distinguishing the roles of autogenic versus allogenic ...web.mst.edu/~yangwa/yang webpage/yang et...

ABSTRACT

Meter-scale transgressive-regressive cyclesof the subsurface Cisco Group are composedof marine and nonmarine carbonate and silici-clastic rocks deposited on the Eastern shelf ofthe Midland basin during Late Pennsylvanianand Early Permian time. Five cycle types arecharacterized by thickness, magnitude, order,and principal lithofacies. Cycle magnitude isdefined as the maximum facies shift in a cycle,indicating extent of shoreline migration. Ciscocycles belong to three orders—minor, interme-diate, and major—and they are superimposedand form a stratigraphic hierarchy. Each or-der of cycles has a distinct range of thicknessand possibly duration. A cycle is also dividedinto a lower sand-poor interval, during whichcoarse siliciclastic supply at the depositionalsite was diminishing, and an upper sand-richinterval, during which coarse siliciclastic sup-ply was high. Regional thickness and lithofa-cies variations of sand-rich intervals indicatethat progradational infilling at a depositionalsite lagged marine regression, suggesting a de-lay in sediment supply from the upland sourcerelative to the time of base-level fall.

Regional systematic variations in cycleabundance, continuity, and characteristicsalong depositional dip and strike record the in-terplay among regional topography, pattern ofsiliciclastic supply, and shelf subsidence, whichcontrolled distribution of depocenters and by-pass zones and, thus, stratigraphic complete-ness and resolution. Regional persistence of cy-cles suggests a eustatic control on regional,ordered transgressive-regressive events. Incontrast, local variations of cycle characters

suggest controls by local topography and dep-ositional dynamics, which determined deposi-tional loci, differential compaction, and ero-sion. A predominantly autocyclic Cisco recordin the upper platform does not imply the ab-sence of allogenic processes. An allocyclicCisco record in the lower platform containsabundant autocyclic imprints, because allo-genic controls on cyclic sedimentation were ac-complished through local autogenic processes.Distinguishing the roles of autogenic versus al-logenic processes in cyclic sedimentation is animportant step in establishing a high-resolu-tion (meter-scale) chronostratigraphy of anysedimentary record1.

INTRODUCTION

Meter-scale depositional cycles are fundamen-tal stratigraphic entities. Cyclicity at this scale iscommonly defined by stratal repetition of physi-cal and chemical characters of sedimentary rocks,such as lithofacies, biofacies, and stable isotopiccomposition. Temporal regularity of depositionalcyclicity, which may be derived from strati-graphic regularity, is the basis for establishing ahigh-resolution cyclostratigraphy (e.g., Herbert,1992; Hinnov and Goldhammer, 1991; Houseand Gale, 1995; Yang et al., 1995). Stratigraphicregularity, however, is commonly altered or de-stroyed by the complex dynamics of processescontrolling deposition and erosion, as demon-strated in this study.

Stratigraphic regularity is controlled by the interplay of many autogenic and allogenicprocesses in cyclic sedimentation (e.g., Wanless

and Shepard, 1936; Galloway, 1971). The twotypes of processes are separable by their physicalscales. Allogenic processes operate at a basin-wide or global scale, such as sea-level change,basin-wide tectonics, and regional climaticchange, whereas autogenic processes operate lo-cally, such as those intrinsic to specific deposi-tional or geomorphic environments. Autogenicand allogenic processes interact and produce cy-cles of variable characteristics. Thus, differentia-tion of local vs. regional and short-term vs. long-term variations of cycle characteristics is criticalto establishing a reliable high-resolution cy-clostratigraphy (Schwarzacher, 1993).

In this study we analyzed the formation, de-struction, and preservation of meter-scale cyclicityof the Cisco Group by examining three-dimen-sional variation of cycle characteristics, includingcycle abundance, continuity, type, magnitude,thickness, and variation in siliciclastic supply. Themechanisms of allogenic and autogenic processesin cyclic sedimentation are demonstrated throughcycle correlation. This study also displays thestratigraphic variability of late Paleozoic cy-clothems that could be missed in a one- or two-di-mensional analysis.

GEOLOGIC SETTING AND PREVIOUS WORK

The Cisco Group is composed of nonmarineand marine, mixed carbonate and siliciclasticrocks deposited on the shallow and stable Easternshelf of the Midland basin, north-central Texas,during Late Pennsylvanian and Early Permiantime (Figs. 1, 2, and 3; Lee, 1938; Wermund andJenkins, 1969; Brown et al., 1990). Deposition ofthe Cisco Group was influenced by a variety oflocal and regional processes, such as carbonateand siliciclastic depositional dynamics, climate,

1333

Distinguishing the roles of autogenic versus allogenic processes in cyclic sedimentation, Cisco Group (Virgilian and Wolfcampian), north-central Texas

Wan Yang* University of Texas, Bureau of Economic Geology, Austin, Texas 78713-8924

Michelle A. Kominz Department of Geology, Western Michigan University, Kalamazoo, Michigan 49008

R. P. Major Department of Geology and Geological Engineering, University of Mississippi, University, Mississippi 38677

GSA Bulletin; October 1998; v. 110; no. 10; p. 1333–1353; 21 figures; 1 table.

*E-mail: [email protected].

1Please contact the senior author to obtain sixcomplete stratigraphic and structural cross sections ofthis study.

YANG ET AL.

1334 Geological Society of America Bulletin, October 1998

and sea-level change, which controlled sedimentsupply and accommodation space with an evolv-ing depositional topography (e.g., Lee, 1938;Brown, 1969; Galloway, 1971; Harrison, 1973;Brown et al., 1987, 1990; Boardman and Heckel,1989; Yang, 1995, 1996).

Previous cyclostratigraphic studies of theCisco Group have concentrated on cycle deline-ation and regional correlation of fragmentaryoutcrop sections, resulting in limited under-standing of the interplay among autogenic andallogenic processes (e.g., Lee, 1938; Boardman

and Malinky, 1985; Boardman and Heckel,1989; Yancey, 1991). This study uses subsurfacedata to provide a more complete three-dimen-sional description of Cisco cyclostratigraphy anda more thorough assessment of autogenic and al-logenic controls on cyclic sedimentation of theCisco Group.

DATA AND METHODOLOGY

Three dip and three strike cross sections of theCisco Group covering ~35 × 100 km2 (Fig. 1)

were constructed, using 71 wells. Average wellspacing is 5 km. Dip cross sections begin at theoutcrop belt in the updip part of the shelf and ex-tend basinward to the west. The study area is alsocovered by cross sections of Brown et al. (1987;Fig. 1B). They correlated 23 regional limestonesin ~5000 wells on the Eastern shelf with those inthe outcrop to establish a stratigraphic frameworkcomposed of 16 depositional sequences (Figs. 2and 3). In our study we identified and correlatedthese limestones in 71 wells with the ~270 wellsof Brown et al. (1987) and with outcrop counter-

��

QAc1970c

000

000

10 mi10 mi10 mi

15 km15 km15 km

JONESJONESJONES SHACKELFORDSHACKELFORDSHACKELFORD STEPHENSSTEPHENSSTEPHENS

TAYLORTAYLORTAYLOR CALLAHANCALLAHANCALLAHAN EASTLANDEASTLANDEASTLAND

A'A'A' AAA

B'B'B'

C'C'C'

BBB

CCC

111

1'1'1'

2'2'2'

3'3'3'

333

222666666

717171

656565646464 636363

737373

707070696969

686868

676767747474

116116116

118118118

117117117

115115115

119119119

103103103

102102102

101101101

100100100999999 989898

979797 969696

959595949494 939393 929292

313131303030 363636

292929

282828

272727

262626343434

252525

242424232323

222222

525252

515151505050 494949

484848 474747464646

454545

444444434343

999555

666111222

333

444121212

171717 161616 151515 141414 131313

626262

909090

888888 898989

878787868686

858585

Well number and location, and dip cross section in this studyWell number and location, and dip cross section in this studyWell number and location, and dip cross section in this study

Well number and location, and strike cross section in this studyWell number and location, and strike cross section in this studyWell number and location, and strike cross section in this study

Well location and cross section of Brown et al. (1987).Well location and cross section of Brown et al. (1987).Well location and cross section of Brown et al. (1987).

1'1'1' 111636363

A'A'A' AAA222222

EASTLANDEASTLANDEASTLANDCounty line and nameCounty line and nameCounty line and name

(A)(A)(A)

(B)(B)(B)

OU

AC

HIT

A T

HR

US

TBE

LTS

OU

AC

HIT

A T

HR

US

TBE

LTS

OU

AC

HIT

A T

HR

US

T B

ELT

S

FO

RT

WO

RT

H B

AS

IN

FO

RT

WO

RT

H B

AS

IN

FO

RT

WO

RT

H B

AS

IN

MARA THONMARA THONMARATHONTHR USTBEL TSTHR USTBEL TSTHRUST BELTS

MIDLANDBASIN

LLANOUPLIFT

RED RIVER UPLIFT

NEW MEXICOTEXAS

OKLAHOMATEXAS Modified from Brown et al. (1990)

100˚ W103˚ W

33˚ N

30˚ N

36˚ N

97˚ W

EASTER

N SHEL

FEAS

TERN S

HELF

BBBEEE

NNNDDD

AAARRR

CCCHHH

D'D'D' DDD

99˚40' W 99˚00' W

33˚50'N

33˚30'N

33˚20'N

33˚40'N

Honaker 77

WICHITA MOUNTAINS

EA

ST

ER

N S

HE

LF

UNITED STATES

MEXICO

Figure 1. (A) Tectonic elements of north-central Texas and location of study area (hachured). (B) Location of wells and cross sections in thisstudy, including those of Brown et al. (1987).

CYCLIC SEDIMENTATION, CISCO GROUP, NORTH-CENTRAL TEXAS

Geological Society of America Bulletin, October 1998 1335

parts to identify major stratigraphic intervals.Subsequent cycle delineation and correlationwere conducted within this framework.

Lithofacies were interpreted mainly on gamma-ray and resistivity logs. Typical signatures of vari-ous lithofacies and inferred depositional environ-ments were established by log calibration of

~1500 m of Cisco Group core (Fig. 4). Lithofaciesinclude nonmarine paleosol, fluvial, delta-plain,and marginal marine siliciclastic rocks, shallow-marine siliciclastic and carbonate rocks, and basin-slope shales. Log patterns established by Brown et al. (1987) and other workers (e.g., Dresser Atlas,1974; Schlumberger, 1987; Serra, 1985, 1986) andcharacteristics of 188 Cisco cycles in outcrop(Yang, 1995) also serve as important guidelinesfor log interpretations.

CYCLE CHARACTERISTICS

Observed and interpreted characteristics ofmore than 1000 cycles include component litho-facies, type, magnitude, thickness, order, and in-tracycle sediment-supply variations. Correlationof these characteristics along depositional dipand strike establishes the three-dimensional cyclearchitecture and trends of cyclic sedimentation tobe used to analyze the roles of autogenic and al-logenic processes in cyclic sedimentation.

Component Lithofacies

A composite Cisco cycle has a sharp base.Early-transgressive, thin carbonaceous shalesand lignites are overlain by sandstones and shalesthat become increasingly calcareous. Some cleanand porous sandstones coarsen upward, suggest-ing that they were well washed and reworked intransgressive, upper shoreface to barrier bar en-vironments (Fig. 5, A and C). Limestones overliethese siliciclastic rocks, many of which consist ofa lower transgressive limestone and an upper re-gressive limestone separated by a thin shale orshaly carbonate (Fig. 5, B and D). The shale isequivalent to the maximum-transgressive coreshale of midcontinent “Kansas-type” cyclothemsand separates the transgressive interval from theregressive interval of a cycle (Heckel, 1977).

Regressive calcareous shales overlie lime-stones and commonly contain thin to moderatelythick, upward-fining sandstones. The sandstonesare probably lower shoreface or shelf sand ridgedeposits. They are overlain by thin to thick,shoreface and prodeltaic shales (Fig. 5, B and D).They become increasingly sandy, grade into up-ward-coarsening and upward-thickening, delta-front sandstones and shales, and finally, massivechannel-mouth-bar sandstones.

Some channel-mouth-bar sandstones are mul-tistoried. They are overlain by delta-plain shalesintercalated with sandstones and, in some cases,by thick distributary channel sandstones (Fig.5A). Fluvial sandstones and shales of levee,crevasse-splay, and flood-plain environments oc-cupy the upper regressive interval. Thick, up-ward-fining and upward-thinning, or massive,fluvial channel sandstones are common; some aremultistoried, and some are in sharp contact with

underlying limestones (Fig. 5C). Calcareous andsandy shales, which have irregular log patternsand may contain thin, very calcareous layers inthe uppermost regressive interval, are calcareouspaleosols formed on fluvial and delta-plain sedi-ments (Fig. 5D). It should be emphasized thatmany cycles consist of only a subset of the litho-facies in a composite cycle.

Cycle Type, Magnitude, Thickness, Duration, and Order

A couplet of upward-deepening and upward-shallowing trends of depositional environmentsdefines a transgressive-regressive cycle (Figs. 4and 5). Principal component lithofacies, i.e.,transgressive and regressive limestones, andmaximum-transgressive core shale, define fivecycle types, as defined from outcrops (Yang,1996) (Fig. 6A). Type I cycles consist of all threeprincipal lithofacies, type II cycles do not containcore shale, type III cycles do not contain trans-gressive limestone, type IV cycles do not containregressive limestone, and type V cycles do notcontain any limestones (Figs. 5 and 6A). TypesIII, IV, and V may or may not contain core shale;they are not as common in the subsurface as inoutcrop, partly owing to uncertain log interpreta-tion of core shales (Fig. 5). Cycle type signifieslithofacies variations among cycles (Yancey,1991; Yang, 1996).

Cycle magnitude is defined as the maximumfacies shift over the transgressive or regressiveinterval of a cycle (Fig. 6B; Yang, 1996). It indi-cates the maximum extent of shoreline move-ment over a cycle. Type I cycles generally havehigh magnitude, whereas type V cycles have lowmagnitude.

Cycle thickness ranges from less than 3 to 90 m. It approximates, to the first order, cycle du-ration. Assuming steady rates of shoreline trans-gression and regression and a constant sedimen-tation rate for all lithofacies, cycles of highmagnitude and large thickness should have longduration. Cycle thickness is also a first-order ap-proximation of accommodation space.

Three orders of cyclicity—major, intermediate,and minor—are observed. Cycle magnitude,thickness, and stacking pattern determine the or-der of a cycle. Cycle duration is not used explic-itly to define cycle orders because it is a functionof cycle magnitude and thickness. Within a majorcycle, however, minor cycles have smaller dura-tion than intermediate cycles, which have smallerduration than the major cycle. Orders of cyclicityprobably represent the degree of perturbation ofprocesses driving transgression and regression.

The relationship among minor, intermediate,and major cycles is one of superimposition interms of thickness and magnitude (Fig. 6B). Thisrelationship, however, is commonly complicated

Figure 2. Operational nomenclature, Vir-gilian and Wolfcampian Series in the subsur-face Eastern shelf. No vertical scale is intended.Depositional sequences bounded by uncon-formities (wavy lines) of Brown et al. (1990) areshown. Modified from Brown et al. (1990).

YANG ET AL.


by lateral variations in cycle magnitude, thick-ness, and cycle absence. Minor cycles are thin,only several meters thick. Their facies shifts oc-cur mainly in marine environments and are gen-erally small (Figs. 6B and 7). Intermediate cyclesare several to tens of meters thick, with large fa-cies shifts from marginal marine or nonmarine tomarine (Figs. 6B and 7).

An intermediate cycle commonly contains twoto three minor cycles. For example, the Ivan no. 1and no. 2 minor cycles and the Breckenridge no. 1and no. 2 minor cycles in well 27 constitute two in-termediate cycles, respectively (Fig. 7). The minorcycle located in the regressive interval of an inter-mediate cycle, such as the Gunsight no. 2 cycle inwell 30, has a large regressive facies shift or mag-nitude, which is equal to that of the intermediatecycle itself (Figs. 6B and 7). The minor cycles inan intermediate cycle may be absent in the adja-cent well, where only one minor cycle remains todefine the intermediate cycle (Fig. 7). This inter-mediate cycle can be regarded as consisting of nominor cycles or only one minor cycle. Here the lat-ter view is taken. For example, the intermediatecycle containing Bunger no. 1 and no. 2 minor cy-cles in well 27 contains only the Bunger no. 1 cycle in well 24 (Fig. 7).

Major cycles are tens to hundreds of metersthick with large facies shifts. They contain one ormore intermediate cycles. Some contain only oneintermediate cycle that itself contains only oneminor cycle, such as the Ivan and Blach Ranchmajor cycles (Fig. 7). On the platform, major cy-cles usually conform with the third-order deposi-tional sequences of Brown et al. (1990).

In summary, minor cycles are the building

blocks of intermediate and major cycles. Thethree orders of cycles form a stratigraphic hierar-chy. Complication of cycle ordering by variationsin type, magnitude, thickness, and cycle absencecan be resolved through cycle correlation (Fig. 7).

Trend of Siliciclastic Sediment-Supply Variations

Many Cisco cycles can be divided into two in-tervals, sand rich and sand poor, as an alternativesubdivision to the more interpretive transgressiveand regressive intervals of a cycle. Commonly,but not always, a significant increase of coarsesiliciclastic supply at a depositional site indicatesthe approach of a clastic sediment source to thatsite, whereas a decrease indicates retreat of thesource. This subdivision scheme uses deposi-tional site as the reference point, emphasizingprocesses operating at the site. It facilitates dis-cussion of the timing, type, and amount of silici-clastic supply in cyclic sedimentation in mixedcarbonate and siliciclastic environments.

The sand-poor interval occupies the lower partof a cycle (Figs. 4, 5, and 6). It consists of trans-gressive siliciclastic and carbonate rocks, as wellas regressive carbonate rocks and marine to mar-ginal marine siliciclastic rocks. The siliciclasticsare mostly shale and siltstone; sandstones are thinand shaly. An overall upward-fining log patternis typical of this interval. The presence of car-bonate rocks and a small amount of sandstone in-dicates minimal siliciclastic supply at the site,and the upward-fining pattern indicates a dimin-ishing supply of coarse sand over this interval,especially after the deposition of regressive lime-

stones (Figs. 4, 5, and 6B). Sandstones of marineand marginal marine origin are probably derivedmainly from coarse-grained sediments of the un-derlying cycle and were transported by longshorecurrents to the depositional site.

The sand-rich interval occupies the upper partof a cycle. It consists of late and maximum re-gressive siliciclastics of deltaic and fluvial origin.An overall upward-coarsening and upward-thickening log pattern is typical. Sands weremainly land derived and transported to the depo-sitional site by rivers. The boundaries betweensand-poor and sand-rich intervals are commonlygradational and generally placed at the base ofdelta-front sandstones. Gradational boundariessuggest a gradual approach of a clastic sedimentsource toward the site.

The sand-rich interval is thinner than the re-gressive interval (Figs. 4, 5, and 6B). Transgres-sion is defined as landward shoreline movement,and regression is defined as seaward shorelinemovement. This relationship suggests that signif-icant siliciclastic supply to a depositional site oc-curred sometime after shoreline regression. Thisagrees with the outcrop observation that sedi-ment supply at a site lags sediment yield in thesource area (Yang, 1996). Moreover, the sand-rich interval is absent in some cycles, especiallyminor cycles, indicating that landward sedimentsupply is insignificant and/or that the site is dis-tant from major depositional loci during late andmaximum regression.

Most intermediate and major cycles contain asand-rich interval because they have a relativelylong period of regression for significant progra-dation of coarse sediments to the depositional

Figure 3. Dip stratigraphiccross section of the Cisco Groupshowing the progradational andaggradational configurations ofthe Eastern shelf and depositionalsystems of the Cisco Group. Lo-cation is shown in Figure 1A.Horizontal scale is approximate.Highly simplified from sectionD–D′ of Brown et al. (1990).



site. Regional correlation of the lithofacies andthickness of sand-rich intervals displays thechanging pattern of sediment supply on the East-ern shelf in cycle development.

CYCLE CORRELATION

Cycle Correlation on Stratigraphic Dip Cross Sections (Figs. 8, 9, and 10)

Topography.The top of the Home Creek inter-val outlines the pre-Cisco depositional topography(Fig. 11). The shelf consists of platform, shelf

edge, and basin slope. The platform dips gently tothe west and has broad highs and lows. A promi-nent high at well 29 on A–A′ is caused by a shelf-edge carbonate buildup underneath the HomeCreek interval; the high at the east end of B–B′corresponds to an underlying carbonate bank(Brown et al., 1987; Figs. 3 and 11). The shelfedge dips abruptly basinward, and two shelfbreaks are present. The area between the breaks ismonoclinal in B–B′ and C–C′, and is a deep troughin A–A′ (Fig. 11). The steep basin slope shallowsto the west. A complete basin-to-slope topographyis seen in the lower Cisco Group (Fig. 11).

The shelf edge migrated westward in steps(Fig. 11). It aggraded during carbonate-rich dep-osition, generating a large basin-slope accom-modation space. Shelf-edge progradation fol-lowed as deltaic and fluvial sedimentsprogressively filled the space. Apart from theoverall shelf edge migration, some major pre-Cisco topographic features, such as the two shelfbreaks on A–A′, persisted during most of theCisco Group time (Fig. 8).

The upper Cisco Group topography could berelated to the precedent topography or could havebeen caused by regional structural upwarping or

Marginal marine

Lagoonal storm deposits

Restricted lagoon

Algal mound

Landward lagoon

Semi-open lagoon

Open marine

Marginal marine

Open marine

Shoreface

Middle-upper shoreface

Delta plain

Delta front, mouth bar

Semi-open lagoonShallow open marine

Channel mouth bar

Marginal marine

Shallow open marine

Upper shoreface

Strandline

Beach barrier

Beach barrier

sandf m c g

0

30

Gamma ray Neutron Depositionalenvironments

Upper shoreface

Radiation increase

Limestone

Arenaceous limestone

Argillaceous limestone

Cross-laminated sandstone

Planar-laminated sandstone

Shale

Sandy shale

Carbonaceous shale

Calcareous paleosol

Lignite

Cycle boundary

Boundary betweensand-rich and -poorintervals

Bioturbated sandstone

Convolution structure

Calcareous mudstone

Tabular or trough cross-bedded sandstone

Thin, interbeddedsandstone and shale

L

M

M

M

M

L

L

L

M

L

Regressive interval

Transgressive interval

Sand-rich interval

Sand-poor interval

L Land-derived sand

M Marine-sourced sand

Increasingwater depth

QAc1973c

ft

Sup

ply

tren

d

Sho

relin

em

ovem

ent

San

d so

urce

Figure 4. Lithology, deposi-tional environments, and wire-line logs of Honaker 77 core inWichita County (Fig. 1A) as anexample of log calibration. Cy-cles and their subdivisions aredelineated from lithofacies andtrends of environment and wa-ter-depth changes. All cycles aretype II according to the classifi-cation scheme in Figure 6A.

YANG ET AL.


2200

2100

2200

2100

0 150GR (API)2 20 200

Resistivity (ohm-m) Lithofacies and depositionalenvironments

Limestone, normal shallow marine

Shale, shorefaceShale and sandstone, delta plain

Sandstone, distributary channel

Sandstone, channel mouth bar anddelta front

Shale, interdistributary

Sandy shale, prodelta to distaldelta front

Regressive limestone

Shale and sandstone, uppershoreface to marginal marineShale, interdistributary/strandline

0 150GR (API) Lithofacies and depositionalenvironments2 20 200

Resistivity (ohm-m)

SP (mv)

SP (mv)

Well 119 — A type I cycle

Well 115 — A type I cycle

Core shale

Shale, shoreface

(A)

(B)

Limestone, normal shallow marineShalestone and shale, strandline andupper shoreface

Shale and sandy shale, delta plain

Sale and sandstone, marginalmarine

Sandstone, channel mouth barand delta front

Regressive limestone

Shale, marginal to shallow marine

Shale and sandstone, delta plain

Core shale

Shale, shoreface

Sandy shale, prodelta and distaldelta front

Transgressivelimestone


QAc1974c

Depth(ft)

Depth(ft)

Sho

relin

em

ovem

ent

Sup

ply

tren

d

Incr

easi

ngw

ater

dep

th

Sho

relin

em

ovem

ent

Sup

ply

tren

d

Incr

easi

ngw

ater

dep

th

Figure 5. Typical well-log signatures of various lithofacies. Cycles and their subdivisions are delineated from trends of environment and water-depth changes. See text and Figure 6A for cycle type classification. See Figure 4 for more lithologic keys.

differential compaction. Two broad highs in themiddle parts of B–B′ and C–C′ are ~16 to 19 kmwide and 90 to 120 m thick (Figs. 9 and 10). Thehigh in B–B′ is located at or near a major deposi-tional locus with abundant coarse siliciclastic de-

posits. Differential compaction between sand-stones in the high and the finer grained rocks inadjacent lows created the relief. It was accentu-ated by continuous deposition of coarse silici-clastics over the high until a later time. A high

similar to those in B–B′ and C–C′ is not presentin A–A′, because coarse siliciclastics were fillingthe deep pre-Cisco trough (Fig. 8).

Preexisting and evolving topography definesthe general framework of cycle architecture,


within which basin-wide tectonics and sedimenta-tion interact with local topography and deposi-tional dynamics. This will be demonstrated in thefollowing sections.

Cycle Abundance and Continuity.Cycle

abundance and continuity vary systematically ondip sections. The number of minor cycles in-creases in the downdip direction except in thebasin slope. For example, 32 to 36 minor cyclesare found in updip wells, but 44 to 46 cycles are

found in downdip wells in A–A′ (Fig. 8). Thereare a total of 68 minor cycles in the Cisco Group,whereas the maximum number of minor cycles ina single well is 46. These discrepancies are causedby common cycle absence on the Eastern shelf.

1800

1700

550

2 20 2000 150GR (API)

0 200SP (mv)

Trans. limestone, normal shallowmarine

Sandy shale, flood plain

Shale and sandstone, marginalmarine and delta plain

Sandstone, channel mouth bar anddelta front

Shale and sandstone, prodelta anddistal delta front

Shale, marginal marine

Sandstone, fluvialchannel

Shale and sandstone,lower shoreface

Sandstone and shale, strandline to upper shoreface

Resistivity (ohm-m)

Shale and sandstone,lower shoreface

Shale, marginal marine

Sandy shale, fluvial and delta plain

Shaly sandstone, interdistributary anddistributary channel

Sandstone, channel mouth barand delta front

Shale and sandstone, distal delta front

Shale, prodelta

Shale, marginal or deep(?) marineLimestone, shallow marine

0 150GR (API)

SP (mv) 2 20 200

Resistivity (ohm-m)

Well 102 — A type II cycle*

Well 22 — A type II cycle*

Calcareous paleosol (?)

Gamma ray log (GP)

Self potential log (SP)

Coarse sandstone

Fine sandstone

Calcareous sandstone

Shallow focusing log (SFL)

Medium induction log (ILM)

Deep induction log (ILD)

Trans. and reg.limestone


Core shale

* Type IV if core shale is interpreted.

Lithofacies and depositionalenvironments

Lithofacies and depositionalenvironments

Core shale

* Type IV if core shale is interpreted.

QAc1975c

(C)

Reg. sandstone,shoreface

Trans. & reg. limestone Trans. limestone

(D)

Core shale

Shale, shorefaceTrans. limestone

Reg. sandstone

650

Depth(ft)

Depth(ft)

Sho

relin

e m

ovem

ent

Sup

ply

tren

d

Incr

easi

ngw

ater

dep

th

Sho

relin

e m

ovem

ent

Sup

ply

tren

d

Incr

easi

ngw

ater

dep

th

Figure 5. (Continued).



Three processes, amalgamation, siliciclasticsediment suppression, and erosion, caused cycleabsence. Amalgamation occurs where two,rarely three, minor cycles merge into one in anadjacent well, where regressive siliciclastics ofthe subjacent cycle and transgressive siliciclas-tics of the superjacent cycle are absent (Fig.12A). It occurs commonly in the lower plat-form, where cycles merge mostly in the updipdirection. Amalgamation of intermediate cyclesis not observed. In some cases, the resolution ofwireline logs prohibits pinpointing the exact lo-cation of amalgamation.

Cycles amalgamate toward either topographichighs or lows. In the case of amalgamation to-ward topographic highs, the absence of regres-sive siliciclastics of the lower cycle and trans-gressive siliciclastics of the upper cycle wascaused by sediment bypassing, erosion, and/ornondeposition in the updip location. These strataare represented by a surface separating the amal-gamated limestones of the lower and upper cy-cles (Fig. 12A). In most cases, the upper cyclehas a smaller magnitude than the lower one andthus represents a transgressive-regressive eventof smaller extent. In the case of amalgamation to-ward topographic lows, the absence of regressiveand transgressive siliciclastics in the downdip lo-cation was most likely caused by sediment star-vation or bypassing.

Many minor and intermediate cycles pinch outinto siliciclastics in the updip direction or towardtopographic lows on the platform (Fig. 12B).Pinch-outs are gradational or abrupt. Gradationalcycle absence was caused by suppression of car-bonate deposition in nearshore siliciclastics-richenvironments due to siliciclastic contamination inthe form of reduced light penetration, nutritionalpoisoning of calcite-secreting organisms, unsuit-able substrate, or carbonate dilution (Mount,1984). As a result, limestones in a cycle, the prin-cipal lithofacies in cycle delineation, were not de-posited and the cycle could not be defined. Thegradual change from marine carbonate environ-ments to marine and nonmarine siliciclastic envi-ronments in the updip direction suggests that gra-dational cycle absence is controlled by the extentof shoreline transgression.

Abrupt cycle absence was caused by fluvialchannel cutting into underlying limestones fortens of meters, as is commonly observed in out-crops (Fig. 12B; Yang, 1995). It is not an indica-tion of the extent of marine transgression. Manyminor cycles are absent before reaching the up-per platform. As a result, the absence of interme-diate cycles composed of one minor cycle pre-dominates in the middle and upper platform.

Cycle absence results in cycle discontinuity.Most minor cycles persist more than 16 km.Many intermediate cycles persist more than 48km and extend farther landward than minor cy-

Cycle typesa

Fluvial and delta-plainsandstone and shale

Marginal marine and deltaicsandstone and shale

Shallow marine carbonates

b

Shallow marine sandstoneand shale

Fairly deep to deepmarine core shale

Shallow to fairly deepmarine carbonates

Shallow to marginal marinesandstone and shale

Marginal marine andnonmarine sandstone andshale

Componentlithofacies

b -- Principal lithofacies used to define cycle types.

a -- "X", the lithofacies which must be present in a certain type of cycles; "x", commonly present; "*", rarely present; "blank", not present.

PhaseI II III IV V

x

x

x

X

x

X

x

x

x

x

X

X

* *

x

x

x

X

*

x

x

x

x

x

*

X

x

*

x

x

X

*

X

xR

egre

ssio

nT

rans

gres

sion

Ear

ly

Lat

e

M

axim

umE

arly

Late

Max

imum

II

II

No. 1

INo.1

Cycletype

No. 1

Cycle order

FL SL S D

Depositional environment

Limestone

Sandstone

Shale~10

Cycle boundary

Boundary separating sand-rich from sand-poor intervals

VSDP

Regressive faciesshift/magnitude

QAc1976c

0ft

No. 3

No. 2

No. 2

Transgressive faciesshift/magnitude

Lith

olog

y

Sho

relin

em

ovem

ent

Sup

ply

tren

d

Min

or

Inte

r-m

edia

te

Maj

or

Figure 6. (A) Cycle type classification by principal component lithofacies. (B) Illustration ofcycle characteristics including component lithofacies, magnitude, type, transgressive and re-gressive intervals, sand-rich and sand-poor intervals, and order. Cycles are delineated fromtrends of depositional environment and/or water-depth changes, which are displayed as a faciescurve on a grid with the horizontal axis as environments and vertical axis as thickness. Envi-ronmental keys: FL—fluvial; DP—delta plain; SL—strandline marginal marine; VS—veryshallow marine; S—shallow marine; FD—fairly deep marine around normal wave base; D—deep marine about storm wave base. See Figures 4 and 5 for lithologic patterns and Figure 4 forpatterns of supply trend and shoreline movement.

YANG ET AL.


24

27

30

65

0 5 mi

0 8 km

100

0

30

FL

ft m0

Cycle hierarchy

West

Brecken-ridge

Inte

rm.

cycl

es

Maj

or

cycl

es

Min

or

cycl

es

BlachRanch

Ivan

Bunger

No. 3

No. 1No. 1

No. 2

No. 3

No. 7No. 7

No. 1

No. 1

No. 2

No. 1No. 1

No. 2

No. 1 No. 1

No. 1 No. 1No. 2No. 3No. 3

VS

D Crystal Falls No. 1

Breckenrige No. 2

Breckenrige No. 1

Blach Ranch

Ivan No. 1

Bunger No. 7

Bunger No. 3

Bunger No. 2

Bunger No. 1

Breckenrige No. 3

Ivan No. 2

FL

VS

D

FL

VS

D

FL

VS

D

East

QAc1978c

Gunsight

Gunsight No. 1Gunsight No. 2

Vertical exaggeration x 340

Breckenridge No. 3

Crystal Falls No. 1

Crystal Falls No. 5

Figure 7. Part of cross section A–A′ illustrating cycle ordering and continuity. Minor, intermediate, and major cycles are indicated by thin,medium, and thick correlation lines. The left panel illustrates the hierarchical relation among the three orders of cycles. Continuity of minor andintermediate cycles varies. Some cycles identified in other sections, such as Bunger numbers 4, 5, and 6 cycles, are not present in this part of A–A ′,indicating common cycle absence. See Figure 6 for environmental keys.

Figure 8. Highly simplified dipstratigraphic cross section A–A′of Cisco Group, Eastern shelf,north-central Texas.


cles, suggesting small extent and magnitude ofminor transgressive-regressive events (Figs. 7 and11). However, where the platform is flat, as in theupper Cisco Group, minor cycles persist over along distance (Figs. 8, 9, and 10). Laterally per-sistent cycles represent regional transgressive-re-gressive events controlled by regional processes.

Nonpersistent minor and intermediate cyclesoccur in three modes. First, some cycles occur

fragmentarily from place to place, such as manyFlippen cycles (Figs. 7 and 11). They are regionalbut nonpersistent. Second, some cycles are onlypresent in the lower platform (Fig. 11). They mayextend farther downdip out of the study area andthus could represent regional transgressive-regres-sive events that did not reach the upper platform.Lastly, very few cycles are of only local extent.

Cycle Type and Magnitude.Cycle type is de-

termined by water depth, lithofacies, and deposi-tional dynamics, and cycle magnitude is deter-mined by the change of water depth over a cycleinterval. Thus, regional trends of cycle type andmagnitude indicate changes of water depth andlithofacies, modified by depositional dynamics.

Minor cycles exhibit two regional trends. First,~20 regional cycles change from type II to type Iand increase in magnitude in the downdip direc-

Figure 9. Highly simplified dipstratigraphic cross section B–B′of Cisco Group, Eastern shelf,north-central Texas.

Figure 10. Highly simplifieddip stratigraphic cross sectionC–C′ of Cisco Group, Easternshelf, north-central Texas.

YANG ET AL.



tion. Many minor cycles in the lower platformalso show this trend.

The type of intermediate cycles is not definedbecause they are composed of several minor cy-cles. A minor cycle in the lower platform becomesthe sole minor cycle in an intermediate cycle in theupper platform, where other minor cycles in thesame intermediate cycle are absent. Therefore, inthe upper platform, the magnitude of this minorcycle is that of the intermediate cycle. The magni-tude of the intermediate cycle in the lower plat-form should be higher than that of the minor cycle,or be equal to that of the minor cycle if this minorcycle has the highest magnitude among all the mi-nor cycles in the intermediate cycle. Thus, themagnitude of the intermediate cycles containingthe 20 regional minor cycles also increases in thedowndip direction. The downdip change from

type II to type I of minor cycles and increase inmagnitude of minor and intermediate cycles indi-cate a larger change of water depth over a cycle in-terval, and more core shale deposition in the lowerplatform than in the upper platform. They wereprobably caused by the downdip increase in ac-commodation space and decrease in land-derivedsiliciclastic sediment supply.

The second trend is that ~10 regional minor cy-cles have consistent type and magnitude acrossthe shelf. This trend, as reasoned above, suggeststhat the magnitude of intermediate cycles contain-ing these minor cycles increases in the downdipdirection. The 10 cycles are either type I or type II.Type I cycles indicate that shelf deepening waslarge and fast enough to prohibit large landwardsiliciclastic influx and core shales were depositedon the upper platform. Type II cycles represent

fast, low-magnitude shelf deepening, duringwhich a large amount of siliciclastic material wasyet to be delivered to the depositional site to sup-press carbonate production.

About 10 regional minor cycles, however, donot show any systematic variation. They repre-sent low-magnitude minor transgressive events,during which local topography and depositionaldynamics largely controlled water depth and dis-tribution of carbonate-suppressing, land-derivedsiliciclastics. The other possible scenario is that alarge and irregular distribution of siliciclasticsupply, combined with irregular local topogra-phy, could have caused these variations regard-less of the magnitude of marine transgressions.The magnitude of intermediate cycles containingthese minor cycles still increases in the downdipdirection.

0 5 mi

0 8 km

20050

0ft

0m

Gonzales

Bunger

GunsightIvan

Crystal Falls

Flippen

98103116117118

29

25

Major cycle boundary

Minor and intermediate cycle boundaryShelf-edge migration pathand direction

A'

A

B

B'

C'

C

West East

Home Creek

Home Creek

Shelf break

0˚0.1˚

0.5˚1.0˚

QAc197

Cycle-amalgamation line

Breckenridge Blach Ranch


Salem School

Home Creek

Figure 11. Pre–Cisco Group depositional topography outlined by the top of Home Creek interval in dip cross sections. Stepwise progradationand aggradation of shelf edge are shown in C–C′. Stratigraphic datum is the top of Cisco Group. A datum in the lower Cisco Group only slightlychanges the pre-Cisco topography.

YANG ET AL.


In a few cases, a cycle on the platform is oftype I in local lows and type II on highs, or viceversa. Any subtle imbalance in the interplayamong accommodation space, siliciclastic sup-ply, local topography, and depositional dynamicscould have caused these contrasting cases.

Cycle Thickness.Thickness is the most vari-able attribute of Cisco cycles. It is related to shelfposition, local topography, component lithofaciesand their compactability, and cycle order; these inturn are controlled by pre-Cisco topography,shelf subsidence, depositional dynamics, sedi-ment supply, sea-level changes, and regional cli-mate. To differentiate these controls from thick-ness trends is difficult (Yang, 1996). Here wespeculate on the implications of generalizedthickness trends on these controls.

Major and intermediate cycles contain one ormore minor cycles and thus are thicker. Minorcycles are less than 3 m to more than 15 m thick,commonly 6 to 9 m. The minor cycle in the re-gressive interval of an intermediate cycle usuallyhas thick regressive siliciclastic deposits and istherefore thicker than those in the transgressiveinterval of the intermediate cycle (Figs. 7 and12B). Many minor cycles have fairly consistentthickness for more than 16 km. The less-consist-ent cycles commonly thicken toward topographiclows or in the downdip direction (Figs. 8, 9, and10). Cycle thickness approximates the accommo-dation space if it is filled by sediments (Yang,1996). Consistent thickness therefore implies thataccommodation space is fairly uniform on a low-relief platform and space variations are con-trolled by local topography. A minor cycle thick-ens abruptly when the overlying cycles are absent(Figs. 7 and 12B). The abrupt thickening does notprovide any information on variation of accom-modation space.

Intermediate cycles are ~3 m to less than 60 mthick, commonly 15 to 30 m. They generallythicken in the downdip direction and toward lo-cal lows on the platform (Figs. 8, 9, and 10). Thisis because accommodation space was large in thelower platform and lows, and depositional lociwere directed downslope by gravity, depositingthick regressive siliciclastic sediments. Interme-diate cycles thin or thicken from shelf edge tobasin slope. The thickness change is largely de-termined by the availability of regressive silici-clastic sediment supply because accommodationspace is ample in the basin slope region.

Rarely, intermediate cycles thicken towardhighs. The thickening is controlled by deposi-tional dynamics and has to be explained on acase-by-case basis. For example, the thickeningof Blach Ranch no. 1 cycle in A–A′ is caused bya thick, maximum-regressive sandstone filling afluvial channel that developed on the underlyinghigh (Fig. 13A). A fluvial channel situated on atopographic high is geomorphologically puz-

zling. A possible scenario is that the topographyat the time of channel initiation may have beenfairly flat. Once the initial small channel occu-pied or migrated into the location, it could deepenand stabilize during subsequent relative sea-levelfall. In addition, this thickening was accentuatedby later differential compaction between thesandstone and adjacent interfluve shale. Thethickening of Gouldbusk no. 1 cycle in B–B′ iscaused by a thick carbonate buildup on the un-derlying high during a maximum transgression(Fig. 13B). The submerged high was devoid ofsiliciclastic contamination, well lit, and well oxy-genated, and had a suitable substrate for organ-isms to colonize.

The thickness of major cycles ranges widelyfrom less than 15 to 60 m, increasing in thedowndip direction on the platform and decreas-ing on the shelf edge and basin slope. It is con-trolled by sea-level changes, sediment supply,and local and regional topography (Brown et al.,1987, 1990; Galloway, 1971).

Sand-rich Intervals. Sand-rich intervals varygreatly in thickness and lithofacies at a localscale, although systematic variations exist. Fiveintermediate cycles in the Gunsight to Brecken-ridge interval in A–A′ are used to demonstratethese variations (Fig. 14).

Sand-rich intervals are 0 to ~30 m thick. Theythicken in the upper and lower platform and are

��

��

��

��

��

��

��

��

��

��Cycle

amalgamation

Cycle absence

due to erosion

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

Downdip Updip

Transgression Transgression

Minorcycle 2

Minorcycle 1

Cycle

amalgamation

Am

alga

mat

ion

surf

ace

ofse

dim

ent b

ypas

sing

, ero

sion

,or

non

depo

sitio

n

Am

alga

mat

ion

surf

ace

ofse

dim

ent s

tarv

atio

n an

dno

ndep

ositi

on

Transgression

Transgression

Cycle

amalgamation

Minorcycle 2

Minorcycle 1

An

inte

rmed

iate

cyc

le c

ompo

sed

of o

ne m

inor

cycl

e (M

inor

cyc

le 1

)

TransgressionTransgressionTransgression

Min

or c

ycle

1M

inor

cyc

le 3

An

inte

rmed

iate

cyc

le c

ompo

sed

of tw

o m

inor

cyc

les

An

inte

rmed

iate

cyc

le c

ompo

sed

of th

ree

min

or c

ycle

s

Minorcycle 3

Minorcycle 2

Minorcycle 1

Well 3 Well 2 Well 1

Upper platformLower platform Limestone Sandstone Shale

(A)

(B)

QAc1980c

0

~10

ft

Figure 12. (A) Cycle amalga-mation in updip and downdip di-rections and associated processes.(B) Cycle absence and abruptthickening. An intermediate cyclecomposed of three minor cycles inwell 3 contains only one minor cy-cle in well 1, owing to updip amal-gamation and fluvial erosion. Mi-nor cycle 1 thickens from well 3 towell 2 because of amalgamation,and from well 2 to well 1 becauseof fluvial erosion.


25 2426

0

100

0

30

Topographic high

Topographic lowTopographic low

Stabilized fluvial channel

0 1 mi

0 1 km

(A)

(B)

Coalesced transgressive-regressive limestone in type II cycles

Gamma ray log, increasingto the right

Q A c 1 9 8 1 c

mft

Sedwick No.1

6867 52

51

50

Gouldbusk No. 1

Broad topographic high0

0

2 mi

3 kmSelectively illustrated sand-stone

Undifferentiated shale andsandstone

Limestone and core shale in type 1 cycles

Breckenridge No. 1

Blach Ranch No. 1

Carbonate buildup on a broad topographic high

Figure 13. Examples of cyclethickening on highs: (A) thicken-ing due to fluvial channel fillingand stabilization; (B) thickeningdue to a carbonate buildup on theunderlying high.

��

��

��

��

��

��

��

Intermediate boundary cycle

WEST

0

200

0

50

��Sand-rich interval of an intermediate cycle

Sand-poor interval of an intermediate cycle

6466 65 25 24 23 2230 29 28 27 26

EAST456

Crystal Falls interval

Bk No. 2

Bl No. 2

Bk No. 2

Iv No. 2

Gs No. 2

Bk No. 3

Bk No. 1

Bl No. 1

Iv No. 1

Gs No. 1

5 mi

0 8 km

0


Bk - Breckenridge

Bl - Blach Ranch

Gs - Gunsight

Iv - Ivan

Boundary between sand-rich and -poor intervals

LowerMiddle

zone

Minor cycle boundary

Boundary separating depocenters from the bypassing zone

QAc1982c

ft m

Cycle-amalgamation line

platformbypassing

Upper

platform

Bk No. 3

depocenter(fluvial dominated)

(deltaic dominated)depocenter

platform

Figure 14. Part of simplified cross section A–A′ illustrating thickness and lithofacies variations of sand-rich intervals of intermediate cycles.See text for discussion.


��

��

��

��

UplandPiedmontShelfedge

Maximumregression

Maximumtransgression

Shoreline positions

Sealevel

Gradedriver

profiles

Lower platformdepocenter

Upper platformdepocenter

Sedimentbypassing

zone

Shelf profile

QAc1983c

Slope

(3) (2) (1)

Figure 15. Model explaining formation ofdepocenters and bypassing zones on the East-ern shelf, due to changing river gradient andsubaerial accommodation space capped by thegraded river profile. When shoreline regressesfrom the maximum-transgressive position toposition (1) in the upper platform, the subaer-ial accommodation space increases to form adominantly fluvial depocenter. From (1) to (2)in the middle platform, the river profile is be-low the shelf surface, creating negative accom-modation space (hachured area) prone to ero-sion and sediment bypass. From (2) to (3) inthe lower platform, the subaerial accommoda-tion space increases to form a dominantlydeltaic depocenter. When shoreline descendson the slope, the subaerial accommodationspace is negative over the entire shelf. No scaleis intended. Modified from Yang (1996).

Figure 16. Highly simplifiedstrike structural cross section1–1′ of Cisco Group, Easternshelf, north-central Texas.


thicker in the upper platform. They thin in the rel-atively steep middle platform and toward theshelf edge (Fig. 14). The thickness variations out-line a fluvial and a deltaic depocenter in the up-per and lower platform, respectively, and a zonedominated by coarse sediment bypass in the mid-dle platform during marine regression (Fig. 14).The sand-rich intervals are upward-shallowing,marine deltaic to fluvial facies successions (Figs.

4 and 5). Fluvial deposits, especially channelsandstones, are more common in the upper plat-form than in the lower platform.

Regional thickness and lithofacies variations ofthese intervals suggest diminishing, but persistent,progradation of coarse siliciclastics in the downdipdirection during late and maximum regression.Coarse sediments were first deposited in the upperplatform; remaining sediments mostly bypassed

the middle platform and were deposited in thelower platform.

As discussed previously, progradation ofcoarse sediments lags shoreline regression at agiven location (Figs. 4 and 5). This lag increasesin downdip wells, as indicated by the downdipdecrease in thickness of these intervals (Fig. 14).The increased lag was caused by delayed progra-dational infilling of coarse sediments in downdip

Figure 17. Highly simplifiedstrike structural cross section2–2′ of Cisco Group, Easternshelf, north-central Texas.


YANG ET AL.


sites or by extensive shoaling due to aggrada-tional infilling of fine sediments in downdip sitesbefore the progradational front of coarse sedi-ments arrived. The delay of coarse siliciclastic in-filling at depositional sites after shoreline regres-sion is consistent with the outcrop observationthat maximum sediment supply at a depositionalsite occurred during late regression, whereasmaximum sediment yield in the source area oc-curred during maximum transgression and earlyregression (Yang, 1996).

The pattern of shelfwide progradational infill-ing suggests regional controls on shoreline regres-sion and sediment infill. Possible controls are (1) sea-level change, which mainly controlledshoreline movement and submarine accommoda-tion space; (2) episodic platform subsidence,which controlled submarine and subaerial accom-modation space as well as shoreline movement;and/or (3) episodic sediment influx into the plat-form due to tectonic and/or climatic changes in up-land source areas. Regional sea-level changes are alikely cause of regional repetitive transgressionand regression. There is no direct evidence ofepisodic subsidence in the study area. On the con-trary, some studies have suggested tectonic stabil-ity of the Eastern shelf (e.g., Wermund and Jenkins, 1969; Brown et al., 1990). Episodic sedi-ment influx caused by upland climatic changes issuggested in the outcrop study (Yang, 1996).

Preservation of thick nonmarine and mar-ginal marine rocks and the distribution of depo-centers and bypassing zones indicate large butunevenly distributed submarine and subaerialaccommodation space on the platform duringlate and maximum regression. Shelf configura-tion and changes in base level, including ab-solute sea level and graded river profile, deter-mine the spatial and temporal variations ofsubaerial and submarine accommodation spaceon the platform (Fig. 15; Yang, 1996). In addi-tion, the thick sand-rich intervals suggest signif-icant platform subsidence, which is the ultimatecontrol on sediment preservation. In summary,the interplay among topography, siliciclasticsupply, base-level change, subsidence, and dep-ositional dynamics controlled the depositionand preservation of coarse siliciclastic rocks ofsand-rich intervals.

Cycle Correlation on Stratigraphic and Structural Strike Cross Sections (Figs. 16, 17, and 18)

Topography.The pre-Cisco Group topographyalong strike is essentially a north-dipping mono-cline (Figs. 16, 17, and 18). Its relief decreasesfrom ~150 m in the upper platform and ~60 m inthe middle platform to ~45 m in the lower plat-form. Thus, the overall pre-Cisco Group topogra-phy of the study area dips to the northwest.

Pre-Cisco Group lows were largely filled dur-ing the beginning of deposition of the CiscoGroup. Lower Cisco Group cycles thicken to thenorth owing to apparent northward regressivefluvial and deltaic progradation, suggesting a ma-jor sediment source to the south and southeast(Fig. 19). Some local features persisted or wereenhanced during deposition of the Cisco Group.For example, the high at well 90 was accentuatedby regressive fluvial channel sandstones (Fig.

19). The persistent low at well 56 of section 2–2′shallowed where it was filled by less-com-pactable, thick sandstones and deepened wherethe sandstones are thin or carbonate buildups inadjacent highs are thick (Fig. 20). The type oflithofacies, depositional dynamics, and differen-tial compaction had controlled the shallowingand deepening.

Shelf topography gradually flattened upward.The relief is much lower in the downdip section

Figure 18. Highly simplified strike stratigraphic cross section 3–3′ of Cisco Group, Easternshelf, north-central Texas.



than in the updip sections (Figs. 16, 17, and 18).As a result, topographic control on cycle abun-dance, continuity, and thickness diminishes up-ward and basinward.

Cycle Abundance and Continuity.Manyminor cycles are absent toward highs as a resultof limited accommodation space, siliciclasticsuppression of carbonate deposition, and fluvialerosion (Figs. 19, 20, and 21). Some cycles, how-ever, are absent in lows where depositional lociof coarse sediments were directed (Fig. 21).These observations suggest a combined controlof topography, accommodation space, and depo-sitional dynamics on cycle abundance.

Gradational and abrupt cycle pinch-out causedby siliciclastic suppression and erosion is com-

mon in updip sections, whereas cycle amalgama-tion is common in downdip section 3–3′, as in thedip sections. Moreover, there are more minor cy-cles in section 3–3′. For example, the Bunger in-terval has three cycles in section 1–1′, but six insection 3–3′ (Figs. 16 and 18). The lack of fluvialerosion in section 3–3′ results in increased cyclecontinuity and abundance.

Most minor cycles are continuous throughoutsection 3–3′. More than half of the minor cyclesare continuous for more than 8 km in sections1–1′ and 2–2′. Intermediate cycles are more con-tinuous than minor cycles. It is surprising thatlower Cisco cycles, such as these in the Gonzalesto Gunsight interval, are more persistent andabundant in section 1–1′ than in section 2–2′

(Figs. 16 and 17). This is because section 2–2′ islocated in the upper sediment bypass zone, whereregressive sediment bypassing and erosion wereextensive (Figs. 14 and 21).

Cycle Type and Magnitude.Contrastingtrends are present between the lower and upperCisco Group and between updip and downdipsections. In the lower Cisco Group, low-magni-tude, type II minor cycles are common on highs,whereas high-magnitude, type I cycles are com-mon in lows, such as the minor cycles in section1–1′ (Fig. 19). This trend suggests that the mag-nitude of high-order transgressive-regressiveevents decreases toward highs where accommo-dation space is small. Because many minor cy-cles in the lower Cisco Group are the only cycle

��

��

��

��

��

��

��

��

��

��

15 8712 90 894

South

0 5 mi

0 8 km

��

��

��

yyyyyyyy

��

yyyyyy

��

��

��

��

��

��

��

��

��

��

��

�

��

��

��

��

��

yyyyyyyy��

��

��

��yy

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

yyyyyyyy

��

��

��

��

yyy

��

��

��

Gunsight No. 1

Bunger No. 3

Bunger No. 1

Gonzales No. 1

Gonzales No. 2 (?)

Deltaicprogradation

Sedimentbypassing

Inferred geometry of deltaic progradation in the Bunger no. 3 intermediate cycle

Area of major depositional loci during regression

Inferred fluvial channel

Direction of regressive sediment transportation

150

0 0

40

QAc1984c

ft m

9

Bunger no.7

Fluvial depositionand erosion

Vertical exaggeration 164

North

Figure 19. Part of cross section 1–1′ showing variations in cycle continuity and thickness, and in lithofacies and thickness of sand-rich intervals,as controlled by local and regional topography, timing, type, and amount of siliciclastic sediment supply, and depositional dynamics. A sea-leveldatum is used because of extreme topographic variations and small tectonic disturbance in the Eastern shelf since late Paleozoic time (Wermundand Jenkins, 1969; Brown et al., 1987). In the lower Cisco Group, this section is at a high angle to depositional strike, reflecting the nature of evolv-ing topography and sediment supply pattern. See text for discussion. See Figures 11, 13, and 14 for more keys.

YANG ET AL.


in an intermediate cycle, this trend indicates thatthe magnitude of intermediate transgressive-re-gressive events also decreases toward highs.

Topographic control, however, weakens in theupper Cisco Group, where gentle topography re-sulted in many minor cycles of persistent or vary-ing type and magnitude along strike (Figs. 17 and18). Any subtle imbalances among local topogra-phy, siliciclastic influx, magnitude of transgres-sion, and depositional dynamics could cause lat-eral variations of cycle type and magnitude. Thepersistent type and magnitude, however, couldoccur because (1) the topography was flat, suchas in the Sedwick and Santa Anna Branch inter-vals, (2) the magnitude of transgression waslarge, or (3) the siliciclastic influx was minimal.

Cycle type and magnitude are much less per-sistent in 2–2′ than in 3–3′. This is because sec-tion 2–2′ is located in the sediment bypass zonewhere sedimentation was erratic. In fact, somecycles in section 2–2′ are type I on highs and typeII in lows, such as the Saddle Creek no. 1 andStockwether no. 1 cycles (Fig. 21). The highswere most likely devoid of major depositionalloci, so that siliciclastic suppression and contam-ination of carbonate and core shale depositionwere greatly reduced.

Cycle Thickness.The consistency of cyclethickness varies greatly between sections and be-tween different orders of cycles. Minor cyclesthicken toward lows or have constant thicknessin section 2–2′, but they have constant thicknessor thicken gently toward lows in section 3–3′.

Intermediate cycles thicken toward lows orhighs or have constant thickness. Large thicknesschanges occur commonly in updip sections 1–1′and 2–2′. For example, from south to north, theBunger no. 3 cycle in section 1–1′ thickens wherethick fluvial sediments were deposited, thins overa local high due to sediment bypassing, thickensagain in a low due to deltaic progradation, and fi-nally thins again where sediment supply from thesouth diminished (Fig. 19).

Topographic and depositional dynamics con-trols on thickness variations of intermediate cy-cles are also well demonstrated in section 2–2′(Fig. 21). The depositional loci were located inpreexisting lows, where thick sandstones weredeposited. The sandstones formed highs owing totheir large thickness and small compactability, to-ward which the overlying cycles thin. This greatlateral and vertical thickness change is in accord-ance with the common cycle absence and poorcycle continuity. They were caused by extensiveerosion, frequent switching of depositional loci,and the relatively steep topography in the sedi-ment bypass zone on the middle platform.

In contrast, the consistent thickness of bothminor and intermediate cycles in section 3–3′ in-dicates diminished influences of fluvial anddeltaic sedimentation and topography on cycle

thickness in the lower platform. As a result, re-gional high-frequency transgressive-regressiveevents were well recorded, strongly suggesting aeustatic origin for these events.

Sand-rich Intervals. Variations in lithofa-cies and thickness of sand-rich intervals in in-termediate cycles along strike are closely re-lated to the location of depositional loci and thepattern of siliciclastic supply. This relationshipis evident in the progradational interval ofBunger no. 3 cycle in section 1–1′ (Fig. 19).This interval thickens in the updip part, wherefluvial sandstones dominate, and in the downdip

part, where deltaic sandstones dominate. It thinsover a high separating the fluvial depocenterfrom the deltaic depocenter, and thins fartherdowndip, where deltaic progradation ceased.Thickening of sand-rich intervals toward fluvialand deltaic depocenters also occurs throughoutsection 2–2′ (Fig. 21).

Sand-rich intervals in section 3–3′ are thin andlaterally consistent and, in some cases, thickengently toward topographic lows. This is attrib-uted to the diminished supply of coarse sedi-ments, especially fluvial deposits, and the attenu-ated topography in the lower platform.

Maximum-transgressive sandstone of type V cycles

9546 56

South

0 1 2 mi

0 3 km

50

100

150

40

��

��

��

��

��

��yz{|

��

yyyyyyyyyyyyyy

��

��

yyyyyyyyyy

��

��

yyyyyyyyyy

��

�

��

��

��

�

��

��

��

yyyyyyyyy

��

��

��

yyyyyyyyy

��

yyyyyyyyyyyy

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

yyyyyyyyyyyy

��

��

��

yyyyyyyyy

��

��

��

��

��

yyyyyyyyyyyyyyyyyyyyy

��

��

��

yyyyyyyyyyyyyyyyyyyyy

��

��

��

��

��yy

zzzz{||

Stockwether No. 3

Shallowing

Deepening

Shallowing

Deepening

Shallowing

Sedwick No. 2Sedwick No. 1

Gouldbusk No. 1

Dothan No.

Dothan No.

Stockwether No. 1

QAc198

0 0mft

Camp Colorado No. 5

Camp Colorado No. 4

Camp Colorado No. 2

Camp Colorado No. 1

Stockwether No. 5

Stockwether No. 3


North

Figure 20. Part of cross section 2–2′ showing varying cycle continuity and repeated shallow-ing and deepening of a topographic low, as controlled by the dynamics of siliciclastic and car-bonate deposition and differential compaction. See Figures 11, 13, and 14 for more keys.



DISCUSSION

Autogenic and Allogenic Processes

Allogenic processes operate at a basin orglobal scale and are external to the depositionalsite; autogenic processes operate within the dep-ositional site at a local scale. These two types ofprocesses are distinguished by their end products,i.e., regional systematic vs. local nonsystematicvariations of cycle characters identified throughcycle correlation and stratigraphic and sedimen-tologic analyses. Distinguishing these processesenables us to better understand the mechanismsof individual processes in cyclic sedimentation.

Regional systematic variations of Cisco cyclecharacters indicate allogenic controls (Table 1).They include high-frequency sea-level changes,shelfwide subsidence, regional pattern of silici-clastic supply, and regional topography. Regionaltopography determines the regional geometry ofprogradation and aggradation and location of de-pocenters and sediment bypass zones. Timing,type, and amount of siliciclastic sediment supplymainly determine lithofacies distribution andthickness, which change from dominantly fluvialand deltaic in the upper platform to dominantlyshelf siliciclastic and carbonate in the lower plat-form because of diminishing coarse siliciclasticsediment supply. Shelf subsidence is the ultimatecontrol on sediment preservation. Shelf configu-

ration, supply pattern, and subsidence combine todetermine stratigraphic completeness and resolu-tion of the Cisco Group. Climatic controls onsediment yield in source areas may have con-trolled the timing of regressive progradation ofintermediate cycles, as inferred from the outcropstudy (Yang, 1996).

Local variations of cycle characters indicateautogenic controls (Table 1). They include localtopography, differential compaction, and deposi-tional dynamics, which control lateral facieschanges, depositional locus switching, and car-bonate and siliciclastic deposition and erosion.Local faulting and folding are not evident in thestudy area.

Autogenic processes operate within the frame-work and boundary conditions set up by allo-genic processes. Autogenic processes locallymodify or even destroy the framework throughgrain-by-grain sedimentation at a specific siteand time, to record the ambient conditions ofdepositional environments at different scales.Therefore, autogenic processes modulate andmodify the detailed character of allogenic prod-ucts. For example, fluvial sediment influx anddeltaic progradation could outpace sea-level rise,causing shoreline regression and cycle absence;fluvial channel erosion could excavate the under-lying sediments, fragmenting cycles and destroy-ing the previous record (Figs. 19 and 21). As a re-sult, the Cisco record in the upper platform and

sediment bypass zone displays poor cyclicity be-cause of cycle absence and discontinuity and isessentially an autocyclic record, whereas therecord in the lower platform is dominantly an al-locyclic record, where signatures of allogenicprocesses were clearly preserved by relativelycontinuous and uniform shelf siliciclastic andcarbonate sedimentation.

Allogenic and autogenic signatures, however,are highly mixed. This study demonstrates that apredominantly autocyclic record in the upper plat-form does not imply the absence of allogenicprocesses. Allogenic controls on sedimentation inthe lower platform were accomplished via auto-genic processes. Therefore, an allocyclic recordmay inherit many autogenic imprints.

High-Frequency Sea-Level Changes

Persistent and repetitive transgression-regres-sion in the study area can easily be explained byhigh-frequency sea-level changes. This, however,can also be caused by cyclic shelf subsidence orvariations in siliciclastic influx when accommo-dation space steadily increases (Yang, 1996).

There is no direct evidence in the Cisco Groupsuggesting episodic regional subsidence, in par-ticular, different orders of subsidence. To the con-trary, tectonic stability of the Eastern shelf duringCisco Group deposition has been suggested inthis and previous studies (e.g., Brown et al.,

TABLE 1. VARIATIONS OF CYCLE CHARACTERS AND THEIR CONTROLS, SUBSURFACE CISCO GROUP, EASTERN SHELF

Cycle Variations of cycle characters in the study area ControlscharactersAbundance More minor and intermediate cycles in lower platform

Absence Amalgamation of minor cycles in highs or lows, common in lower platformSiliciclastic sediment suppression of carbonate deposition common in upper platform Allogenic

Fluvial erosion common in middle and upper platform Eustatic sea-levelCommon absent in highs or lows where depositional loci present changes of different orders

Continuity Persistent: most minor cycles continuous more than 16 km, many Regional topography and/or

intermediate cycles more than 48 km shelf configuration

Intermediate cycles extend farther landward than minor cycles Timing, type, and amount

Nonpersistent cycles: continuous but fragmentary; only in lower platform; of regional siliciclastic

only of local extent sediment supply

Most continuous on lower platform, least in bypass zone Shelf subsidence

Type 20 minor cycles: type II to type I, increased magnitude in downdip direction Possible climatic

and 10 minor cycles: consistent type and magnitude across the shelf changes

magnitude Other minor cycles: No systematic changes

Intermediate cycles: increased magnitude in downdip directionHigh magnitude, type I in lows, low-magnitude, type II in highs in lower

Cisco Group, more consistent in upper Cisco GroupMore consistent in lower platform than in middle and upper platform Autogenic

Thickness In a major cycle, intermediate cycles are thicker than minor cycles Local depositionalMinor cycles: many have consistent thickness for more than 16 km; some topography

thicken in downdip direction or lows; some thicken abruptly Local subsidenceIntermediate cycles: thicken in downdip direction or lows; thin or thicken caused by differ-

from shelf edge to basin slope; rare thickening in highs ential compactionMore consistent in lower platform than in middle and upper platform Carbonate and

Thickness Thickens toward upper- and lower-platform depocenters, thins in middle- siliciclastic deposi-

platform bypass zone tional dynamics

The lag between coarse siliciclastic progradation and regression Local variations of

increases in downdip direction depositional locus

Lithofacies Upward-shallowing, dominantly deltaic in lower-platform depocenter,dominantly fluvial in upper platform depocenterS

and-

rich

inte

rval

s

YANG ET AL.


1990; Elam, 1969; Wermund and Jenkins, 1969;Yang and Dorobek, 1992).

Cyclic siliciclastic influx onto the shelf couldbe caused by episodic tectonic uplift or cyclic cli-matic changes in source areas. There is no evi-dence from either the subsurface or the outcrop tosuggest episodic uplift of the source areas (Yang,1995). High-frequency, regional episodic upliftof different orders over a period of several mil-lion years is yet to be documented, and it is un-likely to have occurred in the study area.

Cyclic waxing and waning of sediment yieldin the source areas caused by high-frequency cli-matic changes are suggested in the outcrop studyof the Cisco Group (Yang, 1996). Assuming asea-level stillstand and constant shelf subsidence,high siliciclastic sediment supply on the shelfwould cause shoreline regression and low supplywould cause shoreline transgression (Curray,1964; Yang, 1996). The effect of varying silici-clastic supply on shoreline movement should bemore prominent in the upper platform, close tothe sediment source, than in the lower platform.However, all characteristics of Cisco cycles be-come much more persistent in the downdip di-rection, away from the influence of land-derivedsiliciclastics. Therefore, cyclic siliciclastic supplycannot adequately account for high-frequency,multiorder transgression-regression throughoutthe Cisco Group.

Exclusion of tectonic and sediment-supplycauses for regional transgression and regressionleaves high-frequency sea-level changes as themajor cause of cyclicity. Continental glaciationsimilar to that of the Pliocene-Pleistocene has beendocumented for late Paleozoic time (Crowell,1978; Veevers and Powell, 1987) and probablycaused high-frequency, high-amplitude sea-levelchanges. They have been suggested as the causefor many Pennsylvanian and Permian cycles(Wanless and Shepard, 1936; Wilson, 1967;Heckel, 1977, 1986; Goldhammer et al., 1994), in-cluding transgressive-regressive cycles of theCisco Group in outcrop (e.g., Lee, 1938; Harrison,1973; Boardman and Malinky, 1985; Boardmanand Heckel, 1989; Yancey, 1991; Yang, 1996).

In addition, cycle ordering indicates orderedhigh-frequency sea-level changes related toMilankovitch climatic cycles. A 2:1 or 3:1 ra-tio between minor and intermediate cycles iscomparable to the ratio between Milankovitchshort-eccentricity and obliquity cycles, as inthe outcrop cycles (Yang, 1995, 1996). Moresignificantly, spectra of 15 Cisco records dis-play distinct Milankovitch eccentricity, obliq-uity, and precessional-index peaks (Yang andKominz, 1996). In summary, geologic andquantitative evidence strongly suggests, butdoes not prove, a major high-frequency sea-level control on cyclic sedimentation of theCisco Group.

Cyclostratigraphy and Autogenic and Allogenic Processes

A major goal in meter-scale cycle studies isto establish a high-resolution cyclostratigraphycalibrated to the periods of Milankovitch cli-matic cycles (e.g., Herbert, 1992; Hinnov andGoldhammer, 1991; House and Gale, 1995;Schwarzacher, 1993; Yang et al., 1995). This isonly possible if Milankovitch orbitally inducedphysical processes, such as climatic and sea-level changes, are indeed the major controls onthe stratigraphic and temporal regularities of a cyclic record. Because non-Milankovitch

processes could also produce meter-scale cycles, it is imperative to distinguish theseprocesses by examining their stratigraphic sig-natures (Brown, 1969). In addition, understand-ing the mechanisms and interplay of theseprocesses will undoubtedly increase the predic-tive power of cyclostratigraphy.

In some studies, average cycle periods arematched with the periods of Milankovitch cli-matic cycles, to suggest a Milankovitch origin forancient cycles, and even to calibrate the chrono-stratigraphy. The results of this study suggest ex-treme caution in this practice because of the pres-ence of different orders of cycles and local

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

9525 4634 56

South

��

��

yyyyyy

��

��

��

yyyyyy

��

yyyyyy

��

��

��

yyyyy

��

��

��

��

�

Flippen No. 1

��

��

��

��

��

��

��

��

��

yyy

��

��

��

��

��

��

yyyy

��

��

��

��

��

��

��

��

��

��

��

��

yyyyyy

��

��

��

��

��

yyyyyyyyyy

��

��

��

��

��

��

��

��

��

��

��

yyyyyyyyy

��

��

��

yyyyyy

��

��

��

��

��

��

��

��

��

��

��

��

��

yyyyyyyyyyyyyyyy

��

yyyyyyyyyy

��

yyyyyyyy

��

��

��

��

��

��

�

��

50

10

0

0 8 km

5 mi

Stockwether No. 5

Stockwether No. 3

Stockwether No. 1

Saddle Creek No.

Flippen No. 7

Flippen No. 6

Flippen No. 5

??

?

QAc1986c

0 0mft ?

Flippen No. 4

Flippen No. 3

wether No. 3

Stock-


Area of major depositional loci during regression

North

Camp Colorado No. 1

Figure 21. Part of section 2–2′ showing frequent depositional locus switching during regres-sion to demonstrate the relation between local topography and depositional locus and their con-trols on cycle thickness, type, and continuity. See Figures 11, 13, 14, and 19 for more keys.



autogenic cycles, and the common cycle absencecaused by amalgamation, siliciclastic sedimentsuppression, and erosion.

SUMMARY

1. Minor, intermediate, and major transgres-sive-regressive cycles of the Cisco Group vary intype, magnitude, thickness, and probably dura-tion. They form a stratigraphic hierarchy.

2. Regional systematic variations in cycleabundance, continuity, type, magnitude, thick-ness, and the lithofacies and thickness of sand-rich intervals indicate allogenic controls oncyclic sedimentation of the Cisco Group. High-frequency sea-level changes controlled the extentand order of regional transgressive-regressiveevents. Regional topography controlled the dis-tribution of depocenters and bypass zones. Re-gional siliciclastic supply patterns controlledcomponent lithofacies and thickness of cycles.Shelf subsidence is the ultimate control on cyclepreservation.

3. Autogenic processes caused local variationsof cycle characters. Local topography, differen-tial compaction, and depositional dynamics com-bined to control lateral facies changes, local dep-ositional loci, and erosion. Regressive fluvial anddeltaic sedimentation was extensive in the upperand middle platform, causing common cycle ab-sence and obliterating the signatures of allogenicprocesses. In contrast, shelf carbonate and silici-clastic sedimentation in the lower platform haveclearly recorded allogenic signatures.

4. Comparison of cycle characters across theshelf indicates that a predominantly autocyclicrecord does not imply the absence of allogenicprocesses, and a seemingly allocyclic record maycontain many autocyclic imprints.

ACKNOWLEDGMENTS

Acknowledgment is made to the Donors of thePetroleum Research Fund, administrated by theAmerican Chemical Society under grants ACS-PRF 20852-AC8 and ACS-PRF 27102-AC8 toM. A. Kominz, which provided partial supportfor this research. Publication was authorized bythe Director, Bureau of Economic Geology, Uni-versity of Texas at Austin.

This paper reports part of Yang’s dissertationresearch at University of Texas at Austin. We aregrateful to W. Galloway, G. Bond, and J. Bannerfor thoughtful discussions and reviews. Criticalreviews by T. Cross, F. Read, and B. Wilkinsonhelped tremendously in shortening the length andsharpening the focus of this paper. We thank thefollowing individuals and institutions for accessto data and use of equipment: P. Knox and Bu-

reau of Economic Geology, University of Texasat Austin, for access to cores, well-log digitiz-ing equipment, and wireline logs; V. Jenkins and Mobile Oil Company for access to cores; M. Dusing and Coda Energy Company for wire-line logs; and Geomap Company for discountedwireline logs and maps. This work is supportedby grants from Grant-in-Aid of the GeologicalSociety of America, the DeFord Field GeologyFellowship of the Geology Foundation, Univer-sity of Texas at Austin, and Grant-in-Aid of theGulf Coast Associations of Geological Societies.

REFERENCES CITED

Boardman, D. R., II, and Malinky, J. M., 1985, Glacial-eustaticcontrol of Virgilian cyclothems in north-central Texas, inMcNulty, C. L., and McPherson, J. G., eds., Transactions:American Association of Petroleum Geologists, South-west Section: Fort Worth, Texas, Fort Worth GeologicalSociety, p. 13–23.

Boardman, D. R., II, and Heckel, P. H., 1989, Glacial-eustaticsea-level curve for early Late Pennsylvanian sequence innorth-central Texas and biostratigraphic correlation withcurve for midcontinent North America: Geology, v. 17,p. 802–805.

Brown, L. F., Jr., 1969, Virgil-lower Wolfcamp repetitivedepositional environments in north-central Texas, inElam, J. G., and Chuber, S., eds., Cyclic sedimentationin the Permian Basin: Midland, West Texas GeologicSociety, p. 115–134.

Brown, L. F., Jr., Iriarte, R. F. S., and Johns, D. A., 1987, Re-gional stratigraphic cross sections, Upper Pennsylva-nian and Lower Permian strata (Virgilian and Wolf-campian Series), north-central Texas: University ofTexas at Austin, Bureau of Economic Geology CrossSections, 27 p.

Brown, L. F., Jr., Iriarte, R. F. S., and Johns, D. A., 1990, Re-gional depositional systems tracts, paleogeography, andsequence stratigraphy, Upper Pennsylvanian and LowerPermian strata, north-central Texas: University of Texasat Austin, Bureau of Economic Geology Report of Inves-tigations 197, 116 p.

Crowell, J. C., 1978, Gondwanan glaciation, cyclothems, con-tinental positioning, and climate change: American Jour-nal of Science, v. 278, p. 1345–1372.

Curray, J. R., 1964, Transgressions and regressions, in Miller,R. L., ed., Papers in marine geology: New York, Macmil-lan, p. 175–203.

Dresser Atlas, 1974, Log review 1: Dresser Atlas Division,Dresser Industries, Inc., 150 p.

Elam, J. G., 1969, Tectonic style in the Permian Basin and itsrelationship, in Elam, J. G., and Chuber, S., eds., Cyclicsedimentation in the Permian Basin: Midland, West TexasGeologic Society, p. 55–80.

Galloway, W. E., 1971, Depositional systems and shelf-sloperelationships in uppermost Pennsylvanian rocks of theEastern shelf, north-central Texas [Ph.D. dissert.]: Uni-versity of Texas at Austin, 116 p.

Goldhammer, R. K., Oswald, E. J., and Dunn, P. A., 1994,High-frequency, glacio-eustatic cyclicity in the MiddlePennsylvanian of the Paradox Basin: An evaluation ofMilankovitch forcing, in de Boer, P. L., and Smith, D. G.,eds., Orbital forcing and cyclic sequences: InternationalAssociation of Sedimentologists Special Publications 19,p. 243–283.

Harrison, E. P., 1973, Depositional history of Cisco-Wolfcampstrata, Bend Arch, north-central Texas [Ph.D. dissert.]:Lubbock, Texas Tech University, 189 p.

Heckel, P. H., 1977, Origin of phosphatic black shale facies inPennsylvanian cyclothems of Midcontinent North Amer-ica: American Association of Petroleum Geologists Bul-letin, v. 61, p. 1045–1068.

Heckel, P. H., 1986, Sea-level for Pennsylvanian eustatic ma-rine transgressive-regressive depositional cycles along

Midcontinent outcrop belt, North America: Geology,v. 14, p. 330–334.

Herbert, T. D., 1992, Paleomagnetic calibration of Mi-lankovitch cyclicity in Lower Cretaceous sediments:Earth and Planetary Science Letters, v. 112, p. 15–28.

Hinnov, L. A., and Goldhammer, R. K., 1991, Spectral analysisof the Middle Triassic Latemar Limestone: Journal ofSedimentary Petrology, v. 61, p. 1173–1193.

House, M. R., and Gale, A. S., 1995, Orbital forcing timescalesand cyclostratigraphy: Geological Society [London] Spe-cial Publication 85, 210 p.

Lee, W., 1938, Stratigraphy of the Cisco Group of the BrazosBasin: Austin, University of Texas Publication 3801,p. 11–90.

Mount, J. F., 1984, Mixing of siliciclastic and carbonate sedi-ments in shallow shelf environments: Geology, v. 12,p. 432–435.

Schlumberger, 1987, Log interpretation principles/applica-tions: Houston, Texas, Schlumberger Educational Serv-ices, 198 p.

Schwarzacher, W., 1993, Cyclostratigraphy and the Mi-lankovitch theory: Developments in Sedimentology 52:London, Elsevier, 225 p.

Serra, O., 1985, Sedimentary environments from wirelinelogs: Houston, Texas, Schlumberger Educational Serv-ices, 211 p.

Serra, O., 1986, Fundamentals of well-log interpretation, 2. theinterpretation of logging data: Developments in Petro-leum Science 15B: New York, Elsevier, 684 p.

Veevers, J. J., and Powell, C. M., 1987, Late Paleozoic glacialepisodes in Gondwanaland reflected in transgressive-re-gressive depositional sequences in Euramerica: Geologi-cal Society of America Bulletin, v. 98, p. 475–487.

Wanless, H. R., and Shepard, F. P., 1936, Sea level and climaticchanges related to late Paleozoic cycles: Geological Soci-ety of America Bulletin, v. 38, p. 1177–1206.

Wermund, E. G., and Jenkins, W. A., Jr., 1969, Late Pennsyl-vanian Series in north-central Texas, in Wermund, E. G.,and Brown, L. F., Jr., eds., A guidebook to the Late Penn-sylvanian shelf sediments, north-central Texas: DallasGeological Society, American Association of PetroleumGeologists-Society of Economic Paleontologists andMineralogists Annual Meeting, p. 1–11.

Wilson, J. L., 1967, Cyclic and reciprocal sedimentation in Vir-gilian strata of Southern New Mexico: Geological Soci-ety of America Bulletin, v. 78, p. 805–818.

Yancey, T. E., 1991, Controls on carbonate and siliciclasticsediment deposition on a mixed carbonate-siliciclasticshelf (Pennsylvanian Eastern shelf of north Texas), inFranseen, E. K., Watney, W. L., Kendall, C. G. St. C.,and Ross, W., eds., Sedimentary modeling: Computersimulations and methods for improved parameter defi-nition: Kansas Geological Survey Bulletin 233,p. 263–272.

Yang, K.-M., and Dorobek, S. L., 1992, Mechanisms for latePaleozoic synorogenic subsidence of the Midland andDelaware Basins, Permian Basin, Texas and New Mex-ico: West Texas Geological Society Bulletin, v. 32,p. 45–60.

Yang, W., 1996, Cycle symmetry and its causes, Cisco Group(Virgilian and Wolfcampian), Texas: Journal of Sedimen-tary Research, v. 66B, p. 1102–1121.

Yang, W., 1995, Depositional cyclicity of the Cisco Group(Virgilian and Wolfcampian), north-central Texas[Ph.D. dissert.]: Austin, University of Texas at Austin,268 p.

Yang, W., and Kominz, M. A., 1996, Quantitative assessmentof causal mechanisms for cyclicity of the Cisco Group,Eastern shelf, Midland basin, in 1996 AAPG AnnualConvention Official Program: San Diego, California,p. A156.

Yang, W., Harmsen, F., and Kominz, M. A., 1995, Quantita-tive analysis of a peri-tidal carbonate sequence, the Mid-dle and Upper Devonian Lost Burro Formation, DeathValley, California—A possible Milankovitch climaticrecord: Journal of Sedimentary Research, v. 65B,p. 306–322.

MANUSCRIPTRECEIVED BY THESOCIETY JANUARY 17, 1997REVISEDMANUSCRIPTRECEIVEDSEPTEMBER24, 1997MANUSCRIPTACCEPTEDJANUARY 23, 1998

distinguishing the roles of autogenic versus allogenic ...web.mst.edu/~yangwa/yang webpage/yang et...

Documents