Soil Organic carbon (SOC) is vital to the soil’s ecosystem functioning as well as improving soil fertility. Slight variation in C in the soil has significant potential to be either a source of CO₂ in the atmosphere or a sink to be stored in the form of soil organic matter. However, modeling SOC spatiotemporal changes was challenging due to lack of data to represent the high spatial heterogeneity in soil properties. Less expensive techniques, digital soil mapping (DSM) combined with space-for-time substitution (SFTS), were applied to predict the present and projected SOC stock for temperature and rainfall projections under different climate scenarios represented by the four Representative Concentration Pathways (RCPs): RCP2.6, RCP4.5, RCP6, and RCP8.5). The relationship between environmental covariates (n = 16) and measured SOC stock (148 samples) was developed using a random forest model. Then, the temporal changes in SOC stock over the baseline were developed for the top 30 cm soil depth of the selected districts (Chiro Zuria, Kuni, Gemechis and Mieso) of West Hararghe Zone at 30 m resolution. The model validation using the random sample of 20% of the data showed that the model explained 44% of the variance (R²) with a root mean square error (RMSE) of 8.96, a mean error (ME) of 0.16, and a Lin’s concordance correlation coefficient (CCC) of 0.88. Temperature was the most important predictor factor influencing the spatial distribution of SOC stock. An overall net gain of SOC stock over the present C stock was expected in the study area by 2050. The gain in areas with the lower baseline SOC stock counterbalanced the loss in areas with the higher baseline stock. The changes in the SOC stock depended on land use land cover (LULC), soil type, and agro-ecological zones. By 2050, cropland is supposed to lose its SOC stock under all RCPs; therefore, appropriate decisions are crucial to compensate for the loss of C.

Location-specific information on soil organic carbon (SOC) helps to identify potential sources and sinks of carbon. The objective of this study was to produce a 30m-resolution digital map of SOC stock for selected districts in West Hararghe Zone of Ethiopia. 148 soil samples(0-30cm) were collected for SOC analysis and BD estimation. Eighteen environmental covariates were acquired from satellite sources, digital elevation model (DEM), and maps. A random forest model was fitted to the data. The accuracy of the prediction was tested using the 10-fold cross-validation method. The model explained 36% of the variance (R2) with a root mean square error of 9.51. The most relevant predictors were normalized difference vegetation index (NDVI), temperature, and elevation. The predicted total SOC stock value was 240.5TgC and ranged between 17.91 and 70.16 t/ha, with a mean of 38.82 t/ha. The soils under agricultural land had a higher amount of carbon than forestland. The mean SOC stock was relatively higher in fine-textured soils and lower in soils, which are dominantly found in drier areas. This study could help provide information on a less expensive method of SOC stock estimation. In addition, the study produced an updatable baseline SOC stock map.