TITLE:
Oil Spill-Related Coastal Pollution in Madingo-Kayes, Republic of Congo
AUTHORS:
Jean Bienvenu Dinga, Christ Mokomba Madzo Makouezi, Auguste Merveilles Bawatokobuta Nanitelamio, Wilfrid Innocent Ndebeka
KEYWORDS:
Dynamics, Coastal Pollution, Hydrocarbons, Hyperspectral Images, Kouilou
JOURNAL NAME:
Journal of Water Resource and Protection,
Vol.17 No.9,
September
24,
2025
ABSTRACT: The environmental impacts caused by frequent oil spills along the Congolese coast pose a significant concern. The main objective of this work is to study the dynamics of coastal pollution due to hydrocarbons using hyperspectral images in the Madingo-Kayes area, located in the Kouilou department in the southwest of the Republic of Congo. To achieve this, two multi-temporal Hyperion hyperspectral datasets, acquired by the EO-1 satellite on February 18, 2014, and December 30, 2016, respectively, were retrieved from the USGS database. These images were then analyzed and processed using ENVI 5.0 and QGIS 3.16 software to detect and map surface hydrocarbon indicators in the study area using three methods. The first method is based on the calculation of sub-band ratios in the absorption wavelength ranges of hydrocarbons (1700 - 1800 and 2300 - 2350 nm). The HD ratio detected 443 hydrocarbon indicators in the 2014 Hyperion image, while 629 indicators were detected in the 2016 image. The HI ratio detected 155 indicators in 2014, compared to 133 in 2016. The second method is the SAM classification, based on prototype hydrocarbon absorption signatures. It detected 387 indicators in the 2014 image and 73 in the 2016 image. The third method is the SSM classification, based on vegetation stress signatures caused by hydrocarbons within the 680–740 nm wavelength range. It detected 168 hydrocarbon-related indicators in vegetation in 2014 and 18 in 2016.