View the PDF here
Exploring ‘wait and see’ responses in French and Australian WUI wildfire emergencies
By: Sandra Vaiciulyte (Departamento de Sismología, Instituto de Geofísica, Universidad Nacional Autonoma de Mexico, Col. Ciudad Universitaria, Alcaldía Coyoacan, Ciudad de Mexico 04510, Mexico)
This article is the short version of the published paper “Exploring ‘wait and see’ responses in French and Australian WUI wildfire emergencies”.
Vaiciulyte, S., Galea, R. E., Veeraswamy, A., Hulse, L. M., (2022). Exploring ‘wait and see’ responses in French and Australian WUI wildfire emergencies. Safety Science Journal. Volume 155, November 2022, 105866 https://doi.org/10.1016/j.ssci.2022.105866
Introduction
In the event of wildfire, a timely response upon the receipt of the first fire cues is vital to human survival in wildfire-prone urban areas. From the past wildfire events in Australia, for example, evidence has shown that residents often ‘wait and see’ how the situation unfolds before taking protective action, such as evacuation or shelter-in-place [1]. Such decision-making delay has in the past been detrimental to people’s survival [2], thus advising largely against a ‘wait and see’ strategy [3]. Yet, media reports from as recent as Victoria’s 2019–2020 bushfire season reveal that some people are still willing to ‘wait and see’, even after receiving an evacuation order, and could be tempted to do so even more as the COVID-19 pandemic has seen many people self-isolate and be reluctant to leave [4;5]. Importantly, the ‘wait and see’ responses in populations reactions to wildfires have received research attention in Australia and North America. However, it is unclear whether the findings extend to European regions, given the scarcity of such research there.
Thus, research is needed to better understand the circumstances under which a ‘wait and see’ response manifests during wildfires, the circumstances under which this response diminishes, or where it is unlikely to occur. Such knowledge would improve wildfire emergency response planning (illustrated in recent wildfires in Greece, see [6]), including support the use of urban-scale evacuation modelling [7;8], simulating population responses beyond the binary ‘stay’ or ‘go’.
Influence in decision-making
Several factors have been found across the literature to impact ‘wait and see’ responses, namely demographic characteristics of residents (e.g. gender, age, medical impairments, etc.); differences in risk perception in relation to culture and other contextual factors [9]; contextual factors such as a region’s official wildfire policy, by showing a preference for one type of protective action [10].
Different wildfire policies exist across geographies, as for example, in the South of France where ‘shelter-in-place is preferred over evacuation. Moreover, residents are accustomed to rely upon instruction from authorities during an incident [11]. Therefore, it is possible that relatively more residents in South of France would choose to ‘wait and see’ upon a wildfire threat. Nonetheless, little research exists on the wildfire-related behaviours of European populations [12], and no research has investigated ‘wait and see’ responses in Europe.
The current research effort specifically looked at risk from wildfire populations in South of France and Australia and with the help of two kinds of questionnaires: actual experience and hypothetical wildfire, assessed the participants’ ‘wait and see’ behaviors generally, as well as under circumstances such as social cues (media notifications, seeing other people evacuate), environmental cues (smoke, embers, fire), and a combination of the two.
Key findings
‘Wait and see’ responses under different circumstances
There were several findings pertaining to ‘wait and see’ behaviors in Australia and South of France.
Firstly, on who was more or less likely to ‘wait and see’: Australian participants in a hypothetical wildfire intending to wait were in line with the summarised findings from literature (in [13]), i.e. between 3% to 32%. Similarly, Australian participants who actually experienced a wildfire, reported waiting closer to the lower end of the previously reported results (25% to 58% – see [13]). In contrast, there were more participants who reported intending to wait or actually waiting in South of France, but far lower than that reported previously for American communities (intending = approximately 71%, actually waiting = between 54% and approximately 85%) [14;15]. These differences show that while past research from outside Europe has been important in highlighting a global problem, those findings cannot be wholly generalised to other parts of the world. Also, Australian participants’ more decisive behaviour was likely influenced in part by an awareness of the tragic 2009 Black Saturday bushfires – an event of a magnitude that France has not experienced this century, and so the French population would likely have less media exposure to information regarding the negative effects of ‘wait and see’ responses.
Secondly, our findings show that different circumstances, such as social, environmental cues and the mixture of the two also influence the ‘wait and see’ decision. For example, the receipt of just social cues was not found to have a decisive effect on all participants’ decision-making (e.g., wildfire notification via the media, and unofficial cues such as seeing neighbours’ reactions). However, official cues in the form of a direct order to evacuate did appear to reduce waiting, as has been noted elsewhere (e.g. [14]). As for the environmental cues, they were more effective in reducing intentions to wait, both when presented alone and in combination with social cues. However, influence of environmental cues was nuanced. The results suggested that, in both the hypothetical and real-life situations, participants were not entirely sure what to do in response to seeing smoke, while more decisiveness was usually displayed in response to seeing embers and certainly in response to seeing flames. So, practitioners should be made aware that their expectations are not necessarily reflected in the manner in which the public is likely to respond. On the other hand, embers and flames are indicators of a fire being in very close proximity, meaning a potential danger to life and property and very little time to act.
Thirdly, pre-event risk perception did not appear to uniformly influence participants’ decision-making, and neither did preparedness (i.e. having a plan). Nonetheless, having insurance was significantly associated with taking action rather than waiting, but only for the South of France sample with actual wildfire experience. It could be that having appropriate insurance cover means that the financial cost of losing property is covered. As such, people’s priorities could be less conflicted during a fire, and attention could be immediately turned to acting to protect self and family from harm. Thinking ahead about one’s evacuation destination was significantly associated with less waiting, but only in the Australian hypothetical wildfire sample. There, more participants planned to go to a nearby town during evacuation rather than escape to closer locations such as a public hall in their own town.
Finally, when it comes to socio-demographic factors, there was no overlap between the two study regions or experience samples, suggesting that there could be both regional differences and other ways in which wildfire experiences shape diverse populations’ responses. Age being an important factor in the Australian sample for a hypothetical fire can potentially indicate that younger and older adults, who have not yet experienced a wildfire and might be prone to reacting slowly, perhaps because they feel less able to make a critical decision or take action for themselves in a situation without any sign of support or involvement from authority figures. Therefore, interventions targeting different age groups could be formed more intentionally. For the South of France participants with actual experience, having pets/livestock was significantly associated with choosing to take action, as was having dependents, indicating the importance of livestock and dependents on evacuation decision-making (also reported in [16;17]).
Study implications
The findings of this study have implications for practitioners and policy makers. They call for their attention when designing effective communication during wildfire emergencies, highlighting a need to ensure instructions are clear about action to be taken, authoritative, and incentivising. They also call for attention when designing preparedness initiatives for wildfire events, especially in at-risk areas where populations may be largely inexperienced. Preparedness should focus on assisting residents to swiftly and correctly interpret environmental cues such as smoke, or identify physiological symptoms and their implications for the ability to carry out protective action safely. Residents should also be helped to identify safe refuges within their locality and alternate routes to reach the safe locations, which can be factored into fire plans. Additionally, more consideration should be given as to how to reach out to residents in their homes – not just residents deemed vulnerable, such as those living with young children or the elderly, but also those without dependents. Educational safety campaigns may be effective but need to capture interest and are likely to be conducted for a limited time. Thus, policy makers should seek to identify and collaborate with alternative information sources such as insurance companies – their marketing campaigns and policies are likely to be viewed by more people and more frequently when policies need renewing, and these could contain reminders about the benefits of wildfire preparedness.
The observed effects of environmental and social cues on ‘wait and see’ responses also have implications for urban-scale evacuation modelling. Firstly, the differences in the extent of waiting seen in France compared to Australia (and the USA) mean caution should be exercised when applying research findings to simulation models across geographies, even when the models are capable of simulating such environmental and social effects. Secondly, there is a potential for over-reliance on the effects of social cues in initiating a protective action response, as it seems that some people will ‘wait and see’ under certain official and unofficial social cues. Thus, developers of evacuation models should consider the integration of hazard data within their models, to enable scenarios where evacuation is triggered by different progressing environmental cues in addition to social cues. Finally, for evacuation simulations to more accurately reflect reality, it is important to consider that response behaviour of individuals with experience of wildfires and those without.
Future research efforts
The study confirms the urgent need to conduct further human behaviour research in European contexts, based on the evidence that South of France participants exhibited more prominent ‘wait and see’ behaviors. To enable this, policy makers should commission and use local research. This will support evidence-based planning, which in turn should help authorities direct resources to where they may have the greatest effect in minimising waiting.
References
[1] Strahan, K., Gilbert, J., 2021. The Wait and See Literature: A Rapid Systematic Review. Fire 4, 4. https://doi.org/10.3390/fire4010004.
[2] Handmer, J., O’Neill, S., 2016. Examining bushfire policy in action: preparedness and behaviour in the 2009 Black Saturday fires. Environ. Sci. Policy 63, 55–62. https://doi.org/10.1016/j.envsci.2016.05.011.
[3] Emergency Management Victoria, 2016. Victorian Bushfire Handbook.
[4] Chan, B., Middleton, J., 2019. More than 30,000 tourists in Australia who defied orders to leave area east of Melbourne amid raging bushfires are ’trapped’ as roads are cut off, after; locals warned staying would be ’suicide’. Daily Mail Australia, Mail online, and AFP [online]. Available at https://www.dailymail.co.uk/news/article-7835 431/Tourists-REFUSE-evacuate-Australian-wine-region-bushfires-cut-road-out.html [Accessed 01 June 2020].
[5] Hollingsworth, J., 2021. First came Covid lockdown. Now a bushfire is forcing these Australians to evacuate. CNN [online]. Available at: https://edition.cnn.com/2021/02/03/australia/perth-evacuation-fires-intl-hnk/index.html [Accessed 30.08.2021].
[6] Smith, H., 2021. Firefighters battling wildfires to save homes on Greek island of Evia. The Guardian [online]. Available at: https://www.theguardian.com/world/2021/aug /10/firefighters-battling-wildfires-to-save-homes-on-greek-island-of-evia?fbclid=IwAR3SOT4mR4LcGkTw_S51QaQilWJZt6Sci8CF96ih60uA0n5q3hKmNtAQ9ME [Accessed: 16 November 2021].
[7] Veeraswamy, A., Galea, E.R., Filippidis, L., Lawrence, P.J., Haasanen, S., Gazzard, R.J., Smith, T.E.L., 2018. The simulation of urban-scale evacuation scenarios with application to the Swinley forest fire. Safe. Sci. 102, 178–193. https://doi.org/ 10.1016/j.ssci.2017.07.015.
[8] Wahlqvist, J., Ronchi, E., Gwynne, S.M.V., Kinateder, M., Rein, G., Mitchell, H., B ́enichou, N., Ma, C., Kimball, A., Kuligowski, E., 2021. The simulation of wildland- urban interface fire evacuation: The WUI-NITY platform. Saf. Sci. 136, 0925–7535. https://doi.org/10.1016/j.ssci.2020.105145.
[9] Hitchcock, J.L., 2001. Gender differences in risk perception: broadening the contexts. Risk: Health Saf. Environ. 12 (3), 179–204.
[10] Hudson, B., Hunter, D., Peckham, S., 2019. Policy failure and the policy-implementation gap: can policy support programs help? Policy Des. Pract. 2 (1), 1–14. https://doi.org/10.1080/25741292.2018.1540378.
[11] Ministère de la Transition écologique, 2020. Feux de forêt : préservons la végétation face aux incendies [online]. Available at: https://www.ecologie.gouv.fr/campagne-fe ux-forets.
[12] Vaiciulyte, S., Galea, E.R., Veeraswamy, A., Hulse, L.M., 2019. Island vulnerability and resilience to wildfires: a case study of Corsica. Int. J. Disaster Risk Reduct. 40, 101272. https://doi.org/10.1016/j.ijdrr.2019.101272.
[13] McLennan, J., Elliott, G., 2013. “Wait and See”: The Elephant in the Community Bushfire Safety Room; Proceedings of Bushfire CRC & AFAC 2012 Conference Research Forum, (August 2012), 1–184. Retrieved from http://www.bushfirecrc.com/sites/de fault/files/managed/resource/2012_research_forum_proceedings_final_0.pdf#page =135.
[14] McCaffrey, S., Wilson, R., Konar, A., 2018. Should I stay or should I go now? Or should I wait and see? Influences on wildfire evacuation decisions. Risk Anal. 38 (7), 1390–1404.
[15] Paveglio, T., Prato, T., Dalenberg, D., Venn, T., 2014. Understanding evacuation preferences and wildfire mitigations among Northwest Montana residents. Int. J. Wildland Fire 23 (3), 435–444. https://doi.org/10.1071/WF13057.
[16] Kuligowski, E., 2020. Evacuation decision-making and behavior in wildfires: Past research, current challenges and a future research agenda. Fire Saf. J. 120, 103129. https://doi.org/10.1016/j.firesaf.2020.103129.
[17] McNeill, I.M., Dunlop, P.D., Skinner, T.C., Morrison, D.L., 2016. A value- and expectancy-based approach to understanding residents’ intended response to a wildfire threat. Int. J. Wildland Fire 25 (4), 378–389. https://doi.org/10.1071/ WF15051.