InSAR数据中的轨道和大气误差研究一直是该领域的难点，尤其对于震间和震后产生的微小长波地表形变信号的测量，很容易被误差淹没，如何精确地描述、改正和评价这两种误差的影响，对于减小其对地球物理参数的影响以及对观测结果作出可靠性评价是非常重要的。本项目从大量历史SAR存档数据本身入手去深入研究影像获取时刻的轨道、大气误差分布规律。首先，基于大量存档影像反演有序的时间序列干涉图，并建立一种新的物理轨道模型，研究利用同轨道的相邻非形变区域内时间序列干涉图或者形变区域内有限的CGPS观测数据（>6个）求解轨道模型；采用空间域低通滤波紧接时间域高通滤波实现大气和轨道误差的分离，对轨道误差和大气误差的尺度特性和空间分布进行精确的数学描述和统计分析，稳健地评价其对干涉形变测量产品的贡献。最后通过海原断裂带的实例应用，建立一套利用时序雷达影像获取活动断层震间微小形变速率场方法及相应误差水平的评价机制。

The research on orbital and atmospheric uncertainties from InSAR data has always been a challenge in this field, especially for the measurement of subtle, long-wavelength deformation (>50km), such as inter-seismic and post-seismic deformation, which can easily be confused with errors. How to accurately characterize, correct and assess their effects, is critical for decreasing the effect on geophysical parameters and interpreting reliably subtle InSAR deformation signals. this project exploits for the first time a large amount of SAR archives in an effort to study deeply orbital and atmospheric error distributions and characterize the technical limitations of the InSAR technique. First of all, a set of phase-unwrapped interferograms, with respect to the first acquisition, are subsequently inverted from a sequence of SAR acquisition, then we develop a physical orbital model, and resolve it by using interferometric time-series phase data of non-deforming area adjacent to the area of interest, or CGPS data (>6) in the presence of deformation. The atmospheric components are separated from orbital ones by applying a low-pass spatial filter followed by a high-pass temporal filter to unwrapped interferograms. We characterize mathematically and analyze statistically scale characteristics and spatial distribution of orbital and atmospheric errors and robustly assess their contributions in InSAR deformation products. Finally through applying to Haiyuan fault zone, we eventually establish a set of methods to acquire subtle deformation in active fault area by using archived SAR images and corresponding error evaluation mechanism.