TS-InSAR Atmospheric Delay Estimating Based On Terrain 3D Information And Its Application On Urban Subsidence Monitoring

Land subsidence is a geological phenomenon that ground surface sink to a lower level caused by nature or human activities.Though land subsides slowly in urban area,the influence of land subsidence is large and the duration is long,which will make a long-term adverse effects on urban economic development and lives of residents.Differential synthetic aperture radar interferometry(DInSAR)has been widely used in the field of subsidence monitoring with advantage of pantoscopic observing,high efficiency and high precision.Based on DInSAR,Time Series In SAR(TS-InSAR)techniques were proposed.With TS-InSAR techniques developing,not only the problems of temporal and spatial decorrelation are overcome,but also the effects of atmospheric delay and noise are reduced.Spatio-temporal filtering method can only have an effect on turbulence delay,because of low temporal correlation.But the vertically stratified tropospheric delay has high temporal correlation,atmospheric delay can't be eliminated if only spatio-temporal filtering method is used.As the research object,vertically stratified tropospheric delay is estimated through TSInSAR atmospheric delay modeling based on terrain 3D information of persistent scatterers(PS)points.In order to solve the problem that there are interference factors in the process of modeling,an optimized atmospheric delay estimating method based on resampling and Gaussian filter is proposed to achieve a robust and reliable vertically stratified tropospheric delay estimation.The details are as follows:(1)Under the condition that external data for atmospheric correction is insufficient or with low quality,atmospheric delay is modelled based on terrain 3D information of PS points for solving the problem that spatio-temporal filtering method in TS-InSAR can't alleviate vertically stratified tropospheric delay.(2)In order to solve the problem that there is a deviation in the vertically stratified tropospheric delay estimation because of interference factors,an optimized atmospheric delay estimating method is proposed.The main idea is to optimize observation value by using the methods of resampling,Gaussian filtering and differencing.There are two experiments are designed to compare the results of the directly estimating method and the optimized method.In one experiment elevation distribution is more concentrated,and the other one elevation is evenly distributed.The results show that,no matter what the elevation distribution is in the monitoring region,the optimized method can weaken the influence of the interference factors.Compared with the direct method,the optimized method more likely to present the average level of vertically stratified tropospheric delay in this region.(3)12 TerraSAR-X images acquired from 2010 to 2015 and 14 Sentinel-1A images acquired from 2016 to 2017 are used as the experimental data.The surface subsidence of Chengdu central urban area is monitored by using TS-InSAR technique,and the direct method and optimized method are applied to estimate the vertically stratified tropospheric delay.The experiment based on the stable region shows that optimized method proved to be effective and reliable in the application of TS-InSAR vertically stratified tropospheric delay correction.The monitoring result shows that the mean subsidence velocity of Chengdu central urban area is within 6 mm/yr,which proves to be nearly no subsidence.In particular,the monitoring status of area within the second ring road is stable in the past seven years.