Mapping Agricultural Land Use Patterns in Benue State Using Sentinel-2 Imagery

Benue State, widely recognised as Nigeria's food basket, sustains diverse agricultural activities across its 23 local government areas. This study maps agricultural land use patterns in three selected LGAs, namely Gboko, Makurdi, and Gwer-East, using Sentinel-2 multispectral imagery acquired during the 2022 cropping season. Object-based image analysis is applied to segment and classify distinct land use categories including cropland, mixed agroforestry, fallow land, and irrigated agriculture. Spectral indices including the Normalised Difference Vegetation Index and the Soil-Adjusted Vegetation Index are computed to distinguish crop types at peak growth stages. Ground truth data collected during field campaigns are used to train and validate the classification model, achieving an overall accuracy of 87 percent. Results reveal that smallholder rainfed farming dominates the landscape, occupying over 68 percent of the mapped area. Spatial patterns indicate significant fallow rotation practices that differ markedly across soil types and proximity to seasonal streams. The maps produced serve as decision-support tools for the Benue State Agricultural Development Programme in optimising input distribution and extension service deployment. Keywords: land use mapping, Sentinel-2, agriculture, Benue State, object-based analysis