The potential of Sentinel-1 imagery for flood event detection : a satellite vs. hydraulic model comparison

Breznik, Jana (author), Oštir, Krištof (author), Rak, Gašper (author)

Comparing flooded area detection between the hydraulic modelling and Sentinel-1 radar imagery processed by Automatic Detection of Water Bodies (Sentinel-1 ADWB) confirmed that SAR (Synthetic Aperture Radar) imagery shows good potential for improving flood studies. The study was performed in a study area of the Krka River floodplain (Slovenia). The focus was to analyse the overlapping areas between the Sentinel-1 ADWB and numerical hydraulic modelling and to recognise the factors that play a role in the differences between those areas. At the same time, the study also demonstrated the applicability of Sentinel-1 ADWB. The overlapping between the two data sources was up to 66%. The most significant differences between the data were in areas simulated by the hydraulic modelling but not detected by satellite imagery processed by Sentinel-1 ADWB. These were mainly areas with dense vegetation, such as forests, where the satellite (as expected) had difficulty detecting floods, and detection borders, where detection was difficult due to lower water depths and/or detection limitations due to spatial resolution or region growing process. Areas less well modelled by the hydraulic model than satellite imagery were also identified, indicating regions where Sentinel-1 ADWB can improve flood studies as additional data for hydraulic simulation. The study shows that users need to understand the drawbacks and advantages of modes for obtaining the data – radar satellite imagery (problems under dense vegetation, spatial resolution, etc.) and hydraulic model (lack of quality input data, the precipitation directly above the studied area, water inflow from the hinterland, etc.). Based on the knowledge of the areas where Sentinel-1 ADWB data can be useful, the data’s potential for calibration and verification processes of hydraulic studies is suggested.

• Language: English
• Keywords: flooded areas, Sentinel-1, radar satellite imagery, hydraulic modelling, SAR, automatic detection, flood data comparison
• Work type: Article
• Typology: 1.01 - Original Scientific Article
• Organization: FGG - Faculty of Civil and Geodetic Engineering
• Publication status: Published
• Publication version: Version of Record
• Year: 2025
• Number of pages: 16 str.
• Numbering: Vol. 651, art. 132587
• PID: 20.500.12556/RUL-166297
• UDC: 627.53
• ISSN on article: 0022-1694
• DOI: 10.1016/j.jhydrol.2024.132587
• COBISS.SI-ID: 221040387
• Publication date in RUL: 06.01.2025

Record is a part of a journal

• Title: Journal of hydrology
• Shortened title: J. Hydrol.
• Publisher: North-Holland, Elsevier
• ISSN: 0022-1694
• COBISS.SI-ID: 25750784

Licences

• License: CC BY-NC 4.0, Creative Commons Attribution-NonCommercial 4.0 International
• Link: http://creativecommons.org/licenses/by-nc/4.0/
• Description: A creative commons license that bans commercial use, but the users don’t have to license their derivative works on the same terms.

Secondary language

• Language: Slovenian
• Keywords: poplavljena območja, Sentinel-1, radarski satelitski posnetki, hidravlično modeliranje, SAR, samodejno zaznavanje, primerjava podatkov o poplavah