Lithofacies Classification: From Sedimentologic Analysis to 3D Reservoir Modelling

Abstract

Abstract Deepwater depositions are complicated processes reflecting the combined effects of global eustatic sea-level variation, regional sediment input, climate, paleo-topography, and many other factors. The complex depositional processes give rise to various depositional facies, which significantly impacts reservoir connectivity and quality. The description of deepwater depositional facies can be performed in details at sedimentologic level based on corematerials, in which complex sedimentation feature such as grain and matrixcomposition, grain-size distribution, color, sorting, roundness, climbingripple, cross bedding, various amount of mud clasts within massive sandstone, degree of bio-alteration, and so on, are considered to classify the facies. Such a detailed facies analysis is necessary because it provides information regarding the geological processes and associated environment that is responsible for the accumulation of the reservoir rocks. However, when dealing with reservoir modeling, it can be too sturdy to include all detailed facies types from sedimentologic description. Main attention should be paid to identify just the major facies types that bear geological environment signature, yet simple enough for any reservoir simulators to handle. This study integrates outcrop data with subsurface data, compares patterns of facies variations from outcrops to several deepwater fields, and suggests that for reservoir modeling purpose most of the deepwater turbidite fields can be described by four major facies types, channel/lobe axis, off-axis, margin, and background. Each of these facies has its range of reservoir properties, and the overall performance of the reservoirs depends on the relative proportions of various facies types and their spatial arrangement. By applying advanced technology, log data are trained to recognize facies types based on the patterns defined from core studies. The detailed log facies types can also be represented by four major groups. As a result, integration of core, log, and outcrop data leads to a robust solution for handling the complex lithofacies issues in 3D geological model, enabling better development of strategy forfield development.