Subseasonal forecasting—predicting temperature and precipitation 2 to 6 weeks ahead—is critical for effective water allocation, wildfire management, and drought and flood mitigation. Recent international research efforts have advanced the sub-seasonal capabilities of operational dynamical models, yet temperature and precipitation prediction skills remains poor, partly due to stubborn errors in representing atmospheric dynamics and physics inside dynamical models. AER Principal Sceintist Judah Cohen worked with data scientsits to develop a machine learning based adaptive bias correction (ABC) method that combines state-of-the-art dynamical forecasts with observations. When applied to the leading subseasonal model from the European Centre for Medium- Range Weather Forecasts (ECMWF), ABC improves temperature forecasting skill by 60-90% and precipitation forecasting skill by 40-69% in the contiguous U.S. We couple these performance improvements with a practical workflow, based on Cohort Shapley, for explaining ABC skill gains and identifying higher-skill windows of opportunity based on specific climate conditions.
Figure1. Spatial distribution of skill across the contiguous U.S. and the years 2018–2021, with average skill displayed above each map. While dynamical model skill drops precipitously at sub-seasonal timescales (weeks 3-4 and 5-6), ABC attenuates the degradation, doubling or tripling the skill of CFSv2 and boosting ECMWF skill 40-90%. Taking the same raw model forecasts as input, ABC also provides consistent improvements over operational debiasing protocols, tripling the precipitation skill of debiased CFSv2 and improving that of debiased ECMWF by 70%.
Citation: Adaptive Bias Correction for Improved Subseasonal Forecasting
S. Mouatadid, P. Orenstein, G. Flaspohler, J. Cohen, M. Oprescu, E. Fraenkel, L. Mackey
Nature Communications (to appear).