Recent Advances in Modeling of the Atmospheric Boundary Layer for Air Quality Modeling

Air quality models have been shown to have persistent tendencies for high biases in near surface concentrations of ground level emitted chemical species (e.g. NOx, CO, primary PM) and, conversely, low biases in ozone during the evening transition and overnight on many days. Meteorology models often show similar tendencies to over-predict ground level water vapor mixing ratios during evening transition times. Such biases suggest that PBL models become too stable too early in the evening and have insufficient mixing in the stable boundary layer (SBL) during the night. Model errors in either surface heat flux and/or the parameterization of PBL turbulent mixing may be contributing to these biases. Note that erroneously high concentrations are typically mitigated through the use of artificially large minimum values for eddy diffusivities in chemical transport models. In some models, the large values are only used in urban areas following the logic that meteorology models have inadequate treatment of urban heat island effects. Thus, improvements to urban parameterizations, particularly increased surface heat capacity, have reduced premature and excessive stabilization. We have also modified the stomatal response to solar radiation to delay the evening transition to stable conditions. In addition to these modification to the land surface model (LSM), we have modified the Asymmetric Convective Model version 2 (ACM2) that is used in both the Weather Research and Forecasting (WRF) model and the Community Multiscale Air Quality (CMAQ) model. The new version of ACM2 has a new local eddy diffusivity algorithm for stable conditions that reduces high biases in water vapor mixing ratios in WRF and high biases in ground level emitted chemical species in CMAQ. Model evaluations with various grid cell resolutions will be described.