sequence-stratigraphic controls on complex reservoir architecture of

AUTHORS

William A. Ambrose � Bureau of EconomicGeology, John A. and Katherine G. JacksonSchool of Geosciences, University of Texas atAustin, Austin, Texas, 78713-8924;[email protected]

William A. Ambrose is a research scientist atthe Bureau of Economic Geology, John A. andKatherine G. Jackson School of Geosciences atthe University of Texas at Austin. His areas ofinterest include unconventional energy minerals,clastic depositional systems, and stratigraphy.He holds M.A. and B.S. degrees in geosciencesfrom the University of Texas at Austin.

Tucker F. Hentz � Bureau of EconomicGeology, John A. and Katherine G. JacksonSchool of Geosciences, University of Texas atAustin, Austin, Texas, 78713-8924;[email protected]

Tucker F. Hentz is a research associate at the Bureauof Economic Geology, John A. and Katherine G.Jackson School of Geosciences at the Universityof Texas at Austin. His areas of interest includesequence-stratigraphic analysis and clastic depo-sitional systems. He holds an M.S. degree in ge-ology from the University of Kansas.

Florence Bonnaffe � Bureau of EconomicGeology, John A. and Katherine G. JacksonSchool of Geosciences, University of Texas atAustin, Austin, Texas, 78713-8924;[email protected]

Florence Bonnaffe is a research scientist associ-ate at the Bureau of Economic Geology, John A.and Katherine G. Jackson School of Geosciencesat the University of Texas at Austin. Her previouswork was with Elf Exploration Production andCompagnie Generale de Geophysique. She re-ceived her M.S. degree in applied geophysics fromthe University of Paris in 1996.

Robert G. Loucks � Bureau of EconomicGeology, John A. and Katherine G. Jackson Schoolof Geosciences, University of Texas at Austin,Austin, Texas, 78713-8924;[email protected]

Robert G. Loucks is a senior research scientist atthe Bureau of Economic Geology, John A. andKatherine G. Jackson School of Geosciences atthe University of Texas at Austin. His research in-terests include carbonate and siliciclastic sequencestratigraphy, depositional systems, diagenesis,and reservoir characterization. He holds a Ph.D.from the University of Texas at Austin.

Sequence-stratigraphiccontrols on complex reservoirarchitecture of highstandfluvial-dominated deltaic andlowstand valley-fill depositsin the Upper Cretaceous(Cenomanian) WoodbineGroup, East Texas field:Regional and local perspectivesWilliam A. Ambrose, Tucker F. Hentz,Florence Bonnaffe, Robert G. Loucks,L. Frank Brown Jr., Fred P. Wang, and Eric C. Potter

ABSTRACT

An analysis of 31 whole cores (�1600 ft, �490 m) and closelyspaced wireline logs (�500 wells) penetrating the Lower Cre-taceous (Cenomanian) lower Woodbine Group in the matureEast Texas field and adjacent areas indicates that depositionalorigins and complexity of the sandstone-body architecture inthe field vary from those inferred from previous studies. Hetero-geneity in the lower Woodbine Group is controlled by high-stand, fluvial-dominated deltaic depositional architecture, withdip-elongate distributary-channel sandstones pinching outover short distances (typically <500 ft [<150 m]) into delta-plain and interdistributary-bay siltstones and mudstones. Thishighstand section is truncated in the north and west parts ofthe field by a thick (maximum of 140 ft [43 m]) lowstand,incised-valley-fill succession composed of multistoried, coarse-gravel conglomerate and coarse sandstone beds of bed-load

AAPG Bulletin, v. 93, no. 2 (February 2009), pp. 231–269 231

Copyright #2009. The American Association of Petroleum Geologists. All rights reserved.

Manuscript received May 7, 2008; provisional acceptance July 22, 2008; revised manuscript receivedSeptember 3, 2008; final acceptance September 18, 2008.

DOI:10.1306/09180808053

fluvial systems. In some areas of the field, this valley fill di-rectly overlies distal-delta-front deposits, recording a fall inrelative sea level of at least 215 ft (65 m).

Correlation with the Woodbine succession in the EastTexas Basin indicates that these highstand and lowstand depos-its occur in the basal three fourth-order sequences of the unit,which comprises a maximum of 14 such cycles. Previous studiesof the Woodbine Group have inferred meanderbelt sandstonesflanked by coeval flood-plain mudstones and well-connected,laterally continuous sheet sandstones of wave-dominated del-taic and barrier-strand-plain settings. This model is inappro-priate, and a full assessment of reservoir compartmentaliza-tion, fluid flow, and unswept mobile oil in East Texas fieldshould include the highstand, fluvial-dominated deltaic andlowstand valley-fill sandstone-body architecture.

INTRODUCTION

The mature, super-giant East Texas field has been the mostproductive oil field in the U.S. lower 48 states and the secondlargest in the country. From its discovery in 1930 (Minor andHanna, 1933; Alexander, 1951; Hudnall, 1951) through mid-2007, it produced 5.42 billion stock tank barrels of oil (BSTB)from the siliciclastic Upper Cretaceous (Cenomanian) Wood-bine Group (Ambrose et al., 2007). Its calculated ultimaterecovery of approximately 5.49 BSTB and advanced degree ofwater encroachment indicate that it is in the waning stages ofproduction. Given these figures, about 70 million stock tankbarrels (MMSTB) are still most likely producible under cur-rent production practices. However, recent closer evaluationof the amount of bypassed pay, deeper Woodbine pay, andpoorly swept oil indicates that the field has remaining reservesof as much as 550 MMSTB (Ambrose et al., 2007). Becauseof this large estimated remaining-reserves volume, the abun-dant wells that exist in the field for potential recompletionand/or deepening, and currently favorable price of oil, pro-ducers are now aggressively targeting recompletions and espe-cially deeper pay zones in the Woodbine section. Unlike pre-vious studies (Oliver, 1971; Turner and Conger, 1981; Halboutyand Halbouty, 1982; Phillips, 1987; DeDominic, 1988; Jasperand Wagner, 1989), our investigation integrates a sequence-stratigraphic analysis of the East Texas Basin and the adjacentfield with reservoir-scale depositional-facies interpretationusing log data from closely spaced wells (~150–1200 ft, ~46–366 m) and core data to document more precisely the regional

L. Frank Brown Jr. � Bureau of Economic Ge-ology, John A. and Katherine G. Jackson Schoolof Geosciences, University of Texas at Austin, Austin,Texas, 78713-8924; [email protected]

Frank Brown is a research professor at the Bureauof Economic Geology, John A. and Katherine G.Jackson School of Geosciences at the Universityof Texas at Austin. His research interests includesequence stratigraphy, depositional systems, andreservoir characterization. He holds a Ph.D. fromthe University of Wisconsin at Madison.

Fred P. Wang � Bureau of Economic Geology,John A. and Katherine G. Jackson School of Geo-sciences, University of Texas at Austin, Austin,Texas, 78713-8924; [email protected]

Fred Wang is a research scientist at the Bureauof Economic Geology, John A. and Katherine G.Jackson School of Geosciences at the Universityof Texas at Austin. He has experience in reservoircharacterization, shale gas production, deep-shelf and deep-water fields, CO2 sequestration,and enhanced oil recovery. He received a Ph.D.in petroleum engineering from Stanford University.

Eric C. Potter � Bureau of Economic Geology,John A. and Katherine G. Jackson School of Geo-sciences, University of Texas at Austin, Austin,Texas, 78713-8924; [email protected]

Eric C. Potter is an associate director of the Bureauof Economic Geology, John A. and Katherine G.Jackson School of Geosciences at the Universityof Texas at Austin. He worked for 25 years forMarathon Oil Company as an exploration ge-ologist. He holds an M.S. degree in geology fromOregon State University.

ACKNOWLEDGEMENTS

This study was funded by the State of Texas Ad-vanced Resource Recovery project. The authorsare indebted to the East Texas Engineering Asso-ciation for log and engineering data, as well asto William E. Galloway for his insights and obser-vations from the core data. Manuscript editingwas by Lana Dieterich. The manuscript benefitedfrom the reviews of Brian W. Horn, William A.Hill, and Michael L. Sweet. David Stephens, JohnAmes, and Jana Robinson prepared the illustra-tions under the direction of Joel Lardon, manager,Media Information Technology. Partial supportof this research was received from the John A.and Katherine G. Jackson School of Geosciencesand the Geology Foundation at the Universityof Texas at Austin. Publication was authorized bythe director, Bureau of Economic Geology.

232 Woodbine Group of East Texas Field

chronostratigraphic framework of the Woodbineand sandstone origins, trends, and continuity inthe field.

The field encompasses approximately 134,000 ac(�210 mi2, �542 km2) in parts of Gregg, Rusk,Smith, and Upshur counties (Figure 1), throughoutwhich more than 31,200 wells have been drilled(Halbouty and Halbouty, 1982; Galloway et al.,

1983; Ambrose et al., 2007). The average wellspacing is 4.3 ac (1.7 ha) (range: 0.05–15 ac [0.02–6.1 ha]) (East Texas Engineering Association, 1953,and data files of the Railroad Commission of Texas).Because of the high density of wells and the field’slong-term trend of depletion, few new wells target-ing the Woodbine section have been drilled since1990. Moreover, despite the field’s long history,

Figure 1. East Texas field, major regional structural elements, and outline of the study area (modified from Siemers, 1978). The eastTexas salt diapir province extends throughout the East Texas Basin: from the Sabine uplift to within 20 mi (32 km) of the Mexia-Talcofault zone (Seni and Jackson, 1984). The location of cored wells cited in this study: (A) Cities Service 2B Killingsworth; (B) Arco B142King; (C) Shell 55 Watson; and (D) Arco C-21 Pinkston, Arco C19 Pinkston, and Arco 18 Griffin, collectively. Specific locations for wellsin group D are shown in Figure 17. NPA = north pilot area; SPA = south pilot area.

Ambrose et al. 233

it has never been unitized to promote maximumrecovery efficiency. As a result, large secondary-recovery projects, such as significant waterfloods(1 mi2 [2.6 km2] or larger), have rarely been im-plemented because coordination of such operationsis difficult over multiple leases.

Primary hydrocarbon accumulation in the fieldis in west-dipping Woodbine sandstones that aretruncated on the west flank of the Sabine uplift by asubregional unconformity below the Austin Chalk(hereafter termed the ‘‘base-of-Austin unconfor-mity’’) that formed during the Late Cretaceous af-ter uplift and erosion. Impermeable calcareous silt-stones of the Austin form the seal throughout thefield. Reservoirs are associated with a strong wa-ter drive resulting from a combination of region-ally tilted Woodbine strata, inferred excellentsandstone-body continuity, and consistently highporosity (averaging 25.2%) and permeability (av-eraging 2.1 d) values (East Texas Engineering As-sociation, 1953). Excellent vertical and horizontalsweep efficiency contributes to an overall high re-covery factor of approximately 77%.

Although the East Texas field has been produc-ing for more than 75 yr, no modern comprehen-sive study of the field exists. This study is the firstto integrate core data from the field and adjacentareas using well-log analysis; our main objective isto understand the function of sequence stratigraphyin the depositional origin of the producing inter-vals in the East Texas field. An additional objectiveis to document facies distribution in the field andto describe its controls on the production potentialfor bypassed pay, deeper Woodbine pay zones, andincompletely swept zones.

Tasks that achieved these objectives included(1) applying sequence-stratigraphic analysis to theWoodbine Group in the East Texas Basin, wherethe succession is complete, and extending it to thetruncated Woodbine section on the adjacent Sa-bine uplift to identify principal chronostratigraphicboundaries in the East Texas field; (2) using theseboundaries to map coeval sandstone units, identi-fying reservoir-facies trends in the pilot areas, andinterpreting depositional-facies origin and distribu-tion and Woodbine paleogeography; and (3) infer-ring sequence-stratigraphic and facies controls on

fluid flow, incompletely swept reservoir zones,potential bypassed pay and deeper pay zones, andenhanced oil recovery by integrating our findingsfrom the large core and log data set with engineer-ing data.

DATABASE AND METHODS

This subsurface study of the Woodbine Group inthe East Texas field and adjacent East Texas Basinused approximately 1600 ft (�490 m) of sectionfrom 31 whole cores and closely spaced log sectionsfrom approximately 500 wells. Project time con-straints and limitations on acquisition of commonlyold, privately owned data prevented us from con-ducting a fieldwide analysis using the more than31,000 available well logs in the field. Therefore, weselected two pilot areas in the northern and south-ern parts of the field (Figure 1). The north pilot area(NPA) and south pilot area (SPA) encompass ap-proximately 12.25 and 2.9 mi2 (~31.7 and 7.5 km2),respectively, with average well spacings of approx-imately 750 and 500 ft (�229 and 152 m), respec-tively. These areas were chosen for a detailed studybecause of the high density of well control and theavailability of cores. They also represent contrast-ing depositional styles and differences in reservoirperformance. For example, permeability is lower(Adair, 1960) and sandstone-body architectureis more complex in the southern part of the field,where the Woodbine Group is composed of mul-tiple, subregionally discontinuous sandstone units.These sandstones, termed ‘‘stringers’’ by field oper-ators, produce under solution-gas drive and haveexperienced relatively greater pressure declines dur-ing water flooding because of their poor contactwith injected water and low degree of stratigraphiccontinuity (Galloway et al., 1983). In contrast, thestringer zone in the northern part of the field isconsistently thinner and overlain by a massive, high-permeability, conglomerate-and-sandstone unitnamed by operators the ‘‘Main sand’’ (historically,the field’s primary reservoir), which produces un-der a strong water drive.

In the study, we documented the continuity ofsignificant stratigraphic surfaces (unconformities


and flooding surfaces) and interpreted the lateraland vertical extent of sequences and parasequencesin each pilot area. Moreover, we provided paleo-geographic reconstructions based on detailed net-sandstone maps and analysis of log facies. Core in-terpretation enabled us to document the verticallithologic succession and infer Woodbine deposi-tional facies, identify and confirm significant strati-graphic surfaces and bed contacts interpreted fromwell logs, and provide rock evidence of systems-tract interpretations.

To resolve depositionally significant and time-equivalent surfaces and zones within the Wood-bine of the East Texas field, we first examined theunit’s sequence-stratigraphic framework in theEast Texas Basin, where the succession is com-plete. Proper examination of the Woodbine’s se-quence succession in the field necessitated its re-gional characterization first to (1) establish a firmreservoir-scale chronostratigraphic basis for accu-rate correlation to and within the field; (2) identifythe precise Woodbine interval and the systemstracts that extend into the field and that containthe reservoir facies; (3) determine possible regionalstructural influences, and their timing, on Wood-bine deposition; and (4) better define the origin ofthe subregional base-of-Austin unconformity, thehydrocarbon seal that caps the Woodbine in thefield.

Published micropaleontologic data (plank-tonic foraminifera, palynomorphs) from the shalyunits immediately above and below the Woodbine(Maness Shale, lower Eagle Ford Group) wereused primarily to place the study interval within thethird-order worldwide coastal onlap curve of Haqet al. (1988).

Correlation efforts in the basin required select-ing wire-line logs away from salt diapirs wherethe Woodbine interval is commonly incomplete(Jackson and Seni, 1984; Seni and Jackson, 1984).We used techniques for interpreting sequence stra-tigraphy from wire-line logs that were discussedby Van Wagoner et al. (1990) and Mitchum et al.(1993). For genetic correlation within the EastTexas Basin and on the Sabine uplift, we used pri-marily gamma-ray logs to ensure consistent andaccurate representation and interpretation of sand-

stone and mudstone intervals within the Wood-bine Group.

GEOLOGIC SETTING

Regional Tectonic Setting

The Sabine uplift is a low-relief regional anticli-norium lying astride the Texas-Louisiana borderbetween the East Texas Basin and north Louisianadiapir province (Figure 1) (Ewing, 1991a). The struc-ture most likely originated in the late Mesozoic;it overlies and is probably genetically related tobasement blocks that formed as midrift highsduring the Triassic rifting phase of the reopeningof the Gulf of Mexico (Jackson, 1982; Nunn, 1990).The Sabine uplift is inferred to have been period-ically active during the Mesozoic and early Ter-tiary (Jackson and Laubach, 1991). Its presentform is related to a Gulf-wide series of middle toLate Cretaceous disturbances and uplifts (Ewing,1991a, b). Halbouty and Halbouty (1982) specif-ically described two episodes of uplift of the struc-ture: just prior to Woodbine deposition and dur-ing late Woodbine and Eagle Ford sedimentation.Jackson and Laubach (1991) also proposed an up-warp of the structure just before accumulation of theWoodbine. Woodbine sediments were derived fromthe Ouachita and Arbuckle Mountains of Okla-homa and Arkansas (Stehli et al., 1972), and thesuccession thins and pinches out eastward againstthe west flank of the Sabine uplift.

The East Texas Basin, a structural embaymentof the Gulf Coast Basin, is bounded on the northand west by the Mexia-Talco fault system (Figure 1).In its deepest part, the basin is filled with morethan 13,000 ft (>3960 m) of Mesozoic and Ter-tiary strata (Wood and Guevara, 1981) that werestructurally modified by mobilization of the Mid-dle Jurassic Louann Salt, most commonly as diapirs,throughout Cretaceous and early Tertiary Basin sed-imentation (Seni and Jackson, 1984). The Angelina-Caldwell flexure, a low-relief monocline that wasmost prominent as a depositional feature duringthe Late Cretaceous and Paleogene (Stehli et al.,

Ambrose et al. 235

1972), separates the embayment from the GulfCoast Basin.

Stratigraphy and Regional Depositional Setting

The middle and upper Cenomanian WoodbineGroup represents the dominant, most widespreadepisode of coarse-siliciclastic deposition during theLate Cretaceous in the East Texas Basin (Figure 2).Stratigraphically, the Woodbine occurs betweenthe older, limestone-dominated Washita Group(upper Albian to middle Cenomanian) and theyounger, areally extensive Eagle Ford Group (lowerTuronian), the principal source rock of East Texasfield hydrocarbons (Surles, 1985; Dzou et al., 2000).The uppermost Washita section comprises the 170–200-ft-thick (52–61-m) Buda Limestone, a deepershelf limestone, and the overlying thick (60–75 ft,

18–29 m) Maness Shale. The Woodbine Group isdivided into the Dexter sand (dominated by sand-stone) and the younger and relatively shalier Lewis-ville Formation.

The Woodbine Group comprises on-shelf fa-cies, mostly fluvial-deltaic deposits, in the East Tex-as Basin (Oliver, 1971) (Figure 3). The WoodbineGroup reaches a maximum thickness of approxi-mately 890 ft (�270 m) in the basin and thins grad-ually toward the Sabine uplift, where it is as muchas 250 ft (76 m) thick at the west (downdip) edge ofthe East Texas field. In the southern part of the ba-sin, these deposits grade into primarily shale (PepperShale) of the outer shelf and slope (Figure 2). Thesouth margin of the Angelina-Caldwell flexureand adjacent Edwards reef trend mark the approx-imate location of the Woodbine depositional shelfedge (Figures 1, 3a) (Siemers, 1978), south of whichthe unit produces from deep (>11,000 ft, >3350 m)Woodbine/Eagle Ford–equivalent slope turbiditefacies, such as those in Seven Oaks (Polk County)and Sugar Creek (Tyler County) fields (Siemers,1978; Foss, 1979). The Woodbine is equivalent tothe lower Tuscaloosa Formation of south-centralLouisiana (Mancini and Puckett, 2005); the pro-ductive deep occurrence of the two units com-poses the Tuscaloosa-Woodbine trend (e.g., Dubielet al., 2003).

Early outcrop studies of the Woodbine Grouptentatively recognize it to be at least partly com-posed of deltaic facies (Scott, 1926). Oliver (1971)presented the only comprehensive regional studyof the unit based on outcrop and well-log data. Heinterpreted three principal Woodbine deposition-al systems: meanderbelt fluvial, shelf strand plain,and high-destructive deltaic. His fluvial system inthe Dexter sand of the northern East Texas Basin(Figure 3b) was inferred to be a succession of thicksandstones with blocky log responses (180–380 ft,55–116 m), flanked by coeval flood-plain mud-stones with baseline and serrate log responses. Hisshelf-strand-plain system of the Lewisville Forma-tion, also only in the north half of the basin, com-prises fewer and thinner sandstones. Log facies areupward coarsening, serrate, and blocky. A high-destructive (wave-dominated) deltaic system wasdeposited south of the other systems and is coeval

Figure 2. Regional Lower to Upper Cretaceous stratigraphy ofthe East Texas Basin. The figure depicts the commonly usedlithostratigraphic nomenclature in the basin, which does not nec-essarily coincide with formal usage as defined by the U.S. Geo-logical Survey. Compiled from Childs et al. (1988), Salvador andMuneton (1989), and Sohl et al. (1991).


with both. Major sand accumulations of this sys-tem developed primarily in stacked coastalbarriers with thick (�100 ft, �30 m), blocky,and blocky-serrate log responses.

Within the lower Woodbine Dexter sand in theEast Texas field, Oliver (1971) inferred stacked me-anderbelt sandstones in the northern part of the fieldgrading to sandstones of equivalent wave-dominated

Figure 3. (a) Paleogeographicreconstruction of the northwestGulf Coast Basin during Wood-bine deposition (modified fromSohl et al., 1991). (b) Regionaldepositional-facies distributionof the lower Woodbine (Dextersand equivalent) inferred byOliver (1971). Only the Dextersand facies (and not those ofthe Lewisville Formation of theupper Woodbine) are illustratedbecause it is the unit presentin the East Texas field.

Ambrose et al. 237

deltaic and coastal-barrier systems in the southernpart (Figure 3b). In contrast, we interpret Oliver’sblocky fluvial and coastal-barrier facies to collective-ly compose those of basin-scale, lowstand, incised-valley river systems. Instead of being flanked bycoeval flood-plain mudstones, the valley systemsare eroded into older highstand, fluvial-dominateddeltaic facies. Therefore, as subsequently discussed,we infer no genetic linkage between the fluvial(incised-valley) and deltaic systems in the field.Our interpretation of sandstone trends and faciesreflects a new depositional model and provides amore realistic understanding of reservoir compart-mentalization, fluid flow, flow-unit geometry, andlocation of unswept mobile oil in the field.

SEQUENCE-STRATIGRAPHIC FRAMEWORK

Regional Chronostratigraphy

The Woodbine Group was deposited during a ma-jor middle and late Cenomanian regressive eventfollowing a pronounced lowering of the relative sealevel after Buda and before Woodbine depositionthat affected the entire Gulf Coast Basin (Salvador,1991; Mancini and Puckett, 2005). This relativesea level fall exposed shelves and platforms aroundthe basin to subaerial exposure (Salvador, 1991).Embayments, such as the incipient East Texas Ba-sin, however, remained mostly submerged. Thehiatus recorded by the middle Cenomanian uncon-formity (Salvador, 1991) occurs at the top of theBuda shelf limestone (SB 10), which is conform-ably overlain by the deeper water, transgressiveManess Shale (Figure 4) throughout the basin andits eastern flank. The unconformity is inferred to

coincide with the third-order sequence boundary(SB) at approximately 94 Ma defined by Haq et al.(1988) (Figure 4). The subaerial exposure of theBuda Limestone did not occur in the east Texasarea (Salvador, 1991) most likely because its rateof subsidence exceeded that of the general GulfCoast Basin. Syndepositional salt mobilization, aprincipal mechanism for Cretaceous subsidencein the East Texas Basin (Seni and Jackson, 1984),created more accommodation space than couldbe countered by the eustatic fall, thus preservinga fully submerged middle Cenomanian Basin inthe east Texas area. Middle Cenomanian (pre-Woodbine) uplift along the north margin of theGulf Coast Basin (Sohl et al., 1991), also termedthe southern Arkansas uplift (Ewing, 1991a, b), pro-vided the catalyst for subsequent coarse-siliciclasticWoodbine deposition.

The top of the Buda Limestone represents anSB and transgressive surface of erosion (TS) and ei-ther a surface of nondeposition or a very low rate ofdeposition above a carbonate highstand systemstract. The thick retrogradational lower half of theoverlying Maness Shale (�50 ft [�15 m] in most ofthe basin, thinning to�35 ft [�11 m] at the Sabineuplift) records the subsequent third-order trans-gressive systems tract, and its associated maximumflooding surface (MFS 10) is dated at approximate-ly 93.5 Ma (Haq et al., 1988) (Figure 4). Micropa-leontologic studies of the Maness Shale (Loeblichand Tappan, 1961) and lower Eagle Ford Group(Pessagno, 1969; Christopher, 1982), integratedwith regional chronostratigraphic compilations(e.g., Childs et al., 1988), support these ages andtermination of Woodbine deposition at approxi-mately 92 Ma. Some imprecision, however, existsin the correlation of these formation-specific dates

Figure 4. Coastal-onlap curve derived from Haq et al. (1988) correlated with the Cenomanian and Turonian succession in the EastTexas Basin. All ages of third-order sequence boundaries are inferred to coincide with those of Haq et al. (1988), except that at thetop of the Woodbine succession (�92 Ma). Haq et al.’s global analysis places the age at approximately 93 Ma, thereby giving theirmajor overlying early Turonian third-order transgressive systems tract a duration of approximately 1.5 Ma (not �0.5 Ma as depictedhere). We infer the variation from the global average to be caused by local tectonic influences onWoodbine deposition. Continuing high-siliciclastic sediment supply and creation of accommodation space in the subsiding East Texas salt basin enabled on-shelf, keep-upWoodbine depositional conditions to extend to the end of the Cenomanian. Compiled from Childs et al. (1988), Salvador and Muneton(1989), and Sohl et al. (1991). Surfaces depicted at the preserved tops of highstand systems tracts in the left log (e.g., sequence boundary[SB] 30) are both SBs and transgressive surfaces (TSs). SB 100 is locally cut out by SB 110. Well location coincides with well 1 in Figure 5a.GR = gamma ray; SP = spontaneous potential; Res = resistivity; MFS = maximum flooding surface; TS = transgressive surface.

Ambrose et al. 239

(and others given in Figure 4) with the worldwideages presented by Haq et al. (1988), and our agecorrelations should be considered to be a best-fiteffort using the sparse micropaleontologic data forthe Upper Cretaceous of east Texas that exist inpublished sources. The top surface of our study in-terval, an MFS 150 above a thick shaly transgressivesystems tract in the basalmost Eagle Ford section(Figure 4) (�130 ft [�40 m] in most of basin,thinning to �30 ft [�9 m] at the Sabine uplift), rep-resents the major early Turonian third-order MFSand the widespread Cenomanian–Turonian anoxicevent (e.g., Herbin et al., 1986; Arthur et al., 1987)dated at approximately 91.5 Ma by Haq et al. (1988).The Woodbine succession was, therefore, depositedover a period of approximately 1.5 m.y. (�93.5–92 Ma) as a single third-order sequence. Each fourth-order sequence composing the Woodbine was de-posited over an average of approximately 110 k.y.

On the basis of a sequence-stratigraphic analy-sis of approximately 225 well logs distributed acrossthe central and south-central parts of the East Tex-as Basin (Figure 1), we identified a maximum of 14fourth-order sequences within the greater Wood-bine succession (top of Buda Limestone to lower-most Eagle Ford Group) in the central part of thebasin. Sequence boundaries, TSs , and MFSs withinthe sequences were inferred primarily from thelogs’ gamma-ray signatures (Figure 4), supportedby whole-core data from the field area. The Wood-bine succession in the basin is dominated by in-ferred lowstand incised-valley fills, which occur aslow-gamma-ray zones as much as 150 ft [45 m]thick forming blocky log responses with abrupt hori-zontal basal (erosional) surfaces. Sequence bound-aries at the base of the incised-valley fills werecorrelated with those at the tops of adjacent upward-coarsening cycles, which represent older highstandsystems tracts into which the valley fills were cut.

Transgressive surfaces of erosion define tops of theaggradational valley fills and coincide with sequenceboundaries atop the highstand systems tracts. Maxi-mum flooding surfaces cap upward-fining succes-sions (transgressive systems tracts) at gamma-raymaxima above the lowstand incised-valley-fill andhighstand successions.

Sequences, Systems Tracts, and Relation toSabine Uplift

The relative rise of the Sabine uplift was initiatedsoon after deposition of the Maness Shale, and weinfer that it was a gradual and continuous process,not episodic, during the entire period of Wood-bine and Eagle Ford deposition (�2.5 m.y). Its de-velopment most likely coincides with that of themiddle Cenomanian southern Arkansas uplift thataffected most of south Arkansas and north Loui-siana (Ewing, 1991a, b). Halbouty and Halbouty(1982), however, deduced that rejuvenation of theSabine uplift occurred as a major, short-term epi-sode, resulting in the development of a regionalangular unconformity after Buda, but before Wood-bine, deposition. Jackson and Laubach (1991) sim-ilarly invoked the formation of a sub-Woodbineangular unconformity by a pre-latest Early Creta-ceous upwarp. Sequence-stratigraphic relations inour study do not support these conclusions andenable a more focused interpretation of the struc-tural influences and their timing on Woodbine de-position in the East Texas Basin and field.

The highstand Buda Limestone carbonates, re-gionally transgressive lower Maness Shale, and basalhighstand upper Maness Shale are each isochronous,maintaining nearly consistent thickness (system-atically thickening by�25% toward the protobasincenter), except where they are truncated by themiddle Turonian base-of-Austin unconformity along

Figure 5. (a) Lithostratigraphic cross section from the axis of the East Texas Basin to the Sabine uplift. The eastern four wells lie on thewest flank of the Sabine uplift. The datum is the top of the Austin Chalk. (b) Schematic rendering of the sequence-stratigraphic frameworkof Woodbine Group, East Texas Basin, and Sabine uplift. The datum is the top of the Woodbine Group. Systems-tract interpretations fromlog facies of the eastern two-thirds of the cross section (wells 3–6) are shown in Figure 6. Systems tracts of well 1, which records themaximum number of fourth-order sequences (14) composing the Woodbine in the deep axial part of the East Texas Basin, are depictedin Figure 4. The line of section closely coincides with that of a. The shaded area in the index maps is the East Texas field. No horizontalscale. GR = gamma ray; SP = spontaneous potential; Res = resistivity. SW = SW Operating, Inc.; NOE = T.C. Noe Oil Account.

Ambrose et al. 241

the flank and crest of the Sabine uplift (Figures 5a,6). Because the Buda and Maness successions ex-hibit no significant internal stratal changes on theSabine uplift, this structure exhibited little or no

relief during deposition of these units. Ewing(1991a) similarly concluded that the uplift is notexpressed in sub-UpperCretaceous (pre-Woodbine)strata.

Figure 6. Details of sequence-stratigraphic and systems-tract framework of Woodbine Group in the eastern East Texas Basin–westernSabine uplift area. The datum is sequence boundary (SB) 10. Wells coincide with the eastern four wells (well 3–6) shown schematicallyin Figure 5b. Two particularly salient features are illustrated: (1) only the oldest five fourth-order sequences S1–S5 (SB 10–60) aretruncated by the base-of-Austin unconformity, and (2) all younger Woodbine sequences depositionally pinch out below the shalythird-order transgressive systems tract (capped by maximum flooding surface [MFS] 150) of the lowermost Eagle Ford Group. Nohorizontal scale. GR = gamma ray; SP = spontaneous potential; Res = resistivity. NOE = T.C. Noe Oil Account.


The Woodbine succession gradually thins fromthe axis of the East Texas Basin westward to theMexia-Talco fault zone and eastward to the Sabineuplift. This thinning coincides with a systematic de-crease in the number of fourth-order Woodbine se-quences in both directions away from the deep axialpart of the basin, where 14 sequences are present.The oldest five sequences extend to the west flankof the Sabine uplift and are truncated by the base-of-Austin unconformity beginning approximately10 mi (16 km) west of the downdip edge of the EastTexas field (Figures 5b, 6). At the westernmostextent of the unconformity, it incises the Eagle Fordstrata and progressively cuts downward stratigraphi-cally toward the uplift (Figure 6). The generallythinner upper nine sequences, in contrast to theolder sequences, are conformably overlain by theEagle Ford mudstones and therefore depositionallypinch out between eastern Henderson and west-central Smith counties more than 10 mi (>16 km)from the field.

No more than the oldest five fourth-orderWoodbine sequences (S1–S5) were ever depositedin the area of the Sabine uplift (Figure 5b). Minorthinning of these sequences toward the uplift re-cords an incipient relative rise of the structure dur-ing earliest Woodbine deposition (Figure 7). Ceno-manian salt mobilization (dome growth) in theEast Texas Basin coincided with rapid accumula-tion of voluminous Woodbine siliciclastic sedi-ments (Seni and Jackson, 1984) and helped sustainthe concurrent relative subsidence and uplift. Be-ginning about the time of deposition of S5 or S6,deposition of the upper Woodbine (sequences S6–S14) was limited to the central part of the EastTexas Basin, a zone approximately 35 mi (�56 km)wide centered on the basin axis (Figures 1, 5b). Re-striction to basin-center deposition records the pe-riod during which the Sabine uplift probably hadbecome a positive (emergent) feature and startedto influence Woodbine depositional patterns. LowerWoodbine sequences S1–S5 were likely exposedon the uplift and composed a proximal, secondarysediment source for the younger, onlapping Wood-bine deposits (S6–S14) (Figure 7). Unlike previ-ous workers, most notably Halbouty and Halbouty(1982), we do not infer short-term major episodes of

Figure 7. Schematic depictions of the development of the Sabineuplift and adjacent East Texas Basin during (a) early Woodbine(S1–S5) deposition, (b) late Woodbine (S6–S14) deposition, and(c) Austin Chalk deposition. The area depicted is approximately50 mi (80 km) wide; vertical exaggeration is significant. ETOF =East Texas oil field.

Ambrose et al. 243

rise of the Sabine uplift immediately before Wood-bine deposition and during late Woodbine deposi-tion. The sequence-stratigraphic architecture of theunit documents a fairly consistent rate of subsi-dence and creation of accommodation space with-in the basin throughout accumulation of lower andupper Woodbine sequences. An episodic, acceler-ated rise of the adjacent uplift during Woodbinedeposition would have produced prominent sig-natures in the basin’s stratigraphic record, such asbasin-scale angular unconformities and abrupt sub-regional depositional-facies transitions within thicksuccessions (e.g., fan deltas at the uplift grading toshallow-marine deposits in the basin).

The Eagle Ford Group gradually thins towardthe Sabine uplift west of the area where the base-of-Austin unconformity is developed (Figure 5).The uplift was, therefore, also a positive featureduring Eagle Ford deposition and was subaeriallyexposed (Jackson and Laubach, 1988) as it prob-ably was during the preceding late Woodbine (S6–S14). Over an extended period before depositionof the Austin Chalk, however, at least a few hun-dred feet of Washita, Woodbine, and Eagle FordGroup strata had been gradually eroded from theuplift and its west flank. Erosion likely began justbefore or during deposition of the middle Wood-bine (�S5 or S6) and culminated no later than dur-ing deposition of the upper Eagle Ford (middleTuronian). A series of local unconformities and dis-conformities within the Woodbine (Figure 7a, b)and Eagle Ford groups probably developed on theuplift and its flanks, recording the structure’s pe-riod of gradual emergence. These unconformitiesare inferred to have been removed by subsequentuplift and erosion. The base-of-Austin unconfor-mity, therefore, represents a significant hiatus andextended period of stratal erosion from the uplift,perhaps lasting as much as 2 m.y.

Development of the middle Turonian base-of-Austin unconformity does not coincide with anydocumented regional or subregional tectonic activ-ity. This development predates regional Laramidecompression, which most likely later affected theSabine uplift (Jackson and Laubach, 1991) and thegreater Gulf Coast Basin beginning in the latestLate Cretaceous (Maastrichtian) (Salvador, 1991).

Culmination of sub-Austin Chalk erosion of theuplift most likely occurred when termination of theperiod of high rate of siliciclastic sedimentation inthe East Texas Basin coincided with the third-ordermiddle Turonian eustatic lowstand immediatelyfollowing Eagle Ford deposition at approximately91 Ma (Haq et al., 1988) (Figures 4, 7c). Relativesea level fall intensified erosion and beveling of theuplift area, accentuating the stratigraphic expres-sion of the resulting SB (unconformity). Minimalthinning of the Austin Chalk (Figure 5) and over-lying Upper Cretaceous carbonates and shales overthe uplift, also documented by Granata (1963),indicates that the relative rise of the structure hadnearly stopped before Austin deposition.

Only the oldest three fourth-order sequences(S1–S3) of the lower Woodbine Group are pre-served below the base-of-Austin unconformity in theEast Texas field (Figures 5b, 6). Detailed mappingof principal sandstone bodies and description ofwhole cores from the NPA and SPA (Figure 1) al-lowed the interpretation of depositional facies andkey depositional surfaces and provided a corrobo-ration of our sequence-stratigraphic interpretation.

Key Chronostratigraphic Surfaces: East Texas Field

Although the same sandstone-bearing sequencesare represented in the NPA and SPA, key differ-ences in their log expression, core attributes, anddepositional facies exist. These differences pro-foundly affected (and continue to affect) their res-ervoir characteristics and production histories. Weherein provide an initial summary of these basicdifferences. Throughout the NPA, the S3 lowstandincised-valley fill (Main sand of operators) overliesthe S1 highstand and transgressive (Maness Shale)systems tracts and SB 10 (top of Buda Limestone)(Figure 8a). The entire S2 succession has beenremoved by valley incision in this area. The base-of-Austin unconformity truncates the top of the S3incised-valley fill across the NPA. Our regional se-quence correlations show that this succession char-acterizes the entire west margin of East Texas field(Figures 6, 8, 9). We infer that it provides a recordof increased regional downcutting to adjust to alower base level created by the initial relative rise


Figure 8. (a) Representative structural-dip cross section of the north pilot area (NPA), showing inferred fourth-order sequence-stratigraphic surfaces and systems tracts. The datum is the base-of-Austin unconformity. No horizontal scale. (b) Representativestructural-dip cross section of the southwestern part of East Texas field and the south pilot area (SPA), showing inferred fourth-ordersequence-stratigraphic surfaces and systems tracts. The datum is the base-of-Austin unconformity. Local lower Woodbine (LWB)chronostratigraphic surfaces LWB 10–40 depicted in SPA wells. LWB 10 coincides with regional maximum flooding surface (MFS) 10.See Figure 1 for the location of the NPA and SPA within the East Texas field. No horizontal scale. GR = gamma ray; SP = spontaneouspotential; Res = resistivity; SB = sequence boundary. TXOK = TXOK Energy Resources Co., EOG = EOG Resources, Inc.

Ambrose et al. 245

of the Sabine uplift during earliest Woodbine (S3)deposition. The Woodbine section of the SPA oc-curs just east of the approximate depositional limit

of the S3 incised-valley-fill system and comprisesmost or all of the S1 highstand deltaic succession(MFS 10–SB 20) (Figures 8b, 9).

Figure 9. (a) Regional areal extent of the S3 lowstand incised-valley fill (SB 30–TS 30, Main sand of operators), (b) cross section ofwest the edge of the incised-valley fill (line of section shown in a; the datum is TS 30), and (c) areal extent of the incised-valley fill inEast Texas field. The cross-sectional view of the eastern depositional or erosional limit of the valley fill is depicted in Figure 8b. Nohorizontal scale. NPA = north pilot area; SPA = south pilot area; GR = gamma ray; SP = spontaneous potential; Res = resistivity; SB =sequence boundary; MFS = maximum flooding surface.


Sequence Boundaries and Transgressive Surfacesof ErosionSequence boundaries are the most significantWoodbine surfaces that define sandstone units andreservoir-facies trends in the pilot areas. In theNPA, they include SB 10 at the top of the BudaLimestone (middle Cenomanian unconformity),SB 30 (base of incised-valley fill), and the base-of-Austin unconformity. Well represented in wholecores from wells just outside the area (Figure 1),SB 30 marks the boundary between the field’s pri-mary reservoir (lowstand fluvial Main sand) fromthe underlying highstand deltaic stringer zone, theprimary completion target today. The distinctly ero-sional surface occurs as fluvial chert-clast conglom-erate or granular, coarse-grained sandstone overly-ing distal-delta mudstones and siltstones, recordinga pronounced drop in relative sea level (Figures 10,11, 12a, b). At the top of shelf limestones of theBuda Limestone throughout the field, as in thebasin, SB 10 exhibits no evidence of erosion and isin conformable contact with the overlying ManessShale (e.g., Figure 10). The base-of-Austin uncon-formity in the NPA and SPA, the middle Turo-nian third-order SB (Figure 4), marks the base ofthe top seal in the field and records a TS. Chert,milky quartz, and other clasts derived from incised-valley-fill conglomerates of the underlying Wood-bine were incorporated in the basal part of thetransgressive Austin Chalk in the NPA (Figure 12c).Well-developed paleosols in lower Woodbine fa-cies immediately below the Austin contact in theSPA and in cores outside the field record a periodof long-term subaerial exposure associated withthis unconformity (Figure 13b). Note that SB 20exists only in the SPA, where incision by SB 30 didnot occur, and even in this area, the surface is pre-served only in the western (downdip) part of thearea (Figure 8b).

Flooding SurfacesThe MFS 10 in the lower Maness Shale (Figures 6,10) is significant because it represents a regionalchronostratigraphic horizon that extends into EastTexas field. One of the local flooding surfaces(lower Woodbine [LWB] 10, Figure 8b) that weused to correlate coeval reservoir sandstone units

in the SPA coincides with MFS 10. These four lo-cal horizons are inferred to represent surfaces ofdelta-lobe abandonment or avulsion (Figure 8b)that bound three sandstone-bearing zones (LWB10–20, LWB 20–30, LWB 30–40) in the upward-coarsening S1 highstand systems tract (MFS 10–SB 20). Although these flooding surfaces are hy-pothesized to reflect the avulsion of major deltaiclobes, nodal points of avulsion are not documentedin this study because a larger area for detailed net-sandstonemaps than that shown in theSPA(Figure1)would be required to fully define the geographicextent of major deltaic lobes in the S1 highstandsequence.

Net-sandstone values were calculated for eachzone in the S1 highstand deltaic succession, whichwere mapped separately to identify reservoir-faciestrends and depositional facies in the pilot areas.Units are generally isopachous in the SPA, althoughlocally developed thicker sandstones occur and areinferred to represent distributary-channel fills (sub-sequently discussed). The same succession in theNPA exists between MFS 10 and the base of thelowstand valley fill (SB 30), but its thickness variesaccording to differences in depth of valley incision(Figure 8a).

DEPOSITIONAL SYSTEMS IN EASTTEXAS FIELD

Highstand Fluvial-Dominated Deltaic System

Cores recording the Woodbine’s lowest fourth-order sequence (S1), herein referred to as Woodbinestringers (after operators’ usage), from the centraland southern parts of the field (Figure 1) exhib-it many sedimentary features typical of fluvial-dominated deltaic systems. For example, a corefrom the Shell 55 Watson well in the central partof the field contains a nearly complete successionof the stringer interval, from the top of the BudaLimestone at 3728 ft (1136 m) to the base of theS3 incised-valley fill at approximately 3625 ft(�1105 m) (Figure 10). This overall upward-coarsening succession is interpreted to include

Ambrose et al. 247

Figure 10. (a) Core description and log response of the S1 highstand (lower Woodbine stringer) section and the lower part of theS3 lowstand incised-valley-fill section in the Shell 55 Watson well. (b) Distal-delta-front siltstones and very fine-grained sandstones at3698 ft (1127 m). (c) Proximal-delta-front very fine-grained, rippled siltstones and sandstones at 3687 ft (1124 m). (d) Reddish-brown, carbonaceous mudstone overlain by a zone with abundant pedogenic mottling at 3661 ft (1116 m). The well is located inFigure 1. SP = spontaneous potential; Res = resistivity; SB = sequence boundary; MFS = maximum flooding surface. CMT = cement.


Figure 11. Core description and photographs of S3 incised-valley bed-load fluvial deposits directly overlying very fine sandstonesand siltstones from highstand delta-front deposits in the Cities Service B2 Killingsworth well. (a) Description and log response from3610 to 3646 ft (1100 to 1111 m). (b) Contact between gravelly and pebbly sandstone and burrowed, very fine-grained sandstone at3636 ft (1108 m). (c) Stacked gravel-bar deposits at 3629 ft (1106 m). (d) Coarse-grained sandstone with large-pebble clasts at 3614 ft(1101 m). Well is located in Figure 1. GR = gamma ray; Res = resistivity; SB = sequence boundary. CMT = cement.

Ambrose et al. 249

Figure 12. Core description and photographs of S3 incised-valley bed-load fluvial deposits directly overlying very fine-grainedsandstones and siltstones from highstand delta-front deposits in the Arco 142B King well. (a) Description and log response from 3420to 3450 ft (1042 to 1051 m). (b) Erosional contact between gravel beds and cross-bedded, coarse-grained sandstone at 3437.4 ft(1047.7 m). (c) Transgressive-lag deposits at the base of the Austin Chalk at 3422.4 ft (1043.1 m). The well is located in Figure 1. GR =gamma ray; Res = resistivity; SB = sequence boundary. CMT = cement.


Figure 13. Core description and photographs of stacked distributary-channel deposits of the S1 highstand succession in the Arco 18Griffin well. (a) Description and log response of interval from 3570 to 3655 ft (1088 to 1114 m). (b) Paleosol developed below the base-of-Austin unconformity (3574.5 ft, 1089.5 m). (c) Cross-bedded, upper-fine sandstone with scour surfaces and soft-sediment deforma-tion at 3599 ft (1097 m). (d) Rippled and scoured, organic-rich fine sandstone with dewatering structures at 3588.5 ft (1093.7 m). Well islocated in Figure 17. GR = gamma ray; Res = resistivity. CMT = cement.

Ambrose et al. 251

prodelta, delta-front, delta-plain, and distributary-channel facies. These facies and other geneticallyrelated facies from other cores in the field areaare described separately. The upper part of theShell 55 Watson core (3600–3625 ft, 1097–1105 m), distinctly coarser grained than the un-derlying section, consists of medium-grained andcoarse-grained, gravelly sandstone interbedded withthin siltstones and very fine-grained sandstones(Figure 10). This upper section is interpreted torepresent nonmarine, bed-load fluvial deposits,which are discussed in the section titled LowstandIncised-Valley Fluvial System.

Reservoir sandstones of the Woodbine stringers(for example, oil-stained sandstones in proximal-delta-front and distributary-channel sandstones inFigure 13) are interpreted to have been depositedin the S1 highstand deltaic succession (Figure 14).Inferred dimensions of sandstone bodies and inter-pretations of facies relationships in both the S1highstand deltaic and S3 incised-valley-fill succes-sions are based on detailed net-sandstone mapsand cross sections in the NPA and SPA areas (seeFigures 15–18 in the following sections of this re-port). They were deposited in high-constructive,fluvial-dominated deltaic systems, with narrow,dip-elongate distributary-channel sandstones ex-hibiting southward-bifurcating net-sandstone pat-terns. Distributary-channel depositional axes areflanked locally by lobate crevasse-splay depositsinferred to pinch out into interdistributary-bay anddelta-plain siltstones (Figure 14a). Sandbodies inmodern fluvial-dominated deltaic systems are dis-tributed in narrow distributary channels that bi-furcate seaward, forming bar-finger sands (Fisk,1961; Galloway, 1975). These distributaries arecommonly flanked by abundant crevasse splaysformed by levee breaching and subsequent infill-ing of the muddy interdistributary bay (Colemanand Gagliano, 1964; Elliott, 1974). The continuityof sandstone bodies in fluvial-dominated deltas istypically poor to moderate because of the presenceof narrow distributaries and lobate crevasse splaysthat pinch out into muddy interdistributary areas(Coleman and Wright, 1975). Sandstone hetero-geneity in fluvial-dominated deltas is also increasedby abundant soft-sediment deformation related

to compaction of distributary-channel sandstonesinto soft delta-front mudstones, as well as dewater-ing of delta-plain sediments (Frazier, 1967; Kellingand George, 1971; Brown et al., 1973). Early dia-genetic siderite (e.g., in the lower part of the coredinterval in Figure 10) may be locally abundant, asobserved in studies of other fluvial-dominated del-taic systems by Coleman and Prior (1982) andBhattacharya and Walker (1991).

Prodelta, Delta Front, and Channel-Mouth Bar

Prodelta, delta-front, and channel-mouth-bar depos-its together comprise the most commonly observedfacies in cores from the S1 highstand deltaic suc-cession in the East Texas field. Prodelta and inner-shelf deposits in the Shell 55 Watson core, extend-ing from approximately 3710 to 3728 ft (�1131to 1136 m) (Figure 10), are composed of calcare-ous, laminated mudstone interbedded with sparse,thin (<1 in., <2.5 cm) beds of very fine-grainedsandstone and coarse-grained siltstone with sharpbases and burrowed and rippled tops. This sectiongrades upward into an upward-coarsening succes-sion of siltstones interbedded with rippled, veryfine-grained sandstone beds that extend from ap-proximately 3684 to 3710 ft (�1123 to 1131.1 m).The lower part of this upward-coarsening sec-tion consists commonly of sideritic siltstone withabundant interbeds of very fine-grained sandstone(Figure 10a). Many of these sandstone beds havesharp but undulating bases and are laminated andrippled; some sandstone beds contain small clayclasts. These thin, sharp-based sandstones are in-terpreted to record pulses of sediment-laden dis-charge from the distributary mouth, representingdeposits of dilute turbidity currents in the deltafront. They are similar to thin, erosive-based, coarse-grained siltstones and very fine-grained sandstonebeds in fluvial-dominated deltaic deposits in theWestphalian in the United Kingdom (De Raafet al., 1965), Upper Cretaceous–lower Paleocenedeltaic deposits in northwest Mexico (McBrideet al., 1975), and the modern Yellow River delta(Wright et al., 1988). In contrast, the upper part ofthe upward-coarsening section (�3684–3693 ft,


Figure 14. Block diagrams summarizing sequences and depositional systems in the lower Woodbine Group in the East Texas field.(a) S1 highstand fluvial-dominated deltaic deposits in the basal lower Woodbine section. (b) Truncation of the S1 highstand (preserved)and S2 (removed) successions by the S3 lowstand incised-valley system comprising multistoried, coarse-grained bed-load fluvialdeposits. Inferred dimensions of sandstone bodies and interpretations of facies relationships in both the S1 highstand deltaic and S3incised-valley-fill successions are based on detailed net-sandstone maps in the NPA and SPA (Figure 1).

Ambrose et al. 253

Figure 15. Core description and photographs of delta-front and channel-mouth-bar deposits of the S1 highstand succession in the ArcoC-21 Pinkston well. (a) Description and log response of cored interval from 3658 to 3708 ft (1115 to 1130m). (b) Sharp-based, distal-delta-front sandstone bed at 3689.3 ft (1124.4 m). (c) Slump in medial-delta-front sandstone bed at 3675.6 ft (1120.3 m). (d) Photograph of cross-bedded channel-mouth-bar sandstone at 3669 ft (1118m). The well is located in Figure 17. GR = gamma ray; Res = resistivity. CMT = cement.


Figure 16. Core description and photographs of crevasse-splay deposits of the S1 highstand succession in the Arco C19 Pinkstonwell. (a) Description and log response of the interval from 3580 to 3625 ft (1091 to 1105 m). (b) Current ripples in lower-splaydeposits at 3611.8 ft (1100.8 m). (c) Deformed ripples and small-scale cross-beds in upper-splay deposits at 3605.7 ft (1099 m). Thewell is located in Figure 17. SP = spontaneous potential; Res = resistivity. CMT = cement.

Ambrose et al. 255

Figure 17. Net-sandstone maps of Woodbine progradational parasequences of the S1 highstand stringer sandstones in the SPA:(a) LWB 30–40, (b) LWB 20–30, (c) and LWB 10–20 intervals, respectively. The SPA is located in Figure 1. Core descriptions andphotographs of selected S1 intervals in the Arco 18 Griffin, Arco C-21 Pinkston, and the Arco C-19 Pinkston wells are shown inFigures 13, 15, and 16, respectively.


Figure 18. (a) Net-sandstone mapof the undivided lower Woodbine stringersandstones (S1 highstand) in the NPA.(b) Net-sandstone map of the S3 low-stand incised-valley-fill system in the NPA.The location of NPA is shown in Figure 1.The northwest–southeast stratigraphicsection in the northwestern part ofthe NPA is shown in Figure 20.

Ambrose et al. 257

�1123–1125 m) consists of ripple-bedded, brownto purple siltstone and very fine-grained sandstone(Figure 10b). The section is sparsely burrowed byPlanolites. This upward-coarsening section of silt-stones and ripple-laminated, very fine sandstonesin the Shell 55 Watson core from 3684 to 3710 ft(1123 to 1131 m) is interpreted to represent delta-front deposits.

A similar vertical facies succession is observedin the Arco C-21 Pinkston core (Figure 15). Thelower part of the succession consists of thin (com-monly 1–4 in. [2.5–10 cm] beds of very fine,sparsely burrowed and laminated sandstone inter-bedded with fine siltstone. Individual sandstonebeds in the lower section are commonly sharp basedand laminated (Figure 15b), and some contain small,millimeter-scale clay clasts and Planolites bur-rows. Higher in the delta-front section, individualvery fine to fine-grained sandstone beds range inthickness from 6 in. to 1 ft (15 cm to 0.3 m). Strat-ification in these sandstone beds consists of ripplesand horizontal laminations commonly distorted byload structures and slumps (Figure 15c). The upperpart of the succession features 1–2-ft (0.3–0.6-m)fine sandstone beds with horizontal and low-angle,inclined laminations; internal scour surfaces; andcross-beds, although soft-sediment deformation hasobscured some stratification (Figure 15d).

Delta Plain, Crevasse Splay, and Distributary Channel

Woodbine delta-plain facies in the East Texas fieldconsist of interbedded organic-rich sandstonesand siltstones. In the Shell 55 Watson core, delta-plain and crevasse-splay deposits are interpretedto extend from 3640 to 3680 ft (1109 to 1122 m)(Figure 10). Intermittently exposed soil horizonsand peat swamps that formed on splay platformsare illustrated by a 6-in. (15-cm) zone of massivelydeformed and mottled organic mudstone and silt-stone with abundant carbonaceous filaments at ap-proximately 3661 ft (�1116 m) (Figure 10d).Crevasse-splay deposits in other Woodbine coresconsist of 5–10-ft (1.5–3-m), upward-coarseningintervals of laminated siltstone grading upward intomillimeter-scale, very fine, ripple-laminated sand-stone beds interbedded with siltstone (Figure 16a,

b). The log response of the crevasse-splay facies istypically upward coarsening and serrate (Figure 16a),reflecting an upward decrease in clay and muddymatrix. The top of the crevasse-splay section is com-posed typically of organic-rich, rippled and cross-bedded, fine sandstone with multiple internal scoursurfaces and soft-sediment deformation (Figure 16c),features commonly observed in crevasse-splay de-posits in the modern Mississippi delta plain (Cole-man et al., 1964; Arndorfer, 1973). Splay deposits,inferred from net-sandstone maps of individual S1highstand parasequences in the SPA, are digitatein shape and extend less than 2500 ft (<760 m)from inferred distributary-channel depositionalaxes (see the section titled Net-Sandstone Geom-etry: South Pilot Area).

Individual Woodbine distributary-channel de-posits are typically composed of 10–20-ft (3–6-m)intervals of erosion-based, cross-bedded and plane-bedded, fine sandstone fining upward into very fine,rippled, and finely laminated sandstone (Figure 13a).These individual channel-fill deposits commonlyoccur in multistoried complexes up to 40 ft (12 m)thick. Channel-fill successions are internally com-plex, with multiple internal scour surfaces, clayclasts, dispersed organic fragments, and abundantsoft-sediment deformation (Figure 13b). Stratifi-cation in the lower part of the channel fill is dom-inantly cross-bedded, low-angle, inclined lamina-tions and large-scale ripples, whereas the upperpart is marked by asymmetric current ripples withmudstone drapes, fine-scale laminations, and dewa-tering structures (Figure 13c). Similar distributary-channel successions are observed in the modernNiger Delta (Oomkens, 1967, 1974), Mississippidelta (Frazier, 1967), and lacustrine deltas in theAtchafalaya Basin (Tye and Coleman, 1988), as wellas in outcrops from Pennsylvanian deltaic depositsof the Eastern shelf of the Midland Basin (Brownet al., 1973; Brown et al., 1990).

Net-Sandstone Geometry: South Pilot AreaA series of net-sandstone maps of three Woodbinefluvial-dominated deltaic parasequences, the LWB10–20, LWB 20–30, and LWB 30–40 intervals(Figure 8b), illustrate sandstone-body and inferredfacies geometry of a progradational system in the


SPA (Figure 17). Maps of successively youngerWoodbine stratigraphic units in the SPA exhibitincreasing net-sandstone thickness patterns, fromwhich a southward progradational trend is inferred.The Woodbine LWB 40–30 interval, a prograda-tional parasequence in the upper part of the Wood-bine offlapping deltaic succession, provides an ex-ample of proximal fluvial-dominated deltaic faciesin the SPA (Figure 1). The sandstone-body geom-etry of this interval, defined by net-sandstone con-tours greater than or equal to 30 ft (�9 m), is dipelongate and southward bifurcating (Figure 17c).Narrow depositional axes (commonly <1000 ft[<300 m]) pinch out laterally (eastward and west-ward) and distally (southward) into relatively mud-dier deposits, with net-sandstone values locallyless than 5 ft (<1.5 m). Depositional axes in theLWB 40–30 interval comprise multistoried suc-cessions of erosionally based fine sandstone con-taining cross-beds and ripples, as well as abundantsoft-sediment deformation (Arco 18 Griffin core[Figure 13b]; the location is shown in Figure 17c).

The LWB 40–30 interval is interpreted to repre-sent multiple crevasse-splay and interdistributary-bay deposits within a larger fluvial-dominated del-taic complex (Figure 17c). In this detailed area,individual elongate sandstone-body trends definedby values of more than 30 ft (>9 m) extend overdistances of only 1500–2500 ft (460–760 m), sug-gesting that they represent individual splay-channeldeposits within subdelta complexes that are com-parable in morphology but smaller in scale thanthose occurring in the modern Mississippi Riverdelta (cf. Coleman and Gagliano, 1964).

Net-Sandstone Geometry: North Pilot AreaThe undivided S1 highstand stringer succession inthe NPA is interpreted to represent updip, fluvialto lower delta-plain feeder systems for deltaic de-posits in the SPA. A net-sandstone map of this in-terval in the NPA, which consists dominantly ofthe LWB 20–30 interval, partly reflects trunca-tion by S3 lowstand incised-valley-fill deposits(Figure 18a). Although this interval is incompletebecause of truncation, sufficiently coherent andwell-defined net-sandstone patterns exist to infera dip-elongate sandstone-body geometry, with de-

positional axes corresponding to values of morethan 10 ft (>3 m). These dip-elongate depositionalaxes display both anastomosing patterns in thesouthwestern part of the NPA, whereas tributarynet-sandstone patterns are inferred in north andnorthwest areas (Figure 18a). However, variationsin degree of incision by the overlying section couldaccount for some of these anastomosing and trib-utary patterns.

Lowstand Incised-Valley Fluvial System

The S3 lowstand incised-valley fill in the East Texasfield consists of multiple successions of chert- andquartz-clast conglomerates and conglomeratic tocoarse sandstones that grade upward into fine tocoarse sandstone. These successions, observed incores from the northern part of the field, are inter-preted to represent multistoried fluvial-channeldeposits within the incised-valley interval. In theEast Texas field they truncate prodelta and distal-delta facies of the older highstand deltaic deposits(Figure 14b). The contact (a regional unconformity)and the valley-fill system itself can be correlatedmore than 35 mi (>56 km) into the East TexasBasin. The valley fill is truncated throughout mostof the field by the base-of-Austin unconformity.Where the entire interval is preserved immediatelydowndip (west) of the field, however, the valleyfill is as much as 140 ft (43 m) thick, recording arelative sea level drop of no less than 215 ft (65 m)(corrected for a conservative 35% porosity loss fromburial compaction alone) (e.g., Houseknecht, 1987).This magnitude of sea level fall seems high underthe Cretaceous greenhouse conditions that existed.Eustatic fall, however, was most likely enhancedby a rate of subsidence in the East Texas Basin thatexceeded that in the adjacent Gulf Coast Basin andother global Cenomanian basins, mostly because ofsyndepositional salt mobilization (Seni and Jackson,1984) and concurrent rapid Woodbine sedimentinflux.

Channel-fill successions of modern bed-loadfluvial deposits are commonly composed of sandyand gravelly longitudinal and transverse bars thatform by downstream migration in braided-river sys-tems (Ore, 1963; Smith, 1970; Boothroyd, 1972;

Ambrose et al. 259

Smith, 1974). Most channel-fill deposits in thesecoarse-grained fluvial systems are caused by migra-tion of sand and gravel bars on the channel floor,with minor slack-water suspension sedimenta-tion of fine-grained material draping the bar forms(Rust, 1972). Woodbine gravel-bar deposits, pre-sent in cores of the Cities Service 2B Killingsworth,Shell 55 Watson, and the Arco B142 King wells,are composed of 30–60 ft (9–18 m) multiple 2–4-ft (0.6–1.2-m) beds of gravel capped by thin(commonly <4 in. [<10 cm]) beds of medium tovery coarse sandstone grading upward into medi-um to very coarse pebbly sandstone (Figure 11). Acore from the Cities Service 2B Killingsworth wellcontains an abrupt, basal contact defined by a 1-ft(0.3-m) bed of gravelly and pebbly, very coarse sand-stone overlying a 10-ft (3-m) section of burrowedsiltstone with thin (centimeter-scale), laminatedand rippled, very fine sandstone (Figure 11b). Peb-bles in this basal, very coarse sandstone are com-posed of subrounded and subangular pink quartziteand microporous (degraded) chert. The overlyingsection consists of multiple 1–2-ft (0.3–0.6-m)gravel beds with thin (1–2-in., 2.5–5.0-cm), coarsesandstone interbeds. Pebble and coarse-granuleclasts in these gravel beds are composed dominantlyof chert with lesser amounts of quartz (Figure 11c).Gravel clasts in some zones display crude imbrica-tion, although this type of stratification is not evi-dent in many other zones. The gravel-dominated,lower section from 3622 to 3637 ft (1104.3–1108.8 m) in the Cities Service 2B Killingsworthwell is overlain by a 12-ft (3.7-m), oil-stained sec-tion composed of laminated and cross-bedded, me-dium sandstone interbedded with a 2.5-ft (0.8-m)bed composed of coarse sandstone with pebbles(Figure 11d). The gamma-ray response of the grav-elly and coarse sandstones in the Cities Service 2BKillingsworth core is sharp based and blocky and ex-hibits a significant leftward deflection with respectto the underlying fine-grained section (Figure 11a).This log response, typical of the Woodbine fluvialincised-valley section in the East Texas field, is afeature on which our interpretation of bed-loadfluvial deposits is partly based.

A short (30 ft, 9 m) section of core from theArco B142 King well includes parts of all three ma-

jor stratigraphic intervals in the East Texas field: theS1 Woodbine stringers, the S3 incised-valley-fillinterval, and the Austin Chalk (Figure 12). As in theCities Service B2 Killingsworth core, a fine-grainedburrowed section of the Woodbine stringer inter-val is truncated by nonmarine, multistoried gravel-bar and sandy channel-fill deposits (Figure 12a, b).Stacking patterns in the valley-fill section are non-systematic, with no net vertical grain-size trend,although much of the upper valley fill is inferredto have been eroded. The Austin Chalk composesthe top 4 ft (1.2 m) of the core. The base-of-Austinunconformity is overlain by large rip-up clasts fromthe underlying Woodbine section, as well as abun-dant shell fragments, reflecting marine transgressionover the exposed Woodbine surface (Figure 12c).

Woodbine S3 incised-valley-fill deposits in theNPA display southwest-trending contour patterns(Figure 18b). Systematic southeastward-decreasingnet-sandstone values in this map reflect bevelingby the base-of-Austin unconformity and resultingsoutheastward thinning of this section toward theSabine uplift (Figure 8a); they do not correspondto original depositional thinning or greater mud-stone content. However, in downdip parts of theNPA, primary depositional axes expressed as tribu-tary patterns containing more than 75 ft (>23 m) ofnet sandstone within a bed-load fluvial valley-fillsystem can be inferred.

DEVELOPMENT STRATEGIES

Although the East Texas field has produced 5.42BSTB and has a high recovery efficiency of 77%, theamount of remaining oil (approximately 1.6 BSTB)is great, partly because of the large original oil inplace (OOIP) of 7.03 BSTB, the poor sweep effi-ciency and variability in reservoir properties of dis-continuous stringer sandstones, and large volumes(1.05 BSTB) of residual oil. Production character-istics of the S1 highstand deltaic and S3 lowstandincised-valley-fill successions vary greatly and arepartly a function of contrasting sandstone architec-ture and depositional origin (Table 1). For example,reservoir sandstone geometry in the S1 highstand


deltaic succession is dominated by dip-elongate,distributary-channel sandstone bodies, whereasbed-load, fluvial-channel-fill sandstones in the S3lowstand incised-valley-fill succession, associatedwith a strong water drive, are inferred to be well-connected vertically and laterally (Table 1). Fewerinfill or well-deepening opportunities exist relativeto those in the S1 highstand deltaic succession.Recovery strategies in the S3 lowstand incised-valley-fill succession are focused mainly on plug-ging off lower, water-saturated zones to avoid up-ward water encroachment, as well as identifying thin(commonly <5 ft [<1.5 m]) zones of oil-saturated,coarse sandstones interbedded with sandy con-glomerates. In contrast, sandstone bodies in therelatively complex S1 highstand deltaic succession,associated with both solution-gas and water driverecovery mechanisms, exhibit great variation inreservoir quality (Table 2). Sweep efficiency andsecondary recovery vary greatly between individ-ual leases in the S1 highstand deltaic succession inthe SPA.

Two main development strategies in the EastTexas field, well deepening and optimized water-floods, are options for increasing recovery efficiencyand are described here in detail. Successful imple-mentation of these strategies will depend on athorough understanding of the reservoir sandstonearchitecture and its control on fluid flow. Otherstrategies, including polymer flooding and enhancedoil recovery (EOR), which are not described in de-tail in this study, can also be used to improve re-covery efficiency in the field. Polymer flooding canrecover additional remaining oil in the field by di-verting waterfloods into poorly swept stringer sand-stones where high-permeability sandstones in theoverlying incised-valley-fill section have served asthief zones. The EOR would target the remaining1.05 BSTB of residual oil in the field. The presenceof high API-gravity oil (38j) in permeable reser-voirs is an incentive for pursuing EOR in the EastTexas field. However, CO2 flooding in the fieldwould likely be immiscible because the reservoirsare low pressure (�1250 psi). The minimum misci-bility pressure for 38jAPI oil at 130–146jF (54.44–63.33jC) in the field is 1850–2000 psi (Ambroseet al., 2007). Moreover, the potential for CO2Ta

ble

1.SummaryofReservoirSandstoneThickness,Geometry,H

eterogeneity,D

rive

Mechanism

,ProductionCharacteristics,andRecoveryStrategies

intheS1

Highstand

Deltaic

andS3

LowstandIncised-Valley-FillSuccessionsin

EastTexasField

Sequence

Net-Sandstone

Values

andGeometry

ReservoirHeterogeneity

Drive

Mechanism

ProductionBehavior

Recovery

Strategies

S3lowstand:

incised-valleyfill

30–125ft;

broad,

well-connected

sheetsandstones

Thin

zonesof

interchannel

mudstones

andsandyconglomerates

Water

drive

Significant

water

encroachment

andverticalmigrationof

oil-w

ater

contact

Plugging

lower,water-producing

zones;recompletionzonesof

interbeddedconglomeraticand

potentially

oil-saturated

sandy

zones(NPA)

S1highstand:

fluvial-dom

inated

delta

5–25

ft;narrow

,

dip-elongate

sandbodies

crosscutting

structuralgradient

Sandbody

pinch-outs,muddy

andsilty

matrix,zonesof

soft-sedimentdeform

ation

Solution-gasand

water

drive

Steepdeclinein

deepened

wellscaused

bylim

ited

reservoir-compartmentsize

(NPA);variablepressure

supportfrom

water

flooding

ofcomplex

sandstones

(SPA)

Welldeepening

(NPA),strategically

targeted

water

flooding,and

limitedpolymer

flooding(SPA)

Ambrose et al. 261

leakage is great in the field because of wellbore in-tegrity issues; casing leaks have been reported fromapproximately one-fourth of the approximately31,000 wells in the field (East Texas EngineeringAssociation, 1953).

Well Deepening

Production of oil in the East Texas field by deepen-ing of existing wells is primarily from stringer sand-stones in the S1 highstand deltaic succession thatare inferred to contain limited, untapped reservoircompartments because of abrupt lateral and verti-cal changes in thickness of sandstone bodies. Per-meability and porosity data in Table 2, in con-junction with net-sandstone maps in Figure 17,indicate that that primary reservoir facies in theS1 highstand deltaic succession in the SPA occurin thick (>25 ft, >7.6 m) distributary-channel andchannel-mouth-bar stringer sandstones. In the NPA,the thickest stringer sandstones (>10 ft, >3 m)occur in southwest-trending belts composed ofchannel-fill deposits in an upper delta-plain setting(Figures 18a, 19). Early production in the field fo-cused on thick, permeable sandstones in the up-per, incised-valley-fill interval, with many wells notpenetrating the underlying stringer sandstones. Inaddition, some stringer sandstones, particularly inthe NPA, have been commingled with the upperincised-valley-fill sandstones and, after being shutin, may have been resaturated with oil resultingfrom updip migration. For potential targets in thestringer sandstones that could be produced eitherthrough well deepening or recompletions to beranked, all commingled and shut-in wells must beidentified, and locations where relatively thicksandstone bodies are distributed in areas wherefewer already producing stringer wells exist mustbe determined.

Several examples of production from recentlydeepened wells in the NPA illustrate the potentialfor additional oil recovery that still exists in theEast Texas field (Figure 19). In October 2007, therecently deepened Danmark 12 Moncrief well, lo-cated in the southwest part of the NPA, produced69 barrels of oil per day (BOPD). This well occursTa

ble

2.Summaryof

Selected

ReservoirProperties(Porosity

andKlinkenbergPerm

eabilityataConfiningPressure

of2000

psi)andLateralExtent

ofPrincipalFacies

intheS1

Highstand

DeltaicandS3

LowstandIncised-Valley-FillSuccessionsInferred

from

Net-Sandstone

Mapsin

theSouthern

Partof

EastTexasField

Facies

Porosity

Rangeand

Mean(%

)

KlinkenbergPerm

eability

RangeandMean(m

d)

Net-Sandstone

Values

(ft)

andPatterns

Strike

Continuity

(ft)

Dip

Continuity

(ft)

Incised-valleyfill(S3lowstand)

4.7–25.4

(18.3)

8.8–954.0(285.1)

Depositionalaxes

75–100;

broad

sheetswithin

valley-fillcomplex

>5000

>10,000

Delta

front(S1highstand)

2.9–37.4

(17.2)

<1.0–678.0(142.8)

5–20;elongate

trendflanking

distributary-channelaxes

1000

–3000

�3000

Channel-mouth

bar(S1highstand)

13.0–27.5

(22.3)

<1.0–1650.0

(847.0)

10–25;elongate-to-arcuatetrend

flankingterm

iniof

distributary-

channelaxes

20–40

ft:digitate

�500

Crevassesplay(S1highstand)

9.3–25.5

(17.7)

<1.0–1360.0

(262.1)

5–20

ft:digitate

500–1000

500–1000

Distributarychannel(S1highstand)

14.2–30.0

(24.5)

<1–2320.0

(897.5)

20–40

ft:digitate

20–40

ft:digitate

�4000


Figure 19. Selected recently deepened and successfully producing wells in S1 highstand stringer sandstones in the NPA. Thenorthwest–southeast stratigraphic section in the northwestern part of the NPA is shown in Figure 20. STB = stock tank barrels. SND =SND Energy Company, Inc.

Ambrose et al. 263

in a southwest-trending, anastomosing, fluvial string-er depositional axis with 12 ft (3.7 m) of net sand-stone. Although the overlying, low-resistivity, upperincised-valley-fill sandstone was inferred to be de-pleted, an upper sandstone bed in the lower stringersandstone interval, more than 20 ft (>6 m) belowthe incised-valley-fill sandstone, was inferred fromrelatively higher resistivity values to contain addi-tional untapped oil (Figure 19). The Danmark C-17Bumpas well is also located in a major stringerfluvial depositional axis (defined by >10 ft [>3 m]of net sandstone) in the western part of the NPA(Figure 19). This deepened well initially produced125 BOPD, resulting in a sharp increase in themonthly production of the Bumpas C lease fromless than 400 bbl to more than 3500 bbl beforeproduction began to decline again (Figure 19).The SND 18 Satterwhite well, deepened in late2005, demonstrates that additional oil can be pro-duced in minor fluvial stringer depositional axeswith thin (�5 ft, �1.5 m) net-sandstone values(Figure 19). The SND 18 Satterwhite well, locatedin a southwest-trending stringer fluvial tributarysystem, initially produced 20 BOPD and tempo-

rarily halted production decline in the Satterwhitelease (Figure 19).

The SND 3 Spurrier and SND 6 Spurrier wellsare two neighboring wells in the northwestern partof the NPA that were deepened in October 1996and November 1997, respectively (Figure 19). Bothwells are on the west margin of a south-trendingtributary, where net-sandstone values range from 5to 15 ft (1.5 to 4.5 m). As a result of these twowells being deepened and completed in Woodbinestringer sandstones, monthly oil production inthe Spurrier lease initially climbed from approx-imately 480 bbl of oil per month to more than1500 bbl of oil per month (Figure 19). A northwest–southeast stratigraphic cross section in the north-western part of the NPA shows that the deepenedSND 3 Spurrier well is located on the west mar-gin of a complex of multistoried channel-fill sand-stones preserved below the base of the S3 lowstandincised-valley-fill succession (Figure 20). Two sand-stone intervals were perforated in this deepenedwell: the lower one in a channel-fill sandstone in-ferred to pinch out northwestward into relativelymuddy channel-margin deposits, and the upper

Figure 20. Northwest–southeast stratigraphic cross section that includes the deepened SND 3 Spurrier well in the northwesternpart of the NPA, illustrating inferred facies architecture of stringer sandstones in the S1 highstand deltaic succession. The lineof section is located in Figures 18a and 19. GR = gamma ray; SP = spontaneous potential; Res = resistivity; SB = sequenceboundary. TXOK = TXOK Energy Resources Co.; SND = SND Energy Company, Inc.


Figure 21. Net-sandstone maps of two genetic depositional units in the S1 highstand stringer section in the SPA, with monthlyproduction per lease in 2007, distribution of water-injection wells, and an example of increased oil recovery by water injection inthick (>20 ft, >6 m), continuous, distributary-channel sandstones in the Mason lease. Monthly production of the combined stringersandstones varies greatly between individual leases in SPA and is partly related to local facies and net-sandstone distribution.Secondary recovery could be made more cost effective in these reservoir examples by discontinuing water injection in sandstone-poor areas where production has ceased from interdistributary and distal-delta-front facies. (a) Net sandstone in the LWB 30–40unit. (b) Net sandstone in the LWB 20–30 unit. (c). Monthly oil production in stock tank barrels (STB) from the Mason lease from1993 to present. Increased oil production in the Mason lease in 1997 was caused by water injection into transmissive, southwest-trending distributary-channel sandstones in the LWB 30–40 unit. The location of SPA is shown in Figure 1. The west–east structuralcross section in the Kinney lease is shown in Figure 22.

Ambrose et al. 265

one in a thin (�5 ft,�1.5 m), channel-margin sand-stone inferred to be in poor lateral communica-tion with thicker and sandier channel-fill depositseastward.

Optimized Waterfloods

Pressure support from injected water has beenunder way in many areas in the East Texas field since1938, increasing recovery efficiency to 77%. How-ever, water flooding could locally be made morecost effective by reducing the number of water-injection wells in sandstone-poor areas, where pro-duction has declined to as little as 2 BOPD. Otheroptions include designing mini waterfloods to pro-vide pressure support in areas where discontinu-ous stringer sandstones have not been penetratedby existing well bores or where relatively thickdistributary-channel sandstones are favorably ori-ented along the structural gradient. Two geneticdepositional units in the Woodbine stringer section(LWB 20–30 and LWB 30–40) in the SPA illus-trate relationships between facies and geometry ofsandstone bodies, oil production, and how waterflooding patterns could be redesigned to increaseefficiency (Figure 21a, b). In addition, the complexlateral and vertical geometry of sandstone bodiesand facies architecture in parasequences in the S1highstand deltaic succession such as the LWB 40–

30 interval (Figure 21a) are inferred to limit path-ways for injected water where sandy distributary-channel sandstones pinch out laterally into thinner,muddier, and poorly transmissive delta-front depos-its (Figure 22).

Monthly production from individual leasesvaries greatly throughout the area and is partly re-lated to local facies and net-sandstone distribu-tion. For example, relatively greater monthly pro-duction in the S1 highstand deltaic successionoccurs in leases in the northern and west-centralpart of the SPA where water-injection wells con-tact thick (>20 ft, >6 m) distributary-channel sand-stones (Figure 21). In contrast, there has been noproduction from sandstone-poor interdistributaryor delta-front facies in the Strickland lease in thesoutheast corner of the SPA in either the LWB 20–30 or LWB 30–40 units since 1996. Secondaryrecovery could be made more cost effective in theSPA by discontinuing water injection in the Strick-land lease, as well as in other muddy areas wherewater injection into thin, discontinuous sandstoneswill provide ineffective pressure support and resultin poor sweep efficiency. In contrast, water injec-tion into thick (>20 ft, >6 m) distributary-channelsandstones in the Mason lease in the southwestcorner of the SPA has been successful in boostingmonthly oil production by providing pressure sup-port in other wells to the northeast that are in the

Figure 22. West–east structural cross section in the Kinney lease in the SPA, showing the location of water-injection wells and thecomplex deltaic facies architecture in the lower Woodbine (LWB) 40–30 interval in the S1 highstand deltaic succession. The line ofsection is shown in Figure 21a. GR = gamma ray; Res = resistivity.


same distributary-channel trend and that are alsoup the structural gradient (Figure 21).

CONCLUSIONS

� This study presents a sequence-stratigraphic andnew depositional model for the lower WoodbineGroup in the East Texas field. Characterization ofthe sequence stratigraphy of the complete Wood-bine succession in the East Texas Basin enabled(1) the interpretation of structural influencesand their timing on Woodbine deposition in theEast Texas field and (2) the correlation of fourth-order sequences to the field where the unit istruncated, thus enabling the construction of aprecise chronostratigraphic framework withinthe field for accurate mapping of reservoir-faciestrends. Dip-elongate sandstones and conglomer-ates of lowstand incised-valley (bed-load fluvial)and highstand (high-constructive deltaic) sys-tems tracts compose the field’s productive zones.This interpretation contrasts with the seminalWoodbine study of Oliver (1971), which infersmeanderbelt-fluvial and strike-elongate stackedcoastal-barrier facies of high-destructive deltasin the field area.

� The lowermost Woodbine Group in the EastTexas field was deposited in a series of dip-elongate, fluvial-dominated deltas in an overallhighstand succession. This highstand successionis truncated by a lowstand succession of incised-valley-fill fluvial deposits that are composed ofmultistoried, conglomeratic sandstones.

� The greatest relative reservoir heterogeneity inthe East Texas field is inferred to occur in theS1 highstand, fluvial-dominated deltaic succes-sion, where narrow distributary-channel andlobate crevasse-splay sandstones pinch out overshort distances (commonly a few hundred feet)into interdistributary mudstones and siltstones.Reservoir continuity is inferred to be relativelygreater in the overlying S3 lowstand incised-valley fluvial succession, composed of gravellyand sandy bedload channel-fill deposits.

� Great potential exists for production of un-drilled, deeper Woodbine S1 highstand deltaic

sandstones and poorly swept pay in the EastTexas field despite the field having produced oilsince 1930. Additional mobile oil can be pro-duced in discontinuous lower Woodbine stringersandstones through deepening of existing wellsand recompletions. Waterfloods can be better de-signed to take advantage of the discontinuous res-ervoir sandstone geometry. Costs can be reducedby shutting off water-injection wells in muddyareaswherenoappreciablepressure support exists.

REFERENCES CITED

Adair, J. K., 1960, East Texas oil field, Gregg, Rusk, Upshur,Smith, and Cherokee counties, Texas: Compass, v. 37,no. 3, p. 174–181.

Alexander, C. I., 1951, History of discovery and developmentof Woodbine oil fields in East Texas, in The Woodbineand adjacent strata: History of discovery and develop-ment of Woodbine oil fields in east Texas: Dallas Petro-leum Geologists Fondren Science Series, p. 12–20.

Ambrose, W. A., T. F. Hentz, F. P. Wang, F. Bonnaffe, R. G.Loucks, E. C. Potter, and S. Clift, 2007, Pilot study ofEast Texas field: Geology, engineering, and potential ex-ploitation workshop: University of Texas at Austin,Bureau of Economic Geology, State of Texas AdvancedResource Recovery and Petroleum Technology TransferCouncil Workshop No. SW0017, variously paginated.

Arndorfer, D. J., 1973, Discharge patterns in two crevassesin the Mississippi River delta: Marine Geology, v. 15,p. 269–287.

Arthur, M. A., S. O. Schlanger, and H. C. Jenkyns, 1987, TheCenomanian–Turonian anoxic event: II. Paleoceano-graphic controls on organic matter production and pres-ervation, in J. Brooks and A. J. Fleet, eds., Marine petro-leum source rocks: Geological Society (London) SpecialPublication 26, p. 401–420.

Bhattacharya, J., and R. G. Walker, 1991, River- and wave-dominated depositional systems of the Upper CretaceousDunvegan Formation, northwestern Alberta: CanadianPetroleum Geology Bulletin, v. 39, p. 165–191.

Boothroyd, J. C., 1972, Coarse-grained sedimentation on abraided outwash fan, northeast Gulf of Alaska: Univer-sity of South Carolina, Coastal Research Division, Tech-nical Report No. 6, 127 p.

Brown Jr., L. F., A. W. Cleaves II, and A. W. Erxleben, 1973,Pennsylvanian depositional systems in north-central Texas:University of Texas at Austin, Bureau of Economic Ge-ology Guidebook 14, 122 p.

Brown Jr., L. F., R. F. Solıs-Iriarte, and D. A. Johns, 1990,Regional depositional systems tracts, paleogeography,and sequence stratigraphy, upper Pennsylvanian and lowerPermian strata, north- and west-central Texas: Universityof Texas at Austin, Bureau of Economic Geology Reportof Investigations 197, 116 p., 27 plates.

Ambrose et al. 267

Childs, O. E., G. Steele, and A. Salvador, 1988, Correlationof stratigraphic units of North America (COSUNA)project, Gulf Coast region: AAPG, oversize chart.

Christopher, R. A., 1982, The occurrence of theComplexiopollis-Atlantopollis zone (palynomorphs) in the Eagle Ford Group(Upper Cretaceous) of Texas: Journal of Paleontology,v. 56, no. 2, p. 525–541.

Coleman, J. M., and S. M. Gagliano, 1964, Cyclic sedimen-tation in the Mississippi River delta plain: Gulf CoastAssociation of Geological Societies Transactions, v. 14,p. 67–80.

Coleman, J. M., and D. B. Prior, 1982, Deltaic environments,in P. A. Scholle and D. R. Spearing, eds., Sandstone de-positional environments: AAPG Memoir 31, p. 139–178.

Coleman, J. M., and L. D. Wright, 1975, Modern river deltas:Variability of processes and sand bodies, in M. L.Broussard, ed., Deltas: Models for exploration: Houston,Texas, Houston Geological Society, p. 99–149.

Coleman, J. M., S. M. Gagliano, and W. G. Smith, 1964,Minor sedimentary structures in a prograding distribu-tary: Marine Geology, v. 1, p. 240–258.

DeDominic, J. R., 1988, Deposition of the Woodbine-Eagleford sandstones, Aggieland field, Brazos County,Texas: Gulf Coast Association of Geological SocietiesTransactions, v. 38, p. 61–68.

De Raaf, J. F. M., H. G. Reading, and R. G. Walker, 1965,Cyclic sedimentation in the lower Westphalian of NorthDevon, England: Sedimentology, v. 4, p. 1–52.

Dubiel, R. F., J. K. Pitman, and D. Steinshouer, 2003, Seismic-sequence stratigraphy and petroleum system modelingof the downdip Tuscaloosa-Woodbine, Louisiana and Tex-as, inG. Stone, J. Wrenn, and S. Bentley, eds., Gulf CoastAssociation of Geological Societies/Gulf Coast SectionSEPM Transactions, 53rd Annual Convention, v. 53,p. 193–203.

Dzou, L., S. Petmecky, and A. Pepper, 2000, East Texas Basinpetroleum systems revisited: Is long distance migration tothe giant East Texas field really necessary? (abs.): AAPGAnnual Convention Program, v. 9, p. A44.

East Texas Engineering Association, 1953, The East Texasoil field, 1930–1950: Kilgore, Texas, East Texas Engi-neering Association, 644 p.

Elliott, T., 1974, Interdistributary bay sequences and theirgenesis: Sedimentology, v. 21, p. 611–622.

Ewing, T. E., 1991a, The tectonic framework of Texas: Aus-tin, Texas, University of Texas at Austin, Bureau ofEconomic Geology, 36 p.

Ewing, T. E., 1991b, Structural framework, in A. Salvador,ed., The Gulf of Mexico Basin: Geological Society ofAmerica, Decade of North American Geology, v. J, p. 37.

Fisk, H. N., 1961, Bar finger sands of the Mississippi delta,in J. A. Peterson and J. C. Osmond, eds., Geometry ofsandstone bodies—A symposium, 45th Annual Meeting,Atlantic City, NJ: AAPG Special Publication, p. 29–52.

Foss, D. C., 1979, Depositional environment of Woodbinesandstones, Polk County, Texas: Gulf Coast Associationof Geological Societies Transactions, v. 29, p. 83–94.

Frazier, D. E., 1967, Recent deltaic deposits of the Missis-sippi River: Their development and chronology: Gulf

Coast Association of Geological Societies Transactions,v. 17, p. 287–315.

Galloway, W. E., 1975, Process framework for describingthe morphological and stratigraphic evolution of deltaicdepositional systems, in M. L. Broussard, ed., Deltas:Models for exploration: Houston, Texas, Houston Geo-logical Society, p. 87–98.

Galloway, W. E., T. E. Ewing, C. M. Garrett Jr., N. Tyler, andD. G. Bebout, 1983, Atlas of major Texas oil reservoirs:Austin, Texas, University of Texas at Austin, Bureau ofEconomic Geology, 139 p.

Granata Jr., W. H., 1963, Cretaceous stratigraphy and struc-tural development of the Sabine uplift area, Texas andLouisiana, in L. A. Herrmann, ed., Report on selectednorth Louisiana and south Arkansas oil and gas fields—Reference volume 5: Shreveport, Louisiana, ShreveportGeological Society, p. 50–95.

Halbouty, M. T., and J. J. Halbouty, 1982, Relationships be-tween East Texas field region and Sabine uplift in Texas:AAPG Bulletin, v. 66, no. 8, p. 1042–1054.

Haq, B. U., J. Hardenbol, and P. R. Vail, 1988, Mesozoic andCenozoic chronostratigraphy and cycles of sea levelchange, in C. K. Wilgus, C. A. Ross, H. Posamentier, andC. G. St. C. Kendall, eds., Sea-level changes: An inte-grated approach: SEPM Special Publication 42, p. 71–108.

Herbin, J. P., L. Montadert, C. Muller, R. Gomez, J. Thurow,and J. Wiedmann, 1986, Organic-rich sedimentation atthe Cenomanian–Turonian boundary in oceanic and coast-al basins in the North Atlantic and Tethys, in C. P. Sum-merhayes and N. J. Shackleton, eds., North Americanpaleoceanography: Geological Society (London) SpecialPublication 21, p. 389–422.

Houseknecht, D. W., 1987, Assessing the relative impor-tance of compaction processes and cementation to re-duction of porosity in sandstones: AAPG Bulletin, v. 71,no. 6, p. 633–642.

Hudnall, J. S., 1951, East Texas field, in F. A. Herald, ed.,Occurrence of oil and gas in northeast Texas: Universityof Texas at Austin Publication No. 5116, p. 113–118.

Jackson, M. P. A., 1982, Fault tectonics of the East TexasBasin: University of Texas at Austin, Bureau of EconomicGeology Geological Circular 82-4, 31 p.

Jackson, M. L. W., and S. E. Laubach, 1988, Cretaceous andTertiary compressional tectonics as the cause of the Sa-bine arch, east Texas and northwest Louisiana: Gulf CoastAssociation of Geological Societies Transactions, v. 38,p. 245–256.

Jackson, M. L. W., and S. E. Laubach, 1991, Structural his-tory and origin of the Sabine arch, east Texas and north-west Louisiana: Bureau of Economic Geology GeologicalCircular No. 91-3, 47 p.

Jackson, M. P. A., and S. J. Seni, 1984, Atlas of salt domesin the East Texas Basin: University of Texas at Austin,Bureau of Economic Geology Report of InvestigationsNo. 140, 102 p.

Jasper, A. K., and W. O. Wagner, 1989, Kerens, South (Wood-bine) field, in F. A. Herald, ed., Occurrence of oil andgas in northeast Texas: Tyler, Texas, East Texas Geolog-ical Society, p. 73-78.


Kelling, G., and G. T. George, 1971, Upper Carboniferoussedimentation in the Pembrokeshire coalfield, in D. A.Bassett and M. G. Bassett, eds., Geological excursions inSouth Wales and the Forest of Dean: Cardiff, Wales,Geological Association South Wales Group, p. 240–259.

Loeblich Jr., A. R., and H. Tappan, 1961, Cretaceous plank-tonic foraminifera: Part I. Cenomanian: Micopaleontol-ogy, v. 7, no. 3, p. 257–304.

Mancini, E. A., and T. M. Puckett, 2005, Jurassic and Creta-ceous transgressive-regressive (T-R) cycles, northern Gulfof Mexico, U.S.A.: Stratigraphy, v. 2, no. 1, p. 31–48.

McBride, E. F., A. E. Weidie, and J. A.Wolleben, 1975, Del-taic and associated deposits of Difunta Group (Late Cre-taceous to Paleocene), Parras and La Popa Basins, north-western Mexico, in M. L. Broussard, ed., Deltas, modelsfor exploration: Houston, Texas, Houston GeologicalSociety, p. 485–522.

Minor, H. E., and M. A. Hanna, 1933, East Texas oil field:AAPG Bulletin, v. 17, no. 7, p. 757–792.

Mitchum, R. M., J. B. Sangree, P. R. Vail, and W. W. Wornardt,1993, Recognizing sequences and systems tracts fromwell logs, seismic data, and biostratigraphy: Examplesfrom the Late Cenozoic of the Gulf of Mexico, in P.Weimer and H. Posamentier, eds., Siliciclastic sequencestratigraphy: Recent developments and applications:AAPG Memoir 58, p. 163–197.

Nunn, J. A., 1990, Relaxation of continental lithosphere: Anexplanation for Late Cretaceous reactivation of the Sa-bine uplift of Louisiana-Texas: Tectonics, v. 9, no. 2,p. 341–359.

Oliver, W. B., 1971, Depositional systems in the WoodbineFormation (Upper Cretaceous), northeast Texas: Univer-sity of Texas at Austin, Bureau of Economic Geology Re-port of Investigations No. 73, 28 p.

Oomkens, E., 1967, Depositional sequences and sand distri-bution in a deltaic complex: Geologie en Mijnbouw,v. 46, p. 265–278.

Oomkens, E., 1974, Lithofacies relations in the late Quater-nary Niger Delta complex: Sedimentology, v. 21, p. 195–222.

Ore, H. T., 1963, Some criteria for recognition of braidedstream deposits: Wyoming University Department of Ge-ology Contributions to Geology, v. 3, p. 1–14.

Pessagno Jr., E. A., 1969, Upper Cretaceous stratigraphy ofthe western Gulf Coast area of Mexico, Texas, and Ar-kansas: Geological Society of America Memoir 111, 139 p.

Phillips, S., 1987, Shelf sedimentation and depositional se-quence stratigraphy of the Upper Cretaceous Woodbine-Eagle Ford groups, east Texas: Ph.D. dissertation, Ithaca,Cornell University, 507 p.

Rust, B. R., 1972, Structure and process in a braided river:Sedimentology, v. 18, p. 221–245.

Salvador, A., 1991, Origin and development of the Gulf ofMexico Basin, in A. Salvador, ed., The Gulf of MexicoBasin: Geological Society of America, Decade of NorthAmerican Geology, v. J, p. 389–444.

Salvador, A., and J. M. Q. Muneton, compilers, 1989, Strat-igraphic correlation chart, Gulf of Mexico Basin, the GulfCoast Basin: Geological Society of America, The Geologyof North America, v. J, plate 5.

Scott, G., 1926, The Woodbine sand of Texas interpreted asa regressive phenomenon: AAPG Bulletin, v. 10, no. 6,p. 613–624.

Seni, S. J., and M. P. A. Jackson, 1984, Sedimentary recordof Cretaceous and Tertiary salt movement, East TexasBasin: University of Texas at Austin, Bureau of EconomicGeology Report of Investigations No. 139, 89 p.

Siemers, C. T., 1978, Submarine fan deposition of theWoodbine-Eagle Ford interval (Upper Cretaceous), Ty-ler County, Texas: Gulf Coast Association of GeologicalSocieties Transactions, v. 28, p. 493–533.

Smith, N. D., 1970, The braided stream depositional envi-ronment: Comparison of the Platte River with someSilurian clastic rocks, north-central Appalachians: Geo-logical Society of America Bulletin, v. 81, p. 2993–3014.

Smith, N. D., 1974, Sedimentology and bar formation in theUpper Kicking Horse River, a braided outwash stream:Journal of Geology, v. 82, p. 205–224.

Sohl, N. F., E. R. Martınez, P. Salmeron-Urena, and F. Soto-Jaramillo, 1991, Upper Cretaceous, in A. Salvador, ed.,The Gulf of Mexico Basin: Geological Society of Amer-ica, Decade of North American Geology, v. J, p. 205–244.

Stehli, F. G., W. B. Creath, C. F. Upshaw, and J. M. ForgotsonJr., 1972, Depositional history of Gulfian Cretaceousof east Texas embayment: AAPG Bulletin, v. 56, no. 1,p. 38–67.

Surles Jr., M. A., 1985, Petroleum and source rock potentialof Eagle Ford Group (Upper Cretaceous), East TexasBasin (abs.): AAPG Bulletin, v. 69, no. 2, p. 309.

Turner, J. R., and S. J. Conger, 1981, Environment of depo-sition and reservoir properties of the Woodbine sand-stone at Kurten field, Brazos County, Texas: Gulf CoastAssociation of Geological Societies Transactions, v. 31,p. 213–232.

Tye, R. S., and J. M. Coleman, 1988, Evolution of Atchafalayalacustrine deltas, south-central Louisiana: SedimentaryGeology, v. 65, p. 95–112.

Van Wagoner, J. C., R. M. Mitchum, K. M. Campion, andV. D. Rahmanian, 1990, Siliciclastic sequence stratigra-phy in well logs, cores, and outcrops: Concepts for high-resolution correlation of time and facies: AAPG Meth-ods in Exploration Series 7, 55 p.

Wood, D. H., and E. H. Guevara, 1981, Regional structuralcross sections and general stratigraphy, East Texas Ba-sin: Austin, Texas, University of Texas at Austin, Bureau ofEconomic Geology, 21 p. plus 8 plates.

Wright, L. D., W. J. Wiseman, B. D. Bornhold, D. B. Prior,J. N. Suhayda, G. H. Keller, Z. S. Yang, and Y. B. Fan,1988, Marine dispersal and deposition of Yellow Riversilts by gravity-driven underflows: Nature, v. 332, p. 629–632.

Ambrose et al. 269

sequence-stratigraphic controls on complex reservoir architecture of

Documents