Geospatial Analysis of Landuse and Landcover change during 2001-2022, Nigeria

Reliable, spatially explicit evidence on land-cover change is essential for climate adaptation, agricultural planning, and monitoring of SDG targets in West Africa. We map and quantify national-scale land-cover dynamics in Nigeria between 2000/2001 and 2022 using the ESA C3S/CCI Land Cover Level-4 product (300 m). Annual categorical layers were assembled and harmonized in Google Earth Engine. To preserve class integrity, we applied nearest-neighbor resampling for grid alignment, masked no-data (0), and restricted accounting to “present classes only.” We computed per-class areas, net changes, and a pixel-wise transition matrix, and summarized transitions into policy-oriented groups (e.g., forest/savanna → cropland; cropland → urban; stable). Results show broad national stability with targeted reconfiguration. Rainfed cropland and woody savanna/deciduous cover dominate both epochs; nonetheless, rainfed cropland expanded modestly at the national scale, urban land approximately tripled in share (from ~0.3% to ~1.1%), and woody vegetation reorganized toward more open/deciduous physiognomies. Shrub and grass classes contracted, while permanent water remained largely stable at 300 m. Dominant flows include shrubland/grassland → cropland and cropland/mosaics → urban near major corridors, alongside internal shifts among woody classes across the Sudanian–Guinean belt. Cross-sensor validation with MODIS MCD12Q1 (IGBP) indicates only moderate agreement after legend grouping, reflecting differences in spatial resolution, seasonality, and ontologies; accuracy improves when restricting to homogeneous cores and coarser support, framing discrepancies as uncertainty bounds rather than simple error. The workflow provides a reproducible national evidence base for Nigeria, suitable for routine monitoring and state-level planning. Findings highlight the need to balance cropland expansion and rapid urban growth with ecosystem connectivity, flood mitigation, and restoration in savanna regions. Future work should integrate multi-sensor fusion, probabilistic agreement metrics, and driver analyses to strengthen attribution and decision relevance.