DIGITAL MAPPING OF ORGANIC CARBON CONTENT AND STOCKS IN SOILS ON THE SLOPE (USING THE EXAMPLE OF A KEY SITE IN THE CIS-SALAIR DRAINED PLAIN)

Digital mapping of the content and stock of soil organic carbon (SOC) in the arable layer (0–30 cm) of soils based on a set of geospatial data characterizing soil formation factors was carried out on the territory of the Cis-Salair drained plain . For comparison, maps of the content of the SOC were compiled on the basis of various sets of geospatial data (selected according to the correlation table): 1) BIO11+RVI; 2) Longitude+CNBL; 2) CAT10+KK+Texture. The modeling method is multiple linear regression. The result of modeling of the SOC content with the data set (CAT10+KK+Texture) according to visual assessment and with lower values ​​of the model performance indicators (RMSE, MAPE) was selected as the best compared to other datasets. Based on the best model of the SOC content, a soil bulk density map was compiled using the pedotransfer function, which was used to compile a map of SOC stocks. In the soils of the upper part of the slope, the highest content and stocks of SOC were revealed, whereas in the soils down the slope, a gradual decrease in these values ​​was found in the series of soils agrochernozems → agro-dark gray → agro -gray. The soil bulk density, on the contrary, increased downhill.