The Study On Interference Conditions Analysis For InSAR In Deformation Monitoring

Since the1960s, the ability of obtaining high-precision spatial information by SAR interferometry (InSAR) has been proven. Based on InSAR technology, Differential-InSAR (D-InSAR) has gradually developed as one of the most powerful tools for terrain elevation and surface deformation measurement. As the surface deformation is threatening the human survival environment, surface deformation monitoring has been attracted the attention of governments. The conventional measures, such as GPS and leveling network, are limited in the scale of area, time, weather, cost, accuracy and so on. InSAR/D-InSAR can monitor a large area with low cost and high efficiency, and improve the accuracy from centimeter to millimeter.However, the technology is restricted by the atmospheric effects and de-correlation factors during D-InSAR processing, and will affect the accuracy of monitoring results and its promotion. Although there are some achievements on the removal of atmospheric effects and other research areas, D-InSAR is still restricted by some factors. Therefore, based on the current status of few studies in the analysis of the D-InSAR constraint conditions, we make an intensive study of several key factors which affect D-InSAR deformation monitoring. This study further extends conditions for InSAR measurements, which also has important reference value and application significance for understanding the interferometry mechanism, establishing improved InSAR numerical model and helping researchers to determine the most appropriate data by different deformation characteristics(1) We analyze the range of critical slope for each sensor and the blind areas monitored by SAR are estimated. We also volume scattering component must be considered while using coherence decomposition technique in vegetation area and the model is modified. We illuminate the difference between deformation gradient and phase gradient and their function is proposed. Meanwhile, angle of incidence is introduced to the function of deformation gradient.(2) We use a series of representative data to validate the conclusion. The simulation data is used to validate the constraints of deformation gradient, SAR geometric imaging characteristics and all limited conditions in practical application. We also use ENVISAT ASAR and TerraSAR-X data in Three Gorges area to anylze coherence.(3) We present a novel approach for accurate temporal coherence decorrelation analysis of InSAR echo signal. This method is to integrate several data processing methods for parameter estimation of multi-temporal datasets, and extract the precise temporal coherence decorrelation component. The proposed algorithm is divided into three steps:Firstly, using the modified Intensity-Driven Adaptive Neighborhood (IDAN) algorithm to estimate interferometric phase; then using least squares fitting to remove deviation; finally, obtaining accurate temporal coherence decorrelation component with coherence decomposition technique by separating approximate unbiased coherence.(4) We take the experimental area of Los Angeles as an example in this paper, and a comparative study of the new method and the existing methods is analyzed as well. The result shows that the new fusion algorithm is able to obtain more reliable and accurate temporal coherence decorrelation component. Meanwhile, the characteristics for the new method of non-threshold and almost adaptive are proved. This study will improve the inversion of geophysical parameters relating to temporal coherence, such as surface classification, change detection and biomass recovery. Moreover, this research is also conducive to monitor the periodical changes of surface and climatic and environmental changes over time.(5) We demonstrate the unique capabilities of ALOS PALSAR data for monitoring ice flow and deformation using D-InSAR and Offset-tracking technique. The test area of Grove Mountains in East Antarctica is selected, and we compare the velocities derived in this paper with NASA/JPL results and GPS points. The result shows using L-band data in fast motion area is very efficient.