Key points are not available for this paper at this time.
When shallow landslides are triggered by sequences of earthquakes or storms, we need to know when in the sequence they occurred and whether they were later reactivated in order to better understand and model the hazard. In tropical environments, cloud cover often obscures all or part of the optical imagery acquired during the sequence, so that it is necessary to use images acquired after the sequence of events to map the landslides. Because of this, we often cannot tell which earthquake or storm triggered a particular landslide. This limits our understanding of how landslide hazard and mass wasting evolve in space and time during such sequences. Synthetic Aperture Radar (SAR) images can be acquired through cloud cover and since 2015, Sentinel-1 has acquired data globally every 6-12 days. Landslides alter the scattering properties of the Earths surface and so can be detected in SAR images. SAR amplitude time series can be used to constrain the timings of individual landslides. We apply these methods to three sequences of triggers in order to better understand how landsliding evolved during them: the 2018 Lombok, Indonesia earthquake sequence; the 2019 Cotabato Davao del Sur earthquake and the earthquake-hurricane sequence that occurred in Haiti in 2021. The 2018 Lombok earthquake sequence also offers an ideal opportunity to test new methods of using InSAR coherence (the level of noise in an interferogram) to detect multi-stage failure. High resolution cloud-free images were acquired halfway through this sequence of four Mw > 6.0 earthquakes and many landslides mapped after the first two earthquakes were then observed to have grown in size or changed shape by the end of the sequence. We demonstrate that for large, favourably oriented landslides, InSAR coherence can be sensitive to this reactivation and so could potentially be used in the future for cases where no cloud-free images are available.
Burrows et al. (Fri,) studied this question.