A Model-Agnostic Physics-Guided Adapter for Few-Shot Transfer of Coastal Flood Prediction Models to Unseen Regions

2026-09-29

AI Summary

This study addresses the substantial data requirements and high costs associated with fine-tuning deep learning flood forecasting models when transferring them across coastal regions. To this end, we propose the Physics Adapter (PA), which incorporates terrain elevation as an architectural inductive bias and integrates a differentiable wet-dry response with a gating mechanism. PA introduces physical priors through lightweight parameters without relying on PDE residuals or conservation losses. As a model-agnostic architecture, it can be seamlessly integrated into diverse heterogeneous backbone networks. Experimental results demonstrate that, in few-shot scenarios, PA reduces root mean square error by 11.5%–22.9%, significantly enhancing both generalization capability and predictive accuracy in previously unseen regions.

Abstract

Deep learning surrogates can produce high-resolution coastal flood maps orders of magnitude faster than physics-based hydrodynamic simulators, yet transferring them to new coastal regions remains costly, since generating target-region data for fine-tuning typically requires numerous time-consuming simulations. To tackle this bottleneck, we introduce the Physics Adapter (PA), a compact, architecture-agnostic adaptation interface that enables efficient few-shot transfer of flood prediction models across diverse coastal regions. PA predicts peak water level through a differentiable wet/dry response that compares terrain elevation against a learned water level, and blends this physics-structured prediction with a data-driven branch through a learned gate. Unlike physics-informed formulations, PA imposes no PDE-residual or conservation losses and instead exploits elevation as an architectural inductive bias, adding a negligible number of trainable parameters. We integrate PA into 12 heterogeneous models, and evaluate them on two coastal regions with markedly distinct geometries, topographies, and shoreline protection configurations. The performance of PA is benchmarked against a no-physics baseline, full fine-tuning, and standard parameter-efficient fine-tuning (PEFT) methods, considering both within-region generalization to unseen sea level rise values and between-region transfer. In low-shot regime (K=3), and averaged over all backbones and transfer settings, adding PA reduces root mean square error by 11.5% when only the output head is adapted on a frozen backbone, by 15.4% when combined with PEFT methods, and by 22.9% under full fine-tuning, compared to matched configurations without PA. Taken together, the findings of this work offer practitioners a concrete recipe for extending DL-based coastal flood predictors to new, data-scarce regions.

Problem

• coastal flood prediction
• few-shot transfer
• deep learning surrogate
• unseen regions
• data scarcity

Innovation

• Physics-Guided Adapter
• Few-Shot Transfer
• Model-Agnostic
• Coastal Flood Prediction
• Parameter-Efficient Fine-Tuning