Dating individual rainfall-triggered landslides with Sentinel-1 SAR time series: Application to the Nepal monsoon

Heavy rainfall events in mountainous areas can trigger thousands of destructive landslides, which pose a risk to people and infrastructure and significantly affect the landscape. Inventories of these landslides are used to assess their impact on the landscape and in hazard mitigation strategies and modelling. Optical and multi-spectral satellite imagery can be used to generate rainfall-triggered landslide inventories over wide areas, but cloud cover associated with the rainfall event can obscure this imagery. This delay means that for long rainfall events, such as the monsoon or successive typhoons, landslide timing is often poorly constrained. This lack of information on landslide timing limits both hazard mitigation strategies and our ability to model the physical landslide triggering processes.

Synthetic aperture radar (SAR) data represent an alternative source of information on landslides and can be acquired in all weather conditions. The removal of vegetation and movement of material caused by a landslide alters the radar scattering properties of the Earth’s surface. Landslides therefore have a signal in SAR imagery and the Sentinel-1 satellite constellation acquires SAR images every 12 days on two tracks globally, offering an opportunity to greatly improve the temporal resolution of individual landslides within an inventory whose trigger is poorly constrained in time, typical in regions with long periods of cloud cover. Here we present methods of using Sentinel-1 SAR amplitude time series to constrain landslide timing. Our approach combines three methods based on the change within the mapped landslide in (i) median amplitude versus the background,  (ii) amplitude spatial variability and  (iii) surface geometry. When applied to triggered landslides of known timing in Japan, Nepal and Zimbabwe, we achieved an overall accuracy of 80% when combining ascending and descending SAR tracks.

Further we apply our methods to inventories of monsoon-triggered landslides in Nepal (from 2015, 2016 and 2017) to decipher the relationship between landsliding and  local hydrometeorological conditions. Specifically, we first analysed the spatial and temporal clustering of timed landslides. Then we calibrated satellite-based rainfall with rainfall and/or river discharge gauges to understand the rainfall intensity over various timescales preceding the landslide occurrence retrieved by our method. We conclude with implications for empirical and physical modelling of monsoon-induced landsliding.