Graph neural network for colliding particles with an application to sea ice floe modeling

TL;DR

This paper presents a Graph Neural Network-based model for sea ice dynamics that captures collisions, offering a more efficient alternative to traditional numerical methods, validated with synthetic data.

Contribution

The paper introduces a Collision-captured Network (CN) that integrates GNNs with data assimilation for scalable, accurate sea ice modeling based on physical interactions.

Findings

The GNN model accelerates sea ice trajectory simulations.

The model maintains accuracy with synthetic data, even with limited observations.

It demonstrates potential for improved forecasting in marginal ice zones.

Abstract

This paper introduces a novel approach to sea ice modeling using Graph Neural Networks (GNNs), utilizing the natural graph structure of sea ice, where nodes represent individual ice pieces, and edges model the physical interactions, including collisions. This concept is developed within a one-dimensional framework as a foundational step. Traditional numerical methods, while effective, are computationally intensive and less scalable. By utilizing GNNs, the proposed model, termed the Collision-captured Network (CN), integrates data assimilation (DA) techniques to effectively learn and predict sea ice dynamics under various conditions. The approach was validated using synthetic data, both with and without observed data points, and it was found that the model accelerates the simulation of trajectories without compromising accuracy. This advancement offers a more efficient tool for forecasting in marginal ice zones (MIZ) and highlights the potential of combining machine learning with data assimilation for more effective and efficient modeling.