| Known as the"Asian Water Tower",the Qinghai-Tibetan Plateau(QTP)is the headwater source regions of several large rivers,and its water conservation is extremely important.As a link between solute transport and energy transmission,soil moisture plays an important role in regional and global climate change through its impact on water cycles and ecosystems.Its alpine ecosystem is extremely fragile and highly sensitive to climate change.In the past few decades,global temperatures have risen significantly.And the permafrost degradation caused by climate warming has also significantly changed the soil moisture in the alpine regions.At the same time,climate warming will also affect the soil moisture in seasonally frozen soil areas.So what is the impact of climate warming on permafrost and seasonally frozen areas,is there any difference?Understanding these conditions helps to deepen people's understanding of the impact of climate change,and can also provide a reference for the formulation of climate change adaptation strategies.The QTP has poor natural conditions,so there is a lack of long-term and large-scale in-situ observations of soil moisture.Therefore,remote sensing products potentially provide useful data sets for the Earth System Models.Among them,the Climate Change Initiative(CCI)soil moisture products issued by the European Space Agency(ESA)have been widely used worldwide,and it has good applicability in the QTP.In order to understand the difference in the long-time series of soil moisture in the permafrost and seasonally frozen ground areas of the QTP,and to reveal the impact of climate change on soil water contents,this study selected the CCI remote sensing products,and the applicability was assessment,and the calibration using statistical method calibration were performed.The inter-annual changes of soil moisture contents at the regional scale were used to analyze the response of soil moisture to climate under different frozen soil conditions.In addition,its relationship with meteorological factors and NDVI(Normalized Difference Vegetation Index)were also examined.The main findings are as follows:(1)All the three CCI products can capture temporal and spatial distribution of soil moisture in the growing season on the QTP.Among them,the CCI combined products have the highest correlation in the three networks,and the error is relatively small.In semi-arid areas dominated by alpine meadows,the CCI combined product has the highest accuracy in retrieving soil moisture,but it underestimated in the area with dense vegetation(Maqu)and overestimated in the barren ground(Shiquanhe).From the spatial distribution,the active product overestimates the soil moisture on the QTP,and the passive product shows that there are great fluctuations in soil moisture.The CCI combined products are closest to the actual distribution of soil moisture in the QTP.Linear regression analysis is an effective correction method to improve the precision of soil moisture inversion of CCI products.Based on the measured soil moisture data,a linear regression analysis of the CCI combined products was performed to establish calibration models.The calibrated products shows that the soil moisture in the QTP gradually decreases from southeast to northwest,which better reflects the soil moisture distribution in the regional distribution of the QTP.(2)Based on the linear regression method,we calibrated the CCI active and passive combination products at various observation points under different frozen soil conditions,and further analyzed the spatial-temporal changes of soil moisture from1986 to 2016.The results show that:(1)From the time trend of 1986-2016,the soil moisture at the permafrost sites largely remained steady,fluctuating from 0.050 to0.055 m~3/m~3 during 1986-2016.Instead,great changes and more frequent change-points in soil moisture were observed at the permafrost-free sites,with a variation from 0.066 to 0.117 m~3/m~3.Among them,the soil moisture in Naqu and Liangdaohe sites in the semi-arid climate area showed a significant decrease in the past 30 years,and the Shiquanhe in the arid area and Maqu in the humid area showed a significant increase.(2)In terms of spatial distribution from 1986 to 2016,the soil moisture changes in the permafrost regions of the central QTP are relatively stable,and the seasonally frozen ground regions are more violent,the soil in semi-arid areas becomes drier,while the soil in arid and humid areas has a tendency to wet.The results show that the existence of permafrost can stabilize the surface soil moisture.In addition,the semi-arid areas have a tendency to desertification in recent decades.(3)The correlation analysis of soil moisture with meteorological factors and vegetation coverage of the QTP from 1986 to 2016 was performed.It was found that in terms of spatial distribution,the precipitation and NDVI in QTP were highly correlated with soil moisture while the correlation between air temperature and soil moisture was relatively low.On the time trend of the observation sites of permafrost and seasonally frozen ground from 1986 to 2016,the correlation coefficients between precipitation and soil moisture at seasonally frozen ground observation sites were higher,up to 0.7,and the permafrost area is relatively low,between 0.4 and 0.5.The correlation coefficients between temperature and soil moisture were generally low,and most of them were negative.The correlation between NDVI and soil moisture in seasonally frozen ground is higher,and the permafrost area is relatively low,and the temperature has a more significant effect on NDVI than precipitation.According to the relationships among environmental factors,precipitation,air temperature and NDVI all have different effects on soil moisture.Among them,the precipitation and NDVI were the most important factors in predicting soil moisture contents.This was especially true in seasonally frozen ground areas.The relationship between air temperature and soil moisture showed great heterogeneity.In this study,the accuracy of CCI soil moisture products in different climate networks on the QTP was evaluated,and it was found that the CCI active-passive combination products had the highest accuracy in retrieving soil moisture in the region.Based on the measured data,linear regression calibration was made on the combined products to obtain products that can better reflect the changes in soil moisture.According to the calibrated products,the analysis found that the presence of permafrost on the QTP can stabilize the surface soil moisture,and the soil moisture in the seasonally frozen areas change more drastically.Among them,precipitation and NDVI are significantly related to soil moisture in seasonally frozen soil areas,but have a lower impact on permafrost areas,and temperature generally has a lower impact on soil moisture.The research results have improved people's understanding of the changes in soil moisture in the QTP under the influence of climate warming,and also provided a reference for the application of remote sensing soil moisture products in surface models. |
PDF Full Text Request