Preprint/Prétirage CHARACTERIZING AND MONITORING ROCKSLIDES FROM SAR TECHNIQUES V. Singhroy 1 and K. Molch 2 1 Canada Centre for Remote Sensing (CCRS), 588 Booth Street, Ottawa, ON Canada, K1A 0Y7 2 MIR Télédétection on assignment at CCRS, 588 Booth Street, Ottawa, ON Canada, K1A 0Y7 ABSTRACT Worldwide, thousands of landslides occur annually moving millions of tons of material. Based on estimates from the International Federation of Red Cross and Red Crescent Societies there are on average 1,550 landslide-related deaths per year. Developing new remote sensing techniques to identify, characterise and monitor motion of landslides will assist in the current national landslide inventory and hazard mapping in mountainous areas. Recent research has shown that interferometric SAR techniques can be used to monitor landslide motion under limited conditions. In this study we used interferometrically derived images, to monitor post slide motion and a RADARSAT fine mode image to characterise the debris size and distribution of a 30x 10 6 m 3 rock avalanche, in the Canadian Rockies. These techniques will assist in the understanding of landslide processes, post failure mechanism and mobility. INTRODUCTION Traditionally, the field of civil engineering has been involved in the assessment of landslide hazards. Slope stability analysis has been used to assess landslide hazards, and more recently remote sensing techniques are being used in stability assessment (Murphy and Inkpen, 1996; Singhroy et al., 1998; Singhroy and Mattar, 2000; Bulmer et al., 1999). Two distinct approaches can be used to determine the characteristics of landslides from remotely sensed data. The first approach determines the number, distribution, type and character of landslides using high- resolution stereo and fused images. The second approach complements the first one, by measuring dimensions (length, width, thickness and local slope, motion, and debris distribution) along and across the landslides using stereo SAR, InSAR and topographic profiles (e.g. laser altimeter profiles). Where possible these dimensional data are compared to field information and previous studies. Several case studies have reported the use of differential interferometry to monitor landslide motion (Fruneau et al., 1996; Vietmeier et al., 1999; Rott et al., 1999). Provided coherence is maintained over longer periods, as is possible e.g. in non-vegetated areas, surface displacement of a few cm per year can be observed. Rott (1999), reported on successfully mapping landslide motion above the treeline from interferograms covering time spans of up to three years. Depending of course on the rates of movement expected, in data pairs with short perpendicular baselines and short time intervals between acquisitions the effect of topography on the differential interferogram is minimized, and coherence is more likely to be maintained in non-vegetated areas, therefore allowing for more reliable measurements of surface displacement. Contrary to motion on the detachment zone, roughness and distribution of landslide debris and their post slide stability has not been studied in detail using remote sensing. This is due in part to the lack of topographic data for blocky landslides and therefore the link between debris roughness and radar backscatter (σ 0 ) has remained elusive. Roughness is defined as the topographic expression of surfaces at horizontal scales of centimetres to a few hundred meters. Landslide surface structures and roughness provide information on flow emplacement parameters (such as emplacement rate, velocity, and rheology). Laser altimeters are used to calculate surface roughness. Digital image analyses of large-scale photographs were used to examine grain size distribution of rock avalanche debris (Couture