Incorporating Geological Conceptual Models and Interpretations into Reservoir Modeling Using Multiple-Point Geostatistics

Abstract

Geological conceptual models and prior knowledge play an important role in reservoir modeling when geologists and reservoir modelers try to predict the spatial heterogeneity of geological sedimentation between wells based on available sparse data in 3D. This process of incorporating the prior models or interpretations can be implicit, poorly defined as in hand contouring or it can also be explicitly used through an algorithm as in digital mapping. The newly introduced multiple-point geostatistics provides a powerful tool for geologists and reservoir modelers to bring such prior models, both explicitly and quantitatively, into reservoir modeling using training images. A training image is the numerical representation of a prior geological model that contains the patterns believed to exist in realistic reservoirs under study. By reproducing high-order statistics, the multiple-point algorithm can capture curvilinear structures or complex features from the training image and anchor them to specific well locations. This article presents the principle of multiple-point geostatistics and its applications to building 3D reservoir models with a stress on the importance of training image concept.