RST Instruments Measurand 3vGeomatics SYSCOM Instruments
Terra Insights
Back to News

News

Landslide at Elliot Lake, British Columbia, Canada

A landslide occurred near Bute Inlet, British Columbia, Canada, in November 2020. The landslide caused some 13 million cubic metres of debris to crash into Elliot Lake, lying below the unstable rock face, causing a wave over 100 metres high and 15 million m3 to be ejected from the lake. The resulting flash flood proceeded down the fjord to the ocean at Bute Inlet, leaving behind a 60 kilometre long sediment plume that disrupted the turbidity, temperature, and chemistry of the waterway.

Bute Inlet, on the southwest coast of British Columbia, Canada experienced a large landslide in November 2020.

Bute Inlet, on the southwest coast of British Columbia, Canada experienced a large landslide in November 2020.

Using Synthetic Aperture Radar (SAR) imagery to perform interferometry (InSAR), areas of instability can be found, measured, and reported on for risk management. Using historical imagery, we discovered a considerable increase in the rate of displacement for the landslide slope in the summer prior to failure. In previous years, the area had been slowly subsiding. These InSAR results were included in the recently published paper, The 28 November 2020 Landslide, Tsunami, and Outburst Flood – A Hazard Cascade Associated With Rapid Deglaciation at Elliot Creek, British Columbia, Canada examining the landslide and debris flow events. 

InSAR exploits the phase difference between two SAR images acquired at different times over the same area to estimate ground deformation with centimeter to millimeter precision. Images must be acquired using the same satellite parameters and orbit position in order to accurately perform the phase difference. 

Sentinel-1A and Sentinel-1B (collectively referred to as “S1”) , were launched under the European Space Agency’s (ESA) Copernicus programme. They collect near continuous SAR imagery all over the world and this data is made publicly available. S1 typically acquires images in 250 km wide swaths with 5 m by 20 m spatial resolution, every six days in Europe and 12 days across the rest of the world. This means, every 12 days, we are able to capture an image of the area of interest with the same imaging parameters. 

For this study, we used images collected from two neighboring ascending orbits (orbits 35 and 137) to perform two separate analyses. Due to the topography at the head of Elliot Creek, no descending orbit had good sensitivity over the landslide area, thus no data from the descending orbits were used. The ascending orbits were used to perform independent line-of-sight (LOS) analyses. 

Sentinel-1B

Sentinel-1A and Sentinel-1B orbit as a constellation and are nearly identical satellites. InSAR can be performed using data from either satellite. Photo: ESA; Image: Artist’s impression of Sentinel-1B

Sentinel-1A and -1B radar scans combined

This interferogram over Romania demonstrates how Sentinel-1A and Sentinel-1B images can be combined. Photo: ESA; Image: Sentinel-1A and -1B radar scans combined

Only snow-free SAR images were used for the study; images with snow cover were identified and excluded based on historical optical satellite imagery. A total of 127 images were collected over the study area from June 2018 to November 2020. Of these, 73 were removed due to snow, yielding 54 images for this analysis (25 for orbit 35 and 29 for orbit 137).

Left and right images show the estimated displacement rate for orbits 35 and 137, respectively. By using multiple orbits, InSAR results can be corroborated between image stacks. Both ascending orbits show more than 5cm of displacement per year in the failure area. 

An overview of the estimated displacement rate over the study area is seen above for orbits 35 and 137, respectively. Individual InSAR measurements are displayed as coloured points where the color represents the component of displacement along the satellite’s LOS. The red end of the colour scale represents displacement away from the satellite, while the blue end of the scale denotes displacement towards the satellite.

Individual InSAR measurements are displayed as coloured points where the color represents the component of displacement along the satellite’s LOS. By graphing 40 points within the slide area reveals the average rate of cumulative displacement. Near the end of the time series, in the most recent months, the rate of displacement increased significantly up to the point of failure. 

The InSAR results show precursory displacement leading up to the failure of the slope. A pronounced increase in the displacement rate is evident in summer 2020 compared to previous years. A LOS displacement of up to 150 mm was estimated from June to October 2020, yielding a LOS velocity of 300 mm per year, in summer 2020. The two ascending analyses show comparable displacement rates and displacement area extents. 

Displacement areas, such as those at Elliot Lake, can be discovered ahead of failure by using the wide coverage and high precision of InSAR.

We use Sentinel data, continuously downloaded and processed in the background, for all ongoing clients. This background data is used in our Emergency Response program and is designed to quickly provide geotechnical staff with the most recent displacement information. Once initiated, we can acquire the next available Sentinel imagery over the site and provide analysis ahead of the next contracted product delivery.

To learn more about how 3vG uses Sentinel data, or for a free site assessment, contact us below:

Free Site Assessment
2022-05-03T14:06:36-07:00April 19th, 2022|

Share This Story, Choose Your Platform!

Go to Top