Land-Use Dynamics in the N’Zi Watershed at M’Bahiakro, Côte d’Ivoire: A Multi-temporal Geospatial Analysis (2015–2024)
BAÏ Ruth *
Unit Training and Research in Earth Sciences and Mining Resources, Félix Houphouët Boigny University, Abidjan, Ivory Coast, Côte d'Ivoire and Wiss Center for Scientific Research of Ivory Coast, Abidjan, Côte d’Ivoire.
ANOUMAN Djoro Gauthier-Lopez
Unit Training and Research in Science and Environment Management, University Nangui Abrogoua, Abidjan, Ivory Coast, Côte d’Ivoire.
ABO N'Guessan Kouamé Emmanuel
Unit Training and Research in Science and Environment Management, University Nangui Abrogoua, Abidjan, Ivory Coast, Côte d’Ivoire.
N’CHO Achié Hervé
Unit Training and Research in Science and Environment Management, University Nangui Abrogoua, Abidjan, Ivory Coast, Côte d’Ivoire.
KOUASSI Kouakou Lazare
Unit Training and Research in Science and Environment Management, University Jean Lorougnon Guédé, Côte d'Ivoire.
KOUAME Kouassi Innocent
Unit Training and Research in Science and Environment Management, University Nangui Abrogoua, Abidjan, Ivory Coast, Côte d’Ivoire.
ETTIEN Djetchi Jean-Baptiste
Unit Training and Research in Earth Sciences and Mining Resources, Félix Houphouët Boigny University, Abidjan, Ivory Coast, Côte d'Ivoire and Wiss Center for Scientific Research of Ivory Coast, Abidjan, Côte d’Ivoire.
*Author to whom correspondence should be addressed.
Abstract
This study analyzes land cover dynamics in the N’Zi watershed at M’Bahiakro (Central-Eastern Côte d’Ivoire) between 2015 and 2024. Satellite images acquired in 2015, 2020, and 2024 were preprocessed to correct for radiometric and atmospheric effects and then classified into six land cover categories using supervised maximum likelihood classification: bare soil and urbanized areas, dense forests, degraded forests, savanna, cultivated land and fallow land, and water bodies. The classification accuracy, assessed using a confusion matrix, shows an overall accuracy greater than 85% and a high Kappa coefficient, confirming the reliability of the results. The diachronic analysis reveals major transformations. Between 2015 and 2020, savannas (-14.84 ha; -75.64%) and dense forests (-6.19 ha; -35.43%) decreased significantly, while degraded forests (+7.00 ha; +95.37%) and cultivated land (+6.09 ha; +45.58%) increased. Bare soils and urbanized areas declined slightly (-2.71 ha; -12.68%), while bodies of water increased (+10.65 ha; +51.13%). Between 2020 and 2024, the changes are even more pronounced: bare soil and urbanized areas increased significantly (+29.82 ha; +159.72%), while degraded forests (-9.65 ha; -67.29%) and cultivated land (-15.89 ha; -81.70%) decreased. Dense forests (+2.69 ha; +23.85%) and savannas (+1.08 ha; +22.59%) increased slightly, while water bodies decreased (-8.04 ha; -25.54%). These trends reflect increasing human pressure and the ongoing conversion of natural areas into agricultural land. The study recommends the use of predictive models (Markov, CA-Markov, Land Change Modeler, Random Forest) to anticipate future changes and support agricultural planning and environmental sustainability. Limitations include the lack of comprehensive field data, seasonal variations between images, and some classification uncertainties, despite the high Kappa coefficient.
Keywords: Land-Use Dynamics, land cover change, geospatial analysis, N’zi watershed, M’bahiakro, Côte d’ivoire