Acta Scientiarum Naturalium Universitatis Pekinensis ›› 2021, Vol. 57 ›› Issue (2): 241-249.DOI: 10.13209/j.0479-8023.2021.002

Previous Articles     Next Articles

InSAR Time Series Analysis Technique Combined with Sequential Adjustment Method for Monitoring of Surface Deformation

WANG Hui, ZENG Qiming, JIAO Jian, CHEN Jiwei   

  1. Institute of Remote Sensing and Geographical Information System, School of Earth and Space Sciences, Peking University, Beijing 100871
  • Received:2020-02-24 Revised:2020-05-28 Online:2021-03-20 Published:2021-03-20
  • Contact: ZENG Qiming, E-mail: qmzeng(at)pku.edu.cn

结合序贯平差方法监测地表形变的InSAR时序分析技术

王辉, 曾琪明, 焦健, 陈继伟   

  1. 北京大学遥感与地理信息系统研究所, 北京大学地球与空间科学学院, 北京 100871
  • 通讯作者: 曾琪明, E-mail: qmzeng(at)pku.edu.cn
  • 基金资助:
    国家自然科学基金(41571337)资助

Abstract:

Based on the SAR satellite imagery with short revisit periods, this paper proposes a novel SBASInSAR time series analysis technique for efficient and continuous monitoring of surface deformation in the study area. First, standard interferometric processing is carryed out with the existing SAR image set in the study area to obtain the interference map. Then the atmospheric and orbital errors are removed, and the traditional SBAS (small baseline subset) method is used to obtain the surface deformation. On this basis, when a new SAR image is added, the progressive SBAS is adopted to invert the surface deformation at the new moment. The progressive SBAS method integrates the idea of sequential adjustment based on the obtained results derived from existing data set, and combines the newly acquired data to implement incremental calculations, finally achieves the equivalent effect of overall processing. Compared with the traditional SBAS method which needs to resolve all the calculations every time when a new image is added, the progressive SBAS method can reduce redundant operations and improve computing efficiency. The experiment proves that based on the Sentinel-1 satellite SAR data acquired in the Yellow River Delta from May 2018 to August 2016, the surface deformations retrieved by the progressive SBAS method are almost the same as the results of the measured ground level. The correlation coefficient (R) is 0.82, and the difference between the ground deformation rate and the traditional SBAS method is within 1 mm/a. The solution time is shortened by about 40%, and the ground deformation can be efficiently and continuously monitored.

Key words: InSAR time series analysis, small baseline set (SBAS), progressive SBAS, sequential adjustment; surface deformation, continuous monitoring

摘要:

基于短重访周期SAR卫星影像, 对黄河三角洲地表形变进行高效和持续监测的SBAS-InSAR时序分析。首先对研究区已有的SAR影像集进行干涉处理, 得到干涉图, 并进行大气效应校正和轨道误差去除, 然后利用传统的SBAS (small baseline subset)方法获取地表形变。在此基础上, 当增加新的SAR数据时, 采取渐进式SBAS方法处理, 反演新时刻的地表形变。渐进式SBAS方法融合序贯平差的思想, 以已有的解算结果为基础, 结合新的观测数据进行增量解算, 可以达到整体解算的效果。相对于传统的SBAS方法, 每次增加新影像都要采用重新全部解算的方式, 能够减少冗余运算, 提高计算效率。实验证明, 基于2018年5月—2016年8月在黄河三角洲地区获取的 Sentinel-1 卫星SAR数据, 利用渐进式SBAS方法反演的地表形变与地表实测控制点结果相近, 相关系数(R)为0.82, 且与传统的SBAS方法反演得到的地表形变速率差异在1 mm/a内, 解算时间缩短约40%, 能够持续高效地监测地表形变。

关键词: InSAR时序分析, 小基线集(SBAS), 渐进式SBAS, 序贯平差, 地表形变, 持续监测