Process-based modelling of turbidity-current hydrodynamics and sedimentation

Abstract

The production potential of deep-water reservoirs is primarily determined by rock bulk volume, porosity and permeability. Quantification of the geometry and spatial distribution of reservoir sands in deep-water deposits can provide crucial information to assess sand body volume, connectivity and the distribution of permeability baffles. This study aims to investigate the influence of turbidity-current process, sediment composition and basin-floor relief, on the geometry and spatial distribution of reservoir sands in turbidite fans. For this purpose, a process-based model has been developed which simulates turbidity-current flow, erosion, and deposition based on principles of fluid dynamics that can deal with arbitrary basin-floor topography and accommodates various grain sizes. It employs the depth-averaged shallow-water approximation in combination with the Boussinesq approximation for density-driven flow in three dimensions. Sediment transport is modelled by an advection-diffusion type equation, and exchange with the bed is largely based on existing empirical models for sediment entrainment and deposition. The model is solved numerically on a rectangular grid representing topography by means of a second-order finite-difference approximation, and employs a shock-capturing technique to accurately model the discontinuous flow front characteristic of density-driven flows. Results are presented of laboratory-scale model validation tests, in which modelling results are quantitatively and qualitatively compared to experimental data. Laboratory experiments involve small-scale flows interacting with complex topographic features as well as multiple successive flows over the same erodible bed. Results indicate that the model is capable of simulating turbidity-current hydrodynamics and sedimentation with an acceptable degree of accuracy under a wide range of conditions.