Generative AI models enable efficient and physically consistent sea-ice simulations

2025-08-20

AI Summary

Accurately and efficiently simulating sea ice dynamics—characterized by scale invariance, anisotropy, and physical complexity—remains a major challenge for Earth system models. To address this, we propose GenSIM, the first generative AI model designed for pan-Arctic sea ice prediction. Given only atmospheric forcing as input, GenSIM jointly forecasts 12-hour evolutions of sea ice concentration, thickness, and drift while preserving physical consistency, reproducing brittle dynamical behaviors, statistical properties, and long-term decline trends. Its key innovations include: (i) the first implicit generative modeling of the sea ice–ocean coupled system, enabling automatic learning of slow-varying oceanic processes; and (ii) robust out-of-distribution generalization across diverse climate scenarios. Compared to conventional numerical models, GenSIM achieves substantial computational speedup, enabling ensemble forecasting and large-scale climate simulations.

Technology Category

• Cognitive Modeling & Cognitive Systems: Simulating Human Behavior
• Search and Optimization: Sampling/Simulation-based Search
• Multiagent Systems: Agent-Based Simulation and Emergent Behavior

Application Category

• Social Networks and Social Media: Generative AI / large language models and their impact on social systems
• User Modeling, Personalization and Recommendation: User modeling and simulation for interactive and conversational systems
• Web Mining and Content Analysis: Web data generation and simulation

Abstract

Sea ice is governed by highly complex, scale-invariant, and anisotropic processes that are challenging to represent in Earth system models. While advanced numerical models have improved our understanding of the sea-ice dynamics, their computational costs often limit their application in ensemble forecasting and climate simulations. Here, we introduce GenSIM, the first generative AI-based pan-Arctic model that predicts the evolution of all relevant key properties, including concentration, thickness, and drift, in a 12-hour window with improved accuracy over deterministic predictions and high computational efficiency, while remaining physically consistent. Trained on a long simulation from a state-of-the-art sea-ice--ocean system, GenSIM robustly reproduces statistics as observed in numerical models and observations, exhibiting brittle-like short-term dynamics while also depicting the long-term sea-ice decline. Driven solely by atmospheric forcings, we attribute GenSIM's emergent extrapolation capabilities to patterns that reflect the long-term impact of the ocean: it seemingly has learned an internal ocean emulator. This ability to infer slowly evolving climate-relevant dynamics from short-term predictions underlines the large potential of generative models to generalise for unseen climates and to encode hidden physics.

Problem

• Improving sea-ice simulation accuracy and physical consistency
• Reducing computational costs for ensemble forecasting
• Predicting key sea-ice properties efficiently

Innovation

• Generative AI model for sea-ice simulations
• Predicts key properties with improved accuracy
• Learns internal ocean emulator from atmospheric forcings