A Continental‐Scale Estimate of Soil Organic Carbon Change at NEON Sites and Their Environmental and Edaphic Controls

Current carbon cycle models focus on the effects of climate and land‐use change on primary productivity and microbial‐mineral dependent carbon turnover in the topsoil, while less attention has been paid to vertical soil processes and soil‐dependent response to land‐use change along the profile. In this study, a spatial‐temporal analysis was used to estimate soil organic carbon (SOC) change in topsoil/A horizon and subsoil/B horizon at National Ecological Observatory Network (NEON) sites, USA over 30 years. To separate the effects of land‐use, environmental, and edaphic factors on SOC change, space‐for‐time substitution was used in combination with the Continuous Change Detection and Classification algorithm and Structural Equation Modeling. Results showed that (a) under natural vegetation, Spodosols and Inceptisols found in the eastern NEON sites had substantial topsoil SOC accumulation (+0.4 to +1.2 Mg C ha−1 year−1), while Inceptisols and Andisols in the west had a comparable magnitude of topsoil SOC loss (−0.5 to −1.8 Mg C ha−1 year−1); (b) Mollisols and Alfisols in the Central Plains sites were susceptible to significant SOC loss under farming and grazing; (c) Runoff/erosion and leaching potential, vertical translocation, and mineral sorption were the most important factors controlling SOC variation across the NEON sites. Our work could be used to parameterize ecosystem models simulating SOC change. Plain Language Summary We estimated the 30‐year soil organic carbon (SOC) change in topsoil/A horizon and subsoil/B horizon at National Ecological Observatory Network (NEON) sites based on their soil measurements to identify where, in which soil type, and under which land‐use change scenario large SOC increase and/or decrease occurred. Within each eco‐climatic domain, the soil plots where soil samples were taken were categorized into different land‐cover subgroups from the remote sensing images over 30 years. By controlling the eco‐climatic domain and the soil type, the SOC change for each land‐use change scenario was estimated. In order to understand the mechanisms of SOC change, we modeled the SOC concentrations in both topsoil and subsoil with their environmental and soil factors. We found that most NEON sites presented SOC loss over the 30 years and more prominently in the topsoil. Under the land‐use change scenario ofconverting natural vegetation to croplands and pastures, there will be significant SOC loss after 30‐year of cultivation and grazing. Soi