GEOPHYSICAL CHARACTERIZATION OF RESOURCE PLAYS

Abstract

The shale exploration and production in the United States have changed the dynamics of the oil and gas business in the world. The production heterogeneity associated with shale resource plays demands recording different kinds of data during the lifecycle of a shale reservoir. Surface seismic, microseismic, well logs, vertical seismic profile (VSP), and core data are some of the most common data acquired for subsurface characterization. Due to the large number of wells drilled in some of the most prolific basins such as Anadarko and Permian Basins, geoscientist resort to well correlations and statistical analysis to plan optimum well locations and the surface seismic data is considered unsought. Well planning with well log correlation entails high uncertainty due to high inherent rock heterogeneity. In this dissertation, I show how incorporating the surface seismic data with log/core data can decrease the uncertainty of mapping producible rock types, and aid in avoiding perilous drilling location such as those that can cause induced earthquakes.

In this dissertation, I show a methodology that combines core and seismic data to delineate petrophysically defined rock types away from the cored well. In the case study presented in the dissertation, the rock types were delineated over an area of 477 square miles from measurements conducted on one cored well. The rock types were defined using porosity and permeability values and estimated away from the well by combining elastic measurements from seismic and core samples. The seismic elastic properties P-impedance (ZP), S-impedance (ZS), and density (ρ) were estimated by simultaneous prestack inversion. One of the limitations of estimating elastic parameters from prestack data is ZS, and ρ experiencing a decrease in resolution compared to ZP due stretching of the non-zero offset data caused by NMO corrections. Hence, we propose a method to compensate for the NMO stretch that balances the spectrum across the reflector, increasing the resolution of ZS and ρ after prestack inversion.

In the final chapter of the dissertation, I propose methodologies to image seismogenic faults. The strike-slip faulting is the dominant deformation style in Anadarko Basin, and the faults get unrecognized due to their low offset in the sedimentary formation while planning water injection or hydrocarbon production well. These faults might act seismogenic to injection or production activities based on their orientation and regional stress conditions. I propose a new method called band-limited multispectral coherence to image the strike-slip faults in basins with similar deformation style as Anadarko Basin and with a record of active induced seismicity. In Anadarko Basin, the strong visual correlation between recorded earthquakes and the faults delineated by the proposed methodology signals the seismogenic nature of the faults. The faults with no associated induced seismicity, geomechanical modeling is proposed to investigate their reactivation potential.