An integrated study of a Mississippian tripolitic chert reservoir Osage County, Oklahoma, USA Benjamin L. Dowdell*, Atish Roy, and Kurt J. Marfurt, The University of Oklahoma Summary With the advent of horizontal drilling and hydraulic fracturing in the Midcontinent, U.S.A., fields once thought to be exhausted are now experiencing renewed exploitation. However, traditional Midcontinent seismic analysis techniques no longer provide satisfactory reservoir characterization, and new seismic analysis methods are needed to properly characterize these radically innovative play concepts. Seismic attributes such as impedance inversion are sensitive to lithology while coherence and curvature are sensitive to lateral changes in waveform and structure. Our objective is to map tripolitic high porosity chert sweet spots within a highly fractured Mississippian lime reservoir, located in Osage County, Oklahoma, which also contains tight nonporous chert using impedance inversion correlated with surface seismic attributes and well log information in the survey. Introduction Osage County, Oklahoma, U.S.A., has an extensive history of hydrocarbon exploration and production. Figure 1 shows the location of the study area (Elebiju et al., 2011; Walton, 2011). In 1896, the first well was drilled in Osage County with oil discovery occurring a year later (Bass, 1942). Concurrent with the exploitation of Burbank field, economic oil and gas production in Osage County has occurred continuously since the early 1920s (Bass, 1942). Recently, interest has been renewed in the Mississippian Lime play with the advent of horizontal drilling and hydraulic fracturing. Figure 1. Map showing location of Osage County and the major geologic provinces in Oklahoma. Top right inset shows location of oil producing formations within the Mississippian lime. Blue box location of study area (Modified from Elebiju et al., 2011; Walton, 2011). The Mississippian section contains highly fractured limestone and nonporous chert with sweet spots of highly porous tripolitic chert. The play concept is to drill horizontal wells perpendicular to the fractures in the lime and nonporous chert and then provide migration pathways by hydraulically fracturing and/or acidizing the formation in order to produce from multiple tripolitic chert sweet spots. Successful mapping of tripolitic chert sweet spots lowers drilling risks and costs and enhances economic success. Relatively little has been published on the Mississippian lime play. Rogers (1996, 2001) presents a study and model for the diagenesis and deposition of the Mississippian chert in Kay County, Oklahoma, directly to the west of Osage County. Nissen et al., (2006) describes the application of seismic attributes for fracture trend detection to Mississippian carbonate reservoirs in Kansas. Yenugu et al., (2010, 2011) presents a study of Osage County Mississippian chert reservoir property prediction using Gray Level Co-Occurrence Matrix (GLCM) seismic texture analysis and AVO inversion correlation with observed well log properties. Yenugu and Marfurt (2011) present a correlation of seismic curvature with fractures observed on borehole image logs. Elebiju et al., (2011) studies the basement structures of Osage County using seismic amplitude and gravimetric data, concluding that Mississippian chert deposition is controlled by basement deformation trends. Matos et al., (2011) use GLCM seismic texture analysis and Kohonen self-organizing maps (SOM) to map chert facies in an Osage County study area. Geologic Setting A warm, shallow sea with plentiful marine life covered most of Oklahoma, which was close to the equator during the Mississippian, approximately from 359 Ma to 318 Ma (Elebiju et al., 2011; Rogers, 2001). An extensive shelf margin existed, trending east-west along the Oklahoma- Kansas border (Watney et al., 2001). Osage County, Oklahoma is on top of the Cherokee Platform and bounded to the west by the Nemaha Uplift and to the east by the Ozark Uplift (Johnson, 2008). Figure 2 shows a generalized stratigraphic column for Osage County (Elebiju et al., 2011). SEG Las Vegas 2012 Annual Meeting Page 1
(Rogers, 2001). Note the tripolitic chert s characteristic response of high porosity, low permeability, low resistivity and low density. Figure 2. Generalized stratigraphic column for Osage County, Oklahoma (from Elebiju et al., 2011). The zone of interest is the Mississippian chert, which represents an unconformable surface. The Mississippian Limestone underlies the whole of Osage County and is roughly 90 m thick (300 ft) in three-fourths of the county (Bass, 1942; Elebiju et al., 2011). An unconformity between the Mississippian and Pennsylvanian represents a period of uplift and erosion during which Osagean cherty limestone underwent diagenetic alteration (Rogers, 2001). Alteration resulted in the deposition of a tripolitic chert, which exhibits high porosity, low permeability, low resistivity and low density, as well as a tight, dense, nonporous chert (Elebiju et al., 2011; Rogers, 2001). Figure 3 shows a type log for the chert from Rogers (2001) study area in Noble County, Oklahoma, located nearby to the southwest of our study area in Osage County. Occurrence of the chert is widespread, but unlike the lime it is altered from, it is not continuous throughout the Mississippian section (Rogers, 2001). The chert is up to 60 m (200 ft) thick and is an important reservoir rock in north-central Oklahoma (Rogers, 2001). When encountered, sweet spots of the tripolitic, high porosity chert makes for good reservoir rock (Rogers, 2001). Figure 3. A representative log highlighting the response of the Mississippian tripolitic chert (informally called chat ) Methodology We begin by applying time processing and filtering to the 3D prestack seismic survey, including deconvolution and velocity analysis to enhance data quality. Once the data are prestack time migrated and stacked, we perform data conditioning, including acquisition footprint removal, and also to calculate attributes such as coherence and curvature, which are key elements in fracture detection and mapping. We correlate the coherence and curvature volumes with interpreted borehole image logs. An impedance inversion is then run on the conditioned prestack time migrated data for mapping porosity and tripolitic chert sweet spots. We then correlate the impedance inversion results with seismic facies derived from SOM analysis and compare predicted zones of high porosity with predicted zones of tripolitic chert facies. Figure 4 shows a diagram with the planned workflow. Figure 4. Proposed workflow for mapping porosity and fractures in a Mississippian lime reservoir from Osage County, Oklahoma. Analysis Figure 5 shows the log response at Well B, which was used to perform the inversion. While the gamma ray response is similar to that of tripolitic chert, the porosity signature is low through the zone and the density is high. Figure 6 shows a cross plot of P-Impedance against porosity, and the rocks from the zone of interest show a trend of decreasing impedance with increasing porosity. Figure 7 shows the results of the postack acoustic impedance inversion and shows pockets of low impedance encased in zones of high impedance. Figure 8 shows coherence corendered with k1 and k2 principle curvatures. Areas of low coherence indicate areas of potential faulting, karsting, and paleotopographic edges. k1 curvature indicates ridge-like features while k2 curvature indicates valley-like features. When corendered with coherence, areas of deformation can be better identified. 2 SEG Las Vegas 2012 Annual Meeting Page 2
Figure 9 shows acoustic impedance corendered with coherence. We observe that zones of low impedance/high porosity correspond to zones of high coherence. Additionally, areas of low coherence correspond to areas of high impedance and appear to outline low impedance zones. Figure 10 shows acoustic impedance corendered with k1- most positive principle curvature, displayed here in grayscale. We observe that the strongest ridge-like lineaments, which k1 is most sensitive, correspond with zones low impedance/high porosity. Figure 5. Log response at Well B through the Mississippian. Figure 6. Cross-plot of P-impedance vs Porosity. Inset logs show the zone of interest. Note that highlighted magenta points within zone of interest have a trend of decreasing impedance with increasing porosity. Figure 7. Postack Acoustic Impedance Inversion (AI). Time slice through Mississippian interval at t=575 ms. Well location denoted by B. Note pockets of low impedance surrounded by zones of higher impedance. Figure 8. Coherence corendered with k1 and k2 principle curvature. Note that the curvature lineaments outline zones of low energy similarity ratio. 3 SEG Las Vegas 2012 Annual Meeting Page 3
porosity and high density, which is characteristic of the tight limestone and nonporous chert. Figure 9. AI corendered with coherence. Note that areas of low impedance correspond to high coherence. Figure 10. AI corendered with k1curvature. Note that ridelike lineaments correspond with zones of low impedance. After performing a postack acoustic impedance inversion, pockets of low impedance/high porosity encased by zones of high impedance/low porosity are observed on a time slice through the Mississippian (t=575 ms). These pockets of low impedance also have high energy similarity ratio and occur on ridge-like lineaments detected by k1-most positive principle curvature. This behavior agrees with observations made by Elebiju et al. (2011), who noted that zones of low impedance/high porosity, which may correlate to tripolitic chert, fell primarily on ridge-like lineaments. Well B is mostly in a zone of higher impedance, implying it is in a zone of tight lime or nonporous chert. When the log response is compared to Figure 3, the signature shows low These preliminary results from this 3D postack seismic survey are encouraging and suggest that further work with a 3D prestack seismic survey will render even better correlations. The main objective of the Osage County, Oklahoma survey is to improve the understanding of the seismic and petrophysical response of tripolitic chert that forms sweet spots in a larger fractured carbonate and tight chert reservoir. Fractures observed on borehole image logs are correlated with computed coherence and curvature volumes, and then are mapped and compared to impedance inversion results. The Mississippian lime contains highly fractured, tight, dense, nonporous limestone and chert as well as pockets of highly porous tripolitic chert. Mapping sweet spots of tripolitic chert in zones of highly fractured lime and nonporous chert greatly improves drilling success and aids the economic recovery of hydrocarbons. Rogers (2001) model for chert diagenesis and deposition suggests that tectonic uplift events influence the location and thickness of deposited chert. Elebiju et al. (2011) observe a good correlation between basement structure faulting and chert deposition, with low-impedance high-porosity material coinciding with lineaments in the Mississippian, which run parallel to basement structure lineaments. Matos et al., (2011) show success in identifying seismic facies correlating to both the porous tripolitic chert and the tight nonporous chert and lime and calibrating these seismic facies to well logs. Conclusions Based on these observations, we expect to find zones of high porosity (low-impedance) predicted from impedance inversion encased in areas of intense faulting and fracturing observed in coherence and curvature volumes and on horizontal borehole image logs. Additionally, we anticipate a correlation between zones of high porosity and predicted zones of tripolitic chert from seismic facies analysis. Ultimately, we anticipate mapping overlapping zones of high porosity, tripolitic chert seismic facies, and highly fractured nonporous chert and lime, resulting in targeted zones. We anticipate enhanced results with prestack impedance inversion and analysis. Acknowledgements The authors wish to thank the Osage Nation and Spyglass Energy for providing us with data and interest in this project. We also wish to thank the sponsors of the AASPI Consortium, the ConocoPhillips School of Geology & Geophysics, and the University of Oklahoma. 4 SEG Las Vegas 2012 Annual Meeting Page 4
http://dx.doi.org/10.1190/segam2012-1563.1 EDITED REFERENCES Note: This reference list is a copy-edited version of the reference list submitted by the author. Reference lists for the 2012 SEG Technical Program Expanded Abstracts have been copy edited so that references provided with the online metadata for each paper will achieve a high degree of linking to cited sources that appear on the Web. REFERENCES Bass, N. W., 1942, Subsurface geology and oil and gas resources of Osage County, Oklahoma: U. S. Geological Survey Bulletin 900-K, 343 393. Elebiju, O. O., S. Matson, G. R. Keller, and K. J. Marfurt, 2011, Integrated geophysical studies of the basement structures, the Mississippian chert, and the Arbuckle Group of Osage County region, Oklahoma: AAPG Bulletin, 95, 371 393. Johnson, K. S., 2008, Geologic history of Oklahoma: Oklahoma Geological Survey. Matos, M. C., M. Yenugu, S. M. Angelo, and K. J. Marfurt, 2011, Integrated seismic texture segmentation and cluster analysis applied to channel delineation and chert reservoir characterization: Geophysics, 76, no. 5, P11 P21. Nissen, S. E., T. R. Carr, and K. J. Marfurt, 2006, Using new 3-D seismic attributes to identify subtle fracture trends in Mid-Continent Mississippian carbonate reservoirs: Dickman field, Kansas: 13th Annual 3-D Seismic Symposium, RMAG-DGS, Expanded Abstracts. Rogers, S. M., 1996, Depositional and diagenetic history of the Mississippian chat, north-central Oklahoma: M.S. thesis, University of Oklahoma. Rogers, S. M., 2001, Deposition and diagenesis of Mississippian chat reservoirs, north-central Oklahoma: AAPG Bulletin, 85, 115 129. Walton, R., 2011, Horizontal drilling breathes new life into Mississippi Lime oil region: Tulsa World, http://www.tulsaworld.com/business/article.aspx?subjectid=49&articleid=20110924_49_e1_cutlin91 9814, accessed 26 March 2012. Yenugu, M., and K. J. Marfurt, 2011, Relation between seismic curvatures and fractures identified from image logs Application to the Mississippian reservoirs of Oklahoma, USA: 81st Annual International Meeting, SEG, Expanded Abstracts, 995 998. Yenugu, M., K. J. Marfurt, and S. Matson, 2010, Seismic texture analysis for reservoir prediction and characterization: The Leading Edge, 29, 1116 1121. Yenugu, M., K. J. Marfurt, C. Wickstrom, and S. Matson, 2011, Correlation of AVO inversion methods with porosity seen on logs and cores: A case study for Mississippian chert reservoirs of Oklahoma, USA: 81st Annual International Meeting, SEG, Expanded Abstracts, 1938 1942. SEG Las Vegas 2012 Annual Meeting Page 5