Exploring the spatial dimension of community-level flood risk perception: a cognitive mapping approach

ABSTRACT

Environmental perceptions are central to individuals’ behavioural interactions with the environment. Cognitive maps, portraying a spatial representation of an individual’s environmental perception, can be aggregated to gain insight into the collective environmental perception of groups and populations. This paper uses cognitive mapping techniques to examine one aspect of environmental perception, flood risk perception, within a residential population (n = 305). Flood risk perception was examined for the whole sample and six subgroup pairs. Using subgroups allowed examination of how factors previously shown to influence flood risk perception influence the cognitive map production in this population. We use a novel technique (slope analysis) to examine how the population’s perception of flood risk compares with expert assessments of flood risk, and compare the results of this novel technique with a commonly used cognitive map analysis technique (majority threshold method). Both methods identify areas where there is consensus within the population as to which areas are at risk of flooding. However, slope analysis usefully identifies areas where the population’s perception of flood risk lacks consensus, and is at odds with expert assessments of flood risk, without the loss of information inherent in the majority threshold method. Thus, this technique provides a novel approach to studies of environmental perception that can be widely applied within many fields.

KEYWORDS:

• Cognitive maps
• risk perception
• boundary analysis
• flooding

Acknowledgements

This work could not have been accomplished without the extensive manual and digital processing input of Richard Geoghegan and Sean Judge. The authors are very grateful for their assistance. Additionally, the authors would very much like to thank Dr Alex Hagen-Zanker of the University of Surrey for his expertise on the use of Fuzzy Kappa statistics. His input is greatly appreciated.

Disclosure statement

No potential conflict of interest was reported by the authors.

ORCiD

Michael Brennan http://orcid.org/0000-0002-5646-028X

Eoin O’Neill http://orcid.org/0000-0003-3476-161X

Finbarr Brereton http://orcid.org/0000-0002-7040-0128

Ilda Dreoni http://orcid.org/0000-0001-8420-522X

Harutyun Shahumyan http://orcid.org/0000-0001-6247-0954

Notes

1. Vector data are representations of the world using points, lines, and polygons. They are useful for storing data that have discrete boundaries, such as country borders, land parcels and streets.

2. Rasters consist of matrices of pixels (called cells) arranged in a grid where each cell contains a value representing information, for example, elevation. Examples include digital photographs, satellite imagery or scanned maps (www.esri.com).

3. Hurricane Charley is the official designation assigned by the World Meteorological Organisation. However, it is referenced locally as Hurricane Charlie.

4. The scheme construction is expected to be completed during 2016.

5. Georeferening an image means to establish its location in terms of map projections or coordinate systems (Hackeloeer et al., 2014).

6. http://www.rte.ie/archives/2013/0826/470307-one-of-the-worst-hurricanes-in-living-memory/

Additional information

Funding

This work formed part of the FLOODPAP project and was supported by UCD Planning & Environmental Policy.