A validation of simplified CFD approach for modeling urban dispersion with Joint Urban 2003 data

We have recently developed and demonstrated the utility of a high-resolution building scale computational fluid dynamics (CFD) model, FEM3MP, for simulating flow and dispersion of chemical/biological agents released in the urban environment (Humphreys, et al., 2003). While high-resolution CFD models are very useful for emergency planning of special events, vulnerability analyses, post-event assessments, and development of mitigation strategies, such models generally require large computer resources and relatively long turnaround times, and are thus unsuitable for emergency response situations. Our simplified CFD approach was developed to meet such needs. With it, only targeted buildings are explicitly treated with fine grid resolution and no-slip boundary conditions, while the remaining buildings are represented as drag elements (or virtual buildings) with much coarser grid resolution. Early test results (Chan, et al., 2004) indicated the approach is potentially very cost-effective.

Here we further evaluate this scheme via a quantitative comparison to data collected in the Joint Urban 2003 field experiment in Oklahoma City, July 2003. We use near- and medium-field data in a validation study in which the accuracy of predicted concentration between simulations using mostly “virtual” and completely “virtual” buildings are assessed and compared. We also compare the computational costs of these simulations.

This work was performed under the auspices of the U.S. Department of Energy by Lawrence Livermore National Laboratory under contract number W-7405-ENG-48.