An approach for incorporating sub-grid variability information into air quality modeling

Air quality grid models simulate only a portion of the spatial and temporal variabilities in the concentration fields. The variability at sub-grid (SGV) scales while always present, it is not explicitly represented by grid models. Thus, there will always be differences between what is simulated and what is seen. In this paper, we suggest that explicit representation of the SGV has merit for its potential use in air quality regulatory applications as a weight of evidence factor in modeled attainment demonstrations, for model evaluation studies, and for its utility in air toxics exposure assessments. For weight of evidence analyses, the SGV distributions can provide qualifying concentration bounds for the model grid simulations. Similarly, for model evaluation, knowing the magnitude and characteristics of the SGV provides a basis for constructing a valid statistical design for comparing what is simulated with what is observed. For air toxic exposure assessments, the introduction of sub-grid variability to the gridded concentration fields provides a more complete description of the total variability (distribution of possible outcomes).

This paper discusses descriptions (parameterizations) of SGV and an operational methodology for incorporating SGV description with the gridded fields. We identify two major contributors to SGV in CMAQ. They include primary source contributions from within and from neighboring grids that have not achieved fully dispersed status (but is nonetheless uniformly dispersed in grid models such as CMAQ, and heterogeneities arising from within-cell photochemistry and turbulent interactions. Local scale dispersion modeling can provide information of SGV contributions from the primary emissions and fine scale grid modeling can provide the means to describe the SGV from heterogeneous photochemistry. We introduce and discuss a formulation for incorporating SGV descriptor information to the CMAQ gridded fields. For this effort, CMAQ was run at multi-scales, (36, 12-, 4, and 1-km grid sizes) for July 2001 centered over Delaware. Our initial formulation utilizes the coefficient of variance (COV) as the sub-grid descriptors for 12 and 4-km cells based upon results of local-scale modeling for the primary species and 1-km CMAQ simulations for the secondary species.

Disclaimer: The research presented here was performed under the Memorandum of Understanding between the U.S. Environmental Protection Agency (EPA) and the U.S. Department of Commerce's National Oceanic and Atmospheric Administration (NOAA) and under agreement number DW13921548. Although it has been reviewed by EPA and NOAA and approved for publication, it does not necessarily reflect their policies or views.