Atmospheric Correction For InSAR And Its Application In Mapping Ground Motion Due To Interseismic Strain Accumulation

InSAR has proved to be a powerful tool for measuring crustal deformation,and be capable of providing displacement information with high spatial resolution and wide coverage.InSAR has been successfully applied in various geophysical applications,e.g.coseismic displacement extraction.However,detection of interseismic activity of faults with InSAR methods still faces challenges,due to slow movement of the fault with a trivial rate which could be easily covered by the existing phase errors.Atmospheric artifacts and orbital errors have been considered as the main error sources in extracting interseismic deformation signals.Thus,we will concentrate on the evaluation and analysis of different error correction methods,so as to find the optimal correction methods,and then to reconstruct interseismic rate map more accurately.The accurate rate map of interseismic deformation is fundamental information to carry out more reliable understanding of active faults and seismic hazard assessment.In this study,the stacking InSAR technique combined with an optimal atmospheric and orbital correction strategy based on external atmospheric data,short-time-baseline ASAR interferograms and simulations is developed to extract the fine surface deformation velocity field.We applied it to derive the interseismic deformation field of Haiyuan Fault System with high accuracy and high resolution,which is located in the northeast margin of Tibetan Plateau.A fine regional 3D crustal movement field is estimated by inverting jointly the InSAR LOS ratemap and GPS velocities.The assessment of the earthquake hazard will be carried out comprehensively by combining the surface observation and inverting parameter in the deep.The main works can be summarized as follow:1.Based on the analysis on characteristics of atmospheric and orbital errors in interferogram,we found the difficulty to separate them from deformation signal is due to the phase coupling beween long-wavelength interseismic signal and the atmospheric and orbital component with similar frequency.Thus,we choose to use indepedent external meteorological data to eliminate atmospheric phase,and assess the pros and cons of different atmospheric and orbital correction methods with short-temporal-baseline interferograms and simulated datasets.Based on the short-temporal-baseline interferograms,we evaluate three external atmospheric data(MERIS,ERA-I,WRF),and our result shows that MERIS and ERA-I is more suitable to be used to correct the atmospheric effect in ASAR interferograms than WRF in Haiyuan Fault System.Based on the simulated experiment,we evaluate the reliability of two orbital calibrating methods,i.e.‘fitting orbital model by using far-field or the whole-field interferometric phase on interferograms' and ‘estimating the orbital model constraining by a prior reference deformation model',and our results show that(1)the method based on information of the far field is prior to that of the whole field,(2)using the prior deformation model or not doesn't matter a lot on the basis of the far-field information.2.Based on the research on InSAR error correction above,we propose a new method to derive the accurate interseismic rate map – the stacking InSAR technique combined with an optimal atmospheric and orbital correction strategy.This method is applied to derive the interseismic rate map of the Haiyuan fault system.The proposed method has been validated by comparing the InSAR rate maps from two adjacent tracks,different correction methods,and with GPS observations.In addition,to reduce the effect of systematic error from plane coordinate system,we realize the algorithm of calculating continuous deformation and strain rate field in spherical coordinate system,which performs better than previous methods in the simulated experiment.3.Following the method stated above,we make use of 6 tracks data from Envisat ASAR to extract the accurate interseismic rate map of the Haiyuan Fault System.The profiles across the fault and locking depth inverted reveal the existence of shallow creep at the Laohushan Fault segment.A fine rate map and strain rate field with high spatial resolution is achieved by inverting jointly the InSAR with GPS observations.The result shows two zones with unusual strain rate:(1)a low strain rate zone located at the Laohushan Fault with shallow creep,in between two high strain zones – Tianzhu seismic gap and 1920 Haiyuan earthquake rupture segment,(2)a high strain rate zone located in the connection area of Xihuashan-Nanhuashan Fault and Liupanshan Fault,where Haiyuan Fault transfers from strike-slip fault to thrust fault.