Votre recherche
Résultats 5 ressources
-
Abstract To increase the resilience of communities against floods, it is necessary to develop methodologies to estimate the vulnerability. The concept of vulnerability is multidimensional, but most flood vulnerability studies have focused only on the social approach. Nevertheless, in recent years, following seismic analysis, the physical point of view has increased its relevance. Therefore, the present study proposes a methodology to map the flood physical vulnerability and applies it using an index at urban parcel scale for a medium-sized town (Ponferrada, Spain). This index is based on multiple indicators fed by geographical open-source data, once they have been normalized and combined with different weights extracted from an Analytic Hierarchic Process. The results show a raster map of the physical vulnerability index that facilitates future emergency and flood risk management to diminish potential damages. A total of 22.7% of the urban parcels in the studied town present an index value higher than 0.4, which is considered highly vulnerable. The location of these urban parcels would have passed unnoticed without the use of open governmental datasets, when an average value would have been calculated for the overall municipality. Moreover, the building percentage covered by water was the most influential indicator in the study area, where the simulated flood was generated by an alleged dam break. The study exceeds the spatial constraints of collecting this type of data by direct interviews with inhabitants and allows for working with larger areas, identifying the physical buildings and infrastructure differences among the urban parcels.
-
This paper investigates local-scale social vulnerability to flood hazards in Romania, aiming to identify the most vulnerable social and demographic groups across a wide range of geographical locations by considering three dimensions: demographic, socioeconomic, and the built environment. The purpose of the paper is threefold: first, it strives to improve the Social Vulnerability model (SoVI®) by applying a different weighting method adapted to the Romanian context, taking into consideration the municipalities exposed to flood movements. Second, it aims to develop an assessment model for the most vulnerable communities by measuring the heterogeneity according to local indicators related to disaster risks. Third, it aims to facilitate emergency managers to identify community sub-groups that are more susceptible to loss and to increase the resilience of local communities. To perform local-level vulnerability mapping, 28 variables were selected and three aggregated indexes were constructed with the help of the ArcGIS software. Moreover, a model of Geographically Weighted Regression (GWR) between communities directly affected by floods and localities with high- and very high values of the Local Social Vulnerability Index (LoSoVI) was used to explore the spatial relationship among them and to compare the appropriateness of Ordinary Least Square (OLS) and GWR for such modelling. The established GWR model has revealed that the negative effects of flood hazards are often associated with communities with a high degree of social vulnerability. Thus, the analysis is able to provide a more comprehensive picture on communities in desperate need of financial resources in order to have the ability to diminish the negative impacts of flood hazards and to provide a more sustainable society.
-
Mapping the delineation of areas that are flooded due to water control infrastructure failure is a critical issue. Practical difficulties often present challenges to the accurate and effective analysis of dam-break hazard areas. Such studies are expensive, lengthy, and require large volumes of incoming data and refined technical skills. The creation of cost-efficient geospatial tools provides rapid and inexpensive estimates of instantaneous dam-break (due to structural failure) flooded areas that complement, but do not replace, the results of hydrodynamic simulations. The current study implements a Geographic Information System (GIS) based method that can provide useful information regarding the delineation of dam-break flood-prone areas in both data-scarce environments and transboundary regions, in the absence of detailed studies. Moreover, the proposed tool enables, without advanced technical skills, the analysis of a wide number of case studies that support the prioritization of interventions, or, in emergency situations, the simulation of numerous initial hypotheses (e.g., the modification of initial water level/volume in the case of limited dam functionality), without incurring high computational time. The proposed model is based on the commonly available data for masonry dams, i.e., dam geometry (e.g., reservoir capacity, dam height, and crest length), and a Digital Elevation Model. The model allows for rapid and cost-effective dam-break hazard mapping by evaluating three components: (i) the dam-failure discharge hydrograph, (ii) the propagation of the flood, and (iii) the delineation of flood-prone areas. The tool exhibited high accuracy and reliability in the identification of hypothetical dam-break flood-prone areas when compared to the results of traditional hydrodynamic approaches, as applied to a dam in Basilicata (Southern Italy). In particular, the over- and under-estimation rates of the proposed tool, for the San Giuliano dam, Basilicata, were evaluated by comparing its outputs with flood inundation maps that were obtained by two traditional methods whil using a one-dimensional and a two-dimensional propagation model, resulting in a specificity value of roughly 90%. These results confirm that most parts of the flood map were correctly classified as flooded by the proposed GIS model. A sensitivity value of over 75% confirms that several zones were also correctly identified as non-flooded. Moreover, the overall effectiveness and reliability of the proposed model were evaluated, for the Gleno Dam (located in the Central Italian Alps), by comparing the results of literature studies concerning the application of monodimensional numerical models and the extent of the flooded area reconstructed by the available historical information, obtaining an accuracy of around 94%. Finally, the computational efficiency of the proposed tool was tested on a demonstrative application of 250 Italian arch and gravity dams. The results, when carried out using a PC, Pentium Intel Core i5 Processor CPU 3.2 GHz, 8 GB RAM, required about 73 min, showing the potential of such a tool applied to dam-break flood mapping for a large number of dams.