An analysis of generally accepted guidelines for firefighter safe work limitations in association with fire suppression effectiveness, relative to wildland fire intensity is presented. Research in firefighter safety zones and wildland – urban interface defensible space is evaluated in context with fire suppression effectiveness as a method of formulating risk and reward relationships. Fire behaviour prediction systems are incorporated in these analyses and as a tool for evaluating risk and reward relationships in the practice of proactive risk management and a variety of values at risk.
On December 1, 1997 Justice K. Peter Richard released his 750 page public inquiry report on the Westray mine disaster in Plymouth, Nova Scotia with the statement “never let the risks outweigh the benefits”. An explosion of methane gas and coal dust on May 9, 1992 killed 26 miners. In his report titled “The Westray Story – A Predictable Path to Disaster” Justice Richard identified no less than 27 problems contributing to the fatal explosion (Richard 1999). Many of the problems identified were consistent with the disaster incubation period highlighted by Sociologist Barry Turner (1976) in his 6 stage disaster assessment model. A model which has been used in objective analysis of the causal factors contributing to wildland fire disasters (Mutch 1982) and proactive insight into disaster potential (Beaver 2001 unpublished report). Remarkably and regrettably, one can draw many parallels between the Westray mine operation and the wildland fire management business.
The argument of risk and reward management in wildland fire litigation can be traced back at least to the testimony of Henry Thol Sr. in the 1949 Mann Gulch fire (McLean 1992). Mr. Thol was apparently adamant in his claim of negligence towards the U.S. Forest Service for “jumping the crew on a fire in such rough and worthless country and in such abhorrent heat and wind.”
There have been great advances in fire environment/behaviour research since 1949 yet a review of today’s fatalities, injuries and near hits draws into question its operational application.
As fire intensity increases, fire suppression effectiveness decreases while the exposure of the firefighter to health and safety risks increases concurrently. The body of research in this respect is consistent with a review of wildland fires in which serious burn injuries or fatalities have occurred. In terms of risk management and determining a standard of care, the relationship between the two fields of study would appear obvious.
Research into the impacts of fire intensity on fire suppression effectiveness has spanned at least 5 decades (Hirsch 1993) while the impacts on firefighter health and safety have been ongoing since the early 1960’s (Butler and Cohen 1998). It seems however that the two areas of study have been conducted or at least presented, largely independent of each other.
What is presented is an analysis of fire intensity as it relates to the risk of personal injury versus related fire suppression capabilities. The argument is presented that if the fire environment conditions are likely to preclude the firefighter of achieving the fire suppression mission objective (usually containment and control) it might be considered an unsafe or unhealthy work environment. In this context one can only speculate the legal arguments of standard of care and due diligence.
This is not a technical paper on fire behavior or fire behavior prediction systems. As such some liberty is taken in the interest of generalizing SI and imperial measures to flame lengths/heights, fire intensities and radiant heat fluxes.
Making a Stand
Fogarty (1996) was apparently subjected to substantial criticism from his publication relating burning conditions to fire control strategies for two New Zealand wildland – urban interface fires. In his analysis Fogarty pointed out that the fire environment conditions presented by these two fires produced head fire intensities far beyond direct attack capabilities and radiant heat flux levels exceeding the limits of personal protective equipment (PPE). The criticism he received from the fire suppression community was largely aimed at his critical assessment of “Making a Stand” type tactics under the prevailing fire environment conditions.
Making a stand is defined by Fogarty as “a NZ Fire Service equivalent to a direct attack on a head fire, but it usually involves pumper units spaced along the length of line where a fire will cross a track or reach houses”. While not present in any wildland fire management glossary this concept can be found in wildland fire publications, videos, the culture itself, and it is not unique to the New Zealand Fire Service.
While his analysis focused primarily on direct attack and head fire intensities it should be noted that both flank and back fires can produce flames of lethal intensities. All in all it was a good attempt at incorporating the available fire intensity and suppression capability research in a risk and reward management framework.
Experiments were conducted in the Australian, Project Aquarius from 1983 to 1985 to examine the effects of heat load from exertion, weather and fire on firefighters suppressing wildland fires (Budd et al 1997a). Dr. Budd reported that firefighters engaged in direct and parallel attack would maintain a sufficient distance from the flames to avoid painful intensities of radiant heat (>2 kW/m2) on bare skin and usually experienced intensities little more than the intensity of direct sunlight. The highest radiant heat flux recorded was 8.6 kWm2 with a mode of 1.6 kW/m2. Also reported was that while dressed in standard PPE of hardhats, work gloves and cotton coveralls firefighters could stay engaged in direct and parallel attacks up to the limiting degree of fire intensity on suppression success.
An important finding of these experiments was that heat load from exertion is more than twice the combined heat load from the fire and the weather. “Contrary to popular belief, the main task for firefighters’ clothing is not to keep heat out but to let it out (Budd 1997b).”
Fire intensity expressed as kW/m is from Byram (1959) (Equation 1) who defines fire intensity as the rate of heat energy release per unit time per unit length of fire front regardless of its depth or width.
Equation 1. From Byram 1959
I = Fire intensity as kW/m
H = Net low heat of combustion as kj/kg
w = Quantity of fuel burned in the active flame front as kg/m2
r = Linear rate of fire spread as m/sec
Calculations of radiant heat flux are from Equation 2, from Leicester (1985) and mid-flame length calculations for fires in scrub fuels from Thomas (1963), Equation 3.
Equation 2. Radiant heat flux from Leicester 1985 where:
Q = Radiant heat flux as kW/m2
I = Fire intensity as kW/m
D = Distance from vertical heat source
Equation 3. Flame length from Thomas 1963
L = Flame length in metres
I = Fire intensity as kW/m
For the purpose of this analysis a radiant heat flux limit of 2 kW/m2 is used for sustained direct and parallel attack, 5.0 m (16.4 feet) from what is assumed to be a vertical flaming fire edge. In this respect flame length will equal flame height.
By these constants the maximum fire intensity at which a firefighter with standard PPE could sustain direct and/or parallel attack would be approximately 500 kW/m (145 Btu/ft/s) or a flame length/height of approximately 1.7m (5.5 ft). In a reference to fire suppression capability Rothermel (1983) references Roussopoulos and Johnson (1975) for the fire suppression interpretations in Table 1 and provides a graphic display in Figure 1. Alexander and Cole (1995) reference similar fire intensity and fire suppression limitations in Table 2.
Crowning, spotting, and major fire runs are probable.
Control efforts at head of fire are ineffective.
8 – 11
2.4 – 3.4
500 - 1,000
1,730 – 3,460
Fires may present serious control problems – torching out, crowning, and spotting.
Control efforts at the fire head will probably be ineffective.
4 – 8
1.2 – 2.4
100 – 500
350 – 1,730
Fires are too intense for direct attack on the head by persons using handtools.
Hand line cannot be relied upon to hold fire.
Equipment such as dozers, pumpers, and retardant aircraft can be effective.
Fires can generally be attacked at the head or flanks by persons using handtools.
Hand line should hold the fire.
Table 1. From Rothermel 1983.
Figure 1. Fire behavior fire characteristic chart (adapted from Rothermel 1983)
Head Fire Intensity
The situation should be considered as “explosive” or super critical in this class. The characteristics commonly associated with extreme fire behaviour (e.g., rapid rates of spread, continuous crown fire development, medium to long-range spotting, firewhirls, massive convection columns, great walls of flame) is a certainty. Fires present serious control problems as they are virtually impossible to contain until burning conditions ameliorate. Direct attack is rarely possible given the fire’s probable ferocity except immediately after ignition and should only attempted with the utmost caution; an escaped fire should in most cases, be considered a very real possibility. The only effective and safe control action that can be taken until the fire run expires is at the back and along the flanks.
Intermittent crown fires are prevalent and continuous crowning is also possible as well in the lower end of the spectrum. Control is extremely difficult and all efforts at direct control are likely to fail. Direct attack is rarely possible given the fire’s probable ferocity except immediately after ignition and should only be attempted with the utmost caution. Otherwise, any suppression action must be restricted to the flanks and back of the fire. Indirect attack with aerial ignition (I.e., helitorch and/or A.I.D. dispenser), if available, may be effective depending on the fire’s forward rate of advance.
Burning conditions have become critical as intermittent crowning and short range spotting is common place and as a result control is very difficult. Direct attack on the head of a fire by ground forces is feasible for only the first few minutes after ignition has occurred. Otherwise, any attempt to attack the fire’s head should be limited to “medium” or “heavy” helicopters with buckets or fixed-wing aircraft, preferably dropping long-term retardants; control efforts may fail. Until the fire weather severity abates, resulting in the subsidence of a fire run, the uncertainty of successful control exists.
Both moderately and highly vigorous surface fires with flames up to just over 1.5 m (≈ 5 ft) high or intermittent crowning (i.e., torching) can occur. As a result, fires can be moderately difficult to control. Hand-constructed fire guards are likely to be challenged and the opportunity to “hotspot” the perimeter gradually diminishes. Water under pressure (e.g., fire pumps with hose lays) and heavy machinery (e.g., bulldozers, “intermediate” helicopter with a bucket) are generally required for effective action at the fire’s head.
From the standpoint of moisture content, surface fuels are considered sufficiently receptive to sustained ignition and combustion from both flaming and glowing firebrands. Fire activity is limited to creeping or gentle surface burning with maximum flame heights of less than 1.3 m (≈ 4 ft). Control of these fires is fairly easy but can become troublesome as adverse fire impacts can still result, and fires can become costly to suppress if not attended to immediately. Direct manual attack by “hotspotting” around the entire perimeter by firefighters with only hand tools and water from back-pack pumps is possible; a “light” helicopter(s) with bucket is also very effective. Fireguard construction with hand tools should hold.
New fire starts are unlikely to sustain themselves due to moist surface fuel conditions. However, new ignitions may still take place from lightning strikes or near large and prolonged heat sources (e.g., camp fires, windrowed slash piles) but the resulting fires generally do not spread much beyond their point of origin and if they do, control is very easily achieved. Mop-up or complete extinguishment of fires that are already burning may still be required provided there is sufficient fuel and it is dry enough to support smouldering combustion.
Table 2. From Alexander and Cole 1995 Firebreak Breaching The fire suppression limitations identified in Tables 1 and 2 is further supported by research on the probability of a fire breaching a firebreak as a function of fire intensity (Byram 1959, Wilson 1988 and Alexander 2000). Byram (1959) suggests firebreaks of at least 1.5 times the flame length are needed to control breaching in the absence of spotting. For a fire intensity of 500 kW/m (145 Btu/ft/sec) this would require a firebreak of 2.5 m (8.3 ft), a considerable task for firefighters with handtools especially in heavy fuels and/or adverse terrain.
Wilson (1988) and Alexander (2000) included spotting in their assessments of a grass fire breaching a mineralized firebreak as a function of fire intensity and firebreak width. Their assessments included grass fuels with the presence of trees within 20 m (66 ft) upwind of the firebreak and without trees present. For grass fuels with trees present there is still a high probability (> 60%) of a fire of 500 kW/m (145 Btu/ft/sec) intensities breaching a 2.5 m (8.3 ft) firebreak (Table 3).
Firebreak Width (m)
Table 3. Percent probability of a firebreak being breached by a grass fire, trees present within 20 m of firebreak (adapted from Alexander 1994) When assessing the radiant heat flux limitations for firefighters engaged in sustained direct and/or parallel attack in combination with the fire intensity limitations on fire suppression it supports those observations by Budd et al (1997a) and Budd (1997b) previously presented.
It draws into question fire suppression strategies such as presented by Fogarty (1996) that tactically deploy firefighters into fire environment conditions that may preclude their mission success.
Butler and Cohen (1998 and 2000) through a more sophisticated model than that presented in Equation 2 suggest a minimum firefighter safety zone radius of 4 times the maximum flame height. Quantified largely upon experiments conducted at the International Crown Fire Modeling Experiment (ICFME) (Alexander et al 2001) Butler and Cohen concluded that firefighters with PPE consisting of protective head and neck equipment plus Nomex of 210 g/m2, could withstand radiant heat fluxes of up to 7 kW/m2 for durations of less than 90 seconds. At a distance of 5.0 m (16.4 ft) this would translate to a vertical flame length of 4.0 m (13.2 ft) and a corresponding safety zone radius requirement of 16.0 m (52.8 ft).
It must be underscored that this level of radiant heat is survivable only for short periods of time (< 90 sec). It is well in excess of what a firefighter engaged in extended attack could sustain, and beyond direct and/or parallel attack capability with handtools.
Applying this model to the fire suppression interpretations of Tables 1 and 2 and Figure 3, firefighters could be effective in direct and/or parallel attack in fire environment conditions that forecasted a required safety zone radius of no greater than 6.7 m (22.0ft).
Margin of Safety
At a direct/parallel attack distance of 5.0 m (16.4 ft) the firefighter would have a small margin of safety between flame lengths of 1.7 m (5.6 ft), the limit at where suppression operations could be successful, and flame lengths of 4.0 m (13.2 ft) where the firefighter would be either forced off the line or incur burn injuries. Under critically dry fuel conditions it would take only small increases in slope or wind to span this margin in short order.
Mutch (1982) and Thomas (1994) both advocate the need to think the worst case scenario in regards to prescribed fire safety, a notion that could justifiably include wildfire suppression operations as well. However, thinking and planning for the worst case scenario raises an interesting dilemma for the fire manager. If the worst case scenario calls for a safety zone radius of greater than 6.7 m (22.0 ft) is there logic in deploying firefighters into this environment? Should firefighters feel anxious about a fire control plan that calls for a safety zone radius of 10 m (32.8 ft), 20 m (65.6 ft) or 30 m (98.4 ft)? At what degree might it breach the employer’s standard of care?
Should the very size of the constructed safety zones from the 1985 Butte Fire burn-over (Rothermel and Mutch 1986 and Alexander 1990) have raise a red flag to the 73 firefighters who had to deploy fire shelters for their survival? Safety zones, which ultimately became shelter deployment zones were constructed with diameters of 90 m (300 ft) to 120 m (400 ft) (Rothermel and Mutch 1986). Using a mean radius of 53 m (175 ft), Butler and Cohen’s safety zone model would have predicted an adequate safety zone size for flame length/heights of 13 m (44 ft). Eye witness accounts of the fire reported “a wall of flame” 60 m (200 ft) to 90 m (300 ft) high.
Albeit the tactic for this fire was indirect attack and burn-out as opposed to a direct or parallel attack. Nonetheless, the fire intensities associated with flame lengths of even 13 m (44 ft) are still off the scale of suppression success for Tables 1 and 2, and Figure 1.
Defensible Space Using this same principle, what of the National Fire Protection Association (NFPA) standard of 10 m (30 ft) defensible space surrounding structures in the wildland – urban interface? The Butler and Cohen model (without consideration of hazardous smoke emissions, convective heat transfer or the potential showering of embers) would suggest that firefighters could be provided some degree of safety from radiant heat in the defensible space zone providing the flame heights did not exceed 2.5 m (7.5 ft). At these maximum flame heights however the firefighter would be backed against the structure engaging in little more than personal survival.
Albeit fire suppression with engines plus water and foam mixtures can be more effective than firefighters with hand tools alone but the risk and reward relationships still exist. It highlights a troubling debate about the tactical use of defensible space in fire suppression operations in the context of making a stand as previously described by Fogarty (1996). While there is no outright expression of defensible space as providing a tactical advantage for firefighters there are certainly suggestions and depictions of it being so.
Greg Esnouf from Victoria, Australia (personal communication) at a breakout session from the 2000 International Fire Safety Summit was in disbelief that North Americans would entertain the thought that defensible space would be for the tactical benefit of the firefighter. His position was made in no uncertain terms that defensible space is for the sole protection of the structure and no place for firefighters.
In consideration that a wood sided structure is many times more tolerant to radiant heat than firefighters plus infinitely more tolerant of smoke it would be difficult to take issue with Mr. Esnouf’s judgment. This position can be further supported by Jack Cohen’s wildland – urban interface research (Cohen 2000) concluding that high intensity wildfires do not necessarily ignite structures. Mr. Cohen points the blame at the many thousands of embers or pilot ignitions that shower the area in advance of the flaming front as being largely responsible for structure ignitions. All that is required is for one of these potential ignition sources to land in a favorable fuel bed.
One need only review the investigations of near hits, injuries and/or fatalities to understand that fire suppression tactics and firefighter safety considerations have to be made on what a fire is expected to do, not what it has done (Wilson 1977, Ensley 1997 and Mangun 1999). It is the essence of “Situational Awareness” (Putnam 1995).
"Situational awareness is the understanding of what the fire is doing and what you are doing in relation to the fire and your goals. It involves an awareness of fire behavior and terrain and the ability to predict where the fire and you will be in the future (Putnam 1995).” It requires systems for forecasting fire behaviour.
Lawyer Rick Krehbiel (1999) provides the following legal assessment of foreseeablity and its relationship to a plaintiff’s duty and standard of care. “The importance of Fire Weather Indexes and Fire Behavior Prediction Systems as a measure of foreseeability cannot be overstated. If the statistics suggest a blow-up, and appropriate caution is not exercised, those making suppression decisions may find themselves up against a very persuasive argument.”
How many fatalities and or serious injuries the 10 Standard Fire Orders have prevented since their inception in 1957 will never be known with any certainty. One might draw a similarity to the use of canaries in underground mining. A simple but effective tool that saved the lives of many miners and was only displaced as improved technology became available.
As well as these orders may have served the firefighting community in the past, Mike Johns (1996), Assistant U.S. Attorney for the District of Arizona reports the following from the litigation following the 1990 Dude fire fatalities. “Fire managers would shudder at the legal arguments made in the Dude Fire litigation which demonstrate the great amount of discretion which the Standard Fire Orders and Watchouts permit. There is no objective standard against which to measure the risk against the propriety of the action.” What the orders and watchouts equate to are rules of engagement with only subjective or implied rules for non-engagement or disengagement, set in the culture of fire control.
There are however systems such as the Canadian Fire Behaviour Prediction (FBP) System (Forestry Canada Fire Danger Group 1992) and BEHAVE (Andrews 1986) that can be applied in enhancing the level of objectivity involved in risk and reward management. It may be argued however that such systems are imperfect in that there are too many variables that need to be accounted for. Diurnal effects, wind speed increases and direction changes, down draft winds, cold fronts, inversions, smoke, flame tilt angle and convective heat transfer as some examples.
These systems however, have been used in short and long-term strategic and tactical fire management planning for many years. The argument can be made that if they can be used in more effective fire suppression planning they can certainly be used in proactive firefighter safety. The British Columbia Ministry of Forests, as an example, has produced fire intensity related Occupational Safe Work Standards and issues advisories and warnings based upon diurnal forecasts of fire intensity (Beck et al 2001).
Alberta Forest Protection has established a policy that all fire suppression resources must work from an anchor point on any fire where the Head Fire Intensity (HFI) is forecasted to be greater than 2,000 kW/m (578 Btu/ft/sec) (Thorburn and Alexander 2001).
In the application of “wildfire and wisdom” as described by Dr. Karl Weike (1998) “knowledge of a fire should be used not just to fight it, but also to decide how and when to walk away from it.” In the application of risk and reward management, fire environment knowledge needs to be incorporated in proactive deployment decisions.
Should we go back to canaries?
This analysis speaks largely to rewards as it relates to the success of the fire suppression mission objective. The fire management business however is generally not that simplistic. In a single wildland fire there can exist numerous values at risk as well as many wildland fire benefits. In looking to apply risk and reward management the fire manager and firefighters alike have a truly daunting but nonetheless important task.
Norman McLean in his account of the 1949 Mann Gulch fire (1992) cites the backlash as to the propriety of risking and ultimately loosing 13 firefighters lives for what was assessed a negligible resource value. It is more troubling yet in consideration that there was no real property at risk and the only threat to life arrived with the firefighters themselves. But can we say in good conscience that things are different now? Can we say that great advancements have been made in risk and reward management over the past 52 years?
Fire safety workshops, seminars, summits and fatality investigations consistently affirm and reaffirm the sanctity of life in the wildland fire management business. “Trees regrow, houses can be rebuilt, but the loss of a life is forever. What has unfolded in the aftermath (Storm King Mountain) is a reaffirmation that people are first. All else is secondary in wildland firefighting” (Putnam 1995). This is but one of many of quotations that categorically state human life is first and foremost. What is conspicuous by its absence in these reports however is an assessment of the values at risk for which these firefighters risked and ultimately lost their lives. How many of these fatalities might have occurred in wildlands, in fire dependant ecosystems, under high risk and low reward fire environment conditions would be little more than intriguing conjecture at this point.
A comparison of the NFPA 1500 Standard on Fire Department Occupational Safety and Health (NFPA 1992) and the NFPA 295 Standard for Wildfire Control (NFPA 1991) suggests that the health and safety of wildland firefighters does not garner the same importance as does the health and safety of structure firefighters. The NFPA 1500 Standard on Fire Department Occupational Safety and Health endorses the use of risk management using the following principles.
Activities that present a significant risk to the safety of members shall be limited to situations where there is a potential to save endangered lives.
Activities that are routinely employed to protect property shall be recognized as inherent risks to the safety of members, and actions shall be taken to reduce or avoid these risks.
No risk to the safety of members shall be acceptable when there is no possibility to save lives or property.
The NFPA 295 Standard for Wildfire Control lists protective clothing, first aid, aircraft safety plus the 10 Standard Orders and 18 Watchouts, but nothing in terms of risk management as in NFPA 1500.
So why do wildland firefighters continue to be placed, and to place themselves in fire environment conditions for which the risks are high and the rewards low? Pat Withen may have summarized the firefighter culture best in his 1994 article “there is no way to just say no in firefighting that doesn’t carry some formal or informal sanctions (Withen, 1994).
Dr. Gerald Wilde (1997) provides an insightful view of risk and reward in his analysis of the perceptions, motivations and actions leading up to the 1992 Westray mine disaster. Many of the factors cited by Dr Wilde have a remarkable parallel in the wildland fire management culture. Benefits/rewards such as, the pay cheque, promotion, job stability, overtime, and pride of association all contributed to the acceptance of risky working conditions.
Justice Richard’s 1997 report on the same mining disaster was highly critical of the Nova Scotia government’s conflict of interest and how it impacted the mine’s safety program. The Nova Scotia government had invested $100 million of public money in the mine and it had to be kept running. “Instead of acting like an impartial occupational safety and health regulator, willing to impose penalties and costs that might have driven the mine out of business, the Nova Scotia government hesitated, trapped by conflicting agendas.”
It takes little insight or imagination to identify similar conflicts in the fire and forest management business. Consider the conflict that the resource manager struggles with, whose responsibilities include fire suppression strategies and tactics plus timber supply for the local mill.
Admittedly this has been a simplistic analysis that does not account for many fire environment factors and variations of fire suppression tactics. The objective has been to establish that risk and reward relationships exist regardless of the fire environment conditions and at times the perceived benefits simply do not justify the risks.
If the work place environment precludes the achievement of the work place objectives a strong argument can be made that an unsafe and unhealthy condition exists. In applying this test, fire managers must evaluate present and expected safety and health risks against the expected suppression rewards all relative to the values threatened. In such an evaluation, the safety and health of the firefighter must never be subordinated to other values.
Current firefighter PPE provides sufficient protection in suppressing wildland fires up to the intensity for which direct and parallel suppression tactics may be effective. Further armoring the firefighter in this respect serves only to increase their exposure/risk to heat stress without an increase in the fire suppression reward. The argument could also be made that firefighters may stay engaged in futile efforts longer, effectively reducing valuable escape time when disengaging become imminent. It is better to get out 5 minutes too soon than 5 seconds too late. As life’s lessons generally show, it is usually easier to get into trouble than it is to get out of it. There will be times when the conditions of risk and reward prescribe “don’t go”.
The health and safety of the firefighter cannot be left to gut feeling and face saving. The record speaks for itself. Fire behavior prediction systems can be, and need to be incorporated in the application of risk and reward management. As such, there must be conscious commitment to research in the understanding of the fire environment. It is the essence of “situational awareness” and a method of addressing the subjectivity associated with the 10 Standard Fire Orders and Watchouts.
Lastly, “Making a Stand” has no place in fire suppression tactics or vernacular. It needs to be expressed both formally and informally throughout the culture. Times exist when fire suppression technology is simply no match for nature. Wildfire losses are inevitable, the lives of firefighters should not be included in the accounting.
“Never let the risks outweigh the benefits.” Literature Cited Alexander M.E. 1990. The 1985 Butte Fire in central Idaho: A Canadian perspective on the associated burning conditions. Proceedings of an International Symposium. Fire and the Environment: ecological and Cultural Perspectives. Knoxville, Tennessee. 335 – 343.
Alexander M.E. 2000. Fire behaviour as a factor in forest and rural fire suppression. Forest Research, Rotorua, in association with the New Zealand Fire Service Commission and the National Rural Fire Authority, Wellington. Forest Research Bulletin No. 197, Forest and Rural Fire Scientific and Technical Series, Report No. 5: 1 – 28.
Alexander M.E. and F.C. Cole 2000 ‘Predicting and interpreting fire intensities in Alaskan black spruce forests using the Canadian system of fire danger rating. Managing Forests to Meet People’s Needs. Proceedings of Society of American Foresters/Canadian Institute of Forestry Convention. SAF Publication 95-02. (Society of American Foresters [SAF]: Bethesda, MD): 185 – 192.
Alexander M.E., B.J. Stocks, B.M. Wotton, and R.A. Lanoville. 2001. Taking the mystery out of crown fires. Initial Attack 2001(Fall): In Press.
Andrews, P.L. 1986. BEHAVE: fire behavior prediction and fuel modeling system-burn subsystem. Part1. Gen. Tech. Rep. INT-194. U. S. Department of Agriculture, Forest Service, Intermountain Research Station.
Beck, J.A., M.E Alexander, S.D Harvey and A.K. Beaver 2001. Forecasting diurnal variation in fire intensity for use in wildland fire management applications. Proceedings of the Fourth AMS Symposium on Fire and Forest Meteorology. American Meteorology Society. Boston, MA. In Press.
Budd, G.M., J.R. Brootherhood, A.L. Hendrie, N.P. Chaney, and M.P. Dawson. 1997a. Stress, strain, and productivity in wildland firefighters: heat load from exertion, weather, and men suppressing wildland fires. International Association of Wildland Fire. 7: 119 – 131.
Budd, G. 1997b. Heat and the wildland firefighter. Wildfire. 6(8): 14 – 19.
Butler, B.W. and J.D. Cohen. 1998. Firefighter Safety Zones: A Theoretical Model Based on Radiative Heating. International Association of Wildland Fire 8: 73-77.
Butler, B.W. and J.D Cohen. 2000. Field verification of a firefighter safety zone model. Proceedings of the 4th International Fire Safety Summit. International Association of Wildland Fire. Edmonton, Alberta Canada. 54 – 61.
Byram, G.M. 1959. Combustion of forst fuels. K.P. Davis, Ed. Forest Fire Control and Use, McGraw-Hill, New York, NY, 61 – 89.
Cohen J.D. 2000. Preventing disaster: home ignitability in the wildland – urban interface. Journal of Forestry 98(3): 15 – 21.
Ensley, P. (ed.). 1997. Historic wildland firefighter fatalities 1910 – 1996. National Wildfire Coordinating Group. U.S Department of Agriculture, Forest Service. National Interagency Fire Center. Boise, ID. 1 – 19.
Fogarty, L.G., 1996: Two rural/urban interface fires in the Wellington Suburb of Karori: assessment of associated burning conditions and fire control strategies. For. Res. Bull. No. 197, For. Rural Fire Sci. Tech. Ser. Rep. No. 1. N.Z. For. Res. Instit., Rotorua in assoc. N.Z. Fire Serv. Comm. and Natl. Rural Fire Authority, Wellington, New Zealand.
Forestry Canada Fire Danger Group, 1992. Development and structure of the Canadian Forest Fire Behavior Prediction System. Inf. Rep. ST-X-3. Forestry Canada, Ottawa, Ont.
Hirsch. K.G. 1996. Initial attack fire crew effectiveness in the boreal forest region of western and central Canada. Masters thesis. Department of Forestry. University of Toronto. Ont. 7 – 35.
Johns, M. 1996. Dude fire still smokin’. Wildfire 5(2): 39 – 42.
Krehbiel, R. 1999. Safety concerns and operational strategies: Legal implications. Unit V. Wildland Fire Behavior Specialist Training Course: 1999 Edition. Environmental Training Centre (ETC), Hinton, AB and Canadian Interagency Forest Fire Centre, Winnipeg, MB. 5.
Leicester, R.H. 1985. Building technology to resist fire, flood, and drought. In proceedings of the Ninth Invitation Symposium. Australia Academy of Techincal Science. 221 – 236.
Mangun R.J. 1999. Wildland fire fatalities in the United States 1990 – 1998. U.S. Department of Agriculture, Forest Service, Missoula Technology & Development Center. Technical Report 9951-2808-MTDC Missoula, MT.
Maclean, N. 1992. Young men and fire. The University of Chicago Press Ltd., London. 149.
Mutch, Robert W. 1982. The safety element of a prescribed burn plan. Lesson plan in Prescribed Fire Management. National Advanced Resource Technology Centre. Marana, Arizona.
National Fire Protection Association. 1991. NFPA 299 standard for protection of life and property from wildfire. 1991 ed. Natl. Fire Protection Assoc., Quincy, Massachusetts. 13 – 15.
National Fire Protection Association. 1991. NFPA 295 standard for wildfire control. 1991 ed. Natl. Fire Protection Assoc., Quincy, Massachusetts. 5 – 14.
National Fire Protection Association. 1992. NFPA 1500 standard on fire department occupational safety and health program. 1992 ed. Natl. Fire Protection Assoc., New Orleans, Louisiana. 21.
Putnam, T. 1995. editor. Findings from the wildland firefighters human factors workshop. U.S. Department of Agriculture, Forest Service. 9951-2855-MTDC: 2 – 4.
Richard, K.P. 1999. The Westray mine explosion – aftermath: did we learn anything from it?. http://www.iwh.on.ca/Pages/99/Papers/P_1B6_1.html.
Rothermel, R.C. 1983. How to predict the spread and intensity of forest and range fires. U.S. Department of Agriculture, Forest Service, Gen. Tech. Report INT – 143: 59 – 62.
Rothermel, R.C., and R.W. Mutch. 1986. Behavior of the life threatening Butte Fire: August 27 – 29, 1985. Fire Management Notes. 47: 14 – 24.
Thomas, D. 1994. Emerging safety issues in ecosystem burning. Proceedings of the Interior West Fire Council Meeting. Coeur d'Alene, Idaho.
Thomas, P.H. 1963. The size of flames from natural fires. Proceedings of the Ninth International Symposium on Combustion. Academic Press. New York. N.Y. 844 – 859.
Thorburn W.R. and M.E. Alexander. 2001. Laces versus lces: adopt and “A” for “anchor points” to improve wildland firefighter safety. Proceedings of the 5th International Fire Safety Summit. International Association of Wildland Fire. Missoula, MT. In Press.
Turner, B.A. 1976. The development of disasters – a sequence model for the origin of disasters. Sociological Review: 753 – 774.
Weick, K.E. 1998. Wildfire and wisdom. Wildfire, 7(1): 14 – 19.
Wilde, G.J.S. 1997. Risk awareness and risk acceptance at the Westray coal mine: an attempt to understand miners’ perception, motivations and actions prior to the accident. http://pavlov.psyc.queensu.ca/faculty/wilde/westray.html.
Wilson A.A.G. 1988. Width of fire break that is necessary to stop grass fires: some field experiments. Canadian Journal of Forest Research 18. 682 – 687.
Wilson, Carl C. 1977. Fatal and near-fatal forest fires, the common denominators. International Fire Chief 43(9): 9 – 15.
Withen, P. 1994. Fire culture. Inner Voice, September/October: 12 – 13.