29 Aug 2022
29 Aug 2022
Status: this preprint is open for discussion.

HIDRA2: deep-learning ensemble storm surge forecasting in the presence of seiches – the case of Northern Adriatic

Marko Rus1, Anja Fettich2, Matej Kristan1, and Matjaž Ličer2,3 Marko Rus et al.
  • 1Faculty of Computer and Information Science, Visual Cognitive Systems Lab, University of Ljubljana, Ljubljana, Slovenia
  • 2Slovenian Environment Agency, Office for Meteorology, Hidrology and Oceanography, Ljubljana, Slovenia
  • 3National Institute of Biology, Marine Biology Station, Piran, Slovenia

Abstract. We propose a new deep-learning architecture HIDRA2 for sea level and storm surge modeling, which is extremely fast to train and apply, and outperforms both our previous network design HIDRA1 and the state-of-the-art numerical ocean model (a NEMO engine with sea level data assimilation), over all sea level bins and all forecast lead times. The architecture of HIDRA2 employs novel atmospheric, tidal and SSH feature encoders, as well as a novel feature fusion and SSH regression block. HIDRA2 was trained on surface wind and pressure fields from a single member of ECMWF atmospheric ensemble and on Koper tide gauge observations during years 2006–2018, and evaluated on the data from June 2019 to December 2020. Compared to HIDRA1, the overall mean absolute forecast error is reduced by 13.9 %, while on storm surge events it is lower by even a larger margin of 25.1 %. Consistent superior performance over HIDRA1 as well as NEMO is observed in both tails of the sea level distribution. Power spectrum analysis indicates that HIDRA2 most accurately represents the energy density peaks centered on the two lowest Adriatic wind-induced free oscillation eigenmodes (seiches) among all tested models. To assign model errors to specific frequency bands covering diurnal and semi-diurnal tides and the lowest two basin seiches, sea level band-pass filtering of several historic storm surge events is applied. HIDRA2 performs well across all frequency bands and most accurately predicts amplitudes and temporal phases of the Adriatic basin seiches. This is shown to be an important forecasting benefit due to the high sensitivity of total Adriatic storm surge sea level to the phase lag between peak tide and peak seiche.

Marko Rus et al.

Status: open (until 24 Oct 2022)

Comment types: AC – author | RC – referee | CC – community | EC – editor | CEC – chief editor | : Report abuse
  • CEC1: 'Comment on egusphere-2022-618', Juan Antonio Añel, 21 Sep 2022 reply
    • CC1: 'Reply on CEC1', Matjaz Licer, 22 Sep 2022 reply
    • CC2: 'Reply on Editor comment. License and Input Training/Testing datasets.', Matjaz Licer, 30 Sep 2022 reply
  • RC1: 'Comment on egusphere-2022-618', Anonymous Referee #1, 25 Sep 2022 reply

Marko Rus et al.

Marko Rus et al.


Total article views: 173 (including HTML, PDF, and XML)
HTML PDF XML Total BibTeX EndNote
119 44 10 173 4 2
  • HTML: 119
  • PDF: 44
  • XML: 10
  • Total: 173
  • BibTeX: 4
  • EndNote: 2
Views and downloads (calculated since 29 Aug 2022)
Cumulative views and downloads (calculated since 29 Aug 2022)

Viewed (geographical distribution)

Total article views: 166 (including HTML, PDF, and XML) Thereof 166 with geography defined and 0 with unknown origin.
Country # Views %
  • 1
Latest update: 30 Sep 2022
Short summary
We propose a new fast and reliable deep-learning architecture HIDRA2 for sea level and storm surge modeling. HIDRA2 features new feature encoders and a fusion-regression block. We test HIDRA2 on Adriatic storm surges, which depend on an interaction between tides and seiches. We demonstrate that HIDRA2 learns to effectively mimic the timing and amplitude of Adriatic seiches better than other tested models. This is essential to reliable HIDRA2 predictions of total storm surge sea levels.