Intercomparisons of upper air and surface winds in an urban region

During the Regional Air Pollution Study (RAPS), winds in the St. Louis region were recorded by surface stations (hourly averages) and by multiple upper air releases (on the hour). This study analyzes differences 1) between upper air and surface resultant winds, 2) among upper air winds at four sites, and 3) among surface winds at 25 sites. The extensive data set provides a statistical basis for indicating the representativeness of individual observations, and the results show considerable variation between simultaneously measured winds determined by alternate means. Based on gaussian plume dispersion, the spatial variations in wind direction are translated into expected departures between predictions and observations of pollutant concentrations. Inaccurate specification of winds in air quality simulation models is likely to be a substantial contributor to differences between short-term predictions and observations on an urban scale.