Rapid landslide mapping by means of post-event polarimetric SAR imagery

Kurzfassung

This paper presents a fast and transferable procedure for rapid mapping of landslides, with the advantage of requiring only post-event SAR imagery. The methodology was developed using dual-polarimetric TerraSAR-X imagery acquired over a landslide at the Yeager Airport near Charleston, West Virginia, USA. Then, the transferability of the methodology was successfully demonstrated by applying it to a further landslide event, a huge mining waste landslide which occurred on 1 April 2015 near Bolshaya Talda, Kemerovo Oblast, Russia. On 12 March 2015, a large-scale landslide occurred at an artificial slope at the Yeager Airport. The airport, completed in 1947, was constructed atop seven semi-connected hilltops. In 2006, due to new Federal Aviation Administration (FAA) safety regulations, the construction of an Engineered Material Arrestor System (EMAS) was necessary. Therefore, the Yeager Airport had to be extended, leading to the construction of a large artificial slope, being the tallest geosynthetic reinforced slope in North America (of ratio horizontal/vertical 1:1). Very High Resolution (VHR) archive SAR imagery recorded shortly before a landslide event is commonly not available. Modern VHR SAR missions such as TerraSAR-X, COSMO-SkyMed or Radarsat-2 do not systematically cover the entire world. Each acquisition has to be programmed manually. Furthermore, due to limited disk space on board the satellites and especially due to limited downlink transmission rates, these sensors are not able to provide worldwide coverage within a short time period – i.e., no archive image recorded shortly before the event. Therefore, we present a methodology for rapid landslide mapping based solely on post-event SAR imagery. Shortly after the first failure of the slope at the Yeager Airport a dual-polarimetric TerraSAR-X HighResolution SpotLight dataset was programmed. About one month later a second failure of the slope occurred, which was then recorded by a second TerraSAR-X scene acquired using the same imaging geometry as the first one. Assuming land cover changes due to the landslide event, i.e. destruction and removal of the vegetation cover – as it was the case at the Yeager Airport landslide – the first step of our object-oriented procedure is the pre-selection of formerly vegetated areas based on the Normalized Difference Vegetation Index (NDVI) of freely available pre-event optical imagery (e.g. Landsat-8 or the Sentinel-2). Next, after polarimetric speckle filtering, the entropy/anisotropy/alpha (H/A/alpha) decomposition is applied to the post-event polarimetric SAR image to detect (within the pre-selected areas) regions characterized by low entropy values, i.e. an evidence of bare soil or rock (landslide material). Then, assuming a certain minimum slope value as a necessary requirement for a landslide event, the landslide detection map is refined accordingly. The developed methodology was applied independently to both aforementioned TerraSAR-X images showing the landslide at different stages of development. As a second failure of the landslide occurred between the two acquisition dates, change detection procedures show high accuracies in mapping the active landslide.