Print this page
Click
to article in PDF format.
Rapid Identification and Ranking of Reservoir Flow Units, Happy Spraberry Field, Garza County, Texas*
By
John Layman III1 and Wayne Ahr2
Search and Discovery Article #40144 (2005)
Posted March 1, 2005
*Adapted from extended abstract, entitled “Porosity Characterization Utilizing Petrographic Image Analysis: Implications for Rapid Identification and Ranking of Reservoir Flow Units, Happy Spraberry Field, Garza County, Texas,” prepared for presentation at AAPG International Conference & Exhibition, Cancun, Mexico, October 24-27, 2004.
1Amerada Hess Corporation, Houston, TX
2Texas A&M University, College Station, TX ([email protected])
Abstract
Carbonate reservoirs may be heterogeneous and exhibit lateral and vertical variations in porosity and permeability. New technology and an improved understanding of carbonate reservoirs have led to more detailed reservoir description, flow unit delineation, and flow unit ranking. Petrographic image analysis (PIA), a relatively new method, was used to analyze the carbonate porosity of the reservoir interval at Happy field, Garza County, Texas. The reservoir produces from depths of -4900 to -5100 feet and consists of Lower Permian oolitic grainstones and packstones. Associated floatstones, rudstones, and in situ Tubiphytes bindstones are also present in the interval.
Reservoir pore characteristics and their corresponding degrees of connectivity (“quality”) were determined using standard petrography, PIA, core analyses, and mercury injection capillary pressures. The PIA method enables rapid measurements of pore size, shape, frequency of occurrence, and abundance. Common pore characteristics were used to identify stratigraphic and diagenetically similar intervals, within which four pore facies were observed. Pore facies were defined and ranked as to quality by comparing PIA data with measured porosity, permeability, and, in a limited number of samples, median pore throat diameters. Pore facies exhibiting oomoldic and solution-enhanced interparticle porosity ranked best in quality. Rocks with incomplete molds and dispersed interparticle pores ranked second; rocks with mainly separate molds ranked third, and rudstones, floatstones, and bindstones with dispersed separate vugs and matrix porosity ranked fourth. The PIA technique is a viable and fast alternative to standard petrography. It yields data that compares with petrophysical measurements and, when properly used, is a valuable method for reservoir characterization in heterogeneous carbonate pore systems.
|
|
IntroductionThis study tests the applicability of PIA as a tool for identifying reservoir flow units in the carbonate reservoir at Happy Spraberry field. The reservoir interval is interpreted to be lower Leonardian in age and part of the Lower Clear Fork Formation, which is shelf equivalent of the Dean Formation (Handford, 1981; Mazzullo and Reid, 1989). The depositional model for Happy field, interpreted by Hammel (1996) and Roy (1998), is an oolitic grainstone shoal complex with floatstone and rudstone debris aprons around patches of in situ Tubiphytes bindstone buildups. Their interpretation is supported by core descriptions, thin-section examination, and wireline log analyses done in this study.
Study Area and MethodsHappy field is located in south central Garza County, Texas, on the eastern shelf that flanks the Midland Basin (Figure 1). Data used in the study include core from five wells, 52 petrographic thin-sections, capillary pressure measurements, and wireline log data. Cores were described for sedimentary structures, constituent grains, and depositional fabric (Layman II, 2002). Thin-sections were examined by standard petrographic methods for total porosity, per abundance, pore types, grain constituents, and cements. Wireline logs were used to calculate porosity on reservoir intervals where core analyses were absent. Cross plot porosity from CNL-FDC logs, general log signature, as well as pore facies information were used to correlate subdivisions of the reservoir zones across the field. Petrographic image analysis of carbonate pores has been used to predict reservoir quality (Anselmetti, 1998; Ehrlich, 1987; Ehrlich et al., 1991). In this study, pore geometry was measured using PIA, and the resulting measurements were related to reservoir quality. Petrographic image analysis was performed on the thin-section data set using Image Pro Plus, an image acquisition and analysis software program. Images from thin-sections were captured by a Sony DXC-290 digital video camera that relayed the signal to a PC. Ten images per thin-section were viewed at 12.5X magnification and analyzed with the software. Porosity was identified and measured for pore size, shape, frequency, abundance, and pore origin. Pores were auto-classified by the software according to geometry, and measurement data were plotted into histograms. These “porosity fingerprints” have implications as to reservoir quality and petrophysical characteristics. This method allowed for a much faster and cheaper alternative to reservoir quality assessment and flow unit mapping based on pore data obtained from PIA.
Stratigraphy and LithologyThe carbonate interval at Happy field is interpreted as Lower Clear Fork Formation (lower Leonardian). This is time-equivalent to the basinal Dean sandstone (Montgomery, 1998). Primary production is from a grainstone shoal complex with associated lithofacies (Figure 2). The shoal is composed of well sorted, medium-grained oolitic grainstones and packstones; the interval averages about 20 feet in thickness. Lithoclastic rudstones and floatstones containing fragmented and whole mollusks, crinoids, and fenestrate bryozoans are common as fringe deposits around the small skeletal buildups. The buildups are composed mainly of encrusting organisms and Tubiphytes-rich bindstone that grew mainly in the central part of the field between two larger grainstone bodies (Ahr and Hammel, 1999).
ResultsThe types of data obtained from PIA studies include pore size, shape, frequency, and abundance (total porosity). In addition, pores in each sample were classified according to their geological origin. Total porosity from PIA was compared to porosity values obtained from standard petrographic methods, log calculations, and core analyses. The comparisons showed that the accuracy of PIA estimates of porosity are comparable to the other methods. Porosity histograms were constructed from the pore data to assess all pore characteristics rapidly (Figure 3). Samples were then correlated to determine trends and patterns in the pore data that defined pore facies of the reservoir. Pore facies are combinations of pore data that have predictable reservoir potential and petrophysical characteristics. Four pore facies were identified in the reservoir and associated carbonate section at Happy Spraberry field. These pore facies serve as the basis for the quality classification scheme. The highest quality or “best” pore facies occurs in the oolitic/skeletal grainstones that typically exhibit 15-25% porosity and 12-25 millidarcies (md) of permeability. This pore facies consists mainly of moldic and solution-enhanced intergranular pores that were produced by diagenetic leaching of grains and interstitial cement. Intermediate quality pore facies typically exhibits 15-25% porosity and 5-12 md permeability. Rock types consist of moderately cemented skeletal grainstones and packstones in which the dominant pore types are incomplete moldic and solution-enhanced intergranular. The pore facies with the lowest reservoir quality is composed of two subfacies. The higher quality subfacies includes isolated molds in highly cemented oolitic skeletal grainstones where leaching only affected metastable grains. Other scattered grains underwent micritization, stabilization, and neomorphism. As a result, pores are commonly isolated, disconnected, and may be classified as separate vugs (Lucia, 1983). Porosity histograms of this pore facies typically show the influence of large (greater than 10,000 mµ2), separate molds, and less than 20% of any other pore type or size. Porosity averages 10-14%, but it may be as high as 25%. Permeability is typically less than 5 md. Overall the lowest quality pore facies is present in silty, skeletal packstones, siliciclastic siltstones, and rudstones. Typically, porosity is less than 10%, and permeability is less than 10 md. This rock type has abundant quartz silt that is locally more porous and permeable than its surrounding carbonate rocks. Large, blocky vugs are also typical in this pore facies. Table 1 is a summary and ranking of the pore facies and associated pore data of each one.
Happy Spraberry field produces from heterogeneous, shallow-shelf carbonates where lateral and vertical variations in porosity and permeability are common. Porosity is predominantly a diagenetic overprint on depositional texture (grain-moldic in oolitic grainstones). Utilizing PIA as a method for characterizing carbonate reservoirs is a relatively new procedure. Data on pore characteristics is obtained much faster than standard petrographic methods. Image analysis data were interpreted to identify 4 distinctive pore facies which, in turn, are predictors of rock type, petrophysical properties, and production characteristics. Image analysis was proven to be a good substitute for more time-consuming methods for determining porosity and provided results with accuracy comparable to results obtained from core analyses, wireline log calculations, and standard petrographic methods. The highest quality reservoir rocks occur in oolitic grainstones and packstone where large (greater than 10,000 mµ2) moldic pores dominate. Also, the highest combined values of porosity and permeability were associated with the presence of large, solution-enhanced intergranular pores in addition to the moldic pores (oomoldic and skelmoldic). We interpret that storage capacity existed in the moldic pores and that solution-enhanced porosity provided connectivity. Pore size data obtained from petrographic image analysis is a useful predictor of median pore throat size, which would otherwise only be available by performing expensive mercury injection capillary pressure tests. The list to avoid pitfalls of petrographic image analysis includes choosing a magnification that gives appropriate and accurate images of porosity, quality control on preparation of thin-section samples, and consistent sampling of thin-sections.
AcknowledgmentsThis study was part of a Master’s Thesis at Texas A&M University. I would like to thank the members of my committee: Wayne Ahr, Tom Blasingame, and Steve Dorobek. I would also like to thank Bob Berg for substituting at my defense. I would also like to thank those who funded this research: AAPG Grant-in-Aid, Texas A&M University Graduate Fellowship, and the late Mr. Michel T. Halbouty for a generous scholarship.
ReferencesAhr, W.M. and B.S. Hammel, 1999, Identification and mapping of flow units in carbonate reservoirs: An example from the Happy Spraberry (Permian) Field, Garza County, Texas USA: Energy Exploration & Exploitation, v. 17, p. 311-334. Anselmetti, F.S., S. Luthi, and G.P. Eberli, 1998, Quantitative characterization of carbonate pore systems by digital image analysis: AAPG Bulletin, v. 82, p. 1815-1836. Ehrlich, R., S.J. Crabtree, K.O. Horkowitz, and J.P. Horkowitz, 1991, Petrography and reservoir petrophysics I: Objective classification of reservoir porosity: AAPG Bulletin, v. 75, p. 1547-1562. Ehrlich, R., and J.P. Horkowitz, 1987, Estimation of petrophysics from thin-section; petrographic image analysis, in AAPG 1987 Southwest Section meeting abstracts: AAPG Bulletin, p. 238. Hammel, B.S., 1996, High resolution reservoir characterization of the Permian (upper Leonardian) Spraberry Formation, Happy Spraberry Field, Garza County, Texas: Unpublished Master’s Thesis, Texas A&M University, 1996, 157 p. Handford, C.R., 1981, Sedimentology and genetic stratigraphy of Dean and Spraberry formations (Permian), Midland Basin, Texas: AAPG Bulletin, v. 65, p. 1602-1616. Layman II, J.M. 2002, Porosity characterization utilizing petrographic image analysis: Implication for rapid identification and ranking of reservoir flow units, Happy Spraberry field, Garza County, Texas: Unpublished Master’s Thesis, Texas A&M University, 2002, 112 p. Lucia, F.J., 1983, Petrophysical parameters estimated from visual description of carbonate rocks: a field classification of carbonate pore space: Journal of Petroleum Technology, v. 35, p. 626-637. Mazzullo, S.J. and A.M. Reid, 1989, Lower Permian platform and basin depositional systems, northern Midland Basin, TX, in P.D. Crevello, J.L. Wilson, J.F. Sarg, J.F. Read, eds., Controls on carbonate platform and basin development: SEPM Special Publication no. 44, p. 305-320. Montgomery, S.L., and W.H. Dixon, 1998, New depositional model improves outlook for Clear Fork infill drilling: Oil & Gas Journal, v. 96, p. 94-98. Roy, E., 1998, High resolution mapping of flow units for enhanced recovery program planning, Happy Spraberry Lime Field, Garza County, Texas: Unpublished Master’s Thesis, Texas A&M University, 1998, 106 p.
|
