Study On Retrieval Of Ground Deformation Using Time-series InSAR And Atmospheric Phase Correction

The ground deformation has become one of the worst geological disasters in China. Therefore, many domestic researchs are faced with a difficult task, which is how to monitor the ground deformation effectively and timely, and provide technical support for relevant departments. As a new advanced means of monitoring ground deformation, DInSAR technique got rapid development in recent years. Nevertheless, it's difficult to meet current accuracy requirements for traditional DInSAR due to the influences of spatial decorrelation, temporal decorrelation and atmospheric effect. This paper aimed to overcome the limitations of DInSAR technique using time-series SAR data and water vapor data, and retrieve the information of ground deformation in high precision. Therefore, the following studies were made in this paper:(1) It studied the basic principles and data processing methods of InSAR and DInSAR, and then made a detailed analysis of the de-coherent sources in InSAR processing.(2) It studied the method of atmospheric correction in InSAR measuring. Atmospheric effects were caused mainly by water vapors, so it made a deep analysis and comparison among the atmospheric correction methods using satalite water vapors such as GPS, MODIS, MERIS, FY. Based on these, the MODIS/ASAR atmospheric correction method was proposed, and the validity and reliability of the method was proved through experiment.(3) It studied the method of selecting high coherent points. Currently, the selection method mainly based on amplitude discrete characteristics, coherence and phase stability, it analyzed the applying condition and deficiency of each method. Finally, phase stability analysis method was determined to select high coherent points in this paper by comparing and anslyzing the experimental result.(4) It studied the time-series InSAR technique to retrieve ground deformation. PS and coherent targets methods had advantages and disadvantages in retrieving deformation respectively, small baseline DInSAR technique which based on high coherent points and combined these two methods was put forward. Based on this method, an experiment over Taiyuan area in the time span of 2004-2007 was conducted, and the result of derivable deformation showed good agreement with the leveling data and meet the accuracy requirement in our country.