Anthropogenic controls over soil organic carbon distribution from the cultivated lands in Northeast China

Both natural and anthropogenic variables affect soil C distribution its pool, however studies about influence on are very limited in the literature. This study investigated effects organic carbon (SOC) changes cultivated lands of Northeast China. A total 196 topsoil samples (0–30 cm) were collected, analyzed for SOC content, 12 environmental (natural anthropogenic) selected as predictors. Natural factors included elevation, slope gradient, aspect (SA), topographic wetness index (TWI), mean annual temperature, precipitation, normalized difference vegetation index, while population (POP), gross domestic product (GDP), distance to socioeconomic center, roads, reclamation period (PER) represented variables. Three different boosted-regression trees models with combination predictors constructed, model performance was evaluated 10-fold cross-validation. We found that all had best performance, followed by topography climate variables, only However, adding greatly improved performance. Results showed PER, POP GDP key affecting content agroecosystems suggests should be main variable predicting agroecosystem a higher human influence. believe accurate prediction mapping will help formulate farmland management policies promote sequestration.