Computational Modeling of Near-Source Deposition ofFugitive Dust on Vegetative Surfaces

摘要

Near-source deposition of vehicle-generated fugitive dust on vegetation, fences, buildingsrnand other vertical surfaces has been proposed as a mechanism that can explain thernobserved discrepancy between the contribution of suspended geological materialrnpredicted by emission inventories and the composition measured on receptor filters. Tornmodel the near-source (3-100 m downwind) deposition of fugitive dust, a quasi-twodimensionalrncomputational model was developed that includes the effects of atmosphericrnstability and surface roughness. The model is based on an Eulerian atmospheric diffusionrnequation (ADE) and includes advection, turbulent diffusion, and particle depositionrnvelocity on both the horizontal surface (i.e. the ground) and on roughness elements suchrnas buildings and vegetation. Model parameters for both wind profile and deposition werernset a priori using typical field data and literature values. Depending on site conditions,rnthe model predictions for PM10 range particles indicated large variations in the fraction ofrndust that was transported. The model was used to analyze the particle deposition datarnfrom field experiments that involved 1) an unstable atmosphere at a site with sparse lowrnvegetation 2) a stable atmosphere at a site with an array of 2.5 m high shipping containersrnsimulating buildings.rnThe model showed that atmospheric stability and surface roughness are importantrnvariables affecting the near-source transport and deposition of vehicle-generated dust.rnThe greatest uncertainty in the model is the parameter for the particle deposition ratern(mass deposited / particle concentration time). Further experiments and theoreticalrnanalysis are needed to develop a submodel to predict this parameter as a function ofrngroundcover, vegetative surface geometry, and turbulent wind field properties.