Evaluation of PBL models compared to GABLS experiments and testing in meteorology and air quality models

The GEWEX Atmospheric Boundary Layer Study (GABLS) has, thus far, involved three intercomparison experiments where large eddy simulations (LES) and single column models (SCM) have been compared to either idealized (GABLS1) or field experiment (GABLS2 and GABLS3) data. The Asymmetric Convective Model version 2 (ACM2), which is a first-order combined local and non-local closure model, has been one of many SCM that participated in the GABLS2 and GABLS3 intercomparisons. Hitherto, evaluation of the ACM2 has focused mostly on the convective boundary layer (CBL) where non-local transport by buoyant convective plumes is a key element. However, realistic representation of stable boundary layer (SBL) conditions, where the depth and intensity of turbulent mixing are much reduced, is also essential for accurate meteorology and air quality simulations. Modeling the magnitude, height and direction of nocturnal low level jets is particularly vital for accurate transport of pollutants over long distances at night. Thus, this study is focused on the evaluation and improvement of SBL modeling in the air quality modeling system that includes the Weather Research and Forecast (WRF) model and the Community Multiscale Air Quality (CMAQ) model. First, the currently implemented PBL model in both WRF and CMAQ (ACM2) is evaluated as part of the GABLS2 and GABLS3 imtercomparison experiments. Second, modified SBL algorithms are tested and evaluated through comparison to the GABLS1 experiment, which was an idealized weakly stable case in the arctic defined by an LES. Third, the modified SBL treatment is further evaluated for the GABLS3 experiment, which included a strong low level jet that was observed at the Cabauw meteorological tower, and compared to the current ACM2 implementation and the other SCM that participated in the GABLS3 intercomparison. Finally, the ACM2 with the modified SBL treatment is applied in the WRF and CMAQ models and evaluated over a regional modeling domain in the eastern US through comparison to radar wind profilers. These regional air quality model simulations are evaluated in terms of nocturnal long range transport.10-->