While AI weather models now rival operational forecasts, how they represent the atmosphere internally remains an open question: feature attribution reveals which input patterns matter, not what the mo...
arXiv:2608. 20117v1 Announce Type: new Abstract: The emergence of large-scale Weather and Climate (W&C) datasets offers new opportunities for modeling extreme Earth events (ExEE) and their impacts using deep learning.
The paper investigates how tropical cyclone attributes are represented within the sparse autoencoder (SAE) features of the AI weather model GraphCast. By analyzing 2,115 six‑hourly timesteps from 205 cyclones, the authors map SAE activations to cyclone position, intensity, and pressure structure across multiple layers and decompositions, finding that each attribute is encoded in multiple features and that deeper layers better predict future position and pressure structure. The study demonstrates that the organization of physical‑attribute relations in SAEs is reproducible and can aid in interpreting cyclone‑related information in GraphCast.
GNN‑VIEW is an interactive, browser‑based tool designed to aid the development and interpretation of graph‑based Earth‑system models. It visualizes the graph’s information pathways to help developers debug the structure before training, and after training it applies attribution‑based explainability to show which input features influence specific forecasts. The authors demonstrate its utility in building an ocean forecasting graph and in diagnosing poor precipitation predictions in a weather model during Hurricane Laura.
The paper introduces GraphRiverCast (GRC), a topology-informed foundation model that predicts daily multivariate hydrodynamics across the global river network using sparse gauge data. By leveraging river connectivity and energetics, GRC outperforms both physics-based and traditional machine-learning models, maintaining skill over long lead times and generalizing to ungauged reaches and finer-resolution networks. The approach demonstrates how incorporating system structure can improve predictions in observation-limited Earth systems.
arXiv:2602.22293v2 Announce Type: replace Abstract: Accurate river prediction is essential for water, food and energy security, yet remains challenging across entire river networks. Machine learning...
We introduce a stretched-grid artificial intelligence (AI) weather forecasting model with 2-km resolution over the western United States and part of the Northeast Pacific and approximately 31-km resol...
The paper presents Laxmi, a retrained version of the AIFS weather model that uses satellite-based precipitation observations instead of ERA5 reanalysis data. Laxmi achieves a 19% improvement in global probabilistic accuracy, reduces drizzle overprediction by 33%, and boosts the 95th percentile Brier skill score by 57%. In a case study of 10 Indian tropical storms, Laxmi accurately forecasted 150 mm event-total precipitation in 7 events, outperforming both the original AIFS and the leading physical model IFS.
The paper presents a graph-transformer AI weather model that is fine‑tuned with high‑resolution IMERG precipitation observations, moving beyond the traditional reliance on the ERA5 reanalysis dataset. This approach yields up to a 19% improvement in medium‑range continuous ranked probability scores and a 57% better Brier skill score for extreme rainfall compared to leading operational models, while also excelling in tropical storm and drizzle prediction. The study demonstrates that directly incorporating observation‑based precipitation data into AI training can markedly enhance forecast accuracy, though physics‑based models still outperform for the heaviest events.
arXiv:2607. 23237v1 Announce Type: cross Abstract: Effective flood risk management relies on accurate forecasting, yet the "black box" nature of stateof-the-art Deep Learning models creates a barrier to trust and accountability in high-stakes public safety decisions.
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.
arXiv:2602. 16579v2 Announce Type: replace-cross Abstract: Reliable global streamflow forecasting is essential for flood preparedness and water resource management, yet data-driven models often suffer from a performance gap when transitioning from historical reanalysis to operational forecast products.