In this study, eleven mathematical models estimating the minimum possible atmospheric dilution of rooftop exhaust gases around buildings are evaluated and reviewed using data obtained from 72 full-scale tracer tests at two buildings. Quantities measured were the stack height, stack-tointake distance, stack diameter, wind speed and direction, exhaust flux, and the stack-to-intake dilution. These models are all based on wind-tunnel simulations and are designed to account for a worst-case scenario so that they can be used to provide a conservative safeguard for design of new stacks or for modification purposes. Although none of the models gave outstanding performance relative to the others, Halitsky's model, derived from the prism tests and Clinical Center data incorporating the exhaust momentum ratio, gave particularly good predictions. Further, all models led to very conservative estimates of the dilution in the case of gases exhausted close to a receptor, suggesting that the models reviewed here should not be used in such cases. IMPLICATIONSThis study examines the selected mathematical models which determine the dilution ratio of a contaminant concentration for an exhaust stack discharge compared to the contaminant concentration at the building intakes. The literature contains a wide range of semi-empirical models which have been developed based on wind tunnel studies. The models are worst-case scenarios and can be used as safeguards for the design modification of ventilation facilities. The present research is of significance because it represents a detailed field measurement program at two full-scale laboratory buildings to determine the dilution of exhaust gases between the stack and the intake. The results provide guidelines for policy makers and designers of industrial and laboratory facilities to use in selecting realistic models to determine appropriate separation distances from the stack discharge and intake for the building ventilation systems. INTRODUCTIONThe effect of buildings on pollutant dispersion is often an important aspect of air quality impact analyses. A plume that is entrained in the flow region around a building can result in concentration fields that are considerably different from those measured in the absence of the building, and which may significantly affect building occupants and equipment operation by contaminating the fresh air intakes.The complexity of these fields is so great that the use of conventional diffusion formulae for their quantitative description often produces misleading results. Such formulae contain the implicit assumptions that the flow field has straight parallel streamlines, modest velocity gradients, and uniform turbulence distributions. Although this is typical of flow over uniformly rough ground, flow around buildings is characterized by curved streamlines, sharp velocity discontinuities, and highly non-homogeneous turbulence. All of these factors disperse effluents in a complicated manner uniquely related to source location and building configuration.A...