Regional climate simulations of cold-season precipitation and snowpack over the US northern Rockies: validation and examination of factors controlling the precipitation distribution

This work examines the performance of a regional climate model in capturing orographic precipitation and snowpack dynamics in the northern US Rockies. The Weather Research and Forecasting (WRF) model is run at a sufficiently fine resolution (4-km horizontal grid spacing), over a sub-continental domain driven by the Climate Forecast System Reanalysis (CFSR), to examine WRF's ability to simulate the observed cold-season precipitation and snowpack dynamics. Observations from Snow Telemetry (SNOTEL, providing precipitation rate and snowpack snow water equivalent (SWE)) and the Parameter-elevation Regressions on Independent Slopes Model (PRISM, providing fine-scale monthly mean values of precipitation and other variables) are used for validation. Three different land surface schemes (Noah, Noah-MP, and CLM) are evaluated. The results show that Noah-MP outperforms the other two schemes regarding precipitation, whereas the CLM performs best in terms of the seasonal evolution of the snowpack SWE. This work focuses on the specific question of controls on precipitation distribution across mountain ranges, which not only capture most precipitation, but also separate watersheds. For this purpose WRF is run with the Noah-MP land surface model over a 30 year period (1980-2009). The results show that most precipitation falls upwind of significant ranges that act as barriers to the flow (Fr<1). For ranges or flow situations where the flow is unblocked (Fr>1), the partitioning depends mostly on precipitation type: stratiform precipitation tends to fall on the upwind side, while convective tends to fall more in the lee .