THE BUSHFIRE RISK: COMMUNITY AWARENESS AND PERCEPTION ON THE RURAL-URBAN FRINGE OF MELBOURNE

GEOGRAPHY 3001

FULL UNIT DISSERTATION

The bushfire risk: community awareness and perception on the rural-urban fringe of Melbourne.

by John B. Gilbert

March 2004

(Supervisor: Dr. Carl Sayer)

Abstract

Bushfires pose a clear and present danger to residents living on the rural-urban fringe of cities in South Eastern Australia. They cause widespread damage to property, put people’s lives at risk and have environmental, social and economic repercussions. This research looks at two such communities on the margins of Melbourne: Upper Beaconsfield and Diamond Creek. It examines the level of awareness and perception of the fire risk amongst the communities and relates this to preparedness for such an event. The major focus of the research is on Upper Beaconsfield, where 200 residents were surveyed and six in-depth interviews undertaken. The Ash Wednesday fires of February 1983 badly affected this community. Diamond Creek offered an interesting comparative study, as there had been no serious fire in the region since 1962. Here 50 residents were surveyed to see if significant differences were apparent. The results derived from a mixture of quantitative and qualitative methodologies. They highlight a greater level of awareness of the bushfire hazard in Upper Beaconsfield in comparison to Diamond Creek. However, they also demonstrate clear spatial differences in the level of awareness amongst the Upper Beaconsfield community. Perception of the risk is influenced by several variables, most notably past experience of major fires. New residents are more likely to be less concerned about adopting appropriate bushfire strategies and are also more prone to underestimating the fire risk. Efforts to increase levels of bushfire education and self-reliance appear to have had some effect. However, there is still a degree of complacency amongst some and over-reliance on the Country Fire Authority by others. Overall, this study suggests a need for considerable improvement in bushfire awareness amongst the Diamond Creek community. In addition, whilst clear progress has been made in Upper Beaconsfield over the past twenty years, a significant number of residents are not as prepared as they should be.

The number of words in the main text amounts to 10,000.

(Source of picture on cover: www.ema.gov.au)

Acknowledgements

The author would like to thank all those have helped make this research possible. Special thanks go to:

1 David Packham for his advice and encouragement.

2 Elinor Niall at the Country Fire Authority for providing me with valuable information.

3 Emily, Judy, Sarah, Laura and Robert Ballantyne-Brodie for their help with the distribution and collection of the questionnaires and for providing me with somewhere to stay.

4 Mary and Brian McDonald for their help and local knowledge of Diamond Creek.

5 The communities of Upper Beaconsfield and Diamond Creek for their willingness to participate.

6 Dr. Carl Sayer for his support and advice.

Table of Contents

1.	Introduction ………………………………………………………….…	1
1.1.	Context…………………………………………………………………..	1
1.2.	The Need For Research………………………………………………….	1
2.	Bushfires: Risk Awareness and Perception …………………………..	3
2.1.	Risk Awareness and Perception………………………………………….	3
2.2.	Natural and Man-Made Hazards…………………………………………	4
2.3.	Bushfires in a Global Context…………………………………………..	5
2.4.	Bushfires in Australia ……………………………………………………	6
2.5.	Aims…………………………………………………………………….	12
3.	Study Areas …………………………………………………………….	13
3.1.	Upper Beaconsfield ……………………………………………………..	13
3.2.	Diamond Creek………………………………………………………….	14
4.	Methodology ……………………………………………………………	15
4.1.	Approach………………………………………………………………..	15
4.2.	Questionnaires…………………………………………………………..	15
4.3.	Interviews……………………………………………………………….	17
5.	Results ………………………………………………………………….	18
5.1.	Initial Findings…………………………………………………………..	18
5.2.	Risk Awareness and Perception………………………………………….	19
5.3.	Behavioural Response to the Bushfire Risk…………………………….	23
5.4.	Spatial Patterns………………………………………………………….	26
6.	Discussion ………………………………………………………………	31
6.1.	Initial Findings…………………………………………………………..	31
6.2.	Risk Awareness and Perception………………………………………….	32
6.3.	Behavioural Response to the Bushfire Risk…………………………….	36
6.4.	Spatial Patterns………………………………………………………….	41
7.	Conclusions …..…………………………………………………………	44
7.1.	The Research…………………….………………………………………	44
7.2.	Suggestions for Future Work…………………………………………….	45
8.	Appendices ……………………………………………………………..	47
8.1.	Pakenham -Berwick Gazette Article: “ Fighting back after flames ”…….	47
8.2.	Sample Questionnaire…………………………………………………..	48
8.3.	Results Summary……………………………………………………….	50
8.4.	Interview Schedule………………………………………………………	52
8.5.	Interview Summaries……………………………………………………	53
9.	Auto-Critique ………………………………………………………….	56
10.	Bibliography ……………………………………………………………	58

1. Introduction

1.1. Context

In January 2003 bushfires engulfed the suburbs of Canberra causing widespread devastation to the people living in Australia’s capital. The scenes of destruction challenged the notion of Australian suburban bliss. By February the fires were also threatening communities on the rural-urban periphery of three other Australian cities: Sydney, Melbourne and Adelaide. The severity of the fire season in 2003 was exceptional, especially in terms of the loss of property and livelihood. Scenes such as those shown in Figure 1 brought the situation to international attention. It prompted a House of Representatives Select Committee Report entitled “A Nation Charred” (October 2003). This collected evidence and received over five hundred submissions examining what had gone wrong and what lessons could be learnt. The bushfires were certainly not a one-off event. Australia has endured other catastrophic bushfires in the past, most notably the Ash Wednesday fires in February 1983. In fact, no other continent is quite as susceptible to fire as Australia (McKnight, 1995: 38). However, it has reopened the debate on bushfire management and provided an opportunity to devise strategies to mitigate future bushfire disasters.

Figure 1: The Canberra Fires: Destruction and Loss

Source: BBC, 2003

1.2. The Need for Research

A major part of improving the management of bushfires is gaining a more detailed understanding of the issues at hand. Research is a vital element in this process. Much of the existing literature on bushfires has explored the scientific aspects of their nature and occurrence (Luke and McArthur, 1978; Wilson, 1988). However, there is also a growing literature that is examining the issue from a social perspective (Lazarus and Elley; 1984; Wilson, 1984; Beringer, 2000). A good deal has been written about the need for further studies in this area with calls in particular for more research into the social responses to fire survival (Packham, 1992: 12). In addition, the Country Fire Authority (CFA) has suggested that more research studies into community responses to emergencies are required (Reinholdt et al., 1999a: 2). Consequently, there is a need to examine community awareness and perception with regards to bushfires. In doing so, it can be related to the wider literature on environmental risk perception that links people’s knowledge and environment to how risks are perceived (Burton et al. 1978, 1993, Smith, 2001).

With this in mind the present study sets out to examine two suburbs on the rural-urban fringe of Melbourne: Upper Beaconsfield and Diamond Creek. Through the use of both quantitative and qualitative primary data analysis, a profile of people’s attitudes towards the risk posed by bushfires will be developed. It will also examine the spatial and temporal factors that influence bushfire risk perception. The study discusses the findings in relation to other similar research and relates it to the wider geographical issues. It concludes by drawing from the empirical findings some practical suggestions for improving bushfire awareness in vulnerable communities.

2. Bushfires: Risk Awareness and Perception

2.1. Risk Awareness and Perception

Risk is a part of daily life at varying magnitudes and in different ways. It is commonly distinguished from hazard, the latter being the general vulnerability from an adverse event and the former being the probability of the hazard occurring (Royal Society, 1983, cited in Adams, 1995; Johnston et al., 2000). People are constantly subjected to risks, some self-inflicted and others impossible to avoid. The human response to risk is influenced by people’s awareness and perception of it (Clark, 1991: 8, Burton et al., 1993: 47, Löfstedt et al., 1998: 3). This significantly influences whether people are suitably prepared for a risk they are exposed to and determines whether they can make a rational decision in response to the impending threat.

How aware people are of a risk and how they perceive it depends on a range of factors including the type, regularity, potential outcome and importance of the risk. It is also dependent on an individual’s level of risk tolerance, knowledge, communication and of changing social views of risk (Smith, 2001: 59). When the risk is associated with a daily occurrence such as crossing a road, people are very aware of the dangers and take appropriate actions to minimise the risk. For example, they cross the road when it is safe to do so (Adams, 1995: 1). However, many risks are not daily events and might only be distant possibilities, which may be difficult to comprehend and for which there is only limited or partial understanding. In such a situation the question arises as to how aware people are of the potential risk they are exposed to? Many hazards fall into this category, both natural and man-made.

Risk is a multi-faceted research area. The particular focus of this research is on social aspects of risk perception. Since there is a need for risks “to be assessed in a more qualitative way” (Smith, 2001: 55), questionnaires and interviews are common research tools for examining people’s views and experiences of risk, not least amongst environmental management. There is a history of using surveys in environmental awareness studies. Saarinen and Cooke (1970) used this method in a study of the perception of environmental quality in Tucson, Arizona (cited in Edgell, 1973: 28). More recently Williams et al. (1999) used a large-scale telephone survey to examine risk perception concerning a range of environmental hazards as part of the Savannah River Stakeholder Study in the United States. Their findings implicated that researchers explore the determinants of risk perception with regards to environmental risks (Williams et al., 1999: 1033). The importance of examining social perspectives in the study of risks is further purported by Brown and Damery (2002) in their examination of managing flood risks. They call for the examination of social vulnerability and the importance of a more “socially informed approach” in risk management (Brown and Damery, 2002: 422).

2.2. Natural and Man-Made Hazards

Natural phenomena are diverse both spatially and in terms of their impact. They only become natural hazards when they interact with humans. The presence of a community transforms an event into a threatening situation in human terms (Keys, 1991: 1). Natural hazards, otherwise known as environmental hazards, are generally defined as geophysical events that can potentially cause large-scale economic damage and physical injury or death (Johnson et al, 2000: 216). Man-made hazards make up a second category of hazards, which are the result of human activities that generate a level of risk. Other hazards fall in a third category somewhere between the two, displaying characteristics of both a naturally occurring and human induced hazard (Clark, 1991: 7). It is into this category that bushfires fall, with fires generated both by lightning strikes and human activity. When interaction with humans occurs bushfires present a considerable potential risk to people. The implication of this is that the rural-urban fringe is extremely susceptible and often at the front line of danger.

Handmer and Penning-Roswell (1990) suggested a model of human response to a natural hazard and how the perceived threat is influenced by three main factors: non-hazard benefits, sense of community and ignorance to the threat (Figure 2). This model helps to illustrate why people live in risk-prone areas. Their perception of the risk is influenced by several factors. It is rare for people to be entirely unaware of the existence of a natural hazard. However, perceptions differ markedly between different people and between the public and expert opinion (Burton et. al., 1993: 47). The unpredictability, lack of alternatives, changing nature of the risk, belief that ‘lightning will not strike twice’ and the other non-hazard benefits are all factors in people’s perception (Clark, 1991: 12, Chapman: 1999, 1). The importance of community is also highlighted in the model, which illustrates that communities can adapt to the risk and implement effective strategies.

Figure 2: The behavioural response to a natural hazard

Source: Model based on Handmer and Penning-Roswell (1990)

2.3. Bushfires in a Global Context

Antarctica is the only continent that is totally devoid of the threat posed by bushfires. Globally, there are relatively few places that are highly susceptible to extreme fires and that are a danger to large concentrations of people. Although the risk to major population centres around the world that are at danger is increasing (Smith, 2001: 248). The coverage of bushfires in the media over the past year has provided an accurate geography of the major fire risk zones. In January and February of 2003 headlines such as “Four die as fires ravage Canberra” (The Daily Telegraph, January 20th 2003) and “Australia brace for new fire threat” (BBC, January 23rd 2003) highlighted the plight of many residents living in the suburbs of Canberra and other parts of Australia. By July reports of large fires in Mediterranean Europe, especially Southern France and Portugal were in the news coinciding with the prolonged heat wave. Most recently in October 2003 it has been California that has been suffering from bushfires. Thus, there is a spatial diversity of extreme fire events around the globe where large periods of dry, hot weather and large amounts of litter (fuel) fanned by strong winds result in high intensity, highly devastating fires.

Australia is widely regarded as the most fire-prone country in the world (Pyne, 1991; Packham, 1992: 1; Beringer, 2000: 2; Smith, 2000: 240). Some of the most intense conflagrations have been witnessed there (Bryant, 1990, 158). This made it the ideal focus for the research undertaken here. The lack of major bushfires in Australia over the most recent summer period (December 2003 to March 2004) demonstrates that the events are fortunately not yearly occurrences. However, this also shows their unpredictability and the difficulty in planning for their eventuality.

2.4. Bushfires in Australia

South-Eastern Australia is particularly vulnerable to the threat posed by bushfires due to a combination of highly flammable sclerophyllous vegetation (especially Eucalyptus), high summer temperatures (reaching 40ºc), frequent drought periods and large accumulations of dry litter (Beringer, 2000: 2). Fire is by no means a new phenomenon in Australia (Chapman, 1999: 27). Vegetation has adapted to regular burning over the past 100 million years. Whereas Aboriginals used fire to manage their environment, Europeans in the past have tried to control it. The result has been a large build up of fuel and an increased potential for fire (Smith, 1999: 103), as shown in Figure 3. As it builds up the potential for destruction from future catastrophic bushfires increases and the vulnerability to nearby communities is exacerbated. The net result can be a high intensity fire that is very difficult to manage. To put it in context, the energy generated by the fires around Omeo (North East Victoria) in February 2003 was the equivalent of five nuclear explosions (Packham, 2003). This is a remote part of Victoria, with a very low population density but it destroyed thousands of hectares of forest (Figure 4) and affected rural farming communities.

As the population has grown in Australia over the past couple of centuries the risk that fire poses has become increasingly evident. Increasing numbers of people live in close proximity to the bush due to its aesthetic appeal (Chapman, 1999: 42) and other reasons such as those put forward by Handmer and Penning-Roswell in Figure 2. In addition, there are growing population pressures pushing the suburbs further into the bush. It is estimated that 80% of the Australian population live in urban or semi-rural areas (Bentley, 2003, cited in House of Representatives Select Committee, 2003: 259). The net result is an ever-increasing risk of property destruction, loss of life, loss of livestock and loss of livelihoods on the rural-urban fringe.

Figure 3: Dense sclerophyllous vegetation (Critchley Park, Upper Beaconsfield)

Source: Author’s Collection

Figure 4: Vegetation in the aftermath of a fire (Omeo, August 2003)

Within Australia, Victoria is known as the “fire-state”, a reputation based on its history of devastating fires including ‘Black Friday’ (13^th January 1939) and the ‘Ash Wednesday’ fires on February 16^th 1983. The effects of the latter are summarised in Table 1.

Table 1: The cost of ‘Ash Wednesday’ in Victoria

Death Toll	45
Property Loss	1719
Estimated Direct Losses	A$200 million
Farmland destroyed	330,000 hectares
Livestock loss	25,000

Source: Compiled from The Age: 1983

These figures help to illustrate the inherent dangers faced by communities living on the rural-urban fringe of Melbourne. Perhaps because of the fire history in this state it is here that a lot of research on the topic has centred. A further important element to consider regarding Ash Wednesday was that it turned out to be a landmark event in bushfire management. Out of the devastation came a series of pieces of legislation and a greater emphasis on educating fire-prone communities so as to empower them with the necessary knowledge to minimise future risks. A number of awareness campaigns and local community schemes have been initiated in recent years, such as Community Fireguard. This is a scheme that seeks to empower local communities at street level (Boura, 1998). In doing so people more actively partake in learning. This gradually develops knowledge and perception over time (Reinholdt et al., 1999a: 58).

A large amount of the Ash Wednesday devastation occurred in the small community of Upper Beaconsfield (Figure 5). Consequently, in the aftermath of the event considerable attention was paid to this rural suburb, both by the media and management groups. The immediate priority was to learn lessons from this experience in order to reduce the risks from future events. Krusel and Petris (1992) examined thirty-two of the deaths in the area concluding that strategies based on community groups would best help to minimise future deaths from such disasters (Krusel and Petris, 1992: 14). Lazarus and Elley (1984) suggested that an effective survival strategy could be based around staying at home, which offers the advantage of increasing the chances of saving properties. This is an important consideration, as people are not only putting their lives at threat on the rural-urban fringe but also their livelihoods. If homes can be saved at minimal risk to residents then it is a better solution as it reduces the cost of a fire still further. Certain stipulations such as appropriately constructed homes and access to sufficient water supplies are emphasised in the study. However, if these conditions can be met then saving lives, many of which were lost due to last minute evacuations, and saving property can go hand-in-hand. Packham (1992) further advocates that an appropriately designed home “protects the occupants, who then protect the house” (Packham, 1992: 11). Thus, the result was a change in bushfire management and post-Ash Wednesday community involvement became the ‘buzzword’. This led to new bushfire survival strategies being devised. The question remains as to how much better prepared people living in Upper Beaconsfield are today.

Figure 5: A typical residence in Upper Beaconsfield

Source: Author’s Collection

Edgell (1973: 45) discovered that people’s level of experience and their place of residence had an important impact on levels of awareness and perception in the Dandenong Ranges. Other factors he examined, such as gender, appeared to have less of an influence. The study also notes that the research was undertaken in the winter months of June and July, in other words the low fire-risk period in the state of Victoria (as shown in Figure 6). This may also be a factor in the study being undertaken here and, as such, is worth taking into account in terms of how it might affect the results.

Figure 6: Bushfire seasons in Australia

Beringer (2000) looked at North Warrandyte, another rural-urban fringe community in Melbourne, which had suffered from a fire in 1991. Via an extensive questionnaire he examined community fire safety and once again the need for community education was highlighted as an area needing attention. A further interesting issue to arise was that of the expectation that “the fire authorities would protect individual homes during a bushfire” (Beringer, 2000: 1). Half the respondents said they did expect this. Whilst the CFA protect as many homes as they can during a serious fire, they have to prioritise their resources on tackling the fire front and saving people’s lives. Through education, communities become aware of this and see the benefit in such schemes as Community Fireguard and also owning fire-fighting equipment to save their own homes. Therefore, Beringer’s paper further illustrates the need for a well thought out bushfire strategy and this is a further area that the present research will examine. Nearby to North Warrandyte is the suburb of Diamond Creek (Figure 7). Two qualitative assessments of the fire hazard were undertaken here last year at the request of the Shire of Nillumbik’s Ratepayers Association (Incoll, 2003; Packham, 2003). This work emphasised the need for reductions in accumulations of ground fuels, the scraping of prohibitive planning policies and a need for greater understanding of the levels of awareness and preparedness of the residents. In his report, Packham recommends that “an independent survey of attitudes of residents to the bushfire threat and their preparedness” should be undertaken (Packham, 2003: 12). It is in reaction to this that Diamond Creek is included within the parameters of this research.

Figure 7: A typical street in Diamond Creek

Source: Author’s Collection

2.5. Aims

With this all in mind the following questions are set out in order to gain a deeper understanding of specific communities and their levels of awareness and perceptions of the bushfire risk.

1. How aware are the communities of Upper Beaconsfield and Diamond Creek to the risk of bushfires?

2. How do residents in the Upper Beaconsfield and Diamond Creek communities perceive the risk from bushfires and rationalise it?

3. How is the behavioural response to the potential risk of bushfires affected by levels of awareness and people’s perceptions of the hazard?

Figure 8: The remnants of a burning house on Ash Wednesday (Emerald-Beaconsfield Road)

Source: Murray and White, 1995

3.0 Study Area

3.1. Upper Beaconsfield

Upper Beaconsfield is a rural suburb located 48 km south-east of Melbourne’s CBD. (DOI, 2000). It was one of the worst affected areas on Ash Wednesday. Twenty-one people were killed and two hundred homes and businesses were destroyed. Despite this tragic past the village today is a thriving community, with a population of nearly 3000. A large proportion of the population is comprised of high-income households (DOI, 2000). Due to its location on the extreme edge of the rural-urban fringe of Melbourne, its manageable population size and its fire history, Upper Beaconsfield offered an ideal location for the research herein.

Map 1: The location of Upper Beaconsfield:

3.2. Diamond Creek

Diamond Creek is a larger suburb located 23 km north-east of the CBD of Melbourne. It is composed predominately of young families, with almost three quarters of the population under fifty years old (DOI, 2000). Diamond Creek is similar in character to Upper Beaconsfield in terms of its rural-urban fringe location and undulating relief. However, it has a larger population of about 12,000 (DOI, 2000) and has much more of a suburban design. Due to this the research focused on the part of the town considered at most risk (the area within the brown box), as it is adjacent to dense bush and a large amount of ground fuel found in PlentyValley.

Map 2: The location of Diamond Creek

4. Methodology

4.1. Approach

A mix of quantitative and qualitative methods have been utilised in this study. The advantage of such an approach is that both forms of data complement each other. For example, interviews cannot be as rigorously analysed as quantitative data. However, they do offer possible interpretations to trends found within the numbers. Likewise, questionnaires provide a large amount of nominal data to highlight the major trends and empirical findings. However, they do not enable more abstract relationships to be interpreted, for which the qualitative dimension is more suited (Flowerdew and Martin, 1997: 195). Therefore, combining the two approaches gives a more realistic interpretation of people’s awareness and perceptions of bushfires. Furthermore, it helps minimise the limitations of each form of data collection.

Lazarus and Elley (1984) adopted a qualitative approach for a bushfire study in Upper Beaconsfield. For their work they believed an in-depth investigation of a small number of people would enable them to best interpret their results (Lazarus and Elley, 1984: 3). This study had a narrow focus: the effect of household occupancy during the Ash Wednesday bushfire in Upper Beaconsfield. Edgell (1973) used quantitative analysis to examine the bushfire hazard in the Dandenong Ranges (close to Upper Beaconsfield) as it enabled comparisons between the sample populations (Edgell, 1973: 28). Clearly, Edgell’s scope was larger and therefore benefited from a wider range of data. Given the focus of this project lies somewhere between these two studies it seemed to be logical to combine the two approaches.

4.2. Questionnaires

A copy of the questionnaire used in this study is given in Appendix 2. The questionnaire was designed to provide a large amount of nominal data to analyse using a chi-square measure. The questions were based on background reading of CFA literature on bushfire issues, past surveys on risk perception (especially Beringer, 2000) and consultation with David Packham (a Rural Fire Consultant). The survey was designed to take no longer than two or three minutes to complete, in order to encourage as many people as possible to fill it in. The layout of the questions, which predominantly involved choosing from a list of options also made the questionnaire easy to respond to. Provision for additional responses was made where appropriate so as not to restrict people’s responses unduly.

A systematic method of sampling was used in which a questionnaire was delivered to one in three houses in every street in Upper Beaconsfield. In total three hundred questionnaires were delivered to resident’s mailboxes over a two-day period (August 13^th and 14^th 2003). Respondents were asked to leave their forms in their mailbox for collection over the next three days (August 15^th – 17^th). The logic of collecting over the weekend was to maximise the number of possible returns. The benefit of delivering questionnaires was that it enabled respondents to answer them without the pressure or bias of an interviewer looking on. To compensate for the lower response rate expected from this method reminder notes were delivered where questionnaires had not been completed. Additional copies of the questionnaire were also made available where necessary.

Once the information had been collated it was coded and entered into an Excel spreadsheet for analysis. Exploratory data analysis was undertaken and calculations of basic descriptive measures and cross tabulations were drawn up. The major statistical test to be used was the chi-squared statistic (χ²). Relationships were considered significant at a probability (p level) of <0.05 in line with Edgell (1973) and Beringer (2000). The use of a non-parametric test had the advantage of making fewer assumptions about the parameters of the populations from which the samples were drawn. Thus, providing a more robust numerical analysis (Flowerdew et al., 1997: 178).

Chi-square test

χ² = (O-E)²/ E

(O = observed frequencies, E = expected frequencies)

Degrees of freedom (df) = number of cells -1

The Diamond Creek survey was essentially the same with a few place specific alterations. The section on Ash Wednesday was not needed and was replaced with a question designed to test people’s knowledge of fire behaviour. Eighty questionnaires were delivered over a two-day period (September 3^rd and 4^th). A systematic sampling technique was applied but limited to the north west area of the suburb for practicality reasons (Map 2). Fifty responses were collected in total. The small sample size meant that statistical analysis was not appropriate for this data. However, it is possible to compare it with the Upper Beaconsfield survey in percentage terms. Thus it adds an interesting comparative aspect to the work.

4.3. Interviews

A copy of the interview schedule is given in Appendix 4. A question on the survey had asked for volunteers for in-depth interviews. Six were selected from the questionnaire responses from Upper Beaconsfield out of thirty-nine who had expressed an interest. Due to time limitations a cross section of the people and responses from the questionnaire was taken. Arrangements were made to conduct the half-hour long semi-formal interviews at the respondents homes during the week of August 18^th to 25^th. The conversations were taped and brief notes were also taken (in case of technical failure). The interview schedule guided the conversation but where additional interesting and useful points emerged these were further pursued. In addition several questions were tailored to each interviewee.

Interviews were subsequently transcribed to enable qualitative analysis of the data. Recurring themes were noted from the interviews and important quotes were compiled. The analysis involved using an interpretivist methodology (Reinholdt et al., 1999a; Lazarus and Elley, 1984). This is based on building a model of behaviour from the information collected. By doing so a social construct of people’s awareness and perception is developed (Reinholdt et al. 1999a: 4). As a result summary sheets of the major points arising from each interview were compiled (Appendix 5). In addition, quotes and findings were used to backup the quantitative element of the study. Interviews were also conducted with Elinor Niall, a Community Development Coordinator at the CFA and David Packham, a Rural Fire Consultant with forty years of experience in this field. This helped with gaining a deeper understanding of the topic and provided useful suggestions. For example, David Packham suggested the comparative study in Diamond Creek. These two interviews were not recorded but were important for background understanding of the issues in relation to Victoria.

5. Results

5.1. Initial Findings

A breakdown of the questionnaire collection is given in Table 2.

Table 2: Questionnaire Distribution and Collection

Upper Beaconsfield			Diamond Creek
Day	Delivered	Collected	Day	Delivered	Collected
Wed. 13//08/03	200		Wed. 3/09/03	80
Thu. 14/08/03	100		Thu. 4/09/03		50
Fri. 15/08/03		68
Sat. 16/08/03		82
Sun. 17/08/03		50
Total	300	200	Total	80	50

This shows that a return rate of 66.7% was achieved for Upper Beaconsfield and 62.5% for Diamond Creek. These figures compare favourably with Beringer (2000), who had a return rate of 45%, albeit with a larger distribution. However, 200 responses provided a good sample of the population of Upper Beaconsfield and the 50 responses from Diamond Creek make for a useful comparison. Overall, it shows an encouraging level of interest in these important issues.

The results of the hazard-rating question are compared in Table 3 with two other similar surveys: the North Warrandyte study (Beringer, 2000) and a study in the Dandenong Ranges (Raseta et al. 1987, cited in Beringer 2000).

Table 3: A Comparison of Studies on the Question of Hazard Rating

Hazard Rating	Upper Beaconsfield	Diamond Creek	North Warrandyte	Dandenong Ranges
Very High	39.5%	20%	52%	36%
High	39%	42%	42%	N/A
Moderate / Less	21.5%	38%	6%	N/A

A full breakdown of the questionnaire results for both Upper Beaconsfield and Diamond Creek are given in Appendix 3.

5.2. Risk Awareness and Perception

Several variables were explored to see what affect they had on people’s perception of the fire hazard and their understanding of the fire risk. These were the age of residents, the length of residency, gender, whether they had experienced Ash Wednesday and whether they had built their own home. The results that are given in Figures 9-14 highlight interesting differences between the findings in Upper Beaconsfield and Diamond Creek.

Age

The perceived risk from bushfires appears to be lower for young (18-25) and old residents (71+) of Upper Beaconsfield, with 37.5% and 39% respectively seeing the hazard rating as moderate (Figure 9). In contrast, the risk is perceived as greatest by the 41-55 age group, with 81% of this group considering the rating high or very high. The 71+ age group do also demonstrate the largest contrast with 46% considering the rating very high.

The results are somewhat different for Diamond Creek in Figure 10. Here a pattern more akin to that of Beringer (2000) materialises, with a decreasing perceived fire hazard with age.

Length of Residency

60% of residents who had been in the area for less than a year rated their knowledge of bushfires as moderate or less (Figure 11). In fact several mentioned on their survey form that they had not been living in Upper Beaconsfield for long enough to really know the situation. The percentage dropped gradually as the years of residency increased. For those who had been in Upper Beaconsfield for more than 21 years only 15% considered their knowledge as moderate and in contrast 42% believed their knowledge was very good. It was the only age group where a large percentage classified their knowledge in the highest category.

Other Variables

67% of women compared to 58% of men rated their knowledge as good or better in Upper Beaconsfield (Figure 12). However, a larger proportion of men rated their knowledge as very good. The result for Diamond Creek shows a slight advantage to the men, with 52% of males compared with 44% rating their knowledge as good or better. This result was reflected in Table 4, which tested respondent’s knowledge of fire behaviour. 74% of male respondents chose the right answer (upslope) compared to only 52% of women.

Table 4: The responses of males and females to the question: “when does fires travel fastest?”

	Female	Male
Upslope	52%	74%
Downslope	26%	9%
Level ground	22%	17%

76% of respondents who had built their own homes considered their knowledge as good or better in Upper Beaconsfield (Figure 13). Similarly in the Diamond Creek study found 79% of those who had, perceived their knowledge as at least good, compared to 69% of those who had not.

The chi-squared statistic was used to test the significance of these relationships, the results of which are given in Table 5. Relationships were considered statistically significant at the p=0.05 level. The data for Diamond Creek does not meet the criteria for being tested statistically due to the small sample size and is therefore omitted from this table.

Table 5: Summary of Chi-Squared Statistic Findings for Upper Beaconsfield

Variables	Tested Against:		Degrees of Freedom
	Bushfire Knowledge	Hazard Rating
Gender	6.07	5.23	5
Age	11.5 (p=0.2)	2.23	8
Length of residence	43.5 (p=0.001)	6.02	8
Ash Wednesday	31.8 (p=0.001)	1.66	5
Built home	19.46 (p=0.0025)	1.50	5
Bushfire Strategy	36.04 (p=0.001)	3.83	11

Ash Wednesday

One of the most statistically significant results to emerge from this table is the importance of Ash Wednesday. This is explored further in Figure 14 that compares bushfire awareness between residents who were present at the time of Ash Wednesday in Upper Beaconsfield and those who have moved in since. The Diamond Creek results are also included to add a further dimension.

5.3. Behavioural Response to the Bushfire Risk

This section examines the strategies that residents have in place to cope with a bushfire and their confidence in the management of bushfires. It also looks at the non-hazard benefits of living in Upper Beaconsfield and Diamond Creek.

Survival Strategies in Upper Beaconsfield

The percentage of people planning to evacuate, in Table 6, is less than in 1983 (31% compared to 59.7%). However, there are also 25.5% planning partial evacuations, where most of the family leave but someone stays. However, 17% are undecided and plan to ‘wait and see’.

Table 6: Have people’s survival strategies changed in the past 20 years

in Upper Beaconsfield?

	Strategies at the time of Ash Wednesday	Current Strategies
Stay at home	40.3%	26.5%
Evacuate	59.7%	31%
Wait and See	N/A	17%
Partial Evac .	N/A	25.5%

An important aspect of any survival strategy is preparation. Regular maintenance of properties and gardens helps minimise the risk of homes being destroyed in severe fires. The questionnaire asked how regularly residence carried out bushfire maintenance in the summer and the results are given in Figure 15.

These results show an encouraging number of people (41%) undertaking regular maintenance every one or two weeks. Some even added on the form that it was a part of their daily routine during the summer months. A further 35% cleared ground litter and guttering once a month. However, the remaining 24% said they only carried out such maintenance once during the summer period.

Confidence in the management of bushfires

The lower average ratings indicate a higher degree of confidence in the management levels. The results show that the Upper Beaconsfield community (Figure 16) have most confidence in the CFA (1.715) and far less confidence in the Government’s response to bushfire issues (3.285). A similar result is found for Diamond Creek (Figure 17). However, the CFA score a better average rating (1.56) compared with Upper Beaconsfield. Meanwhile, the local community average rating is worse (2.32 compared to 2.01).

Non-hazard benefits

There are two responses that stand out in Figure 18 as the most popular reasons for living in such an area. The first of which is the environmental benefit (34%) and the second reason is that living in Upper Beaconsfield means one is away from the risks posed by living in city or more suburban surroundings (29%). It is a little surprising that only 11% of people chose ‘close to friends and family’ as the major benefit.

5.4. Spatial Patterns

For the purpose of analysing the results at different spatial scales Upper Beaconsfield was broken down into three regions and 9 sub-regions given in Table 7. Firstly, the regions are looked at in Figures 19 and 20 and Tables 7 and 8 and secondly the sub-regions in Figures 21 and 22. The region of Diamond Creek that data was collected for is also featured in the figures that follow.

Table 7: Regions and sub-regions for analysis

Upper Beaconsfield
North West Region	Southern Region	North East Region
Sub-region 1 Harpfield Road St Georges Road	Sub-region 3 Abeckett Road Armstrong Road Buchannan Road Quamby Road Telegraph Road	Sub-region 6 Burton Road Morris Road Young Street
Sub-region 2 Brennan Avenue Fraser Avenue Knapton Avenue	Sub-region 4 Corringham Road Fairhazel Court Leppitt Road Salisbury Road	Sub-region 7 Beaconsfield – Emerald Road McArthur Road Stoney Creek Road
	Sub-region 5 Funnell Road Surgarloaf Road Tower Road	Sub-region 8 Paul Grove Rosebank Lane
		Sub-region 9 Grant Court Lenne Road Sutherland Road

Regions

Interesting spatial patterns emerge by looking at the data at regional levels. As is clear from Figure 19, the north west region of Upper Beaconsfield appears to have the highest levels of awareness and perception of the bushfire risk. 58% of the respondents in this region considered the hazard rating as very high compared to 28% and 39% respectively in the north east and southern sections. Confidence with bushfire strategies appears to be universally high in all parts of Upper Beaconsfield. This is not as true for Diamond Creek, which shows a larger percentage of unconfident residents (26%). Perception of bushfire knowledge is also marked by a higher percentage (76%) rating their knowledge as good or better in the north west region compared with 54% in the north east region and 63% in the southern region.

It is interesting to note from Table 8 that the highest percentage of people with a ‘wait and see’ strategy (20%) are in the north east region. There is also a greater willingness to stay and defend homes amongst residents in the other two regions. The results for the strategies adopted by Ash Wednesday residents in 1983 all appear to be fairly similar.

Table 8: A comparison of the strategies adopted by residents in different regions of Upper Beaconsfield at the time of Ash Wednesday compared with today

Residents	Strategy	NW	NE	S
Today	Wait and see	15.79%	20.00%	14.29%
	Evacuation	28.07%	32.50%	31.75%
	Stay at home	24.56%	27.50%	30.16%
	Partial evacuation	35.09%	20.00%	23.81%
Ash Wednesday	Stayed at home	42.86%	40.00%	37.93%
	Evacuated	57.14%	60.00%	62.07%

The spatial distribution of the residents of Upper Beaconsfield that were living here at the time Ash Wednesday shows that almost half of them lived in the southern region (Table 9).

Table 8: Where do those who were present at the time of Ash Wednesday live?

	NW	NE	S
Percentage	22.22%	31.75%	46.03%

The responses to Figure 20 highlight the spatial differences very clearly. The pattern is similar to that found in the other regional comparisons thus far. Residents in the north west region of Upper Beaconsfield are much less likely to expect a CFA tanker and crew to protect their home (26% ‘yes’) than residents in the north east region (44% ‘yes’). Meanwhile the southern region falls between the two (38% ‘yes’) and the figure for Diamond Creek eclipses all of them (58% ‘yes’).

Sub-Regions

Breaking the regions down still further into individual clusters of streets enables micro-scale differences to be seen. Key regions to emerge are Rosebank Lane and Paul Grove (sub-region 8), which come out with the lowest perceived knowledge of the fire risk (Figure 21) and are least confident in the bushfire strategy they have in place (Figure 22). At the other end of the scale, sub-region 2, consisting of Brennan, Fraser and Knapton Avenue, has the highest level of perceived bushfire knowledge (83%). The highest level of confidence was to be found in sub-region 4, which includes Leppitt Road and Salisbury Road.

6. Discussion

6.1. Initial Findings

The devastating Ash Wednesday fires highlight that Upper Beaconsfield is situated in an extremely fire-prone area. This was reflected in the results by the 39.5% of respondents who recognised the hazard rating as very high (Table 3). A further 39% believed the hazard rating to be high, which overall shows almost four out five people appreciate the bushfire risk. The percentage of residents who selected ‘very high’ falls between two other similar surveys undertaken in close geographical proximity to Upper Beaconsfield. In North Warrandyte 52% chose this classification (Beringer, 2000). Meanwhile, only 36% of people in a survey of the Dandenong Ranges gave the maximum hazard rating (Raseta et al., 1987, cited in Beringer 2000). All three studies fall within areas that are considered as extremely fire prone by the CFA. The 21.5% who considered the hazard rating as moderate or less in Upper Beaconsfield is an interesting finding. This is considerably higher than the figure of 6% that was found in the North Warrandyte survey.

Thee most likely explanation for the differences in findings of these two surveys is fire experience. The North Warrandyte survey was undertaken only two years after the 1991 Warrandyte fire. Although this was not on the scale of Ash Wednesday it meant that the events were still fresh in people’s minds. In contrast it has been twenty years since the last major fire in Upper Beaconsfield. A larger number (73%) of the respondents in North Warrandyte had been present in the 1991 fires, whilst only 31.7% of respondents in this study had experienced the fires in 1983. This could well have led to a larger number of residents underestimating the fire hazard. Another factor that may have been of significance is the time of year that this research was undertaken. The North Warrandyte survey took place in March, at the end of the fire season. This survey was carried out in August, during the winter time. As a result people’s general awareness levels could have been quite different. This was a factor encountered by Edgell (1973) and thus is important to recognise. However, if this is the case it highlights the need to reinforce awareness throughout the year so people do not underestimate the risk.

The Diamond Creek results reveal far less awareness from the residents of the hazard rating than in Upper Beaconsfield. Only 20% of the respondents considered it to be very high and 38% thought it was moderate or less. The last major fire to affect this suburb was in 1962 long before most of the current residents would have been living there. Despite a longer-term history of fire events in the vicinity, the relatively long period without such an event appears to have resulted in a degree of complacency amongst residents. The issues surrounding previous fire experience are explored in more depth further in section 6.2.

6.2. Risk Awareness and Perception

Age

It is generally accepted that there is a decreasing perception of the fire hazard with age. The explanations for this are that older people may consider themselves less vulnerable due to their increased experience (Beringer, 2000: 4). An alternative prognosis is that elderly residents are often more sceptical about the danger and thus more likely to discount the risk (Hodge et al., 1979 cited in Reinholtd et al., 1999). To a certain extent the results for Upper Beaconsfield do follow this trend (χ² = 11.5, p = 0.2). Whilst this figure falls outside the probability boundaries set for this research there is still a reasonable level of certainty that it is of significance. However, if this is the case then one might expect the youngest age group (18-25) to perceive the threat as greatest. The Upper Beaconsfield findings contradict this, as 38% of this age group perceive the hazard as moderate. This may highlight a larger degree of naivety amongst younger respondents who are less likely to have had first-hand fire experience. It might also reflect the need for more bushfire education at a younger age. This was certainly the view of one of the interviewees who believed “not enough is done in schools to educate children about bushfires” (Interview 4). However, they make up only 4% of the total respondents to the survey so a more in-depth examination of this age group would be needed to ascertain the full extent of their risk perception and awareness.

The results in Upper Beaconsfield also seem to reveal the converse side to age and that is the benefit of experience. The highest proportion of respondents perceiving the threat as very high were in the 71+ age group. This could derive from a life spent in fire-prone areas and the greater chance that older people have experienced fires. On the other hand, the Diamond Creek results conform more to the pattern found in North Warrandyte. The perception of the fire risk is highest by the younger age groups and significantly lower with age. It is interesting to note that in both Diamond Creek and Upper Beaconsfield the proportion of each age group perceiving the risk as very high is similar (around 20% in the former and 40% in the later). This indicates a core group at the heart of each area who are very aware of the fire prone nature of their environment. These are most likely to correspond to those who have appropriate strategies in place and who involve themselves in community fire protection. They also represent the most community-minded residents.

Length of Residency

There appears to be a clear relationship between length of residence and perceived bushfire knowledge in Upper Beaconsfield (χ² = 43.5, p = 0.001). Length of residency does not necessarily translate into actual experience. However, in the case of Upper Beaconsfield it does as those who have been there for more than twenty years experienced Ash Wednesday. This is very well reflected in Figure 11, with a much larger proportion of respondents perceiving their knowledge as very good in this category. It also appears to be the case that those who have moved into Upper Beaconsfield since the fires but have been there for a number of years perceive their knowledge as better than more recent arrivals. A certain amount of knowledge can be gained over the years from fire scares as it increases to reality of the danger for residents. However, very few people in the categories other than 21+ years perceived their knowledge as very good. This is a strong indication that the severity of the Ash Wednesday has imprinted on their memories the reality of a severe fire. It was very evident from those interviewed that unless one had actually experienced a fire it was very hard to be fully prepared (Figure 23).

Figure 23: Recollections of Ash Wednesday

-“You could here it coming; you couldn’t see anything until it was on top of you like a fast running train.”(Interview 4)

-“That was the most devastating fire I’ve ever seen.” (Interview 5)

-“Disasters of that magnitude – I don’t know if anybody could be prepared. The only preparation you can have for a disaster like that is evacuation.” (Interview 6)

Gender

There are marginally more women in Upper Beaconsfield perceiving their knowledge as good or better compared to men. The reverse is true of the results for Diamond Creek. However, an interesting result was found for the Diamond Creek survey. The residents were asked a question designed to test their actual knowledge of fire behaviour concerning when fire travels fastest (Table 4). This revealed that 52% of women knew the answer compared to 74% of men. This may reflect a tendency amongst some to rely on the knowledge of their partners, with several people who were spoken to saying that this was the case. It is clearly vital for all family members to be equally aware of the fire risk and fire behaviour. However, no significant statistical relationship could be determined.

Home Builders

A significant relationship between those who had built their own homes in Upper Beaconsfield and their levels of bushfire knowledge materialised (χ² = 19.46, p = 0.0025). When constructing homes in bushfire prone areas there are building requirements that have to be adhered to. Consequently the process of building a home can draw people’s attention to the issues of the fire hazard and increase people’s awareness. It is also likely that people who invest their time and money into building their own home have a greater level of interest in ensuring they are protected in the event of a bushfire. This has been taken to an extreme by one couple that were interviewed in this research. They had lost their home in the Ash Wednesday fires and had decided to rebuild on the same site. They decided to design an earth-covered house that is devoid of flammable materials (Figures 24 and 25). The result is a house very well suited to the environment they live in. Clearly they learnt a great deal from the process of constructing such a fire-resistant building, which they feel confident to stay in when another fire comes through.

However, there are also new people who have moved into the area in recent years and built their own homes in the new estate areas such as Rosebank Lane. The perceived bushfire knowledge is considerably lower in this area than many other areas of Upper Beaconsfield. This is discussed further in 6.4.

Figure 24: An earth-covered home in Upper Beaconsfield

constructed of fire retardant materials

Figure 25: Fire shutters to prevent burning embers coming through the windows

Ash Wednesday

The significance of the Ash Wednesday experience has already been mentioned in relation to the other variables. Upper Beaconsfield residents who experienced Ash Wednesday (31.7% of respondents) are more likely to have a greater bushfire understanding than those who have not (χ² = 31.8, p = 0.001). In fact, they are more than 2.5 times more likely to perceive their knowledge of fires as good or very good. In addition, a far higher percentage rate their knowledge in the highest category reflecting the awareness of the fire risk and its behaviour that only comes from actually experience.

A good example of this is the response to the question ‘would you expect a CFA tanker and crew to protect your home in the event of a bushfire’. People in Upper Beaconsfield who were not resident at the time of the fires were more likely to answer ‘yes’ to this question (40%). However, those who have experienced the fires are more likely to understand that the CFA cannot be relied on to save a home during a major fire (70% of Ash Wednesday residents). The response to this question is even more striking when one examines the figure for Diamond Creek. 58% of the respondents replied ‘yes’ which is more in line with the North Warrandyte figure of 53% (Beringer, 2000). Instead it is essential to have a realistic strategy in place and put it into action in good time. Overall the response to this question from Upper Beaconsfield respondents was encouraging especially compared to past surveys and the results from Diamond Creek. Consequently the Ash Wednesday experience clearly does play a significant role in bushfire awareness and perception in the Upper Beaconsfield community.

6.3. Behavioural Response to the Bushfire Risk

Survival Strategies and Planning

The benefit of people, who are suitably prepared, staying and defending their homes during a bushfire has been widely recognised (Lazarus and Elley, 1984; Boura, 1998). The benefits of such a strategy are that more homes could be saved and consequently some loss of livelihood could be thwarted and lives saved. In addition last minute evacuations are extremely risky and were a contributing factor to several of the lives lost during Ash Wednesday (Krusel and Petris, 1992). This was recognised by a considerable number of the respondents in Upper Beaconsfield (52%). As one interviewee put it: “policy now has moved a lot more towards staying and defending your house” (Interview 4). However, for some a far more sensible strategy is evacuating provided it is in good time. One interviewee said, “The only preparation you can have for a disaster like that is evacuation”. Clearly this is still a valid option, especially for those who are less experienced in fire situations.

87% of respondents in Upper Beaconsfield said they were confident in the strategy they had in place. This is clearly important as decisive action and thinking can often be crucial in hazardous situations. However, it does not entirely reflect the responses found in Table 6, which compares people’s strategies at the time of Ash Wednesday with the strategies they have in place today. A significant percentage (17%) do not appear to have a strategy in place as they have decided to wait and see. This could be a cause for concern as it is risky not to a have well thought out plan. On the other hand, it may reflect the unpredictable nature of the bushfire threat that makes it hard to stick to a rigid plan. Certainly the experience of Ash Wednesday highlighted that plans can become hard to put into action. For instance, roadblocks prevented some people from getting back to their homes to protect them (Lazarus and Elley, 1984: 20).

An important aspect of the basic prevention in ensuring the safety of properties during a bushfire is regular house and garden maintenance. A majority of respondents in Upper Beaconsfield were doing this at least once a month, in line with the CFA guidelines. A minority were only carrying out maintenance once during the summer, which is dangerous strategy. The consequence can be a large build up of ground fuel, which can severely jeopardise an attempt to save a house during a serious fire. This suggests that while in general the message is getting across, there is still a need to encourage more people to take the risks seriously. However, as one interviewee put it “some people are never going to be all that interested in it; they just think it will never happen” (Interview 5). Thus, there is a certain level of apathy towards these issues that is faced when trying to encourage a community to become more self-reliant. This corresponds with the third strand of the Behavioural Response Model (Figure 2).

Figure 26: Long grass and overhanging trees – regular summer maintenance is essential to reduce the build up of fuel

Confidence in the Management of Bushfires

The results in Figures 16 and 17 make it immediately clear that there is more confidence in the emergency response aspect of organisation than the policies implemented at governmental levels. This is interesting because “in practice, the acceptance of risk by the public depends crucially on the degree of confidence placed in the organisation charged with its management” (Smith, 2001: 74). The low level of confidence in Federal Government probably reflects the lack of a holistic approach with regard to bushfires. The CFA achieved the best average from the findings, which is most likely a reflection of the efforts that have been made to communicate with the local community through such schemes as Community Fireguard. There generally seems to be the feeling that enough information is distributed in Upper Beaconsfield with only 16.5% of respondents saying they would like to receive more. This contrasts dramatically with the North Warrandyte findings that found 71% of people wanted more information (Beringer, 2000). It could be the case that information in Upper Beaconsfield is more readily available. For instance, at the Ash Wednesday Memorial Park in the centre of the village there is bushfire awareness information (Figure 27 and 28).

Figure 27: The Bushfire Awareness information

in the Ash Wednesday Memorial Park

Figure 28: Practical advice helps residents plan their bushfire strategy

Non-Hazard Benefits

The vast majority of people make a conscious decision to live in areas such as Upper Beaconsfield despite the risk that is posed by bushfires. 96.5% of respondents agreed that the benefits of living in Upper Beaconsfield outweigh the risks. The 34% of respondents who believed environmental benefits were the major non-hazard benefit emphasise the aesthetic appeal of the rural surroundings (Figure 18). It is evident that many people have lived in relatively rural communities for much of their life and that is their preference to continue to do so. A cross-section of views is given in Figure 29.

Figure 29: the bushfire risk in perspective

– “The possibility of a bushfire is part of the price. But in a place like here where the scrub is not dense and is rarely that dry even the risk itself is no that great if you carry out some basic preparations.” (Interview 3)

-“I grew up on a farm, fires were part of the reality of every summer. They were very rarely serious. That (Ash Wednesday) was a rare event. I see no reason to move.” (Interview 4)

-“I’ve lived in other communities where one is constantly aware of the bushfires and having to fight fires during the summer months. So it’s something that’s just been apart of where I’ve lived” (Interview 5)

29% cited being away from city risks as the major benefit. This is very interesting because it indicates that for many the perceived day-to-day risks that people get exposed to in a city environment outweigh the potentially more catastrophic, but less prevalent, risk posed by a serious bushfire. Both these factors fall into strand 1 Behavioural Response Model (Figure 2). Less important in this survey was strand 2, as friends and family was mentioned by only 11% of respondents. This reflects the diminishing importance of community with increasing numbers of new arrivals to the area.

6.4. Spatial Patterns

Regions

The perceived increased awareness in the bushfire risk in the north west region of Upper Beaconsfield could well reflect the fact that this area was the worst hit during Ash Wednesday. Critchley Park (Figure 30) is located here, which was the sight where eleven firemen were killed in addition to many homes being lost. Consequently there is a better overall level of awareness amongst the residents. In contrast the north east section was damaged less and has a lower perceived risk awareness. In addition, the emergence of newer developments such as in Rosebank Lane mean that there is probably less of a general awareness of the past events in this part of the suburb. Comparisons with the region of Diamond Creek examined highlight that residents here have underestimated the hazard rating despite their proximity to Plenty Valley (Figure 31). This further backs up the importance of the first hand experience gained by many in Upper Beaconsfield during the Ash Wednesday fires. It also shows that confidence levels are higher in all three regions of Upper Beaconsfield in comparison with a lower figure for Diamond Creek. There may well be a need for more advice and awareness campaigns here to help people formulate strategies or become more confident in existing ones.

Figure 30: Critchley Park in Upper Beaconsfield

Figure 31: Plenty Valley in Diamond Creek

The varying pattern of bushfire awareness in different sections of Upper Beaconsfield is further emphasised by Figure 20. It is the north east section, which again shows the lowest level of awareness with regard to expectations of a CFA tanker and crew protecting ones home during a serious fire (nearly 44% expecting this). This is almost twice the expectancy of residents in the north west region. However, Diamond Creek is far in excess of this figure with 58%.

Sub-regions

It is also possible to break the regions down to street level and examine which particular roads have the highest levels of awareness. It is not surprising to discover that streets that fall in the north east region do have some of the lowest levels of awareness regarding bushfire issues. Rosebank Lane and Paul Grove (sub-region 8) come out with the lowest perceived knowledge of the fire risk (Figure 21) and are least confident in the bushfire strategy they have in place (Figure 22). There is a perception amongst some residents that the new arrivals are not as concerned or aware of the bushfire risk. One interviewee said, “because there’s more houses going up now, they start to think a bushfire can’t come here” (Interview 6). Areas such as Rosebank were also the hardest to get responses from, indicating a general lack of interest. Therefore, this is clearly an area that needs to be targeted to increase awareness and self-reliance with regard to bushfires.

Figure 32: Rosebank Lane: An area for concern?

Figure 33: New developments under construction in Upper Beaconsfield

7. Conclusions

7.1. The Research

Residents in Upper Beaconsfield and Diamond Creek live in the most fire-prone region of the world. The natural consequence of this is that from time to time they will be subjected to an extreme hazard. The history of devastating fires in Victoria is testament to this. It is therefore essential that people are aware and prepared for such an eventuality when, not if, it arises. In its broadest sense the results of this research demonstrate that a reasonable number of people are aware of the nature of the risk they are exposed to. How this translates into developing bushfire strategies is dependent on the perception of the fire hazard at several levels: individual, household and community. However, the results also identify groups and areas where improvement appears to be needed. Distinctions can be drawn between the two suburbs. Residents in Upper Beaconsfield on the whole are more aware of the risk than those living in Diamond Creek. The experience of Ash Wednesday is the single most important factor in this. However, even within the confines of the Upper Beaconsfield community spatial differences do emerge with the more established and community orientated streets and regions far more aware than the newer areas. As more subdivisions of the land take place the nature of Upper Beaconsfield is changing. It is therefore essential to devise ways of encouraging a seemingly more risk averse population to adopt more appropriate levels of understanding.

Risk perception of the bushfire threat is made up of a complicated set of interrelated factors. This research has highlighted the importance of place specific issues most notably Ash Wednesday. The effect of the 1983 fires, linked closely to other variables such as length of residency, has had a dramatic impact on the Upper Beaconsfield community. In contrast to Diamond Creek where the last major fire was over forty years ago. With this in mind it appear paramount efforts are made to educate the population of the dangers as soon as possible.

The results also highlight a considerable level of confidence in the CFA and the local community in contrast to a fairly widespread mistrust in the Government handling of the fire issues. Perhaps this is symptomatic of a more general public discontent with politics. However, it shall be interesting to see how changes implemented at a national level, as a result of the recent fire review, impact on the perception of the handling of these issues. A range of factors were identified that showed the reasons why people wanted to live in Upper Beaconsfield and Diamond Creek. The environmental benefits of the surroundings were a major factor. In addition being away from city-associated risks was also another important reason. This emphasises the extreme rarity of extreme fire events and the fact that the fire risk is fairly low as far as most people are concerned in comparison to some of the other risks one faces in life. Preparedness is vital to bushfire survival. Adopting a well planned bushfire strategy and keeping properties and the surroundings well maintained are two vital elements. However, the strategy also needs to be realistic as there is no point in planning to stay and defend a property without the necessary fire-fighting equipment such as a pump.

The changes brought about in the management of bushfires post Ash Wednesday proved to be a dramatic turning point in community awareness of the fire hazard. Lessons were learnt and a whole range of policies adopted to help minimise the effects of future extreme fires. Likewise, the upshot of the 2003 fire season has been similar, as it has brought home the realities of the fire risk to rural-urban fringe communities and those charged with its management. However, there is an important difference between passively being more aware and proactively enhancing bushfire preparedness. In Upper Beaconsfield and Diamond Creek this involves greater community action to educate residents and to mitigate the dangers. Schemes such as Community Fireguard appear to work well amongst the core group of residents who take the issues seriously. It is finding ways to encourage more people to get involved in such activities that appear to be more allusive. This is clearly the most pressing challenge for organisations such as the CFA. Of course, empowering communities to be ready and safe in the eventuality of a major fire is just one part of the bushfire management process. It is however, a very significant one because if it is achieved successfully, it means the CFA can concentrate on many of the other areas demanding their attention. In doing so, the chances of minimising the devastation of future extreme fires are increased considerably.

7.2. Suggestion for Future Work

The inclusion of the smaller study in Diamond Creek has provided an extremely useful comparison. Levels of awareness are considerably lower than those found in Upper Beaconsfield even comparing unfavourably with the most apathetic areas. In light of the findings in this research there is an urgent need for a more thorough investigation. The scope should be as wide as possible, not just in Diamond Creek but across the Shire of Nillumbik more generally. If the small sample of the population explored in this research proves to be representative of the town then there are clear areas that attention needs to be focused on. Bushfire education is a part of this as well as simply making people aware of the reality of living on the rural-urban fringe.

There is also the need in both Upper Beaconsfield and Diamond Creek to find ways to encourage the community to become more involved in fire preparedness. A series of focus group discussions with a wide range of people in the two suburbs could be of great benefit in highlighting how the community can be empowered and what issues put individuals off getting involved. The obvious challenge here would be finding a way to get a wide range of people to participate and not just those who are already involved. Nonetheless, this study has successfully examined the issues surrounding perception and awareness of the bushfire risk. In doing so it provides an up-to-date examination of current levels of bushfire preparedness in two fire-prone communities.

8.1. Appendix 1

Pakenham-Berwick Gazette Article, September 15^th 2003

8.2. Appendix 2 – Sample Questionnaire

Section A: Basic Details									
1. Sex:				Female
				Male
2. Age:				18-25
				26-40
				41-55
				56-70
				71 +
3. How long have you lived in Upper Beaconsfield ?				Less than 1 year
				1-5 years
				6-10 years
				11-15 years
				16-20 years
				21 + years
4. Do you own your own property?				Yes – GO TO QU. 5
				No – GO TO QU.7
5. Did you build your own home?				Yes – GO TO QU. 6
				No – GO TO QU. 7
6. To what extent did you consider the fire hazard when constructing and designing you home?				Sought advice
				Didn’t seek advice, but took it into consideration
				Not really
				Not at all
Section B: Risk Awareness
7. What would you consider the hazard rating of the area to be?				Very High
				High
				Moderate
				Low
				Very Low
8. What type of bushfire strategy do you have in place?				Wait and see
				Evacuation
				Stay at home
				Partial evacuation (some family members stay, others go)
9. Do you feel confident with the strategy you have in place?				Yes
				No
10. How would you rate your knowledge of bushfires?				Very Good
				Good
				Moderate
				Poor
				Very Poor
11. In the event of a bushfire would you expect a CFA tanker and crew to protect your home?				Yes
				No
12. Are you happy with the level of information about bushfires you receive?				Yes
				No
13. How often during the summer do you carry out house and garden maintenance on your property, for example, ground litter removal and guttering clearance?				Never
				Once during the summer
				Once a month
				Once a fortnight
				Once a week
Section C: Risk perception								
14.   The benefits of living in Upper Beaconsfield outweigh the risk from bushfires?					Agree – GO TO QU. 15
					Disagree – GO TO QU. 16
15. What is the major benefit of living in Upper Beaconsfield ?					Environmental benefits
					Cost of living compared to the city
					Close to friends and family
					Away from city risks (pollution, crime etc.)
					Convenience for work
					Other: ……………………………….
16. Were you a resident in Upper Beaconsfield at the time of the Ash Wednesday fires?					Yes – GO TO QU. 17
					No – GO TO QU.   19
17. Did you stay at home or evacuate?					Stayed at home
					Evacuated
18. How much better prepared do you feel today than during the Ash Wednesday fires?					Much better prepared
					Better prepared
					The same
					Less well prepared
19. How much confidence do you have in the following interest groups with regard to the handling of bushfires ? Circle as appropriate. 1 = Total confidence, 2 = Confidence up to a point, 3 = Concerns, 4 = Little confidence, 5 = No confidence
The Federal Government	1	2	3			4	5
The State Government	1	2	3			4	5
The CFA	1	2	3			4	5
The Police	1	2	3			4	5
The Upper Beaconsfield community	1	2	3			4	5
20. It would be of great help to be able to speak to a small group of residents in more detail regarding these issues, would you be willing to be interviewed on an individual basis?					Yes – please fill out the form below
					No

PLEASE ONLY FILL IN THIS SECTION IF YOU HAVE SAID YES IN QUESTION 20.

NAME…………………………………………………………………………………

ADDRESS..……………………………………………………………………………

CONTACTNUMBER………………………………………………………………….

Please note: the information you have supplied will remain totally confidential.

I will be collecting your form onThursday 14^th August  Friday 15^th August 

Once completed please put the questionnaire in the envelope provided and leave it in either your mailbox or on your doorstep on the morning of collection.

Question	Upper Beaconsfield		Diamond Creek
1. Sex:
Female	109	54.8%	27	54%
Male	91	45.2%	23	46%
2. Age:
18-25	8	4%	5	10%
26-40	42	21%	15	30%
41-55	96	48%	21	42%
56-70	41	20.5%	7	14%
71 +	13	6.5%	2	4%
3. How long have you lived in Upper Beaconsfield / Diamond Creek?
Less than 1 year	15	7.5%	3	6%
1-5 years	49	24.5%	7	14%
6-10 years	28	14%	6	12%
11-15 years	25	12.5%	6	12%
16-20 years	26	13%	7	14%
21 + years	57	28.5%	21	42%
4. Do you own your own property?
Yes	185	92.5%	43	86%
No	15	7.5%	7	14%
5. Did you build your own home?
Yes	58	30.9%	14	32.6%
No	130	69.1%	29	67.4%
6. To what extent did you consider the fire hazard when constructing and designing you home?
Sought advice	11	17.5%	1	7.1%
Didn’t seek advice, but took it into consideration	34	54%	4	28.6%
Not really	11	17.5%	7	50%
Not at all	7	11.1%	2	14.3%
7. What would you consider the hazard rating of the area to be?
Very High	79	39.5%	10	20%
High	78	39%	21	42%
Moderate	42	21%	14	28%
Low	0	0%	4	8%
Very Low	1	0.5%	1	2%
8. What type of bushfire strategy do you have in place?
Wait and see	34	17%	13	26%
Evacuation	62	31%	19	38%
Stay at home	53	26.5%	8	16%
Partial evacuation (some family members stay, others go)	51	25.5%	10	20%
9. Do you feel confident with the strategy you have in place?
Yes	174	87%	37	74%
No	26	13%	2	4%
Unsure	N/A	N/A	11	22%
10. How would you rate your knowledge of bushfires?
Very Good	35	17.5%	13	26%
Good	91	45.5%	11	22%
Moderate	68	34%	20	40%
Poor	6	3%	5	10%
Very Poor	0	0%	1	2%
11. In the event of a bushfire would you expect a CFA tanker and crew to protect your home?
Yes	73	36.5%	29	58%
No	127	63.5%	21	42%
12. Are you happy with the level of information about bushfires you receive?
Yes	167	83.5%	37	74%
No	33	16.5%	13	26%

8.3. Appendix 3 – Results Summary

Question	Upper Beaconsfield		Diamond Creek
13. How often during the summer do you carry out house and garden maintenance on your property, for example, ground litter removal and guttering clearance?
Never	0	0%	1	2%
Once during the summer	49	24.5%	10	20%
Once a month	71	35.5%	20	40%
Once a fortnight	47	23.5%	10	20%
Once a week	38	16.5%	9	18%
14. The benefits of living in Upper Beaconsfield / Diamond Creek outweigh the risk from bushfires?
Agree	193	96.5%	50	100%
Disagree	7	3.5%	0	0%
15. What is the major benefit of living in Upper Beaconsfield / Diamond Creek?
Environmental benefits	67	34.9%	18	36%
Cost of living compared to the city	7	3.6%	3	6%
Close to friends and family	21	10.9%	9	18%
Away from city risks (pollution, crime etc.)	57	29.7%	11	22%
Convenience for work	7	3.6%	3	6%
Other	33	17.2%	6	12%
16. Were you a resident in Upper Beaconsfield at the time of the Ash Wednesday fires?
Yes	63	31.7%	N/A	N/A
No	136	68.3%	N/A	N/A
17. Did you stay at home or evacuate?
Stayed at home	25	40.3%	N/A	N/A
Evacuated	37	59.7%	N/A	N/A
18. How much better prepared do you feel today than during the Ash Wednesday fires?
Much better prepared	21	35%	N/A	N/A
Better prepared	22	36.7%	N/A	N/A
The same	17	28.3%	N/A	N/A
Less well prepared	0	0%	N/A	N/A
19. When does fire travel fastest?
When it burns upslope	N/A	N/A	31	62%
When it burns on level ground	N/A	N/A	10	20%
When it burns downslope	N/A	N/A	9	18%
20. How much confidence do you have in the following interest groups with regard to the handling of bushfires?
	Average score (1 = very confident,   5 = not confident)
The Federal Government	3.285		2.92
The State Government	3.15		2.98
The CFA	1.715		1.56
The Police	2.38		2.08
The Community	2.01		2.32

8.4. Appendix 4 – Interview Schedule

This is the general structure and content of the questions asked:

RISK AWARENESS AND PERCEPTION

1 When did you moved to Upper Beaconsfield? Were you living in a fire prone area before then?

2 Did you receive any information on your arrival?

3 Was the fire hazard something that you considered before moving here?

4 So what was the motivation behind designing your home in this way?

5 In general do you feel that the community around you does enough to be prepared and aware?

6 Have you attended any bushfire meetings such as bushfire blitz?

o How successful? What did it cover? Was there anything that should have been talked about that wasn’t?

7 Is community fireguard active in your area?

o Is it a success?

8 Is enough being done to get the message across to the community?

9 How do think the media deals with the bushfire issues?

10 Tell me about your (family) routine on a total fire ban day.

11 Tell me about how much of a risk you see bushfires being in this area:

o To you and your home? To your road? To your village?

12 How do you see the risk being best managed in the area? (Is enough being done, e.g. fuel reduction)

13 Are you confident you will receive enough warning in order to make a reasoned decision?

14 In regards to you home / street, where would you see the major threat from a bushfire coming?

THE ASH WEDNESDAY EXPERIENCE

15 What are the major changes in your awareness since Ash Wednesday?

16 Thinking back are there things that you would do differently if you had the chance again?

8.5. Appendix 5 – Interview Summaries

Interview 1

Awareness of bushfires:

1 “There’s always publicity going out but there’s still people who will say that doesn’t interest me and chuck it out without reading it”

2 “Its too quick to get away”

3 “I bet you there will be (another fire). Our block, in terms of plants is just about ready for another fire”

Awareness of community schemes:

1 “There’s one that covers done to, across White Lane and all the way along. Along Leppitt Road and Carpenter Road to the top of the hill there. So everybody who’s been in that group is pretty aware.”

2 “There were 30 names on the list”

3 “After the groups got going they (CFA) encouraged us to run them ourselves”

4 “What Community Fireguard tries to get across is those experiences that you can share”

Perception of community:

1 “They haven’t come to the meetings, but they understand the risks.”

Survival strategy:

1 “You’ve got to learn to live with it … making your decisions and making your preparations knowing that on a decent fire day there’s no way anybody’s going to stop the fire”

2 “Its just working around it and if you don’t want to face up to bushfires you go and live in the city”

3 “Its really a case of being prepared for all sorts of eventualities, you can’t just have one fixed strategy in place”

Ash Wednesday:

1 “There wasn’t the education”

Changes in awareness:

2 “I think we would probably go earlier, knowing what we know now”

CFA:

3 “The fire brigade works their fingers to the bone.”

4 “The CFA nowadays set themselves up so they’re prepared for the fire in three of four directions … its what’s known as a strike force”

Interview 2

Awareness of bushfires:

1 “Its only something you think about in summer”

2 “I’ve grown up with the risk, so I guess it’s a part of life for me”

Awareness of community schemes:

1 “The CFA give a yearly talk to the UBA (Upper Beaconsfield Association). They discuss fire meeting points and that sort of thing.”

2 “There was Fireguard scheme a while ago but its fizzled out”

Perception of community:

1 “ A lot of people don’t do enough … but there are those who do discuss the issues”

Survival strategy:

2 “ We don’t have much choice but to evacuate … there are too many trees and access isn’t great”

3 “”No matter what you do to prepare, if the person next door doesn’t then it makes no difference.”

Interview 3

Awareness of community schemes:

1 “Bushfire Blitz is a one and half hour talk that covers the basics about fire behaviour”

2 “Its down to the willingness of individual streets for Community Fireguard to take off”

3 “Individuals are left to run them but the CFA do provide support”

Perception of community:

1 “You have to remember that people lead busy lives, single parents, or both parents working … bushfires even in high risk areas are unlikely. So people prioritise elsewhere. For instance is it better to have a fire pump or spend the money on school fees?

2 “Its hard to get people to prioritise fires”

Interview 4

Awareness of bushfires:

1 “You don’t have to be a genius to fight fire as long as you are with people who know what they’re doing”

2 “I grew up on a farm, fires were part of the reality of every summer. They were very rarely serious. That (AW) was a rare event. I see no reason to move.”

3 “The possibility of a bushfire is part of the price. But in a place like here where the scrub is not dense and is rarely that dry even the risk itself is no that great if you carry out some basic preparations.”

Awareness of community schemes:

1 Referring to Community Fireguard and Bushfire Blitz: “I have to say I had not heard of either them until you mentioned them”

2 “Involvement not in a wide community sense. Among the neighbours here”

Perception of community:

1 “I hadn’t expected my community to respond in such a disorganised way”

Survival strategy:

2 “In the next fire I will defend this place until it is safe and then deploy myself somewhere else.”

Ash Wednesday:

3 “I was ordered to evacuate twice by a policeman who had no idea where I was to evacuate to and on one occasion by a policeman who had no idea where Upper Beaconsfield was”

4 “In military terms, they didn’t use the available intelligence. They knew the big winds were coming, they knew that gully was deep, and they should have known it would be very difficult for a crew who didn’t know the road”

5 “I said to a fireman where do you want the volunteers … He just said I wish all you volunteers would get out of the place. I’m not accusing him of anything other than incompetence and only of the incompetence you would expect from almost anyone and that’s an important distinction”

6 “You could here it coming; you couldn’t see anything until it was on top of you like a fast running train.”

7 “There were all sorts of people who took advantage of welfare availability. Other people who carried out various insurance frauds.”

8 “The chain of command up here got very confused … We just got attacked in so many places at once, the system had to collapse.”

Concerns:

1 “Think it’s (CFA) a politics ridden body. I still think that in fairness to them a fire of that magnitude is never going to be well handled by small groups of volunteers”

Changes in awareness:

2 “Policy now has moved a lot more towards staying and defending your house”

Areas for improvement:

3 “If more people really had a fire fighting pump and a source of water, um, (a) we could stop some fires before they got big and (b) we could defend properties far more readily.”

4 “We don’t make enough use of our military forces.”

Interview 5

Perception of community:

1 “There would only be a couple of people I know who would be geared up for it”

2 “The majority as far as I can see rely on the CFA”

3 “The rest of the people I know up and down the road wouldn’t have a clue. My perception is that they wouldn’t have much of an idea, I might be wrong”

4 “Its all very well to be community orientated but when the bushfires came through here the people over the road were very helpful”

5 “Some people are never going to be all that interested in it; they just think it will never happen”

6 “The other neighbours who are new to the area, I spoke to them and asked what they were doing and they said they hadn’t thought about it”

Survival strategy:

1 “Its every man for himself”

Ash Wednesday:

2 “That was the most devastating fire I’ve ever seen”

3 “You wouldn’t expect a devastating fire like that to go through more than once in a lifetime”

4 “No matter how much planning you do its just, you wouldn’t have stopped it with anything”

5 “The bush comes back, but the dwellings are what people lose“

Concerns:

1 “Probably my only concern would be, the water supply”

2 “With that in mind its up to the individual to have their own supply of water”

3 “Something I had never experienced before, and that was the last fire I was involved with, and that was a fireball came through … when that had past, the rest of the fire was something you could do something about”

4 “You could here the roar three or four kilometres away”

Benefits of living in Upper Beaconsfield:

1 “Works good, it’s a good spot to live, nice people live around the area.”

Interview 6

Awareness of bushfires:

2 “People have just got to be aware … they sit here and watch the TV coverage of the Canberra fires and say ah gees that’s terrible, but it’ll never happen here”.

3 “You have to be prepared and you know, if it’s going to be too much for you, you’ve got to evacuate”

Awareness of community schemes:

1 “Look I don’t know if there’s been any up here. I can’t recall of any meetings up here”

Perception of community:

2 “I don’t think they are prepared or aware”

3 “I’m disappointed in a community that’s gone through a tragedy”

4 “I think its just people having the attitude it will never happen again, it will happen”

5 “I just think as a community, I don’t think they are prepared any more than they were in 1983.”

6 “They just say are look we’ll just drive off. But it’s probably not as easy as that ‘cause you have traffic management”

Survival strategy:

1 “There’ll always be people who won’t go. That’s a decision you have to make”

Ash Wednesday:

2 “Some went, yet they still died. So there’s no real answer”

3 “Disasters of that magnitude – I don’t know if anybody could be prepared. The only preparation you can have for a disaster like that is evacuation.”

Concerns:

1 “Over the last four or five years watching things grow, you know where trees should be cleared or there should be clearance”

2 “Look at all the houses going up now. People may think that because there’s more houses going up, they start to think a bushfire can’t come here”

3 “There’s no clearance down in Critchley Park. It’s basically how it was back in 1983 … people don’t realise that the trees that went through the bushfires and grew back are now more shallow rooted. So in actual fact they have more intensity to fall and stuff like that, so there are other dangers involved.”

Changes in awareness:

1 “Clearance of your property and having an evacuation plan”

2 “There ought to be somebody with local knowledge who can be brought to bear on these issues”

Auto-Critique

The extensive literature on environmental risk perception offered a wide scope for exploring the reasons behind people living in areas prone to extreme natural hazards. I chose to examine bushfires for a number of reasons. Firstly, it was an extremely topical issue at the time. Fires had been causing widespread damage in several parts of Australia between January and March 2003. Therefore, there appeared be a need to understand what the risks were and how they could be managed. Secondly, when I began to explore the possibilities there appeared to a real need to add to the literature on bushfires. Outside of Australia little has been written on the topic despite the fact that areas of Southern Europe and the United States are extremely vulnerable. Even within the literature in Australia the need for more investigations into all aspects of fires, including the social aspects, was evident. The decision to investigate this topic in Upper Beaconsfield was due to my own knowledge of the area. I had contacts in place to help facilitate this undertaking and I was also confident that the research was feasible.

The approach I adopted enabled a large amount of nominal data to be collected in Upper Beaconsfield. Ideally the views of all residents would have been obtained but this was not possible due to the restraints of time and resources. Nonetheless, 200 responses represented a good cross-section of the community and as such added validity to the research. Whilst the same robustness was not achieved from Diamond Creek, the 50 residents surveyed made a worthwhile comparison that added a further dimension to the research. The downside of using a quantitative methodology when examining social perceptions is that it can pigeon-hole responses to a degree. This is why the in-depth interviews with 6 respondents were useful. They enabled the issues to be discussed in greater depth and put the quantitative findings into context. Thus, the two contrasting methodologies complemented each other to further the overall understanding of the issues.

With hindsight there are a few adjustments to the research that could have been made. In the questionnaire for Upper Beaconsfield a couple of questions to test knowledge of fire behaviour would have been useful to determine how educated the respondents were on these matters. This was to some extent rectified for the Diamond Creek survey. In addition, the lower age range (under 25’s) was under-represented in the research sample. Consequently, in a future study it would be worthwhile to focus more on them. They are, after all, the people who will be involved in the formulation of bushfire management in the years to come. Nonetheless, overall this research has served to highlight some important areas for improving people’s bushfire awareness in Upper Beaconsfield, whilst providing a useful comparison with past research in this field. It has also uncovered the need for a more wide-ranging investigation in Diamond Creek to determine the full extent of people’s perceptions of these important issues.

Bibliography

Adams, J. (1995) Risk. London: UCL Press.

Aplin, G. (2002) Australians and their environment: an introduction to environmental studies, Oxford: Oxford University Press.

Baxter, J. (1984) Who Burned Australia? The Ash Wednesday Fires, Kent: New English Library.

Beringer, J. (2000) Community Fire Safety At The Rural/Urban Interface: The Bushfire Risk. Fire Safety Journal, 35, pp. 1-23.

Boura, J. (1998) Community Fireguard: Creating Partnerships With The Community To Minimise The Impact Of Bushfires. Australian Journal of Emergency Management, 13, pp. 59-64.

Brown, J. and Damery, S. (2002) ‘Managing flood risk in the UK: towards an integration of social and technical perspectives’. TIBG, 27, pp. 412-426.

Bryant, E. (1990) Natural hazards. Cambridge: Cambridge University Press, pp. 157-170.

Burton, I., Kates, R. and White, G. (1993) The Environment As Hazard, London: Guildford Press. Chapters 2,4-5.

Byron, I. And Curtis, A. (2001) Landcare in Australia: burned out and browned off. Local Environment, 6.3, 311-326

CFA (2002) Community Bushfire Preparedness: Executive Summary. Victoria: CFA.

Chapman, D. (1999) Natural Hazards. Melbourne: Oxford University Press. Chapters 1-2 and 7-8.

Cheney, N. (1995) Bushfires: An Integral Part Of Australia’s Environment. http.//www.abs.gov.au/Austats/abs@.nsf…75496a5ad1ca2569de00267e48!OpenDocument. 05/02/2003.

Clark, C. (1991) Environmental Hazards. Surry: Nelson

Department of Infrastructure (2000) Suburbs In Time. http://www.doi.vic.gov.au/doi/internet/research.nsf/0/42da1270c1c7a8d9ca256cc600044216?opendocument. 21/01/04.

Edgell, M. (1973) Nature and Perception of the Bushfire Hazard In South Eastern Australia. Victoria: Monash University Publications in Geography.

Flowerdew, R. and Martin, D. (1997) Methods In Human Geography. Harlow: Prentice Hall.

Handmer, J. and Penning-Roswell, E. (eds.) (1990) Hazards And The Communication Of Risk. Aldershot: Gower Technical.

House of Representatives Select Committee On The Recent Australian Bushfires (2003) A Nation Charred. http://www.aph.gov.au/house/committee/bushfires/inquiry/report.htm. 24/11/03.

House of Representatives Standing Committee on Environment and Conservation (1984) Bushfires and the Australian Environment, Canberra: AGPS.

Incoll, R. (2003) An Overview of Fire Hazard and Vegetation Controls in the Shire of Nillumbik (unpublished).

Johnston, R., Gregory, D., Pratt, G. and Watts, M. (eds.) (2000) The Dictionary of Human Geography. Oxford: Blackwell, pp. 229-300.

Keys, C. (1991) ‘Community analysis: Some considerations for disaster preparedness response’. Mount Macedon, Australian Counter Disaster College.

Krusel, N. and Petris, S. (1992) Staying Alive: Lessons Learnt From A Study Of Civilian Deaths In The 1983 Ash Wednesday Bushfires. Fire Management Quarterly, 2, pp. 1-20.

Lazarus, G. and Elley, J. (1984) A Study of the Effect Household Occupancy During the Ash Wednesday Bushfire in Upper Beaconsfield, Victoria, February 1983. Paper No. 3, The National Centre for Rural Fire Research, Victoria: Technical.

Löfstedt, R. And Frewer, L. (eds) (1998) The Earthscan Reader in Risk and Modern Society, London: Earthscan Publications.

Luke, R. and McArthur, A. (1978) Bushfires in Australia, Canberra: AGPS.

McKnight, T.L. (1995) Oceania: The Geography Of Australia, New Zealand, And The Pacific Islands. Sydney: Prentice Hall.

Melways (2003) Australia Map.

http://www.street-directory.com.au/aus_new/index.cgi. 02/11/03.

Miller, S.I., Carter, W. And Stephens, R.G. (1984) Report on the Bushfire Review Committee on Bushfire Preparedness in Victoria, Australia, following the Ash Wednesday Fires 16 February 1983, Melbourne: Government Printer.

Miller et al. (1984) Report On Bushfire Disaster Preparedness, pp. 150-174.

Murray, R. and White, K. (1995) State of Fire: A History of Volunteer Firefighting and the Country Fire Authority in Victoria. Victoria: Hargreen Publishing.

Niall, E., Rhodes, A. and Russell, S. (2003) Community Safety Post Incident Analysis: North East Victoria and Gippsland Fires 2003. Victoria: CFA.

O’Connor, Kevin (2001) Australia’s changing economic geography: a society dividing. Melbourne: Oxford University Press.

Packham, D. (1992) Bushfires In Australia: What Is The Risk? Australian Planner, 30.1, pp. 8-12.

Packham, D. (2003) Bushfire Threat to Nillumbik Shire (unpublished).

Pakenham-Berwick Gazette (2003) Ash Wednesday: 20 Years On. Victoria.

Pyne, S. (1991) Burning bush – A fire history of Australia, New York: Henry Holt.

Reinholtd, S., Rhodes, A. and Scillio, M. (1999a) Stay or Go? Understanding Community Response to Emergencies. Victoria: CFA.

Reinholtd, S., Rhodes, A. and Scillio, M. (1999b) Preparedness. Victoria: CFA

Robinson, Guy M., Loughran, Robert J., Tranter, Paul J. (2000): Australia and New Zealand: economy, society and environment, London: Arnold.

Rogers, A., Viles, H. and Goudie, A. (eds) (1999) The Student’s Companion to Geography, Blackwell: Oxford

Smith, D. (1999) Saving A Continent, Sydney: University of New South Wales Press, pp. 102-104.

Smith, K. (2001) Environmental Hazards: Assessing Risk And Reducing Disaster. London: Routledge.

The Age and The Adelaide Advertiser (1983) Ash Wednesday. Clayton: Wilke and Company Ltd.

Young, Ann R. M. (2000) Environmental change in Australia since 1788, Melbourne: Oxford University Press, pp. 85-88 and pp. 154-156.

Webster, J. (1986) The Complete Australian Bushfire Book. Melbourne: Nelson.

Williams, B., Brown, S., Greenberg, M. and Kahn, M. (1999) ‘Risk Perception in Context: The Savannah River Site Stakeholder Study’. Society for Risk Analysis, 200, pp. 1019-1035.

Wilson, A. and Ferguson, I. (1984) ‘Fight or flee? A case study of the Mount Macedon bushfire’. Australian Forestry, 47, pp. 230-236.