International Journal of Disaster Risk Reduction 6 (2013) 67 77

Contents lists available at ScienceDirect

International Journal of Disaster Risk Reduction

journal homepage: www.elsevier.com/locate/ijdrr

Defining adequate means of residents to prepare property for protection from wildfire
T.D. Penman a,n, C. Eriksen a,b, R. Blanchi c, M. Chladil d, A.M. Gill e, K. Haynes f, J. Leonard c, J. McLennan g, R.A. Bradstock a
a Centre for Environmental Risk Management of Bushfires, Institute of Conservation Biology and Environmental Management, University of Wollongong, NSW 2522, Australia b Australian Centre for Cultural Environmental Research, University of Wollongong, NSW 2522, Australia c CSIRO Ecosystem Sciences, Urban Systems Program, PO Box 56, Highett, Vic. 3190, Australia d Tasmania Fire Service, GPO Box 1526, Hobart, Tasmania 7000, Australia e Fenner School of Environment and Society, The Australian National University, Canberra, ACT 0200, Australia f Risk Frontiers Natural Hazards Research Centre, Department of Environment & Geography, Macquarie University, North Ryde, NSW 2109 Australia g School of Psychological Science, La Trobe University, Bundoora, Vic. 3086, Australia

a r t i c l e i n f o
Article history: Received 24 May 2013 Received in revised form 4 September 2013 Accepted 4 September 2013 Available online 13 September 2013 Keywords: Fire management Bushfire Preparedness Community engagement Resident Fire risk

abstract
Wildfire can result in significant loss of property and lives. Evidence shows that residents can decrease the risk of loss when they stay to defend their property. In order to safely defend a property, residents need to be adequately prepared for the wildfire conditions they face. Residents who wish to evacuate prior to the arrival of a wildfire also need to prepare their property and themselves for such an action. Despite the importance of preparation, there are no clear and quantifiable definitions of what it means to be prepared for different exposures to wildfire. Here we develop a model and definitions of what it means to be prepared for wildfire. The model considers the exposure of the property, the ability of the structure to withstand such an exposure and whether the resident(s) are adequately prepared. Preparation considers the physical and mental capacity of the residents, the condition of the grounds and the equipment available to defend the property. The model and definitions presented here focus on identifying points of weakness that should be addressed. An improved model and definitions will provide a benchmark for those residents who do prepare for wildfire, potentially reducing the risk of loss of property and life. However they are unlikely to address the large proportion of the at risk population that elect not to prepare for wildfire. Crown Copyright & 2013 Published by Elsevier Ltd. All rights reserved.

1. Introduction Wildfire, or unplanned fire, can result in significant loss of property and lives in many parts of the world [1,2]. Approximately 300,000 people were evacuated and 2223 houses were lost during wildfires in California, USA in

Corresponding author. Tel.: 61 2 4298 1232. E-mail address: tpenman@uow.edu.au (T.D. Penman).

2007 [3]. Wildfires in Greece in 2007 resulted in 76 fatalities and approximately 850 buildings destroyed [4]. In 2009, more than 2000 houses and 173 lives were lost during the Black Saturday fires in Victoria, Australia [5,6]. While fire management agencies deploy suppression resources in an attempt to protect property and lives, there are simply insufficient resources to protect every house from such destructive wildfires [7,8]. Residents can increase the probability of the survival of built structures by 36 times when they stay and defend

2212-4209/$ - see front matter Crown Copyright & 2013 Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ijdrr.2013.09.001

68

T.D. Penman et al. / International Journal of Disaster Risk Reduction 6 (2013) 67 77

their property from wildfire [912]. In Australia, residents have historically been encouraged to stay and defend their property where they are sufficiently prepared [13,14]. This is in contrast to the USA where major evacuations preceding wildfires have been common, with the implication of greater house loss [15]. However, there has been increasing recognition in the USA of the capacity of residents in some areas to actively prepare to stay and defend or shelter in place [3,1619]. Residents who stay to defend their property from wildfire must be adequately prepared in order to do so safely [12,14]. Residents who have defended their property from wildfire commonly report that the conditions faced were far more difficult than they expected [2023]. This has, in many cases, resulted in residents attempting to evacuate at the very last minute as the fire front approaches. However, attempting to escape an approaching fire front is extremely dangerous and is one of the primary causes of wildfire fatalities [13,14,24,25]. This points to the importance of residents preparing adequately for wildfire as well as fire management agencies providing adequate advice on how to adequately prepare for a wildfire [26]. There is considerable confusion regarding what it means to be adequately prepared for wildfires practically and mentally [27,28]. One of the main difficulties for residents arises when fire management agencies provide checklists regarding how to prepare a property for wildfires [2931]. Such approaches imply equivalent ratings of factors and rarely discuss mental preparedness [28,32]. Residents believe their property is well prepared, i.e. in good condition, by being able to tick off a large number of factors on the checklist [33]. In such cases, the simplest and cheapest actions (such as mowing lawns or clearing gutters) are more likely to be adopted. However, these actions do not necessarily result in any reduction in risk to the property [3436]. Information provided by management agencies generally has not accounted for variation in the type of fires a property may be exposed to. Fuel load, fuel structure, weather and topographic features will alter the fire intensity and severity (e.g., [37]), and hence impact on the ability of a resident to safely attempt to defend a structure. Actions which may assist residents in low severity fires are not necessarily going to be sufficient in high severity fires. For example, a 10 m wide fuel break may provide a high level of protection for a structure situated in a grass paddock in the absence of spotting but it is unlikely to provide significant protection for a structure in a contiguous forest with a high level of spotting. Similarly, it is far easier to defend a property under mild conditions compared to extreme fire weather. Clear and quantifiable definitions of what it means to be prepared for different exposures to wildfire are required to reduce the number of houses and lives lost during wildfires. In this paper we develop a model for determining whether it is safe to stay and attempt to defend a property and provide definitions for each of the components of preparedness. The model and definitions discussed in this paper were developed during a 1-day expert workshop attended by all authors, except JM and

RB who provided additional input prior to and after the workshop. The members of the workshop were selected on the basis that they have published papers regarding fire risk at the urban interface and/or house loss. All participants are professional researchers with specialist fire experience in the fields of engineering, ecology, fire behaviour and human geography. The model and associated definitions were developed through an open discussion to achieve a group consensus supported by published and grey literature. The model is to be used to estimate when it is safe for a resident(s) to safely stay and attempt to defend a property during a wildfire. It is beyond the scope of the model to predict the probability of a successful defence. The model deals with preparation within a time frame of weeks and months prior to the fire and does not consider in detail the actions of the resident on the day of a wildfire. For example, dehydration can be a major health issue for individuals defending against wildfire. Our model requires the resident to have sufficient drinking water available to combat potential dehydration, but does not provide recommendations regarding the rate of water consumption on the day or the decision to consume diuretics (e.g. alcohol).

2. Conceptual model A conceptual model was developed to determine whether it is safe to stay and attempt to defend a property (Fig. 1). The ability to safely stay and attempt to defend a property depends on the exposure of the property, the vulnerability (construction, design, material, and sitting) of the house [38] and the preparedness of the resident(s). An interaction between construction type and property exposure will determine whether it is safe for the resident to shelter in place. Similarly, an interaction between preparedness and the property exposure will determine if the resident is likely to be capable of attempting to defend the property. Central to the model structure is the predicted exposure of the property that can be expected from a wildfire.

House construction standard

Exposure

Capacity to defend

Safe to shelter in place

Prepared to exposure

Safe to stay and defend

Fig. 1. Conceptual model for determining whether it is safe to stay and attempt to defend.

T.D. Penman et al. / International Journal of Disaster Risk Reduction 6 (2013) 67 77

69

Exposure to the wildfire is estimated using radiant heat flux (RHF), a measure which has been used in a range of exposure studies (e.g. [39,40]). Calculations for RHF in our model are based on the equations presented in Appendix 3 of AS3959-2009 [41] which predict RHF based on the fire weather, distance to vegetation, slope, topographic position and the fuel load of the vegetation type. Fire weather in these models is based on the McArthur forest fire danger index (FFDI) which is based on a combination of temperature, humidity, wind speed and longer term drying through a drought factor [42]. RHF can be loosely translated into three types of exposure to the primary structure [41]: 1. Embers/firebrands and limited radiation only (RHF 012.5 kW/m2), hereafter the ember zone. 2. Radiant heat and embers/firebrands (RHF 12.529 kW/m2), hereafter the RH zone. 3. Flame contact, radiant heat and embers/firebrands (RHF 4 29 kW/m2), hereafter the flame zone.

scope of this paper. Australian construction standards are based on the one in 50 year FFDI values that have been estimated for each region. These standards do not consider situations where the fire weather exceeds 100 FFDI. For this reason, under our model it is never considered safe to stay and attempt to defend under conditions exceeding 100 FFDI.1 The other major component of the model relates to the issue of preparedness (Fig. 1). Commonly, preparedness is considered on an unquantified scale (e.g. [28,32,35,50]) which is problematic as residents are incapable of determining whether they are appropriately prepared for the situation to which they may be exposed. We provide definitions for preparedness based on the level of exposure the property is expected to experience from the fire front. As above, we consider three exposure levels ember zone, RH zone and flame zone. Failure to achieve preparedness according to definitions means that a household is not deemed safe to stay and attempt to defend their property. 3. Dening preparedness

These definitions are based on the expected exposure from the adjacent vegetation for a given fire weather. Exposure can vary if the values for weather or fuel loads in the adjacent vegetation are higher than the predicted values. For example, weed invasion can alter fuel structure and fuel load, and hence the exposure on the property [43]. Furthermore, the condition of the grounds (see below) will impact upon how the exposure from the adjacent vegetation translates to the ultimate exposure on the house. For these reasons, the predicted exposure of the property is used here as a guide to determine the suitability of the structure's construction and the level of preparedness. Residents attempting to defend properties will generally be required to shelter in place while the main fire front passes [14,24,44]. This is necessary as a RHF load as low as 4 kw/m2 is considered sufficient to cause pain to exposed skin after only 1020 s [45] and exposure to 7 kw/m2 is likely to result in burn injuries [46,47]. For the purpose of this model, we assume that the goal of a resident is to defend a house or similar structure. This will therefore be the place in which the resident(s) will seek shelter as the fire front passes. To do this safely the house must be built and maintained to a construction standard, which is able to withstand the predicted exposure from the wildfire. If the house is not built and maintained to a suitable standard the house may ignite during the passage of the main fire front. This will force the resident(s) to flee into potentially unsafe environments, or be trapped within the burning building. Nevertheless, where a house is built to a suitable construction standard, the resident(s) must ensure that at least one safe passage from the house to a place of last resort has been identified. In Australia, the construction standard AS3959-2009 Construction of Buildings in Bushfire Prone Areas [41] provides the basis for construction standards, although analogues exist elsewhere (e.g. [48]). We acknowledge that there are a number of limitations to this standard (for a detailed discussion see [5]) but these are beyond the

There are a number of components of preparedness (e.g. [28,51]). We have reclassified these as: personal capacity of the resident(s) defending the property; available equipment; and the condition of the property's grounds. Personal capacity relates to both the practical, physical and mental capacity of the resident(s) that are required to endure the demanding conditions that will be faced when protecting a property from wildfire. The available equipment defines the nature and state of equipment that are required to defend the property for the predefined exposure. Condition of the grounds primarily deals with the fuels within the property that may ignite during a wildfire. Each of these is discussed more fully below with summaries of the classifications by exposure level presented in Tables 13. An overall measure of preparedness is determined as the level of exposure to which a resident(s) is capable of defending their property (Fig. 2). This is calculated by determining the minimum capacity across personal capacity, available equipment and condition of the grounds. For example, if a resident is determined to have personal capacity and condition of the grounds at appropriate levels to deal with a flame zone exposure, but only has available equipment to cope with an ember zone exposure, then the capacity to defend is limited to properties determined to be in an ember zone. 3.1. Personal capacity Residents who attempt to defend their property must be aware of the physical and mental demands that the fire will place on them and how this may affect their decisionmaking processes [20,52,53]. Preparing a plan of action or
1 Fire weather indices can exceed this value and forecasts are not always accurate. Given the many uncertainties in the system it may be prudent for residents to prepare for the worst that the region has experienced in the past while keeping in mind that with changing global environments, so-far unprecedented weather events are to be expected [49].

70

T.D. Penman et al. / International Journal of Disaster Risk Reduction 6 (2013) 67 77

Table 1 A summary of the variables and their necessity or otherwise for the personal capacity component of preparedness. This table is a summary only and must be read in conjunction with Section 3.1 for complete definitions of the components. Flame zone Personal ability Good physical health to work in a fire environment, see Section 3.1 Mobility to move capably around the property unassisted Physical ability to work in an environment with heavy smoke from wildfires Mental ability to endure dangerous conditions Physical stamina to work for at least 6 h Mental stamina to work for at least 6 h Experience or detailed understanding in defending property from bushfires Assistance from other residents Dependents No human dependents (young, immobile, elderly) No animal dependents at the house, e.g. pets No animal dependents away from the house, e.g. stock Plan A written fire survival plan Identified roles and communicated actions Plan deals with multiple contingencies Identified place of last resort Battery powered radio RH zone Ember zone

Required Required Required Required Required Required Required Required Required Required Required Required Required Required Required Required

Required Required Required Required Required Required Optional Optional Required Optional Required Required Required Required Required Required

Optional Required Required Required Optional Optional Optional Optional Optional Optional Optional Required Required Required Required Required

Table 2 A summary of the variables for the equipment component of preparedness. This table is a summary only and must be read in conjunction with Section 3.2 for complete definitions of the components. Flame zone PPE Goggles Helmets Boots Dust mask Clothing Gloves Woollen blanket Structure and grounds Gutter protection Pump 10,000 L static water 300 L water holding vessels Reliable mains Hoses to reach around the structure Spare hoses stored indoors Multiple water points Rake and spade Bucket and mop RH zone Ember zone

Required Required Required Required Required Required Required Required Diesel pump required Required Desirable Desirable Required Required Required Required Required

Required Required Required Required Required Required Required Required Diesel or petrol pump required Required Desirable Desirable Required Required Required Required Required

Required Required Required Required Required Required Required Required Pump desirable Desirable Required Required if no static water available Required Desirable Desirable Required Required

a bushfire survival plan prior to the fire season is vital as it guides the decision making processes on the day of a fire [36,54]. The plan must identify actions and assign roles to all individuals in the household whether they are staying to defend or leaving early [55]. A survival plan should also account for multiple contingencies due to the highly variable nature of wildfire, such as fallen trees or power lines resulting in closure of potential escape routes [56], fires approaching the property from an unexpected direction, failure of equipment such as water pumps and hoses, and resident(s) not being at home when the fire starts [21,54].

Actions in the plan may be triggered by a variety of sources, such as actual visual cues (smoke or flames), television, social media or radio. While television and social media rely on power sources and telecommunications, which may be lost during a wildfire [56,57], a battery operated radio will continue to work under most circumstances. For this reason, a battery operated radio is included as a necessity for all residents in fire risk areas. A key component of any survival plan is an identified means of exiting the property and travelling to places of last resort safely [35]. A place of last resort is a location that is safe for the resident(s) to shelter should the built

T.D. Penman et al. / International Journal of Disaster Risk Reduction 6 (2013) 67 77

71

Table 3 A summary of the variables for the condition of the grounds component of preparedness. This table is a summary only and must be read in conjunction with Section 3.3 for complete definitions of the components. Flame zone Landscape design Any wooden fences at least one times the height of the fence from the house Any brushwood fences at least three times the height of the fence from the house No wooden sleepers within 1 m of the structure No dry lawns within 1 m of the house No leaf or bark mulch within 2 m of the house No shrubs within 2 m of wooden or glass components of the structure Property maintenance A 2 m gap between tree branches and the ground Gutters clear of leaves and twigs No trees overhanging the roof Underfloor and underdeck spaces enclosed Underfloor and underdeck spaces clear of leaves and twigs Roof maintained in good condition Heavy fuels Flammable liquids stored in sheds 4 6 m from the house Release valves of all gas bottles facing away from the house BBQ gas bottles stored securely No combustible doormat No combustible outdoor furniture within 2 m of the house or deck No woodpiles adjacent to the structure RH zone Ember zone

Required Required Required Required Required Required Required Required Required Required Required Required Required Required Required Required Required Required

Required Required Required Required Required Required Required Required Required Required Required Required Required Required Required Required Required Required

Required Required Required Required Required Required Required Required Required Desirable Required Required Required Required Required Required Required Required

Personal capacity

Available equipment

Condition of the grounds

Capacity to defend

Fig. 2. Graphical depiction of the preparedness sub-model.

structure (e.g. the house) be alight to a point where the fire becomes unmanageable for the resident(s) to attempt to defend. Without a place of last resort, it is considered too risky for a resident to stay and attempt to defend a property regardless of the exposure. Places of last resort will vary depending on the surrounding landscape. Examples of places of last resort may include a beach, sports field, gravel or concrete car park, cleared paddock, or a well-prepared building all of which should be away from vegetation [58]. It should be noted that these are all conditional on the surrounding environment and personal protective equipment, particularly clothing, may still be required. The role(s) any resident plans to take in the defence of their property will be limited by their physical and mental capacity. Defending a property from wildfire can be physically and mentally demanding [59]. The property may need to be defended continuously under extreme temperatures for

many hours, starting several hours before the fire front arrives and continuing well after the front has passed [60]. Dehydration is a serious concern and residents must have 10 L of drinking water for every person who will stay and defend the property. Residents should not rely on mains water for drinking water as this can be lost during a wildfire [35]. The environment in which they attempt to defend their property can be dangerous, with heavy smoke resulting in low visibility, strong winds carrying flying debris, high temperatures and loud noise from the fire [12,28,61]. Mobility is required to move to extinguish any fires on the property and to move to the place of last resort if necessary. Physical health of a resident is also important as the conditions of the fire may exacerbate existing ailments, such as respiratory conditions that temporarily incapacitate residents [6264]. The residents must also be mentally capable of working in these intense conditions. Stress and anxiety may impair the ability to make safe and rational decisions [65]. The level of physical and mental stamina required increases with houses with higher exposures (Table 1). Residents attempting to defend properties in the flame zone are exposed to the most extreme fire conditions. Flame contact with the house will mean that much of the grounds will also be exposed to flame contact, resulting in a large number of potential ignitions within the property boundaries that require attention. To cope with such a fire residents require detailed knowledge or experience of fire behaviour (Table 1). This experience is likely to improve their ability to make appropriate decisions under stress and in a timely manner to ensure the safety of all involved [66,67]. Furthermore, due to the difficulties in defending under these conditions, it is considered vital that more than one physically and mentally capable resident is

72

T.D. Penman et al. / International Journal of Disaster Risk Reduction 6 (2013) 67 77

defending the property. It should be noted that it is desirable to have more than one capable resident defending property regardless of the exposure. Planning roles and actions within a survival plan needs to ensure that all residents have accounted for the needs of all human and non-human dependents during the fire. Human dependents include the young, immobile, disabled and elderly. Non-human dependents include domestic pets, horses and livestock that reside in paddocks. Residents defending properties in the flame and RH zone should have no dependents as the risk to human dependents and pets is too high in these zones, particularly if the resident needs to escape to a place of last resort. Furthermore, the attention of the resident is split between defending property and ensuring the safety and wellbeing of the dependents. Livestock may also require the resident(s) to move away from the structure, where a greater risk to life may occur [13,14]. While it is preferable to have no dependents in the ember zone, older children, immobile, elderly and pets could shelter in the structure if it is built and maintained to a suitable construction standard considering all sources of exposure to the structure. However, infants should not be present without additional residents due to their requirement for more constant supervision. 3.2. Available equipment To defend a property from wildfire a minimum set of equipment is required for each exposure zone. Equipment can be categorised into three main groups: personal protective equipment (PPE); equipment to protect the built structure; and equipment to address fire on the grounds. These measures can be considered a hierarchy, which reflects the need to firstly protect the individual from injury, secondly to protect the house from igniting and finally to minimise the extent of fire on the grounds. Due to the inherent dangers the fire environment presents, residents attempting to defend their property from wildfire must have adequate equipment and clothing to protect themselves from injuries, regardless of the exposure of the property. Goggles are required to prevent smoke and fine particles from entering or irritating the eyes. A helmet is necessary to protect the head from falling debris and radiation. Dust masks are required to prevent smoke and fine particles from entering the airways and minimising the risk of respiratory problems [62]. The skin needs to be protected from embers, radiant heat and flame exposure: this can be achieved in part by covering up with clothing made from low flammability materials such as thick cotton, wool or denim [30,68,69]. Similarly, gloves can also protect the skin from burns and allow residents to move fallen debris away from the built structure. Boots are required to protect the feet from embers, radiant heat, flames and falling debris. A wet woollen blanket should also be available to cover each resident as protection from radiant heat and flame contact if required. Defending directly against the fire front in the flame zone or the RH zone is unlikely due to the extreme risk to residents. Residents protecting a structure from ignition should focus primarily on preventing the ingress of

embers/firebrands [10,67]. The main areas of concern is the accumulation of leaf litter and twigs in gutters and around other building details, that can then be ignited by embers, or embers entering the built structure through gaps in walls or the roof [10,11,70]. Once ignition has occurred, fire can spread through building cavities, such as the roof space, involving combustible building elements such as the house frame. To protect against gutter based ignition residents can take a variety of approaches. The first is to simply not have gutters on the roof, thereby eliminating the potential for the problem to occur. However, this needs to be balanced against the issue of storm water drainage in other seasons. A second solution is to install gutter guards, i.e. a fine mesh which prevents the accumulation of litter in the gutters. The remaining solutions rely on equipment to respond to the threat of the fire. Residents could install sprinklers on the roof of the structure. When a fire is approaching the resident could block the down pipes and then start the sprinklers. The sprinklers then fill the gutters with water thereby preventing the ignition of any litter that is present [71,72]. A simpler version of the same approach is to use hoses to fill the gutters with water. Both these approaches rely on a suitable enclosed pump, a water source or reliable mains water supply,2 hoses to reach all of the gutters and ladders to allow the able-bodied resident to fill the gutters. In the flame zone the only suitable pump is a diesel pump, due to the lower flammability of the fuel, whereas in the RH and ember zones both diesel and petrol pumps are considered adequate. Electric pumps are not considered suitable due to the high potential for power failures during a wildfire [56]. A third approach is building construction, such as steel frames, that avoids combustible building cavities (http://www.nash.asn.au, accessed May 2013). Controlling fires in the property grounds is dependent on having sufficient water, an appropriate delivery system and basic tools. Fire behaviour in the flame and RH zones is expected to be extremely demanding on water resources. We conservatively estimate that a minimum of 10,000 L of water would be required to safely attempt to defend a house, but house size, environmental circumstances and the specific water delivery systems used will strongly influence this. It is necessary to have sufficient hoses to ensure that all of the structure can be reached with a hose. It is important to have spare hoses stored inside the built structure in case of damage caused by the fire to the hoses (e.g. melting). Similarly it is important to have multiple water points so that if one point becomes damaged or is engulfed by fire, the defence of the property can continue. In the case of a property in the ember zone, 300 L of water holding capacity would be an absolute minimum requirement provided the tools described below are available to extinguish small fires ignited by embers or fire brands. Note that this amount of water and the

2 A reliable mains water supply is rare. It occurs when the water supply can be guaranteed during the course of the fire even if there is a power failure. An example of where this may occur is when the town water supply is of a higher elevation than the town and gravity will ensure there is sufficient pressure for the taps and hoses. However, high demand during a fire may reduce water pressure below useful levels [7].

T.D. Penman et al. / International Journal of Disaster Risk Reduction 6 (2013) 67 77

73

associated tools would only be effective to defend the house against most types of ignition but not sufficient to address a house fire if it is fully ignited. For all exposures, it is recommended that basic tools are available, i.e. rakes, spades, bucket and a mop. Rakes and spades can be used to put out embers and to create breaks in fuels to contain the spread of small fires within the grounds. A bucket of water and a mop is an effective means of extinguishing small fires that start from embers or firebrands that fall within the grounds. 3.3. Condition of the grounds Preparing the grounds to an appropriate condition is primarily concerned with reducing the available fuels for combustion. There are three broad categories that need to be considered landscaping, property maintenance and miscellaneous heavy fuels. Landscaping refers to elements of the garden that have developed as part of an active or passive garden design process. Alterations to these elements result in a long term change in the appearance of the property. Property maintenance refers to actions that are required at least annually and could be colloquially referred to as cleaning up. Finally, miscellaneous heavy fuels are a group of flammable substances that commonly occur within a property and their placement could have positive or negative consequences for fire behaviour. Distances from the structure are provided for elements within each of these categories. These values were determined as the minimum distance for which the estimated heat load that would be transferred from all sources (including houses and other fuel sources on adjoining properties) acting simultaneously on the structure is less than 70% of the radiant heat flux tolerance of the construction standard of the house (see above). Landscaping of a garden can have significant influence on the probability of a property igniting [1,10]. Selection of appropriate elements can not only reduce the risk of ignition, but also can improve a resident's ability to safely attempt to defend their property from wildfire. Combustible fences have the potential to move fire quickly through a property and to transfer significant heat loads to a built structure, potentially igniting it [67,73]. Wooden fences should be at least the height of the fence from a built structure, whereas any brushwood fences should be three times the height of the fence from the built structure. Similarly, wooden sleepers used as garden edges or retaining walls should not be within 1 m of the built structure. Fire can easily move through dry lawns to a structure [67]. Removal of dry lawns or maintenance of green lawns within 1 m of a structure can reduce or eliminate this risk. Leaf or bark mulch will ignite easily from embers and should not be within 2 m of the built structure [74]. Shrubs should not be within 2 m of any wooden or glass elements of a house. A property that is appropriately maintained needs to have minimised fine fuels (i.e. leaf litter and twigs) and have a discontinuous fuel structure in an attempt to reduce the risk of uncontrollable fire behaviour [60,74]. Dry fine fuels are a high risk as they will more readily ignite if exposed to embers during a wildfire. These fuels

pose the greatest risk when they occur on or adjacent to flammable elements of the built structure. The removal of fine fuels from roof valleys, gutters, underfloor or underdeck spaces is therefore required to prevent a secondary ignition of the structure [5,74]. Similarly, maintaining the roof so as to prevent the accumulation of fine fuels and prevent embers from penetrating the roof structure is important [5]. Fuel structure is important as it can influence flame height and hence the ability of a resident to safely attempt to defend their property. A gap of 2 m is recommended between tree branches and the ground, which includes clearing of shrubs [74]. This gap is intended to prevent the transfer of flames into the tree canopy and any resulting crown fire which cannot safely be suppressed by residents. We note that some bark types can carry flames as well so plants with non-ignitable barks are recommended. There is a wide diversity of fuel sources and configurations on properties that have the potential to contribute to the thermal exposure of a home. These range from the familiar mix of heavy and fine vegetation, through to complex combustible man-made heavy fuel objects that form structures (structural fuel) [75]. Objects immediately surrounding a house, such as fences, adjacent houses, outbuildings, ground cover, vegetation, vehicles, rubbish bins, and woodpiles can significantly influence asset exposure [10,60]. These elements can increase or reduce the risk of house loss by playing a role as either a heat source or attenuator of flame, radiant heat flux and/or embers [75]. These fuels are usually dry and may be relatively ignitable by embers. Once they have ignited residents may not be able to extinguish burning heavy fuels with the equipment described above. Outdoor furniture is often made of combustible materials and represents a significant fire risk if it is within 2 m of a structure or situated on a wooden deck. Embers or firebrands can ignite outdoor furniture [76] and the resulting radiant heat or flame contact may be sufficient to ignite the built structure or deck, resulting in fire that cannot be suppressed by a resident. The risk from outdoor furniture ignition could be minimised if it is moved more than 2 m from the structure and decks. A combustible doormat can also readily ignite and subsequently ignite adjacent timber doors or doorframes [77]. As with furniture, moving the doormat away from the house will minimise this risk. Flammable liquids, such as paint and petrol, are difficult to suppress once alight. The safest manner is to store these liquids in a garden shed greater than 6 m from the built structure of concern. Under certain circumstances, gas bottles can become explosive thereby creating an unsafe environment for residents to be in. Fixed gas bottles must be wellsecured and have the release valve facing away from built structures so that if it vents it does not result in any secondary ignitions. Gas bottles that fall over and continue to be heated have the potential to explode. Gas bottles associated with barbecues should be removed from the barbecue and stored upright on stable ground in an open area well away from other heat sources with the pressure relief valve facing away from other structures and egress paths. Storage of these bottles in sheds or on decks is not advised.

74

T.D. Penman et al. / International Journal of Disaster Risk Reduction 6 (2013) 67 77

4. Discussion There is a significant body of research examining why people do or do not prepare [50,53,78,79] but very little defining what it means to be prepared [28,74]. The information regarding preparation generally focuses on generic lists of actions that are considered applicable to all situations. Here we have developed a model of what is required of residents to prepare for wildfire based on the fire conditions the property can expect to be exposed to. Guidelines, even quantitative ones, as we have provided, are often surrogates necessary for public discourse. Considering the multitude of circumstances of people, assets (social, environmental, economic) [2], weather, terrain, fires, local fuels and structural designs and conditions providing a set of universally-applicable and practical instructions is impossible; simplifications are essential. The definitions presented here are based on published and grey literature and bring together experiences from Australia and North America. We believe these definitions represent a major advance in our understanding, specifically for Australia, but also more broadly across all fire prone environments. The model seeks to identify and address any potential points of weakness in the resident's preparation. Our model was developed specifically to determine the conditions under which it will be safe to stay and attempt to defend. We deliberately prioritise the safety of residents over the survival of built structures. Failure to completely achieve just one aspect of the definitions provided indicates that the residents are not considered safe to stay and defend. All steps are necessary to ensure it is safe to stay and defend. By setting a minimum standard, we have overcome the major limitations of the checklist approach, where residents can believe they are prepared because they have fulfilled most of the criteria, but may not have undertaken the critical actions [3335]. Definitions provided here will be of greatest immediate benefit to residents who elect to prepare for wildfire. These definitions provide a benchmark to determine adequate preparation and will be of greatest immediate benefit to residents who elect to prepare for wildfire. Many residents who have defended their property commonly report feeling underprepared because the conditions they faced were far more difficult than they expected [2023]. By having a model which predicts likely exposure, as well as providing definitions for preparations residents can take the appropriate measures (or because of greater awareness make the decision to leave early), thereby reducing the risk of loss of life and property. An improved definition for preparedness is unlikely to have any significant impact on the proportion of residents which adequately prepare for wildfire. Only a small proportion of people prepare adequately for wildfire [20,80]. There are a range of reasons affecting individuals decision to prepare or not to prepare, which include direct personal wildfire experience [50,78,81], a preference for a more natural environment [82,83], neighbourhood attitudes to preparation [50,78,79,84,85] or the resident(s) perception of risk from wildfire [27,84]. Many only begin to prepare their property once a wildfire has ignited in the local area.

Tibbits and Whittaker [35, p. 286] argue that it is perhaps inevitable that most people will wait until there is a fire before they begin to make preparations that go beyond basic maintenance. Using the model presented here, we would strongly argue that residents who only intend to prepare once the fire has started are not safe to stay and defend their properties. Residents need to plan and prepare or maintain their property for wildfire prior to and during the fire season [32,86]. Fires can ignite and impact upon property rapidly within a single day, e.g. the Black Saturday fires in Victoria 2009. In such cases, residents will have little or no time for adequate planning and preparation. Even when residents have several days warning there is often insufficient time to prepare adequately according to our definitions. Residents in fire risk areas need to design, maintain and landscape gardens accordingly (Table 3). Similarly, informed decisions about where and how to locate heavy fuels are vital in fire risk areas in order to maintain a path to safe egress and place of last resort (Table 3). Equipment that is required to defend the property needs to be purchased prior to the fire season and regularly tested so that it is available and in working order whenever a fire occurs (Table 2). Routine maintenance such as cleaning gutters, trimming trees and shrubs, removing dead materials and mowing lawns need to be done regularly throughout the fire season (Table 3). Attempting to do so as a fire approaches is dangerous for two main reasons. Firstly, the time taken to undertake maintenance will distract residents from the implementation of a suitable fire plan. Secondly, these actions will only allow residents to move fuels within the property and not allow for their disposal, potentially creating additional threats to the residents and the structure. Households must also maintain the construction standard of the house throughout the year as failure to do so may increase the vulnerability of the property to ignition [5], thereby making it unsafe to shelter in during the fire front (Fig. 1). Adequate preparation is clearly going to be associated with measurable time and costs to each household which needs to be balanced with other environment risks and everyday priorities [53,87]. Preparation for wildfire should not be limited to those who elect to stay and defend their property. Many residents fail to appreciate that planning and preparation is required even when the plan is to leave early. Often in such cases residents do not invest directly in preparation because they believe the property will be adequately protected by fire suppression, disaster funding or insurance [8891]. However, for these residents planning is required to establish trigger points, list personal belongings and valuables to pack, determine routes of ingress and egress, and identify points of contact during the emergency [28]. Furthermore, preparation of the property is required even when the plan is to leave early as fires can start nearby and move quickly preventing escape. Communications can also fail resulting in people being unable to safely leave. In both scenarios, residents may be forced to shelter at the house, or nearby, and defend their property from the fire. There is only limited information available at present regarding the extent to which preparing a property actually reduces the risk of house loss in the absence of residents

T.D. Penman et al. / International Journal of Disaster Risk Reduction 6 (2013) 67 77

75

during the wildfire. Relationships between some aspects of preparedness and risk reduction have been addressed in the literature, such as the altering of the risk of loss through changed vegetation types and coverage with landscaping [1,10]. Other important aspects, such as the presence of heavy fuels, have not received the same attention (although see [1]) due, in part, to routine post fire assessments of house loss not being inclusive of all the variables we discuss here. An overall assessment is required of how being prepared affects the probability of a built structure surviving varying exposures to wildfire in the presence and absence of residents. An empirical analysis of this issue is likely to be difficult to achieve due to the low frequency and spatial distribution of house loss events and the number of variables involved in any given wildfire [2]. Simulation models have the potential to overcome this as the number of replicates and variables considered is limited only by the available data, software and hardware [92]. Key results of such studies could then be validated against existing or future house loss datasets. However, calibration of such models will be extremely challenging as there are significant contingencies in the system. For example, tree crowns can shield a structure to some extent from radiation and ember storms unless they catch alight and become an even more local ignition source for the structure. Whatever form of analysis takes place, providing context-specific advice from it, along with reasons for doing so, is necessary for people to be able to evaluate it against the specifics of their situation. 5. Conclusion Residents in wildfire-prone areas who wish to maximise their chance of survival need to prepare themselves and their property for wildfire regardless of whether their plan of action is to leave early, wait and see, or stay and defend. Here we have developed a model that identifies how residents should prepare based on the type of fire that the structure is predicted to be exposed to. This represents a significant advancement in preparedness definitions which have traditionally been generic checklists for all properties in fire risk areas. Comparing our model with published data suggests that the level of preparation currently undertaken is generally insufficient to safely stay and attempt to defend a property [20,21]. Providing measurable standards, as we have done here, may help householders understand what is required to prepare in order to reduce their risk of loss from wildfire. However, we acknowledge that information is not the only hurdle to increasing the preparation of residents. Measurable standards will provide a basis for scientists to assess resident's preparation for wildfire and also a semiquantitative basis for estimating the role of preparation in reducing losses during a wildfire.

using funding from National Disaster Resilience Program. Discussions with Simon Heemstra, Peter Jackson, Stuart Midgely, Melissa O'Halloran and Owen Price aided in the development of this paper. Bronwyn Horsey provided editorial assistance in preparing the manuscript. References
[1] Gibbons P, van Bommel L, Gill AM, Cary GJ, Driscoll DA, Bradstock RA, et al. Land management practices associated with house loss in wildfires. PLoS One 2012;7:e29212. [2] Gill AM, Stephens SL, Cary. GJ. The worldwide wildfire problem. Applied Ecology 2013;23:43854. [3] McCaffrey S, Rhodes A. Public response to wildfire: Is the Australian stay and defend or leave early approach an option for wildfire management in the United States? Journal of Forestry 2009;107: 915. [4] Bassi S, Kettunen M. Forest fires: causes and contributing factors in europe. Brussels: European Parliament: Policy Department Economic and Scientific Policy, 2008 http://www.ieep.eu/assets/392/ forest_fires_report.pdf. [5] Leonard J, Blanchi R, Lipkin F, Newnham G, Siggins A, Opie K, et al. Building and land-use planning research after the 7th February Victorian bushfires: preliminary findings, 2009. Melbourne: CSIRO and Bushfires CRC Report No.: Final Report USP2008/018-CAF122-2-12. [6] Victorian Bushfires Royal Commission. Final report: volume 1, the fires and fire-related deaths. Melbourne: Parliament of Victoria; 2009. http://www.royalcommission.vic.gov.au/commission-reports/ final-report. [7] Gill AM, Stephens SL. Scientific and social challenges for the management of fire-prone wildlandurban interfaces. Environmental Research Letters 2009;4:034014. [8] Gill AM. Underpinnings of fire management for the conservation of biodiversity in reserves. Melbourne: Victorian Department of Sustainability and Environment, 2008 http://www.dse.vic.gov.au/fireand-other-emergencies/publications-and-research/fire-research-re ports/research-report-73. [9] Blanchi R, Leonard J. Property safety: judging structural safety. In: Handmer J, Haynes K, editors. Community bushfire safety. Melbourne: Bushfire Co-operative Research Centre; 2008. p. 7785. [10] Ramsay GC, McArthur NA, Dowling VP. Preliminary results from an examination of house survival in the 16 February 1983 bushfires in australia. Fire and Materials 1987;11:4951. [11] Wilson AAG, Ferguson IS. Predicting the probability of house survival during bushfires. Journal of Environmental Management 1986;23: 25970. [12] Whittaker J, Handmer J, Mercer D. Vulnerability to bushfires in rural Australia: a case study from East Gippsland Victoria. Journal of Rural Studies 2012;28:16173. [13] Tibbits A, Handmer J, Haynes K, Lowe T, Whittaker J. Prepare, stay and defend or leave early. Evidence for the australian approach. In: Handmer J, Haynes K, editors. Community bushfire safety. Collingwood: CSIRO Publishing; 2008. p. 5971. [14] Haynes K, Handmer J, McAneney J, Tibbits A, Coates L. Australian bushfire fatalities 19002008: exploring trends in relation to the prepare, stay and defend or leave early policy. Environmental Science & Policy 2010;13:18594. [15] Stephens SL, Adams MA, Handmer J, Kearns FR, Leicester B, Leonard J, et al. Urbanwildland fires: how California and other regions of the US can learn from Australia. Environmental Research Letters 2009;4:014010. [16] Cova T, Drews F, Siebeneck L, Musters A. Protective actions in wildfires: evacuate or shelter-in-place? Natural Hazards Review 2009;10:15162. [17] Mutch R, Rogers M, Stephens S, Gill AM. Protecting lives and property in the wildlandurban interface: communities in Montana and southern California adopt Australian paradigm. Fire Technology 2011;47:35777. [18] Paveglio TB, Carroll M, Jakes PJ. Alternatives to evacuation protecting public safety during wildland fire. Journal of Forestry 2008;106:6570. [19] Paveglio TB, Carroll M, Jakes PJ. Adoption and perception of shelterin-place in California's rancho santa fe fire protection district. International Journal of Wildland Fire 2010;19:67788. [20] McLennan J, Elliott G, Omodei M. Householder decision-making under imminent wildfire threat: stay and defend or leave? International Journal of Wildland Fire 2012;21:91525.

Acknowledgements The work was funded by a commercial research contract to the Centre for Environmental Risk Management of Bushfires, University of Wollongong from the NSW Rural Fire Service,

76

T.D. Penman et al. / International Journal of Disaster Risk Reduction 6 (2013) 67 77 [47] Butler B, Cohen J. Firefighter safety zones: a theoretical model based on radiative heating. International Journal of Wildland Fire 1998;8: 737. [48] CBC. California build code materials and construction methods for exterior wildfire exposure title 24, part 2, vol. 1 of 2. Sacramento, CA; 2012 http://publicecodes.cyberregs.com/st/ca/st/b200v10/. [49] Clarke HG, Smith PL, Pitman AJ. Regional signatures of future fire weather over eastern Australia from global climate models. International Journal of Wildland Fire 2011;20:55062. [50] McGee TK, Russell S. It's just a natural way of life an investigation of wildfire preparedness in rural Australia. Global Environmental Change B: Environmental Hazards 2003;5:112. [51] Morrissey SA, Reser JP. Natural disasters, climate change and mental health considerations for rural Australia. Australian Journal of Rural Health 2007;15:1205. [52] Krusel N, Petris SN. A study of civilian deaths in the 1983 ash wednesday bushfires Victoria, australia. Mt Waverley, Victoria: Country Fire Authority. [53] Eriksen C, Gill N. Bushfire and everyday life: examining the awareness-action gap in changing rural landscapes. Geoforum 2010;41:81425. [54] O'Neill SJ, Handmer J. Responding to bushfire risk: the need for transformative adaptation. Environmental Research Letters 2012;7: 014018. [55] Eriksen C. In Press. Gender and wildfire: landscapes of uncertainty. New York, London: Routledge; 2013: ISBN: 978-0-415-50270-2 [In press]. [56] Matthew SC, Yoshitaka K, Steven ED, John CB, John AE. Causal reasoning processes of people affected by wildfire: implications for agencycommunity interactions and communication strategies. Western Journal of Applied Forestry 2004;19:18494. [57] Marchetti N. Telecommunications in disaster areas. River publishers series in comunications. Aalborg, Denmark: River Publishers; 2010. [58] McLennan J. Use of informal places of shelter and last resort on 7 February 2009. Melbourne, Australia: School of Psychological Science La Trobe University and Bushfire Cooperative Research Centre. http://www.bushfirecrc.com/research/vicfires-taskforce/ research-reports. [59] Mangan RJ. Improving firefighter safety in the wildland-urban intermix. USDA Forest Service, Technology and Development Program; 2000. [60] Blanchi R, Leonard JE. Investigation of bushfire attack mechanisms resulting in house loss in the ACT bushfire; 2003. http://www. bushfirecrc.com/managed/resource/act_bushfire_crc_report.pdf. [61] Fromm M, Tupper A, Rosenfeld D, Servranckx R, McRae R. Violent pyro-convective storm devastates Australia's capital and pollutes the stratosphere. Geophysical Research Letters 2006;33:L05815. [62] Mirabelli MC, Knzli N, Avol E, Gilliland FD, Gauderman WJ, McConnell R, et al. Respiratory symptoms following wildfire smoke exposure: airway size as a susceptibility factor. Epidemiology 2009; 20:4519, http://dx.doi.org/10.1097/EDE.0b013e31819d128d. [63] Bowman DMJS, Johnston FH. Wildfire smoke, fire management, and human health. EcoHealth 2005;2:7680. [64] Reisen F, Blanchi R, Tibbits A. Potential health impacts to residents from smoke exposure during bushfires. Forest Ecology and Management 2006;234(Supplement):S148. [65] McLennan J, Omodei M, Elliott G, McNeill I, Dunlop P, Suss J. Bushfire survival-related decision-making: what the stress and human performance literature tells us. In: Proceedings of the Bushfire Cooperative Research Centre & AFAC 2011 Conference Science Day. Melbourne: Bushfire Cooperative Research Centre; 2011. p. 30719. [66] Lazarus G, Elley J. A study of the effect of household occupancy during the ash wednesday bushfire in upper beaconsfield victoria. Victoria, Australia: Chisholm Institute of Technology; 1983. [67] Gill AM. Landscape fires as social disasters: an overview of the bushfire problem. Global Environmental Change Part B: Environmental Hazards 2005;6:6580. [68] McLean AD. Burns and military clothing. Journal of the Royal Army Medical Corps 2001;147:97106. [69] Behnke WP. Predicting flash fire protection of clothing from laboratory tests using second-degree burn to rate performance. Fire and Materials 1984;8:5763. [70] Ahern A, Chladhil M. How far do bushfires penetrate urban areas? In: Proceedings of Australian disaster conference. Canberra, ACT; Emergency Managemenent of Australia; 1999. p. 216. [71] Mitchell JW. Wind-enabled ember dousing. Fire Safety Journal 2006;41:44458. [72] Potter M, Leonard J. Spary system design for ember attack: research findings and discussion paper. Melbourne, Australia: CSIRO

[21] Whittaker J, Haynes K, Handmer J, McLennan J. Community safety during the 2009 Australian Black Saturday bushfires: an analysis of household preparedness and response. International Journal of Wildland Fire 2013;22:841-9. [22] Whittaker J, McLennan J, Elliot G, Gilbert JB, Handmer J, Haynes K. Human behaviour and community safety. Melbourne: Bushfire Cooperative Research Centre; 2009. [23] Hyland A. Kinglake-350. TTP Company, TTP Company editor. Melbourne; 2011. [24] Handmer J, Tibbits A. Is staying at home the safest option during a bushfire? Historical evidence for an Australian approach Natural Hazards Review 2005;6:8191. [25] Handmer J, O'Neil S, Killalea D. Review of fatalities in the February 7, 2009, bushfires prepared for the Victorian Bushfires Royal Commission April. Melbourne: Bushfire CRC; 2010. [26] Eriksen C, Prior T. The art of learning: wildfire, amenity migration and local environmental knowledge. International Journal of Wildland Fire 2011;20:61224. [27] Paton D, Wright L. Preparing for bushfires: the public education challenges facing fire agencies. In: Handmer J, Haynes K, editors. Community bushfire safety. Collingwood: CSIRO Publishing; 2008. p. 11726. [28] Prior T, Eriksen C. What does being well prepared for wildfire mean? In: Paton D, Tedim F, editors. Wildfire and community: facilitating preparedness and resilience. Springfield, Illinois, USA: Charles C Thomas Publisher Ltd; 2012. p. 190206. [29] Victorian Country Fire Authority. Prepare your property for bushfire, http://www.cfa.vic.gov.au/firesafety/bushfire/prepare-your-prop erty/prepare-property.htm; 2012 [accessed November 2012]. [30] NSW Rural Fire Service. Prepare. Act. Survive. for bush fire season, http://www.rfs.nsw.gov.au; 2012 [accessed November 2012]. [31] CAL-FIRE. Wildfire is coming. Are you set? http://readyforwildfire. org/; 2012 [accessed November 2012]. [32] Paton D, Tedim F. Wildfire risk management: building on lessons learnt. In: Paton D, Tedim F, editors. Wildfire and community: facilitating preparedness and resilience. Springfield, Illinois, USA: Charles C Thomas Publisher Ltd; 2012. p. 32336. [33] Prior T. Householder bushfire preparation: decision making and the implications of risk. University of Tasmania; 2010 ([Ph.D. thesis]). [34] McFarlane BL, McGee TK, Faulkner H. Complexity of homeowner wildfire risk mitigation: an integration of hazard theories. International Journal of Wildland Fire 2011;20:92131. [35] A. Tibbits, J. Whittaker Stay and defend or leave early: policy problems and experiences during the 2003 Victorian bushfires. Environmental Hazard7(2007):283-290. [36] Bushnell S, Balcombe L, Cottrell A. Community and fire service perceptions of bushfire issues in Tamborine Mountain: what's the dfference? Australian Journal of Emergency Management 2007;22: 39. [37] Keeley JE. Fire intensity, fire severity and burn severity: a brief review and suggested usage. International Journal of Wildland Fire 2009;18:11626. [38] Mell WE, Manzello SL, Maranghides A, Butry D, Rehm RG. The wildlandurban interface fire problem current approaches and research needs. International Journal of Wildland Fire 2010;19: 23851. [39] Cohen JD. Relating flame radiation to home ignition using modeling and experimental crown fires. Canadian Journal of Forest Research 2004;34:161626. [40] Zrate L, Arnaldos J, Casal J. Establishing safety distances for wildland fires. Fire Safety Journal 2008;43:56575. [41] Standards Australia, A.S.3959-2009. Construction of buildings in bushfire prone areas. Sydney, Australia: SAI Global Limited; 129. [42] Noble IR, Bary GAV, Gill AM. McArthurs fire danger meters expressed as equations. Australian Journal of Ecology 1980;5:2014. [43] Tran C, Gilroy J, Williams PR. Quantifying the fuel load and fire hazard in a wet sclerophyll community of southeast Queensland, following invasion by the woody weed lantana camara. In: Proceedings of FIRE2008 international fire congress, 36 September, 2008. Adelaide, Australia; 2008. p. 11. [44] Webster J. Essential bushfire safety tips. 3rd ed.Collingwood, VIC: CSIRO Publishing; 2012. [45] Standards Australia, A.S.1530.4-2005. Methods for fire tests on building materials, components and structures. Part 4: Fireresistance test of elements of construction.Sydney, Australia: SAI Global Limited; 129. [46] Braun E, Cobb D, Cobble V, Krasny J, Peacock R. Measurement of the protective value of apparel fabrics in a fire environment. Journal of Consumer Product Flammability 1980;7:1525.

T.D. Penman et al. / International Journal of Disaster Risk Reduction 6 (2013) 67 77 Sustainable Ecosystems. http://www.bushfirecrc.com/sites/default/ files/managed/resource/ember_attack_spray_research_report_final. pdf. Leonard JE, Blanchi R, White N, Bicknell A, Sargeant A, Reisen F, et al. Research and investigation into the performance of residential boundary fencing systems in bushfires. Melbourne: CSIRO Manufacturing and Infrastructure Technology, Fire Science and Technology Laboratory. Nader G, Nakamura G, Lasaux MD, Quarles S, Valachovic Y. Home landscaping for fire. ANR Publications Report no.: 1601074468. Oakland, CA. Leonard J, Blanchi R, Newnham G, Culvenor D, Siggins A, Opie K. Characterisation of interface fuels literature review. Melbourne: CSIRO Ecosystems Sciences Land and Water Bushfire CRC; 2011. Manzello S, Foote ED. Characterizing firebrand exposure from wildlandurban interface (wui) fires: results from the 2007 Angora Fire. Fire Technology 2012:120. Groenhart L, March A, Holland M. Shifting Victoria's emphasis in land-use planning for bushfire: towards a place-based approach Australian. Journal of Emergency Management 2012;27:337. Bihari M, Ryan R. Influence of social capital on community preparedness for wildfires. Landscape and Urban Planning 2012;106: 25361. Paton D, Buergelt P. Community engagement and wildfire preparedness: the influence of community diversity. In: Paton D, Tedim F, editors. Wildfire and community: facilitating preparedness and resilience. Springfield, Illinois, USA: Charles C Thomas Publisher Ltd; 2012. p. 24159. Paton D, Kelly G, Burgelt PT, Doherty M. Preparing for bushfires: understanding intentions. Disaster Prevention and Management 2006;15:56675. Eriksen C, Gill N, Head. L. The gendered dimensions of bushfire in changing rural landscapes in australia. Journal of Rural Studies 2010;26:33242.

77

[73]

[74]

[75]

[76]

[77]

[78]

[79]

[80]

[81]

[82] Nelson KC, Monroe MC, Johnson JF. The look of the land: homeowner landscape management and wildfire preparedness in minnesota and florida. Society & Natural Resources 2005;18:32136. [83] Daniel T, Weidemann E, Hines D. Assessing public tradeoffs between fire hazard and scenic beauty in the wildlandurban interface. North Central Research Station Report no.: USDA Forest Service General Technical Report NC-231. St. Paul, MN: U.S. Department of Agriculture, Forest Service, 2003. [84] Gordon JS, Matarrita-Cascante D, Stedman RC, Luloff AE. Wildfire perception and community change. Rural Sociology 2010;75: 45577. [85] Jakes P, Kruger L, Monroe MC, Nelson KC, Sturtevant V. Improving wildfire preparedness: lessons from communities across the U.S. Human Ecology Review 2007;14:18897. [86] Handmer J, Haynes K. In: Community bushfire safety.Collingwood, Victoria: CSIRO Publishing; 2008. [87] Cottrell A, Bushnell S, Spillman M, Newton J, Lowe D, Balcombe L. Community perceptions of bushfire risk. In: Handmer J, Haynes K, editors. Community bushfire safety. Collingwood: CSIRO Publishing; 2008. p. 11726. [88] McKee M, Berrens R, Jones M, Helton R, Talberth J. Using experimental economics to examine wildfire insurance and averting decisions in the wildlandurban interface. Society and Natural Resources 2004;17:491508. [89] Winter G, Fried J. Homeowner perspectives on fire hazard, responsibility, and management strategies at the wildlandurban interface. Society and Natural Resources 2000;13:3349. [90] Collins TW. Influences on wildfire hazard exposure in arizona's high country. Society and Natural Resources 2009;22:21129. [91] Beringer J. Community fire safety at the urban/rural interface: the bushfire risk. Fire Safety Journal 2000;35:123. [92] Leonard J, Blanchi R, Leicester RH, Lipkin F, Black J. Profiling urban interface vulnerability. In: Proceedings of fire and material 2009 11th conference, January 2009. San Francisco, USA; 2009. p. 268.