EVALUATION OF A FAST-RESPONSE URBAN WIND MODEL - COMPARISON TO SINGLE-BUILDING WIND TUNNEL DATA

Prediction of the 3-dimensional flow field around buildings and other obstacles is important for a number of applications, including urban air quality studies, the tracking of plumes from accidental releases of toxic air contaminants, indoor/outdoor air pollution problems, and thermal comfort assessments. Various types of computational fluid dynamics (CFD) models have been used for determining the flow fields around buildings (e.g., Reisner et al., 1998; Eichhorn et al., 1988). Comparisons to measurements show that these models work reasonably well for the most part (e.g., Ehrhard et al., 2 ; Johnson and Hunter, 1998; Murakami, 1997). However, CFD models are computationally intensive and for some applications turn-around time is of the essence. For example, planning and assessment studies in which hundreds of cases must be analyzed or emergency response scenarios in which plume transport must be computed quickly. Several fast-response dispersion models of varying levels of fidelity have been developed to explicitly account for the effects of a single building or groups of buildings (e.g., UDM - Hall et al. (2000), NRC-Ramsdell and Fosmire (1995), CBP-3 - Yamartino and Wiegand (1986), APRAC - Daerdt et al. (1973)). Although a few of these models include the Hotchkiss and Harlow (1973) analytical solution for potential flow in a notch to describe the velocity field within an urban canyon, in general, these models do not explicitly compute the velocity field around groups of buildings. The EPA PRIME model (Schulman et al., 2000) has been empirically derived to provide streamlines around a single isolated building