents from the rumen of deer obtained after feeding by inserting a rumen fistula have been reported (Ikeda et al. 1992). Their feed consisted of raw grass, alfalfa hay cubes, compound feed for dairy cows, hay, grass silage, and apple lees. In the fluid content of the rumen, the pH was 5.7 to 7.6, ammonia concentration was 3.1–8.2 mg/dl, and protozoa count was 5.0 × 104/ml to 2.5 × 105/ml. These values were similar to the above-described values in cattle and sheep but differed from those in Yeso sika deer reported by Masuko et al. (1992). Fermentation in the rumen widely varies according to the characteristics of consumed feed components (McCullough 1979). Therefore, this inconsistency between the two studies may be associated with differences in the rumen fluid between Yeso sika deer fed prepared feeds and wild Yeso sika deer that selectively grazed on feeds suiting their preferences.

Digestive Ability of Sika Deer Preference for Feeds Yeso sika deer. Souma et al. (1995) evaluated the preferences of Yeso sika deer for hay and silage, as generally fed to livestock, and Sasa senanensis, which is heavily grazed by wild Yeso sika deer throughout the year. Yeso sika deer favored roll baled grass hay and corn silage according to preparation types (Fig. 5.4), and alfalfa

5 Nutritional Physiology of Wild and Domesticated Japanese Sika Deer

71

100% Sasa senanensis 80% Corn Silage 60% High-moisture grass silage 40% Roll baled grass silage 20% Roll baled grass hay 0% Sheep

Deer

Fig. 5.4 The ratio of each preparation type in total dry matter intake. 100% Alfalfa

80%

Perennial ryegrass 60% Orchardgrass 40%

20% 0% Sheep

Deer

Fig. 5.5 The ratio of each legume/grass type in total dry matter intake.

(Medicago sativa L.) and perennial ryegrass (Lolium perenne L.) according to legume/ grass types (Fig. 5.5). High-moisture grass silage was least favored. Another study on the preferences for grasses in sheep showed a preference of perennial ryegrass to orchardgrass (Sawada 1994). These types of roughage are fed to dairy cattle but considered to be also suited for domesticated Yeso sika deer. Sasa senanensis, together with Sasa kurilensis and Sasa nipponica, are widely distributed in Hokkaido (Toyooka et al. 1983) and are preferred by Yeso sika deer, suggesting their usefulness as food resources. When sasa is used as feed, autumn and winter are appropriate seasons. However, when the leaves of Sasa nipponica are grazed on in pasturage in summer, its subsequent reproduction is known to decrease (Okubo et al. 1990). Where sika graze on Sasa senanensis, deer numbers should be maintained low enough to not reduce its reproduction.

72

T. Masuko, K. Souma

Comparison between Yeso sika deer and sheep in terms of feed intake showed similar preferences (Figs. 5.4 and 5.5); feed types mainly grazed by Yeso sika deer were also frequently grazed by sheep. Honshu sika deer. Ikeda et al. (1991) studied preferences of farmed Honshu sika deer for basic foods commonly given them. The animals began eating concentrated feed, wheat bran, and apple lees immediately when they were offered, indicating high preference for them. The preference for raw grass, hay, and corn silage was also good, although hard parts of stalks were left uneaten. Rice straw and grape leaves were least preferred.

Feed Intake Yeso sika deer. In general, feed intake differs among the seasons in deer (Barry et al. 1991; Ikeda et al. 1991; Odajima et al. 1993; Aagnes et al. 1996). In Yeso sika deer, seasonal changes were also observed in hay intake (Souma et al. 1998) (Table 5.6). The feed intake of hay reached a peak in summer, decreased in autumn, was lowest in winter, and recovered in spring. The feed intake relative to body weight was 3.01% per day, and its ratio to metabolic body size was 82.2 g/kgW0.75/day in summer but dropped to 1.60% and 43.9 g/kgW0.75/day in winter. Changes in feed intake were not associated with age or sex. The average body weight of deer was heaviest at 62.6 kg in autumn and was 57.4 kg in winter, 58.4 kg in spring, and 58.1 kg in summer. In Yeso sika deer, the intake of Sasa senanensis differed among autumn, winter, and spring (Masuko et al. 1999a, b) (Table 5.7). The dry matter intake of Sasa Table 5.6 Seasonal changes in the mean hay intake of domesticated Yeso sika deer (N = 5). Intake (g/day) Intake (% of BW/day)e Intake (g/kgW0.75/day)f

Spring

Summer

Autumn

Winter

SEMa

1370.4b, c 2.35b, c 64.8b, c

1705.4b 3.01b 82.2b

1172.4c, d 1.88c, d 52.7c, d

908.0d 1.60d 43.9d

91.0 0.16 4.2

a

SEM: Standard error of the means Means within the same row with different superscripts differ: P < 0.05 e BW: Body weight f W0.75: Metabolic body size b, c, d

Table 5.7 Seasonal changes in the dry matter intake of Sasa senanensis in domesticated Yeso sika deer (N = 4). Autumn Mean ± SE Spring Mean ± SE Difference Intake (DMg/day)a Intake (% of BW/day)b Intake (DMg/kgW0.75/day)c a DM: Dry matter b BW: Body weight c W0.75: Metabolic body size

956.8 ± 146.5 1.66 ± 0.16 45.7 ± 5.0

1,008.5 ± 109.5 1.95 ± 0.18 52.3 ± 4.8

NS P < 0.01 P < 0.01

5 Nutritional Physiology of Wild and Domesticated Japanese Sika Deer

73

senanensis per day was the highest in winter, followed in order by autumn and spring. The dry matter intake per body weight was the highest in autumn, followed in order by winter and spring. This change in the order of autumn and spring is because dry matter intake was most markedly decreased in winter, but body weight was decreased in winter compared with autumn. The dry matter intake of Sasa senanensis per body weight in spring was lower, but that in autumn and winter was slightly higher than the dry matter intake of hay per body weight in each season. These results suggest that the dry matter intake of Sasa senanensis in Yeso sika deer is similar to or higher than that of hay, and Sasa senanensis is an important feed resource for nutrient intake in winter. These seasonal changes in feed intake may be partly due to seasonal differences in the types of plants that can be grazed by wild sika deer. Deer such as red deer are known to show seasonal differences in behavior. In autumn, during their reproduction season, male deer show active reproductive behavior such as defense of harems, and the resulting marked decrease in the grazing time reduces feed intake. In female deer, grazing is interrupted by male deer that form harems and guard females, which also reduces the grazing time compared with the other seasons (Souma et al. 1994). In ruminants, changes in feed intake are considered to be affected by temperature and other aspects of the physical environment (Okamoto 1970; Mimura and Morita 1990). In particular, feed intake is associated with atmospheric temperature. Since body temperature is adjusted according to environmental temperature, energy consumption increases in the cold. However, in deer such as Yeso sika deer, feed intake was the lowest in winter, suggesting the importance of factors other than atmospheric temperature. Concerning such factors, Barry et al. (1991) suggested an association between feed intake and hormones. Melatonin secreted from the pineal body is known as a hormone associated with seasonality (Tomioka 1996). The major roles of melatonin are inhibition of gonad activity, development of biological rhythms, and adjustment of the circadian rhythm. The secretion of melatonin is markedly affected by light/dark cycles, i.e., day length (Ishida 1995; Tomioka 1996). In an experiment with red deer, feed intake increased in summer and decreased in winter, showing a pronounced seasonal cycle. Melatonin administration depressed feed intake in late spring and summer and increased feed intake in autumn and winter, relative to control deer, and appeared to move the cycles by approximately six months (Françoise-Domingue et al. 1992). Administration of melatonin was reported to decrease feed intake (Barry et al. 1991). Thus, Yeso sika deer show seasonality in feed intake even when fed adequate rations under farming conditions, and such seasonal changes appear to be associated with the endocrine system. Since feed intake and body weight of Yeso sika deer decrease in winter, the feeding method and fattening period in each season should be evaluated, and inadequate nutritional intake associated with decreased feed intake from autumn to spring should be avoided in farming of Yeso sika deer. Honshu sika deer. When Honshu sika deer were continuously fed on alfalfa hay cubes, the feed intake showed annual changes associated with seasons (Ikeda 2000). It increased suddenly in the mating season in males although they were not

74

T. Masuko, K. Souma

kept with females. The feed intake of deer was 2.5–3.5% of the body weight during the high-feeding period between March and September, but 1.2–3.1% during the low-feeding period between October and February. The annual mean of the feed intake of red deer was reported to be 3.0 kg/day, or 2.4% of the body weight (125 kg). The annual feed intake of wapiti was reported to be 1–3% of the body weight. The feed intake of red deer and wapiti converted to the ratio relative to the metabolic body size was 30–40 g/day and 35–85 g/day respectively (Takatsuki 1994). The feed intake of Honshu sika deer was 60–90 g/day in females and 38–100 g/day in males, showing a considerable variation (Ikeda 2000). On comparison among seasons, the body weight was largest in autumn, decreased toward winter, and increased from spring to summer. In domesticated Honshu sika deer fed continuously, seasonal changes were observed in the quantity of back fat deposit. The back fat thickness was 1.31, 0.91, 0.90, and 0.81 cm in autumn (November), winter (February), spring (May), and summer (August), respectively (Ikeda 2000). The quantity of back fat deposit tended to increase in autumn and decrease in winter although its seasonal differences were not significant.

Digestibility and Nutritive Value of Feeds Masuko et al. (1997) evaluated the digestibility and nutritive value of hay, silage, and hay cubes for Yeso sika deer. In sika deer fed roll baled grass hay, roll baled grass silage, and alfalfa hay cubes corresponding to 1.8–2.2% of body weight, the digestion rates of crude fibers and hemicellulose were slightly lower than those in sheep, but the digestion rates of the other components were similar. However, since the findings on the digestion rate of ADF differ from those on NDF, further detailed studies on the digestibility of fiber components are necessary. The contents of digestible crude protein (DCP), total digestible nutrients (TDN), and digestible energy (DE) in the three feed types were similar between Yeso sika deer and sheep. Therefore, the nutritive values in grasses fed to sheep are available for Yeso sika deer. Some Yeso sika deer with a particularly high urine volume showed a high urinary nitrogen excretion and a high nitrogen accumulation amount and rate compared with sheep. However, in other Yeso sika deer, the nitrogen accumulation amount and rate were similar to those in sheep (Masuko et al. 1997). Yeso sika deer fed a mixture of hay and wheat bran or soybean meal showed higher digestion rates of all components than did sheep (Masuko et al. 1998). Also, when a single grass type is fed alone, the digestion rates of fiber components are lower in Yeso sika deer than in sheep. However, when hay mixed with wheat bran with lower fiber content or soybean meal with a higher crude protein content is fed, the total fiber digestion rate is higher in Yeso sika deer than in sheep. This suggests high digestibility of a grass-concentrate mixture in Yeso sika deer. This tendency was more marked using soybean meal than using wheat bran, indicating that the combination with a feed with high protein content markedly improves the digestibility of fiber components as reported by McCullough (1979).

5 Nutritional Physiology of Wild and Domesticated Japanese Sika Deer

75

After feeding hay or silage alone, the DCP, TDN, and DE contents were similar between Yeso sika deer and sheep (Masuko et al. 1997). After feeding of a mixture of each type of grass and wheat bran or soybean meal, the DCP content was similar between Yeso sika deer and sheep, but the TDN and DE contents were higher in deer. This tendency was marked using the combination with soybean meal; the TDN content differed by 10.1% and the DE content by 1.06 Mcal/kg (Table 5.8). The fecal nitrogen excretion in Yeso sika deer was similar to or slightly lower than that in sheep while the urinary nitrogen excretion in Yeso sika deer was higher than that in sheep. Therefore, both the nitrogen accumulation amount and rate in Yeso sika deer were high. The urine volume in Yeso sika deer was about 1.6 to 2.2 times that in sheep. However, due to the low nitrogen concentration in urine, the urinary nitrogen excretion in Yeso sika deer was lower than that in sheep (Table 5.9). This may cause a high nitrogen accumulation in Yeso sika deer. Detailed analysis of urinary nitrogen is necessary to evaluate differences between Yeso sika deer and sheep. The digestion rates of the dry matter, organic matter, and energy of Sasa senanensis were low (44.5–48.6%) (Table 5.10). These values were similar to those of roll baled grass silage consisting mainly of second cut orchardgrass (36.6–43.7%) among feed types used in previous digestion tests (Masuko et al. 1997). The digestion rates of crude fibers, ADF, NDF, and hemicellulose after feeding roll baled grass silage were 37.1–51.6%, which were similar to those of Sasa senanensis

Table 5.8 Digestibility and nutrient content of a mixture of hay and concentrate fed to sheep (N = 4) and Yeso sika deer (N = 4). (Significant difference between deer and sheep in each treatment: * P < 0.05 and ** P < 0.01.) Hay + wheat bran Digestibility (%) Dry matter Crude protein NFEb NDFc Energy Nutrient content DCPd (%DMe) TDNf (%DM) DEg (Mcal/kgDM) a

Deer

SEMa

Sheep

Deer

SEMa

56.3 58.0 61.6 52.9 56.5

63.9 60.1 66.3 61.5 62.1

4.3 3.9 3.8 4.8 4.2

56.5 70.1 54.9 55.4 55.3

68.3** 74.7** 67.5** 67.5** 66.4**

2.9 1.2 2.9 3.1 3.0

7.4 57.9 2.96

7.5 63.1 3.22

0.5 3.9 0.38

14.8 55.5 2.37

15.0 65.6** 3.43*

0.3 10.4 0.32

SEM: Standard error of the mean NFE: Nitrogen-free extract c NDF: Neutral detergent fiber d DCP: Digestible crude protein e DM: Dry matter f TDN: Total digestible nutrients g DE: Digestible energy b

Hay + soybean meal

Sheep

76

T. Masuko, K. Souma Table 5.9 Nitrogen balance in sheep (N = 4) and Yeso sika (N = 4) deer fed a mixture of hay and concentrate. Hay + wheat bran Intake of Nb (g/kgW0.75/day)c Fecal N (g/kgW0.75/day) Urinary N (g/kgW0.75/day) Retained N (g/kgW0.75/day)

Hay + soybean meal

Sheep

Deer

SEMa

Sheep

Deer

SEMa

1.03 0.44 0.47 0.13

1.07 0.43 0.43 0.21

0.03 0.05 0.02 0.04

1.57 0.47 0.80 0.30

1.74 0.44 0.75 0.55**

0.06 0.02 0.03 0.07

a

SEM: Standard error of the means N: Nitrogen c W0.75: Metabolic body size ** P < 0.01 b

Table 5.10 Digestibility and nutrient content of Sasa senanensis fed to Yeso sika deer (N = 3). Mean SEMa Digestibility (%) Dry matter 44.5 Crude protein 74.5 NFEb 42.2 NDFc 45.9 Energy 48.4 Nutrient content 12.7 DCPd (%DMe) TDNf (%DM) 44.7 DEg (Mcal/kgDM) 2.21 a SEM: Standard error of the means b NFE: Nitrogen-free extract c NDF: Neutral detergent fiber d DCP: Digestible crude protein e DM: Dry matter f TDN: Total digestible nutrients g DE: Digestible energy

1.6 0.7 0.6 0.5 0.4 0.1 0.4 0.02

(43.7–47.5%). However, the digestion rate of crude protein in Sasa senanensis (74.5%) was much higher than that in roll baled grass silage (53.5–55.2%) (Masuko et al. 1999a). Masuko et al. (1998) reported a high digestion rate of crude protein (74.7%) after feeding of a hay-soybean meal mixture. The digestion rate of crude protein in Sasa senanensis was similar to this value. Similar findings were also reported in experiments in which sika deer were fed Sasa palmata (broadleaf bamboo) (Matoba et al. 1987). The DCP content in dry matter was markedly high (12.7%). This value was only slightly lower than that after feeding of a hay-soybean meal mixture, suggesting a high nutritive value of Sasa senanensis in terms of protein. However, the TDN content in dry matter was low (44.7%) since the digestion rate of each component other than protein was low (Table 5.10), and was only slightly higher than the TDN content in roll baled grass silage.

5 Nutritional Physiology of Wild and Domesticated Japanese Sika Deer

77

Table 5.11 Nitrogen balance in Yeso sika deer fed Sasa senanensis (N = 3). Mean SEMa Intake of Nb (g/kgW0.75/day)c Fecal N (g/kgW0.75/day) Urinary N (g/kgW0.75/day) Retained N (g/kgW0.75/day) a SEM: Standard error of the means b N: Nitrogen c W0.75: Metabolic body size

1.20 0.31 0.81 0.08

0.08 0.01 0.09 0.05

Concerning nitrogen balance, the fecal nitrogen excretion was low, but the nitrogen accumulation amount and rate were high because of a high urinary nitrogen excretion (Masuko et al. 1999a) (Table 5.11).

Body Weights of Sika Deer Uchida et al. (2001) measured body weights of artificially-reared Honshu sika deer, including 10 males and 11 females from birth until 24 to 36 months of age. The monthly changes in body weights from birth until 36 months of age are shown in Fig. 5.6, and the least square means of body weights at birth and at 4, 6, 12, 24 and 36 months of age are shown in Table 5.12. The actual body weights at birth were 4.3 ± 0.8 kg and 4.3 ± 0.6 kg for females and males, respectively. In the first three months, the increase in body weight of males and females was almost the same. However, the rate of increase in body weights of males became greater than that of females after four months of age; the actual body weights for females and males were 28.3 ± 3.4 kg and 31.1 ± 4.0 kg at six months of age, 39.9 ± 4.7 kg and 47.1 ± 6.8 kg at 12 months of age, 53.9 ± 7.0 kg and 66.4 ± 8.3 kg at 24 months of age, and 61.6 ± 8.8 kg and 83.4 ± 12.8 kg at 36 months, respectively. These differences in rates of body weight increase became greater with advancing of age. The rate of increase in body weights of females, but not males, became asymptotic after 30 months of age. The body weights of both sexes recorded in this study were greater than those of artificially reared sika deer reported by Yamane et al. (1997), which were measured in early winter at one and three years of age. Body weights of both males and females tended to stop increasing or to decrease during the breeding season, and this tendency was more pronounced in males. The sika deer were born from June to September and their first breeding season appeared from November of the following year to late February two years later. These correspond to 17 to 20 months and 29 to 32 months of age, respectively (Fig. 5.6). A reduction in feed intake and a 20% reduction in body weights of wild male sika deer in the breeding season was reported by Shiraishi et al. (1996) who found that the body weights of wild male sika deer increased and decreased repeatedly from September to June of the following year. In contrast, the body weights of wild female

78

T. Masuko, K. Souma 90

Male

Body weight (kg)

80

Female

70 60 50 40 30 20 10 0

1

4

7

10

13

16

19

22

25

28

31

34

37

Month of age Fig. 5.6 Changes in body weight means of male and female sika deer.

Table 5.12 Comparison of body weight (least square mean in kg ± SE) by sex and family (classified according to paternal and maternal pedigrees) in Honshu sika deer. Age in months Sex Female Male Family 1 2 3 4

Birth

4

6

12

24

36

4.2 ± 0.22 4.3 ± 0.26

24.2 ± 0.84a 27.2 ± 0.95c

28.6 ± 0.88a 32.5 ± 0.95c

39.5 ± 1.33b 47.4 ± 1.58d

54.0 ± 2.45b 59.7 ± 3.58a 67.1 ± 2.69d 87.1 ± 6.86c

3.9 ± 0.36 4.5 ± 0.27 4.2 ± 0.36 4.6 ± 0.37

22.9 ± 1.36a,b 23.7 ± 0.97a,b 29.2 ± 1.36d 26.9 ± 1.40a,d

28.4 ± 1.43a 27.7 ± 0.96b 32.4 ± 1.43d 33.6 ± 1.47c,d

36.9 ± 2.52a,b 42.2 ± 1.45a 45.3 ± 2.16c 49.5 ± 2.22c,d

57.7 ± 4.00 59.0 ± 2.66 63.2 ± 4.00 62.6 ± 4.08

68.1 ± 5.69 69.5 ± 7.92 73.3 ± 5.69 82.8 ± 6.90

a,c

Least square means in the same column with different superscript letters differ significantly (P < 0.05) b,d Least square means in the same column with different superscript letters differ significantly (P < 0.01)

sika deer did not change or in some cases actually increased slightly during the winter period from December to March of the following year.

Discussion The digestive tract of ruminants can be classified according to morphological characteristics into the grass-eater type showing high fiber digestibility, concentrateselector type showing low fiber digestibility, and the intermediate type. Sika deer

5 Nutritional Physiology of Wild and Domesticated Japanese Sika Deer

79

are classified as the intermediate type/concentrate-selector type while sheep are classified as the grass-eater type (Hofmann 1988). Deer are distributed in forests and grasslands and have feeding habits intermediate between the forest type (leafeaters) and grassland type (grass-eaters), and most closely resemble the feeding habits in goats among domestic ruminants such as cattle, sheep, and goats (Takatsuki 1992). Terada et al. (1987) observed differences in the digestibility of the fiber fraction of feed between sheep and goats, suggesting that this is due to differences in the activity of rumen microorganisms and the digestive tract feed passage rate. Odajima et al. (1991) and Katoh et al. (1991) compared the digestive tract feed passage rate between Honshu sika deer and sheep and observed a significantly higher rate in deer, suggesting that this is a cause of the low digestibility. After feeding a hay-concentrate mixture, the presence of the concentrate may have affected rumen microorganisms, improving the digestibility of all fiber components. With an increase in digestibility, the TDN and DE contents also improved. Japan has had only a short history of research on nutritional physiology of domesticated Japanese sika deer, and this field needs further development. Both research objectives and clear industrial needs must be present for sika deer to become an attractive subject for many researchers. How the deer farming industry will develop is unclear; nevertheless, certain preparations will be necessary for success. Since the nutritive values of DCP and TDN in Yeso sika deer were similar to those in sheep, feed menus for sheep can be utilized (Masuko et al. 2000). Therefore, in small-scale deer farming, the feeding amount can be determined using feeding standards for sheep. In large-scale farming, sika deer grazed on various types of plants or herbage. Thus, from spring to autumn when wild plants can be used, adequate nutritional intake may be achieved. However, in early winter, use of mainly Sasa senanensis is considered. When only Sasa senanensis is fed, TDN intake may be inadequate, and supplementary feed may be necessary. Sika deer graze on various types of plants. Therefore, sika deer may be able to effectively utilize plant biomass resources in addition to existing feeds for ruminant livestock (Secretariat of Agriculture, Forestry, and Fisheries Research Council, Ministry of Agriculture, Forestry, and Fisheries 1991). Many problems must be evaluated before the deer farming industry can grow in Japan. In addition to administrative support, research results that enhance deer farming technology must be accumulated as quickly as possible. On the basis of the above research results on the nutritional physiology of Japanese sika deer, analysis of factors that affect fattening and meat quality of deer is necessary.

Literature Cited Aagnes, T. H., A. S. Blix, and S. D. Mathiesen. 1996. Food intake, digestibility and rumen fermentation in reindeer fed baled timothy silage in summer and winter. Journal of Agricultural Science 127:517–523. Barry, T. N., and P. R. Wilson. 1990. Development and present status of deer farming industry in New Zealand – International Deer Symposium Report (II). Animal Husbandry 44:1015–1021.

80

T. Masuko, K. Souma

Barry, T. N., J. M. Suttie, J. A. Milne, and R. N. B. Kay. 1991. Control of food intake in domesticated deer. Pages 385–401 in T. Tsuda, Y. Sasaki, and R. Kawashima, editors, Physiological aspects of digestion and metabolism in ruminants. Academic Press, Inc., San Diego, California, USA. Drew, K. R., Q. Bai and E. V. Fadeev. 1989. Deer farming in Asia. Pages 309–345 in R. J. Hudson, K. R. Drew, and L. M. Baskin, editors, Wildlife production systems: Economic utilization of wild ungulates. Cambridge University Press, Cambridge, United Kingdom. Fennessy, P. F., and P. G. Taylor. 1989. Deer farming in Oceania. Pages 309–345 in R. J. Hudson, K. R. Drew, and L. M. Baskin, editors, Wildlife production systems: Economic utilization of wild ungulates. Cambridge University Press, Cambridge, United Kingdom. Fletcher, T. J. 1989. Deer farming in Europe. Pages 309–345 in R. J. Hudson, K. R. Drew, and L. M. Baskin, editors, Wildlife production systems. Economic utilization of wild ungulates. Cambridge University Press, Cambridge, United Kingdom. Françoise-Domingue, B. M., P. R. Wilson, D. W. Dellow, and T. N. Barry. 1992. Effect of subcutaneous melatonin implants during long daylength on voluntary feed intake, rumen capacity and heart rate of red deer (Cervus elaphus) fed on a forage diet. British Journal of Nutrition 68:77–88. Hashimoto, I. 1986. Illustrated book of delicious wild plants 2. Seibundo Shinkosha, Tokyo, Japan. (In Japanese.) Henke, S. E., S. Demarais, and J. A. Pfister. 1988. Digestive capacity and diet of white-tailed deer and exotic ruminants. Journal of Wildlife Management 52:595–598. Hofmann, R. R. 1988. Aspects of digestive physiology in ruminants. Comstock Publishing Associates, Ithaca, New York, USA. Hokkaido Institute of Environmental Science. 1995. Report of survey of inhabitant status of brown bears and Yeso sika deer (I). (In Japanese.) Horowitz, W., editor. 1980. Official methods of analysis of the Association of Official Analytical Chemists, 13th edition. AOAC, Washington, DC, USA. Ikeda, S. 2000. Studies on the seasonal productivity of Japanese sika deer. Bulletin of the Miyagi Agricultural College 10:3–12. (In Japanese.) Ikeda, S., and S. Takatsuki. 1999. Seasonal changes in nutritive compositions of the major food plants of sika deer (Cervus nippon) and Japanese serow (Capricornis crispus)—a case study of the Sendai area. Tohoku Journal of Animal Science and Technology 49:1–8. (In Japanese.) Ikeda, S., T. Takeda, M. Ishida, and T. Saito. 1991. Feeding and digestion tendencies of sika deer (Cervus nippon). Report of Miyagi Agricultural College 38:27–36. (In Japanese.) Ikeda, S., T. Sugata, T. Takeda, M. Ishida, and T. Saito. 1992. Properties of rumen contents of sika deer (Cervus nippon) fed on various feeds. Report of Miyagi Agricultural College 40:51–58. (In Japanese.) Ishida, N. 1995. What is the biological clock? Kogyo Gijutsu 36:52–57. (In Japanese.) Katoh, K., Y. Kajita, M. Odashima, M. Ohta, and Y. Sasaki. 1991. Passage and digestibility of lucern (Medicago sativa) hay in Japanese sika deer (Cervus nippon) and sheep under restricted feeding. British Journal of Nutrition 66:399–405. Masuko, T., Y. Kameyama, M. Yokohama, and Y. Ishijima. 1992. Characteristics of rumen contents in Yeso sika deer (Cervus nippon yesoensis). Journal of Agricultural Science, Tokyo Nogyo Daigaku 37:162–165. (In Japanese.) Masuko, T., K. Souma, and Y. Ishijima. 1996. The amount of rumen contents in wild Yeso sika deer (Cervus nippon yesoensis). Grassland Science 42:176–177. (In Japanese.) Masuko, T., K. Souma, H. Kumagai, K. Takasaki, Y. Kameyama, and Y. Ishijima. 1997. Digestibility and nitrogen balance in Yeso sika deer (Cervus nippon yesoensis) fed round baled hay, alfalfa hay cube and round baled silage. Grassland Science 43:32–36. (In Japanese.) Masuko, T., K. Souma, M. Fujii, K. Takasaki, and Y. Ishijima. 1998. Digestibility and nitrogen balance in Yeso sika deer (Cervus nippon yesoensis) fed mixtures of hay and wheat bran or soybean meal. Hokkaido Animal Science and Agriculture Society 40:22–26. (In Japanese.) Masuko, T., K. Souma, K. Miyairi, T. Komatsu, and Y. Ishijima. 1999a. Intake, digestibility and nitrogen balance of sasa (Sasa senanensis) in Yeso sika deer (Cervus nippon yesoensis). Hokkaido Animal Science and Agriculture Society 41:72–75. (In Japanese.)

5 Nutritional Physiology of Wild and Domesticated Japanese Sika Deer

81

Masuko, T., and K. Souma. 1999b. Studies of nutrition in Yeso deer feeding. Hokkaido Animal Science and Agriculture Society 41:1–9. (In Japanese.) Masuko, T., T. Taji, K. Nakamura, M. Sekikawa, and S. Masaoka. 2000. Deer products utilization handbook. Zennipon Youroku Kyoukai, Tokyo, Japan. (In Japanese.) Masuko, T., K. Souma, R. Kitahara, N. Sawada, K. Miyairi, and Y. Ishijima. 2001. Chemical composition and in vitro dry matter digestibility of woody plants eaten by Yeso sika deer (Cervus nippon yesoensis) during from winter to spring season. Hokkaido Animal Science and Agriculture Society 43:41–47. (In Japanese.) Matoba, K., T. Nakamura, S. Sato, T. Watanabe, M. Odajima, K. Usa, and H. Tamate. 1987. Feed utilization in sika deer. Kawatabi Farm Report 3:158–159. (In Japanese.) McCullough, Y. 1979. Carbohydrate and urea influences on in vitro deer forage digestibility. Journal of Wildlife Management 43:650–656. Mimura, K., and T. Morita. 1990. Kachiku Kanrigaku, 6th edition. Yokendo Limited, Tokyo, Japan. (In Japanese.) Miyazaki, A., S. Kasagi, and T. Mizuno. 1984. Digestibility of Zoysia-type grass by Japanese deer. Japanese Journal of Zootechnical Science 55:661–669. Morimoto, H. 1989. Revised nutrition, 17th edition. Yokendo Limited, Tokyo, Japan. (In Japanese.) Nagy, J. G., and W. L. Regelin. 1975. Comparison of digestive organ size of three deer species. Journal of Wildlife Management 39:621–624. Nakai, S. 1988. Methods of identifying 100 familiar poisonous herbs. Kinensha, Tokyo, Japan. (In Japanese.) Nakamura, R. 1977. New feed study, Volume 1. General remarks. Chikusan Shuppansha, Tokyo, Japan. (In Japanese.) Odajima, M., Y. Kajita, K. T. Nam, S. R. Lee, H. Senge, K. Katoh, Y. Shoji, M. Ohta, and Y. Sasaki. 1991. Seasonal changes in food passage and digestibility in Japanese sika deer (Cervus nippon) and sheep under restricted feeding condition. Animal Science Technology (Japan) 62:308–313. (In Japanese.) Odajima, M., K. Nakajima, Y. Ohtomo, S. Oda, Y. Shoji, K. Katoh, M. Ohta, and Y. Sasaki. 1993. Year-long changes in food intake and body weight in group-fed Japanese sika deer (Cervus nippon). Animal Science Technology (Japan) 64:421–423. (In Japanese.) Okubo, T., H. Hirota, Y. Takasaki, A. Ueno, S. Saiga, K. Ataku, H. Kobayashi, T. Shimada, M. Kikuchi, and I. Nakanishi. 1990. Grassland Study 24:28–32. (In Japanese.) Okamoto, S. 1970. Environment and physiology of livestock and poultry. Yokendo Limited, Tokyo, Japan. (In Japanese.) Ramanzin, M., L. Bailoni, and S. Schiavon. 1997. Effect of forage to concentrate ratio on comparative digestion in sheep, goats and fallow deer. Animal Science 64:163–170. Sawada, Y. 1994. Study on improvement and utilization of pasture, and on evaluation of grazing characteristics of temperate grasses. Grassland Science 28:1–5. (In Japanese.) Secretariat of Agriculture, Forestry, and Fisheries Research Council, Ministry of Agriculture, Forestry, and Fisheries. 1991. Biomass change plan—utilization of rich biological resources. Korin, Tokyo, Japan. (In Japanese.) Secretariat of Agriculture, Forestry, and Fisheries Research Council, Ministry of Agriculture, Forestry, and Fisheries. 1995. Japan standard feed components table (1995). Central Association of Livestock Industry, Tokyo, Japan. (In Japanese.) Shiraishi, T., Y. Nagaguchi, S. Hayama, N. Tokita, K. Furubayashi, and M. Yamane. 1996. Seasonal changes of body weight and food intake in captive sika deer (Cervus nippon). Japanese Journal of Zoo and Wildlife Medicine 1:119–124. (In Japanese.) Souma, K., T. Masuko, and Y. Ishijima. 1994. General behavior of the sika deer (Cervus nippon) under housing. Hokkaido Animal Science and Agriculture Society 36:57–62. (In Japanese.) Souma, K., Y. Honda, T. Masuko, and Y. Ishijima. 1995. The palatability of hay, silage, and sasa (Sasa senanensis) on the Yeso sika deer (Cervus nippon yesoensis). Hokkaido Animal Science and Agriculture Society 37:28–34. (In Japanese.)

82

T. Masuko, K. Souma

Souma, K., T. Masuko, R. Kitahara, and Y. Ishijima. 1996. Intake of wild herbaceous and arborous plant in the Yeso sika deer (Cervus nippon yesoensis) and its chemical composition. Hokkaido Animal Science and Agriculture Society 38:98–104. (In Japanese.) Souma, K., T. Masuko, Y. Kobayashi, and Y. Ishijima. 1998. Seasonal alteration of hay intake in the Yeso sika deer (Cervus nippon yesoensis). Hokkaido Animal Science and Agriculture Society 40:27–30. (In Japanese.) Souma, K., T. Masuko, K. Miyairi, R. Kitahara, T. Komatsu, and Y. Ishijima. 1999. Changes in chemical composition and in vitro dry matter digestibility of sasa (Sasa senanensis) on different growth stages. Hokkaido Animal Science and Agriculture Society 41:76–79. (In Japanese.) Takatsuki, S. 1986. Analysis of gastric contents of dead deer in mass deaths in Nikko in 1984. Hakkenpo, Tochigi Prefecture 4:15–22. (In Japanese.) Takatsuki, S. 1992. Deer in the north—ecology of deer, sasa, and snow. Dobutsusha, Tokyo, Japan. (In Japanese.) Takatsuki, S. 1994. Report on the investigation of deer in Mt. Goyo (1992–1993). Pages 45–59. Nature Conservation Division, Environmental Health Department, Iwate Prefecture, Japan. (In Japanese.) Terada, F., R. Tano, K. Iwasaki, and T. Haryu. 1987. Comparison of nutritive values among cattle, sheep and goats fed the same diets. Japanese Journal of Zootechnical Science 58:131–137. (In Japanese.) Tomioka, K. 1996. Creatures that know time. Shokabo, Tokyo, Japan. (In Japanese.) Toyooka, K., A. Sato, and M. Ishizuka. 1983. Distribution map of the sasa group in Hokkaido: Explanatory note. Hokkaido Branch Forest Products Research Institute, Hokkaido, Japan. (In Japanese.) Tsuda, T., I. Ito, T. Hoshino, S. Nishiguchi, Y. Sasaki, and M. Ohta. 1987. Physiological ecological study of productivity of sika deer. Report of a grant-supported study of meat. Itoh Science Foundation, Tokyo 5:147–153. (In Japanese.) Tsuda, T. 1990. Livestock physiology, 9th edition. Yokendo Limited, Tokyo, Japan. (In Japanese.) Uchida, H., S. Ikeda, M. Ishida, T. Inoue, and T. Takeda. 2001. Growth characteristics of artificially reared sika deer (Cervus nippon). Animal Science Journal 72:461–466. Yamane, M., S. Hayama, T. Shiraishi, I. Yoshimura, and K. Frubayashi. 1997. Body weight changes of sika deer (Cervus nippon) under different nutritional conditions—comparison between free-ranging deer and captive deer from Tanzawa mountains. Japanese Journal of Zoo and Wildlife Medicine 2:59–66. (In Japanese.) Yokohama, M., Y. Kameyama, T. Masuko, T. Komatsu, R. Hashizume and Y. Ishijima. 1991. An investigation on the merits of the animal resouces of the Yeso sika deer (Cervus nippon yesoensis). Journal of Agricultural Science, Tokyo Nogyo Daigaku 35:185–191. (In Japanese.)

Chapter 6

Endocrinology of Sika Deer Kiyoshi Yamauchi and Yukiko Matsuura

Abstract There have been few studies of endocrinology of sika deer in Japan, although several studies have been conducted on reproductive physiology. In recent years, we have obtained basic information on sika deer endocrinology by applying fecal steroid analysis as a noninvasive method. In this chapter we explore hormonal changes during the estrous cycle and pregnancy in female sika deer and in addition consider what is known about “silent” ovulation. For male sika deer, we provide the annual testosterone pattern and relate it to aggressive behavior.

Introduction It is an indisputable fact that reproductive organs and mechanism are regulated by the endocrine system in which hormones play a key role. The reproductive physiology of mammals is controlled by the hypothalamo-hypophysial-gonadal axis. The secretion of the gonadotropic hormones (LH and FSH) from the anterior pituitary is controlled by gonadotropin releasing hormone (GnRH) produced in the hypothalamus. Gonadotropins stimulate reproductive organs to secrete steroid hormones, which in turn regulate reproductive physiology such as development and maintenance of sex characteristics, estrus, ovulation, and gestation. By monitoring the hormonal changes, we can determine the length of the estrous cycle and the gestation period. Additionally, steroids released from the gonads not only regulate the function of reproductive organs but also effect secondary sex characteristics, such as antlers, pelage, and behavior. Much research has been conducted into the relationship between testosterone and dominance hierarchy or reproductive success of several species, as it was suggested that testosterone has a direct inductive effect on aggressive and sexual behavior. It is of interest to understand the function of testosterone and its relation to deer behavior. This chapter will review primarily the basic information of endocrinology in sika deer and compare it to other Cervidae, as well as provide new findings based on fecal steroid analysis.

D. R. McCullough et al. (eds.), Sika Deer: Biology and Management of Native and Introduced Populations, DOI: 10.1007/978-4-431-09429-6_6, © Springer 2009

83

84

K.Yamauchi, Y. Matsuura

Fecal Steroid Analysis Several studies have been conducted on the reproductive physiology of sika deer. Most of the subjects used in anatomical and morphological analyses were individuals shot for population control programs, killed in a traffic accident, or hunted for sport (Yamauchi et al. 1982, 1983, 1984; Koizumi 1991; Suzuki et al. 1992a, b, 1996, 2001b; Suzuki and Ohtaishi 1993; Kameyama et al. 2000;Yokoyama et al. 2000a, b, 2001a, b). Compared to ecological information for this species, however, very little is known about the endocrine system. Measurements of hormone levels in the blood have been the most popular technique for the estimation of reproductive status. In intractable species such as Cervidae, however, blood sampling is extremely troublesome because it requires capture, immobilization, and/or restraint. Further, the chemical immobilizations and restraint associated with blood sampling are known to affect the hormone levels in blood (Wesson et al. 1979; Seal and Bush 1982) and thus could compromise the reliability of the endocrine information. The possibility of accidental death by immobilization and restraint also has been reported (Kaji et al. 1991; Suzuki et al. 2001a). Since the 1980s, fecal steroid hormone measurement has been used in several captive and wild species to monitor their reproductive profiles noninvasively (Bamberg et al. 1991; Lasley and Kirkpatrick 1991; Wasser et al. 1991, 1994; Kirkpatrick et al. 1992, 1993; Monfort et al. 1993; Hirata and Mori 1995; Schwartz et al. 1995). As this method allows sampling without disturbing the focal animals, detailed behavioral observations have been applied to several species. In sika deer, this method for fecal progesterone and testosterone analyses were developed and confirmed that fecal concentrations of gonadal steroid show parallel changes to those in plasma. Moreover cyclical fluctuations during the estrous cycle and a pregnancy-related increase in fecal progesterone in the doe, as well as an annual fecal testosterone profiles in the buck, were revealed (Yamauchi et al. 1999, 1997).

Seasonal Breeding Since growth of vegetation, especially grass, fluctuates with seasonal variation, the breeding of carnivores and herbivores is limited according to the seasonal change. In most of these animals the period of parturition corresponds to spring, when ambient temperature rises and food resources increase, so that newborn offspring will grow and prosper. Animals have mechanisms to detect the change of day length, and melatonin plays an important role in relation to photoperiodic modification. Briefly, the information on the change of day length is mediated by the pineal gland through the secretion of melatonin which reaches peak levels during the night (Bubenik and Smith 1987). When melatonin reaches the medial basal hypothalamus (MBH) directly or indirectly, the hypothalamic GnRH pulse generator activity is controlled and, consequently, the pattern of gonadotropin secretion is modified (Mori 1992).

6

Endocrinology of Sika Deer

85

No information is yet available regarding melatonin in sika deer. In some other cervids, functional mechanisms of melatonin for antler cycle or reproductive development have been elucidated by administration, implants, and intake of melatonin (Bubenik 1983; Adam and Atkinson 1984; Whitsett et al. 1984; Webster and Barrell 1985; Adam et al. 1986, 1989; Asher et al. 1987, 1988; Loudon and Curlewis 1988; Webster et al. 1991). From these results it was concluded that the effect of reduced photoperiod is mediated by melatonin. Melatonin influences the seasonal secretory pattern of prolactin which then regulates secretion of pituitary gonadotropins (Bubenik et al. 1985; Bubenik 1990). The same mechanism probably exists in sika deer. In many countries farming deer for the production of venison and/or antler velvet these studies have been conducted, as seasonal breeding could be manipulated by melatonin. Very few sika deer are farmed in Japan, and thus little is understood about melatonin in sika deer. Deer living in temperate latitudes are seasonal breeders (Barrell et al. 1985; Kelly et al. 1985; Lincoln 1985, 1992; Loudon and Brinklow 1992). Sika deer are short-day seasonal breeders as well as other cervid species, and the breeding season begins in early autumn (Miura 1983, 1984a, b). It is expected that luteinizing hormone (LH) or follicle-stimulating hormone (FSH) concentrations increase during the breeding season via secretion of melatonin. Yamaji et al. (1994) observed gonadotropic cells in the pituitary by immunohistochemical methods in wild sika deer (C. n. yesoensis) which were collected from Hokkaido in August (the last month of the nonbreeding season) and October (the peak of rutting season). They found that LH and FSH cells showed seasonal changes; the immunostaining reactivity of LH and FSH cells in October was more intense than that in August in both sexes, and these cells were abundant even in August. These results indicated that gonadotropic hormones in sika deer are activated before breeding season as with other deer species. The seasonal changes of gonadotropic hormones were investigated in roe deer (Schams et al. 1980), white-tailed deer (Mirarchi et al. 1978; Plotka et al. 1980) and red deer (Lincoln 1985), showing that gonadotropic hormones are increasing several months before peak levels of testosterone are achieved and began to decline already some time before the rutting season. C. n. yesoensis also showed a similar tendency in LH and testosterone patterns (Kameyama et al. 2002). Determination of seasonal breeding of deer species not only assists in fundamental research on evolution or environmental adaptation, but could be also utilized in industries for the production of venison and antler velvet.

Female Sika Deer Estrous Cycle Like most cervid species, the sika deer is polyestrous (Sadleir 1987). There are a variety of lengths of estrous cycle in deer species. Though the length of estrous cycle in sika deer had been estimated to be around 20 days based on only behavioral

86

K.Yamauchi, Y. Matsuura

observation, the result of detailed research had not been reported. Monitoring of estrous cycle based on hormonal profiles is difficult to apply in sika deer, because repeated blood sampling is needed. To cope with this difficulty, a noninvasive method for monitoring the reproductive function has been developed by using fecal steroid analysis (Yamauchi et al. 1997; Matsuura et al. 2004a). Additionally, detailed behavior observation was also conducted and its results were compared with hormonal changes. Figure 6.1 Shows the pattern of cyclic fluctuation of two nonpregnant female deer (C. n. centralis) during the breeding season. All estrous behaviors were observed at the time when fecal progesterone concentrations were low. Consequently, the date of the nadirs of progesterone concentration is considered the follicular phase. The mean length of estrous cycle estimated based on the fecal progesterone concentrations (days between the nadirs) and that based on the estrous behavior (days between the first sign of estrus) are shown in Table 6.1. The ranges as well as the mean lengths based on the two parameters coincided well, and the estrous cycle was

80 #1 60

Fecal Progesterone (ng/g)

40

20

0 #2 40

30

20

10

Oct

Nov

Dec

Jan

Fig. 6.1 Changes of fecal progesterone concentration in two female sika deer (C. n. centralis) during the breeding season. In this study, fecal samples were collected every day or every other day during the period between September 1994 and February 1995 to monitor estrous cyclicity at the zoo in Urawa city (35°51′N, 139°39′E), Saitama Prefecture, Japan. Behavioral observations were conducted on the same day as fecal sampling. Behavioral changes including an active approach to the male deer, frequent urination, vaginal discharge, and mating were regarded as a sign of estrus. Arrows indicate the day of behavioral estrus (from Yamauchi et al. 1997).

6

Endocrinology of Sika Deer

87

Table 6.1 Comparison of estrous cycle lengths estimated from fecal progesterone profile and those from behavior changes in two female sika deer (from Yamauchi et al. 1997). Hind Based on Mean ± SEM (n) Range (days) #1 #2

Fecal Progesterone Estrous Behavior Fecal Progesterone Estrous Behavior

13.0 ± 0.9 (6) 12.8 ± 0.7 (10) 12.5 ± 0.3 (6) 12.4 ± 0.4 (9)

10–17 10–16 10–14 10–14

shown to be around 12 to 13 days. The pattern of gradual increase followed by an abrupt decrease was similar to fecal progesterone profiles reported in other ruminant species and cervid species (Hirata and Mori 1995; Schwartz et al. 1995). The cycle length was considerably shorter compared to the goat and cow (21 days), ewe (16 days), and deer species such as moose (24–25 days) (Schwartz and Hundertmark 1993), fallow deer (22 days) (Asher 1985), red deer (18–21 days) (Guiness et al. 1971; Adam et al. 1985; Kelly et al. 1985), white-tailed deer (25–30 days) (Plotka et al. 1977a), caribou (24 days) (Bergerud 1975), and Père David’s deer (19 days) (Curlewis et al. 1988). Duration of estrous cycle in Formosan sika deer (Cervus nippon taiouanus), which belong to Cervus nippon, was investigated during breeding season based on the cyclic basal serum progesterone levels and the observations of the cyclic signs of swollen and moistened external genitalia. Mean duration of estrous cycle was estimated to be 19.3 ± 1.8 days with a range of 17–21 days (Liu et al. 2002). This result was also longer than sika deer (C. n. centralis).

“Silent” Ovulation The length of ovarian cycle in C. n. yesoensis has been examined based on fecal progesterone measurement and behavioral observation of copulation and estrous symptoms such as frequent urination, tail lifting, and changes in external genitalia including color, swelling, and vaginal discharge (Matsuura et al. 2004a). From this study, it was revealed that the days between ovulations, which occurred from the first ovulation to the ovulation accompanied by the first copulation, was 9.8 ± 4.6 days, ranging from five to 24 days (Fig. 6.2), and that ovulation without estrus/ copulation occurred early in the breeding season in most cases. This “silent” ovulation usually has a shorter and more irregular luteal phase than normal estrus (Asher 1985; Curlewis et al. 1988); e.g., 6–12 days in red deer (Asher et al. 2000), within 11 days in elk (Cook et al. 2001), 2–5 days in Eld’s deer (Monfort et al. 1990), and 8–9 days in black-tailed deer (Thomas and Cowan 1975). A detailed study of dairy cows revealed that the average interval between the first (silent) and second ovulation was 12.7 ± 1.7 days, ranging from 6–32 days, and it was shorter than normal intervals. The length obtained in C. n. yesoensis is similar to those lengths and interpreted as the interval indicating luteal phase during “silent” ovulation in sika deer. “Silent” ovulation occurred before the first behavioral estrus, and the increasing length also occurred as the season progressed in red deer (Asher 1985). Since most females in C. n. yesoensis conceived in the early stage of the breeding season

88

K.Yamauchi, Y. Matsuura

Fig. 6.2 Ovulation patterns revealed by fecal progesterone profiles, estrous symptoms, and dates of copulation in 12 female sika deer (C. n. yesoensis). Fecal samples were collected daily, every other day, or every third day from October 3 to December 30, 2000 at a deer farm in Hokkaido (143°27′E, 42°56′N), Japan. Behavioral observations were made during the daytime, and estrous symptoms, copulation and parturition dates were recorded. In this study, behavioral changes of frequent urination and tail lifting and changes in external genitalia including color, swelling, and vaginal discharge were defined as estrous symptoms. Open circles indicate the dates of nadirs of progesterone concentration (ovulation) without any estrous symptoms or copulation, double circles indicate the dates of nadirs with estrous symptoms, and closed circles indicate the dates of nadirs with copulation. Arrowheads indicate the dates of copulation. The last copulation (conception) dates varied by up to 22 days, ranging from October 26 to November 16 (the time interval indicated by the line at the bottom marked by an*), except for female no. 5 (from Matsuura et al. 2004a).

(Matsuura et al. 2004a), it was suggested that the interval between ovulations (9.8 ± 4.6 days) was shorter than mean estrous length (12–13 days) estimated by Yamauchi et al. (1997) in C. n. centralis. Concerning the length of estrous cycle in sika deer, we need to elucidate accurate length and the estrous length in other subspecies by using a vasectomized male and monitoring the receptivity of females, although many animals are required for such research.

6

Endocrinology of Sika Deer

89

Progesterone Profile in Pregnant Sika Deer Figure 6.3 shows changes of fecal progesterone concentrations in pregnant and nonpregnant sika deer (C. n. centralis). Fecal progesterone concentrations of nonpregnant deer were maintained at low levels throughout a year, while that of the pregnant deer increased immediately after mating. There was a conspicuous difference in progesterone profiles between pregnant and nonpregnant deer. Annual profiles of serum progesterone were investigated in red deer (Kelly et al. 1982) and white-tailed deer (Plotka et al. 1977b). Their progesterone concentrations showed basal levels during nonbreeding season, increased rapidly after mating, and then were sustained at the high level. Several days before parturition, there was a marked decrease in progesterone levels. A similar pattern of fecal progestagen was also reported in moose (Schwartz et al. 1995), in which fecal progestagen rapidly increased following conception and maintained high levels after eight weeks of gestation when fecal progestagen levels exceeded peak luteal phase concentration. Annual changes of fecal progesterone concentrations were measured in C. n. yesoensis by K. Jinma et al. in the Laboratory of Theriogenology at Hokkaido University (personal communication 1995); fecal progesterone profiles in pregnant deer were similar to those in pregnant moose (Schwartz et al. 1995).

Fig. 6.3 Changes of fecal progesterone concentrations in three female sika deer (C. n. centralis) (modified from Yamauchi et al. 1997).

90

K.Yamauchi, Y. Matsuura

Regarding the gestation period in sika deer, estimated results based only on behavioral observation have been described in a few papers in Japan (230–236 days (Iimura 1980)), (228 days (Miura 1984b)). To clarify exact gestation period further, conception dates were decided by hormonal profiles and behavioral observation (Matsuura et al. 2004a). Gestation period in 11 Hokkaido sika deer females whose conception were known by hormonal profiles was 224.5 ± 3.7 days, ranging from 216–228 days. The gestation period of sika deer is similar to that of the moose, 216–240 days (Schwartz and Hundertmark 1993) and the caribou, 208–240 days (Leader-Williams 1988), but slightly longer than that of the white-tailed deer, 196– 213 days (Verme 1969) and slightly shorter than that of the red deer, 240–252 days (Haigh 2001). Although gestation period was not related to the fawn’s sex or the female’s age, young females and those in poor condition appear to have a higher probability of prolonged gestation periods following a severe winter (Matsuura et al. 2004b). As to gestation period in sika deer, further research is needed to determine the length in other subspecies, because their body size is influenced by geographical and ecological factors (Ohtaishi 1986).

The Timing of Conception in Sika Deer The timing of conception in C. n. yesoensis was estimated to occur over about 100 days from fetal analysis of culled deer (Suzuki et al. 1996). In Hyogo prefecture, the duration of the calving period was estimated 83 days (Koizumi 1991). The study of captive sika deer showed that lactating or young females were more likely to conceive later than other females (Matsuura et al. 2004b). The timing of breeding is important for reproductive success in wild animals, because the parturition period influences fawn or infant mortality (Clutton-Brock et al. 1982) and subsequent fertility rates (Clutton-Brock et al. 1983). In general, high ranking male deer have more opportunities for copulation than low ranking male deer. In red deer, rut of dominant stags precedes that of bachelor or younger groups (Lincoln and Guiness 1973). Consequently, it was thought that time differences of rut in male deer caused the 100-day variation of conception dates. Fecal testosterone concentrations were measured from captive male sika deer (C. n. nippon) of various body and antler size. From this result, the peak of testosterone levels occurred around the same time among all deer (K. Yamauchi et al. unpublished data, 2001), regardless of dominance hierarchy position or extreme difference in body and antler size. In other deer species it has been reported that late conception is due to a failure to conceive at first estrus (Bergerud 1975; Guiness et al. 1978) or a delay in the onset of ovulation (Cook et al. 2001). In C. n. yesoensis, delay in the onset of ovulation is not likely the factor affecting the delayed conception (Matsuura et al. 2004a). Some females had repeated “silent” ovulation early in the breeding season creating a three- to four-cycle variation in the timing of conception. Furthermore, a few females conceived very late in the breeding season probably due to uterine disease or the failure

6

Endocrinology of Sika Deer

91

to maintain the pregnancy. Therefore, it appears that the variation of conception dates is in most cases caused by the female.

Male Sika Deer Annual Testosterone Rhythm Concerning the endocrinology of males in deer species, there have been many detailed studies of antler development, mostly investigating the role of testosterone (Bubenik 1990; Mourik and Stelmasiak 1990). Plasma testosterone concentrations were measured in wild sika deer (C. n. yesoensis) and peak testosterone levels were detected in late October and early November (Suzuki et al. 1992a). In another study, blood sampling at monthly intervals was carried out on captive sika deer (C. n. yesoensis), and testosterone concentrations peaked in September and October (Kameyama et al. 2000). Moreover, the annual testosterone profile has been measured by fecal analysis (Yamauchi et al. 1997). There was a clear annual change of fecal testosterone concentration; it reached the peak in late September and then decreased rapidly by November, around the estimated time for the breeding season. This relatively low concentration was maintained during the nonbreeding season (Fig. 6.4). In other cervid species, annual testosterone concentrations in serum/ plasma have been examined in white-tailed deer (McMillin et al. 1974; Mirarchi et al. 1978), wapiti (Haigh et al. 1984), roe deer (Schams and Barth 1982; Sempere and Lacroix 1982), fallow deer (Asher et al. 1989), and red deer (Barrell et al. 1985; Lincoln 1985; Bubenik 1990). Those testosterone concentrations attained a peak in the early stage of the breeding season and began to decline during the rut. Seasonal changes of reproductive organs such as testis or seminiferous tubules were examined in C. n. yesoensis (Suzuki et al. 1992a). In this study, spermatogenic activity reached its height in late October and had already begun to decline in late December, and spermatogenesis had stopped by February. However as a very few spermatozoa were observed in February, it seemed that fertility might be maintained in this season. Similar results were found in white-tailed deer (McMillin et al. 1974) and wapiti (Haigh et al. 1984), in which the increase in testosterone correlated with increasing testicular size or scrotal circumference. Yamauchi et al. (1997) found that reproductive behavior was maintained until February when testosterone levels were low (Fig. 6.4). This indicates that reproductive behavior coincides well with annual testicular activities just described above. From these results, it was suggested that temporary secretion of testosterone might be responsible for triggering the aggressive or sexual behaviors in the male deer. However, it appears more likely that a secretion of a large quantity of testosterone may be important for the initiation of the reproductive activities rather than for the maintenance of it, since fecal testosterone rapidly decreased in early breeding season.

92

K.Yamauchi, Y. Matsuura

Antler Cycle Annual testosterone rhythms in sika deer have relevance to annual antler cycles. As shown in Fig. 6.4, antler casting was observed in April when fecal testosterone reached the lowest level and velvet shedding occurred in late August when fecal testosterone began to increase. Similar results were shown in other cervid species (McMillin et al. 1974; Schams and Barth 1982; Barrell et al. 1985; Lincoln 1985; Bubenik and Bubenik 1987; Bubenik 1990; Mourik and Stelmasiak 1990). In general, antler cycles, namely calcification, cleaning, casting, and regeneration, are regulated by the changes in testosterone. In other cervid species, it is also shown that prolactin or insulin-like growth factor 1 (IGF-1), which is under photoperiodic changes, are implicated in antler growth (Mirarchi et al. 1978; Barrell et al. 1985; Lincoln 1985; Bubenik 1990; Mourik and Stelmasiak 1990; Suttie and Fennessy 1992; Li et al. 1999, 2003).

Testosterone and Behavior in Sika Deer Wild male sika deer begin to show aggressive behavior in early autumn to obtain as many females as possible in their territories. Testosterone secreted from testes has been considered to play an essential role in this aggressive behavior (Lincoln et al. 1970; Hart 1974). By experimental manipulation such as castration or implantation with testosterone, a close relationship between androgen levels and social dominance has been reported in ungulates and primates (Bouissou 1983). Similarly, dominant males tend to have higher testosterone levels in red deer (Lincoln et al. 1972).

Antler Casting

Fecal Testosterone (ng / g)

Antler 2500

Antler Regrowth

Velvet

Velvet Shedding

2000 1500 Antler Casting

1000

Velvet Shedding

500 0 9

10

11

12

1

2

3

4

5

6

7

8

(Month)

Fig. 6.4 Annual pattern of fecal testosterone concentration in a male sika deer (C. n. centralis). The shaded area represent the breeding season during which the female deer showed regular estrous cyclicity (redrawn from Yamauchi et al. 1999).

6

Endocrinology of Sika Deer

93

It is known that deer antlers are visual signals of male rank (Bubenik 1990). For example, changing social rank alters antler growth in fallow deer (Bartos and Losos 1987). In sika deer (C. n. centralis), large antlers were related to dominance (Miura 1984a). To elucidate the relation between dominance, antler characteristics, and testosterone levels, C. G. Silva et al. used the fecal testosterone analysis method in wild male sika deer (C. n. centralis) (unpublished data 2001). They determined social rank among the males by observation and measured antler length or weight and fecal testosterone levels at the same time in summer and in autumn. Consequently, it was shown that social rank was influenced by antler characteristics rather than fecal testosterone levels. Further research was conducted to clarify the relationship between fecal testosterone concentration and aggressive behaviors in sika deer. K. Yamauchi et al. collected feces weekly from captive male sika deer (C. n. nippon) which have various body and antler size throughout the year and observed aggressive behaviors among them (unpublished data 2001). From this result, the peak of testosterone levels occurred around the same time in all deer. Although there was not a significant difference in fecal testosterone concentrations between adult male and young male deer during rutting season from September to November, aggressive behaviors were extremely different. Taken together, it appears that, although testosterone induces aggressive or sexual behavior, it does not always coincide with social hierarchy. Moreover, K. Yamauchi et al. have found that fecal testosterone concentrations of adult male deer were significantly higher than those of young male deer in the period from December to March of the following year when normal rut terminated. However, there was a lot of aggressive behavior between them, and young male deer were attacked frequently compared with adult male deer. That is, adult male deer kept their behavioral advantage by their bigger body and antler size, so that testosterone was also kept at high levels. It was suggested that testosterone secretion in young male deer was suppressed because they suffered many attacks. For example, when two male macaques fight, the testosterone levels of the winning male elevate above those of the defeated male (Bernstein et al. 1974). In humans, rise or fall in testosterone levels among players or fans was affected by the winning or losing of sporting events (Booth et al. 1989; Bernhardt et al. 1998). The fluctuation of testosterone levels with physical and psychological effects, as mentioned above, was discussed in a recent report. Testosterone levels in all male sika deer indicate a clear annual rhythm in the long term according to short-day seasonal breeding; however, in the short term, they would vary with the status of population or habitat and result in various subsequent behaviors.

Subject of Future Investigation Japan is a long landmass from south to north, and its climate ranges dramatically. There is a geographical variation among subspecies in aspects such as body size, antler shape, social structure, and eating habits (Miura 1986; Ohtaishi 1986; Takatsuki 1992). Currently, six species of sika deer in Japan exist (Ohtaishi 1986).

94

K.Yamauchi, Y. Matsuura

Although only a few studies have been reported on endocrine information in sika deer, we are gradually clarifying them by means of fecal steroid analysis without disturbing the focused animals. Because marked variations in morphology and ecology are shown, we need to investigate the estrous cycle or gestation period in detail. Furthermore, as reproductive behavior could vary, hormonal secretion is also likely to differ accordingly group by group, as mentioned above in Testosterone and Behavior in Sika Deer. Recently, cortisol levels have been measured as a useful index of stress. By noninvasive methods such as fecal or urinary analysis, the estimation of habitat and the relation to social dominance (Creel et al. 1996; Robbins and Czekala 1997; Bartos et al. 1998; Dehnhard et al. 2001; Huber et al. 2003) were investigated. By applying the fecal steroid method to a population of sika deer, we look forward to contributing to a wide range of management or biological studies in the future.

Literature Cited Adam, C. L., and T. Atkinson. 1984. Effect of feeding melatonin to red deer (Cervus elaphus) on the onset of the breeding season. Journal of Reproduction and Fertility 72:463–466. Adam, C. L., C. E. Moir, and T. Atkinson. 1985. Plasma concentration of progesterone in female red deer (Cervus elaphus) during the breeding season, pregnancy and anoestrus. Journal of Reproduction and Fertility 74:631–636. Adam, C. L., C. E. Moir, and T. Atkinson. 1986. Induction of early breeding in red deer (Cervus elaphus) by melatonin. Journal of Reproduction and Fertility 76:569–573. Adam, C. L., C. E. Moir, and P. Schiach. 1989. Melatonin can induce year-round ovarian cyclicity in red deer (Cervus elaphus). Journal of Reproduction and Fertility 87:401–408. Asher, G. W. 1985. Oestrous cycle and breeding season of farmed fallow deer, Dama dama. Journal of Reproduction and Fertility 75:521–529. Asher, G. W., A. M. Day, and G. K. Barrell. 1987. Annual cycle of liveweight and reproductive changes of farmed male fallow deer (Dama dama) and the effect of daily oral administration of melatonin in summer on the attainment of seasonal fertility. Journal of Reproduction and Fertility 79:353–362. Asher, G. W., G. K. Barrell, J. L. Adam, and L. D. Staples. 1988. Effects of subcutaneous melatonin implants on reproductive seasonality of farmed fallow deer (Dama dama). Journal of Reproduction and Fertility 84:679–691. Asher, G. W., A. J. Peterson, and J. J. Bass. 1989. Seasonal pattern of LH and testosterone secretion in adult male fallow deer, Dama dama. Journal of Reproduction and Fertility 85:657–665. Asher, G. W., K. T. O’Neill, I. C. Scott, B. G. Mockett, and M. W. Fisher. 2000. Genetic influences on reproduction of female red deer (Cervus elaphus) (1) seasonal luteal cyclicity. Animal Reproduction Science 59:43–59. Bamberg, E., E. Mostle, M. Patzl, and G. J. King. 1991. Pregnancy diagnosis by enzyme immunoassay of estrogen in feces from nondomestic species. Journal of Zoo and Wildlife Medicine 22:73–77. Barrell. G. K., P. D. Muir., and A. R. Sykes. 1985. Seasonal profiles of plasma testosterone, prolactin, and growth hormone in red deer stags. Pages 185–190 in P. F. Fennessy and K. R. Drew, editors, Biology of deer production. Royal Society of New Zealand Bulletin 22. Wellington, New Zealand. Bartos, L., and S. Losos. 1987. Response of antler growth to changing rank of fallow deer buck during the velvet period. Canadian Journal of Zoology 75:1934–1939.

6

Endocrinology of Sika Deer

95

Bartos, L., E. Reyes, D. Schams, G. Bubenik, and A. Lobos. 1998. Rank dependent seasonal levels of IGF-1, cortisol and reproductive hormones in male pudu (Pudu puda). Comparative Biochemistry and Physiology, Part a 120:373–378. Bergerud, A. T. 1975. The reproductive season of Newfoundland caribou. Canadian Journal of Zoology 53:1213–1221. Bernhardt, P. C., J. M. Dabbs, Jr., J. A. Fielden, and C. D. Lutter. 1998. Testosterone changes during vicarious experiences of winning and losing among fans at sporting events. Physiology and Behavior 65:59–62. Bernstein, I. S., R. M. Rose, and T. P. Gordon. 1974. Behavioral and environmental events influencing primate testosterone levels. Journal of Human Evolution 3:517–525. Booth, A., G. Shelley, A. Mazur, G. Tharp, and R. Kittok. 1989. Testosterone, and winning and losing in human competition. Hormones and Behavior 23:556–571. Bouissou, M. F. 1983. Androgens, aggressive behaviour and social relationships in higher mammals. Hormone Research 18:43–61. Bubenik, A. B., And S. P. Smith. 1987. Circadian and circannual rhythm of melatonin in plasma of male white-tailed deer and the effect of oral administration of melatonin. Journal of Experimental Zoology 241:81–89. Bubenik, G. A. 1983. Shift of seasonal cycle in white-tailed deer by oral administration of melatonin. Journal of Experimental Zoology 225:155–156. Bubenik, G. A. 1990. Neuroendocrine regulation of the antler cycle. Pages 265–297 in G. A. Bubenik and A. B. Bubenik, editors, Horns, pronghorns, and antlers. Springer, New York, USA. Bubenik, G. A., And A. B. Bubenik. 1987. Recent advances in studies of antler development and neuroendocrine regulation of the antler cycle. Pages 99–109 in C. M. Wemmer, editor, Biology and management of the Cervidae. Smithsonian Institution Press, Washington, DC, USA. Bubenik, G. A., D. Schams, and J. F. Leatherland. 1985. Seasonal rhythm of prolactin and its role in the antler cycle of white-tailed deer. Pages 257–262 in P. F. Fennessy and K. R. Drew, editors, Biology of deer production. Royal Society of New Zealand Bulletin 22. Wellington, New Zealand. Clutton-Brock, T. H., F. E. Guiness, and S. D. Albon. 1982. Red deer: The behavior and biology of two species. University of Chicago Press, Chicago, Illinois, USA. Clutton-Brock, T. H., F. E. Guiness, and S. D. Albon. 1983. The costs of reproduction to red deer hinds. Journal of Animal Ecology 52:367–383. Cook, R. C., D. L. Murray, J. G. Cook, P. Zager, and S. L. Montfort. 2001. Nutritional influences on breeding dynamics in elk. Canadian Journal of Zoology 79:845–853. Creel, S., N. M. Creel, and S. L. Montfort. 1996. Social stress and dominance. Nature 379:212. Curlewis, J. D., A. S. Loudon, and A. P. Coleman. 1988. Oestrous cycles and the breeding season of the Père David’s deer hind (Elaphurus davidianus). Journal of Reproduction and Fertility 82:119–126. Dehnhard, M., M. Clauss, M. Lechner-Doll, H. H. Meyer, and R. Palme. 2001. Noninvasive monitoring of adrenocortical activity in roe deer (Capreolus capreolus) by measurement of fecal cortisol metabolites. General and Comparative Endocrinology 123:111–120. Guiness, F. E., G. A. Lincoln, and R. V. Short. 1971. The reproductive cycle of the female red deer, Cervus elaphus L. Journal of Reproduction and Fertility 27:427–428. Guiness, F. E., S. D. Albon, and T. H. Clutton-Brock. 1978. Calving times of red deer (Cervus elaphus) on Rhum. Journal of Zoology, London 185:104–114. Haigh, J. C. 2001. The gestation length of wapiti (Cervus elaphus) revisited. Animal Reproduction Science 665:89–93. Haigh, J. C., W. F. Cates, G. J. Glover, and N. C. Rawlings. 1984. Relationships between seasonal changes in serum testosterone concentrations, scrotal circumference and sperm morphology of male wapiti (Cervus elaphus). Journal of Reproduction and Fertility 70:413–418. Hart, B. L. 1974. Gonadal androgen and sociosexual behavior of male mammals: a comparative analysis. Psychological Bulletin 81:383–400. Hirata, S., and Y. Mori. 1995. Monitoring reproductive status by fecal progesterone analysis in ruminants. Journal of Veterinary Medical Science 57:845–850.

96

K.Yamauchi, Y. Matsuura

Huber, S., R. Palme, and W. Arnold. 2003. Effects of season, sex, and sample collection on concentrations of fecal cortisol metabolites in red deer (Cervus elaphus). General and Comparative Endocrinology 130:48–54. Iimura, T. 1980. An ecological study on the Japanese deer, Cervus nippon centralis, in the Tanazawa Mountains from the view point of forest protection. Dainipponsanrinkai, Tokyo, Japan. Kaji, K., T. Koizumi, N. Ohtaishi, T. Tsubota, and M. Suzuki. 1991. Evaluation of mass capture methods for sika deer. Honyurui Kagaku (Mammalian Science) 30:183–190. Kameyama, Y., R. Takahashi, M. Ito, R. Maru, and Y. Ishijima. 2000. Seasonal changes of testis volume, scrotal circumference and serum testosterone concentrations in male sika deer (Cervus nippon). Animal Science 71:137–142. Kameyama, Y., A. Miyamoto, S. Kobayashi, T. Kuwayama, and Y. Ishijima. 2002. Annual changes in serum LH and testosterone concentrations in male sika deer (Cervus nippon). Journal of Reproduction and Development 48:613–617. Kelly, R. W., K. P. McNatty, G. H. Moore, D. Ross, and M. Gibb. 1982. Plasma concentrations of LH, prolactin, oestradiol and progesterone in female red deer (Cervus elaphus) during pregnancy. Journal of Reproduction and Fertility 64:475–483. Kelly, R. W., K. P. McNatty, and G. H. Moore. 1985. Hormonal changes about oestrus in female red deer. Pages 181–184 in P. F. Fennessy and K. R. Drew, editors, Biology of deer production. Royal Society of New Zealand Bulletin 22, Wellington, New Zealand. Kirkpatrick, J. F., K. Bancroft, and V. Kincy. 1992. Pregnancy and ovulation detection in bison (Bison bison) assessed by means of urinary and fecal steroids. Journal of Wildlife Diseases 28:590–597. Kirkpatrick, J. F., D. F. Gudermuth, R. L. Flagan, J. C. McCarthy, and B. L. Lasley. 1993. Remote monitoring of ovulation and pregnancy of Yellowstone bison. Journal of Wildlife Management 57:407–412. Koizumi, T. 1991. Reproductive characteristics of female sika deer, Cervus nippon, in Hyogo Prefecture, Japan. Pages 561–563 in F. Spitz, G. Janeau, G. Gonzalez, and S. Aulagnier, editors, Ongules/Ungulates, 1991, Societé Française pour l’Étude et la Protection des Mammiferes, Institute de Recherche sur les Grands Mammiferes, Paris and Toulouse, France. Lasley, B. L., and J. F. Kirkpatrick. 1991. Monitoring ovarian function in captive and free-ranging wildlife by means of urinary and fecal steroids. Journal of Zoo and Wildlife Medicine 22:23–31. Leader-Williams, N. 1988. Reindeer on South Georgia. Cambridge University Press, Cambridge, United Kingdom. Li, C., R. P. Littlejohn, and J. M. Suttie. 1999. Effects of insulin-like growth factor 1 and testosterone on the proliferation of antlerogenic cells in vitro. Journal of Experimental Zoology 284:82–90. Li, C., R. P. Littlejohn, I. D. Corson, and J. M. Suttie. 2003. Effects of testosterone on pedicle formation and its transformation to antler in castrated male, freemartin and normal female red deer (Cervus elaphus). General and Comparative Endocrinology 131:21–31. Lincoln, G. A. 1985. Seasonal breeding in deer. Pages 165–179 in P. F. Fennessy and K. R. Drew, editors, Biology of deer production. Royal Society of New Zealand Bulletin 22, Wellington, New Zealand. Lincoln, G. A. 1992. Biology of seasonal breeding in deer. Pages 565–574 in R. D. Brown, editor, The biology of deer. Springer, New York, USA. Lincoln, G. A., and F. E. Guiness. 1973. The sexual significance of the rut in red deer. Journal of Reproduction and Fertility, Supplement 19:475–489. Lincoln, G. A., R. W. Youngson, and R. V. Short. 1970. The social and sexual behaviour of the red deer stag. Journal of Reproduction and Fertility, Supplement 11:71–103. Lincoln, G. A., F. E. Guiness, and R. V. Short. 1972. The way in which testosterone controls the social and sexual behavior of the red deer stag (Cervus elaphus). Hormones and Behavior 3:375–396. Liu, B.T., S. P. Cheng, M. C. Huang, and J. Y. L. Yu. 2002. Serum progesterone changes in luteal cyclicity and duration of estrous cycle in Formosan sika deer (Cervus nippon taiouanus) hinds. Zoological Science 19:1033–1037.

6

Endocrinology of Sika Deer

97

Loudon, A. S., and J. D. Curlewis. 1988. Cycles of antler and testicular growth in an aseasonal tropical deer (Axis axis). Journal of Reproduction and Fertility 83:729–738. Loudon, A. S., and B. R. Brinklow. 1992. Reproduction in deer: Adaptations for life in seasonal environments. Pages 261–278 in R.D. Brown, editor, The biology of deer. Springer, New York, New York, USA. Matsuura, Y., Y. Sasamoto, K. Sato, Y. Takahashi, M. Suzuki, and N. Ohtaishi. 2004a. Monitoring ovarian function and conception by fecal progesterone analysis in sika deer. Ecological Research 19:397–404. Matsuura, Y., K. Sato, M. Suzuki, and N. Ohtaishi. 2004b. The effects of age, body weight and reproductive status on conception dates and gestation periods in captive sika deer. Mammal Study 29:15–20. McMillin, J. M., U. S. Seal, K. D. Keenlyne, A. W. Erickson, and J. E. Jones. 1974. Annual testosterone rhythm in the adult white-tailed deer (Odocoileus virginianus borealis). Endocrinology 94:1034–1040. Mirarchi, R. E., B. E. Howland, P. F. Scanlon, R. L. Kirkpatrick, and L. M. Sanford. 1978. Seasonal variation in plasma LH, FSH, prolactin, and testosterone concentration in adult male white-tailed deer. Canadian Journal of Zoology 56:121–127. Miura, S. 1983. Grouping behavior of male sika deer in Nara Park, Japan. Honyurui Kagaku (Mammalian Science) 9:279–284. Miura, S. 1984a. Social behavior and territoriality in male sika deer (Cervus nippon Temminck 1838) during the rut. Zeitschrift für Tierpsychologie 64:33–73. Miura, S. 1984b. Annual cycles of coat changes, antler regrowth, and reproductive behavior of sika deer in Nara Park, Japan. Journal of the Mammalogical Society of Japan 10:1–7. Miura, S. 1986. A note on the evolution and social system in Cervidae. Honyurui Kagaku (Mammalian Science) 53:19–24. Monfort, S. L., C. Wemmer, T. H. Kepler, M. Bush, J. L. Brown, and D. E. Wildt. 1990. Monitoring ovarian function and pregnancy in Eld’s deer (Cervus eldi thamin) by evaluating urinary steroid metabolite excretion. Journal of Reproduction and Fertility 88:271–281. Monfort, S. L., C. C. Schwartz, and S. K. Wasser. 1993. Monitoring reproduction in captive moose using urinary and fecal steroid metabolites. Journal of Wildlife Management 57:400–407. Mori, Y. 1992. Central integration of photoperiodicity for gonadotropin release in ruminants. Pages 93–118 in A. Yokoyama, editor, Brain control of the reproductive system. Japan Scientific Societies Press, Tokyo, Japan. Mourik, S., and T. Stelmasiak. 1990. Endocrine mechanisms and antler cycles in rusa deer, Cervus (Rusa) timorensis. Pages 416–425 in G. A. Bubenik and A. B. Bubenik, editors, Horns, pronghorns, and antlers. Springer, New York, New York, USA. Ohtaishi, N. 1986. Preliminary memorandum of classification, distribution and geographic variation on sika deer. Honyurui Kagaku (Mammalian Science) 53:13–17. Plotka, E. D., U. S. Seal, G. C. Schmoller, P. D. Karns, and K. D. Keenlyne. 1977a. Reproductive steroids in the white-tailed deer (Odocoileus virginianus borealis). I. Seasonal changes in the female. Biology and Reproduction 16:340–343. Plotka, E. D., U. S. Seal, L. J. Verme, and J. J. Ozoga. 1977b. Reproductive steroids in the white tailed deer (Odocoileus virginianus borealis). II. Progesterone and estrogen levels in peripheral plasma during pregnancy. Biology and Reproduction 17:78–83. Plotka, E. D., U. S. Seal, L. J. Verme, and J. J. Ozoga. 1980. Reproductive steroids in deer. III. Luteinizing hormone, estradiol and progesterone around estrus. Biology and Reproduction 22:576–581. Robbins, M. M., and N. M. Czekala. 1997. A preliminary investigation of urinary testosterone and cortisol levels in wild male mountain gorillas. American Journal of Primatology. 43:51–64. Sadleir, R. M. F. S. 1987. Reproduction of female cervids. Pages 123–144 in C. M. Wemmer, editor, Biology and management of the Cervidae. Smithsonian Institution Press, Washington, DC, USA. Schams, D., and D. Barth. 1982. Annual profiles of reproductive hormones in peripheral plasma of the male roe deer (Capreolus capreolus). Journal of Reproduction and Fertility 66:463–468. Schams, D., D. Barth, and H. Karg. 1980. LH, FSH and progesterone concentrations in peripheral plasma of the female roe deer (Capreolus capreolus) during the rutting season. Journal of Reproduction and Fertility 60:109–114.

98

K.Yamauchi, Y. Matsuura

Schwartz, C. C., and K. J. Hundertmark. 1993. Reproductive characteristics of Alaskan moose. Journal of Wildlife Management 57:454–468. Schwartz, C. C., S. L. Monfort, P. H. Dennis, and K. J. Hundertmark. 1995. Fecal progestagen concentration as an indicator of the estrous cycle and pregnancy in moose. Journal of Wildlife Management 59:580–583. Seal, U. S., and M. Bush. 1982. Capture and chemical immunobilization of cervids. Pages 480– 497 in C. M. Wemmer, editor, Biology and management of the Cervidae. Smithsonian institution Press, Washington, DC, USA. Sempéré, A. J., and A. Lacroix. 1982. Temporal and seasonal relationship between LH, testosterone and antlers in fawn and adult male roe deer (Capreolus capreolus): A longitudinal study from birth to four years of age. Acta Endocrinology 99:295–301. Suttie, J. M., and P. F. Fennessy. 1992. Recent advances in the physiological control of velvet antler growth. Pages 471–486 in R. D. Brown, editor, The biology of deer. Springer, New York, New York, USA. Suzuki, M., and N. Ohtaishi. 1993. Reproduction of female sika deer (Cervus nippon yesoensis Heude, 1884) in Ashoro District, Hokkaido. Journal of Veterinary Medical Science 55:833–836. Suzuki, M., K. Kaji, and H. Nigi. 1992a. Annual changes of testis size, seminiferous tubules and plasma testosterone concentration of wild sika deer (Cervus nippon yesoensis Heude, 1884) in Hokkaido. Journal of Veterinary Medical Science 54:551–556. Suzuki, M., T. Koizumi, and M. Kobayashi. 1992b. Reproductive characteristics and occurrence of accessory corpora lutea in sika deer Cervus nippon centralis in Hyogo Prefecture, Japan. Journal of the Mammalogical Society of Japan 17:11–18. Suzuki, M., K. Kaji, M. Yamanaka, and N. Ohtaishi. 1996. Gestational age determination, variation of conception date, and external fetal development of sika deer (Cervus nippon yesoensis Heude, 1884) in eastern Hokkaido. Journal of Veterinary Medical Science 58:505–509. Suzuki, M., Y. Nakamura, M. Onuma, J. Tanaka, H. Takahashi, K. Kaji, and N. Ohtaishi. 2001a. Acid-base status and blood gas arterial values in free-ranging sika deer hinds immobilized with medetomidine and ketamine. Journal of Wildlife Diseases 37:366–369. Suzuki, M., M. Onuma, M. Yokoyama, K. Kaji, M. Yamanaka, and N. Ohtaishi. 2001b. Body size, sexual dimorphism, and seasonal mass fluctuations in a larger sika deer subspecies, the Hokkaido sika deer (Cervus nippon yesoensis Heude, 1884). Canadian Journal of Zoology 79:154–159. Takatsuki, S. 1992. Sika deer living in the north. Dobutsusya Publishing, Tokyo, Japan. (In Japanese.) Thomas, D. C., and I. McT. Cowan. 1975. The pattern of reproduction in female Columbian black-tailed deer, Odocoileus hemionus columbianus. Journal of Reproduction and Fertility 44:261–272. Verme, L. J. 1969. Reproductive patterns of white-tailed deer related to nutritional plane. Journal of Wildlife Management 33:881–887. Wasser, S. K., S. L. Monfort, and D. E. Wildt. 1991. Rapid extraction of fecal steroids for measuring reproductive cyclicity and early pregnancy in free-ranging yellow baboons (Papio cynocephalus cynocephalus). Journal of Reproduction and Fertility 92:415–423. Wasser, S. K., S. L. Monfort, J. Southers, and D. E. Wildt. 1994. Excretion rates and metabolites of oestradiol and progesterone in baboon (Papio cynocephalus cynocephalus) faeces. Journal of Reproduction and Fertility 101:213–220. Webster, J. R., and G. K. Barrell. 1985. Advancement of reproductive activity, seasonal reduction in prolactin secretion and seasonal pelage changes in pubertal red deer hinds (Cervus elaphus) subjected to artificially shortened daily photoperiod or daily melatonin treatments. Journal of Reproduction and Fertility 73:255–260. Webster, J. R., J. M. Suttie, and I. D. Corson. 1991. Effects of melatonin implants on reproductive seasonality of male red deer (Cervus elaphus). Journal of Reproduction and Fertility 92:1–11. Wesson, J. A., 3rd, P. F. Scanlon, R. L. Kirkpatrick, H. S. Mosby, and R. L. Butcher. 1979. Influence of chemical immobilization and physical restraint on steroid hormone levels in blood of white-tailed deer. Canadian Journal of Zoology 57:768–776.

6

Endocrinology of Sika Deer

99

Whitsett, J. M., H. Underwood, and J. Cherry. 1984. Influence of melatonin on pubertal development in male deer mice (Peromyscus maniculatus). Journal of Reproduction and Fertility 72:287–293. Yamaji, A., Y. Kiso, M. Suzuki, M. Yokoyama, and F. Sasaki. 1994. Seasonal changes in the gonadotropic cells in the pituitary of wild sika deer (Cervus nippon yesoensis). Journal of Reproduction and Development 40:159–165. Yamauchi, K., T. Murai, H. Tanaka, T. Yamamoto, and Y. Nishitani. 1982. Study on male reproductive organs in Japanese deer, Cervus nippon (Nara Park) - postnatal development and seasonal changes of testis and epididymis. Japanese Journal of Animal Reproduction 28:81–90. Yamauchi, K., Y. Ukai, M. Yaguchi, and Y. Nishitani. 1983. Study on male accessory sex glands in Japanese deer, Cervus nippon (Nara Park), particularly on the seasonal changes. Japanese Journal of Animal Reproduction 29:32–40. Yamauchi, K., T. Murai, and Y. Nishitani. 1984. Studies on the ovary of the Japanese deer, Cervus nippon in the Nara Park - postnatal development and seasonal changes. Japanese Journal of Animal Reproduction 30:162–173. Yamauchi, K., S. Hamasaki, Y. Takeuchi, and Y. Mori. 1997. Assessment of reproductive status of sika deer by fecal steroid analysis. Journal of Reproduction and Development 43:221–226. Yamauchi, K., S. Hamasaki, Y. Takeuchi, and Y. Mori. 1999. Application of enzyme immunoassay to fecal steroid analysis in sika deer (Cervus nippon). Journal of Reproduction and Development 45:429–434. Yokoyama, M., K. Kaji, and M. Suzuki. 2000a. Food habits of sika deer and nutritional value of sika deer diets in eastern Hokkaido, Japan. Ecological Research 15:345–355. Yokoyama, M., H. Uno, M. Suzuki, K. Kaji, and N. Ohtaishi. 2000b. Indices for nutritional condition and thresholds for winter survival in sika deer in Hokkaido, Japan. Japanese Journal of Veterinary Research 48:119–127. Yokoyama, M. M. Onuma, M. Suzuki, and K. Kaji. 2001a. Seasonal fluctuations of body condition in northern sika deer on Hokkaido Island, Japan. Acta Theriologica 46:419–428. Yokoyama, S., I. Maeji, T. Ueda, M. Ando, and E. Shibata. 2001b. Impact of bark stripping by sika deer, Cervus nippon, on subalpine coniferous forests in central Japan. Forest Ecology and Management 140:93–99.

Chapter 7

Reproduction of Female Sika Deer in Japan, with Special Reference to Kinkazan Island, Northern Japan Nobumasa Ohnishi, Masato Minami, Rie Nishiya, Kotoyo Yamada, Hiroyuki Nishizuka, Hiroshi Higuchi, Azusa Nara, Masatsugu Suzuki, and Seiki Takatsuki Abstract Reproduction of female sika deer (Cervus nippon) is explored in detail for a population on Kinkazan Island in northern Japan and compared to the variation in timing and value of reproductive parameters across the Japanese Archipelago. Age at sexual maturity usually occurs at yearling age, but some females, particularly in high nutritional environments, breed in their first year. Timing of the breeding season varies from north to south depending on the climate and vegetation, in relation to the most favorable season for giving birth. Late conceptions are commonplace, some due to young females maturing later in the season, and others due to older females failing to conceive in first estrous periods. Estrous cycles are variable, ranging from five to 25 days, with 15 days being most common. Usually a single calf is born after a 231-day gestation period, but twins occur, especially in nutritious environments. The sex ratio of offspring is balanced. Gestation period varies little across Japan despite northern animals being substantially larger than southern ones. The results for Japanese sika are discussed in relation to sika deer in other areas of the native and introduced range.

Introduction Until the 1980s, studies about the reproductive biology of female sika deer had been conducted on only some limited wild or captive populations (Chapman and Horwood 1968; Chapman 1974; Miura 1978, 1980; Feldhamer 1980; Iimura 1980; Yamauchi et al. 1982; Yamauchi et al. 1983). Therefore, descriptions of sika deer were often lacking in previous reviews of cervid reproduction such as Sadleir (1987) and Loudon and Brinklow (1992). Large-scale research of sika deer reproduction in Japan started in the late 1980s. Now a considerable amount of basic data has accumulated, and it is possible to evaluate variation in reproductive parameters that are influenced by geological, meteorological, and vegetational variables among local populations. Simultaneously, detailed studies of reproduction in particular populations of sika deer were conducted. Kinkazan Island is one area where reproduction of the D. R. McCullough et al. (eds.), Sika Deer: Biology and Management of Native and Introduced Populations, DOI: 10.1007/978-4-431-09429-6_7, © Springer 2009

101

102

N. Ohnishi et al.

population has been followed in detail over a number of years. The focus of this chapter, therefore, is to report the results of reproduction in the Kinkazan population and to relate these findings to variation in reproductive parameters across the Japanese Archipelago. All the deer at the Kinkazan study area were identified by individual natural characteristics and artificial marks such as hair staining and cuts on the ears. We weighed the deer by live-capturing them in mid-March from 1990 to 2002. Deer were captured by inducing them to enter a paddock by offering food and then confining them there. They were driven by people and tangled in a net. Each deer was immobilized by fastening its legs with rubber rope and placed on a flat electronic balance to measure body weight to the nearest 0.5 kg. Estrous periods were determined by observing female copulatory behavior. Rutting behavior, particularly copulation, was observed during daytime from September to November, though we did not observe at night. Some of the females copulated twice or more (Minami et al. chapter 22), and such copulations were also recorded. Data on copulations taken from 1993 to 2002 were used in this analysis. Parturition was observed from May to July, and some additional observation was made thereafter. Since we observed most copulations, we carefully observed the behavior of particular females prior to parturition dates. Females leave their home ranges immediately before parturition and go into forests or shrub lands. We carefully tracked them from a distance so as to not disturb them and confirmed the parturition dates. Dates of parturition taken from 1994 to 2002 were determined (N = 133). To calculate the length of the gestation period, we used the data where both the copulation and parturition were observed between the period of 1993–2002 (N = 82).

Age at Sexual Maturity First ovulation and/or pregnancy of wild sika deer in the Japanese Islands usually occurs during the breeding season at yearling age. That this is typical is confirmed in many populations in Japan such as Hokkaido (Suzuki and Ohtaishi 1993), Iwate Prefecture (Yamauchi and Matsuura chapter 6), Chiba Prefecture (Asada and Ochiai chapter 28), Hyogo Prefecture (Suzuki et al. 1992), and Kumamoto Prefecture (Koizumi et al. chapter 24). Pregnancy rate of this age is as high as that of older age classes in some populations. However, age at first reproduction, while typically at the yearling age, does vary due to environmental quality and population density. For example, fawns may become pregnant in low density populations (Chapman and Horwood 1968; Swanson and Putman chapter 40); however, it is not known if such pregnancies produce surviving offspring or not. There may be failure during gestation, or fawn mortality at or shortly after birth. However, if fawn pregnancy in sika deer is similar to that in white-tailed deer (Odocoileus virginianus), their ecological equivalent in North America, then the results of McCullough (1979) at the George Reserve in Michigan, that conception rate of fawns was

7

Reproduction of Female Sika Deer in Japan

103

matched virtually one-to-one by recruited young in the fall, may apply to sika deer as well. Apparently fawn females reach sexual maturity only under environmental conditions in which the mother is nutritionally capable of successfully rearing the offspring. Similarly, age at first reproduction can be set back to ages older than yearling in high density populations. For example, on Nakanoshima Island of Lake Toya on Hokkaido, age of first pregnancy is later due to their high density and poor nutritional condition (Kaji et al. chapter 30). Females did not give birth until two years of age on Kinkazan Island. Only 1.3% of three-year-old females gave birth, but birth rates increased at four years old and older (see Minami et al. chapter 23).

Timing and Length of Breeding Period The breeding season of free-ranging sika deer has been determined by direct observation (Nara Park) or calculating back gestational age from the date killed (studies in Hokkaido, Chiba Prefecture, Hyogo Prefecture, and Kumamoto Prefecture) assuming a 231-day gestation period as found in captive sika deer females (Hama 1990; Ikeda 1998). According to these studies, the seasonal spread in breeding is from September to January in Japan (Table 7.1). Breeding in northeastern populations tends to take place later than in southwestern populations (Fig. 7.1), probably due to phenological differences in the vegetation and climate (see Yabe and Takatsuki chapter 20 for maps of vegetation and snow depth). These study areas cover approximately 14 degrees of latitude. Large variations in conception date are found within populations (Table 7.1). However, results from Kinkazan Island to be presented below show that most of the breeding occurs in a cluster approximating a normal distribution over about a one month period, with a smaller skew of later breeding dates, including occasional isolated cases much later. Late conception tends to occur in young females (Suzuki et al. 1996). Difference in average or median date of conception between yearlings and females aged greater than two years is significant in Hokkaido (Suzuki et al. 1996), Hyogo (Koizumi 1992) and Kumamoto Prefecture (Koizumi et al. chapter 24). Besides later maturation of young females, failure to conceive followed by repeated estrous cycles results in some mature females breeding later. Yamauchi et al. (1999) Table 7.1 Timing and length of the breeding season for various sika deer populations in Japan, arranged in geographic order from northeast to southwest. Mean date Prefecture of conception Range Days Reference Hokkaido Chiba

29 Oct 23–24 Sep

7 Oct–17 Jan 8 Sep–11 Dec

78 94

Hyogo Kumamoto

29 Sep 1 Oct

9 Sep–23 Jan 12 Sep–22 Dec

136 83

Suzuki et al. 1996 Asada and Ochiai 1996 Koizumi 1992 Koizumi chapter 24

104

N. Ohnishi et al.

Fig. 7.1 Map of Japan showing the location of prefectures mentioned in the text and the location of Kinkazan Island.

reported that the cycle was 12–13 days according to progesterone concentration. Ikeda (1998), on the other hand, reported that it was 20.6 days (range: 18–22 days). Both of these studies were done on female deer in captivity and may be influenced by artificial factors. However, in the wild, since most of the females conceive during the first estrus, the second or later estruses are less frequently observed. Details of estrous cycle and reproductive endocrinology are described in Yamauchi and Matsuura chapter 6. Observed copulations by females in the Kinkazan Island study area are shown in Fig. 7.2. As noted earlier, the distribution of copulations over time approximated a normal distribution over the month of October. We observed 222 copulations through the study period. Copulations were first observed in early October, peaked in mid-October (mean ± SD = October 22 ± 10.4 days), and continued until late November. The pattern was not symmetric but long-tailed to the latter half (Fig. 7.2). Second and later copulations were distributed from mid-October until late November. As many as 85.0% of the copulations were successful in achieving

7

Reproduction of Female Sika Deer in Japan

105

20 15 10 5 0 1

6

11 16 21 October

26

31

5

10 15 November

20

25

30

Fig. 7.2 Percentage frequency of copulations by date of “single-estrus” i.e., pregnancy was achieved (black bar) and “multi-estrus” females (gray bar) of sika deer on Kinkazan Island.

pregnancy during the first estrus, and these females (“single-estrus females,” hereafter) did not come into heat again. The rest of the females (“multi-estrus females,” hereafter) showed a second (10.2%), third (3.9%), or fourth estrus (0.8%). Multiestrus females did not show a clear peak of copulation date; some females became estrus as soon as five days after the first copulation, but approximately 15-day intervals were most common. A few females showed as long as 25-day intervals between estrous cycles. Although the mean body weights of successful females (37.0 kg, N = 99) and multi-estrus females (37.1 kg, N = 15) were not significantly different (t-test, p > 0.97), the successful females were significantly younger (7.3 years old, N = 135) than the multi-estrus females (9.2 years old, N = 23, t-test, p = 0.008). Success of pregnancy of the successful females was 85.0% (87/108), which was significantly higher than that of the multi-estrus females (68.5%, 9/19, χ2 test, p = 0.0081).

Pregnancy Rates Pregnancy rate of adult sika deer (two years and older) in rich nutritional condition is generally higher than 80%. This is reported in Hokkaido, Iwate Prefecture, Chiba Prefecture, Hyogo Prefecture, etc. This high pregnancy rate is maintained even in yearlings and populations that show a reduction of body size due to higher population density. However, the pregnancy rate declines in populations that are at severe high density and/or poor nutritional condition such as Nakanoshima Island, Lake Toya (Kaji et al. chapter 30). In Chiba Prefecture, depression of pregnancy rates appear in the areas that exceed 15 individuals/km2. Takatsuki (1992) suggested that there is a decline of sexual activities in aged females (>12 years old, see Minami et al. chapter 23 for Kinkazan females).

106

N. Ohnishi et al.

Sika deer typically have a single offspring per pregnancy (Sadleir 1987). However, twinning does occur in sika deer, and in Japan, has been observed in Hokkaido and Iwate Prefecture. Suzuki (personal communication) found only two cases of twins over 15 years of research in Hokkaido. In Iwate Prefecture, Takatsuki (1998) reported the twinning rate was only 0.29%. Similarly, twins have been reported occasionally in wild sika deer introduced to areas outside of Japan (Feldhamer and Marcus 1994). So twinning in wild sika deer is relatively rare. On the other hand, in a captive (penned) population in China, the rate was reported to attain 10.4% (Sheng 1992). This may be due to long-term selective breeding to improve the deer farm stock. The sex ratio of newborn fawns was reported to be 1:1 in Hokkaido (Suzuki 1993) and Hyogo Prefecture (Koizumi 1992) and 1.08M:0.92F on Kinkazan Island. In white-tailed deer, females in populations under highly nutritious conditions have a disproportionate number of female offspring (Verme 1965; McCullough 1979), whereas in red deer (Cervus elaphus), in the British Isles, the fetal sex ratio (% of males) rises in populations showing high reproductive performance (Clutton-Brock and Albon 1989). However, comparable deviations from unity of the sex ratio of fetus in sika deer, so far, have not been confirmed in sika deer in the Japanese islands.

Parturition Date The calving season, of course, is affected by variation in the timing of the breeding season; in the Japanese Archipelago it is mainly from April to June. Previous reports indicate calving outside of the regular season occurred until September in Hokkaido (Suzuki et al. 1996), and November in Nara Park (Miura 1980). Observed parturitions on Kinkazan Island began in mid-May, peaked in early June (mean ± SD = June 4 ± 9.2 days), and declined to a low level by early July. This distribution again was nearly normal, but with a skew due to a few later dates, including some exceptional parturitions as late as 2 August (Fig. 7.3). The 2 August parturition would suggest a copulation date by this female in mid-December, which was not observed. This result shows how the length of the breeding seasons, as for example shown in Table 7.1, can be easily over- or underestimated by observing or missing a few copulations at scattered outlier dates.

Gestation Period Whereas several studies cited above determined the length of the gestation period in sika deer in captivity, accurate determination of gestation periods of wild female sika deer had not been determined before our study of the wild sika deer population living on Kinkazan Island. Based on our observations on known females on

7

Reproduction of Female Sika Deer in Japan

107

Fig. 7.3 Percentage frequency of parturition by date of sika deer females on Kinkazan Island.

Kinkazan Island observed to both copulate and give birth, the mean gestation period was 231.6 ± 4.5 days (N = 82). This closely approximates the estimates of 231 days reported for captive sika deer populations in Japan.

Neonatal Weights Just as adult weights of sika deer are greater in the north than the south, weights of neonatal calves vary among local populations. It is estimated at about 6 kg in Hokkaido in the north (Suzuki et al. 1996), and under 4 kg in Hyogo and Kumamoto Prefecture in the south (Koizumi et al. chapter 24). The mean of actual measurements from a Hokkaido captive population was 5.6 kg in males and 4.5 kg in females (Matsuura et al. 2004). For the wild population on Kinkazan Island the weights were 3.38 kg in males and 3.07 in females. The relatively low neonatal weights is related to comparable lower adult weights in this population due to high density.

Discussion The breeding season on Kinkazan Island occurs when it becomes cool in late September and frost begins in October. Plants begin withering in early October, and almost all plants wither in November. Sika deer deposit fat in summer, and the fat deposit of males peaks in September while the peak of females is delayed to November (Takatsuki unpublished data). From these results, it is likely that the rut occurs immediately after the peak of fat deposit of male sika deer. Parturition,

108

N. Ohnishi et al.

in turn, occurred mainly from late May through the first half of June. Plants begin to grow in mid-April on Kinkazan Island. During the parturition period, their leaves are succulent and fresh. This is the best season in terms of nutrition for the sika deer, particularly for offspring and lactating mothers. Thus, both the rut and parturition periods seem to synchronize with plant phenology affecting the nutritional conditions for sika deer. These results explain why the phenology of reproduction varies across Japan. In the north, the breeding season is set back so that parturition coincides with the later spring green-up, and consequent nutritious forage for raising the young. Parturition is concentrated in the north while the peak is less sharp in the south, though intensive studies are limited in the latter. Our results on the gestation period of sika deer were similar to those reported by Hama (1990) for captive animals. However, the Japan results differed from other areas of the sika deer range. The gestation periods of sika deer in Russia and China were longer: 242 and 237.2 days, respectively (Geist 1998), both of which were recorded for captive populations. The female weights for these deer were as much as 71–73 kg while ours were about 40 kg. In general, gestation period is related to body size, and it is likely that since the sika deer on Kinkazan Island are smaller than other populations (partly because of poor food conditions), the gestation period is shortened. In spite of great variations in the weights of mothers and newborn fawns of cervid species, the gestation periods do not differ greatly. For example, it is 250 days for wapiti (Cervus elaphus canadensis) in which mother’s body weight is as heavy as 254 kg and fawn’s weight is 14.8 kg (Geist 1998). Similarly, it is 234 days for red deer (Cervus elaphus) whose mother and fawn body weights are 84 and 7.9 kg, respectively. Interestingly, the data for fallow deer (Dama dama) are very close to ours: gestation period is 230 days and mother and fawn weights are 39 and 4.2 kg, respectively. These comparisons suggest that for the temperate cervids, it is more advantageous to rut immediately after the peak of fat deposition and give birth in the fresh green season, and therefore the gestation periods are less variable. Since late May and early June is the best season of food quality and abundance, parturition is concentrated in this period regardless of mothers’ body weight. If the gestation period is relative to body weight of mother deer, sika deer females “elongate” the duration depending on the plant phenology. The estrous cycle of sika deer females on Kinkazan Island was about 15 days, though the majority (85%) of females became estrous only once. The cycle of captive females was 12–13 days (Yamauchi et al. 1999) or 21 days (Ikeda 1998). Although we do not know the reason for this difference, it is noteworthy that the captive females were kept apart from males (Ikeda 1998). For comparison, it is known that the estrous cycle of red deer is 18–22 days (Morrison 1960a, b; Lincoln et al. 1970; Guiness et al. 1971) and that of wapiti is 21 days (Morrison 1960a). According to body size, a 21-day cycle seems to be too long for sika deer females. The multi-estrus females in our study were older and may have some limitations in their reproductive physiology. The distribution of estrous cycles was not concentrated at any particular length, but was spread from five days to 25 days without any clear peak. Thus, we should be cautious about comparing captive females with

7

Reproduction of Female Sika Deer in Japan

109

artificially controlled reproduction with wild females whose estrous cycles have been determined only from multi-estrus females. Late conception of multi-estrus females automatically results in late parturition, which would shorten the growth of offspring before winter. The growth rate of early-born fawns of Hokkaido sika deer was the same as that of late-born ones, and consequently the mean body weight of the former before winter was greater than that of the latter (Takatsuki and Matsuura 2000). It is likely that mortality would be higher in the late-born fawns. More study is needed on reproduction to answer these questions. For example, are multi-estrus females particular individuals or do many females become multiestrus with age? And, does the behavior of rutting males or their hierarchical position influence females to be single- or multi-estrus?

Literature Cited Asada, M., and K. Ochiai. 1996. Conception dates of sika deer on the Boso Peninsula, central Japan. Mammal Study 21:153–159. Chapman, D. I. 1974. Reproductive physiology in relation to deer management. Mammal Review 4:61–74. Chapman, D. I., and M. T. Horwood. 1968. Pregnancy in sika deer calf, Cervus nippon. Journal of Zoology, London 155:227–228. Clutton-Brock, T., and S. D. Albon. 1989. Red deer in the Highlands. BSP Professional Books, Oxford, United Kingdom. Feldhamer, G. A. 1980. Cervus nippon. Mammalian Species 128:1–7. Feldhamer, G. A., and M. A. Marcus. 1994. Reproductive performance of female sika deer in Maryland. Journal of Wildlife Management 58:670–673. Geist, V. 1998. Deer of the world: Their evolution, behavior, and ecology. Stackpole Books, Mechanicsburg, Pennsylvania, USA. Guiness, F. E., G. A. Lincoln, and R. V. Short. 1971. The reproductive cycle of the female red deer, Cervus elaphus L. Journal of Reproduction and Fertility 27:427–438. Hama, N. 1990. Judgment of pregnancy by supersonic waves, fetal growth, and measurements of blood progesterone for Hokkaido sika deer. Thesis for Veterinary School, Hokkaido University, Japan. (In Japanese.) Ikeda, S. 1998. Management of reproduction. Pages 16–21 in Guidebook for sika deer farming. Ministry of Agriculture, Forestry, and Fishery, Japan, and the Tohoku Branch of Ministry of Agriculture, Forestry, and Fishery. (In Japanese.) Iimura, T. 1980. An ecological study on the Japanese deer, Cervus nippon centralis, in the Tanzawa Mountains from the view point of forest protection. Dainippon-Sanrinkai, Tokyo, Japan. (In Japanese with English summary.) Koizumi, T. 1992. Reproductive characteristics of female sika deer, Cervus nippon, in Hyogo Prefecture, Japan. Ongules/Ungulates 91: 561–563. Lincoln, G. A., R. W. Youngson, and R. V. Short. 1970. The social and sexual behaviour of the red deer stag. Journal of Reproduction and Fertility Supplement 11:71–103. Loudon, A. S. I., and B. R. Brinklow. 1992. Reproduction in deer: Adaptations for life in seasonal environments. Pages 261–278 in R. D. Brown, editor, Biology of deer. Springer, New York, USA. Matsuura, Y., K. Sato, M. Suzuki, and N. Ohtaishi. 2004. The effects of age, body weight, and reproductive status on conception dates and gestation periods in captive sika deer. Mammal Study 29:15–20.

110

N. Ohnishi et al.

McCullough, D. R. 1979. The George Reserve deer herd: Population ecology of a K-selected species. University of Michigan Press, Ann Arbor, Michigan, USA. Miura, S. 1978. A study of sika deer in Nara Park. I. Rutting period. Pages 2–13 in Kasuga Kensho Kai, editor, Annual report of Nara Deer Research Association. Nara, Japan. (In Japanese.) Miura, S. 1980. Annual cycles of sika deer with special reference to birth season. Pages 87–94 in Kasuga Kensho Kai, editor, Annual report of Nara Deer Research Association. Nara, Japan. (In Japanese.) Morrison, J. A. 1960a. Characteristics of estrus in captive elk. Behaviour 16:84–92. Morrison, J. A. 1960b. Ovarian characteristics in elk of known breeding history. Journal of Wildlife Management 24:297–307. Sadleir, R. M. F. S. 1987. Reproduction in female cervids. Pages 123–144 in C. M. Wemmer, editor, Biology and management of the Cervidae. Smithsonian Institution Press, Washington DC, USA. Sheng, H-L. 1992. Sika deer. Pages 202–212 in H-L. Shen, editor, The deer in China. East China Normal University Press, Shanghai, China. (In Chinese with English summary.) Suzuki, M., and N. Ohtaishi. 1993. Reproduction of female sika deer (Cervus nippon yesoensis Heude, 1884) in Ashoro District, Hokkaido. Journal of Veterinary Medical Science 55:833–836. Suzuki, M., T. Koizumi, and M. Kobayashi. 1992. Reproductive characteristics and occurrence of accessory corpora lutea in sika deer Cervus nippon centralis in Hyogo Prefecture, Japan. Journal of the Mammalogical Society of Japan 17:11–18. Suzuki, M., K. Kaji, M. Yamanaka, and N. Ohtaishi. 1996. Gestational age determination, variation of conception date, and external fetal development of sika deer (Cervus nippon yesoensis Heude, 1884) in eastern Hokkaido. Journal of Veterinary Medical Science 58:505–509. Takatsuki, S. 1992. A sika deer herd living in the north. Dobutsu-sha, Tokyo, Japan. (In Japanese.) Takatsuki, S. 1998. The twinning rate of sika deer, Cervus nippon, on Mt. Goyo, northern Japan. Mammal Study 23:103–107. Takatsuki, S., and Y. Matsuura. 2000. Higher mortality of smaller sika deer fawns. Ecological Research 15:237–240. Verme, L. J. 1965. Reproduction studies on penned white-tailed deer. Journal of Wildlife Management 29:74–79. Yamauchi, S., S. Miura, H. Tanaka, T. Yamamoto, and Y. Nishitani. 1982. Study on male reproductive organs in Japanese deer, Cervus nippon (Nara Park): Postnatal development and seasonal changes of testis and epididymis. Japanese Journal of Animal Reproduction 28:81–88. (In Japanese with English summary.) Yamauchi, S., Y. Ukai, M. Yaguchi, and Y. Nishitani. 1983. Study on male accessory sex glands in Japanese deer, (Nara Park), particularly on the seasonal changes. Japanese Journal of Animal Reproduction 29:32–38. (In Japanese with English summary.) Yamauchi, K., S. Hamasaki, Y. Takeuchi, and Y. Mori. 1999. Application of enzyme immunoassay to fecal steroid analysis in sika deer (Cervus nippon). Journal of Reproduction and Development 45:429–434.

Chapter 8

Food Habits of Sika Deer on Kinkazan Island, Northern Japan with Reference to Local Variations, Size Effects, and Comparison with the Main Island Seiki Takatsuki and U. K. G. K. Padmalal

Abstract Food habits of sika deer (Cervus nippon) on Kinkazan Island were studied to show (1) whether local variations exist on this small island (9.6 km2); (2) which kinds of plants are the staple foods for the deer living in the plant communities heavily affected by deer gazing; and (3) whether dietary differences exist or not among males, females, and fawns. Plant availability on Kinkazan Island was characterized by higher frequency of forbs and lower frequency of browse and graminoids than other areas on the adjacent mainland. Local variations in foods were great, but graminoids such as Zoysia japonica, Miscanthus sinensis, and/or Pleioblastus chino were consistently the staple foods. Forbs included many unpalatable plants and did not contribute greatly to the deer diets. Males tended to feed on lower quality foods, while fawns fed on nutritious foods. Food differences among the deer sex and age classes were greatest in spring and autumn, while they were smaller in summer and winter.

Introduction The distribution of sika deer in northern Honshu, the main island of Japan, is mainly limited by winter snow and hunting pressure, particularly on the Japan Sea side. On the Pacific side of northern Honshu, the main island, although a major population exists at Mt. Goyo and the surrounding area, deer distribution occurs in isolated patches; this differs from southern Japan where deer distribution is more continuous. Kinkazan Island, a small island (960 ha) on the Pacific side of northern Honshu (Fig. 26.1 in Takatsuki chapter 26), is an exception, for it is inhabited by many sika deer. The island is regarded as a sanctuary, and wildlife has been conserved. Hunting is prohibited and because no large predator lives there, the deer population is close to the carrying capacity. This has led to prevention of forest regeneration (Takatsuki and Gorai 1994; Takatsuki and Hirabuki 1998), dominance of unpalatable and grazing-tolerant plants (Yoshii and Yoshioka 1949), and mass-mortality of deer (Takatsuki et al. 1994). Also, the forests of Kinkazan Island lack shrub layers and have more gaps than comparable areas on the mainland so that open patches are more available. This affords a high diversity in physiognomy on this island, and the deer can utilize these diversified habitats. D. R. McCullough et al. (eds.), Sika Deer: Biology and Management of Native and Introduced Populations, DOI: 10.1007/978-4-431-09429-6_8, © Springer 2009

113

114

S. Takatsuki, U. K. G. K. Padmalal

It is expected, therefore, that deer behavior and plant responses are reflected in various aspects of the life history of the sika deer. Food habits, explored in this chapter, are a cogent example. We were interested to know what kinds of plants the deer living in this distinctive vegetation characterized by unpalatable plants and grazing-tolerant plants chose as foods. Since the high density of deer limits food, the deer must eat poorer quality foods, particularly in winter, which determines the maximum number of deer that can overwinter. In summer, by contrast, foods are superabundant. Such clear seasonal differences in food supply should be also reflected in their food habits. Also of interest is how food habits relate to the body size of the deer. It is generally accepted that the larger ungulates tend to feed on less nutritious foods because they can survive on lower quality foods which are more abundant, while smaller ungulates tend to feed on more nutritious foods because they require high quality foods which are often less abundant. This is often referred to as the Jarman-Bell principle (Bell 1970; Jarman 1974). It is instructive to know how different sized deer feed under a condition where food quality is poor and seasonal changes in food supply are marked. This chapter describes the food habits of sika deer on Kinkazan Island, focusing on local variations on this small island and comparisons among different sized deer classes based on the former studies by Takatsuki (1980) and Padmalal and Takatsuki (1994).

Methods Plants appearing in 1 m × 1 m quadrats in six sites representing habitats of sika deer on Kinkazan Island were recorded (Fig. 8.1). The feeding effects and plant species lists are described in detail by Takatsuki (1980). In this chapter we summarize those records to compare the feeding of different food plant types: graminoids, forbs, ferns, lianas, shrubs, and trees. The six sites sampled were as follows: Site 1 was chosen in the shrine park in the northwestern part of the island, where the Zoysia japonica (Japanese lawngrass) community develops in open lands and the Poa annua (annual bluegrass) community develops under cherry trees. A very high density of deer live here. Site 2 was selected in a small patch of the Zoysia community among the Pleioblastus chino (dwarf bamboo) community on a flat located north of the shrine. Site 3 was selected in the above Pleioblastus community. Miscanthus sinensis (Japanese silver grass), a tall species, is also abundant here, and shrubs like Viburnum dilatatum (linden arrowwood) and Berberis thunbergii (Japanese barberry) occur. Site 4 was at a higher elevation than Site 3. This site is composed of various plant communities including the forest patches dominated by Carpinus tschonoskii (Yeddo hornbeam), Abies firma (Japanese fir), and Zelkova serrata (gray-bark elm), the Miscanthus community, the Senecio cannabifolius (Aleutian ragwort) community, and the Hypolepis punctata (bramble fern) community.

8

Food Habits of Sika Deer on Kinkazan Island, Northern Japan

115

Fig. 8.1 Map of Kinkazan Island showing the locations of sampling sites for fecal pellets. Contour lines show 100 m intervals.

Site 5 was selected in the Miscanthus community on the peninsula at the northern end of the island. Viburnum shrubs are abundant. Site 6 was selected in the Fagus crenata (Japanese beech) forest. The undergrowth is sparse and only short grasses and forbs grow there, though some unpalatable shrubs of Leucothoe grayana (fetterbush, Ericaceae) and Zanthoxylum piperitum (Japanese pepper tree) grow in clumps. Quantitative evaluation of food habits was done using the point frame method of examining fecal pellets (Stewart 1967). To show local variations of food habits, fecal pellets were collected at the six sites on the island (Fig. 8.1). For sampling, 20 pellets were collected from 20 different fecal piles in August, October, and December, 1976 and in April, 1977. Only fresh pellets were collected. Plants were identified by epidermal features of the plant fragments and placed into six categories: (1) Miscanthus sinensis, (2) Zoysia japonica, (3) Pleioblastus chino, (4) other graminoids, (5) “others” including forbs, browse, and ferns, and (6) unidentified materials. Fields of each microscopic slide were viewed at 200× or 600× magnification. Crossing points (1 mm aperture) on slides covered by plant fragments were scored for each food category and a total of 500 identifications was made for each sample. Percentage frequency of feeding (Fi) was determined as: Fi = fi /Ni where fi = the number of quadrats including the species i eaten by deer, and Ni = the total number of quadrats including the species i.

116

S. Takatsuki, U. K. G. K. Padmalal

In order to compare the food habits of different size classes of sika deer, we designated Site 7 in the western part of the island where a place (0.2 ha) composed of patches of the Zoysia japonica community, forb communities, and deciduous broadleaf forests was available (Fig. 8.1). We adopted the same method as above. Fecal pellets (four pellets each from six to eight fecal piles) were collected in October, 1990 and February, April, and July, 1991. They were treated similarly for determination of plant categories, but in this analysis about half of the samples were used to determine the nitrogen concentrations as an indicator of food quality by the micro-Kjeldahl method.

Results Plants Eaten by Sika Deer Table 8.1 summarizes the results of occurrence of plants (indicating their availability) and their consumption by sika deer. Among the 44 plots, a total of 158 plant species was recorded. Forbs were most abundant, with 674 occurrences (60.9%). These were followed by graminoids (295 occurrences, 26.6%) and browse (126 occurrences, 11.4%). Only four ferns appeared and none were eaten by sika deer. Feeding on graminoids was most frequent (84.7%) and least frequent on forbs (16.8%) except ferns whose sample size was very small.

Food Composition Seasonal changes in food composition determined by fecal analysis are shown in Fig. 8.2. At the shrine park (Site 1), Zoysia japonica was dominant in the feces, making up more than half of the diet in summer and fall. However, it significantly

Table 8.1 Frequencies of feeding (F) of five plant categories. Comparable results from Mt. Goyo are also shown. (From Takatsuki 1996). Kinkazan Island

Graminoid Forb Fern Browse Total

Mt. Goyo

Frequency of occurrence

Frequency of feeding

F(f/N)

Frequency of occurrence

Frequency of feeding

F(f/N)

N

(%)

f

(%)

N

(%)

f

(%)

295 674 12 126 1107

26.6 60.9 1.1 11.4 100

250 113 0 61 424

84.7 16.8 0.0 48.4 38.3

436 224 26 350 1,036

42.1 21.6 2.5 33.8 100

260 12 2 58 332

59.6 5.4 7.7 16.6 32.0

8

Food Habits of Sika Deer on Kinkazan Island, Northern Japan

117

Fig. 8.2 Seasonal changes in the botanical compositions of sika deer feces from six sites on Kinkazan Island. Miscanthus: Miscanthus sinensis, Zoysia: Zoysia japonica, Pleioblastus: Pleioblastus chino. Site numbers correspond to those of Fig. 8.1.

decreased in winter and spring diets, constituting only 10.2% and 25.6%, respectively. Pleioblastus chino appeared consistently although the proportion was small. At the grassland on a flat in the western part (Sites 2 and 3), fecal compositions were similar each other. Zoysia japonica made up 20–30% of the fecal composition in summer and fall but decreased in winter to only 1.2%. Pleioblastus chino formed 20–40% of the diet except in spring when it decreased to 13.2% and 8.4% at Sites 2 and 3, respectively. Pleioblastus chino was constantly grazed in winter and made up 30% of fecal composition. At the deciduous forest in the western part of the island (Site 4), Pleioblastus chino formed 17.4% of fecal composition in spring. In summer, Miscanthus sinensis occupied 13.1%, followed by Zoysia japonica (11.3%) and Pleioblastus chino (9.9%). In fall, Zoysia japonica occupied 15.4%, followed by P. chino (12.3%) and Miscanthus sinensis (8.8%). In winter, Pleioblastus chino increased again to reach a level comparable to that of spring (13.2%), while Zoysia japonica decreased markedly (2.5%). Seasonal changes in the fecal composition were less marked here than other sites. At the grassland in the northern peninsula (Site 5), Miscanthus sinensis was the most important forage in summer and fall. Total amount of graminoids contributed to 70–90%, although identification was difficult because many plant fragments recovered in the feces were sheaths and culms. In the Fagus forest (Site 6), Miscanthus sinensis accounted for 18.7% of the diet in summer and maintained around 10% in fall and even in winter (7.8%). This site differed from others in that “others” and “unidentified” materials occupied considerably large portions of the fecal composition. These categories included browse, forbs, and ferns, but most of the fragments were unidentifiable. It is probable that at least in winter a considerable part of the diet would be bark and twigs, most of which are not identifiable by epidermal features.

118

S. Takatsuki, U. K. G. K. Padmalal

Comparison Among Size Classes of Deer Botanical composition of the feces at Site 7, where feces were sampled to size-class differences on nitrogen content, were similar to those of the shrine park (Site 1) in that Zoysia japonica occupied large portions of the diet particularly in summer and fall, though Pleioblastus chino did not appear at Site 7, although it was present at Site 1. There were apparent differences in botanical compositions among the deer classes (Fig. 8.3). The proportion of graminoids was greatest in males and least in fawns (ANOVA, p < 0.05) except in winter (p > 0.05). Culms and sheaths constituted much of the diet in fall and winter, and these non-leaf parts were found more often in the larger deer classes. Contrary to graminoids, leaves of dicotyledonous plants showed a reversed pattern: their proportions were greater in smaller deer classes (p < 0.05). The exception was between fawns and females in fall when no difference was found (p > 0.05). The “others” category did not show any consistent tendency among the deer classes. They were greater in larger deer in spring and winter while they were smaller in spring and fall. Marked differences were found in seeds in fall when smaller deer consumed more seeds. Similarity indices (Whittaker 1952) between the diets of the deer of different classes indicated that the dietary overlaps were greater in summer and winter and less in spring and fall (Table 8.2). The indices also showed that the dietary overlaps were highest between females and fawns, and lowest between males and fawns (Table 8.2). Nitrogen concentrations of the feces were low in spring, increased in summer and fall, and decreased in winter (Fig. 8.4). They were consistently highest in fawns and lowest in males in every season (Fig. 8.4).

Comparison of Kinkazan Island with Mt. Goyo On Kinkazan Island, from the 158 plant species, forbs composed 60.9% of the plants in quadrats in the sika deer habitats (Table 8.1). However, the frequency of

Fig. 8.3 Seasonal changes in the botanical compositions of diets of male, female, and fawn sika deer at Site 7.

8

Food Habits of Sika Deer on Kinkazan Island, Northern Japan

119

Table 8.2 Percent similarities (Whittaker 1952) of dietary compositions between different classes of sika deer in different seasons. Deer class Spring Summer Fall Winter Males and females Females and fawns Males and fawns

83.6 89 72.7

90.3 96.4 83.7

80.9 82.8 69.5

85.7 90.4 77.2

Fig. 8.4 Seasonal changes in nitrogen concentrations in fecal pellets of male, female, and fawn sika deer at Site 7.

feeding on forbs was only 16.8% (Table 8.1). As a result, the feeding frequency for all the plants was 38.3%. By comparison 88 plant species appeared in the study plots at Mt. Goyo where the sika deer density was low. Among the 1,036 occurrences of plants in these plots, 332 cases contained eaten plants. Although this total frequency of feeding (32.0%) was significantly smaller than that of Kinkazan Island (38.3%) (χ2 value = 352.1, p < 0.0001, df = 1), the difference was small. This is unexpected because the deer density is much higher on Kinkazan Island (50 deer/km2) than at Mt. Goyo (<10 deer/km2), and consequently grazing effects are heavier on Kinkazan Island. In fact, for each plant category except ferns, the feeding frequency was greater on Kinkazan Island than at Mt. Goyo (Table 8.1). For example, the frequencies of graminoids on Kinkazan Island and Mt. Goyo were 84.7% and 59.6%, respectively. Similarly, those of browse and forbs were 48.4% and 16.6%, and 16.8% and 5.4%, respectively. Nevertheless, the total frequencies of all the plants were similar. This is because although the floral composition was

120

S. Takatsuki, U. K. G. K. Padmalal

more biased toward forbs on Kinkazan Island (61.4%) than at Mt Goyo (36.4%), the frequency of forbs was smaller than those of other plant categories. Consequently, in spite of the greater frequency value for forbs on Kinkazan Island (16.8%) than that of Mt. Goyo (5.4%), the total frequency for all plants was only slightly greater than that of Mt. Goyo. On the contrary, although the frequency of feeding for graminoids was as high as 84.7% on Kinkazan Island, frequency of occurrence for graminoids was only 26.6% and consequently did not greatly contribute the total feeding frequency.

Discussion These results suggest that the plant communities on Kinkazan Island have been greatly altered by intensive grazing of the sika deer. They contain many forbs and fewer woody plants and graminoids. The forb category also contains many unpalatable plant species such as Senecio cannabifolius (Aleutian ragwort), Paeonea obovata (a peony), Arisaema spp. (arums), Perilla frutescens (beefsteak plant), etc. In fact, dicotyledonous plants occupied only small portion in the fecal composition of Kinkazan deer. The fecal analyses have shown that even on a small island like Kinkazan (less than 10 km2 in size), local variation in food habits was fairly great. This suggests that it is dangerous to conclude the characteristics of food habits of a deer population from limited samplings. It is worthy of note that despite consistent dominance of grasses in the diets of sika deer on Kinkazan Island, the occurrence of these staple grass species in the deer habitats was not always great. The preferred grasses were Zoysia japonica (Japanese lawngrass), Miscanthus sinensis (Japanese silver grass), and/or Pleioblastus chino (dwarf bamboo). Among these, Zoysia japonica showed great seasonal variation in the fecal compositions, while Pleioblastus chino was stable across the year. These grasses often are capable of regrowing after being grazed, and indeed are rather facilitated by deer grazing because deer eliminate other taller plants that are competitors of these grasses. These results, together with other food habit studies of sika deer populations in northern Japan (e.g., Takatsuki 1983, 1986; Ueda et al. 2003; Yokoyama et al. 2000), suggest that the sika deer in northern Japan are “grazer-type” deer. The fact that the rumino-reticulums of a northern population of sika deer are relatively large (Takatsuki 1988) is consistent with this type of food. Intraspecific variations in food habits among deer of different sizes are often explained by the differences of habitat selection among them, for example, red deer, Cervus elaphus (Charles et al. 1977; Watson and Staines 1978; Clutton-Brock et al. 1987). Although fecal pellets were collected on exactly the same Zoysia community at Site 7, some differences in habitat selection were noticed. We often observed that males stayed on the Zoysia community at the coast where females and fawns were rarely found. It is possible that this spatial difference would result in further differences in foods among sex and age classes.

8

Food Habits of Sika Deer on Kinkazan Island, Northern Japan

121

Botanical analyses have shown that larger deer classes fed on more graminoids in all the seasons, and more culms and sheaths in fall and winter. In contrast, however, smaller deer classes fed on more leaves of dicotyledons in all the seasons and more seeds in fall. These results suggest that larger deer classes fed on less nutritious foods. In fact, nutritional values, particularly protein content, are know to be higher in fruits and seeds than in leaves (Central Association of Livestock Industry 1975). Nitrogen concentrations of the fecal pellets also support the conclusion that the quality of diet is related to body size (Fig. 8.4). Our results support the Jarman-Bell principle (Geist 1974) that larger-bodied ungulates consume greater amounts of lower quality foods, whereas smaller ungulates consume smaller amounts of food but select for higher quality. Body weights of males, females, and fawns were 44, 37, and 19 kg, respectively. The male/female ratio is 1.2, which is the critical value of sexual dimorphism reported by Putman et al. (1993). They suggested that the dietary compositions of ungulates would be different between the sexes when dimorphism exceeded 1:1.2. These differences in body size and food habits are further related to other anatomical differences. Anatomical studies have shown that larger animals that require active fermentation in the foregut to process less digestible foods have developed larger rumino-reticulums (Hofmann 1973, 1989). Sika deer have relatively larger rumino-reticulums than red deer (Takatsuki 1988). This also explains the food selection differences among the different size classes of sika deer. It is also known that a bite size affecting food choice is determined by incisor breadth because a narrow incisor breadth enables ungulates to eat plants selectively (Clutton-Brock and Harvey 1983; Gordon and Illius 1988). The incisor breadth of sika deer on Kinkazan Island was 24.5, 21.3, and 17.0 mm in males, females, and fawns, respectively, and the difference between sexes (male/female = 1.15) was greater than in red deer (male/female = 1.04, Illius and Gordon 1990). Marked seasonal changes in food habits of Kinkazan sika deer are best interpreted by the seasonal difference in food supply. It is interesting that the foods of deer sex and age classes were more different in spring and fall and more similar in summer and winter. This suggests that dietary differences were least when foods were both super-abundant and scarce, and they were pronounced when foods were in intermediate availability. When foods are super-abundant, it is expected that deer can consume adequate amounts of good quality foods, while when foods are limited they are forced by lack of choice to consume poor quality foods. Both situations would result in smaller food differences among the deer classes. In fact, it has been reported that greater interspecific overlaps in foods were observed during the period of low forage availability between mule deer (Odocoileus hemionus) and white-tailed deer (O. virginianus) (Anthony and Smith 1977) and between elk and black-tailed deer (Leslie et al. 1984). On the other hand, when foods are neither super-abundant nor limited but are variable both in quality and quantity, deer can select preferable plants depending on their requirements; larger deer feed on greater amounts of foods that are often less nutritious and smaller deer feed on smaller amounts of nutritious foods. This would be advantageous for the deer because they can use resources more efficiently by differentiating or avoiding overlap (Geist 1974).

122

S. Takatsuki, U. K. G. K. Padmalal

Literature Cited Anthony, R. G., and N. S. Smith. 1977. Ecological relationship between mule deer and whitetailed deer in southern Arizona. Ecological Monographs 47:255–277. Bell, R. 1970. The use of the herb layer by gazing ungulates in the Serengeti. Pages 111–124 in A. Watson, editor, Animal populations in relation to their food resources. Blackwell Scientific, Oxford, United Kingdom. Central Association of Livestock Industry. 1975. Standard chemical compositions of forages in Japan. Central Association of Livestock Industry, Tokyo, Japan. (In Japanese.) Charles, W. N., D. McCowan, and K. East. 1977, Selection of upland swards by red deer (Cervus elaphus L.) on Rhum. Journal of Applied Ecology 14:55–64. Clutton-Brock, T. H., and P. H. Harvey. 1983. The functional significance of variation in body size among mammals. Pages 632–668 in J. F. Eisenberg, editor, Mammal Behavior and Ecology. Smithsonian Institution Press, Washington, DC, USA. Clutton-Brock, T. H., G. R. Iason, and F. E. Guiness. 1987. Sexual segregation and density-related changes in habitat use in male and female red deer. Journal of Zoology 211:275–289. Geist, V. 1974. On the relationship of social evolution and ecology in ungulates. American Zoologist 14:205–220. Gordon, I. J., and A. W. Illius. 1988. Incisor arcade structure and diet selection in ruminants. Functional Ecology 2:15–22. Hofmann, R. R. 1973. The ruminant stomach: Stomach structure and feeding habits of East African game ruminants. East African Literature Bureau, Nairobi, Kenya. Hofmann, R. R. 1989. Evolutionary steps of ecophysiological adaptation and diversification of ruminants: A comparative view of their digestive system. Oecologia 78:443–457. Illius, A. W., and I. J. Gordon. 1990. Variation in foraging behaviour in red deer and the consequences for population demography. Journal of Animal Ecology 59:89–101. Jarman, P. J. 1974. The social organization of antelope in relation to their ecology. Behaviour 48:215–266. Leslie, D. M. Jr., E. E. Starkey, and M. Vavra. 1984. Elk and deer diets in old-growth forests in western Washington. Journal of Wildlife Management 48:762–775. Padmalal, U. K. G. K., and S. Takatsuki. 1994. Age-sex differences in the diets of sika deer on Kinkazan Island, northern Japan. Ecological Research 9:251–256. Putman, R. J., S. Culpin, and S. J. Thirgood. 1993. Dietary differences between male and female fallow deer in sympatry and allopatry. Journal of Zoology 229:67–275. Stewart, D. R. M. 1967. Analysis of plant epidermis in faeces: A technique for studying the food preferences of grazing herbivores. Journal of Applied Ecology 4:83–111. Takatsuki, S. 1980. Food habits of sika deer on Kinkazan Island. Science Reports Tohoku University, Series IV (Biology) 38:7–31. Takatsuki, S. 1983. The importance of Sasa nipponica as a forage for sika deer (Cervus nippon) in Omote-Nikko. Japanese Journal of Ecology 33:17–25. Takatsuki, S. 1986. Food habits of sika deer on Mt. Goyo. Ecological Research 1:119–128. Takatsuki, S. 1988. The weight contributions of stomach compartments of sika deer. Journal of Wildlife Management 52:313–316. Takatsuki, S., and T. Gorai. 1994. Effects of sika deer on the regeneration of a Fagus crenata forest on Kinkazan Island, northern Japan. Ecological Research 9:115–120. Takatsuki, S., and Y. Hirabuki. 1998. Effects of sika deer browsing on the structure and regeneration of the Abies firma forest on Kinkazan Island, northern Japan. Journal of Sustainable Forestry 6:203–221. Takatsuki, S., K. Suzuki, and I. Suzuki. 1994. A mass-mortality of sika deer on Kinkazan Island, northern Japan. Ecological Research 9:215–223. Ueda, H., S. Takatsuki, and Y. Takahashi. 2003. Seasonal change in browsing by sika deer on hinoki cypress trees on Mt. Takahara, central Japan. Ecological Research 18:355–364.

8

Food Habits of Sika Deer on Kinkazan Island, Northern Japan

123

Watson, A., and B. Staines. 1978. Differences in the quality of wintering areas used by male and female red deer (Cervus elaphus) in Aberdeenshire. Journal of Zoology 186:544–550. Whittaker, R. H. 1952. A study of summer foliage insect communities in the Great Smoky Mountains. Ecological Monographs 22:1–44. Yokoyama, M., K. Kaji, and M. Suzuki. 2000. Food habits of sika deer and nutritional value of sika deer diets in eastern Hokkaido, Japan. Ecological Research 15:345–355. Yoshii, Y., and K. Yoshioka. 1949. The vegetation of Kinkazan Island. Ecological Review (Sendai) 12:84–05. (In Japanese.)

Chapter 9

Plants and Plant Communities on Kinkazan Island, Northern Japan, in Relation to Sika Deer Herbivory Seiki Takatsuki and Takehiko Y. Ito

Abstract Due to an extremely high density of sika deer population (50 deer/km2), the vegetation on Kinkazan Island, northern Japan, is heavily affected. Unpalatable plants including forbs, ferns, and shrubs containing secondary compounds such as Senecio cannabifolis (Aleutian ragwort), Primula japonica (primrose), Pteridium aquilinum (bracken fern), and Leucothoe grayana (Ericaceae) are abundant on Kinkazan Island. The forest structure is modified, the shrub layer is quite poor, and regeneration of the forest is prevented by deer browsing. Forest gaps are formed by strong winds, and the gaps are invaded by Miscanthus sinensis (silver grass). At the density of 100–200 deer/km2, the Miscanthus community is maintained, but it is often concentrated on by sika deer because of great forage biomass, and the deer density often exceeds this level. Thus, Miscanthus cannot tolerate heavy grazing, and the Miscanthus community is replaced by the Zoysia japonica (Japanese lawngrass) community which is the most grazing-tolerant lawn-type community. The Zoysia community is favored by deer grazing because grazing removes other competitive tall plants while Zoysia japonica can produce leaves under a good light condition. Z. japonica also owes seed dispersal to sika deer. Thus, forest is decreasing and grassland communities, particularly the Zoysia community, are increasing on Kinkazan Island.

Introduction Cervid herbivory sometimes asserts tremendous effects on vegetation. Selective foraging by deer can affect plant species composition, structure, and succession (Healy 1997). Deer can be a contributing factor in the extinction of endangered plants (Miller et al. 1992). Several scientists think that deer effects on forests are much more serious than commonly thought (Whitney 1984; Anderson and Katz 1993; Boerner and Brinkman 1996; deCalesta 1997). This seems to be the case for Kinkazan Island, a small island (9.6 km2) located on the Pacific side of the Tohoku (northeastern) District of Japan (Fig. 9.1). The island has religious significance, and plants and animals have been protected. D. R. McCullough et al. (eds.), Sika Deer: Biology and Management of Native and Introduced Populations, DOI: 10.1007/978-4-431-09429-6_9, © Springer 2009

125

126

S. Takatsuki, T. Y. Ito 1398E

1408E

1418E

1428E

418N

418N

Honshu

408N

398N

408N

398N

Kawatabi Farm Oshika Peninsula

388N

388N

Kinkazan Island

1398E

1408E

1418E

1428E

Fig. 9.1 Location map of Kinkazan Island, Japan and other locations mentioned in this chapter.

Sika deer (Cervus nippon) and Japanese macaques (Macaca fuscata) occur there, and old-growth forests of Fagus crenata (Japanese beech) and Abies firma (Japanese fir) still remain (Yoshii and Yoshioka 1949). The population of sika deer is estimated at about 500 (Takatsuki and Ito 1999), and the density is as high as 50/km2. This is because hunting is prohibited and no predators occur on the island. Because of the high density and large body size of sika deer (males average 50 kg and females 35 kg), grazing and browsing heavily affect the island vegetation. In a pioneer study on the vegetation of this island, Yoshii and Yoshioka (1949) pointed this out and listed plants unpalatable to the deer. We describe these unpalatable plants and add some plants newly discovered to be unpalatable to sika deer on this island. A second important effect of deer on the Kinkazan vegetation is prevention of forest regeneration. Because of the high density of deer, almost all plants are exposed to deer grazing and browsing. Although the density of sika deer is lower in the forests (Takatsuki 1983, Takatsuki chapter 16), plants in the undergrowth are also affected by the deer. The forests on Kinkazan Island are composed of big trees,

9

Plants and Plant Communities on Kinkazan Island, Northern Japan

127

which are rarely seen on the main island. These trees produce seeds and fruits, and seedlings are seen, but few saplings are found. Consequently, the shrub layer is often sparse (Yoshiola and Kashimura 1959). It is presumed, therefore, that young trees are eaten and killed by the deer. In order to evaluate this assumption, we compared the densities of seedlings, saplings, and trees in the forest of Kinkazan Island and other areas (Takatsuki and Gorai 1994; Takatsuki and Hirabuki 1998). Here we describe the effects of the deer on the forest regeneration with some additional data on Kinkazan plant communities. Thirdly, Yoshioka and other colleagues made important contributions to plant ecology in Japan by pointing out the importance of effects of deer on plant communities, which had been ignored by Japanese plant ecologists. They had concluded, however, that some plants were more abundant on this island simply because of their unpalatability; that is, they were more abundant because sika deer avoided them. It was only recently that plant ecologists recognized that some groups of plants increase not by avoidance of herbivory but rather being favored by herbivory. Some savanna grasses in East Africa show higher productivity at moderate grazing intensity than under nongrazing, which is called “overcompensation” (McNaughton 1983). A typical example is found in short grasses which are tolerant to grazing. Under grazing conditions, they can be comparatively more dominant than other plants which are more vulnerable to grazing. Such grasses are termed as “lawn grasses” (McNaughton 1984) and can dominate others which are eliminated by grazing. On Kinkazan Island there are short-grass communities where many sika deer concentrate to forage. A typical example is the Zoysia japonica (Japanese lawngrass) community (Ito et al. chapter 10). These short-grass communities have been spreading on the island because of continuous heavy grazing by sika deer (Takatsuki 1999). The objective of this chapter is to describe the above-mentioned three effects of sika deer on the plants and plant communities on Kinkazan Island by reviewing the former studies and adding several new study results.

Study Area Kinkazan Island is only 9.6 km2 in extent and 600 m off the coast of Oshika Peninsula in the eastern part of the Tohoku District (Fig. 9.1). The mean monthly temperature fluctuates from 0.5 °C in January to 23.4 °C in August. The annual precipitation is 1,200 mm, mostly concentrated in summer. Topography of the island is steep with sharp ridges. Climatic climax forests of Fagus crenata and Abies firma cover the upper and the lower parts of the island, respectively. Forest structures are simple and composed of large trees with poorly developed undergrowth. Among these forests there are many patches of grassland and forb communities. About 500 deer live on the island (Takatsuki and Ito 1999). After World War II when United States soldiers hunted the deer the population was reduced to about 100 deer. It rapidly recovered during the 1950s and 1960s, and the first census in 1966 showed that there were 450 deer (Asahi et al. 1967). The population

128

S. Takatsuki, T. Y. Ito

stabilized at around 500 deer (50 deer/km2), but several mass mortalities occurred due to hard winters. One was in the spring of 1984 (Takatsuki et al. 1994). The population recovered thereafter, but another mass mortality occurred in 1997 (Takatsuki unpublished). Because of the extremely high density, the deer always face food shortage and chronic malnutrition, which results in size reduction of the deer compared with mainland sika (30% reduction for males and 20% for females) and in low female fecundity (Takatsuki et al. unpublished). The density is higher in grassland communities than in the forest where deer are dependent on graminoids such as Zoysia japonica, Miscanthus sinensis (silver grass), and Pleioblastus chino (dwarf bamboo) as foods (Takatsuki 1980).

Unpalatable Plants There are many plants unpalatable to sika deer on Kinkazan Island. Communities dominated by Senecio cannabifolius (Aleutian ragwort) (Fig. 9.2a), Pteridium aquilinum (bracken fern), and Hypolepis punctata (bead fern) are relatively common. The Senecio community occurs in open and humid habitats. The Pteridium community often occurs in open, dry habitats. Leucothoe grayana (Ericaceae), an unpalatable shrub, grows in the understory of the Fagus crenata forest (Fig. 9.3). Many other unpalatable plants are more abundant on this island than in other areas of Japan, but they do not form conspicuous communities like the above-mentioned associations. For example, Cirsium amplexifolium var. muraii (a thistle) (Fig. 9.2b) often grows in open lands and forest edges. It has very sharp spines on the leaf edges and flowers (involucres). Zanthoxylum piperitum (Japanese pepper tree), a spiny shrub, is frequently found in humid habitats. Berberis thunbergii (Japanese barberry), another spiny shrub (Fig. 9.2c), grows abundantly in open lands, particularly in grassland communities. The twigs are trimmed by sika deer, and consequently the shrubs have become “bonsai” in form. Unpalatable forbs include Paeonia obovata (Japanese peony) (Fig. 9.2d), Arisaema undulatifolium var. ionostemma and Arisaema urashima (Araceae, arums), Primula japonica (primrose) (Fig. 9.2e), Phytolacca esculenta (pokeweed) (Fig. 9.2f), Leonurus japonicus (Labiatae) (Fig. 9.2g), Caryopteris divaricata (blue spirea) (Fig. 9.2h), and Chloranthus serratus (Chloranthaceae). Paeonia obovata grows in forest gaps, and Primula japonica grows along streams. Perilla frutescens (beefsteak plant), a cultivated fragrant forb, grows along roadsides. The exotic forbs Erechtites hieracifolia (fireweed) and Crassocephalum crepidioides (redflower ragleaf) grow on open, disturbed lands. There are other unpalatable plants although they are not particularly abundant on Kinkazan Island: Datura stramonium (Jimsonweed), Elsholzia ciliate (Vietnamese balm), and Sambucus racemosa var. sieboldiana (red elderberry). The above listed plants are strongly defended by spines and/or secondary compounds, and the deer rarely feed on them. Some others, however, are sometimes eaten, although they are not preferred, for example, Rubus microphylla and R. palmatus (bramble), Aralia elata (angelica tree), and Boehmeria sieboldiana (Urticaceae).

9

Plants and Plant Communities on Kinkazan Island, Northern Japan

129

Fig. 9.2 Plants unpalatable to sika deer on Kinkazan Island. (a) Senecio cannabifolius, (b) Cirsium amplexifolium var. muraii, (c) Berberis thunbergii, (d) Paeonia obovata, (e) Primula japonica, (f) Phytolacca esculenta, (g) Leonurus japonicus, (h) Caryopteris divaricata.

130

S. Takatsuki, T. Y. Ito

Forest Regeneration Yoshii and Yoshioka (1949) pointed out that the forest of Kinkazan Island is different from others in Japan. The subcanopy and the shrub layers are poorer (Figs. 9.3 and 9.10). They attributed this unusual forest structure to deer browsing, although they did not present quantitative data. Since the undergrowth in deciduous forests of adjacent mainland areas are covered by Sasamorpha borealis, a dwarf bamboo, it is quite probable that the undergrowth of the Fagus forest on Kinkazan Island also was once covered by it. Now it has been replaced by L. grayana (Fig. 9.3) and S. borealis is found on Kinkazan only in the cracks of rocks (Fig. 9.4). Takatsuki and Gorai (1984) showed the size distributions of trees and the densities of seedlings and saplings in the Fagus crenata forest of Kinkazan Island and compared them with the forests of the main island as controls (Nakashizuka and Numata 1982a, b). Contrary to the control forests where frequency distributions of tree size showed an L-shape, the tree size distribution of the Fagus forest of Kinkazan Island showed a dome-shape with the peak at around 40–50 cm DBH, with a few very large trees (>80 cm) (Fig. 9.5). A remarkable difference was the scarcity of young trees smaller than 10 cm DBH. We counted some young Fagus trees at four different sites: two were outside a deer-proof fence, another two were inside the fence. One plot outside the fence was located in the unpalatable Leucothoe grayana mat in the understory of the Fagus forest (Fig. 9.3) and another was not in a Leucothoe mat. Very few young trees grew outside of a Leucothoe mat, where most regeneration occurred (Fig. 9.6). Inside the fence, we took sampling plots both beneath and away from the Fagus crenata canopy. There were more young trees under the canopy. Although fewer individuals occurred where there

Fig. 9.3 View of the Fagus crenata forest on Kinkazan Island. Note the poor subtree and shrub layers, and the development of a Leucothoe grayana “mat” in the understory.

9

Plants and Plant Communities on Kinkazan Island, Northern Japan

131

Fig. 9.4 A clump of Sasamorpha borealis surviving in a crack of a rock in the forest where it is protected from sika deer browsing.

Fig. 9.5 Frequency distributions of DBH (diameter at breast height) of composite trees in the Fagus forest on Kinkazan and other two control forests (Mt. Moriyoshi and Kayano Flat) on the main island (from Nakashizuka and Numata 1982a, b).

was no canopy, some of these were very large (Fig. 9.6). The height distributions were different within and outside the Leucothoe mat. All the young trees outside were shorter than 40 cm, the canopy height of the Leucothoe grayana mat, while some

132

S. Takatsuki, T. Y. Ito

Fig. 9.6 Frequency distributions of DBH of young Fagus trees in different habitats. (A) outside the Leucothoe grayana mat outside a deer-proof fence, (B) within the Leucothoe mat outside the fence, (C) under the Fagus canopy inside the fence, (D) not under the Fagus canopy inside the fence. DBH classes are –5 = 0–5 mm, –10 = 6–10 mm, etc.

grew taller than 40 cm within the Leucothoe mat (Fig. 9.7). These data demonstrate that Fagus does produce seeds but that seedlings and saplings are eaten by deer. To follow seedling survival, we traced the survival of Fagus seedlings both inside and outside a fence from 1994 to 1996. The autumn of 1993 was a masting year for Fagus crenata. In May, 1994, we labeled the seedlings in 56 1 m × 1 m quadrat inside and 60 quadrats outside the fence. The initial densities were 2.1 and 1.6/m2, respectively. Figure 9.8 shows the survival curves of the Fagus seedlings. Decrease was more rapid outside; they had decreased by more than half as early as September, 1994, and to 11% by October, 1995. In contrast, the decrease was more

9

Plants and Plant Communities on Kinkazan Island, Northern Japan

133

Fig. 9.7 Height distributions of Fagus trees outside of (left) and within (right) the Leucothoe grayana mat in the Fagus crenata forest on Kinkazan Island.

Fig. 9.8 Survival curves of young Fagus trees inside and outside a deer-proof fence in the Fagus crenata forest on Kinkazan Island.

134

S. Takatsuki, T. Y. Ito

gradual inside the fence, and seedlings had only decreased to 60% by October 1995. In 1996, no seedling survived outside while 42% of the seedlings survived inside. These results indicate that the deer feeding reduced the seedlings and eliminated them within three years. We also examined the Abies firma forests on Kinkazan Island and on the opposite Oshika Peninsula (Fig. 9.1; Takatsuki and Hirabuki 1998). Among the three plots on Kinkazan Island, two plots were composed of very large Abies trees and smaller deciduous broad-leaved trees while one plot (Kinkazan 1) was composed of comparatively smaller Abies trees with many other coniferous trees (mainly Torreya nucifera (Japanese nutmeg-yew) ) (Fig. 9.9). The undergrowth of the Abies forest on Kinkazan Island is quite sparse (Fig. 9.10). In contrast, there were various types of the Abies forests on Oshika Peninsula. Oshika plot 1 was composed of numerous small Abies trees with similar sized broad-leaved trees, Oshika plots 2 and 3 were composed of fewer small Abies trees with many broad-leaved trees, and Oshika plot 4 was composed of larger Abies trees. Despite this variation, young Abies trees were more abundant on Oshika Peninsula than on Kinkazan Island, and the undergrowth was densely covered by lianas, shrubs, forbs, and ferns. Thus, the Abies forest on Kinkazan Island has been strongly affected by deer browsing, and forest regeneration has been prevented.

Grassland Communities The forest on Kinkazan Island contains gaps formed by wind-thrown old trees (Fig. 9.11). Gaps often are enlarged by blowdown of an adjacent tree exposed to winds which formerly were blocked by the continuous forest canopy. Consequently, small gaps often are enlarged. When the diameter of a gap exceeds about 30 m, light conditions on the ground are improved, and Miscanthus sinensis, a tall grass, invades these larger gaps (Fig. 9.12). M. sinensis is not always a good forage because the leaf has sharp prickles containing silica at the leaf edge. However, due to its nontoxicity and abundance, the deer are attracted by the Miscanthus community and the density on this community is as high as 200 deer/km2 (Takatsuki 1983). They subsist on it to the extent it is available in the habitat (Takatsuki 1980). The leaves of Miscanthus sinensis are often clipped by the deer in spring when young succulent leaves develop. With removal, new leaves often regrow beside the clipped leaves. Because of repeated grazing, the Miscanthus community on Kinkazan Island is shorter than is usual for Miscanthus communities. The vertical structures of the Miscanthus community on Kinkazan Island and on Oshika Peninsula clearly demonstrate how heavily the former is grazed (Fig. 9.13). Heavy grazing also resulted in size reduction of Miscanthus leaves. We compared the width of Miscanthus leaves of Kinkazan Island with those at Kawatabi Farm of Tohoku University in northern Miyagi Prefecture (Fig. 9.1). The former was significantly narrower (mean = 32.8 mm) than the latter (mean = 56.3 mm, t-test, DF = 38, t = −7.564, p < 0.001). When grazed M. sinensis on Kinkazan Island is protected

9

Plants and Plant Communities on Kinkazan Island, Northern Japan 100

100

Kinkazan 2

Kinkazan 1 50

50

0

0- 10- 20- 30- 40- 50- 60- 70- 80- 90- 100-110-120-130-140-150-

0

0- 10- 20- 30- 40- 50- 60- 70- 80- 90- 100-110-120-130-140-150-

100

Abies firma Other conifers Broad-leaved trees

Kinkazan 3

Density (number / ha)

135

50

0 0- 10- 20- 30- 40- 50- 60- 70- 80- 90- 100-110-120-130-140-150722 289 389

556 300

300 250

Oshika 1

250

200

200

150

150

100

100

50

50

0

Oshika 2

0 0- 10- 20- 30- 40- 50- 60- 70- 80- 90- 100-110-120-130-140-150-

0- 10- 20- 30- 40- 50- 60- 70- 80- 90- 100-110-120-130-140-150411 300 250

300

Oshika 3

250

200

200

150

150

100

100

50

50

Oshika 4

0

0 0- 10- 20- 30- 40- 50- 60- 70- 80- 90- 100-110-120-130-140-150-

0- 10- 20- 30- 40- 50- 60- 70- 80- 90- 100-110-120-130-140-150-

Fig. 9.9 Frequency distributions of DBH (diameter at breast height) of composite trees in the Abies firma forest on Kinkazan (1–3) and on the Oshika Peninsula (1–4, Takatsuki and Hirabuki 1998). Note that scales of the density (Y axis) are different between Kinkazan Island and Oshika Peninsula.

by a fence, it recovers the leaf width within two years after protection. In 1995, we measured the widths of Miscanthus leaves inside and outside a fence that was established in 1993. The mean width inside was 49.3 mm which was significantly wider than that of outside (mean = 32.3 mm, t-test, DF = 38, t = −5.000, p < 0.001). Due to tolerance to grazing, the Miscanthus community is maintained at many places on Kinkazan Island. The community is often accompanied by the shrubs Viburnum dilatatum (linden) and Berberis thunbergii, graminoids like Brachypodium sylvaticum, Agrostis clavata, Festuca rubra, and Carex humilis var. nana, and forbs like Potentilla freyniana (cinquefoil), Viola grypoceras (violet), and Hydrocotyle ramiflora (Apiaceae). Table 9.1 shows the species composition of an area at the northern end of the island dominated by Miscanthus sinensis, but where Cirsium

136

S. Takatsuki, T. Y. Ito

Fig. 9.10 Inside view of the Abies firma forest on Kinkazan Island.

Fig. 9.11 Fallen trees of Fagus crenata in the Fagus forest on Kinkazan Island.

9

Plants and Plant Communities on Kinkazan Island, Northern Japan

137

Fig. 9.12 Large forest gap invaded by Miscanthus sinensis, a tall grass (gap formation is accelerated by wind throw in a domino effect).

Fig. 9.13 Component structures of the Miscanthus sinensis community on Oshika Peninsula and on Kinkazan Island.

amplexifolium var. muraii, an unpalatable forb, grew abundantly. Coverage and height of the composite species were measured within 10 1 m × 1 m plots. As many as 31 forbs species were found, but in small amounts; 14 graminoids appeared and were more abundant. Browse species were less dominant and most of these were shrubs (Takatsuki 1980).

138

S. Takatsuki, T. Y. Ito

Table 9.1 Dominance of composite species in the Miscanthus sinensis community in a western part of Kinkazan Island. Figures are Summed Dominance Ratio (SDR) (Numata and Yoda 1957). Growth forms are: b = browse, bl = woody liana, bu = woody unpalatable, fe = fern, fo = forb, fol = herbaceous liana, fou = unpalatable forb, g = graminoid (modified from Takatsuki 1980). Species Growth form Plot 2 Miscanthus sinensis Cirsium amplexifolium var. muraii Carex lanceolata Pleioblastus chino Brachypodium sylvaticum Stephanandra incisa Lonicera gracilipes Zanthoxylum piperitum Caryopteris divaricata Rubus microphyllus Viburnum dilatatum Potentilla freyniana Bromus pauciflorus Agrostis clavata Pennisetum alopecuroides Calamagrostis arundinacea Paederia scandens var. mairei Agrimonia pilosa Smilax china Oxalis corniculata Clinopodium chinense var. parviflorum Zelkova serrata Clinopodium micranthum Clematis apiifolia Galium trachyspermum Viola grypoceras Ixeris dentata Clinopodium gracile var. multicaule Trisetum bifidum Lysimachia japonica Lespedeza pilosa Perilla frutescens var. citriodora Carpinus tschonoskii Zoysia japonica Festuca rubra var. rubra Rhododendron juponicum Chamaele decumbens Geranium tripartitum Phryma leptostachya var. asiatica Desmodium oxyphyllum Hydrocotyle ramiflora Lycopus maackianus Plectranthus inflexus Paspalum thunbergii

G Fo G G G B B Bu Fo Bu B Fo G G G G Fo Fo Bu Fo Fo B Fo Bl Fo Fo Fo Fo G Fo Fo Fo B G G Bu Fo Fo Fou Fo Fo Fo Fo G

100 62.8 29.6 29.2 29.1 28.0 23.1 18.7 18.4 17.1 16.5 15.6 15.1 15.0 14.6 14.2 9.9 9.6 8.1 7.8 7.0 6.6 6.5 5.5 5.3 5.1 4.8 4.5 4.2 4.1 4.0 3.8 3.5 3.5 3.2 2.9 2.8 2.5 2.5 2.2 2.2 2.2 2.2 2.0 (continued)

9

Plants and Plant Communities on Kinkazan Island, Northern Japan

Table 9.1 (continued) Species Moehringia lateriflora Polygala japonica Hypericum erectum Oplismenus undulatifolius Galium paradoxum Lapsana humilis Cerastium holosteoides var. hallaisanense Luzula capitata Geranium thunbergii Leibnitzia anandria Ajuga decumbens

Growth form Fo Fo Fo G Fo Fo Fo G Fo Fo Fo

139

Plot 2 1.6 1.6 1.6 1.6 1.4 1.4 1.2 1.0 0.8 0.8 0.6

Fig. 9.14 View of an openland in the western part of Kinkazan Island where a Miscanthus sinensis community in 1983 (left) was replaced by the Zoysia japonica community by 1999 (right) due to heavy grazing by sika deer. Berberis thunbergii (shrubs) and Cirsium amplexifolium var. muraii survived.

At a higher deer density, Miscanthus sinensis cannot survive and is replaced by Zoysia japonica, a lawn grass. This short grass is quite productive under good light conditions (Kira 1952; Inoue and Sasaki 1958; Ito et al. chapter 10). That is, heavy grazing removes other taller plants and facilitates the growth and development of Zoysia japonica. In the 1970s, Zoysia japonica contributed to the deer diets as an important forage in habitats where it was available (Takatsuki 1980). During the last two decades, the Zoysia community has enlarged its distribution on Kinkazan Island (Fig. 9.14). Accordingly, the contribution of Z. japonica to the deer diet increased in the 1990s as the Zoysia community developed (Ito et al. chapter 10). Zoysia japonica seems to be a preferred forage for the deer. For example, the bite rate of sika deer is much higher on the Zoysia community than in other plant communities (Ito et al. chapter 10). Seed dispersal by sika deer also enhances the development of the Zoysia community (Takatsuki 1999). Z. japonica produces abundant large seeds at the top of short peduncles, most of which are eaten by the deer. A feeding experiment showed that the survival rate of the seeds passing through the digestive tract of sika deer was about 40%, and the germination of the

140

S. Takatsuki, T. Y. Ito

recovered seeds was higher than that of intact seeds (Imaei 1992). This seems to be a good example of the “foliage is a fruit” theory of Janzen (1984). He proposed that foliage is ecologically a fruit in that the plants afford foliage as bait foods to herbivores and consequently utilize animals as seed dispersal agents, as berrybearing trees do with birds. The Zoysia community and the Miscanthus community are interchangeable depending on grazing intensity of the deer. When a Zoysia community is protected by a fence, it is replaced by Miscanthus within several years. Table 9.2 shows the species compositions of the communities inside (Miscanthus community) and outside (Zoysia community) five years after establishment of a fence. It is noticeable that biomass increased sixfold inside the fence. Zoysia japonica was the superdominant species (93.1% of the total biomass) outside, whereas inside Miscanthus sinensis accounted for 67.1% and some other important plants such as Stephanandra incisa (Rosaecea) and Pleioblastus chino (a bamboo) also grew inside the fence. Growth form composition was also different; only two woody plants and no lianas grew outside whereas five woody plants and four lianas grew inside. A view of the fence 10 years after establishment is shown in Fig. 9.15. Thus for the Zoysia community, grazing of sika deer functions to (1) remove other competitive taller plants and (2) disperse seeds away from mother plants. It is

Table 9.2 Dry weight of each species inside and outside a deerproof fence five years after establishment on Kinkazan Island. Outside Inside Zoysia japonica Carex breviculmis Brachypodium sylvaticum Paspalum thumbergii Viola obtusa Liriope minor Agrostis clavata var. nukabo Haloragis micrantha Luzula capitata Festuca rubra Botrychium ternatum Polygala japonica Lespedeza juncea var. subsessilis Gnaphalium japonicum Lysimachia japonica Ixeris dentata Hydrocotyle ramiflora Plantago asiatica Viburnum dilatatum Galium pogonanthum Oxalis corniculata Miscanthus sinensis

Dry weight (g/m2)

(%)

144.6 ± 53.0 2.2 ± 1.7 1.5 ± 1.2 0.8 ± 1.5 0.7 ± 1.3 0.8 ± 0.2 0.8 ± 0.3 0.4 ± 0.8 0.5 ± 0.4 0.3 ± 0.2 0.2 ± 0.2 0.2 ± 0.2 0.1 ± 0.2 0.1 ± 0.1 0.1 ± 0.1 0.0 ± 0.1 0.0 ± 0.1 0.0 ± 0.0 0.0 ± 0.0 0.0 ± 0.0 0.0 ± 0.0

93.1 1.8 1.2 0.8 0.6 0.6 0.5 0.4 0.4 0.2 0.1 0.1 0.1 0.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0

Dry weight (g/m2)

(%)

0.8 ± 1.5 0.6 ± 0.8 0.8 ± 1.3 9.1 ± 9.7

0.1 0.1 0.1 0.9

0.0 ± 0.0 0.7 ± 0.8

0.0 0.1

0.0 ± 0.1

0.0

0.0 ± 0.0

0.0

41.7 ± 83.4 0.1 ± 0.2 0.1 ± 0.2 609.7 ± 204.3

3.6 0.0 0.0 67.1 (continued)

9

Plants and Plant Communities on Kinkazan Island, Northern Japan

141

Table 9.2 (continued) Outside Dry weight (g/m2) Stephanandra incisa 106.2 ± 117.9 Pleioblastus chino Smilax china Carex lanceolata Calamagrostis epigeios Zanthoxylum piperitum Cirsium amplexifolium var. muraii Rubus microphyllus Akebia quinata Lespedeza pilosa Dioscorea tokoro Potentilla freyniana Phryma letostachya var. asiatica Scilla scilloides Moehringia lateriflora Geranium nepalens ssp. thunbergii Chamaele decumbens Muhlenbergia longiatolon Abies firma Total 153.2

Inside (%)

Dry weight (g/m2)

(%)

11.2

100

53.5 ± 43.1 42.6 ± 85.3 21.7 ± 7.7 11.6 ± 18.0 6.0 ± 11.9 5.4 ± 7.0 2.1 ± 2.7 1.8 ± 2.4 1.4 ± 1.2 1.1 ± 2.3 0.8 ± 0.6 0.5 ± 1.1 0.5 ± 0.3 0.4 ± 0.5 0.0 ± 0.0 0.0 ± 0.0 0.0 ± 0.0 0.0 ±0.0 920.5

6.2 4.1 2.6 1.1 1.2 0.5 0.2 0.3 0.1 0.2 0.1 0.1 0.1 0.0 0.0 0.0 0.0 0.0 100

Fig. 9.15 A fence established in the Zoysia japonica community on Kinkazan Island. This photo was taken in June, 2002, 12 years after establishment.

142

S. Takatsuki, T. Y. Ito

also plausible that Zoysia is fertilized by sika defecation, which may enhance establishment of new plants. Overall, despite the “rich green” appearance of Kinkazan Island, the plants are not always good forage for the deer. Many common palatable plants have been replaced by unpalatable plants. Forest regeneration is prevented because seedlings and saplings are fed on by the deer. Only old, big canopy trees remain but they are lost as they become decadent and fall down. This results in gap formation, and small openings are enlarged as further trees, exposed to wind, fall in turn. At an early phase of gap formation, shrubs and tall grasses grow, but subsequently they are replaced by the Zoysia community because the deer concentrate in these openings and the balance of the former plant groups cannot tolerate deer feeding effects and they decrease. Acknowledgements This series of studies were done with the following personnel: T. Gorai (Tohoku University), Dr. Y. Hirabuki (Miyagi Educational University), F. Shibata (Tohoku University), I. Imaei (Tohoku University), A. Kimura, M. Kikuchi, T. Miura, and H. Kawakatsu (Macky School). All photos were taken by S. Takatsuki.

Literature Cited Anderson, R. C., and A. J. Katz. 1993. Recovery of browse-selective tree species following release from white-tailed deer (Odocoileus virginianus Zimmerman) browsing pressure. Biological Conservation 63: 203–208. Asahi, M., S. Azuma, T. Ito, M. Kawai, and K. Hayashi. 1967. A faunal survey on large mammals in Kinkazan Island, Miyagi Prefecture - sika deer. Annual Report of International Biological Program, Japan Branch-CTS for 1966: 189–196. Boerner, R. E., and J. A. Brinkman. 1996. Ten years of tree seedling establishment and mortality in an Ohio deciduous forest complex. Bulletin of the Torrey Botanical Club 123:309–317. deCalesta, D. S. 1997. Deer and ecosystem management. Pages 267–279 in W. J. McShea, H. B. Underwood, and J. H. Rappole, editors, The science of overabundance: Deer ecology and population management. Smithsonian Institution Press, Washington, DC, USA. Healy, W. M. 1997. Influence of deer on the structure and composition of oak forests in Central Massachusetts. Pages 249–266 in W. J. McShea, H. B. Underwood, and J. H. Rappole, editors, The science of overabundance: Deer ecology and population management. Smithsonian Institution Press, Washington, DC, USA. Imaei, H. 1992. Seed ecology of the Zoysia japonica community on Kinkazan Island. Graduate thesis, Department of Biology, Faculty of Science, Tohoku University, Sendai, Japan. (In Japanese.) Inoue, T., and T. Sasaki. 1958. The effects of shading on sod. Report of Tohoku Agricultural Institute 14:92–103. Janzen, D. H. 1984. Dispersal of small seeds by big herbivores: Foliage is the fruit. American Naturalist 123:338–353. Kira, T. 1952. So called “overgrazed pastures” in ecological context. Bulletin of the Society of Plant Ecology 1:209–213. McNaughton, S. J. 1983. Compensatory plant growth as a response to herbivory. Oikos 40:329–336. McNaughton, S. J. 1984. Grazing lawns: animals in herds, plant form, and coevolution. American Naturalist 124:863–886. Miller, S. G., S. P. Bratton, and J. Hadidian. 1992. Impacts of white-tailed deer on endangered and threatened vascular plants. Natural Areas Journal 12:67–74.

9

Plants and Plant Communities on Kinkazan Island, Northern Japan

143

Nakashizuka, T., and M. Numata. 1982a. Regeneration process of climax beech forests, I. Structure of a beech forest with the undergrowth of Sasa. Japanese Journal of Ecology 32:57–67. Nakashizuka, T., and M. Numata. 1982b. Regeneration process of climax beech forests, II. Structure of a climax beech forest under the influence of grazing. Japanese Journal of Ecology 32:473–482. Numata, M., and K. Yoda. 1957. The community structure and succession of artificial grassland, I. Journal of Japanese Society of Grassland Science 3:4–11. Takatsuki, S. 1980. Food habits of sika deer on Kinkazan Island. Science Reports Tohoku University, Series IV (Biology) 38:7–31. Takatsuki, S., 1983. Habitat selection by sika deer on Kinkazan Island. Journal of the Mammalogical Society of Japan 9:183–191. Takatsuki, S. 1999. The Zoysia japonica community maintained by sika deer. Pages 65–85 in K. Ueda, editor, Processes of mutualism, 2. Seed dispersal. Tsukiji Publishing Company, Tokyo, Japan. Takatsuki S., and T. Gorai. 1994. Effects of sika deer on the regeneration of a Fagus crenata forest on Kinkazan Island, northern Japan. Ecological Research 9:115–120. Takatsuki, S., and Y. Hirabuki. 1998. Effects of sika deer browsing on the structure and regeneration of the Abies firma forest on Kinkazan Island, northern Japan. Journal of Sustainable Forestry 6:203–221. Takatsuki, S., and T. Ito. 1999. Population dynamics of sika deer on Kinkazan Island. Pages 85–97 in Miyagi Prefecture, editor, Conservation of Kinkazan Island. Sendai, Japan. (In Japanese.) Takatsuki, S., K. Suzuki, and I. Suzuki. 1994. A mass-mortality of sika deer on Kinkazan Island, northern Japan. Ecological Research 9:215–223. Whitney, G. G. 1984. Fifty years of change in the arboreal vegetation of Heart’s Content, an oldgrowth hemlock-white pine-northern hardwood stand. Ecology 65:403–408. Yoshii, Y., and K. Yoshioka. 1949. The vegetation of Kinkazan Island. Ecological Review (Sendai). 12: 84–105. (In Japanese.) Yoshioka, K., and T. Kashimura. 1959. Plant communities induced by deer grazing and browsing. Science Reports of the Faculty of Arts and Sciences, Fukushima University 8:9–14.

Chapter 10

Productivity and Foraging Efficiency of the Short-Grass (Zoysia japonica) Community for Sika Deer Takehiko Y. Ito, Mariko Shimoda, and Seiki Takatsuki

Abstract A sika deer population at an extremely high density (814 deer/km2) on Kinkazan Island in northeastern Japan heavily used a lawn grass, Zoysia japonica, community. Field experiments and behavior observations were done to determine the reason for this high use. The productivity of the Zoysia community was as high as 370 g/m2/year. The biomass concentration (150–180 g/cm3 below 5 cm) and the bite rate (about 50 times/min) were higher on the Zoysia community than those in the adjacent forest understory. These characteristics of the Zoysia community seem to attract the sika deer. The potential deer density supported was estimated as 1,840–2,875 deer/km2 during the growing period. This was twice the achieved deer density.

Introduction Because of its warm and humid climate, the vegetation of Japan is dominated by forests, and grassland communities appear only under disturbances such as mowing or grazing (Numata 1973). On small islands, for example, where the density of sika deer (Cervus nippon Temminck) sometimes reaches as high as 30–130 deer/km2 (e.g., Nakanoshima Island, Takahashi 1998; Nozaki Island, Doi and Endo 1995), small grassland communities appear (e.g., Takatsuki 1977, 1980), although sika deer usually live in forests at a low density (<5 deer/km2, Japan Wildlife Research Center 1999). On Kinkazan Island in northeastern Japan, an extremely high density deer population uses a short-grass community dominated by Zoysia japonica (Takatsuki 1983). As early as 1952, Kira (1952) showed the high productivity of the Z. japonica community to resolve the misunderstanding that plant communities of small biomass are less productive. It was shown in later years in other countries that “lawn grasses” with low stature are often highly productive (e.g., McNaughton 1984; Jefferies 1997). The Zoysia community seems to be a typical “lawn grass” and attracts sika deer. Our question was why this community attracted such a high density of sika deer. Therefore, during the plant growing period we quantified D. R. McCullough et al. (eds.), Sika Deer: Biology and Management of Native and Introduced Populations, DOI: 10.1007/978-4-431-09429-6_10, © Springer 2009

145

146

T. Y. Ito et al.

the productivity of the Zoysia community and compared the biomass concentration (plant biomass per unit volume, McNaughton 1976, 1984) and the bite rate as an index of foraging efficiency of the Zoysia community with those in the forest understory.

Methods Study Site Kinkazan Island (38°17′ N, 141°39′ E) is a small island (960 ha) lying off the coast of the Oshika Peninsula of Japan in the Pacific Ocean (Fig. 10.1). The island is regarded as a sacred place, and plants and animals have been conserved for over 300 years for religious reasons. The climate is warm and humid; the mean annual precipitation is 1,176 mm and the mean annual temperature is 12.3 °C. Most parts of the island are covered by climax forests of Fagus crenata (Japanese beech) and Abies firma (Japanese fir) (Yoshii and Yoshioka 1949). The density of sika deer is high because of the absence of predators and the prohibition of hunting. The population fluctuated from 400 to 600 after the 1970s, although it experienced two drastic decreases in 1984, which followed an unusually cold winter, and 1997, for which the factors were unclear (Ito 1985; Takatsuki et al. 1991, 1994; Takatsuki and Ito 1999). In 1999, the deer number was about 500 (Takatsuki and Ito 1999). Because of the high deer density, the plant communities have been heavily affected (Takatsuki and Gorai 1994; Takatsuki and Hirabuki 1998). The present study was carried out on a gentle slope in the western part of Kinkazan Island (Fig. 10.1). The size of the study area was 34.7 ha, about a half of which was occupied by the Zoysia community (17.1 ha; Table 10.1). Shrubs of Berberis thunbergii (Japanese barberry) grow sporadically on the Zoysia community. There are no large Zoysia communities on the island other than the study area and only a few small patches of Zoysia are found on ridges. The deer density in the study area is higher than the average on the island (Takatsuki 1983). Most of the forests in the study area were evergreen and deciduous forests dominated by Abies firma and Carpinus tschonoskii (Korean hornbeam) (Fig. 10.1, Table 10.1). Dominant species in the understory of these forests are Brachypodium sylvaticum (false brome), Carex spp., and Oplismenus undulatifolius (basketgrass).

Deer Distribution We compared the densities of the deer on the Zoysia community with those in other plant communities by recording deer numbers on a fixed route (Fig. 10.1).

10 Productivity and Foraging Efficiency of the Short-Grass

147

Fig. 10.1 Location of Kinkazan Island and vegetation map of the study area, delimited by the bold line. Numbers on contours indicate elevation in meters. F is an exclosure fence. The deer census route is shown.

Counts were conducted for five days in summer (11–16 August 1996), fall (9–18 November 1996), and spring (30 May–7 June 1997). Rainy days were excluded because of low visibility.

148

T. Y. Ito et al.

Table 10.1 The densities of deer (deer/km2) in each plant community in the study area. Standard deviations are shown in parentheses.

Area (ha) Spring Summer Fall

Zoysia community

Shrub and tall grass community

Evergreen forest

Deciduous forest

Pebbly shore

Total

17.1 686.7 (96.9) 813.9 (178.5) 512.9 (214.7)

1.8 101.6 (316.1) 564.5 (886.3) 135.5 (276.6)

10.3 3.9 (7.7) 21.3 (43.6) 112.3 (146.2)

4.0 20.1 (45.3) 22.7 (55.4) 57.9 (93.2)

1.5 0.0 (0.0) 0.0 (0.0) 0.0 (0.0)

34.7 348.2 (44.1) 19 440.2 (66.1) 18 300.6 (74.1) 18

n

Productivity of the Zoysia japonica Community The above-ground productivity of the Z. japonica community was measured by the movable-cage method (e.g., Frank and McNaughton 1993; McNaughton et al. 1996). Nine cages were placed in the Zoysia community from May 1994 to May 1995. The cage size was 70 × 70 × 70 cm. All above-ground plant biomass in four quadrats (10 × 10 cm) inside the cages and two quadrats outside the cages was clipped at ground level every month. The samples were brought to the laboratory, dried at 70 °C for 48 h, and weighed. On each sampling occasion, the cages were moved. Removal in month i (Ri) by deer was defined as: Ri= Ui – Gi where Ui is inside biomass (ungrazed) and Gi is outside biomass (grazed). Total biomass removed (grazed) (R) during the growing period is defined as: 11

R = ∑ Ri . i=5

Productivity in month i (Pi) is represented as the difference between Gi–1 and Ui: Pi= Ui – Gi–1. Because the plants in the Zoysia community begin to grow in mid-April, we regarded the standing biomass in the cage in May (U5) as the growth between April and May. However, because the stem of Z. japonica remains above the ground even in the nongrowing period, we subtracted the stem biomass of Z. japonica (S5) from the whole biomass in the cage (U5) to determine the productivity in May (P5): P5= U5 – S5 . Therefore, productivity throughout the growing period was calculated as:

10 Productivity and Foraging Efficiency of the Short-Grass

149

11

P = ∑ Pi . i=5

Potential Deer Densities on the Zoysia japonica Community We calculated the potential deer densities that could be supported by productivity of the Zoysia community from the monthly productivity and intake rate of the deer. The intake rate of a sika deer with a body weight of 22–30 kg was reported as 0.56–1.19 kg/day (Takatsuki 1976; Miyazaki et al. 1977). However, this body weight is smaller than the average of the sika deer on Kinkazan Island. Mould and Robbins (1982) reported that the intake rate of wapiti (Cervus elaphus nelsonii) with a body weight of W kg was from 0.035 to 0.085W0.75 kg dw/day. If this is also valid for sika deer, the intake would be 0.66–1.60 kg/day, as the body weight of adult sika deer on Kinkazan Island is around 50 kg (Takatsuki 2006). The potential deer densities on the Zoysia community can be calculated by dividing the daily removal of the Zoysia community by the daily intake of sika deer (median value: 1.13 kg/day). Although intake rate varies seasonally and that of fawns must be smaller, we used this as an approximate average value.

Comparison of the Zoysia Community and the Forest Understory In order to compare the food availability of the Zoysia community with that of the forest understory, all above-ground plant parts in 18 quadrats (10 × 10 cm) on the Zoysia community and eight quadrats (50 × 50 cm) in the forest understory were collected. Samples were dried at 70 °C for 48 h and weighed. Sampling was carried out on 19 August (summer), 24 November, 1994 (fall), and 19 May, 1995 (spring) on the Zoysia community, and on 27 November, 1997 (fall), 27 May, 1998 (spring), and 14 August, 1998 (summer) in the forest understory. It is known that high biomass concentration (forage mass per unit volume) leads to a high bite size (Stobbs 1973, 1975; McNaughton 1976, 1984). Therefore, higher biomass concentration is expected to be a factor attracting the deer. Thus, we compared biomass concentrations at 5 cm height intervals from the ground between the two communities.

Bite Rate The higher bite rate would be another factor to attract deer. We therefore arbitrarily selected females of sika deer and counted their bite rates (bites per minute) by observation with binoculars on the Zoysia community and in the forest understory. Observations were carried out in spring (May), summer (August), and fall (November) in 1999.

150

T. Y. Ito et al.

Results Selection of Plant Community by Deer The deer densities were always higher in the Zoysia community than in the other communities in the three seasons (Fisher’s exact probability test: p < 0.0001; Table 10.1, Fig. 10.2). The deer densities on the Zoysia community were highest in summer and lowest in fall, whereas those on the other communities showed an opposite pattern (Fig. 10.2).

Productivity of the Zoysia Community Biomass both inside and outside the cages changed seasonally (one-way repeated measures ANOVA; inside: F6,48 = 47.37, p <0.0001; outside: F6,48 = 22.20, p < 0.0001); it was high in spring and summer, and low in fall (Fig. 10.3a). It was highest in July both inside (251.2 ± 26.4 g/m2 [mean ± SD], n = 36) and outside (187.8 ± 29.6 g/m2, n = 18) of the cages. The biomass was consistently higher inside than outside the cages (two-way repeated measures ANOVA; F1,16 = 40.73, p < 0.0001; Fig. 10.3a). Productivity showed more prominent seasonal changes than biomass (one-way repeated measures ANOVA; F6,48 = 31.249, p < 0.0001; Fig. 10.3b). Productivity was high from May to August, and decreased sharply in September. The total productivity in the growing period was 366.4 ± 82.5 g/m2 (n = 9).

900

*

800 Density (deer / km2)

700

*

600

*

500 400 300 200 100 0 Spring

Summer

Fall

Fig. 10.2 Seasonal changes in deer densities on the Zoysia community (dotted) and other communities (solid). Asterisks indicate significant differences between the density of deer in each community and expected values calculated from the areas of the communities (Fisher’s exact probability test; *: p < 0.0001). Vertical bars show SE.

10 Productivity and Foraging Efficiency of the Short-Grass

151

Fig. 10.3 Seasonal changes in the productivity of the Zoysia japonica community. a: Aboveground biomass (solid: outside the cage, Gi; dotted: inside the cage, Ui). b: Productivity (Pi = Ui–1 – Gi). c: Removal by deer (Ri = Ui – Gi). Vertical bars show SE.

152

T. Y. Ito et al.

Removal of plant biomass by deer showed a similar pattern to that of standing biomass (F6,48 = 9.76, p < 0.0001; Fig. 10.3c). It was over 60 g/m2/month from June to August and about 30 g/m2/month from September to November. The total mass of removal in the growing period was calculated to be 329.6 ± 78.8 g/m2 (n = 9). This was 90.0% of the total productivity. The potential deer densities calculated from productivity measures on the Zoysia community were 3,480 deer/km2 in May and 1,840–2,875 deer/km2 from June to August, and then decreased to about 60–180 deer/km2 in September and October.

Food Availability and Biomass Concentration Plant biomass on the Zoysia community was over three times greater than that of the forest understory from spring to fall (t test: spring; t = −9.016, d.f. = 15, p < 0.0001, summer; t = −14.487, d.f. = 15, p < 0.0001, fall; t = 8.848, d.f. = 15, p < 0.0001; Fig. 10.4). The Zoysia community was short in height and almost all parts of the aboveground biomass were concentrated below 5 cm height (Fig. 10.5). This was mainly the case in the forest understory, also, where most biomass occurred under 5 cm height. However, a small part reached 20 cm height in summer (Fig. 10.5). Biomass concentrations under the 5 cm stratum were greater on the Zoysia community than in the forest understory (t test: spring; t = −9.273, d.f. = 15, p < 0.0001, summer; t = −15.951, d.f. = 15, p < 0.0001, fall; t = −9.807, d.f. = 15, p < 0.0001).

Fig. 10.4 Seasonal changes in the above-ground biomass of the Zoysia community (open circles) and in the forest understory (solid circles). Vertical bars show SE.

10 Productivity and Foraging Efficiency of the Short-Grass

153

Fig. 10.5 Productive structures of the Zoysia community (dotted) and the forest understory (solid) in different seasons. Horizontal bars show SE.

154

T. Y. Ito et al.

Fig. 10.6 Seasonal changes in bite rate of sika deer on the Zoysia community (open circles) and in the forests (solid circles). Vertical bars show SE.

Bite Rate Bite rates were significantly higher on the Zoysia community than in the forest understory (t-test: spring; t = −7.984, d.f. = 37, p < 0.0001, summer; t = 5.090, d.f. = 45, p < 0.0001, fall; t = 12.328, d.f. = 53, p < 0.0001; Fig. 10.6).

Discussion Deer density on the Zoysia community in summer (814 deer/km2; Fig. 10.2) was much greater than that on the rest of the island (50 deer/km2), which means that the deer are disproportionately concentrated on the Zoysia community. This value is extraordinarily high as a sika deer density. The next highest reported density was on grassland on Nozaki Island, southern Japan (400 deer/km2; Doi and Endo 1995). This concentration of deer is due to the high productivity of the Zoysia community, although confinement to a small island and the absence of predators and culling may also be factors. The productivity of the Zoysia community was quite high, and the removal for the year by the deer was 330 g/m2. The possible deer densities on the Zoysia community from May to August (1,840–3,480 deer/km2) were much higher than the achieved deer density (814 deer/km2; Fig. 10.2). This high productivity is due to the morphological character of the Zoysia community. Z. japonica can vigorously develop new rhizomes which produce many erect stems. Its productivity is amazingly high in spite of its small standing biomass.

10 Productivity and Foraging Efficiency of the Short-Grass

155

The annual productivity was about three times greater than the maximum standing biomass, and as much as 90% of the annual productivity was grazed by the sika deer. Besides the high productivity, the Zoysia community has more advantages as a feeding site than the forest understory. First, biomass was also more concentrated on the Zoysia community than the forest understory (Fig. 10.5). Greater biomass concentration increases food intake per bite (Stobbs 1973, 1975). Similar phenomena under heavy grazing also have been reported in East Africa (McNaughton 1976) and in Argentina (Sala et al. 1986). Second, bite rate was also higher on the Zoysia community (Fig. 10.6). It was more than 50 times/min in spring and summer. Experiments using moose (Alces alces), caribou (Rangifer tarandus), and white-tailed deer (Odocoileus virginianus) showed that the bite rates ranged from 27–50 times/min (Shipley and Spalinger 1992), which were lower than ours. There are some other studies under artificial conditions. The values were 40.6–61.4 times/min for wapiti (Cervus elaphus, Gedir and Hudson 2000) and 36.2–64.4 times/min for sheep (Newman et al. 1992). It is said that under such artificial conditions, bite rates tended to be higher than in natural conditions. It is noteworthy that the bite rate of sika deer on the Zoysia community was similar or even higher than these artificial bite rates. It is likely that this high bite rate is possible because the Zoysia community is exclusively composed of Zoysia japonica and therefore the structure is uniform. This would afford a high bite rate because sika deer can pick up the same plants without checking various heights, or testing the smell and taste of different plants, which they must do in mixed plant communities. There are some other possible explanations why the Zoysia community is a better feeding site than the other plant communities. The increased leaf/stem ratio would improve the food supply for the deer, since leaves are more nutritious and digestible than stems. Grazing reduces the community height and increases the proportion of plants of prostrate form that allocate less production to supporting organs. This change increases the leaf/stem ratio. Such a plant community must be more efficient for feeding. Besides, the ability of Z. japonica to produce new leaves under grazing pressure would also be favorable for the deer, since new leaves are more nutritious and digestible than old leaves (e.g., Coppock et al. 1983; Jaramillo and Detling 1988). The high productivity and high foraging efficiency of the Zoysia community attract deer, and in turn, deer grazing facilitates the development of the Zoysia community because heavy grazing eliminates tall plants and consequently benefits Zoysia (Takatsuki 1993). This interrelation seems to be an example of a “feeding loop” (Danell et al. 1985). We conclude that the very high density of sika deer on the Zoysia community is attributed to both its high productivity and high foraging efficiency thanks to its uniform structure and high biomass concentration. Acknowledgements We thank T. Hirose, S. Sakai, K. Hikosaka, M. Kawata, M. Minami, and N. Ohnishi for helpful comments; members of the Laboratory of Plant Ecology at Tohoku University

156

T. Y. Ito et al.

and many students for their help in the field work; K. Sato for the equipment design; and K. Izawa and the staff of Koganeyama Shrine on Kinkazan Island for facilitation of the field survey. This study was partly supported by the Sasakawa Scientific Research Grant from the Japan Science Society and the Pro Natura Fund from the Nature Conservation Society of Japan. Productivity of the Zoysia community and plant community selection by sika deer has been published as Ito, T. Y., and S. Takatsuki. 2005. Relationship between a high density of sika deer and productivity of the short-grass (Zoysia japonica) community: A case study on Kinkazan Island, northeastern Japan. Ecological Research 20:573–579.

Literature Cited Coppock, D. L., J. K. Detling, J. E. Ellis, and D. L. Coppock. 1983. Plant–herbivore interactions in a North American mixed-grass prairie. I. Effects of black-tailed prairie dogs on intraseasonal aboveground plant biomass and nutrient dynamics and plant species diversity. Oecologia 56:1–9. Danell, K., K. Huss-Danell, and R. Bergström. 1985. Interractions between browsing moose and two species of birch in Sweden. Ecology 66:1867–1878. Doi, T., and A. Endo. 1995. A reports of influences of Nozaki Dam in the Odika area. Laboratory of Ecology, Faculty of Science, Kyushu University, Fukuoka, Japan. (In Japanese.) Frank, D. A., and S. J. McNaughton. 1993. Evidence for the promotion of aboveground grassland production by native large herbivores in Yellowstone National Park. Oecologia 96:157–161. Gedir, J. V., and R. J. Hudson. 2000. Seasonal foraging behavioural compensation in reproductive wapiti hinds (Cervus elaphus canadensis). Applied Animal Behaviour Science 16:137–150. Ito, T. 1985. Population dynamics of sika deer on Kinkazan Island. Pages 11–25 in: Report on continuing research for conservation of Kinkazan Island. Miyagi Prefecture, Sendai, Japan. (In Japanese.) Japan Wildlife Research Center. 1999. Management of sika deer population: Proceedings of a workshop on sika deer management in 1998. Japan Wildlife Research Center, Tokyo, Japan. (In Japanese.) Jaramillo, V. J., and J. K. Detling. 1988. Grazing history, defoliation, and competition: Effects on shortgrass production and nitrogen accumulation. Ecology 69:1599–1608. Jefferies, R. L. 1997. Herbivores, nutrients and tropic cascades in terrestrial environments. Pages 301–330 in H. Olff, V. K. Brown, and R. H. Drent, editors, Herbivores: Between plants and predators. Blackwell Science, Oxford, United Kingdom. Kira, T. 1952. So-called “overgrazed pastures” in ecological context. Bulletin of the Society of Plant Ecology 1:209–213. (In Japanese.) McNaughton, S. J. 1976. Serengeti migratory wildebeest: Facilitation of energy flow by grazing. Science 191:92–94. McNaughton, S. J. 1984. Grazing lawns: Animals in herds, plant form, and coevolution. American Naturalist 124:863–886. McNaughton, S. J., D. G. Milchunas, and D. A. Frank. 1996. How can net primary productivity by measured in grazing ecosystems? Ecology 77:974–977. Miyazaki, A., Y. Kojima, and J. Nishimura. 1977. The digestibility of grass hay by Japanese deer in Nara Park. Pages 143–158 in Report on sika deer in Nara Park for 1976. (In Japanese with English abstract.) Mould, E. D., and C. T. Robbins. 1982. Digestive capabilities in elk compared to white-tailed deer. Journal of Wildlife Management 46:22–29. Newman, J. A., A. J. Parson, and A. Harvey. 1992. Not all sheep prefer clover: diet selection revisited. Journal of Agricultural Science 119:275–283. Numata, M. 1973. Succession. Pages 74–92 in M. Numata, editor, Ecology of grasslands. Tsukijishoten Publishing Company, Tokyo, Japan. (In Japanese.)

10 Productivity and Foraging Efficiency of the Short-Grass

157

Sala, O. E., M. Oesterheld, R. J. Leon, and A. Soriano. 1986. Grazing effects upon plant community structure in subhumid grasslands of Argentina. Vegetatio 67:27–32. Shipley, L. A., and D. E. Spalinger. 1992. Mechanics of browsing in dense food patches: effects of plant and animal morphology on intake rate. Canadian Journal of Zoology 70:1743–1752. Stobbs, T. H. 1973. The effect of plant structure on the intake of tropical pasture. II. Differences in sward structure, nutritive value, and bite size of animals grazing Setaria anceps and Chloris gayana at various stages. Australian Journal of Agricultural Research 24:821–829. Stobbs, T. H. 1975. The effect of plant structure on the intake of tropical pasture. III. Influence of fertilizer nitrogen on the size of bite harvested by Jersey cows grazing Setaria anceps cv. ‘Kanzungula’ swards. Australian Journal of Agricultural Research 26:997–1007. Takahashi, H. 1998. Reproduction and habitat use of sika deer under food limitation. PhD dissertation, Graduate School of Environmental Earth Science, Hokkaido University, Sapporo, Japan. Takatsuki, S. 1976. An ecological study on the interrelationships between sika deer and vegetation. Ph. D. dissertation, Faculty of Science, Tohoku University, Sendai, Japan. (In Japanese.) Takatsuki, S. 1977. Ecological studies about effect of sika deer (Cervus nippon) on vegetation. I. Evaluation of grazing intensity of sika deer on the vegetation in Kinkazan Island, Japan. Ecological Review 18:233–250. Takatsuki, S. 1980. Ecological studies about effect of sika deer (Cervus nippon) on vegetation. II. The vegetation of Akune Island, Kagoshima prefecture, with special reference to grazing and browsing effect of sika deer. Ecological Review 19:123–144. Takatsuki, S. 1983. Habitat selection by sika deer on Kinkazan Island. Journal of the Mammalogical Society of Japan 9:183–191. (In Japanese with English abstract.) Takatsuki, S. 1993. Food habits of ungulates and anti-herbivory adaptation of plants. Pages 104–128 in I. Washitani and T. Ohgush, editors, Relationships between animals and plants. Heibon-sha Publishing Company, Tokyo, Japan. (In Japanese.) Takatsuki, S. 2006. Ecological History of Sika Deer. University of Tokyo press, Tokyo, Japan. (In Japanese.) Takatsuki, S., and T. Gorai. 1994. Effects of sika deer on the regeneration of Fagus crenata on Kinkazan Island, northern Japan. Ecological Research 9:115–120. Takatsuki, S., and Y. Hirabuki. 1998. Effects of sika deer browsing on the structure and regeneration on the Abies firma forest on Kinkazan Island, northern Japan. Journal of Sustainable Forestry 6:203–221. Takatsuki, S., and T. Ito. 1999. Population changes of sika deer on Kinkazan Island. Pages 85–97 in Management of Kinkazan Island. Miyagi Prefecture, Sendai, Japan. (In Japanese.) Takatsuki, S., S. Miura, K. Suzuki, and K. Ito-Sakamoto. 1991. Age structure in mass mortality in the sika deer (Cervus nippon) population on Kinkazan Island, northern Japan. Journal of the Mammalogical Society of Japan 15:91–98. Takatsuki, S., K. Suzuki, and I. Suzuki. 1994. A mass-mortality of sika deer on Kinkazan Island, northern Japan. Ecological Research 9:215–223. Yoshii, Y., and K. Yoshioka. 1949. Plant communities of Kinkazan Island. Ecological Review 12:84–105. (In Japanese.)

Chapter 11

Home Range, Habitat Selection, and Food Habits of the Sika Deer Using the Short-Grass Community on Kinkazan Island, Northern Japan Takehiko Y. Ito and Seiki Takatsuki Abstract The home range size and changes by season were studied for four radiocollared adult female sika deer in the high-density population at the shrine area of Kinkazan Island, northern Japan. Size of home ranges was quite small, the greatest of the four females being about 20 ha, much smaller than for sika deer in other areas, or as predicted by body size relationships to home range size in other species by McNab (1963). The female home ranges overlapped broadly and were centered on the short-grass (lawnlike) Zoysia grasslands, their major feeding area. There was no consistent pattern in home range sizes over the seasons, with some individuals showing increases and others declines. However, Zoysia grassland use was greatest in summer and least in winter, which agreed with the foods consumed, which was 70% Zoysia japonica in summer and 14% in winter.

Introduction Home ranges of large herbivores, such as ungulates, usually include several plant communities (e.g., Clutton-Brock et al. 1982; Schoen and Kirchhoff 1985; Gordon 1989a, b; Thirgood 1995; Tufto et al. 1996), and they select some of these according to requirements for foods (e.g., Albon and Langvatn 1992; Spalinger and Hobbs 1992; Mysterud et al. 1999), cover (e.g., Staines 1976; Conradt et al. 2000), and/or other requirements. In the temperate areas, food supply in their home ranges fluctuates seasonally, which often results in change in home range use. Of course, other factors affect the habitat selection. For example, snow restricts the movement of ungulates, thus resulting in reduction of home range size (e.g., Telfer and Kelsall 1984), or predators sometimes affect home range use (Cederlund and Lindström 1983; Festa-Bianchet 1988; Lima and Dill 1990; Kie 1999). Studies on the home range of sika deer (Cervus nippon) are limited. There are several studies on seasonal migration (Maruyama 1981; Takatsuki et al. 2000; Uno and Kaji 2000; Igota et al. chapter 19), which is affected by snow. On Kinkazan Island this is not the case because there is no snow; the climate is moderated by the Pacific Ocean. Sika deer on Kinkazan Island also include several plant communities in their home ranges, and, therefore, it is expected the selection of plant communities in D. R. McCullough et al. (eds.), Sika Deer: Biology and Management of Native and Introduced Populations, DOI: 10.1007/978-4-431-09429-6_11, © Springer 2009

159

160

T. Y. Ito, S. Takatsuki

their home ranges will change seasonally. This population seems to be suitable to study the effects of food supply on home range size, because (1) Kinkazan Island is located in the cool temperate zone, and therefore the plant communities show marked seasonal changes in food supply for the deer, and (2) since snow is rare and no large predator lives on the island, factors affecting home range use are limited. Consequently, analyses are comparatively simple. We choose sika deer using the Zoysia community, a lawnlike grass community, for the following reasons. It is quite productive during the growing period and attracts a high density of deer. However, its value as a foraging site decreases abruptly in the nongrowing period when it has low biomass (Ito et al. chapter 10). It is expected therefore that the effects of the Zoysia community on home range size and habitat selection would be more pronounced than for other plant communities that show less marked seasonal changes in food supply. Takatsuki (1980) studied the food habits of the sika deer in the same study area in 1976 and 1977 and showed that Zoysia japonica was one of the main foods for the deer. The Zoysia community was a small patch at that time, but has subsequently enlarged since the 1980s. In contrast, other dominant plant species such as Pleioblastus chino (dwarf bamboo) and Miscanthus sinensis (silver grass) decreased drastically. These facts should change the deer’s food habits and affect their habitat choices. The objectives of the present study were (1) to determine the seasonal change in home range and habitat selection of sika deer on Kinkazan Island, and (2) to compare the food habits of the sika deer from the same place in the late 1970s to those in 1995–96, in order to show effects of the seasonal change of food supply in the Zoysia community on their habitat selection following vegetative change.

Materials and Methods Home Range and Habitat Selection Four adult female sika deer were captured in the Zoysia japonica community with an immobilizing gun and were fitted with radio collars (transmitter: 144 MHz, ATS Co. Ltd., Minnesota, USA, 230 g, receiver: FT-290 mkII, Yaesu Radio Co Ltd., Japan, antenna: hand-held 3-element Yagi). We selected adult females since they are less influenced by rutting behavior and dispersal as compared to males. Table 11.1 shows the age, date of capture, and observation periods for the radiocollared deer. Deer ages were estimated by wear of incisors (Ohtaishi 1980; Takatsuki 1998). Observations were carried out from March 1995 to August 1998. We tracked only deer No. 1 throughout the entire period. We began tracking No. 2 later, and Nos. 3 and 4 died in the middle of the study period. The observation period was

11 Home Range, Habitat Selection, and Food Habits

161

Table 11.1 Information on tracked female sika deer. Deer| number

Capture day

Age (years)*1

1

20 Nov. 1994

2*2

24 June. 1995

–

–

3

22 Nov. 1994

10

52

4

22 Nov. 1994

10

46

9

Weight (kg)

Observation period

Total home range size (ha)

52

Winter 1995–summer 1998 Summer 1995–summer 1998 Winter 1995–winter 1997 Winter 1995–fall 1995

15.6 20.5 16.1 5.7

1

* Ages were estimated according to teeth wear. *2Age and weight of deer No. 2 were not determined, though she looked similar to other individuals in size.

divided into three seasonal intervals: (1) first interval: 31 March–6 April 1995 (winter), 9–15 June 1995 (spring), 6–12 August 1995 (summer), 22–27 October 1995 (fall); (2) second interval: 3–9 September 1996 (summer), 10–15 November 1996 (fall), 9–14 February 1997 (winter), 2–12 June 1997 (summer); and (3) third interval: 15–25 November 1997 (fall), 28 February–7 March and 11 April 1998 (winter), 16–26 May 1998 (spring), 9–13 August 1998 (summer). In the first and second seasonal intervals, we recorded the deer locations approximately each hour from 0800 to 1800 for about six days in a time interval. On each occasion, the deer were located by telemetry, and thereafter the locations were confirmed by direct observations. This was possible because the deer were relatively tame and approachable to within a short distance without disturbing their behavior. We expressed all home range areas by the minimum convex polygon method (Mohr 1947). If home ranges included an exclosure fence (Figs. 11.1 through 11.4), the area of the enclosure was subtracted from the home range size.

Food Habits Ten fecal pellets group were collected on the Zoysia community in the study at monthly intervals from May to November 1995, and at two-month intervals from January to March 1996. Fecal samples were washed over a 0.5 mm mesh and retained fragments were analyzed using a microscope (Stewart 1967). Plant fragments were spread over a glass slide with 1 mm grids, and crossing points were scored for four categories: leaves of Z. japonica, leaves of other graminoids, other plant parts, and unknown, until the total count reached 200 points. These results were compared with the results of food habits obtained in 1976 and 1977, in which fecal samples were collected in the same area and analyzed by the same method (Takatsuki 1980).

162

T. Y. Ito, S. Takatsuki

Spring

Summer

? ? ?

?

?

? ?

? ?

? ?

?

F

F

?

?

?

? ?

0

100m

Fall

Winter

? ?

?

? ?

? ?

?

?

?

?

?

F

F ?

Home range

Plant community

first term (winter 1995–fall 1995)

Zoysia japonica community

second term (summer 1996–spring 1997)

Miscanthus sinensis community

third term (fall 1997–summer 1998)

shrub community,

evergreen forest

deciduous forest,

marsh,

pebbly coast

Fig. 11.1 Seasonal changes in the home range of the deer No.1. F shows the location of an exclosure fence.

Results Each deer used a particular part of the study area but overlapped extensively with other radioed deer. Deer No. 1 used the northern and northeastern part (Fig. 11.1), No. 2 and No. 3 used the eastern and the southern part (Figs. 11.2 and 11.3), and No.4 used the northwestern part (Fig. 11.4). Home ranges for the three seasonal intervals were stable for all four individuals (Figs. 11.1–11.4).

11 Home Range, Habitat Selection, and Food Habits

163

Summer

Spring

?

F

F

? ?

?

? ?

?

? ? ?

?

?

?

?

?

?

0

?

100m

Winter

Fall

F

F

?

?

?

?

?

? ? ?

?

?

?

Home range

Plant community

first term (winter 1995–fall 1995)

Zoysia japonica community

second term (summer 1996–spring 1997)

Miscanthus sinensis community

third term (fall 1997–summer 1998)

shrub community,

evergreen forest

deciduous forest,

marsh,

pebbly coast

Fig. 11.2 Seasonal changes in the home range of the deer No.2. F shows the location of an exclosure fence.

Changes in home range size showed different seasonal patterns among individuals (Fig. 11.5). Summer ranges of deer Nos. 1, 2, and 4 were larger than winter ranges, while that of No. 3 was smaller than the winter range. Number 1 had the largest range in spring. Number 2 had larger ranges in spring and fall than in summer and winter. Number 3 had the smallest range in summer, and had the largest range in fall and winter in the first and the second intervals, respectively. Number 4 had the

164

T. Y. Ito, S. Takatsuki

Spring

Summer

F

F ?

?

? ?

0

100m

Fall

Winter

F

F

?

?

Home range

Plant community

first term (winter 1995–fall 1995)

Zoysia japonica community

second term (summer 1996–spring 1997)

Miscanthus sinensis community shrub community,

evergreen forest

deciduous forest,

marsh,

pebbly coast

Fig. 11.3 Seasonal changes in the home range of the deer No.3. F shows the location of an exclosure fence.

largest range in summer. Thus, there was no consistent pattern of seasonal change in home range size among the four deer studied. However, changes in the proportion of the Zoysia community in the home ranges showed a consistent seasonal pattern (Fig. 11.6). The proportions of the Zoysia community changed seasonally in every interval for all individuals (χ2 test: p < 0.0001): they were largest in summer, and smallest in winter (Fisher’s exact probability test with Bonferroni test: p < 0.0001) except No. 4 (p = 0.19).

11 Home Range, Habitat Selection, and Food Habits

Spring

Summer

F

0

165

F

100m

Fall

Winter

F

Home range first term (winter 1995–fall 1995)

F

Plant community Zoysia japonica community Miscanthus sinensis community shrub community, deciduous forest,

evergreen forest marsh, pebbly coast

Fig. 11.4 Seasonal changes in the home range of the deer No.4. F shows the location of an exclosure fence.

Zoysia japonica was the main foods of the sika deer and comprised about 70% of the plant fragments in summer, but decreased to 14% in winter (Fig. 11.7). However, comparison of food habits over time showed a marked change in the 20 years from Takatsuki’s (1980) work in the 1970s to this study. In 1976 and 1977, Z. japonica was also one of main foods of the deer, but it occupied only 20–30% in summer and less than 10% in winter (Takatsuki 1980). Two other Gramineae

166

T. Y. Ito, S. Takatsuki

Area (ha)

12

12

No.1

10

10

8

8

6

6

4

4

2

2

0

0 Spring

Area (ha)

12

No.2

Summer

Fall

Winter

Spring Summer 12

No.3

10

10

8

8

6

6

4

4

2

2

Fall

Winter

No.4

0

0 Spring

Summer

Fall

Winter

Spring Summer

Fall

Winter

Fig. 11.5 Seasonal changes in home range size of the four individual deer. White columns: first interval (winter 1995–fall 1995); black columns: second interval (summer 1996–spring 1997); gray columns: third interval (fall 1997–summer 1998).

species, Miscanthus sinensis and Pleioblastus chino, were other main food plants in the 1970s (Takatsuki 1980), but they had decreased greatly by 1995 and 1996.

Discussion The sika deer living in northern Japan migrate seasonally: i.e., Nikko (Maruyama 1981), Mt. Goyo (Takatsuki et al. 2000), and Hokkaido (Uno and Kaji 2000). In contrast, however, the deer on Kinkazan Island did not change their home ranges seasonally (Figs. 11.1 through 11.4). Because there are neither predators nor snow on Kinkazan Island, home range use of the deer is mainly affected by food supply. It is known that home range size is positively related to body size (McNab 1963). The correlation equation between body weight and home range size presented by McNab (1963) predicts that the home range of sika deer on Kinkazan Island would be 95 ha. This is five times larger than the observed size. This is probably because it is not necessary for the deer on Kinkazan Island to migrate because little snow falls and food is not covered by snow in winter. Besides, there may be no space to move during the period of food shortage because of the high deer density.

11 Home Range, Habitat Selection, and Food Habits

Relative area (%)

100

No.1

100

80

80

60

60

40

40

20

20

0

1 2 3 Spring

1 2 3 Summer

1 2 3 Fall

1 2 3 Winter

0

No.3 100

80

80

60

60

40

40

20

20

Relative area (%)

100

167 No.2

1 2 3 Spring

1 2 3 Summer

1 2 3 Fall

1 2 3 Winter

1 2 3 1 2 3 Spring Summer

1 2 3 Fall

1 2 3 Winter

No.4

0

0 1 2 3 Spring

1 2 3 Summer

1 2 3 Fall

1 2 3 Winter

Zoysia japonica community, shrub community,

evergreen forest, deciduous forest, Miscanthus sinensis community

Fig. 11.6 Relative areas of plant communities in seasonal home ranges of the four individuals. Numbers above seasons indicate the study period: (1) first interval (winter 1995 – fall 1995), (2) second interval (summer 1996 – spring 1997), (3) third interval (fall 1997 – summer 1998).

It is noteworthy that the home range of the deer was small, the largest being about 20 ha (Table 11.1). The few studies that have been done on annual home range sizes of nonmigratory sika deer populations did not show home ranges this small. For example, the home range of females on Nakanosima Island in Hokkaido ranged from 48 to 86 ha (Takahashi 1998), and that of adult females in Boso Peninsula, Chiba Prefecture ranged from 46 to 246 ha (Shigematsu et al. 1994). The only exception was females on Nozaki Island, Nagasaki Prefecture whose ranges were only 5–11 ha (Endo and Doi 1996). Since all the populations on Nozaki Island (Endo and Doi 1996), on the Boso Peninsula (Shigematsu et al. 1994), and on Kinkazan Island used the Zoysia community, it is probable that high productivity of the Zoysia community (Ito et al. chapter 10) would result in the small home range of the deer. The Zoysia community is open and highly visible from a distance. Some cervids prefer plant communities with a closed canopy to those with abundant food availability (e.g., roe deer, Mysterud et al. 1999). This is interpreted as an antipredator strategy.

168

T. Y. Ito, S. Takatsuki

Zoysia japonica other graminoids other plants unknown

Fig. 11.7 Seasonal change of fecal composition of the sika deer in 1995–96.

There are many reports that predators and hunting influence the habitat use of cervids (predators: Festa-Bianchet 1988; Lima and Dill 1990; Kie 1999; hunting: Vercauteren and Hygnstrom 1998; Kilpatrick and Lima 1999). The fact that the deer spent most of the time on the Zoysia community although other communities were available in their home ranges probably is attributable to the absence of predators on Kinkazan Island. Although Takatsuki (1983) predicted that the seasonal home range size of sika deer using the Zoysia community would be larger in winter than in summer, no consistent pattern was found in the seasonal home range size among individuals (Fig. 11.5). This is difficult to explain by food supply, but it is important that at least some deer left this area in winter and the deer density decreased (Ito et al. chapter 10). There was a consistent pattern that the proportions of Zoysia community in the summer home ranges were greater than other seasons for all the deer, and this correlated with food habits of the deer (Fig. 11.6). This suggests that seasonal change of the habitat selection of the deer in the study area is strongly affected by the characteristics of the Zoysia community; the productivity of the Zoysia community is high in summer, but low in winter (Ito et al. chapter 10). The Zoysia community on Kinkazan Island has been enlarged in the last 20 years by the continuous heavy grazing of the sika deer. A patch of this community occupied only 100 m2 in the 1970s, but it has now largely replaced the Miscanthus/ Pleioblastus community. The increase of Zoysia japonica in the deer diet well-represents the vegetational change in this area. This means deer dependency on the Zoysia community has intensified during the last two decades. Since the habitat selection by the deer in the study area seems to be strongly affected by food supply in the Zoysia community, this enlargement of the area of the Zoysia community would change deer habitat use over the last two decades. Since it appears that the Zoysia community is still expanding on Kinkazan Island, home range use together with increased carrying capacity for deer also would be affected.

11 Home Range, Habitat Selection, and Food Habits

169

Literature Cited Albon, S. D., and R. Langvatn. 1992. Plant phenology and the benefits of migration in a temperate ungulate. Oikos 65:502–513. Clutton-Brock, T. H., F. E. Guiness, and S. D. Albon. 1982. Red deer: Behavior and ecology of two sexes. The University of Chicago Press, Chicago, Illinois, USA. Conradt, L., T. H. Clutton-Brock, and F. E. Guiness. 2000. Sex differences in weather sensitivity can cause habitat segregation: Red deer as an example. Animal Behaviour 59:1049–1060. Cederlund, G., and E. Lindström. 1983. Effects of severe winter and fox predation on roe deer mortality. Acta Theriologica 287:129–145. Endo, A., and T. Doi. 1996. Home range of female sika deer Cervus nippon on Nozaki Island, the Goto Archipelago, Japan. Mammal Study 21:27–35. Festa-Bianchet, M. 1988. Seasonal range selection in bighorn sheep: conflicts between forage quality, and forage quantity, and predator avoidance. Oecologia 75:580–586. Gordon, I. J. 1989a. Vegetation community selection by ungulates on the Isle of Rhum. II. Vegetation community selection. Journal of Applied Ecology 26:53–64. Gordon, I. J. 1989b. Vegetation community selection by ungulates on the Isle of Rhum. III. Determinants of vegetation community selection. Journal of Applied Ecology 26:65–79. Kie, J. G. 1999. Optimal foraging and risk of predation: Effects on behavior and social structure in ungulates. Journal of Mammalogy 80:1114–1129. Kilpatrick, H. J., and K. K. Lima. 1999. Effects of archery hunting on movement and activity of female white-tailed deer in an urban landscape. Wildlife Society Bulletin 27:433–440. Lima, S. L., and L. M. Dill. 1990. Behavioral decisions made under the risk of predation: A review and prospects. Canadian Journal of Zoology 68:619–640. Maruyama, N. 1981. A study of the seasonal movements and aggregation patterns of sika deer. Bulletin, Faculty of Agriculture, Tokyo University of Agriculture and Technology 23:1–85. (In Japanese with English abstract.) McNab, B. K. 1963. Bioenergetics and the determination of home range size. American Naturalist 97:133–140. Mohr, C. O. 1947. Table of equivalent populations of North American small mammals. American Midland Naturalist 37:223–249. Mysterud, A., P. K. Larsen, R. A. Ims, and E. Østbye. 1999. Habitat selection by roe deer and sheep: Does habitat ranking reflect resource availability? Canadian Journal of Zoology 77:776–783. Ohtaishi, N. 1980. Determination of sex, age and death-season of recovered remains of sika deer (Cervus nippon) by jaw and tooth-cement. Archaeology and Natural Science 13:51–74. (In Japanese with English abstract.) Schoen, J. W., and M. D. Kirchhoff. 1985. Seasonal distribution and home-range patterns of Sitka black-tailed deer Odocoileus hemionus sitkensis on Admiralty Island, southeast Alaska, USA. Journal of Wildlife Management 49:96–103. Shigematsu, Y., K. Ochiai, and M. Asada. 1994. Animal tracking by radio-telemetry. Pages 27–32 in Reports of management for sika deer of Boso Peninsula in Chiba Prefecture, II. Chiba Prefecture, Japan. (In Japanese.) Spalinger, D. E., and N. T. Hobbs. 1992. Mechanisms of foraging in mammalian herbivores: New models of functional response. American Naturalist 140:325–348. Staines, B. W. 1976. The use of natural shelter by red deer (Cervus elaphus) in relation to weather in North-east Scotland. Journal of Zoology 180:1–8. Stewart, D. W. M. 1967. Analysis of plant epidermis in faeces: A technique for studying the food preferences of grazing herbivores. Journal of Applied Ecology 4:83–111. Takahashi, H. 1998. Reproduction and habitat use of sika deer under food limitation. Ph.D. thesis, Graduate School of Environmental Earth Science, Hokkaido University, Sapporo, Japan. Takatsuki, S. 1980. Food habits of sika deer on Kinkazan Island. Science Reports Tohoku University, Series IV (Biology) 38:7–31.

170

T. Y. Ito, S. Takatsuki

Takatsuki, S. 1983. Habitat selection by sika deer on Kinkazan Island. Journal of the Mammalogical Society of Japan 9:183–191. Takatsuki, S. 1998. A life of sika deer read from a tooth. Iwanami-shoten Publishing, Tokyo, Japan. (In Japanese.) Takatsuki, S., K. Suzuki, and H. Higashi. 2000. Seasonal elevational movements of sika deer on Mt. Goyo, northern Japan. Mammal Study 25:107–114. Telfer, E. S., and J. P. Kelsall. 1984. Adaptation of some large North American mammals for survival in snow. Ecology 65:1828–1834. Thirgood, S. J. 1995. The effects of sex, season and habitat availability on patterns of habitat use by fallow deer (Dama dama). Journal of Zoology 235:645–659. Tufto, J., R. Andersen, and J. Linnell. 1996. Habitat use and ecological correlates of home range size in a small cervid: The roe deer. Journal of Animal Ecology 65:715–724. Uno, H., and K. Kaji. 2000. Seasonal movements of female sika deer in eastern Hokkaido, Japan. Mammal Study 25:4 9–57. Vercauteren, K. C., and S. E. Hygnstrom. 1998. Effects of agricultural activities and hunting on home ranges of female white-tailed deer. Journal of Wildlife Management 62:280–285.

Chapter 12

Shift to Litterfall as Year-Round Forage for Sika Deer after a Population Crash Masami Miyaki and Koichi Kaji

Abstract We determined that litterfall was utilized as the main food throughout the year after the first crash of a sika deer population in 1984 on Nakanoshima Island, Toya Lake, Hokkaido, Japan. Palatable plants and leaves of broad-leaved trees decreased to 0.1% of biomass within the browse category in 1994. Meanwhile, unpalatable plants such as senecio (Senecio cannabifolius), chloranthus (Chloranthus japonicus), and spurge (Pachysandra terminalis) occupied 99.9% of biomass on the forest floor. Species richness of the forest floor plants declined greatly. Litterfall is a stable and abundant resource although its consumption requires more time than browsing on leaves of trees if they are available. We propose that a shift to litterfall as alternative food will continue to support a relatively high density sika deer population. However, the high deer population will come at the cost of few palatable plants and tree seedlings within the browsing reach of deer, dominance of unpalatable plants, and a decline of plant species richness for a long time in the post-peak stage.

Introduction Litterfall (fallen leaves) has been reported as important deer food in many studies (Prisyazhnjuk and Prisyazhnjuk 1974; Crawford 1982; Nugent 1988; Picard et al. 1991; Gray and Servello 1995; Mitani 1995; Makino 1996; Ditchkoff and Servello 1998; Takahashi and Kaji 2001; Miyaki and Kaji 2004; Tremblay et al. 2005; Ichinohe et al. 2006; Lefort et al. 2007). For example, Prisyazhnjuk and Prisyazhnjuk (1974) showed that shrub leaves in deciduous forests composed 40% of the weight of rumen contents in winter for a sika deer population on Askold Island, Russia. For a tame white-tailed deer (Odocoileus virginianus), dried leaves of hardwoods were important foods for late autumn and winter (Crawford 1982). The consumption of litterfall by white-tailed deer in unharvested stands was high relative to the availability of browse in winter (Ditchkoff and Servello 1998). Many of these studies emphasized the importance of litterfall as a autumn and/or winter forage. However, studies emphasizing the importance of litterfall even in summer for high-density populations D. R. McCullough et al. (eds.), Sika Deer: Biology and Management of Native and Introduced Populations, DOI: 10.1007/978-4-431-09429-6_12, © Springer 2009

171

172

M. Miyaki, K. Kaji

under food-limited conditions are more limited (Makino 1996; Takahashi and Kaji 2001; Miyaki and Kaji 2004). Tremblay et al. (2005) stated that the abundance of litterfall from mature trees is independent of heavy browsing over a long time, which contributes to sustaining high deer populations in balsam fir stands on Anticosti Island, Quebec, Canada. In New Zealand’s native forest, the ruminants are able to consume virtually all of the foliage of preferred native plants, particularly where they can supplement their diet with fallen leaves from the forest canopy (Nugent et al. 2001). Existence of an alternative food after a population crash, such as litterfall, has a great influence on deer population dynamics and plant species richness in the habitat. On Nakanoshima Island, Toya Lake, Hokkaido, Japan, the main diet for deer shifted from leaves and twigs on the trees and herbaceous plants (Kaji et al. 1988; Kaji and Yajima 1992) to litterfall even in summer (Takahashi and Kaji 2001; Miyaki and Kaji 2004) after the population crash in 1984. At the same time, there was a drastic change in the forest floor vegetation (Sukeno and Miyaki 2007). In this chapter, we summarize the changes of flora and the amount of biomass of summer forage and its contribution to deer forage based on Miyaki and Kaji (2004) and Sukeno and Miyaki (2007). Further, we discuss the relationship among continuing high sika deer population density after the population crash, litterfall resources, few palatable plants within deer reach, and dominance of unpalatable plants. We divided time into pre-peak and post-peak stages in 1984 when the first population crash was observed.

Study Area We conducted our study on Nakanoshima Island in Toya Lake located in southwest Hokkaido, Japan (140°51′ E, 42°36′ N). The island is 497.8 ha in size of which 457.1 ha (91.8%) is deciduous broad-leaved forest, 31.5 ha (6.3%) is coniferous plantation, 7.8 ha (1.6%) is grassland (located in the central part of the island), and 1.4 ha (0.3%) is builtup area. The sika deer are distributed throughout the whole island except the built-up area. The highest temperature was 24.7 °C in August, the lowest temperature was -5.6 °C in February, and the mean yearly precipitation was 1,112 mm from 1988 to 1993 at the Toya Limnological Station for Environmental Biology, Hokkaido University, 4 km west of the island. The maximum snow depth was approximately 50 cm. The major canopy species in the deciduous forest were oak (Quercus crispla), castor-aralia (Kalopanax pictus), magnolia (Magnolia obovata), maple (Acer mono var. glabrum) maackia (Maackia amurensis var. buergeri), linden (Tilia japonica), and hop hornbeam (Ostrya japonica). The forest floor was dominated by unpalatable species for sika deer such as senecio (Senecio cannabifolius), chloranthus (Chloranthus japonicus and Chloranthus serratus), spurge (Pachysandra terminalis), daphne (Daphne kamtschatica), arisaema (Arisaema peninsulae), and peony (Paeonia japonica) (Takatsuki 1989; Kaji et al. 1991). Plum-yew (Cephalotaxus harringtonia) increased temporarily after the population crash and then almost disappeared from deer habitat. Bark-stripping was not observed on large trees except elm (Ulmus japonica) and yew (Taxus cuspidate),

12 Shift to Litterfall as Year-Round Forage for Sika Deer

173

whereas young deciduous trees had been almost eliminated by bark-stripping (Kaji et al. 1991). The short-grass community in the grassland, which was used as a feeding site for the deer in summer, covered 2.3 ha. The major short-grass community species were bluegrass (Poa pratensis), timothy grass (Phleum alpinum), pearl wort (Sagina japonica), sneezeweed (Centipeda minima), dandelion (Taraxacum officinale), and wood sorrel (Oxalis corniculata), all of which were about 3 cm in height. The sika deer population had reached a peak of 299 deer (58 deer/km2) in 1983. After the population crash of 1984, due to mass mortality owing to food shortage, deep snow, and artificial removal, it has maintained a high density of 26–87 deer/ km2 through the second crash in 2004 (Kaji et al. 1988; Hokkaido Institute of Environmental Sciences 2006). Winter foods such as dwarf bamboos (Sasa senanensis and S. kurilensis) and deciduous trees were markedly reduced by 1984 (Kaji et al. 1988; Kaji and Yajima 1992).

Methods Vascular Flora and Plant Species Richness Vascular flora was surveyed in five fenced exclosures established in 1984, five unfenced controls, and the whole area of the island between 2002 and 2004 (Sukeno and Miyaki 2007). One plot of 4 m2, which was divided into four small plots of 1 m2, was set in each 200 m2 exclosure and 200 m2 control. We used the number of species identified in 1977 (Ozaki 1997) for the data in pre-peak stage on the whole area. We evaluated the plant species richness from species-area relations between pre-peak stage (before 1984) and post-peak stage (after 1984) under the assumption that the flora in exclosures had recovered to the pre-peak stage.

Biomass and Consumption of Deer Foods in Summer We set movable cages in the short-grass community in July 1994. We trimmed plants down to the same height as those outside of the cages. Samples were dried at 60 °C for 48 h and weighed. We measured the biomass of deciduous leaves on branches and forest floor plants in July 1994, except conifer leaves which were not utilized by deer. As the height of the browsing line, the reach of the deer, was 212 ± 10 cm (mean ± SD, n = 66) on the island, we collected samples of deciduous leaves up to 220 cm in height. In order to estimate consumption of litterfall by deer and other herbivores, we set litter-traps at 1.5 m height and on the ground respectively in two stands of deciduous forest. Samples were collected every month from June to November 1994. Furthermore, to estimate consumption of litterfall by herbivores other than

174

M. Miyaki, K. Kaji

deer, we put mesh bags containing 10–15 g of leaves in litter-traps and on the ground and collected them after one month in July. We estimated consumption of litterfall by deer L to be: L = Ta{(1-To/ Ta) – (1-Bo / Ba)} = Ta(Bo / Ba) – To

(1)

where Ta is litterfall in litter-traps above the ground, To is litterfall on the ground where deer could access, and Ba and Bo are the residual rates of leaves in mesh bags above the ground and on the ground, respectively. In Equation (1) (1-To/Ta) shows the total consumption rate by herbivores and (1-Bo/Ba) shows the consumption rate by herbivores other than deer. Methods and results are reported in detail in Sukeno and Miyaki (2007) for the change of vascular flora and in Miyaki and Kaji (2004) for biomass and deer consumption.

Results Vascular Flora and Plant Species Richness During 2002–2004, 246 species of 83 families were listed in the islands (Sukeno and Miyaki 2007). Plant species had declined to 32.4% of the list surveyed in 1977 (Ozaki 1997). Between the surveys, the reduction rate in herbaceous species (18.7%, n = 359) and shrub species (35.0%, n = 40) were higher than in tree species (62.5%, n = 64). Dominant species were senecio, pearl wort, spurge and chloranthus (Chloranthus serratus) in the unfenced areas, and plum-yew, ostrich fern (Zatteuccia struthiopteris), shisandra (Shisandra chinensis) and common horsetail, (Equisetum hyemale) in the fenced areas. We compared species-area relations for vascular plants between fenced area (exclosures) and unfenced area (controls) (Fig. 12.1). We used the number of species identified in 1977 (Ozaki 1997) instead of the exclosure data on the whole area in Fig. 12.1. The average number of plant species in 1 m2-plots was 8.35 in exclosures and 2.35 in controls. The slopes and the coefficients of log transformed regression were 0.24 and 8.57 in exclosures and 0.26 and 3.07 in controls, respectively. The coefficients were significantly different (p < 0.01), while slopes were not significantly different.

Biomass and Consumption of Deer Foods in Summer In deciduous forests, biomass of unpalatable species occupied 99.9% (1,018.3 kg/ ha) of the total biomass within the browse layer in July of 1994 (Fig. 12.2). Palatable understory plants at height <20 cm (0.872 kg/ha) and deciduous leaves

12 Shift to Litterfall as Year-Round Forage for Sika Deer

175

Fig. 12.1 Species-area relation for vascular plants between exclosures and controls in 2004. We set the number of species identified in 1977 instead of the exclosure data on the whole area (the outline circle of right side) in the whole area (the open circle in the upper right of the figure) as the maximum exclosure value.

Fig. 12.2 Biomass of forest floor plants and deciduous leaves in deciduous forests on Nakanoshima Island, Toya Lake in July, 1994.

at height ≥40 cm (0.225 kg/ha) had extremely low density and only 0.1% of the total biomass. Difference between litterfall in litter-traps (28.7 kg/ha/month) and that above the ground (4.5 kg/ha/month) means the total consumption of deer and other herbivores on the ground, was significant (p < 0.01) in July 1994 (Fig. 12.3). Consumption rate of deer was estimated at 18.4% of total litterfall.

176

M. Miyaki, K. Kaji

Fig. 12.3 Seasonal change of litterfall above the ground and on the ground in deciduous forests on Nakanoshima Island in 1994. Capped vertical lines represent 1 SE. ** significant at 1% level, *significant at 5% level, and ns not significant.

Fig. 12.4 Biomass and consumption of deer food on Nakanoshima Island in July, 1994. (a) productivity per unit area, (b) total biomass and consumption on the island.

Productivity of summer forage per unit area was the highest in the grassland, 228 ± 55 kg/ha/month (mean ± SE; Fig. 12.4a). Litterfall had a second productivity, 28.7 ± 5.3 kg/ha/month during this period. Deciduous leaves and palatable plants within the browse layer produced 1.08 kg/ha. The total biomass of litterfall in July was 12,499 ± 2,308 kg or 92.4% of the deer food supply on the island (Fig. 12.4b). We estimated consumption of litterfall by

12 Shift to Litterfall as Year-Round Forage for Sika Deer

177

deer at 2,300 ± 425 kg/month or 75.6% of the total consumption. Other consumption of deer food was 17.2% short grasses, 4.1% deciduous forest understory, 3.0% deciduous leaves within the browse layer, and 0.1% conifer plantation understory.

Discussion Change of Flora and Species Richness between Pre-peak Stage and Post-peak Stage We assume that the flora in the exclosures had mostly recovered to that in the prepeak stage of sika deer population because the exclosures were set immediately after the population crash in 1984. We estimated that the number of plant species decreased to about 30% of the beginning flora. Furthermore, more than 20 years of deer browsing after the population crash has impacted especially herbaceous and shrub plants. The dominant species now are restricted to unpalatable plants for deer. They occupy most of the biomass within the browse layer. A large number of studies clearly show that selective ungulate herbivory leads to the dominance of unpalatable, chemically defended plant species in communities (e.g., Augustine and McNaughton 1998). In old-growth hemlock-northern hardwood forest in Pennsylvania, USA, 33 species (80%) were missing from the hemlock-beech stand, and 16 species (59%) were missing from the hemlock stand during 66 years between 1929 and 1995 when deer density maintained a level two to four times higher than presettlement densities after the decline around 1900 (Rooney and Dress 1997). Exclosure studies indicate that direction of change is readily reversible, but the continued impacts of deer on preferred species even at low density indicate that their regulatory influence on forest composition is strong (Nugent et al. 2001). High density of sika deer populations for long time periods also causes a drastic decline of the species richness of herbaceous and shrub plants.

Change of Deer Food and Its Effects for Habitat and Deer Population Makino (1996) reported that fallen leaves were highly utilized by sika deer in winter in the Tanzawa Mountains in Honshu Island, where there is no dwarf bamboo. Litterfall is present not only in autumn and winter, but also in summer, though it is less in summer (Kikuzawa 1983). In our study, we estimated that litterfall contributed about three-fourths of the total consumption of sika deer in summer. The contribution of deciduous leaves to the summer rumen contents of sika deer on the island was as great as in winter (Takahashi and Kaji 2001), whereas browsing from twigs in summer was rare. Thus, litterfall was the major food on the island, even in summer.

178

M. Miyaki, K. Kaji

After the population crash of 1984, body size and body weight of sika deer decreased, and the age of first giving birth delayed (Kaji et al. 1988, Kaji et al. chapter 30). Further, dystocia (Takahashi et al. 2005) and the drop of survival rate of fawns (Kaji and Takahashi 2006) were observed. In this post-peak stage, however, the deer population maintained a high density of 26–87 deer/km2 (Hokkaido Institute of Environmental Sciences 2006). In summer, the nutritional value of litterfall may be similar to leaves on the tree because much of litterfall in summer is still green. On Nakanoshima Island, the number of deer in the grassland decreased remarkably after strong winds in summer because of rich litterfall in the forest (H. Takahashi 2005). As litterfall is scattered on the ground, feeding cost of litterfall might be higher than grazing and browsing when grass and leaves are abundant. Litterfall is an important food resource to compensate for the high sika deer density in the post-peak population. It is supplied all year round in deciduous forests. Furthermore, total litterfall biomass was stable among years (Tremblay et al. 2005). Seedlings and young trees disappear or are severely decreased under high density deer populations (Gill 1992; Nugent et al. 2001; Russell et al. 2001; Dumont et al. 2005; Tremblay et al. 2005). However, life span of large trees is so long that the change of the canopy forest is slow in the post-peak stage. Litterfall is provided for as long as the mature forest exists. Following the population peak, when deer had depleted the plants growing within the browse layer, the continued availability of food produced in the canopy is likely to have decreased the size of the post-peak population crash (Nugent and Challies 1988). Tremblay et al. (2005) stated that the shift to feeding on litterfall introduces a temporal uncoupling between the impact of deer browsing on seedlings and the negative feedback from recruitment failure of mature balsam fir trees on the deer population. In deciduous broad-leaved forest, therefore, a relatively high density of sika deer population after the initial population crash, dependence on litterfall, few palatable plants, dominant unpalatable plants and poor plant species richness might be maintained for a long time unless some irregular incident occurs, such as deep snow or a serious disease outbreak. Acknowledgements This study was partly supported by Hokkaido Government and

the Grant-in-Aid from the Ministry of Education, Science and Culture of Japanese Government (No. 19380084).

Literature Cited Augustine, D. J., and S. J. McNaughton. 1998. Ungulate effects on the functional species composition of plant communities: Herbivore selectivity and plant tolerance. Journal of Wildlife Management 62:1165–1183. Crawford, H. S. 1982. Seasonal food selection and digestibility by tame white-tailed deer in central Maine. Journal of Wildlife Management 46:974–982. Ditchkoff, S. S., and F. A. Servello. 1998. Litterfall: An overlooked food source for wintering white-tailed deer. Journal of Wildlife Management 62:250–255.

12 Shift to Litterfall as Year-Round Forage for Sika Deer

179

Dumont B., P. C. Renaud, N. Morellet, C. Mallet, F. Anglard, and H. Verhyden-Tixier. 2005. Seasonal variations of red deer selectivity on a mixed forest edge. Animal Research 54: 369–381. Gill, R. M. A. 1992. A review of damage by mammals in north temperate forests: 1. Deer. Forestry 65:145–169. Gray, P. B., and F. A. Servello. 1995. Energy intake relationships for white-tailed deer on winter browse diet. Journal of Wildlife Management 59:147–152. Hokkaido Institute of Environmental Sciences. 2006. Results of a survey related to sika deer on Hokkaido (1997–2004). Hokkaido Institute of Environmental Sciences, Sapporo, Japan. (In Japanese.) Ichinohe, T., E. Hosoi, and T. Fujihara. 2006. Seasonal differences in food habits and rumen digestion status of wild sika deer (Cervus nippon) in western Yamaguchi. Japanese Journal of Grassland Science 52:190–197. Kaji, K., and H. Takahashi. 2006. Where do density dependent effects appear? – Relationship between body weight and pregnancy rate for sika deer on Nakanoshima Island. Pages 43–48 in K. Kaji, M. Miyaki, and Y. Uno, editors, Conservation and management of sika deer in Hokkaido. Hokkaido University Press, Sapporo, Japan. (In Japanese.) Kaji, K., and T. Yajima. 1992. Influence of sika deer on forests of Nakanoshima Island, Hokkaido. Pages 215–218 in B. Bobek, K. Perzanowski, and W. Regelin, editors, Transactions 18th Congress of International Union of Game Biologists, Krakow 1987. Swiat Press, KrakowWarszawa, Poland. Kaji, K., K. T. Koizumi, and N. Otaishi. 1988. Effects of resource limitation on the physical and reproductive condition of sika deer on Nakanoshima Island, Hokkaido. Acta Theriologica 33:187–208. Kaji, K., T. Yajima, and T. Igarashi. 1991. Forage selection by deer introduced on Nakanoshima Island and its effect on the forest vegetation. Pages 52–55 in N. Maruyama, B. Bobek, Y. Ono, W. Regelin, L. Bartos, and P. R. Ratcliffe, editors, Proceedings of the International Symposium on Wildlife Conservation, International Congress of Ecology 1990. Tsukuba and Yokohama, Japan. Kikuzawa, K. 1983. Leaf survival of woody plants in deciduous broad-leaved forests. 1. Tall trees. Canadian Journal of Botany 61:2133–2139. Lefort, S., J.-P. Tremblay, F. Fournier, F. Potvin, and J. Huot. 2007. Importance of balsam fir as winter forage for white-tailed deer at the northeastern limit of their distribution range. Ecoscience 14:109–116. Makino, S. 1996. Food habits and habitat selection of tame sika deer Cervus nippon in upper zone in the Tanzawa Mountains. Graduate thesis, Tokyo University of Agriculture and Technology, Tokyo, Japan. (In Japanese.) Mitani, N. 1995. Foraging behavior of tame sika deer on Mt. Tonotake, Tanzawa Mountains. M. Sc. thesis, University of Tokyo, Tokyo, Japan. (In Japanese). Miyaki, M., and K. Kaji. 2004. Summer forage biomass and the importance of litterfall for a highdensity sika deer population. Ecological Research 19:405–409. Nugent, G., and C. N. Challies. 1988. Diet and food preferences of white-tailed deer in northeastern Stewart Island. New Zealand Journal of Ecology 11:61–73. Nugent, G., W. Fraser, and P. Sweetapple. 2001. Top down or bottom up? Comparing the impacts of introduced arboreal possums and ‘terrestrial’ ruminants on native forests in New Zealand. Biological Conservation 99:65–79. Ozaki, T. 1997. The dying forest: A report from Nakanoshima Island, Toya Lake. The Society of Natural Plant Lovers in Iburi. Muroran, Japan. (In Japanese.) Picard, J. F., P. Oleffe, and B. Boisaubert. 1991. Influence of oak mast on feeding behaviour of red deer (Cervus elaphus L). Annals of Forest Science 48:547–559. Prisyazhnjuk, V. E., and N. P. Prisyazhnjuk. 1974. Sika deer (Cervus nippon Temm.) on Askold Island. Bulletin of the Moscow Society for Natural Research Department of Biology 79:16–27. (In Russian with English summary.) Rooney, T. P., and W. J. Dress. 1997. Species loss over sixty-six years in the ground-layer vegetation of Heart’s Content, an old-growth forest in Pennsylvania, USA. Natural Areas Journal 17:297–305.

180

M. Miyaki, K. Kaji

Russell, F. L., D. B. Zippin, and N. L. Fowler. 2001. Effects of white-tailed deer (Odocoileus virginianus) on plants, plant populations and communities: A review. American Midland Naturalist 146:1–26. Sukeno, M., and M. Miyaki. 2007. Impacts of an excessive sika deer population on vascular flora on Nakanoshima Islands, Toya Lake, Hokkaido, Japan. Wildlife Conservation Japan 11:43–66. (In Japanese with English summary.) Takahashi, H., and K. Kaji. 2001. Fallen leaves and unpalatable plants as alternative foods for sika deer under food limitation. Ecological Research 16:257–262. Takahashi, H., Y. Matsuura, M. Ueno, E. Shima, Y., Tanaka, J., Tanaka, and K. Kaji. 2005. Dystocia in free-ranging sika deer Cervus nippon under food limitation. Mammal Study 30:77–81. Takatsuki, S. 1989. Effects of deer on plants and plant communities. Japanese Journal of Ecology 35:67–80. (In Japanese with English summary.) Tremblay, J.-P., I. Thibault, C. Dussault, J. Huot, and S. D. Côté. 2005. Long-term decline in white-tailed deer browse supply: Can lichens and litterfall act as alternative food sources that preclude density-dependent feedbacks? Canadian Journal of Zoology 83:1087–1096.

Chapter 13

The Dynamics of Forest Stands Affected by Sika Deer on Nakanoshima Island—Change of Size Structure Similar to the Thinning Effect Masami Miyaki and Koichi Kaji

Abstract The influences of sika deer damage on forest dynamics after 1984 were monitored inside and outside exclosures for 16 years on Nakanoshima Island, Toya Lake, Hokkaido, Japan. The occurrences of bark-stripping were restricted to a few species and happened mostly in the early 1980s. Because the trees killed by bark-stripping were scattered throughout the stand, large gaps were not formed. In deer exclosures, seedlings ≤10 cm decreased after about 10 years following gap formation because of canopy closing. We analyzed the effect of deer on forest stands using yield-density diagrams. The decrease of tree density by deer damage accelerated the growth of remaining trees, having an effect similar to artificial thinning. It is important to set up specific goals of forest structure for sustainable forest management in sika deer habitat.

Introduction Studies of sika deer impact on forests using fenced exclosures have demonstrated that high deer density causes suppression of the recruitment of seedlings and small trees and reduction of plant species diversity (Stewart and Burrows 1989; Kaji and Yajima 1992; Takatsuki and Gorai 1994; Liang and Seagle 2002; Horsley et al. 2003). However, there is no quantitative information available on overall structural changes in forest dynamics. On Nakanoshima Island in Toya Lake, Hokkaido, Japan, forest damage by sika deer drastically increased after 1981. Tree density and basal area in 1983 in comparison to 1981 decreased to 48% and 78% respectively because of bark-stripping (Kaji and Yajima 1992). We have been monitoring forest dynamics and vegetation inside and outside the exclosures since 1984. In this chapter we predict the changes of the forest structures with high deer density using yield-density diagrams shown by Y-N curves and analyze forest dynamics in sites monitored for 16 years.

D. R. McCullough et al. (eds.), Sika Deer: Biology and Management of Native and Introduced Populations, DOI: 10.1007/978-4-431-09429-6_13, © Springer 2009

181

182

M. Miyaki, K. Kaji

Methods Study Area The research was conducted on the Nakanoshima Islands in Toya Lake, Hokkaido, Japan (140°51′ E, 42°36′ N), which are composed of one major island, Nakanoshima (497.8 km2) and three relatively small islands, Bentenjima, Kannonjima, and Manjujima. Deciduous forest dominates 87.1% of the islands. Forest gaps in deciduous forest, artificial conifer forest, and grassland are 4.7%, 6.3%, and 1.6% of the total area of the islands respectively. The major canopy species in deciduous forests were Quercus crispla (mizunara (white) oak), Kalopanax pictus (castor-aralia), Magnolia obovata (magnolia), Acer mono var. glabrum (painted maple), Maackia amurensis var. buergeri (maackia), Tilia japonica (linden), and Ostrya japonica (hop hornbeam). The forest floor was dominated by unpalatable species for sika deer such as Cephalotaxus harringtonia (plum-yew), Senecio cannabifolius (Aleutian ragwort), Chloranthus japonicus (chloranthus), Pachysandra terminalis (spurge), Daphne kamtschtica (daphne), Arisaema peninsulae (Japanese cobra lily), and Paeonia japonica (peony) (Takatsuki 1989; Kaji et al. 1991). Unpalatable species account for 99.99% of the total biomass of forest floor plants and tree leaves within the browsing layer (Miyaki and Kaji 2004). High population density of deer (20–80 deer/km2) has continued over the last 20 years (Kaji and Takahashi 2006). The principal diet of deer was fresh litterfall, even during summer, because of shortage of palatable plants on the forest floor (Takahashi and Kaji 2001; Miyaki and Kaji 2004, chapter 12).

Field Measurements Preferred tree species for bark-stripping were surveyed in plots with total areas of 2.71 ha from 1980 to 1983, when remarkable bark-stripping was observed. Sets of fenced exclosures (10 m × 20 m) and unfenced controls (10 m × 10 m) were set up at Manjujima (Exclosure-A and Control-A), Manjuzawa on Nakanoshima Island (Exclosure-B and Control- B), on Bentenjima (Exclosure-C and Control-C), and at Oowan on Nakanoshima Island (Exclosure-D) in 1984. The distribution of trees including those killed by bark-stripping were recorded in these plots in 1984. Tree diameters were measured in the plots in 1984, 1994, and 2000 except in Control-D which was damaged by the construction of a road. The tree volumes were then calculated using the allometry between these diameters and volumes previously determined on the islands. In order to show the light conditions in each plot, relative photosynthetically active radiation (PAR) on the forest floor was measured in exclosures and controls in 1999.

13 The Dynamics of Forest Stands Affected by Sika Deer

183

Evaluation of Forest Structure We used the yield-density diagram that is composed of Y-N curves and other accompanying curves to describe forest structure (Kikuzawa 1982, 1999). Y-N curves express the relation between cumulative summation (Y) of individual tree volume from maximum sized tree to one of a certain given volume, and number (N) of these trees on a double logarithmic scale. The curve is expressed by the reciprocal equation: 1/ Y =B / N+A where A and B are parameters (Kikuzawa 1982). The position of the Y-N curve is determined by the coordinates of the B-point (NB, YB). This point is the position on the Y-N curve where YB is one-half of its asymptotic value. YB and NB are expressed as: YB = 1 / (2A) NB = B / A. With comparison of the Y-N curves of exclosure stands and control stands, or the changes within each stand in a time series, we can comprehend not only stand volumes and tree densities, but also forest structures and dynamics such as even compact stands or uneven loose stands, tree size classes with high mortality, and effects of the growth of remaining trees after thinning (Kikuzawa 1982). We predicted changes of the forest structure and their Y-N curves with high deer density as follows (Fig. 13.1): Stage 1: While the deer density increases at the early stage, the forest loses small trees because of browsing and bark-stripping of young trees. The Y-N curve, therefore, becomes shorter at the right side of the curve. Stage 2: As the deer density becomes higher, bark-stripping extends to medium and large sized trees. The B-point moves down from upper right to lower left, assuming that mortalities and growth rates are equal among tree size classes. Stage 3: Bark-stripping decreases because of fewer palatable trees available, even though deer density is still high. The remaining trees grow in good condition and the total stand volume increases. The B-point moves upward assuming that all trees survive and are equal in growth rate. Stage 4: After the crown layer is closed, the number of trees decrease because of competition among trees, yet the stand volume still increases. The B-point moves from lower right to upper left assuming that mortalities and growth rates are equal among size classes. Stage 5: Lastly, both number of trees and the stand volume decrease. The areas of forest gaps expand in this stage. The B-point goes down again. Parameters A and B in Y-N curve of each stand were estimated using the least square method.

184

M. Miyaki, K. Kaji

Cumulative Volume, Y

1

4 2

5 3

Cumulative number, N

Fig. 13.1 Prediction about the change of forest structure with high deer density. Five stages are shown by the movements of the Y-N curves and B-points (•). Arrows indicate the order over time and numerals show the stages as follows: Stage 1: Loss of small-sized trees. Stage 2: Death caused by bark-stripping of medium- and large-sized trees. Stage 3: Growth of individual trees and increase of stand volume. Stage 4: Decrease of number of trees and increase of stand volume. Stage 5: Decrease of number of trees and stand volume.

Results Distribution of Bark-Stripping Trees and Light Condition on Forest Floor Out of 34 tree species observed in the islands from 1980 to 1983, only six species, Hydrangea petiolaris (climbing hydrangea), Ulmus davidiana (Japanese elm), Picrasma quassioides (India quassiawood), Cornus controversa (giant dogwood), Styrax obassia (fragrant snowbell), and Taxus cuspidata (Japanese yew), comprised 67.9% of trees killed by bark-stripping, whereas the number of trees of these species was only 17.5% of the total number of trees (Fig. 13.2). The densities of dead trees at height ≥1.3 m caused by bark-stripping were different among the plots in 1984 (Fig. 13.3). However, dead trees tend to be scattered in each plot. By six years after establishment of the plots, seedlings in the unfenced controls were almost totally restricted to those ≤10 cm in height, which were current-year seedlings (Fig. 13.4). In the exclosures, the number and ratio of seedlings ≤10 cm decreased after about eight years since exclosure establishment or after about eight to 10 years of gap formation by bark-stripping (Fig. 13.4). In unfenced controls the recruitment of saplings to tree class at height ≥1.3 m was not observed. Meanwhile in Exclosure-A and Exclosure-B the recruitment

13 The Dynamics of Forest Stands Affected by Sika Deer

185

Fig. 13.2 Tree species preferred by sika deer for bark-stripping from 1980 to 1983.

Fig. 13.3 Distribution of trees in 10 × 20 m and 10 × 10 m plots in 1984 • = living tree, x = tree killed by bark-stripping.

186

M. Miyaki, K. Kaji 80 70 60 50

Control C (Bentenjima)

60

Seedling height (cm)

50 < 10 cm

30

30

10 ⬉ ⬍ 50

20

20

20

50 ⬉ ⬍ 130

10

10

10

0

0

0

Exclosure D (Oowan)

100 90

Exclosure C

90

80

80

70

70

70

60

60

60

50

50

50

Exclosure B

Exclosure A

40

40

30

30

30

30

20

20

20

20

10

10

10

10

0

0

0

0 1984 1986 1988 1990 1992 1994 1996 1998

40

1984 1986 1988 1990 1992 1994 1996 1998

40

1984 1986 1988 1990 1992 1994 1996 1998

1984 1986 1988 1990 1992 1994 1996 1998

1988 1990 1992 1994 1996 1998

30

1984 1986

40

1984 1986 1988 1990 1992 1994 1996 1998

40

1984 1986 1988 1990 1992 1994 1996 1998

No. of seedlings / 4m2

Control A (Manjujima)

Control B (Manjuzawa)

Fig. 13.4 Changes in number of seedlings in the unfenced control plots and fenced exclosures.

from saplings increased after 1984 when the exclosures were set up. After 1994, however, the number of trees ≤5 cm decreased in Exclosure-A (Fig. 13.5). Further, in Exclosure-C and Exclosure-D where numbers of seedlings has decreased, there were few or no recruits from sapling to small tree class. Figure 13.6 shows the relationship between relative photosynthetically active radiation (PAR) at the forest in 1999 and growth rate of stand volume/year from 1994 to 1999. Relative PAR and growth rate of stand volume showed a significant positive correlation (r = 0.808, p < 0.01). Differences between exclosures and controls in terms of both factors were not significant. Relative PARs were <2% and growth rates of stand volumes were nearly 2% in six of nine stands.

Evaluation of Forest Structure Figure 13.7 shows the movements of Y-N curves from 1984 to 2000. The damage by bark-stripping was relatively small in Exclosure-A and Control-A located on a small island where deer density was lower than the other areas. The B-points on Y-N curves in both stands moved from right to left. The number of medium- and largesized trees decreased. In Exclosure-A, total stand volume increased slightly, and

13 The Dynamics of Forest Stands Affected by Sika Deer

187

Fig. 13.5 Changes of number of trees in the exclosures in 1984, 1994, and 2000.

Fig. 13.6 The relationship between relative photosynthetically active radiation (PAR) in 1999 and growth rate of stand volume from 1994 to 1999.

188

M. Miyaki, K. Kaji

Fig. 13.7 Movements of Y-N curves in the controls and the exclosures in 1984, 1994, and 2000. (A) Manjujima, (B) Manjuzawa, (C) Bentenjima, (D) Oowan.

13 The Dynamics of Forest Stands Affected by Sika Deer

189

the number of trees also increased due to recruitment from saplings. In Control-A, there was a small decrease in total stand volume. Therefore, we diagnosed the impacts of deer on forest structures as stage 2 to stage 3 for Exclosure-A and stage 2 for Control-A. For Exclosure-B, Control-B, Exclosure-C, and Control-C, we diagnosed these to be stage 3 from 1984 to 2000 because the B-points moved upward. The growth in stand volume was evident after 1994. Numbers of trees decreased except for Exclosure-B. In Exclosure-D, the B-points moved upward until 1984. But after that, the growth of the stand volume was slower than B and C plots although there were a few recruits from saplings. This tendency suggests competition among trees. We diagnosed Exclosure-D as being in stage 3 or stage 4.

Discussion Impact of Deer on Forest Dynamics Bark-stripping was restricted to few species such as U. davidiana, S. obassia, and H. petiolaris. These tree species are scattered in deciduous forest, so that few large gaps were formed when death was caused by bark-stripping. In the unfenced controls, large-size seedlings (≥50 cm) became rare or disappeared after 1988, because high deer density was maintained. In the exclosures, small seedlings (≥10 cm) increased immediately after fencing. However they had decreased dramatically eight to 10 years after gap formation by bark-stripping. First-year seedlings of shade-intolerant hardwood species with 2% light showed virtually no growth and had a high rate of mortality, while second-year survivors of shade-tolerant hardwood seedlings had a high rate of mortality (Walters and Reich 1996). Few saplings, therefore, could regenerate in most of the exclosures due to the fact that tree canopies had closed, and light was in short supply on the forest floor. Among the stands where seedling densities were low, recruitment of small trees was also very low in spite of the absence of deer. This analysis of forest dynamics using Y-N curves presented useful information. Stage 1 and Stage 2 of the Y-N curves would have already passed when exclosures were established in 1984 because there were almost no small trees left for browsing, nor large trees for bark-stripping. Vertically translocation of the curve in Stage 3 is similar to that of a stand immediately after artificial thinning (Kikuzawa 1982). In this stage, the forest layer opens and the growth rate of trees is almost equal regardless of their size. Translocation in Stage 3 would be typical after the large-scale damage caused by browsing and bark-stripping in the deciduous broad-leaved forests. Stage 4 has a similar pattern as a stand in which growth reaches full density, reducing its density by natural thinning along the full density

190

M. Miyaki, K. Kaji

curve from lower-right to upper-left (Kikuzawa 1982). This means that the forest layer is closed and the number of trees is reduced even though stand volume is increasing. We can, therefore, conclude that the growth of the stand has recovered in Stage 4. Stage 5 was not observed in the islands. If a high density of deer lasts for long time, the curve will move to Stage 5. It may take as long as the tree life span to reach stage 5.

Target of Forest Management and Evaluation of Deer Damage Damage, or negative effects of browsing and bark-stripping are loss of seedlings and young trees (Stewart and Burrows 1989; Shimoda et al. 1994; Takatsuki and Gorai 1994; Akashi and Nakashizuka 1999; Gill and Beardall 2001; Rooney 2001) and reduction of species diversity (Gill and Beardall 2001; Liang and Seagle 2002; Horsley et al. 2003). These tendencies were remarkable on Nakanoshima Island in Toya Lake. However, the objective of managing the extent of damage must be established by comparing the current status against a target (Reimoser et al. 1999). Generally, recruitment of seedlings is also negligible in a mature forest where the light is limited on the forest floor. Furthermore, in the case of species that are lost, but have long seed dispersal distances, it is possible for them to reestablish quickly after the deer density declines. Therefore, if the target of management is to maintain a given forest structure for a certain period of time, it may not be necessary to reduce deer density immediately. The other effect imposed by deer is that it accelerates the growth of remaining trees in the same way as artificial thinning. This can be regarded as a benefit for the remaining trees. The deer density that has a minimal impact on vegetation is very low compared with the density that yields the maximum sustained yield of deer population (Bailey 1984; Rooney 2001). However, it is not always necessary to reduce the deer population to a level at which the forest can regenerate. The time scale of forest dynamics is much longer than that of deer population, and many forests can survive for several decades without recruitment of seedlings. Therefore, forest managers can select various deer densities to maintain the forest and to manage deer at the same time. A suitable deer density will have to be chosen according to the present forest structure, and various targets for management, such as the immediate recruitment of saplings, conservation of rare species, maximum deer harvest, deer watching or esthetic use of deer, prevention of erosion of the forest floor, etc. Acknowledgements The authors thankfully acknowledge Mitsuru Saito for completing

the early survey and members of Hokkaido Institute of Environmental Sciences and research group of Forest Resource Science, Hokkaido University for assistance in the field. Takashi Yajima, Kiyoshi Umeki, and Saiha Sung made constructive comments on earlier versions of this chapter.

13 The Dynamics of Forest Stands Affected by Sika Deer

191

Literature Cited Akashi, N., and T. Nakashizuka. 1999. Effects of bark-stripping by sika deer (Cervus nippon) on population dynamics of a mixed forest in Japan. Forest Ecology and Management 113:75–82. Bailey, J. A. 1984. Principles of wildlife management. Wiley, New York, New York, USA. Gill, R. M. A., and V. Beardall. 2001. The impact of deer on woodlands: the effects of browsing and seed dispersal on vegetation structure and composition. Forestry 74:209–218. Horsley, S. B., S. L. Stout, and D. S. DeCalesta. 2003. White-tailed deer impact on the vegetation dynamics of a northern hardwood forest. Ecological Applications 13:98–118. Kaji, K., and T. Yajima. 1992. Influence of sika deer on forests of Nakanoshima Island, Hokkaido. Pages 215–218 in B. Bobek, K. Perzanowski, and W. Regelin, editors, Global trends in wildlife management. Transactions 18th International Union of Game Biologists Congress, Krakow 1987. Kaji, K., and H. Takahashi. 2006. Where do density dependent effects appear? Relationship between body weight and pregnancy rate for sika deer on Nakanoshima Island. Pages 43–48 in K. Kaji, M. Miyaki, and Y. Uno, editors, Conservation and management of sika deer in Hokkaido. Hokkaido University Press, Sapporo, Japan. Kaji, K., T. Yajima, and T. Igarashi. 1991. Forage selection by deer introduced on Nakanoshima Island and its effect on the forest vegetation. Pages 52–55 in N. Maruyama, B. Bobek, Y. Ono, W. Regelin, L. Bartos, and P. R. Ratcliffe, editors, Proceedings of the International Symposium on Wildlife Conservation, International Congress of Ecology 1990, Tsukuba and Yokohama, Japan. Japan Wildlife Research Center, Tokyo, Japan. Kikuzawa, K. 1982. Yield-density diagram for natural deciduous broad-leaved forest stands. Forest Ecology and Management 4:341–358. Kikuzawa, K. 1999. Theoretical relationships between mean plant size, size distribution and self thinning under one-sided competition. Annals of Botany 83:11–18. Liang, S. Y., and S. W. Seagle. 2002. Browsing and microhabitat effects on riparian forest woody seedling demography. Ecology 83:212–227. Miyaki, M., and K. Kaji. 2004. Summer forage biomass and the importance of litterfall for a highdensity sika deer population. Ecological Research 19:405–409. Reimoser, F., H. Armstrong, and R. Suchant. 1999. Measuring forest damage of ungulates: what should be considered? Forest Ecology and Management 120:47–58. Rooney, T. P. 2001. Deer impacts on forest ecosystems: A North American perspective. Forestry 74:201–208. Shimoda, K., K. Kimura, M. Kanzaki, and K. Yoda. 1994. The regeneration of pioneer tree species under browsing pressure of sika deer in an evergreen oak forest. Ecological Research 9:85–92. Stewart, G. H., and L. E. Burrows. 1989. The impact of white-tailed deer Odocoileus virginianus on regeneration in the coastal forests of Stewart Island, New Zealand. Biological Conservation 49:275–293. Takahashi, H., and K. Kaji. 2001. Fallen leaves and unpalatable plants as alternative foods for sika deer under food limitation. Ecological Research 16:257–262. Takatsuki, S. 1989. Effects of deer on plants and plant communities. Japanese Journal of Ecology 35:67–80. (In Japanese with English summary.) Takatsuki, S., and T. Gorai. 1994. Effects of sika deer on the regeneration of a Fagus crenata forest on Kinkazan Island, Northern Japan. Ecological Research 9:115–120. Walters, M. B., and P. B. Reich. 1996. Are shade tolerance, survival, and growth linked? Low light and nitrogen effects on hardwood seedlings. Ecology 77:841–853.

Chapter 14

Biology of Sika Deer in Hyogo: Characteristics of Reproduction, Food Habits, Growth, and Condition Mayumi Yokoyama

Abstract The breeding condition, nutritional status, body mass and measurements, food habits, and habitat quality of sika deer in a high-density population in Hyogo Prefecture in southwestern Japan have been measured since 2002 as part of an ongoing monitoring program. These data are necessary for sound management to maintain the quality of both the deer population and their habitat. Compared to more northern populations, sika deer in Hyogo are smaller in stature and show less response to seasonal changes: sika deer can get evergreen broad-leaves even in winter and there is little change in fat deposits in both sexes. Estimated conception dates are September to October; the same as of ten years ago. The conception timing in Hyogo is about three weeks earlier than in Hokkaido. These and other physiological adaptations are related to the more benign climate and vegetation characteristics of the evergreen broad-leaved forest habitats of Hyogo as compared to those of northern Japan.

Introduction Biological attributes of ungulates that affect vital rates, such as fertility and survival, can change with habitat and climate conditions and population density (Hanks 1981; Albon et al. 1992; Keech et al. 2000). Therefore, to properly manage sika deer in Hyogo Prefecture it is necessary to maintain a healthy, high quality population to retain its high value as a hunting resource. Biological features of sika deer in Hyogo Prefecture reflect adaptations to a temperate environment in southwestern Japan (Fig. 14.1). Density of the sika deer population increased mainly in central Honshu in Hyogo Prefecture at the beginning of the 1990s, followed by expansion of the distribution to the surrounding area (Yokoyama and Sakata 2007; Sakata et al. chapter 31). Now Hyogo Prefecture is a core area in the distribution of the sika deer population in the region and has the highest population in Honshu Island. In recent years it has had the highest hunting take (per unit of forested area) in the country. Also, the prefecture has a small closed population on Awaji Island. D. R. McCullough et al. (eds.), Sika Deer: Biology and Management of Native and Introduced Populations, DOI: 10.1007/978-4-431-09429-6_14, © Springer 2009

193

194

M. Yokoyama

Fig. 14.1 Map showing the location of Hyogo Prefecture.

Because of excessive feeding by deer, impacts on the vegetation, such as bark peeling and disappearance of the understory of the forest (Fujiki et al. 2006), have worsened in the high density areas of both central Honshu and Awaji Island. This situation reduced the nutritional condition of deer and resulted in smaller body size and lower pregnancy or survival rates; these changes might eventually cause a population reduction. In order to monitor the condition of the sika deer population over time, in addition to the monitoring of availability of food plants, there needs to be a qualitative measurement of deer condition such as individual nutrition, breeding status, and body size. In this chapter I describe sika deer breeding, body size, food habits, and nutritional condition based on research results on individual deer captured in Hyogo Prefecture since 2002.

Habitat Hyogo Prefecture is located in the southwestern part of Honshu, at latitude 35°15¢ N and longitude 135° E (Fig. 14.1) and consists of part of Honshu and Awaji Islands. The Chugoku Mountains pass through the middle of Honshu and the north faces the Sea of Japan while the south faces the Seto Inland Sea. Therefore, the climate and vegetation differ from north to south. In the center of deer distribution in the central Honshu area the average annual temperature is 13.4 °C, while in the southernmost habitat on Awaji Island it is 16.1 °C. Mostly in the central part of the Honshu there is a watershed called the Hikami corridor which is renowned as the lowest elevation watershed in Japan. Forest occupies about 67% of whole prefecture, of which 94%

14 Biology of Sika Deer in Hyogo

195

is privately owned. From the 1950s to the 1960s, there was intense planting of Japanese cedar, Cryptomeria japonica, and Japanese cypress, Chamaecyparis obtusa, so still about 40% of the forest is plantation. The northern part is mainly deciduous broad-leaved trees such as Quercus serrata (an oak) and Clethra barbinervis (sweet pepperbush). The central part is occupied by plantation and mixed forests of evergreen and deciduous broad-leaved trees, including Q. serrata, Ligustrum japonicum (privet), and Eurya japonica (Theaceae). In the part that faces Seto Inland Sea and Awaji Island there are evergreen broad-leaved trees, mainly Castanopsis sieboldii (Fagaceae) and Quercus glauca (blue oak). Sika deer cause severe agricultural and forestry damage. Almost all commercial agricultural areas have been protected from sika deer by fencing. In most natural forests, there is a deer high-line and sparse understory vegetation layer (Fujiki et al. 2006). In the southern part of the area and along the Seto Inland Sea coast there are no sika deer.

Materials We collected sika deer from 2002 to 2007 during hunting and damage control activities, mainly in Honshu and on Awaji Island of Hyogo Prefecture. Most animals were shot at random. Ages were determined by tooth replacement and counts of cementum annuli in the first incisor using criteria for sika deer (Koike and Ohtaishi 1985).

Reproduction According to studies done since 1990, the pregnancy rate of female deer aged two years and older in Hyogo Prefecture was high (83–100%). The yearling female pregnancy rate was around 70%, however since 1996 it has declined to about 60–70% (Ozaki et al. 2001). The pregnancy rate of adult females taken in 2003 was 100%, the sex ratio of the embryos was 1:1, and the mean mass of the embryos just before birth was about 4 kg (Yokoyama et al. 2003). The estimated conception date (Koizumi et al. chapter 24) was concentrated in two weeks from the last week in September to the first week in October in Honshu and in early October on Awaji (Fig. 14.2). More than 90% of the conceptions by adult females were achieved by the end of October (Fig. 14.2) (Yokoyama et al. 2003). Only one two-year-old female conceived in mid-January. Late conception might occur in young females in which body condition is poor. When compared with the result of about ten years ago (Koizumi 1991), there was no difference in the distribution of the conception dates of 2002–2004 in Honshu (t = 0.617, n = 170, p > 0.1) and, even under the latest high-density population, timing of conception date was not delayed. The peak of the birth date is near the end of May, about three weeks earlier than in Hokkaido Island (Suzuki

196

M. Yokoyama

Fig. 14.2 Estimated conception date of sika deer in Hyogo Prefecture. This was estimated from fetal body mass collected in 2002–2007 using the formula derived by Koizumi et al. (chapter 24).

et al. 1996; Ohnishi et al. chapter 7). This is related to the time of spring greenup, the period of highest nutritive value of plants, which in Hyogo is about three weeks earlier than in Hokkaido. However, the result of fetal growth in 2005–2007 tends to be delayed compared with the result of 2002–2004. Continued high-density over time could lead to serious habitat deterioration and negative influences on fetal growth, as well as other population parameters.

Food Habits Three categories, evergreen broad-leaves, deciduous broad-leaves/forbs, and graminoids, made up the main diet of sika deer in Honshu, accounting for more than 70% of rumen contents of sika deer taken year-round in 2002 and 2003 (Fig. 14.3) (Yokoyama et al. 2003). Graminoids, such as sasa bamboo (Sasa nipponica) or sedge (Carex spp.) and deciduous broad-leaves or forbs were the main diet in spring and summer. Since damage by bark peeling increased, nonassimilated parts that occupied about 10% of rumen contents in summer consisted not only of branches but also bark of trees. The use of nuts/seeds and agricultural products increased in autumn. In winter, the leaves of evergreen broad-leaved trees and the nonassimilated parts such as bark was greater than at other seasons, occupying about 20% of the diet respectively (Figs. 14.3 and 14.4a). For Awaji, evergreen broad-leaves occupied 43% of rumen contents in winter. This indicates that sika deer on Awaji could get high quality food even in winter (Fig. 14.4b). Hyogo sika deer diets have a balance of browse and forbs even in winter.

14 Biology of Sika Deer in Hyogo

197

Fig. 14.3 Seasonal diets of sika deer in Hyogo. Rumen contents were analyzed for each season in the years 2002 and 2003.

Fig. 14.4 Winter diets compared for central Honshu (a) and Awaji Island (b) in 2002–2003 (Yokoyama et al. 2003), and central Honshu in 1988 (c) (Koizumi et al. 1993).

198

M. Yokoyama

Compared to the food analysis in the central Honshu area in the winter of 1988 (Koizumi et al. 1993), the proportion of leaves of broad-leaved trees has increased, whereas that of the leaves of coniferous trees and nonassimilated parts has decreased (Fig. 14.4a, c). It is thought that this was due to the high level of damage to young plantation trees of Japanese cedar and Japanese cypress at that time. According to our field observations in high-density areas, browsing was concentrated on evergreen broad-leaved trees such as Eurya japonica, Aucuba japonica (Japanese laurel), Lindera umbellata (spice bush), Camellia sinensis (tea), deciduous broad-leaved trees such as Clethra barbinervis, Ilex pedunculosa (holly), Rubus palmatus (Rosaceae), graminoids such as Sasa nipponica, Carex morrowii, Miscanthus sinensis (silvergrass), and coniferous trees such as Cryptomeria japonica and Chamaecyparis obtusa. Although Pieris japonica, Neolitsea sericea (Lauraceae), Caesalpinia sepiaria var. japonica (Leguminosae), and Acer rufinerve (redvein maple)were common in the study area, we found no incisor marks on them. Average crude protein of rumen contents in the Honshu area was 17% in spring and declined to 11% in winter (Fig. 14.5). Estimates of the minimum dietary protein requirement for maintenance in ruminants are from 5% to 9% (McCullough 1979; Schwartz et al. 1987; Robbins 1993), while maximum growth requires from 13% to 20% (Ullrey et al. 1967; Robbins 1993). Our results suggest that dietary protein in the sika deer diets in Hyogo is probably higher than the minimum requirements for ruminants, even in winter. The above results show that 70% of the deer foods such as deciduous broadleaved trees or graminoids have high nutritive value throughout the year. Moreover, the nonassimilated parts, which are low quality foods, increase more in winter than other seasons. However, since the use of leaves of evergreen broad-leaved trees also increases in winter, the qualitative decline is thought to be small as compared with the decline in northern Japan (Yokoyama et al. 2000; Ichimura et al. 2004). In autumn the use of agricultural products and nuts/seeds were confirmed, thereby supporting the claim of agricultural damage by deer. It is thought that since 1988

Fig. 14.5 Crude protein of rumen contents (%) of males and females in Hyogo sika deer.

14 Biology of Sika Deer in Hyogo

199

deer food habits have diversified, and thus diet quality has been maintained in the Honshu area of Hyogo.

Growth and Condition Body Size and Growth The sika deer of Hyogo Prefecture are medium in size among Honshu deer. Growth curves (von Bertalanffy equation Y(t) = A{1 - 1/3 exp [-K (t - I)]}; Zullinger et al. 1984) were fitted to body mass (Fig. 14.6), body length, body height (measured at the shoulder), and hind foot length (Fig. 14.7). The growth curve equation provided a good description of the pattern of growth for each measurement (Table 14.1). Body mass did not fit sine or cosine waves like Hokkaido sika deer (Suzuki et al. 2001). Rather body mass fluctuation indicated individual variation in Hyogo sika deer. Growth of body height and hind foot ceased after two years of age in both sexes. However, body mass continued to grow until five years of age in males, and three years of age in females, at which time growth was completed. Although the body length of males and females at maturity was different (p < 0.01), there was no significant difference between females and males in either body height or hind foot length (Table 14.2). Sika deer show greater sexual differences in Hokkaido (Suzuki et al. 2001) than in Hyogo (Table 14.3). However, these results might be affected by the young age of males in Hyogo because of high hunting pressure, and the characteristics of the von Bertalanffy equation in the calculation of upper asymptotes for growth. Nonetheless, these results show that for Hyogo males, mass increases relatively greater with age than do body dimensions. And, further, that Hyogo deer are not only smaller in stature than Hokkaido deer, but our results strongly suggest that reduction in Hyogo male size is caused by density-dependent effects as well.

Fig. 14.6 Body mass of male and female Hyogo sika deer.

200

M. Yokoyama

Fig. 14.7 Growth curves for sika deer shot in Hyogo in 2002–2003 fitted by the von Bertalanffy equation for body length, shoulder height, and hind foot length.

Antler Size Growth of antlers in adult males is completed late in August. Antlers have one point at the age of one, and 70% become four points at the age of two. The maximum length of antlers is about 50 cm, and they tend to be nearly as wide as high (Fig. 14.8).

Body Condition Mandible marrow fat (MCF) showed no seasonal variation, being about 50% on the average; however, the individual difference is large (Fig. 14.9; Yokoyama et al.

14 Biology of Sika Deer in Hyogo

201

Table 14.1 Growth parameters in body length, body height, and hind foot length of sika deer in Hyogo. Parameter estimates for the von Bertalanffy equation, which takes the form: size (mm) = A {1-1exp[-k(t-I)]}, where t is age (month), A is the asymptotic value, K is the growth rate constant, and I is age at the inflection point. Measurement N Sex A K I R2 Body length Body height Hind foot length

40 39 42 38 42 40

Male Female Male Female Male Female

87.37 80.66 80.90 77.60 41.37 39.94

0.09 0.14 0.13 0.17 0.16 0.21

5.50 1.85 2.53 0.95 0.64 -0.82

0.91 0.88 0.87 0.87 0.87 0.93

Table 14.2 Mean and standard deviation of body length, body height, and hind foot length in Hyogo sika deer calculated for the age class where growth was essentially complete. Measurement N Sex x– SD Body length Body height Hind foot length

16 28 17 27 17 29

Male (≥3 years old) Female (≥2 years old) Male (≥2 years old) Female (≥2 years old) Male (≥2 years old) Female (≥2 years old)

86.90 80.93 79.90 78.10 40.40 40.10

8.15 4.73 8.98 5.50 3.14 1.91

P < 0.01 n.s.* n.s.*

*Not significant Table 14.3 Male/female ratios derived from asymptotic vales (A) from the von Bertalanffy equation. Male/female ratio Body length Body height Hind foot length

1.08 1.04 1.04

2003). Therefore, marrow fat was not useful in this study to follow seasonal changes in nutritional condition. It is known that marrow fat is the last body fat deposit that is mobilized (Ransom 1965) and, thus, is useful mainly to indicate starvation conditions; it remains stable under lesser nutritional stress, so other measures are used for seasonal fluctuations in body condition. Kidney fat mass (KFM), conversely, is a commonly used measure, the virtual standard of seasonal changes in body fat content (Leader-Williams and Ricketts 1981; Albon et al. 1986). Growth parameters for KFM were fitted to the von Bertalanffy confined cosine wave (Yokoyama et al. 2001), but indicated low seasonal fluctuation and low coefficient of determination for both males (R2 = 0.23) and females (R2 = 0.22). Compared to Hokkaido deer, Hyogo deer had no remarkable seasonal variation and remarkably low levels of KFM (Fig. 14.10). Given the breeding situation, intake of food, and the low fat accumulation seen in sika deer in Hyogo Prefecture, there is little to indicate a condition of chronic

202

M. Yokoyama

Fig. 14.8 Antler size of male sika deer in Hyogo. Antler width is the greatest outside measurement across the two antlers.

Fig. 14.9 Seasonal changes of kidney fat mass (KFM) (a) and mandible marrow fat (MCF) (b) of adult male and female in sika deer in Hyogo.

Fig. 14.10 Growth and seasonal fluctuations in kidney fat mass (KFM) of sika deer in Hyogo (a) and in Hokkaido (b) of males (°) and females (•). (Hokkaido sika deer modified from Yokoyama et al. 2001.)

14 Biology of Sika Deer in Hyogo 203

204

M. Yokoyama

nutritional stress. They do not accumulate fat in preparation for a seasonal change of the food environment, as is typical of deer in the north. This is considered a physiological adaptation to a stable food supply throughout the year in the evergreen broad-leaved forest. Similarly, sika deer inhabiting the evergreen broadleaved forest in Chiba Prefecture in central Japan show less accumulation of body fat (Asada 1996), because it has a mild winter and good diet quality year-round (Asada and Ochiai 1996). Unlike northern Japan, sika deer inhabiting temperate zone evergreen broad-leaved forest are rarely exposed to nutritional stress and low temperature throughout the year. However, their physiological characteristic of inefficient energy conversion, which does not include body fat accumulation for overwintering or calving, might cause a rapid fall of nutrition due to unusual weather events. If a sudden shortage of food happens during periods of low temperature or snow fall, which lower the quantity and quality of foods, deer might be temporarily stressed. Especially in cases where sika deer distribution has expanded into usually more severe climatic areas during the recent warming trend (north Hyogo), or remained at high density for more than ten years (Awaji Island) there are possibilities of mortality in winter, if snow and low temperature become severe even if for a short period.

Literature Cited Albon, S. D., B. Mitchell, B. J. Huby, and D. Brown. 1986. Fertility in female red deer (Cervus elaphus): The effects of body composition, age and reproductive status. Journal of Zoology (London) (A) 209:447–460. Albon, S. D., T. H. Clutton-Brock, and R. Langvatn. 1992. Cohort variation in reproduction and survival: Implication for population demography. Pages 423–428 in R. D. Brown, editor, The biology of deer. Springer, New York, New York, USA. Asada, M. 1996. Ecological status of sika deer in Boso Peninsula, central Japan. Ph.D. thesis, University of Tokyo, Tokyo, Japan. (In Japanese.) Asada, M., and K. Ochiai. 1996. Food habits of sika deer on the Boso Peninsula, central Japan. Ecological Research 11:89–95. Fujiki, D., M. Suzuki, F. Goto, M. Yokoyama, and H. Sakata. 2006. Tree community structure in secondary forests affected by sika deer (Cervus nippon) herbivory. Japanese Journal of Conservation Ecology 11:21–34. Hanks, J. 1981. Characterization of population condition. Pages 47–73 in C. W. Fowler and T. D. Smith, editors, Dynamics of large mammal populations. Wiley, New York, New York, USA. Ichimura, Y., H. Yamamoto, T. Takano, S. Koike, Y. Kobayashi, K. Tanaka, N. Ozaki, M. Suzuki, H. Okada, and M. Yamanaka. 2004. Rumen microbes and fermentation of wild sika deer on the Shiretoko Peninsula of Hokkaido Island, Japan. Ecological Research 19:389–395. Keech, M. A., R. T. Bowyer, J. M. Ver Hoef, R. D. Boertje, B. W. Dale, and T. R. Stephenson. 2000. Life-history consequences of maternal condition in Alaskan moose. Journal of Wildlife Management 64:450–462. Koike, H., and N. Ohtaishi. 1985. Prehistoric hunting pressure estimated by the age composition of excavated sika deer (Cervus nippon) using the annual layer of tooth cement. Journal of Archaeological Science 12:443–456. Koizumi, T. 1991. Reproductive characteristics of female sika deer, Cervus nippon, in Hyogo Prefecture, Japan. Ongules/Ungulates 91:561–563.

14 Biology of Sika Deer in Hyogo

205

Koizumi, T., S. Yamazaki, and M. Kobayashi. 1993. Food habit of sika deer in winter: Comparison of Hyogo and Kochi Prefecture. Japanese Forestry 104:691–692. (In Japanese.) Leader-Williams, N., and C. Ricketts. 1981. Seasonal and sexual patterns of growth and condition of reindeer introduced into South Georgia. Oikos 38:27–39. McCullough, Y. 1979. Carbohydrate and urea influences on in vitro deer forage digestibility. Journal of Wildlife Management 43:650–656. Ozaki, S., S. Shiomi, and Y. Kamiyama. 2001. Population dynamics of sika deer in South Tajima area in Hyogo Prefecture, (II). Study of Applied Forestry 10:105–109. Ransom, A. B. 1965. Kidney and marrow fat as indicators of white-tailed deer condition. Journal of Wildlife Management 29:397–398. Robbins, C. T. 1993. Wildlife feeding and nutrition, 2nd edition. Academic, New York, New York, USA. Schwartz, C. C., W. L. Regelin, and A. W. Franzmann. 1987. Protein digestion in moose. Journal of Wildlife Management 51:352–357. Suzuki M., K. Kaji, M. Yamanaka, and N. Ohtaishi. 1996. Gestational age determination, variation of conception date, and external fetal development of sika deer (Cervus nippon yesoensis Heude, 1884) in Eastern Hokkaido. Journal of Veterinary Medical Science 58:505–509. Suzuki, M., M. Onuma, M. Yokoyama, K. Kaji, M. Yamanaka, and N. Ohtaishi. 2001. Body size, sexual dimorphism, and seasonal mass fluctuations in a larger sika deer subspecies, the Hokkaido sika deer (Cervus nippon yesoensis Heude, 1884). Canadian Journal of Zoology 79:154–159. Ullrey, D. E., W. G. Youatt, H. E., Johnson, L. D. Fay and B. L. Bradley. 1967. Protein requirements of white-tailed deer fawns. Journal of Wildlife Management 31:679–685. Yokoyama, M., and H. Sakata. 2007. Current status and issues of the Specified Wildlife Conservation and Management Plans for sika deer in Hyogo Prefecture, Japan. Mammalian Science 47:73–79. Yokoyama, M., K. Kaji, and M. Suzuki. 2000. Food habits of sika deer and nutritional value of sika deer diets in eastern Hokkaido, Japan. Ecological Research 15:345–356. Yokoyama, M., M. Onuma, M. Suzuki, and K. Kaji. 2001. Seasonal fluctuations of body condition in northern sika deer on Hokkaido Island, Japan. Acta Theriologica 46:419–128. Yokoyama, M., H. Sakata, S. Hamasaki, and M. Mitani. 2003. Condition assessment of sika deer population in Hyogo Prefecture: Characteristics of reproduction, food habits and nutritional condition. Humans and Nature 14:21–31. Zullinger, E. M., R. E. Ricklefs, K. H. Redford, and G. M. Mace 1984. Fitting sigmoidal equations to mammalian growth curves. Journal of Mammalogy 65:607–636.

Chapter 15

Bark-Stripping Preference of Sika Deer and Its Seasonality on Mt. Ohdaigahara, Central Japan Masaki Ando and Ei’ichi Shibata

Abstract The recent increase in the sika deer (Cervus nippon) population has caused dieback of overstory trees due to bark-stripping on Mt. Ohdaigahara, central Japan, resulting in decline of the forest and expanding of open dwarf bamboo (Sasa nipponica) grassland. We evaluated bark-stripping preference of sika deer and its seasonality. Deer stripped bark selectively: some species, such as Hondo spruce and Nikko fir, were often debarked while some other species, such as Siebold’s beech, not often were. Bark stripping was most intensive during summer when the deer’s main forage, S. nipponica, was abundant, suggesting that bark-stripping was not due to food shortages. The nutritive value of bark was lower than that of S. nipponica, which had high crude protein and hemicellulose contents in summer but an inadequate mineral balance in summer. Sika deer seems to eat bark either to offset the too rich summer forage and/or to attain a proper mineral balance in summer.

Introduction Mt. Ohdaigahara in central Japan is a major habitat for sika deer. It contains the southernmost forest of Hondo spruce (Picea jezoensis var. hondoensis) and the largest forest of Siebold’s beech (Fagus crenata) in western Japan. The recent increase in the deer population has caused dieback of overstory trees due to bark-stripping, resulting in decline of the forest (Akashi and Nakashizuka 1999; Yokoyama et al. 2001). Bark-stripping affects 37 tree species in 29 genera with coniferous and some broadleaf species suffering most (Shibata et al. 1984; Sekine and Sato 1992; Akashi and Nakashizuka 1999; Ando et al. 2003). This suggests that sika deer in Mt. Ohdaigahara exhibit species preference for bark-stripping. Gill (1992) demonstrated the bark-stripping preference of red deer (Cervus elaphus) in Europe for Norway spruce (Picea abies), lodgepole pine (Pinus contorta), and ash (Fraxinus spp.) but not for Sitka spruce (Picea sitchensis). Bark-stripping by sika deer in northern Japan is frequent in winter and spring due to food shortage (Kaji et al. 1984; Oi and Itoya 1997; Ueda et al. 2002). However, analysis of deer feces on Mt. Ohdaigahara shows that bark-stripping occurs frequently in summer (Yokoyama et al. 1996). D. R. McCullough et al. (eds.), Sika Deer: Biology and Management of Native and Introduced Populations, DOI: 10.1007/978-4-431-09429-6_15, © Springer 2009

207

208

M. Ando, E. Shibata

To conserve the primeval forest of Mt. Ohdaigahara, we must evaluate the extent of the damage, clarify which tree species are selectively debarked and when, and determine why this behavior occurs. This chapter mainly describes the bark-stripping preference of sika deer and its seasonality on Mt. Ohdaigahara. Previous reports describe damage caused by deer to primeval forest on Mt. Ohdaigahara (Shibata et al. 1984; Akashi and Nakashizuka 1999; Yokoyama et al. 2001; Ando et al. 2003), deer population density (Maeji et al. 1999), the estimated home range (Maeji et al. 2000; Yajima et al. 2002), the food habits based on fecal analysis (Yokoyama et al. 1996), the characteristics of dwarf bamboo (Sasa nipponica) grassland used as summer forage (Yokoyama and Shibata 1998b), the effect of deer grazing on the biomass and morphology of S. nipponica (Yokoyama and Shibata 1998a), and deer bark-stripping preferences (Ando et al. 2003, 2004).

Study Site Mt. Ohdaigahara is in southeast Nara Prefecture on the Kii Peninsula in central Japan (Fig. 15.1) within Yoshino Kumano National Park. Elevations vary from 1,400 m in the west to 1,600 m in the east with gentle slopes. The climate is characterized by cool

Fig. 15.1 Location of Mt. Ohdaigahara.

15 Bark-Stripping Preference of Sika Deer

209

Fig. 15.2 Two views of open S. nipponica grassland on eastern area of Mt. Ohdaigahara, showing deadfalls from typhoon damage and lack of sapling regrowth due to deer browsing. (Original data: true color, 1,600 × 1,200 pixels.)

temperatures and heavy rainfall; the mean annual temperature is 5.7 °C and the mean annual precipitation is >4,000 mm (Doei et al. 1989). The vegetation of the primeval forest can be divided into two major types (Doei et al. 1989): the eastern area is covered with coniferous forest, dominated by Hondo spruce (Picea jezoensis var. hondensis) and Nikko fir (Abies homolepis) while the western area is mixed forest, dominated by Siebold’s beech (Fagus crenata) and Nikko fir. Dwarf bamboo (S. nipponica) covers most of the forest floor in the eastern area. Fecal analysis shows that this species is the primary forage of sika deer (Yokoyama et al. 1996). Since S. nipponica is protein-rich (Yokoyama and Shibata 1998b) and can tolerate heavy grazing (Yokoyama and Shibata 1998a), deer tend to concentrate in the open S. nipponica grasslands on the eastern area of Mt. Ohdaigahara (Fig. 15.2) (Yokoyama and Shibata 1998b; Yajima et al. 2002). Ando et al. (2006) demonstrated that the open S. nipponica grasslands on Mt. Ohdaigahara originated from large gaps in the forest created by a huge typhoon. Akashi and Nakashizuka (1999) and Yokoyama et al. (2001) pointed out that dieback of canopy trees due to barkstripping and disappearance of seedlings through deer browsing accelerates the creation of gaps in primeval forest. These gaps expanded year by year probably due to dieback of surrounding canopy trees caused by bark-stripping by deer. Moreover, heavy grazing by deer causes a decrease in S. nipponica biomass coupled with dwarfing and densification (Yokoyama and Shibata 1998a).

Deer Density The density of sika deer on Mt. Ohdaigahara was investigated by the block count method 11 times from 1982 to 1997 (Table 15.1). The density increased gradually from about 20 deer/km2 in 1982 (Fukushima et al. 1984) to 40 deer/km2 in 1993 (Koizumi et al. 1994). The average density of deer in spring of 1997 was about 28

210

M. Ando, E. Shibata

Table 15.1 Deer densities on Mt. Ohdaigahara from 1982 to 1997. Year (month)

Census area (ha)

Observed Deer density (deer/km2) Total number of Eastern areaa Western areaa area deer

1982 (Oct)

459.5

102

–

–

22.2

1983 (Apr)

242.0

29

–

–

12.8

1983 (Aug) 494.4

130

–

–

26.3

1989 (Apr)

122

–

–

25.1

1989 (Nov) 485.5

84

–

–

17.3

1993 (May) 486.3

192

–

–

39.5

1993 (Oct)

259.2

79

–

–

30.5

1996 (July) 566.9

99

81

18

17.5

1996 (Oct)

566.9

175

119

56

30.9

1997 (Apr)

566.9

159

98

61.0

28.0

1997 (July) 566.9

129

93

36

22.8

a

485.5

Study Fukushima et al. (1984) Fukushima et al. (1984) Fukushima et al. (1984) Maeda et al. (1989) Koizumi et al. (1994) Yokoyama et al. (1995) Yokoyama et al. (1995) Maeji et al. (1999) Maeji et al. (1999) Maeji et al. (1999) Maeji et al. (1999)

Eastern and western study area are 258.3 and 308.6 ha, respectively.

deer/km2 and the population has possibly stabilized (Maeji et al. 1999). The density of deer in the eastern area is higher than in the western area (Maeji et al. 1999). There is a positive relationship between deer density and bark-stripping, so there is more bark-stripping damage in the eastern than western area (Sekine and Sato 1992; Ando et al. 2003).

Deer Damage More than half the coniferous trees, including Hondo spruce, Nikko fir, northern Japanese hemlock (Tsuga diversifolia) and Hinoki cypress (Chamaecyparis obtusa) have been debarked by sika deer (Yokoyama et al. 2001). Completely bark-stripped larger overstory conifers were killed (Yokoyama et al. 2001); 75.9% of Hondo spruce trees have suffered debarking with resultant mortality. No bark-stripping was found on trunks of Siebold’s beech (Akashi and Nakashizuka 1999; Ando et al. 2004). To conserve the primeval forest, especially the Hondo spruce, the Ministry of

15 Bark-Stripping Preference of Sika Deer

211

the Environment began a forest protection program in 1986. From 1991 to 2001, the trunks of about 22,000 trees were protected against bark-stripping by metal mesh, and areas with remaining Hondo spruce forest were enclosed by a permanent deerproof fence. Since 2002, the sika deer population has been culled to achieve a density of 10 deer/km2 by 2006 (Ministry of the Environment of Japan 2001).

Bark-Stripping Preference on Mt. Ohdaigahara Sika deer on Mt. Ohdaigahara show bark-stripping preferences (Shibata et al. 1984; Sekine and Sato 1992; Akashi and Nakashizuka 1999; Ando et al. 2003; Ando et al. 2004), stripping Hondo spruce, Nikko fir, Clethra barbinervis (sweet pepperbush) and Fraxinus lanuginosa var. serrata, (aodamo) but not Siebold’s beech, Acer shirasawanum (golden maple) or Pourthiaea villosa var. laevis (Oriental photinia (Rosaceae) (Table 15.2). Ando et al. (2003) listed the bark-stripping preference for 25 tree species using Ivlev’s selection index (Ivlev 1965), showing that the most preferred species was C. barbinervis followed by Ilex geniculata (holly). Barkstripping preference is not related to bark chemical content, such as crude protein, fiber (hemicellulose, cellulose, and lignin) or minerals (Ca, Mg, Na, and K), suggesting that the analyzed contents are not main factors in bark-stripping preference (Ando et al. 2003). Table 15.2 Bark-stripping of common trees on Mt. Ohdaigahara.

Species Coniferous

Broadleaved

Akashi and Shibata Sekine and Nakashizuka Ando et al. et al. (1984) Sato (1992) (1999) (2004) ++

++

+

++

+++

+++

NF

+++

+

++

–

+

Clethra barbinervis

+++

+

+

+++

Pourthiaea villosa var. laevis Fraxinus lanuginosa f. serrata Symplocos coreana Acer shirasawanum

–

NF

–

–

+++

++

+

++

NF –

NF +

– –

– –

Fagus crenata (Siebold’s beech)

NF

–

–

–

Abies homolepis (Nikko fir) Picea jezoensis var. hondoensis (Hondo spruce) Chamaecyparis obtusa (Hinoki cypress)

+++ = 75–100% of observed trees barked; ++ = 50–75%; + = 25–50%; – = 0–25%. NF = Not found.

212

M. Ando, E. Shibata

Gill (1992) pointed out that Norway spruce, lodgepole pine, and ash are debarked in every red deer habitat and also noted that silver fir (Abies alba) and rowan (Sorbus spp.) show different susceptibilities. Sika deer prefer Nikko fir and C. barbinervis at Sutama in Yamanashi Prefecture (Nagaike and Hayashi 2003) and at Nikko in Tochigi Prefecture (Kanzaki et al. 1998). Although Hondo spruce is preferred on Mt. Ohdaigahara, Yezo spruce (P. jezoensis) is hardly debarked at all on the Shiretoko Peninsula in Hokkaido Prefecture (Sakabe et al. 1998). Frequent debarking of F. lanuginosa var. serrata was observed on Mt. Ohdaigahara and the Shiretoko Peninsula, but not at Sutama. These comparisons show that the barkstripping preference of sika deer might be regional, as in red deer. In primeval forest, selective bark-stripping by deer (and close relatives) may change the forest composition (Miquelle and Van Ballenberghe 1989; Khan et al. 1994; Yokoyama et al. 2001), because deer strip some tree species selectively and the tolerance of each species is different. Those species that are debarked selectively and are not tolerant will decline and the species that are avoided and/or are tolerant will be able to grow more densely. Therefore, understanding deer barkstripping preferences in each habitat is important for evaluating forest composition and forest conservation concerns.

Seasonality of Bark-Stripping Occurrence on Mt. Ohdaigahara Ando et al. (2003) measured the area of bark wounds in each season and demonstrated that bark-stripping peaks in summer (mid-July to mid-September) (Fig. 15.3), concurring with the results of fecal analysis (Yokoyama et al. 1996). This suggests that there are reasons other than food shortages for bark-stripping in this area. Bark-stripping by red deer in western Scotland on Sitka spruce is most frequent in winter (Welch et al. 1987); bark-stripping on Scots pine (Pinus sylvestris) by moose (Alces alces) in south-central Sweden peaks in April and May (Faber 1996). In Japan, Ueda et al. (2002) demonstrated the bark-stripping peaks in January on Mt. Takahara in Tochigi Prefecture when food is at its shortest; Oi and Itoya (1997) also showed that food availability is lowest when bark-stripping peaks in May in Iwate Prefecture. However, on Mt. Ohdaigahara, Yokoyama and Shibata (1998a) showed that the biomass of S. nipponica peaks in August when bark-stripping also peaks (Fig. 15.3). Excluding food shortage, McIntyre (1972) suggested some hypotheses explaining why deer eat bark: (1) it contains important trace elements, vitamins, alkaloids and tannins, (2) it contains lignin, an indigestible fiber required for proper rumen function, and (3) it may have higher energy and mineral contents than other available forage. Ando et al. (2004) analyzed the chemical content of S. nipponica leaves and the bark of Hondo spruce, Nikko fir, and C. barbinervis, showing that: (1) the crude protein and hemicellulose contents of S. nipponica are always higher than that of bark and peaks in summer; (2) the lignin content of bark is always higher than that of S. nipponica, indicating that bark contains more indigestible roughage;

15 Bark-Stripping Preference of Sika Deer

213 Coniferous Broadleaf

1.0

1999

0.5

0

Area of bark stripped (m2/month)

1.0

2000

0.5

0 1.0

2001

0.5

0 1.0

2002

0.5

0 Spring

Fall Summer

Winter

Fig. 15.3 Seasonal changes in the monthly averaged area of trunk and butt swelling bark stripped by sika deer between July 1999 and July 2002 along six transects (2,400 m2 in total) on Mt. Ohdaigahara, central Japan. (Spring = mid-May to mid-July; Summer = mid-July to midSeptember; Fall = mid-September to mid-November; and Winter = mid-November to mid-May.) * Outside of study period.

(3) the calcium content of bark is always higher than that of S. nipponica; and (4) the K/(Ca+Mg) ratio of S. nipponica always exceeds 2.2 while that of bark is always less than 2.2. The fiber requirement of sika deer provides a plausible hypothesis for the occurrence of bark-stripping. Faber (1996) presumed that moose eat pine bark especially in April and March as roughage for proper rumen function to offset the high

214

M. Ando, E. Shibata

moisture and low fiber content of spring browse. Sika deer on Mt. Ohdaigahara may eat bark for similar reasons, because the crude protein and hemicellulose contents of S. nipponica peak in summer due to the new flush of leaves after June (Yokoyama and Shibata 1998b). A poor mineral balance in S. nipponica might explain bark-stripping on Mt. Ohdaigahara. Kemp and t’Hart (1957) reported that a high K/(Ca+Mg) ratio exceeding 2.2 tends to cause grass tetany in milking cows, which are ruminants like deer. The K/(Ca+Mg) ratio of S. nipponica peaks and exceeds 2.2 in summer, suggesting that the mineral balance of S. nipponica is inadequate in summer. In addition, summer is when males grow antlers and females nurse fawns, both requiring high calcium intake. Perhaps sika deer on Mt. Ohdaigahara feed on calcium-rich bark to supplement minerals in their forage. To clarify the detailed reason(s) for summer bark-stripping by sika deer, we must further investigate their nutritional physiology, especially rumen functions and mineral pathways.

Conclusion Since tree bark provides poor grade food with low crude protein and high lignin levels, bark-stripping by sika deer is very interesting behavior. The intentional ingestion of poor nutritional food such as bark to maintain proper rumen function suggests additional factors for food selection by sika deer as a ruminant. Moreover, bark-stripping selectivity and seasonality show regional differences, suggesting that deer eat bark for different reasons. Bark-stripping in each habitat needs further investigation to clarify regional difference in food selection by sika deer.

Literature Cited Akashi, N., and T. Nakashizuka. 1999. Effect of bark-stripping by sika deer (Cervus nippon) on population dynamics of a mixed forest in Japan. Forest Ecology and Management 113:83–89. Ando, M., H. Yokota, and E. Shibata. 2003. Bark stripping preference of sika deer, Cervus nippon, in terms of bark chemical contents. Forest Ecology and Management 177:323–331. Ando, M., H. Yokota, and E. Shibata. 2004. Why do sika deer, Cervus nippon, debark trees in summer on Mt. Ohdaigahara, central Japan? Mammal Study 29:73–83. Ando, M., A. Itaya, S. Yamamoto, and E. Shibata. 2006. Expansion of dwarf–bamboo, Sasa nipponica, grassland under feeding pressure of sika deer, Cervus nippon, on subalpine coniferous forest in central Japan. Journal of Forest Research 11:51–55. Doei, T., H. Doei, and T. Suganuma. 1989. Flora of the summit area of Mt. Ohdaigahara in the Kii Peninsula, Central Honshu, Japan, I. Nanki Biology 31:31–35. (In Japanese.) Faber, W. E. 1996. Bark stripping by moose on young Pinus sylvestris in south-central Sweden. Scandinavian Journal of Forest Research 11:300–306. Fukushima, N., S. Miura, Y. Kikuchi, N. Maruyama, and H. Tanaka. 1984. Effects of density of sika deer (Cervus nippon) on floral change and its management in Mt. Ohdaigahara. Pages 29–37 in Nara Natural Environment Society. Nara, Japan. (In Japanese.)

15 Bark-Stripping Preference of Sika Deer

215

Gill, R. M. A. 1992. A review of damage by mammals in north temperate forests, 1. Deer. Forestry 65:145–169. Ivlev, V. S. 1965. Experimental ecology of the feeding of fishes. Kodama, Y. and T. Yoshiwara, translators. Tatarasyobo, Yonago, Japan. (In Japanese.) Kaji, K., N. Ohtaishi, and T. Koizumi. 1984. Population growth and its effect upon the forest used by sika deer on Nakanoshima Island in Lake Toya, Hokkaido. Acta Zoologica Fennica 172:203–205. Kanzaki, N., N. Maruyama, M. Koganezawa, and M. Taniguchi. 1998. Bark stripping by sika deer in Nikko, Tochigi Prefecture. Wildlife Conservation 3:107–117. (In Japanese with English abstract.) Kemp, A., and M. L. t’Hart. 1957. Grass tetany in grazing milking cows. Netherlands Journal of Agricultural Science 5:4–17. Khan, J. A., W. A. Rodgers, A. J. T. Johnsingh, and P. K. Mathur. 1994. Tree and shrub mortality and debarking by sambar Cervus unicolor (Kerr) in Gir after a drought in Gujarat, India. Biological Conservation 68:149–154. Koizumi T., E. Shibata, and K. Tabata. 1994. Condition of sika deer population on Mt. Ohdaigahara. Pages 35–42 in Report on conservation of spruce forest on Mt. Ohdaigahara (1989–1993). Environmental Agency, Tokyo, Japan. (In Japanese.) Maeda, M., T. Koizumi, S. Miura, E. Shibata, and E. Kitahara. 1989. Report on sika deer population on Mt. Ohdaigahara. Pages 41–60 in Report on conservation of spruce forest on Mt. Ohdaigahara (1986–1989). Environmental Agency, Tokyo, Japan. (In Japanese.) Maeji, I., S. Yokoyama, and E. Shibata. 1999. Population density and range use of sika deer, Cervus nippon, on Mt. Ohdaigahara, central Japan. Journal of Forest Research 4:235–239. Maeji, I., T. Kurosaki, S. Yokoyama, and E. Shibata. 2000. Home range of sika deer (Cervus nippon) on Mt. Ohdaigahara, central Japan. Nagoya University Forest Science 19:1–10. (In Japanese with English abstract.) McIntyre, E. B. 1972. Barkstripping – A natural phenomenon. Scottish Forestry 26:43–50. Ministry of the Environment of Japan. 2001. Management of deer population on Mt. Ohdaigahara. Tokyo, Japan. Miquelle, D. G., and V. Van Ballenberghe. 1989. Impact of bark stripping by moose on aspenspruce communities. Journal of Wildlife Management 53:577–586. Nagaike, T., and A. Hayashi. 2003. Bark-stripping by sika deer (Cervus nippon) in Larix kaempferi plantations in central Japan. Forest Ecology and Management 175:563–572. Oi, T., and Y. Itoya. 1997. Nutritional evaluation of Cryptomeria japonica damaged by sika deer (Cervus nippon). Pages 284–291. Proceedings of the 2nd Regional Workshop (Forest Protection in Northeast Asia) of IUFRO. Seoul, Korea. Sakabe, T., T. Yabe, T. Yajima, M. Shibuya, and K. Takahashi. 1998. Tree damage by sika deer in a wintering area in the Iwaobetsu district on the Shiretoko Peninsula. Research Bulletin of Hokkaido University Forests 55:113–122. (In Japanese with English abstract.) Sekine, T., and H. Sato. 1992. Tree barking by sika deer, Cervus nippon Temminck, on Mt. Ohdaigahara. Japanese Journal of Ecology 42:241–248. (In Japanese with English abstract.) Shibata, E., N. Katayama, and H. Kataoka. 1984. Damage to coniferous forests by sika deer (Cervus nippon) on Mt. Ohdaigahara. Nara Botany 7:1–6. (In Japanese.) Ueda, H., S. Takatsuki, and Y. Takahashi. 2002. Bark stripping of hinoki cypress by sika deer in relation to snow cover and food availability on Mt. Takahara, central Japan. Ecological Research 17:545–551. Welch, D., B. W. Staines, D. Scott, and D. C. Catt. 1987. Bark stripping damage by red deer in a Sitka spruce forest in Western Scotland. Forestry 60:249–262. Yajima, K., Y. Yamamoto, I. Maeji, T. Kurosaki, T. Yokota, H. Sato, and E. Shibata. 2002. Seasonal change in home range of female sika deer (Cervus nippon) on Mt. Ohdaigahara, central Japan. Nagoya University Forest Science 21:1–7. (In Japanese with English abstract.) Yokoyama, S., and E. Shibata. 1998a. The effects of sika-deer browsing on the biomass and morphology of a dwarf bamboo, Sasa nipponica, in Mt. Ohdaigahara, central Japan. Forest Ecology and Management 103:49–56.

216

M. Ando, E. Shibata

Yokoyama, S., and E. Shibata. 1998b. The characteristics of Sasa nipponica grassland as a summer forage resource for sika deer on Mt. Ohdaigahara, central Japan. Ecological Research 13:193–198. Yokoyama, S., T. Koizumi, and E. Shibata. 1995. Population density and distribution of sika deer, Cervus nippon, in Mt Ohdaigahara. Pages 145–146 in Transaction of the 43rd annual meeting of the Chubu branch of the Japanese Forestry Society. (In Japanese.) Yokoyama, S., T. Koizumi, and E. Shibata. 1996. Food habit of sika deer as assessed by fecal analysis in Mt. Ohdaigahara, central Japan. Journal of Forest Research 1:161–164. Yokoyama, S., I. Maeji, T. Ueda, M. Ando, and E. Shibata. 2001. Impact of bark stripping by sika deer, Cervus nippon, on subalpine coniferous forests in central Japan. Forest Ecology and Management 140:93–99.

Chapter 16

North-South Variations in Sika Deer Ecology as a Forest-Dwelling Cervid Seiki Takatsuki

Abstract Sika deer on the Japanese Archipelago inhabit a variety of habitats with different climatic conditions. I review the significance of this north-south variation for the ecology of sika deer. Because of warm and humid summers, plant growth is vigorous in Japan, resulting in a prevalence of forests. The keen hearing and agility of sika deer and their dependence on forest plants as foods suggest that the sika deer is a cervid adapted to forest environments. The northern deciduous forests are characterized by marked seasonal changes in temperature and consequent plant growth. They are densely covered with Sasa species (dwarf bamboos). The abundant Sasa species are evergreen and are particularly important foods for overwintering sika deer. Snow hinders deer locomotion and reduces food availability. Seasonal events such as rut and parturition are synchronized within short periods. Larger body size and greater fat deposits are advantageous to overwintering. The sika deer show seasonal elevational movements depending on snow. In contrast, southern evergreen forests are often dark, and shrubs and forbs in the understories are less abundant and often unpalatable. The southern sika deer are browsers and eat fruits and seeds in autumn and winter. They are sedentary and live in small groups. The sika deer demonstrates how an ungulate can change its ecology in different environmental conditions.

Introduction Sika deer (Cervus nippon) is the only cervid species living in the Japanese Archipelago. Recent genetic studies have shown that the sika deer is divided into only two groups (Tamate and Tsuchiya 1995; Nagata et al. 1995; Tamate et al. 1998; Goodman et al. 2001; Nagata chapter 3; Tamate chapter 4). In spite of the simple genetic division, because of high environmental diversity of the archipelago, considerable size variations are recognized among the local sika populations. Furthermore, sika ecology is fairly different in the northern and southern areas of Japan. For example, food habits are very different between the northern and southern populations (Campos-Arceiz and Takatsuki 2005; Takatsuki chapter 17). These D. R. McCullough et al. (eds.), Sika Deer: Biology and Management of Native and Introduced Populations, DOI: 10.1007/978-4-431-09429-6_16, © Springer 2009

217

218

S. Takatsuki

variations can be attributed to environmental differences. A substantially important factor is temperature which affects the composition and growth patterns of the flora, which would directly or indirectly affect deer ecology. I consider the significance of these environmental differences on variations among sika deer.

North and South Biodiversity of the Japanese Archipelago The Japanese Archipelago is situated in middle latitudes ranging between 24° N and 46° N, which extends from the subtropical zone to the boreal zone. The mean temperature of Naha in Okinawa is 22.4° C and that of Wakkanai in Hokkaido is 6.4° C. Within this range, four big islands and numerous small islands are distributed. Throughout geological history, they were either bridged or isolated in various ways. In addition, they received “waves” of plants and animals from the Asian continent (Dobson 1994). Accordingly, the biodiversity of the archipelago is high and the endemism is also high. By comparison, species numbers of higher plants for Japan and for Great Britain are 7,000 and 1,000. Similarly, mammal species are 200 and 46, birds are 700 and 210, reptiles are 97 and six, amphibians are 64 and six, and fresh water fishes are 300 and 36. The topography of Japan is steep and flat lands are limited. Summers are quite warm and rainy. These climatic conditions promise luxurious plant growth, and the land is covered with forests. The northern half of the archipelago is covered with deciduous broad-leaved forests, while the southern one supports evergreen broadleaved forests. Some parts of Hokkaido, the northern island, and higher elevations of Honshu, the main island, are covered with coniferous forests. The sika deer ranges over almost all areas of the archipelago. Because the origin of the southern-most population, Kerama sika deer in Okinawa, was artificial, this should be excluded from the natural distribution. Thus, the distribution ranges from Yaku Island (30° N, mean annual temperature 19° C) to Hokkaido (46° N, mean annual temperature 6.6° C).

Stability of Environment In terms of food supply, the north is much more seasonably variable than the south. In the southern evergreen forests, food abundance in summer is not very great, but in winter is not very limited. In contrast, in the northern deciduous forests, sika deer are afforded superabundant foods in summer while almost all the plants wither and foods become poor in winter. It is therefore adaptive for the northern deer to intake much food in summer and deposit fat to overwinter. In fact, it is known that northern

16 North-South Variations in Sika Deer Ecology as a Forest-Dwelling Cervid

219

cervids increase food intake in summer (red deer, Cervus elaphus, Mitchell et al. 1977; Simpson et al. 1978; Loudon et al. 1989; wapiti, Cervus elaphus, Robbins et al. 1981; Geist 1982; white-tailed deer, Odocoileus virginianus, Holter et al. 1977; Warren et al. 1981; Wheaton and Brown 1983; roe deer, Capreolus capreolus, Drozdz and Osiecki 1973; Drozdz et al. 1975; reindeer, Rangifer tarandus, McEwan and Whitehead 1970; White and Fancy 1986; moose, Alces alces, Verme 1970; Gasaway and Coady 1974; Pollack 1974) and deposit fat in autumn prior to winter (red deer, Batcheler and Clarke 1970; Mitchell et al. 1976; wapiti, Flook 1970; white-tailed deer, Johns et al. 1984; moose, Fong 1981; reindeer, McEwan and Whitehead 1970; Dauphine 1976; White and Fancy 1986). Because of marked reduction of food supply in winter, the northern deer face food shortage. The sika deer on Nakanoshima Island in Hokkaido were observed to consume fallen leaves in winter (Miyaki et al. 1996). It is known that sika deer reduce intake even if they are afforded enough foods (Ikeda et al. 1998). Similar behavior was observed in red deer (Pollack 1974), Odocoileus spp. (Wood et al. 1962), moose (Heptner and Nasimovitch 1968; Renecker and Hudson 1985; Schwartz et al.1987), reindeer (McEwan and Whitehead 1970), and roe deer (Drozdz et al. 1975). It is also characteristic of the northern deciduous forests that the trees bear nuts, particularly acorns. Acorns are rich in starch and lipid and are good food for wildlife (Goodrum et al. 1971; Pekins and Mautz 1988; Feldhamer et al. 1989; Johnson et al. 1989; Wentworth et al. 1989; Kirkpatrick and Pekins 2002). It is reported that some cervids move to oak forests in autumn looking for acorns (red deer, Heptner and Nasimawitch 1968; Van de Veen 1979; wapiti, Murphy 1963). An important phenomenon is masting or great yearly variations of acorn production. Because acorns affect fat deposition of wildlife, masting is important for nutrition. Furthermore, snow, which shows great yearly variations, strongly affects food availability. In essence, food supply is unstable in the northern forests, and therefore elasticity or an opportunistic strategy seems to be more adaptive for the deer living there.

Differences in Forest Undergrowth Another difference between the northern and southern forests is that toxic plants are more abundant in the southern forests (Levin 1976). Plants containing toxic substances or secondary compounds which are not preferred by ungulates, such as Lauraceae, Symplocos spp. (sweetleaf), and Rutaceae, are more abundant in the southern forests. Woody plants contain more tannin and lignin which lower digestibility for herbivorous mammals (Robbins et al. 1987a, b, 1991, 1995; Hanley et al. 1992) than do herbaceous plants (Iason and Van Wieren 1997). Against these plant defenses, browsers are more capable of detoxification than grazers (Powell et al. 1974). For example, browsers possess protein in saliva to detoxify tannin, but grazers do not (Robbins et al. 1991). Because of shading, forest floors are generally poor in plant biomass. In Japanese forests, however, this is not always the case. The northern deciduous forests of Japan

220

S. Takatsuki

are often vegetated by dense dwarf bamboos (Sasa spp.) on the forest floor. The biomass is particularly great in the Sasa palmata and S. kurilensis communities (Oshima 1961). This is not only different from those in the Chinese mainland where dwarf bamboos do not grow in the areas north of the Jangzi River but also unique among the temperate deciduous forests of the world. In terms of foods for ungulates, these dwarf bamboos are unique. The larger Sasa species are distributed in snowy areas. These areas are not inhabited by the sika deer which is vulnerable to snow (Takatsuki 1992a), but instead by the Japanese serow (Capricornis crispus). It is known that the Japanese serow, a browsing bovid, does not preferably eat these dwarf bamboos (Ochiai 1999). On the contrary, Sasa nipponica, another Sasa species (see Fig. 26.2 in Takatsuki chapter 26) which is distributed in the snowless Pacific side, is a very important food for sika deer, particularly in winter (Takatsuki 1983, 1986; M. Yokoyama et al. 1996; S. Yokoyama et al. 1996). This dwarf bamboo meets all the criteria for a key forage plant: (1) abundance, (2) nutritional content, (3) stability in supply (being an evergreen), and (4) tolerance to grazing (Standing 1938). The leaves of S. nipponica are rich in protein (Ikeda and Takatsuki 1999), but they are coarse and less digestible than pasture grasses (Takatsuki, S. and S. Ikeda unpublished). This seems to be a reason that the rumino-reticulum volume of the northern sika deer is biggest among the cervids (Fig. 16.1, Hofmann 1985; Takatsuki 1988).

Marked Seasonal Changes The northern habitats are characterized by marked seasonal changes. Synchronized phenology of plants results in the synchronized deer phenology. For example, rut and parturition concentrate in short periods and, consequently, battles among males are serious. Clear seasonality accelerates the rapid growth of offspring in summer before the onset of winter when they cannot grow. Food composition of the northern deer shows marked seasonal variations (Takatsuki 1983, 1986; M. Yokoyama et al. 1995), while that of the southern sika deer is comparatively stable (Asada and Ochiai 1996; Ikeda 2002). Snow affects various aspects of deer life. For example, snow affects group formation. Ungulates generally form larger groups in open environments. Deciduous trees in the northern forests drop their leaves, and the trees without leaves afford effects similar to open habitats. Openness is more prominent in snowy environments. Group sizes of sika deer at Mt. Goyo in northern Japan showed marked seasonal change with group size increasing in winter (Fig. 16.2). Such seasonal changes would be different from those in the southern evergreen forests, such as group sizes of sika deer on a small island (Iyo-kashima Island, southern Shikoku; Takatsuki 1982).

Concentrate selectors

Intermediate (opportunistic adaptable selective)

Roughage feeders Selective NonSelective

Musk deer Pere David’s deer

Chinese water deer

Muntjak Caribou

Sika deer

Roe deer Domestic goat

Domestic sheep

White-tailed deer Eurasian red deer Pudu

Fallow deer Maned deer Mule deer

Sambar deer Axis deer

Mazama

American Elk

Domestic cattle Moose

Barasingha deer

Feeding frequency

Fig. 16.1 Classification of Cervidae on the basis of morphophysiological feeding type (Hofmann 1985). Some bovids are also shown for comparison. Note the position of sika deer. By courtesy of Royal Society of New Zealand.

Fig. 16.2 Seasonal changes in group size of sika deer in the foothills of Mt. Goyo, northern Japan (from Takatsuki 1992b).

222

S. Takatsuki

Snow inevitably increases winter mortality of sika deer, particularly that of fawns, which would require increased natality to compensate in order to maintain the population. This consequently results in high turnover of the population. Snowfall differs among years; deer mortality is less in more open winters and higher in snowy winters. Effects of snow on ungulate populations have been demonstrated in many species (wapiti, mule deer, bighorn sheep Ovis ovis, Picton 1984; moose and white-tailed deer, Mech et al. 1987). In order to survive in deep snow, it is advantageous to have a bigger body, longer legs, and to deposit more pre-winter fat (Geist 1983). Although home range use of sika deer is not well-studied, some case studies have shown the habitat use patterns of the northern sika deer. For example, sika deer in Nemuro (Kaji 1981) and in Shiretoko (Yabe et al. 1990) in Hokkaido ranged widely in deciduous forests from spring to autumn, but were concentrated in a small area of coniferous forest in winter. Such shifts in home range use become traditional migrations with high predictability. Elevational movements are more common in Japan because of the steep topography. The pioneer study of Maruyama (1981) firstly demonstrated the updown movements of sika deer in Omote-Nikko, central Honshu. Similar up-down movements were also found in Mt. Goyo in northern Honshu (Takatsuki et al. 2000). In eastern Hokkaido, however, sika deer not always showed up-down but sometimes long distance horizontal movements (Uno and Kaji 2000; Igota et al. 2004). In summary, sika deer living in the Japanese Archipelago have been affected by its steep topography, warm and humid summer climate, and consequent rich vegetation. Throughout their long history, they have diversified into two types: one in the stable environment of the warm temperate zone of southeastern Japan and the other in the unstable environment with contrasting summer and winter of the cool temperate zone of northeastern Japan.

Forest and Grassland Prevalence of Forests in the Japanese Archipelago The Japanese Archipelago is primarily covered with forests. Given the warm and humid summers, vegetation succession proceeds rapidly. Bare lands are immediately invaded by pioneer plants like Miscanthus sinensis (silver grass), a tall grass. It is thereafter replaced by weeds such as Oenothera biennis (common evening primrose) and Erigeron canadensis (fleabane) and by pasture grasses such as Poa pratensis (bluegrass) and Agrostis alba (bentgrass). After several years, pioneer shrubs such as Rhus javanica var. roxburghii (sumac), Mallotus japonicus (Euphorbiaceae), and Aralia elata (Japanese angelica) invade to form dense scrubs. If the place is not a bare mineral soil but is a logged-over area with rich organic soil and numerous seed banks, the succession proceeds more rapidly. In such a case, logged tree stumps also begin sprouting. Thus, Rubus spp., Stachyurus praecox

16 North-South Variations in Sika Deer Ecology as a Forest-Dwelling Cervid

223

(spiketail), Viburnum spp., Prunus spp., Quercus spp. and Carpinus spp. (hornbeam) grow vigorously to become a coppice forest within 20 years. In such places, grasslands and scrublands disappear as if they are “swallowed” by forests. In this warm and humid land, grasslands are maintained only by artificially impeding vegetational succession, such as by prescribed burning or mowing. Vegetation characteristics shape the evolution of ungulates adapted to different types of habitats (Geist 1983) as outlined in Fig. 16.3. In terms of deer habitat, a forest floor provides fewer food plants because of shading, while grasslands are rich in plant biomass because of good light conditions. Plant species composition is also different between forests and grasslands. Forest plants are generally slower in photosynthesis than grassland plants (Bjorkman 1968). This relates to succession, and late successional plants are slower in photosynthesis than pioneer plants (Bazzaz 1979). Grasses or Gramineae (Poaceae) species are dominant in grasslands. Grasses have high biomass and afford rich foods for ungulates. When a gap is formed in a forest, particular species of plants such as Miscanthus sinensis, Rubus spp., Aralia elata, Rhus javanica var. roxburghii and Clerodendrum trichotomum (glorybower) invade. It is known in the tropical forests that gap invaders are more readily eaten by herbivores than late successional species (Coley 1983). This is the case in Japan, and we often observe that plants in gaps are more frequently clipped by sika deer. Plant biomass is, of course, greater in forests than in grasslands, but the majority of the biomass is composed of supporting trunks and branches that are less nutritious and inedible for ungulates. The leaves are positioned in the canopy, and therefore ground dwellers cannot feed on them during the growing season. Since trees and shrubs have bigger woody supporting structures, they can allocate their production into leaves. In contrast, herbaceous plants have to allocate

Fig. 16.3 Body color patterns of wapiti (Cervus elephus) and sika deer (C. nippon).

224

S. Takatsuki

their production into not only leaves but also supporting stems as well. Grasses often have tough leaves and culms. Grass leaves are long and have tough central veins to strengthen the structure. Culms are hollow which structure is quite strong in spite of limited amounts of materials. The high content of cellulose in the stems and leaves of grasses also makes them tough. Because plants containing much cellulose are less digestible (Van Soest 1982; Robbins 1983), ruminants have developed the rumination system in order to digest these grasses. Grasses, in turn, are nontoxic, that is, they are low in secondary compounds which defend against herbivory. Among as many as 8,000 grass species, only 21 species or 0.3% contain alkaloids (Culvenor 1973). This is in contrast to dicotyledonous plants like Labiatae and Ranunculaceae that contain considerable toxic and/or odiferous substances. Contrary to grasses, long-lived leaves of forest plants are slower in photosynthesis, more metabolically costly to produce, and more vulnerable to herbivory (Mooney and Gulmon 1982). That is, forest plants grow more slowly and invest more for defense than grassland plants. For ungulates, therefore, grasses are safe and abundant food plants, though they are not ideally nutritious. Another characteristic of grasses is that they are more tolerant of grazing or defoliation. Some forbs and woody plants are more vulnerable to defoliation and die because of clipping, while grasses can recover. This is attributed to the morphology of grasses. A young grass plant has a culm covered with sheaths appearing from nodes. Younger sheaths are situated inside older ones and develop by breaking through the outer sheaths. Older leaves are, therefore, situated at lower nodes. Even if these older leaves are removed, younger leaves can survive and extend if only shoot apices are alive. Since shoot apices are situated at the lower parts of grasses, they can often escape grazing. Typical examples are found in lawn grasses that are maintained by repeated mowing. In contrast, forbs have growing points at various height of stems, and removal of them is hazardous and often fatal. Since woody plants allocate more products to supporting organs, they are more vulnerable to mowing and grazing. According to such mechanisms, grasses become relatively more abundant in Japanese grasslands that are maintained by mowing. In several cases when “grasslands” appear by gap formation, e.g., forest logging, or overabundance of sika deer on islands, sika deer concentrate into these grasslands and maintain them by gazing. Grazing is a kind of defoliation and therefore it works in favors of grasses and grasslands. Consequently, although the Japanese Archipelago is dominated by forests, grasslands appear only temporarily or in small scale due to disturbance. However, such grasslands afford good foraging grounds for ungulates and may be maintained by their grazing.

Forest and Grassland as Sika Deer Habitats Visibility is generally poor in forests and good in grasslands. Geist (1987) has pointed out that the auditory sense of ungulates is more important in forests while visual sense is more important in open lands. Males of sika deer, red deer, and

16 North-South Variations in Sika Deer Ecology as a Forest-Dwelling Cervid

225

wapiti bugle during the rut. Habitats of the three Cervus species become more open in this order, but comparison of their vocalization has not been done, though male sika deer respond keenly to other males’ bugling (Minami and Kawamichi 1992). Relating to the visual sense, body color is more contrasting in grassland ungulates. The wapiti as a grassland dweller has a dark front body and light hind body, while the sika deer as a forest dweller has less distinct color difference (Fig. 16.3). Because of high visibility, ungulates are more readily located by predators in grasslands than in forests. Ungulates form larger groups in grasslands than in forests. This is interpreted to mean that ungulates in larger groups are more likely to detect predators, and the group dynamics makes it more difficult for predators to make a kill (Hamilton 1971; Hirth 1977; Risenhoover and Bailey 1985). For grassland ungulates, the ability to run fast and long is quite important. Steppe ungulates like the pronghorn (Antilocapra americana) and the Mongolian gazelle (Procapra gutturosa) are excellent runners. Forests are safer habitats for species that escape predators by stealth. When a sika deer encounters people, it swiftly escapes. However, it is not always a good runner; when it runs, the body moves up and down. However, it can run even on steep slopes and through vegetation and forestfloor debris. It seems to be adapted to steep topography and dense vegetation where long escapes are not necessary but short distance dashes and concealment are more important. Lower visibility in forests also relates to the behavior of sika deer during rut. A rutting male sika deer tries to hold as many females as possible (Fig. 16.4). The numbers of females a male can guard is fewer in the forest and more in grasslands. In fact, Miura (1986) has found differences in rutting behavior of sika males in

Fig. 16.4 A territorial male of sika deer herding several females on Kinkazan Island. (Photo by S. Takatsuki.)

226

S. Takatsuki

different habitats. It seems more advantageous for a male on Kinkazan Island in northern Japan, which is vegetated more by grassland communities, to form a rutting territory since females are gregarious; while it is more effective for a male on Nozaki Island in southern Japan, which is densely covered by evergreen forests, not to form a rutting territory but to access individual females. Forests canopies function to moderate strong winds and snow fall. Microclimate fluctuates less in forests than in open habitats (Ozoga 1968). Storms in winter rob herbivorous mammals of body heat (Blaxter et al. 1963; Moen 1973). Deer often stay in forests to avoid storms (Staines 1974, 1976). Ungulates are vulnerable to deep snow since they sink into the snow because their small hooves (Gilbert et al. 1970; Gates and Hudson 1981; Pudd et al. 1983; Telfer and Kelsall 1984). Canopy trees interrupt much of the snow fall, and less snow accumulates in the forest floor. This is more marked in coniferous than deciduous forests. Thus, forests function as refuges for deer in winter.

Literature Cited Asada, M., and K. Ochiai. 1996. Food habits of sika deer on the Boso Peninsula, central Japan. Ecological Research 11:89–65. Batcheler, C. L., and C. M. H. Clarke. 1970. Note on kidney weights and the kidney fat index. New Zealand Journal of Science 13:663–668. Bazzaz, F. A. 1979. The physiological ecology of plant succession. Annual Review of Ecology and Systematics 10:351–371. Bjorkman, O. 1968. Carboxyldismulatase activity in shade-adapted and sun-adapted species of higher plants. Physiologica Plantarum 21:1–10. Blaxter, K. L, J. P. Joyce, and F. W. Wainmann. 1963. Effect of air velocity on the heat losses of sheep and cattle. Nature 197:1115–1116. Campos-Arceiz, A., and S. Takatsuki. 2005. Food habits of sika deer in the Shiranuka Hills, eastern Hokkaido - A northern example among the north-south variations of food habits in sika deer. Ecological Research 20:129–133. Coley, P. D. 1983. Herbivory and defensive characteristics of tree species in a lowland tropical forest. Ecological Monographs 53:209–233. Culvenor, C. C. 1973. Alkaloids. Pages 375–446 in G. W. Butler and R. W. Bailey, editors, Chemistry and biochemistry of herbage, Vol. 1. Academic, London, United Kingdom and New York, New York, USA. Dauphine, T. C., Jr. 1976. Biology of the Kaminuriak population of barren ground caribou. Part 4: Growth, reproduction and energy reserves. Canadian Wildlife Service Report Series 38, Ottawa, Canada. Dobson, M. 1994. Patterns of distribution in Japanese land mammals. Mammal Review 24:91–111. Drozdz, A., and A. Osiecki. 1973. Intake and digestibility of natural feeds by roe-deer. Acta Theriologica 18:81–91. Drozdz, A., J. Weiner, Z. Gebczynska,and M. Kvasinka. 1975. Some bioenergetic parameters of wild ruminants. Polish Ecological Studies 1:85–101. Feldhamer, G. A., T. P. Kilbane, and D. W. Sharp, 1989. Cumulative effect of winter on acorn yield and deer body weight. Journal of Wildlife Management 53:292–295. Flook, D. R. 1970. A study of sex differential in the survival of wapiti. Canadian Wildlife Service Report Series 11, Ottawa, Canada.

16 North-South Variations in Sika Deer Ecology as a Forest-Dwelling Cervid

227

Fong, D. W. 1981. Seasonal variation of marrow fat content from Newfoundland moose. Journal of Wildlife Management 45:545–548. Gasaway, W. C., and J. W. Coady. 1974. Review of energy requirements and rumen fermentation in moose and other ruminants. Canadian Naturalist 101:227–262. Gates, C. C., and R. J. Hudson. 1981. Weight dynamics of wapiti in the boreal forest. Acta Theriologica 26:407–418. Geist, V. 1982. Adaptive behavioral strategies. Pages 219–277 in J. W. Thomas and D. E. Toweill, editors, Elk of North America. Stackpole Books, Mechanicsburg, Pennsylvania, USA. Geist, V. 1983. On the evolution of ice age mammals and its significance to an understanding of speciations. Walton Distinguished Lecture Series, Mountain Lake Biology Station (ASB Bulletin) 30:109–133. Geist, V. 1987. On speciation in Ice Age mammals, with special reference to cervids and caprids. Canadian Journal of Zoology 65:1067–1084. Gilbert, P. F., O. C. Wallmo, and R. B. Gill. 1970. Effect of snow depth on mule deer in Middle Park, Colorado. Journal of Wildlife Management 34:15–23. Goodman, S.J., H. B. Tamate, R. Wilson, J. Nagata, S. Tatsuzawa, G.. Swanson, J. M. Pemberton, and D. R. McCullough. 2001. Bottlenecks, draft and differentiation: The population structure and demographic history of sika deer (Cervus nippon) in the Japanese archipelago. Molecular Ecology 10:1357–1370. Goodrum, P. D., V. H. Reid, and C. E. Boyd. 1971. Acorn yields, characteristics, and management criteria of oaks for wildlife. Journal of Wildlife Management 35:520–532. Hamilton, W. D. 1971. Geometry for the selfish herd. Journal of Theoretical Biology 31:295–311. Hanley, T. A., C. T. Robbins, A. E. Hagerman, and C. McArthur. 1992. Predicting digestible protein and digestible dry matter in tannin-containing forages consumed by ruminants. Ecology 73:537–541. Heptner, W. G., and A. A. Nasimovitch. 1968. Der Elch. Die Neue Brehm Bucherei, No. 386. A. Ziemseu Verlag, Wittenberg-Lutherstadt, Germany. Hirth, D. H. 1977. Social behavior of white-tailed deer in relation to habitat. Wildlife Monographs 53:1–55. Hofmann, R. R. 1985. Digestive physiology of deer – their morphophysiological specialization and adaptation. Pages 393–407 in P. F. Fennessy and K. R. Drew, editors, Biology of deer production. Royal Society of New Zealand Bulletin 22, Wellington, New Zealand. Holter, J. B., W. E. Urban, and H. H. Hayes. 1977. Nutrition of northern white-tailed deer throughout the year. Journal of Animal Science 45:365–376. Iason, G. R., and S. E. Van Wieren. 1997. Digestive and indigestive adaptations of mammalian herbivores to low-quality forage. Pages 337–369 in H. Olff, V. K. Brown, and R. H. Drent, editors, Herbivores: Between plants and predators. Blackwell Science, Oxford, United Kingdom. Igota, H., M. Sakuragi, H. Uno, K. Kajio, M. Kaneko, R. Akamatsu, and K. Maekawa. 2004. Seasonal migration patterns of female sika deer in eastern Hokkaido, Japan. Ecological Research 19:169–178. Ikeda, K. 2002. The food habits of sika deer in a coniferous plantation area, northern Kyushu. Journal of the Japanese Forestry Society 84:175–179. (In Japanese with English summary.) Ikeda, S., T. Takeda, and M. Murai. 1998. Deer farming. Pages 6–15 in A guide for sika deer farming. Ministry of Agriculture, Forestry and Fishery of Japan, and Tohoku Branch of the Ministry of Agriculture, Forestry and Fishery of Japan. (In Japanese.) Johns, P. E., M. H. Smith, and R. K. Cheser. 1984. Annual cycles of the kidney fat index in a southeastern white-tailed deer herd. Journal of Wildlife Management 43:969–973. Johnson, A. S., J. M. Wentworth, and P. E. Hale. 1989. Cumulative mast needs of forest wildlife. Pages 18–23 in C. E. McGee, editor, Proceedings, workshop on Southern Appalachian mast management. University of Tennessee, Knoxville, Tennessee, USA. Levin, D. A. 1976. Alkaloid-bearing plants: An ecogeographic perspective. American Naturalist 110:261–284.

228

S. Takatsuki

Loudon, A. S. I., J. A. Milne, J. D. Curlewis, and A. S. McNeilly. 1989. A comparison of the seasonal hormone changes and patterns of growth, voluntary food intake and reproduction in juvenile and adult red deer (Cervus elaphus) and Père David’s deer (Elaphurus davidianus) hinds. Journal of Endocrinology 122:733–745. Kaji, K. 1981. Range use of sika deer (Cervus nippon yesoensis Heude) in the Nemuroshibetsu District, Hokkaido. Journal of the Mammalogical Society of Japan 8:226–236. Kirkpatrick, R. L., and P. J. Pekins. 2002. Nutritional value of acorns for wildlife. Pages 173–181 in W. J. McShea and W. M. Healy, editors, Oak forest ecosystems: Ecology and management for wildlife. Johns Hopkins University Press, Baltimore, Maryland, USA. Maruyama, N. 1981. A study of the seasonal movements and aggregation patterns of sika deer. Bulletin of Faculty of Agriculture, Tokyo University of Agriculture and Technology 23:1–85. (In Japanese with English summary.) McEwan, E. H., and P. E. Whitehead. 1970. Seasonal change in the energy and nutritive intake in reindeer and caribou. Canadian Journal of Zoology 48:905–913. Mech, L. D., R. E. McRoberts, R. O. Peterson, and R. E. Page. 1987. Relationship of deer and moose populations to previous winters’ snow. Journal of Animal Ecology 56:615–627. Minami, M., and T. Kawamichi. 1992. Vocal repertoires and classification of the sika deer Cervus nippon. Journal of the Mammalogical Society of Japan 17:71–94. Mitchell, B., D. McCowan, and I. A. Nicholson. 1976. Annual cycles of body weight and composition in Scottish red deer, Cervus elaphus. Journal of Zoology, London 180:107–127. Mitchell, B., B. W. Staines, and D. Welch. 1977. Ecology of red deer. Institute of Terrestrial Ecology, Cambridge, United Kingdom. Miura, S. 1986. The society of sika deer. Pages 90–93 in Encyclopedia of animals. Heibonsha Publishing Company, Tokyo, Japan. (In Japanese.) Miyaki, M., S. Hori, Y. Nishikawa, and K. Kaji. 1996. A study on the carrying capacity of Nakanoshima Island in Lake Toya. Pages 143–147 in Studies on landscape ecology of management of riparian ecosystem in terms of conservation of well-conserved natural ecosystem. Forestry Technology Center of Hokkaido, Japan. (In Japanese.) Moen, A. N. 1973. Wildlife ecology. Freeman and Company, New York, New York, USA. Mooney, H. A., and S. L. Gulmon. 1982. Constraints on leaf structure and function in reference to herbivory. BioScience 32:198–206. Murphy, D. A. 1963. A captive elk herd in Missouri. Journal of Wildlife Management 27:411–414. Nagata, J., R. Masuda, and M. C. Yoshida. 1995. Nucleotide sequences of the cytochrome b and the 12S rRNA genes in the Japanese sika deer Cervus nippon. Journal of the Mammalogical Society of Japan 20:1–8. Ochiai, K. 1999. Diet of the Japanese serow (Capricornis crispus) on the Shimokita Peninsula, northern Japan, in reference to variations with a 16-year interval. Mammal Study 24:91–102. Oshima, Y. 1961. Ecological studies of Sasa communities, I. Productive structure of some of the Sasa communities in Japan. Botanical Magazine, Tokyo 74:189–210. Ozoga, J. J. 1968. Variation microclimate in a conifer swamp deeryard in Northern Michigan. Journal of Wildlife Management 32:574–585. Pekins, P. J., and W. W. Mautz. 1988. Digestibility and nutritional value of autumn diets of deer. Journal of Wildlife Management 52:328–332. Picton, H. D. 1984. Climate and the prediction of reproduction of three ungulate species. Journal of Applied Ecology 21:869–879. Pollack, A. M. 1974. Seasonal changes in appetite and sexual condition in red deer stags maintained on a six-month photoperiod. Journal of Physiology 244:95–96. Powell, G. M., J. J. Miller, A. H. Olavesen, and C. G. Curtis.1974. Liver as major organ of phenol detoxification? Nature 252:234. Pudd, W. J, A. L. Ward, and L. L. Irwin. 1983. Do split hunting seasons influence elk migrations from Yellowstone National Park. Wildlife Society Bulletin 11:328–331. Renecker, L. A., and R. J. Hudson. 1985. Estimation of dry matter intake of free-ranging moose. Journal of Wildlife Management 49:785–792.

16 North-South Variations in Sika Deer Ecology as a Forest-Dwelling Cervid

229

Risenhoover, K. L., and J. A. Bailey. 1985. Foraging ecology of mountain sheep: Implications for habitat management. Journal of Wildlife Management 49:797–804. Robbins, C. T. 1983. Wildlife feeding and nutrition. Academic, San Diego, California, USA. Robbins, C. T., R. S. Radbielancik-Norman, D. L. Wilson, and E. D. Mould. 1981. Growth and nutrient consumption of elk calves compared to other species. Journal of Wildlife Management 45:172–186. Robbins, C. T., T. A. Hanley, A. E. Hagerman, O. Hjeljord, D. L. Baker, C. C. Schwartz, and W. W. Mautz. 1987a. Role of tannins in defending plants against ruminants: Reduction in protein availability. Ecology 68:98–107. Robbins C. T., S. Mole, A. E. Hagerman, and T. A. Hanley. 1987b. Role of tannins in defending plants against ruminants: Reduction in dry matter digestion? Ecology 68:1606–1615. Robbins, C. T., A. E. Hagerman, P. J. Austin, C. McArthur, and T. A. Hanley. 1991. Variation in mammalian physiological responses to a condensed tannin and its ecological implications. Journal of Mammalogy 72:480–486. Robbins, C. T., D. E. Spalinger, and W. Hoven. 1995. Adaptation of ruminants to browse and grass diets: Are anatomical-based browser-grazer interpretation valid? Oecologia 103:208–213. Schwartz, C. C., W. L. Regelin, and A. W. Franzmann. 1987. Nutritional energetics of moose. Swedish Wildlife Research Supplement 1:301–310. Simpson, A. W., J. S. Webster, J. S. Smith, and C. A. Simpson. 1978. The efficiency of utilization of dietary energy for growth in sheep (Ovis ovis) and red deer (Cervus elaphus). Comparative Biochemistry and Physiology 59A:95–99. Staines, B. W. 1974. A review of factors affecting deer dispersion and their relevance to management. Mammal Review 4:79–91. Staines, B. W. 1976. The use of natural shelter by red deer (Cervus elaphus) in relation to weather in North-east Scotland. Journal of Zoology, London 180:1–8. Standing, R. A. 1938. Use of key species, key areas, and utilization standards in range management. American Forester 29:9–19. Takatsuki, S. 1982. A note on the herd size of sika deer on Kashima Island. Journal of the Mammalogical Society of Japan 9:54–56. Takatsuki, S. 1983. The importance of Sasa nipponica as a forage for sika deer (Cervus nippon) in Omote-Nikko. Japanese Journal of Ecology 33:17–25. Takatsuki, S. 1986. Food habits of sika deer on Mt. Goyo. Ecological Research 1:119–128. Takatsuki, S. 1988. The weight contributions of stomach compartments of sika deer. Journal of Wildlife Management 52:313–316. Takatsuki, S. 1992a. Foot morphology and distribution of sika deer in relation to snow depth. Ecological Research 7:19–23. Takatsuki, S. 1992b. Sika deer living in the north: Ecology of deer, bamboo, and snow. Dobutsusha Publishers, Tokyo, Japan. (In Japanese.) Takatsuki, S., K. Suzuki, and H. Higashi. 2000. Seasonal up-down movements of sika deer at Mt. Goyo, northern Japan. Mammal Study 25:107–114. Tamate, H. B., and T. Tsuchiya. 1995. Mitochondrial DNA polymorphism in subspecies of the Japanese sika deer, Cervus nippon. Journal of Heredity 86:211–215. Tamate, H. B., S. Tatsuzawa, K. Suda, M. Izawa, T. Doi, K. Sunagawa, F. Miyahira, and H. Tado. 1998. Mitochondrial DNA variations in local populations of the Japanese sika deer, Cervus nippon. Journal of Mammalogy 79:1396–1403. Telfer, E. S., and J. P. Kelsall. 1984. Adaptation of some large North American mammals for survival in snow. Ecology 65:1328–1334. Uno, H., and K. Kaji. 2000. Seasonal movements of female sika deer in eastern Hokkaido, Japan. Mammal Study 25:49–57. Van de Veen, H. 1979. Food selection and habitat use in the red deer (Cervus elaphus L.). Dissertation. Rijksuniversiteitte Groningen, The Netherlands. Van Soest, P. J. 1982. Nutritional ecology of the ruminant. O & B Books, Corvallis, Oregon, USA.

230

S. Takatsuki

Verme, L. J. 1970. Some characteristics of captive Michigan moose. Journal of Mammalogy 51:403–405. Warren, R. J., R. L. Kirkpatrick, A. Oelschlaeger, P. F. Scankon, and F. C. Gwazdauskas. 1981. Dietary and seasonal influences on nutritional indices of adult male white-tailed deer. Journal of Wildlife Management 45:926–936. Wentworth, J. M., A. S. Johnson, and P. E. Hale. 1989. Influence of acorn abundance on whitetailed deer in the Southern Appalachians. Pages 2–6 in C. E. McGee, editor, Proceedings, workshop on Southern Appalachian mast management. University of Tennessee, Knoxville, Tennessee, USA. Wheaton, C., and R. D. Brown. 1983. Feed intake and digestive efficiency of south Texas whitetailed deer. Journal of Wildlife Management 47:442–450. White, R. G., and S. G. Fancy. 1986. Nutrition and energetics of indigenous northern ungulates. Pages 259–269 in O. Gundmundsson, editor, Grazing research at northern latitudes. Plenum, New York, USA. Wood, A. J., I. McT. Cowan, and H. C. Nordan. 1962. Periodicity of growth in ungulates as shown by deer of the genus Odocoileus. Canadian Journal of Zoology 40:593–603. Yabe, T., M. Suzuki, M. Yamanaka, and N. Ohtaishi. 1990. Population trend, food habits, wintering area preference of sika deer on Shiretoko peninsula, and the application of them to nature education (1988). Report of Shiretoko Museum 11:1–20. Yokoyama, M., N. Maruyama, K. Kaji, and M. Suzuki. 1996. Seasonal changes of body fat reserves in sika deer of East Hokkaido, Japan. Journal of Wildlife Research 1:57–61. Yokoyama, M., K. Kaji, and M. Suzuki. 2000. Food habits of sika deer and nutritional value of sika deer diets in eastern Hokkaido, Japan. Ecological Research 15:345–355. Yokoyama, S., T. Koizumi, and E. Shibata. 1996. Food habits of sika deer as assessed by fecal analysis in Mt. Ohdaigahara, central Japan. Journal of Forest Research 1:161–164.

Chapter 17

Geographical Variations in Food Habits of Sika Deer: The Northern Grazer vs. the Southern Browser Seiki Takatsuki

Abstract Based on the fall/winter food compositions of 19 local sika populations from Hokkaido to Kyushu, covering the cool temperate zone to the warm temperate zone, a clear tendency was found. The northern sika deer (>36° N) were dependent on dwarf bamboos and other graminoids, while the southern deer (<35° N) were dependent on browse and fruits. That is, the northern deer were grazers while the southern one were browser types in food habits. This largely corresponded to the genetic groups as well as vegetational zones. The populations around 35° N were variable and the proportions of graminoids ranged from 20% to 80%. The food compositions of sika deer on the Japanese Archipelago vary greatly, suggesting elasticity of this cervid. The background of this difference is discussed.

Introduction The Japanese Archipelago covers a wide range of latitude, from Soya Peninsula in Hokkaido (45° N) to Iriomote Island of Okinawa (26° N). The vegetation includes boreal coniferous forests in Hokkaido, subalpine coniferous forests in Hokkaido and northern Honshu, cool-temperate deciduous broad-leaved forests in many areas, warm-temperate evergreen broad-leaved forests in central and southwestern Japan, and some subtropical forests in southern islands. Because of this great environmental variation, the flora and fauna of the Japanese Archipelago are very rich. In spite of this high biodiversity, there are only three ungulate species occurring in the Japanese Archipelago, all belonging to different families: serow (Capricornis crispis), Bovidae; wild boar (Sus scrofa), Suidae; and sika deer Cervus nippon; Cervidae. The Jarman-Bell principle predicts that smaller ungulates tend to live in forests and feed on high quality plants which often grow patchily, whereas larger ungulates tend to live in grasslands and feed on less digestible plants, mainly grasses, which often grow abundantly and uniformly (Bell 1970; Geist 1974; Jarman 1974). This principle includes various aspects of ungulate ecology, such as energy metabolism, digestive physiology, anti-predator behavior, grouping behavior, and many others. D. R. McCullough et al. (eds.), Sika Deer: Biology and Management of Native and Introduced Populations, DOI: 10.1007/978-4-431-09429-6_17, © Springer 2009

231

232

S. Takatsuki

In terms of foraging, body size is a key factor. It affects basal metabolic rate, with smaller ungulates requiring relatively more energy per unit body mass, and vice versa. Thus, smaller ungulates necessarily require higher quality foods. Higher quality food plants, however, are generally only sporadically available and consequently less abundant. Therefore, smaller ungulates selectively feed on them. Larger ungulates, in turn, require relatively less energy per body weight, but since their absolute body mass is great, absolute intake must be large. Since high-quality food plants are less abundant, the larger ungulates usually feed on abundant food plants, which are less digestible. The combination of energy requirements in relation to body size and food availability in relation to nutritional quality determines that smaller ungulates selectively feed on nutritious foods (browsers) while larger ungulates nonselectively feed on roughage foods (grazers). The physio-morphology of digestive organs of ungulates, particularly ruminants, explains apparent gastrointestinal adaptations to varying food plants. Larger ruminants that eat roughage foods tend to have not only absolutely but relatively larger rumino-reticulums, longer small intestines, and longer retention time, while smaller ruminants who eat nutritional foods tend to have smaller rumino-reticulums, larger omasums, shorter small intestines, and shorter retention time (Hofmann 1973, 1988). Although these comparative studies were developed to explain the diversified adaptation of different ungulate species to various environments, they seem valid for explaining intraspecific variations as well. Takatsuki (1988) has shown that the contribution of the rumino-reticulum of sika deer changes in relation to body size and pointed out that the Jarman-Bell principle is valid within species as well. Sika deer show great geographical variation in body size: the northern sika deer are larger than southern ones. It is, therefore, expected that the northern sika deer are grazers and the southern ones are browsers. However, this principle cannot be simply adopted to explain the relationships between sika deer foods and the habitat vegetation. A unique feature of the deciduous broad-leaved forests of Japan is that the forest floors are densely covered by dwarf bamboos (Sasa spp.) (Miyawaki 1986). The dwarf bamboos are evergreen and therefore very important for the sika deer in winter (Takatsuki 1983, 1986). The Jarman-Bell principle, originally proposed based on the observations of East African ungulates, predicts that larger ungulates are dependent on grasses growing in openlands (Jarman 1974). Dwarf bamboos belong to the Gramineae (Poaceae), but they are different from openland grasses in that (1) they grow in the forest floor, (2) the culms are woody and live longer than other grasses, and (3) they are evergreen. According to these traits, it is expected that food habits of sika deer in the Japanese Archipelago would show geographical variations, because (1) vegetation and ecosystems in northern and southern Japan are greatly different, (2) in spite of such great variations in the habitat, there is only one cervid species, (3) body sizes of sika deer are different (larger in the north and smaller in the south), (4) larger ungulates generally eat roughage food plants, and (5) the northern forests supply dwarf bamboos while southern one are vegetated by shrubs and supply berries and acorns in autumn. Thus, sika deer are suitable for studying how an ungulate species varies according to different habitats. The objective of this study is to review the

17 Geographical Variations in Food Habits of Sika Deer

233

food habit study of sika deer from northern to southern Japan, with attention paid to the contribution of graminoids, particularly dwarf bamboos.

Materials and Methods I reviewed published papers of my own and others that adopted quantitative analyses of rumen contents and feces. Since many samples were collected in autumn or winter, and since food habits show considerable seasonal variations, I used only autumn and winter data, though I used the summer data for the Yakushima Island population (Takatsuki 1990). Foods were simply categorized into graminoids and others. The proportions of graminoids were plotted against latitude (Fig. 17.1).

Fig. 17.1 North-south variation in the contributions of graminoids in the foods of sika deer ( autumn, ° winter, summer) along the Japanese archipelago. (1) Ashoro (Yokoyama et al. 2000); (2) Shiranuka Hills (Campos-Arceiz and Takatsuki 2005); (3) Mt. Goyo (Takatsuki 1983); (4) Kinkazan Island (Takatsuki 1980); (5) Takahara (Ueda et al. 2002); (6) Oze (Takatsuki 2003); (7) Omote-Nikko (Takatsuki 1986); (8) Shimane Peninsula (Takatsuki and Sato 1988); (9) Ashuu (Tanaka, Y. unpublished); (10) Shizuoka (Jiang, Z. and S. Takatsuki unpublished); (11) Boso Peninsula (Asada and Ochiai 1996): (12) Minoo (Takatsuki 1985); (13) Nara Park (Takatsuki and Asahi 1978); (14) Ohdaigahara (S. Yokoyama et al. 1996); (15) Yamaguchi (Jayasekara and Takatsuki 2000); (16) Tshushima Island (Suda 1997); (17) Fukuoka (Ikeda 2002); (18) Nozaki Island (Takatsuki et al. 1984); (19) Yaku Island (Takatsuki 1990).

234

S. Takatsuki

Results It is apparent that northern sika deer are highly dependent on graminoids as their food while southern ones are less dependent on them (Fig. 17.1). The southern deer feed on fruits, seeds, and browse. The northern sika deer from Hokkaido through the northern Kanto district living in the deciduous broad-leaved forests subsist on Sasa nipponica from autumn to spring. An exception was a study in Ashoro, Hokkaido (Point 1 in Fig. 17.1) that showed S. nipponica accounted for only 24.1% and dead leaves of broad-leaved trees, twigs, and bark occupied as much as 49.3% of the diet (Yokoyama et al. 2000). The comparatively small representation of S. nipponica in the diet of Ashoro deer in winter suggests that they probably could not reach S. nipponica because it was buried under snow. Thus, I treated the plot of the Ashoro deer differently from other northern deer. The sika deer populations of Shiranuka Hills (Point 2 in Fig. 17.1; Campos-Arceiz and Takatsuki 2005) in eastern Hokkaido, Mt. Goyo in northern Honshu (Point 3 in Fig. 17.1; Takatsuki 1986), Nikko (Point 6 in Fig. 17.1; Takatsuki 1983) and Mt. Takahara (Point 5 in Fig. 17.1; Ueda et al. 2002) in the northern Kanto district fed exclusively on S. nipponica. The population of Oze at the junction of Fukushima, Gunma, and Niigata prefectures fed on another Sasa species, S. palmata (Point 6 in Fig. 17.1; Takatsuki 2003). Subsistence on S. nipponica is also found in the Kinki district, central Japan (Point 14 in Fig. 17.1; S. Yokoyama et al. 1996). The population on Kinkazan Island in the Tohoku district fed on grasses such as Zoysia japonica (Japanese lawn grass), Miscanthus sinensis (silver grass), Brachypodium sylvaticum (false brome), and Pleioblastus chino (a dwarf bamboo) (Point 4 in Fig. 17.1, Takatsuki 1980). This is quite different from southern deer where the contribution of graminoids were small in the whole diet (Fig. 17.1). The southern deer do not subsist on graminoids not because they are not available but because they can find better foods like fruits, seeds, and other evergreen leaves. The sika deer population living in the evergreen broad-leaved forests in the areas west to the Kanto district feed on leaves of evergreen trees and shrubs and fruits/seeds. The population on the Boso Peninsula feeds on leaves of sedges and evergreen shrubs such as Aucuba japonica (Japanese laurel), Eurya japonica (Theaceae), and acorns of Lithocarpus edulis) (an evergreen oak) (Asada and Ochiai 1996). The population in Yamaguchi Prefecture, western Honshu, fed on leaves of Eurya japonica and acorns of oaks (Point 14 in Fig. 17.1; Weerasinghe and Takatsuki 1999; Jayasekara and Takatsuki 2000). For the population on Tshushima Island, browse was important, occupying as much as 63–64% of the diet and graminoids accounted for only 2–3% (Suda 1997). For the population in northern Kyushu, leaves of herbs and browses were important and graminoids appeared only slightly in the feces of sika deer (Point 17 in Fig. 17.1; Ikeda 2002). On Yakushima Island, the deer in the lower elevation were browsers but those above the timber line subsisted on a dwarf bamboo (Pseudosasa owararii, Point 19 in Fig. 17.1; Takatsuki 1990).

17 Geographical Variations in Food Habits of Sika Deer

235

Discussion As predicted, there was a clear tendency that the northern sika deer are grazers and southern ones are browsers (Fig. 17.1). The deer of the northern group are larger in body size and have well-developed rumino-reticulums, smaller omasums, and longer small intestines (Takatsuki 1988). They live in deciduous broad-leaved forests and feed on Sasa species. Since Sasa species are evergreen, they provide a high protein intake for sika deer (Takatsuki 1983; Ikeda and Takatsuki 1999). In contrast, the deer of the southern group are smaller in body size and probably have less developed rumino-reticulums, relatively larger omasums, and shorter small intestines. They live in evergreen broad-leaved forests and feed on nutritional foods like fruits/seeds in autumn and winter. The boundary between the two groups exists at around 35° N. This corresponds well with the boundary between the evergreen and deciduous broad-leaved forests in Honshu. As is seen in Fig. 17.1, the proportions of graminoids in the diet are variable around this transitional zone. A recent important finding of sika deer genetics is that the sika deer of the Japanese archipelago are divided into two groups (Nagata et al. 1995; Tamate et al. 1998; Goodman et al. 2001), a northern group and a southern group, and the former seven named subspecies are invalid. The population of Yamaguchi Prefecture at the western end of Honshu belongs to the southern group, but the population of Hyogo Prefecture belongs to the northern group in spite of latitudinal proximity and continuity (west of Point 9 in Fig. 17.1). It is noteworthy that the Boso Peninsula population belonging to the northern genetic group shows intermediate type food habits. In contrast, the population living at higher elevations of Yaku Island in southern Kyushu are grazers (Takatsuki 1990). These results give us a hint to consider the relationships between genetic populations and ecological populations (Fig. 17.2). If we consider these results to mean that genetic populations correspond with ecological populations represented by food habits, the results suggest that the genetic differences affect digestive morphology and physiology, and consequently different food selections. The browser-type food habits of the Yamaguchi population (a southern genetic group) support this. However, a contradiction is that the Boso Peninsula population (a northern genetic group) show intermediate type food habits (Asada and Ochiai 1996). Nevertheless, although food habits reflect genetic differences overall, sika deer are elastic in food habits, locally depending on habitat variability. In this case, genetic characters are “masked.” The browser-type food habits of the Boso Peninsula population support this. The results of the Yakushima Island population where the deer at lower elevation were browsers while those above the timber line were grazers also support this interpretation. Furthermore, the food habits along the transitional zone are quite variable, and more complete analyses are needed for these local populations.

236

S. Takatsuki

Fig. 17.2 A conceptual comparison among genetic groups, food habits, and habitats (vegetation) of sika deer in the Japanese archipelago.

Literature Cited Asada, M., and K. Ochiai. 1996. Food habits of sika deer on the Boso Peninsula, central Japan. Ecological Research 11:89–95. Bell, R. 1970. The use of the herb layer by gazing ungulates in the Serengeti. Pages 111–124 in A. Watson, editor, Animal populations in relation to their food resources. Blackwell Scientific, Oxford, United Kingdom. Campos-Arceiz, A., and S. Takatsuki. 2005. Food habits of sika deer in the Shiranuka Hills, eastern Hokkaido - A northern example among the north-south variations of food habits in sika deer. Ecological Research 20:129–133. Geist, V. 1974. On the relationship of social evolution and ecology in ungulates. American Zoologist 14:205–220. Goodman, S. J., H. B. Tamate, R. Wilson, J. Nagata, S. Tatsuzawa, G., Swanson, J. M. Pemberton, and D. R. McCullough. 2001. Bottlenecks, drift and differentiation: The population structure and demographic history of sika deer (Cervus nippon) in the Japanese Archipelago. Molecular Ecology 10:1357–1370. Hofmann, R. R. 1973. The ruminant stomach. East African Literature Bureau, Nairobi, Kenya. Hofmann, R. R. 1988. Anatomy of the gastro-intestinal tract. Pages 14–43 in C. Church, editor, The ruminant animal. Digestive physiology and nutrition. Prentice-Hall, Upper Saddle River, New Jersey, USA. Ikeda, K. 2002. The food habits of sika deer in a coniferous plantation area, northern Kyushu. Journal of the Japanese Forestry Society 84:175–179. (In Japanese with English abstract.) Ikeda, S., and S. Takatsuki. 1999. Seasonal changes in nutritive compositions of the major food plants of sika deer (Cervus nippon) and Japanese serow (Capricornis crispus) - a case study of the Sendai area. Tohoku Journal of Animal Science and Technology 49:1–8. Jarman, P. J. 1974. The social organization of antelope in relation to their ecology. Behaviour 48:215–266.

17 Geographical Variations in Food Habits of Sika Deer

237

Jayasekara, P., and S. Takatsuki. 2000. Seasonal food habits of a sika deer population in the warm temperate forest of the westernmost part of Honshu, Japan. Ecological Research 15:153–157. Miyawaki, A., editor. 1986. Vegetation of Japan, Vol. 7. Kanto. Shinbundo, Tokyo, Japan. Nagata, J., R. Masuda, and M. C. Yoshida. 1995. Nucleotide sequences of the cytochrome b and the 12S rRNA genes in the Japanese sika deer Cervus nippon. Journal of the Mammalogical Society of Japan 20:1–8. Suda, K. 1997. Rumen contents and food selectivity of sika deer (Cervus nippon) on Tsushima Islands. Wildlife Conservation Japan 2: 25–134. (In Japanese with English abstract.) Takatsuki, S. 1980. Food habits of sika deer on Kinkazan Island. Science Reports of Tohoku University, Series IV (Biology) 38(1):7–31. Takatsuki, S. 1983. The importance of Sasa nipponica as a forage for sika deer (Cervus nippon) in Omote-Nikko. Japanese Journal of Ecology 33:17–25. Takatsuki, S. 1986. Food habits of sika deer on Mt. Goyo. Ecological Research 1:119–128. Takatsuki, S. 1988. The weight contributions of stomach compartments of sika deer. Journal of Wildlife Management 52:13–316. Takatsuki, S. 1990. Summer dietary compositions of sika deer on Yakushima Island, southern Japan. Ecological Research 5:253–260. Takatsuki, S. 2003. Use of mires and food habits of sika deer in the Oze Area, central Japan. Ecological Research 18:331–338. Tamate, H. B., S. Tatsuzawa, K. Suda, M. Izawa, T. Doi, K. Sunagawa, F. Miyahira, and H. Tado. 1998. Mitochondrial DNA variations in local populations of the Japanese sika deer, Cervus nippon. Journal of Mammalogy 79:1396–1403. Ueda, H., S. Takatsuki, and Y. Takahashi. 2002. Bark stripping of hinoki cypress by sika deer in relation to snow cover and food availability on Mt. Takahara, central Japan. Ecological Research 17:545–551. Weerasinghe, U. R., and S. Takatsuki. 1999. A record of acorn-eating by sika deer in western Japan. Ecological Research 14:205–209. Yokoyama, M., K. Kaji, and M. Suzuki. 2000. Food habits of sika deer and nutritional value of sika deer diets in eastern Hokkaido, Japan. Ecological Research 15:345–355. Yokoyama, S., T. Koizumi, and E. Shibata. 1996. Food habits of sika deer as assessed by fecal analysis in Mt. Ohdaigahara, central Japan. Journal of Forest Research 1:161–164.

Chapter 18

What Is “Natural” Vegetation? A Reconsideration of Herbivory by Wild Ungulates Seiki Takatsuki

Abstract Traditionally, Japanese plant ecologists treated grazing effects on vegetation lightly. When they considered the grazing effects, they regarded them as artificial because their frame of reference was the effects of livestock. Plant ecologists recognized that the oak-Sasa nipponica (a dwarf bamboo) or the birch-S. nipponica forests were plant communities characterized by simple forest structures with poor shrub layers and dense S. nipponica cover in the understories. Nobody realized that they are affected by grazing of sika deer. A long-term study (20 years) using a deer-proof fence, has shown, however, that the oak-S. nipponica forest is influenced by continuous grazing of sika deer. If deer grazing is excluded, the understory becomes densely covered by woody plants which suppress the growth of S. nipponica. It is necessary, therefore, to reconsider the effects of wild ungulates for the maintenance of several natural communities.

Introduction Japanese plant ecologists have treated effects of herbivores as unimportant or often ignored them (Miyawaki 1967; Numata 1969; Ito 1977). Some note grazing effects of livestock and emphasize that grazing functions to slow or stop vegetation succession. Or, they regard grazing as a factor maintaining grasslands or early successional vegetation (Numata 1969, 1973, 1974; Ito 1977). Grazing is regarded as artificial stress on plant communities. The deciduous broad-leaved forests in the Pacific side of Japan are dominated by oaks (Quercus spp.) or birch (Betula spp.) in the tree layers and Sasa nipponica in the forest floors (Miyawaki 1986). The physiognomy is characterized by the simple forest structure with poor shrub layers and uniform undergrowth of Sasa nipponica (Fig. 18.1). They are considered as “natural” forests in terms of not being affected by livestock herbivory. I evaluated this phenomenon by establishing deer-proof fences in Mt. Goyo in northern Honshu. It seems to be an important opportunity to reconsider the attitude of traditional Japanese plant ecologists who overlooked the role of ungulates in shaping forest ecosystems. D. R. McCullough et al. (eds.), Sika Deer: Biology and Management of Native and Introduced Populations, DOI: 10.1007/978-4-431-09429-6_18, © Springer 2009

239

240

S. Takatsuki

Fig. 18.1 View of the oak-Sasa nipponica forest with poor shrub layers at Mt. Goyo.

Study Area The study was done at Mt. Goyo situated in the Pacific side in northern Honshu (see Fig. 26.1 in Takatsuki chapter 26). The vegetation of northern Japan is quite different between the Pacific and the Japan Sea sides because of snow. The Pacific side is less snowy, and evergreen plants are sparse because plants above the snow are exposed to chilly, dry winds, while many evergreen shrubs grow on the Japan Sea side because they are buried and preserved in snow. This is the case for dwarf bamboos that are often dominant in the undergrowth of the deciduous broadleaved forests in northern Japan. Sasa nipponica, a short dwarf bamboo with high turnover of the above-ground parts, grows rapidly on the Pacific side, while S. palmata and S. kurilensis which are taller with slow turnover of the above-ground parts, grow on the Japan Sea side (Suzuki 1961). The oak-S. nipponica or the birch-S. nipponica forests are found from the Pacific side of middle Honshu through eastern Hokkaido. The distribution of sika deer in northern Japan is also affected by snow (Takatsuki 1992). They are confined to area with snow depths of less than 1 m, and abundant in the area with snow of less than 50 cm. Consequently, the distributions of sika deer and Sasa nipponica correspond. In the Sasa nipponica range, sika deer subsist on S. nipponica as a food, particularly in winter (Takatsuki 1983, 1986). In Mt. Goyo, the deer show elevational migrations according to snow depth (Takatsuki et al. 2000). The wintering areas in the lower elevations are concentrated on by sika deer in winter, and the leaves of Sasa nipponica there are heavily grazed (Takatsuki 1986). Though S. nipponica is tolerant to grazing, it is reduced in height by continuous heavy grazing.

18 What Is “Natural” Vegetation? A Reconsideration of Herbivory

241

Deer-Proof Fences A deer-proof fence was built in an oak forest at 600 m elevation in the reserve of Mt. Goyo in November, 1988. This is a place where sika deer concentrate in winter and Sasa nipponica is heavily grazed. The oak forest is a coppice which before the 1960s was used as a charcoal producing forest. The fence was a 5 m × 5 m square; wooden posts were set at the corners, and nylon net 2 m high with 5 cm aperture mesh was stretched between the poles. A larger fence (7 m × 7 m) was placed surrounding the 5 m × 5 m fence to further prevent invasion by the deer. The lower skirt was further strengthened by wire mesh net with a rust-proof coating. Sasa nipponica was sampled by clipping to the ground level in a 1 m × 1 m quadrat inside and outside the fence in every October from 1989 to 2001. The height of the clipped culms was measured to the nearest 0.5 cm. The leaves and culms were separated, and dried at 70° C for 48 hours in a drying oven. Plant materials were weighed to the nearest 0.1 g on an electronic balance. In 1990, 1993, 1997 and 2001, productive structures of Sasa nipponica both inside and outside the fence were measured for height interval (20 cm). In advance of fence establishment, the same samplings were done in 1981 and 1982, and the data were used for comparison. In 2001, 13 years after fencing, twenty 1 m × 1 m quadrats were randomly selected inside and outside the fence, and the height of all the woody plants in them were measured to the nearest centimeter.

Results The effect of the fence was pronounced (Fig. 18.2). The height of Sasa nipponica averaged about 70 cm in 1981 but decreased to about 50 cm in 1982 (Fig. 18.3). Data were not available from 1983 to 1988. Height had become as short as about 15 cm in 1989 when the fence was established. It recovered to about 40 cm by fencing as early as the next year and peaked at about 60 cm in 1993, 5 years after fencing (Fig. 18.3). However, it became shorter again after 1996, probably because of shading by larger plants (see below). In 1981 and 1982, biomass of inside and outside the fence was 250 g/m2, with culms being 150 g/m2 and leaves 100 g/m2 (Fig. 18.4). However, by 1989, biomass was reduced both inside and outside the fence. The inside value, however, remained greater than the outside value. Thereafter, outside the fence, culm weight fluctuated between 10 and 40 g/m2, and leaf weight fluctuated around 50 g/m2. Both the leaf and culm weights changed in synchrony. Inside the fence, the culm weight gradually decreased after 1993 from 50 to 100 g/m2 to around 50 g/m2. Decrease of leaf generally paralleled culm weight but not in synchrony. The difference in the production of culms and leaves inside and outside the fence is apparent in Fig. 18.5. Inside, S. nipponica culms were taller and leaves were distributed above 20 cm and more than half were above 40 cm before 1993. After

242

S. Takatsuki

Fig. 18.2 Difference inside and outside the fence. Since the fence used in the present study was double-layered, it was difficult to show the inside view. Thus, another single-layered fence is shown to demonstrate the effects of the fence.

Fig. 18.3 Changes in culm height of Sasa nipponica inside and outside the deer-proof fence at Mt. Goyo. The fence was established in 1988. Vertical bars show SD.

1997, however, leaves inside the fence were distributed between 20 and 40 cm high because of shading by shrubs (see below). In 2001, many other species of woody plants grew taller than S. nipponica inside the fence (Table 18.1). As many as 20 species appeared inside the fence while only seven species were found outside. All the woody plants outside the fence were shorter than 10 cm, but some of them inside were taller than 2 m. The height distribution of tree, shrub, and liana species show that they grew well inside the fence, though the variations were great (Fig. 18.6).

18 What Is “Natural” Vegetation? A Reconsideration of Herbivory

243

g / m2 Inside

100 ?

2000

2001 2001

1999

2000

1998

Outside

150

Biomass

1997

1996

1995

1994

1993

1992

1991

1990

1989

1988

1987

1984

1983

1982

1981

0

1986

?

50

1985

Biomass

150

100 ? ?

50

1999

1998

1997

1996

1995

1994

1993

1992

1991

1990

1989

1988

1987

1986

1985

1984

1983

1982

1981

0

Fig. 18.4 Changes in biomass of Sasa nipponica inside (top) and outside (bottom) the deer-proof fence at Mt. Goyo. Circles: leaf, squares: culm.

Discussion Height and biomass of Sasa nipponica in 1981 and 1982 were much greater than those in 1989. This period coincided with the rapid increase of the sika deer population in this area. In the early 1990s, the deer density became quite high; trees and shrubs in the wintering areas were heavily barked (Fig. 26.4 in Takatsuki chapter 26) and many deer starved to death in late winter and early spring. Though data are not available for the middle 1990s, it is likely that size reduction of S. nipponica occurred in this period. Because of intensification of deer culling, the density became lower and barking less frequent after the late 1990s. The recovery of the plants inside the fence was complicated. The objective of the fence establishment was to record how Sasa nipponica would recover after deer

Height

cm 80

1990 Outside

60

60

40

40

20

20

0

0

60 40 20 cm 80

Height

Inside

cm 80

0

20 40 60 80

1997 Outside

Inside

cm 80 60

40

40

20

20

0

0 0

20 40 60 80 g /m2

Outside

60 40 20

60

60 40 20

1993

0

20 40 60 80

2001 Outside

60 40 20

Biomass

Inside

Inside

0

20 40 60 80 g /m2

Biomass

Fig. 18.5 Productive structures of Sasa nipponica inside (right) and outside (left) the fence at Mt. Goyo. Dark gray: leaf, dotted: culm. Table 18.1 Numbers in square meters and mean height of woody plants in 2001 inside and outside a deer-proof fence established in 1989 in the foothills of Mt. Goyo. Inside Outside

Symplocos chinensis Swida controversa Fraxinus sieboldiana Prunus verecunda Carpinus tschonoskii Magnolia obovata Acer amoenum var. matsumurae Zelkova serrata Quercus serrata Rubus palmatus Weigela hortensis Callicarpa japonica Lonicera gracilipes Lindera umbellata Viburnum dilatatum Zanthoxylum piperitum Rubus microphylla Rosa multiflora Euonymus sieboldianus Rubus koehneanus Actinidia arguta Celastrus orbiculatus

Life form

Number

Mean height

SD

Number

Mean height

SD

Tree Tree Tree Tree Tree Tree Tree

7 7 3 2 1 1 1

21.6 49.1 128.3 17.5 10.0 25.0 160.0

9.3 66.6 117.7 2.1 – – –

13 0 4 0 0 0 0

4.4 – 3.5 – – – –

0.8 – 0.7 – – – –

Tree Tree Shrub Shrub Shrub Shrub Shrub Shrub Shrub Shrub Shrub Shrub Shrub Liana Liana

0 0 12 6 5 4 4 3 1 1 1 1 1 12 5

– – 38.9 171.7 184.0 76.5 81.3 112.7 120.0 6.0 25.0 20.0 18.0 36.3 15.4

– – 30 89.8 53.7 69.1 79.6 94 – – – – – 45 2.9

2 14 0 0 0 0 0 9 0 0 4 0 71 0 0

5.0 5.7 – – – – – 4.8 – – 7.0 – 6.7 – –

– 1.4 – – – – – 0.5 – 2.8 – 3.2 – –

18 What Is “Natural” Vegetation? A Reconsideration of Herbivory

245

Fig. 18.6 Frequency distributions of height of different types of woody plants in 20 m2 inside and outside the fence.

246

S. Takatsuki

exclusion. It rapidly recovered, starting as early as the next year after fencing. However, eight years after fencing, other woody plants exceeded S. nipponica. Since S. nipponica is intolerant to shading (Agata and Kubota 1976), it became shorter again inside the fence (Fig. 18.2) due to the shading by other woody plants. Although the coverage of S. nipponica was not accurately measured, it was obviously reduced. In 1988, the fenced plot was placed on a site where S. nipponica was continuous and dense (coverage was almost 100%). By 2001, however, the coverage inside the fence declined to about 30%. The results of the present study strongly suggests that the physiognomy of the oakSasa nipponica forest at Mt. Goyo is maintained by continuous browsing by sika deer. It is apparent that Japanese plant ecologists have overlooked the effects of deer. From the results of this experiment, it is probable that the physiognomy of these forests are maintained by sika deer herbivory. If the deer disappeared, shrub layers would become thicker and would suppress the growth of Sasa nipponica. Even if the floral compositions of the forests would be similar, the physiognomy would be quite different. Japanese plant ecologists traditionally regard effects of ungulates as “artificial,” because they consider herbivory is asserted only by livestock while ignoring wild ungulates. The wild ungulates are no doubt natural components of Japanese ecosystems. Existence and maintenance of not only early successional plant communities but also some forest communities like the oak-Sasa nipponica community should be understood with consideration of wild ungulates. If deer herbivory reduces the shrubs, it would facilitate the growth of Sasa nipponica. This would favor the deer because their foods increase. This relationship seems to be a good example of a “grazing loop” (Danell et al. 1985). Such a point of view on ungulate-plant community interactions has been overlooked by Japanese plant ecologists. For our understanding of the dynamics of these forests to improve, (1) Japanese plant ecologists need to evaluate the effects of wild ungulates as factors in maintaining some plant communities and (2) they should consider herbivory of ungulates as natural and they need remember that some wild ungulates assert significant effects on the maintenance of plant communities.

Literature Cited Agata, W., and F. Kubota. 1976. The effect of light intensity on the habitat differentiations of Sasa nipponica and S. pupurascens in deciduous broad-leaved forests. Bulletin Grassland Ecology Research Group 15:38–44. Danell, K, K. Huss-Danell, and R. Bergström. 1985. Interactions between browsing moose and two species of birch in Sweden. Ecology 66:1867–1878. Ito, S., editor. 1977. Composition and structure of plant communities. Plant Ecology Series 2. Asakura Publishing, Tokyo, Japan. (In Japanese.) Miyawaki, S., editor. 1967. Vegetation of Japan. Gakushu-kenkyuusha Publishing, Tokyo, Japan. (In Japanese.) Miyawaki, A., editor. 1986. Vegetation of Japan, Vol. 7. Kanto. Shinbundo, Tokyo. Japan. Numata, M., editor. 1969. Illustrated book of plant ecology. Asakura Publishing, Tokyo, Japan. (In Japanese.)

18 What Is “Natural” Vegetation? A Reconsideration of Herbivory

247

Numata, M. 1973. Succession. Pages 74–92 in M. Numata, editor. Ecology of grasslands. Tsukijishoten, Tokyo, Japan. (In Japanese.) Numata, M., editor. 1974. Flora and vegetation in Japan. Kodansha and Elsevier, Tokyo, Japan. (In Japanese.) Suzuki, S. 1961. Ecology of bambusaceous genera Sasa and Sasamorpha in the Kanto and Tohoku districts of Japan, with special reference to their geographical distribution. Ecological Review 15:131–147. Takatsuki, S. 1983. The importance of Sasa nipponica as a forage for Sika deer (Cervus nippon) in Omote-Nikko. Japanese Journal of Ecology 33:17–25. Takatsuki, S. 1986. Food habits of sika deer on Mt. Goyo. Ecological Research 1:119–128. Takatsuki, S. 1992. Foot morphology and distribution of sika deer in relation to snow depth. Ecological Research 7:19–23. Takatsuki, S, K. Suzuki, and H. Higashi. 2000. Seasonal up-down movements of sika deer at Mt. Goyo, northern Japan. Mammal Study 25:107–114.

Chapter 19

Seasonal Migration of Sika Deer on Hokkaido Island, Japan Hiromasa Igota, Mayumi Sakuragi, and Hiroyuki Uno

Abstract We review the large-scale seasonal migration of sika deer on Hokkaido Island, Japan. A total of 57 female deer wintering in the Shiranuka Hills were radiotracked between April 1997 and May 2002, yielding 7,765 relocations. Ten (18%) of the deer were upward migrants that had summer home ranges lower than or at similar elevation as their winter home ranges, 29 (51%) were downward migrants that had summer home ranges higher than their winter home ranges, and 12 (21%) were the nonmigrants that had overlapping seasonal home ranges. The summer home ranges of migrants were widely scattered over a total area of 5,734 km2. Migration distance of all migrants averaged 35.1 ± 3.6 km (mean ± SE, n = 39, range = 7.2–101.7 km). Deer showed strong site fidelity to their seasonal ranges. The results suggest that snow cover and bamboo grass are the factors affecting seasonal migration of sika deer in eastern Hokkaido. In addition, coniferous cover can be another important factor in the case of the reverse altitudinal migration by the upward migrants. We think the three migration types developed during the process of deer expanding their distribution after a population bottleneck 130 years ago.

Introduction Many animals, e.g., insects, fishes, birds, and mammals, exhibit a pattern of movement that is termed a migration by the vast majority of zoologists (Baker 1978). Many species of large mammals perform movements, which can be categorized into local movements, migration, dispersal, and nomadism (McCullough 1985). Local movements are those encompassed by daily travels to obtain food, water, escape or concealment cover, or den site, and seasonal shifts as resource availability changes. Migration is a regular, round-trip movement of individuals between discrete habitats (White and Garrott 1990). Dispersal is a one-way movement from one geographic area to another, not related to seasonal resource availability. Nomadism is broad-scale movement that is erratic or random in space, but not necessarily in time. The boundaries among local movement, migration, and dispersal are sometimes unclear, and the categories may be arbitrary in a sense, and a matter of convenience (McCullough 1985). D. R. McCullough et al. (eds.), Sika Deer: Biology and Management of Native and Introduced Populations, DOI: 10.1007/978-4-431-09429-6_19, © Springer 2009

251

252

H. Igota et al.

Cervids that inhabit temperate and boreal regions often exhibit seasonal migration between summer and winter ranges (e.g., elk (Cervus elaphus), Adams 1982; white-tailed deer (Odocoileus virginianus), Nelson and Mech 1987; mule deer (O. hemionus hemionus), Garrott et al. 1987; Nicholson et al. 1997; blacktailed deer (O. hemionus columbianus), Loft et al. 1984; caribou (Rangifer tarandus), Cumming and Beange 1987; moose (Alces alces), Histøl and Hjeljord 1993). Seasonal migration might have allowed cervids to disperse toward areas of high altitude or latitude, where the environment seasonally changes drastically and becomes worse, especially, in winter. In boreal mountainous regions, the activities of deer are physically or physiologically restricted by snow in winter (Verme 1973; McCullough 1985). Foraging beneath the snow becomes difficult (Parker et al. 1984; Takatsuki 1992) and energy expenditure increases at low temperatures (Silver et al. 1971). These disadvantages may affect reproductive success and cause high mortality in years of heavy snow (Kaji et al. 1988; Takatsuki et al. 1994; Uno et al. 1998). Deer, therefore, may aggregate in limited refuges (i.e., wintering areas) with less snow and warmer temperatures than in more widespread summer ranges where females give birth and nurse fawns (Adams 1982). In Hokkaido, Inukai (1952) reported that sika deer migrated seasonally between the northwest where snow was heavy, and the southeast where snow was less in the Edo era (1603–1867) and Ainu people (native people on Hokkaido Island, see Uno et al. chapter 29) hunted deer in spring and autumn near by the migration routes. Overharvest and heavy snow falls led to decrease the population size of sika deer in the late nineteenth century (Matsuda et al. 1999). At least three populations of sika deer survived the bottleneck and expanded from a few natural refuges in the mountainous areas of Akan, Hidaka, and Daisetsu districts by the 1990s (Nagata et al. 1998; Kaji et al. 2000). Large-scale migration of deer between the northwest and the southeast was not observed in the 1950s (Inukai 1952), and that suggested most of the deer that survived the bottleneck were sedentary in the wintering areas. The deer surviving in the wintering areas thereafter might have dispersed to the surrounding areas. If the new land was not suited for wintering, the dispersers might have moved back to the familiar land in winter. Migratory behavior is considered to develop as the result (Igota et al. 2004). As their distribution expanded, the scale of the migration also enlarged over several generations. The summering areas of sika deer in eastern Hokkaido are extensive and dispersed, on the other hand, while the potential wintering areas are restricted to a few places because of heavy snow and limited coniferous forests (Kaneko et al. 1998; Sakuragi et al. 2003a). Therefore, sika deer in eastern Hokkaido have a large-scale seasonal migration. A previous study first described such large scale migration from the Lake Akan wintering area to the northeastern area (Uno and Kaji 2000). On the other hand, in western Hokkaido, the deer distribution has expanded to the area facing the Sea of Japan and the population has increased during the 1990s and 2000s (Biodiversity Center of Japan 2004; Hokkaido Institute of Environmental Sciences 2006). Large-scale migratory behavior between northwest and southeast of Hokkaido may be restored in the near future if their habitat is not fragmented.

19 Seasonal Migration of Sika Deer on Hokkaido Island, Japan

253

In this chapter, we review the seasonal migration of sika deer on Hokkaido Island, Japan. Firstly, we consider the seasonal migration patterns and examined the factors affecting seasonal migration of the population in the eastern part of Hokkaido according to Sakuragi et al. (2003a, b) and Igota et al. (2004). Secondly, we discuss mechanisms and ecological significance of migration of sika deer.

Hokkaido Island Hokkaido (78,037 km2) is the northeastern-most island of Japan, which is located between latitude 41° and 45° N (Fig. 19.1). The island is mountainous and extensively forested (61% of total area), with cold, snowy winters and cool, humid summers. Average temperature was 7–10° C, annual precipitation was 820–1,470 mm, and snow cover remained for 50–150 days in 2006 (Hokkaido Government, http:// www.pref.hokkaido.lg.jp). It is surrounded by the Sea of Japan, the Pacific Ocean, and the Sea of Okhotsk (Fig. 19.1). Since a warm current (the Tsushima Current) runs through the Sea of Japan and a cold current (the Kuril Current) along the coast of Pacific, it is warm in the western part of Hokkaido and cool in the eastern part in summer. In winter, the snow accumulation in the eastern part is less than that in the western part, because the prevailing westerly wind is obstructed by the high mountains area in the central part of Hokkaido.

Fig. 19.1 Map of Hokkaido Island.

254

H. Igota et al.

Study Area for Radio-Telemetry The study area was approximately 7,500 km2 in eastern Hokkaido, Japan. The capture site (43°13′ N, 143°53′ E; Fig. 19.2) was located along the upper stream (200 m elevation) of the Shoro River in the Shiranuka Hills, which is a representative wintering area of sika deer in eastern Hokkaido (Hokkaido Institute of Environmental Sciences 1995). The mean annual precipitation was 1,399 mm and the mean annual temperature at the capture site was 5.2° C with monthly means of −8.2° C in February and 18.6° C in August (National Land Agency of Japan 1992). Snow cover persists from December to late March or April. Mean snow depth in February in the wintering area in the study years varied between 34 and 80 cm. The major vegetation of the study area is mixed forest with evergreen conifers and deciduous broad-leaved trees (Tatewaki 1958; Fig. 19.3). The dominant tree species are Acer mono (painted maple), Tilia japonica (linden), Quercus crispula (white oak), and Abies sachalinensis (Sakhalin fir), with undergrowths of bamboo grasses Sasa senanensis (Ss), S. nipponica (Sn), S. kurilensis (Sk), and S. borealis (Sb) (Toyooka et al. 1983). The areas higher than 700 m are covered by evergreen coniferous forests of Abies sachalinensis, Picea jezoensis (Yeddo spruce), and P. glehnii (Sakhalin spruce). The areas lower than 200 m are covered by deciduous broad-leaved forests of Acer mono, T. japonica, Q. crispula, and Q. dentata (Daimyo oak). Plantation forests composed of Abies sachalinensis and P. glehnii are scattered in the mountainous area. The eastern part of the study area is flat lowland (<200 m elevation), dominated by pastures. In this part, the stream courses are

Fig. 19.2 Map of the study area in eastern Hokkaido, Japan. The radio-collared sika deer were captured on the Shiranuka Hills wintering area (´).

19 Seasonal Migration of Sika Deer on Hokkaido Island, Japan

255

Fig. 19.3 The vegetation map of the study area in eastern Hokkaido, which is also the area of captures.

covered by deciduous broad-leaved forests, and plantation forests mainly composed of the deciduous conifer Larix leptolepis (Japanese larch) are also scattered in some parts. The population of sika deer in eastern Hokkaido drastically increased during the 1980s, peaked around 1995 and, thereafter, gradually decreased owing to relaxation of hunting regulations (and consequent increase in hunting kill) and an increase of nuisance control kill with firearms to reduce serious damage to agriculture and forestry (Kaji et al. 2000; Uno et al. 2006). Brown bears (Ursus arctos) and domestic dogs (Canis familiaris) also are probable predators on sika deer.

Capture and Radio-Tracking A total of 60 female deer were captured between 1997 and 2000 (Igota et al. 2004). Deer were drug immobilized, weighed, blood sampled, and equipped with radiocollars and ear tags (Hamasaki 1998; Tsuruga et al. 1999). Three of the 60 deer captured died just before tracking commenced. The remaining 57 deer were categorized as fawns (n = 11), yearlings (n = 6), and adults (≥2 years old) (n = 40) according to tooth replacement (Ohtaishi 1980). Radio-collars had an estimated three-year battery life and were equipped with motion-sensitive mortality sensors. Radiocollared deer were relocated by ground triangulation (White and Garrott 1990) or visual observations. Tracking was carried out within every seven days for each deer. When deer were missed on the ground, aerial tracking was conducted by airplane following the methods of Mech (1983).

256

H. Igota et al.

Data Analysis We calculated the geographic centers of activity (COA) as the average UTM of locations of an individual deer within each seasonal home range (Hayne 1949). The characteristics of the seasonal home ranges (elevation, snow depth, bamboo grass forage value, coniferous cover ratio, winter temperature, and southern slope ratio) at the individual-landscape level were calculated for each “COA grid (a 1-km grid that included an individual seasonal COA).” Elevation was determined as the mean elevation of 50-m grids within a COA grid (National Land Agency of Japan 1992). Southern slope ratio (%) was determined by calculating the proportion of 50-m grids of slope aspect of 135–225° in a COA grid (National Land Agency of Japan 1992). Snow depth was determined as the average during the 30-year period (1955–1984) of the mean snow depth in February in a COA grid (National Land Agency of Japan 1992). Winter temperature was determined as the average during the 30-year period (1955–1984) of the minimum temperature in February in a COA grid (National Land Agency of Japan 1992). Bamboo grass species are important components in winter diet for sika deer in eastern Hokkaido (Yokoyama et al. 2000). Bamboo grass forage value was determined following the methods of Kaji et al. (2000). Bamboo grass species were ranked in the order: Sn > Ss > Sb > Sk based on availability, grazing tolerance, and nutritional value. The values (scores in parentheses) were none (0), Sk(1), Sk + Ss (2), Sb or Sn + Sb or Ss + Sb (3), Ss (4), Ss + Sn (5), and Sn (6). Coniferous cover ratio (%) was determined by calculating the proportion of 30-m grids of coniferous forest or mixed forests in a COA grid (Environment Agency of Japan 1997). Site fidelities to the seasonal home ranges were assessed by whether individual deer summer or winter home ranges overlapped or not in successive years, and by calculating the distances between individual COAs in successive years for summer and winter.

Migration Information We collected 7,765 location points between April 1997 and May 2002 (Igota et al. 2004). Monitored duration for each individual averaged 714 ± 67 (SE) days (range = 20–1,877). Ten (18%) of 57 radio-collared deer were upward migrants (UM) that had summer home ranges lower than or at similar elevation as their winter home ranges, 29 (51%) were downward migrants (DM) that had the summer home ranges higher than their winter home ranges, and 12 (21%) were nonmigrants (NM) that had overlapping seasonal home ranges (Table 19.1). The remaining 6 deer (11%) indicated migratory behaviors but died or were missed due to radio-collar troubles before their summer home ranges were determined. Upward migrants and DMs migrated from winter home ranges to summer home ranges between March and April, and migrated from summer home ranges to winter

Upward migrants Downward migrants Non-migrants Unknownb

Migration direction East 9, West 1 North 25, East 4 – –

Mean ± SE 58.7 ± 8.9 27.0 ± 2.3 – –

Range 18.4–101.7 7.2–53.4 – –

125 ± 25 (9) 123 ± 29 (19) 66 ± 4 (11) –

Mean ± SE (n) 26–264 19–602 48–94 –

Range

Summer home range size (ha)a

74 ± 19 (9) 107 ± 17 (14) 197 ± 36 (7) –

Mean ± SE (n)

b

9–163 27–230 69–353 –

Range

Winter home range size (ha)a

Seasonal home range sizes were calculated when deer were located more than 30 times in a season. They showed migratory behavior but were missed due to radiocollar failures or died before their summer ranges were reached.

a

N

10 29 12 6

Migratory type

Migration distance (km)

Table 19.1 Migration patterns of upward migrant, downward migrant, and nonmigrant sika deer in eastern Hokkaido, Japan (Igota et al. 2004).

19 Seasonal Migration of Sika Deer on Hokkaido Island, Japan 257

258

H. Igota et al.

0

20

40

60

80

100 km

Elevation (m) 0 – 100 100 – 200 200 – 300 300 – 400 400 – 500 500 – 600 600 – 700 700 – 800 800 – 900 900 – 1000 1000 – 1100 1100 – 1200 1200 <

Fig. 19.4 Distribution of individual summer home ranges (n = 51) of radio-collared sika deer in eastern Hokkaido, Japan, 1997–2001. Blank squares: the summer home ranges of the downward migrants (n = 29), solid squares: the summer home ranges of the upward migrants (n = 10), blank triangles: the summer home ranges of non-migrants (n = 12) (Igota et al. 2004).

home ranges between October and January. The summer home ranges of UMs and DMs were widely scattered over a total area of 5,734 km2 (Fig. 19.4), while the winter home ranges were concentrated in area of 821 km2 in the vicinity of the capture site. Migration direction of most UMs (90%) was east, whereas most DMs (86%) moved north (Fig. 19.4, Table 19.1). Migration distance of all migrants averaged 35.1 ± 3.6 km (mean ± SE, n = 39, range = 7.2–101.7 km). UMs migrated significantly greater distances than did DMs (p = 0.001; Fig. 19.1, Table 19.1). Mean seasonal home range sizes of the three migratory types varied from 66 to 125 ha in summer and 74–197 ha in winter (Table 19.1). Igota et al. (2004) monitored 33 deer in two or more summers and 24 deer in two or more winters. As the proportion of individuals with overlapping and nonoverlapping seasonal home ranges did not differ among the three migratory types in summer or winter (p > 0.05), they were pooled for each season. Thirty-two (97%) of 33 deer had overlapping summer home ranges in successive years. Seventeen (71%) of 24 deer had overlapping winter home ranges in successive years. The proportions of individuals with overlapping or nonoverlapping seasonal home ranges differed between summer and winter (p = 0.005; Fig. 19.5). As the distance between individual COAs in successive years did not differ among three migratory types in summer or winter (p > 0.05), they were pooled for each season. The distance between individual COAs in two successive years in summer (mean ± SE = 814 ± 626 m, n = 58) were significantly shorter than in winter (mean ± SE = 2,340 ± 768 m, n = 42; p = 0.008; Fig. 19.6). The winter home range characteristics and migration distances

19 Seasonal Migration of Sika Deer on Hokkaido Island, Japan

259

Fig. 19.5 The proportion of individuals with overlapping seasonal home ranges in successive years for sika deer in eastern Hokkaido, Japan, 1997–2001. As the proportion of individuals with overlapping and nonoverlapping seasonal home ranges did not differ among the three migration types in summer or winter, the bars were pooled for each season (Igota et al. 2004).

Fig. 19.6 The distance between individual geographic centers of activity (COAs) in two successive years in summer and winter. The bars indicate SE. As the distance between individual COAs in successive years did not differ among three migration types in summer or winter, they were pooled for each season (Igota et al. 2004).

of the individuals with nonoverlapping winter ranges (n = 7) did not vary among years (p > 0.05). The summer home ranges of UMs were at significantly lower elevations than their winter home ranges, whereas DMs had the summer home ranges at significantly higher elevations than their winter home ranges (Table 19.2). The summer and winter home ranges of NMs did not vary in elevations (Table 19.2). The elevations in summer home range differed among three migratory types (p < 0.0001), whereas the elevations of winter home range were similar among three migratory types (p = 0.17). The snow depths in the summer home ranges of UMs were significantly higher than in their winter home ranges (Table 19.2). Although only one UM migrated to a winter home range with deeper snow than its summer home range, the difference was only 1 cm. The snow depths in the summer home ranges of DMs were significantly higher than in their winter home ranges (Table 19.2). No DM migrated to a winter home range with deeper snow than its summer home range. There was no significant difference in snow depth between the seasonal home ranges of NMs (Table 19.2). The snow depths in summer home range differed among three migratory types (p < 0.0001), and those of DMs were significantly higher than of NMs or UMs (p < 0.0167). The snow depths in winter home range were similar among three migratory types (p = 0.58). According to the weather data of Japan Meteorological Agency, the snow depths during November and May in Naka-Shibetsu in the eastern part of the study area (see Fig. 19.2 for locations) were higher than those

260

H. Igota et al.

Table 19.2 Comparison of the characteristics (mean ± SE) of the seasonal home ranges of upward migrants (UM, n = 10), downward migrants (DM, n = 29), and nonmigrants (NM, n = 12) for sika deer in eastern Hokkaido, Japan (Igota et al. 2004). Summer

Wilcoxon’s signed-ranks test

Winter

Characteristics

Typea

Mean

SE

Mean

SE

Z

P

Elevation (m)

UM DM NM UM DM NM UM

137 489 242 67 93 55 3.2

26 26 6 4 3 1 0.9

250 262 241 53 53 55 6.0

11 7 5 2 1 1 0.0

−2.65 −4.70 −0.37 −2.71 −4.70 −0.58 −2.27

0.008 <0.0001 0.72 0.007 <0.0001 0.56 0.02

DM NM UM

4.1 6.0 5

0.3 0.0 4

6.0 6.0 65

0.0 0.0 7

−4.63

<0.0001

b

b

−2.81

0.005

DM NM UM

84 67 −15.4

6 5 0.3

67 69 −16.1

4 5 0.2

−2.87 −0.73 −1.54

0.004 0.47 0.12

DM NM UM

−17.5 −16.0 28

0.2 0.2 3

−16.2 −16.1 24

0.1 0.2 4

−4.35 −1.60 −1.38

<0.0001 0.11 0.17

DM NM

20 26

−1.96 −1.46

0.05 0.14

Snow depth (cm)

Bamboo grass forage value

Coniferous cover ratio (%)

Winter temperature (°C)

Southern slope ratio (%)

2 12

26 24

2 12

a

Tests for differences between summer and winter ranges. Statistical comparison could not be done because the bamboo grass values of all seasonal home ranges of NMs were the same.

b

of Naka-Teshibetsu in the Shiranuka Hills (p = 0.03), although the former (50 m) is in lower elevation than the latter (80 m). The bamboo grass forage values in the summer home ranges of both UMs and DMs were significantly lower than in their winter home ranges, respectively (Table 19.2). Statistical comparison could not be done because the bamboo grass forage values of all seasonal home ranges of NMs were the same (Table 19.2). The bamboo grass forage values in summer home ranges differed among three migratory types (p < 0.0001) and those of NMs were significantly higher than those of UMs or DMs (p < 0.0167). All of the bamboo grass forage values in winter home ranges of three migratory types were equivalent (n = 6). The coniferous cover ratios in the summer home ranges of UMs were significantly lower than in their winter home ranges (Table 19.2). In contrast, the coniferous cover ratios of the summer home ranges of DMs were significantly higher than in their winter home ranges (Table 19.2). There was no significant difference in the coniferous cover ratios between the seasonal home ranges of NMs (Table 19.2).

19 Seasonal Migration of Sika Deer on Hokkaido Island, Japan

261

The coniferous cover ratios in summer home ranges differed among three migratory types (p < 0.0001) and those of UMs were significantly lower than of NMs or DMs (p < 0.0167). The coniferous cover ratios in winter home ranges were similar among three migratory types (p = 0.95). There was no significant difference in the winter temperatures between the seasonal home ranges of UMs (Table 19.2), but most (seven of 10 deer) migrated to the winter home ranges 0.1–2.6 °C colder than their summer home ranges. The winter temperatures in the summer home ranges of DMs were significantly lower than in their winter home ranges (Table 19.2). There was no significant difference in the winter temperatures between the seasonal home ranges of NMs (Table 19.2). The winter temperatures in summer home ranges differed among three migratory types (p < 0.0001), and those of DMs were significantly lower than of NMs or UMs (p < 0.0167). The winter temperatures in winter home ranges were similar among three migratory types (p = 0.98). There was no significant difference in the southern slope ratios between the seasonal home ranges of UMs (Table 19.2). The southern slope ratios in the summer home ranges of DMs were significantly lower than in their winter home ranges (Table 19.2). There was no significant difference in the southern slope ratios between the seasonal home ranges of NMs (Table 19.2). The southern slope ratios in summer home ranges were similar among three migratory types (p = 0.09), as also were the winter home ranges (p = 0.71).

Discussion Migration Patterns The reverse altitudinal migration exhibited by the UMs has not been previously reported in sika deer or other cervid species. Our previous study on the seasonal habitat selection based on the population-landscape level (i.e., pooling all of three migratory types), revealed that the winter habitat selection might be determined by snow depth and amount of conifer cover (Sakuragi et al. 2003a). Large-scale movement. The migratory deer, especially the UMs, moved repeatedly over long distances between discrete seasonal home ranges, although they utilized relatively confined areas during summer or winter. The summer home ranges of all individuals were widely scattered over eastern Hokkaido (Fig. 19.4). Individuals from a different wintering herd adjacent to the capture site also migrated over great distances (mean = 20 km, maximum = 42 km; Uno and Kaji 2000) between the Lake Akan wintering area and summer ranges scattered in Abashiri subprefecture (Fig. 19.7). Thus, the scale of seasonal migration of sika deer in eastern Hokkaido was large compared with other migratory populations of sika deer on mainland (Honshu), Japan (e.g., Miura 1974; Maruyama 1981; Honma 1995; Takatsuki et al. 2000; Yabe and Takatsuki chapter 20).

262

H. Igota et al.

Since the human population in Japan has increased four-fold during the last 150 years and is especially high in Honshu, the habitat of sika deer has became fragmented, and the distribution area of deer reduced in Honshu (Tokida 2006). On the other hand, the summering areas are extensive and more contiguous in eastern Hokkaido (Kaneko et al. 1998). The deer population size is relatively large and they

Fig. 19.7 Seasonal migrations of female sika deer captured on Akankohan in April 1993. Capture and release site (n); center of activity for winter ranges (®); center of activity for summer ranges (l); location points from aerial tracking in wintering areas (*); location points from aerial tracking in summer ranges (°) (Uno and Kaji 2000).

19 Seasonal Migration of Sika Deer on Hokkaido Island, Japan

263

are distributed across almost the entire area of the island (Kaji et al. 2000). Nevertheless, even in Hokkaido, the potential wintering area for sika deer is restricted to a few constrained areas because of heavy snow and limited coniferous forests (Kaneko et al. 1998; Sakuragi et al. 2003a). Therefore, sika deer in eastern Hokkaido are forced to make large-scale seasonal migrations. Site fidelity to seasonal home ranges. Deer rarely changed their summer home ranges no matter which migratory type they belonged to. High site fidelities to summer home ranges have been reported for many cervid populations (Schoen and Kirchhoff 1985; Tierson et al. 1985; Brown 1992; Smith and Robbins 1994; Uno and Kaji 2000). As female deer give birth and nurse offspring in their summer ranges (Adams 1982), use of familiar sites can be advantageous for them. In our study only one two-year-old female deer changed her summer home range to 36.5 km away from the previous range. She used the new range for at least two years after the change. Her summer home range shift between yearling and two years of age appears to be a dispersal movement from her natal range. For whitetailed deer, fawns usually remain in their natal ranges at least until the beginning of their second summer (Nelson 1998), with dispersal occurring among yearlings or two-year-olds (Nelson and Mech 1992). Site fidelity to winter home ranges was strong but relatively lower than for summer home ranges, as found in previous studies (Schoen and Kirchhoff 1985; Tierson et al. 1985; Brown 1992; Uno and Kaji 2000). A part of a mule deer population remained outside the defined wintering area during mild winters (Brown 1992). In contrast, winter storms forced black-tailed deer to temporarily leave their normal winter home ranges and to inhabit a limited area in lower elevation (Loft et al. 1984). However, we could not find meaningful annual variation in the winter home range characteristics or migration distances of the individuals that shifted their winter home ranges, although the winter weather conditions varied among years during the study. Note, however, we do not suggest that the weather conditions are unimportant to deer winter ecology. In order to examine the factors affecting winter home range shifts, it may be necessary to monitor the relationship between winter range shifts and environmental conditions for a longer time, or for a larger sample of deer. Non-migratory strategy. The NM deer were sedentary at the wintering area in intermediate elevations with little snow, sufficient coniferous cover, and high quality of bamboo grass. The NMs may adopt a resident strategy, as migration is unnecessary if they spent summer in areas where wintering is possible. The NM strategy, existing along side of the migratory strategy in a single population, also has been reported for other cervid populations (Larson et al. 1978; Kufeld et al. 1989; Nicholson et al. 1997; Ball et al. 2001). Kufeld et al. (1989) suggested that the proportion of nonmigrants in a population was greatest in habitat of sufficient quality to support deer year-round. Timing of fall migration. Several studies have found that snow cover, rather than temperature decline, is the factor affecting onset of fall migration (Tierson et al. 1985; Smith and Robbins 1994). White-tailed deer begin to migrate from summer

264

H. Igota et al.

home range as snow depths reach around 38 cm (Tierson et al. 1985). In our study, however, most deer departed from their summer home ranges before the first snowfall or before the first cover of snow, whereas the remainder departed after the first cover of snow but before snow depth was great (Igota et al. 1997–2001; Fig. 19.8). We suggest that snow cover is not the factor initiating fall migration in this population. With increasing depth of snow, forage availability declines, and the energetic costs of foraging and locomotion increases (Parker et al. 1984). Indeed, all deer departed from their summer home ranges much earlier than the date that the snow depth in summer home range reached its peak in each year. On the other hand, most deer arrived at their winter home ranges before the first cover of snow or when only a shallow cover of snow was present. All deer arrived at their winter home ranges before the snow cover in the wintering area reach its maximum depth. Thus, most deer departed from their summer home ranges well before snow cover was established there, and arrived at their winter home ranges when there was no or only shallow snow.

Fig. 19.8 Timing of seasonal migrations of sika deer in eastern Hokkaido, Japan, in fall 1997/98–2000/01 and spring 1998–2001. The boxes and the vertical lines indicate the 25–75% and the 10–90% ranges of the samples, respectively (solid box: departure from seasonal range; blank box: arrival to another seasonal range). The horizontal lines in the boxes show the medians. The circles indicate the values out of 10–90% ranges. The numbers show the sample sizes (Igota et al. unpublished data).

19 Seasonal Migration of Sika Deer on Hokkaido Island, Japan

265

Deer show high site fidelity to both their seasonal home ranges (Schoen and Kirchhoff 1985; Tierson et al. 1985; Brown 1992; Uno and Kaji 2000) and their migratory routes (Sakuragi et al. 2004). That is, deer exhibit a regular round-trip movement between seasonal home ranges (White and Garrott 1990). The migratory patterns tend to be transferred from mothers to their fawns via learning (Adams 1982; Nelson 1998). Memory of distance and direction between seasonal home ranges was suggested to be part of the navigation process for seasonal migration (Nelson 1994). Thus, learning and memory are the important elements for the migratory behavior of deer. Individual sika deer may decide the timing of the onset of fall migration based on memory of the migration distance and may practice yearly migrations, stimulated by the weather conditions and/or other factors (e.g., photoperiod: Garrott et al. 1987; relative humidity: McCullough 1964). Timing of spring migration. The timing of departure from winter range corresponded to the timing of snow melt (Igota et al. 1998–2001). This suggests that snow melt triggered the onset of spring migration. Previous studies also suggested that timing of spring migration coincided with snow melt (Rongstad and Tester 1969; Hoskinson and Mech 1976; Schoen and Kirchhoff 1985; Uno and Kaji 2000). The onset of spring migration occurred when snow depth became ≤30 cm, despite the great variation of annual maximum snow depth of the wintering area, and most deer showed high synchrony in the departure date. In white-tailed deer, departure from winter home range was also associated with snow depth <38 cm (Tierson et al. 1985). Tracked deer were concentrated into the wintering area in the Shiranuka Hills where the population density became extremely high (Hokkaido Institute of Environmental Sciences 1995). Under high deer density, the availability and quality of forage markedly decreases because of strong foraging pressure (Kaji et al. 1988). The increased metabolic rate of deer during spring (Silver et al. 1971), coupled with their relatively poor nutritional condition, would be powerful stimuli to disperse to new food resources. Especially females would have to get back to their summer home ranges well before the impending birth of their fawns (Nelson and Mech 1981). Therefore, when the restriction of deer activities is released by snow melt, deer may depart from the wintering area to avoid a high-density condition. Thus, the timing of departure may show close synchrony despite variation in the environmental characteristics in the individual’s summer home range.

Factors Affecting Seasonal Migration We compared the seasonal home range characteristics between summer and winter for each migratory type, based on the individual-landscape level. The UMs wintered in areas of less snow, higher quality of bamboo grass, and more coniferous cover than their summer home ranges. Whereas, the DMs wintered in areas of less snow, higher quality of bamboo grass, higher winter temperature, and more southern slopes, but less coniferous cover than their summer home ranges. The NMs used

266

H. Igota et al.

year-round ranges with similar characteristics to the winter home ranges of both the UM and the DM migrants. Snow cover. Our data suggest that snow cover is an important factor affecting seasonal migration of sika deer in eastern Hokkaido. Ordinarily, movement to higher elevation in winter would seem unreasonable because, in most cases, snow deepens and the temperature decreases at higher elevations. However, we confirmed that the UMs used the winter home ranges of lower snow cover than their summer home ranges, although the former was higher in elevation than the latter. Their winter home ranges were located in the Shiranuka Hills, while their summer home ranges were located in the eastern part of the study area. The weather data showed that the Shiranuka Hills, facing the Pacific, was less snowy than the eastern part, facing Nemuro Straits. Thus, the Shiranuka Hills of less snow cover is one of the important wintering areas of sika deer in Hokkaido (Kaneko et al. 1998; Kaji et al. 2000). The activities of deer are physically restricted by snow and, therefore, accessing forage beneath the snow becomes difficult (Parker et al. 1984). Deer mortality is higher in years of heavy snow (Kaji et al. 1988; Takatsuki et al. 1994; Uno et al. 1998). Thus, snow cover and its impediment to deer locomotion may be the proximate factor driving seasonal migration (Verme 1973; McCullough 1985; Sakuragi et al. 2003a). Our data, regardless of migratory type, support this hypothesis. Bamboo grass. We also suggest that bamboo grass is an important factor affecting seasonal migration of this population. Bamboo grass forage value, as well as snow cover, was the important variable limiting the sika deer distribution across the island of Hokkaido (Kaji et al. 2000). Bamboo grass, especially S. nipponica, which lies beneath the snow, is an important winter dietary component for sika deer in eastern Hokkaido (Yokoyama et al. 2000), and is characterized by its abundance, stable availability (evergreen), and tolerance to grazing by sika deer (Takatsuki 1983, 1986). The dominant species of the Shiranuka Hills wintering area was S. nipponica. Limited availability of bamboo grass may strongly influence seasonal movement of sika deer. Sika deer in the Tanzawa Mountains concentrated at higher elevation in late autumn and only moved down in winter after the reduction of bamboo biomass, because bamboo grass occurred only in high mountains due to high density of the deer population (Borkowski and Furubayashi 1998). Although our previous study at the population-landscape level could not detect bamboo grass variety as a factor affecting seasonal migration of sika deer in eastern Hokkaido (Sakuragi et al. 2003a), this study at the individual-landscape level revealed that bamboo grass variety contributed to the migration patterns of both the UM and the DM migrants. Coniferous cover. In the case of the UMs, coniferous cover plays an important role in seasonal migration. Evergreen coniferous forest is important cover in wintering areas because of the smaller amount of the snow under the canopy (Bloom 1978; Wallmo and Schoen 1980; Sakuragi et al. 2003a), and thermal cover (Moen 1968; Ozoga 1968). The UMs obtained remarkably more coniferous cover by their ascendant movement. Most UMs had neither coniferous nor mixed forest in their summer home ranges. As the eastern part of the study area where deer summered was low and flat (see Fig. 19.3), humans have facilitated colonization of these areas

19 Seasonal Migration of Sika Deer on Hokkaido Island, Japan

267

over the last 100 years by replacing wilderness composed of mainly deciduous broad-leaved forest with pastures and plantations of deciduous conifers (Larix leptolepis) (Kaneko et al. 1998). In contrast, the DMs moved to less conifer cover by the descendent movement. However, the winter home ranges of all DMs included more than 28% conifer cover, which can provide sufficient shelter for deer from heavy snow and storms. Vegetation cover provides deer with not only habitat with less snow but also shelter from cold temperatures, and concealment from predation and hunting (Hamilton et al. 1980; Nelson and Mech 1981; Takatsuki 1986). Winter temperature. Winter temperature might not necessarily be a factor affecting seasonal migration of the population. The winter home ranges of the DMs were warmer than their summer home ranges because of elevational difference. Migration to warmer areas may reduce their energy expenditure under severe winter conditions (Silver et al. 1971). However, most of the UMs used winter home ranges colder than their summer home ranges, although the difference was not statistically significant. This implies that winter temperature may not contribute to, at least, seasonal migration of the UMs. In areas of cold winter temperature, but limited or no snow, deer may occupy the same range year-round (Tierson et al. 1985). Slope aspect. Slope aspect also might not necessarily be a factor affecting seasonal migration of the population. Snow depth on southern slopes is considered to be lower than on northern slopes due to greater solar radiation. The DMs may use habitat with less snow by migration to areas with more southern slope. Half of them summered in the Okhotsk Sea side of the Shiranuka Hills. The Okhotsk side facing the north is less abundant in southern slopes than the Pacific side facing the south. However, the amount of southern slope was similar between the seasonal home ranges of the UMs. Therefore, slope aspect could not contribute to their migration. The UMs, however, may have enough southern slope within their winter home ranges because proportions of southern slope in the winter home range were similar between the UM and the DM migrants. In conclusion, we suggest that snow cover and bamboo grass are the factors affecting seasonal migration of sika deer in eastern Hokkaido. In addition, coniferous cover can be another important factor in the case of the reverse altitudinal migration by the UMs. Winter temperature and slope aspect may play only minor roles because they did not contribute to the migration patterns of the UMs.

Why Three Migration Strategies? The wintering herd in the Shiranuka Hills originated from the population that survived the bottleneck of 130 years ago (Kaji et al. 2000). The Shiranuka Hills is considered to be a wintering area where deer survived even in a year of the deepest snow depth in a 30-year period (Kaneko et al. 1998). The sika deer surviving in this refugium may have developed into the three migration types during the process of their expanding distribution (Kaji et al. 2000; Sakuragi et al. 2003a).

268

H. Igota et al.

After the historical population bottleneck, some sika deer in the Shiranuka Hills might have dispersed to the eastern lowlands with their large amount of pastures of high quality forage in summer. However, deer could not winter in these areas with more snow, lower quality of bamboo grass, and little or no conifer cover. Therefore, the UMs had to migrate toward the Shiranuka Hills in winter. This accounts for the reverse altitudinal migration. Both the DM and the UM probably developed during the process of the expanding distribution of the population over several decades after the bottleneck. These migration strategies may provide each of the migration types with the following ecological significance. Nicholson et al. (1997) reported that trade-offs about whether or not deer migrate exist for mule deer in mountainous southern California. Migrants used habitats of higher quality than the nonmigrants, whereas migrants were at increased risk of predation during migration and, in years of low precipitation, had higher mortality than the nonmigrants (Nicholson et al. 1997). This suggested that in areas with extremely variable precipitation, a mixed strategy for migration could be maintained. We suggest there is a similar trade-off in sika deer too, despite the recent development of migratory behavior of deer in eastern Hokkaido. Our previous study based on analysis of fecal nitrogen content in this population suggested that the DMs fed on a higher quality diet than do the nonmigrants and the UMs in summer (Sakuragi et al. 2003b). The DMs had higher mortality rate than did the NMs and the UMs (Igota et al. in preparation). Thus, these results indicate that the DMs have risk of migration in compensation for a higher quality food whereas the nonmigrants are free from risk of migration but feed on a lower quality food in summer. The trade-offs may exist between these two types. However, the UMs did not appear to include both risk of migration and rich food by migrating. Firstly, when the DMs inhabited the northern woodland of high altitude in summer, they acquired relatively high quality food but the risk of migration. The quality of food was positively correlated to elevation in the population (Sakuragi et al. 2003b). Migration to higher elevation in summer, where onset of plant growth is later, prolongs access to a diet high in protein and low in fiber (Albon and Langvatn 1992). Furthermore, in mountainous areas, changes in microclimate associated with topographical roughness will increase the variability of snow melt and produce patches of highly digestible forage over a prolonged period (Albon and Langvatn 1992). On the other hand, by downward migration in eastern Hokkaido they obtained less snow, higher quality bamboo grass, higher winter temperature, more southern slopes and less but sufficient coniferous cover during winter (Igota et al. 2004). Thus, such a descendent altitudinal migration, similar to the ordinary patterns reported previously (McCullough 1985), provided the DMs with a rich winter habitat. Secondly, when the NMs adopted the sedentary strategy in the wintering area of intermediate elevation, they were free from risk of migration but acquired relatively low quality food (Sakuragi et al. 2003b). Their year-round home ranges were away from agricultural fields and roads of high traffic volume. However, because all migratory types used the same wintering area, the high browsing pressure caused deterioration of the habitat quality. It has been pointed out that habitat in a winter-

19 Seasonal Migration of Sika Deer on Hokkaido Island, Japan

269

ing area may be relatively poor because of increased browsing pressure at greater deer densities in winter (Sabine et al. 2002). Nonetheless, NMs annually used the wintering area with relatively little snow, high quality of bamboo grass and sufficient coniferous covers for wintering and avoided the potential risk of migrating (Igota et al. 2004). Thus, the nonmigrants may not receive the benefits of migration, but are compensated by no cost or risk of migration. Finally, when the UMs inhabited the eastern open grassland with low altitude in summer, they had lower risk of migration but did not acquire a higher quality food despite the long-distance migration and closeness to agricultural fields. They may achieve higher survival by avoiding attractive agricultural fields which have the greatest potential risk. They may increase their alert behavior against human disturbance in their summer habitat which has less vegetation cover for concealment. On the other hand, they migrated to an area of higher elevation in winter than their summer home ranges; nevertheless, they obtained less snow, higher quality of bamboo grass, and more coniferous cover (Igota et al. 2004). Thus, although such a upward altitudinal migration is different from the ordinary patterns reported previously (McCullough 1985) and seems unreasonable at first, such an ascendant altitudinal migration actually provides the UMs with a rich winter habitat. The numbers of both the UMs and NMs captured were less than half of the DMs when sampled in the wintering area. Assuming that capture was random, this indicates that DMs predominated in the population. A possible reason is that carrying capacity or population growth rate during the process of expanding distribution differs among the three types. The north part of the study area probably had a high carrying capacity for summer range of deer because deer habitat become richer at higher elevation (Albon and Langvatn 1992; Sakuragi et al. 2003b), resulting in the increase of the DM population. Whereas, NMs may have been maintained in a low population size probably owing to relatively confined living area throughout years. In the case of the UMs, although they dispersed into the widest range (Fig. 19.4) with a great amount of pasture (Fig. 19.3), they did not feed on a higher quality diet despite its potential availability, probably because of human disturbance (Sakuragi et al. 2003b). However, to better test the above hypotheses, we need to evaluate the cost and benefit of migration comprehensively, examining other life-history parameters, such as lifetime reproductive success, body condition, and so on (CluttonBrock et al. 1982).

Literature Cited Adams, A. W. 1982. Migration. Pages 301–321 in J. W. Thomas and D. E. Toweill, editors, Elk of North America: Ecology and management. Stackpole Books, Harrisburg, Pennsylvania, USA. Albon, S. D., and R. Langvatn. 1992. Plant phenology and the benefit of migration in a temperate ungulate. Oikos 65:502–513. Baker, R. R. 1978. The evolutionary ecology of animal migration. Hodder & Stoughton, London, United Kingdom.

270

H. Igota et al.

Ball, J. P., C. Nordengren, and K. Wallin. 2001. Partial migration by large ungulates: Characteristics of seasonal moose Alces alces ranges in northern Sweden. Wildlife Biology 7:39–47. Biodiversity Center of Japan. 2004. The national survey on the natural environment, report of the distributional survey of Japanese animals (Mammals). Japan Wildlife Research Center, Tokyo, Japan. (In Japanese.) Bloom, A. M. 1978. Sitka black-tailed deer winter range in the Kadashan Bay area, Southeast Alaska. Journal of Wildlife Management 42:108–112. Borkowski, J., and K. Furubayashi. 1998. Seasonal changes in number and habitat use of foraging sika deer at the high altitude of Tanzawa Mountains, Japan. Acta Theriologica 43: 95–106. Brown, C. G. 1992. Movement and migration patterns of mule deer in southeastern Idaho. Journal of Wildlife Management 56:246–253. Clutton-Brock, T. H., F. E. Guiness, and D. Albon. 1982. Red deer: Behavior and biology of two sexes. University Chicago Press, Chicago, Illinois, USA. Cumming, H. G., and D. B. Beange. 1987. Dispersion and movements of woodland caribou near Lake Nipigon, Ontario. Journal of Wildlife Management 51:69–79. Environment Agency of Japan 1997. Natural environment information GIS. Nature Conservation Bureau, Tokyo, Japan. Garrott, R. A., G. C. White, R. M. Bartmann, L. H. Carpenter, and A. W. Alldredge. 1987. Movements of female mule deer in northwest Colorado. Journal of Wildlife Management 51:634–643. Hamasaki, S. 1998. Capture and immobilization of wildlife—Capturing methods for small and medium mammals. Journal of Veterinary Medicine 51:69–73. Hamilton, G. D., P. D. Drysdale, and D. L. Euler. 1980. Moose winter browsing patterns on clearcuttings in northern Ontario. Canadian Journal of Zoology 58:1412–1426. Hayne, D. W. 1949. Calculation of size of home range. Journal of Mammalogy 30:1–18. Histøl, T., and O. Hjeljord. 1993. Winter feeding strategies of migrating and nonmigrating moose. Canadian Journal of Zoology 71:1421–1428. Hokkaido Institute of Environmental Sciences. 1995. Results of the survey related to sika deer and brown bear in Hokkaido. Volume I. Hokkaido Institute of Environmental Sciences, Sapporo, Japan. (In Japanese.) Hokkaido Institute of Environmental Sciences. 2006. Reports on the survey of sika deer population in Hokkaido (1997–2004). Hokkaido Institute of Environmental Sciences, Sapporo, Japan. (In Japanese.) Honma, K. 1995. Movement patterns and habitat use of sika deer (Cervus nippon) in Oku-Nikko and Ashio area. Master’s thesis, Graduate School of Education, Joetsu University of Education, Niigata, Japan. (In Japanese.) Hoskinson, R. L., and L. D. Mech. 1976. White-tailed deer migration and its role in wolf predation. Journal of Wildlife Management 40:429–441. Igota, H., M. Sakuragi, H. Uno, K. Kaji, M. Kaneko, R. Akamatsu, and K. Maekawa. 2004. Seasonal migration patterns of female sika deer in eastern Hokkaido, Japan. Ecological Research 19:169–178. Inukai, T. 1952. The sika deer in Hokkaido and its rise and decline. Hoppo Bunka Kenkyu Hokoku (Report of North Culture Research) 7:1–45. (In Japanese.) Kaji, K., T. Koizumi, and N. Ohtaishi. 1988. Effects of resource limitation on the physical and reproductive condition on sika deer on Nakanoshima Island, Hokkaido. Acta Theriologica 33:187–208. Kaji, K., M. Miyaki, T. Saitoh, S. Ono, and M. Kaneko. 2000. Spatial distribution of an expanding sika deer population on Hokkaido Island, Japan. Wildlife Society Bulletin 28:699–707. Kaneko, M., K. Kaji, and S. Ono. 1998. An analysis of the change of distribution accompanying the change of habitat of Hokkaido sika deer. Honyurui Kagaku (Mammalian Science) 38:49–59. (In Japanese.) Kufeld, R. C., D. C. Bowden, and D. L. Schrupp. 1989. Distribution and movements of female mule deer in the Rocky Mountain foothills. Journal of Wildlife Management 53:871–877. Larson, T. J., O. J. Rongstad, and F. W. Terbilcox. 1978. Movement and habitat use of white-tailed deer in southcentral Wisconsin. Journal of Wildlife Management 42:113–117.

19 Seasonal Migration of Sika Deer on Hokkaido Island, Japan

271

Loft, E. R., J. W. Menke, and T. S. Burton. 1984. Seasonal movement and summer habitats of female black-tailed deer. Journal of Wildlife Management 48:1317–1325. Maruyama, N. 1981. A study of the seasonal movements and aggregation patterns of sika deer. Bulletin of Faculty of Agriculture, Tokyo University of Agriculture and Technology 23:1–85. (In Japanese with English summary.) Matsuda, H., K. Kaji, H. Uno, H. Hirakawa, and T. Saitoh. 1999. A management policy for sika deer based on sex-specific hunting. Research on Population Ecology 41:139–149. McCullough, D. R. 1964. Relationship of weather to migratory movements of black-tailed deer. Ecology 45:249–256. McCullough, D. R. 1985. Long range movements of large terrestrial mammals. Pages 444–465 in M. A. Rankin, editor, Migration: Mechanisms and adaptive significance. Contributions in Marine Science (Supplement) Volume 27. Marine Science Institute, University of Texas at Austin, Port Aransas, Texas, USA. Mech, L.D. 1983. Handbook of animal radio-tracking. University of Minnesota Press, Minneapolis, Minnesota, USA. Miura, S. 1974. On the seasonal movements of sika deer population in Mt. Hinokiboramaru. Journal of the Mammalogical Society of Japan 6:51–62. (In Japanese with English summary.) Moen, A. N. 1968. Surface temperatures and radiant heat loss from white-tailed deer. Journal of Wildlife Management 32:338–344. Nagata, J., R. Masuda, K. Kaji, K. Kaneko, and C. Yoshida. 1998. Genetic variation and population structure of the Japanese sika deer (Cervus nippon) in Hokkaido Island, based on mitochondrial D-loop sequence. Molecular Ecology 7:871–877. National Land Agency of Japan. 1992. Outline of national land information system. National Land Information Office, Planning and Coordination Bureau, Tokyo, Japan. Nelson, M. E. 1994. Migration bearing and distance memory by translocated white-tailed deer, Odocoileus virginianus. The Canadian Field-Naturalist 108:74–76. Nelson, M. E. 1998. Development of migratory behavior in northern white-tailed deer. Canadian Journal of Zoology 76:426–432. Nelson, M. E., and L. D. Mech. 1981. Deer social organization and wolf predation in northeastern Minnesota. Wildlife Monographs 77:1–53. Nelson, M. E., and L. D. Mech. 1987. Demes within a northeastern Minnesota deer population. Pages 27–40 B. D. Chepko-Sade, and Z. Halpin, editors, Mammalian dispersal patterns. University of Chicago Press, Chicago, Illinois, USA. Nelson, M. E., and L. D. Mech. 1992. Dispersal in female white-tailed deer. Journal of Mammalogy 73:891–894. Nicholson, M. C., R. T. Bowyer and J. G. Kie. 1997. Habitat selection and survival of mule deer: Tradeoffs associated with migration. Journal of Mammalogy 78:483–504. Ohtaishi, N. 1980. Determination of sex, age and death-season of recovered remains of sika deer (Cervus nippon) by jaws and tooth-cement. Archaeology and Natural Science 53:13–17. (In Japanese.) Ozoga, J. J. 1968. Variations in microclimate in a conifer swamp deeryard in northern Michigan. Journal of Wildlife Management 32:574–585. Parker, K. L., C. T. Robbins, and T. A. Hanley. 1984. Energy expenditures for locomotion by mule deer and elk. Journal of Wildlife Management 48:474–488. Rongstad, O. J., and J. R. Tester. 1969. Movements and habitat use of white-tailed deer in Minnesota. Journal of Wildlife Management 33:366–379. Sabine, D. L., S. F. Morrison, H. A. Whitlaw, W. B. Ballard, G. J. Forbes, and J. Bowman. 2002. Migration behavior of white-tailed deer under varying winter climate regimes in New Brunswick. Journal of Wildlife Management 66:718–728. Sakuragi, M., H. Igota, H. Uno, K. Kaji, M. Kaneko, R. Akamatsu, and K. Maekawa. 2003a. Seasonal habitat selection of an expanding sika deer population in eastern Hokkaido, Japan. Wildlife Biology 9:109–121. Sakuragi, M., H. Igota, H. Uno, K. Kaji, M. Kaneko, R. Akamatsu, and K. Maekawa. 2003b. Benefit of migration in female sika deer population in eastern Hokkaido, Japan. Ecological Research 18:347–354.

272

H. Igota et al.

Sakuragi, M., H. Igota, H. Uno, K. Kaji, M. Kaneko, R. Akamatsu, and K. Maekawa. 2004. Female sika deer fidelity to migration route and seasonal ranges in eastern Hokkaido, Japan. Mammal Study 29:113–118. Schoen, J. W., and M. D. Kirchhoff. 1985. Seasonal distribution and home-range patterns of Sitka black-tailed deer on Admiralty Island, southeast Alaska. Journal of Wildlife Management 49:96–103. Silver, H., J. B. Holter, N. F. Colovos, and H. H. Hayes. 1971. Effect of falling temperature on heat production in fasting white-tailed deer. Journal of Wildlife Management 35:37–46. Smith, B. L., and R. L. Robbins. 1994. Migrations and management of the Jackson elk herd. U. S. Department of the Interior National Biological Survey Resource Publication 199. Takatsuki, S. 1983. The importance of Sasa nipponica as a forage for sika deer (Cervus nippon) in Omote-Nikko. Japanese Journal of Ecology 33:17–25. Takatsuki, S. 1986. Food habits of sika deer on Mt. Goyo, northern Honshu. Ecological Research 1:119–128. Takatsuki, S. 1992. Foot morphology and distribution of sika deer in relation to snow depth in Japan. Ecological Research 7:19–23. Takatsuki, S., K. Suzuki, and I. Suzuki. 1994. A mass-mortality of sika deer on Kinkazan Island, northern Japan. Ecological Research 9:215–223. Takatsuki, S., K. Suzuki, and H. Higashi. 2000. Seasonal elevational movements of sika deer on Mt. Goyo, northern Japan. Mammal Study 25:107–114. Tatewaki, M. 1958. Forest ecology of the island of the North Pacific Ocean. Journal of the Faculty of Agriculture of Hokkaido University 50:371–472. Tierson, W. C., G. F. Mattfeld, R. W. Sage, and D. F. Behrend. 1985. Seasonal movements and home-ranges of white-tailed deer in the Adirondacks. Journal of Wildlife Management 49:760–769. Tokida, K. 2006. Sika deer management in nature reserve area. Pages 20–37 in T. Yumoto and H. Matsuda, editors, Ecology for sika deer and forest. Bunichi Sogo Shuppan, Tokyo, Japan. (In Japanese.) Toyooka, K., M. Sato, and S. Ishizuka. 1983. The distribution map of Sasa group in Hokkaido. Explanatory Note. Hokkaido Branch, Forestry and Forest Products Research Institute, Sapporo, Japan. (In Japanese.) Tsuruga, H., M. Suzuki, H. Takahashi, K. Jinma, and K. Kaji. 1999. Immobilization of sika deer with medetomidine and ketamine, and antagonism by atipamezole. Journal of Wildlife Diseases 35:774–778. Uno, H., and K. Kaji. 2000. Seasonal movements of female sika deer in eastern Hokkaido, Japan. Mammal Study 25:45–57. Uno, H., M. Yokoyama, and M. Takahashi. 1998. Winter mortality pattern of sika deer (Cervus nippon yesoensis) in Akan National Park, Hokkaido. Mammalian Science 38:33–246 (In Japanese with English summary.) Uno, H., K. Kaji, T. Saitoh, H. Matsuda, H. Hirakawa, K. Yamamura and K. Tamada. 2006. Evaluation of relative density indices for sika deer in eastern Hokkaido, Japan. Ecological Research 21:624–632. Verme, L. J. 1973. Movements of white-tailed deer in Upper Michigan. Journal of Wildlife Management 37:545–552. Wallmo, O. C., and J. W. Schoen. 1980. Response of deer to secondary forest succession in southeast Alaska. Forest Science 26:448–462. White, C. G., and R. A. Garrott. 1990. Analysis of wildlife radio-tracking data. Academic Press, San Diego, California, USA. Yokoyama, M., K. Kaji, and M. Suzuki. 2000. Food habits of sika deer and nutritional value of sika deer diets in eastern Hokkaido, Japan. Ecological Research 15:345–355.

Chapter 20

Migratory and Sedentary Behavior Patterns of Sika Deer in Honshu and Kyushu, Japan Tsuneaki Yabe and Seiki Takatsuki

Abstract We review migration and home range use in Honshu and Kyushu. Sika deer in the northern areas showed seasonal up-down elevation movements (migration), though some sedentary deer were known. The home ranges often exceeded 1 km2. Annual home range sizes of sedentary deer and seasonal ranges of “shifting” animals in the southern populations did not exceed 1 km2. Home ranges of males were larger than those of females with some exceptions.

Introduction Cervids, as do other herbivorous mammals, select preferable habitats: productive plant communities for foraging or concealing habitats for escaping predators. In places showing great seasonal fluctuation in plant productivity, for example, cervids select habitats depending on the location of favorable conditions. In the temperate areas, cervids utilize forests during winter to avoid chilly winds. When such seasonality is intensified, the home ranges of cervids are often enlarged, and in extreme cases, the cervids utilize separate summer and winter ranges. This is termed migration. In snowy areas, since movements of cervids are restricted during winter, they migrate to avoid snow. The long-distance migration of caribou (Rangifer tarandus) is a typical example (Miller et al. 1977). In areas of steep topography or of great elevational grades, cervids show up-down migration movements (Maruyama 1981; Takatsuki et al. 2000). The distributional range of the sika deer in Japan extends from Hokkaido Island in the north to Kyushu in the south. Climate and vegetation range from the subboreal zone in Hokkaido to the subtropical zone in Kyushu. Migratory behavior of sika deer in Hokkaido is covered in Igota et al. chapter 19. Here we consider the sedentary or migratory behavior of sika deer in the more southerly islands, Honshu and Kyushu. The sika deer range in the northern and middle areas of Honshu is largely composed of cool-temperate deciduous broad-leaved forest, whereas in the rest of the southern part of Japan it is warm-temperate evergreen broad-leaved forest. In the former, foods for sika deer are quite rich in summer because of vigorous plant D. R. McCullough et al. (eds.), Sika Deer: Biology and Management of Native and Introduced Populations, DOI: 10.1007/978-4-431-09429-6_20, © Springer 2009

273

274

T. Yabe, S. Takatsuki

productivity in the hot and rainy climate, while foods are quite poor during winter, particularly in the snowy places. A unique feature of these forests is that they are densely covered by dwarf bamboos on the forest-floor, and sika deer can feed on the evergreen leaves even during winter as long as the snow is not too deep. The distribution of sika deer is biased to less snowy areas, due to the heavy load on their hooves which limits their movements in deep snow (Takatsuki 1992a). On the contrary, foods are generally poor in the southern evergreen forests because of the thick forest canopy. However, most of these forests are fragmented by human activities, and they are more or less accompanied by secondary grassland patches or crop fields, which afford abundant foods for the sika deer. Snow is rare in the latter and, consequently, winter foods are relatively abundant. Thus, the evergreen forests in the south show less seasonality in food availability than the deciduous forests in the north. It is therefore expected that such environmental differences would cause geographical differences in home range use of sika deer. There are a considerable number of studies on the home range use of sika deer, but no attempt has been made to look at them on an archipelago scale. The objective of this chapter is to review the home range use of sika deer, particularly migration, living in different habitats.

Methods The Japanese archipelago is usually divided into eight blocks: (A) Hokkaido, (B) Tohoku, (C) Kanto, (D) Chubu, (E) Kinki, (F); Chugoku, (G) Shikoku, and (H) Kyushu (Fig. 20.1). Besides these blocks, some small islands are inhabited by

A. Hokkaido

B. Tohoku

D. Chubu F. Chugoku

C. Kanto E. Kinki G. Shikoku H. Kyushu

Fig. 20.1 Regional blocks of Japan.

N 200km

20 Migratory and Sedentary Behavior Patterns of Sika Deer in Honshu

275

sika deer. With the exception of Hokkaido, we reviewed all studies of blocks with samples of four or more, and additional samples of less than four, radio-tracked sika deer on home range use and migration and compared the sizes of the home ranges, migration distances, and elevational differences of up-down movements.

Results In the Tohoku block, a population on Mt. Goyo, Iwate Prefecture, was studied (B1, Fig. 20.2; Takatsuki et al. 2000). The study area covered from 200 to 1,400 m in elevation. Most areas were covered by oak, Quercus serrata, forests, but areas higher than 1,200 m were covered by subalpine coniferous forests. Snow covered the area from December to March, and the higher elevations were covered by snow from early December through early May. Among nine radio-tracked sika deer, four showed up-down movements, of which elevational gaps were about 600 m, while other five were sedentary and stayed at the lower areas through the year; elevational differences were less than 200 m. It is also known that males are found several tens of kilometers away from their usual range during the rut (Takatsuki 1992b). In the Kanto block, six studies were done in five areas. In Ashio (C1, Fig. 20.2), all of the five tracked deer were sedentary (Satake et al. 1994). In Omote-Nikko Coniferous forest & sub-alpine vegetation Boreal mixed forest Deciduous broad-leaved forest Evergreen broad-leaved forest

B1 B2

C3 E4 E3

C2

F1 F2 D1 C4

H2 E2 E1 H1 H3

C1 C5

Migratory Mixed Sedentary

N 200km

Fig. 20.2 Vegetation and migratory-sedentary patterns of sika deer in Honshu and Kyushu Islands, Japan. (Vegetation map is based on Yoshioka 1973.)

276

T. Yabe, S. Takatsuki

(C2, Fig. 20.2) which is higher in elevation and consequently more snowy than Ashio, Maruyama (1981) determined the movement patterns of 23 sika deer. He focused on the overlap of the summer and winter ranges and categorized the deer populations into the overlap type, the segregate type, and the dispersal type. The overlap type population included five sedentary animals and four up-down movement animals. They stayed at around 1,600 m in elevation and showed some shortdistance movements. The segregate type deer showed a great up-down movement, staying at over 2,000 m in elevation in summer, moving down to 1,300 m in November, staying at around 1,600 m in spring, and then again climbing the mountain in summer. Only one deer showed this movement. Five young deer did not stay long at any place and always moved; these were categorized as the dispersal type. The other eight deer did not show a consistent pattern. In Oku-Nikko (C3, Fig. 20.2), which is even higher than Omote-Nikko, Honma (1985) determined the movements of ten deer. He showed that all ten deer were migratory. Migration distances were about 10 km. The above three studies showed that even within these three close areas (the distance between Ashio and Oku-Nikko was only 15 km), the populations showed different patterns: the Oku-Nikko population included only migratory deer, the Omote-Nikko population included both migratory and sedentary, and the Ashio population included only sedentary deer. Snow depth varies; Oku-Nikko is very snowy—as much as 2 m of snow accumulates, whereas Ashio is less snowy and many deer remained over the winter. Nagata (2005) radio-tracked three males and two females in a protected area in Tanzawa Mountains (C4, Fig. 20.2) for five to nine years. The study area covered 500–900 m in elevation where snow accumulation is less than 50 cm. The vegetation of the area consisted of deciduous broad-leaved forest and coniferous plantation. Supplemental feeding was done in winter. Although a female temporally shifted her home range, all deer were basically sedentary through the year with average annual home range sizes of 0.5–0.8 km2 for males and 0.2–2.3 km2 for females. More recently, it was found that four sika deer living outside the refuge moved across the boundary for about 10 km (Tanzawa-Ohyama Synthetic Survey Team 2006). Since this movement was affected by human activities, the deer are regarded as sedentary following the conclusion of Nagata (2005). In Boso Peninsula (C5, Fig. 20.2), four males and six females were studied (Shigematsu et al. 1995; Chiba Prefecture and Boso Deer Research Group 2004). The vegetation was composed of evergreen broad-leaved forest, coniferous plantation, and grassland patches at an elevation of 10–350 m. A dispersing young and two adult males were sedentary whereas one male seasonally moved for less than 1 km. All females were considered sedentary while two of them showed temporal movements for 2–3 km. Home range sizes were 0.7–1.4 km2 in males and 0.04–1.4 km2 in females. In Chubu block, central Japan, a male and a female were traced on the slope of Mt. Fuji (D1, Fig. 20.2; Jiang et al. 2006). The female moved horizontally in early winter but did not return to the original place. Her home range size was 12.4 km2 in fall but was reduced to 0.4 km2 or smaller from winter until summer. The male did

20 Migratory and Sedentary Behavior Patterns of Sika Deer in Honshu

277

not migrate from fall to the next spring. Since this district, which includes many other mountains, has the greatest elevational range in Japan, it is quite likely that the sika deer of this area show migration. In the Kinki block (E, Fig. 20.2), six studies were done in four areas. In Odaigahara Mountains (E1, Fig. 20.2), home range use of four males and six females was determined (Maeji et al. 2000; Yajima et al. 2002). The habitat was mountainous including a plateau of 1,600 m in elevation, covered by Abies firma (Japanese fir) and Picea jezoensis (Yeddo spruce) forests, Fagus crenata (Japanese beech) forests, and Sasa (dwarf bamboo) patches. The forest floor used to be covered by mosses, but these were replaced by Sasa nipponica following repeated typhoons that resulted in tree blow-downs, and Sasa patches have been spreading due to tree deaths caused by deer barking (Ando and Shibata chapter 15). Home range sizes of males and females were 2.1 km2 and 0.8–1.3 km2, respectively. Most were sedentary, though a few of the deer showed migratory tendency, covering the elevation ranges from 600 to 1,600 m (Maeji et al. 2000). Yoshida et al. (2007) traced four females in the same place by GPS collars and found that they stayed at 1,400–1,600 m in elevation from spring to fall, but descended to 650–1,100 m in the snowy season. Sika deer in Nara Park (E2, Fig. 20.2; Torii and Tatsuzawa chapter 25) are tame and receive supplementary food. Summer home ranges of the females were as small as 0.1–0.2 km2 (Miura 1984). Those of males were also small: about 0.1 km2 during the rut and 0.8 km2 outside of the rut. Males moved among groups for social reasons, but they did not move for environmental reasons, which means that they were sedentary. Two GPS-collared females were tracked in an area ranging from 250 to 500 m, covered by conifer plantation and cultivated land, in Osaka Prefecture (E3, Fig. 20.2; Kawai et al. 2006). They were sedentary and annual home range sizes were 0.2 and 0.4 km2, respectively. Daily home range sizes of five males and two females in Hyogo Prefecture (E4, Fig. 20.2) living in the habitat covered by conifer plantation and cultivated lands ranging from 300 to 700 m in elevation, were 0.4–0.7 km2 and 0.2–0.8 km2 in size, respectively (Shiomi et al. 1995). In Chugoku block, two case studies are available. The study area in Shimane Prefecture (F1, Fig. 20.2) was selected in hilly land where deciduous forests and conifer plantations coexist in the land lower than 300 m in elevation and a little snow falls. Home range sizes of four males and two females were 0.1–0.3 km2 and 0.5–0.8 km2, respectively (Yokoyama et al. 2002). Larger home ranges of females are not common, though the sample size was small. All of them were sedentary. In Yamaguchi Prefecture (F2, Fig. 20.2) the study area is a mosaic of deciduous forests and conifer plantations, the elevation ranges from 100 to 500 m, and little snow falls. Two males were studied, and the home range sizes of the both males were determined as 1.5 km2, and they were sedentary (Tado and Fukuhara 1997). One female’s range was 0.2 km2 and she was sedentary, while another female had a home range as large as 2.8 km2, and she was considered as migratory. No study was done in Shikoku block (G, Fig. 20.2). One study was available in Kyushu block. Nine males and nine females were radio-tracked in the Kyshu Mountains of Miyazaki Prefecture (H1, Fig. 20.2), where elevation ranged from 600 to 1,200 m, and deciduous broad-leaved forests

278

T. Yabe, S. Takatsuki

and conifer plantation covered the study area. The seasonal home range size of seven males averaged 0.5 km2, and they moved for 2–8 km (Yabe and Koizumi 2003). Subadults and old males were sedentary. Annual home range sizes of seven sedentary females were 0.3 km2 on average. Two females moved for 4–8 km.

Island Populations Kinkazan Island lies off Oshika Peninsula on the Pacific side of the Tohoku block (B2, Fig. 20.2). It is 9.6 km2 in size and forest vegetation prevails with grassland patches. About 400 sika deer inhabit the island. Females utilized both grasslands and forests and their home range sizes were 0.1–0.2 km2 (n = 4, Ito 2000). They were sedentary. Fujita (2005) determined the home range using a GPS collar on a female on Kinkazan Island as 0.23 km2. The island included four types of plant communities, and the forest occupied 65.2%, the herbaceous community accounted for 18.5%, the Zoysia community for 4.4%, and other plant communities for 11.5%. The sika female was sedentary. Nozaki Island in northern Kyushu (H2, Fig. 20.2) is 7.4 km2 in size and inhabited by 500 sika deer. The vegetation is composed both of evergreen broad-leaved forests and grassland patches (Takatsuki 1984). The mean home range size of six females was only 0.03–0.04 km2 (Endo and Doi 1996). They used both forests and grasslands below 200 m in elevation, but forest use was more frequent during winter while grassland use was more frequent in summer. They were sedentary. Mage Island (H3, Fig. 20.2) in southern Kyushu is 9 km2 in size, and is primarily grasslands with small forest patches; it is inhabited by 500 sika deer. All the home ranges of nine males and five females were smaller than 1 km2, and the deer were sedentary (Tatsuzawa 2005).

Discussion This review has shown that northern sika populations tend to be migratory whereas southern ones tend to be sedentary (Fig. 20.2). It is likely that the major factor affecting the migration of the northern populations is snow (Fig. 20.3). A comparison of the three local populations in Tochigi Prefecture (C1, C2, C3) strongly suggests this causation (Fig. 20.3). Hoof loads of sika deer are 0.6–0.8 kg/cm2 (Takatsuki 1992a), which is equivalent to that of mule deer (Odocoileus hemionus, Kelsall 1969) and much lower than that of caribou (Rangifer tarandus, Klein et al. 1987). Since the breast height of sika deer is about 50 cm, snow deeper than 50 cm hinders their locomotion (Takatsuki 1992a). Sika deer in high mountains, therefore, descend when snow falls. The winter climate of Japan is influenced by more than latitude and elevation. Weather patterns result in quite a contrast in snow fall between the Japan Sea

20 Migratory and Sedentary Behavior Patterns of Sika Deer in Honshu

279

100cm< 50-100cm 10-50cm 0-10cm

B1 B2

C3 E4 E3

C2

F1 F2 D1 C4

H2 E2 E1 H1 H3

C1 C5

Migratory Mixed Sedentary

N 200km

Fig. 20.3 Distribution of maximum snow depth and migratory-sedentary patterns of sika deer in Honshu and Kyushu Islands, Japan. (Snow data are from Hokuriku Regional Development Bureau 2005.)

(western) side and the Pacific (eastern) side of the archipelago. Cold air movements from the Chinese mainland bring a moist air mass over the Japan Sea where the warm Tsushima Current flows from south to north, bringing heavy snows to the western side of the divide (Daimaru 2002). This is particular the case in Tohoku District in northern Honshu where snow accumulates 2–3 m or more, and sika deer cannot live in many mountainous areas. Migration patterns of the sika deer in Tochigi Prefecture (C1, C2, C3, Fig. 20.3) seem to be affected by this snow pattern. Thus, migration is an adaptation to a complex interaction between latitude, elevation, and weather patterns. The Shimane population (F1) was sedentary in spite

280

T. Yabe, S. Takatsuki

of its location on the Japan Sea side, which is explained by less snow accumulation because of low elevation. In contrast, a part of the population of Odaigahara (E1) in Kinki District was migratory in spite of its location on the Pacific side. This area is snowy because of high elevation, thus being equivalent to lower elevations on the western side of the mountains in the northern Tohoku block. Besides these three major variables, local habitat and population conditions can further influence migratory versus sedentary behavior. For example, home ranges of sika deer on islands were quite small and they were sedentary. Kinkazan Island (B2) is located in the cool-temperate zone and seasonality is marked, and the summit is as high as 450 m. These factors seem to favor migration. However, the deer were sedentary. This fact suggests that the high density and consequent filling of home ranges on the island do not permit the deer to move up and down. Even in the northern cool temperate zone in Tohoku block, however, sika deer living at low elevation tend to be sedentary. Thus, it is expected that sika deer populations in the central Chubu block, where elevation magnitude is greatest, would move up and down. Studies on migration in this block are needed. Although climbing mountains requires energy, deer densities are often high in wintering areas and, consequently, food plants are often poor due to heavy deer grazing. Because of this, food conditions are often poorer there. Therefore, leaving there and climbing to higher areas would be advantageous. Besides, returning to birth places (“homing”) in higher areas would mean returning to familiar or safe places in terms of behavior. A study on migration in eastern Hokkaido (Igota et al. 2004), which showed that females wintering at a high mountain moved eastward and returned quite precisely to the last-year summer ranges, strongly suggests that sika deer females have homing behavior. However, sika deer do not migrate such long distances as caribou (Miller et al. 1977) or mule deer (Nicholson et al. 1997) in North America. This relates to steep topography of the Japanese Archipelago, which is further reinforced by habitat fragmentation or other human activities which block migration routes at lower elevations. Contrary to the northern populations, the southern Japanese populations in the warm-temperate zone were sedentary (Fig. 20.2). Many evergreen plants grow there and, therefore, deer foods in summer and winter are not so different as in the north. Since snow does not fall there, the difference between summer and winter habitat would be less than in the north. These factors disfavor migration in the south. A difference in forestry damage in Tochigi Prefecture in the north where damage occurred in a short period in spring (Ueda et al. 2002) and in Fukuoka Prefecture in the south where damage continuously occurred throughout the year (Ikeda 1996) strongly supports this conclusion. Thus, the significance of winter is very different between the north and the south. Sika deer in the south live in small groups, spaced from one another, at low densities because: (1) forests are generally dark throughout the year, (2) consequently, foods are less abundant and lower in nutrition, and (3) visibility is not good. However, once forests are logged and open habitats are created, sika deer concentrate there and form larger groups. There were some short-distance movements in

20 Migratory and Sedentary Behavior Patterns of Sika Deer in Honshu

281

the southern populations in the warm-temperate zone. It is not proper to call these movements migrations; better that they be called local “shifts” in habitat use. Factors affecting these movements such as foods, visibility, and reproductive behavior need to be further studied. Although this review has shown such a consistent tendency towards migration or sedentary behavior of sika deer between the northern and the southern populations, local variations were also found. Home range sizes and the proportions of migratory and sedentary animals varied, which seems to relate to resource distributions which change in small scales depending on the complexity of habitats lying on diversified topography and a mosaic of vegetation. Kie et al. (2002) found that diversity of home ranges of mule deer is often explained by heterogeneity of habitats. Ford (1983) predicted a home range model based on optimal foraging rules that the optimal home range shape in a patchy environment is elongate rather than circular. Many home range samples of birds Contopus sordidulus (wood pewee), Seiurus aurocapillus (ovenbird), rodents Microtus ochrogaster (prairie vole), Microtus pennsylvanicus (meadow vole), Eutamias minimus (least chipmunk), Peromyscus leucopus (white-footed mouse), Pitymus pinetorum (pine vole), Lemmus trimucronatus (brown lemming), and fox (Vulpes vulpes) have agreed with this. In Japan the topography is steep and vegetation is patchy, both because of natural factors and activities of agriculture and forestry. Further studies on habitat use and migration of sika deer based on such a point of view are needed.

Literature Cited Chiba Prefecture and Boso Deer Research Group. 2004. Research report on sika deer management in Boso Peninsula, Chiba Prefecture (1992–2003 summary): 60–69. (In Japanese.) Daimaru, H. 2002. The snow-rich mountains famous to the world. Pages 13–26 in T. Kajimoto, H. Daimaru, and H. Sugita, editors, Snow environment and plant ecology of Japanese northern mountains. Tokai University Press, Hadano, Japan. (In Japanese.) Endo, A., and T. Doi. 1996. Home range of female sika deer Cervus nippon on Nozaki Island, the Goto Archipelago. Japan Mammal Study 21:27–35. Ford, R. G. 1983. Home range in a patchy environment: optimal foraging predictions. American Zoologist 23:315–326. Fujita, Y. 2005. A possibility of seed dispersal by sika deer between different plant communities. Ms thesis, School of Agricultural Sciences, The University of Tokyo, Tokyo, Japan. Hokuriku Regional Development Bureau, Ministry of Land, Infrastructure and Transport. 2005. Distribution of averaged maximum snow depth. http://www.hrr.mlit.go.jp/library/hokuriku2005/ s1/1–04/03sekisetusin/03sekisetusin.html (In Japanese, cited on 28 Jul. 2006.) Honma, K. 1985. Analysis of movement pattern and habitat selection of sika deer in Oku-Nikko and Ashio Area. Ms thesis of Joetsu Educational College, Joetsu, Japan. (In Japanese.) Igota, H., M. Sakuragi, H. Uno, K. Kaji, M. Kaneko, R. Akamatsu, and K. Maekawa. 2004. Seasonal migration pattern of female sika deer in eastern Hokkaido, Japan. Ecological Research 19:169–178. Ikeda, K. 1996. Characteristics of deer damage in Fukuoka Prefecture and trials of damage prevention by repellents. Forestry and Chemicals 137:13–18. (In Japanese.)

282

T. Yabe, S. Takatsuki

Ito, Y. T. 2000. A study on the short-grass community maintained by grazing of the sika deer on Kinkazan Island. PhD thesis, Faculty of Science, Tohoku University, Sendai, Japan. Jiang, C., M. Kitahara, S. Takatsuki, M. Sugita, and H. Yoshida. 2006. An ecological study on sika deer on the foot of Mt. Fuji. Abstracts of 2006 Annual Meeting of the Mammalogical Society of Japan 55. Kawai, Y., S. Ohtani, Y. Ishizuka, W Ishii, and Y. Matsushita. 2006. Spatial use of two sika deer (Cervus nippon) individuals which have overlapped home ranges. Bulletin of Agricultural, Food and Environmental Sciences Research Center of Osaka Prefecture 42:16–19. (In Japanese with English summary.) Kelsall, J. P. 1969. Structural adaptations of moose and deer for snow. Journal of Mammalogy 50:187–208. Kie, J. G., R, T. Bowyer, M. C. Nicholson, B. B. Boroski, and E. R. Loft. 2002. Landscape heterogeneity at differing scales: effects on spatial distribution of mule deer. Ecology 83:530–544. Klein, D. R., M. Meldgaard, and S. G. Fancy. 1987. Factors determining leg length in Rangifer tarandus. Journal of Mammalogy 68:642–655. Maeji, I., T. Kurosaki, S. Yokoyama, and E. Shibata. 2000. Home range of sika deer (Cervus nippon) on Mt. Ohdaigahara, central Japan. Nagoya University Forest Science 19:1–10. (In Japanese with English summary.) Maruyama, N. 1981. A study of the seasonal movements and aggregation patterns of sika deer. Bulletin of Faculty of Agriculture, Tokyo University of Agriculture and Technology 23:1–85. (In Japanese with English summary.) Miller, F. L., R. H. Russell, and A. Gunn. 1977. Inter-island movements of Peary caribou (Rangifer tarandus pearyi) on western Queen Elizabeth Islands, Arctic Canada. Canadian Journal of Zoology 55:1029–1037. Miura, S. 1984. Social behavior and territoriality in male sika deer (Cervus nippon Temminck 1838) during the rut. Zeitschrift für Tierpsychologie 64:33–73. Nagata, K. 2005. Home range characteristics of sika deer in Fudakake of Tanzawa Mountains. Honyurui Kagaku (Mammal Science) 45:25–33. (In Japanese with English summary.) Nicholson, M. C., R. T. Bowyer, and J. G. Kie. 1997. Habitat selection and survival of mule deer: tradeoffs associated with migration. Journal of Mammalogy 78:483–504. Satake, C., T. Tanaka, and M. Koganezawa. 1994. Home range and habitat selection of sedentary sika deer in Ashio Mountains. Abstract of Mammalogical Society of Japan, 1994:64. (In Japanese.) Shiomi, S., K. Takaya, and Y. Ueyama. 1995. The research on the behavior of sika deer with radio telemetry (I): the results in the two different areas. Transactions of Kansai Branch of Japan Forestry Society 4:147–150. (In Japanese.) Shigematsu, Y., M. Asada, and K. Ochiai. 1995. Radio-tracking of individuals. Chiba Prefecture and Boso Deer Research Group, Research report on sika deer management in Boso Peninsula, Chiba Prefecture 3:54–72. (In Japanese.) Tado, H., and N. Fukuhara. 1997. Research on wildlife ecology and forest damage prevention, 1: Home range and vegetation use of sika deer. Report of Yamaguchi Prefectural Institute of Forestry 11:66–75. (In Japanese.) Takatsuki, S. 1984. Ecological studies on effect of sika deer (Cervus nippon) on vegetation on Nozaki Island, the Goto Islands, northwestern Kyushu. Ecological Review 20:223–235. Takatsuki, S. 1992a. Foot morphology and distribution of sika deer in relation to snow depth. Ecological Research 7:19–23. Takatsuki, S. 1992b. Sika deer living in the north. Dobutsu-sha Publishers, Tokyo, Japan. (In Japanese.) Takatsuki, S., K. Suzuki, and H. Higashi. 2000. Seasonal elevational movements of sika deer on Mt. Goyo, northern Japan. Mammal Study 25:107–114. Tanzawa-Ohyama Synthetic Survey Team. 2006. Atlas Tanzawa 2006. Sika deer management— present status and problem. http://e-tanzawa.jp/atlas/2–3_shika_1genjo.html. (In Japanese, cited on 4 Jul. 2007.)

20 Migratory and Sedentary Behavior Patterns of Sika Deer in Honshu

283

Tatsuzawa, S. 2005. How do we cope with sika deer in the evergreen broad-leaved forest area with reference to the efforts in Yakushima Island? Transactions of Kanto Branch, The Ecological Society of Japan 54:41–53. (In Japanese.) Ueda, H., S. Takatsuki, and Y. Takahashi. 2002. Bark stripping of hinoki cypress by sika deer in relation to snow cover and food availability on Mt Takahara, central Japan. Ecological Research 17:545–551. Yabe, T., and T. Koizumi. 2003. Sedentation and migration of sika deer in Kyushu Mountains. Forest and Forestry in Kyushu 65:1–3. (In Japanese.) Yajima K., Y. Yamamoto, I. Maeji, T. Kurosaki, T. Yokota, H. Sato, and E. Shibata. 2002. Seasonal changes in home range of female sika deer (Cervus nippon) in Mt. Ohdaigahara, central Japan. Nagoya University Forest Science 21:1–7. (In Japanese with English summary.) Yokoyama, N., S. Katagiri, and H. Kanamori. 2002. Relationship between home range of sika deer (Cervus nippon) and species composition of woody plants in Misen Mountains, Shimane Peninsula. Applied Forest Science 11:27–38. (In Japanese with English summary.) Yoshida, T., T. Suzuki, R. Araki, T. Kurosaki, and K. Tokida. 2007. Habitat use of sika deer in Mt. Ohdaigahara revealed by GPS telemetry. Abstracts of 118th Annual meeting of The Japanese Forest Society. CD-ROM, ISSN:1349–8517. P2g19. (In Japanese.) Yoshioka, K. 1973. Plant geography. Kyoritsu Publishers, Tokyo, Japan. (In Japanese.)

Chapter 21

Variation in Mating Behavior of Sika Deer: Mating Behavior of Sika Deer on Nozaki Island Akira Endo

Abstract Variation in the mating tactics of male sika deer (Cervus nippon) associated with female distribution was studied on Nozaki Island in the Goto Islands, Japan. In 1990, 1991, and 1993, observations were carried out mainly in an open grassland (about 5 ha) in the central part of the island, which was regularly used by 20 adult females and their calves for daily feeding. A majority of the observed females were not associated with other females. The estrous days of these females were not synchronous. Several females accepted multiple copulations. In the rutting season of 1991, five dominant males (DMs) established and kept exclusive home ranges, while those of subordinate males (SMs) considerably overlapped with other males. The mean home range size of dominants (4.28 ± 0.74 ha) was significantly smaller than that of subordinates (10.14 ± 0.59 ha). The open grasslands where females stayed longest when foraging during the day were occupied by dominant males, but subordinates were not chased out violently by dominants. We, therefore, designated the home range of a dominant male as “dominance area.” Most females copulated with dominant males. The mating success of dominants was significantly greater than that of subordinates (U = 0, p < 0.01). Frequently, dominant males did not search for estrous females, but intercepted estrous females from subordinates. The “mating aggregation” was formed more frequently by subordinates and might be an alternative mating tactics of subordinates. Almost all copulations were followed by guarding behavior. The postcopulatory guarding by DMs appears to be more effective in the prevention of additional female copulations with other males than guarding by SMs. Dominant males often copulated repeatedly with the same female. If the paternity of a male increases with the amount of sperm, repeated copulation seems to be an adaptation to increase his reproductive success. SMs decreased the duration of the precopulatory phase to achieve copulation before having to give way to DMs.

Introduction Social systems of vertebrates have been believed to be stable within species. However, some recent studies report intraspecific variation in social systems (Lott 1984, 1991; Langbein and Thirgood 1989). Since Emlen and Oring (1977), D. R. McCullough et al. (eds.), Sika Deer: Biology and Management of Native and Introduced Populations, DOI: 10.1007/978-4-431-09429-6_21, © Springer 2009

285

286

A. Endo

the mating system, a central component of any social system, has been seen as an outcome of the reproductive tactics of both sexes to maximize their reproductive success. The factors affecting reproductive success are often different between two sexes; female reproductive success is limited by the resources available for reproduction, while male reproductive success is limited by the number and the distribution of females available for mates (Bradbury and Vehrencamp 1977; Emlen and Oring 1977; Davies 1991). In widely distributed mammal species, the factors affecting female and male reproductive success often vary significantly between populations and, thus, intraspecific variation in mating system is expected as an outcome of evolution to local optima. Even within a population, the availability of estrous females for mature males is reported to vary both with local habitat condition (Alvarez et al. 1990) and over the course of a rutting period (Schaal 1985a, b; Langbein and Thirgood 1989; Hirth 1997). Also, the optimal tactics for males may vary according to the social status of the different individuals (Moore et al. 1995). In the Cervidae, intraspecific variation in the mating system is known in widely distributed species such as fallow deer (Dama dama) and sika deer (Cervus nippon). Three major types of mating system have been recognized for fallow deer: harem (Alvarez et al. 1975), mating territory (Alvarez et al. 1990; Moore et al. 1995), and lek (Schaal and Bradbury 1987; Clutton-Brock et al. 1988; Apollonio et al. 1989; Hirth 1997). Alternatively, Langbein and Thirgood (1989) have classified the mating system of fallow deer into seven types: stand (reproductive territory, isolated), temporary stand (reproductive territory, defended only 1 or 2 h of the day), multiple stand (two or three reproductive territories abutting), lek (assemblies of territorial males, for display and mating only), following (nonterritorial, follow the group containing reproductive females), harem (nonterritorial, defends a group of females), and dominance groups (nonterritorial, multi-male mixed sex group, access to estrous female monopolized by dominant male). Langbein and Thirgood (1989) also suggested that ecological factors such as male density, female density, habitat structure, and tree cover affect the mating system. In the case of sika deer, geographically separated populations are classified into 14 subspecies which are distributed widely from Vietnam to Ussuri (Ohtaishi 1986). Six of the 14 subspecies are native to the islands of Japan and these subspecies differ in body size, antler shape, and antler length (Ohtaishi 1986). Geographical variation in feeding ecology is also reported (Takatsuki 1991). For the mating system of sika deer, the “mating territory” is found in the seminatural population in Nara Park (Miura 1984). Miura (1986) also reported three additional types of mating system in the Japanese sika deer: “harem” for the Kinkazan population in northeastern Japan, “tending bond” for the Nakanoshima population in northern Japan, and “mating aggregation” for the Nozaki population in southwestern Japan. Mating aggregation is an unusual mating system in which many males rush for an estrous female and the most dominant male finally copulates. Miura (1986) suggested that the mating system of sika deer might be affected by the gregariousness of females. As an index of female gregariousness, previous studies assessed the group size by including all mature females (Miura 1986). However, there have been no studies into the variation of male mating behavior with the spatio-temporal distribution of estrous females.

21 Variation in Mating Behavior of Sika Deer

287

In a previous study of sika deer on Nozaki Island, Endo and Doi (1996) showed that females had small home ranges (3.0–3.6 ha), stable over the course of a year, which covered both open vegetation and forested areas. We also showed no dependency in the simultaneous movements of females although their home ranges overlapped each other to a considerable extent (Endo and Doi 1996). Moreover, Endo et al. (1997) found postcopulative guarding by dominant males in the enclosed population on the island. In this study, we investigate the mating tactics of male sika deer in the Nozaki population and assess how they are influenced by the features of estrous females (such as gregariousness, distribution patterns, and multiple copulation) and by male guarding tactics. We also examine what alternative tactics subordinate males use.

Material and Methods Study Area Nozaki Island (740 ha in size) is a small island in the Goto Islands, located in the west of Nagasaki Prefecture (33°10´ N, 129°8´ E). Most land on the island is covered with secondary evergreen broad-leaved forest, dominated by Castanopsis cuspidata (Japanese chinquapin), Camellia japonica (Japanese camellia), and Machilus thunbergii (Lauraceae). The remainder is covered with young forests of Pinus thunbergii (Japanese black pine), bushes of Glochidion obovatum (Euphorbiaceae), or seminatural grasslands of Miscanthus sinensis (Chinese silvergrass), Imperata cylindrica (cogon grass) and Zoysia japonica (Japanese lawngrass) (Kawahara 1983). The seminatural grasslands spaces are used frequently by deer as feeding places. About 700 deer live on the island, and in 1991 the density of deer varied from 0.6/ha in forest to 3.1/ha in open grasslands (Doi and Endo 1992). Hunting is not allowed on the island and there are no natural predators. In 1991, mating behavior of sika deer was studied mainly in an open grassland (about 5 ha) in the central part of the island, which was regularly used by 20 adult females and their offspring for daily feeding.

Field Observation Most observations were carried out from dawn to dusk during the October rutting season of 1990, 1991, and 1993. Female gregariousness and mating behavior were both recorded by direct observation using 8 X 30 binoculars and a telescope with zoom lens from X15–X60. Total observation time was 150 h in 1990, 250 h in 1991, and 54 h in 1993. In August 1991, just before rutting season, we developed a means of identifying all the males observed in the study area by recording antler length and shape.

288

A. Endo

Female Gregariousness In October 1991, we scored the number of adult females within a group as an index of the gregariousness of females. When an adult female was associated with her calf, for example, the value of the female gregariousness is one, while the group size is two. Following Miura (1978), we defined a group as being an assemblage of animals which move synchronously within 10 m of one another.

Mating Behavior We recorded the movements of a focal male using the focal animal sampling method (Altmann 1974), and all male-female sexual interactions were noted. Whenever a copulation was observed, we identified the male, measured the time duration of sexual interactions, and recorded whether or not multiple copulation and postcopulative guarding by a male occurred. To examine synchrony of estrus in females, we counted the number of females that had copulated in the observation area in 1991 (Endo and Doi 2001). According to studies of red deer where the duration of estrus is one day (Clutton-Brock et al. 1982; Kelly et al. 1985), the related sika deer is considered to have the same duration of estrus. Female reproductive condition is very difficult to distinguish in wild populations without a detailed physiological examination, and we had to assume that the day a female copulated was the day of her estrus. The home ranges of males in 1991 were determined using the minimum convex polygon method (Mohr 1947). Mating success of each male was calculated by the formula as follows: the number of females copulated by a male/the number of days that he appeared in the study area. We examined multiple copulation, male guarding success, and the processes to acquire estrous females (mating aggregation, searching, and appropriation) using the records of all copulation observed over three years.

Results Mating Behavior of Females Female gregariousness and synchrony of estrus. Frequency distribution of female gregariousness in October 1991 is shown in Fig. 21.1. The mean size was 1.03 females per group. The majority of females were not associated with other females. The date of copulation was determined for ten females by direct observation and in the case of eight additional females by observations made after copulation: since mate guarding by males only occurred after copulation, we assumed the guarded female had been copulated. Figure 21.2 shows the distribution of the date of copulation

21 Variation in Mating Behavior of Sika Deer

289

Fig. 21.1 Frequency distribution of the index values of female gregariousness.

Fig. 21.2 Daily number of estrous females observed in October 1991. (Redrawn from Endo and Doi 2001.) Table 21.1 Proportion of single- or multiple-copulated females. (Data from Endo and Doi 2002.) Percent N Single-copulated female Multiple-copulated female With one male With multiple males

54.6 45.4

12 5 5

in 18 females in October 1991. There is no tendency toward synchronization (Endo and Doi 2001). Multiple copulations by females. Table 21.1 shows the proportion of multiplecopulated females to the total number of copulated females, observed during three years. In this table, the number of females estimated to be copulated in Fig. 21.2 have been excluded. It was founded that ten of 22 females in estrus (45.4%) made multiple copulations (Endo and Doi 2002).

Mating Behavior of Males Spacing patterns of males in rutting season. In rutting season of 1991, five dominant males (DMs) established and kept their own home ranges in and around the

290

A. Endo

Fig. 21.3 Home ranges of dominant males (DMs) and subordinate males (SMs) in October 1991.

Table 21.2 Home range size (ha) of males. Mean ± S.D. Dominant males Subordinate males

4.28 ± 0.74 (N = 3) 10.14 ± 0.59 (N = 3)

observation area. Fifteen subordinate males (SMs) were also observed in the same area. The home ranges of eight males (five DMs and three SMs) were determined, but the range borders of two DMs (DM4 and DM5) and the home ranges of 12 SMs could not be determined. Figure 21.3 shows the spacing pattern of eight home ranges of males in rutting season of 1991. The ranges of the DMs were exclusive to each male, while those of SMs considerably overlapped with ranges of both DMs and SMs. However, the mean home range size of the DMs was significantly smaller than those of SMs (Table 21.2; Mann-Whitney U test: U = 0, z = 2.248, p < 0.05). Dominant males occupied the open grasslands where females spent long periods foraging in the day time. Subordinate males did not occupy their own open grassland and ranged over all open grassland within their home ranges. Subordinates were tolerated by the resident DMs and were not chased violently. Mating success of dominant males and subordinate males. On an open grassland occupied by two DMs (DM1 and DM2) in 1991, 18 females were copulated. The mating success of the DMs and SMs are shown in Table 21.3. Most females were copulated by DMs. Only four of 15 SMs were able to copulate successfully. Mating success of DMs was significantly greater than that of SMs (U test, U = 0, z = −1.964, p < 0.01). Guarding success of dominant and subordinate males. During three years of this study, 27 copulations were observed. Twenty-six were seen to be followed by

21 Variation in Mating Behavior of Sika Deer

291

Table 21.3 Mating success of males. N females / N days

N

Dominant males

2

Subordinate males

0.46 (DM 1) 0.24 (DM 2) 0.03 ± 0.05

15

Table 21.4 Guarding success or failure of males. (Data from Endo and Doi 2002.) Success (%) Failure (%) Dominant males Subordinate males

94.0 (N = 17) 12.5 (N = 1)

6.0 (N = 1) 87.5 (N = 7)

Table 21.5 Number of observations for three mating tactics of dominant and subordinate males. No aggregation Mating aggregation Dominant Subordinants

1 4

Interception 10 2

Searching 5 0

guarding behavior. For the remaining one case, we could not determine whether the male guarded the female after copulation because the pair moved into cover. The success or failure of male guarding after copulation is summarized in Table 21.4. Dominant males guarded the females successfully in most cases, but SMs usually failed to guard a copulated female (Fisher’s exact test: p < 0.001). Subordinate males also failed to guard a female against DMs or other SMs of higher rank. However, one case of a SM successfully guarding after copulation did occur, at a time when the DM was guarding another female. Process of acquisition of estrous female. During three years of this study, we observed the process leading to acquisition of a female by individual males in 22 of the 27 observed copulations (Table 21.5). Dominant males did not frequently search for estrous females that came into the open grassland, but appropriated estrous females from SMs. In four to six cases, SMs obtained females by “mating aggregation” as reported by Miura (1986). The mating aggregation was most frequently composed of SMs (Table 21.5; comparison of the number of mating aggregations versus of nonaggregation between DMs and SMs, Fisher’s exact test: p < 0.01). Copulatory behavior of dominant and subordinate males. The number of mounts before ejaculation for DMs and SMs are shown in Fig. 21.4. Subordinate males mounted significantly less often before ejaculation than did DMs (U test, U = 41.0, z = −2.28, p < 0.05). The median intervals between the first mount and the copulation in a sequence of sexual interactions were 2.00 min (n = 26, Q1 = 1.00, Q3 = 2.75) and 0.25 min (n = 7, Q1 = 0.18, Q3 = 2.00), for DMs and SMs respectively. Hence, SMs commenced copulation significantly earlier than DMs (U test, U = 48.0, z = −1.96, p < 0.05).

292

A. Endo

Fig. 21.4 Number of mountings before ejaculation by dominant males (DMs) and subordinate males (SMs). Boxes represent interquartile ranges, horizontal lines within boxes show medians. Lower and upper bars represent 10th and 90th percentiles respectively. (Redrawn from Endo and Doi 2002.)

Table 21.6 Number of observations for single and repeated copulations in dominant and subordinate males. (Data from Endo and Doi 2002.) Single Repeated Dominant males Subordinate males

12 8

7 0

Males often copulated with the same female several times in a day. We observed seven repeated copulations. These repeated copulations were only made by DMs, and not by SMs (Table 21.6).

Discussion Dominant males in the Nozaki sika population appeared to adopt a “sit and wait” tactic, waiting for estrous females to come to the open grassland every day. It was found that dominant males avoided conflict with other dominant males by spacing themselves out to a sufficient degree that encounters between dominants were rare. The spacing pattern of dominant males is similar to “multiple stand,” as classified by Langbein and Thirgood (1989), but it does not agree exactly with territory. In a population of territorial sika deer, the dominant males do not tolerate the presence of other males and attempt to drive opponents away from the territory in rutting season (Miura 1984). At the same time, dominant males keep many females to themselves by great efforts at herding behavior (Miura 1984). In this study, dominant males were found to tolerate the presence of subordinates, and only in a few instances were subordinates chased. The dominant males also showed no herding behavior toward females. However, this behavior had been observed for the dominant male in an enclosure on Nozaki Island (Endo et al. 1997). While subordinates could move freely and hence had a chance of access to females, estrous females preferentially copulated with dominant males. The dominant males, without any conflict, could each monopolize access to about 20 estrous females in their area.

21 Variation in Mating Behavior of Sika Deer

293

Moore and Marchinton (1974) reported a similar mating system in white-tailed deer, and designated it as a “dominance area.” We, therefore, also called the mating system in Nozaki a “dominance area,” not a “territory.” Langbein and Thirgood (1989) showed that territoriality by dominant males increased with male and female densities in fallow deer populations. Territorial activity of dominant males in the present study was less than those of the multiterritorial populations of fallow deer, while the density of males in our study (1.8 males/ha) was significantly higher than the maximum density among these populations (0.5 males/ha). Female density in our study (4.0 females/ha) was also greater than that of fallow populations (2.0 females/ha). Hirth (1997) also showed the mating system of a fallow deer population in Texas was different from the prediction of Langbein and Thirgood (1989). Since density varies temporally and spatially in a population, and the total numbers of each sex also affects the mating system, it is difficult to make an evaluation of the mating system based on density. A major factor in the cost-benefit trade-off for male deer is the actual numbers of females synchronized in estrus, as mentioned by Langbein and Thirgood (1989). The distribution, both spatial and temporal, of estrous females in a population will affect decisions on mating tactics by males. Previous studies, however, have used the approximate density or the group size including nonestrous mature females to estimate the availability of females. In this study, by contrast, we excluded the nonestrous females and used the spatio-temporal distribution of estrous females to evaluate the male mating tactics of sika deer. Females in the Nozaki population had small home ranges (3.0–3.6 ha), stable over the course of a year, which covered both open vegetation and forested areas and showed no synchronization of movements although their home ranges overlapped each other to a considerable extent (Endo and Doi 1996). Moreover, the estrous times of females on Nozaki Island were scattered, not synchronous, and the rutting season was consequently prolonged for a month without a peak (Endo and Doi 2001).The daily availability of females (number of estrous females per day) for rutting males was quite low. When the estrous times of females are high synchronized, territory and harem defending provides high reproductive benefit, while it imposes heavy energetic investment on dominant males. A successful fallow deer territory holder, for example, achieved 68 matings in the rut, but he died prior to the following rut season (Moore et al. 1995). In red deer a harem holder that mated during the peak of rut season showed higher success than those that rutted before and after the peak (Clutton-Brock et al. 1982). Dominant males of the Nozaki population, on the other hand, achieved high mating success without the energetic cost of herding females or chasing away other males. Moreover, dominant males appropriated estrous females from subordinates more frequently than they searched for them by themselves in their dominance areas (Table 21.5). Searching tactics are considered to be more costly than the appropriation tactics, because it requires more energy and time to approach and to assess the condition of all females that have come to the dominance area, than does appropriation of an estrous female already identified by a subordinate male.

294

A. Endo

In the rut season, dominant males in the Nozaki population spent more time feeding than those in other populations of sika deer (S. Miura personal communication). They often maintained their social status and held a “dominance area” more than two consecutive years (Endo unpublished data), while all territorial males of fallow deer in a given year failed to hold territories the following year (Moore et al. 1995). Dominance area and appropriation tactics might be due to the low and constant daily availability of estrous females. These tactics would allow the dominant males in the Nozaki population to save energy, would prevent the dominants from exhaustion, and hence turnover of rank among males might occur less frequently. After the rutting season, however, one dominant male abandoned his dominance area and switched his tactics from “appropriation” to “searching” in response to decreased female availability (Endo unpublished data). Dominant males show a plasticity in mating tactics and change behavior in response to the spatio-temporal distribution of estrous females. For deer species, estrus in gregarious populations of both red deer and sika deer showed high synchronicity. A majority of female red deer conceived during the first two weeks of October (Clutton-Brock et al. 1982) and the peak of conception for sika deer was in the last ten days of October (Miura 1980). Similarly, synchronous estrus of females has been reported for some fallow deer populations (CluttonBrock et al. 1988; Hirth 1997). Iason and Guiness (1985) showed estrus for red deer was more synchronized within groups than between groups and suggested the effect of social cues among females within groups as the most important variable. Social cues are more effective at advancing the synchronization of estrus in gregarious females inhabiting open habitats. The home ranges of females on Nozaki Island were found to overlap with each other to a considerable extent, although individual females were observed to move freely without notable interference (Endo and Doi 1996). In the absence of interference, females might have little chance to respond to social cues and that might explain the asynchronous estrus of females in the Nozaki population. Another surprising finding of this study is the multiple copulations of female sika deer. It has been reported previously only for the Kinkazan population (Minami 1992). In the Kinkazan population, Minami (1992) reported that some females escaped from a guarding male and copulated with other males and suggested that females preferred to mate with different males to ensure conception. In the present study, females had the opportunity to accept multiple copulations, with five of 22 females mating with several males. Moreover, four of five females, which undertook multi-male copulations, copulated last with dominant males, and the remaining one finally copulated with a subordinate male of higher rank than her previous mate. Consequently, females might be able to choose the stronger mates. The estrus of a female lasts for one day and females are able to copulate repeatedly with several males during the day. Postcopulative guarding of a dominant male in enclosure lasted more than 10 h (Endo et al. 1997). Postcopulative guarding might be the optimal tactic for a male in the Nozaki population to ensure his paternity. In the population in which estrus is asynchronous and density of estrous females is low, additionally, postcopulative guarding by males would be the most

21 Variation in Mating Behavior of Sika Deer

295

profitable tactic. In harem-forming red deer populations in which many females in a group became estrous on the same day, dominant males attempted to mate with estrous females to maximize their reproductive success (Clutton-Brock et al. 1982). This seems to be the case for males observed in this study, most of which guarded females after copulation. Guarding success was significantly different between dominant males (94%) and subordinates (12.5%). Subordinate males also have a chance of mating, even though they are successfully exploited as an “estrous female detector” by dominant males. In the dominance area, subordinates had little chance for mating. However, they might have an opportunity to obtain estrous females by means of mating aggregation, which disturbs the order in the dominant area. As the frequency of the mating aggregations of subordinates was significantly higher than that of dominants (Table 21.5), it might be a successful mating tactic of subordinates. Although the highest ranked male in the mating aggregation succeeded in copulation, lower ranking males could successfully copulate by sneaking a female for which higher-ranked males are fighting. A decrease in the numbers of mounts and the duration of the precopulatory phase also could be an optimal strategy by surbordinates to achieve a copulation before being driven away by a dominant male (Fig. 21.4). Finally, dominant males often copulated the same female repeatedly. It suggests the possibility of sperm competition in sika deer. In this study, females that came into the open grassland might have already copulated with other males. If the dominant male could flush out the sperm of other males, and if the paternity of a male increases with the amount of sperm, repeated copulation seems to be adaptive in increasing his reproductive success.

Literature Cited Altmann, J. 1974. Observational study of behavior: sampling methods. Behaviour 49:227–265. Alvarez, F., F. Braza, and A. Norzagary. 1975. Estructura social del gamo (Dama dama, Mammalia, Cervidae), en Donana. Ardeola Especial 21:1119–1142. Alvarez, F., F. Braza, and C. San Jose. 1990. Coexistence of territoriality and harem defense in a rutting fallow deer population. Journal of Mammalogy 71:692–695. Apollonio, M., M. Festa-Bianchet, and F. Mari. 1989. Correlates of copulatory success in a fallow deer lek. Behavioral Ecology and Sociobiology 25:89–98. Bradbury, J. W., and S. L. Vehrencamp. 1977. Social organization and foraging in emballonurid bats, III: Mating systems. Behavioral Ecology and Sociobiology 2:1–17. Clutton-Brock, T. H., F. E. Guiness, and S. D. Albon. 1982. Red deer: Behavior and ecology of two sexes. University of Chicago Press, Chicago, Illinois, USA. Clutton-Brock, T. H., D. Green, M. Hiraiwa-Hasegawa, and S. D. Albon. 1988. Passing the buck: Resource defence, lek breeding and mate choice in fallow deer. Behavioral Ecology and Sociobiology 23:281–296. Davies, N. B. 1991. Mating systems. Pages 263–294 in J. R. Krebs and N. B. Davis, editors, Behavioural ecology: An evolutionary approach. Blackwell Scientific, Oxford, United Kingdom. Doi, T., and A. Endo. 1992. A report on a census of sika deer in Nozaki Island, the Goto Islands. Ojika Town Office, Ojika, Japan. (In Japanese.)

296

A. Endo

Emlen, S. T., and L. W. Oring. 1977. Ecology, sexual selection and the evolution of mating systems. Science 197:215–23. Endo, A., and T. Doi. 1996. Home range of female sika deer (Cervus nippon) on Nozaki Island, the Goto Archipelago, Japan. Mammal Study 21:27–35. Endo, A., and T. Doi. 2001. Asynchronous estrus of female sika deer (Cervus nippon) during the rutting season. Mammal Study 26:69–72. Endo, A., and T. Doi. 2002. Multiple copulations and post-copulatory guarding in a free-living population of sika deer (Cervus nippon). Ethology 108: 39–747. Endo, A., T. Doi, and A. Shiraki. 1997. Post-copulative guarding: mating behavior of non-territorial male sika deer (Cervus nippon) in an enclosure. Applied Animal Behaviour Science 54:257–263. Hirth, D. H. 1997. Lek breeding in a Texas population of fallow deer (Dama dama). American Midland Naturalist 138:276–289. Iason, G. R., and F. E. Guiness. 1985. Synchrony of oestrus and conception in red deer (Cervus elaphus L.). Animal Behaviour 33:1169–1174. Kawahara, H. 1983. Vegetation of Nozaki-jima, the Goto Island. Bulletin Nagasaki Institute of Applied Science 24:239–247. (In Japanese with English summary.) Kelly, R. W., K. P. McNatty, and G. H. Moore. 1985. Hormonal changes about oestrus in female red deer. Pages 181–184 in P. F. Fennessy and K. R. Drew, editors, Biology of deer production. The Royal Society of New Zealand Bulletin 22, Wellington, New Zealand. Langbein, J., and S. J. Thirgood. 1989. Variation in mating systems of fallow deer (Dama dama) in relation to ecology. Ethology 83:195–214. Lott, D. F. 1984. Intraspecific variation in the social systems of wild vertebrates. Behaviour 88:266–325. Lott, D. F. 1991. Intraspecific variation in the social systems of wild vertebrates. Cambridge University Press, Cambridge, United Kingdom. Minami, M. 1992. Peculiar mating behavior of sika deer. The Earth and Animals 55:204–207. Asahi-Shinbun, Tokyo, Japan. (In Japanese.) Miura, S. 1978. A year of sika deer in Nara Park: I. Rutting period. Annual Report of Nara Deer Research Association 4:3–13. (In Japanese with English summary.) Miura, S. 1980. A year of sika deer in Nara Park: II. Birth season. Annual Report of Nara Deer Research Association. 5:87–94. (In Japanese with English summary.) Miura, S. 1984. Social behavior and territoriality in male sika deer (Cervus nippon Temminck 1838). Zeitschrift für Tierpsychologie 64:33–73. Miura, S. 1986. A note on the evolution and social system in Cervidae. Honyurui Kagaku (Mammalian Science) 53:19–24. (In Japanese.) Mohr, C. O. 1947. Table of equivalent populations of North American small mammals. American Midland Naturalist 37:223–249. Moore, N. P., P. F. Kelly, J. P. Cahill, and T. J. Hayden, T. J. 1995. Mating strategies and mating success of fallow (Dama dama) bucks in a non-lekking population. Behavioural Ecology and Sociobiology 36:91–100. Moore, W. G., and R. L. Marchinton. 1974. Marking behavior and its social function in whitetailed deer. Pages 447–456 in V. Geist and F. Walther, editors, The behaviour of ungulates and its relation to management. IUCN New Series, No.24. International Union for the Conservation of Nature and Natural Resources (IUCN), Morges, Switzerland. Ohtaishi, N. 1986. Preliminary memorandum of classification, distribution and geographic variation on sika deer. Honyurui Kagaku (Mammalian Science) 53:13–17. (In Japanese.) Schaal, A. 1985a. Observations preliminaires sur le cycle sexuel du daim, Cervus (Dama) dama. Mammalia 49:288–291. Schaal, A. 1985b. Variation of mating system in fallow deer (Dama dama). Abstracts 19th International Ethological Conference 1:277. Toulouse, France. Schaal, A., and J. W. Bradbury, 1987. Lek breeding in a deer species. Biology of Behaviour 12:28–32. Takatsuki, S. 1991. Feeding ecology of herbivores. Pages 119–137 in M. Asahi and T. Kawamichi, editors, Modern mammalogy. Asakura-shoten, Tokyo, Japan. (In Japanese.)

Chapter 22

Reproductive Ecology of Sika Deer on Kinkazan Island, Northern Japan: Reproductive Success of Males and Multi-Mating of Females Masato Minami, Nobumasa Ohnishi, Ayumi Okada, and Seiki Takatsuki

Abstract We recorded growth, reproductive behavior, survival, and reproductive success of 458 identified tame sika deer (Cervus nippon) in a park-like place on Kinkazan Island (9.6 km2 in area), northern Japan for 15 years (1989 to 2004). This island is covered by old-growth forests and no predator lives there. The males held mating territories. It seems advantageous for them to wait for females in their territories in this closed habitat where females formed loose groups rather than to follow the moving females passing through several territories. Males increased body weight up to five to six years old, and some of them became territory holders. No male lived longer than 14 years. Three ranks were recognized among the males during the rut: TD (territorial dominants), ND (nonterritorial dominants), and SB (subordinates). Categorization for 15 years showed that TD, ND, and SB accounted for 19.4%, 10.8%, and 69.8% of the males, respectively. Males who became TD at least once accounted for 16.5% among the all males. Males that were dominant but did not have territories were 2.9% of all males. The remaining males (80.6%) lived as SB through their lives. Territoral dominant males had heavier body weights than ND and SB males. Reproductive success of TD was high: they monopolized as many as 67.2% of the matings. The females mated with multiple males several times during a rut, which has not been reported in red deer (Cervus elaphus). Territorial dominant male sika deer guarded estrous females both before and after mating. Male behaviors, rank promotion related to fighting ability, and reproductive success of sika deer were similar to those of red deer. However, the rutting territory was different from the mobile harem of red deer which live in more open habitats. We discuss the significance of multimating and multimale mating and point out the need to determine the occurrence of this behavior in other cervid species.

Introduction In order to understand the reproductive strategy and social structure of a population of long-lived animals in relation to reproductive success, a long-term observational study based on individual identification is necessary. Such studies on red deer D. R. McCullough et al. (eds.), Sika Deer: Biology and Management of Native and Introduced Populations, DOI: 10.1007/978-4-431-09429-6_22, © Springer 2009

297

298

M. Minami et al.

(Clutton-Brock et al. 1982), Soay sheep (Ovis aries, Clutton-Brock and Pemberton 2004), bighorn sheep (Ovis canadensis, Festa-Bianchet et al. 1997), mountain goat (Oreamnos americanus, Festa-Bianchet et al. 1994), and northern elephant seal (Mirounga angustirostris, Le Boeuf and Reiter 1988) have brought tremendous contributions to our understanding of behavior and ecology. They yield fruits to socio-ecology as well as evolutionary ecology. Such studies, however, are limited because of difficulties in collecting data on sexual behavior and reproductive success of large mammal individuals over long lifetimes, and more case studies are needed. We began a long-term study of sika deer which, we hoped, would be equivalent to the above-mentioned studies. The target population inhabits Kinkazan Island, a small island (9.6 km2) in northern Japan which has been free from predation and hunting. The deer are tame, which permitted us close observation and livecapturing. We identified all the members of the population that we observed for 15 years, being equivalent to their life span, and have collected data on growth, social relations, reproductive behavior, and reproductive success. Here we report the results of our observations of male sika deer. There are a few studies on reproductive behavior of sika deer. Detailed observation was done at Nara Park, western Japan (Takaragawa and Kawamichi 1977a, b; Miura 1984), on Nozaki Island, west-central Japan (Endo and Doi 2002; Endo chapter 21) and on Nakanoshima Island, Hokkaido in northern Japan (H. Takahashi et al. unpublished).

Methods On Kinkazan Island, our study site, plants and animals have been conserved because of religious reasons. About 500 sika deer live on the island, but there are no predators and hunting is prohibited. Our intensive study area is situated at the northwestern part of the island where a shrine exists. About 150 sika deer live in the park of the shrine (Fig. 22.1). The park is dominated by a lawn-like Zoysia japonica community and cherry and pine trees grow sporadically. This park is surrounded by natural forests such as Abies firma (Japanese fir) and Carpinus tschonoskii (hornbeam). The sika deer subsist on Zoysia japonica (Japanese lawngrass, but they are afforded cookies by visitors and partially subsist on garbage. The deer are tame, and we could observe them from a close distance. We identified all the deer of the population at the park by natural characteristics like facial appearance, scars, spot patterns, and hair color, as well as by liquid hair dying, freeze branding, ear notching, and microchip insertion. We observed the deer from 1989 to 2004. Intensive observations were done in autumn (from late September to November of 1989 to 2004) and in early summer (from May to June of 1994 to 2004). During these periods, we determined target animals and continuously traced and recorded their behavior. Rutting behavior was recorded following the methods of Miura (1984). Male behaviorial interactions with females and other males, marking behavior, and vocalization were recorded. We did not observe at night time.

22 Reproductive Ecology of Sika Deer on Kinkazan Island, Northern Japan

299

Fig. 22.1 Sika deer grazing on the Zoysia japonica lawn in the park of the shrine on Kinkazan Island, northern Japan (August 1996).

Fig. 22.2 A sika deer male laid on the platform of an electronic balance.

We live-captured the deer in March and October/December (Minami et al. 1992). We used a tranquilizing rifle when necessary. We weighed the deer by a platform-type electronic weighing scale designed for livestock (Fig. 22.2). This enabled us to weigh several tame animals without capture when they stood on the scale platform. It is known that antlers function to determine social rank and fighting success of males, but the males of the target population were confined in a paddock in summer for the traditional autumn festival. At the festival, the antlers were

300

M. Minami et al.

removed by saw, and thereafter the males were released from the paddock. Age of the deer born before 1988 was unknown, but for all the younger deer, we knew not only their age but the identity of their mothers.

Results and Discussion Population The size of the study area is about 30 ha in which about 150 deer live (mean density is about 470/km2). The population was largely stable for the 15 years of the study, with the exception of a mass mortality in 1997, which was reflected in the study population (Table 22.1). During the mass mortality more males died than females. Over the study the sex ratio (male/female) was biased towards females with an average of 0.84 males/female (range from 0.51 to 1.03). The sex ratio was lower (0.67) for deer older than four years reflecting a male bias in mortality.

Male and Female Life in the Non-rut Season There were four major groups in the population. On a daily basis they came to the park from the surrounding forests, stayed at the park from morning to evening, Table 22.1 Yearly changes in the composition of the study population of sika deer on Kinkazan Island. 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 Male Over 4 years old 1–3 years old 0 years old Subtotal Female Over 4 years old 1–3 years old 0 years old Subtotal Total

25

28

33

33

45

36

33

21

26

31

31

29

37

23

28

25

19

14

19

18

13

12

22

26

25

17

16

14

11

8

12

6

12

4

14

10

10

4

14

64

70

69

60

71

61

63

38

52

63

67

58

68

48

49

52

49

54

55

54

48

43

45

44

39

44

5

17

24

23

17

17

21

25

16

19

17

24

16

13

16

8

5

9

9

14

2

12

4

11

4

6

66 130

82 152

84 153

77 137

80 151

81 142

89 152

75 113

71 123

68 131

72 139

67 125

66 134

22 Reproductive Ecology of Sika Deer on Kinkazan Island, Northern Japan

301

and returned to the forests at night. They used foraging grounds and resting sites in the park during the daytime hours, then moved to night bedding sites in the forests. Females lived in family groups. The places they used were very small in size (1 km2) and stable. That is, females did not move away from the park, but rather lived out their lives at the place of their birth. Sika deer living in mountain areas of other places in Japan often show up-down migratory movements (Maruyama 1981; Takatsuki et al. 2000; Yabe and Takatsuki chapter 20) or horizontal long-distance migration in Hokkaido (Igota et al. 2004, Igota et al. Chapter 19), but the sika deer on Kinkazan Island were sedentary. The four groups staying in four different places in the forests came to the park from four different directions and returned there every evening. The group members were stable. Each group was composed of several maternal units, and these units commonly used specific foraging grounds and resting sites and took the same travel routes each day to move from forest to park, and back again. Fawns and yearlings, both males and females, stayed together. Other females stayed in the family groups. Though young females giving birth temporarily left the groups and became independent, they stayed in the same home ranges. The first parturition occurred at three years old at the earliest and after four years old in most cases (Minami et al. chapter 23). This was apparently later than in most other populations in Japan. For example, at Mt. Goyo on the mainland, close to Kinkazan Island, about half of the two-year-old females gave birth, and the pregnancy rate was more than 80% at ages older than three years (Takatsuki 1992). Males usually stayed in family groups during the non-rut season. Sometimes they formed male-only groups. The members of such male groups were unstable and changed daily.

Growth and Life of Males Males at the age of two or three years stayed with their mothers but gradually enlarged their home ranges. Some of them followed older males and entered maleonly groups during the non-rut season. Males three and four years old became independent from family groups, and their home ranges became larger. Some males were found in different places in the park or even outside the park. Some fast-growing young males showed rutting behavior as early as two years of age. They joined rutting males when the latter approached females and attempted to mate. The youngest age of males that were observed to successfully mate was three years old. There were 83 males that lived longer than three years, but only two males mated as three year olds. One of them was the heaviest male among his cohort. Heavier males showed more active rutting behavior than lighter males of the same age. These heavy young males approached estrous females and attempted to copulate, occasionally gaining a few opportunities.

302

M. Minami et al.

Male Rank Categories There were several dominant males in the study population. We termed them “territorial dominant males” (TD). A TD male showed defense behavior at the borders of his territory and did not permit intrusion by other TD males. A TD was socially dominant in his territory and had priority access to estrous females. Among the males who did not hold territories, there were males who behaved dominantly except for the lack of defense of a territory. We termed them “non-territorial dominant males” (ND). All other males were categorized as “subordinate males” (SB). This category included young males, old males beyond their prime, males that did not show strong rutting activities, and rutting males of lower rank.

Body Weight and Growth Males captured and weighed in October whose ages were known are shown in Fig. 22.3. Male body weight increased up to around five years of age, and TD males were first observed at this age. TD and ND males were heavier than SB males at the same age. Body weights of SB males peaked at seven years old and became lighter thereafter. Older SB males included those who greatly lost weight during the rut. There were no males older than 15 years. Sika deer males continued to be involved in rut until the year of their death.

Fig. 22.3 Body weight growth of different male categories. (SB = subordinate male, ND = nonterritorial dominant male, TD = territorial dominant male).

22 Reproductive Ecology of Sika Deer on Kinkazan Island, Northern Japan

303

Age and Rank of Males Male categories at each age are shown in Fig. 22.4. Territorial dominant and ND males appeared at five years of age and older and became more predominant after 10 years old. Among 17 males that became TDs, 11 (64.7%) attained their highest social rank during the ages of seven to nine years. Four males attained the highest rank at five or six years, while another two did at 10 or more years. There were two types of males. One type became TDs earlier, maintained their rank, and ended their life as dominants, whereas a second type became TDs or NDs later, maintained the rank for a short period or, in extreme cases, they did not become dominants at all. Among the 13 males older than eight years that died during the study period, three finished their lives as SBs, and three never rose above being SBs. This means that older TD and ND males were not dominant because they lived long, but that those that become dominants also lived long. In other words, a high proportion of dominants in older age classes was because they were inherently good quality in all respects. Males born after 1990 that died before 2004 accounted for 103 individuals. Among them, only 17 (16.5%) became TDs at least once. Three males (2.9%) did not become TD, but only SB males. Only the 17 males that reached heavy body weights and lived long became TDs or NDs. The other 83 males (80.6%) become neither TDs or NDs. It is an apparently advantageous for males to attain high social rank earlier in life, but at the same time it is accompanied by risk. It is usual that young TD males

90

Number of animals

80

unknown

70

SB

60

ND

50

TD

40 30 20 10 0

0

1

2

3

4

5

6 7 8 Age (yr)

9

10 11 12 13

Fig. 22.4 Numbers of each social class of sika deer males at each year of age on Kinkazan Island.

304

M. Minami et al.

would face great energy expenditure to maintain their rank. This could cause growth inhibition or increased winter mortality. One way to avoid this is by shortening the period of rut to reduce the cost. Most of the TD males reached that category at ages from six to 11 years. This corresponded with the age of red deer where males holding harems of females increased between six and 11 years old (Clutton-Brock et al. 1982). Males of both species maintained harems or territories until their deaths. It seems that success in reproduction becomes strongest at this age in both species.

Rut and Territory Formation We heard rutting calls (color plate 2a) from late August, which was the same time as shown by the population in Nara Park, western Japan (Miura 1984). Detailed descriptions of the rutting call is given elsewhere (Minami 1997). Velvet antlers turned hard in late September, and antagonistic behavior and marking behavior (color plate 2b) became frequent at that time. Uncommon movements of males were observed in this period: strange males that lived outside the park visited, and known males left the park. Dominant males established territories in early October and kept them until early November. Towards the end of this period some males were exhausted and the rut declined. The number of territory holder decreased in middle and late November, and borders of the remaining territories became less clearly defined. A few rutting dominant males wandered over a wider area containing several former territories. Rutting behavior of young males was delayed and extended later than that of older males. The earliest copulation during the 15-year study was observed on October 4 while the latest one was observed on November 27. The peak of mating occurred in middle October (Ohnishi et al. chapter 7). According to the dates of parturition and pregnancy duration, there were a few earlier and later matings than these, but they were uncommon.

Comparison of Male Reproductive Systems There are four reproductive systems among Cervidae. The harem is the system where a male defends not an area but females; it is often found in situations where female groups are gregarious and stable (Fig. 22.5, right). A male progressively mates with estrous females in the harem. This system is advantageous in an open habitat since females in under such circumstances are more gregarious. The harem system of red deer on Rhum, Scotland, is a typical example (Lincoln and Guiness 1977; Clutton-Brock et al. 1982). A territory is defined as an area having any resources that are defended against rivals, and there are several types including reproductive and foraging territories

22 Reproductive Ecology of Sika Deer on Kinkazan Island, Northern Japan

305

Fig. 22.5 Diagram of reproductive systems of Cervidae. Left shows a territory, right a harem.

(Nobel 1939; Burt 1943; Hinde 1956; Brown 1964). A reproductive territory is a system to defend a piece of land in order to copulate with females who visit the land for food, water, and/or resting (Fig. 22.5, left). Male sika deer on Kinkazan Island adopted the reproductive territory. Territorial dominant males (TD) defended their areas which included foraging grounds and resting sites for female groups. TDs defended temporary female foraging or resting groups within their territories. The females did not form larger tight groups because of the park-like landscape and absence of predation and hunting; they moved freely across territories, and the composition and the size of the female groups continually changed, even within a day. The TDs looked for females passing through their own territories and mated with estrous females among them. In a “lek system”, a group of males forms a cluster of quite small adjacent territories and each attempts to mate with a female visiting the lek and entering his small territory. This system is found when female density is quite high at a locality (Clutton-Brock 1989). This seems similar to the territory system shown by sika deer, but is different in that a lek does not contain resources (e.g., food and resting sites), that attract females; instead, females visit the lek expressly in order to mate with a male. Among cervids, this system is shown by fallow deer (Dama dama, Clutton-Brock et al. 1988; Apollonio et al. 1990). In the fourth system, “roving,” a male wanders to find females, and once he finds an estrous female, he temporarily defends her. This system is sometimes referred to as a “tending bond.” It is usually found when the female density is low and dispersed, and consequently determining the location of females is not predictable. They must be searched for and located. A typical example of roving is found in moose (Alces alces, Peterson 1955). From these comparisons, it is apparent that reproductive systems of cervids are affected by the density, gregariousness, and movements of females.

306

M. Minami et al.

The reason why males on Kinkazan Island adopted the territory system seems to be because female groups were less gregarious and unstable. The females were composed of dispersed individuals or of small groups, and the members continually joined and left these groups. If a male formed a harem and followed a female group, because the female groups change the chance of mating would be reduced by the harem system. The TD males tried to herd females into a group within their territories (color plate 2d). This would be because it is necessary for them to check the reproductive statuses of the females and increase the possibility of successful mating. Territorial dominants repeatedly herded the females within their territories by side-stretch behavior (color plate 2e), head-up display, and herding, but once a female left the territory, the TD did not follow her onto other territories. A female group sometimes became as big as 30 deer, but quickly broke down into smaller units composed of only a few individuals who entered the forest separately in the evening. Because female groups frequently fragmented, it was impossible for the males to defend females for long periods as males in the harem system do. Since the females on Kinkazan Island show mating with multiple males as their estrus continues for about a day, it is necessary for males to guard them at least for a day in order to ensure paternity (see below). However, if a male guards an individual female he has copulated with, he cannot search for other estrous females. It is more efficient to find estrous females within his territory than to try to herd a whole group of females. Since the deer density is quite high in the study area, females including estrous ones frequently pass the male territories. The high density of females is probably an important factor favoring the males adopting the territory system. In contrast, red deer on Rhum adopted the harem system. Harem sizes were variable, and many harems were composed of only fewer than four females (Lincoln and Guiness 1977). This figure is much smaller than the number of females staying in the territory on Kinkazan Island. Since the deer density on Rhum (13.9 deer/km2, Clutton-Brock et al. 1982) is less than 10% of that on Kinkazan Island, numbers of females staying around males must be fewer. It is reported that joining and leaving of red deer females to and from harems is frequent (Lincoln and Guiness 1977), but the frequency must be lower than that on Kinkazan Island according to the lower deer density. Besides density, the open habitat of Rhum would cause gregariousness of females, and more permanent group formation. In such a habitat, it would be more advantageous for males to defend relatively stable existing female groups rather than to hold a territory and wait for the females to enter it. It is known that social system of cervids during the rut is flexible depending upon environmental conditions (red deer, Carranza et al. 1990; fallow deer, Apollonio et al. 1990). Clutton-Brock (1989) has reported that sociality of cervids, both within and among species, is affected by home range size of females, size and stability of female groups, and densities and distributions of females. For example, red deer form harems in open habitats while they form territories in the mosaic habitat composed of forest and openland (Carraza et al. 1990). Fallow deer form leks in less forested habitat, while they form territories in forested habitats (Apollonio et al.

22 Reproductive Ecology of Sika Deer on Kinkazan Island, Northern Japan

307

1990). They also form leks at high population density, while they form harems or territories at a low density (Clutton-Brock et al. 1988). Studies on sika deer have also shown the flexibility of the social system during the rut. The males on Nozaki Island, western Japan, where dense forests develop, adopt the “consorting female” system, while the males in Nara Park in western Japan, where park-land landscape is developed, adopt the territory system (Miura 1986) like those on Kinkazan Island (this study). Our study has shown that the best option in an environment where a population at a quite high density inhabits a parkland habitat is to form a territory and mate with females passing through the territory rather than to form a harem. It is plausible that a population living in forests at a low density would have a different social system. If a population lives in an open habitat at a low density, the harem system to guard one or two maternal unit(s) would be more advantageous, whereas for a population living in a dense forest at a low density, the “consorting female” system would be more advantageous because holding a territory in such an environment would be costly. It is suggestive that males on Kinkazan Island whose territories were situated at the peripheral parts of the park, and included forests, often waited for females in the park in the morning and evening when the females moved from and into forests, while they stayed in the forest to look for females (consorting female system) during daytime when the majority of the females stayed in the park to forage where they were herded by stronger TDs. This observation suggests that a male changes his rutting behavior depending on the density of females and visibility within the habitat. In a red deer population in Spain, both the territory system and the mobile harem system coexisted (Carranza et al. 1990).

Social Rank of Males and Reproductive Success Males showed rut-related behaviors such as rut call (color plate 2a), aggressive approach, chasing, head-up display (Fig. 22.6), head-down display, antler clash (color plate 2c), and parallel walk. All these behaviors were described in Nara Park, western Japan by Miura (1984). We recognized rank among the males through these behaviors. There were differences in vocalization among the three male categories. TD males frequently gave two kinds of rut calls that NDs called less frequently, and SBs rarely called (Minami 1997). NDs had home ranges overlapping one to three territories of other TDs. NDs sometimes showed defensive behavior against neighbor TDs in their territories. When a TD was absent, an ND male behaved as if he were the TD and approached females. A TD aggressively and persistently expelled NDs when they approached females. However, he did not when NDs stayed on the peripheral parts of his territory. Some SBs eagerly approached females, while others did not show interest in females. TDs expelled SBs, but if SBs did not approach females, TDs permitted them to stay in their territories. Thus, it was the behavior of other males more than their presence that TD males responded to.

308

M. Minami et al.

Fig. 22.6 Sika deer male giving a head-up display.

Dominance was affected by intensity of rutting motivation. TD males did not intensively expelled NDs and SBs near the end of the rut. TD and ND males became SBs after cessation of the rut. Some ND males became dominant and behaved as if they were TDs after TDs abandoned their territories.

Body Weight and Rank Category We compared the body weights during the rut of the three categories of males older than five years (Fig. 22.7). The mean body weights of TD, ND, and SB males were significantly different from one another (TD:ND p < 0.05, ND:SB p < 0.001, TD: SB p < 0.0001, U-test). There seems to be a trade-off between the cost of holding territory and the benefits from doing so. The cost would be the energy burden of expelling rivals, while the benefits would be success in mating with females. As territory size increases, the number of females would increase, but so would the numbers of rival males. This would be particularly the case under a high deer density. In such a situation, discrimination of rivals would be meaningful in order to reduce energy expenditure. It is noteworthy that Miura (1984) who carried out an intensive observation of sika deer in Nara Park, did not describe ND males in the population. One of the present authors (Minami) also observed Nara males and did not recognize NDs. Thus, we think that the existence of ND males in the Kinkazan population is unique and attributable to the high population density. Under such a condition, complete expelling of rival males is quite costly and too exhaustive or even dangerous for TD males.

22 Reproductive Ecology of Sika Deer on Kinkazan Island, Northern Japan

309

Fig. 22.7 Mean body weights of social classes of male sika deer on Kinkazan Island. Vertical lines indicate SD.

Thus, allowing ND males to stay in the territory if they did not approach females avoided much expenditure of energy.

Rank Category of Males and Reproductive Success Territorial dominant males are heavy, experienced, and their antlers are long (Takatsuki et al. unpublished) and, therefore, their social rank is high. Furthermore, in the territory system, males at higher rank (not always top ranked) can avoid unnecessary fighting with other males while securing greater opportunities of mating. High rank promises more mating, which increases reproductive success. This results in monopolization of females by the top-ranked males (Table 22.2). Among the 692 copulations observed during the 15 years of the study, TD males accounted for 67.2% while NDs and SBs accounted for only 10.3% and 19.6% respectively. The mean number of copulations per male of these categories was 4.8, 1.3, and 0.4 for TD, ND, and SB males, respectively. However, variation between individuals within each male category was great. Within TD males the maximum number of copulated females by a given male was 22, while the minimum was zero. This was similar to a red deer harem system where male copulations varied from one to 22 (Lincoln and Guiness 1977). Many of the SBs did not mate. And, not all males advanced in category from SB through ND to TD; some males remained SBs throughout their life. A long-term study to determine lifetime reproductive success of each individual is needed.

310

M. Minami et al.

Mating A TD male always monitored the estrous status of females by smelling their vagina or urine. When no estrous female was available, the TD protected female groups to form a temporary “harem” on his territory and continuously monitored the members. Once the TD became aware of a female approaching estrus, his attention was mainly on her, and later, when she entered estrus, he concentrated on guarding her. He conducted specific behaviors such as sexual approach (Fig. 22.8a), low–stretch follow, smelling perineum, and grooming (Fig. 22.8b). Eventually this led to mounting (Fig. 22.8c) and copulation (color plate 2f), as described in the Nara Park population by Miura (1984). Females were typically receptive to copulations for only 24 h (Minami 1997). Many of them mated multiple times during the estrous period with the same male, and also often mated with multiple males, which revealed promiscuous mating relationships. Mating relationships in 1990, when we observed the largest number of matings are shown in Table 22.2. A total of 117 matings were observed among 27 males and 17 females. As many as 82.4% of females copulated with multiple males. On Kinkazan Island, promiscuous mating was common. We observed a total of 147 cases of estrus for nine years from 1990 to 1998. Among the 147 cases, multiple mating with a given male occurred in 90 cases (61.2%), and multi-male mating occurred in 56 cases (38.2%) (Fig. 22.9). The maximum number of different males a female mated in one duration of estrus was eight. Endo et al. (1997) also reported multi-male mating by females on Nozaki Island, western Japan, where sika deer live in as high a population density as on Kinkazan Island. No information is available on the mating system of sika deer at low population density.

Fig. 22.8 Behaviors relating to mating: (a) low-stretch; (b) grooming; (c) mounting.

Status

TD TD TD TD TD TD ND SB SB SB SB SB SB SB SB SB SB

SB

SB

SB

SB

Name of male

Moyasi Sansen Meki Nori Akaoni Sirou Nakio Maro Make Tama Chime Yonsen Bio Kosako Dai Deka Gorou

Kakuo

Oten

Furaku

Go

Inoshishi 2

Urawa 1

4

1

En 1

2

2

Obakyu 3

Kaku 2 3

Kasa 1

1

3

Giga 1

1

Kibushi 1

Gibusu 3

Kuten 1

3

Ko 1 1

1

Tabaa

Name of female Tensen 1

1

1

3 1

1

1

Nanawaru 1

1 1 1

1

Nami 1

1

2 1

1 1

3

Niji 2

Nobi 1

1

2

Bui 1

1 1

2

1

2

Peke 1

Madara

1

1

1 4 1 1 1 1 1 1 2 1

8 1

(continued)

1

Mimigire

Table 22.2 Mating relationships among male and female sika deer of the study population during the 1990 rut. Figures in the columns indicate the number of copulations between the male and female. (TD = territorial dominant male, ND = nonterritorial dominant male, SB = subordinate male.)

22 Reproductive Ecology of Sika Deer on Kinkazan Island, Northern Japan 311

SB SB SB Unk Unk Unk

27

Status

Name of male

Haten Syamo Ugo Yamanogosen Unknown A Unknown B No. of copulations No. of males

Inoshishi 2 1

Urawa 6 3

En 5 3

Obakyu 4 2

1

Kaku 5 2

Kasa 5 3

Giga 2 2

Kibushi 1 1

Gibusu 3 1

Kuten 4 2

2 2

Ko

Name of female

1 1

Tabaa

Table 22.2 (continued)

Tensen 10 8

1

Nanawaru 5 5

Nami 11 8

1

Niji 2 1

Nobi 5 4

1

Bui 1 10 8

1

Peke 1 1

Madara 1 1

Mimigire 25 14

312 M. Minami et al.

22 Reproductive Ecology of Sika Deer on Kinkazan Island, Northern Japan

313

Fig. 22.9 Composition of mating. Left bar: percent of females copulating with one to four or more different males; right bar: percent of females copulating one to four or more times.

Mate Guarding Males often showed postcopulatory mate guarding (Fig. 22.10) in addition to precopulatory guarding. They protected the mated female from access by the other males and also prevented the female from moving out of their control. A TD male chased and kept the female within his territory when she attempted to escape. Dominant males guarded the estrous female for longer periods and copulated repeatedly with her at intervals of several hours. Postcopulatory guarding is important since the male can assure his paternity if he can protect the female for the duration of the estrous period. However, the guarded females occasionally mated with other males that escaped the attention of the guarding male. The guarding male, once he recognized the mating of the guarded female with another male, immediately copulated with her again. This suggests that a male would be able to reduce the possibility of fertilization by other males through internal sperm competition (Fig. 22.11). One reason SBs obtained a smaller number of matings was probably because of their lack of success in postcopulatory guarding. Subordinate males occasionally made “sneaking” approaches to the estrous female and guarding male couple and tried to sneak a mating without the notice of the guarding male. Another tactic used by sneaking SBs took advantage of alarm behavior. We observed an SB approach the pair while stotting (stiff-legged bounding gait), give alarm calls, roughly stamp the ground, and dash bodily against the female. Then the SB attempted to mate with her during the running and confusion, which disrupted the usual territorial behavior. Such a chaotic situation, a “rut panic,” attracted the surrounding males, and ten or more males pursued the female attempting to copulate (Fig. 22.12). Multiple mating would provide SB males with the chance of reproduction and promote their sneaking behavior.

314

M. Minami et al.

Fig. 22.10 Guarding of an estrous female by a territorial dominant male.

Fig. 22.11 Reproductive strategies. (A) Mono-mating, (B) Multiple mating, (C) Guard the copulated female to avoid other matings, (D) Repeated copulations.

Similar sneaking behavior with alarm calls was reported also for red deer (Clutton-Brock et al. 1982). We observed several successful cases of this tactic by SBs, but in most cases, the guarding TD regained her again or another TD acquired her. Estrous females sometimes escaped guarding by their own initiative and entered the neighboring territory, or escaped by running a long distance.

22 Reproductive Ecology of Sika Deer on Kinkazan Island, Northern Japan

315

Fig. 22.12 Males running in a “rut panic”.

Multimating and Reproductive Success We observed multiple mating in sika deer on Kinkazan Island, which was not reported in red deer. The multiple mating in sika deer provides males with many alternative tactics. For a dominant male, there can be alternative behaviors: whether he guards a female through the estrous period to assure the reproductive success with the female (Fig. 22.11), or he switches to guarding another female to increase the number of females he mates with. Consequently, a TD sometimes protected the female for a long time, for a short time, or sometimes completely abandoned guarding her. The standard of decision for males is unclear, but the males may decide on the basis of whether or not the female had mated previously with other males. Additionally, duration of an estrus is 24 h, although the possibility of fertilization is not likely to be equal throughout the period. Males might change their decision about whether, and how long to guard by knowing the time in the estrous period of the female, and likelihood of conceiving. The benefit of the multiple mating for females is unclear. Presumably multiple mating leads to assured fertilization for females. Multiple mating covers the best timing of fertilization at higher possibility than a single one, therefore, multiple mating should raise the fertilization success. Another possible explanation is that by securing guarding by a given TD, females avoid the risk or cost of refusing mat-

316

M. Minami et al.

ing. We often observed that females were exhausted by being pursued by males, and in one case a female was injured. However, females cannot avoid being chased by males because many males are attracted by seeing mating behavior in an open habitat. It may be better for females to reduce the risk of being chased by males by opting to be guarded by a strong male in his territory. Nevertheless, we often observed that females intentionally escaped the guarding of a male by moving into the control of other males (in most cases, another TD), where they mated. The female may be able to obtain successful genes as a result of internal sperm competition, assuming there is a positive correlation of male physical strength with sperm activity, or with success in fertilization. The guarding TD male usually mated a female at several hour intervals under successful guarding, but mated immediately after finding the guarded female had mated with other males. Thus, multiple mating appears adaptive for both sexes, and evolution in the direction of increasing the number of matings is favored. As studies on sika deer populations at lower densities have not been done, we do not know if the behavior observed on Kinkazan Island and Nozaki Island (Endo and Doi 2002) is common or not. However, when females have the physical capacity of mating multiple times, and if the multiple mating is beneficial for the females, it is quite expected that females at a lower density do multiple mating. Additionally, the observed multimale mating in sika deer must not be exceptional as it was also reported in reindeer that form multi-male groups (Hirotani 1994). Before assessing the universality of multiple mating, we need to investigate the evolutionary significance of multiple mating for both males and females. Unfortunately, studies on individual variations in fertilization success among males, and possible mate choice by females for sika are limited. It is necessary to confirm whether such matings exist in other Cervus species or not. Such studies are necessary to understand the significance of multiple mating and evolution of these behaviors.

Literature Cited Apollonio, M., M. Festa-Bianchet, F. Mari, and M. Riva. 1990. Site-specific asymmetries in male copulatory success in a fallow deer lek. Animal Behaviour 39:205–212. Brown, J. L. 1964. The evolution of diversity in avian territorial systems. Wilson Bulletin 76:160–169. Burt, W. H. 1943. Territoriality and home range concepts as applied to mammals. Journal of Mammalogy 24:346–352. Carranza, J., F. Alvarez, and T. Redondo. 1990. Territoriality as a mating strategy in red deer. Animal Behaviour 40:79–88. Clutton-Brock, T. H. 1989. Mammalian mating systems. Proceedings of the Royal Society of London, Series B 236:339–372. Clutton-Brock, T. H., and J. Pemberton. 2004. Soay sheep. Cambridge University Press, Cambridge, United Kingdom. Clutton-Brock, T. H., F. E. Guiness, and S. D. Albon. 1982. Red deer: Behavior and ecology of two sexes. University of Chicago Press, Chicago, Illinois, USA.

22 Reproductive Ecology of Sika Deer on Kinkazan Island, Northern Japan

317

Clutton-Brock, T. H., D. Green, M. Hiraiwa-Hasegawa, and S. D. Albon. 1988. Passing the buck: Resource defense, lek breeding and mate choice in fallow deer. Behavioral Ecology and Sociobiology 23:281–296. Endo, A., and T. Doi. 2002. Multiple copulations and post-copulatory guarding in a free-living population of sika deer (Cervus nippon). Ethology 108:739–747. Endo, A., T. Doi, and A. Shiraki. 1997. Post-copulative guarding; mating behavior of non-territorial male sika deer (Cervus nippon) in an enclosure. Applied Animal Behavior Science 54:257–263. Festa-Bianchet, M., M. Urquhart, and K. G. Smith. 1994. Mountain goat recruitment: Kid production and survival to breeding age. Canadian Journal of Zoology 72:22–27. Festa-Bianchet, M., J. T. Jorgenson, C. H. Berube, C. Portier, and W. D. Wishart. 1997. Body mass and survival of bighorn sheep. Canadian Journal of Zoology 75:1372–1379. Hinde, R. A. 1956. The biological significance of the territories of birds. Ibis 98:340–369. Hirotani, A. 1994. Dominance rank, copulatory behaviour and estimated reproductive success in male reindeer. Animal Behaviour 48:929–936. Igota, H., M. Sakuragi, H. Uno, K. Kajio, M. Kaneko, R. Akamatsu, and K. Maekawa. 2004. Seasonal migration patterns of female sika deer in eastern Hokkaido, Japan. Ecological Research 19:169–178. Le Boeuf, B. L., and J. Reiter. 1988. Lifetime reproductive success in northern elephant seals. Pages 344–362 in T. H. Clutton-Brock, editor, Reproductive success. University of Chicago Press, Chicago, Illinois, USA. Lincoln, G. A., and Guiness, F. E. 1977. Sexual selection in an herd of red deer. Pages 33–38 in J. H. Calaby and C. H. Tyndale-Biscoe, editors, Reproduction and evolution. Australian Academy of Sciences, Canberra, Australia. Maruyama, N. 1981. A study of the seasonal movements and aggregation patterns of sika deer. Bulletin of the Faculty of Agriculture, Tokyo University of Agriculture and Technology 23:1–85. (In Japanese with English summary.) Minami, M. 1997. Vocal repertoire and the functions of vocalization in the rutting season in sika deer Cervus nippon. PhD thesis, Osaka City University, Osaka, Japan. Minami, M., N. Ohnishi, S. Takatsuki, and N. Hama. 1992. Capturing sika deer on Kinkazan Island. Honyurui Kagaku (Mammal Science) 3:23–30. (In Japanese.) Miura, S. 1984. Social behavior and territoriality in sika deer (Cervus nippon Temminck 1838) during the rut. Zeitschrift für Tierpsychologie 64:33–73. Miura, S. 1986. Sika deer, their diversity of ecology and society. Pages 90–93 in The encyclopedia of animals of the world IV. Heibon-sha Publishing Company, Tokyo, Japan. (In Japanese.) Noble, G. K. 1939. The role of dominance in the social life of birds. Auk 56:264–273. Peterson, R. L. 1955. North American moose. University of Toronto Press, Toronto, Ontario, Canada. Takaragawa, N., and T. Kawamichi. 1977a. Rutting behavior of Cervus nippon in Nara Park. Annual Report of the Nara Deer Research Association 3:43–61. (In Japanese.) Takaragawa, N., and T. Kawamichi. 1977b. Home ranges and social organization of rutting males of Cervus nippon in Nara Park. Annual Report of the Nara Deer Research Association 3:63–80. (In Japanese.) Takatsuki, S. 1992. Sika deer living in the north. Dobutsu-sha Publishers, Tokyo, Japan. (In Japanese.) Takatsuki, S., K. Suzuki, and H. Higashi. 2000. Seasonal up-down movements of sika deer at Mt. Goyo, northern Japan. Mammal Study 25:107–114.

Chapter 23

Life-Time Reproductive Success of Female Sika Deer on Kinkazan Island, Northern Japan Masato Minami, Nobumasa Ohnishi, Naoko Higuchi, Ayumi Okada, and Seiki Takatsuki

Abstract Life-time reproductive success (LRS) of female sika deer (Cervus nippon) on Kinkazan Island (9.6 km2 in area), northern Japan, was studied based on individually identified females for 17 years from 1989 to 2005. Birth rate was almost zero for females under four years age, while at four it suddenly increased to 39%. Birth rates were about 50% thereafter. Individual females gave birth usually every second year. In the years when mothers were nursing offspring they neither gained body weight nor came into estrus. In the next “yeld” year (no fawn), they came into estrus after gaining body weight. When a mother lost a fawn within 10 days after birth, she recovered body weight. As many as 63% of females died without giving birth to any offspring. A LRS of three offspring was most frequent (16.3%), and more or fewer offspring were less common. No female gave birth to more than six fawns. LRS was greater in longer-living females, and neither concentrating parturition in the prime age classes nor skipping parturition to achieve a longer life was observed. These results seemed to reflect poor food-quality habitat due to the extremely high deer density on the island.

Introduction The principle of natural selection is to select advantageous traits to maximize the number of offspring (Trivers 1985) and, therefore, finding the factors affecting traits is important to understand the role of natural selection in the evolution of a species. It is, however, generally difficult to determine these factors. Such studies require concrete information of the life histories of a large number of individual animals and their habitat characteristics. A few studies of such cases are red deer (Cervus elaphus) on Rhum (Clutton-Brock et al. 1982) and Soay sheep (Ovis aries) on St. Kilda (Clutton-Brock and Pemberton 2004) in Scotland. We need more case studies on more animal species in different habitats to determine patterns for large mammals. Determining females’ reproductive success in mammals is easier than males’ because mother-infant relations can be determined by directly observing parturition D. R. McCullough et al. (eds.), Sika Deer: Biology and Management of Native and Introduced Populations, DOI: 10.1007/978-4-431-09429-6_23, © Springer 2009

319

320

M. Minami et al.

and lactation. The reproductive success of females is affected by various innate and external (environmental) factors. Among them, nutritional condition of females relating to food resources and predation of offspring are particularly important (Saether 1997). There is no study of life-time reproductive success (LRS) in sika deer because of difficulties of individual identification and long-term study required to cover lifetimes of the sample animals. Sika deer on Kinkazan Island, northern Japan, are tame and close observation is possible. Weighing them is also possible, and we can know their nutritional condition. Close-up observation enables us to determine offspring mortality. No predators live on Kinkazan Island (although crows, Corvus macrorhynchos, sometimes attack new-born fawns), which enables us to specify factors affecting reproductive success. The objective of the present study was to describe how female sika deer under poor food conditions maximized their LRS. Since the deer density on the island is high (Takatsuki 2006), the deer always face food shortage and nutritional conditions are often poor. Therefore, the burden of reproduction for females is heavy. Therefore, we should be cautious to note that the Kinkazan population under such conditions is not representative of sika deer populations in other areas with different population densities and habitat qualities.

Study Population We studied the females of a local sika deer population living around a shrine in the northeastern part of Kinkazan Island. The target animals were 137 females among 472 females inhabiting the study site for the 17 years between 1989 and 2005. Since no predators live there, and hunting is prohibited for religious reasons, the deer density is as high as 45/km2 over the whole island, and consequently the deer always face food shortages. The population subsists on Zoysia japonica, a lawn grass (Takatsuki 1980). The females are sedentary and move about in small home ranges every day. Parturition occurs in June, and a single fawn is born. Rut occurs in October when dominant males form rutting territories (Minami et al. chapter 22). The females copulate with multiple males multiple times. Natural mortality across the population is concentrated in March and April when the deer die of starvation.

Methods We have identified all the sika deer in the study population since 1989. Ages of the deer born before 1989 were unknown. We censused deer in every March and October and deaths, if any, were checked. New-born fawns were captured since 1994, and survival within 10 days after birth of the fawns was determined. Body weights of individual deer were determined in March and November since 1991.

23 Life-Time Reproductive Success of Female Sika Deer on Kinkazan Island

321

From 1996 onward, body weights of some deer were determined without capture because they stepped on the platform of an electronic weighing scale. Body weights of copulated females and non-copulated females were compared by the U-test. Body weights of yeld females, milk females, and females losing fawns (the fawn died within 10 days after birth) in March and those in March of the next year were compared by the Wilcoxon test.

Results Age of First Parturition and Age-Specific Birth Rates A total of 137 females was studied. The oldest female was 15 years old. Eighty-one females died before 2005 and 56 were still alive at that date. Forty-nine females (60.5%) among the dead females never gave birth. Females did not give birth until two years of age (Fig. 23.1). Only 1.3% of threeyear-old females gave birth, but birth rates increased at four years old and older. Birth rates varies around 50% from five years up to 14 years old. In the population of Mt. Goyo, northern Japan females became pregnant at two years old (about 50%) and over 80% of females older than three years became pregnant (Takatsuki 1992), and in Hyogo, western Japan a similar pattern was reported (Koizumi 1992). These differences illustrate the consequences of high deer density on Kinkazan Island and the resulting food limitations on the population.

Parturition Interval The low birth rate of Kinkazan females can be attributed to two possibilities. One is that the population contains a mix of both highly reproductive females and low

Fig. 23.1 Age-specific birth rate of female sika deer on Kinkazan Island (1989–2005).

322

M. Minami et al.

reproductive ones, and another is that most of the females alternate “pregnant years” and “rest years.” When we define a “parturition chance” as the case when a female older than four years is alive during a parturition period, we observed 227 parturition chances and the birth rate was 50.2%. For parturition chances among the 37 females that died, no birth in the year following a birth year was most common (n = 50), while three years of continuous parturition or three years of continuous “rest” was very rare. It is common for this population that females rest in the year following a parturition year. However, in the cases of early death of fawns (within 10 days after birth), 75% of the females gave birth in the next year (n = 20). These results suggest that most of the females on Kinkazan Island, where foods are always in short supply, principally give birth every second year. Repeated parturitions happened for females between four and nine years of age (one threeyear repetition and five two-year repetitions) with an exception of one female that gave birth for a continuous three-year period between 12 and 14 years of age. The ages between four and nine years old, when females achieve the best nutrition, seem to be the most reproductive, or “prime” years, since mortality of females increased after eight years of age (see Minami et al. chapter 27).

Copulation and Parturition A copulation does not directly promise pregnancy and parturition. The parturition rate of females that copulated (parturition/copulation) shown in the present study based on 129 samples was 88.4%; thus, 11.6% of the copulations were unsuccessful or fetuses were lost. This value was largely equivalent to the values (10–15%) of previous studies of cervids (Robinette et al. 1955; Teer et al. 1968; Ransom 1967; Nellis 1968; Markgren 1969). Abortion and fetal loss of red deer is “not common” (Mitchell et al. 1977). However, the accuracy of observation and parturition, as well as reproductive physiology of pregnant females were not the same, so comparison is difficult. Copulation rates varied among years, ranging from 10.3% (1996, n = 58) to 62.1% (1995, n = 58). However, 1996 was the year before a mass-mortality during the 1996–1997 winter. Other values were greater than 40% except for 29.6% in 1992 and 29.4% in 1994. The copulation rate and deer density were not correlated (r2 = 0.257, p > 0.05), suggesting it was not density-dependent. The density at the study site was as high as 500 deer/km2, and consequently grazing was quite heavy. No tall grass grew, and only Zoysia japonica, a productive lawn grass, and other comparable grazingresistant plants grew there. The deer density seemed to have reached the maximum, and the copulation rate was suppressed. It was very low in 1996, but the density was not different from the former several years, suggesting the low birth rate was not caused by a high density. However, the copulation rate recovered in the next year after the density-independent mass mortality in the winter of 1996–1997, suggesting that the copulation rate can be increased by reduced population density.

23 Life-Time Reproductive Success of Female Sika Deer on Kinkazan Island

323

Copulation and Body Weight Nutrition affects estrus, and it is therefore expected that body weight of females, which reflects their nutritional status, relates to copulation. In early October copulating females were heavier (52.5 kg, SD: 4.7, n = 69) than non-copulating females (47.5 kg, SD: 4.4, n = 67, p<0.001, U-test). These results suggest that gaining body weight is essential to achieving estrus.

Parturition, Fawn Care, and Body Weight Parturition and taking care of fawns are costly for the mother, and it is expected that these burdens would be reflected in body weight. Body weights in March of consecutive years were compared among milk females (females which gave birth), yeld females (females which did not give birth), and females that gave birth but the fawn died (Fig. 23.2). Body weight for yeld females increased by an average of 4.4 kg (12.7% of the mean body weight of 34.7 kg, n = 224), while milk female body weight decreased by 6 kg (14.7% of a mean body weight of 40.9 kg, n = 140). Females that lost their fawns decreased by 0.9 kg (2.1% of mean body weight of 41.6 kg, n = 16). Since the burden for parturition is common between milk females and females that lost fawns, the body weight difference between them seem to show the cost of taking care of fawns. This was 5.1 kg (12.5% of body weight of milk females and 12.3% of body weight of fawn-lost females). Among the 37 females that died during the study, 31 (83.8%) gave birth and produced fawns in the preceding year. This suggests the heavy burden of parturition and lactation.

Fig. 23.2 Body weights of females in March. Body weights of the same females in consecutive years are compared. Vertical lines indicate the SD (p < 0.001).

324

M. Minami et al.

Four females among 137 did not give birth over many years. Among them, two females looked unhealthy. One did not give birth up to nine years of age. She seemed to be blind in one eye, and her body hair color looked unhealthy. The other unhealthy female did not give birth up to 10 years of age. Her body weight was only 33.7 kg over five years. The other two females gave birth at a young age but did not reproduce thereafter, despite looking healthy. A female which did not give birth for 11 years maintained heavier weights (mean 41.7 kg). Another female which did not give birth for eight years lost body weight for two years after parturition (31.3 kg), but recovered thereafter (41.4 kg over five years). We observed these two to copulate several times. It is likely, therefore, that they had physiological failures to not to give birth. The fact that these “pseudo-yeld” females lived long (older than 17 years) suggests that being yeld may elongate life. There was no correlation between body weight of mothers and reproductive success, or between decrease in body weight in mothers and reproductive success. No clear relationship was found between life-time reproductive success (LRS) and body weights in March in three- and four-year olds, parturition years, and nonparturition years. Therefore, it is likely that although body weights of mothers affect estrus, copulation, and parturition, they do not affect survival of fawns.

Life-Time Reproductive Success (LRS) of the Kinkazan Females According to the data on parturitions and “parturition chances”, the mean parturition rate was determined as 43.6% (336 parturitions among 770 chances for 121 females). When we adopt the definition of reproductive success as the numbers of fawns surviving up to one year of age (Clutton-Brock et al. 1982), because the survival rate of the fawns was 74.4% the mean reproductive success for the Kinkazan population was still 32.4% (43.6% of 74.4%). The relatively high value of fawn survival on Kinkazan Island, despite poor habitat quality, can be attributed to no predation on this island. Life-time reproductive success (LRS) were determined for naturally dying females (n = 86). Since they contained as many as 49 females that produced no fawns at all, the LRS was only 0.98. However, the mean LRS for females that experienced parturition at least once was 3.08, ranging from 1 to 5 (n = 37). The value for non-aged females who experienced parturition was 3.27, ranging from 0 to 7 (n = 44). The frequency distribution of LRS shows that females that produced no fawns made up as many as 63% of all females (Fig. 23.3). Females with LRS of 3 were most abundant (16.3%) and those of more or less LRS were less frequent. Those of LRS of 4 and 5 combined accounted for only 4.7%. No female produced more than six fawns. Information on LRS for other sika deer populations is not available. LRS of red deer on the island of Rhum, Scotland, ranged more widely between individual females, from 0 to 13 (Clutton-Brock et al. 1982). The mean LRS of all the females of red deer on Rhum and sika deer on Kinkazan Island were: 4.5 and 0.98, respectively, and those of the females who fawned at least once were 7.38 and 2.63, respectively. Both values were much lower in the

23 Life-Time Reproductive Success of Female Sika Deer on Kinkazan Island

325

Fig. 23.3 Frequency distribution of lifetime reproductive success (LRS) of female sika deer on Kinkazan Island (1985–2005) (n = 86).

Kinkazan population. However, these values for Kinkazan are not representative of sika deer in general because the Kinkazan population was apparently malnourished, as suggested by low body weight, delayed first parturition, and low pregnancy rates (Fig. 23.1). Mortality of fawns is not known in other sika populations. If this value as well as those of life span and age of first parturitions were similar to those of the Kinkazan population, the LRS would be double of that of the Kinkazan population, which is close to that of red deer on Rhum.

Discussion In general, there seem to be two contrasting strategies in order to increase LRS in deer. One strategy is to concentrate reproduction at a young age, even though fawning burdens the females. Another strategy is “skipping” reproduction which lightens the burden by spreading it over time, and this enables a longer life. Sika deer on Kinkazan Island seemed to adopt the latter strategy. Although “concentrated” reproduction during a particular period of life was not found in the Kinkazan population, LRS varied greatly. Females that live a long life showed high LRS. Among the long-lived females, those that could maintain good nutritional conditions could achieve high LRS, while those which could not maintain good nutritional conditions obtained high LRS before 10 years old. For example, LRS of six females that lived up to 13 or 14 years of age ranged from three to five. Four of them continued to reproduce before their deaths, whereas the other two could not reproduce after 11 years of age. Their body weights at old ages (five of six weighings) were lighter than those at their past parturition times. This suggests that females that could not obtain enough body weight by accumulating fat deposits could not come into estrus. Mortalities of

326

M. Minami et al.

females that died young are not well-known, but many of them died during the fawning season. This suggests that mortality related to nutritional conditions. To compare the reproductive success per year between older females and those dying at the prime age, the values of LRS divided by parturition chances (“annual reproductive success”) were calculated. The annual reproductive success was not different between them (maximum 0.57, mean 0.35 for >10 years old; maximum 0.6, mean 0.45 for
Literature Cited Clutton-Brock, T. H., F. E. Guiness, and S. D. Albon. 1982. Red deer: Behaviour and ecology of two sexes. University of Chicago Press, Chicago, Illinois, USA. Clutton-Brock, T. H., and J. M. Pemberton. 2004. Soay sheep: Dynamics and selection in an island population. Cambridge University Press, Cambridge, United Kingdom. Koizumi, T. 1992. Reproductive characteristics of female sika deer, Cervus nippon, in Hyogo Prefecture, Japan. Pages 561–563 in F. G. Spitz, G. Janeau, G. Gonzalea, and S. Aulagnier, editors, Ongules/Ungulates 91, Société Française pour l’Étude et la Protection des Mammifères, Paris, France. Markgren, G. 1969. Reproduction of moose in Sweden. Viltrevy 6:127–285. Mitchell, B., B. W. Staines, and D. Welch. 1977. Ecology of red deer: A research review relevant to their management. Institute of Terrestrial Ecology, Cambridge, United Kingdom. Nellis, C. H. 1968. Productivity of mule deer on the National Bison Range, Montana. Journal of Wildlife Management 32:344–349. Ransom, B. A. 1967. Reproductive biology of white-tailed deer in Manitoba. Journal of Wildlife Management 31:114–123. Robinette W. L., J. S. Gashwiler, D. A. Jones, and H. S. Crane. 1955. Fertility of mule deer in Utah. Journal of Wildlife Management 19:115–136. Saether, B.-E. 1997. Environmental stochasticity and population dynamics of large herbivores: A search for mechanisms. Trends in Ecology and Evolution 12:143–149. Teer, J. G., J. W. Thomas, and E. A. Walker. 1968. Ecology and management of white-tailed deer in the Llano Basin of Texas. Wildlife Monographs 15:1–62. Takatsuki, S. 1980. Food habits of sika deer on Kinkazan Island. Science Reports Tohoku University, Series IV (Biology) 38:7–31. Takatsuki, S. 1992. Sika deer living in the north. Dobutsusha, Tokyo, Japan. (In Japanese.) Takatsuki, S. 2006. Ecological history of sika deer. University of Tokyo Press, Tokyo, Japan. (In Japanese.) Trivers, R. 1985. Social evolution. Benjamin/Cummings, Menlo Park, California, USA.

Chapter 24

Reproduction of Female Sika Deer in Western Japan Toru Koizumi, Shin-ichiro Hamasaki, Mayumi Kishimoto, Mayumi Yokoyama, Masato Kobayashi, and Aiko Yasutake

Abstract We describe the reproductive characteristics of female sika deer as determined from fetuses in hunter-killed samples from Hyogo Prefecture, Honshu Island in southwestern Japan, and from similar samples from Kumamoto Prefecture in Kushu Island, the southwestern-most main island in Japan. No fawns were pregnant in either sample. In Hyogo 76.6% of yearling and 89.4% of older females were pregnant, as compared to 84% and 92% for Kumamoto. The pregnancy rate did not decline with age of female in either population. Fetal development is described. Reproductive parameters were similar to those of populations in other parts of Japan. However, timing varied from northeast to southwest, with the latest birth times and shortest duration of fawning occurring in Hokkaido. The timing of fawning was correlated with a warmth index used to describe vegetation growth, with fawning being later in the north and at higher elevations.

Introduction Deer of various species adapt to a wide range of climates from the cold polar regions to the tropics (Whitehead 1993). Most are seasonal breeders, with breeding occurring during only a limited period of the year. Seasonal breeding can be interpreted, from the viewpoint of reproductive fitness, as a mechanism to ensure that births occur in the optimal season when environmental conditions favor the survival of the offspring. Findings from red deer (Cervus elaphus) on the Isle of Rhum indicate that calves born out of regular season have a lower probability of survival. Since late-born calves are insufficiently nursed, they have low body weights through winter (Clutton-Brock et al. 1982). Similarly, early-born calves are significantly lighter and have a lower probability of surviving over the summer. If the same facts can be observed in other species or populations, timing of calving would vary according to the regional differences in suitable periods for producing young. In fact, in Newfoundland, where the favorable season for calving is limited, caribou (Rangifer tarandus) breed between 9 and 19 October and 90% of the calves are born in about a 12-day period, usually between 24 May and 5 June (Bergerud 1975). D. R. McCullough et al. (eds.), Sika Deer: Biology and Management of Native and Introduced Populations, DOI: 10.1007/978-4-431-09429-6_24, © Springer 2009

327

328

T. Koizumi et al.

Similarly, births were highly synchronous with 80% of 64 births occurring in seven days from 4 to 10 June in the Kaminuriak caribou population (Dauphine and McClure 1974). In contrast, tropical species show a different seasonal pattern. In the free-ranging axis deer (Axis axis) herd kept in southern England, calving is aseasonal with a similar number of births occurring in the winter and summer months, although mortality is highly seasonal and calves born in the winter seldom survive (Loudon and Curlewis 1988). The feral Reeves’ muntjac (Muntiacus reevesi) introduced into central southern England show births that are distributed throughout the 12 months (Chapman et al. 1984). The white-tailed deer (Odocoileus virginianus) has the widest distribution range extending from about 50° N in North America to 15° S in South America. There is a latitudinal effect on the breeding seasons in the area above 30° N (Sadlier 1987), while the peak of the season is in spring and early summer in the Caribbean region (Bronson 1989). In the southernmost range (Colombia), fawning occurs through a seven-month period from September to March (Blouch 1987). Sika deer originally ranged from Vietnam to southern Ussuri in Far East Russia and have been introduced in many European countries, the United States, Australia, and New Zealand. In Japan, they occur on all major islands (Hokkaido, Honshu, Shikoku, and Kyushu) and the surrounding smaller islands including Kerama Islands of the Ryukyus, covering from subtropical to subarctic zones. Therefore, regional difference in the timing of calving seasons can be observed over their wide range. This chapter describes the reproductive characteristics of female sika deer in western Japan, with an emphasis on the regional difference in calving season. Materials were collected from two populations, in Hyogo and Kumamoto Prefectures. Both populations are open populations with medium densities, but are increasing even under exploitation. Since calving season is also influenced by population condition, such as nutritional state, density, and age structure, results should be compared among populations under similar conditions. We selected two well-documented populations on Boso Peninsula (35° N) (Asada and Ochiai 1996) and in Hokkaido (43° N) (Suzuki and Ohtaishi 1993; Suzuki et al. 1996).

Hyogo Prefecture Population Hyogo Prefecture is located in the southwest of Honshu Island, the largest of the main islands of Japan. During our study from 1988 to 1992, deer range covered about two-thirds of the prefecture with a large continuous range in the north. Deer damage to forestry and agriculture has occurred since the mid-1970s and the loss reached 722 million yen in 1989 (Hyogo Prefecture, unpublished data). Despite a decrease in the number of hunters, deer harvest also has been increasing since the mid-1970s, suggesting that the population might have increased. To cope with an increase in both population and amount of damage, the prefectural government has conducted intensive control since 1985 in the eight northern municipalities where deer were most abundant at that time.

24 Reproduction of Female Sika Deer in Western Japan

329

Fig. 24.1 Location of the study area in Hyogo Prefecture.

The investigation was conducted in these eight municipalities from 1988 to 1992. The area includes 825 km2, located around 35°20′ N and 134°40′ E (Fig. 24.1). Since the Chugoku Mountains lie in the north, the terrain is generally complex and dissected by streams. The climate is temperate with a mean annual temperature of 14 °C. Temperature seldom falls to −10 °C even in mid-winter. The annual rainfall is about 2,000 mm. Snow depth is less than 30 cm except for the high mountains. Forests cover approximately 90% of the region. As the result of intensive afforestation after World War II, Cryptomeria japonica (Sugi cedar) and Chamaecyparis obtusa (Hinoki cypress) plantations are widely distributed and constitute over 60% of the forests. Deer densities, which were estimated using the block count method in 1990, ranged from 6.6 to 19.8 deer/km2 with an average of 12.8 deer/km2 (Hyogo Prefecture, unpublished data). Fetuses and reproductive tracts were collected from 152 female deer shot by hunters during intensive control conducted in February and March. Deer were hunted by groups of 10–15 hunters and the leader of each group was requested to submit a data sheet in which the site, date, sex, and pregnancy status were recorded on each kill, along with a pair of first incisor teeth. Although we could not confirm all deer killed, we visited all groups once or twice during the period to instruct the

330

T. Koizumi et al.

hunters on how to examine for the presence of a fetus. Therefore, we believed the reports on the reproductive status of females were accurate. Ages of females were determined by extent of tooth replacement and a count of the growth layers in the cementum of the collected incisors. Additionally, in May 2004 two fully grown fetuses and a neonate were collected to estimate the weight of neonates.

Pregnancy Rate Pregnancy status was reported for 780 females from1988 to 1992. Of these females, 518 were pregnant. Age-specific pregnancy rate is shown in Fig. 24.2. No fawns were pregnant. There was no significant difference in pregnancy rates among the years (Chi-square test, p = 0.95), although the annual pregnancy rates of yearlings and older females varied from 84.2% to 90.5%. The combined pregnancy rate was 86.6%; it was relatively low for yearlings (76.6%) and high for two years old and older (89.4%). Pregnancy rate of older females at 10 years old and older was high (86.6%) and showed no indication of a decline in pregnancy rate with age until very old. The oldest female was 21 years old and not pregnant. The oldest pregnant female was 17 years old.

Fetal Growth A total of 152 fetuses was collected between 1989 and 1994. A 16-year-old female had twins. Since both were males, it was unknown if they were monovular or

Fig. 24.2 Pregnancy rate by female age in Hyogo Prefecture. Numbers above the data points are sample sizes.

24 Reproduction of Female Sika Deer in Western Japan

PERCENTAGE OF FREQUENCY

30

331 FEB MAR

25

MAY

20 15 10 5 0

0

400

800

1200 1600 2000 2400 2800 3200 3600 4000 FETAL WEIGHT IN GRAMS

Fig. 24.3 Distribution of fetal weight of sika deer in Hyogo Prefecture.

biovular. Although the sex ratio of fetuses was slightly biased toward females (73 males: 76 females) excluding twins, the difference was not significant (Chi-square test, p = 0.81). Despite being collected over a two-month period, the size of each fetus varied widely, ranging from 4.59 to 1,825 g in weight. Weights of two fetuses and a neonate collected in May indicated that the birth weight in this region was nearly 4,000 g (Fig. 24.3). Although males (mean ± SE; 1,012 g ± 38) were slightly heavier than females (967 g ± 37), there was no significant difference (T-test, p = 0.80). Therefore, all data were combined. Depending on the fetal weight, some external changes in fetuses were observed. These changes are summarized in Fig. 24.4. The head, neck, and body were well-differentiated in the smallest fetus of 4.59 g. The digits had taken shape as hooves. The mouth was open. A 40 g fetus had complete eyelids. The nares were well developed and close together. The nipples had formed ventrally. Tactile hairs first appeared on the tip of the lower jaw and then above and below the eyes. When the fetal weight exceeded 800 g, all fetuses had tactile hairs on the tip of the lower jaw and muzzle and around the eyes. The white spots first appeared along the median line of the dorsum and then extended over the both side of thigh, rump, trunk, and brachia. Appearance of white spots varied among individuals. The smallest fetus with white spots was 385 g, while the spots could not be observed in a fetus of 680 g. White spots occurred in all fetuses over 800 g in weight. The appearance of eyelashes was slightly delayed. Eyelashes were present in only 35.3% of the 600–800 g class, in over 50% of the 800–1,000 g class, and reached 100% in the 1,200 g class. In the class of 800–1,000 g, 20% of the fetuses had general body hairs on their parietal region. General body hairs were observed at elbow, thigh, shin, rump, and tail with the development of fetuses. The entire body was well-haired in fetuses over 1,500 g.

332

T. Koizumi et al.

Fig. 24.4 External development of sika deer fetuses in Hyogo: (a) tactile hairs at lower jaw, (b) white spots, (c) eyelashes, (d) general body hair.

Suzuki et al. (1996) classified the development of fetus into four stages based on the external changes in fetuses observed in Hokkaido. Since the external changes in Hyogo did not completely follow the same pattern as those in Hokkaido, we slightly modified the Suzuki et al. standards as follows. Fetuses with tactile hairs and without white spots were classified as stage 1. Fetuses with white spots and no general body hairs were assigned to stage 2. Fetuses partially covered with general body hairs were ranked stage 3. Stage 4 included all fetuses entirely covered with general body hair although their eyes were not open. Frequencies of stage 1–stage 4 were 9.9%, 49.6%, 38.2%, and 2.3%, respectively (Table 24.1).

24 Reproduction of Female Sika Deer in Western Japan

333

Table 24.1 Change in fetal weight with developmental stage in Hyogo Prefecture. Stage 1 Stage 2 Stage 3 Stage 4 N

13

65

50

3

Mean weight (g) SE Minimum weight Maximum weight

376.8 64.73 4.59 680

863.6 25.86 385 1,220

1,238.7 24.84 869 1,590

1,724.7 52.98 1,645 1,825

Calving Season Calving season was described from the distribution of calving dates. The calving date of each female was estimated using a linear relationship between the cuberoot of fetal weight and gestational age (Huggett and Widdas 1951). Unfortunately, we could not find adequate data for known-age fetuses of sika deer which is required to apply this method. Therefore, we first calculated a fetal growth line for elk (Cervus elaphus canadensis) using the data from nine known-age fetuses given by Morrison et al. (1959). Correlation analysis gave the following fetal growth line: W1/3 = 0.119T – 3.80 (R2 = 0.994, p < 0.01), where W was the fetal weight in grams and T was the age in days of the fetus. This equation had an x-intercept of 32.01 days, corresponding to 12.96% of the gestation length (247 days). Then, following the method used by Mitchell and Lincoln (1973), the regression line was redrawn to pass through the positions of two points on the coordinate, the cuberoot of fetal weight at the end of the gestation period and 12.96% of the gestation period on the time axis. We used a gestation length of 231 days (Hama 1988), a neonatal weight of 3,860 g (the mean weight of fully grown fetuses and two neonates collected in May) and an x-intercept of 29.94 days (12.96% of 231 days). The fetal growth line of sika deer in Hyogo was expressed by W1/3 = 0.078T – 2.34. Rewriting the equation, age in days of each fetuses was estimated from T = 12.78W1/3 + 29.94. Finally, birth date was calculated by adding the gestational age at the collection date. The earliest calving occurred on 27 April in 1990 and the latest on 10 September in 1993, covering a period of 134 days. Over the six-year period, the mean calving date was 22 May. Over 30% of calvings occurred in mid-May, with large numbers in early and late May also (Fig. 24.5). Estimated calving distribution suggested that conception frequently occurred over 20 days between late September and early October, with a range from 9 September to 23 January. During the period of study, there was only small year-to-year fluctuation in calving dates. Difference in mean dates was six days and gave no significant difference (Kruskal-Wallis test, p = 0.90). Calving dates seemed to be related to female age although the pattern was indistinct because of small sample size. Yearlings and two-year-old females showed later calving dates, while the mean dates of calving became earlier with increasing age. In the females 10 years old and older, however, the latest calving date occurred and dates were widely scattered (Table 24.2).

334

T. Koizumi et al.

PERCENTAGE OF FREQUENCY

35 30 25 20 15 10 5 0 1 21 APR

11 21 MAY

1

11

21 JUN

1

11

21 JUL

1

11

21 AUG

1

11 SEP

CALVING DATE

Fig. 24.5 Distribution of calving dates of sika deer in Hyogo Prefecture.

Table 24.2 Difference in calving date with age of female in Hyogo Prefecture. Age of female Sample size Mean SE (days) Earliest

Latest

1 2 3 4 5 6 7 8 9 ≤10

19 June 22 June 7 June 21 June 7 June 27 June 27 May 7 June 21 May 10 Sep

9 10 13 12 13 12 7 7 4 13

28 May 25 May 17 May 22 May 12 May 24 May 15 May 16 May 9 May 4 June

4.63 5.13 2.60 4.05 2.88 5.48 2.95 5.04 4.54 11.22

6 May 4 May 4 May 4 May 29 Apr 3 May 7 May 30 Apr 1 May 5 May

Kumamoto Prefecture Population Kumamoto Prefecture is located in the middle west of Kyushu Island, the southernmost of the main islands of Japan (Fig. 24.6). Deer occur over about a quarter of the prefecture, with a large range in the south. The prefectural government permitted bucks-only hunting after World War II. The harvest had been low until early the 1980s and then greatly increased. Damage has also become extensive since the 1980s. To manage the increase in population and the amount of damage, the prefectural government revised the hunting regulations in 1994 to allow the control of females in nine southern municipalities where deer damage was most serious. Investigation was carried out in these municipalities in February and

24 Reproduction of Female Sika Deer in Western Japan

335

Fig. 24.6 Location of the study area in Kumamoto Prefecture.

March between 1995 and 1998. The study area compasses 1,338 km2, located at about 32°20′ N and 130°50′ E (Fig. 24.6). The climate is temperate, with moderately cold winters and hot humid summers. Snow seldom accumulates in most of the area and rarely exceeds 30 cm in depth even in the mountains. The topography is characterized by steep-sloped ridges and valleys. Forests account for 89% of the land. They were originally comprised of evergreen broad-leaved forests and coniferous-deciduous mixed forests. However, as the results of intensive reforestation after World War II, 70% of the forests consists of artificial plantations of Sugi cedar and Hinoki cypress. Deer densities in the region were surveyed at 69 sites in 1994 by the pellet-count method. They ranged from 0/km2–87.7/km2 with an average of 19.2/km2 (Kumamoto Prefecture, unpublished data). The method for collecting reproductive materials was similar to that in Hyogo Prefecture. Each leader of a hunter group was requested to check the presence of fetuses and to submit a pair of the first incisors when the group shot females in the nuisance control. Ages of females were determined by the extent of tooth replacement and by counting the growth layers in the cementum of the collected incisors.

336

T. Koizumi et al.

N = 112 84 46

40 37

27 21 17 17

15 12 13 15 13

7

4

3

1

1

PREGNANCY RATE

100 80 60 40 20 0

0

2

4

6

8

10

12

14

16

18

AGE IN YEARS

Fig. 24.7 Pregnancy rate by age of female in Kumamoto Prefecture.

Pregnancy Rate Pregnancies were confirmed in 485 females shot during 1995–1998, excluding 1996 when the collection was incomplete. No pregnant fawns were observed. The oldest female was 18 years old and not pregnant. The oldest pregnant females were 16 years old. Although the pregnancy rates indicated a year-to-year fluctuation ranging from 88.6% to 92.3%, there was no significant difference (Chi-square test, p = 0.66). Therefore, all data were combined. The pregnancy rates were generally high in all ages; 84% of yearlings, 93% of two-year olds, and 92% of older females were pregnant (Fig. 24.7). Although the sample size was smaller in the class 10 years old and older (N = 69), there was no indication of a decline in pregnancy with age.

Fetal Growth A total of 118 fetuses was collected in February and March between 1995 and 1996. No twinning was observed. Sex ratio (67 males: 51 females) was not significantly different from equality (Chi-square test, p = 0.14) though the ratio was slightly biased toward males. Weights of fetuses varied widely with the lightest being 62.1 g (male) and the heaviest being 1,494 g (female) (Fig. 24.8). Mean weight (mean ± SE) of females (854.6 g ± 42.1) was slightly heavier than that of males (849.9 g ± 43.7). However, since the difference was not significant (T-test, p = 0.94), all data were combined. External changes in fetuses were examined in relation to their weights. As described in the section for Hyogo Prefecture, appearance of tactile hairs, general

24 Reproduction of Female Sika Deer in Western Japan

337

25 PERCENTAGE OF FREQUENCY

FEB MAR 20

15

10

5

0

0

200

400

600

800

1000

1200

1400

1600

FETAL WEIGHT IN GRAMS

Fig. 24.8 Distribution of fetal weight of sika deer in Kumamoto Prefecture.

body hairs, and white spots were used as criteria (Fig. 24.9). Then each fetus was classified into four categories according to the degree of development. Appearance of tactile hairs is a characteristic of fetuses in the middle stage of gestation (Morrison et al. 1959). Tactile hairs could first be observed on a 367 g fetus. All fetuses exceeding 800 g had tactile hairs on the tip of the lower jaw, muzzle, and around the eyes, the same patterns as observed in Hyogo. The appearance of white spots was delayed, compared with that in Hyogo. In Kumamoto, only 14.3% of the fetuses of 400–600 g in weight had white spots, while the white spots occurred in 60% of the fetuses in the same range in Hyogo. General body hairs also appeared first in the heavier weight class. The smallest fetus with general body hairs weighed 869 g in Hyogo and 984 g in Kumamoto. Only 19.1% of the 1,000–1,200 g weight class in Kumamoto had developed general body hairs compared to 47.2% in Hyogo (Fig. 24.9). The frequency of each development stage was 8.6% at stage 1, 53.6% at stage 2, 15.2% at stage 3, and 2.7% at stage 4, suggesting that the development of fetuses progressed more slowly in Kumamoto (Table 24.3).

Calving Season Calving season was determined by the distribution of calving dates based on the fetal weight. The method for estimating calving date was similar to that used for Hyogo Prefecture. Gestation period and x-intercept of fetal growth line were assumed to be 231 and 29.94 days, respectively. Since there was no available information on the weight of neonates or full-term calves in Kumamoto Prefecture, we first calculated the regression line between the cube-root of fetal weight and the

338

T. Koizumi et al.

Fig. 24.9 External development of sika deer fetuses in Kumamoto: (a) tactile hairs at lower jaw, (b) white spots, (c) eyelashes, (d) general body hair.

Table 24.3 Change in fetal weight with developmental stage in Kumamoto Prefecture. Stage 1 Stage 2 Stage 3 Stage 4 N

32

60

17

3

Mean weight (g) SE Minimum weight Maximum weight

465.32 26.23 62.10 728

925 21.74 576 1,313

1,263.94 31.58 984 1,446

1,427.33 35.67 1,372 1,494

number of days from an arbitrary day to the date of collection. We used 1 January, as the arbitrary day. The regression line was expressed by W1/3 = 0.077T* – 1.37 (R2 = 0.32, p < 0.01), where W is the fetal weight and T* the number of days

24 Reproduction of Female Sika Deer in Western Japan

339

PERCENTAGE OF FREQUENCY

between 1 January and the date of shooting. Since this line had an x-intercept of 17.79, we adjusted the x-intercept to be 29.94 sliding the line parallel on the x-axis. Then the following equation was obtained as a fetal growth line in Kumamoto: W1/3 = 0.077T – 2.30. The equation gave an estimation of birth weight to be about 3,700 g. Rewriting the equation, gestational age of fetus (T) was estimated from T = 12.99W1/3 + 29.94. Calving date of each fetus was calculated in the same manner as used for Hyogo Prefecture deer. The estimated calving dates indicated that the earliest calving occurred on 30 April and the latest on 9 August, with a season duration of 101 days (Fig. 24.10). The mean date of the whole sample was 22 May with no significant difference between Hyogo and Kumamoto (Mann-Whitney test, p = 0.51). Sixty percent of the calvings occurred during the 20 days between mid- and late May. Backdated conceptions ranged from 12 September to 22 December with a peak around late September. There was no significant difference in mean date between 1995 and 1996 (Mann-Whitney test, p = 0.73) because the difference was only one day. Table 24.4 summarizes the differences in the timing of calving by age of females. Calving

40 30 20 10

0

21

1

11

APR

21

1

11

MAY

21

1

11

21 JUL

JUN

1

11 AUG

CALVING DATE

Fig. 24.10 Distribution of calving dates of sika deer in Kumamoto Prefecture.

Table 24.4 Difference in calving date with age of female in Kumamoto Prefecture. Age of female Sample size Mean SE (days) Earliest

Latest

1 2 3 4 5 6 7 8 ≥10

6 July 9 Aug 24 June 26 May 29 May 26 May 29 July 25 June 24 June

9 16 13 10 5 5 9 6 25

1 June 26 May 23 May 14 May 19 May 12 May 19 May 22 May 16 May

6.89 5.38 3.25 2.14 3.15 4.01 7.88 6.91 2.95

30 Apr 8 May 10 May 6 May 12 May 1 May 13 May 10 May 30 Apr

340

T. Koizumi et al.

by yearlings tended to be later than those for two-year-old and older females. The mean dates of calving became earlier with increasing age.

Discussion No pregnant calves were observed in either the Hyogo or Kumamoto samples. Histological examination of ovaries collected in Hyogo indicated no fawns had ovulated whereas all yearlings and adults that ovulated were pregnant (Suzuki et al. 1992). Therefore, puberty of females in these populations occurs at least after 12 months old. Zhuo and Wu (1981) described that the female sika deer in Sichuan, China sexually matured at 1.5 years of age and their conception rate was about 80%. Pregnancy in sika calves has been reported in some herds introduced into foreign countries. Chapman and Horwood (1968) found an eight-month-old calf shot in England had an embryo. Mullan et al. (1988) also found that corpora lutea occurred in two of five calves and one calf had an embryo. Since a pregnant sika calf was recently found in Hokkaido (Y. Matsuura personal communication 2004), it is possible that this phenomenon may occur in the western Japanese populations. However, we believe it will be unusual. Increasing and expanding populations of sika deer generally have high pregnancy rates. Davidson (1976) reported that 80% of yearlings and 85% of adults were pregnant in New Zealand. Ratcliffe (1987) also described that pregnant yearlings were common in most populations in Great Britain and at least of 80% of adults were pregnant. In contrast, low productivity related to low pregnancy rates has been reported in high density populations. Kaji et al. (1988) found that the ratio of calves/100 females was negatively correlated with population density on Nakanoshima Island (42°36′ N, 140°51′ E), declining rapidly from 65.3 at 38.3 deer/km2 to 6.8 at 52.5 deer/km2. The ratio of calves/100 females on Nozaki Island, located at 33°12′ N and 129°8′ E, ranged from four to seven with a density of 62 deer/km2–102 deer/km2 (Doi et al. 2004). Compared to these reports, both populations in Hyogo and Kumamoto have high pregnancy rates, suggesting that most of females are in good nutritional condition and the population densities have not yet attained the level of resource limitation. This may be related to the spread of young forest plantations which provide abundant forage for deer. The forests in Japan had been degraded by overcutting during World War II. After the war, the government commenced the “expansion of reforestation” program, aiming at converting natural forests to artificial forests that have high productivity. The total area of artificial forest exceeded 9 million ha in 1975 and reached 10 million ha in 1980 (Koizumi 1998). Spatial analysis of deer damage to forestry indicated that browsing damage by sika deer was correlated with the area of clear-cutting, suggesting that clear-cutting over large areas was likely to have triggered the deer irruption (Murakami and Koizumi 2003). Only one set of twins was observed in Hyogo among 152 pregnancies. In Kumamoto, none of the 118 pregnant females had twins. Suzuki and Ohtaishi (1993)

24 Reproduction of Female Sika Deer in Western Japan

341

described that twinning was exceptional in Hokkaido based on the examination of more than 100 pregnant females. Haensel (1980) also reported only one set of twins in 83 births in a captive group in Berlin. Therefore, sika deer regularly produce only one offspring at parturition. Results from Hyogo and Kumamoto Prefectures showed that calving season of sika deer in these two regions did not vary from year to year or from locale to locale. On the other hand, the timing of calving was different from that in Hokkaido (Suzuki et al. 1996). If yearlings, which tend to conceive later, are excluded, mean date of calving in Hyogo and in Kumamoto was 21–22 May, 20 days earlier than in Hokkaido. This difference was not attributed to the difference in fetal size. Fetal development indicated 30% of the fetuses collected in Hyogo and in Kumamoto attained stages 3 and 4 in February and March, whereas there were no March fetuses classified into these stages in Hokkaido (M. Suzuki personal communication 2004). This shows that fetuses in western Japan had progressed to more advanced stages in early spring than those in Hokkaido. Late calving in Hokkaido was not due to population condition. Pregnancy rate of yearlings and older females in Hokkaido, calculated from Suzuki and Ohtaishi (1993), was approximately 90%, similar to that in Hyogo and Kumamoto. This was obviously a characteristic of a high performance population. Therefore, it is reasonable to consider the calving season in western Japan to have commenced earlier than in northern Japan. Asada and Ochiai (1996) reported calving seasons for other high-performance populations of sika deer on the Boso Peninsula (35° N, 140° E). Calving seasons estimated from their description ranged from 26 April to 2 August, with a median date of 15–16 May, about five days earlier than in western Japan. To compare the regional difference in calving season, we used the Warmth Index, a cumulative temperature of monthly mean temperatures which exceeded 5 °C. This has been widely used to assess the phenological differences in vegetation of Japan. Mean temperature at each location was based on the data from the nearest weather station, corrected by applying the rule which equated a change of 100 m in elevation with a temperature of 0.6 °C. Elevations were obtained from the Digital National Land Information provided by Ministry of Land, Infrastructure, and Transport (http://www.nla.go.jp/ksj/). Warmth Indices indicated that the vegetative growth in three southerly regions commenced in March and lasted until December, while the period was limited during six months from May to October in Hokkaido. The timing of calving season was closely related to the Warmth Index, tending to be later in the populations living in the northerly or high-elevated regions (Fig. 24.11). Distinct regional differences in calving season have also been reported for mountain sheep (Bunnell 1982) and reindeer (Leader-Williams 1988). Bunnell (1982) found northern populations of mountain sheep gave birth later and over a shorter period than populations at more southerly latitudes. Leader-Williams (1988) also reported the timing of the reindeer calving season is delayed at northerly latitudes. These findings revealed that females produced their offspring in the period in which their energy requirements were satisfied. Over the temperate and cold zones where the environments change more regularly,

342

T. Koizumi et al. 120

Boso Hyogo Kumamoto

WARMTH INDEX

100

80

60 Hokkaido 40

20 1 MAY

1JUN

1JUL CALVING DATE

1AUG

1SEP

Fig. 24.11 Comparison of calving periods in sika deer. Solid lines represent calving periods; circles represent the mean or median date. (Data for Hokkaido and Boso are cited from Suzuki et al. (1996) and Asada and Ochiai (1996), respectively.)

predictability of seasonal fluctuation in forage production will cause the regional difference in calving season. If information about calving seasons is obtained from other high performance populations, more definite tendencies may be revealed. Acknowledgements We are grateful to the members of the hunting associations in Hyogo and Kumamoto Prefectures for providing the valuable materials from sika hunts. We thank the staff of the prefectural government of Hyogo and Kumamoto for assistance during the study and for permission to use unpublished information. We also thank Drs. Masatsugu Suzuki and Dale R. McCullough for their comments on a draft of this chapter. The study was funded by grants from the Forestry and Forest Products Research Institute and from the Ministry of Agriculture, Forestry and Fishery.

Literature Cited Asada, M., and K. Ochiai. 1996. Conception dates of sika deer on the Boso Peninsula, central Japan. Mammal Study 21:153–159. Bergerud, A. T. 1975. The reproductive season of Newfoundland caribou. Canadian Journal of Zoology 53:1213–1221. Blouch, R. A. 1987. Reproductive seasonality of the white-tailed deer on the Colombian llanos. Pages 339–343 in C. E. Wemmer, editor, Biology and management of the Cervidae. Smithsonian Institution Press, Washington, DC, USA. Bronson, F. H. 1989. Mammalian reproductive biology. University of Chicago Press, Chicago, Illinois, USA.

24 Reproduction of Female Sika Deer in Western Japan

343

Bunnell, F. L. 1982. The lambing period of mountain sheep: Synthesis, hypothesis, and tests. Canadian Journal of Zoology 60:1–14. Chapman, D. I., and M. T. Horwood. 1968. Pregnancy in a sika deer calf, Cervus nippon. Journal of Zoology 155:227–228. Chapman, D. I., N. G. Chapman, and O. Dansie. 1984. The periods of conception and parturition in feral Reeves’ muntjac (Muntiacus reevesi) in southern England based upon age of juvenile animals. Journal of Zoology, London 204:575–578. Clutton-Brock, T. H., F. E. Guiness, and S. D. Albon. 1982. Red deer: Behavior and ecology of two sexes. University of Chicago Press, Chicago, Illinois, USA. Dauphine, T. C., and R. L. McClure. 1974. Synchronous mating in Canadian barren-ground caribou. Journal of Wildlife Management 38:54–66. Davidson, M. M. 1976. Season of parturition and fawning percentages of sika deer in New Zealand. New Zealand Journal of Forestry Science 5:355–357. Doi, T., A. Endo, H. Kawahara, and M. Baba. 2004. Annual report on assessment of a storage dam in Ojika district in 2003. Kyushu University Graduate School of Science Department of Biology, Fukuoka, Japan. (In Japanese.) Haensel, J. 1980. Zur biologie der Vietnam-sikas (Cervus nippon pseudaxis Eydoux and Souleyet, 1838) Untersuchungen an der Zuchtgruppe im Tierpark Berlin 5:69–99. (Cited from Sadlier 1987.) Huggett, A. St. G., and W. F. Widdas. 1951. The relationship between mammalian foetal weight and conception age. Journal of Physiology, London 114:306–317. Hama, N. 1988. Pregnancy diagnosis by ultrasonic scanning and measurement of serum progesterone level in Ezo sika does (Cervus nippon yesoensis Heude). Graduation thesis, Hokkaido University, Sapporo, Japan. (In Japanese with English summary.) Kaji, K., T. Koizumi, and N. Ohtaishi. 1988. Effects of resource limitation on the physical and reproductive condition of sika deer on Nakanoshima Island, Hokkaido. Acta Theriologica 33:187–208. Koizumi, T. 1998.Transition of forestry and wildlife damage in Japan. Pages 9–18 in B.-Y. Lee, S.-G. Lee, and B.-H. Yoo, editors, Forest protection in northeast Asia. Forestry Research Institute, Seoul, Korea. Leader-Williams, N. 1988. Reindeer on South Georgia. Cambridge University Press, Cambridge, United Kingdom. Loudon, A. S. I., and J. D. Curlewis. 1988. Cycles of antler and testicular growth in an aseasonal tropical deer (Axis axis). Journal of Reproduction and Fertility 83:729–738. Mitchell, B., and G. A. Lincoln. 1973. Conception dates in relation to age and condition in two populations of red deer in Scotland. Journal of Zoology 171:141–152. Morrison, J. A., C. E. Trainer, and P. L. Wright. 1959. Breeding season in elk as determined from known-age embryos. Journal of Wildlife Management 23:27–34. Mullan, J. M., G. A. Feldhamer, and D. Morton. 1988. Reproductive characteristics of female sika deer in Maryland and Virginia. Journal of Mammalogy 69:388–389. Murakami, T., and T. Koizumi. 2003. Broad spatial scale study about the relationship between the appearance trend of clear cutting area and the distribution of new planting area damaged by deer browsing. Kyushu Journal of Forest Research 56:101–104. (In Japanese with English summary.) Ratcliffe, P. R. 1987. Distribution and current status of sika deer, Cervus nippon, in Great Britain. Mammal Review 17:39–58. Sadlier, R. M. F. S. 1987. Reproduction of female cervids. Pages 123–144 in C. E. Wemmer, editor, Biology and management of the Cervidae. Smithsonian Institution Press, Washington, DC, USA. Suzuki, M., and N. Ohtaishi. 1993. Reproduction of female sika deer (Cervus nippon yesoensis Huede, 1884) in Ashoro District, Hokkaido. Journal of Veterinary Medical Science 55:833–836. Suzuki, M., T. Koizumi, and M. Kobayashi. 1992 Reproductive characteristics and occurrence of accessory corpora lutea in sika deer Cervus nippon centralis in Hyogo Prefecture, Japan. Journal of the Mammalogical Society of Japan 17:11–18.

344

T. Koizumi et al.

Suzuki, M., K. Kaji, M. Yamanaka, and N. Ohtaishi. 1996. Gestational age determination, variation of conception date, and external fetal development of sika deer (Cervus nippon yesoensis Heude, 1884) in eastern Hokkaido. Journal of Veterinary Medical Science 58:505–509. Whitehead, K. G. 1993. Encyclopedia of deer. Swan Hill Press, Shrewsbury, United Kingdom. Zhuo, S. L., and S. Z. Wu. 1981. Observations on the sika deer’s reproductive physiology and breeding capability in Sichuan. Zoology Magazine August:18–21.

Chapter 25

Sika Deer in Nara Park: Unique Human-Wildlife Relations Harumi Torii and Shirow Tatsuzawa

Abstract Sika deer have had a long history of cultural importance in Nara Park, beginning in the eighth century with a legend that a god rode into the park on the back of a white deer. With protection for religious reasons, the population built up and became tame because of its frequent interaction with people visiting the religious shrines at the park. The interface of sika and humans at such close proximity over the years inevitably led to harmony or conflict depending on the goals and motivations of people. Nara Park in its modern form was established in 1880, and these conflicting values of sika deer in the park have continued into modern times. However, the long history of known numbers, and approachable tame deer, have yielded an unusually long and detailed record of population dynamics, ecology, and behavior. It has also led to high populations of deer with consequent impacts on their habitat. In this chapter we review and summarize this unique record of cultural and biological interrelations between sika deer and humans.

Introduction Nara Park is located adjacent to the urban area of Nara city, which is one of the most beautiful old cities in Japan and contains many historical places and much cultural heritage that attract two million tourists from abroad every year. Nara Park covers approximately 6.6 km2, including flat areas where tourists visit temples and other attractions and the adjacent mountains (Fig. 25.1), such as Mt. Kasuga that, thanks to their long history of protection, contain unique ecosystems. Sika deer have lived in Nara Park through the ages and have shaped the specific ecosystem and scenery of the park. Sika deer are symbols of Nara Park. They are widely accepted as essential to the park, and in 1957, they were designated a national natural treasure, the “Deer of Nara.” The deer, as main subjects in many historical artworks, have long been tamed to human presence, but basically they are free-ranging. They continue to be a major source of attraction to tourists, and this has not motivated people to control the population of deer. In Japan where forests are very dense, sika deer are difficult to observe for more than a fleeting moment. This has made the tame deer at Nara good subjects for long-term population censuses and behavioral studies by direct observation at close range. D. R. McCullough et al. (eds.), Sika Deer: Biology and Management of Native and Introduced Populations, DOI: 10.1007/978-4-431-09429-6_25, © Springer 2009

347

348

H. Torii, S. Tatsuzawa

Fig. 25.1 Map of Nara Park. Tourists encounter several hundred deer in the daytime in the area surrounded by the black line. Adapted from the CD map, 1/25,000 [Nara] published by the Geographical Survey Institute.

On the other hand, the sika deer at Nara Park have been in close relationship with the local people both in a positive and negative way. Today the population of deer in Nara Park has increased to around 1,200, and they have caused increases in traffic accidents and damage to crops. Discussions about the management of these animals are needed to face the problems of not only damage to agriculture and forestry, but also the serious effects on the habitat of the area. In this chapter, we will review the history and current circumstances of Nara sika deer.

History of Deer and People in Nara Park Ancient Times to the Edo Era The following history is taken from Fujita (1997). The first record of the idea of “Shin-Roku” (deer as messengers of God) is found in the year 1006. After that, several old documents from the eleventh and twelth centuries show deer in Kasuga area were not hunted, for the place was a holy land, and some of them tell a legend which said a god from the Kashima shrine entered Mt. Kasuga riding on a white deer in the year 768 (Fig. 25.2).

25 Sika Deer in Nara Park: Unique Human-Wildlife Relations

Fig. 25.2 Painting showing the legend that a god came to Mt. Kasuga riding a white deer from Kashima Shrine (provided by Nara National Museum).

349

350

H. Torii, S. Tatsuzawa

After the thirteenth century, Nara deer were protected by Kofuku-ji Temple, which controlled the whole Yamato district, including Nara, from any hunting or capturing. At this time killing of deer was consider the same as killing of priests. This helped the population of deer to increase and, at the same time, forced people to contain their frustrations about damage to their crops and other deer-human conflicts. In 1602, Tokugawa Ieyasu, the first shogun of the Edo Era, clearly designated Nara deer as “Shin-Roku” and prescribed punishments for persons killing them. This ordinance assured deer of protection with the government’s approval. However, in 1671, antler-cutting (ceremonial sawing off the antlers of males) was started to prevent accidents with people and destruction of night lanterns by antler trashing. This indicates that times were starting to change, and governments could no longer ignore public outcries. Because one or two hundred male deer were antler-cut annually for 46 years from 1672, the deer populations of those times can be estimated roughly from expected sex ratios to range from a few hundred to one thousand. From the above history it is apparent that deer were deified from ancient to modern times and were used politically by authorities. As Fujita (1997) notes, this unique relationship between deer and people did not derive from beliefs of local people, nor adoration of deer in the real sense of the term; nevertheless, it is clear that protection for religious reasons was the main cause of the increase in the population of deer, with resultant damage to crops and other human-deer conflicts.

From the Meiji Era until before World War II Nara Park was established in 1880, and this spelled the end of the “Shin-Roku” myth. With the restoration period (the Meiji Era starting from 1868), the general trend of thought in Japan was modernization, and little further credence was given to traditional religion which was considered only superstition. Deer were regarded as just one more species of wild animal, and not sacred. Adding to this, guns became available in large quantities to the public, so wild animals faced hard times. Nara deer were not an exception and they lost their special religious status, denoted by a record that government officials ate deer meat in a stew. In 1873, the local government confined 700 deer in an enclosure to prevent damage to crops. However, in only a few years the population in the enclosure decreased to 38 due to overcrowding. However, some deer remained free in the fields, and the Kasuga area, including Kasuga Shrine and Mt. Kasuga, was designated a conservation area. This made it difficult to distinguish between wild deer and protected “Shin-Roku.” Deer crossed into areas of human habitation and caused agricultural damage. To prevent this, the Kasuga Shin-Roku Preservation Society (“Kasuga Shin-Roku Hogo Kai;” later the name was changed to the Foundation for the Protection of Deer in Nara Park) was organized by 1897. The local government paid a subsidy to

25 Sika Deer in Nara Park: Unique Human-Wildlife Relations

351

the society from 1902 to 1918. Private lands were surrounded by deer-proof fences and ditches, and the society paid compensation for crop damages. Moreover, the society started to herd and protect deer at night to decrease crop damage and prevent poaching and wild dog attacks. However, damage kept on increasing, resulting in the ongoing capture of deer in crop fields, to be held permanently in captivity. This method is still used today.

After World War II until the Present As World War II intensified, conservation became more difficult, and it was neglected as labor and funds were redirected to the war effort. Consequently, the deer population decreased rapidly. By the end of the war in 1945 the population was as low as 79 individuals. In the years following the war the population recovered under rigorously enforced protection. In 1957 “Deer of Nara” were recognized as a national natural treasure as they “blend in with the appealing scenery of the park and provide the most beautiful natural scenery of Japan with wildlife.” In addition, the conservation area was expanded to the whole of Nara city. The law for protection of cultural properties designates an administrator, but the sika deer was left without a designated administrator for many years. In the meantime, the Foundation for the Protection of Deer in Nara Park (the Nara Deer Fund) had been working for their conservation. Nara city government had been making partial payment for the crop damage, but in 1979, the farmers brought a lawsuit to get complete compensation for damage. This action resulted in an arbitrated settlement which designated a smaller protected area and allowed the control of deer outside the protected area. Nevertheless, in 30 years after the arbitration nothing changed in practice; local authorities continued only the capture of deer in crop fields. No population control was done, and the level of crop damage did not decrease. This is probably because people have the deep-seated idea that deer are a protected species, and the local authorities fear damaging the tourist’s image of deer at Nara, which might impact negatively on visitation. Major management actions include keeping pregnant females in captivity to prevent accidents at the time of birth and cutting males’ antlers at the end of summer. Males with larger antlers are chosen and isolated until the beginning of October to cut their antlers at an event for tourists. Deer captured in crop fields are kept in permanent captivity. Moreover, injured or sick individuals are confined until they are recovered and released, or die. Studies on behavior and ecology of Nara deer began quite early in the history of Japanese mammalogy. The most impressive was the study of social behavior done by Kawamura (1950, 1957) right after World War II. Kawamura used an individual-identification method and did behavioral experiments in Nara Park

352

H. Torii, S. Tatsuzawa

from 1948 to 1950. He demonstrated the basic social structure and habitat use patterns of the deer. In addition, annual population censuses of deer numbers were conducted by the Nara Deer Fund. This is one of the rare instances in which the same census method was applied over a long time period, thus yielding a long record of numbers. Further ecological research of the deer was done in Nara Park in the 1970s, such as socio-biological studies (e.g., Miura 1976; Takaragawa and Kawamichi 1977), the study of age determination methods and population structure of deer (Ohtaishi 1978), and analysis of food habits and effect of deer browsing on the vegetation (Takatsuki 1980). These pioneering research projects established the dynamics of the deer population and the relationship between deer and vegetation, not only in the flat areas of Nara Park, but also the adjacent mountainous area including Mt. Kasuga (Fig. 25.1).

Behavior of Nara Deer and Their Social Structure Grouping, Social Rank, and Leaders According to Kawamura (1950, 1957) at the time he began research in 1948, about 140 deer were using the flat grass land (32 males above the age of one year and 110 females and fawns), and females and fawns formed 12 groups based on blood relationships. Each of these groups had its own well-organized home range. A few groups gathered at daytime and fed together in the fields of Tobihino and Asajigahara, but at night each group rested in its own resting sites (Kawamura 1950). Some individuals were seen to move away from their home range in mating seasons; compared to the loose-bond male groups, however, female groups were quite stable (Kawamura 1950, 1957). Kawamura (1957) studied experimentally whether the female groups were structured under a ranking system and whether there was any individual group leader. He threw a piece of sweet potato in the middle of two targeted individuals and observed their agonistic behaviors to determine their rank order. To determine the leader of the group, he disguised himself as a horse and approached the group to observe which individuals gave the alarm call (“pya!”) first. As a result, he showed that: (1) a specific individual always gives alarm calls first, and (2) the same specific individual usually leads the group when they move. He concluded, therefore, that female groups have a specific leader. On the other hand, males seemed to join groups of females readily, and males above the age one year possessed home ranges that overlapped those of females. Their interests towards female groups were subtle, and except for mating seasons, they usually formed groups consisting of only males. However, in the mating season, males’ home ranges changed and boundaries between mountains and the flat lands disappeared. The number, location, and configuration of male territories where

25 Sika Deer in Nara Park: Unique Human-Wildlife Relations

353

mating took place did not change during the five years of Kawamura’s study although their occupiers switched once in a while.

Territory and Home Range Following Kawamura’s (1950, 1957) pioneering work, Miura (1983) continued social structure studies. After two years of direct observation with individual identification, he pointed out four types of male social structure and noted that the tendency for males to gather with each other was the highest in spring and summer and the lowest in autumn during the mating season. The 82 known individual males were divided into territorial or non-territorial individuals. The former were all over five years old. The home ranges of territorial males were smaller and non-overlapping compared to those areas occupied by non-territorial males. Territories were divided into two areas: the core area where males showed exclusive behaviors and the surrounding area (Miura 1984). Miura (1984) also found that territorial males showed 57.4% of sexual behavior, and 76.2% of mating behavior. Thus, territorial males had a better chance of mating than non-territorial males and territoriality contributed to higher mating success. Moreover, he showed that: (1) the location and configuration of territories were stable; (2) the dominant individual of the year occupied a given territory; (3) dominance and ranking was positively correlated with the size of antlers; and (4) there was a positive correlation between the length of antlers and body weight. Miura (1984) also analyzed the male vocal repertory and categorized vocalizations into six types. Minami and Kawamichi (1992) subsequently reported 13 vocalization in five groups of Japanese sika deer. Minami (1993) noted the similarity of deer calls with the sounds of traditional Japanese deer calling instruments, which mimic the natural calls.

Other Social Behaviors Because close direct observations are possible, other notable behavioral studies have been done in Nara Park that would have been difficult elsewhere. Inoue and Kawamichi (1976) studied the development of social behavior by observing fawns from birth to five months of age. They noted that (1) suckling lasts six months; (2) fawns join the female group during the suckling period; and (3) contacts with human during this time affect later habituation to humans. Other relationships between individuals have been studied, such as (1) resting time of females becomes relatively shorter during mating seasons due to many disturbances by males (Yabu and Wada 1996); (2) allogrooming behavior is more likely to be initiated by females (Matsuno and Urabe 1999); and (3) the frequency of behavior between both sexes is not affected by their mating relationships or the rank of the male (Yamada and Urabe 1998; Matsuno and Urabe 1999).

354

H. Torii, S. Tatsuzawa

Population Trend Dynamics of Nara Deer Population Trends In the flat plain of Nara Park, a population census is taken every summer by the Nara Deer Fund (Fig. 25.3). These counts, combined with historic estimates, indicate that a population recorded at above 700 individuals in Meiji Era dropped to 79 at the end of World War II, and with conservation efforts recovered up to 1,000 by 1965. After some fluctuations, the population became stable around 1,200 (Nara Deer Fund 2005) (Fig. 25.3). This gives an incredibly high density calculated at 1,000 individuals per km2 in the 120 ha plain. Tatsuzawa et al. (2002), who studied daily activity and its seasonal variation for 10 years (1990 through 1999), showed that the population was steady at high density: mean = 961.1 per km2 (SD = 60.0) in summer and 907.7 (SD = 62.8) in autumn. In addition, the summer population density showed a strong negative correlation with the rate of population increase in the following year. In other words, a negative density-dependent relationship was confirmed. Yearly fluctuation of the number of fawns (birthrate, early mortality, or dispersal rate) seemed to contribute most to the density effect (Tatsuzawa et al. 2002). Nara Park deer mainly feed on short-grass vegetation (mainly Zoysia japonica). Calculated from the productivity and digestibility of Z. japonica, the deer capacity of Nara Park grassland is 12–14 per ha in the growing season of this plant (April to September). Thus, a population of 1,100–1,200 per km2 (Miyazaki 1980; Miyazaki et al. 1984) suggests that the Nara deer are at or near their maximum number based on food resources. Food habits will be covered in greater detail in a later section. From at least the end of the 1940s (Kawamura 1950) to the end of the 1980s (Tatsuzawa and Fujita 2001) hundreds of Nara deer continued daily dynamic movements back and forth across the flat from the base of Mt. Kasuga to around Koufuku-ji Temple (Fig. 25.1). However, developments in the flat (for instance,

Fig. 25.3 Population trend of Nara deer based on yearly censuses by the Nara Deer Fund.

25 Sika Deer in Nara Park: Unique Human-Wildlife Relations

355

installation of railings and fences, shrinkage of the woodland area) terminated this movement. Although about 20–50% of the population traversed the flat even in the 1990s, the percentage has decreased every year (Tatsuzawa et al. 2002). Infrastructure of the park has become a physical barrier and caused cautious females to stay in the woodland, speeding the increase of the population in the woodland area.

Early Mortality and Life Expectancy In Nara Park pregnant females are captured from the end of March to April and kept in captivity until the middle of July after they deliver fawns. The estimated mortality of young animals during this period can be derived from the records kept by the Nara Deer Fund. The average mortality of young of the year over a five-year period was 40.9% (±3.9%), and for one-year-olds 22.4% (±4.1%). Because the bodies of dead fawns disappear soon after death it is possible that fawn mortality was higher than estimated. An analysis by Ohtaishi (1976) of individuals buried after death led to an estimate of 50% fawn mortality. We constructed a the life table of Nara Deer by age determination of individuals found dead (unpublished), and it showed that the mean life expectancy at birth was 4.0 years for females and 3.8 years for males. Deer that lived to one year of age had a longer life expectancy for both sexes, again reflecting a high mortality in early life. Both mean life expectancy and the greatest length of life were higher in females; thus, females had greater longevity than males.

Age-Specific Pregnancy Rate We calculated the pregnancy rate of 235 females that died between February and May (unpublished). There was only one pregnant individual out of 22 one-year-olds, giving a pregnancy rate of 4.2%. The pregnancy rates of two- and three-year-olds were 52.6% and 86.7% respectively. The highest rate was achieved by three-year-olds. The average rate for females over the age of three was 73.7%, and rates decreased rapidly after age 12; however, one pregnancy was noted even in a 23-year-old female. However, pregnancy rates were calculated from the recovered bodies of individuals that died from accidents occurring during capture of pregnant individuals in spring, so they may be overestimated. Because pregnant females are targeted by this capture, the actual pregnancy rate across all females is probably around 50%. From the censuses of the Nara Deer Fund over the last several decades there was a total of approximately 1,200 individuals; 300 were males, 700 were females, and 200 were fawns (Fig. 25.3). As stated earlier, the survival rate at the end of age one is about 40% and 20% at age two. This indicates that the population of females of age one and two is 140 in total, and, therefore, there are about 560 fawn-bearing females. Every year, around 200 females in captivity and 50 females in the wild are

356

H. Torii, S. Tatsuzawa

estimated to give birth (Nara Deer Fund, personal communication). If these estimates are correct, it means that more than 250 individual females give birth every year; thus almost half of females above the age of three years give birth, which is not so different from the 50% pregnancy rate estimated above for the total female population (both in captivity and free-roaming).

Feeding Habits and Nutritional Status Food Habits In the 1970s, Takatsuki and Asahi (1977) examined deer droppings at Nara Park using microhistological techniques and found that deer feed mainly on monocotyledon plants, such as graminoids, indicating they were grazers. Zoysia japonica and Miscanthus sinensis (silver grass) covered the whole area of Nara Park at that time. After 30 years of extremely high population density, the vegetation changed. Miscanthus sinensis decreased in Mt. Wakakusa, whereas tourists and the local people who artificially fed deer increased, therefore changing the feeding environment of deer. Visitors and tourists feed deer routinely at Nara Park (Fig. 25.4). Still, from the rumen content analysis by Torii et al. (2000) deer were still using the same food plant categories as reported by Takatsuki and Asahi (1977).

Fig. 25.4 Tourists at Nara Park feeding deer rice crackers.

25 Sika Deer in Nara Park: Unique Human-Wildlife Relations

357

From examining the rumen contents of dead deer we also found that Nara deer feed mainly on graminoids (mostly Z. japonica) and broad-leaved trees. They also fed on herbaceous plants, coniferous leaves, seeds, and nuts. We also found Sasa spp. bamboo leaves in the rumen contents, although there is little Sasa left around Nara Park due to the continuous feeding by deer on these highly preferred species. Bamboo leaves in the rumen probably come from deer eating fresh bamboo fronds set out as decorations during religious events at temples or shrines. Because Nara Park deer feed mainly on short-grass vegetation they face food shortage in winter when the short-grass withers. Thus, for example, the percent of crude protein in rumen contents increased in spring, stabilized in summer and autumn, and then decreased in winter. In winter deer mainly feed on vegetables provided by humans. Nara deer are not completely dependant on food provided by humans; but such foods partially compensate for food shortages which are especially severe in winter. Clearly, artificial feeding, and also feeding on farmers’ crop fields, help support the continuous high density of deer at Nara.

Nutritional Status We collected the femur bones from dead deer found around Nara Park and evaluated the color and material of the bone marrow in accordance with the categories of Takatsuki (2001). In this analysis, we were able to separate the marrow into five types, including a yellow gelatinous texture that was not observed by Takatsuki (2001). The content of fat in this yellow gelatinous texture averaged 6.3% (SD = 2.9); this was similar to deer that starve to death in spring in northeastern Japan and Hokkaido. This indicates that Nara deer are in extremely in poor condition, and many of them would not be able to survive in a truly wild situation. Certainly survival of less well-nourished individuals is largely due to the active involvement of the Nara Deer Fund and supplemental feeding by visitors. Also, the more benign conditions for Nara deer include being spared stresses such as predation and migrations. These findings all indicate that Nara deer are a high density, low performance population. Reproduction usually starts at the age three in Nara Park, which is one year later than in most other areas. The age-specific pregnancy rate is lower and individuals are in poorer nutritional condition than in comparable populations. Poor physical condition, which is specific to the end of winter in deer in Hokkaido and western Japan, is seen all year round in Nara Park deer.

Alteration of Park Vegetation by the High-Density Population It has been 1,300 years since deer protection started in Nara and the deer population has been very dense for at least a few hundred years. This has had a great effect on the vegetation of Nara Park and surrounding areas. The most obvious feature is a

358

H. Torii, S. Tatsuzawa

pronounced browse line. There are very few branches of favorite tree species below the height of 2 m, which is the upper limit of deer reach. Podocarpus nagi and Pieris japonica (andromeda), plant species that deer do not eat, cover the land around the Kasuga grand shrine. The well-known short grass vegetation covers the ground of many areas and, of course, is maintained as a lawn under the high grazing pressure of deer. Takatsuki (1980) classified the vegetation of Nara Park into three types. The first type is vegetation unpalatable to deer. It occurs on the flat, but unpalatable vegetation grows abundantly on the uplands of Mt. Kasuga and Mt. Wakakusa (Fig. 25.1) as well. Unpalatable vegetation is typified by Podocarpus nagi and Sapium sebiferum (Chinese tallow), both alien invasive species. Podocarpus nagi was first brought into the Kasuga grand shrine in the ninth century as a sacred tree. Notably, there now are pure P. nagi forests in Tobihino. Sapium sebiferum was planted in the park because deer do not feed on it. S. sebiferum grows also in Tobihino and Mt. Wakakusa. Pieris japonica grows as an understory plant of P. nagi and S. sebiferum stands. In addition a fern, Hypolepis punctata, that deer dislike, grows on both sides of the approach to Kasuga grand shrine. The second type is short-grass vegetation established by deer browsing pressure. The Z. japonica grass community is a typical short-grass vegetation type. Many deer forage day and night on these grass communities, virtual lawns, and many tourists gather at such landscapes, which are the outstanding scenic areas of the park. Zoysia japonica is the main food item of deer as noted earlier. Poa annua (bluegrass) and Hydrocotyle maritima (a wetland plant) can also be found in the area, and occasionally Trifolium repens (white clover), Sisyrinchium angustifolium (blue-eyed grass), and other species. Of concern is that in some areas Sapium sebiferum has started to invade. It is possible that the Z. japonica community vegetation will be lost in the future if this invasion continues. The third type is Miscanthus sinensis growing where deer browsing pressure is low in the upland areas. This community is also shrinking rapidly. Traditionally, burning was carried out to favor this community, but brackens establish more rapidly after fire, so this practice may have to be discontinued. Currently, fences are being erected to preserve the remaining areas of M. sinensis on Mt. Wakakusa. Miyazaki (1980) estimated the appropriate population in the flat of Nara Park to be 1,100–1,200 individuals based on the short-grass production. This was based on the height of the short-grass, which at the time was higher than 4 cm. However, present production is lower because the height is less than 4 cm. This estimation is for the population of spring to autumn, so the forage biomass must be much lower in winter. Moreover, just after World War II the height of the plants in the flat was high enough to hide a rugby ball (Kawamura 1957). This suggests that the vegetation and biodiversity of the grassland, even in the flat area, was probably much richer then, and it has changed for the worse over time. From these results it is clear that long-term deer browsing has established a unique vegetation resistance to deer feeding. It has evolved to be unpalatable, so it persists despite heavy foraging pressure.

25 Sika Deer in Nara Park: Unique Human-Wildlife Relations

359

Conservation of Mt. Kasuga Primeval Forest and Nara Deer The Effect of Deer Browsing on the Vegetation of Mt. Kasuga The Mt. Kasuga primeval forest lies next to the flat area in Nara Park. It is one of the few broad-leaved evergreen forests remaining in Kinki district, which is the northern limit of such forests. Hunting and logging in the area have been prohibited since the year 841 and the forest has been left almost untouched by humans. This virgin forest was designated as a “National Natural Treasure” and as a special “National Natural Treasure” in 1924 and 1956 respectively. In 1998, it was also recognized by UNESCO as a World Cultural Heritage Site along with the Kasuga grand shrine. However, it is known that the forest is affected by deer foraging. Maesako (2001a) randomly selected 22 study sites, and of 2,351 trees of 56 different species identified, bark stripping marks were found on 251 trees (10.7%) of 36 species (64.3%). In 1986, the Environmental Agency established 20 permanent meter-square quadrats in the same area within Mt. Kasuga forest. According to the results, in 1986 there were six species of canopy trees (Castanopsis cuspidata (chinquapin), Neolitsea aciculata (Lauraceae), etc.), 15 species of shrubby plants (C. cuspidata, Quercus sessifolia, Eurya japonica, etc.), and 12 species of forest floor vegetation (C. cuspidata, Pieris japonica, etc.). Plant composition was measured again in 2003 (Maesako 2004). The species composition of the canopy trees had not changed, while on the forest floor, six species, such as Quercus salicina, Q. glauca (blue oak) and holly, disappeared and five species, including Symplocos prunifolia and Camellia japonica appeared. Four new species of shrubs, such as C. japonica and Q. sessifolia, appeared in the intermediate, sub-canopy layer. In total, the number of species observed in the area decreased from 18 species in 1986 to 15 species in 2003. These findings may show that the effects of deer browsing began only rather recently. Four species of trees, C. cuspidata, Q. sessifolia, Symplocos prunifolia, and Neolitsea ariculata, were the only species larger than 10 cm dbh, and we compared their diameter distributions. Larger C. cuspidata grew steadily, but below 20 cm dbh the number of trees decreased. However, S. prunifolia, a species that deer do not bark, increased. Regeneration is difficult for broad-leaved evergreen forest species, and this threatens the perpetuation of the pristine natural forest. The next generation of the forest will likely be largely unpalatable species. This process is apparent in seedlings also. Shimoda et al. (1994) followed the survival rate of seedlings and amount of deer browsing inside and outside a deerexclusive fence in a new forest gap. The effects of deer browsing on the survival rates differed among plant species. The foraging rates were high and the survival rates were low for Mallotus japonicus (Euphorbiaceae), Aralia elata (angelica tree), and Zanthoxylum ailanthoides (prickly ash). Q. salicina had a marked low foraging rate and low survival rate while Neolitsea aciculata and Sapium sebiferum were low in foraging rate and high in survival rate. Moreover, Maesako (2002) counted 42 seedlings of new species not part of the pristine forest in the area.

360

H. Torii, S. Tatsuzawa

They are likely to form the canopy of the future. These results show that not only the natural death of seedlings through competition and crowding, but also deer browsing on seedlings has a great effect on the regeneration of forests. The effects of deer are most pronounced in tree-fall gaps. New gaps caused by typhoons or some other main factor are produced on average once in 6.6 years on Mt. Kasuga. It takes about 70 years for the gaps to close in and 110 years for the forest floor vegetation to grow into canopy. The average turnover rate to the next new gap is estimated to be 180 years (Naka 1994). All these findings indicate that in any forest gap created on Mt. Kasuga, the next generation of the forest will regenerate with only species unpalatable for deer. Yamakura et al. (2001) predicted the future of Mt. Kasuga using the survival rate and seedling production of Zanthoxylum ailanthoides. According to their results, the Z. ailanthoides in the area will disappear within 185 years, and most of the seedling death is due to deer browsing. Seeds buried in the soil will run out before existing individuals disappear. Furthermore, it is possible that species that do not produce buried seeds disappear earlier due to the effects of deer browsing. Nanami et al. (2002) predicted that in 250 years the forests will be formed by only Neolitsea aciculata, Podocarpus nagi, and Sapium sebiferum with a shrub understory of N. aciculata. However, recently deer began to forage on Maesa japonica, Urtica thunbergiana (nettle), and N. aciculata, all of which were thought to be unpalatable (Maesako 2001b). Deer are known to feed on other species when more favored plants are used up, so there is fear that the whole forest might disappear.

Deer Density in the Forest of Mt. Kasuga To protect the vegetation and biodiversity, we must first determine the population density of deer. In 2005, using a block count method, an estimate of deer density on Mt. Kasuga was 15.9 and 22.8 per km2 in March and November respectively. The density in autumn is not so different from that of Odaigahara in Nara where forest degradation through deer browsing is an issue. Other high density records for sika deer in Japan are 30–50 deer per km2 on Nakanoshima Island in Lake Toya, Hokkaido (Kaji 1986), 80 on Nozaki Island, Nagasaki Prefecture (Doi et al. 1985), and 35 in Nikko, Tochigi Prefecture (Koganezawa and Satake 1996). These areas have in common a marked alteration and impoverishment of vegetation. It is clear that further deer population growth must be prevented to protect the precious forests on Mt. Kasuga.

Management of the Nara Deer Population The present deer management at Nara Park creates many problems. Feeding by local people and tourists is the most important factor for deer winter survival.

25 Sika Deer in Nara Park: Unique Human-Wildlife Relations

361

Feeding makes up for the natural food shortage and, at the same time, encourages deer to remain in the flat areas waiting for the tourists to buy them deer crackers. This prevents dispersion of the deer to the surrounding areas such as Mt. Kasuga. Capturing pregnant females and keeping them in an enclosure is probably the main factor lowering the mortality of newborn fawns, thus weakening an important regulatory mechanism in wild populations. Group feeding and breeding in captivity may be causing the attenuation of mother-fawn relationships. After captive females are freed, they often suckle a few, sometimes even four or five fawns, behavior not observed elsewhere. Antler-cutting of males in rutting periods might disrupt rank and mate selection and is likely to disturb territory occupation. This may lead to less-adaptive individuals being more successful in breeding and, thus, interfere with natural selection. The effect of antler-cutting on fighting behavior is still unclear. Moving some antler-cut males to surrounding areas (e.g., Ikoma city, 10 km away) is definitely altering natural dispersal and innate behaviors. Yamakura et al. (2001) suggested putting up a fence to control deer browsing and conserve vegetation in the Mt. Kasuga primeval forest. However, Tatsuzawa and Fujita (2001) pointed out that fencing would lead to more human-deer conflicts in the flat area and more detrimental effect on the vegetation in Mt. Kasuga area outside of the primeval forest. What is needed is simultaneous management of the deer and vegetation of the total park. To recover the biodiversity of the park a scientific and adaptive management plan to control deer is needed. People of Nara have lived with deer for more than 1,000 years. Although the deer cause inconvenience in daily life as well as serious damage, people still feel it is natural that deer live within the same area. It is valuable for both deer and people that Nara Park be used as a biological study field and environmental education facility, a place where we can learn to solve people-deer problems to the benefit of both.

Literature Cited Doi, T., K. Inakazu, Y. Ono, and H. Kawahara. 1985. A preliminary study on the effects of sika deer on natural regeneration of forest. Bulletin of the Nagasaki Institute of Applied Science 26:13–18. (In Japanese.) Fujita, K. 1997. The millennium history of the relationship between deer and people in Nara. Deer My Friend, Nara, Japan. (In Japanese.) Inoue, Y., and T. Kawamichi. 1976. Behavioral development of the fawns on Nara deer Cervus nippon. Annual report of the study of deer in the Nara Park 1975:31–46. (In Japanese.) Kaji, K. 1986. Population dynamics and management of sika deer introduced into Nakanoshima Island in Lake Toya. Honyurui Kagaku (Mammalian Science) 5:25–28. (In Japanese.) Kawamura, S. 1950. Preliminary report of social life of Nara deer. Seiri Seitai (Physiology and Ecology) 4:75–87. Kawamura, S. 1957. Deer in Nara Park. Japan Mammal Series 4:1–166. (In Japanese.) Kawamura, S. 1959. Personality in the community and behavior of mammals. Pages 6–30 in K. Imanishi, editor, Animal community and individuals. Iwanami Shoten, Tokyo, Japan. (In Japanese.) Koganezawa, M., and C. Satake 1996. Effect of grazing by sika-deer on the vegetation of Oku-Nikko and their management. Annual Report of Pro-Nature Fund 5:57–66. (In Japanese.)

362

H. Torii, S. Tatsuzawa

Maesako, Y. 2001a. Fraying by sika deer (Cervus nippon TEMMINCK) and tree species preference in a warm-temperate evergreen forest on Mt. Kasugayama, Nara, Japan. Bulletin of Stories Nara Saho College 9:9–15. (In Japanese.) Maesako, Y. 2001b. Vegetational changes effected by sika deer in the Nara Park and Kasugayama. Nara Botany 23:21–25. (In Japanese.) Maesako, Y. 2002. Current-year seedlings in a warm-temperature evergreen forest Mt. Kasugayama, a World Heritage Site in Nara, Japan. Bulletin of Stories Nara Saho College 10:29–36. (In Japanese.) Maesako, Y. 2003. Community dynamics during a 17 year period in a specific plant community (a Castanopsis cuspidata forest) in Kasugayama Forest Reserve, Japan. Bulletin of Stories Nara Saho College 11:37–43. (In Japanese.) Maesako, Y., and H. Torii. 2000. Tree barking by sika deer Cervus nippon in warm temperate evergreen forest in Mt. Kasugayama, Nara Prefecture, Japan. Bulletin of Kansai Organization for Nature Conservation 22:3–11. (In Japanese.) Matsuno, K., and M. Urabe. 1999. Male-female interactions of sika deer (Cervus nippon) in Nara Park through allogrooming during breeding and rutting seasons. Journal of Ethology 17:41–49. Miura, S. 1976. Ecological studies on sika deer in Nara park with reference to spatial structure. Annual report of the study of deer in the Nara Park 1975:47–61. Miura, S. 1980. Correlation between body weight and antler length of sika deer bucks. Journal of the Mammalogical Society of Japan 8:78. Miura, S. 1983. Grouping behavior of male sika deer (Cervus nippon) in Nara Park, Japan. Journal of the Mammalogical Society of Japan 9:279–284. Miura, S. 1984. Social behavior and territoriality in male sika deer (Cervus nippon) during the rut. Zeitschrift für Tierpsychologie 64:33–73. Miura, S. 1984. Annual cycles of coat changes, antler regrowth and reproductive behavior of sika deer (Cervus nippon) in Nara Park, Japan. Journal of the Mammalogical Society of Japan 10:1–8. Miyazaki, A., S. Kasagi, and T. Mizuno. 1984. Digestibility of Zoysia-type grass by Japanese deer (Cervus nippon). Japanese Journal of Zootechnical Science 55:661–669. (In Japanese.) Miyazaki A. 1978. On the productivity of Zoysia-type grass land: Carrying capacity for deer based on the productivity of Zoysia-type grass in Nara Park (from the results of the study during 1976–1977. Annual report of the study of deer in the Nara Park 1977: 45–148. Kasuga Manifestation Society. (In Japanese.) Minami, M., and T. Kawamichi. 1992. Vocal repertoires and classification of the sika deer Cervus nippon. Journal of the Mammalogical Society of Japan 17:71–94. Nanami S., T. Yamakura, A. Itoh, and H. Kawaguchi. 2002. Population structure of Podocarpus nagi and Neolitsea aciculata at Mikasayama hill, Nara, Japan. Bulletin of Kansai Organization for Nature Conservation 24:29–43 Ohtaishi, N. 1978. Ecological and physiological longevity in mammals from the age structures of Japanese deer. The Journal of the Mammalogical Society of Japan 7:130–134. Shimoda, K., K Kimura, M. Kanzaki, and K. Yoda. 1994. The regeneration of pioneer tree species under browsing pressure of sika deer in an evergreen oak forest. Ecological Research 9:85–92. Takaragawa N., and T. Kawamichi. 1977. Rutting behavior of Cervus nippon in Nara Park. Annual report of the study of deer in the Nara Park 1976:43–61, Kasuga Manifestation Society. (In Japanese.) Takatsuki S. 1979. The vegetation of Nara Park with reference to the grazing effects of sika deer (Cervus nippon). Nara Park, annual report of the study of deer in the Nara Park 1978:113–132, Kasuga Manifestation Society. (In Japanese.) Takatsuki, S. 2001. Assessment of nutritional condition in sika deer by color of femur and mandible marrows. Mammal Study 26:73–76. Takatsuki S., and M. Asahi. 1977. Food habits of sika deer in Nara Park, assessed by fecal analysis (I). Nara Park, Annual report of the study of deer in the Nara Park 1976:129–141, Kasuga Manifestation Society. (In Japanese.)

25 Sika Deer in Nara Park: Unique Human-Wildlife Relations

363

Takatsuki S., and M. Asahi. 1978. Food habits of sika deer in Nara Park, assessed by fecal analysis (II). Nara Park, Annual report of the study of deer in the Nara Park 1977:25–37, Kasuga Manifestation Society. (In Japanese.) Tatsuzawa S., K. Fujita, and M. Itoh. 2002. Population dynamics of sika deer Cervus nippon on the flat part of the Nara Park. Bulletin of Kansai Organization for Nature Conservation 24: 3–14. (In Japanese.) Tatsuzawa S., and K. Fujita. 2001. Conservation and integrated management of deer of Nara: Ecological and historical viewpoints from citizens’ researches. Bulletin of Kansai Organization for Nature Conservation 23:127–140. (In Japanese.) Torii, H. 2006. Life table and population dynamics. Research Report of Natural Monument “Deer of Nara” Nara Prefecture Board of Education:1–9. (In Japanese.) Torii H., K. Suzuki, Y. Maesako, and N. Ichimoto. 2000. Stomach contents of the sika deer in Nara Park, Nara Prefecture. Bulletin of Kansai Organization for Nature Conservation 22:13–15. (In Japanese.) Yabu N., and K. Wada. 1996. Spatial patterns of individuals within resting groups of sika deer in Nara Park. Nankiseibutsu 38:79–86. (In Japanese.) Yamada M., and M. Urabe. 1998. Allogrooming of sika deer Cervus nippon in Nara Park. Bulletin of Kansai Organization for Nature Conservation 20:59–66. (In Japanese.) Yamakura T., T. Kawasaki, N. Fujii, T. Mizuno, D. Hirayama, H. Noguchi, S. Nanami, A. Itoh, K. Shimoda, and M. Kanzaki. 2001. Predictive discussion on the fate of an evergreen broadleaf forest at Kasugayama. Bulletin of Kansai Organization for Nature Conservation 23:157–167. (In Japanese.)

Chapter 26

A 20-Year History of Sika Deer Management in the Mt. Goyo Area, Northern Honshu Seiki Takatsuki

Abstract The biology and management of sika deer on Mt. Goyo in the Tohoku district of northeastern Japan is reviewed. This is a relatively isolated deer population in this part of Japan. The deer subsist primarily on dwarf bamboo; 60–80% of winter diet determined by fecal analysis is Sasa nipponica. Deer migrate in elevation in response to winter snowfall. Density on the winter range is high and the deer are forced to feed on fallen leaves and tree bark. Deer numbers were low in the early 1900s and received legal protection as a cultural resource from 1919 to 1929. Because the deer population increased, with consequent damage to agricultural crops, hunting of males was resumed in the early 1930s. Thereafter the population continued to increase, despite the increasing kill of males, and winter starvation began to occur. Therefore, kill of females and cull hunting were instituted, with the female kill approaching the male kill by the mid-1990s. This hunting plan has stabilized the situation and must be continued to maintain a balance between deer numbers, agricultural damage, effects on vegetation, and winter starvation. Deer population control will continue to be required to manage deer and human interaction at Mt. Goyo, including cultural and esthetic values.

Introduction Because of a heavy hunting pressure and snow which promotes hunting efficiency, the distribution of sika deer in the Tohoku (northeastern) district is the most fragmented in Japan. The population of Mt. Goyo is almost the only major population in the district and it is isolated from other populations. The closest small populations are on the Oshika Peninsula, 100 km south of Mt. Goyo and on Kinkazan Island off the Oshika Peninsula (Fig. 26.1). The Mt. Goyo population is characterized by (1) isolation in distribution, (2) severe winters, (3) exposure to hunting, and (4) problems with agricultural and forestry damage. In addition, the historical fact that it once faced the danger of extinction, common in many areas of Japan, in the early half of this century should be recalled. It is worthwhile to record the history of this population since it illustrates several characteristics of wildlife management in Japan. D. R. McCullough et al. (eds.), Sika Deer: Biology and Management of Native and Introduced Populations, DOI: 10.1007/978-4-431-09429-6_26, © Springer 2009

365

366

S. Takatsuki

Fig. 26.1 Map of the Tohoku district showing the distribution of the sika deer population in the Mt. Goyo area and location of other populations.

Environment and the Sika Deer Mt. Goyo is situated at the southern part of the Kitakami Mountains on the Pacific side of the Tohoku district (Fig. 26.1). The summit is 1,341 m in elevation and, like other mountains in this area, the topography is relatively gentle and the peaks and ridges are not sharp like those of other Japanese mountains. The higher elevations have been preserved since the Edo era, and most are covered by natural or seminatural forests. The higher areas are covered by alpine vegetation, and below that coniferous forests occur. Below the coniferous forests, beech forests predominate, and lower areas are covered by secondary oak forests. The foothills are covered by coniferous forest plantations or used as pastures. Below the protected forest areas the bottomland alluvial areas are used for rice paddy fields. Most of the forest areas are densely covered by Sasa nipponica, a dwarf bamboo (Fig. 26.2). This bamboo is unique because of its simple morphology. Many dwarf bamboos evolved in the Japanese Archipelago (Suzuki 1961), but only this species has no branch on its culm. This is an adaptation to the thin-snow environment. Because of dry, cold winds, it is harmful for plants to be exposed above the snow. This is the reason why evergreen plants are limited in these areas. Since bamboos are evergreen, taller and branched bamboos are damaged above snow. It is adaptable to be shorter and not to have branches above the ground. Sasa nipponica is such a dwarf bamboo. By doing so, a new culm appears every year, so the turn-over rate is rapid; the life-span of the leaves is only one year. In the deep-snow areas, plants are buried in snow where humidity is high and temperature does not go much below 0 °C. This environment is favorable for plants

26 A 20-Year History of Sika Deer Management in the Mt. Goyo Area

367

Fig. 26.2 Inside view of an oak forest with dense cover of Sasa nipponica at the foothill of Mt. Goyo.

Fig. 26.3 Morphology of dwarf bamboos for several genera taxa showing branching culms (from Suzuki 1961).

to overwinter. In such an environment, it is adaptable for dwarf bamboos to be taller and have branches to allocate products into supporting organs and to possess longer-lived leaves. Morphology relates to ecology: taller and branched bamboos are long-lived and the life-span of the leaves is longer (Fig. 26.3). They accumulate leaves of several cohorts, and the leaves gradually turn over in contrast to the thin-snow areas where plants are exposed above snow.

368

S. Takatsuki

Fig. 26.4 A dogwood tree from which the bark was peeled by wintering sika deer.

The deer in the Mt. Goyo area subsist on Sasa nipponica in winter. In most of the area of Mt. Goyo, fecal composition included 60–80% S. nipponica whereas it only contained 20% in the foothill areas where pasture grasses were more important (Takatsuki 1986). The deer move up and down the mountain depending on snow (Takatsuki et al. 2000). It begins to snow in late December and snow melts in April in the low areas and in May in the high areas. The “mountain deer” or migratory deer come down to the lower areas, often valleys, and stay there during winter. Such wintering areas are often situated at the outskirts of the reserve. The densities in the wintering areas are often very high, and the deer even eat fallen dead leaves of trees and peel bark off the trunks (Fig. 26.4). In winters of heavy snow, the deer are forced to go further down. Since hunting is permitted and pest control is performed outside the reserve, more deer are killed in such winters.

26 A 20-Year History of Sika Deer Management in the Mt. Goyo Area

369

Deer Management The status of the sika deer at Mt. Goyo before the 1970s is unclear. Old hunters report that sika deer were quite uncommon before World War II. For example, in the 1910s hunters left home in early morning, walked to Mt. Goyo, and stayed there for several days. Such visits were repeated many times in winter, but they could find few sika deer. It is fairly certain that there were no sika deer in the areas surrounding Mt. Goyo. It was thought in the 1910s that the population faced extinction. Given this situation, Iwate Prefecture adopted a preservation campaign for this population as “the northern-most population in Honshu.” The prefecture completely prohibited hunting anywhere in the prefecture from 1919 to 1929. However, the deer population did not show recovery. The prefecture then adopted another 10-year hunting prohibition from 1933. Information during the war is quite limited. Though only one deer was officially reported as killed during 1945–1949, it is doubtful because many records during the war are not accurate, and there is a record that a pioneer village in Mt. Goyo suffered damage by deer and the villagers finally abandoned it. This strongly suggests abundant deer. It was only after 1950 when accurate reports of hunted deer were recorded; about 20–30 deer were killed every year until 1970. Thereafter, the record shows a rapid increase: the deer number exceeded 100 in 1971, 200 in 1973, and 400 in 1976. Damage to agriculture became apparent after 1965. Pest control was begun from 1970, although very few deer were culled before then. Control was intensified in 1976 when as many as 150 deer were culled. At this time, only males were culled, but in 1978 females were also targeted. The population was estimated to be less than 1,000 in the late 1970s, but more than 1,000 males were shot in 1980. This winter was a record-breaking snowy one, and the deer were forced to move down into valleys where they could not move about, and hunters could easily shoot them. Nevertheless, the prefecture did not change the principle. Hunting was prohibited, and the protected area was enlarged from 833 ha in 1948 to 32,000 ha in 1972. This “complete protection” resulted in overpopulation in and around the reserve. In the late 1970s and early 1980s, the deer distribution expanded, and damage to agriculture and forestry became serious. The western limit of the distribution crossed the Kesen River in the 1980s. The distribution of the deer earlier was estimated to be 320 km2, but it increased to be 600 km2 by 1985 (Fig. 26.5). Planted trees, pasture grasses, and crops were badly damaged. People tried to establish fences around crop fields. The deer density increased in the reserve, and many starved deer were found in wintering areas. The prefecture performed supplementary feeding at the lower boundary of the reserve on another side (Fig. 26.6). They fed blocks of hay grasses to the deer. This supplementary feeding was intended to stop deer from going down to the crop fields at lower elevations. However, it lowered the mortality of the deer through starvation. This policy was contradictory because the prefecture, keeping the slogan to protect the northern-most Honshu deer to present their conservationist face to the public, at the same time permitted population control to mitigate the

370

S. Takatsuki

Mt. Goyo to 1950

1970 s 2001 1985

Fig. 26.5 Expansion of deer distribution in the Mt. Goyo area over time.

complaints by local farmers. When damage became quite heavy and farmers’ complaints became quite serious, the prefecture at last changed their policy, increasing population control and ending supplementary feeding. They also established fences around the pastures (Fig. 26.7). A considerable number of deer died in the spring following the fence establishment around the fenced pastures. In the spring of 1993, for example, as many as 65 carcasses were found in an area of 0.83 km2, or the carcass density was 78/km2. Since in Japan only male deer were legal as game animals, hunting females was prohibited. In the late 1980s and the early 1990s, local hunters hunted almost every weekend, but the damage did not decline, which means that the capability of the hunters reached the upper limit under males-only regulations. Male-only hunting is not effective at stopping population growth, and the damage continued to increase. The deer spread to cover 700 km2 (Fig. 26.5). In view of this situation, the Ministry of Environment, Japan permitted hunting females as game animals in 1994. Because of this, the proportion of females in the deer kill abruptly increased after 1994 to just less than the male kill, although the kill of male deer remained relatively stable during this time (Fig. 26.8).

26 A 20-Year History of Sika Deer Management in the Mt. Goyo Area

371

Fig. 26.6 Supplementary feeding of dried hay bales for the deer at the foothill of Mt. Goyo.

Fig. 26.7 Fence established at a pasture to prevent deer invasion.

The combined kill of the sexes resulted in a nearly doubling of the total kill over that of the 1980s. After a time lag of several years, the effects of female-hunting population control appeared. The damage peaked in 1993 and thereafter declined every year (Fig. 26.9). By 2002, the damage had decreased to 5% of the amount in 1993. It seems that the most serious phase of damage has passed. However, the population control should be continued, otherwise the population will rapidly recover.

372

S. Takatsuki 2000

Number killed

1500

1000 male 500 female 0 1980

1985

1990 Year

1995

2000

Fig. 26.8 Changes over time in male (•) and female (°) sika deer killed by hunters and culled in the Mt. Goyo area.

Fig. 26.9 Changes over time in damages (in Japanese yen) on agriculture and forestry by sika deer in the Mt. Goyo area.

In fact the fecundity of yearling females, which was very low in the late 1980s and early 1990s when the deer density was highest, recovered after 1994. Fecundity of females older than two years also decreased in the early 1990s and recovered after 1994 but it was higher in the late 1980s (Fig. 26.10). It is important to remember at this phase that the objective of the population control of the deer is to reduce damage and not to cause the extinction of the population. It should be remembered that this is the only area in the Tohoku district where the sika deer population has been conserved. The population should be conserved as a flagship species of the high biodiversity of the Kitakami Mountains.

26 A 20-Year History of Sika Deer Management in the Mt. Goyo Area

373

Fig. 26.10 Changes in fecundity of female sika deer in the Mt. Goyo area (• central part of the distribution, ° peripheral part of the distribution).

Literature Cited Suzuki, S. 1961. Ecology of bambusaceous genera Sasa and Sasamorpha in the Kanto and Tohoku districts of Japan, with special reference to their geographical distribution. Ecological Review 15:131–147. Takatsuki, S. 1986. Food habits of sika deer on Mt. Goyo. Ecological Research 1:119–128. Takatsuki, S., K. Suzuki, and H. Higashi. 2000. Seasonal up-down movements of sika deer at Mt. Goyo, northern Japan. Mammal Study 25:107–114.

Chapter 27

Survival Patterns of Male and Female Sika Deer on Kinkazan Island, Northern Japan Masato Minami, Nobumasa Ohnishi, and Seiki Takatsuki

Abstract We constructed life tables for sika deer males and females on Kinkazan Island in northeastern Japan to follow survivorship and examine patterns of mortality by sex, age, and social status in this population. Because there are no predators or human hunting on Kinkazan Island, these results represent background, natural mortality, which is characteristic of this protected situation. Mortality was high in both sexes among fawns, with most occurring shortly after birth or near the end of winter of the first year. Females survived better than males at all ages: they had a mean lifespan of 4.0 years compared to 3.1 years in males; both were short because of the high mortality of fawns. Dominant males survived better than subordinate males. Most mortality of adults occurred in late winter and was related to starvation. The close connection of mortality to food shortage is reflective of the high deer density in relation to carrying capacity in this protected population.

Introduction Life table construction is useful for population analysis as well as population management for wildlife in determining whether a population is stable, the age structure is stable, and if age-specific mortality and fecundity are stable (Caughley 1977). However, these assumptions are unrealistic in the real world (McCullough 1979; Seber 1982). For example, birth rate and mortality are not stable in most habitats, particularly high latitude areas where climatic conditions greatly vary among years, and recent studies are revealing that such assumptions are unrealistic in many situations (Lebreton et al. 1993; Jorgenson et al. 1997; Unsworth et al. 1999). Besides lack of stability in natural systems, precise age structure is often difficult to determine for wildlife populations. For example, age determination from many individuals without capturing or sampling some body parts that are indicative of age of the animals is almost impossible. If sample size is small (e.g. <150), age structure is imprecise (Caughley 1977). When a life table is constructed from carcasses, it is also often accompanied by difficulties. For example, remains of D. R. McCullough et al. (eds.), Sika Deer: Biology and Management of Native and Introduced Populations, DOI: 10.1007/978-4-431-09429-6_27, © Springer 2009

375

376

M. Minami et al.

offspring are fragile and decay rapidly, which results in biases in collection (Sinclair 1977). When animals are sampled from hunter-killed animals, sampling is often biased towards larger animals (Leader-Williams 1988). These practical limitations make precise life table construction difficult. Therefore, it is often preferable to trace animals through their life-spans rather than to construct life tables using unrealistic assumptions, with consequent inaccuracies. However, this is again difficult for long-lived animals such as sika deer. It naturally takes a long time to complete such a data set. A life table indicates the life history characteristics of each species, although it is also affected by environmental factors. For ungulates, for example, factors affecting mortality include predation and hunting, which can influence survival patterns greatly. If a population is free from these variable factors, the life table would show the life history characteristics in a more simple way, but valid only under those circumstances. Although there are some good long-term studies of other species (red deer, Cervus elaphus, Clutton-Brock et al. 1982; Soay sheep, Ovis aries, Clutton-Brock and Pemberton 2004; bighorn sheep, Ovis canadensis, Jorgenson et al. 1997), information on sika deer is still limited. In this study we tried to fill this gap by constructing a realistic life table for the sika deer population living on Kinkazan Island (see location map in Ohnishi et al. chapter 7), where it was possible to follow individuals through their lifetimes. The sika deer population on Kinkazan Island lacks predators, and hunting has been prohibited for religious reasons. The population for the whole island (9.6 km2) is about 450 individuals. Most of them live in a wild state, but the individuals in a herd living in the western part of the island are tame, so observation at close distance is possible. We began a study of this herd in 1990, and all the births and nearly all of the deaths were known up to the spring of 2004. All the deer are identified, and mother-fawn relations were known (Minami et al. chapter 23). Birth and death could be accounted for, and the mortality factors were determined. It was possible, therefore, to construct precise life tables for both males and the females.

Methods We identified all the deer living around the shrine in 1990, and thereafter all the new-born deer were also identified. Deer were identified by natural marks such as scars and white spot patterns on the body, as well as by hair staining and inserting microchips (Trovan Electronic Identification System) under the skin. We live-captured the deer every March since 1991 to weigh, measure, and mark them. Natality and mortality of the deer were determined during the period from 1990 to 2004. In autumn from 1990 to 1993, mothers and their fawns were recorded. Since 1994, mothers and their newborn fawns immediately after parturition were observed. Survival of the fawns was checked in March, May, and October. When

27 Survival Patterns of Male and Female Sika Deer on Kinkazan Island

377

we found carcasses of deer that had died we identified them by natural marks and microchips. When we could not find individual deer for longer than several months, we regarded them as “disappeared” and included them in the tally of deer that had died. Intensive surveys to recover carcasses were carried out in May, October, and March–April. Numbers of deer at every age were determined, and the age-specific mortalities of deer born during the study period were calculated. We observed rutting behavior of the males during the rut (Minami et al. chapter 22). We categorized the males into “dominant males” (DMs) and “subordinate males” (SMs). DMs included two types: territory holders who exclude other males and nonterritorial males who engage in reproduction by other strategies. When SMs fight with territory holders, they fight to a draw, or even sometimes win, but do not hold territories and are subordinate to DMs. From these results we could compare survivorship of males of different dominance categories.

Results A total of 276 deer (149 males and 127 females) was born from 1990 to 2003. Among them, 170 deer (97 males, 59 females, and 14 sex-unknown animals) died from 1990 to 2004. Among these dead deer, 14 sex-unknown deer were fawns who died immediately after birth.

Age-Specific Mortality The mortality of the males (65.1%) was significantly higher than that of the females (46.5%, p < 0.002, χ2-test). As many as 22.7% of the females and 31.4% of the males died during the first year of age. The age-specific mortalities were determined by using the data taken in the whole study period, but the mortality of fawns was determined only for those whose “just-born” deaths were known after 1994. According to the age-specific mortalities, survival curves of all the males and the females were calculated (Fig. 27.1). The mortalities in the first year of life were particularly high. The mortalities of the males three to six years of age were relatively low (about 5%) while it increased after seven years old. For the females, the mortality was high only in the first year and, thereafter, only few females died before eight years of age. Mortality increased gradually after nine years old. Figure 27.2 shows the age-specific mortalities of the males and the females. The mortalities of the males were higher than that of females except for 9–10 years old and 11–12 years old. The differences were significant at the ages of 0–1 years (χ2 test, p < 0.05), 6–7 years (P < 0.02), and 7–8 years old (p < 0.05). The age-specific mortality of females did not reach to 100% because four 14-year-old females were still living in the spring of 2004.

378

M. Minami et al.

Fig. 27.1 Survival curves of males (•) and females (°) of sika deer on Kinkazan Island constructed from the information of all the births and almost all of the deaths.

Fig. 27.2 Age-specific mortalities of males (•) and females (°) of sika deer on Kinkazan Island. Four 14-year-old females were alive in the spring of 2004.

27 Survival Patterns of Male and Female Sika Deer on Kinkazan Island

379

The mean life spans of males and females were 3.1 years (n = 104) and 4.0 years (n = 53) respectively, which were not significantly different (Mann-Whitney U-test, p = 0.193).

Social Status of Males Social status of 97 males who were born and died between 1990 and 2004 was determined. All the males were SMs up to the autumn of four years old. Four males (13.6%) became DMs at five years old, and thereafter the proportions of them increased because more males became DMs at the same time some males died. As many as 77.7% and 83.3% of the males became DMs at 10 and 11 years old, respectively. Among the 14 males who lived longer than eight years, 13 males (92.9%) became DMs at least once in their lives. Mortalities of the SMs and DMs are shown separately according to their social status in the previous years in Fig. 27.3. The mortality of the SMs was always higher than that of the DMs. The life spans of the males who became a DM at least sometime during in their lives and those of males who never became a DM were compared. The life span of the former was 9.47 +/− 2.17 years (mean +/− SD), whereas that of the latter was 6.08 +/− 1.56 years. The former was significantly longer than the latter (Mann-Whitney U-test, p = 0.003).

Fig. 27.3 Age-specific mortalities of subordinate males (SM, °) and dominant males (•) of sika deer on Kinkazan Island. * Values of SMs older than eight years are excluded because of small sample size.

380

M. Minami et al.

Table 27.1 Mortality of male and female sika deer (age older than one year) on Kinkazan Island in different periods. Deaths Mortality rate/month May–Sept. Oct.–early Mar. Mid-Mar.–Apr. Total

Male

Female

Male

Female

1 (2.3%) 27 (62.8%) 15 (34.9%) 43 (100%)

1 (4.0%) 17 (68.0%) 7 (28.0%) 25 (100%)

0.2 5.1 9.0

0.2 3.2 4.2

Seasonal Mortality Years were divided into three periods according to intensive carcass recoveries (Table 27.1). Both the males and the females older than one year old died mainly from October to early March (62.8% males and 34.9% females) and from midMarch to April (34.9% males and 28.0% females), while only a few deer died during the rest of the year. Mortality rates per month were highest during the period from mid-March to April both in males and females as well as in fawns and older deer (Table 27.1). Fawn mortality occurred more evenly over time than for older deer because deaths immediately after births were frequent, while many deaths of older deer were concentrated in early spring. Monthly mortalities of fawns were much higher than those of older deer.

Discussion Given that deer on Kinkazan Island are free from predation and hunting, the life tables derived in our study seem to reflect the innate nature of the sika deer population dynamics. Mortalities were high in the first year of life for both males and females. This is known in many cervid species: red deer (Ahlen 1965; Lowe 1969; Guiness et al. 1978), wapiti (Cervus elaphus canadensis, McCullough 1969), moose (Alces alces, Peterson 1977; Franzmann et al. 1980; Ballard et al. 1981, 1991; Albright and Keith 1987; Larsen 1989; Gasaway et al. 1992), mule deer (Odocoileus hemionus, White et al. 1987), white-tailed deer (Odocoileus virginianus, O’Pezio 1978), roe deer (Capreolus capreolus, Pielowski 1984; Gaillard et al. 1993). In many cases, predation is one of the main mortality factors: red deer (Jedrzejewski et al. 1992), moose (Gasaway et al. 1983; Messier and Créte 1985; Ballard et al. 1987), mule deer (White et al. 1987), roe deer (Danilkin 1996), but since there were no predators on Kinkazan Island, the high mortality of the fawns of this population is attributed to other factors. Ohnishi et al. (chapter 7) carefully observed parturitions and the growth of fawns on Kinkazan Island and showed that there are two peaks in fawn mortality. One happens immediately after birth caused by accidents and crow (Corvus macrorhynchos) attacks while another is due to starvation during the first winter after birth.

27 Survival Patterns of Male and Female Sika Deer on Kinkazan Island

381

The deaths of deer older than one year were concentrated in winter and spring, particularly during March and April (Table 27.1), which strongly suggests that the most important mortality factor for this population was starvation. Body size reduction of the deer, poor antler development of the males, and low pregnancy rate of the females (S. Takatsuki unpublished) suggest poor nutritional condition of the population and support this explanation. Mortality was significantly higher in males than in females (Fig. 27.1). This is known in many other cervids: red deer (Lowe 1965), moose (Peterson 1977; Albright and Keith 1987; Boer 1988), Odocoileus spp. (Taber and Dasmann 1957; Klein 1970; Fuller 1990; McCorquodale 1999), and roe deer (Strandgaard 1972; Kaluzinski 1982). The mortality of the males thereafter was low up to six years old, but increased rapidly after seven years old (Fig. 27.1). The most marked difference in mortality between the males and the females was at ages from six to eight years old (Fig. 27.2). These ages seem to be important in the life of males. It is known that the body weight of males increases at younger ages and attained a maximum at these ages (S. Takatsuki et al. unpublished), when some males begin to hold their harem territories during the rut. These males are subject to fighting against other males, which would result in heavy stresses. Dominant males do not have enough time to feed and, consequently, lose body weight (Minami 2008). Not all males can become DMs, and the mortality between four and seven years old was higher than for those who survived to become DMs (Fig. 27.3). That is, all the males who live longer than eight years became DMs. Consequently, the mean life span of the experienced DMs was significantly longer that that of others who did not live to become DMs. This suggests there are great differences in reproductive success between them. Social activities relating to the rut, such as exclusion and watching of other males and fighting with them, bugling to advertise themselves to other males, and holding and guarding females are intensified at these ages, and therefore the burden on them is increased. Nevertheless, mortality of males decreased at seven years old. After 12 years old, mortality increased again, and only a few males survived thereafter. Although the mean life span of males was shorter than that of females, the difference was not statistically significant. This seems to be due to the great contribution of fawn mortality to all deaths (56% for males and 53% for females). The males’ life span (3.1 years) was longer than those of other populations subject to hunting and possibly predation of bears (brown bears, Ursus arctos in Hokkaido and black bears, Ursus thibetanus in Honshu). Some studies on Hokkaido populations have shown that the life spans of males were 2.9–3.1 years (Hokkaido Institute of Environmental Sciences 1997), and a study on the population of Mt. Goyo, a northern Honshu population, has shown that it was only 1.0 years (Takatsuki et al. 1991), although these studies required some assumptions in life table construction. The life span of the females on Kinkazan Island (4.0 years), in turn, was similar to or a little longer than those of other populations. Those of Hokkaido populations were, for example, 3.6–3.9 years old (Hokkaido Institute of Environmental Sciences 1997), and that of Mt. Goyo was 2.9 years old (Takatsuki et al. 1991). It seems to be typical that the life spans of the males and the females in other populations were similar to those of the Kinkazan population. Although the reason is not clear,

382

M. Minami et al.

it seems to relate to the fact that size reduction of the Kinkazan deer is more marked in the males and, therefore, sexual dimorphism is smaller in this population (S. Takatsuki unpublished). It is expected that pregnancy and care of fawns would cause stresses on females. In spite of the expectation, mortality of females on Kinkazan Island was not high at four to six years old when the females began reproduction. The body weight of the females stabilizes at around four years old (S. Takatsuki et al. unpublished). After this age, body weight and pregnancy rate were stable until nine years old. That is, mortality was not different between either the growing ages and maturity weight ages, or between milk females and yeld females. Female mortality increased after nine years old. Since reproduction continued even at this age, the burden of parturition and lactation may increase after this age. Though this population lives in a habitat where natural vegetation is conserved and the deer population is not artificially managed, it should be remembered that (1) some of the deer, particularly DMs, feed on garbage mainly in winter, which may lower mortality, and (2) males’ antlers are removed in early October in order to avoid accidents to visitors, which may affect the social interrelations among the males, and possibly reduce serious injuries in fighting. It is also noteworthy that deer density has long been close to the carrying capacity of the island because of no population control. This has resulted in reduction of the deer (Ohnishi et al. chapter 7) and low pregnancy rates in females (Takatsuki et al. 1991).

Literature Cited Ahlen, I. 1965. Studies on the red deer, Cervus elaphus L., in Scandinavia, III. Ecological investigations. Viltrevy 3:1–376. Albright, C. A., and L. B. Keith. 1987. Population dynamics of moose, Alces alces, on the southcoast barrens of Newfoundland. Canadian Field Naturalist 101:373–387. Ballard, W. B, T. H. Spraker, and K. P. Taylor. 1981. Causes of neonatal moose calf mortality in southcentral Alaska. Journal of Wildlife Management 45:335–342. Ballard, W. B., J. S. Whitman, and C. L. Gardner. 1987. Ecology of an exploited wolf population in south-central Alaska. Wildlife Monographs 98:1–54. Ballard, W. B., J. S. Whitman, and D. J. Reed. 1991. Population dynamics of moose in south-central Alaska. Wildlife Monographs 114:1–49. Boer, A. H. 1988. Mortality rates of moose in New Brunswick: A life table analysis. Journal of Wildlife Management 52:21–25. Caughley, G. 1977. Analysis of vertebrate populations. Wiley, New York, New York, USA. Clutton-Brock, T. H., and J. Pemberton. 2004. Soay sheep: Dynamics and selection in an island population. Cambridge University Press, Cambridge, United Kingdom. Clutton-Brock, T. H., F. E. Guiness, and S. D. Albon. 1982. Red deer: Behavior and ecology of two sexes. Chicago University Press, Chicago, Illinois, USA. Danilkin, A. 1996. Behavioural ecology of Siberian and European roe deer. Wildlife Ecology and Behaviour Series 2, Chapman & Hall, London, United Kingdom. Franzmann, A. W., C. C. Schwartz, and R. O. Peterson. 1980. Moose calf mortality in summer on the Kenai Peninsula, Alaska. Journal of Wildlife Management 44:764–768. Fuller, T. K. 1990. Dynamics of a declining white-tailed deer population in north-central Minnesota. Wildlife Monographs 110:1–37.

27 Survival Patterns of Male and Female Sika Deer on Kinkazan Island

383

Gaillard, J. M., D. Delorme, J. M. Boutin, G. V. Laere, B. Boisaubert, and R. Pradfel. 1993. Roe deer survival patterns: A comparative analysis of contrastive populations. Journal of Animal Ecology 62:778–791. Gasaway, W. C., R. D. Stephenson, J. L. Davis, P. E. K. Shepherd, and O. E. Barris. 1983. Interrelationships of wolves, prey, and man in interior Alaska. Wildlife Monographs 84:1–50. Gasaway, W. C., R. D. Boertje, D. V. Grangaard, D. G. Kelleyhouse, R. D. Stephenson, and D. G. Larson. 1992. The role of predation in limiting moose at low densities in Alaska and Yukon and implications for conservation. Wildlife Monographs 120:1–59. Guiness, F. E., T. H. Clutton-Brock, and S. D. Albon. 1978. Factors affecting calf mortality in red deer (Cervus elaphus). Journal of Animal Ecology 47:817–832. Hokkaido Institute of Environmental Sciences. 1997. Report of the study on the present status of brown bears and sika deer in Hokkaido, III. Hokkaido Institute of Environmental Sciences, Sapporo, Hokkaido, Japan. Jedrzejewski, W., B. Jedrzejewska, H. Okarma, and A. L. Ruprecht. 1992. Wolf predation and snow cover as mortality factors in the ungulate community of the Bialowieza National Park, Poland. Oecologia 90:27–36. Jorgenson, J. T., M. Festa-Bianchet, J. M. Gaillard, and W. D. Wishart. 1997. Effects of age, sex, disease, and density on survival of bighorn sheep. Ecology 78:1019–1932. Kaluzinski, J. 1982. Dynamics and structure of a field roe deer population. Acta Theriologica 27:385–408. Klein, D. R. 1970. Food selection by North American deer and their response to over-utilization of preferred plant species. Pages 25–44 in A. Watson, editor, Animal populations in relation to food resources, Blackwell Scientific, Oxford, United Kingdom. Larsen, D. G., D. A. Gauthier, and R. L. Markel. 1989. Causes and rate of moose mortality in the southwest Yukon. Journal of Wildlife Management 53:548–557. Leader-Williams, N. 1988. Reindeer on South Georgia. Cambridge University Press, Cambridge, United Kingdom. Lebreton, J. D., R. Pradel, and J. Clobert. 1993. The statistical analysis of survival in animal populations. Trends in Ecology and Evolution 3:91–95. Lowe, V. P. W. 1969. Population dynamics of the red deer (Cervus elaphus) on Rhum. Journal of Animal Ecology 38:425–457. McCorquodale, S. M. 1999. Movements, survival, and mortality of black-tailed deer in the Klickitat Basin of Washington. Journal of Wildlife Management 63:861–871. McCullough, D. R.1969.The tule elk: Its history, behavior, and ecology. University of California Publications in Zoology 88:1–209. McCullough, D. R. 1979. The George Reserve deer herd: Population ecology of a K-selected species. University of Michigan Press, Ann Arbor, Michigan, USA. Messier, F., and M. Créte. 1985. Moose-wolf dynamics and the natural regulation of moose populations. Oceologia 65:503–512. Minami, M. 2008. Individual life history and reproductive success: Sika deer. pp. 123–148 in S. Takatsuki and J. Yamigawa, editors, Mammalogy in Japan 2: Middle- and large-sized mammals, including primates. University of Tokyo Press, Tokyo, Japan. O’Pezio, J. P. 1978. Mortality among white-tailed deer fawns on the Seneca Army Depot. New York Fish and Game Journal 25:1–15. Peterson, R. O. 1977. Wolf ecology and prey relationships on Isle Royale. U. S. National Park Service Monograph Series 11:1–210. Pielowski, Z. 1984. Some aspects of population structure and longevity of field roe deer. Acta Theriologica 29:17–33. Seber, G. A. F. 1982. The estimation of animal abundance and related parameters, 2nd ed. Griffin, London, United Kingdom. Sinclair, A. R. E. 1977. The African buffalo: A study of resource limitation of populations. University of Chicago Press, Chicago, Illinois, USA. Strandgaard, H. 1972. The roe deer (Capreopus capreolus) population at Kalo and the factors regulating its size. Danish Review of Game Biology 7:1–205.

384

M. Minami et al.

Taber, R. D., and R. F. Dasmann. 1957. The dynamics of three natural populations of deer Odocoileus hemionus columbianus. Ecology 38:233–246. Takatsuki, S., S. Miura, K. Suzuki, and K. Ito-Sakamoto. 1991. Age structure in mass mortality in the sika deer (Cervus nippon) population on Kinkazan Island, northern Japan. Journal of the Mammalogical Society of Japan 15:91–98. Unsworth, J. W., D. F. Pac, G. C. White, and R. M. Bartmann. 1999. Mule deer survival in Colorado, Idaho, and Montana. Journal of Wildlife Management 63:315–326. White, G. C., R. A. Garrott, R. M. Bartmann, L. H. Carpenter, and A. W. Alldredge. 1987. Survival of mule deer in northern Colorado. Journal of Wildlife Management 51:852–859.

Chapter 28

Sika Deer in an Evergreen Broad-Leaved Forest Zone on the Boso Peninsula, Japan Masahiko Asada and Keiji Ochiai

Abstract Meteorological variation throughout the year causes seasonal variations (seasonality) in the availability of food resources for herbivores. This seasonality creates latitudinal and altitudinal variations not only in food habits, but also in body size (Lindstedt and Boyce 1985; Geist 1986) and reproduction (reviewed by Bronson 1989) of the herbivores within a reaction norm (Stearns and Koella 1986). As sika deer have a broad latitudinal distribution (15–45° N; Ohtaishi 1986; Whitehead 1993), by comparative studies of nutritional ecology of sika deer in each habitat they occupy we can examine how they react to seasonality as evolved through adaptive radiation from warmer habitats near the equator to colder northern habitats. We present results on deer on the Boso Peninsula in central Japan to show the influence of seasonality on their nutritional ecology.

Introduction Over the past decade we studied sika deer on the Boso Peninsula of Honshu Island, Japan (near Tokyo), including population density, body size (Ochiai and Asada 1995, 1997), food habits (Asada and Ochiai 1996a), food quality (Asada and Ochiai 1999), body condition (Asada 1996), reproduction (Asada and Ochiai 1996b), and deer management (Asada and Ochiai 1998). In this research, the effect of seasonality of food resources for sika deer dwelling in evergreen broad-leaved forests has been obvious in their nutrition. The purpose of this paper is to review studies on nutritional ecology of sika deer on the Boso Peninsula and to discuss what consequences can be inferred from the seasonality in their food environment.

Study Area The Boso Peninsula is located in Chiba Prefecture in central Japan (35° N, 140° E; Fig. 28.1). Elevation ranges only from 0 to about 300 m, and the topography is very steep. Annual precipitation is 2,000–2,400 mm, and the mean monthly D. R. McCullough et al. (eds.), Sika Deer: Biology and Management of Native and Introduced Populations, DOI: 10.1007/978-4-431-09429-6_28, © Springer 2009

385

386

M. Asada, K. Ochiai N Honshu

Tokyo The Boso Peninsula

Distribution of sika deer in the Boso Peninsula 2001 1978-80 1973-74

The Pacific Ocean 0

10km

Fig. 28.1 Distribution of sika deer on the Boso Peninsula, central Japan in 1973–74 (Koganezawa et al. 1976), 1978–80 (Chiba Prefecture 1981), and 2001 (Chiba Prefecture and Boso Deer Research Group 2002).

temperature is about 4 °C in midwinter and 25 °C in mid-summer (University of Tokyo 1988). Vegetation on the peninsula is characterized by evergreen broad-leaved forests of Machilus thunbergii (Tabu-no-ki tree, Lauraceae), Castanopsis cuspidate var. sieboldii (Fagaceae), and planted Lithocarpus edulis (Japanese tanbark oak); natural coniferous forests of Abies firma (Momi fir) and Tsuga sieboldii (southern Japanese hemlock); and plantations of the conifers, Cryptomeria japonica (Japanese cedar) and Chamaecyparis obtusa (Hinoki false cypress).

Sika Deer Population Dynamics Before the 1860s (Edo era), many sika deer were distributed throughout the Kanto Plain (the coastal region north and east of Tokyo) (Furubayashi and Sinoda 2001) and on the Boso Peninsula (Ochiai 1995). With the logging of forests on the peninsula before World War II, the deer population on Boso became isolated from the others on the Kanto Plain. After World War II, construction of roads and hydroelectric dams in the area was vigorously pursued. Along with poaching for food and recreation, these activities greatly reduced the size and distribution of the deer herd. Thus, the Boso population experienced a bottleneck in the late 1950s. In 1961, to avoid extinction of the deer, the Chiba Prefecture government prohibited deer hunting. The population gradually recovered in the late 1970s and 1980s, and its range expanded rapidly (Fig. 28.1). In the early 1980s the population size was approximately 200–500 (Iimura and Chiba Prefecture 1981; Chiba

28 Sika Deer in an Evergreen Broad-Leaved Forest Zone on the Boso Peninsula

387

Prefecture 1987). From that time on the deer began causing frequent agricultural and forest damage, so legal hunting to control numbers was reintroduced. About 400–900 deer have been harvested annually since the late 1990s. Since 1992, we have estimated the deer population size by two methods: fecal pellet group counts on line transects throughout the distribution range and the block-count method every year in 14 research areas (Fig. 28.2; Nakama et al. 1980; Maruyama and Furubayashi 1983). The population size was about 2,000 in the 1990s and approximately 2,500–3,500 in 2002. The deer were distributed over 440 km2 in 2001. Local differences in deer density and yearly fluctuations are apparent within the Boso Peninsula (Fig. 28.2). The density in Amatsu-Kominato (areas 6, 11, 13, and 14), where deer have occurred since the 1950s, increased rapidly to a high of around 30–40 deer/km2 by 1991, and with hunting has since decreased to around 10 deer/km2. During the high density period around 1991, calf size, doe weight, and fat reserves decreased temporally (Fig. 28.3). However, no obvious mass mortality occurred during winters at the time.

Food Habits Seasonal Changes in Food Habits According to an analysis of rumen contents (Asada and Ochiai 1996a), graminoids and woody plants are the primary foods of sika deer on the Boso Peninsula throughout the year (Fig. 28.4). Evergreen species are important foods all year round, and deciduous broad-leaved species are important when in leaf. Interspecific variations in food habits of ruminants are due to body size, metabolic rate, and digestive capacity (Bell 1971; Geist 1974; Jarman 1974; Hofmann 1989). Hofmann (1985, 1989) classified ruminants by food habits into three categories: concentrate selectors, grass and roughage eaters, and intermediate feeders. He listed sika deer among intermediate feeders. Takatsuki (1988, 1990), however, placed sika deer in every category. Sika deer in northern Japan feed mainly on graminoids and are thus grass and roughage eaters (Takatsuki 1980, 1983, 1986). On the other hand, on Tsushima Island, western Japan, the deer consume far more leaves of woody species than graminoids (68.5% of total in fall; Takatsuki 1988), thus suggesting they are concentrate selectors. The mixed diet of the Boso deer we studied would identify them as intermediate feeders. It is apparent that sika deer show an unusual amplitude in food habits depending on environmental and habitat circumstances. Acorns of Lithocarpus edulis (tanbark oak), Quercus acuta (Japanese evergreen oak), Q. glauca (Japanese blue oak), Q. myrsinaefolia (an evergreen oak), and Castanopsis cuspidate var. sieboldii (Fagaceae) were consumed from September to April, with a peak in October (Asada and Ochiai 1996a; Fig. 28.4). Unlike consumption of leafy material, the consumption of mast does not influence the subsequent production of these tree species. Even if deer consume all acorns that

1990

10 0 1986

10 0 1986 2000

20

20

1995

30

1990

40

30

1990

1995

7

2000

20

12

10 0 1986

30

10 0 1986 2000

40

20

1995

5 50

1990

1990

1995

10 0 1986

20

30

40

50

30

2

1990

40

1990

3

10 0 1986

20

30

40

50

50

40

8

2000

2000

50

2000

1995

1995

50

1995

1990

1990

20 10 0 1986

10 0 1986

2000

30 20

9

1990

2000

40

1995

2000

30

11

1990

13

10

1995

50

1995

10 0 1986

20

30

40

50

1990

40

1990

2000

6

2000

10 0 1986

20

30

40

50

50

14

1995

1995

4

2000

1

1995

10 0 1986

20

30

40

50

2000

1990

Distribution of sika deer in the Boso Peninsula

1995

2000

Fig. 28.2 Density changes of sika deer on the Boso Peninsula, Central Japan from 1986 to 2002. Deer densities were estimated by the block-count methods at 14 sites (numbered dots). Y-axes refer to the deer densities (0–50 deer/km2) and X-axes refer to research year (1986–2002). Data from Sites 4–6 were obtained in collaboration with the University of Tokyo.

10 0 1986

20

30

40

50

10 0 1986

20

30

40

50

10 0 1986

20

30

40

50

N

388 M. Asada, K. Ochiai

28 Sika Deer in an Evergreen Broad-Leaved Forest Zone on the Boso Peninsula

389

Deer density (deer / km2)

50 Deer density in no.14

40 30 20

Deer density in no.11

10 0

180

)

45 9 10

35

10

12

6 11

7

140

21

8

30

120

Body weight of adult females

100

25 20

11

RKFl of adult females 6 7 9 3 9

10

60 )

15

80 4

40

9

10

20

5

0

0 )

2 100

Body weight of calves 2

15

RKFl of calves

10

80

2

11 12

5

11

8

11

8

7

5

6

5

40 20

)

1

60

RKFl (

20 Body Weight (kg

160

RKFl (

Body Weight (kg

8

11

40

8 10 16

7

0

Jaw length (cm)

Jaw length of calves 15 2 14

10 25

11 4

13 12

1986 1987 1988 1989 1990 1991 1992 19931994 19951996 1997

Fig. 28.3 Yearly fluctuations of deer density in winter, body weight and fat reserve (Riney’s kidney fat index, RKFI) of adult females and calves, and jaw length of calves in October from 1986 to 1997 in Amatu-Kominato town on the Boso Peninsula, central Japan. Numbers within figure indicate sample size.

390

M. Asada, K. Ochiai 50

Graminoids Non-bamboo

25 Bamboo 0 Woody Plants

50 Composition(%)

Evergreen broad leaves Deciduous broad leaves 25

0 Dead leaves 50

25

Others

Forbs

Acom

0 Apr. May Jun. Jul. Aug. Sep. Oct. Nov. Dec. Jan. Feb. Mar. Month

Fig. 28.4 Seasonal change in the food habits of sika deer on the Boso Peninsula (from Asada and Ochiai 1996a) as indicated by mean percentage composition of rumen contents.

fall to the forest floor in one year, they will not change the acorn production in the following year. Thus, the production of this food is independent of the density of deer, unlike that of leaves, which decline in biomass with foraging. Harlow et al. (1975) and McCullough and Ullrey (1985) examined the nutritive value of food items for white-tailed deer and showed that acorns had a high content of fats and soluble carbohydrates. As Moen (1973) pointed out, acorns have a significant effect on fat reserves in fall and mortality in winter. Thus, we consider acorns to be a key food resource for deer on the Boso Peninsula. The Boso Peninsula deer consumed little bark, unlike other populations, such as the Hokkaido (M. Yokoyama et al. 2000), Mt. Goyo (Takatsuki and Ikeda 1993),

28 Sika Deer in an Evergreen Broad-Leaved Forest Zone on the Boso Peninsula

391

Nikko (Maruyama et. al. 1975), Mt. Tanzawa (Furubayashi and Maruyama 1977), Mt. Ohdaigahara (S. Yokoyama et al. 1996; Ando and Shibata chapter 15), and Tsushima Island (Takatsuki 1988).

Sexual Differences in Food Habits in Fall Sympatric differences in food habits among deer grouped by age and sex have been reported in fallow deer (Dama dama) (Putman et al. 1993), white-tailed deer (Odocoileus virginianus) (Warren and Krysl 1983; Beier 1987), red deer (Cervus elaphus) (Staines et al. 1982), and sika deer (Padmalal and Takatsuki 1994). According to several studies, females consumed higher quality foods than males, reflecting an allometric relationship among body size, digestive capacity, and metabolic requirements (Staines et al. 1982; Clutton-Brock and Harvey 1983; Beier 1987; Padmalal and Takatsuki 1994). However, males on Boso consumed significantly more acorns and less browse than females in fall (Asada and Ochiai 1996a). Putman et al. (1993) observed that male fallow deer consumed more digestible forage than females in fall and winter. Warren and Krysl (1983) found that male white-tailed deer fed on more acorns than females. It has been considered that, in fall, males have a different nutritional requirement from females because of rutting behavior (Warren and Krysl 1983; Asada and Ochiai 1996a). Acorns have high fat and soluble carbohydrate content that is easily converted into energy, but lower protein content than that of evergreen broad-leaves (Harlow et al. 1975; Asada and Ochiai 1999). Adult males require a lot of energy in the rutting season for wallowing, chasing females, fighting other males, and other sexual behaviors. Therefore, males require food with higher energy content and higher digestible content than do females.

Dietary Protein Content Protein is the most important nutrient for animals. In ungulates, the protein content in the diet affects the growth pattern (Wood et al. 1962; Kay 1985) and seasonal fluctuations in body weight (Gates and Hudson 1981; Leader-Williams 1988) and varies among regions even within the same species (Johns et al. 1984; Mayer et al. 1984). Since the food habits of the sika deer on the Boso Peninsula differ from those in the northern deciduous broad-leaved forest zone (Asada and Ochiai 1996a), it is likely that there are regional differences in seasonal fluctuations of protein intake. According to rumen analysis of culled deer on the Boso Peninsula (Asada and Ochiai 1999), the mean crude protein content of the diet was 10.4% in summer and 8.4% in winter (Fig. 28.5). Robbins (1993) reviewed the dietary protein requirements for maintenance of adult ruminants, giving a range of 5–9%. This indicates

392

M. Asada, K. Ochiai

Fig. 28.5 Rumen nitrogen content (%, SD) of sika deer on the Boso Peninsula, central Japan (revised from Asada and Ochiai 1999). Numbers within the figure indicate sample size. The nitrogen content multiplied by 6.25 is termed the crude protein content.

Fig. 28.6 Mean nitrogen content (%, SD) of each food item of sika deer on the Boso Peninsula, central Japan (revised from Asada and Ochiai 1999): evergreen leaves; graminoids; deciduous broadleaves; n acorns. Numbers within the figure indicate sample size. The nitrogen content multiplied by 6.25 is termed the crude protein content.

that dietary protein on Boso is above the maintenance level even in winter. Evergreen broad-leaves eaten by deer had relatively high crude protein content throughout the year (Fig. 28.6). The annual mean (and SD) of the crude protein content was 13.1% (4.1%), in the range required for maximum growth (13–20%; Robbins 1993). Graminoids, mainly evergreen sedges, had a stable protein content of around 10%. The Boso sika deer consume these leaves as their primary food throughout the year (Asada and Ochiai 1996a). Therefore, we consider that sika deer on the Boso Peninsula have a stable high protein intake throughout the year. The mean fecal nitrogen (FN) content is an index of dietary protein content (Raymond 1948; Holechek et al. 1982). In summer, the FN of the Boso population was lower than that of the Iwate population (northern Honshu) in the northern deciduous broad-leaved zone (Fig. 28.7; Watanabe and Takatsuki 1993; Asada and Ochiai 1999). In winter, on the other hand, the FN in Boso was higher than that in Iwate. Thus, the magnitude of the seasonal fluctuation is smaller in the Boso population than in the

28 Sika Deer in an Evergreen Broad-Leaved Forest Zone on the Boso Peninsula

393

Fig. 28.7 Mean fecal nitrogen content (%, +/-SD) in summer (open circles) and winter (solid circles) on the Boso Peninsula (left, data from Asada and Ochiai 1999) and Iwate (right, data from Watanabe and Takatsuki 1993). Numbers within the figure indicate sample size.

northern population. In other words, the negative energy balance period is shorter in Boso. This smaller fluctuation is a result of the stable and high nitrogen content of the primary foods (evergreen leaves) of the Boso deer throughout the year.

Density-Dependent Limitation of Dietary Nitrogen FN had a negative relation with deer density in both summer and winter on the Boso Peninsula (Fig. 28.8; Asada and Ochiai 1999), indicating a density-dependent decrease of dietary nitrogen, as reported in white-tailed deer by Sams et al. (1998) and bighorn sheep by Blanchard et al. (2003). The preference ranking of a food by ungulates was correlated with its nutrient contents (Westoby 1974). As density increases, deer begin consuming less nutritious diets as a result of the reduction in high-quality forages. For example, the availability of Aucuba japonica (Japanese laurel) leaves, which have a relatively high nitrogen content (more than 2% in winter; Yamamura and Kimura 1992), decreased as the deer density increased (Asada et al. 1991). Therefore, the density-dependent decrease of FN was due to a foraging-driven decrease of plants having high nitrogen contents.

Body Size and Fat Reserves The mean body weight of sika deer on the Boso Peninsula in October was 60.4 kg in adult males (four years and older) and 39.5 kg in adult females (two years and older) (Ochiai and Asada 1995), smaller than similar weights of deer in Hokkaido (Suzuki et al. 2001) and northern Honshu (Koganezawa et al. 1986; Takatsuki 1992).

394

M. Asada, K. Ochiai 0.20 August: y=0.19-0.0017x

Fecal nitrogen content

0.19

4

10

0.18 0.17

13

0.16

14

10

0.15

5

0.14

11

10

0.13 Arcsin

N=41, R2=0.360, p<0.001

10

0.12

7

February: y=0.17-0.0014x

0.11

N=53, R2=0.505, p<0.001

0.10 0

5

10 Deer density

15

20

25

30

(individuals / km2)

Fig. 28.8 Regression of fecal nitrogen content in August (solid circles) and February (open circles) on deer density on the Boso Peninsula, central Japan (from Asada and Ochiai 1999). The nitrogen content of each fecal sample is arcsine transformed. Mean values, SD, sample size, and respective regression equations are shown.

Seasonal Fluctuation Body weight and fat reserves of adult males (two years and older) on the Boso Peninsula showed a maximum in summer and decreased until early winter (Fig. 28.9), as seen in Hokkaido (Suzuki et al. 2001; M. Yokoyama et al. 1996, 2001). It has been considered that this loss in weight arises from the decrease in foraging time due to sexual behavior in rutting season (Dauphine 1976; Mitchell et al. 1976; Leader-Williams and Ricketts 1981; Clutton-Brock et al. 1982; Bunnell and Gillingham 1985). Fat accumulation before winter and winter weight loss have been reported in many large temperate-zone ungulates (Bandy et al. 1970; Anderson et al. 1974; Mitchell et al. 1976; Franzmann et al. 1978; Moen 1978; Miura and Maruyama 1986), including sika deer in Hokkaido (M. Yokoyama et al. 1996, 2001). In contrast, the weight of adult males is constant during winter on the Boso Peninsula (Fig. 28.9). Moreover, the body weight and fat reserves of the adult females are almost constant throughout the year. Boyce (1979) and Lindstedt and Boyce (1985) suggested that seasonality appears to be a very important factor in the evolution of large body size and induces the storage of fat. That is, the longer the negative energy balance period that animals experience, the bigger their body size is. We consider that the shorter negative energy balance period in Boso due to the good food quality in winter (Asada and Ochiai 1996a, 1999) results in the small and constant body size.

28 Sika Deer in an Evergreen Broad-Leaved Forest Zone on the Boso Peninsula

Body weight (kg)

70

6 29

129

60 50

395

40

237 251 12 25

6

426 454

125

40 30 20 May 200

Aug

Oct

Jan Feb Mar

5

190 180 170 160

Riney’s kidney fat mass (g)

150 140 130

25

120 110

117

100 90 80

17

70 60

4

35

333 197 181

106

50

370

10

40 30 20 10 0 May

Aug

Oct

Jan Feb Mar

Month

Fig. 28.9 Seasonal changes of body weight and Riney’s kidney fat mass for adult male (solid circles) and adult female (open circles) sika deer culled in Boso Peninsula from 1993 to 2003. The horizontal line represents the mean, the vertical bar the 2SD range. Numbers above the range lines indicate sample sizes.

Early Growth Growth of young cervids ceases in winter when food availability decreases (Wood et al. 1962; Bandy et al. 1970; Kay 1985). Figure 28.10 compares the early growth pattern of sika deer on the Boso Peninsula with that in Hokkaido

396

M. Asada, K. Ochiai Male

120

Hokkaido 100 80 19 34

60 51

Body weight (kg)

40 35

20

35 19

3 4

8 3 13

30 22

5

2

12

Boso

2

2

0 100 Female Hokkaido

80 60 40

47

20

4

0 May

42

Oct

1139

25

4 25

10

23

29 12

Boso

2 14

Mar

0-year-old

6 2015

Oct

Oct

Mar

1-year-old

Mar

2-year-old

Oct

Mar

3-year-old

Month Fig. 28.10 Growth fluctuations in body weight of sika deer culled on the Boso Peninsula from 1992 to 1996 (solid circles) and in Hokkaido (cyclic curves cited from Suzuki et al. 2001). The solid circle represents the mean weight and the vertical bar represents the 2SD range. Numbers above the range lines indicate sample size.

(Suzuki et al. 2001). Between fall and late winter, there is no obvious weight loss in calves on the Boso Peninsula, although growth ceases between January and March. On the other hand, calves in Hokkaido lost body weight between fall and winter. This means that the deer on the Boso Peninsula can grow more steadily over a longer period than those in Hokkaido. This regional difference in seasonal growth pattern is also seen in yearlings. However, since the rate of growth is lower on the Boso Peninsula than in Hokkaido, the adult size is smaller. It is likely that this prolongation of growth period on the Boso Peninsula is due to the relatively small decrease in food availability in fall and winter owing to the supply of evergreen leaves and acorns.

28 Sika Deer in an Evergreen Broad-Leaved Forest Zone on the Boso Peninsula

397

Reproduction Conception Date Seasonal characteristics of reproduction in mammals are partly related to seasonal dietary conditions (Lincoln 1985; Sadleir 1987; Bronson 1989). Birth and lactation occur in spring in conjunction with the peak in available vegetation above 30° latitude (Bronson 1989). Since this seasonal pattern of food availability varies with latitude, the breeding season also varies with latitude in mountain sheep (Bunnell 1982), reindeer (Leader-Williams 1988), and deer of the genus Odocoileus (Bronson 1989). Offspring of sika deer on the Boso Peninsula are conceived between early September and mid-December, with a median of late September (Fig. 28.11; Asada and Ochiai 1996b). As the mean gestation length is 235 days on the Boso Peninsula (Nakajima 1929), the calving dates are estimated to be between late April and early August, with a median of mid-May (Fig. 28.12). This conception period is one month earlier than in Hokkaido (Suzuki et al. 1996), which is about 10° north of the Boso Peninsula. The breeding season is also later at more northerly latitudes in mountain sheep (Bunnell 1982) and reindeer (Leader-Williams 1988). This variation in the breeding season is most likely because of phenological differences in the vegetation. Leaves of deciduous trees on the Boso Peninsula (from early April to early May; Watanabe 1978) develop a month before those in Hokkaido (from early May to mid-May; Sasaki 1983). As in sika deer, the calving time of reindeer populations is one month later at 10° further north (Leader-Williams 1988).

Fig. 28.11 Distribution of conception date of sika deer on the Boso Peninsula, central Japan (from Asada and Ochiai 1996b). The date of conception was estimated from the culled date and the gestational age estimated from the crown-rump length.

398

M. Asada, K. Ochiai 50 40

Bud opening of deciduous trees

Leaf coloring of deciduous trees

30

Frequency

20 Calving dates

10

Hokkaido (43.58 N)

0

Bud opening of evergreen trees Bud opening of deciduous trees

60

Leaf coloring of deciduous trees

40 Calving dates

20

Boso (35.08 N) 0

E ML E ML E MLE M L E M L E M L E M L EM L E ML E ML E ML Mar.

Apr.

May

Jun.

Jul.

Aug.

Sep.

Oct.

Nov.

Dec.

Jan.

Month Fig. 28.12 Calving dates of sika deer and plant phenology in Hokkaido (upper) and Boso (lower). The histograms indicated calving dates in Hokkaido (from Suzuki et al. 1996, excluding calving by 1-yearold-females) and Boso (from Asada and Ochiai 1996b). The horizontal lines indicate plant phenology in Hokkaido (Sasaki 1983) and Boso (Watanabe 1978).

Pregnancy Rate The pregnancy rate in adult (two years and older) female sika deer depends on deer density in the previous year (p = 0.001) and in the current year (0.01 < p < 0.05; Fig. 28.13). When the deer density exceeds 15 individuals/km2, the rate of pregnancy is depressed. We consider that this density dependency of the pregnancy rate is derived from the decline in food quality in summer (Fig. 28.8). Annual variations in pregnancy rate also are related to acorn consumption by deer (Fig. 28.14). Because Quercus and Lithocarpus trees show large yearly variations in acorn production, the pregnancy rate in sika deer on the Boso Peninsula appears to be affected not only by density-dependent factors, but also by densityindependent acorn production, as in white-tailed deer in the southern Appalachians, USA (Wentworth et al. 1990, 1992).

Discussion In winter, sika deer on the Boso Peninsula consume acorns, which have a high content of fats and soluble carbohydrates, and evergreen leaves, which have high protein content. Furthermore, snow seldom falls in the Boso Peninsula, so plants and fallen

28 Sika Deer in an Evergreen Broad-Leaved Forest Zone on the Boso Peninsula

399

Fig. 28.13 Relationship between the pregnancy rate and deer density in the current (top) and the previous year (bottom) on the Boso Peninsula, central Japan. The solid and dashed lines represent regressions fit to the data and upper and lower 95% confidence intervals of slope, respectively. Pregnancy rate declined as deer density increased: r = -0.482, F1,22 = 6.648, p = 0.0171, Y = 1,476–0.015x in the current year, r = 0.627, F1,22 = 14.288, p = 0.0010, Y = 1.616–0.022x in the previous year.

acorns on the forest floor are not covered with snow. On the other hand, in northern deciduous forests, deer feed mainly on graminoids throughout the year. Moreover, heavy snow reduces food availability in winter. Therefore, the period of winter food deficit (the negative energy balance period) is shorter on the Boso Peninsula than in more northern habitats. These factors allow sika deer to reach a smaller adult size than deer in more northerly populations and to not need a winter fat reserve. In summer, unlike in winter, Boso deer eat food with lower protein content than more northerly deer. The growth rate of the Boso deer is lower than that of more northerly deer. The dietary nitrogen content in summer decreases as deer density increases. These facts show that the magnitude of the seasonal fluctuation of food availability is smaller on the Boso Peninsula. It is likely that sika deer in the evergreen broad-leaved forest zone are shaped by low seasonality and are characterized by these responses in nutritional ecology.

400

M. Asada, K. Ochiai

Fig. 28.14 Relationship between the pregnancy rate and mean percentage of acorns in rumen of sika deer on the Boso Peninsula in winter from 1993 to 2000. Samples were collected in the AmatsuKominato area from a large Lithocarpus plantation.

Literature Cited Anderson, A. E., D. E. Medin, and D. C. Bowden. 1974. Growth and morphometry of the carcass, selected bones, organs, and glands of mule deer. Wildlife Monographs 39. Asada, M. 1996. Ecological status of sika deer on the Boso Peninsula, central Japan. Ph.D. Thesis, University of Tokyo, Tokyo, Japan. (In Japanese). Asada, M., and K. Ochiai. 1996a. Food habits of sika deer on the Boso Peninsula, central Japan. Ecological Research 11:89–95. Asada, M., and K. Ochiai. 1996b. Conception dates of sika deer on the Boso Peninsula, central Japan. Mammal Study 21:153–159. Asada, M., and K. Ochiai. 1998. Sika deer management in Chiba Prefecture. Honyurui Kagaku (Mammalian Science) 38:324–331. (In Japanese.) Asada, M., and K. Ochiai. 1999. Nitrogen content in feces and the diet in sika deer on the Boso Peninsula, central Japan. Ecological Research 14:249–253. Asada, M., H. Kabaya, and I. Yamanaka. 1991. Foraging for Aucuba trees by sika deer on the Boso Peninsula. Shinrin Boeki (Forest Pests) 40:206–210. (In Japanese.) Bandy, P. J., I. McT. Cowan, and A. J. Wood. 1970. Comparative growth in four races of blacktailed deer (Odocoileus hemionus). Part 1. Growth in body weight. Canadian Journal of Zoology 48:1401–1410. Beier, P. 1987. Sex differences in quality of white-tailed deer diets. Journal of Mammalogy 68:323–329. Bell, R. 1971. A grazing ecosystem in the Serengeti. Scientific American 225:86–93. Blanchard, P., M. Festa-Bianchet, J. Gaillard, and J. T. Jorgenson. 2003. A test of long-term fecal nitrogen monitoring to evaluate nutritional status in bighorn sheep. Journal of Wildlife Management 67:477–484. Boyce, M. S. 1979. Seasonality and patterns of natural selection for life histories. American Naturalist 114:569–583.

28 Sika Deer in an Evergreen Broad-Leaved Forest Zone on the Boso Peninsula

401

Bronson, F. H. 1989. Mammalian reproductive biology. University of Chicago Press, Chicago, Illinois, USA. Bunnell, F. L. 1982. The lambing period of mountain sheep: Synthesis, hypotheses and tests. Canadian Journal of Zoology 60:1–14. Bunnell, F. L., and M. P. Gillingham. 1985. Foraging behavior: Dynamics of dining out. Pages 53–79 in R. J. Hudson and R. G. White, editors, Bioenergetics of wild herbivores, CRC Press, Boca Raton, Florida, USA. Chiba Prefecture. 1987. Science report on the status of sika deer in Chiba prefecture. Chiba, Japan. (In Japanese.) Chiba Prefecture and Boso Deer Research Group. 2002. Science report on the status of sika deer in Chiba prefecture. Vol. 10. Chiba, Japan. (In Japanese.) Clutton-Brock, T. H., and P. H. Harvey. 1983. The functional significance of variation in body size among mammals. Special Publication of the American Society of Mammalogists 7:632–663. Clutton-Brock, T. H., F. E. Guiness, and S. D. Albon. 1982. Red deer: Behavior and ecology of two sexes. University of Chicago Press, Chicago, Illinois, USA. Dauphine, T. C. 1976. Biology of the Kaminuriak population of barren-ground caribou. Canadian Wildlife Service Report Series 38:1–69. Franzmann, A. W., R. E. LeResche, R. S. Rausch, and J. L. Oldmeyer. 1978. Alaskan moose measurements and weights and measurement-weight relationships. Canadian Journal of Zoology 56:298–306. Furubayashi, K., and N. Maruyama. 1977. Food habits of sika in Fudakake, Tanzawa Mountains. Honyudobutugaku-Zassi (Journal of the Mammalogical Society of Japan) 7:55–62. (In Japanese with English abstract.) Furubayashi, K., and Y. Sinoda. 2001. Distribution of sika deer (Cervus nippon) around Edo in Edo era. Wildlife Conservation Japan 7:1–24. (In Japanese.) Gates, C. C., and R. J. Hudson. 1981. Weight dynamics of wapiti in the boreal forest. Acta Theriologica 26, 27:407–418. Geist, V. 1974. On the relationship of social evolution and ecology in ungulates. American Zoology 14:205–220. Geist, V. 1986. On speciation in Ice Age mammals, with special reference to cervids and caprids. Canadian Journal of Zoology 65:1067–1084. Harlow, R. F., J. B. Whelan, H. S. Crawford, and J. E. Skeen. 1975. Deer foods during years of oak mast abundance and scarcity. Journal of Wildlife Management 39:330–336. Hofmann, R. R. 1985. Digestive physiology of the deer. Pages 393–407 in P. F. Fennessy and K. R. Drew, editors, Biology of deer production. Royal Society of New Zealand Bulletin 22, Wellington, New Zealand. Hofmann, R. R. 1989. Evolutionary steps of ecophysiological adaptation and diversification of ruminants: A comparative view of their digestive systems. Oecologia 78:443–457. Holechek, J. L., M. Vaura, and R. D. Peiper. 1982. Method for determining the nutritive quality of range ruminant diets: A review. Journal of Animal Science 54:363–376. Iimura, T., and Chiba Prefecture. 1981. The population of sika deer on East Boso Hill and its management. Chiba, Japan. (In Japanese.) Jarman, P. J. 1974. The social organization of antelope in relation to their ecology. Behaviour 48:215–266. Johns, P. E., M. H. Smith. and R. K. Chesser. 1984. Annual cycles of the kidney fat index in a southeastern white-tailed deer herd. Journal of Wildlife Management 48:969–973. Kay, R. N. B. 1985. Body size, patterns of growth, and efficiency of production in red deer. Pages 411–422 in P. F. Fennessy and K. R. Drew, editors, Biology of deer production. Royal Society of New Zealand Bulletin 22, Wellington, New Zealand. Koganezawa, M., N. Katai, and N. Maruyama. 1976. Distribution of sika deer on eastern Boso Hill. Nihonzaru (Japanese Macaques) 2:115–121. (In Japanese.) Koganezawa, M., T. Inui, and M. Kitahara. 1986. Body weight and external carcass measurements of sika deer (Cervus nippon TEMMINCK) in Nikko - Ashio Mountains, Tochigi Prefecture, Japan. Memoirs of the Tochigi Prefectural Museum 4:29–53. (In Japanese with English abstract.)

402

M. Asada, K. Ochiai

Leader-Williams, N. 1988. Reindeer on South Georgia. Cambridge University Press, New York, New York, USA. Leader-Williams, N., and C. Ricketts. 1981. Seasonal and sexual patterns of growth and condition in introduced reindeer on South Georgia. Oikos 38:27–39. Lincoln, G. A. 1985. Seasonal breeding in deer. Pages 165–179 in P. F. Fennessy and K. R. Drew, editors, Biology of deer production. Royal Society of New Zealand Bulletin 22, Wellington, New Zealand. Lindstedt, S. L., and M. S. Boyce. 1985. Seasonality, fasting endurance, and body size in mammals. American Naturalist 125:873–878. Maruyama, N., and K. Furubayashi. 1983. Preliminary examination of block count method for estimation numbers of sika deer in Fudakake. Journal of the Mammalogical Society of Japan 9:274–278. Maruyama, N., Y. Totake, and N. Katai. 1975. Seasonal change of food habits of the sika deer in Omote-Nikko. Honyudobutugaku-Zassi (Journal of the Mammalogical Society of Japan) 6:163–173. (In Japanese with English abstract.) Mayer M. W., R. D. Brown, and M. W. Graham. 1984. Protein and energy content of white-tailed deer diets in the Texas coastal bend. Journal of Wildlife Management 48:527–534. McCullough, D. R., and D. E. Ullrey. 1985. Chemical composition and gross energy of deer forage plants on the George Reserve, Michigan. Michigan State University Agricultural Experiment Station East Lansing Research Report 465. Mitchell, B., F. McCowan, and I. S. Nicholson. 1976. Annual cycles of body weight and condition in Scottish red deer, Cervus elaphus. Journal of Zoology, London 180:107–127. Miura, S., and N. Maruyama. 1986. Winter weight loss in Japanese serow. Journal of Wildlife Management 50:336–338. Moen, A. N. 1973. Wildlife ecology. W. H. Freeman, San Francisco, California, USA. Moen, A. N. 1978. Seasonal change in heart rates, activity, metabolism, and forage intake of white-tailed deer. Journal of Wildlife Management 42:715–738. Nakajima, M. 1929. Experimental report of penned sika deer at the University Forest, Chiba. Miscellaneous Information, the Tokyo University Forests 8:95–114. (In Japanese.) Nakama, S., N. Maruyama, S. Hanawa, and O. Moril. 1980. On estimating the serow number by new visual methods in Wakinosawa Village, Aomori Prefecture. Journal of the Mammalogical Society of Japan 8:59–69. (In Japanese with English abstract.) Ochiai, K. 1995. Status of sika deer on the Boso forests. Pages 138–146 in M. Osawa and S. Ohara, editors, Science of bio-global environments, Asakura-shoten, Tokyo, Japan. (In Japanese.) Ochiai, K., and M. Asada. 1995. Growth in the body size of sika deer (Cervus nippon) on the Boso Peninsula, central Japan. Journal of the Natural History Museum and Institute, Chiba 3:223– 232. (In Japanese with English summary.) Ochiai, K., and M. Asada. 1997. Growth pattern of the skull and limb bone size of sika deer, Cervus nippon, on the Boso Peninsula, central Japan. Journal of the Natural History Museum and Institute, Chiba 4:159–172. (In Japanese with English summary.) Ohtaishi, N. 1986. Preliminary memorandum of classification, distribution and geographic variation on sika deer. Honyurui Kagaku (Mammalian Science) 53:13–17. (In Japanese.) Padmalal U. K. G. K., and S. Takatsuki. 1994. Age-sex differences in diets of sika deer on Kinkazan Island, northern Japan. Ecological Research 9:251–256. Putman, R. J., S. Culpin, and S. J. Thirgood. 1993. Dietary differences between male and female fallow deer in sympatry and in allopatry. Journal of Zoology, London 229:267–275. Raymond, W. F. 1948. Evaluation of herbage for grazing. Nature 161:937–938. Robbins, C. T. 1993. Wildlife feeding and nutrition. Academic Press, San Diego, California, USA. Sadleir, R. M. F. S. 1987. Reproduction of female cervids. Pages 23–144 in C. M. Wemmer, editor, Biology and management of the Cervidae. Smithsonian Institution Press, Washington, DC, USA. Sams, M. G., R. L. Lochmiller, C. W. Qualls, Jr., and D. M. Leslie, Jr. 1998. Sensitivity of condition indices to changing density in a white-tailed deer population. Journal of Wildlife Diseases 34:110–125.

28 Sika Deer in an Evergreen Broad-Leaved Forest Zone on the Boso Peninsula

403

Sasaki, C. 1983. Phenology of woody plants and temperatures in central Hokkaido. Review of Forest Culture 4:77–86. (In Japanese.) Staines, B., J. M. Crisp, and T. Parish. 1982. Differences in quality of food eaten by red deer (Cervus elaphus) stags and hinds in winter. Journal of Applied Ecology 19:65–77. Stearns, S. C., and J. C. Koella. 1986. The evolution of phenotypic plasticity in life-history traits: prediction of reaction norms of age and size at maturity. Evolution 40:893–913. Suzuki, M., K. Kaji, M. Yamanaka, and N. Ohtaishi. 1996. Gestation age determination, variation of conception date, and external fetal development of sika deer (Cervus nippon yesoensis Heude, 1884) in eastern Hokkaido. Journal of Veterinary Medical Science 58:505–509. Suzuki, M., M. Onuma, M. Yokoyama, K. Kaji, M. Yamanaka, and N. Ohtaishi. 2001. Body size, sexual dimorphism, and seasonal mass fluctuations in a larger sika deer subspecies, the Hokkaido sika deer (Cervus nippon yesoensis Heude, 1884). Canadian Journal of Zoology 79:154–159. Takatsuki, S. 1980. Food habits of sika deer on Kinkazan Island. Science Report of Tohoku University, Series IV (Biology) 38:7–31. Takatsuki, S. 1983. The importance of Sasa nipponica as a forage for sika deer (Cervus nippon) in Omote-Nikko. Japan. Journal of Ecology 33:17–25. Takatsuki, S. 1986. Food habits of sika deer on Mt. Goyo, northern Honshu. Ecological Research 1:119–128. Takatsuki, S. 1988. Rumen contents of sika deer on Tsushima Island, western Japan. Ecological Research 3:181–183. Takatsuki, S. 1990. Summer dietary composition of sika deer on Yakushima Island, southern Japan. Ecological Research 5:253–260. Takatsuki, S. 1992. Research report of sika deer on Mt. Goyo (1988–1991). Iwate Prefecture. (In Japanese.) Takatsuki, S., and S. Ikeda. 1993. Botanical and chemical composition of rumen contents of sika deer on Mt. Goyo, northern Japan. Ecological Research 8:57–64. University of Tokyo. 1988. An outline of the university forest in Chiba 1988. Chiba, Japan. (In Japanese.) Warren, R. J., and L. J. Krysl. 1983. White-tailed deer food habits and nutritional status as affected by grazing and deer-harvest management. Journal of Range Management 36:104–109. Watanabe, R. 1978. Seasonal division based on the phenological records in two different climatical regions of Japan. Bulletin of Institute of Nature Education in Shiga Heights, Shinshu University 17:19–32. Watanabe, T., and S. Takatsuki. 1993. Comparison of nitrogen and fiber concentrations in rumen and fecal contents of sika deer. Journal of the Mammalogical Society of Japan 18:43–48. Wentworth, J. M., A. S. Johnson, and P. E. Hale. 1990. Influence of acorn use on nutritional status and reproduction of deer in the Southern Appalachians. Proceedings of Annual Conference of Southeastern Association of Fish and Wildlife Agencies 44:142–154. Wentworth, J. M., A. S. Johnson, P. E. Hale, and K. E. Kammermeyer. 1992. Relationships of acorn abundance and deer herd characteristics in the Southern Appalachians. Southern Journal of Applied Forestry 16:5–8. Westoby, M. 1974. An analysis of diet selection by large generalist herbivores. American Naturalist 108:290–304. Whitehead, G. K. 1993. The Whitehead encyclopedia of deer. Swan Hill Press, Shrewsbury, United Kingdom. Wood, A. J., I. McT. Cowan, and H. C. Nordan. 1962. Periodicity of growth in ungulates as shown by deer of the genus Odocoileus. Canadian Journal of Zoology 40:593–603. Yamamura, Y., and M. Kimura. 1992. Matter-economical roles of evergreen leaves in Aucuba japonica, an understory shrub in the warm-temperate region of Japan 2. Dynamics and budgets of nutrients. Botanical Magazine, Tokyo 105:95–104. Yokoyama, M., N. Maruyama, K. Kaji, and M. Suzuki. 1996. Seasonal changes of body fat reserves in sika deer of east Hokkaido, Japan. Journal of Wildlife Research 1:57–61.

404

M. Asada, K. Ochiai

Yokoyama, M., K. Kaji, and M. Suzuki. 2000. Food habits of sika deer and nutritional value of sika deer diets in eastern Hokkaido, Japan. Ecological Research 15:345–355. Yokoyama, M., M. Onuma, M. Suzuki, and K. Kaji. 2001. Seasonal fluctuations of body condition in northern sika deer on Hokkaido Island, Japan. Acta Theriologica 46:419–428. Yokoyama, S., T. Koizumi, and E. Shibata. 1996. Food habits of sika deer as assessed by fecal analysis in Mt. Ohdaigahara, central Japan. Journal of Forest Research 1:161–164.

Chapter 29

Sika Deer Population Irruptions and Their Management on Hokkaido Island, Japan Hiroyuki Uno, Koichi Kaji, and Katsumi Tamada

Abstract Sika deer populations on Hokkaido Island, northern Japan, have often showed irruptive behavior in the recent past. During the last two decades the irruption of sika deer has caused extensive agricultural damage and severely affected forest vegetation. In response, in 1998 the Hokkaido government initiated aggressive population control based on the “Conservation and management plan in eastern Hokkaido.” Since institution of this plan both the sika deer population size indices and the amount of damage have clearly decreased. We consider that decreasing the population size, extending use of net fences, and applying chemical repellents are main contributors to reducing damage. We recognize that there are many uncertainties with estimating the population parameters and population size and that uncertain information must be replaced with more accurate data. Because the current deer density in eastern Hokkaido remains extremely high, we must continue to control and manipulate sika deer density. Long-term monitoring and evaluation of population parameters are important. More information is needed as well to evaluate the effects of sika deer population levels on biodiversity.

History of Sika Deer on Hokkaido Island Inukai (1952) reported about the status of sika deer (Cervus nippon yesoensis) on Hokkaido Island in the Edo era (1603–1867) as follows: “There were large herds of sika deer throughout Hokkaido and they migrated seasonally between the west, where snows were heavy, and the southeast, where snows were less. The eastern part of Hokkaido, for example Shiranuka Hills, was a good hunting area for native Ainu people. In the 1780s, the sika deer population dramatically decreased because of severe winter conditions and consequently 300–400 Ainu people died from starvation.” Thus, Inukai thought the deer population size had occasionally fluctuated in numbers during the Edo era. Prior to the Japanese colonization of Hokkaido Island in the late nineteenth century sika deer coexisted with natural predators such as wolves (Canis lupus) and brown bears (Ursus arctos) as well as with the indigenous Ainu people (Inukai 1952). D. R. McCullough et al. (eds.), Sika Deer: Biology and Management of Native and Introduced Populations, DOI: 10.1007/978-4-431-09429-6_29, © Springer 2009

405

406

H. Uno et al.

Overexploitation of sika deer for skins, meat, and antlers, along with periodic severe winter conditions, threatened the sika deer population during the 1870s and 1880s (Fig. 29.1). Between 1873 and 1875 more than 100,000 deer were harvested and more than 50,000 skins were exported each year (Hokkaido Government 1987). Heavy snow falls in 1879 and 1903 led to high winter mortality. Thereafter, until 1925, sika deer were restricted in distribution to several small districts in eastern Hokkaido (Fig. 29.2). Wolves were thought to be extinct by 1890, killed to reduce depredation on livestock (Inukai 1952). To conserve the deer population, hunting was completely banned between 1889 and 1899 and again between 1920 and 1952. The population of deer rebuilt rapidly because of the extinction of the wolf, the two bans on hunting and bucks-only hunting restrictions between 1953 and 1993. Thanks to these measures, the sika deer numbers and their distributional range gradually recovered (Koizumi et al. 1984), having survived the population bottleneck of about 120 years ago.

Distribution and Current Status In 1974 the sika deer distribution was concentrated in the eastern half of the island and expanded westward thereafter (Fig. 29.2) (Kaji et al. 2000). No deer were noted on the southern part of Oshima Peninsula in the 1970s, but were reported there in the 1990s following their reintroduction. Kaneko et al. (1998) analyzed the habitat of sika deer using GIS (geographic information system) and reported that coniferous plantations and pasture lands have increased in the eastern part of Hokkaido during the last three decades. Therefore, biomass of forage plants (e.g., Sasa bamboo and other graminoids) has increased and the vegetative cover of coniferous forests has enlarged. Kaji et al. (2000) considered that elimination of wolves, replacement of native mixed hardwood forests with conifer plantations, and increased pasture acreage likely contributed to the expansion in deer distribution. Sika deer occupied their entire potential habitat by the 1990s (Kaji et al. 2000). Currently there are at least three populations of sika deer on Hokkaido Island— Akan, Hidaka, and Taisetsu—that have different haplotype compositions (Nagata et al. 1998; see Fig. 29.2). The relative density of sika deer is indicated by the number of deer harvested by hunting. For recording of hunting statistics Hokkaido is divided into 3,599 blocks with each block about 5 × 4.6 km in size (Fig. 29.3). As shown by kill statistics the core areas of sika deer distribution for the Akan population are in the Shiranuka Hills (SH) and Akan National Park (ANP), in southeastern Hokkaido (Fig. 29.3).

Population Size Indices In order to obtain additional population measures independent of hunter harvest we determined relative population size indices based on the results of the following surveys: spotlight census, aerial survey, catch per unit effort (CPUE), sighting per unit effort

29 Sika Deer Population Irruptions and Their Management on Hokkaido Island

407

Fig. 29.1 Changes in sika deer harvest and loss incurred. The bar indicates the number of deer harvested and the line indicates the cost of agricultural and forest damage (Data from Hokkaido Government, unpublished).

Fig. 29.2 Sika deer range expansion from 1925 to 1991 on Hokkaido Island, estimated from observations reported in personal interviews and mail surveys (Kaji et al. 2000).

408

H. Uno et al.

Males

Females

Akan National Park (ANP) Shiranuka Hills (SH)

Fig. 29.3 The number of deer harvested by hunting in 1999. Each block is 5 × 4.6 km (Hokkaido Government 2002).

(SPUE), and damage to agriculture and forestry. Population surveys were conducted in various seasons from October to March: spotlight census in late October and aerial survey in February-March. We considered the population indices obtained from April in year t to March in year t + 1 as the index in year t. Using the relative population indices, we compared observations in year t to those in 1993 because we estimated the absolute population of sika deer in eastern Hokkaido to be 120,000 ± 46,000 deer as of March 1994 (Hokkaido Institute of Environmental Sciences 1995). We conducted pre-hunting season spotlight counts of deer from vehicles annually from 1992 to 2000. In 1992, we established 61 survey routes close to farmlands and rivers throughout the eastern part of Hokkaido (Kaji and Tomizawa 1993). We calculated the population index in eastern Hokkaido (Akan population) using the results of these standard routes. Aerial surveys, CPUE, and SPUE were also sampled from the eastern part of Hokkaido. We began the spotlight counts in the western part of Hokkaido in 1993. Numbers of observed deer in these fixed routes were stable for repeated observations (coefficient of variance was less than 25% in each year) (Kaji and Tomizawa 1993). We calculated the number of deer observed per 10 km in each survey route (Fig. 29.9). From the ratio estimator method (Cochran 1977), we obtained the average, bias, and variance and estimated the 95% confidence interval of the population size index (Matsuda et al. 2002). In 2002, we conducted the census on 143 survey routes throughout Hokkaido Island. The 143 routes included the 61 routes previously used in eastern Hokkaido. Night spotlight routes on these 143 survey routes, divided into 12 management units, were used to index deer population size. The data were expressed as the average number of deer observed per 10 km by spotlight count (Tamada et al. unpublished data). By this index the observed numbers of deer in Units 10 and 11 increased from 1992 to 1997 (Fig. 29.4). The CPUE and SPUE independently

29 Sika Deer Population Irruptions and Their Management on Hokkaido Island

409

estimated from questionnaire surveys are shown in Fig. 29.5 (see below for further details). The CPUE and SPUE indices for the Akan population (Units 9, 10, 11 and 12) increased from 1990 to 1997 (Kaji et al. unpublished data). We conducted aerial surveys in February–March of 1993, 1994, and 1997 through 2001. A helicopter (Aerospacial AS-350B) was used for aerial counts in ANP and SH. We divided the study area (about 224 km2) into 24 survey units and estimated the average density per unit as the relative index (Hokkaido Institute of Environmental Sciences 1995; Uno et al. 2006). We collected records of hunting activities (i.e., numbers of deer killed or observed) from questionnaire surveys to hunters and used these data to estimate the CPUE (number of deer harvested per hunter-day) and SPUE (number of deer sighted per hunter-day). Because the quota for deer harvested per day was doubled in the fiscal year 1998 hunting season, the CPUE index in 1998 was approximately twice as high as that in 1997 (Fig. 29.10). We therefore excluded the fiscal year 1998 and 1999 CPUE indices from the analysis. Value of damage caused by deer to agriculture and forestry was obtained from Hokkaido government statistics (Hokkaido Government 2000). All independent measures (harvest statistics, spotlight counts, and hunter questionnaires) showed that the Akan population irrupted and caused severe damage to agriculture and forestry over the last twenty years (Figs. 29.1, 29.4, and 29.5). In response, a ten-day hunt of female sika deer was begun in restricted areas (the ten municipalities around ANP) from 1994. Despite the female removal, the deer population size continued to increase until 1997. The amount of damage by deer to agriculture and forestry in 1996 was estimated to be greater than 5 billion Japanese yen. The irruption of sika deer extremely affected both the over-story and under-story of forest vegetation. Many larger deciduous trees died from bark stripping by sika deer (Fig. 29.6), and most of the small-sized trees (dbh less than 10 cm) have disappeared from the forest because of deer damage and suppression of tree reproduction (Uno et al. 1995). Miyaki and Nishikawa (2001) also reported the disappearance of small-sized trees in ANP and SH. Hardwood seedlings taller than 20 cm have been damaged by deer browsing (Terasawa and Akashi 2001). The biomass of under-story vegetation has similarly been reduced. Species of high palatability to deer such as Sasa kurilensis (dwarf bamboo) have decreased due to deer foraging in ANP (Uno 2000; Miyaki and Nishikawa 2001), and Sasa nipponica (dwarf bamboo) also decreased in SH (Terasawa and Akashi 2001). Other sika deer populations on Hokkaido Island are also irruptive. The population on Nakanoshima Island (Kaji et al. 1988) and that on Cape Shiretoko (Kaji et al. 2004) have undergone irruption and crashes. McCullough (1997) considered that any number of factors, such as increasing the carrying capacity and reducing mortality, can contribute to irruptive behavior in ungulates. Elimination of wolves, reduction in hunting pressure, and increase in the carrying capacity through habitat alteration are all considered to have contributed to irruptive behavior of sika deer in eastern Hokkaido.

1992

1992

1992

1992

1994

1994

1994

1994

1996

Unit 1

1996

Unit 2

1996

Unit 3

1996

Unit 7

1998

1998

1998

1998

2000

2000

2000

2000

2002

2002

2002

2002

160 140 120 100 80 60 40 20 0

160 140 120 100 80 60 40 20 0

1992

1992

1994

1994

1996

Unit 6

1996

Unit 8

1998

1998

2000

2000

2002

2002

160 140 120 100 80 60 40 20 0

160 140 120 100 80 60 40 20 0

1992

1992

1994

1994

1996

Unit 4

1996

Unit 9

1998

1998

2000

2000

2002

2002

160 140 120 100 80 60 40 20 0

160 140 120 100 80 60 40 20 0

160 140 120 100 80 60 40 20 0

160 140 120 100 80 60 40 20 0

1992

1992

1992

1992

1994

1994

1994

1994

1996

Unit 5

1996

Unit 10

1996

Unit 11

1996

Unit 12

1998

1998

1998

1998

2000

2000

2000

2000

2002

2002

2002

2002

Fig. 29.4 Changes in number of deer observed by spotlight count per 10 km in 12 management units. Solid bar indicates average value and error bar indicates SE (Tamada et al. unpublished data).

160 140 120 100 80 60 40 20 0

160 140 120 100 80 60 40 20 0

160 140 120 100 80 60 40 20 0

160 140 120 100 80 60 40 20 0

410 H. Uno et al.

0.8

6

1.0

0.6

2

4

0

Unit 7 8

10

0.6

0.8

1.0

6

0.2

0

2

Unit 8

1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001

Unit 6

8

10

0.6

0.8

1.0

6

8

10

0.8

1.0

CPUE 0.0

0.2

0

2

1.0

0.8

0.6

0.4

0.0

4

SPUE

CPUE 0.2

0.4

10

8

6

4

2

SPUE

0.0

1.0

0.8

0.6

0.4

1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 0

4

CPUE

CPUE 0.2

0.4

SPUE

SPUE

SPUE

0.4

0.0 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001

10

8

6

4

2

0

10

8

6

4

2

0

0.2

Unit 3

1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001

1.0

2000 2001

0.8

1997 1998 1999

0.4

Unit 2

1994 1995 1996

CPUE

0.6

0.0

1991 1992 1993

0.2

1.0

0.8

0.6

0.4

0.2

0.0 1990

0.0

1990

1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001

1991 1992 1993

Unit 9

Unit 4

1994 1995 1996 1997 1998 1999 2000 2001

10

8

6

4

0

2

SPUE

SPUE

CPUE

CPUE

CPUE 0.6

4

6

2

0.4

0

1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001

Unit 12 10

8

12

16

8

0

4

10

SPUE

0.2

Unit 11

1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001

0.0

1.0

0.8

0.6

0.4

Unit 10 8

6

4

2

SPUE

0.2

0.0

1.0

1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 0

10

8

6

4

SPUE

29 Sika Deer Population Irruptions and Their Management on Hokkaido Island

2

0

SPUE

CPUE 0.8

1999 2000 2001

0.6

Unit 5

1993 1994 1995 1996 1997 1998

0.4

0.0

1990 1991 1992

0.2

1.0

0.8

0.6

0.4

0.2

0.0

Fig. 29.5 Changes in number of deer harvested per hunter-day (CPUE) and sighted per hunter-day (SPUE). Bold line indicates CPUE and broken line indicates SPUE (Kaji et al. unpublished data).

CPUE

CPUE

411

412

H. Uno et al.

Fig. 29.6 Bark stripping by sika deer in Akan National Park, Hokkaido (Photo by H. Uno).

Life History Female deer on Hokkaido usually give birth to a single offspring in June (Fig. 29.7). In examination of fetuses twins were rarely observed (Suzuki and Ohtaishi 1993). Suzuki et al. (1996) estimated conception dates that ranged from 7 October to 17 January, with most concentrated between mid-October and early November. Many females attain sexual maturation at yearling age; the pregnancy rate of females surveyed in Ashoro town in 1991 was 83.3% for yearlings and 100% for ≤2-year-old females (Suzuki et al. 1993). The pregnancy rates of yearlings and ≤2 years old were 92.9% and 98.4%, respectively in Akan town, 1998/1999 (Uno and Tamada 2001). These results show that sika deer have very high fertility on Hokkaido. Deer mortality is concentrated in three annual periods. Deer are killed by sport hunting between November and January, die of natural mortality on wintering areas mainly during February and April, and are killed in nuisance control efforts (culling) mainly between March and October (Fig. 29.7). Nuisance culling is done mainly in farmland and pasture land to reduce crop damage. Seasonally, female sika deer in the Akan population migrate from the summer range to the winter range in November to January, and return mainly in May (Fig. 29.7) (Uno and Kaji 2000; Sakuragi et al. 2003). Age-specific mortality rates in winter were high for fawns and six-year-old and older females in ANP (Uno et al. 1998). Male mortality was unknown because of sexual segregation in the wintering area (Hokkaido Institute of Environmental Sciences 1995). We consider that long snow duration and decrease availability of food resources due to impacts on plants by deer feeding

29 Sika Deer Population Irruptions and Their Management on Hokkaido Island

413

Fig. 29.7 Timetable for the life history of sika deer on Hokkaido.

were important factors of winter mortality (Takatsuki chapter 16). Harvest mortality of adult females was higher than natural mortality except in years with more severe winters (Uno and Kaji 2006).

Population Management Management Plan In 1998, the Hokkaido government implemented the “Conservation and management plan for sika deer in eastern Hokkaido” (CMPS) and started aggressive population control based on feedback management (Hokkaido Government 1998; Matsuda et al. 1999). The management goal is to maintain moderate population levels to avoid population irruptive behavior, limit damage to crops and forests, and sustain a moderate yield without endangering the population. We considered three levels of relative population sizes and four levels of hunting pressure, and chose one of four actions, based on the estimates of relative population size (Fig. 29.8 and Table 29.1). If the recent deer population index (%P(t)) is above the irruption level (%P+), we apply “emergency decrease action” (EDA); i.e., maximize the harvest of female

414

H. Uno et al.

Fig. 29.8 Concept diagram for population management in eastern Hokkiado; %P(t) = population size index observed, %P + = the irruption level, %P* = the optimal level, %P- = the critical level, EDA = emergency decrease action, GDA = gradual decrease action, GIA = gradual increase action, and BH = ban on hunting (Hokkaido Government 1998).

Table 29.1 Four management programs based on relative population size index. Program Population size index Action Emergency decrease action (EDA) %P + < %P(t) Maximum harvest of females Gradual decrease action (GDA) %P* < %P(t) < %P+ Hunting for males and females Gradual increase action (GIA) %P– < %P(t) < %P* Males-only hunting Ban on hunting (BH) %P(t) < %P– Ban on hunting %P(t) denotes population size index observed, %P+ denotes the irruption level, %P* denotes the optimal level and %P– denotes the critical level (Matsuda et al. 1999).

deer by hunting and control kill. If %P(t) is larger than the optimal level (%P*) but smaller than %P+, we apply “gradual decrease action” (GDA) by hunting for males and females. If %P(t) is larger than the critical level (%P−) but smaller than the optimal level (%P*) we apply “gradual increase action” (GIA), by males-only hunting. If %P(t) is smaller than %P−, we institute a total ban on hunting (BH). Sex-specific hunting is considered to be effective for population management (Matsuda et al. 1999).

Population Trend Population size indices continued to increase until 1996 (Fig. 29.10). The Hokkaido government has initiated EDA from the fiscal year 1998 based on the CMPS. Under this action, 37,800 female deer and 34,700 male deer were harvested or culled in fiscal year 1998 (Fig. 29.9). The number of females harvested by hunting in 1998 increased four times over 1997. We consider that doubling the hunting quota and extending the hunting periods for females increased female harvest

29 Sika Deer Population Irruptions and Their Management on Hokkaido Island

415

Fig. 29.9 Changes in observed number of deer per 10 km in autumn (solid line) and number of deer harvested from 1992 to 2001 in eastern Hokkaido. The vertical solid bar is the male harvest and the open bar is the female harvest. Error bars indicate SE.

200

262

175

Popolation indices

150 125 100 75

spotlight census

50

aerial survey

25

catch per unit effort observation frequency by hunters damage to agriculture and forestry

0 1989

1991

1993

1995

1997

1999

2001

Fig. 29.10 Changes in population size indices of sika deer in eastern Hokkaido (Matsuda et al. 2002).

effectively. In fiscal year 1999, 31,550 females and 25,950 males were harvested or killed by hunting and nuisance culling. Thereafter population size indices decreased (Fig. 29.10). The relative size indices, estimated by spotlight census, in 1996 and 2000 compared to 1993 were 123% (95% CI = 97–150%) and 79% (95% CI = 59–100%), respectively. Hunters cooperated with government to control the population.

416

H. Uno et al.

Reducing Crop Damage The value of damage in 2002 was estimated to be 2,952 million Japanese yen. Damage to pasture, sugar beets, wheat, potatoes, rice, and maize represented 50%, 12%, 7%, 6%, 6% and 3% of total damage, respectively (Hokkaido Government unpublished data). To reduce crop damage, 2.5 m high net fences have been erected to exclude sika deer from agricultural fields. Fences extending for about 2,200 km along the edge of farmlands in eastern Hokkaido were built from 1995 to 2000 (Hokkaido Government 2001). Net fences are efficient, but set-up and maintenance costs are very high. Electric fences are normally cheaper than net fences, and electric fences about 1.5 m high have been used to exclude sika deer from relatively smaller enclosures. In addition, chemical repellents such as Ziram (dimethyldithiocarbamate) have been sprayed to reduce deer browsing in conifer (Abies sachalinensis and Larix kaempferi) plantations. The value of damage to agriculture and forestry decreased from 1996 to 2000 (Fig. 29.1). We believe that the combination of reducing the population size, erecting net fences, and using chemical repellents were the main contributors to reduced damages.

Other Management Problems Underestimating the absolute population size. The absolute deer population was estimated to be between 74,000 and 166,000 individuals (90% CI) in the fiscal year 1993 based on aerial surveys (Hokkaido Institute of Environmental Sciences 1995). This population size was incompatible with the known statistics on hunter kill, given the life-history parameters of sika deer. Because of the low visibility of the deer in the dense forests of Hokkaido, the absolute population size was underestimated. In 2000 the Hokkaido government announced that the population size in 1993 had been underestimated and extended the hunting season and expanded the hunting areas to increase harvest in the winter of 2000 (Hokkaido Government 2000). We recognize there is uncertainty, and uncertain information must be replaced with more accurate information. Therefore, we applied a population-dynamics model based on population trend and harvest data (Matsuda et al. 2002) and assumed that ranges of biological parameter values were consistent with the 16–21% rate of natural increase per year (Kaji et al. 2004). Using this approach we re-estimated the population size in 1993 to be between 170,000 and 330,000 (Matsuda et al. 2002). We continue to work to improve our estimates in order to manage sika deer more precisely. Lead poisoning in sea eagles. A serious side effect of hunting on sika deer is the inadvertent lead poisoning of large birds of prey. Steller’s sea eagles (Haliaeetus pelagicus) and most white-tailed sea eagles (Haliaeetus albicilla) migrate seasonally between Hokkaido Island and the Kamchatka Peninsula and Sakhalin Island, Russia (Working Group of White-tailed Eagles and Steller’s Sea Eagles 1996;

29 Sika Deer Population Irruptions and Their Management on Hokkaido Island

417

McGrady et al. 2000). Both eagles are listed as endangered species. In the winter of 1996–1997 18 eagles (15 Steller’s sea eagles and three white-tailed sea eagles), and in 1997–1998 26 eagles (16 Steller’s and ten white-tailed) were found dead (Hokkaido Government unpublished data). They died from ingesting lead fragments of rifle bullets (Kurosawa 2000). In response, the Hokkaido government prohibited hunters from using the lead bullets in rifles in the 2000–2001 season and in shotguns in 2001–2002. Still 11 sea eagles died from lead poisoning in the winter of 2001–2002. This problem remains unsolved.

Moving from Population Management to Ecosystem Management We recognize there are a lot of uncertainties with estimating the population parameters and population size and we have worked to improve our methods. Monitoring sika deer population dynamics and population parameters over the long term is a very important first step in wise sika deer population management. Nevertheless, population management of sika deer at the species level is not sufficient to address all of the ramifications of sika deer in the Hokkaido landscape. Ecosystem management at a larger scale is required (Cederlund et al. 1998). These authors noted that more information was needed to evaluate the full effects of roe deer (Capreolus capreolus) on biodiversity conservation. Similarly, Waller and Alverson (1997) pointed out that white-tailed deer (Odocoileus virginianus) in the United States had reached sufficient densities to cause manifold and substantial ecological impacts on forest diversity and proposed that monitoring efforts should focus on the impacts of such keystone species on other elements of the ecosystem. DeCalesta and Stout (1997) proposed “the relative deer density displayed on McCullough’s (1984) recruitment curve provided managers with broaden approach to ecosystem management.” Kaji (2001) considered the current relative density of sika deer in relation to the density appropriate for sustained timber yield and agricultural crop production. Because the current deer density in eastern Hokkaido is much higher than the relative density associated with sustaining biological diversity, we have to continue to reduce sika deer density. Monitoring of not only deer density, but also habitat conditions, crop and forest damages, and other elements of ecosystem are also important.

Literature Cited Cederlund, G., J. Bergqvist, R. Gill, J. M. Gaillard, B. Boisaubert, P. Ballon, and P. Duncan. 1998. Managing roe deer and their impact on the environment: Maximizing the net benefits to society. Pages 337–372 in R. Andersen, P. Duncan and J. D. C. Linnell, editors, The European roe deer: The biology of success. Scandinavian University Press, Oslo, Norway.

418

H. Uno et al.

Cochran, W. G. 1977. Sampling techniques, 3rd edition. Wiley, New York, New York, USA. DeCalesta, D. S., and S. L. Stout. 1997. Relative deer density and sustainability: A conceptual framework for integrating deer management with ecosystem management. Wildlife Society Bulletin 25:252–258. Hokkaido Government. 1987. A report of an ecological survey for sika deer on Hokkaido. Hokkaido Nature Preservation Division, Sapporo, Japan. (In Japanese.) Hokkaido Government. 1998. Conservation and management plan for sika deer (Cervus nippon) in eastern Hokkaido. Department Environment and Lifestyle, Hokkaido Government, Sapporo, Japan. (In Japanese.) Hokkaido Government. 2000. Conservation and management plan for sika deer (Cervus nippon) in Hokkaido. Department Environment and Lifestyle, Hokkaido Government, Sapporo, Japan. (In Japanese.) Hokkaido Government. 2001. A report of exploitation of venison and measures for reducing damage to agriculture. Department of Agriculture, Hokkaido Government, Sapporo, Japan. (In Japanese.) Hokkaido Government. 2002. Conservation and management plan for sika deer (Cervus nippon) in Hokkaido. Department Environment and Lifestyle, Hokkaido Government, Sapporo, Japan. (In Japanese.) Hokkaido Institute of Environmental Sciences. 1995. Results of a survey related to sika deer and brown bear in Hokkaido. Hokkaido Institute of Environmental Sciences, Sapporo, Japan. (In Japanese.) Inukai, T. 1952. The sika deer in Hokkaido and its rise and decline. Hoppo Bunka Kenkyu (Report of North Culture Research) 7:1–45. (In Japanese.) Kaji, K. 2001. Relative deer density, deer and habitat. Pages 245–251 in K. Kaji, editor, Study on conservation and management of sika deer in Hokkaido, 1996–2000. Hokkaido Institute of Environmental Sciences, Sapporo, Japan. (In Japanese.) Kaji, K., and M. Tomizawa. 1993. Census methods and evaluation of sika deer population in Hokkaido. Honyurui Kagaku (Mammalian Science) 32:127–134 (In Japanese.) Kaji, K., T. Koizumi, and N. Ohtaishi. 1988. Effects of resource limitation on the physical and reproductive condition of sika deer on Nakanoshima Island, Hokkaido. Acta Theriologica 33:187–208. Kaji, K., M. Miyaki, T. Saitoh, S. Ono, and M. Kaneko. 2000. Spatial distribution of an expanding sika deer population on Hokkaido Island, Japan. Wildlife Society Bulletin 28:699–707. Kaji, K., H. Okada, M. Yamanaka, H. Matsuda, and T. Yabe. 2004. Irruption of a colonizing sika deer population. Journal of Wildlife Management 68:889–899. Kaneko, M., K. Kaji, and S. Ono. 1998. Analysis of habitat change and sika deer distribution on Hokkaido. Honyurui Kagaku (Mammalian Science) 38:49–59. (In Japanese with English summary.) Koizumi, T., N., N. Ohtaishi, and K. Kaji. 1984. Population trend of sika deer in Hokkaido. Acta Zoologica Fennica 172:79–80. Kurosawa, N. 2000. Lead poisoning in Steller’s sea eagles and white-tailed sea eagles. Pages 107–109 in M. Ueta and M. J. McGrady, editors, First symposium on Steller’s and white-tailed sea eagles in East Asia. Wild Bird Society of Japan, Tokyo, Japan. Matsuda, H., K. Kaji, H. Uno, H. Hirakawa, and T. Saitoh. 1999. A management policy for sika deer based on sex-specific hunting. Research on Population Ecology 41:139–149. Matsuda, H., H. Uno, K. Tamada, K. Kaji, H. Hirakawa, and T. Saitoh, K. Kurumada, and T. Fujimoto. 2002. Harvest-based estimation of population size for sika deer on Hokkaido Island, Japan. Wildlife Society Bulletin 30:1160–1171. McCullough, D. R. 1984. Lessons from George Reserve, Michigan. Pages 211–242 in L. K. Halls, editor, White-tailed deer ecology and management. Stackpole Books, Harrisburg, Pennsylvania, USA. McCullough, D. R. 1997. Irruptive behavior in ungulates. Pages 69–98 in W. J. McShea, H. B. Underwood, and J. H. Rappole, editors, The science of overabundance: Deer ecology and population management. Smithsonian Institution Press, Washington, DC, USA. McGrady, M. J., M. Ueta, E. Potapov, I. Utekhina, V. B. Masterov, M. Fuller, W. S. Seegar, A. Ladyguin, E. G. Lobkov, and V. B. Zykov. 2000. Migration and wintering of juvenile and

29 Sika Deer Population Irruptions and Their Management on Hokkaido Island

419

immature Steller’s sea eagles. Pages 83–90 in M. Ueta and M. J. McGrady, editors, First symposium on Steller’s and white-tailed sea eagles in East Asia. Wild Bird Society of Japan, Tokyo, Japan. Miyaki, M., and Y. Nishikawa. 2001. Availability of summer forage for sika deer in the broadleaved forests in eastern Hokkaido. Pages 51–70 in K. Kaji, editor, Study on conservation and management of sika deer in Hokkaido, 1996–2000. Hokkaido Institute of Environmental Sciences, Sapporo, Japan. (In Japanese.) Nagata, J., R. Masuda, K. Kaji, M. Kaneko, and M. C. Yoshida. 1998. Genetic variation and population structure of the Japanese sika deer (Cervus nippon) in Hokkaido Island, based on mitochondrial D-loop sequences. Molecular Ecology 7:871–877. Sakuragi, M., H. Igota, H. Uno, K. Kaji, M. Kaneko, R. Akamatsu, and K. Maekawa. 2003. Seasonal habitat selection of an expanding sika deer Cervus nippon population in eastern Hokkaido, Japan. Wildlife Biology 9:141–153. Suzuki, M., and N. Ohtaishi. 1993. Reproduction of female sika deer (Cervus nippon yesoensis Heude, 1884) in Ashoro district, Hokkaido. Journal of Veterinarian Medical Science 55:833–836. Suzuki, M., K. Kaji, M. Yamanaka, and N. Ohtaishi. 1996. Gestational age determination, variation of conception date, and external fetal development of sika deer (Cervus nippon yesoensis Heude, 1884) in eastern Hokkaido. Journal of Veterinarian Medical Science 58: 505–509. Terasawa, K., and N. Akashi. 2001. Effects of sika deer on preservation and regeneration of natural forest. Pages 111–134 in K. Kaji, editor, Study on conservation and management of sika deer in Hokkaido, 1996–2000. Hokkaido Institute of Environmental Sciences, Sapporo, Japan. (In Japanese.) Uno, H. 2000. The biomass and availability of Sasa kurilensis for sika deer in Akan district. Pages 34–40 in M. Miyaki, editor, Report of effects of sika deer on vegetation in Akan National Park. Maeda Ippoen Foundation, Akan, Japan. (In Japanese.) Uno, H., and K. Kaji. 2000. Seasonal movements of female sika deer in eastern Hokkaido, Japan. Mammal Study 25:49–57. Uno, H., and K. Kaji. 2006. Survival and cause-specific mortality rates of female sika deer in eastern Hokkaido, Japan. Ecological Research 21:215–220. Uno, H., and K. Tamada. 2001. Population characteristics of sika deer in Akan area. Pages 18–31 in K. Kaji, editor, Study on conservation and management of sika deer in Hokkaido, 1996– 2000. Hokkaido Institute of Environmental Sciences, Sapporo, Japan. (In Japanese.) Uno, H., Y. Takashima, and H. Tomizawa. 1995. Sika deer damaged the forest. Forest Protection 249:36–38. (In Japanese.) Uno, H., M. Yokoyama, and M. Takahashi. 1998. Winter mortality pattern of sika deer (Cervus nippon yesoensis) in Akan National Park, Hokkaido. Honyurui Kagaku (Mammalian Science) 38:233–246. (In Japanese with English summary.) Uno, H., K. Kaji, T. Saitoh, H. Matsuda, H. Hirakawa, K. Yamamura, and K. Tamada. 2006. Evaluation of relative density indices for sika deer in eastern Hokkaido, Japan. Ecological Research 21:624–632. Waller, D. M., and W. S. Alverson. 1997. The white-tailed deer: A keystone herbivore. Wildlife Society Bulletin 25:217–226. Working Group of White-tailed Eagles and Steller’s Sea Eagles. 1996. Wintering status of Steller’s sea eagles and white-tailed eagles in northern Japan. Pages 1–9 in Survey of status and habitat conditions of threatened species. Environment Agency, Tokyo, Japan. (In Japanese.)

Chapter 30

Irruptive Behavior of Sika Deer Koichi Kaji, Hiroshi Takahashi, Hideaki Okada, Masao Kohira, and Masami Yamanaka

Abstract A dominant paradigm of large herbivores is that following introduction to new range, or release from harvesting, the herbivore population at a low level will increase rapidly to a peak, followed by a crash, then recover to a lower density than peak abundance. However, supporting evidence has tended to be anecdotal. We have been monitoring two sika deer (Cervus nippon) populations and their habitats over 20 years: a deer population introduced to Nakanoshima Island (NKI) and a naturally colonizing deer population on Cape Shiretoko (CS) on Hokkaido, Japan. Both populations built to peak abundance followed by a crash, which resulted in significant effects on the vegetation. There were, however, marked differences in post-crash behavior between the two populations. Following the crash, the NKI herd continued to increase with a lower growth rate and reached a higher peak population size than the first irruption, while the CS herd showed repeated irruptions and crashes with no decline in carrying capacity (K). The NKI herd exhibited densitydependent changes in population parameters such as delayed sexual maturity, lower calf:female ratio, and lower body and antler growth as deer exceeded carrying capacity in the initial irruption. As a result of the irruption there was a decline in both winter- and summer-range quality. Thus, competition for high-quality food among sika deer in the initial irruption could have been a limiting factor, whereas unlimited abundance of poor-quality forage permitted a slower growth to even higher density in the subsequent buildup. In contrast, the CS herd exhibited a high adult survival rate and calf:female ratio and good antler growth, which indicated high quality of summer range. In addition, mortality patterns in crash years were also different between the populations; for the NKI herd, mortality was composed of both sexes in all age classes throughout years, while for the CS herd, mortality was composed mainly of calves and adult males, with few adult females. Although density-dependent resource limitation through interaction with winter climate was the important limiting factor of peak density for both populations, the carrying capacity differences and lags between summer and winter might generate different fluctuations in numbers for the two populations.

D. R. McCullough et al. (eds.), Sika Deer: Biology and Management of Native and Introduced Populations, DOI: 10.1007/978-4-431-09429-6_30, © Springer 2009

421

422

K. Kaji et al.

Introduction Cervid species have expanded their range and increased dramatically in abundance across Europe (roe deer, Capreolus capreolus, and moose, Alces alces) and North America (white-tailed deer, Odocoileus virginianus, and mule/black-tailed deer, O. hemionus) in recent decades due to reductions in hunting and natural predators, habitat changes due to agricultural and silvicultural activities, and moderate climates (McShea et al. 1997; Linnell et al. 1998; Côte et al. 2004). Similar factors might contribute to range expansion of sika deer in Japan, where a 70% expansion of the range was recorded during the last three decades (Biodiversity Center of Japan 2004). Wolves (Canis lupus) were extirpated by 1890 on Hokkaido Island (Inukai 1933) and by 1905 in Honshu Island (Hiraiwa 1981). Hunters in Japan have decreased from half a million in the mid-1970s to 150,000 in the mid-2000s. Bans on hunting and bucks-only hunting following overexploitation in the early 1900s, mild winters in recent decades, and human alteration of habitat have contributed to both a range expansion and increase in abundance of sika deer populations. On Hokkaido Island, sika deer that once decreased to near-extinction due to overexploitation and heavy snowfall recovered under government protection and irrupted in eastern Hokkaido in the mid-1990s (Uno et al. chapter 29). Thereafter the deer population has been continuously increasing in western Hokkaido (Hokkaido Institute of Environmental Sciences 2006). Overabundant sika deer populations have caused severe damage to agriculture and to plantation and natural forests as well. Feeding by sika deer has heavily affected vegetation on the forest floor of natural forests, leading to erosion in some parts of Japan. To reduce damage to agriculture and forestry, aggressive population control of sika deer is conducted all over Japan. Associated with sika deer range expansion, the deer have influenced the natural vegetation at nearly half of the country’s 83 national parks and quasi-national parks (Tokida 2006). Greatest concerns with deer damage are in Shiretoko National Park (Tokida et al. 2004; Kaji et al. 2006) and Yakushima National Park (Yahara 2006), both registered as World Natural Heritage Sites, where some endemic and rare plant species are threatened with extinction. The most problematic issues concerning irruptive behavior of ungulates occur in parks and reserves where the values of nature are supposed to supersede those of humans (McCullough 1997). A long, continuing debate has taken place over the need for human intervention to control the numbers (and environmental effects) of the North American elk (Cervus elaphus) herd in Yellowstone National Park, USA, where elk numbers are purported to be inducing changes in the vegetation (Wagner et al. 1995; Huff and Varley 1999; National Research Council 2002). Similar controversy has arisen for sika deer management policy in Shiretoko National Park (Kaji et al. 2006) and Yakushima National Park (Yahara 2006). Major concerns are the effects of natural regulation of sika deer populations and, subsequently, the effects on vegetation. After release from human hunting, deer populations have been commonly exhibited explosive population increases, and this phenomenon has usually been called “irruption.” Since Leopold (1943) first formally conceptualized the irruptive phenomenon in

30 Irruptive Behavior of Sika Deer

423

ungulate populations, irruptive behavior of ungulates has been a great concern of the ecology and management fields. A dominant paradigm of large herbivores is that following introduction to new range or release from harvesting, an herbivore population at low level increases rapidly to a peak followed by a crash, then recovers to a lower density than peak abundance (Leopold 1943; Riney 1964; Caughley 1970; McCullough 1997). However, evidence of a build-up and crash of ungulates based on long-term data is very limited, including cases such as reindeer (Rangifer tarandus) introduced to St. Matthew Island in the Bering Sea in Alaska (Klein 1968) and South Georgia Island in the South Atlantic Ocean (Leader-Williams 1988), and sika deer (Cervus nippon) introduced to Nakanoshima Island (Kaji et al. 1988) and naturally colonizing the Shiretoko Peninsula on Hokkaido (Kaji et al. 2004). Although the dominant paradigms predict a first irruption, followed by a recovery to a reduced carrying capacity through a dampened oscillation to K (Caughley 1976) or to a stable equilibrium at a lowered K (Leopold 1943), inferences about irruptions have tended to be anecdotal (McCullough 1997). Forsyth and Caley (2006) developed mathematical models to better describe irruptive dynamics of large herbivores and evaluated the dynamics of seven ungulate populations either introduced to new range or released from harvesting. Although the latter study supported the widespread existence of irruptive dynamics, none of the published studies measured food. Thus, the key issue in irruptive dynamics is how food diminishes after the herbivores were introduced or released from harvesting (Forsyth and Caley 2006). McCullough (1997) provided a hypothesis test for three irruptive paradigms; two were the classic paradigms proposed by Leopold (1943) and Caughley (1976): if the previous effects of ungulate feeding on vegetation have been detrimental, then both or either (1) the population growth rate and (2) the population size achieved should be lower following the initial irruption. The third paradigm has been derived from Clutton-Brock and his colleagues’ (Clutton-Brock et al. 1991; Clutton-Brock 1994) study of Soay sheep (Ovis aries) on Hirta and St. Kilda Islands, Scotland, and similar results were found for white-tailed deer in Michigan (McCullough 1982, 1983; Case and McCullough 1987) and black-tailed deer in California (McCullough 1997). This paradigm involves repeated irruptions and crashes with no accompanying decline in K (McCullough 1997). In this chapter, we first review case studies of irruptive behavior of sika deer and its consequences on the habitat in which it occurred based on long-term monitoring over 20 years and discuss classic irruptive paradigms following hypotheses tests proposed by McCullough (1997).

Case Studies Nakanoshima Island Study site. Nakanoshima Island, along with two islets, has a total area of 520 ha and is located in Lake Toya, southwestern Hokkaido, the northernmost island of

424

K. Kaji et al.

Japan. The major vegetation of NKI (497.8 ha) is composed of deciduous broadleaved trees (467 ha) such as Acer mono (painted maple) and Tilia japonica (linden), a coniferous plantation (18.3 ha) of Abies sachalinensis (Sakhalin fir) and Larix leptolepis (Japanese larch), and an open grassland (7.1 ha). This island is part of the wildlife reserve of Shikotsu-Toya National Park, and no predator effective against sika deer is present. NKI is characterized by mild winters and mainly deciduous broad-leaved trees which provide poor thermal cover. Emigration and immigration of sika deer are negligible. Nakanoshima Island has been the long-term study site of a sika deer population since 1980. The initial irruption (Kaji et al. 1988), its consequent effects on habitat changes (Kaji et al. 1991; Kaji and Yajima 1992; Kaji 2003), on food habit changes (Takahashi and Kaji 2001; Miyaki and Kaji 2004), and long-term effect on the habitats (Miyaki and Kaji chapter 13) have been previously reported. Population changes. The population grew from three deer (one male and two females) introduced during 1957–1966 to a peak of 273 animals in March 1984 and declined to 137 animals (26.3 deer/km2) by May 1984 due to a population crash (n = 67 deer) and removal (n = 95 deer) (Kaji et al. 1988; Fig. 30.1). Because we found 26 carcasses and 41 carcasses before and after the drive count, respectively, we derived a conservative pre-crash population estimate of 299 deer. Thus, around 22% of the population died in 1983–1984. This mortality rate might be underestimated because removal probably prevented additional mortality that otherwise would have occurred. A major response of the population during the first irruption was reduction of the calf to female ratio (calf:100 females), which dropped sharply from 65 to seven, and was negatively correlated with population density (Kaji et al. 1988). Although the first irruption of the NKI population imposed irreversible changes on the vegetation (Fig. 30.2), the population during the post-crash recovery

Fig. 30.1 Population changes of sika deer on Nakanoshima Island, Hokkaido, Japan, 1957–2006. Black bars show the removals, open circles show estimated population size, black circles show minimum population ground counts, and arrows show population crashes following a peak (Kaji et al. in preparation).

30 Irruptive Behavior of Sika Deer

425

Fig. 30.2 Sika deer irruption and changes in forest vegetation on Nakanoshima Island, Lake Toya, Hokkaido, Japan (Kaji 2003).

Fig. 30.3 Log-linear regressions from 1966 through 1984 indicating a 16% annual rate of increase (r = 0.15, λ = 1.16, R2 = 0.9979), and from 1986 through 2001 indicating a 6% increase (r = 0.067, λ = 1.07, R2 = 0.7498). Xs show population size from 2002 to 2006 (Kaji and Takahashi in preparation).

continued to increase with significantly lower growth rate (1986–2001: rm = 0.07, 95% CI:0.04–0.08) than the initial irruption (1964–1984; rm = 0.15, 95% CI:0.15; Fig. 30.3) and reached a second peak of 434 animals (83.5 deer/km2) in March 2001, which was 1.6 times higher than the first peak (Fig. 30.1). The population declined a second time to 304 animals (58.5 deer/km2) by May 2001 due to natural mortality (n = 42 deer) and removal (n = 102 deer). The population quickly recovered once again to 437 animals in March 2003 and a third crash followed. We found a total of 114 carcasses by May 2004, which was comprised of 33 carcasses before the count, 59 carcasses after the count, and 22

426

K. Kaji et al.

carcasses that could not be determined to time due to condition. When we allocate the 22 undetermined season of death according to the known mortality dates before and after the count, we derived a conservative pre-crash population of 478. Thus around 24% of the population died in 2002–2003. After the 2004 population crash, the estimated number of deer decreased to 159 deer in March 2006 (Fig. 30.1). Vegetation changes. In 1980, when the sika deer density was around 30 deer/km2, deer browsing or bark-stripping affected saplings, dwarf bamboos, and small trees on wintering grounds; however, there was ample available food. With increasing population density, deer had a marked influenced on forest structure and composition by bark-stripping and browsing of trees. Also, they eliminated dwarf bamboos (Sasa senanensis and S. kurirensis), the major understory plants on the island before 1984, and important winter foods for sika deer in Japan (Fig. 30.2; Kaji et al. 1991; Kaji and Yajima 1992). The first crash in the winter of 1984 coincided with the elimination of dwarf bamboos, which had been the most important forage in winter (Kaji et al. 1988). The open grassland in the central part of NKI consisted of forbs, grasses (≥30 cm in height), and tall plants (≥100 cm), and there was no obvious effect of grazing on the vegetation by 1981. However, in the following years heavy grazing by sika deer changed the vegetation into a short-grass community, except that Senecio cannabifolius (Aleutian ragwort), an unpalatable plant which was not grazed by deer in 1982 remained (Kaji et al. 1988). After the population crash in 1984, when available twigs, bark, and dwarf bamboos were eliminated, the deer shifted to fallen leaves and unpalatable plants such as Cephalotaxus harringtonia (Japanese plum yew), Senecio cannabifolius, and Cynanchum caudatum (Asclepiadaceae) as winter foods, which previously were untouched by deer (Takahashi and Kaji 2001; Miyaki and Kaji 2004, Miyaki and Kaji chapters 12 and 13). Among these three unpalatable plants, Japanese plum yew was most widely distributed on NKI and served as an important forage in winter, but it was eliminated by 2003. Even thought the winter of 2003–2004 was relatively mild, the most severe population crash occurred, which coincided with the elimination of Japanese plum yew. Thereafter the NKI herd, dependent mainly on fallen leaves throughout year (Ueno et al. 2007) has been decreasing (Fig. 30.1). Body mass and population parameters. Although mean body mass and skeletal size of NKI deer initially were almost the same as those of mainland deer, the resource limitation caused by high deer density was reflected in altered physical characteristics of the deer. Most noticeable changes between pre-crash (1982) and post-crash (1984) were in body mass and antler size. Compared to the 1982 body mass, the deer captured in 1984 were smaller in body mass for all age classes in both sexes, and in hind foot length for calves (Kaji et al. 1988). In 1988, four years after the first crash in 1984, the body mass of females recovered temporarily. It decreased again thereafter, however, and reached the lowest recorded body mass in 2004 when the second crash occurred in the population (Fig. 30.4; Kaji and Takahashi in preparation). Between 1982 and 2004 body mass of females decreased 29% for calves, 17% for yearlings, 9% for 2.5-year olds, and 15% for ≥3.5-year olds. The most significant reduction of body mass occurred in

30 Irruptive Behavior of Sika Deer

427

Fig. 30.4 Body mass of female sika deer age by age classes during 1982–2004 on Nakanoshima Island, Hokkaido. Bars show SD when sample size was greater than three (Kaji and Takahashi in preparation).

young age classes, but corresponding changes in body mass occurred in all age classes, suggest that the effect was consistent. There was an inverse relationship between spring body mass in each age class and the previous winter population count (Kaji and Takahashi in preparation). This suggests density-dependent changes in body mass occurred through resource limitation. Sika deer generally conceive a calf when they are yearlings, however, associated with body mass decline, the age at maturity delayed from yearling to two years old during 1982 and 1984 (Kaji et al. 1988). Thereafter age at maturity was further delayed to three years old. There was, however, no obvious change in pregnancy rates for adult females ≥3 years old (Kaji and Takahashi in preparation). The mortality during crash years consisted of both sexes and all age classes. The first crash of 1983–1984 (n = 67 deer) involved 27% males, 25% females, 13% yearlings, and 39% calves; 4% were of unknown sex and age. The second crash of 2003–2004 (n = 114 deer) was composed of 45% males, 33% females, 4% yearlings, and 18% calves.

Cape Shiretoko Study site. Sika deer, once locally extirpated, recolonized the Shiretoko Peninsula, Hokkaido in the early 1970s from a nearby source population, the Akan herd (Kaji 1988). The population has been expanding its range and size since the mid-1980s, resulting in chronic strong effects on the natural vegetation of forests and grasslands

428

K. Kaji et al.

on wintering grounds. Since the Shiretoko Peninsula (Shiretoko National Park and peripheral areas) has been designated a World Natural Heritage Site of UNESCO, management of the overabundant deer population is a major concern. Among four major wintering grounds on Shiretoko Peninsula, Cape Shiretoko (CS) was the largest one, and this is where our study site of 5 km2 is located. Sika deer on CS were virtually isolated from other wintering populations in the park. Although migration was possible, dispersal was restricted because of natural barriers to movement, such as abrupt mountainous topography and steep cliffs, and overpopulation by sika deer on much of Shiretoko National Park (Kaji et al. 2004). Recent study using radio-collared sika deer on CS showed that the CS herd was resident (M. Kohira, Shiretoko Nature Foundation, personal communication 2007). Cape Shiretoko was the site of sika deer study reported by Kaji et al. (2004, 2006), from which this account is derived. Cape Shiretoko is characterized by cold winters, heavy snow fall, deciduous and coniferous mixed forests, and it is inhabited by a potential predator, the brown bear (Ursus arctos). The vegetation of CS is characterized by a natural grassland in a long strip along a terrace above the sea cliffs that is about 200–500 m wide and 4,500 m long (about 2 km2 in total area) (Fig. 30.5). Prior to the sika deer invasion, Calamagrostis langsdorfii (reed grass), Miscanthus sinensis (silvergrass), Artemisia montana (mugwort), and Sasa senanensis (dwarf bamboo) were widespread. The grassland is adjacent to deciduous–coniferous mixed forests dominated by Acer mono (painted maple), Quercus crispula (Mongolian oak), Ulmus davidiana (elm), and Abies sachaliensis (Sakhalin fir) (Tatewaki et al. 1966). Three species of dwarf bamboo occurred on CS: S. senanensis, S. kurilensis, and S. shikotanensis. Sasa senanensis, the major winter forage, was found in forests, whereas S. kurilensis and S. shikotanensis were distributed in small areas outside the forest. All of the bamboos in the forest disappeared due to heavy browsing by deer.

Fig. 30.5 Wintering sika deer on Cape Shiretoko (photo by Hideaki Okada).

30 Irruptive Behavior of Sika Deer

429

Fig. 30.6 Changes in the sika deer population on Cape Shiretoko, Hokkaido, during 1986–2005, based on aerial photographic censuses (Kaji et al. 2004 and unpublished data).

Population change. The naturally colonizing deer population on CS increased from 54 animals (11 deer/km2) in April 1986 to 592 animals (118 deer/km2) in February 1998 and declined to 177 animals (35 deer/km2) in March 1999 (Kaji et al. 2004; Fig. 30.6). Following the 1999 crash, the population rapidly recovered and reached a second peak of 626 animals in 2003, decreased to 341 deer in 2004, and reached a third peak of 603 in 2005 (Fig. 30.6). Thus, the CS herd showed repeated irruptive behavior with no apparent lessening of the peak population size. Systematic carcass counts of dead deer were carried out each May by Shiretoko Nature Foundation from 1999 to 2005 (except in 2001), and mass mortality was observed in 1999, 2004, and 2005. Most mortality occurred in April after population counts. Thus, we could estimate that minimum mortality rates were 37%, 26%, and 19% in 1999, 2004, and 2005 respectively. Vegetation change. In 1983, when sika deer density was over 8 deer/km2, barkstripping of large elms (Ulmus laciniata) trees occurred (Fig. 30.7). There were, however, no obvious impacts of sika deer on forest vegetation as yet. Woody vegetation showed signs of overbrowsing in 1987 when deer density was greater than 15 deer/km2 (Fig. 30.7). Small trees (dbh <15 cm) disappeared due to browsing and bark-stripping, and a browse line was formed. Bark peeling was concentrated in the smaller dbh classes in 1987 and expanded to larger trees thereafter. Sika deer foraging strongly affected forest species composition and structure. With increasing deer density, the coverage, height, and culm density of dwarf bamboo in forests decreased (Kaji et al. 2004). In 1993, when deer density was 80 deer/km2, elm, Prunus ssiori (Japanese bird cherry), and dwarf bamboo in forests were eliminated. Aleutian ragwort, an unpalatable plant for deer, expanded into sites previously occupied by dwarf bamboo. In 1998, when the sika deer population reached a peak density of 118 deer/km2, we found that large oak (Quercus crispula) trees

430

K. Kaji et al.

Fig. 30.7 The initial sika deer irruption and crash and related changes in forest vegetation on Cape Shiretoko, Hokkaido. (Kaji 2003).

(dbh >40 cm) were being debarked by deer, an indication of a severe winter food shortage. The sika deer herd crashed the following winter. Over 20 years of grazing effects on herbaceous plants on CS has drastically changed the plant community. Compared to the 1960s vegetation status, tall plants (Hemerocallis yezoensis (daylily), Filipendula kamtschatica (meadowsweet), Artemisia montana, and Empetrum nigrum (crowberry)) drastically decreased to near extinction, while unpalatable plants (Senecio cannabifolius and Ligularica hodgsonii (Compositae)) and invasive plants (Cirsium vulgare (bull thistle)) greatly expanded their range (Tokida et al. 2004). Body mass and population parameters. Comparison between the population structure of CS sika deer in 1980s and 2000s revealed that sex ratios (100 females: males) drastically shifted towards females, from 86 males:100 females in the 1980s to 53 males:100 females in the 2000s (Chi-square test, p = 0.03). There were, however, no significant differences in calf to female ratios between the two periods (59 calves:100 females in 1980s versus 50 calves:100 females in 2000s, Chi-square test, p > 0.05), which indicates that there was no major change in productivity between the two periods. Because antler length of sika deer is closely related to body mass and lengthens as mass increases (Kaji et al. 1988), we compared length of cast antlers on CS with those of adult males dying of natural causes on CS in 1999, and those of harvested males in eastern Hokkaido during 1979–1980 (Kaji et al. 2004). We found no significant differences in size between adult male cast antlers collected during 1984–1987 and antlers of adult males dying from natural causes during 1999 on CS. Males dying from natural causes at CS had larger antlers than harvested males in eastern Hokkaido.

30 Irruptive Behavior of Sika Deer

431

Discussion Both populations of sika deer we studied reached peak abundance following introduction (NKI) or recolonizing (CS) to new range, and crashes followed. Densitydependent resource limitations through interaction of winter climate coincided with the apparent peak and subsequent population crash of sika deer for both herds. These initial irruptive patterns are typical for ungulate introductions to previously unoccupied areas (Caughley 1970). There were, however, marked differences in post-crash behavior between the two populations. Although the initial irruption of both populations resulted in significant effects on their habitats, neither of those populations showed a diminution of peak abundance in the second (NK) or subsequent (CS) peaks, as proposed by Leopold (1943) and indicated by Caughley’s (1976) model. The differences in irruptive post-crash behavior between NKI and CS were reflected in available food, body mass, and mortality patterns. The NKI herd exhibited density-dependent changes in population parameters, such as delayed sexual maturity, lower calf:female ratio, and lower body and antler growth as a result of the decline in both winter and summer range quality at close to carrying capacity (Kaji et al. 1988). Conversely, the CS herd exhibited high adult survival rate, calf: female ratio, and good antler growth (Kaji et al. 2004). Klein (1968) suggested that for ungulates in northern latitudes, winter-range carrying capacity determines the upper limit of population numbers, while summer-range quality determines the size of individuals. Thus, the poor and good physical condition of the NKI and CS herds, respectively, reflected summer-range quality for each habitat although both were considerably changed following the initial irruption, whereas peak population size was determined by density-dependent resource limitation combined with winter climate (Kaji et al. 1988, 2004). For the NKI herd, the first crash occurred in 1984 during a relatively severe winter, while the second crash occurred in 2001 during a warm winter. For the CS herd, although snow depth affected the wintering population size, we found no evidence that snow depth and/or snowfall duration were sole factors in mass mortality of sika deer, because the 1999, 2004, and 2005 mass mortalities of the CS herd occurred during a period of “average” winter conditions, as measured during the 1979–2005 period. Thus winter climate was not a sole limiting factor, but worked in combination with resource limitation, which played a more important role for determination of peak density. In a similar case, sika deer on Kinkazan Island, off the east coast of Honshu Island in northern Japan, were reduced by hunting in the 1950s and subsequently recovered to around 500 deer. Later the population crashed when nearly half of the population died in 1984 due to malnutrition combined with exceptional cold and late snow (Takatsuki et al.1994). Thereafter the population recovered to over 500 deer, and nearly half of the population crashed again in the “normal winter” of 1997 (Takatsuki 2006). For mule deer, Hobbs and Swift (1985) proposed that ranges with low quality resources of greater extent or amount can have higher carrying capacities for animals than ranges with more limited high quality forage, even though the latter

432

K. Kaji et al.

supports higher population growth rates and better physical condition of individuals. Similarly, Messier (1991) emphasized that, as a limiting factor, forage quality may influence moose population growth. Poor-quality forages, however, if in unlimited abundance, actually may permit slow continuous moose population growth and lead to consequent high density (Van Ballenberghe and Ballard 1997). Following Hobbs and Swift’s (1985) argument, habitat on NKI changed from a previous smaller range of good quality resources (dwarf bamboos and browse species) to a large amount of poor quality resources (fallen leaves and unpalatable resources). Thus, competition for high-quality food among sika deer at high density during the first irruption could be a regulating and limiting factor, while unlimited abundance of poor-quality forage permitted slower growth and higher density during the first post-crash increase. After elimination of the woody plant, Japanese plum yew, fallen leaves were the major resource, which alone could not support high density, so the population has been decreasing thereafter. For the CS herd, as with the NKI herd, the first crash was coincident with elimination of dwarf bamboos and woody plants species in forest. However, a large amount of dwarf bamboo still remained on the natural grassland, where the availability was strongly affected by snow depth. The difference of habitat quality for the two populations might also be reflected in mortality patterns. For the NKI herd, both sexes in all age classes starved in all seasons (Takahashi et al. in preparation), while for the CS herd, starvation occurred only in late winter and was biased towards calves and males. Although climatic conditions and scarcity of food resources during winter might regulate both the NKI and CS herds, the CS population density could remain high due to the high female survival rate as well as a high intrinsic rate of increase. This caused repeated irruptions and crashes similar to the Soay sheep on St. Kilda reported by Clutton-Brock et al. (1991). Fluctuations in the numbers of Soay sheep are the result of over-compensatory density-dependent mortality in winter, following the lack of any substantial reduction in fecundity at high density resulting from superabundant summer resources and early weaning (Clutton-Brock 2004). Population crashes occur in years when density is high and winter weather conditions are unfavorable (Clutton-Brock 2004). The cause and consequences of instability of Soay sheep are similar for sika deer on CS, except for early weaning. We have shown that in the initial irruptive population of sika deer, simple dynamics may appear as a result of the combination of density-dependent resource limitation and environmental stochasticity. In contrast to the initial irruption, the sequent irruptions were caused by a complex interaction of the climate-vegetationherbivore system, as shown by recent long-term studies, which have revealed a complex interplay between the effects of climate and density on population dynamics (Clutton-Brock et al. 1997; Loison and Langvaton 1998; Post and Stenseth 1998, 1999; Solberg et al. 1999, 2001). Acknowledgements The Nakanoshima Island studies reported here were supported by Hokkaido Government and Agriculture, Forestry and Fisheries Research Council and the Cape Shiretoko studies by Ministry of the Environment, Hokkaido Government, Shari town, and Fuji Film Green Fund. Numerous individuals contributed to these studies, but extensive field work for this project

30 Irruptive Behavior of Sika Deer

433

was carried out by Hiroshi Takahashi, Junpei Tanaka, Mayumi Ueno, and Eriko Shima for NKI, and Hideaki Okada, Masami Yamanaka, and Masao Kohira for CS. I thank Dale McCullough for improving an earlier version of this chapter. This study was supported in part by a Grant-in-Aid for Scientific Research (B) No. 19380084 from the Ministry of Education, Culture, Sports, Science and Technology.

Literature Cited Biodiversity Center of Japan. 2004. Research on species diversity: Report on mammal distribution. Nature Conservation Bureau, Ministry of the Environment, Japan. (In Japanese.) Case, D. J., and D. R. McCullough. 1987. The white-tailed deer of North Manitou Island. Hilgardia 55:1–57. Caughley, G. 1970. Eruption of ungulate populations, with emphasis on Himalayan tahr in New Zealand. Ecology 51:53–72. Caughley, G. 1976. Wildlife management and the dynamics of ungulate populations. Pages 183–246 in T. H. Coaker, editor, Applied biology 1. Academic Press, London, United Kingdom. Clutton-Brock, T. 1994. Counting sheep. Natural History 103:29–35. Clutton- Brock, T. 2004. The causes and consequences of instability. Pages 276–310 in T. H. Clutton-Brock and J. M. Pemberton, editors, Soay sheep: Dynamics and selection in an island population. Cambridge University Press, Cambridge, United Kingdom. Clutton-Brock, T., O. F. Price, S. D. Albon, and P. A. Jewell. 1991. Persistent instability and population regulation in Soay sheep. Journal of Animal Ecology 60:593–608. Clutton-Brock, T. H., A. W. Illius, K. Wilson, B. T. Grenfell, A. D. C. MacColl, and S. D. Albon. 1997. Stability and instability in ungulate populations: An empirical analysis. American Naturalist 149:195–219. Côté, S. D., T. P. Rooney, J. P. Tremblay, C. Dussault, and M. Waller. 2004. Ecological impact of deer overabundance. Annual Review of Ecology, Evolution and Systematics 35:113–147. Forsyth, D. M., and P. Caley. 2006. Testing the irruptive paradigm of large-herbivore dynamics. Ecology 87:297–303. Hiraiwa, Y. 1981. Wolves: Their ecology and history. Ikeda-Shoten, Tokyo, Japan. (In Japanese.) Hobbs, N. T., and D. M. Swift. 1985. Estimates of habitat carrying capacity incorporating explicit nutritional constraints. Journal of Wildlife Management 49:814–822. Hokkaido Institute of Environmental Sciences. 2006. Results of a survey related to sika deer on Hokkaido. Hokkaido Institute of Environmental Sciences, Sapporo, Japan. (In Japanese.) Huff, D. E., and J. D. Varley. 1999. Natural regulation in Yellowstone National Park’s Northern Range. Ecological Applications 9:17–29. Inukai, T. 1933. Review on extirpation of wolves on Hokkaido. Shokubutu to Dobutsu (Plants and Animals) 1:1091–1098. (In Japanese.) Kaji, K. 1988. Sika deer. Pages 155–180 in N. Ohtaishi and H. Nakagawa, editors, Animals of Shiretoko. Hokkaido University Press, Sapporo, Japan. (In Japanese with English summary.) Kaji, K. 2003. Sika deer and damages caused by the deer on Hokkaido: How to cope with deer. Sinrin Kagaku (Forestry Science) 39:28–34. (In Japanese.) Kaji, K., and T. Yajima. 1992. Influence of sika deer on forests of Nakanoshima Island, Hokkaido. Pages 215–218 in B. Bobeck, K. Prezaowski, and W. Regelin, editors, Global trends in wildlife management, Transactions Vol 1: 18th IUGB Congress, Krakow 1987. Swiat Press, KrakowWarsaw, Poland. Kaji, K., T. Koizumi, and N. Ohtaishi. 1988. Effects of resource limitation on the physical and reproductive condition of sika deer on Nakanoshima Island, Hokkaido. Acta Theriologica 33:187–208. Kaji, K., T. Yajima, and T. Igarashi. 1991. Forage selection by introduced deer on Nakanoshima Island, and its effect on the forest vegetation. Pages 52–55 in N. Maruyama, B. Bobek, Y. Ono,

434

K. Kaji et al.

W. Regelin, L. Bartos, and P. R. Ratcliffe, editors, Proceedings of the International Symposium on Wildlife Conservation, INTECOL 1990. Japan Wildlife Research Center, Tokyo, Japan. Kaji, K., H. Okada, M. Yamanaka, H. Matsuda, and T. Yabe. 2004. Irruption of a colonizing sika deer population. Journal of Wildlife Management 68:889–899. Kaji, K., H. Okada, M. Kohira, and M. Yamanaka. 2006. The Shiretoko sika deer herd: Management actions and natural regulation. Pages 43–55 (Japanese) and 229–231 (English) in D. R. McCullough, K. Kaji, and M. Yamanaka, editors, Wildlife in Shiretoko and Yellowstone National Parks: Lessons in wildlife conservation from two world heritage sites. Shiretoko Nature Foundation, Shari-chou, Japan. Klein, D. R. 1968. The introduction, increase, and crash of reindeer on St. Matthew Island. Journal of Wildlife Management 32:350–367. Leader-Williams, N. 1988. Reindeer on South Georgia: The ecology of an introduced population. Cambridge University Press, Cambridge, United Kingdom. Leopold, A. 1943. Deer irruptions. Wisconsin Conservation Bulletin 8:3–11. Linnell, J. D. C., P. Duncan, and R. Andersen. 1998. The European roe deer: A portrait of a successful species. Pages 11–22 in R. Andersen, P. Duncan, and J. D. C. Linnell, editors, The European roe deer: The biology of success. Scandinavian University Press, Oslo, Norway. Loison, A., and R. Langvaton. 1998. Short- and long-term effects of winter and spring weather on growth and survival of red deer in Norway. Oecologia 116:489–500. McCullough, D. R. 1982. Population growth rate of the George Reserve deer herd. Journal of Wildlife Management 46:1079–1083. McCullough, D. R. 1983. Rate of increase of white-tailed deer on the George Reserve: A response. Journal of Wildlife Management 47:1248–1250. McCullough, D. R. 1997. Irruptive behavior in ungulates. Pages 69–98 in W. J. McShea, H. B. Underwood, and J. H. Rappole, editors, The science of overabundance: Deer ecology and population management. Smithsonian Institution Press, Washington, DC, USA. McShea, W. J., H. B. Underwood, and J. H. Rappole, editors. 1997. The science of overabundance: Deer ecology and population management. Smithsonian Institution Press, Washington, DC, USA. Messier, M. 1991. The significance of limiting and regulating factors on the demography of moose and white-tailed deer. Journal of Animal Ecology 60:377–393. Miyaki, M., and K. Kaji. 2004. Summer forage biomass and the importance of litterfall for a highdensity sika deer population. Ecological Research 19:405–409. National Research Council. 2002. Ecological dynamics on Yellowstone’s Northern Range. National Academy Press, Washington, DC, USA. Post, E., and N. C. Stenseth. 1998. Large-scale climatic fluctuation and population dynamics of moose and white-tailed deer. Journal of Animal Ecology 67:537–543. Post, E., and N. C. Stenseth. 1999. Climate change, plant phenology, and northern ungulates. Ecology 80:1322–1339. Riney, T. 1964. The impact of introductions of large herbivores on the tropical environment. International Union for the Conservation of Nature Publication New Series 4: 261–273. Solberg, E., B. E. Sæther, O. Strand, and A. Loison. 1999. Dynamics of a harvested moose population in a variable environment. Journal of Animal Ecology 68:186–204. Solberg, E. J., P. Jordhøy, O. Strand, R. Aanes, A. Loison, B. E. Sæther, and J. D. C. Linnell. 2001. Effects of density-dependence and climate on the dynamics of a Svalbard reindeer population. Ecography 24:441–451. Takahashi, H., and K. Kaji. 2001. Fallen leaves and unpalatable plants as alternative foods for sika deer under food limitation. Ecological Research 16:257–262. Takatsuki, S. 2006. Ecological history of sika deer. University of Tokyo Press, Tokyo, Japan. (In Japanese.) Takatsuki, S., K. Suzuki, and I. Suzuki. 1994. A mass-mortality of sika deer on Kinkazan Island, northern Japan. Ecological Research 9:215–223. Tatewaki, M., K. Itoh, M. Tohyama, and T. Nigi. 1966. Vegetation community. Pages 2–33 in M. Tatewaki, editor, Vegetation on Cape Shiretoko, Japan. Forest Vegetation Research Group, Sapporo, Japan. (In Japanese.)

30 Irruptive Behavior of Sika Deer

435

Tokida, K. 2006. Sika deer problems in national parks—Histories on relationship between people and deer. Pages in 20–37 in T. Yumoto and H. Matsuda editors, Deer eat world heritage sites: Ecology of deer and forests. Bunichi Sogo Shuppan, Tokyo, Japan. (In Japanese.) Tokida, K., T. Torii, M. Miyaki, H. Okada, M. Kohira, Y. Ishikawa, K. Sato, and K.Kaji. 2004. A deer management approach to promote ecosystem management in National Parks: A case study of sika deer in Shiretoko, Hokkaido Island, Japan. Journal of Conservation Ecology 9:193–202. (In Japanese with English summary.) Ueno, M., C. Nishimura, H. Takahashi, K. Kaji, and T. Saitoh. 2007. Fecal nitrogen as an index of dietary nitrogen in two sika deer Cervus nippon populations. Acta Theriologica 52:119–128.. Van Ballenberghe, V., and W. B. Ballard. 1997. Population dynamics. Pages in 223–245 in A. W. Franzmann, and C. C. Schwartz, editors, Ecology and management of the North American moose. Smithsonian Institution Press, Washington, DC, USA and London, United Kingdom. Wagner, F. H., R. Foresta, R. B. Gill, D. R. McCullough, M. R. Pelton, W. F. Porter, and H. Salwasser. 1995. Wildlife policies in the U.S. national parks. Island Press, Washington, DC, USA. Yahara, T. 2006. Increases of deer and extinction risk of wild plants. Pages 168–187 in T. Yumoto and H. Matsuda, editors, Deer eat world heritage sites: Ecology of deer and forests. Bunichi Sogo Shuppan, Tokyo, Japan. (In Japanese.)

Chapter 31

The Management of Sika Deer Populations in Hyogo Prefecture, Japan Hiroshi Sakata, Shin-ichiro Hamasaki, and Hiromune Mitsuhashi

Abstract We describe the present status of sika (Cervus nippon) deer management in Hyogo Prefecture and our efforts to predict the future trend of sika deer populations and agricultural damage for management options. Using monitoring data from 1999 to 2001 we analyzed the relationships between sika deer density, hunting pressure, and damage to agriculture in Hyogo Prefecture. By regression analysis of the changing rate of a density index, we detected a negative relationship to the density index in the preceding year and proportion of urban area of the landscape and a positive relationship to neighboring deer abundance and proportion of agricultural area. Amount of agricultural damage was correlated to deer abundance, proportion of area in forest plantation, and the amount of boundary between cultivated lands and forests. These variables for the area within 2.5 km distance of focal cultivated land explained the amount of damage better than those within 5 km distance. Then, using the derived formula, we could predict the future trend of deer population according to harvesting plans. If 10,000 deer per year were harvested (approximately the harvest in 2001), the deer density and agricultural damages would not be reduced. Even harvesting 20,000 deer for a few years does not seem to reduce the population under the target population of the management plan. The accuracy of this model was examined by comparing the prediction to the monitoring data in the next year. Finally, we discuss the contribution of our effort to the advancement of deer management.

Introduction Deer Management in Hyogo Prefecture Hyogo Prefecture (135° E 35° N) is located in the center of Honshu Island, Japan (for location, see Fig. 14.1 in Yokoyama chapter 14). Density of sika deer in Hyogo is believed to have been low in the mid-twentieth century but to have increased since the 1970s. At that time, deer began to cause serious damage to agriculture and D. R. McCullough et al. (eds.), Sika Deer: Biology and Management of Native and Introduced Populations, DOI: 10.1007/978-4-431-09429-6_31, © Springer 2009

437

438

H. Sakata et al.

10000 8000 6000 4000

Harvested males

Harvested females

Nuisance-controlled males

Nuisance-controlled females

2000

1995

1990

1985

1980

1975

0

1970

2000 1965

Number of sika deer

12000

Fig. 31.1 Number of sika deer harvested in Hyogo Prefecture by year from 1961 to 2001.

forestry. In the late 1970s, the number of deer removed by nuisance control and harvest began to increase (Fig. 31.1). However, the prefecture government still prohibited shooting and trapping females except for permitted nuisance controls until 1994 (Sakata et al. 2001). The lack of a systematic deer management plan has increased human-deer conflicts in Hyogo Prefecture. Many hunters, agriculturists, and foresters have felt the necessity of radical control. However, the government has not changed its protection policy due to the absence of accurate, objective data on the deer population trend and amount of damage and the lack of analysis necessary for decision-making. In 1990, the prefecture government began a preparatory investigation (Hyogo Prefecture Government 1991), but systematic monitoring of the population trend did not begin until annual fecal censusing on over 60 plots and gathering hunter reports were established in 1999 (Hyogo Prefecture Government 2000). In 1999, the Hyogo Prefecture Government implemented a sika deer management plan (Hyogo Prefecture Government 1999). To reduce damage to agriculture and forestry, the plan aimed to decrease deer numbers to half of those in 1999. The government estimated the deer population could be reduced to 33,500 in five years by harvesting 8,000 deer annually. However, this plan was not based on science. The actual number of deer harvested in 1999 was 9,000, exceeding the harvest proposed by the plan. Still, the deer density index and agricultural damage rose a little in 2000. In 2000, 9,700 deer were harvested, but the index and damage still rose a little in 2001. Finally, in 2002, projects were begun for controlling population and building protection fences.

31 The Management of Sika Deer Populations in Hyogo Prefecture

439

Scientific Approach to Population Management Management should be based on well-grounded information. To contribute to the management of sika deer we try to determine the factors influencing population dynamics of deer and damage to agriculture and to predict their trends. The success of a management plan depends on the validity of its formulation not only for the total amount of deer population and damage but also its distribution. Therefore, we need to be able to predict changes in spatial distribution, as well as the trends of total population numbers and amount of damage. Data and models for analysis of spatiotemporal dynamics are required. In addition, for practical management, predictor variables must be measured in advance, and necessary data must to be simple and low-cost enough to monitor continuously over a wide area. Using available monitoring data, we analyzed the relationships between sika deer density, hunting pressure, and damage to agriculture in Hyogo Prefecture from 1999 to 2001. We constructed a model to predict deer population trends, which included spatiotemporal effects, using regression analysis. Based on a model similar to the time series density-dependent model (Dennis and Taper 1994), we added predictor variables concerning neighboring density, harvest size, climate, and human land use. We selected effective predictors among these variables with a model selection method using information criterion (Mallows’ Cp (Mallows 1973) ). This process allows us to detect the factors which influence trends in population and damage (Sakata et al. 2001). Based on the results, we simulated the changes in deer density and damage according to variable harvesting plans, which differ in total amount of harvest and spatial design. Successful prediction of trends in the near future may support governmental decision-making. Finally, we examined the accuracy of our prediction with data from 2001 to confirm the current limitations of our results and to improve the model.

Study Area and Monitoring Monitoring Deer Density In Hyogo Prefecture, an index of sika deer density has been annually monitored since 1999 with 408 4 (NS) × 5 (EW) km plots placed over the entire prefecture area. The index is estimated according to a line-transect census of fecal pellet groups and reports from hunters. It is difficult and costly to count deer directly in this area because of steep topography and dense vegetation.

440

H. Sakata et al.

The feces census is undertaken in early November, immediately before hunting season. To reduce the influence of seasonal fluctuation of fecal decomposition rate, we conduct it at the same season every year. A census taker walks about a 5 km predetermined census route along ridges in the mountainous forest in each grid and counts the number of groups of feces within 1 m on both sides of the transect. Given the variation in fecal density on a small scale, the census routes should be as long as possible. The 5 km distance was selected as the distance a census-taker could reasonably trace in a day. We treated the number of fecal deposits consisting of 10 or more pellets per 1 km census route as an index of sika deer density. We selected the census route locations to be as representative of the grid vegetation as possible. Surveys were conducted on 119 grids in 1999, and on 60–70 in 2000–2002. Because of budget limitations, not enough investigations of feces density were conducted to estimate densities in all areas. We confirmed that the results of feces census were correlated with a block count method used in a neighboring area (n = 20, R2 = 0.539, p = 0.0002; Hamasaki et al. 2007). In addition, the number of observed deer per unit effort was calculated from reports submitted by individual hunters. The reports included information about the date and grid hunted and the number of deer that they observed and harvested in each grid and on each day. About 80% of hunters reported their results, a total of 18,688–21,268 man-days in 259–300 grids for seasons from 1999 to 2002.The hunter reports cover most of the deer habitat area in Hyogo Prefecture. These reports depend on optional cooperation by individual hunters and were not based on a strict standard that designated a standard hunting method, time, or area. To confirm the consistency between fecal censuses and hunter reports and to build a formula to convert observation per unit effort into feces density, we conducted a regression analysis with the following model: Feces density(/ Km) = a ×

( No. of observed deer )b . ( Hunting man − days )c

We estimated a, b, and c, because the relation might not be linear. The data were pooled by grids and hunting seasons and regressed on the feces density surveyed on corresponding grids and seasons. The result showed that the estimations from fecal censuses and from hunter reports were consistent to some extent (R2 = 0.61, n = 120). The conversion formula is Feces density(/ Km) = 2.89 ×

( No. of observed deer )0.73 . ( Hunting man − days)0.47

The standard errors of estimated parameters are shown in Table 31.1. We used feces density per 1 km as the index of deer density, because this is easy to confirm by forest transects. Numbers of observed deer per unit effort calculated from hunters’ reports were used after being converted into feces density by this formula.

31 The Management of Sika Deer Populations in Hyogo Prefecture

441

Table 31.1 The estimation of coefficients in the formula to convert hunter report data into fecal density.

Feces density(/ Km) = a ×

( No. of observed deer )b . ( Hunting man − days )c 95% confidence limits

Coefficient

Estimation

Standard error

Lower

Upper

a b c

2.896 0.727 0.468

0.179 0.159 0.179

2.541 0.413 0.114

3.253 1.040 0.823

Hunting and Nuisance Control In Hyogo Prefecture, the hunting season is from 15 November to 15 February. Nuisance controls are carried out at almost all times of year except during hunting season if permission from the prefecture is issued upon request of the individual suffering damage. Although the number of harvested deer has increased, the population of hunters showed a peak at 15,000 in 1977 and steadily declined to 6,000 in 2001 (Fig. 31.2). The average age of hunters has increased (60 years old in 2001). Hunting is unlikely to be an effective means of control for a widely increased deer population (Suzuki et al. 2003). According to hunters’ reports, the hunting pressure on sika deer was more intense in the grids where deer density was high (Fig. 31.3). Hunters were not systematically allocated to areas but tended to go to the high deer density areas.

Environment of the Study Area Land use and vegetation. In Hyogo, the human population has been high for a long time, and the remaining wild area is very sparse. Figure 31.4 shows a land-use map of Hyogo. The geometrical data were provided by Japan Integrated Biodiversity Information System (http://www.biodic.go.jp/english/J-IBIS.html). Urban areas, cultivated land, and forest are mixed in a fine mosaic. Of prefecture area, 9.7% is cultivated and 67.2% is forest. The forests were traditionally harvested for lumber, firewood, leaf mold, and other products. After World War II these were provided by industrial substitutes or imports, and biological forest products are no longer utilized. About 44% of forest area is plantation forests containing Japanese cedar (Cryptomeria japonica) and Hinoki cypress (Chamaecyparis obtusa). The plantations were actively cultivated in the 1960s and 1970s. As these densely planted trees have grown the understory has become dark and understory vegetation sparse. There are few available foods for deer in these mature forests that have not been maintained.

H. Sakata et al. 16000 14000

Trap

Gun

2000

1995

1990

1985

1980

1975

10000 8000 6000 4000 2000 0

1970

12000

1965

Number of licensed hunters

442

Air Gun

Fig. 31.2 Number of licensed hunters in Hyogo Prefecture from 1961 to 2001.

Fig. 31.3 A. Density index (feces density per 1 km line census) of sika deer in 1999. B. Harvesting pressure (number of harvested female deer/km2) in the 1999–2000 season.

Most remaining forests are altered communities consisting mainly of summergreen forests whose canopy layer is dominated by broad-leaved deciduous oaks (e.g., Quercus serrata and Q. variabilis) and forests dominated by an evergreen conifer, Japanese red pine (Pinus densiflora). In some forests, patches of dwarf bamboo species (e.g., Sasa veitchii and S. nipponica) cover the forest floors. Secondary succession has progressed in these forests; broad-leaved evergreen shrubs and subcanopy trees (e.g., Eurya japonica (Theaceae), Ilex crenata (Japanese holly), and Illicium anisatum (Japanese star anise)) have tended to increase in the understory.

31 The Management of Sika Deer Populations in Hyogo Prefecture

443

Fig. 31.4 Land use in Hyogo Prefecture.

In the following analysis, we adopted four broad land-use categories: urban (including residential, industrial, and commercial areas), cultivated, artificial forest (Japanese cedar or Japanese cypress plantations), and other forest. Climate. In Kobe city in mid-southern Hyogo Prefecture, the monthly average temperature is 5.2 °C in January (coldest month) and 27.5 °C in August (hottest month). The average precipitation per year is 1,264.7 mm and maximum snow depth is 20 mm. In Toyo-oka city in northern Hyogo, the monthly average temperature is 2.9 °C in February (coldest month) and 26.3 °C in August (hottest month). The annual precipitation is 1,987.9 mm and snowfall is 590 mm. These data are averages for 1971–2000. In the following analysis we use maximum depth of snow cover and sum of monthly averaged temperatures in a year, which were estimated on every 1 km2. These meteorological data were provided by the Japan Meteorological Agency (http://www.jma.go.jp/jma/indexe.html).

444

H. Sakata et al.

Analysis of Population Dynamics Methods of Regression Analysis for Population Dynamics We carried out regression analysis in order to construct a model that predicts change in deer population by clarifying of the factors that influence it. Our aim is to predict the change in deer density on every 4 (NS) × 5 (EW) km grid. Because the grids are an open system, population can change due to four factors: birth, death, immigration, and emigration. However, we cannot evaluate the specific factors. Only the total change rate can be estimated. We try to relate the total change rate to available variables of management, environment, climate, and land use. We used the time series density-dependent model (Dennis and Taper 1994) after modification as follows. To extend the basic model, variables representing effects of neighboring density (migration), harvesting, climate, and human land use were prepared as candidates for predictors (Table 31.2). The variables that we needed for spatial analysis were calculated in a geographic information system (GIS). We selected the effective predictors using model selection method with an information criterion, Mallows’ Cp. It is unclear what extent of spatial scale is appropriate for analysis to evaluate the effects of environment and harvesting. In this analysis, we prepared two scales of variables concerning harvesting, climate, and human land use to select one or the other. One scale is variables in the focal grid, and the other is the average or sum of the nine grids including the grids around the focal grid (Narrow and Wide in Table 31.2). In a model selection method using information criterion, the model with the lowest information criterion is regard as the best of all models consisting of any combinations of candidacy variables. In this analysis, any model including two variables representing the same factor in a different scale was excluded, because these variables have intense co-linearity that makes it difficult to interpret their ecological meaning. Our aim was not to adjust the formula to fit data by including as many variables as possible, but to build an ecologically reasonable model. Table 31.2 List of candidate variables used in regression analysis to predict the dependent variable rate of change of the sika deer density index. “Narrow” refers to variables in the focal grid and “Wide” to variables in the nine grids including the focal grid. Candidates for predictor variables: Deer density: Density of fecal groups in the focal grid (no./km) Neighboring deer density: Density of fecal groups in adjoined eight grids (no./km) Neighboring deer abundance: Index of adjoined eight grids density x forest area (km2) Reserved area: Proportion of game preservation area (%) No. of harvested females/forest area (no./km2) (Narrow or Wide) No. of harvested males/forest area (No./km2) (Narrow or Wide) Proportion of urban districts (%) (Narrow or Wide) Proportion of agricultural area (%) (Narrow or Wide) Proportion of plantation area/forest area (%) (Narrow or Wide) Temperature index Maximum snow depth

31 The Management of Sika Deer Populations in Hyogo Prefecture

445

The expanded model was modified as follows to carryout linear multiple regression: log(Dt;1/ Dt) = a + bD1+ c1F1+ c2F2 +.....+ cpFp where Dt is density index in year t, Fn is the prepared candidacy of predictors variable, and a, b, and cn are estimated coefficients using the method of least squares. In addition, to confirm the covariance structure of selected variables, we drew a path diagram.

Results of Regression Analysis for Population Dynamics By the model selection method, the following five variables were selected as factors predicting the rate of change in density index (Table 31.3, Fig. 31.5), as follows: 1. The negative coefficient of “Deer density” in the focal grid suggests a density dependent-effect. 2. The positive coefficient of “Neighboring deer abundance” is considered to represent the amount of dispersal of deer from high-density grids to low-density grids. 3. “No. of harvest females/forest area (Wide)” was selected rather than a male variable or a narrow range. The negative coefficient is considered to represent the intensity of harvesting effect on the population. The results suppose that harvesting males has less effect on population. And, to reduce local density, hunting pressure is best spread over a wide range. 4. “Proportion of agricultural area (Narrow)” had a positive coefficient. 5. “Proportion of urban districts (Wide)” was negative for the rate of change in the deer density index. The last two variables had opposite effects in spite of their covariance. They suggest that deer increase by feeding on agricultural products or are attracted to them. On the other hand, conceivably, the deer could reproduce less or avoided moving around in urban districts. Table 31.3 Estimated coefficients in the prediction formula for change in sika deer density index. Factor Estimation Standard error t P Intercept Deer density Neighboring deer abundance No. of harvested females/forest area (Wide) Proportion of agricultural area (Narrow) Proportion of urban districts (Wide)

0.269571888 −0.015476575 0.000001108 −0.170674043

0.09414754 0.00339019 0.00000046 0.06364442

2.86 −4.57 2.42 −2.68

0.0046 0.0001 0.0165 0.0078

0.745106076

0.32310708

2.31

0.0220

−3.340138311

1.62651066

−2.05

0.0411

446

H. Sakata et al.

Fig. 31.5 How selected factors contributed to the change in density index.

The predicting formula is: log(D1;tDt) = 0.270 – 0.0155 × Dt + 0.00000111 × NDAt – 0.171 × HFW + 0.745 × PAgN – 3.34 × PUW where NDA is the “Neighboring deer abundance,” HFW is the “No. of harvested females/forest area (Wide),” PAgN is the “Proportion of agricultural area (Narrow),” PUW is the “Proportion of urban districts (Wide).” These detected relationships between the population trend and selected factors can be interpreted in an ecologically reasonably manner. In a path diagram (Fig. 31.5) deer density affects the population trend in at least two ways. First, high deer density indirectly causes negative effects on the population through inducing more hunting or nuisance kill in areas of high density and, second, other density-dependent effects (direct arrow in the path diagram) conceivably act through decreasing food resources or motivating deer dispersal. The contribution of the density dependence by the first route (– 0.30) was quantitatively much less than the latter (– 0.44 × 0.18 = 0.08), suggesting internal effects within the population played a larger role than the distribution of hunting and nuisance control effort.

Analysis of Agricultural Damage Methods of Regression Analysis for Agricultural Damage The model predicting agricultural damage by sika deer was built by a method similar to that for population trend. Amount of damage was tallied not by grids, but by administrative districts (Fig. 31.6). Data by grids were retallied into 88 administrative

31 The Management of Sika Deer Populations in Hyogo Prefecture

447

Fig. 31.6 Proportion of cultivated area damaged by sika deer in 1999.

Table 31.4 List of candidate variables used in regression analysis to predict agricultural damage. “Narrow” refers to variables of area within a 2.5 km distance from the cultivated area, and “Wide” to area within a 5 km distance from the focal cultivated area. Neighboring deer abundance: index of density x forest area (km2) (Narrow or Wide) Proportion of urban districts (%) (Narrow or Wide) Proportion of agricultural area (%) (Narrow or Wide) Proportion of plantation area/forest area (%) (Narrow or Wide) Boundary between cultivated lands and forests (km) Accumulative length of protection fence (m)

districts. To detect the factors influencing damage value per area (hectare), regression analysis was conducted. Predictor variables chosen by model selection methods using Mallows’ Cp are listed in Table 31.4. In the model selection, the model including two variables representing the same factor at a different scale was excluded.

448

H. Sakata et al.

Results of Regression Analysis for Crop Damage By the model selection method, three variables were selected as factors predicting agricultural damage (Table 31.5, Fig. 31.7). 1. The positive coefficient of “Deer abundance” within a 2.5 km distance from the focal cultivated area showed that the intensity of damage depends on deer abundance. 2. The proportion of plantation area in forests within a 2.5 km distance influenced agricultural damage. This shows that even if deer abundance was at the same level, damage was greater in areas where the proportion of plantation forest was higher. This is thought to be the case because there tends to be little deer food in these forests. Most plantation forests were planted with Japanese cedar and Japanese cypress. The understories of these forests are dark with little vegetation, and the lower branches have been trimmed or have withered due to shading.

Table 31.5 Estimated coefficients in the prediction formula for amount agricultural damage (thousand yen/ha). Factor Estimation Standard error t P Intercept Neighboring deer abundance (Narrow) Proportion of plantation error (Narrow)

−1.550450602 0.000027557

1.22518082 0.00000625

−1.27 4.41

0.2076 0.0001

5.510880446

1.85359906

2.97

0.0034

Fig. 31.7 Factors affecting agricultural damage by sika deer.

31 The Management of Sika Deer Populations in Hyogo Prefecture

449

3. The length of the boundary interface between cultivated lands and forests also related to the amount of damage. This was probably because of the behavior of sika deer in this area, hiding in the forest to avoid exposing themselves on open lands in the daytime and coming out at night to feed on grassland or cultivated land. As predictors, the method selects variables within a 2.5 km distance from the focal cultivated area, rather than a 5 km distance. This fact suggests that deer abundance and environments within (at the most) 5 km or shorter distance from the focal cultivated area influence the amount of agricultural damage by deer. This variable needs to be analyzed at a finer scale, but we could not do so because the spatial resolution of our data was limited. These detected relationships between amount of agricultural damage and selected factors also can be reasonably interpreted. The results suggest that landscape arrangement is important to reduce damage, as well as to achieve population control. The predicting formula is: Dam = –1.55 + 0.0000275 × DA25 + 5.51 × PP25 where Dam is amount of agricultural damage (thousand yen/ha), DA25 is the index of deer abundance within 2.5 km distance of focal cultivated land, PP25 is the proportion of plantation area in forest within 2.5 km distance of focal cultivated land.

Simulations To plan for the optimal amount of deer harvesting, multiple alternative plans should be examined. The predicted outcomes must be shown to all concerned parties in an understandable form. Using the predicting formulas described above, we simulated changes in sika deer density and agricultural damage according to the following four hunting plans: (1) harvesting no deer, (2) harvesting 10,000 deer (approximately the same number and distribution as 2001), (3) harvesting 20,000 deer distributed density-dependently, and (4) hunting 20,000 deer distributed depending on the increase rate of population growth (Fig. 31.8 shows the case of (2) and (3)). The result, if no deer are hunted, is that the population will increase in most of the area. Harvesting 10,000 deer, the deer density and agricultural damages will not be reduced. Harvesting 20,000 deer density-dependently will reduce the population. However, harvesting 20,000 for one year will not reduce the deer population so greatly that it cannot recover (Sakata et al. 2002). In decision-making, the reliability of simulation is important. We examined our predictions of change in the sika deer density. Figure 31.9 shows the difference between the predictions using data in 1999–2000 and the results of monitoring in 2001. The difference tends to be small in the grids where many man-days of

450

H. Sakata et al.

Fig. 31.8 Comparison of two hunting plans for simulated changes in sika deer density and agricultural damage. For Plan 2, hunt 10,000 deer: (a) predicted change in rate of deer population increase; (b) predicted agricultural damage. For Plan 3, hunt 20,000 deer: (c) predicted change in rate of deer population increase; (d) predicted agricultural damage.

31 The Management of Sika Deer Populations in Hyogo Prefecture

451

Fig. 31.9 (a) Prediction of change in deer density using data from 1999–2000 and observations from monitoring data in 2001. The scale is natural logarithm of changing rate of density index. We predicted that the index will increase 7.25% on average, and the observed average change was 7.90% increase. (b) Histogram of errors in the above predictions.

Fig. 31.10 Relationship between number of reported man-days of hunting and prediction errors by grid. These methods need to be understandable to every resident, as well as accurate to the extent possible.

hunting are reported (Fig. 31.10). Note the convergence to zero when the man-days increase. In summary, our predictions can be the basis for decision-making in deer population management for Hyogo Prefecture, although they are not complete. However, no predictions more reliable than ours are available. Even though these predictions have limitations, we feel it is important to publish them for use in management. These methods need to be understandable to every resident, as well as

452

H. Sakata et al.

accurate to the extent possible. If local residents know the potential of predictions that may encourage more accurate prediction in the future and, simultaneously, better appreciate the necessity for the data that they provide. Not only researchers, but all concerned parties, should consider factors affecting human-deer relationships on the same objective basis. Scientists are required to show the methods that contribute to social decision-making and reaching policy agreements. We recognize the necessity of simulating other variables concerning deer management such as status of natural vegetation, number of road-killed deer, production of venison, etc. An inclusive balance of costs and benefits should be examined. Our work proposes a framework for deer population control as a starting point for appropriate systematic management in the future.

Literature Cited Dennis, B., and M. L. Taper. 1994. Density dependence in time series observations of natural populations: Estimation and testing. Ecological Monographs 64:205–224. Hamasaki, S.-I., M. Kishimoto, and H. Sakata. 2007. Evaluation of three monitoring methods for sika deer population density: Block count, pellet group count and sighting per unit effort. Honyurui Kagaku (Mammalian Science) 47:65–71. (In Japanese.) Hyogo Prefecture Government. 1991. Report of research on sika deer status in 1990. Kobe, Japan. (In Japanese.) Hyogo Prefecture Government. 1999. Sika deer conservation and management plan. Kobe, Japan. (In Japanese.) Hyogo Prefecture Government. 2000. Report of research on sika deer status in 1999. Kobe, Japan. (In Japanese.) Mallows, C. L. 1973. Some remarks of Cp. Technometrics 15:661–675. Sakata, H., S.-I. Hamasaki, M. Kishimoto, H. Mitsuhashi, A. Mitsuhashi, M. Yokoyama, and M. Mitani. 2001. The relationships between sika deer density, hunting pressure and damage to agriculture in Hyogo Prefecture. Humans and Nature 12:63–72. (In Japanese.) Sakata, H., S.-I. Hamasaki, H. Mitsuhashi, M. Yokoyama, and M. Mitani. 2002. The examination of plural scenarios for management of sika deer population in Hyogo Prefecture. Humans and Nature 13:21–28. (In Japanese.) Suzuki, M., H. Sakata, and T. Tanaka. 2003. Dynamics of hunter population in Hyogo Prefecture. Humans and Nature 14:33–41. (In Japanese.)

Chapter 32

Management Strategy of Sika Deer Based on Sensitivity Analysis Shingo Miura and Kunihiko Tokida

Abstract We conducted sensitivity analysis of sika deer (Cervus nippon) based on demographic parameters obtained from a variety of previously studied populations and applied a life-stage matrix model. Sensitivity analysis by a stochastic model revealed that adult female survival was the most effective in the population growth rate; this was also supported by an analytical model. Survival of young and fecundity of adults had an impact on the growth rate in a rapidly increasing population. Based on the life history characteristics of sika deer, long-term warm winters were the major factor causing recent increases of population sizes and range expansion, which was accompanied by enlargement of habitat due to drastic changes in human society. Female hunting was effective for controlling these populations, but careful management in the newly proposed scheme is required.

Introduction The sika deer is the most familiar and widely distributed large mammal in Japan. In recent years, the population has drastically increased its range and size in almost all parts of the major islands (Biodiversity Center of Japan 2004). From 1979 to 2003, it increased its spatial distribution more than 70% in terms of the 5 × 5 km grid bases (Fig. 32.1). Consequently, the deer have caused serious damage to agricultural crops and forest plantations in various parts of their range. The recent total area of damage rose from 20,000 to 50,000 ha with a peak cost exceeding five billion yen in 1995 (Ministry of Agriculture, Forestry and Fisheries 2004). Prefectural governments have conducted large-scale population controls and constructed protection fences. Although the total number of culled deer increased from around 53,000 in 1992 to 135,000 individuals in 2000, the area of forestry and agricultural damage also increased. Aside from economic damage, bark stripping or browsing on natural vegetation, including rare and endangered plant species is also widespread and urgently requires a plan for conservation of forest ecosystems and landscapes, especially in the national parks.

D. R. McCullough et al. (eds.), Sika Deer: Biology and Management of Native and Introduced Populations, DOI: 10.1007/978-4-431-09429-6_32, © Springer 2009

453

454

S. Miura, K. Tokida

Fig. 32.1 Spatial distribution of sika deer in Japan according to the Basic National Survey on the Natural Environment (Biodiversity Center of Japan 2004). Dark grid represents range where sika deer occurred in 1979 and 2002, light one in 2002, and mid-tone inhabited in 1979 but disappeared in 2003.

Why does the population of sika deer increase its size and range despite large-scale control? What factors are responsible for such population growth? How should we manage and control the population based on its ecology and life history? Life history and demographic studies are of considerable importance not only for understanding population dynamics of the species but also for implementing conservation and management of the population. Demographic vital rates such as age-specific mortality or fecundity are the basis for evaluating the current status of the population and for predicting population growth (Caughley 1994). Researchers have attempted to estimate these rates in many ways, for example, by recovering shooting samples, constructing life tables, and conducting long-term direct observations in many areas. As a result, spatial and temporal variations in vital rates are found in response to habitat conditions, population sizes and densities, and local history of population dynamics. We also recognize that a high level of uncertainty in vital rate estimation may be involved due to sampling and measurement errors. Despite spatial and temporal variations and uncertainty, an integrated analysis of vital rates is needed as a basis for constructing a proper management strategy for sika deer that includes the best ways to regulate population growth.

32 Management Strategy of Sika Deer Based on Sensitivity Analysis

455

Sensitivity analysis, which highlights the effect of changes in vital rates in each life-stage on population growth rate (λ), is a strong tool for analyzing natural populations. Sensitivity, defined as the absolute change in λ with infinitesimal absolute change in vital rate (Caswell 1989), has been widely used in conservation to identify ways to increase the populations concerned (Crouse et al. 1987; Brault and Caswell 1993; Crowder et al. 1994; Silvertown et al. 1996; Grand and Beissinger 1997; Benton and Grant 1999). It can, on the other hand, be equally valuable for management options to effectively decrease the populations (Shea and Kelly 1998; Citta and Mills 1999). A general form of the analysis has been applied to deterministic models of population growth using stage- or age-based matrices. Elasticity, defined as the proportional change in λ with proportional changes in vital rates (Caswell 1989), is also calculated. However, both sensitivity and elasticity calculated from point estimates of mean or representative values are not appropriate because they ignore potential variations and fluctuations in such rates, which usually contain a variety of errors (Benton and Grant 1996; Wisdom and Mills 1997; Crooks et al. 1998; Gaillard et al. 1998; Mills et al. 1999). Instead of this technique, recent studies have explored the use of an alternative approach, logistic or simple regression, to analyze the sensitivity of stochastic models (McCarthy et al. 1995; Wisdom and Mills 1997; Wisdom et al. 2000; Cross and Beissinger 2001; Mills and Lindberg 2002). This is a simulation-based method, using life-stage simulation analysis, to estimate the potential effect of each vital rate on λ in the face of uncertain and changing, or even “best guess” vital rates (Wisdom and Mills 1997). The relative importance of each vital rate is indicated by the value of its standardized regression coefficient (Wisdom et al. 2000). To establish sound management strategies for sika deer populations, we assess the effects of life history traits on population growth by reviewing vital rates of sika deer populations obtained so far in various parts of Japan. The purposes of this chapter are to propose a classification of sika deer life-stages based on evaluations of its life history and vital rates, to use a simulation-based regression approach, together with conventional approaches, for directly evaluating vital rates effects on population growth, and to examine factors responsible for recent increases in the populations.

Methods Vital Rates of Sika Deer We used all available papers and reports, both published and unpublished, to evaluate the life history of sika deer. The vital rates were collected under a variety of habitat conditions and sampling methods and, thus, include sampling and measurement errors, and spatial and temporal variation due to environmental effects. The range of each vital rate was defined as the lowest and highest values in the

456

S. Miura, K. Tokida

study samples, but plausible values were also introduced from other cervid species when the samples were insufficient to define. Sika females usually give birth to single offspring, and multiple births (twins and triplets) are rare, estimated at less than 2% (Iimura 1980; Sadleir 1987; Takatsuki and Suzuki 1990; Hayssen et al. 1993). Thus litter size can be regarded as 1.0. Agespecific fecundity rate was defined as the proportion of fertile females to all females in each age class. This rate was widely monitored as pregnancy rate in many study samples by autopsy of individuals shot in winter. The fecundity rate was judged to be very close to the pregnancy rate in this season. So, we converted the age-specific pregnancy rates into the life-stage fecundity. Age-specific survival rate has been estimated by constructing time-specific or horizontal life tables based on an age distribution by assuming a stable or stationary population. The age distribution obtained either from carcasses or shooting samples was converted directly or indirectly (smoothing the frequency) into a dx or lx column (Caughley 1977). However, this assumption of a stable age distribution cannot be valid for temporally fluctuating populations (Caughley 1977; Menkens and Boyce 1993). Substantial yearly variations in survival rate were reported in many wild cervid populations (for example, McCullough 1979; Unsworth et al. 1999). Despite this limitation, demographic parameters obtained by this method may be used as ancillary data in stochastic modeling (Menkens and Boyce 1993). Then, we arranged stage-specific survival rates based on age-specific data obtained by constructing life tables. Aside from this approach, an attempt to estimate age-specific survival rates has been made by longterm direct observations on cohorts in some study areas (Uno et al. 1998; Minami 2003 personal communication; see other chapters in this book). The oldest age at death of free-ranging sika deer populations was recorded by age determination of shooting or natural death samples in some study areas. The longevity was longer in females than in males with a maximum 20 years for females and 16 years for males at Nara Park (Ohtaishi 1975).

Matrix Construction We created a Leslie-Lefkovitch matrix (Caswell 2001) with a life-stage structure that we propose here for sika deer populations. This matrix, shown below, was composed of values of Fi, Si, and Pi. We parameterized these values using a prebirth pulse approach (Caswell 1989) and a one-year time step because a prebreeding census (early spring) was often conducted in many study areas. Under the prebirth pulse model, Fi (each element of the top row of the matrix) equals the product of (S0) × (mi), where mi is age- or stage-specific fertility probability and S0 is survival probability of calves. S1,2 were survival probabilities for yearlings and juveniles, respectively. The annual probabilities of adult and senescence stages are Pa and Ps, respectively. Sa and Ss are annual transition probabilities from juvenile to adult, and from adult to senescence, respectively. The density-dependent effect is not considered in this model.

32 Management Strategy of Sika Deer Based on Sensitivity Analysis

0 S  1 0 0   0

0 0 S2 0 0

F1 0 0 Sa 0

Fa 0 0 Pa Ss

457

Fs  0  0 0  Ps 

Sensitivity and Regression Analysis The total variation in estimates of demographic parameters, including spatial and temporal variations and sampling variation (Wisdom et al. 2000; White 2002), is unclear for the sika deer population. For inputting the variation of vital rates, we defined the range of each parameter according to the highest and lowest values obtained so far in the study samples. Although a variety of distributions (uniform, beta, log-normal, etc.) can be applied for vital rates (Wisdom et al. 2000; Cross and Beissinger 2001), to simplify our approach we used a set of random combinations of vital rates chosen independently from uniform probability distributions. The dominant eigenvalue (λ) of the matrix was calculated in 500 matrix replicates of vital rates and matrix elements resulting from randomly selected values. Then, for the purpose of determining which life history parameter was most likely to affect λ, we used simple linear regression to assess the relationship between each vital rate and λ (Wisdom and Mills 1997). Logistic or nonlinear regression can be applied (McCarthy et al. 1995; Horvitz et al. 1997; Cross and Beissinger 2001), but we used simple regression as the first step. We compared the regression coefficients (R2) as a measure of relative effect of each vital rate on λ, which is approximated by the product of sensitivity2 (Brault and Caswell 1993). The slope of the line equals the analytical sensitivity (Mills and Lindberg 2002). In addition to this, the analytical sensitivities (sij) and elasticities (eij) of the population were calculated in each matrix composed of the highest, midpoint, and lowest values according to the following equations (Caswell 1989, 2001): Sij =

vi wi ∂λ = ∂α ij < W,V >

where vi and wi refer to the ith and jth elements of the age- or stage-specific reproductive value (V) and stable stage distribution (W) vectors, respectively, and is the scalar product of W and V. The elasticity of λ to element aij is simply the sensitivity rescaled to account for the magnitude of both λ and the matrix element: eij =

∂logλ . aij ∂λ = λ ∂α ij ∂ log aij

458

S. Miura, K. Tokida

Thus, we compared the ranking of demographic parameters estimated by regression analysis with the analytical values obtained from deterministic models.

Results Life-Stages of Sika Deer Although body dimensions and weight of sika deer associated with age varied considerably with geographical location, no distinct differences were detected in growth patterns. Judging from data on body weight and age in some study areas, body growth curves may generally be divided into three phases: increasing rapidly in weight until three years, stabilizing until about 10 years, and gradually declining thereafter. Like many other cervid species (for example, Clutton-Brock et al. 1982; Gaillard et al. 1992), sexual maturity of sika deer appeared to depend on body weight. However, pregnancy was not recorded for calves irrespective of their size in the wild populations. Immunological assay showed the onset of ovulation was unlikely to occur in this age (Yamauchi et al. 1997). Age at puberty, which appeared to vary with habitat conditions, ranged from 1.5 to 3.5 years old. Accordingly, age at first reproduction varied between two and four years old. The pregnancy rates of yearlings and juveniles varied largely from area to area, which was probably the reflection of habitat quality. It was well known in some study areas that the pregnancy rate increased as body weight increased regardless of age (Takatsuki 1992; Asada and Ochiai 2001). The first and subsequent pregnancies, therefore, were associated with body mass rather than age. Some critical thresholds of body weight to initiate reproduction were reported, 40 kg for Kinkazan Island (Minami 2003 unpublished data) and 43 kg for Nakanoshima Island (Kaji 2003 unpublished data). The annual probability of survival (or mortality) varied with age in sika deer. Calves experienced the highest rate of mortality and were particularly sensitive to environmental conditions. Calf mortality in ungulate species is known to occur mainly within the first few weeks and during the first winter (Guiness et al. 1978; Clutton-Brock and Albon 1982; Gaillard et al. 1993). The former mortality is probably linked to mother’s condition or experience (Ohnishi 2003 unpublished data), whereas the latter is closely associated with the rate of maturity and body weight prior to winter. Survival of yearlings was lower than juveniles and adults, because they are still maturing and are affected strongly by their size as calves. Survival of juveniles is sometimes low or similar to that of adults and is relatively insensitive to external conditions, but pregnancy success depends on maturity. Survival and fecundity rates of senescence are different from those of younger adults. These differences result from a decline of body weight probably due to the effect of tooth wear (Ohtaishi 1975; Takatsuki 2000). Based on the above evaluation, we propose here to classify the life-stages of sika deer, which was partly combined with age, to construct a population model

32 Management Strategy of Sika Deer Based on Sensitivity Analysis

459

Fig. 32.2 Life-stage diagram used in the stochastic and deterministic analysis of sika deer populations. See methods and results for details on structure and parameterization.

Table 32.1 Range of parameter estimates of sika deer used in the stochastic and deterministic models. (Life history data from Ohtaishi 1975; Miura 1991; Takatsuki et al. 1991; Koizumi 1992; Takatsuki 1992, 1998; Kaji 1997; Asada and Ochiai 2001; Uno et al. 1998; Okada 2003 unpublished data; Minami 2003 unpublished data; Tatsuzawa 2002 unpublished data). Symbol Demographic parameter Minimum Midpoint Maximum S0 S1 S2 Sa Ss Pa Ps M1 ma ms

Calf survival Yearling survival Juvenile survival Transition to adult Transition to senescence Adult survival Senescence survival Juvenile fecundity Adult fecundity Senescence fecundity

0.30 0.50 0.60 0.65 0.05 0.70 0.05 0.10 0.40 0.30

0.55 0.68 0.75 0.82 0.08 0.84 0.10 0.35 0.70 0.58

0.80 0.85 0.90 0.98 0.10 0.97 0.15 0.60 1.00 0.85

(Fig. 32.2). The vital rates and their ranges that we used here for calculations are listed in Table 32.1.

Sensitivity by Regression Analysis Of 10 combinations of regressions between demographic parameter and λ, the results of life-stage simulation analysis for six parameters are illustrated (Fig. 32.3). Regression analysis indicated that adult survival accounted for the most effective vital rate in variation of the population growth. Calf survival was the next important parameter affecting λ. However, adult survival was about four times as important as calf survival. Survival of both yearlings and juveniles showed low values of the slope although the values of R2 were low. Fecundity of adult stage showed weak impact on λ, but that of the juvenile stage, which was age at first reproduction, accounted for no variation in λ. All other vital rates accounted for almost no variation in λ.

460

S. Miura, K. Tokida

Fig. 32.3 Life-stage simulation analysis for 500 replicates of a Leslie-Lefkovitch matrix model, which were selected randomly from the range of demographic parameter values (Table 32.1), and the resulting population growth rate (λ). Sensitivity of each of the six parameters was expressed by simple linear regression.

Comparison of Analytical Sensitivity Conventional sensitivity and elasticity were calculated according to the highest, midpoint, and lowest value of vital rates found among study population samples (Fig. 32.4). The population growth rate changed remarkably with the level of vital rates. Highest elasticity was associated with adult survival across the range of vital rates, indicating that λ was consistently affected by changes in adult survival. In the high growth rates, survival of yearlings and juveniles were also important. Calf survival that was linked with fecundities for all stages, as well as adult fecundity, may have a great impact on changes in λ. These results indicated strong agreement between the sensitivity and elasticity calculations and the regression analysis in identifying the vital rate having the greatest effect on λ.

32 Management Strategy of Sika Deer Based on Sensitivity Analysis

461

Fig. 32.4 Conventional sensitivity and elasticity calculated from the deterministic population model for the minimum, midpoint, and maximum values of parameters.

Discussion Life-Stage Importance Although sika deer, like other large ungulates, have highly age-structured populations, we adopted the stage-based population model to calculate sensitivity. The main reason for using this model is that it is applicable to free-ranging populations to monitor survival and fecundity rates by direct observations because we can count the number of individuals categorized according to life-stages. The fundamental assumption is that all individuals within a stage are subject to the same survival and fecundity schedules. There is no particular relationship between stage and age, but a strong link is identified for young stages. Life histories of mammals and birds can often be represented in terms of three life-stages: newborn, juvenile, and adult. However, especially in large ungulates, four categories of life-stage are very common, because both yearlings and juveniles are easily distinguished from adults.

462

S. Miura, K. Tokida

In sika deer, calves, yearlings, and juveniles were readily distinguishable from adults by their sizes. For sensible use of the stage-based model, more detailed information on lifestage of the sika deer is needed. The three-year old animals were combined with the adult stage in this model, but it appeared to be appropriate that this age class could be separated as adolescent from adult stage for the populations with low growth rate, high density, and low quality habitat, such as Kinkazan and Nakanoshima Islands, because the vital rates of three-year old individuals were different from the adult stage (Minami et al. 2003 unpublished data). Furthermore, we have divided the final stage into adult and senescent stages. The senescent stage of more than 10 years old was evident in the increase in mortality and decline in fertility in sika deer (Ohtaishi 1975; Takatsuki 1992). Senescence in reproduction and survival has been recognized in large ungulates (Skogland 1988; Gaillard et al. 2000; Ericsson et al. 2001; Loison et al. 2002). Despite the importance of the senescent stage, there were, in general, no visual criteria to distinguish senescent adults from younger adults. When applying this model to wild populations, it was necessary to estimate the proportion of senescence based on the proportion of other stages in the age-structure of the population. Since Nelson and Peek (1982) pioneered the method of evaluating life-stage importance to measure the sensitivity of population for elk (Cervus elaphus) using the Leslie-Lefkovitch matrix, this approach has become increasingly popular in the studies on ungulate populations (Escos et al. 1994; Milner-Gulland 1994; Gaillard et al. 1998). Matsuda et al. (1999) initially used this sort of matrix for predicting a Yezo sika deer population. However, their matrix, consisting of only two stages of both sexes and incorporating hunting pressure, was too simple to analyze life history. We believe that our categories of life-stages are a useful, fairly robust first step in a wider framework of model construction for life history and population dynamics of sika deer.

Sensitivity in Sika Deer Populations Population management requires particular actions that are most likely to change the population dynamics of the target species. Sensitivity analysis is a strong tool for this purpose. Several types of sensitivity analysis approaches are now available, from traditional analytical to simulation-based models (Mills and Lindberg 2000). We used two types of approaches to sensitivity analysis of sika deer populations: analytical elasticity analysis and life-stage simulation analysis. As a result, there was a strong correspondence between the two analyses. The simulation-based approach revealed that survival of adults had the greatest effect on population growth and that survival of calves was of secondary importance. Survival of yearlings and juveniles and fecundity of adults had a weak correlation with λ, whereas no correlations were detected for all other demographic parameters. It was notable that fecundity of juveniles, which was age at first reproduction, accounted for no variation of λ.

32 Management Strategy of Sika Deer Based on Sensitivity Analysis

463

In contrast to the simulation-based approach, elasticity of analytical analysis varied largely with the parameter values used in the matrices. Adult survival was consistently the most important variable regardless of the parameter value, but survival of yearlings and juveniles, and fecundity of adults were important in the higher level of values. In other words, they had greater impact on population growth in good years. The elasticity of adult fecundity was combined with calf survival as an element of the matrix in this analytical model. Changes in elasticity according to the level of parameter values resulted from the difference in the range of vital rates (Mills et al. 1999). It is reasonable to assume that vital rates seldom change by equal amounts in nature. Despite minor differences between the two approaches, adult survival was the prominent variable influencing the population growth in sika deer. In general, many demographic studies of large mammals have shown that adult survival has higher elasticity than juvenile survival and fecundity (Nelson and Peek 1982; Ecos et al. 1994; Walsh et al. 1995; Crooks et al. 1998). This conclusion is in accord with our results. Ecological theory related to life history strategy suggested that adult survival had the most important effect on population growth for vertebrates with relatively low mortality, long lives, delayed maturity, low fecundity, and iteroparity (Pianka 1970; Read and Harvey 1989; Heppell et al. 2000). Contrary to this suggestion, Eberhardt (1977) and Fowler (1987) pointed out that under increasing density of the population, a negative effect on population growth would occur in following order: juvenile survival, age at first reproduction, adult fecundity, and adult survival. Gaillard et al. (1998, 2000) suggested that population dynamics of large ungulates was strongly affected by juvenile survival especially under a combination of stochastic environmental variation and density dependence. Indeed, juvenile survival was likely to be more sensitive to temporal variation and density dependence than adult survival. Many studies on ungulate populations indicated that adult survival was less variable than juvenile survival, probably because adult survival buffered against temporal environmental variation (Skogland 1985; Festa-Bianchet et al. 1995; Singer et al. 1997; Gaillard et al. 1998, 2000). Although our methods took account of such variations to some extent, the overall process of population dynamics containing temporal variation and density dependence was not fully evaluated. To determine which demographic variables affect population dynamics, we need further studies on the relationship between temporal variation in vital rates and its sensitivity. Sensitivity analysis cannot be the sole tool to evaluate demographic parameters (de Kroon et al. 2000; Caswell 2001). It may, however, give us a robust guideline to identify the life-stages which should be targeted to manage sika deer populations. Like other ungulate populations (McCullough 1979, 1999; Fowler 1981; Kie et al. 1983; Bowyer 1999; Gaillard et al. 2000), density dependence is likely to occur for sika deer populations, especially at population sizes close to carrying capacity (McCullough 1979; Clutton-Brock et al. 1982; Fowler 1984). Density dependence is usually presumed to have its greatest effect on fecundity through food resource availability. Environmental factors, on the other hand, have a greater influence on the survival of young than adults when the populations are close to their carrying capacity (Geist 1971; Clutton-Brock et al. 1982; Picton 1984;

464

S. Miura, K. Tokida

Skogland 1985; Ecos et al. 1994). Density dependence in ungulates often interacts with other variables and density independent factors, such as climate, and thus provides varying results. High adult mortality was observed for a sika deer population near carrying capacity. In 1984, when mass mortality occurred due to harsh winter and food shortage on Kinkazan Island, we recovered a total of 265 carcasses and estimated overall mortality rate of more than 40% of the population. Of 132 female carcasses, calves and yearlings accounted for 13.8%, whereas adults and senescents accounted for 61.4% and 24.5%, respectively (Takatsuki et al. 1991). Mortality was biased toward prime-aged stages relative to the proportion of age-structure. We need longterm studies on the density dependence of sika deer populations to detect changes in fecundity, survival, and recruitment (McCullough 1999; Minami 2003 personal communication). Indeed, a model incorporating density dependence is needed for such populations. However, the model we used here is applicable to almost all sika deer populations because their densities are thought to be considerably lower than carrying capacity.

Factors Causing Current Increase of Population and Range Expansion Sensitivity analysis showed the relative importance of vital rates to population growth. Our model demonstrated that adult survival had the strongest impact on λ, followed by calf and yearling survival. Many factors, such as warm winters, low hunting pressure, reduction of human activities, changes in habitat qualities, etc., appear to be responsible for the recent rapid increase in spatial distribution and population size of sika deer in Japan. Among these factors, warm winters are most likely to be the dominant factor, because it is not a site-specific but rather, an archipelago-wide phenomenon. Warm winters may both directly and indirectly influence the survival of females, regardless of life-stage. Direct effects promoting survival of female sika deer result from reduction of cost of movement and foraging in deep snow and improvement of physical conditions during energetically expensive seasons. Warm winters also caused temporal and spatial changes in plant phenology; in particular, early timing and longer duration of growth substantially increase the survival of deer and improve fecundity as well (Picton 1984; Mech et al. 1987; Post and Stenseth 1998, 1999). Under the strong effect of density dependence, responses of vital rates to warming have varied with environmental conditions, sex, and life-stage (Post and Stenseth 1999). However, winter mortality of adult and young females generally decreases in many ungulate populations with low density or no distinct effect of density dependence. In Japan, the climate has drastically changed since the mid-1980s. Long-term trends of temperature anomalies recorded in 17 meteorological stations showed that

32 Management Strategy of Sika Deer Based on Sensitivity Analysis

465

the temperature has rapidly increased especially in winter (Fig. 32.5). Furthermore, snowfall also drastically decreased to about half that of 1975–1985 (Fig. 32.5). Evidently, warm winters have been occurring continuously for about 20 years all over Japan. The date of cherry blooming, which is a symbol of the coming of spring to Japanese, was 4.2 days earlier, and that of leaf tint of Japanese maple in autumn was 15.6 days later on average in 2005 than in the previous five decades (Japan Meteorological Agency 2007). According to meteorological stations, no snow depth of more than 150 cm has been recorded since 1985, but before that it had been recorded about 1.5 times in 10 years in northern parts of Japan. Winter severity and snow accumulation caused malnutrition, directly influencing the survival of deer due to restricted mobility and limitations on forage intake. Mass mortality, in which one-third to one-half of the population died, was reported to occur after a severe winter with heavy snow, such as in Kinkazan Island and the Nikko area in the spring of 1984 (Maruyama and Takano 1985; Takatsuki et al. 1991). Mass mortality due to heavy snow was also recorded in 1879 and 1903 in Hokkaido, where almost all populations of Yezo sika deer were nearly extinct thereafter (Inukai 1952). Historically, the population of sika has been regulated by repeated heavy snow accumulation in Japan. Continuing warm winters have greatly raised the survival rate of sika deer, making it a major factor in promoting the growth rate of deer populations. Current overpopulation and range expansion of sika deer may have been accelerated by the warm winters of the late 1980s and early 1990s. According to comparison of spatial distributions between 1979 and 2003, remarkable range

Fig. 32.5 Long-term trends of temperature from 1898 to 2003 in Japan (Japan Meteorological Agency 2004). Changes in average frequency of temperature anomalies against normal (1971–2000 average) recorded in 17 meteorological stations. The dark and light lines represent the above and below anomalies, respectively. The thick lines indicate five-year running mean. Arrows indicate occurrence of mass mortality of sika deer populations recorded in Japan (see text).

466

S. Miura, K. Tokida

expansions were observed especially in Hokkaido, Tohoku, and Hokuriku regions (Fig. 32.1), which were all high snowfall areas in the past. Hunting pressure directly reduces the survival of wildlife. The recent harvest level of sika deer has been very high, exceeding 100,000, of which about half were taken on Hokkaido Island (Fig. 32.6). Sika females had been protected by law for a long time, with some exceptions in specific areas where hunting was permitted to prevent nuisance behavior. However, since 1999 hunting females has been permitted widely in the areas under the Specific Wildlife Management Scheme (SWMS). Currently, females account for about 20–30% of all harvest. This ratio substantially influences the trend of population growth. Yet despite the strong impact on the population, there were no managed areas showing a distinct trend of population decline. Apparently, it was too little too late for this policy to achieve population control. Hunting pressure was probably too low relative to the population size,

Fig. 32.6 Changes in harvested numbers of sika deer (a), and number per hunter (b) in Japan according to annual hunting records.

32 Management Strategy of Sika Deer Based on Sensitivity Analysis

467

especially in the initial phase of the population increase in the late 1980s. Unfortunately, sound management practices were not established, and lack of countermeasures based on evaluations of population and other trends appear to have led to critical status. In relation to range expansion, it is notable that with advancing age of farmers, relocation to cities and abandonment of farming have proceeded rapidly in the low elevation mountain regions in Japan. The number of farming families has drastically decreased by about one-half in the past two decades (the total agricultural population is less than 10% of Japan’s population), and 30% of local rural communities have disappeared. This structural change in the agricultural society has occurred widely, especially in the western part of Honshu (Senda et al. 2002). Reduction of human activities has weakened the ability of local communities to prevent wildlife invasion. Consequently, wildlife damage to agricultural products has accelerated, creating a vicious cycle. As a result, a total of over 3,900 km2 of cultivated areas, including paddy fields and orchards has been abandoned in the past five years, which is equivalent to the total cultivated area of the major agricultural prefectures of Ibaraki and Niigata. In addition, about 80% of forest plantation areas have been abandoned after clear-cutting without replanting because of high costs and shortages of manpower. Other previous plantations have reverted to natural forests. Hence, space for wildlife habitat has now expanded. New habitat usually provides rich foods with high nutritional value for sika deer. Judging from spatial distribution, sika deer originally occurred in ecotone ecosystems mixed with broad-leaved forests, such as Quercus spp. in the low elevation mountainous areas. Sika deer were more like neighbors of villagers than pure mountain residents isolated from villages. So, they were very familiar with people and nuisance behavior was common. Without a doubt, the drastic change in mountain village society has provided new habitat to wildlife, particularly to sika deer, wild boar (Sus scrofa), and Japanese macaques (Macaca fuscata). The latter two species have also shown rapid range expansion. According to our regression analysis, fecundities of all stages had less effect on this population growth than did survival rates. A weak regression with low slope was detected for the fecundity of adults. Analytical analysis, in contrast, revealed that the effect of fecundity was relatively high in the populations with high growth rates, while fecundity of all stages was negligible in the low and middle population growth rates. These results suggest that high fecundity might promote the growth rate in the populations showing high growth rates under good years. Habitat expansion with good quality of foods was likely, of course, to influence the survival, but was more strongly associated with improvement of fecundity. The factor of habitat enlargement in relation to human social change, therefore, was secondary in triggering overpopulation of sika deer. In conclusion, the sika deer population has increased in size through increased survival rate due to long-term continuing warm winters and, in turn, a high fecundity rate has resulted from habitat enlargement and access to better food condition due to drastic social changes that have occurred in the past two decades in Japan.

468

S. Miura, K. Tokida

Hunting pressure was too low early on in this expansion and still is insufficient to control the current, very high population.

Recommendations for Management With an alteration of the Wildlife Protection Law, a Specific Wildlife Management Scheme (SWMS) was proposed and established by the Ministry of Environments in 1999. Female hunting has been permitted in specific areas under the provisions of this scheme. Prefectural governments have the right to set the number of deer that can be hunted per day. This scheme is now very popular, having been adopted in 32 prefectures, almost all having sika deer populations. Large-scale female hunting is common all over Japan and is strongly expected to control population size. Our analysis of life history of sika deer, indicating that adult female survival had the greatest impact on the population growth, supported, in part, the management option of this scheme. Undoubtedly, female hunting has profound effects on the trends and structure of a population, and is a quick and easy option to decrease or control population size. Indeed, it is reasonable and necessary to adopt this option in many regions confronting a state of emergency with sika damage to agricultural fields and forests. It should be emphasized accordingly that careful hunting plans are needed based on correct evaluations of population sizes and trends derived from information collected on the current status. To control populations, optimal size or density of the population should be estimated from the viewpoint of biological safety for sustaining local populations, taking into consideration of the amount of agricultural damage by deer. Optimal size or density is variable according to land use, amount of damage, relationship between local people and wildlife, objective of ecosystem management, and environmental conditions. From the economic aspects, there are no co-existence densities with agriculture and forestry because deer cause damage even in low densities of less than 2 individuals/km2 (Miura 1999; Asada and Ochiai 2001). It is necessary, therefore, for managers to adopt other options to avoid damage, such as fencing or tree shelters. Because, as already mentioned, mass mortality or sudden drop of population size has occurred due to heavy snow or winter severity, catastrophic events should be considered in an evaluation of optimal size of the populations (Matsuda et al. 1999). After establishing a goal of population size and density, a plan for annual optimal numbers of deer to be harvested should be calculated to avoid serious risk to the population. Projection matrix models we used here may be useful tools to address these numbers. In some areas, for example Goyo Mountain and Hokkaido, where long-term heavy hunting pressure has been applied, drastic changes were observed in the sex and age structures of the populations. Hunters have a tendency to shoot primarily large antlered trophy males, though female hunting is strongly required. They hesitate to shoot females and calves because of their psychological resistance. As a

32 Management Strategy of Sika Deer Based on Sensitivity Analysis

469

consequence, the sex ratio of these populations has been skewed remarkably toward females, reaching less than one male per 20 females. Almost all males more than five years old have been removed completely from the population. This extremely biased sex ratio creates serious problems for maintaining the population. Like other large cervids, the sika deer is a polygynous species. Dominant males in aggressive interaction with other males mate a large number of females during the rut (Miura 1984; chapters in this book). Dominants are thought to be both physically and genetically healthy according to strong sexual selection. Contrary to this mating system, the disturbance of the sex ratio due to hunting might pose a lethal risk for sustaining healthy populations because only a small number of young males copulate with a majority of females. It is widely accepted that effective population size (Ne), and thus viable population size, is affected greatly by sex ratio. We recommend that strong regulation of male hunting is needed for the populations subjected to long-term hunting pressure. Finally, we want to point out the fact that the number of hunters has steadily decreased from the peak of 450,000 in the early 1970s to 150,000 in 2000, and 70% of these are at least 60 years old. Consequently, the number of harvested deer/ hunter has increased rapidly with a decline in the number of hunters and increase in the size of sika deer populations (Fig. 32.6). Few members of the young generations (20- and 30-year-olds) are becoming hunters, numbering only 2,000–3,000 in 2000, and declining since then. Hunters are essentially an “endangered species” and will be extinct within 20 years. Aside from establishing sound management practices for sika deer populations, we are now facing a big question: who will control the sika population in the future?

Literature Cited Asada, M., and K. Ochiai. 2001. Annual report on conservation and management of sika deer in Boso Peninsula, Chiba Prefecture: 9. Chiba Prefecture, Japan. (In Japanese.) Benton. T. G., and A. Grant. 1996. How to keep fit in the real world: Elasticity analyses and selection pressures on life histories in a variable environment. American Naturalist 147:115–139. Biodiversity Center of Japan. 2004. The national survey on the natural environment. Report of the distribution of Japanese animals (mammals). Ministry of Environment, Tokyo, Japan. Bowyer, R. T., M. C. Nicholson, E. M. Molvar, and J. B. Faro. 1999. Moose on Kalgin Island: Are density-dependent processes related to harvest? Alces 35:73–89. Brault, S., and H. Caswell. 1993. Pod-specific demography of killer whales (Orcinus orcus). Ecology 74:1444–1454. Caswell, H. 1989. Matrix population models. Sinauer, Sunderland, Massachusetts, USA. Caswell, H. 2001. Matrix population models, 2nd edition. Sinauer, Sunderland, Massachusetts, USA. Caughley, G. 1977. Analysis of vertebrate populations. Wiley, New York, New York, USA. Caughley, G. 1994. Directions in conservation biology. Journal of Animal Ecology 63:215–244. Citta, J. J., and L. S. Mills. 1999. What do demographic sensitivity analyses tell us about controlling brown-headed cowbirds? Studies in Avian Biology 18:121–134. Clutton-Brock, T. H., and S. D. Albon. 1982. Winter mortality in red deer (Cervus elaphus). Journal of Zoology 198:515–519.

470

S. Miura, K. Tokida

Clutton-Brock, T. H., F. E. Guiness, and S. D. Albon. 1982. Red deer: Behavior and ecology of two sexes. University of Chicago Press, Chicago, Illinois, USA. Crooks, K. R., M. A. Sanjayan, and D. E. Doak. 1998. New insights on cheetah conservation through demographic modeling. Conservation Biology 12:889–895. Cross, P. C., and S. R. Beissinger. 2001. Using logistic regression to analyze the sensitivity of PVA models: A comparison of methods based on African wild dog models. Conservation Biology 15:1335–1346. Crouse, D. T., L. B. Crowder, and H. Caswell. 1987. A stage-based population model for loggerhead sea turtles and implications for conservation. Ecology 68:1412–1423. Crowder, K. B., D. T. Crouse, S. S. Heppell, and T. H. Martin. 1994. Predicting the impact of turtle excluder devices on loggerhead sea turtle populations. Ecological Applications 4:437–445. de Kroon, H., J. van Groenendael, and J. Ehrlen. 2000. Elasticities: A review of methods and model limitations. Ecology 81:607–619. Eberhardt, L. L. 1977. Optimal policies for conservation of large mammals with special reference to marine ecosystems. Environmental Conservation 4:205–212. Ecos, J., C. L. Alados, and J. M. Emlen. 1994. Application of the stage-projection model with density-dependent fecundity to the population dynamics of Spanish ibex. Canadian Journal of Zoology 72:731–737. Ericsson G., K. Wallin, J. P. Ball, and M. Broberg. 2001. Age-related reproductive effort and senescence in free-ranging moose, Alces alces. Ecology 82:1613–1620. Festa-Bianchet, M., J. T. Jorgenson, M. Lucherini, and W. D. Wishart. 1995. Life-history consequences of variation in age of primiparity in bighorn ewes. Ecology 76:871–881. Fowler, C. W. 1981. Density dependence as related to life history strategy. Ecology 62: 602–610. Gaillard, J.-M., A. J. Sempere, J.-M. Boutin, G. van Laere, and B. Boisaubert. 1992. Effects of age and body weight on the proportion of females breeding in a population of roe deer (Capreolus capreolus). Canadian Journal of Zoology 70:1541–1545. Gaillard, J.-M., D. Delorme, and J. M. Jullien. 1993. Effects of cohort, sex, and birth date on body development of roe deer (Capreolus capreolus) fawns. Oecologia 94:57–61. Gaillard, J.-M., M. Festa-Bianchet, and N. G. Yoccoz. 1998. Population dynamics of large herbivores: Variable recruitment with constant adult survival. Trends in Ecology and Evolution 13:58–63. Gaillard, J.-M., M. Festa-Bianchet, N. G. Yoccoz, A. Loison, and C. Toigo. 2000. Temporal variation in fitness components and population dynamics of large herbivores. Annual Review of Ecology and Systematics 31:367–393. Geist, V. 1971. Mountain sheep: A study in behavior and evolution. University of Chicago Press, Chicago, Illinois, USA. Grand, J., and S. R. Beissinger. 1997. When relocation of loggerhead sea turtle (Caretta caretta) nests becomes a useful strategy. Journal of Herpetology 31:428–434. Guiness, F. E., T. H. Clutton-Brock, and S. D. Albon. 1978. Factors affecting calf mortality in red deer. Journal of Animal Ecology 47:812–832. Hayssen, V., A. van Tienhoven, and A. van Tienhoven, editors. 1993. Asdell’s patterns of mammalian reproduction. Cornell University Press, Ithaca, New York, USA. Heppell, S. S., H. Caswell, and L. B. Crowder. 2000. Life histories and elasticity patterns: Perturbation analysis for species with minimal demographic data. Ecology 81:654–665. Horvitz, C. C., D. W. Schemske, and H. Caswell. 1997. The relative “importance” of life-history stages to population growth: Prospective and retrospective approaches. Pages 247–272 in S. Tuljapurkar and H. Caswell, editors, Structured-population models in marine, terrestrial, and freshwater systems. Chapman and Hall, New York, New York, USA. Iimura, T. 1980. An ecological study on the Japanese deer, Cervus nippon centralis, in the Tanzawa Mountains from the view point of forest protection. Dainipponsanrinkai, Tokyo, Japan. (In Japanese with English summary.) Inukai, T. 1952. The sika deer of Hokkaido and its prosperity and decline. The Report of Northern Cultural Research (Hoppo Bunka Kenkyu Houkuku) 7:1–45. (In Japanese.)

32 Management Strategy of Sika Deer Based on Sensitivity Analysis

471

Japan Meteorological Agency. 2004. Report on climate change. Japan Meteorological Agency, Tokyo, Japan. (In Japanese.) Japan Meteorological Agency. 2007. Long-term trends of phenological events in Japan. Japan Meteorological Agency, Tokyo, Japan. (In Japanese.) Kaji, K. 1997. Annual report on conservation and management of Yezo sika deer in Hokkaido, 2. Hokkaido, Japan. (In Japanese.) Kie, J. G, M. White, and D. L. Drawe. 1983. Condition parameters of white-tailed deer in Texas. Journal of Wildlife Management 47:583–594. Koizumi, T. 1992. Reproductive characteristics of female sika deer, Cervus nippon, in Hyogo Prefecture, Japan. Pages 561–563 in F. Spitz, G. Janeau, G. Gonzalez, and S. Aulagnier, editors. Ongulés/Ungulates 91. Societé Française pour l’Étude et la Protection de Mammifères, Paris, France. Loison A, C. Toigo, J. Apollinaire, and J. Michallet. 2002. Demographic processes in colonizing populations of isard (Rupicapra pyrenaica) and ibex (Capra ibex). Journal of Zoology, London 256:199–205. Maruyama, N., and K. Takano. 1985. Basic research on large mammals and changes in forest environment, pages 248–253, Environmental Agency, Tokyo, Japan. (In Japanese.) Matsuda, H., K. Kaji, H. Uno, H. Hirakawa, and T. Saitoh. 1999. A management policy for sika deer based on sex-specific hunting. Researches on Population Ecology 41:139–149. McCarthy, M. A., M. A. Burgman, and S. Ferson. 1995. Sensitivity analysis for models of population viability. Biological Conservation 73:93–100. McCullough, D. R. 1979. The George Reserve deer herd: Population ecology of a K-selected species. University of Michigan Press, Ann Arbor, Michigan, USA. McCullough, D. R. 1999. Density dependence and life-history strategies of ungulates. Journal of Mammalogy 80:1130–1146. Menkens, G. E., Jr., and M. S. Boyce. 1993. Comments on the use of time-specific and cohort life tables. Ecology 74:2164–2168. Mech, L. D., R. E. McRoberts, R. O. Peterson, and R. E. Page. 1987. Relationship of deer and moose populations to previous winters. Journal of Animal Ecology 56:615–627. Mills, L. S., and M. S. Lindberg. 2002. Sensitivity analysis to evaluate the consequences of conservation actions. Pages 338–366 in S. R. Beissinger and D. R. McCullough, editors, Population viability analysis. University of Chicago Press, Chicago, Illinois, USA. Mills, L. S., D. F. Doak, and M. J. Wisdom. 1999. Reliability of conservation actions based on elasticity analysis of matrix models. Conservation Biology 13:819–829. Milner-Gulland, E. J. 1994. A population model for the management of the saiga antelope. Journal of Applied Ecology 31:25–39. Ministry of Agriculture, Forestry and Fisheries. 2004. Annual report on agricultural damage caused by wildlife in Japan. Ministry of Agriculture, Forestry and Fisheries, Tokyo, Japan. Miura, S. 1984. Social behavior and territoriality in male sika deer (Cervus nippon Temminck 1838) during the rut. Zeitschrift für Tierpsychologie 64:33–73. Miura, S. 1991. Life history strategies of Japanese ungulates and their management. Pages 244–273 in M. Asahi and T. Kawamichi, editors, Contemporary mammalogy. Asakura-Shoten, Tokyo, Japan. (In Japanese.) Miura, S. 1999. Ecology of wildlife and agricultural and forestry damage. ZenRinKyo, Tokyo, Japan. (In Japanese.) Nelson, L. J., and J. M. Peek. 1982. Effect of survival and fecundity on rate of increase of elk. Journal of Wildlife Management 46:535–540. Ohtaishi, N. 1975. Life table of sika deer in Nara Park. Pages 25–35 in Annual report sika deer in Nara Park, 1974. Kasuga-Kenshokai, Nara, Japan. (In Japanese.) Pianka, E. R. 1970. On r and K selection. American Naturalist 104:592–597. Picton, H. D. 1984. Climate and the prediction of reproduction of three ungulate species. Journal of Applied Ecology 21:869–879. Post, E., and N. C. Stenseth. 1998. Large-scale climatic fluctuation and population dynamics of moose and white-tailed deer. Journal of Animal Ecology 67:537–543.

472

S. Miura, K. Tokida

Post, E., and N. C. Stenseth. 1999. Climatic variability, plant phenology, and northern ungulates. Ecology 80:1322–1339. Read, A. F., and P. H. Harvey. 1989. Life history differences among the eutherian radiations. Journal of Zoology, London 219:323–353. Sadleir, R. M. F. S. 1987. Reproduction in female cervids. Pages 123–144 in C. M. Wemmer, editor, Biology and management of the Cervidae. Smithsonian Institution Press, Washington, DC, USA. Senda, M., Y. Tanimoto, and N. Koyama. 2002. Economic results of grazing on less-utilized arable fields in LFAs. Bulletin of National Agricultural Research Center for Western Region 2:41–58. (In Japanese with English summary.) Shea, K., and D. Kelly. 1998. Estimating biocontrol agent impact with matrix models: Carduus nutans in New Zealand. Ecological Applications 8:824–832. Silvertown, J., M. Franco, and E. Menges. 1996. Interpretation of elasticity matrices as an aid to the management of plant populations for conservation. Conservation Biology 10:591–597. Singer, F. J., A. Harting, K. K. Symonds, and M. B. Coughenour. 1997. Density dependence, compensation, and environmental effects on elk calf mortality in Yellowstone National Park. Journal of Wildlife Management 61:12–25 Skogland, T. 1985. The effects of density-dependent resource limitations on the demography of wild reindeer. Journal of Animal Ecology 54:359–374. Skogland, T. 1988. Tooth wear by food limitation and its life history consequence in wild reindeer. Oikos 51:238–242. Takatsuki, S. 1992. Sika deer living in the north. Doubutsusha Publishing, Tokyo, Japan. (In Japanese.) Takatsuki, S. 1998. The twinning rate of sika deer, Cervus nippon, on Mt. Goyo, northern Japan. Mammal Study 23:103–107. Takatsuki, S. 2000. Life of sika deer reconstructed by teeth. Iwanami-Shoten, Tokyo, Japan. (In Japanese.) Takatsuki, S., and K. Suzuki. 1990. Reproductive performance of female sika deer in Goyo Mountain. Annual Meeting of the Mammal Society of Japan. Abstract 79. (In Japanese.) Takatsuki, S., S. Miura, K. Suzuki, and K. Ito. 1991. Age structure of mass mortality in the sika deer (Cervus nippon) population on Kinkazan Island, northern Japan. Journal of the Mammalogical Society of Japan 15:91–98. Uno, H., M. Yokoyama, and M. Takahashi. 1998. Winter mortality pattern of sika deer (Cervus nippon yesoensis) in Akan National Park, Hokkaido. Mammalian Science 38:233–246. (In Japanese.) Unsworth, J. W., D. F. Pac, G. C. White, and R. M. Bartmann. 1999. Mule deer survival in Colorado, Idaho, and Montana. Journal of Wildlife Management 63:315–326. Walsh, N. E., B. Griffith, and T. R. McCabe. 1995. Evaluating growth of the Porcupine caribou herd using a stochastic model. Journal of Wildlife Management 59:262–272. White, G. C. 2002. Population viability analysis: Data requirements and essential analyses. Pages 288–331 in L. Boitani and T. K. Fuller, editors, Research techniques in animal ecology: Uses and misuses. Columbia University Press, New York, New York, USA. Wisdom, M. J., and L. S. Mills. 1997. Sensitivity analysis to guide population recovery: Prairie chickens as an example. Journal of Wildlife Management 61:302–312. Wisdom, M. J., L. S. Mills, and D. F. Doak. 2000. Life-stage simulation analysis: Estimating vitalrate effects on population growth for conservation. Ecology 81:628–641. Yamauchi, K., S. Hamasaki, Y. Takeuchi, and Y. Mori. 1997. Assessment of reproductive status of sika deer by fecal steroid analysis. Journal of Reproductive Development 43:221–226.

Chapter 33

Sika Deer in Russia Vladimir V. Aramilev

Abstract Sika deer (Cervus nippon hortulorum) were originally distributed over much of Primorsky Krai (an administrative division in the Russian Far East similar to a U.S. state) in Far East Russia, occupying coastal and inland valley deciduous and mixed deciduous-pine forests up to about 500 m elevation, with isolated individuals in favored habitat at higher elevations. Their limits to the north extend into southern Khabarovsky Krai along the Ussuri River Valley, and along the seacoast to the village of Malaya Kema. These limits are set by winter snow depth, which also limits their occupation of higher elevations in the Sikhote-Alin mountain range. Sika deer were translocated in the late nineteenth and early twentieth centuries to other countries and to European Russia where they have shown considerable adaptive capacity by acclimatizing to severe winter conditions. Sika deer feed primarily on forest understory vegetation and on herbaceous plants in forest openings. Their native forage is deficient in sodium so they seek sources of salt at the seashore, from mineral soils and springs, or from aquatic freshwater plants with high sodium content. Sika deer displace other ungulate competitors from their preferred habitats and, in high numbers, can have serious impacts on vegetation. There is some interbreeding between sika deer and wapiti, but intermediate phenotypes are rare. Overhunting beginning in the late 1800s and extending through World War II resulted in a severe decline in the sika deer numbers and distribution. At the same time, many sika deer were raised in farms for commercial production of antlers and other products. In the last 25 years there has been an increase in the wild population and expansion of the distribution across Primorsky Krai. Now sika deer occupy most of their original range, apparently colonizing all suitable habitats. Serious anthropogenic pressure, both a loss of habitat and hunting, has not hindered this growth in the number and dispersal of sika deer, although some former habitat along the seacoast has been lost to human development. Sika deer in farms are not very economically viable at present. The re-occupation of the original range consisted of both wild and farm-escaped deer, but genetic studies show that they are all of the original genetic stock. Their increase has occurred despite serious predation pressure by tigers and leopards, and illegal hunting; they are currently managed for sustainable harvest.

D. R. McCullough et al. (eds.), Sika Deer: Biology and Management of Native and Introduced Populations, DOI: 10.1007/978-4-431-09429-6_33, © Springer 2009

475

476

V. V. Aramilev

Original Distribution The first references to wild sika deer in Primorsky Krai are found in the works of explorers and researchers investigating the south of the Russian Far East (Maak 1861; Przhevalsky 1870; Yankovsky 1882; Emelyanov 1927; Abramov 1928; Zolotarev 1937; Bromley 1956; Geptner et al. 1961; Bromley and Kucherenko 1983; Arsenev 2003). The numbers and distribution of sika deer in Far East Russia have shown considerable fluctuation over the last century, due primarily to human influences, especially uncontrolled hunting. Climatic fluctuations, while influencing occupation by sika deer of the more extreme parts of the distribution, were minor in relation to the broad-scale impacts of humans that affected the population over their entire range. Bromley and Kucherenko (1983) provided a map of the approximate range of sika deer in the mid-nineteenth century (Fig. 33.1). The historical distribution of sika deer was principally in Primorsky Krai where sika deer were distributed along the coast and inland along lower elevation river courses (Fig. 33.2). To the west, the original range extended along the Ussuri River as a tongue into southern Khabarovsky Krai and adjacent Heilongjiang Province of China, where the northern edge fluctuated depending on the vagaries of climate and sequences of mild and hard winters. Until the mid-nineteenth century sika deer, in what is modern Primorsky Krai, occupied primary and secondary oak (Quercus) broad-leaved deciduous forests in river valleys and on adjacent mountain slopes to an altitude of about 500 m. On the western slopes of the Sikhote-Alin Mountains deer were regularly encountered as far north as 30 km south of the confluence of the Bolshy Ussurka River with the Ussuri River. From here, and apparently from adjacent areas in China, individual groups, and most likely, lone males, made their way to the foothills and along river courses south of Dalnerechensk. Lone sika deer were often encountered along the Ussuri River valley as far north as Dalnerechensk. This was apparently the extreme northern edge of their range for, since the end of the nineteenth century, there have not been regular sightings of sika deer farther north than Dalnerechensk. Wild sika deer did not range at this time along the Bolshy Ussurka River Valley east of Dalnerechensk. To the south of Dalnerechensk, until the 1880s sika deer were encountered in almost all oak broad-leaved deciduous forests. At the confluence of the Arsenevka and Ussuri Rivers a mosaic-like distributional range ran along the valleys of the middle and upper drainages of the Artemovka, Ilistyi, Komissarovka, and Nesterovka Rivers that flow into the Sea of Japan and Peter the Great Bay south of 44°N latitude. Sika deer occurred sporadically along these river valleys (Fig. 33.1). On the eastern slopes of the Sikhote-Alin Mountains, facing the Sea of Japan, the mouth of the Bolshy Kema River was the northern extreme of sika deer distribution although it is not out of the question that these were feral deer that had escaped from farms. Wild sika deer ranged in separate herds farther south of the mouth of the Cheremukhovy, Zerkalny, and Rudny Rivers. A more or less solid, though

33 Sika Deer in Russia

477

Fig. 33.1 Sika deer range in Primorsky Krai as restored in the nineteenth century (stippled area), and in 1970 (solid squares) (taken from Bromley and Kucherenko 1983).

mosaic-like range, was found along the seacoast to the south of 44°N latitude along the Avvakumovka, Margaritovka, Milogradovka, Chernyi, and Kievka Rivers to the south of Primorsky Krai where deer lived in almost all oak broad-leaved deciduous forest, ranging up mountain slopes to the lower edge of Korean pine (Pinus koraiensis)-broad-leaved deciduous forest belts. Sika deer made their way to the large islands in Peter the Great Bay—Russky, Putyatin, and even the relatively distant Askold—apparently by swimming from the mainland or crossing the ice in winter. There are no data on when sika deer arrived on Askold Island. Information on the Askold Island sika deer population was lacking until the development of mining on the island in the mid-nineteenth century. At that time the deer were hunted by lighthouse workers and miners and, beginning

478

V. V. Aramilev

Fig. 33.2 Current distribution (cross-hatched area) of sika deer in Primorsky Krai.

in 1885, by members of the Vladivostok Society of Sport Hunters. In a report on sika deer written by the Vladivostok Society of Sport Hunters (1897), “… deer numbers on Askold Island are high. Their annihilation in other areas of the krai calls for even more dedicated protection on Askold Island. There is no doubt that the Society can effectively carry out such a program since its charter obliges members to establish strict and correct hunting, to protect and breed valuable animals.” An isolated population of deer lived to the west and southwest of Lake Khanka, along the tributary valleys of the Razdolny River and in the Komisarovka and Nesterovka river valleys that flow into Lake Khanka (Fig. 33.1). This population moved along the valleys of rivers (the Kedrovka, Narva, and Poima) flowing from the west into Peter the Great Bay. In the nineteenth century a rare, two-way interchange apparently continued to operate in the tall grass of the Ussury-Khanka plains; this area connected the wild deer of this southwest range with the SikhoteAlin Mountains. Deer from the western and southwestern areas of Primorsky Krai

33 Sika Deer in Russia

479

made irregular contact with individuals in adjacent areas of China (Bromley and Kucherenko 1983). Additional references to sika deer in other areas of Primorsky Krai in the nineteenth century do exist but they do not contain detailed habitat descriptions. With the turn of the twentieth century records become more available. For example, Arsenev’s diary for 1906 (2003) contains the following: “Earlier the deer’s range extended to Svyaty Olga Bay, but as immigrants and Chinese settled southern Ussuriisky Krai, hunters began to aggressively track the animal, and in the last 60 years the animal has been observed moving slowly to the north. Lone deer have now made their way to the Tkhe-Tu-Be (Dzhigitovka) River.”

Farmed Deer When N. M. Przhevalsky visited Primorye in the mid-nineteenth century, the southern portion of the region, where sika deer lived, was sparsely settled by humans. During this period the Chinese began to actively trade in non-timber forest products, including animal products. Antler velvet (“panty”), deer fetuses (“lutai”), tails, and sinew were highly valued. The number of sika deer began to decline markedly and this provided an incentive to establish deer farms. It is likely that deer farming and captive breeding techniques were obtained from the Chinese. S. Y. Ponosov, the first deer farmer in Primorsky Krai, began breeding deer in Shkotovsky District and subsequently moved his operation to Olginsky District where, in 1912, his herd numbered approximately 500 head. By the beginning of the twentieth century M. I. Yankovsky farmed around 2,000 head: 1,500 on Putyatin Island and 500 at Cape Gamov. There were deer farms at Cape Peschanyi and on Popov, Rikord, and Rimsky-Korsakov Islands, with a number of smaller operations spread out along the coastline and in the Prikhankaisky lowlands; the total number of deer at these latter operations was around 1,500 head. During the Russian Civil War (1917–1924), and the following period of collectivization, many of the deer farms were destroyed. By the end of 1922 there were around 3,000 head at all the remaining farms. During the Russian Civil War and the consolidation of Soviet power, the number of farm-bred sika deer declined to 3,000 head (Abramov 1928; Menard 1930; Bromley 1956). Subsequent immigrants to the Primorsky region recognized the value of sika deer antler velvet in East Asian markets and they began to re-establish deer farms. By the mid-1920s there were at least 10,000 sika deer in captivity in Primorsky Krai (Ryashchenko 1976). Deer numbers were saved from destruction at farms during World War II. Sika deer breeding at other locations in Russia began after the war, with the breeding stock having been derived from wild and farm-bred deer in Primorsky. The number of deer at farms was around 20,000 by the mid-1950s (Ilina 1956), and in 1968 that figure reached 36,000. In 1978 there were 56,000 deer, of which 48,000 were located in Primorsky Krai (Sidorov 1980). At the beginning of the 1980s, Bromley and Kucherenko (1983) wrote that “Antler velvet, at the current time, is successfully

480

V. V. Aramilev

obtained from animals living in captivity. Practice shows that for the future acclimatization and re-acclimatization of sika deer, it is significantly simpler, from a technical point of view, to use animals from deer farms than wild animals. Wild sika deer should be kept and used only to occasionally revitalize the genetic fund of farm-bred deer.” By the mid-1980s the number of farm raised deer reached 60,000, with 15,000 deer located in the Altai region (Bogachev et al. 1983). Sika deer farm collectives in Primorye were located along the coast of southwest Primorsky, in the Prikhankaisky Lowlands, and along the Sea of Japan coastline to the east and north of Vladivostok to Port Plastun. Deer held at farms regularly escaped and mixed with wild deer, while at the same time deer farmers regularly set live traps along the perimeters of farms and successfully captured wild deer to supplement their captive herds. Farm-raised deer today are not experiencing the best of times in Primorsky Krai. This is a consequence of a drop in the price for antler velvet on foreign markets and the lack of demand on domestic markets. Antler velvet and hides currently have almost no value. All government and collective deer farms have been purchased by private owners. The sector, due to low prices for its products, is not expanding, deer numbers are dropping markedly, and the number of captive deer in Primorsky Krai is currently less than several thousand. Although they face similar problems, deer farms in the Altai region are surviving. Sika deer released in the late twentieth century at hunting collectives in western Russia are a common commercial hunting target.

Relocation to European Russia and Other Countries Two hundred forty head from Primorsky Krai deer farms were released in Russian zapovedniks (nature reserves): Khopersky, Teberdinsky, Zhigulevsky, Buzuluksy Bor, Oksky, Ilmensky, and Mordovsky in 1938 (Arsenev 1949; Ilina 1956; Pavlov et al. 1974). These animals acclimatized in different ways. In the second half of the twentieth century farm-bred deer from Primorsky Krai and the Altai were introduced at the Khopersky and Askaniya Nova Zapovedniks in the European portion of Russia. The current number of sika deer outside their natural range is 3,700 (Danilkin 1999). In the late nineteenth and early twentieth centuries, sika deer were exported to other countries, including Ireland, Denmark, Czechoslovakia, and other European countries. In 1900 Primorsky sent 15 head to New Zealand where there are now tens of thousands of sika deer (Bromley and Kucherenko 1983).

Genetics Following the dramatic decline in wild sika deer numbers in the 1940s and 1950s, several Russian wildlife specialists asserted that genetically pure sika deer were to be found only in Lazovsky Reserve and adjacent areas (Prisyazhnyuk 1974,

33 Sika Deer in Russia

481

1978; Bromley and Kucherenko 1983). They held that there were two forms of sika deer in the population in Primorsky Krai: wild sika deer of the original stock, and individuals derived from farm-raised animals. According to Prisyazhnyuk (1978), the domestic sika had a smaller body and antlers, its internal organs weighed less, and it differed in several biochemical indicators in comparison to wild deer. Fifteen years later, however, the first tentative statements appeared that the differences cited above “fall into a range of differences and modifications within a single subspecies” (Bogachev and Alymov 1990). Modern DNA analysis techniques now allow us to rigorously examine the relationships among sika deer in Primorsky Krai. In 1999 we collected skin samples from sika deer at various locations throughout their modern range in Primorsky Krai. Samples were taken from 42 deer: 24 deer at the western extreme of their range, four from the wild and 20 from deer farms; 18 samples were taken from their eastern range in the Sikhote-Alin Mountains, 10 from the wild and eight from deer farms (Fig. 33.3). Three of the samples came from wild deer in the Lazovsky Reserve. According Wilson (2000) and Simon Goodman et al. (unpublished data), analysis of DNA samples for sika deer in Primorsky Krai showed a high degree of genetic diversity for this subspecies in comparison with other subspecies. However, the results indicated that there are no genetic distinctions between farmed and wild deer. Taking genetic, demographic, and geographic factors together, we consider the sika deer in Primorsky Krai to be a single native population. This was to be expected since sika deer initially captured for farms in Primorsky Krai were taken from the wild around 100–150 years ago. Genetic change in such a short time would not be expected, especially since farm-bred deer have regularly intermixed with wild deer. Neither is it likely that a native form of sika deer would arise in Lazovsky and Olginsky Districts since in the late nineteenth and early twentieth centuries both areas contained privately owned deer farms and in the 1970s two

Fig. 33.3 Location of tissue sample collections (dots, numbers) in Primorsky Krai, Far East Russia for DNA analysis.

482

V. V. Aramilev

large, state-run farms, one in the village of Glazkovka (Lazovsky District) and another in the village of Vesely Yar (Olginsky District) were operating. The breeding stock for these collectives was Primorsky-derived from deer farms located in southwest Primorsky Krai.

Conservation Status Despite the lack of genetic distinction between native and farm-bred sika deer in Primorsky Krai, wild sika deer are included on the Russian Federation Endangered Species List (IUCN Red Book) and noted as a “native population.” Hunting sika deer is prohibited in Lazovsky and Olginsky Districts of Primorsky Krai. The sika deer on Askold Island, located in Peter the Great Gulf, are also listed in the Red Book. At the time sika deer were added to the USSR Red Book, many scientists believed they survived only in Olginsky and Lazovsky Districts, and on Askold Island, so these were the only regions described in the Red Book.

Population Size Changes over Time According to longtime residents and hunters, until the mid-nineteenth century there were around 25,000 sika deer in Primorsky Krai. The highest density was along the shores of Peter the Great Bay and to the north along the Sea of Japan coast to Olga Bay (Fig. 33.1). The oak and the mixed broad-leaved deciduous forests preferred by sika deer still dominate in these areas (Bromley and Kucherenko 1983). Populations declined sharply in the first half of the twentieth century although there are no solid census figures. Abramov (1939), at the end of the 1920s, estimated the number of sika deer in the Russian Far East at 900–1,100 head. As a result of the Civil War, localized military conflicts (during the1930s and 1940s), and World War II (1941–1945), the number of wild deer dropped precipitously. This was partially a consequence of no control of hunting during war times and also the creation of special brigades to shoot wild ungulates to supply the army. Heavy winter snows and predator impacts also took a toll on deer numbers. By 1949 the number had fallen to about 300 (Bromley 1956), 150 of which were located in the Sudzukhinsky (later named Lazovsky) Reserve, about 20 in the Sikhote-Alin Reserve, up to 30 in both the Suputinsky (later named Ussuriisky) and the Kedrovy Pad Reserves, with 60 or more outside of the reserves (Geptner et al. 1961). The creation of reserves provided protection for the remaining deer until the outbreak of World War II. In 1938, as conflict with Japan escalated, the Red Army opened the Far Eastern Front. Initially manned by two army units, an additional unit was added two years later. The term “front” in Soviet military language meant army units numbering from several hundred thousand to a million soldiers. A single unit required up to 60,000 head of cattle a month for rations (Suvorov 1998).

33 Sika Deer in Russia

483

On April 13, 1941 Russia signed a neutrality agreement with Japan. Even so, the army units in Far East Russia were not decommissioned. The Far Eastern Front ran from TransBaikal to Primorsky and one of the three units was stationed in the south and southwest of Primorsky, that is, in sika deer habitat. Special brigades operated between 1938 and 1947 whose mission was to shoot wild game to supply the army with meat. These brigades operated year round, and despite their small size—four or five soldiers with a vehicle and a horse-drawn trailer— they took several hundred ungulates a year (I. N. Nesterenko 1985 personal communication). Local people also hunted wild ungulates since livestock was annexed in full for the war effort (N. P. Ponomareva 2002 personal communication). The pressure from such unregulated hunting was bound to have an impact on sika deer numbers. The war years nearly wiped out the sika deer population in the southwest and west of Primorsky Krai; in the southern end of the Sikhote-Alin Mountains sika deer remained alive only in the areas difficult to access along the coastline between Port Olga south to Vrangel on Nakhodka Bay. In Kedrovy Pad Reserve, created in 1916, there were 90 head in 1924, and by 1928 around 200. Later the population grew to 300 but by 1948 less than 30 remained. By 1967 there were no sika deer at all left in Kedrovy Pad Reserve (Bogachev 1982). In the Sudzukhinsky (Lazovsky) Reserve the number was around 500 by 1936, but by 1944 the population had decreased by half and in 1948 there were only 150 (Bromley 1956, 1981). Census data for 1974 in Lazovsky Reserve showed 280 deer (Prisyazhnyuk 1975a, b). There were around 30 deer in the Sikhote-Alin Reserve in 1935 (Pavlov et al. 1974). The approximate distribution of sika deer in the 1970s is shown in Fig. 33.1. In the 1950s and 1960s farm-raised deer from collectives were introduced into Primorsky Krai reserves. Deer were introduced into the Suputinsky (Ussuriisky) Reserve in 1950 through 1952 (Pavlov et al. 1974). Seventeen deer from the Silinsky deer farm collective were introduced into the Kedrovy Pad Reserve in 1968. By 1974 there were fewer than 40 deer in Kedrovy Pad, including deer that may have escaped from the Amursky deer farm collective. Fifteen to 18 deer were released into the Ussuriisky Reserve in 1960 and 10 more in 1962 (Bogachev 1982). The deer population increased thereafter. Dormidontov (1977) estimated there were 1,000 wild sika deer in Primorsky and Pikunov et al. (1973) reported 670–700 in the 1960s all mainly in the Lazovsky Reserve, Khasanskii District, and on Askold Island. Tracking sika deer dispersal across their entire range is problematic since no special research of this kind has been undertaken and survey data are contradictory. Sika deer recovered from low population numbers and restricted distribution following excessive hunting in earlier years, and this process continued over the last few decades. By the end of the 1960s in Olginsky District, groups of sika deer remained in locations difficult to access along the coast from Moryak-Rybolov Bay to Olga Bay. According to V. M. Mashutin, the manager of the Nizmenny lighthouse, at the beginning of the 1960s the number of sika deer was less than 100. By the early 1980s sika deer were widely dispersed, their numbers difficult to estimate, and the wapiti (Cervus elaphus xanthopygus) had by then all but disappeared near the lighthouse. Makovkin (1999) also reported that sika deer replaced wapiti in

484

V. V. Aramilev

Partizansky Creek during their expansion, thus showing the competitive dominance of sika deer as presented in greater detail below. In the early 1980s sika deer began to intensively colonize the Vasilkovka River Valley, which is 10–15 km from the coast. By the early 1990s they began to colonize the left tributaries of this river, and by 2000 they had moved up into elevations of 600–700 m, 25–30 km from the sea. Sika deer dispersed with equal rapidity into the interior of the coastal mainland. Not just wild sika deer colonized the area in the 1960s and 1970s, but also deer that escaped from deer farms at the villages of Valentin (Lazovsky District), Vesely Yar (Olginsky District), and Plastun (Terneisky District). They also dispersed along river valleys. In 1983 sika deer occupied the steep southern slopes in the middle drainage of the Avvakumovka River, and in 10 years had stretched upwards along the river and its branches for a distance of 30–35 km. Dispersal across Olginsky District at this time was 3–3.5 km a year. Observations on sika deer since 1982 for Olginsky and Lazovsky Districts reveal that deer distribution has moved 5–10 km inland from the coast. Over time, sika deer have spread along river valleys and across low mountain passes in directions leading away from the sea and, in the last 20 years, they have moved 60–80 km inland along river valleys. According to Makovkin’s (1999) observations, as soon as the number of sika deer in Lazovsky Reserve exceeded 500 and the forage in the coastal areas was exhausted, the deer began to colonize adjacent areas. This process was gradual and the rate was not uniform for all areas. Indicative of the process is Benevka River, where colonization occurred from two directions: from the sea coast across the Zapovednyi Ridge and along the Kievka and Benevka River Valleys. The deer colonized the area much more quickly from the coast, the rate reaching 5 km a year. From the south, the valleys were colonized only half as quickly.

Current Distribution Sika deer currently occupy the range that they occupied in the mid-nineteenth century (compare Fig. 33.1 with Fig. 33.2). A comparison of the modern range with a reconstructed range reveals some differences in the survey data from the nineteenth century that can logically be explained as data gaps. The sika deer apparently now occupy all suitable habitats in the south of the Russian Far East. Most sika deer in Far East Russia live in Primorsky Krai, although in recent years there have been regular encounters with groups of animals and lone deer in southern Khabarovsky Krai in the eastern tributaries of the Ussuri River. The northern most location for sika deer in the Ussuri River valley is the city of Vyazemsky in Khabarovsky Krai. Their range in Primorsky Krai consists of three interconnected areas. The southwestern area is in southwest Primorye, in the Cherny Mountains, and on the Borisovsky Plateau where sika deer range from the seashore to the border with

33 Sika Deer in Russia

485

China. Isolated populations of sika deer are found on the left bank of the Tumangan River, in the mountains along the southern portions of Ekspeditsy Bay. A northwest area is located to the west of Lake Khanka, on the Pogranichny Ridge, and in the forested portion of the Prikhankaisky lowlands. Contact with the southwest group takes place across forested areas in the headwaters of the Slavyanka and Nesterovka Rivers. The main group is located in the Sikhote-Alin mountain system. The southern boundary of the range crosses from the mouth of the Razdolny River and runs along the coast to the east, and then up the coast, to the north, to the port of Rudny Pristan. Sika deer densities decrease to the north toward the left bank of the Dzhigitovka River. Sika deer are found on a regular basis starting at the left bank of the Dzhigitovka River and running to the village of Ternei, this being the territory of the Sikhote-Alin Biosphere Reserve (Zapovednik). North of Ternei the deer range to the village of Malaya Kema, and individuals are spotted near the village of Amgu. See Voloshina and Myslenkov (chapter 34) for a more detailed discussion of sika deer at the northern edge of their range in Russia. A purported natural break in sika deer range in Primorsky Krai along the Razdolny River Valley (between Vladivostok and Lake Khanka; Figs. 33.1 and 33.2) is not an insurmountable barrier. Although the valley is characterized by urbanization and agriculture, there is sufficient remaining habitat to allow passage of some deer. A broad-leaved deciduous forest runs along the Vladivostok–Ussuriisk highway and railroad right-of-way and the Razdolny River valley is largely meadow containing areas of wetlands overgrown with reeds 2–2.5 m tall, and areas of sparse forest and scattered individual trees. Sika deer swim well and the Razdolny River is not a serious obstacle; in winter deer cross the river on the ice. During recent Amur leopard (Panthera pardus orientalis) field surveys, fresh sika deer tracks were regularly encountered crossing the Razdolny-Khasan highway and moving into the Razdolny River flood plain.

Sika Habitat and Its Condition Sika deer in Primorsky Krai live primarily in oak broad-leaved deciduous forests and mixed broad-leaved-Korean pine forest of the Manchurian flora subregion. The distribution of sika deer and these mixed forest types are nearly identical. These forests have well-stocked undergrowth. The region’s slopes support abundant broad-leaved deciduous forests of Quercus mongolica (oak), Tilia amurensis (linden), Acer tegmentosum (maple), Fraxinus mandshurica (ash), Phellodendron amurense (Amur cork tree), Juglans mandshurica (Manchurian walnut), Aralia elata (aralia), Lespedeza bicolor (lespedeza), and other heat-loving plants. They grade into Korean pine stands at the higher elevations. These forests contain numerous small meadows and light gaps, and shady valleys and gullies that are very suitable for sika deer (Bromley and Kucherenko 1983). Sika deer are less often encountered in Korean pine-broad-leaved deciduous forests above 500 m and are

486

V. V. Aramilev

entirely absent from Korean pine-dark coniferous taiga. In river valleys sika deer are found in broad-leaved deciduous valley forests and low-light coniferous valley forests. On the eastern macro-slope of the Sikhote-Alin Mountains sika deer are found in both broad-leaved deciduous and narrow-leaved forests. Sika deer also like forests where fire promotes young growth as well as narrowleaved deciduous forests. Extensive or repeated fire, conversely, is a genuine habitat threat and is, as a rule, human-caused to promote agricultural or forestry purposes. Fire completely destroys the forest understory, and herbaceous species composition decreases with repeated fires. Depending on forage availability and degree of protection, deer density changes in various habitat types. Deer distribution is restricted by elevation and degree of slope. They prefer gentler slopes. Sika deer currently live year round in southwest Primorsky Krai at elevations up to 800–900 m. On the eastern macro-slope of the Sikhote-Alin, depending on the latitude, sika deer use habitats at 600–700 m, and on the western macro-slope 400–500 m.

Winter Snow and Sika Deer The key factor limiting sika deer distribution is snow depth. Wild sika deer are typical representatives of thermophilic fauna of the south of the Russian Far East. The subspecies C. n. hortulorum occurs at the extreme distribution of its ecological range in Primorsky Krai. Prime sika deer conditions occur where: (1) winter snow totals do not exceed 800–1,000 cm; (2) the first snowfalls are comparatively late in the season; (3) there are fewer than 45 days of snow cover; and (4) the average snow pack is less than 25–30 cm. Sika deer are encumbered in snow deeper than 40–50 cm and are helpless in very deep snow. Consequently, in winter sika deer seek the southern and southeastern coastal areas of Primorsky, a 10–15 km wide forest belt adjacent to the Sea of Japan. The deer prefer south and southeastern facing slopes where the snow pack as a rule does not accumulate. Southern Primorsky generally has little snow and what does fall usually remains on the ground for relatively short periods (usually not more than five to eight days), and is often melted by rain (Abramov 1928, 1939; Bromley 1956). Topographic relief plays a significant role since on steep slopes snow depth is always less, and the snow is more quickly dispersed by wind and sublimated by the sun. The severity of winter influenced deer numbers, particularly during periods when their populations were high. Bromley and Kucherenko (1983) examined the data for 70 years during the twentieth century, and deep-snow winters have been registered 11 times: 1909–1910, 1914–1915, 1924–1925, 1932–1933, 1941–1942, 1947–1948, 1957–1958, 1961–1962, 1965–1966, 1972–1973, 1978–1979. From our data in Sikhote-Alin there was a deep-snow winter in 1985–1986, but it did not extend to the coast where the major part of sika deer range was located. We have not observed mortality of deer due to deep snow in recent years.

33 Sika Deer in Russia

487

Migration Long migrations of sika deer in Primorsky Krai have not been described. In the opinion of many authors, sika deer in Primorsky Krai are a resident, non-migratory species (Bromley 1956; Prisyazhnyuk 1974; Bromley and Kucherenko 1983). Although some sika deer in the interior distribution at highest elevations may migrate seasonally like sika deer in Japan (Igota et al. chapter 19; Yabe and Takatsuki chapter 20) and as do some wapiti (32% of the population) in the Sikhote-Alin Reserve in Far East Russia (D. Miquelle 2001 personal communication), most parts of the sika range in Primorsky Krai are low elevation habitats and coastal environments. Thus, sika movements in response to snow tend to be short in distance, and often reverse repeatedly in direction as snow falls and melts over the winter. From our observations, there are short vertical local shifts when depth of snow increases. Sometimes heavy wet snow breaks down young trees, and incidentally results in bringing food within reach of sika deer, which counteracts the difficulty of deer moving around in deep snow. Sometimes sika deer simply move to steep slopes with a southern exposure where snow does not accumulate. These movements are from several kilometers up to several tens of kilometers long and are directed downwards into valleys of springs and rivers. Most of these movements are probably best viewed as local shifts rather than seasonal migrations. The longer movements are observed in deep snow winters, when the general direction of movement is toward the seashore, where the depth of a snow cover is always lower. On our counting areas at the seashore, where average population density is 2.5–3.0 per km2, in deep snow winters population density is seven to 10 per km2. Because such movements are irregular depending on snow depth, it is debatable if they constitute migrations. Movements of sika deer on the interior margins of their range in Primorsky Krai have hardly been investigated. Very likely a large sample of radio-tracked sika deer would reveal some small percentage of the population in the highest elevations of their distribution that performs what could be considered seasonal migrations, but that must await further study.

Food Habits and Use of Salt Sika are predominantly a “disturbance” species, using wooded areas for escape and shelter, and coming into forest gaps or other disturbed areas for feeding. Postharvest oak stands are rich in herbaceous and shrub layers and in sedges, grass, and mixed grass meadows. Our observations show that sika deer are not fastidious feeders, and this is especially obvious in winter. In contrast to wapiti and roe deer (Capreolus pygargus), a herd of 25–30 sika deer can live in an area of 10–15 ha for 15–20 days. Wapiti and roe deer occur in small groups and move about more regularly to obtain preferred forage species.

488

V. V. Aramilev

Since sika deer habitat in Primorsky Krai is sodium-deficient, the deer must find ways to supplement sodium in their diet. If additional sources of sodium are not available, habitat quality declines (Aramilev 2000). There are three natural sources of sodium in Primorsky Krai: (1) seaweed, (2) natural mineral salts (solonetz) from mountain rock and soil deposits or from springs, and (3) halophytic freshwater vegetation in rivers. All three sources have high sodium concentrations, 10 or 100 times that of the sika deer’s typical forage (Aramilev 2002). Artificial salt licks set up by hunters provide additional non-natural sources of sodium. The most readily available source of sodium, seaweed on coastal beaches, coincides with their preferred habitat. Deer regularly visit beaches in coastal bays covered with seaweed provided these bays have not undergone extensive anthropogenic impact. Makovkin (1999) repeatedly refers to the intensive use of seaweed both in Lazovsky Reserve and in areas outside the boundaries of the reserve. The deer living along river valleys and in low elevation areas 40–50 km from the sea journey to the sea. The deer living at the more remote territories seek out natural mineral solonetz deposits; these occur on the Lazovsky and Ussuriisky Reserves. Another sodium source for the deer is freshwater vegetation in rivers like Vasilkovka and Avvakumovka Rivers in Olginskii District. We have studied natural mineral solonetz formed from outcrops of zeolite, a mountain rock at Lazovsky Reserve, and solonetz deposits at mineral springs in Ussuriisky Reserve. In both cases, based on tracks and visual sightings, our observations confirm that sika deer actively visit these locations. According to Makovkin (1999), sika deer use the mineral springs on Sukhoi Creek year-round, although the most active period of sodium uptake is April and May. As many as 15–20 deer visit a site at a given time. In the area surround Melkovodny Bay most deer replenish sodium needs at a natural solonetz located about 2 km from the sea. The deer fearlessly visit this site, even during the day. In mid-May 1984, in 3 h of observations between 1400 and 1700 h, 17 deer made their way to the solonetz. Further evidence that this solonetz is used very intensively is the enormous pit that has been dug where deer eat the “salty” clay. During field research on the Avvakumovka River and its tributaries, we discovered bays with freshwater vegetation containing elevated concentrations of sodium; sika deer actively visit these areas.

Impacts on Sika Habitat by Humans Sika deer habitat has undergone serious anthropogenic changes since the midnineteenth century, but timber harvest is not the primary problem. The broad-leaf deciduous and coniferous-broad-leaved deciduous forests that sika deer prefer, as a rule, have very little commercial-grade ash or oak, the major timber species at present in Primorsky Krai. In fact, small patches of logging increase the forage opportunities for deer because grasses and sedge, oak seedlings, and young, broadleaved softwood species grow rapidly. Selective cutting of oak and the harvest of birch for firewood by local people has the same effect.

33 Sika Deer in Russia

489

The main human impact was the establishment of towns and villages, vacation areas, and camping spots along the seacoast in the last 150 years. The bays on the eastern shoreline of Primorsky Krai, from Vrangel to the village of Malaya Kema, have experienced human development in some sections so that access to the sea is blocked. This has seriously reduced the primary sika deer habitat. The bays suitable for the deer as additional sources of sodium are now intensively used by commercial fishermen, farmers, and vacationers. In last 150 years sika deer have lost around 15–20% of their best habitat.

Impact of Sika Deer on Their Environment Sika deer are conservative in selection of habitat. If left undisturbed, they will remain for a long period in a limited area. To collectively overcome difficult winter conditions, dozens, and sometimes hundreds of deer will gather in some areas. There are extensive hectare-sized fields in coastal oak groves on the eastern macroslope of the Sikhote-Alin Mountains where the deer give no relief to the young oak trees, foraging on young shoots. Some of the oaks die, others continue to attempt to put out new shoots. Along the seacoast from south of Olga to Nakhodka is a belt of oak forests with a park-like appearance that lacks the shrub and herbaceous strata. On southern, steep slopes protected from winter sea breezes, where deer group up in winter, the grass cover, herbaceous growth, and shrub layer are destroyed by deer feeding and trampling. Deer incessantly gnaw the bark of deciduous tree species such as maple, linden, and elm, similar to deer in Japan (Ando and Shibata chapter 15). Deer also congregate when seeking seaweed at the coast, resulting in destruction of forage areas and damage to soil cover. Thus, sika in Russia can have impacts on the environment as severe as those reported for Japan. According Makovkin (1999), deer have completely destroyed the understory on test plots set up by the Preobrazhensky Forest District in the coastal area of Lazovsky Reserve. Forest-forming species such as linden, Amur cork tree, Manchurian walnut, Japanese ash, Mongolian oak, various maple species, and others are entirely absent. Such species as aralia, Lonicera maackii (honeysuckle), Rubus komarovii (raspberry), Actinidia kolomikta (magnolia vine), Sumphoricarpos vulgaris (snowball berry), Philadelphus tenuifolius (syringa), Panax ginseng (Siberian ginseng), Acanthopanax sessiliflorus (acanthopanax), Euonymus sacrosancta (wahoo), and Ribes komarovii (currant) have disappeared. Corylus heterophylla and C. mandshurica (Siberian and Manchu) filbert trees are browsed down everywhere. To forage on the thinnest, and thus, the highest shoots, sika deer break the filbert stems, thus causing significantly more damage than by simply feeding. According to Konkov (1999) the area of very great damage to vegetation in Lazovsky Reserve now covers 33,000 ha, or one quarter of the reserve.

490

V. V. Aramilev

Makovkin (1999) also notes that in the coastal zone of the Lazovsky Reserve there has been no reforestation for almost 20 years, and a number of tree species have begun to disappear from the composition of the forest. Some simply reach a certain age limit and die; other small trees are weakened by frequent browsing, and are unable to withstand the autumn and winter winds and are blown down, disturbing the surrounding ground. Since the root area of an upturned tree can cover 10 m2, the damage to the forest is significant. Ugolny Bay illustrates the impacts of sika deer on a landscape. Herds of up to 90 deer can be found seeking sodium in April and May when the snow is deep elsewhere. The gentle coastal terrace, with its abundant grass and thin stands of bushes, is a preferred deer foraging area. Under the pressure of endless grazing and trampling, the vegetation at Ugolny Bay has disappeared, and following two catastrophic floods, the characteristic 25–30 cm thick dark brown topsoil layer has washed away to the bedrock (Makovkin 1999). Sika deer also exert a serious impact on the young growth of coniferous species. According to Makovkin (1999), deer have damaged 99% of the 1,532 Korean pines left at Ugolny Bay. Sixty percent have antler damage. Fourteen percent of the total damaged trees have died. The level of damage outside the reserve depends upon population density. Even so, when a deer herd remains in a limited area for an extended period of time, forage is damaged. A herd of 20–25 deer has repeatedly spent seven to 10 winter days on a 5-ha area near our field station in Avvakumovka River (Olginsky District). Following such extensive forage damage, wapiti and roe deer refrain from visiting the area for five or six years, until such time as the understory regrows.

Recent Sika Deer Density and Numbers Since the mid-1970s, in large part due to improved protection and several winters with little snow, there has been a significant and rapid increase in sika deer, both in its natural range and in reintroduction areas. By the early 1980s there were 1,400– 1,600 head in Primorsky Krai (Petrashev 1993), and by mid-decade the number was 2,500 (Khakhin and Prisyazhnyuk 1985), although these reported figures were significantly lower than official data which estimated sika deer numbers at 10,000. Numbers in the 1990s declined dramatically: 1991 = 7,100, 1992 = 6,000, 1993 = 5,000, 1994 = 4,700, and 1995 = 5,900 head. From 1989 until 1994 the number of sika deer steadily declined by an average of 10% per year. There was a significant 25% increase in 1995. Typical sika deer densities in 2002 (Aramilev and Aramileva 2002) are given in Table 33.1. Density declines as one move from south to north in the sika deer distribution because of natural reductions in the carrying capacity of the habitat. This general dynamic for sika deer numbers in Primorsky Krai is similar to what was also observed for roe deer, which were subjected to similar hunting pressures.

33 Sika Deer in Russia

491

Table 33.1 Sika deer densities in various habitat types (site survey in Olginsky and Lazovsky Districts, Primorsky Krai). Average density/1,000 ha (minimum–maximum) Survey Habitat type

area (ha)

North

Central

Hilly broad-leaf 20,000 2.0 (1.0–3.0) 9.0 (4.0–30.0) deciduous forest Low elevation broad20,000 7.3 (5.0–11.0) 6.8 (1.0–16.0) leaf deciduous forest Low elevation coniferous 20,000 1.0 (1.0–1.0) 3.0 (1.0–9.0) deciduous forest Note: North – Northern and central portion of Olginsky District Central – Southern portion of Olginsky District South – Lazovsky District (outside of the Lazovsky Reserve)

South 16.0 (7.0–26.0) 25.4 (12.0–77.0) 17.0 (1.0–61.0)

Competition with Other Ungulates A notable feature of sika deer in the south of the Russian Far East is competition with roe deer and wapiti (Makovkin 1999). The sika deer displace wapiti and roe deer from the sika’s preferred habitat. In some cases the sika deer have displaced goral (Nemorhaedus caudatus) as well. Sika deer compete with goral only in coastal areas. A high sika deer density will force the goral to forage on cliffs inaccessible to the deer. Habitats of roe deer in places where sika deer are found are broad-leaved forests, cedar-broad-leaved forests, and narrow-leaved forests on hills and low mountains. Also roe deer live in valley forests, including farmlands. Roe deer in Primorsky Krai live in open landscapes where sika deer do not live. Therefore, when sika deer occupy roe deer habitats, roe deer use areas and forages less suited to sika deer. Roe deer also began to shift into cedar-broad-leaved forests in mid-mountain elevations and in logged areas less used by sika deer. Another roe deer strategy to avoid sika deer is to retreat into low and mid-elevation coniferous forests that have a shallower snow pack; something significant for roe deer. Forage availability is less than in broad-leaf deciduous forests, but the underbrush and herbaceous growth along creeks provide a way for roe deer to survive in winters with deep snow. Even though sika deer are smaller in size and weigh less than wapiti, sika deer displace the latter into the mountains or to the north. There have not, however, been observations of direct aggression, so the displacement appears to be driven primarily by competitive exclusion. Observations of sika deer foraging amidst groups of wapiti are common. We have repeatedly found sika deer and wapiti foraging together with no aggression between the species being observed. We have also encountered wapiti bedding areas adjacent to the bedding areas of sika deer. Where sika have been introduced to areas of Europe and New Zealand, hybridization with red deer and wapiti has been a major concern (Bartos chapter 39; Swanson and Putman chapter 40; Banwell chapter 42). Here in Far East Russia, the only place where the natural ranges of sika and wapiti overlap, hybridization

492

V. V. Aramilev

also occurs. Female sika deer have been encountered in the wild with young male wapiti, and a hybrid between sika deer with wapiti, with the latter’s rack and other mixed features, has been encountered. Because the sika lineage split originally from the wapiti lineage (Tamate chapter 4) it is likely that the two species were isolated for considerable times during climatic changes in the past, and the current overlap on the extremes of their respective ranges represents a secondary contact. DNA analysis of 42 sika deer previously noted in the section on genetics showed that hybridization with wapiti does occur occasionally, but it is fairly rare (S. Goodman 1999 personal communication), as would be expected from the fact that sika deer usually displace wapiti from their common ranges. Thus, hybridization must be occurring only on the fringes of the sika deer distribution. This is similar to the situation reported of sika and red deer in the British Isles (Swanson and Putman chapter 40). No F1 hybrids (first generation) were found, only backcrosses, which indicate that the hybrids were fertile. Both species show introgression of alleles from the opposite species suggesting that hybridization is not a recent phenomenon, but has occurred over a long time period. Despite occasional hybridization, most individual deer retain the phenotype of sika or wapiti, much like has been reported when hybridization occurs where sika has been introduced into red deer (Swanson and Putman chapter 40) or wapiti (Banwell chapter 42) ranges in other countries. These results points to natural (“disruptive”) selection against hybrids with intermediate phenotypes. Thus, even though hybridization does occasionally occur, hybrid swarms of intermediate phenotypes have not developed, even though hybridization has been happening for a long time. This suggests that the sika and wapiti are “good” species, and should be retained despite some difficulties with the biological species concept in the strictest sense. Hybrids of sika deer and roe deer do not occur, and as previously noted, roe deer do abandon forage areas used by sika deer to occupy forest free valley areas and agricultural fields.

Predation As their range and numbers increased, the significance of sika deer as prey for the Siberian tiger (Panthera tigris altaica) and Amur leopard (P. pardus orientalis) also increased. Food habits studies in southwest Primorsky Krai in 2000–2001 show that fur and bones of sika deer are present in 67% of the scats of the leopards (n = 87); sika deer fur was found in 100% of the tiger scats (n = 27) in this region (author’s unpublished data). In sika deer habitat in the Sikhote-Alin Mountains 70% of the tiger kills were of sika. The other 30% of kills examined were wild boar and wapiti. These findings, from 2000 through 2002, were during a period when wild boar numbers in this area were high. It would appear that when wild boar numbers are low, the percentage of sika deer killed by tigers is even higher.

33 Sika Deer in Russia

493

Kucherenko (1985), who worked in sika deer habitat in Olginsky District in the 1970s, does not include the sika deer in his list of tiger prey. In a book about the Lazovsky Reserve (Valova et al. 1989), it is written that “… with an increase in tiger numbers within the Zapovednik, the role of the sika deer as prey also increases. Annually, within the Zapovednik and in adjacent areas, there are from several to a dozen incidents of sika deer killed by tigers.” This same publication states that “… lynx (Lynx lynx) inflict considerable damage in years with heavy snow, when the deer concentrate in small areas making hunting easier. In 1966, in three winter months, in the area of Proselochnyi, several lynx killed 12 deer, which was more than 10% of the herd that was overwintering in that area.” Pikunov and Korkishko (1990) showed that in the 1970s, in southwest Primorsky Krai, leopards only occasionally preyed on sika deer, but in the 1980s, the sika deer became its main prey. Incidents of leopards and tigers killing sika deer are regularly recorded during Amur leopard surveys (Aramilev and Fomenko 1999, 2000). In the 1960 and 1970s, when tiger numbers in the Sikhote-Alin were lower, wolves were reported to attack sika deer. As a rule, wolves preyed on sika deer at the end of winter when overflow ice appeared. Wolf attacks on sika deer in southwest Primorsky Krai have also been reported in recent years. However, as tigers have increased from earlier over-hunting, wolves have declined. In both Lazovoski Reserve (Makovkin 1999) and Sikhote-Alin Reserve (Miquelle et al. (2005) as tiger populations came back, wolves went to extinction, suggesting competitive exclusion. Thus, the predation picture for sika deer is complicated by variations in numbers of competitors and predators, and the human influences on all of them.

Hunting The first information on sika deer hunting in Russia is found in Przhevalsky (1870), Yankovsky (1882), Baikov (1915), and Arsenev (2003), in reports and other archive materials. According to Przhevalsky (1870), a herd of 40–60 sika deer was located in the outskirts of Vladivostok and sika deer meat for sale on local markets was significantly less expensive than beef. According to Yankovsky (1882), there were so many sika deer in the south of Primorsky Krai that up to 400 pairs of antlers could be taken. Herds reached several hundred head in some areas of southern Primorsky Krai and even the most mediocre of hunter could take five to 10 sets of antlers without much trouble; to shoot a couple of deer in the winter for meat presented no problem (Ryashchenko 1976). The Vladivostok Society of Sport Hunters annually shot deer on Askold Island to finance its activities. Between 14 and 19 sets of antlers were taken annually, with a pair selling for 80–300 rubles, this at a time when a cow cost 30 rubles. This was the Society’s main source of income (Vladivostok Society of Sport Hunters 1897). Few hunting rules or other restrictions existed at the turn of the twentieth century and sika deer were taken year round. Male deer were the primary target, taken for

494

V. V. Aramilev

their antlers, and this not only reduced total deer numbers but also changed the sex ratio in the population. Sika deer were a stable source of income for local people at the turn of the twentieth century, with people selling antler velvet, deer fetuses, tails, sinew, skins, and other parts. Meat and deer fat were used as food, tails and fetuses as medicine. A sika deer provides 40–50 kg of dressed game meat per animal taken. The hides were turned into chamois and sewn into clothing, including hunting wear. Deer were trapped in pits, driven onto overflow ice, and with the introduction of rifles, the harvest increased markedly. From then until the mid-1980s sika deer played little practical role in the local economy. Hunting was banned in Primorsky Krai, although poaching continued at a low level in many areas. Poaching occurred in Khasansky, Nadezhdinsky, Ussuriisky, Lazovsky, and Olginsky Districts and on Askold Island. At this point sika deer densities began to increase in several areas, reaching critical numbers that resulted in depletion of forage and massive die-offs in years with heavy snow (Lazovsky Zapovednik, Nezhinsky, and Borisovsky Hunting Collectives). Hunting of so-called farm bred sika deer living in the wild was again permitted in 1985. Following the opening of licensed hunting, the number and the range of the sika deer continued to grow and forage problems in areas where hunting was banned continued. Today in Russia all wild animals are the property of the state. In each administrative area of Primorsky Krai there are some hunting facilities; in some areas there are also general purpose hunting lands (not rented by any organization of hunters). In these lands it is possible to hunt deer, but regional hunting managers have to supervise the activity. In all hunting facilities annually in winter, when the hunting season is over, an estimate of harvested animals, including sika deer, is performed. Proceeding from these data, hunting managers of each facility estimate autumn numbers and recommend a hunting quota. On general purpose lands this work is done by the regional hunting manager. The data on number and required quotas is sent to the Primorsky Hunting Department in Vladivostok. The Department checks and analyzes the data and finally asserts a quota on harvesting of sika deer and other species across Primorsky Krai. They produce a general report for all Primorsky Krai and send the data to the Department of Hunting in Moscow. In Vladivostok the Primorsky Hunting Department distributes the determined quota to hunting facilities and to general purpose lands. The hunting facilities receive a quantity of licenses corresponding to the quota that are sold to hunters. The license is a legal form that gives a hunter the right to hunt, but also is a report form to fill in information on any animals taken. The information includes the sex, approximate age, weight of an animal, and the number of embryos if it is female. The hunter has to hand over the license to the hunting facility within five days after the end of the hunting season, and the facility turns them all in to the Primorsky Hunting Department in Vladivostok. There the data are analyzed to determine the kill. This system works for legal hunters. However, in Primorsky Krai there also are poachers who do not buy licenses and do not report animals taken.

33 Sika Deer in Russia

495

By our estimate, poaching of sika deer in different properties makes up from 10–40% of the total kill. Owing to massive, induced resettlement of more than 4,000 head of sika deer, few extremely snowy winters, and improved protection in Primorsky Krai, the number of sika deer in Russia in the second half of the 1980s grew to approximately 20,000, and in introduced areas alone, the number reached 10,000. The number of licenses issued for the 1994–1995 season was 648, which is 7.7% of the total postseason number of sika deer in Russia (Mirutenko 1996). The number of sika deer in the early 1990s, as with other ungulate species, dropped sharply due to decreased protection measures, increased poaching, and greater natural predation. According to Hunting Resources Census Bureau data, there were at this time 9,000–10,000 wild sika deer in Russia, including 3,700 wild sika located outside of their natural range (Mirutenko 1996) that is, fewer than were released in the European portion of Russia (Danilkin 1999). According to Primorsky Hunting Department data, the number of sika deer in the years 1996–2000 has tended to increase: 1996 = 7,900; 1997 = 7,400; 1998 = 9,590; 1999 = 8,160; 2000 = 12,000 head. Data on current numbers are derived from a census conducted in February and March at the end of the commercial hunting season (Mirutenko 1996). Primorsky Hunting Department data are from a “winter trail census.” (At winter routes ungulates prints of daily prescription are counted and by Formozov’ formula transferred into population density per unit area.) This method uses a relative count of numbers that, given conditions in Primorsky Krai, results in a total that is 1.5 times lower than an absolute count on fixed census plots. If official data are used as the baseline, then the post-commercial hunting season number for the year 2000 was around 12,000 deer (Table 33.2). These are figures for sika deer at hunting leases where hunting is allowed. There are also sika deer in all three nature reserves, on four wild-life refuges and in two administrative districts where hunting is not allowed. In the winter of 2001 a “census plot count” was conducted in various types of sika deer habitat in Olginsky and Lazovsky Districts. Sixty census plots, totaling 60,000 ha, were set up at random. The plots recorded 1,460 sika deer in the north of Olginsky District and 825 in south, with 4,950 in Lazovsky District (outside of the reserve), for a total of 7,235 deer. The winter trail method was used in three reserves and the numbers obtained were: Lazovsky, 700; Usssuriisky, 250; and Sikhote-Alin Biosphere Reserve, 150, for a total of 1,100 sika deer. Table 33.2 Pre-hunting season number of ungulates in Primorsky Krai, 1996–2000 (data from Primorsky Hunting Department). Winter trail census Species

1996

1997

1998

1999

2000

Sika deer Red deer Wild boar Roe deer Musk deer Moose

7,900 25,580 27,400 28,000 9,700 3,500

7,400 21,000 20,720 27,100 10,930 3,500

9,590 23,500 18,300 26,700 15,000 3,700

8,160 27,810 19,280 27,200 21,000 3,700

12,000 26,440 22,430 33,170 21,000 3,700

496

V. V. Aramilev

Counts were held at monitoring sites at wildlife refuges, but it is impossible to assess the quality of the counts. According to the Primorsky Hunting Department, the number for the Barsovy wildlife refuge is around 700, for the Borisovsky Plateau wildlife refuge at least 300, for the Vasilovsky wildlife refuge around 400, around 40 at the Poltavsky wildlife refuge and about 200 recorded in the Chernyie Cliffs wildlife refuge, for a total number of 1,640 sika deer. Thus, the total number of sika deer in Primorsky Krai is estimated at 22,000. We should note, however, that this is a post-hunting season count. With a 25% increase in numbers by the following autumn, the number of sika deer would increase to 31,250. If correct, the nineteenth century peak estimate of 25,000 head has now been exceeded. The goal of management is to have a sustainable harvest while maintaining relatively high, stable populations. Table 33.3 shows targets and actual take of sika deer in Primorsky Krai based on licenses returned to the Primorsky Hunting Department. Other species are provided for comparison. According to the Primorsky Hunting Department, due to the low rate of return of used licenses, these data underestimate the kill. When working at model hunting territories where all licenses are returned, the actual take indicated on the licenses varies, for various species, from 40–60% of the permitted take. This occurs because in Primorsky Krai a license to hunt ungulates has an element of prestige, so some hunters, those with a lot of money or influence, obtain licenses even if they don’t go hunting or have little ability. At the same time, local hunters will use one license to shoot several animals. Estimating the illegal take of sika deer is extraordinarily difficult, but judging by the continued dispersal of deer and an increase in their numbers, both the legal and illegal take, as well as natural mortality and predation, do not exceed the population’s replacement capacity. Our estimate is that hunters from the city currently take around 500 sika deer, with local hunters taking an equivalent number. A practical estimate is that between 500 and 700 licenses can be issued annually, taking into account the hunting success rate. According to data obtained at census plots in Olginsky and Lazovsky Districts, the post-hunting season number of sika deer for the region is around 7,200, and for

Table 33.3 Hunting targets and take for licensed species (data from Primorsky Hunting Department). 1996–1997 Species

Target

Sika deer 500 Red deer 1,300 Wild boar 2,100 Roe deer 2,000 Musk deer 550 Moose 100

1997–1998

1998–1999

1999–2000

Take

Target

Take

Target

Take

367 632 940 1,166 69 17

390 1,400 2,100 1,500 550 100

270 650 1,080 992 148 80

540 1,500 2,100 1,600 550 100

417 600 695 2,000 1,175 2,100 1,167 1,630 124 550 65 100

Target

2000–2001

Take

Target

Take

494 1,003 1,533 1,304 462 32

700 2,000 2,100 1,700 500 100

292 399 577 619 107 2

33 Sika Deer in Russia

497

the rest of Primorsky Krai, according to the data of the Primorsky Hunting Department, the number is around 12,000, for a total of 19,200. Given annual recruitment, set by the Primorsky Hunting Department at 25%, in autumn the number of sika deer in Primorsky Krai (excluding reserves) is 24,000. If hunting is allowed in Olginsky and Lazovsky Districts, with a take limit of 10% of the population, 2,400 sika deer can annually be taken in Primorsky Krai. This is almost five times the current limit. Such a volume would help to regulate deer numbers in relationship to forage capacity and increase the practical value of sika deer for sport and commercial hunting in Primorsky Krai. Sika deer populations currently have recovered, and management efforts appear to be holding the population approximately stable, with a sustainable harvest, both legal and illegal and predator kill. We can hope that these conditions continue so that sika deer contribute to the local economy while remaining in reasonable balance with their habitat, and support a natural take by native carnivores, including the threatened Amur leopard and Siberian tiger.

Literature Cited Abramov, K. G. 1928. Spotted deer—Basic information on antler horn management. AO Primorsky Zoopitomnik, Vladivostok. 149 p. (In Russian.) Abramov, K. G. 1939. Distribution patterns for spotted deer in Primorsky Krai. Practical Scientific Notes of the Chief Department of Zapovedniks. Issue 3, Pages 34–42. (In Russian.) Aramilev, V. V. 2000. Hunting management and large predators. Pages 116–123 in Reference book of hunting users and hunters. Vladivostok, Russia. (In Russian.) Aramilev, V. V. 2002. Pages 120–123 in Added sources of sodium in the ecology of the Primorye ungulates. Kirov, Russia. (In Russian.) Aramilev, V. V., and P. V. Fomenko. 1999. Simultaneous census of the Far Eastern leopard (Panthera pardus Linnaeus) in Southwest Primorsky Krai. Pages 12–13 in VI Congress of the Mammal Society. Moscow, Russia. (In Russian.) Aramilev, V. V., and P. V. Fomenko. 2000. Distribution and number of Far Eastern leopard in southwest Primorsky Krai. Pages 50–63 in Protection and rational use of plant and animal resources. Irkutsk, Russia. (In Russian.) Aramilev, V. V., and T. S. Aramileva. 2002. Pages 125–128 in Issues of protecting and using native sika deer populations in Primorsky Krai. Kirov, Russia. (In Russian.) Arsenev, V. A. 1949. Acclimatization of sika deer in European zapovedniks of the USSR. Collection 7, Pages 79–96. Environmental protection: Izd. VOOP, Moscow. (In Russian.) Arsenev, V. K. 2003. Expedition diaries of 1906. Notes of Society of Research of Primorsky Krai 30:1–88. (In Russian.) Baikov, N. A. 1915. In the mountains and forests of Manchuria. Petrograd. 120 p. Manchuriankrai Research Society. (In Russian.) Bogachev, A. S. 1982. Protection, acclimatization and domestication of deer. Nauka, Moscow. (In Russian.) Bogachev, A. S., and A. N. Alymov. 1990. Eco-genetic mechanism in the sika deer population. Pages 116–117 in Evolution and genetic research of mammals, Part 2. Dalnauka, Vladivostok. (In Russian.) Bogachev, A. S., G. Vakhreev, A. Velizhanin, and V. Troinina. 1983. Sika deer of Primorye. Hunting and Wildlife Management 2:12–13. (In Russian.)

498

V. V. Aramilev

Bromley, G. F. 1956. The ecology of wild sika deer. Pages 148–215 in Materials on research findings in government reserves. Nauka, Moscow. (In Russian.) Bromley, G. F. 1981. Sika deer (Cervus nippon Temmink, 1838) of Primorsky Krai—past and present population condition. Pages 93–103 in Rare and endangered terrestrial animals of the Far East of the USSR. Dalnauka, Vladivostok. (In Russian.) Bromley, G. F., and S. P. Kucherenko. 1983. Ungulates of the Russian Far East. Nauka (Science), Moscow. (In Russian.) Danilkin, A. A. 1999. Mammals of Russia and adjacent regions. Deer. GEOS, Moscow. (In Russian.) Dormidontov, R. V. 1977. Sika deer. Pages 10–28 in Ungulates. Lesnaya Promyshlennost, Moscow. (In Russian.) Emelyanov, A. A. 1927. Commercial animal species of the land of the Orochi. Production capacity of the Far East. Volume 4. Knizhnoe Delo, Khabarovsk-Vladivostok. (In Russian.) Geptner, V. G., A. A. Nasimovich and A. G. Bannikov. 1961. Mammals of the Soviet Union. Volume 1: Artiodactyls and perissodactyls. Vysshaya Shkola (Higher Education), Moscow. (In Russian.) Ilina, G. I. 1956. Ecological features of sika deer and possibilities for acclimatization in the European portion of the USSR. Scientific Notes of the Potemkin Moscow City Pedagogic Institute 61(4–5):77–89. (In Russian.) Khakhin, G. V., and V. E. Prisyazhnyuk. 1985. Current status and number of sika deer in the USSR. Ecological features of wildlife protection. Nauka, Moscow. (In Russian.) Konkov, A. J. 1999. Infringement of forest regeneration in places with intensive grazing by sika deer. Pages 238–239 in Woods and forest forming processes in the Far East. Dalnauka, Vladivostok. Kucherenko, S. P. 1985. Tiger. Agropromizdat, Moscow. (In Russian.) Maak, R. K. 1861. Travel through the Amur. 1855. Historical Account. Publishing House of Geographical Society, Saint Petersburg. (In Russian.) Makovkin, L. I. 1999. Wild sika deer in the Lazovsky Zapovednik and surrounding areas. Almanac “Russki Ostrov.” Dalpress, Vladivostok. (In Russian.) (Simultaneously published in English as The sika deer of Lazovsky Reserve and other territories.) Menard, G. I. 1930. Antler horn management. Gostorgizdat, Moscow-Leningrad. (In Russian.) Miquelle, D. G., P. A. Stephens, E. N. Smirnov, J. M. Goodrich, O. J. Zaumyslova, and A. E. Myslenkov. 2005. Tigers and wolves in the Russian Far East: Competitive exclusion, functional redundancy, and conservation implications. Pages 179–207 in J. C. Ray, K. H. Redford, R. S. Steneck, and J. Berger, editors, Large carnivores and the conservation of biodiversity. Island Press, Washington, DC, USA. Mirutenko, V.S. 1996. Sika deer. Pages 121–124 in Resources of main hunting species and hunting colectives of Russia: 1991–1995. Press of Rosselkhozacademy, Moscow. (In Russian.) Pavlov, M. P., I. B. Korsakova, and N. P. Lavrov. 1974. Acclimatization of game animals and birds in the USSR. Part 2. Volgovytskoe Knizhnoe, Kirov. (In Russian.) Petrashev, V. V. 1993. Introduction to neocenology. MSKhA Izdatelstvo, Moscow. (In Russian.) Pikunov, D. G., V. K. Abramov, and A. A. Skripchinsky. 1973. Several features of the distribution and protection of rare mammals in the south of the Far East of the USSR. Pages 25–28 in Rare mammal species of the USSR and their protection. Nauka, Moscow. (In Russian.) Pikunov, D. G., and V. G. Korkishko. 1990. The leopard in the south of the Far East. Nauka, Moscow. 192 p. (In Russian.) Prisyazhnyuk, V. E. 1974. Several biochemical indicators for the sika deer of southern Primorye. Sika deer of southern Primore. Izddatel’stvo Kyrgystan, Frunze. (In Russian.) Prisyazhnyuk, V. E. 1975a. Population census of sika deer in Lazovsky Reserve. Pages 1–58 in Ungulates of USSR fauna. Nauka, Moscow. Prisyazhnyuk, V. E. 1975b. Unique population of native sika deer in Primorsky Krai. Pages 240– 254 in Scientific articles of Nature Conservation Lab. Ministerstvo Selskogo khozyaistva SSSR, Moscow.

33 Sika Deer in Russia

499

Prisyazhnyuk, V. E. 1978. Morphological features of wild native sika deer and means to protect the animal. 24 p. Ph.D. dissertation, Institute of Ecology, Evolution and Morphology of Animals. (In Russian.) Resources of the main hunting species and hunting collectives of Russia: 1991–1995. 1996. L. K. Lomanov, editor. Press of Okhotdepartament, Moscow. Przhevalsky, N. M. 1870. Traveling around Ussorisky Krai in 1868–1869. Publishing House of Geographical Society, Saint Petersburg. Ryashchenko, L. 1976. Deer farming for antlers in Primorsky Krai. Dalnevostochnoe Knizhnoe Izdatelstvo, Vladivostok. (In Russian.) Sidorov, S.V. 1980. Dispersal of sika deer in the northern Caucasus. Pages 259–260 Ungulates of the USSR. Nauka, Moscow. (In Russian.) Suvorov, V. 1998. Icebreaker. Moscow. (In Russian.) Valova, Z. G., N. G. Vasilev, V. I. Zhivotchenko, L. I. Marovkin, T. I. Oliger, V. E. Prisyazhnyuk, N. P. Prisyazhnyuk, N. V. Solomkina, V. S. Khramtsov, and S. L. Shaldybin. 1989. Lazovsky Zapovednik. Agropromizdat, Moscow. (In Russian.) Vladivostok Society of Sport Hunters. 1897. Report for his imperial majesty Aleksandr Mikhailovich of the Vladivostok Society of Sport Hunters 1888–1896. Dalny Vostok, Vladivostok. (In Russian.) Wilson R. L. 2000. An investigation into the phylogeography of sika deer (Cervus nippon) using microsatellite markers. M.Sc. thesis, University of Edinburgh, Scotland, UK. Yankovsky, M. I. 1882. Sika deer, leopards and tigers of Ussuriiky Krai. Bulletin of the VostochnoSibersky Branch of the Geographic Society, Irkutsk. 13(3):76–79. (In Russian.) Zolotarev, N. T. 1937. Mammals of the Imana River basin. Izdatelstvo AN USSR, Leningrad. (In Russian.)

Chapter 34

Sika Deer Distribution Changes at the Northern Extent of Their Range in the Sikhote-Alin Mountains of the Russian Far East Inna V. Voloshina and Alexander I. Myslenkov

Abstract In this chapter we describe the shifts in the range of sika deer (Cervus nippon hortulorum) at the northern extreme of their range in the Russian Far East as influenced by natural limitations in habitat and the extremes in climate, and by human activities. We compare results from the Lazovsky Nature Reserve in the south Sikhote-Alin Mountains in the heart of the best deer habitat with those from the Sikhote-Alin Nature Biosphere Reserve and Zheleznyakovsky Refuge in the central Sikhote-Alin at the northern extreme of the sika deer range. Historical records from the 1920s show that later sika deer distribution and population numbers have been severely reduced, primarily by hunting and capture of animals for deer farms. Sika deer populations reached a low in the 1940s, at which time they were very rare in the north. Recovery followed in the subsequent decades, first by an increase in the pockets of survivors in the best habitat near the coast, with subsequent colonization of inland and more northern areas by dispersing animals. First dispersers were individual deer, some of which showed apparent seasonal migratory behavior. Permanent populations developed in areas with suitable yearlong habitat. The population is more dispersed in the south, where climate is more favorable and habitat more extensive. In the north deer are in fewer, more concentrated places near the sea coast in the best habitat and less severe winter conditions. A few sporadic, apparently migratory individuals occur during summer at inland locations. Average group size increased over time, as did maximum winter group sizes, which exceeded 100 individuals in the best wintering areas along the coast. By the 2000s, the original range, and most of the suitable habitat had been reoccupied.

Introduction In the Russian Far East, the range of sika deer, according to many authors (Abramov 1930, 1954; Bromley 1956; Heptner et al. 1961), has always had a mosaic structure. The areas of occupation are characterized by the best habitats for the deer, whereas the unoccupied habitats are deficient in some respect (Becklemishev 1928; Korenberg 1979). Still, patterns of distribution of sika deer vary over time due to D. R. McCullough et al. (eds.), Sika Deer: Biology and Management of Native and Introduced Populations, DOI: 10.1007/978-4-431-09429-6_34, © Springer 2009

501

502

I. V. Voloshina, A. I. Myslenkov

both natural (climate, vegetation change, etc.) and anthropogenic (human hunting, alteration of habitats for forestry and agriculture, capture or escape of farmed deer, etc.) factors. The Russian Far East, principally Primorsky Krai, is a particularly instructive place to observe the changes in distribution and numbers of sika deer over time. First, this is at the northern limits of the sika deer distribution, so natural factors have an especially important influence, with ebbs and flows in the distributional pattern being driven by climatic changes. Because the deer are at their physiological limits to adapt to local conditions, year-to-year variation in climate (especially winter snowfall) results in survival or mortality patterns that influence the distribution at the margins. Longer term trends influence the distribution on a broader scale, forcing deer downward in elevation, and southward in direction. The northern extreme of the distribution, thus, is usually a narrow strip, sometimes with interruptions due to locally inappropriate habitat, along the sea coast of the Sea of Japan. In the ice-age period of the Pleistocene, major climatic shifts certainly resulted in huge latitudinal shifts of the northern part of the distribution. It likely that such shifts due to climate change isolated the distributions of different subspecies of the red deer (Cervus elaphus) distribution and, consequently, led to differentiation of the sika deer from the parental stock (Tamate chapter 4). This isolation had to be of considerable duration of time to allow sufficient differentiation so that little hybridization occurred when climates allowed secondary contact of the two genetic stocks; indeed, the sika deer competitively displaces the red deer in Primorsky Krai, the only parts of the original ranges where the two species come into natural contact (Makovkin 1999). In addition to natural processes, the activities of humans have dramatically influenced the numbers and distribution of sika in the Russian Far East (Aramilev chapter 33). The two main factors were capture of deer for confinement to deer farms to produce antler (“panty”) for the Chinese medicine market and hunting for food from the late 1800s up to the end of World War II. The numbers of sika deer were severely reduced, and the distribution retreated far to the south in Far East Russia (Aramilev chapter 33). Shifts in the numbers and distribution of sika in Far East Russia due to natural fluctuations (local winter die-offs and the flux of the northern limits of the range) interacting with anthropogenic influences are particularly informative about the ecology and behavior of the species. In many ways the ebb and flow of the sika deer distribution in the Russian Far East is like an “experiment” which reveals the influence of various factors. In this work we attempt to extract those lessons. For information prior to the 1960s we are dependent on the historical records reported by field biologists, specifically the work of Abramov (1930), who covered much of Primorsky Krai, and Bromley (1956), who concentrated on the area currently included in Lazovsky Nature Reserve. These records suffer from the lack of a systematic approach and usually were of short duration. Still, they are helpful for setting baseline values that can be related to more recent work. In recent years we used two study areas to understand the changes in sika deer over time: the Sikhote-Alin Nature Biosphere Reserve and surrounding areas at the

34 Sika Deer Distribution Changes at the Northern Extent

503

northern extreme of the distribution, and Lazovsky Nature Reserve in the southeastern part of Primorsky Krai. In this area the sika deer survived in low numbers the severe hunting pressure up until the end of World War II and subsequently expanded to reoccupy the areas from which they were extirpated. Earlier literature was directed more broadly over the southeastern part of Primorsky Krai (Aramilev chapter 33), and we review this literature, particularly the distribution maps, to fit our work in Lazovsky and Sikhote-Alin Reserves into a larger time and geographical context. The objective of this chapter, therefore, is to present the depression of sika deer distribution at the beginning of the twentieth century and reestablishment by the beginning of the twenty-first by constructing a series of maps from 1928 to 2005.

Abramov’s Survey in the 1920s Abramov (1930) surveyed the distribution of wild sika deer and produced a map of their range in 1928. He estimated the number of sika deer in the wild at the time as around 1,000 individuals. He reported that the sika deer range had a mosaic structure with large clusters near the sea coast and smaller areas in the inland, as is apparent on his map (Fig. 34.1). He also stated that the sika deer mosaic of areas became progressively smaller from south to north, and the distance between areas occupied similarly became greater to the north. He suggested that sika deer moved between local areas only irregularly. At the time many deer were held in captivity in small peasant sika deer farms; Abramov (1930) reported that 105 farms held 1,530 deer in 1928. He learned that most sika deer in farms originated with wild deer that were driven into enclosures from the local forests. It is apparent that small peasant sika deer farms occurred in the same areas as wild sika deer (points on the map, Fig. 34.1). The sika deer range in the area planned for Lazovsky (Sudzukhinsky) Reserve had six small peasant farms at that time and the sika deer range in Mai-che basin (near the present Ussurisky Nature Reserve) had five peasant farms. The famous hunter and writer V. Yu Yankovsky reported that his grandfather M. Yu Yankovsky established the sika deer farm at the Sedimi (southern part of Primorsky Krai) peninsula in the same way; i.e., the deer were driven into the enclosure from the forest. This farm was expropriated by the Soviet government and was present throughout the twentieth century (Yankovsky 1993). In the thinking of the time, the wild population in the forest had to be retained as a refuge to maintain stocks of sika deer to establish new large deer farms planned by the Soviet regime. Consequently, in 1928 K.G. Abramov organized the SouthUssurian Refuge (called Sudzukhinsky Reserve from 1935 and Lazovsky Reserve from 1970), and in 1934 the Ussurisky Nature Reserve (near the town of Ussurisk in southwestern Primorsky Krai) for that purpose. Abramov’s investigation of the various districts of Primorsky was not equal. More observation was conducted in the Kievka and Chermaya River basins (the

Fig. 34.1 Abramov’s (1930) map of sika deer distribution in 1928 for the southern portion of Primorsky Krai.

504 I. V. Voloshina, A. I. Myslenkov

34 Sika Deer Distribution Changes at the Northern Extent

505

current location of Lazovsky Reserve), so he declared that population the largest. The Avvakumovka, Pfusung, and Vasilkovka Rivers were little surveyed, so large sika deer areas were absent on Abramov’s map. No observations at Yodzikhe River (Jigitovka), Sakhombe (Serebryanka) and Tavaiza (Russkaya) in Sikhote-Alin Nature Reserve resulted in blanks on his map, but small peasant sika deer farms also were absent there. The sika farm near Amgu village (Saion Bay) about 200 km north of Terney Bay was well known to Abramov, and he put the northern sika deer boundary near there. Despite these limitations, the value of the Abramov’s early map as a starting point is very great. As a result of Abramov’s map, we can observe 33 sika deer local groups and understand that it described only the southern part of the sika deer range in Primorsky Krai. A second map for 1948 was published by Bromley (1956). A third map for 1981 was published by Bogachev et al. (1983). Finally, Makovkin (1999) published a map of sika deer distribution according to his long-term investigations in Lazovsky Reserve from 1981 to 1996.

Lazovsky Nature Reserve Study Area Lazovsky Nature Reserve lies on the coast of the Sea of Japan (43° N, 135° S) between the Kievka and Chernaya Rivers. Chinese names of geographical places dominated from 1920 to 1970, so few Russian names were used by early biologists. The name of this reserve, Sudzukhinsky, came from the Sudzukhe River basin (now the Kievka basin) and the Chermaya River was previously the Taukhe River. Similarly, the name of the Sudzukhinsky Nature Reserve was changed to Lazovsky Nature Reserve in the 1970s, after the settlement of Lazo to the northeast of the reserve; the reader should be aware that Lazovsky is the same reserve that was previously called Sudzukhinsky. This and other reserves sometimes include “State” in the title, but for ease of understanding, the simpler forms (Lazovsky Nature Reserve or Lazovsky Reserve) are used here. The Lazovsky Reserve data base is 6,495 sika deer sightings by staff on the reserve from 1964 to 2006. The data base of our sika deer visual observations was compiled in the Microsoft Excel program. All sightings were described by X and Y coordinates. We constructed distribution maps of sika deer by decade in the Arc GIS 3.2a program. We then superimposed all maps over Abramov’s (1930) map for 1928 and examined the overlapping of existing range or newly established range with successive new maps. In addition, Makovkin (1999) did sika deer surveys with a KA-26 helicopter in 1988, 1989, and 1990. We also determined the distribution and numbers of sika deer by helicopter survey on 5–9 February, 2004 using a Mi-2 helicopter following the sample methods of Kuzmin et al. (1984) and Kuzyakin and Chelintsev (1987). This survey was conducted on Lazovsky Nature Reserve and the lands administered as hunting grounds adjacent to the reserve in Lazovsky and Olginsky Districts. This area includes the main habitats of sika deer in the southeastern Sikhote-Alin region.

506

I. V. Voloshina, A. I. Myslenkov

Fig. 34.2 Location of helicopter sample plots on Lazovsky Nature Reserve surveyed in February 2004.

The area was divided into 70 sample plots in systematic order and then in each plot a continuous (total) count was conducted (Fig. 34.2). Average size of the plots was 820 ha. In addition, the observers counted all animals seen from the helicopter while flying between plots. Helicopter flight elevation was about 100 m, and speed was 70–80 km/h. There were two qualified observers on each side of the helicopter. Aerial surveying began two days after a snowfall of 20 cm, so tracks were visible. Immediately after flights, all participants gathered together to discuss results, check records, and define the number of different individual deer counted, which were mapped on a large scale map (1:25,000). Statistical error of the sample counts was calculated by Chelintsev’s formula (Chelintsev 1984) using STATISTICA software for Windows (Stat Soft, Inc. 1999). Significance was calculated using Kolmogorov-Smirnov two-sample tests.

Results for Lazovsky Reserve Range of Sika Deer in the 1940s G. F. Bromley (1956) studied the sika deer range and ecology from 1944 to 1948 and constructed a map of local distribution on Lazovsky Reserve (Fig. 34.3). Bromley’s map showed the configuration of local groups in the reserve at a time

34 Sika Deer Distribution Changes at the Northern Extent

507

Fig. 34.3 Bromley’s (1956) map of the sika deer distribution in the Lazovsky Nature Reserve and adjacent areas.

508

I. V. Voloshina, A. I. Myslenkov

when small peasant sika deer farms were absent. He observed that the sika deer groups’ size was typically five to six animals; the largest group he observed in five years was 14 animals. Because population size and group size are usually correlated, this suggests a relatively low population. Bromley reported that the total population declined from 320 to 160 deer, which he attributed to the deep snow during the winter of 1947–1948. After he published these results, all researchers thought that the population decline applied to all wild sika deer, and the subspecies needed urgent protection. These results suggested that depression of the sika deer numbers took place at the end of the 1940s.

Range of Sika Deer since the 1960s Investigations of sika deer by the reserve’s staff began in 1964, and sightings have been systematically recorded ever since. From these records we assembled a summary of deer sightings by decade, which generally reflects changes in the population size (Table 34.1) and distribution of sika deer (Fig. 34.4) despite obvious variation in sightings over time due to the lack of systematic control of methods of observation. Not all sightings contained adequate information to assign a specific location, so the number of mapped locations is less than the total sightings as given in Table 34.1. From Table 34.1 and the distribution figures it is apparent that the sika deer population increased substantially from the 1960s to the 2000s, by nearly an order of magnitude based upon the increase in sika deer sightings (341 to 3,288). The low numbers of sika deer in the 1960s is further supported by the first sika deer aerial survey by helicopter which took place in March 1965 when 170 km were surveyed, and 71 sika deer were observed (Prisyazhnyuk 1966). This is much below the sightings by helicopter in the 2000s as discussed further below. The increase in sika deer numbers was accompanied by a spread from concentrated areas near the sea coast to inland locations, which accelerated in the latter decades. Animals that migrated to the continental inland were relatively few. In the 1960s most observations (89%) were concentrated in areas near the coast of the Sea of Japan. Only 11% was sightings of deer far from the coast in the summer time, all of single individuals. By the 2000s, approximately 40% of the population was found at inland locations (Table 34.1). Similarly, mean group size of sika deer increased until the 1980s and declined somewhat thereafter as deer were more dispersed over the interior parts of the range and relatively less concentrated in a few coastal areas. The largest group seen in the 1960s was 20 deer, whereas in subsequent decades groups of over 100 deer were commonly seen in winter concentration areas near the coast. By the 1970s sika deer were concentrated on the sea coast near Glazkovka and Valentin villages in the north and extended to the Ostrovnoi Cape and Chekhunenko Lake at the south of the reserve. One animal was observed to swim to Petrov Island. A preexisting local group had been on the island, but a newly established group appeared on the mainland coast around Petrov Creek.

34 Sika Deer Distribution Changes at the Northern Extent

509

Fig. 34.4 Distribution of sika deer in Lazovsky Nature Reserve in the 1960s through 2000s based upon field sightings by reserve staff.

510

I. V. Voloshina, A. I. Myslenkov

Table 34.1 Number, group size, and location (coastal or inland) of sightings of sika deer in the Lazovsky Nature Reserve. Decade Sika deer sightings Mean group size Coastal (%) Inland (%) 1960s 341 3.2 1970s 593 4.4 1980s 1,705 6.9 1990s 1,661 5.3 1,644 (3,288)a 5.2 2000sa a 2000–2005 years only, so the number of sightings was decades.

89 11 93 7 84 16 64 36 61 39 doubled to be equivalent to other

In contrast, in the 1970s the lowest number of sika deer was recorded in the Sikhote–Alin Reserve and surrounding areas (see below), so the recovery of the sika deer population began in the southern part of range and later spread to the north.

Sika Deer Range after the 1980s in the Lazovsky Reserve By the 1980s Uglovaya Bay was still a center of concentration, and large herds appeared in Proselochnaya Bay and Ezhovaja Bay (Fig. 34.4). Figure 34.4 shows both the growth of the area of concentration along the coast and the formation of new groups on the coast and, in addition, the appearance of new groups inland along the Krivaja River, the right tributary of the Kievka River. This group area appears on Abramov’s 1928 map (Fig. 34.1), so the process of reoccupation of the previous range began in the 1980s. Newly established groups at Kammenny Creek and the Benevka River appeared after 1985. This group was shown on Abramov’s (Fig. 34.1) map, so they were there in 1928, had disappeared by the 1940s, and had reappeared by 1985. Makovkin (1999) also discusses this spread of sika deer both to new and often previously occupied areas during the 1980s. A wild sika deer survey in Primorsky was initiated by Glavokhota RSFSR, Russian Game Services in 1981 (Bogachev et al. 1983; Bogachev and Zhoga 1985). They estimated the total number of sika deer in the Lazo District as 1,670, but L. I. Makovkin thought this estimate was too high. He estimated the total number in Lazovsky District was nearer 800 individuals. Expansion and increase in numbers of deer continued throughout the 1990s, and a greater percentage occurred at inland locations (36%; Table 34.1). These groups lived far from the sea coast all year round as winters were not hard during this time. There was a herd of 100 deer at the Maralovaya River, right tributary of the Sokolovka River, and 40 individuals in one herd on the Zvezdochka Mountain slopes. Large herds became common, and the largest herd observed in 80 years was seen in April 1995 when 250 sika deer were recorded at Petrov Bay near the sea coast. In the 2000s deer remained largely in the same areas, but the colonizing groups along the tributaries of the Kievka River extended farther inland along the valleys. In just five years of observation (2000–2005) 79 groups of more than 20

34 Sika Deer Distribution Changes at the Northern Extent

511

Fig. 34.5 Distribution of the sika deer in the central Sikhote-Alin Nature Reserve and adjacent areas by decades since 1960.

sika deer were observed. Most of the area occupied by sika deer on Abramov’s (1930) 1928 map (Fig. 34.1) had been re-colonized and most of the suitable habitat had been occupied.

2004 Helicopter Survey Estimates Although the relative increases in the size and distribution are useful, at best they are only an index to the actual numbers of sika deer. To produce a more definitive estimate of numbers systematic helicopter counts were conducted from 5 to 9 February 2004. A total of 25 plots scattered broadly over Lazovsky Reserve were sampled. Area of the sample plots was 185.25 km2, which was approximately 38% of the total area of 488 km2. The timing of helicopter surveys was planned to coincide with the formation of large groups of deer on the wintering areas, and following a fresh snowfall so recent tracks would be exposed. We did not place plots in mixed (pinebroad-leaved) forest because of low visibility and low density of sika deer there.

512

I. V. Voloshina, A. I. Myslenkov

In calculating the land area to which the counts were applied we excluded areas with no sika deer or with very low densities (settlements, fields, marshes, strip widths of 0.5 km along the roads, and ranges over 600 m above sea level). A strip width 1 km along the reserve’s border, where deer density is was less than on plots, was excluded from calculations of density and population size. The total sika deer number in the reserve was estimated at 4,139 ± 488 individuals, for an average density of 8.6 sika deer/km2. Density on Lazovsky Reserve was the highest of all areas similarly sampled in surrounding areas in Lazovsky and Olginsky Districts. Vasilkovsky Refuge to the north along the coast was next highest (6.2 deer/ km2), with adjacent hunting grounds having densities from 1.5 to 3.5 deer/km2. Deer density usually decreased with distance moved towards the north and toward inland areas. Thus, Lazovsky Reserve could be divided into three strata with significantly different (p < 0.01) deer densities: a coastal stratum in the Sokolovka and Proselochnaya River valleys (the original source group for expansion of the sika deer range over time), 14.1 ± 2.4 deer/km2 (n = 8 plots), a middle stratum of the valley of the Kievka River, 7.0 ± 1.3 deer/km2 (n = 11 plots), and an inland stratum in the upper reaches of the Kievka and Chernaya Rivers, 2.4 ± 0.7 deer/km2 (n = 6 plots). The helicopter survey was useful in giving specific numbers and density of deer to compare with the less systematic staff sighting records. However, the overall patterns of density and habitat preference were similar, giving credence to the usefulness of accumulated field observations. They give confidence that the results covering five decades shown in Table 34.1 are reasonably reflective of changes in sika deer numbers over time. In coastal habitats sika deer densities have exceeded the optimal level, and they are influencing the quality of their habitat. There is a visible lack of shrubs and tree seedlings, which have been suppressed by deer browsing, especially in winter. Makovkin (1999) discussed the impacts that sika deer were having on their habitat as well, so it is not just a reflection of very high deer numbers in recent years. This situation has persisted for 30–40 years, even though deep snow conditions were recorded in several winters. At first this might seem like a paradox: habitats with the minimum of woody plant winter forage support the maximum deer density. However, our observations during winter tracking of sika deer showed that, like the sika deer on Nakanoshima Island in Hokkaido, Japan (Miyaki and Kaji chapter 12), they consume the falling leaves. Also, oak crops of acorns are a significant part of the winter sika deer diet. The coastal habitats in Lazovsky Reserve are characterized by very high production of acorn crops, about 700 kg/ha on average. Low snow cover and southern exposures in coastal habitats provide the opportunity to feed from the ground during most of the winter.

The Sikhote-Alin Nature Reserve Study Area In previous publications we have reported that in the northern part of Primorsky Krai in the Sikhote-Alin Nature Biosphere Reserve and surrounding territories the numbers of wild sika deer have increased as has the size of groups as well (Voloshina and Myslenkov 1982, 1983; Voloshina 1997, 2001, 2002). The administrative center

34 Sika Deer Distribution Changes at the Northern Extent

513

of the Reserve is Terney, a settlement in the northern part of Primorsky Krai (45° N, 136°30′ E). Our results are based on spatial distribution of sika deer studied by the inspection of habitats from Jigit Bay to Blagodatnoe Bay and from the Skrytaya River to Zheleznyakovsky Refuge using route transects (Fig. 34.5). The routes were surveyed in January, March, and December. The total distance was about 470 km. The density of tracks was determined on each route. Visual observations of sika deer were conducted by aid of 12x binoculars and a 60x ZRT–457 telescope. The following information was collected in the database: date, place, X and Y coordinates, group size, sex and age composition, forest type, topographical relief, biological notes, and names of observers. Thus, 1,100 sightings were described. Visual observations of large mammals were carried out as part of the program “Letopis Prirody” (Chronicles of Nature) by the field staff of the reserve. From these 1,100 individual sightings a database of all sika deer sightings in Sikhote– Alin Reserve and Zheleznyakovsky Refuge from 1962 to 1999 was used to build maps of sika deer ranges every ten years from 1960 to 1999. Further processing was conducted by the programs ArcInfo, ArcView, and CalHome (Kie et al. 1994). In the program ArcView, a map of point range of sika deer without borders for 40 years was constructed. The program CalHome was used to calculate distances between sightings of sika deer and for the calculation of areas occupied by groups. The program allows use of the Adaptive Kernel and Minimum Convex Polygon methods. As a rule, we used the 95% inclusion map of available sighting points. These studies showed that there were two groups of wild sika deer in the Sikhote-Alin Reserve (Voloshina 1989). The south group occupied a coastal area about 21 km long and 2–6 km wide, while the north group inhabited an area 6 × 3 km near Abrek Mountain. In addition to changes in distribution we studied habitat use, group sizes, and use of eastern and western slopes of the Sikhote-Alin Mountains. In addition, helicopter surveys of the sika population were conducted in 1987, 1990, 1991, 1994 in the north and south parts of the Sikhote-Alin Reserve and Zheleznyakovsky Refuge.

Range of Sika Deer in the 1930s to 1940s Researcher Yu. A. Salmin recorded the first reliable visual sighting of sika deer in the northern part of Sikhote-Alin Reserve in 1937; he also observed sika deer along the Kunaleika River that is in the southern part of the reserve (Bromley 1981). Bromley saw a group of sika deer in the Tavaisa Bay in 1940. It is apparent that sightings of sika deer were very rare in Sikhote-Alin Reserve during this time.

Range of Sika Deer in the 1960s For the 1960s we have in total 14 points of visual sightings of sika deer (Fig. 34.5). They show three centers of distribution: around Jigit Bay, Blagodatnoe Valley, and

514

I. V. Voloshina, A. I. Myslenkov

Upolnomochennaya Valley. Exceedingly important were two sightings near Utesnaya Mountain. This location is to the north of Plastun settlement. These sightings were made in January 1965, before establishment of the Plastunskaya sika deer farm, so they represent a natural expansion of sika deer. The sighting of sika in Blagodatnoe Valley in September 1962 is also important because it happened as well long before establishment of the Plastunskaya sika deer farm. A series of sightings along the Kunaleika River in 1965 and 1966 suggest that a sika group inhabited this part of the Sikhote-Alin Reserve. A single sighting of a sika female in Upolnomochennaya Valley suggested the presence of a north group in the 1960s.

Range of Sika Deer in the 1970s For the 1970s we have 12 locations of sika; most of them are situated in the northern part of Sikhote-Alin Reserve, in Upolnomochennaya Valley (Fig. 34.5, Table 34.2). Therefore, the greatest depression of the sika deer population was probably in the 1970s. Of special interest are sightings of three single sika deer at extreme inland locations: (1) Spornyi Creek (basin of the Serebryanka River), (2) Ammonitnyi Creek in the same basin, and (3) Shandui Creek (basin of the Zabolochennaya River). These sightings suggest dispersal of sika far from the coastal group sources. For the calculation of area, the remote points were removed from analysis, so habitat area of the southern sika group was 320 ha. When we conducted an inspection of the northern area in 1974 for the establishment of the Zheleznyakovsky Refuge, we did not find any tracks or feces of sika deer. This refuge was founded in 1976. Occupation of Zheleznyakovsky Refuge by sika deer began only in the 1980s (Myslenkov and Voloshina 1989).

Range of Sika Deer in the 1980s In the 1980s, an increase in sika deer numbers and active dispersal was characteristic. These processes were accompanied by a significant rise of visual sightings. In the Sikhote-Alin Reserve and the Zheleznyakovsky Refuge, a total of 171 sightings of sika was recorded (Fig. 34.5). Blagodatnoe Valley and Upolnomochennaya Valley were centers of local groups. One of them, the source group in Upolnomochennaya Valley, Table 34.2 Habitat areasa of sika deer in the Sikhote-Alin Reserve and the Zheleznyakovsky Refuge. Decade Area of the south group Area of the north group Total area 1960s 24,970 700 1970s 320 1,077 1980s 16,760 8,997 1990s 15,450 16,320 a Results (in ha) are given by Minimal Convex Polygon analysis.

25,670 1,397 25,757 31,770

34 Sika Deer Distribution Changes at the Northern Extent

515

continued to grow. The dispersal into the Zheleznyakovsky Refuge had only begun (three locations). There was a first sighting of sika deer there in summer of 1987. A significant difference in habitat use between south and north groups occurred during the 1980s. At this time, the center of the south source group was formed around Ozernyi Creek. Habitat area of the south group, based on 95% confidence intervals, was 16,760 ha, but in the north only 8,997 ha (Table 34.2). Therefore, low densities and wide dispersion of sika deer were observed in the south group, and high aggregation in the north group.

Range of Sika Deer in the 1990s In the 1990s an increase of the sika deer numbers and dispersal of animals continued. In the surveys there was a clear trend of population increase (Fig. 34.6). At the end of the 1980s, the total number of the sika deer in the reserve was 80 animals, but in the 1990s it increased to about 360 animals. In 1998 and 1999 visual counts were made of only the north group, and the total number was estimated. The north group increased from 25 to 140 animals. The number of sika on the Zheleznyakovsky Refuge in 1994 was 28 animals, and in 1999, 38 individuals. During this decade a total of 540 sightings was recorded in the Sikhote-Alin Reserve and Zheleznyakovsky Refuge. Additionally, there were several sightings recorded outside the protected territories. The general range of sika deer was extended, but the centers of local groups remained in the former places. As a result, there were two new dispersal groups formed in the Zheleznyakovsky Refuge and in the southern part of the Sikhote-Alin Reserve (Fig. 34.5). The area of the north group was extended from 8,997 to 16,320 ha (Table 34.2), but the area of the south group did not show significant changes. 160

Number of animals

140 120 100 80 60 40 20 0 1987

1990

1991

1993 1998 Year Southern group Northern group Total

Fig. 34.6 Sika deer numbers by visual counts in the Sikhote-Alin Reserve.

1999

516

I. V. Voloshina, A. I. Myslenkov

It is further noteworthy that in Upolnomochennaya Valley very large groups of sika were observed in January and February of 1998, when herds of 100–125 animals were recorded. In the south group, group sizes were lower, with a maximum of 30 animals. We noted a distant location in the basin of the Armu River, which is a tributary of the Great Ussurka River on the western side of the Sikhote-Alin Mountains. This point of dispersal was more than 65 km removed from the nearest source group. In view of the disjuncture between the north and south groups, we calculated the land areas occupied by sika deer separately. The habitat area given by Bogachev and Zhoga (1985), 75,000 ha, is overestimated because these authors used the total range calculation and did not divide the sika range into two parts. Similarly we removed the summer isolated inland sightings from the calculation because these places were only temporarily occupied and their inclusion would have biased the area upward. The smallest habitat areas were recorded in the 1970s because of the depression of sika numbers in the south group (Table 34.2). The increase of the north group area occurred by the formation of a new colonizing group (Table 34.2). In the 1980s and 1990s, the only growth of sika density without an accompanying increase in area was recorded in the south group. During the 40 years we have had 12 sightings of sika far inland from coastal habitats. They were situated about 60–65 km from the source groups and observed only during the summer season. The characteristic features of inland habitats are relatively high elevations (up to 700 m above sea level) and coniferous forest cover. We suggest that the regularity of sightings of sika deer at these inland locations demonstrates a migration tendency in this local population. According to the literature, group size of sika deer in Primorsky Krai in the nineteenth century was a maximum of 60 animals. In the beginning of the twentieth century it was 25–30 animals (Przhevalsky 1870; Abramov 1930). In the Sikhote-Alin Reserve, the percent frequency of sika seen as single animals throughout the year varied around 20% for the decade 1990–1999 with the trend declining to 13% in 1997. Percentage of groups of two to ten individuals varied from 57% up to 76%. For this decade the percentage of groups of 11–20 individuals increased. Besides that, large groups of more 100 individuals were observed for the first time. A group of 125 animals was registered in January of 1998 in Upolnomochennaya Valley. To estimate average group size we use all data available for each year. Comparison of seasonal dynamics of aggregation shows that the average group size was highest in winter. Figure 34.7 shows the comparison of average group size over time from 1980 to 1999. In accordance with the increase in total numbers and the observation of very large groups, the average group size showed a general upward trend. Comparison of group size of sika deer in the Sikhote-Alin Reserve in the north with Lazovsky Reserve (data from Makovkin 1999) in south-eastern Primorsky Krai showed similar results (Fig. 34.7). The average group size varies from four to six individuals per group. It is slightly higher in general for Sikhote-Alin Reserve than Lazovsky Reserve, due to the somewhat more severe climate and restricted habitat near the sea coast; thus, sika deer in the Sikhote-Alin Reserve show a more concentrated pattern.

34 Sika Deer Distribution Changes at the Northern Extent

517

Fig. 34.7 Dynamics of average group size of sika deer in the Sikhote-Alin Reserve and Lazovsky Reserve.

Discussion In 1928 Abramov (1930) estimated the sika deer number in the wild as 1,000 and 1,500 in captivity. After his work the number of sika deer decreased to a minimum at the end of 1940s and remained depressed during 1950s and 1960s. The succession of maps from 1928 to 2005 for Lazovsky Reserve shows that the source areas remained in the same locations near the sea coast, but the population numbers began to build up. The reestablishment of sika deer groups in the depleted areas began in 1960s with the dispersal of single animals and summer migration. These dispersing deer formed permanent groups after 1975. Large and very large herds were present at the sea coast from 1970s to 1990s. Large herds appeared after 2000 in the inland colonization groups. The sika deer number in Lazovsky Reserve grew from 160 individuals (during the low period) to 4,139 ± 488 individuals according to the 2004 helicopter survey. The highest density of sika deer was in protected coastal areas of Lazovsky Reserve (8.6 deer/km2). Our long-term investigations and range analysis show that sika deer returned to the most of the original range indicated on Abramov’s 1928 map. For sika deer of the central Sikhote-Alin as typified by Sikhote-Alin Reserve, the growth of sika deer numbers happened first by an increase of population density in occupied areas, but not by extension of habitat area. Only later were new areas occupied by dispersing sika deer, usually first by single individuals. The sightings of single animals in the inland far from source group areas confirm the presence of migrations in sika deer through the formation of new dispersal groups, which become permanent if the habitat allows. The appearance of new local groups and the significant increase of population numbers over the last 40 years in Primorsky Krai illustrate the high dispersal capability of the species.

518

I. V. Voloshina, A. I. Myslenkov

Winter severity sets limits on the extent to which the sika deer can expand its range, as shown by the range of the most northern population extension of sika deer in the Sikhote-Alin Reserve and Zheleznyakovsky Refuge. Except for sporadic migratory individuals, sika deer are found in coastal areas only, while in the Lazovsky Reserve permanent groups occupy both the coastal and inland areas. Acknowledgements The authors thank the technician staff of the Sikhote-Alin State Reserve, V. Voronin, M. E. Borisenko, and N.V. Shershikov, for their help in the field, and S. Chihunov for making the sika deer distribution maps. Preliminary preparation of the Geographical Information System was conducted by A. M. Shevlyakov and V.A. Sushko on licensed programs from Hornocker Wildlife Institute, USA. Our thanks to assistants of Lazovsky Reserve, N. Tregub and D. Rtischev for constructing the data base.

Literature Cited Abramov, K. G. 1930. Sika deer. Elementary introduction to raising deer for antlers. Primorsky Zoopitomnik, Vladivostok, Russia. (In Russian.) Abramov, K. G. 1954. Ungulate animals of the Far East. Knizhnoe Izdatelstvo, Khabarovsk, Russia. (In Russian.) Becklemishev, V. N. 1928. Organism and community. Proceedings of the Biological Scientific Institute and Biological Station at Permsky University. Volume 1, Number 2–3. (In Russian.) Bogachev, A. S., and A. I. Zhoga. 1985. Restoration and rational use of the sika deer in the Central and South Sikhote-Alin. Pages 82–86 in A. A. Astafiev, editor, Jubilee session, devoted to 50-year Sikhote-Alin Biosphere Reserve (Abstracts of reports). Primorsky Research Society, Vladivostok, Russia. (In Russian.) Bogachev, A. S., G. I. Vakhreev, A. C. Velizhanin, and V. P. Troinina. 1983. Piatnistyi olen’ Primor’ia (Sika deer in Primorsky). Zh. Okhota I okhotnich’e khoziaistvo (Journal of Hunting and Game Farming) 2:12–13. (In Russian.) Bromley, G. F. 1956. Ecology of the wild sika deer. Pages 148–215 in P. B. Yurgenson, editor, Collection of results of study mammals in the state reserves. Izdatelstvo Ministerstva Selskogo Khozyastva, Moscow, Russia. (In Russian.) Bromley, G. F. 1981. Sika deer (Cervus nippon Temminck, 1838) in Primorsky Krai (Last and recent status of population). Pages 93–103 in Rare and endangered terrestrial animals of the USSR Far East. Dalnauka, Vladivostok, Russia. (In Russian.) Chelintsev, N. G. 1984. Mathematical bases of aerial survey of game species: Using of aviation for conservation and utilization of wildlife. Nauka, Moscow, Russia. (In Russian.) Heptner, V. G., A. A. Nasimovich, and A. G. Bannikov. 1961. Mammals of the Soviet Union, Volume 1. Vysshaya Shkola, Moscow, Russia. (In Russian.) Inukai T. 1952. The sika deer in Hokkaido and its prosperity and decline. Hoppo Bunka Kenkyu (Report of Northern Cultural Research) 7:1–45. (In Japanese.) Kie, J. G., J. A. Baldwin, and C. J. Evans. 1994. CALHOME home range program electronic user’s manual. U.S. Forest Service, Pacific Southwest Research Station, Fresno, California, USA. Korenberg, E. I. 1979. Biohorological structure of species. Nauka, Moscow, Russia. (In Russian.) Kuzmin, I. F., G. V. Khakhin, and N. G. Chelintsev. 1984. Aviation in the game industry. Forest Industry, Moscow, Russia. (In Russian.) Kuzyakin, V. A., and N. G. Chelintsev. 1987. Methodical notes about aerial survey of forest ungulates. Nauka, Moscow, Russia. (In Russian.) Makovkin, L. I. 1999. The sika deer of Lazovsky Reserve and surrounding areas of the Russian Far East. Almanac Russki Ostrov, Vladivostok, Russia.

34 Sika Deer Distribution Changes at the Northern Extent

519

Myslenkov, A. I., and I. V. Voloshina. 1989. Aerial surveys of ungulates in the Sikhote-Alin Reserve. Pages 66–67 in T. B. Sablina, editor, Ecology, morphology, use and protection of wild ungulates. VTO, Moscow, Russia. (In Russian.) Prisyazhnyuk, V. E. 1966. O vozmozhnosti aviaucheta piatnistykh olenei v Sudzukhinskom zapovednike (On the possibility of aerial counts of sika deer in Sudzukhinsky Reserve). Sbornik: Voprosy ekologii (Abstracts: Questions of ecology). Pages 240–241, Tomsk, Russia. (In Russian.) Przhevalsky, N. M. 1870. Journey in Ussuriisky Krai from 1868 - 1869. St. Petersburg, Russia. (In Russian.) Voloshina, I. V. 1989. Territorial distribution, numbers and density of population of wild sika deer in Sikhote-Alin Reserve. Pages 90–91 in T. B. Sablina, editor, Ecology, morphology, wild ungulates use and protection. VTO, Moscow, Russia. (In Russian.) Voloshina, I. V. 1997. Population dynamics of the wild sika deer in the central Sikhote-Alin. Rare mammals of Russia and surrounding territories. RTO, Moscow, Russia. (In Russian.) Voloshina, I. V. 2001. Sika deer (Cervus nippon hortulorum) range in Sikhote-Alin, Russian Far East. 25th International Congress of the IUGB, Nicosia, Cyprus. (Abstract) Voloshina, I. V. 2002. Extension of range and population dynamics of sika deer in Sikhote-Alin. Pages 180–181 in Capacity building for sustainable management of East Asia biosphere reserves. The 7th Meeting of UNESCO-MAB East Asian Biosphere Reserves Network (EABRN-7). Dalnauka, Vladivostok, Russia. Voloshina, I. V., and A. I. Myslenkov. 1982. Sika deer. Pages 263–264 in N. G. Vasiliev and E. N. Matyushkin, editors, Plant and animal world of the Sikhote-Alin Reserve. Nauka, Moscow, Russia. (In Russian.) Voloshina, I. V., and A. I. Myslenkov. 1983. Sika deer in the central Sikhote-Alin. Pages 165–167 in V. E. Sokolov, editor, Rare mammals and their protection. Nauka, Moscow, Russia. (In Russian.) Yankovsky, V. Yu. 1993. Tiger, sika deer, ginseng. Tropa, Moscow, Russia. (In Russian.)

Chapter 35

Sika Deer in Mainland China Dale R. McCullough, Zhi-Gang Jiang, and Chun-Wang Li

Abstract Of the 13 commonly recognized subspecies of sika deer (Cervus nippon) in the world, five were originally found in mainland China. Due to hunting and habitat loss and fragmentation, populations of wild sika deer declined to near-extinction. Today there are only three subspecies, C. n. hortulorum, C. n. sichuanicus, and C. n. kopschi, still living in mainland China; C. n. mandarinus and C. n. grassianus are thought to be extinct. The total number of wild sika deer in China today is approximately 8,500. In addition, about 290,000 sika deer are farmed in China. Of the three subspecies, the south China sika deer (C. n. kopschi) is the most endangered. Sika deer were listed by the International Union for Conservation of Nature and Natural Resources (IUCN) as Endangered in 1996 and 2003, and in the China Red Data Book of Endangered Animals as endangered. Therefore, they have been listed as a National Class I Protected Wild Animal Species of China. Since the 1970s more than 17 nature reserves have been established for saving the wild sika deer in Sichuan, Jiangxi, Zhejiang, and Gansu provinces, and in the northeastern part of China. Studies of genetics indicated that the small sika deer populations occurring in Qingliangfeng (Zhejiang Province) and Taohongling (Jiangxi Province) are somewhat different from the other Chinese populations, and they are deserving of special conservation effort. Human population pressures continue to impact the remnants of sika deer in the wild, and loss of genetic diversity is likely to occur in the remaining deer in isolated, small subpopulations.

Introduction Mainland China constituted the major part of the original range of sika deer, not only in geological times but also in recent years. Based on the fossil and subfossil record Guo and Zheng (2000) summarized the geographic history of sika deer in China. During the early Pleistocene, sika deer were only found in north China and Taiwan (Fig. 35.1). In the middle Pleistocene to Holocene they expanded their distribution over a broad region of central and eastern China. After the ice age, due to D. R. McCullough et al. (eds.), Sika Deer: Biology and Management of Native and Introduced Populations, DOI: 10.1007/978-4-431-09429-6_35, © Springer 2009

521

522

D. R. McCullough et al.

Fig. 35.1 The palaeogeographical distribution of sika deer in China (after Guo and Zheng 2000).

the continuous rising of Qinghai-Tibet plateau and developing agricultural civilization, sika deer lost habitat and their distribution area shrank once again. Interpretation of the map in Fig. 35.1 is complicated by the fact that climate changed repeatedly and dramatically during the Pleistocene and, undoubtedly sika deer distributions followed suit, morphing northward in latitude and upslope in elevation during warm periods, and reversing in cold periods. Thus extreme “outliers” in the northwest and west in Fig. 35.1 are a reflection of favorable periods far exceeding recent climate norms. A second problem in interpretation traces to the element of chance in discovering buried remains. This is influenced even further by preservation of sites, and given that sika deer were primarily adapted to lower lands and river valleys where natural erosion and agricultural modification were prevalent, physical remains were not likely to be retained to the same degree as in more protected sites. By the same token, however, it could be argued that deposits are more likely to be discovered because they are exposed by these forces more frequently and are in proximity of people.

35 Sika Deer in Mainland China

523

However these factors of preservation and discovery play out, we can conclude that the sika deer distribution expanded and then contracted again due to climate changes during the Pleistocene. The Holocene sites shown in Fig. 35.2 (after Guo and Zheng 1992) represent a minimum picture of the sika deer distribution over the historic human period during which climates were relatively stable. He (1994) suggested that Songming County in Yunnan Province was the southern edge of the sika deer distribution in mainland China during this time. Based on morphological characters, sika deer in the world were classified by Whitehead (1993) into 13 subspecies. Until the early nineteenth century, five subspecies were abundant throughout mainland China (Guo and Zheng 1992, 2000; Sheng and Ohtaishi 1993). Since the 1920s, two of these subspecies, C. n. mandarinus and C. n. grassianus, have gone to extinction in the wild (and perhaps in captivity as well) (Sheng and Ohtaishi 1993; Guo and Zheng 2000). The validity of the subspecific designation of these two forms is questionable, given the strong similarity of all northeastern China and Far Eastern Russian sika deer (Wilson 2000), but the

Fig. 35.2 Changes of distribution of Chinese sika deer in the nineteenth and twentieth centuries (after Guo and Zheng 2000).

524

D. R. McCullough et al.

absence of suitable samples for DNA analysis prevents a rigorous evaluation of the question. The surviving subspecies, C. n. hortulorum, C. n. sichuanicus, and C. n. kopschi, appear valid based on DNA evidence (Wilson 2000). Although C. n. hortulorum is still common in Far East Russia, in China all three subspecies currently survive in small, fragmented, and isolated habitats (Guo and Zheng 2000). Two recent studies found that genetic diversity of Chinese sika deer is low compared to that of sika deer in Japan (Wu et al. 2004; Lü et al. 2006). Most sika deer populations in China do not differ much from each other. Lü et al. (2006) related the low genetic diversity to the past history of Chinese populations being small and isolated. Thus, low effective population sizes led to genetic drift or founder effects. The genetic consequences of small population size have been profound over the long time period at which populations have been isolated, and this can only be overcome in the modern world by the purposeful translocation of deer between the remaining subpopulations, as proposed for tule elk (Cervus elaphus nannodes) by McCullough et al. (1996). Tule elk are distributed in some 22 small populations in California with virtually no probability of their dispersing naturally between populations because of the intervening human developments. Clearly any management of such subpopulations for retention of genetic diversity across the isolated populations requires detailed knowledge of the genetic structure of the set of populations. Most sika deer in captivity are of the northeastern (C. n. hortulorum) subspecies, and shows reasonable diversity (Wu et al. 2006). An exception to all Chinese sika deer being genetically similar, however, is the subpopulation near Lin’an in Zhejiang Province which, according to Wu et al. (2004), is sufficiently distinct from other Chinese subpopulations as to warrant separate conservation measures to preserve its uniqueness. This difference suggests a long period of isolation from other sika deer populations, including those in adjacent Jiangxi Province near Pengze, some 350 km west, samples of which were included in Wu et al. (2004). This result is surprising given the short distance between the two locations. There is a low mountain range separating Pengze and Lin’an. But before extensive human alterations of the habitat prevented dispersal of deer, there would have been a lowelevation connection eastward from Lin’an to the coastal plain surrounding Shanghai and neighboring cities and then extending westward to the Pengze area along the Yangtze River Valley and flood plain. No samples from the subpopulation in Anhui Province were included in the Wu et al. (2004) study. This is the population that is nearest to the Lin’an sika deer population (about 100 km distance across an approximately 1,000 m high mountain ridge). Usually such adjacent subpopulations, separated by minimal barriers, would be expected to be similar in genetic inheritance. The cause of the differing genetic cluster of sika deer at Pengze found by Wu et al. (2004) remains to be explained, but in the meantime, this subpopulation deserves special conservation attention. It is further notable that Wu et al. (2004) found that all Chinese sika deer are more closely related to the southern Japanese populations than to the northern Japanese population (Nagata chapter 3). This argues against two separate invasions of the Japanese archipelago from the mainland ancestral stock of sika deer, one from the north via land bridge through Sakhalin Island and one from the south via

35 Sika Deer in Mainland China

525

the Korean Peninsula. Rather, the results of Wu et al. (2004) suggest that there were two separate southern invasions via the Korean Peninsula over time, an earlier one that subsequently diverged into the northern Japanese sika deer type, and a later one, the southern type that retained more similarity with the current mainland and Taiwan populations. This hypothesis of two southern invasions at different time periods is similarly supported by nuclear microsatellite markers (Wilson 2000).

Historic Decline and Protection Wild populations of the Chinese sika deer in the late 1980s comprised no more than 1,500 individuals (Sheng and Ohtaishi 1993). The main factors that led to sika deer decline were human utilization and habitat loss. Sika deer have been hunted as a game species since ancient times in China. There are records of hunting of sika deer in the Shang dynasty (1562 to 1066 B.C.). Some parts of the sika deer body, such as antler, blood, penis, and placenta are used in Chinese traditional medicine. This plus the demands for meat and skins led to sika deer being heavily hunted over a long time. Due to their endangered status, hunting of wild sika deer is now legally prohibited. However, continuing impacts of human alterations of the environment and poaching threaten the survival of wild sika deer and, therefore, the species was listed in Category I of the State Key Protected Wildlife List in China and on CR/EN Category of the IUCN (Hu 1998). Under the national strategy of environmental protection and sustainable development planning, the State Forestry Administration of China (the national wildlife management authority) proclaimed 15 taxa of wild flora and fauna as key state protected taxa during the period from 2000 until 2050. About 17 nature reserves have been established for protecting wild sika deer and their habitats; most of these are national-level nature reserves. Almost all known wild sika deer and their habitats are under protection at some governmental level, and recovery programs for wild deer have been carried out. Thus, the immediate prospects of sika deer on the Chinese mainland are bright, due to improving habitat and wildlife protection law enforcement. Nevertheless, it will be a challenge to maintain free-ranging populations in the nature reserves in China because of continuing human encroachment and lack of sufficient management. In the meantime the reserves and deer farms almost certainly will maintain stocks of sika deer as sources of re-establishment to former ranges in the wild if, or when, that becomes possible. Because of the economic value of farmed sika deer and the long history of sika deer farming, most studies in Chinese sika deer are focused on physical and biochemical characters, velvet antler processing methods, and medical effectiveness of deer products, while the knowledge of biology and ecology of wild sika deer is limited (Song and Liu 2005). However, recently studies on ecology, genetics, and management of wild sika deer have increased and there is a growing literature on these topics.

526

D. R. McCullough et al.

Sika deer were domesticated in ancient China, beginning no later than 370 A.D. However, captive sika deer were first exploited on a large scale as an economic animal in the Qing dynasty from 1733 (Zhao 1990). The first farming of sika deer for esthetic purposes (similar to zoo animals) occurred in Jilin Province in the northeastern part of China. The deer were captured from the wild. As captive raising of sika deer gradually developed, people began to use sika velvet antlers as traditional Chinese medicine and to consume the game meat. From then on, the captive sika deer were introduced from the northeastern part of China into many other regions of China. Therefore, most sika deer raised in farms in China were derived principally from the subspecies C. n. hortulorum (Wu et al. 2004). Considerable numbers still exist in deer farms today. With a few exceptions, deer farms are typically small, being mainly family enterprises with fewer than several dozens of animals, most of which are held in small confined pens rather than freeranging on pastures. Still, the long history of captive rearing of sika deer and the commercial trade in stocks means that interchange of genetic stocks may have occurred. While most of these shifts of sika deer stocks were probably on a local scale for ease of transportation that does not mean that longer translocations did not occur, even in earlier times. Barges moved along the major rivers, the Yellow and Yangtze, for example, and a canal system over 300,000 miles in total length was developed early in history for moving products long distances. The Grand Canal was started in the sixth century and eventually connected Beijing in the north to Hangshou (Hangzhou) (near Shanghai) in the south (Warshaw 1994). Clearly the potential to move sika deer long distances existed for centuries. These translocations underscore the importance of establishing the natural genetic structure of sika deer over their whole range before such intermixing complicates the picture even further (Wu et al. 2004). The gross pattern of the Holocene distribution shown in Fig. 35.1 largely matches what we know about sika deer in other parts of their range at similar latitudes. Sika deer were most prevalent in the low western plains of the Yangtze and Yellow River valleys, and coastal areas. This was the “rice-bowl” of China from ancient times. We can infer fairly reasonably that the northern and northeastern limits, extending mainly along river valleys, were set by winter snow depth and can be approximated by the 500–600 m elevation contour, much like the modern distribution of sika deer in Far East Russia. An anomaly in the distribution to the west occurs in the Sichuan Basin, where deer were known to have occurred in good numbers historically, and remnants still persist yet today. The elevation of the basin, around 2,500–3,000 m, is much higher than other parts of the sika deer range. Sika deer can withstand cold temperatures; their difficulty is winter snow depth. This adaptation was probably enhanced by the fact that the continuous rising of Qinghai-Tibet plateau, which isolated the Sichuan Basin occurred gradually. The reason sika deer can occupy this area seems to be related to relatively moderate climate, despite the elevation, and low winter precipitation. Sichuan Province has the highest human population of any province in China and has been a major agricultural area from ancient times. The same variables that made the area suitable for humans made it good habitat for sika deer.

35 Sika Deer in Mainland China

527

In coastal and south China climatic conditions favored sika deer, but much of this area is broken topography, with low to moderately high mountains intersected by relatively narrow river valleys. Thus, sika deer distribution was broken as well, isolated into patches and stringers along river courses. Thus, on a mapping scale, what is shown as a “continuous” distribution of sika deer over the region is actually at ground level a complex mosaic of bits of sika deer occupation interspersed with marginally suitable or unoccupied matrix. With human development the landscape occupied by sika deer was further fragmented, probably resulting in a metapopulation structure until total isolation of remaining fragments of occupied areas in the late nineteenth century.

Current Status While the sika population in deer farms is booming, wild sika deer are extinct over most of the original range, but still exist in northeast China, southern China, and western China. Wild sika deer have two origins: those that survived in their original habitat remnants and those that escaped from deer farms and established wild populations. The State Forestry Administration (SFA) organized a nationwide terrestrial vertebrate survey from 1998 to 2000 (SFA 2001). According to results there were 8,427 sika deer of three subspecies left in the wild in eight provinces (Table 35.1). They reported 4,500 in the Gansu Province, 720 in Sichuan Province, around 700 in Jiangxi Province, 680 in south Anhui, 200 in north Zhejiang, around 500 in Jilin Province, and 27 in Heilongjiang Province. They also reported 287,857 sika deer in deer farms in 28 provinces, municipal regions, and autonomous regions (Table 35.1).

Northeastern China Wild sika deer are distributed in the Changbaishan region, Dongfeng County, Liaoyuan County, the Meihekou region where they were on royal hunting grounds with high sika deer density in the Qing Dynasty (1644 to 1912), Shuiyang County, Dongning County, and Ningan County of Heilongjiang Province, which neighbors Jilin Province. This was once the major range of the subspecies C. n. hortulorum. The existing habitats of sika deer are rather fragmented. In Jilin Province, wild sika deer occur in northeastern China adjacent to the Russian and North Korean borders. Sheng and Ohtaishi (1993) considered this population questionable and functionally extinct. It consists of sporadic isolated individuals in remote habitat that are likely dispersers across international borders (Russia and North Korea) in a source-sink system (Pulliam 1988). Because the populations are more stable in the countries across the Chinese border in this area, particularly in Russia, individual sika deer disperse from higher density across the border to lower density, where they face a high mortality rate. There is little dispersal in the reverse direction, so the Chinese isolates wink on and off depending on the frequency, place, and timing of dispersal events.

528

D. R. McCullough et al. Table 35.1 Sika deer in mainland China (SFA 2001). Province or region Number in wild Number in farms Beijing Shanghai Tianjing Chongqing Hebei Shanxi Inner Mongolia Liaoning Jilin Heilongjiang Jiangsu Zhejiang Anhui Fujian Jiangxi Shandong Henan Hubei Hunan Guangdong Guangxi Sichuan Guizhou Yunnan Tibet (Xizang) Shaanxi Gansu Ningxia Qinghai Total a For display in a zoo.

0 0 0 0 0 0 0 0 500 27 0 200 680 0 700 300 0 0 0 0 0 720 0 0 0 0 5,300 0 0 8,427

2,695 64 4,079 646 6,527 50,000 70 15,000 178,000 6,200 3.348 1,190 4,200 1,000 1,000 5,182 556 2,000 923 643 276 358 272 2,000 13a 322 354 560 379 287,857

Consequently, sika deer on the Chinese side of the border are locally quite sporadic and unpredictable in location and abundance. For example, a track survey done in 1999 in Heilongjiang Province adjacent to the Russian border only found sika deer tracks on two survey lines (3% of all survey lines): there were three sika deer tracks on one survey line, and 43 sika deer tracks on the other survey line (Sun et al. 1999). In Dongfeng and Lishui Counties of Jilin Province there are feral sika deer populations that were established by escaped individuals from sika deer farms (SFA 2001). Wild sika deer are distributed in the state-owned forests in the Shuiyan-Mulins area at the extreme southern tip of Heilongjiang Province, just across from the Russian border. The range of wild sika deer in Heilongjiang is about 1,000 km2. Compared with the distribution range of sika deer recorded in a survey on the state-owned forest in the Heilongjiang Province in 1992, there were 41 (range 21

35 Sika Deer in Mainland China

529

to 56) wild sika deer on a range of about 2,000 km2. At the end of last century, the number of wild sika deer decreased to 27 (range 21 to 33), with average density of 0.027 ± 0.006 deer/km2. Given the presence of sika deer on the border further north in the vicinity of the Russian city of Dalnerechensk (Aramilev chapter 33), cross-border dispersal into China probably occurs occasionally in this area as well. Sika deer also occur near the border in Jilin Province, which lies to the south of Heilongjiang Province along both the Russian and North Korean borders. Wu (2000) reported that sika deer occur adjacent to the Chinese-Russian border occasionally, and a similar cross-border intrusion occurs in the Paektu Mountain area in the Changbaishan Mountain range on the Chinese border with North Korea (Sheng 1990; Korean People’s Democratic Republic 1998; McCullough chapter 36). These results point to an unsustainable population in the northeastern part of China, and it persists only because of cross-border dispersal of sika deer. Only if China institutes more effective protection measures will a persistent population be possible. Fortunately, some reserves, for example, Jilin Changbaishan National Nature Reserve and Jilin Baihe Nature Reserve, have been established for protecting endangered animals (including sika deer) and the natural landscape of this region of northeastern China adjacent to the Russian and North Korean borders (Piao et al. 1999). These reserves should be beneficial to saving northeast China sika deer in coming years.

Western China Wild sika deer are distributed in Gansu Province and Sichuan Province. Originally sika deer occurred over most of the Sichuan Basin and southern Gansu Province, but today they have been reduced to several populations in the area near the small extension of Sichuan Province into Gansu Province. During the field survey organized by the SFA (2001) a large number of wild sika deer were found in Wenxiang County and Wudu County of Gansu Province. The deer farm in Tianshui County and the Ziwuling Land Reclamation Bureau introduced sika deer from Jilin Province in 1958 and 1963, respectively. However, some of the introduced sika deer escaped to the wild. The estimated number of sika deer in the wild was 5,300, belonging to the Sichuan sika deer and northeast sika deer subspecies (Table 35.1). However, further detailed information of the sika deer in Gansu Province is unavailable (G.-L. Zhang, personal communication 2007). The largest population in Sichuan Province occurs on Tiebu Nature Reserve (Li and Zhao 1989). This population has received the most attention and has been studied by Chinese biologists for some time (Guo 2003). The other two populations are more isolated and smaller, and much less is known about them. One is estimated to contain 130 to 150 sika deer, whereas the most distant population, which survives in a mountain valley, is estimated at 30 to 45 deer (Guo 2000). The most recent estimate of sika deer in Tiebu Nature Reserve is 287 (Guo 2003). SFA gives a total of around 720 sika deer remaining in Sichuan Province (Table 35.1).

530

D. R. McCullough et al.

Southern China The south China subspecies of sika deer (C. n. kopschi) is the most endangered of all the sika deer subspecies in the world. The distribution area of south China sika deer is divided into three main isolated areas in Jiangxi Province, Anhui Province, and Zhejiang Province. South China Sika deer are distributed in Jingxian, Jinde, Yixian, Shexiang, Huangshan, Qimeng, Tiantai, Guichi, Qingyang, Nanling and Ningguo Counties of Anhui Province, Pengze, Hukou, Jiujiang, Yongxiu, Boyang, Anyi, Fenxing, Leping, Fuliang Counties of Jiangxi Province, and Lin’an County of Zhejiang Province. These populations live in valleys in low, broken mountainous habitat at relatively low elevation in the range of 300–600 m. Little is known about sika in Anhui Province due to its sporadic distribution. Wu et al. (2003) conducted a field survey in 2002 and reported that 70 to 90 sika deer occupied about 20 km2 at around 800 m elevation near the town of Ningguo. They noted that protection measures were essential to save this population. A total of around 1,580 sika deer still live in southern China (Table 35.1), but they occur at isolated places too far apart for natural dispersal and, in most places, still require more effective protection from poaching and alteration of habitat by people and their livestock.

Eastern China There are two main feral sika deer populations in Shandong Province. In 1984, a group of sika deer introduced from Jilin Province was released into the Liugongdao Island National Forest Park; recent census indicated there were about 200 sika deer in 223 ha forests in the park on the island. At Weideshan Mountain in Rongcheng City, Shandong Province (about 231 km2 in area) sika deer escaped from a deer farm into the mountains. Recent census found about 100 feral sika deer in the area.

Farmed Sika Deer Sika deer are farmed all over China except the Tibetan Autonomous Region, Xijiang Urger Autonomous Region, and Hainan Province (Table 35.1). Sika deer bred in captivity number more than 280,000 head. The largest scale sika deer farming is in Jilin Province, where more than 50% of the farmed sika deer are bred. Next most important are Shanxi and Liaoning Provinces, with 50,000 and 15,000 sika deer on farms, respectively. The remaining provinces, such as Hebei, Heilongjiang, Shandong, Tianjing, and Yunnan, have less than 10,000 farmed sika deer each. Some wellestablished varieties of farmed sika deer, such as Shuangyang sika deer, Changbaishan sika deer, and Xifeng sika deer, were selected and bred. Those sika deer breeds have been introduced to many sika deer farms in China (Li and Xue 2000).

35 Sika Deer in Mainland China

531

Although sika deer farms in China are prosperous, there is still concern about the degeneration of farm-bred sika deer (Wu and Zhang 2001; Lei 2007). To maintain genetic diversity of the farmed sika deer, capture of wild sika deer to recruit wild blood to farmed sika deer occurred frequently over the history of sika deer farming. However, now that the population of wild sika deer is low, how the genetic health of the farmed sika deer will be maintained remains a question. Nevertheless, we should enhance the protection of the wild sika deer through in situ and ex situ conservation practice as the first priority. Wu and Zhang (2001) suggested that some endangered and small populations of wild sika deer may be enhanced by release of farmed deer. Where wild sika deer populations are secure, they can help farmed deer to acquire some good characteristics again by crossbreeding the wild and farmed sika deer.

Conservation and Management In 2000 the State Forestry Administration announced 15 wild flora and fauna taxa as the key protected taxa in the 2000 to 2050 State Wildlife Species Protection Program. Deer are among the 15 key protected taxa, and the sika deer is one major conservation target species of the deer in China. Altogether, 17 nature reserves have been established in China to protect wild sika deer and their habitat, covering an area of 424,485 ha, although some of the nature reserves are not solely for the protection of wild sika deer (Table 35.2). Among these reserves, ten are national while seven are overseen by provincial, county, or city level wildlife management authorities. The function of all existing nature reserves for protecting southern China sika deer will be enforced through habitat restoration, in-service training of the reserve staff, and strengthening logistic support, according to the 2000 to 2050 State Wildlife Species Protection Program (SFA 2001). Now, seven years later, part of the goal set in 2000 to 2050 State Wildlife Species Protection Program has been achieved.

Northeastern China Sika deer live in temperate broad-leaved or mixed broad-leaved and coniferous forests and meadows in northeastern China. Now, the Tiexi Nature Reserve of 7,200 ha in Heilongjiang Province and the Siberian Tiger Nature Reserve have been established. The two reserves are not inclusively for the protection of wild sika deer; however, the reserves provide protection of wild sika deer and its habitats. According to the State Wildlife Species Protection Program of 2000 to 2050, the natural habitat in southern Laoyeling will be under protection by the state. A population of wild sika deer is planned for the area, with captive-bred sika deer released into the wild. The natural habitats of wild sika deer in Hunchun, Changbaishan,

Baishuijiang NR

Gahai-Zecha NR

Jiangxi Taohongling NR Longtangshan NR Qinglianfeng NR Jixi Qiangliangfeng NR Laoshan NR Lianhuashan NR

Guiqingshan NR Xaingshan NR Tiexi NR

Siberian Tiger NR

Tiebu NR

Longwangshan NR

Yaoli NR

Weideshan NR

2

3

4

8 9

10 11 12

13

14

15

16

17

5 6 7

Guniujiang NR

1

Name of the Reserve

3,767

1,224

20,587

80,000

2,134 11,330 7,200

18,000 11,691

3,900 10,800 3,000

12,500

10,800

213,751

7,134

Area (ha)

Rongchen City, Shandong Province 6,667

Fuliang County, Jiangxi Province

Ruoegai County, Sichuan Province Anji County, Zhejiang Province

Guichi City, Anhui Province Lintan, Zuoni, Kangping, Weiyuan County, Gansu Province Zhang County, Gansu Province Li County, Gansu Province Mishan City, Hewilongjiang Province Hunchun City, Jilin Province

Pengze County, Jiangxi Province Linan City, Zhejiang Province Linan City, Zhejiang Province Jixi, County, Anhui Province

Qimeng County, Shitai County, Anhui Province Wenxiang, Wudu County, Gansu Province Luqu County, Gansu Province

Location

1985

2000

1985

1965

2000

1992 1992 1996

2001 1983

1998 1998 1986

1981

1998

1978

1988

Year of establishment

County NR

County NR

Provincial NR

National NR

National NR

Provincial NR National NR Provincial NR

Provincial NR Provincial NR

National NR National NR National NR

National NR

National NR

National NR

National NR

Status

Sika deer, birds

Sika deer, evergreen broad-leaved forest Sika deer, black muntjac

Sika deer, musk deer and its habitat

Tiger, red deer, musk deer, red deer

Forest ecosystem, sika deer Forest ecosystem, sika deer Red deer, sika deer, secondary forests

Sika deer and its habitat Sika deer, black muntjac Sika deer, black muntjac, subtropical evergreen broad-leaved forests Sika deer, black muntjac Red deer, sika deer

Sika deer, black muntjac, subtropical evergreen broad-leaved forests Giant panda, sika deer, forest ecosystem Sika deer, black-necked cranes, wetlands Sika deer and its habitat

Main conservation targets

Table 35.2 Nature Reserves in China designated for protecting sika deer or that contain sika deer within their boundaries.

532 D. R. McCullough et al.

35 Sika Deer in Mainland China

533

Dunhua, Antu, and Fusong counties of Jilin Province, have been planned or have already been established as nature reserves. A nature reserve for protecting the feral sika deer population in Dongfeng County, Jilin Province has been planned. Five wildlife protection stations and six monitoring posts will be established in Heilongjiang Province, and an additional five wildlife protection stations and six wildlife monitoring posts will be established in Jilin Province. Their mission is to protect and monitor wildlife including wild sika deer.

Western China A recent survey indicated that a large number of wild sika deer, including feral ones, live in Gansu Province. They live in habitat at the edge between alpine coniferous and broad-leaved forest and mountain meadows with sparse shrubs, at elevations ranging over 2,000 m. They occupy the edges of coniferous broad-leaved mixed forests, where in winter they forage on the southern slopes in shrublands. In spring and autumn sika deer are often seen grazing on open grassland patches on mountain slopes. Baishuijiang National Nature Reserve is located in southern Gansu Province; its area is over 2,000 km2. The primary target species for the reserve are giant panda (Ailuropoda melanoleuca) and takin (Budorcas taxicolor), but the reserve also protects the wild sika deer within its boundaries (Dong et al. 2002). The main target of the Gahai-Zecha National Nature Reserve of about 108 km2 is to protect wild sika deer and their habitat. Other nature reserves, the Lianhuashan National Nature Reserve and Zehai-Zecha Nature Reserve in southern Gansu Province, also protect 116 and 108 km2 habitats of sika deer, respectively. Guiqingshan Nature Reserve in the Zhang County and Xiangshan Nature Reserve in Li County of Longnan City, Gansu Province also have sika deer. However, the population status of the sika deer in Gansu is the least known among all wild sika deer populations in China. So far no research has been done on the wild sika deer in Gansu Province, or on the feral sika deer in either Gansu and Shandong Provinces. A field survey on population and habitat status is needed. Lack of information obviously impedes management. More attention and resources need to be allocated to the wild sika deer in Gansu and Shandong Provinces in the future. In contrast, the Tiebu Nature Reserve population is the best-studied sika deer herd in China. A field survey in 1975 showed that there were about 500 sika deer in the Tiebu area of Sichuan Province (Sheng 1991), so the Tiebu Nature Reserve was established in 1983 to protect this population. The reserve encompasses 205 km2. Tiebu has good sika deer habitat, being a mix of forests and meadows, and the edges between the two are prime sika deer areas. The reserve lies at 2,500–3,000 m, unusually high in elevation for sika deer, but it lacks deep snows because of the low winter precipitation. Unfortunately, the reserve also contained over a thousand human residents (villages are scattered over the reserve), and over 3,500 domestic livestock. Continuing impacts of poaching, agriculture, and livestock grazing (cattle and goats) threaten the sika deer population in the reserve.

534

D. R. McCullough et al.

According to State Wildlife Species Protection Program of 2000 to 2050 by SFA, the Tiebu Nature Reserve will be enlarged in order to protect more sika deer habitat. The function of the Tiebu Nature Reserve will be improved during the period. In most ways, the sika deer in the reserve reflect the ecological and behavioral characteristics of sika deer in other parts of the range as covered by other chapters in this book. Sex and age ratios in 1987 by Sheng (1991) are as expected: 22.7% male, 58.4% female, and 18.9% fawns. The social structure consisted of 15 different aggregations distributed in the valleys and forest edges of two local drainages. Rut extends from September to November, with a peak in October (Sheng 1991). Major foods are grasses, forbs, and palatable shrubs, selected by season to maximize nutritional intake (Sheng 1991; Guo 2001). Sika deer spend 52.1% of their time active, with peaks in the morning and at midnight (Guo 2003), and similar results were reported by Liu et al. (2004). Sika deer visit salt licks and eat salty soil in spring (Guo 2001) just as they do in Far East Russia (Aramilev chapter 33). Liu et al. (2002) found that grazing behavior accounts the most for activity time in spring, ruminating and bedding behaviors accounted for the next, and the least for alerting and moving behaviors. For northeast China sika deer, there are three grazing peaks during a day (Liu et al. 2002). The males and females showed different activity patterns. The females spent more time moving than the males, while the males were more alert and bedded more than the females. Weather is another factor which influences the time budget of sika deer. Sika deer spent more time grazing during cloudy days than during sunny and rainy days, while they were less alert and moved less during cloudy days. According to Liu et al. (2004) the activity pattern of Sichuan sika deer in Tiebu Nature Reserve differs from south China sika deer. Sika deer in Tiebu had two peaks of activity including grazing: the first peak took place at 0830 and the second peak at 1900 hours. Furthermore, they were more active during the night than during daytime. In conclusion, many factors such as sex, breeding conditions, season, sunlight, food, and human disturbance are factors that affect the daily activity rhythm and time budget of Sichuan sika deer (Guo 2003; Liu et al. 2004). From 1987 to 1991 Guo and Zheng (2005) marked 111 newborn fawns and followed their survival and reproduction thereafter to construct a life table. The oldest male was 14 years and the oldest female was 15 years. The sex ratio at birth was 1:1 but among adults it was 1:3 in favor of females, showing the lower survivorship of males. The life table analysis (Guo and Zheng 2005) showed that the population over this time was just more than replacing itself (λ = 1.032; 1.000 equals exact replacement), which suggests that the Tiebu Nature Reserve population did not further decline in the years following 1987. Females reproduced from ages four to 11 years, and many births, particularly by prime aged females (four to nine years) were twins (6.5%). This is a very high rate of twinning, which is relatively uncommon in sika deer. This suggests that the Tiebu population is well below the natural carrying capacity of the habitat, and that reproduction is quite high due to high resource availability. Conclusions from the population dynamics of the population are in agreement with direct measurements on the vegetation. Guo (2002) measured vegetation

35 Sika Deer in Mainland China

535

productivity and calculated a theoretical maximum carrying capacity of around 1,500 deer. He recommended a lower population, however, of 450 to 750 deer. From these two lines of evidence it is apparent that if the mortality rate could be reduced, the population would recover quickly and could be maintained at a higher level, even without improving the habitat.

Southern China Altogether, eight nature reserves have been designated for protection of wild sika deer and their habitats in Zhejiang, Jiangxi, and Anhui Provinces. Of those nature reserves six are national, one provincial, and one a county reserve. These reserves protect over 600 km2 of habitat. Taohongling Nature Reserve was established in 1981, but the area encompassed (12,500 ha) was unclear until the nature reserve settled the management rights of land in 2001. Sheng (1990, 1991) reported the number of sika deer in the reserve as 150 to 200 head. More recent estimates of the number obtained in a sika deer survey and biodiversity investigation, which was conducted by scientists from the Institute of Zoology, Institute of Botany and Southern China Institute of Botany of the Chinese Academy of Sciences, and South-central China University of Forestry Science and Technology from 2004 to 2006, is around 300 (Jiang in press), but the trend seems to be gradually downward because sika deer are dispersing to the adjacent areas out of the nature reserve (Wu and Zhang 2001; Liu 2007). Sika deer in the Taohongling Nature Reserve prefer gentle slopes covered with shrubs, close to water but at a distance from human disturbance (Fu et al. 2006). Xu et al. (1998) stated that more effective protection measures are needed, not only here, but for all of the south China sika deer populations in Jiangxi Province. This recommendation was adopted by the wildlife management authority, and more nature reserves have been established; at least one of them has wild sika deer (Table 35.2). After the establishment of the Taohongling National Nature Reserve cutting of firewood and hunting were banned. Burned areas and frequency of wild fire have been greatly reduced, and the succession of the secondary vegetation has accelerated inside the reserve. Consequently, the heights of plants has increased and shrubs are spreading quickly; the habitat suitability for sika deer has been reduced, particularly in the center of the reserve. Some sika deer have even moved out of the reserve. Clearly, an understanding of habitat use by sika deer is necessary for effective management of Taohongling Reserve for deer. Fu et al. (2006) found that, in Taohongling Nature Reserve, the environmental factors preferred by sika deer are shrub and shrub-meadow, lesser canopy and shrub overstory, high food abundance, shade, mid and upper slopes with less steepness, distances from water sources of 100–400 m, and distance from human disturbance exceeding 800 m. They also suggested that it is necessary to increase community acceptance and carry out habitat improvement.

536

D. R. McCullough et al.

To further understand habitat use and selection by sika deer, Liu (2007) collected data on seasonal habitat use by sika deer based on sampling quadrats placed along line transects in Taohongling Nature Reserve and the Efeng area which lies near the Taohongling Nature Reserve. The Selection Index Method was used to reveal what kind of habitat had been used and selected by sika deer. Then, Principal Component Analysis (PCA) was employed to identify the key factors affecting the habitat use by sika deer. Based on those analyses and the information on diet of sika deer at Taohongling Nature Reserve, Liu also developed a Habitat Suitability Index Model of sika deer and evaluated the habitat suitability of the sika deer with a GIS (geographical information system) model. Liu (2007) found at the Taohongling National Natural Reserve, the area with a Habitat Suitable Index (HSI) over 0.75 was 2,326, 1,014, 1,892, and 829 ha in spring, summer, autumn, and winter, respectively. These suitable habitats amount to 18.57%, 8.09%, 15.11% and 6.55% of total area of the reserve, respectively. Most of those habitats are located at the fringe, or the buffer zone of the reserve, which means that due to the succession of vegetation, the habitat in the center of Taohongling Reserve is no longer suitable for sika deer in some degree. Finally, Liu also found prescribed burning and slash logging had similar effects on vegetation. Both measures increased the species diversity and biomass of forbs and grasses and reduced the heights of shrubs. However, effects of stimulating the growth of herbs and grasses of prescribed burning were more profound than slash logging, while prescribed burning also retarded the growth of woods longer than slash logging. Similar results also were found in another population of the south China sika deer in Qingliangfeng Nature Reserve (Yang et al. 2002; Ma et al. 2004). Liu (2007) found the sika deer in Taohongling Nature Reserve foraged on a wide variety of plant species; 37 plant species had been accumulatively recorded as food items during different seasons. Sika deer were primarily browsers, feeding mainly on twigs, shoots, stems, and leaves of woody plants. In the diets of sika deer, there were 21 species of herbs and 16 species of shrubs. Sika deer at both study sites showed no selection for aspect, but preferred habitat with slopes between 15° and 45°. Water was another factor that affected the habitat selection by sika deer. Sika also selected habitats with the open views, and with less human disturbance. Yang et al. (1990) studied the behavior of south China sika deer in Taohongling Nature Reserve. Behaviors of sika deer were grouped into eight categories: grazing, ruminating, bedding, moving, standing, drinking, alert, agonistic, and other behaviors. Although it is a primary study, this report benefited deer management and habitat improvement in Taohongling Nature Reserve. For example, dams have been built to store water in two reservoirs for deer drinking. Qingliangfeng Mountain is located between Zhejiang and Anhui Provinces, and is inhabited by southern China wild sika deer. Thus, two national nature reserves, Qingliangfeng and Jixi Qingliangfeng, have been established to protect the wild sika deer on Qingliangfeng Mountain. The Qingliangfeng Nature Reserve is 108 km2 in area and is located in the 2,595 km2 forested area of Lin’an City (Yu 2007). Yang et al. (2002) used direct observation and track survey methods to study

35 Sika Deer in Mainland China

537

that sika deer population and found in winter the deer used the gentler south- and west-facing slopes where they fed on a mix of browse and herbaceous vegetation. They noted that the habitat is deteriorating. Furthermore, DNA evidence from skin in a local hunter’s home proved to be from animals poached in the nature reserve (Wu et al. 2005). Based on winter track count surveys Yu et al. (2006) estimated the population at 104 sika deer. According to the State Wildlife Species Protection Program of 2000–2050, the Jisi Qingliangfeng Nature Reserve will be promoted to the national nature reserve level. A nature reserve in the Jiangxi Province will also be promoted to the national level, and a county nature reserve will be elevated to the provincial level. The feral sika deer in Shandong Province attract less conservation attention. However, the feral sika deer in the Weideshan area, Rongcheng City are now under the protection of the Weideshan Nature Reserve. The reserve is a county-level nature reserve. Reintroduction of sika deer on Liugongdao Island is under management of the Liugongdao National Forest Park. Acknowledgements We acknowledge the financial support of the Natural Scientific Foundation of China (No. 30670353, 30430120), and the Knowledge Innovation Project of the Chinese Academy of Sciences (CXTDS2005-4). We sincerely thank all people and organizations who made our work possible, in particular, Dr. Jian Liu, Ms. Hongxia Fang, and Ms. Xiaoge Ping for their various kinds of help.

Literature Cited Dong, Y.-L, A.-Q. Li, X.-Y. Ding, J. Wang., C.-Y. Liu, and X.-K. Xu. 2002. Conservation and applications of life resource in Baotianman Natural Reserve. Henan Science 20:61–65. (In Chinese with English summary.) Fu, Y.-Q., X.-D. Jia, J.-C. Hu, Y.-S. Guo, H.-B. Zhu, W.-H. Liu, and Y.-S. Wang. 2006. Summer habitat selection by sika deer in the Taohongling Nature Reserve, Jiangxi Province. Sichuan Journal of Zoology 25:863–865. (In Chinese with English summary.) Guo, Y.-S. 2000. Distribution, numbers, and habitat of Sichuan sika deer (Cervus nippon sichuanicus). Acta Theriologica Sinica 20:81–87. (In Chinese with English summary.) Guo, Y.-S. 2001. Study on the food habits of Sichuan deer (Cervus nippon sichuanicus). Journal of Sichuan Teachers College (Natural Sciences) 22:112–119. (In Chinese with English summary.) Guo, Y.-S. 2002. Determination of sika deer’s food resources and loading capacity in Tiebu Nature Reserve Sichuan, China. Acta Theorologica Sinica 22:254–263. (In Chinese with English summary.) Guo, Y.-S. 2003. Daily activity rhythm and time budget of Sichuan sika deer. Acta Theriologica Sinica 23:104–108. (In Chinese with English summary.) Guo, Y.-S., and H.-Z. Zheng. 1992. Geographic history of sika deer in China. Journal of Sichuan Teachers College (Natural Sciences) 13:1–9. (In Chinese with English summary.) Guo, Y.-S., and H.-Z. Zheng. 2000. On the geological distribution, taxonomic status of species and evolutionary history of sika deer in China. Acta Theriologica Sinica 20:168–179. (In Chinese with English summary.) Guo, Y.-S., and H.-Z. Zheng. 2005. Life table and the rate of natural increase in Sichuan sika deer. Acta Theriologica Sinica 25:150–155. (In Chinese with English summary.)

538

D. R. McCullough et al.

He, X.-R. 1994. Studies on geographic distribution and disappearance causes of sika deer, rhinoceroses and sarus crane in the history of Yunnan. Journal of Yunnan University (Natural Sciences) 16:294–298. (In Chinese with English summary) Hu, J.-C. 1998. Sika deer. Pages 266–269 in S. Wang, editor, China red data book of endangered animals. Science Press, Beijing, Hong Kong, China, and New York, New York, USA. Jiang, Z.-G., editor. In press. Biodiversity in the Taohongling Sika Deer National Nature Reserve. Tsinghua University Press, Beijing, China. (In Chinese with English summary.) Korean People’s Democratic Republic. 1998. A survey of tigers and prey resources in the Paektusan area, Lyangan Province, North Korea, in winter, 1998. Academy of Sciences, Institute of Geography, in association with the Russian Academy of Science Far East Branch, Institute of Geography. Lei, Q.-B. 2007. Current situation and developing prospect of sika cultivation. Journal of Economic Animal 11:61–62. (In Chinese with English summary.) Li, S.-C., and H.-L. Xue. 2000. Varieties of farmed sika deer in China. Sichuan Animal and Veterinary Sciences 27:39. (In Chinese.) Li, W.-H., and X.-Y. Zhao. 1989. China’s nature reserves. Foreign Language Press, Beijing, China. Liu, H., H.-Y. Shi, and J.-C. Hu. 2004. Daily rhythm and time budget of Sichuan sika deer (Cervus nippon sichuanicus) in spring. Acta Theriologica Sinica 24:282–285. (In Chinese with English summary.) Liu, J. 2007. Food and habitat selection by sika deer (Cervus nippon kopschi) and habitat improvement in the Taohongling Nature Reserve, Jiangxi. PhD Dissertation, Institute of Zoology, Chinese Academy of Sciences, Beijing, China. (In Chinese with English summary.) Liu, Z.-S, J.-P. Wu, and L.-W. Teng. 2002. Time budget and behavior pattern of semi-free Cervus nippon in spring. Chinese Journal of Ecology 21:29–32. (In Chinese with English summary.) Lü, X.-P., F.-W. Wei, M. Li, Y. Guang, and H. Liu. 2006. Genetic diversity among Chinese sika deer (Cervus nippon) populations and relationships between Chinese and Japanese sika deer. Chinese Science Bulletin 51:433–440. Ma, J.-F., E. Zhang, S.-Y. Zhang, and D.-M. Weng. 2004. Preliminary analysis on the habitat use by sika deer in Qingliangfeng National Natural Reserve in autumn. Chinese Journal of Zoology 39:35–39. (In Chinese with English summary.) McCullough, D. R., J. K. Fischer, and J. D. Ballou. 1996. From bottleneck to metapopulation: Recovery of the tule elk in California. Pages 375–403 in D. R. McCullough, editor, Metapopulations and wildlife conservation, Island Press, Washington, DC and Covelo, California, USA. Piao, Z.-J., B.-H. Qiu, X.-L. Zhang, J.-J. He, and Y.-H. Zhang. 1999. Current status and conservation of endangered animals in Changbaishan Nature Reserve. Changbaishan Nature Conservation 58–59:1–12. (In Chinese.) Pulliam, H. R. 1988. Sources, sinks, and population regulation. American Naturalist 132:652–661. Sheng, H.-L. 1990. Status of deer harvest and conservation in China. Pages 165–166 in N. Ohtaishi, K. Kaji, and T. Mano, editors, Proceedings of the deer and bear forum Hokkaido 1990. Wildlife Information Center, Sapporo, Japan. (In Japanese and English.) Sheng, H.-L. 1991. Sika deer. Pages 202–212 in H.-L. Sheng, editor, The deer in China. East China Normal University Press, Shanghai, China. (In Chinese with English summary.) Sheng, H.-L., and N. Ohtaishi. 1993. The status of deer in China. Pages 1–11 in N. Ohtaishi, and H.-L. Sheng, editors, Deer of China: Biology and management. Elsevier Science, Amsterdam, The Netherlands. Song, Y.-L., and Z.-T. Liu. 2005. Rare animal species – Sika deer and its studies. Bulletin of Biology 40:1–3. (In Chinese.) State Forestry Administration of PR China. 2001. National engineering construction plan for deer protection. State Forestry Administration of People’s Republic of China, Beijing, China. Sun, B., D. G. Miquelle, Y. Xiaochen, E. Zhang, S. Hiyai, G. Goshen, D. G. Pikunov, Y. M. Dunishenko, I. G. Nikolaev, and L. Daming. 1999. 1999 survey of Amur tigers and Far Eastern leopards in eastern Heilongjiang Province, China, and recommendations for their conservation. A final report to The Wildlife Conservation Society, Bronx, New York, USA.

35 Sika Deer in Mainland China

539

Warshaw, S. 1994. China emerges: A concise history of China from its origin to the present. Diablo Press, Berkeley and San Francisco, California, USA. Whitehead, G. K. 1993. The encyclopedia of deer. Swan Hill Press, Shrewsbury, United Kingdom. Wilson, R. L. 2000. An investigation into the phylogeography of sika deer (Cervus nippon) using microsatellite markers. M. S. Thesis, University of Edinburgh, Scotland, United Kingdom. Wu, H., Q.-H. Wan, and S.-G. Fang. 2004. Two genetically distinct units of the Chinese sika deer (Cervus nippon): Analyses of mitochondrial DNA variation. Biological Conservation 119:183–190. Wu, H., Q.-H. Wan, S.-G. Fang, and S.-Y. Zhang. 2005. Application of mitochondrial DNA sequence analysis in the forensic identification of Chinese sika deer subspecies. Forensic Science International 148:101–105. Wu, H., J. Hu, S.-G. Fang, L.-L. Kong, and F. Jia. 2006. Genetic diversity and genetic structure of domestic sika deer in China. Journal of Zoology 41:41–47. Wu, H.-L., X.-B. Wu, and G.-B. Gong. 2003. Current status of sika deer resource in Wanjia Town, Ningguo City, Anhui Province. Chinese Journal of Zoology 38:54–57. (In Chinese with English summary.) Wu, P.-J., and E.-D. Zhang. 2001. The resource conservation and utilization of wild sika deer in China. Journal of Chinese Medicinal Materials 24:552–554. (In Chinese with English summary.) Wu, Z. 2000. Tiger conservation plans in Jilin Province. Pages 38–39 in D. Miquelle, E. Zhang, M. Jones, and T. Jin, editors, Workshop to develop a recovery plan for the wild North China tiger population. Heilongjiang Forestry Administration, Harbin, China, and The Wildlife Conservation Society, Bronx, New York, USA. Xu, H., H. Lu, L. Sheng, C.-M. Gu, H.-F. Xu, H.-J. Lu, H.-L. Sheng, and G.-M. Gu. 1998. Status and current distribution of South China sika deer. Biodiversity Science 6:87–91. (In Chinese with English summary.) Yang, J., T.-M. Ding, and P.-X. Hu. 1990. Primary report of ecological studies in south China sika deer. Chinese Wildlife 55:17–19. (In Chinese.) Yang, Y.-W., S.-Y. Zhang, and A.-X. Cheng. 2002. Characteristics of habitats used by sika deer in winter and spring in South China. Journal of Northeast Forestry University 30:57–60. (In Chinese with English summary.) Yu, J.-A., Q. Lu, Q. Zhou, and S. Zhang. 2006. Study on the population number and distribution of Cervus nippon kopschi in Qingliangfeng Nature Reserve. Journal of Zhejiang Forestry Science and Technology 26:1–4. (In Chinese with English summary.) Yu, L. 2007. Lin’an County: A model for the future? Pages 10–19 in In search of excellence: Exemplary forest management in Asia and the Pacific. http://www.fao.ogr/docrep//007/ ae542e/ae54208.htm. Zhao, D.-S. 1990. Deer thremmatology. Chinese Forestry Press, Beijing, China. (In Chinese.)

Chapter 36

Sika Deer in Korea and Vietnam Dale R. McCullough

Abstract Sika deer have suffered greatly on the Asian mainland from the close association of their primary habitat to prime agricultural lands and from loss through direct killing and alteration of habitats by humans. Sika deer populations are extinct in the wild in Vietnam and South Korea. North Korea still has wild populations of sika deer in the northern part of the country along the Chinese and Russian borders. There are several thousand in captivity in Vietnam, but no native stocks are present in South Korea. Small numbers cross over the borders between North Korea and Far East Russia, and northeastern China. The northern and central populations in China and North Korea are genetically similar and related to the southern clade in Japan. Human population pressures continue to impact the remnants of sika deer in the wild, and loss of genetic diversity is likely given the fragmentation of the remaining deer into isolated, small subpopulations.

Introduction Sika deer originated on the mainland of southeastern Asia and subsequently spread to Taiwan and the Japanese Archipelago. Sika deer still thrive on the mainland in Russia (Aramilev chapter 33; Voloshina and Myslenkov chapter 34) but to the south, in what almost certainly was the heart of their range over most of the time since the Pleistocene, they are either extinct in the wild or reduced to small and insecure remnants. China has been covered in another chapter (McCullough et al. chapter 35), so in this chapter I review the other parts of the original distribution of sika on mainland Asia, Korea and Vietnam. Unfortunately, these countries have done little in the way of research on sika deer, and there is generally a paucity of information and few publications. I have used the historical record and existing literature to the extent it is available. Where specific results are unavailable some expectations reasonably can be inferred from what is known about general patterns of sika deer in adjacent areas with similar environmental and habitat characteristics. The southeast Asian civilizations extend far back into history, and the landscapes they occupy have been modified for human benefit for millennia. Because D. R. McCullough et al. (eds.), Sika Deer: Biology and Management of Native and Introduced Populations, DOI: 10.1007/978-4-431-09429-6_36, © Springer 2009

541

542

D. R. McCullough

sika deer and humans have similar habitat requirements, these modifications have, for the most part, been detrimental for sika deer. Low, more level lands are most suitable for agricultural fields, and high rainfall amounts favor marshy and swampy vegetation types. These variables are the factors favoring rice culture in the wetter southern parts of southeast Asia, and wheat and other upland grain crops in the drier northern parts of the range. These staple foods fostered the growth of human populations to high densities, which further encouraged more intensive agriculture. Sika deer either were hunted out in the early stages of the conversion process or eliminated when fields became too extensive and escape cover disappeared from the landscape. It is ironic that so many of the activities of humans—creation of high quality foods and elimination of large predators—favored sika deer, but at some stage of development the human population presence became overwhelming. As one moved from the prime, lowland landscapes that were the favored habitat of the sika deer, to the upland areas, natural forest vegetation became more common. These landscapes were less amenable to agriculture and were inherently less productive. Still human population pressures, in conjunction with cultural and economic forces, induced people to occupy these areas as well. Although ecologically the modification of forests to create agricultural fields benefits sika deer if the scale is small and patches of disturbance are interspersed with undisturbed habitat, the long history of human population pressure in Asia has left little unaltered habitat. The combination of conversion of habitats to a severe degree with continued direct hunting on the confined, remaining natural refugia on the fringes of the range has resulted in the current poor status of sika deer in Korea and Vietnam. Although in a few isolated patches in Korea near the Russian border sika deer disperse across the border, hunting of such animals is continuous and uncontrolled. These isolated populations often wink on and off as new dispersers arrive, establish a small colony, and are subsequently extinguished by hunting.

Original Distribution Korea Sika deer in Korea were of the northeastern subspecies Cervus nippon hortulorum, in common with those of northeastern China and adjacent Far East Russia. Although little specific data are available about the original distribution of sika deer on the Korean peninsula, they were known to extend across most of the country (Won and Smith 1999). The specific distribution can be inferred from their distribution in adjacent areas in Japan, China, Taiwan, and Far East Russia. Most of the Korean peninsula is mountainous, and sika deer would have been found mainly along the coastal areas and extended inland along river courses and into the higher elevations where hardwood and mixed hardwood-conifer forests occurred up to a limit set by winter snow depth (Fig. 36.1). No doubt sika deer in the higher mountains were seasonally migratory, as they are in northern Japan (Yabe and Takatsuki

36 Sika Deer in Korea and Vietnam

543

Fig. 36.1 Map of the Korean Peninsula showing the high mountain (dashed line) area where sika deer originally were probably scarce, and Lyangan Province (stippled area) where sika deer are known to still exist in the wild.

chapter 20) and that there is some suggestion of in Far East Russia (Voloshina and Myslenkov chapter 34). This would allow them to occupy all but the highest mountains which are found in the north, near the border with China. The mountain chain extends southward mainly on the eastern side of the peninsula. Sika deer no doubt reached their highest original density on the west coast facing the Yellow Sea where more level topography and mild climate favored them. Once again, the distribution of sika deer and human agriculture was highly correlated.

Vietnam Sika deer in Vietnam are commonly recognized as a separate subspecies, C. n. pseudaxis. Microsatallite DNA studies show that they are a distinct clade, separate from other sika populations, and a valid subspecies (Wilson 2000). Her results suggest that a common ancestral population arose in the northeastern part of the mainland distribution and then spread southward along the south coastal area of China, to Vietnam, and eastward to Taiwan during a land bridge period (McCullough

544

D. R. McCullough

chapter 37). Both the Vietnam and Taiwan populations, probably through isolation, differentiated to a greater (Vietnam) or lesser (Taiwan) degree. Taiwan sika deer group more closely with northeastern populations, but 10,000 years of isolation caused by the closing of the land bridge has resulted in the formation of a separate clade that warrants a subspecific designation, C. n. taiouanus. Along the coast from China southward to Vietnam sika deer distribution would have been sporadic and scattered in patches of good habitat. Historically, sika were known to occur near the Chinese border in Quang Ninh Province (near the towns of Cai Bau, Binh Khe, Hoanh Bo, and Dinh Lap) (Huynh et al. 1992). However, it is in the Red River Valley (Hanoi and adjacent landscape) where sika deer were prevalent (Huynh et al. 1990, 1992, 1994) (Fig. 36.2). Note that the commonly cited map in Whitehead (1972, his Map 12) showing the distribution of the subspecies of sika deer is incorrect in placing Vietnamese sika deer in the central, rather than the northern part of the country. Sika deer would have been quite abundant originally in this rich agricultural area. They would have extended westward along river valleys into the surrounding mountains on the borders with China and Laos, with the upper limits determined more by habitat suitability rather than winter snow depth, given the mild conditions of this southerly latitude.

Fig. 36.2 Map of northern Vietnam showing the probable original distribution (stippled area) of sika deer.

36 Sika Deer in Korea and Vietnam

545

From the Red River Valley the sika distribution extended southward on the east side of the Ammonite Mountains approximately to North Parallel 16°20′ near the present city of Hue (Huynh et al. 1992), where the mountains meet the sea. Although narrow coastal strips occur to the south, the mountains remain close to the seacoast southward from here until the Mekong Delta. Sika deer were not known from historical records or fossils to have occurred further south in Vietnam in the Hoh Chi Minh City (Saigon) area, or to the west in Laos and Cambodia. It is probable that during the Pleistocene glacial cycles, sika deer extended southward as sea levels fell and the continental shelf was exposed. What factors set a limit to the southern distribution during the Holocene is not clear. However, tropical climates being too hot, and competition with a plethora of other large ungulates, including other species of deer better adapted to tropical forests, seem likely candidates.

Current Status In Korea Sika deer are extinct in the wild in South Korea (Won and Smith 1999), and the zoo stocks were imported from elsewhere (C. Won personal communication 1998). Sika still exist in the wild in North Korea, but little information is available on the current distribution. They are found mainly in the northern part of the country, near the Russian border, and it is likely that some cross-border movement occurs between both Russia and China. Ironically, but probably not accidentally, sika deer still persist in the higher mountains where originally they would have been least common. The only field data available are from Lyangan Province (Fig. 36.1) which contains the Paektu Mountain Range adjacent to the Chinese border. Field surveys for tigers and their prey reported the presence of sika deer (Democratic People’s Republic of Korea 1998). Based on track surveys, sika deer were found in all five administrative districts at low densities. Highest density, approximately 0.5–0.7 deer/km2, was found in the higher elevations near the Chinese border in Samdzhien District. Despite elevations often above 1,500 m this is an area with broad-leaved forests, intermixing with conifers at higher elevations, much like the sika deer habitat in Far East Russia, or those in the Sichuan basin in China. According to the World Commission on Protected Areas web site, the Samjiyon Deer National Monument, a 3,000 ha protected area, was established in 1980. Deer elsewhere in Lyangan Province were less abundant, presumably because of greater human presence in the lower elevations. Deer are said to cross the border with China during migrations (Democratic People’s Republic of Korea 1998) and probably move to lower elevations to the east as well during winter snows. Tigers still occur sporadically in Lyangan Province and apparently follow sika deer and other ungulates in these movements. Given that sika deer occur this far south in North Korea, they probably are thinly distributed through the low mountains northeastward to the Russian border. Presumably

546

D. R. McCullough

they occur sporadically further south from Lyangan Province, but information is not available to further define the limits of the range.

Current Status in Vietnam Sika deer are extinct in the wild in Vietnam. I conducted an extensive survey of sika deer in Vietnam in 1998, and all informants said sika disappeared from the wild “… long ago.” According to Thévenon et al. (2004) the major decline occurred during the early 1900s. It is reasonable to presume that sika disappeared earliest from the heart of the Red River Valley around Hanoi, Hai Phong, Nam Dinh and other cities founded on the rich agricultural plains where humans flourished. They survived longest in the mountains on the fringes of the range where humans had the least impact. Huynh et al. (1992) noted that as late as the 1960s, sika deer were still seen commonly in the lower foothills of Ba Vi Mountain and along the river at Bat Bat in Ha Tay Province, and at Luong Son Mountain in Hoa Binh Province. These locations are only about 40 km from the capital city of Hanoi. Huynh et al. (1990; as cited by Thévenon et al. 2004) reported that the last wild sika deer were located in the southern part of the distribution in the Nghe Tinh Mountains in Nghe An Province. It is in this general region that the largest numbers of farmed deer exist today. Sika deer are currently found mainly in deer farms, with smaller numbers having been relocated to enclosures in parks, reserves, and zoos. Whereas there is virtually no information on the original deer in the wild, some work has been done on captive deer to help deer farmers (Huynh et al. 1992) with practical matters of husbandry. No one knows how many farmed deer there are, and estimates range from 3,000 to 15,000, with the latter referring to earlier times. In the 1980s sika deer were very valuable for their antlers, and a mature female was worth about US $5,000. With the collapse of the antler market by 2000, females were worth about US $200, and the number of deer held in farms declined substantially. Estimates by the most informed people at the present time were from 3,000 to 5,000. Most farmed deer are in Nghe An and Ha Tinh Provinces (near the city of Vinh; Fig. 36.2), with the largest number, several hundred, held in a large enclosure at a collective farm near Pho Chau in Ha Tinh Province. Other areas of local deer abundance in family-held farms are near the towns of Do Luang and Quyah Luu in Nghe An Province. Other farm deer probably occur on scattered private farms in the Red River delta, the major agricultural area of the northern part of Vietnam, and the heart of the original sika deer range. The largest sika deer population in national parks is in Cuc Phong National Park in Ninh Binh Province (about 80 km south of Hanoi) where about 200 deer are held in three large pens. Small groups (five to 10) are held in small enclosures at Ba Vi National Park in western Ha Tay Province (about 60 km west of Hanoi), and Cat Ba National Park on Cat Ba Island in Hailong Bay, east of the coastal city of Haiphong. The Hanoi Zoo maintains a herd of sika deer, but according to my informants, they have been interbred with sika deer imported from Vladivostok, Russia. They are,

36 Sika Deer in Korea and Vietnam

547

therefore, genetic hybrids between C. n. pseudaxis and C. n. hortulorum, and not pure Vietnam deer. Although there is no shortage of stock for reintroducing sika deer to the wild, and Cuc Phong National Park represents an appropriate place for such a release, the current situation makes such a program futile. Because of the poverty level, especially in the more remote rural areas, in combination with the wide availability of firearms left over from the Vietnam-American war, poaching for subsistence and trade is prevalent. It is beyond the capabilities of the central authorities to control this activity. Consequently, any sika deer released to the wild are likely to be poached out in a short time. Hopefully with the resurgence of the Vietnam economy, a release program will become possible sometime in the future, and the sika deer can once again become a part of the wild native fauna. The remaining genetic diversity in captive populations is low, but not severely so, suggesting a bottleneck (Thévenon et al. 2003, 2004). Thévenon et al. (2004) suggest that the current captive stocks arose from less than 200 sika deer in Ha Tinh Province brought into captivity. In the last half of the twentieth century many of the farmed deer were kept in large collective farms where interbreeding between relatively large numbers of animals was common. However, in my visit to these farms in 1998 showed that collective farm herds were being broken up and distributed into family-managed small groups of deer, usually fewer than 10 individuals, and often only one to three. Only one collective farm with around 100 sika deer existed in 1998. With this fractionation of the breeding stock in isolated small pens, retention of genetic diversity will require frequent trading of animals between farmers, although Vietnam sika deer zoo stocks around the world retain similar diversity to the farmed deer (Thévenon et al. 2004). The immediate prospects of sika deer on the Asian mainland outside of Russia are dim. High human populations and severe poverty render recovery programs for wild deer infeasible. Hopefully over time, economic growth will reach a point in these rural areas where it is possible to restore and protect wild populations of sika deer. In the meantime natural reserves, hopefully, and deer farms almost certainly, will maintain stocks of sika deer as sources of re-establishment in the wild if, and when, that becomes feasible.

Literature Cited Democratic People’s Republic of Korea. 1998. A survey of tigers and prey resources in the Paektusan area, Lyangan Province, North Korea, in winter, 1998. Academy of Sciences, Institute of Geography, in association with the Russian Academy of Science Far East Branch, Institute of Geography. Huynh, D. H., T. V. Duc, and H. M. Khien. 1990. The status of endangered species of deer in Vietnam. National Centre for Scientific Research, Hanoi, Vietnam. (In Vietnamese.) Huynh, D. H., D. N. Can, T. V. Duc, and P. T. Trong. 1992. Raising sika deer in Vietnam. Nghe An Publishing House, Vinh, Vietnam. (In Vietnamese.) Huynh, D. H., D. V. Tien, C. V. Sung, P. Trong, and H. M. Khien. 1994. Checklist of mammals in Vietnam. Science and Technics Publishing House, Hanoi, Vietnam. (In Vietnamese.)

548

D. R. McCullough

Thévenon, S., A. Bonnet, F. Claro, and J.-C. Maillard. 2003. Genetic diversity analysis of captive populations: The Vietnamese sika deer (Cervus nippon pseudaxis) in zoological parks. Zoo Biology 22:465–475. Thévenon, S., L. T. Thuy, L. V. Ly, F. Maudet, A Bonnet, P. Jarne, and J.-C. Maillard. 2004. Microsatellite analysis of genetic diversity of the Vietnamese sika deer (Cervus nippon pseudaxis). Journal of Heredity 95:11–18. Whitehead, G. K. 1972. Deer of the world. Viking Press, New York, New York, USA. Wilson, R. L. 2000. An investigation into the phylogeography of sika deer (Cervus nippon) using microsatellite markers. M. S. Thesis, University of Edinburgh, Scotland, United Kingdom. Won, C., and K. G. Smith. 1999. History and current status of mammals of the Korean Peninsula. Mammal Review 29:3–33.

Chapter 37

Sika Deer in Taiwan Dale R. McCullough

Abstract The Taiwan sika deer (Cervus nippon taiouanus) migrated to the island during land-bridge periods (40,000 to 10,000 years before present), presumably from an adjacent mainland China stock that later became extinct. On Taiwan they were found mainly at low elevation habitats around the island and reached greatest abundance on the large western coastal plain. They were heavily exploited during the European colonial period, serving as currency and export items for international trade and, consequently, disappeared early from all but the more remote parts of their historic range. At the same time, their habitat was usurped for agricultural development. Exploitation for subsistence and velvet antler for the Chinese medicine market continued on the remote remnant pockets of survivors until the last known sika deer in the wild was killed in 1969. Fortunately there were many sika in private ownership, including stocks at the Taipei Zoo and on Green Island off the east coast of Taiwan. Establishment of a captive breeding facility at Kenting National Park at the southern tip of the island, stocked mainly with animals from the Taipei Zoo, resulted in increased numbers, and the eventual release of deer to the wild where they now number around 400 head (Pei chapter 38). A separate release of sika deer from captivity on Green Island has resulted in a second free-roaming population of several hundred animals. Although not now seriously threatened with extinction, to achieve full recovery it is desirable to establish a third population in the central or northern part of the island further removed from the two existing wild populations.

Introduction Geologically, Taiwan was formed by mountain uplift created by the collision between the Asian mainland tectonic plate and the subduction of the westwarddrifting Philippine plate. Modern DNA studies show that sika on the island of Taiwan were derived from populations on the adjacent mainland, from stocks of sika that have been long extinct in southern coastal China (Wilson 2000).

D. R. McCullough et al. (eds.), Sika Deer: Biology and Management of Native and Introduced Populations, DOI: 10.1007/978-4-431-09429-6_37, © Springer 2009

549

550

D. R. McCullough

Sika reached Taiwan during land bridge periods when sea levels were low and the island was connected broadly to mainland Asia. The continent at those times was substantially larger than at present, extending to the eastern margins of Japan, Taiwan, and the Philippine Islands (Ho 1986). However, a deep-water channel separated Taiwan from the Philippine Islands, and there was little fauna exchange beyond fruit bats and perhaps some small rodents that reached Taiwan by rafting (Kano 1940). The Sea of Japan and the South China Sea were large inland seas at the time. The mainland Asian fauna spread out geographically across the low plain connecting these “islands,” which were mountains on the edge of the continental shelf. Many fossils of terrestrial mammals have been found on the floor of the Taiwan Straits (Hu and Tao 1993; Ho et al. 1997) showing that the fauna was continuous from the mainland across to Taiwan. As a lowland species, the sika were undoubtedly favored by the extensive lowlands, marshes, and plains of the land bridge areas. Pollen records show that grasslands were extensive on the more level lowlands, and semi-open forests occupied the uplands (Liew et al. 1998). These are the kind of mixed habitats with extensive edges and ecotones that are favored by sika deer. Probably the exposed Taiwan Straits at the last glacial maximum looked much like the recent western plains of Taiwan, where sika deer thrived when Europeans first arrived in the fourteenth century. Sika deer fossils that are around 40,000 years old found on Taiwan show that sika also reached Taiwan during an earlier land bridge period. The current population reached Taiwan during the last glacial period, approximately 10,000 to 12,000 years ago. Again, DNA studies by Wilson (2000) showed that all current stocks of sika are essentially the same (although a few unique alleles are found in only one or another isolated population, as will be discussed below), thus making it highly unlikely that some isolated pockets (e.g., in the eastern Hau-Tung Valley isolated from the western plain by the great central mountain range) of sika might have survived from the earlier invasion. The sika in Taiwan were derived from the southern major clade of sika (Wilson 2000; Nagata chapter 3; Tamate chapter 4) as are the Vietnam sika, and presumably the stocks that originally occupied the coastal plain of mainland China before their extinction there. Sika deer of the southern clade are semitropically adapted animals found mainly at lower elevations.

Distribution of Sika in Taiwan In Taiwan, the major original range of sika deer was the extensive lowland plain on the western side of the island, where they reached high abundance (Fig. 37.1). However, they also extended along lower valleys into the mountains. Zoogeographic studies by Kano (1940) showed that in the eastern plains sika occupied elevations up to about 300 m. From historical records we know that sika occupied the lower landscapes all the way to the south tip of the island, and the major Hau-Tung Valley (shaped like a rift valley, but formed by the Central Mountain Range to the west and

37 Sika Deer in Taiwan

551

Fig. 37.1 Former distribution of sika deer in Taiwan (stippled areas, after Su 1985), current distribution of populations (circles), and proposed introduction areas (squares).

the uplifting of the east coastal valley by subduction of the Pacific plate) on the western side of Taiwan between the high Central Mountains and the much lower western Coast Range. Sika occupied most of the smaller valleys stringing off in both directions from the Hau-Tung Valley and probably occupied much of the eastern Coast Range, which is mainly low in elevation. The original genetic stock of sika which reached Taiwan by spreading across the exposed plains from China was adapted to lowlands, and clearly physiological and behavioral factors played a role in the sika in Taiwan remaining a lowland species. Nevertheless, 12,000 years is a relatively long time for further adaptive modification not to have occurred so what prevented sika from extending their range into higher elevations? Three factors probably played a role. First is the very steep and unstable character of the mountains. Due to recent rapid uplifting of the high Central Mountain Range (with more than 200 peaks over 3,000 m) most of the central and western part of the island is particularly rugged, rising to a high point at Yu Shan (Jade Mountain) at 3,952 m.

552

D. R. McCullough

Taiwan continues to be one of the most rapidly uplifting areas of the world. Between 1907 and 1964 there were 15,088 earthquakes felt (an average of 269 per year), most frequently on the eastern side of the island (Lee 1981). There are some volcanic, basalt parent materials in small areas near the northern tip of the island. However, most of the mountains are composed of former seabed slate and shale substrates which, under force of earthquakes and gravity, quickly shatter into scree, thus creating unstable talus slopes that rest at the angle of repose. Landslides are ubiquitous in Taiwan (Lee 1981). Besides earthquakes shattering the substrates, severe typhoons sweep up from the South Pacific and hit the island, three per year on average, often dumping huge amounts of rainfall; supersaturated soils contribute further to the instability of the mountains. Second, the mountains are cloaked with thick, dense, virtually impenetrable forests in the lower and middle elevations, forming openings mainly on the alpine peaks where lightning strikes create gaps. These sorts of closed habitats do not favor the sika deer, which is predominantly an edge species, relying on openings for feeding at night, and forests for concealment cover during the day. The third factor is that the less severe slopes in the mountains were occupied by native sambar deer (Cervus unicolor), a large deer that shows extreme flexibility in physiological tolerances from north to south in Taiwan. In the northern and central parts of the island sambar occupied open, low-growing bamboo grasslands at the highest elevations, where they would have had to show elevational migration to deal with winter snows. In the mountains towards the south end of the island sambar thrive in thick, tangled, subtropical vegetation. It is notable that in the zoogeographic studies of Kano (1940) the upper distribution of sika deer at 300 m was the lower limit of sambar deer. It is probable that competition with sambar on the gentler slopes set an upper limit to the distribution of sika, just as a similar limit holds for wapiti and sika in Far East Russia (Aramilev chapter 33) and between introduced red deer and sika in New Zealand (Banwell chapter 42). Muntjac (Muntiacus reevesi) and wild boar (Sus scrofa) cohabited with the sika in the lower elevations and probably added to the competitive pressure in these areas as well. Steep ridges and rocky slopes are the habitat for the native serow (Capricornis crispus) and are not very promising habitat for a lowland-adapted sika deer to invade. An approximate map of the original distribution of sika is given in Fig. 37.1. Perhaps the range extended a bit beyond the limits shown, but not by much. In the major exception, further intrusion into the mountains would have been along stream courses, particularly the larger ones where a river flood plain would have contained suitable habitat. Furthermore, because sika deer are mixed feeders, taking grasses, forbs, and browse (Hsia 1990), they thrive in disturbed habitats; therefore, the activities of humans in altering vegetation to create crop fields would have favored sika and may have allowed intrusion further into mountains where, otherwise, they would not have done well in natural vegetation (Severinghaus and McCullough 1996). Indeed, over their continental range, sika populations and human populations prospered equally in the same areas—at least up until the last 150 years or so when guns and other technology gave humans the capability to overexploit the deer (McCullough chapter 1).

37 Sika Deer in Taiwan

553

Similarly, natural and anthropogenic fires would have favored sika, and anthropological studies suggest that aboriginals used fire to favor more open habitats that supported game populations, and the sika would have been a prime target of such activities. But even without humans, the frequent summer lightning storms, particularly in the southwestern part of the island, would have favored sika on the plains and lower mountains (Su 1985). This area has a dry summer, and droughts occur here on average every two years (Severinghaus 1989). Thus, it is not surprising that smoke over the Formosan Straits from fires on Taiwan was reported by a Chinese ship captain in the year 605 AD. He noted that such smoke was commonly seen every spring and fall (Hirth and Rockhill 1966).

Sika and Humans As noted earlier, sika deer have thrived simultaneously with humans over most of their distributional range, and over most of the history of the species (McCullough chapter 1). Probably no other wild ungulate has been so intimately tied to the presence and activities of humans for such a long time—tens of thousands of years at minimum. Prior to the arrival of Europeans, sika deer were widespread and abundant on the island of Taiwan. The origin of aboriginal people on Taiwan is in question. It is logical that humans crossed the same land bridge on foot just as much of the fauna did in the last glacial maximum. There is some evidence that humans may have lived on Taiwan for perhaps 6,000 to 7,000 years, but the results are equivocal. However, it is much clearer that the aboriginal people present on Taiwan when the earliest Chinese and Europeans arrived reached there by sea and were derived from Malaysian/Polynesian ancestry. These people have been on Taiwan for perhaps 3,000 to 4,000 years. Sika deer figured prominently in aboriginal art and culture (Chen 1968).Other species were hunted, but deer were probably most important. Sika were taken for meat and tools, but this had little impact on the overall population. The number of aboriginals was low. The census of 1910 showed there were 122,100 aboriginals (Chen 1968), and giving allowance for their higher numbers on the western plains prior to be driven off by early Chinese settlers, their population in prehistoric times still was probably less than 200,000. Surprisingly, the Chinese empire on the Asian mainland had little contact with Taiwan before the seventeenth century (Hirth and Rockhill 1966; Kuo 1973). They knew of the existence of the island from early times, but made no effort to incorporate it into the Middle Kingdom (Roy 2003). Although there were regular contacts with the island from about 230 A.D. onward there was little move to establish a permanent presence there. This was probably due to both a perception that Taiwan held no treasures of particular note, and the fact that the waters of the straits between Taiwan and mainland were rife with pirates (particularly the Penghu Islands), which made travel there dangerous (Davidson 1903). Thus, for example when a

554

D. R. McCullough

Chinese explorer named Cheng Ho became shipwrecked on Taiwan in 1430, he was forced to spend some time there; when he finally managed to return to the mainland he reported that there were only aboriginal people on the island, and no Chinese settlements (Roy 2003). Chinese from Fukian Province on the adjacent mainland began to colonize the western plains of Taiwan in the 1600s, at about the same time that European traders— soon to be colonists—came on the scene (Mackay 1895; Davidson 1903). Mainland Chinese were driven to move to Taiwan because of population pressures on the mainland, and the generally small amount of suitable crop lands in the broken topography of southeastern China. As the Chinese population on Taiwan built up, some Taiwanese intermarried with the Han Chinese early on and gave rise to the mixed ethnicity people currently recognized as Ping-pu (Chen 1968). However, cultural conflicts with the indigenous peoples increased and competition for land intensified; through warfare most of the aboriginal tribes were either eliminated or displaced from the plains to the mountains. When Europeans arrived, strife occurred where the mountains met the plains (including head-hunting of Chinese by aboriginal warriors, a practice common between rival tribes), so few people of one ethnic group dared to encroach on the other’s territory (Mackay 1895; Davidson 1903). Despite their population buildup, the Chinese must not have had too great an impact on sika populations, for historical records speak of the very high abundance of sika on the western plains when European first arrived. The Portuguese were the first Europeans to encounter the island in 1590, and they named it Ilha Formosa (Beautiful Island). Later the Chinese authorities forced the Dutch from the Penghu (Pescadores) Islands and invited them to move to Formosa instead. Consequently, the Dutch established settlements in 1624 and took total possession of Taiwan in 1642 when they drove the Spanish from a small settlement in Keelung at the north end of the island (Roy 2003). Sika deer were sufficiently common that their hides became the first major export commodity, serving as currency. In the absence of gold or silver coins, taxes levied on the local Chinese and aboriginal peoples were paid in deer hides, and these in turn were exported by the Dutch East India Company in trade to Japan and other countries (Chiang 1987; Patel and Lin 1989). Around 40,000 sika deer skins were exported per year during the Dutch period (Hsieh 1964) with a peak of over 60,000 in 1650 (Chiang 1987).

Extinction of Sika Deer in the Wild This heavy exploitation in the mid-1600s probably resulted in the extirpation of sika from most of the western lowlands in the following decades. There is an absence of historical records to follow these events, but when the Japanese obtained the island after defeating the Chinese in the Sino-Japanese War in 1895, the only reports of sika are from small pockets in the interior of the island. Kano’s (1940)

37 Sika Deer in Taiwan

555

studies from 1925 to 1933 show that sika were still present at that time but were rare. Thereafter, once again, the sika probably paid a price during World War II when the Japanese exploited all possible resources in pursuing the war. Scientific study of wildlife stopped, and even Kano himself became a victim of the war (Kuroda 1952). Probably by the end of World War II, sika deer were largely confined to the smaller valleys in the eastern part of the island in areas removed from larger cities. In 1966 Ruhle (1966) reported that few, if any, sika still existed in the wild, and in 1969 Eu (1969) listed the sika deer as endangered in Taiwan. In the same year Wayre (1969) reported they were on the verge of extinction. In the summer of 1973 I went to Taiwan to conduct the first faunal survey since the Japanese work in the 1930s, concentrating on large mammals (McCullough 1974). All parts of the island were driven by car or hiked on foot to ascertain the status of the larger mammals, particularly the sika deer and clouded leopard (Neofelis nebulosa), which were known to be under serious threat of extinction. Valleys in the eastern part of the island near Taitung were scoured for the existence of any remaining sika deer, but without success. Interviews with local aboriginal villagers, particularly hunters, showed that they were very familiar with sika and that in many smaller valleys they had still been abundant up through the 1950s. Aboriginal hunters saved lower jawbones as trophies and for spiritual reasons, and under the eaves of their homes would be strings of jaws of sika and other species of game, but nearly all of the sika jaws were covered with smoke and suet, showing that they were from long ago. Only a few more recent specimens were represented. The most recently killed animal was reported in the small village of Tien-kuang Li as taken in 1969, and the villagers and hunters there were convinced they were all gone. My field searches in the local area where the last animals in the wild were known to have occurred failed to turn up any sign of sika tracks or droppings. At the same time, however, many sika deer were held in captivity by Chinese to produce antlers for the Chinese medicine market.

Recovery Program In the final report on my faunal survey (McCullough 1974) one of my recommendations was for the Taiwan government to establish a captive breeding program for sika deer with the goal of reestablishing the species in the wild. I proposed that the newly established Kenting National Park at the southern tip of the island was a favorable location for such a facility. Fortunately, two captive stocks thought to be genetically pure Taiwanese sika were still in existence. One was in a small herd in the Taipei Zoo, and the other was on Green Island, a small island in the Pacific Ocean 33 km off the east coast of Taiwan. Deer were not native to the oceanic Green Island, a volcanic outcrop on the Philippine plate that had drifted closer to Taiwan over eons, but was never connected by a land bridge. Somewhere around

556

D. R. McCullough

1900 to 1910 local people in the Taitung area captured deer from the wild and, in two separate groups, introduced the captive animals to breeding pens on Green Island to raise them for the antler trade (McCullough 1974; McCullough and Severinghaus 1998). The deer were owned by the local Hsiang (a governmental organization more-or-less equivalent to a U.S. county), and 197 deer were released to the wild in 1986 after the price for velvet antler declined substantially, and the farming enterprise was no longer economically viable. Sika deer have been raised in captivity in Chinese societies for over a thousand years, primarily for the antler velvet as medicine. Because they have been traded, bought, and sold as a commodity for so long, with virtually no records kept, the genetic provenance of stocks of sika are always suspect. The sphere of influence that Taiwan found itself in changed repeatedly over history. First it was China in the sixteenth century, then European colonizing powers driven by trade in the seventeenth century, then back to China once again by the eighteenth, then to Japan in 1895, then back to China once more in 1945, and finally independence since the Communist revolution on the mainland in 1949. In Taiwan’s checkered history it is easily imagined that sika deer might easily have been shipped from China or Japan to Taiwan, and either released purposefully or escaped from captivity. It was desired to base the captive breeding effort on native Taiwan stock, partly for esthetic reasons and partly for retention of local adaptation, which could be presumed to exist because sika had been isolated from other mainland stocks for at least 10,000 years. In 1974 there was no way to ascertain the status of these two captive stocks. It was only later, prompted by this question, that in 1998 I organized an international group of scientists to explore that genetics of all sika deer throughout their native range. This effort lead to Wilson’s (2000) work that showed the two Taiwan stocks to be virtually the same genetically, and original to Taiwan. However she found that some unique alleles were found in each population but not the other. In the absence of this information in 1973, I suggested using the Taipei Zoo stock to establish the captive breeding population. Although zoo records leave much to be desired, and there is no way keepers could have known if introgression of mainland or Japanese sika had occurred (their original source was taken from captive animals held for the antler production), at least their intention to display native sika deer was clear. In contrast, the Green Island stock was not backed by any records, none of the people involved were still alive (indeed, most informants were third generation), and the intent of the people who brought the deer to the island was solely commercial production of velvet antler without any preservation motive. In 1973 Taiwan was a very poor country, and most people were hard pressed to make a living. All available land was being used to raise crops commercially and for subsistence. People were forced by necessity to attempt terracing the steep mountain slopes for farming. Consequently, no area in the western lowlands was available to create a captive breeding facility for sika deer. Eventually this breeding facility was established at the Chiting area in the southwestern part of Kenting National Park in 1984 (Kenting National Park 1984; Wang 1991).

37 Sika Deer in Taiwan

557

The population at the captive breeding facility increased, and eventually deer were released to the wild in the surrounding area (Pei chapter 38). Ironically, tuberculosis broke out in sika deer at the Taipei Zoo (as well as in a number of farmed deer), and that population had to be sacrificed (Severhinghaus and McCullough 1996; McCullough and Severinghaus 1998). The zoo was subsequently restocked with deer from the Kenting population.

Current Status on Green Island Green Island is 16 km2 in area, relatively rough in topography despite not being very high in elevation, and covered by dense forests except where cleared by humans. Soils are poor, and agriculture is limited to small areas, mainly scattered along coastal areas where most people live. There are only about 2,700 people on Green Island, and traditionally they made their living mainly from fishing, which constituted about 70–85% of the economy. Due to declines in the fishing catch in recent years, other sources of income are being pursued, principally recreation. Sika deer are found at low density over most of the island. Lui (1992) did systematic surveys for tracks and droppings and found them virtually everywhere outside of the villages, and our briefer field work in 1996 and 1998 gave the same result. Night spotlight surveys along the main perimeter road in this later work gave counts between 20 to 40 per night. The Hsiang estimates the number of deer as between 200 and 250, and this seems reasonable. Deer density is highest in the northeastern part of the island where domestic water buffalo pastures have been cleared, and the continuing buffalo grazing maintains the grassland. These low, lawn-like grasses are prime sika habitat, and sika are attracted to them where they concentrate their feeding at night. The deer are not habituated and are easily disturbed by human presence. But our spotlight work showed that they quickly returned to the pastures once the source of disturbance leaves. Some tourists from the local hotels try to observe the deer in this area, and there is potential to increase ecotourism using the attractiveness of the deer. It is notable that grazing by water buffalo is responsible for maintaining these pastures, and their removal would likely work to the detriment of the sika deer. It is unlikely that a sufficient population of sika could be tolerated on the island to maintain the pastures themselves because of problems with agricultural damage, for which by Taiwanese law the Hsiang is responsible. The other areas where sika congregate are adjacent to the small agricultural areas where local people raise crops on a small scale, mainly for the home or local market. As in Japan, sika deer and crops do not mix well on Green Island. The Hsiang has an active program to prevent deer damage, and they pay compensation to the local farmers. Nylon-net fences have been set up to try to keep the deer out, but these are easily breeched. A number deer die from entanglement in the fence, and it appears that at times local people may chase the deer that penetrate the nets (perhaps with a bit of human help) back into the nets to poach them. It is a complicated situation with positive and

558

D. R. McCullough

negative aspects. Sika deer have the potential to be a tourist attraction, but it is inevitable that they will come into conflict with local people who do not want to give up on their ability to raise subsistence crops, and Hsiang officials who do not want to spend funds on protective measures and compensation. At one time the Hsiang proposed to capture all of the deer (an impossibility) and put them in an enclosure of about 500 animals as a tourist attraction, similar to Nara Park in Japan as described by Torii and Tatsuzawa (chapter 25). However, this plan was soon abandoned as impractical. It is ironic that the conflict is the result of an inadvertent sika deer recovery program. It is largely true, as Hsiang authorities claim, that they did as much or more for sika recovery, without any intent or budget, as the captive breeding program at Kenting National Park.

Other Possible Release Sites It is desirable to establish a third wild population somewhere in the middle or northern end of the island, both to increase the total number of sika deer and to guard against a catastrophic loss of the two wild populations, both of which are near the south end of the island. Although unlike to both go to extinction simultaneously, it is notable that typhoons strike this area frequently, and it is not an impossibility. In addition, the new population can be established using both Kenting National Park and Green Island stock, thus mixing the genetic material to retain diversity, and recombining the rare alleles found only in one population or the other (Wilson 2000). A serious proposal was made to establish a new population of sika deer in Yangmingshan National Park in the northern part of Taiwan. This national park is located in the hills just to the north of Taipei. In a review of this proposal Severinghaus and McCullough (1996) recommended that this introduction not be made. The area suitable for sika deer was a small set of terraced former agricultural fields, which would not support more than a token number of animals. It would have to be fenced to prevent deer from dispersing through out the area and, inevitably, getting into agricultural fields and neighborhoods causing damage. The costs did not justify the benefits, given that the outcome would have been little more than an outdoor zoo, hardly necessary with the ready availability of the Taipei Zoo nearby. In 2002 I visited a possible area for a reintroduction at Mataian in the Hua-Tung Valley about 70 km south of Hualien. The land is in public ownership, held by the Minister of Transportation. It is a flood plain with marshes and many water birds and small mammals. There is considerable habitat suitable for sika deer, but the area is surrounded by agricultural areas. A sika deer reserve would require a strong fence that would contain the population. But the same would be true of any area within the major original habitat, which in the meantime has been converted to agricultural uses. Other areas, including on the western plain, the original core range, might become available in the future. The economic success of Taiwan, the original “Asian tiger” in development terms, has been having a pronounced impact on the

37 Sika Deer in Taiwan

559

ways of making a living, and the amount of marginal lands being abandoned for agriculture has increased; between 1977 and 1995 croplands decreased by 11.65% (Taiwan Provincial Forestry Bureau 1995). This trend promises to lead to areas suitable for an additional sika deer reserve in the future–thus completing the major requirements to declare the recovery program a success.

Literature Cited Chen, C.-L. 1968. Material culture of the Formosan aborigines. The Taiwan Museum, Taipei, Taiwan. Chiang, S. S. 1987. The relationship between sika deer and early Taiwan history. Pages 2–24 in The 1985 annual report of the Formosan sika deer restoration study. Conservation Research Report Number 38, Kenting National Park, Taiwan. (In Chinese.) Davidson, J. W. 1903. The island of Formosa: Historical view from 1430 to 1900. Macmillan, New York, New York, USA. Eu, H. H. T. 1969. Forest recreation and wildlife conservation in Taiwan. Forest and Forest Industry Development Project, Republic of China, Taipei, Taiwan. Hirth, F., and W. W. Rockhill. 1966. Chau Ju-Kua: his work on the Chinese and Arab trade in the twelfth and thirteen centuries entitled Cu-tan-Chi. Paragon Book Reprint Corporation, New York, New York, USA. Ho, C. S. 1986. A synthesis of the geologic evolution of Taiwan. Tectonphysics 125:1–16 Ho, C.-K., G.-Q. Qi, and C.-H. Chang. 1997. A preliminary study of the late Pleistocene carnivore fossils from the Penghu Channel, Taiwan. Annual of the Taiwan Museum 40:195–224. Hsia, L. C. 1990. Feeding behavior of deer. Pages 49–73 in 1998 Sika Deer Restoration Report. Kenting National Park, Taiwan. Hsieh, C.-M. 1964. Taiwan–Ilha Formosa: A geography in perspective. Butterworth, London, United Kingdom. Hu, C.-H., and H.-J. Tao. 1993. Monograph of fossil fauna in Penghu Peninsula. Culture Center of Penghu Hsien, Penghu, Taiwan. (In Chinese.) Kano, T. 1940. Zoogeographical studies of the Tsugitaka Mountains of Formosa. Shibusawa Institute of Ethnographic Research, Tokyo, Japan. Kenting National Park. 1984. The 1984 annual report of the Formosan sika deer restoration study. Conservation Research Report Number 18, Kenting National Park, Taiwan. Kuo, T.-Y. 1973. Early stages of the Sinicization of Taiwan. Pages 21–29 in P. K. T. Sih, editor, Taiwan in modern times. Asia in the modern world series Number 13, St. John’s University Press, St. John’s, Newfoundland, Canada. Kuroda, N. 1952. Mammalogical history of Formosa, with zoogeography and bibliography. Quarterly Journal of the Taiwan Museum 5:267–304. Lee, S.-W. 1981. Landslides in Taiwan. Pages 195–206 in South East Regional Symposium on Problems of Soil Erosion and Sedimentation. January 27–29, 1981, Bangkok, Thailand. Liew, P.-M., C.-M. Kuo, S.-Y. Huang, and M.-H. Tseng. 1998. Vegetation change and terrestrial carbon storage in eastern Asia during the last glacial maximum as indicated by a new pollen record from central Taiwan. Global and Planetary Change 16–17:85–94. Liu, H. Y. 1992. Study of the released sika deer on Green Island. East Coast Scenic Area, Taiwan Tourism Bureau, Taiwan. (In Chinese.) MacKay, G. L. 1895. From far Formosa: The island, its people, and missions. Fleming H. Revell Company, New York, New York, USA. McCullough, D. R. 1974. Status of the larger mammals in Taiwan. Tourism Bureau, Taipei, Taiwan. McCullough, D. R., and L. L. Severinghaus. 1998. Recovery program for the endangered Taiwan sika deer. Pages 177–184 in Z. Zomborszky, editor, Advances in deer biology. Proceedings of the 4th International Deer Biology Congress, Kaposvar, Hungary.

560

D. R. McCullough

Patel, A. D., and Y.-S. Lin. 1988. History of wildlife conservation in Taiwan. Zoology Department, National Taiwan University, Taipei, Taiwan. Patel, A. D., and Y.-S. Lin. 1989. Zoology council on agriculture. Forestry Series No. 20, Taipei, Taiwan. Roy, D. 2003. Taiwan: A political history. Cornell University Press, Ithaca, New York, USA. Ruhle, G. C. 1966. Advisory report on national parks for Taiwan 1965. American Committee for International Wild Life Protection, Special Publication Number 19, Bronx, New York, USA. Severinghaus, L. L. 1989. Natural Resources. Pages 49–127 in The Steering Committee for Taiwan 2000 Study, Taiwan 2000. Institute of Ethnology, Academia Sinica, Taipei, Taiwan. Severinghaus, L. L., and D. R. McCullough. 1996. A comprehensive review of the sika deer restoration program in Taiwan. Report to Yangminshan National Park, Taiwan, ROC. (In Chinese and English.) Su, H. J. 1985. Vegetation analysis on the native habitat of Formosan sika deer and proposal of its reintroduction area in Kenting National Park. The 1984 Annual Report of the Sika Deer Restoration Study, Conservation Research Report 18:63–101. Kenting National Park, Taiwan ROC. Taiwan Provincial Forestry Bureau. 1995. The third forest resources and land use inventory in Taiwan. Taiwan Provincial Forestry Bureau, Taiwan, ROC. Wang, Y. 1991. Current status of Formosan sika deer restoration program. Pages 277–288 in Lin, Y. S. K.-H. Chang, editors, Proceedings of the First International Symposium on Wildlife Conservation. Council on Agriculture Forestry Series Number 39, Taipei, Taiwan, ROC. Wilson, R. L. 2000. An investigation into the phylogeography of sika deer (Cervus nippon) using microsatellite markers. M. Sc. Thesis, University of Edinburgh, Edinburgh, Scotland, United Kingdom. Wayre, P. 1969. Wildlife in Taiwan. Oryx 10:46–56.

Chapter 38

The Present Status of the Re-introduced Sika Deer in Kenting National Park, Southern Taiwan Kurtis Jai-Chyi Pei

Abstract The captive breeding program for the Formosan sika deer (Cervus nippon taiouanus) was initiated in 1986 with a founder group of 22 deer in the Cheting Sika Deer Sanctuary in Kenting National Park. Their productivity was estimated to be 0.73/adult female/year according to three total counts between 1990 and 1996 in different sub-regions within the sanctuary. A total of 50 individuals was released into the adjacent area of the sanctuary in 1994, 1995, and 1997, after the captive population was believed to number more then 100 deer. In the year 2000 a mark-resighting estimation using camera traps suggested density within the 2–3 km2 distribution area was 27.6/km2, for a total of 55.3 to 82.9 deer in the freeliving population, which was lower then expected. The free-living sika deer used more secondary forest than natural broad-leaved forest, probably due to domestic goat use of the latter. The poor food quality provided by the secondary forest and predation by local domestic dogs were most likely responsible for the low recruitment rate. Accordingly, a strict policy toward goat herding and dog control inside of the national park was adopted. By 2003 the total area occupied by the free-living population was more than 20 km2, although most deer were still concentrated in the area surrounding the sanctuary. An additional 128 deer from the sanctuary were released into four scattered locations in the park between 2000 and 2003. The total population, including those inside of the sanctuary, in the Kenting National Park is likely still less than 500 deer. A more active management plan is required.

The Cheting Sika Deer Sanctuary In his survey of the larger mammals of Taiwan in 1973, McCullough (1974) reported that Formosan sika deer became extinct in the wild around 1969, and he recommended the establishment of a captive breeding facility to preserve the species. There are many small herds, usually a few individuals, held in private ownership in Taiwan to supply the velvet market, but long-term conservation of the unique subspecies, C. n. taiouanus, requires larger populations in public ownership. Subsequently, a recovery project and a sanctuary was established in 1984 in D. R. McCullough et al. (eds.), Sika Deer: Biology and Management of Native and Introduced Populations, DOI: 10.1007/978-4-431-09429-6_38, © Springer 2009

561

562

K. J.-C. Pei

Kenting National Park at the southern tip of Taiwan (Fig. 38.1). The distribution of major land-use types in the Kenting National Park and the location of the Cheting Sika Deer Sanctuary are shown in Fig. 38.1. The total area of the sanctuary is approximately 120 ha. In addition to a research center and an indoor captive facility, the open lands of the sanctuary were subdivided into a number of fenced enclosures of various sizes. Management intensity varied in the different enclosures; some were heavily managed for pasture growth, some were provided with supplementary food regularly, and some only provided minimum efforts beyond maintenance of the surrounding fence and protection of the habitat. The recovery project population was established with individuals from the Taipei Zoo and a number of private deer farms in central and southern Taiwan. Initially 40 individuals were brought to the Cheting Sika Deer Sanctuary in 1986 to form the founder group for captive breeding. However, only 22 individuals (seven males, 15 females) of this founder group participated in the breeding during the following progression (Wang and Yang 1988).

Fig. 38.1 Distribution of major land-use types in Kenting National Park showing landscape features and the location of the Cheting Sika Deer Sanctuary.

38 The Present Status of the Re-introduced Sika Deer

563

Table 38.1 Results of the three total counts in the Cheting Sika Deer Sanctuary, Kenting National Park, between 1990 and 1996 (information provided by Kenting National Park). Number of birth Number of initial Number of Description (duration) seasons females total births Indoor facility (October, 1986–December, 1990) 6-ha open enclosure (October, 1989–January, 1994) 25-ha open enclosure (January, 1991–February, 1996)

4

15

48

4

4

13

5

6

29

Three total counts of deer numbers were conducted between 1990 and 1996 for three different kinds of enclosures within the sanctuary (Table 38.1). The number of adult females was maintained the same during the whole monitoring duration, and mating opportunities were the same for all females. No deaths were recorded from these monitoring efforts. On average, every 100 adult females produced 73 calves each year in this semicaptive facility (Pei 2001). Although this productivity was lower than regular deer farms in Taiwan (usually more than 90%, Shih et al. 1984) it is higher than those in the protected areas in Russian Far East (21–45 calves/100 adult female/season, Makovkin 1999) and in Poland (60 calves/100 adult females/season, Matuzewski 1988).

Release of Deer to the Wild The first release of 10 sika deer (three males, seven females) into an area adjacent to the Cheting Sika Deer Sanctuary took place in 1994, by which time the number of deer was believed to have increased from the original 22 individuals into a total of 100–200 individuals inside of the sanctuary (Wang et al. 1994). Later two more groups, consisting of 10 and 30 deer respectively, were released into the same area in 1995 and 1997 (Table 38.2). Table 38.2 Number of sika deer released into the surrounding area from the Cheting Sika Deer Sanctuary and translocated to other locations in the Kenting National Park. Adult Location

Date

Male

Female

Sub-adult

Cheting

1/23/1994 4/23/1995 1/31/1997 6/24/2000 9/29/2002 9/29/2002 10/26/2002 3/15/2003 12/24/2003 12/25/2003

3 5 5 3 2 7 15 10 3 10 63

7 5 17 5 4 8 15 10 6 15 92

0 0 8 0 0 0 5 0 0 0 13

Longluan Tan Jioupeng Camp

Mt. Baishami Chuhuo Area Total

564

K. J.-C. Pei

All 50 released deer were followed closely by radio-tracking and direct observation to monitor their dispersal. The distribution of the free-living population was restricted within the immediate area around the sanctuary in the first several years; the total area occupied was only 200–300 ha in extent (Wang et al. 1996, 1997, 1998). Eight deer (two males, six females), which belonged to the initial group released, were found dead between 1994 and 2000, but the cause of death was not determined (Ming-Hsiung Pang, personal communication). Although most free-living deer settled in the area north of the sanctuary during the period, there was no evidence of deer crossing the Gangkou River and dispersing into the area with the largest natural forest in the national park (Fig. 38.2) (Wang et al. 1996, 1997, 1998). The low population density was considered to be responsible for the limited distribution and lack of greater expansion of the free-ranging population.

Fig. 38.2 The present distribution of the free-living population of sika deer surrounding the Cheting Sika Deer Sanctuary (shaded area) and the four additional release sites (black dots): Longluan Tan, Jioupeng Camp, Mt. Baishami, and Chuhuo. The darker shaded area close to the Sanctuary is the area with a much higher population density of the sika deer then the lighter shaded area (Chen and Wang 2003).

38 The Present Status of the Re-introduced Sika Deer

565

Increase of the Free-living Population in the Cheting Area A survey was conducted in the year 2000 to estimate the number of deer in the freeliving population and to determine the factors limiting recruitment. Camera traps were used to estimate the size of the deer population and also to evaluate the relative use of the secondary and natural broad-leaved forests in the Cheting area (Pei 2001). These two forest types are major habitats in this area and both contain significant edge habitat that is considered the prime habitat for sika deer. The secondary forest is dominated by Acacia confuse (Formosan koa), Leucaena leucocephala (white leadtree), and Lantana camara (lantana), while the natural broad-leaved forest is dominated by Ficus gibbosa (strangler fig) and Bischofia javanica (bishopwood). Deer density was estimated by using the same method developed by McCullough et al. (2000). This method uses a mark-resighting protocol (White 1996) to estimate the number of deer that share the same distributional range with the radio-collared (marked) individuals. Estimated density for the free-living population was 27.6 animals per 100 ha, for a total of approximately 55.3–82.9 deer in the whole 200–300 ha area outside of the sanctuary (Fig. 38.2) (Pei 2001). This figure was not only much lower than one might expect based on the productivity of the population inside the sanctuary; this density was also in the lower range of sika deer population densities in other protected area such as 147–183 deer/100 ha in Russian Far East (Ohtaishi 1999), 20–100 deer/100 ha in Japan (Ohtaishi 1988), and 33 deer/100 ha in Scotland (Whitehead 1988). For habitat selection, the same study also showed that, although the natural broad-leaved forest is a better habitat for the sika deer in Cheting area (Su 1985), they used secondary forest significantly more than natural broad-leaved forest (Table 38.3). Interestingly, domestic goats used significantly more of the natural broad-leaved forest, while the unleashed domestic dog, like the deer, tend to occur more in the secondary forest, although the difference for domestic dog was not statistically significant (Table 38.3).

Table 38.3 Comparison of the Occurrence Index* (mean with SD in parentheses) of sika deer, domestic dogs, and domestic goats from camera trapping in the two major forest types in the Cheting area of Kenting National Park. Secondary forest Natural broad-leaved (n = 3) forest (n = 8) Total camera trap working hours 3,685 10,935 0.6 (1.1)b Sika deer 2.5 (2.4)a Domestic dog 1.7 (3.0)a 0.2 (0.5)a Domestic goat 0.0a 1.4 (2.0)b *Occurrence Index = (number of pictures taken/total operating hours of automatic cameras) × 1,000. Values marked with different letters for the same species in different habitats were significantly different (t-test, p < 0.1).

566

K. J.-C. Pei

Because herds of domestic goats in this area are always accompanied by their owners, their presence will inevitably interfere with the willingness of the wild sika deer to share the same habitat. Moreover, because 75% of the food items foraged by goats and deer overlap in this area (Wang et al. 1997), the intensive consumption of goat herds within the natural broad-leaved forest will also inevitably reduce good quality food for the deer. Packs of domestic dogs can also be seen or heard quite frequently in the release area. These results suggest, therefore, that the lower recruitment rate of the free-living population might be a result of a combination of poor nutritional intake and, most likely, predation by domestic dogs. These two population limiting factors were later supported by studies on the sika deer’s food quality and carcass examinations, and management measurements were adopted accordingly. Pei (2001) recommended continuing release of adult sika deer from the sanctuary into the adjacent areas to accelerate the increase of the free-living population and, hence, encourage further dispersion to the north into larger areas of natural habitat (Fig. 38.2). Although the northern part of the Kenting National Park is suitable habitat for sika deer, the expansion of the wild population is expected to be slow enough to allow monitoring of their impact on the vegetation (Pei 2001).

Food Quality of Sika Deer in the Cheting Area Food quality of the sika deer in the Cheting area was evaluated from February 2002 to January 2003 based on the fluctuation of the fecal nitrogen content (Pei and Chen 2004). A total of 356 fresh fecal groups were collected from four sampling areas: the enclosed pasture within the sanctuary (EnPa), enclosed secondary forest within the sanctuary (EnSe), secondary forest outside of the sanctuary (FrSe), and the mature broad-leaved forest outside of the sanctuary (FrMa). Pangola grass (Digitaria decumbens) is the only grass species grown in the enclosed pasture (EnPa) sampling area, and intensive grazing rotation management of deer herds was made approximately once every month. Results showed that the mature broad-leaved forest provided food with the highest quality, followed by the secondary forest, while the managed pangola grass pasture was the poorest (Table 38.4, Fig. 38.3). Deer food from all forest habitats in the Cheting area maintained protein content near or around an ideal level (14%) year-round with little seasonal variation (Fig. 38.3). Low variation across the months probably was due to the downwind location from the heavy seasonal wind in winter of the Cheting area so that the sea breezes moderated the climate. However, probably because of the accelerated growth and maturation induced by high precipitation, food nitrogen content generally was lower during summer months (Fig. 38.3). In addition to the recommendation to control domestic goat herding within the national park to improve sika deer feeding and nutritional intake in the broadleaved forest, a better pasture management within the sanctuary, such as applying nitrogen fertilizer directly or introducing suitable leguminous grasses to the pasture, was also recommended (Pei and Chen 2004).

38 The Present Status of the Re-introduced Sika Deer

567

Table 38.4 Over-all average percent nitrogen for sika deer fecal samples collected from different areas in Kenting National Park between February 2002 and January 2003. Area* Average fecal percent nitrogen ± SD** n 117 EnPa 1.77 ± 0.30a EnSe 1.95 ± 0.21b 85 FrSe 1.93 ± 0.25b 83 FrMa 2.13 ± 0.14c 71 *EnPa = pasture within the sika deer sanctuary; EnSe = secondary forest within the sanctuary FrSe = secondary forest outside of the sanctuary; FrMa = mature broad-leaved forest outside of the sanctuary. **Average fecal percent nitrogen values marked with different letters were significantly different (ANOVA; p < 0.05). Fig. 38.3 Fecal percent nitrogen by month (means with 1 SD confidence limits) for sika deer samples collected from the pasture (EnPa) and secondary forest (EnSe) within the Sika Deer Sanctuary, and the secondary forest (FrSe) and mature broad-leaved forest (FrMa) of free-range areas outside of the sanctuary in Kenting National Park from February 2002 to January 2003. Upper and lower dotted lines represent, respectively, the typical fecal-nitrogen content of deer consuming food with the ideal (14%) or minimum required (7%) protein content.

Dog-kills of Sika Deer in the Cheting Area Dead sika deer found inside and outside of the sanctuary between January 2002 and June 2003 were examined as soon after death as possible to determine cause (Chen 2003). Dog-related kills were determined by the researcher, a licensed veterinarian, and experienced keepers of the sanctuary based on the location and appearance of the wounds and blood. Traces of attack or chase nearby and unusual dog activities

568

K. J.-C. Pei

Table 38.5 The details of the dog-related kills of sika deer in Cheting area, Kenting National Park, from January 2002 to June 2003. Ageb Locationc Sexa Date

F

M

Un

A

J

Un

In

Jan. 24, 2002 1 1 March 11, 2002 3 3 3 March 30, 2002 1 1 April. 23, 2002 1 2 3 3 Oct. 15, 2002 1 1 1 Jan. 27, 2003 5 5 5 March 10, 2003 1 1 2 2 March 24, 2003 6 1 6 1 7 March 27, 2003 1 1 1 April 22, 2003 2 2 May 5, 2003 1 1 1 May 15, 2003 1 6 1 6 7 June 7, 2003 4 4 4 June 21, 2003 1 1 1 Total 10 6 23 16 1 22 35 a F = female; M = male; Un = unknown. b A = adult; J = juvenile; Un = unknown. c In = Inside of the sika deer sanctuary; Out = outside of the sika deer sanctuary.

Out

Sum

1

1 3 1 3 1 5 2 7 1 2 1 7 4 1 39

1

2

4

(such as grouping or barking) before the body was found were also recorded to facilitate the judgment about cause of death. Fourteen cases of dog-related attack incidents causing a total of 39 sika deer deaths were recorded during the 18-month study period (Table 38.5). A majority of the kills (35 out of 39) were found within the sanctuary itself. Less than half of the cases (six out of 14) involved only one deer, while groups of more than three deer killed at one time all happened inside of the sanctuary (Table 38.5). All larger groups of deer found dead were near the corner of the fence suggesting that local domestic dogs were already skillful in chasing and driving the group of deer along the fences and trapping them in the corner (Chen 2003). Although the data were not sufficient to evaluate the actual impact of these deaths on the growth of the deer population, diminishing these kills would definitely help to increase the growth rate of the present population. Obviously, a stricter policy toward control of free-ranging behavior of domestic dogs in nearby villages is desirable.

Present Status The latest survey of deer numbers in 2003 revealed a minimum of 147 deer within the sanctuary (Chen and Wang 2004). Although no reliable population estimation was made for the free-living deer population, the same survey also showed that the

38 The Present Status of the Re-introduced Sika Deer

569

free-living population continued to expand its distribution south-eastward into the coastal areas of the right-hand peninsula (Fig. 38.2), and the total area occupied was more than 2,000 ha by then (Chen and Wang 2004). However, the majority of the population still concentrated in the area surrounding the sanctuary (Fig. 38.2) (Chen and Wang 2004). Although a gradual expansion of the free-living deer population might be advantageous for closely monitoring the interaction between sika deer and vegetation in the Kenting National Park (Pei 2001), authorities transferred and released more deer from the sanctuary into four more distant locations inside or right on the edge of the park starting in the year 2000 (Fig. 38.1). A total of 128 additional deer were released between 2000 and 2003 (Table 38.2). Monitoring of these scattered newly formed small populations is necessary. Kenting National Park has now adopted stricter policies toward goat herding and dog control inside of the park, which should be significant to the sika deer population. However, considering the slow growth rate for the free-living population, the total population size of the sika deer (including those inside of the sanctuary) in Kenting National Park is likely still less than 500. A more active management plan is required.

Literature Cited Chen, C. Y. 2003. The potential impact of free-ranging dog to sika deer in Cheting area, Kenting National Park. Master thesis, National Pingtung University of Science and Technology, Pingtung, Taiwan. (In Chinese with English abstract.) Chen, S. C. and Y. Wang. 2004. Population distribution of Formosan sika deer (Cervus nippon taiouanus) in Kenting National Park. Bulletin of National Park 14:81–102. (In Chinese with English abstract.) Makovkin, L. I. 1999. The sika deer of Lazovsky Reserve and surrounding areas of the Russian Far East. Almanac Russki Ostrov, Vladivostok, Russia. Matuzewski, G. 1988. Polen. Pages 5.2-PL-2:1–9 in E. Eick, J. A. Willett, R. König, and K.-H. Schulze-Schwefe, editors. Sika. Internationale Arbeitsgemeinschaft Sikawild, Kurkölner, Germany. McCullough, D. R. 1974. Status of larger mammals in Taiwan. Taiwan Tourism Bureau, Taipei, Taiwan, R.O.C. McCullough, D. R., K. C. J. Pei, and Y. Wang. 2000. Home range, activity patterns, and habitat relations of Reeves’ muntjacs in Taiwan. Journal of Wildlife Management 64:430–441. Ohtaishi, N. 1988. Memorandum of the status of sika deer in Japan. Pages 5.1-J: 1–8 in E. Eick, J. A. Willett, R. König and K.-H. Schulze-Schwefe, editors. Sika. Internationale Arbeitsgemeinschaft Sikawild, Kurkölner, Germany. Pei, K. J. C. 2001. The present status of the re-introduced Formosan sika deer (Cervus nippon taiouanus) in Kenting National Park. Quarterly Journal of Chinese Forestry 34:427–440. (In Chinese with English abstract.) Pei, K. J. C. and Chen, T. J. 2004. Food quality of the Formosan sika deer in Cheting area, Kenting. Taiwan Journal of Forest Science 19:353–362. (In Chinese with English abstract.) Shih, C. H., S. K. Yang, S. M. Sung, and C. S. Hwang. 1984. A survey on the feeding and management of Formosan sika deer in Taiwan. Kenting National Park Conservation Research Report 18:248–75. (In Chinese.)

570

K. J.-C. Pei

Su, H. J. 1985. Vegetation analysis of the natural habitat of the Formosan sika deer and the siteselection for its sanctuary in the Kenting National Park. Kenting National Park Conservation Research Report 18: 63–101. (In Chinese.) Wang, Y., and H. J. Yang. 1988. The Sika Deer Recovery Project. Pages 4–25 in 1988 Conference on the Formosan Sika Deer Recovery Project. Kenting National Park Conservation Research Report 63. (In Chinese.) Wang, Y., G. W. Kuo, C. H. Hu, and M. T. Chen. 1994. The monitoring of the released Formosan sika deer. Kenting National Park Conservation Research Report 91:1–83. (In Chinese.) Wang, Y., S. C. Chen, Y. S. Sun, C. T. Lin, and S. F. Liao. 1996. Ecology of the released Formosan sika deer at Kenting National Park. Kenting National Park Conservation Research Report 93:1–63. (In Chinese.) Wang, Y., S. C. Chen, C. T. Lin, S. C. Chan and R. C. Chang. 1997. Population and environmental monitoring of the released Formosan sika deer at Kenting National Park. Kenting National Park Conservation Research Report 98:1–45. (In Chinese.) Wang, Y., S. C. Chan, S. C. Chen, and F. H. Chen. 1998. Population and environmental monitoring of the released Formosan sika deer at Kenting National Park—Application of GIS system. Kenting National Park Conservation Research Report 100:1–65. (In Chinese.) White, G. C. 1996. NOREMARK: Population estimation from mark-resight surveys. Wildlife Society Bulletin 24:50–52. Whitehead, G. K. 1988. British Isles. Pages 5.2-GB-3: 1–11 in E. Eick, J. A. Willett, R. König and K-H Schulze-Schwefe, editors. Sika. Internationale Arbeitsgemeinschaft Sikawild, Kurkölner, Germany.

Chapter 39

Sika Deer in Continental Europe Ludeˇ k Bartoš

Abstract In this chapter I review the history of introduction of sika to continental Europe by country and summarize their acclimatization to the local environment. Beginning about 150 years ago sika deer were introduced into most of the countries in western, central, and eastern Europe; northern and southern countries were less affected. Besides free-ranging feral populations, there are an unknown number of deer parks, zoological gardens, and farms containing sika deer. With consistent growth of deer farming in Europe, there is increased interest in captive breeding of sika. Due to anatomical and behavioral features, sika appear to be successful competitors with local deer species. Although characterized as sedentary, they have the capability to migrate long distances. Frequently lone individuals, usually males, appear in areas with no established sika deer where they join red deer. In the beginning, imported sika occasionally interbred with other species, such as red deer, hog deer, and axis, mostly by coincidence. Most published records described the hybrids as due to a break of the geographical isolation between species. Nevertheless, sika and red deer (Manchurian wapiti, Cervus elaphus xanthopygus) hybridize naturally where they overlap in Far East Russia. In many areas where sika coexist with red deer, it is still believed by many hunters that no interbreeding has occurred, despite massive evidence to the contrary, including modern genetic techniques. Disregard of hybridization has resulting in introgression of sika and red deer genes in many areas. Still, there is no general wildlife management strategy in continental Europe to realize the danger and to solve the situation, and free-ranging sika populations further increase in numbers.

History The history of sika deer in Europe began some 150 years ago. At that time many exotic animal and plant species were introduced as enrichment to European fauna and flora. Various deer species and subspecies were introduced to make new game animals available. However, in several European countries for example, red deer was a “royal species” which could be hunted only by privileged individuals. D. R. McCullough et al. (eds.), Sika Deer: Biology and Management of Native and Introduced Populations, DOI: 10.1007/978-4-431-09429-6_39, © Springer 2009

573

574

L. Bartoš

The first translocations of sika deer were from their distant native range (Fig. 39.1). According to Eick (1995d), the main importer and supplier of sika deer in continental Europe was Hagenbeck of Hamburg. The geographical origin of these deer can no longer be ascertained, as the Hagenbeck archive was destroyed during World War II. In addition, the company of Mohr, in Ulm, has been occasionally reported as a supplier. Nevertheless, it is not clear if this company ever imported any sika from their native territories. In the beginning, it was not recognized that the imported animals could have belonged to various subspecies; in fact, it was entirely ignored (Eick 1995e). In part this was because the first imports of sika subspecies to Europe were parallel with the discovery of the species itself. First descriptions of the holotype were often based on fragmented material. For example, Swinhoe (1864) described his Cervus nippon hortulorum from “three skins of two bucks and one doe.” Moreover, after import to Europe, various subspecies had been mixed and intercrossed. As a result, in Europe generally only the large hortuloid types and the smaller nipponoid types can be distinguished (Eick 1995e). Recent studies using microsatellite and mitochondrial DNA analyses have revealed the origin of sika deer in the United Kingdom (Goodman et al. 2001). This is promising for further reconstruction of the history of sika introductions. The history of individual countries is presented in alphabetical order. Figure 39.2 shows distribution of sika deer in Europe with earliest dates of the first occurrences of the species. Estimates of numbers of sika deer are shown in Table 39.1.

Fig. 39.1 A map showing the distribution of the basic locations from which sika deer were imported and/or acclimatized during the translocation to Europe.

39 Sika Deer in Continental Europe

575

Fig. 39.2 Distribution of sika deer in Europe with earliest dates of the first occurrences of the species. (1-Portugal; 2-Spain; 3-Ireland; 4-United Kingdom; 5-France; 6-Belgium; 7-The Netherlands; 8-Switzerland; 9-Germany; 10-Denmark; 11-Norway; 12-Sweden; 13-Italy; 14Slovenia; 15-Austria; 16-Czech Republic; 17-Croatia; 18-Poland; 19-Slovak Republic; 20Hungary; 21-Bosnia-Hercegovina; 22-Finland; 23-Lithuania; 24-Estonia; 25-Latvia; 26-Albania; 27-Yugoslavia; 28-Macedonia; 29-Greece; 30-Romania; 31-Belarus; 32-Ukraine; 33-Bulgaria; 34-Russia; 35-Moldova).

Austria There are two feral populations of sika in Austria, both located along the Danube River, west and north-west from Vienna, in the Persenburg-Ostrong and PreuwitzTulln Danube Meadows (Eick 1995a). The animals of the first population, said to come originally from Japan, were imported from the Kluk enclosure, near Podeˇbrady (east from the capital, Prague), Czech Republic (Niethammer 1963). Because various sika deer subspecies were crossed in this enclosure (Kokeš 1970), however, the stock can be hardly called of Japanese origin. Subsequently Emperor Franz Josef I of Austria received two stags and five hinds as a gift from the Emperor

576

L. Bartoš

Table 39.1 Numbers of free-living sika deer in Continental Europe. (Adapted from Eick et al. 1995 if not stated otherwise.) Country Year of census Area Number Austria

1983 1997a

Czech Republic

1988 1988 1988 2003b 1977c

Denmark Estonia France

1980 1980

Germany

1995 1990 1995 1989 1991 1977d 1984

Hungary Lithuania Poland

1985 1970e 1981 1981

Persenburg–Ostrong Preuwitz-Tulln Total: Plzenˇ North Bouzovsko Total: Total: Total:

150 700 850 1,000 450 1,450 4,954 500

Rambouillet Forêt de la Harth Total: Glücksburg Hochrhein Hütten and Duvenstedt Mountains Möhnesee Ostangeln, Schwansen Schlitz Weserbergland Total: Fehérvárcsurgó Total: Total: Kadyny Kobiór Total: Total: Südranden, Total:

90 70 160 25 350 70 800 150 35 70 1,500 115 704 160 30 190 12,000 90

Russia and Ukraine 1984f Switzerland 1988 a Actualized in Bartoš et al. (2003). b According to changes in the system of monitoring population dynamics, data for the numbers according to areas of sika deer distribution has no longer been available. Hence, for the purpose of this chapter, numbers for the two main Czech sika areas were adapted from earlier records; the total numbers were obtained from the recent official statistics of the country. c Compiled from Bennetsen (1977). d Probably now extinct (Eick 2003, personal communication). e Population is probably now extinct (Baleisis et al. 2002). f Compiled from Uloth (1984).

of Japan in 1907 via Hagenbeck in Hamburg and released them into the enclosure in Persenburg. When the enclosure fell into disrepair, the animals dispersed into the wooded areas of the Ostrong (Eick 1995a). The second population is based on sika imported from the Czech Republic into the enclosure at Neuaigen, north of Tulln. The enclosure was destroyed at the end of the war in 1945 and the deer escaped into the wild (Eick 1995a).

39 Sika Deer in Continental Europe

577

Czech Republic In 1891 sika were first released into the Kluk enclosure near Podeˇbrady (Kokeš 1970). In the late nineteenth and early twentieth century, sika were frequently imported for deer parks. The origins were allegedly Japan, eastern China, Korea, and southeastern Siberia, Russia (around the Ussuri River). These sika were regarded to belong to both Japanese (Cervus nippon nippon) and Dybowski’s sika (Cervus nippon hortulorum) subspecies (Wolf and Vavruneˇk 1975–1976). Additional sika were also imported through Hagenbeck; these originated for the most part from Japan and Russia (Komárek 1945). There were also imports from estates in England and Austria (Kokeš 1970). Sika were maintained in enclosures where they prospered until the 1940s. During World War II and subsequent political changes in the country after the war, some deer parks were destroyed and sika escaped (Vavruneˇk and Wolf 1977). This was the case of two parks, Lipí in Maneˇtín, Bohemia, with sika kept since 1897 (Doležal 1960) and Cˇemíny, with sika kept since the beginning of the century. These escaped sika were the base of the west Bohemian sika population around Plzenˇ (Pilsen) North. Migrating sika colonized further territories, such as Slavkovský les (Wolf and Vavruneˇk 1975–1976; Vavruneˇk and Wolf 1977). In the 1970s, numbers of feral sika in this area reached 1,000. Since that time sika numbers have multiplied. Sika deer in the second main area in Czechia, Bouzovsko, northern Moravia, came from a deer park adjoining a castle on the estate Žádlovice belonging to the Dubský family. Sika deer were allegedly imported from Hagenbeck (Babicˇka et al. 1977). The first deer who escaped from the park were harvested in 1918 (Hošek 1982). In 1945, after World War II, the park fences were destroyed and 14 sika deer (four stags and 10 hinds) escaped and became the base of the current population (Babicˇka et al. 1977; Hošek 1982).

Denmark Sika were first introduced in Denmark as park deer in 1900 (Bennetsen 1977). One stag and one hind were released in an enclosure on the estate of Svenstrup. The origin of these sika is not known for certain, but some evidence suggests that they may have come from Ireland. In 1909 some sika from Hagenbeck in Hamburg were released in a deer park on the estate of Knuthenborg. These two releases seem to have formed the basis for all other Danish populations. Yet some doubt exists about the origin of the population on the estate of Frijsenborg. This was the first freeranging population in Denmark. It seems that sika were released there several times between 1900 and 1910. They may have come from Hagenbeck, from Svenstrup, or from both. Apart from two herds, Mejgaard and Ormstrup-Tange, founded by sika that in all probability immigrated from Frijsenborg, all other sika herds were established by stocking or by deer which broke out of deer parks. It is evident that the stags are migratory.

578

L. Bartoš

France In 1890, the Emperor of Japan gave a gift of one stag and three hinds from Nara Park, Japan, to President Carnot of the Third Republic. The animals were released into a 900 ha enclosure in Rambouillet. The population reached 200 animals, but has been reduced to about 90. Sika share space with roe deer (Capreolus capreolus) (Legrand 1988; Eick 1995c); the environment is said to be too dry for red deer (Eick 2003, personal communication). Sources of sika in the second location, Forêt de la Harth, were Mulhouse Zoo and the Huttenheim enclosure (southwest of Strasbourg). In 1953 a pair of sika from Mulhouse Zoo was placed into a 3 ha enclosure and further deer were imported from the Huttenheim enclosure two years later. The Forêt de la Harth enclosure was later enlarged to 230 ha and further in 1976 to 290 ha. In 1970, 10 sika were released into the wild. The herd reached some 250 deer of which 30 were free-ranging. These numbers were then reduced to some 70 animals. Surplus sika were used to establish new herds in Alsace, Massif Central, Burgundy, and Switzerland (Cailmail 1988).

Germany Seven areas with sika are recognized in Germany: Möhnesee (east of Dortmund) (Eick 1988), Hochrhein (near Waldshut, in the south close to the border with Switzerland) (Wicki 1988), Ostangeln and Schwansen (in the north, close to the Denmark border) (Rumohr-Rundhof 1988), and several smaller areas—Schlitz (Ueckermann 1988), Glücksburg (north of Ostangeln) (Eick 1995d), the Hütten and Duvenstedt Mountains (Eick 1995d), and Weserbergland (southwest of Hannover) (Hake 1988). This information has been summarized by Eick (1995d). The largest German sika population in Möhnesee was derived from sika from Hagenbeck and from the Hellabrun Zoo, Munich. In 1893, Baron von Donner released six to 10 sika into an 800 ha enclosure near Neuhaus, Möhnesee. The enclosure also contained red deer, axis (Cervus axis), sambar (Cervus unicolor), hog-deer (Cervus porcinus), barasingha (Cervus duvauceli), mouflon (Ovis musimon), and antelope (Feaux De Lacroix 1913, cited in Eick 1995d). Eick (1995d) further cited information from Holtkotte (1941 in Eick 1995d) reporting “… Dybowski and sika stags were imported from Manchuria… In 1928 and 1930 several Dybowski were imported into the same park… Unfortunately, after two years, the Dybowski stags were inadvertently shot.” Sika got out of the enclosure first in 1936, when the fence was broken by snow. Allegedly, the majority of the escaped animals returned to the enclosure. Sika finally left the enclosure in 1945; at first they spread eastward and later westward as well. They share the environment with red deer, roe deer, and wild boar (Eick 1995d). The second largest population is located at Hochrhein (Wicki 1988). In 1911 the sika were imported and released into a 180 ha enclosure with fallow deer (Dama dama),

39 Sika Deer in Continental Europe

579

mouflon, and ibex (Capra ibex) at Rohrof, near Waldshut. These sika originated from Basel Zoo, St. Peter Game Park in Freiburg, and Hagenbeck, Hamburg. In 1939 the sika were released from the enclosure (Eick 1995d). The third largest German population, at Ostangeln, Schwansen, was started by Paul von Schiller in 1928. He purchased seven sika deer from Hagenbeck and released them into a 9 ha enclosure at Buckhagen. The first deer escaped at the beginning of World War II, but were caught again. As the war continued, it was difficult to maintain the enclosure and the deer were released into the wild. The escaped sika have established themselves quite well. The other deer species in the area are fallow and roe deer (Rumohr-Rundhof 1988). The Weserbergland population was established by Karl Freiherr von WolffMetternich of Haus Amelunxen, who bought in 1933 four sika deer from Carl Hagenbeck. He released them into a 31 ha enclosure originally used for keeping roe deer. The first individuals escaped here during World War II as well. In 1945 up to 40 deer were released into the wild. Further sika were imported later—in 1968 from the Frankfurt Zoo, in 1972 from Nürnberg, and in 1976 from Zoo Wuppertal and from the deer park Schirmecke (Hake 1988). In the Hütten and Duvenstedt Mountains, sika were imported as a substitute for fallow deer. In 1964, Hans Peter Greve of Holzbunge, introduced three sika from Mohr, Ulm, into a small 0.30 ha enclosure. One hind escaped shortly afterwards, but remained in the vicinity of the pen. In 1965 another enclosure, owned by Claus Kuhr in Hegelholt, Neu-Duvenstedt, was stocked with three sika from Mohr, Ulm. Since that time, offspring of sika in these enclosures have been continually released into the wild (Eick 1995d). Almost no information is available for the origin of the population at Schlitz. The origin of the deer is unknown, but sika appeared in the wild in 1960 (Ueckermann 1988). It is not sure, however, if the population still exists (Eick 1995d). The most recently established feral population is the one in Glücksburg. In the 1970s a display was erected and stocked with sika probably from Denmark. Due to heavy snow, the animals escaped in the winter of 1978/79. Since the beginning, migration between Glücksburg and Ostangeln has been recorded (Eick 1995d).

Hungary The only place with sika in Hungary is Fehérvárcsurgó (southwest of Budapest). In 1910, sika deer from Hagenbeck were introduced into the deer park at Fehérvárcsurgó belonging to the Count Gyula Károlyi. As in other areas, sika were doing very well and their numbers reached some 200 deer by the 1940s. Nevertheless, during World War II the population was decimated and a few remaining deer escaped into the Bakony Mountains. In 1946 a further pair of sika was released. Throughout following two decades, the population was supported by further imports from Budapest Zoo—seven in 1958 and five in 1969. The origin of these sika was the former USSR. In 1975 they imported a further 28 deer from the USSR (Markovic 1988).

580

L. Bartoš

Poland Sika deer were introduced into the present territory of Poland at the turn of the nineteenth century, perhaps in 1895 a smaller nipponoid type in Kobiór (southern Poland), and in 1910 a larger hortuloid type in the Kadyny (northern Poland) (Matuszewski 1988). Other reports give 1905 for the first introduction of sika to Kobiór from England via Hagenbeck (Niethammer 1963; Matuszewski and Sumin´ski 1984; Eick 1995f). The original Kadyny sika were imported via Hagenbeck from Japan (Niethammer 1963; Eick 1995f). The acclimatization in both locations was similar: individuals were first kept in enclosures and then released into the wild (Matuszewski 1988).

Russia and Ukraine According to Eick (1995b), the translocation of sika deer from the Russian far east was realized in several steps. The first sika deer came from Primorsky Krai, an area of natural sika occurrence, mostly from deer farms around Vladivostok (Fig. 39.1). Further imports were a mix of continuing transports from the far east. Around 1909 deer were brought to Askania-Nova, Ukraine (Fig. 39.1), which became the source for many releases in the former Soviet Union (USSR) and internationally. There were two basic reasons for translocation of sika deer from the Russian far east. In the second half of the nineteenth century, the numbers of wild sika deer began to decline. In 1924, harvesting free-living sika deer became prohibited (Zhirnov et al. 1978). Because the situation did not improve, in 1938 240 deer were released into several nature reserves in central Asian regions of the former USSR, southwest of Chelyabinsk and further to the west. These reserves were in Busuluk Bor (northwest of Orenburg), Kuibyshev, an area northeast of Riazan, in Mordvin, in the Khoper reserve, an area southwest of Borisoglebsk, and in Teberda in the Caucasus. In some of the locations climate conditions were extremely hostile, snow cover in particular, and the deer did not establish themselves. The second reason for translocation was increasing demand for medical products from antlers. Around 1933, sika were used for velvet production in Altai region, where they were crossed with marals (Siberian wapiti, Cervus elaphus sibiricus). More recently, further transports followed (Eick 1995b). From 1918 to 1972, approximately 2,400 sika deer were released in eastern Europe (Pavlov et al. 1974). In its natural range, sika was decreasing in number and its hunting was banned (Makovkin 1999; Baskin and Danell 2003). In the European part of the former USSR, it has been bred for game and used for sport hunting (Baskin and Danell 2003). Today, sika deer are well-established in the European part of Russia and in Ukraine in a number of locations including near Moscow and near Kiev (Eick 1995b). According to Baskin and Danell (2003), recent introduced populations of introduction are in Voronezh Oblast, Oksky Nature Reserve, Khopersky Nature

39 Sika Deer in Continental Europe

581

Reserve, Mordovsky Nature Reserve, and Ilmensky Nature reserve (the latter located in the Asian part of Russia).

Switzerland In Südranden in northern Switzerland, sika immigrated from the population in Hochrhein, Germany. The first animal was recorded in 1917. Nevertheless, the main immigration occurred after the Rohrof deer park was opened in 1939 and sika deer were released. A relatively stable population has established. Wandering sika deer may be seen also in other areas (Schoenberger et al. 1988).

Other Countries According to anonymous information on the internet, in Estonia sika deer were introduced in 1956 in Vääna Forest District. Later single individuals dispersed to Alutaguse from the population in neighboring areas of Russia, introduced there in 1987. At present the introduced sika population is probably extinct. In eastern Estonia single immigrants from neighboring areas may occur. According to Baleisis et al. (2002), 24 sika deer were brought to Lithuania to be acclimatized in 1954 from the Chelyabinsk farm in the Gorno-Altai Region (Russia; Fig. 39.1) and released in the Dubrava forest (Kaunas district). Several years later the numbers of sika in the forest was augmented to 60 to 70 individuals. In the 1970s, however, the population settled at this level and later started decreasing. Dispersal of the sika deer to the neighboring forests was not observed. Since 1992 there are no data on the numbers of this species in the Dubrava forest. Sika are also listed among six nonnative species introduced to Lithuania (Baškyte et al. 1997). On the other hand, they are not included among “legal game” species nor among “natural resources” (as are, e.g., fallow deer). Therefore, it is unlikely that sika deer would have any significant meaning these days in the wild in Lithuania. A well-studied sika population exists in Moldova (Prisyazhnyuk and Tchegorko 1990), but no data are available on its current situation. There are internet advertisements for hunting sika deer in other countries not mentioned here, such as Finland, Bulgaria, and Slovenia, suggesting the existence of the species there as well.

Captive Sika in Europe Besides free-ranging sika populations, there are an unknown number of deer parks, zoological gardens, and other enclosures and farms containing sika deer of various origin in Europe. These numbers and locations continually change over time. Only

582

L. Bartoš

fragmentary information is available and it is difficult to get any reliable data and/ or trend estimates. According to the ISIS (International Species Information System) website, there are 29 zoological gardens in Europe keeping 417 sika of various subspecies (listed as: Cervus nippon, C. n. nippon, C. n. dybowskii, C. n. hortulorum, C. n. pseudoaxis, and C. n. taiouanus). Most countries with feral populations of sika deer have this species also in enclosures. For example, sika can still be found in most deer parks in Denmark (Herlevsen 2003, personal communication). Switzerland has an insignificant number of free-ranging sika (Table 39.1). In 1997, a survey of captive deer in that country reported 13 deer parks with about 130 sika deer and almost 30 farms with over 200 additional sika (Daenzer 2003, personal communication). With consistent growth of deer farming in Europe (Bartoš and Šiler 1994; Reinken 1998; Audenaerde 2002), there is increased interest in the captive breeding of sika. An exception is Denmark. In a new regulation issued in 1993, the species of deer which could be legally farmed were reduced to red and fallow deer. Existing herds of sika deer could stay, but it was prohibited to sell livestock or increase numbers of deer on farms (Herlevsen 2003, personal communication). No constraints exist in the majority of other countries. In Belgium, during the last six years there have been several imports of sika from Ireland and Germany. These sika have been distributed over several hobby parks in the country. The largest holding has some 60 animals now (Audenaerde 2003, personal communication). In Lithuania, about 1,000 sika deer are farmed in the Kaunas and Klaipeda districts (Malakauskas and Grikieniene 2002 and Grikieniene 2003, personal communication).

Interspecific Interactions Ungulate species from outside Europe have often been introduced and become naturalized. These species are not likely to have co-evolved with the indigenous species, and as such may have overlapping niches which result in competition. Accounts of such apparent competition often involve sika deer (Latham 1999). Due to anatomical and behavioral features, sika deer appear to be particularly efficient in breaking down fibrous food (even of poor quality) and in building up energy reserves (Hofmann 1988, 1989; Takatsuki 1988). Interspecific sika behavior is rather flexible. Intolerance by sika deer toward other species has been reported from several places in Europe (Bartoš and Žirovnický 1982). Primarily, they dominate the smaller roe deer (e.g., Rowland 1967; Cadman 1980; Opluštil 1980). Danilkin (1996) cited instances of sika deer chasing roe deer from feeding sites and links this to a decline in roe deer numbers as sika deer increased following their introduction. Sika appeared aggressive and a successful competitor with fallow deer in the Czech Republic. Within 12 years after sika appeared in the wild, the formerly well-established fallow deer in the West Bohemia area have almost entirely disappeared, surviving only at the periphery of the area (Wolf and Vavruneˇk 1975–1976; Vavruneˇk and Wolf 1977). On the

39 Sika Deer in Continental Europe

583

contrary, in Ostangeln, Germany, fallow deer seem to be successful competitors with the feral sika deer, outnumbering sika in the area (Rumohr-Rundhof 1988). Sika are also reported to disturb red deer rutting (Wolf and Vavruneˇk 1975–1976). Sumin´ski (1964 cited in Matuszewski 1988) suggests that Kadyny sika dominate red deer “because of their greater bravery.” According to Matuszewski and Sumin´ski (1984), mainly during the rut, young red stags have been attacked by older sika males. In their natural habitat, sika seem to compete successfully with much larger Manchurian wapiti (Cervus elaphus xanthopygus). Although the Manchurian wapiti is behaviorally dominant, sika deer is the dominant ecological competitor (Makovkin 1999). As with red deer, wherever there is a high population density of sika deer there is a decline in the population of roe deer. This occurs because sika deer rapidly exhaust the food supply (Makovkin 1999). The interspecific competition usually occurs on a “non-visible” basis, which may lead human observers to the opinion that no interspecific competition between local species and exotic sika deer exists. Putman (1996) has a good illustration from the New Forest area based on unpublished results of his students. In Boxall’s fouryear study of fallow, roe, sika, and red deer in Roydon Wood, individuals of two or more species were recorded in the same sampling unit (1 ha grid square) on only 30 occasions in 2,580 observations (Boxall 1990, cited in Putman 1996). However, animals of different species were commonly recorded within different patches of the same habitat, and there was no evidence of interference to exclusion. Similar conclusions are reported by Sharma (1994, cited in Putman 1996), who examined the potential for direct interaction between fallow and roe deer where they occur in sympatry, assessing from transect data the expected number of transect walks on which no deer of either species were seen, the number on which only roe, or only fallow were seen, and the number in which both roe and fallow might be encountered within the same site. Observed frequencies were compared against these expected values; no tests showed a significant difference between observed and expected values, offering no evidence that either species makes any attempt to avoid a site because of the presence of the other (Sharma 1994, cited in Putman 1996). Levels of overlap calculated (0 = none, 1 = complete) in relation to habitat use were universally high, with overlaps in excess of 0.66 recorded between all pairs of deer species in all seasons, with summer values being highest.(0.89 for roe, 0.82 for fallow, and 0.72 for red deer). The potential for interaction between sika and other deer in the New Forest was also indicated by overlap in diets based on fecal analysis and direct feeding observations. Again, the overlaps were high ( (0.49 for roe, 0.89 for fallow, and 0.95 for red deer). Large variation in rutting activity for sika deer has been reported across Europe and also within very small areas. Matuszewski (1988) reported the start of the rut in mid-September in the south, while mid-October in the north of Poland. According to Bennetsen (1977), in Denmark the rutting season varies considerably from one herd to another. In Frijsenborg the rut normally starts about 1 September, whereas it starts as late as 1 November or even later in other districts. In addition to this it seems that the rut covers a considerably longer period in each individual herd than is the case for red deer and fallow deer. As a result, the calving period may range

584

L. Bartoš

from 27 March to 5 October (Bennetsen 1977). This is very similar to the situation in the south of Switzerland (Schoenberger et al. 1988). In the Czech Republic, in Weserbergland and Möhnesee, Germany, Forêt de la Harth, France, and in the areas of origin, rutting usually starts in mid-October and persists often until December (Wolf and Vavruneˇk 1975–1976; Cailmail 1988; Hake 1988; Baskin and Danell 2003). An extended period of rut resulting in late births may be a good indication of adaptability of the species in the given conditions. In general, late deliveries may be caused by the fact that females who fail to conceive in the early stage must be in a good condition to persist in estrous cycling as approaching winter brings worsened conditions (e.g., Bartoš 1982). Thus, sika may be able to overcome severe environmental conditions sometimes better than the local species. Introduced sika deer have often higher condition ratings than local species in other continents also, such as North America (Armstrong 1980; Butts et al. 1982; Keiper 1985; Harmel and Armstrong 1987; Richardson and Demarais 1992; Feldhamer and Armstrong 1993) and elsewhere.

Migration Although sika have been characterized across Europe as sedentary and faithful to a given location, they have the capability to migrate a long distance. In Poland, the Kadyny sika were recorded to migrate 80–160 km (Matuszewski and Sumin´ski 1984). Movement over long distances during the rut in particular has also been reported from Russia (Sokolov 1959; Yevtushevskiy 1974; Makovkin 1999). Nevertheless, published evidence based for example on radiotelemetry of sika in continental Europe is rare (Kistler 1995). It happens frequently that a wandering single animal, mostly a stag, occurs in an area with no established sika deer. Such an individual usually approaches and joins red deer. In the Czech Republic it has happened a number of times that a sika stag breaks into an enclosure containing red deer. His origin is usually not known (Bartoš 2008). Those sika stags that are migratory are often seen roaming together with red deer (Bennetsen 1977). Sika stags joining red deer groups have been reported also from Poland (Bartoš 1982; Matuszewski 1988) and the Czech Republic (Bartoš and Žirovnický 1982; Bartoš 2008).

Hybridization At the beginning, imported animals were not recognized according to their subspecific origin. They were kept in mixed stocks and bred in zoos and game preserves. Occasionally, sika were interbred with other species, such as red deer, hog deer, and axis (Powerscourt 1884; Benirschke 1967; Bartoš 1991). Eick (1995d) cited from his personal correspondence with H. Heck from early 1940s:

39 Sika Deer in Continental Europe

585

“I myself know of hybrids of sika both with red deer and six-pointers (hog-deer, axis) both as half-blood and three-quarters blood. It can also be assumed that in the case of such crossbreeding infertility does not occur … under natural conditions it is rather exceptional for stags to exhibit such infidelity. Hybrids have always occurred in small reserves.” Sika were hybridized both intra- and interspecifically for improvement in antler size to increase velvet production on farms. As such, it has been well-documented and published, though usually only on a local basis (Menard 1930; Mirolyubov 1936, 1949; Mirolyubov and Ryashchenko 1948; Makovkin 1999). More recently in Ireland Harrington has experimentally produced sika and red deer hybrids under controlled research conditions (Harrington 1973, 1974, 1982). In Europe various breeders intentionally crossed red and sika deer for trophy improvement of sika as well, but have rarely documented or published this (Niethammer 1963; Whitehead 1995; Eick 1995e). Instead it has been frequently better kept in secret. My experience illustrates this approach. Seeking material for our study on hybridization in the early 1980s, I received photos of F1 hybrids between sika and red deer from a small enclosure in Germany. Later the hybrid male was sold to another estate as a “good trophy sika stag” (Eick, 1985). Most of the published records on hybridization describe the hybrids as a natural curiosity caused by a break of the geographical isolation of both initial species. Nevertheless, red deer (Manchurian wapiti) and sika deer are in natural contact along the Ussuri River on the Russia-China border. Hybridization is a common occurrence in that region (Flerov 1952; Sokolov 1959; Heptner et al. 1961). Chinese hunters have a special name for it, chin-da-guiza, and are fully aware that it is a hybrid (Mirolyubov 1949). The first European explorers such as Maak and Przewalski who saw the hybrid even described it as a subspecies of sika deer (Maak 1859; Przhevalski 1870). In Europe, in mixed populations a sika stag occasionally joins a red deer harem during the rut. Under such circumstances, red deer males usually ignore the presence of the sika stag and only chase away other red stags. They will even ignore sika stag attempts to mate a hind in the harem and produce hybrids (Bartoš and Žirovnický 1982; Bartoš 2008). In Europe, hybrids were recorded as far back as 1884, when Powerscourt (1884) reported that, “The Japanese deer here have undoubtedly interbred with the red deer; there are three to four deer in the park here which are certainly hybrids, the red hind in each case being the dam.” Since than hybridization has been widely recorded from many countries such as Ireland (Harrington 1973, 1974, 1982; Herzog 1987; Herzog and Harrington 1991; Herzog and Krabel 1993; Herzog and Krabel 1993; Herzog and Herzog 1995; Nagata et al. 1998; Gehle and Herzog 1998; Nagata et al. 1998; Gehle and Herzog 1998); Britain (Millais 1897; Brooke 1898; F.W.B. 1902; Lydekker 1915; Whitehead 1950, 1964; Delap 1968; Whitehead 1972; Lowe and Gardiner 1974, 1975; Ratcliffe 1987; Hunt 1987; Whitehead 1988; MacNally 1988; Abernathy 1994; Goodman et al. 1999); Czech Republic (Bartoš and Žirovnický 1981, 1982; Bartoš et al. 1981; Bartoš and Vítek 1993); Germany (Rocholl 1967; Herzog 1987; Gehle et al. 1998); Lithuania (Baleisis et al. 2002); the former Soviet Union (Menard 1930; Sarkisov 1944; Mirolyubov and

586

L. Bartoš

Ryashchenko 1948; Mirolyubov 1949; Flerov 1950, 1952; Sokolov 1959; Heptner et al. 1961; Salganskiy et al. 1963; Steklenyev 1986; Prisyazhnyuk and Tchegorko1990; Makovkin 1999); China (Yu 1986); New Zealand (Davidson 1973, 1987; Challies 1985); and elsewhere. In New Zealand hybrid zones have also developed between North American wapiti (Cervus elaphus nelsoni) and sika deer (Nugent et al. 1987). Only occasionally genetic analysis may have shown there is not yet large-scale in situ hybridization occurring between the sika and red deer feral populations (Diaz et al. 2006). Despite the massive evidence, a high proportion of European hunters still do not seem at all convinced of the possibility of this interspecific hybridization. In hunting magazines in countries with sika in their territory, we can see recent articles repeatedly doubting any possibility of the cross. In many areas where sika coexist with red deer, it is obstinately believed that no interbreeding has ever occurred (e.g., Bennetsen 1977; Eick 1995d). Where a detailed investigation was made, evidence appeared that hybridization in that particular region did occur, sometimes to an extent of progressed introgression between the two species (Bartoš 1991). This happened for example in Great Britain (Lowe and Gardiner 1975; Abernathy 1994; Goodman et al. 1999), in Ireland (Harrington 1973, 1974, 1982), and in the Czech Republic (Bartoš et al. 1981). For decades it had been believed that sika in the largest German population at Möhnesee has never crossed with sympatric red deer (Eick 1995d), until Gehle and Herzog (1998) showed evidence of the opposite. Even in Ireland, where a hybrid population at Wicklow is famous and well-documented (Harrington 1973, 1974; Herzog and Harrington 1989; Herzog and Harrington 1991; Herzog and Herzog 1995), in other Irish areas such as the Killarney National Park with sympatric sika and red deer, it is still believed that “there is no evidence to date to suggest that hybridization has occurred in Killarney” (O’Donoghue 1991). In 1994, we visited Killarney National Park for other purposes and observed various hybrid signs especially in the local red deer population. We even videorecorded a young red deer stag flicking his extraordinary elongated tail in a way characteristic for sika. Still, local rangers argued there is no sign of hybridization. Disbelieve in the existence of interspecific hybridization, so widespread across continental Europe, is perhaps because only F1 hybrids look really suspicious. According to European hunters’ tradition of performing a “selective harvest,” such F1 hybrid males roaming with red deer may be culled as “poor red deer,” without any particular interest of the hunter in unusual phenotype traits other than smaller body size and possibly “poor antlers.” F1 hybrid females integrated in red deer herds may be culled as well, but traditional lower interest in females may cause them to survive unnoticed. Also, extra-nuclear heredity, called a maternal effect, may have been involved (Bartoš and Žirovnický 1981). As known from reciprocal crosses of farm animals remarkably different in size (Walton and Hammond 1938 and others), hybrids are often closer in size to the mother’s dimension, thus reducing the difference in appearance of the hybrid from its mother. In addition, heterosis effect (Ashby 1937) may also be involved. It may further decrease the body size difference if the mother of the calf is a red deer (Bartoš 2008 unpublished). If the hybrid comes from a sika mother, such an individual has good chances for survival.

39 Sika Deer in Continental Europe

587

For European hunters large body size and unusually larger and abundantly branched antlers of such a “sika” individual are signs of “good genes” worthy of protection. In the F2 generation and later, these hybrids are not necessarily recognizable according to their phenotype (Harrington 1974; Bartoš and Žirovnický 1981, 1993). Moreover, some of the traits of red deer designated of hybrid origin with sika deer (Bartoš and Žirovnický 1981) may be duplicated from other species or subspecies, e.g., wapiti, maral, etc., the species imported to Europe for centuries and contaminating the gene pool of the original populations (Powerscourt 1884; Benirschke 1967; Gray 1971; Herzog and Harrington 1991). Another firmly held tradition in Europe is that an unusual phenotype and/or antler appearance of a “sika deer” has come from previous crosses between various sika subspecies. Concurrently, no attempts have usually been made to ask how the result of such crosses should look. Earlier descriptions of large Cervus nippon hortulorum refer to only eight-point antlers, with more points being an exception (Flerov 1950, 1952; Sokolov 1959; Heptner et al. 1961; Prisyazhnyuk 1972). During the period between 1936 and 1949, Bromley found (1956) only two exceptions of antlers having more than eight points in the Tsudsukhinskiy Zapovednik (Tsudsukhinsk Reserve). In Mordovskiy Zapovednik (Mordovsk Reserve) only one antler with five and two antlers with six points out of 53 was found (Shtarev 1966). In contrast, Eick (1995e) cited his correspondence with H. Heck, who confirmed that since the beginning of sika introductions, over eight-pointed antlers were exceptional in Japanese sika, whereas with Dybowski stags 10- and 12-pointers were not at all rare. Moreover, the sika populations in the former USSR have changed over time. Antlers with five points on a single antler could sometimes be found in the introduced population of Tchernomorskiy Zapovednik (Black Sea Reserve) (Verestiennikov 1968). Even in the natural range of sika deer, Lazov district, Primorsky Krai in the Far East, we can follow gradual changes in antler appearance. Despite the earlier descriptions of eight-point antlers as the sika deer standard (Flerov 1950, 1952; Sokolov 1959; Heptner et al. 1961), Prisyazhnyuk (1971) analyzed 40 pairs and 97 single antlers and recognized 35 antlers with various anomalies including those bearing numerous points. Twenty per cent of these showed extra points in distal and/or top points. In another study from the same area, Prisyazhnyuk (1972) found 12% of antlers bearing more than eight points, the maximum being 13 points. The problem is, however, that in South Primor’ye (Primorsky Krai), there are numerous farms containing sika and Manchurian deer together (Prisyazhnyuk 1972). Occasional escapes from the farms are unavoidable. Even the seven or more kilometers separating the Isle of Askold in the Sea of Japan from the Primorsk mainland is not a barrier for the deer (Prisyazhnyuk 1978, personal communication). Thus, changes in the number of antler points in sika deer may result from interspecific hybridization. Alternatively, the status of sika deer subspecies on mainland Asia is a matter of debate, with some authors suggesting that the larger “hortuloid” subspecies are principally of hybrid origin (e.g., Lowe and Gardiner 1975). More recent studies dealing with karyotypes would support the suggestion about the hybrid origin of the “hortuloid” subspecies (Bartoš and Žirovnický 1981; Herzog 1987, 1995; Herzog and Harrington 1991).

588

L. Bartoš

Larger antlers with surplus points may be also indicate previous hybridization with red deer for the smaller “Japanese” sika deer. In the Czech Republic, in the late1970s, sika deer in the Plzenˇ North population still differed according to their origin more than 30 years after the deer established themselves in the wild (Vavruneˇk 1978, personal communication). Those coming from the Cˇemíny enclosure had strong antlers with a maximum of eight points, while those coming from the Lipí deer park had longer antlers tending to have more tines. The difference between the two enclosures was that in Cˇ emíny sika shared the enclosure with mouflon and roe deer, but no red deer. In contrast, in Lipí there were red deer living together with sika (Doležal 1960; Wolf and Vavruneˇk 1975–1976; Vavruneˇk and Wolf 1977). Many other signs of previous or recent hybridization in the population have been documented (Bartoš and Žirovnický 1981, 1982; Bartoš and Vítek 1993). The disregard of scientific evidence characteristic of hunters’ approach across many European countries has led to the fact that introgression of sika and red deer gene pools has been occurring in many areas without receiving any attention.

Conclusion Sika have established themselves in the European fauna and seems to be a stabilized component. With its high potential to compete with autochthonous species and readiness to hybridize with native red deer, the sika deer has represented a real threat to original European deer populations for over a century. European hunters’ interests may be the main factor leading to ignoring the threat of sika deer presence. Warnings about the danger have not been taken too seriously. The first discoveries of the danger of sika red deer hybridization on a scale of whole populations were reported around 30 years ago (Harrington 1973, 1974; Lowe and Gardiner 1975; Bartoš et al. 1981; Bartoš and Žirovnický 1981, 1982). Conclusions of these earlier studies have been fully supported by recently available genetic techniques (Herzog and Harrington 1991; Abernathy 1994; Herzog and Herzog 1995; Goodman et al. 1999). Still, there is no general wildlife management strategy in continental Europe to rectify the situation. In some European countries, such as the Czech Republic, harvest tends to increase (according to published statistics, harvested numbers have been almost doubled over the last seven years). On a national level, however, the reports on total numbers suggest an increase in sika population rather than reduction (Table 39.1), despite the fact we could question the reliability of the estimates of free-ranging populations used in official statistics. Acknowledgements Dedicated to the memory of my late colleague and friend, Jirˇí Vavruneˇk. Personal communications of various information from Jadvyga Grikieniene, Ejvind B. Herlevsen, Tony G. Armitage, Paul Audenaerde, Hanspeter Daenzer, Stanislav Hekele, Stanislav Žbánek and others are highly acknowledged. I greatly appreciate the comments and suggestions made by Ernst Eick, an Honorary Chairman of the International Sika Society, and Jan Pluháček. Completing the chapter was supported by grants from the Czech Ministry of Agriculture (MZE0002701402) and the Czech Science Foundation. (523/08/0808).

39 Sika Deer in Continental Europe

589

Literature Cited Abernathy, K. 1994. The establishment of a hybrid zone between red and sika deer (genus Cervus). Molecular Ecology 3:551–562. Armstrong, W. E. 1980. The influence of exotic Artiodactyls on white-tailed deer performance and survival. Pages 1–14 in D.E. Harmel, Principal Investigator, Performance Report Job No. 20. Federal Aid Project No. W-109-R-3, Texas Parks and Wildlife Department, Austin, Texas, USA. Ashby, E. 1937. The physiology of heterosis. American Naturalist 71:5141–520. Audenaerde, P. M. F. 2002. European view on the deer farming industry of the world. Pages 73–76 in J. B. Wood, editor, Proceedings of the 2002 North American Deer Farmers Association Annual Conference and The World Deer Farming Congress III. North American Deer Farmers Association, Austin, Texas, USA. Babicˇka, C., M. Drábek, J. Štika, and J. Ženožicˇka. 1977. Poznatky z chovu jelena siky. Myslivost 11:270–271. (In Czech.) Baleisis, R., P. Bluzma, and L. Balciauskas. 2002. Lietuvos kanopiniai zverys. Akstis. (Lithuania.) Bartoš, L. 1982. Reproductive and social aspects of the behaviour of ‘white’ red deer. Säugetierkundliche Mitteilungen 30:89–117. Bartoš, L. 1991. Sika/red deer hybridization — Recognition, consequences and present status. Pages 191–195 in N. Maruyama, B. Bobek, Y. Ono, W. Regelin, L. Bartoš and P. R. Ratcliffe, editors. Wildlife conservation: Present trends and perspectives for the 21st century. Japan Wildlife Research Center, Yushima, Bunkyo-ko, Tokyo, Japan. Bartoš, L., and J. Žirovnický. 1981. Hybridization between red and sika deer. II. Phenotype analysis. Zoologischer Anzeiger 207:271–287. Bartoš, L., and J. Žirovnický. 1982. Hybridization between red and sika deer. III. Interspecific behaviour. Zoologischer Anzeiger 208:20–36. Bartoš, L., and M. Vítek. 1993. Cluster analysis of red and sika deer phenotypes. Pages 15–21 in N. Ohtaishi and H.-I. Sheng, editors. Deer of China: Biology and management. Elsevier, Amsterdam, The Netherlands. Bartoš, L., and J. Šiler. 1994. Game farming in Europe. Pages 23–48 in J. Boyazoglu and L. Hetényi, editors. Game farming in Europe. Food and Agriculture Organization of the United Nations, Rome, Italy. Bartoš, L., J. Hyánek, and J. Žirovnický. 1981. Hybridization between red and sika deer. I. Craniological analysis. Zoologischer Anzeiger 207:260–270. Bartoš, L., P. Šustr, P. Janovský, and J. Bertagnoli. 2003. Sika deer (Cervus nippon) lekking in a free-ranging population in Northern Austria. Folia Zoologica 52:1–10. Baskin, L., and K. Danell. 2003. Ecology of ungulates: A handbook of species in Eastern Europe and Northern and Central Asia. Springer, Berlin and Heidelberg, Germany. Baškyte, R., P. Mierauskas, J. Virbickas, P. Kavaliauskas, S. Paltanavicˇius, M. Alakevicˇius, J. Balevicˇiene, L. Balcˇiauskas, A. Brukas, A. Gutkauskas, P. Kurlavicˇius, R. Pakalnis, V. Urbonas, S. Oleninas, P. M. Rogers, P. Brylski, and M. Dumbliauskiene. 1997. Republic of Lithuania: Biodiversity conservation: Strategy and action plan. Environmental Protection Ministry of the Republic of Lithuania, Vilnius, Lithuania. Benirschke, K. 1967. Sterility and fertility of interspecific mammalian hybrids. Pages 218–234 in K. Benirschke, editor. Comparative aspects of reproductive failure. Springer, New York, New York, USA. Bennetsen, E. 1977. Some facts about sika deer in Denmark. Deer 4:153–157. Bromley, G. F. 1956. Ekologiya dikogo pyatnistogo olenya. Pages 148–215 in Sbornik materialobv po rezul’tatam izutcheniya mlekopitayushtchikh v gosudarstvennykh zapovednikakh. Izdatel’stvo Ministeria sel’skogo xozyaystva (Ministry of Agriculture Publishers, Moscow, USSR. (In Russian.) Brooke, D. 1898. Hybrid red deer. The Field 92–182.

590

L. Bartoš

Butts, G. L., M. J. Anderegg, W. E. Armstrong, D. E. Harmel, C. W. Ramsey, and S. H. Sorola. 1982. Food habits of five exotic ungulates on Kerr Wildlife Management area, Texas. Technical Series No. 30. Texas Parks and Wildlife Department, Austin, Texas. Cadman, A. 1980. Roe in the New Forest. Deer 5:51. Cailmail, F. 1988. Frankreich: Forêt de la Harth, Haut-Rhin, Gatter Salzlecke. Pages 5.2-F-2/1–5 in E. Eick, J. A. Willett, R. König and K.-H. Schulze-Schwefe, editors. Sika, Cervus nippon Temminck, 1838. Internationale Arbeitsgemeinschaft Sikawild, Möhnesee-Körbecke, Germany. Challies, C. N. 1985. Establishment, control, and commercial exploitation of wild deer in New Zealand. Pages 23–36 in P. F. Fennessy and K. R. Drew, editors, Biology of deer production. The Royal Society of New Zealand Bulletin 22. Wellington, New Zealand. Danilkin, A. A. 1996. Siberian and European roe deer. Chapman & Hall, London, United Kingdom. Davidson, M. M. 1973. Characteristics, liberation, and dispersal of sika deer (Cervus nippon) in New Zealand. New Zealand Journal of Forestry 3:153–180. Davidson, M. M. 1987. Sika deer. Pages 468–477 in C. King, editor. The handbook of New Zealand mammals. Oxford University Press, Auckland, New Zealand. Delap, P. 1968. Observations on deer in North/West England. Journal of Zoology 156:531–533. Diaz, A., S. Hughes, R. Putman, R. Mogg, and J. M. Bond. 2006. A genetic study of sika (Cervus nippon) in the New Forest and in the Purbeck region, southern England: Is there evidence of recent or past hybridization with red deer (Cervus elaphus)? Journal of Zoology 270:227–235. Doležal, V. 1960. Z Maneˇtínska. Myslivost 4:174. (In Czech.) Eick, E. 1988. Bundesrepublik Deutschland. Möhnesee. Pages 5.2-D-3/0 in E. Eick, J. A. Willett, R. König and K.-H. Schulze-Schwefe, editors. Sika, Cervus nippon Temminck, 1838. Internationale Arbeitsgemeinschaft Sikawild, Möhnesee-Körbecke, Germany. Eick, E. 1995a. Austria. Pages 30.1–30.2 in E. Eick, R. König and J. Willett, editors. Sika, Cervus nippon Temminck, 1838. Volume I. Second Edition. International Sika Society, Möhnesee, Germany. Eick, E. 1995b. Commonwealth of Independent States. Pages 3.1–3.10 in E. Eick, R. König and J. Willett, editors. Sika, Cervus nippon Temminck, 1838. Volume I. Second Edition. International Sika Society, Möhnesee, Germany. Eick, E. 1995c. France. Pages 2.3-F/1–3 in E. Eick, R. König and J. Willett, editors. Sika, Cervus nippon Temminck, 1838. Volume I. Second Edition. International Sika Society, Möhnesee, Germany. Eick, E. 1995d. Germany. Pages 2.2-D/1–13 in E. Eick, R. König and J. Willett, editors. Sika, Cervus nippon Temminck, 1838. Volume I. Second Edition. International Sika Society, Möhnesee, Germany. Eick, E. 1995e. A history of naturalisation. Pages 9.1–9.14 in E. Eick, R. König and J. Willett, editors. Sika, Cervus nippon Temminck, 1838. Volume I. Second Edition. International Sika Society, Möhnesee, Germany. Eick, E. 1995f. Poland. Pages 18.1–18.5 in E. Eick, R. König and J. Willett, editors. Sika, Cervus nippon Temminck, 1838. Volume I. Second Edition. International Sika Society, Möhnesee, Germany. Eick, E., R. König, and J. Willett, editors. 1995. Sika, Cervus nippon Temminck, 1838. Volume I. Second Edition. International Sika Society, Möhnesee, Germany. F.W.B. 1902. Hybrid deer at Powerscourt. The Field 100–1012. Feldhamer, G. A., and W. E. Armstrong. 1993. Interspecific competition between four exotic species and native Artiodactyls in the United States. Transactions of the 58th North American Wildlife and Natural Resources Conference 468–478. Wildlife Management Institute, Washington, DC, USA. Flerov, K. K. 1950. Morfologiya i ekologiya olenoobraznikh v protsese ich evolyutsii. Materialy po Tshetvertitshnom Periodu SSSR 2:50–69. (In Russian.) Flerov, K. K. 1952. Fauna SSSR. Tom 1 (2) Kabargi i oleni. Izdatelstvo Akademii Nauk SSSR, Moskva (Academy of Science of the USSR Publishers), Leningrad, USSR. (In Russian.)

39 Sika Deer in Continental Europe

591

Gehle, T., and S. Herzog. 1998. Is there evidence for hybridisation between red deer and sika deer in Germany? Pages 121–123 in Z. Zomborszky, editor. 4th International Deer Biology Congress, June 30-July 4, 1998, Final Program - Abstracts. Pannon Agricultural University, Faculty of Animal Science, Kaposvar, Hungary. Goodman, S. J., N. H. Barton, G. Swanson, K. Abernethy, and J. M. Pemberton. 1999. Introgression through rare hybridization: A genetic study of a hybrid zone between red and sika deer (Genus Cervus) in Argyll, Scotland. Genetics 152:355–371. Goodman, S. J., H. B. Tamate, R. Wilson, J. Nagata, S. Tatsuzawa, G. M. Swanson, J. M. Pemberton, and D. R. McCullough. 2001. Bottlenecks, drift and differentiation: The population structure and demographic history of sika deer (Cervus nippon) in the Japanese archipelago. Molecular Ecology 10: 1357–1370. Gray, A. G. 1971. Mammalian hybrids. A check-list with bibliography. Technical Communication No. 10 (Revised) of the Commonwealth Bureau of Animal Breeding and Genetics, Edinburgh, United Kingdom. Hake, F. 1988. Bundesrepublik Deutschland: Weserbergland. Pages 5.2-D-6/1–9 in E. Eick, J. A. Willett, R. König and K.-H. Schulze-Schwefe, editors. Sika, Cervus nippon Temminck, 1838. Internationale Arbeitsgemeinschaft Sikawild, Möhnesee-Körbecke, Germany. Harmel, D. E., and W. E. Armstrong. 1987. The influence of exotic Artiodactyls on white-tailed deer production and survival. Pages 1–11, Performance report - Federal Aid Project No. W109-R-10. Texas Parks and Wildlife Department, Austin, Texas, USA. Harrington, R. 1973. Hybridization among deer and its implications for conservation. Irish Forestry Journal 30:64–78. Harrington, R. 1974. The hybridization of red and sika deer in Northern Ireland. Irish Forestry Journal 31:2–3. Harrington, R. 1982. The hybridization of red deer (Cervus elaphus L. 1758) and Japanese sika deer (C. nippon Temminck 1838). Pages 559–571 in F. O’Gorman and J. Rochford, editors. Transactions XIVth International Congress of Game Biologists. Irish Wildlife Publications for the Organising Committee of the XIVth Congress, Dublin, Ireland. Heptner, V. G., A. A. Nasimovitch, and A. G. Banikov. 1961. Mlekopitayushtchie Sovetskogo soyuza. Tom pervyy. Parnokopytnye i neparnokopytnye. Gosudarstvennoe izdatel’stvo ‘Vysshaya shkola’, Moskva. (In Russian.) Herzog, S. 1987. Mechanisms of karyotype evolution in Cervus nippon Temminck. Caryologia 40:347–353. Herzog, S. 1995. Hybridization and the species concept: Implications for wildlife management strategies. Pages 3.4:15–20 in E. Eick, R. König and J. Willett, editors. Sika, Cervus nippon Temminck, 1838. Volume II. Second Edition. International Sika Society, Möhnesee, Germany. Herzog, S., and R. Harrington. 1989. Cytogenetic markers of sika deer (Cervus nippon) introgression into red deer (Cervus elaphus) populations. Pages 47–49 in 2. Symposium zur Planung eines CIC-Rothirschgenetikprojektes. Forschungsinstitut für Wildtierkunde, Veterinärmedizinische Universität Wien, Vienna, Austria. Herzog, S., and R. Harrington. 1991. The role of hybridization in the karyotype evolution of deer (Cervidae, Artiodactyla, Mammalia). Theoretical and Applied Genetics 82:425–429. Herzog, S., and A. Herzog. 1995. Cytogenetic and biochemical-genetic studies on hybridization betwen red deer and sika deer. Pages 58.1–58.6 in E. Eick, R. König and J. Willett, editors. Sika, Cervus nippon Temminck, 1838. Volume II. Second Edition. International Sika Society, Möhnesee, Germany. Herzog, S., and D. Krabel. 1993. Haemoglobin variants within the genus Cervus. Small Ruminant Research 11:187–192. Hofmann, R. R. 1988. Morphological classification of sika deer within the comparative, morphophysiological system of ruminant feeding types. Pages 6.2/1–8 in E. Eick, J. A. Willett, R. König and K.-H. Schulze-Schwefe, editors. Sika, Cervus nippon Temminck, 1838. Internationale Arbeitsgemeinschaft Sikawild, Möhnesee-Körbecke, Germany. Hofmann, R. R. 1989. Evolutionary steps of ecophysiological adaptation and diversification of ruminants: A comparative view of their digestive system. Oecologia 78:443–457.

592

L. Bartoš

Hošek, E. 1982. O zvˇe rˇ i a myslivosti na panstvích Zˇádlovice a Dubravice. Severní Morava 44:31–35. (In Czech.) Hunt, E. 1987. Hybridization between red deer (Cervus elaphus L.) and sika deer (Cervus nippon) with particular reference to Argyll, North Scotland and the New Forest. B.Sc. thesis, University of Southampton, Southampton, United Kingdom. Keiper, R. R. 1985. Are sika deer responsible for the decline of white-tailed deer on Assateague Island, Maryland? Wildlife Society Bulletin 13:144–146. Kistler, R. 1995. Studies on the problems between forest and game in the canton of Schaffhausen Switzerland. Pages 66.1–66.10 in E. Eick, R. König and J. Willett, editors. Sika, Cervus nippon Temminck, 1838. Volume II. Second Edition. International Sika Society, Möhnesee, Germany. Kokeš, O. 1970. Asijští jeleni na území Cˇeskoslovenska. Ochrana Fauny 4:158–161. (In Czech.) Komárek, J. 1945. Pages 218–219 in Lesnická zoologie III. Prague, Czechoslovakia. Latham, J. 1999. Interspecific interactions of ungulates in European forests: An overview. Forest Ecology and Management 120:13–21. Legrand, B. 1988. Frankreich, Centre de Rambouillet. Pages 5.2-F-1/1–2 in E. Eick, J. A. Willett, R. König and K.-H. Schulze-Schwefe, editors. Sika Cervus nippon Temminck, 1838. Internationale Arbeitsgemeinschaft Sikawild, Möhnesee-Körbecke, Germany. Lowe, V. P. W., and A. S. Gardiner. 1974. A re-examination of the subspecies of red deer (Cervus elaphus) with particular reference to the stocks in Britain. Journal of Zoology 174:185–201. Lowe, V. P. W., and A. S. Gardiner. 1975. Hybridization between red deer (Cervus elaphus) and sika deer (Cervus nippon) with particular reference to stocks in N.W. England. Journal of Zoology 177:553–556. Lydekker, R. 1915. Catalogue of the ungulate mammals in the British Museum (Natural History). Vol. IV. British Museum, London, United Kingdom. Maak, R. K. 1859. Puteshestvie po Amuru, sovershenoe v 1855 godu. Saint-Petersburg. (In Russian.) MacNally, L. 1988. Great Britain: Loch Ness, Invernesshire, Scotland. Pages 5.2-GB-3/1–7 in E. Eick, J. A. Willett, R. König and K.-H. Schulze-Schwefe, editors. Sika Cervus nippon Temminck, 1838. Internationale Arbeitsgemeinschaft Sikawild, Möhnesee-Körbecke, Germany. Makovkin, L. I. 1999. The sika deer of Lazovsky Reserve and surrounding areas of the Russian Far East. Almanac “Russki Ostrov,” Vladivostok, Russia. Malakauskas, M., and J. Grikieniene. 2002. Sarcocystis infection in wild ungulates in Lithuania. Acta Zoologica Lituanica 12:372–380. Markovic, L. 1988. Fehérvárcsurgó, Ungarn. Pages 5.2-H/1–4 in E. Eick, J. A. Willett, R. König and K.-H. Schulze-Schwefe, editors. Sika, Cervus nippon Temminck, 1838. Internationale Arbeitsgemeinschaft Sikawild, Möhnesee-Körbecke, Germany. Matuszewski, G. 1988. Polen. Pages 5.2-PL/1–9 in E. Eick, J. A. Willett, R. König and K.-H. Schulze-Schwefe, editors. Sika, Cervus nippon Temminck, 1838. Internationale Arbeitsgemeinschaft Sikawild, Möhnesee-Körbecke, Germany. Matuszewski, G., and P. Sumin´ski. 1984. Sika deer in Poland. Deer 6:74–75. Menard, G. A. 1930. Pantovoe olenevodstvo. Gostorizdat, Moskva-Leningrad. (In Russian.) Millais, J. G. 1897. British deer and their horns. Henry Southeran & Company, London, United Kingdom. Mirolyubov, I. I. 1936. Biologiya odomashnenogo pyatnistogo olenya. Vestnik Dalnevostotchnogo Filiala Akademii Nauk SSSR 16:155–174. (In Russian.) Mirolyubov, I. I. 1949. Gibridizatsiya pyatnistogo olenya s izyubrem. Karakulevodstvo i Zverovodstvo 1:74–75. (In Russian.) Mirolyubov, I. I., and L. P. Ryashchenko. 1948. Pyatnistyi olen (spotted deer) Vladivostok. Pages 153 in G. F. Bromley, editor. Ecology of the wild spotted deer in the Maritime Territory. Ministry of Agriculture U.S.S.R., Moscow, USSR. Nagata, J., R. Masuda, K. Kaji, M. Kaneko, and M. C. Yoshida. 1998. Genetic variation and population structure of Japanese sika deer (Cervus nippon) in Hokkaido Island. Pages 125 in Z.

39 Sika Deer in Continental Europe

593

Zomborszky, editor. 4th International Deer Biology Congress, June 30-July 4, 1998, Final Program - Abstracts. Pannon Agricultural University, Faculty of Animal Science, Kaposvar, Hungary. Niethammer, G. 1963. Die Einbürgerung von Säugetieren in Europa. Paul Parey, Hamburg and Berlin, Germany. (In German.) Nugent, G., J. P. Parkes, and K. G. Tustin. 1987. Changes in the intensity and distribution of red deer and wapiti in northern Fiordland. New Zealand Journal of Ecology 10:11–21. O’Donoghue, Y. A. 1991. Growth, reproduction and survival in a feral population of Japanese sika deer (Cervus nippon nippon Temminck, 1836). Ph.D. thesis, Department of Zoology, University College, Dublin, Ireland. Opluštil, S. 1980. Výskyt siky v Severomoravském kraji. Myslivost 1:11. (In Czech.) Pavlov, M. P., I. V. Korsakova, and N. P. Lavrov. 1974. Aklimatizatsiya okhotnitch’e promyslovykch zverei i ptits v SSSR. Volgo-Vyatskoe Knizhnoe Izdatiel’stvo, Kirov. (In Russian.) Powerscourt, V. 1884. On the acclimatization of the Japanese deer at Powerscourt. Proceedings of the Zoological Society London 207–209. Prisyazhnyuk, V. E. 1971. Slutchayi asimetrii, nenormalnogo stroyeniya i poverezhdyeniya rogov dikogo pyatnistogo olenya (Cervus nippon T.). Zoologichesky Zhurnal 50:1380–1387. (In Russian.) Prisyazhnyuk, V. E. 1972. Vozrastnaya i individualnaya izmentchivost rogov pyatnistykch olenyey Primor’ya. Trudy Moskovskogo Obshtchestva Ispytatyeley Prirody XLVIII:135–149. (In Russian.) Prisyazhnyuk, V. E., and P. T. Tchegorko. 1990. Gibridizatsiya blagarodnogo i pyatnistogo oleney v Moldavii. V Syezd Vsesoyuzhnogo Teriologitcheskogo Obshtchestva AN SSSR 3:169–171. (In Russian.) Przhevalski, N. M. 1870. Puteshestvie v Ussuriiskom krae v 1868–1869. Saint Petersburg. (In Russian.) Putman, R. J. 1996. Competition and resource partitioning in temperate ungulate assemblies. Chapman & Hall, London, United Kingdom. Ratcliffe, P. R. 1987. Distribution and current status of sika deer, Cervus nippon, in Great Britain. Mammal Review 17:39–58. Reinken, G. 1998. Landwirtschaftliche Hirschhaltung - eine Alternative zur umweltfreundlichen Grünlandnutzung. Zeitschrift für Jagdwissenschaft 44:78–84. (In German.) Richardson, M. L., and S. Demarais. 1992. Parasites and condition of coexisting populations of white-tailed and exotic deer in south-central Texas. Journal of Wildlife Diseases 28:485–489. Rocholl, W. 1967. Gebt ‘grünes Licht’ fur Sika-Wild. Wild und Hund 134–137. (In German.) Rowland, R. 1967. A history of the deer at Beanlien Hampshire. Deer 1:123–126. Rumohr-Rundhof, W.-H. v. 1988. Bundesrepublik Deutschland. Ostangeln/Schwansen. Pages 5.2-D-4/1–9 in E. Eick, J. A. Willett, R. König and K.-H. Schulze-Schwefe, editors. Sika Cervus nippon Temminck, 1838. Internationale Arbeitsgemeinschaft Sikawild, MöhneseeKörbecke, Germany. Salganskiy, A. A., I. S. Sles, V. D. Treus, and G. A. Uspenskiy. 1963. Zoopark ‘Askaniya Nova’ (Opyt akklimatizatsii dikikh kopytnykh i strausov). Gosozdat S.-Kh. Literatury USSR, Kiev. (In Russian.) Sarkisov, A. A. 1944. Trudy Jerevanskogo. Zoologitcheskogo Parka 1–2:91–97. (In Russian.) Schoenberger, H. U., H. Matzinger, and G. Schwyn. 1988. Schweiz: Südranden. Pages 5.2CH/1–11 in E. Eick, J. A. Willett, R. König and K.-H. Schulze-Schwefe, editors. Sika Cervus nippon Temminck, 1838. Internationale Arbeitsgemeinschaft Sikawild, Möhnesee-Körbecke, Germany. Shtarev, Y. F. 1966. Rezul’taty akklimatizatsii pyatnistogo olenya v Mordovskoy ASSR. Trudy Mordovskogo zapovednika (Proceedings of Mordovian State Natural Reserve) 3:64–70. (In Russian.) Sokolov, I. I. 1959. Fauna SSSR. Mlekopitayushtchie. Kopytnye zveri (Otryad Perissodactyla i Artiodactyla). Izdatel’stvo Akademii nauk SSSR, Moskva, Leningrad. (In Russian.) Steklenyev, E. P. 1986. Mezhvidovaya gibridizatsiya blagarodnogo (Cervus elaphus L.) i pyatnistogo olenya (Cervus nippon hortulorum Temm.). Citologiya i Genetika 20:138–142. (In Russian.)

594

L. Bartoš

Swinhoe, R. 1864. Letters in. Proceedings of the Royal Society of London 168–169. Takatsuki, S. 1988. The weight contributions of stomach compartments of sika deer. Journal of Wildlife Management 52:313–316. Ueckermann, E. S. J. 1988. Bundesrepublik Deutschland. Schlitz. Pages 5.2-D-5/0 in E. Eick, J. A. Willett, R. König and K.-H. Schulze-Schwefe, editors. Sika Cervus nippon Temminck, 1838. Internationale Arbeitsgemeinschaft Sikawild, Möhnesee-Körbecke. Uloth, W. 1984. Der Sikahirsch in der UdSSR. Unsere Jagd 34:377. (In German.) Vavruneˇk, J., and R. Wolf. 1977. Chov jelení zvˇe ˇre v Západocˇeském kraji. Sborník veˇdeckého lesnického ústavu VŠZ v Praze 20:97–115. (In Czech.) Verestiennikov, D. S. 1968. Olen’ pyatnistiy v Tchernomorskom zapovednike. Vestnik Zoologii 1:30–36. (In Russian.) Walton, H., and J. Hammond. 1938. The maternal effects on growth and conformation in Shire horse-Shetland pony crosses. Proceedings of the Royal Society of London Series B - Biological Sciences 125:311–335. Whitehead, G. K. 1950. Deer and their management in the deer parks of Great Britain and Ireland. Country Life Ltd., London, United Kingdom. Whitehead, G. K. 1964. The deer of Great Britain and Ireland. Routledge & Kegan Paul, London, United Kingdom. Whitehead, G. K. 1972. Deer of the world. Constable, London, United Kingdom. Whitehead, G. K. 1988. British Isles. Pages 5.2-GB/1–11 in E. Eick, J. A. Willett, R. König and K.-H. Schulze-Schwefe, editors. Sika Cervus nippon Temminck, 1838. Internationale Arbeitsgemeinschaft Sikawild, Möhnesee-Körbecke, Germany. Whitehead, G. K. 1995. British isles. Pages 22.1–22.9 in E. Eick, R. König and J. Willett, editors. Sika, Cervus nippon Temminck, 1838. Volume I. Second Edition. International Sika Society, Möhnesee, Germany. Wicki, D. 1988. Bundesrepublik Deutschland. Hochrhein. Pages 5.2-D-1/0 in E. Eick, J. A. Willett, R. König and K.-H. Schulze-Schwefe, editors. Sika Cervus nippon Temminck, 1838. Internationale Arbeitsgemeinschaft Sikawild, Möhnesee-Körbecke, Germany. Wolf, R., and J. Vavruneˇk. 1975–1976. Sika východní Cervus nippon Temm v Západních Cˇechách. Sborník Veˇdeckého Lesnického Ústavu VŠZ v Praze 18–19:185–199. (In Czech.) Yevtushevskiy, N. N. 1974. Razmnozheniye olenya pyatnistogo (Cervus nippon hortulorum Sw.) v usloviyakch Pridneprovya. Vestnik Zoologii 8:23–28. (In Russian.) Yu, X. 1986. Comparative observations on the karyotype of north-eastern red deer and north-eastern sika deer and karyotype analysis of their hybridized combinations. Acta Genetica Sinica 13:125–131. Zhirnov, L. V., V. A. Bychkov, V. A. Orlov, V. S. Pokrovski, V. E. Prisyazhnyuk, and G. V. Khakhin. 1978. Ussuriyskiy pyatnistyy olen’ (aborigennaya populyatsiya) Cervus nippon hortulorum Swinhoe, 1864. Pages 65–67 in A. G. Bannikov, editor. Red Data Book of USSR, Part I, Mammals. Lesnaya Promyshlennost Publishers, Moscow. (In Russian.)

Chapter 40

Sika Deer in the British Isles Graeme M. Swanson and Rory Putman

Abstract Since their introduction to the British Isles approximately 150 years ago, sika deer have expanded their distributional range and established many large freeranging populations in Scotland, Ireland, and England. As such, they have become a significant resource, or pest, depending on various local, regional, or national objectives. In Ireland, initial concern was their ability to hybridize with local populations of red deer (Cervus elaphus) (e.g., Harrington 1973, 1982). Examples of both complete admixture of species (Wicklow) and strong within-species breeding are found within Ireland. Explanations for these different outcomes have mostly centered on the genetic status of the original founders (Lowe and Gardiner 1975; Harrington 1982) rather than effects of the abundance of each species and differences in ecology that are equally likely to have contributed (Swanson 1999; Putman 2000). More recently, attention in Ireland as elsewhere within the United Kingdom has turned to their damage of habitats and commercial forestry (e.g., Lowe 1994). The largest populations of sika deer within the United Kingdom are currently found in Scotland, partly due to the number and extent of introductions (Ratcliffe 1987a), but also because of the widespread availability of suitable habitat. Populations in Scotland have shown significant increase in number and steady expansion in distribution. As in southern Ireland, management has been focused primarily on concerns about hybridization with red deer and damage to tree crops and habitats. There is difference of opinion about the seriousness of hybridization and also whether it is now perhaps too late to achieve effective management. Culls of sika deer have increased steadily across Scotland recently, but there is little sign that these have affected their abundance or distribution. In England, sika deer have remained essentially local to their introduction locations until recently. Patterns of expansion in Dorset and Hampshire in the south and Lancashire in the north suggest that populations may have to reach a critical density before expanding. Hybridization with red deer has been reported in Lancashire and Dorset/Devon and recent work has established the genetic origin of a number of English as well as Scottish populations.

D. R. McCullough et al. (eds.), Sika Deer: Biology and Management of Native and Introduced Populations, DOI: 10.1007/978-4-431-09429-6_40, © Springer 2009

595

596

G. M. Swanson, R. Putman

In this chapter we review the general ecology of sika deer in the British Isles (habitat use, feeding ecology, population ecology), the impacts of sika deer on habitats and on native deer through competition and hybridization, and the management approaches applied.

History Sika deer were first brought to the British Isles in 1860 when a pair of Japanese and a pair of Manchurian sika were presented to London Zoo (Whitehead 1964). Later that year, six Japanese deer were delivered to Viscount Powerscourt for his deer park at Enniskerry in Ireland (Powerscourt 1884). These animals were subsequently bred and distributed, either directly or indirectly, to parks throughout Ireland, Scotland, and England, and were kept alongside many other deer species (Ratcliffe 1987a). Only two further introductions are recorded to have come directly from Japan; one to Dawyck in the Scottish Borders (southern Scotland near the border with England) and the other (now possibly extinct) to Pixton in Devon, England (Ratcliffe 1987a). However, the origins of some introductions are not recorded; for example, those of the extensive population in the Poole Basin in Dorset and those in the New Forest. Most of the introduced deer remained fenced within park boundaries until around 1910–1920, at which point they were either deliberately released or escaped to form feral populations of varying success. The origin and subspecific status of the sika stocks introduced to the British Isles has generated much interest because of the possible role it may have played in the hybridization of sika with red deer (Lowe and Gardiner 1975; Harrington 1982; Ratcliffe 1987a; Ratcliffe et al. 1992; Abernethy 1994a). Most populations are thought to have originated directly or indirectly (i.e., via Powerscourt) from the southern Japanese archipelago and belong to the subspecies C. n. nippon (Ratcliffe 1987a; Abernethy 1994a). However, mainland Chinese (C. n. hortulorum) and Taiwanese (Formosan) (C. n. taiouanus) sika stocks were known to have been kept at the same time in some park collections, and it is entirely possible that these may have contributed genetically to various park and feral populations (Whitehead 1964; Ratcliffe 1987b; Ratcliffe et al. 1992). Even the population in the Scottish borders, which was documented as having come directly from Japan, may have had introgression of genes from a stag of unknown provenance introduced from Whipsnade Zoo in 1939 (Whitehead 1950). The benefits of using molecular techniques to address such systematic and phylogenetic problems have been well documented (e.g., Avise 1994). Although there are some differences in appearance between British sika populations (e.g., coat color and skull shape), genetic analysis of several populations (including Argyll, Dorset, Fife, and Scottish Borders indicates that all are essentially of Japanese origin (C. n. nippon) and, interestingly, have all originated from the same area around Nagasaki on Kyushu Island, Japan (Swanson 1999; Goodman et al. 2001; Diaz et al. 2006). The molecular markers found in both the Wicklow and Killarney

40 Sika Deer in the British Isles

597

populations in Ireland also show a similar pattern to those found elsewhere in Scotland and England (G. Swanson 1999 unpublished data).

Current Distribution Following their introduction to deer parks and private collections between 1870 and 1930, sika deer subsequently escaped or were released into the surrounding habitat (Whitehead 1964; Ratcliffe 1987a). In Ireland, sika deer occur in the wild in both the northern and southern parts of the island, with major populations in the south in Killarney, centered on the National Park, in Donegal, and around the mountains of Wicklow (Fig. 40.1).

Fig. 40.1 Current distribution of sika deer in the British Isles (after Putman 2000).

598

G. M. Swanson, R. Putman

In Northern Ireland, there are significant populations associated with the old deer park of Colebrooke in County Fermanagh, and others in County Limerick and County Tyrone, probably continuous with populations in Donegal. Smaller populations were established as escapees from deer parks in County Monaghan, and County Down, but the current status of these is unknown. In Scotland, introductions have resulted in well-established populations in Sutherland, based around Shin Forest and Torrachilty, in the Great Glen and Glenmazeran, Argyll and Borders, and smaller separate populations in Fife and Loch Morar (Fig. 40.1). In 1993, the British Deer Society (BDS) conducted a questionnaire survey of deer sightings in order to map distributions of all deer species in Scotland (Rose 1994). This survey was updated in 2001 and provides a picture of sika expansion and distribution to date (Ward 2005). Large population ranges in Scotland are sometimes the result of two populations coalescing, but all extant populations are expanding their range and, presumably, their numbers within that range. Despite their widespread introduction to numerous deer parks in the nineteenth century (see Whitehead 1964), sika deer are not widely established in the wild in England. They are restricted to a relatively small number of localized populations (Fig. 40.1). Large populations are found in the south, within the New Forest and northern Hampshire, in south-east Dorset, within the Poole Basin, in Bowland (Lancashire) and in the Lake District. Smaller populations are found in Northamptonshire and Bedfordshire, with recent colonization of an area around the Oxfordshire/Buckinghamshire/Wiltshire and Gloucestershire borders. Sika deer are found on Lundy Island and on Brownsea Island in the Poole Basin. No populations are yet established in Wales. In addition to these wild, or feral, populations, sika are still widely maintained in deer parks, with an estimated total of about 1,000 Japanese sika and a further 500 Manchurian or Formosan sika maintained in private parks (Harris et al. 1995).

Population Size Although there has been no systematic survey of the British Isles sika population there are local estimates (of varying quality) and some attempts at bringing these together to give an overall approximation (e.g., Harris et al. 1995; Staines 1998). In Scotland, sika have become extremely well-established and widespread, with some 14,000 km2 colonized by the species and an estimated count of around 10,000 individuals. The main centers of population are in Argyll, Inverness-shire, Peeblesshire, Ross and Cromarty, and Sutherland, although sika in Scotland are actively extending their range at the present time (Harris et al. 1995; Abernethy 1998; Livingstone 2001; Pemberton et al. 2006). At current estimates the New Forest population numbers about 150–200 animals. Numbers in Bowland are also estimated at around 200 animals (Harris et al. 1995) and fewer than 100 animals are maintained on Lundy Island. The most extensive population in England is that originating in Dorset; this population, which

40 Sika Deer in the British Isles

599

is now extending its range into parts of Devon, probably numbers less than 2,000 (Harris et al. 1995; Staines 1998).

Local Population Size and Density Local estimates of minimum population size are available in many areas of Scotland through Deer Management Group and Forestry Commission records. Dung counts have been used to estimate sika populations in some areas. Table 40.1 shows some estimates of local sika population densities and total sizes. Densities vary dramatically between forests and counting method, but it is clear that densities are also strongly influenced by forest structure and growth stage (Chadwick et al. 1996). It is likely that vantage point counts may underestimate the number of deer, especially of a secretive species like sika. However, dung counts also may be inaccurate due to a number of reasons including poor species identification, missed groups, inaccurate decay rate estimation, etc. (Putman 1984; Laing et al. 2003). Densities are highest in thicket forests in all cases.

Table 40.1 Local sika population estimates in Scotland. Wide method as well as true variation. Data from Chadwick et al. 1996 from Fernanda et al. 2001). Density Forest Year Forest type per km2 Mixed forest and open

variation is due to counting (except South Scotland 1997a Total population

How estimated

456 550–650 – – –

Reconstruction Reconstruction Dung counts Dung counts Vantage point count Dung counts ” ” ” ” ” Dung counts ” Vantage point counts ” ” ” ” ” Dung counts Dung counts

South Scotland ” ” ” ”

1985 1990 1991 1991 1991

Pre-thicket conifer Thicket conifer Thicket conifer

– – 4.2 12.6–34.8 10.3–15

” ” ” ” ” ” Craggan Loch Coire Carradale

1991 1997a 1997a 1997a 1997a 1997a 1993 1993 1991

Pre-felling conifer Open ground Pre-thicket Thicket Pole stage Pole stage thinned Thicket lodgepole pine Thicket lodgepole pine Establishment conifer

3.6–10.3 9.1–20.6 9.7–48.6 14.9–19.6 2.6–4.4 1.7–2.4 30.7 ± 4.6 55.2 ± 8.1 2.3

– – – – – – 160 ± 24 250 ± 37 –

” ” Thicket conifer 18–29.8 ” ” Pre-felling conifer 8–11 Achaglachgach 1991 Establishment conifer 4 ” ” Pre-thicket conifer 7.6 Knapdale 1991 Pre-thicket conifer 4–14 Scaniport 1995 Birch woodland 69.3–93.7 Scaniport 1998 Birch woodland 6.2–8.4 a South Scotland estimates from Fernanda et al. 2000

– – – – – 370–500 33–45

600

G. M. Swanson, R. Putman

Numbers of sika for other parts of Britain are less well-documented. In a managed population at Lulworth in Dorset densities have been recorded in the region of 11–12 deer/km2 (Putman and Clifton-Bligh 1997). As elsewhere, substantial populations may build up in suitable habitat. In second rotation forests in County Wicklow, densities of sika ranged from 14 to 44 deer/km2; in younger forests, estimated densities were equivalent to those recorded in high density Scottish populations, at 42–45 deer/km2 (Lowe 1994).

Life History Traits Breeding Across Britain, sika are seasonal breeders with the rut beginning towards the end of September or early October (depending on location and latitude). However, the sika rut is often more protracted than that of other species; whistling males may be heard from the end of August through until mid-December (or exceptionally, as late as mid-February). Early records of rutting behavior of sika deer in Britain suggested that males mark and defend mating territories in woodland within the female range (Horwood and Masters 1970, 1981). These territories are marked by fraying and bole-scoring of perimeter trees and thrashing ground vegetation such as heather bushes. More recent study makes it clear that the mating strategy within sika is extremely flexible, with males adopting a number of different strategies depending on circumstance (Putman and Mann 1990; Putman 1993; Thirgood et al. 1998). In various populations studied males have been found to defend rutting territories, as described, but in other cases to collect and defend a harem, as do red deer; or simply patrol areas of superior food quality within the female range and cover estrous females when encountered (Putman and Mann 1990; Putman 1993). The development of a simple breeding “lek” has also been reported in some central European populations (Bartos et al. 1992). It seems probable that, as in other deer species, males adopt differing strategies depending on the male’s age and dominance status, the density and distribution of females, and the degree of competition experienced from other breeding males (Langbein and Thirgood 1989; Putman 1993; Thirgood et al. 1998). Calves are born from early May to late June after a gestation of around 220 days. As noted, however, sika have a less well-synchronized breeding cycle than many other British deer species, and it is common to find newborn calves in August or September and, infrequently, as late as October. Normally a single calf is born, but infrequent cases of twin births have been reported (Davidson 1990; Clinton et al. 1992). Most females breed successfully for the first time as yearlings, and thereafter breed each adult year. In three English populations, New Forest, Lulworth (East Dorset), and Bovington, conception rates among yearling females were around 80%. Those among adult females were 80% (New Forest) and 90% (Dorset populations) respectively (R. Putman unpublished data; Putman and Clifton-Bligh 1997). In six sika populations

40 Sika Deer in the British Isles

601

studied across Scotland, fertility rates were equally high with yearling conception rates 80% or above, and pregnancy rates among adults mostly between 85% and 100% (Chadwick et al. 1996). Pregnant calves were recorded occasionally in animals culled throughout most Scottish and many English populations (Chadwick et al. 1996). Whether or not these precocious breeders would have been able to maintain the pregnancy to its full term, or successfully rear the resultant calf, is less certain. Reproductive rates are clearly extremely high and there is no clear evidence for any density-dependent reduction in fecundity among British sika (Chadwick et al. 1996; Putman et al. 1996; Putman and Clifton-Bligh 1997) with recorded densities up to 35 per km2. Recruitment rates to the adult population, however, are not as high as such high fecundity rates might suggest. Studies in their native habitat in Japan (Kaji et al. 1988) and in Killarney (O’Donoghue 1991) suggest there is a high early juvenile mortality and only 40–50% of calves may survive to the beginning of their first winter. At present it is unclear to what extent the various populations suffer from density-dependent responses in fertility or survivorship. Because most sika populations in Britain are now managed intensively by hunting or culling, the effects, if any, are likely to be less common.

Population Dynamics and Dispersal While populations in northern Scotland are expanding relatively rapidly, both in numbers and distribution, populations in England and in some parts of western Scotland are spreading relatively slowly. This difference in rates of spread—with rapid increases in some populations but not others—seems to reflect availability of suitable habitat for colonization, particularly the availability of young coniferous plantations. To a lesser extent, landscape features such as roads, railways, and urban settlements impede dispersal (Livingstone 2001). In continuous areas of good habitat sika show a steady expansion in range, estimated in Argyll at between 3 and 5 km per year. In other areas, where rather localized populations occupy smaller pockets of suitable habitat, there appears a rather different pattern of dispersal with long periods of no movement at all beyond the established range, followed by a sudden and rapid irruption from this source (Putman 2000). In either case, it is characteristically young males that disperse first. At the leading edge of a wave of expansion young males are encountered at a considerable distance away from the main population center and adult males may typically become established in a new area about 10–15 years before the first females are noted (Ratcliffe 1987a; Staines 1998).

Social Organization Outside the period of the rut, sexes are strongly segregated, as is common among ungulates (Main et al. 1996). In most populations males and females occupy

602

G. M. Swanson, R. Putman

distinct geographical ranges for much of the year. Although commonly considered a “herding” species, sika deer are actually one of the less social of the deer species (Putman and Mann 1990). From the end of winter through until September, the majority of animals are generally encountered alone, or in the case of females, a single animal accompanied by a calf and perhaps a yearling. Thus in studies in Dorset and Hampshire 95% of females seen through spring and summer were encountered as solitary females or single females and followers, and most males were also encountered individually (Mann 1983; Putman and Mann 1990). There is evidence, however, that group size varies with the amount of cover in given habitats (Putman and Mann 1990; Putman 2000). Whereas forest and woodlands had group sizes of one or two, heath land and agricultural fields had group sizes of two to three and two to seven, respectively. Within forested habitats relatively little forage of high quality is available, so deer move into more open habitats to gain access to better food, usually at night. The rut in September causes an increase in aggregation and increases the number of groups encountered containing adults of both sexes. Group size increases in winter, but even during this period sika are rarely observed in groups of more than five or six. Groups appear to be temporary associations formed as animals coincide in favored feeding areas. Observations of marked sika females in Wareham Forest in Dorset showed little consistency of group composition or individual association from one day to the next (Horwood and Masters 1970). Social “groups,” therefore, are extremely fluid in composition. These larger winter aggregations persist until March or April when females drift away from the groups to calve.

Home Range Individual deer have relatively small home range areas. In the Killarney National Park in Ireland (a semi-agricultural landscape with open mixed woods of 150– 250 ha bordering on larger areas of mixed conifer/broad-leaved plantations) mature females were found to use ranges of only 18–22 ha. Adult males had somewhat larger ranges of 45–55 ha, and young males ranged more widely still, within ranges of between 60 –70 ha. Little work has been done on ranging behavior elsewhere in Britain, but preliminary radio-tracking studies in two areas of northern Scotland suggest similar home range areas (C. MacLean 1998 unpublished data).

Habitat Selection British sika deer seem primarily associated in their distribution with acid soils, with the majority of populations established in areas of coniferous plantations and adjacent heath. Some idea of the relative preference shown by sika for different forest

40 Sika Deer in the British Isles

603

structures may be seen in an analysis of the range of densities which have been recorded in coniferous forest areas of different growth stage (Chadwick et al. 1996). In young plantations, typical sika deer density was 2.0–4.0 per ha, in pre-thicket 4.0–14.0, in thicket 10.3–29.8, and in pole stage and mature forest 8.0–11.0. Sika are adaptable, however, and may be encountered in a wide range of other habitats, including estuarine reed beds and similar wetland areas (e.g., at Arne in Dorset). Nevertheless, sika deer appear to be dependent on some presence of woodland cover and seem less able to adapt to completely treeless conditions than red deer in upland Scotland (Staines 1998). In their most “typical” British habitat of acid coniferous woodland, sika show a very predictable pattern of habitat use, lying up in dense thickets during the day and moving out to feed during the night in more open communities within the forest itself or on open ground beyond. This regular pattern is maintained throughout the year; indeed the overall pattern of use of available habitats changes little between seasons (Mann and Putman 1989a). Within the New Forest and associated range in southern Hampshire, sika deer occupy a more varied environment of acid grasslands, heathland, and extensive areas of broad-leaved as well as coniferous woodland. While the general pattern of habitat use remains much the same as in other areas, the animals make greater use of deciduous woodlands for feeding and make far less use of open fields and heaths (Mann and Putman 1989a).

Feeding Most studies of sika deer diets have been carried out in coniferous forest and heathland habitat. All show a high intake of grasses and heather (Calluna vulgaris and Erica tetralix) in all seasons, although the proportion of grasses to heather differed between studies: 30:50 (Mann 1983; Mann and Putman 1989b), 60:20 (Quirke 1991), and 70:20 (Mann 1983). A variety of other dietary components contributed to the remainder of the diet (pine needles, bark, and gorse Ulex europaeus), but with no single item comprising more than about 8% at any time (Mann and Putman 1989b). Few comparable data are available for (e.g.) reedbed populations or others, but see Diaz et al. (2004). New Forest sika, however, showed striking differences in diet (Mann and Putman 1989b). New Forest animals consumed considerable quantities of both deciduous and coniferous browse, particularly in the winter, when it comprised up to 23% of the total food intake. In addition, the animals showed striking seasonality in diet, feeding opportunistically on a number of foods as they become available. In spring and summer, New Forest sika fed extensively on grasses and heather—as do populations elsewhere in Britain—but their diet was far more varied. It included significant amounts of forbs, deciduous browse, gorse, and conifer needles. In autumn, only 25% of the diet was composed of heather (primarily Calluna vulgaris, but including Erica tetralix and E. cinerea) and grass, with the bulk of the food intake being composed of coniferous browse, gorse, holly, and

604

G. M. Swanson, R. Putman

acorns. In winter, there was a further increase in the intake of pine needles when less than 20% of the diet was made up of grasses (Mann and Putman 1989b). These results suggest that sika deer shift diets to obtain the most nutritious food available at any given time in a given place.

Impacts of Sika on their Habitat Because sika deer eat both deciduous and coniferous trees, and heather, they are capable of having browsing impacts on forestry similar to red deer. However, sika deer appear to preferentially be a grazing species, and they have a smaller body size than red deer. Consequently, they may cause less browse damage per capita than do red deer, although there is little question that significant damage may occur in areas of high population density (e.g., Lowe 1994). Where suitable woodland cover is available, sika tend not to be observed so commonly in open habitats, so at present they are not responsible for trampling damage to upland moss communities as red deer may be (Scottish Natural Heritage 1994). However, significant impacts have been recorded in saltmarsh and open fen communities in southern England (e.g., Diaz et al. 2004). Sika deer do graze on agricultural fields that border woodlands in which they are resident. Local populations have been recorded as causing damage to arable crops in England, although here, because populations are only locally distributed, damage to crops is likewise only likely to be of very local significance (e.g. Putman and Moore 1998; Packer et al. 1999).

Damage to Timber Crops Like red deer, sika may cause considerable damage to commercial forestry (Ratcliffe 1987a; Lowe 1994; Chadwick et al. 1996; Abernethy 1998). Damage may be caused through browsing of both lateral and leading shoots, much as by red deer in similar contexts, and also by bark-stripping in hard winters. The economic significance of such damage may be locally very considerable. One formal assessment of the extent of damage which may be caused by sika in commercial forestry has been carried out in five coniferous forests in Wicklow (Ireland; Lowe 1994) These forests were planted primarily with Sitka spruce (Picea sitchensis), Douglas fir (Pseudotsuga menziesii), Scots pine (Pinus sylvestris), and larch (Larix europaea). Between 22% and 76% of newly planted trees were found to have had their leading shoots damaged by sika. Damage proved most serious in newly established forest in its first rotation. Within second rotation forests of more varied age-structure, leader browsing was recorded on 22–47% of trees, with damage of actual economic significance caused in up to 29% of cases. Older trees also suffered significant bark-stripping damage on the main stem and lower branches,

40 Sika Deer in the British Isles

605

with between 14% and 27% of trees affected in the so-called “thicket” stage of growth before thinning and up to 32% of mature trees still showing significant damage just prior to felling (Lowe 1994). An assessment of bark-stripping damage by sika in Craggan and Loch Coire forests, Sutherland, showed that the percentage of damaged trees varied between 10.0% ± 9.3% and 75.5% ± 6.9%, dependent on forest block. Lodgepole pine (Pinus contorta) (mean 51.8% damaged) appeared more vulnerable than Sitka spruce (10% damaged) or Scots pine (0.0% damaged) (Forestry Commission for Scotland 1998 unpublished data). Mature trees may also suffer additional damage in some areas through “bolescoring” when sika stags gouge deep vertical grooves into the bole of particular trees during defense, marking, and advertisement of mating territories in the rut. Such bole-scoring damage appears to be a peculiarity of sika. A study of the incidence of bole-scoring damage in deciduous woodlands in Killarney, for example, showed for stems with a girth greater than 30 cm at breast height, 90% of yew trees showed some signs of damage, with high levels of damage also recorded for rowan (Sorbus aucuparia), ash (Fraxinus excelsior), and holly (Ilex aquifolium). From his study of five sika populations in England Carter (1984) concluded that conifers, especially spruces, were harder hit than broadleaf trees. Consistently higher damage was recorded on Norway (Picea abies) and Sitka spruce, with between 80% and 90% of trees damaged in susceptible sites. Much lower frequency of damaged stems was recorded for beech (Fagus sylvatica), hornbeam (Carpinus betulus), alder (Alnus glutinosa), sycamore (Acer pseudoplatanus), or birch (Betula spp.). However it is clear that the behavior is unpredictable in occurrence; only three of 10 sites monitored in these five populations showed any sign of bole-scoring damage. The damage done to woodlands overall is thus much less significant than that of browsing or bark-stripping, but locally it can be a serious problem.

Damage to Conservation Habitats Despite the equal potential for damage to native woodlands, there are comparatively few reports to date of sika causing significant damage to woodlands in conservation areas in United Kingdom (e.g., Putman 1995) although damage is considered significant within oak woods and yew woods in Killarney, Republic of Ireland (Larner 1977). Despite the greater abundance of sika in Scotland than elsewhere within the United Kingdom (above), there are likewise no reports to date of damage on open habitats of upland systems (moorlands, upland bogs, arctic-alpine, or montane assemblages). However, sika populations at high density are considered to be causing significant damage to internationally-recognized saltmarsh and fenland communities in Poole Harbour in Southern England (Diaz et al. 2004) where heavy grazing caused pronounced structural changes in the vegetation and led to exposure of bare ground.

606

G. M. Swanson, R. Putman

Competition with Other Deer Species There is obvious potential for competition between all deer species in Britain as they overlap in geographical location and in use of food plants. Sika deer are sympatric with red deer in most areas with the exception of the Borders in Scotland and in parts of England. They are generally sympatric with roe deer (Capreolus capreolus) in England and Scotland, and with fallow deer (Cervus dama) in most of their English range, in North Argyll and Perthshire. In all these cases there is potential for feeding competition, and with red-sika populations there is also the possibility of mate competition. There are, however, very few definitive studies of the effects of sika on other deer species. The potential interaction between red and sika has been indicated by correlative results by several studies, suggesting suppression of productivity, or geographical displacement of red deer by sika, but competition has not been unequivocally demonstrated in these cases (McKelvey 1959; Dzieciolowski 1979; Feldhamer and Armstrong 1993; Abernethy 1994b). Again, based on correlation, Makovkin (1999) suggested complete displacement of wapiti by an expansion of sika deer in Primorsky Krai, Far East Russia. There is likewise no direct evidence of competition between sika and other deer species, although many people believe that both roe and red deer decline in forests inhabited by sika, and some data show lower than expected densities of roe in sika inhabited forest (Chadwick et al. 1996). Only one study has so far been published that more objectively explores patterns of habitat use and diet of sika in sympatry with other deer species. In the New Forest, in southern England, Putman (1986, 1996) considered overlaps in habitat use and diet of New Forest sika deer with sympatric populations of fallow, red, and roe deer (and free-ranging populations of horses and cattle). Overlaps in habitat use were highest with fallow deer and roe, most notably in autumn and winter (niche overlap 0.80–0.90, falling to 0.65–0.75 in spring and summer). Overlaps in diet were highest with fallow and red deer (fallow, consistently 0.80–0.90; red 0.90–0.95), while overlaps with roe deer were lower (between 0.52 and 0.63 in different seasons). Despite this overlap in habitat use and diet, no direct evidence was found to suggest direct competitive effects in terms of population dynamics (Putman and Sharma 1987), perhaps because population numbers of all species are strongly controlled by (human) management and, thus, kept below levels at which resources become limiting and competition would be apparent.

Hybridization Distribution It should be recognized that there is a strong possibility that some of the initial sika stocks released Britain were already genetically introgressed in captivity. The records kept by early zoological parks and private collections were incomplete and

40 Sika Deer in the British Isles

607

not very reliable. Many exchanges of sika stocks were made through traders who were business men, not biologists or zoo keepers. Furthermore, it is clear that many of the releases in the United Kingdom were secondary introductions from Viscount Powerscourt’s deer park at Enniskerry in Ireland (Powerscourt 1884), where these animals were maintained in close proximity to populations of red deer within the same park. These animals—by this stage of very uncertain genetic provenance— were subsequently the source for many introductions to other parks throughout Ireland, Scotland, and England (Ratcliffe 1987a). Whatever the original stocks, sika deer are known to be able to interbreed with red deer under certain circumstances (Putman 2000; Pemberton et al. 2006) wherever they have been introduced into native red deer range. So the large question becomes: what is the nature of the hybridization and will the phenotypes of each species be completely swamped in a large, hybrid swarm? Or, will the integrity of the species phenotypes by retained by behavioral (assortative mating), ecological (niche partitioning, habitat requirements, etc.), and/or selective pressures (disruptive selection) on the hybrid progeny. So far, whatever the genetic introgression, various processes seem to be leading to retention of “sika” and “red” deer phenotypes in most parts of the British Isles (although there is one established hybrid swarm in County Wicklow, in southern Ireland). Thus, in the paragraphs which follow, the terms “sika” and “red” deer should be understood as referring to the phenotype of the individual or population and not the underlying genetic makeup. In Ireland, all populations, with the exception of Killarney are thought to be hybrid to some extent. Genetic analysis of the Wicklow sika (which is the one British population which presents a hybrid “swarm” of phenotypic intermediates between red and sika) has confirmed extensive hybridization, even within local populations that appear superficially unaffected, e.g., Lugalow Estate (G. Swanson 1999 unpublished data). Analysis of samples from Killarney has so far remained inconclusive. Alleles typical of the opposite species were found in the small number of animals sampled; however, it was possible that they were ancestral polymorphisms rather than introgressed alleles. Nevertheless, recent reports of hybridization may result in further investigation of this population. The most detailed analysis of hybridization between red and sika deer in wild populations has been carried out in Argyll, Scotland (Ratcliffe et al. 1992; Abernethy 1994a; Goodman et al. 1999; Swanson 1999; Pemberton et al. 2006). The pattern is one of a moving wave of hybridization related to the expansion of sika into resident red deer areas showing locally high proportions of the population carrying small amounts of hybrid DNA (Pemberton et al. 2006). Across Scotland, Swanson (1999) used samples collected between 1991 and 1997 to carry out a genetic survey of hybridization on the mainland. This survey was then extended to include samples from a number of Hebridean Islands (Pemberton et al. 2006). A definite pattern emerged of strong assortative, withinspecies mating (red phenotype and sika phenotype), but occasional hybridization where sika deer have expanded their range into resident red deer populations. The best genetic and phenotypic evidence for recent hybridization in Argyll comes from the north of Kintyre. A number of sites showed a non-random association

608

G. M. Swanson, R. Putman

of introgressed alleles and in one population (Loch Avich), 14 (32%) of the sika deer contained multiple alleles normally found in red deer. Phenotypic hybrids were reported in Inverliever and 12% of the Loch Avich samples were identified as hybrid by the forest ranger. In Galloway a pocket of hybridization was identified in Dundeugh Forest, northeast of the main Galloway forest, at the leading edge of sika colonization. A number of phenotypic hybrids have been reported in recent years (J. Wykes 1996 personal communication), one of which was sampled and genotyped as an F1 or F2 hybrid. A further three red deer samples from the same forest were genotyped as F2 or F3 backcrosses. Evidence from Sutherland, Easter Ross, and Argyll suggests that once the initial wave had passed, there has been little disruption of the overall appearance of each phenotypic species. However, considerable numbers of both red and sika deer coexist with small amounts of hybrid DNA. The amount of hybrid DNA was consistently greater in sika populations than in red populations and this may be due to many factors, including biases in the direction of hybridization and backcrossing. Contrary to a previous genetic study (Abernethy 1994a, b) which used some markers now known to be not diagnostic in Scottish populations (Goodman et al. 1999) no evidence for extensive hybridization was found in the Great Glen samples. This is perhaps surprising as phenotypic hybrids have previously been reported there (Ratcliffe 1987a). In England, early studies of cranial morphology (Lowe and Gardiner 1975; Ratcliffe et al. 1992; Putman and Hunt 1994) suggested that most populations of sika deer were of hybrid status, with the possible exception of the population in the New Forest of Hampshire. Diaz et al. (2006) subsequently investigated the genetic status of a number of sika populations in southern England with molecular markers. Based on analysis of 329 samples collected from all over Purbeck, they found extensive evidence that the Dorset population is of hybrid status; populations from the New Forest in Hampshire showed far less evidence of introgression, with any red deer markers of ancient, rather than more recent origin and at levels no greater than might be accounted for by natural variation/mutation. Although no overall significant genetic differences were found between the New Forest and Purbeck populations of sika, New Forest deer were more tightly clustered around the range of genetic variation exhibited by control samples obtained from Japan. It is further clear that relatively small numbers of individuals of the New Forest sika contained detected introgressed red DNA. While it is not possible to attach confidence intervals to these findings, these small, but perhaps important, genetic differences between the populations support earlier deductions based on cranial morphometrics (Putman and Hunt 1994) that New Forest deer may be more pure sika than other populations of feral sika in Britain. The sika “component” of both Purbeck and New Forest animals was statistically similar to control samples from Kyushu (Island), Japan and is consistent with the suggestion by Goodman et al. 2001 that the United Kingdom sika population originated from Kyushu, and more specifically, the Nagasaki region of Japan.

40 Sika Deer in the British Isles

609

Causes of Hybridization The reason why some populations have hybridized while others have not is of interest because it may provide information upon which management strategies can be based. Two main hypotheses have been put forward to explain why some British red and sika deer populations are more vulnerable to hybridization than others. The first involves the genetic provenance of the local Cervus deer, and the relative body size of the populations (Lowe and Gardiner 1975; Harrington 1979, 1982), which can be complicated by prior hybridization. Past hybridization was thought to have increased the body size of Asian mainland (Lowe and Gardiner 1975) and deer park sika populations (Harrington 1979, 1982), making them more reproductively compatible with red deer and, therefore, vulnerable to further hybridization. These theories were suggested following breeding experiments (Harrington 1979) and skull morphometric analysis (Lowe and Gardiner 1975). As part of the Red/Sika Hybridization Project, University of Edinburgh, Scotland (see Goodman et al. 1999; Swanson 1999) various Scottish red and sika deer populations were compared genetically to establish whether there was any association between genetic background and post-introduction hybridization (Swanson 1999). The analysis revealed that genetically very similar populations showed different patterns of hybridization. In those areas where it was possible to establish the pre-introduction hybrid status of the sika population through the chronology of introductions, there was no evidence that previous hybridization resulted in more recent hybridization events. As another measure, carcass weights of culled deer were compared as a rough measure of body size to investigate whether populations with smaller body size differences were more likely to hybridize. Carcass weights varied greatly within and between populations, almost certainly due to age of culled samples and variation in habitat quality (Mitchell and Crisp 1981; Mitchell et al. 1981; Feldhamer et al. 1985; Ratcliffe 1987b; Albon and Clutton-Brock 1988). Still, the comparison revealed no association between either current average size or current size differences between potential mates and recent hybridization. These results suggest that the average body size of the sympatric populations may not be a factor in fostering, or preventing, hybridization. There is some evidence to suggest that probability of hybridization is lower in areas where populations of both phenotypic species are of sufficient size that they may breed “true” (i.e., the probability of hybridization is increased in areas where one or other species is rare, and thus denied the opportunity to breed readily with conspecifics). Ratcliffe (1987a) suggested that hybridization is thus more likely where sika or red deer males are colonizing areas with resident populations of the opposing taxon. The evidence from the Hybridization Project suggests that hybridization does tend to occur in these circumstances as all three newly hybridized populations identified across Scotland (Argyll, Easter Ross, and Borders) are at the leading edge of sika colonization into resident red deer areas (Swanson 1999; Goodman et al. 1999, 2001). The colonization process (and lack of conspecifics with which to breed) may, therefore, be one factor involved in hybridization.

610

G. M. Swanson, R. Putman

Nevertheless, hybridization is clearly not inevitable in these circumstances, as some sika populations in this survey (Loch Morar, Great Glen, and Fife) expanded without evidence of recent hybridization (Swanson 1999; Pemberton et al. 2006). There has been some debate regarding the direction of crossing between wild red and feral sika populations (Harrington 1979; Abernethy 1994a). This question was investigated by Goodman et al. (1999) and Swanson (1999) by studying the pattern of nuclear (passed on by male and female) and mitochondrial (mtDNA, passed on only by the female) genetic markers found in Scottish red and sika populations. As mtDNA is inherited from the mother only, it provides a record of each deer’s matrilineal taxon. Differences in the distribution of genotypes, therefore, can provide information about the direction of crossing (Arnold 1993). Across Scotland, all hybrids of very recent origin contained red mtDNA and were therefore descended from a lineage of red females (Swanson 1999). Although some sika-like hybrids did contain sika mtDNA in the south of Kintyre, it is very likely that they are the result of backcrossing by hybrids into the sika population (Goodman et al. 1999). If there are no practical ways of stopping the transfer of DNA, what are the likely consequences for the Cervus deer population in Britain? In some areas hybridization is already almost complete in the sika-like population (e.g., Wicklow and Sutherland), in the sense that most individuals have some red DNA, and considerable in the red population (e.g., Argyll), with many other red populations show some evidence of possible past hybridization with sika and/or wapiti. Despite this, phenotypic disruption has not followed (even with characters normally considered as diagnostic, e.g., metatarsal gland) apart from subtle changes in body measures in some populations (Swanson 1999). Abernethy (1994a) suggested that the absence of obvious phenotypic disruption in Argyll was possibly only temporary, and that in time, following sufficient gene flow, assortative mating may break down causing a chain reaction of hybridization. The principle example of where this has occurred is in Wicklow, Ireland where hybridization has resulted in the virtual extinction of completely red-like deer following approximately 20 more generations of crossing than in Argyll (Ratcliffe 1987a). However, the history of Wicklow red and sika deer is not comparable to most other populations in Britain, and the same outcome should not be considered inevitable. Hybridization occurred within the park prior to release and the local red deer population was relatively small, of poor quality, and was culled heavily resulting in a disturbed social organization (Delap 1967). In virtually all British populations, strong within-species mating predominates and hybridization is thus relatively rare.

Management Implications Given the overall phenotypic similarities of hybrid and non-hybrid deer by species in most British Isles populations, it is not surprising that managers have found differentiating between the two almost impossible (Swanson 1999). In Argyll, only

40 Sika Deer in the British Isles

611

2% of the genetically diagnosed hybrids were identified by the forest ranger as showing hybrid traits, and across Scotland only one out of the eight samples submitted as hybrid-like was confirmed genetically (Swanson 1999). Nevertheless, deer managers in both Scotland and Ireland have correctly identified several pockets of recent hybridization. As most hybrids and non-hybrids cannot be identified visually, the selective culling of hybrid deer is not a practical option, unless it is targeted at a local population containing obvious hybrids, rather than at specific individuals. Even then, it is only likely to slow down the rate of transfer of DNA rather than stop it entirely. As the spread of sika, and subsequent hybridization, is associated with the distribution of forests, it appears that the only practical way to slow down the spread of sika and their genes is to restrict aforestation in some areas, as previously suggested by Ratcliffe (1987b). Because of their potential for hybridization with red deer and the damage that may be caused to forestry interests once populations reach significant population density there is growing concern about control of sika populations within Britain (Deer Commission for Scotland 1998). Many populations are already closely managed in an attempt to maintain low population sizes, prevent further increases, or actively reduce populations. In Scotland, the major governmental organizations concerned with deer regard sika as a considerable threat, particularly to forestry, and are actively promoting a policy of rigorous control (Deer Commission for Scotland 1998). More recently sika deer have been added to Schedule 9 of the Wildlife and Countryside Act (1981, as amended 1998), thus translocation of sika to areas outside their existing range is now illegal in any part of Britain. In Ireland, the main control operations are carried out by the Irish Parks and Wildlife Service (Duchas) and the state forestry organization (Coilte). However, despite these increased control measures sika continue to expand in most areas and are likely to eventually colonize all suitable habitat. Although managers of some populations have managed to decrease local densities considerably, there are few examples of total eradication. Thus, sika deer are likely to remain part of the British woodland landscape for a considerable time to come.

Literature Cited Abernethy, K. 1994a. Establishment of a hybrid zone between red and sika deer (genus Cervus). Molecular Ecology 3:551–562. Abernethy, K. 1994b. The introduction of sika deer, Cervus nippon nippon, to Scotland. Ph.D. thesis, University of Edinburgh, Edinburgh, United Kingdom. Abernethy, K. 1998. Sika deer in Scotland. Deer Commission, Scotland, The Stationery Office. Albon, S. D., and T. H. Clutton-Brock. 1988. Climate and the population dynamics of red deer in Scotland. In M. B. Usher and D. B. A. Thompson, editors, Ecological change in the Uplands. Special Publication Number 7 of The British Ecological Society. Blackwell Scientific, Oxford, United Kingdom. Arnold, J. 1993. Cytonuclear disequilibria in hybrid zones. Annual Review of Ecology and Systematics 24:521–544.

612

G. M. Swanson, R. Putman

Avise, J. C. 1994. Molecular markers, natural history and evolution. Chapman and Hall, London, United Kingdom. Bartos, L., U. Zeeb, and J. Mikes. 1992. Lekking behaviour in sika deer. Pages 205–208 in N. Maruyama, B. Bobek, Y. Ono, W. Regelin, L. Bartos, and P. R. Ratcliffe, editors, International Symposium on Wildlife Conservation - Present trends and perspectives for the 21st century. Japan Wildlife Research Center, Tokyo, Japan. Carter, N. A. 1984. Bole scoring by sika deer (Cervus nippon) in England. Deer 6:77–78. Chadwick, A. H., P. R. Ratcliffe, and K. Abernethy. 1996. Sika deer in Scotland: Density, population size, habitat use and fertility - Some comparisons with red deer. Scottish Forestry 50:8–16. Clinton, T., T. J. Hayden, J. M. Lynch, and P. Murphy. 1992. A case of twin foetuses in a sika hind (Cervus nippon) from County Wicklow, Ireland. Deer 8:437–438. Davidson, M. M. (1990) Sika deer. Pages 468–477 in C. M. King, editor, The handbook of New Zealand mammals, Oxford University Press, Auckland, New Zealand. Deer Commission for Scotland. 1998. A policy for sika deer. Deer Commission for Scotland, His Majesty’s Stationary Office, London, United Kingdom. Delap, P. 1967. Hybridisation of red and sika deer in north-west England. Deer 1:131–133. Diaz, A., E. Pimm, and J. Hannaford. 2004. Ecological impacts of sika deer on Poole Harbour saltmarshes. Pages 175–188 in J. Humphreys and V. May, editors, The ecology of Poole Harbor. Elsevier, London, United Kingdom. Diaz, A., S. Hughes, R. J. Putman, R. Mogg, and J. M. Bond. 2006. A genetic study of sika (Cervus nippon) in the New Forest and in the Isle of Purbeck, Southern England: Is there evidence of recent or past hybridisation with red deer? Journal of Zoology 207:227–235. Dzieciolowski, R. 1979. Structure and spatial organisation of deer populations. Acta Theriologica 24:3–21. Feldhamer, G. A., J. B. Stauffer, and J. A. Chapman. 1985. Body morphology and weight relationships of sika deer (Cervus nippon) in Maryland. Zeitschrift Saugetierkunde 50:88–106. Feldhamer, G. A., and W. E. Armstrong. 1993. Interspecific competition between four exotic species and native artiodactyls in the United States. Transactions of the North American Wildlife and Natural Resources Conference 58:468–478. Fernanda, F. C. Marques, S. T. Buckland, D. Goffin, C. E. Dixon, D. L. Borchers, B. A. Mayle, and A. J. Peace. 2001. Estimating deer abundance from line transect surveys of dung: Sika deer in southern Scotland. The Journal of Applied Ecology 38:349–363. Goodman, S., N. Barton, G. Swanson, K. Abernethy, and J. Pemberton. 1999. Introgression through rare hybridization: A genetic study of a hybrid zone between red and sika deer (Genus Cervus) in Argyll, Scotland. Genetics 152:355–371. Goodman, S., H. Tamate, R. Wilson, J. Nagata, S. Tatsuzawa, G. Swanson, J. Pemberton, and D. McCullough. 2001. Bottlenecks, drift and differentiation: The population genetic structure and demographic history of sika deer (Cervus nippon) in the Japanese archipelago. Molecular Ecology 10:1357–1370. Harrington, R. 1973. Hybridisation among deer and its implications for conservation. Irish Forestry Journal 30:64–78. Harrington, R. 1979. Some aspects of the biology and taxonomy of the deer of the County Wicklow region, Ireland. Ph.D. thesis, National University of Ireland, Dublin, Ireland. Harrington, R. 1982. The hybridisation of red deer (Cervus elaphus L. 1758) and Japanese sika deer (Cervus nippon nippon Temminck 1838). Transactions of the International Congress of Game Biology 14:559–571. Harris, S., P. Morris, S. Wray, and D. W. Yalden. 1995. A review of British mammals: Population estimates and conservation status of British mammals other than cetaceans. Joint Nature Conservation Committee, Peterborough, United Kingdom. Horwood, M. T., and E. H. Masters. 1970. Sika deer. British Deer Society, Reading, United Kingdom. Horwood, M. T., and E. H. Masters. 1981. Sika deer (Cervus nippon) with particular reference to the Poole Basin (2nd edition). British Deer Society, Lower Basildon, United Kingdom.

40 Sika Deer in the British Isles

613

Kaji, K., T. Koizumi, and N. Ohtaishi. 1988. Effects of resource limitation on the physical and reproductive condition of sika deer on Nakanoshima Island, Hokkaido. Acta Theriologica 33:187–208. Laing, S.E., S. T. Buckland, R. W. Burns, D. Lambie, and A. Amphlett. 2003. Dung and nest surveys: Estimating decay rates. Journal of Applied Ecology 40:1102–1111. Langbein, J., and S. J. Thirgood. 1989. Variation in mating systems of fallow deer in relation to ecology. Ethology 83:195–214. Larner, J. B. 1977. Sika deer damage to mature woodlands of southwestern Ireland. Proceedings of the X111th Congress of Game Biology:192–202. Livingstone, S. R. 2001. The application of GIS to the spread of introduced Japanese sika deer (Cervus nippon) in Scotland. M.Res. thesis, University of Edinburgh, Edinburgh, United Kingdom. Lowe, R. 1994. Deer management: Developing the requirements for the establishment of diverse coniferous and broadleaf forests. Unpublished report, Coilte, Bray, Co. Wicklow, Ireland. Lowe, V. P. W., and A. S. Gardiner. 1975. Hybridisation between red deer (Cervus elaphus) and sika deer (Cervus nippon) with particular reference to stocks in N.W. England. Journal of Zoology, London 177:553–566. Main, M. B., F. W. Weckerly, and V. C. Bleich. 1996. Sexual segregation in ungulates: New directions for research. Journal of Mammalogy 77:449–461. Makovkin, L. I. 1999. The sika deer of Lazovsky reserve and surrounding areas of the Russian Far East. Almanac Russki Ostrov, Vladivostok, Russia. Mann, J. C. E. 1983. The social organisation and ecology of the Japanese sika deer (Cervus nippon) in southern England. Ph.D. thesis, University of Southampton, United Kingdom. Mann, J. C. E., and R. J. Putman. 1989a. Habitat use and activity patterns of British sika deer (Cervus nippon Temminck) in contrasting environments. Acta Theriologica 34:83–96. Mann, J. C. E., and R. J. Putman. 1989b. Diet of British sika deer (Cervus nippon Temminck) in contrasting environments. Acta Theriologica 34:97–110. McKelvey, P. J. 1959. Animal damage in North Island protection forests. New Zealand Science Review 17:28–34. Mitchell, B., and J. M. Crisp. 1981. Some properties of red deer (Cervus elaphus) at exceptionally high population-density in Scotland. Journal of Zoology, London 193:157–169. Mitchell, B., W. Grant, and J. M. Cubby. 1981. Notes on the performance of red deer, Cervus elaphus, in a woodland habitat. Journal of Zoology, London 194:279–284. O’Donoghue, A. 1991. Growth, reproduction, and survival in a feral population of Japanese sika deer (Cervus nippon). Ph.D. thesis, University College, Dublin, Ireland. Packer, J. J., J. Doney, B. A. Mayle, C. F. Palmer, and M. Cope. 1999. Field and desk studies to assess tolerable damage levels for different habitats and species of deer. Report to the Ministry of Agriculture, Fisheries and Foods, on project VC 0315, London, United Kingdom. Pemberton J., G. Swanson, N. Barton, S. Livingstone, and H. Senn. 2006. Hybridisation between red and sika deer in Scotland. Deer 13:22–26. Powerscourt, V. 1884. On the acclimatisation of the Japanese deer at Powerscourt. Proceedings of the Zoological Society of London 1884:207–209. Putman R. J. 1984. Facts from faeces. Mammal Review 14:79–97. Putman, R. J. 1986. Competition and coexistence in a multi-species grazing system. Acta Theriologica 31:271–291. Putman, R. J. 1993. Flexibility of social organisation and reproductive strategy in deer. Deer 9:23–28. Putman R. J. 1995. Deer on national nature reserves: Problems and practices. English Nature Research Report 173, English Nature, Peterborough, United Kingdom. Putman R. J. 1996. Competition and resource partitioning in temperate ungulate assemblies. Chapman and Hall, London, United Kingdom. Putman R. J. 2000. Sika deer. British Deer Society/Mammal Society, London, United Kingdom. Putman R. J., and S. K. Sharma. 1987. Long term changes in New Forest deer populations and correlated environmental change. Symposia of the Zoological Society of London 58:167–179.

614

G. M. Swanson, R. Putman

Putman, R. J., and J. C. E. Mann. 1990. Social organisation and behaviour of British sika deer in contrasting environments. Deer 8:90–94. Putman, R. J., and E. J. Hunt. 1994. Patterns of hybridisation and introgression between red and sika deer in different populations of the North of Scotland and Argyll. Deer 9:104–110. Putman, R. J., and J. R. Clifton-Bligh. 1997. Age-related bodyweight, density and fecundity in a south Dorset sika population (Cervus nippon), 1985–93. Journal of Natural History 31:649–660. Putman, R. J., and N. Moore. 1998. Impact of deer in lowland Britain on agriculture, forestry and conservation habitats. Mammal Review 28:141–164. Putman, R. J., J. Langbein, A. J. M. Hewison, and S. K. Sharma. 1996. Relative roles of densitydependent and density-independent factors in population dynamics of British deer. Mammal Review 26:81–101. Quirke, K. 1991. The diet of red deer, sika deer and Scottish blackface sheep in Killarney National Park, Co. Kerry. M.Sc. thesis National University of Ireland, Dublin, Ireland. Ratcliffe, P. R. 1987a. Distribution and current status of sika deer (Cervus nippon) in Great Britain. Mammal Review 17:39–58. Ratcliffe, P. R. 1987b. The management of red deer in upland forests. Her Majesty’s Stationery Office, London, United Kingdom. Ratcliffe, P. R., A. J. Peace, M. Hewison, E. Hunt, and A. H. Chadwick. 1992. The origins and characterisation of Japanese sika deer populations in Great Britain. Pages 185–190 in N. Maruyama, B. Bobek, Y. Ono, W. Regelin, L. Bartos, and P. R. Ratcliffe, editors. International symposium on wildlife conservation - Present trends and perspectives for the 21st century. Japan Wildlife Research Center, Tokyo, Japan. Rose, H. 1994. Scottish deer distribution survey, 1993. Deer 9:153–155. Scottish Natural Heritage. 1994. Red deer and the natural heritage. Scottish Natural Heritage, Edinburgh, United Kingdom. Staines, B. W. 1998. Sika deer: Their status, distribution and ranging behaviour. In B. W. Staines, S. C. F. Palmer, I. Wyllie, R. Gill and B. Mayle, editors, Desk and limited field studies to analyse the major factors influencing regional deer populations and ranging behaviour. Final Report to Ministry of Agriculture, Fisheries and Food on contract VC 0314, London, United Kingdom. Swanson, G. M. 1999. The genetic and phenotypic consequences of translocations of deer (Genus Cervus) in Scotland. Ph.D. thesis, University of Edinburgh, Edinburgh, United Kingdom. Thirgood, S. J., J. Langbein, and R. J. Putman. 1998. Intraspecific variation in ungulate mating strategies: The case of the flexible fallow deer. Advances in the Study of Behaviour 28:333–361. Ward, A. I. 2005. Expanding ranges of wild and feral deer in Great Britain. Mammal Review 35:165–173. Whitehead, G. K. 1950. Deer and their management in the deer parks of Great Britain and Ireland. Country Life, London, United Kingdom. Whitehead, G. K. 1964. The deer of Great Britain and Ireland. An account of their history, status, and distribution (1st edition). Routledge and Kegan Paul, London, United Kingdom.

Chapter 41

Free–Ranging and Confined Sika Deer in North America: Current Status, Biology, and Management George A. Feldhamer and Stephen Demarais

Abstract In North America, the largest numbers of free-ranging sika deer (Cervus nippon) currently occur in Texas and Maryland, with smaller populations on Chincoteague Island, Virginia; there are very small populations about which little is known in North Carolina and Kentucky. Sika deer in these five states are sympatric with native white-tailed deer (Odocoileus virginianus). A few individual sika deer escapees from game farms or ranches occur occasionally in other states and may be taken during the hunting season, but there are no other reproductively viable, free-ranging populations. Likewise, there are no known free-ranging sika deer in any of the Canadian provinces or territories. Confined populations of sika deer on game farms or ranches are much more widely distributed in North America, with animals currently held in at least 39 states and four Canadian provinces. In addition to other management concerns related to exotics, the perceived magnitude of problems associated with the potential spread of chronic wasting disease (CWD) has resulted in very restrictive regulations against transporting sika deer (and other ungulates) in most states and provinces.

Introduction Considering microbes, invertebrates, plants, and vertebrates, Pimentel et al. (2000) estimated that approximately 50,000 exotic species have been introduced into the United States, and “… invasion biology has burgeoned in the last decade into a main focus of conservation biology and ecology” (Simberloff 1997:3). Although many of these introductions have been accidental, most nonindigenous mammalian species, including numerous species of ungulates, have been intentionally released. These releases have included “… a variety of motives: for sport; for sentimental or nostalgic reasons; as an aesthetic amenity; as a potential source of food; as furbearers; or as a form of biological control” (Lever 1985:1). Today, free-ranging populations of exotic cervids in North America include axis deer (Cervus axis), fallow deer (Dama dama), red deer (C. elaphus), sambar (C. unicolor), as well as sika deer (C. nippon). Free-ranging sika deer, like any other D. R. McCullough et al. (eds.), Sika Deer: Biology and Management of Native and Introduced Populations, DOI: 10.1007/978-4-431-09429-6_41, © Springer 2009

615

616

G. A. Feldhamer, S. Demarais

established exotic, produce a mosaic of interacting direct and indirect effects on native communities. Unfortunately, relatively little work has been done on sika deer in North America, compared to the body of literature from Asia and Europe. Many questions remain unanswered and much remains to be done to place sika deer, and the biology of nonindigenous ungulates in general, “… in the context of wellestablished paradigms that characterize a mature science” (Simberloff 1997:17). This chapter includes biological aspects of free-ranging populations of sika deer in North America within the larger context of exotic ungulates. We also include current management considerations, questions, and controversies surrounding sika deer confined on game farms or ranches, and current state and provincial regulations and initiatives.

Free-Ranging Populations Texas No state has more exotic ungulates in terms of species diversity and abundance than Texas (Teer et al. 1993; Demarais et al. 1998; Teer 2003). In addition to sika deer, fallow deer and axis deer are the most numerous cervids introduced to Texas. Large numbers of confined and free-ranging nilgai antelope (Boselaphus tragocamelus) and numerous other species of ungulates also occur (Mungall and Sheffield 1994; Teer 2003). There is little public land in Texas; 99% of rangeland and 97% of forests are privately owned (White 2000). In terms of exotics, there are few regulations that encumber landowners. For many Texans, Baccus (2002:281) suggested that one of the lures of game ranching in Texas may have been reaction against state and federal government, because “the exotic game rancher did not have to conform to bag limits, seasonal rules or other hunting regulations of the state.” As such, it is imperative to view sika deer in Texas within the larger context of the total number of introduced ungulates and the cumulative impact exotics have on habitats and native white-tailed deer. Sika deer in Texas were first released by Richard Friedrich in 1932 on the Bear Creek Ranch in Kerr County, in the Edwards Plateau ecological region (Fig. 41.1). Ranches in others areas of the state were stocked from these animals (Mungall and Sheffield 1994). In response to increased numbers, censuses of exotic ungulates were conducted at about five-year intervals beginning in 1963. The total number of exotic ungulates has increased dramatically since the initial survey, with an average increase of 57% between 1963 and 1988. The last survey of exotics in Texas was conducted in 1994. It was estimated that about 195,000 individuals representing 71 species occurred. Of these, approximately 77,000 animals were free-ranging (Traweek 1995). Currently, free-ranging sika deer occur in 11 counties in the central Edwards Plateau and South Texas Plains ecological regions (Fig. 41.1). Based on the 1994 survey, there were approximately 3,000 free-ranging sika deer in Kimble County,

Fig. 41.1 States and counties with known populations of free-ranging sika deer (Cervus nippon) as of 2003.

41 Free-Ranging and Confined Sika Deer in North America 617

618

G. A. Feldhamer, S. Demarais

and 1,000 in Kerr County. Estimates were 400–550 sika deer in Real, Bandera, and Sutton counties, and ≤100 in Medina, Gillespie, Llano, Bexar, Bell, and Duval counties (Traweek 1995). This probably is an underestimation, however, given previous growth rates and the reliance on a voluntary survey of landowner-generated population estimates. As noted by Demarais et al. (1998:50), growth rates may have slowed but another possibility is that “the reliability of the survey instrument may no longer be at an acceptable level.” Many ranchers refused to participate in the 1994 survey, or were reluctant to disclose information, possibly because of regulatory concerns and private property issues (Demarais et al. 1998; Baccus 2002). Although previous subspecies designations of sika deer worldwide were problematic at best (Tamate and Tsuchiya 1995; Cook et al. 1999; Nagata et al. 1999; Randi et al. 2001), recent genetic studies have clarified this issue (Nagata chapter 3). Introductions in Texas include the small Japanese subspecies (C. n. nippon), Formosan sika (Cervus n. taiouanus), and possibly melanistic Ryuku sika (C. n. keramae), with interbreeding and a resulting variety of color morphs, as well as Manchurian (Dybowski’s) sika (C. n. hortulorum). Mungall and Sheffield (1994) suggested that 90% of sika in the United States are Japanese x Formosan crosses. Larger sika bucks are preferred by many trophy hunters, as well as for venison production and by-products.

Maryland Sika deer currently occur in four counties on Maryland’s eastern shore. They are primarily in southern Dorchester County, with smaller numbers in Wicomico, Somerset, and Worcester counties, including Assateague Island (Fig. 41.1). Throughout the eastern shore of Maryland, the sika deer population is estimated at 5,000–10,000 individuals and is sympatric with white-tailed deer. These deer originated from an unknown number of individuals, likely brought from the Woburn Abbey population in England. These likely were the first sika deer in North America. Clement Henry kept them for several years in an enclosure in Cambridge, Dorchester County. In 1916, he released four or five animals (possibly only one male) on James Island in Chesapeake Bay. Approximately two km from mainland Dorchester County, James Island is only about 113 ha in size but provided excellent habitat. The population grew to about 300 animals by 1958, with individuals occasionally leaving the island and swimming to the mainland. In the late 1950s, about 60% of the individuals on James Island died, probably because of malnutrition caused by high population density (Christian et al. 1960). Sika deer on mainland Dorchester County had established a viable population by then. They were rare until the late 1960s, when they began to increase in abundance. Significant growth of the population occurred throughout the 1970s (Flyger and Davis 1964; Feldhamer et al. 1978) despite heavy legal harvest and poaching pressure. Based on Sex/Age/Kill Model estimates, the sika deer

41 Free-Ranging and Confined Sika Deer in North America

619

population in Dorchester County from 1988 through 2000 has ranged from about 3,800 to 7,800 (B. Eyler, 2002, Maryland Department of Natural Resources, personal communication) and currently is estimated at about 5,500. In 2002, there were 923 antlered sika deer and 951 antlerless sika taken during the firearms, archery, and muzzleloader hunting seasons, all from the southern half of the county. This compares with 947 white-tailed bucks and 1,997 antlerless white-tailed deer harvested throughout the county (D. Hotten, 2003, Maryland Department of Natural Resources, personal communication). Sika deer hunting in Maryland has grown tremendously in popularity. There is relatively little public land in Dorchester County, but two wildlife management areas offer hunting opportunities. Also, Blackwater National Wildlife Refuge (NWR) allows hunting by permit. Most hunting in the area is done through private clubs that lease exclusive rights to hunt from landowners. The Maryland Department of Natural Resources (DNR) has been encouraged by landowners and outfitters to attempt to increase the number of sika deer to enhance opportunities for additional revenue through hunting. However, the management objective of Maryland DNR for sika deer is to maintain populations at their current levels (D. Hotten, 2003, Maryland DNR, personal communication). Bag limits have been reduced from the 1970s; currently, one antlered and one antlerless sika deer may be taken in the firearms, muzzleloader, and archery seasons. In 1930, Charles Law made a separate introduction of approximately six sika deer to Assateague Island, Worcester County. These deer were likely from the same stock as those introduced to James Island (Flyger 1960). Sika deer on Assateague Island now outnumber white-tailed deer by about 3:1 (Keiper 1990). Deer from this second introduction eventually moved south from Assateague Island onto Chincoteague Island National Wildlife Refuge, Virginia.

Virginia Sika deer in Virginia are found only on Chincoteague Island (Fig.41.1), the southern extension of Assateague Island, Maryland. A barrier island approximately 5,800 ha in size, Chincoteague NWR is a major staging area for migratory shorebirds. As noted, sika deer moved south following the 1930 introduction on Assateague Island, Maryland. Population densities of sika deer on Chincoteague have remained stable since at least the early 1990s with an estimated 800–1,200 animals. Sika deer represent a unique hunting opportunity in the area, with average annual harvest of about 300 animals. Abundance of white-tailed deer on the island is about one-third that of the sika deer, but harvest pressure is directed almost entirely at the exotic. As in Maryland and Texas, sika deer on Chincoteague prefer fresh and saltwater marsh habitats, whereas whitetails occur more often in fields and forests (T. Penn, 2003, Chincoteague NWR, personal communication).

620

G. A. Feldhamer, S. Demarais

North Carolina A free-ranging population believed to be Dybowski’s sika resides in northeast Hyde County (Fig. 41.1). This population resulted from an illegal release in the early 1980s that was intended to establish a commercial hunting facility. The animals were purchased in Missouri and initially kept in a large fenced area of about 200 ha. An unknown number eventually escaped or were released. These have moved 50–60 km from the release site and are expanding their range. A number of sika deer have been harvested since then. However, a local hunt club in Hyde County attempts to maintain the population by not harvesting females. Only one or two stags are killed every few years; the average live weight is about 110 kg. The population is believed to be fairly stable at about 50 animals (J. S. Osborn, 2003, North Carolina Wildlife Resources Commission, personal communication).

Kentucky Populations of free-ranging C. n. nippon in Kentucky are small and of relatively recent origin. Documented sightings of sika deer have occurred in Jefferson County east of Louisville. This group of about 20 individuals is the result of an intentional release by a private landowner about 10 years ago. A few sika deer, escapees from captive facilities, also occur in Campbell and Bracken counties (Fig. 41.1) where individuals occasionally are harvested (J. Day, 2003, Kentucky Department Fish and Wildlife Resources, personal communication). Little is known about either of these populations.

Sika Deer on Game Farms or Ranches Sika deer are confined on game farms or ranches in at least 39 states and four Canadian provinces (Table 41.1). Teer et al. (1993:448) defined game farming as “intensive husbandry of wild stocks in penned conditions,” whereas game ranching “involves free-ranging, managed wildlife usually on private property.” Ranched deer may or may not be enclosed with fences (White 2000). Proponents of game farms and ranches consider commercial exploitation of confined cervids an “alternative agricultural resource.” The commercialization of native and exotic wildlife on game farms or ranches remains a relatively new phenomenon in North America. Numerous associations exist to promote the industry, such as the North American Elk Breeders Association and the Exotic Wildlife Association. The North American Deer Farmers Association (NADeFA) claims members in 45 U.S. states, seven Canadian provinces, and Mexico. According to NADeFA, there were more than 10,000 deer farms/ranches

41 Free-Ranging and Confined Sika Deer in North America

621

Table 41.1 Presence or absence of confined sika deer (subspecies undifferentiated) on game farms/ranches in the United States and Canada. Data are from responses to mail surveys and/or personal communication to wildlife management agencies, agriculture departments, and/or extension agents in all states and provinces or territories in 1993 (from Feldhamer 1995) and 2003 and from an internet search in 2008. Augmented with data from the North American Deer Farmers Association (NADeFA). Also, see Buck (2002). Except for Manitoba, data for Canada are from 2003 only. State/province/ territory Alabama

Alaska Arizona

Arkansas

California

Colorado

Connecticut

Delaware

Presence, absence, regulations, and comments 1993: Regulations do not allow game ranching for deer, but deer are held for other purposes. No known sika deer in state. Hunters take sika deer, however. 2003: Regulations do not allow importation of sika deer; no changes since 1993. 1993: Exotics allowed under permit. No permit ever issued for sika deer. 2003: No changes since 1993; no sika deer known in Alaska. 1993: Unregulated. May be possessed by anyone in any quantity. No specific information on whether or not sika deer are in the state. 2003: Sika are not restricted (the only exception in the genus Cervus), are confined in at least one location, and believed to be in numerous other locations. Possession and transportation of live cervids will become much more restrictive because of CWD. 1993: Permits required only if animals are for sale. Only one game breeder known to have sika deer. 2003: Two breeding facilities with a total of seven sika deer, although no data on facilities that do not breed or sell. Moratorium on new deer facilities since April 2002; no importation of cervids allowed since November 2002. 2008: Restrictions on importation of whole cervid carcasses from CWDpositive states and provinces imposed in 2005. 1993: Moratorium on importation of cervids adopted 1992; apparently only one game farm with a single sika deer. 2002: Sika deer can be sold and transported between permit holders in the state. There are approximately 125 individual sika deer held in commercial facilities. 2008: Carcass transportation regulations. 1993: As of 1992, there were 59 sika deer distributed among six breeders. 2003: Records of species held are no longer kept; probably still the same number of ranches holding sika deer. Import regulations in place and veterinary certificate required for negative CWD, bovine Tb, brucellosis. 2008: Carcass transportation regulations. 1993: Approximately six to eight sika deer held in the state. 2003: There are 12 facilities with sika deer. Current ban on importation of any live cervids, because of disease concerns. 2008: Carcass transportation regulations. 1993: No known individuals in captivity; there are abundant free-ranging sika deer in Maryland, approximately 50 km from Delaware border. (continued)

622

G. A. Feldhamer, S. Demarais

Table 41.1 (continued) State/province/ territory

Florida

Georgia

Hawaii

Idaho

Illinois

Indiana

Iowa

Kansas

Kentucky

Presence, absence, regulations, and comments 2003: Confined sika deer are known from one location; free-ranging sika from Maryland uncommon but do occur. Current regulations prohibit importation. 1993: As of 1992, 67 sika deer on hunting preserves, 449 on game farms, 26 in wildlife exhibits. Total represents about 9% of the number of exotic ungulates in the state. 2003: Confined sika deer present in state but no current data on location or numbers. Because of CWD, current ban on all cervid importation unless they are from herds disease-free for five years. 1993: State does not regulate holding sika deer or maintain records on game farms. Believed to be none but unknown. 2003: There are a few sika deer held for breeding or exhibit. Because of CWD, import of any cervid illegal without departmental authorization since August 2002. 1993: Not surveyed. 2003: No sika deer held. USDA and state Forestry and Wildlife permits required to import; unlikely these would be granted for any new species. 1993: Occur on several licensed game farms; the exact number is not known. State now prohibits entry of any more sika deer because of concerns over hybridization with elk. 2003: Similar to 1993; importation of sika and other exotic cervids regulated. 1993: Only records necessary are for holding white-tailed deer; no permit necessary for exotic deer. There are two elk breeders that hold sika deer. 2003: Limited numbers of sika deer probably held. No cervid may be imported into the state for any reason. 2008: Carcass transportation regulations. 1993: State does not require game breeders to provide information on exotics. Sika deer occur based on data provided by the NADeFA. 2003: Game breeder licenses issued for white-tailed deer only. Board of Animal Health prohibits importation of all cervids because of CWD. 1993: Approximately 100 game breeders hold deer in the state; since they are not required to provide a list of species held, there is no information on sika deer. 2003: Some game breeders hold sika deer; reports of some escapees each year. Imports must be certified free of CWD, brucellosis, and Tb. 2008: Carcass transportation regulations. 1993: State has no control over exotic deer; one game breeder is known to have sika deer, but others are believed to illegally export the species to other states. 2003: Confined sika deer are held. State Department of Animal Health is responsible for confined deer. Import of cervids from certified herds is allowed. 1993: Five game breeders known to have sika deer; unknown numbers. 2003: Confined sika are held (in addition to one free-ranging herd). No new captive sika allowed to enter the state due to CWD. 2008: Carcass transportation regulations. (continued)

41 Free-Ranging and Confined Sika Deer in North America

623

Table 41.1 (continued) State/province/ territory Louisiana

Maine

Maryland

Massachusetts

Michigan

Minnesota

Mississippi

Missouri

Montana

Presence, absence, regulations, and comments 1993: Possession of exotic deer regulated by state Department of Agriculture and Forestry. They have no records for any species of deer held but are in the process of developing standards. Sika deer are held in several locations in the state. 2003: Sika deer are on less than 20% of the 142 licensed deer farms. Sika allowed if tested for brucellosis and Tb; elk, red deer, mule deer, and white-tailed deer currently prohibited. 1993: There are 25 to 50 sika deer in the state, among six to10 breeders. 2003: Penned sika deer are raised for commercial purposes. Current ban on importation of any cervid probably will be extended. 1993: Only three individuals are known to possess confined sika deer; unknown numbers. 2003: Only three facilities with sika deer. Regulations against new facilities in place since 1992. Currently, importation of any cervid is banned primarily due to concerns over CWD. 1993: A few (“less than 25”) sika deer held under propagator’s permits, along with red deer and fallow deer. 2003: Five licensed deer farms with sika. Currently, importation of all cervids is restricted. 2008: Importation of carcasses forbidden from CWD-positive states or provinces in 2005. Boned meat allowed. 1993: No regulations regarding possession of exotic deer. Three permit holders for white-tailed deer have indicated they have sika deer; only about 20 in the state. Sika deer have escaped from captivity but there are no known wild populations. 2003: Sika deer held in registered facilities, regulated by state Department of Agriculture. Current ban on importation of all cervids due to CWD. 2008: Carcass transportation regulations. 1993: Regulations began in 1992 for exotics. Several game farms hold sika deer in the state. 2003: Several farms/ranches with sika deer. As of 1 January 2004, new restrictions from Board of Animal Health on importation of cervids due to CWD. 2008: Carcass transportation regulations. 1993: Approximately 25 sika deer on two game ranches in the state. 2003: Board of Animal Health moratorium on importation/movement of cervids due to CWD. 1993: No permit necessary to hold sika deer. They are held in the state but there is no information on numbers or distribution. 2003: Sika deer are held on licensed facilities. Restricted import of cervids from areas with known CWD; others from herds with CWD surveillance for three years. 1993: Sika deer are prohibited from entering the state under new rules adopted in 1992, because of concerns of hybridization with elk, competition, and difficulty of controlling escapees. At that time, there were about 25 sika on six licensed game farms. These can no longer be propagated. (continued)

624

G. A. Feldhamer, S. Demarais

Table 41.1 (continued) State/province/ territory

Nebraska

Nevada

New Hampshire

New Jersey

New Mexico

New York

North Carolina

North Dakota

Ohio

Presence, absence, regulations, and comments 2003: There are no sika deer on licensed game farms or ranches. They are currently on the list of species prohibited for importation. 1993: No permit required for most exotics, including sika deer. About 35 sika known on at least seven game farms. 2003: About 120 sika deer on 10 licensed farms/ranches; state Department of Agriculture regulates and monitors importation, including for brucellosis and Tb. 1993: No confined sika deer held. 2003: Sika deer are prohibited from importation, although there can be exceptions. Currently there are two commercial facilities with sika deer. 1993: Sika deer may be raised and sold for venison. Only two individuals hold a total of 12 sika deer. There have been escapees; these have been recaptured or killed. 2003: Two propagators licensed to hold sika deer. State Agriculture Department currently prohibits importation of any live cervid. 1993: Ten to 12 individuals hold sika deer for breeding or hobby purposes; unknown numbers. 2003: Five people hold 12 sika deer as pets; a zoo holds18 sika deer for educational purposes. As of April 2002, all cervids banned from importation due to CWD. 1993: One person in state holds three sika deer. 2003: No sika deer known to be held. Exotic deer are not allowed on landowner game parks, and all exotic cervids currently prohibited from importation. 2008: Carcass transportation regulations. 1993: Forty-six exhibitors/breeders known to hold sika deer in the state. 2003: Sika deer are confined in state. Current prohibition on import of sika deer and other Cervus spp. 2008: Carcass transportation regulations. 1993: Permit system does not designate species of deer held. A “number” of people are known to possess sika deer for breeding purposes; farming for venison is restricted to fallow deer. 2003: Four of 85 licensed facilities hold sika deer; individual records now kept for each facility. As of 15 May 2002 all cervids prohibited from importation. 1993: State does not regulate exotic deer and has no records of sika deer, although people with permits for white-tailed deer may hold them. 2003: No private farms are licensed for sika deer. Import and testing requirements including for CWD are now in place. 2008: Carcass transportation regulations. 1993: Apparently the permit system does not designate the species of deer held. One person known to have sika deer based on information from NADeFA, there are probably several others. 2003: Sika deer are held in private ownership. Imported cervids must be certified for brucellosis and Tb. (continued)

41 Free-Ranging and Confined Sika Deer in North America

625

Table 41.1 (continued) State/province/ territory Oklahoma

Oregon

Pennsylvania

Rhode Island

South Carolina

South Dakota

Tennessee

Texas

Utah

Presence, absence, regulations, and comments 1993: Nine people are licensed and hold sika deer; unknown number of deer. 2003: Seven licensed breeders hold sika deer. Since June 2002, no import of cervids from states with CWD, or other states without approved CWD monitoring program. 1993: State adopted new rules on private holding of deer in 1993. There are at least eight operators holding a total of 66 sika deer. There have been several reported escapes. 2003: Sika deer are held in licensed facilities. Import ban on parts or carcasses from states with CWD. 2008: Carcass transportation regulations. 1993: Confined sika deer present. 2003: Sika deer held on approximately 12 licensed game farms and by menagerie permit holders. Importation of all cervids banned 1 August 2002 until establishment of health testing and monitoring protocols. Sika deer occasionally harvested during hunting season. 2008: Carcass transportation regulations. 1993: State has no information pertaining to individual game farms or whether sika deer are held; they believe they are not. 2003: No facilities approved to hold sika deer. Emergency ban on importation of all cervids currently in place. 2008: Carcass transportation regulations. 1993: Permit required to import sika deer; none currently held in captivity. 2003: No sika deer held; no regulatory provisions allow deer farming/ ranching. 2008: Carcass transportation regulations. 1993: State had no license requirements for game farms in the past. They are currently formulating regulations. Sika deer are known on game farms. Several escapes have occurred; status of resulting wild populations is unknown. 2003: Two permittees hold confined sika deer. Import regulations regarding brucellosis, Tb, and CWD testing. 1993: State does not regulate sika deer and their status is unknown to regulatory agencies. One breeder in state known from NADeFA. 2003: Sika deer are held on licensed facilities. No importation of cervids from areas with CWD. 2008: Carcass transportation regulations. 1993: Sika deer on 207 game ranches in 77 counties; 6,225 confined sika. 2003: Hundreds of game ranches, many with sika deer. No importation of cervids from areas with CWD. 1993: Game farms and ranches for big game species are prohibited, as is importation and possession of sika deer other than for zoos. 2003: Sika deer are prohibited from importation or possession, except for educational or scientific use. Three sika deer held under certificate for exhibition. (continued)

626

G. A. Feldhamer, S. Demarais

Table 41.1 (continued) State/province/ territory

Presence, absence, regulations, and comments

2008: Carcass transportation regulations. 1993: The only exotic deer raised legally are red deer and fallow deer; sika are not legal but may occur on one shooting preserve. 2003: Sika deer are prohibited on game farms; since May 2002 importation ban on all live cervids. 2008: Carcass transportation regulations. Virginia 1993: Only fallow deer are allowed on game farms. 2003: Game farms/ranches currently are illegal; three people hold sika deer for exhibit purposes. Importation of all cervids prohibited since 1994. 2008: Carcass transportation regulations. Washington 1993: As of 1993, 699 sika deer were held at 19 locations. State is unaware of any wild sika deer, but one was harvested during 1992 deer season. 2003: Deer/elk farms regulated out of existence in 1994; no selling, transfer, or propagation since then. One case (eight to 10 sika deer) still under litigation. 2008: Carcass transportation regulations. West Virginia 1993: No records are available as no permits for sika deer are required. State knows one operator with sika deer; other people hold them but no records. 2003: No known sika deer on any game farms/ranches. Intra- and interstate movement of cervids prohibited due to concerns for CWD. Wisconsin 1993: In 1992, 32 deer farms held a total of 210 sika deer. 2003: Eight registered deer farms with sika deer. All imported cervids must be five-years certified free of brucellosis, Tb, and CWD. Wyoming 1993: State does not allow importation or possession of any exotic big game. 2003: Only one game farm (NX Bar Ranch) in state; elk only. No sika deer permitted in state. Alberta No exotics are allowed on game farms/ranches. Borders are closed to importation of any cervid due to CWD concerns. British Columbia No sika deer known to be held on licensed game farms. Fallow deer, reindeer, and bison are the only exotic ungulates allowed; native cervids are precluded from import into the province. Manitoba A number of facilities may hold sika deer. Until recently, the species was not regulated, monitored, or recorded. In 2001, all cervids were subject to import, export, and possession permit requirements. In 2002, a moratorium on import of any cervid due to CWD. 2008: Carcass transportation regulations. New Brunswick No sika deer are known on licensed or unlicensed game farms. Only four exotic ungulate species allowed on game farms: red deer, elk, fallow deer, and bison. Newfoundland/ No sika deer; there are no game farms/ranches in the province. Labrador No licensed game farms or ranches in NWT. Northwest Territories Nova Scotia Sika deer are on two licensed deer farms. Strict regulations on imports/ exports of all cervids. Vermont

(continued)

41 Free-Ranging and Confined Sika Deer in North America

627

Table 41.1 (continued) State/province/ territory Nunavut Ontario

Prince Edward Island Quebec

Saskatchewan Yukon Territory

Presence, absence, regulations, and comments No game farms or sika deer are in the territory. No regulations specific to the territory currently are in place regarding importation, however. Sika deer are allowed on game farms: probably less than 100 individuals. Currently no licensing/regulatory mechanism for deer/elk farming under Ontario Ministry of Agriculture and Food. Exotic species are not covered in legal framework for wildlife in Ontario. No sika deer are held. Game farm regulations currently do not preclude import. Importation of sika deer is not prohibited and approximately 50 are kept on five facilities. Monitored by provincial Department of Agriculture, Fisheries, and Food, but no permits are necessary for captive cervids. One or two sika deer escapees killed in last 10 years. No sika deer on game farms. Importation of sika deer is not allowed. No sika deer held on game ranches; only elk and reindeer. New regulations make it difficult to import exotic species.

in North America in 2003 with an annual production value of about $111 million. Game farming in Canada has expanded since the 1980s, especially in western provinces, with fallow deer the primary cervid (Renecker 1989) but sika deer are farmed as well. Nonetheless, the commercial importance of the game industry in North America remains a small fraction of that of the domestic cattle industry (Coon et al. 2002), and increasing regulatory restrictions probably have reduced the number of practitioners during the past decade. Certainly, “clear distinctions in popular culture between domestic livestock, free-ranging wildlife and household pets … now are blurred” (Diez et al. 2002:290). Nonetheless, confined exotic ungulates should not be considered “domesticated” (sensu Clutton-Brock 1999) given their retained behavioral, morphological, and physiological characteristics. Sika deer on game ranches initially were raised for sport trophy hunting, and this remains a significant purpose. The primary goal of game farms is production of brood stock, venison, and by-products such as antler velvet and hides. Commercial enterprises also may include animal parks for visitor viewing and photography. Among exotic species of cervids in North America, sika deer probably rank after fallow deer, red deer, and possibly axis deer in the number of confined individuals and account for only about 2.3% of the deer raised (B. Cahill, 2003, NADeFA, personal communication). The numbers of commercial operators and confined sika deer in each state or province remain variable, ranging from one or two individuals with a few animals in some states to hundreds of operators and thousands of confined sika deer in Texas. However, most game farms are small; many involve 10 or fewer individual sika deer. Many states now ban importation and do not allow breeding. It is difficult to determine which subspecies are held, with breeders often considering their

628

G. A. Feldhamer, S. Demarais

sika deer simply as large- or small-bodied. As noted, several subspecies occur in Texas, with the larger C. n. hortulorum more valuable than the small C. n. nippon. Mungall (2000:757) noted that “Large-scale sika farming has not developed in North America because sika meat can be stronger tasting and less tender than axis venison.… The discovery that sika will breed with elk (wapiti, Cervus elaphus nelsoni) has led to development of “American silk” for the deer farmer…” It is easier now to determine which state agencies regulate confined sika deer than it was 10 years ago. In the early 1990s, state biologists sometimes did not know, or were unaware, if confined sika deer were present. Exotic cervids essentially were “below the radar” and surveys to determine the presence of confined deer produced conflicting data (Teer 1991; Oregon Department of Fish and Wildlife 1992; Feldhamer 1995). Today, state biologists are much more aware of the status of exotic cervids within their jurisdictions, if for no other reason because of the threat of chronic wasting disease (CWD). Perceived positive and negative aspects of sika deer and other introductions, including the potential threat of CWD, are discussed under Management Considerations.

Biology of Sika Deer in North America Body Weight, Antler Characteristics, Genetics Osborn et al. (1995) established regression equations to determine live weight and dressed weight from chest girth dimensions based on a sample of 90 subadult and adult female sika deer from Kerr and Real counties in the Edwards Plateau region of Texas. Different equations were necessary seasonally (winter vs. summer) for both age classes. The mean whole weight of 37 adult females in the summer was 38.2 kg; for seven subadults it was 31.5 kg. In the winter, mean whole weight of 40 adults was 39.1 kg, and 33.4 kg for six subadults. The subspecies of sika deer in Maryland is C. n. nippon. These deer are smaller than those from Texas. Mean dressed weight of yearling males is 24.0 kg, whereas for yearling females it is 20.5 kg. Growth rates, average body weight, and measurements of both sexes were given by Feldhamer et al. (1985) based on a sample of 606 individuals harvested during a five-year period. Males grow more rapidly than females; sexual dimorphism is apparent by 1.5 years of age. Males attain maximum weight at 5.5 years of age, and females at 7.5 years of age. Although these deer are small, even for the subspecies, they are in excellent condition as indicated by large amounts of visceral fat throughout the year. Good body condition results from interspersion of forest, marsh, and agricultural lands, and relatively mild winters in the region. Although winter temperatures in Maryland generally are too mild to affect sika deer through thermal stress, body weights of males and females were significantly correlated with mean ambient temperatures in February (Feldhamer 1985). This relationship may have been associated with

41 Free-Ranging and Confined Sika Deer in North America

629

molting from winter to summer pelage. Mean body weights of yearlings also were inversely related to total precipitation in July. Rainfall, tidal action, and low elevation in Maryland often result in flooded fields, which significantly reduce availability of cultivated crops to the deer. Because of the small number of individuals introduced to James Island, Maryland, including possibly only a single male, sika deer exhibit several indications of a founder effect. One of the most striking is that the maximum number of antler points in mature males is six (three on a side; Feldhamer et al. 1985). This contrasts with adult C. n. nippon in Japan, most of which have eight-point racks (four on a side; Ohtaishi 1976), as do three- to five-year-old males in Texas (Mungall and Sheffield 1994). Antler length of sika deer in Texas is variable, but generally ranges from 28 to 48 cm. It reaches 60 cm or more in Dybowski’s deer (Mungall and Sheffield 1994). In addition to antler structure, phenotypic homogeneity is evident in skulls of sika deer from Maryland. Based on 24 skull measurements taken on 106 cleaned skulls of females from Dorchester County, and 81 skulls of female C. n. hortulorum from Russia, there was significantly less variation in the introduced sample, again probably because of a founder effect (Feldhamer and Brtalik 1988). The introduced deer in Maryland also had significantly less cranial variation when compared to a native population from Japan (Matsumoto et al. 1984). Besides phenotypic homogeneity in cranial and antler structure, there also is genotypic homogeneity in the population of sika deer in Maryland. Feldhamer et al. (1982) collected muscle and liver tissue from 120 sika deer from Dorchester County. Ten enzymes and 19 scorable loci were consistently resolved, with 25– 40 individuals examined for each locus. No heterozygosity or polymorphism was found. Interestingly, no polymorphism occurred in 43 C. n. hortulorum from a native population in Russia scored for eight loci by Markov et al. (1992). Unlike introduced sika deer, these results were unexpected in a native population and may be indicative of low genetic diversity in the species. Morphological indications of founder effects in other introduced sika deer populations in North America have not been investigated. Polymorphism in the diploid chromosome number occurs in sika deer (Gustavsson and Sundt 1969; Koulischer et al. 1972). The karyotypes of four sika deer from Dorchester County were identical (2n = 68), however. Thirty-two chromosome pairs in addition to the sex chromosomes were acrocentric. There was one metacentric pair, probably the result of centric fusion. Large amounts of pericentric heterochromatin were evident on all chromosomes except the metacentric (van Tuinen et al. 1983). Asher et al. (1999) discussed an unusual case of natural hybridization between a male sika deer and female axis deer on a game farm in Tennessee. The offspring, a female, had a karyotype of 2n = 67, intermediate between sika deer and axis deer (2n = 66). Pelage coloration of free-ranging sika deer is variable in Texas, Maryland, and elsewhere but typical for the species. However, very unusual white calves have been produced on a game farm in Lakewood, Pennsylvania (B. Cahill, 2003, NADeFA, personal communication).

630

G. A. Feldhamer, S. Demarais

Reproduction Reproduction is distinctly seasonal, with breeding during September through November and calving during April through June in Texas (Howery et al. 1989). Mullan et al. (1988) first described the occurrence of accessory corpora lutea in sika deer based on specimens from Dorchester County, Maryland. They also suggested that sika deer were capable of breeding as calves, that is, when six months old. Sample sizes in that study were small, however. In a follow-up study, Feldhamer and Marcus (1994) examined reproductive tracts of 54 adult females, 25 yearlings, and 15 calves. They found corpora lutea of pregnancy or estrus in 51 adults (94.4%) and all the yearlings. Four of the calves had a blastocyst present, while four others had a corpus luteum of estrus. Based on mean size of fetuses, they concluded that adult sika deer generally bred earlier in the season than yearlings, and both these age classes bred earlier than calves. Excellent body condition and mild climate probably contribute to breeding by calves, as did probable moderate population density. Approximately 50% of fawn white-tailed deer in this area also breed. This greater-than-expected potential reproductive contribution of the calf age group may help facilitate maintenance of sika deer densities in Maryland, despite the fact that sika deer have a single calf (Howery et al. 1989), although they rarely do produce twins (Feldhamer and Marcus 1994). Thus, neonatal weights and overall condition should be less affected by nutritional stress in sika deer than in white-tailed deer, which usually have twins (Demarais et al. 1998; Miller et al. 2003). Because sika deer calves breed late in the season, progeny from calves may be small and suffer higher mortality rates (Takatsuki and Matsuura 2000). Nonetheless, there is no information on whether fetuses of pregnant calves reach full term or on subsequent body condition or survival of neonates. Immunocontraception is important in zoo management as well as in wild herds under certain conditions. Kirkpatrick et al. (1996) found that porcine zonae pellucidae (PZP) vaccine was an effective contraceptive in Formosan sika deer. Raphael et al. (2003) reported that melengestrol acetate fed at a concentration of 0.000154% in pelleted feed produced 100% contraception in Formosan sika deer

Diseases and Parasites Transmission of new diseases and parasites between exotic and native ungulates is commonly cited by resource managers as a significant concern. Richardson and Demarais (1992) compared the parasite fauna of 28 white-tailed deer and three species of exotics, including 16 sika deer, from the YO Ranch in Kerr County, Texas. There were 26 species of exotic ungulates on the ranch, with an estimated density of one ungulate per 5 ha. All cervids surveyed had relatively few parasites. Other than the louse Haematopinus suis, usually found on feral hogs (Sus scrofa), the parasites on sika deer were the same as those of white-tailed deer. Interestingly, sika

41 Free-Ranging and Confined Sika Deer in North America

631

deer had a higher mean condition rating than the white-tailed, axis, and fallow deer examined (Richardson and Demarais 1992). Bovine tuberculosis is commonly tested for in confined ungulates. The etiological agent of tuberculosis (Tb) in deer is Mycobacterium bovis, which has been documented in captive and free-ranging sika deer (Dodd 1984; Mirsky et al. 1992). Rhyan and Saari (1995:220) described “abundant bizarre giant cells” from lymph and lung lesions as a consistent feature of Tb in sika deer, which differed from lesions found in other cervids. Keel et al. (2003) reported mortality in three confined female sika deer that had lesions consistent with malignant catarrhal fever (MCF). Although MCF has been reported previously from sika deer (Sanford et al. 1977), the deer in this instance were infected with caprine herpesvirus 2 (CpHV-2), usually found in goats. Crawford et al. (2002) noted that CpHV-2 can cause more chronic disease in sika deer than is expected for MCF. Shah et al. (1965) detected antibodies to the virus Myxovirus parainfluenza 3 in a sika deer from the Maryland/Virginia area. The number of human cases of Lyme disease continues to increase in the United States, with 23,305 cases reported in 2005 and 19,931 cases in 2006 (Center for Disease Control 2008). The causative agent for Lyme disease is the spirochete Borrelia burgdorferi, with the common vector being ixodid ticks. Although the most common reservoirs for the ticks are white-footed mice (Peromyscus leucopus) and white-tailed deer, Oliver et al. (1999) found antibodies to B. burgdorferi from 14 diverse mammalian species, including sika deer from Assateague Island.

Feeding Habits and Rumen Ciliates Sika deer have very flexible feeding habitats, as might be expected for a species adapted to a range of habitats and regions worldwide. The interspersion of forests, brushland, and grassland in central and south Texas is ideal for sika deer, as it is for many other ungulates. In the Edwards Plateau region of Texas, seasonal variation followed general availability patterns. Grasses were readily consumed when succulent, and forbs were eaten when available. In times of reduced forage supply due to season or drought, browse consumption increased considerably (Jackley 1991). Grasses taken include Texas wintergrass (Stipa leucotricha), vine mesquite (Panicum obtusum), and witchgrass (Leptoloma cognatum). Leavenworth vetch (Vicia leavenworthii), velvet bundleflower (Desmanthus velutinus), bighead evax (Evax prolifera), and wild onion (Allium canadense) are among the numerous species of forbs taken. Favored browse species include acorns and leaves of live oak (Quercus virginiana) and white shin oak (Q. durandii), hackberry (Celtis sp.), wild plum (Prunus sp.), mustang grape (Vitis mustangensis), greenbriar (Smilax sp.), and Texas sotol (Dasylirion texanum) (Mungall and Sheffield 1994). In bottomland conifer forests in Maryland, poison ivy (Toxicodendron radicans), Japanese honeysuckle (Lonicera japonica), and greenbriar are heavily browsed. Pokeweed

632

G. A. Feldhamer, S. Demarais

(Phytolacca americana), wax myrtle (Myrica sp.), American holly (Ilex opaca), and cordgrass (Spartina patens) also are taken (Flyger and Warren 1958). Ciliated protozoans from the rumens of 26 sika deer from the Edwards Plateau region of Texas were described by Dehority et al. (1999). In addition to rumen protozoa previously noted for sika deer in Texas (Sybert 1990) and Japan (Imai et al. 1993), they found five new host records, including Entodinium exiguum, E. laterospinum, Epidinium ecaudatum, Isotricha intestinalis, and Dasytricha ruminantium.

Interspecific Competition Interspecific competition—the use or defense of a limited resource by a species that reduces the availability of that resource to other species—often is difficult to quantify between free-ranging populations of exotic and native cervids. Difficulties relate to the effects of intraspecific interactions as well as “temporal and spatial complexities of study areas, and unique attributes and characteristics of populations involved” (Feldhamer and Armstrong 1993:468). Nonetheless, available empirical evidence usually involves: (1) overlap in resource use; (2) changes in resource use; or (3) changes in population density, age structure, fecundity, or survival in one or both of the competing species. Evidence of competitive interaction between sika deer and white-tailed deer on Assateague Island was based on significant dietary overlap (Keiper 1990). Similar dietary overlap also was noted between sika deer and white-tailed deer on overgrazed ranges in Texas (Kelley 1970; Butts et al. 1982; Henske et al. 1988; Jackley 1991). Sika deer outcompete white-tailed deer, especially in degraded habitats. Digestibility of three forages common on overgrazed range in Texas was greater for sika deer than white-tailed deer; greater digestibility gives sika deer a competitive advantage due to their ability to more completely use these forages when highly digestible forages are limited (Demarais et al. 2003). This is consistent with Hofmann’s (1985) comparisons of rumen anatomy, feeding behavior, and digestive physiology of deer. Sika deer, considered “intermediate feeders,” are more adaptable in forage selection, less specialized, and more opportunistic than are “concentrate selectors” such as white-tailed deer. Thus, sika deer not only take higher-quality forage when available, but they also can survive on lower quality forage that will not sustain white-tailed deer. Reproductive characteristics have been proposed as a second, possibly additive competitive mechanism in Texas (Demarais et al. 1998). As noted, sika deer usually produce only one calf with peak calving during May, in contrast to white-tailed deer which often produce two fawns with peak fawning in June. The frequency of summer drought and the loss of forage to other competing species may limit the availability of nutrition capable of supporting recruitment of healthy, twin white-tailed deer fawns. The combination of earlier calving with better condition at birth of a singleton may increase the competitive advantage sika deer exhibit over whitetailed deer (Demarais et al. 1998).

41 Free-Ranging and Confined Sika Deer in North America

633

Trends in hunter-harvested sika deer and white-tailed deer in Dorchester County, Maryland and nearby Chincoteague Island, also suggest competitive interaction, assuming that harvest trends parallel relative population density (Feldhamer and Armstrong 1993). Since the mid-1970s, the number of sika deer harvested in southern Dorchester County has been about three times that of white-tailed deer. Similarly, as previously noted for Chincoteague Island, sika deer represent almost the entire deer harvest each year, yet continue to outnumber white-tailed deer. It is possible that competition is not operating, and there are alternative explanations for the observed population trends. For example, habitat changes may have occurred that favor the exotic, or disease may be affecting native white-tailed deer and not sika deer. These factors do not appear to be operating, however, and interspecific competition is the most feasible explanation. Throughout southern Dorchester County, there are several large hunt clubs that 30 years ago had only white-tailed deer on the property. These clubs harvest the same number of deer today, but almost all are sika deer; rarely is a white-tailed deer seen. Direct interspecific competition may be minimized, however, by different use of habitats and resulting avoidance. Sika deer in Dorchester County spend much time in marshes, dominated by saltgrass (Distichlis spicata), needlerush (Juncus sp.), three-square rush (Scirpus olneyi), and cordgrass (Spartina alterniflora), as well as wet, lowland forests dominated by loblolly pine (Pinus taeda). White-tailed deer occur more often in upland conifer cover and agricultural areas. Thus, there is spatial segregation, with populations syntopic. For example, Eyler (2001) found average interspecific home range overlap of 18 radiocollared sika females and 15 female white-tailed deer was significantly less than the degree of intraspecific overlap for each species. Sika deer are much more physically aggressive than whitetailed deer, and although they are larger, white-tailed deer generally avoid contact with the exotic.

Management Considerations Few issues in wildlife management today are as controversial as the introduction of nonindigenous ungulates. Holding wildlife for commercial purposes has generated debate among the public, within regulatory and wildlife management agencies, and in the scientific community among those favoring farming and ranching deer (Yorks 1989; White and Valdez 1991) and those opposing such activities (Geist 1985, 1988, 2002; Teer 1989; Peyton 2000). There certainly are positive and negative aspects to the introduction of sika deer and other cervids in North America (Demarais et al. 1990; Samuel and Demarais 1993). Positive aspects include hunter access to popular trophy species without the necessity of overseas travel. In addition to fee hunting, exotics such as sika deer, fallow deer, and red deer commonly are stocked because of their tractability. They provide a diversified, alternative agricultural base, with production of venison, antler velvet and other medicinal by-products, and breeding stock. All provide needed revenues to producers.

634

G. A. Feldhamer, S. Demarais

Propagation of threatened or endangered species, as well as the aesthetics and other nonconsumptive values of these animals are additional arguments for introductions (Mungall 2000; Baccus 2002). Conversely, valid arguments for opposing introductions have been made. Concerns include the philosophical issue of private property rights of citizens and the positive aspects of developing a new industry versus the traditional view that wildlife belongs to the people of the state with management entrusted to regulatory management agencies. Problems which relate specifically to sika deer include regulatory authority and potential ecological impacts and are expanded below. Nonetheless, as Baccus (2002:276) stated, whether these introductions are “good or bad is in the eye of the beholder.” Regardless of the arguments for and against exotic ungulates, in states like Texas and Maryland, free-ranging populations at this point in time are a fait accompli. Complete eradication, even if agreed upon by all the diverse stakeholders, would be practically impossible.

Regulatory Authority As noted, in the early 1990s, many states did not have legislation that clearly defined “wildlife” or distinguished between native and exotic species (Ervin et al. 1992; Wheaton et al. 1993). There often was confusion among different agencies within a state concerning regulation of exotic deer. Ervin et al. (1992:251) felt the most important result of their 1989 questionnaire to all the states on the legal status of exotic deer was “the general lack of knowledge among state agencies which either are or could be directly involved in regulating the exotic deer industry.” Uncertainty regarding legislative and regulatory authority is much reduced in most states and provinces. The rapid increase in the number of game farms/ranches in the 1990s focused attention on many regulatory and social issues related to confined ungulates, including ownership and commercialization of native wildlife as well as exotics. As of 2002 (Diez et al. 2002), farmed cervids were regulated by the state department of agriculture in 25 states, by the fish and wildlife agency in 19 states, and in six states by both departments. Nonetheless, relations between state agencies are not always smooth, and “most state regulations are not consistent or comprehensive” (Coon et al. 2002:263). There also are questions over disposition of animals from defunct operations, animal welfare, and responsibility for escapees and the cost of recovery (Kahn 1993). Other regulatory concerns involve the impact of ranching on law enforcement and differentiating between farmed and illegally harvested deer during the legal deer seasons. As indicated in Table 41.1, several states have initiated much more restrictive regulations during the past decade governing possession of exotic deer, in many cases banning importation entirely. Animal health in Canadian provinces and territories is regulated by the Canadian Food Inspection Agency (CFIA), which also has responsibility for game farms. CFIA administers the Farmed Cervid Program under the Health of Animals Act and

41 Free-Ranging and Confined Sika Deer in North America

635

Regulations which encompasses reportable diseases such as Tb, brucellosis, and CWD. The Farmed Cervid Program requires at least one negative tuberculosis herd test every three years and at least one negative brucellosis test in the herd’s history. With negative test results, deer herd owners can move deer in Canada (barring interprovincial restrictions) with the required movement permits issued by CFIA (B. Tapscott, Alternative Livestock Specialist, Ontario Ministry of Agriculture and Food, personal communication, 2003). Provincial standards are variable. Some provinces allow exotic cervids while others do not, and some provinces simply ban game farms (Renecker 1989; White 2000; see Table 41.1).

Ecological Impacts This is the area of greatest concern about sika deer in North America and includes concerns such as habitat degradation, interspecific competition, hybridization, and disease transmission. The potential for habitat degradation is well-documented. On overgrazed habitats, as previously noted, sika deer can survive more easily than white-tailed deer because of their rumen morphophysiology and feeding habits, further exacerbating habitat decline. Likewise, in conjunction with feeding habits, the direct and indirect interspecific competitive abilities of sika deer have been documented. The problem of potential hybridization between escaped or free-ranging sika deer and elk (red deer), well-known in Europe (Lowe and Gardiner 1975; Bartos et al. 1981; Goodman et al. 1999, Bartos chapter 39), is of concern in western states and provinces. The hybridization of sika deer and elk in Texas (Mungall 2000) producing reproductively capable offspring lent credence to this concern. Sika deer escape from confinement in numerous states where individuals occasionally are taken during the hunting season. Genetic testing currently is done in several states to detect sika/elk hybrids. Disease transmission may occur between captive deer and livestock, and between captive deer and free-ranging wildlife populations adjacent to the fences (see Davidson and Crow 1983; Miller and Thorne 1993). There is potential crosstransmission of several diseases between elk/red deer and sika deer. Strict quarantine procedures in place today help mitigate transmission, but the potential for widespread problems exists. No disease in free-ranging or confined cervids currently has received greater attention than chronic wasting disease (CWD). Chronic wasting disease is a transmissible spongiform encephalopathy (TSE) that causes progressive, neurodegenerative disease that is always fatal. The etiological agent is believed to be prions. Prions are very unusual infectious agents because they may be both infectious and hereditary. They are small, modified cellular proteins without any nucleic acids. Other prion-caused TSEs include scrapie in domestic sheep and goats, bovine spongiform encephalopathy or “mad cow disease” in cattle and other hosts, and Creutzfeldt-Jacob disease in humans. Currently, white-tailed deer, mule deer (Odocoileus hemionus) and Rocky Mountain elk are the only cervids known to

636

G. A. Feldhamer, S. Demarais

contract CWD (Williams et al. 2002a). Distribution of CWD has spread substantially in the past decade in confined herds and free-ranging animals because of movement of infected, farmed elk. The implications of CWD obviously are significant for the cervid farming industry and for management of free-ranging populations. As such, surveillance and monitoring programs are in place in most jurisdictions, limiting or eliminating import of cervids from known foci of CWD. There has been a CWD surveillance, depopulation, and indemnity program for farmed elk in the United States since September 2001 with millions of dollars of funding. The U.S. Department of Agriculture Animal and Plant Health Inspection Service currently has 15 laboratories under contract assisting the National Veterinary Services Laboratories testing for CWD. In some states, such as Wisconsin, importation is banned unless the animals are from a herd monitored for at least five years. This amounts to a de facto ban because no herd can meet the standard. Thirty two states restrict importation of cervid carcasses from states and provinces positive for CWD (Arkansas Game and Fish Commission 2008), although boned meat is allowed. Interestingly, as noted by Williams et al. (2002a), several species of ungulates appear to be resistant to CWD, including moose (Alces alces), pronghorn antelope (Antelocapra americana), bighorn sheep (Ovis canadensis), and mountain goats (Oreamnos americanus). Nonetheless, because of the close phylogenetic relationship of sika deer and elk/red deer, regulations and restrictions in many states and provinces often specifically exclude importation of C. nippon (Table 41.1). As noted by Williams et al. (2002b:87–88), the future challenge of CWD will be balancing “complex and often competing and conflicting interests of the general public, sportspeople, the game farming industry, traditional livestock industries, and many state and federal animal and public health agencies.” The biological, social, and political aspects of privately held sika deer, within the larger context of exotic deer, are generating increasing controversy and legal challenges in North America. Regardless of whether the number of game farms and ranches continues to increase or remains stable, controversy probably will continue because the issues are complex. The private sector will promote the positive aspects of economic diversification while public wildlife management agencies, concerned with the potential negative aspects of all exotic deer in general and sika deer in particular, lobby for more restrictive regulations. Such regulatory limitations have increased dramatically in the past decade and this trend is likely to continue. They also are somewhat unique because “Although the rearing and marketing of these cervids is an agricultural activity, the process and potential consequences are inextricably linked to their wild counterparts, the wildlife management system, and the ecosystem upon which wildlife species depend” (Coon et al. 2002:262–3). Acknowledgements Thanks to Phyllis Menden and Brian Cahill, North American Deer Farmers Association for information on members raising sika deer. We thank the individuals from the departments of natural resources and/or agriculture from all states and provinces or territories who provided information on the occurrence within their jurisdiction of free-ranging sika deer or those on game farms or ranches. Special thanks to Doug Hotten and Brian Eyler, Maryland Department of Natural Resources, Tom Penn, Chincoteague National Wildlife Refuge, Brian Richards, Texas Parks and Wildlife Department, Scott Osborne, North Carolina Wildlife Resources Commission,

41 Free-Ranging and Confined Sika Deer in North America

637

and Jonathan Day, Kentucky Department Fish and Wildlife Resources. Figure 41.1 was produced courtesy of J. Kevin Davie, Southern Illinois University Carbondale Library. This is publication WF256 of the Forest and Wildlife Research Center, Mississippi State University.

Literature Cited Arkansas Game and Fish Commission. 2008. Carcass importation restrictions. www.agfc.com/ hunting/deer/programs-deer/cwd-monitoring-deer/carcass.aspx Accessed February 28, 2008. Asher, G. W., D. S. Gallagher, M. L. Tate, and C. Tedford. 1999. Hybridization between sika deer (Cervus nippon) and axis deer (Axis axis). Journal of Heredity 90:236–240. Baccus, J. T. 2002. Impacts of game ranching on wildlife management in Texas. Transactions of the North American Wildlife and Natural Resources Conference 67:276–288. Bartos, L., J. Hyanek, and J. Zironicky. 1981. Hybridization between red and sika deer. I. Craniological analysis. Zoologica Anzeiger Jena 207:260–270. Buck, J. M. 2002. Status and management implications of captive cervid farming in the northeast. Transactions of the North American Wildlife and Natural Resources Conference 67:297–307. Butts, G. L., M. J. Anderegg, W. E. Armstrong, D. E. Harmel, C. W. Ramsey, and S. H. Sorola. 1982. Food habits of five exotic ungulates on Kerr Wildlife Management Area, Texas. Texas Parks and Wildlife Department Technical Series No. 30. Center for Disease Control. 2008. Reported cases of Lyme Disease by year, United States, 1991– 2006. www.cdc.gov/ncidod/dvbid/lyme/. Accessed 4 March 2008. Christian, J. J., V. Flyger, and D. E. Davis. 1960. Factors in the mass mortality of a herd of sika deer, Cervus nippon. Chesapeake Science 1:79–95. Clutton-Brock, J. 1999. A natural history of domesticated animals, 2nd edition. Cambridge University Press, New York, New York, USA. Cook, C. E., Y. Wang, and G. Sensabaugh. 1999. A mitochondrial control region and cytochrome b phylogeny of sika deer (Cervus nippon) and report of tandem repeats in the control region. Molecular Phylogenetics and Evolution 12:47–56. Coon, T. G., H. Campa, A. B. Felix, R. B. Peyton, S. R. Winterstein, F. Lupi, M. Schulz, and J. Sikarskie. 2002. Farming captive cervids: A review of social, economic and ecological opportunities and risks in Michigan and North America. Transactions of the North American Wildlife and Natural Resources Conference 67:251–268. Crawford, T. B., H. Li, S. R. Rosenburg, R. W. Norhause, and M. M. Garner. 2002. Mural folliculitis and alopecia caused by infection with goat-associated malignant catarrhal fever virus in two sika deer. Journal of the American Veterinary Medical Association 221:843–847. Davidson, W. R., and C. B. Crow. 1983. Parasites, diseases, and health status of sympatric populations of sika deer and white-tailed deer in Maryland and Virginia. Journal of Wildlife Diseases 19:345–348. Dehority, B. A., S. Demarais, and D. A. Osborn. 1999. Rumen ciliates of white-tailed deer (Odocoileus virginianus), axis deer (Axis axis), sika deer (Cervus nippon) and fallow deer (Dama dama) from Texas. Journal of Eukaryotic Microbiology 46:125–131. Demarais, S., D. A. Osborn, and J. J. Jackley. 1990. Exotic big game: A controversial resource. Rangelands 12:121–125. Demarais, S., J. T. Baccus, and M. S. Traweek, Jr. 1998. Nonindigenous ungulates in Texas: Longterm population trends and possible competitive mechanisms. Transactions of the North American Wildlife and Natural Resources Conference 63:49–55. Demarais, S., J. J. Jackley, B. K. Strickland, and L. W. Varner. 2003. In vitro digestibility of forages by coexisting deer species in Texas. Texas Journal of Science 55:175–182. Diez, J. R., M. Gilsdorf, and R. Werge. 2002. The federal role in regulating alternative livestock operations. Transactions of the North American Wildlife and Natural Resources Conference 67:289–296.

638

G. A. Feldhamer, S. Demarais

Dodd, K. 1984. Tuberculosis in free-living deer. Veterinary Record 115:592–593. Ervin, R. T., S. Demarais, and D. A. Osborn. 1992. Legal status of exotic deer throughout the United States. Pages 244–252 in R. D. Brown, editor, The biology of deer. Springer, New York, New York, USA. Eyler, T. B. 2001. Habitat use and movements of sympatric sika deer (Cervus nippon) and whitetailed deer (Odocoileus virginianus) in Dorchester County, Maryland. Master’s thesis, University of Maryland Eastern Shore, Princess Anne, Maryland, USA. Feldhamer, G. A. 1985. Climatic factors and body weight of yearling sika deer. Mammalia 49:291–293. Feldhamer, G. A. 1995. Status of sika deer on game farms and game ranches in the United States. Pages 2.3:1–9 in E. Eick, R. König, and J. A. Willett, editors, Sika Cervus nippon Temminck, 1838. Vol. I. International Sika Society, Möhnesee, Germany. Feldhamer, G. A., and W. R. Brtalik. 1988. Female skull characteristics in two subspecies of sika deer. Mammalia 52:410–414. Feldhamer, G. A., and W. E. Armstrong. 1993. Interspecific competition between four exotic species and native artiodactyls in the United States. Transactions of the North American Wildlife and Natural Resources Conference 58:468–478. Feldhamer, G. A., and M. A. Marcus. 1994. Reproductive performance of female sika deer in Maryland. Journal of Wildlife Management 58:670–673. Feldhamer, G. A., J. A. Chapman, and R. L. Miller. 1978. Sika deer and white-tailed deer on Maryland’s eastern shore. Wildlife Society Bulletin 6:155–157. Feldhamer, G. A., R. P. Morgan, P. E. McKeown, and J. A. Chapman. 1982. Lack of polymorphism in liver and muscle enzymes from sika deer (Cervus nippon). Journal of Mammalogy 63:512–514. Feldhamer, G. A., J. R. Stauffer, and J. A. Chapman. 1985. Body morphology and weight relationships of sika deer in Maryland. Zeitschrift für Säugetierkunde 50:88–106. Flyger, V. 1960. Sika deer on islands in Maryland and Virginia. Journal of Mammalogy 41:140. Flyger, V., and J. Warren. 1958. Sika deer in Maryland – An additional big game animal or a possible pest. Proceedings of the Annual Conference of the Southeastern Association of Game and Fish Commissioners 12:209–211. Flyger, V., and N. W. Davis. 1964. Distribution of sika deer (Cervus nippon) in Maryland and Virginia in 1962. Chesapeake Science 5:212–213. Geist, V. 1985. Game ranching: Threat to wildlife conservation in North America. Wildlife Society Bulletin 13:594–598. Geist, V. 1988. How markets in wildlife meat and parts, and the sale of hunting privileges, jeopardize wildlife conservation. Conservation Biology 2:15–26. Geist, V. 2002. Why game ranching and the North American system of wildlife conservation are incompatible. Transactions of the North American Wildlife and Natural Resources Conference 67:269–275. Goodman, S. J., N. H. Barton, G. Swanson, K. Abernethy, and J. M. Pemberton. 1999. Introgression through rare hybridization: A genetic study of a hybrid zone between red and sika deer (Genus Cervus) in Argyll, Scotland. Genetics 122:355–371. Gustavsson, I., and C. O. Sundt. 1969. Three polymorphic chromosome systems of centric fusion type in a population of Manchurian sika deer (Cervus nippon hortulorum Swinhoe). Chromosoma 28:245–254. Henske, S. E., S. Demarais, and J. A. Pfister. 1988. Digestive capacity and diets of white-tailed deer and exotic ruminants. Journal of Wildlife Management 52:595–598. Hofmann, R. R. 1985. Digestive physiology of the deer – Their morphophysiological specialisation and adaptation. Pages 393–407 in P. F. Fennessy and K. R. Drew, editors. Biology of deer production. Royal Society of New Zealand Bulletin 22. Wellington, New Zealand. Howery, L. D., J. A. Pfister, and S. Demarais. 1989. Seasonal reproductive activity of 4 exotic ungulates in Texas. Journal of Wildlife Management 53:613–617.

41 Free-Ranging and Confined Sika Deer in North America

639

Imai, S., M. Matsumoto, A. Watanabe, and H. Sato. 1993. Rumen ciliate protozoa in Japanese sika deer (Cervus nippon centralis). Animal Science and Technology 64:578–583. Jackley, J. J. 1991. Dietary overlap among axis, fallow, sika and white-tailed deer in the Edwards Plateau region of Texas. Master’s Thesis, Texas Tech University, Lubbock, Texas, USA. Kahn, R. 1993. Wildlife management agency concerns about captive wildlife: The Colorado experience. Transactions of the North American Wildlife and Natural Resources Conference 58:495–503. Keel, M. K., J. G. Patterson, T. H. Noon, G. A. Bradley, and J. K. Collins. 2003. Caprine herpesvirus-2 in association with naturally occurring malignant catarrhal fever in captive sika deer (Cervus nippon). Journal of Veterinary Diagnostic Investigation 15:179–183. Keiper, R.R. 1990. Biology of large grazing mammals on the Virginia Barrier Islands. Virginia Journal of Science 41:352–363. Kelley, J. A. 1970. Food habits of four exotic big-game animals on a Texas “Hill Country” ranch. Master’s thesis, Texas A& I University, Kingsville, Texas, USA. Kirkpatrick, J. F., P. P. Calle, P. Kalk, I. K. M. Liu, and J. W. Turner. 1996. Immunocontraception of captive exotic species. 2. Formosan sika deer (Cervus nippon taiouanus), axis deer (Cervus axis), Himalayan tahr (Hemitragus jemlahicus), Roosevelt elk (Cervus elaphus roosevelti), Reeve’s muntjac (Muntiacus reevesi), and sambar deer (Cervus unicolor). Journal of Zoo and Wildlife Medicine 27:482–495. Koulischer, L., J. Tyskens, and J. Mortelmans. 1972. Mammalian cytogenetics. VII. The chromosomes of Cervus canadensis, Elaphurus davidianus, Cervus nippon (Temm.) and Pudu pudu. Acta Zoologica Pathologica 56:25–30. Lever, C. 1985. Naturalized mammals of the world. Longman, London, United Kingdom. Lowe, V. P. W. and A. S. Gardiner. 1975. Hybridization between red deer (Cervus elaphus) and sika deer (Cervus nippon) with particular reference to stocks in N. W. England. Journal of Zoology 177:553–566. Markov, G., A. Danilkin, and G. B. Hartl. 1992. Lack of biochemical-genetic variation in native sika deer (Cervus nippon hortulorum) from the far east of the Asian continent. Zeitschrift für Säugetierkunde 57:118–119. Matsumoto, M., H. Nishinakagawa, and J. Otsuka. 1984. Morphometrical study on the skull of Cervus pulchellus, Cervus nippon megeshimae and Cervus nippon yakushimae. Journal of the Mammalogical Society of Japan 10:41–53. Miller, K. V., L. I. Muller, and S. Demarais. 2003. White-tailed deer, Odocoileus virginianus. Pages 906–930 in G. A. Feldhamer, B. C. Thompson, and J. A. Chapman, editors, Wild mammals of North America: Biology, management, and conservation. Johns Hopkins University Press, Baltimore, Maryland, USA. Miller, M. W., and E. T. Thorne. 1993. Captive cervids as potential sources of disease for North America’s wild cervid populations: Avenues, implications and preventive management. Transactions of the North American Wildlife and Natural Resources Conference 58:460–467. Mirsky, M. L., D. Morton, J. W. Piehl, and H. Gelberg. 1992. Mycobacterium bovis infection in a captive herd of sika deer. Journal of the American Veterinary Medicine Association 200:1540–1542. Mullan, J. M., G. A. Feldhamer, and D. Morton. 1988. Reproductive characteristics of female sika deer in Maryland and Virginia. Journal of Mammalogy 69:388–389. Mungall, E. C. 2000. Exotics. Pages 736–764 in S. Demarais and P. R. Krausman, editors, Ecology and management of large mammals in North America. Prentice-Hall, Saddle River, New Jersey, USA. Mungall, E. C., and W. J. Sheffield. 1994. Exotics on the range: The Texas example. Texas A & M University Press, College Station, Texas, USA. Nagata, J., R. Masuda, H. B. Tamate, S. Hamasaki, K. Ochiai, M. Asada, S. Tatsuzawa, K. Suda, H. Tado, and M. C. Yoshida. 1999. Two genetically distinct lineages of the sika deer, Cervus nippon, in Japanese islands: Comparison of mitochondrial D-loop region sequences. Molecular Phylogenetics and Evolution 13:511–519.

640

G. A. Feldhamer, S. Demarais

Ohtaishi, N. 1976. Developmental variation of the antlers of Japanese sika deer at Nara Park (Preliminary). Nara City: Kasuga Fund, pp. 107–128, Nara, Japan. Oliver, J. H., L. A. Magnarelli, H. J. Hutcheson, and J. F. Anderson. 1999. Ticks and antibodies to Borrelia burgdorferi from mammals at Cape Hatteras, NC and Assateague Island, MD and VA. Journal of Medical Entomology 36:578–587. Oregon Department of Fish and Wildlife. 1992. Private holding of cervids. White Paper, Oregon Wildlife Commission, Portland, Oregon, USA. Osborn, D. A., S. Demarais, and R. T. Ervin. 1995. Weight estimation for axis, fallow, sika and white-tailed deer in Texas. Texas Journal of Science 47:287–294. Peyton, R. B. 2000. Wildlife management: Cropping to manage or managing to crop? Wildlife Society Bulletin 28:774–779. Pimentel, D., L. Lach, R. Zuniga, and D. Morrison. 2000. Environmental and economic costs of nonindigenous species in the United States. BioScience 50:53–65. Randi, E., N. Mucci, F. Claro-Hergueta, A. Bonnet, and E. J. P. Douzery. 2001. A mitochondrial DNA control region phylogeny of the Cervinae: Speciation in Cervus and implications for conservation. Animal Conservation 4:1–11. Raphael, B. L., P. Kalk, P. Thomas, P. P. Callie, J. G. Doherty, and R. A. Cook. 2003. Use of melengestrol acetate in feed for contraception in herds of captive ungulates. Zoo Biology 22:455–463. Renecker, L. A. 1989. Overview of game ranching in Canada. Pages 47–62 in R. Valdez, editor, Proceedings of the first international wildlife ranching symposium. New Mexico, State University, Las Cruces, New Mexico, USA. Rhyan, J. C., and D. A. Saari. 1995. A comparative study of the histopathologic features of bovine tuberculosis in cattle, fallow deer (Dama dama), sika deer (Cervus nippon), and red deer and elk (Cervus elaphus). Veterinary Pathology 32:215–220. Richardson, M. L., and S. Demarais. 1992. Parasites and condition of coexisting populations of white-tailed and exotic deer in south-central Texas. Journal of Wildlife Diseases 28:485–489. Samuel, W. M., and S. Demarais. 1993. Conservation challenges concerning wildlife farming and ranching in North America. Transactions North American Wildlife and Natural Resources Conference 58:445–447. Sanford, S. E., P. B. Little, and W. A. Rapley. 1977. The gross and histopathologic lesions of malignant catarrhal fever in three captive sika deer (Cervus nippon) in southern Ontario. Journal of Wildlife Diseases 18:29–32. Shah, K. V., G. B. Schaller, V. Flyger, and C. M. Herman. 1965. Antibodies to Myxovirus parainfluenza 3 in sera of wild deer. Bulletin of the Wildlife Disease Association 1:31–32. Simberloff, D. 1997. The biology of invasions. Pages 3–17 in D. Simberloff, D. C. Schmitz, and T. C. Brown, editors, Strangers in paradise: Impact and management of nonindigenous species in Florida. Island Press, Washington, DC, USA. Sybert, V. L. 1990. A study of the ciliated protozoa of white-tailed deer (Odocoileus virginianus), axis deer (Axis axis), and sika deer (Cervus nippon) in the Edwards Plateau of central Texas. Master’s thesis, Southwest Texas State University, San Marcos, Texas, USA. (Cited in Dehority et al. 1999.) Takatsuki, S., and Y. Matsuura. 2000. Higher mortality of smaller sika deer fawns. Ecological Research 15:327–240. Tamate, H. B., and T. Tsuchiya. 1995. Mitochondrial-DNA polymorphism in subspecies of the Japanese sika deer, Cervus nippon. Journal of Heredity 86:211–215. Teer, J. G. 1989. Commercial utilization of wildlife resources: Can we afford it? Pages 1–7 in R. Valdez, editor, Proceedings of the first international wildlife ranching symposium. New Mexico State University, Las Cruces, New Mexico, USA. Teer, J. G. 1991. Non-native large ungulates in North America. Pages 55–66 in L. A. Renecker and R. J. Hudson editors, Wildlife production: Conservation and sustainable development. Agriculture and Forestry Experiment Station Miscellaneous Publication 91–6, University of Alaska, Fairbanks, Alaska, USA.

41 Free-Ranging and Confined Sika Deer in North America

641

Teer, J. G. 2003. Non-native large mammals in North America. Pages 1180–1187 in G. A. Feldhamer, B. C. Thompson, and J. A. Chapman, editors, Wild mammals of North America: Biology, management, and conservation. Johns Hopkins University Press, Baltimore, Maryland, USA. Teer, J. G., L. A. Renecker, and R. J. Hudson. 1993. Overview of wildlife farming and ranching in North America. Transactions of the North American Wildlife and Natural Resources Conference 58:448–459. Traweek, M. S. 1995. Statewide census of exotic big game animals. Progress Report Federal Aid Project W-127-R-3 No. 21, Texas Parks and Wildlife Department, Austin, Texas, USA. van Tuinen, P., T. J. Robinson, and G. A. Feldhamer. 1983. Chromosome banding and NOR location in sika deer. Journal of Heredity 74:473–474. Wheaton, C., M. Phybus, and K. Blakely. 1993. Agency perspectives on private ownership of wildlife in the United States and Canada. Transactions of the North American Wildlife and Natural Resources Conference 58:487–494. White, R. J. 2000. Big game ranching. Pages 260–276 in S. Demarais and P. R. Krausman, editors, Ecology and management of large mammals in North America. Prentice-Hall, Saddle River, New Jersey, USA. White, R. J., and R. Valdez. 1991. Benefits of wildlife ranching in the United States. Pages 40–45 in L. A. Renecker and R. J. Hudson editors, Wildlife production: Conservation and sustainable development. Agriculture and Forestry Experiment Station Miscellaneous Publication 91–6. University of Alaska, Fairbanks, Alaska, USA. Williams, E. S., M. W. Miller, T. J. Kreeger, R. H. Kahn, and E. T. Thorne. 2002a. Chronic wasting disease of deer and elk: A review with recommendations for management. Journal of Wildlife Management 66:551–563. Williams, E. S., M. W. Miller, and E. T. Thorne. 2002b. Chronic wasting disease: Implications and challenges for wildlife managers. Transactions of the North American Wildlife and Natural Resources Conference 67:87–103. Yorks, T. P. 1989. Ranching native and exotic ungulates in the United States. Pages 268–285 in R. J. Hudson, K. R. Drew, and L. M. Baskin, editors, Wildlife production systems: Economic utilisation of wild ungulates. Cambridge University Press, New York, New York, USA.

Chapter 42

The Sika in New Zealand D. Bruce Banwell

Abstract The first introduction of sika into New Zealand took place in 1885, but they failed to establish themselves in the open, snow-grass mountain ranges of the South Island. It appears another lot, not officially recorded, arrived around 1900, but was not released to the wild, and probably retained in one of the zoological enclosures. A gift of six animals to the New Zealand Government by the 11th Duke of Bedford in 1904 were eventually liberated in January 1905 in the Poronui area of the Kaimanawa Ranges that lie to the east of the North Island’s Lake Taupo. This introduction was successful, and the herd now occupies a considerable range. It is believed the origin of these animals was of mixed genetics, involving at least two subspecies, perhaps as many as five, the number of races present at Woburn Abbey Park at the time of the presentation. Visual characteristics tend to suggest this to be the case. Once established, the herd’s impact and relationship with the introduced European red deer (Cervus elaphus) in the same area has created some interesting observations and theories concerning the co-existing processes. Introduced with the view to sporting purposes, the stags of the herd have produced some outstanding sets of antlers, comparable with any sikine trophies from elsewhere in the world.

Sika Introduction to New Zealand During early days of British settlement in New Zealand, acclimatization societies were responsible for the import and establishment of mammals and birds, most of which were imported from the “Homeland.” New Zealand had no indigenous land mammals other than the native bats. The following account for sika is lifted primarily from Banwell (1999). The first sika to be introduced to New Zealand was in 1885 when three animals consisting of a male and two females were imported by John Bathgate, of Dunedin City and presented to the Otago Acclimatisation Society. According to Mavis Davidson (1998 personal communication), a New Zealand biologist who carried out a long-term study of the local sika, these animals were claimed to have come D. R. McCullough et al. (eds.), Sika Deer: Biology and Management of Native and Introduced Populations, DOI: 10.1007/978-4-431-09429-6_42, © Springer 2009

643

644

D. B. Banwell

from the Ussuri district of eastern Siberia and were recorded as survivors of a turbulent voyage from Manchuria on board a ship, the Tamsui. This information suggests they were probably of the Dybowski race, Cervus nippon hortulorum. Liberation of these deer took place in the South Island on a property known as the Otekaieke Estate near the small township of Kurow, situated in North Otago’s Waitaki Valley. By 1888 the group had increased to seven animals and by 1890 the local newspapers were stating their numbers as increasing, no doubt following along the lines of the Otago Acclimatisation Society’s report that they “…were doing well and growing into a nice little herd.” However, after 1892 little or nothing was heard of the Otekaieke deer. It seems that in time, the increasing population became a nuisance to local farmers. The animals lived in a very open environment and, as a consequence, were eventually wiped out by the local settlers. As Mavis Davidson (1973) later remarked, it appears strange that the Society, which presumably had acquired the animals for recreational reasons, failed to monitor the progress of the herd very closely in relation to its welfare, herd structure, and trophy antler potential. As she pointed out at the time, there is little doubt that someone was aware of their fate, but the threat of prosecution created a conspiracy of silence. Davidson (1973) then went on to state, “There are other possibilities as to the fate of at least some of these sika, the most likely scenario being of dispersal westwards. The main factors influencing dispersal are considered to be population pressures, or the innate urge of animals to leave their area of origin on reaching puberty. As it is very unlikely the Otekaieke sika ever came under any population pressures, any movement away from the region of origin would appear to be due to innate dispersal behavior, perhaps accelerated by local settler harassment.” Assuming they did, in fact, disperse, they could have then entered a hostile environment. The Otekaieke deer may not have thrived under such conditions above and away from the area of liberation, the region being quite mountainous and subject to adverse weather conditions, particularly during winter. Both Russian (Makovkin 1999) and Japanese (Takatsuki 1991) biologists agree with research carried out here in New Zealand, that sika tend to be predominantly grazers, the Russians proposing a steppe, or grass plain, as place of origin (Flerov 1952). In Far East Russia sika congregate in snow-free areas, unable to cope very well with heavy snow conditions (Makovkin 1999). Even in New Zealand’s Kaweka Range, where the successfully introduced sika range today, it is claimed (Davidson 1973), that they migrate to lower reaches when the snow covers the “tops,” or alternatively, to high elevation grasslands. During the 1890s, human population to the west of the release area was rather sparse and a few deer could have escaped detection (Banwell 1999). Alternatively, a sighting would, undoubtedly, have inspired considerable comment. If perhaps, the log books of the old sheep stations behind the North Otago townships of Oamaru and Kurow had survived, it may have been possible to find some reference to the animals in question. In any event, they seemed to have disappeared not many years after their release. It was after the turn of the century when sika were once again liberated in New Zealand (Donne 1924). In 1904, some six years after their importation into Woburn

42 The Sika in New Zealand

645

Abbey, the then Duke of Bedford presented six head to the New Zealand government. At the time of the gifting it was claimed the six animals belonged to the Manchurian race, Cervus nippon mantchuricus. The genetic background of these six deer introduced to New Zealand has created great interest (Banwell 1995, 1999). Their descendants display a great diversity in morphological characteristics. However, sika are phenotypically variable and capable of producing quite a wide range of coloration and spotting, even within one particular subspecies, let alone in a group evolving from a mixture of races. A study of the various introductions and relative variations in those races enclosed at Woburn Abbey, seat of the Dukes of Bedford, at the time in question and the stories that evidently circulated following the liberation of the animals in New Zealand, reveals some interesting facts (Glover 1956; Banwell 1999). In the manifest of deer held at Woburn Abbey during 1903, the year before the six animals were shipped to New Zealand, there were 106 “Japanese,” 34 Manchurian, and 12 Formosan (Taiwan) sika in the park, making a total of 152 enclosed there. By 1904 the total had increased to 171. “Japanese” sika are recorded as having been acquisitioned by Woburn Abbey during 1894 from several different animal dealers, one from a Captain Marshall, five from a company calling itself W. Jamrach, two from a man named Cross, and one from the London Zoo, making a total import constituting two males and seven females. Later that same year further acquisitions were effected from the property of Lord Powerscourt in Ireland. Powerscourt had been credited with having imported the first sika into Britain (as a matter of fact, the first into Europe) as early as 1860. These he had enclosed in his Powerscourt Park in Ireland’s County Wicklow (Whitehead 1964). This first import, together with another consisting of a male and a female introduced to Regent’s Park Zoological Gardens, are the earliest recorded introductions of sika into the European arena. “Manchurian” sika arrived at Woburn Abbey during August 1894, one male and two females from the London-based dealers, W. Jamrach. By October of that year mortality had reduced the total to one individual. Nevertheless, by the end of 1897 the group had increased to ten, including three animals of the Formosan subspecies, by purchase of animals from a German firm, Carl Hagenbech. It would appear the retention of records was rather imperfect over this particular period, as other sources of information reveal that there were small numbers of both Kerama, Cervus nippon keramae, and Dybowski’s sika, Cervus nippon hortulorum, in the park as well (Whitehead 1972; Banwell 1995, 1999). It appears there were, in fact, at least five subspecies of sika at Woburn Abbey at the time of the delivery of the shipment to New Zealand and there is no evidence they had been separately enclosed (Glover 1956). Alternative sources of information, including Lydekker (1901) and Whitehead (1950), confirm the presence of the five forms. However, New Zealand historian, T. E. Donne (1924) presented the facts quite differently. As a consequence of this conflicting information, one can not separate fact from conjecture. This confused situation has led to the widely accepted conclusion that the New Zealand herd has arisen from a mixture of forms, claimed to be from a minimum of two and the possibility of all five subspecies present at Woburn at the time

646

D. B. Banwell

(Davidson 1990; Banwell 1999). The least likely contributor is the Kerama, a diminutive form from the Ryukyu Islands, which can offer a considerable influence of total melanism. Nevertheless, it has been claimed that smaller, black animals have been sighted within the New Zealand sika range. Whatever the case, it is generally accepted that the sika in this country are of mixed origin, most certainly influenced by genetics from both sides of the Sea of Japan, as is clearly evident in morphological and color variations, including antler configuration. The genetic background of the New Zealand sika is certainly an interesting subject and it is to be hoped that someday in the not too distant future, by the application of DNA, the true genetic sources will be determined. The animals consigned to the New Zealand government from Woburn Abbey in 1904 comprised three of each sex and were shipped in five crates out of the London Docks on the S. S. Kaikoura under the command of Captain Reginald C. Clifford (Donne 1924). The sea journey took 70 days over which time they were fed on clover hay and carrots. Two calves were born at sea, only one surviving the journey. The ship docked at the Port of Wellington during early September 1904, where the animals were immediately placed in quarantine on Somes Island in Wellington Harbor. There they were to remain for the statutory six months. In the New Zealand Times of January 18, 1905 it was reported, “…the half dozen pretty little Japanese deer that have been quarantined for the past six months on Somes Island, have at length obtained their release. The Tourist Department has dispatched them north and they are to be liberated on the Kaimanawa Range, about 15 miles south of Lake Taupo. The range is heavily timbered with open valleys and good water.” Because the deer arrived in early September and were released in January it would seem the animals were excused from the full six months of quarantine. The liberation took place in January 1905 at a place known as Merrylees Clearing (Donne 1924), part of which was then Taharua Station, a large farm property known today as Poronui (Fig. 42.1). It is situated on the eastern periphery of the North Island’s Kaimanawa Forest Park. An unpublished Tourist Department report of events between 1900 and 1907 states that release of the three females and one male animal was successfully accomplished; two males were “lost” in the operation. By the word “lost,” it can only be assumed they were killed or died of stress during the exercise. Mention was also made of two calves at foot the following year and the fact the “little hind” had failed to conceive that season. She almost certainly bred the following season and was undoubtedly the major key to the presence of the obvious “Japanese” element in the herd structure today (Banwell 1999). Little information concerning the herd was recorded in the period between the release and when the first licenses to hunt were issued in 1925, particularly in regard to establishment and dispersal. In fact, this was the case up until protection of all introduced game animals was lifted in 1930, when the New Zealand Government saw fit to declare all such introduced species as “pests.” Nevertheless, by 1930 the herd numbers had increased considerably and Davidson (1973) recorded an estimated dispersal rate of 0.6 km per year, assessed over the period of the initial 65 years. The rate of spread was probably underestimated in the early

42 The Sika in New Zealand

647

Fig. 42.1 Map of the North Island of New Zealand showing the site of release and the recent (as of 1998) distribution of sika in the wild.

years of establishment and ultimately slowed down over later years. Current distribution is shown in Fig. 42.1. Obviously, some of the animals remained in and around the area of the initial release, but before very long, as was anticipated, a proportion had spread to the north and west towards Tongariro National Park. An interesting outcome in the establishment of this species is a claim relating to the fact that two distinctive types of sika now exist in specific areas of the herd’s present range. There are those displaying the characteristics that suggest they are predominantly of a Chinese form, while in some pockets of the overall range, others distinctly resemble a Japanese form. The majority of the latter appear to have moved further eastwards towards the Hawkes Bay side of the range, while the Chinese characteristics seem more evident nearer the point of liberation and throughout the Kaimanawa Ranges. This may well suggest the Japanese form to be more aggressive and, as a result, dispersed more rapidly onto fresh domain.

648

D. B. Banwell

Labeled as a “pest” in New Zealand, similar to all introduced fauna, regardless of its intrinsic value as a source of recreation, there have been several illegal translocations effected in quite recent years, some far beyond the natural spread and establishment of the main herd. These translocations threaten the purity of some of our red deer (Cervus elaphus), although it does not appear that hybridization is as widespread as some advocates claim. Nevertheless, such releases have potential grave consequences.

Current Distribution The spread and establishment of the original sika herd has not been as extensive as was first anticipated. During the years of 1929 and 1930, J. G. Holmes of the Tourist Department at the City of Rotorua, conducted a survey in which he set out to determine the status of axis deer (Axis axis) that had been unsuccessfully introduced to Tongariro National Park. He subsequently submitted a report during April of each year recording that his efforts proved to be unsuccessful in regard to the axis, but made mention of sika as having spread at least as far as the “bald tops” of the Kaimanawa Range by way of a track cut between the Waiotaka stream and the Tongariro River and had traveled along these open tops above the forest to a point past the Waipakihi stream. He had also surveyed the country above Tokaanu. Although it was the period of the rut, he failed to report about sika behavior, or trophy antler potential. His only comment was in relation to the fact that not all sportsmen going after the sika were licensed to do so, some claiming to be employed as camp cooks, but who were armed with a rifle to hunt deer. The areas surveyed by Holmes are located to the east and south of Lake Taupo. According to Mavis Davidson (1973), contrary to expectations, the sika had dispersed in different directions about the time they had been sighted near Lake Taupo. By that time they had reached the Kaweka Range to the south-east. The Hartree and Whittle families, living in that area, were probably the first people to be aware of the animals in that particular region. A member of the Whittle family was in possession of a set of antlers from a stag killed in 1934 or 1935. R. H. Hartree encountered a sika stag on Turangakumu, a high point on the Ahimanawa Ranges to the east during the 1939 through ’45 war years, and a government field officer, Ron Fraser, recorded the shooting of a sika stag in the vicinity of Lake Waikaremoana in 1939, obviously a wanderer. During 1938, Evan Wilson, a South Island-based government hunter, doing one of three stints in the Kaweka Ranges, discovered a small herd of sika on the subsidiary Black Birch Range. He assessed the group as comprising a “very handsome stag” and two male subordinates. This central North Island herd has, over the hundred-odd years since the initial introduction, established itself over a large range of habitat, incorporating the Kaweka, Kaimanawa, and Ahimanawa Ranges and has, in more recent years, occupied areas beyond this tract of North Island high-country (Fig. 42.1). According to Cam Speedy (personal communication 1998), who has made a study of the sika of

42 The Sika in New Zealand

649

the area and on whom I was obliged to depend while compiling the history of the herd as far as the extent of the range today, the natural dispersal has slowed down quite considerably over the past 20 years. Nevertheless, expansion due to illegal and unofficial liberations elsewhere continues to see sika springing up well away from their feral range. Evidently, there are something like 14 established populations in various parts of New Zealand today, including at least two in the South Island. In the opinion of the writer, the introduction of this species into the South Island is irresponsible and, as already stated, puts red deer herds at risk, particularly the unique Otago herd, perhaps the sole surviving population of pure indigenous British red deer (C. e. scoticus) anywhere in the world!

Habitat Selection In the central North Island, the range of the sika is mainly restricted to the central hill country with its poorer volcanic soils and resulting quality of vegetation offered as habitat. Cam Speedy (personal communication 1998) believes, that to the north of the established sika range, on the Kaingaroa Plains, ease of access and excessive hunting pressures in a huge area of exotic pine forest, has reduced the sika’s ability to effectively disperse onto new country. According to him game density appears to remain low throughout a good deal of this area and as a consequence, habitat quality is high, thus allowing red deer to remain competitive, despite poor, mineral-deficient soils. To the north-east, around the area along the highway between the city of Napier and Lake Taupo, the superior forest may also have provided red deer with an environment in which they could effectively compete. In this same region, sika managed to establish themselves where the forest has been previously felled, eventually replaced by scrub. To date, the sika has not advanced significantly to the north-east towards the heavily forested Urewera area, where other species such as rusa (Cervus timorensis) and red deer would pose a threat to its co-existence. Competition with red deer as a factor in the spread of sika will be discussed further in a later section. In 1992, John Mason, of the Pureora Field Centre of the Department of Conservation, informed me by letter that occasional reports of sika having been shot or sighted within the confines of the Pureora Forest Park have filtered into his office since the mid-1970s. It is believed these animals were the result of illegal liberation. Mason stated that it was interesting why they have failed to colonize the Pureora Forest Park to the west of Lake Taupo, because geographically, there is little to prevent them doing so. At the time of writing, there were reasonable numbers residing in the area known as the Tongariro National Park to the south-west, and it is a relatively short distance from there to Pureora. The northern Ruahine Ranges to the south of the established sika range also appear to present a natural barrier. Cam Speedy believes their lack of ability to displace resident red deer from that area has resulted in minimal dispersal in that direction. According to Speedy, the higher rainfall of the Ruahine Ranges may well present a natural barrier. Consequently, he suspects the key to be possibly climate and soil chemistry.

650

D. B. Banwell

The limestone outcrops along the highway between the city of Napier and Lake Taupo area, together with the local weather patterns, present red deer with good feeding conditions, thus allowing them to remain competitive with the sika. Although this particular area allows sika to grow superior antlers, for some reason they have failed to make significant progress in dispersal in that direction over the past 20 years or more. To the east of the original area of liberation, pastoral farming prevents sika from spreading any further in that direction. To the west of the current range lies Lake Taupo and farm development around the lake and the significant township of Taupo. Collectively, they have prevented dispersal in that general direction. South of the lake, however, there exists a different situation. Here sika have established themselves on the western side of the Desert Road, or State Highway No. 1, which passes through that area and have been there for a number of years. They have occupied an exotic pine forest known as Karori and the adjacent native forest of Rangataua on the south-eastern side of Mount Ruapehu, a well-known volcano in the district. They have been in that locality for some considerable time. In more recent years they have moved around the southern slopes of Ruapehu as far as Horopito, some having been sighted on the western side of State Highway 4 in the Eura Forest during 1995. In these areas, because of the domination of the resident red deer, sika are found only in small pockets. Sika appear to have done very well in the manuka scrub (Leptospernum scoparium) country of northern and eastern Tongariro National Park and have been sighted as far west as Mount Hauhungatahi on the western side of Mount Ruapehu, for at least 20 years. They are, nevertheless, rare on the ranges south-west of Turangi, the country lying between Lakes Taupo and Rotoaira, being podocarp and hardwood forest. Cam Speedy records that the papa country (a blue soft rock) appears to stop their spread. Here also, red deer competition is strong with ample feed available, thus affording them the opportunity of remaining the dominant species. In total, sika deer currently occupy some 8,000 km2 of the central North Island higher country. It is believed this will not change significantly by natural dispersal unless more illegal liberations take place, or red deer competition is lowered by drastic reduction of their population levels, or their elimination by hunting. So, it appears that established red deer populations in their preferred habitat can play a major role in restricting the further dispersal of sika onto new ground and act as an effective controlling element.

Competition with Red Deer As has already been established elsewhere in the world, sika can flourish in country where other species of deer fail. In New Zealand their competition with red deer has proved successful, eventually displacing the latter by forcing them to seek more compatible rangeland, in particular the higher altitude country. The sika and wapiti evolved from a common ancestor (Nagata chapter 3), but when the two closely related species have attempted to share common ground it is not always in harmony (Kiddie 1962; Challies 1985).

42 The Sika in New Zealand

651

Perhaps the answer lies in the theory put forward by Cam Speedy that vegetation competition is the most significant factor influencing sika dispersal and establishment in new areas. Sika appear to be more adept at surviving in beech forest (Nothofagus), their digestive anatomy allowing them to be more competitive in such an environment than is the case with the red deer. It seems sika lose the advantage in podocarp forest (kahikatea, white pine (Podocarpus dacrydioides), totara (Podocarpus totara), matai, black pine (Podocarpus spicatus), miro (Podocarpus ferrugineus) ) and hardwood (northern rata (Metrosideros robusta) ) where the red deer can compete on a more equal footing. Taking this factor into consideration, it is not surprising that sika have not spread into the Pureora Forest at the anticipated rate. From what I have discovered in regard to “forest” in their native lands, it does not surprise me that such factors here in New Zealand have such an effect on selected habitat. The open forest of northern China, for example, is much more like our open beech forest than is our heavy podocarp forest. As already mentioned, sika in New Zealand are predominantly grazers, depending mainly on the forest for cover over much of the year. The displacement of red deer by sika in New Zealand is similar to the displacement of wapiti reported by Makovkin (1999) in Primorsky Krai of the Russian Far East. It also fits in with the observation by Kiddie (1962) that the upper range of elevations occupied by sika is limited where red deer are better adapted to high elevation habitats, but in mountains where red deer are absent, sika successfully occupy these habitats. Most authors agree, that in habitats most suited to sika, they outcompete red deer by virtue of their ability to thrive on heavily fed-upon vegetation. What role behavioral dominance plays in competition between the two species is unknown, but could be relevant. Kiddie (1962) reported on a fight between stags of the two species that was won by the much smaller sika stag. Whereas the red deer successfully “bulldozed” the smaller sika, the latter’s aggressiveness and agility allowed it to parry and thrust against its larger opponent, thus delivering the more telling antler stabs. Nevertheless, the writer disagrees with this generalization, having witnessed the process both in the wild and in the seclusion of an English park. The rutting seasons of both species overlap, the red deer beginning a little earlier than that of the sika. As a consequence, later in the red deer rut, the stag is tired and reluctant to get involved with the aggressiveness of his smaller counterpart which, by that time is full on. It is the opinion of the writer, that initial hybridization between the two species is initiated by the sika stag mating with a young red deer hind which has missed conception in her early cycles during the red deer rut.

Hybridization with Red Deer Perhaps the most controversial subject that arises in relation to the sika in New Zealand is the subject of hybridization with red deer. Unfortunately, a considerable amount of this information is expounded by unqualified observers, and up to this

652

D. B. Banwell

time, no genetic studies incorporating DNA techniques have been done. Evidence available confirms hybridization certainly occurs (Challies 1985), but not to the levels some commentators would have us believe. Unusual characteristics, such as malformed antler configuration, always attract such opinion, much of which is misleading about causality. One needs to take into account the morphological variability that exists in pure sika. When a first-class sika trophy is taken, the question is inevitably asked if its large size is due to introgression of red deer genes. In fact, judging on morphological grounds, hybridization tends to be quite rare in the established sikine areas, most convincing reports coming from the regions along the periphery of the accepted range where a greater proportion of red deer exist. This result is consistent with the results of Swanson and Putman (chapter 40) that earlier dispersers in the British Isles are most likely to interbreed. Clear cases of hybridization are usually easily recognized, with several obvious signs apparent in the progeny. Nevertheless, Kiddie (1962) stated that of the 800–1,000 sika he shot over a period of many years, only four animals displayed any evidence of hybridization or, alternatively, created some suspicion. Mavis Davidson (1973) was of the same opinion and supported Kiddie’s findings. Still, it makes sense that the proportion of questionable animals would possibly increase as the years pass by and it should be appreciated that the findings of Kiddie and Davidson were based on observations from 30 to 40 years ago. However, the two species tend to retain their phenotypic identity, and perhaps due to niche separation or assortative mating, as with hybridization in the British Isles (Swanson and Putman chapter 40) they maintain a typical appearance of sika or red deer despite past hybridization, although the sikine characteristics appear to be dominant in most cases. A genetic study using DNA is needed to answer these questions.

Adaptation of Sika Phenotypes to Local Conditions In describing the visual characteristics of the New Zealand sika, an acute problem immediately arises because of the mixed origin of the herd. Over the many years they have occupied New Zealand, however, some interesting consistent patterns over the distributional range of sika in the central North Island have emerged. It is obvious there are pockets of animals that clearly reflect the different characteristics of either the Chinese, or the Japanese types (phenotypes representative of northern and southern subspecies), or in some instances intermixing of both. For example, to the north and east of the existing range towards the Hawke’s Bay side, there is a tendency for the animals to show a predominance of Japanese features, while towards Lake Taupo, or the Kaimanawa Range in the west, Chinese characteristics appear to prevail. As early as the 1960s, Kiddie (1962) recorded two distinct types of sika occupying the Kaimanawa Ranges, referring to them as two “subspecies” rather than two different phenotypes. Each group was, of course, probably affected by

42 The Sika in New Zealand

653

crossbreeding between the original animals liberated and the interim period since but, nevertheless, notable differences were being observed. Kiddie noted some three inches difference at shoulder height, and I noted variation in coloration and spotting to be different as well. In the Ahimanawa Ranges it was obvious that both size and color of the winter pelage tended towards the Japanese type. In fact, while studying a good number of animals from various sources on deer farms (Banwell 1993), animals could be identified that were distinctly Japanese in character, others resembling those from China’s Heilongjiang Province, while others displayed traits of both. It appeared to be more distinct in the males and certainly more so during the winter months when the pelage color was more distinct. Among these animals from varying sources were dark, charcoal-brown and unspotted specimens, while others were a rich, dark chestnut with pronounced spotting and “bibs,” or throat patches. Some were intermediate. Within the females there were grayish animals so typical of the Japanese races, while others were of the typical Chinese brownishgrey with quite distinct spotting. Although the largest of our sika are not as big or heavy as some of the northern Chinese, or Russian specimens, nonetheless, some of the stags taken from the western extremity of the range are much larger than any of the Japanese subspecies with the exception perhaps of the Hokkaido form, C. n. yesoensis (Banwell 1996, 1997). There have been several interesting theories put forward in relation to this significant variation in size within the New Zealand herd. Ernst Eick (personal communication 1996), of the International Sika Society, suggested to me that the significant variations could be due to the fact that the smaller Japanese type was possibly harder to find and offered a much smaller target to the hunter, a factor which could account for them being in larger numbers in certain areas in contrast to the Chinese type with their larger bodies and floral coat. It would be interesting indeed if this segregation was based on each type preferring a different environment. There is some evidence that this is the case. Davidson (1973) in her long-term study in the area claimed the larger, Chinese type tended to keep to the fringes of the forest more so than the smaller, Japanese-like form which tended to venture into more open areas to feed. This difference would parallel the behavior of the northern and southern types in their native ranges in Asia. Just as in Japan and the Asian mainland, the southern types are adapted to fields and disturbed areas and, being in lowlands, are non-migratory and not winter hardy (Takatsuki 1991). In contrast, the northern types are more adapted to forests, are winter hardy, and show migratory movements in response to snowfall. As previously noted, some New Zealand observers believe that the sika in higher elevations move up and down in response to snow. The rutting calls of the New Zealand stag, as would be expected, comprise the full spectrum of the sikine group, including the single, descending call as well as the typical Japanese “hee-haw.” Some males only emit the single descending call, or moan-like signal, while others produce both styles, reflecting their Japanese influence. The subsidiary calls, such as the “hiss” and the “gurgle” the characteristic of all sikines, are also given by New Zealand animals, and the vocal scope of the female and calf, both in communication and alarm are commonly heard.

654

D. B. Banwell

An interesting variation in the New Zealand males is the coloration of the antler velvet. Clear evidence of both black (Japanese) and red (Chinese) colors are produced by various stags and, as would be expected, there are intermediate shades. The configuration of the antlers, the variations between the Chinese and Japanese forms, is evident and in almost every case coincides with the color of the velvet, red in the case of the Chinese style, black in the case of those displaying Japanese conformation.

Health Wild New Zealand sika, like most of the deer family established in this country, have been relatively free of health problems. Little research has been carried out in this field in New Zealand due to the current “pest” philosophy, most information being left to the initiative of the local hunters to report. In captivity, as part of the deer-farming scene, they are susceptible to malignant catarrhal fever. Endoparasites, including lung worm and nematodes of the abomasums, have been recorded in wild specimens. Polydactylism (extra hooves erupting above the normal, or multi-hoof components on the regular form) appears to occur more frequently in New Zealand than elsewhere, and is apparent in sika, red deer, sambar (Cervus unicolor), and fallow deer (Dama dama). In addition to this deformity, campylognathie (twisted nose) has been reported in several specimens of sika and other species. Whether these deformities occur more frequently in New Zealand because of low genetic diversity due to the small number of founders is not known.

Hunting and Trophy Value The intrinsic value of the sika herd is very high in regard to the recreation it offers, as well as having a profound effect on the economy of those who depend on income generated by the visiting hunters. The sika is a highly respected quarry amongst the New Zealand hunting fraternity (Fig. 42.2). Trophy potential is world class, some of the antlers secured comparing favorably with the very best taken in their countries of origin and elsewhere, particularly when the proportionate body size is given consideration. Quality has improved considerably since the population density has been reduced by heavier hunting pressure. Some fine specimens have been secured since 1990 (Banwell 1999). Furthermore, the sika has been capable of producing quality antler in areas where the soil chemistry is quite poor, and where red deer antler quality is comparatively poor. This trend is in tandem with the inability of red deer to compete with sika in this area of poor volcanic soils and, as a consequence, inferior sustenance. Hunting the sika in New Zealand has changed dramatically since the first licenses were issued in 1925. During the early years of trophy hunting, the ethics

42 The Sika in New Zealand

655

Fig. 42.2 A fine trophy taken by Glenn McRae in the Sparrowhawk Range in 1996.

and etiquette of New Zealand’s native Britain were strictly exercised, practiced only by the privileged and wealthy, but in “A Kiwi version of that” as Mavis Davidson (1973) once put it. She recounted that early sportsmen could encounter something in the order of a dozen potential trophy stags in the course of a morning without the need of “squeezing the trigger.” Following the removal of protection in 1930, a period of heavy kills took place, provoking a lot of penned criticism and outrage in the sporting journals of the day. Today hunting sika is free with no license required but, in some areas, a permit from the appropriate authority is necessary; or alternatively, courteous permission requested of the landowner. Hunting is open all the year round. Three completely different hunting factions operate, some chasing the coveted trophy, others hunting merely for the “pot,” while the third element simply enjoys the atmosphere of the great outdoors. Helicopters, fixed-wing aircraft, and four-wheel drive vehicles all make access a lot easier than those far off days when “men were men,” and long tramps carrying a fortnight’s gear on their back was the order of the day. Comfortable huts have replaced tenting in some selected areas. Things are not what they used to be.

Literature Cited Banwell, D. B. 1993. The Sikine mess. Journal of the British Deer Society, Deer 9:39–41. Banwell, D. B. 1995. The sikas. Journal of the British Deer Society, Deer 9:446–450. Banwell, D. B. 1996. The sikas of Japan, Part I. Journal of the British Deer Society, Deer 9:638–642.

656

D. B. Banwell

Banwell, D. B. 1997. The sikas of Japan, Part II. Journal of the British Deer Society, Deer 10:38–43. Banwell, D. B. 1999. The sika. New Zealand Big Game Records Series, Volume One, New Zealand Deerstalkers’ Association, Inc., Halcyon Publishing Limited, Auckland, New Zealand. Challies, C. N. 1985. Establishment, control, and commercial exploitation of wild deer in New Zealand. Pages 23–36 in P. F. Fennessy and K. R. Drew, editors, Biology of deer production. The Royal Society of New Zealand, Bulletin 22. Wellington, New Zealand. Davidson, M. M. 1973. Characteristics, liberation and dispersal of sika (Cervus nippon) in New Zealand. New Zealand Journal of Forestry Science 3:53–180. Davidson, M. M. 1990. The sika deer. Pages 468–477 in C. M. King, editor, The handbook of New Zealand mammals. Mammal Society of New Zealand, Oxford University Press, Auckland, New Zealand. Donne, T. E. 1924. The game animals of New Zealand. John Murray, London, United Kingdom. Flerov, K. K. 1952. Fauna of U.S.S.R., Mammals, Volume 1. Institute of Zoology, U.S.S.R. Academy of Sciences, Moscow, USSR. Glover, R. 1956. Notes on the sika deer. Journal of Mammalogy 37:99–104. Kiddie, D. G. 1962. The sika deer (Cervus nippon) in New Zealand. New Zealand Forest Service, Wellington, New Zealand. Lydekker, R. 1901. The great and small game of Europe, Western and Northern Asia and America. Rowland Ward, London, United Kingdom. Makovkin, L. I. 1999. The sika deer of Lazovsky Reserve and surrounding areas of the Russian Far East. Almanac Russki Ostrov, Vladivostok, Russia. Takatsuki, S. 1991. Food habits of sika deer in Japan with reference to dwarf bamboo in northern Japan. Pages 200–204 in N. Maruyama, B. Bobek, Y. Ono, W. Reglin, L. Bartos, and P. R. Ratcliffe, editors, Wildlife conservation: Present trends and perspectives for the 21st century. Japan Wildlife Research Center, Yushima, Bunkyo-Ku, Tokyo, Japan. Whitehead, G. K. 1950. Deer and their management. Country Life, London, United Kingdom. Whitehead, G. K. 1964. The deer of Great Britain and Ireland. Routledge and Kegan Paul Limited, London, United Kingdom. Whitehead, G. K. 1972. Deer of the world. Constable, London, United Kingdom.

Plate 1 (a–c) Male, female, and calf on Kinkazan Island, Japan. (d) Males in Shiretoko National Park, Hokkaido, Japan. (e) Female and calf on Nozoki Island, Japan. (f) Female on Kinkazan Island, Japan. (Photos by Dale R. McCullough.)

Plate 2 (a) A sika deer male giving a rutting call. (b) A wallowing male. (c) Males engaging in an antler clash. (d) A territorial dominant male guarding females. (e) Male showing sidestretch behavior towards a female. (f) Copulation. (Photo (a) by Seiki Takatsuki, (b) and (e) by Nobumasa Ohnishi, (c) and (d) by Masato Minami, and (f) by M. Nakamura.)

Plate 3 (a) Sika deer prefer to feed on lush herbaceous vegetation. (b) Calf feeding on herbaceous vegetation. (c) Sika deer feeding on Zoysia lawn grasses on Kinkazan Island. (d) Forest with Sasa bamboo understory, a preferred food. (e) Stag in winter with Sasa bamboo. (f) Winter is a time of low food availability in the northern parts of the sika deer range. (Photos by Dale R. McCullough.)

Plate 4 (a) Sika deer stripping bark. (b) Sika deer eliminated small trees by bark-stripping in the forest of Nakanoshima Island, in 1982. (c) Bark stripped from a felled tree in Akan National Park. (d) Deer barked large trees just before a population crash on Cape Shiretoko. (e) Winter grazing area where vegetation was damaged by deer overgrazing. (f) Sika deer observed during a winter aerial census by helicopter. (Photos (a), (c), (e) by Hiroyuki Uno, (b) and (f) by Koichi Kaji, and (d) by Hideaki Okada.)

Plate 5 (a) Severe damage by sika deer to vegetation and soil on Nozaki Island, Japan. (b) Soil erosion caused by vegetation removal by sika deer on Nozaki Island. (c) Sika deer on Nozaki Island feeding on Zoysia lawn. Note poor body condition. When the sod is broken, soil erosion follows. (d) Impact on woody vegetation on Nozaki Island. (e-g) Views of a meadow on Nakanoshima Island, Hokkaido showing vegetation change over time due to feeding of sika deer. (e) View in 1984 after the elimination of tall grasses during the first sika deer population irruption in 1980-83. Short grasses predominate and unpalatable plants have begun invading. (f) Nearer view of the same area in 1992 showing little change. (g) Same view in 2002 after a second deer population irruption. Note the elimination of short grasses and replacement by unpalatable and poisonous plants. (Photos (a-d), (f), (g) by Dale R. McCullough and (e) by Koichi Kaji.)

Plate 6 (a) Sika resting on Kinkazan Island. (b) Japanese macaque and sika calf on Kinkazan. (c) Running male on Kinkazan. (d) Stags feeding on Zoysia lawn on Kinkazan. Note wrinkled noses from pressing into the grass. (e) Sika stags at Nara Park. (f) People feeding crackers to sika at Nara Park. (Photos by Dale R. McCullough.)

Plate 7 (a) Japanese sport hunters dressing deer. (b) Collecting a DNA tissue sample in Vietnam. (c) Restrained dangerous captive rutting stag in Taiwan. (d) Cutting deer antlers in Vietnam. (e) Antler seqments in rice wine in Taiwan. (f) Velvet antlers in freezer in Vietnam. (Photos by Dale R. McCullough.)

Plate 8 (a) Sika deer on a large collective farm near Pho Chau, Vietnam. (b) Sika group on the farm shown in (a). (c) Sika deer in small enclosure at Cat Ba National Park, Vietnam. (d) Stag sika deer in the captive breeding facility, Kenting National Park, Taiwan. (e) Sika deer group by the seashore in Far East Russia. (f) The tracks of a tiger being followed by a person. (g) Female sika deer killed by a tiger in Far East Russia. Note tiger tracks on right side of carcass. (Photos (a–d) by Dale R. McCullough and (e–g) by A. Myslenkov.)

Index

A Accessory corpora lutea, 630 Acetic acid molar ratio, 67 Acid detergent fiber (ADF), 64, 65, 74, 75 Acorns, 219, 387, 390, 391, 396, 398, 399, 512 Acquisition of estrous female, 291 Activity patterns, 534 Adaptation of phenotypes, 652–654 Admixture, 43, 44, 47–49, 57 Aerial survey, 406, 408, 409, 416, 511, 505, 506 Age at maturity, 102, 321, 340, 382, 427 Age structure, 300–304 Age-specific birth rates, 321 Age-specific mortality, 377–379 Age-specific survival, 456 Aggressive behavior, 92, 93, 304 Agricultural damage, 350, 365, 369–370, 409, 416, 437, 438, 446–450 Ainu, 252, 405 Alarm calls, 313, 352 Alces alces (moose), 12, 87, 155, 219, 252, 305, 380, 422, 636 Alfalfa, 64, 65, 70, 73, 74 Allele frequency, 50 Allelic richness, 49 Allogrooming, 353 American silk, 628 Ammonia, rumen concentration, 67, 68, 70 Analysis of spatiotemporal dynamics, 439 Annual reproductive success, 326 Antler-cutting (for medicine), 3, 61, 62, 350, 361, 556 Antlers, 11–17, 20, 90–93, 200, 202, 629 beams and tines 13, 14, 20, 21, 629 clash, 307 cycle, 85, 91 growth, 421, 431 size, 426

Antilocapra americana (pronghorn antelope), 225, 636 Apple lees, 70, 72 Arboreal plants, 62–64, 68 Archaeological sites, 17 Artificial thinning, 189, 190 Asiatic black bear, 7, 37, 381 Askania-Nova (Ukraine), 580 Askold Island (Russia), 477, 482 Assateague Island (Maryland), 618, 619, 631, 632 Asynchronous estrus, 294 Austria, 575, 576 Average group size, 516, 517 Axis axis (axis deer), 55, 328, 578, 584, 615, 616, 627, 629 Axis japonicus, 13, 21

B Bamboo grass (Sasa), 64–66, 68, 71–73, 79, 173, 196, 208–209, 213, 220, 232, 234, 239–244, 246, 251, 254, 256, 260, 263, 265–269, 357, 366–368, 409, 426–428 Barasingha, 578 Bark chemical content, 211 Bark-stripping, 3, 4, 172, 173, 181–186, 189, 190, 194, 207–214, 359, 426, 429, 489, 600, 604, 605 Bears, 7, 37, 255, 381, 405 Belgium, 582 Bighorn sheep (Ovis canadensis), 298, 341, 376, 636 Biodiversity, 218, 358, 360, 405, 417 Biogeographic boundary, 38 Biomass of vegetation, 146, 149, 152–154, 173–177. Birth date, see parturition Birth rate, 321, 322 657

658 Bison bison (American bison), 36 Bite rate, 146, 149, 154, 155 Bite size, 149 Black-tailed deer (Odocoileus hemionus columbianus), 6, 87, 121, 252, 422, Blakiston’s line, 37, 38 Block-count method, 387 Body condition, 193, 199-203 Body mass (weight), 28, 77–78, 107, 195, 196, 199, 302, 303, 308, 309, 320, 323–326, 336, 381, 382, 391, 393, 394, 396, 426, 427–431, 628 Body size, 385, 387, 391, 393–396 Bole-scoring, see bark-stripping Boso Peninsula (Japan), 328, 385-400 Bottlenecking, 49–52, 252, 267, 268, 386 Breeding behavior, 92, 93, 225, 285, 288-292, 294–298, 301–305, 307, 310, 313, 314, 352, 379–382, 600, 601 Breeding season, 103–105, 304, 307, 327, 397, 600 Breeding territory, 287, 290, 293, 294 British Isles, 596–598, 607, 610 Brow tine, 13, 14, 20 Brown bear (Ursus arctos), 7, 255, 381, 405 Browsers, 232, 234, 235 Browsing, 182, 183, 189, 190, 198, 234–235, 246, 426, 428, 429, 512, 604, 605 Browsing line, 173, 425 Bucks-only hunting, 334

C Calf to female ratios, 430 Calving, 327, 328, 387, 630, 632 Calving season, 333, 334, 337–341, 397, 632 Camera traps, 565 Canadian Food Inspection Agency (CFIA), 634 Canis lupus (wolf), 3, 7, 406, 422, 493 Cape Shiretoko (Japan), 421, 427–430 Capra ibex (ibex), 579 Capreolus capreolus (European roe deer), 2, 219, 422, 578, 579, 582, 583, 588, 606 Capreolus pygargus (Asian roe deer), 2, 487, 491 Capricornus crispus (serow) 37, 231, 552 Captive breeding, 61, 62, 79, 479, 480, 483, 502-508, 517, 525, 526, 530–531, 546, 555–558, 561-563, 581, 620–628, 634, 635 Carcass counts, 429

Index Caribou (reindeer) (Rangifer tarandus), 87, 155, 219, 252, 273, 278, 280, 327, 328, 341, 423 Carrying capacity, 168, 382, 421, 423, 431, 463, 464 Catch per unit effort (CPUE), 406, 408, 409 Centers of activity (COA), 256, 259 Cervus axis, 578 Cervus duvauceli (barasingha), 578 Cervus elaphus, 6, 43, 44, 54, 61, 108, 120, 155, 219, 223, 327, 252, 376, 422, 492, 502, 552, 615 C. e. canadensis, 108, 121, 333, 380 C. e. nannodes, 52, 524 C. e. nelsoni, 586 C. e. sibiricus, 580 C. e. xanthopygus, 32, 483, 487, 491, 573, 583 Cervus harbinensis, 20 Cervus horturolum, 53, 54 Cervus kazusensis, 13, 21, 22 Cervus nippon C. n. centralis, see 27, 34-36, 62, 88 C. n. dybowskii, 582, 618 C. n. grassianus, 523 C. n. hortulorum, 524, 542, 574, 577, 587, 596, 618, 645 C. n. keramae, 27, 54, 618, 645 C. n. kopschi, 524, 530 C. n. magishimae, 27 C. n. mandarinus, 523 C. n. nippon, 27, 36, 46, 53–55, 90, 577, 596, 618, 628 C. n. pseudoaxis, 543, 582 C. n. sichuanicus, 524 C. n. taiouanus, 544, 549–559, 561–569, 582, 596, 618 C. n. yakushimae, 27, 28, 54 C. n. yesoensis, 27, 28, 34-36, 46, 62, 65, 70, 90 Cervus porcinus (hog-deer), 578 Cervus pulchellus, 35, 54 Cervus timorensis (rusa deer), 61 Cervus unicolor (sambar deer), 552, 578, 615, 654 Cheting sika deer sanctuary (Taiwan), 561–569 Chiba (Japan), 36, 105 China, 1, 6, 20, 32, 33, 36, 37, 54, 55, 62, 108, 485, 521–537, 545, 551 Chincoteague Island (Virginia), 619, 633 Chromosomes, 46, 47, 629 Chronic wasting disease (CWD), 628, 635, 636

Index Chronological data (fossil record), 13, 14 Chubu block (Japan), 276, 280 Chugoku block (Japan), 277 Ciliated protozoans, 632 Classification (subspecies), 27, 28, 34–36 Climate-vegetation-herbivore system, 432, 464, 523 Colonizing (populations), 421, 423, 429 Coloration and characteristics, 645, 653, 654 Competition, 491, 552, 554, 582, 583, 606, 632–633, 649–651 with red deer, 649–651 Concealment (of young), 2 Conception, 89–91, 333, 339, 340 Conception date, 103, 195, 196, 397 Coniferous cover, 251, 256, 260, 261, 263, 265, 266–269 Conservation and use of population, 482, 524, 531, 537 Consorting female, 307 Consumption of litterfall, 173, 174, 176 Control of females, 334 Control region, 27–29, 32, 33, 35, 36 Copulation, 87, 90, 287–295, 304, 309, 310, 322–324 Cortisol, 94 Crop damage, see damage Crude fat, 64 Crude fiber, 64–67, 74 Crude protein, 64, 198, 211, 212, 214 CWD, see chronic wasting disease Czech Republic, 576, 577, 582, 584, 585, 586, 588

D Dama dama (fallow deer), 61, 62, 87, 103, 108, 286, 293, 305, 578, 579, 581–583, 606, 615, 627, 631, 654 Damage, 190, 195, 198, 208–210, 255, 365, 369-373, 412, 416, 438, 446–449, 489, 490, 604, 605, 611 Deciduous forest (cool temperate broadleaved), 171–178, 182, 239, 273 Deciduous forest understory, 177 Deciduous leaves, 173–177 Decision-making, 438, 439, 449–451, 451, 452 Deer population density, 149, 209, 210, 305– 308, 310, 322, 329, 335, 354, 356, 360, 382, 385, 387–390, 393, 394, 398, 399, 490, 511–513, 517, 564, 565, 599–601, 603, 605, 611 Deer farming, see captive breeding

659 Deer populations, 172, 177, 178 Deer-proof fence, 130, 241, 416 Delayed sexual maturity, 421, 431 Denmark, 577–579, 582, 583 Density, 209–210, 360, 565, 599–600 Density dependency, 393, 398, 431 Density index, 437, 438, 445 Development of fetus, 331, 332, 337 Diet, 116–120, 177, 536, 603, 604, 606 Dietary protein content, 74–77, 391–393 Digestibility, 65, 74–77 Diseases, 630, 635-637, 654 Dispersal, 251, 263, 483, 484, 514–518, 601, 646 Distribution, sika populations, 2, 3, 5, 7, 365, 484, 501–509, 511, 513-515, 521–523, 526–528, 530, 541–546, 550–553, 597, 598, 600–602, 606–608, 648, 649 Divergence time, 36, 37 DNA markers, 43, 47 Dog-kills, 255, 567, 568 Dominant males (DM), 93, 286, 287, 289–295, 320–304, 377, 381 Donegal (Ireland), 597, 598 Drive count, 424 Dung counts, see fecal census Dwarf bamboo, see bamboo grass Dybowski’s deer, 620, 629

E Ecosystem management, 417 Edge species, 2, 3 Edwards Plateau (Texas), 616, 628, 631, 632 Effective population size, 50 Ejaculation, 291 Elaphurus bifurcatus, 13, 21 Elasticity, 455, 457, 460, 462, 463 Elevation magnitude (of distribution), 273, 275–280 Elk, see wapiti, see also Cervus elaphus Endocrinology, 83 Energy balance, 393, 394, 399 England, 596–598, 601, 604–608 Environmental stochasticity, 432 Eradication, 611 Estonia, 581 Estrous cycle, 85–87, 103–105, 108, 109 Estrous periods, 102, 294–285, 310, 313, 315 Evergreen broad-leaved forests, 196, 204, 273, 287, 385, 386 Evolutionary significant unit, 52 Exclosures, 173–175, 177, 181, 182, 184, 186–189

660 Exotic Wildlife Association, 620 Extinction events, 5, 11, 22, 546, 554, 555

F Factors predicting agricultural damage, 448, 449 Fallen leaves, 62, 68, 426, 432, 512 Fallow deer (Dama dama), 61, 62, 87, 103, 108, 286, 293, 305, 578, 579, 581–583, 606, 615, 616, 627, 631, 633, 654 Far East Russia, see Russia Farmed deer, see captive breeding, deer farming Fat deposit, 107, 108, 387, 390, 393–396 Fawning, see calving season Fecal census, 335, 387, 438, 440, 599 Fecal nitrogen, 75, 268, 387, 392–393, 566 Fecal steroid analysis, 84 Fecundity rate, 372, 456, 458, 461, 467, 601 Feed intake, 72–74 Feeding, 603–604 Fertility rates, 193, 601 Fetal growth, 196, 330–333, 336, 337, 341 Fetal growth line, 333, 339 Finland, 581 Fluid content (rumen), 67, 68, 70 Follicle-stimulating hormone (FSH), 83, 85 Food habits, 70-73, 113–121, 160, 161, 165, 168, 194, 196–199, 201, 204, 232, 233, 235, 236, 273, 274, 280, 356, 357, 385, 387–391, 536, 603, 631 Food limitation, 121, 172, 212, 321 Food quality, 78, 198, 432, 566, 567, 664 Food, seasonal change, 121, 218, 273, 274, 387, 391, 392, 394, 395, 399 Foraging efficiency, 146, 155 Forest dynamics, 4, 178, 181, 186, 189, 190, 224, 486 Forest regeneration, 130–134, 178, 490 Forestry damage, see damage Forest structure, 181, 183, 186, 189, 190 Forest undergrowth, 219, 220 Formosan sika deer, 561, 618, 630 Fossil record, 13–17, 550 Founder effect, 562, 629 France, 578, 584 Freeze branding, 298

G Gene flow, 45, 47, 48, 50–52, 57 Genetic boundary, 5, 37–39 Genetic differentiation, 43, 47, 48, 52, 57 Genetic distance, 32, 37–39

Index Genetic divergence, 33, 36 Genetic diversity, 28, 43, 44, 49–52, 57, 481, 524, 531, 541, 547, 549, 556, 629 Genetic drift, 50–52, 524 Genetic origin, 595 Geographical variation, 217, 218, 232, 274 Germany, 578, 579, 581–585 Gestational age, 333, 339 Gestation, 89, 90, 333, 397 Gestation period, 89, 90, 102, 106, 107, 333, 397 Goat herding, 566, 569 Gonadotropic hormones (LH and FSH), 83, 85 Gonadotropin releasing hormone (GnRH), 83, 84 Goral (Nemorhaedus caudatus), 491 Goto Islands (Japan), 287 Graminoids, 114–121, 196, 198, 356, 357 Grass silage, 70, 71, 74–76 Grazing, 120, 125–127, 134, 135, 139, 140, 232, 235, 239, 240, 246, 604, 605 Grazing tests, 64 Green Island (Taiwan), 555, 557, 558 Group size, 487, 499, 505, 508, 510, 513, 516, 517 Groups of females, 286, 288, 306, 601, 602 Groups of males, 301, 304-306, 310 Growth, 301, 302, 395 curves, 199–200 fluctuations, 395, 396 parameters, 201 rate, 109, 568, 569 Guarding (of females by males), 287–291, 294, 295

H Habitat, 1–6, 224–226, 273, 274, 280, 421–424, 431, 432,467, 475, 479, 483–493, 495, 497, 602, 603, 649, 650 disturbance, 2–5 fragmentation, 52, 280 selection, 159–161, 168, 536, 565, 602, 603, 649–650 suitability, 535, 536 use, 603, 606 Hagenbeck, 574, 576–580 Hanaizumi site (Japan), 16, 17 Haplotypes, 32, 33, 37, 38 Harem, 304–307, 310, 600 Harvesting, 437–439, 444, 445, 449 Head-up display, 306–308 Health of herd, 654 Helicopter counts, 511 Hemicellulose, 211, 212, 214

Index Herbaceous plants, 62–64, 68 Heterozygosity, 49, 51 Hog deer, 578, 584 Hokkaido Island, 11, 14, 21,33, 37, 38, 46, 50, 53, 103, 251–253, 328, 405–417, 422 Holocene, 17, 20, 22, 521 Home range, 159–168, 251, 256–261, 263–269, 287–290, 293, 294, 352, 353, 602 Home range use, 274, 275, 277 Hondo spruce, 207, 209–212 Honshu-Shikoku-Kyushu, 11, 12, 14–16, 20–22, 38, 39, 44, 46-48, 273–281 Honshu sika deer, 37, 66-68, 72, 73 Hormones, 83–85 Human-deer impacts, conflicts, 3, 274, 369–370, 438, 467, 488, 541, 552–554 Hungary, 579 Hunting, 4, 5, 7, 193, 195, 199, 252, 255, 369–371, 438, 441, 446, 466, 483, 493–497, 553, 555, 616, 634–637, 654, 655 Hunting and trophy value, 654, 655 Hybridization, 46, 54, 491–492, 584–588, 606–611, 629, 635, 651, 652 with red deer and wapiti, 492, 635, 648, 651, 652 Hydropodes sp. (water deer), 12 Hyogo Prefecture, 193, 328–330, 335–337, 339, 437–452

I Ibex, 579 Ice bridge, see also land bridge, 21 Immigration, 12, 20, 21, 37 Immunocontraception, 630 Insulin-like growth factor 1 (IGF-1), 92 Intermediate feeder, 79, 387 Intraspecific variation, 13, 285, 286 Intrinsic rate of increase, 432 Introductions, see relocation Introgression (genetic), 596, 607, 608 In vitro dry matter digestibility (IVMD), 64 Ireland, 596, 597, 602, 604, 605, 607, 610, 611 Irruption (population), 3, 173, 405, 409, 413, 421–425, 430–432 Island populations, 278

J James Island (Maryland), 618, 619, 629 Jarman-Bell principle, 114, 121, 231–232 Japanese Archipelago, 217, 218, 222, 224 Japan Sea, 278–280

661 K Kanto block (Japan), 275 Kashima shrine (Japan), 348 Kenting National Park (Taiwan), 561–569 Kentucky, 620 Kerama Island (Japan), 38 Kidney fat mass (KFM), 201–203, 389, 395 Killarney (Ireland), 596, 597, 601, 602, 605, 607 Kinkazan Island (Japan), 50, 51, 101, 103–108, 125–130, 134–142, 159, 160, 166–168, 278, 298–316, 320–326, 375–382 Kinki block (Japan), 277 Kofuku-ji temple (Japan), 350 Korea, 1, 525, 527, 529, 541–547 Korea Strait, 11, 12, 20, 21 Kumamoto Prefecture (Japan), 328, 334–336 Kyushu, 38, 39, 51, 53, 596, 608 Kyushu block (Japan), 277

L Land bridge, 5, 11, 12, 20, 21, 27, 28, 37, 524, 543, 550 Land use, 439, 441, 444 Late conception, 103, 109 Lazovsky Nature Reserve (Russia), 502, 503, 505–507, 509, 510 Lead poisoning (eagles), 416, 417 Legislation (regulatory), 634 Lek, 286, 305–307 Leopard (Amur), 492, 493 Leslie-Lefkovitch matrix, 456, 462 Leucothoe, 128, 130 Life history, 355, 412, 453–455, 457, 461–463, 468, 600 Life span, 325, 326, 377, 379, 381 Life-stages, 455, 458, 461–463 Life tables, 355, 375, 376, 381, 454, 456 Life-time reproductive success (LRS), 319–326 Lignin content, 65 Lithuania, 581, 582, 585 Litterfall (fallen leaves), 171–178, 182 Live-capture, 102, 298, 299 Long-term study, 298, 309, 320, 376 Low-stretch approach, 310 Luteinizing hormone (LH), 83, 85 Lyme disease, 631 Lynx lynx (lynx), 493

662 M Macaca fuscata (Japanese macaque), 37, 126, 467 Major histocompatibility complex, 44 Male groups, see groups, male Male rank, 302 Malignant catarrhal fever, 631 Mallows’ Cp, 439, 444, 447 Management, 405, 413, 414, 437–452, 468, 469, 494–497, 524, 525, 531, 533, 535–537, 558, 569, 610, 611, 633–635 Manchurian (Dybowski’s) sika, see C. n. dybowskii Manchurian wapiti, 583, 585 Mandible marrow fat (MCF), 200–202 Marine isotope stages, 15 Marking behavior, 298, 304 Mark-resight, 565 Maryland, 618, 619 Mass mortality, 429, 431, 464–465 Mate choice, 316 Mate guarding, 313 Maternal units, 301 Mating aggregation, 286, 288, 291, 295 behavior, 286–295 success, 288, 290, 293 system, 286, 293, 352 territories, 297 Medial basal hypothalamus (MBH), 84 Medium-sized deer, 11–14, 17, 21, 22 Melatonin, 84, 85 Metapopulation, 52, 57 Microsatallites (nuclear DNA), 47–51, 543 Middle Pleistocene, 15, 16, 20, 21 Migration, 159, 251–252, 256–258, 260, 261, 263, 266–269, 273–281, 487, 542, 584 Migration strategy, 267–269 Milk females, 321, 323 Mineral balance, 214, Mineral intake, 68, 69, 488 Minimum convex polygon method, 288 Minimum protein requirements, 198 Mirounga angustirostris (northern elephant seal), 298 Miscanthus (silver grass), 66–67, 114–118, 128, 134, 135, 139, 140, 160, 166, 168, 222, 234, 356, 358 Mitochondrial DNA, 28, 43–45, 47–48 Mixed forest, 254, 256, 266 Model selection method, 439, 444, 445, 447, 448 Moldova, 581 Molecular phylogeny, 33

Index Mongolian gazelle 225 Monitoring (population), 194, 405, 417, 437–439, 449, 563, 566 Moose (Alces alces), 12, 87, 155, 219, 252, 305, 380, 422, 636 Morphological variation, 28 Mortality, 320, 322, 325, 355, 376-381, 412, 424–427, 429, 431, 432 Mouflon (Ovis musimon), 578, 579, 588 Mountain goat (Oreamnus americanus), 298, 636 Mounting (copulation), 291, 310 Movement patterns, 276 mtDNA, 28, 32, 37, 39, 43, 47, 610 Mt. Goyo (Japan), 118-120, 234, 239, 240, 365–373 Mt. Kasuga (Japan), 347, 359, 360 Mt. Ohdaigahara (Japan), 207–214 Mule deer (Odocoileus hemionus hemionus), 121, 252, 278, 280, 380, 422, 635 Multi-estrus females, 105, 108, 109 Multiple mating (copulation), 287–289, 293, 310, 313–316 Muntiacus (muntjac), 2, 328, 552

N NADeFA. See North American Deer Farmers Association Nagasaki, 596, 608 Nakanoshima Island (Japan), 103, 172, 175, 178, 181, 182, 219, 298, 421, 423–427 Nara deer fund, 351, 352, 354, 355, 357 Nara Park (Japan), 1, 298, 347–361 Natural regulation, 422 Natural selection, 319 Natural thinning (forest), 189 Natural vegetation, 239, 240 Negative energy balance period, 393, 394 Neighbor-joining method, 33 Nemorhaedus caudatus (goral), 491 Neonatal weights, 107 Neutral detergent fiber (NDF), 64, 65, 74, 75 New Zealand, 62, 172, 328, 552, 643-645 Nikko fir, 209–212 Nilgai antelope (Boselaphus tragocamelus), 616 Nishikigaoka (Japan), 16 Nitrogen-free extract (NFE), 64 Nomadism, 251 Nonindigenous ungulates, 616, 633 Nonmigrant, 256, 263, 268, 269 Non-territorial dominant males (ND), 302, 303, 307–309

Index North American Deer Farmers Association, 620, 621 North American Elk Breeders Association, 620 North Carolina, 620 Northern elephant seal, 298 Northern Ireland, 598 Nozaki Island (Japan), 167, 278, 285-287, 292–294, 298, 307 Nucleotide diversity, 32, 33 Nuisance control, 438, 441, 446 Nutrient intake, 65, 73 Nutrition, 74–77, 194, 201, 204, 357 Nutritional physiology, 62, 108, 320, 325, 326, 385, 399

O Occurrence index, 565 Odocoileus hemionus columbianus (blacktailed deer), 6, 87, 121, 252, 422, Odocoileus hemionus hemionus (mule deer), 121, 252, 278, 280, 380, 635 Odocoileus virginianus (white-tailed deer), 2, 87, 102, 121, 155, 171, 219, 252, 263, 328, 380, 422, 615, 632 Okhotsk (Russia), 253, 267 Okinawa Prefecture (Japan), 38 Oldest pregnant females, 336 Optimal strategy, 295 Orchardgrass, 62, 64, 71, 75 Oreamnus americanus (mountain goat), 298, 636 Ovis aries (Soay sheep), 298, 319, 376, 423 Ovis canadensis (bighorn sheep), 298, 376, 636 Ovis musimon, (mouflon), 578, 589, 588 Ovulation, 87, 88, 90 Oxygen isotope fluctuation curve, 15, 20

P Pacific Ocean, 278–280 Palatable plants, 172, 176–178 Paleobiology, 4, 5, 37 Paleolorodon naumanni (fossil elephant), 20–22 Pangola grass, 566 Panthera pardus orientalis (Amur leopard), 492, 493 Panthera tigris altaica (Siberian tiger), 492 Parasites, 630–631 Parturition, 77, 84, 89, 90 chances, 322-324, 326

663 date, 102, 106, 195, 301, 321, 325, 336 interval, 321, 322 rate, 322, 324 repeated, 322 Pellet-count, see fecal census Père David’s deer, 55, 87 Perennial ryegrass, 71 Photosynthetically active radiation (PAR), 182, 186, 187 Phylogenetic trees, 27, 29, 32, 33, 35, 43, 47–49, 53–55 Phylogeny, 33, 34 Phylogeography, 47, 57 Physical condition, 431, 432 Plant communities, 125–142, 145–156 Plantation forest, 195, 198 Plant phenology, 108, 220, 398 Plant species richness, 173, 174 Pleistocene, 5, 16, 27, 28, 37, 38, 43, 55, 521, 522 Pleioblastus (dwarf bamboo), 114–118, 128, 140, 160, 166, 168, 234 Pliocene, 55 Poland, 580, 583, 584 Polymerase chain reaction (PCR), 29, 32 Population, 193–196, 437–452, 494–497, 527–530 bottleneck, 406 census, 352, 354, 495, changes, 424–426, 429 conditions, 280 control, 369–372, 405, 413, 438 crash, 172, 173, 177, 178, 424, 432 density, 173, 177, 178, 181–184, 189, 195, 196, 198, 357, 487, 490, 495, 512 dynamics, 386, 387, 534, 601, 606 growth rate, 423, 425, 432, 444-446, 534, 565, 566, 601 history, 44, 47–53, 354, 414, 429, 482–484, 495, 525, 544 irruptions, 3, 405–417, 421–432 parameters, 421, 426, 427, 430, 431 size, 421, 423–425, 429, 431, 483, 576, 598–600 size index, 387, 408, 414 structure, 43, 44, 52, 57, 430 survival, 193, 194, 377 Postcopulative guarding, 287, 288, 294 Post-peak population, 178 Post-peak stages, 172 Powerscourt, 596, 607 Predation, 492, 566 Pregnancy, 89, 194, 195

664 Pregnancy rate, 102, 105, 106, 195, 330, 336, 355, 356, 398–400, 412, 456, 458 Prime ages, 326 Primeval forest, 208–210, 212, 359 Primorsky Krai (Russia), 6, 476, 502–505, 512, 513, 516, 517 Proboscidean species, 16, 20 Productivity, 145, 146, 148–152, 154, 155, 340 Procapra gutturosa (Mongolian gazelle), 225 Progesterone, 84, 86–89 Prolactin, 85, 92 Promiscuous mating, 310 Pronghorn antelope (Antilocapra americana), 225, 636 Propionic acid molar ratio, 67 Protozoa count, 67, 70 Proximate analysis, 62 Pseudaxis grayi (fossil deer) 20 Pseudaxis hortulorum (= C. n. hortulorum), 20 Puberty of females, see age at maturity

Q Quantitative trait loci (QTL), 46 Quaternary, 12

R Radio-telemetry, 219, 252, 254, 255, 273, 275, 276, 564 Rangifer tarandus (caribou), 87, 155, 219, 252, 273, 278, 280, 327, 328, 422 Raw grass, alfalfa hay cubes, 70 Reaction norm, 385 Recovery, sika deer in Taiwan, 555-559, 561-563 Red deer, see also Cervus elaphus, 6, 43–46, 49, 54, 55, 62, 73, 87, 108, 120, 219, 224, 306, 319, 327, 376, 492, 502, 552, 573, 606, 615, 627, 633, 635, 636 Regional difference, 391, 396 in calving season, 328, 341, 342 in food selection, 214 in weight, 3, 91, 396 Reoccupation of the previous range, Russia, 510 Reproduction, 101–103, 105, 328-342, 377, 385, 397, 398, 630 Reproductive behavior, see breeding behavior Reproductive strategy, 297 Reproductive success, 297, 298, 307–309, 315, 319, 320, 324 Reticulum, 69, 70

Index Roe deer (Asian, Capreolus pygargus) 2, 487, 491 Roe deer (European, Capreolus capreolus) 2, 219, 422, 578, 579, 582, 583, 588, 606 Roughage (in diet), 71, 212, 213 Rumen anatomy, 632 ciliates, 631 contents, 67, 68 function, 212–214 nitrogen content, 392 Russia, 328, 475–497, 501, 502, 527, 545, 552, 580, 581, 584, 585, 644 Russia, range recovery, 475–481, 483–487, 490–492, 494, 495 Rut call, 224, 304, 307 Rut panic, 313 Rut season, see breeding season Ryuku sika deer, 618

S Sahama (Japan), 15, 16 Sakhalin Island (Russia), 37 Sambar deer (Cervus unicolor), 552, 578, 615, 654 Sasa, see bamboo grass Scotland, 595–611 Sea eagle, 416, 417 Sea level changes, 12 Searching, 294 Secondary compounds 128, 219 Sedentary (non-migratory), 273–281 Seedlings, 181, 184, 186, 189, 190 Semicaptive, 563 Sensitivity analysis, 453–469 Sequencing (DNA), 29, 32 Serow (Capricornus crispus), 37, 231, 552 Sex ratio, 106, 195, 300, 430 of fetuses, 331 of newborn fawns, 106 Sexual dimorphism, 121, 199, 382 Sexual maturity, see age at maturity Shikoko Island (Japan), 43 Shiranuka Hills (Japan), 234, 251, 254, 260, 265–268 Shiretoko National Park (Japan), 422, 428 Shiretoko Peninsula (Japan), 423, 427, 428 Short-grass community, see Zoysia Siberian tiger (Panthera tigris altaica), 492 Sichuan Province, basin (China), 526, 533 Sighting per unit effort (SPUE), 406–409, 411 Sika–human interactions, 553, 554

Index Sikhote-Alin Mountains (Russia), 476, 487, 501-513, 516 Sikhote-Alin Nature Biosphere Reserve (Russia), 502, 512 Silent ovulation, 87–88, 90 Simulation-based method, 455 Simulations, population, 449–452 “Sit and wait” tactics, 292 Site fidelity (to home range), 251, 263, 265 Slovenia, 581 Sasa nipponica grasslands, 208, 209, 212–214 Snow, 221, 222, 226, 240, 251, 253, 254, 263, 264, 266, 267, 273–281, 366–368, 428, 443, 465, 486 Soay sheep (Ovis aries), 298, 319, 376, 423 Social system, 285, 286, 306, 307, 352, 379, 601–02 Soluble carbohydrates, 390, 398 Spacing pattern (of rutting males), 289, 290, 292 Species-area relations, 173, 174 Species concept, 6, 44, 53–56 Sperm competition, 295, 313, 316 Spotlight count, 408–410 Spring forage, 147–150, 152, 154, 155 Stotting, 313 Subordinate males (SB), 287, 290–295, 302, 303, 307–309, 313, 314, 377 Subspecies, 27–30, 34, 36, 37, 39, 54, 523, 618, 627, 628, 645, 652, 653 Summer forage, 147–150, 152, 154, 155, 172, 176, 516 Supplementary feeding, 369, 370 Surveys in Russia, 505, 511, 513, 515 Sus scrofa (wild boar), 231, 467, 492, 552 Survival, 193, 194 Survival curves, 377 Switzerland, 578, 581, 582, 584

T Taipei Zoo, 562 Taiwan, 1, 5, 6, 62, 543, 549–559, 561–563 Tandem repeat, 27, 32 Taohongling Nature Reserve (China), 535, 536 Tategahana site (Japan), 16, 17 Taxonomy, 6, 28, 30, 34, 36, 39, 44, 53–55. Tending bond, 305 Territorial dominant (TD) males, 225, 302–310, 313–316, 379–381 Testosterone, 83–85, 90–93 Texas, 616–618 Tiebu Nature Reserve (China), 533–534 Time series density-dependent model, 439, 444

665 Tohoku District, block (Japan), 125, 275, 278, 280 Tokara Strait (Japan), 37 Total volatile fatty acid (VFA) concentration, 67 Tourism, 360, 557, 558 Toya Lake (Japan), 172, 175, 181, 182, 190 Trace elements, 68 Tsugaru Strait (Japan), 11, 12, 21 Tsushima Island (Japan), 28 Tuberculosis, 631, 635 Tule elk, 52, 524 Twinning, 106, 330, 331, 336, 340, 341, 630

U Understory plants, 174 Ungulate herbivory, 177 Ukraine, 580 United Kingdom, 574 Unpalatable plants, 125, 126, 128, 129, 142, 172, 174, 177, 182, 224, 358, 426, 429, 430 Urinary nitrogen, 75 USSR, 579, 580, 587

V Vascular flora, 173, 174 Vegetation changes, 426, 429 Vietnam, 1, 5, 6, 328, 541–547 Virginia, 619 Vital rates, 454, 455, 457, 459, 460, 462–464 Vocalization, 307 alarm calls, 313, 352 rut calls, 224, 301, 307

W Wales, 598 Wapiti, 6, 44, 54, 55, 87,108, 121, 155, 219, 223, 224, 252, 380, 422, 483, 487, 491, 492, 583, 585, 586, 620, 628, 635, 636 Warm-temperate evergreen broad-leaved forest, 273 Watase line (Japan), 37, 38 Water deer, 12 Wheat bran, 72, 74–76 White-tailed deer (Odocoileus virginianus), 2, 6, 87, 102, 121, 155, 171, 219, 252, 263, 293, 328, 380, 422, 615, 632 Wicklow (Ireland), 596, 597, 600, 604, 607, 610

666 Wild boar (Sus scrofa), 231, 467, 492, 552 Winter climate, 256, 261, 265, 267, 268, 421, 428, 431, 453, 464, 465, 467, 486 Winter concentration areas, 426, 428, 508 Wolf (Canis lupus), 3, 7, 405, 406, 422, 493 World Cultural Heritage Site, 359 World Natural Heritage Site, 428

Y Yeld females, 321, 323, 324 Yeso sika deer, 62–66

Index Yield-density diagrams, 181, 183, 184, 186, 188, 189 Yoshinoda (Japan), 15

Z Zheleznyakovsky Refuge (Russia), 513–515, 518 Zhoukoudian (China), 17 Zoysia (lawn grass), 66, 67, 114–118, 120, 125, 127–128, 139–142, 145–146, 148–155, 160, 164–165, 167–168, 234, 298, 322, 354, 356, 357, 358