Evaluating a Hybrid Prognostic–Diagnostic Model That Improves Wind Forecast Resolution in Complex Coastal Topography

Abstract

The results from a hybrid approach that combines the forecasts of a mesoscale model with a diagnostic wind model to produce high-resolution wind forecasts in complex coastal orography are evaluated. The simple diagnostic wind model [Winds on Critical Streamline Surfaces (WOCSS)] was driven with forecasts (on a 9-km grid) from the Coupled Ocean?Atmosphere Mesoscale Prediction System (COAMPS) to obtain detailed near-surface wind forecasts with 3-km horizontal spacing. Forecasts were produced by this hybrid model for four cold-season cases?two frontal and two nonfrontal?over the central California coastal region. They were compared with 3-km forecasts from the innermost COAMPS full physics model nest and with winds observed at 35 surface sites scattered throughout the study domain. The evaluation sought to determine the conditions for which the hybrid approach performs well and those for which it does not. The performance (in terms of bias and root-mean-square error) was evaluated 1) when there were and were not fronts and 2) for the early (6?18 h) and late (21?36 h) periods of the mesoscale model forecasts. The geographic distribution of performance was also examined to see if forecasts were affected by mountains and oceans. The hybrid approach performed best during stable, nonfrontal conditions. There were no clearly defined geographic effects on hybrid performance. The computation requirements of the full physics mesoscale model nested down to 3 km are substantially greater than those of the hybrid approach. Suggestions are given for further improvements.