This is the authors' abstract. We don't add key points for this paper.
Northern China has experienced some of the most intensive land use changes globally, yet the consequent impacts on soil carbon dynamics and regional source-sink patterns, particularly the divergent responses of soil organic carbon (SOC) and soil inorganic carbon (SIC), remain poorly quantified. This study examined how land use change reshapes SOC and SIC pools and modulates their sensitivity to environmental drivers. Using nationwide soil survey data from the 1980s and 2010s, we applied random forest models and partial least squares structural equation modeling to disentangle the responces SOC and SIC pools for land use and environmental change. Results revealed that Northern China functioned as a net soil carbon source over the 30-year interval, losing approximately 0.56 Gt of SOC and 0.56 Gt of SIC. Land use type emerged as the dominant control on carbon source-sink patterns. Grassland transitioned into strong carbon sources driven primarily by SIC losses of 1.51 Gt, whereas unutilized land exhibited estimated net carbon gains with apparent SIC accumulation of 2.06 Gt. Beyond altering carbon stocks, land use change reversed regional source-sink patterns and markedly heightened the sensitivity of soil carbon dynamics to environmental changes with explanatory strength increment of 0.11 for SOC and 0.09 for SIC, respectively. Specifically, SOC responded primarily to biotic drivers such as vegetation productivity, while SIC exhibited tighter coupling with increasing atmospheric CO2 fluctuations. The main contribution of this study is revealing how SOC and SIC respond differently to environmental drivers following land use change. Practically, these results indicate that regional carbon management strategies must fully consider land use change effects and explicitly include soil inorganic carbon in carbon neutrality accounting frameworks.
Answers are grounded in this paper and cite their sources.
No takes yet. Share an insight, caveat, or question.
Tang et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: