Elements for an Integrated Geostatistical Modeling of Heterogeneous Reservoirs

Abstract

The geostatistical approach for modeling heterogeneous reservoirs allows, on one hand, to integrate data of various natures and scales, and on the other hand, to evaluate uncertainties by generating multiple possible scenarios of the reservoir heterogeneity. Building a geostatistical reservoir model must account for the geological depositional environment of the reservoir, so as to represent the major heterogeneities that control fluid flow. The quantitative information from wells, seismic and well tests, etc., must be used for the inference of the model structural parameters. Constraining model realizations to the hydrodynamic data from production allows to further increase the model reliability for production forecasts. This paper presents the elements of an integrated methodology for modeling heterogeneous reservoirs. We first introduce the basic geostatistical models used to describe the heterogeneous reservoirs. This is followed by an outline on the inference of the model structural parameters. Then, we present an inverse approach based on the gradual deformation method for history matching.