Integration of optical remote sensing data flows in landslide early warning systems : Status, Limitations, Perspectives

Stumpf, A. (1,2), Malet, J.-P. (1), Travelletti, J. (1,3), Gance, J. (1,4)

(1) Institut de Physique du Globe de Strasbourg, CNRS UMR 7516, Université de Strasbourg, 5 rue Descartes, F-67084 Strasbourg Cedex, France
(2) Laboratoire Image, Ville, Environnement, CNRS UMR 7362, Université de Strasbourg, France
(3) BEG SA, Bureau d’Etudes Géologiques, Aproz, Switzerland
(4) BRGM, Bureau de Recherches Géologiques et Minières, Orléans, France

Due to technological advances in digital photography, UAV (unmanned aerial vehicle) remote sensing, digital image correlation and multi-view photogrammetry, terrestrial and low-altitude optical remote sensing become increasingly useful to observe small scale surface changes that can be important early warning indicators. During recent years, the Super-Sauze landslide (Southern French Alps) has been continuously surveyed with low altitude UAVs, fixed terrestrial cameras and mobile handheld cameras. A number of automatic and semi-automatic analysis methods to extract relevant features and monitor their evolution has been developed and tested.
 Digital image correlation to derive displacement fields from terrestrial cameras during acceleration and deceleration phases ;
 Semi-automatic image filtering for the detection of surface fissures in UAV images ;
 Time-lapse photogrammetry for the 3D tracking of installed targets and other prominent surface features with fixed terrestrial cameras ;
 Multi-temporal multi-view photogrammetry with photographs recorded with handheld cameras to reconstruct 3D models of the landslide and quantify deformation and erosion.
The objective of this work is to provide an overview of the temporal and spatial resolution of those techniques and their potential to reveal quantitative and qualitative information relevant for the identification of landslide geo-indicators and for early warning. Current limitations of those techniques regarding temporal frequencies, weather conditions, etc. are discussed, as well as future directions for field installations, real-time information transfer and more robust processing techniques.