Tailoring spatial reference in early warning systems to administrative units

Abstract

This paper presents a novel approach for early warning systems to transform the spatial reference of dispersion models into addressable administrative units enabling an intelligible warning message reception and spatial awareness. The spatial reference along the entire information chain from sensor measurements via dispersion model to warning dissemination is discussed. Core concept behind this approach is the application of the Egenhofer-Model for analysis of binary topologic relations. It is used to identify topologic relations between dispersion model forecasts and administrative units. The result set of intersection matrices defines the relations that are counted for area classification with hazard specific parameters. Following the design for a generic and hazard-independent web service based on Open Geospatial Consortium (OGC) standards is presented in combination with results of the reference implementation.