Resolving fine scale in air toxics modeling and the importance of its sub-grid variability for exposure estimates

Atmospheric processes and the associated transport and dispersion of atmospheric pollutants are known to be highly variable in time and space. Typical air-quality models that characterize atmospheric chemistry effects, e.g. the Community Multiscale Air Quality (CMAQ), provide volume-average concentration values for each grid cell in the modeling domain. There are many processes that result in sub-grid variability, such as variability in time and space of the transport and diffusion conditions, or the distribution of emission sources, or the efficiency of chemical transformation. If CMAQ were to be used to assess air toxic exposures, the question arises of the importance of sub-grid concentration variations on exposure assessment results. Exposure models typically estimate inhalation exposure for various demographic groups as they reside and move among different geographic locations. As exposure assessments are typically performed on a census tract level, using ambient predictions calculated on 12-km or 4-km regularly spaced grids. In this study, we used the Hazardous Air Pollutant Exposure Model (HAPEM) to estimate inhalation exposures using ambient annual average concentrations predicted by CMAQ on both 12-km and 4-km grids to investigate how within-grid variability due to distribution of emission would affect inhalation exposure estimates. An urban plume dispersion model was used to estimate the variability of annual average benzene concentration values within CMAQ grid cells for a modeling domain centered on Houston, TX. Then, a series of HAPEM model simulations using the estimated within-grid variability in ambient concentrations provided an assessment of the sensitivity of benzene exposure estimates to sub-grid annual average concentration variability. The large increases in maximum exposure impacts seen in the exposure estimates in comparison to exposure estimates generating using grid-average concentration values are preliminary in that only one source of sub-grid variability was treated, namely the discrete location and distribution of emissions, but they do suggest the importance and value of developing improved characterizations of sub-grid concentration variability for use in air toxic exposure assessments.