Predictions in public: understanding the design, communication and dissemination of predictive maps to the public - Work Package 5 First national community survey

Research in Australia has been conducted on the public’s response to risk and warning communication (Dootson et al. 2019, 2021). However, less research effort has focused exclusively on maps, and even less has focused on fire spread prediction maps and the Australian context. The purpose of the research reported here is to assess the extent to which community members use, comprehend, perceive, and act upon maps, including fire spread prediction maps in bushfires. A secondary purpose is to offer jurisdiction-specific feedback to fire agencies on community engagement with their current map products to inform their respective communication strategies and map design. In collaboration with a project Steering Committee, maps currently in use across the states and territories of Australia were tested in a nationwide survey (N = 3007). The maps showed a hypothetical bushfire scenario. The sample included over 52% female respondents, with over 51% aged 18 to 44 years old. The sociodemographic characteristics of the sample are provided in Table 1 on page 9. In the overall sample, approximately 11% of the respondents indicated that someone in their household was a member of a state emergency service agency. Just over 42% indicated that they had previously experienced a bushfire, with 34% indicating experience within the past five years. A summary of the results is provided below. The intended audience for this report is the project team, the Steering Committee, and agency people involved in map production and disseminating public information and warnings.

Preferred, trusted sources and platforms. Respondents indicated they typically received information about bushfires from the formal fire agency (e.g., QFES, NSW RFS, ACT ESA, CFA VIC, TFS, SA CFS, WA DFES, NTFRS), local government, state government, Bureau of Meteorology, and media. However, when asked about who a trusted source of bushfire information was, media was usually replaced with police services or family and friends.

Comprehension. Respondents for the most part comprehended the purpose of the maps and associated warning messages, as well as the intended prompted action. There were some points of confusion, including when the map contained multiple polygons and warning levels, when the respondent was placed outside the polygon, and when it was hard to locate oneself on the map due to the design of the map (e.g., too many roads and no road names; or the map was too bare, with limited geographical information). Moreover, there was moderate-high self-reported perceived map effectiveness, such as that the maps and associated warning messages were worth remembering, grabbed attention, and were powerful, informative, or meaningful (Davis et al., 2017; Dillard et al., 2007).

Risk perceptions and emotions. Both risk perceptions and negative emotions, overall, tended to increase when the map and associated warning message were visualising and describing a higher level of threat escalation, for example, Map 1 may have been a ‘Watch and Act’ level of bushfire escalation whereas Map 2 may have then been an ‘Emergency Warning’ level of bushfire escalation, a higher level of warning in the national Australian Warning System. Map 2 would thus be associated with stronger self-reported risk perceptions and negative emotions, congruent with the higher level of warning escalation.

Protective action intentions. The national sample was largely compliant with the agency-issued instructions (usually stated in the associated warning message). While there were cases of respondents indicating that they would do specific protective actions implied but not explicitly mentioned by the warning message, these were usually aligned with protective behaviour. There were instances, however, where respondents indicated that their top five protective action intentions would include an action that could potentially put the individual (or their property) at risk (e.g., waiting for a firefighter to tell them to evacuate, waiting for police to knock on the door).

Coping appraisal. The maps and associated warning messages had two impacts on coping appraisal, either: (1) the map and associated warning message saw an increased coping appraisal from the initial self-reported general perceived coping appraisal of responding to bushfires; or (2) the more complex, higher escalation of warning (e.g., Emergency Warning) saw a higher coping appraisal assessment by the respondent than the simple map of lower escalation of warning (e.g., Watch and Act). Pairing a high coping appraisal with high-risk perceptions is helpful to incite protective action over maladaptive decision-making (e.g., wishful thinking, denial, inaction) as per protective motivation theory (Rogers, 1975).

Feedback. Common feedback across the whole sample of respondents on the maps included calls for:

• Directionality of hazard: use arrows showing directionality of the bushfire spread.
• Legend: use a legend or key to help interpret the information presented on the map.
• Timing: indicate when the map was developed and for how long it is valid; time estimates on how fast it is tracking.
• Landmarks: show on the map key landmarks, including evacuation centres, to help people locate themselves on the map.
• Routes: show clear routes out to help people navigate their evacuation.
• Interactive capabilities: provide the ability to engage with the map directly, i.e., zoom functionality.
• Sizing and legibility: increase the prominence of hazard markers alongside placenames, roadways, and other landmarks.
• Colours: increase contrast of colours to clearly delineate multiple warning areas.

The results from this study combine with the other work packages in the Predictions in Public research program to cumulatively underpin the future design of maps for use in the public information and warnings milieu in Australia, under the Australian Warning System. This report should not be read in isolation to other work packages.