AI model improves weather forecasts weeks ahead with simpler targets
S2S-JEPA: Predicting the Predictable at Subseasonal-to-Seasonal Timescales
Artificial IntelligenceMachine Learning
Summary
Weather forecasts for two weeks to two months ahead are very hard because many details can't be predicted reliably. The authors propose focusing on the stable, predictable parts of the weather instead of trying to get every small detail right. They adapted a computer vision technique that predicts these main patterns rather than all details. Their AI model performs as well as or better than the best physics-based forecasts at 5 to 6 weeks ahead. This approach may help farmers, energy managers, and others who need longer-range weather predictions.
What this means in practice
- •For agriculture planners: Use improved AI forecasts to anticipate weather conditions weeks ahead for crop planning and risk management.
- •For energy grid operators: Incorporate more accurate subseasonal weather predictions to better manage energy supply and demand over several weeks.
Authors
Chenyu Dong, Gianmarco Mengaldo
Abstract
The subseasonal-to-seasonal (S2S) timescale, roughly from two weeks to two months ahead, is a critical forecast window for sectors such as agriculture, energy, and water management. Yet, it is widely known as the `predictability desert'. Recent AI weather models excel up to two weeks ahead but deteriorate beyond, largely because they are trained to predict fine-scale details that are neither predictable nor essential at S2S timescales. We argue that a more physically grounded objective is to forecast only the slowly varying components that remain predictable. Computer vision reached the same conclusion with the Joint-Embedding Predictive Architecture (JEPA), which predicts in latent space, discarding unpredictable details. In this work, we introduce S2S-JEPA, which brings the JEPA paradigm to S2S forecasting. It is tailored to this task through design elements from state-of-the-art AI weather models. S2S-JEPA achieves comparable skill to the gold-standard ECMWF physics-based ensemble and surpasses it on multiple metrics at weeks 5 to 6.