Key points are not available for this paper at this time.
Vulnerability is the most important predictive variable in the risk equation, but it isn't easy to evaluate the best objective approach to quantify it. Another hot topic of debate among scientists is whether vulnerability analysis describes only patterns or can also produce a quantitative value. The need to streamline and provide comparable and easy-to-use results has led to developing vulnerability indicators. Generally, these provide some form of aggregation of underlying factors, often including hazard exposure. Factor selection varies from deductive approaches, based on theoretical understanding, to inductive ones, based on statistical relationships. For the past thirty years, there have been significant efforts to measure vulnerability, but up to now, the field of vulnerability assessments has been dominated by hierarchical versus inductive approaches. The hierarchical analysis is a transparent approach, more accessible to stakeholders due to its logical structure and statistical support, and capable of functioning with more available datasets for assessing vulnerabilities in the studied areas. These are the most eloquent reasons for preferring the hierarchical approach in stakeholder territorial management and mitigation policies. The inductive, statistical approach developed by Cutter (based on the hazards-of-place model, Cutter, 1996) uses the principal component analysis (PCA) to establish vulnerability factors over time and eliminates the biases from aggregated decisions. Against this background, our study proposes a new model for quantifying vulnerability using an Impact Chain-based approach, taking as an initial case study the powerful flood events and the COVID-19 pandemic that affected Romania in 2020-2021. The hazards, impacts, vulnerability, exposed elements, and adaptation options pertaining to the case study are integrated into a comprehensive Impact Chain that is used as the foundation for the model. The proposed model relies on factorial techniques and ANOVA, with a focus on identifying statistically significant multiple regressions. It also integrates an optimization procedure that enables either a maximum value response or a minimum accepted value. This new framework allows for identifying vulnerability's influencing role in unfolding a multi-hazard and pinpointing the potential ways in which vulnerability can be affected by this unfolding. Thus, the model looks at vulnerability with a double lens, assessing its power to induce change by conditioning impacts and adaptation options and its propensity to change by certain impacts and adaptation options working in asynergy. Only by thoroughly analyzing both of these facets and understanding their implications can we produce bias-(more) free vulnerability assessments, particularly in multi-hazard contexts. Keywords: vulnerability, vulnerability approaches, hierarchical approach, inductive approach, Impact Chain
Armaș et al. (Sat,) studied this question.