High-Resolution Models Overestimate California Wintertime Precipitation


Predicting precipitation over California is a critical challenge for the modeling community and a critical need for a state facing uncertainty in current and future water resources. DOE-funded scientist Peter Caldwell of Lawrence Livermore National Lab evaluated the ability of many models with differing spatial resolution to predict California precipitation. Predicted wintertime precipitation was systematically overestimated by higher-resolution models for both regional and global climate models due to over-predicted rainfall intensity during large storms. Increasing the resolution of model topography, clouds, and hydrology was expected to increase precipitation predictability, but the study reveals that this is not currently the case. These results suggest that improvements in high-resolution simulation of west-coast precipitation are needed and that caution should be used when interpreting current-generation regional model output in the complex coastal and mountainous terrain of the western United States.


Caldwell, P. 2010. “California Wintertime Precipitation Bias in Regional and Global Climate Models,” Journal of Applied Meteorology and Climatology, 49:2147–58. doi:10.1175/2010JAMC2388.1.