Evaluation of HRRR Wind Speed Forecast and WindNinja Downscaling Accuracy during Santa Ana Wind Events in Southern California

Abstract Santa Ana winds are dry offshore downslope windstorms, commonly observed during autumn and winter across southwestern California in the United States. The accuracy of real-time Santa Ana wind forecasts is crucial for wildfire-related emergency management and decision-making. This research utilizes the U.S. Department of Agriculture Forest Service’s diagnostic wind flow model WindNinja (WN) and evaluates 1) the accuracy of 10-m sustained wind speed from an operational coarse resolution mesoscale forecast model and 2) the relative accuracy of high spatial resolution WN downscaled simulations during six Santa Ana wind events. NOAA High-Resolution Rapid Refresh (HRRR) 6-h forecasts with 3-km grid spacing were used as inputs to WN to downscale sustained wind speeds to 500-m horizontal grids. Validation with weather stations shows that WN improved the overall forecast accuracy by 13%, on average, relative to HRRR forecasts. Improvements were also recorded in 71.6% of all weather stations used. However, overall WN skill scores declined at higher observed wind speeds. HRRR wind speed forecasts have an overall tendency to overpredict at lower observed wind speeds but underpredict at higher wind speeds. Downscaling increased negative wind speed biases of input HRRR forecasts even more at stations located in wind-prone lee-slope canyons. In contrast, stations located at well-exposed ridgetop sites benefitted from downscaling when the stations had negative input HRRR forecast biases, given that the ridgetops were sufficiently resolved in WN.