Preprints
https://doi.org/10.5194/egusphere-2024-3691
https://doi.org/10.5194/egusphere-2024-3691
29 Nov 2024
 | 29 Nov 2024
Status: this preprint is open for discussion.

Application of HIDRA2 Deep Learning Model for Sea Level Forecasting Along the Estonian Coast of the Baltic Sea

Amirhossein Barzandeh, Marko Rus, Matjaž Ličer, Ilja Maljutenko, Jüri Elken, Priidik Lagemaa, and Rivo Uiboupin

Abstract. Sea level predictions, typically derived from 3D hydrodynamic models, are computationally intensive and subject to uncertainties stemming from physical representation and inaccuracies in initial or boundary conditions. As a complementary alternative, data-driven machine learning models provide a computationally efficient solution with comparable accuracy. This study employs the deep learning model HIDRA2 to forecast hourly sea levels at five coastal stations along the Estonian coastline of the Baltic Sea, evaluating its performance across various forecast lead times. Compared to the regional NEMOBAL and subregional NEMOEST hydrodynamic models, HIDRA2 consistently delivers superior results, particularly across all sea level ranges and stations. While HIDRA2 struggles to capture high-frequency variability above (6 h)-1, it effectively reproduces energy in lower-frequency bands below (18 h)-1. Errors tend to average out over longer time windows encompassing multiple seiche periods, enabling HIDRA2 to surpass the overall performance of the NEMO models. These findings underscore HIDRA2’s potential as a robust, efficient, and reliable tool for operational sea level forecasting and coastal management in the Eastern Baltic Sea region.

Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this preprint. The responsibility to include appropriate place names lies with the authors.
Amirhossein Barzandeh, Marko Rus, Matjaž Ličer, Ilja Maljutenko, Jüri Elken, Priidik Lagemaa, and Rivo Uiboupin

Status: open (until 24 Jan 2025)

Comment types: AC – author | RC – referee | CC – community | EC – editor | CEC – chief editor | : Report abuse
Amirhossein Barzandeh, Marko Rus, Matjaž Ličer, Ilja Maljutenko, Jüri Elken, Priidik Lagemaa, and Rivo Uiboupin
Amirhossein Barzandeh, Marko Rus, Matjaž Ličer, Ilja Maljutenko, Jüri Elken, Priidik Lagemaa, and Rivo Uiboupin

Viewed

Total article views: 112 (including HTML, PDF, and XML)
HTML PDF XML Total BibTeX EndNote
81 26 5 112 1 3
  • HTML: 81
  • PDF: 26
  • XML: 5
  • Total: 112
  • BibTeX: 1
  • EndNote: 3
Views and downloads (calculated since 29 Nov 2024)
Cumulative views and downloads (calculated since 29 Nov 2024)

Viewed (geographical distribution)

Total article views: 112 (including HTML, PDF, and XML) Thereof 112 with geography defined and 0 with unknown origin.
Country # Views %
  • 1
1
 
 
 
 
Latest update: 13 Dec 2024
Download
Short summary
We evaluated a deep-learning model, HIDRA2, for predicting sea levels along the Estonian coast and compared it to traditional numerical models. HIDRA2 performed better overall, offering faster forecasts and valuable uncertainty estimates using ensemble predictions.