Landscape historic trough spatial changes of mangroves in Benin Coastal area (West-Africa) using random forest algorithm applied to Landsat images
DOI :
https://doi.org/10.52638/rfpt.2026.736Résumé
Mangrove forests, recognized as the main ecosystem of coastal zones and providing various goods and services, are facing are facing increasing anthropogenic pressures and require continuous investigation. Ensuring the sustainable management of these ecosystems implies permanent monitoring of their dynamics. In line with this, the present study carried out on the mangrove forests of Benin (West-Africa) aimed to evaluate the effect of anthropic perturbation on their cover dynamics over the past four decades (including both Ramsar sites 1017 and 1018). The adopted method focused on remote sensing techniques using Landsat images acquired between 1986 and 2023. The Random Forest algorithm was used for image processing. Results revealed that whatever the period, the overall accuracy values obtained after post-classification analysis ranged from 84.56% to 96.52% indicating a correct image classification. From these results, a decline in mangrove cover was observed between 1986 and 2013 followed by an increase from 2013 to 2023. In Ramsar site 1017, mangroves lost about 61.90% (3261.06 ha) of their initial area from 1986 to 2000. Between 2000 and 2014, mangrove cover loss was negligible and estimated at 2.29% (59.04 ha). In contrast, from 2014 to 2023, mangroves experienced an increase estimated at about 16.66% (324.63 ha). The same trend was observed in Ramsar site 1018, with a loss of 31.12% (459.63 ha) and 27.44% (279.18 ha) respectively from 1986 to 2001 and from 2001 to 2013 followed by a gain estimated at 39.89% (279.18 ha) between 2013 and 2023. Landscape changes analysis revealed that regardless of the site or period, mangroves were mainly converted into meadow, other plant and water body (e.g. in Ramsar site 1017, from 1986 to 2001, about 568.26 ha of mangroves were converted into meadows, 33.66 ha into other plant, and 35.46 ha into water body). The principal changes observed in this study focused on Benin coastal zones confirm the vulnerability of mangrove forests facing anthropogenic pressure.
Keywords: Mangrove, Dynamics, Remote sensing, Benin.
Téléchargements
Téléchargements
Publié-e
Comment citer
Numéro
Rubrique
Catégories
Licence
(c) Tous droits réservés MAHOUTIN GILDAS SERGE ZANVO, Nicolas DELBART, Paul PASSY, Kolawolé Valère SALAKO, Romain GLELE-KAKAI 2026

Cette œuvre est sous licence Creative Commons Attribution 4.0 International.
Les auteurs conservent les droits d'auteur et accordent à la revue le droit de première publication de l'œuvre sous une licence Creative Commons Attribution (CC-BY) 4.0 qui permet à d'autres de partager l'œuvre avec une reconnaissance de la paternité de l'œuvre et de sa publication initiale dans cette revue.