Sar Interferometry In Atmospheric Vertical Stratification Delay Correction

Synthetic aperture radar interferometry is an important means of monitoring regional land subsidence. Compared with the traditional methods of geodetic survey, such as Leveling, bedrock bench mark, GPS, etc., InSAR interferometric techniques can obtain high spatial resolution surface deformation details in the large-scale, which precision can reach to the centimeter level. These advantages help to reveal the geological phenomena, and explore the mechanism of surface deformation process.However, one of main limiting factors when using InSAR to extract slow surface deformation in large area is tropospheric delays. Traditional methods based on the interferometric phase evaluation of atmospheric delay only consider the atmospheric levels of turbulent delays, which did not consider the atmospheric stratified delay signal, orbital error and deformation phase. Current interferometry methods are sometimes difficult to separate atmospheric stratified delay signal which is combination of orbital error, deformation signal, the residual atmospheric stratified delay can reach to deformation amplitude in some regions and be easily to fuzzy deformation phase, and even impact the accuracy of SAR interferometry measurement. Based on this outstanding issue, We use Meteorological model and lessen regional and Gaussian multi-scale decomposition technique combined with the weighted least squares method to process the interferometric phase unwrapping in this study. With the deep analysis on scale features and frequency characteristics of deformation phase, atmospheric delay signal and other signals, We ultimately determine the appropriate Gaussian function and decomposition scale, and separate long-wavelength signal from the phase unwrapping. Finally, we evaluate the stratified atmospheric delays from the long-wavelength signal.We apply improved atmospheric stratified delay algorithm to the United States Las Vegas, and then using local GPS monitoring data to verify the InSAR time-series measurements.The main methods and conclusions of this study are the following four aspects:(1) We gain a long time series information of land subsidence deformation using PS/SBAS methods for SAR Interferometry. In the primary results, We identify Orbital errors, residual topographic errors, atmospheric delay. We then analyse the spatial distribution and frequency characteristics of noise, which provide a basis analysis of assessment methods for the error components. (2) We corrected Stratified atmosphere delay in sar interferometry according to meteorological model data, based on the statistics of amplitude and frequency characteristics in ZTD, ZHD,ZWD. Although the calibration results is poor, it does help us better understand the delay component characteristics, which then determin assessment steps of each component.(3) Atmospheric delay is an obvious problem in large area when we obtain slow surface deformation. We try to use lessen regional method and find the delay in small area is less affected. Also we validate the idea with GPS data.(4) Through image multi-scale decomposition, we determine the appropriate Gaussian parameters and decomposition scale. Finally, we successfully separate stratified atmosphere delay from the long-wavelength signal delay signal. In this study, we use the weighted least squares method to access the Stratified atmosphere delay. In addition to north-east mountain of the study area, this method successflly eliminate Stratified delay from north, south-west, south-eastmountain. The accuracy analysis results show that multi-scale decomposition is a feasible method to elimate Stratified atmosphere delay.