Soil organic carbon (SOC) plays an important role in terrestrial ecosystems. In this study, the typical black soil area of Tieling City, Liaoning Province, was used as the study area, and digital soil mapping (DSM) technology was utilized to estimate the SOC content in Tieling City. To improve the accuracy and reduce the uncertainty of predicting the organic carbon content of topsoil (0-20 cm) in the plains region by applying machine learning methods and adding remote sensing variables to the traditional predictors.Based on 2708 soil samples and the corresponding environmental covariates, the spatial distribution of organic carbon content in topsoil (0-20 cm) in Tieling City in 2021 was mapped for the first time using ordinary kriging, regression kriging and random forest models. The primary environmental variables influencing the variability of SOC content are mean annual temperature, altitude, mean annual precipitation and vegetation cover and the interaction of the two factors enhances the effect of a single factor on the spatial heterogeneity of SOC content. We then applied the RF model to predict the spatial distribution of SOC content in 1991 and 2021 based on natural environmental variables and remote sensing variables. In the past three decades, the SOC content has always been higher in the eastern region than in the western region. The SOC content in the eastern and northern part of the study area showed an increasing trend with an area of 8245.8522 km², while the SOC content in the central and southern part of the study area showed a decreasing trend with an area of 4717.089km² and average organic carbon content increased from 21.88g/kg to 22.90g/kg, compared with 2021. Our high-resolution and high-precision estimation of the spatial distribution of SOC can provide a basis and reference for regional soil quality improvement and ecosystem optimization.
 

Soil organic carbon; Digital soil mapping; Dynamic changes; spatial heterogeneity