| Description |
The result of this dissertation enables routine calculation for removal of vehiclegenerated particulate matter with a mean aerodynamic diameter less than 10 microns (known as PM10) downwind of unpaved roads given the height and density of vegetation downwind of the road. At present, the calculation of PM10 removal given vegetative height and density requires an expensive field study or classifying the vegetation as being one of five very general vegetative types. The estimation of PM10 removal downwind of an unpaved road is important in developing net PM10 emission inventories for use in regional air-quality models. Current methodologies for estimating PM10 removal by downwind roughness elements are based on the results of a small number of field studies measuring removal under limited roughness and atmospheric conditions. To significantly increase the data relating PM10 removal rate to site roughness and atmospheric stability, numerical modeling is employed and an additional field study is performed. The simulations utilize Lagrangian dispersion and Atmospheric Diffusion Equation (ADE) techniques. The new field study site features roughness and meteorological conditions distinct from those previously documented in the existing peer-reviewed literature. The PM10 removal measured in the field studies compared well, within a relative 10% error, to the numerical simulation predictions of PM10 removal for the field study site conditions. The simulation results indicate that PM10 removal is related to roughness and atmospheric stability by: ( 1 - CF) = ( 1 - exp( - 2 . 8H*)) exp( - 2 . 0 T**0■ 64) + exp( - 2 . 8H*), where CF is the captured fraction of PMio within the first 100 m and H * and T* parameterize site roughness and meteorological conditions. Qualitatively, this equation indicates that CF increases with atmospheric stability, canopy height, and canopy density. Simple transport models for mean horizontal advection and vertical turbulence within downwind roughness are developed for use in numerical simulations. The models are applicable for both canopies (an array of horizontally homogeneous roughness elements with infinite fetch) and windbreaks (an array of nonhorizontally homogeneous roughness elements located in rows with finite fetches). These transport models have simple inputs, such as vegetative mean height, leaf area index (for canopies), and optical porosity (for windbreaks). These models are also applicable for all practical atmospheric stabilities, roughness heights, and most roughness densities (canopies with frontal area indexes greater than 0.075 and windbreaks with optical porosities less than 0.9). For sparsely distributed roughness elements, traditional atmospheric surface layer parameterizations are more appropriate. |
