Geospatial Modelling and Remote Sensing Assessment of Flood-Induced Land Degradation in Nigeria’s Middle Belt

Abstract

Flooding is a natural hydrological phenomenon, caused by natural and human-induced factors, that occurs when water temporarily inundates land that is normally dry. The direct impact of this could include the loss of human lives, displacement of populations, destruction of homes, roads, bridges, farmlands, and public utilities, and damage to economic assets. Meanwhile, the indirect effect could include disruption of livelihoods, food insecurity due to crop loss, degradation of soil quality, and long-term socio-economic setbacks. This study employed geospatial techniques, including remote sensing, field survey, and GIS-based Multi-Criteria Decision Analysis (MCDA), to assess flood-induced land degradation in the Middle Belt of Nigeria. The methodology involved analysing Landsat 9 imagery for land use/land cover (LULC) classification, computing the Normalized Difference Water Index (NDWI) for water body detection, and using Shuttle Radar Topography Mission (SRTM) data for terrain analysis. Field-collected GPS coordinates of historically flooded areas were overlaid for validation. LULC Classification revealed that vegetation dominated the landscape (72.64%), followed by bare land (11.6%), forest (11.37%), built-up areas (3.75%), and water bodies (0.64%). Built-up areas, though minimal in coverage, were categorized as high risk due to impermeable surfaces, which accelerate runoff. Through the MCDA weighted overlay, the final flood risk model classified the region into Low (0.3%), Medium (84.8%), and High (13.8%) flood risk zones. High-risk areas were largely associated with built-up, low-lying, and river-adjacent areas. The integration of remote sensing, terrain data, and field surveys provided a scientifically validated model for flood risk and land degradation mapping in the Middle Belt area of Nigeria, therefore it is recommended for use.