Sedimentology and Sequence Stratigraphy of Reefs and Carbonate Platforms: A Short Course (AAPG Course Notes 34)

SEDIMENTOLOGY AND SEQUENCE STRATIGRAPHY OF REEFS AND CARBONATE PLATFORMS A Short Course by

Wolfgang Schlager

Free University, Amsterdam

Continuing Education Course Note Series #34

Published by The American Association of Petroleum Geologists Tulsa, Oklahoma U.S.A.

PREFACE Classical sequence stratigraphy has been developed primarily from siliciclastic systems. Application of the concept to carbonates has not been as straightforward as was originally expected even though the basic tenets of sequence stratigraphy are supposed to be applicable to all depositional systems. Rather than force carbonate platforms into the straightjacket of a concept derived from another sediment family, this course takes a different tack. It starts out from the premise that sequence stratigraphy is a modern and sophisticated version of lithostratigraphy and as such is a sedimentologic concept. "More sedimentology into sequence stratigraphy" is the motto of the course and the red line that runs through the chapters of this booklet. The course sets out with a review of sedimentologic principles governing the large-scale anatomy of reefs and platforms. It then looks at sequences and systems tracts from a sedimentologic point of view, assesses the differences between siliciclastics and carbonates in their response to sea level, evaluates processes that compete with sea level for control on carbonate sequences, and finally presents a set of guidelines for application of sequence stratigraphy to reefs and carbonate platforms. In compiling these notes, I have drawn not only on literature but also on as yet unpublished materials from my associates in the sedimentology group at the Free University, Amsterdam. I acknowledge in particular Hemmo Bosscher, Ewan Campbell, Juul Everaars, Arnout Everts, Jeroen Kenter, Henk van de Poel, John Reijmer, Jan Stafleu, and Flora Vijn. The Industrial Associates Program in Sedimentology at the Free University was instrumental in getting this course underway-through stimulating discussions as well as by providing financial support.

CHAPTER 1

PRINCIPLES OF CARBONATE SEDIMENTATION latitude carbonate environments, the base of the euphotic zone usually lies at 50-120m (Figure 1-3). Temperature is the second dominant control on carbonate production. Generally, warmer is better, but there exists an upper temperature limit for all carbonate organisms that sets important limits to carbonate production, particularly in restricted lagoons. The most important effect of temper-ature, however, is the global zonation of carbonate deposits by latitude (Figures 1-4 through 1-7). In spite of what has just been said about the role of light, the boundary of tropical and temperate carbonates seems to be largely a function of winter temperature rather than radiation.Throughout the Phanerozoic, this boundary lay at approximately 30 latitude. Nutrients. Contrary to common expectations, highnutrient environments are unfavorable for carbonate systems (Figure 1-8). Nutrients, to be sure, are essential for all organic growth, including that of carbonate-secreting benthos. However, the carbonate communities, particularly reefs, are adapted to life in submarine deserts. They produce their organic tissue with the aid of sunlight from the dissolved nitrate and phosphate in sea water and are very efficient in recycling nutrients within the system. In highnutrient settings, the carbonate producers are outpaced by soft-bodied competitors such as fleshy algae, soft corals or sponges. Furthermore, the destruction of reef framework through bio-erosion increases with increasing nutrient supply. Growth potential of carbonate systems. Siliciclastic depositional systems depend on outside sediment supply for their accumulation. In carbonate settings, the ability to grow upward and produce sediment is an intrinsic quality of the system, called the growth potential. Conceptually, one should distinguish between the ability to build up vertically and track sea level, the aggradation potential, and the ability to produce and export sediment, the production potential. The aggradation potential is particularly critical for survival or drowning of carbonate platforms (Figure 19); the production potential is a crucial factor for the progradation and retreat of carbonate platforms and the infilling or starvation of basins. The growth potential is different for the different

In order to prepare ourselves for application of sequence stratigraphy to reefs and carbonate platforms, we first have to understand how shallowwater carbonate accumulations are formed. We therefore start out with an overview of basic rules that govern production, distribution and deposition of carbonate material in the marine environment. This overview is selective and focuses on those principles that are immediately relevant for the large-scale anatomy and seismic signature of reefs and platforms. Three basic rules capture the peculiar nature of carbonate depositional systems-carbonate sediments are largely of organic origin,the systems can build wave-resistant structures and they undergo extensive diagenetic alteration because the original minerals are metastable. The implications of these rules are pervasive. We will encounter them throughout the chapters of this booklet, starting with this first section, which briefly reviews the principles of growth of reefs and production of sediments as well as basic patterns in the anatomy of carbonate accumulations.

0

Growth and production "Carbonates are born, not made." This little phrase by Noel P. James illustrates the fact that carbonate growth and production are intimately tied to the ocean environment with light, temperature and nutrients exerting the most dominant controls. Light intensity decreases exponentially with water depth. Growth of organic tissue and carbonate fixation can be related to this exponential decay via a hyperbolic tangent function (Figures 1-1, 1-2). This implies that there is a shallow zone of light saturation, usually 10-20m, where light is not a growth-limiting factor; below it, light is the dominant control. The base of the euphotic zone is defined by biologists as the level where oxygen production by photosynthesis and oxygen consumption by respiration are in balance. In the geologic record, all environments with abundant growth of photosynthetic organisms, such as green algae or corals, are considered euphotic. In low1

2

Schlager

facies belts of a platform. Most important is the fact that the growth potential of the rim is significantly higher than that of the platform-interior facies. When confronted with a relative sea-level rise that exceeds the growth potential of the interior, a platform will raise its rim and leave the lagoon empty ("empty bucket"). Carbonate production commonly follows the law of sigmoidal growth shown in Figure 1-10. Communities respond to opening up of new living space with a sigmoidal growth curve. Growth is slow at first, then accelerates, often exceeding the rate of change in the forcing function; finally, growth decreases as the systems reaches the limits of the newly formed niche. In carbonate sedimentology, this pattern is known as the start-up, catch-up, keepup stages of growth (Neumann & Macintyre;1985) and reef response to the Holocene sea-level rise is a typical example of this rule (Figure 1-11). The law of sigmoidal growth implies that the growth potential of the system is significantly lower in the start-up phase. In the Holocene, the effect lasts only 2000 to 5000 years. However, much longer lag effects have been postulated after mass extinctions (e.g. Hottinger, 1989, p.270).

Anatomy of reefs and carbonate platforms In-situ production is the hallmark of carbonate platforms. Any parcel of sea floor in the photic zone acts as a source of sediment. Since production is highest in the uppermost part of the water column and the overlying terrestrial environment is inimical to carbonates, platforms have a very strong tendency to develop flat tops at sea level. Rim-building. The tendency towards flat tops is enhanced by the ability of tropical carbonate systems to build wave-resistant structures, particularly at the platform margins. Reefs are the most common building blocks of platform rims, but carbonate sand shoals, too, are able to build stable barriers by the interplay of storm deposition and rapid lithification. Thus reefs and shoals may alternate in the construction of a platform rim. The presence of a rim, often built to sea level, disrupts the seaward-sloping surface normally developed by loose sediment accumulations on a shelf. Rimmed platforms differ fundamentally from siliciclastic shelves in this respect (Figure 1-12). The growth anatomy of a rimmed platform is that of a bucket-a competent, rigid rim protects the loose sediment accumulation of the platform interior (Figure 1-13). Because of the higher growth potential of the rim, "empty buckets," i.e., raised rims and empty lagoons, are cohunonly found in the geologic record. The rapid decrease of carbonate production with depth, in conjunction with the ability of reefs to hold

much of this growth in place, enables carbonate systems to build rather steep relief. Depth-dependent, in-situ growth may be an explanation for the formation of steep forereef walls (Figures 1-14 through 1-16). Despite the fact that platform rims can aggrade faster than lagoons, many platforms backstep when they come under stress. Figure 1-17 shows why backstepping can be advantageous for platforms. The basic principle is that the platform moves the margin to a position where growth potential and relative sealevel rise are again in balance. Platforms built from loose sediment without reef construction or lithification at the shelf break are called carbonate ramps (see Read, 1985 for review). Ideally, ramp profiles are in equilibrium with the seaward-dipping wave base and parallel those of sediment-stuffed siliciclastic shelves (Figure 1-12a). However, the term is often used loosely and some rim-building is tolerated by most authors. Slopes and rises. Slopes and debris aprons around platforms are important elements of the edifice and largely determine the extent and shape of the top. These areas also act as sinks for much of the excess sediment produced by the platform top. In sequence stratigraphy, platform margins and slopes play a crucial role as they hold much of the information on lowstands of sea level. Geometry and facies of slopes and rises are governed by several rules that are summarized below. a) The volume of sediment required to maintain a slope as the platform grows upward increases as a function of platform height. The increase is proportional to the square of the height for conical slopes of isolated platforms, such as atolls, and it is proportional to the first power of the height for linear platform slopes. b) The upper parts of platform slopes steepen with the height of the slope, a trend that siliciclastics abandon at early stages of growth. As a consequence, the slopes of most platforms, notably the high-rising ones, are steeper than siliciclastic slopes (Figures 1-19, 1-20). Changes in slope angle during platform growth change the sediment regime on the slope, shifting the balance between erosion and deposition of turbidity currents. This in turn profoundly influences sediment geometry on the slope and the rise (Figure 1-21). c) The angle of repose of loose sediment is a function of grain size. Engineers have quantified this relationship for man-made dumps (Figure 1-22) and the same numerical relationships have recently been shown to apply to large-scale, geological features (Figures 1-23 through 1-26; Kenter, 1990). The important property seems to be the degree of cohesion of the sediment. Changes in composition of the sediment dumped on a slope may produce unconformities as shown in Figure 1-27.

Principles of Carbonate Sedimentation 3

-- ----

I I

--\ion\

I I

I

I

1

I

.C I

I :

.,. I

a.,

~I

~:

I

~

1:t :,

,...-

-

~

o

0

0

o

op' oo OJ

,/o

0

o

0

0,,/

,,o"'


~/

~0

o~~9

'

0

I

or;~

N

~

l ----------------------- ---~u_p_!l~!!~.!.9~~

/

g

-

I

E.

em

Fig.l-1. Change of light intensity (I) and organic production (P) with water depth. Light displays a simple exponential decrease with water depth. The curve of organic production can be related via a hyperbolictangent function to light intensity; production shows a shallow zone of light saturation, where light is not a growth-limiting factor, followed by rapid decrease of organic growth with water depth. After Bosscher & Schlager (in press), modified.

0

c:

q

I

o/

o ,//

I

/ ,/

0

0

o o

/,/'

,/ ,l' _./

o

/,

/

,,~

0

i-1'

0

°o /

'~0 ;

('11

/

,''

tD

0 0

r 0

o

,,l,

:

tJ "0

0 J

,,'

0

I' c

q

0

0,/

c9

/

~/J'o 0/ 0

0: 00

1

o

,l

0

1

0

1

,'

t>'>"

i

0

0

o ;o o o

~~ /cP

/

/

80

I

q,

~::: ____________________ jilJ~!.~~~~r:!l_!i~~

0 00 0

I

0

,_ , /

;: / : I 1 :I

00~

0

0 ~
!

'

/

,'

/

r!

r.,!

,.b

0

0

I

f

I

:

~I """ : ' f'

I

f'

ra

E >

:

"'u

0

I'

: f I

\n :~t :

t::

E

50

100

150

=-=::.

::~{

:;:::::

~

ill

~~~~:~~

:r:: ::

::

~r

~·i: :~\

·=··

::;t :}:·

]::;'

:::::::

.l1

£

i

~ ~ li~

5.0

10.0

growth rate (mm/yr)

;::::;: -:·:· ;:

··:·:·: ~:=:::

t;;:;Z

,·: .;.:

Jf:

:::;: :

~ --1 active reef growth strongly reduced growth max1mum depth

m Cl •

Fig.l-2. Predicted and observed values of coral growth vs. depth. Dots: measured growth rates of Caribbean reef coral Montastrea annularis; curves: growth rates predicted by a light-growth equation for common values of water turbidity in the Caribbean. Note zone of light saturation and lower zone of decrease of growth with depth. The good correspondence of observed and predicted values is very encouraging. It indicates that below the saturation zone, light is indeed the dominant control on carbonate production and that existing equations are reasonable approximations of reality. After Bosscher & Schlager (in press), modified.

Fig.l-3. Depth limits of reef growth in the Caribbean. Reef growth and growth of green algae are commonly used to define euphotic zone in a geologically reproducible way. After Vijn &: Bosscher, written comm.

4

Schlager Fig.l-4. Coral growth versus solar radiation (a) and water temperature (b). Note increase of growth with temperature and poor correlation of coral growth with radiation. After Bosscher & Schlager, written comm.

DATA FROM WEBER and WHITE 1974

4~----------------------------------~-420 480 510 450 AVERAGE SOLAR RADIATION

I It

U'l z

IN G CAUM2/DAY

..

8

6

::J

23

24

25

26

27

28

AVERAGE SURFACE WATER TEMPERATURE

29

30

IN'C

Fig.l-5. Distribution of recent coral reefs is limited in the north and south by minimum winter temperatures. After Bosscher & Schlager, written comm.

DISTRIBUTION OF CORAL REEFS

40

40

20

20

...

···I'· ·-:.•.

·:·

20

·~

..··.

···:-~

·..·.......

40

Q

CORAL REEFS

6J

20° ISOTHERM

··.

... ~·!·

C)'t?

0

'\

·:

20

40

Principles of Carbonate Sedimentation 5

A .....

z

<{

0

z w :::> u m z <{


z

::J

z

ID
0

w

0

>

~ ~

0

1
w w a: a::

<{

a::

oo

B

20° LATITUDE

REEF GROWTH I-----OPTIMUM--- •MARGINAL- -RARE----..!

MASSIVE

CORAL-ALGAL REEF COMPLEXES

(/) U)

w

z

REEF COMPLEXES INTERMIXED WITH BRYOZOAN- ALGAL BIOSTROMES, SANDS, BIOCLASTIC DEBRIS

BEDDED BRYOZOAN-ALGAL BIOCLASTIC DEBRIS ~ REEF GROWTH DURING CLIMATIC OPTIMA

~

u

J: 1-

w

>

1
w

a:

o

1oo

zo•

3o•

4o•

LATITUDE INDONESIA PHILIPPINE-MARIANAS TAIWAN RYUKYUS -JAPANCORAL-ALGAL

REEF

-fy.(y~y

' ... ·':·· ... ·. -'

BRYOZOAN-ALGAL BIOCLASTIC DEBRIS

Fig.l-6. Latitudinal change from tropical to temperate carbonate facies in the northern Pacific (after Schlanger, 1981). Note disappearance of reefs and decrease in thickness, indicating reduced production in temperate climate. (Reproduced with permission of Society for Sedimentary Geology)

6

Schlager

A Salini1y.

B Maximum temperature. •c

~

D Skeletll associations

C Minimum

tamper~ture,

a) PHYSICAL WATER VARIABLES vs. LINEAR GROWTH

Fig.l-7. Carbonates in temperate latitudes and tropical latitudes - a comparison. Figs A,B,C illustrate changes in environmental conditions; Fig.D illustrates difference in skeletal carbonate - temperate carbonates dominated by benthic foraminifers and molluscs ("foramol" association), tropical latitudes by green algae and corals ("chlorozoan" association); Fig.E shows that non-skeletal grains are absent in temperate-water carbonates. After Lees (1975). (Reproduced with permission of Elsevier Science Publishers)

"C

b)

FOR MONT ASTREA ANNULARIS DATA FROM TOMASCIK aM BANDER 196!1

DATA FROM CORTES and RISK 1985

e

DODGE, ALLER and THOMSON 1974 b.

2

0~--~------------~--------~--~~~-----.1 .2 2 --------10 20 100 200 RESUSPENBION

MGICM210AY

2 B.P. M. MG/1.

Fig.l-8. Coral growth versus amount of resuspended sediment (a) and amount of suspended particulate organic matter (b). Both variables are negatively correlated with reef growth. Suspended matter reduces light and, in the case of suspended organic matter, also stimulates growth of organisms that compete with carbonate benthos. After Bosscher & Schlager (written comm.).

Principles of Carbonate Sedimentation 7 FRI!QUENCY Of PLATFORM ACCUMULATION RATES

:

i "

10

1

···j~,

..••

cooling oru11 Hatooane ••• tewel

.,, •• dtnt

0

Q

0

I

cL

······-

lont-tlf'IR IMbllde•c•

l

Tr Alpl

Apan•tn• 'lalf.

Recent ra•t a.rala

Hotoc... ,.... (

H.ree I

I

.L

a

c

.

10-2010

0

()

eatta ..aa .Jur.-Rec.

c

oJHio

:;9

"""'

RU!S IN MM/IOOOYR ISUSNOI"I" UNITII

Fig.1-9. Carbonate growth rates and rates of (relative) sea-level rise. Open bars- Holocene, black- distant geologic past. Short-term Holocene rates exceed long-term Phanerozoic rates by one order of magnitude, in agreement with the rule that sedimentation rates increase as the duration of the observation interval decreases (e.g. Sadler, 1981). Based on the Holocene and cyclostratigraphy in ancient formations, Schlager&: Bosscher (written comm.) estimate the average growth potential of platforms in the time range of 1ol-1oS years at 300 -lOOOm/Ma. This short-term growth potential is relevant for platform drowning: whether a platform drowns or survives a relative sea-level rise is determined in a short-term race through the photic zone, whereby the zone of light-saturation in the top 10-20 m is most critical. It is, therefore, unlikely that long-term, steady subsidence and third-order sea-level cycles, both typically at 10 -100 m/Ma, can drown healthy reefs and platforms. Modified after Schlager (1981), Bosscher & Schlager (in press,a).

LAW OF SIGMOIDAL GROWTH

•

,~--------·:-::-:.:-::-:SI!!k~..,~up~

,, ,,/

I

rapid

creation of niche J:

i

e 0

I

time-

I

Fig.l-10. Populations of organisms respond to the opening up of new living space in three steps - first, growth lags behind the creation of living space, second, population growth exceeds the rate of change in space, finally, population growth is limited by the rate of growth in living space. Most carbonate systems are controlled by organic growth and thus follow this law.

,,/ log phase

c.tch up

I
Ita-! uP "'

8

Schlager

CD

>

... _..

as CD

C/)

llal rubble

c

CD

C/) Q)

....Q

1!!.

'

''

,_,. .,j

10

~ 0 Q)

.D

nis lacles

C/)

.....

Q) Q)

E

10

5

Q)

20

''

''

''

\

\

•

'

'

\

\

Alacran, Yucatan

\

\

\

\

I I

\ \

\

Gt.Barrier Reef

\ \

\

'\

\

\

\

\

\

\ \

\

\

\

\

\

coral rubble and sand lades

30

''

\

coral head • lacies

!

' ''

\

\

10 \

'

\

'

5

t t t

\

\

\ \

\

\

\

\

\

\ \ \

\

\

\

\ \ \ \

'1'

\

'

'

St.Crohc, Virgin Is.

\\ "''\\

Fig.1-11. Growth history of Holocene reefs. Note start-up, catch-up, and keep-up phase of growth, following the law of sigmoidal growth. After Schlager (1981), Neumann & Macintyre (1985).

a)

/high onergy deposits, close to shore

SILICICLASTICS

b)

high-energy sand shoals and reels

moderato to high on orgy shore (dependant on width and depth of lagoon)

protocted lagoon

elevated by organic framebulfdlng and ayndepositionallithification

\

I

···

'

\

.... · SOFTFtANK\

'

mudJ rubble \ I

\

RIMMED PLATFORMS

Fig.1-12. Shore-to-slope profiles in carbonates and siliciclastics. (a) Siliciclastics, abundant sediment supply. Result is seaward dipping surface in equilibrium with deepening wave base. (b) Rimmed carbonate platforms. Wave-equilibrium profile grossly distorted by construction of wave-resistant reefs and quickly lithifying sand shoals, mainly at platform margin but also elsewhere on platform (e.g. lagoonal patch reefs). Platforms are basically dish-shaped and equilibrium profiles develop only locally in parts of the lagoon.

I 1

soFrnu ' \sand,mud. ,' ',, __ ...... '

!

I

STIFF~IM

····· .......

/ I

1

\ organic: frt:IJI't&, / ' cement '

Fig.1-13. Bucket principle. The growth anatomy of rimmed carbonate platforms resembles a bucket, held together by stiff rims of reefs or rapidly cemented sand shoals, and filled with less consolidated sediments of lagoons and tidal flats. Growth potential of a platform is largely determined by the growth potential of the rim. After Schlager (1981). (Reproduced with permission of Geol. Society of America)

Principles of Carbonate Sedimentation 9

•

•

•

'

'

Fig.1-14. Reef wall of the barrier reef in Belize. It is unclear whether these walls are Pleistocene sea cliffs mantled by Holocene reef or whether they are constructional features, formed by upward and outward growth of corals. In any case, the walls illustrate the tendency of platforms to oversteepen their upper slopes. After James & Ginsburg (1979).

Schlager

10

200~----------------~--------------~

years B.P.

0 Belize Reef Wall

80000

k:O.OS;Ik:250;10:2000 Om:0.005 Rllllime 110000 yn SUrfoce plocted ..ay 1000 yn

20.0

40.0

60.0

IO.U

100.0

140.0

120.0

meten

Fig.l-15. The rapid decrease of reef growth with depth may explain the near-vertical reef walls. This computer simulation attempts to duplicate the reef wall of Belize as a series of reef shoulders formed during minor highstands of sea level. After Bosscher & Schlager, in press.

,.., 0

0

40 fl)


eo

E

120

160

10

~

YEARS B.P. &10 20 I

30 I

40

50 I

I

SeaLevel~

\ "' ~ !!

,i

I•

...

•• ••• t• • •• ••

:•

'. \

,

7{l

eoI

(\ f\

160

Fig.1·16. Conceptual model of James & Ginsburg (1979) that formed the basis for computer simulation in Fig.1·5. Fig.l-17. Backstepping is a typical response of carbonate platforms to a relative rise of sea level that slightly exceeds the platform's growth potential. There are several reasons why backstepping may be advantageous for a platform under stress. (a) Backstepping reduces wave destruction of the margin, making it easier for the platform rim to grow upward. (b) Platforms may step back to higher ground, such as an old shoreline. (c) Differential subsidence can be a reason for backstepping of platforms when the outer, most rapidly subsiding part of the platform can no longer cope with sea-level rise.

Principles of Carbonate Sedimentat ion 11

A)

BACKST EPPING/ WAVE POWER

C construc tion

Ga •C-D

D destruct ion

G reef growth

t?i:dZiSMHi:'1~:;,\(ittr!(t1~i~~f!!.-i\}J/XC-''??/!.!~/):\/:UJ(/:~~\i(i!_ ';,...

1

=decay of wave power by bottom friction Airy waves, depth constant (1)

B)

1 dH =-gH4 dxV

H wave height

v wave phase velocity g gravity accelera tion

BACKST EPPING ONTO HIGHER GROUND

sealevel rise

reefs

beach ridges &dunes

lagoon reef

BACKSTEPPING TO AREA OF LOWER SUBSIDENCE

C)

f

otential

CD

...en

0

distance from hinge line --+ hinge

zone

lws 1982

· -:::;·,;:··<·:·.:·:·:,.::·:.

12

Schlager Atoll "coM"

Platform "prJsm"

tana

W'Jlume y - ta':.a-

.oV,,.o~oincrrmtnfJ in •olum• ctlU5«1 by ~rtical growth All of platform or atoll

rolum~ofaloi/COM V• 7Th'

oolumtof5i01Wpri•m y• ...l!.:...l of Jtngth "} 21CIIIfl

Jiiin1ii

growth a1 rolumo

IH •ik h•l

QroW/11 tJf ro/utM

If* •:;;;k-hj

Fig.1·18. Upward growth of carbonate platforms with constant slope requires deposition of ever larger volumes of sediment on the flanks. In the case of a cone-shaped atoll, the growth of volume (V) is proportional to the square of the height of the cone. In case of a linear platform margin, the growth of V is proportional to the height of the platform. Real platforms probably fall between these two extremes. The steady increase in slope volume may limit the growth of very tall platforms such as oceanic atolls. Smaller platforms seem to compensate for this effect by increase in slope angle and/or rapid filling of the basin. Schlager (1981). (Reproduced with permission of Geol. Society of America)

A

_., __

...

Fig.1·19. Modem submarine slope angles of carbonate platforms and siliciclastic systems in Atlantic and Pacific. Contours of 1, 2 and 4% of total sample (N) in unit area. N (carbonates) = 413, N (clastics) = 72. Carbonates steepen with height by building slump-resistant slopes. After Schlager &c Camber (1986). (Reproduced with permission of Society for Sedimentary Geology)

pic.lo pordol

aaa pordol

•

•

600

2600

orm

400 200

2200

0

2000 alOO

B

2400

1500

1000

..

Ia locomotive

plz clavaces

•

500

0

..

.z; tl

2600

400

2400 2200

basin

0

2000 2000

1500

1000

500

-

.E

t.re del sella

platform

2500

-., E

600

200

.,

.,...

0

distance In meters Fig.1-20. Decrease in slope angle with decreasing slope height in a shoaling basin (Triassic, Southern Alps, Italy). This fossil example confirms the trend observed in modem slopes (Fig.1-18). (Reproduced with permission of Blackwell Scientific Publ.)

"i .z;

Principles of Carbonate Sedimentation 13

Angle vs. depositional regime EROSION Bahama Escarpment

> 25°

~====~Jt----~~~---7\' ,~~0 \

\ \

\

NEUTRAL (sand bypassilg) Exuma transect ODP

.?""

gravel

sand/silt

10-12°

Fig.1-22. Grainsize vs. angle of internal friction for dry sediments. (Angle of repose is considered equal to angle of internal friction at zero confining pressure). This relationship explains much of the difference in slope angle of coarse, mud-free slope sediments and muddy slopes. After Kirkby (1987), modified.

ACCRETION Blake transect ODP

2-3°

0

WELL DOCUMENTED EXAMPlES (GROUP A)

0

EXAWPLES LACKING PRECISE CONTROL ON GEOMETRY (GROUP B)

6,

FLANKS STABILIZED BY ORGANIC FRAMEBUILDING OR

CEMENTAT;~:OUP

Fig.l-21. Slope angle and the balance of erosion and deposition on slopes. As slope angle increases the vigor of turbidity currents increases too and changes the depositional regime on the slope from accretion to erosion. By-pass slopes represent an intermediate stage. They receive mud from the perennial rain of sediment, whereas coarse material travelling in large turbidity currents is swept farther into the basin. After Schlager & Camber (1986).

DOMINANT SEDIMENT.FABRIC

Fig.l-23. Sediment composition strongly influences slope angles of carbonate platform flanks. Cohesionless sediments, such as clean sand and rubble, build up to angles of over 40°. Muddy, cohesive sediments tend to develop large slumps that maintain a low slope angle. After Kenter (1990). (Reproduced with permission of Blackwell Scientific Publ.)

14

Schlager

.-------------------------------------------------------------------------------------------~~~ akm

f---1---+-~·-il

common dopth pOinla

LIIB-111

modern gully

erosion surface

A

slump mass

Fig.1-24. Profile of slope with muddy cohesive sediment. Characteristics are gentle slope angle and large-scale slumps with toe thrusts at the distal end of the slope. Northern flank of Little Bahama Bank, after Harwood &: Towers (1988) and Kenter (1990). (Reproduced with permission of Ocean Drilling Program and Blackwell Scientific Publishers) .

. ~ ..

;

!.

blae-a-Mo~M·

Mdlmlnt

ipfon

mid-slope

....

,

Fig.l-25. Large-scale slumps on muddy slope in plan view. Note combination of slump scars on the slope and turbidite aprons at the basin margin. After Harwood & Towers (1988). (Reproduced with permission of Ocean Drilling Program)

·,

Principles of Carbonate Sedimentation 15

NNE 024

ssw

INTERPRETATION CROSS-SECTION VAL DE MESOI, NORTHERN SELLA

204

platform atructureleaa zone

meters

200

100

500

300

200

100

0

Fig.1-26. Profile of platform slope composed of non-cohesive sediment, i.e. sand and rubble with little or no mud. Slumps are small and express themselves as numerous small-scale truncations. Overall bedding remains steep and sub-parallel. Triassic Sella platform, Dolomites, Italy. After Kenter (1990). (Reproduced with permission of Blackwell Scientific Publ.)

16

Schlager

a CARBPLAT

runtime: 2000 yrs timestep: 100 yrs platform height: 50 m platform width: 400 m Gmargin: 10 mmtyr Ginlerior: 4 mm/yr k: 0.2/m initial depth: 2 m linear sea-level rise: 3 mm/yr wavebase: 10 m minangle/maxangle: 5135 degrees width of section: 330 m

unconformity caused by change in sediment composition

b sea level

windward slope

leeward slope

Fig.l-27. Grain-size-related unconformity in model and outcrop. (a) Computer program CARBPLAT (Bosscher & Southam, 1992) takes results of Kenter (1990) into account and assigns different angles of repose to rubbly sand and mud. This produces an unconformity on the slope of a continuously growing platform. At first, the platform grows with an empty lagoon and thus sheds only sand and rubble from its reef margin. Subsequently, the lagoon fills up and exports large volumes of mud; this mud buries the reef talus at a more gentle slope angle. (b) Sediment geometry of the modem fore-reef slopes of Bahamian platforms as observed in submersible dives by Grammer et al. (1990) and Grammer (1991). Mud wedge unconformably overlies reef rubble and the situation closely resembles the model predictions of Bosscher & Southam (1992).

CHAPTER2

THE STANDARD FACIES BELTS

Irrespective of geologic time and setting, shallowwater carbonates have a strong tendency to develop similar facies belts. These facies can be grouped into a platform model (Figure 2-1), best summarized by Wilson (1975). Important specifics on particular facies have been added from other authors. 1A) DeepSea Setting: Below wave base and below euphotic zone; part of deep sea, i.e. reaching through the thermocline into the realm of oceanic deep water. Sediments: Entire suite of deep-sea sediments such as pelagic clay, siliceous and carbonate ooze, hemipelagic muds including turbidites; adjacent to platforms we find mixtures of pelagic and platformderived materials in the form of peri-platform oozes and muds. Biota: Predominantly plankton, typical oceanic associations. In peri-platform sediments up to 75% shallow-water benthos. 1B) Cratonic deep-water basins Setting: Below wave base and below euphotic zone but normally not connected with the oceanic deepwater body. Sediments: Similar to 1A but in Mesozoic-Cenozoic rarely ever pelagic day; hemipelagic muds very common; occasionally anhydritic; some chert; anoxic conditions common (lack of bioturbation, high organic content). Biota: Predominantly nekton and plankton, coquinas of thin-shelled bivalves (Posidonia type), sponge spicules. 2) Deep shelf Setting: Below fair-weather wave base but within reach of storm waves, within or just below euphotic zone; forming plateaus between active platform and deeper basin (these plateaus are commonly established on top of drowned platforms). Sediments: Mostly carbonate (skeletal wackestone, some grainstone) and marl, some silica; well bioturbated, well bedded. Biota: Diverse shelly fauna indicating normal marine conditions. Minor plankton. 3) Toe-of-slope apron Synonyms: lower slope, deep shelf margin. Setting: Below £air-weather wave base, below euphotic zone; debris apron formed by sediment

gravity transport at the toe-of-slope in an oceanic or cratonic basin or on the proximal part of a deep shelf. Sediments: Lime mud, somewhat siliceous, with intercalations of debris flows and turbidites in the form of microbreccias and graded beds of lime rubble, sand and silt. Biota: Mostly re-deposited shallow-water benthos, some plankton and deep-water benthos. 4) Slope (Figures 1-18 through 1-27) Synonym: foreslope of platform Setting: Strongly inclined sea floors (over 1.4°) seaward of the platform margin. Sediments: Mostly pure carbonates, rare intercalations of terrigenous mud. Grain size highly variable from mud to rubble (end members are gentle muddy slope with much slumping and sandy slope with steep, planar foresets). Biota: Mostly redeposited shallow-water benthos, some deep-water benthos and plankton. 5) Reefs of platform margin (Figures 2-2, 2-3) Setting: (I) Organically stabilized mud mounds on upper slope or (II) ramps of knoll reefs and skeletal sands or (III) wave-resistant barrier reefs rimming the platform. Sediments: Almost pure carbonate of very variable grain size. Most diagnostic are masses or patches of boundstone or framestone, internal cavities with fillings of cement or sediment, multiple generations of construction, encrustation and boring and destruction. Biota: Almost exclusively benthos. Colonies of framebuilders, encrusters, borers along with large volumes of loose skeletal rubble and sand. 6) Sand shoals of platform margins (Figure 2-3) Setting: Elongate shoals and tidal bars, sometimes with eolianite islands; above £air-weather wave base and within euphotic zone, strongly influenced by tidal currents. Sediments: Clean lime sands, occasionally with quartz; partly with well-preserved cross bedding, partly bioturbated. Biota: Worn and abraded biota from reefs and associated environments, low-diversity in-fauna adjusted to very mobile substrate. 7) Platform interior-normal marine Setting: Flat platform top within euphotic zone and normally above fair-weather wave base; called lagoon 17

18

Schlager

when protected by sand shoals, islands and reefs of platform margin; sufficiently connected with open sea to maintain salinities and temperatures close to those of adjacent ocean. Sediments: Lime mud, muddy sand or sand, depending on grain size of local sediment production and the efficiency of winnowing by waves and tidal currents; patches of bioherms and biostromes. Terrigenous sand and mud may be common in platforms attached to land, absent in detached platforms such as oceanic atolls. Biota: Shallow-water benthos with bivalves, gastropods, sponges, arthropods, foraminifers and algae particularly common. 8) Platform interior-restricted Setting: As for facies 7, but less well connected with open ocean so that large variations of temperature and salinity are common. Sediments: Mostly lime mud and muddy sand, some clean sand; early diagenetic cementation common; terrigenous influx common.

usr•

OPEN SU SHHF

I - 1110£ lflTS

!

OUP SHUF

•uau

,

fORESlOPf

ORGANIC IUILD UP

•

~

,......._

~-

-

r-"\

' ,.......,.

'

I

···"'"':·· ·........ •••••••• '

II\

• •. • • • • : •

•

.

~

••••••

VI RY MAR ROll tH TS DElliS FlOIIS AND TURIIDITU ,. fll[ UHIWATl STUU. MOUlDS 011 IOE OF SlOPE.

GIANT TAlUS ILOCU, UfllUD lUG[ UYiliU.

DOMISLOP[ IIOUIOS.

mn lw;mm.

~m:nmN m~r

I

DOWNSLOPE NOUWOS.

•ur ••ous.

IOUIDSTDIE PUCHU. ru•cuc AID UIIIU flAil£ IIOU RUF SPUI UD IlDDY£,

UlJNAS

... • 1 ' ~ ... "~ "'.."' .. ". ..... ;''2 ~

• • I ·~~ ,-..... •• ••••• ,..,..• -~ ,. • ~ ..L :I~~~-

. . . . . . ,-. .

r---~

SHElF UGOON OP£1 CIRCUlATION

IIINROIIfD EDGE SUDS

s

r"

.......-:::.-::CS~ ~

Biota: Shallow-water biota of reduced diversity, but commonly with very large numbers of individuals; typical are cerithid gastropods, miliolid foraminifers. 9) Platform interior-evaporitic Setting: as in facies 7 and 8, yet with only episodic influx of normal marine waters and an arid climate so that gypsum, anhydrite or halite may be deposited besides carbonates. Sabkhas, salt marshes and salt ponds are typical features. Sediments: Lime mud or dolomite mud along with nodular or coarse-crystalline gypsum or anhydrite; intercalations of red beds and terrigenous eolianites in land-attached platforms. Biota: Little indigenous biota except blue-green algae, occasional brine shrimp, ostracodes, mollusks. The standard model says nothing about windwardleeward differentiation. Most platforms develop asymmetries in response to dominant wind directions (Figures 2-4, 2-5). Seismic surveys reveal these asymmetries better than most other techniques.

~

A

A

!;

IIIDE tflTS ISLUOS, DUNES, IAIIIU IUS, 'ASSES All tHAllUS.

Tl DAL DELTAS. LUOONAL PDIDS, TY'IUl SKUF IIOUIDS,

COUIIWU AlCAL !IUS, CHARNELS liD TIDAL IUS OF lllll SAIID.

TIDAL fUTS. CNUIELS IATUIAL lUllS, POIDS. AliAL IIAT

arns.

ARHYDIIU DOIIES. JEP[[ SnutTlM£1 UIIIWATU UUSTS OF IYPSUII. SAUIAS (UAPOWAfJU

li=·

cmmmn ~ PU.DS)

SAUIIAS

Z IIICIICIIIOCWTII 4 IIOCUSTJC 1 IGUIDSTONI I ~~~~~~TIC CALCJSILT I COATED, ll TIIOCI.ASTIC I IIHOLE SHELLS J PIUCIC IIORII, "IC-CCIA. Ill IIICRITE IJCIITE IIOCUSIIC liTIIDCI.ASIIC 9 IIOCLASTIC 4 IIIIClASTIC· GUJISTOI£. CDIIGI.OIIIIAT£. !COQUINA JIIICKESTOIII: liTHDCUSJIC 0 COAT to "JCIIOeii[CC lA (SHELL USN) 5 IIOCLASTIC IAOIOUIII CUllS II IUIISTDIE· SHAlE lUCillE PACISTOIII. FlOAT STOll.

I SPICULIT£ Z IICIDIIOCUSTIC tALC !Silt 3 PElAIIC IUCIIT(

I IEIF

RUDSTOII.

'mm"~g·· CUINSTOIII

ltDQUUA (SNIU MSN) 3 OIIOIDAl IJOtlASTIC UAIISTOIE e Ul IIIECCIA S OOliTE

1B VHOll SMELLS II, !r, II ~m:m?LITIC ,. llltalll ltr£1ESTUl ~ IIOCUSTIC PUOIIAL ~ ••• u•t•Au, VAUESTOII£ PHil IIICIITE UIIIIUI otOA1£0 llltllll WODUUI-PUil QUillS II HUDSfOIIE U UTUOLITMIC UAII[lS "IUITE AIIIYDIIIU. I 'IUPUIT[ 1 PGIIIOSTIM 711APUTOII SELUITE IJCIIIU OIIOJDS II ILADES IN J 101 lAIIIIAT£ IIICIITI. IIICIIf£ PUll[ IIICIITI UllOIIAl MSYCUIIM:EM IIICIITE CUINSTOIIE

.,.....,

..

I !

! .."' ~

Fi~ure 2-1. Synopsis of standard facies belts reviewin~ second-order bodies of sediment and standard m1crofacies associated with each belL (Reproduced with permission of Springer Verlag)

Figure 2-3. Rimmed margins are common in carbonate platforms and re~uire certain adjustments in seismic statigraphic techniques. Reefs are important rim builders (a), but early-hthified stacked sand shoals can be equally efficient in defending platform margins (b). (After James & Macintyre, 1985. Reproduced with permission of Colorado School of Mines Press)

The Standard Facies Belt 19

a)

b)

REEF MOSAIC

ATTACHI!D II!GIIENTI!D CAI.CAJIIEOUI IIENTHOI

BAFFLES TONE

BINDSTONE

FRAMESTONE

CAVITIES WITH INTE-L

llf:DIIIENT

OOIIIAL a IIAIIIYI III!TAZOA

REEF LIMESTONE

FLOATSTONE

RUDSTONE

Figure 2-2. Recognition of ecologic reefs in the geologic record depends on a variety of features; most of them require outcrops or cores as an observational basis. This sketch illustrat~s characteristic aspects of reefs. Note particularly the multiple successions of different encrusters and frame builders as well as cavities with mtemal sediment and cement. Obviously, outcrops or cores, not seismic profiles, are required to apply these criteria. (After James & Macintyre, 1985. Reproduced with permission of Colorado School of Mines Press)

SHOAL - RIMMED PLATFORM

REEF - RIMMED PLATFORM

REEF - RIMMED PLATFORM

20

Schlager

WINDWARD

z

I!DGI!

PRO FILl!

A -0

I GEN(RAI..ISEDI

Figure 2-4. Sediment tvpes and patterns on Bu Tini Shoal. (From Purser et al., 1973). Windward-leeward asymmetries such as this one probably are common in the geologic record. Seismic grids are one of the best techniques to reveal them. (Reproduced with permission of Springer Verlag)

The Standard Facies Belt 21

ROCK RIDGE (EOLIAN?)

D

HCX...OCEf\E SAND

g

HOLOCEI\E REEF

~ PLEISTOCENE ROCK

500M~:

I}}DI

HOLOCENE SAIII>

~ HOLOCEI\E REEF ~ PlEISTOCENE ROCK

Figure 2-5. Windward and leeward platform mar~ins in the Bahamas-a comparison. Note active reef growth along windward margin (upper panel), reefs buned in sediment on leeward margin (off-bank sediment transport). After Hine & Neumann (1977).

CHAPTER3

RHYTHMS AND EVENTS IN CARBONATE DEPOSITIO N Shallow water carbonates rarely ever accumulate in a uniform, steady fashion. Rather, the record shows a hierarchy of rhythms on time scales of thousands to hundreds of millions of years. These rhythms are punctuated by singular events and overprinted by the unidirectional changes of organic evolution. Below we briefly discuss some of these patterns and processes. They fall outside the domain of classical sequence stratigraphy but we believe that the seismic stratigrapher should be aware of them as they may well influence sediment production and sequence anatomy in reefs and carbonate platforms.

in many instances (e.g., Fischer, 1964; Read & Goldhammer, 1988; Goldhammer et al., 1990; ). These observations imply that the frequency bands of orbital cycles and stratigraphic sequences broadly overlap and the two approaches complement each other.

Autocycles Under most circumstances, depositional systems respond rhythmically to linear forcing, i.e. they fall behind, then catch up, overshoot, and fall behind again (Figures 3-2 and 3-3). Besides this non-linear response of an entire system in the time domain, we also find space rhythms within a system-mud banks, tidal passes, delta lobes etc. typically form rhythmic patterns in space. Migration of these spatial patterns can also produce a rhythmic sediment record (see Figure 3-4). Both effects have been invoked to explain rhythmicity in the geologic record. The notion of autocyclicity is !Particularly popular for short time periods of 103-10 years, the typical domain of parasequences. However, there is no reason to dismiss autocyclicity as an explanation for longer-term rhythms in the geologic record. The emphasis on short-term autocyclicity is probably an artefact of our strong observational bias towards the Holocene.

Milankovitch cycles The perturbations of the Earth's orbit, their influence on climate, sea level and the sediment record have received much attention recently. Reefs and platforms are particularly sensitive to Milankovitch forcing because they respond to both sea-level cycles as well as environmental change driven by the orbital perturbations. Sea-level signals are best recorded in the shallow lagoons and tidal flats of the bank tops. Figure 3-1 illustrates that this record strongly depends on the interference of various frequencies (as well as on subsidence). At slow subsidence (and large amplitudes in the low frequencies) the record may be erratic in spite of regular orbital forcing. The most pronounced orbital cycles have periods of ca. 20,000 to 400,000 years, but the full spectrum of orbital perturbations ranges from 101 to 106 years. Orbital cyclostratigraphy is a rather sophisticated technique and we lack the time to adequately cover it in this course (see Fischer & Bottjer, 1991 and Goldhammer et al., 1991 for overviews). However, the approach has great potential in stratigraphy and is rapidly expanding into the realm of sequence stratigraphy. Recently, many sequence stratigraphers reported much higher numbers of sequence boundaries than previously observed. For instance, Van Wagoner et al. (1990, p.52) estimate that type-1 unconformities are spaced at intervals of 100,000-150,000 years and that global curves such as the one by Haq et al. (1987) refer to sets of sequences. The carbonate record confirms this notion

Organic evolution The organic origin of most carbonate sediment makes carbonate depositional systems especially sensitive to evolutionary changes in the biota. Figures 3-5 and 3-6 depict some of these changes for reef biota. It is likely that they modulate the growth potential and the growth geometry of reefs and platform margins, governing the distribution of, for instance, drowning events or progradation phases through time. Bosscher & Schlager (in press) found that the observed maximum accumulation rates of platforms go through a minimum in the wake of mass extinctions of reef biota, suggesting some evolutionary control on the carbonate growth potential. James & Bourque (in press) emphasize the difference between reef-building

22

Rhythms and Events in Carbonate Deposition 23

and mound-building platforms and their irregular distribution through time. Chem~cal

evolution

The composition of the ocean and atmosphere is determined largely by the rates of recycling and plate motion. Fischer (1982) proposed that during the Phanerozoic, the Earth has oscillated between an icehouse state with cold poles and maximum latitudinal temperature gradients and a greenhouse state with

warm poles and reduced temperature gradients. Fischer argued for an endogenic control of the icehouse-greenhouse cycle, because it seems to correlate with variations of magmatic activity and sea-floor spreading. Figure 3-7 shows that the calcite/aragonite ratio in carbonates also seems to vary in accordance with the icehouse-greenhouse cycle. In this way, the growth potential of carbonate systems as well as their potential for diagenetic alteration may be tied to changes in the ocean environment. (See Chapter 6 for follow-up.)

Fl SCHER CYCLES PERIODIC PLATfORM CARBONATE DEPOSITION IN TUNE TO MILANKOVITCH RHYTHMS

£> P SUAE

DEPOSITION

Of PLAlFOIIM

ON PLATfORM

APERIODIC PLATFOIU4 CARBONATE DEPOSIT I ON WITH MILANKOVITCH SEA LEVEL OSCILLATIONS TIME --200,000 Vll---1•00,000 Y l l - - - - 0 V I I -

Figure 3-1. Scenarios for deposition of carbonate cycles dictated by Milankovitch rhythms. The top panel shows the drowning and exposure of a carbonate platform due to glacio-eustasy produced by superimposition of 20,000-year and 100,000-year Milankovitch pulses. Small variations in the amplitude of the 100,000-year pulse produce bundling of tidal-flat cycles into clusters of five. With large variations in the amplitude of the 100,000-year pulse, as in the Pleistocene, deposition on carbonate platforms is likely to be erratic and will not easily reveal the Milankovitch control, as depicted in the lower panel. (After Hardie & Shinn, 1986. Reproduced with permission of Colorado School of Mines Press)

24

Schlager

Q) DROWNING (SUBSIDENCE »CARBONATE <:;:::J SHORE LINE TRANSGRESSES

@ PROGRADATION

PRODUCTION)

(CARBONATE PRODUCTION > SUBSIDENCE)

.. , ...···· LAGOON CARBONATE FACTORY AND DISPERSAL SYSTEM IN FULL OPERATION

w u 2 w 0 (f)

w ~

1-

CD ~ (f)

@PROGRADATION STALLED !CARBONATE PRODUCTION~ SUBSIDENCE) CARBONATE FACTORY TOO SMALL SUPRATIDAL CAP FF'I lENT OPERATION ..... ·. ,• ·.··.···~-. A-B-C SHALLOWING UPWARD SEQUENCE

(f)

:J 0 ~

2

i=

z

0 u

@

DROWNING (SUBSIDENCE »CARBONATE PRODUCTION)

~SHORELINE TRANSGRESSES WATER TOO SHALLOW FOR EFFICIENT OPERATION OF' CARJ_ONATE FACTORY

@PROGRADATION (CARBONATE PRODUCTION > SUBSIDENCE) ~ SHORELINE PROGRADE$ OMISSION SURFACE (SUBTIDAL L_ DIRECTLY ON SUPRATIDAL) _.. •.. 'I'. . ..

STACKING OF A-8-C SHALLOWING UPWARD SEOUEttCES

Figure 3-2. The autocyclic model of R.N. Ginsburg. Ginsburg (1971) proposed an autocyclic model that elegantly explains superimposed A-B-C shallowing-upward sequences m platform carbonates. The model calfs on continuous, steady basinal subsidence coupled with carbonate sediment supply rates that are selfregulated by the extent of progradation. The crux of the model lies in the recognition that carbonate sediment production rate will fall behind subsidence rate as progradation drastically reouces the size of the shelflagoon "factory'' (see above), and will not recover until the lagoon once again becomes deep enough (due to subsidence) to allow efficient carbonate production and sediment dispersal (perhaps 1-2 meters). This factor introduces an automatic lag time in deposition so that subtidal sediments lie directly on supratidal sediments. (Hardie & Shinn, 1986. Reproduced with permission of Colorado School of Mines Press)

Rhythms and Events in Carbonate Deposition 25

AUTOCYCLES OF SEDIMENT SYSTEMS

linear forcing by

t

sea level

\\

rythmlc response

of sedimentation

horizontal distance - .

EveRAARs et at.

time-. Figure 3-3. Examples of autocycles on platforms generated by the computer program MAPS of Demicco & Spencer (1989). C)_•cles consist of shoaling-upward sequences with supratidal flats (dark) and subtidal (light gray). Autocyclictty is poorly understood and may operate also on time scales of millions rather than thousands of years.

Erosion of foreshore

& shoreface _ _ _ Newly prograded mud coastline

5m

layers

Discontinuities

VERTICAL

IXAOGf."'ATtON

10 0.

Figure 3-4. Autocycles generated by migrating mud banks on the coast of Suriname. Stacking of sediment from migrating mud banks creates on the shoreface a vertical sequence of laminated and bioturbated muds with discontinuity features and on the coastal plain a horizontal sequence of mud marshes and sand cheniers. Average duration of a cycle is 30 years. (After Rine & Ginsburg, 1985. Reproduced with permission of Society for Sedimentary Geology)

26

Schlager

PERIODS

MAJOR SKELETAL ELEMENTS

BIOHERMS 6

tn

z

5

..J ..J

4

~

3

0

-

ztn

CORALS

T

I

I I

-a: 4(

~~

uo -

300

REEFS

t-~o~.

2

I

1

I

CJ

0 ..J 0

REEF

C)

OUNDS

UJ

PRECAMB Figure 3-5. An idealized stratigraphic column representing the Phanerozoic and illustrating times when there appear to have been no reefs or bioherms (gaps), times when there were only reef mounds,· and times when there were both reefs and reef mounds. Bars on right side indicate occurrence of "empty buckets.'' They tend to coincide with intervals of full reef development. Arrows represent times of important faunal extinctions with the potential to generate sequence boundaries (after James, 1983).

CYCLE II CYCLE I

UOISTS

Figure 3-6. A generalized plot of the main biotic constituents of carbonate buildups against time for the Phanerozoic. The balloons represent the relative abundance and importance of the different taxa (from James, 1983). (James and Macintyre, 1985. Reproduced with permission of Colorado School of Mines Press)

Rhythms and Events in Carbonate Deposition 27

ICEHOUSE

I

GREENHOUSE "- CLIMATIC EPISOOES '-.:.FISCHER SEA LEV£L FLUCTIJA TIONS HALLAM VAIL. ET AL.

~

Hlgh-Mg Calcl11 and, 1111 abvndatttly. Aragonite

II Cttclle; Mg content oenarelly lower. lncraa••no towtrd

"ThtllhOid"

Figure 3-7. Icehouse-greenhouse cycles of the Phanerozoic are correlated to change in composition of inorganic carbonate precipitates and to mineralogy of organic carbonate production. This illustrates the close connection between carbonate production and ocean environment. (After Fischer, [1982], Sandberg [1985]. Reproduced with permission of Macmillan Magazines Ltd.)

CHAPTER4

SEQUENCES AND SYSTEMS TRACTSA SEDIMENTOLOGIC VIEW Stratigraphic sequences and systems tracts as defined in sequence stratigraphy (Figure 4-1) are lithostratigraphic concepts and as such interpretable in sedimentologic terms. We believe that sedimentologic analysis is the most appropriate way to reveal the primary controls behind a certain feature of sequence stratigraphy. This, in turn, allows one to assess the role of sea level and other factors in building the sequence record. When sedimentologic analysis is performed, it turns out that very often sea level is not the only process that may generate sequences and systems tracts. The classical interpretation of sequences and systems tracts in terms of sea-level cycles (Figure 4-2) is a possible, but generally not a unique solution. In particular, changes in volume and composition of sediment will often have similar effects as sea-level fluctuations (Figure 4-3). Whether these alternative controls do apply, has to be determined specifically for each instance. Chapter 6 discusses examples of sequences created not by sea level but by other processes. Here, we will examine some pivotal concepts and terms of sequence stratigraphy and seismic stratigraphy to improve our understanding of their genesis.

formities." Sequence boundaries are defined as "observable discordances in a given stratigraphic section that show evidence of erosion or nondeposition with obvious stratal terminations, but in places they may be traced into less obvious paraconformities recognized by biostratigraphy or other methods." At this point, it is again useful to consider the basic sedimentologic meaning of seismic unconformities: they represent changes in the pattern of sediment input and dispersal in the basin (Schlager et al. 1984, p.725; Schlager 1989; see Galloway, 1989, for similar emphasis on "reorganization"). Sea level is an important control on input and dispersal patterns, but so are tectonic changes in sea-floor topography, tectonic changes in the hinterland [drainage patterns], Figure 4-6), changes in ocean environment (Figures 4-5, 6-6) and autocyclic processes in sedimentation (Figures 3-3, 3-4). This sedimentologic interpretation of sequence boundary is compatible with the definition of sequence and sequence boundary by Vail et al. (1977, p.53). It is meant as a complement rather than a replacement of the original definition. A growing number of basin studies reveal a rather complex interplay of eustasy, regional tectonics and sedimentologic processes as control on depositional sequences (e.g. Underhill, 1991). The original definition along with the above interpretation are broad, yet specific enough to address this superposition of effects. Van Wagoner et al. (1988) restrict the term unconformity and sequence boundary to surfaces that include significant subaerial exposure. This burdens the term with a genetic connotation that is often difficult to verify in the field and nearly impossible to verify in seismics. Furthermore, this restrictive definition of sequence boundary "defines away" any alternative to sea level in the interpretation of sequences and sequence boundaries (see Schlager, 1991, for discussion). The successful application of sequence-stratigraphic techniques to terrestrial deposits far removed from fluctuating sea level or lake levels, merits mention in this context (e.g. Hanneman & Wideman, 1991). It illustrates that packaging of sediment accumulations into unconformity-bounded sequences is a

The sequence boundary Despite the subdivision of sequences into systems tracts and the recognition of other surfaces (such as the maximum flooding surface), the unconformable sequence boundary has remained the key element in sequence stratigraphy. The limited resolution of seismic profiles very often does not allow one to identify the systems tracts. What generally can be done, however, is to recognize sequences and sequence boundaries as unconformities and sediment packages with internally coherent bedding patterns respectively. We strongly recommend using the original definition of sequence and sequence boundary by Mitchum in Vail et al. (1977, p.531). Mitchum defines sequence as "a stratigraphic unit composed of a relatively conformable succession of genetically related strata and bounded by unconformities or their correlative con-

28

Sequence and Systems Tracts 29

general principle that only happens to have been discovered in the sea-level domain.

Ecologic reef, stratigraphic reef, seismic reef

Seismic unconformity .vs. outcrop unconformity

"Was this reef really a reef?" This question continues to fuel heated discussions among geologists (see James & Bourque, in press, for review). Dunham (1970) introduced a novel perspective in this debate when he proposed to qualify the term and distinguish between ecologic reefs, built and bound by organisms, and stratigraphic reefs where the binding is done by cementation. This distinction has proved very useful but the increasing use of seismics to identify reefs has complicated matters again. The "seismic reefs" defined by geometry and reflection character do not correspond to either ecologic reefs or stratigraphic reefs. Seismics tends to overlook certain reefs and, in other circumstances, add non-reef deposits to the reef. Living parts of reefs are small by seismic standards, easily destroyed and quickly buried by their own debris. The limited resolution of the seismic tool reveals reefs only where many generations of reef growth were stacked vertically to thicknesses of hundreds of meters. Where the locus of reef growth has been forced to shift laterally, reefs form small lenses embedded in detritus sediment. These lenses will be seen by the field geologist but not by the seismic tool. (Where porous, these reefs will also be "seen" by migrating hydrocarbons, of course.) Equally important are those situations where the seismic tool shows more "reef" than is actually there in a stratigraphic or ecologic sense. Reefs defined by their geometry and the incoherent reflection character normally include, besides ecologic reefs, also the reef aprons on the landward side as well as the coarse forereef debris where it is unstratified (Figure 4-14). In addition, sand shoals interfingering with reefs may be included in the seismic unit.

Perhaps the most significant contribution of sequence stratigraphy is the discovery that sedimentary basin fills are dissected by unconformities that can be used to subdivide the sediment accumulations into mappable, genetically coherent units. Unconformities, of course, have long been recognized in outcrop. One must emphasize, though, that seismic unconformities and outcrop unconformities do not match. Three different situations may occur: (1) Boundaries where in outcrop and seismic profile the time lines abut against a surface. This type is unproblematic because both the field geologist and the seismic interpreter will recognize this feature as an unconformity. (2) There are outcrop unconformities that cannot be seen by the seismic tool because their geometric expression consists of a microrelief on the scale of centimeters or decimeters, while the large-scale bedding of the two units remains parallel. This is common in deep-sea sediments. (3) Finally, the seismic image may show unconformities that correspond to transitional boundaries in outcrop. Here, two radically different situations must be distinguished. (3a) Seismic unconformities where time lines converge into an interval of continuous, but very slow sedimentation. The limited resolution of the seismic tool may portray this thin interval as a lap-out surface. The condensed interval meets the definition of unconformity by Vail et al. (1977) in so far as all deposits below it are older than the oldest deposits above the interval. The difference with this definition lies solely in the fact that "non-deposition" needs to be replaced by "slow deposition". For the seismic interpreter this is a minor difference, but one that has to be kept in mind when seismics is tied to boreholes or outcrops. The deepening-upward succession on many drowned platforms is an example of this "transitional seismic unconformity" (Figure 5-10; Erlich et al., 1990, figure 11; Campbell, in press). (3b) The most disturbing kind of seismic unconformity is the pseudo-unconformity, where a rapid facies change occurs in each bed at a similar position and the seismic tool merges these points of change into one reflection. Time lines cross this reflection, thus it is not an unconformity in the sense of Vail et al. (1977) and needs to be recognized and eliminated before sequence analysis is done (Figures 4-8 through 4-13). Carbonate platforms are particularly prone to this effect because rapid facies changes may occur on several places in the edifice and may go hand in hand with significant changes in physical properties (e.g. carbonate-shale transition at the toe of slope, or the boundaries of reefs).

Valley cutting, fan building The model of systems tracts assumes that sea level controls the presence or absence of submarine fans as well as their position higher or lower on the toe-ofslope. A view of modern oceans and hydrodynamic theory indicate that fan development is largely a function of the transport capacity and competence of turbidity currents. These parameters, in turn, depend on sediment supply and topography (slope angle, size of valleys to channel the flows etc.). Sea level can influence sediment supply and, to a lesser extent, topography but in doing so it must compete with other processes. For carbonate platforms, Schlager & Camber (1986) have proposed a model of slope evolution as a function of the height of the platform (Figure 1-21). This model predicts that the depocenter will shift basinward as the slope angle increases. Steep slopes can be bypassed and even eroded by turbidity currents and fans can onlap these slopes irrespective of sea-level position. The position and shape of the gravity-dis-

30

Schlager

placed sediment bodies is largely a function of topography (Figures 4-15, 4-16). Carbonate sediments, in particular reefs, are known for their ability to rapidly build steep relief (e.g. Figures 1-14, 1-15). Depocenters of gravity-displaced sediments will shift equally rapidly in response to the changes in sea-floor relief. Thus, fans and submarine canyons in carbonate terrains are probably rather unreliable indicators of sealevel movements.

Abrupt breaks versus gradual shifts in sequences The classical sequence model is characterized by abrupt breaks at the sequence boundary, and, to a lesser extent, at the transgressive surface and the maximum flooding surface. The punctuation of the record and the concomitant asymmetry is well displayed in the sawtooth pattern of the coastal onlap curve (Vail et al. 1977) as well as the pattern of systems tracts (Figure 4-1). There can be no doubt that these punctuations and asymmetric patterns exist (e.g. Van Wagoner et al., 1990). However, there also exist many well-documented examples of near-symmetrical cycles and gradual shifts of facies belts in the 105-106 year domain, e.g. Figures 4-10a, 4-17, or Kauffman et al., 1977 on the Cretaceous of the Western Interior. The tendency towards gradual shifts and symmetrical cycles is most pronounced in deeperwater settings, whereas near-shore environments

favor the development of erosional breaks that, in turn, generate asymmetries by preferentially destroying one limb of a cycle (e.g. Kauffman, 1984 on the Western Interior Basin). Recently, the case for gradual shifts and symmetrical cycles has been strengthened by detailed studies on stacking patterns within loworder cycles. It turns out that the stacking patterns in low-order cycles often reveal gradual changes even where the individual high-order cycles are punctuated and asymmetric (Figure 4-18; Read & Goldhammer, 1988; Goldhammer et al., 1990; Goldhammer et al., 1991). It seems that the sediment record in the millionyear domain is less punctuated and more symmetric than the cycles in the ten-thousand year domain. In view of what has been said above about the seismic tool displaying rapid transitions as unconformities, one may ask to what extent the punctuated nature and pervasive asymmetry of the sequence-stratigraphic models is an exaggeration inspired by the peculiarities of the seismic tool. Sequence stratigraphy, to be sure, can be practiced with symmetric cycles too. Goldhammer et al. (1991) have demonstrated this in exemplary fashion. However, in the absence of drastic events, correlation of sequence boundaries and systems tracts becomes much more arbitrary and sequence-stratigraphic analysis converges with the tried and true sedimentologic practice of delineating shoaling, coarsening and thickening trends or their reverse, in a section (e.g. Goldhammer et al., 1991).

Sequence and Systems Tracts 31

INCISED \lALLEY

livfl

A)

IN DEPTH CORRELATIVE CONFORMITY (SEQUENCE BOUNDARY)

UNCONFORMITY

HSTI

CONDENSED SECTION

TSTI

~ SUBAERIAL HIATUS

:

:

UNCONFORMITY~

;: cy: ' ;~-zfii3B -

.,

SBI .r

DISTANCE

....,,.,.

,.,,.

'---

j;* CORRELATIVE CONFORMITY (SEQUENCE BOUNDARY)

BJ R GEOlOGIC TIME

LEGEND

t,:;·::;,.j

if!l C) ( i

J

ALLUVIAL COASTAL PLAIN ESTUARINE/FLUVIAL SHOREFACE/DELTAIC SANDS

l ,JJN)

MARINE SILT, MUDSTONE f=::.::=:=J MARINE SHALE (:·:·:{{!;:) DEEP-WATER SANDS

Figure 4-1. Stratigraphic sequences and their systems tracts as defined in sequence stratigraphy. (After Vail,

1987)

32

Schlager

TIME

--.....-:-t---t:'-r-~

EUSTACY

~ Lowstand SystarM ll'act (lST)

[;j lowstand Basin Floor Fan (bf) !]] Lowstand Slope Fan (If) ~Lowstand Wedge-Prograding Compte• (law)

TECTONIC SUBSIDENCE

~ Transgressive SyaterM ll'act (TST)

0 HlgMtand SyaterM ll'act (HST) D Systenw Tract (SMST)

RELATIVE CHANGE OF SEA LEVEL

Shelf Margin

LST TST HST SMST

LEGEND SYSTEMS TRACTS

SURFACES (SB) SE~NCE BOUNDARIES (SB 1) =TYPE 1 (SB 2) ~ TYPE 2

(Dl.S) DOWNLAP SURFACES (mts) ~maximum flooding (Ibis) ~ top 118M\ noor tan (tats) = top slope tan ...race (TS) TRANSGRESSIVE SURFACE (First lloodlng above

-lac:•

-tac•

-tac•

m..lmum progr..s.tlon}

HST = HIGHSTAND SVSTEMS TRACT TST ~ TRANSGRESSIVE SVSTEMS TRACT lvf = lnciMd wiley fll LST = LOWSTAN) SVSTEMS TRACT 1¥1 = Incited wiley .. llw = lowaUnd wedge-prograding complex If lowetand elope f8n bl lowetand baetn floor tan fc =tancMnnela " ~ f8n Iobei SMST = SHELf MARGIN SVSTEMS TRACT

= =

Figure 4-2. Relation of systems tracts to sea-level cycle as proposed by sequence stratigraphy. (After Vail, 1987)

a)o

bP

E ,.,_

E ,.,_ c .c

c:

.c

a. Cl)

c.

lJ

lJ

Cl)

L

L

0

distance in basin ( k m l -

90

Figure 4-3. Variations in sediment supply and sea level fluctuations have a similar effect on sequence patterns. This example shows the synthetic strati~raphy of the Permian-Triassic Akhdar Group in Arabia. (After Aigner et al., 1991, modified.) (a) Synthetic stratigraphy generated by sinusoidal variations in sediment supply. (b) Synthetic stratigraphy generated by sinusoidal fluctuations of sea level.

Sequence and Systems Tracts 33 -

NW

Figure 4-4. Sequence boundary in deef Gulf of Mexico related to change in sediment mput and dispersal. The carbonate platform on the right was drowned in the wake of a mid-Cretaceous anoxic event and later covered with pelagics. As carbonate input ceased, the debris apron of platform was gradually buried by clashc sediments transported at right angle to the profile. (Campeche Bank, Gulf of Mexico, line courtesy of R.T. Buffler, Texas Institute for Geophysics.)

-

GULF OF MEXICO UTIG Line GT2-3b

4.0

8.0 sec

twt

.2

.1

.3

.4

CONTOURS post-MCU siliciclntics Canom-l.l'lleacane

pre-MCU Cllrbonlltes Albian-early Cenom.

FLORIDA PLATFORM

60

0 KM

CAMPECHE PLATFORM

CUBA .3 .2

.1 0.0

1.0

Figure 4-5. Isopachs above and below the mid-Cretaceous unconformity-a major sequence boundary in the Gulf of Mexico characterized by drowning of most platforms in the Gulf rim. Pre-MCU isopachs reflect the input of carbonate debris from the platforms, post-MCU isopachs reflect the growth of a flysch wedge, deposited in front of the advancing Cuban island arc. The unconformity represents a fundamental change in the sediment-input pattern of the Gulf. (Modified after Phair in Schlager, 1989.)

34

Schlager

"ttlckness• in two-way traveltime (seconds)

SIGSBEE Pleistocene

CNCO DE MAYO late Mocene

tiYOI.Ul Plocene

lPPER

I'.EXICAN RDGES

midde Tertiary (?) to late IViocene

LOWER I'.EXICAN RIDGES earty(?) to mldde

Tertiary(?)

after ShatJ) et al, 1984 Figure 4-6. Isopachs of Cenozoic seismic sequences in the Gulf of Mexico (After Shaub et al., 1984). Sequences were defined and mapped following the principles of Vail et al. (1977). Note that each sequence shows a different pattern of sediment input. The number of sequences and their ages do not correspond well with the Cenozoic sea-level curve of Haq et al. (1987). We postulate that they are controlled by changes in sediment input that reflect tectonic events in the hinterland of the Gulf.

Sequence and Systems Tracts 35 WEST

...

RIMBU MUDOIVIROO«

,, ,...,. WEST

ALBERTA

"DEEP BASIN""

EAST

CH,ji#J

WEST SHALE BASIN

EA$1 SHALE BASIN

Fi~ure 4-7. Generalized sequence stratigraphic cross section of Upper Devonian deposits in central Alberta. Dtrection and location of clinoforms as well as distribution of reefs, platforms and shale wedges indicate that all sequence boundaries shown represent important changes in sediment input and dispersal patterns. This does not exclude sea level as a possible control, but it strongly suggests that other factors besides sea level strongly influenced the sequence pattern. (After Stoakes & Wendte, 1988. Reproduced with permission of Canadian Society of Petroleum Geologists.)

Figure 4-8. The difference between seismic unconformity and outcrop unconformity in the Sella mountains of the Southern Alps. (After Stafleu & Schlager [1991] and J. Stafleu, written communication.) The Late Triassic platform of the Upper Schlem Fm. prograded by many hundreds of meters while aggrading by tens of meters only. Platform growth was ultimately terminated by terrigenous deposition of the Raibl Fm. This situation was surveyed in the field and subsequently modelled seismically (see Figure 4-9).

36

Schlager

a N

0

~

~

0

0

llll.

500.

IIIII.

IIIII.

lllll.

1500.

mi.

lllll.

15111.

2500.

m.

.Jill.

lilll.

MI.

4GIIl.

41111.

b 0

500.

IIIII 1500. 2IJI)

511.

1110. Ill!. 2IJI)

Zlll.

2m

.1111.

Jill

Jill.

».

mi.

41111.

Figure 4-9(a). Outcrops shown in Figure 4-8 were used to construct a lithologic model indicating rapid progradation and an alternative model based on the unlikely assumption of genuine toplap. (After Stafleu & Schlager [1991] and J. Stafleu, written communication.) Figure 4-9(b). Synthetic seismic profiles of the progradation model and the toplap model, using vertical incidence and 20Hz wavelet (top right). Note that the thin undaform beds (=topsets) are not resolved and the synthetic of either model portrays the clinoforms toplapping against the terrigenous unit. (After Stafleu & Schlager [1991] and J. Stafleu, written communication.)

Sequence and Systems Tracts 37

Figure 4-9(c). Synthetic profile of the progradation model (time section) at 20Hz (top) and 50 Hz (bottom). At 50 Hz the transition from clinoforms into undaform beds is visible as a separate event below the base of the terrigenous umt. However, the relationship between dinoform and undaform bedding is shown as toplap, i.e., as a pseudounconformity. (After Stafleu & Schlager [1991) and J. Stafleu, written communication.)

Schlager

38

a

RAIILGROUP

0 fl) fl)

w

z

~

0

:E

a:

5 1-

1-

DISTANCE km

b

0.1

w :&

~ 0.2--

>-

-c ~I

0 0.3

3:

1Cl)

Q

z

0 0

w

o;:;

(I)

0.5

Figure 4-10. Pseudo-unconformities in seismic models of outcrops (after Rudolph et al., 1989; Schlager et al., 1991; Biddle et al., 1992). (a) Platform-to-basin transition, Picco di Vallandro, Triassic, Southern Alps. (b) Seismic model of platform-basin transition, showing several unconformable relationships. Only one of these seismic unconformities (open arrows) was portrayed as an unconformity in the lithologic model; all other lapout patterns correspond to rapid lateral facies changes and represent pseudo-unconformities (black arrows).

Sequence and Systems Tracts 39

F

100M

B

G

F

.650

D ......

fl)

w .7001 ~

i=

.750 '.· f.'.

i

100 M

Figure 4-11. Example of pseudo-downlap surface in the Triassic platform-basin transition. Change from carbonate muds and breccias into (mostly terrigenous) basin sediments at base of a prograding platform slope is shown as pseudo-downlap. (After Rudolph et al., 1989. Reproduced with permission of Geological Society of America)

40

Schlager

o-

D A

-

200m

IOULDERSI"--J DOLOMITE 'r-:i. LIMESTONE

~

INTERBEDDED LIMESTONE,

ARGILlACEOUS LIMESTONE, AIIIO MARL ~

100M

Figure 4-12. Example of pseudo-onlap in the Triassic platform-basin transition. Toe-of-slofe retreats but keeps shedding debris in the form o meter-thick tongues of megabreccias that extend into the basin. The seismic model images this situation as onlap of basin sediments on the clinoforms. (After Rudolph et al., 1989. Reproduced with permission of Geological Society of America)

Figure 4-13. Increased seismic resolution can solve the problem of pseudo-unconformities. When frequency is increased from 25 Hz to 100Hz, the pseudo-onlap of the Triassic example is correctly displayed as an interfingering pattern. (Mter Staileu & Schlager et al., 1991)

Sequence and Systems Tracts 41 AG-4

sw

"'uw

NE

2

"'w ::1!

f: z

3

0

f:

~

..J

4

IL

w a:

·~-5

8

Fi~ure 4-14.

Margin of the West Florida :platform in seismic profile. The margin shows a "seismic reef" (labeled as "margin facies") with elevated top and incoherent reflections, but dredge hauls and submersible dives revealed this "reef'' to consist of platform-interior deposits (after Corso et al., 1988). Seismic reefs are not necessarily ecologic reefs in outcrop.

t3o~---SLOPE

120

e

•• ,,

&,,,,

fl·~;,

c: tOO

,,'

10~

j :: 10

, ...

1

,~---

,.,

E

!i

,,. ..

~,~':

. >: 110 .§

. . . 1. . ............... ______________ ._I

-;.. BASIN

•

:

: : :

~

...__.LIB TRANSECT, 2-3•

•--•I!XUMA TRANIECT, 10-11°

g,~~~ ~.~ . -· . ...

~~~!!!------

\

-------~-~. ••

------·

•

Km

Figure 4-16. Accumulation rates of various Neogene time intervals of the Little Bahama Bank and Exuma Sound slope transects. Note that on the gentle slope (i.e., LBB), sedimentation rates are highest on the slope and decrease basinward. On the steep Exuma slope, rates in the basin are higher than on the slope, which is interpreted as evidence for more efficient basinward transport by turbidity currents with increasin~ slope angle. This effect explains the different positions of the highstand wedges in Figure 4-12. (After Austin, Schlager et al., 1988. Reproduced with permission of Ocean Drilling Program)

ISQm

U

ns

300

Figure 4-15(a). Holocene highstand wedge on a gentle, accretionary slope. Note maximum thickness on upper slope. (Western margin of Great Bahama Bank, from unpublished surveys by W. Schlager and A.W. Droxler, interpretation as in Wilber et al. 1990.) Figure 4-15(b). Holocene highstand wedge at the foot of a steep, erosional slope. Pedro Bank on Nicaragua Rise. (After Glaser et al., 1991). In comparing (a) and (b), note that position of sediment wedge is a function of slope angle rather than sea-level position. Both wedges formed in the last 8000 years. Reproduced with permission of Elsevier Science Publishers.

42

Schlager

FACIES MOSAIC ACROSS DEPOSITIONAL STRIKE LOWER ORDOVICIAN, CENTRAL APPALACHIANS EAST

275 KM ------------~

(RESTORED)

v;0:: 500 w w

1-

TIDAL

~ 400 (/) (/)

w

~ 300

(/)

w u

u

~ z

J: II-

z

200

w (/) w

:f

I 00

CLEAR SPRING

MD

HAGERSTOWN

MD

FREDERICK

MD

Figure 4-17. Gradual facies shifts in Early Ordovician carbonates of the Appalachians. Note absence of abrupt breaks that could represent sequence boundaries. The gradual shifts in the million-year domain contrast with the largely asymmetric, punctuated cycles in the ten-thousand-year domain. (After Hardie &: Shinn, 1986. Reproduced with permission of Colorado School of Mines Press)

Sequence and Systems Tracts 43

b history of sea level sedimentation subsidence synthetic stratigraphy

-..., I ll ~

TST

...

-

-sb•!

0 Ill

c

~

0

i1 Ii :J Condensed Megacycle

E II>

E

=: J Megacyle Rhythmic

HST

-J

I'QfS- -

-

Amalgamated Megacycle

TST

0~~

-25m

+25m

Sea Level Amplnude

Figure 4-18(a). Gradual and largely symmetrical changes in cycle-stacking patterns, Triassic Latemar platform, Dolomites, Northern Italy. (After Goldhammer et al., 1990.) Cycles change from mainly subtidal deposits at the bottom (LPF) to subtidal-supratidal cycles (LCF) to tepee-rich cycles with thin subtidal members (TF) and back to subtidal-supratidal cycles (UCF). The pattern shows a rather symmetrical swing from subtidal to exposure-dominated facies and back. Figure 4-18(b). Model of the Latemar cycle stacking produced by computer program "Mr. Sediment" (after Golhammer et al., 1990). Synthetic stratigraphy on the left and model of sea level, subsidence and sedimentation on the right. Sequence-stratigraphic classification of synthetic stratigraphic column: TST, HST =transgressive and higlistand systems tract, respectively; sb =sequence boundary; mfs =maximum flooding surface. Note that with gradual change such as these, the sequence boundary and the maximum flooding surface are no longer unique horizons, their position becomes to some extent arbitrary.

CHAPTERS

CARBONATES VERSUS SILICICLASTICS IN SEQUENCE STRATIGRAPHY The principle of depositional bias

leeward margin in Figure 5-1). However, even in these loose-sediment dominated prograding margins, one finds reefs or sand rims. We believe that the shelf-margin elevation of these prograding platforms yields the most reliable sea-level record in carbonate seismic stratigraphy (e.g. Eberli & Ginsburg, 1988; Sarg, 1988).

In chapter 4 we concluded that sea level fluctuations are but one of several ways to generate sequences and systems tracts. This claim is based on a sedimentologic evaluation of sequence anatomy. Where sea level does exert the dominant control on sequence stratigraphy, sedimentology must be called upon again, this time to assess the differences of the sea-level records of depositional systems. Siliciclastics and carbonates are the two most prominent examples that come to mind, but evaporites, too, have their specific ways of recording sea level. Depositional systems resemble newspapers that all report on the events of the day, but each with different editorial bias. It behooves the reader to learn about the editorial bias of his paper. Similarly, the geologist ought to know about the bias of depositional systems in recording changes of sea level (and other environmental factors). Below, we discuss some peculiarities of carbonates vis a vis siliciclastics.

Highstand shedding It is a well-established fact that in the Pleistocene, siliciclastic sediment supply to the deep sea was at its maximum during glacial lowstands of sea level. Rimmed carbonate platforms were in antiphase to this rhythm. They produced and exported most sediment during interglacial highstands when the platform tops were flooded. This "highstand shedding" is best documented for the Bahama Banks (Figure 5-2; Droxler et al., 1983; Mullins, 1983; Reijmer et al., 1988). The pattern also appears in platforms of the Caribbean, the Indian Ocean (Droxler et al., 1990) and the Great Barrier Reef (Davies et al., 1989). See Schlager et al. (in press), for review. Highstand shedding is a general principle of carbonate sedimentation because of the combined effect of sediment production and diagenesis. Sediment production of a platform increases with its size, and the production area of a platform with its top exposed is normally smaller than the production area of a flooded platform. In the Bahamas, the flooded top is one order of magnitude larger than the belt of shallowwater production during lowstands (Droxler et al., 1983; Schlager et al., in press). The effect of increased highstand production is enhanced by the rapid lithification of carbonates during lowstands. Siliciclastics owe part of the high sediment input during lowstands to erosion of the preceding highstand deposits. Carbonate diagenesis largely eliminates this effect. In the marine environment, there is a strong tendency for the sea bed to lithify and develop hardgrounds wherever waves and currents interrupt continuous sedimentation (Schlager et al., in press). Thus, widespread hardgrounds will protect the highstand deposits when sea level starts to fall and wave base is lowered. When sediments finally become exposed, they tend to devel-

Defended platform margins We have mentioned above that carbonate platforms have a strong tendency to form elevated margins that build to sea level right at the shelf break, protecting a slightly deeper lagoon that gently rises to the littoral zone farther landward. When these rims are overwhelmed by sea level, they cause facies belts to jump and interrupt the gradual shift in onlap. This jump must not be confused with a sea-level fall and sequence boundary. Furthermore, elevated rims have a strong tendency to stack vertically, putting reef on reef, sand shoal on sand shoal, because the environmental conditions favor the establishment of a new reef on top of an old reef, an oolite shoal on top of an older shoal. In this way, the lateral migration of systems tracts in response to sea level is interrupted (e.g. windward margin in Figure 5-1). Prograding margins dominated by offshore sediment transport more closely resemble the classical sequence model. They are controlled by loose sediment accumulation and approach the geometry of siliciclastic systems (e.g.

44

Carbonates Versus Siliciclastics 45

op an armor of lithified material within hundreds to a few thousands of years as borne out by the numerous rocky Holocene islands on extant platforms (e.g. Halley & Harris, 1979). Subsequent erosion acts largely through chemical dissolution, enlarging porosity and creating cave networks but keeping surface denudation lower than in siliciclastics. Highstand shedding creates highstand wedges of sediment, displayed in exemplary fashion by the Holocene sediment of extant carbonate platforms (Figure 4-15). Furthermore, turbidites are more frequent in highstand intervals, forming highstand bundles, (Figure 5-6). Platforms are not completely shut down during lowstands. Sediments continue to be produced from a narrow belt of sands and fringing reefs, and from eroding sea cliffs that cut into the exposed platform. There is no evidence, however, that the lowstand input reaches anywhere near the volume of highstand sediments and the postulated apron-building around Pacific atolls during lowstands of sea level has not withstood close scrutiny (see Thiede et al. [1981] vs. Dolan [1989]). Furthermore, lowstand input is compositionally different and can be recognized by petrographic analysis (e.g. Everts, 1991). . The limitations of highstand shedding can be deduced from the above discussion of its causes. The difference between highstand and lowstand prod uction depends on the hypsography of the platform. Flat-topped, rimmed platforms with steep slopes show more pronounced highstand shedding than ramps. If there is no flat top at all and sea-level simply moves up and down an inclined plane, the difference between highstand and lowstand production will be zero. Figure 5-4 shows that besides platform morphology, the duration of sea-level cycles is important. When sea-level cycles are long (e.g. millions of years) and flanks gentle, the platform can build a lowstand wedge that partly substitutes for the production area lost on the platform top (see next section for further discussion). Finally, lithification and the resistance against lowstand erosion vary with latitude and possibly with age. Figure 5-3 illustrates the decrease of marine cementation with latitude. Cool-water carbonates are less prone to submarine lithification than their tropical counterparts. Lithification upon exposure, too, is reduced because of the low content of metastable aragonite and magnesian calcite. It is possible that Paleozoic carbonates with their generally lower content of metastable minerals are also more prone to lowstand flushing. Lowstand shedding of Tertiary cool-water carbonates has been described by Driscoll et al. (1991).

Sea-level records specific to carbonates With their flat tops built to sea level and their biota very sensitive to water depth, carbonate platforms are one of the most reliable dip sticks in the ocean. This quality is enhanced by the resistance to erosion of

exposed platforms. Reefs are "born" as rock-hard structures, other platform deposits usually lithify within a few thousand years when exposed. Subsequent erosion is largely chemical and directed into the rock rather than at its surface (see above). Surface denudation is generally less than in siliciclastics and a reasonable sea-level record can be gleaned from platforms by determining overall subsidence and measuring the thickness of marine intervals plus position and timing of exposure horizons (see Figure 5-5 and Ludwig et al., 1988; McNeill et al., 1988). The combination of defended margins and enhanced resistance to erosion creates some special opportunities for sequence stratigraphy. Rapidly prograding platform margins tend to preserve the original elevations of the shelf surfaces particularly well, including the very important lowstand systems tracts. Eberli & Ginsburg (1988) and Sarg (1988) have contributed excellent examples of sea-level curves gleaned from carbonates using the technique of the fluctuating shelf surface (Figure 5-6). Recently, sea-level studies of platform tops have been complemented by compositional analysis of platform-derived turbidites on the platform flanks. The depositional environment of these calciturbidites is below the range of sea-level fluctuations so that sedimentation is not interrupted during lowstands. Highstand turbidites differ not only in abundance but also in composition from their lowstand counterparts as they contain more ooids, pellets, and grapestones-grains that require flooded bank tops for their formation. This has been well documented for Pleistocene turbidites in the Bahamas (Figure 5-7). Reijmer et al. (in press) report on a Triassic example (Figure 5-8), Everts (1991) on a Tertiary one. Carbonate petrography reveals not only changes in the spectrum of contemporary grains. Lithoclasts derived from erosion of older, lithified parts of the platform are easily recognized. We must emphasize, however, that erosion may cut into older material on the flanks at any time during platform history, regardless of sea-level position (see Figure 1-21). During the present highstand, old limestone is exposed and eroded in many places on the Bahamian slopes and rim collapse caused high admixture of lithoclasts in a debris flow from the Sangamonian highstand (Crevello & Schlager, 1980). Limitations on the compositional tool are the same as for highstand shedding. A platform that builds a lowstand wedge during an extended sea-level cycle may well be able to produce highstand material on this auxiliary platform top.

Drowning and drowning unconformities Siliciclastic systems can build to sea level from any depth, the rate of aggradation being simply a function of sediment input. For reefs and carbonate platforms, on -the other hand, there exists a maximum depth from which they can rebound. This maximum depth of

46

Schlager

rebound is the limit of the photic zone because carbonate production in the aphotic depths of the ocean is negligibly small. Thus, platform growth can be terminated by a relative rise whose rate exceeds the growth potential of the platform in question. A sequence boundary will be formed by the drastic change in sediment composition and in sediment dispersal patterns that accompany platform termination. Geometrically, this sequence boundary resembles a lowstand unconformity. In reality, however, it must form during a rise or highstand of sea level because drowning can occur only when the platform top is flooded (Schlager, 1989). Drowned platforms and drowning unconformities are common in the geologic record (Figures 5-9, 5-10, 5-12). At least some of them have been interpreted as the result of major lowstands and this may explain some discrepancies between the sea-level curve from sequence stratigraphy and curves derived by other techniques (Figure 6-5). Raised rim and empty bucket are common with drowning unconformities and provide a good criterion for their recognition. One problem with the seismic interpretation of carbonate-siliciclastic boundaries is the question of relief on the basinward side. Interfingering and onlap relationships, i.e. pseudo-unconformities and genuine unconformities, are sometimes extremely difficult to distinguish (Figures 5-9 through 5-12). The sediment record of platform drowning is highly variable. Campbell (1992) pointed out that drowning unconformities, too, illustrate the mismatch between outcrop unconformities and seismic unconformities. Only a fraction of the drowning unconformities includes a significant hiatus and thus meets the definition of unconformity in the classical sense.

Many drowning events are represented by a gradational succession of facies that record the increasing water depth on the platform top. A characteristic element of this drowning succession is high-energy grainstones, often rich in echinoderm debris, that appear sandwiched between muddy lagoonal deposits below and deep-water muds above. Drowning succession may include numerous small hiatuses but the overall pattern is one of gradational, albeit rapid, change from platform interior to pelagic or hemipelagic deposition. Erlich et al. (1990) illustrate how recognition of the drowning succession critically depends on seismic resolution (Figure 5-10).

Structural control on reefs and platforms There are numerous examples for structural control of the position of reefs and platform margins. Rather than adding to these examples, we would like to emphasize the opposite quality, the "healing power" of reefs and platforms. Through their vast in-situ production of sediment, platforms are able to quickly fill in structural relief and prograde away from inactive structures, responding to environmental factors such as wind and currents. In this way, structural control soon is lost unless continually renewed by synsedimentary movements (Figure 5-13). After deposition, reefs and platforms often attract deformation along their flanks and over their margins because of preferential compaction of the surrounding sediments. (Figures S-9, 5-10,5-12). This phenomenon, caused by the extensive early lithification of platform rims, is a specific attribute of rimmed platforms.

Carbonates Versus Siliciclastics 47

Fi9ure 5-1. Windward (left) and leeward (right) bank margins in the Bahamas. Note strong evidence for se1smic reefs where margin is stationary and reefs are allowed to stack vertically for extended periods. Apparent lack of reefs on the rapidly migrating leeward margin is probably caused by rapid shift of reef growth and frequent interruption by sediment dumping (compare with the leeward margin of Figure 2-5). Relative elevation of the shelf edge in the prograding margin is probably the best seismo-stratigraphic sealevel indicator in carbonate platforms. The profile also illustrates the effect of sediment supply on sequence geometry. Supply is hi~h on the leeward margin (right) and therefore the margin progrades. The windward margin (left) receives httle sediment and builds up vertically. (Line courtesy of Texaco Inc.; interpretation after Eberli & Ginsburg, 1988)

48

Schlager

a Peri-platform mud Bahamas

Carbonate

30

en

Terrigenous mud

w 30

N Atlantic

A

MARINE PRECIPITATES

(..)

N-56

20

Figure 5-2(a). Sedimentation rates during highstands and lowstands of sea level in various depositional systems. In terrigenous muds from the continental rise, rates are low and uniform during inter~lacial highstands and high and variable (turbtdite-controlled) during glaciallowstands ("lowstand sheddin§"). In peri-platform muds, the pattern is reversed ( 'highstand shedding"). In pelagic carbonate ooze, rates remain low and uniform in both highstands and lowstands. (After Droxler & Schlager, 1985, modified. Reproduced with permission of Elsevier Science Publishers)

z w a: a:

20

10

::> (..) (..)

Glacial

at

0

0

b

...

-

0......_~~40::........._,__....;8::.:0::..........._.~0

-40

-20

0

20

40

60

PALEOLATITUDE

Figure 5-3. Inorganic carbonate precipitation as estimated by the formation of ooids and submarine cements is most intensive in the low latitudes; submarine cementation of cool-water carbonates is only a fraction of the cementation in tropical carbonates. This pattern has persisted thoughout the Phanerozoic. Rapid cementation reduces reworking and sediment export from the platforms during lowstands of sea level and is partly responsible for the pronounced highstand shedding of tropical carbonate platforms. (After Opdyke & Wilkinson, 1990) Reproduced by permission of Elsevier Science Publishers•

50 Sedimentalion rates In mm/ky -

mmlka 40

-60

...... 40

Figure 5-2(b). Highstand and lowstand sedimentation rates during the late Quaternary in the Bahamas (Atlantic) and Maldives (Indian Ocean). In both instances, rates during glacial lowstands (shaded) are generally lower than during the interglacial highstands of sea level. Other effects may modulate this pattern (e.~. upwarddecreasing rates in the Bahamian basm site), but they do not erase it. Bahama store =ODP Site 633, Bahama basin =ODP 632, Maldtves slope =ODP 714. Solid bars indicate means with 90% confidence intervals. (After Reijmer et al. [1988], Droxler et al. [1990), modified.)

Carbonates Versus Siliciclastics 49 a) STEEP SLOPE, SHORT CYCLES (modern Bahamaa)

a)

TIME (M.Y. belore present)

~~~;-.--,......!;'15~~"T"""""""'.---!l!IO--..--.--.,......,--:r5-.--.--,.......,.-;0I.O

100m

Enewetak

1.5

plat'f'orm

40

30-90. upper elope

811o -40

-80 2.5 b) GENTLE SLOPE. LONG CYCLES (Neogene Bahamas)

-120 oxyaen-1sotope

b) 20

15

sea

level

TIME (M.Y. before present) 10 5

0 sequence

low-atand wedge

-...

'.........

200M

100

i

Figure 5-4. Highstand sheddin~ is expected to be most pronounced with steep-stded platforms and short cycles (a). Gentle flanks and long cycles probably lead to growth of a broad lowstand wedge that produces bank-top sediment and may dampen the effect of highstand shedding. Figure 5-5. Sea-level record of carbonate r.latform in the Pacific compared with two other sea- eve) indicators: the eustatic curve of Haq et al. (1987) based on sequence stratigraphy and the oxygen isotope ratio, in this time interval largely a eustatic sea-level signal reflecting changes in ice volume. The platform record consists of the subsidence path of exposed bank top (dotted) and short intervals of marine sedimentabon during highstands (arrows). The overall similarity of all three curves points to a strong eustatic signal in all of them, but correspondence between platform stratigraphy and isotopes is significantly better than with the sequence-stratigraphy curve. (After COSOD Report, 1987, modified)

50

Schlager

sea-level

sea-level Fall

----·

b)

Figure 5-6. Sea-level changes recorded in the fluctuating shelf surface of a prograding Bahama platfonn in the Tertiary. The fact that many platfonns develop hard, erosion-resistant rims makes this classical sequence-stratigraphic technique especially useful in carbonates. (After Eberli & Ginsburg, 1989, Cenozoic progradation of northwestern Great Bahama Bank, a record of lateral platfonn growth and sea-level fluctuations. In Crevello, P.D., Wilson, J.L., Sarg, J.F., & Read, J.R (eds.), Controls on carbonate platforms and basin development, SEPM Special Publication, v. 44, p. 33-351).

Carbonates Versus Siliciclastics 51

12

a

14

13

! :r

...0ti:

... Q:

0

u

8

TURBIDITE~

GRADED PERIPLATFORM OOZE

UNGRADED TURBIDITE

Figure 5-7. Highstand bundling and compositional signals in calciturbidites. Quaternary cores from Tongue of the Ocean, a basin surrounded by the Great Bahama Bank. Turbidites are most abundant during interglacial highstands of the sea, when bank tops are flooded and produce sediment; turbidites are thin and scarce during slaciallowstands. Turbidites also vary in composition: Hi~hstand layers are rich in pellets and ooids-t.e., grains that form on the shallow tianks by the interaction of tidal currents and winnowing from waves. Lowstand turbidites consist mainly of skeletal material, includin~ reef detritus, because fringing reefs and skeletal sand can migrate down-slope with fallin~ sea level. Glacial-mterglacial strati~aphy is provided by variation in aragonite/calcite ratio, a property that ts closely correlated witli the oxygen tsotope curve. (After Haak & Schlager, 1989. Reproduced with permission of Geologische Vereinigung.)

Schlager

52

a Biota, non-specified Platform-interior Shallow reef Deep reef or forereef Open ocean biota Embedding sediment

1. 2. 3. 4. 5. 8.

0.4

.. . . . ' e·.•... . . .. . ·..·.··' .. . .. -... . .. ... .. . ... ....... · . . . . . . . . .... .!3· . .. . . ··. .. . . 0

•• t- , • '.,

.~

• •

I

• .• •.

•

..... : .

•

:·.:•• :•

.: • :

•, • •,

" •

'It

II

•

I

•

• "-

~1

.,

0

~

.·

0

o.o

:

0

.

. . 0

0

·4

..

. .

:

.

.. ..

> 0

""'0

:rJ

.

0

••

.

•

0

•

•

0-2

Oo

,~

.. .... • .I

••

00

:

:

•

..

0

0

....

0

0

1\)

-0.2

•5

.

. .

-0.4

~----~----~----~----~--~~-----T----~----~-----;----~----·~-0-6 }.3 -0 .} o.s 1•I 0.9 0.7 0.3 o.J -0.7 -0.9 -o.s -0-3

FACTOR 1

b 2. BASIN LACKE, GOSAU

1. PLATFORM

TOTES GEBIRGE Wavelength

Ratioa

Wavelength

in m

in m 93.87 18.17 17.07

--------·

14.44 13.74 8.94 8.66 5.03 4.54 4.27 3.91

---------------------------------

24.0 23.68 13.0 11.84 6.5 4.7 ---------·8.61 4.4 7.28 4.0 3.7 3.5 4.98 2.7 4.11 2.3 2.2 2.42 1.3 2.12 1.2 1.1 1.82 1

---------------------------------

Figure 5-8. Compositional signals in calciturbidite-a Triassic example from the northern Alps (after Reijmer et al., 1991; Reijmer, 1991). Point counts and subsequent cluster and factor analysis on 277 turbidites revealed that turbidite composition varies between two extremes: (1) high content of platform material (grains from the platform interior and shallow reefs) and (2) high content of open-ocean biota such as radiolarians and halobiid bivalves. The variations of turbidite composition display a hierarchy of cycles and the inferred periods of the cycles correspond very well with the exposure cycles of the adjacent platform (the Lofer cycles of Fischer, 1964). Reijmer et al. conclude that the variation of turbidite composition is driven by the flooding and exposure of the platform tops. The inferred duration of the most prominent cycles is consistent with orbital control. (a) Factor analysis of the compositional data of 277 turbidites. Factor 1, explaining 56% of total variance, represents an oscillation between platform input (during highstands and platform flooding) and open-ocean input (during lowstands and platform exposure). (After Reijmer et al., in press.) (b) Ratios of the wavelengths of major cycles on the platform and in the basin as determined by Walsh spectral analysis. Wavelengths are expressed as multiples of the shortest cycle, that is, 3.91 m on the platform and 1.82 min the basin. Sedimentation rates ~leaned from biostratigraphy indicate a duration of 15,000....18,000 years for this basic cycle, compatible with its interpretation as the short cycle of the Earth's orbital precession. The similarity of the two spectra supports the notion that turbidite composibon varies m step with the flooding and exposure of the platform source area. (Reijmer et al., written communication.)

Carbonates Versus Siliciclastics 53

a

WILMINGTON PLATFORM

silici-

clastics Cret.-Tert.

plat for•

Late Jur . ... B. Cret.

elevated platro~m margin with drowning unconformity on seaward side

b

Figure 5-9(a). Drowning unconformity on Wilmington Platform accordin9 to Meyer (1989) and Schlager (1989). The seaward flank is over 2500 m high and is onlapped by more gently dtpping, almost transparent siliciclastics. Clastics prograde over platform after drowning has shifted the point of sediment onlap by tens of kilometers shoreward. Erlich et al. (1990) propose a much-reduced relief and interfingering of carbonates and siliciclastics. The debate illustrates a typical problem with drowning unconformities. (Profile courtesy of Shell Oil Co.) Figure 5-9(b). Deep lagoons and raised rims (empty buckets) often precede final drowning of a platform and can be used to distinguish drowning unconformities from lowstand unconformities. In the final stages of growth, the rim of the Wilmington platform rose 200 m above lagoon floor.

54

Schlager

a)

Cii

1000 METERS

1.4

0

z

0 0

w

(/)

w ~

i=

1.6

b)

Platform 3 (Late·growth buildup)

I Platform 2

A€; : : : : : : :.:- ~. -!r;:. .- L-r- J-;--.--JIL-.--=0~,

Figure 5-lO(a). Seismic profile of a "late growth reef" rising above the drowned platfonn top. Miocene, Pearl River Mouth Basin, South China Sea. In this instance the amount of relief at the time of drowning was confinned by drilling. Figure 5-lO(b). Successive stages of drowning and backstepping in Miocene platfonn, Pearl River Mouth Basin. (After Erlich et al., 1990). Successive backstepping of platfonns prior to complete drowning is a very common pattern of carbonate platfonns and commonly produces excellent stratigraphic traps.

Carbonates Versus Siliciclastics 55

S.E.

N.W.

"JOB CREEK" WF19 0

G.A.·3 WF 16

G.A.-6 1mi

G.A.-4

G.M.-1 WF 14

D.J.-1

G.M.-2 ALEXO

1

I

500m

1:1

l:J·

Job Creek reef-edge (Shell fieldwork 1979), incorporating unpublished Shell measured sections.

Figure 5-11. Margin of Devonian Leduc platform, showing back-stepping mar~in onlapped by shale below and carbonate-shale interfingerin~ at the top (Alberta, Canada). Thts example ts based on outcrop observations. In seismic profiles, mterfingering and onlap relationships may be virtually indistinguishable. (After Andrews, 1988; published with permission of Canadian Society of Petroleum Geologists)

011 Morocco

10 km

line cour1esy Euon Co.

Figure 5-12. Unconformity on a Cretaceous-Jurassic platform off Morocco (dotted). This se9-uence boundary was attributed to a Valangiman lowstand by Vail et al. (1977). Schlager (1989) proposed an alternative interpretation as a drowning unconformity. Note the similarity with the coeval Wilmington platform in the nearly conjugate American margin. Note furthermore the raised rim and differential compaction, causing folds in the overlying siliciclastics. Arrows mark the position of boreholes.

56

Schlager

--

/

Ortist-i 0

-----

0

Cortina d'Ampezzo

Platform margins ~

Latest Anisian

~ Early

Ladinian

10km

Figure 5-13. Synsedimentary fault control on platforms is quickly modified and obliterated by progradation and basin-filling when fault movements cease. Triassic, Southern Alps, Italy. (After Bosellini, 1984, modified)

CHAPTER6

ALTERNATIVES TO EUSTASY IN SEQUENCE CONTROL ment are particularly important for reefs and carbonate platforms and will be discussed here. Marine carbonate sediment is extracted directly from the dissolved load of sea water. Thus, carbonate production is intimately tied to the chemistry and other environmental conditions of the ocean. This is in stark contrast to siliciclastics, for which the ocean is merely a transport medium of material ultimately derived from land. The rate of carbonate production is an important control on sequences and systems tracts. Changes in ocean environment can directly modulate the growth and thus the sequence geometry of reefs and carbonate platforms. Environmental change may, but need not, operate in tandem with sea level. Platform drowning is a case in point. Drowning requires that the reef or platform be submerged to subphotic depth by a relative rise that exceeds the growth potential of the carbonate system. The race between sea level and platform growth goes over a short distance, the thickness of the photic zone. Holocene systems indicate that their short-term growth potential is an order of magnitude higher than the rates of long-term subsidence or of third-order sea level cycles (compare Figure 1-15 and Schlager, 1981). This implies that drowning events must be caused by unusually rapid pulses of sea level or by environmental change that reduced the growth potential of platforms. With growth reduced, drowning may occur at normal rates of rise. Examples from the Holocene and the Cretaceous will be called upon to illustrate the role of environmental change. In the Holocene, many reefs drowned at a time when the rate of sea-level rise had already passed its maximum and was slowing down. The drownings occurred when the sea started to invade the flat bank tops, generating shallow lagoons with highly variable temperatures and salinities as well as high suspended loads due to soil erosion. The ebb flow carried these "inimical bank waters" seaward and killed the reefs (Figure 6-1). Neumann & Macintyre (1985) coined the phrase "reefs shot in the back by their own lagoons" for this example of drowning by environmental change. A modern example of a reef crisis induced by environmental change is the coral mortality in the wake of the El-Nifio event of the 1980s (Glynn, 1991). In this

Classical sequence stratigraphy holds that succession and geometry of sequences are controlled by relative changes of sea level (e.g. Vail et al., 1977; Haq et al., 1987). The theory postulates furthermore that a global stack of relative sea-level curves from sequence stratigraphy reveals a strong eustatic signal that dominates the record even in tectonically mobile settings such as convergent ocean margins. Consequent application of the above concepts has produced the most widely cited eustatic sea level curve to date (Haq et al., 1987). This curve has proved very useful in many instances, particularly in periods with independent evidence for strong glacio-eustasy. However, the curve has also generated a growing wave of concern and reports of serious discrepancies. We refer to Miall (1986), Hubbard (1988) and Underhill (1991) for critique of the general concept. This chapter restricts itself to questions specifically related to sequence stratigraphy in reefs and carbonate platforms. We contend that, particularly in carbonates, eustasy is less dominant than postulated by classical sequence stratigraphy and that other processes contribute significantly to the formation of stratigraphic sequences. The alternatives are of two kinds: First, many sea-level fluctuations reflected in sequence stratigraphy are related to local or regional tectonics. Second, there are controls beyond the realm of sea level that can, at times, dominate sequence anatomy. This line of reasoning builds on the arguments presented earlier that the sedimentologic interpretation of sequences and systems tracts leaves room for other controls besides sea level. In the first category, regional changes instead of eustatic ones, we mention intraplate deformation as a particularly attractive alternative to eustasy (Cloetingh, 1988). The past decade produced mounting evidence that the model of rigid, internally undeformed plates is inadequate. It has been shown how plates buckle and warp in response to changes of the intraplate stress field. The resulting sea-level changes are not global but super-regional, i.e., they can imprint synchronous events on entire ocean basins including their margins. For carbonates, this would provide an explanation for the similarity of sea-level curves from mid-ocean settings and ocean margins without necessarily implying eustasy. In the second category, changes in ocean environ-

57

58

Schlager

instance, the damage can be traced to unusually high water temperatures that exceeded the tolerance of many corals. The El-Nino crisis in the eastern Pacific may not have a lasting effect on reefs. However, one can easily imagine that more frequent occurrences of severe El-Nifio events may damage some reefs to extent where they loose the race with subsidence and drown. A Holocene example of coral reef growth limited by ocean circulation has been described by Roberts & Phipps (1988). They noticed that coral reefs in the eastern Java Sea are relatively scarce and restricted to water depths of 15 m or less; below this depth, coral reefs are replaced by bioherms of green algae that are devoid of hard framework. These bioherms produce much sediment but would be unable to build truly wave-resistant structures. The limitation of coral growth is attributed to upwelling of nutrient-rich waters that increase bioerosion and stimulate the growth of competitors to scleractinian corals. The damaging effect of excess nutrients on coral reefs has been documented in numerous involuntary experiments with sewage outfalls and fertilizer runoff. In the Cretaceous, spectacular growth of platforms alternates with extensive drowning events (Arthur & Schlanger (1979); Schlager, 1981). These drownings coincide in space and time with oceanic anoxic events (Figure 6-2). A similar coincidence was observed in the Devonian and the Liassic. In the Cretaceous, the switch from carbonate to siliciclastic deposition produced some spectacular unconformities and sequence boundaries (e.g., Figures 4-4, 5-12). There is, however, no indication that sea-level rises at the time of drowning were unusually rapid. There seems to be no correlation between the rates of (relative) rise and the fate of the platform. Rates of drowned platforms show the same broad scatter as platforms overall and platforms growing at 20 m/Ma were drowned just as well as others growing at 200 m/Ma (Figure 6-3). Furthermore, third-order cycles, the prime cause of sequence boundaries according to classical sequence theory, show rates of rise one order of magnitude below the growth potential of healthy platforms (compare Figures 6-4b and 1-9). We conclude that the majority of Cretaceous drownings were not caused by abnormally high rates of relative sea-level rise but rather by abnormal reduction of growth through environmental change. The coincidence of drowning with anoxia may point to a common cause of the two phenomena or to a cause-and-effect relationship. Hallock & Schlager (1986) have argued for the latter. In analogy to the modem reef killings by sewage outfalls and fertilizer runoff, they speculated that the carbonate benthos was overwhelmed by faster-growing heterotrophic biota when excess nutrients from the anoxic depths of the ocean were swept to the surface in open-ocean overturn. Vogt (1989) proposed poisoning by hydrothermal plumes. The two concepts can be

combined: Volcanism may have provided the energy source for open-ocean upwelling. This upwelling, in tum, could cause fertility pulses with blooms of noncarbonate organisms that damaged the reefs and platforms. The notion of oceanic events terminating platform growth and creating sequence boundaries may explain some peculiarities of the curve by Haq et al. (1987). Figure 6-5 shows the Early-Cretaceous curve to be out of step with other sea-level records. The discrepancies can be reconciled if one assumes that the postulated lowstands really are drowning unconformities. A similar situation is seen on Figure 5-5, where the curve of Haq et al. (1987) postulates two abrupt sea-level falls, at 16.5 and 15.5 Ma respectively. Neither the oxygen isotopes nor the platform stratigraphy of Enewetak confirm these events. Ocean currents are yet another process by which environmental change exerts sequence control. Figure 6-6 summarizes a rather instructive case study from the Blake Plateau. Pinet & Popenoe (1985) mapped a number of well-defined sequences and related them to variations in the course and current intensity of the Gulf Stream. These variations, in tum, were tentatively correlated to sea-level cycles of the curve by Vail et al. (1977). O.D.P. drilling confirmed the strong influence of bottom currents on these sediments but found a very poor correlation with the eustatic curve from sequence stratigraphy (Austin, Schlager et al., 1988). It seems that, at least in this particular case, the meandering Gulf Stream shapes sequences without demonstrable connection to sea-level cycles. Yet another example of a current- controlled sequence boundary is shown in Figure 6-7 from the Gulf of Mexico (Mullins et al., 1987). Finally, changes in sediment input merit mention as important controls of sequence boundaries. These changes are typically linked to tectonic rearrangements of the hinterland and thus pertain to siliciclastics more than to carbonates. Examples include the Cenozoic sequences in the central Gulf of Mexico (Figure 4-6) and the Mesozoic sequences described by Embry (1989) from the Sverdrup Basin. Carbonates may be influenced indirectly by such changes in terrigenous input as certain areas become favorable, others inimical for platform growth. The Devonian of Western Canada reflects this interplay of shifting siliciclastic supply and the carbonate response to it (see Figure 4-7 and Stoakes & Wendte, 1988). We feel that alternatives to eustasy deserve more attention in sequence stratigraphy. The sedimentological underpinnings of the concept clearly indicate that first, relative and absolute (i.e. eustatic) changes of sea level have the same effect on sequences and, second, sequences can be generated by processes other than sea level. The fascination with, and exclusive emphasis on, eustasy may have reduced rather than improved the role of sequence stratigraphy as a predictive tool.

Alternatives To Eustasy In Sequence Control 59

a) .... ,.., ~

lank

1101

--~.··,

~

llloll

.:::·.. ~:!.e.... -

.....

--~

Outer l'oetureDro!IOff - -

Hen

Channel

0

b) .,c.:

. ..

prase11t

!JOOOBP~

II

6000 BP •

"

7000 BP

"

8000

10

LU

1-

w

BP~

~ ro

II

z ~"''

G.. IIJ

0

...,,.

o... ,

l: 1-

'

,.

,,

.,Jll

BRANCHING CORALS

a

MASSIVE CORALS


ACROPORA PRIMATA SAND RADIOCARBON DATE

...

'" HORIZONTAL

... DISTANCE

..

.,. IN

...

METERS

Figure 6-1. Caribbean reefs drowned during the Holocene trans~ession. Rate of sea-level rise cannot be the cause of drowning, because the reefs drowned when sea-level nse was decreasing. The demise is related to reduction of growth potential by environmental stress: the crisis coincides with tlte flooding of the bank tops and ~eneration of lagoon waters with hidt suspended loads and extremely variable temperatures and salinities. (After Adey et al., 1977. Repro'ltuced with permission of Rosenstiel School of Marine & Atmospheric Science) (a) Map of St. Croix, Virgin Islands, showing drowned reef at shelf ed~e, labeled "shelf edge feature." (b) Cross section of drowned reef. Radiocarbon dates indicate drowning coincided with flooding of the shelf.

60

Schlager

140

120

100

SOMe

a)

ORO~ING

BY SEA-LEVEL PULSF ?

Numbers: ampfilude (In meters) at

effeclive rise at subsidence 50 m/Ma

ANOXIA N=39 10

5

b)

Ratel of rise In Cretaceous 3.order cycles

DROWNING N=34

!!!r:

.,

•..

Figure 6-2. Frequency distribution through time of oceanic anoxic events (top) and platform drownings (bottom). The coincidence in time may well point to a causal link. This mar be one example of environmental contro on sequences and sequence boundaries. (After Schlager & Philip, 1990. Reproduced with permission of Kluwer Academic Publishers)

GROWTH RATES OF DROWNED PLATFORMS (CRETACEOUS)

g"'

I

N:46

1"\,

moving average (numerous sources)

\Medlen

~

C"

~

~----~~------~------~------~~

m/Ma-

50

100

150

200

Figure 6-3. Growth rates of drowned Cretaceous platforms show broad scatter and are similar to growth rates of platforms in general; drowning shows no bias towards fast-growing platforms, indicating that the rate of long-term subsidence had little influence on Cretaceous drownings. (After Schla~er, 1991. Reproduced with permission of Elsev1er Science Publishers)

tlo

,,r=J

z&o ...

rate o1 rise-

Figure 6-4. Sea level pulse as cause of platform drowning-an evaluation. (a) Postulated eustatic sea level curve (Haq et al., 1987) in the midCretaceous, a time of global platform drowning. A platform subsiding at 50 m/Ma must have been exposed during the dotted intervals of the curve and experienced only the fluctuations drawn in full. Their amplitudes, shown in meters on the graph, are too small to move a platform out of the photic zone. (b) Rates of third order cycles on the curve of Haq et al. (1987) based on straight-line connections of maxima and minima. The rates are one order of magnitude lower than the short-term growth potential of healthy platforms (Schlager, 1991, and Figure 1-15. Reproduced with permission of Elsevier Science Publishers)

Alternatives To Eustasy In Sequence Control 61

Figure 6-5. Earll Cretaceous sea-level curves from various parts o the world show strong similarity, indicating long-term eustatic control. The curve from sequence stratigraphy is at variance with the conventional record, probably because widespread drowning unconformities in the Valan&inian were misinterpreted as lowstand unconformities (e.g. Figure 5-12). (After Schla~er, 1991. Reproduced with permission of Elsevier Science Publishers)

Sea Level and Valanginian Drowning IEDM.t.~~

ET AL H.t.Q ET AL.

HARRIS

KAUFFMAN

Aroblo

N Amerlco

125

uo Ma

W Atrlco

Tith.

World

se~

a) w

b)

lewetrlslnq

~

28

~.,

....

R•llttve c:N"'"'OI .........

Sri1mic Mq~Aneft

aM

lhll27

N

hl.rus•~

Sitt Ill

......

Figure 6-6. Environmental control of sequences. (a) Cretaceous and Cenozoic stratigraphic sequences on the Blake Plateau. (After Pinel & Popenoe, 1985). (b) Ages of sequences and hiatuses on Blake Plateau after calibration by Ocean Drilling Project Sites 627 and 628. Correlation with tl:ie sea-level curve of Haq et al. (1987) is poor. Sedimentology of ODP holes indicates that erosion and deposition on Blake Plateau are largely determined by the shifting course of the Gulf Stream, i.e. by changes in ocean environment rather than fluctuations of sea level. (After Austin, Schla~er et al., 1988.) Reproduced with permission of Ocean Drilling Program)

62

Schlager n'

...

14'

u'

10'

...

0

400 km

BATHYMETRIC CONTOUR: 200m

Figure 6-7. Environmental control of sequences, Gulf of Mexico. (After Mullins et al., 1987.) (a) Study area on

West Florida Shelf is touched by the "Loop Current," a segment of the Gulf-Stream system. (b) Several sequence boundaries can be mapped in the Cenozoic. (c) Boundary 1/11 does not correspond to a significant

event on the curve of Haq et al. (1987). Rather, it correlates with the closing of the Straits of Panama and the establishment of the Loop Current-a tectonic event that induces environmental change.

Alternatives To Eustasy In Sequence Control 63

PROFILE ~2-44

b

PROGRADING ' CLINOFORMS

MID-MIOCENE UNCONFORMITY u

: w 0

:1

.,"'....

,..

X

;.

!

•

0

1-

I c)

TIME GlOBAl 0

~

~

f~

-=-_.:_;s-t="--..:-

... ......

......

~

\.

~

0

~~ 2

,..

., .... ~

AGEIIII.J.I

0

W.l'lORIOA

~.

\

I

II.

\..

\...

.......

5

'

\ \

SEA-lEV£L HIGH

\

c

...z ... ... u g

ONlAP

n

CHAPTER7

CONCLUSIONS

(1987). Relative sea level enters into this system because it controls accommodation. However, systems tracts are faithful recorders of sea-level only where variations in sediment supply are small compared to variations in accommodation. In all other situations, the distribution of exposure features in the sediment record provides a more reliable record of sea-level positions than systems-tract geometry. In carbonates, the situation is analogous, except that outside sediment supply has to be replaced by G, the in situ growth and production of carbonate material. Sediment geometry and systems tracts are again controlled by two a priori independent rates: A' = the rate of change in accommodation as defined above, and G' = dG/dt, the rate at which a platform produces sediment and builds wave-resistant structures. The maximum rate of growth that the system can sustain, the growth potential, varies across the platform. Therefore, we must distinguish between Gr'= the growth rate of the platform rim, and G '= the growth rate of the platform interior. Bfsed on the relationship between A' and G', five common patterns can be distinguished; they are illustrated in Figure 7-2. The term A' is very closely tied to relative sea-level change. One caveat must be made, though: A' as well as S' represent changes of volume in time, the derivative dV I dt. Sequence stratigraphic models consider either the change in the vertical dimension only, the partial derivative (}z/ at, or the change in two dimensions, ((}z/(}t).(()xj()t). One tacitly assumes these partial derivatives to be good approximations of the change in accommodation volume-a premise that does not always hold. The term S', rate of sediment supply, is only remotely related to sea level and is a major cause of extraneous controls on sequences, independent of sea level. In siliciclastic systems, sediment supply is governed largely by conditions in the hinterland; the growth term G' of carbonates is intimately tied to the ocean environment and to organic evolution. The essence of the above considerations is that depositional sequences represent a composite record of eustasy, tectonically driven, regional change in sea

Accommodation and sediment supply-two basic controls on sequences Application of sedimentologic principles to sequence stratigraphy allows one, among others, to better understand the various factors that cause depositional systems to generate sequences. Recognition of direct and indirect controls is of the utmost importance here. For instance, relative sea level has direct control on the shifting strand line and the exposure and flooding of depositional surfaces; it has indirect control (namely via changes in accommodation) on progradation and retreat of clinoforms. Indirect controls can be very powerful but more often than not they must compete with other factors that have a similar effect. Control on depositional sequences is no exception. Classical sequence stratigraphy postulates that systems tracts are essentially controlled by sea-level change. Sedimentologic analysis leads to the conclusion that direct control of sea level extends only to a small portion of the features in the sequence and systems tract model, such as exposure and flooding phenomena. For most other features, sea-level control is exerted indirectly, primarily by changing the accommodation. And here sea level faces competition from another factor-sediment supply. Progradation and retreat of shelf margins, occurrence of condensed intervals as well as development of turbidite accumulations are influenced at least as much by sediment supply as they are by changes in accommodation. Below we summarize the "double-control concept of systems tracts" first for siliciclastics, then for carbonates. Siliciclastics. We define S as the volume of sediment in the system and A as the accommodation space, i.e., the space available for sedimentation. Then sequences and systems tracts are primarily controlled by two rates, A' and S'. A'=dA/ dt, the rate of change in accommodation, i.e. the sum of the rates of subsidence by crustal cooling, sediment loading, sediment compaction, structural deformation, eustasy etc. S' = the rate of sediment supply. Figure 7-1 summarizes the relationship of A' and S' for the systems tracts and major surfaces of the classical sequence stratigraphic model as described by Vail 64

Conclusions 65

level and environmental change. This last term stands largely for the factors modulating sediment supply. The situation is analogous to that of the oxygen-isotope record of pelagic sediments, known to reflect the combined effects of temperature, salinity, ice volume and biogenic fractionation etc. The succession of sequences, much like the oxygen isotope curve, is a proxy indicator of sea level only when all other effects are minor. However, even where this is not the case, sequence stratigraphy can be a powerful tool for stratigraphic correlation and prediction, offering information on eustatic and regional changes of sea level as well as events in the ocean environment. The record of environmental change again contains global events (such as certain mass drownings of platforms) besides regional ones. To study this composite record, the carbonate sequence stratigrapher should take a three-step approach: 1. Establish unconformity-bounded sequences and correlate them in a region, including platform-tobasin transitions if at all possible. 2. Establish a calendar of regional events, based on the sedimentologic meaning of these sequences and sequence boundaries. 3. Interpret these events in the context of known (!) eustasy, regional tectonics as well as regional and global environmental change.

Figure 7-1. Sedimentologic interpretation of systems tracts in siliciclastics. They are generated by the interplay of the rate of change in accommodation, A', and the rate orchange in sediment supply, S'. Exposure surfaces that extend beyond the shelf break are diagnostic of relative sea-level falls and cannot be generated by variations of supply.

Messages for the seismic stratigrapher Unconformity-bound sequences are well developed in carbonates but eustatic sea level competes with tectonics and environmental change for sequence control. Sequence boundaries are best defined as originally proposed by Vail et al. (1977). Sedimentologically, they represent important changes in the pattern of sediment input and dispersal in a basin. Optimum interpretation of sequences in carbonate platforms requires more ground truth than in siliciclastics. (This situation is similar to interpretation of wireline logs, where carbonates require more core control for optimum results.) Best seismic sea-level records on platforms can be gleaned from the change in deviation of the shelf break in prograding rimmed platforms. Vertically stacked rims yield poor seismic records but provide good sea-level curves via bore holes (from overall subsidence plus position of exposure surfaces). Ideal carbonate ramps obey the same rules as siliciclastics in sea-level response. However the ideal carbonate ramp is rare, and the transition to rimmed platforms very gradual and not easily recognizable in seismic lines. Besides lowstands with subsequent transgressions and wide-spread reefs, drowning of platforms during highstands produces pronounced and well traceable sequence boundaries, often tied to oceanic events rather than pulses of sea level. The drowning unconformity is a sequence boundary formed during a highstand of sea level.

a;, GP >A'

a;, a.;< A'

A'
Figure 7-2. Sedimentologic interpretation of basic ~eometries in carbonates, generated by the mterplay of the rate of change in accommodation, A', and the rate of carbonate growth, G'. (Gr' = growth rate of platform rim, Gp' =growth rate of platform interior).

66 Conclusions

Rimmed carbonate platforms shed most sediment during highstands when the bank tops are flooded. They form highstand wedges of sediment on the basin margins. Associated turbidites may record exposure and flooding of the banks as changes in composition. Seismic reefs differ from stratigraphic reefs and ecologic reefs. Reefs defined by seismic reflection character and geometry will normally include the unstrati-

fied fore-reef debris as well as the back-reef apron. Seismic unconformities are not congruent with outcrop unconformities. The seismic tool commonly misses transitions involving condensed (pelagic) units and shows unconformities instead. More significant are pseudo-unconformities generated by rapid facies change in adjacent beds, for instance at the base of platform slopes.

CHAPTERS

REFERENCES Universiteit), Amsterdam, p. 1-187. Cartwright, R.A., 1985, Provenance and sedimentology of carbonate turbidites from two deep-sea fans, Bahamas: Thesis, University of Miami, 129 p. Cloetingh, S. (1988). Intraplate stresses: A tectonic cause for third-order cycles in apparent sea level, in: Sea-level changes-an integrated approach, SEPM Spec.Publ., 42, p. 19-29. Corso,W., Buffler,R.T. & Austin,J.A. (1988). Erosion of the Southern Florida Escarpment. in: Bally, A.W.(ed), AAPG Atlas of Seismic Stratigraphy, 2, AAPG Studies in Geology, no. 27, p.149-151. COSOD Report (1987). Report of the Second Conference of Scientific Ocean Drilling "COSOD II", Strasbourg, France, p. 1-142 Crevello, P.D. & Schlager,W. (1980). Carbonate debris sheets and turbidites, Exuma Sound, Bahamas. Journal Seimentary Petrology., v. 50, p. 1121-1148. Cross, T.A. & Lessenger, M.A. (1988) Seismic stratigraphy. Annual Reviews Earth Planetary Science, v.l6, p. 319-354. Davies,P.J., Symonds, P.A., Feary, D.A. & Pigram, C.J. (1989) The evolution of the carbonate platforms of northeast Australia. In: Crevello,P.D., Wilson, J.L., Sarg, J.F. & Read, J.F. (eds.) Controls on carbonate platform and basin development, SEPM Spec.Publ.,v. 44, p.233-258 Demicco,R.V. & Spencer,R.J. (1989). MAPS-a BASIC program to model accumulation of platform sediments. Computers and Geosciences, v.lS, p.95-105 Dillon, W.P., Trehu, A.M., Valentine, P.C. & Ball, M.M. (1989). Eroded carbonate platform margin-the Blake escarpment off southeastern United States, in: Bally, A.W.(ed), AAPG Atlas of Seismic Stratigraphy, v. 2, AAPG Studies in Geology, no.27, p.140-148. Dolan, J.F. (1989). Eustatic and tectonic controls on deposition of hybrid siliciclastic/ carbonate basinal cycles: discussion with examples: AAPG Bull.,v. 73, p.1233-1246. Driscoll, N.W., Weissel, J.K., Karner, G.D. & Mountain, G.S. (1991) Stratigraphic response of a carbonate platform to relative sea level changes: Broken Ridge, southeast Indian Ocean: AAPG

Adey,W.H., I.G. Macintyre, R. Stuckenrath, and R.F. Dill, 1977, Relict barrier reef system of St. Croix: Its implications with respect to late Cenozoic coral reef development in the western Atlantic: Proc.Third Int. Coral Reef Symposium, v. 2, p. 15-21. Aigner,T., A. Brandenburg, A. Van Vliet, M. Doyle, D. Lawrence, and Westrich,1991, Stratigraphic modelling of epicontinental basins: Two case studies, in R.H. Dott, and T. Aigner, eds., Processes and patterns in shelf and epeiric basins: Sedim.Geol., v. 69, p. 167-190. Andrews, G.D., 1988, Devonian Leduc outcrop reefedge models and their potential seismic expression, in N.J. Embry, and A.F. Glass, D.J., eds., Devonian of the World, v. ll, Canadian Society of Petroleum Geologists, Mem. 14, p. 427-450. Arthur,M.A., and S.O. Schlanger, 1979, Cretaceous "oceanic anoxic events" as causal factors in development of reef-reservoired oil fields: AAPG Bulletin, v. 63, p. 870-885. Austin, J.A., Jr., W. Schlager, et al., 1988, Leg 101-an overview: Proceedings ODP, Scientific Results, v. 101, p. 455-472. Beach, D.K., and R.N. Ginsburg, 1980, Facies succession, Plio-Pleistocene carbonates, northwestern Great Bahama Bank: AAPG Bulletin, v. 64, p. 1634-1642. Bosellini,A., 1984, Progradation geometries of carbonate platforms: Examples from the Triassic of the Dolomites, northern Italy: Sedimentology, v. 31, p. 1-24. Bosscher, H., and W. Schlager, (in press,a) Accumulation rates of carbonate platforms: Journal of Geology Bosscher,H., and W. Schlager,(in press, b). Computer simulation of reef growth: Sedimentology, Bosscher,H., and J. Southam, (1992) CARBPLAT-a computer model to simulate the development of carbonate platforms: Geology, v. 20, p. 235-238. Brandner, R., E. Fliigel, R. Koch, and L.A. Yose, 1991, The northern margin of the Schlern/Sciliar-Rosengarten/Catinaccio platform: Dolomieu Conference, Guidebook A, (Tourist Office Ortisei) Ortisei, p. 1-61. Campbell, A.E. (1992) Unconformities in seismic records and outcrops: Ph.D. Dissertation, (Vrije 67

68

Schlager

Bull., 75, p.BOB-831 Droxler, A.W., Schlager,W. & Whallon, C.C. (1983). Quaternary aragonite cycles and oxygen-isotope record in Bahamian carbonate ooze: Geology, v.ll, p.235-239 Droxler, A.W. & Schlager,W. (1985). Glacial versus interglacial sedimentation rates and turbidite frequency in the Bahamas: Geology; v.l3, p.799-802 Droxler, A.W., Haddad, G.A., Mucciarone, D.A. & Cullen, J.L. (1990). Pliocene-Pleistocene aragonite cyclic variations in ocean drilling program holes 714A and 716B (the Maldives) compared to hole 633A (the Bahamas): Records of climate-induced CaC03 preservation at intermediate-water depths: Proceedings ODP (Scientific Results), v.115, p.539-577 Dunham, R.J. (1970). Stratigraphic reefs versus ecologic reefs: AAPG Bull., v.54, p.1931-1950 Eberli, G. (1991) Calcareous turbidites and their relationship to sea-level fluctuations and tectonism, in: Einsele, G. et al. (eds.), Cycles and Events in Stratigraphy, (Springer) Berlin, p.340-359 Eberli, G. & Ginsburg,R. (1988). Aggrading and prograding infill of buried Cenozoic seaways, northwestern Great Bahama Bank, in: Bally,A.W.(ed), AAPG Atlas of Seismic Stratigraphy, v. 2, AAPG Studies in Geology, no. 27, p. 97-103 Embry, A.F. (1989) A tectonic origin for third-order depositional sequences in extensional basins-implications for basin modeling, in: Cross, T.A. (ed.), Quantitative dynamic stratigraphy, (Prentice Hall), New York, p.491-501 Epting, M. (1989). Miocene carbonate buildups of central Luconia, offshore Sarawak, in: Bally, . A.W.(ed), AAPG Atlas of Seismic Stratigraphy,v. 3, AAPG Studies in Geology, no.27, p. 168-173. Erlich, R.N., Maher, K.P., Hummel, G.A., Benson, D.G., Kastritis, G.J., Linder, H. D., Hoar, R.S. & Neeley, D.H. (1988). Baltimore Canyon Trough, Mid-Atlantic OCS: Seismic stratigraphy of Shell/ Amoco/Sun wells, in: Bally, A.W. (ed.) Atlas of seismic stratigraphy,v. 2, p.51-65 Erlich, R.N., Barrett, S.F. & Guo, B.J. (1990). Seismic and geologic characteristics of drowning events on carbonate platforms: AAPG Bull., v.74, p. 1523-1537. Everts, A.J. (1991). Interpreting compositional variations of calciturbidites in relation to platform stratigraphy: an example from the Paleogene of SE Spain.- Sedimentary Geology, v. 71, p. 231-242. Fairbanks, R.G. (1989) A 17,000-year glacio-eustatic sea level record: influence of glacial melting rates on the Younger Dryas event and deep-ocean circulation: Nature, v.342, p. 637-642. Fischer, A.G. (1982). Long-term oscillations recorded in stratigraphy, in: Berger,W. & Crowell,J.C.(eds), Climate in earth history, (Nat.Ac.Press), Washin gton, p.97-104 Fischer, A.G. & Bottjer, D.J. (1991)(eds) Orbital forcing and sedimentary sequences. Special issue, Journal

of Sedimentary Petrology, v.61, p.l063-1252 Galloway, W.E. (1989). Genetic stratigraphic sequences in basin analysis 1: Architecture and genesis of flooding-surface bounded depositional units: AAPG Bull., v. 73, p.125-142 Ginsburg, R.N. (1971). Landward movement of carbonate mud: new model for regressive cycles in carbonates: AAPG Bull., v.55, p.340, Abstract. Glaser, K.S., Droxler,A.W. & Baker,P.A. (in press). Highstand shedding off two "semi-drowned" shallow carbonate systems, Pedro Bank and the southern shelf of Jamaica, Northeastern Nicaragua Rise. Journal of Sedimentary Petrology, 25 p. Glynn, P.W. (1991) Coral reef bleaching in the 1980s and possible connections with global warming: Trends in Ecology and Evolution, v. 6, p.175-179 Goldhammer, R.K, Dunn, P.A, & Hardie, L.A. (1990) Depositional cycles, composite sea level changes, cycle stacking patterns, and the hierarchy of stratigraphic forcing: examples from platform carbonates of the Alpine Triassic: Geol. Soc. America Bull., v. 102, p. 535-562. Goldhammer, R.K., Lehmann, P.J., Todd, R.G., Wilson, J.L., Ward, W.C. & Johnson, C.R. (1991) Sequence stratigraphy and cyclostratigraphy of the Mesozoic of the Sierra Madre Oriental, northeast Mexico: Field Guidebook, (Gulf Coast Section SEPM), Austin, p. 1-85. Grammer, G.M., Ginsburg, R.N. & McNeill, D.F. (1990) Morphology and development of modern carbonate foreslopes, Tongue of the Ocean, Bahamas, in: Larue, O.K. & Draper, G. (eds.) Transactions 12. Caribbean Geol. Con£., St.Croix (Miami Geol. Soc.), Miami, p. 27-32. Grammer, G.M. (1991) Formation and evolution of Quaternary carbonate foreslopes, Tongue of the Ocean, Bahamas. Ph.D. dissertation, Univ. of Miami, Coral Gables, Florida, p. 1-314. Grigg, R.W. (1982). Darwin Point: a threshold for atoll formation: Coral Reefs, v.1, p. 29-34. Haak, A.B. & Schlager,W. (1989). Compositional variations in calciturbidites due to sea-level fluctuations, late Quaternary, Bahamas: Geol.Rundschau, v.78, p. 477-486. Halley, R.B. & Harris, P.M. (1979) Freshwater cementation of a 1,000-year old oolite. Journal of Sedimentary.Petrology, v.49, p. 969-988. Hallock, P. & Schlager,W. (1986). Nutrient excess and the demise of coral reefs and carbonate platforms: Palaios, v.l, p. 389-398. Hanneman, D.L. & Wideman, C.J. (1991) Sequence stratigraphy of Cenozoic continental rocks, southwestern Montana. GSA Bulletin, v. 103, p. 1335-1345. Haq, B. (1991) Sequence stratigraphy, sea-level change, and significance for the deep sea. Intl. Assoc. Sediment. Special Publication, v.12, p. 3-39. Haq, B., Hardenbol,J. & Vail,P.R. (1987). Chronology of fluctuating sealevels since the Triassic (250 million years ago to present): Science, v.235, p.

References 69 1156-1167. Hardie, L.A. & Shinn,E.A. (1986). Tidal flats. in: Carbonate depositional environments, modern and ancient. Part 3. Colorado School of Mines Quarterly, v.80, no. 3, 74 pp. Harwood, G.M. & Towers,P.A. (1988). Seismic sedimentologic interpretation of a carbonate slope, North margin of Little Bahama Bank. in: Austin,Jr., J.A., Schlager, W. et al., Proceedings ODP, Scientific Results, v. 101, p. 263-277. Hine, A.C. & Neumann,A.C. (1977). Shallow carbonate-bank-margin growth and structure, Little Bahama Bank, Bahamas: AAPG Bull., v. 61, p. 376-406. Hottinger, L. (1989) Conditions for generating carbonate platforms: Memorie Societa . Geologica Italiana, v.40, p. 265-271. Hubbard,R.J. (1988). Age and significance of sequence boundaries on Jurassic and Early Cretaceous rifted continental margins: AAPG Bull., v. 72, p. 49-72. Jacquin, T., Arnaud-Vanneau, A. & Arnaud, H. (1991) Systems tracts and depositional sequences in a carbonate setting: a study of continuous outcrops from platform to basin at the scale of seismic lines. Marine and Petroleum Geol., v. 8, p. 122-139. James, N.P. (1983). Reef environment. in: Scholle,P.A., Bebout,D.G. & Moore C.H.(eds), AAPG Memoirs,v. 33, p. 346-440. James, N.P. & Bourque, P.A. (in press) Reefs and mounds. Geoscience Canada James, N.P. & Ginsburg,R.N. (1979). The seaward margin of Belize barrier and atoll reefs: I.A.S. Special Publications.,v. 3, (Blackwell), Oxford, 191 pp. James, N.P. & Macintyre,I.G. (1985). Reefs. Carbonate depositional environments, modern and ancient. Part 1. Colorado School of Mines Quarterly, v. 80, no.3, 70 pp. Kauffman, E. (1977) Geological and biological overview: Western Interior Cretaceous Basin. Mountain Geologist, Rocky Mountain Assoc. Geologists, v. 14, p. 75-99. Kauffman, E.G. (1984) Paleobiogeography and evolutionary response dynamic in the Cretaceous Western Interior seaway of North America. Geol. Assoc. Canada Special Publications, v.27, p.273306 Kenter, J.A.M. (1990). Carbonate platform flanks: slope angle and sediment fabric: Sedimentology, v. 37, p. 777-794. Kirkby, M.J. (1987) General models of long-term slope evolution through mass movement, in: Anderson, M.G. & Richards, K.S., Slope Stability (Wiley), Chichester, p. 359-380. Ladd, H.S., Newman,W.A. & Sohl,N.F. (1974). Darwin Guyot, the Pacific's oldest atoll: Proeedings.Second Int. Coral Reef Symposium, 2, Great Barrier Reef Committee, Brisbane, p. 513-522. Lees,A. (1975). Possible influence of salinity and temperature on modern shelf carbonate sedimentation: Marine Geology, v.19, p. 159-198.

Ludwig, K.R., Halley, R.B., Simmons, K.R. & Peterman, Z.E. (1988). Strontium-isotope stratigraphy of Enewetak Atoll: Geology, v.l6, p. 173-177. McNeill,D.F., Ginsburg,R.N., Chang, S-B.R. & Kirschvik, J.L. (1988). Magnetostratigraphic dating of shallow-water carbonates from San Salvador, Bahamas: Geology, v. 16, p. 8-12. Meyer, F.O. (1989). Siliciclastic influence on Mesozoic platform development: Baltimore Canyon Trough, western Atlantic, in: Crevello, P.O., Wilson,J.L., Sarg, J.P. & Read, J.F.(eds), Controls on carbonate platform and basin development. SEPM Special Publications, v. 44, p.211-232 Miall, A.D. (1986). Eustatic sea level changes interpreted from seismic stratigraphy: A critique of the methodology with particular reference to the North Sea Jurassic record: AAPG Bull.,v. 70, p.l31-137 Middleton, M.F. (1987). Seismic stratigraphy of Devonian reef complexes, northern Canning Basin, western Australia: AAPG Bull., v.71, p.1488-1498 Mullins, H.T. et al. (1991) Retreat of carbonate platforms: response to tectonic processes: Geology, v.l9, p. 1089-1092. Mullins, H.T., Gardulski,A.F., Wise,Jr.,S.W. & Applegate,}. (1983). Middle Miocene oceanographic event in the eastern Gulf of Mexico: Implications for seismic stratigraphic succession and Loop Current/Gulf Stream circulation: GSA Bull., v. 98, p. 702-713. Neumann, A.C. & Macintyre,l. (1985). Reef response to sea level rise: Keep-up, catch-up or give-up: Proceedings Fifth Int. Coral Reef Congress, Tahiti, v. 3, p.lOS-110. Opdyke, B.N. & Wilkinson, B.H. (1990) Paleolatitude distribution of Phanerozoic marine ooids and cements: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 78, p. 135-148. Pinet, P.R. & Popenoe,P. (1985). Shallow seismic stratigraphy and post-Albian geologic history of the northern and central Blake Plateau: GSA Bull., v.96, p. 627-638. Purdy, E.G. & Bertram, G.T. (in press) Atoll and carbonate platform development in the Maldives, Indian Ocean: AAPG Bull. Purser, B.H. (ed) (1973). The Persian Gulf. Holocene carbonate sedimentation and diagenesis in a shallow epicontinental sea. Springer-Verlag, New York, 471 pp. Read, J.F. (1985) Carbonate platform facies models: AAPG Bull., v. 69, p. 1-21. Read, F. & Goldhammer, R.K. (1988) Use of Fischer plots to define third-order sea-level curves in Ordovician peritidal cyclic carbonates, Appalachian: Geology, v. 16, p.895-899 Reijmer, J.J.G. (1991) Sea level and sedimentation on the flanks of carbonate platforms: Ph.D. Dissertation, (Free University) Amsterdam, p. 1162.

70

Schlager

Reijmer, J.J.G., Schlager,W. & Droxler,A.W. (1988). Site 632: Pliocene-Pleistocene sedimentation in a Bahamian basin. Proceedings ODP, Scientific Results, v. 101, p. 213-220. Reijmer, J.J.G., Ten Kate,W.G.H.Z., Sprenger,A. & Schlager,W. (1991). Calciturbidite composition related to the exposure and flooding of a carbonate platform (Triassic, Eastern Alps). Sedimentology, v. 38, p. 1049-1074. Rine, J.M. & Ginsburg,R.N. (1985). Depositional facies of a mud shoreface in Suriname, South America-a mud analogue to sandy, shallowmarine deposits: Journal Sedimentary Petrology, v.55, p. 633-652. Roberts, H.H. & Phipps, C.V. (1988) Proposed oceanographic controls on modern Indonesian Reefs: a turn-off/turn-on mechanism in a monsoonal setting. Proceedings 6th Intl. Coral Reef Symposium, 1988, v. 3, p.529-534, Townsville Rudolph, K.W., Schlager,W. & Biddle,K.T. (1989). Seismic models of a carbonate foreslope-to-basin transition, Picco di Vallandro, Dolomite Alps, northern Italy: Geology, 17, p. 453-456. Sandberg, P. (1985). Nonskeletal aragonite and pC02 in the Phanerozoic and Proterozoic, in: Sundquist, E. and W. Broecker The carbon cycle and atmospheric C02: Natural variations Archean to Present, Geophysical Monographs, v. 32, p. 585594. Sarg, J.F. (1988). Middle-Late Permian depositional sequences, Permian Basin, West Texas and New Mexico, in: Bally,A. W.(ed), Atlas of Seismic Stratigraphy, v. 3, AAPG Studies in Geology, no.27, p.140-155 Schlager, W. (1981). The paradox of drowned reefs and carbonate platforms: GSA Bull. v. 92, p.197-211 Schlager, W. (1989). Drowning unconformities on carbonate platforms, in: Crevello,P.D., Wilson, J.L., Sarg,J.F. & Read,J.F.(eds), Controls on carbonate platforms and basin development, SEPM Special Publication, v. 44, p. 15-25. Schlager, W. (1991). Depositional bias and environmental change-important factors in sequence stratigraphy, in: Biddle,K.T. & Schlager,W.(eds.). The record sea-level fluctuations, Sedimenary Geology., v. 70, p. 109130. Schlager, W. & Ginsburg,R.N. (1981). Bahama carbonate platforms-the deep and the past: Marine Geology, v.44, p.1-24 Schlager, W., Buffler, R.T. & Phair, R. (1984). Geologic history of the southeastern Gulf of Mexico. in: Buffler,R.T., Schlager,W. et al., Initial Reports.DSDP, v.77, Washington, p.715-738 Schlager, W. & Camber,O. (1986). Submarine slope angles, drowning unconformities and self-erosion of limestone escarpments: Geology, v.14, p.762-765 Schlager, W. & Philip, J. (1990). Cretaceous carbonate platforms. in: Ginsburg,R.N. & Beaudoin,B. (eds), Cretaceous resources, events and rhythms, NATO ASI Series, C, v. 304, p.173-195

Schlager, W., Reijmer, J,J.G. & Droxler, A.W. (in press) Highstand shedding of rimmed carbonate platforms-an overview. SEPM Special Publication (Major, R. & Halley, R., eds.) Schlanger, S.O. (1981) Shallow-water limestones in oceanic basins as tectonic and paleoceanographic indicators. in: Warme,J.E., Douglas,R.G. & Winterer,E.L.(eds), The deep sea drilling project: a decade of progress, Soc.Econ.Pal.Min., Spec.Publ., 32, p.209-226 Shaub, F.J., Buffler, R.T. & Parsons,J.G. (1984). Seismic stratigraphic framework of deep central Gulf of Mexico basin: AAPG Bull., v. 68, p. 1790-1817. Stafleu, J. and Schlager, W. (1991) Pseudo-toplap in seismic models of the Schlern-Raibl contact: Dolomieu Conference on Carbonate Platforms and Dolomitization, Abstracts, (Tourist Office Ortisei) Ortisei/Italy, p.362 Stoakes, F.A. (1980) Nature and control of shale basin fill and its effect on reef growth and termination: Upper Devonian Duvernay and Ireton Formations of Alberta, Canada: Bull. Canad. Petrol. Geol., v.28, p.345-410 Stoakes, F.A. & Wendte, J.C. (1988) The Woodbend Group. In: Krause, F.:f. & Burrows, O.G., Devonian lithofacies and reservoir styles in Alberta, 13th Core Conference,(Canadian Soc. Petrol. Geol.), Calgary, p.153-170 Thiede, J., Dean,WE. & Claypool,G.E. (1982). Oxygen deficient depositional paleoenvironments in the Mid-Cretaceous tropical and subtropical central Pacific Ocean, in: Schlanger,S.O. & Cita, M.B.(eds), Nature and origin of Cretaceous carbon-rich facies, Acad.Press, London, p.79-100 Underhill, J.R. (1991) Controls on Late Jurassic seismic sequences, Inner Moray Firth, UK North Sea: a critical test of a key segment of Exxons's original global cycle chart: Basin Research, v.3, p.79-98 Vail, P.R. et al., (1977). Seismic stratigraphy and global changes of sea level, in: Payton,C.E.(ed), Seismic stratigraphy-applications to hydrocarbon exploration, AAPG Mem., 26, p.49-212 Vail, P.R. (1987). Seismic stratigraphic interpretation using sequence stratigraphy. Part 1: Seismic stratigraphy interpretation procedure, in: Bally,A.W.(ed), AAPG Atlas of Seismic Stratigraphy, v.1, AAPG Studies in Geology, no.27, p. 1-10. Vail,P .R., Audemard,F., Bowman,S.A., Eisner,P.N. & Perez-Cruz,G. (1991). The stratigraphic signature of tectonics, eustasy, and sedimentation, in: Einsele,G., Rieken W. & Seilacher,A. (eds.), Cycles and events in stratigraphy ( Springer Verlag) Berlin, 22 pp. Van de Poel,H. & Schlager, W. (submitted). MesozoicCenozoic variations in mineralogy of skeletal carbonate: Geologie Mijnbouw. Van Wagoner,J.C., Mitchum, R.M., Campion, K.M. & Rahmanian, V.D. (1990). Siliciclastic sequence stratigraphy in well logs, cores and outcrops:

References 71

concepts for high-resolution correlation of time and facies: AAPG Methods in Exploration Series, v. 7,55 pp. Vogt, P.R. (1989). Volcanogenic upwelling of anoxic/nutrient rich water-a possible factor in carbonate bank/reef demise and benthic faunal extinctions?: GSA Bull., v. 101, p. 1225-1245. Wanless, H.R. & Dravis, J.J. (1989) Carbonate environments and sequences of Caicos platform:

28th Inti. Geol. Congress, Field Trip Guidebook T374, p.l-75 Wilber, R.J., Milliman, J.D. & Halley, R.B. (1990). Accumulation of bank-top sediment on the western slope of Great Bahama Bank: Rapid progradation of a carbonate megabank: Geology, v.l8, p.970-974 Wilson, J.L. (1975). Carbonate facies in geologic history.(Springer Verlag), New York, 471 p.