Versatile Risk Indicator Maps using SVG

drs Barend Köbben
International Institute for Geo-information Science and Earth Observation (ITC)
PO Box 6, 7500 AA Enschede, The Netherlands

Keywords: Web mapping, Presenting GML with SVG, Data-driven graphics, Case studies


This paper presents results of a proof-of-concept implementation, using SVG for Risk Indicator Maps. These maps are to be be part of an urban risk management system, and therefore need to be fit for a multitude of tasks, ranging from providing the general public with information about risks to providing local authorities an interface to the underlying risk assessment databases and models. Furthermore, the maps need to be useable on a wide range of platforms, from the office systems of the local authorities, through public access via the WWW, to handheld devices providing location based services to field personnel. The required versatility, scalability and the need for the mapping solution to fit in a framework of component-based interoperable web services led to the choice for SVG.

The SLARIM project

This research is part of the larger framework of the project “Strengthening Local Authorities in Risk Management” (SLARIM). The objective of this ITC internal research project is the development of a methodology for the implementation of risk assessment and spatial decision support systems for risk management by local authorities in flood and earthquake threatened urban areas in developing countries. The large disaster risk in cities in developing countries is due to a number of factors, such as high vulnerability, lack of resources for proper planning and lack of spatial information in order to make the right decisions. Even if such spatial information is available, it may be dispersed through different local authorities, without an operational procedure for information sharing. In order to be able to effectively take measures on risk reduction, local authorities must be supplied with reliable, up-to-date, and interpreted information on the nature and geographical distribution of hazard and risk, and the possible risk scenario's.
Risk information, presented spatially, is needed by local authorities to take decisions on how to reduce the risk for particular areas, either by reducing the hazard probability (eg. structural measures like dikes) or by reducing the vulnerability (eg. restrictive zoning, building codes). Risk information also forms the basis for a proper emergency response planning.
The project focuses on the two types of natural hazards that cause most damage in urban areas: flooding and earthquakes. The methodology will be developed in a number of case study cities in different countries.

Visualization and Use of Risk Indicator Maps

Within the SLARIM project, one work package deals with the spatial data infrastructure and base data aspects of a flood/seismic risk assessment system, directed towards the needs of medium sized local authorities in developing countries. The activities within this work package include the development of a methodology to handle interoperability aspects. Distributed processing, interoperability research and the impact of the OpenGIS specification will play an important part in this activity. The author of this paper is responsible for “recommendations concerning the methods for the online mapping and presentation of risk indicators”. The main activities include:

  1. the cartographic design of risk indicator maps (RIMs), and
  2. the design of technical concepts of RIM Web services using SVG.

The first requires research into the cartographic grammar for visualisation of real and perceived risks. For this, a theoretical review into perception properties for RIMs is needed and the visual variables best matching these should be determined, combined with a best practices review into the real solutions already in use.
For the second activity, visualisation will be considered as part of a system of interoperable web services. The reason for this is that such a system could support on-line, distributed data from various local authorities, and it can be scalable (ranging from in-office planning to rapid response in the field) and time-aware (supporting comparisons, real-time views and extrapolation into future scenarios). All these requirements have led to the decision to use XML-based solutions in this workpackage, and thus GML for spatial data exchange and SVG for visualisation.
It was decided to try out the concepts developed in the workpackage in a small proof-of-concept implementation. The application should deliver scalable, on-line RIMs to a number of possible clients, including PDA’s. The maps are generated server-side, from on-line distributed data into SVG-based maps offering visualisations of risks as well as access to underlying data.
The implementation of this proof-of-concept application has started is March 2003, this paper will present the results and discuss its conceptual and technical setup.