FIREFIGHTER SAFETY ZONES: HOW BIG IS BIG ENOUGH?

Bret W. Butler and Jack D. Cohen

ll wildland firefighters Greene 1959; Behnke 1982). Stud- working on or near the A safety zone should be ies by Braun and others (1980) A fireline must be able to large enough so that suggest that when a single layer of identify a safety zone. Further- the distance between 6.3 oz/yd2 (210 g/m2) Nomex cloth more, they need to know how is worn, second degree burns will “big” is “big enough.” the firefighters and occur after 90 seconds when a flames is at least four firefighter is subjected to radiant Beighley (1995) defined a safety times the maximum fluxes greater than 0.6 Btu/ft2/s zone as “an area distinguished by flame height. (7 kW/m2). characteristics that provide free- dom from danger, risk, or injury.” The Nomex shirts and trousers The National Wildfire Coordinating relationship between convective currently used by wildland Group proposed that a safety zone heating and safety zone size in firefighters have fabric weights of be defined as “a preplanned area of future work. 5.7 and 8.5 oz/yd2 (190 and 280 sufficient size and suitable location g/m2), respectively. Few studies, that is expected to prevent injury What Do We Know? however, have explored relation- to fire personnel from known haz- ships between flame height and the Two questions are important when ards without using fire shelters” safety zone size necessary to specifying safety zone size: 1) What (USDA/USDI 1995). prevent burn injury. is the radiant energy distribution in front of a flame? and 2) How In our study of wildland firefighter much heat can humans endure Theory Versus Reality safety zones, we focused on radiant before injury occurs? Concerning We formulated a theoretical model heating only. In “real” wildland the first question, Fogarty (1996) to predict the net radiant energy fires, convective energy transport and Tassios and Packham (1984) arriving at the firefighter wearing in the form of gusts, fire whirls, or related the energy received by a Nomex clothing as a function of turbulence could contribute sig- firefighter to fireline intensity and flame height and distance from the nificantly to the total energy distance from the flame front. flame (Butler and Cohen [In received by a firefighter. However, Green and Schimke (1971) pre- press]). Figure 1 displays the convection is subject to buoyant sented very specific information results. forces and turbulent mixing, both about fuel break construction on of which suggest that convective slopes and ridges in the Sierra The amount of radiant energy heating is important only when a Nevada mixed-conifer forest type. arriving at the firefighter depends firefighter is relatively close to the Others have discussed the perfor- both on the distance between the fire. One reason that firefighters in mance of fire shelters under differ- firefighter and the flame and on potential entrapment situations ent heating regimes (for example, the flame height. The information are told to lie face down on the King and Walker 1964; Jukkala and shown suggests that in most cases ground is to minimize their expo- Putnam 1986; Knight 1988). As safety zones must be relatively sure to convective heating. We one would expect, there is not large to prevent burn injury. hope to define more clearly the much information related to the second question. The available We compared safety zone sizes pre- Bret Butler and Jack Cohen are research information suggests that 0.2 Btu/ dicted by our model against those scientists in the Fire Behavior Research 2/ 2 Unit, Rocky Mountain Research Station, ft s (2.3 kW/m ) is the upper limit reported on four wildfires: the Intermountain Fire Sciences Laboratory, that can be sustained without Missoula, MT. injury for a short time (Stoll and Continued on page 14

Volume 58 •Ê No. 1 • 1998 13 designated safety zone. When the Distance from flame (ft) fire overran them, they were lying 650 face down on the ground without 600 fire shelters in a 25-foot- (8-m-) wide clearing near the top of a 550 No injury ridge. Tragically, only one of the 500 Burn injury probable 0.5 four survived, and he suffered 450 severe burns over most of his body. 400 Figure 1 suggests that for this fire, Burn injury limit 2 350 1 Btu/ft /s the safety zone should have been 300 2 large enough to separate fire- 250 fighters from flames by 150 feet 200 (46 m). Clearly, the 25-foot- (8-m-) 150 wide clearing did not qualify as a 100 safety zone. 50 5 0 Flame heights were reported to be 0 50 100 150 200 250 300 350 200 to 300 feet (62 to 92 m) high Flame height (ft) Mann South on the Butte Fire that burned on Canyon Gulch Butte Fire steep slopes covered with mature and Fire Battlement lodgepole pine and Douglas-fir Creek Fires during August of 1985 (Mutch and Rule-of-thumb = 4 x maximum flame height Rothermel 1986). Figure 1 indi- 2 Burn injury limitÐ0.6 Btu/ft /s cates that a cleared area greater than 1,200 feet (370 m) across would have been needed to prevent Figure 1—Lines represent predicted radiant energy arriving at the firefighter as a function of flame height and distance from the flame. It is assumed that the firefighter is injury to the firefighters standing wearing fire-retardant clothing and protective head and neck equipment. The heavy in its center. In fact, safety zones shaded line represents the burn injury threshold of 0.6 Btu/ft2/s (7 kW/m2). The heavy solid 300 to 400 feet (92 to 123 m) in black line indicates the rule of thumb for the size of the safety zone. diameter were prepared (Mutch and Rothermel 1986). This Mann Gulch Fire, the Battlement the burned area, with its width diameter was not sufficiently large Creek Fire, the Butte Fire, and the approximately half the length, the enough to meet the definition of South Canyon Fire. safety zone created by Dodge’s a safety zone, as indicated by the escaped fire would have been about fact that 73 firefighters had to The Mann Gulch Fire overran 16 150 feet (46 m) wide. Figure 1 deploy in fire shelters to escape the firefighters on August 5, 1949. Wag indicates that the safety zone radiant heat. As the fire burned Dodge, one of only three survivors, needed to be large enough to sepa- around the edges of the deploy- lit a fire and then lay face down in rate the firefighters and flames by ment zone, the intense heat forced the burned-out area as the main 90 to 150 feet (27 to 46 m) or the firefighters to crawl while fire burned around him. The Mann approximately the same width as inside their shelters to the opposite Gulch Fire occurred in an open the area created by Dodge’s fire. side of the clearing. stand of scattered, mature pon- derosa pine (60 to 100+ years old) The Battlement Creek Fire On July 2, 1994, the South Canyon with a grass understory. Flame occurred in western Colorado dur- Fire was ignited by a lightning heights of 10 to 40 feet (3 to 12 m) ing July of 1976 (USDI 1976). The strike to a ridgetop in western were estimated to have occurred at fire burned on steep slopes covered Colorado. During the afternoon of the time of entrapment. Rothermel with 6- to 12-foot- (2- to 4-m-) July 6, the South Canyon Fire (1993) indicates that Dodge’s fire high Gambel oak. Flames were “blew up,” burning across the pre- burned about 300 feet (92 m) estimated at 20 to 30 feet (6 to dominately Gambel-oak-covered before the main fire overran it. 9 m) above the canopy. Four slopes with 50- to 90-foot- (15- to Assuming an elliptical shape for firefighters were cut off from their 28-m-) tall flames (South Canyon

14 Fire Management Notes Fire Accident Investigation Team the safety zone. In such fires, the ever, this model does not account 1994). Tragically, 14 firefighters separation distance suggested in for convective heating that could were overrun by the fire and died figure 1 is the radius of the safety significantly increase the total while attempting to deploy their zone, meaning the safety zone energy transfer to shelters fire shelters. Twelve of the diameter should be twice the value deployed within a few flame firefighters died along a 10- to indicated. lengths of the fire. 12-foot- (3- to 4-m-) wide fireline on a 55-percent slope, the other What About Fire Conclusions two in a steep narrow gully. Eight Shelters? Radiant energy travels in the same other firefighters deployed their We calculated the net radiant form as visible light, that is, in the fire shelters in a burned out area energy transferred through a fire line of sight. Therefore, locating approximately 150 feet (46 m) shelter like those used by fire- safety zones in areas that minimize wide. They remained in their shel- fighters in the USDA Forest firefighters’ exposure to flames will ters during three separate crown Service. The fire shelter is based on reduce the required safety zone fire runs that occurred 450 feet the concept that the surface will size. For example, topographical (138 m) away from them; none of reflect the majority of the incom- features that act as radiative these eight firefighters was injured ing radiant energy. An average shields are the lee side of rocky (Petrilli 1996). One firefighter esti- emissivity for the aluminum-foil outcroppings, ridges and the tops mates that air temperatures inside exterior of a fire shelter is 0.07, of ridges, or peaks containing little ° ° the shelters reached 115 F (46 C) indicating that approximately or no flammable vegetation. Safety and remembers smoke and glow- 93 percent of the energy incident zone size is proportional to flame ing embers entering the fire shel- on a fire shelter is reflected away height. Therefore, any feature or ters during the crown fire runs. (Putnam 1991). Model predictions action that reduces flame height Survivors felt they were far enough shown in figure 2 suggest that heat will have a corresponding effect on from the flames that survival with levels remain below the injury the required safety zone size. Some minor injuries would have been limits for deployment zones wider examples are burnout operations possible without the protection of than 50 feet (15 m), even with that leave large “black” areas, thin- a fire shelter (Petrilli 1996). A 300-foot- (92-m-) tall flames. How- ning operations that reduce fuel firefighter who did not deploy in a shelter but remained on a narrow ridge below the eight firefighters Distance from flame (ft) during the “blowup” experienced 650 no injuries (South Canyon Fire 600 Burn injury limitÐ0.6 Btu/ft2/s Accident Investigation Team 1994). 550 Figure 1 suggests that in this situ- 2 500 0.05 Btu/ft /s ation, the safety zone must be large enough to separate the 450 firefighters and flames by 250 to 400 0.1 350 feet (77 to 115 m). 350 300 A general rule of thumb can be 250 derived from figure 1 by approxi- 200 mating the injury limit with a 150 0.2 straight line. After doing so, it 100 0.3 appears that a safety zone should 50 0.6 be large enough that the distance 0 between the firefighters and flames 0 50 100 150 200 250 300 350 Flame height (ft) is at least four times the maximum flame height. In some instances— Figure 2—Predicted radiant energy on a fire shelter as a function of distance between the such as the Mann Gulch, Battle- fire shelter and flames, and flame height. The heavy shaded line represents the burn ment Creek, and Butte fires—the injury threshold for a firefighter inside a deployed fire shelter. fire may burn completely around Continued on page 16

Volume 58 •Ê No. 1 • 1998 15 load, and retardant drops that Beighley, Mark. 1995. Beyond the safety Petrilli, A. May 21, 1996. [Personal com- decrease flame temperatures. zone: Creating a margin of safety. Fire munication with B. Butler]. Missoula, Management Notes. 55(4): 22-24. MT. Braun, E.; Cobb, D.; Cobble, V.B.; Krasny, Putnam, T. 1991. Your fire shelter: Pub. We emphasize that while this study J.F.; Peacock, R.D. 1980. Measurement NFES 1570. Boise, ID: National Inter- addresses the effects of radiant of the protective value of apparel fabrics agency Fire Center, National Wildfire in a fire environment. Journal of Con- Coordinating Group, National Fire energy transfer, convection is not sumer Product Flammability. 7: 15-25. Equipment System, 18 p. addressed. Convective energy Butler, B.W.; Cohen, J.D. [In press]. Rothermel, R.C. 1993. Mann Gulch Fire: A transfer from gusts, fire whirls, or Firefighter safety zones: A theoretical race that couldn’t be won. GTR INT-299. model. International Journal of Ogden, UT: U.S. Department of Agricul- turbulence could significantly Wildland Fire. ture, Forest Service, Intermountain increase the total heat transfer to Fogarty, L.G. 1996. Two rural/urban inter- Research Station. 10 p. the firefighter and thus the face fires in the Wellington suburb of South Canyon Fire Accident Investigation required safety zone size. Further Karori: Assessment of associated burn- Team. 1994. Report of the South Can- ing conditions and fire control strate- yon Fire Accident Investigation Team. work in this area is needed. gies. FRI Bulletin No. 197, Forest and Atlanta, GA: U.S. Department of Agricul- Rural Fire Scientific and Technical ture Forest Service, Southern Region. Series, Rep. No. 1. Rotorua and 39 p. plus appendices. Acknowledgments Wellington, NZ: New Zealand Forest Stoll, Alice M.; Greene, Leon C. 1959. Rela- The United States Department of Research Institute in association with tionship between pain and tissue dam- the Interior’s Fire Coordinating the National Rural Fire Authority. 16 p. age due to thermal radiation. Journal of Green, L.R.; Schimke, H.E. 1971. Guides Applied Physiology. 14(3): 373-382. Committee, Boise, ID, provided for fuel-breaks in the Sierra Nevada Tassios, S.; Packham, D. 1984. An investi- financial assistance for a portion mixed-conifer type. Berkeley, CA: U.S. gation of some thermal properties of of this study. Ted Putnam of the Department of Agriculture, Forest four fabrics suitable for use in rural Service, Pacific Southwest Forest and firefighting. National Center for Rural Forest Service’s Missoula Tech- Range Experiment Station. 14 p. Fire Research. Tech. pap. no. 1. nology and Development Center, Jukkala, A.; Putnam, T. 1986. Forest fire Canberra, ACT, Australia: Forest Missoula, MT, provided valuable shelter saves lives. Fire Management Research Institute, Forestry and Timber information and advice on the Notes. 47(2): 3-5. Bureau. 13 p. King, A.R.; Walker, I.S. 1964. Protection U.S. Department of the Interior. [Unpub- effects of heat on human tissue. of forest firefighters. Unasylva. lished July 17, 1976, report]. Accident 18(1): 29-32. report of Battlement Creek Fire fatali- Knight, Ian. 1988. What intensity of ties and injury. U.S. Department of the Literature Cited fire can a fire fighter survive in a Interior. 125 p. Behnke, W.P. 1982. Predicting flash fire reflective shelter? Fire Technology. USDA/USDI. [Review copy, 10/31/95]. Glos- protection of clothing from laboratory 24(4): 312-332. sary of wildland fire terminology. Boise, tests using second degree burn to rate Mutch, R.W.; Rothermel, R.C. 1986. 73 ID: National Interagency Fire Center, performance. London: International firefighters survive in shelters. Fire National Fire and Aviation Support Conference on Flammability; 30 p. Command. 53(3): 30-32, 48. Group. 160 p. n

16 Fire Management Notes