Last Updated : June 25, 2021

SBI (Split-bandwidth Interferometry) for displacement measurement

Released: June 25, 2021 JAPANESE

SBI (Split-bandwidth Interferometry) is a technique to measure ground displacements using SAR images split into two frequency subbands. This method is useful for measuring large crustal deformation as an applied technology of the standard InSAR. Here the technical background is described.

Signals in a SAR image have an extent of bandwidth with a certain central frequency (e.g., 1250 MHz in the range direction in the case of ALOS-2). While the full-bandwidth image is used for interferometry in general, it is also possible to exploit the subband images in which the bandwidth is split. SBI uses the subband images. The figure below depicts the concept of SBI. First, two images with lower (fL) and higher (fH) frequencies are produced by splitting the band from a full-bandwidth SAR image. Two interferograms with two different frequencies (ΔφH、ΔφL) are then be created from two SAR images acquired at different epochs by the same approach as the standard InSAR. In general, the amount of phase change due to displacement (pixel shift) is proportional to the frequency (i.e., inversely proportional to the wavelength). Hence, the interferogram between the two subband interferograms (ΔφH-ΔφL) corresponds to a small effective frequency (long effective wavelength).

Since a SAR image has each bandwidth for the range and azimuth directions, SBI is applicable in both directions. The range and azimuth SBI can measure displacement in the line-of-sight and along-track direction, respectively. The subband images along azimuth correspond to forward and backward looking images and the azimuth SBI is also called MAI.

While the measurement accuracy of SBI is worse than the standard InSAR, SBI obtains better coherence in areas with large displacement gradients than InSAR and can measure large displacements with even a few meters. This enables us to measure large crustal deformation caused by a large inland earthquake. Although SBI is applicable to every SAR image in principle, a certain amount of Δf is required to obtain a sufficient accuracy*. As modern spaceborne SAR data such as ALOS-2 tend to have wider bandwidth than before, larger Δf can be taken and the opportunities that SBI gives significant results have increased.




Advantage: Able to measure large ground displacement unlike InSAR
Disadvantage: Lower accuracy than InSAR (order of 10 cm)

* While the larger Δf, the shorter the effective wavelength, it is not a significant problem because the effective wavelength is still at least a few meters in general.

References

  • Bamler, R., & Eineder, M. (2005). Accuracy of differential shift estimation by correlation and split-bandwidth interferometry for wideband and Delta-k SAR systems. IEEE Geoscience and Remote Sensing Letters, 2(2), 151–155. https://doi.org/10.1109/LGRS.2004.843203
  • Jiang, H., Feng, G., Wang, T., & Bürgmann, R. (2017). Toward full exploitation of coherent and incoherent information in Sentinel-1 TOPS data for retrieving surface displacement: Application to the 2016 Kumamoto (Japan) earthquake. Geophysical Research Letters, 44(4), 1758–1767. https://doi.org/10.1002/2016GL072253