In the event of chemical or biological (CB) agent attacks or accidents, first-responders need hazard prediction data to launch effective emergency response action. Accurate and timely knowledge of the wind fields in urban areas is critically important to identify and project the extent of CB agent dispersion to determine the hazard-zone. In their 2008 report (GAO-08-180), U.S. Government Accountability Office has reported that first responders are limited in their ability to detect and model hazardous releases in urban environments. The current set of modeling tools for contaminant dispersion in urban environments rely on empirical assumptions with diagnostic equations (Wang et al. 2003, Williams et al. 2004). The main advantage of these models is their relatively fast turn-around times, although their predictive capabilities can be limited. As part of the Joint Effects Model (JEM), funded by the Department of Defense, urban transport and dispersion models have been evaluated for their rapid-response capabilities. As discussed in Heagy et al. (2007), majority of the urban transport and dispersion models considered in the evaluation study fell short of satisfying the JEM key performance parameter of maximum 10-minutes run-time on a desktop computer, and the models that were able to satisfy the performance parameter were employed at low resolutions.
Senocak, Inanc; Thibault, Julien C.; and Caylor, Matthew. (2009). "Rapid-Response Urban CFD Simulations Using a GPU Computing Paradigm on Desktop Supercomputers". Eighth Symposium on the Urban Environment, .