Détails Publication
Modelling past and future land use and land cover dynamics in the Nakambe River Basin, West Africa,
Auteur(s): Gnibga Issoufou Yangouliba, Benewindé Jean‑Bosco Zoungrana, Kwame Oppong Hackman, Hagen Koch, Stefan Liersch4, Luc Ollivier Sintondji, Jean‑Marie Dipama, Daniel Kwawuvi, Valentin Ouedraogo, Sadraki Yabré, Benjamin Bonkoungou, Madou Sougué, Aliou Gadiaga, Bérenger Koffi
Auteur(s) tagués: DIPAMA Jean-Marie
Renseignée par : DIPAMA Jean-Marie
Résumé

Understanding land use and cover (LULC) dynamic is of great importance to sustainable development in Africa where
deforestation is a common problem. This study aimed to assess the historical and future dynamics of LULC in the Nakambé
River Basin. Landsat images were used to determine LULC dynamics for the years 1990, 2005 and 2020 using Random
Forest classifcation system in Google Earth Engine while the predicted LULC of 2050 was simulated using the Markov
Chain and Multi-Layer-Perceptron neural network in Land Change Modeler. The fndings showed signifcant changes in
LULC patterns. From 1990 to 2020, woodland and shrubland decreased by − 45% and − 68%, respectively, while water
body, cropland and bare land/built-up increased by 233%, 51%, and 75%, correspondingly. From 2020 to 2050, the results
revealed that under the Business-as-usual scenario, bare land/built-up and water bodies could continue to increase by 99%
and 1%, respectively. However, cropland, shrubland, and woodland could decrease by − 32.61%, − 33.91%, and − 46.86%,
respectively. Under the aforestation scenario, the contrary of Business-as-usual could occur. While woodland, shrubland,
and cropland would increase by 22.24%, 51.57%, and 18.13%, correspondingly, between 2020 and 2050, the area covered by
water bodies and bare land/built-up will decrease by − 6.16% and − 39.04%, respectively. The results of this research give
an insight into past and future LULC dynamics in the Nakambé River Basin and suggest the need to strengthen the policies
and actions for better land management in the region.

Mots-clés

Land use/land cover, Random forest, Markov chain, Multi-layer-perceptron neural network, · Land change modeler, Nakambé River Basin

939
Enseignants
5607
Publications
49
Laboratoires
84
Projets