Click to view PDF images of illustrations that accompanied oral presentation.
Reservoir Geology of ERT (Penn Sand) Field
Potter County, Texas: A Case For Cores*
By
Robert W. Von Rhee1
Search and Discovery Article #20010 (2002)
*Adapted as online version from oral presentation to Tulsa Geological Society, October 1, 2002
1Rockford energy Partners, LLC, Tulsa Oklahoma ([email protected])
Abstract
The ERT (Penn Sand) Oil Field was discovered in 1991 by Horizon Oil and Gas. It is located in the northern Palo Duro Basin. By the end of 1994, seven wells were drilled with a 71% success ratio (5/7). At this time the field remained incompletely developed with only a southern limit evident. In 1995 Tide West Oil Company acquired the field based on an estimated EUR of 1MMBO from producing wells and another 460 MBO recoverable via two proved undeveloped locations. The Penn Sand log signatures indicated a low-contrast shaly sand ranging from 70 to 100 feet thick and broken by 1- to 3-foot-thick beds with little or no porosity. Because of the internal tight beds, the reservoir was interpreted to be highly heterogeneous and therefore not valued as a suitable candidate for secondary waterflood recovery. Tide West continued to develop the field in 1995, drilling eight wells with an 88% success ratio (7/8). The analysis of whole core taken from two of these wells indicated the reservoir is a ripple laminated, shaly sandstone turbidite. The cores revealed that the “tight beds” interpreted from the logs were actually irregularly shaped regions of cemented sandstone in which nearly all the porosity was occluded by calcite cementation. The contact between cemented and non-cemented sand was about three sand grains wide, or less than one millimeter. The contacts cut across sedimentary structures and bed boundaries and ranged from sub-horizontal to vertical within the four-inch dimension of the core face. Porosity distribution was bimodal, clustering from 11 to 16% for uncemented zones and 3 to 6% for the cemented intervals. The pore system is very fine (pore throats 5-10 micrometers) yet homogeneous. The sharp transition between porosity regimes resulted in significant open-hole log understatement of porosity thickness and magnitude in zones less than 1.5 feet thick. Schlumberger’s Formation Micro-Imager tool (FMITM) successfully resolved the complex contacts between cemented and uncemented sandstone porosity. The dominance of a single sedimentary process creates a level of lateral continuity and isotropy. The reservoir is highly heterogeneous on a small scale, but more laterally and vertically continuous on the inter-well scale. Because of the core studies, Tide West eventually concluded a waterflood was feasible. Estimated secondary recovery potential was 3 MMBO, with a 1996 present value of nearly $20 million.
uDevelopment of ERT field after acquisition tWell-log & Penn Sand interpretation
uDevelopment of ERT field after acquisition tWell-log & Penn Sand interpretation
uDevelopment of ERT field after acquisition tWell-log & Penn Sand interpretation
uDevelopment of ERT field after acquisition tWell-log & Penn Sand interpretation
uDevelopment of ERT field after acquisition tWell-log & Penn Sand interpretation
uDevelopment of ERT field after acquisition tWell-log & Penn Sand interpretation
uDevelopment of ERT field after acquisition tWell-log & Penn Sand interpretation
uDevelopment of ERT field after acquisition tWell-log & Penn Sand interpretation
uDevelopment of ERT field after acquisition tWell-log & Penn Sand interpretation
uDevelopment of ERT field after acquisition tWell-log & Penn Sand interpretation
uDevelopment of ERT field after acquisition tWell-log & Penn Sand interpretation
uDevelopment of ERT field after acquisition tWell-log & Penn Sand interpretation
uDevelopment of ERT field after acquisition tWell-log & Penn Sand interpretation
uDevelopment of ERT field after acquisition tWell-log & Penn Sand interpretation
|
Figure Captions (1-17)
Click here for sequence of Figures 12 and 13.
ERT Field
Purchasing an oil field successfully requires that that its size be accurately estimated. Determination of the field size is literally the million-dollar question.
An overview map of the Texas Panhandle, with counties and geologic features, provides the setting of ERT Field (Figure 1). In 1994, Tide West Oil Company had an opportunity to acquire this partly developed oil field. The operator was Horizon Oil &Gas of Spearman, Texas. They had drilled seven wells with one dry hole and one marginal well. They had discovered what they called a “Tonkawa” sandstone reservoir. This field was located south (“on the backside”) of the Amarillo Uplift, on the northern rim of the Palo Duro Basin.
The log of the Penn Sand discovery well, Horizon O&G Bivins Ranch No. 1-3, completed December, 1991, with IPP 112 BO, 0 BW, Gas TSTM, shows the position of the producing reservoir relative to overlying Permian?/Pennsylvanian shale below Permian carbonates and underlying granite wash with mixed lithologies. Based on the position of the sand relative to the Permian carbonates, a geologist familiar with the Anadarko basin would probably think of the Penn Sand as “Tonkawa” or even “Douglas,” essentially Virgilian in age. The problem with assigning the sand as “Tonkawa” is that it attached a whole string of attributes based on the technical staff’s experience with that unit in the Anadarko Basin. The result was that there was not too much excitement about the added potential of the producing sand.
By naming it “Penn Sand,” we recognized how little we knew about the reservoir and how little we knew about its producing attributes and ultimate field size potential. This made it “new” and generated much more open-minded thinking about its potential.
At the time we evaluated ERT Field for purchase, it consisted of seven wells (Figure 2) that had been drilled in the following order:
1-2 = Granite wash oil discovery 1-3 = Penn Sand oil discovery 2-3 = T/A-- Penn sand tight 3-3 = Penn Sand oil 2-2 = Penn Sand oil 3-2 = Penn Sand oil 4-2 = Dry hole; Penn Sand absent.
For assessing the size of the field, we organized the geologic research as follows:
• Analyze field size of existing fields Field size vs. discovery date Field size distribution • Map existing data • Synthesize and draw conclusions
Also, a strong sense of urgency was given to these tasks. Palo Duro Basin references were identified and studied from personal libraries, Petroleum Abstracts, which was cost effective, AAPG Library and AGI’s GeoRef, Oklahoma Well Log Library, and miscellaneous sources (e.g., colleagues). Regional Geologic Setting
Understanding the stratigraphic and
tectonic settings (Figures 3 and 4) were considered to be
fundamental for study of the field. The tectonic framework (after Budnik,
1989) and stratigraphic
ERT Field is in the Whittenburg Trough, north of Bush Dome in the northern part of the Palo Duro Basin (Figure 8). Principal routes of sand transport are from the Amarillo Uplift and secondarily from the Bush Dome and Bravo Dome (Rose, 1986).
In ERT Field, the non-arkosic Penn Sand accumulation is positioned within major transport routes and major accumulation of sediments. Hydrocarbon traps are present in Missourian and Virgilian “wash” on the north side of the Amarillo Uplift. There is potential for distal shale-out and dip reversal against the Bush Dome in SW Potter County.
In Potter County, Texas (Figure 9), with area of 961 sq. mi., about 200 acres of the 22,000-acre lease had been developed when Tide West bid on ERT Field. About a dozen wells had been drilled within the 100 sq. mi. centered on the field. In the county, there were only 53 wells drilled deeper than 4800 feet to penetrate pre-Permian; some of these were deep enough to have encountered the Penn Sand.
Assessing Field Size
In assessing the field size, research was
undertaken to (1) analyze the field size of existing fields by plotting
field size vs. discovery date and field size distribution, (2) prepare
To analyze field size of existing fields, the efforts were undertaken to: (1) Calculate Estimated Ultimate Recovery (EUR) of each field/reservoir (2) Identify reservoir lithology (3) Identify reservoir age (4) Plot field size vs. discovery date (5) Plot field size distribution and curve
Ultimate recovery of 20 pre-Permian reservoirs in fields in northern Palo Duro Basin, as of the third quarter, 1995, was 14.5 MMBO (Figure 10). Average field size was 725 MBO, with maximum being 4.3 MMBO; median was 86 MBO.
Based on a plot of EUR vs. discovery year, there is no decrease of field size with time (Figure 11). This probably indicates that the basin probably is not fully mature.
For the distribution of pre-Permian field size, a 5th order trend line is close to fitting the data. The trend line shown in Figure 12 is “eyeballed,” based on analysis of the distributions of field sizes in more mature basins. The normal truncation of the “bell curve” for the very small fields is the economic effect of drilling for oil and gas. At some low-end economic point, the “field” turns into a “show” and drops off the chart (Figure 12). Based on this curve and the position of the field on the curve in its undeveloped state, a range of field sizes can be estimated. The most likely Penn Sand field size is estimated to be 1.5-2.5 MMBO (Figure 13); the size of a “one-of-a-kind” field is 3.2-4.75 MMBO.
In
Completion data for the first two wells
drilled by Horizon, Bivins Ranch No. 1-2 and Bivins Ranch No. 1-3 (Figure
14), are IPP 105BO/48BW/TSTM (granite wash) and IPP 112BO/0BW/TSTM,
respectively. Additional development wells drilled by Horizon are Bivins
Ranch No. 2-2, with IPP 116BO/12BW/TSTM, Bivins Ranch No. 3-2, with IPP
122BO/BW16/TSTM (Figure 15), and Bivins Ranch No. 3-3, with IPP281BO/-/TSTM
(Figure 16). A stratigraphic
At the time we purchased ERT Field, its EUR was 1 MMBO. Only one side of the field was delimited, and FSD strongly suggested the most likely size would be 1.5 to 2.5 MMBO. Further, the “one-of-a-kind” OOAK (say “oak” effect) was significant, and that helped bolster confidence in the premium paid. Two 230 MBO wells were equated to about 2.8 BCFE with an approximate PV10 of $1.5MM.
We thought that paying for two locations was a good bid, because we had an 80% confidence level that we would drill two, perhaps three times that many wells. Further, two wells would make the field a 1.5 MMBO field, which fit the FSD.
This reasoning and strategy worked. We bought the field for $1.5MM, based on primary development and recovery. The potential for secondary recovery was discounted to zero. Development of ERT Field after Acquisition
Click here for sequence of Figures 25 and 26.
Location of well in Figure 22.
Click here for sequence of Figures 33 and 34.
Click here for sequence of Figures 37 and 38.
Click here for sequence of Figure 49A and 49B.
Click here for sequence of Figures 17 and Figure 50A and 50B.
Well-Log and Penn Sand Interpretation
At the time ERT Field was acquired, we needed to learn how to interpret accurately the well logs of this low-contrast reservoir and its architecture. The Penn Sand porosity logs, as envisioned by the technical staff, showed laterally persistent thin beds of shale (Figures 18, 19, 20, and 21). Based on a conventional interpretation of open-hole log signature, the engineering reservoir “model” when ERT Field was acquired was a horizontally layered reservoir with tight vertical boundaries. This would tend to create early or premature water break-through, with failure of a waterflood. In other words, the reservoir was considered to be a poor candidate for waterflooding due to the probability of premature water breakthrough and inaccessible compartments and potential for trapped oil that never sees the flood. In order to test this interpretation, well-log data from the existing control together with those from the seven Penn Sand completions drilled in 1995 and cores from two of those wells at opposite ends of the field (Figure 22) needed to be examined in terms of detailed correlation, depositional environment, and diagenesis.
The cored interval in Bivins Ranch No. 4-3 is of the lower 46 feet (of 61 feet) of the Penn Sand and 70 feet of the underlying strata (Figure 23). The Penn Sand is a shaly sand that is ripple-laminated and is thought to be a turbidite (Figures 24, 25, 26, 27, and 28). The underlying unit is also a turbidite deposit that is a sandy shale; it in turn is underlain by a crinoid-rich shale (Figures 24, 29, and 30). The sandstone is brown, oil-stained, with gray, calcite-cemented patches that are irregularly distributed. The predominance of disarticulated crinoid stems in the shale suggests that they have experienced some transport but have not been greatly worn during transport. Crinoids lived in relatively shallow marine waters. However, the overall core and basinal setting suggest the possibility that emplacement of the massive crinoidal clay shale was by sediment gravity flow and that the fossils are “out of place.”
The cored interval in Bivins Ranch No. 6-2 is of the lower 62 feet (of 78 feet) of the Penn Sand and 61 feet of the underlying beds (Figure 31). The Penn Sand, as in Bivins Ranch No. 4-3, is a shaly, ripple-laminated sand that is also interpreted as a turbidite (Figures 32, 33, and 34). The underlying section is a sandy shale that is also a ripple-laminated turbidite (Figures 35 and 36).
Reservoir Properties
The oil-stained sandstone, with significant porosity, shows sharp vertical and horizontal contacts between calcite-cemented (gray) and less well cemented rippled sandstone (brown) (Figures 37 and 38). Representative samples of the oil-stained sandstone have porosity of 12 to 14% and permeability of 3-4 md in both Bivins Ranch No. 4-3 and No. 6-2 (Figures 38, 39, and 40). Clay laminae in the sandstone tend to be microscopic to macroscopic flow barriers; however, they are very unlikely to persist as inter-well vertical flow barriers. Representative porosity in the highly cemented sandstone in Bivins Ranch No. 4-3 (Figure 41) is 3%; permeability in this type of sandstone is 0.05-0.4 md. Cementation boundaries are very sharp and do not follow any bed boundary. The cementation boundary is only three grains wide, with less than one millimeter transition, well below the discrimination of any open-hole porosity tool. The patches of highly cemented sandstone are represented by “bright spots” of high resistivity on the FMI tool (Figures 42 and 43).
Porosity from core analyses shows a distinct bimodal distribution, with almost 25% of the samples showing 3.5-6.5% and approximately 70% showing 11.5-16.5% (Figure 44). The former, of course, corresponds to the highly cemented sandstone, and the latter corresponds to the oil-stained sandstone.
A plot of porosity vs. permeability for Bivins Ranch No. 4-3 and No. 6-2, with 125 data points, also shows the bimodality of the data set, with the relatively tight cluster of pay-zone porosity from 11 to 16% (Figure 45). This cluster (within 5 PU) reflects the dominance of single grain size (very fine) and single sedimentary structure (ripple). Bimodality reflects the sharp boundary (within three sand grains) between porous and cemented sand with porosity occluded. There is little to no gradational porosity – consistent with visual interpretation of cores. Diagenetic occlusion of porosity by calcite cementation has created a petrographic group with higher grain density (Figure 46).
Comparison of porosity from core analyses and from log (density) determination shows that logs are good in resolving porosity for porous and tight intervals thicker than 1.5 feet (Figure 47). However, for intervals less than 1.5 feet thick, log porosity is too high for tight intervals and too low for porous intervals (Figures 48 and 49). This effect is the direct result of thin beds. The tight beds fall below the cut-off of 10%. Some of the porous beds are netted, and if so, the magnitude of porosity is understated.
The original engineering reservoir model, which was based on a conventional interpretation of open-hole log signature, was that the reservoir is a layered system and that it was a poor waterflood candidate due to probability of premature water breakthrough and inaccessible compartments and potential for trapped oil that never sees the flood. However, from integration of core studies with log interpretation, it can be shown that lateral continuity on an inter-well scale should be present. There should be acceptable vertical paths for fluids to flow throughout reservoir and avoid premature water breakthrough if flooded. Conclusions
Update as of August, 2002
Cumulative production through June 1, 2002, was 1.13 MMBO, with hyperbolic decline from 15 Penn Sand producers (Figure 51). Remaining reserves are estimated to be about 0.95 MMBO. EUR of the field is currently estimated to be 2.1 MMBO, or nearly exactly the estimate based on original FSD curves. Original estimate was that the most likely field size would be 1.5 to 2.5 MMBO.
References
Budnik R.T., 1989, Tectonic structures of the Palo Duro Basin, Texas Panhandle: University of Texas Bureau of Economic Geology Report of Investigations (RI) 187, 43 p. Dutton, S.P., 1980, Petroleum source rock potential and thermal maturity, Palo Duro Basin, Texas: University of Texas at Austin, Bureau of Economic Geology Geological Circular 80-10, 48 p. Rose, P. R., 1986, Petroleum geology of the Palo Duro Basin, Texas Panhandle: Office of Nuclear Waste Isolation, Battelle Memorial Institute, BMI/ONWI-589, 45 p.
Acknowledgments
Acknowledgment is made to the following for assistance in this study:
• Questar E&P, Tulsa, Oklahoma • Joe Campbell (regional geology) • Mike Kuykendall (petrography) • Rod Tillman (core interpretation & project supervision) • Garner Wilde (biostratigraphy) • Technical Staff, Tide West Oil Company, Tulsa, Oklahoma. |