Developing a localized vegetation classification system for sustainable land use management in Kebbi State, Nigeria

Accurate vegetation classification is crucial for environmental monitoring, natural resource management, and climate change modelling. This study develops a localized vegetation classification system using the Normalized Difference Vegetation Index (NDVI) and machine learning algorithms for Kebbi State, Nigeria. Landsat 8 imagery and field observations were used to train a Random Forest model, achieving an overall accuracy of 88.2%. The results show significant differences in NDVI values across vegetation types, effectively distinguishing between grasslands, shrubs, and barren lands. The classification system demonstrates the potential of NDVI for vegetation classification in Kebbi State, supporting sustainable land use management practices such as reforestation, crop selection, and land degradation monitoring. This study contributes to developing localized vegetation classification systems, addressing regional specificities in vegetation characteristics and promoting informed decision-making for environmental conservation.