aiqbench analysis
This study demonstrates the successful transfer of Earth-based GraphCast foundation models to Martian atmospheric forecasting through fine-tuning, enabling accurate prediction of diurnal cycles and seasonal structures within just 10 epochs. The research validates that AI weather models trained on terrestrial data can be rapidly adapted for extraterrestrial environments, offering a scalable pathway for mission operations and dust storm mitigation without requiring massive new datasets from scratch.
MarsCast: Transfer Learning of AI Weather Foundation Models to Planetary Atmospheres
arXiv:2608.05054v1 Announce Type: cross Abstract: We investigate the transferability of Earth weather foundation models to planetary atmospheres by adapting the GraphCast graph neural weather forecasting model to Mars. While GraphCast achieves state-of-the-art performance for terrestrial forecasting, its applicability to non-Earth environments remains unexplored. Using the Mars Climate Database (MCD), which provides global atmospheric fields across vertical altitude levels (similar to Earth pressure levels), we evaluate zero-shot and fine-tuned GraphCast predictions of Martian temperature and wind fields. Zero-shot forecasts produce a surprisingly accurate depiction of current conditions but fail to reproduce diurnal variability and rapidly decay toward climatological mean states. To address this limitation, we fine-tune GraphCast using MCD variables and top-of-atmosphere solar radiation forcing while holding humidity constant. Fine-tuning enables rapid learning of Martian thermal variability. Within as few as 10 training epochs, the model begins to capture the diurnal cycle and forecasts up to 10 days reproduce seasonal and vertical temperature structure. Prediction quality improves with training sample size and exhibits sensitivity to seasonal initialization. These results demonstrate that Earth-trained AI weather models can be adapted to simulate Martian atmospheric dynamics, providing a pathway toward rapid planetary weather prediction to support mission operations, dust storm risk mitigation, and future human exploration.
Abstract:We investigate the transferability of Earth weather foundation models to planetary atmospheres by adapting the GraphCast graph neural weather forecasting model to Mars. While GraphCast achieves state-of-the-art performance for terrestrial forecasting, its applicability to non-Earth environments remains unexplored. Using the Mars Climate Database (MCD), which provides global atmospheric fields across vertical altitude levels (similar to Earth pressure levels), we evaluate zero-shot and fine-tuned GraphCast predictions of Martian temperature and wind fields. Zero-shot forecasts produce a surprisingly accurate depiction of current conditions but fail to reproduce diurnal variability and rapidly decay toward climatological mean states. To address this limitation, we fine-tune GraphCast using MCD variables and top-of-atmosphere solar radiation forcing while holding humidity constant. Fine-tuning enables rapid learning of Martian thermal variability. Within as few as 10 training epochs, the model begins to capture the diurnal cycle and forecasts up to 10 days reproduce seasonal and vertical temperature structure. Prediction quality improves with training sample size and exhibits sensitivity to seasonal initialization. These results demonstrate that Earth-trained AI weather models can be adapted to simulate Martian atmospheric dynamics, providing a pathway toward rapid planetary weather prediction to support mission operations, dust storm risk mitigation, and future human exploration.
Earth and Planetary Astrophysics (astro-ph.EP); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
arXiv:2608.05054 [astro-ph.EP]
(or arXiv:2608.05054v1 [astro-ph.EP] for this version)
https://doi.org/10.48550/arXiv.2608.05054 arXiv-issued DOI via DataCite (pending registration)
Full text under the publisher's open licence · 229 words
Related coverage
Integrating GNSS-Derived Zenith Wet Delay into a Weather Foundation Model Improves Precipitation Forecasting
arXiv · Jul 28, 2026
Aurora 1.5: Extending open foundation models for weather and Earth-system applications
Microsoft Research · Jul 9, 2026
Earth observation embeddings are effective sub-grid descriptors for probabilistic weather downscaling
arXiv · Sep 4, 2026
Xaurora: Generative Weather Forecasting with Denoising Stochastic Interpolants from a Foundation Model Prior
arXiv · Oct 6, 2026