Mapping stocks of soil organic carbon and soil total nitrogen in Liaoning Province of China

• Published in: Geoderma 2017-11, Vol.305, p.250-263
• Main Authors: Wang, Shuai, Zhuang, Qianlai, Wang, Qiubing, Jin, Xinxin, Han, Chunlan
• Format: Article
• Language: English
• Subjects: Boosted regression tree; Digital soil mapping; Environmental variables; Soil organic carbon; Soil total nitrogen
• ISSN: 0016-7061; 1872-6259
• DOI: 10.1016/j.geoderma.2017.05.048

Estimation of carbon and nitrogen stocks is important for quantifying carbon and nitrogen sequestration as well as greenhouse gas emissions and inventorying national carbon and nitrogen balances. For Liaoning province of China, we estimated the vertical distribution of soil organic carbon (SOC), soil total nitrogen (STN), bulk density (BD), and mapped their spatial distribution at five standard soil depth intervals (0–5, 5–15, 15–30, 30–60 and 60–100cm) using nine environmental variables as predictors including precipitation, temperature, land use, elevation, system for automated geoscientific analyses (SAGA) wetness index, and Normalized Difference Vegetation Index (NDVI). The highest average contents of SOC and STN were 15.2gkg−1 and 1.6gkg−1 in the 0–5cm soil layer, and 1.5gkg−1 SOC and 0.4gkg−1 STN in the 60–100cm soil layer, respectively. The prediction precision for SOC, STN and BD all decreased with soil depth. Average SOC and STN stocks for 0–30cm were 3.1kgm−2 and 0.5kgm−2, respectively. For the top 1m, SOC and STN were 4.5kgm−2 and 0.9kgm−2, respectively. In total, the soils stored approximately 588Tg SOC and 128Tg STN within the top 1m. The soils under forest had the highest amount of carbon (356Tg) and nitrogen (58Tg) followed by agriculture and wetland that contributed 34% and 48% of the total stock, respectively. >91% of the total SOC and STN stocks were in Argosols and Cambosols. We adopted a digital soil mapping method to map the spatial distribution of SOC and STN stocks and predict their uncertainties. The estimation was validated with a 10-fold cross-validation procedure. The data and high-resolution maps from this study can be used for future soil carbon and nitrogen assessment and inventorying. •Constructing soil depth functions to describe vertical distribution of SOC, STN and BD•Deriving predictive models to map the soil depth functions across the study area•Evaluating the importance of environmental covariates in controlling the spatial patterns of SOC and STN•Producing the spatial distribution of SOC and STN storages in Liaoning Province of China