Vertical distribution of soil organic carbon and influential factors along vegetation transect on west slope of Mount Segrila, southeastern Tibet

Objectives The montane ecosystems located in the southeastern part of Tibetan Plateau possess great potential of soil carbon sequestration. It is therefore necessary to investigate the variations of soil organic carbon (SOC) with different ecosystems and the influential factors, which can help to improve understanding of soil carbon cycling and the regional balance between carbon source and sink on the Qinghai-Tibetan Plateau.

Methods The present study conducted an intensive soil sampling from 3000 m to 4600 m above sea level along the west slope of Mount Segrila. Vertical variations in SOC storage under different vegetation types along the elevation gradient were investigated, and the effects of various environmental factors were analyzed.

Results We found that SOC concentration in 0–5 cm surface soil increased with the increasing of elevation. The averaged SOC content for the four vegetation types was alpine meadow (8.31% ± 0.77%) > dark coniferous forest (7.20% ± 0.90%) > alpine shrub meadow (6.74% ± 0.80%) > coniferous and broad-leaved mixed forest (3.88% ± 0.46%). SOC contents of 5–10 cm, 10–15 cm, 15–20 cm, 20–30 cm, 30–40 cm and 40–60 cm first increased and then decreased with elevation, and the averaged SOC content was dark coniferous forest > alpine shrub meadow > alpine meadow > coniferous and broad-leaved mixed forest. SOC concentration decreased significantly with the increasing of depths along soil profile. The vertical distribution of SOC content showed a surface-accumulated pattern under alpine meadow and alpine shrub meadow, but a gradually decreasing pattern under dark coniferous and coniferous and broad-leaved mixed forests. The variations in SOC stock of 0–20 cm, 20–40 cm and 40–60 cm with elevation also presented a unimodal pattern. The SOC stock of surface 0–20 cm in alpine meadow (C 95.66 ± 4.81 t/hm²) was the highest among all four vegetation types on west slope of Mount Segrila. Dark coniferous forests had the highest SOC stock in 20–40 cm and 40–60 cm, and the total SOC stock of 0–60 cm (C 199.14 ± 11.10 t/hm²) was significantly higher compared to other vegetation types. The SOC stock in coniferous and broad-leaved mixed forests was significantly lower at each profile depth, making the total SOC stock (C 111.45 ± 10.30 t/hm²) the lowest among the four vegetation types. The variation in SOC stock was negatively correlated with mean annual temperature (MAT) and litter C/N, but positively correlated with elevation, mean annual precipitation (MAP) and soil moisture. Results of stepwise regression analysis suggested that soil moisture was the key influential factors controlling the variations of SOC in each soil layer and the whole soil profile. Random forest models explained 50.32%–65.82% of variation in SOC. Soil moisture was the most important predictor of variations in SOC storage in the topsoil. MAT, MAP and litter quality were significant predictors of SOC variations in both topsoil and subsoil. The relative importance of vegetation type in predicting SOC increased with soil depth.

Conclusions The storage of SOC along an elevation gradient of Mount Segrila varies significantly with vegetation type and depth of soil profile. Environmental factors such as soil moisture are the key factors that control the variations of SOC in the topsoil, whereas vegetation types significantly contribute to the prediction of changes in SOC in deep layer of soil profile.