3D geomechanical models have seen a rapid increase in usage within the oil and gas industry. This is driven by the industry's focus on unconventional reservoirs and the subsequent need to understand both wellbore instability during drilling and the drivers for hydraulic fracturing efficiency. Geomechanical models can assist in addressing these questions. In this paper, we demonstrate how broadband AVA inversion models are employed to characterize the Earth's mechanical properties. We then compare stresses computed from our broadband AVA model with those of a computed model utilizing only well log interpolated values. Finally, we show that the employed property model in geomechanical calculations has a significant impact on wellbore stability calculations and fracture propagation pressures. We argue that the use of elastic properties calculated from AVA inversion results in a 3D geocellular model hold significant advantages over the use of properties inferred from well-log interpolation. In addition, the greater fidelity of the elastic properties estimated from broadband seismic AVA inversion has a positive impact on the final geomechanical model. Geomechanical property population from broadband 3D seismic AVA inversionWhen employing 3D geomechanical models, one of the tasks is to build a model of properties representing the elastic response (both dynamic and static) and strength of rock contained in each cell of a geocellular model. Figure 1 shows a gridded model with x and y dimensions of 150m and a variable cell height with average thickness in reservoir layers of 15m. The majority of the model was populated by acoustic impedance (AI) derived from AVA inversion of recently acquired full azimuth broadband data; these data were depth-converted and resampled to the geomechanical modelling grid. Model cells representing the shallow overburden were filled with acoustic impedance interpolated from well logs, while the deep section depth trends, acoustic impedance was calculated from bulk density and stacking velocity trends. In a similar manner, a gridded model of Vp/Vs ratio is created using pre-stack AVA inversion. Using a series of transformsthat relate the models of acoustic impedance and Vp/Vs ratio to static elastic properties (Young's modulus and Poisson's ratio), as well as strength properties (unconfined compressive strength, friction angle and tensile strength) -the geocellular mechanical property models were generated.The quality of our geomechanical model will thus be directly related to the accuracy of estimation of elastic parameters from the seismic volume. One of the most difficult tasks in any AVA inversion process is to obtain a Vp/Vs vs. AI relationship that is compatible with similar log-derived values. Using as an initial source a seismic volume with a broader frequency spectrum rich in low frequency content, as compared with conventional bandwidth land data Giroldi, 2013, Wallick et al, 2012), produces results with greater fidelity to true reservoir properties. This can be seen in Figure 2, wh...
Interpretation of conventional wide azimuth 3D seismic data over deep reservoirs in Saudi Arabia has reached its limits. Impedance calculated from these data and long used as a predictive tool is low in both resolution and fidelity from the standpoint of reservoir properties. Further seismic reprocessing and recalculation of inversion is considered to yield limited upside.A state-of-the-art, high-density full azimuth seismic dataset was acquired over a producing field in eastern Saudi Arabia to attempt to address these issues. The southern one-third of the total volume was selected with well control to develop, optimize, and quality control processing workflows. This "pilot" volume was processed through anisotropic prestack time migration followed by poststack spatially-adaptive wavelet processing and spectral balancing. Analysis of the new reflection seismic data volume shows an increase in bandwidth with improved signal-to-noise ratio when compared with legacy 3D data.Colored inversion was prepared from the pilot volume and compared with the legacy colored inversion conventional 3D volume. The increase in signal bandwidth significantly improves the relative impedance resolution when compared with the legacy inversion. Use of colored inversion simplifies the process of horizon picking and was used to generate horizon control for use in calculation of absolute impedance.Seismic data-driven relative inversion compares favorably with the well-log-derived low frequency model. Acoustic impedance inversions were computed from the new data and benefited significantly from the increase in low frequency signal. In the final absolute acoustic impedance inversion, a portion of the bandwidth typically reserved for well-generated models was replaced by seismic low frequencies.As the methodology for incorporating seismic low frequency into the inversion process improves, so will the prediction of reservoir properties and geometry. Extraction of geobodies at this time varies among the several inversion results and this variability reflects uncertainty in methodology. Additional work will be required to determine which methodology produces the most accurate predictive capability.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.