Paper Title
CFD Simulation of Dense Gas Dispersion Over Rough Terrain: Assessment of the SST Turbulence Model

Abstract
Computational Fluid Dynamics (CFD) is currently an important tool to predict gas dispersion for risk analysis purposes. However, the accuracy of CFD simulations can be compromised when inappropriate settings are used. Special attention to turbulence modeling and grid refinement need to be taken when simulating gas dispersion over rough terrains. While the standard k-ε turbulence model is the most widely used, the SST model is sometimes selected in order to improve prediction of the wall region. In the present work the standard k-ε and SST turbulence models are assessed and compared against experimental data of dense gas dispersion over a rough terrain. The first near-wall grid node placement has been investigated, since low Reynolds number turbulence models require small y+ values and rough terrains require the wall-adjacent cell centroid to be greater than the sand-grain roughness of the terrain. Computations have been carried out using CFX 18.0 and experimental data from Prairie Grass trials have been used for validation. Computational results showed good agreement with experimental measurements for the standard k-ε model, while it was found that the SST model is not able to provide reliable predictions for gas dispersion over rough terrain because it underpredicts turbulent viscosity regardless of the near-wall grid refinement. Keywords - Computational Fluid Dynamics (CFD), Dense Gas Dispersion, Rough Terrain, SST Turbulence Model, k-ε Turbulence Model