Sequence Stratigraphy of the Upper Jurassic Mixed Carbonate ...

Sequence Stratigraphy of the Upper Jurassic Mixed Carbonate/

Siliciclastic Haynesville and Bossier Shale Depositional Systems in

East Texas and North Louisiana

Andrea D. Cicero, Ingo Steinhoff, Tony McClain, Kimberly A. Koepke, and Jim D. Dezelle BP America, 501 Westlake Park Blvd., Houston, Texas 77079

ABSTRACT Recent discoveries in the Haynesville and Bossier shales have dramatically increased unconventional gas exploration activity in the mature petroleum provinces of East Texas and North Louisiana. The antecedent topography shaped by underlying carbonates and subsequent sediment budgets strongly influenced (1) facies development and stacking patterns that vary along the northern rim of the young Gulf of Mexico Basin during Haynesville and Bossier time, and (2) the depositional processes, total organic carbon richness, and preservation of the self-sourcing Haynesville and Bossier Shale units. The Haynesville Shale depositional system is unique in that it contains both retrogradational and progradational facies that are contemporaneous with each other. On the western shelf of the East Texas Salt Basin, the time-equivalent Gilmer (Haynesville) Lime consists of backstepping carbonate facies, whereas to the east strong progradational stacking patterns dominate in the North Louisiana and western Mississippi salt basins due to increased sediment supply from the ancestral Mississippi River which outpace subsidence and eustasy. Hence, major bounding stratigraphic events such as maximum flooding surfaces and condensed sections critical for shale gas exploration appear to be diachronous along depositional strike. During Bossier time, the western carbonates are drowned out and siliciclastics become increasingly dominant, expanding westward from North Louisiana into East Texas and ultimately across most of the northern Gulf of Mexico shelf as the Cotton Valley sandstones and its distal shale equivalents. Depending on paleophysiography, some areas of the Haynesville-Bossier system are restricted and relatively sediment starved. These correspond with areas of total organic carbon enrichment and in turn, lower shale gas exploration risk.

INTRODUCTION The recent interest in unconventional shale gas resources has caused a resurgence of activity in mature hy drocarbon-bearing basins of onshore North America. In the East Texas and North Louisiana salt basins, histori cal plays focused on the conventional clastics and carbonates (e.g., Rodessa, Travis Peak/Hosston, Pettet/Sligo, and Smackover), and since the late 1970s, the tight-gas sandstones of the Cotton Valley. With heritage fields facing a production decline, and recent advancements in completion and drilling technology, exploration and production activity has turned to Upper Jurassic Haynesville and Bossier shales. Given the relative infancies of these plays, lack of long-term production, uncertainty in estimation of gas reserves in unconventional plays, and Cicero, A. D., I. Steinhoff, T. McClain, K. A. Koepke, and J. D. Dezelle, 2010, Sequence stratigraphy of the Upper Jurassic mixed carbonate/siliciclastic Haynesville and Bossier Shale depositional systems in East Texas and North Louisiana: Gulf Coast Association of Geological Societies Transactions, v. 60, p. 133-148.

133

Copyright © 2010, The Gulf Coast Association of Geological Societies. All Rights Reserved.

Cicero et al.

complications due to the nuances of colloquial nomenclature and “state-line” geology (i.e., Coleman and Cole man, 1985), a regional assessment was undertaken to further our understanding of the stratigraphy of these plays in an effort to decrease exploration shale gas risk. Utilizing a classic sequence stratigraphic approach that integrates well log, seismic, benthic foraminiferal biostratigraphy, and variations in eustatic sea level (Mitchum et al., 1995), nomenclature as described by Van Wagoner et al. (1988), combined with additional insights from regional structural reconstructions, proprietary multi-disciplinary biostratigraphy (which integrates nannofossil, benthic foraminifera, palynological, and am monite data), basin modeling, and geochemical variations, a dual clastic-carbonate model has been proposed for the Haynesville-Bossier shale system. To the west in the East Texas Salt Basin, a retrogradational carbonate system was in place, whilst to the east in the North Louisiana and Western Mississippi salt basins, a prograda tional clastic system dominates. Additionally, underlying carbonate platform physiography greatly influenced the depositional settings and sediment supply of the Haynesville-Bossier system. Integration of all methodologies described above was paramount to our overall understanding of the system, as incomplete interpretations are drawn when a minimal number of subsurface disciplines are used.

METHODOLOGY The well and seismic database and area of interest of this study is shown in Figure 1. Fourteen hundred re gional wells with digital log curves were considered, of which approximately 600 were correlated in a series of strike and dip oriented cross-sections, incorporating mudlogs and core data when available. Multi-discipline bio stratigraphic analysis was available for 20 wells that helped define absolute ages to each of our correlated se quences. Well correlations were further refined using observations from a robust 2D seismic database of more than 2,500 seismic lines and several smaller 3D seismic surveys. To tie in our correlations with regional Gulf of Mexico stratigraphic variations, several long regional 2D lines were interpreted. Additionally, industry reports and studies, joint industry projects, and the literature aided in our interpretations. Combined insights from litho-, bio-, and seismic stratigraphic interpretations were used to define our regional chronostratigraphic framework (Fig. 2).

STRATIGRAPHIC RELATIONSHIPS A chronostratigraphic framework for the Upper Jurassic–Lower Cretaceous of East Texas and North Louisi ana is given in Figure 2, with accompanying interpreted and uninterpreted seismic lines in Figure 3. The type log in Figure 2 is located in Blocker Field, Harrison County, Texas, and used throughout as this work as a reference. Supersequences, sequence boundaries (SB), and maximum flooding surfaces (MFS) discussed in this paper are indicated on the log, as well as the lithological formation names to tie in with previous work and colloquial no menclature and understanding. Also shown are geological time intervals, absolute age (Gradstein and Ogg, 2004), Wheeler diagrammatics, systems tract interpretations, and the sea level curve of Haq et al. (1987, 1988). Given the duration of proposed orders of sequences throughout the geologic record by Vail et al. (1991), and the frequency at which sequence boundaries are observed within the sedimentary record of the study area, the follow ing time spans have been applied in this chronostratigraphic framework: 1st order, 50-100s m.y.; 2nd order, 4-50 m.y.; 3rd order, 0.5-4 m.y.; 4th order, 0.08-0.5 m.y.; and 5th order, 0.02-0.08 m.y. The chronostratigraphic system (Fig. 2) can be divided into three supersequences: the first (A) spans from the Werner Anhydrite/Louann Salt equivalent (SSB1 [supersequence boundary]) to the latest Cotton Valley clas tics (SSB2), representing over 20 m.y. of deposition. The second supersequence, B (SSB2-SSB3), is relatively short-lived (~8 m.y.) in comparison to the first, beginning with deposition of Late Cotton Valley lowstand fans followed by the precipitation of the Knowles Lime. The third supersequence (C) begins with the Valanginian Unconformity (SSB3), followed by deposition of the Travis Peak (Hosston) clastics and Pettet (Sligo) formations. Given the scope of this paper, discussion will focus on the sequence stratigraphy of the Haynesville and Bossier systems of Supersequence A. The maximum flooding surface (mfs) of Supersequence A occurs in the Bossier, just above SB3. It marks a transition from a generally transgressive system during Haynesville time to a generally progradational system during Bossier time. This correlates well with the eustatic sea level fluctuations described by Haq et al. (1987),

134

Sequence Stratigraphy of the Haynesville and Bossier Shale Depositional Systems in East Texas and North Louisiana

Figure 1. Location of study and well and seismic database utilized.

which depict rising sea level followed by a retreat during Lower Bossier time. Additionally, five third-order (or higher-order) sequences are identified within Supersequence A. The first sequence (SSB1-SB1) includes the Norphlet sandstones, the transgressive Lower Smackover carbonates, marls, and shales, and the Upper Smack over platform carbonates and shoals. The second sequence (SB1-SB2) is comprised of later Smackover carbon ates and clastics, and the Lower Gilmer (Cotton Valley) Lime and its evaporative Buckner equivalents. The third sequence (SB2-SB3) includes the Buckner clastics, followed by the transgressive Upper Gilmer Lime and Haynesville Shale. The Haynesville Shale comprises a transgressive systems tract (TST) of a third-order se quence, with the top of the Haynesville marked by a major flooding surface. This interpretation is similar to that of Hammes et al. (2009) and Goldhammer and Johnson (2001), who interpret the Haynesville as a TST within a second-order regional Gulf of Mexico supersequence. The Haynesville can be further broken down into three additional fourth-order cycles; in the central part of the study area where the Haynesville is predominantly shale facies, the cycles correlate well with lithological picks and often coarsen-upward in profile. Similarly, in the East Texas Salt Basin and to the west, Haynesville carbonates can be divided into three fourth-order coarsening upward (“cleaning-upward”) cycles that are correlatable with stages of pinnacle reef growth (Hammes et al., 2009; Hammes, 2009). As such, we have indicated timelines of the Haynesville to continue through, rather trun cate against, positive carbonate highs, such as the Strickland high and Sabine Island in Figure 3. Varying litholo gies within the Haynesville system are driven in part by underlying carbonate paleotopography (Fig. 4) and the sources and rates of sediment flux into the basin.

135

Cicero et al.

Figure 2. Chronostratigraphic diagram including a type log, sequence stratigraphy, absolute ages (Gradstein and Ogg, 2004), and eustatic sea level curves (Haq et al., 1987). Location of corresponding cross-section line A-A’ is indicated on location map (Fig. 1).

The overall prograding log profile (see type log in Figure 2, and Figure 5) of the Bossier Shale is indicative of a highstand systems tract (HST) that can be separated into a lower (up to SB3) and upper (up to SB4) cycle. These cycles, in turn, coarsen upward, and are often capped with carbonates. In the central portion of the study area, carbonate stringers mark the top of the Lower Bossier (SB3), but along the western shelf and in the East Texas Salt Basin, the Lower Bossier can include reef facies, platform carbonates of mixed lithologies, or shale. Norwood and Brinton (2001) suggested that the latest stage of pinnacle reef formation takes place during Bossier time along a steep platform edge whose paleotopography is driven by local tectonics and salt dissolution. Fol lowing a short-lived lowstand systems tract (LST) and/or minor transgressive systems tract (TST), some of the earliest “Bossier sandstones” are deposited in the East Texas Salt Basin. Sediment flux of clastics into the system increases throughout Upper Bossier time (SB3-SB4), followed by the major onset of Cotton Valley clastic depo sition (Taylor sands, SB5) sourced from the ancestral Red and Ouachita rivers (Klein and Chaivre, 2002).

PALEOGEOGRAPHY The influence of antecedent topography on the evolution of depositional systems in the East Texas and North Louisiana Salt Basins has been described by previous workers (Ewing, 2001; Klein and Chaivre, 2002; Ewing, 2009; Martin and Ewing, 2009). To derive a proxy of basin paleophysiography, a regional isochron map was

136


Figure 3. (Top) Uninterpreted seismic line and (bottom) integrated seismic and chronostratigraphy of line A-A’. Supersequence boundaries are indicated in red, sequence boundaries in black, and maximum flooding surfaces in blue. Onlap and downlap indicated with the use of arrows. Seismic data courtesy of Seismic Exchange, Inc.

137

Cicero et al.

Figure 4. Proxy of shelf paleophysiography derived from a regional isochron of Pettet to Upper Gilmer Lime from seismic time structure maps. Isochron thicks are interpreted as paleobasins, whereas isochron thins are interpreted as paleohighs. Intrashelfal highs and sub-basins are named based on relative geographic position for communication purposes. The southern margin of the map has been interpreted more rigorously to reflect the continental slope. Depth of the East Texas Salt Basin may be accentuated due to later salt movement and changes in shelf physiography.

derived from Pettet and Upper Gilmer Limestone time structure surfaces interpreted from our 2D seismic grid, where isochron thins are interpreted as relative highs, and thicks represent relative lows, or paleobasins (Fig. 4). Although this map does not take into account late salt movement and varying shelf physiography during Gilmer through Pettet time, it does show relative basin and northern Gulf of Mexico shelf physiographies, including basement highs and intra-shelfal sub-basins. The relative positions of the East Texas Salt and Haynesville basins (approximate North Louisiana Salt Basin) can be approximated from this isochron map, although depth variations in the East Texas Salt Basin may not have been as deeply accentuated as it appears on the map, as it was greatly affected by late salt evacuation and diapirism. The map was calibrated to 3D seismic in a small area over Sabine Island, where relatively little movable salt was deposited, and suggests a high degree of confidence of physi ographic nuances there. The extent of the Jurassic carbonate platform system, and erosional updip limit of Jurassic sediments, gener ally corresponds with the Mexia-Talco and Southern Arkansas fault systems. The East Texas Salt Basin is sepa rated from the Haynesville Basin by a system of paleohighs, including the Gilmer Ridge, Strickland High, and a series of carbonate islands along the continental shelf break. Unlike the East Texas Salt Basin, which are open to marine waters, the paleohighs surrounding the Haynesville Basin served as natural barriers to marine circulation. Periodic increased influx of marine waters through conduits between the islands was likely when sea levels were relatively high. The details of paleobasin structure in the eastern part of the Haynesville Basin are not well con strained due to a lack of deep well data and poor seismic coverage. Regional structural flexures such as the an cestral LaSalle Arch could have provided additional restriction and influence sedimentary processes and budgets, even as subtle seafloor topographic features during periods of relative sea level lows.

138


Figure 5. Subsurface cross-sections A-A’ and B-B’ showing supersequence and sequence boundaries, positions of maximum flooding surfaces, and varying stacking patterns in the Haynesville and Bossier shales. Profile A-A’ corresponds with interpreted seismic line in Figure 3. The relative position of the mfs of Supersequence A (SSA) is constant above SB3. In contrast, the MFS of SSA in cross-section BB’ shifts from the top of the Haynesville to within SB3-SB4 from east to west. Also note that Haynesville wire-line log signatures are retrogradational to the west (backstepping) and progradational (coarsening-upward) to the east. For details, see text.

DISCUSSION Figure 6 is a generalized east-west cross-section spanning from the western shelf of the East Texas Salt Ba sin to the ancestral Mississippi River that illustrates the relationships between eustasy, subsidence and sedimenta tion rates at hypothetical times (T0, T1, T2, etc.). Progradational, retrogradational, and aggradational facies stacking patterns are denoted with colored arrows. Maximum flooding surfaces are indicated along the profile, and vary with paleobasin physiography and the interplay between subsidence rates and sediment input. In the East Texas Salt Basin, given a relatively constant eustatic sea level, Late Jurassic subsidence exceeds sedimentation rates. Carbonate precipitates in preferential marine environments with minimal clastic sediment influx into the basin. During this time, proto-fluvial and deltaic systems of the Red and Ouachita rivers are not in place nor well developed in the region; thus, clastic sediment inputs are low. Reef buildups and shallow marine platform facies are common, and an overall backstepping profile is observed in Western Shelf carbonates (Goldhammer and Johnson, 2001). The maximum flooding surface of Supersequence A appears to coincide with

139

Cicero et al.

Figure 6. Schematic of gradational facies stacking patterns from the Smackover to Taylor sands in the Cotton Valley section across the East Texas Salt Basin to the Haynesville Basin in the east, where the sediment input through the ancestral Mississippi delta overwhelms subsidence at a constant eustatic sea level causing regression and progradation, whereas facies belts along the western East Texas Salt Basin are still retrograding until much later in geologic time (see T0 to T7 relative time scale). The Gilmer Ridge area records a transition from retrogradational to progradational systems between the two end member locations in the west and in the east. The schematic is based on well log observations in the respective areas, which suggests a diachronous nature of the maximum flooding surface between the Upper Haynesville and Lower Bossier.

the top of the Lower Bossier indicating a change from retrogradational to progradational conditions resulting from increasing siliciclastics from the developing fluvial-deltaic systems to the north and west. Massive sand stone deposition begins in early Cotton Valley time. To the east in the Haynesville Basin, sediment input exceeds subsidence rates, creating an overall prograda tional system. The ancestral Mississippi River system was in place as early as Smackover time, growing increas ingly stronger through the Jurassic. Thick isopachs and fluvial-deltaic log signatures indicate clastic influx was very high during Haynesville and younger time (Buller and Dix, 2009; Hammes et al., 2009). With subsidence at a minimum in the basin, the system is overwhelmed with sediment input, and exhibits an aggradational to progra dational pattern following deposition of deeper water clastics. Facies stacking patterns suggest the maximum flooding surface here to be coincident with the top of the carbonates, or within the lowermost cycle of the Haynesville Shale. An overall coarsening-upward log profile characterizes the Haynesville through Cotton Val ley systems, punctuated with minor flooding events associated with fourth and fifth order cycles (Fig. 5). Highest potential for enrichment and preservation of total organic carbon exists in areas where clastic and carbonate (or “inorganic”) input is subdued or non-existent. This is most likely in areas that are either sheltered from detrital input and/or conditions that allow for carbonate precipitation, build-up, or both, and that receive a disproportionately large amount of organic carbon. Dysoxia or even anoxic bottom water conditions are also

140


required to allow total organic carbon to be preserved and available as input for hydrocarbon maturation (Loucks and Ruppel, 2007). All of these geologic requirements are usually met in the deeper parts of restricted sub-basins during times of detrital input starvation and diminished carbonate precipitation and/or build up during flooding periods of respective parts of the depositional system. Between the East Texas Salt Basin and the Haynesville Basin, which represent two such intrashelfal sub basins, the Gilmer Ridge, which remains a positive topographical feature throughout Haynesville and Bossier time, is perhaps the most challenging to interpret in that it bridges both basinal systems and includes retrograda tional, aggradational, and progradational facies (defined as “omnigradational” in Figure 6). Along the eastern edge of the East Texas Salt Basin, shallow marine platform carbonates develop along the Gilmer Ridge. On the opposite side of the Gilmer Ridge, thick Haynesville and Bossier shales are deposited within the Haynesville Basin, which can be divided into smaller cycles each with a coarsening-up log profile. This cyclical pattern con tinues through Late Bossier time, after which massive clastic input leads to deposition of the Taylor sands and its distal equivalents. Because the Gilmer Ridge is influenced both by a progradational system to the east and a ret rogradational system to the west, variations within the omnigradational facies during Haynesville and Bossier time give rise to multiple MFS interpretations. Trying to define a single regional MFS for Supersequence A (Fig. 2) can be quite complex. Log and seismic correlations are relatively simple within the shale facies of the Haynesville Basin proper; however, as you approach the boundaries of the basin, of which the Gilmer Ridge is one, the stratigraphic rela tionships are more complex. If correlations originate in the center of the basin and are carried toward the east, a more progradational stacking pattern may be presented, whereas, if, with the same origin, a correlation spans westward, an overall retrogradational system may present itself. One of the results of having opposing stacking patterns in tandem is the existence of apparently diachronous maximum flooding surfaces. Maximum flooding surfaces are fundamental components of sequence stratigraphic analysis; their relative positions influence se quence boundaries of systems tracts, cycles, and are commonly used to aid in the overall understanding of a ba sin’s history, which in turn, drive exploration decisions. Multiple chronostratigraphic scenarios can be discussed for even a single basin or shelf, depending upon the location and orientation of a given stratigraphic cross section. Numerous closely-spaced dip sections combined with intersecting strike sections help to demonstrate and quantify diachrony in the relative turning points of retrogradational to progradational stacking patterns in the sedimentary record (Fig. 5).

DEPOSITIONAL ENVIRONMENTS Depositional environment interpretations are captured in a series of maps for the Upper and Lower Haynes ville and Bossier systems given variations in paleogeography, sea level, and subsidence and sedimentation rates.

These smaller-scale stratigraphic systems correspond with fourth-order (or higher) sequences and are generally

more reflective of local variances in the factors listed above.

Lower Haynesville Figure 7 depicts the depositional environment model for the Lower Haynesville. This interval consists of backstepping carbonates that grade into “hot shales.” Carbonate and clastic depositional environments are most distinctive during Lower Haynesville time; a direct result of carbonate paleotopographic highs, such as the Gil mer Ridge and Strickland High, that separate the Haynesville from the East Texas Salt Basin. A series of carbon ate islands, including Sabine and Natchitoches islands located inboard of the continental shelf edge help restrict and define the boundaries of the Haynesville Basin. The carbonate system to the west consists of thick shallow marine platform carbonates, and a zone of patch reefs likely propagated along the edges of rifted horst blocks associated with salt dissolution and/or withdrawal (Fig. 7) (Norwood and Brinton, 2001). Behind the band of shallow marine carbonates, evaporites precipitated in a sabkha environment to the northwest resulting from ma rine restriction. A transition zone separates the carbonate-dominated from the clastic system to the east. Eastern sediment supply is from the fluvial-deltaic ancestral Mississippi River system. The Gray sands are a deepwater sandstone producing unit in the Lower Haynesville, located along the depositional shelf edge and are particularly prolific in

141

Cicero et al.

Figure 7. Gross depositional environment map of the Lower Haynesville. Erosional updip limit from Ewing (2001). Jackson and Lincoln parishes (Judice and Mazzullo, 1982; Ewing, 2001). Early activity along the Rodessa fault system (coincident with the Lower Haynesville depositional shelf break) could have provided accommodation space and a conduit by which Gray sand sediments could have been transported from the ancestral Mississippi River system along the shelf edge, with additional westward transport from longshore drift or climatically-driven wind currents (Moore et al., 1995). Given the proposed size and strength of the ancestral Mississippi River sys tem, basin-floor fan sedimentation outboard of the continental shelf break is possible and indicated by dashed boundaries in Figure 7. However, current-day structure does not allow sufficient well control to confirm this. Additional clastic input sources could include aeolian deposits. Thick, organic shales were deposited in the deepest parts of the Haynesville Basin, while penecontemporane ous carbonates precipitated on paleotopographic highs. Additional restriction could have been provided by subaqueous highs, such as an ancestral La Salle Arch in central Louisiana, during periods of sea level lows. The thickest section of the Lower Haynesville shale is located in northeast Texas, and has a “hot shale” gamma ray log response of up to 150 API [American Petroleum Institute] units. Geochemical analyses confirm the Lower Haynesville Shale to be the most anoxic part of the depositional system, which reflects highest preservation po tential for total organic carbon, likely due to the higher levels of restriction provided by carbonate topographic highs along basin edges. However, periodic influx of marine waters through conduits between these topographic features could have occurred.

Upper Haynesville During Upper Haynesville time, a high-energy carbonate system remains in place in the west, with patch reefs dotting the shelf edge in the southwestern portion of the East Texas Salt Basin (Fig. 8). Shallow marine

142


Figure 8. Gross depositional environment map of the Upper Haynesville. Erosional updip limit from Ewing (2001). carbonates become more expansive in some areas (i.e., Camp and Upshur counties), while bands of evaporites become narrower with increasing transgression and marine incursion. In the deepest portion of the East Texas Basin, the Upper Haynesville consists of offshore/shelfal carbonates grading into condensed shales that expand basinward and beyond the continental shelf break. To the east, the ancestral Mississippi River system gets stronger, increasing clastic sediment supply into the Haynesville Basin. Marginal marine deposition expands basinward and to the west. A unique depositional envi ronment exists in which marginal marine deposition occurs behind a zone of robust shallow marine carbonates in Bossier and Webster parishes, Louisiana. Shallow marine carbonates precipitate along a thin, structural high related to movement along the Rodessa fault system. This local carbonate high, arcuate in shape, is located along a roll-over structure on the downthrown side of the Rodessa fault, and can contain extremely thick sections of marine carbonates, which grade into shaly carbonate mixes along the peripheries. Marginal marine clastics fill the accommodation space along the fault zone behind (to the north of) the carbonates. Although Sabine and Natchitoches islands are now submerged, these carbonate subaqueous highs continue to restrict marine circulation. The Strickland High and Gilmer Ridge persist as carbonate topographic highs, pro viding restriction along the western flanks of the Haynesville Basin. Local changes in relative sea level would allow for periodic flooding and restriction of the basin. The top of the Upper Haynesville marks a major flooding surface, and the transition of a TST to HST.

Lower Bossier The colloquial moniker, “Bossier Shale” has historically referred to most, if not all, of the shale facies that occurs between the base of the Cotton Valley siliciclastics and the top of the Upper Gilmer Lime (Collins, 1980;

143

Cicero et al.

Klein and Chaivre, 2002; Williams et al., 2001). This thick occurrence of shale can be subdivided into the Upper, Lower, and most recently, “Mid-Bossier” shale. In this paper, using sequence stratigraphic principles, the Lower Bossier is defined as the HST of a third-order sequence, SB2-SB3 (Fig. 2). A major progradation commences in Lower Bossier time and persists throughout deposition of the Cotton Valley sandstones, with the first major onset of clastic sediments coincident with the deposition of the Cotton Valley Taylor sands. The overall highstand systems tract in Bossier time is punctuated with smaller-scale se quences of local flooding and is generally capped with carbonate stringers that are easily correlated around the Haynesville Basin. The Lower Bossier is characterized by flooding shale at its base, coarsening-upward log pro file, and localized carbonate precipitation at the top, where structural and/or marine conditions were favorable for carbonate precipitation. A retrogradational carbonate system still persists during Lower Bossier time to the west, however, isolated local clastic systems associated with proto-river systems of the Red, Ouachita, and other rivers, begin to prograde across the carbonate shelf following structural lows. Structural lows formed from early faulting associated with salt withdrawal and dissolution (Norwood and Brinton, 2001) and are conduits for clastic transport throughout Bossier time. Deepwater fans associated with these clastic systems are deposited in the East Texas Basin, and are correlative with the first pulse of “Bossier Sandstone” deposition, a highly productive play in East Texas (Klein and Chaivre, 2002). Conversely, shallow marine reef development continues on local highs of horst and graben structures that result from the aforementioned salt withdrawal and dissolution. A narrow fringe of patch reefs in the area of Robertson, Leon, and Freestone counties (Norwood and Brinton, 2001; Goldhammer and Johnson, 2001) persists where water conditions are conducive to reef development. Shallow marine carbonate develop ment continues to backstep landward, and zones of evaporites continue to narrow. Along the western shelf, car bonates are intermixed with clastics. The carbonate-siliciclastic transition zone continues to migrate westward as the ancestral Mississippi River gains strength and expands westward, widening the band of marginal marine deposition. A secondary fluvial input source may have been present in southwestern Arkansas. Another unique structural and depositional setting occurs in a zone spanning west to east from Harrison County, Texas, to Caddo Parish, Louisiana. Structural highs resulting from movement along the Rodessa fault zone create optimal conditions for carbonate marine deposition in the Lower Bossier. Carbonates, particularly ooids, discernable in core thin sections and noted in mud logs, form along an arcuate band, propagating on structural highs along the depositional shelf break. Behind (to the north of) this limited band of ooid development, shaly and mixed clastics are deposited. By Lower Bossier time, much of East Texas and North Louisiana experience widespread shale deposition. All carbonate paleogeographic highs have now been flooded, although some bathymetric relief is still associated with these structures. The Lower Bossier Shale contains a high volume of carbonate, although clastics dominate closer to the ancestral Mississippi River system. The Bossier Shale thins over the former Sabine Island and bas inward of the continental shelf break. Changes in depositional environment of the Lower Bossier are also re flected in lithological, mineralogical, and geochemical differences in comparison with the Haynesville Shale (Buller and Dix, 2009; Dix et al., 2010). Increased clastic input increases clay, chlorite, and aluminosilicate con tent, along with decreased total organic carbon content, and may render the Bossier Shale less organic-rich (and less prospective) than the Haynesville Shale in the Haynesville Basin proper.

Upper Bossier The Upper Bossier is here defined as a third order sequence that spans from SB3 to SB4 (Fig. 2). The last pulse of carbonate deposition in the system occurs during this sequence, after which marginal marine and fluvial Cotton Valley sandstones are deposited into the basin. This sequence corresponds well with the interval com monly known as the “Mid-Bossier Shale.” In this paper, we attribute shale above SB4 to be part of the next stratigraphic sequence, SB4-SB5, otherwise recognized as the deposition of the Taylor sands, but historically any shale located just below the Taylor sands would have been called “Bossier.” Hence, there are sometimes discrep ancies when trying to relate a colloquially-known interval with a chronostratigraphic framework sequence. In general, the “Mid-Bossier” corresponds with the Upper Bossier sequence presented in this paper. During latest Bossier time, widespread shale deposition continues. Although the ancestral Mississippi River system continues to strengthen, a last pulse of carbonate and evaporite deposition occurs during this time in the west (Fig. 10). Mixed carbonate-siliciclastics are common, given increasingly prograding fluvial-deltaic systems

144


Figure 9. Gross depositional environment map of the Lower Bossier. Erosional updip limit from Ewing (2001). along the western shelf of the East Texas Basin. Deepwater fan deposition in the southwestern portions of the East Texas Basin likely represent later pulses of “Bossier Sandstone,” which continue through early Cotton Val ley “Taylor”-aged time. The major onset of clastic deposition of the Cotton Valley Group begins in Upper Boss ier time, and continues through to the next marine transgression and deposition of the Knowles Lime. The “Mid-Bossier,” a prospective facies within the Bossier Shale, has similar characteristics and properties to the Haynesville shale (Hammes, 2009). On well logs, this zone is identifiable by increased porosity, and its core properties and total organic carbon content are similar to those of the Haynesville Shale. Lithological corre lations show the “Mid-Bossier” seemingly convergent with the Haynesville Shale in the southern portions of the Haynesville basin, however, careful chronostratigraphic correlations and mapping suggest this particular facies is time correlative with the Upper Bossier. Additional core and multi-disciplinary biostratigraphic data could pro vide additional insight into its ultimate age or time-transgressive nature.

CONCLUSIONS The Haynesville and Bossier depositional systems of East Texas and North Louisiana are influenced by ante cedent topography and paleobasin physiography, as well as varying rates of subsidence, sediment supply and eustasy, resulting in a dual system of retrogradational, backstepping carbonate facies in the west and prograda tional clastics in the east. The interplay of these two systems can be observed in wire-line log signatures, lithological variations, seismic stratigraphy, and ultimately, the depositional environments of each sequence. Given the diachronous nature of maximum flooding surfaces in the Haynesville-Bossier system, discerning the sequence stratigraphic history can be quite challenging. Whilst correlations within the shale facies in the central basinal region are straightforward, the direction in which one extrapolates to (west or east), and incorporating a

145

Cicero et al.

Figure 10. Gross depositional environment map of the Upper Bossier. Erosional updip limit from Ewing (2001). variety of subsurface tools can lead to widely varied stratigraphic and depositional interpretations. The integra tion of multi-discipline biostratigraphic data, along with insights from well logs, mud logs, and seismic data is critical in understanding this complex system.

ACKNOWLEDGMENTS The authors would like to acknowledge the support of BP America to publish this paper and thoughtful re views by Scott Hamlin and Ursula Hammes that improved this manuscript. Seismic Exchange, Inc., and Bailey Banks is thanked for permission to publish seismic data in Figure 3. We thank Scott Staerker for biostratigraphic insight and discussions, and Janelle DeFreitas-Duncan for ArcGIS assistance. We would also like to acknowl edge Art Donovan, Ursula Hammes, and the “Jurassic Café” crew for lively discussions about the complex se quence stratigraphy of this area.

REFERENCES CITED Adams, R. L., 2009, Basement tectonics and origin of the Sabine Uplift: Gulf Coast Association of Geological Sci ences Transactions, v. 59, p. 3-19. Buller, D., and M. C. Dix, 2009, Petrophysical evaluation of the Haynesville Shale in northwest Louisiana and northeast Texas: Gulf Coast Association of Geological Societies Transactions, v. 59, p. 127-143.

146


Coleman, J. L., Jr., and C. J. Coleman, 1981, Stratigraphic, sedimentologic and diagenetic framework for the Jurassic

Cotton Valley Terryville massive sandstone complex, northern Louisiana: Gulf Coast Association of Geological

Sciences Transactions, v. 31, p. 71-79.

Collins, S. E., 1980, Jurassic Cotton Valley and Smackover reservoir trends, East Texas, North Louisiana, and Southern Arkansas: American Association of Petroleum Geologists Bulletin, v. 64, p. 1004-1013. Dix, M., D. Spain, C. Walling, J. Sano, N. Casarta, and A. Richardson, 2010, Stratigraphy and depositional dynamics of the Haynesville-Bossier sequence: Inferences from whole-rock elemental data: American Association of Petro leum Geologists, Annual Convention and Exhibition, April 2010, New Orleans, Louisiana. Ewing, T. E., 2001, Review of Late Jurassic depositional systems and potential hydrocarbon plays, northern Gulf of

Mexico Basin: Gulf Coast Association of Geological Sciences Transactions, v. 51, p. 85-96.

Ewing, T. E., 2009, The ups and downs of the Sabine Uplift and the northern Gulf of Mexico Basin: Jurassic basement blocks, Cretaceous thermal uplifts, and Cenozoic flexure: Gulf Coast Association of Geological Sciences Transac tions, v. 59, p. 253-269. Goldhammer, R. K., and C. A. Johnson, 2001, Middle Jurassic-Upper Cretaceous paleogeographic evolution and se quence-stratigraphic framework of the northwest Gulf of Mexico rim, in C. Bartolini, R. T. Buffler, and A. Cantu-

Chapa, eds., The western Gulf of Mexico Basin: Tectonics, sedimentary basins, and petroleum systems: Ameri can Association of Petroleum Geologists Memoir 75, Tulsa, Oklahoma, p. 45-81.

Gradstein, F. M., and J. G. Ogg, 2004, Geological time scale 2004—Why, how, and where next!: Lethaia, v. 37, p. 175-181. Hammes, U., 2009, Sequence stratigraphy and core facies of the Haynesville Mudstone, East Texas: Gulf Coast Asso ciation of Geological Sciences Transactions, v. 59, p. 321-324.

Hammes, U., R. M. Reed, H. D. Rowe, and R Eastwood, 2009, Addressing conventional parameters in unconventional

shale-gas systems: Geochemistry, petrography, petrophysics, and depositional environment of the Haynesville

Shale: Proceedings of the 29th Annual Gulf Coast Section of the Society of Economic Paleontologists and Miner alogists Foundation Bob F. Perkins Conference, Houston, Texas, p. 181-203.

Haq, B. U., J. Hardenbol, and P. R. Vail, 1987, Chronology of fluctuating seaLevels since the Triassic: Science, v. 235, p. 1156-1167. Haq, B. U., J. Hardenbol, and P. R. Vail, 1988, Mesozoic and Cenozoic chronostratigraphy and cycles of sea level

change, in C. K. Wilgus, B. S. Hastings, C. G. St. C. Kendall, H. W. Posamentier, C. A. Ross, and J. C. Van Wag oner, eds., Sea level change: An integrated approach: Society of Economic Paleontologists and Mineralogists

Foundation Special Publication 42, Tulsa, Oklahoma, p. 71-108.

Judice, P. C., and S. J. Mazzullo, 1982, The Gray sandstones (Jurassic) in Terryville Field, Louisiana: Basinal deposi tion and exploration model: Gulf Coast Association of Geological Sciences Transactions, v. 32, p. 23-43.

Klein, G. D., and K. R. Chaivre, 2002, Sequence and seismic stratigraphy of the Bossier Formation (Tithonian), western East Texas Basin: Gulf Coast Association of Geological Sciences Transactions, v. 52, p. 551-561. Loucks, R. G., and S. C. Ruppel, 2007, Mississippian Barnett Shale: Lithofacies and depositional setting of a deepwa ter shale-gas succession in the Fort Worth Basin, Texas: American Association of Petroleum Geologists Bulletin,

v. 91, p. 579-601. Martin, B. J., and T. E. Ewing, 2009, Ancestral basin architecture: A possible key to the Jurassic Haynesville trend:

Gulf Coast Association of Geological Sciences Transactions, v. 59, p. 511-515.

147

Cicero et al.

Moore, G. T., E. J. Barron, and D. N. Hayashida, 1995, Kimmeridgian (Late Jurassic) general lithostratigraphy and source rock quality for the western Tethys Sea inferred from paleoclimate results using a general circulation model: American Association of Petroleum Geologists Studies in Geology 40, Tulsa, Oklahoma, p. 157-173. Norwood, E. M., and L. Brinton, 2001, Reexamination of Late Jurassic reef building in the East Texas Basin; a matur ing gas play: Gulf Coast Association of Geological Sciences Transactions, v. 51, p. 259-271. Mitchum, R. M., J. B. Sangree, P. R. Vail, and W. W. Wornardt, 1995, Recognizing sequences and systems tracts from well logs, seismic data, and biostratigraphy: Examples from the Late Cenozoic of the Gulf of Mexico, in H. A. Posamentier and P. Weimer, eds., Siliclastic sequence stratigraphy: American Association of Petroleum Geologists Memoir 58, Tulsa, Oklahoma, p. 163-197. Vail, P. R., F. Audemard, S. A. Bowman, P. N. Eisner, and C. Perez-Cruz, 1991, The stratigraphic signatures of tecton ics, eustasy, and sedimentation—An overview, in G. Einsele, W. Ricken, and A. Seilacher, eds., Cycles and events in stratigraphy: Springer-Verlag, New York, New York, p. 611-659. Van Wagoner, J. C., H. W. Posamentier, R. M. Mitchum, P. R. Vail, J. F. Sarg, T. S. Loutit, and J. Hardenbol, 1988, An overview of the fundamentals of sequence stratigraphy and key definitions, in C. K. Wilgus, B. S. Hastings, C. G. St. C. Kendall, H. W. Posamentier, C. A. Ross, and J. C. Van Wagoner, eds., Sea level change: An inte grated approach: Society of Economic Paleontologists and Mineralogists Special Publication 42, Tulsa, Okla homa, p. 39-45. Williams, R. A., M. C. Robinson, E. G. Fernandez, and R. M. Mitchum, 2001, Cotton Valley/Bossier of East Texas: Sequence stratigraphy recreates depositional history: Gulf Coast Association of Geological Sciences Transactions, v. 51, p. 379-388.

148