PACT: Peak-Aware Cross-Attention Graph Transformers for Efficient Storm-Surge Emulation

TL;DR

PACT introduces a novel peak-aware cross-attention graph transformer that efficiently predicts storm surges at the station level, capturing extreme events and outperforming existing models in accuracy and speed.

Contribution

The paper presents PACT, a new graph transformer architecture with a peak-aware training strategy for efficient and accurate storm-surge prediction from atmospheric forcing data.

Findings

PACT outperforms baseline models in RMSE and MAE.

Improved peak fidelity and tail preservation in surge predictions.

Requires about 3.5 seconds to generate a full seasonal surge trajectory.

Abstract

Accurate and efficient storm-surge emulation is essential for coastal hazard assessment, yet high-fidelity hydrodynamic models remain too expensive for large scenario ensembles and rapid evaluation under heterogeneous climate forcings. We present PACT, a peak-aware cross-attention graph transformer for efficient station-level storm-surge prediction from atmospheric forcing fields. PACT represents each forcing patch as a graph, encodes spatial structure with GraphSAGE, and uses a learned station query to aggregate node information through cross-attention rather than uniform pooling. A Transformer encoder models temporal dependence across the forcing history, and a horizon-query decoder generates lead-specific forecasts from a shared temporal memory. To better capture extreme events, we introduce a peak-aware learning strategy that couples a lightweight auxiliary peak-aware head with a tailored training objective, including a tail-focused loss on peak-dominated samples and a horizon-wise slope regularizer to encourage coherent multi-step evolution. Across multiple tide-gauge stations along the US Northeast coast, PACT outperforms a strong spatio-temporal graph neural network baseline in both RMSE and MAE. Diagnostics show improved peak fidelity and tail preservation for reanalysis and most CMIP6 datasets. PACT is also computationally efficient, requiring about 3.5~s to generate a full winter-season surge trajectory for one year after training. Under distribution shift across five CMIP6 forcings, PACT transfers well within the CMIP6 family but degrades markedly when transferring from reanalysis to climate-model forcings, highlighting a persistent reanalysis--GCM gap.