永久散射体网络建模及地表形变与大气影响分析

Modeling with Permanent-Scatterer Network and Analysis of Ground Deformation and Atmospheric Influence

为了更有效、更可靠地探测地表线性和非线性形变,估计SAR成像时的大气影响,本文提出了基于优化永久散射体网络的邻域差分建模方法,详细讨论了PS组网、网络优化以及估计地表形变与大气影响等参数的数学模型与计算过程,并基于Matlab环境编程实现了这一方法。最后以上海陆家嘴地区为例,选取26幅ERS-1/2卫星SAR影像进行实验,探测出实验区在1992—2002年间的线性形变速率、非线性形变量、总的区域形变场及各成像时刻的大气影响值,与地面观测结果具有较高的一致性。实验结果表明,本文提出的PS邻域差分建模方法是可行的,相应的计算程序是可靠的。

This paper proposes an approach of modeling with a permanent-scatterer(PS) network(MPSN) based on the neighborhood differential idea,aiming at effectively and reliably detecting both linear and nonlinear ground deformations and estimating atmospheric effects by differential synthetic aperture radar interferometry with permanent scatterers(PS-DInSAR).Firstly,the triangular irregular network(TIN) of PSs is generated based on the PSs' image coordinates.Deformations are correlative both temporally and spatially,but atmospheric delays are only correlative spatially.Furthermore,the shorter the intervals of space and time series are,the more the correlations of deformations and atmospheric delays will strengthen.So the phase difference of neighboring PSs can depress the atmospheric influence.Then the neighboring differential phase model(NDPM) is constructed based on the neighborhood of PSs among the TIN,and the corresponding set of neighboring differential phases is generated.What are unknown in NDPM are residual elevation difference Δh_ij and mean velocity difference Δv_ij between two neighboring PSs i and j.Secondly,the TIN is twice optimized according to distance(<1km) and multi-image phase coherence(γ) between two neighboring PSs,and the neighboring differential phases hardly influenced by atmosphere delays are selected to resolve the NDPM by using a simple periodogram,i.e.to estimate the residual elevation difference Δh_ij and the mean velocity difference Δv_ij between two neighboring PSs i and j.After the estimation of Δh_ij and Δv_ij,the elevation Δh and the mean velocity v of every PS are respectively calculated through network adjustment based on the optimized TIN.Finally,the PS's phase residue Δ_res can be calculated.Then the temporal and spatial correlations of deformations and atmospheric delays are taken into account again,and the appropriate filters are selected to discriminate the following phases from phase residue Δ_res,(1)PS's atmospheric phases in differential interferograms,(2)PS's atmospheric phases in SAR images,(3)PS's no-linear deformation phases.A set of computer programs are developed in Matlab environment to implement the above approach.Lujazui in Shanghai is selected as the study site,and the experiments are performed using 26 SAR images taken by ERS-1/2 from 1992 to 2002 to estimate linear motion rate,nonlinear motion magnitude,total deformation distribution,and atmospheric delay for each SAR image.The obtained measurements are very consistent with the ground-based measurements.Based on modeling with PSs' TIN,the neighboring differential phases hardly influenced by atmosphere delays are selected to resolve the NDPM.The results are more reliable and precise.The experiment results show that the proposed modeling and methods are viable,and the computation software is reliable.