Stochastic Joint Simulation of Fades and Diagenesis: A Case Study on Early Diagenesis of the Madison Formation (Wyoming, USA)

摘要

The aim of this paper is to propose an integrated approach to reproduce both fades and diagenetic trends in a static reservoir model based on an outcrop case study. In Wyoming (USA), the Madison Formation (Mississippian) is a thick (up to 350 m) carbonate series, outcropping in several locations of the Bighorn foreland basin. Within these series, nine sedimentary facies have been identified. Based on their vertical stacking pattern, they are organized in small-scale facies sequences: 1) intertidal to supratidal fades sequence; 2) shallow subtidal to intertidal fades sequence; 3) deep subtidal fades sequence. These fades associations have been integrated in a synthetic depositional model, which corresponds to a carbonate ramp progressively evolving towards the most inner part of a platform. This enables to propose a sequence stratigraphy framework for the studied series, that represents at least six third-order sequences (some of them being locally eroded). The diagenetic study has been focused on the identification of the early diagenetic phases. Results from these analyses show the occurrence of several successive early diagenetic phases (micritization, marine calcite cementation, dolomitization, etc.). For modeling purposes, seven "diagenetic imprints" have been defined, each of them corresponding to a succession of diagenetic phases that can coexist in the same sedimentary facies. Moreover, as each sedimentary fades may be affected by several diagenetic imprints, a quantification of these imprints has been realized. A 3D gridded model designed for geostatistical modeling has been constructed in order to reproduce the fades organization of the three first third-order sequences (that are the best documented). The gridding is then based on the four sequence boundaries which have been recognized on every section. The relationships between sedimentary facies and diagenesis have been used to define lithofacies simulation rules. The simulations are based on the plurigaussian and nested algorithms. Finally, a discussion on the distribution of the potential reservoir heterogeneities is proposed, taking into account the sedimentary characteristics (fades, architectures, etc.) and the diagenetic impact.