Factors affecting community-agency trust before, during and after

Journal of Environmental Management 130 (2013) 10e19

Contents lists available at ScienceDirect

Journal of Environmental Management journal homepage: www.elsevier.com/locate/jenvman

Factors affecting community-agency trust before, during and after a wildfire: An Australian case study Emily A. Sharp a, *, Rik Thwaites a, Allan Curtis a, b, Joanne Millar a a b

Institute for Land, Water and Society, Charles Sturt University, PO Box 789, Albury, NSW 2640, Australia National Centre for Groundwater Research and Training, Adelaide, South Australia, Australia

a r t i c l e i n f o

a b s t r a c t

Article history: Received 14 January 2013 Received in revised form 30 June 2013 Accepted 14 August 2013 Available online

Trust has been identified as a critical relationship component in contexts of high uncertainty and complexity such as wildfire management, and as a primary factor in public support for wildland fire management strategies. However, little attention has been paid to identifying and comparing factors across fire management stages (i.e. before, during, after a fire) that may influence trust between community members and fire management agencies. This paper attempts to address this gap by exploring factors affecting community-agency trusting relationships before, during and after a wildfire event. We draw upon 26 semi-structured interviews with 38 residents of a community directly impacted by fires in December 2006 and January 2007 in Victoria, Australia. Communication, cooperation, trustworthiness, and integration of local concerns and knowledge influenced trust in more than one fire management stage. Institutional structures and reduction of uncertainty were particularly strong influences during a fire. After a fire, resolving negative outcomes and immediately meeting perceived needs arising from the fire were factors unique to this stage. Ó 2013 Published by Elsevier Ltd.

Keywords: Trust Trustworthiness Fire and fuels management Communication Shared responsibility Australia

1. Introduction The state of Victoria in south-eastern Australia is one of the most fire-prone environments in the world because of its climate, vegetation and geography. In both the 2002e03 and 2006e07 fire seasons, wildfires burnt over one million hectares of public and private land. The 2006e07 wildfires were the longest recorded fires in Victoria’s history, running for 69 days (Smith, 2007). In February 2009, severe wildfires burnt fewer hectares (450,000 ha) but claimed 173 lives, injured more than 5000 people and destroyed more than 2000 homes. External reviews following the 2002e03, 2006e07 and 2009 fires have highlighted the need for wildfire management agencies to work in partnership with communities to prepare for and respond to wildfires. Historically, public land managers and other managing agencies have provided public awareness and education campaigns, and community consultation and development, in regards to wildfire planning and management. More recently in Australia and internationally, environmental managers responsible for wildfire have placed a greater emphasis on community engagement and community-agency partnerships to build * Corresponding author. Tel.: þ61 2 60259894; fax: þ61 2 60519897. E-mail address: [email protected] (E.A. Sharp). 0301-4797/$ e see front matter Ó 2013 Published by Elsevier Ltd. http://dx.doi.org/10.1016/j.jenvman.2013.08.037

community capacities to prepare for and respond to wildfire. Trust is important in these partnerships because it encourages members to communicate openly, agree to work together, and seek to maintain relationships (Beunen and de Vries, 2011; Cooke et al., 2011; Six, 2005). Trust is considered to be a significant resource in social relationships, such as partnerships, because it can reduce uncertainty and complexity and save expenditures of time and emotion (Bachmann, 2006). Trust is also identified as a key component of the social acceptance of fire and fuels management strategies (e.g. Brunson and Evans, 2005; Vogt et al., 2005). Similarly, trust in information sources like government agencies may shape public opinion about environmental disasters (Safford et al., 2012) and influence individuals’ wildfire mitigation actions when a hazard is unfamiliar (Brenkert-Smith et al., 2006; Paton, 2007; Siegrist and Cvetkovich, 2000). However, Absher and Vaske (2011) argue that mitigation actions are not directly influenced by trust but mediated by other constructs such as past behaviour and perceived effectiveness of the actions. The importance of trust in building and maintaining relationships between fire-prone communities and fire management agencies (i.e. community-agency trust) has emerged from studies examining various community-agency interactions before (McCaffrey, 2004; Winter et al., 2004), during (Kumagai et al.,

E.A. Sharp et al. / Journal of Environmental Management 130 (2013) 10e19

2004) and after wildfires (Burns et al., 2008; Olsen and Shindler, 2010). Some research also recognises that relationship interactions in a particular fire management stage (i.e. before, during or after wildfire) influence community-agency trust in other stages and over time. For example, Mendez et al. (2003) suggest that trust in agency actions during a fire may be reduced for people who were already alienated from government agencies before the fire. Diminished trust stemming from before and during the fire may then compromise public support for restoration activities after the fire. Similarly, trusting relationships built after a fire can influence public support for future management actions (Toman et al., 2008) and collaborative planning between agencies and communities (Burns et al., 2008). Although recent studies have identified factors influencing trust in general, or for a specific fire management stage (i.e. before, during or after a wildfire), previous research has not explicitly attempted to identify factors affecting trust across all three management stages in a single study. Additionally, little attention has been paid to how identified factors that are important in one stage (e.g. before) may be similar or different to factors identified in the other stages (e.g. during or after). Therefore, our research contributes to the literature by: 1) identifying factors affecting community-agency trust before, during and after a wildfire through a single study; and, 2) by exploring the similarities and differences of these factors among the management stages (i.e. before, during and after a fire) using one data set. In this paper, we begin by describing previous research that defines trust and identifies factors affecting trust for fire management in general or for a specific fire management stage. After providing this conceptual background, we present and discuss our case study examining factors affecting community-agency trust before, during and after a prolonged fire event in rural Victoria, Australia. Finally, we discuss the implications of this research. 2. Conceptual background Because trust is considered fundamental to human relationships (Cook, 2001), numerous conceptualisations of trust have been developed in different disciplines, such as economics and psychology. While there is no consensus among disciplines about the definition of trust (Kramer, 1999), definitions by Mayer et al. (1995) and Rousseau et al. (1998) are frequently cited and used in the natural resources management (NRM) literature (e.g. Davenport et al., 2007; Liljeblad et al., 2009). In the present study, trust was defined as ‘a willingness to rely upon another person or organisation based upon positive expectations of their intentions or behaviour’, adapted from Mayer et al. (1995, p. 712) and Rousseau et al. (1998, p. 395). This definition of trust makes a distinction between trust and trustworthiness. First, a willingness to rely represents trust or trusting intentions. Second, these trusting intentions are based on positive trusting beliefs or expectations (i.e. beliefs or expectations about another’s trustworthiness). Trustworthiness is a quality of the trustee (i.e. person being trusted) while trusting is something that the trustor (i.e. person doing the trusting) does (Mayer et al., 1995). These constructs are linked in that the trustor’s perception of the trustee’s trustworthiness contributes to the trustor’s intention to trust the trustee. Several authors (e.g. Hosmer, 1995; Hudson, 2004; Mollering, 2006; Rousseau et al., 1998) have suggested that trust, or trusting intentions, is usually associated with several elements: 1) interdependence among actors; 2) uncertainty regarding the intentions or behaviour of the trustee; 3) risk, that is, the trustor could experience negative outcomes if the trustee proves untrustworthy;

11

4) vulnerability of the trustor through taking on that risk; and, 5) expectations that the trustee will not abuse the trustor’s vulnerability. Trust is often operationalised as reliance on the other party for certain actions that carry some sort of risk or potential for negative outcomes. Most of the trust literature views trustworthiness as a multidimensional construct. Mayer et al. (1995) suggest that trustworthiness is comprised of three characteristics, including: ability (i.e. trustor perceptions of the trustee’s knowledge, skills and competencies); benevolence (i.e. the extent to which a trustor believes that a trustee will act in the best interest of the trustor); and, integrity (i.e. the extent to which the trustor perceives the trustee as acting in accord with a set of values and norms shared with or acceptable to the trustor). A trustor makes a decision about whom to trust on the basis of information about the trustee’s trustworthiness. According to Rus and Iglic (2005), this information can be obtained directly through personal experience or indirectly through perceptions of the quality of institutional structures, processes and outcomes (e.g. policies and procedures). These sources of information distinguish interpersonal trust from institutional trust. Interpersonal trust is based on the form and characteristics of the relationship itself (e.g. formal vs informal relationships, relationship history) (Lewicki et al., 2006). Institutional trust represents systems operating in a predictable way and transcends interpersonal relationships (Luhmann, 1988; Zucker, 1986). Institutional trust can precede interpersonal trust by providing an environment which helps people to engage in a relationship when personal experience with others is lacking (Blomqvist, 1997; Luhmann, 1988). Although studies relating to trust in wildfire and NRM often do not explicitly differentiate between trust and trustworthiness, antecedents important to trust-building in these contexts have been identified. For example, in an NRM context, Shindler and AldredCheek (1999) identified inclusiveness, sincerity, commitment, continuity, sound organisation and planning skills, and efforts that lead to action, as factors of agency trustworthiness in communityagency relationships. Gray et al. (2012) suggest trust is influenced by whether those affected by decisions are involved in the decisionmaking process and that trust varies with scale, such that state and regional agencies are more trusted than federal ones. Other potential trust influences include: agency performance; agency staff understanding of local culture; equitable distribution of benefits; level of dependence on the agency (Baral, 2012); levels of community engagement; and, clearness of communication (Davenport et al., 2007). In a fire and fuels management context, Winter et al. (2004) found that public trust in the government’s ability to manage fuels is tied to perceptions of risks and benefits and the agency characteristics of care, competence, and credibility. Cvetkovich and Winter (2007) contend that trust is centred on the similarity of values between an individual and an agency, the agency’s consistency in their actions with those values and the legitimacy of any inconsistencies. Liljeblad et al. (2009) found that the primary dimensions of trust lie not only in shared norms and values, but also in an individual’s willingness to endorse the agency to act on their behalf and in perceptions of efficacy (i.e. beliefs about how others will act and their capacity to act). In collaborative wildfire planning processes, LaChapelle and McCool (2012) identify transparency, leadership, framing of risk assessments, prior relationship history and the scale of the plan as factors influencing trust. Given the breadth of factors identified as being important to trust, we sought to identify and compare those factors which were most important to trust for each fire management stage (i.e. before, during and after) according to fire-affected residents of a case study area in Victoria, Australia.

12


3. Wildfire management context and study area In rural and regional Victoria, the Country Fire Authority (CFA) and the Department of Sustainability and Environment (DSE) are the two main government bodies responsible for wildfire planning, preparation and response. The CFA is largely a volunteer organisation at local brigade level, and is responsible for managing fires on private land; the DSE is responsible for fires on public land, including state forests and national parks. The agencies work together to manage large fires which cross public-private boundaries. Community fire recovery programs and services are coordinated by the Department of Human Services (DHS) and local government (Council). However, the DSE and CFA jointly coordinate the repair of private assets damaged as a result of efforts to control wildfire (e.g. fencing damaged in constructing fire breaks). The DSE is also responsible for rehabilitating public land and assets affected by wildfire. During the 2006e07 fire season, Victoria adhered to a ‘Stay or Go’ policy which advised community members to make a decision prior to the fire season to prepare, stay and defend their properties from wildfires or leave well before a fire arrived in their area. The policy was underpinned by legislation which, instead of mandating forced evacuations, grants community members the right to stay and defend their properties. The legislation remains current; however, the extensive loss of life and property in wildfires in February 2009 has led to extensive revisions to the ‘Stay or Go’ policy. The current policy states that the protection of human life is paramount and places greater emphasis on the ‘leave early’ option,

as well as community members having back-up plans and access to a range of safety options, including places to shelter during a fire. Our research focused on fire-affected communities in the King Valley within the Rural City of Wangaratta (RCOW) local government area in Victoria. This rural area, located approximately 250 km northeast of the state capital Melbourne, is bound on the south and east by extensive tracts of densely forested public land (Fig. 1). The study area comprises small rural localities and mixed farms and covers approximately 1350 square kilometres, with a total population just under 2000 (Australian Bureau of Statistics, 2006). The climate features hot, dry summers and cool, wet winters. For several years prior to wildfires in 2006e07, the area had experienced drought conditions with substantially lower than normal rainfall and above average temperatures. Lightning strikes ignited multiple fires on public and private land in and around the King Valley in early summer, December, 2006. Not all ignitions could be contained, and the fires burned throughout December and eventually joined with other fires to become the Great Divide Complex which burnt over one million hectares of land. Lightning ignited a new round of fire in early January 2007, burning approximately 33,000 ha, including 750 ha of plantation timber, before being extinguished in early February. Although no one was killed and few primary residences lost, the two rounds of fire caused substantial economic and social disruption to the communities. Smoke taint resulted in the loss of the 2007 wine vintage locally, and private pine plantation losses were estimated at over $20 million (AUD) dollars (Plantations Northeast, 2008).

Fig. 1. Rural City of Wangaratta local government area, Victoria, Australia. Fire-affected localities shown in relationship to the regional centre of Wangaratta.


4. Methods

13

Table 2 Interview guide used with community member participants.

The research aimed to identify factors affecting communityagency trust before, during and after a wildfire and to explore how the factors were similar or different among these management stages. We used semi-structured interviews with purposively sampled participants to gather qualitative data that was analysed using inductive and deductive procedures. A qualitative approach allowed us to capture detailed descriptions of trusting relationships which were important to participants with similar and different experiences before, during and after one common event. 4.1. Study participants Participants were purposively selected from residents in RCOW who were affected by wildfires in December 2006 and January 2007. Purposive sampling allows researchers to select participants who can be expected to clarify and deepen understanding of the research objectives because of those individuals’ specific characteristics, knowledge and experiences (Rubin and Rubin, 2005). Participants were recruited from lists provided by the RCOW Bushfire Recovery Officer and a Community Safety staff member from CFA Regional Headquarters at Wangaratta. The lists included fire-affected residents who had positive (i.e. expressed accolades to agencies or described interactions as ‘good’, ‘reliable’, ‘helpful’, etc.), negative (i.e. expressed concerns to agencies or described interactions as ‘poor’, ‘bad’, ‘unreliable’, etc.) or mixed (i.e. expressed both positive and negative interactions) experiences with the agencies in the 2006e07 fires and who were willing to talk about their experiences. Several individuals who had submitted personal photos and stories to a local photographic exhibition featuring images from the fires were also recruited. Our sample was not intended to represent the statistical distributions of population demographics. Instead, we selected participants from a range of demographic categories who would reflect diversity in fire experiences and social and farming backgrounds. Participant characteristics are shown in Table 1. In all, we interviewed 38 community members (21 males, 17 females) in 26 interviews (12 interviews involved couples) between March and July 2008.

Question 1: Please tell me about your experiences with the 06/07 fires. Probe 1a) Had you experienced a bushfire before the 06/07 fires? Probe 1b) How were your experiences in the 06/07 fires similar or different to previous fires? Question 2: Please describe an event before, during or after the recent fires which either strengthened or diminished your trust in a particular person, organisation or agency. Probe 2a) What were the main factors that made you feel like you could/could not trust him/her/them? Probe 2b) What other things contributed to/diminish your trust in him/her/them? Question 3: What do you think is most important in being able to trust others: a) before the fire; b) during the fire; and, c) after the fire? Question 4: Trust is a very commonly used term, but we may not all be talking about the same thing when we use the word. If you had to write a dictionary definition for how you define trust, what would that definition be?

4.2. Data collection and analysis An interview guide was used in semi-structured interviews with fire-affected community members. The guide provided predetermined topics but allowed participants to elaborate on topics of particular importance to them or bring forth other related topics not specifically covered in the interview guide (Punch, 2005). Question probes were also used to further explore and/or clarify concepts brought forth by the interviewees. The same guide was used with all participants (Table 2). We began each semi-structured interview by explaining to the participant(s) that the purpose of the research was to learn about their relationships with government agencies and agency staff during the 2006e07 fires and how trust, or its absence, may have influenced these relationships. We first asked participants to describe their fire experiences and any particular incidents in which they felt their trust was diminished or increased in a specific person, agency or organisation before, during or after the fires. Participants were then asked to identify what they regarded as critical to the creation or loss of community-agency trust for each

Table 1 Social characteristics of 38 community members gathered through 26 interviews. Number of community member participants Gender (n ¼ 26) Couples (1M/1F) Male Female

12 9 5

Age (n ¼ 38) Under 65 Over 65

31 7

Primary Occupation (n ¼ 38) Agriculture 12 Professional 11 Trades 6 Retired 4 Other 5 CFA volunteer or family member in CFA at time of 06/07 fires (n ¼ 26) Yes 17 No 9 Had fire burn onto property during06/07 fires (n ¼ 26) Yes 16 No 10

Number of community member participants Length of residence in region (n ¼ 38) 40 years

8 10 6 6 8

Property size (n ¼ 26) 40 ha

16 10

Farm enterprise type (n ¼ 16) Grazing- beef or dairy Cropping or orchard or nuts Grapes/grazing Income source (n ¼ 26) On-farm only Some/all off-farm

9 3 4 9 17

14


fire management stage. Some concern may be raised about a respondent’s ability to recall events and feelings for each stage (i.e. before, during and after) 14 months after the events occurred. However, research indicates that in major events (e.g. a wildfire) recall does not present an issue in this time period (Berney and Blane, 1997). Finally, participants provided their own definition of trust to help clarify how they understood the abstract concept. We did this at the end of the interview so that their definition would not influence their other responses. We did not provide a definition at any point in the interview so that we would not influence their use or understanding of the term. However, at the end of the interview, when discussing participant definitions, we did provide our understanding of the term if participants asked. Interviews typically lasted 1e1.5 h, although some lasted as long as 2.5 h. All interviews were recorded with consent of the participant, transcribed verbatim and analysed using both an inductive and deductive iterative coding process (Rubin and Rubin, 2005). We used a standard coding procedure described by Richards (2005) as descriptive, topic and analytic coding. In the first pass, numerous codes were developed directly from the data (e.g. ‘difficulty in accessing assistance’; ‘fencing issues’; ‘emotional needs’). In the second pass, categories were developed by looking for associations between the numerous codes (e.g. the first pass codes above were developed into a category labelled ‘personal needs’; another category was ‘community needs’). In the third pass, categories were compared and contrasted to explore commonalities and differences between management stages and responses from different participants (e.g. participants with considerable vs little experience with wildfire). The final categories were combined through an iterative process to develop ‘factors’ which explained the relationships between the categories and between the categories and the research objectives (e.g. ‘personal needs’ and ‘community needs’ were categories which partly comprised the factor ‘perceived needs met immediately AFTER’). The qualitative data analysis program NVIVO 7 was used during the analysis process to help develop categories and code the data. We took care to distinguish between factors critical to trust and ‘issues’ that arose from the fires but were not necessarily critical to trust. To do this, we ensured that the final list of factors promoted a ‘willingness to rely on another person or organisation’ which reflected how the majority of participants defined trust (Sharp et al., 2012). 5. Findings As described below, communication, perceived cooperation among agencies, integration of local concerns and knowledge, and trustworthiness were inter-related factors affecting trust in more than one management stage (Table 3). Institutional structures and the reduction of community member uncertainty were additional factors influencing trust during a fire. After a fire, agency resolution

Table 3 Factors affecting trust described by participants for each fire management stage. Italicised factors represent those described for more than management stage. Before

During

After

Communication Interagency planning Local concerns addressed/included Trustworthiness

Communication Interagency cooperation Local knowledge/assistance integrated Trustworthiness Institutional structures

Communication Interagency coordination

Reduction of uncertainty

Trustworthiness Perceived needs met immediately Negative outcomes resolved

of negative outcomes which had occurred during the fire and undelayed provision of assistance for perceived community and individual needs were important to community-agency trust. 5.1. Factors affecting trust before a fire Communication was the most frequently mentioned factor as being important to trust before a fire. Many participants thought trust had been built because the agencies did a good job of providing information about making the ‘stay or go’ decision and agency activities occurring in fire planning and preparation. However, many participants also described how information seemed only to flow from agencies to residents, with some participants stating that there was little opportunity to provide feedback to agencies or work together with agencies on fire management that spanned the public-private boundary, as expressed in this quote. I’d like to talk, have someone offer to say, in partnership, not this is the rule and that’s the rule.. I want to say, ‘what [interface management] could we mutually agree to that would mean I feel adequately protected in my circumstances’? CM16 Most participants suggested that visible cooperation among agencies before a fire could build trust because it showed communities that agencies had defined roles and responsibilities which would work in the communities’ best interests during a fire. Conversely, many participants also said trust could be lost if it appeared that agencies conducted planning in isolation and did not share information with each other before the fire. One woman suggested that trust could be strengthened if agencies demonstrated to the public how they worked together before a fire, ‘so people know how this fits in with that and that sort of thing’. Similarly, several CFA volunteers discussed how the agencies had made considerable efforts in the past few years at regional and state level to work together conducting planned burns and producing joint public education publications. However, as this local CFA volunteer noted, cooperation in fire planning and preparation is not always visible to the public. .the strong attribute is that the agencies need to be able to work together. And I know that’s worked on, but . a lot of people don’t see the levels at which they work.. People quite often don’t understand that. CM9 Many participants described how trust could be built if managing agencies responded to local concerns addressed directly to the agency or raised in public meetings, and included those concerns in planning where appropriate. Before the fire, many participants felt that their concerns related to land management practices, permit processes, the condition of roads and fire access tracks, and identification of personal and community assets were not effectively addressed by Council and fire management agencies. Most participants considered land management and fire management issues as inseparable concerns. Several participants described how they had ‘resorted to local members of Parliament to try to get some form of dialogue [with Council and DSE]’ regarding fire and land management issues. Other participants felt they did not have a voice in matters which had the potential to directly impact them. Public meetings and consultation were often viewed as ‘polite listening exercises’ which did not result in visible change. Nothing that we can do or want is going to change.what these outside people and events are going to do.. It’s no point in jumping up and down and saying they should have done this or whatever because it’s not gonna change..CM19 Before the fire, trust was also influenced by trustworthy characteristics primarily related to the integrity of the agencies and how they responded to and included local residents’ needs and concerns


15

in fire planning and preparation (Table 4). The characteristics most commonly described included: responsive to needs; credible; transparent; open-minded; and, inclusive. Some participants who felt that the DSE, as an organisation, was not trustworthy did not feel that the DSE shared their values for land or fire management. For example, this participant questioned the integrity of the DSE because she felt it did not weigh her concerns for personal safety equally with agency conservation goals and objectives.

agency trust. For example, some participants described how residents had lost trust in the agencies because of visible conflict between them during the fire.

We’ve always been sort of fighting for retaining the natural environment, and here’s the one instance where it almost feels like we’re competing with the wildlife to have sanctuary.. That’s the thing I don’t trust about them. I don’t trust them to see us as equally important to the bush that’s at the back of the land. CM16

The perceived lack of cooperation led several participants to describe the fire response as ‘joint’ rather than ‘integrated’. These participants suggested that visible cooperation and ‘working together as a whole rather than as separate parts’ were important to community-agency trust in this stage. Although most CFA members felt that cooperation among agencies had greatly improved in recent years, rumours of conflict and disagreement persist, and participants thought this damaged trust.

5.2. Factors affecting trust during a fire As in the before a fire stage, communication was described by all participants as essential for community members being able to rely on, or trust, fire agencies and their staff during a fire. Many participants felt the community information meetings during the fire and their primarily one-way flow of accurate, timely and reliable information was a strong contributor to trust. Most participants stated that during a fire there ‘was no such thing as too much information’ related to the fire situation. This participant noted the importance of trustworthy information from managing agencies during a fire. You trust the information we’re getting is right. You rely on that.because it’s all you’ve got. You’ve got what you can see going on, and if the two don’t match up then you’re going, ‘well, hold on a minute’.. CM9 Many participants stated that a lack of cooperation between agencies during the fire could potentially diminish community-

There was so much conflict. What I noticed was that.you had all these groups, and they were all at loggerheads with each other.. And I know a lot of people have lost their trust in the system, and I know a lot of people are angry.. CM3

You hear a lot that there’s this incredible tension between them. and that doesn’t give me any confidence or trust.. CM13 During the fire, some participants felt that local knowledge about local conditions (e.g. local weather patterns, location and condition of fire access tracks, water access, etc.) was not wellintegrated into the fire response and that this diminished trust in the agencies. .the local CFA captain. should be involved in the strategy in how to go about fighting the fire, given that he knows the area so well. And I think there were very many decisions sort of made over his head or without even consulting him. That would certainly diminish trust and respect. CM22 Many participants, including CFA members, stated that persons with extensive local knowledge could be identified before a fire and more effectively integrated into response strategies. This was seen to engender trust because ‘locals trust locals and their knowledge’

Table 4 Differences among stages for trust factors common to more than one stage. Before Communication trust factor e direction and amount of information Direction of information flow Multi-directional Example of flow direction

Agency

Agency

Community Member Amount of information Not described Integration of local concerns and knowledge trust factor e who participates Who participates in Agencies decision-making Local community members in general Local community member experts Interagency cooperation trust factor e visibility and responsibility Visibility Visible through joint publications, training, public participation and education processes Trustworthiness trust factor e most common descriptors Most common descriptors Integrity e Inclusive Open-minded Responsive to needs Transparent Credible

Duringi

After

One-way with limited need for question and answer

Two-way with need for question and answer

Agencies

Agencies

Community Member

Community Member

‘No such thing as too much’ in decision-making Agencies Local brigade captains Local experts identified before the fire

Too much ¼ overwhelming Not described

Visible at the firefront and staging ground

Not directly visible: behind-the-scenes coordination of one-stop shop

Abilitye Strong leadership Good decision-making skills Consistent actions Competent BenevolenceeReassuring Kept residents’ best interests at heart

Benevolencee Compassionate Empathetic Caring Sensitive manner Abilitye Knowledge Anticipate needs Integritye Follow through Keep promises/word

16


more so than decisions made from personnel perceived to be ‘outsiders’. Trustworthiness also played a role during the fire with the characteristic of ability frequently described. For example, one CFA participant noted that community-agency trust resulted from having ‘the right people to make the right decision, that have the skills and personality to do it’. Another participant described how his trust increased in a particular agency member who displayed outstanding leadership and decision-making skills.

did not perceive that adequate warnings or assistance had been rendered.

He’s just an absolute natural leader.inspiring on a whole host of levels. He was able to do every job that was asked of him.. Where things are a really fluid situation, he was able to step up into those roles naturally.. He was someone you could trust.. CM18

After a fire, communication was still an influential trust factor. Many participants said trust was enhanced if communication was through one-on-one contact and limited to targeted information specific to individuals’ needs because it showed agency understanding of resident needs in this stage. Because most residents were mentally, physically and emotionally exhausted, too much information in the recovery phase was viewed as overwhelming, as described by this participant.

However, trustworthiness was not entirely based on ability. Some participants gave examples of individual CFA volunteers and DSE staff members who demonstrated benevolence when they took actions which were ‘reassuring’ during the emergency and ‘made decisions that had the residents’ best interests at heart’. Institutional structures, such as policies and procedures, were seen by many participants to provide a foundation for trust during a fire. Participants stated that residents are more willing to trust (i.e. rely on) agencies if they know there are policies and procedures in place that will minimise losses and ensure public and CFA volunteer safety. For example, one participant stated that ‘you just have to trust that there is order behind all this and it will become apparent’. Another participant said, ‘you trust there’s a set of protocols or a chain of command that should be followed..’ Another participant stated that positive outcomes underpinned trust-building, further explaining in this quote that institutional structures underpinned those outcomes. The structures to support the fire are as important [as trust itself]. The structures allow you to get good outcomes.. CM20 Trust, however, could be diminished because of specific management policies, such as crew changeovers conducted away from the fire front. For example, several participants felt abandoned or angry, thereby leading to a loss of trust, when crew changeovers left fires and backburns unattended and/or threatening private property. A few participants also described how trust could be diminished if agency personnel did not have the appropriate training to perform the tasks assigned, or CFA volunteers were not held accountable to response protocols because of their volunteer status, and this resulted in property damage or loss. While the literature suggests that reducing a trustor’s uncertainty about a trustee’s actions and responses to external circumstances is an important condition for trust in general, it was particularly important to participants when talking about the fire response stage. Many participants described how their uncertainty was reduced and their trust, or willingness to rely on the agencies, was increased when they were reassured that they were not alone during the fire. This reduction of uncertainty came in several forms: direct on-property assistance from the fire agencies; individual phone calls providing fire updates from CFA to individuals in remote areas; and, the visible presence of firefighting personnel and equipment, even if not on the participant’s property. There was the comfort.that there was a response in the valley and that we were there.the trucks going past and.just that they could see red and blue lights at night and just stuff going ‘round.. CM18 However, feelings of uncertainty and a consequent feeling of abandonment caused a few participants to harbour strong feelings of distrust towards the agencies. During the fire, none of these participants could visibly see an agency response in their area and

We knew we were on our own.but the helicopters never came.and people worse off than what we had, because they’ve got a lot more trees around them.they were left to their own defences as well.where were they [the agencies]? CM27 5.3. Factors affecting trust after a fire

You just felt like you were left.. And then all of a sudden, it was overkill. CM11 One couple noted that the ‘show-bag’ of information they received was ‘good value’ as a communication medium but could be seen as overwhelming if there was no further one-on-one contact to further explain the assistance available in a time of great stress and exhaustion. Agency cooperation, evidenced indirectly through ‘one-stop shop arrangements’ which provided information about recovery programs and assistance, was seen to strengthen trust. Participants suggested that ‘knowing that agencies and organisations have worked together’ to plan a coordinated response reduced community member frustration and increased community members’ willingness to rely on, or trust, agency advice. This participant explained how trust was enhanced through agency cooperation. Trust was built through having one place to go to. So you’re not going from this agency to that agency to that agency, trying to find the fact that you’ve got to go back to the first one.’ CM23 The trustworthiness characteristics of benevolence and integrity were frequently mentioned in the after stage. Several participants explained that agency staff needed to understand the overwhelming mental, emotional and physical stress residents had undergone. Consequently, these participants frequently described trustworthy personnel as individuals who responded to residents’ needs and concerns in a sensitive, compassionate and empathetic manner. Both an agency and its staff were described as trustworthy when they ‘followed things through’. For example, the participant quoted below said that the most critical aspect to communityagency trust after a fire related to agencies ‘keeping their word.’ Following up on what they say they’re going to do. And that’s particularly with an organisation. Like one arm of the organisation says, ‘yes, we’ll reimburse you’ and the other arm says, ‘no, we won’t.’ For us, we’re dealing with an organisation. And to me, that’s important. CM17 However, competence (i.e. ability) did play a role in trustworthiness after the fire. Trust was increased when agency personnel were ‘knowledgeable about all of the various services available’. Trust also increased when personnel could anticipate community members’ needs and recommend particular recovery services on a one-on-one basis. Participants described community-agency trust as being strengthened when recovery assistance programs were deployed without delay and met immediate and basic needs at personal and community levels. These programs demonstrated to residents that


the government and its agencies could be trusted, or relied upon, because they were ‘were looking after’ communities and ‘cared about the community and whether they recovered’. Trust was further strengthened when agencies and agency personnel followed through on funding promises, assistance requests and resident enquiries. You trusted *** because she followed it through.whatever she said.. She was the only one who did that. CM11 Some participants raised concern that recovery periods commenced and concluded for residents at different times. These participants felt trust could be lost in the future if on-going and long-term needs resulting from the fire weren’t met. I’m aware of people [nine months after the fires] saying, ‘oh yeah we went up to the back block last week and it was the first time we’ve been up there to look at the damage from the fire’, and they obviously couldn’t pull themselves to do that earlier. So I don’t know if it’s recognised that it really does take people such a long time. CM26 Negative outcomes experienced during the fire had the potential to influence trust after a fire if participants desired a sense of closure but did not feel they were able to reach that. Providing closure did not necessarily mean compensation for losses suffered as a result of the fires or extensive rehabilitation and repair of damaged property. Most of these participants sought explanations for why certain events and actions, or lack of action, occurred. Most participants who felt they had not received adequate closure after the fire for these outcomes described a complete loss of trust in the agency they perceived as responsible. .in a couple of words, why.. That’s all I want.. Twelve months down the line, that’s probably my biggest issue that I still can’t get over. Not even a letter of e not necessarily apology, but you know, an explanation: ‘this is why.’. CM14 5.4. Similarities and differences among factors One of our objectives was to compare the factors affecting trust among stages. The findings showed that participants described institutional structures and reducing resident uncertainty as particularly influential in building or losing trust during a fire. After a fire, factors related to community recovery, such as resolving negative outcomes suffered during the fire and providing immediate assistance for community and individual needs, were important to trust. While these factors may reflect the different contexts of each stage, some factors were important among all three stages. However, the findings also showed some subtle differences in these factors across the stages (Table 4). Most participants who described communication as important to community-agency trust in all management stages also described how the amount and direction of information flow which contributed to trust differed in each stage. For example, before a fire, participants desired a multi-directional flow of information between agencies (e.g. planning), from agencies to community members (e.g. provision of property protection education) and from community members to agencies (e.g. identifying local assets, fire hazards and other concerns). However, during a fire, participants described an increase in trust with a one-way communication process where unlimited amounts of timely, reliable and accurate information flowed from agencies to communities. In contrast, after a fire, too much information was seen as overwhelming, with participants stating that residents would be more likely to rely on, or trust, agencies if information was provided one-on-one and residents were able to ask questions specific to their circumstances.

17

Many participants described how community-agency trust was strengthened when agencies recognised the importance of local concerns in fire planning and preparation (i.e. before a fire) and integrated local knowledge into the fire response (i.e. during a fire). Conversely, community-agency trust was diminished when fire planning and preparation was ‘managed by outsiders’ and the fire response was ‘organised away from the fire front’. The differences in this factor for each stage related to who participated in decisionmaking. Trust was strengthened before a fire when community members felt they ‘had a say’ in fire planning and land management decisions. In contrast, during the fire, most, but not all, participants were willing to ‘let the experts make the decisions’. However, these participants were more likely to describe positive trusting relationships when they felt that local knowledge was integrated into fire management decisions made during the emergency. Participants described how trust could be built if agency cooperation was visible to the public. However, the level of this visibility and how it was demonstrated subtly differed across stages. Before a fire, participants described agency cooperation being demonstrated through joint-branded publications and a multi-agency presence in public participation and education processes. During a fire, cooperation could be shown by DSE and CFA crews working together and with other relevant agencies at the fire front and staging ground. In contrast, after a fire, participants stated that joint visits by multiple agency representatives could be overwhelming given the exhaustion present in that stage, but coordination of ‘one-stop shop’ arrangements with information available from multiple agencies via one representative was conducive to building trust. Characteristics of trustworthiness were common across agencies and were also similar whether participants were describing neighbours, agency staff members, or agencies and organisations as a whole. However, the most common trustworthy characteristics described by participants were different in each management stage. Before a fire, characteristics related to integrity, such as transparency and credibility, were most frequently described. During a fire, the findings showed that participants most frequently described ability attributes. Participants also frequently described benevolence in this stage. Agencies were considered trustworthy when they demonstrated competence and skill in fighting fires and made decisions perceived to have the community’s best interests at heart. Finally, after a fire the characteristics most frequently described tended to fall into the category of benevolence, followed by ability and integrity (e.g. follow-through). 6. Discussion Numerous authors in the wildfire management context have noted the significance of trust to positive community-agency relationships and public support for wildfire management strategies. Findings from this research confirm the importance of communication, agency cooperation and trustworthiness in all stages of fire management and identify some factors, such as institutional structures, as important to trust for specific management stages in an Australian context. First, our research confirms that communication is a critical factor influencing trust in each management stage. Importantly, our research suggests that our participants desired a multi-directional information flow before the fire. This confirms previous research which has found that even though community members find oneway communication approaches (e.g. brochures, television ads) useful and trustworthy, interactive approaches (e.g. interpretive centres and guided field trips) are rated as significantly more useful sources of information (Bushnell and Cottrell, 2007; Shindler et al., 2009; Toman et al., 2008). A multi-directional information flow also encourages community members to ask questions and express

18


concerns when communicating with fire management agency and government staff (McCaffrey, 2004) thereby building a sense of community, commitment to action and a sense of collective and individual efficacy (Paton, 2000). Previous research also suggests that interactive forms of communication, such as agency-led field trips to view fire-affected areas of public land (Olsen and Shindler, 2010; Toman et al., 2008), promote positive agency-community relationships after a fire which can encourage community recovery and healing (Burns et al., 2008). Our research differs in that our participants primarily reflected on actions and events which promoted community-agency trust in the immediate aftermath of the fires. Our research suggests that in this context, attempting to develop community-agency partnerships for forest rehabilitation and planning in the immediate aftermath of a fire may be seen as premature when community members are still exhausted and overwhelmed by the event and its impacts on their own properties. However, participants in our research also showed concern that there was not enough attention paid to long-term recovery assistance for emotional needs. It is possible this concern could be eased by integrating interactive forms of communication, as suggested by Toman et al. (2008) and Olsen and Shindler (2010), and community participation in forest restoration (Burns et al., 2008) with existing recovery programs. Therefore, our research suggests that the timing of different forms of communication after a fire is critical. For situations similar to our research context, the key challenge may be gauging when community members can move forward from requiring targeted and limited information directed at individual needs to coping with the demands of participating in forest rehabilitation and planning processes. Second, findings from the trustworthiness factor support the Mayer et al. (1995) typology of ability, benevolence and integrity as overarching dimensions of trustworthiness. Mayer et al. (1995) propose that individuals determine the trustworthiness of others based on their beliefs in the trustee’s ability (i.e. knowledge, skills and competencies), benevolence (i.e. the extent to which a trustor believes that a trustee will act in the best interest of the trustor), and integrity (i.e. the extent to which the trustor perceives the trustee as acting in accord with a set of values and norms shared with or acceptable to the trustor). These dimensions encompass the characteristics of care and competence identified by Winter et al. (2004) in their study of trust’s influence on the social acceptance of fuel management strategies. Our research also suggested that even though all three dimensions of trustworthiness could be found in participant descriptions for each management stage, the dimensions of trustworthiness most frequently described differed in each stage. These findings appear to confirm previous theoretical work which suggests that context may determine which trustworthy dimension contributes most to perceptions of trustee trustworthiness when building and maintaining trust (Mayer et al., 1995; Gill et al., 2005). For example, it may be that aspects of integrity (i.e. shared norms and values) are most important to trustworthiness before a fire when community members seek open and inclusive planning processes which integrate their concerns. Previous research supports the idea that shared values play a central role in trust when community members feel agency actions and decisions reflect similar values to their own (Cvetkovich and Winter, 2007; Davenport et al., 2007; Liljeblad et al., 2009). Third, our participants suggested that institutional structures were important in providing a base for trusting the agencies in the response stage through the perception that agencies had sound policies for responding to a fire. This supports previous research that states that reliable institutions, such as policies and procedures, create a more trustworthy environment at all levels (Rus

and Iglic, 2005) which can contribute to the formation of interpersonal trust (Zucker, 1986) or substitute for interpersonal trust when it is absent (Luhmann, 1988). This was particularly important in our research context since not all community members had relationships with fire agency staff prior to the 2006e07 fires. While institutional structures were particularly important to our research participants during the fire, Ryan and Klug (2005) state that trust is needed at all levels, including at institutional and interpersonal levels, for actions and partnerships to endure. This suggests that an emphasis on community engagement before a fire that provides community members opportunities to build relationships with local staff and participate in fire preparation and planning activities may help trust relationships endure negative outcomes encountered during a fire. Similarly, our research highlighted the importance of addressing local concerns and including local input into fire planning before a fire, a finding supported by research in other contexts which has associated trust with inclusion in decision-making processes seen as fair and equitable (Leahy and Anderson, 2008; Shindler et al., 2002). Finally, our findings show that trust is influenced by several factors in each stage. This confirms a growing consensus in the broader literature (e.g. Dietz and den Hartog, 2006; Lewicki et al., 2006; Mayer et al., 1995; Rousseau et al., 1998) and NRM literature (e.g. Davenport et al., 2007; Liljeblad et al., 2009) that trust is a multi-dimensional concept which is more inclusive than decision heuristics related to shared values (e.g. Cvetkovich and Winter, 2003; Cvetkovich and Winter, 2007; Vaske et al., 2007). 7. Conclusion Trust is an important part of community-agency partnerships and public support for wildland fire policy, but little attention has been paid to simultaneously identifying factors affecting trust in each stage of a wildfire event. In this study, we identified and compared factors affecting trust before, during and after a prolonged fire event in Victoria, Australia. Communication, cooperation, trustworthiness, and integration of local concerns and knowledge influenced trust in more than one fire management stage. Institutional structures and reduction of uncertainty were particularly strong influences during a fire. After a fire, resolving negative outcomes and immediately meeting perceived needs arising from the fire were factors unique to this stage. At a practical level, these sets of factors provide agencies an opportunity to influence community-agency trust in each management stage. Agencies can concentrate on factors affecting trust identified for each stage, as well as the different ways common factors work in a particular management stage. In doing so, agencies can target their community engagement strategies to each stage while still maintaining the ‘big picture’ of community-agency trust for the event as a whole. Acknowledgements Charles Sturt University funded this study. The lead author wrote the paper while undertaking a post-doctoral fellowship with the Australian National Centre for Groundwater Research and Training. We would like to thank Simon McDonald, Spatial Analysis Unit, Charles Sturt University for assistance in creating the study area map. References Absher, J.D., Vaske, J.J., 2011. The role of trust in residents’ fire wise actions. Int. J. Wildland Fire 20, 318e325.

E.A. Sharp et al. / Journal of Environmental Management 130 (2013) 10e19 Australian Bureau of Statistics, 2006. 2006 Census Quick Stats (Postal Area). http:// www.censusdata.abs.gov.au (accessed 17.07.08.). Bachmann, R., 2006. Trust and/or power: towards a sociological theory of organizational relationships. In: Bachmann, R., Zaheer, A. (Eds.), Handbook of Trust Research. Edward Elgar, Cheltenham, UK, pp. 393e408. Baral, N., 2012. Empirical analysis of factors explaining local governing bodies’ trust for administering agencies in community-based conservation. J. Environ. Manage. 103, 41e50. Berney, L., Blane, D., 1997. Collecting retrospective data: accuracy of recall after 50 years judged against historical records. Soc. Sci. Med. 45, 1519e1552. Beunen, R., de Vries, J.R., 2011. The governance of Natura 2000 sites: the importance of initial choices in the organization of planning processes. J. Env. Plan. Manage. 54 (8), 1041e1059. Blomqvist, K., 1997. The many faces of trust. Scand. J. Manage. 13 (3), 271e286. Brenkert-Smith, H., Champ, P.A., Flores, N., 2006. Insights into wildfire mitigation decisions among wildland-urban interface residents. Soc. Nat. Res. 19, 759e768. Brunson, M.W., Evans, J., 2005. Badly burned? Effects of an escaped prescribed burn on the social acceptability of wildland fuels treatments. J. For. 103, 134e138. Burns, M.R., Taylor, J.G., Hogan, J.T., 2008. Integrative healing: the importance of community collaboration in postfire recovery and prefire planning. In: Martin, W.E., Raish, C., Kent, B. (Eds.), Wildfire Risk: Human Perceptions and Management Implications. Resources for the Future, Washington, DC, pp. 81e97. Bushnell, S., Cottrell, A., 2007. Increasing community resilience to bushfire e implications from a north Queensland community case study. Aust. J. Emer. Manage. 22 (2), 3e9. Cook, K.S., 2001. Trust in society. In: Cook, K.S. (Ed.), Trust in Society. Russell Sage Foundation, New York, pp. xiexxvii. Cooke, B., Langford, W.T., Gordon, A., Bekessy, S., 2011. Social context and the role of collaborative policy making for private land conservation. J. Env. Plan. Manage. 55 (4), 469e485. Cvetkovich, G., Winter, P.L., 2003. Trust and social representations of threatened and endangered species. Environ. Behav. 35 (2), 286e307. Cvetkovich, G.T., Winter, P.L., 2007. The what, how, and when of social reliance and cooperative risk management. In: Siegrist, M., Earle, T.C., Gutscher, H. (Eds.), Trust in Cooperative Risk Management: Uncertainty and Scepticism in the Public Mind. Earthscan, London, UK, pp. 187e209. Davenport, M.A., Leahy, J.E., Anderson, D.H., Jakes, P.J., 2007. Building trust in natural resource management within local communities: a case study of the Midewin National Tallgrass Prairie. Environ. Manage. 39, 353e368. Dietz, G., den Hartog, D.N., 2006. Measuring trust inside organisations. Pers. Rev. 35 (5), 557e588. Gill, H., Boies, K., Finegan, J.E., McNally, J., 2005. Antecedents of trust: establishing a boundary condition for the relation between propensity to trust and intention to trust. J. Bus. Psychol. 19 (3), 287e302. Gray, S., Shwom, R., Jordan, R., 2012. Understanding factors that influence stakeholder trust of natural resource science and institutions. Environ. Manage. 49, 663e674. Hosmer, L.T., 1995. Trust: the connecting link between organizational theory and philosophical ethics. Acad. Manage. Rev. 20 (2), 379e403. Hudson, B., 2004. Trust: towards conceptual clarification. Aust. J. Polit. Sci. 39 (1), 75e87. Kramer, R.M., 1999. Trust and distrust in organizations: emerging perspectives, enduring questions. Annu. Rev. Psychol. 50, 569e598. Kumagai, Y., Daniels, S.E., Carroll, M.S., Bliss, J.C., Edwards, J.A., 2004. Causal reasoning processes of people affected by wildfire: implications for agencycommunity interactions and communication strategies. West. J. Appl. For. 19 (3), 184e194. LaChapelle, P.R., McCool, S.F., 2012. The role of trust in community wildland fire protection planning. Soc. Nat. Resour. 25, 321e335. Leahy, J.E., Anderson, D.H., 2008. Trust factors in community-water resource management agency relationships. Landsc. Urban Plann. 87, 100e107. Lewicki, R.J., Tomlinson, E.C., Gillespie, N., 2006. Models of interpersonal trust development: theoretical approaches, empirical evidence, and future directions. J. Manage. 32, 991e1022. Liljeblad, A., Borrie, W.T., Watson, A.E., 2009. Determinants of trust for public lands: fire and fuels management on the Bitterroot National Forest. Environ. Manage. 43, 571e584.

19

Luhmann, N., 1988. Familiarity, confidence, trust: problems and alternatives. In: Gambetta, D. (Ed.), Trust: Making and Breaking Cooperative Relations. Basil Blackwell, New York, pp. 94e107. Mayer, R.C., Davis, J.H., Schoorman, F.D., 1995. An integrative model of organizational trust. Acad. Manage. Rev. 20 (3), 709e734. McCaffrey, S., 2004. Thinking of wildfire as a natural hazard. Soc. Nat. Res. 17, 509e 516. Mendez, S.R., Carroll, M.S., Blatner, K.A., Findley, A.J., Walker, G.B., Daniels, S.E., 2003. Smoke on the hill: a comparative study of wildfire and two communities. West. J. Appl. For. 18 (1), 60e70. Mollering, G., 2006. Trust: Reason, Routine, Reflexivity. Elsevier, London. Olsen, C.S., Shindler, B.A., 2010. Trust, acceptance and citizen-agency interactions after large fires: influences on planning processes. Int. J. Wildland Fire 19 (1), 137e147. Paton, D., 2007. Preparing for natural hazards: the role of community trust. Disaster Prevent. Manage. 16 (3), 370e379. Paton, D., 2000. Emergency planning: integrating community development, community resilience and hazard mitigation. J. Americ. Soc. Prof. Emer. Manage. 7, 109e118. Public Submission No. 243 Plantations Northeast, 2008. Report on the Impact of Public Land Management Practices on Bushfires in Victoria. State of Victoria, Melbourne, Victoria, Australia. Punch, K.F., 2005. Introduction to Social Research: Qualitative and Quantitative Approaches, second ed. Sage Publications, Thousand Oaks, CA. Richards, L., 2005. Handling Qualitative Data: a Practical Guide. Sage, London. Rousseau, D.M., Sitkin, S.B., Burt, R.S., Camerer, C., 1998. Not so different after all: a cross-discipline view of trust. Acad. Manage. Rev. 23 (3), 393e404. Rubin, H.J., Rubin, I.S., 2005. Qualitative Interviewing: the Art of Hearing Data, second ed. Sage Publications, Thousand Oaks, CA. Rus, A., Iglic, H., 2005. Trust, governance and performance: the role of institutional and interpersonal trust in SME development. Int. Sociol. 20 (3), 371e391. Ryan, C.M., Klug, J.S., 2005. Collaborative watershed planning in Washington state: implementing the Watershed planning act. J. Env. Plan. Manage. 48 (4), 491e 506. Safford, T.G., Ulrich, J.D., Hamilton, L.C., 2012. Public perceptions of the response to the Deepwater Horizon oil spill: personal experiences, information sources, and social context. J. Environ. Manage. 113, 31e39. Sharp, E., Thwaites, R., Curtis, A., Millar, J., 2012. Trust and trustworthiness: conceptual distinctions and their implications for natural resources management. J. Env. Plan. Manage.. http://dx.doi.org/10.1080/09640568.2012.717052. Shindler, B., Aldred-Cheek, K., 1999. Integrating citizens in adaptive management: a propositional analysis. J. Conserv. Biol. 3 (1), 9e13. Shindler, B., Brunson, M., Stankey, G.H., 2002. Social Acceptability of Forest Conditions and Management Practices: a Problem Analysis. USDA Forest Service, Pacific Northwest Research Station, Portland, OR. General Technical Report PNW-GTR-537. Shindler, B., Toman, E., McCaffrey, S., 2009. Public perspectives of fire, fuels and the forest service in the Great Lakes Region: a survey of citizen-agency communication and trust. Int. J. Wildland Fire 18 (2), 157e164. Siegrist, M., Cvetkovich, G., 2000. Perception of hazards: the role of social trust and knowledge. Risk Anal. 20 (5), 713e719. Six, F., 2005. The Trouble with Trust: the Dynamics of Interpersonal Trust Building. Edward Elgar, Cheltenham, UK. Smith, R., 2007. Key Issues Identified from Operational Reviews of Major Fires in Victoria 2006/07. State of Victoria, Melbourne. Toman, E.L., Shindler, B., Absher, J., McCaffrey, S., 2008. Postfire communications: the influence of site visits on local support. J. For. 106 (1), 25e30. Vaske, J.J., Absher, J.D., Bright, A.D., 2007. Salient value similarity, social trust and attitudes toward wildland fire management strategies. Hum. Ecol. Rev. 14 (2), 223e232. Vogt, C.A., Winter, G., Fried, J., 2005. Predicting homeowners’ approval of fuel management at the wildland-urban interface using the theory of reasoned action. Soc. Nat. Res. 18, 1e18. Winter, G.J., Vogt, C., McCaffrey, S., 2004. Examining social trust in fuels management. J. For. 102 (6), 8e15. Zucker, L.G., 1986. Production of trust: institutional sources of economic structure, 1840e1920. Res. Org. Behav. 8, 53e111.