Searching for landslides in California with open-access InSAR data processed by the Advanced Rapid Imaging and Analysis (ARIA) Project for Natural Hazards

Abstract:

Landslides are a major natural hazard and are often the dominant process that erode mountainous landscapes. Both their hazardous impact and erosive potential depend on the landslide properties including the velocity, geometry, and frequency of occurrence. Here we use standardized open access interferometric synthetic aperture radar (InSAR) data processed by the Caltech-Jet Propulsion Laboratory Advanced Rapid Imaging and Analysis (JPL-ARIA) to identify and monitor slow-moving landslides in central and northern California. The SAR data were collected by the Copernicus Sentinel-1 A/B satellites between 2015-2020 and processed to geocoded unwrapped interferograms with 3-arcsecond (~90 m) posting by JPL-ARIA. Data products are freely available from the NASA Alaska Satellite Facility archive. To download, manipulate, and analyze these data, we use the open-source ARIA-tools (https://github.com/aria-tools/ARIA-tools) and construct deformation time series using the open source Miami INsar Time-series software in PYthon (MintPy) software package (https://github.com/insarlab/MintPy). We then apply a newly developed InSAR detection approach which uses both local and regional spatial filters to reduce long-wavelength noise (e.g., tectonic processes, ionospheric and tropospheric noise) and reveal localized deformation features such as landslides. Our preliminary results show hundreds of active slow-moving landslides, with particularly high clusters of landslides on the Big Sur coast, central San Andreas Fault, and in the Eel River catchment, all areas well-known for high landslide activity. We directly compare the displacement time series for dozens of landslides to local precipitation and find a direct relation between cumulative rainfall and landslide movement. Our work shows how open source standardized InSAR data can be used to better understand landslide processes by measuring their size, motion, and frequency of occurrence over large areas and long time periods.