Extended Abstract (712K)13.3
Comparison between LES and RANS computations for the study of contaminant dispersion in the MUST field experiment
Anne Dejoan, CIEMAT Centro de Investigaciones Energéticas, Medioambientales y Tecnológicas, Madrid, Spain; and J. L. Santiago, A. Pinelli, and A. Martilli
The MUST (Mock Urban Setting Test) is an experimental study set up in the great basin desert (USA) to investigate the dispersion of a passive contaminant over an urban environment represented by a almost regular array of rectangular containers (Biltoft, 2001). The present study simulates the turbulent flow over the MUST configuration by making use of two different approaches, the Large Eddy Simulation (LES) and the Reynolds Averaged Navier-Stokes equations (RANS). The first approach (which is based on a filtering operation applied to the Navier-Stokes equations) consists of resolving explicitly the dynamics of the unsteady large eddies of turbulence while the small scale motions are modelled. In the second approach, the whole spectrum of turbulence is modelled. LES is in general more accurate than RANS but has the drawback to require large computational times. The purpose of this study is to compare the results obtained from LES and RANS in the aim to assess to what extent each approach can be used to investigate the dispersion of a tracer contaminant in an urban environment. In the present LES the standard Smagorinsky model is used to represent the subgrid scale stress tensor and the subgrid scalar stress while in the RANS approach the k-epsilon model is used. The comparative analysis will be based on mean and turbulence fields (velocity profiles, kinetic energy of turbulence) and on contaminant concentration, quantities for which the available experimental data will be used as reference for the assessment. Also the flow structure in terms of mean size of the recirculation zones forming around the obstacles will be considered. Finally, spatial averaged properties over horizontal slabs will be extracted from the LES and RANS results to get informations on urban parameterizations used in mesoscale models.
Biltoft, C. A. ``Abbreviated Test Plan for Customer Test: Mock Uraban Setting Test (MUST)'', DPG Document WDTC-TP-01-028, Defense Threat Reduction Agency, 2001->
Session 13, Physical and Fine-Scale Modeling II
Thursday, 13 September 2007, 3:30 PM-4:30 PM, Boardroom
Previous paper Next paper
Recorded presentation