Revealing the scale- and location-specific controlling factors of soil organic carbon in Tibet

Abstract

Soil organic carbon (SOC) leads to a significant impact on global carbon (C) cycling and soil quality. Variations in SOC are controlled by vegetation, geomorphic, geological and climatic factors, but the dominant environmental differs. In the Qinghai -Tibet Plateau, which contains large amount of low-latitude permafrost, the impact of environmental factors for the variations of SOC may be different due to the unique and complicated geographical condition. In this study, the two-dimension empirical mode decomposition (2D-EMD) is applied to examine the variations of SOC at different scales and locations, and the correlations between SOC and environmental factors are explained. The spatial distribution of SOC in Tibet was decomposed into three intrinsic mode functions (IMFs) under different scales, with spatial variation scales of approximately 7 km, 109 km and 338 km, which represented the small, medium and large scale, respectively. The remaining residual represented the variation trend of SOC across Tibet. The correlations between SOC and environmental factors (elevation, radiation, evapotranspiration and temperature) are distinguished by the physiographic zone at small and medium scales. Temperature is weekly or nonsignificantly correlated to SOC in cold-dry western Tibet at large scale. Normalized difference vegetation index (NDVI) and precipitation influenced SOC mainly at small scales, while the effects of precipitation and evapotranspiration on the distribution of SOC were due to geomorphology and type of permafrost. The combined effect of climate on SOC was larger than other factors at large scale while factors refer to DEM, evapotranspiration, water erosion and NDVI accounted for more contribution at small scale. The results indicated that the environmental factors influence SOC under a combination of scale and location effect. These findings are of great significance for future studies in SOC dynamic modelling under the influence of natural changes and human activities.