Erosion in Surface-based Modeling Using Tank Experiment as Analog

Characterizing deepwater reservoirs, their stratigraphy and structure can be highly uncertain due to the extremely limited data. The existence of subseismic scale fine-grained layers can significantly affect oil production of a reservoir, which requires explicitly modelling the uncertainty of these fine-grained fluid barriers. Conventional modeling techniques, such as twopoints statistics, multipoint statistics, object-based modeling, and process-based model, are incapable of providing a satisfying solution due to their inherent shortages and the lack of data. This work focuses on surface-based modeling of the deepwater system. The contribution of this work is the use of a tank experiment of delta basin to extract relevant statistics of erosion that can be used as analogs for modeling deepwater fans. The advantage of tank experiment data is the availability of the intermediate dynamic topographies of a depositional process, which is a rich source of understanding not only the deposition geometries but also the processes. Proper statistics can be inferred from these dynamic data, and will be applied in the surface-based framework to construct geologically realistic models. The benefits from this research will be a workflow of applying tank experiment data as a data source of modeling deepwater system. A highly geologically realistic model is constructed with information inferred from the data and geological knowledge.