TOWARDS THE CONSTRUCTION OF A SPATIAL DATABASE TO MANAGE LANDSLIDES WITH GIS IN MOUNTAINOUS ENVIRONMENT
Résumé
While different GIS-based statistical algorithms of landslide predictive modeling exist, few studies explain (i) the best way to acquire the input data and (ii) the overall influence of the input datasets on the quality of the output results. For most of the studies, the main sources of information are aerial photographs and topographic maps. As the analysis of these documents is often delicate, especially for the recognition of the unstable zones (shadow, morphology hidden by vegetation), a field verification (somehow costly) is essential. One other difficulty stands in the integration of data that have different origins (image data, field observations, technical reports) and unequal precision. Even though some authors have proposed a roadmap defining the best source of documents to integrate in the database (according to the working scale), any sound evaluation of the source information, or any assessment of the acquisition exist. A methodological reflexion has to be carried out to define the best source of information according to the objectives of the study, the scale of interest, the complexity of the environment (topography, land-use) and the time available.
Domaines
Géomorphologie
Origine : Fichiers éditeurs autorisés sur une archive ouverte
Loading...