Potential for Downscaling Precipitation Forecasts Using Orographic Precipitation Gradients in the Western United States

Precipitation forecasting over complex terrain remains challenging despite advances in numerical weather model resolution. For example, current operational ensemble modeling systems in the United States [e.g., the Short Range Ensemble Forecast system (SREF) and Global Ensemble Forecast System (GEFS)] lack sufficient resolution to adequately resolve terrain effects on precipitation over the contiguous western United States. Historical techniques for downscaling such forecast system output across sub-grid scale topography involves using day-of-year or monthly climatological relationships between precipitation and elevation. However, these relationships perform poorly when the precipitation and elevation relationship differs from climatology. Variations in precipitation-elevation relationships have not been widely studied for large regions or extended time periods, especially over the western U.S. interior.

One approach to quantifying the precipitation-elevation relationship is to calculate the orographic precipitation gradient (OPG), usually in units of millimeters per meter. Here we develop a quality controlled daily OPG data set for the contiguous western United States based on a linear regression framework of gauge precipitation observations over a 39-year period. This data set is then used to identify spatial and temporal patterns of winter-season (DJF) OPG variability across the western United States on multiple timescales. We then explore drivers of this variability on seasonal OPG patterns using empirical orthogonal function (EOF) analysis and daily OPG variability locally using atmospheric variables as predictors in an effort to develop an improved method for downscaling precipitation forecasts in complex terrain of the western U.S.