Changes Of Permafrost Thermal State And Its Impact On Carbon Release In The Qilian Mountains

Snow cover has an important influence on the thermal state of the active layer and near-surface permafrost,and even carbon exchange between land and atmosphere.Presently,researches on the impact of snow cover on the thermal state of permafrost is mainly concentrated in the Arctic regions with thick snow cover,but few studies in the mid and low latitudes with thin snow cover.In this study,based on field observation data,we are trying to analyze the influence of snow cover on the thermal state of permafrost in the Qilian Mountains by using one dimensional heat transfer model with phase change.The aims of study are to explore the change of permafrost thermal states and its impact on carbon release in the Qilian Mountains in the different Shared Socioeconomic Pathways?SSPs?within Phase 6 of the Coupled Model Intercomparison Project?CMIP6?.In the paper,firstly we analyzed the influence of snow cover on soil surface energy balance,surface heat flux at a depth of 5 cm and the active layer thickness based on the monitoring data from two observation sites in the Eboling?EB?and Yeniugou regions?PT1?during 2012–2019.Secondly,we revised the one-dimensional heat transfer model with phase change by using the model of parametric volumetric unfrozen water content?VWC?.Then,the revised model was used to simulate the thermal states of permafrost by incorporating snow cover.Lastly,we simulated the soil thermal states in the Eboling Mountain in different SSPs within CMIP6 from 2015–2100 based on the revised one-dimensional heat transfer model with phase change,and discussed the characteristics of soil organic carbon release under the background of climate warming.The detailed results and main conclusions are as follows:From the observation data,snowfall mainly occurs during October–November and April–May in the Qilian Mountains.Different snow cover time and the variation trend of snow depth have different effects on soil temperatures.During the spring transitional period,snow cover reduces the temperature rise of the ground surface when snow depth is approximately 18 cm.From autumn to winter and spring,the snow insulation effect may prevail when snow depth is about 21 cm.From autumn to winter and spring,if snow depth continues to increase,the snow insulation effect will be more significant.While,if snow depth continues to decrease,its insulation effect will be weakened.The average active layer thickness at PT1 and EB monitoring sites were 162.1 and 74.2 cm from 2013 to 2018,respectively.The insulation effect plays a crucial role in the influence of snow cover on active layer thickness.Even in years with a larger air freezing index?FI_a?and a smaller air thawing index?TI_a?,the maximum snow cover height index?SCHI?will lead to the maximum of the active layer thickness in current or following year.Based on the soil temperatures of EB and PT1 monitoring sites,the active layer thickness of two sites have no obvious change trend.Due to the influence of snow insulation effect,the active layer thickness changed by 25 cm from 2015 to2017 at EB site.According to the field observation data of soil temperature and VWC,we established the new parametric VWC model,which effectively improved the simulation accuracy of soil VWC near zero curtain.We revised the one-dimensional heat transfer model with phase changes by using the parametric VWC model.Based on the revised model,simulated soil temperatures at EB site at the depths of 0.1,0.2,0.4,0.6 and 0.77m were in good agreement with the observation results.The simulated temperature was approximately 0.09°C higher than the average measured value.The simulated temperature was about 0.23°C lower than the borehole temperature at the depths of1¹6 m.Meanwhile,we simulated the sensitivity of soil temperatures at depths of0.05⁰.77 m to the changes of air temperature and snow depth.The results showed that the mean ground temperature variation were-0.07?-0.03?0.04?0.07?0.11 and 0.14°C under the patterns of air temperatures increasing-1,-0.5,0.5,1,1.5 and 2°C.And the mean ground temperature variation were-0.19?-0.11?-0.03?0.02?0.03 and 0.04°C under the patterns of snow depth coefficient of 0.25,0.5,0.75,1.25,1.5 and 1.75.The same range of air temperature rise or fall basically caused the same amplitude of mean ground temperature,while the decrease range of mean ground temperature caused by the decrease of snow depth was larger than the increase of snow depth.We simulated soil temperatures based on the revised one-dimensional heat transfer model.In SSP126,SSP245 and SSP585,the increasing rate at the depth of 0.1 m were0.07,0.18 and 0.22°C·?10a?^-1 from 2015 to 2100,respectively.While there was no significant warming trend at depths of 0.6,1.0 and 2.0 m in SSP126.And in SSP245and SSP585,the increasing rates were 0.08 and 0.09°C·?10a?^-1 for the depth of 0.6 m,0.07 and 0.09°C·?10a?^-1 for the depth of 1.0 m,and 0.06 and 0.08°C·?10a?^-11 for the depth of 2.0 m,respectively.With the increases of soil temperatures,the active layer thickness also increased gradually.In SSP126,SSP245 and SSP585,the active layer thickness deepened at the rates of 0.03,0.04 and 0.08 m·?10a?^-1,with the maximum active layer thickness of 1.30,1.35 and 1.90 m from 2015–2100.Based on carbon decomposition models,we simulated the initial carbon loss ratios were 0.25%,0.31%and 0.28%at the depth of 0.6,1.0 and 2.0 m.The loss ratio of organic carbon was the largest at the permafrost table,followed by permafrost layers.From 2015 to2100,the loss ratio of carbon at depths of 0.6,1.0 and 2.0 m showed no significant change trend in SSP126.While in SSP245 and SSP585,the increase rate was 0.002%·?10a?^-1 fot the depths of 0.6 m and 2.0 m,and 0.002%and 0.003%·?10a?^-11 for the depths of 1.0 m.The carbon loss leads to the gradual decrease of organic carbon density,and compared with the shallow active layer,the carbon loss amount and the proportion are the largest at the permafrost layers and bottom of the active layer.From 2015 to 2100,the organic carbon density of 0.05⁰.6 m,0.6¹.0 m,1.0².0 m and 0.05².0 m decreased with the reduction rates 15%,19%,20%and 18%.This paper preliminarily studied the influence of snow cover on permafrost thermal states,and simulated the changes of permafrost thermal state and the loss of soil organic carbon in the Qilian Mountains based on CMIP6 by using the revised one-dimensional heat transfer model.While under natural conditions,the change of permafrost thermal state and its impact on soil organic carbon is very complicated.It is necessary for the future study on the mechanisms of permafrost degradation with climate warming,and the impacts of hydrothermal process on carbon release.The results will provide basic data for the parameters,development and improvement of Earth System Model,in order to provide scientific basis for the understanding of the response and feedback of mountain permafrost regions to global climate change.