作者: Aldo Clerici , Susanna Perego , Claudio Tellini , Paolo Vescovi
DOI: 10.1016/S0169-555X(02)00079-X
关键词: Data mining 、 Cartography 、 Class (computer programming) 、 Natural hazard 、 Geographic information system 、 Landslide 、 Point (geometry) 、 Subdivision 、 Geology 、 Multivariate statistics 、 Simplicity
摘要: Numerous methods have been proposed for landslide probability zonation of the landscape by means a Geographic Information System (GIS). Among multivariate methods, i.e. those which simultaneously take into account all factors contributing to instability, Conditional Analysis method applied subdivision territory Unique Condition Units is particularly straightforward from conceptual point view and suited use GIS. In fact, working on principle that future landslides are more likely occur under conditions led past susceptibility defined computing density in correspondence with different combinations instability factors. The simplicity this method, however, does not necessarily imply it simple implement, especially as requires rather complex operations high number GIS commands. Moreover, there possibility that, order achieve satisfactory results, procedure has be repeated few times changing or modifying class subdivision. To solve problem, we created shell program which, combining commands, Geographical Research Support (GRASS) commands gawk language carries out whole automatically. This makes construction Landslide Susceptibility Map easy fast large areas too, even when spatial resolution adopted, shown application Parma River basin, Italian Northern Apennines.