Variation of soil organic carbon across different land covers and land uses in the greater Gaborone region of Botswana

Abstract

Soils are a potentially viable sink for atmospheric carbon and could contribute to mitigating global climate change. Soil Organic Carbon (SOC) content exhibits considerable spatial variability both horizontally (land use) and vertically (soil profile). Land cover land use (LCLU) is one of the key determinants of SOC stock, hence there is a significant variation of SOC across different LCLUs. This study aimed to investigate the spatial variability of SOC across different LCLUs in the Greater Gaborone region of Botswana. Remotely sensed data used for image classification was obtained from the United States Geological Survey (USGS) Earth Explorer (www.usgs.gov). The imagery used in this study was Sentinel-2A obtained in the month of March with cloud content of less than 10% for easy interpretation. Image classification was done using a supervised classification method based on a Maximum Likelihood classifier. The major LCLU types identified in the area included water bodies, trees dominated, cropland, shrubland, bare land, and built-up. The Walkley and Black method, core method, Bouyoucos hydrometric method and pH meter were used to determine SOC content, bulk density, soil texture and pH, respectively. Soil bulk density, pH and sand fraction showed a negative correlation with SOC content, while silt and clay showed a positive correlation. The total SOC stock in the study area was estimated to be 4.36 MtC, with trees dominated areas accounting for 1.13 MtC (25.9%), shrubland 2.83 MtC (64.9%), cropland 0.14 MtC (3.2%), built-up 0.22 MtC (5.1%), and bare land 0.04 MtC (0.9%), hence indicating that trees dominated and shrubland were good sequesters of carbon in the Greater Gaborone.