International audienceStatistical assessment of landslide susceptibility has become a major topic of research in the last decade. Most progress has been accomplished on producing susceptibility maps at meso-scales (1:50,000–1:25,000). At 1:10,000 scale, which is the scale of production of most regulatory landslide hazard and risk maps in Europe, few tests on the performance of these methods have been performed. This paper presents a procedure to identify the best variables for landslide susceptibility assessment through a bivariate technique (weights of evidence, WOE) and discusses the best way to minimize conditional independence (CI) between the predictive variables. Indeed, violating CI can severely bias the simulated maps by over- or under-estimating landslide probabilities. The proposed strategy includes four steps: (i) identification of the best response variable (RV) to represent landslide events, (ii) identification of the best combination of predictive variables (PVs) and neo-predictive variables (nPVs) to increase the performance of the statistical model, (iii) evaluation of the performance of the simulations by appropriate tests, and (iv) evaluation of the statistical model by expert judgment. The study site is the north-facing hillslope of the Barcelonnette Basin (France), affected by several types of landslides and characterized by a complex morphology. Results indicate that bivariate methods are powerful to assess landslide susceptibility at 1:10,000 scale. However, the method is limited from a geomorphological viewpoint when RVs and PVs are complex or poorly informative. It is demonstrated that expert knowledge has still to be introduced in statistical models to produce reliable landslide susceptibility maps

International audienceThe objective of this work is to present a low-cost methodology to monitor the displacement of continuously active landslides from ground-based optical images analyzed with a normalized image correlation technique. The performance of the method is evaluated on a series of images acquired on the Super-Sauze landslide (South French Alps) over the period 2008-2009. The image monitoring system consists of a high resolution optical camera installed on a concrete pillar located on a stable crest in front of the landslide and controlled by a datalogger. The data are processed with a cross-correlation algorithm applied to the full resolution images in the acquisition geometry. Then, the calculated 2D displacement field is orthorectified with a back projection technique using a high resolution DEM interpolated from Airborne Laser Scanning (ALS) data. The heterogeneous displacement field of the landslide is thus characterized in time and space. The performance of the technique is assessed using differential GPS surveys as reference. The sources of error affecting the results are then discussed. The strongest limitations for the application of the technique are related to the meteorological, illumination and ground surface conditions inducing partial or complete loss of coherence among the images. Small movements of the camera and the use of a mono-temporal DEM are the most important factors affecting the accuracy of the ortho-rectification of the displacement field. As the proposed methodology can be routinely and automatically applied, it offers promising perspectives for operational applications like, for instance, in early warning systems

No abstract