Estimating Large Gradient Ground Deformation With Dinsar And Offset-Tracking

Geological disasters bring huge calamities to economic life, and most of them are caused by the deformation of the earth’s surface. It would result in irreversible surface displacement of great gradient, once landslides, earthquakes and other geological disasters of great hazards occurred. These deformation has become an element factor that affects the sustainable development of regional economy and society. Therefore, it is necessary conduct timely and comprehensive monitoring and analysis of these geological disasters, so as to provide some reference for the preventive measures and post-disaster management. Based on this, it’s of great vital to develop a scientific and effective new type of space geodesy positively.Differential Interferometric Synthetic Aperture Radar technology (the DInSAR), a new type of space geodesy developed from the end of the last century, is widely used in areas like volcanoes, earthquakes, landslides, ice, urban ground settlement with various advantages of high precision, wide monitoring range, fast and efficient, less limit from time and weather etc. While there exist defects of DInSAR for it’s vulnerable to time and space and may lose valid data, is limited by radar imaging geometry. Therefor DInSAR results could just reflect the relative deformation from the ground objects to the radar line of sight (LOS), and the true three-dimensional displacement of surface is impossible to restore with single platform SAR data alone.Offset-tracking technology is used to acquire two-dimension of range-azimuth surface deformation information with sub-pixel accuracy based on SAR image information and multi-level registration, which can overcome the insensitiveness of one-dimensional Los deformation of North-South in DInSAR. And this technology is not sensitive to the coherence of the SAR image, this, to some extent eases problem caused by coherence in information acquiring. But compared with DInSAR, Offset-tracking is not mature and its accuracy is highly reliant on the resolution of SAR image. Therefore, with the joint of the two to complement the insufficiency of each other, in the absence of external data, the real three-dimensional surface deformation recovery can be realized by merely using SAR images.Based on phase and intensity information of SAR image, combined with modern space geodesy:DInSAR and Offset-tracking, the synthetic aperture radar differential interferometric and pixel offset estimation, this paper studies on two representative large deformation gradient geological disasters:landslide-prone mountainous areas in Ya’an and Nepal earthquake with magnitude 7.8 in April,2015. The main work is summarized as the following points:1. With the use of the RadarSAT-211 scene SAR data and Offset-tracking technology, get information of sequential range-azimuth surface deformation in Ya’an and make detailed analysis on the monitoring landslide spots. And the extraction process of sequential deformation by Offset-tracking technology is improved during the study.2. Conduct detailed analysis on one-dimensional LOS deformation by DInSAR technology and two-dimensional range-azimuth surface displacement by Offset-tracking technology according to radar imaging geometry. Based on Least Square, the three-dimensional surface deformation restoration model is established with the combination of DInSAR and Offset-tracking techniques. In view of this, experiment grounded on the real seismic deformation field and satellite data is conducted under simulation data, and the experiment result shows that the reliability and usefulness of combination of multi-measurement three-dimensional model and multi-system, multi-angle measurement.3. Acquire co-seismic deformation field in studying area by selecting Sentinel-1A lifting data from the Nepal earthquake-prone area, and respectively, using the DInSAR method and Offset-tracking technology. Restore the true three-dimensional co-seismic deformation field in the lifting public area with combination of multi-angle measurements. And it turns out to prove that the absolute three-dimensional deformation field can be calculated according to the SAR image alone, without introducing the external data.