Permafrost Surface Deformation Monitoring And Active Layer Thickness Inversion Based On InSAR Technology

The Qinghai-Tibet Plateau is the world’s largest permafrost area at high altitude and low-middle latitudes.The continuous warming of the Qinghai-Tibet Plateau has caused the degradation of permafrost,which is manifested by the intensification of surface deformation,increasing of active layer thickness（ALT）,posing a threat to the ecological environment and engineering facilities in the Qinghai-Tibet Plateau region.Therefore,surface deformation monitoring and active layer thickness inversion studies in permafrost areas provide important decision bases for local ecological environment and geological disaster prevention.Traditional surface deformation monitoring methods can only obtain single-point information and are costly.With the development of interferometric synthertic aperture radar technology（InSAR）,deformation information can be obtained from a macroscopic perspective.In the permafrost zone,the surface deformation is mainly due to the periodic changes caused by the freeze-thaw process of the ALT,which require long time series monitoring.The multitemporal interferometric synthetic aperture radar（MT-InSAR）technology can achieve high accuracy,large range and long time series monitoring.Therefore,this paper is based on SBAS-InSAR technology to carry out research on surface deformation monitoring and ALT inversion studies in the permafrost zone of the upper reaches of the Heihe River basin in Qilian Mountains.The main work and conclusions of this paper are as follows.（1）The surface deformation is a combination of long-term trends and seasonal cyclic variations in the permafrost zone.The InSAR deformation results were decomposed using the model to obtain the vertical annual average deformation rate reflecting the long-term variation and the seasonal deformation magnitude reflecting the periodic variation.The results show that the annual average vertical deformation rate is-40～30mm/a and the seasonal deformation amplitude is 10～60mm in the study area.Analyzing the influence of environmental factors on surface deformation,it is found that the annual average vertical deformation rate and seasonal deformation amplitude are decreasing with increasing elevation and slope.The control of temperature on deformation is to influence the time of freezing and thawing,and then the depth of thawing.Differences in vegetation cover lead to different soil moisture,and the seasonal deformation is larger in areas with high soil moisture.Hydraulic erosion mainly occurs in the vicinity of the alluvial gully,and the surface deformation process shows a stepped pattern.Water accumulation acts in the downstream of rivers,and has obvious seasonal characteristics.（2）In order to improve the accuracy and resolution of SMAP data,which are biased dry in the study area,the ANN method was used to downscale the SMAP data.The mean RMSE is 0.05,which is 0.09 improvement in RMSE compared to the original SMAP L4 data accuracy.The deep soil moisture is not a stable value,but changes with increasing depth of soil moisture.Therefore,based on the Exp F method,the optimum T value with regional homogenization is used to achieve the estimation of deep soil moisture.The accuracy of the results gradually decreases with increasing depth.The accuracy verification RMSE of SMAP L4 deep data is 0.2.The accuracy of soil moisture results at the same depth obtained,RMSE is 0.16.The estimation results of Exp F have smaller errors and much higher spatial resolution than SMAP L4 data,which provide important data for the subsequent ALT inversion.（3）The surface deformation in the permafrost zone is influenced by temperature and soil moisture.Therefore,two models are used to carry out the ALT inversion and compare the applicability of the two models in the Yeniugou area of Qilian Mountains.The temperature inversion method includes two data one is the maximum thawing time of permafrost using InSAR deformation results,and the other is the maximum regional temperature.The results of ALT were obtained using a regionally homogeneous parameter inversion.The mean error of the inversion results is 1.09 m,the median is 0.86 m,and the standard deviation is 0.82 m.The soil moisture inversion method assumes that the vertical surface deformation is caused by the volume change of the water ice phase change.The inversion is obtained by using the seasonal deformation magnitude and soil moisture estimation results.The mean error of the inversion results is 0.07 m,the median is 0.28 m,and the standard deviation is 0.44 m.A comparative analysis of the results of the two models shows that the inversion method based on soil moisture has higher accuracy.It is more applicable to the study area.The spatial distribution of the ALT shows a high southeast and low northwest,ranging from 0.3 to 6.5m,with an average value of 2.47 m.In this study,we investigate the influence of environmental factors on surface deformation in permafrost areas based on the InSAR technique.Based on the ANN and Exp F method,we obtain the high-resolution soil moisture data.Combined with the ALT inversion model of InSAR,we achieve high-resolution active layer thickness inversion.The results of these studies can help to understand the response of permafrost to climate,and are useful for engineering construction.