Mapping and assessment of land use/land Cover Changes in West Kordofan State - Sudan, using remote sensing and GIS

Abstract

This study was conducted in three localities (El Salam, El Snout, and Babanousa) in West Kordofan State, Sudan. The main objective was to identify, analyse, monitor, and map land use/land cover dynamics during the period from 1986 to 2014. Three Landsat images with multi-temporal dates: Thematic Mapper (TM) (1984), Enhanced Thematic Mapper (ETM)+ (2003), and Landsat 8 OLI (2014) were acquired during the dry season. The images were geo-referenced and radiometrically corrected. Image classification, change detection, and accuracy assessment were applied. A total of 200 control points was registered using the Global Positioning System. Remotely sensed data were processed and analysed using ERDAS 9.1 and ArcGIS 10.0 software. The land use/land cover classification results were grouped into six categories: dense forests, agricultural and bare lands, shrublands, rangelands, low-density forests, and water bodies. The overall transformation of LULC during the studied period (1986-2014) is indicated by the increase of water bodies from 0.01% to 0.19%, agricultural and bare lands from 31.39% to 44.6%, and range lands from 12.69% to 39.24%. However, during the same period, dense forests, low-dense forests, and shrublands showed a decreasing rate from 20.97% to 8.93%, 8.85% to 3.61%, and 26.09% to 3.41%, respectively. The overall accuracy assessment of classified imagery for TM, ETM+, and Landsat 8 OLI indicated Kappa statistics of 0.75, 0.81, and 0.88, respectively. The study recommended conducting periodic assessments and monitoring for LULC changes using remote sensing techniques to support strategic land-use planning.