Scholarlink

Time series of interferometric SAR (InSAR) images offer the potential to detect and monitor surface deformation with high spatial resolution, even for slow deformation processes. However, many different sources contribute to phase changes which are used in InSAR to estimate displacements. Complex displacement mechanisms or strong atmospheric contributions can complicate the separation of these contributions and even cause problems when unwrapping the phase. A preliminary model of expected displacements can support this process but requires information about all involved deformation processes. However, as these processes are often the main subject of the investigation, they are not sufficiently understood in advance. In this contribution, we approach this issue by analyzing InSAR time series results of regions with complex deformation behavior with the established statistical methods of principal and independent component analysis to identify dominant displacement patterns. We study Sentinel-1 InSAR data from 2015 to 2022 above the storage cavern field Epe in North Rhine Westphalia, Germany. Epe displays a spatially and temporally complex surface deformation field, which was described in previous studies as consisting of a linear signal relating to the cavern convergence as well as of seasonal and cavern pressure-dependent contributions. Our resulting displacement components can be clearly separated and appointed to different sources. This is supported by ground truth data and supplemental measurements of cavern pressure levels and groundwater levels. We also find that the previously described linear parametrization of displacements related to cavern convergence is no longer sufficient for longer time series. Our results show that we can obtain source-dependent displacement models from long and complex InSAR time series when using ICA. These can then either be used to refine time series processing or to describe the physical processes causing to the surface displacements with a geophysical source model. Both will be the subject of future investigations.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Signal Decomposition with InSAR Displacement Time Series Above a Storage Cavern Field: Example Epe (NRW, Germany)

  • 会议论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年

查看全文
  • Alison Seidel,
  • Malte Westerhaus,
  • Markus Even,
  • Hansjörg Kutterer

摘要

Time series of interferometric SAR (InSAR) images offer the potential to detect and monitor surface deformation with high spatial resolution, even for slow deformation processes. However, many different sources contribute to phase changes which are used in InSAR to estimate displacements. Complex displacement mechanisms or strong atmospheric contributions can complicate the separation of these contributions and even cause problems when unwrapping the phase. A preliminary model of expected displacements can support this process but requires information about all involved deformation processes. However, as these processes are often the main subject of the investigation, they are not sufficiently understood in advance. In this contribution, we approach this issue by analyzing InSAR time series results of regions with complex deformation behavior with the established statistical methods of principal and independent component analysis to identify dominant displacement patterns. We study Sentinel-1 InSAR data from 2015 to 2022 above the storage cavern field Epe in North Rhine Westphalia, Germany. Epe displays a spatially and temporally complex surface deformation field, which was described in previous studies as consisting of a linear signal relating to the cavern convergence as well as of seasonal and cavern pressure-dependent contributions. Our resulting displacement components can be clearly separated and appointed to different sources. This is supported by ground truth data and supplemental measurements of cavern pressure levels and groundwater levels. We also find that the previously described linear parametrization of displacements related to cavern convergence is no longer sufficient for longer time series. Our results show that we can obtain source-dependent displacement models from long and complex InSAR time series when using ICA. These can then either be used to refine time series processing or to describe the physical processes causing to the surface displacements with a geophysical source model. Both will be the subject of future investigations.