Towards Cartographic Support for Risk Assessments of Civil Infrastructures
ISBN 978-85-88783-11-9
Authors
1Heitzler, M.; 2Iosifescu-enescu, I.; 4Adey, B.T.; 5Hackl, J.; 5Hurni, L.
1ETH ZURICH, INST OF CARTOGRAPHY & GEOINFORMATION Email: hmagnus@ethz.ch
2ETH ZURICH, INST OF CARTOGRAPHY & GEOINFORMATION Email: iosifescu@ethz.ch
4ETH ZURICH, INST OF CONSTRUCTION & INFRASTRUCTURE MANAGEMENT Email: adey@ibi.baug.ethz.ch
5ETH ZURICH, INST OF CONSTRUCTION & INFRASTRUCTURE MANAGEMENT Email: hackl@ibi.baug.ethz.ch
5ETH ZURICH, INST OF CARTOGRAPHY & GEOINFORMATION Email: lhurni@ethz.ch
Abstract
Recent developments in civil engineering aim at incorporating dynamic models into risk assessment processes in order to simulate the consequences that arise from the impact of natural hazards on infrastructures. The results of these assessments help infrastructure managers to reduce risk by helping them determine and implement suitable treatment strategies. The simulation of the behaviour of several interacting systems, however, such as river systems, infrastructure systems and the society, which is required in such risk assessment processes, results in large amounts of heterogeneous datasets with multivariate, multidimensional (i.e. spatial and temporal) and often abstract characteristics. This makes it useful to have effective visualization and analysis methods that enable stakeholders and analysts to make sense of these data. There are three ways in particular that cartographic concepts could be beneficial. First, in order to efficiently interpret risk assessment data, appropriate cartographic visualizations need to be generated that take into account the special characteristics of these data. Examples for a road network that may be affected by natural hazards are 1) the amount of physical damage that an infrastructure object may incur during the event, 2) the estimated reconstruction costs following the event, and 3) the consequences that can be related to the distribution to traffic flow due to a disturbed infrastructure network. The first two measures can for example be directly encoded into color of the corresponding geometries. However, consequences can come in different shapes and therefore need individual considerations. For example, the consequence of increased travel time to hospitals can be displayed in the form of colored polygons overlying the road network. Second, comparative visualizations should be provided that allow to compare different states of infrastructure networks based on varying initial simulation conditions. For example, strategies to reduce risks can include modifications to infrastructure networks (e.g. increasing the robustness of a bridge or adding more scour protection). An estimation of the benefits of including a particular intervention in such a strategy, therefore, requires a comparison of the performance of the network with and without the intervention. Another example would be to compare the difference in consequences that are related to the maximum and minimum threshold of an expected flood intensity. In particular, multiple aligned maps (i.e. depicting system states on separate maps) and difference maps (i.e. depicting different system states within the same map) are suggested. Third, in order to understand the behaviour of the infrastructure system, suitable navigation methods need to be provided that take into account the possibility of multiple alternative timelines due to changes before or during a simulation. Existing navigation techniques for this purpose are therefore extended so that they additionally represent the states of multiple systems. To exemplify these concepts, they are applied to data originating from the implementation of a civil engineering methodology for risk assessment of civil infrastructure for the region of Chur, Switzerland.