Subseasonal forecasting of the weather two to six weeks in advance is critical for resource allocation and climate adaptation but poses many challenges for the forecasting community. At this forecast horizon, physics-based dynamical models have limited skill, and the targets for prediction depend in a complex manner on both local weather variables and global climate variables. Recently, machine learning methods have shown promise in advancing the state of the art but only at the cost of complex data curation, integrating expert knowledge with aggregation across multiple relevant data sources, file formats, and temporal and spatial resolutions.To streamline this process and accelerate future development, we introduce SubseasonalClimateUSA, a curated dataset for training and benchmarking subseasonal forecasting models in the United States. We use this dataset to benchmark a diverse suite of models, including operational dynamical models, classical meteorological baselines, and ten state-of-the-art machine learning and deep learning-based methods from the literature. Overall, our benchmarks suggest simple and effective ways to extend the accuracy of current operational models. SubseasonalClimateUSA is regularly updated and accessible via the https://github.com/microsoft/subseasonal_data/ Python package.

Inspired by the demands of real-time climate and weather forecasting, we develop optimistic online learning algorithms that require no parameter tuning and have optimal regret guarantees under delayed feedback. Our algorithms—DORM, DORM+, and AdaHedgeD—arise from a novel reduction of delayed online learning to optimistic online learning that reveals how optimistic hints can mitigate the regret penalty caused by delay. We pair this delay-as-optimism perspective with a new analysis of optimistic learning that exposes its robustness to hinting errors and a new meta-algorithm for learning effective hinting strategies in the presence of delay. We conclude by benchmarking our algorithms on four subseasonal climate forecasting tasks, demonstrating low regret relative to state-of-the-art forecasting models.

The prediction of drought is of major importance in climate-related studies, hydrologic engineering, wildlife or agricultural studies. This study explores the ability of two machine learning methods to predict 1, 3, 6 and 12 months standardized precipitation and evapotranspiration index (SPEI) for the Wilsons Promontory station in Eastern Australia. The two methods are multiple linear regression (MLR) and artificial neural networks (ANN). The data-driven models were based on combinations of the input variables: mean precipitations, mean, maximum and minimum temperatures and evapotranspiration, for data between 1915 and 2012. Two performance metrics were used to compare the performance of the optimum MLR and ANN models: the coefficient of determination (R2) and the root mean square error (RMSE). It was found that ANN provided greater accuracy than MLR in forecasting the 1, 3, 6 and 12 months SPEI.