Approccio metodologico per la valutazione modulare della vulnerabilità finalizzata alla riduzione dei rischi naturali antropici 2021

Abstract

Vulnerability is an important component of risk assessment and represents the main element in the risk perception. Typically, the characteristics related to social, cultural, physical and institutional factors increase the susceptibility of an individual or a community, to the impacts of hazards. Vulnerability is described as a dynamic phenomenon that can vary significantly across time and space; it is greatly influenced by human actions and behaviors, by the emergency response related to road accessibility. As a consequence, there is a continuing need for risk reduction disaster strategies to shift attention from assessing hazard events toward reducing vulnerabilities within social systems. Describing and quantifying vulnerability is an important challenge along this path. Our current understanding of vulnerability is guided by methodologies, indicators and measurement standards derived from different schools of thought. This thesis presents a methodological approach to describe and to assess the vulnerability index at the inter-municipal scale, using three indices. Spatial analysis is conducted on the basis of census zones in an area defined as “Territorial Context” (TC) characterized by the union of municipalities. A measure of modular vulnerability is evaluated on the basis of inductive methods. Vulnerability is defined as the conditions determined from social and economic factors from human and climatic territorial pressures, from critical issues generated by past events and from the functioning of road infrastructures during an event. The three modular components of the vulnerability are: TCVIpeople (Territorial Context Vulnerability index-people); TCVIexposure (Territorial Context Vulnerability index-exposure); and TCVIemergency (Territorial Context Vulnerability index-emergency). Thirty-eight variables are selected and geoprocessed for each of the 195 census analysis units in the Mediterranean study area of southern Italy. Using Principal Components Analysis (PCA) with varimax rotation and Kaiser criterion for component selection, the social and territorial vulnerability index, are identified. The third vulnerability index, TCVIemergency, is processed through the transport modeling technique. In the latter case, a contextual interruption of all road network exposed to the highest level of hazard is assumed. Models are implemented to assess the forest fire, flood and landslide hazards. The TCVIemergency index is calculated (on the basis of the differences in travel time, after and before the event, from the origin (centroids of the census areas) to the destination points (strategic buildings in emergency planning and civil protection operational structures), using the shortest paths network model. This index can provides useful information for evacuation planning and rescue operation during emergency situations. A fuzzy logic model is used to evaluate the vulnerability classification, while the fuzzy overlay function is used to calculate the final aggregate TCVI index. The performance of classification models is measured by some statistical metrics. A dedicated Geographic Information System (GIS) is used to capture, geo-process and display spatial data recorded at different scales. The GIS technology allows to evaluate and visualize the results, through maps, as a realistic representation and to identify and manage the process. The results contribute to debates in contemporary literature on vulnerability in many ways. First these analyses constitute an attempt to quantify and mapping vulnerability at census area in a natural or handmade scenario. Secondly new variables like the road network representing the category most damaged during the events, with the greatest repercussions on the community and on the economy, are introduced. The current methods of vulnerability assessment are in fact mainly based on social aspects, the built environment and climatic factors, leaving out the importance of the road infrastructure. The model is developed at the census area, which is the smallest geographic unit that the National Institute of Statistics uses to aggregate demographic data, in an inter-municipal area. It is well known that vulnerability is a scale-dependent variable and it could be very accurate for larger spatial scales than the TC area. Moreover, new classification criteria for vulnerability maps are investigated, using fuzzy set theories. Finally, working with the territorial contexts TC, a new approach for risk reduction is defined, in order to better meet the needs of the Civil protection activities. This is the first national attempt to calculate the spatial distribution of vulnerability in a territorial context functional to emergency planning. Through this study, a comprehensive understanding of the relative driving components contributing to the overall vulnerability is achieved. Results show significant differences in the spatial distribution of the social vulnerability, highlighting the multidimensionality and heterogeneity of the municipal characteristics. The TCVI in the southern and central part of TC is higher than that its northern and western parts. In general, by analyzing the results of the vulnerability values it must be noted that about 56% of census areas are characterized by low and low-medium, while 35 % fall into categories labelled with high, very high and the remaining 9% falls into the moderate vulnerability category. The vulnerability maps provide useful territorial information, that can support policy-makers for prevention and emergency management. Within the context of natural and handmade hazards, the TCVI could be used to manage the repartition of resource, helps to determine which places may need specialized attention during immediate response and long-term recovery after an extreme event. It can provide an indication of the housing areas that need development and humanitarian aids and can provide guidance for better preparedness, response and mitigation strategies. The vulnerability maps can also be used as guidance to road administrations in the planning and in their investment to prioritize interventions and for normal maintenance and control activities. Actions and emergency measure are directly connected with resilience, then this work can help to strong intent to increase capacity building of human resources, better land use management, increasing preparedness and emergency measures that are taken during and after event. Following the introduction section, the present study is composed by two main sections that delve into: 1) conceptual frameworks for vulnerability and hazards assessments. This is accomplished by discussing the relevant primary research literature and analyzing the events recorded in the past; 2) methodological approaches to model natural and anthropic hazards and for vulnerability measuring in a Territorial Context. An application in the Territorial Context of Marina of Gioiosa Ionica in Southern Italy, is developed. Finally, the last section presents the main conclusions of the study and potential developments. Keywords: forest fire, landslide and flood hazard, vulnerability index, territorial context, indices and maps, social and territorial vulnerability, road susceptibility.