Distributional Regression U-Nets for the Postprocessing of Precipitation Ensemble Forecasts

Abstract Accurate precipitation forecasts have a high socioeconomic value due to their role in decision-making in various fields such as transport networks and farming. We propose a global statistical postprocessing method for grid-based precipitation ensemble forecasts. This U-Net-based distributional regression method predicts marginal distributions in the form of parametric distributions inferred by scoring rule minimization. Distributional regression U-Nets are compared to state-of-the-art postprocessing methods for daily 24-h forecasts of 3-h accumulated precipitation over the south of France. Training data come from the Météo-France weather model AROME–ensemble prediction system (AROME-EPS) and span 3 years. A practical challenge appears when consistent data or reforecasts are not available. Distributional regression U-Nets compare favorably with the raw ensemble. In terms of continuous ranked probability score, they reach a performance comparable to quantile regression forests (QRFs). However, they are unable to provide calibrated forecasts in areas associated with high climatological precipitation. In terms of predictive power for heavy precipitation events, they outperform both QRF and semiparametric QRF with tail extensions.