Generative assimilation and prediction for weather and climate

2025-03-04

AI Summary

Existing weather–climate prediction models primarily focus on pure forecasting and lack organic integration with data assimilation, leading to error accumulation in long-term predictions and limiting applicability from seasonal to millennial timescales. This paper proposes GAP, a unified deep generative framework that— for the first time—embeds generative modeling (via latent-variable variational inference) into the end-to-end assimilation–prediction pipeline, jointly enforcing observational constraints, dynamical evolution modeling, and responses to external forcings. By incorporating physics-informed priors, multi-source observational fusion, and long-range temporal modeling, GAP breaks the traditional paradigm of decoupled assimilation and forecasting, unifying probabilistic state inference with long-term stable simulation. Experiments demonstrate that GAP matches state-of-the-art ensemble methods in data assimilation, probabilistic weather forecasting, and seasonal prediction, and—critically—achieves the first stable millennial-scale climate simulation, accurately reproducing variability from diurnal to decadal timescales.

Technology Category

• Machine Learning: Deep Generative Models & Autoencoders
• Natural Language Processing: Generation
• Computer Vision: Generative Adversarial Networks (GANs) for Vision

Application Category

• Web Mining and Content Analysis: Web data generation and simulation
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Social Networks and Social Media: Generative AI / large language models and their impact on social systems

Abstract

Machine learning models have shown great success in predicting weather up to two weeks ahead, outperforming process-based benchmarks. However, existing approaches mostly focus on the prediction task, and do not incorporate the necessary data assimilation. Moreover, these models suffer from error accumulation in long roll-outs, limiting their applicability to seasonal predictions or climate projections. Here, we introduce Generative Assimilation and Prediction (GAP), a unified deep generative framework for assimilation and prediction of both weather and climate. By learning to quantify the probabilistic distribution of atmospheric states under observational, predictive, and external forcing constraints, GAP excels in a broad range of weather-climate related tasks, including data assimilation, seamless prediction, and climate simulation. In particular, GAP is competitive with state-of-the-art ensemble assimilation, probabilistic weather forecast and seasonal prediction, yields stable millennial simulations, and reproduces climate variability from daily to decadal time scales.

Problem

• Improves weather and climate prediction accuracy
• Addresses error accumulation in long-term forecasts
• Integrates data assimilation with prediction models

Innovation

• Generative Assimilation and Prediction (GAP) framework
• Deep generative model for weather and climate
• Quantifies atmospheric states under multiple constraints