Detecting Mountainous Landslides by SAR Polarimetry: A Comparative Study Using Pi-SAR-L2 and X-band SARs

Abstract

In this paper, we evaluate the polarimetric sensitivity of multi-frequency synthetic aperture radars (SARs) for detecting landslide areas in forest-covered mountains. We tested L-band airborne and X-band spaceborne SARs, i.e. the airborne Polarimetric Interferometric SAR in L-band (Pi-SAR-L2), Terra SAR X, and Cosmo Skymed, at the Totsukawa-mura test site in Nara, Japan. We found that three parameters—the coherence of HH and VV, polarimetric entropy, and the power ratio of HH/HV—are very effective, especially with L-band SAR, for detecting land cover changes from a forest to a landslide. Results show that X-band SARs are less sensitive to landslide areas because the X-band penetrates less through a forest compared to the L-band.