EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Water-State-Aware Spatiotemporal Graph Transformer Network for Water-Level Prediction

Ziang Li, Wenru Zhang, Zongying Liu, Shaoxi Li, Jiangling Hao, Chu Kiong Loo

Journal of Marine Science and Engineering · 18.11.2025 · 13 (11) · 2187

DOI: 10.3390/jmse13112187 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Marine Science and Engineering, online erschienen am 18.11.2025. Verfasst von Ziang Li und 5 weiteren Autor:innen. Die Arbeit wurde bisher 4-mal zitiert (Crossref) und nennt 35 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China, the 2023 DMU Navigation College First-Class Interdisciplinary Research.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Accurate water-level prediction is a critical component for ensuring safe maritime navigation, optimizing port operations, and mitigating coastal flooding risks. However, the complex, non-linear spatiotemporal dynamics of water systems pose significant challenges for current forecasting models. The proposed framework introduces three key innovations. First, a dual-weight graph construction mechanism integrates geographical proximity with Dynamic Time Warping (DTW)-derived temporal similarity to better represent hydrodynamic connectivity in coastal and estuarine environments. Second, a state-aware weighted loss function is designed to enhance predictive accuracy during critical hydrological events, such as storm surges and extreme tides, by prioritizing the reduction in errors in these high-risk periods. Third, the WS-STGTN architecture combines graph attention with temporal self-attention to capture long-range dependencies in both space and time. Extensive experiments are conducted using water-level data from five stations in the tidal-influenced lower Yangtze River, a vital artery for shipping and a region susceptible to coastal hydrological extremes. The results demonstrate that the model consistently surpasses a range of baseline methods. Notably, the WS-STGTN achieves an average reduction in Mean Squared Error (MSE) of 27.6% compared to the standard Transformer model, along with the highest coefficient of determination (R2≈0.96) across all datasets, indicating its stronger explanatory power for observed water-level variability. This work provides a powerful tool that can be directly applied to improve coastal risk management, marine navigation safety, and the operational planning of port and coastal engineering projects.

Förderung

  • National Natural Science Foundation of China
  • the 2023 DMU Navigation College First-Class Interdisciplinary Research

Zitieren

APA

Li, Z., Zhang, W., Liu, Z., Li, S., Hao, J., & Loo, C. K. (2025). Water-State-Aware Spatiotemporal Graph Transformer Network for Water-Level Prediction. Journal of Marine Science and Engineering, 13(11), 2187. https://doi.org/10.3390/jmse13112187

Vancouver

Li Z, Zhang W, Liu Z, Li S, Hao J, Loo CK. Water-State-Aware Spatiotemporal Graph Transformer Network for Water-Level Prediction. Journal of Marine Science and Engineering. 2025;13(11):2187. doi:10.3390/jmse13112187

BibTeX
@article{Li2025,
  author = {Li, Ziang and Zhang, Wenru and Liu, Zongying and Li, Shaoxi and Hao, Jiangling and Loo, Chu Kiong},
  title = {Water-State-Aware Spatiotemporal Graph Transformer Network for Water-Level Prediction},
  journal = {Journal of Marine Science and Engineering},
  year = {2025},
  volume = {13},
  number = {11},
  pages = {2187},
  doi = {10.3390/jmse13112187}
}
Kostenlos starten →