Population density modelling in support of seismic risk assessment

Abstract

Demographic data is a fundamental component of an earthquake loss model for the estimation of human exposure and social vulnerability. Fine scale population distribution information is needed for the assessment of causalities, determination of shelter needs and proper implementation of evacuation plans in pre- and/or post-disaster phases, i.e. earthquake scenario modelling and rapid emergency response. This paper describes the techniques that are used to map the population distribution, and to integrate the building damage information with demographic data for casualty estimation. Methods to map population density are described focusing on different downscaling techniques and the contribution of ancillary data. An application for the Larger Urban Zone (LUZ) of Vienna is illustrated. The case study disaggregates the residential population from a local and a country level census at the level of single building blocks. The downscaling is based on a dasymetric approach using an urban land use map as ancillary information. The model was applied after testing two different methods: a limiting variable and a fixed-ratio method. The latter performed slightly better and was applied on the entire study area. The results of the proposed methodology can be used to perform population vulnerability analysis for night time scenarios which deals mainly with residential building typologies. The enhanced spatial detail influences the accuracy of the information on human exposure when the population map is used in risk assessment models.