A method for automated snow avalanche debris detection through use of synthetic aperture radar (SAR) imaging

Abstract

Avalanches are a natural hazard that occur in mountainous regions of Troms County in northern Norway during winter and can cause loss of human life and damage to infrastructure. Knowledge of when and where they occur especially in remote, high mountain areas is often lacking due to difficult access. However, complete, spatiotemporal avalanche activity data sets are important for accurate avalanche forecasting, as well as for deeper understanding of the link between avalanche occurrences and the triggering snowpack and meteorological factors. It is therefore desirable to develop a technique that enables active mapping and monitoring of avalanches over an entire winter. Avalanche debris can be observed remotely over large spatial areas, under all weather and light conditions by synthetic aperture radar (SAR) satellites. The recently launched Sentinel-1A satellite acquires SAR images covering the entire Troms County with frequent updates. By focusing on a case study from New Year 2015 we use Sentinel-1A images to develop an automated avalanche debris detection algorithm that utilizes change detection and unsupervised object classification methods. We compare our results with manually identified avalanche debris and field-based images to quantify the algorithm accuracy. Our results indicate that a correct detection rate of over 60% can be achieved, which is sensitive to several algorithm parameters that may need revising. With further development and refinement of the algorithm, we believe that this method could play an effective role in future operational monitoring of avalanches within Troms and has potential application in avalanche forecasting areas worldwide.

Regional snow-avalanche detection using object-based image analysis of near-infrared aerial imagery

“…7). Our results show that avalanches were also detected on slopes steeper than 35 • , thus highlighting the limits of detection methods using arbitrary slope thresholds (Bühler et al, 2009;Vickers et al, 2016). Out of 193, 139 of the highest-lying avalanches (> 2800 m a.s.l.)…”

“…Previous studies have demonstrated that a combination of aerial images and digital elevation models (Bühler et al, 2009), or aerial (Lato et al, 2012 and satellite (Larsen et al, 2013), or SAR images (Vickers et al, 2016) favours the automatic detection of snow-avalanche deposits; however, identifying release zones and tracks remains challenging. We propose an OBIA algorithm that tracks release zones, tracks, and run-out zones in NIR images.…”

“…They found that release zones and tracks were mostly difficult to detect. In a similar context, the automatic method by Vickers et al (2016) evaluates backscatter in 50 × 50 pixel regions of Sentinel-1A images, subsequently masking out areas with a predicted zero probability of snow-avalanche occurrence; pixels with a DEM-derived local slope of > 35 • were ignored. From the test pixels, they selected those with a backscatter difference above a specified threshold.…”

Where are the avalanches? Rapid mapping of a large snow avalanche period with optical satellites

“…The mapped avalanches described in this paper are a unique training and validation dataset as it contains a large number of individual avalanches in differing topography and illumination conditions 5 over a very large region. The same applies for the validation of existing mapping product from radar satellite data (Eckerstorfer et al, 2017;Vickers et al, 2016;Wesselink et al, 2017).…”

“…Radar satellites have the advantage of acquiring data despite clouds and without daylight. Therefore, radar data has also been applied to generate avalanche maps (Eckerstorfer and Malnes, 2015;Vickers et al, 2016;Eckerstorfer et al, 40 2017;Wesselink et al, 2017). Due to the coarser spatial resolution (3 -30 m) and limitations by the observation geometries (radar shadow and layover), considerable parts of mountain regions cannot be covered.…”