Many transport and dispersion (T&D) models need meteorological data and source characteristics to forecast concentration and dosage fields. To do this the models use observational data or mesoscale-model-generated forecast winds as the prevailing winds for a given scenario. This research examines how errors in these input wind fields translate into T&D model solution errors. In particular, this study focuses on street-level plume errors that occur in building aware T&D models for a set of scenario's where the release location varies relative to the building locations. First a “truth” plume was created for a given release location and wind direction. Then the T&D model errors associated with input wind errors were determined by comparing plumes calculated using wind directions varied at 2 degree increments to the truth plume. The errors for a given level dosage are quantified as fraction of overlap, figure of merit in space, measure of effectiveness, and normalized absolute difference. These errors were evaluated for a non urban, an urban grid, and urban radial domain. Two case studies modeled after common city designs were also evaluated. Results show that the relative impact of input errors vary significantly with the release location and the wind direction relative to buildings.