Improving Seasonal Forecast Using Probabilistic Deep Learning

• Published in: Journal of advances in modeling earth systems 2022-03, Vol.14 (3), p.n/a
• Main Authors: Pan, Baoxiang, Anderson, Gemma J., Goncalves, André, Lucas, Donald D., Bonfils, Céline J. W., Lee, Jiwoo
• Format: Article
• Language: English
• Subjects: Air temperature; Benchmarks; Big Data; Climate; Climate models; Climate variability; Deep learning; ENVIRONMENTAL SCIENCES; Forecast verification; Forecasting; General circulation models; GEOSCIENCES; MATHEMATICS AND COMPUTING; Methods; Neural networks; Precipitation; Random variables; seasonal forecast; Seasonal forecasting; Seasons; Simulation; Spatial data; Variability; variational inference; Weather forecasting
• ISSN: 1942-2466; 1942-2466
• DOI: 10.1029/2021MS002766

The path toward realizing the potential of seasonal forecasting and its socioeconomic benefits relies on improving general circulation model (GCM) based dynamical forecast systems. To improve dynamical seasonal forecasts, it is crucial to set up forecast benchmarks, and clarify forecast limitations posed by model initialization errors, formulation deficiencies, and internal climate variability. With huge costs in generating large forecast ensembles, and limited observations for forecast verification, the seasonal forecast benchmarking and diagnosing task proves challenging. Here, we develop a probabilistic deep learning‐based statistical forecast methodology, drawing on a wealth of climate simulations to enhance seasonal forecast capability and forecast diagnosis. By explicitly modeling the internal climate variability and GCM formulation differences, the proposed Conditional Generative Forecasting (CGF) methodology enables bypassing crucial barriers in dynamical forecast, and offers a top‐down viewpoint to examine how complicated GCMs encode the seasonal predictability information. We apply the CGF methodology for global seasonal forecast of precipitation and 2 m air temperature, based on a unique data set consisting 52,201 years of climate simulation. Results show that the CGF methodology can faithfully represent the seasonal predictability information encoded in GCMs. We successfully apply this learned relationship in real‐world seasonal forecast, achieving competitive performance compared to dynamical forecasts. Using this CGF as benchmark, we reveal the impact of insufficient forecast spread sampling that limits the skill of the considered dynamical forecast system. Finally, we introduce different strategies for composing ensembles using the CGF methodology, highlighting the potential for leveraging the strengths of multiple GCMs to achieve advantgeous seasonal forecast. Plain Language Summary Seasonal forecast benefits a broad range of societal sectors. However, current dynamical seasonal forecast systems are considerably hindered by observation, model, and computation limitations. We develop a machine learning probabilistic forecast model that learns from climate simulations to infer possible climate patterns a season ahead. The model achieves competitive performance for global precipitation and temperature forecast, compared to the costly dynamical forecasts. More importantly, it offers crucial implications for understanding the limitations and imp