Extreme Seas: Next-Generation Forecasting of Significant Wave Height

Extreme sea states amplify risks to maritime operations, offshore infrastructure, and vulnerable coastlines globally. Spectral models like WAVEWATCH III (WW3) underpin operational wave forecasting. Recent innovations leverage the convergence of physical modeling and machine learning, particularly CNN/Transformer architectures coupled with WW3, to unlock unprecedented capabilities in predicting wave dynamics quickly and precisely. Meanwhile, the proliferation of multiple validation sources (satellite altimetry, SAR, and buoy networks) provides new pathways to quantify and reduce predictive uncertainty. This Topic is positioned at the forefront of Extreme Seas science, forging next-generation methodologies for Significant Wave Height forecasting through integrated physics-ML frameworks and multi-source validation.

The goals of this Researhc Topic are the following:

-Develop scalable physics-ML frameworks (e.g., Transformer-based WW3 surrogates) to enhance gridded, track-aware predictions beyond single-point outputs, enabling spectral correction and downscaling.

-Develop robust methodologies for calibrated ensembles, tail modeling, and return-level estimation using extreme value theory, enhancing reliability across engineering design and hazard warnings.

-Assess WW3 and other spectral models using harmonized multi-source data (satellite altimetry/SAR, buoy networks, HF radar), enhancing global-to-coastal wave forecasting accuracy.

-Create benchmark datasets, reproducible workflows, and FAIR practices for community reuse.

-Leverage next-generation significant wave height forecasting to ship routing and operability, offshore design criteria, early warning, and coastal impact chains.

The scope and themes are:

Original Research, Reviews, Perspectives/Policy briefs, Technology/Methods, and Brief Reports across:

• Spatial–temporal SWH prediction beyond single-point outputs; gridded and track-aware forecasts.

• Hybrid physics–ML: CNN/Transformers; WW3–ML coupling; spectral correction and downscaling.

• Uncertainty & extremes: calibrated ensembles, tail modeling, return-level estimation, EVT.

• Multi-source validation: WW3 and other spectral models, satellite altimetry/SAR, buoy networks, HF radar; cross-sensor harmonization.

• Data/Code: benchmark datasets, reproducible workflows, and FAIR practices for community reuse.

• Applications: ship routing and operability, offshore design criteria, early warning, and coastal impact chains

Article Types

• Original Research, Review, Mini Review

• Perspective, Policy Brief, Opinion

• Technology & Code, Methods, Brief Research Report

• Data Report (where appropriate)

Keywords: Significant wave height, Wave Modelling, Machine Learning, Hybrid Forecasting, ​Extreme Events, Uncertainty, Multi-Source Validation

Important note: All contributions to this Research Topic must be within the scope of the section and journal to which they are submitted, as defined in their mission statements. Frontiers reserves the right to guide an out-of-scope manuscript to a more suitable section or journal at any stage of peer review.