NIVA: A Multimodal Foundation Model for Actionable Earth System Intelligence
This work addresses the need for AI models that can model coupled Earth system dynamics for extended-range weather and climate prediction, providing a proof of concept for foundation models in this domain.
NIVA is a multimodal foundation model that learns unified representations across Earth system components, demonstrating in a two-modality setting (ocean and atmosphere) that it can capture coupled dynamics and accurately predict major climate indices, enabling subseasonal-to-seasonal prediction.
Recent advances in AI-driven weather and climate modeling have improved forecast skill while reducing computational cost. However, existing data-driven approaches are limited in their ability to model coupled Earth system dynamics, which is required for extending predictability beyond the ~2-week horizon. To address this, we introduce NIVA, a multimodal foundation model designed to learn unified representations across Earth system components. While the full framework targets atmosphere, ocean, ice, and land interactions, we focus here on a two-modality setting (ocean and atmosphere) as a controlled proof of concept to evaluate whether foundation models can learn coupled dynamics. Trained on large-scale Earth system simulations, NIVA learns physically meaningful cross-modal structure, providing a foundation for subseasonal-to-seasonal prediction. As initial validation, we show that NIVA captures key modes of climate variability through accurate prediction of major climate indices.