P54 Evaluation of weather research and forecasting model's physical parameterizations for regional climate studies over Southern Ontario

Kamal, Mostofa, University of Waterloo, Candad, John C. Lin, University of Utah

In this study, the Weather Research and Forecasting (WRF) Model is used as a nested regional climate model to investigate the sensitivity of total precipitation, daily maximum, minimum, and mean temperature on physical parameterization schemes. The main objective is to identify the optimal choice of physical parameterizations for WRF over Southern Ontario region. Two sets of twelve off-line configurations were simulated for the month of January and July 2002, and compared against DAYMET gridded observations. Simulated precipitation shows overestimation in July but underestimation in January, and more sensitive to the selection of convective and PBL parameterizations. Moreover, WRF performs poorly while simulating extreme precipitation. In case of surface air temperature, model captured spatial heterogeneity well; however, the magnitude is systematically underestimated. The underestimation is higher in January and lower in July. WRF underestimates the daily minimum temperature while overestimates maximum temperature. Temperature is most sensitive to PBL schemes followed by Microphysics. Taylor diagram analysis shows that WRF precipitation and temperature are in good agreement with the DAYMET observations with correlation coefficient 0.6~0.8 and >0.9, respectively. These results indicate that optimum selection of parameterization combinations are thus indispensable before configuring WRF for a regional climate change assessment.