Estimates and determinants of soil organic carbon and total nitrogen stocks up to 5 m depth across a long transect on the Loess Plateau of China

摘要

Purpose Carbon (C) and nitrogen (N) soil profiles are influenced by several environmental factors. However, the contents and distributions of these elements in deep soils and sediments are largely underestimated. We aimed to estimate the stocks, patterns, and driving factors of deep soil C and N on the Chinese Loess Plateau (CLP) after large-scale ecological restoration projects. Materials and methods Soil organic carbon (SOC) and total nitrogen (TN) contents in different soil layers were measured directly at 86 sites along a regional transect across the CLP. Results and discussion SOC and TN contents ranged from 1.97 to 6.83 g C kg(-1) and 0.24 to 0.72 g N kg(-1), respectively, as the soil depth varied from 0 to 5 m. The mean contents and degrees of variability of SOC and TN decreased with the increasing of soil depth. Based on SOC and TN content patterns, we divided the 0-5-m soil profile into layers of 0-0.1, 0.1-0.4, 0.4-1, and 1-5 m. In the 1-5-m soil layer, approximately 70% of the mean SOC stock (14.97 kg C m(-2)) and 71% of the mean TN stock (1.75 kg N m(-2)) were stored. A partial least square regression model showed satisfactory predictive performance, with R-2 and Q(2) 0.5 for SOC and TN stocks in the 0.1-0.4-m soil layer. Climatic factors, soil water content (SWC), and field capacity strongly affected SOC and TN stocks in all soil layers. The significance of clay content, SWC, and normalized difference vegetation index varied with soil depth and became the strongest in the 1-5-m soil layer. The highest proportion of SOC and TN stocks for this soil layer were found in grassland and in 450-550 mm rainfall zone. Conclusion Considerable amounts of SOC and TN stocks were stored in the 1-5-m-deep soils. Land-use types and rainfall zones can significantly affect the SOC and TN stocks. This information is helpful for identifying local land uses associated with high SOC and TN stocks and is essential for accurately estimating and predicting regional C and N stocks and cycles in terrestrial ecosystems.