Mapping Lithofacies in a Siliciclastic and a Carbonate Reservoir Using Statistical Rock Physics

In order to characterize two highly heterogeneous reservoirs (one siliciclastic and other carbonate), this paper shows how through the integration of well-log data and 3D pre-stack seismic data, using a statistical rock physics approach, it was possible to extrapolate the lateral continuity of the target lithofacies present in studied reservoirs. The implemented methodology consists of three phases. In the first phase, seismic lithofacies were defined and Vp, Vs and density logs were used to compute elastic attribute logs. This information was used to generate cross-plots of elastic attributes colored by lithofacies to understand the relationship between seismic attributes and lithofacies. In the second phase, 3D pre-stack seismic data and well-log data was used to estimate volumes of AI and EI(20) using a post-stack inversion algorithm. In the last phase, by means of Bayesian analysis, probability maps of the target lithofacies were made with the purpose of identifying the zones with the best petrophysical properties in the reservoirs. The resulting probability maps of lithofacies characterize the heterogeneity of the reservoir and help to reduce the risk associated with the location of future development wells in the field.