Person oriented exposure modeling for ambient air toxics: High resolution modeling for local scale air quality
S.S. Isukapalli, P. George, S. Stamatelos, S.W. Wang, P.G. Georgopoulos
Environmental and Occupational Health Sciences Institute, Piscataway, NJ
Person-oriented modeling for assessing individual and population exposures to air toxics requires ambient concentration estimates at high resolutions. However, due to theoretical constraints, the resolution of grid models is often limited to 2 km. Further limitations arise due to the complexity in developing meteorology and emissions inputs at high resolutions, thus resulting in typical simulations at 4 km resolution or coarser. Several approaches have been used in the past for incorporating photochemical grid model outputs into exposure modeling at the resolution of a census tract and a census block. These include statistical techniques such as the Spatio-Temporal Random Field (STRF) and the Bayesian Maximum Entropy (BME), and hybrid methods that use local atmospheric dispersion models in conjunction with outputs from grid models. For even finer scales, street canyon models or detailed Computational Fluid Dynamics (CFD) simulations can be used with outputs from grid models. This study presents a critical evaluation of techniques for estimating concentrations of air toxics at different scales using available estimates at the next coarser resolution. Evaluation criteria include: (a) performance as a surrogate for obtaining finer scale model outputs from coarser scale model outputs, evaluated through model outputs at different resolutions; and (b) performance w.r.t. estimating point concentrations at specific point locations, evaluated using available measurements. Case studies are presented using CMAQ simulations in conjunction with statistical techniques, local scale models, and CFD simulations. The corresponding estimates of population exposures and study-specific, individual exposures are comparatively assessed.
This work is funded in part by the USEPA (Cooperative Agreement CR-83162501) and the American Chemistry Council (ACC). This work has not been reviewed by and does not represent the opinions of the funding agencies.