Detect and Deter: Can Countries Verify the Nuclear Test Ban?

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

Ola Dahlman • Jenifer Mackby Svein Mykkeltveit • Hein Haak

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

Ola Dahlman OD Science Applications Fredrikshovsgatan 8 115 23 Stockholm Sweden [email protected]

Jenifer Mackby Center for Strategic and International Studies 1800 K St. NW Washington, DC 20006 USA [email protected]

Svein Mykkeltveit NORSAR POB 53 2027 Kjeller Norway [email protected]

Hein Haak KNMI POB 201 3730 AE De Bilt The Netherlands [email protected]

Cover illustration: Cover photo is taken from a boat heading through the ice for the radionuclide station (RN73), Palmer Station in Antarctica, symbolizing the long journey of the CTBT.

The facts and opinions expressed in this work are those of the authors and not necessarily those of the publisher. Every effort has been made to contact the copyright holders of the figures and tables which have been reproduced from other sources. Anyone who has not been properly credited is requested to contact the publishers, so that due acknowledgement may be made in subsequent editions. ISBN 978-94-007-1675-9 e-ISBN 978-94-007-1676-6 DOI 10.1007/978-94-007-1676-6 Springer Dordrecht Heidelberg London New York Library of Congress Control Number: 2011935151 © Springer Science+Business Media B.V. 2011 No part of this work may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, microfilming, recording or otherwise, without written permission from the Publisher, with the exception of any material supplied specifically for the purpose of being entered and executed on a computer system, for exclusive use by the purchaser of the work. Printed on acid-free paper Springer is part of Springer Science+Business Media (www.springer.com)

Contents Acknowledgments. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1 Setting the Political Stage. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 1.1 Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 1.2 Provisions of the CTBT. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 1.2.1 Preparatory Commission Phase . . . . . . . . . . . . . . . . . . . . . . . . . . 12 1.2.2 Entry into Force. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 1.2.3 Scope — Zero Yield. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 1.3 Testing as Part of Nuclear Weapon Development. . . . . . . . . . . . . . . . 23 1.4 Political Development . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25 1.5 CTBT in the Context of Nuclear Non-Proliferation and Disarmament . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 2 Monitoring Underground Nuclear Explosions. . . . . . . . . . . . . . . . . . . . . 2.1 Underground Nuclear Explosions and Weapon-Related Experiments. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1.1 Characteristics of an Underground Nuclear Explosion . . . . . . 2.1.2 Nuclear Weapon-Related Experiments . . . . . . . . . . . . . . . . . . . . 2.2 Seismological Monitoring. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2.1 Measurements. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2.2 Detection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2.3 Event Location. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2.4 Event Characterization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2.5 New Methods to Analyze and Exploit Data. . . . . . . . . . . . . . . . . 2.2.6 New Paradigm for Seismic Analysis . . . . . . . . . . . . . . . . . . . . . . . 2.3 Radionuclide Xenon Monitoring. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3.1 Release of Xenon from an Underground Explosion . . . . . . . . . 2.3.2 Measurements. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3.3 Background Radiation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3.4 Detection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3.5 Localization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3.6 Discrimination . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.4 Satellite-Based Monitoring. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.5 The North Korean Nuclear Explosions . . . . . . . . . . . . . . . . . . . . . . . . . 2.6 States and “Precision Monitoring”. . . . . . . . . . . . . . . . . . . . . . . . . . . . .

29 29 29 30 32 33 35 48 54 57 60 63 63 66 67 68 73 74 75 77 84

v

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

3 Monitoring Atmospheric Nuclear Explosions . . . . . . . . . . . . . . . . . . . . . 89 3.1 Thirty Years since Last Nuclear Explosion in the Atmosphere . . . 89 3.2 Many Detectable Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90 3.3 Infrasound Monitoring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91 3.3.1 Infrasound Observations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93 3.3.2 Infrasound Sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95 3.3.3 Detection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98 3.3.4 Localization and Characterization . . . . . . . . . . . . . . . . . . . . . . 100 3.4 Seismological Monitoring. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101 3.5 Radionuclide Monitoring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102 3.5.1 Monitoring Radionuclide Particles . . . . . . . . . . . . . . . . . . . . . 103 3.5.2 Particulate Detection and Tracking . . . . . . . . . . . . . . . . . . . . . 104 3.5.3 Monitoring Noble Gases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107 3.6 National Technical Means . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107 3.6.1 U.S. Satellite-Based Systems to Monitor Atmospheric Nuclear Explosions . . . . . . . . . . . . . . . . . . . . . 108 4 Monitoring Nuclear Explosions in the Oceans . . . . . . . . . . . . . . . . . . . 4.1 Few Explosions Under Water . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2 Oceans Are Vast and Vastly Transparent . . . . . . . . . . . . . . . . . . . . . 4.3 Hydroacoustic Monitoring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.3.1 Eleven Stations Monitor the Oceans . . . . . . . . . . . . . . . . . . . . 4.3.2 Detection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.3.3 Location . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.3.4 Multitude of Sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.4 Seismic Monitoring of Oceanic Explosions . . . . . . . . . . . . . . . . . . 4.5 Radionuclide Observations of Underwater Explosions . . . . . . . . 4.6 Satellite Observations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

111 111 112 113 114 117 118 121 123 125 126

5 On-Site Inspections . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.1 The Political Frame . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.2 Reflections on the Frame . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.3 Conduct of an OSI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.3.1 Operational Manual . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.3.2 Systemic Approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.3.3 Inspection Team . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.3.4 Logistics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.4 OSI Technologies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.4.1 Visual Observations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.4.2 Radionuclide Techniques . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.4.3 Seismological Aftershock Monitoring . . . . . . . . . . . . . . . . . . . 5.4.4 Geophysical Exploration Techniques . . . . . . . . . . . . . . . . . . . 5.5 OSI: A Key Tool for States to Verify and Deter . . . . . . . . . . . . . . . . .

129 130 132 136 136 138 140 142 143 143 145 150 154 155

vi

Contents

6 Synergy with Science. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.1 Science and the Nuclear Test Ban Treaty— A Long-Standing Relationship . . . . . . . . . . . . . . . . . . . . . . . . . . 6.2 Rapid S&T Development since the Treaty Opened for Signature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.3 Synergy among Different Sciences . . . . . . . . . . . . . . . . . . . . . . . . . . 6.3.1 Seismology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.3.2 Radionuclide Observations . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.3.3 Infrasound . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.3.4 Hydroacoustics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.3.5 Data Mining . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.4 Science and the CTBT—Perspectives for the Future . . . . . . . . . . . 6.4.1 CTBTO and Scientific Community Perspective. . . . . . . . . . . 6.4.2 States and the Scientific Community . . . . . . . . . . . . . . . . . . .

159

7 Verifying the CTBT— A State Perspective. . . . . . . . . . . . . . . . . . . . . . . . 7.1 Verification—A Political Process . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.1.1 How Much Verification Is Enough? . . . . . . . . . . . . . . . . . . . . . 7.1.2 Deterring Non-Compliance . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.1.3 Verifying Compliance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.1.4 Building Confidence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.2 Monitoring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.2.1 The International Monitoring System (IMS) . . . . . . . . . . . . . 7.2.2 Precision Monitoring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.2.3 Wide-Area Monitoring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.2.4 Non-Governmental Monitoring . . . . . . . . . . . . . . . . . . . . . . . . 7.3 Consultation and Clarification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.4 On-Site Inspections . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.5 Ready for Entry into Force? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.5.1 The CTBTO—A “Common House” for States Parties . . . . . . 7.5.2 States Parties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.5.3 Regional Cooperation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.6 Detect and Deter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

181 182 182 186 188 189 191 191 193 198 200 201 203 206 207 209 209 212

160 161 163 164 165 168 170 174 177 177 178

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 215 Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 243

vii

Acknowledgments All four of us were closely engaged in the negotiations on the Comprehensive Nuclear-Test-Ban Treaty at the Conference on Disarmament in Geneva and its Group of Scientific Experts. During the implementation of the treaty, we have all been occupied with the CTBTO Preparatory Commission since it was established in 1996. During all these years we have been privileged to work with many members of the scientific and diplomatic communities in Geneva and Vienna, and at a number of research institutions around the world. They have all, in one way or another, influenced our thinking and contributed to this work. In particular, we would like to mention our close friends in the Group of Scientific Experts, in the CTBT negotiations, and in the Working Group on Verification in the CTBTO Preparatory Commission in Vienna. We also wish to acknowledge the stimulating interaction over the years with friends and colleagues at the Provisional Technical Secretariat (PTS) of the CTBTO Preparatory Commission, and express appreciation for their contributions to the book. In particular, we gratefully acknowledge the assistance of Lassina Zerbo, Bob Pearce, John Coyne, Annika Thunborg, and Todd Vincent in providing illustrations produced by the staff of the PTS. The International Scientific Studies project that culminated in a large scale conference in June 2009 (ISS09) provided a vast amount of valuable information on the recent scientific and technological developments that served as essential background material for this book. We are indebted to many scientists who kindly provided us illustrative material they presented in posters or lectures at the ISS09 Conference. These contributions are identified in captions and in the list of references. A number of colleagues and friends have contributed significantly to this book by reviewing the text and providing valuable comments and

ix

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

suggestions. Our deep appreciation goes to Elisabeth Blanc, Malcolm Coxhead, Lars-Erik De Geer, Läslo Evers, Steven Gibbons, Anders Ringbom, Richard Starr, David Simpson, Bernard Massinon, Vitaly Schukin, Aili Bi, Edward Ifft, Wolfgang Weiss, and Kyoshi Suyehiro, who helped improve different parts of the manuscript. We also thank all who have contributed in informal discussions and exchanges. We express gratitude to colleagues at our home institutions for their support and encouragement. Tormod Kværna and Frode Ringdal at NORSAR have contributed with illustrations. At KNMI, Marieke Laagland, Torild van Eck and Ko van Gend have creatively helped out with the many figures. At CSIS, Andrew St. Denis, Anna Newby and Talitha Dowds assisted in shepherding through the volume. We are grateful for the excellent cooperation and support from Petra van Steenbergen, Hermine Vloemans, and their team at Springer publishing company. The interest and generous support for this project from the Ministry of Foreign Affairs of the Netherlands and the Nuclear Threat Initiative are also highly appreciated. The responsibility for any errors or ambiguities that remain in the book rests solely with us. The views presented in this book are those of the authors and may not reflect, and do not represent, those of the institutions, organizations, and national authorities with which the authors are or have been associated. Stockholm, Sweden Washington, D.C., USA Kjeller, Norway De Bilt, The Netherlands

Ola Dahlman Jenifer Mackby Svein Mykkeltveit Hein Haak

Ola Dahlman served as Chairman of the Group of Scientific Experts at the Conference on Disarmament, and he also chaired the CTBTO Preparatory Commission work on implementing the verification regime. Jenifer Mackby served as Secretary of the CTBT negotiations, the Group of Scientific Experts, and the CTBTO Preparatory Commission work on implementing the verification regime. Svein Mykkeltveit represented Norway in the Group of Scientific Experts and serves as Friend of the Chair in the CTBTO Preparatory Commission work on implementing the verification regime. Hein Haak represented The Netherlands in the Group of Scientific Experts and the CTBT negotiations, and serves as Chairman of the CTBTO Preparatory Commission work on implementing the verification regime.

x

INTRODUCTION How can countries verify compliance with the Comprehensive NuclearTest-Ban Treaty (CTBT) and detect and deter violations? How can the dramatic developments since the end of the 20th century in science and technology provide countries improved or even new tools? How can states, by adopting new concepts and procedures, efficiently monitor areas of interest or concern to them, individually or together with other countries? These are the issues addressed in this book, written at a time when the political developments seem to favor a decline in the role of nuclear weapons and a priority placed on entry into force of the CTBT. This book aims to provide constructive input to countries around the world on how they could arrange their efforts to verify compliance with the CTBT and how they could cooperate with other countries in fulfilling their functions as states parties in an efficient, cost-effective manner. The task of verifying compliance with the CTBT rests with the states parties to the treaty once the treaty has entered into force. This is a serious responsibility that countries need to examine and prepare for in advance. Countries will likely have different priorities and requirements as well as widely diverse national capabilities to monitor the CTBT. Few countries have established the national resources or expertise needed to efficiently verify compliance with the treaty. In this book we describe the verification tools that are rapidly developing and argue that the time has come for countries to act on their priorities, either individually or in cooperation with other like-minded countries. The CTBT provides states with two verification elements: an International Monitoring System (IMS) and an on-site inspection (OSI) regime. The IMS, composed of 321 seismological, radionuclide, hydroacoustic, and infrasound stations, is approaching completion and

O. Dahlman et al., Detect and Deter: Can Countries Verify the Nuclear Test Ban?, DOI 10.1007/978-94-007-1676-6_0, © Springer Science+Business Media B.V. 2011

1

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

provides equal coverage of the world, to the extent possible. A country is likely to focus its monitoring efforts on one or a few countries of concern, or on limited areas of those countries. Such focused efforts, which we call “precision monitoring,” can improve the detection and deterrence capabilities far beyond those of the IMS. In addition to the information from the IMS, states can utilize data from the thousands of seismic and other stations that are not part of the IMS, as well as modern techniques to analyze, integrate, and exploit different kinds of data that are optimal for the chosen target areas. Countries can also make use of observations from readily available high-resolution satellites, and from other national technical means. Under the treaty, states are allowed to present results derived from such data in a request for an OSI. The CTBT OSI regime is the most far-reaching inspection regime of any international arms control treaty. It adds to the deterrence against clandestine nuclear explosions and is an important measure to determine compliance. An OSI can be requested by a state party in order to clarify whether a nuclear explosion has been carried out in violation of the treaty and to gather facts to help identify the possible violator. An inspection area of up to 1,000 square kilometers can be requested and, once an inspection is approved by the Executive Council of the CTBT Organization (CTBTO), the international inspection team can deploy sophisticated inspection equipment for an inspection period of up to 130 days. We suggest that countries having similar political agendas and concerns regarding the CTBT create joint verification centers, perhaps on a regional basis. Such cooperation will make it possible for many countries to take an active part in CTBT monitoring and will increase engagement globally. This will also help create a global knowledge base that will be essential when addressing questions related to OSI. The European Union, having a common foreign and security policy, and other regional organizations such as those related to nuclear-weapon-free zone treaties, might be well served by such joint centers. The impressive developments in monitoring technologies were reflected in the findings of the International Scientific Studies Conference in 2009 (ISS09) organized by the Provisional Technical Secretariat of the CTBTO. The ISS09 Conference engaged hundreds of scientists from around the world to address scientific and technical issues related to the verification of the CTBT. They then presented their results at a conference of 600 scientific experts, diplomats and academics from 99 countries in Vienna in June 2009. The scientific contributions to that conference provided essential background material for this book. We summarize these scientific and technological developments and discuss how they can be used by countries to improve their monitoring capabilities and the effectiveness of OSIs.

2

Introduction

This book approaches CTBT verification from the perspective of the states parties and discusses their roles and functions. The CTBTO established by the treaty and its future Technical Secretariat will have the essential role of supporting states parties in achieving the object and purpose of the treaty. A book on the tasks and functions of the CTBTO and its Technical Secretariat was published by three of us in 2009. This book has five main elements. Chapter 1 provides a review of important recent political developments in international security and arms control and how these may relate to the CTBT. This chapter also summarizes the political positions on the CTBT of the nine states whose ratification is still needed for the entry into force of the treaty. Chapters 2 through 4 examine technologies and capabilities to monitor explosions under ground, in the atmosphere, and under water. The monitoring and identification of underground explosions—in our view the key verification issue—are discussed comprehensively in Chapter 2, with a focus on seismological and xenon observations and the synergy between the two. Benefiting from the concept of precision monitoring, a state can achieve a very high capability to detect even weak events and locate and interpret them with high precision in areas of concern or interest. Chapters 3 and 4 discuss the monitoring of explosions in the atmosphere and under water, respectively. We share the prevailing view that test explosions in the atmosphere as well as the oceans are unlikely, and show that there is a high monitoring capability in both of these environments down to low explosion yields. We describe radionuclide, infrasound, and satellite monitoring as efficient tools for states to monitor explosions in the atmosphere. Hydroacoustic monitoring is so efficient that the oceans can be confidently monitored with only the few hydroacoustic stations that are part of the IMS. Chapter 5 is about OSI, an essential part of the verification regime from a political and technical perspective, with regard to both deterrence and the ability to clarify events that may be of concern to one or more countries. We address the many political, technical, logistic, and organizational issues related to an OSI. We promote the need for more states to become involved, provide their expertise to develop procedures, and prepare for the ability to launch properly equipped and competent inspection teams upon entry into force of the treaty. Science and technology have played an important role in the development of the CTBT verification system, and throughout this book we show how recent scientific and technological developments have improved the verification tools available to countries. In Chapter 6 we consider the synergy between CTBT verification activities and the scientific community. We also discuss how science and technology can enhance cooperation on CTBT verification among countries.

3

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

Chapter 7, the last chapter, summarizes our conclusions and suggestions in non-technical language and can be read independently of the other chapters. In this chapter, we stress that verification is a political process in the hands of the states. We further summarize the verification tools available to states: the IMS, the OSI regime, and additional technical means that a state may choose to use. We discuss how these tools can be integrated into new concepts such as precision monitoring, and how they can be used to deter non-compliance, verify compliance, and build confidence. We stress the need for global engagement by countries in verifying the treaty and argue that it would be cost effective for countries that share similar political goals and concerns to create joint verification centers, perhaps on a regional basis. The book is intended for professionals in the political, diplomatic, scientific, and military fields who deal with international security, non-proliferation, arms control and disarmament, in particular with the implementation of such agreements. It is also intended for non-governmental organizations and journalists seeking a deeper understanding of the nuclear test ban issue and how countries can effectively verify compliance with the CTBT. This book could also be used as a textbook to train diplomats and other experts dealing with the role of science and technology in international security.

4

Chapter 1

SETTING THE POLITICAL STAGE 1.1 Introduction Calls for the cessation of nuclear testing can be traced back to the beginning of the nuclear age. Over the years a number of attempts to negotiate an end to testing failed, usually because of an inability to agree on verification provisions, in particular on-site inspections. However, in 1963, the United States, United Kingdom, and Soviet Union negotiated the Treaty Banning Nuclear Weapon Tests in the Atmosphere, Outer Space and Under Water, known as the Partial Test Ban Treaty (PTBT), a precursor to the Comprehensive Nuclear-Test-Ban Treaty (CTBT). Because underground testing was excluded from the ban, onsite inspections were not called for. However, the PTBT did include in its Preamble and Article I a commitment to negotiate “the permanent banning of all nuclear test explosions” (PTBT 1963). In 1974 the United States and the Soviet Union signed the Threshold Test Ban Treaty (TTBT), which prohibited tests having a yield exceeding a threshold of 150 kilotons (equivalent to 150,000 tons of TNT). Although the treaty did not enter into force until the two countries completed a verification protocol in 1990, both parties to the TTBT also undertook an obligation in the Preamble and Article I to continue negotiations toward the cessation of all underground nuclear weapon tests (TTBT 1974/1990). Following negotiations at the Conference on Disarmament (CD) in Geneva from 1994 to 1996 (Figure 1.1), the United Nations General

O. Dahlman et al., Detect and Deter: Can Countries Verify the Nuclear Test Ban?, DOI 10.1007/978-94-007-1676-6_1, © Springer Science+Business Media B.V. 2011

5

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

Fig. 1.1  The Council Chamber of the Palais des Nations, venue for the League of Nations; the 1955 Four Power Conference on the reunification of Germany; and negotiations of the NPT, the Chemical Weapons Convention, and the CTBT.

Assembly adopted the CTBT (CTBT 1996) in September 1996 by a vote of 158 to 3 (Bhutan, India, and Libya), with five abstentions (Cuba, Lebanon, Mauritius, Syria, and Tanzania) (UN 1996). Since the conclusion of the negotiations (Figure 1.2), there has been a moratorium on nuclear testing among the five nuclear weapon states recognized by the Treaty on the NonProliferation of Nuclear Weapons (NPT). The five include China, France, the Russian Federation, the United Kingdom and the United States. India and Pakistan both conducted a number of nuclear weapon tests in May 1998, and the Democratic People’s Republic of Korea (DPRK) conducted two nuclear weapon tests, in 2006 and 2009. Following the tests by those three countries, the United Nations Security Council unanimously adopted resolutions condemning them. In the case of the DPRK, the Council acted under Chapter VII of the UN Charter and imposed a range of strict sanctions in both instances (UNSC 2006, UNSC 2009a). In the case of India and Pakistan, the resolution (UNSC 1998) led to sanctions by the United States, and 14 countries suspended bilateral aid programs to both countries. Why is a ban on nuclear testing desirable? While a country could develop a basic bomb like the one used in Hiroshima without testing, and the nuclear weapon states sustain their weapon arsenals by means other than testing, a treaty such as the CTBT makes it difficult for a state to develop advanced nuclear weapons (Perry and Scowcroft 2009). This impedes a nuclear arms race and is seen as a measure to strengthen the NPT, which calls for nuclear disarmament in Article VI. The Preamble of the CTBT also notes that the treaty could contribute to the protection of the environment.

6

Setting the Political Stage

As of June 2011, 153 countries have ratified the CTBT, including each of the European Union (EU) countries, three of the five recognized nuclear weapon states, and 82 of the 118 members of the Non-Aligned Movement (NAM); all but seven of the 114 signatories of the nuclear-weapon-free zone treaties have signed the CTBT. Yet nine more countries specified in the treaty must ratify the CTBT for it to enter into force, as will be discussed below: China, the DPRK, Egypt, India, Indonesia, Iran, Israel, Pakistan, and the United States (Mackby 2011). 1.2  Provisions of the CTBT The basic obligations of the CTBT are contained in Article I of the treaty: 1. Each State Party undertakes not to carry out any nuclear weapon test explosion or any other nuclear explosion, and to prohibit and prevent any such nuclear explosion at any place under its jurisdiction or control. 2. Each State Party undertakes, furthermore, to refrain from causing, encouraging, or in any way participating in the carrying out of any nuclear weapon test explosion or any other nuclear explosion. The treaty provides for the establishment of a Comprehensive NuclearTest-Ban Treaty Organization (CTBTO) in Vienna to implement the treaty’s provisions and support countries in their efforts to verify compliance with the treaty. The CTBTO will include a Conference of the States Parties, the principal decision-making organ, which will meet annually, and an Executive Council to promote implementation of and compliance with

Fig. 1.2  Between meetings of the negotiations of the CTBT in 1996 (left to right): Ambassador Munir Akram of Pakistan and Ambassador Sha Zukang of China; Indian Ambassador Arundhati Ghose and U.S. Ambassador Stephen Ledogar.

7

Fig. 1.3  The 321 seismic, radionuclide, hydroacoustic, and infrasound stations of the International Monitoring System are distributed around the globe. In addition, 16 radionuclide laboratories analyze the samples of particular interest.

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

8

Setting the Political Stage

the treaty, including approving requests for on-site inspections (OSI). A Technical Secretariat will assist the states parties, the Conference, and the Executive Council in the implementation of the treaty, in particular in regard to the verification regime. A Preparatory Commission (PrepCom) for the CTBTO with a Provisional Technical Secretariat (PTS) was established in 1997 to carry out the necessary preparations to implement the operation of the treaty’s verification regime. The aim is to make a seamless transition to the CTBTO upon entry into force. The verification provisions of the CTBT are more far-reaching than those of other treaties. The CTBT provides for an International Monitoring System (IMS)—comprised of 337 high-quality stations and laboratories in 89 countries—to monitor for compliance (Figure 1.3). The IMS includes 50 primary and 120 auxiliary seismological stations to detect seismic events. The IMS radionuclide network is the first global network, comprising 80 radionuclide stations, 40 of which will be capable of detecting noble gases upon entry into force of the treaty; 16 radionuclide laboratories analyze samples of filters from the stations. In addition, a unique network of 60 infrasound stations are designed to detect nuclear explosions conducted

Fig. 1.4  A schematic depiction of CTBT monitoring illustrates sensors of the International Monitoring System providing data via a global communications infrastructure to the International Data Center (IDC) in Vienna. The IDC then transmits results of its analysis to the national authorities.

9

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

in the atmosphere, and 11 hydroacoustic stations are to detect such explosions in the oceans (Figure 1.4). The IMS stations are to be certified according to specifications agreed upon by the PrepCom. More than 85 percent of the IMS stations have been installed, and almost 80 percent were certified by the end of 2010. Under the treaty, data from these stations of the IMS are transmitted in real time via a global communications infrastructure to an International Data Center (IDC) in Vienna, and the data from the individual stations are authenticated to ensure they are not manipulated. The IDC is to receive, process, and analyze the data in a standardized way to produce bulletins containing information about, inter alia, the origin time, location, and strength of detected events. These bulletins are then sent to states parties. The IDC has been operating and sending bulletins provisionally since September 1999. How countries can verify compliance with the treaty is more thoroughly examined in Chapters 2 through 7 of this book. In addition to data provided by the IMS, countries are able to use data from thousands of other stations located around the world that are not part of the IMS. Countries can also use information from other technologies, such as satellites and other intelligence assets, as part of their national technical means of verification. If there is a concern about possible non-compliance with the treaty, countries may first request a process of consultation and clarification, in which the Director-General and the Executive Council are to assist by providing relevant information. In addition, countries will be able to request an OSI to clarify if a nuclear weapon test explosion was carried out. Such a request would be based on information from the IMS or derive from national technical means of verification. The area of an OSI will not exceed 1,000 square kilometers, and the duration will not exceed 60 days, unless the Executive Council authorizes an extension of a maximum of 70 more days. The decision to conduct an OSI will require at least 30 affirmative votes of the 51 members of the Executive Council. These and other issues regarding OSI are further examined in Chapter 5. In respect to verification, the CTBT stands in contrast to a number of other international treaties that control the spread of weapons of mass destruction (WMDs) and contain limited or no provisions for verification. This is true for the PTBT, considered the predecessor of the CTBT, which prohibited the testing of nuclear weapons in all environments except underground (Figure 1.5). Efforts to develop a verification protocol for the Biological Weapons Convention (BWC 1972) were rejected in 2001 after six years of negotiations (Bolton 2002, Findlay 2006). Other agreements, including the Antarctic Treaty (of 1959), the Outer Space Treaty (of 1967), and the Seabed Treaty (of 1971) denuclearize and demilitarize specific

10

Setting the Political Stage

Fig. 1.5  President John F. Kennedy signs the Partial Test Ban Treaty in 1963, flanked by Vice President Lyndon B. Johnson, Secretary of State Dean Rusk, and U.S. senators.

areas of the globe as well as outer space, but do not include provisions for verification, although the Antarctic Treaty contains provisions for inspections without an administrative institution to implement them. The five treaties establishing nuclear-weapon-free zones in Latin America, the South Pacific, South East Asia, Africa, and Central Asia contain no specific provisions regarding verification beyond the safeguards put in place by the International Atomic Energy Agency (IAEA), though they prohibit the testing of nuclear weapons. The NPT includes a provision for safeguards in an agreement between each non-nuclear state party and the IAEA. An Additional Protocol, which grants the Agency broader information and access rights, has been adopted by more than 100 countries (IAEA 2011). If routine inspections are not sufficient, the IAEA may request a special inspection; however, this has rarely been exercised, as examined in Chapter 5. Verification of the Chemical Weapons Convention (CWC), which entered into force in 1997, relies primarily on routine on-site inspections. The right to request a challenge inspection of a state party has never been exercised under the CWC. A number of treaties between the United States and the Soviet Union or the Russian Federation have extensive verification provisions. The TTBT, which entered into force in 1990, the Intermediate-Range Nuclear Forces Treaty (INF), which entered into force in 1988, the Strategic Arms Reduction Treaty (START), which entered into force in 1994, and New START, which entered into force in 2011, include detailed inspection and counting provisions. However, the Strategic Offensive Reductions Treaty (also known as the Moscow Treaty), which entered into force in 2003, contains no specific verification provisions.

11

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

1.2.1  Preparatory Commission Phase The CTBTO PrepCom is meant to be a temporary organization that will present a final report on the operational readiness of the regime to the first session of the Conference of the States Parties upon entry into force of the CTBT. The PrepCom is to carry out the progressive commissioning, technical testing, and provisional operation of the IMS and IDC; assure support of certified laboratories and means of communications; prepare for the conduct of OSIs; and, if requested by States Signatories, provide legal and technical advice to facilitate ratification. It is also entrusted with the development of operational manuals for the seismological, radionuclide, hydroacoustic, and infrasound monitoring, as well as manuals for the operation of the IDC and the carrying out of OSIs; these manuals are to be adopted by the first session of the Conference of the States Parties. The PrepCom thus oversees the implementation of the verification regime, adopts an annual program and budget, and has developed administrative and financial regulations. As noted, the PrepCom established a PTS and appointed an Executive Secretary to assist with its activities at the Vienna International Center in Vienna, where states members meet twice a year. There members consider questions in two Working Groups, one related to legal and administrative matters (Working Group A) and the other to establishing the verification regime (Working Group B), which meet in extended sessions to determine working plans and consider the progress made toward implementation of the treaty. The Working Groups’ recommendations, once approved by the PrepCom, delegate responsibilities to the PTS. The PTS provisionally operates the IMS and the IDC, maintains a global communications infrastructure, assists states members with the installation and operation of monitoring facilities, and reports back to the PrepCom on its progress in implementing the verification regime. Pursuant to Article XIV of the CTBT, a biannual conference (called the Article XIV Conference) has been held since 1999 to consider measures that countries can undertake to accelerate the ratification process and the treaty’s entry into force. Because of the difficulties involved in ratification in certain countries, the CTBT has taken longer than anticipated to enter into force, and the PrepCom has been in existence longer than expected. The capabilities of the verification regime and issues related to sustaining it until entry into force are examined in greater detail in the rest of this book.

12

Setting the Political Stage

1.2.2  Entry into Force To enter into force, the CTBT requires the ratification of 44 specific countries. The 44 include those which participated in the negotiations in the CD and which had nuclear power and research reactors in 1996, as listed in Annex 2 of the CTBT (Figure 1.6). Of the nine previously mentioned countries that are still required to ratify before the treaty will enter into force, all but three have signed the CTBT: the DPRK, India, and Pakistan.

Ratifications for Entry into Force: 44 States Algeria 11/7/03

Bulgaria 29/9/99

Finland 15/1/99

ISRAEL

PAKISTAN

South Korea

USA Vietnam

24/9/99

Argentina

Canada

France

Italy

Peru

Spain

Australia

Chile

Germany

Japan

Poland

Sweden

Austria

CHINA

Hungary

Mexico

Romania

Switzerland

Bangladesh

Colombia

INDIA

Netherlands

Russia

Turkey

4/12/98

9/7/98

13/3/98

18/12/98

12/7/00

6/4/98

20/8/98

13/7/99

1/2/99

8/7/97

5/10/99

12/11/97

25/5/99

5/10/99

31/7/98

10/3/06

2/12/98

1/10/99 8/3/00

29/1/08

23/3/99

30/6/00

16/2/00

Belgium

Congo DR INDONESIA

NORTH KOREA

Slovakia

Ukraine

Brazil

EGYPT

Norway

South Africa

UK

29/6/99

24/7/98

28/9/04

IRAN

15/7/99

3/3/98

23/2/01

6/4/98

30/3/99

Fig. 1.6  Ratifications required for entry into force of the CTBT, pursuant to Article XIV.

China China signed the CTBT the day it opened for signature in September 1996. The treaty has been in the National People’s Congress for the process of ratification since 2000. Although China has not yet ratified the treaty, the representative said at the UN General Assembly in 2008, “China commits itself to the early ratification of the CTBT. ... Before the entry into force of the CTBT, China will honor its commitment of moratorium on nuclear test[ing]” (Kang 2008). China has expressed its desire for early entry into force at numerous international conferences, including the 2010 NPT Review Conference (Li 2010). When Chinese President Hu Jintao met U.S. President Barack Obama in Washington in January 2011, they issued a Joint Statement in which “both sides support early entry into force of the CTBT” and they agreed to work together to reach this goal (Joint Statement 2011).

13

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

China contributed actively to the negotiations in Geneva and has also been participating in the work of the PrepCom. It has been especially engaged in the efforts on OSI, on mobile noble gas detectors and the preparation of procedures to guide inspectors. Democratic People’s Republic of Korea (DPRK) Although the DPRK (North Korea) is considered an isolated country, the delegation at the CD participated in the negotiations on the treaty. The country’s representatives also voted yes on the United Nations resolution that adopted it; however, it did not sign the treaty. After withdrawing from the NPT in 2003, the DPRK conducted a nuclear weapon test in 2006 and again in 2009. It is under sanctions for violating UN resolutions on its nuclear activities. In November 2010 North Korea showed a new modern uranium enrichment facility to U.S. scientific experts, resulting in further debate about its nuclear activities. The DPRK has not spoken officially about the CTBT for a number of years. The six-party talks (involving the DPRK, South Korea, China, the United States, Russia, and Japan) could possibly become a venue where the issue of CTBT ratification might be discussed; however, these talks have not taken place for several years. Egypt Egypt was also active in the CTBT negotiations; one of its representatives served as Chair of the Working Group on legal issues and Friend of the Chair on the Preamble and Review of the Treaty. The country plays a key role in the Middle East as well as among the NAM, which it chairs from 2009 to 2012. It has steadfastly contended that it has been in good standing with the NPT and will not support further arms control agreements, including the CTBT, the Chemical Weapons Convention, the African Nuclear-WeaponFree Zone Treaty, or the IAEA Additional Protocol until Israel ratifies the NPT. This condition is meant to address what Egypt sees as the regional considerations associated with the Middle East and the universal adherence to related treaties, commencing with the NPT (Mubarak 2001). The decision made in 1995 to indefinitely extend the NPT included a provision to establish a Middle East zone free of nuclear weapons and other weapons of mass destruction, as well as to conclude the negotiations on the CTBT. Egypt stated at the 2009 Article XIV Conference to Facilitate Entry into Force of the CTBT that the decision about ratification of the CTBT would be linked to the positive outcome of the May 2010 NPT Review Conference, particularly regarding the issue of a zone in the Middle East. The 2010 NPT Review Conference decided that a conference should be held in 2012 on a

14

Setting the Political Stage

Middle East zone free of nuclear weapons and all other weapons of mass destruction, and, in Egypt’s view, a certain amount will depend on what can be achieved there. However, it remains to be seen whether uprisings in the Middle East in the spring of 2011 will have an impact on the 2012 Conference. On a regional level, if progress is made toward establishing a zone free of weapons of mass destruction in the Middle East, Egypt, Israel, and Iran may look at ratification of the CTBT in a different light. In addition, the DPRK, Egypt, India, Indonesia, Iran, and Pakistan are all members of the NAM, which stressed achieving universal adherence to the CTBT at its 2009 summit (NAM 2009). India India holds a historic role on the issue of nuclear testing, as Prime Minister Jawaharlal Nehru called for a nuclear test ban as early as 1954 (Figure 1.7). In 1978 India took the initiative at the 33rd session of the UN General Assembly to secure a resolution calling upon all states, in particular all nuclear weapon states, to refrain from testing, pending the conclusion of a comprehensive test ban treaty. Although India has not joined the NPT, it did sign the PTBT. The country’s representatives actively participated in the negotiations on the CTBT in Geneva, serving as Friend of the Chair on seismic techniques and on the future organization to implement the treaty. One of the aspirations of India was to include in the treaty a time-bound framework for nuclear disarmament, and this was not forthcoming. In addition, when the negotiators placed India on the list of those required for entry into force, the Indian ambassador said that the country would not accept any language in the treaty that would affect its sovereign right to decide, in the light of its supreme national interest, whether it should accede to a treaty. Fig. 1.7  Prime Minister Jawaharlal Nehru of India first called for the end to nuclear testing in 1954.

15

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

India would not sign “not now, not ever” (Frontline 2009). In June 1996 it also removed its four monitoring stations (primary and auxiliary seismic, radionuclide, and infrasound) from the list of the IMS in the protocol to the treaty. As a result, there are four monitoring stations currently listed in the protocol as “To be determined.” India continued to attend the negotiations but did not join consensus on the treaty in the CD. Since the nuclear weapon tests of India and Pakistan in 1998 they have been observing a moratorium. Prime Minister Atal Behari Vajpayee told the UN General Assembly in October 1999 that India “would not stand in the way” of the treaty coming into force. This sentiment was echoed a decade later, in December 2009, by Prime Minister Manmohan Singh when he told visiting Japanese President Yukio Hatoyama that if the United States and China ratify the CTBT it “would create a new situation” (Singh 2009). The joint statement from Prime Minister Singh and President George W. Bush issued at the beginning of the India civil nuclear cooperation agreement process on 18 July 2005 included a commitment from India to continue its unilateral moratorium on nuclear testing. This is a condition of the United States for full cooperation. However, in a subsequent response to domestic queries, Singh stated that India had the sovereign right to test (Gill 2009). Indonesia Indonesia, speaking in 2009 on behalf of the NAM at the third session of the 2010 NPT Preparatory Committee, said, “We support the objective of the CTBT, which is intended to enforce a comprehensive ban on all forms of nuclear tests without exception and to stop the development of nuclear weapons, in the direction of the total elimination of nuclear weapons” (Indonesia 2009). Indonesia said in June 2009 that it would ratify the CTBT right after the United States; however, it announced at the NPT Review Conference on 3 May 2010 that it was proceeding to initiate the process of ratification. Foreign Minister R.M. Marty M. Natalegawa stated that Indonesia shared the vision of a world free of nuclear weapons and expressed the hope that “this further demonstration of our commitment to the nuclear disarmament and nonproliferation agenda will encourage other countries that have not ratified to do so” (Natalegawa 2010). Subsequently, the foreign minister said that Indonesia was planning to play a more central role in nuclear disarmament and expected to ratify the CTBT in 2011. “God willing, Indonesia will complete the ratification process in 2011 and encourage various parties to support the implementation of the CTBT,” he said (Natalegawa 2011). As a state party to the Treaty on the Southeast Asia Nuclear-Weapon-Free Zone, Indonesia has undertaken under Article 3 not to test or use nuclear weapons.

16

Setting the Political Stage

Iran Iran participated very actively in the negotiations on the CTBT, serving as Friend of the Chair on on-site inspections report writing, follow-up action, and sanctions. Although it did not join the final consensus on the treaty in the CD, it signed the treaty on the first day it opened for signature at the United Nations. Iran voted in favor of the CTBT resolution at the 2009 General Assembly, and the country has been a strong advocate of the treaty in the PrepCom. Given the contentious situation in the UN Security Council and the IAEA surrounding Iran’s nuclear activities, Iran has not voiced its views on CTBT ratification. In 2012 Iran will become the next chair of the NAM, which has voiced its support of the CTBT on a number of occasions. Israel Israel has been actively engaged in both the negotiations and the PrepCom in Vienna. As an Observer in the CD negotiations, it was not allowed to break consensus, but it could contribute papers and thereby make its contributions felt in the treaty, in particular on the issue of OSI. Israel’s considerations for ratification have been expressed on a number of occasions and include the non-abusive nature of the on-site inspection regime; equal status in the policy-making organs of the organization, in particular the Executive Council; and adherence to the treaty by other Middle Eastern countries. The treaty stipulates that the Executive Council will make the decision about whether an OSI will take place, and this body will be composed of six geographical regions, the composition of which is enumerated in the annex to the treaty. Israel is in the Middle East and South Asia regional group on the Executive Council. The equal status of Israel in the policymaking organs of the treaty was a strong requirement in the end game of the CTBT negotiations in 1996 and it remains so today. Due to the political situation in the area, countries in that group have been unable to meet, although this issue has been addressed in a number of conferences. During the PrepCom phase, as during the negotiations, Israel has been especially interested in conditions for on-site inspections, indicating that it was concerned that a frivolous charge might lead to an OSI and might be used to gain access to sensitive information (Ramaker et al. 2003). This said, it was among the first to sign the treaty. In the deliberations in the PrepCom in Vienna it has also focused on the IMS and IDC build-up.

17

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

Pakistan Pakistan was also actively involved in the CTBT negotiations. Like India, Pakistan wanted the goal of nuclear disarmament to be included in the treaty. It strongly endorsed provisions for entry into force that would include the eight nuclear-capable states, as did a number of others in the negotiations. As Ambassador Munir Akram said at the time, “To those who live in the real world, it is clear that if one of those states is out of the treaty, all of them will be out.… Those who sincerely desire an early entry into force with or without these eight states ignore fundamental strategic and political realities” (Ramaker et al. 2003). Contrary to India, Pakistan voted in favor of the CTBT at the United Nations. Not long thereafter, in May 1998, India carried out a number of nuclear weapon tests, and Pakistan followed suit. After the 1998 tests of both countries, Prime Minister Nawaz Sharif told the UN General Assembly, “Pakistan has consistently supported the conclusion of a CTBT for over 30 years. ... There is no reason why the two countries cannot adhere to the CTBT. In a nuclearized South Asia, CTBT would have relevance if Pakistan and India are both parties to the Treaty” (Sharif 1998). Pakistan repeated at the United Nations in 2005 and 2007 that it would not sign the treaty unilaterally. Nevertheless, since the U.S.-India civil nuclear deal, the linkage of actions by Pakistan to those of India may not be a foregone conclusion. “The conclusion of the U.S.-India nuclear agreement, which fails to extract any favorable commitment from India about its intentions towards the CTBT, has not created optimism in Islamabad about its prospects,” said former Ambassador Shahbaz of the Permanent Observer Mission of Pakistan to the PrepCom. “The deal upset our threat perception by aggravating the imbalance in our capabilities” (Personal communication 2010). Pakistan is following the developments regarding the prospects of the CTBT entry into force, and when it sees movement it will review its policy in the interest of regional peace and security, Shahbaz added. Pakistan is maintaining a unilateral moratorium on nuclear testing and has said that it will not be the first to resume testing. Since the establishment of the CTBTO PrepCom, Pakistan has attended a number of the meetings of the Working Group on Verification as an observer and participated as observer at the 1999, 2007, and 2009 Article XIV Conferences. United States When the treaty opened for signature at the United Nations in September 1996, President Bill Clinton was the first to sign it. However, when the U.S.

18

Setting the Political Stage

Senate declined to provide its advice and consent to ratification in 1999, the treaty lost a sense of urgency in the United States as well as in much of the international community. Subsequently, the George W. Bush administration stated its opposition to the treaty. In the following administration, President Barack Obama, in a benchmark speech in Prague in April 2009, called for the elimination of nuclear weapons and said that he would “immediately and aggressively” pursue U.S. ratification of the CTBT (Obama 2009). An examination of the issues surrounding the ratification and entry into force of the CTBT revolves around a number of key questions, most of which are captured in the 2009 report of the Congressional Commission on the Strategic Posture of the United States, a bipartisan, congressionally appointed group chaired by former Secretary of Defense William J. Perry (Congressional Commission 2009). The CTBT is the only item on which the Commission did not reach a consensus view, and thus the arguments presented by both sides illustrate the debate in other circles, in particular the U.S. Senate, regarding the ratification of the treaty. For example, commissioners opposing the treaty believed that a CTBT would diminish the confidence in the reliability of the U.S. nuclear weapons stockpile, thereby reducing the credibility of America’s nuclear deterrent. Proponents argue that the Stockpile Stewardship Program (SSP) has ensured that the United States can maintain a safe, secure, and reliable stockpile without testing. Opponents further believe that a zero-yield ban is unverifiable and that countries could conduct tests without being detected. Treaty supporters maintain that the CTBT is effectively verifiable and that potential violators could extract little, if any, military value from clandestine testing at levels that are undetectable. In this regard, the 2010 Department of State report on compliance with arms control agreements said that there were no indications during the reporting period, 2004– 2008, that any of the NPT nuclear weapon states “engaged in activities inconsistent with its declared moratorium” (State Department 2010). In spite of the lack of agreement on the CTBT, the Congressional Commission recommended a number of actions to prepare the way for a new Senate review of the CTBT, including securing agreement among the five nuclear weapon states recognized by the NPT on a definition of the activities that are banned and permitted under the treaty and defining a diplomatic strategy for securing entry into force. “Many of the members of the commission would strongly support the CTBT if there were clarification,” Vice Chair James Schlesinger said (Grossman 2009). In another approach, former Los Alamos National Laboratory director Sigfried Hecker said, “The single most important reason to ratify the CTBT is to stop other countries from improving their arsenals. ... [W]e gain substantially more from limiting other countries than we lose by giving up testing” (Weeks 2009).

19

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

As for concerns about the lifetime of nuclear stockpiles, a 2007 report by the JASON Group, a leading independent scientific advisory group established in 1960 to provide consulting services to the U.S. government on defense science and technology matters, states that “the primaries of most weapons system types in the stockpile have credible minimum lifetimes in excess of 100 years and that the intrinsic lifetime of Pu in the pits is greater than a century” (JASON 2007). The SSP, life extension programs (LEP), and computer simulations have extended the reliability of the weapons. The administrator of the National Nuclear Security Administration (NNSA), Tom D’Agostino, said, “The SSP over the past decade has provided improved scientific and analytic tools, including advanced supercomputer simulation and sophisticated experimental capabilities, which were not available to the previous generation of designers/engineers. ... We know more about the complex issues of nuclear weapons performance today than we ever did during the period of nuclear testing” (D’Agostino 2008). A technical analysis of the treaty by the U.S. National Academy of Sciences in 2002 stated, “The worst-case scenario under a no-CTBT regime poses far bigger threats to U.S. security interests-sophisticated nuclear weapons in the hands of many more adversaries-than the worst-case scenario of clandestine testing in a CTBT regime, within the constraints posed by the monitoring system” (NAS 2002a). The Department of State, the NNSA, and the National Academy of Sciences sponsored a project (NAS 2010) to review and update the 2002 report on technical issues related to the treaty, to be published in 2011. More recently, on 6 April 2010, the United States released the third Nuclear Posture Review, a comprehensive review of U.S. nuclear weapons strategy and policy for the next five to 10 years. As regards the CTBT, the review stated that ratification of the CTBT “is central to leading other nuclear weapons states toward a world of diminished reliance on nuclear weapons, reduced nuclear competition and eventual nuclear disarmament” (DOD 2010a). Precedents It is worth noting that there are precedents for late arrivals in the arms control arena. The landmark nuclear arms control treaty, the NPT, did not obtain the ratification of certain key countries, most notably China and France, until 1992, more than 20 years after it opened for signature and well after most other countries had ratified. This is significant because the central NPT tenet not to transfer nuclear weapons is directed at the five acknowledged nuclear weapon states, which include China and France. Neither China nor France participated in the NPT negotiations. The same

20

Setting the Political Stage

two countries did not join the PTBT of 1963 (China became a nuclear weapon state in 1964). Yet both countries were among the first to sign the CTBT, and France ratified it in April 1998. The NPT entered into force in 1970, upon ratification of 40 countries in addition to the depositories—the United States, United Kingdom, and Soviet Union. Although the TTBT was signed in 1974, it did not enter into force until 1990, following additional negotiations on the verification protocol. As noted, all of the nine countries that still need to ratify the CTBT for it to enter into force participated actively in the negotiations. With a moratorium on nuclear testing among the five for more than 14 years, and the subsequent moratorium by India and Pakistan, there is an established norm of non-testing in effect, with the exception of the DPRK. 1.2.3  Scope — Zero Yield In the negotiations on the CTBT from 1994 to 1996, delegates recalled that in the NPT and the PTBT negotiations of the 1960s it was not possible to define a nuclear weapon test explosion. There are no definitions in those treaties, and this absence of definitions was not questioned over the years. As for the CTBT, there was concern that to define a nuclear weapon test in technical terms in a treaty could expose sensitive information and would ultimately prove unnecessary because it would take so much negotiating time and effort. The decision was made early in the negotiations in the Conference on Disarmament not to define nuclear weapon test explosion (Ramaker et al. 2003). The lack of such a definition in the CTBT has become a contentious issue in the United States (Kyl 2009), although the U.S. State Department’s article-by-article analysis of the treaty says, “The U.S. decided at the outset of negotiations that it was unnecessary, and probably would be problematic, to seek to include a definition in the Treaty text of a ‘nuclear weapon test explosion or any other nuclear explosion’ for the purpose of specifying in technical terms what is prohibited by the Treaty” (U.S. Government 1997). During the negotiations, the five nuclear weapon states consulted often among themselves on the scope of the treaty. Although these consultations were shrouded in secrecy, the general parameters of the discussions became well known: desires ranged from allowing four pounds of nuclear yield to several hundred tons (Ramaker et al. 2003). However, they finally agreed that the language in Article I of the treaty should mean that no tests that produced a nuclear yield should be allowed to anyone under the treaty, i.e., there should be a zero yield. The five announced this decision to all 60 Members plus Observers sitting in the Conference on Disarmament

21

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

on different dates, as follows: France on 10 August 1995, the United States on 11 August 1995, the United Kingdom on 14 September 1995, the Russian Federation on 23 October 1995, and China on 21 March 1996 (Ramaker et al. 2003). On 7 October 1999 Ambassador Stephen Ledogar, Chief U.S. negotiator for the CTBT, testified before the Senate Foreign Relations Committee that Russia and China had committed themselves to a zero-yield ban. He said that fact was “substantiated by the record of the negotiations at almost any level of technicality (and national security classification) that is desired and permitted. A ban should be a ban. The answer to this dilemma should be no threshold for anybody, i.e., zero means zero. ... If what you did produced any yield whatsoever, it was not allowed. If it didn’t, it was allowed” (U.S. Senate 1999). In the same hearing Secretary of State Madeleine Albright expressed the same view. Russian negotiator in the Conference on Disarmament, Victor Slipchenko, recently suggested that ratification by the United States corresponds to Russia’s interests. He said that it might be possible to confirm at a high level the official position from the ratification of the treaty in the State Duma in 2000 by Foreign Ministry official Yuri Kapralov that “in accordance with the CTBT, all test explosions of nuclear weapons are banned, including hydro-nuclear experiments, whatever the level of energy release” (Slipchenko 2009). Russian President Dmitry Medvedev noted that “Under the global ban on nuclear tests, we can only use computer-assisted simulations to ensure the reliability of Russia’s nuclear deterrent” (Global Security Newswire 2009). The chief negotiator of Australia, Ambassador Richard Starr, who played a leading role in the formulation of the language on scope, said, “I had no doubt whatsoever that what we were promoting meant zero yield, and we understood this was accepted by each of the nuclear weapon states when they announced their intention to adhere to this formula. Our work in Geneva was backed up by talks in the major capitals, including each of the nuclear weapon states” (Personal communication 2010). Despite the clear political declaration by all five nuclear weapon countries, there has been a great deal of discussion about nuclear weapon–related experiments. It is important not to lose track of what the treaty does and does not prohibit. In simple terms, as one ambassador in the negotiations put it, “We’re banning the bang, not the bomb.” A number of experiments are carried out with the stated purpose of securing the safety and reliability of existing weapons and weapon designs. These subcritical, hydrodynamic and other experiments are discussed in Chapter 2.

22

Setting the Political Stage

1.3  Testing as Part of Nuclear Weapon Development During a period spanning more than five decades, 2,052 nuclear explosions were carried out in the atmosphere, under ground, under water, and in outer space. Some 1,500 of these took place under ground. There are slight inconsistencies regarding the numbers of explosions reported in different publications, however, the overall picture is much the same. The United States and the former Soviet Union conducted most of them, with the United States conducting 1,030 tests and the Soviet Union 715. France conducted 210 tests, and the United Kingdom and China each carried out 45 tests. India announced it had conducted a nuclear explosion for peaceful purposes in 1974. In May 1998 it announced two tests involving five nuclear devices, and in the same month Pakistan announced that it had conducted two tests involving a total of six nuclear devices (a test may involve two or more nuclear devices detonated simultaneously).

Fig. 1.8  Test sites of the world: 1. Amchitka Island, Alaska (U.S.), 2. Central Nevada Test Area (U.S.), 3. Nevada Test Site (U.S.), 4. Fallon, Nevada (U.S.), 5. Trinity Site, New Mexico (U.S.), 6. Hattiesburg, Mississippi (U.S.), 7. Reggan, Sahara Desert, Algeria (France), 8. In Ekker, Algeria (France), 9. Novya Zemlya (USSR), 10. Semipalatinsk (USSR), 11. Lop Nor, Western China (China), 12. Chagain Hills, Baluchistan (Pakistan), 13. Pokharan, Rajastan Desert (India), 14. Monte Bello Islands, Australia (UK), 15. Emu Field, Australia (UK), 16. Maralinga and Woomera Test Sites, Australia (UK), 17. Eniwetok Atoll, Marshall Islands (U.S.), 18. Bikini Atoll, Marshall Islands (U.S.), 19. Johnston Island (U.S.), 20. Eastern Pacific Ocean (U.S.), 21. CEP, Muroroa and Fangataufa Atolls, French Polynesia (France), 22. Christmas Island, Kiribati (U.K. and U.S.), 23. P’unggye (North Korea). The U.S. carried out three atmospheric tests in the South Atlantic Ocean (Operation Argus). The Soviet Union conducted peaceful nuclear explosions at more than 46 sites on its territory and the U.S. did so in five locations in the United States.

23

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

North Korea has announced two tests, and Israel is not believed to have tested (Dahlman et al. 2009). Some 517 nuclear tests were conducted in the atmosphere by the nuclear weapon states recognized under the NPT. The PTBT of 1963 prohibited nuclear weapon testing in the atmosphere, although, as mentioned, not all countries signed it. Nuclear weapon testing was conducted in more than 20 locations around the world (Figure 1.8). Subsequently, tests were confined to sites in the United States, Soviet Union, China, and French Polynesia; India, Pakistan, and the DPRK tested within their territories. The purpose of most tests was to research and refine new nuclear weapons or to study weapons effects (DOE 2000). The non-nuclear components can be tested and replaced without detonating a warhead (Garwin 2011). The Soviet Union and the United States carried out what are called peaceful nuclear explosions (PNEs) for various purposes: to create dams, stimulate oil and gas recovery, study the Earth’s structure, and produce underground storage space, among others. From 1965 to 1988 the Soviet Union conducted 124 PNEs, some of which involved multiple devices, at more than 46 sites (Mikhailov 1996, Nordyke 2000). The United States conducted 27 such explosions, three of which involved multiple nuclear devices, from 1961 until 1973. All but four of these were conducted at the Nevada Test Site (renamed the Nevada National Security Site in 2010). Since the Cold War, the number of nuclear warheads has been greatly reduced. In 1986, the Soviet Union held some 45,000 nuclear weapons and the United States had 24,400; as of 2011, the estimated numbers were 12,000 and 9,400, respectively (Kristensen 2010). In May 2010 the United States revealed the number of nuclear weapons it held available for use in war: 5,113 as of 30 September 2009 (DOD 2010b). The New START treaty signed by the United States and Russian Federation in Prague on 8 April 2010 limits their numbers of deployed warheads to 1,550 and deployed intercontinental ballistic missiles (ICBMs) and submarine launched ballistic missiles (SLBMs) to 700, with a combined limit of 800 deployed and non-deployed ICBM launchers, SLBM launchers, and heavy bombers equipped for nuclear armaments. They also declared their intention to follow this agreement with further reductions. France announced that it holds 300 operationally deployed warheads. The United Kingdom announced in its Strategic Defence and Security Review of October 2010 that it will reduce its stockpile of operationally deployed warheads from fewer than 160 to no more than 120 and the overall nuclear warhead stockpile from not more than 225 to not more than 180 by the mid-2020s (U.K. Government 2010). China has an estimated 240 warheads: about 175 active nuclear warheads and 65 warheads in reserve (Norris and Kristensen 2010). The nuclear arsenals of Israel, Pakistan, and India are unknown but

24

Setting the Political Stage

Fig. 1.9 About 50 B61 nuclear bombs stored under ground in the United States.

are estimated at about 80 each. North Korea is estimated to possess fewer than 10 nuclear devices. Figure 1.9 shows arsenals in underground storage. 1.4  Political Development There has been significant political development and a number of activities on nuclear-related issues. As mentioned, in his Prague speech President Obama called for the elimination of nuclear weapons and said that he would “immediately and aggressively” pursue U.S. ratification of the CTBT. Subsequently, he presided over the UN Security Council Summit in September 2009 that unanimously adopted Resolution 1887, which covered a number of actions to strengthen the NPT (UNSC 2009b). The resolution calls on states to “refrain from conducting a nuclear test explosion and to sign and ratify the Comprehensive Nuclear-Test-Ban Treaty (CTBT), thereby bringing the treaty into force at an early date” (UNSC 2009c). Also pursuant to the resolution, President Obama hosted a Global Nuclear Security Summit in April 2010 to focus on securing nuclear materials worldwide and combating nuclear terrorism. The summit brought together 49 world leaders—an unprecedented number—in Washington to spotlight the goal of securing all vulnerable nuclear material within four years and examine how to prevent terrorist groups from gaining nuclear materials. The danger that weapons-grade nuclear material and the technology needed to develop nuclear weapons may spread to terrorists and non-state actors has largely replaced the concerns of the Cold War. It is possible that terrorists could make a low-yield nuclear explosive device using weapons-grade or reactor-grade plutonium (NAS 2002b). Former Director-General of the IAEA Mohamed El Baradei reported in 2009, “We still have 200 cases of illicit trafficking of nuclear material a

25

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

year reported to us. ... Pretty soon you will have nine weapons states and probably another 10 or 20 virtual weapons states” (El Baradei 2009). The possible confluence of terrorism, nuclear material, and upheavals in international security has made many question the relevance of nuclear weapons in today’s security environment, and a new movement has evolved calling for the elimination of nuclear weapons. This was triggered by an op-ed in The Wall Street Journal in January 2007 by former U.S. Secretaries of State George Shultz and Henry Kissinger, former Secretary of Defense William Perry, and former Chairman of the Senate Armed Services Committee Sam Nunn. They called for eight urgent steps that would provide the basis for a world free of the nuclear threat. One of these was “to achieve ratification of the Comprehensive Test Ban Treaty, taking advantage of recent technical advances, and working to secure ratification by other key states” (Shultz et al. 2007). This was followed a year later with another Wall Street Journal oped, in which the same senior statesmen noted the interest and support for urgent action generated around the world by their first article. They went on to call for bringing the CTBT into effect, which would strengthen the NPT (Shultz et al. 2008). This was followed by articles in The Times by former U.K. foreign and defense secretaries, in Le Monde by four French statesmen, and in the Frankfurter Allgemeine Zeitung by four German statesmen, as well as articles by senior statesmen of Australia, Belgium, Italy, Japan, the Netherlands, Norway, Poland, the Russian Federation, and South Korea, along with many organizations around the world associated with this movement (e.g., Nuclear Security Project 2010, Global Zero 2010). The CTBT figures prominently among most of these articles and organizations. Countries may have a different perspective than the senior statesmen about the elimination of nuclear weapons. Under the NPT, states parties are committed to pursue nuclear disarmament, and at the NPT Review Conferences nuclear weapon states describe the activities that they have undertaken to reduce the size of their nuclear arsenals. Nevertheless, debate continues over the conditions under which the elimination of nuclear weapons will take place. As noted, President Obama endorsed the abolition of nuclear weapons, though he added, “This goal will not be achieved quickly—perhaps not in my lifetime” (Obama 2009). At the signing ceremony of the New START agreement (Figure 1.10), Russian President Dmitry Medvedev acknowledged about the two countries, “Yes, we have 90 percent of all the stockpiles which is the heritage of the Cold War legacy and we’ll do all that we have agreed upon. ... [W]e do care about what is going on with nuclear arms in other countries of the world, and we can’t imagine a situation when the Russian Federation and the

26

Setting the Political Stage

United States take efforts to disarm and the world would move towards a principled different direction away” (Medvedev 2010). China, along with the United Kingdom and France, has indicated that when the two countries with the largest numbers of nuclear weapons reduce to their levels, it will join the debate. “China has consistently stood for the complete prohibition and thorough destruction of nuclear weapons,” said the representative to the 2010 United Nations First Committee (PRC 2010). France has said that the common goal is to create the conditions that will make nuclear weapons ultimately unnecessary (France Diplomatie 2010). “Nuclear deterrence remains an essential concept of national security,” stated the 2008 French White Paper. “The sole purpose of the nuclear deterrent is to prevent any State-originating aggression against the vital interests of the nation...” (French White Paper 2008). The 2010 Strategic Defence and Security Review of the U.K. said, “It is right that the United Kingdom should retain a credible, continuous and effective minimum nuclear deterrent for as long as the global security situation makes that necessary.” It also said that as a party to the NPT, it remains committed to the long term goal of nuclear disarmament (U.K. Government 2010). 1.5  CTBT in the Context of Nuclear Non-Proliferation and Disarmament The CTBT has been linked to the NPT since the latter was negotiated in 1968. The NPT is considered the cornerstone of the international nonproliferation regime and has 189 states parties, more than any other arms control treaty. Under the NPT, the non-nuclear weapon states are obliged not to acquire nuclear weapons, while the nuclear weapon states are obliged in Article VI to “pursue negotiations in good faith on effective measures relating to cessation of the nuclear arms race at an early date and to nuclear disarmament, and on a treaty on general and complete disarmament under strict and effective international control.” The nonnuclear weapon states see the CTBT as a benchmark in that provision, a tangible goal that will constitute a good faith effort by the nuclear weapon states to work toward nuclear disarmament. The nuclear weapon states believe that the CTBT will contribute to non-proliferation. The CTBT notes both objectives in the Preamble, which says that the cessation of nuclear weapon test explosions, by constraining the development and qualitative improvement of nuclear weapons, “constitutes an effective measure of nuclear disarmament and non-proliferation.” As mentioned, at its 2009 summit, the NAM stressed the significance of universal adherence to the CTBT (NAM 2009). The CTBT has figured prominently in most Review Conferences of the NPT, and the absence of agreement on the CTBT was considered a prime reason for the failure of

27

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

the 1990 and 2005 Conferences to adopt a Final Document. The call for a CTBT also figured prominently in the 1995 decision to extend the NPT indefinitely. The five nuclear weapon states wrote a letter to the 1995 NPT Review Conference in which they pledged their support for the conclusion of the CTBT and, at the same time, called upon all states parties to the NPT to make the NPT permanent. “This will be crucial for the full realization of the goals set out in article VI,” the letter said (NPT 1995). The CTBT was included in the “13 Steps” to implement Article VI of the NPT encompassed in the final document of the 2000 Review Conference. The first step was the “urgency of signatures and ratifications, without delay and without conditions,” to achieve the early entry into force of the CTBT. The May 2010 Review Conference of the NPT heard more than 60 states parties note in their opening statements that the CTBT was key to nonproliferation. In the consensus final document the conference resolved that all nuclear weapon states “undertake to ratify the CTBT with all expediency,” noting that this would have a positive effect on others, and that they should encourage Annex 2 countries in particular to ratify. The Final Document stated, “The Conference calls on all States to refrain from any action that would defeat the object and purpose of the Comprehensive Nuclear Test-Ban Treaty pending its entry into force” (NPT 2010).

Fig. 1.10  U.S. President Barack Obama and Russian President Dmitry Medvedev sign the New START treaty at Prague Castle on 8 April 2010. The treaty limits the number of nuclear warheads to 1,550 for each country.

28

Chapter 2

Monitoring Underground Nuclear Explosions 2.1  Underground Nuclear Explosions and Weapon-Related Experiments The first underground nuclear tests were carried out in 1957 by the United States and the Soviet Union. As discussed in Chapter 1, in 1963 the Soviet Union, United Kingdom, and United States signed the Partial Test Ban Treaty (PTBT), which prohibited testing in the atmosphere, outer space, and under water, and since 1980 all nuclear explosions have been conducted under ground. Of the 2,052 nuclear explosions that have been carried out, 1,501 have been conducted under ground, though as noted in Chapter 1, there are minor divergences in the number of reported explosions. Most of the explosions involved the development and testing of nuclear weapons, although about 150 were part of programs for socalled peaceful nuclear explosions (PNEs) (Nordyke 2000). 2.1.1  Characteristics of an Underground Nuclear Explosion Underground testing is a well-established technique, with the explosive device placed in either a tunnel or a shaft built into a hill or mountainside or in a wide and deep borehole drilled into the ground. Explosions have been conducted both in hard rock, such as that found at the former Soviet test site at Semipalatinsk, and in softer rock, such as that found at the

O. Dahlman et al., Detect and Deter: Can Countries Verify the Nuclear Test Ban?, DOI 10.1007/978-94-007-1676-6_2, © Springer Science+Business Media B.V. 2011

29

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

Nevada Test Site (NTS) in the United States. An underground explosion creates a cavity the size of which is dependent on explosion yield, bedrock, and placement depth. A 1 kiloton explosion will form a melted cavity with a radius of 4 to 12 meters, depending on the structure and lithology (strength, compressibility, and sound speed) of the bedrock. The volume of the cavity increases in proportion to the energy release, or “yield,” of the explosion. The cavity, which initially is well sealed by melted rock, is surrounded by crushed and cracked rock that extends to about 10 times the cavity radius (Hawkins and Wohletz 1996). In many cases the cavity collapses, and a chimney is formed with a height several times the cavity radius. The source effects are further discussed in relation to on-site inspections in Chapter 5. Two main features of an underground explosion that are detectable at a distance are seismic signals and radionuclide gases, in particular xenon. In strength, the seismic signals are approximately proportional to the explosion yield. The strength of the signals also depends on the coupling of the shock wave to the surrounding bedrock, which in turn relates to the composition and the physical properties of the bedrock. Because of the melted rock in the walls of the cavity, radionuclide material is to a great degree trapped, if the cavity stays intact. If the cavity cracks or collapses, radioactive gases may leak out. Noble gases, like xenon, that do not easily react chemically with surrounding materials are more likely to escape from the cavity compared to other gases such as iodine. Leakages of noble gases have been observed from underground nuclear test sites in both the former Soviet Union and the United States (Dubasov 2010). A major leakage might occur if an unknown geological feature happens to be located close to the explosion, as in the case of the United States test Baneberry (Figure 2.1). In addition to observing signals generated by the explosion, satellite measurements, either photographic or by radar, can be used to observe the extensive logistic activities prior to detonation that are associated with preparations for a nuclear test. Observations by satellite may also reveal effects on the Earth’s surface above the explosion. 2.1.2  Nuclear Weapon–Related Experiments As explained in Chapter 1, there has been a great deal of discussion about the fact that there is no definition of a “nuclear explosion” in the Comprehensive Nuclear-Test-Ban Treaty (CTBT). This discussion has also addressed what nuclear weapon–related experiments the treaty does and does not prohibit. Such experiments all relate to the “physics package” that contains the nuclear material. The many non-nuclear parts of a nuclear weapon, numbering 4,000 to 6,000 (Shalikashvili 2001), can be tested without the nuclear material. The three types of nuclear-related tests

30

Monitoring Underground Nuclear Explosions

Fig. 2.1 Underground nuclear explosion Baneberry, with a yield of 10 kilotons, conducted at the Nevada Test Site on 18 December 1970. The device was exploded at a depth of about 270 meters beneath the surface. The event released radioactivity into the atmosphere, resulting in a cloud of radioactive dust that reached an altitude of 3 kilometers.

discussed below aim at increasing the understanding of the behavior of nuclear material in an explosion process. Are these tests prohibited under the CTBT? Subcritical experiments (SCE), which are conducted as part of the Stockpile Stewardship Program in the United States, Russia, and China, aim at studying the material properties and the equation of state of fissile material at high pressure. In an SCE, a small amount of fissile material is exposed to a strong shock wave generated by conventional explosives. The experiments are material-related and can be conducted in different ways using materials in a configuration with no resemblance to nuclear warheads (JASON 1997). From 1997 until 2006, 23 SCEs were reported in the United States (Medalia 2008). Two of those experiments were conducted jointly with the United Kingdom. The U.S. SCEs have all been conducted in a tunnel complex 300 meters underground at the Nevada Test Site (NTS). On 15 September 2010, the first subcritical experiment in a new U.S. series was conducted at the NTS. During a visit to the Russian test site at Novaya Zemlya on 28 June 2002, the Russian Defense Minister Sergey Ivanov announced that Russia intended to maintain that test site and continue conducting subcritical experiments at a rate of four to six

31

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

a year (NTI 2002). It has also been reported that China conducted four subcritical experiments at the Lop Nor test site in 2001 (Lewis 2009). As subcritical tests are designed not to release any fission energy, they are not banned under the CTBT. The second type of nuclear weapon–related experiment is the hydrodynamic experiment, or test. This type of experiment also explores the behaviour of material under high pressure and high temperature. A mock-up of the nuclear part of a weapon, with the fissile material replaced by a non-fissile material possessing similar material properties, can be used. The detonating high explosive sends a strong shock wave through the non-fissile nuclear material, and its behaviour is studied (Global Security 2005). As hydrodynamic experiments do not involve any fissile material, they are not banned under the CTBT. The third type of experiments, hydronuclear, involve fissile material that could result in a “very slight degree of super criticality” and the release of a small amount of fission energy (Thorn and Westervelt 1987). In 1960 and 1961, during the nuclear test moratorium, 35 such hydronuclear experiments were reported at Los Alamos and a smaller number at the NTS. The Los Alamos experiments released nuclear explosion yields ranging from less than a few grams to 100 grams of equivalent conventional explosives, compared to billions or trillions of grams or kilotons or megatons in nuclear weapons. The tests were conducted to address nuclear weapon safety issues and gain data on the behavior of the fissile materials involved. Reports indicate that hydronuclear experiments can address only a limited range of questions. A similar program of thermonuclear experiments with “no considerable nuclear energy release” was conducted by the Soviet Union at the Semipalatinsk test site from 1958 to 1989 (Mikhailov 1998). Under the CTBT, these hydronuclear and thermonuclear experiments are prohibited. Thus there is agreement among the nuclear weapon states on these nuclear material–related tests; the CTBT does not prohibit subcritical tests in which no chain reaction occurs or hydrodynamic tests in which the fissile materials are replaced by non-fissile materials. The treaty does not allow hydronuclear tests. 2.2  Seismological Monitoring Seismology is a well-established science on a global scale, and CTBT verification is only one of many applications. Seismological verification is a discipline in its own right that has been pursued for more than 50 years as part of the preparations for test ban negotiations. The Group of Scientific Experts of the Conference on Disarmament (CD) in Geneva developed

32

Monitoring Underground Nuclear Explosions

the seismological component of the verification system for the CTBT. This system also served as a template for the other networks (hydroacoustic, infrasound, and radionuclide) that are part of the International Monitoring System (IMS) (Dahlman et al. 2009). 2.2.1 Measurements Seismological measurements have been carried out for more than 100 years, and currently more than 16,000 stations are operating worldwide. A rapidly growing number of these stations are of high quality and are part of global, regional, or national networks established for earthquake monitoring, from which data are centrally collected and easily available. In some parts of the world very dense networks are established, either permanently or on a temporary basis. The USArray, for example, deploys 400 stations for a period of two years, with an inter-station separation of 70 kilometers. The stations are then moved in order to cover the whole of the United States over a period of 10 years. This will provide detailed information on the structure of the Earth and on wave propagation properties (Woodward et al. 2009). The seismological part of the IMS comprises a global network of 50 primary stations and 120 auxiliary stations (see Figure 1.3 in Chapter 1). Most of the primary stations are array stations, for which a number of sensors are deployed within an area with an extent of a few kilometers and sometimes more. This placement enhances the performance of a station and also makes it possible to characterize seismic wave propagation features. The array stations are currently the most capable stations in operation globally. The primary stations are continuously reporting online to the International Data Center (IDC) of the Provisional Technical Secretariat (PTS), and they are the basis for IDC seismic event detection. The IDC uses data from the auxiliary stations to improve on the analysis of events that have already been determined, based on data from the primary stations. Auxiliary data are requested both as part of the automatic data processing cycle and by the analysts on a case-by-case basis during the analysts’ interactive review of the data. This procedure is partly a reflection of the situation at the time of the CTBT negotiations, when there was good reason to limit the volume of data transmitted because of cost implications. However, the current version of the Global Communications Infrastructure that has been established for CTBT verification purposes has the capacity to provide all the auxiliary station data online to the IDC without added costs. The intent during the CTBT negotiations was that the auxiliary station network should be based essentially on seismic stations that already existed

33

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

for scientific purposes. During deliberations at the CTBTO PrepCom, operational technical requirements, similar to those for the primary stations, were established. To meet these requirements, the PrepCom has over the years invested a considerable amount toward upgrading many of the existing auxiliary stations. The responsibility of operating these stations and covering the associated costs rests with the host country, however. This arrangement has created some difficulties; there have been countries hosting auxiliary stations that were not prepared to carry these costs, and several auxiliary stations have stopped transmitting data to the IDC. Some stations are provided by members of the Federation of Digital Seismograph Networks (FDSN), such as the Global Seismographic Network (GSN). This two-tiered concept of primary and auxiliary stations was a good one at the time of the negotiations, when there were few modern, highquality stations operating. Today the situation is quite different: highquality data are readily available from many of the thousands of stations around the world outside of the IMS, and national and international organizations are providing rapid and comprehensive bulletins based on these data (see Figure 2.2). Data from these non-IMS stations cannot be used by the PTS nor the future TS, because the stations are not officially part of the verification regime defined in the CTBT. However, any state is free to use any data it finds useful in the monitoring of the CTBT as part of its national technical means (NTMs), and it can use such data to support a request for an on-site inspection. Countries thus have the potential to

Fig. 2.2  World map of seismic stations, shown as gray triangles, with IMS stations shown as red triangles (Storchak 2009).

34

Monitoring Underground Nuclear Explosions

achieve a monitoring capability far beyond what can be provided by the IMS data. As a state is likely to be interested in monitoring only a few states and areas of concern, it can focus its efforts and resources on those limited areas and achieve monitoring capabilities significantly above what can be obtained at a global level. The concept of “precision monitoring” for such focused efforts is introduced later in this chapter (section 2.6) and further discussed in Chapter 7. 2.2.2 Detection Signals and Noise Today, the limiting factor in detecting seismic signals is not the resolving power of the instrumentation but the ever-present noise in the Earth. The ability to detect a signal depends on the signal-to-noise ratio, defined as the strength of the signal relative to the strength of the background noise. The signal-to-noise ratio varies with frequency, and signals and noise can have quite different frequency content. Frequency filtering is a good way to enhance the signal-to-noise ratio, as illustrated in Figure 2.3. The

Filtered 6.0 - 12.0 Hz

Filtered 6.0 - 12.0 Hz

Filtered 2.0 - 4.0 Hz

Filtered 2.0 - 4.0 Hz

Filtered 0.5 - 1.0 Hz

Filtered 0.5 - 1.0 Hz

Unfiltered Waveforms

Unfiltered Waveforms

0.0

60.0

120.0

180.0

0.0

60.0

120.0

180.0

Fig. 2.3  The effect of filtering of seismic records. The left panel is the recording of a Nevada Test Site nuclear test on 14 September 1991 at the NORSAR array site NO602 in Norway. Different frequency bands are indicated. The right panel is a recording on the same seismometer of a nuclear test at Semipalatinsk on 6 February 1988. The two records show clear distinctions in frequency content which are due to the different source locations and travel paths (Figure preparation courtesy of Steven Gibbons, NORSAR).

35

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

frequency content of signals observed at a station depends on where the sources are located. Hence, to optimize detection, stations should be tuned with different filters for different source regions. The frequency content is one way of differentiating between signals and noise; propagation velocity is another. Array stations provide excellent tools to study background noise in some detail. Over a few kilometers, corresponding to the aperture of many of the array stations, the noise can be quite coherent and propagates with speeds and directions that change over time. A comprehensive study of short-period background noise has been conducted using 18 array stations (Koper et al. 2009). Figure 2.4 illustrates the noise characteristics at two of the world’s most sensitive array stations. Statistics on the observed noise propagation velocities show that more than 70 percent of the noise phases have observed apparent velocities below 5 km/s. Most seismic signals of interest for verification have substantially higher velocities of up to 25 km/s across the Earth’s surface. Array stations use this velocity difference to facilitate signal detection and characterization. One should keep in mind, however, that seismic signals with lower velocities, like S-phases recorded in the regional distance range, are also used to define events. The strength and shape of a recorded signal depend on the source and the transmission properties of the signal from source to receiver. Magnitude is used to describe the size of a seismic event and is estimated as the logarithm of the measured signal amplitude during the first few seconds of the signal, to which a correction for the distance to the source is added (see Figure 2.9 for an example of a standard amplitude-distance curve). Magnitude scales are thus logarithmic, and one magnitude unit corresponds to a factor of 10 in observed signal strength. Magnitudes are estimated for each station of the network observing an event but are typically provided as a magnitude for the network as a whole. For an underground explosion the signal strength is approximately proportional to the explosive yield. This means that there is by and large a linear relation between magnitude and the logarithm of the yield, and a magnitude 4 event is generally considered to correspond to a yield of the order of 1 kiloton. To make more accurate conversions between magnitudes and yields—for example, when converting detection thresholds from magnitudes to yields—requires great care. There are three main sources of uncertainty involved with this conversion; one is related to the way the Earth’s structure influences the transmission of seismic waves. Also, for a symmetrical and reproducible explosion source, there can be significant differences among the signals observed at one given station from explosions at different places (Figure 2.3). This is due to strong inhomogeneities in the Earth. Individual station magnitudes

36

Monitoring Underground Nuclear Explosions

will therefore vary, and this is also true for different stations at the same distance from the source. In addition, there can be significant differences among observations at the same station from different sources at the same distance, even if they are located relatively close to one another. These differences in the strength and appearance of the signals are greater for signals recorded within 2,000 kilometers of the event than for observations at greater distances. Signals recorded at shorter distances from the source are propagating through the upper, more heterogeneous, part of the Earth,

Fig. 2.4  Short period seismic noise measured at the IMS arrays in Makanchi, Kazakhstan (top) and Yellowknife, Canada (bottom). For each figure, the left part of the frame shows the location and the layout of the array. The right part is a plot of the noise intensity in different directions and for different slowness. The plot is best interpreted as a picture looking down from the array. The center of the plot provides the noise intensity looking straight down. Slowness is the reciprocal seismic velocity. The two figures illustrate that the noise field is quite different for the two stations (Koper et al. 2009).

37

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

whereas signals observed at large distances travel mostly through the deeper, more uniform, part of the Earth’s interior. Events might thus be given different magnitudes depending on where the observing stations are located because of inhomogeneities in the Earth’s structure. The differences in transmission properties introduce an uncertainty when comparing magnitudes of events located at different points on the globe. To accurately estimate yields from seismological observations requires calibration explosions and observations from a well defined network of stations. Scientific developments over the last decades have greatly increased the understanding of the Earth’s crust and upper mantle structure, thanks to new methods of analysis and an ever-increasing number of high-quality stations. This has made it possible to develop region-specific models for travel times and signal characteristics at regional and local distances and to use such data for event detection, location, and characterization (see Figure 2.5). The design of the global seismological IMS system benefited from this improved understanding of regional seismic wave propagation. In fact, many of the arrays in the IMS primary seismic network are regional or highfrequency arrays designed for optimum detection and characterization of signals originating from events in the regional distance range of up to 2,000 kilometers. Still, the total number of IMS seismological stations limits the network’s regional coverage, and it is clear that close-in observations from stations that are not part of the IMS network are playing an increasingly important role in improving seismological monitoring capability. The second source of uncertainty is that different organizations use somewhat different procedures for estimating magnitude. This results in many different magnitude scales with systematic biases among them and is a source of some confusion (Bergman and Engdahl 2009). The magnitudes reported by the IDC are, for example, 0.3–0.5 units lower than those reported by the scientific community in general. The difference may be most pronounced when comparing local magnitude scales applied to observations at local and regional distances up to some 2,000 kilometers from an event. Magnitudes based on many observations from many stations around the world are more consistent. The third uncertainty in converting magnitudes into explosion yields or vice versa is the variation in seismic signal strength with the properties of the bedrock in which the explosion took place. These differences can be significant: an explosion detonated in the alluvium deposits at NTS generates seismic signals that are an order of magnitude weaker than those from an explosion of similar yield in hard rock at the Semipalatinsk test site (Rodean 1971). The problem of converting seismological observations into explosion yield was illustrated by the controversy over the Threshold Test Ban Treaty

38

Monitoring Underground Nuclear Explosions

(TTBT), which limits explosions to yields below 150 kilotons. Experts in the United States reported that the Soviets had carried out test explosions with yields beyond the 150 kiloton limit. In 1988, the Joint Verification Experiment (JVE) was conducted, in which the United States observed a test at Semipalatinsk, Kazakhstan, where there is old, hard granite, and the Soviets monitored a test at the Nevada Test Site in the United States, where the rocks are softer and younger. The JVE demonstrated that estimates of yields based on seismological measurements at a distance from the two test sites differed and that explosions with the same yield produced seismic events with lower magnitudes at the U.S. test site than at the Soviet test site. The JVE and the development of improved means to estimate yields resolved some of the technical issues standing between the two countries over the TTBT (Sykes and Ekstrom 1989) and paved the road for development of the protocol for the TTBT and the subsequent ratification of this treaty by the two countries.

Fig. 2.5  World map of upper mantle seismic shear wave velocities. Seismic wave propagation is complex due to the heterogeneous structure of the earth. The accuracy of the epicenter determination depends strongly on the knowledge of the regional seismic velocities. The lower five panels and diagram on the right show the depth distribution of the velocities for the respective areas indicated on the map (Lekic and Romanowicz 2009).

39

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

Seismic signals from an underground explosion can be reduced if the test is conducted in a large cavity (e.g., Stevens et al. 1991). Experience with such decoupling techniques is limited, however. The United States conducted a nuclear explosion in 1966 with a yield of less than half a kiloton in a cavity generated by a substantially larger nuclear explosion. Seismic signals observed at distances of up to 110 kilometers were some 70 times weaker than if the explosion had been fully coupled (Springer et al. 1968). The Soviet Union conducted a partially decoupled nuclear explosion in 1976 with a yield of 8 to 10 kilotons, which gave a decoupling factor of about 15. The cavity used had a volume of about 200,000 cubic meters, sufficient to fully decouple a 3 kiloton explosion. A number of decoupling studies using small-scale chemical explosions have been carried out that have given decoupling factors in the 10 to 30 range. A recent experiment reported by Gitterman (2009) that used conventional charges of about 1 ton gave decoupling factors in the three to seven range. The extent of the signal reduction, reflected in the different decoupling factors observed, depends on several factors, including the size of the cavity in relation to the yield of the explosion and the exact shape of the cavity. It has been found that high-frequency signals decouple less effectively than lower frequency signals. Detection Processes The detection capability of an individual station also depends on the procedures used to detect a signal. In general, the methods used are quite simple: comparing a short-term average of a signal recording to a longterm average in a particular frequency band. For array stations the signals are enhanced by adding the signals from the sensors together. By inserting appropriate time delays between signals from the different array elements, the array effectively becomes a tuned antenna that selectively looks in real time at several hundred different spots on the Earth. The more information that is available about the signal being looked for and the noise that needs to be reduced, the more efficient will be the detection process. Using observations from a particular region as a reference in the detection process increases the efficiency of detecting signals from that region. This has been demonstrated by correlating signals from repeated mining explosions and by detecting aftershocks of earthquakes recorded on arrays (Gibbons et al. 2007). The application of new data analysis methods can improve detection capability considerably and reduce the likelihood of false alarms (Figure 2.6). The detection process is a balancing act between correctly detecting a signal and creating a false alarm. With a lower threshold weaker signals

40

Monitoring Underground Nuclear Explosions

Fig. 2.6  No signal from the ARCES array was reported by the PTS for the 2006 DPRK explosion. So-called full least square (LS) processing of ARCES data resulted, however, in a clear signal detection at the expected time of arrival for such a signal (Selby and Bowers 2009).

can be detected, but, on the other hand, more false alarms are generated. The relation between the probability of correct detections and false alarms defines the receiver operating characteristics (ROC) for a receiver (Heeger 1997). Figure 2.7 shows an illustration of an ROC curve. The issue of probability of detection is important in the subsequent discussion of confident detection versus deterrence. Signal detections are currently carried out at individual stations, and the detections from multiple stations in a network are then combined to form Fig. 2.7 These notional ROC curves illustrate the relation between the probability of detection and the false alarm rate. Such a curve can be estimated for any verification system and the ambition is to obtain ROC curves that come as close to the upper left corner of the diagram as possible.

41

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

events. The number of false alarms that can be accepted from the individual stations depends on how this network analysis process is set up. Discussion of a new paradigm for network processing is included in Section 2.2.6. Individual Station Capabilities There are many factors that influence the detection capability of a station and many ways of estimating and presenting that capability. Generally, stations in the interior of continents have lower background noise and are more sensitive than those close to oceans. Array stations normally have better detection capabilities due to the signal-to-noise ratio improvement achievable from array data processing. The relative detection performance of 111 IMS stations is presented in Kværna and Ringdal (2009) with estimates based on their globally averaged detection capabilities. There is a difference of 1.5 magnitude units between the most and the least sensitive stations. By measuring how the stations contribute to the PTS bulletin, a somewhat different station ranking is obtained (Coyne et al. 2009). This ranking also depends on the location of stations relative to the main earthquake areas. Six stations are among the 10 most highly ranked in both studies.

Fig. 2.8  World map of noise thresholds for the seismic array ARCES (Norway) for events in grid cells in areas of 200 by 200 km expressed in terms of equivalent body wave magnitude. Signals from the various source regions (the grid cells) with strength 0.5 magnitude units above these noise thresholds would be detected by this station (Kværna and Ringdal 2009).

42

Monitoring Underground Nuclear Explosions

The complex signal transmission properties, with possible contributions from the radiation pattern of the sources, give stations very different detection capabilities for different parts of the world. These differences can be significant, as illustrated in Figure 2.8 for the ARCES station in Norway. This is partly a distance effect, but also for events at the same distance from the station the detection capability differs by one order of magnitude, or a factor of 10 in signal strength. It has been noted above that there can be great differences among the detection capabilities of stations when events are at a considerable distance and that weak events can be detected at large distances. However, being close to an event yields substantially better detection capabilities, and the improved understanding of wave propagation at distances of less than 2,000 kilometers makes it possible to use such observations for monitoring. At such distances, differences between various regions are even greater than discussed above. In addition, the dependence of the strength of the signal on distance shows that signals at a few hundred kilometers distance are two to three orders of magnitude—or 100 to 1,000 times—stronger than for distances beyond 2,000 kilometers (Figure 2.9). The significant impact of distance between events and stations is clearly shown by the difference between the detection capability of a very capable station, MKAR in Kazakhstan, which was among the 10 best

Fig. 2.9  The black dots correspond to the noise thresholds shown in Fig. 2.8 and are here plotted versus epicentral distance from ARCES. The red/blue curve is a best-fitted standard amplitude distance curve (Kværna and Ringdal 2009).

43

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

Fig. 2.10a  Events (yellow dots) detected by the Tehran network between 2002 and 2008. The larger red dots are events detected by the IMS station MKAR in Kazakhstan. The black triangles are the locations of the seismic stations of this network (Engdahl and Bergman 2009).

Fig. 2.10b Plot of the number of events detected by the Tehran network (white bars) and by the IMS station MKAR in Kazakhstan (grey bars) versus magnitude as determined by the Tehran network. Above magnitude 4 the majority of the events is detected by both systems; for magnitudes lower than 4, an increasing number of events is missed by the MKAR station (Engdahl and Bergman 2009).

44

Monitoring Underground Nuclear Explosions

Fig. 2.11 The plot shows the number of seismic events in ISC (blue) and PTS (red) bulletins versus magnitude for the period May 2004 – April 2007. The figure shows that the PTS bulletin is complete above ISC magnitude 4.0. After the Great Sumatra event on 26 December 2004, PTS bulletins were not produced for several days, thus some aftershocks are missing in the PTS bulletin. The spike in the number of events just below magnitude 3 is for events of unknown magnitude (Storchak 2009).

stations in both studies discussed above, and the local detection capabilities of the Iranian Telemetered Seismic Network (also known as the Tehran network) for events in the Tehran region of Iran, as illustrated in Figures 2.10 a and b. MKAR, which is located at a distance of more than 2,000 kilometers from these events, starts to lose events in the magnitude range of 3.5 or 4; below 3.5 it has lost most of the events detected by the local network. The Tehran network detects a large number of events in the Tehran region down to a local magnitude threshold of about 1.5 (Engdahl and Bergman 2009). Network Capabilities The International Seismological Center (ISC) compiles the most comprehensive list of seismic events worldwide, based on reporting from all available global seismological stations, which now number around 16,000 (Storchak 2009). A comparison of the number of events reported by the ISC and the PTS for a three-year period shows that the IMS is capable of detecting all events above magnitude 4 reported by the ISC (Figure 2.11). Even if the IMS capability drops rapidly below magnitude 4, it can still detect a few percent of all ISC events at magnitude 3. A 90 percent detection capability has been estimated for the complete IMS network, and this illustrates that events down to magnitude 3.5 can be detected in most of the Northern Hemisphere (Figure 2.12, top). A similar calculation at 10 percent detection capability, which might still be considered a high enough probability to deter any clandestine nuclear explosion, shows detection thresholds generally half a magnitude lower (Figure 2.12, bottom).

45

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

Fig. 2.12  The figure shows the detection capability at 90 % and 10 % detection probability levels for the IMS primary seismic network as implemented in late 2007. This capability is represented by the magnitude of the smallest seismic event that would be detected with the indicated probability by three stations or more (Figure preparation courtesy of Tormod Kværna and Frode Ringdal, NORSAR).

46

Monitoring Underground Nuclear Explosions

A 2002 U.S. National Academy Study (NAS 2002a) concluded that the IMS monitoring thresholds “would drop generally by about 0.25 magnitude units in Europe, Asia and North Africa, and by about 0.5 magnitude units in some regions (such as Iran)” if data from IMS auxiliary seismic stations were also continuously examined for detections. Threshold Monitoring Areas covered by in-region seismic networks can be monitored down to very low magnitudes, for example, down to magnitude 1.5 for the Tehran region of Iran, as mentioned above, or even lower for dense regional networks such as those in California and Japan. Explosions of a few tons would generate a seismic event of magnitude 1.5. If a state wants to monitor a specific region or country when only stations located outside that region or country are available to that state, low monitoring thresholds can also be achieved. The concept of “threshold monitoring,” developed by Ringdal and Kværna (1989, 1992), is relevant in this context. Figure 2.12 shows the detection capability of the IMS seismic network by presenting, in the traditional way, global maps of detection thresholds, which are estimates of the smallest magnitude of seismic events that can be detected. Detection levels shown in this figure are for detection at a minimum of three stations, which is needed to provide a meaningful location of the event. The IMS three-station detection capability for the Korean peninsula at the 90 percent probability level at the time of the North Korean test on 9 October 2006 was at approximately magnitude 3.8. One might say that the requirement of detection at three stations or more is too conservative, or stringent, in the sense that its application does not show the full potential of the network. The threshold monitoring technique represents a supplement to traditional event detection analysis in which the following question is addressed: Given the data available, what is the largest seismic event at a given time for a given site or in a given region that could possibly have occurred without being detected? By subjecting the North Korean test site to threshold monitoring, Kværna et al. (2007) conclude that, using data from the IMS network available at the time of the test, this site can be monitored—in the threshold monitoring sense—at the 90 percent probability level down to a magnitude of between 2.3 and 2.5. This corresponds to explosive yields of only around 10 tons for explosions in hard rock. Not surprisingly, use of data from the IMS seismic array station in South Korea was essential in obtaining such low values. The threshold monitoring approach provides an upper limit of the magnitude of events that have not been detected. It shows at each moment the largest events that could have taken place in an area without being detected.

47

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

The main purpose of the threshold monitoring method is thus to draw attention to those instances when a given threshold (e.g., 2.3 to 2.5 for the North Korean test site at the 90 percent probability level) is, in fact, exceeded. For such instances, analysts will then apply other traditional analysis tools to clarify the source of the observed disturbance and, if possible, confirm that there was in fact an event in the monitored area at the time indicated. The threshold monitoring technique can be said to provide a more complete picture of actual capabilities than the traditional approach based on detection at three stations, and in that sense this technique represents an optimal usage of the available data. This technique can be useful to a state that might find reason to subject specific areas of interest to special scrutiny, continuously or during limited periods of time. Kværna et al. (2002) were able to continuously monitor Novaya Zemlya during November and December 1997 down to magnitude 2 most of the time and were confident that no event exceeding magnitude 2.5 took place. Data from non-IMS stations can be included in the analysis, and this will contribute to a further lowering of the detection threshold. These low thresholds will represent a considerable deterrence to clandestine nuclear testing under the CTBT. 2.2.3  Event Location Location is a critical parameter for the further assessment of an event. It can serve to determine in which country an event occurs, and it is a critical input to deciding on an inspection area, should an on-site inspection (OSI) be requested (see Chapter 5). Even if there might be additional input when defining an inspection area, a good event location based on seismological observations is a very important parameter. The closer the initial estimate of the location is to the actual site of an event, the higher the probability is that an OSI will be successful. According to the CTBT, an on-site inspection area shall not exceed 1,000 square kilometers, corresponding to a circular area with a radius of 18 kilometers. If location uncertainties significantly exceed this number, the likelihood increases of requesting and conducting an inspection in an area that will not contain the actual event. There are two kinds of uncertainties in locating events: a statistical uncertainty, which depends on observational uncertainties, the number of observing stations, and their distribution around the event; and a systematic bias caused by differences between the velocity model used in the computations and the actual seismic wave velocities. The statistical uncertainty is estimated as part of the location process. It is normally presented as the area, in the form of an uncertainty ellipse, within which

48

Monitoring Underground Nuclear Explosions

the event should be found with 90 percent probability. The systematic bias goes unnoticed in the location process and has to be addressed and corrected using calibration events that have locations that are known with high precision. A number of such calibration, or “Ground Truth” (GT), events have been identified that have a location uncertainty of less than five kilometers (GT5). If uncalibrated, the bias can be large and exceed 50 to 100 kilometers. In the absence of accurate three-dimensional Earth models, each station must be calibrated for events in each region for which it contributes observations. This is a time-consuming task, especially to establish the calibration events needed.

Case Studies A number of regional studies of location accuracy were reported at the International Scientific Studies Conference in 2009 (ISS09) . One valuable source for such studies includes mining explosions with high-precision GT locations . As some of these explosions are quite weak, it is also possible to study location uncertainties of small events . A study of eight mining regions in Europe, Asia, Africa, and North America comprises a total of 56 events in the magnitude range of 2 .5 to 4 .5 (Bergkvist and Johansson 2009) . Table 2 .1 summarizes the results, and Figure 2 .13a shows the PTS location error distribution for all eight regions .

All regions, except Wyoming in the United States, show a similar pattern, with errors in the range of 1 to 30 kilometers . The systematic component of the errors is in the range of 3 to 16 kilometers . The errors for Wyoming are significantly larger . It is interesting to note that the estimated 90 percent confidence areas are quite small; in almost all cases they are less than 1,000 square kilometers . However, for almost all regions the 90 percent confidence areas do not contain the ground truth locations of 90 percent of the events . For some areas, like Kiruna in Sweden, only 5 percent of the confidence areas contain the GT solutions . Once the systematic error is

49

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

Fig. 2.13a  The location error distribution for all eight mining areas (see Table 2.1) analyzed by Bergkvist and Johansson (2009).

Fig. 2.13b  Confidence error ellipses, before and after removal of systematic errors for mining explosions near Kiruna, Sweden (Bergkvist and Johansson 2009).

removed, 95 percent of the confidence areas contain the GT, as illustrated in Figure 2.13b. A study by Cichowicz (2009), based on 36 seismic events in gold mines in South Africa located with high precision by local seismological networks in and around each mine, shows a similar result. The location errors decreased significantly when data from several calibration events were used in the computation of the event locations. The Italian Seismological Network, comprising more than 200 stations, provides event locations with errors of typically less than 2 kilometers for

50

Monitoring Underground Nuclear Explosions

events within Italy. A comparison of Italian network locations with those of the PTS for 63 events in Italy shows that 50 percent of the events had a location difference of less than 20 kilometers, and 90 percent of the events had a difference less than 40 kilometers (Giuntini et al. 2009a). Very similar numbers were obtained in a comparison of PTS locations with those determined by the local Romanian network (Ghica et al. 2009). National Data Centers are established in most countries as a point of contact for data exchange with the PTS. Some of these national centers are fairly large facilities, and are often part of scientific organizations. NDCs undertake preparedness exercises to test the processing and analysis capabilities of NDCs and the IDC, the performance of the IMS, and the efficiency of data exchange. These are annual exercises, pioneered and organized by the German NDC, and involve NDCs around the world. The exercise in 2008 focused on an event off the coast of Chile. The NDCs’ analyses show that the final location determined by the PTS differs by more than 50 kilometers from those obtained by NDCs. The initial automatic PTS location was some 200 kilometers away. The exercise in 2009 used a mining event in eastern Kazakhstan, from which seismic and, for the first time in such an exercise, infrasound observations were also used. The 2010 exercise was driven by synthetic radionuclide observations, and the issue at hand was to find a real seismic event in a region of high seismicity that would be consistent with the synthetic radionuclide observations. Global Capabilities In an ISS09 Conference study by Bergman and Engdahl (2009), 302 events having well-established locations with an uncertainty of less than 5 kilometers (GT5 events) from 37 areas around the globe were used as reference events, and their locations were compared with those reported by the PTS. The distributions of the location differences are shown in Figure 2.14a. Fifty percent of the PTS events differ from the reference events by less than 16.5 kilometers, and 90 percent differ by less than 44.5 kilometers. The mean difference is 22 kilometers. Figure 2.14b shows the location difference vectors, giving the relative size and direction of the errors. This shows that there is a mixture of systematic and random errors in most of the areas. In assessing these results, Bergman and Engdahl conclude that the distribution of observing stations in azimuth around an event is the single most important parameter influencing location uncertainty. The better the coverage, the less the uncertainty. The ISC maintains a similar collection of GT5 events. At present there are 62 such events. A comparison of the PTS locations with those

51

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

Fig. 2.14a Histogram of location differences between GT5 events and PTS reports for 302 globally distributed earthquakes. GT5 events have uncertainties in their true locations that are less than 5 km (Bergman and Engdahl 2009).

comparatively few events shows a very similar result, with 50 percent within 16 kilometers and 90 percent within 40 kilometers (Storchak 2009). On a global scale, the PTS locations can be compared with the locations provided by the ISC. ISC locations cannot be considered “ground truth” as they are biased by the uneven configuration of the network of stations

Fig. 2.14b  Location difference vectors of Reviewed Event Bulletin (REB) locations relative to locations in the GT5 data set. GT5 events have uncertainties in their true locations that are less than 5 km. The small arrow in the lower left corner provides the scale for the size of the error (Bergman and Engdahl 2009).

52

Monitoring Underground Nuclear Explosions

Fig. 2.15 Histogram of differences between PTS focal depth estimates and focal depths for GT5 events. Shaded columns are events for which the REB reported a non-zero depth (Bergman and Engdahl 2009).

used by the ISC and uncertainties in the velocity models used. Yet the sheer volume of stations used by the ISC, including those very close to events, compared to the PTS, should in general provide a reasonably high degree of accuracy. In a study comprising 62,000 events located by the PTS and the ISC, 50 percent of the PTS locations were within 16 kilometers of the corresponding ISC locations, and 90 percent were within 60 kilometers (Storchak 2009). Depth Estimation Depth estimates of earthquakes are more uncertain than estimates of the coordinates at the surface of the Earth (latitude, longitude). There are two ways of making depth estimates: using stations at very short distances and making a three-dimensional location, or using signals reflected at the Earth’s surface above the event, a pP phase. To confidently identify an observed phase as pP is difficult, and for most events at shallow depths it proves impossible. The experience so far, supported by the examples below, is that depth estimates of events down to depths of 30 to 50 kilometers are uncertain, unless the event is observed by stations at close distances. Agencies such as the U.S. Geological Survey use large amounts of data from stations close to the events to estimate and report depths of shallow earthquakes. Other institutions, primarily using observations at greater distances, may report a nominal depth value for shallow events; the PTS gives 0 kilometers. Bergman and Engdahl used the 302 GT5 events referred to above to study the differences between the depth of these reference events and the estimates in the PTS bulletin. Figure 2.15 shows that there is a large scatter in the PTS estimates relative to the calibration events. Most of the reference

53

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

events are at shallow depths, and for almost all such events the PTS assigns a depth of 0 kilometers. A comparison of earthquake depths in Italy provided by the National Institute of Geophysics and Volcanology (INGV), using data from the Italian network, and the PTS using data from IMS, also shows a great scatter for depths of less than 50 kilometers and a fairly good correlation for depths greater than about 100 kilometers (Giuntini et al. 2009a). 2.2.4  Event Characterization A key question in seismological monitoring has been how to distinguish observations of nuclear explosions from those of the large number of earthquakes occurring around the world. In the time period leading up to the CTBT negotiations, a large number of discrimination studies were conducted, based primarily on nuclear explosions and earthquakes in the former Soviet Union and the United States. Several discrimination parameters were developed, based on frequency content, the complexity of the seismic signals, and comparison of the strength of different waves. In general, good discrimination could be achieved, with few overlapping events, when nuclear explosions and earthquakes from the same region were compared. Results were less successful when applying standard criteria to events globally. Most early studies aimed at separating explosion and earthquake populations. Currently the objective is to characterize observed events to establish if they are part of the population of earthquakes in the source region or if they need further consideration. Few studies have been conducted on how to describe or characterize a particular earthquake population and determine if a new event belongs to that population. This assessment must be done on a regional basis and will be discussed further below. Despite the fact that discrimination should be carried out on a regional scale, the protocol to the CTBT states that standard event screening criteria should be applied to screen out earthquakes on a global scale and this has not proven successful. Case Studies Few new scientific studies on event discrimination have been carried out since the beginning of the 21st century. However, the nuclear explosions in the Democratic People’s Republic of Korea (DPRK) in 2006 and 2009 provided data to address discrimination, and some studies of these events are introduced below. A discrimination analysis by Materni et al. (2009) compares the magnitudes estimated from body waves (mb) with those estimated

54

Monitoring Underground Nuclear Explosions

Fig. 2.16 Density distribution of the discrimination scores between two groups; earthquakes (blue curve) and explosions (red curve) (Materni et al. 2009).

from surface waves (Ms). The study uses a statistical method based on a learning set of observations from more than 100 explosions and some 250 earthquakes. The discriminant distribution of these events is shown in Figure 2.16. This figure illustrates that discrimination, as well as detection, is statistical in nature and that the probability distributions may in general be well separated but the “tails” may overlap, creating “false alarms.” Applying the parameters established from the learning set to a set of 806 earthquakes globally, 798 events were classified as earthquakes and eight were misclassified as explosions. This shows that the mb-Ms method is fairly robust, but it still has the limitation that surface waves are rarely observed for weak events. Cepstrum analysis is a non-linear analysis method to detect echoes and other reverberations in a signal. Cepstrum is mathematically defined as the Fourier transform of the logarithm of the power spectrum of a signal (Wei and Li 2003). It has previously been used as a discriminant, and in a study by Wei and Xu (2009), a new discriminant has been developed based on the variability of the cepstrum, where explosions show an approximate linear variation and earthquakes display significant variations. Using a parameter, denoted as “C,” to describe the variability of the cepstrum from signals observed at stations in northeastern China, a good separation is achieved between earthquakes on one side, and chemical explosions in the Korean peninsula and the DPRK nuclear test in 2006 on the other (see Figure 2.17). Another example of using local or regional observations in the distance range of 200 to 2,000 kilometers to discriminate between explosions and earthquakes is a study by Kim et al. (2009b). Local observations from nuclear explosions at the Lop Nor test site in China and nearby earthquakes

55

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

Fig. 2.17 Plot of the local magnitude ML vs. the value of the C-parameter for a series of earthquakes (red dots) and chemical explosions (blue dots). The black dot is the DPRK nuclear test in 2006. The C-parameter measures the variability of the cepstrum of the recorded waveforms (Figure after Wei and Xu 2009).

show that the two kinds of events can be separated fairly well using local P wave to S wave amplitude ratios at different frequencies (see Figure 2.18). Event Screening The protocol to the CTBT notes that: The procedures and standard event screening criteria to be used by the International Data Center in carrying out its agreed functions, in particular for the production of standard reporting products and for the performance of a standard range of services for States Parties, shall be elaborated in the Operational Manual for the International Data Center and shall be progressively developed. The procedures and criteria developed initially by the Preparatory Commission shall be approved by the Conference at its initial session. In a report to Working Group B in August 2009 (CTBT/PTS/INF.1026 2009) on the results of the operational application of event screening criteria since 2001, the PTS noted that only 45.6 percent of these events were screened out as natural events in the Reviewed Event Bulletin (REB) during a one-year period in 2007–2008. Events that could not be considered for screening because they were too small or had insufficient data to allow screening amounted to 49.1 percent. Those not screened out despite available data were 5.3 percent. This class of events is the one that needs special attention by those who worry about compliance with the treaty. However, almost half of the events do not lend themselves to the

56

Monitoring Underground Nuclear Explosions

Fig. 2.18  Plot of the spectral amplitude-ratio between P- and S-waves. The known nuclear explosions plotted as triangles in the pink range have consistently higher values than the earthquakes, plotted as circles in the yellow range. The event on 13 March 2003 caused some confusion but proved to be an earthquake close to the former Chinese nuclear test site at Lop Nor. It falls in the earthquake population, but overlaps with the explosion population of the higher frequencies (Kim et al. 2009b).

screening process. This raises questions about the use of screening using global criteria and shows the need to do screening on a regional basis. 2.2.5  New Methods to Analyze and Exploit Data Almost every part of modern society generates, distributes, and collects enormous amounts of data. Technical developments have facilitated a dramatic increase in data collection since the end of the 20th century, and much more data are being collected than it is possible to analyze and interpret. This is also true for monitoring and surveillance, as well as for geophysical studies. In the area of CTBT monitoring, the IMS generates large volumes of data and so can an OSI. Countries may also want to use the large amount of data and information available outside the IMS from additional observing stations, satellites, and other sources. New analysis tools are needed to take advantage of the available data, and new tools are being rapidly developed. The remarkable increase in open data archives, coupled with sophisticated data management and analysis techniques, has promoted the development of data mining, a catch-word for automated processes that find patterns and relationships in large data volumes. Data fusion procedures are used to combine and draw conclusions from different kinds of data and create benefits from the synergy involved. New data management systems have been developed to intelligently store, retrieve, and process very large amounts of data.

57

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

Russell (2009) describes data mining as computational methods to improve performance based on experience. He also identifies two key questions: On what piece of the system do you want to improve the performance, and how is that piece represented? What data and knowledge are available relevant to that piece? There are two basic approaches to data mining: model-free and model-based. The goal in a model-free approach is to find a hypothesis that maximizes a combination of simplicity and best fit to the data. In the model-based approach, prior knowledge strongly constrains the hypotheses. Applying this new and rapidly developing science of data analysis and exploitation to the verification of the CTBT could substantially improve capability and cost efficiency. The ISS09 Conference was the first time this synergy was explored, and a number of contributions were presented. Kuzma and Vaidya (2009) presented a review of these contributions, and the following comments are based on their paper. A dominant theme of the ISS09 Conference contributions in the area of data analysis was the difference between the results of the automatic IDC processing and those of experienced analysts at the IDC. The current automatic IDC processing generates spurious events that the analysts reject based on their experience. The amount of rejected events is on the order of 50 percent of the number of events retained after analyst review (see Figure 6.5). A number of studies, mostly in their initial stages, explored how the automatic system can be trained to avoid mistakes and thus reduce the amount of spurious events. Carluccio et al. (2009) presented initial results of a supervised training algorithm using a neural network. The purpose of the procedure is to predict the likelihood that an automatically created event is not an artifact. In the analysis, historical, automatically generated events were compared with the results of the analysts’ review, and the events were classified as real or bogus. This information was used to search for hidden patterns among the event parameters, and on the basis of those patterns a neural network was trained to classify new events as real or bogus. As previously noted, all detection and discrimination procedures have similar trade-offs between correct detection or identification and false alarms. Initial results using this supervised training algorithm show correct classification rates in the 75 to 80 percent range. A number of other stateof-the-art classification algorithms, including so-called support vector machines, naïve Bayes classifiers, and decision trees, were also tested. The studies presented achieved a probability on the order of 75 to 90 percent to correctly identify an event as real or bogus; the corresponding probability of correctly identifying individual signal phases was a bit higher and exceeded 90 percent (Kleiner et al. 2009, Schneider et al. 2009, Procopio et al. 2009). So-called statistical, or Bayesian, inference methods can be used

58

Monitoring Underground Nuclear Explosions

Fig. 2.19 Map of Italy showing the detection capability of the Italian Seismic Network on 1 July 2007. Plotted is the magnitude (MP) for which the catalogue is complete with a probability of 99.9 percent (Schorlemmer 2009).

to mimic the behavior of an analyst when he or she decides that an automatically determined event is a “real” event. The Bayesian inference algorithm computes the a posteriori probability of an event, taking into account all the evidence of the event. This makes it possible to test the probability distribution for alternative hypothetical events as part of the location process (Arora et al. 2009a, Arora et al. 2009b, and Myers and Johannesson 2009). Arora et al. (2011) report on a project to use a Bayesian inference system to improve current automatic software for analysis of seismological data. The current version is based on reported detections at some IMS stations and also on negative evidence that other IMS stations have not reported detections. Initial tests show that the inference system, including the use of negative evidence, improves the performance compared to classical analysis systems. It is expected that the detection thresholds will be lowered significantly when the inference system also performs signal detection on recorded waveforms. Schorlemmer (2009) presents a method for computing the spatial variation and temporal evolution of detection capability of seismic networks based on empirical data on detected and missed events, and for estimating probability maps for event detection. Figure 2.19 is an example of the probability maps presented at the ISS09 Conference, and shows that the detection capability of the Italian Seismological Network on 1 July 2007 was in the range of magnitude 1 to 2.

59

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

The IMS is composed of several technologies from which data and information have to be combined to fully utilize the network’s potential. A combination of seismological and hydroacoustic data helps to identify oceanic events (Jepsen and Fisk 2009). Fusion of seismological and gravity data can improve the estimates of Earth structures (Maceira and Ammon 2009). Data fusion is also essential for an OSI, for which a large number of different observations and different data need to be combined. The considerably increased amount of data requires not only new approaches to data analysis but also to data management. Present systems are limited in the way data can be searched and retrieved . More flexible frameworks are used commercially, by Google in particular, and Liu (2009) proposes the use of a distributed database that can take advantage of parallel processing and multiple nodes. New data mining methods and procedures can be used not only to improve on the existing automatic data analysis work but also to form the basis for developing new approaches to the analysis of data from seismological and other monitoring networks. A new paradigm for seismic analysis is discussed below, as well as the concept of precision monitoring, which builds on recent advances in the analysis and exploitation of large amounts of data. 2.2.6  New Paradigm for Seismic Analysis The present automatic seismic analysis is based on methodologies and procedures developed some 25 years ago. Signal detection is carried out independently at each individual station, where an automatic algorithm declares a signal detection when a short-term signal is compared to a long-term average of the waveform signal and exceeds a preset threshold. Signals are normally filtered to increase the signal-to-noise ratio before detection, and for array stations the detection processes are carried out on a number of “beams,” which enhances signals from different areas of the globe. Each detection is described by a number of parameters, such as the arrival time, amplitude, frequency, and complexity of the signal. For array stations the direction and apparent velocity of the incoming signal also are estimated, which provides a rough estimate of the location of an event. A key step in the seismological analysis is the association of observations from the individual observing stations to particular events. This forms the basis for the “creation” and localization of seismic events. An initial fundamental uncertainty is to identify which observations belong to which events. All detections from the stations in the actual network are available to an automatic association and location process. The association process is a search process, in which hypothetical events distributed around

60

Monitoring Underground Nuclear Explosions

the globe are tested against the observations using a non-linear travel time model. When a sufficient number of observations match an event hypothesis, a tentative event is declared. No comparison is made in this automatic process of observed waveforms among stations or with earlier observations in a historical database. The automatic process is followed by an analyst review of the events created in the automatic process. Here the waveforms are brought back again, and the analyst can make a judgment whether the event should be considered to be a real or “spurious” event. With a large number of detections at a fairly large number of stations, the likelihood is quite high that the automatic process associates observations incorrectly and creates spurious events. Roughly one-third of the events in the bulletin created by the automatic processing at the PTS are rejected in the manual analysis. By looking at the waveforms observed from an event, an analyst can also improve on the location by adjusting the detection time and deleting or adding stations to process. The analyst normally does not go back and compare with earlier observations but uses her or his professional experience in the interpretation. This traditional automatic process has some specific characteristics. Signal detection is carried out by a fairly simple detection process at each individual station without any consideration of what is observed at other stations in the network. Experience from earlier observations is taken into account in the detection process in only a general way, such as in station-specific filter settings. Throughout the process each observation and event is treated as if it was the first one ever observed, and no comparisons are made with earlier events or observations. The association process is based on parameter data only, the original waveform signals are not part of the process. The striking development of new data mining methods for the analysis and exploitation of large volumes of data and a corresponding development in computer hardware and software have paved the way for a new paradigm in seismological data analysis. This paradigm is based on three new pillars: to make event detection at the network level and not at individual stations, to use the waveform data throughout the detection and location process, and to use historical data as a reference in the process. A new paradigm for data analysis will exploit well established seismological characteristics. Seismological observations at the same stations from repeated events at the same source location are generally identical and can be easily recognized. Successful tests have been made to automatically recognize repeated mining explosions observed at the same station using correlation techniques (Gibbons et al. 2007). Seismological signals from closely spaced events observed at the same station also normally show large similarity, and trained seismologists can quite often

61

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

recognize a signal as coming from a seismic event in a specific region from its characteristic appearance. On the other hand, observations of a particular event at different stations often have quite different features, as the seismic wave propagation characteristics vary strongly from one region of the world to another. This means that events in different parts of the world show distinct patterns in their waveforms, as observed over a network. The network has, for each region, a specific fingerprint that remains stable as the wave propagation over a defined path does not change over time. A new paradigm should be based on a simultaneous analysis of all the waveform observations from the stations in the actual network. Detection will not be carried out at the station level but at the network level, exploiting the characteristic fingerprints of events in different areas. New observations will be compared or related to earlier observations in a historical database. The better defined the characteristic features of the signals to be detected, the better the detection capability. The new process should also locate the new events in relation to earlier nearby events. The location uncertainty will in this way be reduced as understanding of events in a region increases. Events should also be characterized in relation to earlier observed nearby events. Is the new observed event typical for the region or is it abnormal in some way? This kind of comparison could be based on the waveform observations made or on features extracted from the observations describing the population of events in a region. Eventually, it might be possible to make the interpretation in relation to physical models of earthquake sources in the area. Any event that does not fit the earthquake population, however defined, would then be an event that merits further clarification. Such a new paradigm for data analysis and interpretation might be developed and applied to seismological data on a global scale, but this is a big undertaking as an initial step. It is more realistic to initially apply this paradigm to precision monitoring of selected smaller areas. It is also a tool that perfectly fits the concept of precision monitoring by providing states high detection capability, event location with low uncertainty, and a characterization of events to point out events that may need to be clarified further. To develop new analysis systems based on such a new paradigm is a challenging undertaking and should engage both the seismological and the data mining scientific communities. As such a system would significantly improve the ability of countries to monitor the CTBT, it would be natural that a number of states cooperate in its development, engaging scientific institutions around the world.

62

Monitoring Underground Nuclear Explosions

2.3  Radionuclide Xenon Monitoring Detection of radioactive material characteristic of a nuclear explosion is a key element in confidently identifying a detected event as a nuclear explosion. During the period of atmospheric testing, detection and analysis of radioactive particles were used to monitor testing activities. Since testing moved under ground, only a few examples of release of nuclear particles have been observed at large distances from test sites (Table 2.2), and recent developments in radionuclide monitoring have focused on radioactive noble gases, in particular xenon. Since the beginning of the 21st century, xenon monitoring science has grown considerably. New monitoring equipment has been developed and implemented on a global scale, and the scientific understanding of the possibilities and limitations of xenon monitoring has increased. Almost 30 of the 220 scientific posters and contributions for the ISS09 Conference related to xenon monitoring, and Ringbom and Miley (2009) provided a summary that has been used in creating this part of the chapter. 2.3.1 Release of Xenon from an Underground Explosion In the near field, radioactive releases of noble gases have been observed from a number of underground nuclear explosions . The United States reports that, out of its 723 explosions conducted since the PTBT of 1963, 105 leaked radioactivity due to containment failure and 287 had minor leakages related to operational activity such as drilling back into the explosion area . The drill-back releases were mostly xenon . Another five tests showed latetime seepage; these releases were not considered accidental or operational (Schoengold et al . 1996) . Most of these releases were observed only locally at test sites; 52 of the leakages, or 12 percent, were observed off-site, and

63

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

only a few of these were detected at large distances. It should be noted that the measurement equipment used in the United States at the time of the testing was considerably less sensitive than the IMS noble gas monitoring systems now being deployed. Radioactive noble gas releases have also been reported from underground nuclear explosions in the Soviet Union. A total of 340 underground nuclear tests were conducted at the Semipalatinsk test site from 1961 to 1990. Of these, 145, or 42.6 percent, generated a weak release of radioactive noble gases, and for 12 explosions, or 3.5 percent, a “nonstandard radiation situation” was reported, indicating a stronger release. During the same period, 39 nuclear tests were conducted underground at the Novaya Zemlya test site in Russia. Of these, 23, or 59 percent, were accompanied by a weak release of radioactive noble gases. Anon-standard radiation situation, or a stronger release, was observed for two explosions. In these two cases both early-time release within an hour of the explosion and late-time release from one to 28 hours after the explosion were observed (Dubasov 2010). Table 2.2 summarizes the detections in Sweden from underground explosions in the former Soviet Union and the United States. The measurement of xenon from the DPRK explosion in 2006 was obtained using Swedish equipment placed in South Korea. Given the large number of explosions conducted during the period covered by the table, the number of radionuclide observations is quite small; it should, however, be noted that, with the exception of the xenon observation in South Korea, the reported observations were made in Sweden and the released radioactive material had to reach Sweden to be detected. If a global detection system had been operational at that time, the number of observations would have been larger. Most of these observations are not related to xenon, as few xenon detection systems were in operation during the period when the major nuclear powers conducted nuclear explosions. Four xenon isotopes are produced in sufficient quantities and have suitable half-lives to be relevant for detection of an underground explosion: Xe-131m, Xe-133m, Xe-133, and Xe-135. The different isotopes have different rates of decay, with half-lives ranging from nine hours (Xe135) to 12 days (Xe-131m). There is also one isotope of argon (Ar 37) that might be useful for detecting a nuclear explosion. This isotope is not a fission product created by the explosion but rather is created by neutrons interacting with calcium in the surrounding bedrock. Xenon isotopes can be generated not only by nuclear explosions but also by nuclear reactors and in particular by medical isotope production, as discussed below. Comparing the isotopes will give information on the source, as illustrated in Figure 2.20.

64

Monitoring Underground Nuclear Explosions

Fig. 2.20   The figure illustrates in a schematic way the isotope ratios for xenon releases from nuclear reactors and nuclear explosions (De Geer 2009b). A more comprehensive discussion of these ratios, including observations from the INGE project, is provided by Kalinowski et al. (2009a).

There is genuine uncertainty regarding the expected amount of noble gas release, or the radionuclide source term, from an underground nuclear explosion. In a nuclear explosion an enormous amount of xenon is created, some 15 to 20 percent of the atoms that fission creates generate a radioactive xenon atom at some stage. The uncertainty is related to how much of this xenon is released into the atmosphere where it may be observed. An explosion conducted in close contact with the bedrock, in a borehole or a tunnel, glassifies the surrounding rock wall and essentially seals the cavity. Still, noble gases might escape through essentially three mechanisms with different timescales: • •

•

an immediate vent within hours through the engineered containment; seepage that might follow within the next few days, driven by the initial pressure in the cavity and facilitated by geological- or explosion-generated fractures; and diffusion through the fractured rock and soil, enhanced by variation in barometric pressure, so-called barometric pumping (Annewandter et al. 2009).

Explosions in cavities are likely to increase the risk of radioactive releases. In a decoupled explosion the very purpose is to reduce the pressure on the wall in order to reduce the resulting seismic signal, which means that no glazed surface will be formed. This will increase the likelihood that radionuclide gases might escape. It is expected that only a small fraction of the xenon generated will be released, and efforts can be made to reduce this amount and even prevent

65

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

leakage by creating a carefully contained explosion environment. Given the many uncertainties, however, it is hard for the tester to predict when, how large, or even where leaks might occur. The possible observation of xenon serves not only a monitoring purpose; the nature of its unpredictability is also an important deterrence feature. 2.3.2 Measurements Although atmospheric xenon measurements were carried out already in the 1980s and 1990s in Germany, the Soviet Union, Sweden, and the United States, there were no stations suitable for implementation as part of an IMS when the CTBT opened for signature in 1996. Different types of xenon monitoring equipment have been developed and tested as part of the International Noble Gas Experiment (INGE) since 2000, in which four different systems were produced as prototypes by France, Russia, Sweden, and the United States. In 1997 the U.S. system made the first publicly known detection of Xe-135 in the atmosphere, probably from reactor leakage (Reeder et al. 1997). The three other systems went forward into successive phases of the INGE and have been established at various IMS radionuclide station locations worldwide as part of ongoing testing and evaluation of the systems and to learn about xenon background in the atmosphere. Upon entry into force of the CTBT, it is required that a total of 40 xenon stations be operational; by December 2010, 27 of these stations were in test operation. Several xenon stations have passed the certification stages and are considered operational in the IMS as well as in the INGE. The xenon stations operating within the IMS are complex automatic systems in which air is collected from the atmosphere for periods of 12 to 24 hours. The small amount of xenon present in the collected air is extracted, and the radioactive components are measured. Depending on design, the stations provide one or two sample spectra per day that are transmitted to the PTS. The sensitivities of the three different systems for the four xenon metastable states and isotopes are in the range of 0.15 to 0.3 mBq/cubic meter, which means that the systems are able to detect a few hundred atoms of radioactive isotopes per cubic meter of air (Pakhomov and Dubasov 2009, Nikkinen et al. 2009). Ongoing scientific and technical development might provide stations with even higher sensitivity and time resolution and greater ability to discriminate observations of xenon generated by nuclear explosions from that generated by civilian activities. The noble gas stations are now approaching the same data availability as the less complex particulate stations, and future systems will have higher up-time and be less dependent on operators.

66

Monitoring Underground Nuclear Explosions

Fig. 2.21  Map of the global distribution of the average Xe-133 activity concentration for 2008 (Ringbom et al. 2009b).

2.3.3  Background Radiation Xenon is released by nuclear explosions and by a number of other human activities, including the production of medical isotopes and operation of nuclear power reactors. The establishment of measuring stations over the last few years has made it possible to map, for the first time, the xenon distribution around the globe (Wotawa et al. 2009, Ringbom et al. 2009a, Ringbom et al. 2009b). The dominating effect of a few medical isotope production facilities (Fleurus in Belgium, Chalk River in Canada, and Pelindaba in South Africa) is clearly shown on a global map of the average Xe-133 concentration for 2008 in Figure 2.21. Table 2.3 compares xenon emissions from nuclear reactors and medical isotope production facilities with those from an underground explosion, and this is another way of presenting the dominant effect of the isotope production plants. It also shows the large amount of Xe-133 generated by a nuclear explosion. A 1 percent release from an underground explosion is one to two orders of magnitude larger than the daily release from a large isotope production plant. Studies of releases from the isotope production facilities presented at the ISS09 Conference demonstrated significant changes in xenon background when facilities were switched on and off (Saey et al. 2009a, Ungar et al. 2009a). The scientific understanding of releases from medical isotope production facilities are improving based on close-in measurements at the Fleurus plant in Belgium (Ringbom et al. 2009c) and

67

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

at the Pelindaba plant in South Africa (Saey et al . 2009b) . The influence of these releases on the background xenon level at individual stations can be reasonably well modeled and predicted (Wotawa et al . 2009) . Releases from reactors have also been studied in some detail (Zhang et al . 2009) and could be of interest in specific situations even if they are less important in a global perspective . Saey (2010) provides a review of xenon in the environment . In addition to human-created sources, there is a natural background of xenon on the Earth created by spontaneous fission of uranium in soil and rock . The xenon concentration varies with the type of rock; the greatest concentration is in monazite sand, which contains thorium, and in uranium ore, in which xenon might be detected in subsoil samples . The ratio of the different xenon isotopes is different from that of xenon generated by nuclear explosions (Hebel and Kalinowski 2009) . Overall, this contribution to atmospheric radioxenon from natural background sources is small compared to that from other sources . Emissions from any source act as noise that impairs the xenon detection capability of a station in the same way seismic noise limits seismological detection capability . As xenon is not accumulated in the body, it does not constitute a health risk, and there is therefore no medical reason to reduce or limit releases of the gas . Discussions have, however, been initiated with the medical isotope plant operators to find ways to limit release and thereby increase the detection capability of the xenon network . 2.3.4 Detection The detection of xenon depends on the amount of radionuclide gas coming to a station from an explosion relative to the background noise. Although the understanding of the background noise at stations around

68

Monitoring Underground Nuclear Explosions

the world is rapidly increasing, the main difficulty in estimating detection capabilities is the uncertainty in the amount of xenon that is released from an explosion cavity. In addition, there is uncertainty in predicting the way a xenon release is distributed throughout the atmosphere (Chen and Hogue 2009). This uncertainty also applies to potential evaders, making it difficult to estimate the risk of being detected. In principle, there are two ways of making detections: a priori and a posteriori, and this applies very much to xenon. Xenon may be detected without having any prior knowledge of any particular event that might be a source, or one may look for xenon that might come from a particular event known from independent information. The latter situation is in most cases the relevant one for xenon detection from underground nuclear explosions. After a seismic event has been detected and located the key question may be: Is it a nuclear explosion? Using what is referred to Fig. 2.22 Xe-133 venting from the DPRK event on 9 October 2006 as modeled in the early stages and the further development of the plume until 27 October 2006 (Becker et al. 2009).

69

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

as forward modeling, the dispersion of a possible release from that event is calculated as a function of time, based on the actual meteorological conditions. The observation of xenon at Yellowknife in Canada on 21 October 2006 and several days after has been attributed to the DPRK event, and it is a good example. The atmospheric modeling of the dispersion of a possible release from the DPRK predicted that it should reach Yellowknife on 21 October (Figure 2.22). A weak signal was observed at the expected time that was not easily explained as coming from the Canadian medical isotope facility at Chalk River or any other known source. While the signal was of sufficient strength to be regarded as a detection, it was too weak to independently raise an alarm without prior indication from another source of information. In this sense it resembled the threshold monitoring mode of seismic data analysis. Figure 2.23 shows the Xe-133 activity observations at Yellowknife for a period of more than three years. Figure 2.24 shows a high-resolution presentation of the observations during October 2006 (top part). It also shows a predicted Xe-133 signal at Yellowknife from a release at the site and time of the DPRK explosion on 9 October 2006 (bottom part). Expected detection capability can be modeled using an assumed source term, an atmospheric dispersion model, and background noise levels at detecting stations. D’Amours and Ringbom (2009) conducted a study on the global detection capability of the planned IMS 40-station xenon network for all four relevant xenon isotopes. The study assumes source terms based on a 10 percent release of the total activity generated by a

Fig. 2.23 Activity concentrations of Xe-133 at Yellowknife, Canada, from August 2003 to October 2006 (D’Amours et al. 2009).

70

Monitoring Underground Nuclear Explosions

Fig. 2.24 The top part of the figure shows Xe-133 activity concentrations at Yellowknife, Canada, for October 2006. The lower part of the figure shows the predicted Xe-133 signal at Yellowknife from a release at the time and place of the 9 October 2006 DPRK event. The maximum influence of the known xenon source at Chalk River Labs, Canada, is shown as grey bars in the lower part of the figure (Becker et al. 2009).

1 kiloton plutonium charge after 2 .8 days, which is the time when the activity of Xe-133 reaches its maximum . The source terms and the halflives for the different xenon isotopes are presented in Table 2 .4 . Figure 2 .25 shows the global xenon detection capability in terms of the number of IMS stations that can detect xenon from hypothetical explosions at locations around the globe during a two-week period after the release . The top panels of Figure 2 .25 present the results for Xe-133, which shows the best detection results, and the bottom panels of the figure show the results for Xe-135, which has an order of magnitude fewer stations detecting the explosions . This lower detection capability for Xe135 is a consequence of its short half-life of only nine hours . The results for Xe-131m and Xe-133m are very similar to those for Xe-133, with only

71

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

Fig. 2.25  The four panels show the average number of stations detecting a release of Xe-133 (top) and Xe-135 (bottom) in July and December (D’Amours and Ringbom 2009).

72

Monitoring Underground Nuclear Explosions

marginally fewer station detections. For all xenon isotopes there is a pronounced global variation in the detection capability. The detection capability for explosions in the area around the equator is significantly lower than for explosions in the Northern and Southern Hemispheres. This is due to the pattern of the horizontal global atmospheric circulation depicted in weather maps. Less well known is the vertical component of the circulation pattern. Close to the equator, the air rises, due to the strong radiative component during the daytime; the water in the moist air condensates, and rainfall results. At about latitude 30 degrees north and south the air descends in a high-pressure area. Some of the descending air travels back equatorially along the surface, thus closing the loop and creating the trade winds. This closed loop, or cell, is called a Hadley cell. For surface-based radionuclide stations the cell structure determines their capability to detect sources at a certain latitude. There is also a strong seasonal variation of the Xe-133 detection capability seen in Figure 2.25, with enhanced capability in the winter season (December for the Northern Hemisphere and July for the Southern Hemisphere). The seasonal variation is less pronounced for Xe-135. 2.3.5 Localization To locate the source of xenon observations analysts need to model the way the gas has been transported through a complex and dynamic atmosphere from the source to the recording stations. Stohl (2009) offers a comprehensive description of the current capabilities of atmospheric transport modeling. There are two principal ways of relating observations to a source, and both involve the modeling of the xenon transport through the atmosphere. One is forward modeling, showing the way the release will disperse from a real or assumed event, as discussed above. The other is backtracking, in which one identifies the areas, or the “fields of regard,” from which a release could possibly originate given the current observations and the atmospheric conditions. In this calculation, possible contributions from each 1x1 degree area of the globe are computed every three hours, usually going back two weeks. In this way, possible source areas for observations can be identified (Becker and Wotawa 2009). The possible source area grows as the release is moved back in time. Source locations from radionuclide observations are quite uncertain even when observations from several stations are included. Given a number of uncertainties in the knowledge about and resolution of atmospheric conditions and the fact that a xenon measurement takes up to 24 hours, the uncertainty in the estimated source area is quite large and difficult to quantify (Panday 2009, Seibert 2009). This uncertainty also applies to

73

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

forward modeling and introduces a corresponding uncertainty in linking a xenon observation to a particular event. The uncertainty in obtaining a location from xenon measurements alone could easily exceed 1,000 kilometers. Radionuclide observations thus need corroborating evidence such as seismic or satellite observations to pinpoint an event. For CTBT verification, the most likely supporting scenario is a seismic event and any radionuclide observations that can be associated with that event. An interesting mode of data fusion could be when the individual technologies are working at their extremes; the seismic threshold monitoring might show an increased level in the monitored area, and a corresponding increase of the xenon level might provide supporting evidence of an event that might have taken place that needs further clarification. Such a temporal coincidence of observations by two or more technologies might create a compelling case for a state to seek further clarification. 2.3.6 Discrimination Xenon can be generated by nuclear explosions, several other human activities, in particular medical isotope production, and processes in nature. One way to establish that a xenon observation comes from a nuclear explosion and not from any other source is to use atmospheric modeling to track the observed release to the site of an observed event of interest and thereby exclude that it is coming from a known civilian source. Another option is to compare the isotope composition of the xenon observation with that expected from a nuclear explosion. One suggested discrimination method is based on the relative appearance of the four relevant xenon isotopes. As discussed previously, their relative ratio depends on how they have been generated, and Figure 2.20 illustrates isotope ratios from nuclear explosions and nuclear reactors. As shown in the diagram, there should be a clear difference between releases from explosions and reactors. Releases from isotope production and nuclear explosions might be more difficult to discriminate, as emissions from isotope production facilities are expected to show isotopic characteristics similar to that of nuclear test explosions, due to the short irradiation time (Kalinowski et al. 2009a). The likelihood of detecting Xe-135 is an order of magnitude lower than that of the other three isotopes due to the short half-life of Xe-135. Xe-131m might also be difficult to detect as explosions initially only produce small amounts and it has a long half-life, which means that it is often present as a background. For a number of observations, all four isotopes will thus not be available for discrimination. Studies have been conducted to develop a discriminant based on the two isotopes Xe-133 and Xe-133m, which have

74

Monitoring Underground Nuclear Explosions

Fig. 2.26 Typical calculated isotope ratios Xe-133m/Xe133 as a function of time after a nuclear explosion. The upper curve corresponds to a situation with complete fractionation; the trapping of all of the solid fission products and iodine. The lower curve corresponds to seepage and leakage of xenon without chemical fractionation within the fission chain (Le Petit et al. 2009).

a higher detection probability. Figure 2.26 presents the calculated ratio Xe133m/Xe-133 as a function of time after an explosion. Efforts have also been directed toward how to use statistical approaches to increase the ability to detect lower concentrations of these two isotopes (Le Petit et al. 2009). In addition, studies have been initiated on how to use data mining methods in the analysis of radionuclide observations. A machine learning contest based on observed xenon background data and added simulated explosion data is reported by Ungar et al. (2009b). A number of teams participated in the exercise, using different data mining techniques to classify the samples. This first initiative may well be followed by more. 2.4  Satellite-Based Monitoring Since the start of the 21st century, a growing number of satellites able to make optical or radar observations have been placed in orbit. Technical capabilities have gradually improved and so has data availability. Satellite data of different kinds are available for a number of security-related applications (Jasani et al. 2009). At the time of the CTBT negotiations, satellite images were essentially classified or very expensive and difficult to acquire; that is why satellite observations are not part of the IMS of the CTBT and thus will not be available to the Technical Secretariat (TS). Today satellite observations, both optical and radar, can prove very useful to states as part of their NTMs in monitoring the CTBT. Technical development over the last decade has increased the number of operational optical satellites and also improved their resolution to one meter or less. For example, QuickBird has a resolution of 0.6 meters. The time differences between successive observations of one scene vary from

75

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

one to a several days. A most significant development is that data are readily available to any user at affordable prices or free of charge on the Web. Radar satellites have also developed dramatically, and several such satellites are operational. Synthetic aperture radar (SAR), which forms the antenna by the flight path of the satellite, has improved resolution to the order of 10 meters for most satellite radar systems. Radar satellites can operate at any time of the day and even in cloudy conditions. They can provide images very similar to those from optical systems, but in another frequency range, giving prominence to different features. Radar can be used in an interferometric mode (InSAR), which enables the observation of topographic changes by making a detailed comparison of the radar response at a particular point on the ground from two separate SAR measurements. Topographic changes as small as 0.2 to 0.5 centimeters can be detected. Vincent et al. (2003) report on InSAR measurements of three underground nuclear tests conducted at NTS in June 1992. The source information of the tests is provided in Table 2.5.

Satellite photos, radar interference pattern, and observed displacements are shown in Figure 2.27. The observed ground displacements from the three explosions are in the range of one to two centimeters. Hafemeister (2009b) reports that InSAR can locate events to within 100 meters and that surface deformations from explosions with yields of more than 1 kiloton can be observed for explosions at a depth of 500 meters in many locations. It is to be expected that explosions at other locations having solid bedrock will create smaller surface deformations that are subsequently more difficult to detect. InSAR analysis is fairly complex and requires large amounts of computations, which limits its application to smaller areas. The actual event of interest must therefore be detected by seismological observations or other means, and InSAR can then be a tool to refine the analysis. Observations from space can be used by countries for CTBT monitoring in different applications. One scenario is event driven, when a state wants to look carefully at what might have happened at a particular location where a seismic event has been detected and located. Using satellite observations prior to and after an event, it is possible to detect human activities in the

76

Monitoring Underground Nuclear Explosions

Fig. 2.27  The figure shows coseismic surface-deformation signals (white dots) from three underground tests conducted at the Nevada Test Site in 1992. The top images show nearby craters (red dots) from underground tests prior to 1992. Color interferograms in middle row derived from InSAR data show surface displacement during and after explosions, and profiles in the bottom row depict near-vertical displacement (left scale) and surface topography (right scale) (Science and Technology Review 2005, in Hafemeister 2009).

area, or the absence of such activities. This analysis technique identifies the differences among observations of the same scene and can be applied to optical and radar images. Another application of satellite-based monitoring is to monitor a specific area of interest, e.g., a former or potential test site. By continuous satellite monitoring of an area, any divergence from the usual pattern of activities can be detected and analyzed using additional sources of information. In this situation, modern analysis techniques of identifying differences between successive optical or radar images also facilitate surveillance. Satellite surveillance might be an important component of precision monitoring, to be further discussed later in this chapter and in Chapter 7. 2.5  The North Korean Nuclear Explosions The two underground explosions conducted and announced by the DPRK on 9 October 2006 and 25 May 2009, even if unwanted, provided an opportunity to test the capability of the CTBT monitoring system. Many

77

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

Fig. 2.28  The figure shows estimates by the PTS and the INGV (Italy) of the locations of the two underground nuclear explosions carried out by the DPRK on 9 October 2006 and 25 May 2009, with corresponding uncertainty ellipses. The INGV location for the 2009 event is determined relative to the location estimated by the INGV for the 2006 event, with an uncertainty of about 3 kilometers only (Chiappini et al. 2009).

publications contain information on the tests (e.g., Medalia 2010, Murphy et al. 2010). Both of these explosions were detected and located from seismological observations. A number of estimates of the locations of the two explosions have been reported, all fairly close to each other, as shown in Figure 2.28. Analyzing satellite photos from the area makes it possible to see the build-up of infrastructure and other features of interest. Figure 2.29 shows what might be entrances to a tunnel complex likely used for the DPRK nuclear explosion in 2006. This provides a good illustration of the value of satellite observations to states when interpreting observed events. The seismic signals from the two explosions show characteristics expected from explosions. Even without prior knowledge that the seismologically observed events were explosions, they would be identified as such with high confidence. Figure 2.30 compares seismic P and S wave observations at regional distances from the two DPRK explosions and nearby earthquakes and chemical explosions. The figure shows that the two nuclear tests have a higher P/S signal ratio than earthquakes of similar size, especially at higher frequencies. An analysis of the source model, using signals with 100 times lower frequencies, or 0.02–0.10 Hz, shows that the 2009 explosion can be

78

Monitoring Underground Nuclear Explosions

Fig. 2.29  Satellite picture taken on 13 October 2006 of the DPRK test site. Various infrastructures, including tunnel entrances (in the upper left white section of the photo) that might have been used in connection with the test on 9 October 2006 as well as the test of 25 May 2009, are clearly visible (DigitalGlobe-ISIS 2006). Fig. 2.30 Three-component Pg/Lg spectral ratios at eight discrete frequencies used in the discrimination analysis are plotted for earthquakes, chemical explosions, and nuclear tests. The mean values at each frequency are plotted for earthquakes (solid circles) and chemical explosions (solid triangles), and the scatter is indicated as colored arms. The spectral amplitude ratios of the 9 October 2006 (red squares) as well as the 25 May 2009 (blue squares) nuclear explosion both fall in the explosion population (Kim et al. 2009b).

79

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

Fig. 2.31  Various kinds of seismic events can be grouped in a source type, or Hudson, plot. The source type can be determined based on ground motion. A perfectly spherical symmetric underground nuclear explosion would appear at the apex of the plot. The orange star shows the position of the full moment tensor solution for the 25 May 2009 DPRK nuclear test. Red symbols and ellipses are NTS explosions, blue symbols and ellipses are earthquakes in the western U.S., and green symbols and ellipses are mining collapse events. The yellow star denotes the Crandall Canyon Mine collapse in Utah in 2007 (Hellweg et al. 2009).

explained by a model similar to what has been observed from explosions in the United States (Figure 2.31). The paper by Murphy et al. (2010) contains a detailed study of the locations, source depths, and seismic yields of the two North Korean nuclear explosions. The yield estimates are 0.9 kilotons and 4.6 kilotons for the 2006 and 2009 explosions, respectively. The relative locations of the two events were determined by correlating the waveforms of the signals from 35 seismic stations that recorded both explosions. The authors estimate that the 2009 test was conducted about 2.5 kilometers west-northwest of the 2006 test. They also conclude that the two tests appear to have been conducted in a tunnel complex mined into Mount Mantap, with entrances identified on satellite imagery (Figure 2.29). The absolute coordinates of the entrance point, lat 41.28065 N and long 129.08545 E, were obtained at a very high accuracy from satellite data. The relative seismic locations were integrated with local topographic data and satellite imagery to determine absolute locations for the two explosions. Best-fitting source depths, or overburden thickness, and associated explosive yields were estimated for the two explosions. The authors conclude that the data are best fit with source depths of about 200 meters for the 2006 test and 550 meters for the 2009 test, giving the best location estimate for the 2006 explosion at lat 41.2867 N, long 129.0902 E, and for the 2009 explosion at lat 41.2925 N, long

80

Monitoring Underground Nuclear Explosions

129.0657 E. This illustrates what can be achieved by “precision monitoring” when information from different sources is combined, in this case seismology and satellite observations as well as topographic information. The satellite photo, Figure 2.29, shows clear traces of many activities that have been carried out at the testing area. It is obvious that the DPRK made no effort to diminish or hide any of these traces, and it is equally obvious that it takes great efforts to hide large-scale activities from satellite surveillance of selected areas. Radionuclide observations were made from the explosion in 2006, but no such observations have been reported from the explosion in 2009. Xenon observations in 2006 were made in South Korea, using temporarily operated Swedish equipment put into place after the explosion (Ringbom et al. 2007). A backtracking from the temporary recording station to estimate the area from which the debris could have come is illustrated in Figure 2.32. The best source location, based on backtracking of all five Xe-133 detections made at Kansong, South Korea, is 90 kilometers off the much more accurate location based on seismological observations. The origin time is also off, by 69 hours when based on Xe-133 observations only. If the ratios between several xenon isotopes can be used, the estimated origin time will be less uncertain (Becker et al. 2009). This illustrates the difficulties of making good event locations based on radionuclide observations alone even over short distances. The concentrations of the four xenon isotopes of interest for monitoring observed by the Swedish instruments in South Korea are shown in Figure 2.33. For three of the four isotopes, signals related to the DPRK explosion could be detected. No Xe-135 coming from the test was observed. Due to the long transport time of the sample from South Korea to the laboratory in Stockholm, any Xe-135 had decayed away due to its short half-life. Fig. 2.32 Field-of-regard for the strongest Xe-133 detection encountered by Ringbom et al. (2007) at Kansong in South Korea (yellow square) for the event in DPRK on 9 October 2006. The field-of-regard is the result of backward atmospheric transport modeling. The two ellipses constitute different uncertainty estimates for the location of the event, based on seismic data (Becker et al. 2009).

81

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

Fig. 2.33  Measured concentration time series in South Korea for four xenon isotopes for the period 11 October 2006 – 26 January 2007. The time of the explosion is indicated by the black arrow in the uppermost panel (Ringbom et al. 2009d).

Figure 2.34 shows the Beta spectrum of the observations made in South Korea and analyzed in Stockholm. The distributions of the two metastable isotopes Xe-131m and Xe-133m have been modeled to fit the observed data. A comparison of the measured Xe-133m/Xe-133 ratios, with calculations assuming a plutonium (Pu-239) device, indicates that the release occurred within hours of the explosion (Ringbom et al. 2009d). The fact that no radionuclide releases have been reported for the DPRK explosion in 2009 has been widely discussed. Was it really a nuclear explosion, or was it, despite the official announcements by the DPRK, a large conventional explosion? To set off a conventional explosion of about 4 kilotons is a large-scale and complex operation involving 4,000 tons, or some 2,000 cubic meters, of high explosives that must be transported, loaded, and exploded in an orderly fashion. It is hard to find good reasons to make so much effort to fake a nuclear test. The other explanation is that the DPRK managed to contain the explosion in an efficient way and

82

Monitoring Underground Nuclear Explosions

Fig. 2.34   This Beta spectrum is from a sample obtained in South Korea on 14 October 2006 and analyzed in Stockholm on 21 October 2006. The actual measurement of Beta intensity for different energies is represented as crossed bars (+). The Xe-133 background is shown in green and the Xe-135 signal in blue. The distribution of the two metastable isotopes Xe-131m and Xe-133m have been modeled so that the combined distribution, the black curve, fits the observations (Ringbom et al. 2009d).

prevent major leakages that could be observed outside its borders. With good experience in underground construction and working in solid rock, and with the possibility of obtaining a sufficient overburden coverage, it would be possible to contain the explosion sufficiently to prevent detection by the IMS network and other stations outside the DPRK. As noted above, there are indications that the source depth of the 2009 explosion was deeper than that of the 2006 explosion. The two DPRK explosions created a great deal of interest within the PrepCom among representatives from states and PTS staff, and a number of briefings on the two events were provided by the PTS. The PTS, and after entry into force the TS, should analyze data according to agreed-upon procedures and not draw any conclusions or make any statements on the nature of specific events. This marks a significant difference in responsibility between the International Atomic Energy Agency (IAEA) in relation to the NPT and the PTS/TS in relation to the CTBT. As many Vienna-based diplomats represent their countries at both the PrepCom and the IAEA, it is important to be aware of the differences between the two organizations. Countries must not expect more from the PTS/TS when it comes to monitoring than the organization’s charter allows. As discussed in several places in this book, the CTBT provides for verification of compliance to be carried out by states based on data and information from many sources, including the IMS. It is likely that a country or a group of countries will focus its monitoring on areas that are of concern to it; this is discussed further later in this chapter and in Chapter 7. The two North Korean nuclear explosions served as instructive illustrations of the capabilities of the CTBT verification system. The 2006

83

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

event showed the power of synergies among the different monitoring technologies; seismic stations pointed to an explosive source at a welldetermined location, and data from xenon stations indicated that the source was of nuclear origin. For the 2009 event, seismological data again indicated an explosive source at an even better–determined location, but could not, in the absence of associated radionuclide observations, identify the explosion as nuclear or conventional. The OSI regime could play an important role in such situations; an OSI has the potential to resolve the issue of the nature of an event when considerable uncertainty persists after using all monitoring resources available to states. 2.6  States and “Precision Monitoring” A country is likely to be interested in monitoring countries and areas where it may be concerned that nuclear explosions in breach of the CTBT might be conducted. It is unlikely that a state is interested in monitoring the entire globe with the same ambition. Selection of monitoring targets is a national political decision, and different states might have different priorities. By focusing its monitoring resources and interest on one or a few limited areas, a country or group of countries having the same political priorities can achieve a significantly improved monitoring capability. Recent scientific and technical developments in different monitoring technologies and in the ability to analyze and interpret data have also provided countries with improved tools. High-resolution studies of the crust and earthquake sources have created the concept of “precision seismology,” which strives to improve on the quality of seismological interpretation, particularly on a regional basis. This is very much in line with the needs of states in their monitoring of the CTBT, and the term “precision monitoring” has been adopted in this book for such focused efforts. Seismology is a strong global science that is developing independent of activities related to CTBT verification. Ongoing scientific development increases understanding of the detailed structure of the Earth’s interior. This and the rapid increase in observations of earthquakes and mining explosions will make it possible to more fully utilize seismological observations at local and regional distances from an event. The IMS has demonstrated a better detection capability than anticipated during the CTBT negotiations. Since the start of the 21st century, the number of high-quality seismological stations outside the IMS from which data are readily available has radically increased. There are agreements in place to freely exchange data in real-time from thousands of stations with continuous waveform data available in easily accessible archives.

84

Monitoring Underground Nuclear Explosions

This means that the amount of data available to countries has increased dramatically. Countries can significantly improve their monitoring capabilities, benefiting not only from more available data, but also from an increased understanding of how to use data obtained at short and regional distances from a source. Detection capabilities even below magnitude 2, corresponding to an explosion yield of less than 10 tons, or 0.01 kilotons (NAS 2002a), can be obtained through precision monitoring of selected areas. Given this good detection capability, it is likely that seismic events are also observable in areas with low seismic activity. The assessment of such events will be greatly facilitated by the new paradigm suggested for the analysis of seismic events, in which new observations are related to past experience in an area. The studies presented at the ISS09 Conference provided a coherent picture of event location accuracy. Half of the PTS seismic event locations are within some 20 kilometers of the true source locations; the other half has larger errors. If this is related to a situation in which an OSI may be requested, it means that every second event reported by the PTS has its true location outside the maximum area that can be covered by an OSI. Studies show that the uncertainties in event locations can be substantially reduced by using calibration data and new location techniques, particularly for limited geographical areas in line with the precision seismology concept. Using all available seismological information, and carefully calibrating their areas of interest, countries will no doubt achieve the high location accuracy needed for possible requests for an OSI of an event in those areas. Confident discrimination between explosions and earthquakes, especially for smaller events, must be approached on a regional scale. A key concern for CTBT monitoring is to characterize observed events to determine if they are earthquakes or belong to known human activities such as mining, or if they need further clarification. The fact that less than half of the events reported by PTS can be screened out as earthquakes illustrates that characterization using globally applicable procedures is unsuccessful. A regional approach is needed for event characterization, optimum detection, and localization. A careful examination of the earthquakes that take place in areas of interest and of the seismological signals observed from these earthquakes at the monitoring stations a state chooses to use might gradually create a picture or a model of the seismicity in these areas. Each new event can then be compared with that model to see if it fits the established pattern. The existing methods and procedures for seismological data analysis used by the PTS date back almost three decades, and the time has come to create a new paradigm for such analysis. Developments in data analysis and exploitation have created the tools needed to form a new and integrated

85

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

approach to the detection, location, and characterization of seismic events using data from the rapidly expanding data resources available from regional and global networks. Yet, even if most mathematical tools for such a new paradigm are available, there is still a need for concerted efforts to actually develop the necessary procedures and computer programs. This approach can be used on a global scale, but it is more easily developed and implemented as a key component in precision monitoring of limited areas. The new IMS xenon network provides an important monitoring and deterrence asset down to low yields. It is a new technique that has developed significantly since the early 21st century. The capability of the technique is still not fully understood: xenon observations were made from the first DPRK explosion in 2006 but not from the second in 2009. It is far from understood how much xenon might leak from an explosion in different media and under different circumstances. The leakage is likely to be more critically dependent on the explosion environment than on the yield. This uncertainty is present also for a potential evader, and the xenon technique might thus be an important deterrence against even low-yield tests. The sensitivity of the xenon system, as with any other system, is limited by the operational performance of the individual stations, the extent of the network, and the background noise. There is a potential to improve on the capability of the individual stations by making them more robust and reducing downtime. Sensitivity could be increased by implementing advanced new sensors and by more frequent measurements in denser networks to catch the nuclides with the shortest half-lives. More frequent sampling would also improve the accuracy in the estimates of the time xenon is observed at a station. A more accurate timing would reduce the uncertainties in the results of the atmospheric transport modeling of the nuclides, be it backtracking or forward modeling. In addition to the IMS network, which is limited by the treaty to an initial installation of 40 noble gas stations to be followed after entry into force by another 40 stations, states can install additional stations or use mobile or airborne equipment to make measurements related to events of interest. The xenon background is essentially generated by a few medical isotope production facilities. To limit or stop these releases at the facilities is feasible and should significantly improve the capability of the xenon network. Atmospheric transport models have been developed to either trace a certain observation back to a possible source region or to assess if a xenon or particulate observation might come from a particular seismic event. The uncertainty in either backtracking or forward modeling is considerable and is likely to remain so, even if models are improving as understanding of the dynamics of the atmosphere continues to increase. Radionuclide

86

Monitoring Underground Nuclear Explosions

observations thus need to be interpreted together with additional information, including seismological data, satellite observations, or information from other sources. The possibility for countries to use satellite observations to monitor the CTBT has increased significantly since the treaty was negotiated. This applies both to optical and radar systems. The optical systems have a spatial resolution of 1 meter or less; for radar systems this resolution is about 10 meters. Radar systems have all-weather capability and also the ability to detect small deformations of the Earth’s surface at the centimeter scale. A number of systems are operational, giving good coverage with repeated observations of a scene within one or two days. The most dramatic development is that data from all of these open systems are readily available at a low cost. Computer software is also available to analyze the large amounts of data involved and identify changes over time in a selected area. Satellite observations have developed into a valuable tool for states to use when monitoring limited areas of interest or to analyze a particular event. There are a number of monitoring tools available to countries, with capabilities that have developed appreciably since the CTBT was signed. To use precision monitoring of areas of interest, a state or a group of states can apply these tools in an integrated way. These tools can be used by countries in two principal ways: to continuously monitor that no activity above the prevailing detection threshold takes place and to assess with precision any event observed in the area. Threshold monitoring using seismological and xenon observations will continuously show the largest seismic event or xenon release in the area that otherwise would go undetected. Using satellite observations in a similar mode can help determine that no significant changes are made in the infrastructure of interest. Capable tools are available. What remains is for states to become involved, to create the necessary competence and systems, and to build up the necessary background or models of activities in the areas of interest. This can be done either at a national level or, as discussed in Chapter 7, together with other countries having the same political priorities as regards CTBT monitoring.

87

Chapter 3

Monitoring Atmospheric Nuclear Explosions 3.1  Thirty Years since Last Nuclear Explosion in the Atmosphere A little more than 30 years ago, in October 1980, China conducted its last nuclear test explosion in the atmosphere. This marked the end of atmospheric testing by all countries. The United Kingdom, United States, and Soviet Union, as depositories, had already signed the Partial Test Ban Treaty (PTBT) in 1963, which banned explosions in the atmosphere, in space, and under water. France, the fourth country that conducted atmospheric tests, ended its testing in the atmosphere in 1974. A total of 517 nuclear explosions were carried out in the atmosphere: United Kingdom 21, China 22, France 50, United States 210, and Soviet Union 214. As noted in Chapter 1, the PTBT contains no verification provisions. There might be several reasons for that: explosions in the atmosphere are generally easy to detect, and national monitoring equipment, such as for atmospheric radionuclide surveillance, was already available and operational at the time the treaty was signed. The United States launched its satellite-based Vela project in 1963 to monitor atmospheric nuclear explosions, as discussed under national technical means (NTM) below. The three depositories of the PTBT were also ready to move testing underground to reduce radioactive fall-out and thereby also the information that other states might obtain from analyzing the fall-out.

O. Dahlman et al., Detect and Deter: Can Countries Verify the Nuclear Test Ban?, DOI 10.1007/978-94-007-1676-6_3, © Springer Science+Business Media B.V. 2011

89

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

3.2  Many Detectable Features A nuclear explosion in the atmosphere exposes all the explosion phenomena to observation (Figure 3.1). Such an explosion generates a strong shock wave, ionizing radiation as well as electromagnetic radiation at long wavelengths, and a strong light pulse. Furthermore, a nuclear explosion generates radioactive fall-out from the explosion device itself and fall-out induced in the surrounding material by the initial ionizing radiation. All these phenomena are or generate detectable features. Prompt radiation can be detected at such great distance within line of sight that it is observable by satellite-mounted equipment. The shock wave generates acoustic signals in the atmosphere, and such infrasound signals can be detected at large distances. Seismic signals are generated by the shock wave when it hits the ground; these signals are considerably weaker than those generated if the explosion takes place underground. The shock wave also creates lasting effects over a large area if the explosion is carried out over land, and these effects are easily detectable by satellite photo observations. An explosion over water is likely to create hydroacoustic signals, which are discussed further in the next chapter. The radioactive fall-out, either in the form of noble gases or particles, is distributed in the atmosphere of the hemisphere where the explosion took place and does not cross the equator. This is caused by the global atmospheric circulation pattern, from the equator to the poles, which prevents the air masses in the Northern and Southern Hemispheres from mixing, as discussed in Chapter 2. The certainty that any test explosion in the atmosphere would be detected should serve as a strong deterrence, and clandestine atmospheric nuclear testing may thus no longer be a serious concern to countries. In addition, the efficient verification tools available should make countries confident that any atmospheric test will be detected. An explosion over land would likely be attributed to a testing state by its location. To attribute explosions on the oceans, especially in the high seas, is less trivial. One way to do this is to identify the testing operation and its logistics; another is to use nuclear forensics to identify the fissile material used, using national techniques normally not associated with verification in the Comprehensive Nuclear-Test-Ban Treaty (CTBT). Verification of atmospheric explosions has not attracted much attention since the last atmospheric test 30 years ago, although the United States recently launched a new generation of satellite-based NTMs.

90

Monitoring Atmospheric Nuclear Explosions

3.3  Infrasound Monitoring At the time of atmospheric nuclear explosions, nations conducted infrasound observations in several parts of the world. Some of the very large explosions carried out in the early 1960s, with yields up to 58 megatons, created low frequency waves that were observed circling the Earth several times. These waves, propagating in a channel between the Earth’s surface and the tropopause at an altitude of about 10 kilometers, are called Lamb waves, or acoustic gravity waves. Explosions in the kiloton range were not able to generate this kind of wave, and such explosions were observed only at distances of some hundreds to a 1,000 kilometers from the test site. The observed acoustic waves strongly depend on the atmospheric propagation conditions, as shown in Figure 3.2. The two infrasound recordings from a French test in Polynesia were obtained at two infrasound stations at almost the same distance of around 450 kilometers from the explosion but in nearly opposite directions—one to the east and the other to the northwest. The effect of the stratospheric winds, blowing toward the east at the time of the observation, can be clearly seen on these recordings; the propagation time is some five minutes, or 20 percent, shorter to the station in the east, and the signal at that station also has higher frequencies and three clear signal phases. The detection capability is generally dependent on the stratospheric winds at altitudes of some 50 kilometers, which are part of the global circulation pattern, with westerly winds in the Northern Hemisphere in the winter and easterly winds in the summer.

Fig. 3.1  Atmospheric explosion named Badger, with a yield of 23 kilotons at the Nevada Test Site on 18 April 1953. The energy distribution from a low altitude nuclear explosion is air blast 50 percent, thermal radiation 35 percent, and nuclear radiation 15 percent (Blanc 2009).

91

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

During the CTBT negotiations it was decided that an infrasound network should be part of the IMS. Based on past infrasound recordings obtained at large distances and modeling based on the knowledge of the atmosphere at that time, a 60-station network was designed. This infrasound network is unique and was near completion at the beginning of 2011. It has provided observations from a number of natural and man-made sources, which are discussed below, and demonstrates that acoustic wave propagation in the atmosphere is very complex. The CTBT triggered a renaissance in infrasound. The many new observations provided by the IMS infrasound network are associated with a number of different infrasound sources. Infrasound observations have also increased the understanding of acoustic wave propagation in the atmosphere, which will increase our knowledge of the dynamics of the atmosphere and thereby provide important contributions to climatology and meteorology. Infrasound observations may even prove more interesting and important for studying the atmosphere than for monitoring nuclear tests.

92

Fig. 3.2  Recordings and spectrograms (top panels) of a French nuclear test explosion of a few kilotons in Polynesia at two different infrasound stations: Hao, 440 kilometers to the northwest of the explosion, and Gambier, 450 kilometers to the east, as shown on the map (lower panel) (Blanc 2009).

Monitoring Atmospheric Nuclear Explosions

3.3.1  Infrasound Observations Infrasound observations face two challenges: to detect signals in the presence of disturbing wind at the stations, and to analyze and interpret the observed signals, given the complex wave propagation in the atmosphere. A great deal of effort has gone into coping with those challenges over the last decade, and progress has been made. The geometrical configuration of IMS infrasound stations has been gradually modified and, as far as possible, tailored to local conditions. An IMS infrasound station is an array having four or more elements within a distance of 1 to 3 kilometers. Each station element is connected to a number of pipes, each of which has a number of outlets. The purpose of this arrangement is to average out local pressure disturbances created by the winds, which vary rapidly and are uncorrelated among the outlets. The pattern of the outlets for a station in Greenland is shown in Figure 3.3. The IMS infrasound stations have developed into quite sizable installations, and technical developments continue to improve their capability (Arrowsmith and Whitaker 2009). Data from all IMS infrasound stations are transmitted online to the Provisional Technical Secretariat (PTS) in Vienna, where correlation techniques in a number of frequency bands are used to detect infrasound signals (Brachet et al. 2009). The analysis provides, for each station, arrival times and the direction of arrival and velocity for the detected signals, as well as their frequency content and coherency. Several distinct signals that have propagated along different paths, e.g., to different altitudes, may be observed at a given station from one and the same source. To locate an event, observations from two or more stations must be combined. Experience has shown that only a limited number of events generate signals detectable by two or more stations. Thus, a large number of Fig. 3.3 Pattern of outlets to reduce the wind noise at infrasound station IS18, Qaanaaq (or Dundas), Greenland. The number of these “rosettes” may vary according to the local wind conditions (Campus 2009).

93

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

signals are observed only at one station. For such signals only the azimuth to the source can be estimated, and it is impossible to locate the source from infrasound observations alone. In the spring of 2010, the PTS started to include infrasound observations in its routine bulletins. The automatic processing produces many tentative events that are rejected in the analysts’ review; infrasound data contributes to the definition of an average of three events accepted into the Reviewed Event Bulletin (REB) per day. Most of these accepted events have seismological observations in addition to the infrasound observations. Special kinds of sources, like volcanic eruptions and microbaroms, may continuously emit infrasound signals. The Eyjafjallajökull volcano in Iceland, which erupted in the spring of 2010, produced continuous detections for days at multiple IMS infrasound stations without any corresponding detections at the IMS seismic stations. In February 2008 a meteoroid exploded in Oregon just above the USArray, a dense experimental seismic network discussed in Chapter 2 (Woodward et al. 2009). Signals were recorded at three IMS infrasound stations (Figure 3.4), and the infrasound signals were also recorded by more than 200 of the seismic stations, allowing a detailed analysis of the wave propagation. It is important to gain insight into the propagation of infrasound signals for the understanding of the detection and location capability of the IMS infrasound network (Hedlin et al. 2009).

Fig. 3.4.  Infrasound recording at the IMS station at Newport, Washington, from a bolide over Oregon on 19 February 2008. The recording shows the waveforms at four elements of the station and the estimated speed and azimuth of the observed signal (Hedlin et al. 2009).

94

Monitoring Atmospheric Nuclear Explosions

3.3.2  Infrasound Sources As a young and recently revived science, infrasound scientists are now rediscovering the many sources first detected in the 1970s by a research group at Lamont-Doherty (Donn and Rind 1971, Rind and Donn 1975). Modern instrumentation and a global station network offer today’s scientists observations of higher quality and global coverage. A reference event database has been established by the PTS that contained 750 events as of July 2010. Among those events are mine and quarry blasts, other explosions, volcanic eruptions, shallow earthquakes, and bolides, as well as rocket launches and re-entries. Studies of signals from continuously radiating sources, such as volcanoes, might provide interesting data to study short-term changes in the upper atmosphere. These and other scientific applications are further discussed in Chapter 6. The Earth is constantly bombarded by meteoroids of different sizes that are interesting sources of infrasound. On average, a 10 centimetersized meteoroid hits the Earth every 30 minutes with a source energy corresponding to an explosion of 10 kilograms of TNT. This is visible in the sky as a bright meteor which is called a bolide. A meteoroid of one meter size with a source energy of 0.1-1 kiloton hits the Earth a couple of times per month. A large meteoroid of some 10 meters diameter and an equivalent energy of tens of kilotons enters the Earth’s atmosphere once per decade. These explosion-like events can be used to test and evaluate the infrasound network, especially to estimate the capability to detect high-altitude explosions. Three large bolides with an equivalent yield in the range of 10 to 20 kilotons were observed in 2004 and 2006, and all three were recorded by the IMS infrasound network (Edwards et al. 2009) (Figure 3.5). On 8 October 2009, a meteoroid exploding at high altitude off the

Fig. 3.5 Location determination of a large bolide on 7 October 2004,, using worldwide instrumentation to detect infrasonic waves (Edwards et al. 2009).

95

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

coast of the Indonesian island of Sulawesi generated infrasound signals that were detected at 14 stations at distances up to 14,000 kilometers from the source. This is the largest such event so far detected by the IMS infrasound network (Mialle et al. 2010). Another meteoroid created an impact crater in the border region between Peru and Bolivia on 15 September 2007 (Figure 3.6), when both infrasound and seismic signals were recorded from the event. Rocket launches and re-entries are examples of human activities in the atmosphere that generate infrasound signals. Rocket launches from the Baikonur Cosmodrome in Kazakhstan have been routinely detected at an infrasound station 600 kilometers away and the initial trajectory of the rocket could be observed (Gopalaswamy and Smirnoff 2009). This poster also contains recordings obtained from the re-entry of a Sayuz TMA-11 capsule. The North Korean test of a Taepodong-2 long-distance missile on 5 April 2009 was detected by infrasound stations in South Korea. The missile could be tracked during its initial phase (Kim et al. 2009a). The missile was also a well defined source, providing valuable data on infrasound signatures and wave propagation. Accidental or planned explosions provide good opportunities to test and evaluate the infrasound network as they usually have well defined locations, origin times, and yields. An accidental explosion in an ammunition facility at Gerdec in Albania in 2008, with an estimated yield of the order of 1 kiloton, was detected by several infrasound stations up to a distance of 5,000 kilometers (Green et al. 2011). Other accidental explosions providing infrasound observations include explosions in Chelopechene, Bulgaria, in 2008; Buncefield, United Kingdom, in 2005 (Ceranna et al. 2009a); and Novaky, Slovakia, in 2007. These explosions had estimated yields of 5 to 500 tons and were recorded by several stations at distances up to 5,000 kilometers (Green et al. 2008). Test and calibration explosions have been conducted by Israel at its Sayarim Military Range in the Negev Desert since 2008. Signals from a test explosion Fig. 3.6 Meteorite impact crater created in the border region between Peru and Bolivia in the village of Carancas, Peru, on 15 September 2007. The diameter of the hole is roughly 20 meters (Minaya et al. 2009).

96

Monitoring Atmospheric Nuclear Explosions

recorded at a distance of 277 kilometers are presented in Figure 3.7a, and a photo of a 10 ton test is shown in Figure 3.7b. A calibration explosion, with a yield of 82 tons of conventional explosives, was conducted on 26 August 2009 and another explosion of 100 tons was conducted on 26 January 2011 (NTI 2011). From the explosion in 2009, infrasound signals were recorded at a large number of temporary and permanent infrasound stations out to a distance of 3,000 kilometers, including two of the IMS infrasound stations, one in Germany at a distance of 2,750 kilometers and one in Tunisia at 2,460 kilometers from the explosion (Gitterman 2010, Mialle et al. 2010).

Fig. 3.7a  Infrasound recordings obtained at station MMLI in Israel at 277 kilometers distance from a test explosion at the Sayarim Military Range in the Negev Desert in 2008. The signals arrive at different velocities and with different frequency content (Gitterman et al. 2009). Fig. 3.7b  Photo of a 10 ton test explosion at the Sayarim Military Range in 2008. Insert: a 1-ton shot 10 minutes later (Gitterman et al. 2009).

97

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

Infrasound signals from the explosion in January 2011 were recorded at three IMS infrasound stations out to a distance of 6,200 kilometers to the east. The observations have not been fully evaluated at the time of this writing. These observations of actual explosions support the estimates of global detection capabilities presented below. Seismological signals were also observed from both explosions as discussed below. 3.3.3 Detection The detection capability of infrasound can be estimated in two ways: from observations of actual explosions or through modeling. As discussed above, infrasound signals have been observed from a limited number of explosions with yield in the range of 1 kiloton or below. Also bolides, for which the equivalent yield has been estimated by optical means, might serve as calibration events. Yields can also be estimated from the period and amplitude of the observed infrasound signals. The number of such observations are, however, too few to allow an estimate of the global capability based on those data alone. The observational data would rather serve as a reality check of the results obtained by modeling. Models to estimate the global detection capabilities are generally based on models of the atmospheric conditions, in particular the wind distribution geographically and over time, noise distribution at recording stations, and the relation between explosion yield and infrasound signal strength. As noted above, detection capabilities depend strongly on the stratospheric winds, and stations downwind record stronger infrasound signals. Comparing results of two different models by Green and Bowers (2010) and Le Pichon et al. (2009a, b) shows that by and large they agree: both models arrive at the same general conclusion—that explosions down to 1 kiloton can be detected with high confidence all over the world at any time of the year. The models also show that there is a substantial, but varying, probability that explosions with yield substantially below 1 kiloton can also be detected. Figure 3.8 compares the seasonal variations of the estimated global detection capability during a year (Green and Bowers 2009). This illustrates the dramatic variations in detection capability over the Earth’s surface and over time. The model by Le Pichon et al. (2009a, b) shows a similar pattern, strongly influenced by the seasonal trends in the stratospheric winds, and these authors have also made an attempt to present the average global detection capability and the variation of that capability over the year. They show that 100 ton explosions can be detected over half of the globe and 1 kiloton explosions over the whole globe. These estimates of the detection capability refer to the IMS infrasound network. In addition, states may use data from local and regional stations

98

Monitoring Atmospheric Nuclear Explosions

and networks. The denser infrasound networks established in Europe and South Korea have made it possible to detect a large number of infrasound events in those areas. Around 1,000 events are observed annually in Europe, mostly related to quarry blasting and supersonic aircraft (Ceranna et al. 2009b). The network in South Korea observes about half that number; Figure 3.9 shows the locations of these events, which are mainly

Fig. 3.8 Estimated global detection capability for atmospheric nuclear explosions. These results were obtained through modelling of the atmosphere. The considerable differences in detection capabilities are caused by the large seasonal changes in atmospheric circulation patterns, which lead to equally large changes in the profiles of wind and temperature. In turn, these profiles determine the wave propagation properties of infrasound and therefore the detection capability (Green and Bowers 2009).

99

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

Fig. 3.9  Distribution of seismo-acoustic events in South Korea in the period 1999 to 2008. Most events are concentrated in specific areas: A) construction of a waterway, B) mining, C) public works, D) dam construction, E) quarries and military exercises, F) civil work for an airport, G) limestone quarry, S) mining and military exercises, H) and I) civil works on a highway, and K) quarry mining (Lee et al. 2009).

associated with mining, construction work, and military exercises. This is an example of precision monitoring using infrasound observations and supplementary national observations. 3.3.4  Localization and Characterization The preceding discussion has shown the complexity and the dynamics of infrasound propagation in the atmosphere. This strong dependence of infrasound signals on conditions in the atmosphere has two consequences: first, infrasound signals are fine tools to study atmospheric conditions, a subject discussed further in Chapter 6, and second, it is difficult to analyze and interpret infrasound signals without a good understanding of the actual atmospheric conditions. Chapter 2 discussed the location of seismic events and the importance of calibrating the velocities of seismic waves to obtain precise locations. Seismic waves have the inherent characteristic that they do not change over time or over a given path. Acoustic signals in the atmosphere, on

100

Monitoring Atmospheric Nuclear Explosions

the other hand, propagate through a complex medium that changes its characteristics, sometimes very rapidly. The way infrasound signals are influenced by short term and small scale changes in the atmosphere is illustrated by the different appearance of signals obtained at the same stations from two co-located explosions only minutes apart. Calibration events will thus not improve the ability to analyze infrasound recordings unless they occur very close in time and space to the actual event. Procedures to understand the actual atmospheric conditions and good models to estimate wave propagation contribute to improved locations, but the uncertainties might still be hundreds of kilometers and much larger than those achieved from seismic observations. Infrasound locations are also hampered by the fact that very few infrasound events are observed at more than two or three IMS stations, with each station providing arrival times and the direction of arrival of each signal. Sometimes a recording contains several “phases” due to multi-pathing in the atmosphere. The observed arrival direction at a station might be biased due to wind or topography. The wave propagation velocity varies over time and can be different in different directions from the source as illustrated in Figure 3.2. These conditions make it difficult to locate infrasound events with high accuracy. If the infrasound stations have a good azimuthal coverage around the source, however, location uncertainties could be quite small and, under certain conditions, even better than those obtained by seismic recordings. Military applications of seismic and sound measurements at short distances show the same result. If the observing stations are located in almost the same direction from the source, the location errors might, on the other hand, be very large. Location uncertainties are further discussed in Le Pichon et al. (2009a, b) and in Green and Bowers (2010). As infrasound signals are very much affected by rapidly changing meteorological conditions, there is a great variation in their appearance. The frequency content is, however, a good measure of the source intensity and few sources can generate signals in the frequency range 0.1–0.2 Hz, which is representative for low yield nuclear explosions. 3.4  Seismological Monitoring Atmospheric explosions can also generate seismic signals, created by the interaction of the atmospheric shock wave with the ground. The signals are weaker than those generated by an underground explosion of corresponding yield. As nuclear testing in the atmosphere ended in October 1980, only some old seismological observations exist from such explosions. Seismological observations have also been made from chemical test explosions and from accidental explosions in the atmosphere.

101

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

Seismological signals from 44 atmospheric nuclear explosions conducted at the Nevada Test Site from 1951 to 1958 were recorded at Pasadena, at distances of about 375 kilometers. The explosions had yields in the range of 1 to 60 kilotons and were detonated at altitudes above ground from about 100 to 2,000 meters. The observed signal amplitudes were proportional to the 2/3 power of the explosion yield and to the scaled altitude, that is, altitude divided by the third root of the yield. The explosions were carried out at four different areas at the test site, and a fairly strong variation was observed among these areas (Ericsson and Dahlman 1966). Model experiments, in which seismic signals were recorded at distances up to 1 kilometer from charges ranging from 0.05 kilogram to 5 kilograms and detonated at elevations from 0.1 to 10 meters, showed essentially the same relation among signal amplitude, charge size, and altitude (Dahlman et al. 1968). An interesting observation was that there seems to be an optimal scaled altitude for generating seismic signals. For the actual testing ground, it was 0.5 to 1 meter for 1 kilogram, corresponding to 50–100 meters for a 1 kiloton explosion. More recently, seismic signals were observed from the two Israeli test explosions in 2009 and 2011 of about 100 tons, discussed above. For both the explosions, seismic signals were recorded at a number of Israeli stations at distances up to 340 kilometers. For the 2011 explosion, observations were reported from three IMS auxiliary seismic stations. The local magnitude for the 2009 explosion was estimated at 2.5 (Gitterman 2010) and for the 2011 explosion at 2.7 (Space Daily 2011). Even if only a limited amount of seismological observations of atmospheric explosions have been made, it is fair to conclude that such observations can contribute to the monitoring of explosions in the atmosphere. In the context of precision monitoring, discussed in Chapter 2, when a country focuses its monitoring on one or a limited number of countries and areas, it might be possible to achieve a good capability to detect explosions in the atmosphere as well. 3.5  Radionuclide Monitoring An atmospheric nuclear explosion releases into the air all the nuclear elements generated—radioactive particles as well as radionuclide gases. Additional radioactive elements are created when the prompt radiation interacts with surrounding material. The composition of these induced radioactive particles depends on the material in the explosive device and in the immediate surroundings of the explosion. These induced radioactive particles also form part of the fall-out detectable from an atmospheric nuclear explosion.

102

Monitoring Atmospheric Nuclear Explosions

The IMS radionuclide particulate network is the first global network established, comprising 80 stations distributed fairly evenly around the world (see Figure 1.3). Radionuclide particles are collected and analyzed at these stations, about two thirds of which are already operational as of early 2011. Samples containing particles of special interest are analyzed at one or more of the 16 IMS radionuclide laboratories. Upon entry into force, 40 of the 80 radionuclide stations will also be equipped to observe noble gases, especially xenon, as discussed in Chapter 2. Nations have measured radionuclide particles since the beginning of nuclear testing. For an atmospheric nuclear explosion the prompt radionuclide radiation, such as neutron and gamma radiation, as well as light and heat, are exposed to observation. There is no provision for measuring these types of radiation as part of the CTBT verification regime. Any monitoring of them is to be carried out at the national level as part of NTMs, which will be discussed in section 3.6. 3.5.1  Monitoring Radionuclide Particles During a nuclear explosion, large quantities of radioactive fission products are produced, and their composition depends to some extent on the fissile material, uranium or plutonium, used in the nuclear device. The initial nuclear radiation from an explosion, especially neutrons from thermonuclear devices, will interact with the surrounding materials in the device and in the environment and create radioactive activation products. Many of the radioactive products decay and disappear rapidly, while others have half-lives that fit well into the CTBT verification scheme: long-lived enough to survive the several days time scale of sampling and measurement, but short-lived enough to emit substantial radiation and not stay around long enough to create a natural background. Among the most potent nuclides to signal a nuclear explosion are barium/lanthanum-140 zirconium-95 and -97, molybdenum-99, iodine-131, -132 and -133, and ruthenium-103. Each radionuclide station is a rather simple installation where radionuclide particles are collected by blowing a large amount of air through filters at the rate of 500 cubic meters/hour or more. Filters are changed every day so that each filter is collecting particles for a period of 24 hours. A filter then rests for a day so that some decay products of radon collected by the filter can decay. The radiation from radon decay products would otherwise mask interesting radionuclides during the measurement. The filters are then measured at each individual station by gamma spectroscopy for another 24 hours, which makes the whole measurement cycle last 72 hours (Werzi 2009).

103

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

The observed spectra are transmitted from the individual stations to the PTS, where the information from the entire network is compiled and reported to states. The observed spectra are categorized into five levels: 1) normal background, 2) anomalous background, 3) contains relevant radionuclide(s) typically seen at this station (“relevant” refers to radionuclides that might be part of fall-out from a nuclear explosion), 4) contains one relevant radionuclide not typically seen at this station, and 5) contains multiple relevant radionuclides not typically seen at this station (Nikkinen et al. 2009). All level 5 samples are split and sent to two of the 16 certified radionuclide laboratories to be re-measured. These certified laboratories are equipped and their staffs are trained to make more sensitive measurements and detailed analysis than can be done at individual stations (Comley et al. 2009). There is also ongoing activity to improve the instrumentation and, as mentioned below, there is a large potential to improve the sensitivity (Turunen et al. 2009). As part of a quality assurance program, PTS arranges an annual proficiency test with the certified laboratories, where they identify the different isotopes in a test sample (Duran et al. 2009). The filters are also sent to the PTS, where they are collected and stored. This collection of worldwide samples obtained over a long period of time could prove interesting, not only to examine long-lived nuclear isotopes but also to look at non-nuclear particles. Large amounts of particles, organic and non-organic, are collected in the filters, and filters from highly polluted areas are gray or black. This might be interesting material to study in connection with global pollution as well as the spreading of organic material on a regional and global scale, which is further considered in Chapter 6. In addition to the IMS radionuclide stations, a state can monitor with its own national stations, and several did so prior to the establishment of the IMS. Some states use filter collection systems operating on aircraft, normally installed in outside pods that allow air to stream through during flight. This makes it possible to collect samples and even do in-flight measurements in areas where the plume from a possible explosion is expected to appear, both in national and international airspace. 3.5.2  Particulate Detection and Tracking IMS particulate stations are required to have a sensitivity of 30 microBq/ cubic meter for Ba-140, one of the isotopes of most interest in fall-out from nuclear testing in the atmosphere, and stations generally meet that requirement. The critical step is the measurement of the exposed filter. The sensitivity of the stations could be improved by installing new gamma measuring equipment and by enhancing their shielding

104

Monitoring Atmospheric Nuclear Explosions

to reduce background interference. Having a longer “cool-off” period to further reduce radon would also increase the sensitivity but prolong the measurement cycle. Such actions have the potential to improve the sensitivity of the stations by a factor of 10 and of the laboratories by a factor of 100 (Aalseth et al. 2009a). This would provide a significant potential to increase the capability of the particulate network stations, which already have an impressive sensitivity to detect barium-140 and lanthanum-140. Aalseth et al. (2009a) also show that the detection of some iodine isotopes are somewhat hampered by interference from radiation from beryllium-7. Werzi (2009) has estimated the global capability to detect lanthanum-140 at different source strengths, taking into account the sensitivity of the stations and using atmospheric modeling. His capability maps for different seasons show that lanthanum-140 releases corresponding to explosions in the range of 1 to 10 tons can be detected globally and that even smaller releases can be detected in areas around the globe at some distance away from the equator. This is quite a remarkable monitoring capability that makes it most unlikely that any nuclear explosion in the atmosphere will go undetected. Figure 3.10 shows the level of barium-140 observed in air samples during the period of atmospheric testing. This level has now decreased significantly, but the detection threshold of the IMS particulate network is still more than a factor of 10 below today’s background. Iodine-131 is one of the isotopes likely to be detected from an atmospheric explosion. It is also potentially detectable following underground nuclear explosions, as I-131 was observed in Sweden from some underground nuclear explosions conducted in the former Soviet Union and the United States (De Geer 2009a) (see Table 2.2). A study of I-131 observations

Fig. 3.10  Level of barium-140 observed in air samples during the period of atmospheric testing. Red line shows current IMS detection capabilty (Weiss 2009).

105

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

identified 172 such detections reported at 29 stations of the IMS particulate network from July 2005 to April 2009. Some of the observations likely came from nearby medical or nuclear facilities; other observations made in urban areas probably derived from other local sources. However, for detections at 11 remote stations there seemed to be no plausible explanation (Matthews 2009). Figure 3.11 illustrates that observations at stations in remote locations, in particular sites close to the sea, show unexplained observations of I-131. This demonstrates the need to understand what constitutes typical observations at the individual stations. The procedure to locate the source of a release of radionuclide particles follows the same procedure as with atmospheric modeling for noble gases described in Chapter 2. For a potential atmospheric source location, forward modeling of the transport of the fall-out can be performed to see if it arrives at the observing stations at a time consistent with the observations. The alternative possibility is to perform backtracking to estimate the area from which the fall-out could possibly come, given the pattern of arrival of the debris at observing stations. Both methods carry large uncertainties, but if the nuclear fall-out is detected at several stations, the uncertainties are reduced. To determine a more precise location, a state needs to use additional tools, such as satellite photos or specialized detection equipment that are part of its NTMs, as discussed below.

Fig. 3.11  IMS radionuclide particulate stations where iodine-131 was detected in the period July 2005 to April 2009. Numbers identify IMS stations as listed in the treaty (Matthews 2009).

106

Monitoring Atmospheric Nuclear Explosions

3.5.3  Monitoring Noble Gases The monitoring of xenon has been discussed in relation to underground nuclear explosions in Chapter 2. That chapter presented estimates of the global capability to detect xenon based on the assumption of a 10 percentage leakage from a 1 kiloton nuclear explosion (see Figure 2.25). An atmospheric explosion releases all of the instantaneously produced xenon, and all the parent iodine species, such that all the xenon will be in the atmosphere. The released xenon will be spread between the surface layer of air and traveling more slowly than the fast-moving jet stream. Thus a larger number of stations is likely to detect an atmospheric explosion compared to an underground explosion. 3.6  National Technical Means In addition to information provided by the TS, each state is free to use whatever information it finds useful to address an event or issue of concern. As discussed in Chapter 2, such information could be generally available data from scientific or commercial sources, such as optical or radar satellites. A number of high resolution optical satellites are operational and data from many of these satellites are available to any state. In addition, a smaller number of radar satellites are becoming operational, providing good resolution data even when the sky is cloud covered. Any atmospheric explosion close to the ground will create a large-scale and lasting imprint on the ground that can be observed from satellites. An explosion above water might not give any long-lasting evidence of the explosion itself, but the logistical arrangements surrounding the testing operation is very much exposed to satellite observation of open waters. In addition to using generally available data outside the IMS, a state may use different national assets as part of its NTMs. This might include normal intelligence tools such as information from human sources and from monitoring communications of different kinds. The United States has established a satellite-based system to monitor explosions in the atmosphere and in space that is reviewed below. Some other states might operate similar systems. Russia has an advanced space program, including Cosmos early warning satellite systems (Podvig et al. 2011), to detect missile launches. Such systems or other specialized nuclear explosion monitoring systems probably provide Russia a national capability to detect nuclear explosions in the atmosphere, although little information on possible Russian nuclear explosion monitoring systems is openly available. Likewise, little information is openly available from China, France, and the United Kingdom on national systems they might

107

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

possess to monitor atmospheric nuclear test explosions. The United States has, and other countries are likely to have, the capability to use aircraft to collect radioactive debris and gases in the air over international waters. This can dramatically enhance the amount of data on radionuclides because the planes can fly into the plume to observe the highest concentrations available with the least amount of decay. It is the quickest way to obtain the best possible picture of the radioactive fall-out. 3.6.1  U.S. Satellite-Based Systems to Monitor Atmospheric Nuclear Explosions In 1963, the same year the PTBT was signed, the United States launched its two first Vela satellites to monitor atmospheric nuclear explosions. The satellites, operated at an altitude of more than 100,000 kilometers, were equipped to detect gamma, x-ray, and neutron radiation. A second generation of Vela satellites also had two “bhangmeters” to detect the optical flash with a characteristic double pulse. The pulse, shown in Figure 3.12a with a logarithmic time scale, looks quite symmetric; in fact, it has a very steep beginning, with a rise time of the order of 200 micro-seconds. The later part of the pulse is more extended over time, with a total length exceeding one second. This double pulse is caused by the explosion process and is always present for low altitude explosions (Barasch 1979). Interestingly, the electromagnetic pulse (EMP) can also be a double pulse, where the second pulse is a ground reflection. can be seen in Figure 3.12b, this pulse is much shorter; the total length of the double pulse is about 10 micro-seconds. On 22 September 1979, a Vela satellite detected what appeared to be a nuclear explosion originating within a wide area in the South Atlantic Ocean. This event created a large debate in many countries: Was it really a nuclear explosion or an artifact that was detected? A broad search for supporting evidence of a nuclear explosion provided no clear results, and in May 1980 a U.S. Presidential Panel concluded that the observation was likely caused by a meteorite hitting the satellite and sunlight reflected off particles ejected by that collision. A number of experts disagree with that conclusion, claiming that the observation was from a nuclear explosion (Richelson 2006), and the origin of this observation is still unresolved. Monitoring equipment has been improved and upgraded over time, and several kinds of sensors to detect nuclear explosions in the atmosphere are now operating on several satellite systems, sometimes with dual use functions (NAS 2002a). Most of the monitoring equipment is now on board the 33 GPS satellites, each of which has a nuclear detection package of several detectors: optical, x-ray, EMP, gamma, and neutron radiation.

108

Monitoring Atmospheric Nuclear Explosions

Fig. 3.12a  Light flash signature of a 19 kiloton atmospheric nuclear test conducted at the Nevada Test Site on 1 May 1952. The two distinct light peaks, with a dimmer but still luminous minimum between them, are characteristic of the optical signatures of all atmospheric nuclear explosions below about a 30 kilometer altitude. A non-linear logarithmic time scale is used to display this curve so that details can be shown of the very fast first peak (Barasch 1979, in Hafemeister 2009).

Fig. 3.12b Electromagnetic pulse structure from an atmospheric nuclear explosion. The second pulse is a ground reflection (Hafemeister 2007).

Infrared sensors, to sense the heat from an explosion, are employed as part of the U.S. Defense Support Program (DSP). The prime purpose of this system is to detect missile launches with nuclear explosion detection as an additional task. The first DSP satellite was launched in 1970, and five satellites are currently in geosynchronous orbit at an altitude of about 35,000 kilometers; three are operational and two are backups (Defense Industry Daily 2009). The DSP systems will be replaced during the 21st century by the Space-Based Infrared System (SBIRS), which is comprised

109

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

of satellites in both high and low orbits for strategic and tactical missions (FAS 1998). Monitoring research work is continuing in the United States and a new kind of lightweight satellite, Fast On-Orbit Recording of Transient Events (FORTE) (Los Alamos National Laboratory 2011), was launched in 1997. It has an orbit altitude of about 800 kilometers and carries three broad-band receivers to measure radio frequency signals in the band 30-300 MHz. It also carries an optical imager that can detect and locate light flashes with a resolution of about 10 kilometers. FORTE also has an on-board computer system that can distinguish lightening from nuclear explosions. The system has a dual use application of measuring lightning and nuclear data. In May 2010 the United States launched improved instruments to detect atmospheric nuclear explosions. These instruments, developed by Sandia and Los Alamos National Laboratories, are on board the new IIF series of GPS satellites (NNSA 2010). This illustrates the U.S. ambition to maintain an up-to-date national capability to monitor nuclear explosions in the atmosphere.

110

Chapter 4

Monitoring Nuclear Explosions in the Oceans 4.1  Few Explosions Under Water The United States and the former Soviet Union are the only two countries that have carried out underwater nuclear explosions. They have conducted five explosions each: the United States at test sites in the Pacific Ocean and the former Soviet Union in the Arctic Ocean, close to Novaya Zemlya. The first underwater nuclear explosion, named Baker, was a test in shallow water conducted as early as 1 July 1946 at the Bikini atoll to study the effects of such nuclear explosions on naval ships, as seen in Figure 4.1. The last underwater nuclear explosion was carried out in 1962, so for some 50 years no nuclear explosions have been conducted underwater. This testifies to the lack of past interest in conducting nuclear explosions in the oceans, and it is difficult to find arguments for why any country would find this option interesting in the future, in particular given the exceptional ability to detect explosions in the oceans. The capability to detect such explosions is far greater than for explosions in any other medium. Underwater nuclear explosions have several detectable features: hydroacoustic and seismic signals generated by the explosion shock wave, and radionuclide noble gases and particulates generated by the explosion and its interaction with material close to the device.

O. Dahlman et al., Detect and Deter: Can Countries Verify the Nuclear Test Ban?, DOI 10.1007/978-94-007-1676-6_4, © Springer Science+Business Media B.V. 2011

111

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

Fig. 4.1  The Baker underwater nuclear explosion, conducted on 1 July 1946 at the Bikini atoll, tested the effects of a nuclear explosion on naval ships, seen in the background. The USS Saratoga aircraft carrier sank during this bomb test.

Some argue that attribution and not detection is the key concern in the oceans. It has been suggested that a country might set off a clandestine explosion under water in a remote part of an ocean assuming that the test will be detected but not attributed to the country responsible. However, satellite observations and other national technical means (NTMs) would make it possible for states to detect and track ships and aircraft that might be involved in such an operation. A similar scenario for conducting explosions in the atmosphere is considered in Chapter 3. 4.2  Oceans Are Vast and Vastly Transparent Since oceans cover 71 percent of the earth’s surface, how is it possible to confidently monitor such an enormous area? It is primarily possible because acoustic signals can be transmitted through the oceans essentially without any damping, which will be explained in more detail later in this

112

Monitoring Nuclear Explosions in the Oceans

chapter. Signals from an explosion of just a few kilograms trapped in a unique underwater waveguide—the sound fixing and ranging (SOFAR) channel—can be detected across an ocean. Oceans are transparent not only under water, but also on their surface. The surface of an ocean is wide open to satellite observations, as it is essentially free of the objects and activities that complicate satellite surveillance of land areas. Thus, to monitor ship movements or other activities that might be part of a nuclear test operation in an ocean is a fairly straightforward undertaking. Oceans are transparent also in a third way: most oceanic areas are defined as high seas, which are outside the jurisdiction of any state. This means that it is possible for states to have unfettered and prompt access to observations of an area of interest in great detail using aircraft, drones, or vessels, and to make close-in measurements of radionuclide gases and particulates in the air and water without crossing into another state’s jurisdiction. 4.3  Hydroacoustic Monitoring Hydroacoustic monitoring involves the detection of events or objects using acoustic signals in the water. The technique is most frequently used to detect submarines and as part of geophysical exploration. In these applications, measurements are carried out at fairly short distances from the object of interest. In geophysical exploration and in many submarine applications, active sound sources are used to illuminate the object and the response is recorded. Submarine detection is also carried out passively by listening to the noise generated by the submarine. As noted, such passive measurements are carried out with sensors located fairly close to the object. Extensive military installations have been established to track submarines across the oceans, such as the U.S. Navy Sound Surveillance System (SOSUS) in the Atlantic and Pacific Oceans (NOAA 2011a, FAS 2011). Hydrophone stations, such as the Japan Agency for Marine Earth Science and Technology (JAMSTEC) stations off the coast of Japan, are also operated by countries for scientific purposes (Sugioka et al. 2009). The Comprehensive Nuclear-Test-Ban Treaty (CTBT) application of hydroacoustic monitoring is quite different. It makes use of observations at great distances to monitor events in or close to oceans. Low frequency acoustic signals can travel for very long distances through the oceans in the SOFAR channel. This channel is a low velocity zone in which sound waves become trapped and can propagate without losing significant energy. The depth of the SOFAR channel varies in the different oceans depending on salinity, temperature, and depth of the ocean. The normal range of depth

113

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

of the channel is 600–1200 meters, with the range being deepest in subtropic regions and shallower at higher latitudes (NOAA 2011b). This pronounced low velocity zone is illustrated in Figure 4.2a, which presents acoutsic wave velocity as a function of depth. The figure also shows the lateral variations of the velocities in and close to the SOFAR channel across an east-west profile in the Pacific Ocean (illustrated in Figure 4.2b). In the Antarctic Ocean region, the SOFAR channel becomes more complex. Due to a weaker temperature variation with depth in that part of the ocean, the SOFAR channel tends to decrease in depth as it loses its pronounced structure. An “Antarctic channel” can be found at a depth of some 400 meters (Garner 1967). The propagation in the SOFAR channel is so efficient that explosive charges of one or a few kilograms detonated in the channel can be detected at a distance of 10,000 kilometers (Banister et al. 1993). Small explosions anywhere in an ocean can be detected at global distances even if only a fraction of the energy generated finds its way into the SOFAR channel. How this transfer of signals from outside the SOFAR channel takes place is still unknown, and in a keynote lecture at the ISS09 Conference, Ralph A. Stephen, professor at the Woods Hole Oceanographic Institution, noted, “We do not understand the physical mechanism that gets vertically propagating energy from below the seafloor into horizontally propagating energy in the ocean sound channel” (Stephen 2009). A summary of seismoacoustic applications in marine geology and geophysics is presented in Odom and Stephen (2004). 4.3.1  Eleven Stations Monitor the Oceans The hydroacoustic component of the CTBT monitoring system is a unique global system, comprised of 11 observational stations (Lawrence Fig. 4.2a This graph of sound speed versus depth shows the low velocity SOFAR channel at a depth of approximately 700 meters. The different colors correspond to different source and receiver locations, as indicated in Figure 4.2b (Stephen et al. 2009).

114

Monitoring Nuclear Explosions in the Oceans

Fig. 4.2b  Map of the Pacific Ocean locations between California and Hawaii where the sound velocity profiles were measured, and the corresponding depth profile (Stephen et al. 2009).

and Arranz 1998, Jans 2005, Hall 2000). Six of the stations measure the signal directly in the water with sensors deployed in the SOFAR channel. The remaining five stations are seismic stations on islands. Referred to as T-phase stations, these stations record seismic waves generated by the conversion of hydroacoustic signals to seismic signals at steeply sloping ocean-land boundaries. The 11 stations monitor the world’s oceans, with the exception of the Arctic Ocean and the Mediterranean Sea, which are well covered by the seismological component of the International Monitoring System (IMS). The most efficient way of detecting hydroacoustic signals is to take measurements in the SOFAR channel. Each ocean-based IMS hydrophone station has three hydrophone sensors in a triangular pattern, about two kilometers from each other, operating in the SOFAR channel. The three sensors make it possible for a station to provide not only the time of arrival but also the direction of an observed signal. The hydrophones are anchored to the ocean bottom and attached to a land station by cables. This means that the sensors can move somewhat relative to each other and to a reference point, which introduces some uncertainty both in the estimated direction and in the precise timing of observed signals, as will be discussed later. The hydrophones operate at frequencies of 1 to 100 Hz, a band suitable for detecting signals from events at large distances.

115

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

The hydroacoustic network is the most sensitive of the IMS networks and it is also the most vulnerable, with sensors placed in the harsh oceanic environment. To install sensors at a depth of 1 kilometer offshore in the ocean is a complex and expensive operation, presenting great logistical challenges. Operating and maintaining these stations has also proven difficult, and there are already examples of cables that have been broken by shipping activity. In February 2010 an earthquake off the coast of Chile generated a tsunami that wiped out the hydroacoustic station at Juan Fernández Island. The cost to rebuild the station is estimated at US$15 million and will take a year or more to accomplish, as specialized ships are needed to re-lay the cables. A key issue for the hydroacoustic network is increasing its resilience, perhaps even at the expense of the current very high detection capability. One approach discussed during the CTBT negotiations would be to support the sensor from a buoy connected to shore by wireless data transmission. Such a buoy is presented by Ginzkey et al. (2009). This buoy has only one sensor and cannot estimate the direction of the incoming signal, and the prime purpose of such a buoy system would be to serve as a temporary replacement if a cable-mounted station were to go out of operation for a long period. Given the operational experience, it could be useful to explore alternative and more robust ways of building a hydroacoustic station. The T-phase stations are specially tuned to record T-phases in a frequency band of 0.5 to 20 Hz, a bit higher than other IMS seismic stations. All five T-phase stations in the IMS network are placed on islands that slope steeply into the ocean. Hydroacoustic waves convert efficiently into seismic waves at such steeply sloped ocean-island boundaries. T-phase stations are, however, considerably less sensitive than hydrophone installations, as illustrated in Figure 4.3. On the other hand, T-phase stations are much simpler, cheaper to install and operate, and less vulnerable. Fig. 4.3  A comparison of velocity amplitude envelopes for a hydrophone and a T-phase station (Rodgers and Harben 1999).

116

Monitoring Nuclear Explosions in the Oceans

4.3.2 Detection There are two physical reasons why hydroacoustic observations are most effective in detecting underwater explosions. As mentioned, one is the SOFAR channel that provides hydroacoustic signal transmission across an ocean with little damping. The other reason is that a larger amount of the explosion energy is transformed into acoustic energy by underwater explosions than by similar explosions carried out underground. In other words, the acoustic coupling between an explosion and the surrounding media is higher in water than in bedrock. This also means that seismic signals are generated more efficiently by contained underwater explosions than by contained underground explosions. There are a number of examples illustrating the extreme capability to detect explosions in water using hydroacoustic observations. As part of a 2008 seismic survey, numerous explosions off the coast of Japan were recorded across the Pacific Ocean as far away as the station at Juan Fernández Island off the coast of Chile, a distance of 16,230 kilometers, as can be seen in Figure 4.4. The explosions had a yield of 40 kilograms and were detonated at a depth of 60 meters. Explosions with yields of 20, 40, and 400 kilograms from two other surveys off the Japanese coast in 1996 and 1998 were recorded at hydroacoustic stations in California and at Wake Island. Figure 4.5 shows

Fig. 4.4  Travel paths of hydroacoustic and seismic rays across the Pacific Ocean. A 40 kilogram explosion was detected at the hydroacoustic station at Juan Fernández Island (H03), Chile, at 16,230 kilometers from Japan (Hyvernaud 2009a).

117

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

that a large number of these explosions, with a yield as low as 20 kilograms, were recorded with a signal to noise ratio (SNR) of 10 dB and above. Figure 4.5 contains interesting observations from an explosion of 615 tons contained in a ship off the coast of New Jersey on 25 June 1970, referred to as the Chase 21 explosion. This surface explosion was observed with an SNR of 30 to 45 dB in the most favorable frequency band. This shows that surface explosions also generate fairly strong hydroacoustic signals, although less strong than those from fully contained underwater explosions. If it is assumed that an SNR of 3 dB is sufficient for detecting a signal, and a number of the earthquake observations in Figure 4.5 have even lower SNRs, then it should be possible to detect hydroacoustic signals from surface explosions of 1,000 kilograms or lower. As discussed in Chapter 3, seismic signals have been detected from atmospheric nuclear explosions at altitudes of several hundred meters over land. It is thus likely that explosions over an ocean will generate detectable hydroacoustic signals in a similar way. 4.3.3 Location As noted, because of the extremely efficient wave propagation in the SOFAR channel, oceans can be confidently monitored by a small number of stations down to very low explosion yields. What about the location accuracy? Hydrophones operate at higher frequencies than seismic stations, which normally improves the accuracy of determining the arrival time of a signal. The propagation speed is low, about 1.485 km/s, compared to seismic signals, and this reduces the consequences of timing errors. Fig. 4.5 Plot of frequency energy content for the Chase 21 explosion (dark green circles), 1996 and 1998 Japanese refraction experiment explosions (red, blue and green), and earthquakes reported by the PTS (triangles) recorded in Wake Island and Point Sur, California (Jepsen and Fisk 2009).

118

Monitoring Nuclear Explosions in the Oceans

Fig. 4.6 Observations from Australian hydroacoustic station Cape Leeuwin (HA01) show the effect of horizontal refractions on bearing deviation from true azimuth to sources of noise in the Indian and Southern Oceans (Li and Gavrilov 2009).

IMS hydroacoustic stations also can provide an estimate of the direction (bearing) of an arrival signal. These estimates can suffer from two kinds of uncertainties, however. One relates to the motion of the hydrophones relative to each other, as discussed above, and is of the order of 0.2 degrees. The other is due to the horizontal refraction of the signals and depends on inhomogeneities in the oceans. In most areas this uncertainty is on the order of 0.2 degrees, but it can be as large as 1 degree in certain areas (Li and Gavrilov 2009). Figure 4.6 shows the effect of horizontal refractions on bearing deviation from true azimuth as observed at the hydroacoustic station Cape Leeuwin in Australia. These numbers may seem small, but when considering global distances they are significant; bearing errors of 0.4–1 degree correspond to uncertainties in event location of the order of 100–200 kilometers at 10,000 kilometer distance. Chapter 2 pointed out that location accuracy increases when the observing stations are well distributed in azimuth around an event. The small number of hydroacoustic stations is likely to confine observations to a limited azimuth sector, and this will lead to reduced location accuracy. No comprehensive study of the achievable location accuracy throughout the oceans has been published thus far. The observed location uncertainty will be illustrated with a few examples. In the 2008 Japanese seismic survey discussed above, a large number of the explosions were observed by two IMS stations, on Juan Fernández and Wake Islands, and the estimated locations were reported in the Provisional Technical Secretariat (PTS) bulletin. Hyvernaud (2009a) has estimated the locations of these explosions with an estimated standard uncertainty of just 3 kilometers, using a larger number of hydroacoustic and seismic stations. This high accuracy illustrates what can be achieved when seismic and hydroacoustic data are combined and when stations are

119

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

well distributed around the source region. Figure 4.7 shows a comparison of these locations and those reported by the PTS, based on the two hydroacoustic IMS stations. The hydroacoustic locations are scattered over an area and not lined up as the reference locations, and they also have systematic uncertainty of some 500 kilometers. An event in November 2008 off the coast of Australia that was reported by the PTS had a large uncertainty ellipse with a maximum extent of some 500 kilometers (Prior et al. 2009). The event location was estimated from two hydroacoustic stations at long ranges and showed that event locations are sensitive to small azimuth deviations. In this case, a location based on hydroacoustic and seismological observations would have led to a much smaller uncertainty ellipse, of the order of 20 square kilometers. Even if the hydroacoustic network provides a high detection capability, the location accuracy achievable from the limited number of IMS hydroacoustic stations is in many cases poor because the small number of IMS hydroacoustic stations often provides limited azimuthal coverage around the observed events. To improve accuracy, a country can use seismological observations when available, as well as hydrophone installations established for other purposes, such as the SOSUS and systems established for scientific purposes, in addition to the IMS stations. Such systems covering large oceanic areas would significantly improve the location capability achievable by the state operating those systems.

Fig. 4.7  Map of best locations, red, from the Laboratoire de Geophysique (LDG) in French Polynesia, and IDC REB locations, yellow, of a Japanese seismic refraction profile in September 2008 (Hyvernaud 2009a).

120

Monitoring Nuclear Explosions in the Oceans

Location accuracy can also be improved by calibration events. This is most effective for seismological observations, as noted in Chapter 2, but of very limited use for infrasound observations, as noted in Chapter 3. Would calibration be useful for hydroacoustic observations? In acoustic wave propagation terms, oceans are somewhere in between the solid Earth and the atmosphere in that they are changing but not dramatically and not rapidly. As most oceanic areas are freely accessible to any country, calibration may be carried out after the observation of an event of interest. The procedure would involve dropping a depth charge as a calibration event at the estimated location of the event of interest and making a relative location of the two events. If needed, this procedure can be repeated until the calibration event is very close to the original event. This is a simple procedure that in principle is available to any country; in practice it requires that a country has the ability to operate in or above the oceans. 4.3.4  Multitude of Sources The IMS hydroacoustic network is a unique new global network and is somewhat similar to the infrasound network in that signals are now being observed from a multitude of sources. Sources such as shrimp, dolphin clicks, fish, and rain can be observed only at close distances from the sensors, whereas undersea earthquakes and volcanic eruptions can be observed at global distances, as seen in Figure 4.8.

Fig. 4.8  Illustration shows the noise level range of sounds that can be detected in the oceans (Coates, Seiche Ltd. 2009, in Bradley 2009).

121

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

Fig. 4.9  Spectrogram of whale vocalizations that can propagate over long distances (NOAA 2001, in Bradley 2009).

The signals very often show distinct and characteristic patterns, as illustrated by the spectrogram of whale calls in Figure 4.9. Hydroacoustic observations might be of great interest in studying many of those sources. A number of such studies were presented at the International Scientific Studies Conference in 2009 (ISS09), including hydroacoustic observations of whale songs, ice shelf dislocations and drifting icebergs, and underwater volcanic eruptions. Hydroacoustic observations might also be used in connection with tsunami warnings. Because oceans cover 71 percent of the Earth’s surface, they have a significant influence on global climate, and the variation of the temperature of the oceans is essential when assessing global warming. As the sound speed in water is dependent on temperature, among other things, it is possible to monitor the large-scale variations of the ocean temperature using hydroacoustic measurements over large distances. These scientific applications will be further considered in Chapter 6. Would this multitude of hydroacoustic sources make it difficult to observe signals from an underwater nuclear explosion and identify them? Figure 4.5 shows that explosions in the range of 20–400 kilograms can be detected and generally have higher SNRs than the observed earthquakes. The figure also shows that earthquakes generally have significantly lower SNRs in the higher frequency band of 32–64 Hz compared to underwater explosions with similar SNRs in the low frequency band of 3–6 Hz. This reflects that explosions generate higher frequencies than earthquakes, a well-known phenomenon from seismological observations, as discussed in Chapter 3. Jepsen and Fisk show in Figure 4.10 another illustration of the differences in signal spectra of different events: the Chase 21 surface chemical explosion, an underground nuclear explosion, and an earthquake. The same pattern can be observed: the spectra of the two explosions show much more high frequency content than that of the earthquake. The

122

Monitoring Nuclear Explosions in the Oceans

Fig. 4.10  Waveform and spectrogram of the Chase 21 explosion recorded at Ascension Island (left), a French underground nuclear explosion recorded on Wake Island in 1996 (middle), and an earthquake recorded at Cape Leeuwin, Australia (right) (Jepsen and Fisk 2009).

spectra of the surface explosions show a slow undulation of the spectra that might be due to water motions after the explosion. Explosions fully contained under water generate bubble pulses when the explosion cavity expands and contracts. The frequency of these pulses depends on the explosion depth and the yield of the explosion. In addition to these spectral differences, there is one more fundamental difference—the strength of the signal. All sources, with the exception of strong underwater volcanic eruptions, are extremely weak even compared to a low yield nuclear explosion detonated in an ocean. If a nuclear explosion were conducted under water, it would generate extremely strong hydroacoustic signals that could be easily detected and distinguished from all other sources by the sheer strength of the signals. 4.4  Seismic Monitoring of Oceanic Explosions Explosions in the oceans can be observed not only by hydroacoustic but also by seismic signals. Of the average 100 events determined daily by the International Data Center (IDC), most are based on seismic data and only a handful have been labeled as “hydro only.” The seismic capability to detect an underwater explosion is not as good as that based on hydroacoustic observations. Nevertheless, as mentioned, contained explosions under water generate stronger seismic signals than a corresponding explosion in hard rock. A 5 ton explosion conducted in the Dead Sea in 1999 as

123

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

part of a calibration experiment was detected at distances up to 5,000 kilometers and was assigned a magnitude of 3.9, much higher than expected from a corresponding explosion in hard rock. This stronger coupling for underwater explosions means that it should be possible to detect weaker explosions under water than under ground, given the same signal transmission properties and the same network of observing seismic stations. A large number of non-nuclear explosions have been conducted under water in different parts of the world, but no comprehensive study of the global explosion detection capability has been made. A few examples illustrate current detection capability. The largest of the explosions that resulted in the sinking of the Kursk submarine in the Barents Sea in August 2000 was estimated at about 1–2 tons by NORSAR, and it was observed at many seismological stations (see Figure 4.11) up to a distance of 5,000 kilometers. In the discussion of the global seismological detection thresholds in Chapter 2, it was noted that the detection capability of the IMS primary network was significantly lower, measured in magnitudes, in the oceanic areas in the Southern Hemisphere than elsewhere on the globe.

Fig. 4.11  Recordings of the main explosion that sank the Kursk submarine in the Barents Sea on 12 August 2000 from the seismic stations ARCES in northern Norway, at a distance of 500 kilometers, and more distant stations SPITS on Spitsbergen and NORES in southern Norway. A smaller explosion, about two minutes before the main explosion, was detected only on the closest station (clearly visible by applying a different scaling than used in this figure) (Orfeus 2000).

124

Monitoring Nuclear Explosions in the Oceans

As underwater explosions generate stronger signals and thus higher magnitudes than corresponding underground explosions, the difference in detection capability, measured in explosion yield between the Northern and Southern Hemispheres, might not be that large. The hydroacoustic network will, no doubt, detect event orders of magnitude smaller than those that could be detected by the seismological network, and for many events in the oceans the synergy between the two kinds of observations can be exploited. Seismological observations in particular will improve the location accuracy of events observed by both techniques, primarily by providing a better azimuth coverage around the events, as illustrated in the section on hydroacoustic location. 4.5  Radionuclide Observations of Underwater Explosions An underwater nuclear explosion has a number of detectable radionuclide features that are similar to those of an atmospheric explosion. If an explosion takes place close to the surface, the prompt radiation might be observable by the satellite-based special detection equipment described in Chapter 3. If an explosion takes place at a greater depth, the initial explosion phenomena cannot be directly observed. The radionuclide features of such an explosion would then be radioactive particles and noble gases. The amount and composition of the radionuclide particles generated depend also on the explosion device and on other material close to the explosion. The particles are ejected into the atmosphere with the water column, and a fair amount of the particles might fall back into the ocean with the water. The surface water around the explosion will thus contain particles that eventually will be spread and diluted by currents and wind. The amount of particles released into the atmosphere is in most cases smaller than for an atmospheric explosion. Very few, if any, particles are released from an explosion that is contained within the water. The radionuclide noble gases generated by an underwater explosion will be released into the atmosphere. For a shallow explosion this will happen rapidly; for a deeper and contained explosion it will take a longer time, but the amount might eventually be the same as for an atmospheric explosion, even if all processes are not yet fully understood. The noble gases and the released particles will be spreading around the hemisphere in which the explosion took place and will be detectable by the IMS radionuclide stations. Because most oceanic areas are international waters, a concerned state can fly over an area and take air samples or go there with ships to sample the water. Such observations can then be provided as national technical

125

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

means for possible consideration of the event by the CTBTO Executive Council. Not all countries have the resources needed to carry out operations in remote oceanic areas, and these countries may instead request an on-site inspection (OSI), in accordance with the CTBT. Such a request must be approved by the Executive Council following the procedures discussed in Chapter 5, except, in this instance, there may not be an “inspected state.” An OSI would most likely be limited to measurements of radioactivity in the air and water in the area around the event. One or several states might provide the Technical Secretariat of the CTBTO with logistical support in the form of ships and aircraft for such measurements. Another developing capability, nuclear forensics, deals with tracing nuclear materials, and experts are considering to what extent the debris from an explosion can be traced to the perpetrator, a much more difficult task (Ferguson 2006). Work to establish procedures and databases to conduct such analysis is being carried out at a number of national laboratories and in international cooperation among these laboratories. Such analyses go far beyond the standardized analysis that the 16 certified radionuclide laboratories are tasked to carry out on behalf of the CTBTO. 4.6  Satellite Observations This chapter has noted that oceans are wide open to satellite surveillance; however, it is unlikely that an explosion or its direct effects will be observed by optical or radar satellites, as their orbits and schedules are likely known to whoever wants to conduct an explosion. That said, satellite observations could nonetheless be helpful in identifying and tracking ships that might be part of a testing operation and thus assist in answering the question: who did it? Semi-automatic procedures for the detection of ships through the use of satellite photos have been developed to be applied in several contexts, including surveillance for pollution prevention, fishery control, and security and defense (Willhauck et al. 2005). It is likely that many state authorities monitor parts of the oceans for different purposes. A continuous surveillance of large oceanic areas would not be needed as part of CTBT monitoring. Satellite observations would rather be used to detect and track ship movements once an event of concern has taken place. This means that a country could analyze satellite images over the area where an event took place for a time interval prior to the event to identify ships that might have been involved with test preparation. If such ships are detected, their motions both before and after the event can be tracked. Such backtracking would reveal where a ship is coming from, and

126

Monitoring Nuclear Explosions in the Oceans

tracking the ship after the event might make it possible to locate, identify, and even intercept it. A satellite photo of ship tracks can be seen in Figure 4.12.

Fig. 4.12 Ship tracks south of Alaska show the clouds formed by exhaust from ships (NASA 2009).

127

Chapter 5

On-Site Inspections The on-site inspection (OSI) regime is the ultimate verification measure of the Comprehensive Nuclear-Test-Ban Treaty (CTBT), comprising strong political, technical, and operational elements. It adds to the deterrence against clandestine tests and is an important measure to determine compliance. A state can request an OSI of an event of concern either following a process of consultation and clarification or without going through such a process. Either way, time is of the essence in the decisionmaking and conduct of an OSI. While some effects from a nuclear explosion may be long lasting, others can be short lived and thus time sensitive. This requires that an OSI inspection team and its equipment can be deployed and initiate inspection activities in a timely manner within the six-day time frame specified in the CTBT. Fig. 5.1  The inspection team has the right to conduct overflights to obtain a general orientation of the inspection area, optimize locations for ground-based inspection, and make certain measurements.

O. Dahlman et al., Detect and Deter: Can Countries Verify the Nuclear Test Ban?, DOI 10.1007/978-94-007-1676-6_5, © Springer Science+Business Media B.V. 2011

129

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

This chapter presents the political frame of the OSI regime as established in the treaty, considers the conduct of an OSI, and stresses the importance of an overall well-organized approach and the need for an inspection team with expert knowledge and experience in the different technologies involved. The chapter also reviews and analyzes the different technologies that can be used during an OSI (Figure 5.1) and discusses a few OSI scenarios. 5.1  The Political Frame More than one third of the CTBT text and protocol is devoted to OSI. A summary of the main points of the text is provided to facilitate the following analysis and discussion. Quotes are from the treaty text. The treaty defines the purpose of an OSI and also the procedures to be followed to agree on and carry out an inspection. The treaty states that “the sole purpose of an on-site inspection shall be to clarify whether a nuclear test explosion or any other nuclear explosion has been carried out in violation of Article I and, to the extent possible, to gather any facts which might assist in identifying any possible violator.” A state party will be able to request an OSI based on information collected by the International Monitoring System (IMS) or information obtained by national technical means of verification in a manner consistent with generally recognized principles of international law, or a combination of the two. National technical means are used in this book as a broad concept involving any information relevant to CTBT verification in addition to the IMS data, including data from seismological and radionuclide stations outside the IMS, satellite observations, and technical and other intelligence data available to a state. A request for an OSI is to be presented to the executive body of the CTBTO, the Executive Council, as well as to the Director-General of the Technical Secretariat (TS). It is understood that an OSI is triggered by an event for which a state sees the need for further clarification. It shall be carried out in the area where the triggering event occurred and the inspection area shall be continuous and not exceed 1,000 square kilometers, with no linear distance greater than 50 kilometers. The state requesting the OSI shall, inter alia, provide information about the estimated location of the triggering event, with an indication of the uncertainty of that location and the proposed boundaries of the area to be inspected. An on-site inspection must balance the needs of gathering enough information for the task at hand while avoiding the disclosure of information not related to the purpose of the OSI that the inspected state has the right to protect. The inspected state has the right to take measures

130

On-Site Inspections

to protect sensitive installations and locations and to prevent the disclosure of confidential information not related to the purpose of the inspection. It also has the right to declare restricted-access sites, where an inspection can take place only after negotiations between the inspection team and the inspected state, referred to as “managed access.” Each restricted-access site shall not exceed four square kilometers, and the inspected state has the right to declare up to 50 square kilometers of restricted-access sites. If more than one such restricted-access site is declared, each such site must be separated from any other such site by a small alleyway with a width of no less than 20 meters where access is permitted. A request for an OSI would trigger a number of activities. The treaty provides that the request be communicated to all states parties and that the Technical Secretariat begin preparations for an OSI immediately. The Director-General will seek clarification from the state on whose territory the OSI has been requested to try to “resolve the concern raised in the request.” Unless the requesting state considers its concern resolved and withdraws its request, the Executive Council must make a decision on the OSI request within 96 hours of receipt. The decision to approve a request of an OSI requires 30 affirmative votes among the 51 Executive Council members. If the request for an OSI is approved, the inspection team must arrive at the point of entry in the state to be inspected no later than six days after the request was received. The inspection team must submit a progress report no later than 25 days after the approval of the OSI, and, based on the consideration of that report, a majority of the Executive Council members may decide to discontinue the inspection. If the Executive Council does not make such a decision, the inspection will continue for up to 60 days. If the inspection team sees the need for more time to fulfill its mandate, it may request an extension of an additional maximum of 70 days, to be approved by a majority of the members of the Executive Council. The inspection team may submit a recommendation to terminate the inspection, and such a recommendation shall be considered approved unless two thirds of the members of the Executive Council vote against it. The inspection team also may submit a request to conduct drilling, which would need to be approved by a majority of the members of the Executive Council. An inspection report, containing descriptions of the team’s activities and findings and an assessment of the cooperation granted during the inspection, among other things, will be made available to all states parties and to the Executive Council. The Executive Council shall review the report and address whether any non-compliance with the treaty has occurred. The Executive Council will make a decision on the inspection report by a two thirds majority. Article V of the treaty provides for an escalating

131

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

Fig. 5.2 During an OSI field exercise soil samples are taken for further analysis in a laboratory.

series of measures to redress a situation of possible non-compliance: the Conference of the States Parties may restrict or suspend the rights of a state party; it may recommend to states parties collective measures in conformity with international law; and it or the Executive Council may bring an issue regarding non-compliance to the United Nations. An inspection team will consist of inspectors and inspection assistants coming from the states parties and from the staff of the Technical Secretariat (Figure 5.2). There is a cap on the number of persons on an inspection team: “The total number of members of the inspection team present on the territory of the inspected State Party at any given time, except during the conduct of drilling, shall not exceed 40 persons.” The treaty calls for procedures for the conduct of an on-site inspection to be specified in an operational manual for on-site inspections, to be approved at the initial session of the Conference of the States Parties. A list of equipment for use during OSIs shall also be prepared for approval at that Conference. The development of the operational manual and the technologies and equipment to be used during an OSI are discussed later in this chapter. 5.2  Reflections on the Frame A great deal of experience has been accumulated on inspecting sites and facilities under various arrangements, including International Atomic Energy Agency (IAEA) safeguard agreements, the Chemical Weapons Convention (CWC), the bilateral agreements between the Soviet Union (or Russian Federation) and the United States, and the United Nations Special Commission (UNSCOM) in Iraq, among others. The CTBT negotiators could draw on these experiences when they elaborated the provisions

132

On-Site Inspections

Fig. 5.3  OSI field camp set up for the 2008 Integrated Field Exercise in Kazakhstan.

on OSI. Nevertheless, the CTBT text on OSI contains unique provisions on which there was at the time of the negotiations—and still is—limited technical background. The United States, which possesses a great deal of experience in observing effects of nuclear explosions at close distances, did not participate in CTBTO Preparatory Commission (PrepCom) activities on OSI for eight years, but it resumed sending experts in the subject in 2009. The decision-making process in the Executive Council was also formulated without much experience on how such a process would play out. Concern has been expressed that it might be difficult to obtain the 30 affirmative votes needed for a decision to conduct an OSI. In contrast, a challenge procedure in the Chemical Weapons Convention, so far never used, will take place unless a three quarters majority of its Executive Council votes against it within 12 hours of receipt of the request. The way issues may be addressed in the Executive Council merits further consideration below. The Executive Council will be elected by the Conference of the States Parties. The treaty establishes how seats in the Executive Council will be distributed geographically. States will be elected to serve on the Executive Council on a regional basis. Membership in the CTBTO Executive Council is based in part on selective criteria that will guide the choice within each of the six regional groups towards certain key countries. These criteria include: political and security interests, nuclear capabilities relevant to the treaty, as well as the number of monitoring facilities in the IMS, expertise and experience in monitoring technology, and contribution to the annual budget of the CTBTO. Such criteria basically ensure that the five nuclear

133

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

weapon states will always have a seat on the Executive Council (U.S. Senate 1999). Table 5.1 compares the geographical distribution in the Executive Council of the CTBTO with the geographical distribution of countries in the IAEA Board of Governors for 2010. There is a considerable difference in the composition. The CTBTO Executive Council reflects more the number of countries in each regional group, whereas the Board of Governors reflects more the economic strength of the different regions and their engagement in nuclear issues. Table 5.1  Geographical Distribution of Seats in the CTBTO Executive Council and the IAEA Board of Governors Members CTBTO Africa Eastern Europe Latin America and the Caribbean Middle East and South Asia North America and Western Europe South East Asia, the Pacific, and the Far East

number 10 7 9 7 10 8

% 19 14 18 14 19 16

Members (2010) IAEA number 5 4 6 3 10 7

% 14 11 17 9 29 20

The discussion in the Executive Council could be a purely political exercise, or the members could carefully consider and assess the technical material presented in support of the request for an OSI. To avoid the discussion in the Executive Council becoming a purely political exercise, countries will need to build the capacity to conduct technical analysis and integrate it into the political processes of the treaty. Such widespread knowledge and ability among the Executive Council members would increase the possibility that a decision regarding whether an OSI should be conducted would be based on an analysis of the technical evidence presented in the request. Countries must also develop their capabilities to analyze the results of an OSI as reported by the inspection team. This is essential if the Council is to make sensible decisions on the continuation of an OSI beyond the initial period and the possible conduct of drilling, and in regard to when to cease inspection activities. Chapter 7 suggests establishing such expertise through regional cooperation. It should be in the interest of all states parties to promote and support such cooperation to provide all members of the Executive Council access to the needed expertise, thereby increasing the possibility of a factual approach to considerations in the Council.

134

On-Site Inspections

Given that the Executive Council will make a thorough assessment of the technical evidence presented in support of an OSI, how would its members judge the credibility of the information presented? Data and material obtained from the IMS, which is a common source of authenticated information, will have a high degree of credibility. As discussed in Chapter 2, the IMS stations are certified to fulfill agreed-upon specifications and the data must be authenticated. Data from other available seismological stations analyzed by individual states or regional centers will also carry high scientific credibility because anyone can independently replicate the analysis. In addition, satellite observations are generally available. The association of satellite observations with the location of a triggering event, most likely obtained from seismological observations, will also be highly credible, because the analysis can be replicated. To validate such technical material it is essential that members of the Executive Council have access to the expertise needed to make an independent analysis. It might be more difficult to convince critical Executive Council members of the credibility of information obtained by other national technical means, such as signal or other intelligence. Because an OSI can only be carried out once the CTBT has entered into force, it is difficult to foresee how the discussion and the decision making in the Executive Council may develop. It might be useful for states to consider possible scenarios in order to gain some insight into this important process. If the Executive Council rejects a request for an OSI, what then? An OSI will be an important tool to verify compliance with the CTBT, but further steps could be taken without it. The likelihood of convincing other states parties that an event in non-compliance with the treaty has taken place, without conducting an OSI, would depend on the political situation and the technical evidence that could be presented. If observed radioactive material characteristic of a nuclear explosion could be credibly associated with the event, such evidence might obviate the need for an OSI. The two explosions carried out by the Democratic People’s Republic of Korea (DPRK) may illustrate this situation, keeping in mind that an OSI can be conducted only after entry into force. Xenon was observed from the first explosion but not from the second, raising questions from a few about whether the second explosion was a nuclear or a chemical explosion. The second explosion was well located from seismological and satellite observations; an OSI, which would have included radionuclide measurements in the source region, could well have provided additional evidence. A country could continue to make its case regarding an event of concern even if the Executive Council rejected its request for an OSI. One possible recourse would be to request a special session of the Conference of the States Parties; such a request must be supported by a majority of the states parties. Another option is for a state to directly refer the matter to the

135

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

United Nations Security Council. States could also individually or jointly take actions they saw as appropriate to address the situation in conformity with international law. 5.3  Conduct of an OSI An OSI is a technical operation carried out in a political environment. This may make the operation complex, depending on the reaction, cooperation, or actual level of mandatory support of the inspected state. It is essential that an OSI contribute to the clarification of an event and build confidence in the understanding of a situation. A challenge of any OSI is to balance the needs of gathering enough information for the task while avoiding the disclosure of information not related to the purpose of the OSI that the inspected state has the right to keep protected. If a team were to find convincing evidence of a nuclear explosion in violation of the treaty its task would be fulfilled. What if it found no such evidence? We all know the difficulties of proving that a particular activity has not occurred. So one may ask: If not an explosion, what was it? If the triggering event was seismic, most likely it was an earthquake, but this may not be so easy to prove, as discussed below. Since the CTBTO PrepCom was established in 1997, countries have spent much time and work on developing an operational manual to guide the procedures for inspection and the interaction between the inspection team and the inspected country. To increase the likelihood of confidently clarifying an event, the conduct of an OSI should be based on a well-structured plan of applying the different OSI technologies to look for features of an underground explosion that could help clarify the event. Such a systemic approach would require a good understanding of the observable features of an underground nuclear explosion and of the way the different technologies should be applied. A systemic approach also would need a professional team that could fully utilize the capability of the technologies and equipment available for an OSI and be able to integrate and exploit the information in an efficient way. Lastly, this type of approach also would require conducting the logistical operation of deploying and sustaining the inspection team and its equipment in a prompt, smooth, safe, and secure manner (Figure 5.4). 5.3.1  Operational Manual The operational manual is to contain procedures guiding the conduct of an OSI and the reporting of the results. The manual will address a number of specific issues, such as preparation, coordination, support, and

136

On-Site Inspections

Fig. 5.4  Experimental mobile xenon sampling unit for field deployment in an OSI (Popov et al. 2009).

planning, in addition to containing instructions on the application of the different OSI technologies and on how to report the results. The manual will also address health, safety, and confidentiality issues. The drafting of an operational manual for on-site inspections has deeply engaged many delegations at the CTBTO PrepCom since 1997, and the manual is still not finalized. Its elaboration turned out to be more sensitive than expected, and the process has become overburdened with the volume of information that delegations want to include. A constructive way forward was found by compiling a “test manual,” which was used during the large-scale Integrated Field Exercise (IFE08) in Kazakhstan in October 2008 (CTBT/WGB/TL-18/30 2006). This test manual was based on the rolling text of the operational manual that had been elaborated over the years in the PrepCom. It contains fairly detailed guidance on how to proceed with the different on-the-ground activities during an OSI, as well as guidance on issues such as “health and safety,” “confidentiality,” and “reporting.” In particular, it addresses issues related to the interaction between the inspection team and the inspected state. The test manual provides a detailed timeline for the different activities associated with an OSI (Figure 5.5). Given the experience from the use of the test manual during the IFE08 and with the political will of all parties concerned, it should be possible to finalize a manual to be presented to the initial Conference of the States Parties. In the further work on the manual it might be useful for states to consider the level of details in the document. How much detailed guidance is necessary or desirable? What key issues should be in an operational manual and what might be left for more technical instructions?

137

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

Fig. 5.5  Timelines of an OSI as described in the CTBT, from the test manual used during the Integrated Field Exercise in 2008 (CTBT/WGB/TL-18/30 2008). Timing is essential in the conduct of an OSI, including in decision making, deployment of equipment and inspectors to the site, and the field work itself.

5.3.2  Systemic Approach What is a systemic approach? A number of different technologies can be applied during an OSI. To provide for an efficient inspection, it is important to understand the way different technologies can contribute individually and in synergy with others. A well-staged, or systemic, approach to an OSI essentially means applying the different technologies in an optimal way. Operational analysis is a valuable tool to analyze and compare alternative ways of conducting complex operations, both military and civilian (Wei and Jing 2008). Such analysis could support a systemic approach by facilitating not only the initial planning of an OSI, but also the necessary adjustments to the plan of an ongoing inspection, taking into account the observations made. A systemic approach could prove useful to conclude the remaining work in the PrepCom to establish a functioning OSI regime. Such an approach would prioritize among the different technologies and the different steps to be taken, starting by developing the most effective technologies and the support needed for these technologies (Zucca et al. 2009). One thousand square kilometers is a large area to search. To find evidence of a nuclear explosion, it is essential to proceed in a structured way, taking into account the local infrastructure. To carry out a nuclear test explosion, one needs to access the site of the test with heavy equipment, which in most situations will restrict the number and size of sites where a test could have been conducted. Mathematical methods might help reduce the search area by fusing different kinds of information coming from measurements and expert judgment (Figure 5.6). Above all, a precise location of the triggering event and a good estimate of the uncertainty of

138

On-Site Inspections

Fig. 5.6  To reduce the area of search in an OSI, such criteria as land use (active mining) and demographics (cities, farms, factories) can be used. In this example, the search area is reduced from 225 km2 to less than 1 km2, with green depicting low probability of finding the surface ground zero and red indicating a higher probability (Carrigan and Johannesson 2009).

139

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

that location are essential to direct the inspection. The protocol to the CTBT defines the technologies that can be used during an OSI and also defines when during an inspection a technique can be used. Throughout an inspection a team could use position-finding equipment to define the boundaries of the inspection area. The team also could carry out ground- and air-based visual inspections, make radioactive measurements, take environmental samples, and conduct local seismic monitoring. During a continuation period of an OSI that will follow the initial period, subject to approval of the Executive Council, the team also will be able to use active seismic methods and several geophysical exploration techniques, such as detecting magnetic or gravity anomalies. Drilling, which requires a well-defined target, must be approved by the Executive Council. These individual techniques will be further discussed below. The technologies can be evaluated from different perspectives. Hawkins et al. (2009) suggest several evaluation criteria: relevance; showing the diagnostic value of a technology; sensitivity and confidence indicating the ability to detect and document signals with high confidence; and, finally, equipment and personnel factors indicating how easily the instrument can be deployed and operated and what the requirements are on the personnel operating it. The usefulness of a particular technology is dependent on the environment in which an OSI takes place. Han et al. (2009) have made a comprehensive analysis of the usefulness of different OSI technologies, identifying their possible contributions and limitations. This will be discussed in greater detail under the consideration of individual technologies. During an OSI a great deal of information regarding the geography, visual inspection, and technical observations on a large and small scale is collected, analyzed, and combined, and conclusions must be drawn within a short period of time. New methods to analyze and exploit data— data mining, which is discussed in more detail in Chapter 2—could be used to interpret this great variety of information and significantly increase the efficiency of an on-site inspection, and there is an urgent need to explore how such methods can be applied to an OSI and to develop necessary procedures to include such a tool. 5.3.3  Inspection Team An initial list of inspectors and inspection assistants should be compiled no later than 60 days after entry into force of the treaty based on proposals from states parties and the Director-General of the Technical Secretariat. A state party has the right to object to an inspector or inspection assistant proposed in the initial list, and that person will then not be designated

140

On-Site Inspections

for inspection tasks in that state. During the PrepCom phase, 60 people to date have completed a full training cycle, and some 500 individuals from states signatories have participated in OSI activities and are on a roster of experts to participate in further training courses and training exercises. Many of the people so far engaged in OSI training activities have a general background but are not experts in any of the OSI technologies. Given the complexity of most of the OSI technologies, an inspection team should be manned by experts working in the relevant fields with professional knowledge to conduct observations and analyze and interpret the results. Such experts in the different technologies must be the backbone of an inspection team. In addition, a team needs a team leader who is a skilled negotiator, and he or she might need one or two persons assisting in relations with the inspected state and with the Technical Secretariat. The technical work should be directed by an operational manager who also serves as the deputy team leader, assisted by a small group of people experienced in the integration and interpretation of information, who will create a strategy as well as a plan for taking measurements. The experts will also assist in combining information from different technologies and in making joint interpretations. A possible organization of an inspection team is provided in Figure 5.7. To adopt such an expert approach to the composition of the inspection team would be an important element in preparing for a functioning OSI regime at entry into force. It would put new requirements on states Inspection Team

Inspection Team Leader

Administrative, Legal and Diplomatic Support, Interpretation

Support Activities

Inspection Activities

Head of Support

Health and Safety

Communications

Head of Operations

Logistics

Technical Support

Radionuclide Measurements

Planning and Integration of Results

Passive Seismological Monitoring

Expert Subteams on:

Visual Observations

Active Seismic Survey

Other Geophysical Techniques

Field Support

Fig. 5.7  Diagram of a possible organization of an OSI inspection team.

141

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

to designate suitably qualified inspectors and inspector assistants with expert knowledge. In addition, it would simplify the training, as experts in the field will need training only on the specific application of detecting evidence of a nuclear explosion. An operational manager and his subteam should already have expertise in operational management, data integration, and exploitation and operational analysis; however, they would need training on OSI-specific applications. The team leader and his assistants will need training in the political and procedural perspectives of an OSI, and all team members will need general briefings and training on such specific OSI conditions as confidentiality, health hazards, safety, and security. 5.3.4 Logistics From an arms control perspective, an OSI is a large-scale operation, but compared to military, rescue, and relief operations, as well as commercial geophysical exploration surveys, an OSI is a fairly limited logistical operation. Its main challenge is that it is carried out in a political environment. The IFE08 carried out by the Provisional Technical Secretariat (PTS) in Kazakhstan in 2008 was a valuable test that focused the interest of the CTBTO PrepCom on OSI issues (CTBT/PTS/INF.1021 2009). It provided a good opportunity to test the deployment of an OSI team and equipment for a number of technologies. A large number of lessons were learned on different perspectives of an OSI (CTBT/PTS/INF.999 2009). However, the IFE08 also was an illustration of the need to find a more integrated and efficient logistics concept. The equipment—a total of 51 tons—packed and prepared for transport to the field, is shown in Figure 5.8. Sending hundreds of individually packaged items that need to be unpacked and assembled and made to work together in the field is not the way a field operation should be conducted. The logistics concept needs to be a container-based system of modules in which the equipment is already assembled, connected, and ready to work. This would enable the team to start observing and measuring immediately upon arrival at the deployment site, and it would save valuable time that could be crucial for some of the OSI technologies. The modules should be transportable to the deployment site by aircraft, helicopters, and trucks. There is a wealth of experience involving mobile laboratories for other applications that could serve as models when developing a professional logistics concept for an OSI.

142

On-Site Inspections

Fig. 5.8  OSI equipment was packed and placed in storage before being shipped from Vienna to Kazakhstan for the Integrated Field Exercise in 2008 (Anderson and Gaya-Pique 2009).

5.4  OSI Technologies During the PrepCom period the OSI technologies have been addressed in a number of theoretical and practical workshops and field tests. This work has been conducted mainly within a small community of experts; the vast knowledge and operational experience available in the civilian community and within the different sciences have been included only to a limited extent. This was clearly demonstrated at the International Scientific Studies Conference in 2009 (ISS09), during which some 40 posters related to OSI technologies and procedures were presented. This valuable material forms part of the basis for the discussion in this section of the different OSI technologies to be applied to underground events. 5.4.1  Visual Observations Since the CTBT was opened for signature in 1996, there has been a dramatic development in overhead satellite observations. As discussed in Chapter 2, satellite photos with resolution as high as 1 meter or better are readily available to countries. The satellites also provide good coverage of the Earth, with several observations per day of any particular location. Radar satellites with resolution in the 10 meter range are also available that provide all-weather capability. Radar satellites used in the interferometric mode can also detect small changes in the ground level that might occur as a result of a nuclear test, as discussed in Chapter 2 (see Figure 2.27). Computer software is also generally available that can identify changes in a particular scene from pictures taken at different times. This makes it

143

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

possible to follow work progressing on the ground, for example, on the infrastructure needed to conduct an underground nuclear test. Making detailed analysis of satellite observations of large areas to look for features that might be related to the preparation of a nuclear explosion is time consuming and costly. If the analysis is carried out once an event of particular interest has been observed, however, it can be limited to a small area. The area to be analyzed depends on the accuracy of the estimated location of the triggering event. As a triggering event is most likely detected and located using seismological data, it is important to improve the accuracy and minimize the uncertainty in seismic event location. As discussed, it is likely that once a state finds an event that needs further clarification, it will consult satellite photos to judge if it is logistically possible to carry out an underground nuclear explosion in that area. From satellite recordings it is also possible to observe and analyze activities in the area prior to an event to identify possible changes in the local infrastructure. Chapter 7 discusses how regional cooperation might facilitate such analysis. Satellite observations are of great importance to an inspection team when planning an inspection. With the location of the triggering event and the corresponding uncertainty as to a starting point, satellite observations make it possible to identify which parts of the area around a triggering event might be the most likely site for an explosion when taking terrain and local infrastructure into consideration. Such considerations would be a basis for the planning of the different activities during an OSI. Coxhead (2009) has used satellite photos to illustrate activities related to the conduct of a 12.5 ton triggering event for an OSI exercise in Kazakhstan in 2002 (Figure 5.9). To conduct an underground nuclear explosion is an extensive operation. It requires heavy equipment as part of the preparation for drilling or for digging a tunnel. To bring such equipment to a test site requires either a road or an infrastructure for airlifting. The preparatory work is carried out for a substantial period of time, and it is very difficult to hide these activities from a careful analysis of satellite observations of a selected and limited area. Photos and radar observations from satellites are important both for states in the process leading up to an inspection and for an inspection team planning an inspection. Overhead observations from aircraft and helicopters are significant initial activities of an OSI that provide a more detailed picture in support of the planning. Overhead observations may also reveal traces from preparation of or leftovers from a test. It might also be possible to see lasting minor effects of the shockwave on the ground, such as changed texture, cracks, or other features.

144

On-Site Inspections

Fig. 5.9 Satellite photos taken one year before (left) and after (right) preparations were completed for a triggering event for the OSI exercise in Kazakhstan in 2002. Roads had been added in the interim, which could have indicated to inspectors that preparation had been made for a nuclear test, if this had been a real OSI (Coxhead 2009).

It would be very hard to hide all traces of the conduct of an underground explosion in an area where no previous human activities have occurred given that inspectors find the area where the explosion took place. A possible scenario is to conduct a clandestine test in an area where heavy and large-scale activities, such as mining, occur or have occurred. However, even in that environment it would be difficult to hide a nuclear test from close-in visual inspections and radionuclide measurements. 5.4.2  Radionuclide Techniques Radionuclide measurements are a key component in an OSI. If certain characteristic radionuclides, either gases or particles, are detected, it is likely that an event will be identified as a nuclear explosion. Schoengold et al. (1996) provide a comprehensive summary of radionuclide observations of venting from U.S. tests from 1961 to 1992. These observations are summarized in Table 5.2, listing the top 15 aerosol radionuclides ranked by frequency of detection from a total of 80 tests. Of these top 15 ranked radionuclides, only six have a half-life long enough to be relevant to an OSI. The decay of those isotopes reflecting the likelihood of detection is shown in Figure 5.10.

145

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

Table 5.2   Top 15 Aerosol Radionuclides Ranked by Frequency of Detection (Ely et al. 2009)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

Isotope

Frequency

Half life-times

I-133 I-131 I-135 Cs-138 Ba-140 Rb-88 Ru-103 Te-132 I-132 Sr-91 I-134 Zr-95 Ru-106 Ba-139 Ce-141

63 62 62 27 19 15 14 13 11 8 8 6 5 5 5

20.3 hours 8.04 days 6.61 hours 32.2 minutes 12.8 days 17.8 minutes 39.4 days 77.9 hours 143 minutes 9.7 hours 52.6 minutes 64.4 days 368 days 84.9 minutes 32.4 days

Initial consideration > 1week*

> 1week > 1week

> 1week > 1week > 1week

*  >1 week identifies isotopes of interest in an OSI.

A nuclear explosion produces a large amount of different radionuclide isotopes that initially are in the immediate vicinity of the explosion in the cavity. The amount of radioactivity that escapes and is detectable at the surface depends on the level of containment. Radioactive material can escape from the cavity by three different mechanisms having different time scales. If the explosion is poorly contained, the over-pressure in the cavity might lead to an almost immediate release, not only of radioactive noble gases but also of particulates. Such releases can be big enough to Fig. 5.10 The plot provides an overall estimate of the likelihood of observing each of the indicated radionuclides as a function of time. For times less than 60 days, iodium-131 and barium-140 will be prominent, with other isotopes becoming more optimal at later times (Ely et al. 2009).

146

On-Site Inspections

Fig. 5.11  Plot of the surface pressure against time. After several large barometric depressions (low pressure) during the fall of 1993 sulfur hexafluoride was detected in concentrations of 280–600 ppt at the surface near the Non-Proliferation Experiment (NPE) only 50 days after detonation, and helium-3 breakthrough was detected after 375 days (Carrigan et al. 2009).

be detected at great distances, as discussed in Chapter 2. Another form of release, still driven by the cavity over-pressure, is seepage through fractures or cracks or because of the porosity of the rock. Seepage is normally a slower process that involves only radioactive gases. The third process by which radioactive gases can leave the cavity even in well-contained explosions is through barometric, or atmospheric, pumping, where atmospheric pressure variations drive the transport. This transport mechanism is thousands of times more effective than gaseous diffusion. Atmospheric pumping was studied in the Non-Proliferation Experiment (NPE) conducted at the Nevada Test Site (NTS) in 1993, with a 1 kiloton chemical explosion detonated 400 meters below the surface releasing tracer gases. The first detection of tracer gases was made 50 days after the explosion, and further detections were made as late as 375 days after the explosion (Figure 5.11). Models that were calibrated by the NPE for the release of Xe133 and Ar-37 through atmospheric pumping—and taking into account the detection limits for these nuclides—show a detection window from 50 to 135 days for Xe-133 and from 80 to 215 days for Ar-37 (Figure 5.12). In 1989, a Russian team was able to measure both Xe-133 and Xe-131m at the Semipalatinsk test site. The measurements were taken in 6–12 meter deep boreholes more than 180 days after and more than100 meters away from an underground nuclear explosion. The measured concentrations were as high as 0.8 and 2.6 Bq/cubic meter for Xe-133 and 5.2 and 16 Bq/cubic meter for Xe131m (Dubasov 2003, in McIntyre et al. 2009). These concentrations should be compared with detection limits 1,000 times lower that have been obtained for newly developed systems (Popov et al. 2009), and with the significantly lower background concentrations discussed below.

147

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

Fig. 5.12 Models predict that Xe-133 would arrive 50 days after a 1 kiloton underground nuclear explosion; Ar-37 is detectable after 80 days. The window for detecting Xe-133 is 85 days long and that of detecting Ar-37 is 135 days long (Carrigan et al. 2009).

Radionuclides can also be produced when neutron radiation from a nuclear explosion activates elements in the ground and water. The resulting activation depends on the possible shielding of neutrons from the explosion and on the material available in the water and in the bedrock. Friese et al. (2009) report on a groundwater scenario in which 70 percent of the activation products is Na-22 and 13 percent is Ca-41. As calcium and sodium are common in groundwater, they are likely to interact with neutrons from an underground nuclear explosion. Neutrons interacting with calcium in the soil also generate Ar-37, a noble gas with a half-life of 35 days, which fits nicely within the time frame of an OSI. The natural background of Ar-37 generated by spontaneous fission

Fig. 5.13  Global variation of the natural background of Ar-37 generated by spontaneous fission neutrons from soil and cosmic rays in subsoil samples (Riedmann and Purtschert 2009).

148

On-Site Inspections

neutrons from soil and cosmic rays in subsoil samples is of the order of 1–120 mBq/cubic meter and varies around the world (Aalseth et al . 2009b) . This global variation relative to a reference laboratory in Switzerland has been modeled using information on the cosmic ray radiation and the calcium content (Figure 5 .13) . In general, these background levels are low compared to expected releases from an explosion, and Ar-37 is an interesting target for detection during an OSI (Han et al . 2009) . To detect gamma radiation from different isotopes, both particulates and gases, is an important OSI activity . Gamma surveys can be carried out as airborne surveys at low altitudes (Seifert et al . 2009a) or on the ground using hand-held or vehicle-mounted equipment (Prah and Tanaka 2009) . A wider range of instruments is available today than when the CTBT was negotiated, the instrumentation has been significantly improved in the recent past, and further improvements may be expected (Köble et al . 2009, Pakhomov and Dubasov 2009) . There are also several methods to conduct the analysis . The simplest method is to measure the total number of gamma counts within the energy window of the instrument . An alarm can be triggered when the number of counts exceeds a pre-set threshold . Using more elaborate analysis procedures, the isotopes generating the radiation can be identified from a gamma spectrum, showing the intensity of gamma radiation as a function of the energy (Figure 5 .14) .

Fig. 5.14 Gamma ray spectrum from nuclear fallout, noting important isotopes . The horizontal axis represents the energy of gamma rays in keV, which is the equivalent energy of an electron when it is accelerated in an electric field of 1,000 volts . The vertical axis represents the normalized number of measured counts .

149

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

One issue of concern is that radionuclide measurements might unintentionally reveal sensitive information about other nuclear activities, past or present. Different techniques have been developed that limit the extent of the information to that necessary to detect and identify signs of a nuclear test. One such blinding technique compares the energy in eight different parts of the gamma spectrum and gives an alarm when the criteria are met for detecting radiation from a nuclear test. The intent is that in this way no details are disclosed that are not pertinent to the inspection (Seifert et al. 2009b). To detect noble gases is a more laborious and time-consuming process. To analyze a sample may take from three to 24 hours. A high-speed approach to analyzing multiple samples will allow detection systems to process and measure 32 samples per day (McIntyre et al. 2009). 5.4.3  Seismological Aftershock Monitoring Seismology is another important OSI technology that can cover a large area and provide locations with high accuracy. There is a vast amount of experience in the seismological community on the detection and location of very weak aftershocks and the monitoring of mines and large constructions such as dams. Seismological observations during an OSI may detect signals associated with the deterioration of the cavity or other weak movement in the ground in the vicinity of the cavity of an underground nuclear explosion. Three U.S. nuclear explosions at the NTS—Faultless, Boxcar, and Benham, all conducted in 1968—are well known for generating a large number of seismic aftershocks. The NPE 1 kiloton chemical explosion discussed above and a number of nuclear explosions at the former Soviet Union test site at Semipalatinsk also generated a large number of aftershocks (Table 5.3). Table 5.3  Aftershock Sequence Observed at a Test Site in the Former Soviet Union (Smith et al. 2009)

Date

Shot

Depth (m)

18.10.1981 25.04.1984 18.09.1984 18.09.1987 22.04.1988 06.09.1988 07.08.1989 02.09.1989

Shaft 1236 Shaft 1316 Shaft K-4 (Quartz-4) Tunnel 132R Tunnel 704 Shaft RN-1 (Rubin-1) Shaft 1352 Shaft 1410

~500 ~500 557 69 130 791 523 ~400

150

Tunnel length (m)

Yield (kt)

462 466 -

~100 ~100 10 ~0.1 ~1 8.5 30 ~5

Measurement period (hours) 38-161 12-161 0.13-7

Period of aftershock existence (days)

Number of recorded events

15 20 80 25

154 322 394 51 144 152 570 136

On-Site Inspections

Fig. 5.15 The cumulative number of aftershocks of the nuclear explosion Colwick at the Nevada Test Site (NTS) versus seismic magnitude. Fewer events are detected as the magnitude increases. Hard rock environments such as the former Soviet Union test site at Semipalatinsk in Kazakhstan, for example, had larger aftershocks (Smith et al. 2009).

It is interesting to note that some explosions with yields of the order of and below 1 kiloton have generated a fair amount of aftershocks. The aftershocks are generally weak, and the number of aftershocks versus their magnitude is shown in Figure 5.15 for the Colwick explosion conducted in volcanic tuff. The largest aftershock had a magnitude slightly larger than zero. Hard rock environments such as the former Soviet test site at Semipalatinsk produced larger magnitude aftershocks. The rate of aftershocks from an explosion decreases with time after the event. Their intensity as a function of time for an explosion at Semipalatinsk is shown in Figure 5.16. Explosions in different bedrocks show different behavior; explosions in granite generate stronger and larger Fig. 5.16  The rate of aftershocks from an explosion in a vertical shaft at the former Soviet test site at Semipalatinsk in Kazakhstan decreases over time. The rate is similar to that of an explosion in a Soviet test site tunnel (Smith et al. 2009).

151

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

numbers of aftershocks that decay more slowly over time than in other kinds of rock. This is illustrated in Figure 5.17, which shows a comparison of the aftershock rates from explosions at Nevada, having softer bedrock, and Semipalatinsk, which is hard granite. Seismological observation is the only technique that can clarify that a suspected event is an earthquake. Earthquakes generate aftershocks that generally are two orders of magnitude, or a factor of 100, stronger than those generated by explosions of similar magnitude. The cumulative number of aftershocks as a function of magnitude for explosions and earthquakes at the NTS are shown in Figure 5.18. As noted earlier, explosions in granite are likely to generate stronger aftershocks than those presented in this figure. The possibility of detecting signals at close distances from such weak seismic aftershocks depends very much on the local conditions. Disturbances generated by human activities such as traffic contribute to the seismic noise, which is normally significantly lower during nighttime. The use of several sensors to form a seismic array improves the signalto-noise ratio roughly in proportion to the square root of the number of sensors. Solid bedrock, such as granite, normally has a lower background noise and a better transmission of the signals than less consolidated rock. It was noted above that aftershocks following explosions in granite are stronger than those in softer rock; this can be a source effect, but it might also be that the good transmission properties make them appear stronger. Another uncertainty when estimating the probability of detecting aftershocks of a certain magnitude is the difference in local magnitude scales, as discussed in Chapter 2. This uncertainty may increase when Fig. 5.17 Aftershocks per day following the occurrence of explosions at the former Soviet Union (numbers refer to explosions in Table 5.3) and Nevada test sites. The granite of the Soviet test site produces orders of magnitude more aftershocks (upper red lines) than the alluvium or tuff geology of the NTS (lower blue lines) (Smith et al. 2009).

152

On-Site Inspections

Fig. 5.18  The cumulative number of aftershocks versus magnitude shows that there are far fewer aftershocks from an explosion than from an equivalently sized earthquake at the NTS (Smith et al. 2009).

the events get smaller and are recorded only by stations at very short distances. Earthquakes with magnitude in the range 0 to -0.6 are observed at distances less than 10 kilometers by a local network at the Hayward fault in California (McEvilly et al. 1999). Joswig (2009) has studied the possibilities of detecting small buried explosions. Using a network of small array stations with four elements, he found that a 1 kilogram explosion could be detected at distances up to 10 kilometers in the test area. Procedures for analyzing the data from these four-element arrays were developed and tested during the IFE08 (Giuntini et al. 2009b). Joswig suggests that a 1 kilogram explosion might correspond to a magnitude -1 event. As discussed in Chapter 2, there is considerable uncertainty in any conversion between explosion yields and magnitudes, and vice versa, and this uncertainty is likely to increase for smaller events detonated in a heterogeneous ground surface material. Kovacs et al. (2009) have studied the location accuracy over an area of 8 kilometers by 10 kilometers using explosive sources in the range 50–800 grams recorded by two mini-arrays and eight three-component stations. The location accuracy was found to be of the order of 100 meters for explosions in the center of the area and around 300 meters for explosions at the perimeter of the test area. Military applications show that it is possible to improve the location accuracy further. One efficient means of calibration is a step-by-step procedure using a fixed network of locally deployed stations. Once an aftershock is recorded and located, a calibration shot is set off at that location and the aftershock is re-located relative to that calibration shot.

153

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

The procedure can be repeated and the uncertainty might soon be of the order of meters. As this requires cooperation with the inspected state to conduct explosions, the most likely application is when the triggering event is an earthquake and both parties are interested in clarifying the event. If one or more aftershocks are recorded, the precise estimation of their locations and depths could prove decisive in identifying a triggering event as a natural earthquake. 5.4.4  Geophysical Exploration Techniques During a continuation period of an OSI, the inspection team can use additional geophysical exploration techniques that might provide further information about the event under scrutiny or about phenomena or installations related to it. An advanced exploration technique comprising seismic reflection measurements has been developed in particular for oil exploration. This technique is extensively used on an industrial scale for off-shore exploration, as well as on land. No reports are available on any seismic reflection survey at a site of an underground explosion. Given the dramatic development of this technique over the last decades, it is possible that the explosion cavity and the fracture zone surrounding it would be detected if a survey were conducted on the surface right above the cavity. A reflection survey requires specialized measurement and analysis equipment and techniques and must be conducted by a professional team. It is a large-scale and expensive operation. And while it would be unrealistic to conduct such surveys over large areas during a CTBT OSI, it might be justified as a final check in a limited area to identify and pinpoint a potential cavity prior to drilling. Other geophysical techniques that may be used during an OSI include microgravity, magnetic, and electric surveys. How would a cavity, the surrounding fracture zone, and the borehole or the tunnel used in the operation affect those geophysical fields at the surface? This is another area in which little data are available from measurements that may have been conducted at sites of nuclear tests. These geophysical exploration techniques are all well established, important tools in mineral exploration and bedrock studies. It is generally recognized that the conduct and analysis of such surveys require expert teams. Aeromagnetic surveys are a fast and straightforward method to map magnetic anomalies in an inspection area (Gaya-Piqué et al. 2009). The magnetic field, like the gravity field, is complex, and it may be difficult to interpret the observations in terms that are relevant to an OSI. It might add value when interpreted together with other observations, underlining the importance of data and information integration. Local magnetic

154

On-Site Inspections

observations may reveal the presence of magnetic material left in the ground, such as cables and pipes. Ground penetrating radar is another technique that can be used to discover underground objects and constructions down to a depth of a few meters, depending on soil conditions. Models can be used to predict the way these electromagnetic and gravity fields are affected by underground heterogeneities such as emplacement tunnels or boreholes and explosion generated cavities. When such features are modeled in a homogeneous earth model, surface effects can be clearly shown (Fourie et al. 2009, Pignatelli et al. 2009). The real world is a complex one, however, and these fields are disturbed by a number of local features, such as geological inhomogeneities and strong variation in topography, that make interpretation of the observations most difficult. The usefulness of these techniques, therefore, needs to be demonstrated by making measurements at the sites of underground nuclear tests in different environments. To clarify, this could be an interesting international cooperative project and such measurements would not reveal any sensitive information on the tests conducted. 5.5  OSI: A Key Tool for States to Verify and Deter An OSI provides countries a unique tool that has strong political, technical, and operational components and that can serve countries by enhancing verification and deterrence—but how likely is it that an OSI will ever be conducted? How likely is it that a state or group of states will request an OSI? As mentioned, the experience from the CWC is that, so far, a challenge inspection has not been requested. There have been only two attempts to request the IAEA to conduct a special inspection. In 1992 Romania requested that the Agency clarify discrepancies about unreported plutonium separation that took place during the rule of Nicolae Ceausescu. The IAEA Director-General reported the results in September 1992, and these results did not generate further questions. In 1993 the Agency requested a special inspection of North Korea but was not granted access. This led to a report on North Korean non-compliance to the UN Security Council, and the case is still unresolved (Heinonen 2010, Carlson and Leslie 2005). It is likely that the threshold of requesting an OSI under the CTBT will be much lower, as the supporting evidence from monitoring activities are likely to be much stronger. A country requesting an OSI will need to be able to present convincing and credible information to the Executive Council in support of a request to gain an affirmative vote. Another key condition will be that the states represented in the Executive Council have the scientific and technological expertise to be able to assess the information presented in support of a

155

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

request. To establish the necessary expertise globally is a long-term process and should be in the interest of all countries. Chapter 7 proposes a model of regional cooperation and centers to facilitate such a knowledge build-up. A third hurdle to the conduct of an OSI is the reaction of the state to be inspected; it might refuse to accept an inspection. As with other treaties, there is no way to conduct an OSI by force, and the consent of the inspected state is a prerequisite. Would a country accept an inspection if it has carried out a clandestine test and the requested inspection area includes the site of the explosion? Rather than be caught in an inspection, would it not prefer to reject the inspection and face the consequences? Article V of the treaty identifies “Measures to redress a situation and to ensure compliance, including sanctions.” These measures are similar to those in other treaties and can include “to restrict or suspend the State Party from the exercise of its right and privileges under this Treaty,” or “collective measures which are in conformity with international law,” and “to bring the issue, including relevant information and conclusions, to the attention of the United Nations.” Following the nuclear testing by India (1998), Pakistan (1998), and North Korea (2006 and 2009), the Security Council unanimously adopted resolutions condemning them and sanctions were imposed (UNSC 1998, 2006, 2009a). As discussed in Chapter 7, resolute enforcement is crucial to creating a strong deterrence. If the country to be inspected accepts an inspection there might be three scenarios: there has been a clandestine test within the requested inspection area; there has been a clandestine test, but the requested inspection area does not cover the explosion site; and, finally, there has been no clandestine test—the triggering event was a natural event, most likely an earthquake, that might be located within or outside the inspection area. Starting with the last scenario, it is likely that the inspected state is eager to demonstrate that it was a natural event and not a clandestine explosion. This most likely applies whether the event is within the requested inspection area or outside. To identify an observed event as a natural event might be a difficult task even with good will on all parts involved, however, and the primary step would be to estimate a precise location and depth of the event. The inspected country might be operating a national seismic network that could provide important data to increase the precision of location and depth of the event. Chapter 7 discusses how the credibility of such network data could be enhanced to serve this purpose. Seismic aftershock measurements could detect and locate possible aftershocks and radionuclide measurements could establish the absence of radionuclides, in particular xenon. Still, this might be a difficult situation and it is important that states, including those in the Executive Council, have the expertise to address these issues.

156

On-Site Inspections

The second scenario is that a country has conducted a clandestine test and the explosion or the triggering event is mislocated in such a way that the requested inspection area does not contain the site of the explosion. This might well be a reality, as 50 percent of the locations currently provided by the PTS have associated uncertainties exceeding the dimension of the allowed inspection area. As further discussed in Chapter 7, in this case an OSI will not detect any evidence of a clandestine test; however, there might be observations of infrastructure associated with a testing operation or of xenon or other radionuclides that might have been dispersed. An important issue under these circumstances is for the inspection team and the Executive Council to decide how and when to conclude the inspection. The political issue might remain unsolved. This scenario stresses the importance of precision monitoring to obtain an accurate estimate of the location of the triggering event. In the most frequently discussed scenario, the clandestine explosion is within the requested inspection area. This scenario would be the result of the monitoring activities having been able to locate the triggering event with high accuracy, as discussed in Chapter 2, thus providing an adequate basis for selecting an inspection area that covers the site of the event. The more precisely the triggering event is located, the greater the probability that the inspection will be successful. A clandestine test that has been located with precision within the inspection area is highly likely to be revealed during an OSI. It is highly unlikely that all the evidence of a testing operation could be concealed, given all the tools and the substantial inspection time available. The fact that a clandestine test is likely to be detected during an OSI, if the triggering event can be precisely located, will serve as a strong deterrent. It should also encourage countries to make extra efforts to increase the location accuracy through “precision monitoring,” as discussed in Chapters 2 and 7.

157

Chapter 6

Synergy with Science This chapter reviews the long-standing relationship between science and the Comprehensive Nuclear-Test-Ban Treaty (CTBT) and how this relationship has, on the one hand, helped develop individual sciences and, on the other hand, significantly improved the CTBT verification system and the ability of countries to verify compliance with the treaty. It also discusses how data produced by the International Monitoring System (IMS) can be used for other purposes, in particular, applications related to key global issues such as global warming and environmental protection. Such dual use provides additional benefits to states from their large investments in the IMS and strengthens relations between the Provisional Technical Secretariat/Technical Secretariat (PTS/TS) of the CTBT Organization (CTBTO) and the scientific community. This relationship is important to maintaining the PTS/TS as a viable entity, able to recruit competent staff. Cooperation with the scientific community is also essential to keeping the development of the techniques and procedures used at the PTS/TS in pace with scientific and technological developments, within the constraints set by the CTBT. Countries can also improve their verification capabilities through application of scientific developments and close cooperation with the scientific community in a number of areas, as discussed in Chapters 2–5. There are no formal limitations on states to take advantage of new scientific and technological developments to improve their capability to verify the CTBT, and they can incorporate new technologies and analysis methods as they become available. Furthermore, states need a close connection to the scientific community to develop the expertise necessary to analyze and interpret verification data and information. While activities carried O. Dahlman et al., Detect and Deter: Can Countries Verify the Nuclear Test Ban?, DOI 10.1007/978-94-007-1676-6_6, © Springer Science+Business Media B.V. 2011

159

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

out during the Preparatory Commission (PrepCom) period, including the training programs conducted by the PTS, have increased the general knowledge about CTBT verification globally, further work remains in most states and regions to establish the expertise needed, and the regional cooperation suggested in Chapter 7 would serve as a good frame for such activities. 6.1  Science and the Nuclear Test Ban Treaty— A Long-Standing Relationship The relationship between the scientific community and the nuclear test ban spans more than half a century and has provided many mutual benefits to date. From the beginning of the test ban discussions, verification issues have been prominent, and the application of seismology to detect underground nuclear explosions became a key element in the consideration of the CTBT verification system. The desire to monitor underground explosions promoted the development of the science of seismology, and considerable resources were mobilized in many countries to conduct extensive verification-related research programs and to build new and more sensitive seismological stations. The World Wide Standard Seismograph Network (WWSSN) was initiated in 1961 as part of the U.S. Vela Uniform program to improve the detection of nuclear explosions (Powell and Fries 1966). The WWSSN was a global network built with standardized equipment and high-precision timing systems. Data were recorded photographically, and recordings were made openly available to scientists on microfilm. This data distribution system, the first ever with a global reach, was key to the success of the WWSSN. The concept of seismic array stations also developed at about the same time, and many such stations were established around the world as a component of the verification programs of several countries (see Chapter 2). Underground nuclear testing provided seismology for the first time with a controlled source strong enough to generate seismic signals that could be observed at global distances. When the Long Shot nuclear test was conducted in the seismically active area of the Aleutian Islands in 1965, seismologists were in for a surprise: when the explosion was located using observations from more than 300 stations, the estimated location was 22 kilometers away from the true location, and the 90 percent confidence region had a maximum extent of only 8 kilometers. The estimated depth was 58 kilometers, compared to the actual shot depth of 0.7 kilometers (Lambert et al. 1969). This drew attention to the systematic deviation, or bias, in the determination of seismological event locations. Many more such examples were found later for nuclear explosions in other parts of the

160

Synergy with Science

world. As a result, a new and more accurate travel time model for seismic waves from shallow events was developed (Herrin 1968). Nuclear explosions also were used in the Soviet Union as sources of seismic measurements over long distances to study the Earth’s structure (Nordyke 1974). An even more surprising and important discovery was that seismic signals from an explosion, a fairly simple and symmetric seismic source, were quite complex. The signals observed at different stations from the same explosion could have quite different waveforms and spectra. The same was true when explosions in different parts of the world were observed at the same station. This revealed that the interior of the Earth is far more complex than was previously expected when most of the complexities in seismic signals were attributed to earthquake sources. The dispute that occurred between the United States and the Soviet Union on the verification of the 150 kiloton yield threshold in the Threshold Test Ban Treaty, discussed in Chapter 2, is an interesting example of how observations of explosions differ among different regions of the world. In addition to national research efforts and informal cooperation among institutions in many countries to build new stations, promote research, and share knowledge, the Conference of the Committee on Disarmament (CCD) in Geneva established in 1976 the Group of Scientific Experts (GSE) to consider how seismological observations could facilitate the monitoring of a CTBT (Dahlman et al. 2009). The primary responsibility of the GSE was to design and test a global seismological verification system, which by and large has become the seismological component of the IMS, discussed in Chapter 2. The GSE continued until the CTBT negotiations ended in 1996 and was a successful effort to promote a disarmament treaty by sustained scientific work in a political environment. During the CTBT negotiations in 1994–1996, experts in other technologies joined the negotiations to develop the hydroacoustic, infrasound, and radionuclide components of the IMS, using the seismological component as a template. 6.2  Rapid S&T Development since the Treaty Opened for Signature The work of the GSE documented the significant scientific and technological developments that took place over a number of decades at the end of the 20th century. The scientific and technological developments have continued since the CTBT was opened for signature in 1996—within the different verification technologies specified in the CTBT as well as in other technologies that states may use as part of national technical means (NTMs). At the International Scientific Studies Conference in 2009 (ISS09) in Vienna, a large number of presentations illustrated recent scientific and technological developments in different areas. As might be expected,

161

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

developments were more noticeable in new areas such as radionuclide noble gas detection than in well-established areas such as seismology. Infrasound, which had received little attention since testing in the atmosphere ended, also generated renewed interest, for reasons discussed later in this chapter. The developments have been even more significant in information technology, including communication, computer hard- and software, and information analysis. New approaches to managing, analyzing, and exploiting large volumes of data, usually referred to as data mining, have been developed to support rapidly growing Web applications. Data mining has great potential to improve the processing and analysis of IMS data at the International Data Center (IDC) and of the information collected during on-site inspections (OSIs). It also provides enhanced monitoring possibilities for countries, as it improves their ability to analyze and exploit ever more data and information, including contributing to the development of a new concept of precision monitoring using a number of different techniques to monitor one or a few areas of interest. Easily available overhead satellite observations—both high-resolution optical and radar data—are important monitoring tools for states. Since the beginning of the 21st century there has been significant improvement in the resolution of satellite images. A great deal of such data are generally available at low cost, as are computer programs that allow in-depth interpretation of the data. Software that can identify the differences between images of a scene taken at different times makes it possible to follow activities in progress, such as the construction of infrastructure for the conduct of a nuclear test explosion. The analysis of radar data has also improved, and the Interferometric Synthetic Aperture Radar (InSAR) technique makes it possible to detect small changes in ground elevation from one observation to the next by scanning the same point on the Earth’s surface regularly. These developments in overhead observations have dramatically improved the ability of countries to monitor selected areas subject to precision monitoring and interpret observations made by other sensors. Satellitebased navigation systems that provide high-precision position finding in any part of the world are an important tool during an OSI. How will these developments, which have taken place since the CTBT was negotiated, affect, on the one hand, the IMS and the OSI regime and, on the other hand, the ability of states to verify compliance with the treaty? The treaty is specific in regard to the number and location of the IMS stations; it does not specify the equipment in any way that would limit the possibility of implementing new technologies. However, the PrepCom has established specifications that stations must meet to be certified. The same is true for the IMS data analysis, an area where it is possible

162

Synergy with Science

to benefit from new scientific developments, and small steps in that direction have been taken by the PTS. For reasons explained in Chapter 2, the CTBT does not incorporate the use of data from auxiliary seismological stations in its online event detection process at the IDC, although today’s communication techniques easily allow the collection of all data online from those stations. As noted, countries have no formal limitations on benefiting from scientific and technological developments. In addition to exploiting improvements in the IMS and analyzing these data without any of the restrictions imposed on the PTS/TS, they can take advantage of scientific and technological developments to improve their NTM monitoring capabilities in whatever way they see fit. The growing amount of highquality seismological data readily available outside the IMS, as well as satellite observations and new data mining tools to focus the analysis on areas of interest, will provide countries, individually and in cooperation, with a monitoring capability considerably greater than that envisaged when the treaty was negotiated. The treaty identifies a large number of techniques that may be used during an OSI, although it does not specify any functional or technical requirements. The equipment is being specified during the PrepCom work, and it should thus be possible to include the latest scientific and technological developments. The work to date has demonstrated, however, that states are quite careful and conservative in implementing these. During an OSI it is important to rapidly analyze and exploit the data and observations gathered, and there are no limitations in the treaty on the way the data analysis and interpretation should be carried out; the new data mining techniques could prove valuable. Yet further work is needed to develop, test, and implement such techniques. 6.3  Synergy among Different Sciences To develop a sustainable synergy among the scientific community, the states, and the CTBTO’s Technical Secretariat, all parties should benefit; win-win situations need to be created. To exploit IMS data for non-verification purposes, in particular applications related to such key global issues as global warming and protection of the environment, is an important dual use of investments already made and operational resources allocated by countries. It is also a way to maintain broad scientific interest in IMS data and keep the system up to date over long periods of time, even after providing monitoring data has become a routine activity. Making IMS data available to the scientific community for research purposes is a key element in such cooperation. A good precedent has

163

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

been established in seismology by providing recognized tsunami warning centers online data from selected IMS stations. Given the abundance of data available from seismological stations outside the IMS, access to the IMS seismological data is not likely to reveal anything of concern to states that cannot be observed by other stations. The majority of IMS infrasound and hydroacoustic observations are likewise unlikely to reveal anything that would be of concern to any country. The radionuclide observations might be useful for other applications, such as monitoring accidental nuclear releases from nuclear facilities as discussed below. The PTS has made an analysis of the results of the ISS09 Conference and identified a large number of possible projects that could be realized in cooperation with scientists and scientific institutions internationally (CTBT PTS 2011). This paper also contains a list of all ISS09 Conference posters. 6.3.1 Seismology Seismology is a well developed science that has evolved since the end of the nineteenth century. It is a science of global reach, as seismologists and seismological institutions exist in almost all countries of the world. An open and free exchange of seismological data is of fundamental importance for seismological research, as well as for CTBT verification (Simpson and Willemann 2009). The 170 stations of the seismological IMS network are a small fraction of the 16,000 stations established and operated for scientific and other purposes around the globe. Still, data from the IMS stations would contribute to the scientific community in several ways if they were generally available. The IMS network has modern and highperformance stations with authentication equipment to ascertain that data have not been tampered with. Many IMS stations are among the most sensitive on the globe, and 33 of the 50 stations in the primary network are array stations or will be so when established. Such stations provide not only an enhanced detection capability but also the ability to study wave propagation of signals across the array and focus or beam signals to study the deeper structures of the interior of the Earth (Koper and Xu 2009). Data from all the primary stations are also collected online routinely, and technical arrangements exist to collect data from the auxiliary stations in the same way. Seismological data from the IMS network would be an interesting and valuable addition to data already available for scientific studies and monitoring global seismicity (Kebede and Koch 2009). So far, countries have not agreed to make these data available to the scientific community on a routine basis. Data and bulletins have been provided by the PTS to research organizations on a case-by-case basis as part of the ISS process and through contractual agreements. Access to IMS data is seen

164

Synergy with Science

by many scientists as a prerequisite for enhanced cooperation, and this is a key issue in the relationship between the PTS/TS and the scientific community. As noted, a small step has been taken, as states have agreed to make data from selected stations, both primary and auxiliary, available online to recognized tsunami warning centers. The fact that states agreed to provide data from auxiliary seismic stations online for a non-verification purpose is an important recognition of the dual value of IMS data. The seismological community, states, the CTBTO, and the PTS/TS share a number of scientific challenges. One of the most important of these is to increase the accuracy in locating seismic events and to characterize observed events in relation to each other and to geophysical and tectonic structures. The notion of “precision seismology” has been introduced among seismologists to describe such efforts. This is a concept of focusing activities on which we have modeled “precision monitoring” introduced in Chapter 2. To find the precise location of a triggering event is crucial for an OSI, and high-precision locations are also essential in the analysis of earthquakes in relation to geological and tectonic structures (Bergkvist and Johansson 2009, Bergman and Engdahl 2009, Cichowicz 2009). An event must be located in three dimensions, including the depth below the Earth’s surface, a parameter that is difficult to estimate for shallow events. Cooperative efforts are under way to create a database of well located reference events to be used for calibration. The PTS has taken steps to explore how developments in data mining might be used to improve the analysis of IMS seismological data. The application of modern data analysis and exploitation methods will make it possible to develop a new paradigm for the detection, location, and characterization of seismic events, especially on a regional level. The challenges for seismology have not changed greatly over the last decade. To fully understand the seismic signals and be able to use all the information available in the analysis is still the goal of a large body of research. The aim is to separate the earthquake source characteristics from the influence of wave propagation. Wave propagation information can be inverted to map geological structures on a global, regional, and local scale. The source characteristics provide information on the structure and the dynamics of the earthquake source. An increased combined understanding of geological structures and earthquake sources can contribute to building a coherent picture of the dynamics of the planet. 6.3.2  Radionuclide Observations More than any other observation, radionuclide observations are directly related to the nuclear origin of an explosion. Such observations

165

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

might also contain information on other nuclear activities. Chapter 2 noted that medical isotope production is the main source of the global xenon background radiation. Accidents in nuclear facilities may release radioactive debris that can spread widely. States have been sensitive about sharing radionuclide data with the scientific community or the International Atomic Energy Agency (IAEA), as such observations also may reveal other kinds of nuclear activities. It has therefore been difficult so far to find agreement among states to release radionuclide data for non-CTBT purposes. This attitude may change, however, after the nuclear accident at Fukushima, Japan, caused by the earthquake and tsunami on 11 March 2011. Following this accident, the PTS started to share IMS radionuclide network data and the results of its analysis with the IAEA and the World Health Organization. By mid-April 2011, 35 IMS radionuclide stations had detected radionuclide particles and noble gases from the Fukushima accident. The information collected by the IMS radionuclide network could prove useful not only to assist health authorities in monitoring radioactive releases from nuclear accidents, but also to the IAEA in enforcing its safeguard agreements with states. Such measurements could be seen as a global extension of the wider area environmental sampling that is part of the Additional Protocol to the Safeguards Agreement. In future discussions and negotiations to strengthen nuclear disarmament and nonproliferation, including cessation of the production of fissile materials, or “cut-off” treaty, a possible dual use of data from the IMS radionuclide network might be considered. Synergy and cooperation with the scientific community might also develop to improve the radionuclide observation capabilities in different ways. Developments may be related to new, more sensitive radiation detectors for the detection of explosion debris (Miley et al. 2009, Aalseth et al. 2009a). It may also be related to improved collection equipment that could reduce the sample time from the current 24 hours. This would improve the precision in estimating the arrival time of the observed radionuclides and thus improve the precision in tracking the source of the release. Better understanding of the environmental background radioactivity, especially related to xenon and other noble gases, is an issue of broad interest. There is also widespread interest in improving radionuclide data analysis methods and procedures. Another and more specific monitoring issue is to further develop mobile data collection and measurement systems that can be deployed on airplanes. Such mobile systems will give countries the ability to take measurements in the plume wherever it travels in their national or in international air space. This flexible way of taking measurements will provide states with a valuable

166

Synergy with Science

Fig. 6.1a  Filter from radionuclide station RN08 at Cocos Island, Australia. The filters from this island site station are typically very clean.

Fig. 6.1b  Filter from radionuclide station RN20 in Beijing, China. The filters from this station are usually dark and contain large amounts of non-nuclear particles.

complement to the fixed station network that may also be of great interest to monitor accidental releases (Hafemeister 2009). Scientific work is also proceeding to improve the possibility of observing radionuclides during an OSI. This includes efforts to increase the understanding of the expected target nuclides and possible noise sources. It also involves the development of new technology and sensitive instruments to improve OSI radionuclide measurements (Williams et al. 2010). The filters used at the particulate radionuclide stations collect a large number of non-nuclear particles. No analyses of these particles are conducted currently, but the filters might be an interesting source of information to monitor pollution and the way pollution spreads globally. The aerosols collected may also provide valuable information related to

167

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

climate change. Observations of the filters reveal large difference around the world, as can be seen in Figure 6.1. The filters from all particulate stations that have already been collected at the PTS contain an impressive amount of data. The filter observations are categorized into five levels, ranging from normal background to containing multiple relevant radionuclides, as discussed in Chapter 3. While states are sensitive about sharing data containing radionuclide information, it should be possible to meet that concern by using filters containing normal background only. Analysis of such filters to search for non-nuclear particles can then not be misused by looking for radionuclides, as there will be only the normal background, which is not of a sensitive nature. An essential part of radionuclide monitoring is atmospheric transport modeling to track the path of the radionuclide material from a source to observation stations or from stations back to a source. Development and validation of such transport models is therefore of common interest to states, the CTBTO and its TS, the scientific community, and those organizations working on climate change and global warming. Radionuclide materials, which are present in the atmosphere from a number of sources, can be seen as markers by which the movements of the air masses can be followed from one monitoring station to another. Such tracking information is useful in validating atmospheric transport models both on a regional and global scale. It should be possible to develop the already existing cooperation between the PTS and the scientific community in this area without revealing data and information that countries might consider sensitive. 6.3.3 Infrasound Infrasound has developed into a broad interdisciplinary field that might contribute to progress on many scientific challenges, such as earth-oceanatmosphere interaction and atmospheric circulation, both related to global warming. Infrasound may also be used to monitor naturally occurring events such as volcanic eruptions. The establishment of the IMS infrasound network and the challenge of analyzing the data and understanding observed sources and signal propagation recreated the science of infrasound, which existed at the time of atmospheric testing in the 1960s and then essentially disappeared. Within a period of some ten years, a number of appealing applications of infrasound observations have been identified, in addition to CTBT monitoring. Since 2000 a number of national infrasound stations outside the IMS have been established for scientific purposes. For example, a regional network has been established in Europe with a combination of IMS and national infrasound stations (Ceranna et al. 2009b).

168

Synergy with Science

It is generally considered that there is a very low likelihood of a clandestine nuclear test in the atmosphere. To maintain a long-term interest in infrasound observations and secure a high level of expertise, it is important to further develop the close cooperation that initially existed between the PTS and the growing infrasound scientific community. The vast majority of the signals and the sources observed, such as bolides and volcanic eruptions, are of interest primarily to the scientific community. It is thus important that data are made available to scientists and scientific institutions on a routine basis, and it would make sense to engage scientists and institutions outside the PTS/TS in the work to analyze these observations. Chapter 3 noted that infrasound signals are greatly affected by the atmosphere, which means that they carry a great deal of information on the conditions of the atmosphere, including at high altitudes. Inversion of infrasound observations can be used as a tool to estimate the state of the middle and upper atmosphere at altitudes of 40–120 kilometers (Drob et al. 2009). Infrasound observations can also be used to image the wave activity that contributes to the dynamics of the atmosphere. Annual variation of planetary waves have been observed in the Arctic and Antarctica. The strong amplitude variations of the planetary waves in the winter are produced by gravity waves, which are stronger in the Northern than in the Southern Hemisphere. An anomaly was observed in 2006 over Antarctica that might have been produced by stratospheric warming. A sudden stratospheric warming is an event that occurs at altitudes between 10 and 50 kilometers, during which the polar vortex of westerly winds in the Northern Hemisphere winter suddenly slows or even reverses direction. This is accompanied by a considerable warming of the stratosphere. These are initial results of using infrasound as a tool to study the large-scale dynamics of the atmosphere and thus contribute to the understanding of global circulation and effects of global warming (Blanc et al. 2009). In addition, infrasound observations can be used to monitor events in the lower part of the atmosphere; observations from cyclones, hurricanes, and thunderstorms were reported at the ISS09 Conference (Waxler et al. 2009, Rambolamanana et al. 2009). Cyclonic activities over an ocean generate infrasound due to the interaction of the cyclonic winds with the ocean surface. Cyclones in deep oceans generate more infrasound signals than cyclones on shallow seas. More than 100 incidents between aircrafts and volcanic ash have been reported between 1973 and 2003 (Fee et al. 2009); this illustrates the importance of detecting volcanic eruptions even in remote and unexpected places. The volcanic eruption on Iceland in the spring of 2010, for example, severely interrupted air travel in Europe for weeks. Fee et al.

169

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

Fig. 6.2  Infrasound observations from the Tunguharua volcano eruption in Ecuador on 15 July 2006. The three parts of the figure show: the recorded infrasound signal, with pressure as a function of time (top); the frequency spectrum of the signal (middle); the acoustic source power (black curve) and its correlation with the height of the ash cloud (green curves, bottom) (Fee et al. 2009).

(2009) show the results of infrasound recordings of volcanic eruptions and conclude that their studies demonstrate “the possibility of using infrasound observations over broad regions for automatic, low-latency notification of hazardous eruptions.” Such notification would serve as an input to the Volcanic Ash Advisory Centers (VAAC) (Figure 6.2). The Earth’s atmosphere is constantly bombarded by meteoroids, and observations by the IMS infrasound network can be used to monitor the bolides in synergy with satellite and other observations of such objects. Observations of bolides can be used as a means to evaluate the detection and location capability of the infrasound network, as discussed in Chapter 3. The synergy between observations from infrasound stations and the dense USArray seismic network and an observation of a bolide were presented in Chapter 3 (Hedlin et al. 2009). 6.3.4 Hydroacoustics The vast majority of the signals observed by the IMS hydroacoustic network are related to natural sources in or close to the oceans. Such signals are many orders of magnitude weaker than those that would be observed

170

Synergy with Science

in the unlikely event of a nuclear explosion in an ocean and are thus not of direct interest to CTBT monitoring. Some signals may, however, be useful to calibrate hydroacoustic wave propagation, and many can be of great interest not only to science but also to important civilian applications. Given the shared interest in observations from the IMS hydroacoustic network, it should be possible to develop a synergy with the scientific community in the analysis and use of these data. The PTS is using observations of hydroacoustic signals together with seismological observations to improve the analysis of earthquakes below the ocean floor in areas where the seismic network has limited coverage. The scientific community is primarily interested in hydroacoustic signals from natural events and activities occurring in the ocean. To directly involve interested scientific institutions in the data analysis would facilitate a close cooperation with the PTS/TS, provide useful data to institutions involved, and reduce the routine workload on the PTS/TS. It would also help maintain a highquality hydroacoustic network and develop global expertise, familiarity, and engagement in hydroacoustics. This would ensure a viable hydroacoustic observation and analysis activity in the long term. The IMS network is a valuable tool to record whale sounds around the world throughout the year, even in remote areas where whale studies are difficult to carry out (Flore et al. 2009 ). Deploying the passive hydroacoustic

Fig. 6.3  Hydroacoustic observations of the break-up of a giant iceberg (B15A) during the summer of 2008–2009 in the South Pacific. The red dots with the uncertainty ellipses are the hydroacoustic locations and the green dots are reference locations (Hyvernaud 2009b).

171

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

systems over long periods has allowed scientists to monitor and analyze the movements and migration patterns of whale populations across the ocean basins in a way that does not disturb the animals. The stations record whale vocalizations, even in locations such as sub-Antarctic areas where visual sightings are too costly to conduct (Royer et al. 2009, Harris et al. 2009). A spectrogram from a blue whale is shown in Chapter 4, in Figure 4.9, and Flore et al. (2009) show that whales can be detected at distances ranging up to 100–200 kilometers. Hydroacoustic signals have also been observed from ice calving or iceshelf dislocations in the Antarctic and from drifting icebergs and their breakup (Royer et al. 2009, Hyvernaud 2009b). Figure 6.3 shows the tracking of a giant iceberg, 27 x 122 kilometers—the size of Long Island, New York— in the Pacific Ocean using hydroacousic signals. As shown in the figure, a large number of observed hydroacoustic sources are associated with the gradual deterioration and breakup of the iceberg. The hydroacoustic tracking is compared with precise locations from other sources, indicating a scatter of some 100 kilometers. The Monowai sea mount close to Antarctica has been an active underwater volcano for more than eight years. T-phases have been recorded at the Reseau Sismique Polynesien (RSP), the French seismic network, in and around Tahiti, and hydroacoustic signals have been observed from as far as the IMS station at Juan Fernandez off the coast of Chile. Hydroacoustic signals have provided valuable information on the timehistory of the rupture process of the Sumatra earthquake on 26 December 2004 (Fig 6.4). Salzberg (2009) reports that hydroacoustic observations might be useful for tsunami warnings. He suggests that there is a relation between hydroacoustic signals and tsunami excitation and that it might be possible to determine the likelihood of severe local tsunamis based on hydroacoustic signal spectra. Another application of hydroacoustics is “acoustic thermometry,” which studies ocean warming in the context of climate change (Stephen 2009). The propagation velocities of the acoustic signals in the oceans depend on temperature, pressure, and salinity. Measuring long-distance signal propagation will provide integrated sound speed data that might show small temperature variations over large areas. Studies of the variation of sound speed will require precision measurements that are repeated over the same path for a long time. To provide needed highquality data, however, the locations of hydrophones of the IMS station must be precisely known at the time of the measurements; Chapter 4 noted that hydrophones can move some 10 meters in relation to each other. Precise measurements for environmental studies would require improvements in the IMS hydroacoustic stations that might be

172

Synergy with Science

Fig. 6.4 Snapshots illustrate how the hydroacoustic IMS station at the Chagos Archipelago in the middle of the Indian Ocean tracked over a time span of more than 7 minutes the rupture process of the large Sumatra earthquake on 26 December 2004 that caused the disastrous tsunami in the Indian Ocean. The red star denotes the point of initiation of the earthquake rupture process. The panels to the right of the maps show the “envelope” of the hydroacoustic recording, and the red lines in this envelope indicate the times corresponding to the direction of approach of hydroacoustic waves arriving at the station, indicated by the white lines in the maps.

173

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

implemented in cooperation with institutions engaged in studying global warming. In addition, there is a need to develop a strong enough source that is acceptable to the strict national environmental rules guiding sound sources in the oceans. Thus these technologies provide scientific information that could be valuable to scientists in other fields in addition to the purpose for which they were intended—detecting nuclear weapon test explosions. In the same way that many military applications have civilian or dual purposes, data from the IMS can be applied to other areas. Both hydroacoustic and infrasound observations can be useful in addressing important issues related to global environment and global warming. These issues may be of key importance to states, and it may be in their shared interest that the infrasound and hydroacoustic IMS data are fully used in related scientific studies. 6.3.5  Data Mining Computing techniques have dramatically improved in recent years, and automated methods for text/object identification are becoming more and more commonplace in the commercial sector. Various methods are being applied to uncover meaning in ever-larger data sets. These include a variety of algorithms developed for data mining, a catch-all phrase for the use of automated algorithms such as clustering, classification, and statistical inferencing to find patterns and relationships in large amounts of data. Data fusion techniques are also applied to explore synergy between different types of related data. The first PTS international scientific symposium, “CTBT: Synergy with Science 1996–2006 and Beyond” (CTBTO Preparatory Commission 2006), held in the fall of 2006 showed that data mining could significantly improve the analysis of verification data. Data mining, a relatively new field, gained increased attention during the ISS process and more scientists from that field began to explore how to improve the analysis of IMS data. To connect different sciences is normally a difficult process, given different traditions, reference frames, and languages, but it is also a process that can prove rewarding. It took time, effort, and a number of seminars and workshops to understand and accept how data mining could be effectively utilized (Russell 2009). Chapter 2 discussed a new paradigm for the analysis of seismic events using data mining to relate observed signals to large historical databases of signals and events. It will be a long way, with many steps, to realize such a radically new approach. Studies have been initiated on how to improve the automatic processing of IMS data using different new analyses and

174

Synergy with Science

Fig. 6.5  These maps compare the results of the automatic seismological data analysis (SEL 3) with results of analysis by the expert analysts (LEB) below. The observations cover the period 1 January 2005–31 March 2009. The SEL 3 results contain 202,657 events compared to 142,204 in the LEB. The comparison shows that 30 percent of the SEL 3 events were rejected in the expert analysis and that the distribution of LEB event locations is consistent with the established pattern of earthquake occurrence, whereas the SEL 3 distribution is not (Pearce et al. 2009).

175

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

exploitation techniques. The work and the skills of the analysts are key to preparing a high-quality bulletin based on the results of the automatic processing. Figure 6.5 compares the results of the present automatic analysis and that of the analysts, showing a dramatic difference. One important step is to understand the way experienced analysts build up their empirical knowledge and then exploit that understanding in the automatic analysis. Experts are exploring various aspects of machine learning, including a number of state-of-the-art classification algorithms to improve automatic event processing (Kleiner et al. 2009, Kuzma et al. 2009, Schneider et al. 2009, Procopio et al. 2009). New data mining techniques can also be used to detect signals and identify specific phases in the waveforms (Ohrnberger 2009, Tuma and Igel 2009). Given the enormous amount of already available detected signals, one issue is how such prior information might improve detection (Selby and Bowers 2009). The ISS09 Conference served as a catalyst in bringing together the data mining and the seismic, hydroacoustic, and infrasound communities. The PTS initiated a follow-up, the virtual Data Exploitation Center (vDEC). This will be a development environment that enables researchers from around the world to use state-of-the-art data mining techniques to create new data processing methods and test them on IMS data (Vaidya et al. 2009). This will allow new data analysis and exploitation techniques to evolve, which might be valuable not only for CTBT applications but also for the scientific community. The initial results presented at the ISS09 Conference on the application of data mining to the analysis of seismic, hydroacoustic, and infrasound observation may form the basis for the creation of a new paradigm for the analysis of such data. Given the increasing amounts of available data and the resulting difficulty in performing data analysis and data interpretation, data mining may be a useful tool both for CTBT verification purposes and for scientific work in general. States can also greatly benefit from exploiting new data mining techniques in different ways within the concept of precision monitoring. Data mining is an essential part of a new paradigm to be developed for the analysis of seismic data, and it could also provide procedures for threshold monitoring of seismic and noble gas data. New analysis methods will also be available to analyze satellite observations to identify changes in infrastructures and other construction activities over time. New data mining procedures might benefit the analysis and interpretation of the large amount of different observations made during an OSI. So far, however, very few studies have been conducted on how to integrate and make a joint interpretation of the diverse data obtained during an OSI. This is a challenging area of research.

176

Synergy with Science

6.4  Science and the CTBT—Perspectives for the Future A close future relationship between the CTBTO and the scientific community will be mutually beneficial. States around the globe will need to engage the scientific community in order to establish their capabilities to verify the CTBT nationally or in regional cooperation. The PrepCom and its PTS—and after entry into force the CTBTO and the TS—will need to stay tuned to the scientific community to keep its systems up to date and to be able to maintain a qualified staff and a viable organization. The scientific community will benefit from access to valuable data and interesting research challenges. 6.4.1  CTBTO and Scientific Community Perspective The PTS/TS needs to recruit qualified experts in a number of fields, and that need will continue into the future. This will be possible only if the organization is regarded as a stimulating and challenging environment. Once the IMS is fully established and the data analysis procedures and programs are in place, the organization runs the risk that the tasks at hand might become routine. The same might be true for OSI work—once the preparations have been completed and the task is to maintain readiness for something that might or might not happen in the future. To attract and keep a professional staff, it will be necessary to establish close links to the scientific community. This is also essential to continuously update IMS and OSI technologies, within the frame of the treaty, in pace with scientific developments. Although the seismological IMS network comprises only a small fraction of all seismological stations around the world, it is also valuable for research purposes, as it is a standardized and high quality network. A number of the IMS stations are also array stations providing unique data. The infrasound and the hydroacoustic networks are unique global networks and their data are of great interest to scientists in those areas. In these two areas there is a clear dual benefit: the scientific community will profit from IMS data supporting research that will improve the knowledge of the atmosphere and the oceans, which in turn will facilitate the routine analysis at the PTS/TS. In addition to sharing data, how can cooperation between the scientific community and the PTS/TS develop? A number of national and some international institutions are providing routine bulletins based on available seismological observations in much the same way as the PTS, with the difference that the PTS is using IMS stations only. These institutions currently share with the PTS, and later will share with the TS, a common goal to make their analyses more efficient, increase the detection capability,

177

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

provide locations and depth estimates with high precision, and reduce the number of spurious events. To develop new and more efficient procedures for the analysis of seismological data is important for the PTS/TS and other institutions producing bulletins, and this could be a joint effort. An ongoing process, such as a continued ISS, would focus the interest of many scientists on issues related to the CTBT and would benefit the PTS/ TS more than individual conferences or short-term activities. It would be coordinated by the scientific community in a sustained and uninterrupted fashion to carry out studies on important aspects of the verification system. Conferences such as the one organized by the PTS in June 2011 (CTBT: Science and Technology 2011), would be important elements in such a sustained activity. 6.4.2  States and the Scientific Community Prior to and during the negotiations, many states invested substantial effort and money on national research and development programs to support the creation of a CTBT verification regime. Through this work experts were trained and knowledge was accumulated in many states. Once the treaty was opened for signature, many of those programs were canceled or substantially reduced. States need to re-establish scientific efforts to prepare for the verification of the CTBT after entry into force; this applies both to monitoring and on-site inspection activities. Because the task of verifying compliance with the treaty rests with the states parties and few states have the resources or expertise needed to carry out this task, states should establish their expertise at a national level or in cooperation with other states. Scientific and technological development is providing states additional potential to monitor the CTBT well beyond what the IMS offers and this additional capability is growing. Countries should develop, in cooperation with the scientific community, new analysis methods and procedures needed to carry out advanced analysis such as precision monitoring. Countries also share with the scientific community the challenge to understand natural events, such as earthquakes, and their relation to geophysical and tectonic structures. To increase the understanding of how seismic, infrasound, and hydroacoustic signals propagate and how particles and gases are transported in the atmosphere are areas in which states and the scientific community have common interest. As mentioned in Chapter 2 and developed further in Chapter 7, cooperation on a regional basis might be a good way of creating a global engagement in the monitoring and verification of the CTBT. Countries in a region having similar political priorities regarding verification of the CTBT

178

Synergy with Science

could establish a joint monitoring center to focus monitoring on areas of common interest and to provide independent analysis and assessments as a basis for political consideration and decisions. Experts working at those centers must be able to analyze the different monitoring data provided by the IMS and other available sources; they should also make an assessment of observed events in areas of interest. The monitoring centers could, in addition to carrying out their monitoring task, engage in related scientific issues of interest to the region. Such additional civilian or scientific activities could benefit a region and keep its center viable. Training of experts would be an essential initial element in the establishment of regional monitoring centers. Scientific cooperation within a region and between regional centers would be an effective way to increase and share knowledge and to ensure that expertise will be available in many states. It will be essential to establish close cooperation among the regional centers and between the centers and the broader scientific community. The establishment of such cooperation could benefit from the growing collaboration among National Data Centers. During the PrepCom phase, activities such as the PTS training programs and the program to bring experts from developing countries to Working Group B meetings have increased the general knowledge on CTBT verification globally. This provides a good foundation, but further sustained efforts are needed to establish the expertise required to analyze and interpret verification data of different kinds. Creating provisional regional monitoring centers before entry into force of the treaty in order to train experts and develop methods, procedures, and infrastructure would be an important step to provide states the expertise needed to monitor the CTBT. States could, without delay, take an initiative to develop joint research projects at the monitoring centers on issues of specific interest. One such issue that needs immediate attention is how to develop the concept of precision monitoring to integrate and use all available data for precision monitoring of limited areas.

179

Chapter 7

Verifying the CTBT— A State Perspective How can countries verify compliance with the Comprehensive Nuclear-Test-Ban Treaty (CTBT) and detect and deter violations? The key conclusions in this book hinge on the fact that verification of compliance and non-compliance with the CTBT rests with the states parties. Verification is essentially a political process based on a broad range of verification tools that have improved over the last decade. The CTBT identifies three steps in the verification procedure: monitoring for evidence of clandestine tests, engaging in a consultation and clarification process to resolve issues of concern, and requesting and conducting on-site inspections (OSIs) to clarify whether a nuclear explosion has been conducted in violation of the treaty and to gather facts that might help to identify the possible violator. Should a clandestine activity be identified, the actions to be taken will be addressed on a case-by-case basis. The International Monitoring System (IMS) and the data and information provided by the Technical Secretariat (TS) of the CTBT Organization (CTBTO) are essential and high-performing tools available to all states parties to the CTBT. Countries should develop and sustain competence and resources to assess and interpret the information provided by the TS and by other sources available to them. A country is likely to focus its efforts and resources on only a few states and areas of concern. In Chapter 2 we introduced the concept of precision monitoring for such focused efforts, which will increase detection capability and improve location

O. Dahlman et al., Detect and Deter: Can Countries Verify the Nuclear Test Ban?, DOI 10.1007/978-94-007-1676-6_7, © Springer Science+Business Media B.V. 2011

181

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

accuracy compared to what can be obtained at a global level. Precision monitoring can be conducted at a national level or with international cooperation among countries having similar political priorities regarding CTBT verification. In addition, the on-site inspection (OSI) regime in the CTBT is potentially a powerful verification instrument. Countries have a responsibility to assure that OSI strategies, procedures, and equipment are developed so that an OSI team is properly organized and staffed with the needed experts. The dramatic scientific and technological developments since the CTBT was opened for signature have improved monitoring and OSI capabilities significantly and have also provided countries access to valuable data and information that are generally available in addition to what will be provided by the TS. 7.1  Verification—A Political Process In the political process of verification, a state party to a treaty makes an overall assessment whether other states are in compliance with the provisions of the treaty. Such an assessment is based on the political situation, the relations among the countries concerned, the issue at hand, and the verification information and data available. “Trust but verify” is an old slogan from former U.S. President Ronald Reagan, and it may be true that there generally is an element of verification of compliance in agreements even among trusted partners. The less trust and confidence among countries, the greater is the need for verification. In assessing its need to verify the CTBT, a state will make a political judgment on where its main concern is, and different states may well come to different conclusions. A country’s political will and its technical ability to develop a nuclear explosive device are key elements in such an assessment. Political priorities and the constraints on spreading nuclear weapons imposed by the Treaty on the Non-Proliferation of Nuclear Weapons (NPT) are likely to make a state focus its verification efforts primarily on a very limited number of other states. Even if conditions may change over time, any changes will be gradual enough to allow a state to reassess its verification priorities. 7.1.1  How Much Verification Is Enough? The adequacy of verification is in the eyes of the beholder, as there is no objective way of making a generally applicable assessment of what constitutes “adequate” verification. Several attempts have been made to address the adequacy of a verification regime. During the CTBT negotiations it was suggested that adequate verification is “verification that satisfies all

182

Verifying the CTBT—A State Perspective

concerned,” which underlined the subjective nature of such judgments. During ratification of the 1988 Intermediate-Range Nuclear Forces (INF) Treaty by the U.S. Senate, Ambassador Paul Nitze defined effective verification as follows: “[I]f the other side moves beyond the limits of the treaty in any militarily significant way, we would be able to detect such a violation in time to respond effectively and thereby deny the other side the benefit of the violation” (Lugar 2009). This definition underlines the relationship between verification and the overarching security situation. It recognizes that few, if any, verification systems are able to detect a minor violation of a treaty or an agreement. For most verification systems, there is a minimum capability. Verification measures should be designed to enable the parties to detect evidence of possible non-compliance before such activities threaten the core security objectives of the countries concerned. A high detection capability will also provide a high degree of deterrence against clandestine testing. Nevertheless, it is surprising that countries have judged the need for verification differently in different treaties. For example, the CTBT verification provisions stand in contrast to those of other multilateral disarmament treaties related to weapons of mass destruction. The three key treaties in force related to weapons of mass destruction are the Biological Weapons Convention (BWC), the Chemical Weapons Convention (CWC), and the NPT, as discussed in Chapter 1. The BWC, established to address one of the most serious threats to the world, has no verification provisions, whereas the CTBT has a comprehensive global monitoring system and an intrusive OSI regime. The BWC refers compliance issues to individual states and to the United Nations Security Council. The CWC contains extensive provisions for OSIs not only of chemical weapons storage and destruction facilities but also of chemical industries that produce specified chemicals that might be used as precursors to chemical weapons. OSIs are conducted on a routine basis to verify declarations and ongoing operations in industries and at destruction facilities. The treaty also contains provisions for challenge inspections but, despite suspicions of clandestine activities, no such challenge inspection has been requested to date. This is an example of a situation in which the political threshold for an inspection request has increased because the option has not been exercised. It illustrates the importance of applying the verification tools; otherwise, the international community may face a situation in which the tools may become politically impossible to use. The NPT includes a provision for comprehensive verification or safeguard arrangements concluded between all non-nuclear weapon states parties to the NPT and the International Atomic Energy Agency (IAEA). These agreements provide for different kinds of inspections and

183

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

routine monitoring of nuclear material and activities involving such material at facilities that have been declared by states. The safeguards agreements also contain a provision for special inspections that the IAEA may conduct if it considers the information provided by the state concerned inadequate for the IAEA to fulfill its responsibility. As noted in Chapter 5, this has been seldom applied (Carlson and Leslie 2005). The experience from the clandestine nuclear weapons programs in Iraq, and subsequent discovery of undeclared nuclear material in the DPRK, demonstrated the need to handle undeclared material and activities. Whereas the standard safeguards agreement provides routine access only to specific “strategic points” in declared facilities, an Additional Protocol to the safeguards agreement provides the IAEA access to any place at a nuclear site and to other locations where nuclear material is, or may be, present. Under this protocol, which has been concluded with some 100 states, states are also required to provide the IAEA access to locations that are, or could be, engaged in activities related to the nuclear fuel cycle. The IAEA has a far-reaching mandate to analyze and interpret the results of its inspections and to interact with states to clarify uncertainties. The five regional treaties that created nuclear-weapon-free zones in Africa, Latin America, the South Pacific, South East Asia, and Central Asia are important as they prohibit nuclear weapons in a very large part of the world. Cooperation within the regions and the safeguards agreements between the states and the IAEA are key verification provisions in these treaties, which also prohibit nuclear testing in the zones. Compared with the above treaties, the CTBT has an extensive and unprecedented international verification regime to support countries in verifying compliance with the treaty. The global international monitoring system is unique among the treaties and so is the provision of requesting and conducting an OSI of a large area anywhere on the territory of a state party. Why is this? For the CTBT, verification has been a key issue ever since the test-ban discussions started, and it ended up with the most farreaching provisions of any treaty. A great deal of the expert work on the global monitoring system, including the activities of the Group of Scientific Experts at the Conference on Disarmament (CD), took place at the height of the Cold War, when mistrust increased the need for verification. A nuclear explosion is a significant and observable event that lends itself to verification. The countries having nuclear weapons in particular demanded rigorous provisions to ensure that others would not test when they were giving it up. Why is the TS given a mandate different from that of the IAEA? The TS is responsible for the installation and operation of monitoring stations, receiving IMS data, and providing countries with data and bulletins resulting from its analysis of IMS data. Why was the TS not given the

184

Verifying the CTBT—A State Perspective

responsibility of assessing observed events? As noted in Chapter 2, seismic events alone exceed 100 per day, and a large number of those would not be easily classified as having a natural origin from the observed signals alone. If the task of clarifying events was to be given to an international organization, that organization would have to act with impartiality and treat all observed events in the same way, and geophysical events all over the globe would need to be investigated to an extent that would cause the system to break down. The analysis of observations must be put into a political perspective, and this can be done only by countries, unless an international organization is given a far-reaching political mandate. This is the rationale for the procedure established in the CTBT: states have the task of making assessments in relation to possible noncompliance and of making decisions to seek further clarification through a consultation and clarification process. States can also request an OSI. This responsibility makes it important for countries to acquire expertise regarding verification measures, rather than expecting the TS to provide answers about compliance or possible non-compliance. There is still a widespread misunderstanding that the TS will make such assessments and ring the bell when something is observed that merits further clarification. This misconception has likely been fueled during the CTBTO Preparatory Commission (PrepCom) phase, when countries have expected and requested the Provisional Technical Secretariat (PTS) to report its interpretation of observed events following, in particular, the two DPRK nuclear explosions in 2006 and 2009. In regard to monitoring, the CTBTO and the TS thus have the task to support states by operating the IMS and the International Data Center (IDC) and by providing high-quality data and products. In regard to OSI, the TS assumes more responsibility, as the Director-General of the TS must ascertain that requests meet the requirements for an OSI, seek clarification from the state party to be inspected, and issue the inspection mandate and handle communications regarding inspection reports. As noted, the CTBTO Executive Council must approve a request by a state party for an OSI. Then the TS is responsible for organizing and conducting the OSI, with inspection team members coming from states parties and the TS. The inspection team reports to the Director-General, who forwards the report to the Executive Council, which can make a decision to prolong or to conclude the inspection. The Executive Council will review the inspection report and “shall address any concern as to: (a) Whether any non-compliance with the Treaty has occurred and (b) Whether the right to request an on-site inspection has been abused.” In cases of noncompliance, the Executive Council can defer the issue to the Conference of the States Parties.

185

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

7.1.2  Deterring Non-Compliance An important purpose of a verification regime is to deter countries from conducting clandestine activities. In his report on findings and recommendations concerning the CTBT, the former chairman of the U.S. Joint Chiefs of Staff, General John M. Shalikashvili (Figure 7.1), noted: When evaluating verification, the probability of deterring violations is as important as the probability of detecting them. Thus, the value of a verification system extends well past the range where a monitor has high confidence of detecting, identifying, locating and attributing a violation, and down into the grey area where a potential evader lacks certainty about the likelihood of discovery (Shalikashvili 2001). Deterrence is very much a question of political judgment: How big a risk is a country prepared to take to carry out an activity in breach of a treaty— in the case of the CTBT, to carry out a clandestine nuclear explosion? There are likely a number of considerations behind such a decision: What is the need or benefit of carrying out such a clandestine test? What is the likelihood of it being detected? Finally, what are the consequences if detected? The answers to such questions will depend on the political and military situation, and different states will no doubt look at the issues involved differently. Countries less well integrated in the global community might be prepared to take larger risks. From a technical point of view, as discussed in Chapter 2, a monitoring system such as the seismic network has a considerably lower threshold at 10 percent or 1 percent detection probability than at the 90 percent probability usually used to describe confident detection. The monitoring of xenon is a novel technique and provides a strong deterrent. The release of xenon from underground explosions in different Fig. 7.1  Gen. John M. Shalikashvili, former chairman of the U.S. Joint Chiefs of Staff, Special Advisor on the CTBT to President Bill Clinton.

186

Verifying the CTBT—A State Perspective

geological environments and under different circumstances is still not fully understood and is difficult to predict even for the country conducting the explosion. This uncertainty in estimating whether or how much xenon will be released from a test acts as a deterrent to a potential evader. In addition, national technical means in the hands of several states present a considerable deterrence, as the evader may not know or be able to estimate the capabilities of other states. Some of these means, especially the non-technical ones, may be related to the operation surrounding a test and not to the yield of an explosion. Increased public openness, as seen by reporting of events on the Internet and other media from groups of people and individuals, also adds to the likelihood that clandestine activities may be revealed. Further, whistleblowers from inside the system have, at great personal risks, revealed critical activities in the past. The third element in deterring a country from conducting nuclear tests relates to the consequences it will face if a clandestine activity is detected. The consequences of violating a treaty are not known in advance, with the exception of actions like the one in Article V of the CTBT: “to restrict or suspend the State Party from the exercise of its rights and privileges under this Treaty.” Under the CTBT, the Conference of the States Parties or the Executive Council may refer a violation to the United Nations. Resolute enforcement is likely to effectively induce deterrence even under uncertain detection capabilities, whereas definitively proven noncompliance instances followed by very weak and delayed enforcement measures undermine deterrence value. A number of previous situations illustrate the difficulties the global community faces in enforcing corrective actions to change the behavior of the state in violation of specified treaty terms. A great deal of effort over the years has been devoted to improving the tools available to countries to verify the CTBT and other treaties; less effort has been invested in exploring how the international community should deal with countries in breach of treaties or commitments. Sanctions of different kinds have been a preferred tool and have been applied with varying degrees of success. In recent years they have increased in severity to include such actions as the imposition of financial and banking restrictions, and the international community has become more resolved in enforcing them, with countries previously opposed to sanctions now joining the consensus more often in the United Nations Security Council. Concerted efforts are still required to improve their effectiveness, however. There is no doubt a need for the global community to further develop the ways and means to deal, in a determined and constructive way, with countries in breach of international obligations and norms.

187

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

7.1.3  Verifying Compliance As mentioned, the political nature of verification means that a country may see the need to focus its verification efforts on only a few other countries of concern to it. The ability to verify a treaty might therefore look quite different depending on the political perspectives of individual states. This book has shown that verifying the CTBT from a state’s perspective is quite different from the international global perspective usually discussed. The verification regime of the CTBT is designed to provide a monitoring capability around the world that is as equal as technically possible. The data and the results of the standardized analysis of the PTS/TS aim to give all states a common and solid basis for their verification of events globally. Individual countries may then improve upon this ability by focusing on a small number of countries by using additional data and techniques. Article IV of the CTBT states: “At entry into force of this Treaty, the verification regime shall be capable of meeting the verification requirements of this Treaty.” The IMS is defined by its physical components, which consist of 321 monitoring stations and16 radionuclide laboratories. No efforts were made during the CTBT negotiations to reach agreement on functional verification or monitoring requirements for the IMS, and it would have been very difficult to identify such design criteria, both politically and technically. Normally, when a system is designed, a number of functional performance criteria are established. For a monitoring system, it could include the minimum size of objects or events that should be detected with a certain probability and the maximum acceptable uncertainty in locating these objects or events. There are no such operational performance criteria specified for the IMS. During the negotiations, technical experts made estimates of the expected capability of different station configurations. The experts had different approaches, and a 1 kiloton capability was often expressed as a good target. As has been clearly demonstrated, the currently achievable capability goes well beyond this specific target. There is, therefore, no technical way to certify, at entry into force, that the verification regime meets the verification requirements, as no such requirements have been defined. Countries must make their own assessment and acknowledge that they find the verification provisions adequate by signing and ratifying the treaty. As of June 2011, 182 states had signed the CTBT and 153 had ratified it; by doing so they have demonstrated that from their national perspective the verification requirements are adequately fulfilled. At entry into force of the treaty there are no specific requirements on the extent of the implementation of the IMS or on the readiness of the OSI regime, in part because it was unknown at the time of the negotiations how long it would take for the treaty to enter into force.

188

Verifying the CTBT—A State Perspective

Fig. 7.2  The Vienna International Center, headquarters of the IAEA and the CTBTO PrepCom.

7.1.4  Building Confidence International arms control and disarmament treaties have played an important role in building confidence among countries. This applies both to bilateral treaties on the elimination of intermediate-range missiles and the reduction of strategic nuclear weapons between the Russian Federation/Soviet Union and the United States, and to the global treaties on weapons of mass destruction—the BWC, the CWC, and the NPT. Regional treaties, such as those on nuclear-weapon-free zones and the Treaty on Conventional Forces in Europe (CFE), also have contributed to confidence building in their respective regions. The negotiating process, the related dialogue, and the exchange of information and data already build confidence, and the fact that countries manage to reach an agreement contributes as well. The most important confidence-building elements involve a successful implementation of treaties that may include exchange of specific information; on-site inspections, even at sensitive facilities; and interaction among countries to settle any issues that may arise. Has the CTBT contributed to building confidence, despite the fact that it is not in force? The fact that states possessing nuclear weapons—with the exception of the Democratic People’s Republic of Korea (DPRK)—have renounced nuclear testing and created a moratorium is an important confidence-building element. The by-and-large smooth and constructive cooperation among the states signatories in the CTBTO PrepCom since 1997 (Figure 7.2) is another example of confidence building, as is the

189

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

establishment of the IMS with its more than 300 monitoring stations in 89 countries that will provide data online to all states signatories. The fact that most countries hosting monitoring stations are engaged in an online exchange of data already during the PrepCom phase helps promote confidence. The CTBT includes a provision on confidence-building measures that pertains to man-made large chemical explosions for mining purposes that could conceivably be mistaken for nuclear explosions. Under this provision, states are requested to provide, on a voluntary basis, notification to the TS about the location, time, nature of activities (mining or other purposes), and other details. As a further confidence-building measure, countries may also invite other states parties or members of the TS to visit the sites of these explosions. It is unlikely that uncertainty will prevail on the nature of chemical explosions if all concerned are interested in resolving possible ambiguities. International scientific cooperation related to the CTBT has been going on for half a century and has helped to build a common understanding of the basic issues related to CTBT and in particular of how to verify it. For instance, the work by the Group of Scientific Experts at the Conference on Disarmament was a successful effort to build confidence at the height of the Cold War. It promoted the training of scientists and scientific cooperation among institutions around the world, and a concept of international exchange of verification data was established. The International Scientific Studies (ISS) project and other scientific activities organized by the PTS also contribute to international cooperation, as well as confidence building. The Joint Verification Experiment mentioned in Chapter 2 was undertaken in 1988 to resolve difficulties in estimating nuclear explosion yields of Soviet and U.S. nuclear explosions under the Threshold Test Ban Treaty (TTBT). Those experiments, carried out by the two powers at each other’s test sites, significantly improved the mutual confidence related to nuclear test monitoring. What more can be done? Later in this chapter we suggest that cooperation among countries should be increased; this may take the form of regional centers that engage countries in a region with similar priorities regarding CTBT verification. The creation of such centers would contribute to enhanced regional cooperation, as well as collaboration among such centers in different parts of the world. This would not only increase engagement in the CTBT, but also serve as a strong confidencebuilding measure.

190

Verifying the CTBT—A State Perspective

7.2 Monitoring Because the judgments regarding verification of compliance with the CTBT rest with the states parties, each state must develop a national strategy on how to approach this issue. Such national strategies are likely to be quite different from one country to another and reflect the security concerns of each country and its level of engagement. Some countries or groups of countries are likely to show a high ambition to verify the treaty and will expend considerable effort and resources, whereas others may do very little or nothing. As noted above, for political and security reasons, a country is likely to focus its attention on one or a few countries of particular concern. It also is possible that groups of countries will decide to develop a common strategy for CTBT verification, as discussed below. The most likely monitoring targets within identified countries of concern would be areas suitable for conducting clandestine underground nuclear testing. By concentrating efforts on precision monitoring of such areas, a state will achieve a high monitoring capability in the selected areas. Countries might also be interested in monitoring the oceans for possible nuclear explosions in or above these vast areas, although these environments are less likely to be used for clandestine testing. Monitoring involves three important steps: detecting an event, locating it, and determining if it is of natural origin or if it requires further clarification. Different technologies have different and partly complementary capabilities to contribute to carrying out these steps. 7.2.1  The International Monitoring System (IMS) The IMS was created to provide all states parties with reliable, authenticated information and data to facilitate their monitoring of the CTBT, and it is living up to that goal. To date more than 80 percent of the IMS stations have been installed, and they have proved that they can produce high-quality data and that they possess detection capabilities that in many cases surpass what was expected. There is no doubt that a capable IMS station network will be operational at entry into force of the treaty, whenever this occurs. The IMS global networks of radionuclide, infrasound, and hydroacoustic stations are unique, and few such stations exist outside the IMS. This means that these networks provide data that no country can gain access to on its own. The IMS seismological stations are among the world’s most capable. This especially applies to the array stations, which hardly exist outside the IMS. Yet the IMS seismic network constitutes only a small part of the total number of seismological stations deployed worldwide.

191

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

The many non-IMS seismological stations, taken together, can provide detection and location capabilities significantly beyond those of the IMS in many parts of the world; however, the seismological component of the IMS in itself is a valuable and powerful monitoring tool. Under the CTBT, the data and information produced by the IMS will have a number of inherent advantages. IMS data will have a high credibility as they are produced by a system established for the purpose of verifying the treaty that will be operated by the TS in a fully transparent way, with the cooperation of the states parties hosting the monitoring stations. The IMS consists of high-quality stations built to specifications agreed to by the PrepCom, and the TS will certify that each station operates according to these specifications. IMS data are convenient for states to use, as any observations from IMS stations around the world will be provided by the TS in a standardized format. Each station’s data are also authenticated, to ensure that they are not manipulated when sent from the station to the TS. The IMS data will be routinely analyzed by the TS following agreed upon procedures, which provides states with a convenient overview of what has been observed. The IMS data will be available to all states parties, which may facilitate a dialogue among countries on any observed event. Since most of the IMS stations have been installed and certified, many of the characteristics above are already apparent in the testing phase prior to entry into force. As the main concern most likely will be to monitor and deter clandestine underground explosions, seismological and radionuclide observations, particularly of xenon, will be key. The basic level at which a country can monitor the CTBT is to use the results presented by the TS in its bulletin, which is based on the analysis by the TS of IMS data. States can request that the TS obtain data on a selection of events occurring in defined areas of interest to that state. This will facilitate a national review, but it will not improve the monitoring capability compared to the global bulletin, as it will be only a selection from that bulletin. The TS is tasked to analyze data, according to agreed upon procedures, to detect and locate events from all over the world in an even-handed way and report daily to states parties. The organization is not mandated to focus its attention on a particular country or region or to optimize its procedures accordingly. An evaluation of the seismological IMS network, with data from 80 percent of the stations now available and analyzed by the PTS in the agreed upon standardized way, shows that its detection capability is significantly better than 1 kiloton globally and is below 0.1 kiloton in most parts of the Northern Hemisphere. These numbers refer to explosions conducted in solid bedrock. If explosions are carried out in soft

192

Verifying the CTBT—A State Perspective

material such as alluvium, which is present, for example, at the Nevada Test Site (NTS), or in large cavities, the seismic signals will be weaker and the threshold correspondingly higher. Studies presented at the ISS09 Conference provided a coherent picture of the accuracy of event location from seismic data. Half of the PTS locations currently are within some 20 kilometers of the true source locations and the other half has larger errors. For comparison, the 1,000 square kilometers allowed for an OSI corresponds to a circle with a radius of 18 kilometers. The third step in monitoring is to characterize observed events to determine if they are of natural origin or related to known human activities, or if they need further clarification. Using global criteria to identify, or screen out, seismic events that are of natural origin has not proven successful, and the results are likely of limited use to states parties. Less than half of the events reported by the PTS can be screened out as earthquakes using the global criteria indicated in Annex 2 to the protocol of the treaty. Given the inhomogeneities of the Earth and the corresponding large variations in seismic signal propagation, seismic event characterization should be approached on a regional scale, and states may be able to develop criteria to characterize seismic events in their regions of interest. The IMS xenon observations are also routinely reported by the PTS. The likelihood of observing xenon from an underground nuclear explosion depends largely on local conditions in the immediate vicinity of the explosion and to a lesser extent on explosion yield. The difficulty of estimating the likelihood of leakage serves as an important deterrent to clandestine testing, but it also makes it impossible to give a good estimate of the xenon detection threshold, in the way provided for seismological observations above. To locate the source of xenon observations or to link such observations to observed seismic events requires modeling of the transport of the radioactive material in the atmosphere. Atmospheric modeling has for different reasons limited precision, giving location uncertainties of hundreds of kilometers. Procedures for screening xenon observations have yet to be developed. It might be difficult to separate xenon observations from medical isotope production plants from those of explosions based on the isotope composition, as both sources are short lived, but tracking the observation to possible sources is a possible screening method. 7.2.2  Precision Monitoring Since a country is most likely interested in monitoring only a limited number of other countries and specific areas, it can embark on what we

193

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

have called precision monitoring, which entails the monitoring state using all assets at its disposal. Data provided by the IMS are one element, and data from selected non-IMS scientific stations, mainly seismological, are another. In addition, a country can use satellite observations and information obtained by other national technical means. Such precision monitoring would advance confidence in monitoring by significantly improving detection and location capabilities and the ability to identify the nature of observed events. It will also increase deterrence and extend it to smaller explosions. To embark on precision monitoring of one or several selected areas, a state or group of cooperating states will need to establish its own analysis capability. The seismological analysis will be based on generally available data from a number of the IMS stations, both primary and auxiliary, and also on data from a selection of well-placed stations that have been established for scientific purposes. By focusing the analysis of all these data on a particular region and using modern data mining methods, the monitoring capability will increase significantly compared to that of the IMS. Detection capabilities even below magnitude 2, corresponding to an explosion yield below 10 tons, or 0.01 kiloton, can be achieved through focused monitoring of selected areas (NAS 2002a, Kværna et al. 2002). Using the “threshold monitoring” technique discussed in Chapter 2, countries can further increase their monitoring capability and continuously estimate the size of the largest event that might go unnoticed in an area, which adds to the deterrence capability. Seismologists have created the concept of “precision seismology” in an effort to increase the quality of seismological interpretation, in particular on a regional basis. This is very much in line with the precision monitoring of the CTBT advocated in this book (Figure 7.3). Approaching event location on a regional scale, the use of calibration data and new location techniques can substantially improve location accuracy in line with the precision seismology concept. Part of the precision monitoring procedure includes calibrating the selected regions to reduce systematic bias in the event locations. In their areas of interest, countries will no doubt achieve the high location accuracies that are needed as a good basis for possible OSI requests. To identify whether an observed seismic event is of natural origin or if it needs further clarification, a country must know the characteristics of earthquakes occurring in the areas of interest. Observations of new events should therefore be assessed and compared with earlier events in a particular region. There are in principle two complementary ways to approach this: numerically, or statistically, and geophysically. As discussed in Chapter 2, the numerical approach is based on a comparison of the

194

Verifying the CTBT—A State Perspective

Fig. 7.3 Seismological stations in Europe and adjacent areas . Large concentrations of such stations in several parts of the globe can be used by states to support precision monitoring . A state can select from available stations to create an optimal network to monitor an area of interest .

observed signals or signal features from a new event with corresponding data from earlier events in a reference data set . In the more complicated geophysical approach, a geophysical model is gradually developed of the actual area, based not only on seismological observation, but also other geophysical data . New observed events can then be compared to that model . Dramatic developments in data analysis and data mining and exploitation have created the tools needed to create a new and integrated approach to analyzing seismological observations . Such a new paradigm for data analysis means that the observed waveform data from all stations in the network selected to monitor an area are analyzed together in an integrated process to detect events, to locate them, and to clarify if they are of natural origin (see Chapter 2 for more detail) . In this process of analysis, the historical database of earlier observations of events in the area made by the selected stations is used as a reference . The IMS xenon network provides an important monitoring and deterrence asset down to low yields, as discussed above . The sensitivity of the xenon network, as of any other system, is limited by the operational performance of the individual stations, the extent of the network, and the background noise . The capability of the individual stations could potentially be improved by making them more robust to reduce downtime, and by increasing the sensitivity by replacing existing sensors with new and more sensitive ones . The IMS noble gas network will comprise 40 stations at entry into force of the treaty, and decisions to add noble gas capability to the remaining 40 stations of the radionuclide network will be considered at the first annual session of the Conference of the States Parties . The xenon

195

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

background is essentially generated by a few medical isotope production facilities; to limit or stop these releases at the facilities is feasible and would significantly increase the ability to detect weak xenon releases from other sources, and discussions with these industries have been initiated. Very few xenon stations exist outside the IMS network, although in the future countries may cooperate to establish additional stations close to areas of interest, to be used as part of their NTM. Because the release of xenon might be delayed after an explosion, and it takes time for the gas to be transported through the atmosphere, there might be sufficient time to deploy equipment in a country where the released xenon gas is estimated to arrive. For example, Swedish equipment was moved into South Korea to detect xenon releases from the first DPRK nuclear explosion, which is a good illustration of bilateral cooperation. Taking measurements with airborne equipment to detect xenon at a close distance from the source is another option. Several countries have such equipment, and xenon observations were made from airplanes after the first DPRK explosion. Monitoring of xenon in national and international airspace could be an additional way for states to increase their monitoring capability. The dispersion of xenon can be analyzed by atmospheric transport models. Such models can be used in “backtracking” mode to trace a particular observation back from a monitoring station to a possible source region, or in “forward modeling mode” to predict the way a release would disperse from a given source. This last application can be used to explore if a xenon observation can be associated with a particular seismic event. The uncertainty in both backtracking and forward modeling is considerable and is likely to remain so even at a regional scale, as the atmosphere exhibits a significant element of unpredictable, or chaotic, behavior. The analysis and interpretation of xenon observations and other radionuclide observations should thus be integrated with additional information such as seismological data, satellite observations, or information from other sources. The possibility for countries to use satellite observations to monitor the CTBT has increased dramatically since the treaty was negotiated. This applies both to optical and to radar observations. The optical systems have resolutions of 1 meter and better; for radar systems the resolution is about 10 meters (Figure 7.4). Radar systems have an all-weather capability and, when operating in an interferometric mode, the ability to detect small deformations of the Earth’s surface that might be caused by an underground nuclear explosion. A number of systems that provide good coverage are currently operational, with repeated observations of a particular area within one or two days. The fact that data from most of those satellite systems are openly and readily available at low cost is a significant development. Computer software is also available to analyze the large amount of data

196

Verifying the CTBT—A State Perspective

Fig. 7.4  The Canadian Radarsat-2, launched in 2007, is a 2,200 kilogram satellite expected to operate for seven years in a 798 kilometer sunsynchronous low Earth orbit. It will provide radar images with a ground resolution of 10 meters (Canadian Space Agency 2011).

involved and to identify changes over time in selected areas. Therefore, satellite observations can be used to monitor an area in a continuous way, to follow the build-up of infrastructure that might be associated with nuclear testing, and to observe activities related to particular events. Such observations can also be used to identify observed seismic events located in areas where no infrastructure for nuclear testing exists. In such cases, the events would most likely be earthquakes. Satellite observations would thus be an essential tool to be used by a country in precision monitoring of selected areas. A close synergy between satellite and seismic observations requires that the seismic events can be located with high precision. In addition to technical monitoring systems, a country also can use other sources of information. These could include technical and other intelligence information that the state normally wants to keep secret, or it could be open-source data. A country is likely to prefer not to use its intelligence data as part of the evidence presented in support of a request for further clarification of an event; it is more likely that such information will be used to initiate and direct a careful search for other observations or technical data that are openly available. The amount of open-source data has increased dramatically in today’s interconnected world, and new data mining methods have improved the possibilities of interpreting and making use of it. The likelihood of obtaining either intelligence or opensource data is probably independent of the yield of an explosion but rather related to the testing operation as such. Chapter 2 discussed the possibility of conducting explosions in cavities as a way to reduce the strength of seismic signals. Given that experience with the decoupling technique is limited to only two nuclear explosions of low yields in explosion-generated cavities, a number of questions must be addressed by a state considering this method to escape detection.

197

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

Would the reduction in signal strength be the same if the cavity was excavated and not created by an explosion, as an explosion not only creates a cavity but also cracks a large area of the surrounding bedrock? Other questions relate to the increased risk of radioactive releases from explosions in cavities. During an explosion in a shaft or a borehole, the high pressure and temperature against the wall of the formed cavity create a glazed surface that helps contain the radioactive material, including the radioactive noble gases. In a decoupled explosion, the very purpose is to reduce the pressure on the wall, so no glazed surface will be formed. This will increase the likelihood that radionuclide gases might escape. Adding these uncertainties to the logistical difficulties of creating a suitable cavity without being detected should make decoupling a less attractive scenario for a potential evader. The capability to detect and deter that can be achieved through precision monitoring very much depends on both the effort a country is prepared to devote and the data and information available. Seismological detection capability depends on the number, quality, and sensitivity of the stations well located to monitor the actual area. In most parts of the world there is a large number of seismological stations operating, from which a country can select a specific set optimal for the monitoring task at hand (see Figure 7.3). To select those stations, many of which are likely to be at a fairly close distance to the area of interest, requires careful analysis of how well signals from events in the area to be monitored can be observed at the individual stations. Not only the stations, but also the analysis procedures, must be tailor-made to fit the actual monitoring task, and modern data mining techniques offer good tools. A country would need to establish a data analysis facility to carry out the necessary routine analysis and also engage experts to analyze and interpret the results. In addition, precision monitoring requires the integration of xenon and satellite observations and information from possible NTM sources in the analysis. Given the high resolution and good coverage from satellite observations, the extent of overhead monitoring of areas of interest is likely to be limited only by the amount of effort that a country wants to commit. Few countries are likely to be prepared to dedicate their own resources to precision monitoring, and countries having similar political priorities with respect to CTBT monitoring might, as an alternative, establish joint analysis centers, possibly on a regional basis. 7.2.3  Wide-Area Monitoring Countries may also want to be assured that no nuclear test explosions take place in the oceans or in the atmosphere above the oceans. Data from

198

Verifying the CTBT—A State Perspective

the IMS global network provide a good basis for such wide-area monitoring. As discussed in Chapter 3, a nuclear explosion in the atmosphere, even of very low yield, is likely to generate radioactive fall-out that will be detected with high probability at large distances. The scenario of radioactivity from an atmospheric explosion being washed out in heavy rain does not apply to radioactive noble gases such as xenon. The location accuracies using radionuclide observations alone are quite low, and such observations provide only a rough indication of where an explosion was conducted. Infrasound observations will provide better, though still not very precise, locations for explosions of yields down to 0.1 kiloton. An explosion over land will present effects on the ground that can be easily detected by satellite observations. If the explosion occurs above water there will be local radioactive contamination but no long-term effects at the site that could be detected at a distance. The possibility of detecting atmospheric explosions over oceans using hydroacoustic measurements has not yet been explored. The United States, and possibly other countries, has established national satellite-based systems to monitor atmospheric explosions. The U.S. systems include instruments to detect several features, such as optical flash, electromagnetic pulse, and initial nuclear radiation, as discussed in Chapter 3. The instruments are deployed on a large number of satellites, including 33 GPS satellites, and in May 2010 the United States launched improved instruments on board a new series of GPS satellites. The monitoring capability of U.S. satellite-based instruments has not been released. As elaborated in Chapter 4, the unique hydroacoustic signal transmission properties in the oceans make it possible to detect very small underwater explosions, on the order of 100 kilograms, or 0.0001 kiloton, at global distances. Hydroacoustic location accuracy depends on where an event occurs in relation to the few observing hydroacoustic stations; it is generally on the order of 100 kilometers, but could be several times larger for an event unfavorably located in relation to the stations. A nuclear explosion, even of low yield, conducted in the ocean will generate hydroacoustic signals that are orders of magnitude stronger than what is normally observed from other sources. A strong underwater explosion will therefore be easily recognized. It is thus highly likely that even a very small nuclear explosion conducted in the atmosphere or under water in the oceans will be detected. But who did it? This is a question that has recurred when discussing such a scenario. To conduct a nuclear test explosion in or over the ocean, perhaps in a remote location, and to collect data that make the effort worthwhile, is likely a substantial and complex operation. It is highly doubtful that such an operation will escape the NTMs, including high-resolution satellites,

199

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

available to countries today, and the state responsible for such an activity will be identified from the observed logistical operation. This will serve as a credible deterrence, and it is doubtful that any country would even consider conducting a clandestine explosion in either of these two media. This explains why the verification discussion has focused for decades on underground explosions. 7.2.4  Non-Governmental Monitoring The IMS data are not in the public domain; however, some countries make data from IMS stations on their territory openly available. A growing amount of information that can be used to monitor the CTBT is generally available, as discussed below, and thus countries have comparatively less privileged information than they did previously. It is also possible for non-governmental organizations (NGOs) and institutions to engage in CTBT monitoring. Previous examples of direct involvement by NGOs in a disarmament treaty include the Ottawa Process, which led to the Mine Ban Treaty championed in 1997 by the International Campaign to Ban Landmines (ICBL), the Committee of the Red Cross, other NGOs, and a group of countries led by Canada. The ICBL and other NGOs have also played a role in the implementation of the treaty by engaging in mine clearance in many parts of the world. During periods of nuclear testing, NGOs and scientific institutions engaged in monitoring and reporting on the nuclear tests that were conducted, which created an increased openness and public awareness about nuclear testing. Such monitoring also created confusion sometimes when naturally occurring earthquakes were mistakenly identified as nuclear tests. A number of geophysical institutions, such as the U.S. Geological Survey, rapidly analyze observed seismic events and put the results on their websites within hours of an event. In many parts of the world, such institutions present compilations of observed seismic events that are more comprehensive than those presented by the PTS because they may use data from many thousands of stations. This creates a rich body of material available to those who want to look for seismic events in specific areas. More basic data in the form of recordings from a large number of seismological stations are available online from data archives such as the Incorporated Research Institutions for Seismology (IRIS 2011). Making use of this data to determine if an observed event is an earthquake or an explosion, however, is a more complex task requiring seismological expertise, reference observations, and appropriate software.

200

Verifying the CTBT—A State Perspective

In addition to seismological data, high-resolution satellite photos are also generally available, but analyzing such photos is not a trivial undertaking and requires expertise and experience. That said, numerous satellite observations of features related to issues of political concern, such as the DPRK nuclear explosions and the Iranian nuclear program, have been published by independent institutions (e.g., Albright and Brannan 2006, Jasani et al. 2009). In addition, a great deal of open-source data is by definition available to the public; however, it would likely require a fair amount of effort to collect and analyze such data even if the focus is on one or a few countries only. Large-scale leaks of government documents through Wikileaks illustrate the difficulties of keeping documents or information out of the public domain. Very few, if any, radionuclide observations are publicly available. Within a month after the 11 March 2011 earthquake and tsunami that damaged the Fukushima Daichi power plant in Japan, 35 IMS radionuclide stations in both the Northern and Southern Hemispheres recorded radioactive materials from the leakages at this power plant. Following the event, the PTS shared IMS radionuclide data and analysis results with the IAEA and the World Health Organization (WHO); it remains to be seen whether this development will result in a greater openness on radionuclide data in the future. Information is therefore available that would make it possible for specialized NGOs and non-governmental institutions to engage in monitoring. If there are reasons to believe that a country might conduct a nuclear test, it is likely that some organizations or institutions might start monitoring activities. This could increase public interest in the country concerned and thus increase deterrence. It might also lead to misinterpretations that could be difficult to handle politically. 7.3  Consultation and Clarification The CTBT encourages states to “make every effort to clarify and resolve, among themselves or through the Organization, any matter which may cause concern about possible non-compliance.” States are obliged to provide a clarification within 48 hours of a request for information, and countries may also ask the Director-General of the TS or the Executive Council to assist in obtaining clarification. The requesting country can also present a request for an OSI to the Executive Council and the DirectorGeneral without going through the consultation and clarification process. What might be the practical use of these procedures? If the event of concern is a clandestine explosion, it is not likely that consultations will contribute to resolving the situation. There is, however, one kind of event

201

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

that might be resolved using consultation and clarification, and that is a natural or man-made non-nuclear event that might be mistaken for a clandestine nuclear explosion. Such man-made events would include large chemical explosions for mining purposes, as discussed earlier in this chapter in relation to confidence-building measures. Naturally occurring events, notably shallow earthquakes, might present a bigger challenge. How can a country prove that an observed seismic event actually was an earthquake, if another country for some reason expresses concern? In many areas of the world there is a high concentration of earthquakes, including at shallow depths. Chapter 2 discussed the large number of earthquakes observed in Iran (Figures 2.10a and 2.10b), and large numbers of earthquakes also occur in the western United States and parts of Russia and China. A credible way of confidently identifying a seismic event as a non-explosion is to determine that it has a depth exceeding what is practical for nuclear testing. For logistical reasons it is most unlikely that anyone would conduct a test at a depth beyond 2–3 kilometers. Chapter 2 discussed the difficulties of estimating depths of shallow earthquakes without data from close-in stations. Some countries have dense in-country station networks that can contribute to locating and estimating the depths of earthquakes with high precision (see Chapter 2, section 2.2.3). A key question then is how can such national networks be used, and how credible are such data if presented by a country to resolve an issue? The CTBT contains provisions for “cooperating national facilities.” This provision allows countries to make arrangements for the use of national monitoring stations that are not part of the IMS. Such stations will be operated and paid for by the hosting states and certified by the PTS/TS as fulfilling the operational requirements of an IMS station, including data authentication. Data from such stations could then be used to support a process of consultation and clarification. To date, no cooperating national facilities have been registered with the PTS. The operational quality of national seismological networks is in most cases very high, and data from such networks would be valuable in supporting a cooperative effort to resolve issues related to naturally occurring earthquakes. To increase the credibility of such data it could be useful to consider some mechanism to voluntarily provide information on these national networks and their capabilities. Cooperation in developing a clear picture of worldwide seismicity, using data from national networks, the IMS, and other external stations, would be a confidence-building measure that could significantly improve the possibility of resolving uncertainties that might arise regarding future earthquakes.

202

Verifying the CTBT—A State Perspective

7.4  On-Site Inspections The on-site inspection (OSI) regime provides countries with a powerful tool to detect and deter violations of the CTBT. The regime has strong political, technical, and operational elements and provides for intrusive inspection of an area of 1,000 square kilometers in another state party or any area beyond the jurisdiction or control of any state. The many details in the treaty regulating an OSI are discussed in Chapter 5. It is worth noting that a state can use whatever means are at its disposal to monitor the CTBT, but an OSI can be initiated only within the frame of the treaty provisions and when approved by the Executive Council of the CTBTO. All countries are thus in the very same position when it comes to the application of the OSI regime. An OSI can be requested by a state party to clarify whether a nuclear weapon test explosion or any other nuclear explosion has been carried out in violation of the treaty and to gather facts that might assist in identifying the violator. A country requesting an OSI must be able to present credible evidence to the Executive Council in support of its request in order to gain sufficient support for an affirmative vote. Composed of a specified number of states parties from six geographical regions, the Executive Council is required to approve a request for an OSI by at least 30 out of 51 votes. Members are selected from within each group, taking into account their technical engagement and contributions to the annual budget. This means that in practice the five nuclear weapon states recognized by the NPT—China, France, the Russian Federation, the United Kingdom, and the United States—will always be members of the Executive Council. Further discussion of the issues involved with the Executive Council is contained in Chapter 5. To be able to consider the technical evidence presented for an OSI, it is important that the states represented in the Executive Council have the necessary scientific and technological expertise. Establishing this expertise globally should be in the interest of all countries, and this chapter proposes increased international cooperation among countries to achieve that goal. If a request for an OSI is approved, the country to be inspected is obliged by the treaty to accept it. Interesting questions follow from this obligation: Would a country accept an inspection if it has carried out a clandestine test and the requested inspection area includes the site of the explosion and the evidence presented suggests that the OSI likely will locate the test site? Rather than being caught in an inspection, would it not be preferable to the country to reject the inspection and face the consequences? This might be a tempting option if the consequences of not obeying the provisions of the treaty are not strong enough. Enforcement is therefore a crucial element

203

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

in creating strong deterrence and establishing and maintaining the treaty credibility. As discussed in Chapter 5, if the country to be inspected accepts an inspection, there are three possible scenarios: there has been a clandestine test within the requested inspection area; there has been a clandestine test, but the requested inspection area does not cover the explosion site; and, finally, there has been no clandestine test—the triggering event was a natural event, most likely an earthquake. The third scenario might well be the most likely, as a large number of earthquakes occur every year in countries that have or might have nuclear weapons. To identify an observed event as being of natural origin might be a difficult task even when all parties involved, including the inspected state, show good will. Radionuclide measurements will be important to establish the absence of observable radionuclides, in particular xenon, but they will not be conclusive with respect to the nature of the triggering event. Visual observations, including those made from aircraft or satellites, may, in certain areas, rule out that the event was an explosion due to the lack of needed infrastructure in the area. In other areas, such as former testing grounds, it might not be possible to draw such a conclusion. To estimate the depth of an observed seismic event with high accuracy, either from observations of the event itself or from possible aftershocks, could provide convincing evidence that an event was an earthquake. Precise depth estimations of events at shallow depths require stations close to the event, and some countries operate a national seismic network that could provide important data. It would be of value to explore how such observations could be used in the analysis of observed events in a credible way. Countries might well be faced with a situation in which no clear evidence of a clandestine test is observed during an OSI and no other origin of the event can be established. How would it be possible to leave such an inspection with enough confidence? There are a number of issues to be addressed regarding how to handle inspections of natural events in a way that does not create unfounded suspicion and weaken the credibility of the treaty. Countries should analyze this scenario and the corresponding issues in order to be prepared to handle such situations at a national level, during deliberations with the Director-General of the TS and in the Executive Council. The second scenario involves an observed triggering event that is a clandestine nuclear explosion that has been mislocated in such a way that the requested inspection area does not contain the site of the explosion. This might well be a reality, as 50 percent of the events currently reported by the PTS have location uncertainties exceeding the dimensions of the inspection area. In this case an OSI will not detect any

204

Verifying the CTBT—A State Perspective

evidence of a clandestine test itself; however, there might be observations of infrastructure associated with a testing operation or of xenon or other radionuclides that might be dispersed from a test. Such radionuclide observations may not be conclusive, but might instead increase the uncertainty about the nature of the event and may indicate that the event took place outside the requested inspection area. The question then is if a second request for an OSI can be made, based on the same triggering event, if new evidence is presented. The CTBT does not set a time limit for submitting a request for an OSI, nor does it explicitly prohibit a second request, as long as it is not simultaneous. It is likely that strong evidence would have to be presented to the Executive Council to obtain approval for a second OSI, but there seems to be no formal obstacle in the treaty. For an OSI to be successful it is essential that monitoring data locate the triggering event with high accuracy and thereby provide a good basis for selecting an inspection area that covers the site of the event. The more precisely the triggering event can be located, the greater the probability that the inspection will be successful. This is why we argue in several places in this book that countries should engage in precision monitoring to obtain event locations with high accuracy, in addition to expanding their capabilities to detect weak events. An event location with a small uncertainty matched with high-resolution satellite photos will provide a good basis to define not only an inspection area but also to establish an initial inspection plan. A clandestine test that has been located with a high degree of precision within the inspection area is likely to be revealed during an OSI. Given all the inspection tools and the substantial inspection time available, it is highly unlikely that all evidence of a testing operation can be concealed if the inspection team can focus on a small area. An OSI is a large-scale and fairly complex operation when seen in the context of disarmament treaties. Compared to other operations such as peace keeping or relief operations, however, it is relatively small. The experience from field tests, in particular a month-long exercise carried out in Kazakhstan in 2008 (IFE 08) shows that the logistics concept should be further developed to draw inter alia on the wealth of experience available on mobile laboratories for other applications. A container-based system—in which modules can be transported to the deployment site by aircraft, helicopters, and trucks, and the equipment is already assembled, connected, and ready to work—would significantly speed up deployment. A number of procedures and techniques can be applied during an OSI to search for traces of not only an explosion but also the testing operation. The techniques discussed in Chapter 5 include, among others, close-in visual inspections from overflights and from the ground, on-site measurements of possible seismic aftershocks associated with the cavity

205

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

region, observation of xenon and other nuclear products, and a number of geophysical measurements. Once a likely explosion site has been located and Executive Council approval has been granted, drilling can be conducted to confirm the existence of an explosion cavity and to take radionuclide samples. To provide for an efficient OSI, it is important to understand the way different technologies can contribute individually and in synergy with each other. A systemic approach to an OSI, in which the different technologies are applied in an optimal way, taking into account conditions on the ground, is essential to support the progress of an inspection. Operational analysis, used frequently within military organizations, is a good tool to plan and replan an ongoing OSI, taking into account the observations made. To carry out an OSI is a demanding task, and the composition of the inspection team is critical to a successful outcome. Countries have an important responsibility to identify suitable experts and make arrangements for their rapid deployment following a decision to conduct an OSI. Experts in different technologies must be the backbone of such a team. The team leader, who should have experience from international negotiations, needs one or two persons to assist in relations with the inspected state and with the TS. The operational work should be directed by an operational manager, assisted by a small group of people experienced in the interpretation and integration of the information obtained and in creating a strategy for the progressing inspection. A fair amount of work has been accomplished, but significant efforts still need to be undertaken to finalize the preparations to conduct an OSI at entry into force of the treaty. The experience and knowledge are available in many countries, in particular in the nuclear weapon states. Thus, to advance the preparation for an operational OSI regime, countries with experience in nuclear testing and in the use of the different OSI technologies at nuclear test sites should contribute their expertise. States also should cooperate in the PrepCom to finalize a draft operational manual for the conduct of an OSI and a list of equipment to be used. These two documents will be adopted by the first Conference of the States Parties. If countries are prepared to make OSI a priority and to provide the PTS with the proper guidance and support, it should be possible to establish a workable OSI regime within a fairly short time, possibly within one or two years. 7.5  Ready for Entry into Force? The entry into force of the CTBT is clearly defined in the treaty: “This Treaty shall enter into force 180 days after the date of deposit of the

206

Verifying the CTBT—A State Perspective

instruments of ratification by all States listed in Annex 2 to this Treaty.” The entry into force is thus de-linked from any considerations of the readiness of the verification regime. The PrepCom is, however, requested to present a report on the operational readiness of the regime to the initial session of the Conference of the States Parties. This is likely to be a formal document. Readiness at entry into force can, from a state perspective, be considered at three levels: a “common house” for all states parties, a national level, and an area of cooperation among like-minded states. The CTBTO and its TS, with the IMS and the OSI regime, is a “common house” that serves as a common resource to all states parties, which share the responsibility to build and sustain it. In addition, countries should establish their own verification resources that match their national priorities and needs. Countries may also establish cooperation with other countries that share common political priorities on CTBT verification. Such cooperation, perhaps on a regional basis, could increase global engagement and make verification more cost effective for countries. 7.5.1  The CTBTO—A “Common House” for States Parties The initial session of the Conference of the States Parties to the treaty, to be convened no later than 30 days after entry into force of the treaty, will mark the transfer from a provisional to a permanent organization. The PrepCom will hand over an operational IMS to the permanent treaty organization. The IMS has been in test operation for many years during the PrepCom phase and there is no doubt that a capable IMS will be available at entry into force, whenever this happens, even if a few stations may still need to be installed. The PrepCom will have prepared instructions, in the form of operational manuals, which will regulate in a technical way the operation of the different elements of the IMS and how to analyze the data. Drafts of such manuals have existed for a long time and have been tested, evaluated, and improved. There is little doubt that the PrepCom will be able to present these documents for adoption at the initial Conference of the States Parties. The PrepCom shall also “make all necessary preparations, in fulfilling the requirements of the Treaty and its Protocol, for the support of onsite inspections from the entry into force of the Treaty.” As discussed earlier in this chapter, an important remaining step for the PrepCom is to significantly improve operational readiness to conduct an OSI. In addition, two OSI-related documents shall be prepared for approval by the initial session of the Conference of the States Parties: an operational manual containing all appropriate legal, technical, and administrative procedures and a list of equipment for use during an OSI. The preparation

207

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

of this operational manual has been a major endeavor for the PrepCom, and progress has been slow. Given the political will and engagement by all states concerned, it should be possible to finalize an OSI operational manual in a short time, possibly within one or two years. The same applies to the list of equipment, even if the usefulness during an OSI of some of the equipment considered has not yet been demonstrated. At entry into force of the CTBT, the PTS shall be transformed into a permanent organization, the TS. The PTS was established in 1997 and has functioned for a much longer time than originally anticipated. A substantial amount of experience has been gained that will be useful in moving into a permanent international organization. An external organizational review of the PTS (CTBT/PC-24/INF.9 2005) and external reviews of the individual programs within the CTBTO PrepCom were carried out in the period 2000–2003 (CTBT/WGB-14/INF.3 2000, CTBT/WGB-17/INF.3 2001, CTBT/WGB-21/INF.5 2003). These reviews recommended a number of actions to enhance the effectiveness and cost efficiency of the PTS. To prepare for the transfer of the PTS into the TS at entry into force, it would be prudent to take another look at how the organization works and is managed. Such a review should also address the handover from the PTS to the TS and the need to modify the organization and its method of work when moving from implementation and testing to operation. The evaluation should include a review of the rules and regulations that guide the work of the organization and a review of the desired profile and competence of the personnel. Preparing the ground carefully for the TS would be a good investment to ensure a smooth and cost-effective operation after entry into force. The TS will face the challenge of both sustaining a high-quality technical operation to produce products on a par with national or international institutions and maintaining a highly qualified staff on a long-term basis. This challenge is going to increase over time, as the CTBT verification system is a static one as defined in the treaty, whereas scientific systems are dynamic. To maintain its credibility and relevance and to be an attractive workplace for experts in different fields, the TS will need to stay attuned to scientific and technological developments and incorporate, within the limitations of the treaty, new techniques and procedures as they become available. The scientific community is an indispensable source of scientific input and expertise without which the TS over time will become increasingly outdated. The CTBTO, and before that the PrepCom, must therefore develop a strategy for the TS to interact with the scientific community. Such a strategy might be addressed within the frame of the review mentioned above.

208

Verifying the CTBT—A State Perspective

7.5.2  States Parties Many countries have established a national data center to receive data and information from the PTS during the PrepCom test and evaluation period. Few countries, however, have established the knowledge base or the facilities to be able to monitor the treaty based on the data to be reported by the TS, or to build up the resources needed for precision monitoring. Few countries outside the established nuclear weapon states have such resources, and many states that had significant verification research resources and programs at the time of the CTBT negotiations have reduced or dismantled those activities. To establish and maintain, at a national level, the expertise and the technical facilities to form an independent monitoring capability requires a determined effort and allocation of resources. It also requires that this national activity, however it is organized, maintain a close connection to the scientific community to benefit from ongoing scientific and technological developments. It is likely that many countries will refrain from establishing any national capability to monitor the treaty after entry into force. This is understandable, as the political interest is likely to focus on new additional measures to strengthen the non-proliferation regime and promote international security. Still, it will be unfortunate if only a very small number of countries has the knowledge base to carefully and independently assess and address events that might come up for consideration in the Executive Council, where countries from all parts of the world will be present. To facilitate a factual and constructive discussion, it is important that the members of the Council are not dependent on technical and scientific assessments and judgments from only a few states. The members of the Council will need independent and well-founded judgment from different parts of the world to underpin the political discussion. A well-informed Council, in which the members have access to expert advice, will be in the interest of all states parties. There is thus a great need for countries around the world to address this situation and decide on their level of ambition and contributions to the verification of the CTBT. Because it takes time to establish the necessary expertise and facilities for analysis, it is essential that countries take action now, either nationally or in cooperation with other countries. 7.5.3  Regional Cooperation An attractive alternative to establishing an analysis and assessment capacity at the national level could be to do so in cooperation with other countries in a region that share a common political view on the priorities of CTBT verification. A basic level could be to create a joint knowledge

209

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

base, in the form of a group of experts, that could help countries in the region to analyze and assess the data and information reported by the TS and help identify events that might be of concern. Such a group of experts could also provide scientific advice to regional members of the Executive Council on events that might be brought to the Council. It might also explore how the monitoring data could be used for other, non-verification purposes that might benefit the region. A more ambitious approach would be for a group of countries in a geographical region to establish joint facilities and expertise to engage in precision monitoring. The cooperating states would need to establish common political monitoring priorities and decide on which states or areas to focus. A regional center is a more extensive undertaking and would require facilities and personnel, as discussed above, but it would still be a cost-effective way of doing the work by combining resources from the countries involved. A regional center could, in the same way as a national center, use IMS data and other available scientific data, primarily seismic, satellite observations, and other generally available information. The extent to which a regional center could use restricted information from the NTM available to individual countries depends on how far security cooperation in the region has advanced. Regional centers could also explore how verification-related data could be used to address other issues of regional interest such as environmental protection and natural disaster mitigation. Regional centers should have close links to universities and other research organizations in the region and to the broader scientific community. They could also serve as regional training centers for the sciences involved. The European Union (EU) has established a common foreign and security policy (CFSP). Under this umbrella, the EU could establish a European CTBT monitoring center to support the EU and the EU member states in precision monitoring of states and areas in accordance with their CTBT monitoring tasks and priorities. The EU already has a similar center, Fig. 7.5 The EU Satellite Center (EUSC) in Torrejón de Ardoz, near Madrid. The EUSC is an agency of the Council of the European Union that supports the decision making of the European Union by providing analysis of satellite imagery and collateral data.

210

Verifying the CTBT—A State Perspective

Fig. 7.6  Meeting of a special session of the Agency for the Prohibition of Nuclear Weapons in Latin America and the Caribbean (OPANAL), an intergovernmental agency established in 1967 to ensure that the obligations of the Treaty of Tlatelolco are met.

the EU Satellite Center (EUSC 2011), which was incorporated into the EU in 2002 (Figure 7.5). This center supports, in accordance with the European Security Strategy, the decision making of the EU in the field of CFSP. The EUSC could support, be linked to, or be part of a CTBT monitoring center, contributing satellite data and analysis. Establishing such a joint EU center would be a cost-effective way to set up a credible European CTBT verification capability. Such a center might be a decentralized organization, drawing on existing national monitoring organizations. An EU CTBT monitoring center could also serve as a good model for similar centers in other regions. A possible approach for such centers might be to link them to the nuclear-weapon-free zone treaties and the organizations established for these treaties. The Treaty of Tlatelolco, which covers Latin America, has its headquarters, OPANAL, in Mexico City (Figure 7.6). The Treaty of Rarotonga covers the South Pacific and has located its secretariat with the South Pacific Forum Secretariat headquartered in Suva, Fiji. The Treaty of Bangkok, which covers South East Asia, placed its secretariat with the ASEAN Secretariat and has headquarters in Jakarta, Indonesia. The Treaty of Pelindaba covers the African continent and will headquarter its secretariat with the African Commission on Nuclear Energy in South Africa. The CTBT regional centers could build on cooperative efforts among specialized institutions in several countries in the region and be created gradually. The initial aim might be to analyze and assess the data and information reported by the PTS/TS, and over time an independent capability can be developed to implement precision monitoring of areas of concern to the countries in the region. The establishment of regional centers might be facilitated by fostering international scientific cooperation among interested countries and potential regional centers to develop a common knowledge base and establish the methods and procedures necessary for precision monitoring. Such cooperation might also help create a renewed interest and engagement in the CTBT among

211

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

countries globally and help build mutual understanding and confidence among countries on how to apply the CTBT verification regime. 7.6  Detect and Deter The CTBT has already been on a long and slow journey. It has taken much longer than expected to move toward entry into force, and hurdles still remain en route. That said, much has also been accomplished. The political winds have changed and seem to favor a decline in the role of nuclear weapons and a priority on entry into force of the CTBT. The tools available to states to detect possible violations and to deter clandestine activities are constantly improving far beyond what many expected when the treaty was opened for signature. The IMS, a common resource to all states parties, is approaching full implementation and provides a good monitoring capability globally. The expected synergies among the technologies of the IMS are coming to fruition. Countries can, with precision monitoring of selected areas of interest or concern, reach capabilities well beyond that of the IMS. The OSI regime is an important element to enhance deterrence and to provide clarification of events of concern. Countries should make further joint efforts to prepare for efficient inspections with well-equipped and competent inspection teams. The continuing rapid developments in science and technology make it possible for countries, nationally and in international cooperation, to further develop verification tools, and these developments will provide gradual improvements to the individual monitoring technologies. Together with new methods to analyze, integrate, and exploit data and information, they will make it possible to establish new concepts and paradigms, providing dramatic improvements in capabilities. This scientific and technological development will continue, and countries will be able to harvest and apply the results to further improve their verification capability and their confidence in the treaty. For potential violators of the treaty, we do not see comparable prospects to improve on capabilities to evade detection. Hence, scientific developments will make it increasingly difficult to get away with non-compliant behavior. Countries will likely have different priorities and requirements as well as widely diverse national capabilities with respect to CTBT monitoring. A key challenge for countries is to define their national strategies and priorities and step up their verification activities to meet those needs. Few states have the national resources and expertise needed to efficiently verify compliance with the treaty. Cooperation among like-minded states could be a cost-effective way to increase global engagement and establish efficient monitoring of areas of common concern. The time has come for

212

Verifying the CTBT—A State Perspective

countries to act on their priorities, either individually or in cooperation with other, like-minded countries, and to re-engage in the CTBT verification in a way similar to what took place prior to and during the negotiations. “Detect and deter” will also be important concepts for future arms control and non-proliferation agreements. What can we learn from the CTBT? What are the lessons from decades of work to develop, negotiate, and implement the CTBT verification regime? We may draw some conclusions, despite the fact that the treaty is not yet in force and we do not know how ambitiously countries will monitor it. Verification can prove essential and even decisive for states to accept a treaty or agreement. The adequacy of verification arrangements is in the eyes of the beholder, and different countries may make different judgments. The CTBT demonstrates one way of balancing the tasks and responsibilities between states and the implementing organization; the IAEA safeguards agreements illustrate another. Whatever balance might be agreed upon in future treaties, experience shows that it is essential that this division of tasks is clear and clearly understood by all stakeholders. It is possible to develop, agree upon, and implement extensive verification arrangements, including monitoring systems of global reach and intrusive OSI regimes. To develop verification measures may take a long time and can involve many sciences and technologies. Ongoing scientific and technological development holds the potential of creating ever more efficient verification regimes. Science and technology are evolving, and it is essential to formulate verification regimes in a way that can accommodate such ongoing development. It is important to engage the scientific community on a broad scale early to create a scientific basis on which to build the verification regimes. Prior to negotiations, there is a need to create a process or a forum where the scientific and the political perspectives can interact. Because many remaining disarmament and non-proliferation issues have a global reach, broad international engagement is essential.

213

references* Aalseth CE, Andreotti E, Arnold D, Degering D, Giulian A, Gonzales R, Gurriaran R, Hult M, Keillor M, Laubenstein M, Le Petit G, Margineanu R, Matthews M, Miley H, Osvath I, Pelicciari M, Plastino W, Simgen H, Weber M, Werzi R (2009a) UltraLow Background Measurements of Decayed Aerosol Filters, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (RN-09), http://www.ctbto.org/ fileadmin/user_upload/ISS_2009/Poster/RN-09D%20%28US%29%20-%20Harry_ Miley%20etal.pdf Aalseth CE, Hayes J, Keillor M, McIntyre J, Orrell J, Seifert A (2009b) Argon-37 Background Measurements Supporting On-Site Inspection, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (OSI-24), http://www.ctbto.org/ fileadmin/user_upload/ISS_2009/Poster/OSI-24B%20%28US%29%20-%20Craig_ Aalseth%20etal.pdf Albright D, Brannan P (2006) ISIS Imagery Brief: New Activities at the Esfahan Natanz Nuclear Site in Iran, Institute for Science and International Security (ISIS), 14 April 2006 Anderson J, Gaya-Pique LR (2009) OSI: A Challenging Inspection—Logistics for the IFE, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (OSI02), http://www.ctbto.org/fileadmin/user_upload/ISS_2009/Poster/OSI-02_BOSI%20A%20challengin%20inspetion%20-%20Logistics%20for%20the%20IFE. pdf Annewandter R, Schmidt K, Geiger S, Kalinowski M (2009) The barometric driven xenon-tracer transport of the Non-Proliferation Experiment simulated by finite volume finite element higher-order accurate modeling, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (OSI-25), http://www. ctbto.org/fileadmin/user_upload/ISS_2009/Poster/OSI-25B%20%28UK%20 Germany%29%20-%20Robert_Annewandter%20Karen_Schmidt%20Sebastian_ Geiger%20Martin_Kalinowski.pdf

* Websites were accessed in May 2011

O. Dahlman et al., Detect and Deter: Can Countries Verify the Nuclear Test Ban?, DOI 10.1007/978-94-007-1676-6, © Springer Science+Business Media B.V. 2011

215

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

Arora N, Russell S, Kidwell P, Sudderth E (2011) Global seismic monitoring as probabilistic inference, Advances in Neural Information Processing Systems 23, MIT Press, 2011 Arora NS, Jordan MI, Russell S, Sudderth EB (2009a) Vertically Integrated Seismological Analysis I: Modeling, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (DM-08), http://www.ctbto.org/fileadmin/user_upload/ ISS_2009/Poster/DM-08A%20%28US%29%20-%20Nimar_Arora%20Stuart_ Russel%20and%20Erik_Sudderth.pdf Arora NS, Russell S, and Sudderth EB (2009b) Vertically Integrated Seismological Analysis II: Inference, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (DM-09), http://www.ctbto.org/fileadmin/user_upload/ISS_2009/Poster/ DM-09A%20%28US%29%20-%20Nimar_Arora%20Stuart_Russel%20and%20Erik_ Sudderth.pdf Arrowsmith SJ, Whitaker RW (2009) A robust technique for the automatic detection and location of infrasound events, poster presented at the ISS09 Conference, 10– 12 June 2009, Vienna (INFRA-04), http://www.ctbto.org/fileadmin/user_upload/ ISS_2009/Poster/INFRA-04F%20%28US%29%20-%20Stephen_Arrowsmith%20 and%20Rod_Whitaker.pdf Banister RW, Guthrie KM, Kay JS, Bold GEJ, Johns MD, Tan SM, Tindle CT (1993) ATOC–New Zealand receiver site survey and acoustic test, The Journal of the Acoustical Society of America, April 1993, Vol. 93, no. 4 Barasch G (1979) Light Flash Produced by an Atmospheric Nuclear Explosion, Los Alamos Scientific Laboratory, LASL-79-84 November 1979 Becker A, Wotawa G (2009) On the PTS in-house capacity building in atmospheric transport modelling during the past decade with an outlook on scheduled improvements in support of the noble gas verification regime, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (ATM-03), http://www.ctbto.org/ fileadmin/user_upload/ISS_2009/Poster/ATM-03E%20%28PTS%29%20-%20A_ Becker%20and%20G_Wotawa.pdf Becker A, Wotawa G, Ringbom A, Saey PRJ (2009) A refined backtracking and source reconstruction for the noble gas measurements taken in the aftermath of the announced October 2006 event in North Korea, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (ATM-10), http://www.ctbto.org/fileadmin/ user_upload/ISS_2009/Poster/ATM-10E%20%28PTS%29%20-%20A_Becker%20 and%20G_Wotawa.pdf Bergkvist NO, Johansson P (2009) Spot-checking of IMS location accuracy, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (SEISMO-02), http://www.ctbto.org/fileadmin/user_upload/ISS_2009/Poster/SEISMO-02I%20 %28Sweden%29%20-%20Nils-Olov_Bergkvist%20and%20Peder_Johansson.pdf Bergman EA, Engdahl ER (2009) Analysis of the Location Capability of the International Monitoring System, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (SEISMO-03), http://www.ctbto.org/fileadmin/user_upload/ ISS_2009/Poster/SEISMO-03I%20%28US%29%20-%20Eric_Bergman%20and%20 ER_Engdahl%20%28location%29.pdf 216

References

Blanc E (2009) Atmospheric Nuclear Explosions, presentation at the ISS09 Conference, 10–12 June 2009, Vienna Blanc E, Le Pichon A, Ceranna L, Farges T (2009) Study of the dynamics of the atmosphere with the Infrasound Network, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (INFRA-27), http://www.ctbto.org/fileadmin/ user_upload/ISS_2009/Poster/INFRA-27G%20%28France%29%20-%20Elisabeth_ Blanc%20etal.pdf Bolton, JR (2002) The Biological Weapons Convention: Challenges and Opportunities, Briefing Series, CNS, Washington, 11 January 2002, http://cns.miis. edu/archive/cns/programs/dc/briefs/011102.htm Brachet N, Brown D, Mialle P, Le Bras R (2009) Infrasound data processing for CTBT verification: sources, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (INFRA-03), http://www.ctbto.org/fileadmin/user_upload/ ISS_2009/Poster/INFRA-03F%20%28PTS%29%20-%20Nicolas_Brachet%20 etal%20%28IDCSource%29.pdf Bradley D (2009) Ambient Noise, poster presented at the ISS09 Conference, 10– 12 June 2009, Vienna (HYDRO-01), http://www.ctbto.org/fileadmin/user_upload/ ISS_2009/Poster/HYDRO-01H%20%28US%29%20-%20David_Bradley.pdf BWC (1972) Convention on the Prohibition of the Development, Production and Stockpiling of Bacteriological (Biological) and Toxin Weapons and on their Destruction, http://disarmament.un.org/treatystatus.nsf/44e6eeabc9436b788525 68770078d9c0/ffa7842e7fd1d0078525688f0070b82d?OpenDocument Campus P (2009) The IMS Infrasound Network, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (INFRA-01), http://www.ctbto.org/ fileadmin/user_upload/ISS_2009/Poster/INFRA-01F%20%28PTS%29%20-%20 Paola_Campus.pdf Canadian Space Agency radarsat2/default.asp

(2011)

http://www.asc-csa.gc.ca/eng/satellites/

Carlson J, Leslie R (2005) Special Inspections Revisited. Australian Safeguards and Non-Proliferation Office. Paper presented at the INMM 2005 Symposium, July 2005 Carluccio R, Console R, Chiappini S, Chiappini M (2009) Sel1 vs. REB bulletins comparison: looking for underlying features, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (SP-12), http://www.ctbto.org/ fileadmin/user_upload/ISS_2009/Poster/SP-12C%20%28Italy%29%20-Roberto_ Carluccio%20etal.pdf Carrigan CR, Johannesson G (2009) Application of a Bayesian Approach to Search-Area Reduction During a CTBT On-Site Inspection, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (OSI-10), http://www.ctbto.org/ fileadmin/user_upload/ISS_2009/Poster/OSI-10B%20-%20%28US%29%20-%20 Charles_Carrigan%20and%20Gardar_Johannesson.pdf Carrigan CR, Sweeney JR, Zucca JJ (2009) Detection of Trace Noble Gas Emissions from Underground Nuclear Explosions, poster presented at the ISS09 Conference,

217

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

10–12 June 2009, Vienna (OSI-26), http://www.ctbto.org/fileadmin/user_upload/ ISS_2009/Poster/OSI-26B%20%28US%29%20-%20Charles_Carrigan%20etal.pdf Ceranna L, Le Pichon A, Green DN, Maille PJ (2009a) The Buncefield Explosion: A benchmark for infrasound analysis across central Europe, Geophysical Journal International, 177: 491-508, doi: 10.1111/j.1365-246X.2008.03998.x, http:// onlinelibrary.wiley.com/doi/10.1111/gji.2009.177.issue-2/issuetoc Ceranna L, Le Pichon A, Vergoz (2009b) Towards an automatic and continuous monitoring of the infrasound activity across Europe, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (INFRA-05), http://www.ctbto.org/ fileadmin/user_upload/ISS_2009/Poster/INFRA-05F%20%28Germany%29%20 -%20Lars_Ceranna%20etal.pdf Chen P, Hogue R (2009) Atmospheric Transport Modeling, Science for Security, Verifying the Comprehensive Nuclear Test Ban Treaty, PTS, June 2009 Chiappini M, Carluccio R, Chiappini S, Console R, D’Ajello Caracciolo F, Damiani K, Giuntini A, Materni V, Pignatelli A (2009) The DPRK event of May 25, 2009: a preliminary analysis carried out at INGV from a multidisciplinary perspective, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (SEISMO-66), http://www.ctbto.org/fileadmin/user_upload/ISS_2009/Poster/SEISMO-66K%20 %28Italy%29%20-%20Massimo_Chiappini%20etal%20%28DPRK%20May%20 2009%29.pdf Cichowicz A (2009) Multi-Reference Relocation Technique, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (SEISMO-05), http://www.ctbto. org/fileadmin/user_upload/ISS_2009/Poster/SEISMO-05I%20%28South%20 Africa%29%20-%20Artur_Cichowicz.pdf Coates R, Seiche Ltd. (2009) Examples of undersea noise, National Physical Laboratory, in Bradley D (2009) Ambient Noise, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (HYDRO-01), http://www.ctbto.org/ fileadmin/user_upload/ISS_2009/Poster/HYDRO-01H%20%28US%29%20-%20 David_Bradley.pdf Comley CL, Davies AV, Thompson P, McLarty J, Ball M, Miles C, Collins J (2009) IMS Radionuclide Laboratory GBL#15, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (RN-31), http://www.ctbto.org/fileadmin/user_upload/ ISS_2009/Poster/RN-31D%20%28UK%29%20-%20Christine_Comly.pdf Congressional Commission (2009) America’s Strategic Posture, United States Institute of Peace, http://media.usip.org/reports/strat-posture-report.pdf Coxhead M (2009) On-Site Inspection: A role for satellite imagery, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (OSI-14), http:// www.ctbto.org/fileadmin/user_upload/ISS_2009/Poster/OSI-14_B-OSI%20a%20 role%20for%20satellite%20imagery.pdf Coyne J, Jia Y, Brogan R (2009) Relative Contribution of IMS Stations to the Reviewed Event Bulletin of the IDC, poster presented at the ISS09 Conference, 10– 12 June 2009, Vienna (SEISMO-10), http://www.ctbto.org/fileadmin/user_upload/ ISS_2009/Poster/SEISMO-10J%20%28PTS%29%20-%20John_Coyne%20etal.pdf

218

References

CTBT (1996) Comprehensive Nuclear-Test-Ban Treaty, http://unhq-appspub-01. un.org/UNODA/TreatyStatus.nsf/44e6eeabc9436b78852568770078d9c0/0655d51 a30692632852568770079dda2?OpenDocument CTBT: Science and Technology (2011) http://www.ctbto.org/specials/ctbtscience-and-technology-20118-10-june-2011-vienna-austria/ CTBT/PC-24/INF.9 (2005) Final report on the review of the organizational structure of the Provisional Technical Secretariat of the CTBTO Preparatory Commission, 22 April 2005 CTBT PTS (2011) Possible Projects for CTBTO Arising from the 2009 International Scientific Studies Conference, 10–12 June 2009, May 2011, CTBTO Preparatory Commission, http://www.ctbto.org.fileadmin/user_upload/ISS_2009/ISS_ report_2011.pdf CTBT/PTS/INF. 999 (2009) Lessons learned: Report on review and follow-up process for the 2008 integrated field exercise, 28 April 2009, CTBTO Preparatory Commission CTBT/PTS/INF.1021 (2009) Report on the conduct of the 2008 integrated field exercise, 4 August 2009, CTBTO Preparatory Commission CTBT/PTS/INF.1026 (2009) Event Screening Criteria for Waveform Technologies, 20 August 2009, CTBTO Preparatory Commission CTBT/WGB/TL-18/30 (2006) On-Site Inspection Test Manual, 7 September 2006, CTBTO Preparatory Commission CTBT/WGB-14/INF. 3 (2000) Report of the External Evaluation Team on the International Data Center, 24 November 2000, CTBTO Preparatory Commission CTBT/WGB-17/INF. 3 (2001) Report of the External Evaluation Team on the International Monitoring System, 21 December 2001, CTBTO Preparatory Commission CTBT/WGB-21/INF. 5 (2003) Report of the External Evaluation Team of Major Programme 4: On-Site Inspection, 5 June 2003, CTBTO Preparatory Commission CTBTO Preparatory Commission (2006) CTBT: Synergy with Science 1996–2006 and Beyond, ISBN: 92-95021-34-7, CTBTO Preparatory Commission D’Agostino (2008) Testimony on U.S. Strategic Posture before the House Armed Services Subcommittee Administrator Thomas P. D’Agostino, 27 February 2008, http://nnsa.energy.gov/mediaroom/congressionaltestimony/02.27.08 D’Amours R, Ringbom A (2009) A study on the global detection capability of IMS for all CTBT relevant xenon isotopes, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (RN-16), http://www.ctbto.org/fileadmin/user_upload/ ISS_2009/Poster/RN-16D%20%28Canada%20and%20Sweden%29%20-%20Real_ DAmours%20and%20Anders_Ringbom.pdf D’Amours R, Ek N, Hogue R, Servranckx R, Zaganescu C (2009) Supporting the CTBT Radionuclide Monitoring System at the Canadian Meteorological Centre, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (RN-21), http://

219

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

www.ctbto.org/fileadmin/user_upload/ISS_2009/Poster/RN-21D__Canada___-_ Real_DAmours__Nils_Ek_etal.pdf Dahlman O, Axelsson H, Edin K (1968) Charge mass and elevation influence on seismic waves from small chemical explosions above ground, Research Institute of National Defence, Stockholm, FOA 4 Report C 4375-26, October 1968 Dahlman O, Mykkeltveit S, Haak H (2009) Nuclear Test Ban: Converting Political Visions to Reality, Springer ISBN 978-1-4020-6883-6 De Geer LE (2009a) Swedish Experience Detecting Leakages from Foreign Underground Nuclear Explosions, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (RN-23), http://www.ctbto.org/fileadmin/user_upload/ ISS_2009/Poster/RN-23D%20%28Sweden%29%20-%20Lars-Erik_DeGeer.pdf De Geer LE (2009b) Radioxenon Signatures from Underground Nuclear Explosions, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (RN-22), http://www.ctbto.org/fileadmin/user_upload/ISS_2009/Poster/RN22D%20%28Sweden%29%20-%20Lars-Erik_DeGeer.pdf Defense Industry Daily (2009) DSP Satellites: Supporting America’s EarlyWarning System, 11 October 2009, http://www.defenseindustrydaily.com/dspsatellites-supporting-americas-earlywarning-system-02650/ DigitalGlobe-ISIS (2006) ISIS Imagery Brief: North Korean Site After Nuclear Test, 17 October 2006, http://isis-online.org/uploads/isis-reports/documents/ dprktestbrief17october2006.pdf DOD (2010a) Nuclear Posture Review Report, U.S. Department of Defense, April 2010, http://www.defense.gov/npr/docs/2010%20nuclear%20posture%20 review%20report.pdf DOD (2010b) United States Declassifies Nuclear Stockpile Details to Promote Transparency, American Forces Press Service, 3 May 2010, http://www.defense. gov/news/newsarticle.aspx?id=59004 DOE (2000) United States Nuclear Tests: July 1945 through September 1992, DOE/ NV-209 (Rev. 15), December 2000, http://www.nv.doe.gov/library/publications/ historical/DOENV_209_REV15.pdf Donn WL, Rind D (1971) Natural Infrasound as an Atmospheric Probe, Geophysical Journal of the Royal Astronomical Society, 26: 111-133, http:// onlinelibrary.wiley.com/doi/10.1111/gji.1971.26.issue-1-4/issuetoc Drob DP, Meier R, Picone M, Garces M (2009), Inversion of infrasound signals for passive atmospheric remote sensing, poster presented at the ISS09 Conference, 10– 12 June 2009, Vienna (INFRA- 25), http://www.ctbto.org/fileadmin/user_upload/ ISS_2009/Poster/INFRA-25G%20%28US%29-%20Douplas_Drob%20etal.pdf Dubasov YV (2003) Xe radionuclides in subsoil gas and atmospheric air at Semipalatinsk Test Site, INGE Noble Gas Workshop, Ottawa, September 2003 Dubasov Y (2010) Underground Nuclear Explosions and Release of Radioactive Noble Gases, Pure and Applied Geophysics, 167: 455-461, doi: 10.1007/s00024-0090026-z 220

References

Duran EB, Shen X, McWilliams E, Cella L, Werzi R (2009) Proficiency Test Program for CTBT Radionuclide Laboratories, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (RN-29), http://www.ctbto.org/fileadmin/user_upload/ ISS_2009/Poster/RN-29D%20%28PTS%20-%20Emerenciana_Duran%20etal.pdf Edwards WN, ReVelle DO, Brown P (2009) Infrasonic Detection of Bolides, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (INFRA-17), http://www.ctbto.org/fileadmin/user_upload/ISS_2009/Poster/INFRA-17G%20 %28Canada%29%20-%20Wayne_Edwards%20etal.pdf El Baradei (2009) Mohamed El Baradei Warns of New Nuclear Age, The Guardian, 14 May 2009, http://www.guardian.co.uk/world/2009/may/14/elbaradei-nuclearweapons-states-un Ely JH, Fast JE, Seifert CE, Warren GA (2009) Estimation of Ground-Level Radioisotope Distributions for Underground Nuclear Test Leakage, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (OSI-17), http://www. ctbto.org/fileadmin/user_upload/ISS_2009/Poster/OSI-17B%20%28US%29%20 -%20James_Ely%20etal.pdf Engdahl ER, Bergman EA (2009) Analysis of Detection Capability of the International Monitoring System, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (SEISMO-09), http://www.ctbto.org/fileadmin/user_upload/ ISS_2009/Poster/SEISMO-09J%20%28US%29%20-%20ERobert_Engdahl%20 and%20E_Bergman%20%28detection%29.pdf Ericsson U, Dahlman O (1966) On the dependence of short period P-waves on yield, height and epicenter properties of nuclear air shots, Research Institute of National Defence, Stockholm, FOA 4 Report C 4261–23, August 1966 EUSC (2011) European Union Satellite Centre, http://europa.eu/agencies/ security-agencies/eusc/index_en.htm FAS (1998) Missile Defense and Related Programs, Federation of American Scientists, http://www.fas.org/spp/starwars/program/dote98/98sbirs.htm FAS (2011) Integrated Undersea Surveillance System (IUSS), Federation of American Scientists, http://www.fas.org/irp/program/collect/iuss.htm Fee D, Garces M, Steffke A, McCormack D, Servranckx R, Hetzer C, Hedlin M, Matoza R, Yepes H, Ramon P (2009) The Acoustic Fingerprint of Volcanic Ash Emission: Results from the International ASHE Project, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (INFRA- 28), http://www.ctbto.org/ fileadmin/user_upload/ISS_2009/Poster/INFRA-28G%20%28US%29%20-%20 David_Fee%20and%20Milton_Garces.pdf Ferguson CD (2006) Can Nuclear Forensics Trace a Detonated Nuclear Weapon to its Source? paper presented at the annual meeting of the American Political Science Association, Philadelphia, 2006, http://www.allacademic.com/meta/ p151550_index.html Findlay T (2006) Verification and the BWC: Last Gasp or Signs of Life? Arms Control Today, September 2006, http://www.armscontrol.org/act/2006_09/ BWCVerification 221

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

Flore S, Adam O, Ruzié G, Stafford K, Guinet C (2009) Using IMS hydroacoustic data to monitor whales in the South Western Indian Ocean, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (HYDRO-17), http://www.ctbto.org/ fileadmin/user_upload/ISS_2009/Poster/HYDRO-17H%20%28France%29%20 -%20Samaran_Flore%20etal.pdf Fourie CJS, Murdie R, Gaya-Pique L (2009) The Limitations of the Gravity Technique when Investigating a Possible Ground Zero, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (OSI-38), http://www.ctbto.org/fileadmin/ user_upload/ISS_2009/Poster/OSI-38B%20%28South%20Africa%29%20-%20CJS_ Fourie%20etal.pdf France Diplomatie (2010) Remarks at the Global Zero Conference by Pierre Sellal, Secretary General of the French Ministry of Foreign Affairs, Paris, 2 February 2010, http://www.diplomatie.gouv.fr/en/france-priorities_1/disarmamentarms-control_7359/arms-control-and-arms-trade_1109/events_2129/remarksat-the-global-zero-conference-by-pierre-sellal-secretary-general-of-the-frenchministry-of-foreign-affairs-02.02.10_13610.html French White Paper (2008) The French White Paper on Defence and National Security, June 2008, http://www.cfr.org/france/french-white-paper-defencenational-security/p16615 Friese JI, Payne RF, Arthur RJ, Miley H (2009) Investigations into Radionuclide Signatures at Underground Nuclear Test Sites, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (OSI-18), http://www.ctbto.org/fileadmin/ user_upload/ISS_2009/Poster/OSI-18B%20-%20Judah_Friese%20etal.pdf Frontline (2009) H-bomb or a fizzle, The Hindu, Vol. 26, no. 19, 12-25 September 2009, http://www.hindu.com/fline/fl2619/stories/20090925261908000.htm Garner DM (1967) Oceanic Sound Channels around New Zealand, New Zealand Journal of Marine and Freshwater Research, 1: 3-15, http://www.royalsociety.org. nz/publications/journals/nzjm/1967/001 Garwin RL (2011) Stockpile Stewardship Without Nuclear Explosion Testing, Presentation at ACA Annual Meeting, 10 May 2011 Gaya-Piqué LR, Grant A, Purss M, Tuckwell G (2009) The Use of Aeromagnetic Surveys in the Framework of CTBTO On-Site Inspections—Experience from IFE08, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (OSI03), http://www.ctbto.org/fileadmin/user_upload/ISS_2009/Poster/OSI-03_BThe%20use%20of%20aeromagnetic%20surveys.pdf Ghica D, Popa M, Oancea V (2009) Analysis of system performance applied to the territory of Romania (2004–2008), poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (SP-02), http://www.ctbto.org/fileadmin/user_upload/ISS_2009/ Poster/SP-02C%20%28Romania%29%20-%20Daniela_Ghica%20etal.pdf Gibbons SJ, Böttger Sörensen M, Harris DB, Ringdal F (2007) The detection and location of low magnitude earthquakes in northern Norway using multi-channel waveform correlation at regional distances, Physics of the Earth and Planetary Interiors, 160: 285-309 222

References

Gill A (2009) Possible Indian reaction to U.S. ratification of the CTBT, Bulletin of the Atomic Scientists, 5 May 2009, http://www.thebulletin.org/web-edition/ features/possible-indian-reaction-to-us-ratification-of-the-ctbt Ginzkey L, Kubaczyk C, Ehrlich J (2009) Hydroacoustic Monitoring for CTBTO with a Mobile Buoy System: Past Experience and New Concepts, poster presented at the ISS09 Conference, 10–12 June 2009 (HYDRO-03), http://www.ctbto. org/fileadmin/user_upload/ISS_2009/Poster/HYDRO-03H%20-%20Lothar_ Ginzkey%20Christian%20Kubaczyk%20Jan_Ehrlich.pdf Gitterman Y (2009) Source Phenomenology Experiments with Borehole Explosions of Special Design in Israel, Bulletin of the Seismological Society of America, 99: 1892-1905 Gitterman Y (2010) Sayarim infrasound calibration explosion: Near-source and local observations and yield estimation, 2010 Monitoring Research Review: Ground-Based Nuclear Explosion Monitoring Technologies, https://na22.nnsa. doe.gov/mrr/2010/PAPERS/07-06.PDF Gitterman Y, Garces M, Bowman R, Fee D, Israelson H, Hofstetter R (2009) Infrasound Calibration Experiment at Sayarim, Israel, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (INFRA-09), http://www.ctbto.org/ fileadmin/user_upload/ISS_2009/Poster/INFRA-09%20%28US%29%20-%20 Yefim_Gitterman%20etal.pdf Giuntini A, Console R, Carluccio R, Chiappini M (2009a) Comparison of IDC and National Seismic bulletins in the Italian Region, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (SEISMO-07), http://www.ctbto.org/ fileadmin/user_upload/ISS_2009/Poster/SEISMO-07I%20%28Italy%29%20-%20 Alessandra_Giuntini%20etal.pdf Giuntini A, Pignatelli A, Console R, Chiappini S, Chiappini M (2009b) Event Location through the analysis of seismic waves recorded by mini-arrays in the scenario of an On-Site Inspection, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (OSI-32), http://www.ctbto.org/fileadmin/user_upload/ ISS_2009/Poster/OSI-32B%20%28Italy%29%20-%20Alessandra_Giuntini%20etal. pdf Global Security (2005) Nuclear Weapon Hydrodynamic Testing, GlobalSecurity. org, 27 April 2005, http://www.globalsecurity.org/wmd/intro/hydrodynamic.htm Global Security Newswire (2009) Russia to Use Supercomputers to Test Viability of Nuclear Arsenal, 23 July 2009, http://gsn.nti.org/gsn/nw_20090723_5131.php Global Zero (2010) http://www.globalzero.org Gopalaswamy B, Smirnoff A (2009) Infrasound Detection of Rocket Launches, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (INFRA-14), http://www.ctbto.org/fileadmin/user_upload/ISS_2009/Poster/INFRA-14G%20 -%20Bharath_Gopalawamy%20and%20Alexander_Smirnov.pdf Green D, Bowers D (2009) Estimating the detection capability of the International Monitoring System infrasound network, poster presented at the ISS09 Conference,

223

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

10–12 June 2009, Vienna (INFRA-13), http://www.ctbto.org/fileadmin/user_ upload/ISS_2009/Poster/INFRA-13F%20%28UK%29%20-%20David_Green%20 and%20David_Bowers.pdf Green D, Vergoz J, Le Pichon A, Ceranna L, Drob D, Evers L (2008) Analysing Multiple Explosions in Europe: Signal Characteristics and Propagation Modeling, presentation at 2008 Infrasound Technology Workshop, Bermuda, 3–7 November 2008, http://www.rdss.info/librarybox/infrabox/ITW2008/index.html Green DN, Bowers D (2010) Estimating the detection capability of the International Monitoring System infrasound network, Journal of Geophysical Research, 115, D18116, doi: 10.1029/2010JD014017, http://www.agu.org/pubs/ crossref/2010/2010JD014017.shtml Green DN, Vergoz J, Gibson R, Le Pichon A, Cerrana L (2011) Infrasound radiated by the Gerdec and Chelopechene explosions: propagation along unexpected paths, Geophysical Journal International, 185: 890-910, doi:10.1111/j.1365246X.2011.04975.x Grossman EM (2009) Strategic Posture Panel Split over Nuclear Test Pact Ratification, Global Security Newswire, 7 May 2009, http://www. globalsecuritynewswire.org/gsn/nw_20090507_3983.php Hafemeister D (2007) Physics of Societal Issues: Calculations on National Security, Environment, and Energy, Springer, 2007 Hafemeister D (2009) CTBT Space-Based Monitoring: Interferometric Synthetic Aperture Radar and National Technical Means, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (SEISMO-65), http://www.ctbto. org/fileadmin/user_upload/ISS_2009/Poster/SEISMO-65K__US__-_David_ Hafemeister.pdf Hafemeister D, Weiss W, Miley H, Ringbom A (2009) The Future of Monitoring Radionuclides for the CTBT, presentation at the ISS09 Conference, 10–12 June 2009, Vienna Hall L (2000) Hydroacoustic Monitoring of the CTBT, CTBT Commission, http:// www.ctbtcommission.org/hallpaper.htm Han X, Jia M, Gong B, Tian M, Xu H (2009) The availability and limitation of techniques for on-site inspection, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (OSI-11), http://www.ctbto.org/fileadmin/user_upload/ ISS_2009/Poster/OSI-11B_-_Xiaoyuan_Han.pdf Harris D, Thomas L, Hildebrand J, Wiggins S, Harwood J (2009) Estimating the abundance of blue whales (Balaenoptera musculus) in the northern Indian Ocean using vocalisations recorded by sea-bed mounted hydrophones, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (HYDRO-12), http://www. ctbto.org/fileadmin/user_upload/ISS_2009/Poster/HYDRO-12H%20%28UK%20 US%29%20-%20Danielle_Harris%20etal.pdf Hawkins W, Wohletz KH, Stissman AJ, Sweeney JJ, Milbrath BD, Bowyer TW, Vannon MG (2009) Evaluation of On-Site Inspection Activities for Planning and

224

References

Implementation, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (OSI-07), http://www.ctbto.org/fileadmin/user_upload/ISS_2009/Poster/ OSI-07B%20%28US%29%20-%20Ward_Hawkins%20etal.pdf Hawkins W, Wohletz K (1996) Visual Inspection for CTBT Verification, Los Alamos National Laboratory, November 1996, http://www.ees1.lanl.gov/Wohletz/LAMS13244-MS.pdf Hebel S, Kalinowski M (2009) Natural lithospheric radioxenon background in soil gas samples, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (RN-27), http://www.ctbto.org/fileadmin/user_upload/ISS_2009/Poster/RN27D%20%28Germany%29%20-%20Simon_Henel%20and%20Martin_Kalinowski. pdf Hedlin M, Drob D, De Groot-Hedin C, Le Pichon A, Vergoz J, Walker K (2009) Studies of infrasound event location involving atmospheric events accurately located using dense seismic network data, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (INFRA-7), http://www.ctbto.org/ fileadmin/user_upload/ISS_2009/Poster/INFRA-07F%20%28US%29%20-%20 Micahel_Hedlin%20etal.pdf Heeger D (1997), Signal Detection Theory, New York University, 12 November 1997, http://www.cns.nyu.edu/~david/handouts/sdt-advanced.pdf Heinonen O (2010) The Case for an Immediate IAEA Special Inspection in Syria, Policy Watch, 5 November 2010, Belfer Center for Science and International Affairs, http://belfercenter.ksg.harvard.edu/publication/20507/case_for_an_immediate_ iaea_special_inspection_in syria.html Hellweg M, Dreger D, Romanowicz B (2009) Comprehensive Test Ban Monitoring: Contributions from Regional Moment Tensors to Determine Source Type and Depth, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (SEISMO-21), http://www.ctbto.org/fileadmin/user_upload/ISS_2009/Poster/ SEISMO-21J%20%28US%29%20-%20P_Hellweg%20etal.pdf Herrin E (1968) Introduction to 1968 seismological tables for P phases, Bulletin of the Seismological Society of America, 58: 1193-1195 Hyvernaud O (2009a) Comparison of CEA Hydroacoustic Bulletin and IDC Hydroacoustic Monitoring in the Pacific, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (HYDRO-10), http://www.ctbto.org/fileadmin/user_ upload/ISS_2009/Poster/HYDRO-10H%20%28France%29%20-%20Olivier_ Hyvernaud%20etal%20%28Comparison%20of%20CEA%29.pdf Hyvernaud O (2009b) Monitoring submarine eruptions and drifting icebergs in the South Pacific Ocean, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (HYDRO-11), http://www.ctbto.org/fileadmin/user_upload/ISS_2009/ Poster/HYDRO-11H%20%28France%29%20-%20Olivier_Hyvernaud%20etal%20 %28Monitoring%20submarine%29.pdf IAEA (2011) Safeguards Agreements and Additional Protocols, http://www.iaea. org/OurWork/SV/Safeguards/sv.html

225

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

Indonesia (2009) Statement by the Delegation of the Republic of Indonesia to the United Nations on Behalf of the Group of Non-Aligned States Parties at the Third Session of the Preparatory Committee for the 2010 NPT Review Conference on Specific Issue: Nuclear Disarmament and Security Assurances, 7 May 2009 IRIS (2011) Incorporated Research Institutions for Seismology, http://www.iris. edu.hq Jans W (2005) Hydroacoustic CTBT Monitoring of the World Oceans: Strengths and Weaknesses, presentation at the VERTIC Seminar during the Conference on Facilitating the Entry into Force of the CTBT (Article XIV Conference), 2005 Jasani B, Pesaresi M, Schneiderbauer S, Zeug G (2009) Remote Sensing from Space, Supporting International Peace and Security, Springer, 2009, ISBN 978-14020-8483-6 JASON (1997) Subcritical Experiments, JSR-97-300, March 1997, The MITRE Corporation, JASON Program Office JASON (2007) Pit Lifetime, JSR-06-335, 11 January 2007, The MITRE Corporation, http://www.fas.org/irp/agency/dod/jason/pit.pdf Jepsen D, Fisk M (2009) The Role of the IMS Hydroacoustic Network for characterizing events in the REB, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (HYDRO-06), http://www.ctbto.org/fileadmin/user_upload/ ISS_2009/Poster/HYDRO-06H%20%28Australia%29%20-%20D_C_Jepsen%20 and%20M_D_Fisk.pdf Joint Statement (2011) Full Text of China–U.S. Joint Statement, Xinhua, 19 January 2011, http://news.xinhuanet.com/english2010/china/2011-01/20/c_13698492.htm Joswig M (2009) Dig-into-Dust: Resolving ML–2.0 Microearthquakes in OSI Passive Seismics, Landslide Creep, and Sinkhole Collapse by Nanoseismic Monitoring, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (OSI-30), http://www.ctbto.org/fileadmin/user_upload/ISS_2009/Poster/OSI30B%20%28Germany%29%20-%20Manfred_Joswig.pdf Kalinowski M, Peters J, von Weizsacker CF (2009a) Discrimination of nuclear explosions against civilian sources based on atmospheric xenon isotopic activity ratios, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (RN24), http://www.ctbto.org/fileadmin/user_upload/ISS_2009/Poster/RN-24D%20 %28Germany%29%20-%20Martin_Kalinowski%20Jana_Peters%20etal.pdf Kalinowski M, Tuma M, Hebel S (2009b) Civil Sources of Atmospheric Radioxenon: Estimating Emission Strengths, poster presented at the ISS09 Conference, 10– 12 June 2009, Vienna (RN-01), http://www.ctbto.org/fileadmin/user_upload/ ISS_2009/Poster/RN-01D%20%28Germany%29%20-%20Martin_Kalinowski%20 Matthias_Tuma%20Simon_Hebel.pdf Kang Y (2008) Statement by Mr Kang Yong, Representative of Chinese Delegation at the 63rd Session of the UN General Assembly First Committee, 10 November 2008, Ministry of Foreign Affairs of the People’s Republic of China, http://www. fmprc.gov.cn/eng/wjb/zzjg/jks/kjfywj/t521626.htm

226

References

Kebede F, Koch K (2009) Consistency of the IDC-reviewed event bulletin with other global seismological bulletins and its unique contribution to seismicity studies, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (SEISMO-06), http://www.ctbto.org/fileadmin/user_upload/ISS_2009/Poster/ SEISMO-06I%20%28PTS%29%20-Fekadu_Kebede%20and%20Karl_Koch.pdf Kim TS, Shin JS, Che IY, Lee HI (2009a) Infrasound Signals from the Test of North Korea’s Long-Range Missile, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (INFRA-15), http://www.ctbto.org/fileadmin/user_upload/ ISS_2009/Poster/INFRA-15G%20%28Republic%20of%20Korea%29%20-%20 TaeSung_Kim%20etal.pdf Kim WY, Richards PG, Sykes LR (2009b) Discrimination of Earthquakes and Explosions near Nuclear Test Sites using Regional High-Frequency Data, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (SEISMO-27), http://www.ctbto.org/fileadmin/user_upload/ISS_2009/Poster/SEISMO-27J%20 %28US%29%20-%20Won_Young_Kim%20_Paul_Richards%20and%20Lynn_ Sykes.pdf Kleiner A, Mackey L, Jordan MI (2009) Machine Learning for Improved Automated Seismic Event Extraction, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (DM-02), http://www.ctbto.org/fileadmin/user_upload/ISS_2009/ Poster/DM-02A%20%28US%29%20-%20Ariel_Kleiner%20etal.pdf Köble T, Rosenstock W, Berky W, Friedrich H, Risse M (2009) New radionuclide measurement systems improving search and detection capabilities for OSI deployment, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (OSI-19), http://www.ctbto.org/fileadmin/user_upload/ISS_2009/Poster/OSI19B%20%28Germany%29%20-%20T_Koeble%20etal.pdf Koper KD, Seats K, Benz HM (2009) On the Composition of Earth’s Short Period Seismic Noise Field, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (SEISMO-30), http://www.ctbto.org/fileadmin/user_upload/ISS_2009/ Poster/SEISMO-30K%20%28US%29%20-%20Keith_Koper%20and%20Kevin_ Seats.pdf Koper KD, Xu Y (2009) Using IMS Seismic Arrays to Constrain the Structure of Earth’s Deep Interior, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (SEISMO-50), http://www.ctbto.org/fileadmin/user_upload/ISS_2009/ Poster/SEISMO-50K%20%28US%29%20-%20Keith_Koper%20and%20Yan_Xu.pdf Kovacs AC, Hegedus EZ, Guthy T, Csabafi R (2009) Seismic field exercise at ACIFE08 in Hungary, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (OSI-31), http://www.ctbto.org/fileadmin/user_upload/ISS_2009/Poster/ OSI-31B__Hungary__-_Attila_Csaba_Kovacs_Endre_Zoltan_Hegedus_etal.pdf Kristensen HM (2010) Status of World Nuclear Forces, Federation of American Scientists, 12 January 2010, http://www.fas.org/programs/ssp/nukes/ nuclearweapons/nukestatus.html Kuzma H, Vaidya S (2009) Data Mining, Science for Security, Verifying the Comprehensive Nuclear Test Ban Treaty, PTS, June 2009

227

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

Kuzma HA, Engdahl ER, Rector JW (2009) Support vector regression for phase arrival prediction and SEL3 event evaluation, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (DM-03), http://www.ctbto.org/ fileadmin/user_upload/ISS_2009/Poster/DM-03A%20%28US%29%20-%20Heidi_ Kuzma%20etal.pdf Kværna T, Ringdal F, Schweitzer J, Taylor L (2002) Optimized Seismic Threshold Monitoring Part 1: Regional Processing, Pure and Applied Geophysics, 159: 969-987 Kværna T, Ringdal F, Baadshaug U (2007) North Korea’s Nuclear Test: The Capability for Seismic Monitoring of the North Korean Test Site, Seismological Research Letters, 78: 487-497 Kværna T, Ringdal F (2009) Analysis of the IDC Reviewed Event Bulletin for detection capability estimation of the IMS primary and auxiliary seismic stations, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (SEISMO-15), http://www.ctbto.org/fileadmin/user_upload/ISS_2009/Poster/SEISMO-15J%20 %28Norway%29%20-%20Tormod_Kvaerna%20and%20Frode_Ringdal.pdf Kyl J (2009) Why We Need to Test Nuclear Weapons, The Wall Street Journal, 20 October 2009, http://online.wsj.com/article/SB10001424052748704500604574483 224117732120.html Lambert DG, von Seggern DH, Alexander SS, Galat TA (1969) The Long Shot Experiment, Vol. II, Comprehensive Analysis, Seismic Data Lab, Alexandria, VA, Report no. 234 Lawrence MW, Arranz MG (1998) A Global Hydroacoustic Monitoring System for the Comprehensive Nuclear-Test-Ban Treaty: Plans and Progress, Acoustical Society of America, 25 June 1998, http://www.acoustics.org/press/135th/lawrence. htm Le Petit G, Vivier A, Blanchard X (2009) Measurement of isotopic ratio Xe133m/ Xe133 at very low level counting using Bayesian statistical approach, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (RN-25), http://www. ctbto.org/fileadmin/user_upload/ISS_2009/Poster/RN-25D%28France%29%20 -%20%20Gilbert_LePetit.pdf Le Pichon A, Vergoz J, Blanc E, Ceranna L, Evers L (2009a) Assessing the detection capability of the global IMS infrasound network, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (INFRA-12), http://www.ctbto.org/ fileadmin/user_upload/ISS_2009/Poster/INFRA-12F%20%28France%29%20-%20 Alex_LePichon%20etal.pdf Le Pichon A, Vergoz J, Blanc E, Guilbert J, Ceranna L, Evers L, Brachet N (2009b) Assessing the performance of the International Monitoring System’s infrasound network: Geographical coverage and temporal variabilities, Journal of Geophysical Research, 114, D08112, doi:10.1029/2008JD010907, http://www.agu.org/journals/ ABS/2009/2008JD010907.shtml Lee HI, Che IY, Kim TS, Shin JS (2009) Characteristics of Seasonally Dependent Propagation of Infrasound Wave around the Korean Peninsula, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (INFRA-08), http://www.ctbto.org/

228

References

fileadmin/user_upload/ISS_2009/Poster/INFRA-08F%20%28Republic%20of%20 Korea%29%20-%20Lee_Hee-Il%20etal.pdf Lekic V, Romanowicz B (2009) Upper Mantle Elastic Structure Inferred Using the Spectral Element Method, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (SEISMO-55), http://www.ctbto.org/fileadmin/user_upload/ ISS_2009/Poster/SEISMO-55K%20%28US%29%20-%20Vedran%20_Lekic%20 and%20Barbara_Romanowicz.pdf Lewis J (2009) Subcritical Testing at Lop Nor, ArmsControlWonk, 3 April 2009, http://www.armscontrolwonk.com/2239/subcritical-testing-at-lop-nor Li B (2010) Statement by HE Ambassador Li Baodung to the 2010 Nonproliferation Treaty Review Conference, 4 May 2010, http://www.reachingcriticalwill.org/legal/ npt/revcon2010/statements/4May_China.pdf Li B, Gavrilov A (2009) The significance of horizontal refraction effect for backazimuth estimation from the CTBT hydroacoustic stations, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (HYDRO-14), http://www.ctbto.org/ fileadmin/user_upload/ISS_2009/Poster/HYDRO-14H%20%28Australia%29%20 -%20Binghui_Li%20and%20Alexander_Gavrilov.pdf Liu Y (2009) Seismic search engine, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (DM-16), http://www.ctbto.org/fileadmin/user_upload/ ISS_2009/Poster/DM-16A%20%28US%29%20-%20Yang_Liu.pdf Los Alamos National Laboratory (2011) FORTE (Fast On-orbit Recording of Transient Events), http://www.forte.lanl.gov/index.shtml Lugar (2009) Lugar introduces bill to assure arms control inspections continue until New START Treaty ratified, Senate Floor Statement of Senator Richard G. Lugar, 5 November 2009, http://lugar.senate.gov/news/record.cfm?id=319683& Maceira M, Ammon CJ (2009) Joint inversion of surface wave velocity and gravity observations and its application to Central Asian basins shear velocity structures, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (SEISMO-52), http://www.ctbto.org/fileadmin/user_upload/ISS_2009/Poster/SEISMO-52K%20 US%20-%20M_Maciera%20and%20CJ_Ammon.pdf Mackby J (2011) Nonproliferation, Verification and the Nuclear Test Ban Treaty, Fordham International Law Journal, 34: 697-733 Materni V, Console R, Chiappini S, Chiappini M, Adelfio G (2009) Development and testing of seismic regional discriminants, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (SEISMO-22), http://www.ctbto.org/ fileadmin/user_upload/ISS_2009/Poster/SEISMO-22J%20%28Italy%29%20-%20 Valerio_Materni%20etal.pdf Matthews KM (2009) Spurious Iodine-131 detections at remote IMS radionuclide stations, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (RN06), http://www.ctbto.org/fileadmin/user_upload/ISS_2009/Poster/RN-06D%20 %28New%20Zealand%29%20-%20Murray_Matthews.pdf

229

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

McEvilly TV, Uhrhammer RA, Clymer RW, Karavas B (1999) Northern Hayward Fault Network, Berkeley Seismological Laboratory Annual Report, http://seismo. berkeley.edu/annual_report/ar98_99/node6.html McIntyre JI, Aalseth CE, Bowyer TW, Hayes JC, Keillor ME, Kirkham R, Miley HS (2009) Concept of Operations for Noble Gas On-Site Inspections, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (OSI-27), http://www.ctbto.org/ fileadmin/user_upload/ISS_2009/Poster/OSI-27B%20%28US%29%20-%20Justin_ McIntyre%20etal.pdf Medalia J (2008) Comprehensive Nuclear-Test-Ban Treaty: Issues and Arguments, CRS Report for Congress RL 34394, U.S. Congressional Research Service, 28 February 2008 Medalia J (2010) North Korea’s 2009 Nuclear Test: Containment, Monitoring Implications, U.S. Congressional Research Service 7-5700, 2 April 2010 Medvedev D (2010) Remarks by President Obama and President Medvedev of Russia at New START Treaty Signing Ceremony and Press Conference, 8 April 2010, http://www.whitehouse.gov/the-press-office/remarks-president-obama-andpresident-medvedev-russia-new-start-treaty-signing-ceremony Mialle P, Brachet N, Brown D, Bittner P, Coyne J, Given J (2010) A unique global network that detects very low frequency sound waves in the atmosphere, PTS Spectrum no. 15, November 2010, http://www.ctbto.org/fileadmin/user_upload/ pdf/Spectrum/2010/Spectrum15_web.pdf Mikhailov VN (1996) USSR Nuclear Weapons Tests and Peaceful Nuclear Explosions, 1949 through 1990, The Ministry of the Russian Federation for Atomic Energy, Ministry of Defence of the Russian Federation, RFNC-VNIIEF Mikhailov VN (1998) USSR Nuclear Tests, Hydronuclear Experiments, Plutonium Inventory, Ministry of the Russian Federation for Atomic Energy Miley HS, Aalseth CE, Bowyer TW, Fast JE, Hayes JC, Hoppe EW, Hossbach TW, Keillor ME, Kephart JD, McIntyre JI, Seifert A (2009) Alternative treaty monitoring approaches using ultra-low background measurement technology, Applied Radiation and Isotopes, 67: 746-749 Minaya E, Burgoa B, Le Pichon A, Antier K, Cansi Y, Hernandez B, Drob D, Evers LG, Vaubaillon J (2009) Meteorite detected by Infrasound (IS08) and Seismic (PS06 and Bolivian Network) Stations, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (INFRA-18), http://www.ctbto.org/fileadmin/user_ upload/ISS_2009/Poster/INFRA-18G%20%28Bolivia%29%20-%20E_Minaya%20 etal.pdf Mubarak M (2001) Statement by HE Ambassador Mahmoud Mubarak before the Conference on Facilitating the Entry into Force of the Comprehensive Nuclear-TestBan Treaty, 11–13 November 2001, http://www.un.org/webcast/ctbt/statements/ egyptE.htm Murphy JR, Kohl BC, Stevens JL, Bennett TJ, Israelsson HG (2010) Exploitation of the IMS and other data for a comprehensive, advanced analysis of the North Korean

230

References

nuclear tests, 2010 Monitoring Research Review: Ground-Based Nuclear Explosion Monitoring Technologies, https://na22.nnsa.doe.gov/mrr/2010/PAPERS/04-11.pdf Myers SC, Johannesson G (2009) Physics-based data mining for seismic bulletins, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (DM11), http://www.ctbto.org/fileadmin/user_upload/ISS_2009/Poster/DM-11A%20 %28US%29%20-%20Stephen_Myers%20and%20Gardar_Johannesson.pdf NAM (2009) Final Document, XV Summit of the Heads of State and Government of the Non-Aligned Movement, 16 July 2009, www.namegypt.org/en/ RelevantDocuments/Pages/default.aspx NAS (2002a) Technical Issues Related to the Comprehensive Nuclear Test Ban Treaty, Committee on Technical Issues Related to Ratification of the Comprehensive Nuclear Test Ban Treaty, National Academy of Sciences, National Academy Press, 2002, http://www.nap.edu/openbook.php?isbn=0309085063 NAS (2002b) High Impact Terrorism: Proceedings of a Russian-American Workshop, Russian Academy of Sciences, National Academy of Sciences, National Academies Press, 2002, http://www.nap.edu/catalog.php?record_id=10301 NAS (2010) Review and Update: Technical Issues related to the Comprehensive Nuclear Test-Ban Treaty, National Academy of Sciences, 2010, http://sites. nationalacademies.org/PGA/cisac/PGA_053215 NASA (2009) Ship Tracks South of Alaska, NASA Earth Observatory, 4 March 2009, http://earthobservatory.nasa.gov/IOTD/view.php?id=37455 Natalegawa RMMM (2010) Statement to the NPT Review Conference by HE Dr RM Marty M. Natalegawa, Foreign Minister of Indonesia, 3 May 2010, http://www. reachingcriticalwill.org/legal/npt/revcon2010/statements/3May_Indonesia.pdf Natalegawa RMMM (2011) Indonesia Wants to Advance Peacekeeping Role in 2011, The Jakarta Post, 8 January 2011, http://www.thejakartapost.com/ news/2011/01/08/indonesia-wants-advance-peacekeeping-role-2011.html Nikkinen M, Zähringer M, Werzi R (2009) The radionuclide processing system of the CTBTO, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (RN30), http://www.ctbto.org/fileadmin/user_upload/ISS_2009/Poster/RN-30%20 %28PTS%29%20-%20Mika_Nikkinen%20etal.pdf NNSA (2010) NNSA Launches New Detection Capability, Press Release, 28 May 2010, http://nnsa.energy.gov/mediaroom/pressreleases/satellitelaunch05.28.10 NOAA (2001) Spectogram of a northeast Pacific blue-whale, Sounds in the Sea, 2001, National Oceanic and Atmospheric Administration Ocean Explorer, in Bradley D (2009) Ambient Noise, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (HYDRO-01), http://www.ctbto.org/fileadmin/user_upload/ ISS_2009/Poster/HYDRO-01H%20%28US%29%20-%20David_Bradley.pdf NOAA (2011a) National Oceanic and Atmospheric Administration, Acoustic Monitoring of the Vents Program, Ocean Sound, 2011, http://www.pmel.noaa.gov/ vents/acoustics.html

231

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

NOAA (2011b) National Oceanic and Atmospheric Administration Ocean Explorer, 2011, http://oceanexplorer.noaa.gov/explorations/sound01/background/ acoustics/media/sofar.html Nordyke MD (1974) A review of Soviet data on the peaceful uses of nuclear explosions, Lawrence Livermore Laboratory, Livermore UCRL-51414 Nordyke MD (2000) The Soviet Program for Peaceful Uses of Nuclear Explosions, 1 September 2000, Lawrence Livermore National Laboratory, http://www.osti.gov/ bridge/purl.cover.jsp?purl=/793554-ZAQEpq/native/ Norris RS, Kristensen HM (2010) Chinese nuclear forces, 2010, Bulletin of the Atomic Scientists, November/December 2010 NPT (1995) Declaration Dated 6 April 1995 by France, the Russian Federation, the United Kingdom of Great Britain and Northern Ireland and the United States of America in Connection with the Treaty on the Non-Proliferation of Nuclear Weapons, NPT/Conf.1995/20, 19 April 1995 (annex) NPT (2010) 2010 Review Conference of the Parties to the Treaty on the Nonproliferation of Nuclear Weapons Final Document, United Nations, New York, 2010, http://www.un.org/ga/search/view_doc.asp?symbol=NPT/ CONF.2010/50%20(VOL.I) NTI (2002) Novaya Zemlya Test Site to be Maintained, 28 June 2002, http://www. nti.org/db/nisprofs/russia/treaties/ctbt2.htm NTI (2011) CTBTO Tests Nuke Explosion Monitoring Technology, Nuclear Threat Initiative, Global Security Newswire 28 January 2011, http://www. globalsecuritynewswire.org/siteservices/full_edition.php?Edition=01/28/2011 Nuclear Security Project (2010) http://www.nuclearsecurityproject.org/site/c. mjJXJbMMIoE/b.3534665/k.5828/About_the_Project_Index.htm Obama B (2009) Remarks of President Barack Obama, Prague, 5 April 2009, The White House, http://www.whitehouse.gov/the_press_office/Remarks-ByPresident-Barack-Obama-In-Prague-As-Delivered Odom RI, Stephen RA (2004) Proceedings, Seismo-Acoustic Applications in Marine Geology and Geophysics Workshop, Woods Hole Oceanographic Institution, 24–28 March 2004, Technical report APL-UW TR 0406 Ohrnberger C, Riggelsen M, Scherbaum F (2009) Generative Graphical Models for Classification of Seismic Signals, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (DM-04), http://www.ctbto.org/fileadmin/user_upload/ ISS_2009/Poster/DM-04A%20%28Germany%29%20-%20Carsten_Riggelsen%20 etal.pdf Orfeus (2000) Signals from the accident of the Russian submarine Kursk, Observatories and Research Facilities for European Seismology, Vol. 2, no. 2, August 2000, http://www.orfeus-eu.org/Organization/Newsletter/vol2no2/norsar.html Pakhomov SA, Dubasov YV (2009) Development of improved equipment for measurements of xenon radionuclides in atmospheric air, poster presented at

232

References

the ISS09 Conference, 10–12 June 2009, Vienna (RN-11), http://www.ctbto.org/ fileadmin/user_upload/ISS_2009/Poster/RN-11D%20%28Russia%29%20-%20 Sergey_Pakhomov%20and%20Yuri_Dubasov.pdf Panday A (2009) Review of the Methodology Used for Global Backtracking of Anthropogenic Radionuclides, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (ATM-06), http://www.ctbto.org/fileadmin/user_upload/ ISS_2009/Poster/ATM-06E%20%28US%20Nepal%29%20-%20Arnico%20Panday. pdf Pearce RG, Ndiath AS, Kasmi A, Yassine C, Lombe D, Banganan E, Turyomurugyendo G, Gore J, Sitnikov K, Wiggins-Grandison M, Makhonina M, Fisseha M, Meless O, Polich P, Bittner P, Ali SM, Nikolova S, Tajan, Medinsakya T, Khukuudei U (2009) Exploiting the skills of waveform data analysts in the quest for improved automatic processing, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (DM-12), http://www.ctbto.org/fileadmin/user_upload/ ISS_2009/Poster/DM-12A%20%28PTS%29%20-%20Robert_Pearce%20etal.pdf Perry WJ, Scowcroft B (2009) U.S. Nuclear Weapons Policy, Independent Task Force Report No. 62, Council on Foreign Relations, April 2009, http://www.cfr.org/ proliferation/us-nuclear-weapons-policy/p19226 Pignatelli A, Nicolosi I, Carluccio R, D’Ajello Caracciolo F, Chiappini M (2009) Planning an on-site inspection from a potential fields anomaly perspective, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (OSI-36), http://www. ctbto.org/fileadmin/user_upload/ISS_2009/Poster/OSI-36B%20%28Italy%29%20 -%20A_Pignatelli%20etal.pdf Podvig PL, Bukharin OA, Zhelezov BV, Kadyshev TT, Miasnikov EV, Sutyagin IV, Tarasenko MV (2011) Early warning satellites, Russian strategic nuclear forces, updated in http://www.russianforces.org/sprn (original in Russian: Strategicheskoye yadernoye vooruzhenie Rossii, Moscow, IzdAT, 1998) Popov YS, Popov VS, Prelovskij VV (2009) High Throughput Mobile Xenon Measurement System for OSI Purposes, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (OSI-28) Powell P, Fries D (1966) Handbook of World-Wide Standard Seismograph Network, Institute of Science and Technology, University of Michigan, Ann Arbor, 1966 Prah M, Tanaka J (2009) OSI: Measurement of levels of radioactivity, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (OSI-06), http:// www.ctbto.org/fileadmin/user_upload/ISS_2009/Poster/OSI-06_B-OSI%20 Measurement%20of%20levels%20of%20radioactivity.pdf PRC (2010) Statement by the Chinese Delegation at the Thematic Debate on Nuclear Disarmament at the First Committee of the 65th Session of the UNGA, 15 October 2010, http://www.reachingcriticalwill.org/political/1com/1com10/ statements/15Oct_China.pdf Prior MK, Bittner P, Jepsen D, Yan J, Meless O (2009) Detection of an in-water event on seismic and hydroacoustic sensors, poster presented at the ISS09 Conference, 10–

233

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

12 June 2009, Vienna (HYDRO-07), http://www.ctbto.org/fileadmin/user_upload/ ISS_2009/Poster/HYDRO-07H%20%28PTS%29%20-%20Mark_Prior%20etal.pdf Procopio MJ, Young CJ, Gauthier JH (2009) Applying machine learning methods to improve efficiency and effectiveness of the IDC automatic event detection system, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (DM07), http://www.ctbto.org/fileadmin/user_upload/ISS_2009/Poster/DM-07A__ US__-_Michael_Procopio_etal.pdf PTBT (1963) Treaty Banning Nuclear Weapon Tests in the Atmosphere, in Outer Space and Under Water (also called Partial Test Ban Treaty or Limited Test Ban Treaty), http://www.un.org/disarmament/WMD/Nuclear/pdf/Partial_Ban_ Treaty.pdf Ramaker J, Mackby J, Marshall PD, Geil R (2003) The Final Test: A History of the Comprehensive Nuclear-Test-Ban Treaty Negotiations, CTBTO Preparatory Commission, 2003 Rambolamanana, Ramanantsoa, Andriananaivoarisoa, Randrianarinosy (2009) Observation and events identification from I33S station, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (INFRA-23), http:// www.ctbto.org/fileadmin/user_upload/ISS_2009/Poster/INFRA-23G%20 %28Madagascar%29%20-%20Rambolamanana%20etal.pdf Reeder PL, Bowyer TW, Abel KH, Hubbard CW, Panisko ME, Thompson RC, Warnes RA (1997) Observation of Xe-135 with the PNNL ARSA System, Pacific Northwest National Laboratory, PNNL-11654, August 1997 Richelson J (2006) The Vela Incident; Nuclear Test or Meteorite? The National Security Archive, 5 May 2006, http://www.gwu.edu/~nsarchiv/NSAEBB/ NSAEBB190/index.htm Riedmann R, Purtschert (2009) Detection and activity levels of natural Ar-37 in soil air, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (RN-28) Rind D, DonnWL (1975) Further Use of Natural Infrasound as a Continuous Monitor of the Upper Atmosphere, Journal of the Atmospheric Sciences, 32: 1694-1704, http://journals.ametsoc.org/doi/abs/10.1175/1520-0469%281975%29032%3C169 4%3AFUONIA%3E2.0.CO%3B2 Ringbom A, Elmgren K, Lindh K (2007) Analysis of radioxenon in ground level air sampled in the Republic of South Korea on October 11–14, 2006, Swedish Defence Research Agency, report FOI-R-2273-SE Ringbom A, Miley HS (2009) Radionuclide Monitoring, Science for Security, Verifying the Comprehensive Nuclear-Test-Ban Treaty, PTS, June 2009 Ringbom A, Saey PRJ, Bowyer T, Al Ajmi D, Al-Rashidi MS, Al-Saleh M, Aldener M, Becker A, Cooper MW, Elmgren K, Forrester J, Hansen R, Henriksson H, Hayes JC, Hosticka B, Kirkham R, Lidey LS, Lindh K, McIntyre J, Miley HS, Mumba N, Payne RF, Peterson J, Shalash M, Söderström C, Thompson RC, Tippawan U, Tooloutalaie N, van Der Linde H, Williams R, Wilson N, Zähringer M (2009a) Measurements of the atmospheric radioxenon background at four locations in Asia and Africa,

234

References

poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (RN-02), http://www.ctbto.org/fileadmin/user_upload/ISS_2009/Poster/RN-02E%20 %28Sweden%29%20-%20Anders_Ringbom%20etal.pdf Ringbom A, Henriksson H, Auer M, Hoffman I, Le Petit G, Saey PRJ, Schlosser C, Wotawa G (2009b) Characterization of the global distribution of atmospheric radioxenons, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (RN-19), http://www.ctbto.org/fileadmin/user_upload/ISS_2009/Poster/RN19D%20%28Sweden%29%20-%20Anders_Ringbom%20etal.pdf Ringbom A, Saey PRJ, Andersson P, Ban S, Becker A, De Geer LE, Elmgren K, Lindh K, Peterson J, Söderström C (2009c) Field Measurements around a major European radioxenon release source – a medical isotope production facility, poster presentation at the ISS09 Conference, 10–12 June 2009, Vienna (RN-03), http://www.ctbto.org/fileadmin/user_upload/ISS_2009/Poster/RN-03E%20 (Sweden)%20-%20Anders_Ringbom%20etal.pdf Ringbom A, Elmgren K, Lindh K, Peterson J, Bowyer T, Hayes J, McIntyre J, Panisko M, Williams R (2009d) Measurements of radioxenon in ground level air in South Korea following the claimed nuclear test in North Korea on October 9, 2006, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (RN26), http://www.ctbto.org/fileadmin/user_upload/ISS_2009/Poster/RN-26D%20 %28Sweden%29%20-%20Anders_Ringbom%20etal.pdf Ringdal F, Kværna T (1989) A multichannel processing approach to real time network detection, phase association and threshold monitoring, Bulletin of the Seismological Society of America, 29: 1927-1940 Ringdal F, Kværna T (1992) Continuous seismic threshold monitoring, Geophysical Journal International, 111: 505-514 Rodean HC (1971) Nuclear explosion seismology, US Atomic Energy Commission, Oak Ridge, TN, 1971 Rodgers AJ, Harben PE (1999) Modeling the conversion of hydroacoustic to seismic energy at islands and continental margins; preliminary analysis of Ascension Island data, UCRL-JC-134303 LLNL, 26 July 1999 Royer JY, Dziak RP, Delatre M, Chateau R, Brachet C, Haxel JH, Matsumoto H, Goslin J, Brandon V, Cansi Y, Guinet C, Samaran F (2009) Results from a 14-Month Hydroacoustic Experiment in the Southern Indian Ocean, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (HYDRO-04), http://www.ctbto.org/ fileadmin/user_upload/ISS_2009/Poster/HYDRO-04H%20%28France%29%20 -%20JY_Royer%20etal.pdf Russell S (2009) Detection, Classification and Association, Not Necessarily in That Order, presentation at the ISS09 Conference, 10–12 June 2009, Vienna Saey PRJ (2010) Xenon in Radionuclides in the Environment, John Wiley & Sons Ltd., ISBN 078-0-470-71434-8 Saey PRJ, Auer M, Becker A, Hoffmann E, Nikkinen M, Ringbom A, Tinker R, Schlosser C, Sonck M (2009a) The influence on the radioxenon background during

235

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

the temporary suspension of operations of three major radiopharmaceutical production facilities in the Northern Hemisphere and during the start-up of a radiopharmaceutical production facility in the Southern Hemisphere, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (RN-07), http://www.ctbto.org/fileadmin/user_upload/ISS_2009/Poster/RN-07D%20 %28Belgium%29%20-%20Pausl_Saey%20etal.pdf Saey PRJ, Bowyer T, Aldener M, Becker A, Cooper MW, Elmgren K, Faanhof A, Hayes JC, Hosticka B, Lidey LS, Mumba N, Payne R, Ringbom A, Thompson RC, van Der Linde H, Wortmann G, Zähringer M (2009b) Evaluation of environmental radioxenon signals from a singular large source emitter in Africa, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (RN-04), http://www.ctbto.org/ fileadmin/user_upload/ISS_2009/Poster/RN-04E%20%28Belgium%29%20-%20 Paul_Saey%20etal.pdf Salzberg D (2009) IMS Hydroacoustic Contributions to Tsunami Warning Research, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (HYDRO-16), http://www.ctbto.org/fileadmin/user_upload/ISS_2009/Poster/ HYDRO-16H%20%28US%29%20-%20David_Salzberg.pdf Schneider J, Dubrawski A, Ashokkumar B, Gergely A, Kanade A, Kaushal S, Xu J (2009) Supervised Classification for Improving Automatic Labeling of Phase and Identifying False Associations, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (DM-05), http://www.ctbto.org/fileadmin/user_upload/ ISS_2009/Poster/DM-05A%20%28US%29%20-%20Jeff_Schneider%20etal.pdf Schoengold CR, DeMarre ME, Kirkwood EM (1996) Radiological Effluents Released from U.S. Continental Tests 1961 through 1992, U.S. Department of Energy Nevada Operations Office, DOE/NV-317 (Rev.1), August 1996, https:// www.lanl.gov/history/admin/files/Radiological_Effluents_Released_from_U.S._ Continental_Tests_1961_Through_1992.pdf Schorlemmer D (2009) Probabilistic Estimates of Monitoring Completeness of Seismic Networks, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (SEISMO-42), http://www.ctbto.org/fileadmin/user_upload/ISS_2009/ Poster/SEISMO-42K%20%28US%29%20-%20Danijel%20Schorlemmer.pdf Science and Technology Review (2005) Earth’s Subsurface from Space, Science and Technology Review, Lawrence Livermore National Laboratory, April 2005, p4-11 Seibert P (2009) Current issues in atmospheric transport modeling and source location for the CTBT verification, poster presented at the ISS09 Conference, 10– 12 June 2009, Vienna (ATM-08), http://www.ctbto.org/fileadmin/user_upload/ ISS_2009/Poster/ATM-08E%20%28Austria%29%20-%20Petra_Seibert.pdf Seifert CE, Ely JH, Fast JE, Warren GA (2009a) Investigations into Radiological Overflight Searches for On-Site Inspections, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (OSI-20), http://www.ctbto.org/fileadmin/user_upload/ ISS_2009/Poster/OSI-20B%20%28US%29%20-%20Carolyn_Seifert.pdf Seifert CE, Myjak MJ, Pfund DA (2009b) Detection of Anomalous Gamma-Ray Spectra for On Site Inspection, poster presented at the ISS09 Conference, 10–12

236

References

June 2009, Vienna (OSI-21), http://www.ctbto.org/fileadmin/user_upload/ ISS_2009/Poster/OSI-21B%20%28US%29%20-%20Carolyn_Siefert%20M_ Myjak%20and%20D_Pfund.pdf Selby N, Bowers D (2009) Can the use of prior information improve signal detection? poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (DM-10), http://www.ctbto.org/fileadmin/user_upload/ISS_2009/Poster/DM10A%20%28UK%29%20-%20Neil_Selby%20and%20David_Bowers.pdf Shalikashvili J (2001) Letter to the President and Report on the Findings and Recommendations Concerning the Comprehensive Nuclear Test Ban Treaty, 4 January 2001, http://www.state.gov/www/global/arms/ctbtpage/ctbt_report.html Sharif (1998) Address by HE Mr Mohammad Nawaz Sharif, Prime Minister of the Islamic Republic of Pakistan to the 53rd Session of the UN General Assembly on 23rd September 1998, Pakistan Mission to the United Nations, http://www.fas.org/ news/pakistan/1998/09/12980923.html Shultz G, Perry W, Kissinger H, Nunn S (2007) A World Free of Nuclear Weapons, The Wall Street Journal, 4 January 2007, http://www.hoover.org/publications/ hoover-digest/article/6109 Shultz G, Perry W, Kissinger H, Nunn S (2008) Toward a Nuclear-Free World, The Wall Street Journal, 15 January 2008, http://online.wsj.com/public/article_print/ SB120036422673589947.html Simpson DW, Willemann R (2009) Open data exchange in support of CTBT research, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (SEISMO-56), http://www.ctbto.org/fileadmin/user_upload/ISS_2009/Poster/ SEISMO-56K%20%28IRIS%20-%20US%29%20-%20David_Simpson%20and%20 %20Ray_Willemann.pdf Singh M (2009) Manmohan signals return to Vajpayee line on CTBT, The Hindu, 30 December 2009, http://www.hindu.com/2009/12/30/stories/2009123057570100. htm Slipchenko V (2009) Russia: Former Envoy Outlines Proposals to Help Ratification of Test Ban Treaty, Moscow Carnegie Center for International Peace, 24 July 2009, http://www.carnegie.ru Smith A, Spivak A, Li Z (2009) Passive Seismic Monitoring of Aftershocks during an On-Site Inspection, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (OSI-33), http://www.ctbto.org/fileadmin/user_upload/ISS_2009/Poster/ OSI-33B%20%28US%20China%20Russia%29%20-%20Albert_Smith%20etal.pdf Space Daily (2011) Israel conducts massive controlled blast, UPI Space Daily, 27 January 2011, http://www.spacedaily.com/reports/Israel_conducts_massive_ controlled_blast_999.html Springer D, Denny M, Healy D, Mickey W (1968) The Sterling experiment: Decoupling of seismic waves by a shot-generated cavity, Journal of Geophysical Research, 73: 5995-6012, http://www.agu.org/journals/ABS/1968/ JB073i018p05995.shtml

237

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

State Department (2010) Adherence to and Compliance with Arms Control, Nonproliferation, and Disarmament Agreements and Commitments, U.S. Department of State, 2010, http://www.state.gov/documents/organization/145181. pdf Stephen RA (2009) Long-Range Ocean Acoustic Propagation, T-Phases, Earthquakes and Hydro-Acoustic Networks, presentation at the ISS09 Conference, 10–12 June 2009, Vienna Stephen RA, Bolmer ST, Dzieciuch MA, Worcester PF, Andrew RK, Buck LJ, Mercer LJ, Colosi JA, Howe BM (2009) Deep seafloor arrivals—An unexplained set of arrivals in ocean acoustic propagation, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (HYDRO-18), http://www.ctbto.org/fileadmin/user_upload/ISS_2009/ Poster/HYDRO-18H%20%28US%29%20-%20Ralph_Stephen%20etal.pdf Stevens JL, Murphy JR, Rimer N (1991) Seismic source characteristics of cavity decoupled explosions in salt and tuff, Bulletin of the Seismological Society of America, 81: 1272-1291 Stohl A (2009) Recent Developments in Atmospheric Transport Modeling and Source Identification, presentation at the ISS09 Conference, 10–12 June 2009, Vienna Storchak DA (2009) CTBTO Contribution to the Global Earthquake Data Collection—A View from the International Seismological Centre, presentation at the ISS09 Conference, 10–12 June 2009, Vienna Sugioka H, Suyehiro K, Shinohara M (2009) Detection, location, and characterization of hydroacoustic signals using seafloor cable networks offshore Japan, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (HYDRO-8), http://www.ctbto.org/fileadmin/user_upload/ISS_2009/Poster/ HYDRO-08H%20%28Japan%29%20-%20Hiroko_Sugioka%20etal.pdf Sykes LR, Ekstrom G (1989) Comparison of Seismic and Hydrodynamic Yield Determination for the Soviet Joint Verification Experiment of 1988, Proceedings of the National Academy of Sciences, Vol. 86, no. 10, 3456-3460, May 1989, Geophysics, http://www.pnas.org/cgi/reprint/86/10/3456 Thorn R, Westervelt D (1987) Hydronuclear Experiment, LA-10902-MS, Los Alamos National Laboratory TTBT (1990) Treaty between the United States of America and the Union of Soviet Socialist Republics on the Limitation of Underground Nuclear Weapon Tests (1974, and Protocol Thereto 1990), U.S. State Department, http://www.state.gov/www/ global/arms/treaties/ttbt1.html Tuma M, Igel C (2009) Kernel-based machine learning techniques for hydroacoustic signal classification, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (DM-06), http://www.ctbto.org/fileadmin/user_upload/ ISS_2009/Poster/DM-06A%20%28Germany%29%20-%20Matthias_Tuma%20 and%20Christian_Igel.pdf Turunen J, Peräjärvi K, Pöllänen R, Toivonen H (2009) A novel measuring

238

References

system for detecting and characterizing radioactive particles, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (RN-15), http://www.ctbto.org/ fileadmin/user_upload/ISS_2009/Poster/RN-15D%20%28Finland%29%20-%20 Jani_Turunen%20etal.pdf U.K. Government (2010) Securing Britain in an Age of Uncertainty: The Strategic Defence and Security Review, October 2010, http://www.direct.gov.uk/ prod_consum_dg/groups/dg_digitalassets/@dg/@en/documents/digitalasset/ dg_191634.pdf UN (1996) Comprehensive Nuclear-Test-Ban Treaty, Resolution adopted by the United Nations General Assembly, A/RES/50/245, 17 September 1996, http:// daccess-dds-ny.un.org/do Ungar RK, Bean M, Yi J, Hoffman I, Servranckx R, Zaganescu C, Ek N, Blanchard X, Gilbert L, Brachet G, Achim P, Taffery T (2009a) Changes in radioxenon observations in Canada and Europe during medical isotope production facility shut downs in 2008, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (ATM15), http://www.ctbto.org/fileadmin/user_upload/ISS_2009/Poster/ATM-15E%20 %28Canada%29%20-%20Kurt_Ungar%20etal%20%28Main%29.pdf Ungar RK, Stocki TJ, Li JG, Japkowicz N, Hoffman I, Yi J, Bean M (2009b) Ranking Methods of Classifying Radioxenon Observations using Machine Learning for the Comprehensive Nuclear-Test-Ban Treaty, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (DM-17), http://www.ctbto.org/fileadmin/user_upload/ ISS_2009/Poster/DM-17A%20-%20Kurt_Ungar%20etal.pdf UNSC (1998) Security Council Resolution S/Res/1172, 6 June 1998, http:// daccess-dds-ny.un.org/doc/UNDOC/GEN/N98/158/60/PDF/N9815860. pdf?OpenElement UNSC (2006) Security Council Resolution S/Res/1718, 14 October 2006, http://daccess-dds-ny.un.org/doc/UNDOC/GEN/N06/572/07/PDF/N0657207. pdf?OpenElement UNSC (2009a) Security Council Resolution S/Res/1874, 12 June 2009, http:// daccess-dds-ny.un.org/doc/UNDOC/GEN/N09/368/49/PDF/N0936849. pdf?OpenElement UNSC (2009b) Security Council Resolution S/Res/1887, 24 September 2009, http://daccess-dds-ny.un.org/doc/UNDOC/GEN/N09/523/74/PDF/N0952374. pdf?OpenElement UNSC (2009c) Security Council 6191st meeting, 24 September 2009, S/PV.6191, http://www.una.org.uk/towardszero/Meeting%20Record%20NPT%2024%20 Sept%2009%20N0952314.pdf U.S. Government (1997) Message from the President of the United States transmitting Comprehensive Nuclear-Test-Ban Treaty, Opened for signature and signed by the United States at New York on 24 September 1996, Treaty includes two annexes, a protocol and two annexes to the Protocol, U.S. Government Printing Office, Washington, D.C.

239

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

U.S. Senate (1999) Statement by Ambassador Stephen J Ledogar (Ret) Chief U.S. Negotiator of the CTBT, prepared for the Senate Foreign Relations Committee Hearing on the CTBT, 7 October 1999 Vaidya S, Engdahl ER, LeBras R, Koch K, Dahlman O (2009) Strategic Initiative in Support of CTBT Data Processing: vDEC (virtual Data Exploitation Centre), poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (DM-01), http:// www.ctbto.org/fileadmin/user_upload/ISS_2009/Poster/DM-01A%20%28US%20 PTS%29%20-%20Sheila_Vaidya%20etal.pdf Vincent P, Larsen S, Galloway D, Laczniak RJ, Walter WR, Foxall W, Zucca JJ (2003) New signatures of underground nuclear tests revealed by satellite radar interferometry, Geophyical Research Letters, Vol. 30, no. 22, 2141, doi:10.1029/2003GL018179, 19 November 2003, http://www.agu.org/journals/ ABS/2003/2003GL018179.shtml Waxler R, Hetzer, Assink, Talmadge, Gilbert, Garces and Stopa (2009) On the monitoring of hurricanes using the international infrasound network, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (INFRA-29), http:// www.ctbto.org/fileadmin/user_upload/ISS_2009/Poster/INFRA-29G%20 %28US%29%20-%20Roger_Waxler.pdf Weeks J (2009) Nuclear Disarmament: Will President Obama’s Efforts Make the U.S. Safer? CQ Researcher, Vol. 19, no. 34 Wei FS, Li M (2003) The cepstrum analysis of seismic source character, Acta Seismologica Sinica, Vol. 16, no. 1, January 2003, Article 1000-9116(2003)01-0000-00 Wei FS, Xu ZH (2009) Difference in seismic cepstrum between explosions and earthquakes, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (SEISMO-25), http://www.ctbto.org/fileadmin/user_upload/ISS_2009/Poster/ SEISMO-25J%20%28China%29%20-%20Fu_Sheng_Wei%20and%20Zhong-Huai_ Xu.pdf Wei C, Jing L (2008) Operational analysis on emergency logistical system and emergency response model, Dalian Maritime University IEEE/SODI, 25 November 2008, http://ieeexplore.ieee.org/xpl/freeabs_all.jsp?tp=&arnumber=4686605 Weiss W (2009) The Current Status of the RN Particulate Network, presentation at the ISS09 Conference, 10–12 June 2009, Vienna Werzi R (2009) The operational status of the IMS Radionuclide Particulate Network, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (RN20), http://www.ctbto.org/fileadmin/user_upload/ISS_2009/Poster/RN-20D%20 %28PTS%29%20-%20Robert_Werzi.pdf Willhauck G, Caliz JJ, Hoffmann C, Lingenfelder I, Heynen M (2005) Objectoriented ship detection from VHR satellite images, Definiens Imaging GmnH Munich, http://www.isprs.org/publications/related/semana_geomatica05/front/ abstracts/Dijous10/R32.pdf Williams RM, Humble PH, Hayes JC (2010) Extraction of Xenon using Enriching Reflux Pressure Swing Adsorption, 2010 Monitoring Research Review: Ground-

240

References

Based Nuclear Explosion Monitoring Technologies, https://na22.nnsa.doe.gov/ mrr/2010/PAPERS/06-07.PDF Woodward R, Busby R, Alvarez M, Ahern T, Taber J, Simpson D (2009) Observations from EarthScope’s USArray, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (SEISMO-31), http://www.ctbto.org/ fileadmin/user_upload/ISS_2009/Poster/SEISMO-31K%20%28IRIS-US%29%20 -%20Bob_Woodward%20Bob_Busby%20etal.pdf Wotawa G, Becker A, Kalinowski M, Saey PRJ, Tuma M, Zähringer M (2009) Global distribution of the Radioxenon background caused by known civilian emissions and its consequences for CTBT verification, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (ATM-16), http://www.ctbto.org/fileadmin/ user_upload/ISS_2009/Poster/ATM-16E%20%28PTS%29%20-%20G_Wotawa%20 and%20A_Becker.pdf Zhang W, Ungar RK, Hoffman I, Lawrie R (2009) Krypton isotopic signature study of the primary coolant of CANDU nuclear power plant, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (RN-08.1), http://www.ctbto.org/ fileadmin/user_upload/ISS_2009/Poster/RN-08D%20%28Canada%29%20-%20 Kurt_Ungar%20%28Krypton%29.pdf Zucca J, Sokkappa P, Carrigan C, Bowyer T, Miley H, Milbrath B, Hawkins W, Wohletz K, Vannon M, Arzigian J (2009) A systems approach to establishing an effective CTBT On-Site Inspection (OSI) regime, poster presented at the ISS09 Conference, 10–12 June 2009, Vienna (OSI-09), http://www.ctbto.org/fileadmin/ user_upload/ISS_2009/Poster/OSI-09B%20%28US%29%20-%20Jay_Zucca%20 etal.pdf

241

index References to figures and tables are indicated by f and t following the page numbers. A Aalseth, C. E., 105 Access to sensitive information, 17, 131 Accidental explosions as infrasound sources, 96 Accuracy of event location. See Location accuracy Acoustic gravity waves, 91 Acoustic thermometry, 172 Adequate verification, 182–185, 213 Aeromagnetic surveys, 154 Aerosol isotopes filters collecting, 167–168 ranked by frequency of detection, 145–146, 146t Africa African Commission on Nuclear Energy, 211 Nuclear-Weapon-Free Zone Treaty, 14, 184, 211 Aftershock monitoring, 150–154, 150t, 151–153f, 205 Agency for the Prohibition of Nuclear Weapons in Latin America (OPANAL), 211, 211f Aircraft incidents with volcanic ash, 169 observations, 144–145, 166, 205 Akram, Munir, 18 Albanian accidental explosion (2008), 96

243

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

Albright, Madeleine, 22 Alluvium, 193 Amplitude-distance curve, 36, 43f Antarctica ice calving or ice-shelf dislocations, 172, 173f planetary waves and, 169 Antarctic channel, 114 Antarctic Treaty of 1971, 10–11 ARCES array (Norway), 41–42f, 43, 43f, 124f Argon, 64 Ar-37, 147–149, 148f Arora, N., 59 Array stations, 41–42, 41–42f, 160, 177, 191 Article XIV Conference, 12, 14, 18 Ascension Island. See Chase 21 explosion (1970) Association process, 60–61 Atmospheric nuclear explosions, 89–110 detectable features, 90, 91f, 199 infrasound monitoring, 90, 91–101 detection, 98–100 localization and characterization, 100–101 observations, 93–94 overview, 91–92 sources, 95–97 last explosion in atmosphere 30 years ago, 89 national technical means (NTM), 107–110 overview, 3 radionuclide monitoring, 102–107 noble gases, 107 particles monitoring, 103–104 particulate detection and tracking, 104–106 satellite-based systems monitoring, 89, 90, 108–110 seismological monitoring, 101–102 signals and noise, 90 verification, 90 Atmospheric pumping, 147 Automated data processing vs. experienced analysts, 58, 61, 175f, 176 B Background radiation and xenon monitoring, 67–68, 67f, 166 Backtracking, 73, 86, 106, 126, 196 Baker underwater nuclear explosion, 111, 112f

244

Index

Baneberry nuclear explosion (U.S.), 30, 31f Bangkok Treaty, 211 Barium/lanthanum, 103, 105, 105f, 146f Bayesian inference methods, 58–59 Beijing radionuclide station (China), 167f Benham nuclear explosion (U.S., 1968), 150 Bergman, E. A., 51, 53 Beryllium, 105 Bhangmeters, 108 Biological Weapons Convention (BWC), 10, 183, 189 Bolides, 94f, 95, 95f, 169, 170 Bolivia and Peru border, meteorite impact at, 96, 96f Bowers, D., 98, 101 Boxcar nuclear explosion (U.S., 1968), 150 Bulgaria, accidental explosion (2008), 96 Buoy system, 116 Bush, George W. joint statement with Singh, 16 opposition to CTBT, 19 BWC. See Biological Weapons Convention C Calibration in precision monitoring, 194 in underwater monitoring, 121 Canadian Radarsat-2, 197f Cape Leeuwin hydroacoustic station (Australia), 119, 119f, 123f Carluccio, R., 58 Cavities and underground nuclear explosions effect on signals, 40, 193, 197–198 increasing risk of releases, 65 CD. See Conference on Disarmament Central Asia Nuclear-Weapon-Free Zone Treaty, 184, 211 Cepstrum analysis, 55 Certification of IMS stations, 10, 202 CFE (Conventional Forces in Europe) Treaty, 189 Chagos Archipelago hydroacoustic station (Indian Ocean), 173f Chalk River isotope plant (Canada), 67, 70, 71f Chase 21 explosion (1970), 118, 118f, 122–123, 123f Chemical explosions, 79f, 122, 135, 147, 150, 190, 202 Chemical Weapons Convention (CWC), 11, 14, 132, 133, 155, 183, 189 Chile earthquake (February 2010), 116

245

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

China atmospheric nuclear testing by, 89 Beijing radionuclide station, 167f earthquakes in, 202 engaging in debate of abolition of nuclear weapons, conditions for, 27 Executive Council membership of, 203 monitoring ability for atmospheric nuclear explosions, 107–108 NPT and, 6, 20 nuclear warheads held by, 24 number of nuclear tests by, 23, 24 PTBT and, 21 signing and ratification of CTBT, 13–14, 21 subcritical experiments (SCEs), 32 zero-yield ban and, 22 Cichowicz, A., 50 Clandestine testing, 156–157, 169, 203–205. See also Verification of compliance Clarification. See Consultation and clarification process Climate change, 169, 172 Clinton, Bill, 18 Cocos Island radionuclide station (Australia), 167f Colwick nuclear explosion (U.S.), 151, 151f Committee of the Red Cross, 200 Comprehensive Nuclear-Test-Ban Treaty (CTBT). See also CTBT Organization (CTBTO) adoption of, 6 Article I provisions, 7, 21 Article IV provisions, 188 Article V provisions, 131–132, 156, 187 Article XIV provisions, 12 compared to other treaties, 83, 184 in context of nuclear proliferation and disarmament, 27–28 countries needing to ratify for entry into force, 7 entry into force, 13–21, 206–212 inspections. See On-site inspections environmental protection and, 6 negotiations, 6, 7f preamble of, 6, 27 precedents, 20–21 provisions of, 7–23 ratification of, 7, 12, 13, 13f verification provisions of. See Verification of compliance Computer software, 143–144, 162, 196–197

246

Index

Conference of Committee on Disarmament (CCD), 161 Conference of the States Parties election of Executive Council, 133 Executive Council deferring non-compliance issues to, 185 initial session, 12, 132, 195, 207 role of, 7 special session request, 135 Conference on Disarmament (CD), 5, 6f, 21 Group of Scientific Experts, 32–33, 161, 184, 190 Confidence building, 189–190 Congressional Commission on the Strategic Posture of the United States, 19 Consultation and clarification process, 10, 185, 201–202 Conventional Forces in Europe Treaty (CFE), 189 Cooperation CTBT provisions for “cooperating national facilities,” 202 regional cooperation, 209–212 with scientific research community, 164–168, 177–179, 190 Cosmos early warning satellite systems (Russia), 107 Coxhead, M., 144 CTBT. See Comprehensive Nuclear-Test-Ban Treaty “CTBT: Synergy with Science 1996–2006 and Beyond” (PTS symposium), 174 CTBT Organization (CTBTO) Conference of the States Parties, 7 entry into force, 207–208 establishment of, 3, 7 Executive Council. See Executive Council Executive Secretary, 12 Preparatory Commission. See Preparatory Commission (PrepCom) phase role of, 185 scientific community and, 165, 168, 177–178, 190, 208 Technical Secretariats, 9 Cut-off treaty, 166 CWC. See Chemical Weapons Convention Cyclones, 169 D D’Agostino, Tom, 20 D’Amours, R., 70 Data fusion procedures, 57, 60, 74 Data mining, 57–58, 61, 162, 174–176, 197 Dead Sea explosion (1999), 123–124 Decoupling techniques, 40, 197–198

247

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

Definition of nuclear weapon test explosion, 21, 30 Democratic People’s Republic of Korea (DPRK). See Korea, Democratic People’s Republic of Depth estimation for earthquakes, 53, 54, 202 for underground nuclear explosions, 53–54 Detection. See also Precision monitoring atmospheric nuclear explosions, 90, 91f, 98–100 hydroacoustic monitoring, 117–118, 117–118f infrasound monitoring, 98–100 likelihood of, 199–200 underground nuclear explosions. See Underground nuclear explosions underwater nuclear explosions, 117–118, 117–118f xenon monitoring, 63t, 64, 68–73, 69–72f Deterrence, 182, 186–187, 212–213 Discrimination to separate explosions from earthquakes underground nuclear explosions, 54–57, 55f, 57f, 78, 79f case studies, 54–56 event screening, 56–57 precision monitoring, 85 xenon monitoring, 74–75 DPRK (Democratic People’s Republic of Korea). See Korea, Democratic People’s Republic of Drilling into an explosion cavity, 140 Dual use of IMS data. See Synergy with science E Earthquakes depth estimates of, 53, 54, 202 event discrimination to separate from nuclear explosions, 54–57, 55f aftershock monitoring and, 154 Chase 21 explosion vs. underground explosion vs. earthquake, 122–123, 123f hydroacoustic observations, 122 non-governmental monitoring and, 200 regional approach, 85 screening by generally applicable criteria, 193 spectral ratios, 78, 79f states’ ability to show earthquake as triggering event, 156, 202, 204 Fukushima nuclear reactor accident (2011) and, 166 hydroacoustic signals and, 172 Juan Fernández Island hydroacoustic station and, 116 precision monitoring and, 165, 194, 197

248

Index

satellite monitoring of, 197 seismic signals attributed to, 161 Ecuador volcanic eruption (July 2006), 170f Egypt’s signing and ratification of CTBT, 14–15 El Baradei, Mohamed, 25–26 Electromagnetic pulse (EMP), 108, 109f Enforcement. See Non-compliance Engdahl, E. R., 51, 53 Entry into force, 13–21 Environmental background radioactivity, 67–68, 67f, 166 Equator atmospheric nuclear radioactivity not crossing, 90 detection levels at, 73 Equipment. See On-site inspections (OSI) European Union common foreign and security policy (CFSP), 210–211 infrasound events, detection of, 99 joint verification centers, proposal for, 2, 210–211 regional infrasound network in, 168 Satellite Center (EUSC), 210–211, 210f Event location. See Location accuracy Executive Council membership of, 133–135, 134t, 203 need for diverse input from states parties, 209 OSI approval, 10, 126, 130, 131, 155, 157, 185, 203 referrals to UN, 132 role of, 7 Experts automated data processing vs. experienced analysts, 58, 175f, 176 Group of Scientific Experts (GSE), 32–33, 161, 184, 190 recruitment of, 177, 198 training of, 178, 179, 210 Explosion yield estimates, 38–39, 197 Eyjafjallajökull volcano (Iceland), 94 F False alarms, 40–42, 41f, 55 Fast On-Orbit Recording of Transient Events (FORTE) satellite, 110 Faultless nuclear explosion (U.S., 1968), 150 Federation of Digital Seismograph Networks (FDSN), 34 Fee, D., 169–170 Field tests. See Integrated Field Exercise (IFE08, Kazakhstan)

249

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

Filters to collect radionuclide particles, 103–104, 167–168, 167f Fisk, M., 122 Fleurus medical isotope plant (Belgium), 67 FORTE satellite, 110 Forward modeling, 73–74, 86, 196 France atmospheric nuclear testing by, 89, 91, 92f CTBT and, 21 engaging in debate of abolition of nuclear weapons, conditions for, 27 Executive Council membership of, 203 monitoring ability for atmospheric nuclear explosions, 107–108 NPT and, 6, 20 nuclear warheads held by, 24 number of nuclear tests by, 23 PTBT and, 21 Reseau Sismique Polynesien (RSP, seismic network), 172 Frankfurter Allgemeine Zeitung on CTBT, 26 French Polynesia, atmospheric nuclear testing in, 91, 92f, 172 Frequency filtering, 35, 35f Friese, J. I., 148 Fukushima nuclear reactor accident (2011), 166, 201 G Gamma surveys, 149–150, 149f Geophysical exploration techniques, 154–155, 195, 206 German National Data Center, 51 Ginzkey, L., 116 Gitterman, Y., 40 Global Communications Infrastructure, 33 Global Nuclear Security Summit (April 2010), 25 Global Seismographic Network (GSN), 34 Global Zero, 26 Google, 60 GPS satellites, 108, 110, 199 Green, D. N., 98, 101 Greenland infrasound station, 93, 93f “Ground Truth” (GT), 49, 52 Group of Scientific Experts (GSE), 32–33, 161, 184, 190 GT5 events, 51–52, 52–53f, 53

250

Index

H Hadley cell, 73 Hafemeister, D., 76 Han, X., 140 Hatoyama, Yukio, 16 Hecker, Sigfried, 19 Hu Jintao, 13 Hurricanes, 169 Hydroacoustic monitoring atmospheric nuclear explosions, 90 effectiveness of, 199 global network, 9f, 10 map of station locations, 8f, 9 underwater, 3, 113–114, 170–174, 199. See also Oceans, nuclear explosions in Hydrodynamic experiments, 32 Hydronuclear experiments, 32 Hydrophones, 115, 116f, 118, 119, 172 Hyvernaud, O., 119 I IAEA. See International Atomic Energy Agency ICBL (International Campaign to Ban Landmines), 200 ICBMs (Intercontinental ballistic missiles), 24 Iceberg tracking, 172, 173f Iceland volcanic eruption (spring 2010), 169 IDC. See International Data Center (Vienna) IFE08. See Integrated Field Exercise (Kazakhstan) IMS. See International Monitoring System Incorporated Research Institutions for Seismology (IRIS), 200 India moratorium in nuclear testing by, 21 nuclear testing by, 6, 23, 24 signing and ratification of CTBT, 15–16 unknown nuclear arsenal size of, 24–25 UN Security Council resolution condemning nuclear testing by, 156 U.S.-India nuclear agreement, 18 Indonesia’s signing and ratification of CTBT, 16 Information sharing among countries, 166, 200 of radionuclide observations from Fukushima, 201 with scientific community. See Scientific research community, cooperation with

251

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

Information technology, developments in, 162. See also Data mining Infrared sensors, 109 Infrasound monitoring atmospheric effects and, 169, 199 atmospheric nuclear explosions, 90, 91–101 accidental explosions, 96 meteorites and bolides, 95–96, 95–96f planned explosions for testing and calibration purposes, 96–97, 97f volcanos. See Volcanic eruptions global network, 9–10, 9f map of station locations, 8f, 9 regional network in Europe, 168 synergy with other sciences, 168–170 INGE (International Noble Gas Experiment), 66 INGV (National Institute of Geophysics and Volcanology), 54 InSAR. See Interferometric Synthetic Aperture Radar satellites Inspection and inspection team. See On-site inspections (OSI) Integrated Field Exercise (IFE08, Kazakhstan), 137, 142, 144, 145f, 153, 205 Intercontinental ballistic missiles (ICBMs), 24 Interferometric Synthetic Aperture Radar satellites (InSAR), 76, 76t, 77f, 143, 162, 196 Intermediate-Range Nuclear Forces (INF) Treaty, 11, 183, 189 International Atomic Energy Agency (IAEA) Additional Protocol, 11, 14, 166, 184 headquarters, Vienna, 189f inspections, 11, 184 Iran’s nuclear activities and, 17 North Korea, requests for access to, 155 NPT and, 83, 183 safeguard agreements, 11, 132, 166, 184 states sharing data with, 166, 201 International Campaign to Ban Landmines (ICBL), 200 International Data Center (IDC, Vienna), 10, 33, 162, 185 International Monitoring System (IMS), 1–2, 191–193 array stations. See Array stations benefits of, 192–193 confidence building and, 190 CTBTO role, 185 CTBT provisions, 9 global network, 9–10, 9f, 191 no operational performance criteria specified, 188 readiness upon entry into force of CTBT, 207

252

Index

satellite monitoring not part of, 75 seismological. See Seismological monitoring stations. See Array stations; Monitoring stations TS role, 185, 192 xenon monitoring network, 70, 86, 195 International Noble Gas Experiment (INGE), 66 International Scientific Studies Conference (ISS09) accuracy of seismic data, 193 data mining, 176 data processing, 58 hydroacoustic observations, 122 location accuracy, 49, 51, 85 monitoring technologies, 2 OSI technologies, 143 scientific and technological developments, 161–162 seismo-acoustic applications, 114 xenon monitoring, 63, 67 International Scientific Studies (ISS) project, 190 International Seismological Center (ISC), 45 Internet’s deterrent effect, 187 Iodine, 103, 105–106, 106f, 107, 146f Iran earthquakes in, 202 non-government monitoring of nuclear activities of, 201 signing and ratification of CTBT, 17 Telemetered Seismic Network. See Tehran network Iraq’s clandestine nuclear weapons program, 184 IRIS (Incorporated Research Institutions for Seismology), 200 ISC (International Seismological Center), 45 Israel chemical test explosions by, 102 NPT and, 14 no nuclear testing by, 24 signing and ratification of CTBT, 17 test and calibration explosions in Negev Desert, 96–97, 97f unknown nuclear arsenal size of, 24–25 ISS09. See International Scientific Studies Conference Italian Seismological Network, 50–51, 59, 59f Italy, earthquake depth estimates in, 54 Ivanov, Sergey, 31–32

253

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

J Japan Agency for Marine Earth Science and Technology (JAMSTEC), 113 explosions off coast of, detection of, 117–118, 117f, 119, 120f Fukushima nuclear reactor accident (2011), 166, 201 JASON Group report (2007), 20 Jepsen, D., 122 Joint verification center. See Regional cooperation Joint Verification Experiment (JVE), 39, 190 Joswig, M., 153 Juan Fernández Island hydroacoustic station, 116, 117, 117f, 119, 172 K Kansong, South Korea, observations of DPRK nuclear tests, 81–82f Kapralov, Yuri, 22 Kazakhstan Integrated Field Exercise in (October 2008), 137, 142, 144, 145f, 153, 205 location accuracy for events in mining areas, 49t, 51 rocket launches, 96 Kim, W. Y., 55 Kissinger, Henry, 26 Korea, Democratic People’s Republic of event characterization (discrimination) to separate explosions from earthquakes and, 54, 55, 56f IAEA requests for access to, 155 non-government monitoring of nuclear activities of, 201 non-participation in renouncing nuclear testing, 189 nuclear arsenal size of, 25 nuclear testing by, 6, 21, 24 OSI and, 135 signing and ratification of CTBT, 14 threshold monitoring and, 47–48 undeclared nuclear material and activities of, 184 underground nuclear explosions, 77–84, 78–83f UN Security Council resolution condemning nuclear testing by, 156 uranium enrichment facility in, 14 xenon releases from underground nuclear explosions of, 63t, 64 detection, 69–71f, 70, 86 Korea, South. See South Korea Kovacs, A. C., 153 Kursk submarine (Russia), 124, 124f Kuzma, H., 58 Kværna, T., 42, 47, 48 254

Index

L Lamb waves, 91 Lamont-Doherty (research group), 95 Landmines, 200 Lanthanum. See Barium/lanthanum Latin America Agency for the Prohibition of Nuclear Weapons in Latin America (OPANAL), 211, 211f Nuclear-Weapon-Free Zone Treaty, 184, 211 Ledogar, Stephen, 22 Le Monde on CTBT, 26 Le Pichon, A., 98, 101 Life extension programs (LEP), 20 Liu, Y., 60 Location accuracy hydroacoustic monitoring, 118–121, 120f, 199 on-site inspections (OSI), 48, 138–140, 139f uncertainties in relation to dimensions of inspection area, 204–205 underground nuclear explosions, 48–54 case studies, 49–51, 49t, 50f depth estimation, 53–54 global capabilities, 51–53, 52f underwater nuclear explosions, 118–121, 120f xenon monitoring, 73–74 Long Shot nuclear test (US 1965), 160 Lop Nor test site (China), 32, 55 Los Alamos (U.S.), 32 Low-yield nuclear explosive devices, 25 M Magnitude reporting, uncertainty in, 36–38 Managed access, 131 Materni, V., 54–55 mb-Ms method, 55 Measurements nuclear explosions used for seismic measurements, 160–161 underground nuclear explosions, 33–35 xenon monitoring, 66 Medical isotope production facilities, 67–68, 67f, 166, 196 xenon emissions compared to nuclear reactors, 67, 68f Medvedev, Dmitry, 22, 26

255

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

Meteoroid explosions as infrasound sources, 95–96, 96f, 170 Oregon, 94, 94f Middle East, 14–15, 17 Miley, H., 63 Mine Ban Treaty, 200 Mining explosions, 49, 49t, 50, 50f, 190 MKAR (Kazakhstan), 43–44, 44f, 45 Mobile laboratories, 137f, 205 Molybdenum, 103 Monitoring aftershock. See Aftershock monitoring hydroacoustic monitoring, 3, 113–123. See also Oceans, nuclear explosions in infrasound monitoring, 90, 91–101. See also Atmospheric nuclear explosions; Infrasound monitoring non-governmental, 200–201 precision. See Precision monitoring underground nuclear explosions, 29–87. See also Underground nuclear explosions Monitoring centers. See Regional cooperation Monitoring stations. See also Array stations certification of, 10, 202 “cooperating national facilities,” 202 hydroacoustic monitoring, 8f, 9 non-governmental, 200–201 non-IMS stations, 34, 191–192, 202 number and location of IMS stations, 8f, 9, 162, 188 sites of seismic stations, 34, 34f specifications for, 162, 192 “to be determined” IMS stations, 16, 195 TS role, 192 underground nuclear explosions, 35f, 36–37 underwater nuclear explosions, 114–116 xenon stations, 66, 86, 195–196 Monowai sea mount (Antarctica), 172 Murphy, J. R., 80 N NAM. See Non-Aligned Movement Natalegawa, Marty M., 16 National Academy of Sciences (NAS, U.S.), 20, 47 National Data Centers, 51, 179

256

Index

National Institute of Geophysics and Volcanology (INGV), 54 National Nuclear Security Administration (NNSA), 20 National technical means (NTM), 107–110 data used in, 34 defined, 130 deterrent effect of, 187 satellite-based, 89, 90 verification technologies, 161 xenon stations, 196 Nehru, Jawaharlal, 15, 15f Network capabilities infrasound monitoring, 9–10, 9f radionuclide xenon monitoring, 9, 9f, 66, 86, 195–196 seismological monitoring, 164–165 underground nuclear explosions, 45–47, 46f, 61–62 Nevada Test Site (NTS; now Nevada National Security Site) aftershock rates, 151–152, 151–152f atmospheric explosions, 102 Badger, atmospheric explosion, 91f Non-Proliferation Experiment (NPE, 1993), 147, 147f underground explosions, 24, 30, 31, 35f, 38, 39, 76t New paradigm for seismic analysis, 60–62, 85–86, 174, 212 New Strategic Arms Reduction Treaty (START, 2010), 24, 28f NGOs (non-governmental organizations), 200 Nitze, Paul, 183 NNSA (National Nuclear Security Administration), 20 Noble gases. See also Xenon radionuclide monitoring atmospheric nuclear explosions producing, 107 environmental background radioactivity, 166 Fukushima nuclear reactor accident (2011) and, 166 IMS network, 195. See also Xenon monitoring on-site inspections and, 147–149, 148f, 150 underground nuclear explosions producing, 30, 64, 65 underwater nuclear explosions producing, 125 Non-Aligned Movement (NAM), 14, 15, 16, 17, 27 Non-compliance consultation and clarification process, 10, 185, 201–202 deterring. See Deterrence sanctions, 187 UN role. See United Nations Non-governmental organizations (NGOs), 200 Non-Proliferation Experiment (NPE, 1993), 147, 147f, 150

257

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

Non-Proliferation of Nuclear Weapons, Treaty on the (NPT). See also International Atomic Energy Agency (IAEA) Article VI, 27, 28 confidence building, 189 Congressional Commission recommendations on, 19 Egypt’s compliance with, 14 entry into force, 21 India not signatory, 15 linked to CTBT, 27–28 moratorium, 6 North Korea’s withdrawal from, 14 purpose of, 26 ratification of key countries not completed for 20 years, 20 UN Security Council Resolution 1887 on, 25 verification process, 183–184 NORES station (Norway), 124f NORSAR array site (Norway), 35f, 124 North Korea. See Korea, Democratic People’s Republic of Novaya Zemlya test site (former Soviet Union), 31–32, 48, 64, 111 NPT. See Non-Proliferation of Nuclear Weapons, Treaty on the NPT Review Conferences, 13, 14, 16, 26, 27–28 NTM. See National technical means NTS. See Nevada Test Site Nuclear reactors Fukushima accident, 166 isotope ratios from reactors vs. explosions, 65f, 74 Nuclear Security Project, 26 Nuclear testing. See also Atmospheric nuclear explosions, Underground nuclear explosions ban on, 6, 184 cessation of, 5, 15 moratorium on, 6, 16, 18, 21, 189 Nuclear weapon-related experiments, 30–32 Nunn, Sam, 26 O Obama, Barack Global Nuclear Security Summit (April 2010) and, 25 support for U.S. ratification of CTBT, 19, 25 on time needed to achieve abolition of nuclear weapons, 26 Oceans, nuclear explosions in, 111–127 history of, 111–112

258

Index

hydroacoustic monitoring, 3, 113–123, 199 calibration, usefulness of, 121 detection, 117–118, 117–118f location accuracy, 118–121, 120f, 199 monitoring stations, 114–116 multitude of sources, 121–123, 121f overview, 3 radionuclide observations of, 125–126 satellite observations of, 126–127 seismic monitoring, 123–125 transparency of oceans, 112–113 Ocean warming, 172 Odom, R. I., 114 Oil exploration, 154 On-site inspections (OSI), 129–157 access to sensitive information and, 17, 131 conduct of, 2, 10, 48, 136–142 conclusion of, 157 equipment for, 136, 137f, 142, 143f, 163, 206, 208 logistics, 142 operational manual, 136–137 systemic approach, 138–140 time lines from test manual, 137, 138f data fusion and, 60 defined in CTBT, 130 inspection team, 2, 129f, 132, 132f, 140–142, 141f, 185, 206 location accuracy, 48, 138–140, 139f in oceanic areas, 126 overview, 1, 2, 129–130 political context, 130–132 purpose of, 130 reflections on frame, 132–136 refusal of state to allow, 156, 203–204 requests for activities triggered by, 131 consultation and clarification. See Consultation and clarification process evidence for, 130, 203 new evidence for second request, 205 purpose of, 2, 10 technologies, 140, 143–155 geophysical exploration techniques, 154–155, 195, 206 radionuclide techniques, 145–150, 146f, 146t

259

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

seismological aftershock monitoring, 150–154, 150t synergy with other sciences, 163 visual observations, 143–145 as tool for states to verify and deter, 1, 4, 155–157, 182, 203–206. See also Verification of compliance TS role, 185 voting on, 9, 10, 155, 185, 203 OPANAL (Agency for the Prohibition of Nuclear Weapons in Latin America), 211, 211f Open-source data, 201 Optical flash from atmospheric nuclear explosions, 108, 109f, 199 Optical satellites, 87, 107, 196 OSI. See On-site inspections Ottawa Process, 200 Overhead observations from aircraft. See Aircraft P Pakistan moratorium on nuclear testing by, 21 nuclear testing by, 6, 23, 24 signing and ratification of CTBT, 18 unknown nuclear arsenal size of, 24–25 UN Security Council resolution condemning nuclear testing by, 156 Partial Test Ban Treaty (PTBT), 5, 10, 11f, 15, 21, 24, 29, 89 Particles monitoring, 103–104 filters. See Filters to collect radionuclide particles Fukushima nuclear reactor accident (2011) and, 166 particulate detection and tracking, 104–106 Peaceful nuclear explosions (PNEs), 24 Pelindaba isotope plant (South Africa), 67, 68 Pelindaba Treaty, 211 Perry, William J., 19, 26 Peru and Bolivia border, meteorite impact at, 96, 96f “Physics package,” 30 Planetary waves, 169 PNEs (peaceful nuclear explosions), 24 Poland, location accuracy for events in mining areas, 49t Political context, 5–28, 182–190. See also specific countries Precision monitoring development of, 2, 3, 179 DPRK location of explosions, 81 joint analysis centers for, 198 regional cooperation for, 210

260

Index

satellite-based, 162, 194, 197 seismology using, 165, 194, 195f, 198 underground nuclear explosions, 62, 84–87 verification process using, 181–182, 193–198, 204, 205 Preparatory Commission (PrepCom) certification of IMS stations, 10 confidence building and, 190 educational role of, 160 headquarters, 189f IFE08 and, 142 monitoring stations, specifications for, 162 North Korean explosions and, 83 operational manual drafting, 137, 206, 207–208 PTS. See Provisional Technical Secretariat report on operational readiness, 207 role of, 9, 12 systemic approach to OSI, 138 technologies, 143 upgrading auxiliary stations, 34 U.S. participation in, 133 Probability maps, 59, 59f Provisional Technical Secretariat (PTS) collection and storage of filters by, 104 conversion into TS, 208 cooperation with research community future challenges, 177–179 hydroacoustic signals, 171 infrasound, 169 radionuclides, 164–166, 168 DPRK nuclear explosions and, 83 event screening report, 56 external reviews of, 208 Fukushima leakages and, 201 GT5 events and, 51–54 infrasound monitoring and, 93 network capabilities of, 45 quality assurance program, 104 reference event database established by, 95 role of, 9, 12 TS compared with, 185 underwater nuclear explosions reported by, 119–120 PTBT. See Partial Test Ban Treaty

261

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

Pulse light and electromagnetic flash from atmospheric nuclear explosions, 108, 109f, 199 Pumping, atmospheric, 147 Q QuickBird, 75 R Radar satellites, 76, 107, 143, 196. See also Interferometric Synthetic Aperture Radar satellites (InSAR) Radionuclide monitoring. See also Xenon monitoring atmospheric nuclear explosions, 102–107 filters. See Filters to collect radionuclide particles global network, 9, 9f improving methods for, 166 map of station locations, 8f, 9 non-conclusive nature of, 204 on-site inspections (OSI), 145–150, 146f, 146t role of, 3 synergy with science, 165–168 underground nuclear explosions, 63 underwater nuclear explosions, 125–126 Rarotonga Treaty, 211 REB. See Reviewed Event Bulletin Receiver operating characteristics (ROC), 41, 41f Reflection surveys, 154 Refusal of state to allow on-site inspections, 156, 203–204. See also Noncompliance Regional cooperation, 168, 178–179, 184, 189, 190, 198, 209–212 Regional seismic wave propagation, 38, 39f Research organizations, seismological data provided to, 164 Reseau Sismique Polynesien (RSP, French seismic network), 172 Reviewed Event Bulletin (REB) event screening and, 56 infrasound monitoring and, 94 locations and, 52f Ringbom, A., 63, 70 Ringdal, F., 42, 47 ROC (receiver operating characteristics), 41, 41f Rocket launches and re-entries as infrasound sources, 96 Russell, S., 58

262

Index

Russian Federation. See also Soviet Union earthquakes in, 202 Executive Council membership of, 203 New START treaty with U.S., 24 NPT and, 6 satellite systems of, 107 zero-yield ban and, 22 Ruthenium, 103 S Saey, P. R. J., 68 Safeguards agreements. See International Atomic Energy Agency (IAEA) Salzberg, D., 172 Sanctions for non-compliance, 6, 187 Satellite-based monitoring atmospheric nuclear explosions, 89, 90, 107–110, 199 capabilities of types of, 87 DPRK test site, 79f European Union Satellite Center (EUSC), 210–211, 210f GPS. See GPS satellites increased monitoring via, 196–197 interferometric mode radar satellites (InSAR), 76, 76t, 77f, 196 low-cost systems, 87, 196 non-governmental, 201 optical systems. See Optical satellites planning for OSI, 143–144 precision monitoring, 162, 194, 197 radar. See Radar satellites role of, 2 synthetic aperture radar (SAR), 76 technological improvements in, 162 underground nuclear explosions. See Underground nuclear explosions underwater nuclear explosions, 126–127 Sayuz TMA-11 capsule re-entry, 96 Schlesinger, James, 19 Schoengold, C. R., 145 Schorlemmer, D., 59 Scientific research community, cooperation with, 164–168, 177–179, 190, 208, 209 Scientific synergy. See Synergy with science Scope of zero yield, 21–22 Seasonal variation of detection levels, 73, 98, 99f Seepage, 147

263

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

Seismic wave propagation, 38, 39f, 165 Seismological monitoring aftershock monitoring, 150–154 atmospheric nuclear explosions, 101–102 IMS network, 164–165 map of station locations, 8f, 9 oceanic explosions, 123–125 precision. See Precision monitoring sites of seismic stations, 34, 34f underground nuclear explosions, 30, 32–62. See also Underground nuclear explosions underwater nuclear explosions, 123–125 Seismology, science of, 32, 160, 164–165 Semipalatinsk test site (former Soviet Union), 29, 32, 38, 39, 64, 147 aftershocks, 150, 150t, 151f Shahbaz (Ambassador, Pakistan), 18 Shalikashvili, John M., 186, 186f Sharif, Nawaz, 18 Ship monitoring by satellites, 126–127, 127f Short-period background noise, 36, 37f Shultz, George, 26 Signals and noise. See also Infrasound, Seismological, and Hydroacoustic monitoring aftershocks and, 152 atmospheric nuclear explosions, 90 underground nuclear explosions, 35–40 Singh, Manmohan, 16 Sites of nuclear tests, 23f, 24 SLBMs (submarine launched ballistic missiles), 24 Slipchenko, Victor, 22 Slovakian accidental explosion (2007), 96 Sound fixing and ranging (SOFAR) channel, 113–115, 114–115f, 117, 118 Sound monitoring. See Infrasound monitoring; Signals and noise Sound Surveillance System (SOSUS, U.S. Navy), 113, 120 South Africa, location accuracy for events in mining areas, 49t, 50 South Atlantic Ocean possible nuclear explosion (1979), 108 Southeast Asia ASEAN Secretariat, 211 Nuclear-Weapon-Free Zone Treaty, 16, 184, 211 South Korea’s detection of infrasound events, 99, 100f South Pacific Forum Secretariat, 211 Nuclear-Weapon-Free Zone Treaty, 184, 211 264

Index

Soviet Union atmospheric nuclear testing by, 89 bilateral agreements with U.S., 11, 132, 189 decoupling techniques, 40 hydronuclear experiments, 32 nuclear warheads held by, 24 number of nuclear tests by, 23, 24 peaceful nuclear explosions (PNEs) by, 24 PTBT and, 5, 89 seismic measurements from nuclear explosions, 161 TTBT and, 5 underwater nuclear explosions by, 111 xenon releases from underground nuclear explosions of, 63t, 64 Space-Based Infrared System (SBIRS), 109 Spectrogram of whale vocalizations, 122, 122f SPITS station, 124f Spontaneous fission of uranium creating xenon, 68 Starr, Richard, 22 State Department (U.S.) analysis of CTBT, 21 report on compliance with arms control agreements (2010), 19 States parties, 3, 209 verification. See Verification of compliance Stephen, Ralph A., 114 Stockpile Stewardship Program (SSP), 19, 20, 31 Stohl, A., 73 Strategic Arms Reduction Treaty (START), 11 Strategic Offensive Reductions Treaty (Moscow Treaty), 11 Subcritical experiments (SCEs), 31 Submarine detection, 113 Submarine launched ballistic missiles (SLBMs), 24 Sumatra earthquake (December 2004), 172, 173f Supervised training algorithm, 58 Sweden leakage detection from underground nuclear explosions, 63t, 64, 105 location accuracy for events in mining areas, 49, 49t, 50f Synergy with science, 159–179 CTBT and, 160–161 future relationship between CTBTO and scientific community, 177–178 exploitation for different sciences, 163–176 data mining, 174–176 hydroacoustic monitoring, 170–174, 171f, 177 infrasound, 168–170, 177 265

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

radionuclide observations, 165–168 seismology, 164–165 overview, 3 rapid S&T development, 161–163 scientific community, cooperation with, 164–168, 178–179, 190 seismology, 160, 164–165 Synthetic aperture radar (SAR), 76. See also Interferometric Synthetic Aperture Radar satellites (InSAR) T Taepodong-2 long-distance missile (April 2009), 96 Technical Secretariat (TS) compared with IAEA, 184–185 data analysis by, 34, 75, 83, 192 inspection team and, 140 OSI requests and, 130, 131 PTS converting to, 208 role of, 9 Technologies. See On-site inspections (OSI) Tehran network, 44f, 45 Terrorists, 25–26 Testing. See also specific countries conducting tests ban, desirability of, 6 as part of nuclear weapon development, 23–25 sites of tests, 23f, 24 “Test manual” for on-site inspections, 137–138, 138f Threshold monitoring, 46f, 47–48, 87, 194 Threshold Test Ban Treaty (TTBT), 5, 11, 21, 38–39, 161, 190 Thunderstorms, 169 The Times on CTBT, 26 Tlatelolco Treaty, 211, 211f T-phase stations, 116, 116f, 172 Transport modeling, 168 TS. See Technical Secretariat Tsunami, 166, 172 TTBT. See Threshold Test Ban Treaty Tunguharua volcanic eruption (July 2006), 170f U Underground nuclear explosions, 29–87 background radiation, 67–68 cavity. See Cavities and underground nuclear explosions

266

Index

characteristics of, 29–30 detection, 35–48 amplitude-distance curve, 36, 43f cavity effect on signals, 40 explosion yield estimates, 38–39 false alarms, 40–42, 41f frequency filtering, 35, 35f individual station capabilities, 42–45 magnitude reporting, uncertainty in, 36–38 network capabilities, 45–47, 46f, 61–62 processes, 40–42 regional seismic wave propagation, 38, 39f short-period background noise, 36, 37f signals and noise, 35–40 station variations, 35f, 36–37 threshold monitoring, 46f, 47–48 event characterization to separate explosions from earthquakes. See Earthquakes event location, 48–54. See also Location accuracy case studies, 49–51, 49t, 50f depth estimation, 53–54 global capabilities, 51–53, 52f history, 23 monitoring, 30, 32–62 association process, 60–61 automatic process, 61 measurements, 33–35 new methods to analyze and exploit data, 57–60 new paradigm for seismic analysis, 60–62, 85–86, 174 North Korean nuclear explosions, 77–84, 78–83f nuclear weapon-related experiments, 30–32 overview, 3, 29 precision monitoring, 84–87 preparations for, observation of, 144–145, 145f PTBT and, 5 radionuclide xenon monitoring, 63–75 background radiation, 67–68, 67f calculated isotope ratios after explosions, 75, 75f corroborating evidence needed, 74 detection, 68–73, 69–72f discrimination, 74–75 DPRK nuclear tests, 81–82f

267

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

IMS xenon network, 66, 86, 195–196 isotope ratios from reactors vs. explosions, 65f, 74 leakage detection, 63t, 64 localization, 73–74 measurements, 66, 147, 147f medical isotope production facilities’ compared to nuclear reactors’ emissions, 67, 68f new IMS xenon network, 195–196 release of xenon, 63–66 satellite-based monitoring, 30, 75–77, 87, 144. See also Satellite-based monitoring sites of seismic stations, 34, 34f threshold monitoring, 46f, 87 TTBT and, 5 Underwater nuclear explosions. See Oceans, nuclear explosions in Ungar, R. K., 75 United Kingdom Buncefield accidental explosion (2005), 96 engaging in debate of abolition of nuclear weapons, conditions for, 27 Executive Council membership of, 203 monitoring ability for atmospheric nuclear explosions, 107–108 NPT and, 6 nuclear warheads held by, 24 number of nuclear weapon tests by, 23 PTBT and, 5, 89 subcritical experiments (SCEs), 31 United Nations General Council resolution condemning nuclear testing, 15 non-compliance issues brought to, 132, 156, 187 Special Commission (UNSCOM) in Iraq, 132 United Nations Security Council BWC role, 183 Iran’s nuclear activities and, 17 recourse to, when OSI not approved, 136 resolutions condemning nuclear testing, 6, 156 sanctions for non-compliance, 187 Summit (September 2009) adopting Resolution 1887, 25 United States atmospheric nuclear testing by, 89 bilateral agreements with Soviet Union, 11, 132, 189 decoupling techniques, 40 definition of nuclear weapon test explosion as issue in, 21

268

Index

earthquakes in, 202 Executive Council membership of, 203 hydronuclear experiments, 32 India-U.S. nuclear agreement, 18 location accuracy for events in Wyoming, 49, 49t New START treaty with Russia, 24 NPT and, 6 nuclear warheads held by, 24, 25f number of nuclear weapon tests by, 23, 24 peaceful nuclear explosions (PNEs) by, 24 PTBT and, 5, 89 signing and ratification of CTBT, 18–20 satellite system of, 108–110, 199. See also Satellite-based monitoring subcritical experiments (SCEs), 31 treaties with Soviet Union/Russian Federation, 11 TTBT and, 5 underwater nuclear explosions by, 111 xenon releases from underground nuclear explosions of, 63t, 64 U.S. Defense Support Program (DSP), 109 U.S. Geological Survey, 53 U.S. Navy Sound Surveillance System (SOSUS), 113 USArray, 33, 94, 170 U.S Presidential Panel (1980) on satellite observation of possible nuclear explosion, 108 V VAAC (Volcanic Ash Advisory Centers), 170 Vaidya, S., 58 Vajpayee, Atal Behari, 16 Vela satellites and program (U.S.), 89, 108–110, 160 Verification of compliance, 1, 181–213. See also On-site inspections (OSI) atmospheric nuclear explosions, 90 consultation and clarification, 10, 185, 201–202 CTBT provisions, overview of, 10–11 detection and deterrence, 212–213 DPRK nuclear explosions and, 83–84 joint verification centers, proposal for, 2, 209–212 monitoring, 191–201 International Monitoring System, 191–193. See also International Monitoring System (IMS) non-governmental monitoring, 200–201 wide-area monitoring, 198–200

269

Detect and Deter: Can Countries Verify the Nuclear Test Ban?

on-site inspections, 203–206 political process, 182–190 adequacy of verification, 182–185, 213 confidence building, 189–190 deterring non-compliance, 182, 186–187 verification regime and shortcomings, 188 precision monitoring, 193–198 PTBT lacking verification provisions, 89 readiness for entry into force, 206–212 CTBTO, 207–208 regional cooperation, 209–212 states parties, 209 Vincent, P., 76 Volcanic Ash Advisory Centers (VAAC), 170 Volcanic eruptions hydroacoustic signals, 172 infrasound observations, 94, 95, 169–170, 170f W Wake Island hydroacoustic station, 119, 123f Wall Street Journal op-eds by Shultz, Kissinger, Perry, and Nunn (2007 & 2008), 26 Wave propagation. See Seismic, Infrasound, and Hydoacoustic wave propagation Weapons of mass destruction (WMDs), treaties on, 10–11, 183 Weather events, infrasound monitoring to identify, 169 Wei, F. S., 55 Werzi, R., 105 Whale vocalizations spectrogram of, 122, 122f tracking whale migrations by, 171–172 Whistleblowers, 187 Wide-area monitoring, 198–200 Wikileaks, 201 Working Group on Verification, 18 World Health Organization, 166, 201 World Wide Standard Seismograph Network (WWSSN), 160 X Xenon monitoring, 63–75 atmospheric nuclear explosions, 107 deterrent effect of, 186–187 environmental background radioactivity, 166 experimental mobile sampling unit for, 137f

270

Index

medical isotope production. See Medical isotope production facilities synergy with other sciences, 166 underground nuclear explosions, 63–75, 192 background radiation, 67–68, 67f calculated isotope ratios after explosions, 75, 75f corroborating evidence needed, 74 detection, 63t, 64, 68–73, 69–72f discrimination, 74–75, 204 DPRK nuclear tests, 81–82f, 135 IMS xenon network, 70, 86, 195 isotope ratios from reactors vs. explosions, 65f, 74 localization, 73–74 measurements, 66, 147–148f release of xenon, 63–66 Xe-131m, 64, 71, 71t, 74, 82, 83f, 147 Xe-133, 64, 67, 67f, 68t, 69–71f, 70–71, 71t, 72f, 73–74, 75f, 81–82, 81t, 83f, 147–149, 148f Xe-135, 64, 66, 71, 71t, 72f, 73–75, 81, 83f Xu, Z. H., 55 Y Yellowknife, Canada, detection of xenon release from DPRK event, 70–71f Z Zero yield, 21–22 Zirconium, 103

271