PDE foundation models are skillful AI weather emulators for the Martian atmosphere

TL;DR

This paper demonstrates that PDE foundation models pretrained on numerical PDE solutions can be adapted to create effective weather emulators for the Martian atmosphere, even with limited data and computational resources.

Contribution

The authors extend the Poseidon PDE foundation model from 2D to 3D and show its effectiveness in Martian weather prediction with sparse initial conditions.

Findings

34.4% performance improvement with pretraining and extension

Effective modeling with limited training data (~34 GB)

Model adapts to sparse initial conditions

Abstract

We show that AI foundation models that are pretrained on numerical solutions to a diverse corpus of partial differential equations can be adapted and fine-tuned to obtain skillful predictive weather emulators for the Martian atmosphere. We base our work on the Poseidon PDE foundation model for two-dimensional systems. We develop a method to extend Poseidon from two to three dimensions while keeping the pretraining information. Moreover, we investigate the performance of the model in the presence of sparse initial conditions. Our results make use of four Martian years (approx.~34 GB) of training data and a median compute budget of 13 GPU hours. We find that the combination of pretraining and model extension yields a performance increase of 34.4\% on a held-out year. This shows that PDEs-FMs can not only approximate solutions to (other) PDEs but also anchor models for real-world problems with complex interactions that lack a sufficient amount of training data or a suitable compute budget.