Department of Geography and Regional Research, Vienna, AustriaCorrespondence to: P. Reichenbach (paola.reichenbach@irpi.cnr.it)The special issue of Natural Hazards and Earth SystemSciences entitled “Landslide hazard and risk assessment atdifferent scales” contains 6 of more than 20 oral and postercontributions originally presented in the session NH3.9 –“Landslide risk assessment methods and strategies” held atthe General Assembly of the European Geosciences Union,in Vienna, Austria, on 7 May 2010. Topics discussed in thesession included the following: (i) heuristic, statistical, deter-ministic, and physically based methods and models to eval-uate landslide susceptibility and hazard, and (ii) evaluationof qualitative and quantitative vulnerability and risk assess-ments based on different data. During the session, variouscontributions were discussed dealing with landslide hazardand risk assessment at local, regional or national scales, indifferent physiographic, climatic and geological settings.The meeting proved a valuable opportunity to discuss andcompare methods, techniques, and tools for the identifica-tion, evaluation, and mitigation of landslide hazards and theassociated risks. Despite the large number of contributionspresenting case study applications to evaluate landslide haz-ards and risk, information on the quality, the reliability andthe limitations of the applied models was not analyzed indetail and should be considered a major issue for future re-search. In the following, we summarize the content and themain results of the papers originally presented and discussedat the meeting that are published in this special issue.Floris et al. (2011) exploit spatial data available in Ital-ian WebGIS portals to evaluate landslide susceptibility ofthe Euganean Hills Regional Park, located SW of Padua, NEItaly. In the paper, quality, applicability and possible analysisscales of the online data were investigated and a susceptibil-ity analysis of the study area was completed using a prob-abilistic approach that compared landslide distribution andthe influencing factors. The input factors used in the analysisinclude landslide distribution, morphometric data (i.e., ele-vation, slope, curvature, profile and plan curvature) and non-morphometric data (i.e., land use, distance to roads and dis-tance to rivers). Attention was paid to data pre-processing, inparticular to the re-classification of continuous data that wasperformed following objective, geologic, and geomorpholog-ical criteria. The results show that the simple probabilisticapproach used for the susceptibility evaluation was accurateand precise (repeatability). Heuristic, statistical, or determin-istic methods could be applied to the online data to improvethe prediction. The data available online for Italy allow forthe assessment of landslide susceptibility at medium to smallscales. Morphometric factors, including terrain elevation andslope angle, provide significant information where lithologi-cal and structural data are not available. The main drawbackof the online data sources was the lack of information on thetemporal frequency of landslides. For this reason, a completehazard analysis is not possible.Pereira et al. (2012) use a bivariate statistical model toidentify a combination of landslide predisposing factors thatbest predicted landslide susceptibility in the Santa Marta dePenagui˜ao study area (70km