Towards open service-oriented architectures for risk and crisis management
The effective provision of heterogeneous information stemming from various sources, being geographic or thematic maps, measurement values, damage assessment reports or weather forecasts play a major role in risk and crisis management. Most often these tasks are performed by public civil agencies on different administrative levels which all have their own information and communication systems for the provision of data and services. The possibility of sharing and fusioning all relevant information, especially in case of transboundary natural hazards, is often quite limited. Even in those intra-agency cases in which data exchange should be basically possible, different data formats and service interfaces, but also differences in the thematic views, temporal and spatial resolutions and semantic descriptions about the crisis situation, hinder an efficient interpretation of the existing data observations. This is of particular concern if the data is to be used in time critical decision making situations as they regularly occur in situation rooms for crisis management. This paper first describes how the problem of syntactic and semantic interoperability may be approached by the design of open serviceoriented architectures (SOA) based upon international standards, such as those of the European research projects ORCHESTRA (Open Architecture and Spatial Data Infrastructure for Risk Management, http://www.eu-orchestra.org/ ), SANY (Sensors Anywhere) and TRIDEC (Collaborative, complex and critical decision-support in evolving crises, http://www.tridec-online. eu). These architectural approaches support the design and implementation of risk and crisis management application relying upon standard interfaces of the Open Geospatial Consortium (OGC). The Reference Model of the ORCHESTRA Architecture (RM-OA) [Usländer, T. (ed.) (2007)] was recognized as OGC bestpractices architecture in 2007 and since then has been extended and refined in numerous research and implementation projects. The paper then proceeds with pointing out how human computer interaction in situation rooms for emergency management may be designed in such a way that spatial data and other information can be intuitively presented to crisis managers, relieving them from the chore of performing complex interaction schemes while already overloaded with time-critical decision making.