Who Does What? Shared Responsibility for Wildfire Management and the Imperative of Public Engagement: Evidence from Whistler, Western Canada

Abstract

In Canada and elsewhere, there is an ascendancy of a whole-of-society approach that centres shared responsibility for wildfire management. This article engages the debates on the rise of shared responsibility for wildfire management to argue that this context demands a renewed research focus on understanding how the public allocates responsibility for wildfire management. We illustrate this argument through a case study of public engagement with wildfire risk and shared responsibility in Whistler, British Columbia, western Canada. Our case study draws on evidence from a quantitative survey administered to 1311 participants in the spring and summer of 2024. The study reveals a near-universal concern about wildfires among the participants and a high level of risk perception. This is consistent with community climate and wildfire reports and plans. This level of concern is driving a high level of mitigation activity completion among participants, even though the level of preparedness is mixed. Our study found a marked pattern of responsibility allocation across the phases of wildfire management. Participants put the municipal government at the forefront of mitigation, preparedness, and response. The provincial government was ranked as most responsible for recovery. Homeowner responsibility declined as one moves from mitigation and preparedness through to response and recovery. Private actors, such as insurance, have greater responsibility in the recovery phase. Multivariate General Linear Models (GLMs) show that how respondents allocate responsibility for various aspects of wildfire management is influenced by home ownership, prior wildfire experience, perceived preparedness, and commitment to bearing the costs of FireSmart assessment. We conclude that a sustained research commitment is needed to further elucidate the dynamics of public expectations and attitudes in the context of shared responsibility for wildfire management.

1. Introduction

In Canada and across North America, concern about the wildfire ‘problem’ is growing, as climate change intersects with fast-expanding wildland–urban interface areas (WUIs), increased flammability of fire-deficit landscapes and perennial institutional challenges [1,2,3,4,5,6,7,8,9]. The 2023 wildfire season set a new historical record of area burnt by wildfires in Canada [10]. Just as insured losses from wildfire disasters continue to break insurance records, so too has the public cost of wildfire protection [11]. Governments and wildfire managers are increasingly seeking to limit financial exposure while promoting a collaborative approach that mobilizes the whole of society around wildfire management. Such an approach is partly driven by the assumption that those who benefit from wildfire protection also share in the cost, thereby shifting risk behaviour accordingly. In this context, the need to continually assess and understand patterns of public attitudes toward wildfire management becomes critical for shared-responsibility policies and programs to take hold effectively and equitably.

This paper examines public engagement with wildfire risk and management within a whole-of-society approach in Whistler, a resort town in British Columbia, Canada. Engaging with debates on shared responsibility for wildfire management and disaster risk reduction, the paper traces the critical imperative for a renewed and sustained focus on understanding public perceptions and expectations in this context. Drawing on a multi-modal survey administered to 1311 residents of Whistler, the paper reveals near-universal concern about wildfires among the participants and a high level of risk perception, which is linked to a high level of mitigation activity completion, even though the level of preparedness is mixed. We found a clear pattern of responsibility allocation across the phases of wildfire management, influenced by factors such as home ownership, prior wildfire experience, perceived preparedness, and commitment to bearing the costs of FireSmart assessment.

This paper thus makes two main contributions, the first theoretical and the second empirical. First, the paper argues that the ascendancy of the principle of shared responsibility in wildfire management creates critical imperatives for renewed focus on public engagement research to understand how the public allocates responsibility for wildfire management. Amidst the shift towards a form of wildfire governance that centres pluralities of actors, tools, and objectives [3,6,8,12], an expanded scope emerges for public participation in wildfire governance and management. Yet not only the multiplicity of expectations among the plurality of actors that converge around wildfire management, but also the dynamic influence of climate change and the shifting socio-techno-political context of wildfire management mean that shared responsibility is a matter of continual negotiation, which ought to be informed by a robust, iterative understanding of public perception, attitudes and expectations. In short, the paper suggests that effective and equitable wildfire management, guided by the principle of shared responsibility, would increasingly depend on mutually explicit attitudes and expectations among the public, the state, and all wildfire management stakeholders.

Second, the paper contributes to the growing scholarly insights into public attitude to wildfire risk and management. More crucially, through the case of Whistler, the paper generates novel insights into the dynamics of shared responsibility, which the literature describes as largely unclear in practice [13]. By generating a responsibility ranking for wildfire management stakeholders across the four phases of fire management—mitigation, preparedness, response, and recovery—our study generates insights into the less-studied issues of public allocation of responsibility for wildfire management. The paper also identified important factors that influence the pattern of responsibility allocation. Together, these insights have vital policy and program relevance for the case study community, Whistler. It also holds implications for the discourse and debates of shared responsibility across many other wildfire contexts in North America and beyond.

The rest of the paper is organized as follows: In Section 2, we present an overview of the debate on the rise of the shared responsibility principle in wildfire management in Canada and beyond, drawing attention to the imperative this creates for renewed and sustained efforts to render explicit public attitudes and expectations around wildfire management. Next, we review studies on public engagement with wildfires. We then present background details about the study area and outline the methods employed to collect and analyze data for the research. We present and discuss the findings of the research under four main headings: risk perception, level of preparedness, level of completion of mitigation tasks, and allocation of responsibility. In Section 7, we draw conclusions from the study.

2. Shared Responsibility and the Imperative of Public Engagement

A shift towards shared responsibility increasingly marks the governance of wildfire risk in Canada and across North America. Tymstra ([12], p. xxviii; also pp. 146–157) describes this shift in terms of ‘… a paradigm shift from reactively fighting fire to proactively managing fire and the fire environment … [noting that] this shift means accepting more managed fire on the landscape and strengthening defensive positions by making communities and other values-at-risk FireSmart’ [3,6,8,14]. Canadian ‘FireSmart’, like ‘FireWise’ in the US, emphasizes the principle of risk-sharing and shared responsibility for wildfire protection and disaster risk reduction more generally [12,15]. In Australia, too, debates have been ongoing about shared responsibility for wildfire management and disaster reduction [16,17,18,19,20,21,22,23].

This increased emphasis on shared responsibility underpinned by a whole-of-society approach to wildfire disaster risk reduction reflects the requirements of international frameworks such as the Hyogo Framework for Action and the Sendai Framework for Disaster Risk Reduction [21,24]. These frameworks recognize that effective risk reduction and disaster management require a collaborative approach that draws on the capacities of all stakeholders [25,26]. Across both the public and private sectors, risk managers and policymakers are increasingly emphasizing the need for those who benefit from risk reduction measures to share the costs of these measures so as to shift risk behaviour accordingly. This is particularly so at a time when the cost of wildfire-related disaster risk management and the cost of insured damages from disasters, including wildfires, are at record highs [11,27]. Yet, this growing emphasis on shared responsibility is not unproblematic [16].

Despite this emphasis, who takes what responsibility and how are vital questions that continue to be unclear and contested in practice, whether in Canada [13] or elsewhere [16,17,18,21]. Indeed, academic studies and post-disaster inquiries frequently point to a disjuncture between public expectations and institutional expectations of shared responsibility for wildfire security in Canada ([12,28,29,30,31]; see also [25,26,32] on flood risk) and elsewhere [17,18,21,23,24,33,34]. Not only the high stakes of wildfire disasters but also the complex dynamics of wildfire socio-ecology, the history of wildfire management, the dynamic nature of the wildland–urban interface, and the ‘shifting goalpost of global environmental change’ all make shared responsibility for wildfire security a subject of continual negotiation across time and space ([35], p. 2; see also [17,21,23,34,36]). As such, not only are expectations surrounding shared responsibility dynamic, but these shifting expectations would need to be repeatedly laid bare for the possibility of bridging the expectation gap between the public and wildfire managers. In other words, effectively and equitably sharing responsibility for wildfire management requires a continuous commitment to mutually explicit expectations.

Expectations of wildfire managers and policymakers often reflect what’s enshrined in laws, policies, plans and other statutory documents [16]. These expectations routinely get disseminated through public engagement programs, awareness campaigns, and the media. As such, not only are the expectations of wildfire managers and policymakers readily explicit, but these expectations are also often privileged as vital starting points for the negotiation of rights, responsibilities, and obligations in wildfire management [17,34]. However, as post-disaster studies and institutional reviews often reveal, public expectations become most visible only after wildfire disasters have occurred. In these contexts, public expectations are often revealed as incongruent with those of wildfire managers, thus becoming a source of public frustration and tension. Too late, this heightened visibility of public expectations post-disaster does little to shape disaster risk reduction efforts, even if it contributes to overall institutional accountability and public discourse. The role that such a display of public expectations might play in addressing future risks is limited by the dynamic and shifting nature of wildfire risks and associated expectations around shared responsibility.

This situation presents a critical imperative for renewed and sustained effort to understand the patterns and dynamics of public expectations. Such an effort to proactively assess and monitor shifting public attitudes and expectations is vital for effectively and equitably sharing responsibility for wildfire management with the public. Given the well-established behavioural literature on public attitudes, what sets this new imperative apart? At least three conditions underpin this imperative. First, public attitude studies become radically temporally specific, given the dynamics of risk and expectations. This means that public attitude studies need to be more carefully contextualized in place and time, so that findings are always tentative and anchored to prevailing social, economic, cultural and political and environmental conditions (cf. [36,37,38]). This also means that monitoring of shifting attitudes through longitudinal studies becomes important. Here, determining the intervals of longitudinal series would require balancing how fast risk and expectations are shifting with the consideration of research fatigue among the populations being studied, and the constraints of research resources and time.

Second, public attitudinal studies need to frame public responses comprehensively to include ‘expectations’. This is important since the self-reported responses of the public to attitudinal studies also reflect, at least in part, public expectations. Such an inclusive framing helps situate these studies as part of the broader public discourse surrounding the negotiation of shared responsibility. Third, public perception studies also need to be more cognizant of and embrace the politics of public engagement. Since these studies are not neutral in their effect on wildfire management, researchers need to be more reflexive about the role their research plays in creating more equitable wildfire management systems. So too, research needs to more meaningfully engage local communities and partners, including NGOs, activist groups, local governments and other stakeholders in the communities where research is taking place. Researchers themselves must grapple with their often-complicated positionality within increasingly consequential shared responsibility spaces. The study presented in this paper thus responds to some of these imperatives while building on insights from existing literature on public attitudes to wildfires.

3. Insights from Research on Public Wildfire Engagement

Researchers continue to highlight the vital importance of social aspects, including public behaviour, attitudes and engagement, for effective wildfire management [30,39,40,41,42]. Although lagging compared to research on fire behaviour and ecologies [39], social science research studying public attitudes to wildfire risk and management has grown rapidly over the last two decades, generating significant insights into the social factors affecting wildfire management and fire resilience [30,33,37,41,43,44]. Factors such as perception of wildfire risk, prior experience of wildfires, perception of mitigation efficacy, levels of preparedness, access to funds, and access to information, among many others, have been identified as shaping patterns of public engagement with wildfire management.

In Canada, a growing number of studies continue to document patterns of public engagement with wildfire risk and management. Some of these have focused on public attitudes to wildfires in general. For instance, a qualitative study of residents of Valemount, British Columbia, by Canosa et al. [39] found that participants were more concerned about the indirect impacts of wildfires, such as health effects of wildfire smoke and power outages, even though the direct risk of wildfire was increasing in their community. They found that previous experiences with these wildfire impacts, dependence on public power service and a limited transport network galvanized public action around wildfires in this community. In their study of disaster (including wildfire) awareness and preparedness among older adults in Canada, Bogdan et al. [45] found that while risk perception is generally low among this population due to barriers such as hazard vulnerability misperceptions and cost-related reasons, stronger risk perception is linked to risk specific to geographic location where respondents live. In their survey of community leaders across British Columbia, Copes-Gerbitz et al. [13] found that proactive wildfire management measures, such as enforcement of regulations and education, which were considered effective and had minimal impacts on community values, were also the most supported by respondents. They found engagement with proactive wildfire management was limited by financial and social (time and expertise) barriers, which exist unevenly across community groups.

A relatively older set of studies laid the foundation for understanding patterns of public wildfire engagement in Canada. For instance, Cote and McGee [46] studied the intended wildfire evacuation responses among residents of Mt. Lorne, Yukon. They found that respondents planned to stay on their property despite a wildfire evacuation order as they perceived their property as safe and because of concerns about the impacts of evacuation on everyday life, although some lacked the knowledge required for them to stay in their property safely. In their study of Peavine Métis Settlement in Alberta, Christianson et al. [47] found that local leadership, economic opportunities, community capacity and land and home ownership all affected whether the public supported wildfire mitigation programs.

In their survey of six WUI communities in Alberta, McFarlane et al. [30] found a high level of completion of most mitigation activities. They found that the level of perceived threat, perceived efficacy of mitigation actions, and not lack of funds, have the most effect on respondents’ mitigation actions. The same research also found a significant difference between communities with and without government-led wildfire management activities. They found that respondents who live in communities where wildfire management has taken place have a higher perceived risk and greater awareness of wildfire and mitigation than those who live in communities with no wildfire management activities, even though they did not complete more mitigation activities on their properties [48]. McGee at al. [49] also found in a post-wildfires context in western Canadian communities that differences in how people experienced the wildfire hazard shaped risk perceptions post-event and the adoption of mitigation measures.

Taken together, these studies show that public engagement with wildfire varies across groups and is influenced by many factors, including previous experience, economic factors, locational factors, perceived efficacy of fire management actions, threat assessment, and governance factors, among others. These studies show the importance of demographic and behavioural factors as well as local contextual factors. While some of these studies draw on qualitative evidence, some draw on quantitative evidence and still others employ a mix of these approaches. In the Canadian context, studies using large ‘N’ inferential models are still limited (for an exception, see [30]), despite the importance of such methods for examining the relations among various factors and variables. Our study thus builds on this growing literature by exploring not only patterns of risk and action but also those of responsibility allocation, a critical area of research that requires attention amid the ascendancy of the shared responsibility principle in wildfire management in Canada and beyond.

4. Study Area and Methods

4.1. Study Area

Our research presents a case study of the city of Whistler in British Columbia (see Figure 1). Whistler’s approximately 13,000 residents, tens of thousands of non-resident workers, and 2.5 million annual tourists all contribute to its multibillion-dollar winter and summer tourist economies that depend on surrounding forest landscapes.

Although the fire regime in Whistler, tucked between the coastal maritime rainforests and the drier interior continental forests, is known to have relatively long return intervals and mixed fire severities [50], climate change is likely to alter this fire regime [51]. Whistler is projected to warm by about 3 °C on average by the 2050s compared to the recent past [52]. This is expected to bring longer, hotter and drier summers [52]. Indeed, given the historical pattern of wildfires in the area (the Whistler Forest History Project, led by Don MacLaurin, Peter Ackhurst, John Hammons, who are all registered professional foresters and long-time residents of Whistler, has documented major historical wildfires that have shaped the vegetations surrounding Whistler) and the increasing fire frequency around Whistler (e.g., Downton Lake fire, 2023; Boulder Creek fire 2015, Blackcomb Mountain fire, 2009), the municipal government, the Resort Municipality of Whistler (RMOW), has been taking measures to strengthen the community’s wildfire preparedness and resilience. Community wildfire policy documents point to the extent of, and damage caused by, the Horse River fire in Fort McMurray in 2016 as a warning for Whistler regarding the significant economic investment at risk in a potential fire incident [53].

A community fire hazard assessment conducted in Whistler in 2014 found that 26% of assessed structures had a fire hazard rating of ‘extreme’, and 50% of assessed homes were rated as ‘high’ [53]. The community is surrounded by approximately 4000 ha of high fire-hazard crown land. Also, a ‘large majority of homes within the RMOW on private lands are not FireSmart compliant,’ making them vulnerable to wildfires ([53], p. 11). In this context, Whistler has continued to show leadership in wildfire management, with the municipality and local organizations, such as Cheakamus Community Forest, proactively planning and implementing various strategies to reduce wildfire risk, raise public awareness and collaboratively develop innovative responses to wildfire and other disaster risk challenges [54]. For instance, in 2022, RMOW commissioned and approved its current Community Wildfire Resiliency Plan, which laid out the strategic and operational framework and recommendations for achieving wildfire resilience in Whistler. As part of the implementation of the Plan, the municipality commissioned a Community Wildfire Defence Plan in 2024 that outlines the objectives in pursuit of wildfire policy, emergency planning, and training requirements for Whistler.

The fuel management aspect of Whistler’s fire risk management strategy also includes an elaborate program of ecological and social monitoring to understand the socio-ecological impacts of ongoing landscape fuel treatment interventions. Whistler’s active FireSmart program also engages homeowners in wildfire mitigation activities. In 2024, for instance, the FireSmart Unit facilitated 22 FireSmart Workdays in which 250 homeowners and residents participated, and 352 distinct addresses used the FireSmart Community Chipper Service. About 61 Critical Infrastructure assessments were carried out with 50% of associated landscapes treated by the FireSmart crew. The program also carried out 126 Home Partners Program assessments. Our research was conducted within this context of active wildfire management planning and activities. Our work was supported by wildfire managers in Whistler, although the research process was independently designed and implemented by our research team.

4.2. Methods

The research that informs this article adopts a quantitative method which provides valuable insights into the general patterns of public engagement with wildfire. The total population for this study is the entire 10,065 ‘private dwellings’ in Whistler based on the 2021 census [55]. About half of these private dwellings in Whistler, 5595, are occupied by usual residents, that is permanent residents who are present or temporarily absent [55]. The remaining half of the ‘private dwellings’ house temporary residents, including tourists. For this reason, the population was oversampled to 1311. This sample size also takes into consideration the typical response rate for mailed survey. In their study, McFarlane et al. [30] reported a 38% response rate, while Oulahen [56] reported a 42% response rate using a hand-delivered survey.

We employed a two-stage systematic sampling process to ensure a representative spread of sampled residences across Whistler. In large populations, systematic sampling provides a more even distribution of the sample than simple random sampling can achieve [57]. Using the 2024 Whistler civil address map and a list of Whistler’s 297 residential streets, the first stage involved choosing every fourth street, and the second stage involved choosing every second residence. In total, 1311 residences were selected and sent an invitation to complete the survey online. The invitation was sent out in April 2024. We sent a follow-up letter after six weeks, with a paper copy of the survey included this time, to increase participation. By October 2024, when we ended the survey, it had received 234 responses in total (197 permanent residents and 36 temporary residents), for a response rate of approximately 18%. It’s important to note that the Jasper Fires were ranging over the British Columbia–Alberta border in Jasper National Park in July 2024 as this research was being conducted. We noticed a small spike in survey response rates in the days following the Jasper Fires. For More on the Jasper Wildfire, see NRCan [58].

The survey was carefully designed to include 62 questions, combining open-ended and closed-ended questions. The questions were structured along several themes that were central to the research objectives, including demographic and residency information, wildfire risk perception, level of wildfire preparedness, level of completion of mitigation activities, and perceived allocation of responsibility for wildfire management. The survey was multi-modal, using both an online format (Qualtrics) and a paper format. The primary means of survey delivery was by mail. Mail surveys have many advantages, including good access to sampled participants, good response rates, limited cost, and good quality of answers since respondents have the time and the flexibility to pace their response to the survey while they seek the information that they need to complete the survey. The paper-based survey responses were manually entered into Qualtrics. All responses were auto-coded in Qualtrics and analyzed using SPSS version 31.

The research received ethics approval from the University of British Columbia’s institutional research ethics board. The research posed minimal risk to the respondents. Aside from the time required to participate, no additional risk was anticipated beyond that experienced in daily life. Throughout the research process, all members of the research team approached all research activities reflexively and with a high standard of research practice. Findings from the research were presented at a community workshop in Whistler in May 2025.

4.3. Demographic and Residential Characteristics

A summary of respondents’ demographic and residential characteristics is presented in

Table 1. Respondents’ demographic and residential characteristics.

Demographic and Residential Characteristics		Count	Percent
What is your age range? (n = 191)	Below 18	0	0.0%
	18–34	5	2.6%
	35–44	24	12.6%
	45–54	38	19.9%
	55–64	50	26.2%
	65 and above	74	38.7%
What is your gender? (n = 191)	Male	93	48.7%
	Female	94	49.2%
	Non-binary/third gender	0	0.0%
	Prefer not to say	4	2.1%
What is the range of your earning per annum after tax? (n = 189)	Less than $35,000	2	1.1%
	$35,000–$70,000	39	20.6%
	$70,000–$150,000	54	28.6%
	More than $150,000	54	28.6%
	Undisclosed	40	21.2%
Ownership of residence (n = 233)	Owner	203	87.1%
	Part owner	6	2.6%
	Renting	23	9.9%
	Residing based on other arrangements	1	0.4%
Property type (n = 231)	Detached	143	61.9%
	Semi-detached including duplex and fourplex	42	18.2%
	Row-townhouse	25	10.8%
	Apartment	18	7.8%
	Others	3	1.3%
How long have you lived in the current property in years (n = 232)	Less than 5	59	25.4%
	5–14	91	39.2%
	15–24	30	12.9%
	25 or more	52	22.4%

, showing that about 65% of the respondents are aged 55 and above, of which 39% are 65 and above. The respondents are split nearly evenly between the male and female genders. Nearly 60% of the respondents earn $70,000 and above. Almost 90% of the respondents have a post-secondary certificate. The respondents are mostly homeowners (87%), and 62% live in detached houses. About a quarter of the respondents have lived in their residence for 5 years of less, 22% have lived in their residences for 25 years of more, and slightly over half of the respondents have spent between 5 and 24 years in their current residence. About 43% of the respondents live in residences that are more than 30 years old. Only a small proportion (12%) live in residences that are under 10 years old.

5. Findings

5.1. Wildfire Risk Perception

To understand participants’ perception of wildfire risk, we asked how concerned they were about wildfires in Whistler. As

Table 2. Respondents’ perception of wildfire risk.

	‘1’ Very Concerned	‘2’ Concerned	‘3’ Neutral	‘4’ Unconcerned	‘5’ Very Unconcerned		Mean	Std. Deviation
Concerned about wildfires in Whistler? (n = 230)	51%	40%	7%	2%	0%		1.61	0.74
	‘1’ Extreme	‘2’ High	‘3’ Moderate	‘4’ Low	‘5’ No Risk	‘6’ Unknown
Perceived risk to Whistler from wildfire? (n = 230)	25%	55%	19%	1%	0%	0%	1.95	0.69
Perceived risk property from wildfire? (n = 230)	17%	47%	32%	4%	0%	0%	2.24	0.78
	‘1’ Strongly Agree	‘2’ Agree	‘3’ Neutral	‘4’ Disagree	‘5’ Strongly Disagree
Threat to property warrants mitigation (n = 229)	36%	48%	12%	3%	1%		1.86	0.83

below shows, an overwhelming majority (91%) felt ‘very concerned’ or ‘concerned’. This is consistent with the perceived level of risk to Whistler from wildfire: 80% of participants thought that the risk to Whistler was either ‘extreme’ or ‘high’. A significantly smaller proportion, about two-thirds, thought wildfire risk to their property was ‘extreme’ or ‘high’, with a third stating that risk to their property from a wildfire was ‘moderate’. A significant majority (84%) ‘agree’ or ‘strongly agree’ that the threat to their property from wildfire warrants mitigation action. Overall, there is a high perception of risk with a very high level of concern about wildfires in Whistler. Risk to personal property is considered sufficient to warrant action to mitigate it.

5.2. Level of Preparedness

To assess participants’ perceived level of wildfire preparedness, we asked them to rate their preparedness. Approximately half (46%) of the participant said they were ‘moderately prepared’. Only 23% felt ‘prepared’ or ‘very prepared’. Asked how prepared they felt their community was, only 16% felt it was ‘prepared’ or ‘very prepared’. About half (49%) said that their community was ‘moderately prepared’. To assess the participants’ actual level of wildfire preparedness, we posed eight ‘yes-or-no’ questions on various preparedness activities and cognitive tasks. As

Table 3. Respondents’ level of wildfire preparedness.

	‘2’ Yes	‘1’ No	Mean	S.D.
If evacuated, I have a place to go aside from the emergency shelter (n = 220)	73%	27%	1.73	0.444
I have a grab-and-go bag ready in case of an evacuation (n = 223)	36%	64%	1.36	0.481
I know where in the community to find information during an emergency (n = 222)	64%	36%	1.64	0.480
I have a fireproof safe for valuables (n = 221)	26%	74%	1.26	0.441
My household has a plan in place in case of an evacuation order (n = 221)	40%	60%	1.40	0.491
If I have a suggestion on wildfire management, I know where to provide it (n = 218)	36%	64%	1.36	0.480
I am aware of the wildfire management process in my community (n = 219)	53%	47%	1.53	0.500
I have healthcare coverage in addition to the provincial healthcare plan (n = 218)	69%	31%	1.69	0.462

below shows, the level of preparedness across the activities and cognitive tasks varied significantly. A high proportion (73%) of the participants have a place to go aside from the emergency shelter, if evacuated. A total of 69% have healthcare coverage in addition to the widely accessible provincial healthcare plan. When asked whether they know where in the community to find information in an emergency, 64% of respondents answered ‘Yes’. On the low end of the preparedness completion are tasks such as having a fireproof safe for valuables, completed by only 26%, and having a grab-and-go bag ready in case of an evacuation, completed by 36% of the respondents. Slightly over half of the respondents are aware of the wildfire management process in their community.

5.3. Level of Completion of Mitigation Actions

We also assessed participants’ level of completion of mitigative activities by asking whether or when they plan to complete a list of mitigative activities that correspond to FireSmart recommendations. Overall, the level of mitigation completion is higher than that of preparedness. As shown in

Table 4. Respondents’ level of completion of mitigation actions.

	Completed	To Complete in 1 Year	To Complete in 2 Years	To Complete in 3 Years	Not Applicable
Screen or enclose the undersides of decks and porches (n = 199)	20%	10%	6%	13%	51%
Install stucco, metal, brick or other fire-resistant exterior siding on your house (n = 196)	28%	3%	1%	11%	58%
Screen house vents, gutters and the underside of eaves with metal mesh (n = 206)	32%	11%	7%	11%	39%
Landscape with fire-resistant materials and vegetation (such as rocks, aspen) (n = 206)	39%	11%	6%	7%	37%
Thin shrubs or trees so that nearby plants and trees do not touch (n = 211)	50%	24%	4%	4%	18%
Prune large trees by removing all branches that are close to the ground (n = 209)	55%	22%	4%	3%	17%
Remove needles, leaves and overhanging branches from the roof and gutters (n = 207)	61%	14%	3%	2%	19%
Keep grass short and water frequently during the spring, summer and autumn (n = 214)	63%	4%	0%	0%	33%
Remove shrubs, trees or fallen branches close to your house (n = 212)	65%	19%	2%	3%	11%
Install metal, asphalt, slate, tile or other fire-resistant roofing materials on your roof (n = 205)	70%	1%	0%	6%	22%
Install double or thermal pane or tempered glass in windows and exterior glass (n = 204)	71%	2%	0%	9%	17%

below, some of the mitigative activities with the highest completion rates include installation of double or thermal pane or tempered glass in windows and exterior glass (completed by 71%), installation of fire-resistant roofing material (completed by 70%), removal of shrub, trees and fallen branches close to house (completed by 65%) and removal of leaves, needles and overhanging branches from the roof and gutters (completed by 60%). The least completed mitigative activities include screening or enclosing the underside of decks and porches (completed by 20%), installing fire-resistant exterior siding on the house (completed by 28%) and screening house vents, gutters and the underside of eaves with metal mesh (completed by 32%). This pattern of completion is consistent with the way respondents ranked different parts of the house in terms of their importance in the ability of a house to withstand wildfires, from 1 to 7, with 1 being the most important. The roof was ranked highest (M = 2.22; SD = 1.29) followed by landscape (M = 2.99; SD = 2.07). Eaves were ranked the lowest (M = 4.36; SD = 1.37).

5.4. Allocation of Responsibility

In the context of shared responsibility for wildfire, we assessed how participants were allocating responsibility for the different phases of wildfire management to various actors. As shown in

Table 5. Respondents’ rankings of level of responsibility of actors across the four phases of wildfire management (1—most responsible; 7—least responsible).

Ranking	Preparedness	Mitigation	Response	Relief and Recovery
1	Municipal Government (M = 2)	Municipal Government (M = 1.93)	Municipal Government (M = 1.7)	Provincial Government (M = 2.49)
2	Homeowner (M = 2.82)	Provincial Government (M = 2.76)	Provincial Government (M = 2.26)	Municipal Government (M = 2.89)
3	Provincial Government (M = 3.07)	Homeowner (M = 3.1)	Federal Government (M = 3.91)	Private Sector (e.g., Insurance) (M = 3.52)
4	Local Community (M = 3.72)	Local Community (M = 3.69)	Local Community (M = 3.94)	Federal Government (M = 3.61)
5	Federal Government (M = 4.62)	Federal Government (M = 4.32)	Homeowner (M = 4.39)	Homeowner (M = 4.99)
6	Private Sector (e.g., Insurance) (M = 5.4)	Private Sector (e.g., Insurance) (M = 5.61)	Private Sector (e.g., Insurance) (M = 5.77)	Local Community (M = 5.06)
7	NGOs (e.g., Red Cross) (M = 6.38)	NGOs (e.g., Red Cross) (M = 6.59)	NGOs (e.g., Red Cross) (M = 6.03)	NGOs (e.g., Red Cross) (M = 5.44)

below, respondents ranked municipal governments as having the most responsibility for mitigation, preparedness, and response, and second most responsible for recovery after the provincial government. The homeowner is the third most responsible for mitigation, second most responsible for preparedness, and much less responsible for the response and recovery phases. Conversely, the provincial government’s responsibility increased progressively through the preparedness, response, and recovery phases. The private sector, which remained at the bottom of the ranking across preparedness, mitigation and response, ranked third in responsibility for recovery after the provincial and municipal governments.

We applied a series of Multivariate (multinomial) Generalized Linear Models (GLMs) to identify which of our survey variables explain the pattern of responsibility allocation among participants. For the mitigation phase, we tested ten relevant variables in our survey related to wildfire mitigation, as shown in

Table 6. Variables related to mitigation.

Variable Description	Score	Mean	S.D.
Ownership of residence	1 = Owner, 2 = Part Owner, 3 = Renting, 4 = Other	1.24	0.637
Wildfire threat to property warrants mitigation	1 = Strongly Agree, 5 = Strongly Disagree	1.86	0.830
To what degree is fuel management effective?	1 = Very Effective 5 = Very Ineffective	2.13	0.883
Maximum one-off payment for a FireSmart assessment?	1 = No cost, 2 = $250, 3 = $500, 4 = $750, 5 = $1000, 6 = $1500	1.69	1.017
I have the skills for wildfire mitigation work.	1 = Strongly Agree, 5 = Strongly Disagree	3.05	1.115
Property type	1 = Detached, 2 = Semi-Detached, 3 = Townhouse, 4 = Apartment, 5 = Other	1.68	1.030
How familiar are you with FireSmart activities?	1 = Very Familiar, 5 = Unfamiliar	1.99	1.058
Have you had a FireSmart assessment done for your residence?	1 = No, 2 = Yes	1.44	0.497
It is difficult to find money for wildfire mitigation work	1 = Strongly Agree, 5 = Strongly Disagree	2.40	0.956
Mitigation will significantly reduce damage to my property	1 = Strongly Agree, 5 = Strongly Disagree	2.06	0.887

below. We found that only two variables, namely ownership of residence (Pillai’s Trace value = 0.088; p < 0.05) and the maximum one-off payment for a FireSmart assessment (Pillai’s Trace value = 0.120; p < 0.05), were significantly associated with the overall responsibility allocation in the mitigation phase.

For preparedness, we tested six relevant variables in our survey related to preparedness, as shown in

Table 7. Variables related to preparedness.

Variable Description	Score	Mean	S.D.
How concerned are you about wildfires in Whistler?	1 = Very Concerned, 5 = Very Unconcerned	1.61	0.740
What do you perceive as the risk to Whistler of a potential wildfire?	1 = Extreme, 5 = No Risk, 6 = Unknown	1.95	0.690
What do you perceive as the risk to your property of a potential wildfire?	1 = Extreme, 5 = No Risk, 6 = Unknown	2.24	0.780
To what degree are you prepared for a future wildfire event?	1 = Very Prepared, 5 = Unprepared, 6 = Unsure	3.13	0.951
To what degree do you think your community is prepared for a future wildfire event?	1 = Very Prepared, 5 = Unprepared, 6 = Unsure	3.35	0.996
Have you had a FireSmart assessment done for your residence?	1 = No, 2 = Yes	1.44	0.497

below. We found that only one variable captured in the question ‘To what degree are you prepared for a future wildfire event?’ is significantly associated with the overall responsibility allocation in the preparedness phase (Pillai’s Trace value = 0.143; p < 0.05). In other words, how prepared respondents felt for a future wildfire influenced patterns of responsibility allocation for preparedness.

For the response phase, we tested eight variables in our survey related to the response phase of wildfire management, as shown in

Table 8. Variables related to response.

Variable Description	Score	Mean	S.D.
Have you ever been evacuated or placed on evacuation alert because of a wildfire?	1 = Yes, 2 = No	1.91	0.283
Do you have fire insurance coverage?	1 = No, 2 = Not sure, 3 = Yes	2.75	0.525
If evacuated, I have a place to go aside from the emergency.	1 = No, 2 = Yes	1.73	0.444
I have a grab-and-go bag ready in case of an evacuation order.	1 = No, 2 = Yes	1.36	0.481
I know where in the community to find information during an emergency.	1 = No, 2 = Yes	1.64	0.480
I have a fireproof safe for valuables.	1 = No, 2 = Yes	1.26	0.441
My household has a plan in place in case of an evacuation order.	1 = No, 2 = Yes	1.40	0.491
At what point do you think the resources of Whistler’s fire department are no longer sufficient to address a fire event?	1 = 2 houses on fire, 2 = 5 houses, 3 = 10 houses, 4 = 20 houses, 5 = 50 houses	2.11	0.832

below. The results indicate that only one variable captured in the question ‘Have you ever been evacuated or placed on evacuation alert because of a wildfire?’ is significantly associated with the overall responsibility allocation in the response phase (Pillai’s Trace value = 0.086; p < 0.05). In other words, whether or not participants have experienced wildfire evaluation influences the patterns of responsibility allocation for wildfire response.

For the recovery phase, we tested two relevant survey variables related to recovery, as shown in

Table 9. Variables related to recovery.

Variable Description	Score	Mean	S.D.
Do you have a fire insurance coverage?	1 = No, 2 = Not sure, 3 = Yes	2.75	0.525
I have access to healthcare insurance coverage in addition to the provincial healthcare plan.	1 = No, 2 = Yes	1.69	0.462

below. The results indicate that neither of the two variables is significantly associated with the allocation of responsibility for wildfire relief and recovery.

6. Discussion

Concern about wildfire risk in Whistler is almost universal, with a significant majority of the participants rating risk to their community and their property as extreme or high. This pattern of heightened perception of wildfire risk observed among the participants reflects the context of active wildfire risk management pursued by the municipal government. Faulkner et al. [48] similarly found higher levels of risk perception and awareness in communities with wildfire management activities than in those without. As noted earlier, wildfire vulnerability assessments carried out in Whistler indicate significant levels of wildfire risk. In addition, participants would be keenly aware of the past wildfire events in and around Whistler, which have often been quickly suppressed, especially those that occurred in Whistler and its ski mountains. In discussions with wildfire managers, there were often references to high-profile wildfire disasters such as Fort McMurray in 2016 and Jasper National Park in 2024 as holding lessons for Whistler. The levels of risk perception and concerns reported here are comparable to those reported in other communities in British Columbia, Canada (e.g., [39]).

As the mixed level of wildfire preparedness shows, the high level of risk perception and concern is not translating into high levels of wildfire preparedness. The level of wildfire preparedness among the participants is mixed, with only four tasks completed or affirmed by more than half of them. Fewer than half of the participants have completed the remaining four tasks. A near-universal level of concern did not translate into a high level of completion of preparedness actions. Only 36% and 26% of participants completed activities such as having a grab-and-go emergency bag and a fireproof safe for valuables, respectively. These low levels of completion of basic preparedness measures belie the high level of risk perception demonstrated by the participants. Conversely, the level of completion of mitigation actions was much higher among the participants. More than half of the respondents have completed seven out of eleven mitigation activities. This level of completion is consistent with the participants’ level of risk perception and concern. It also confirms the high level (84%) of respondents who agree or strongly agree that wildfire risk to their property warrants mitigation. This pattern also shows differentiated emphasis and a possible trade-off between preparedness and mitigation among our participants. This potential trade-off might be understood as part of existing trade-offs in wildfire-prone socio-ecological systems [59], and in the values underpinning shared responsibility [34].

There is a remarkable pattern in how participants allocate responsibility for different phases of wildfire management. The municipal government was put at the forefront of preparedness, mitigation, and response, except for relief, for which the provincial government has primary responsibility. This pattern of allocating the most responsibility for disaster management to governments has also been reported in the context of flooding in Canada (see [56]). Homeowner responsibility is shown to decline as one moves from mitigation and preparedness through to response and recovery. This pattern of responsibility allocation is influenced by several factors. Allocation of responsibility for preparedness is influenced by participants’ own perceived level of preparedness for a future wildfire event. This makes sense since participants are likely to rank homeowners as most responsible for preparedness if they themselves are more confident about their level of preparedness, and rank other actors as having higher responsibility for preparedness if they are less confident about their own level of preparedness. It is also remarkable that it was the perceived level of preparedness and not actual preparedness that shaped the allocation of responsibility for preparedness. The allocation of responsibility for mitigation is influenced by whether the participant owns their residence and by the amount of a one-off payment they are willing to make for a FireSmart assessment. Homeowners are likely to feel greater responsibility for mitigating wildfires on their property than non-homeowners, who may delegate this responsibility to other actors. Also, those willing to pay more for a FireSmart assessment are likely to allocate greater responsibility to homeowners compared to those who are willing to pay lower amounts or nothing.

The allocation of responsibility for wildfire response is influenced by firsthand experience of a wildfire evacuation order or alert. This demonstrates the importance of prior wildfire experience in shaping how participants think about responsibility for wildfire response. McGee et al. [49] also found that prior experience with wildfire disasters influenced people’s wildfire behaviours. A simple crosstabulation shows that a major difference between those who have experienced evacuation before and those who have not is down to how they rank municipal and provincial governments. Those with previous experience allocated more responsibility to the provincial government, while those without previous evacuation experience allocated more responsibility to the municipal government.

7. Conclusions

This article has argued that research focused on public engagement with wildfire risk and responsibility allocation gains renewed significance in the context of increased emphasis on shared responsibility for wildfire management. We have illustrated this argument through a case study of public engagement with wildfire risk and management in Whistler BC, Canada. We employed quantitative data processed through both descriptive and inferential statistics. Our findings show a near-universal concern about wildfires among the participants and a high level of risk perception, both of which are consistent with community climate and wildfire reports and plans. This level of risk perception is also reflected in the level of completion of mitigation actions, which remains high among the participants. However, the level of preparedness is mixed, with some actions showing high completion rates and others low. How respondents allocate responsibility for various aspects of wildfire management is shaped by factors such as home ownership, prior wildfire experience, perceived preparedness, and commitment to bearing the costs of hazard assessment.

Our study makes at least three contributions to the debates on public engagement with wildfire risk and the literature on disaster risk reduction. First is the importance of local contextual factors for understanding patterns of public attitude to wildfire risk [36,37,38,45,48]. The context of local fire hazard levels, wildfire resiliency planning and ongoing wildfire management, both at the household and landscape levels, has contributed to the high levels of risk perception, almost universal concern about wildfire risk and familiarity with FireSmart observed among the residents. This calls for more careful attention to context and greater collaboration with local landscape managers to deepen appreciation of the ways that contextual factors explain some of the patterns in public attitudes towards wildfire risk.

Secondly, our study shows the need to pay attention to the differentiated ways in which the public engages with wildfire preparedness and wildfire mitigation. The clear difference between these two aspects of participant engagement in wildfire management actions suggests that there may be a basis for a trade-off between preparedness and mitigation (cf. [16,34,59]). Although the greater emphasis on mitigation among our participants may be linked to some local factors—for instance, most of Whistler’s neighbourhoods are tucked within forests and other vegetation, making mitigation activities appear more necessary—whether this increased focus on mitigation contributes to a decreased concern for preparedness is different question. If a trade-off indeed exists in how the public approaches mitigation and preparedness, then it would be important for future studies to investigate how this trade-off operates, which factors drive it, and how policy interventions can effectively shape it to enhance wildfire resilience.

Third, our study highlights the need for more explicit consideration of shared responsibility in public engagement programs and policies. Not only is it important to track how the public is allocating responsibility for various aspects of wildfire management, but it is also important to understand which factors shape public allocation of responsibility. This is crucial for policymakers and landscape managers seeking to achieve more effective sharing of responsibility, which requires more explicit and democratic negotiation of expectations between the public and decision makers [17,21,22]. For instance, the fact that only prior wildfire evacuation experience matters in how the public allocates responsibility for wildfire response suggests a significant cognitive difference between those who have experienced a wildfire and those who have not. We should not have to wait for a wildfire event for the public to gain a clearer understanding of how responsibility is shared for wildfire response, or indeed for any other phase of wildfire management. As such, programs and policies need to embed more explicit frameworks for articulating (and negotiating) prevailing patterns of shared responsibility among actors operating across scales.