Carbonates and relative changes in sea level

Marine GeoloJty, 44 (1981) 181--212 181 Elsevier Scientific Publishing Company, Amsterdam -- Printed in The Netherlands CARBONATES AND RELATIVE CHANGES IN SEA LEVEL CHRISTOPHER G. St. C. K E N D A L L and W O L F G A N G SCHLAGER Gulf Research Company, P.O. Drawer2038, Pittsburgh,PA 15130 (U.S.A.) Fisher Island Station, Miami Beach, FL 33139 (U.S.A.) (Accepted for publication March 12, 1981) ABSTRACT Kendall, G.St.C. and Schlager, W., 1981. Carbonates and relative changes in sea level. In: M.B. Cita and W.B.F. Ryan (Editors), Carbonate Platforms of the Passive-Type Continental Margins, Present and Past. Mar. Geol., 44: 181--212. In the geologic record some of the most accurate gauges of changes in sea level are the sediment type, geometry and diagenesis of carbonate shelves and platforms. This is because carbonates frequently occur at or very near sea level and are usually less compacted than siliciclastics.World-wide changes in relativesea level (the sum of eustatic sea-level changes, sedimentation and crustal movements) have occurred repeatedly and cyclicly through geologic time, producing characteristic responses in carbonates. (I) Relative rises in sea level (usually caused by the cumulative effect of tectonic subsidence and eustatic rise)may result in the following: (A) Drowned carbonate reefs or platforms. Here carbonate growth potential is exceeded by relative sea-level rise,and is characterized by shallow-water sediments, overlain by hardgrounds and/or deep-water sediments, some of which may be condensed sequences. (B) Platforms where only the fast growing rim and patches of the interior are able to match sea-level rise while the remainder of the platform is drowned (temporarily). (C) Platforms which keep up and maintain a fiat top at sea level and contain shallowwater sediments whose thickness at the least matches the height of the sea-level rise.If terrigenous supply is limited, prograding sheets of shelf carbonate occur (frequently capped by supratidal evaporites), with prograding shelf-margin carbonate clinoforms and turbidites. If terrigenous supply is high, the shelf carbonates encroach on deltaics. (2) Relative drops in sea level (caused by crustal uplift or by subsidence being outpaced by a eustatic drop in sea level) cause karst and soil development over shelves and platforms, deposition of "deep-water" evaporites in adjacent semi-enclosed basins and in open marine basins the deposition of deltaic and aeolian clastics that bypassed the shelf. Falls are accompanied by platform-wide fresh-water diagenesis. During relativesea-level rises marine diagenesis is c o m m o n in the subtidal portions of the shoaling-upward carbonares, and fresh-water diagenesis and dolomitization and sulphate deposition is c o m m o n in their intertidal and supratidal portions. The stratigraphic significance of these reponses to relativesea-level change is that many are tied to eustatic events and so are predictable within a basin of deposition. INTRODUCTION This paper, largely a review and interpretation of existing data, was p r o m p t e d by recent development in seismic stratigraphy, which is rapidly 0025--3227/81/0000--0000/$02.75 © 1981 Elsevier Scientific Publishing Company 182 becoming a universally accepted tool for mapping sedimentary sequences. The interpretation of seismic records becomes more accurate as one reduces the options by specifying what is geologically likely and what is impossible. In recent years, seismic stratigraphy has benefitted from two developments: the establishment of semi-quantitative, eustatic sea-level charts based on the onlap of depositional sequences onto continental margins and cratonic interiors (Vail et al., 1977; Kauffman, 1977; Hancock and Kauffman, 1978; Hallam, 1978; Harris et al., 1980), and the progress in modelling and predicting subsidence (e.g., Sclater et al., 1971; Watts and Ryan, 1976; Watts and Steckler, 1979). Rate and direction of relative sea-level changes (i.e., the sum of subsidence, eustacy and sedimentation) can now be predicted within fairly narrow limits for most Mesozoic and Cenozoic ocean margins. This in turn poses the question: to what degree axe depositional features and facies patterns predictable from known changes in relative sea level? This report is a step towards tackling this question for carbonate sediments. Carbonates, with their environment-dependent in-situ sediment source, their cemented platform margins (which are relatively wave-resistant) and intensive early diagenesis, are different to siliciclastics, although a number of basic principles of deposition apply equally to both sediment families. In this study, it turned out that in order to build the house, we first had to secure the ground and present an inventory of the types of responses to relative changes of sea level in the geologic record. This constitutes the main part of this report. Precise correlations with actual events, and with the rate and amplitude of relative sea-level change, are still plagued by formidable uncertainties in stratigraphy and in our knowledge of sea-level history, but there are some encouraging examples. Stratigraphic sequences of shallow-water deposits and their facies patterns are primarily controlled by the rates and type of sedimentation, local crustal movements, and eustatic sea level (Fig.l). These three controls act in consort with one another, though, locally, one control may be more important than another. The facies anatomy of carbonate shelves and platforms reflects the combined effect of these three parameters. Often it is difficult to separate the relative importance of these three primary controls whatever the sediments. This is because the same geometry and facies distribution may be the result of an infinite number of combinations of these controls. Suess (1906), supported by Grabau (1936) and others, proposed that eustacy is the major cause for the differentiation of stratigraphic sequences. Stille (1924), on the other hand, favored the combined influence of sea-level changes and tectonism. In contrast, Umbgrove (1939} and Kuenen (1939) related relative sea-level changes to sedimentation, eustacy and tectonism. Pennsylvanian and Permian cyclic deposits from the western U.S.A. are related to crustal movement by Feray (1967), eustacy by Wilson (1967), and depositional controls by Brown (1967), and Robertson Handford and Dutton (1980). Similarly, cyclicity in the Pleistocene carbonates of Barbados (Mesolella et al., 1970) and the Permian carbonates of West Texas and New 183 LOW SEA LEVEL LEVEL EUSTATIC CHANGES ~v OF SEA LEVEL -~ 141 AGAINST TIME RELATIVE CHANGES=L IV ~ ,~ s~A SEA OFAS,EN ATL~,VEEL, ~-FIIiI' % lLANe ' \ I ........ F*tL I POINT OF GREATEST IV AGAINST TIME I 8T'~Lo'STI~AI;:I ........ ) ~ \ ~. POINTOF GREATEST COASTALONLA~ PHASE OF R E L A T I V E SEA LEVEL AND C O A S T A L O N L A P DETERMINED BY R A T E S OF SUBSIDENCE SEDIMENT RESPONSE TO CHANGES IN RELATIVE SEA LEVEL. SHELVES FREQUENTLY UNDERGO TECTONIC DEFORMATION AS THEY FORM BUT DESPITE THEIR DIFFERENCES, TEND TO BUILD TO SEA L E V E L A. NARROW SHELF: WI~ERE RATES OF SEDIMENTATION ARE LOW OR TECTONIC SUBSIDENCE ms H I G H DEVELOPED B. WIDE SHELF : - DEVELOPED WHERE TECTONIC SUBSIDENCE IS LOW OR SEDIMENTATION RATES ARE H I G H .................................................................................................. I. RELATIVE FAST RISE. FOINT OF GREATEST ~COASTAL O,LAP I I POINT OF GREATESTCO*STAL OhLAP 12~ - TRANSGRESSION IN RESPONSE TO RELATIVE SEA LEVEL RISE EXCEEDING SEDIMENTATION. ................................................................................................... . & nT RELATIVE RISE ~pOINT OF GREATESTCOASTALO~LAP POINT OF GREATEST ONLAP I~CO*STAL ~f/-I~ LAND REGRESSION IN RESPONSE TO SEDIMENTATION EXCEEDING RELATIVE SE/~ LEVEL RISE. .................................................................................................. l%T RELATIVE FALL A EUSTATICDROP IN SEA LEVEL PRODUCESWIDESPREAD UNCONFORMITIESBECAUSETHE SEDIMENTSURFACE WHICH WAS BUILT TO SEA LEVEL IS NOW EXPOSED. Fig.1. General interrelationship of relative sea-level changes, tectonism and sedimentation. Mexico (Silver and Todd, 1969) are ascribed to the complex interplay of sea level, crustal movement and reef growth. We concur with Wilson (1975) who concludes that "cyclic" sedimentation and sea-level fluctuation, independent of local tectonics, are both the rule in geologic history. So, as Vail et al. (1977) and Pitman (1978) show, when considering sedimentation, relative sea level cannot be divorced from sedimentation rates, eustacy, or subsidence. As understanding of these phenomena improves it is becoming easier to separate the different controls. Thermal subsidence can be approximated by an empirical equation (Parsons and 184 Sclater, 1977) and can be differentiated from the effects of sea-level fluctuations caused by changes in rates of sea-floor spreading (Hays and Pitman, 1973). The isostatic response of the crust to sea-level changes can be isolated by substracting the effects of thermal cooling and sediment loading. The latter are considered in terms of sediment thicknesses and their ages derived from well data (Van Hinte, 1978; Steckler and Watts, 1978). Watts and Steckler (1979) and Vail and Hardenbol (1979), have gone on to refine this system and use it for determining absolute changes in sea level. As absolute changes in sea level are separated, so it will become easier to identify and separate the effects of local tectonics and sedimentation. Objectives Our paper attempts to set forth the general ways that shelf and platform carbonates respond to changes in relative sea level. Armed with these basic concepts, we can explain and sometimes predict carbonate facies relationships. As the timing and magnitude of eustatic changes become better defined so will these guidelines. The paper is based on theoretical considerations, as well as examples from the geologic record for which sea level, crustal movements, and sedimentation rates are usually well constrained. These examples range from those of the Quaternary, where responses are rapid and sea level is known from detailed charts based upon independent evidence, to those from the much older geologic record, where sea-level movements are inferred from a variety of sources, including the sea-level onlap charts proposed by Vail et al. (1977). To illustrate how relative sea level controls carbonate sedimentation, we will briefly summarize some concepts related to sea-level change and then classify and illustrate the most common responses of carbonate sedimentation to relative sea-level changes. CONCEPTS ASSOCIATED WITH FLUCTUATIONS IN RELATIVE SEA LEVEL Causes of sea-level change Relative sea-level changes may have a number of causes, many of which have been reviewed by Pitman (1978} and by Donovan and Jones (1979). They may occur where the depositional setting subsides or is uplifted. These effects may be related to faulting, thermal regime, salt diapirism or isostacy. Local differences occur where the underlying sediment compacts at different rates. Other relative changes can be caused when basins become isolated either by the development of sedimentary or tectonic sills or by eustatic drops in sea level. If this isolation occurs at low latitudes, evaporative draw-down may produce a corresponding ctropin sea level, as in the Mediterranean. Eustatic sea-level changes have two commonly identified causes, Either they are glacially induced or there is a change in the shape and so volume of 185 the ocean basins due to tectonics. Other less common causes include major meteorite impacts (Lemcke, 1975; Smit and Hertogen, 1980), submarine volcanic outpourings (Schlanger and Premoli-Silva, 1981), and desiccation or sudden flooding of enclosed ocean basins such as the South Atlantic in the Cretaceous. Orders of sea-level cycles Vail et al.'s (1977) relative sea-level charts established for the Phanerozoic, are a superposition of several orders of cycles. Unfortunately, the relative importance of the various orders of sea-level cycles to carbonate sedimentation appears to be inversely proportional to the certainty with which they are known. Timing and amplitude of first-order and second-order cycles are well documented by the onlap of stratigraphic sequences on the cratons (Sloss, 1963), by datable unconformities and the relative lateral position of one sedimentary sequence to another (Vail et al., 1977 vs. Hays and Pitman, 1973). First-order cycles may last over hundred million years, which is much longer than the average life span of a carbonate platform. Vail et al.'s (1977) cycles of second and particularly of third order, caused by rates of eustatic sea-level change of tens or hundreds of Bubnoffs, can successfully compete with tectonic subsidence, which commonly ranges from 10 to 100 Bubnoffs (1 Bubnoff = 1 mm/1000 years, Fischer, 1969) and have had a strong influence on carbonate deposition. However, there is still uncertainty as to the exact amplitude of the lower-order cycles and their timing (see Vail and Hardenbol, 1979). Furthermore, we believe that there are fluctuations of still higher frequency than even the third-order cycles of Vail et al. (1977). Some of the most prominent features in carbonate platform stratigraphy, such as drowning events (Schlager, 1980) or high-frequency cyclic deposition (e.g., Fischer, 1964; Wilson, 1975, p.281), may be related to sea-level fluctuations in the 10,000--100,000-yr. domain. These cycles of fourth and fifth order are beyond the resolving power of the state-o¢.the-art seismic stratigraphy as used by Vail et al. in 1977. However, if the system of combining well data and seismic cross-sections outlined by Vail and Hardenbol (1979) includes detailed measured sections, detailed well descriptions and highresolution seismic cross-sections, in some cases these cycles may be resolved. For the present, we can best recognize these higher-frequency cycles in outcrop, from not only the Neogene and Quaternary but older rocks. We should also realize that these high-frequency fluctuations have a most profound impact on shallow.water carbonate deposition. Indeed, we feel that the strongest cases for sea-level control on neritic carbonate sedimentation are associated with short cycles or singular events. In Pennsylvanian and Permian, the cycles are probably of glacio-eustatic origin (see Van Siclen, 1958; Peterson and Ohlen, 1963; Silver and Todd, 1969; Meissner, 1972). In contrast, Huh et al. (1977) and Esteban and Giner {1977} ascribe cycles to a short-term draw-down of sea level in evaporitic basins. 186 Response of carbonate sedimentation to changes in sea level As sea-level charts are accepted by geologists so their approach to stratigraphy has changed. This is because they can use this understanding of sealevel variation predictively and couple it to rates of subsidence and rates of carbonate and clastic sedimentation, and so construct conceptual models for different geologic settings through geologic time (Figs.2--8). The construction of these predictive models is based on the realization that different rates of sea-level rise or fall will affect both carbonates and clastics in different ways and create specific geometries. By looking at Vail et al.'s (1977) charts, particularly the revised curve for the Tertiary (Vail and Hardenbol, 1979} and Kauffman's (1977) charts for the Cretaceous, it is possible to determine when one might expect rapid rises, stillstands, slow rises or falls and to relate these to the geologic record to see what the general response of both carbonate and clastic sediments has been. In some cases the pattern of sediments match the charts and in others, the charts will need to be modified, particularly where a high frequency of sedimentary cycles is common. In the discussion which follows we consider the effects of changes in relative sea level on carbonates. Under favorable conditions, carbonate platforms and, particularly, reefs can be shown to grow at rates of 103--104 Bubnoffs (Macintyre and Glynn, 1976; Adey, 1978; Lighty et al., 1978; Schlager, 1979) (Table I) and thus match even the fastest known rises of sea level such as the glacio-eustatic pulse during the Holocene (at rates of 103-104 Bubnoffs, Adey, 1978) (Table I); healthy platforms easily outpace long-term basin subsidence, generally on the order of 10--100 Bubnoffs (Fischer, 1975; Schwab, 1976) (Table I). However, adverse conditions can reduce the growth potential of carbonate platforms and reefs by several orders of magnitude. Consequently, the known ISOLATED AND ATTACHED CARBONATE PLATFORM DEPOSITIONAL SETTINGS. FtSOLATEDCARBONATEI , I , LATFO.MS f ' @ ' SLOW SUBSIDENCE RAPID SUBSIDENCE Fig,2. General carbonate responses to relative sea-level change. 187 T. PLATFORMS DROWNED DUE TO SLOW CARBONATE ACCUMULATION, POSSIBLY TERMINATED BY CLIMATE, CLAY LADEN WATER OR TOO GREAT A WATER DEPTH. DROWNED REEFS ON PLATFORM ] PELA DROWNED OFF PLATFORM REEFS INITIATED BY BIOLOGICAL ACCRETION OR ON OLD HIGHS I HARDGROUNDS FORMED DUE TO LOW RATES OF SEDIMENT ACCUMULATION 'rr. TERMINATED BY CLASTIC POLLUTION. Fig.3. Response of isolated carbonate platform to a relative sea-levelrise which exceeds carbonate growth potential and causes drowning. responses of reefs and carbonate platforms to sea-level rises cover a spectrum from complete failure to keep pace with the rising sea to situations where the rise is far exceeded by carbonate growth. Response of the carbonate system is not simply a function of the rate of relative sea-level rise but rather the difference between the rate of rise and the growth potential of a platform. To predict the response of a platform we need to know and quantify both the rate of relative rise and the platform growth potential, including its variation during the rise. Data on these variables are just beginning to appear {e.g., on rates of rise: Hancock and Kauffman, 1979; on growth potential of platforms: Schlager, 1980). Except for the Holocene, these data are insufficient to warrant a quantitative treatment of the subject. 188 "r. RATE OF RELATIVE SEA LEVEL RISE APPROXIMATELY MATCHES GROWTH POTENTIAL OF RIM. I ~,~o.~?~,;,; ~°V~.,~ ~'L ~Z;~.',~'.~;TED~ i / 7 / ~ /. T- ///~,\~\ // // /7 \ I ////// / -- I DEE~ CAR,ONATEA=O"°~AT,ON I [SLOWSDURINGRAPIDRISE& GENERALI PLATFORMSEDIMENTATION i \\ I,~;,°',N.T= J DEVELOPMENT OF HARD GROUNDS AND CONDENSED SEQUENCE 11' A.GROWTH POTENTIAL OF INTERIOR BEGINS TO EXCEED RELATIVE SEA LEVEL RISE. PINNACLES BEGIN TO CATCH UP W,TH 1 BUILDUPS CAICH UP wWITH| ~,BU/LDUPSC.A~UCALLO [FAUNA CAP. J ] I B. GROWTH POTENTIAL EXCEEDS RELATIVE SEA LEVEL RISE AND C A R B O N A T E S COVER WHOLE SHELF ON REACHING SEA LEVEL. Fig.4. Response of isolated carbonate platform to a relative sea-level rise which exceeds the carbonate platform growth potential except at the rim and over some patches. 189 BUILDUPS INITIATED ON HARDGROUND SURFACES DURING OR FOLLOWING RAP D RISE. ~ ~ : , I "~, ~i,,~ ~i ~M>~.-~ ~ ~--_~ LZ,-~, ;",~.-~j~-----~ PE'AG,C S.ALES ENVELOPING"I ~ ~ . ~ _ - - - / ' j ~ PREVIOUS I / / k : ~ J ~ ~ - ~ ' I BUILDUPS I // - ~ [AND MARGIN.J / / / / ~ / I// / l/" / ~ U A L L FORMED DURING RAPID RISE/ J : ~ ( ~ ~ ~ ~:~--E::~---T~ ~ T ~ , - - - r - L - r / ~ T" " ~ " ~- --:" / / / / [~ I N C E INITIALLY CARBONATE I ~ CANNOT KEEP PACE WITH SEA Y MARGIN PROGRADES I WHEN CARBONATE PRODUCTION EXCEEDS I SEA LEVEL RISE. Fig.5. Response of carbonate margins to a relativesea-levelrisewhich at firstexceeds the margins carbonate growth potential and then at leastmatches the growth potential. However, within the spectrum of responses of carbonate platforms to sea-level rise, one can identify a few basic types of response and define them in terms of inequalities. It is this approach that will be followed here. In order to define these types of response we have to introduce and explain first some new or unfamiliar terms. Bucket principle. We believe that the basic growth anatomy of a carbonate platform is that of a bucket, held together by stiff rims of "competent" material and filled with "incompetent" lagoonal or tidal deposits (MacNeil, 1954; Klovan, 1974; Wilson, 1975, Fig.3). This pattern is the result of the selective early cementation of the carbonates, which accumulate along the platform margin, while the carbonates which accumulate behind this margin remain uncemented (Land and Goreau, 1970; James et al., 1976; James and Ginsburg, 1980; Longman, 1980; Shinn et al., 1981). The sediments of the rim range from carbonate shoal sands and muds to organic frame reefs. Interior and rim may have different growth potentials. The overall growth potential of a platform is determined by the growth potential of its rim. During rapid relative rises of sea level, the rim may keep pace while the platform interior lags behind. Platforms with deep lagoons result. As the rim rises above the platform floor, the platform interior becomes a sediment trap which is quickly filled, should the rate of relative sea-level rise decrease. As a consequence, the growth potential (or accumulation potential} of the platform interior commonly increases during a rise of sea level. 190 SUBTIDAL MARGIN, MAJOR SOURCE OF CARBONATES 1 FOR PROGRADING FLANK. 1 .~.~ ~ . , ~.. IFILLS TO SEA LEVEL. SHOALING CYCLES CAUSED BY FLUCTUATIONS IN SEA LEVEL CHANGE & RATES OF SEDIMENTATION. ~IAGRAM] NEXT ~~ EACH MAJOR RISE IN SEA LEVEL MAY BE REPRESENTED BY ONE OR SEVERAL UPWARD SHOALING SEQUENCES. (1) (2) _ - v ~ A v ~ ---_ - --~-- ~ A ------ A = KARSTIC SURFACE --~.-~--~-_~_ B : ALGAL STROMATOLITE ~I~,~Is ~ ,~/~, ~1~~ "--"--'--:'::-B------~" ° = OvE"DEE" ,AGOO.A.rRA,SG,ESS.,EOARBO.ATE ~'I ~ C ( , ( c D LIME SANDS t ~ ~ -.---_-__--_DIFFERENCE BETWEEN 1 8. 2 CAUSED BY "DAMPING" OF CARBONATE SEDIMENTATION RATES. Fig.6. Response of isolated platform whose carbonate growth potential at least matches and may exceed relative sea-level rise. Carbonate bodies not built by the bucket principle are the ramps of Ahr (1973). They develop gently sloping flanks which probably reflect the gradual shoreward decrease of wave action, much like a siliclastic shelf. Basin starvation and a relative sea-level rise may cause ramps to evolve into steep carbonate cemented rimmed shelves (Wilson, 1975). 191 G R O W T H P O T E N T I A L M A T C H E S OR E X C E E D S RELATIVE SEA L E V E L RISE. IFOLLOWING RISE. CARBONATES QUICKLY FILL TO / LEVEL RrSES. PRODUCE SHOALING CYCLES. rCA%OBTOALMA"O'N MA"OR SOORCE OE1 /f BONATES ON PRQGRADINGFLANK ~ /~ THE GREATER MARGIN STARVATION MAY STEEPEN rN BASIN NOW ~DUE TO WATER DEEPENSOFF THE MARGN i ~ ~,SDLAFED OEPRESS,ON F,LLEO SY SEA I I EVAPORATESFASTER THAN IT FILLS. I SUPRATIDAL EVAPORITESAND OCCASIONAL CLASTICS THIN BEDDED ON DOLOMITES. ATTACHED SHELVES AEOLIAN CARBONATE SHELF - CLASTIC ENCROACHMENT AND PROGRADATION 1. RELATIVE SEA LEVEL RISE rCARBONATE ENCROACHES ON ] /DELTAICS WHICH RETREAT AS / |SEA LEVEL RISES. / I CELTAIC MIGRATION I )INDEPENDENT OF SEA LEVEL I / C A U S E S INTERDIGITATION I CLASTICS I 2. RELATIVE SEA LEVEL FALL [oDEL~)A/CSRENCN~OEACHFROM LANDWARD] AT A SEA LEVELLOW, COURSECLASTICS MAY BYPASS A NARROWSHELFAND Fig.7. Response of isolated and attached carbonate margins whose growth potential at least matches and may exceed relativesea-levelrise. 192 A, BASIN OPEN TO SEA. I CLASlIC BYPASSING SHELF B, BASIN BARRED SO ONLY NET INFLOW OF S E A I THROUGH LOCAL CANYONS , /' ,' ." ' ,,' / [AEO~i;,~ GL; sTics] ON ATTACHED // I PLATFOAMS i AEOLIAN CLASTICS I ON ATTACHED PL/&TFORMS ~ oF EVAPOR,TE RROOOC ,O. SKS O A T,G..YPA .,NG SHELF, Fig.8. Response of carbonate provinces to a relative sea-level drop. Carbonate reefs. Reefs formed along basin margins are relatively common in the geological column. The two most common reefs are: (1) those formed in the basin, and on and at the base of the foreslopes of basin margins, and (2) those formed in the shoal water of platforms at or back from the basin margin (Table II, A, B). Both forms of reefs nucleate over topographic highs. The origins of the highs are various. Most are accretional and related to local opportunism of the fauna. Less commonly, the highs are tectonic, as they are for the Paleocene Dahra reefs in the Sirte Basin of Libya (Terry and Williams, 1969). They may also be karstic remnants developed where a sea-level fall has exposed a previously shallow-water shelf to subaerial processes (Purdy, 1974). These karst features may be drowned by the next relatively rapid sea-level rise. Though the now deeper shelf may be unable to sustain regional carbonate sedimentation, the isolated karst topographic highs are able to TABLEI Rates of sea-level movement, tectonic movement and carbonate sedimentation in Bubnoffs (1 Bubnoff = 1 mm per 1000 years; Fischer, 1969) A. Sea-level movement (1) 1st-order cycles of Vail et al. (1977) and Vail and Hardenbol (1979): a few Bubnoffs (2) 2nd- and 3rd-order cycles of Vail et al. (1977) and Vail and Hardenbol (1979): 10--100 Bubnoffs (3) Holocene rise (glacio-eustatic): 103--104 Bubnoffs (Adey, 1978) (4) Alpine Triassic falls: in excess of 100 Bubnoffs (Fischer, 1964) B. Tectonic movement (1) Average tectonic subsidence: 10---100 Bubnoffs (Fischer, 1975; Schwab, 1976) (2) Subsidence at Mid-Ocean Ridge over flint 2 m.y.: 250 Bubnoffs (Sclater et al., 1971 ) C. Carbonate sedimentation (1) Holocene reef/phtform growth: 10~--10 ' Bubnofs (Adey, 1978) 193 T A B L E II Examples of carbonate responses to changes in relativesea level A. Foreslope and basin reef buildups G. M o d e m platforms with survivalof rim and patches 1. R e c e n t l i t h o h e r m s of the Straits of Florida 1. The Queensland Shelf Great Barrier Reef and shelf atolls(Maxwell, 1968). ( N e u m a n n et al., 1977). 2. The Jurassic sponge reefs of M o r o c c o (Evans 2. A m e m b e r of Pacific atollswith deep lagoons (E.G. Emery et al.,1954). a n d Kendall, 1977). 3. The C a r b o n i f e r o u s Waulsortian M o u n d s (Lees, 3. The southern part of the Belize Shelf (Wanfland et al.,1975). 1964). 4. The Devonian Keg River Reefs of A l b e r t a 4. The islandshelf of St. Croix, Virgin Islands (Adey and Burke, 1976; Adey et ai.,1977). ( L a n g s t o n a n d Chin, 1967). 5. The Silurian Michigan Basin Reefs (Mesolelia 5. Parts of the western Yucatan Shelf (Logan et al., 1969), where the drowning is almost complete et ah, 1974). and the rim is only marked by widely spaced, 6. The Middle Ordovician offshelf b u i l d u p s of isolated coral reefs. Virginia (Read, 1978). B. Buildups o n shallow w a t e r p l a t f o r m s H. Ancient reefs which show an initiallag followed by catch up and keep up 1. The R e c e n t reefs of Bikini a n d E n i w e t o k in the Pacific (Tracy a n d L a d d , 1974). 2. The Miocene of Indonesia (Vincelette a n d S oeparjadi, 1976). 3. The Paleocene of L i b y a (Terry a n d Williams, 1969). 4. The Devonian S w a n Hills a n d L e d u c of weste m C a n a d a (Stoakes a n d Wendte, 1980). 5. The Devonian of the C a n n i n g Basin of weste m Australia (Play ford, 1980). 6. The Middle Ordvvician of Virginia (Read, 1978). 1. Silurianreefs of the Michigan Basin (Mesolella et al., 1974). 2. The Paleocene of L i b y a (Terry and Williams, 1969). 3. The Jurassic M o r o c c a n Pinnacle Reefs of the rich area of the middle high Atlas (Evans a n d Kendall, 1973). 4. The Devonian reefs of the M o r o c c a n a n d Spanish S a h a r a (Dumestre a n d Illing, 1967). 5. The Winnepegosis Keg River Reefs f r o m the Devonian of C a n a d a ( L a n g s t o n a n d Chin, 1967). 6. The Devonian of the C a n n i n g Basin of western Australia (Piayford, 1980). 7. The Ordovician m o u n d s of Virginia (Read, 1978). C. Facies selectivehardgrounds I. 1. The R e c e n t of the Persian G u l f (Shinn, 1969). 2. The R e c e n t of the B a h a m i a n P l a t f o r m s ( T a f t et al.,1968; Dravis, 1 9 7 7 ; a n d Harris, 1979). 3. The R e c e n t flanks of the B a h a m i a n Platforms (Mullins a n d N e u m a n n , 1 9 7 9 ) . 4. The E u r o p e a n C r e t a c e o u s chalks b y B r o m l e y (1968). 1. The Permian of the G u a d a l u p e Mrs. (King, 1 9 4 8 ; Newell et al., 1953). 2. Triassic of the n o r t h e r n a n d s o u t h e r n Alps (Fischer, 1 9 6 4 ; O t t , 1 9 6 7 ; Zanki, 1 9 6 7 ; Bosellini a n d Rossi, 1974). 3. Devonian of t h e C a n n i n g Basin ( P l a y f o r d a n d L o w r y , 1 9 6 6 ; Piayfo~l, 1980). 4. The C r e t a c e o u s of Mexico a n d the G u l f Coast (Coogan et al., 1 9 7 2 ; Wilson, 1 9 7 4 ; P.325). D. Widespread hardgrounds J. H o l o c e n e reefs w h i c h m a t c h e d sea level rises 1. The Devonian of western Canada (Stoakes, in press;and Wendte, in press). 2. The Callovian of the Paris Basin where they can be shown to become younger continentw a r d (Purser, 1969). 3. O t h e r levels in Jurassic of western E u r o p e (H aliam, 1969). 1. The Galeta Fringing R e e f in the C a r i b b e a n (Macintyre a n d G l y n n , 1976). 2. The Florida Fringing Reef (Lighty et al., 1978). E. D r o w n e d reefs a n d p l a t f o r m s K. A n c i e n t buildups w h i c h m a t c h sea level rise 1. The Devonian of central a n d w e s t e r n E u r o p e (Tucker, 1 9 7 4 ; Krebs, 1974). 2. The Jurassic o f the T e t h y a n R e a l m (Bosellini, 1 9 7 3 ; Bernoulli a n d J e n k y n s , 1974). 3. The Cretaceous o f the Pacific a n d A t l a n t i c 1. The Devonian o f C a n a d a (Stoakes a n d Wendte, 1980). 2. S o m e of the Pleistocene A c r o p o r a - P a l m a t a Reefs o f B a r b a d o s (Mesolleia et al., 1970). Fiat t o p p e d shallow-water c a r b o n a t e p l a t f o r m s 194 TABLE II (continued) E. Drowned reefs and platforms (continued) (Matthews et al., 1974~ Arthur and Schlanger, 1979; Schlanger, in press). 4. The Devonian of the Canning Basin of western Australia to Table 7 (Playford, 1980). 5. Carbonate ramp-to-basin transitions and foreland basin evolution, Middle Ordovician, Virginia Appalachians, (Read, 1980). F. Reefs with fibrous carbonate cement cavity fill L. Basinal evaporltes and shallow water carbonate margins 1. The Mississippian stromatactis reefs of England (Bathtu~t, 1959). 2. The Leduc and Golden Spike Devonian reefs of western Canada (Walls, 1979). 3. Some of the Devonian buildups of Central Europe (Krebs, 1974). 4. Some of the Triassic buildups of the Calcareous Alps (Zankl, 1968). 5. The Dolomites of N. Italy (Bosellini and Rossi, 1. 2. 3, 4. The The The The Paradox Basin (Peterson and Owler, 1963). Sverdrup Basin (Davies, 19'/7). Zechstein Basin (Richter-BelTburg, 1955). Michigan Basin Silurian (Mesolella et al., 1974;Huh et al., 1977). 5. The Tertiary--Messinian evaporites of the Mediterranlan Sea (Hsii, 19'/2). 6. The Miocene of the Red Sea (Heybroek, 1965). 1974). 6. Devonian reef of the Canning Basin, western Australia (Play ford, 1980). 7. Middle Ordovician buildups of Virginia (Read, 1978). promote carbonate sedimentation. Sometimes the shallow-water shelf is drowned by a rapid relative sea-level rise before subaerial exposure can take place. In this case, carbonate sedimentation may again be terminated across most of the previous shallow-water carbonate shelf but topographic highs formed by subaerial islands become the nucleus of the overlying reefs. However, most of the nuclei of ancient build.ups observed by the authors and others are accretional in origin, like the cores of the Devonian build-ups from western Canada (Wendte, 1981). Cementation fabrics within carbonate build-ups are often marine and are, in some cases, dominated by fibrous cements {Table II, F). Marine cementation in carbonate build-ups appears to depend on the steepness and facies of the build-up, which in turn are controlled by rates of change of relative sea level and more particularly sediment accumulation rates. Build-ups with lowangle margins, like the Devonian Swan Hills reef of Alberta have little marine cement, but most frequently contain late burial cements (Wong, 1979). In contrast, build-ups with high-angle margins, particularly in settings exposed to the pumping action of breaking waves, have dominantly marine cements in marginal positions, but little marine cementation in the interior (Playford, 1980). This latter is commonly filled by late-stage burial cements {Walls et al., 1979). The geometric control on the marine cementation of carbonate margins can be seen in the Permian Capitan Limestone of the Guadelupd Mountains. Both Babcock (1977) and Yurewicz (1977) show that marine cementation becomes more prevalent in the younger Capitan, as it progTessively steepens. 195 Drowning. Reefs and carbonate platforms are drowned when the rate of relative rise of sea level exceeds the vertical accumulation rate and the reef or platform becomes submerged below the euphoric zone and terminating prolific carbonate production by photosynthetic organisms. Reefs and platforms suffocated by influx of clastics are excluded from this category. Incipient drowning. Incipient drowning results from deepening within the euphotic zone, and occurs where a relative rise of sea level exceeds the growth potential of the carbonate ecosystem, but before the process goes to completion, the system is able to recover because the rate of sea-level rise decreases relative to the carbonate growth potential. Start-up phase. The Holocene record, as well as numerous examples from the ancient, contain evidence that it takes some time to start-up the "carbonate factory" after periods of exposure. A rapid rise of the sea may thus "take a reef or platform by surprise" and initiate a start-up period during which the system runs below its full growth potential, lagging behind the rise of the sea. In the Holocene, Adey (1978) estimates this lag-period to be on the order of 500--1000 years, but much longer periods of subdued growth seem to be indicated by the geologic record. In ancient platforms, analogues to these modern lag deposits can be found to include extensive hard grounds, condensed sequences and pelagic facies. For instance, carbonate hard grounds associated with carbonate platforms are common in the geologic record. They occur in two forms. One is faciesselective and has a patchy distribution through sequences (Table II, C). The other form is more widespread and cuts across facies. We believe that these latter widespread hard grounds (Table II, D) probably trace sea-level rises during which carbonate sedimentation terminates and cementation, grazing, and boring of the old carbonate sediment/water interface result. In all cases (Table II, D) these latter hard grounds seem to coincide with periods of rapid relative sea-level rise on the charts of Vail et al. (1977). Similarly, condensed sequences commonly occur on sea mounts, on terraces of platform slopes, and on shelf edges. As with hard grounds, we speculate that the drowning of platforms, which precedes deposition of these condensed sequences, is caused by very rapid rises of relative sea level with respect to carbonate gro~, ~h potential. Once a platform or reef is drowned, slow and intermittent pelagic deposition may prevail for considerable time. This is because the former platform represents a topographic high that is often swept by currents and is remote from terrigenous influx. The firstphase cements of these carbonates are commonly marine (Purser, 1969; Lighty, 1977a; Stoakes, 1981; Wendte, 1981). The existence of an initial lag period then implies that the growth potential of a carbonate platform or reef is often considerably reduced during the first stages of a sea-level rise. 196 CARBONATE RESPONSE TO RISING RELATIVE SEA-LEVEL CHANGE The three types of responses to sea-level rise commonly observed in the Holocene as well as in the distant geologic past are listed in Table III, are illustrated in Figs.2--8, and will be discussed in the following section. Type A: drowning (Fig.3) Drowning is the result of the complete failure of a reef or platform to keep pace with a relative rise of sea level so that it leaves the realm of shallow-water carbonate sedimentation becoming submerged below the euphotic zone. The process is complete only when neritic carbonate production has ceased and truly deep-water conditions have been established. However, the onset of drowning expresses itself even within the euphoric zone by a change to deeper-water communities in reefs and on the lagoonal floors. A modern example for incipient drowning of a large platform is the outer Yucatan shelf in the Gulf of Mexico (Logan et al., 1969). Modern examples of drowned reefs include the relict Holoeene barrier reef off St. Croix in the Caribbean, (Adey et al., 1977), off eastern Florida (Lighty, 1977b) and off the N. Bahamas (Hine and Neumann, 1977). The St. Croix barrier consists of shallow-water Acroporapalmata reef capped by deeper-water head corals, while the Florida reef is now an inactive shelfTABLE III Types of response of carbonate platforms to sea-level rise, responses are defined by relatio~hip of rate of rise of relative sea level versus growth potential of platform rim and platform interior Type A: Drowning: Rise > rim, interior Complete failure of both rim and platform interior to keep pace with the rising sea, so platform is drowned Type B: catch up--survival of rim and patches," Rise > rim > interior quickly followed by rim > rise > interior later followed by During start-up phase both rim and interior fail to match rise (incipient drowning) while platform top still resides in euphotic zone; later the rim catches up with rise turning platform interior into deep lagoon that in turn acts as sediment trap rise > rim, interior Type C: Keep up--upbuitding and outbuilding: Rim, interior > rise Growth potential of both rim and interior of platform match or exceed rate of rise; no deep lagoons are developed and platform maintains flat top within few meters of sea level and tends to prograde basinward 197 margin reef with a central zone of Acroporapalmata. Drowning of Holocene reefs was probably brought about by a combination of rapid sea-level rise and adverse water conditions during incipient flooding of the wide flat platform tops (Adey et al., 1977; Lighty et al., 1978). This Holocene rise, however, was not great enough, nor did it last long enough, to establish pelagic conditions on the drowned reefs and platforms. A more complete stage of drowning can be observed on slope terraces in front of the present-day platform margin, such as the Pourtales Terrace east of Florida (Gomberg, 1976, pp.120-130). This feature, currently in 175--300 m of water depth, reached into the photic zone during the Pleistocene low stands when red algae grew on the upper part of the terrace. Whenever the sea rose to interglacial high stands, as for instance during the Holocene, algal growth ceased and pelagic carbonate deposition prevailed. The drowning of the Miami Terrace (Mullins and Neumann, 1979) is even more advanced. This terrace was part of a prograding platform margin until early Miocene time, and became subaerially exposed during the late Miocene drop in sea level (Peck et al., 1979). It was drowned during the subsequent rise and has been dominated by the accumulation of phosphorites and pelagic sediments ever since. Drowned reefs and platforms in the geologic record often have a similar appearance to reefs and platforms that were suffocated by clastic influx. Schlager (1980) recently compiled examples of platforms where "death by clastic pollution" can be ruled out because the termination of neritic carbonate deposition was followed by non-deposition or pelagic deposition. Good examples of drowned reefs and platforms are listed in Table II, E. Build-ups abandoned by the rise frequently are enveloped by a shale cap or deep-water limestone. Rises of relative sea level at rates of 103--104 Bubnoff units may be required to drown healthy platforms (Schlager, 1979, and Table I). This is several orders of magnitude faster than sea-level rises suggested by the secondand third-order cycles of Vail et ai. (1977). The high rate of sea-level rise required to drown a platform implies that these drowning pulses were probably of very short duration. If the sea rose at a rate of 104 Bubnoff units, some 104 years would suffice to increase water depth to 5 0 - 1 0 0 m and thus transfer the platform below the euphotic zone, even if some of the rise were compensated for by carbonate accumulation. In some cases, even less time and subsidence is needed. For instance, Purser (1969) describes cycles no thicker than 20 m, suggesting incipient drowning occurred below this depth. A sea-level pulse of 10,000 years may well be too short to be resolved by Vail et al.'s (1977) stratigraphic method of sea-level reconstruction. Certain mass extinctions of platforms may well be related to short-term sea-level events even if there is no obvious correlation with Vail et ai.'s second- and thirdorder cycles. A likely example of eustatic drowning is the Middle Cretaceous (Arthur and Schlanger, 1979; Schlager, 1980). 198 Type B: survival of rim and patches (Figs.4, 5) This type is a complex but common response that occupies an intermediate position between complete failure and complete success. The rate of rise is such that only the platform rim (normally a reef) and/or isolated patch reefs on the platform interior keep pace while the remainder of the platform is drowned, becoming a deep lagoon or shelf sea. This response typically occurs when the rising sea reaches the platform tops after a period of exposure. The pattern tends to develop in stages: during the "start-up" phase the rate of rise exceeds both growth rate of rim and interior, and the depositional setting shifts to deeper and more open-marine conditions, but the sea floor remains in the euphoric zone. During the second or "catch-up" phase the reef rim and newly established patch-reefs in the interior attain full growth rate and build to sea level. A third phase may follow when the interior lagoon fills up and a flat platform top is re-established. The end result of phase three will then be identical with the "keep-up" response where accumulation rates of rim and interior exceed the rate of rise throughout the event. However, in many cases the lagoon does not fill up at all or fills up after the raised rim and patches have existed for a long time, and for this reason, "survival of rim and patches" is treated as a separate response. Several modern platforms have reacted in this way to the Holocene rise. A good example is the Queensland Shelf of Australia, where only the rim, i.e., the Great Barrier Reef and Faros in the lagoon survived the Holocene transgression. Other examples are listed in Table II, G. Where examined in drill holes, the internal structure of reefs in these areas typically reveals an initial lag period followed by a shoaling-upward sequence of the catch-up phase. Best documented examples of this reef structure are the algal-coral reefs of St. Croix, in the Caribbean {Adey and Burke, 1976, p.98; Adey et al., 1977). Initially outpaced by the rapidly rising sea, the reefs caught up when sea-level rise slowed at 5000 years B.P. The reefs display a shoaling-upward succession of coral communities, presently capped by intertidal algal ridges. Other examples of reefs with lag and catch-up phase during the Holocene rise include the Alacran reef at the rim of the Yucatan shelf (Macintyre et al., 1977) and reefs in the Great Barrier Reef Complex {Davies and Marshall, 1979). Ancient examples of large platforms with survival of rim and patches include the Jurassic--Cretaceous continental shelf of eastern North America. The Blake Nose, for instance, occupies the seaward margin of a large carbonate platform that was drowned in Early Cretaceous (Barremian) time, while the margin continued to grow until Late Cretaceous (probably Campanian} time {Benson et al., 1979, p.107). Similar conditions may have prevailed elsewhere along the margin and an elevated carbonate shelf edge seems to have existed and dammed up clastics for considerable time after the continuous Jurassic platform was drowned {e.g., profiles in Grow et al., 1979). The Middle and Late Devonian of central Europe provide another example of a continuous platform that was drowned and succeeded by shelf atolls lined 199 up along its former margin and scattered over the interior (Krebs, 1974, p.164) as were the shelves on which the Devonian Swan Hills and Leduc reefs of western Canada (Stoakes and Wendte, 1979} and the Devonian of the Canning Basin of Western Australia (Playford, 1980). Other classic examples are the Miocene build-ups of Indonesia (Vincelette and Soeparjadi, 1976). Similarly, the Late Triassic Dachstein platform of the eastern Alps (Fischer, 1964; Zankl, 1971) was partly drowned and inundated by a transgressive sea in such a way that the ocean-facing reef belt in the south survived several million years longer than the remainder of the platform. There again, incipient drowning of the platform interior stimulated growth of patch reefs that first lagged behind the rising sea and later built to sea level through a succession of skeletal sand bodies, mud mounds and finally coral-algal frameworks (Schaeffer, 1979, p.22). Other ancient reefs displaying the combination of lag and catch-up phase in response to relative rise of sea level are numerous (Table II, H). Keep-up type: flat-topped prograding platforms (Figs.6, 7) In this type of response, growth potential of rim and interior matches or exceeds the rate of rise. The platform bucket fills to sea level, and in most cases the platform rim progrades seaward, building on excess sediment dumped on the flanks. These rims may consist of reefs or stacked carbonate sand shoals that record little or no apparent change in water depth during deposition. Cementation of the rim is usually marine but occasional freshwater cements suggest that the sediments built to sea level and sometimes contained a fresh-water lens within the exposed top of the rim. Examples include the Permian Capitan Formation of west Texas (King, 1948; Newell et al., 1953) and the Triassic of the southern Alps (Bosellini and Rossi, 1974). In some cases this rapidly prograding shelf margin is the source of carbonate turbidites (Newell et al., 1953; Thomson and Thomasson, 1969; Evans and Kendall, 1977}. The depositional environment over the platform interior varies from supratidal to just a few meters below low tide. During sea-level rises, shelf width will usually increase and clastics will tend to be confined to the landward side of the shelf by its width. Shelf sedimentation, particularly towards its seaward margin, is usually represented by shoaling-upward carbonate sequences. Occasionally during rises isolated depressions back from the shelf margins become evaporative lagoons. Here net inflow causes the deposition of varied evaporites and carbonates like the Cretaceous Ferry Lake anhydrite of the Gulf Coast (McFarland, 1977) and some Mesozoic evaporites of northern Africa and Europe {Van Houten, 1980). Individual shoaling cycles tend to be extremely widespread and, where clastic supply is low, are frequently terminated by supratidal evaporite sequences. On narrower shelves with low clastic supply, carbonates may dominate the seaward margin of the shelf while clastics cover most of the shelf. Where clastic supply is high and delta migration takes place, carbonates and clastics interfinger rhythmically. 200 When carbonate sedimentation exceeds the rate of sea-level rise the carbonates of the margin ingress over the abandoned delta lobes. This is extremely common in the Pennsylvanian of eastern United States (Ferm and Horn, 1979), and is true of the Pennsylvanian reciprocal sedimentation described by Van Siclen (1958) and Brown {1967) from Texas and the Pennsylvanian of New Mexico described by Wilson (1967). In the Devonian of western Canada clastics drive the basin margin seaward so that during each subsequent sea-level rise, the carbonate shelves are initiated in a more seaward position than the previous carbonate shelf (Stoakes and Wendte, 1979; Exploration Staff, Chevron Standard Limited, 1979; Stoakes, 1981). A carbonate shelf system can be expected to fill to the supratidal zone when a relative sea-level rise slows, and the excess sediment causes the coast to prograde seaward. The deposits are typically rhythmic or cyclic {Wilson, 1975, pp.281--309). Certain types of cycles, such as the shoaling-upward cycles ranging from shallow marine to supratidal, may be caused by some internal mechanism of feed-back, independent of an external dictating force. Ginsburg (1971), for instance, proposed a purely depositional origin for shoaling-upward cycles in platform carbonates based on two principles well documented in the Holocene: (1) after carbonates reach sea level, a lag time exists of several hundred to thousands of years between the flooding of an area caused by a rise of relative sea level and the onset of rapid sedimentation, and (2) though lime mud is the major sediment fraction, and is produced nearly everywhere on the platform, it accumulates preferentially along protected coasts, forming seaward prograding wedges of shallow marine deposits capped by tidal deposits. Ginsburg's prograding mud wedges would explain shoaling-upward cycles in platform carbonates independent of any high-frequency fluctuations of either sea level or rates of subsidence. Other cycles such as the Lofer cycles described by Fischer (1964) include terrestrial episodes that require an outside dictating force. These are best explained by rapid but small oscillations of sea level. Fischer (1975) suggests a correlation with the earth's orbital parameters ("Milankovich Cycles"). Fla~topped and thick shallow-water platforms are common in nearly all periods of the Phanerozoic and the late Precambrian (Table II, I). Neritic carbonate deposition is rather stable once it is initiated on flat-topped carbonate platforms. In the Bahamas (Late Jurassic to Recent) and in the southern Appenines (Middle-Triassic to Paleocene) flat-topped platforms were maintained for nearly 150 Ma (Paulus, 1972; Meyerhoff and Hatten, 1974; D'Argenio et al., 1975). The only interruptions in these sequences were caused by falls of relative sea level; all variations in the rate of rise of relative sea level were buffered within the platform system by varying the rate of upbuilding and outbuilding. Examples of modern platforms maintaining a flat top at sea level are difficult to demonstrate because Holocene sea level has not risen far enough above the Late Pleistocene contour to prove this point. Platforms flooded early during the Holocene, such as the Yucatan 201 shelf or the Queensland shelf, were not able to build up with the rising sea and responded by being drowned or having the rim and some reef patches survive (Table II, J). Some of these Holocene reefs were able to match a relative sea-level rise in the order of 2 • 103 to 9 • 103 Bubnoff units (mm/ 1000 yr.) by upward growth. Ancient examples of these build-ups can be seen in Table II, K). Rate o f sea-level rise vs. carbonate response The responses type A through C discussed above are defined qualitatively and a priori do not provide a key to the absolute rate of sea-level rise that caused them. We consider our classification no more than an intermediate step towards the ultimate goal of predictive models which will allow the stratigrapher, once having considered the rate of relative rise of sea level and other environmental parameters, to predict the response of carbonate platforms. At present, the relationship between rate of sea-level rise and a corresponding carbonate response is poorly defined because of lack of quantitative data and because the growth potential of platforms has probably varied considerably through space and time. The life span of carbonate platforms and reefs is commonly on the order of five to a few tens of millions of years and thus in the same range as the third-order sea-level cycles defined by Vail et al. (1977). Rates of sea-level rises within these cycles are known only to the nearest order of magnitude because of the uncertainty of the absolute amplitude in sea-level excursions (see Vail et al., 1977 vs. Hancock and Kauffman, 1979). Using Vail and Hardenbol's (1979) sea-level curve for the Tertiary we estimate the fast rises within the third-order cycles to be on the order of 25--200 Bubnoffs. Similar rates ( 1 0 - 9 0 Bubnoffs) have been calculated by Hancock and Kauffman (1979) for the transgressive pulses in the Cretaceous which are comparable in length to Vail's third-order cycles (Table I). The growth potential of carbonate reefs and platforms, on the ceher hand, seems to be at least one order of magnitude higher (Wilson, 1975, p.15; Schlager, 1979). Schlager {1980) estimates the growth potential of modem platforms to be near 1000 Bubnoffs (Table I). This estimate is based on the following observations: the rise of 8000-10,000 Bubnoffs during the Early Holocene appears to have been too fast for most reefs and platforms whereas most of them recovered during the later Holocene, when sea level rose at rates of 5 0 0 - 2 0 0 0 Bubnoffs. Many ancient platforms seem to have had similar potential. This is suggested by numerous examples of platforms that prograded while their flat tops kept pace with relative sea level rising at rates of tens to hundreds of Bubnoffs. If these estimates are correct, then the sea-level rises depicted by the thirdorder cycles of Vail et al., (1977) are too slow to explain types A and B responses on carbonate platforms. Rather, these responses, which both include drowning, may be caused by environmental stress accompanying a rise of sea level, or by a pulse in subsidence, or by fast, short-term eustatic 202 fluctuations of sea level superimposed on third-order cycles. A strong argument that the responses are often dictated by eustacy, is their non-random distribution through time, on a global basis, with certain types of responses clustered at specific time intervals but rare at others: For instance, global mass-extinctions of platforms (type-A response) are associated with midCretaceous transgressions, suggesting eustatic pulses of sea level whose rate and amplitude was sufficient to completely drown reefs and platforms (Schlager, 1980). Similarly, fast pulses but with smaller amplitude may have caused the widespread examples of type-B response in the Middle and Late Devonian, when areally widespread carbonate platforms were flooded several times and succeeded by barrier and patch reefs that stood high above the surrounding deep lagoon, like the former platform tops of Central Europe, (Krebs, 1974) and the Alberta basin (Stoakes and Wendte, 1979). Pulses of fast rate but small amplitude are suggested by the succession of hardgrounds (incipient drowning) followed by shoaling.upward sequences (catch-up phase}, which is common in the Middle and Late Jurassic {Purser, 1969; Murris, 1980). Type-C response where the depositional surface remains at sea level allows direct determination of the rate of relative rise of sea level that caused this particular response. Rates of relative rise during stable sea level or platform progradation are found to vary from a few tens to several hundreds of Bubnoffs (Schlager, 1980). Shoaling-upward cycles of carbonate are common in the centres of more stable shelves. The shelf interior appears to tend to have fewer complete shoaling-upward cycles since tongues of water associated with many individual sea-level rises are unable to extend all the way across the shelf interior. Complementing this, hiatuses are common to the shelf interior (Murris, 1980). In contrast, the shelf margin and basin centres may lack shallowwater sediments because the subsidence is rapid and so evidence of the progradational cycles can be obscured. Thus, where subsidence is extremely fast, as at the basin margin, particularly immediately after the break-up of a continent, cycles may be hidden by the effects of rapid subsidence. Similarly, basin interiors may subside so rapidly that tidal flats are prevented from developing. Thus, instead of the asymmetric shoaling-upward cycles common to stable shelves, symmetrical shoaling and deepening cycles can occur, as they do in the Cretaceous basin of the western interior of the U.S.A. (Kauffman, 1977). However, toward shelf centres as carbonates encroach on the continental interior the cyclic nature of the carbonates becomes more and more common. It should be realized that the supratidal evaporites associated with these shoaling-upward carbonates are formed only during stable sea level or sea-level rises and should not mistakenly be interpreted as forming during sea-level falls. It is our contention, the upper parts of many of Vail et al.'s {1977) onlap cycles may be dominated by a series of shoaling-upward carbonate sequences. For example, the Fredricksburg of Central Texas is capped by fourteen shoaling-upward cycles (Mueller, 1975}. Similarly, the shoaling-upward 203 sequence observed in the Oxfordian Smackover Formation of the U.S. Gulf Coast (Croft et al., 1980) or its North African equivalent in Tunisia match the top of a sea-level rise as do many of the shoaling-upward carbonates of the Middle East (Murris, 1980). These cycles tend to become more and more restricted upward. The effect ends when sea level reaches a standstill and then falls. The type of shoaling-upward cycles of the Fredricksburg and the Jurassic of the Middle East clearly differ from the very thick cyclic platform carbonates like that of the Triassic Dachstein Limestone (Fischer, 1975), the Dolomia Principale (Bosellini, 1965), and the Permian of the Guadelupe Mountains (Newell et al., 1953). These latter cycles are of still shorter duration, e.g. 50,000 years for the Dachstein cycles {Fischer, 1975}. Most of the cycles begin with a karst horizon enveloping lower subtidal deposits, which implies that the sea-level falls were on the order of 102--103 Bubnoffs and outpace the combined effect of subsidence and long-term sea-level movement {Table I). We believe, with Fischer (1975) that these cycles are the result of the waxing and waning of ice sheets responding to the variations of the earth's orbital parameters. The early cement paragenesis associated with the shoaling-upward carbonates is dominantly fresh-water, though schizohaline and marine cements are not uncommon. The extent of the cementation and diagenesis is probably controlled by the length of subaerial exposure of the prograding carbonate at sea level. RESPONSE O F C A R B O N A T E S T O D R O P S IN SEA L E V E L A small eustatic drop in sea level will expose shallow, flat carbonate shelves, no matter their width, producing a widespread unconformity (Fig.l). This is because, no matter the tectonic setting, carbonate-shelf sedimentation tends to build to sea level. Should the shelves have undergone tectonic deformation as they formed, some sea-level drops can cause erosion to cut down into the tectonically disturbed sediments. Thus a sea-level drop is like a camera shutter which freezes parts of the continuum of tectonic change and carbonate sedimentation and provides a relative-time scale for dating both widespread synchronous unconformities like those recognized by Suess (1906); Stille (1924); Sloss (1963); Sleep (1976) and Vail et al. (1977) and events less commonly identified with sea-level change like the deposition of shoaling-upward carbonate cycle. While carbonate platforms and reefs usu'ally keep pace with all but the fastest rises in relative sea level, they are very poorly equipped to shift the loci of carbonate production and deposition when there is a relative drop in sea level. The flanks of platforms are usually so steep that reefs or other carbonate facies belts, are unable to gradually migrate down slopes following the retreating sea. What little sediment is deposited during this retreat, is quickly removed by beach erosion and subsequent terrestrial weathering. Consequently, the most common record of sea-level drops on carbonate plat- 204 forms is a subaerial hiatus, associated with karst development and cliff erosion. Fresh-water cements like those described by Meyers (1974) from Mississippian limestones of New Mexico are common to sea-level falls. During standstills in the retreat of the sea, the connections of the basin to the open sea may be closed or nearly closed by fringing reefs or structural highs. This tendency towards isolation of the basin and the lack of clastic influx makes carbonate basins particularly prone to evaporite deposition when sea level drops. These basinal evaporites tend to form where carbonate deposition has previously occurred in confined basins set well within the margin of the continent. As Maiklem (1971) proposed for the Elk Point Basin of Canada, the great thickness of evaporites suggests this basin, and others like it, were still in contact with oceanic water while much of the evaporite formed. The mechanism of evaporite drawdown (Maiklem, 1971) allowed a net inflow of marine water into the basins. Table II, L lists basins where cycles of basinal evaporites and shallow-water carbonates occur and were produced by variations in sea level. The evaporites are associated with the sea-level lows and the carbonates with shelves during the sea-level highs. A sea-level drop usually terminates carbonate deposition at a carbonate margin, and the exposed carbonate of the shelf is cemented, so little detritus is shed, and, furthermore, the cemented carbonate partially inhibits the erosion of the clastics trapped on the shelf. The evaporites of the Paradox Basin, the Delaware Basin, the Michigan Basin and Zechstein, etc., are largely devoid of clastics during the evaporative phase, presumably because the rate of evaporite production masks the clastic influx. However, some basins at sea-level lows are dominated by shelfderived clastics because the basins have greater access to the open sea and so are not evaporative. The Permian Basin of Texas and New Mexico (Kendall, 1969) shows this clastic phase followed by a later evaporative one. Occasionally, a drop in sea level may be accompanied by downslope migration of the carbonate-producing area. For instance, in the Midland Basin during a sea-level low in the early Guadalupian, the previously deep Midland Basin sea-floor formed a shallow-water depositional setting in which the San Andres oolites were formed (Todd, 1976). Similarly, Esteban and Giner (1977) record the progradation of Miocene reefs downslope as sealevel drops in the Messinian of the Mediterranian. CONCLUSIONS (1) In carbonate platforms, relative rises of sea level may produce three different responses: (A) completely drowned platforms, submerged below the euphoric zone and capped by hardgrounds or condensed deep-water deposits; (B) platforms where rim and reef patches of the interior survive and match sea level while the remainder of the platform is drowned; (C) platforms where growth of both rim and interior match sea-level rise and a flat surface is maintained at sea level. (2) Response to rapid relative rise of sea level after exposure often pro- 205 ceeds in three phases: (A) start-up phase when carbonate accumulation lags behind the rising sea; (B) catch-up phase when accumulation exceeds the rate of sea-level rise and the reef or platform builds to sea level; (C) keep-up phase when accumulation closely matches the rate of rise and the top of the reef or platform remains at, or very close to, sea level. The stratigraphic record of these three phases of growth is a shoalingupward sequence overlying a hardground or lag deposit. (3) Vertical growth potential of reefs and platforms is one to two orders of magnitude higher than rate of eustatic sea-level rise in most of Vail et al.'s (1977) second- and third-order cycles. The type of carbonate response seems to be mainly controlled by the presence or absence of rapid sea-level pulse superimposed on these cycles and by tectonic subsidence. (4) Falls of relative sea level are associated with karst and soil development on the platform and the deposition of subtidal evaporites in adjacent semienclosed basins and clastics in more open-marine basins. ACKNOWLEDGEMENTS This paper was presented at the 26th International Geological Congress in Paris (Kendall and Schlager, 1980) and is dedicated to our friend and colleague Riccardo Assereto. The authors would like to express their appreciation of Jack Wendte, with whom they had many conversations and whose ideas they have drawn freely on. Also we thank Stan Frost, Bob Ginsburg, Mitch Harris, Ian Lerche, Robin Lighty, Macomb Jervey, Ken Mesolella, Conrad Neumann, Nahum Schneiderman, Frank Stoakes and Peter Vail for their valuable suggestions and editorial help. Finally, we would extend gratitude to Gulf Research for encouraging the publication of this paper. REFERENCES Adey, W.H., 1978. Coral reef morphogenesis: a multidimensional model. Science, 202: 831--837. Adey, W.H. and Burke, R., 1976. Holocene bioherms (algal ridges and bank-barrier reefs) of the eastern Caribbean. Geol. Soc. Am. Bull., 87: 95--109. Adey, W.H., Macintyre, I.G., Stuckenrath, R. and Dill, R.F., 1977. Relict barrier-reef system off St. Croix: its implications with respect to Late Cenozoic coral-reef development in the western Atlantic. Third Int. Coral Reef Syrup., Proc., 2: 15--22. Ahr, W.M., 1973. The carbonate ramp: an alternative to the shelf model. Trans. Gulf Coast Assoc. Geol. Soc., 23: 221--225. Arthur, M.A. and Schlanger, S.O., 1979. Cretaceous "oceanic anoxic events" as causal factom in development of reef-reservoired giant oil fields. Am. Assoc. Pet. Geol. Bull., 63: 870--885. Babcock, J.A., 1977. Calcareous algae organic boundstones, and the genesis of the upper Capitan Limestone (Permian, Guadalupian) Guadalupe Mountains, West Texas and New Mexico. Field Conf. Guidebk. Permian Basin Sect. SEPM, pp.3--44. Bathurst, R.G.C., 1959. The cavernous structure of some MiasissippianStromatactis reefs in Lancashire, England. J. Geol., 67: 506--521. Benson, W., Sheridan, R.E. et al.,1979. InitialReports of the Deep Sea DrillingProject, 44:1005 pp. 206 Bernoulli, D. and Jen~yns, H.C., 1974. Alpine, Mediterranean, and Central Atlantic Mesozoic faciesin relation to the early evolution of the Tethys. In: Modern and Ancient Geosynclinal Sedimentation, SEPM Spec. Publ., 19: 129--160. Bosellini,A., 1965. Lineamenti strutturalidelle Alpi Meridionali durante ilPermo-Trias. Mew_ Mus. Storia Nat., Venezia Tridentina, 15: 1--72. Bosellini,A., 1973. Modello geodinamico e paleotettonico delle Alpi Meridionali durante il Giurassico-Cretacico~ Accad. Naz. Lincei, 183: 163--205. Bosellini,A. and Rossi, D., 1974. Triassiccarbonate build-ups of the Dolomites, northern Italy. In: L.F. Laporte (Editor), Reefs in Time and Space. SEPM Spec. Publ., 18: 209--233. BoseUini, A. and Winterer, E.L., 1981. Subsidence and sedimentation on a Jurassic passive continental margin, southern Alps, Italy.Am. Assoc. Pet. Geol. Bull. (in press). Bromley, R.G., 1968. Burrows and borings in hard grounds. Medd. Dansk Geol. Foren., 18: 247--250. Brown, L.F., 1967. Virgil--Lower Wolfcamp repetitive depositional environments in north-central Texas. In: J.C. Elam and S. Chuber (Editors), Cyclic Sedimentation in the Permian Basin. West Texas Geol. Soc., 56: 115--134. Colacicchi, R.A., Passeri, L. and Pialli, G., 1970. Nuovi dati sul Giurese Umbro-Marchigiano ed ipotesi per unsuo inquadranento regionale. Mem. Soc. Geol. Ital., 9: 883--874. Coogan, A.H., Bebout, D.G. and Maggio, C., 1972. Depositional environments and geologic history of Golden Lane and Poza Rica Trend, Mexico, an alternative view. Am. Assoc. Pet. Geol. Bull., 56: 1419--1447. Cooper, M.R., 1977. Eustacy during the Cretaceous, its implications and importance. Palaeogeogr., Palaeoclimatol., Palaeoecol., 22: 1--60. Croft, W.Y., Druckman, Y.C. and Moore, C.H., 1980. Jurassic Smackover Carbonate Grainstone Reservoir, North Haynesville Field, Claiborne Parish, Louisiana. In: R. Halley and R.G. Louckes (Editors), Carbonate Reservoir Rocks. SEPM Core Workshop, 1: 120--136. D'Argenio, B., Pescatore, T. and Scandone, P., 1975. Structural pattern of the CampaniaLucanla Apennines. Cons. Naz. Ric., Quadern. Ric. Sci., 90: 313--327. Davies, G.R., 1977. Carbonate--anhydrite facies relationships, Otto Fjord Formation (Mississippian-Pennsylvanian), Canadian Arctic Archipelago. In: Reefs and Evaporites. Am. Assoc. Pet. Geol. Studies in Geology, 5: 145--168. Davies, P.J. and Marshall, J.F., 1979. Aspects of Holocene reef growth--substrate age and accretion rate. Search, 10:(7/8): 276--279. Donovan, D.T. and Jones, E,J.W., 1979. Causes of worldwide change in sea level. J. Geol. Soc. London., 136: 187--192. Dott, R.H. Jr., and Shaver, R.H. (Editors), 1974. Modern and Ancient Geosynclinal Sedimentation. SEPM Spec. Publ., 19~ 380 pp. Dravis, J.J., 1977. Holocene sedimentary depositional environments on Eleuthera Bank, Bahamas. M.Sc. thesis, Univ. Miami, 386 pp. Dumestre, A. and Illing, L.V., 1967. Middle Devonian reefs in the Spanish Sahara. In: D.H. Oswald (Editor), Int. Symp. Devonian System. Calgary, Alta. Soc. Pet. Geol., 2: 333--350. Emery, K.O., Tracey, J.I. and Ladd, H.S., 1956. Geology of Bikini and nearby atolls. U.S. Geol. Surv. Prof. Pap. 260A: 1--259. Esteban, M. and Giner, J., 1977. Field-guide to Santa Pola reef: Messinian Seminar 3, Project 96: Measinian Correlation, Malaga and Granda Univ., Spain. Evans, I. and Kendall, C.G.St.C., 1973. Coral reefs and bioherm morphology -- tools for the interpretation of the Jurassic basin of the central High Atlas Mountains, Morocco (abstr.). Am. Assoc. Pet. Geol. Bull., 57: 778. Evans, I. and Kendall, C.G. St. C., 1977. An intepretation of the depositional setting of some deep-water Jurassic carbonates of the central High Atlas Mountains, Morocco. SEPM Spec. Publ., 25: 249-261. 207 Exploration Staff, Chevron Standard Limited, 1979. The geology, geophysics and significance of the Nisku reef discoveries, west Pembina Area, Alta., Canada. Bull. Can. Pet. Geol., 27: 326--359. Feray, D.E., 1967. Tectonic versus eustatic control of Pennsylvanian cyclical sedimentation in North-Central Texas. In: J.C. Elam and S. Chuber (Editors), Cyclic Sedimentation in the Permian Basin. West Texas Geol. Soc., 2nd ed., pp.81--82. Ferm, J.C., 1968. A review of "Cyclic Sedimentation" by R.H. Duff et al. Geotimes, 13: 9--30. Ferm, J.C. and Horn, J., 1979. Carboniferous Depositional Environments in Appalachian region. Coal Geology, Univ. South Carolina, 760 pp. Fischer, A.G., 1964. The Lofer cyclothems of the Alpine Triassic. Kans. Geol. Surv. Bull., 169: 107--149. Fischer, A.G., 1969. Geological time--distance rates: the Bubnoff unit. Geol. Soc. Am. Bull., 80: 549-552. Fischer, A.G., 1975. Tidal deposits, Dachstein Limestone of the North Alpine Triassic. In: R.N. Ginsburg (Editor), Tidal Deposits. Springer, New York, N.Y., pp.235--242. Ginsburg, R.N., 1971. Landward movement of carbonate mud: new model for regressive cycles in carbonates (abstr.) Am. Assoc. Pet. Geol. Bull., 55: 340. Gomberg, D.N., 1976. Geology of the Pourtales Terrace, Straits of Florida. Doctoral dissertation Univ. Miami, 371 pp. (unpublished). Grabau, A.W., 1920. A textbook of Geology, 1. General Geology. D.C. Heath and Co., New York, N.Y., 83 pp. Grow, J.A., Mattick, R.E. and Schlee, J.S., 1979. Multi-channel seismic depth sections and interval velocities over outer continental shelf and upper continental slope between Cape Hatteras and Cape Cod. Am. Assoc. Pet. Geol. Mere., 29: 65--83. Hallam, A., 1969. A pyritized limestone hardground in the Lower Jurassic of Dorset (England). Sedimentology, 12: 231--240. Hallam, A., 1978. Eustatic cycles in the Jurassic. Palaeontol., Palaeoclimatol., Palaeoecol., 23: 1--32. Hancock, J.M. and Kauffman, E.G., 1979. The great transgressions of the Late Cretaceous. J. Geol. Soc. London, 136: 175--186. Harris, P.M., 1979. Facies anatomy and diagenesis of a Bahamian ooid shoal. University of Miami, Fla, Sedimenta, 7:163 pp. Harris, P.M., Frost, S.H., Shaffer, B.L., Schneidermann, N. and Kendall, G.St.C., 1980. Cretaceous sea levelsand stratigraphy,eastern Arabian peninsula. Am. Assoc. Pet. Geol. Bull. (Abstr.),64: 719. Harris, T.J.,Hay, J.T.C. and Twombley, B.N., 1968. Contrasting limestone reservoirsin the Murban Field, Abu Dhabi. In: 2nd Regional Tech. Symp., Soc. Pet. Eng. AIME, Dhahran, Saudi Arabia, pp.149--182. Hays, J.D. and Pitman, W.C., Ill,1973. Lithospheric plate motion, sea-levelchanges and climatic and ecologicalconsequences. Nature, 246: 18--22. Heckel, P.H., 1974. Carbonate buildups in the geologic record: a review. In: L.F. Laporte (Editor), Reefs in Time and Space. S E P M Spec. Publ., 18: 90--154. Heezen, B.C., Matthews, J.L.,Catalano, R., Natiand, J.,Coogan, A., Tharp, M. and Rawson, M., 1973. Western PacificGuyots. In: B.C. Heezen et al.,InitialReports of the Deep Sea DrillingProject, 20: 653--723. Heybroek, F., 1965. The Red Sea Miocene evaporite basin. In: Salt Basins around Africa. London Inst.Petroleum, pp.17--40. Hine, A.C. and Neumann, A.C., 1977. Shallow carbonate-bank-margin growth and structure, LittleBahama Bank, Bahama. Am. Assoc. Pet. Geol. Bull.,61: 376--406. Hsii, K.J., 1972. Origin of saline giants: a critical review after the discovery of the Mediterranean evaporites. Earth Sci. Rev., 8: 371--396. Huh, J.M., Briggs, I. and Gill, D., 1977. Depositional environments of Pinnacle Reefs, Niagara and Salina Groups, Northern Shelf, Michigan Basin. In: J.H. Fisher (Editor), Am. Assoc. Pet. Geol. Studies in Geology, 5. Reefs and Evaporites, pp.1--22. 208 James, N.P. and Ginsburg, R.N., 1980. The seaward margin of Belize barrier and atolls reefs. Int. Assoc. Sedimentol. Spec. Publ., 3: 1--190. James, N.P., Ginsburg, R.N., Marszalek, D.S. and Choquette, P.W., 1976. Facies and fabric specificity of early subsea cements in shallow Belize (British Honduras) reefs. J. Sediment. Petrol., 46: 523--544. Kauffman, E., 1977. Geological and biological overview: western interior Cretaceous Basin. Mountain Geol., 14: 75--99. Kendall, C.G.St.C., 1969. An environmental re-interpretation of the Permian evaporite/ carbonate shelf sediments of the Guadalupe Mountains. Geol. Soc. Am. Bull., 80: 2503--2526. Kendall, C.G.St.C. and Schlager, W., 1980. Response of carbonate platforms to relative changes in sea level (abstr.). 26th Int. Geol. Congr. Sect., 6: 493. King, P.B., 1948. Geology of the southern Guadalupe Mountains, Texas. U.S. Geol. Surv. Prof. Pap., 215: 1--183. Klovan, J.E., 1974. Development of Western Canadian Devonian reefs and comparison with Holocene analogues. Am. Assoc. Pet. Geol. Bull., 58: 787--799, Krebs, W., 1974. Devonian carbonate complexes of Central Europe. In: L.F. Laporte (Editor), Reefs in Time and Space. SEPM Spec. Publ., 18: 155--208. Kuenen, Ph.H., 1939. Quantitative estimations relating to eustatic movements. Geol. Mijnbouw, 8: 194--201. Land, L.S. and Goreau, T.F., 1970. Submarine lithification of Jamaican reefs. J. Sediment. Petrol., 40: 457--462. Langston, J.R. and Chin, G.E., 1968. Rainbow member facies and related reservoir properties, Rainbow Lake, Alberta. Bull. Can. Pet. Geol., 16: 104--143. Lees, A., 1964. The structure and origin of the Waulsortian (Lower Carboniferous) "reefs" of west-central Eire. Philos. Trans. R. Soc. London, Set. B, 247: 483--531. Lemcke, K., 1975. Possible consequences of the impact of large meteorites in the ocean. Neues Jahrb. Geol. Palaeontol., Monatsh., 12: 719--726. Lighty, R.G., 1977a. Submarine diagenesis in relict Holocene coral reef, southeast coast of Florida, diagenetic control on porosity and stratigraphy. Am. Assoc. Pet. Geol. Bull., 61: 808. Lighty, R.G., 1977b. Relict shelf-edge Holoeene coral reef, southeast coast of Florida. Third Int. Coral Reef Symp., Proc., 2: 215--222. Lighty, R.G., Macintyre, I.G. and Stukenrath, R., 1978. Submerged Early Holocene barrier reef southeast Florida shelf. Nature, 276: 59--60. Logan, B.W., Harding, J.L., Ahr, W.M., Williams, J.C. and Snead, R.G., 1969. Late Quaternary carbonate sediment in Yucatan Shelf, Mexico. In: B.W. Logan et al., Carbonate Sediments and Reefs, Yucatan Shelf, Mexico. Am. Assoc. Pet. Geol. Mere., 11: 5--128. Longman, M.W., 1980. Carbonate diagenetic textures from near-surface diagenetic environments. Am. Assoc. Pet. Geol. Bull., 64: 461--487. Macintyre, I.G. and Glynn, P.W., 1976. Evolution of modern Caribbean fringing reef, Galeta Point, Panama. Am. Assoc. Pet. Geol. Bull., 60: 1054--1072. Macintyre, I.G., Burke, C.A. and Stuckenrath, R., 1977. Thickest recorded Holocene reef section, Isla Perez core hole, Alacran Reef, Mexico. Geology, 5: 749--754. MacNeil, F.S., 1954. Organic reefs and banks and associated detrital sediments. Am. J. Sci., 252: 385--401. Maiklem, W.R., 1971. Evaporative drawdown -- a mechanism for water-level lowering and diagenesis in the Elk Point Basin. Bull. Can. Pet. Geol., 19: 487--503. Matthews, J.L., Heezen, B.C., Catalano, R., Coogan, A., Tharp, M., Natland, J. and Rawson, M., 1974. Cretaceous drowning of reefs on Mid-Pacific and Japanese guyots. Science, 184: 462--464. Maxwell, W.G.H., 1968. Atlas of the Great Barrier Reef. Elsevier, Amsterdam, 258 pp. McFarland, E., Jr., 1977. Lower Cretaceous sedimentary facies and sea-level changes U.S. Gulf Coast. In: D.G. Bebout and R.G. Loucks (Editors), Cretaceous Carbonates of Texas and Mexico. Bur. Econ. Geol. Austin Report, 89: 5--11. 209 Meissner, F.F., 1972. Cyclic sedimentation in Middle Permian strataof the Permian basin, West Texas and N e w Mexico. In: J.C. Elam and S. Chuber (Editors),Cyclic Sedimentation in the Permian Basin. West Texas Geol. Soc., Midland, Texas, 2nd ed., pp.203--232. Mesolella, K.J., Sealy, H.A. and Matthews, R.K., 1970. Facies geometries within Pleistocene reefs of Barbados, West Indies. Am. Assoc. Pet. Geol. Bull., 54: 1899--1917. Mesolella, K.J., Robinson, J.D., McCormick, L.M. and Ormiston, A.R., 1974. Cyclic deposition of Silurian carbonates and evaporites in the Michigan. Am. Assoc. Pet. Geol. Bull., 58: 34--62. Meyerhoff, A.A. and Hatten, C.W., 1974. Bahamas salient of North America. In: C.A. Burk and C.L. Drake (Editors), The Geology of Continental Margins. Springer, New York, N.Y., pp.429--446. Meyers, W.J., 1974. Carbonate cement stratigraphy of the Lake Valley Formation (Mississippian), Sacramento Mountains, New Mexico. J. Sediment. Petrol., 44: 837-861. Mueller, H.W., III, 1975. Centrifugalprogradation of carbonate banks --a model for depositional diagenesis, Fort Terrett Formation, Edwards Group, Lower Cretaceous, Central Texas. Ph.D. thesis Univ. Texas, Austin, 420 pp. Mullins, H.T. and Neumann, A.C., 1979a. Geology of the Miami Terrace and its paleooceanographic implications. Mar. Geol., 30: 205--232. Mullins, H.T. and Neumann, A.C., 1979b. Deep carbonate bank-margin structure and sedimentation in the northern Bahamas. SEPM Publ., 27: 165--192. Murris, R.J., 1980. Middle East stratigraphic evolution and oil habitat. Am. Assoc. Pet. Geol. Bull., 64: 597--618. Neumann, A.C., Kofoed, J.W. and Keller, G.H., 1977. Lithotherms in the Straits of Florida. Geology, 5: 4--10. Newell, N.D., Rigby, J.K., Fischer, A.G., Whiteman, A.J., Hickox, J.E. and Bradley, J.S., 1953. The Permian Reef Complex of the Guadalupe Mountains Region, Texas and New Mexico. Freeman, San Franciso, Ca., 236 pp. Ott, E., 1967. Segrnentierte Kalkschw[imme (Sphinctozoa) aus der alpinen Mitteltrias und ihre Bedeutung als Riffbildner im Wettersteinkalk. Abh. Bayer. Akad. Wiss., Math.Naturwiss. Kl., Neue Folge, 131:96 pp. Parsons, B. and Sclater, J.G., 1977. An analysis of the variation of ocean-floor bathymetry and heat flow with age. J. Geophys. Res., 82: 803--827. Paulus, F.J., 1972. The geology of site 98 and the Bahamian platform. In: C.D. Hollister, J.I. Ewing et al., Initial Reports of the Deep Sea Drilling Project, 11: 877--897. Peck, D.M., Missimer, T.M., Slater, D.H., Wise, S.W. and Odonnell, T.H., 1979. Late Miocene glacial-eustatic lowering of sea level -- evidence from the Tamiami Formation of south Florida. Geology, 7: 285--288. Peterson, J.A. and Ohlen, H.R., 1963. Pennsylvanian shelf carbonates, Paradox basin. In: R.O. Bass and S.L. Sharp (Editors), Shelf Carbonates of the Paradox Basin. Four Comers Geol. Soc. 4th Field Conf. Syrup., Durango, Colo., pp.65--79. Pitman III, W.C., 1978. Relationship between eustacy and stratigraphic sequences of passive margins. Geol. Soc. Am. Bull., 89: 1389--1403. Playford, P.E., 1980. Devonian "Great Barrier Reef" of Canning Basin, western Australia. Anx Assoc. Pet. Geol. Bull., 64: 814--840. Playford, P.E. and Lowry, D.C., 1966. Devonian reef complexes of the Canning basin, western Australia. Geol. Survey W.A. Bull., 118:150 pp. Poag, C.W., 1979. Stratigraphy and depositional environments of Baltimore Canyon Trough. Am. Assoc. Pet. Geol. Bull., 63: 1452--1466. Purdy, E.G., 1974. Karst-determined facies patterns in British Honduras: Holocene carbonate sedimentation model. Am. Assoc. Pet. Geol. Bull., 58: 825--855. Purser, B.H., 1969. Synsedimentary marine lithification of Middle Jurassic limestones in the Paris Basin. Sedimentology, 12: 205--230. 210 Read, J.F., 1978. Geometry, facies and development of large non-reefal shelf and offshelf build-ups, Middle Ordovician, Virginia (abstr.)Am. Assoc. Pet. Geol. Bull.,89: 557. Read, J.F., 1980. Carbonate ramp-to-basin transitions and foreland basin evolution, Middle Ordovician, Virginm Appalachians. Am. Assoc. Pet. Geol. Bull., 64:1575--1612. Richter-Bernburg, G., 1955. Uber salinare Sedimentation. Z. Dtsch. Geol. Ges., 105: 599--645. Roberts, D.G., Montadert, L. and Seaxle, R.C., 1979. The Western Rockall plateau: stratigraphy and structural evolution. In: L. Montadert et al., Initial Reports of the Deep Sea Drilling Project, 48: 1061-1088. Robertson Handford, C. and Dutton, S.P., 1980. Pennsylvanian--Early Permian depositional systems and shelf-margin evolution, Palo Duro Basin, Texas. Am. Assoc. Pet. Geol. Bull., 64: 88--106. Ryan, W.F., Rad, U., et al., 1979. Initial Reports of the Deep Sea Drilling Project, 47: 835 pp. Schaeffer, P., 1979. Facies and paleoecology of two Upper Triassic reef-complexes in the northern Calcareous Alps ("Upper Rhaetian" reef limestones, Salzburg, Austria). Facies, 1 : 3--245. Schlager, W., 1979. Drowning of carbonate platforms. Geol. Soc. Am., Abstr. Programs 11: 511--512. Schlager, W., 1980. Mesozoic calciturbidites in DSDP Hole 416A -- petrographic recognition of a drowned carbonate platform. In: E.L. Winterer et ah, Initial Reports of the Deep Sea Drilling Project, 50 : 733--749. Schlanger, S.O. and Premoli-Silva, I., 1981. Tectonic, volcanic and paleogeographic implications of redeposited reef faunas of Late Cretaceous and Tertiary age from the Naurn Basin and Line Islands. In: R.L. Larson, S.O. Schlanger et al., Initial Reports of the Deep Sea Drilling Project, 61 (in press). Scholle, P.A. and Arthur, M.A., 1980. Carbon isotope fluctuations in Cretaceous pelagic limestones: potential stratigraphic and petroleum exploration tool. Am. Assoc. Pet. Geol. Bull., 64: 67--87. Schwab, F., 1976. Modern and ancient sedimentary basins: comparative accumulation rates. Geology, 4: 723--727. Sclater, J.G., Anderson, R.N. and Bell, M.L., 1971. Elevation of ridges and evolution of the central eastern Pacific. J. Geophys. Res., 76: 7888--7915. Shinn, E.A., 1969. Submarine lithification of Holocene carbonate sediments in the Persian Gulf. Sedimentology, 12 : 109--149. Shinn, E.A., Halley, R.B., Hudson, J.H., Lidz, B. and Robbin, D.M., 1979. Threedimensional aspects of Belize patch reefs (abstr.). Am. Assoc. Pet. Geol. Bull., 63: 528. Shinn, E.A., Hudson, J.H., Halley, R.B., Lidz, B., Robbin, D.M. and Macintyre, I.G., 1981. Geology and sediment accumulation rates: Carrie Bow Cay, Belize, Central America. Smithson, Contrib. ~a~, Sci. (in press). Shipley, T.H., Buffler, R.T. and Watkins, J.S., 1978. Seismic stratigraphy and geologic history of Blake Plateau and adjacent Western Atlantic continental margin. Am. Assoc. Pet. Geol. Bull., 62: 792--812. Silver, B.A. and Todd, R.G., 1969. Permian cyclic strata, northern Midland and Delaware Basins, West Texas and southeastern New Mexico. Am. Assoc. Pet. Geol., 53: 2223--2251. Sleep, N.H., 1976. Platform subsidence mechanisms and "eustatic" sea-level changes. Tectonophysics, 36: 45--56. Sloss, L.L., 1963. Sequences in the cratonic interior of North America. Geol. Soc. Am. Bull., 74: 93--114. Smit, J. and Hertogen, J., 1980. An extraterrestial event at the Cretaceous--Tertiary boundary. Nature, 285: 198--200. Steckler, M.S. and Watts, A.B., 1978. Subsidence of the Atlantic-type continental margin off New York. Earth Planet. Sci. Lett., 41: 1--13. Stille, H., 1924. Grundfragen der vergleichenden Tektonik. Borntr~'ger, Berlin, 443 pp. 211 Stoakes, F.A., 1981. Sea-level control of carbonate-shale deposition during progradational basin-filllng: the Upper Devonian, Duvernay and Ireton formations of Alberta, Canada. Bull. Can. Pet. Geol. (in press). Stoakes, F.A. and Wendte, J., 1979. Sea-level control of the sedimentation patterns during transgression and regression: Late Devonian of western Canada (abstr.). Can. Soc. Pet. Geol. Bailtie Syrup. Recent Advances in Carbonate Sedimentology in Canada, pp.26--27. Suess, E., 1906. The Face of the Earth, 2. Clarendon Press, Oxford, 556 pp. Taft, W.H., Arrington, R., Haimovitz, A., MacDonald, C. and Woolheater, C., 1968. Lithification of modern carbonate sediments at Yellow Bank, Bahamas. Bull. Mar. Sci. Gulf Caribb., 18: 762--828. Terry, X.Y. and Williams, X.Y., 1969. The Idris "A" geoherm and oilfield, Sirte Basin, Libya: its commercial development, regional Paleocene geologic setting and stratigraphy. In: The Exploration for Petroleum in Europe and North Africa. Inst. Petroleum, London, pp.31--48. Thomson, A.F. and Thomasson, M.R., 1969. Shallow- to deep-water facies development in the Dimple Limestone (Lower Pennsylvanian) Marathon Region, Texas. SEPM Spec. Publ., 14: 57--78. Todd, R.G., 1976. Oolite-bar progradation, San Andres Formation, Midland Basin, Texas. Am. Assoc. Pet. Geol., 60: 907--925. Traeey, J.I. and Ladd, H.S., 1974. Quaternary history of Eniwetok and Bikini atolls, Marshall Islands. Proc. 2nd Coral Reef Syrup., 2, Brisbane, pp.537--550. Tucker, M.E., 1974. S edimentology of Paleozoic pelagic limestones: the Devonian Griotte (southern France) and Cephalopodenkalk (Germany). Int. Assoc. Sedimentol. Spec. Publ., 1: 71--92. Umbgrove, J.H.F. 1939. On rhythms in the history of the earth. Geol. Mag., 76:116--129. Vail, P.R. and Hardenbol, J., 1979. Sea-level changes during the Tertiary. Oceanus, 22: 71--79. Vail, P.R., Mitchum, R.M., Jr., Todd, R.G., Wildmier, J.M., Thompson, S. III, Sangree, J.B., Bubb, J.N. and Hatfield, W.G., 1977. Seismic stratigraphy and global changes of sea level. In: Seismic stratigraphy -- Applications to Hydrocarbon Exploration. Am. Assoc. Pet. Geol. Mere., 26: 49--212. Van Siclen, D.C., 1958. Depositional topography -- examples and theory. Am. Assoc. Pet. Geol. Bull., 42: 1897--1913. Van Hinte, J.E., 1978. Geohistory analysis --application of micropaleontology in exploration geology. Am. Assoc. Pet. Geol. Bull., 62: 201--222. Van Houten, F.B., 1980. Latest Jurassic--Cretaceous regressive facies, northeast Africa craton. Am. Assoc. Pet. Geol. Bull., 64: 857--867. Vincelette, R.R. and Soeparjadi, R.A., 1976. Oil-bearing reefs in Salawati Basin of Irian Jaya, Indonesia. Am. Assoc. Pet. Geol. Bull., 60: 1448--1462. Walls, R.A., Mountjoy, E.W. and Fritz, P., 1979. Isotopic composition and diagenetic history of carbonate cements in Devonian Golden Spike reef, Alberta, Canada. Can. Geol. Soc. Bull., 90: 963--982. Wantland, K.W. and Pusey, W.C., 1975. Belize Shelf Carbonate Sediments, Clastic Sediments and Ecology. Am. Assoc. Pet. Geol., Studies in Geology, 2:598 pp. Watts, A.B. and Ryan, W.B.F., 1976. Flexure of the lithosphere and continental margin basins. Tectonophysics, 36:25--44. Watts, A.B. and Steekler, M.S., 1979. Subsidence and eustasy at the continental margin of eastern North America. In: Deep Drilling Results in the Atlantic Ocean, Ewing Series, 3: 218--234. Wendte, J., 1981. Platform evolution and its control on reef inception and localization, Upper Devonian, Redwater, Alberta. Bull. Can. Pet. Geol. (in press). Wendte, J.C. and Gensamer, A.R., 1979. Pore systems in Jurassic carbonate reservoirs, United States Gulf Coast (abstr.) Am. Assoc. Pet. Geol. Bull., 63: 551. Wilson, J.L., 1967. Cyclic and reciprocal sedimentation in Virgilian strata of southern New Mexico. Geol. Soc. Am. Bull., 78: 805--818. 212 Wilson, J.L., 1975. Carbonate Facies in Geologic History. Springer, Berlin, 471 pp. Wong, P.K., 1979. Sequential cementation in the Upper Devonian Kaybob Reef, Alberta (abstr.). Can. Soc. Pet. Geol. Syrup. in honor of Dr. A.D. Baillie -- Recent Advances in Carbonate Sedimentology in Calgary, Canada, pp. 24--30. Yurewicz, D.A., 1977. The origin of the massive facies of the Lower and Middle Capitan (Permian) Guadalupe Mountains, New Mexico and West Texas. Field Conf. Guidebk. Permian Basin Sect. SEPM. pp.45--92. Zankl, H., 1968. Sedimentological and biological characteristics of a Dachsteinkalk reef complex in the Upper Triassic of the northern Calcareous Alps. In : G. Mueller and G.M. Friedman (Editors), Recent Developments in Carbonate Sedimentology in Central Europe. Springer, Berlin, pp.215--218. Zankl, H., 1971. Upper Triassic carbonate facies in the northern Limestone Alps. In: G. Mueller (Editor), Sedimentology of Parts of Central Europe -- Heidelberg, Waldemar Kramer, Frankfurt am Main, pp. 147--185.