Quantitative Prediction of Three-Dimensional Facies Architecture and Heterogeneity in Meandering Fluvial Successions
Fluvial meander bends undertake expansion, translation, rotation and combinations thereof as they evolve. However, relationships between the migratory behavior of a river, the geometry of accumulated sedimentary bodies (e.g., point bars, counter-point bars) that arise from channel migration, and the resultant internal lithofacies distribution within these bodies remain relatively poorly understood. To explore the relationship between fluvial channel evolution and resultant accumulated stratigraphic architecture, a forward numerical stratigraphic model – the Point-Bar Sedimentary Architecture Numerical Deduction (PB-SAND) – has been developed that uses a combined geometric-stochastic approach. The model is applied to predict types of lithological heterogeneity and sandbody connectivity in fluvial successions for a variety of meandering river types.
The modeling approach is constrained by quantified sedimentological data from real-world case-study examples stored in a relational database, the Fluvial Architecture Knowledge Transfer System (FAKTS). The model has the following capabilities: 1) to replicate bar-growth trajectories and sedimentary structures of meandering systems based on real-world data of sedimentary architecture derived from modern rivers and ancient successions that serve as geologic analogs; 2) to examine the sensitivity of intrinsic system behavior to different allogenic controls operating at varying spatial and temporal scales, such as point-bar elements in humid coastal plain vs. dryland fluvial fan settings; 3) to quantify the heterogeneity and compartmentalization arising from intra-bar mud drapes; 4) to predict the sedimentary architecture of meander belts arising from repeated migration and avulsion of river reaches; 5) to predict fluvial sandbody stacking patterns, for example in response to coeval rift basin development.
The grid-free, 3D model provides linkage between local outcrop measurements and large-scale evolutionary behavior, and allows quantitative assessments of possible scenarios depicted in traditional qualitative facies models. Output from PB-SAND can be employed to condition reservoir models at different spatial scales, notably by creating training images for constraining models built through techniques based on Multi-Point Statistics. More realistic architectural geometries and spatial distributions of facies associations markedly enhance conventional reservoir models, thereby improving fluid-flow simulations.
AAPG Datapages/Search and Discovery Article #90323 ©2018 AAPG Annual Convention and Exhibition, Salt Lake City, Utah, May 20-23, 2018