Time Series InSAR Analysis Over The Three Gorges Region: Techniques And Applications

As a coherent imaging sensor, Synthetic Aperture Radar (SAR) actively transmits microwave signals to the earth, and receive, record the echo in a synthetic way. The image of SAR is not affect by ground sun illustration or clouds. Along with the accumulation of the space-borne SAR images, how to fully exploit the information implicated in the time series SAR images is well concerned. Recently, the research has been focused on the deformation measuring dependent on the artificial ground objects in urban areas, namely, Permanent Scaterer SAR Interferometry (PS-InSAR) technique. Nevertheless, for more cases, the stable point-like targets in the area of our interest are very few. This drawback has limited the applications of time series InSAR data in many geologic applications. Thus, in this dissertation, based on the deeply understanding of SAR interferometry principles, the error and coherence model of time series InSAR analysis over the complicated topographic region is studied. The work presented solves the bottleneck problem of time series InSAR analysis in difficult areas and expands its application to the extra-urban regions. More than one hundred InSAR images have been processed to measure and analyze the topography, land use, and deformation over the Three Gorges Region. The results show the unique advantages of time series InSAR data and allow us offering the basic geology information and decision aids in geologic hazards monitoring. The work that has been studied includes:1. By modeling the errors in the InSAR technique, a method is proposed to generalize a reliable Digital Elevation Model (DEM) from short normal baseline interferograms with the facilitations from external topography data. The accuracy of the resultant DEM is improved without any ground control points.2. The principle of InSAR coherence is studied to decompose the observed coherence into temporal and spatial ones. An approximate temporal coherence map can be extracted and be seen as an important data source for land use mapping. In the meanwhile, the point-like targets can be detected from only two InSAR images.3. Based on the theory and model research, the characteristics of the current time series InSAR analysis techniques are overviewed. By maximizing the target temporal coherence factor, a method to extract topography and deformation from partially coherent targets is proposed, namely, Quasi PS-InSAR (QPS-InSAR). The work allows us expanding the time series InSAR data applications in the fields of landslide monitoring, fault deformation measurement etc., and gives possibilities of deformation monitoring over the Three Gorges Region.4. With two different time series InSAR analysis methods, we are able to monitor the slow landslides in Badong. We found two subsidence areas around the new built Badong town by both techniques. The results from the two independent techniques are cross validated. However, since the QPS technique takes advantage of also partially coherent targets, much higher spatial density measurements can be obtained. The results can be seen as the support for city emigration decision5. We carried out the QPS and PS analysis over the Three Gorges Dam site. From the QPS results, we found that there exists no visible temporal subsidence of the Dam from 2003 to 2008. By analyzing the measured deformation trends, we presumed that the dam declined slightly on account of the upriver water pressure over the riverbed crust. Then from the PS results, we found that the time series deformation of the dam strongly related to different water levels. A seasonal deformation pattern can be observed as well. The results fit well with the published Three Gorges Dam deformation measured with conventional survey methods.