An interpretable significant wave height forecasting model using a causal AI framework with error correction

doi:10.1016/j.oceaneng.2025.124135

An interpretable significant wave height forecasting model using a causal AI framework with error correction

2025-12-31

0

PRE

AI

DOI：10.1016/j.oceaneng.2025.124135

Original

Original request for help

Save

Abstract

En 中文

• A causal AI framework integrating a C-Seq2Seq model with XGBoost error correction for SWH forecasting, is developed. • The causal AI framework achieves statistically significant RMSE improvements of 1.08 %–18.68 % over an LSTM baseline across 1–24‑h lead times. • Both PCMCI and SHAP analysis consistently identify wind speed and gust of wind as key causal drivers with significant contributions to SWH forecasts.

An interpretable significant wave height forecasting model using a causal AI framework with error correction

Abstract

Journal

Organization