Because large structures have significant mechanical effects upon the flow, it is necessary to simulate these effects when predicting flow and/or dispersion within urban centers. In order to accomplish this we have implemented urban parameterizations in ADAPT, a diagnostic windfield model, COAMPS, a mesoscale weather prediction model, and LODI, a Lagrangian particle dispersion model. The first step in using these parameterizations is to characterize the urban regions by the fraction that is covered by buildings, the average building heights, etc.

In order to be able to rapidly accomplish this for any site within the United States, we have coupled our models to USGS land use data bases. We currently have the USGS Land Use, Land Cover data base which has seven urban categories and the USGS NLCD data base which has three urban categories. We have developed mapping of these classes to the parameters necessary for use in our models.

The Department of Energy, through the Chemical and Biological National Security Program (CBNP), has sponsored major field campaign, URBAN 2000, that was conducted in Salt Lake City in October 2000. This was intended to provide data to evaluate the CBNP models. We are using the URBAN data to guide our choice of mappings from urban land use categories to the model parameterization variables. We will describe the mapping developed and report on the sensitivity of the models to these mappings.