New depositional model improves outlook for Clear Fork infill drilling

Sept. 14, 1998
For over 30 years, the Clear Fork Group (Leonardian), a major reservoir interval of the Permian Basin with nearly 1 billion bbl of original oil in place, has been interpreted as an arid or semi-arid deposit. Thin-bedded, anhydrite-bearing dolostones of the Clear Fork have been deposited as the result of cyclic sedimentation in an evaporitic 1 or, more specifically, a sabkha/salt pan-type setting. 2 Until very recently, such models have gone unquestioned despite mounting evidence for more
Scott L. Montgomery
Petroleum Consultant
Seattle

William H. Dixon
Fina Oil and Chemical Co.
Houston

For over 30 years, the Clear Fork Group (Leonardian), a major reservoir interval of the Permian Basin with nearly 1 billion bbl of original oil in place, has been interpreted as an arid or semi-arid deposit. Thin-bedded, anhydrite-bearing dolostones of the Clear Fork have been deposited as the result of cyclic sedimentation in an evaporitic1 or, more specifically, a sabkha/salt pan-type setting.2 Until very recently, such models have gone unquestioned despite mounting evidence for more complex depositional origins.

Important interpretations of the last few years, meanwhile, have eliminated the sabkha "requirement," proposing instead a succession of open marine, shoal, lagoon, and tidal flat environments,3 4 but still do not account for certain crucial phenomena observed in cores from various fields-in particular, abundant plant debris and solution-collapse features.

This article summarizes a new, more comprehensive depositional model of the Clear Fork based on over 10,600 ft of core from the North Robertson Unit (NRU), Gaines County, Tex. One of the larger Clear Fork pools, NRU is the subject of an extensive, ongoing reservoir characterization project supported, in part, by the U.S. Department of Energy as part of its Oil Field Demonstration Project. Petrophysical and engineering data from this project have been discussed in other reports.5 6 7 8 9 The new depositional model is considered to have potentially important consequences for strategies regarding future enhanced recovery and exploratory operations.

General characteristics

As shown on Fig. 1 [181,930 bytes], the major Clear Fork producing trend is concentrated along the eastern margin of the Central Basin Platform (CBP) and Northern Shelf of the Midland Basin, with subsidiary production along the Eastern Shelf. Clear Fork strata comprise a portion of the Leonardian platform-shelf margin that prograded basinward between Wichita/Abo and Glorieta deposition (Fig. 2 [155,079 bytes]). Downdip, basinal equivalents to the Clear Fork include the Dean and lower Spraberry formations. The largest Clear Fork pools, such as Fullerton, Goldsmith, and North Robertson, have total productive areas as large as 4,500-6,000 acres, with recoverable reserves of 80-200 million bbl. Reservoirs include multiple, thin (5-35 ft) zones of dolomitized carbonate characterized by secondary pore systems.4 10 11 12

Intercrystalline porosity is dominant, with moldic and vug types present. Increasing evidence also suggests that fracture porosity, related to solution collapse zones, is significant in a number of fields. Non-fracture porosities and permeabilities commonly range from 7-15% and 0.5-30 md, respectively. Reservoirs produce 35-42° gravity oil and are solution-gas driven. Oil entrapment is related to large, gentle, asymmetric anticlines developed over faulted basement structures. Thickness of the oil column in Clear Fork fields varies from 100 ft to over 1,400 ft, with net pay ranging from less than 50 ft to 360 ft. A low or variable net:gross pay ratio reflects both the thinness of the individual reservoir zones and also their lateral discontinuity and heterogeneity.

Most Clear Fork fields were discovered in the 1940s-50s and are now under waterflood. Primary recovery has typically been 6-15% of estimated original oil in place (OOIP), with incremental production (combined waterflooding and infill drilling) of 5-11%. Low recoveries are attributable to the heterogeneous nature of the reservoir zones.

At the same time, wells have typically been perforated across large intervals or, if selectively, on the basis of log-determined cutoffs, leaving significant uncertainty with regard to the specific distribution of productivity. Actual testing to delineate zone productivity has seldom been performed, particularly in the middle and upper Clear Fork.

Both exploration and development have tended to focus particularly on the lower portion of the Clear Fork, where shoal facies are more abundant and production rates appear higher. Data recently acquired at NRU5 6 and also at North Riley field4 indicate that reservoir quality is also significant in other facies interbedded with, and lying above, shoal-type grainstones. Such data, in general, show that reservoir complexity is due to a combination of:

  1. Original diversity and intercalation of depositional facies, and
  2. "Blurring" of original facies boundaries by dolomitization, resulting in a complicated distribution of porosity and permeability. The second of these factors has made it difficult to accurately predict facies distribution on the basis of logs alone. Nonetheless, there is an important degree of correlation between better reservoir quality rock and specific depositional facies, making the question of depositional models significant.

The Clear Fork 'problem'

It is widely recognized among geologists who have studied core from the Clear Fork that large amounts of plant debris exist in many locations. This is true not only of fields on the CBP and Northern Shelf but on the Eastern Shelf as well (e.g. Westbrook field, Mitchell County). Cores from some fields show hundreds of feet of such debris: At NRU, thin coal beds, root traces, large fern fronds, and intact palm leaves are all observed as well. Fig. 3 [133,404 bytes] provides several examples. In general, such debris is volumetrically too large and widespread to be explained as the result of local phenomena such as fresh-water springs or "oases."

In addition, zones of solution-collapse breccia (sometimes oil-stained), often subtle in appearance, have now been identified in a number of fields, where they usually correspond to structurally higher areas. At NRU, zones as thick as 40 ft have been noted with conspicuous fracturing and good permeability.9 It is now suspected that such zones have contributed significantly, at times principally, to production from certain portions of the Clear Fork and may also act as "thief" zones for injected water during secondary recovery.

Finally, detailed core study at NRU and elsewhere suggests that a wider diversity of facies exists within the Clear Fork than commonly recognized. Distinctions among different varieties of grainstone shoal deposit, as well as identification of biohermal and beach facies, open and restricted lagoonal facies, can all be made.

These observations represent a clear "problem" for earlier depositional models and have not been regularly reported in previous publications on the Clear Fork. Their consistency and potential significance at NRU requires that they be integrated into a revised model that fully accounts for core-based study and analysis.

New facies model: Discussion

The model of Fig. 4 [116,391 bytes] has been devised to account for the full diversity of lithotypes observed in core from NRU, encompassing both the Clear Fork and lower Glorieta. It includes a total of 15 separate facies types, some of which are highly localized in occurrence. In general, offshore or seaway areas existed to the east and northeast. The most significant facies and their respective lithologies include the following:
  • Open shelf/forebank.
Wackestone-grainstone deposits showing significant bioturbation and burrowing were deposited in open marine waters, seaward of shoal bank features, mainly below wave base. Grainstones, comprised of a variety of different allochems, are less abundant and reflect deposition in shallower water settings or at/above storm wave base.
  • Fusulinid shoals.
This facies includes grainstones particularly rich in fusulinid allochems. These were apparently worked into low-relief shoals by wave action. Extensive leaching of fusulinid tests has resulted in abundant moldic porosity, often filled by secondary anhydrite cement.
  • Shoal bank.
Well-sorted grainstones and subordinate packstones, containing a variety of allochems (bryozoans, crinoids, bivalves, oncoids, peloids) comprise this lithofacies. Dolomitization has obscured most original sedimentary structures; however, occasional scour-and-fill, graded bedding, rip-up clasts, and cross lamination are observed. Such features, plus the absence of carbonate mud, testify to higher energy conditions.
  • Reef.
Three separate lithofacies are interpreted as reefal deposits. These include:
  1. Rudstones, floatstones, and boundstones (cyclostome bryozoan in growth position, rugose coral, crinoid, sponge, gastropod fragments) comprising the reef core;
  2. Coarse, skeletal grainstones of the reef talus facies, showing bedding planes dipping up to 30° landward; and
  3. Non-bedded, less coarse grainstones composed of reefal material showing no sedimentary structures, interpreted as reef debris aprons. Debris apron strata appear to coalesce between individual reef bodies and grade into shoal deposits.
  • Lagoon.
Extensively burrowed, dolomitized skeletal-peloid wackestones and packstones that occur stratigraphically landward of shoal and reef debris apron deposits are interpreted as reflective of lagoonal sedimentation. Open lagoonal facies lack significant anhydrite, except as a late-stage component; restricted lagoonal facies, in contrast, contain abundant finely distributed and scattered nodular anhydrite, as well as wispy laminations. Open lagoonal strata have moderate-to-good porosities and permeabilities, due in part to leaching of allochems. In total, lagoonal deposits comprise as much as 20-25% of the total rock volume of the middle and upper Clear Fork.
  • Island complex.
A series of deposits occurring within the lagoonal area is identified as possible island interior, margin, beach, and algal mat facies. Wackestone-packstone deposits with root casts, dessication cracks, abundant plant fragments are interpreted to comprise the subaerial portion of the complex, with massive-to-burrowed mudstones representative of high-salinity ponds. Unburrowed packstone-grainstone sediment with rounded grains, finely broken plant debris, and inclined bedding represent beach deposits and are intebedded with occasional thin-bedded algal mat material exhibiting fenestral texture.
  • Tidal flat.
Finely crystalline dolomudstone, wackestone, and packstone, showing evidence of burrowing, occasionally preserved lamination, and nodular anhydrite is considered representative of tidal flat deposits. Where they display high-angle cross bedding, abundant scour surfaces, and burrowing, the sediments are interpreted as tidal channel deposits. Faunal diversity in this facies is low. Interbedded sediments that are finely laminated, non-fossiliferous, and show fenestral texture comprise a tidal flat algal mat facies.
  • Supratidal.
Deposits of this facies consist of cryptocrystalline dolomudstones, with bedded anhydrite, rip-up clasts, soil zones, root casts, and dessication cracks. Supratidal/exposed deposits also contain disseminated detrital quartz and, locally, abundant plant fragments. These sediments are principally restricted to the Glorieta formation.
  • Solution collapse breccias.
These deposits include broken, leached, partly rotated fragments of supratidal, island, and possibly other facies. Coal beds up to several centimeters thick, plant debris zones, lag soils, and other evidence of exposure also occur in association with collapse breccias. These breccias occur in the middle and upper Clear Fork and are seen to be oil stained at NRU, where they appear to have acted as water injection "thief" zones.

The model of Fig. 4 is considered to reflect deposition under hot and relatively humid climatic conditions. An abundance of fresh water is required for the volume of vegetal cover indicated, as well as solution features. It should be noted that the model proposes a shifting mosaic of facies that more accurately accounts for the lateral and vertical heterogeneity observed in the Clear Fork than do previous schemes.

Facies associations

Clear Fork deposits exhibit complex interlayering between facies due to lateral migration of original depositional regimes. As first proposed by Todd and Silver,1 this a direct result of interaction between eustatic sea level cycles and subsidence along the Clear Fork platform and shelf edge or seaway. At NRU, a ramp/platform model appears to fit the general facies distribution, as follows:
  1. Outer ramp/outer shelf-margin = open shelf/forebank;
  2. Shallow ramp/inner shelf-margin = shoal, reef, open lagoon, island; and
  3. Back ramp/platform interior = restricted lagoon, tidal flat, supratidal.
It should be noted that this model proposes the absence of a distinct shelf-edge in the NRU area. Instead, it suggests this area lay adjacent a shallow seaway with water depths within the NRU probably no more than about 20 ft during Clear Fork deposition. The actual platform margin existed some distance to the northeast.

Basinward progradation of the back ramp is clearly indicated by upward transition from open marine and shoal deposits of the lower and lowermost middle Clear Fork to lagoonal, tidal flat, and supratidal deposits of the middle and upper Clear Fork. Island complexes are also particularly abundant in the middle and upper Clear Fork. It is significant that these complexes commonly occur above solution collapse zones, suggesting a possible degree of subtle structural control.

Reservoir quality distribution

While Clear Fork reservoirs at NRU generally have low permeabilities (<5 md), petrophysical and well test data indicate that higher recoveries are associated with better quality reservoir rock. In the lower Clear Fork, such rock consists mainly of shoal grainstones, which exhibit above-average continuity. In the lower part of the middle Clear Fork, better quality reservoirs exist in shoal and reef debris apron facies. These shallow ramp facies are scarce to absent in the upper two-thirds of the Clear Fork, where open lagoon deposits comprise the most important reservoir facies.

As noted, most development to date has either concentrated on the lower part of the Clear Fork or has not been based on detailed knowledge of zonal productivity. This leaves significant, possibly overlooked potential in the middle and upper Clear Fork, particularly in lagoonal facies and solution collapse breccias, which could be selectively perforated in targeted infill wells to recover bypassed reserves.

Conclusions

Systematic study of new and existing core information on the Clear Fork Group, West Texas, suggests that the interval was deposited under hot and humid conditions in a complex array of specific environments ranging from nearshore marine, shoal, reef, and island complexes to lagoonal, tidal flat, and supratidal settings. Vegetal covering of island and supratidal areas was intermittently widespread. In the upper part of the Clear Fork, where peritidal and lagoonal facies dominate, solution collapse breccias testify to fresh water invasion during periods of sea level lowstand.

The implications of this new model for Clear Fork reservoir development are significant. Targeted infill drilling of lagoonal facies and more study of solution breccias in the upper part of the Clear Fork, to determine their reservoir or "thief zone" potential, are among such implications.

References

  1. Silver, B.A., and Todd., R.G., Permian cyclic strata, AAPG Bull., Vol. 53, No. 11, 1969, pp. 2,223-51.
  2. Handford, C.R., Coastal sabkha and salt pan deposition of the Lower Clear Fork formation (Permian), Texas, Journal of Sedimentary Petrology, Vol. 51, 1981, pp. 761-778.
  3. Saller, A.H., Lithologic controls on production history and recovery efficiency in some Paleozoic carbonate reservoirs, West Texas, USA, in Carbonate rocks-hydrocarbon exploration & reservoir characterization, Special Publication, Technology Research Center, Japan National Oil Co., No. 3, 1994, pp. 103-115.
  4. Saller, A.H., and Henderson, N., Distribution of porosity and permeability in platform dolomites: insight from the Permian of West Texas, AAPG Bull., Vol. 82, No. 8, pp. 1,258-82.
  5. Davies, D.K., Vessell, R.K., and Auman, J.B., Improved prediction of reservoir behavior through integration of quantitative geological and petrophysical data, SPE Paper 38914, 1997, 16 pp.
  6. Davies, D.K., Vessell, R.K., Doublet, L.E., and Blasingame, T.A., Improved characterization of reservoir behavior by integration of reservoir performance data and rock type distributions, Fourth International Reservoir Characterization Technical Conference Proceedings, 1997, pp. 645-669.
  7. Montgomery, S.L., Permian Clear Fork Group, North Robertson Unit: Integrated reservoir management and characterization for infill drilling, Part I-Geologic Analysis, AAPG Bull., Vol. 82, No. 10, 1998, in press.
  8. Montgomery, S.L., Permian Clear Fork Group, North Robertson Unit: Integrated reservoir management and characterization for infill drilling, Part II-Petrophysical and Engineering Data, AAPG Bull., Vol. 82, No. 11, 1998, in press.
  9. Kamis, J., Dixon, W., and Vessell, R., Environments of deposition for the Clear Fork and Glorieta formations, North Robertson Unit, Gaines County, Texas, in Johnson, K.S., ed., Platform Carbonates in the Southern Mid-Continent, Oklahoma Geological Society Circular, forthcoming.
  10. Mazzullo, S.J., Stratigraphy and depositional mosaics of Lower Clear Fork and Wichita Groups (Permian), northern Midland Basin, Texas, AAPG Bull., Vol. 66, 1982, pp. 210-227.
  11. Ruppel, S.C., Styles of deposition and diagenesis in Leonardian carbonate reservoirs, West Texas, SPE Paper #24691, 1992, pp. 313-320.
  12. Holtz, M.H., Ruppel, S.C., and Hocott, C.R., Integrated geologic and engineering determination of oil-reserve growth potential in carbonate reservoirs, Journal of Petroleum Technology, Vol. 44, 1992, pp. 1,250-57.

The Authors

Scott L. Montgomery is a Seattle petroleum consultant and author. He is lead author of the "E&P Notes" series in the AAPG Bulletin and the quarterly monograph series "Petroleum Frontiers" published by Petroleum Information/Dwights LLC. His current research interests include frontier plays and field re-development in North America. He holds a BA degree in English from Knox College and an MS degree in geological sciences from Cornell University. E-mail: [email protected]

William H. Dixon is a senior exploration adviser at Fina Oil & Chemical Co. in Houston. He began his career with Marathon Oil Co. and also worked for Tenneco Oil Co. prior to joining Fina in 1990. Over his 26 years in the oil business he has designed drilling, completion, and production facilities and specialized in waterflood oil recovery. His geologic responsibilities are in West Texas and Southeast New Mexico, including petrographic and structural interpretations. He holds BS and MS degrees in geology from the University of Michigan.

Copyright 1998 Oil & Gas Journal. All Rights Reserved.