Spatial Econometric Analysis of Agricultural Land Use Change and Productivity in Benue River Valley

Agricultural land use transformation in the Benue River Valley, spanning Benue and Kogi States in North Central Nigeria, is driven by population growth, land degradation, and the encroachment of pastoral and urban land uses on previously cultivated areas. This study applies spatial econometric methods, including the spatial lag model and the spatial error model estimated using maximum likelihood, to examine the determinants of agricultural land productivity changes across 48 administrative units over three agricultural census periods from 2012 to 2022. Land use change data are derived from supervised classification of Sentinel-2 multispectral imagery, while productivity data are obtained from the Nigeria Agricultural Sample Census. Spatial weight matrices are constructed based on queen contiguity and inverse distance weighting. Moran's I test confirms significant positive spatial autocorrelation in productivity levels, validating the use of spatial econometric specifications over ordinary least squares. The spatial lag model indicates that a 10 percent productivity improvement in a neighbouring unit is associated with a 3.7 percent productivity spillover to the focal unit, reflecting technology diffusion and market integration effects. Land fragmentation, soil degradation index, and distance from flood-irrigated areas emerge as the most significant negative determinants of productivity after spatial dependence is controlled. Keywords: spatial econometrics, agricultural productivity, land use change, Benue River Valley, spatial lag model.