The buildup of surface pollutants has been shown to be a controlling factor in the quality of storm water runoff. In industrial areas particularly, atmospheric fallout is an important component of surface pollutant loadings. Storm water runoff models presently in use do not consider the physics of atmospheric dustfall.Industries, vehicle exhausts, and blowing of wind over unprotected surfaces all introduce pollutants to the atmosphere. Redistribution of this material on the ground depends on local topography and prevailing meteorological conditions. The location of the industrial areas; the direction, velocity, and duration of wind; total precipitation; and source concentrations are important parameters in the prediction of atmospheric dustfall. The paper describes the physical processes of atmospheric fallout that are relevant to water quality modelling. A new model, called ATMDST, to predict dustfall on individual subcatchments in a metropolitan area using prevailing meteorological conditions is developed based on statistical methods. Results from average, one-variable and two-variable linear regression models were statistically compared with observed data. Finally, ATMDST is interfaced with the storm water management model version 3 (SWMM3) to compute runoff water quality. The model is applied to Hamilton, Ontario. Key words: atmospheric dustfall, air pollution, urban runoff, water quality, pollutant buildup, environmental modelling.