CFD simulation of short-range plume dispersion from a point release in an urban like environment

An accurate simulation of the short-range plume dispersion of a hazardous pollutant in a geometrically complex urban region is a prerequisite in emergency preparedness and to assist regulators for developing effective policies. This study critically examines the real predictive capability of a three-dimensional Computational Fluid Dynamics (CFD) model, Fluidyn-PANACHE, to apply it in emergency contexts of an accidental or deliberate airborne release in urban regions. The model is evaluated with the Mock Urban Setting Test (MUST) field experiment of a continuous point source release in an idealized urban geometry of a regular array of shipping containers in various atmospheric stability varying from neutral to stable, and very stable conditions. The simulations are performed using three combinations (cases 1, 2, & 3) of inflow boundary conditions for wind and turbulence profiles. A detailed analysis with statistical measures shows that the performance of the Fluidyn-PANACHE against MUST experiment with all the three cases of the inflow boundary conditions is well achieved within the acceptable standards for air quality applications. The model with three cases 1, 2, & 3 predicts respectively 52.8%, 59.9%, and 67.9% of the total concentrations within a factor of two and shows an overall under-prediction. The sampling line maximum concentrations are better simulated by the CFD model with case-3 (95% within a factor of two) in comparison to other cases 1 & 2. A comparative statistical analysis is also performed with other evaluation studies in the literature for the averaged and sampling line maximum concentrations. The present evaluation of the Fluidyn-PANACHE strengthen the evidence that it is capable of dealing properly with the dispersion phenomena in geometrically complex urban environments.