Modeling The Spatial And Temporal Dynamics Of Soil Organic Carbon In The Oinghai-Tibet Plateau Using Community Land Model

Soil organic carbon(SOC)is an important component of soil that helps to maintain soil fertility,determine yield,evaluate soil quality and ecological environment.Soil is the largest carbon pool in terrestrial ecosystems,stored about 1500 Pg C at a depth of 1 meter,which is three times that in the atmosphere.Thus,under the climate change,the decomposition and release of soil organic carbon will have a significant impact on the global warming.A huge amount of SOC is stored in the Qinghai-Tibet Plateau where is very sensitive to the climate change.SOC model is the only effective and widely used method for the large spatial-and time-scale study of SOC.The simplified model structure and complex parameters bring great uncertainty to the predictions.The Community Land Model(CLM)is the only one among all the earth system models in CMIP5 that considers the interaction between carbon and nitrogen in the ecosystem.It can also accurately simulate the soil carbon cycle process.Therefore,the CLM4.5 was employed to simulate the SOC in the Qinghai-Tibet Plateau.Based on the parameter optimization and accuracy verification,the changes of SOC stock and density in the Qinghai-Tibet Plateau were estimated for the past(1970-2010)and the future(2020-2050).The long-term simulation of SOC has scientific significance for determining the source or sink function of the soil,assessing the potential of soil sequestration and formulating reasonable policies to cope with global climate change.The main contents and results are listed as follows:(1)Sensitivity analysis and optimization of CLM4.5 model's parameters:The OAT(One at a time)method was used to analyze the important parameters of the decomposition module.Results showed that the soil organic carbon modelling was most sensitive to the temperature scalar parameter Q10,respiration fraction and distribution fraction.The range of optimization was determined by gathering literature,and the model accuracy was measured by the profile data collected during the Second National Soil Survey.After selecting the optimal parameters,the model accuracy was improved,and R2 increased from 0.34 to 0.51.(2)SOC storage and density distribution on the Qinghai-Tibet Plateau were simulated based on CLM4.5:The optimized CLM4.5 was used to simulate historical and future SOC in the top 30cm.Between 1970 and 2010,The range of SOC stock was 10.046-10.187 Pg,which was generally declined.A turning point appeared in 2006,after which SOC began to rise slightly.The SOC density increased from northwest to southeast.The future forcing data is from four climate models of CMIP5(RCP4.5).Between 2020 and 2050,the SOC stock changed from 10.561 to 10.892 Pg,showed a continuous rising trend,with a growth rate of 10.5 Tg·y-1.Compared the distribution of SOC density in 2050 with that in 2010,the whole plateau showed an increase,but the soil in southern broad-leaved forest showed a decrease.(3)Identify spatial and temporal control factors of soil organic carbon:Based on the simulations,the spatial and temporal control factors of soil organic carbon were analyzed.The SOC density increases with elevation,then decreases after 4500 m.Above 4500 m,the inhibition of temperature on the SOC decomposition is less than the significant decrease of carbon input from the litter carbon pool.SOC density was positively correlated with temperature and precipitation.Different vegetation types also affect SOC density.