Modification of the Yonsei university boundary layer scheme in the WRF model for stable conditions

The stable stratified atmospheric boundary layer continues to pose challenges for Numerical Weather Prediction and Climate models. Existing parameterizations cannot adequately capture the depth of the Planetary Boundary Layer (PBL), low-level jets and nocturnal near surface temperature because of their poor representation of turbulent fluxes, especially in mountainous terrain. In addition, small scale processes such as collisions between katabatic and valley flows which produce intense mixing, strong vertical velocities and a rapid drop in temperature are distributed in space and time and are unable to be captured. These interactions between flows of different scales, in general, can contribute significantly to sub-grid, short-lived, intense turbulence events, spasmodically producing high fluxes over short periods. Such vigorous mixing episodes need to be included in meso-scale models in order to improve their performance, especially in dealing with near surface flows. High-resolution numerical calculations were performed using the Weather Research and Forecasting (WRF) model to test its ability to predict mountain weather. Data from two comprehensive field experiments conducted under the aegis of Mountain Terrain Atmospheric Modelling and Observations (MATERHORN) Program (www.nd.edu/~dynamics/materhorn) were used in this study. Different PBL options available in the WRF model were tested and evaluated for stable conditions. The Yonsei University (YSU) PBL scheme was modified and implemented in WRF. The performance of the modified and original YSU scheme, were compared. Preliminary results show that the modified version is more capable of predicting the velocity of the observed low-level jet, which was consistently overpredicted by the original version. Further data processing and analysis is under way. The objective is to develop better parameterizations for turbulent fluxes in the complex terrain leading to improved predictability of local scale air flow. This work is a contribution in this direction and numerous applications follow from this research including air quality modelling and aviation.