How to disentangle sea-level rise and a number of other processes influencing coastal floods?

• Published in: Atti della Accademia nazionale dei Lincei. Rendiconti Lincei. Scienze fisiche e naturali Vol. 35; no. 2; pp. 371 - 380
• Main Authors: Orlić, Mirko, Pasarić, Miroslava
• Format: Journal Article
• Language: English
• Published: Cham Springer International Publishing 01-06-2024; Springer Nature B.V
• Subjects: Atmospheric depressions; Atmospheric forcing; Biomedicine; Coastal processes; Decomposition; Earth and Environmental Science; Earth Sciences; Environment; History of Science; Lagoons; Learning algorithms; Life Sciences; Machine learning; Modelling; Physics; Sea level; Storm surges; The Mediterranean System: a Hotspot for Climate Change and Adaptation; Wind; Hindcast; Sea-level data; Decomposition; Flood; Forecast; Venice
• ISSN: 2037-4631; 1720-0776
• DOI: 10.1007/s12210-024-01242-z

On 12 November 2019 at 21:50 UTC, about 85% of the city of Venice was flooded, due to the sea-level height reaching 189 cm—the second largest value ever recorded there. Both the operational modeling system and the machine learning system underestimated the event by about 40 cm. To explain the underestimation, the sea-level data recorded in the area were subjected to the decomposition method that had been gradually developed at the Andrija Mohorovičić Geophysical Institute over the last 40 or so years. The procedure revealed eight phenomena contributing to the sea level maximum: vertical land motion, sea-level rise, variable annual change, surge caused by planetary atmospheric waves, tide, storm surge, meteotsunami, and wind set-up inside the lagoon. It turned out that a combined contribution of the last two phenomena was almost equal to the difference between observed sea-level height and forecasted/hindcasted values. Consequently, the difference was related to a secondary atmospheric depression, which had caused both meteotsunami and wind set-up inside the lagoon but was not adequately captured by the operational modeling system nor was it allowed for by the machine learning system. Since the decomposition method proved to be useful in the Adriatic Sea, it is expected that the method could be applicable in other basins around the world if they are prone to strong and multifaceted atmospheric forcing.
Graphical abstract