About This Item

Share This Item

The AAPG/Datapages Combined Publications Database

AAPG Bulletin

Abstract


Volume: 76 (1992)

Issue: 5. (May)

First Page: 589

Last Page: 606

Title: Oligocene Hackberry Formation of Southwest Louisiana: Sequence Stratigraphy, Sedimentology, and Hydrocarbon Potential (1)

Author(s): STEPHEN P. J. COSSEY (2) and RICHARD E. JACOBS (3)

Abstract:

The Oligocene Hackberry sequence was deposited in a slope environment consisting of an irregular, updip slide scar, a rotational slide zone up to 4 mi (6.5 km) wide, and a downdip region more than 20 mi (32 km) wide where meandering submarine channels deposited thick turbiditic sands. The shelf margin slides probably began during the late stage of a relative fall in sea level and prior to a maximum flood event in the middle Oligocene. The slides probably were caused by a combination of salt withdrawal and an unstable shelf edge. The play has produced more than 374 million bbl of oil equivalent (BOE) up to December 1988. The first fields were discovered in structural/stratigraphic traps on the updip flanks of the salt domes, where channels were forced to meander around pal obathymetric highs. Other fields are located in the paleobathymetric lows many miles downdip of the salt domes. Statistical analysis of field data shows that 41 fields with more than 1 million BOE each and with a total estimated ultimate recovery of 117 million BOE remain to be discovered in the play. Interpretation in southwestern Louisiana has shown that new reserves could be discovered in three potential reservoir sands: (1) lower Frio shelf-edge sands preserved in large slide blocks, (2) onlapping, sandy "fill sequences" restricted to the lows between slide blocks, and (3) meandering, dip-oriented, sandy channel complexes less than 4500 ft (1400 m) wide. These three sandstones cannot be distinguished unless dipmeter, seismic, and paleontologic data are used in combination.

Text:

INTRODUCTION

The Hackberry Formation is a hydrocarbon-bearing, Oligocene slope facies which occurs in a broad subsurface "embayment" of southeast Texas and southwest Louisiana (Figure 1). The Hackberry has always intrigued geologists and discouraged investors because of a reputation as a difficult play to understand and because of an abundance of dry holes. Difficulties in interpretation have been compounded by the fact that the play is often worked by geoscientists in either Louisiana or Texas, but rarely both.

Previous work by Paine (1968) and Benson (1971) has shown that the play can be subdivided into "updip" and "downdip" areas. The setting of the updip Hackberry play is an area of slope failure involving slide blocks; the setting of the downdip play is channels where thick, turbidite sands were deposited. The vague dividing line between the two subplays is an east-west line of salt domes which occur within the play boundary. Ewing and Reed's (1984) subdivision of the Texas Hackberry into "play 1" and "play 2" is essentially similar. Our paper investigates the depositional subenvironments, sequence stratigraphy, and depositional processes of the Hackberry, and defines the boundaries of the two Hackberry subplays in an area of western Calcasieu Parish, Louisiana. Specifically, our paper d cuments the characteristics of the slide blocks and channels on the basis of high-quality seismic data, well logs, dipmeter data, and paleontologic data. Known and potential reservoir sands associated with the subenvironments are classified. The relationship between the salt domes, slide blocks, and submarine channels in western Calcasieu Parish is also investigated.

The play has an abrupt northern boundary where the Hackberry sharply onlaps the unfaulted margin of the lower Frio shelf sediments (Figure 2). The eastern and western boundaries are fairly well defined, and the southern boundary is presently limited by drilling economics

End_Page 589------------------------------

to where the base of the Hackberry Formation is at about 15,000 ft (4600 m). The abrupt and well-defined geographic limits of the play allow accurate tabulation of field sizes and their discovery sequence. The potential sizes of fields yet to be found in the Hackberry are then estimated by statistical methods from the field-size distribution data.

SEQUENCE STRATIGRAPHY

The sequence stratigraphic interpretation of the Hackberry is based on observations made from southwest Louisiana seismic data, well logs, and paleontologic data. A chronostratigraphic diagram was constructed to demonstrate the sequence of events which occurred during most of the Oligocene (Figure 2). It does not imply that all the events described below occurred throughout the "Hackberry embayment," although some evidently did and have been described by previous workers. The interpretation seems to better fit a model where the genetic units are bounded by maximum flood surfaces (Galloway, 1989) rather than type 2 sequence boundaries (Van Wagoner et al., 1990).

In the early Oligocene, a highstand systems tract (the lower Frio shelf sequence) prograded southward and reached a point of maximum progradation just south of the present Beauregard Parish and Calcasieu Parish border. This period is represented by a eustatic rise (Haq et al., 1988) although relative sea level in southwest Louisiana was apparently falling. At the point of maximum progradation of the shelf, the shelf edge became unstable causing rotational sliding and slope regrading (Galloway, 1987). The sediments incorporated in these slides are thin parasequences representative of an outer shelf environment. The vertical log character of these sediments suggests that the shelf sequence changed from progradational to slightly aggradational just prior to the sliding.

The cause of the sliding could have been (1) loading of the underlying lower slope sediments, (2) oversteepening of the shelf edge due to salt withdrawal, or (3) a combination of (1) and (2) whereby the sediment loading actually initiated the movement of the salt. Repeated stratigraphy in at least one well in the study area indicate that multiple failures took place and that the process started at the shelf edge and progressed landward. The failures appear to have occurred along a zone several miles wide and almost 100 mi (160 km) along the shelf edge in Texas and Louisiana (Paine, 1968; Benson, 1971). The updip limit of this zone delineates what is often referred to as the "Hackberry embayment." The effect of the slides was to almost instantaneously produce accommodation space in a s stem which was just keeping pace with subsidence and eustatic rise. It also produced a very hummocky paleobathymetry which subsequently provided gravity flows with many sediment fairways to the upper slope.

Foraminiferal data invariably indicate a deepening of paleoenvironments from the lower Frio to the base of the Hackberry (Benson, 1971; Ewing and Reed, 1984). We independently observed this in many areas. According to Benson (1971, p. 4), "The underlying Frio section contains a shallow-water, middle to inner-shelf fauna." A shelf-edge failure could produce an instantaneous increase in water depth which would then be recorded in the foraminiferal assemblage of the post-failure sediments. A water depth increase of several hundred feet could locally mask the effect of a relative sea level fall, and so paleontologic data in the region of the sliding must be interpreted with care.

Sediment input along the northern boundary of the Hackberry embayment was quite variable. Large volumes of sand were being deposited during the middle Oligocene in eastern Calcasieu Parish and western Jefferson Davis Parish (Figure 1), where sediment was being fed into the embayment via a number of large submarine canyons (Paine, 1968). These canyons in map view are represented by a number of abrupt incursions within the slide scar at the updip limit of the embayment in Jefferson Davis and eastern Calcasieu parishes (Figure 3). However, in western Calcasieu Parish, sediment input was less, and no major submarine canyons were present because the boundary is relatively smooth in outline (Paine, 1968).

Sedimentation caused by turbidity currents and other gravity flow processes in Calcasieu Parish continued in the late middle Oligocene. Sandstones ("fill sequences") were deposited in the lows created by the rotational faulting. Submarine erosion of the slide blocks suggests that considerable volumes of sediment were transported onto the slope by gravity flow processes.

End_Page 590------------------------------

Shortly after the hummocky paleobathymetry of the slide area was filled, turbidity currents flowed further downslope and created channel/levee complexes. The channels were erosional in nature, and evidence in Calcasieu Parish suggests that up to 100 ft (30 m) of strata were removed. This sequence of events produced the tripartite division of the potential Hackberry reservoirs in Calcasieu Parish shown in Figure 2.

During the late Oligocene, relative sea level began to rise and effectively reduce the sediment supply to the slope channel complexes. This rise culminated in a maximum flooding in the late middle Oligocene, and shelf sediments (upper Frio) again started to prograde into the available accommodation space. The thick wedge of "Hackberry shale" of previous workers (Paine, 1968; Benson, 1971; Paine, 1971) is represented by the shale and turbidite sandstones of the slope sediments genetically associated with the late Frio shelf sediments.

The sequence stratigraphy is important because of the confusion which has surrounded the Hackberry terminology for the last 50 years. One of the problems is whether or not to include the rotated slide blocks of lower Frio within the Hackberry. In this paper, the economic and exploration definition of Hackberry play will include the Hackberry Formation and the slide blocks containing lower Frio sediments.

PRODUCTION HISTORY

The Hackberry play has produced more than 374 million bbl of oil equivalent (BOE) over the last 50

End_Page 591------------------------------

years, and it remains a target for many explorationists. Production data are difficult to confirm because of the terminology differences in Texas and Louisiana and because of the difficulty of establishing which fields actually produce from the Hackberry play. Fields which have been previously studied in detail include Port Acres (Jackson et al., 1987) and North Sabine Lake (Eubanks, 1987). Summaries of southeast Texas Hackberry fields have been published by Ewing and Reed (1984) and in map form by Eubanks (1987) and Curtis and Echols (1985).

In order to predict the exploration potential of the Hackberry play, accurate production data for all the fields in the play had to be obtained. Table 1 was produced by manually tabulating data from Petroleum Information's and Dwights' production databases and then comparing the results to other published field summaries as a quality check. Detailed correlation within each field area was sometimes necessary to confirm that the production was actually from the Hackberry sandstones. Estimated ultimate recoveries (EUR) for active fields were calculated by adding six times the 1988 production to the cumulative production. This was assumed to be an accurate estimation for fields on decline. EURs for abandoned fields was assumed to be the cumulative production of the field. Finally, the dat were converted to barrels of oil equivalent using the formula, 5.8 bcf of gas = 1 million BOE.

Historical interest was focussed on fields having an EUR of more than 1 million BOE; these fields were considered to be significant discoveries. Figure 3 was compiled using the data from Table 1 and shows the distribution of the significant discoveries as well as the smaller fields (EUR < 1 million BOE). Table 1 was also used to reconstruct the discovery sequence of all the significant Hackberry fields; this is shown in Figure 4.

Since the Hackberry play has such well-defined boundaries, its development may exemplify the history of a typical play. For example, the largest field in the play was discovered within the first 5 years of drilling and major accumulations (>10 million BOE) have since been discovered every few years, but the size of major

End_Page 592------------------------------
End_Page 593------------------------------

discoveries has decreased over time (Figure 4). The first phase of drilling discovered major accumulations such as Edgerley, Port Neches, and Gillis-English fields on the northern flanks of the known salt domes (Figure 3). A second phase of drilling in the 1960s led to the discovery of large fields such as South Manchester further downdip and not directly associated with the salt dome structures. As exploration progressed, it became evident that there were differences between the updip portion of the play and the downdip. The updip play was first described by Paine (1968) as the area north of the salt dome trend and south of the Hackberry shale pinch-out. The play was known to be associated with small rotational fault blocks, but high quality seismic and dipmeter data were not availab e to interpret the detailed stratigraphy. Paine (1971) also described several

End_Page 594------------------------------

cores from the Manchester area and concluded that the sands were deposited by turbidity currents which may have been deflected around paleotopographic highs in the location of the present-day salt domes.

From the early 1970s to 1985, drilling in the Hackberry play did not discover any fields with more than 8.2 million BOE (Bell City South). This was the longest period in the history of the play during which no field of at least 10 million BOE was discovered. However, interest in the Hackberry play again picked up with the discovery of Morgan Bluff field (Figure 3), which is thought to have an EUR of 12 million BOE. Continuing success was achieved in 1988 when two new-field discoveries were announced in Calcasieu Parish by Prairie Producing and Exploration Company of Louisiana (Figure 3). The initial production of these fields was 496 BOPD + 407 MCFGD and 380 BOPD + 821 MCFGD. These two discoveries were not included in the production statistics, but they give an indication of the high otential remaining in the Hackberry play.

FUTURE EXPLORATION POTENTIAL

With accurate historical data on the play, it is possible to predict its future potential, assuming that a log normal distribution of field sizes is present and that no field larger than Port Neches North (the largest Hackberry field) will be discovered. A glance at the historical exploration (Figure 4) of the Hackberry play indicates that, in the future, fields with EURs more than 10 million BOE will be rare or absent.

This prediction can be quantified by using a program called primes (Lee and Wang, 1986). This program is designed to fit the field EURs to a log normal distribution, match the discovered fields to "slots" in the distribution, and then predict the future field sizes or yet-to-be-found reserves by identifying the empty slots. Figure 5 shows a plot from primes for the Hackberry play, using fields with EURs of more than 1 million BOE. This plot show the most-likely case for finding future Hackberry fields. The results suggest that all the Hackberry fields with EURs of more than 10 million BOE have been found, but at least two fields in the 9-million BOE range remain to be found. Also, a total of 41 fields with EURs of more than 1 million BOE and a mean field size of 2.8 million BOE remain to be discovered. The total yet-to-be-found reserve in these 41 fields is estimated to be 117 million BOE or 31% of what has already been discovered. Not included in this total are the more than 100 yet-to-be-found fields with EURs between 100,000 and 1 million BOE, also predicted by the program.

This analysis predicts a large potential reserve in the Hackberry play, which will probably consist of relatively small exploration targets. A detailed study was conducted in western Calcasieu Parish, Louisiana, to provide some insight as to the nature and distribution of future exploration targets in the Hackberry.

WESTERN CALCASIEU PARISH STUDY AREA

A study area in western Calcasieu Parish (Figures 3, 6) was chosen to investigate the relationship between the slide areas, the downdip Hackberry channels, and the salt domes. This area was chosen because it is relatively sparsely drilled for Hackberry targets, a high quality 2-4 mi (3.2-6.4 km) seismic grid is available, and reasonable seismic-to-well ties are possible.

Relationship to Salt Domes

In detail, the slide scar at the northern limit of the Hackberry play in western Calcasieu Parish is somewhat irregular and can be subdivided into headlands (remnants of lower Frio shelf sediments) and arcuate embayments (areas of slides of lower Frio shelf edge or delta-front deposits) (Figure 6). The embayments occur directly updip of the salt domes, and the headlands occur directly updip of the interdome areas. This pattern suggests that direct slumping could have been triggered by oversteepening of the shelf edge due to salt withdrawal out of the secondary rim synclines of the three salt domes to the south. Similar examples of extensive sliding and slumping associated

End_Page 595------------------------------

with salt movement in the Texas Yegua have been interpreted by Edwards (1990). Seismic line 2 (Figure 7) shows at least three coherent slides which appear to have rotated towards the axis of the secondary rim syncline on the north side of the Starks salt dome. Also evident on this line are the anomalous fill sequences which onlap and fill the rotated hanging walls of the slides. These are inferred to be sand prone by comparison with cross section AA' in the northeastern part of the study area (Figure 6), where close well control exists.

The close association of the salt domes and the slide areas suggests that salt withdrawal played a major role in the initiation of the slides and may have additionally contributed to the rise in relative sea level and maximum flooding event in the middle Oligocene, after deposition of the Hackberry sandstones.

Slide Characteristics

Benson (1971), in a study north of the Manchester area (Figure 3), noted the existence of small rotated blocks and also that they contained anomalous sand sequences. Evidence from western Calcasieu Parish suggests that such blocks probably exist along all the Hackberry trend and that they have not been sufficiently tested by drilling in many areas.

Three seismic lines and two detailed cross sections show the characteristics of the rotational slide features in the study area. Only one of the seismic lines (line 1) is close to a cross section (AA') (see Figure 6). Cross sections were constructed in areas where the well spacing was 3000 ft (900 m) or less in order to document the detailed changes in stratigraphy. Seismic lines show that the individual rotated blocks are 0.5 mi (0.8 km) or

End_Page 596------------------------------

less wide in a dip direction. One seismic line in the study area (not shown) shows the strike dimension of individual blocks to be less than 1.5 mi (2.4 km). The whole zone, containing many rotated blocks, is generally less than 4 mi (6.4 km) wide in a dip direction, but very extensive in a strike direction.

Cross section AA' (Figure 8) is oriented perpendicular to the sliding direction and is subparallel to seismic line 1 (see Figure 6 for location). This seismic line (Figure 9) shows at least three rotated slide blocks west of the no-data zone. Detailed correlations of the lower Frio markers in the wells in the cross section indicate a similar number of slide blocks to be present. The northernmost slide block is eroded down almost to the Nodosaria blanpiedi marker (Figure 8). The erosion in the paleobathymetric lows suggests that gravity flows were a major mechanism for emplacement of the fill sequences (Figure 8). The lower Frio "A" and "B" sands are missing and are thought to have been eroded or removed by subsequent sliding from the crest of the rotated block. Comparison to a complet section indicates that more than 600 ft (185 m) must have been

End_Page 597------------------------------

removed from the crest of this slide block. The missing sands may have been incorporated into mass flows and redeposited downdip in a slope-channel system. In contrast, the southernmost slide block shows an almost complete section of lower Frio sands and has been only partially eroded at the crest. The inconsistency of the crestal erosion may suggest that secondary sliding was the mechanism for strata removal rather than erosion above effective wave base. The resultant pattern of submarine unconformities within the slide area is very variable and extremely localized.

After the sliding, sands were deposited in the lows created by the rotational faulting. These fill sequences (Figure 8) are anomalously thick, very localized, and consist of both fining-upward and coarsening-upward cycles. This sand was almost certainly transported off the shelf and onto the upper slope by turbidity currents. Some of the sand was probably deposited in the lows, whereas the remainder was entrained downslope in the turbidity current. Rapid deposition could have occurred due to the hydraulic jump which would have been created as flows passed over the exposed slide scars, similar to the processes proposed for sand-rich flows by Mutti and Normark (1987), who also suggested that large scour features can also form as a result of hydraulic jump. This may be an alternative exp anation for the erosion of large portions of the lower Frio slide blocks. In map view, the fill-sequence sand bodies form strike-oriented, lens-shaped sand bodies within the area of slumping, reflecting the arcuate nature of the failure surface. Seismic line 1 (Figure 9) shows that the fill sequence thins downdip within each slide block and is thinner overall in the most downdip slide block. The fill sequence is also relatively flat lying and onlaps the rotational slide blocks (Figure 9). A very similar relationship in a shallow glide-plane system has been described in the Yegua by Edwards (1990).

Dipmeter logs are invaluable tools for identifying

End_Page 598------------------------------

slide blocks, but they were not routinely run by the various operators in the study area. A dipmeter from well 7 in cross section AA' shows a dip change from about 4 degrees to the east above the rotational block to about 20 degrees to the northwest within the block (Figure 10). The dipmeter also indicates some drape of the shales near the crest of the slide block between 7470 and 7510 ft (2277-2289 m).

Cross section BB' (Figure 11) is located northeast of Edgerley field (Figure 6) and shows a single rotational fault block in the slide area with an apparent dip dimension of 3000-4000 ft (900-1200 m). The straight hole at well 12 penetrated the footwall of the slide, and the sidetrack penetrated the hanging wall. The strata in the footwall are almost flat lying in the orientation of this cross section because both the sidetrack and the straight hole penetrated the Discorbis "D" marker. The strata in the slide block have an apparent northeast dip of 11 degrees, shown by the correlation of the lower Frio markers. The sidetrack also penetrated a thick sandy fill sequence very similar in character to the fill sequence seen in cross section AA' (Figure 8). Some erosion or sliding of the lo er Frio sand sequence has occurred towards the crest of the rotational block, as in cross section AA'.

Occasionally, a well in the study area shows repeated stratigraphy, indicating that the sliding process did not occur as a single event but progressed updip as a type of headward erosion process. The Halbouty 1 Plauche (in Sec. 2, T9S, R11W) shows at least 270 ft (82 m) of repeated section (Figure 12). The mechanism by which a repeated stratigraphy might occur in this environment

End_Page 599------------------------------

is shown in Figure 13. Similar examples of repeated section have been reported from the North Sea (M.T. Richards, 1990, personal communication).

Seismic line 3 (Figure 14) shows the extent of the slumping in the central part of the study area, where at least four rotated blocks can be distinguished. The orientation of the seismic lines is critical to the interpretation of the slides because of the irregular outline of the northern boundary of the slide zone. For example, seismic lines 1 and 3 are perpendicular to each other, but both are perpendicular to the local slide-failure direction.

Slope Channels

The multiple failures along the shelf edge most likely initiated gravity flows which remobilized some of the lower Frio sediments. The mass movement flows were transported to the south along paleobathymetric lows and diverted around the highs. At least three major channels were recognized in western Calcasieu Parish, and each one can be traced back to an area of sliding updip of a salt dome (Figure 6). Many turbidity currents subsequently followed the same meandering pathways. The salt-controlled paleobathymetric highs at the location of present-day salt domes diverted the flows and caused meanders to form on their northern flanks, creating structural/stratigraphic traps.

The channel complexes were mapped using a combination of strike-oriented seismic lines, well logs, paleontologic data, and one conventional core. The seismic lines show very subtle synclines where channel systems cross, and well logs show blocky or fining-upward character when a channel complex has been penetrated.

End_Page 600------------------------------

However, because of the apparent meandering nature of the systems, regional strike-oriented seismic lines may not always be in an optimum orientation. The channel systems range in width from approximately 0.5 mi (0.8 km) to 1.5 mi (2.4 km), and their thickness ranges from less than 50 ft (15 m) to about 500 ft (150 m). Since the major channel complexes always occur at the base of the Hackberry, paleontologic data is critical for determining where the pre-Hackberry occurs. Proximity to a channel system is sometimes indicated by thin siltstones or sandstones at the base of the Hackberry.

Cross section DD' (Figure 15) was constructed across a slope channel complex where close well spacing, paleontologic data, and a high quality, strike-oriented seismic line were all available. This productive Hackberry channel complex has a maximum width of 4500 ft (1375 m), and correlation of markers shows that it has eroded about 100 ft (30 m) into the underlying sequence. The top of the channel is very flat topped and shows no apparent evidence of differential compaction. Proximity to a channel complex is indicated by the presence of hydrocarbon-bearing silts or thin sands at the edges. Foraminiferal data indicate that the channel sands were deposited on an upper slope on top of outer shelf lower Frio deposits. Seismic line 5 (Figure 16) is located exactly along the line of cross se tion and shows the subtle erosional base and flat-topped nature of the channel complex. Although difficult to interpret without well control, the high-amplitude reflector at the top of the channel appears to onlap the channel edges and seems to coincide with the productive part of the channel. A conventional core from the Arco 1 Hoffpauir (in Sec. 35, T10S, R11W, 3 mi--4.8 km--south of this seismic line) shows the channel fill to be a fining-upward sequence with medium-grained sandstone and many clay intraclasts at the base. The sandstone is also silica cemented, contains carbonaceous material, and shows parallel lamination at the top.

Known Field Locations

The larger fields within the play are found within the channel sands in the downdip portion of the play or in channels on the northern flank of paleohighs. The fields on the northern flanks of the salt domes appear to be associated with large meanders in the channels. In this particular location, turbidity currents would have "banked up" on the outer part of the bend. The upper part of the flow may have left the channel, and the coarser material would have been retained within the channel and would have decelerated and been rapidly deposited. This process, called "flow stripping," is similar to that proposed by Piper and Normark (1983) on the modern Navy Fan.

The rotated slide blocks are sometimes productive, but the two examples in western Calcasieu Parish are single-well producers surrounded by dry holes. Detailed production data on these two wells were not available. Many of the slide blocks could be productive, but their areal extent is small and the dip within them can be very steep.

We found no confirmed production from the sandy fill sequences. This is not surprising considering that the targets are irregularly shaped and less than 1 x 0.5 mi (1.6 x 0.8 km) in plan view. The recently discovered Morgan Bluff field in Texas is along depositional strike of the rotational slide zone

End_Page 601------------------------------

(Figure 6), but well logs and seismic data from this field were not available to us.

CONCLUSIONS

Multiple slope failures prior to the deposition of the Hackberry Formation were probably initiated by a combination of sediment loading during the late stages of relative sea level fall and by oversteepening near the shelf edge, probably caused by concurrent salt withdrawal.

Detailed mapping of the Hackberry must be carried out using high-quality seismic data, dipmeters, and all available paleontologic data. Mapping in western Calcasieu Parish, Louisiana, has shown that some of the

End_Page 602------------------------------

undiscovered reserves within the play may be contained within three types of reservoir sand within the Hackberry and displaced lower Frio sequences:

(1) Lower Frio shelf-edge sands in rotated slide blocks. Individually, their maximum dimensions are 0.5 mi (0.8 km) in a dip direction and 1.5 mi (2.4 km) in a strike direction. They occur in a zone up to 4 mi (3 km) wide and are identified from seismic and dipmeter data and detailed correlation of well logs. Small potential hydrocarbon traps may exist in the many coherent rotated slide blocks in the updip portion of the Hackberry play. Dip-oriented seismic lines may not always show the slide interval due to the arcuate nature of the slide scars and the rotational nature of the slide blocks.

(2) Sandy fill sequences confined to the lows created by the rotated slide blocks. These are thick, anomalous strike-oriented sand bodies identified from seismic and by well log correlation. These sequences are lens shaped in map view and onlap the rotated slide blocks. We did not identify any hydrocarbon production specifically from this type of reservoir.

(3) Channel sands deposited downdip of the rotational slide area. Good quality, strike-oriented seismic lines, nearby well logs, and paleontologic data are the most useful for determining the locations of the channel

End_Page 603------------------------------

sandstones, although their expression is very subtle. The majority of historical Hackberry production occurs from this type of reservoir. Hydrocarbon traps exist where turbidite channels were forced to meander around paleobathymetric highs and where faults intersect the channels.

Statistical analysis of historical production and discovery data shows that total reserves of 117 million BOE in 41 fields of between 1 million and 10 million BOE remain to be found in the Hackberry play. The potential for more than 100 fields of between 0.1 million and 1 million BOE also exists within the play.

References:

Benson, P. H., 1971, Geology of Oligocene Hackberry trend, Gillis English Bayou-Manchester area, Calcasieu Parish, Louisiana: Gulf Coast Association of Geological Societies Transactions, v. 21, p. 1-14.

Curtis, D. M., and D. J. Echols, 1985, Habitat of oil and gas in the middle Frio (Oligocene) Hackberry: Gulf Coast Section, SEPM Fourth Annual Research Conference Proceedings, p. 263-274.

Edwards, M. B., 1990, Stratigraphic analysis and reservoir prediction in the Eocene Yegua and Cook Mountain Formations of Texas and Louisiana: Houston Geological Society Bulletin, December, p. 37-50.

Eubanks, L. G., 1987, North Sabine Lake field: complex deposition and reservoir morphology of lower Hackberry (Oligocene), southwest Louisiana: AAPG Bulletin, v. 71, p. 1162-1170.

Ewing, T. E., and R. S. Reed,1984, Depositional systems and structural controls of Hackberry sandstone reservoirs in southeast Texas: University of Texas Bureau of Economic Geology Circular 84-7, 48 p.

Galloway, W. E., 1987, Depositional and structural architecture of prograding clastic continental margins: tectonic influence on patterns of basin filling: Norsk Geologisk Tiddskrift, v. 67, p. 237-251.

Galloway, W. E., 1989, Genetic stratigraphic sequences in basin analysis I: architecture and genesis of flooding-surface bounded depositional units: AAPG Bulletin, v. 73, p. 125-142.

Halbouty, M. T., 1979, Salt domes; Gulf region, United States and Mexico: Houston, Gulf Publishing Company, 529 p.

Haq, B. U., J. Hardenbol, and P. R. Vail, 1988, Mesozoic and Cenozoic chronostratigraphy and eustatic cycles: an integrated approach: SEPM Special Publication 42, p. 71-108.

End_Page 604------------------------------

Jackson, M. L. W., M. P. R. Light, and W. B. Ayers, Jr., 1987, Geology and coproduction potential of submarine-fan deposits along the Gulf Coast of east Texas and Louisiana: Journal of Petroleum Technology, v. 39, no. 4, p. 473-481.

Lee, P. J., and P. C. C. Wang, 1986, Evaluation of petroleum resources from pool size distributions, in D. D. Rice, ed., Oil and gas assessment: AAPG Studies in Geology 21, p. 33-42.

Mutti, E., and W. R. Normark, 1987, Comparing examples of modern and ancient turbidite systems: problems and concepts, in J. K. Legget and G. G. Zuffa, eds., Marine clastic sedimentology: London,

End_Page 605------------------------------

Graham and Trotman, p. 1-38.

Paine, W. R., 1968, Stratigraphy and sedimentation of subsurface Hackberry wedge and associated beds of southwestern Louisiana: AAPG Bulletin, v. 52, p. 322-342.

Paine, W. R., 1971, Petrology and sedimentation of the Hackberry sequence of southwest Louisiana: Gulf Coast Association of Geological Societies Transactions, v. 21, p. 37-55.

Piper, D. J. W., and W. R. Normark, 1983, Turbidite depositional patterns and flow characteristics, Navy Submarine Fan, California borderland: Sedimentology, v. 30, p. 681-694.

Van Wagoner, J. C., R. M. Mitchum, K. M. Campion, and V. D. Rahmanian, 1990, Siliciclastic sequence stratigraphy in well logs, cores and outcrops: concepts for high resolution correlation of time and facies: AAPG Methods in Exploration Series 7, 55 p.

End_of_Article - Last_Page 606------------

Figure Captions/Table Heads:

Figure 1. Map of southeast Texas and southwest Louisiana showing location of the Hackberry play. Northern limit of outline represents the Hackberry subcrop. The southern boundary is presently limited by drilling economics to where the base of the Hackberry Formation is at about 15,000 ft (4600 m).

Figure 2. Chronostratigraphic diagram of the Oligocene showing the relationship of the Hackberry to the upper and lower Frio prograding sequences. Slumping occurred after the relative fall in sea level caused by the progradation of the lower Frio shelf. Absolute ages are from Curtis and Echols (1985). The three potential Hackberry reservoir sandstones are (1) rotated slide-blocks of shelf-edge sediments, (2) fill sequences in the lows created by the rotational faulting, and (3) narrow, sand-filled submarine channels. For the purposes of this diagram, the channels are drawn strike-parallel; their true orientation is dip-parallel. No horizontal scale is intended.

Click to view image in GIF format. Figure 3. [Color] Simplified hydrocarbon production map showing fields which have produced only from the Hackberry play (includes slide blocks of the lower Frio sediments). Data is synthesized from Table 1. Salt dome locations derived from Halbouty (1979). Green outline shows detailed study area in western Calcasieu Parish of Figure 6. Gas symbols show 1988 Hackberry play discoveries by Prairie Producing and Exploration Company of Louisiana. EUR = estimated ultimate recovery; BOE = barrels of oil equivalent.

Figure 4. Log-normal plot showing discovery sequence of all Hackberry fields with estimated ultimate recoveries of more than 1 million BOE. Note how the largest field was discovered during the first 5 years of exploration and how the size of major discoveries (>10 million BOE) decreased as exploration progressed (arrowed line). The names of the largest and most recent fields discovered in the play are shown.

Click to view image in GIF format. Figure 5. [Color] Plot of a primes output showing log normal distribution. Blue slots show known fields; red slots show the relative sizes of yet-to-be-found (YTF) fields. Plot is created from Table 1 using only fields with an EUR of more than 1 million BOE.

Click to view image in GIF format. Figure 6. [Color] Generalized depositional environments of the Hackberry play and Hackberry production within the western Calcasieu Parish study area. Location of example seismic lines (1, 2, 3, and 5) and cross-sections (AA', BB', and DD') shown in green. Complete seismic grid used for interpretation is not shown. "Hackberry shelf" refers to remnants of lower Frio shelf sediments which were not affected by sliding during deposition of Hackberry sediments. See Figure 3 for location of map.

Click to view image in GIF format. Figure 7. [Color] North-south seismic line 2 across Starks salt dome. The rotated slide blocks of lower Frio shelf-edge sediments remained coherent and slid toward the secondary rim syncline of the salt dome. Note the pinch-out of the Hackberry shale to the north, landward of the approximate location of the late Frio shelf edge. See Figure 6 for location of line.

Figure 8. Structural cross section AA' across the northern boundary of the Hackberry play (see Figure 6 for location). Note the abrupt pinch-out of the Hackberry shale, the anomalous fill sequence(s), and the abrupt change in dip determined from correlation of log character. A dipmeter was available in well 7 (see Figure 10). See Table 2 for well locations.

Click to view image in GIF format. Figure 9. [Color] Seismic line 1, showing rotational slide blocks of lower Frio and shallow decollement surface. The fill sequence appears to onlap the rotated blocks. Line is located within 1 mi of cross section AA' (see Figure 6). Individual slides are about 0.5 mi (0.8 km) wide.

Figure 10. Dipmeter from well 7 in cross section AA' (see Figure 8). Note the abrupt change of dip below 7510 ft (2289 m) to approximately 20 degrees northwest within the rotated slide block. Note also apparent drape on top of slide block between 7470 and 7510 ft (2277-2289 m).

Figure 11. Structural cross section BB' across one slide block (see Figure 6 for location). The vertical hole of well 12 penetrated the footwall, and the sidetrack penetrated the hanging wall and the fill-sequence sandstones. Some erosion of the slide block in the southwest is inferred by correlation of markers in lower Frio sequence. See Table 2 for well locations.

Figure 12. Repeated section in the Halbouty 1 Plauche (Sec. 2, T9S, R11W). Repetition of excellent lower Frio markers shown. The repetition was probably produced by multiple slide events, as shown in Figure 13.

Figure 13. Process of multiple sliding or slope adjustment by headward erosion which could cause repeat section as seen in Figure 12.

Click to view image in GIF format. Figure 14. [Color] Seismic line 3 showing rotational slides of lower Frio sediments and shallow decollement surface. Individual slides are less than 0.5 mi (0.8 km) wide. See Figure 6 for location of line.

Figure 15. Cross section DD' across a productive Hackberry slope channel complex. Note interpreted depositional environment change from outer shelf to upper slope across the basal Hackberry contact and the absence of thick sands just 2000 ft (600 m) laterally from the channel complex center. The top of the sandstones appears flat, and correlation shows the base is eroded about 100 ft (30 m) into the underlying sequence. See Table 2 for well locations, and Figure 6 for cross section location.

Click to view image in GIF format. Figure 16. [Color] Seismic line 5. A strike-oriented line across the same productive Hackberry channel complex as shown in cross section DD' (Figure 15) using the same three wells. Note the slightly higher amplitude at the channel complex top and the subtle onlap of high amplitude onto the margin. See Figure 6 for line location.

Table 1. Texas (TX) and Louisiana Hackberry Fields*

Table 2. Locations of Wells used in Cross Sections AA', BB', and DD'

Acknowledgments:

(1) Manuscript received February 7, 1991; revised manuscript received August 14, 1991; final acceptance December 13, 1991.

(2) BP Exploration, P.O. Box 4587, Houston, Texas 77210. Present address: BP International Ltd., Sunbury Research Centre, Chertsey Road, Sunbury-on-Thames, Middlesex TW16 7LN, United Kingdom.

(3) BP Exploration, P.O. Box 4587, Houston, Texas 77210.

The authors wish to thank Seitel Inc., Houston,for permission to publish line 2 and line 5; Richardson Seismic Services, Inc., Houston, for permission to publish line 3; and Spectrum Surveys Inc., Houston, for permission to publish line 1. Reviews and comments from David G. Roberts (BP Houston) and Dave Webster (BP Glasgow) improved and clarified the manuscript. The authors also thank BP for permission to publish the paper and for financial contributions towards the printing of the color figures. Marc Edwards, Frank Ethridge, and an anonymous reviewer provided constructive comments.

Copyright 1997 American Association of Petroleum Geologists

Pay-Per-View Purchase Options

The article is available through a document delivery service. Explain these Purchase Options.

Watermarked PDF Document: $14
Open PDF Document: $24

AAPG Member?

Please login with your Member username and password.

Members of AAPG receive access to the full AAPG Bulletin Archives as part of their membership. For more information, contact the AAPG Membership Department at [email protected].