Integrating geology, rock physics, and seismology for reservoir-quality prediction

This research focuses on the prediction of reservoir quality from seismic and well-log data, integrating concepts from geology and geophysics. The purpose has been to understand the geologic processes that control lateral variations in acoustic impedance and porosity. The work concentrates on the effect of rock texture and fractures on the elastic properties of sedimentary rocks.; This work improves the understanding of the rock-physics depositional and diagenetic trends. The modified Hashin-Shtrikman lower bound can be used to distinguish between sorting and packing effects. It constitutes an upper bound for the sorting effect and a lower bound for the packing effect. Pressure solution is an alternative mechanism to reproduce the rock-physics diagenetic trend for high-porosity quartzose sands, using the Digby-Rutter model proposed here.; The patterns that clastic sedimentary sequences present, in the rock-physics planes, agree with predictions from rock-physics models. Dispersed sand-clay mixtures predominate in fluvial deposits, whereas laminar mixtures predominate in mud-rich deep-water deposits. Scarcity of mixed lithofacies characterizes sand-rich deep-water deposits, whereas abundance of these lithofacies occurs in low-energy shallow marine deposits. The results demonstrate that the elastic properties of clastic mixed lithofacies strongly vary depending on the mixture's proportion and fabric, and rock-physics models can be used to predict these variations.; The second part of this research deals with the use of outcrop information and seismic data to predict fracture distribution in the subsurface. This work documents a fundamental link between fracture hierarchies and sequence stratigraphy. Fracture spacing and dimensions of different fracture hierarchies are constrained by the thickness of the confining stratigraphic interval. It also documents examples of hierarchical shearing and progressive deformation, a concept that explains the evolution of faults and fracture systems. I evaluate different geostatistical techniques to create digital static models of fractured reservoirs using outcrop data. The stochastic-fault-modeling technique creates maps of fracture density using an object-based indicator approach. Finally, this study presents an integrated approach to the prediction of fracture swarms from seismic data, starting from understanding the mechanisms for fracture localization, analyzing the expected seismic response of fracture swarms, and applying these concepts to the interpretation of seismic attributes.