Landslides Risk Scenarios and Associated Uncertainties in Relatively Scarce Data Environment: a Methodological Approach for Buzau County, Romania

摘要

Global changes are strongly affecting the environment, inducing pattern changes of the environmental hazards and consequently making necessary a more focused attention in modeling future risk through the development of possible alternative scenarios.Scenarios became in the recent period an integrant part within risk studies and can be considered a useful support instrument for decision makings.Quantitative simulation models and analysis procedures currently used in hazard and risk assessment are often not applicable in relatively scarce data environments, where reasonably complete inventory of historical events is commonly not available.Additionally, due to the lack of data and the resulting need to approximate, in such environment is fundamentally important to carry out an estimation of the associated uncertainty.This contribution that is focusing more on a methodological framework and suggests an expert based approach, that can be considered a useful modeling framework for the combination of qualitative and quantitative knowledge.This Formative Scenario Analysis integrates intuitions and qualitative evaluation of impact factors with the quantitative analysis of relations between these factors, using fuzzy sets theory for the uncertainty estimation.The method is applied to Buzau County (Romania), an area belonging to the Curvature Romanian Carpathians and Subcarpathians, a region strongly affected by environmental hazards.The region has been previously involved in numerous episodes of severe hydro-meteorological events that caused considerable damages (1975, 2005, 2006).The paper outlines the importance of developing proper scenarios, focusing on the advantages that such an approach may bring (especially considering uncertainty assessment).This abstract is part of the ongoing 7th Framework Program "CHANGES" (2011-2014, Grant Agreement No.263953) of the European Commission.