20.500.12556/DiRROS-21818 HIDRA3 a deep-learning model for multipoint ensemble sea level forecasting in the presence of tide gauge sensor failures Accurate modeling of sea level and storm surge dynamics with several days of temporal horizons is essential for effective coastal flood responses and the protection of coastal communities and economies. The classical approach to this challenge involves computationally intensive ocean models that typically calculate sea levels relative to the geoid, which must then be correlated with local tide gauge observations of sea surface height (SSH). A recently proposed deep-learning model, HIDRA2 (HIgh-performance Deep tidal Residual estimation method using Atmospheric data, version 2), avoids numerical simulations while delivering competitive forecasts. Its forecast accuracy depends on the availability of a sufficiently long history of recorded SSH observations used in training. This makes HIDRA2 less reliable for locations with less abundant SSH training data. Furthermore, since the inference requires immediate past SSH measurements as input, forecasts cannot be made during temporary tide gauge failures. We address the aforementioned issues using a new architecture, HIDRA3, that considers observations from multiple locations, shares the geophysical encoder across the locations, and constructs a joint latent state that is decoded into forecasts at individual locations. The new architecture brings several benefits: (i) it improves training at locations with scarce historical SSH data, (ii) it enables predictions even at locations with sensor failures, and (iii) it reliably estimates prediction uncertainties. HIDRA3 is evaluated by jointly training on 11 tide gauge locations along the Adriatic. Results show that HIDRA3 outperforms HIDRA2 and the Mediterranean basin Nucleus for European Modelling of the Ocean (NEMO) setup of the Copernicus Marine Environment Monitoring Service (CMEMS) by ∼ 15 % and ∼ 13 % mean absolute error (MAE) reductions at high SSH values, creating a solid new state of the art. The forecasting skill does not deteriorate even in the case of simultaneous failure of multiple sensors in the basin or when predicting solely from the tide gauges far outside the Rossby radius of a failed sensor. Furthermore, HIDRA3 shows remarkable performance with substantially smaller amounts of training data compared with HIDRA2, making it appropriate for sea level forecasting in basins with high regional variability in the available tide gauge data. sea level modeling deep learning storm surges modeliranje višine morske gladine globoko učenje poplavljanje true false true Angleški jezik Slovenski jezik Neznano 2025-04-03 12:59:50 2025-04-03 12:59:50 2025-10-24 03:49:54 0000-00-00 00:00:00 2025 0 0 Soavtorji: Hrvoje Mihanović, Matjaž Ličer, Matej Kristan; str. 605-620 iss. 3 Vol. 18 Feb. 2025 0000-00-00 Zaloznikova Objavljeno NiDoloceno 2024-07-08 2024-12-04 2025-02-04 004.85:622.847:551.461.2 1991-9603 10.5194/gmd-18-605-2025 225239555 https://gmd.copernicus.org/articles/18/605/2025/ 1 c524496e-107a-11f0-963a-001a4af901a5 https://dirros.openscience.si/Dokument.php?lang=slv&id=31453 RAZ_Rus_Marko_2025.pdf RAZ_Rus_Marko_2025.pdf 1 EA0489B4AA961560F61647728049655E 0280d2501f17ab83692a53a39d2258909d0b85946041c728b0a06bff0ce570a8 f2256811-107a-11f0-963a-001a4af901a5 https://dirros.openscience.si/Dokument.php?lang=slv&id=31454 Nacionalni inštitut za biologijo Code for HIDRA3: a robust deep-learning model for multi-point sea-surface height forecasting Training and test datasets, pretrained weights and predictions for HIDRA3 0 0 0