Spatial distribution and severity of soil acidity across land uses in Gozamen District, Upper Blue Nile Basin, Ethiopia

Ethiopia is facing a soil pH problem that has not yet been accurately measured due to limited research on this issue. This study attempts to quantify the range of acid soils and develop land use-specific soil acidity characteristics in the selected kebeles of Gozamen District of northwestern Ethiopia through field soil survey, GIS, and digital soil mapping. Soil samples were taken using stratified random sampling over six kebeles of the Gozamen District. Soil pH was analyzed based on the standard procedure of FAO (2020) and GLOSOLAN (2019). Atmospheric images of Sentinel-2 Level-2 A were analyzed for vegetation cover via Random Forest (RF) Machine Learning Models, producing land-use/land-cover map with an overall accuracy of 92% and a Kappa Coefficient of 0.88. Spatial variability of soil pH was found to range from a minimum of 4.11 to a maximum of 6.58 throughout the selected kebeles of Gozamen District. Approximately 43% of the Gozamen District (selected kebeles) is very strongly acidic, and 39% is strongly acidic, with only 16% exhibiting moderate to slightly acidic pH, making them potentially useful for agricultural production. Agricultural and grazing lands were identified as having the greatest level of acidification, while forest ecosystems retained comparatively higher levels of soil pH, demonstrating the positive impact of natural vegetation in preserving healthy soils. This combination of field soil sampling, Sentinel-2 imaging, RF classification, and geostatistical interpolation offers an effective and economical method to develop accurate high-resolution mapping for soils that will ultimately lead to site-specific liming, precision soil fertility management, and planning for land restoration as part of Ethiopia’s agricultural transformation and sustainability initiatives.