Implementation of a semi-automatic approach for detection and extraction of oil slicks using Sentinel-1: Case study of the Moroccan exclusive economic zone

Due to its strategic geographic location, Morocco experiences heavy ship traffic, which inevitably has consequences. Ship accidents and illegal discharges, commonly referred to as oil spills, are frequent and pose significant harm to the marine ecosystem. Therefore, employing new technologies, such...

Full description

Saved in:
Bibliographic Details
Published inRemote sensing applications Vol. 36; p. 101375
Main Authors Houali, Youssef, Tounsi, Yassine, El Ghandour, Fatima-Ezzahra, Habib, Abir, Labbassi, Yassine, Labbassi, Kamal
Format Journal Article
LanguageEnglish
Published Elsevier B.V 01.11.2024
Subjects
Automatic identification system
Marine pollution
Oil spill
Sentinel-1
Synthetic aperture radar
Oil spill
Automatic identification system
Marine pollution
Sentinel-1
Synthetic aperture radar
Online AccessGet full text

Cover

More Information
Summary:Due to its strategic geographic location, Morocco experiences heavy ship traffic, which inevitably has consequences. Ship accidents and illegal discharges, commonly referred to as oil spills, are frequent and pose significant harm to the marine ecosystem. Therefore, employing new technologies, such as satellite monitoring, is crucial and will greatly enhance the ability to monitor and assess these oil spills. The present work is based on Sentinel-1 SAR images and aims to take into evidence the extent of the hydrocarbon marine pollution in all of the Moroccan EEZ. More than 220 possible cases were detected and mapped during the last decade (2014–2024). A batch process was developed to automate the detection of oil spills by identifying areas with reduced Normalized Radar Cross Section (NRCS) values. This process incorporates corrections, masks, wind field estimation, and oil clustering. The method successfully detected approximately 80% of the cases between Kenitra and Casablanca. The closest detected spill is located 1 km from the coast, the largest spill covers an area of 90 km2, and the most extended spill measures about 145 km. This study can help raise awareness among governmental authorities about potential environmental damage and can be used to train artificial intelligence models, which could simulate oil slick propagation.
ISSN:2352-9385
2352-9385
DOI:10.1016/j.rsase.2024.101375