Data used to drive the Double Layer Carbon Model in the Qinling Mountains.

We used a process-oriented soil carbon model, the Double Layer Carbon Model (DLCM) (Li et al., 2022a), to estimate the spatiotemporal dynamics of SOC in different soil layers and further evaluate the impacts of different climate response functions on SOC estimates in the Qinling Mountains. The DLCM is an advanced and comprehensive tool designed to simulate SOC dynamics for both the top 20 cm layer (SOC20​) and the deeper 20­­–100 cm layer (SOC20–100​). The fundamental principle of the DLCM revolves around accurately predicting SOC dynamics by simulating the interactions between fresh organic matter inputs, microbial activity, and existing soil carbon pools. The DLCM defines four soil carbon pools, categorized based on their location within the soil profile and their decomposition rates. The model divides the soil profile into topsoil (0-20 cm) and subsoil (20–100 cm) layers to match the SOC maps of the corresponding two layers generated by data-driven models. Each of these layers contains a young carbon pool (CY) with a higher decomposition rate and an old carbon pool (CO) with a lower decomposition rate. Specifically, the topsoil contains the young topsoil carbon pool (CYt) and the old topsoil carbon pool (COt), while the subsoil contains the young subsoil carbon pool (CYs) and the old subsoil carbon pool (COs). These compartments help accurately simulate the dynamics of SOC by considering both the fast-cycling and slow-cycling components of organic matter decomposition. The DLCM optimizes initial SOC stocks using extensive spatial simulations, ensuring accurate baseline carbon levels. It also incorporates climate change responses, adjust decomposition rates based on climate and environmental changes, and lead to robust estimates under different climatic scenarios. The simulation process of the DLCM involves initializing SOC stocks with spatially detailed baseline data, adding organic matter inputs based on vegetation production, and simulating microbial decomposition while adjusting for climate variables such as temperature and soil moisture. The model performs layer-specific calculations to capture depth-specific SOC dynamics. It relies on comprehensive input data, including initial SOC stocks, climate data, and vegetation production to drive these simulations.Here we provided the dataset used to drive the DLCM, including annual net primary production (NPP) from 1982 to 2018, SOC maps in the two soil layers during the 1980s, annual mean temperature maps from 1982 to 2018, and annual surface soil moisture and root-zone soil moisture maps from 1982 to 2018. SOC20 and SOC20-100 maps in the Qinling Mountains with a spatial resolution of 1 km × 1 km during the 1980s were extracted from our previous SOC datasets, which were generated by a machine learning algorithm (Li et al., 2022b). The spatial patterns of the two SOC maps are shown in Fig. 1b. Annual vegetation NPP maps with a spatial resolution of 1 km × 1 km during 1982-2018 were calculated by the Carnegie-Ames-Stanford (CASA) model in our previous study (Li et al., 2021). ROOT_NPP_Parameter maps are from our previous study (Li et al. 2022c). We calculated the annual average temperature maps based on the monthly average temperature maps with a spatial resolution of 1 km × 1 km during 1982-2018, which were collected from the high resolution temperature dataset in China (Peng et al., 2019). Annual surface soil moisture and root-zone soil moisture maps with a spatial resolution of 0.1° × 0.1° were extracted from the Global Land Evaporation Amsterdam Model (GLEAM) version 4.1a datasets (Hulsman et al., 2023). We resampled soil moisture maps to a resolution of 1 km × 1 km by using the nearest neighbor resampling method.ReferenceLi, H., Wu, Y., Liu, S., Zhao, W., Xiao, J., Winowiecki, L.A., Vågen, T.-G., Xu, J., Yin, X., Wang, F., Sivakumar, B., Cao, Y., Sun, P., Zhang, G., 2022a. The Grain-for-Green project offsets warming-induced soil organic carbon loss and increases soil carbon stock in Chinese Loess Plateau. Science of The Total Environment 837, 155469. https://doi.org/10.1016/j.scitotenv.2022.155469Li, H., Wu, Y., Liu, S., Xiao, J., Zhao, W., Chen, J., Alexandrov, G., Cao, Y., 2022b. Decipher soil organic carbon dynamics and driving forces across China using machine learning. Global Change Biology 28, 3394–3410. https://doi.org/10.1111/gcb.16154Li, H., Wu, Y., Liu, S., Xiao, J., 2021. Regional contributions to interannual variability of net primary production and climatic attributions. Agricultural and Forest Meteorology 303, 108384. https://doi.org/10.1016/j.agrformet.2021.108384Li, H., Wu, Y., Liu, S., Zhao, W., Xiao, J., Winowiecki, L.A., Vågen, T.-G., Xu, J., Yin, X., Wang, F., Sivakumar, B., Cao, Y., Sun, P., Zhang, G., 2022c. The Grain-for-Green project offsets warming-induced soil organic carbon loss and increases soil carbon stock in Chinese Loess Plateau. Science of The Total Environment 837, 155469. https://doi.org/10.1016/j.scitotenv.2022.155469Peng, S., Ding, Y., Liu, W., Li, Z., 2019. 1 km monthly temperature and precipitation dataset for China from 1901 to 2017. Earth System Science Data 11, 1931–1946. https://doi.org/10.5194/essd-11-1931-2019Hulsman, P., Keune, J., Koppa, A., Schellekens, J., Miralles, D.G., 2023. Incorporating Plant Access to Groundwater in Existing Global, Satellite-Based Evaporation Estimates. Water Resources Research 59, e2022WR033731. https://doi.org/10.1029/2022WR033731