Study On The Spatiotemporal Evolution Characteristics Of Surface Soil Moisture On The Qinghai-Tibet Plateau Based On High-resolution Land Surface Simulatio

With the influence of global climate change,the weather and climate of the Qinghai-Tibet Plateau(QTP)are experiencing warming and wetting climate.In recent years,the average warming rate over the QTP is almost twice of the global mean warming rate.The water bodies in the plateau are mainly characterized by accelerated retreat of glaciers,significant expansion of lakes,increasing of glacier runoff,and the "Asian Water Tower" is experiencing an imbalanced transition from solid to liquid conditions.Soil moisture is an important variable for water cycle,which directly affects vegetation growth,hydrological process and landatmosphere interaction,and it is also an important indicator of the weather and climate prediction in the QTP.Therefore,it is necessary to study the variation characteristics of soil moisture in the QTP.It is a challenge to accurately describe the characteristics of soil moisture because few observation stations are over the QTP.Satellite remote sensing and land model simulation-based soil moisture products are seriously affected by the complex terrain,vegetation and snow cover of the QTP,so these two products have large uncertainties.In view of this,six sets of surface soil moisture products,including high resolution land surface soil moisture products,are evaluated by using the observation data of 140 stations on the QTP and the Triple Collocation(TC)method.On this basis,the variation characteristics and influence factors of soil moisture in the QTP were explored,and the main conclusions are as follows:(1)Compared with several sets of popular global soil moisture simulation products,high resolution CSSPv2 land surface products have the best applicability in QTP.In this paper,the applicability of the GLDAS-2.1 products(three sets,Noah,Catchment,VIC),ESA CCI satellite product,ERA5 reanalysis product and CSSPv2 model product in April-September during 2009-2017 are evaluated against in-situ observations.We find that the CSSPv2 product shows the best simulation,which has the highest correlation coefficient,the lowest root mean squared error and lower bias.The ESA CCI satellite products,ERA5 and Noah model products also have relatively good simulation results.The simulation results of Catchment and VIC products are poor.Except CSSPv2,other products all overestimate the soil moisture.The TC errors of ESA CCI,ERA5 and CSSPv2 are calculated,and the results show that the CSSPv2 has the lowest error,followed by ESA CCI and ERA5 has the highest error.The accuracy of ESA CCI and ERA5 products is higher in the eastern QTP,but lower in the western QTP.However,the accuracy of CSSPv2 products is higher in the western QTP and lower in the eastern QTP.(2)Based on the high-resolution and long-term simulation of CSSPv2,the variation characteristics of surface total soil water and surface soil ice content in the plateau’s warm seasons from 1979 to 2017 and their relationship with climate factors are revealed.The results showed that the total soil water is higher in the southeast and lower in the northwest.But there are spatial differences in the trend of total soil water,with increasing in the northwest and middle of the QTP,and decreasing in the southeast and southwest of the QTP.Through the correlation analysis with climate factors,it is shown that the total soil water in the northwest of the plateau may be more affected by the change of temperature,while the total soil water in the middle and southeast of the plateau may be more affected by the change of precipitation.The areas where the total soil water is positively related to the South Asian summer monsoon are mostly concentrated in the southern part of the QTP and the Hengduan Mountains.Through Ensemble Empirical Mode Decomposition(EEMD),the components obtained have significant periods such as 4 years,5-13 years and 7-20 years.Different from the total soil water,the soil ice content shows the distribution characteristics of higher in the northwest and lower in the southeast.In terms of the spatial variation of the trend,the soil ice content has an increasing trend in the northwest of the plateau and a decreasing trend in the east of the plateau.There is a significant negative correlation between soil ice content and air temperature.In the cooling area,the increase of precipitation is beneficial to the increase of soil ice content,but the correlation between soil ice content and precipitation is poor in the warming area.At the same time,it is also found that the soil ice content has no obvious change period,the EEMD’s component has a similar period to the total soil water.