Journal article Open Access

Flood Hazard and Risk Mapping by Applying an Explainable Machine Learning Framework Using Satellite Imagery and GIS Data

Gerasimos Antzoulatos; Ioannis-Omiros Kouloglou; Marios Bakratsas; Anastasia Moumtzidou; Ilias Gialampoukidis; Anastasios Karakostas; Francesca Lombardo; Roberto Fiorin; Daniele Norbiato; Michele Ferri; Andreas Symeonidis; Stefanos Vrochidis; Ioannis Kompatsiaris

Flooding is one of the most destructive natural phenomena that happen worldwide, leading to the damage of property and infrastructure or even the loss of lives. The escalation in the intensity and number of flooding events as a result of the combination of climate change and anthropogenic factors motivates the need to adopt real-time solutions for mapping flood hazards and risks. In this study, a methodological framework is proposed that enables the assessment of flood hazard and risk levels of severity dynamically by fusing optical remote sensing (Sentinel-1) and GIS-based data from the region of the Trieste, Monfalcone and Muggia Municipalities. Explainable machine learning techniques were utilised, aiming to interpret the results for the assessment of flood hazard. The flood inventory was randomly divided into 70%, used for training, and 30%, employed for testing. Various combinations of the models were evaluated for the assessment of flood hazard. The results revealed that the Random Forest model achieved the highest F1-score (approx. 0.99), among others utilised for generating flood hazard maps. Furthermore, the estimation of the flood risk was achieved by a combination of a rule-based approach to estimate the exposure and vulnerability with the dynamic assessment of flood hazard.

Files (15.2 MB)
Name Size
Sustainability_1551221_REVISED_LR_Round_2.pdf
md5:d2559c7b1bf2665825e8f44905f2262e
15.2 MB Download
27
26
views
downloads
Views 27
Downloads 26
Data volume 395.3 MB
Unique views 18
Unique downloads 19

Share

Cite as