Phisical Vulnerability Proxies from Remotes Sensing: Reviewing, Implementing and Disseminating Selected Techniques

Risk from natural hazards and its direct effect in disrupting human livelihood is of paramount interest for maintaining a sustainable development, and remote sensing can be a valuable tool in collecting relevant information to assess and mitigate risk. This paper focuses on monitoring the vulnerability term of the risk equation, and in particular the physical vulnerability factor using remote sensing imagery. It provides an overview on state-of-the-art methodologies used in extracting a set of selected physical vulnerability indicators. The physical vulnerability of a building is defined as the probability of structural failure in the extreme situation of natural hazards like quakes. Although very difficult to compute precisely, it can be estimated indirectly through a set of representative indicators called "proxies". The literature offers different techniques for extracting a set of relevant information items from remote sensing imagery. When optical satellite images are used, both types of information, geometrical and spectral, are useful to extract features connected to proxies. Therefore, algorithms that combine spectral and spatial information would be the more effective choice in exploring the content of the acquired data. The relationship between the set of indicators and the aggregation method is expected to produce a time-changing synoptic view. The extracted information would then be used for mapping and so would support decision-making and help optimizing the selection of a risk management strategy.