Beyond Ensemble Averages: Leveraging Climate Model Ensembles for Subseasonal Forecasting

What is it about?

Accurately forecasting temperature and precipitation on subseasonal time scales -- two weeks to two months in advance -- is extremely challenging. These forecasts would have immense value in agriculture, insurance, and economics. Our paper describes an application of machine learning techniques to improve forecasts of monthly average precipitation and 2-meter temperature using lagged physics-based predictions and observational data two weeks in advance for the entire continental United States. For lagged ensembles, the proposed models outperform standard benchmarks such as historical averages and averages of physics-based predictions. Our findings suggest that utilizing the full set of physics-based predictions instead of the average enhances the accuracy of the final forecast.

Featured Image

Photo by Abid Shah on Unsplash

Photo by Abid Shah on Unsplash