Fire Management today Volume 64 • No. 1 • Winter 2004
FORECASTING WILDLAND FIRE BEHAVIOR: AIDS AND GUIDES
United States Department of Agriculture Forest Service Editor’s note: This issue of Fire Management Today is the third in a three-part series of reprinted articles related to wildland fire behavior, some of them decades old. Although the articles appear in today’s format, the text is reprinted largely verbatim and therefore reflects the style and usage of the time. We made minor wording changes for clarity and added metric conversions where needed. All illustrations are taken from the original articles.
Fire Management Today is published by the Forest Service of the U.S. Department of Agriculture, Washington, DC. The Secretary of Agriculture has determined that the publication of this periodical is necessary in the transaction of the public business required by law of this Department.
Fire Management Today is for sale by the Superintendent of Documents, U.S. Government Printing Office, at: Internet: bookstore.gpo.gov Phone: 202-512-1800 Fax: 202-512-2250 Mail: Stop SSOP, Washington, DC 20402-0001
Fire Management Today is available on the World Wide Web at
Ann M. Veneman, Secretary April J. Baily U.S. Department of Agriculture General Manager
Dale Bosworth, Chief Robert H. “Hutch” Brown, Ph.D. Forest Service Managing Editor
Jerry Williams, Director Madelyn Dillon Fire and Aviation Management Editor
Carol LoSapio Guest Editor
Martin E. Alexander and Dave Thomas Issue Coordinators
The U.S. Department of Agriculture (USDA) prohibits discrimination in all its programs and activities on the basis of race, color, national origin, sex, religion, age, disability, political beliefs, sexual orientation, or marital or family sta tus. (Not all prohibited bases apply to all programs.) Persons with disabilities who require alternative means for communication of program information (Braille, large print, audiotape, etc.) should contact USDA’s TARGET Center at (202) 720-2600 (voice and TDD).
To file a complaint of discrimination, write USDA, Director, Office of Civil Rights, Room 326-W, Whitten Building, 1400 Independence Avenue, SW, Washington, DC 20250-9410 or call (202) 720-5964 (voice and TDD). USDA is an equal opportunity provider and employer.
Disclaimer: The use of trade, firm, or corporation names in this publication is for the information and convenience of the reader. Such use does not constitute an official endorsement of any product or service by the U.S. Department of Agriculture. Individual authors are responsible for the technical accuracy of the material presented in Fire Management Today. Fire today Management Volume 64 • No. 1 • Winter 2004
On the Cover: CONTENTS Forecasting Wildland Fire Behavior: Aids, Guides, and Knowledge-Based Protocols ...... 4 M.E. Alexander and D.A. Thomas Forest Fires and Sea Breezes...... 12 G.L. Hayes Fundamentals of Fire Behavior ...... 15 H.T. Gisborne Vertical Wind Currents and Fire Behavior...... 24 John S. Crosby Warning Signs for Fire Fighters ...... 27 A.A. Brown Recognizing Weather Conditions That Affect Forest Fire Behavior ...... 29 Owen P. Cramer Meteorological Problems Associated With Mass Fires ...... 34 Aids and guides from the past, DeVer Colson some illustrated here, can Some Principles of Combustion and Their Significance in Forest Fire Behavior ...... 37 help improve the fire behavior George M. Byram forecasting capabilities needed Vortex Turbulence—Its Effect on Fire Behavior...... 45 today in both fire use and fire James B. Davis and Craig C. Chandler The Concept of Fire Environment ...... 49 suppression. See the articles C.M. Countryman in this issue for descriptions. Get the Most From Your Windspeed Observation...... 53 John S. Crosby and Craig C. Chandler Atmospheric Stability Forecast and Fire Control ...... 56 Rollo T. Davis Downbursts and Wildland Fires: A Dangerous Combination ...... 59 Donald A. Haines Estimating Slope for Predicting Fire Behavior ...... 62 Patricia L. Andrews Air Tanker Vortex Turbulence—Revisited ...... 64 Donald A. Haines A Trend Analysis of Fireline “Watch Out” Situations in Seven Fire-Suppression Fatality Accidents ...... 66 Gene A. Morse The FIRE 21 symbol (shown below and on the LCES—A Key to Safety in the Wildland Fire Environment ...... 70 cover) stands for the safe and effective use of Paul Gleason wildland fire, now and throughout the 21st How IC’s Can Get Maximum Use of Weather Information ...... 72 century. Its shape represents the fire triangle Christopher J. Cuoco and James K. Barnett (oxygen, heat, and fuel). The three outer red triangles represent the basic functions of Beyond the Safey Zone: Creating a Margin of Safety ...... 78 wildland fire organizations (planning, operations, Mark Beighley and aviation management), and the three critical Firefighter Safety Zones: How Big Is Big Enough? ...... 82 aspects of wildland fire management (prevention, Bret W. Butler and Jack D. Cohen suppression, and prescription). The black interior represents land affected by fire; the emerging Safety Alert: Watch Out for Aircraft Turbulence! ...... 86 green points symbolize the growth, restoration, Billy Bennett and sustainability associated with fire-adapted The Consumption Strategy: Increasing Safety During Mopup . . . . . 88 ecosystems. The flame represents fire itself as an Tom Leuschen and Ken Frederick ever-present force in nature. For more informa tion on FIRE 21 and the science, research, and Author Index—Volume 63 ...... 94 innovative thinking behind it, contact Mike Apicello, National Interagency Fire Center, Subject Index—Volume 63 ...... 95 208-387-5460. SHORT FEATURES Websites on Fire ...... 28 The Blowup Fire and Firefighter Safety...... 33 Wildland Fire Research’s Raison D’etre ...... 92 Guidelines for Contributors ...... 93 Firefighter and public safety is Photo Contest Announcement ...... 98 First of Its Kind: A Historical Perspective on our first priority. Wildland Fire Behavior Training ...... 99
Volume 64 • No. 1 • Winter 2004 3 FORECASTING WILDLAND FIRE BEHAVIOR: AIDS, GUIDES, AND KNOWLEDGE-BASED PROTOCOLS
M.E. Alexander and D.A. Thomas
an wildland fire behavior really be predicted? That depends on By systematically reflecting upon our fire behavior C how accurate you expect the forecasts and the tools that helped us prepare prediction to be. The minute-by them, we become the masters of fire behavior minute movement of a fire will probably never be predictable—cer models and not their servants. tainly not from weather conditions forecasted many hours before the This issue is devoted to aids, fire. Nevertheless, practice and The Practice of guides, and knowledge-based proto experienced judgment in assessing Predicting Wildland cols involved in predicting wildland the fire environment, coupled with Fire Behavior fire behavior for safe and effective a systematic method of calculating More than 50 years ago, Barrows fire suppression (Alexander 2000). fire behavior, yield surprisingly (1951) outlined the basic concepts It includes 21 articles published good results (Rothermel 1983). of predicting or forecasting wild- from 1947 to 1998. A recent article land fire behavior that are still very by Weick (2002) that emphasizes This is the third and final special valid today (see the excerpt on the importance of human factors in issue of Fire Management Today in pages 6–7). As figure 1 shows, the the field of fire behavior forecasting a series of issues devoted to the process of judging fire behavior can could have easily been included. subject of wildland fire behavior. be divided into five simple steps: The first two issues contained 36 articles dealing with wildland fire behavior case studies and analyses published in Fire Management Today and its predecessors between 1937 and 2000. These two issues contained lead articles on various aspects of those subjects (Alexander and Thomas 2003a, 2003b). Not included in these two issues are two recent articles on fire behavior published in Fire Management Today (Brown 2002; Cornwall 2003).
Marty Alexander is a senior fire behavior research officer with the Canadian Forest Service at the Northern Forestry Centre, Edmonton, Alberta; and Dave Thomas is the regional fuels specialist for the USDA Figure 1—Judging fire behavior requires systematic analysis of many factors Forest Service, Intermountain Region, (from Barrows 1951). Ogden, UT.
Fire Management Today 4 1. Basic knowledge. The founda- We recommend that fire behavior analysts adopt tion for judging probable fire the framework of the After Action Review, as behavior must rest on basic knowledge of the principles of described on the Wildland Fire Lessons Learned combustion: What is necessary Center Website. for combustion to occur? What causes the rate of combustion to increase or decrease? How may ior predictions to actual fire behav by putting their fire behavior fore combustion be reduced or ior. This is the only way one can casts through a reflective scrutiny stopped? truly meet Barrows’ (1951) advice based on four basic questions: 2. Forest knowledge. Three basic to “evaluate the combined effects of factors in a forest area—weather, all significant factors influencing 1. What did your fire behavior fore topography, and fuels—are fire behavior.” cast say would happen? important indicators of fire 2. What actually happened? behavior. Conscious reflection, not as an 3. Why did the fire behavior aid, 3. Aids and guides. Several aids afterthought but as a normal rou guide, or protocol predict accu and guides are available to assist tine in the day-to-day business of rately (or inaccurately)? in evaluating weather, topogra fire behavior forecasting, involves a 4. Finally (and most importantly), phy, and fuels. highly professional method of ques if you had to make this forecast 4. Estimate of situation. The prob tioning whether our fire behavior again, what would you do differ abilities for various patterns of aids, guides, and protocols are ently? How would you change fire behavior are systematically working. By systematically reflect the way you used the aid, guide, explored through an estimate of ing upon our fire behavior forecasts protocol, or model/system in this the situation based upon the and the tools that helped us pre different approach? combined effects of weather, pare them, we become the masters fuels, and topography. of fire behavior models and not Judging the quality of a fire behav 5. Decision. The end product of their servants. ior prediction or forecast solely on the fire behavior analysis is a the outcome can be hazardous. By decision outlining when, where, To paraphrase Dr. Karl Weick chance, good predictions or fore and how to control the fire and (2003)—coauthor of Managing the casts can sometimes have bad out spelling out any special safety Unexpected: Assuring High comes and bad predictions or fore measures required. Performance in an Age of casts can result in good outcomes Complexity (Weick and Sutcliffe (fig. 2). From a practical stand For this third and final issue in the 2001)*—becoming a mindful FBAN point, overpredictions can be easily series dealing with wildland fire is a constant struggle for alertness, readjusted without serious, lasting behavior, we chose articles from and to be alert means to “constant consequences, whereas underpre past issues that reflect the various ly and diligently seek instances dictions can be disastrous (table 1) elements involved in Barrows’ where your model didn’t work and from the standpoint of human safe (1951) process of judging or pre identify indicators you missed that ty (i.e., for the public and for fire- dicting wildland fire behavior. signaled expectations weren’t being filled….” Outcome Comparisons of Fire We recommend that FBANs and Good Bad Behavior Predictions others adopt the framework of the and Forecasts Needed After Action Review, as described on Good Objective Unlucky After 50 years, the only item we the Wildland Fire Lessons Learned would add to Barrows’ (1951) out- Center Website (
Volume 64 • No. 1 • Winter 2004 5 On the Place of Fire Behavior in Wildland Fire Management* Although forestry dates back forces to judge where and when made in advance, the unexpected hundreds of years, organized for fires will start and how they will may become expected and a poten est fire research has been under behave once ignition takes place. tial blow or run may often be antic way less than 30 years. During Every member of the firefighting ipated soon enough to be prevent much of this time the major team from ranger to smokechaser ed. Effective fire control requires efforts have been devoted to stud must be able to make reliable esti that suppression plans and action ies of fire behavior or closely mates of the behavior of fires burn be carried out in accordance with allied fields. As a result, much ing under a wide variety of condi continuing estimates and forecasts has been learned about how fires tions. These estimates must be of what the fire is going to do. act, in spite of the relatively short good enough to provide the basis Analysis of fire behavior is a basic period of organized effort. for decisions which will lead to fast, requirement in firefighting applica Knowledge stemming from any efficient, and safe firefighting. ble equally to the one-man smoke- research projects, plus the experi chaser or the big fire where hun ence gained from the control of Fire Behavior and dreds of men are in action. thousands of fires, provide a good Suppression Methods foundation for a general under The character and difficulty of the Fire Behavior and standing of the complex subject. suppression job on every fire Safety depends largely upon the behavior An important reason for under The main purpose of this publica of the fire. The speed, strength, and standing fire behavior is to provide tion is to summarize the most type of attack are governed by the safety for the firefighters. Every fire important aspects of fire behavior location of the fire and its reaction behavior situation calls for specific as we now know them. The to the surrounding environment. safety measures. Experience gained author recognizes that there are Each change in environment may from fighting thousands of fires has still many unknowns in the change fire behavior and in turn shown that the suppression job behavior of forest and range fires. call for some adjustment in fire may be accomplished with a rea These unknowns will be the tar fighting strategy and techniques. sonable degree of safety. To achieve gets of future research. In the The ability of the man handling the safety it is highly important that all meantime it is important that the suppression job to evaluate the firefighters have a general knowl best available information on fire behavior pattern largely determines edge and the leaders of the fire behavior be placed in the hands the efficiency and economy of the fighting forces have a high degree of the men who must carry on entire firefighting operation. of knowledge of fire behavior. the vital task of fire control … A primary purpose of evaluating the The most dangerous individual in a Knowledge of fire behavior is an behavior of every fire is to reduce suppression organization is usually essential requirement for fire or prevent unexpected “blowups the man who is afraid of fire. Fear is fighters. Successful fire control and runs.” A careful check on largely a result of ignorance. Many operations depend, first of all, everything that will affect the risks can be eliminated from fire upon the ability of the protection behavior of a fire reduces the fighting if each man knows what to chances for the “unexpected.” expect the fire to do. The average *From Barrows (1951) handbook Fire Behavior in Northern Rocky Mountain Forests. When a skilled size-up has been firefighter need not be an expert on
Fire Management Today 6 all phases of fire behavior, but he Judging Fire Behavior and the meaning of the most sig should have a working knowledge of Many complex factors influence the nificant of the many factors that ignition, combustion, and rate of ignition, rate of spread, and general may be observed and to present a spread of fires burning in forest behavior of fires. Some of these fac method of evaluating their com fuels. Equipped with such basic fire tors can be measured more or less bined effects. behavior “know-how” the individual precisely with instruments. Others firefighter can approach his job do not lend themselves to exact Fire Safety Measures without fear and with confidence measurements and therefore must A thorough understanding of fire that he can perform required duties be evaluated in general terms. The behavior is essential to the pro in a safe and efficient manner. combined effects of all factors, motion of safety in firefighting whether measured precisely or not, operations. Accidents often occur Fire Behavior and the determine the behavior of a fire. No when so-called “unexpected fire Forest Manager single factor, such as wind, steep behavior” develops. To avoid In the northern Rocky Mountains ness of slope, or kind of fuel, will these “unexpected events,” the fires influence many phases of the provide the answer to questions of first and most important safety forest management job. The behav when and where fires will start and measure on every fire, regardless ior of fires is an important factor in how fast they will spread. Likewise, of size, is to make the estimate of the growth, harvesting, and regen no single instrument or meter will the fire behavior situation…. eration of forest crops. How often answer these fundamental ques Fires behave according to certain fires occur and how hot they burn tions. Therefore it is necessary for laws. Runaway fires do not just affect both the quality and quantity the fire control man to develop a happen. When keen observations of products harvested from the for system aided by instruments and and evaluations are made of est. The forest manager may influ other guides where available, which weather, topography, and fuels, ence fire behavior by the nature of will help him evaluate the com there are very slim chances for his operations, especially in timber bined effects of all significant fac firefighters to be surprised sud cutting. When a forest is opened up tors influencing fire behavior. denly by an unexpected blowup. by thinning or harvesting opera tions, lower humidities, high tem Keen observation is a fundamental Every fire behavior situation calls peratures, and higher wind veloci requirement for judging fire behav for special safety measures. In ties are created within the stands. ior. Many visible signs are present most cases the best safety meas Fire behavior is thereby affected. in the forest to assist the fire con ure is aggressive and intelligent Sometimes the debris remaining trol man in arriving at reliable firefighting aimed at abating the after logging constitutes a fuel con decisions. These include such danger spot. dition which greatly increases the things as the color of the grass and chance for fires to ignite and burn other annual vegetation, the posi Keen observation and interpreta intensely. For these reasons it is tion of a fire on a slope, the time of tion of weather, topography, and important for forest managers to day, and the amount of sunshine fuels lead to a good understand know fire behavior and to be able to filtering through the forest canopy. ing of fire behavior and to safe, evaluate the influence of forest One of the purposes of this hand efficient firefighting. management operations on it. book is to analyze the importance
Volume 64 • No. 1 • Winter 2004 7 fighters). Underpredictions can also are shown in the list of additional Judging the quality of a render chosen operational strategy references beginning on page 10. fire behavior prediction and tactics useless (Cheney 1981). Because copies of many of these or forecast solely on the In addition to evaluating the out articles are difficult to obtain, even outcome can be come of a forecast, it is wise to look through library sources, they are hazardous. at the fire behavior prediction being scanned and will be made process itself. Russo and available through the World Wide Schoemaker (1989) examine com Web. Many are now available for dedication and outstanding editori mon pitfalls for decisionmakers downloading from the Fire Man al abilities have brought “life” to that are equally valid for FBANs agement Today Website many of the articles contained in and others making fire behavior (
Table 1—The scope of quantitative wildland fire behavior prediction (adapted from Rothermel 1974, 1980).
Resolution Relative Ease of Impact of Fire usefulness/ prediction inaccurate situation Intended use Timeframe Area value accuracy prediction
Possible Training Long-term Not Moderate Extremely to Minor or applicable very easy minimal
Long-range planning Yearly/ State/ Good Easy to Significant (e.g., preparedness seasonal province/ moderately system development) territory Easy
Potential Short-term planning Daily/ Forest/ Very good Moderately Serious (e.g., daily fire weekly district difficult to assessment) difficult
Actual Near real-time Minutes to Stand- or Excellent Very to Critical (e.g., automated hours site-specific extremely dispatch, project difficult fires, escaped fire situation analysis)
Fire Management Today 8 Barrows, J.S. 1951. Fire behavior in north The Ten Most Dangerous ern Rocky Mountain forests. Stn. Pap. No. Decision Traps* 29. Missoula, MT: USDA Forest Service, Northern Rocky Mountain Forest and 1. Plunging in: Beginning to 6. Shooting from the hip: Range Experiment Station. Brown, H. 2002. Thirtymile Fire: Fire gather information and reach Believing you can keep behavior and management response. Fire conclusions without first tak- straight in your head all the Management Today. 62(3): 23–30. ing a few minutes to think information you’ve discovered, Bunton, D.R. 2000a. Creating an index that about the crux of the issue and therefore “winging it” mirrors our past. Fire Management Today. 60(1): 27–31. you’re facing or to think rather than following a sys- Bunton, D.R. 2000b. Subject index— through how you believe deci- tematic procedure when mak Volumes 31–59. Fire Management Today. sions like this one should be ing the final choice. 60(1): 32–94. Cheney, N.P. 1981. Fire behaviour. In Gill, made. 7. Group failure: Assuming that A.M.; Groves, R.H.; Noble, I.R., eds. Fire 2. Frame blindness: Setting out with many smart people and the Australian Biota. Canberra, to solve the wrong problem involved, good choices will Australia: Australian Academy of Sciences: 151–175. because, with little thought, follow automatically, and Cornwall, M. 2003. Dome Peak Fire: you have created a mental therefore failing to manage Witnessing the extreme. Fire framework for your decision the group decisionmaking Management Today. 63(1): 16–18. that causes you to overlook process. Rothermel, R.C. 1974. Concepts in fire modeling. Paper prepared for Advanced the best options or lose sight 8. Fooling yourself about feed- Fire Management Training Course, of important objectives. back: Failing to interpret the National Fire Training Center, 1974 3. Lack of frame control: evidence from past outcomes November 11–22, Marana, AZ. Rothermel, R.C. 1980. Fire behavior sys Failing to consciously define for what it really says, either tems for fire management. In: Martin, the problem in more ways because you are protecting R.E. and others, eds. Proceedings of the than one or being unduly your ego or because you are Sixth Conference on Fire and Forest Meteorology. Washington, DC: 58–64. influenced by the frames of tricked by hindsight. Rothermel, R.C. 1983. How to predict the others. 9. Not keeping track: Assuming spread and intensity of forest and range 4. Overconfidence in your judg- that experience will make its fires. Gen. Tech. Rep. INT–143. Ogden, ment: Failing to collect key lessons available automatical UT: USDA Forest Service, Intermountain Forest and Range Experiment Station. factual information because ly, and therefore failing to Russo, J.E.; Schoemaker, P.J.H. 1989. you are too sure of your keep systematic records to Decision traps: Ten barriers to brilliant assumptions and opinions. track the results of your deci decision-making and how to overcome them. New York, NY: Simon and Schuster 5. Shortsighted shortcuts: sions and failing to analyze Incorporated. Relying inappropriately on these results in ways that Saveland, J.M.; Wade, D.D. 1991. Fire man “rules of thumb,” such as reveal their key lessons. agement ramifications of Hurricane Hugo. In: Andrews, P.L.; Potts, D.F., eds. implicitly trusting the most 10. Failure to audit your decision Proceedings of the 11th Conference on readily available information process: Failing to create an Fire and Forest Meteorology; 1991 April or anchoring too much on organized approach to under- 16–19; Missoula, MT. SAF Publ. 91–04. convenient facts. standing your own decision- Bethesda, MD: Society of American Foresters: 124–131. making, so that you remain Weick, K.E. 2002. Human factors in fire constantly exposed to all the behavior analysis: Reconstructing the * Based on Russo and Schoemaker (1989). above mistakes. Dude Fire. Fire Management Today. 62(4): 8–15. Weick, K.E. 2003. Managing the unexpect ed: A look at the big ideas. Paper present ed at the Georgia Tech’s DuPree College Russo and Schoemaker (1989) examine common of Management and Leadership Center, Carter Presidential Center; 2003 March pitfalls for decisionmakers that are equally valid for 25; Atlanta, GA. Weick, K.E.; Sutcliffe, K.M. 2001. Managing FBANs and others making fire behavior the unexpected: Assuring high perform predictions or forecasts. ance in an age of complexity. San Francisco, CA: Jossey-Bass. ■
Volume 64 • No. 1 • Winter 2004 9 Additional References on Wildland Fire Behavior* Fire Behavior Officer/Fire Moore, W.R. 1959. Training in the ten stan Greenlee, D.; Greenlee, J. 2002. Changes in Behavior Analyst dard fire fighting orders. Fire Control fire hazard as a result of the Cerro Bushey, C.L.; Mutch, R.W. 1990. Fire Notes. 20(2): 58–60. Grande Fire. Fire Management Today. behavior service center for extreme Mutch, R.W. 2002. Why don’t we just leave 62(1): 15–21. wildfire activity. Fire Management the fireline? Fire Management Today. Jemison, G.M. 1939. Determination of the Notes. 51(4): 34–42. 62(4): 22–24. rate of spread of fire in the Southern Chandler, C.C.; Countryman, C.M. 1959. Thomas, D.; Cook, W. 2002. Dude Fire staff Appalachians. Fire Control Notes. 3(1): Use of fire behavior specialists can pay ride. Fire Management Today 62(4): 4–5. 4–7. off. Fire Control Notes. 20(4): 130–132. Thorburn, W.R.; MacMillan, A.; Alexander, Johnson, V.J. 1982. The dilemma of flame Countryman, C.M.; Chandler, C.C. 1963. M.E.; Nimchuk, N.; Frederick, K.W.; Van length and intensity. Fire Management The fire behavior team approach in fire Nest, T.A. 2003. “Principles of Fire Notes. 43(4): 3–7. control. Fire Control Notes. 24(3): Behavior”: A CD-ROM-based interactive Luke, R.H.; McArthur, A.G. 1963. Fire 56–60. multimedia training course. Fire behavior studies in Australia. Fire Dell, J.D. 1966. The fire-behavior team in Management Today. 63(2): 43–44. Control Notes. 24(4): 87–92. action—The Coyote Fire. Fire Control Sackett, S.S.; DeCoste, J.H. 1967. A new Notes. 27(1): 8–10, 15. Research mobile fire laboratory. Fire Control Knutson, K.K. 1962. The place of the fire Alexander, M.E.; Andrews, P.L. 1989. Notes. 28(4): 7–9. behavior officer in the fire suppression Wildland fire occurrence and behavior Stocks, B.J. 1977. Fire behavior research in organization. Fire Control Notes. 23: analysis in the year 2000 and beyond. Ontario. Fire Management Notes. 38(2): 81–82. Fire Management Notes. 50(4): 35–37. 9–11, 19. Weick, K.E. 2002. Human factors in fire Alexander, M.E.; De Groot, W.J.; Hirsch, Traylor, R.E. 1961. Correlation of weather behavior analysis: Reconstructing the K.G.; Lanoville, R.A. 1989. Use of posters to fire spread in grass and brush fuels on Dude Fire. Fire Management Today. for interpreting fire behavior and danger the Snake River plains in southern Idaho. 62(4): 8–15. research. Fire Management Notes. 59(2): Fire Control Notes. 22(4): 118–119. 41–44. Training Alexander, M.E.; Maffey, M.E. 1992–93. Computerized Aids and Decision Alexander, M.E. 2002. The staff ride Predicting fire behavior in Canada’s aspen Support Systems approach to wildland fire behavior and forests. Fire Management Notes. Andrews, P.L. 1986. Methods for predicting firefighter safety awareness training: A 53–54(1): 10–13. fire behavior—You do have a choice. Fire commentary. Fire Management Today. Alexander, M.E.; Yancik, R.F. 1977. The Management Notes. 47(2): 6–10. 62(4): 25–30. effect of precommercial thinning on fire Andrews, P.L.; Bevins, C.D. 1999. BEHAVE Andrews, P.L.; Sackett, S.S. 1989. Fire potential in a lodgepole pine stand. Fire fire modeling system: Redesign and observation exercises—A valuable part Management Notes. 38(3): 7–9, 20. expansion. Fire Management Notes. of fire behavior training. Fire Banks, W.G.; Frayer, H.C. 1966. Rate of for 59(2): 16–19. Management Notes. 50(1): 49–52. est fire spread and resistance to control Andrews, P.L.; Chase, C.H. 1990. Update of Carlton, D.W. 1991. Fire behavior train in the fuel types of the Eastern Region. the BEHAVE fire behavior prediction sys ing—A look at some upcoming Fire Control Notes. 27(2): 10–13. tem. Fire Management Notes. 51(1): changes. Fire Management Notes. Barry, E.F. 1942. How about the esprit de 22–25. 52(2): 15–19. corps. Fire Control Notes. 6(3): Eubanks, R.L.; Bradshaw, R.L; Andrews, P.L. Cochran, A.R. 1957. A training course in 124–125.** 1986. Current status of BEHAVE system. fire safety and fire suppression tech Byram, G.M.; Martin, R.E. 1962. Fire whirls Fire Management Notes. 47(2): 29–31. niques. Fire Control Notes. 18(1): in the laboratory. Fire Control Notes. Finn, M. 2001. British Columbia Forest 33–38. 23(1): 13–17. Service adds new software for wildland Editor. 1958. Safe practices under Countryman, C.M. 1956. Old-growth con firefighting. Fire Management Today. blowup conditions—A training outline version also converts fire climate. Fire 61(2): 43–44. for the fire crew boss. Fire Control Control Notes. 17(4): 15–19. Finney, M.A.; Andrews, P.L. 1999. FAR Notes. 19(1): 3–7. Countryman, C.M. 1973. The fire environ SITE—A program for fire growth simula Euler, D.H. 1946. The sand box as a fire- ment concept. Fire Management. 34(2): tion. Fire Management Notes. 59(2): control training tool. Fire Control 17. 13–15. Notes. 7(1): 37–39. Curry, J.R. 1936. Fire behavior studies on Rothermel, R.C. 1983. BEHAVE and you Giffen, C.A. 1956. Fire control practices: the Shasta Experimental Forest. Fire can predict fire behavior. Fire A training course. Fire Control Notes. Control Notes. 1(1): 12–13. Management Notes. 44(4): 11–15. 17(3): 26–29. Davis, L.S.; Martin, R.E. 1961. Time–tem Scott, J.H. 1999. NEXUS: a system for Keller, P. 2002. What’s a staff ride? Fire perature relationships of test head fires assessing crown fire hazard. Fire Management Today. 62(4): 6–7. and backfires. Fire Control Notes. 22(1): Management Notes. 59(2): 20–24. Keller, P. 2002. Walk back into tragedy: A 20–21. Van Gelder, R.J. 1976. A fire potential quantum leap forward. Fire Davis, W.S. 1949. The rate of spread–fuel assessment model for brush and grass Management Today. 62(4): 16–21. density relationship. Fire Control Notes. fuels. Fire Management Notes. 37(3): 10(2): 8–9. 14–16. * Articles on or related to fire behavior from issues of Gisborne, H.T. 1942. Review of problems Fire Management Today and its predecessors (Fire and accomplishments in fire control and Hand-Held Computer Technology Control Notes, Fire Management, and Fire fire research. Fire Control Notes. 6(2): Bradshaw, R.L.; Dean, W.A. 1980. Adding Management Notes) from 1936 to 2003 that were not 47–63. print capability to your TI–59 fire behav reprinted or referenced elsewhere in this three-part series of special issues (Fire Management Today vol **See C.F. Olsen, “Analysis of the Honey Fire” (Fire ior CROM. Fire Management Notes. umes 63[3], 63[4], and 64[1]). Management Today 63(3) [Summer 2003]: 29–41). 41(4): 7–8.
Fire Management Today 10 Burgan, R.E.; Susott, R.A. 1986. HP–71 automatic weather stations. Fire Notes. 1(7): 395–396. replaces TI–59 for fire calculations in the Management Notes. 47(4): 5–7. Pirsko, A.R. 1961. Alinement chart for field. Fire Management Notes. 47(2): perimeter increase of fires. Fire 11–13. Fuels and Fuel Sampling Control Notes. 22(1): 1–4. Burgan, R.E.; Susott, R.A. 1988. Correcting Altobellis, A.T.; Cooper, R.W. 1963. Moisture an error in the HP–71B fire behavior content of gallberry and palmetto during Fire Weather and Fire CROM. Fire Management Notes. 49(2): a dry period. Fire Control Notes. 24(1): Weather Forecasting 31–32. 10. Countryman, C.M. 1972. This humidity Cohen, J.D.; Burgan, R.E. 1979. Hand-held Blank, R.W.; Simard, A.J.; Eenigenburg, J.E. business: What it is all about and how calculator for fire danger/fire behavior. 1985. A tester for measuring the mois it is used in fire control. Fire Control Fire Management Notes. 40(1): 8–9. ture content of dead fine fuels. Fire Notes. 33(2): 10–11. Management Notes. 46(2): 8–12. Cramer, O.P. 1950. Use your weather Research Instrumentation Bruce, D. 1951. Fuel weights on the records to interpret fire-weather fore Clark, B.; Steuter, A.A.; Britton, C.M. 1981. Osceola National Forest. Fire Control casts. Fire Control Notes. 11(4): 41–43. An inexpensive anemometer frame. Fire Notes. 12(3): 20–23. Cuoco, C.J. 1992–93. Prescribed burns? Management Notes. 42(3): 13–14. Buck, C.C. 1951. Flammability of chaparral Share information with fire weather Dibble, D.L. 1960. Fire climate survey trail depends on how it grows. Fire Control forecasters and involve them in the er. Fire Control Notes. 21(4): 116–120. Notes. 12(4): 27. planning. Fire Management Notes. Little, E.C. 1973. Costs $10—Foolproof Countryman, C.M. 1974. Moisture in living 53–54(3): 10–13. timer measures rate of fire spread. Fire fuels affects fire behavior. Fire Fite, F.M. 1953. Fire weather forecasts. Management. 34(4): 10–12. Management. 35(2): 10–14. Fire Control Notes. 14(1): 18–20. McMahon, C.K.; Adkins, C.W.; Rodgers, S.L. Dieterich, J.H. 1963. Litter fuels in red pine Fujioka, F.M. 1997. High resolution fire 1986. A video image analysis system for plantations. Fire Control Notes. 24(4): weather models. Fire Management measuring fire behavior. Fire 103–106. Notes. 57(2): 22–25. Management Notes. 47(1): 10–15. Miller, R.K.; Schwandt, D.L. 1979. Slash Graham, H.E. 1964. A portable fire- Schaefer, V.J. 1959. Use of the 60-second fuel weights in red pine plantations. Fire weather forecast unit for use on back print camera for stereophotography of Management Notes. 40(1): 6–7. country fires. Fire Control Notes. project fires and related activities. Fire Potts, D.F.; Ryan, K.C.; Loveless Jr., R.S. 25(3): 11. Control Notes. 20: 89–90. 1984. A procedure for estimating duff LaMois, L.M. 1961. Weather and forest depth. Fire Management Notes. 45(2): fires. Fire Control Notes. 22(1): 22–24. Supporting Tools and Equipment 13–15. O’Dell, C.A.; Hammer, L.C. 1979. Fire Anderson, K. 2001. NIFC FIRE RAWS unit Scott, J.H.; Reinhardt, E.D. 2002. weather meteorological support units. survives burnover. Fire Management Estimating canopy fuels in conifer Fire Management Notes. 40(2): 3–5. Today. 61(2): 39–42. forests. Fire Management Today. 62(4): Potter, B.E. 1997. Making sense of fire Clark, B.; Roberts, F. 1982. A belt weather 45–50. weather. Fire Management Notes. kit accessory for measuring woody fuel Weise, D.R.; Saveland, J.M. 1996. 57(2): 26–27. moisture. Fire Management Notes. 43(3): Monitoring live fuel moisture—A task Rodney, E.A. 1964. Forest fires and fire 25–26. force report. Fire Management Notes. weather conditions in the Asheville, Dell, J.D.; Hull, M.K. 1966. A fire-behavior 56(3): 13–14. N.C., fire weather district—Spring sea team field unit. Fire Control Notes. 27(3): son, 1963. Fire Control Notes. 25(3): 6–7. Guidelines and Decision Aids 7–9, 15. Division of Forest Fire Research, Alexander, M.E.; Stam, J.C. 2003. Safety Schroeder, M.J. 1950. The Hudson Bay Intermountain Forest and Range alert for wildland firefighters: Fuel condi High and the spring fire season in the Experiment Station. 1959. Belt weather tions in spruce-beetle-killed forests of Lake States. Fire Control Notes. 11(1): kit. Fire Control Notes. 20(4): 122–123. Alaska. Fire Management Today. 63(2): 1–8. Ellis, G.R. 1965. A combination pocket 25. Svorcek, A.J. 1965. 50 years of fire weath meter for windspeed and duration. Fire Anderson, H.E. 1984. Calculating fire size er service. Fire Control Notes. 26(2): Control Notes. 26(2): 5. and perimeter growth. Fire Management 8–9. Fischer, W.C. 1978. New portable weather Notes. 45(3): 25–30. Watkins, C.H. 1961. The Oregon state instrument shelter performs well. Fire Cargill, G.E. 1970. Table speeds fire spread mobile fire weather unit. Fire Control Management Notes. 39(3): 15–18. estimates. Fire Control Notes. 31(2): Notes. 22(3): 89–92. Maxwell, F.; McCutchan, M.; Roberts, C.F. 15–16. 1974. Automation of fire weather obser Greenlee, J.; Greenlee, D. 2003. Trigger Long-Duration Projections vations. Fire Management. 35(3): 22–25. points and the rules of disengagement. Lukens, D.; Krebs, J. 1986. Long-term Palmer, T.Y.; Pace, G.D. 1974. Microwave 63(1): 10–13. fire behavior projections. Fire oven dries fuels fast. Fire Management. Melton, M. 1989. The Keetch/Byram Management Notes. 47(4): 22–23. 35(2): 22–23. Drought Index: A guide to fire conditions Mohr, F.; Both, B. 1986. Confinement—A Sackett, S.S. 1980. An instrument for rapid, and suppression problems. Fire suppression response for the future? accurate determination of fuel moisture Management Notes. 50(4): 30–34. Fire Management Notes. 56(2): 17–22. content. Fire Management Notes. 41(2): Melton, M. 1996. Keetch–Byram Drought Mohr, F.; Lukens, D.; Terry, D. 1987. 17–18. Index revisited: Prescribed fire applica Managing confinement suppression Warren, J.R. 1980. Remote automatic tions. Fire Management Notes. 56(4): response on the Middle Ridge and weather stations (RAWS). Fire 7–11. Little Granite Fires, August 1986. Fire Management Notes. 41(2): 15–16. Mitchell, J.A. 1936. Rule of thumb for Management Notes. 48(3): 23–25. Warren, J.R. 1986. Very portable remote determining rate of spread. Fire Control
Volume 64 • No. 1 • Winter 2004 11 FOREST FIRES AND SEA BREEZES*
G.L. Hayes
pot fires which started upwind from going forest fires have Coastal surface winds can change direction S been reported by I.S. Stivers, abruptly in mid-afternoon, carrying going fires in Forest Ranger for the New York an unexpected direction and upsetting suppression Conservation Department, whose district covers eastern Long Island. plans. They had been observed on a num ber of occasions, and from a num The conditions described and the and returns at a higher elevation ber of different fires. location, on Long Island, indicate from land to water. During the day that the type of local winds known light hours the air is heated more Suspecting at first that incendiaries as sea breezes was responsible for over the land than over the water. were setting fires behind him, both the upwind spot fires and for This sets up a local pressure system Stivers sent patrols upwind from the rapid changes in direction of that induces the warmer, lighter going fires. The patrols found no the surface wind. Much has been land air to rise and flow seaward incendiaries but they did find new learned about sea breezes in recent and the colder, heavier air over the fires starting. They, and he, also years that should be of very materi water to settle and flow landward. observed that the smoke column, al help in planning fire suppression after rising high in the air, turned in such coastal areas as Long The sea breezes occur on warm and moved back in a direction Island. Obviously, fire suppression days near the shores of large bodies opposite to the surface winds. The is most difficult when rapid and of water. They are strongest and spots were starting from embers unexpected changes in wind condi best developed along the seacoasts which fell from this smoke column. tions occur. If the wind shifts can but occur also along the shores of be anticipated, defensive action can bays and large lakes. In the temper On other occasions, Stivers wrote, be planned in advance. ate zone the landward flowing wind surface winds changed abruptly in current may be from 200 to 2,000 mid-afternoon from a northerly or There is an excellent discussion of feet (60–600 m) thick and may westerly to a southerly or easterly sea breezes in the June 1946 issue reach inland for 20 to 25 miles direction, carrying going fires in an of the bulletin of the American (32–40 km). Above this is the unexpected direction and upsetting Meteorological Society under the return current. Under the same suppression plans. A typical case title “Theory and Observation of conditions it may extend offshore was a fire on Sunday, April 1, 1945, Land and Sea Breezes,” by as far as 60 miles (97 km) over the at 2:30 p. m., that started with a Raymond Wexler. As many fire con ocean. In hotter climates or in northwest wind and began to trol men in coastal areas may not combination with topographic spread to the southeast. Fifteen have access to the Bulletin, the fol winds the inland range is extended. minutes later the wind shifted fast lowing digest of Mr. Wexler’s article The winds from lakes extend short to the southwest and sent the fire has been prepared. The land breeze er distances. over the Radio Corporation is not mentioned as it occurs main Communications plant at ly at night and is felt primarily over Two distinct types of sea breezes are Riverhead. the water. recognized. The first type develops when there is little or no gradient Definition and wind;** the second type develops When this article was originally published, Characteristics of when there is a light offland gradi G. L. Hayes was a forester for the USDA ent wind. The first type develops as Forest Service, Southeastern Forest Sea Breezes Experiment Station, Asheville, NC. A sea breeze is a local circulation in ** The gradient wind is the air movement caused by the which the wind near the surface prevailing pressure differences in the atmosphere. It is * The article is reprinted from Fire Control Notes 8(2/3) blows from the water onto the land the wind that is usually predicted in the Weather [Spring/Summer 1947]: 30–33. Bureau forecasts.
Fire Management Today 12 a small circulation near the shore The most dangerous part of the sea breeze early in the day, soon after the air circulation is the front or surface separating the over the land has become warmer than the air over the water. With landward blowing sea air from the seaward flowing continued heating of the land, the land air. circulation extends progressively farther landward and seaward and grows stronger and deeper. After about a half hour from the 2. Gradient wind. Calm conditions, time the front has passed, the or a light offland gradient wind The second type, which is the more velocity is usually very constant, are favorable for sea breeze for common in temperate latitudes, with little gustiness. As the higher mation. If the gradient wind is develops over the water and usually winds are then flowing opposite to blowing parallel to the shore or comes onto the land suddenly, later the surface winds, the danger of off the water, the sea breeze will in the day. The offland gradient upwind spot fires is present. not develop. wind holds the colder and heavier sea air back and heats it up until Although the sea breeze blows from The velocity of the offland gradi the force of the wind can no longer water to land, it does not always ent wind affects the time of hold it. Then the sea air rushes blow perpendicular to the coast arrival of the sea breeze and the ashore where it is heated until it line. It tends to blow perpendicular distance inland that it will move. rises and joins the gradient wind at first then shift to the right as the Under calm conditions, the sea which is blowing out to sea over day grows older. Thus, along the breeze may develop near the head. The typical sea breeze circu east coast where the shore is direct shore soon after sunup and move lation is then established. ly north and south it would tend to progressively farther inland until start as an easterly wind, shifting to the maximum temperature for The most dangerous part of the sea southerly. Along the west coast it the day is reached, after which is breeze circulation, from the fire would tend to start as a westerly subsides. The stronger the control standpoint, is the front or wind, shifting to northerly. offland gradient wind, the later surface separating the landward in the day the sea breeze comes blowing sea air from the seaward External Factors ashore, and it may never pene flowing land air. The reasons are: Influencing Sea trate far inland. In fact, if the Breezes wind is strong enough, the sea air cannot leave the water. At 1. The winds blow in opposing Several conditions affect sea breeze Danzig a gradient wind of 22 directions on either side of the formation and behavior. front and rise at the front. miles per hour (35 km/h) was observed just to balance the 2. The front moves. The rate of its 1. Character of day. As sea breezes force of the sea breeze. The front advance is less than the velocity occur only when the air over the moved intermittently back and of the sea breeze behind it and it land becomes warmer than over forth across the shore line. decreases as it moves farther the sea, clear, hot days are most inland. When a front moves favorable to their formation. To have a front stall over a fire across a fire, the rear or a flank They can and do occur on over would create a very bad situa suddenly becomes the head of cast days but they form later, are tion. The winds could be strong, the fire. milder, and extend inland for and would certainly be gusty and 3. The winds along the front are shorter distances. In general, the fluctuate wildly in direction, as the strongest and gustiest part of clearer and hotter the day, the the front moved back and forth. the sea breeze circulation. Initial earlier the sea breeze will form, gusts of the sea breeze as strong the stronger it will get, and the 3. Topography. Where there are as 34 miles per hour (55 km/h) farther inland it will penetrate. mountains along a shore line, have been recorded, whereas the With light gradient winds and the sea breeze may combine with average behind the front is only clear skies, it usually starts about an upvalley or upslope wind. about 11 miles per hour (18 2 to 3 hours after sunrise and Such a combination wind is km/h). ends within 2 hours before sun stronger than a straight sea set.
Volume 64 • No. 1 • Winter 2004 13 breeze and may extend much far Where the sea breeze is observed to have ther inland. If the mountains lie important effects on fires, fire control men would several miles back from the coast, separate circulations may profit by observing its characteristics. be set up in the morning which will merge after noon. Such a along our coast are more favor 6. Distance from the shore. The combination in California is able to sea breezes when the veg sea breeze is felt first and has reported to establish a continu etation is dead and the leaves are greatest velocity right at the ous flow of wind for as much as off the deciduous trees than after shore. As distance from shore is 40 miles (64 km) inland. A simi the fields and woods “green up.” increased the sea breeze arrives lar but less extensive flow takes later in the day, has less velocity, place between Great Salt Lake 5. Atmospheric stability. An unsta and the front moves more slowly. and the Wasatch Mountains in ble lower atmosphere is more Utah. favorable for sea breezes than a With so many factors affecting the stable one. In an unsaturated time of arrival and characteristics Along the shores of a bay there atmosphere, stability depends on of the sea breeze, it is impossible to may be two components of the the rate of temperature drop set up definite rules which will tell sea breeze, one from the bay and with increasing elevation. If the when it may arrive or how it will the second from the sea beyond. temperature decreases more behave for any particular place or The bay circulation will usually than 5-1/2 ºF in 1,000 feet of ele day. Where the sea breeze is be the first to affect the land but vation (or 1 ºF in 182 feet), the observed to have important effects may be replaced later by the air is unstable and ascending on fires, fire control men would ocean breeze, accompanied by a convection currents develop eas profit by observing its characteris change in wind direction. ily. If it decreases less than this, tics as related to the factors already it is stable and convectional discussed. Or the local weather 4. Vegetation. A heavy vegetative movement cannot take place. Air forecaster of the U.S. Weather cover retards heating of the land over the land that is very stable Bureau might be induced to predict surface. Hence, the sea breeze in the morning may, through the time of arrival, its range inland, starts earlier and becomes surface heating, become unstable and probable velocity at and behind stronger along desert or semi- later in the day, hence the the front. ■ desert coasts than along heavily hottest part of the day is most forested ones. Likewise, with favorable for sea breezes. other things equal, conditions
Russo and Schoemaker (1989) Decision Trap 1—Plunging In: Beginning to gather information and reach conclusions without first taking a few minutes to think about the crux of the issue you’re facing or to think through how you believe decisions like this one should be made.* * See page 9.
Fire Management Today 14 FUNDAMENTALS OF FIRE BEHAVIOR*
H.T. Gisborne
ur job of fire control can be done, in fact has been done, in There are only three things you can do to stop a O several ways: By brute strength fire—rob it of its fuel, keep it from being heated to and little attention to the condi the ignition point, or shut off the oxygen supply. tions we are attempting to control; by observation of what is happening but with little or no understanding there are only three basic factors or nates as (C6H10O5)y. This means that of why the fire is behaving as it three essentials to this chemical there are 6 atoms of carbon, 10 does; or by practical application of process, it is obvious that we are atoms of hydrogen, and 5 of oxygen knowledge of the basic laws of overlooking a bet if we fail to con in each molecule of cellulose. chemistry and physics that are sider each of these three things in Starch, which is found in the roots, actually determining the rate at our calculating. seeds, and leaves of all plants, is which a fire is spreading. Let us very similar, differing only in the look into the most significant fac Three Essentials of Combustion. subscript. The chemists designate tors that affect fire behavior. Completely controlling the chemi the various starches as (C6H10O5)x. cal reaction called fire are only Fire is a Chemical three essentials. They are: This point is important to remem Process ber because it helps to reduce some Combustion is a chemical process. 1. Fuel or something that will com errors of judgment based on the It is classified that way because bine with oxygen rapidly enough belief that the chemical nature of combustion, with or without flame, to generate heat; our fuels differs very materially.
is a molecular reaction in which 2. Heat enough to raise that fuel to When C6H10O5 burns, every mole molecules of oxygen in the air com the ignition point; and cule of that substance combines bine with molecules of cellulose 3. Plenty of oxygen in contact with with six molecules of oxygen. The and lignin (which make wood) and the fuel or with the gases evolved resultant products are gases, 6 mol thereby change most of the solid from the wood. ecules of carbon dioxide, and 5 into gases. These gases are mole molecules of water vapor. Fire cules of different substances. They Remove the fuel as we do when we makes water out of the hydrogen are no longer cellulose and lignin. dig a fire trench; keep it from being and oxygen atoms that are in every Such changes of substance are heated to the ignition point, as we molecule of wood. The chemist ➞ chemical, not physical, processes. do when we widen the trench or writes it this way: C6H10O5 + 6O2
When these changes occur at such when we use water; or shut off the 6CO2 + 5H2O. Unfortunately, that a rapid rate that heat and flame are supply of oxygen as we do when we water is not of any help to us produced, the process is called throw dirt, use water, or bury a because it exists as a gas, a super combustion or fire. burning log, and you can stop the heated gas, which rises straight up spread of any fire. Every one of our and away from our fuels. The water When you look into the fundamen methods of fighting fire is based on that really counts is the moisture tals of combustion and find that one or more of those three simple content of the grass, trees, or brush essentials. THERE ARE NO OTHER before they burn. WAYS. When this article was originally published, Because of this similarity of chemi H.T. Gisborne was in charge of the Division of Forest Protection for the USDA Forest Fuel. Chemically, all of the fuels cal composition of our fuels, it is Service, Northern Rocky Mountain Forest that carry our fires are practically obvious that we should not calcu and Range Experiment Station. alike. From grass and brush to tree late probabilities on the belief that needles, tree trunks, and rotten different kinds of wood or brush or * The article is reprinted from Fire Control Notes 9(1) wood on the ground, they are all of grass burn differently. The leaves of [Winter 1948]: 13–24. It was used at the 40-man Fire Boss School on May 5, 1947. the type that the chemist desig- grass, trees, and brush and the bark
Volume 64 • No. 1 • Winter 2004 15 and wood of trees are all largely Moderate to large areas of fuel releasing their cellulose. The big variable which energy suddenly are creating conditions that breed produces really significant differ- ences in fire behavior is not the not only higher wind velocities, but twisters or chemicals, it is the moisture con- even big whirlwinds. tent.
There are, however, two other other words the kindling tempera Size of fuel is also worth noting ingredients in wood in addition to ture of grass, wood, cotton batten, from another angle. Take 10 cellulose that are of some, perhaps or cellulose in any natural form is pounds (4.5 kg) of dry grass or dead academic, significance. One of easily produced. It is not an abnor pine needles, 10 pounds (4.5 kg) of these is lignin, a substance for mally high temperature. You will dry branchwood, and 10 pounds which the chemists do not know build more held line and have to (4.5 kg) of log in one chunk and the formula. The significance of charge up less line lost if you ignite each of them. What happens? lignin lies in the fact that it has a remember that simple fact and The needles will liberate their slightly higher heat content than then do something about it. B.t.u.’s (British thermal units) in a cellulose and that it leaches and few seconds, the branches will decays more slowly. Hence old The key to ignition is the ease with release theirs in a few minutes, wood is likely to have lost more cel which a fuel can be heated to 540 while the 10-pound (4.5-kg) log lulose than lignin and therefore °F (282 ºC). That ease naturally may take half an hour to release its will have a slightly higher heat con depends upon one obvious differ heat. Ease of ignition is, therefore, tent per pound of material remain ence in fuels, i.e., their size. The not the only difference in fire ing than fresh cut or freshly killed fine fuel naturally heats clear behavior to expect in accordance wood. Differences in the pitch con through and reaches 540 °F (282 with different sizes of fuel. The rate tent are also known to affect the ºC) far quicker than a heavy fuel of release of the energy is also heat of a fire. like a log. Size of fuel is therefore tremendously different. the significant feature to watch, There are also some other minor other things such as moisture con This feature, combustion rate, is differences in the chemical nature tent being equal. Actually, size and what a football player would call of plant and tree leaves, but a series moisture content influence the the triple threat of fire. And this of tests of the fat and oil content of process of combustion in much the rate of release of energy is the one the leaves of six different genera of same way. Make a stick wetter and feature we fail most often to recog weeds and brush which were made you reduce its ease of ignition. nize. The three threats involved for three consecutive summers Similarly, the bigger the stick, the are: failed to reveal anything significant. harder it is to ignite it. The wet Instead, this chemical study made stick and the big stick both require 1. The more sudden the release of at our Priest River laboratory con more time or more heat to raise all this heat, the farther it will firmed the finding that moisture their surface temperature to the radiate a temperature of more content is THE big variable. ignition point. And that is another than 540 °F (282 ºC). And that good basic fact to keep in mind means something to your tactics. Ignition. When there is plenty of both in sizing up probable fire It means that if the fuels are fuel, the next essential of combus behavior and in deciding on tactics even moderately dry, a wider fire tion is that it must be heated to the to use along the line. Let your fire line is needed wherever you find ignition point. For dry cellulose, a burn through the heavier fuels an appreciable volume of fine temperature of only 400 ºF to 600 where it will burn more slowly. fuels. This applies to both stop ºF (204–316 ºC) is required. The Fight it at those places where it ping a fire and in backfiring. average usually used is 540 ºF would go into finer fuels and spread 2. The faster the release of those (282 ºC). The point that is of practi faster. Also, fire line construction is B.t.u.’s, the greater the volume cal importance is that if your fuels easier in the fine fuels. You gain in of gas suddenly created, hence are even moderately dry they do two ways by using this basic knowl the faster it will rise overhead. not have to be heated very hot to edge. That also means something to reach this ignition temperature. In tactics employed, because the swifter the rise of hot air the
Fire Management Today 16 The most important variable in fire behavior is fuel known as far back as 1939 that in moisture, and when our fuel moisture indicator this splitting tremendous energy was released and that the process sticks are below 5 percent you can expect your then split other uranium atoms fires to blow up and explode. which in turn released more energy and split more atoms, the process continuing and accelerating as long greater the chance of sucking up dirt is superior to dry dirt primarily as there was a supply of a suitable blazing embers and carrying because it lowers the temperature fuel in a proper arrangement and them up and over the line, if the more. But when either moist or dry condition.. The job of the atomic smoke is leaning across the line. soil covers the surface of the fuel physicists was, therefore, to pro That means spot fires. the major benefit is by cutting duce this chain reaction yet control 3. The faster this release of energy down the oxygen supply. Water also it. Our job is simpler. It is merely to and the faster the uprush created does the same thing if enough is control the molecular chain reac by it, the greater the local wind applied to form a film over the sur tion that is fire. velocity created by the fire. face of the fuel. But here too the Moderate to large areas of fuel major benefit is in lowering the As you can see, fire is a similar releasing their energy suddenly fuel temperature below the ignition process in that if you heat one mol are creating conditions that pint. ecule of a fuel to the ignition point, breed not only higher wind its process of changing from velocities, but twisters or even Combustion—A Molecular Chain C6H10O5 into CO2 and H2O may big whirlwinds. I once saw one of Reaction. The public has heard release enough energy to ignite the really big ones whirl like a and read a lot recently about atom several other adjacent molecules of tremendous barrel and move ic fission, so controlled that it 2 C6H10O5. If the fuel is in a critical across 2 square miles (5.2 km ) becomes a chain reaction and condition (dry enough), as com while I was running 200 yards thereby makes possible atomic pared to a critical mass (large (180 m) along the top of Desert bombs. More understanding of the enough), that process then Mountain, on the Flathead fire job and better financial support becomes a chain reaction and not Forest. by the public may follow when we only spreads like wildfire but it show that the job of fire control is really is wildfire in our case. Oxygen. This last essential of com definitely one of stopping a chain Whereas the nuclear physicists bustion is one that we can’t do very reaction which differs from the have to make their fuels, and much about. Combustion engineers bombs primarily in that ours is a arrange them carefully in an atom who design and operate boilers do a molecular instead of an atomic ic pile, our fuels are arranged for us lot by controlling this one of the chain reaction. and then, periodically, are put into three essentials. But under our proper condition (dryness) such conditions there is a1most always A chain reaction may be compared that the chain reaction starts when plenty of oxygen to facilitate com to a chain letter; you receive one ever and wherever the spark is bustion of our fuels. Under free but you send out two or maybe applied. burning conditions such as occur three or four. Each of the recipients on a forest fire, about 10 pounds of one of these letters similarly 3 If this sounds farfetched or academ (4.5 kg) or 133 cubic feet (3.75 m ) sends out two or three or four. The ic, let me call your attention to one of air is needed for the complete thing spreads like wildfire. The first more fact, which I know you will combustion of only 1 pound (0.45 problem in producing an atomic not dispute. It is this: That when kg) of dry fuel. bomb was along this line. That our fuels are in their most critical problem was to obtain certain condition, i.e., their driest, we have The one time when we do some chemicals which, when assembled some molecular chain reactions thing to reduce the oxygen supply in a sufficient quantity and which are so violent that we cannot is in throwing dirt. While that dirt arrangement, known as the “critical stop them, just as there is no stop does lower the temperature of the mass,” would perpetuate the ping an atomic bomb once its chain fuel it lands on, the principal func process of splitting atoms of urani reaction is started. Furthermore, tion of dirt is to shut off or at least um into atoms of two other ele we have occasions when combus reduce the supply of oxygen. Moist ments, barium and krypton. It was tion in the form of a forest fire
Volume 64 • No. 1 • Winter 2004 17 approaches a rate and even a mag tion that developed in those earlier protest, we discontinued used of nitude rivaling an atomic bomb. critical years. Hence, it is evident the 2-inch (5-cm) ones. Finally, in Those of you who were on any of that the critical variable in fire 1942, with the Model 6 Danger our big fires in 1929, 1931, and behavior is moisture content of the Meter, we dropped duff moisture 1934 probably saw some of these fuels. Consequently, I want to call and began to rely solely on the half- explosions. Many of them covered your attention to some of the possi inch (13-mm) sticks. several square miles in only a bilities available to you for improv minute or two. ing your calculation of probabilities From a technical viewpoint these by watching fuel moisture above all half-inch (13-mm) sticks alone fail If you will keep this chain reaction other elements. to represent our fuels in two ways: idea in mind, and if you will size up your fire, either as a whole or on Basis of Fuel Moisture 1. They do not show the true bene your sector, in the light of the three Measurements. You all know about fits of light rains as well as duff basic essentials of combustion, you the fuel moisture indicator sticks moisture measurements did; and may be able to calculate the proba used at some 175 fire danger sta 2. After heavy rains, they dry out bility of one of these explosions. If tions in Region 1. There are some faster than either duff or 2-inch you can do that, you may be able to things those sticks will tell you far diameter (5-cm) sticks. save your own life and the lives of your men, as well as improve your The error is therefore always fire control tactics. A burning index rating is toward showing more danger than essential to calculation would be revealed if all of the sig There is one basic criterion to nificant forest fuels were measured. watch, however, in trying to antici of the probabilities in The half-inch (13-mm) sticks are pate a molecular chain reaction at any fuel type. not too fast, of course, for cheat- an explosive rate. This is moisture grass, but this fuel type does not content of the fuel, for it is mois cover a large percentage of our ture content, not mass, nor vol better, far more accurately than you area. Furthermore, after it has ume, nor size, nor arrangement of can estimate. To make best use of cured, cheatgrass responds so fuel which first determines whether those stick measurements you need closely to changes of relative or not a forest fire can truly to know: Why we use half-inch (13 humidity that humidity measure explode. And you should remember mm) sticks, how they are made, ments can very well be used as an that this moisture content not only and how accurate they are. index of moisture content of that can be, but is being measured. You one fuel type. Finally, cheatgrass can get these measurements every For four consecutive summers, changes moisture and flammability day if you want them. 1922, 1923, 1924, and 1925, I col so rapidly that you might as well lected at periodic intervals samples always be ready for the worst. Moisture Content— of the five major dead fuels that burn in a forest fire. I took these The half-inch (13-mm) sticks which The Critical Variable samples to the laboratory and We have not had any true forest fire we now use are made from new determined their moisture con lumber each year. Any one of sever explosions in Region 1 since 1936. I tents. I found out which fluctuated believe there were a couple of al species of wood could be used, the most, and which the least. On because here again we are dealing minor ones that year on the Little this basis, I selected the top layer of Rockies Fire on the Lewis and primarily with cellulose. We use duff, half-inch (13-mm) sticks, and ponderosa pine because it is readily Clark Forest. But we had several 2-inch-diameter (5-cm) branch really big ones in 1934, 1931, 1929, obtainable in clear stock at a rea wood as the best representations. sonable price. We use only sapwood and one or two in 1926. You have We therefore used duff hygrome all read about those in 1919 and because it is the moisture content ters, half-inch (13-mm) sticks, and of sapwood of twigs, branches, logs, 1910. The main reason why we 2-inch (5-cm) sticks at several fire have not had any explosions in and snags in which we are most danger stations for the next 5 years interested. We can ignore the mois recent years is this matter of mois to measure fuel moisture. Then, at ture content. Our fuels simply have ture content of the heartwood of a the suggestion of the rangers in a log because if the outer sapwood is not dried out to the critical condi regional meeting and despite my
Fire Management Today 18 extremely dry the inner heartwood Within the mid-elevation thermal belt, you can has got to be dry too. We also expect the least benefit from increased fuel ignore the effect that bark has on natural wood, because if we used moisture at night. natural sticks with bark on them some of that bark would soon chip wet here and dry there, but under and one under the half shade left off and then we would no longer widespread and long continued after a moderately heavy logging know the true oven-dry weight of drought it is fully adequate. The operation. The stick moistures our sticks. sticks are exposed on a flat, in the obtained by this method can there open, but under a shading layer of fore be accepted as representing To be sure that moisture measure screen cloth. The reason for this, average-bad conditions. Open south ments made at different stations do preparing to meet “average-bad” slopes will have drier half-inch not differ because of differences conditions, is used in all fire con sticks. Densely timbered north between the sticks or because of trol planning in Region 1. The slopes will have materially higher errors by the danger station opera sticks are therefore always exposed fuel moistures. But when the sticks tor, we go to a lot of work and alike at all stations so that the at our stations have high moisture incur considerable expense. These results are truly comparable. contents, adjacent areas, both open sticks now cost from $1 to $1.75 and timbered, also can be expected per set to manufacture. In making The intention in such an exposure to be moist to wet. When our sticks them they are oven-dried and then is to sample average-bad but not are each day showing lower and cut off at the ends until they weigh the very worst conditions. By sam lower moistures you can depend on exactly 100.0 grams oven dry. This pling average-bad conditions we are it that both the open areas and the is done so that all that is needed to using the sound engineering prin timbered slopes will also be getting determine their moisture content ciple of preparing for the worst drier and drier. in percent is to weigh them and probable but not the worst possible. subtract 100.0 from the total Engineers did not build the Golden Our present sticks and exposures weight. Gate Bridge at San Francisco to therefore give you one definite and withstand the worst possible earth dependable index to watch. They As you can see, this difference in quake. They built it to withstand give you something that you can weight is not only the weight of the the worst probable. Few ditches, use in calculating, instead of guess water in the wood, picked up from storm sewers, or bridges are built ing. the air and from rain, but it is also to withstand the worst possible the moisture content expressed as a flood. To meet worst possible condi The most significant single feature percentage of the oven-dry weight. tions usually costs more than the of stick moistures to watch for is Consequently, when you call for a resource is worth. It is better eco just this: Are they below 5 percent fuel moisture content measure nomics therefore to accept the risk and how long have they been there? ment from any of our stations you of the worst possible flood, earth Your danger of blow-ups and explo can bank on its accuracy probably quake, or fire weather conditions, sions can be really calculated by 95 times out of 100. The other 5 and plan to meet only the average- getting merely that information. If times the scales will be out of bal bad or worst probable. We can get the sticks at both the nearest ance, which is an operator error, or adequate fire control at a justifiable ranger station and some nearby the operator will have read the cost by using this principle. We do lookout have been down below 5 scales wrong. Eliminating that use it, not only in fire danger meas percent for several days you can error is a job for training and urement, but also in all phases of bank on it that every fuel type in supervision. fire control planning in Region 1. that area is in a truly critical condi tion. Fortunately, this does not Application of Stick Moisture. By The double layer of 12-mesh screen happen very often, but it has hap the present practice we measure cloth under which we expose our pened and it will happen again. stick moistures at only two to four sticks provides an amount of shade When it does you will be making occupied stations per ranger dis and a fuel-moisture equivalent to the mistake of your life if you fail to trict. That is not enough under what you would get if you operated know it. You can always find out by some conditions of spotted weather, two danger stations, one in full sun consulting the local ranger station
Volume 64 • No. 1 • Winter 2004 19 fire danger charts or Form 120 R-1. or lower from 2 till after 7 p.m. become much more involved. It If you are already out on a fire a THAT is explosive fire weather. should be evident, nevertheless, phone call will bring you the that fuel moisture is THE major desired information. Differences in rate of spread variable and that if you are to cal between fuel types practically disap culate accurately, your first and If the sticks are reported as at less pear under these explosive condi best bet is to get the stick mois than 5 percent, you should then ask tions. The basic laws of chemistry tures and other measurements for two more things: a check of the take charge when nature produces from the nearest danger stations computations to be sure no errors such conditions, and the molecular before you even start to order men. were made, and a remeasurement chain reaction is actually unstop After you get to the fire, you can of the sticks right then. The dis pable until the wind goes down, the then see to it that you are informed patcher or his assistant can do both humidity goes up, and the fuels each day, preferably twice a day, as of these in 10 or 15 minutes. If absorb a little moisture. If you have to how fuel moisture and other fac these checks verify the original to fight such fires, and you should tors are changing. There are then reports, you can then calculate that be mentally ready for it, you will three other major variables to every fuel type in the area, on both probably do it like Kelley and Ryan watch. These are fuel type, the north and south slopes, and at all fought the Freeman Lake explosion. thermal belt, and wind. altitudes, is in its most explosive You will not build much fire line condition. You can bank on it that that day, but you will calculate Fuel Types fire will spread in all of these types where that fire front will be at mid Some men have a misconception at the fastest rate, that there will be night and you will then have fire about fuel types because they do little difference in rate of spread camps and men well distributed not understand that our four rates between fuel types, and that the around it and ready to begin work of spread—Extreme, High, danger of both spotting and of big at the first crack of dawn. Kelley Medium, and Low—represent dif whirls will be at a maximum. You and Ryan had more than 600 men ferences only on a class 65 to class can expect a chain reaction at its strung around the Freeman Lake 75 day. Obviously, rate of spread worst. Fire front the next morning after will not differ at all in different that fire started, and those men types when the woods are soaking Those of you who have never seen never let that fire make another wet. Also, rate of spread is very fires like the Lost Johnny and Half major run. That is a record to shoot nearly the same in all types after Moon on the Flathead in 1929, the for; it has seldom if ever been several August days with the tem Freeman Lake on the Kaniksu and equaled in this region. perature at 100 °F (38 ºC), humidi the McPherson on the Coeur ty at 10 or 15 percent, and the d’Alene in 1931, and the Pete King The real difficulties and the most afternoon wind at 15 to 20 miles on the Selway in 1934, simply can frequent need of skill and under per hour (24–32 km/h). Hence, we not fully appreciate the significance standing by fire bosses come, how have used the principle of prepar and the danger under these condi ever, in judging gradations between ing for the average-bad in our fuel tions. It may be enough to point this explosive condition and that type classification, and the rates out that the Freeman Lake Fire, easiest of all conditions, when fire given on our fuel type maps are starting at 10:30 a.m. on August 3, will spread, but only so slowly that those to be expected on an average- 1931, exploded almost from the control is largely a problem of how bad day. This is about class 70 on start to cover 20,000 acres (8,100 to do it at the least cost. In between our burning index meter. You can ha) in the next 12-1/2 hours. This is this explosive condition and the not use those fuel type maps cor at the rate of 1,600 acres per hour easiest condition, other factors than rectly, or dependably, on any other (650 ha/h), from a standing start! stick moisture become more and basis. Both duff and 2-inch-diameter (5 more important and all the factors cm) sticks were down to 4 percent moisture content that afternoon. Wind was 13 miles per hour (21 Although fuel moisture is the critical variable for km/h) at 10 a. m., and 18 miles per making fuels explosive, wind velocity is often the hour (29 km/h) at 7 to 8 p.m. straw that breaks the camel’s back. Relative humidity was 10 percent
Fire Management Today 20 Our fuel type classes are therefore “Calculating the probabilities” means careful based on differences in rate of consideration of every available source of spread, not at the explosive point where we can do nothing about it, information concerning each of the basic factors but at combinations of moisture of fire behavior. contents, wind velocity, and vegeta tive conditions just short of the explosive point. These begin early Thermal Belt The next time you have a fire start in August whenever fuel moisture The major significance of this ther ing in late afternoon or early drops to 5 or 6 percent, the humid mal belt is that inside a certain alti evening about 1,000 feet (300 m) ity falls to less than 15 or 20 per tudinal zone burning conditions up from the main valley bottom, I cent, and the wind rises above its change less from daytime to night suggest that you note for your normal afternoon average of 6 or 8 time than they do in either the val selves whether or not that particu miles per hour (10–13 km/h). After ley bottoms or on the mountain lar fire does not run faster and for several days of such weather, espe tops. At Priest River this zone more hours during the night than a cially if the burning index rises to begins about 600 feet (180 m) similar fire in the valley bottom. 75, as it will with fuels under 5 per above the valley bottom and contin Also note whether or not that fire cent, humidity under 10 percent, ues upward for about 1,000 feet picks up and starts to run earlier in and winds of more than 10 miles (300 m). Below and above this zone the morning. I think you will find per hour (16 km/h), differences in fuels pick up more moisture at both of these conditions in almost rate of spread become less and less night than they do within it. Within all thermal belt fires. They are as all fuel types approach the explo the zone the fuels lose a little every essential elements in the equation sive condition. afternoon and pick up a few percent required to calculate the probabili between 6 p.m. and 3 a.m., but the ties. A burning index rating is therefore change is very slight. Up on the essential to calculation of the prob mountain top, however, the same These facts also should be highly abilities in any fuel type. If it shows fuels will pick up 4 percent more at significant to all fire dispatchers. class 65 to 75, you can count on night and lose 4 percent more in Other things being equal, more the differences shown by the fuel the daytime. Down in the valley men should be sent, and they type map, insofar as that map is bottom they will pick up and lose 8 should be speeded on their way well made. The weaknesses in these to 12 percent more than within the faster to every fire in the thermal maps are well recognized and steps thermal belt. This is true on both belt. Furthermore, on a going fire, are being taken to correct them. north and south aspects. The only if night work can be done on any places where it may not hold true sector, it should be planned first on In applying the burning index to a are in steep-sided, deep, east and those portions of the front from correct fuel type map, some guides west canyons like that of the 500 feet to 2,000 feet (152 to 610 have been worked out, but this is Salmon River. In that canyon and m) above the valley bottom, unfortunately a field in which our perhaps in a few other spots like it, because this is the zone of the ther fire research has been woefully the depth of the canyon and its ori mal belt. Within this zone you can weak. Our best contribution is in entation in relation to prevailing expect the least benefit from U.S.D.A. Circular 591, Influences of winds combine to interfere with increased fuel moisture at night. Altitude and Aspect on Daily normal air drainage. There the Variations in Factors of Fire thermal belt effect becomes less Wind Danger, by Lloyd Hayes, published pronounced or even disappears. Although fuel moisture is the criti in 1941. The outstanding new fact Sometimes going fires will also dis cal variable that puts all fuel types resulting from this research was rupt this belt, if the fires are large in an explosive condition, or the discovery and general location enough, but in most places and reduces them to an easy job of fire of what Hayes called the thermal under most conditions you should control, wind velocity is often the belt. calculate your probabilities on the straw that breaks the camel’s back. basis of the known difference of In fact, at fuel moistures of 6 or 7 burning conditions within this percent up to 20 or 25 percent, thermal belt. wind is often the variable which
Volume 64 • No. 1 • Winter 2004 21 Experienced judgment is the final determinant of what you actually do, both in planning to control a fire and out on the fire line where you try to put your plan into effect. finally determines what a fire will Another general law of wind behav been done you will still have to use do. Some basic research by Fons at ior during the ordinary fair weather judgment, and perhaps even do the California station has shown of June, July, and August, that is some pure guessing. Nevertheless, that with fuel moisture at 8 per quite well known, is that during the your batting average is absolutely cent, variations of wind velocity are day the wind usually blows up the certain to increase IF you first do more significant in changing the canyon or creek, while during the the best you can to calculate on the rate of spread than are variations in night it blows down canyon. This basis of facts and known principles. fuel temperature, fuel size, com reversal of direction in the evening pactness, or density. usually takes place just a few min Experienced Judgment utes after sundown. When both the Perhaps I should not close on this Whether or not some fire seasons daytime and the night winds are point; because if by doing that I are, as a whole, windier than others very light—less than 4 miles per cause you to discount any of the I do not know. But we do know that hour (6 km/h)—this reversal may things previously called to your wind is a result of certain meteoro not be of much significance. attention then I weaken my point. logical conditions which change However, in topography and on However, in fire control there are periodically at from 3- to 5- or 6 areas which are materially heated still a lot of basic factors not yet day intervals. If you will watch the by the sun’s rays, the afternoon understood or not yet measured. wind record portion of any fire dan wind created by rising hot air may And even when they are measured ger station chart, particularly for a amount to 8 or 10 miles per hour the basic facts must still be put lookout station, you will see a grad (13–16 km/h). When this is the together, weighted one against ual increase of wind for several case, reversals at sundown may pro another, and a balanced decision days, then a decrease, then another duce a significant down-canyon then reached. Worse yet, sometimes increase. Obviously, by watching wind. This condition is most pro that decision must then be modi this up and down trend you can nounced on south aspects and in fied or even seriously compromised definitely improve your calculation watersheds draining toward the on the basis of what you can do of the probabilities, even though south into a big canyon running about it. you cannot forecast precisely. east and west, like that of the Salmon River. But even under these Experienced judgment is therefore There are a few general rules of conditions, a large fire may create the final determinant of what you wind behavior which can be used such an updraft as to upset the nor actually do, both in planning to locally in Region 1. First is a dis mal reversal of wind. Hence, while control a fire and out on the fire covery, made by Hayes and this generality is worth considering line where you try to put your plan described in Circular 591, that the in your calculations there are other into effect. But if you will stop to places of greatest wind danger at factors which also must be recog examine just what is meant by night are, strange as it may seem, nized before you make your final experienced judgment, you will the north aspects at high altitudes. estimate of rate of spread. come back to the items I have list To put it another way: While you ed above. For what is experienced can usually count on the wind From what has been said it should judgment except opinion based on dying down during the night in the be clearly evident that “calculating knowledge acquired by experience? valley bottom, you should not the probabilities” means doing If you have fought forest fires in count on this if your fire is up on much, very much more than just every different fuel type, under all the high divides between major fight a fire with brute strength and possible different kinds of weather, drainages. Instead, at the higher numbers of men. It means careful and if you have remembered exactly elevations you should expect the consideration of every available what happened in each of these highest winds at night, not in the source of information concerning combinations of conditions, your daytime, and more wind on the each of the basic factors of fire experienced judgment is probably north aspects than on the south. behavior. But even when that has
Fire Management Today 22 very good. But if you have not Furthermore, you have some Conclusion fought all sizes of fires in all kinds knowledge of how both fuel mois Even though there are some holes of fuel types under all kinds of ture and wind velocity differ in our information, we have much weather then your experience does according to altitude and aspect. more than our predecessors. Those not include knowledge of all the The outstanding general differences men had to think of EVERYTHING. conditions. In that case, some of are known. Very few if any of the They even had to go to town to buy the facts and principles described most experienced old-time fire their axes and shovels and grub. above should be helpful to you. fighters knew these things. Then they had to remember out of their own personal experiences Summary And finally you have not only excel what the topography and timber There are only three things you can lent topographic maps to help you and brush types were like, up there do to stop a fire—rob it of its fuel, visualize your fire area, but you at the fire. Finally, they could only keep it from being heated to the have the major differences in fuel feel the wind and kick the duff to ignition point, or shut off the oxy types shown clearly so that you can see how dry the fuels were, right gen supply. where they stood. Finally they could look at the sky and guess at When it comes to fire behavior, Fire control still requires what the weather might be tomor there are likewise only a few basic headwork based on row. Maybe some of them prayed. variables. The big one is fuel mois knowledge. ture, and when our fuel moisture But times have changed. Where indicator sticks are below 5 percent those old timers had to guess at you can expect your fires to blow calculate what you should expect most everything, today, we have up and explode. As that moisture your fire to do on this particular measurements and maps and many content rises above 5 percent your slope in the next few hours. other facilities. While we might like fires become less and less explosive to have more, I doubt that anyone and you know that they are then It is true that you still have to esti ever will be able to sit down to a more and more influenced by mate how much different the fuel machine, punch a key for every fac another major variable, wind. moisture will be at your fire from tor of the situation, and have the what it is at the fire danger station. machine tell him what to do. Fire Both fuel moisture and wind are You also may still have to guess control still requires headwork measured every day of the fire sea what the exact wind direction and based on knowledge. If we will son at numerous stations. Those velocity will be on your fire even make a purposeful attempt to use measurements will show you clear after you find out what they are at all of the knowledge and all of the ly and accurately what the present the nearest ranger and lookout sta facilities that are availa1ble to us moistures and velocities are, and tion. And it is true that there may today we can do one thing the old how they are changing, whether be an acre of High-High fuel right timers could not do: We can come getting better or worse. These are near your fire even though the map mighty close to getting adequate facts. They are available to you. shows Medium-Medium or even fire control, and at an operating They were not available to the Low-Low. But if you have been on cost far below what it used to be. ■ rangers and supervisors who fought your district very many years and the fires of 1910 and 1919, nor to have gotten around, or if you have many men in 1928 and 1929. You someone else there who really therefore have this accurate knowl knows his fuels, you may be able to edge that those men did not have. pick up that important fact.
Volume 64 • No. 1 • Winter 2004 23 VERTICAL WIND CURRENTS AND FIRE BEHAVIOR*
John S. Crosby
orest fires are known to behave in a variety of ways, sometimes Turbulent, gusty winds affect fire behavior by F in quite unexpected ways. fanning the fire in spurts from varying directions, Prompt suppression requires that and by carrying heat and embers to fresh fuels. the fire boss, in estimating the probabilities of control within the allowable period, consider factors of the action of the wind above the high pressure centers with a slight affecting the behavior of the fire as immediate surface and which may drift outward, and counterclockwise well as those fixed by the site. have considerable effect on the fire. around low pressure centers with a slight drift towards lowest pressure. The important variables not deter General Characteristics At any fixed location the wind mined by the specific location are of Wind direction changes as the pressure the weather factors, primarily mois Wind is air in motion. The direc systems migrate and take up new ture and wind. Estimates of fuel positions in respect to that point. moisture and winds are made on tion of motion taken is almost the basis of weather forecasts, or unlimited. Near the ground the wind customarily blows in gusts Many reactions are superimposed through a knowledge of normal on the flow of air, particularly near daily variation and past experience and lulls, seldom as a steady even flow. Because it cannot be seen, it the ground, to modify the pressure based on observation of weather flow. Aloft the wind is stronger, and reactions in the locality. Often the can only be noted by its effect on various objects, and hence it is dif more steady, being changed only by weather forecast must be interpret strong reactions. ed in terms of local topography, or ficult to obtain a complete picture proximity to large water bodies, so of the variations that characterize air flow. Watching the drift of Up and down air currents may exist personal observation may be in the lower atmosphere in a great invaluable. smoke is one way to observe its motion; this is like observing some variety of intensities and steepness of rise or fall. Small eddies in a Although fuel moisture is an what similar currents in a river. Both water and air are fluid, light wind may be on]y a few feet in important factor, the purpose of depth, whereas strong convection this paper is to point out certain though water is more limited in its freedom of motion. The water currents may extend several miles wind features, particularly those in into the atmosphere, or the gentle which vertical currents are con swirls around and over rocks, makes eddies around land projec lift caused by a warm front may cerned, and to present a few gener amount to 10 feet in a mile (2 al rules for recognizing the proba tions, and tumbles over falls. It exhibits motion in many directions m/km), but extend over 1,000 miles bility of their existence. On the (1,600 km). ground, the best information avail besides down stream. Likewise, air moves in a turbulent fashion near able about wind is its surface veloc At ground level the wind tends to ity and direction, both of which the ground while still following a general course. parallel the surface; that is to say, may be constantly changing, because the wind cannot penetrate whereas little if anything is known The general flow of air is deter or go through the solid earth, its larger up-and-down currents must When this article was originally published, mined by the air pressure gradient John Crosby was a forester for the USDA and is modified by the effect of the change to a motion along the sur Forest Service, Lake States Forest face on reaching the surface, Experiment Station, St. Paul, MN. earth’s rotation and the friction caused by the passage of the air though the direction may be vari able, arid small eddies still persist. * The article is reprinted from Fire Control Notes 10(2) over the earth’s surface. The direc [Spring 1949]: 12–15. tion becomes clockwise around
Fire Management Today 24 Sustained vertical motion of the air increasing the air feeding into the moderately stable layers and join or is more prominent at some dis fire, the down-draft increases the set up vertical currents aloft, thus tance above the surface, where, of surface wind velocity, making it giving a new impetus to the fire, course, it is more difficult to more gusty and turbulent. causing it to flare up unexpectedly. observe. When a vertical motion, A study of large fires in relation to such as an eddy or convection cur Once started, convection currents air stability conditions aloft might rent, is superimposed on the exist may be accentuated or depressed by throw new light on unexpected fire ing wind, the result is alternately to the atmosphere, depending on its behavior, and provide a new tool for speed up and slow down the wind, condition of stability. If stable con better forecasting fire behavior. making it gusty and stronger. ditions exist (where the tempera ture decrease with elevation in the When there is marked air stability The stronger the horizontal wind, atmosphere is slight), the convec even during the daytime, the height the more turbulent it becomes in tion currents will be damped. to which convection currents may its passage over a rough surface, rise is of little consequence, and the thus creating stronger eddies and diurnal variation in wind is not more gustiness with frequent The stability of the air important. Thus, a strong daytime changes in direction. Turbulent, layers both near the wind may not die down much at gusty winds affect fire behavior by surface and aloft greatly night because it is driven by the fanning the fire in spurts from pressure gradient alone, and it will varying directions, and by carrying influences fire behavior. decrease only as the pressure gradi heat and embers to fresh fuels. ent decreases. However, in relatively unstable air These considerations are useful Convection Currents (where decrease of temperature only insofar as one is able to plan The motion of the air is also with elevation is great), convection for them and hence a few very gen strongly affected by the heat it currents are increased in speed and eral rules may be helpful. gains from the earth on sunny days. depth. Convection currents some Air in contact with the ground times rise to 8 or 10 miles (13–17 While the actual stability of the air then, because of the additional km) in the atmosphere and develop and the pressure gradient are basic, heat, becomes lighter than air great vertical velocities. above and tends to rise. The rising they are not subject to convenient observation at a fire. Indirectly, warm air sets up convection cur With night-time cooling, the air is however, the condition of stability rents. A forest fire also sets up such stabilized at low levels, and the shows itself in several ways. currents locally because of the convection currents subside. This intense heating of the air by the change is a part of the daily varia Cloud Formations. Cumulus type fire. tion in stability. In flat country the clouds are always an indication of wind then dies down. In mountain rising air currents, and often indi The earth’s surface is not uniformly ous country the wind stops flowing rectly of instability. In mountainous heated. Water surfaces are cooler up-slope and begins to flow down country, the rising currents may be than land, and forested land cooler slope. Along larger water bodies the due to lift over a ridge, while in than exposed soil or rocks, so the daytime landward breeze changes level country it is almost always a surface air is not of uniform tem at night to a seaward breeze. These result of convection if not associat perature. Warmed air tends to rise changes are normal only when the ed with a front. For these clouds to in streams usually localized over pressure gradient is weak. the warmer areas, or hills may help form there must be sufficient water to start the warm air upward. vapor present in the rising air so Influence on Fire that it is cooled to its saturation Behavior Down-drafts occur as complements point before the lift ceases. If the to up-drafts. Both currents have The stability of the air layers both cloud bases are low it is an indica their effect on a forest fire. While near the surface and aloft greatly tion of abundant moisture; if high, an up-draft in a favorable atmos influences fire behavior. Very large water vapor is scarce. This condi phere has the effect of pulling on fires generate intense heat and may tion is indicated at the ground sur the rising smoke column, thus enable the heated air to penetrate face by high or low relative humidi-
Volume 64 • No. 1 • Winter 2004 25 ty respectively. The height of the Thunderstorms with high bases may be dry cloud tops indicates the height to storms—the rain evaporates into the air before it which the convection currents extend, and shows also the stability reaches the ground, and hence lightning strikes of the air, as the currents do not are more dangerous. penetrate stable air layers. Flat- topped cumulus clouds, therefore, indicate stability aloft. extending to great heights with ble air the impurities are carried strong vertical currents. Thun aloft and away, while stable air Often, however, vertical currents derstorms with high bases may be traps impurities and holds them in exist without formation of cumulus dry storms, that is, the rain evapo a shallow layer of air. clouds as the water vapor content is rates into the air before it reaches so low that it cannot be carried the ground, and hence lightning Air Mass. Cool air masses follow high enough to condense. Under strikes are more dangerous. ing cold fronts during the fire sea such conditions, when the sky is son east of the Rockies tend to rap mostly clear, evaporation is speed- Stratiform clouds (fog-like clouds idly develop instability in passing ed, resulting in faster drying of or layer clouds) indicate stable con over warmer areas. This instability fuels. ditions at least at the level of the at first is not deep, but increases clouds, though stratocumulus may with time. The cool air is also dry When relative humidity is low and often form in turbulent surface air air, and visibility is good. It is usu temperature high, strong currents even though the turbulence is shal ally recognized as coming in with may exist to considerable elevations low. In general, the lower the stra fresh northerly winds. without clouds forming. A further tus clouds, if they persist, the more check can be made by watching the stable the air, and the less possibili Winds. Gusty winds with a notice rise of temperature during the ty of vertical currents. Low stratus able decrease in velocity at evening morning. A sharp rise early that clouds in the morning, however, indicate the possibility of strong flattens out and remains high sub often are a better indication of good convection currents during the day. stantiates the prospect of deep ver moisture conditions than of contin Turbulence and gustiness are more tical currents, assuming nearly ued stability during the day, for readily started in unstable air. Such clear skies. Small whirlwinds or they may have formed in a shallow gusty winds usually are accompa dust devils also indicate unstable layer of stable air that will rapidly nied by frequent changes in direc conditions, though they may exist change to an unstable layer during tion. The direction may vary only near the surface. the heat of the day. through 45 or more degrees rapid ly, back and forth, or more moder Thunderstorms and very large Visibility. Good visibility is often a ately within periods of an hour or ■ cumulus clouds indicate instability sign of unstable air in which verti so. cal currents may develop. In unsta
Russo and Schoemaker (1989) Decision Trap 2—Frame Blindness: Setting out to solve the wrong problem because, with little thought, you have created a mental framework for your decision that causes you to overlook the best options or lose sight of important objectives.* * See page 9.
Fire Management Today 26 WARNING SIGNS FOR FIRE FIGHTERS*
A.A. Brown
n 1949, 32 men died as a direct result of forest fires on national- We need to study the large fire from the point of I forest, State, and private lands. view of a local weather phenomenon. Most of them lost their lives because of extreme fire conditions which resulted in blow-ups. These rising and descending air currents 2. There is always danger in placing comments will be confined to these and a corresponding acceleration in men above a large fire and in special situations. the surface air circulation with fighting it from the head down effects similar to those of blowing in steep country. Wherever such Probably it is expecting too much fresh oxygen on a smoldering fire. strategy is necessary, lines of to make fire behavior experts of all retreat and places of refuge fire bosses. Nevertheless, we should In other situations unburned gases become a critical part of the go as far as we can in the interest seem to accumulate, then explode. responsibility of the fire boss. of safety and sound fire strategy. 3. Closely related to No. 2 is the Full analysis of such factors will fact that it is always hazardous to Large Fire Behavior require the help of competent attempt to outrun a fire uphill We need to study the large fire meteorologists and active participa when there is danger of being from the point of view of a local tion and close cooperation by both trapped. Nearly always there are weather phenomenon. As soon as research and administrative groups. safer alternatives. sufficient heat and sufficient area, This will be essential if we are to 4. Special precautions are needed from which heat is rising, have make significant new progress in in assigning men to special crossed a particular threshold, the foreseeing blow-up behavior. It can duties when they are detached fire takes on new potentials in be done. from the main crews or will oth behavior beyond those to be expect erwise be isolated for a time ed by simply extending the dimen from direct supervision and sions of a small fire. Sometimes we Every fire crew boss guidance by an experienced fire say “it begins to write its own tick needs to have a good man. et.” This is because of the air turbu knowledge of fire 5. The danger of being asphyxiated is often overlooked in selecting lence which is set up. Similarly, behavior if he is to be there is good evidence that local places of refuge. The bottom of a atmospheric conditions, beyond the left on his own gulch in the direction of spread already known effects of humidity responsibility. may become a chimney flue even and wind, play a big part. This though it has no fuel to burn, relates to the stability of the air at and most low places directly in the location of a fire. It seems rea Warning Signs the path of the head of the fire sonable, when an existing atmos In the meantime, here are some have such hazards. pheric inversion or ceiling gives warning signs to consider when Effects of ground cover— way under pressure of a mass of hot critical situations arise: air and gases from below, that there The fire front moves much more is a sudden acceleration in both the Manpower placement and safety— rapidly, through grass and open 1. Every fire crew boss needs to cover than through heavy timber. All experienced fire fighters realize When this article was originally published, have a good knowledge of fire A.A. Brown was chief of the Division of Fire behavior if he is to be left on his this, but they often underestimate Research, USDA Forest Service, the contrast in the rate of spread. Washington Office, Washington, DC. own responsibility. The alterna tive is close supervision and The fire perimeter can be expected explicit safety instructions by an to change from 2 to 10 times as * The article is reprinted from Fire Control Notes 11(3) rapidly on the sectors of a fire in [Summer 1950]: 28–29. experienced supervising officer.
Volume 64 • No. 1 • Winter 2004 27 that kind of cover. These two— To an experienced fire 3. The mouth of a canyon in rough cheatgrass and dry bunchgrass— fighter, dust devils are country is always affected by have extremely high rates of speed conflicting air currents. Any fire in steep country, even if the cover an ominous sign for in its close vicinity is likely to is sparse. It is well to recheck the blow-ups. reflect these air currents in its known ratios between contrasting behavior. The head of the fire in but intermingled fuel types and to such cases may not be the most impress them on trainees. ty, will be modified a great deal threatening. by rugged topography. 4. To an experienced fire fighter, Influence of weather and 2. Extremely rugged country is apt dust devils—those local whirl topography— to produce erratic behavior in winds of dust—are an ominous 1. Prevailing wind direction, partic any fire that has gained momen sign. Such whirls account for ularly if the wind is of low veloci tum because of the conflicting many blow-ups. ■ air currents that are set up.
WEBSITES ON FIRE* The Pulaski Project Corporation, the project has many Wildland Fire Developing a fully accessible, planned endeavors, including a Research National Wildfire Education Center world-class hiking trail to the Established in January 2001, the and Museum in Wallace or Nicholson mine (also known as Wildland Fire Operations Silverton, ID. The proposed center the Pulaski Tunnel) and rehabili- Research Group (WFORG) in will link the Pulaski story to the tating the adit itself are chief Hinton, Alberta, provides leader- challenges facing forest manage- goals of the Pulaski Project. ship in fire operational research ment and wildfire issues in 21st Founded in 2002, the project is and technology development. century America. designed to honor the memory The Website describes many of “Big Ed” Pulaski and other areas of research and develop- The project’s Website includes news wildland firefighters and to focus ment, including fire equipment and information on wildland fire attention on issues surrounding and protective clothing, fire history and management, along wildland fire management and management systems, and cur- with a jointly sponsored and mod- forest health. Now an action ele- rent operational issues for fire erated listserv discussion group on ment of the Greater Wallace managers. Research outputs are forest health, conservation, and fire Community Development intended to benefit firefighters, management. The site also provides fire managers, equipment manu links to fire season reports, fire facturers, and fire service agen related publications, and an abun * Occasionally, Fire Management Today briefly cies. The site also posts upcom describes Websites brought to our attention by the dance of other resource sites of ing wildland fire conferences and wildland fire community. Readers should not con- interest. strue the description of these sites as in any way symposiums in the United States exhaustive or as an official endorsement by the USDA Forest Service. To have a Website described, and Canada and links to other contact the managing editor, Hutch Brown, at Found at
Fire Management Today 28 RECOGNIZING WEATHER CONDITIONS THAT AFFECT FOREST FIRE BEHAVIOR*
Owen P. Cramer
iolent or erratic fire behavior often develops as a complete In stable air, both the intensity of the fire and the V surprise even to the more amount of spotting are reduced. Smoke will not experienced fire fighters. Such rise as high, and much drift smoke will remain in behavior usually is not completely explained and is frequently dis the lower layers. missed with the remark that the fire suddenly “blew up.” Unusual of temperature. In stable air, under Convective circulation into the base fire behavior is often closely related lying air is relatively cooler and of a fire and in the column of rising to certain weather conditions that heavier; overlying air is relatively hot gases above a fire is weak. Both can be recognized by visible charac warmer and lighter. If the tempera the intensity of the fire and the teristics. These weather conditions, ture decreases no more than 5 °F amount of spotting is reduced. In some of their characteristics, and per 1,000 feet increase in elevation stable air, smoke will not rise as their relation to fire behavior are in dry air, the air is stable. In high, and much drift smoke will described here. extremely stable air, temperature remain in the lower layers. The may actually increase with height. most common stability phenome The descriptions and terminology non is the inversion layer. used in this discussion agree with There are several indicators of sta definitions in the U.S. Weather ble air. Surface wind is steady or Inversion. An inversion is a hori Bureau Weather Glossary of 1946, frequently calm and smoke tends to zontal layer of air through which with two exceptions. These are fire lie in layers. Clouds are the stratus temperature increases with increas storm, which has been used in pub or stratified type showing no verti ing height. An inversion is the most lished accounts of fires started from cal motion (fig. 1). Visibility is often stable air condition. Inversion lay extensive incendiary bombings, and poor, particularly in the lower lay ers occur at any height and vary fire whirlwind, which is possibly ers. Ground and valley fogs form in greatly in thickness. As the ground used here for the first time. stable layers near the ground. Air in cools at night, a surface layer of air Weather conditions described are the lower layers is usually stable becomes colder than the air above divided into two major groups: phe during calm, clear nights, but and produces a surface inversion. nomena of stable air, of which only becomes unstable in midday when Surface inversions are most pro inversion is discussed; and phe heated by the warm ground. nounced in valley bottoms to which nomena of unstable air, including turbulent, convective, and whirling. Stable Air General Features. Stable air (sta bility) is air in which vertical motions are suppressed primarily because of the vertical distribution
When this article was originally published, Owen Cramer was a meteorologist for the USDA Forest Service’s Pacific Northwest Forest and Range Experiment Station.
* The article is reprinted from Fire Control Notes 15(2) [Spring 1954]: 1–6. Figure 1—Stable air.
Volume 64 • No. 1 • Winter 2004 29 cold air flows from surrounding Spot fires are more likely in unstable air because slopes. This type of inversion is of the more intense drafts in the fire and the readily dissipated by ground heat- ing during the day. greater vertical speed in the smoke column.
Since an inversion tends to sup in unstable air are the cumulus in stable air, resulting in a stronger press any vertical motion, its base type with pronounced vertical indraft at the base of the fire and a is frequently marked by: development and restricted hori hotter burning fire. Spot fires are zontal area (fig. 3). A deep layer of more likely because of the more • The flat top of a cloud or fog moist, unstable air may be marked intense drafts in the fire and the layer, by cumulonimbus clouds or thun greater vertical speed in the smoke • The common height at which ris derstorms. Instability at the cloud column. Unstable air is favorable ing cumulus clouds cease to rise, level does not necessarily mean for the formation of fire whirl and that this condition exists all the winds. These effects are discussed • The height at which a rising way to the ground. If it does exist, in more detail under the several smoke column levels off (fig. 2). it may be indicated by dust whirls instability phenomena described and erratic winds. below. There is often greater wind, or a shift in wind direction, above the Unstable air affects fires in several Turbulence. Turbulence is irregu inversion. An inversion near the ways. Spread of fires may be accel larity in air motion, shown by ground affects a fire in the same erated by gusty wind. The column bumpy air for the pilot and gusty way as stable air but to a greater of smoke over the fire will rise wind for the ground observer. Any degree. In the lower layers it tends faster and to greater heights than obstacle to the wind sets up to weaken drafts into and above a fire, thereby reducing the fire’s intensity and spotting potential. It has been suggested that flammable mixtures of gases liberated by a slow-burning fire might accumu late under a surface inversion, and that these might ignite and burn explosively. Unstable Air General Features. Unstable air (instability) is air that tends to turn over owing to relatively warm, light Figure 2—Inversion. air in the lower layers and relatively cooler, heavy air in the upper lay ers. The decrease in temperature with increasing height is greater than in stable air—5.4 °F or more per 1,000 feet in dry air. Vertical motions are accelerated. Upward and downward currents develop. Indicators are erratic surface winds with gusts and lulls, and a variation in direction and turbulence above the surface layers. Since smoke, dust, and haze are widely dispersed by mixing of high and low layers, visibility is generally good. Clouds Figure 3—Unstable air.
Fire Management Today 30 mechanical turbulence on the lee of the column of rising air. The ris vective circulation caused by a fire, ward side (fig. 4). Intermingled cur ing warm air above a continuing including both indraft at the base rents of rising warm and descend heat source is known as the con and updraft in the smoke column. ing cool air cause thermal turbu vective column. Above a fire this is The more intense the convective lence, which is characteristic of seen as the smoke column. circulation, the hotter and faster unstable air. Turbulence may be Cumulus clouds are convective the fire will burn and the higher accentuated by an uneven surface columns that have become visible embers will be carried. heating that varies with color of because of moisture condensation. soil, amount of shade, and type of The greater the instability of the air Thundersquall. A thundersquall is ground cover. or the greater the source of heat, the sudden wind that blows out the more intense becomes the con ward from beneath a thunderstorm. Gustiness. Gustiness is a charac teristic of wind in unstable or tur bulent air. Gustiness refers to sur face winds that vary rapidly in ver tical and horizontal speed and direction. Increasing instability and increasing turbulence caused by surface obstacles result in corre-
The more intense the convective circulation, the hotter and faster the fire will burn and the higher embers will be carried. sponding increases in gustiness. Since a fire greatly increases sur face instability, the intensity of gusts is likely to be greater in the immediate vicinity of a fire. Gusts Figure 4—Turbulence. usually cause a fire to spread spas modically in unpredictable direc tions. They also cause rapid fluctua tion in fire intensity and rate of spread.
Convection. A convection is motion in the air resulting from temperature differences in adjacent bodies of air. Convective currents are characteristic of unstable air. They consist of rising warm air and descending cool air currents (fig. 5). Heating at the ground either by the sun or by fire may initiate the upward current. Surrounding air descends and flows toward the base Figure 5—Convection.
Volume 64 • No. 1 • Winter 2004 31 Such a wind originates in the area Even a small fire whirlwind may produce of heaviest precipitation in a cumu considerable spotting and local intensification of lonimbus cloud, a convective cloud type that occurs in unstable, moist the fire. air. Air, cooled by precipitation, descends from the cloud and fans dred feet. Though usually not of a whirlwind that easily snaps off out at the surface (fig. 6). The destructive force, dust whirls can large trees. The diameter of its cir thundersquall usually occurs with a throw small debris several yards. culation may vary from 3 to 50 well-developed thunderstorm and The greatest speed is near the cen yards (2–45 m) or more. Fire whirl hits suddenly with speeds averaging ter, where a strong upward current winds encompassing whole fires 30 to 50 miles per hour (48–80 occurs. Dust whirls occasionally 1,000 yards (910 m) or more across km/h) for a period of several min form in the vicinity of fires and have been reported. Besides the utes. The thundersquall may occur move into the fire area, throwing rotating horizontal winds, there is a beneath a thunderstorm from sparks and embers in all directions strong vertical current at the cen which no precipitation reaches the and temporarily intensifying the ter, which may raise burning debris ground, and may extend outward a fire as they pass. to great heights. Even a small fire mile (1.6 km) or more ahead of the whirlwind may produce consider storm edge. Fire Whirlwind. A fire whirlwind able spotting and local intensifica is any whirlwind caused by a fire. tion of the fire. A central spout or These sudden, strong winds may The fire whirlwind may vary in tube may sometimes be present sweep a fire far beyond its confines intensity, from a small dust whirl to (fig. 7). Because of the wind and before the rainy section of the thunderstorm arrives. If the rain evaporates before reaching the ground, the fire may continue to burn unchecked.
Whirlwind. A whirlwind is any revolving mass of air, from the dust whirl to the hurricane. The torna do, a whirlwind associated with thunderstorms, is the most severe, though not the largest type. Whirlwinds are usually associated with extremely unstable air. Fires frequently make the nearby atmos Figure 6—Thundersquall. phere unstable and produce fire whirlwinds. Two types of whirlwind will be described, the dust whirl and the fire whirlwind.
Dust whirl. The dust whirl is the smallest type of whirlwind, fre quently known as a dust devil. Dust whirls indicate unstable air. They occur on sunny days with light sur face wind when the layers of air next to the ground become much hotter than the air immediately above. These whirls are usually 5 to 25 feet (1.5–8 m) in diameter and may extend upward several hun- Figure 7—Fire whirlwind.
Fire Management Today 32 In a fire storm, the surface draft into the base of the fire may be of destructive violence several hundred yards outside the fire. the resulting accelerated combus tion, fire whirlwinds are sometimes accompanied by a roaring noise similar to that produced by a rapid Figure 8—Fire storm. ly burning fire. Duration and behavior are variable. Fire whirls Fire Storm. A fire storm is a vio several hundred yards outside the may occur and recur where the lent convection caused by a large, fire. The fire storm, like other con combination of fire-produced insta continuous area of intense fire. vective phenomena, increases in bility, topography, and wind are This phenomenon was frequently intensity with greater atmospheric favorable. It is sometimes possible observed after extensive firebomb instability. Burning material may to dissipate a small, recurring fire raids in Europe and Japan. The be lifted several miles high. A fire whirlwind by cooling the part of convective system usually encom storm is not likely in the usual the fire over which it forms. passes the entire fire (fig. 8). The wildfire, where only the periphery surface draft into the base of the is actively burning. ■ fire may be of destructive violence
The Blowup Fire and Firefighter Safety* A review of fire fatalities through Hundreds have died from this strategy, more foremen trained the years focuses our attention source, if one considers the historic in handling men on fires, has on four major problems. fires of the past, such as Peshtigo. had much to do with the reduc- Losses of life are becoming fewer tion in the number of fire fatali- The greatest man-killer, of because of organized fire-suppres ties in recent years. But blow-up course, is the blow-up fire which sion efforts. Fast initial action with fires still constitute the worst almost yearly takes its toll. machine-age equipment such as potential killer. Much remains to planes, trucks, and tractors, a bet- be done before the problem is ter understanding of fire behavior, solved. * From Seth Jackson, “Death on the Fire Line (Fire Control Notes 11[3] [July 1950]: 26–27). more thorough planning of control
Volume 64 • No. 1 • Winter 2004 33 METEOROLOGICAL PROBLEMS ASSOCIATED WITH MASS FIRES*
DeVer Colson
eather plays an important role in the behavior of mass Many fires have been designated as “blow-ups” W fires. The knowledge and simply because of a lack of understanding of the understanding of the meteorologi factors controlling the behavior of these fires. cal conditions existing prior to and during these fires are essential for efficient fire fighting and control in ditions. These conditions are usual Surface Wind Patterns both urban and rural situations. ly combined with high surface tem The details of the surface wind pat Ordinary fires or even large fires peratures and low relative humidi which are burning and spreading in terns are necessary for efficient fire ties and often with strong surface fighting operations. These details a regular manner do not present winds. the major control problems. would include: Serious situations often develop One notable example involved the when what seems to be a routine • The actual local surface wind pat famous Chicago fire on October 8, terns; fire suddenly intensifies or begins 1871; and the associated fires in to spread at a greatly increased rate • the diurnal variations in these Wisconsin and Michigan on the patterns; and or changes its direction of spread same day, which burned over 1 mil abruptly. In forestry parlance, the • the dependence of these local pat lion acres (400,000 ha), including terns and their diurnal variations term “blow-up” or “explosive” has the entire town of Peshtigo, where been applied to such forest fires, on the surface pressure patterns, over 600 lives were lost. Weather as well as frontal and storm pas since these fires often seem to data indicate extreme dryness and explode. However, many fires have sages, the upper level weather strong winds on that date. On days patterns, atmospheric stability, been designated as “blow-ups” sim with less hazardous burning condi ply because of a lack of understand wind and temperature profiles, tions, these fires might well have and topography. ing of the factors controlling the been controlled before they had behavior of these fires. reached such disastrous propor A knowledge of the local wind pat tions. There is much to be learned both terns and their variations is even in identifying these factors and in more essential in areas of rugged In the preparation and planning for terrain. In these areas, there are forecasting the occurrence of these the fire bombing raids over the factors. Some of the possible mete the additional effects of general Tokyo area, weather conditions drainage patterns (mountain and orological factors will be discussed were studied in connection with briefly in this paper. valley winds) and the diurnal up- brush fires in North Carolina, a and downslope winds due to the region climatically similar to the differential heating of the slopes. General Burning Tokyo area. The following factors Conditions The relative influences of all these were used: precipitation, relative factors vary greatly with the Most serious fires occur with humidity, and maximum wind ruggedness of the terrain. extremely low fuel moisture caused speed. The maximum wind speed by severe or extended drought con- on the day of the fire was used, Two local wind surveys have been while the precipitation and relative conducted, one by the U.S. Weather When this article was originally published, humidity were weighted over the Bureau in 1949–52 at Oak Ridge, DeVer Colson worked for the U.S. Weather day of the fire and the three previ Bureau. TN, and the other by Operation ous days. These same factors are Firestop in 1954 at Camp used directly or indirectly in most * The article is reprinted from Fire Control Notes 17(1) Pendleton, CA. Unfortunately, [Winter 1956]: 9–11. current fire danger rating systems. much of the data from these sur
Fire Management Today 34 In the Mann Gulch fire, the unusual currents may have been due to the strong surface winds resulting from descending currents from the high-level thunderstorms in that area. veys cannot be applied to other through passes, strong local winds of these fluctuations have been areas because of the influences of will be set up as the air flows down found to be closely associated with the local terrain and weather condi the lee side. Examples of such the degree of atmospheric instabili tions. However, as data from addi strong local winds are the Santa ty. Also, the magnitude and fre tional surveys are accumulated, Ana winds in southern California, quency of these fluctuations will be more and more generalizations can the east winds in western Oregon greater at well-exposed sites than at be made that can be applied to and Washington, and the chinook well-sheltered locations. other areas. Such wind studies are winds on the east slopes of the Mechanical eddies and turbulence important in air pollution and Rocky Mountains. These winds have can be generated as air flows across smog control. a tremendous effect on fires, since and around sharp features of ter they are associated with high tem rain and buildings. It is the unusual cases that cause peratures and low relative humidi the most trouble. Some recent ties. Convection cases are the 1949 Mann Gulch fire Under certain atmospheric condi in Montana and the 1953 Upper Level Winds tions, better convection can be sus Rattlesnake and 1954 Sierra City As fires spread into the crowns of tained which will promote more fires in California. At each of these high trees, a different rate of spread efficient burning. These conditions fires, fire fighters lost their lives can be expected since the wind are usually associated with atmos when the fire spread rapidly in an speed and direction at this level pheric instability, that is, with near unusual and unexpected manner. In may be quite different from that or superadiabatic temperature lapse the Mann Gulch fire, the unusual near the ground. Also, with burn rates. However, the convection col currents may have been due to the ing buildings, the surface winds umn will not attain great heights if strong surface winds resulting from may have little connection with the the wind speed increases too rapid descending currents from the high- fire spread at higher levels. With ly with height. Too strong a wind level thunderstorms in that area. In convection currents carrying burn speed may cause the column to be the other two cases, the rapid ing embers up into even higher lev broken away from its energy spread of the fire may have been els, the rate and direction of the source. due to a combination of the normal spread of the fire due to spotting night downslope air drainage acting may be entirely different from that Temperature inversions tend to act simultaneously with a pressure gra which would be expected from just as a lid on free convection. dient across and through the pass a knowledge of the surface winds However, under these conditions, es. As more is learned about these alone. as the free air temperature reaches wind patterns, more of these a certain value or as the energy of unusual fires and their behavior Turbulence the fire becomes great enough, the patterns can be anticipated. In addition to the actual local wind convection can break through the inversion and can suddenly extend Topography patterns, the turbulence or the fluctuations in both the wind speed to much greater heights, especially With the proper pressure gradient and the direction must be consid if the atmosphere is unstable above across mountain ridges and ered. The magnitude and frequency the inversion. When this break through occurs, sudden changes will take place in the fire behavior As fires spread into the crowns of high trees, a and the spread. different rate of spread can be expected since the Much experimental and theoretical wind speed and direction at this level may be quite work is now in progress on the different from that near the ground. general problems of turbulence, dif-
Volume 64 • No. 1 • Winter 2004 35 fusion, convection, and allied prob- When a temperature inversion breaks, lems at many air pollution and sudden changes will take place in the fire behavior micrometeorological projects. and the spread. Thunderstorm and Lightning around and over the fire will be observed in the vicinity of some The high-level and often dry thun changed. A complete study of this large fires and may become quite derstorms present a great hazard in problem requires accurate and appreciable at times. the Rocky Mountain area because detailed data on temperature, of lightning fires. Project Skyfire humidity, wind speed and direction, Conclusion has been set up in the Northern and the composition of the gases in As more is learned about the mete Rocky Mountain area to study the the convection column. From these orological factors as well as a better origin, development, structure and results, it will be possible to deter knowledge of the fuel distribution intensity, movement, distribution mine the rate of transfer of heat, and efficiency of combustion, fewer of these storms, and the possibility momentum, and the distribution of fires will be designated as “blow of modification of these storms to energy about the fire. In addition to ups.” These fires can be anticipated reduce the lightning hazards. experimental studies at actual fires, and their behavior patterns expect much information has been gained ed. However, a vast amount of diffi Meteorological from model studies. cult experimental and theoretical Phenomena induced by work will be necessary to accom a Large Fire Strong indrafts, usually referred to plish this goal. ■ as the firestorm, have been Once a fire develops, the original wind and temperature distribution
Russo and Schoemaker (1989) Decision Trap 3—Lack of Frame Control: Failing to consciously define the problem in more ways than one or being unduly influenced by the frames of others.* * See page 9.
Fire Management Today 36 SOME PRINCIPLES OF COMBUSTION AND THEIR SIGNIFICANCE IN FOREST FIRE BEHAVIOR*
George M. Byram
lthough a large fire is essen tially a physical or meteorolog Heat makes combustion a chain reaction by letting A ical phenomenon, combustion gases distilled from the fuel react with oxygen in itself is a chemical chain reaction the atmosphere to give off more heat, raising the process, which takes place at high temperatures. In all forest fires, temperature of adjacent fuel. large or small, materials such as leaves, grass, and wood combine tion products are principally invisi monoxide burning. However, if with oxygen in the air to form com ble water vapor and carbon dioxide. combustion is not complete, small bustion products plus large quanti If combustion is not complete, amounts of carbon monoxide ties of heat. Heat, as we shall see, is some of the distilled substances will remain. In this phase the fuel is the most important combustion condense without being burned and burned as a solid, with oxidation product in fire behavior. remain suspended as very small taking place on the surface of the droplets of liquid or solid over the charcoal. Phases of Combustion fire. These condensed substances There are three rather definite are the familiar smoke that accom Even though the three combustion phases of combustion, although panies most fires. Under certain phases tend to overlap, they can be they overlap somewhat and all exist conditions, some of the water vapor plainly seen in a moving fire. First simultaneously in a moving fire. may also condense and give the is the zone in which leaves and First comes the preheating phase, smoke a whitish appearance. grass blades curl and scorch as they in which fuels ahead of the fire are are preheated by the oncoming heated, dried, partially distilled, and In the third or final phase the char flames. Next is the flame zone of ignited. In the second phase, the coal left from the second phase is burning gases. distillation of gaseous substances burned and leaves a small amount continues but is now accompanied of residual ash, which is not a com Following the flames is the third by their burning or “oxidation.” bustion product. If combustion is but less conspicuous zone of burn Ignition might be regarded as the complete and if the charcoal** is ing charcoal. Unless fuels dry to a link between the first, or preheat mostly carbon, the primary com considerable depth (that is, unless ing, phase and the second, or bustion product in this phase will the Build-up Index is high), this gaseous, combustion phase. be carbon dioxide because the ini last zone may be almost absent. If Ignition may also be regarded as tial water is driven off in the first this happens the burned-over area the beginning of that part of the two phases. Some carbon monoxide will appear black instead of gray, combustion process in which heat is formed as an intermediate prod which means that much of the is given off. The flames seen over a uct, which in turn burns as a gas to remaining charcoal, as well as some forest fire or in a fireplace are the form carbon dioxide. The small of the underlying fuel, has not burning of distilled gases; combus- blue flames appearing over the completely burned. With the excep coals in a fireplace are carbon tion of such years as 1947, 1952, When this article was originally published, and 1955, a blackened burned-over George Byram was a physicist for the area has been more common than a ** The composition of charcoal varies, depending on USDA Forest Service, Southeastern Forest the conditions under which it is formed. If the distilla gray ash-covered area in the Experiment Station. tion temperature is low, 400 to 500 ºF (204 to 260 ºC), the charcoal will contain considerable tar coke. Eastern and Southern States. However, in the rapid heating and resultant high tem * The article is reprinted from Fire Control Notes 18(2) peratures existing in a forest fire, the deposits of sec [Spring 1957]: 47–57. ondary products in the charcoal are probably low.
Volume 64 • No. 1 • Winter 2004 37 Heat of Combustion Convection, with some help from radiation, is the The heat of combustion is heat that principle means of heat transfer from a ground makes combustion a chain reac fire to the crowns of a conifer stand. tion. Heat supplied to unburned fuel raises its temperature to the point where the fuel, or the gases smoke, the combustion efficiency woody fuel if it burned with maxi distilled from the fuel, can react might be as high as 80 percent. mum efficiency. About 1 pound with the oxygen in the atmosphere Large fires burning with dense (0.45 kg) of pitch would accomplish and in so doing give off more heat. smoke would be less efficient. the same result. This in turn raises the temperature Combustion efficiency probably of adjacent fuel, and thus the drops somewhat with increasing The rate of heat release in a forest chainlike nature of combustion moisture content. fire can be visualized by comparing becomes established. it with a familiar rate, such as that Heats of combustion are given in required for house heating. For The heat energy released by burn British thermal units per pound of example, consider a hot, rapidly ing forest fuels is high and does not dry fuel. A B.t.u. is the quantity of spreading fire burning with a 20 vary widely between different types heat needed to raise the tempera chain (1,320-foot [400-m]) front and of fuels. Table 1 gives the heats of ture of 1 pound of water 1 ºF. For with a forward rate of spread of 50 combustion for a number of sub example, the above tabulation chains (3,300 feet [1,000 m]) per stances. These materials and heats shows with the help of a little arith hour. If the fire burns 6 tons of fuel were selected from tables in Kent’s metic that the burning of 1 pound per acre (13.4 t/ha), in 1 hour’s time Mechanical Engineers Handbook, (0.45 kg) of an average woody fuel enough fuel would be consumed to 12th edition. Their average is prob gives off enough heat to raise the heat 30 houses for a year if each ably a good approximation for for temperature of 100 pounds (4.5 kg) house yearly required the equivalent est fuels. Fuels do not ordinarily of water about 86 ºF. To raise the of 10 cords (25.5 m3) of wood weigh burn with maximum efficiency, so temperature of 100 pounds (4.5 kg) ing approximately 2 tons per cord the actual amount of heat released of water (about 12 gallons [45 L]) (0.7 t/m3). per pound of fuel in a forest fire from a temperature of 62 ºF (17 ºC) will be somewhat less than shown to the boiling temperature of 212 Occasionally there is a fire in the in the tabulation. For a small fire ºF (100 ºC) would require about 1.7 Eastern States with a rate of spread burning in dry fuels with very little pounds (0.76 kg) of an average exceeding 5,000 acres per hour (2,000 ha/h). If it burns in a dense, Table 1—Heat produced by various fuel types. continuous stand of conifers, which Heat of combustion might have 12 tons (10.9 t) or more of available fuel per acre, such a fire Substance Per pound, dry (B.t.u.) Per kg, dry (kJ) could consume enough fuel in an Wood (oak) 8,316 19,330 hour to heat 3,000 houses for a year. Wood (beech) 8,591 19,969 Heat Transfer Wood (pine) 9,153 21,275 There are three primary ways in Wood (poplar) 7,834 18,209 which heat travels or is transferred from one location to another. These Pine sawdust 9,347 21,726 are conduction, convection, and Spruce sawdust 8,449 19,639 radiation. Although dependent on Wood shavings 8,248 19,172 convection, there is a fourth or sec ondary means of heat transfer in Pecan shells 8,893 20,671 forest fires, which might be Hemlock bark 8,753 20,345 described as “mass transport.” This is the carrying of embers and fire Pitch 15,120 35,145 brands ahead of the fire by convec Average (excluding pitch) 8,620 20,036 tive currents and results in the familiar phenomenon of “spotting.”
Fire Management Today 38 As a heat-transfer mechanism, con Convection, with some help from slash and limbs and logs in blow- duction is of much greater impor radiation, is the principle means of down areas. Materials that are poor tance in solids than in liquids and heat transfer from a ground fire to conductors of heat, such as most gases. It is the only way heat can be the crowns of a conifer stand. Hot forest fuels, ignite more readily transferred within opaque solids. By gases rising upwards dry out the than do good conductors, but they means of conduction, heat passes crown canopy above and raise its burn more slowly. Although the through the bottom of a teakettle temperature to the kindling point. effects of conduction are far less or up the handle of a spoon in a Although convection initiates conspicuous than those of radiation cup of hot coffee. crowning, both convection and and convection, conduction is a radiation preheat the crown canopy very important factor in the com Convection is the transfer of heat ahead of the flames after a crown bustion process. by the movement of a gas or liquid. fire is well established. Convection For example, heat is transferred is also a factor in the preheating of Factors Affecting the from a hot air furnace into the the ground fuels in a surface fire Combustion Rate interior of a house by convection, but to a lesser extent than radia Many factors affect combustion in although the air picks up heat from tion. The effects of both radiation such complex ways that they are the furnace by conduction. and convection in preheating are not yet fully understood even for a simple gas or liquid fuel. Solid fuels Radiation is the type of energy one are even more complex. Even so, feels when sitting across the room Conduction is one of the there are two rather simple factors from a stove or fireplace. It travels main factors limiting the that have obvious and definite in straight lines like light, and it effects on the combustion rate of travels with the speed of light. combustion rate in heavy fuels, such as woody substances and are of great importance in forest fire suppres Most of the preheating of fuels slash and limbs and sion. The first of these is the mois ahead of a flame front is done by logs in blowdown areas. ture content of the fuel, and the radiation. For a fire that occupies a second is fuel size and arrange small area and can be thought of as ment. a “point” (such as a small bonfire considerably increased when a fire or a spot fire), the intensity of radi spreads upslope, because the flames It is difficult to overestimate the ation drops as the square of the dis and hot gases are nearer the fuels. effect of water on the combustion tance from the fire increases. For The opposite is true for downslope rate and, hence, on fire behavior. example, only one-fourth as much spread. Water in a fuel greatly diminishes radiation would be received at 10 the preheating rate in the first feet (3 m) as at 5 feet (1.5 m) from Convection and radiation can trans phase of combustion. Much of the the fire. However, when a fire fer heat only to the surface of heat is used in raising the tempera becomes larger, the radiation inten unburned (or burning) fuel. ture of the water and evaporating it sity does not drop off so rapidly. For Actually, radiant heat may pene from the fuel. The large quantities a long line of fire, the radiation trate a few thousandths of an inch of resulting water vapor dilute the intensity drops as the distance from into woody substances and this oxygen in the air and thus interfere the fire increases; that is, one-half penetration may be of some signifi with the second or gaseous com as much radiation would be cance in the burning of thin fuels, bustion phase. If the initial fuel received at 10 feet (3 m) as at 5 feet such as grass blades and leaves. moisture is high enough, water (1.5 m). For an extended wall of However, radiation, like convection, vapor may make the mixture so flame, radiation intensity drops off for the most part transfers heat “lean” that the gases will not burn. even more slowly. This tendency for only to the surface of fuel material, This dilution of the oxygen in the radiation to maintain its intensity and conduction may be considered air also affects the third or carbon- in front of a large fire is an impor the only means of heat transfer burning phase of combustion. tant factor in the rapid growth of a inside individual pieces of fuel. For Although data are lacking, it is fire’s energy output. this reason conduction is one of the probable that moisture reduces main factors limiting the combus considerably the heat yield or com tion rate in heavy fuels, such as bustion efficiency. This heat loss
Volume 64 • No. 1 • Winter 2004 39 would be in addition to that result- It is difficult to overestimate the effect of water on ing from the water-heating and the combustion rate and, hence, on fire behavior. evaporation requirements.
The effect of size and arrangement The effect of fuel arrangement can brand material available for spot of fuel on combustion can be illus be visualized if a volume of excel ting may be greater than the effect trated by the following example. sior like fuel, such as that just of fuel size and arrangement. This Consider a large pile of dry logs all described, is compressed until it point will be discussed in the sec about 8 inches (20 cm) in diameter. occupies a volume only 4 or 5 tion on applications to fire behav Although somewhat difficult to times that of the original volume of ior. start, the log pile will burn with a logs. The total burning surface and hot fire that may last for 2 or 3 radiative conditions remain the The Fire Triangle hours. The three primary heat- same as before compression, but The principles of combustion may transfer mechanisms are all at both convective heat exchange and be summarized in an effective way work. Radiation and convection oxygen supply are greatly reduced. by means of the fire triangle. This heat the surfaces of the logs, but There will be a corresponding triangle neatly ties together not only conduction can transfer heat decrease in fire intensity. only the principles of combustion inside the individual logs. Since but illustrates their application as conduction is the slowest of the Fuel size and fuel arrangement well. The three sides of the triangle three heat-transfer mechanisms, it have their greatest effect on the are FUEL, OXYGEN, and HEAT. In limits the combustion rate in this lower intensity fires and in the ini the absence of any one of these case. Consider now a similar pile of tial stages of the buildup of a major three sides, combustion cannot logs that have been split across fire. When a fire reaches conflagra take place. The fire triangle repre their diameter twice, or quartered. tion proportions, the effect on fire sents the basic link in the chain Assume that the logs are piled in an behavior of factors such as ignition reaction of combustion (fig. 1). overall volume somewhat greater probability and quantity of fire- Removing any one or more sides of than the first pile, so there will be ample ventilation. This log pile will burn considerably faster than the first one because the combustion rate is less dependent on conduc tion. The surface area was more than doubled by the splitting, so that convection and radiation are correspondingly increased in the preheating effects. The burning surface is also increased by the same amount.
Assume that the splitting action is continued indefinitely until the logs are in an excelsior state and occupy a volume 30 or 40 times as great as in their original form. Convective and radiative heat trans fer will be increased tremendously in the spaces throughout the whole fuel volume, and the combustion rate might be increased to a point where the fuel could be consumed in a few minutes instead of hours.
Figure 1—The fire triangle is the basic link in the chain reaction of combustion.
Fire Management Today 40 Fuel size and fuel arrangement have their greatest properties should be started first on effect on the lower intensity fires and in the initial small-scale fires. Such work might give essential fundamental informa stages of the buildup of a major fire. tion on the relation between the variables controlling the convection the triangle breaks or destroys the This is best illustrated by consider process. chain. Weakening any one or more ing the spatial structure of the two sides weakens the chain and dimin types of fires. The height of the sig Certain properties of the atmos ishes fire intensity correspondingly. nificant vertical structure of a low- phere, such as the vertical wind intensity fire can usually be ex profile and to a lesser extent the The purpose of all fire suppression pressed in tens of feet. This dis vertical temperature profile, appear efforts is to remove or weaken tance is usually small compared to to be the controlling factors in directly or indirectly one or more the surface dimensions of the burn extreme fire behavior if an exten sides of the fire triangle. Con ing area, so that in a physical sense sive area of plentiful dry fuel exists. versely, all conditions that increase the fire is “thin” or two-dimension A discussion of the atmospheric fire intensity operate in such a way al as far as volume structure is con factors is outside the scope of this as to greatly increase or strengthen cerned. On the other hand, the sig paper, but it may be well to exam the sides of the triangle, and hence, nificant vertical structure of a well- ine in some detail those phases of the chain reaction of combustion. developed conflagration may extend the combustion process that permit In a blowup fire the chain becomes thousands of feet into the air, and the atmospheric factors to exert so strong that it cannot be broken this dimension may at times exceed their maximum effect. by the efforts of man. This means the surface dimensions of the burn that when blowup conditions exist, ing area. Fire behavior is an energy phenom the only opportunity to break the enon, and its relation to the com chain is by early strong initial The height that smoke rises above, bustion process can be understood attack. or in the neighborhood of, a fire is by the use of four basic fuel factors not always a true indicator of the relating to energy: Application to Fire height of the active convection col umn above a fire. Smoke from a 1. Combustion period, Behavior 2. Critical burn-out time, It is more difficult to apply our small fire may reach a height of 1,000 feet or (300 m) more, but 3. Available fuel energy, and knowledge of ignition and combus 4. Total fuel energy. tion to the behavior of very high- active convection may reach only a few percent of this height.* intensity fires, sometimes referred This last factor is constant, or near to as conflagrations or “blowups,” ly so, for any given quantity of fuel than to the behavior of the more It is the three-dimensional struc ture of a large fire that causes it to per acre. The first three are vari frequent low-intensity fires. The ables which, even for any homoge ordinary fire behaves for the most take on storm characteristics which, in turn, produce behavior neous component in a given fuel part as one would expect from the type, depend on factors such as fuel principles or combustion. In a con phenomena that one could not expect by scaling upwards the moisture content and fire intensity. flagration or blowup, however, the A fifth fuel factor, the quantity of sides of the fire triangle are greatly behavior of a low-intensity fire. However, this does not mean that firebrand material available for strengthened by factors that are spotting, is more or less independ absent, or nearly so, in small fires. scale-model fires, including small fires in the laboratory under con ent of the other four and will be Although these factors work treated separately. through the basic combustion prin trolled conditions, would not be useful in preliminary convection ciples, they so greatly modify the The combustion period may be expected effects of the basic column studies. Probably experi mental work on convection column defined as the time required for a processes that a high-intensity fuel to burn up completely, and erratic fire cannot be considered as * Although it is too involved to discuss in a paper on depends primarily on fuel size, fuel a large-scale model of a low-inten combustion, the height of the convection zone depends arrangement, fire intensity, and on the rate of heat output of the fire, the wind speed, sity fire. the vertical wind shear, and the stability of the atmos fuel moisture. It may range from a phere.
Volume 64 • No. 1 • Winter 2004 41 few seconds for thin grass blades to affecting the behavior of the fire critical burn-out time. The avail several hours or longer for logs and front.* able fuel energy and fire intensity heavy limbs. Critical burn-out time will therefore drop as fuel moisture is defined as the maximum length From the standpoint of fire behav increases. For most fires there are of time that a fuel can burn and ior, a crown fire in a dense conifer some fuel components which do still be able to feed its energy into stand could have more available not burn because of their high the base of the forward traveling fuel energy than a fire in an area of moisture content; in other words, convection column; its magnitude heavy logging slash. However, these components may be regarded depends primarily on fire intensity unless large portions of a heteroge as having infinitely long combus or the rate of a fire’s energy output. neous fuel have very long combus tion periods. The available fuel energy is that tion periods, fuel size and fuel part of the total fuel energy which arrangement should not have as An increase in fire intensity can is fed into the base of the convec much influence on the behavior of greatly reduce the combustion peri tion column. For fuels with a com major fires as on smaller fires. In a od for those fuel components with bustion period equal to or less than major fire a larger proportion of the higher moisture contents. For the critical burn-out time, the the heavier fuels take on the char some components the combustion available fuel energy is equal to the acteristics of flash fuels. This is a period might be infinite for a low- total fuel energy. If the combustion combined result of the shorter intensity fire, but perhaps only a period is longer than the critical few minutes, or even less, for a burn-out time, then the available high-intensity fire. For example, in fuel energy is less than the total The purpose of all fire the high-intensity Brasstown fire fuel energy. Total fuel energy is suppression efforts is to on March 30, 1953, in South determined by the quantity of fuel Carolina, as well as in other large per acre and the combustion effi remove or weaken fires in the Southeast in the last ciency. If the combustion efficiency directly or indirectly one few years, green brush often is assumed to be constant, the or more sides of the burned, leaving blunt pointed terms “available fuel energy” and fire triangle. stubs. In a similar manner a reduc “total fuel energy” can be replaced tion of the combustion period from by the terms “available fuel” and infinity to a few seconds for green “total fuel.” combustion periods and longer conifer needles takes place when a critical burn-out times for the fire crowns. An example will illustrate how fire high-intensity fires. Nevertheless, behavior relates to the four preced fuel size and fuel arrangement con The fifth fuel factor, the quantity of ing quantities. Consider a fire tribute heavily to the rate of firebrand material available for spreading in an area of plentiful buildup of fire intensity, especially spotting, becomes increasingly heterogeneous fuel, a considerable in the early stages, and are there important as fire intensity increas part of which is in the form of fore an important part of the fire es. Equally important is the rela flammable logs and heavy slash and behavior picture. tion between surface fuel moisture the rest a mixture of smaller mate and the probability of ignition from rial such as twigs, pine needles, and Much of the effect of fuel moisture embers or firebrands dropped from grass. Assume that the critical can be interpreted in terms of the the air. This relation has not as yet burn-out time is about 20 minutes. four basic fuel factors. Because been determined experimentally, Those fuel components with a com moisture decreases the combustion but ignition probability increases bustion period less than 20 minutes rate, it increases the length of the rapidly with decreasing fuel mois will have an available fuel energy combustion period. This, in turn, ture—hence with decreasing rela equal to their total fuel energy. means that a smaller fraction of a tive humidity. We know that the However, logs and heavy limbs may heterogeneous fuel will have a ignition probability for most fire require several hours to burn out, combustion period less than the brands is essentially zero when fuel so their available energy may be moisture is 25 or 30 percent (on an comparatively low; they could still oven-dry weight basis). We also * Heat sources a considerable distance behind the main be burning after the fire had moved flame front could possibly have indirect effects on fire know that not only ignition proba several miles, so would not be behavior by slightly modifying the structure of the wind bility but also combustion rate is field.
Fire Management Today 42 In a blowup, the sides of the fire triangle are so greatly strengthened that a high-intensity erratic fire cannot be considered as a large-scale model of a low-intensity fire. greatest for oven-dry material. In face fuel in which firebrands fall series of burning experiments to addition, both of these phenomena and the fraction of the ground area measure some of the factors and in the lower moisture content covered by the fuels. their response to variables such as range appear to be considerably moisture content and fire intensity. affected by a change of fuel mois Fuel characteristics that make However, once this was done, the ture content of only a few percent. plentiful and efficient firebrands are classification system itself might be not definitely known. The material comparatively simple. Probably its The importance of the relation would have to be light enough to greatest value would be in estimat between fuel moisture and ignition be carried aloft in updrafts, yet ing the conflagration potential of probability in the behavior of large capable of burning for several min different fuel and cover types for fires can be illustrated by a hypo utes while being carried forward by different combinations of weather thetical example. Suppose that the upper winds. Decayed punky conditions. from the convection column over a material, charcoal, bark, clumps of large fire, 10,000 embers per square dry duff, and dry moss are probably There is an important difference in mile per minute are dropping in efficient firebrands. Leaves and the energy conversion process for a front of the fire. Suppose that the grass are more likely to be ineffi low-intensity fire and a high-inten surface fuel moisture content is cient firebrands except over short sity fire. In the “thin” or two- such that only 0.1 percent of these distances. dimensional fire, most of the ener firebrands catch and produce spot gy remains in the form of heat. At fires, thus giving only 0.1 spot fires The initial phases of the blowup the most, such a fire cannot con per square mile. On the other hand, phenomenon are directly related to vert more than a few hundredths of if we assume that the surface fuel the combustion process and the one percent of its heat energy into moisture is low enough for 5 per basic fuel factors. A decreasing fuel the kinetic energy of motion of the cent of the embers to catch, then moisture means higher combustion updraft gases and the kinetic ener there would be 500 spot fires per rates and shorter combustion peri gy of the convection column square mile. As they burn together, ods. There will, therefore, be an eddies.* On the other hand, a these spot fires would greatly increase in the available fuel ener major conflagration may convert 5 increase the rate of spread and gy, or available fuel, accompanied percent or more of its heat energy intensity of the main fire. by an increase in fire intensity. The into kinetic energy which appears increase in fire intensity lengthens in the form of strong turbulent Thus, relative humidity (working the critical burn-out time, which updrafts, indrafts, convection col through fuel moisture) has a two means a further increase in avail umn eddies, and whirlwinds which fold effect on rate of spread in cer able fuel. A cycle of reinforcement can carry burning material aloft. tain types of extreme fire behavior. is thus established which favors The efficiency of the energy conver First is the effect on fuel combus growth of fire intensity. As the sion process, and hence the kinetic tion rate and rate of spread of the intensity increases, the atmospheric energy yield, increases rapidly with ordinary flame front. This effect factors become increasingly impor increasing fire intensity. This is would be present on small and tant. It is at this stage that spotting brought about by the mutual rein large fires alike. Second is the effect and ignition probability may forcement action in the basic fuel in accelerating rate of spread and become dominant fire behavior fac fire intensity by increasing the tors. * Although a detailed discussion is outside the scope of this paper, energy conversion processes in a fire can be probability of ignition from falling studied by a thermodynamic procedure in which a large embers. This latter effect would be By using the basic fuel factors it is fire, like a thunderstorm, can be treated as a heat engine. The efficiency of a heat engine is measured by present only on fires where spot possible that a fuel classification the fraction of heat or thermal energy that can be con verted into the kinetic energy of motion. A two-dimen ting was abundant. Ignition proba method could be developed to clas sional fire has an efficiency as a heat engine that is very bility will also depend on other fac sify fuel in terms of expected fire nearly zero or, at the most, only a few hundredths of one percent. A major high-intensity fire has an efficien tors, such as the nature of the sur behavior. It would first require a cy as a heat engine that may reach 5 percent or more.
Volume 64 • No. 1 • Winter 2004 43 factors plus favorable atmospheric It is the three-dimensional structure of a large fire conditions. that causes it to take on storm characteristics. In addition to the difference in the energy conversion processes in the Heat is transferred by conduction, fuel factors, which are (1) combus two types of fires, there is an enor convection, and radiation. A fourth tion period, (2) critical burn-out mous difference in rate of energy means of heat transfer might be time, (3) available fuel energy, (4) yield. For example, there were peri defined as mass transport and is the total fuel energy, and (5) quantity ods in the Buckhead fire in north familiar phenomenon of spotting, of material available for spotting. Florida in March 1956 when the which becomes increasingly impor Such a group of factors might be rate of spread probably exceeded tant on high-intensity fires. used to classify fuels in terms of 8,000 acres (3,200 ha) per hour. expected fire behavior. The rate of energy release from this Fuel moisture has more effect on fire would compare favorably with the ignition and combustion If atmospheric conditions are such the rate of energy release from a process than any other factor. that one or more strong convection summer thunderstorm. columns can form, the following Low-intensity fires are essentially appear to be the main combustion Summary two-dimensional phenomena, and factors that determine the intensity Combustion is basically a chemical major high-intensity fires three- and rate of spread of a major fire: chain reaction that can be divided dimensional. The third dimension into three separate phases: of a high-intensity fire permits the 1. The quantity of available fuel conversion of part of its heat ener energy, or available fuel, per 1. Preheating and distillation, gy into the kinetic energy of acre. The magnitude of this 2. Distillation and the burning of motion, which changes the relative quantity depends on a reinforc volatile fractions, and significance of the various combus ing relationship between the 3. The burning of the residual tion factors and greatly modifies basic fuel factors. In turn, this charcoal. their expected effects. For this rea relationship is regulated primari son a high-intensity fire cannot be ly by fuel size and arrangement, For a forest fuel, ignition is the link regarded as a magnified version of a fuel moisture, and the intensity between phase 1 and phase 2 of the low-intensity fire. of the fire itself. combustion process. For most for 2. Quantity of firebrand material est fuels the heat of combustion is The relation of fire behavior to the per acre available for spotting. between 8,000 and 9,000 B.t.u.’s combustion process can be under 3. Probability of ignition from fire per pound on a dry weight basis. stood by the use of a group of basic brands dropping ahead of the main burning area. This proba bility depends on several factors, The initial phases of the blowup phenomenon are the most important of which is directly related to the combustion process and the the prevailing relative humidity determining the surface fuel basic fuel factors. moisture. ■
Fire Management Today 44 VORTEX TURBULENCE— ITS EFFECT ON FIRE BEHAVIOR*
James B. Davis and Craig C. Chandler
he fire wasn’t doing much until the air tanker went Vortex turbulence consists of a pair of miniature “T over, and then it spotted all whirlwinds trailing from the wingtips of any over the place,” complained the fire aircraft in flight. crew foreman.
Such reports have caused fire con tion between them tends to make How Important are trol officers to ask, “Can air tankers them move first downward, then Vortex Wakes? really cause erratic fire behavior?” outward along the surface of the The Flight Safety Foundation, Inc., The answer is yes—under some ground. conditions. The gremlin is “vortex reports: “In recent years, there have turbulence,” a pair of whirlwinds streaming out behind the wingtips. What is Vortex Turbulence? Vortex turbulence is a sheet of tur bulent air that is left in the wake of all aircraft. It rolls up into two strong vortices, compact fast-spin ning funnels of air, and to an observer on the ground appears to trail behind each wingtip (fig. 1). Because it moves out at right angles to the flight path, vortex tur bulence can be distinguished from propeller wash, which is largely localized to a narrow stream lying approximately along the flight path. Unfortunately, however, vortex tur bulence is usually invisible.
Under certain conditions the two vortices may stay close together, sometimes undulating slightly as they stretch rearward. The interac-
When this article was originally published, James Davis was a forester for the USDA Forest Service, Pacific Southwest Forest and Range Experiment Station; and Craig Chandler was a fire behavior specialist for the Forest Service, Forest Fire Research. Washington Office.
* The article is reprinted from Fire Control Notes 26(1) Figure 1—Low-flying spray plane. Note funneling effect of spray trailing behind each [Winter 1965]: 4–6, 16. wingtip. This is vortex turbulence.
Volume 64 • No. 1 • Winter 2004 45 been increasingly frequent reports The vortex in the form of a horizontal whirlwind by pilots encountering severe dis can cause sudden and violent changes in fire turbances of another airplane even when separated from it by distances behavior on calm days in patchy fuels. of several miles. There also are an increasing number of fatal acci wing than on top, some air tries to the more severe the turbulence will dents to lighter airplanes, resulting flow around the end of the wing to be. The B-17 is a heavier airplane from upsets near the ground or the lower pressure area. Because of than the PBY. Thus, when the vor structural failures which are being the flow around the tip, the main tex wake immediately behind a B ascribed to encounters with wakes stream—instead of flowing straight 17 is 29 m.p.h. (46 km/h), the of large airplanes. It is now general back across the top and bottom of lighter PBY’s vortex will be only 16 ly accepted that the only distur the wing—tends to fly inward m.p.h. (26 km/h) under the same bance which an airplane can pro toward the fuselage on the top of flying conditions, since both planes duce that is powerful and persistent the wing and outward on the bot have the same wingspan. enough to account for these inci tom. As a result, the air doesn’t “fit dents arises from the vortices together” at the trailing edge but How Does Air Tanker which trail from the wingtips of forms a vortex sheet that rolls up any airplane in flight.” Speed Affect into two large whirlwinds that trail Turbulence? from each wingtip (fig. 2). Ordinarily, vortex turbulence does It may seem surprising, but turbu lence is inversely related to air not pose any difficulties to fire con Is Turbulence the Same trol forces. But under special cir speed (fig. 3). cumstances vortex wakes may for All Air Tankers? cause a fire to act most unexpected Vortex severity and persistency vary Other factors being equal, an air ly. Line personnel should become with several factors. Most impor craft with a high wingspan loading familiar with the vortex problem tant are the type and size of the air at slow airspeed is the source of the and the situations where it is likely craft and the condition of the air. strongest vortices. In terms of air to affect fire behavior. Vortex turbulence is greatest when safety, one of the greatest hazards is produced by a large aircraft with a a heavily loaded aircraft flying at What Causes Vortex heavy wingspan loading. slow speeds before landing or after Turbulence? takeoff. Essentially, this is the con Thus, the heavier the aircraft or dition when an air tanker slows Vortex turbulence is a byproduct of payload per unit of wing surface, down for an accurate airdrop. the phenomenon that gives lift to an airplane. Air flowing the longer route over the top of the wing has to travel faster than the air flowing across the bottom in order to reach the trailing edge simultaneously. The difference in speed causes a dif ference in pressure between the top and bottom of the wing with a resultant upward force, or lift. If you want to demonstrate this effect, hold the back of a spoon in a stream of water from a faucet. The spoon will be pulled into the stream as soon as the water touch es it.
However, here is where the trouble starts. Since the air pressure is greater on the under surface of the Figure 2—Airflow over wing with distortion of flow and vortex formation.
Fire Management Today 46 The air tanker pilot should be aware of the 1. Wind can blow the vortices away problem his aircraft can cause through the effect from the drop area. For example, a 10-m.p.h. (16-km) wind can of vortex wakes on a fire. blow the vortices more than 800 feet (240 m) in the short time How Does Aircraft motion may be better termed “skid required to drop from 150 feet ding” than “rolling,” for at the (45 m). Height Affect 2. Vortices weaken rapidly with Turbulence? ground contact point the direction of rotation is opposite the core’s time. Under average air condi At high altitude, the two vortices lateral movement (see fig. 2). The tions, the turbulence may lose remain separated by a distance downward movement may require its danger potential in less than a slightly less than the aircraft’s only 10 seconds from a TBM flying minute. In rough air, the funnels wingspan. However, the interaction at 50 feet (15 m), but a minute or break up and weaken even more of the two causes them to drop. As more from the same aircraft flying rapidly. Calm air is the worst sit they approach within approximately at 150 feet (45 m). The time uation because it permits the a wingspan of the ground, they required for downward movement turbulence to persist for a longer begin to move laterally outboard is important for two reasons: period. from each wingtip. The lateral How Does Vegetation Affect the Vortex? Natural surfaces are more or less rough and, therefore, cause fric tional resistance to air movement above them. The rougher the sur face, the greater the friction. Timber, for example, has a much greater slowing effect on wind than does open grassland. Whereas a vortex turbulence is more like a horizontal whirlwind than what we normally think of as a wind, the same frictional considerations apply. A heavy stand of timber would dissipate most of the force of a vortex; the same vortex would be only slightly weakened in grass or scattered timber. How Do Vortex Wakes Affect Fire Behavior? Although there are many observa tions on the effect of vortex wakes on other aircraft, we have only two or three on forest fires. However, what is known about the vortex and about fire behavior can lead to some pretty good guesses.
Because wind tends to break up the Figure 3—Relation of vortex velocity to air tanker speed. The tanker’s altitude was 75 feet vortex and is normally accompa (23 m); vortices took about 15 seconds to reach the ground, where their velocities were nied by much natural turbulence, obtained.
Volume 64 • No. 1 • Winter 2004 47 the chances are that vortex turbu lence will probably be noticeable only on a calm day. Not only will the vortex wake be stronger on quiet days, but because the fire will usually be spreading slowly, the sudden air turbulence will be even more unexpected and potentially serious.
On the ground, the effect of vortex turbulence will be felt as a sudden gust which may last only a few sec onds or for up to half a minute. In litter, grass, or light brush the result will be a sudden but brief Figure 4—Wake from a DC-3 and pronounced vertical motion of the vortex. flareup or increase in local fire intensity and rate of spread. In over long distances by the upper brush or scattered timber. heavy timber or brush fuels with a winds. Only rarely would one 3. The air tanker is large or heavily continuous overstory, vortex turbu encounter a fire in patchy timber loaded. lence will usually not reach the and brush under precisely these 4. The air tanker is flying low and ground and so will have no notice weather conditions; yet this was slow. able effect on fire behavior. apparently the case on one well- documented fire in California in The air tanker pilot should be In patchy fuels, where timber or 1962. aware of the problem his aircraft brush is interspersed with open can cause. He may know the effect grassy areas, the effects of vortex Summary of vortex wakes on his or other air turbulence may be extremely seri Vortex turbulence consists of a pair craft, but may not know the effect ous. Although the vortex wake will of miniature whirlwinds trailing on a fire. He can abide by the fol not reach the ground beneath a from the wingtips of any aircraft in lowing rules during situations of timber canopy, it may in the open flight. The more heavily loaded the possible danger from vortex wakes: ings. Because the core usually aircraft, and the lower and slower it remains above ground, the true flies, the stronger the vortex turbu 1. Don’t fly parallel to the fireline wind direction at the surface is not lence will be and the more likely to more than necessary. parallel to the ground but slightly reach the ground. The vortex will 2. Keep high except when making upward (fig. 4). Thus, both flames be in the form of a horizontal the actual drop. and burning embers tend to be whirlwind with velocities up to 25 3. Ensure that ground crews are swept upward as well as outward. m.p.h. (40 km/h)—sufficient to alert to the presence of the air Thus, vortex turbulence, compared cause sudden and violent changes tanker and the pilot’s intentions. with a natural gust of the same in fire behavior on calm days in velocity, has a greater potential for patchy fuels. Acknowledgments triggering crowning and spot fires The authors have received technical because flames and embers are Wind, gustiness, and surrounding guidance from many sources but driven up into the crowns. high vegetation will tend to break are especially grateful to Richard C. up or diminish vortex intensity. Rothermel, Aeronautical Engineer, The most serious situation is calm Northern Forest Fire Laboratory; air on the ground but a light, The fire crew should be alert for Herbert J. Shields, Supervisory steady wind aloft. Under these con trouble when: Engineer, Arcadia Equipment ditions the vortex may be carried Development Center; and Alan W. far from the aircraft to strike the 1. The air is still and calm. McMasters, Operations Analyst, ground in an unexpected location, 2. The fire is burning in open Pacific Southwest Forest and Range with ember showers being moved Experiment Station. ■
Fire Management Today 48 THE CONCEPT OF FIRE ENVIRONMENT*
C.M. Countryman
ebster** defines “environ ment” as “the surrounding Fire environment is the complex of fuel, W conditions, influences or topographic, and airmass factors that forces that influence or modify.” influences or modifies the inception, growth, This definition applies to “fire envi and behavior of fire. ronment” very well. For fire envi ronment is the complex of fuel, The amount of heating of fuel by The north slope also has its distinc topographic, and airmass factors the sun on a slope depends partly tive diurnal trend (fig. 2D). The that influences or modifies the on aspect. A slope facing east data illustrated in figure 2 were inception, growth, and behavior of begins to warm first, and its maxi obtained from observations taken fire. mum temperature occurs early in on a clear day on 45-degree slopes the day (fig. 2A). A slope facing early in July at 42° N. For a differ Fire environment may be repre south reaches its maximum tem ent combination of cloud cover, sented by a triangle (fig. 1). The perature about 2 hours later, and it slope, time of year, and latitude, a two lower sides of the triangle rep is higher than the maximum of the different pattern would be observed. resent the fuel and topographic east-facing slope (fig. 2B). A slope This differential heating of different components of fire environment. facing west reaches its maximum aspects affects the probability of fire The top side represents the airmass temperature still later, and this starts, and also fire growth and component; this is the “weather” maximum is higher than those of behavior. part of the fire environment. the east and south slopes (fig. 2C). Interrelationships of Components Fire environment is not static, but varies widely in horizontal and ver tical space, and in time. The fire environment components and many of their factors are closely interrelated. Thus, the current state of one factor depends on the state of the other factors. Also, a change in one factor can start a chain of reactions that can affect the other factors.
For example, consider the simple topographic factor of slope aspect.
When this article was originally published, C.M. Countryman was a research forester for the USDA Forest Service, Pacific Southwest Forest and Range Experiment Station.
* The article is reprinted from Fire Control Notes 27(4) [Fall 1966]: 8–10. ** Webster’s Third International Unabridged Dictionary Figure 1—Fire environment may be represented by a triangle. Each side represents a (1961: G. & C. Merriam Co.), p. 760. component of fire environment.
Volume 64 • No. 1 • Winter 2004 49 When the surface of a slope is heat tive or urban, on the surface is almost the only source of heat. This ed, it transmits this heat to the air affected by airmass conditions such energy heats the earth’s surface and above it by conduction, convection, as clouds, moisture content, and to a minor extent the air above it. and radiation. The resulting windspeed. Most of the energy that directly and increase in air temperature changes indirectly modifies the airmass and the relative humidity. In addition, fuel components of fire environ local winds also are often strongly Because fire behavior ment comes from the heated earth affected by the differences in air and fire environment surface. Because of differences in temperature resulting from the dif are interdependent, slope, aspect, and ground cover, ferential heating of slopes of differ heating by the sun is not uni ent aspects. These winds are further changes in one will form—some areas become much modified by the configuration of cause changes in the warmer than others. This variation the topography and by the surface other. in the local heat sources creates the fuels. Since the moisture content of variability in local weather and fuel fine dead woody fuels depends pri conditions. marily on the relative humidity of Fire and Fire the air, the differences in heating of Environment Perhaps we can most simply con slopes can affect both fuel moisture Where does fire fit into this pic sider fire as just another local heat content and fuel temperature. The ture? In an environment without source. As a heat source it reacts amount of heating of fuels, vegeta fire, radiant energy from the sun is with its surroundings in the same
Figure 2—Relationship of temperature to time of day on 45-degree slopes facing in four directions: A, east; B, south; C, west; and D, north. Data were taken on a clear day in early July at 42° N.
Fire Management Today 50 way as other local heat sources: A fire burning under a dense timber stand is interacting with the airmass to cre burning in a closed environment that may be ate changes in local weather, and with the fuel to a modify fuel mois much different than the more open environment ture and temperature. Because of above or outside the stand. the high temperatures in a fire, however, the reaction can be much more violent. By adding fire to the 1. Closed environment, and For example, a fire burning under a center of the fire environment tri 2. Open environment. dense timber stand (fig. 3C) is angle (fig. 1), this symbol becomes burning in an environment that the fire behavior triangle. It is the Inside a building, for example, the may be much different than that current state of each of the envi fire environment is nearly inde above or outside the stand. Fuel ronmental components—topogra pendent of outside conditions. Fuel moisture is often higher, daytime phy, fuel, and airmass—and their characteristics are determined by temperature is lower, and wind- interactions with each other and the construction of the building speed is much slower. In this situa with fire that determines the char and by its contents. The climate tion the fire is burning in a closed acteristics and behavior of a fire at and, hence, the moisture content of environment. any given moment. the hygroscopic fuels are controlled by the heating and cooling systems. If the fire builds in intensity and Fire Environment Air movement and topographic breaks out through the crowns of effects are nearly nonexistent. This the trees (fig. 3D), it is burning in Patterns is confined or “closed” environ an open environment and can come Because fire behavior and fire envi ment. However, the environment under an entirely different set of ronment are interdependent, outside buildings is not confined. controls. Fire behavior and charac changes in one will cause changes Current airmass characteristics teristics can change radically. in the other. To understand or pre vary with the synoptic weather pat dict fire behavior, we must look at terns and local conditions. Wind Open and closed environments the fire behavior and the fire envi movement and topographic effects exist in other fuels as well as tim ronment at all points of the fire. prevail. This is “open” environ ber, such as grass and brush. Thus, both fire behavior and fire ment. Because of the short vertical extent environment are pattern phenome of these fuels, the probability of fire na. Fire burning inside a building is burning entirely in a closed envi controlled by the fire environment ronment is much less. But the The scope of the fire environment within the building. The outside closed fire environment in a fuel depends primarily on the size and environment has little effect. As bed influences fire behavior, even if characteristics of the fire. For a long as the fire remains within the only part of the fire is burning in a very small fire, the environment is building (fig. 3A), there can be no closed environment. a few feet horizontally and vertical spread to adjacent fuel elements. ly. For a large fire, it may cover The fire is confined. The most obvious use of the con many miles horizontally and extend cept of fire environment and fire thousands of feet vertically. An If the fire breaks out of the build behavior patterns is probably in intensely burning fire will involve a ing, it is no longer burning in a understanding and predicting wild larger environmental envelope than closed environment. Outside condi fire behavior, but the concept can one burning at a lower combustion tions can influence its behavior, also be used in prescribed burning. rate. and the fire can spread to other fuel In fires of low or moderate intensi and grow in size and intensity (fig. ty, which are usually desired in pre Open and Closed Fire 3B). scribed burning, the fire environ Environments ment pattern largely controls the From a fire behavior standpoint, Closed and open environments also behavior pattern. Thus, by knowing fire environment can be separated exist in wildland fuels; however, the the fire environment pattern for into two general classes: boundaries between the two envi the area, the fire behavior pattern ronments are not as clear as they can be predicted. And by selecting are in urban areas.
Volume 64 • No. 1 • Winter 2004 51 Figure 3—These fires are burning in the following fire environments: A, closed urban; B, open urban; C, closed wildland; D, open wildland. the proper environment pattern, the desired type of behavior can he For a prescribed fire, by knowing the fire obtained and dangerous points can environment pattern for the area, the fire behavior be alleviated. pattern can be predicted. Summary Fire environment is the complex of Fire environment is not static, but of fire burning in a closed environ fuel, topographic, and airmass fac varies widely in space and time. ment may be vastly different from tors that influences or modifies the Both fire environment and fire one burning in an open environ inception, growth, and behavior of behavior are pattern phenomena, ment. The concept of fire environ fire. It is the current state of these and both patterns for the area of ment and fire behavior patterns is factors and their inter-relationship the fire must be considered in useful for the understanding and with one another and with fire that order to understand and predict a prediction of fire behavior for both determines the behavior and char fire’s behavior. wildfires and prescribed fires. ■ acteristics of a fire at any given moment. Because of the difference in the fire environment patterns, the behavior
Fire Management Today 52 GET THE MOST FROM YOUR WINDSPEED OBSERVATION*
John S. Crosby and Craig C. Chandler
urface windspeed is often the most critical weather element Peak windspeeds that persist for 1 minute can S affecting fire behavior and fire affect gross fire behavior, including rate of spread danger. It is also the most variable and fire intensity. and, consequently, the hardest to evaluate. ior, including rate of spread and fire Thermal turbulence occurs when What Is Gustiness? intensity. For example, a surface horizontal wind meets convective Air moving across the surface of fire in pine litter spreading at 10 currents produced by unequal heat land is constantly changing speed chains (660 feet [201 m]) per hour ing or cooling at the ground. Its and direction. Standing still, one with the wind averaging 5 miles per magnitude depends mostly on observes a series of gusts and lulls. hour (8 km/h) would spread 11 feet topography, ground cover, solar Because of gusts, trying to measure (3.3 m) farther than expected dur radiation, and atmospheric stability. windspeed is much like trying to ing a minute when the wind was The maximum thermal turbulence measure the speed of a car on a blowing at 9 miles per hour (14 occurs above rough topography winding mountain road. It slows on km/h). During that minute it would with patchy ground cover during the turns, speeds up on the burn with twice its average intensi sunny afternoons in unstable air. straightaways, and slows to a crawl ty and would be nearly three times on bumpy stretches. To obtain a as likely to jump a prepared fireline. Gustiness Problem reliable average speed, one must Gustiness is a serious problem for determine the time required to Momentary gusts have little effect both fire researchers and fire-con travel at least 2 miles (3.2 km). And on the overall rate of fire spread trol planners. Because of gustiness, the rougher and more crooked the and intensity, but they do produce wind measurements at two loca road, the longer is the distance large fluctuations in flame height tions cannot be compared unless required to obtain a reliable aver and can easily trigger crowning or they are taken at the same height age. This same principle applies to throw showers of sparks across the above the ground and for the same wind measurements. The greater fireline when other weather factors length of time. For maximum com the gustiness (the ratio between the are in critical balance. Gusts will parability, measurements should be range in momentary windspeeds usually be close to the average taken as high above the ground as and the average speed), the longer value and will rarely exceed the possible and for as long as possible. it takes to determine a reliable maximum value. But high towers and long observa windspeed. tions are expensive. Therefore, for Gustiness is caused by mechanical fire-danger rating we have estab Peak windspeeds that persist for 1 and thermal turbulence. lished a standard anemometer minute can affect gross fire behav- height of 20 feet (6.1 m) and a stan Mechanical turbulence is produced dard observation time of 10 min When this article was originally published, by friction as the air flows over the utes. John Crosby was a research forester for the ground surface. Its magnitude USDA Forest Service, North Central Forest depends on the height above the Experiment Station, Columbia, MO; and While these standards are fine for Craig Chandler was Forest Service ground where measurements are fire-danger rating, they often con Assistant Chief, Forest Fire Research made, the roughness of the ground fuse the firefighter on the ground. Branch, Division of Forest Protection surface, and the windspeed. The Research, Washington, DC. Rapid changes in fire behavior are maximum mechanical turbulence determined by rapid changes in the is found close to the surface in wind blowing on the burning fuel, * The article is reprinted from Fire Control Notes 27(4) [Fall 1966]: 12–13. rough topography on windy days. and not by changes in the long-
Volume 64 • No. 1 • Winter 2004 53 term average windspeed 20 feet (6.1 Often the firefighter must estimate the variations m) above ground. Often the fire in windspeed that may be expected for the fighter loses confidence in his meteorologist or his weather sta average speed that is reported. tion, or both, because he is told to expect a 16-mile-per-hour (26 Table 1—Wind gust estimating table. km/h) wind and found the fire fanned by 35-mile-per-hour (56 Standard Probable Probable momentary gust speed km) gusts. He often must estimate 10-minute maximum the variations in windspeed that average 1-minute speed Average Maximum may be expected for the average mph km/h mph km/h mph km/h mph km/h speed that is reported. 1 1.6 3 4.8 6 9.7 9 14.5 Tool for Estimating 2 3.2 5 8.0 8 12.9 12 19.3 Gustiness 3 4.8 6 9.7 11 17.7 15 24.1 4 6.4 8 12.9 13 20.9 17 27.4 To help firefighters estimate gusti ness, we determined the 10-minute 5 8.0 9 14.5 15 24.1 18 29.0 average speed, the probable fastest 6 9.7 10 16.1 16 25.7 20 32.2 1-minute average speeds, and the 7 11.3 11 17.7 17 25.7 21 33.8 probable average and highest 8 12.9 12 19.3 19 30.6 23 37.0 momentary speed or gust during 9 14.5 13 20.9 20 32.2 24 38.6 the fastest 1-minute speed (table 1). 10 16.1 14 22.5 22 35.4 26 41.8 The table values were determined 11 17.7 15 24.1 23 37.0 27 43.5 from several hundred noon and afternoon observations made at 12 19.3 17 27.4 25 40.2 29 46.7 Salem, Missouri, during fire sea 13 20.9 18 29.0 26 41.8 30 48.3 sons. They were taken when gusti 14 22.5 19 30.6 28 45.1 32 51.5 ness was likely to be greatest, as it 15 24.1 20 32.2 29 46.7 33 53.1 often is on difficult fires. Thus, the 16 25.7 21 33.8 30 48.3 35 56.3 estimates are most accurate when 17 27.4 22 35.4 32 51.5 36 57.9 they are needed the most. 18 29.0 23 37.0 33 53.1 38 61.2 It is difficult to convert windspeeds 19 30.6 24 38.6 34 54.7 39 62.8 taken by firefighters to the standard 20 32.2 25 40.2 35 56.3 40 64.4 windspeed. In preparing spot fore 21 33.8 26 41.8 37 59.5 42 67.6 casts for project fires, wind meas 22 35.4 27 43.5 38 61.2 43 69.2 urements are often made with a 23 37.0 28 45.1 39 62.8 44 70.8 hand-held anemometer. This 24 38.6 29 46.7 40 64.4 46 74.0 instrument indicates gust speed accurately, but it is almost impossi 25 40.2 30 48.3 41 66.0 47 75.6 ble to accurately determine average 26 41.8 31 49.9 43 69.2 49 78.9 speed with it. Consequently, the 27 43.5 32 51.5 44 70.8 50 80.5 windspeed reported from the fire- 28 45.1 33 53.1 45 72.4 51 82.1 line almost invariably is the average 29 46.7 34 54.7 46 74.0 53 85.3 gust speed rather than the accepted 30 48.3 35 56.3 47 75.6 54 86.9 20-foot (6.1-m), 10- minute stan dard. Therefore, another table was Note: All readings were taken in the afternoon 20 feet (6.1 m) above the ground.
Fire Management Today 54 Table 2—Standard windspeed estimates based on maximum gusts a Gustiness is a serious Fastest gust Standard windspeed when problem for both fire observed on atmospheric condition is: researchers and fire- hand-held c d e control planners. anemometer b Stable Neutral Unstable mph km/h mph km/h mph km/h mph km/h 0–3 0–4.8 0 0 0 0 0 0 developed to convert gust speed 5 4–6 6.4–9.7 1 1.6 1 1.6 1 1.6 feet (1.5 m) above the ground to 7 11.3 2 3.2 1 1.6 1 1.6 the standard 20-foot (6.1-m), 10 8 12.9 2 3.2 2 3.2 1 1.6 minute speed for stable, neutral, and unstable conditions (table 2). 9 14.5 3 4.8 2 3.2 2 3.2 This conversion should be used 10 16.1 4 6.4 3 4.8 3 4.8 when fire-danger indexes are deter 12 19.3 6 9.7 4 6.4 4 6.4 mined from fireline observations or 14 22.5 8 12.9 6 9.7 5 8.0 when wind information consists of 16 25.7 10 16.1 8 12.9 7 11.3 a mixture of hand-held and tower ■ 18 29.0 12 19.3 9 14.5 8 12.9 observations. 20 32.2 15 24.1 11 17.7 10 16.1 22 35.4 17 27.4 13 20.9 12 19.3 24 38.6 19 30.6 15 24.1 14 22.5 26 41.8 22 35.4 17 27.4 16 25.7 28 45.1 24 38.6 19 30.6 18 29.0 30 48.3 27 43.5 21 33.8 20 32.2 32 51.5 29 46.7 23 37.0 22 35.4 34 54.7 32 51.5 25 40.2 23 37.0 36 57.9 34 54.7 27 43.5 25 40.2 38 61.2 37 59.5 29 46.7 27 43.5 40 64.4 39 62.8 31 49.9 29 46.7 a. Standard windspeed is 10-minute average speed 20 feet (6.1 m) above the ground b. Readings were taken 5 feet (1.5 m) above ground. For best results observations should be made for several minutes. c. This column usually should be used for observations between 8 p.m. and 8 a.m. d. This column usually should be used for observations between 8 a.m. and noon, and between noon and 8 p.m. on overcast days. e. This column usually should be used between noon and 8 p.m. on clear or partly cloudy days.
Russo and Schoemaker (1989) Decision Trap 4— Overconfidence in Your Judgment: Failing to collect key factual information because you are too sure of your assumptions and opinions.* * See page 9.
Volume 64 • No. 1 • Winter 2004 55 ATMOSPHERIC STABILITY FORECAST AND FIRE CONTROL*
Rollo T. Davis
Unstable air masses increase chances of big fires. Relative Most large fires occur when the temperature humidity seems to play a smaller profiles through the lower levels of the role than thought before. atmosphere exhibit some degree of instability. Atmospheric stability forecasts, projecting stability for 36 to 48 hours, can warn fire control per face winds. Papers have been pre Relation of Air Turbulence to sonnel when to expect erratic fire sented about the relationship Erratic Fire Behavior in the behavior and an increase in blow between relative humidities below Southeast.”† In this paper, they up potential.** 30 percent and large fires. Daniel J. pointed out the possibility of a Kreuger, former Georgia Fire direct relationship existing between ave you ever wondered why Weather Supervisor, made a study unstable low-level air and extreme some forest fires are extremely of forest fires in Georgia for the fire behavior in the Southeast. H difficult to control while oth years 1950–59. He reported in the ers, under seemingly like weather Georgia Forest Research Paper #3 A review of the weather conditions and fuel conditions, are relatively that 77 percent of the fires burning at the time of the larger fires easy to curb? Even during dry peri 300 acres (121 ha) or more occurring in Mississippi during ods when winds are high and occurred when the relative humidi 1967 revealed that large, hard-to humidities low, some fires show no ty was 25 percent or less. Ninety- control fires did not necessarily erratic behavior or blow-up poten two percent of the large fires occur on the days with the lowest tial and are easily checked. But at occurred when the relative humidi relative humidities. In fact, the other times, under apparently the ty was 30 percent or less. Mr. largest fires occurred 24 to 48 same conditions, the wildest blow Kreuger concluded: hours after a day with desert-like up develops. Still more puzzling is humidities. This pattern seemed to the fact that some fires are almost 1. Fires, when promptly and ade be begun by the passage of a cold impossible to control and become quately attacked (barring equip front. With cold, dry, continental conflagrations even though the soil ment failure), rarely, if ever, arctic air overspreading the State is wet, humidities are relatively become large unless the relative behind the front, the relative high, and surface winds outside the humidity is 30 percent or less at humidities often dropped below 20 fire zone are light. Why the differ the fire. percent. One to 3 days later, relative ence? 2. Potential for large fires increases humidities started climbing, but rapidly as humidities fall below fire severity and size also increased. Blow-up characteristics of forest 25 percent. Fire fighters should fires have been attributed to low increase their vigil whenever Hoping that this unexpected fire relative humidities and strong sur- these low relative humidities pattern might be explained, the exist or are forecast. daily surface weather maps and the When this article was originally published, temperatures from the surface to Rollo Davis was a Forestry Meteorologist Atmospheric the 5,000-foot (1,524-m) level were for the ESSA Weather Bureau, Jackson, critically examined for all days on MS. Turbulence The relationship of atmospheric which fires of more than 300 acres (121 ha), classed as “E” fires, * The article is reprinted from Fire Control Notes 30(3) turbulence to erratic fire behavior [Summer 1969]: 3–4, 15. has also been studied and dis burned. The examination of the ** Editor’s note: Beginning with the Summer 1969 temperature profiles aloft strongly issue of Fire Control Notes, most articles in each issue cussed. As early as 1951, George M. started with a short summary paragraph. In 1972, the practice was largely discontinued. Byram and Ralph M. Nelson pre sented a paper titled “The Possible † Fire Control Notes 12(3) 1–8.
Fire Management Today 56 suggested that the atmospheric unstable, and fifteen others burned behavior of forest fires once they instability in the lower atmosphere during unstable conditions. The got started. played a significant role in erratic greatest number, by a significant behavior of fires. percentage, occurred when the air It seems reasonable that air mass mass was classified as absolutely stability should play a very impor To investigate further, information unstable. Thirty four of the big tant role in the behavior of forest on all 1967 fires of the class “E” fires, nearly one-half of the 70 cases fires. Unstable air, from the meteor and larger was requested from the studied, burned when the air mass ological viewpoint, is also convec Fire Control Directors of the States at the fire site was absolutely unsta tively unstable. Once the air starts surrounding Mississippi. The ble. to rise, it will be warmer than its requested information was supplied surroundings. The air continues to by Louisiana, Arkansas, Tennessee, Relative Humidities rise until it reaches a level where and Alabama, and a total of 70 fires Relative humidities in the area of the temperature of the surrounding were investigated. No attempt was the fires ranged from 18 percent to air is the same. When unstable air made to investigate weather condi 80 percent. A large percent of the is displaced upward, it is replaced tions for fires when fire control fires during periods when the by air moving laterally, creating an personnel were unable to attack the atmosphere was absolutely unstable indraft of air, which is also unsta fire shortly after it started. burned when relative humidities at ble. This air rises. With the heat of the surface were above the level the fire being the initiating force to Atmospheric stability in the layer normally associated with big or start and maintaining convection, a between the surface and the 5,000 erratic fires. Nearly 60 percent of chain reaction is begun. The con foot (1,524-m) level was categorized the large fires studied took place vective column increases in size, for the investigations as follows: when the relative humidity in the and the indrafts increase in velocity area was above 30 percent. Air mass to fan the flames which then 1. Stable—Temperatures aloft stability, therefore, appears to be as increase the heat to intensify con decreasing with increase in alti significant, if not more significant, vection, and so on (fig. 1). Fire con tude at a rate about 3.5 degrees than low-level moisture in the trol personnel are well aware of F or less per 1,000 feet. 2. Conditionally unstable— Temperature decrease with increase in altitude at a rate of 3.5 to 5.4 degrees F per 1,000 feet. (Conditionally unstable air tends to become unstable if forced to rise. Additional heat supplied at the surface is suffi cient to produce the needed rise.) 3. Unstable—Temperature decrease with increase in altitude of 5.5 degrees F per 1,000 feet. 4. Absolutely unstable— Temperature decrease with increase in altitude greater than 5.5 degrees F per 1,000 feet.
Only six of the 70 fires studied occurred when the conditions in the low levels of the atmosphere were classified as stable. Fifteen, or 21 percent, occurred when the air mass was classified as conditionally Figure l—Convection currents visibly at work on a forest fire.
Volume 64 • No. 1 • Winter 2004 57 many of the direct and indirect The atmospheric stability forecast should be a effects of air mass instability on for routine product of all weather offices, and fire est fires. Some of the more spectac ular effects are rapid crowning, control personnel should be trained to use it. long-distance spotting, erratic movement, and blow-up potential. Upper air temperature data are future and come up with a forecast readily available at all ESSA of the atmospheric stability for the Conclusions and Weather Bureau Offices where following 36 to 48 hours. Recommendations forestry meteorologists are sta Considering the value of such fore Most large fires occur when the tioned. These data enable the casts to the forestry industry, the temperature profiles through the forestry meteorologist to determine atmospheric stability forecast lower levels of the atmosphere the degree of atmospheric instabili should be a routine product of all exhibit some degree of instability. ty. Using other meteorological weather offices, and fire control Fire control foresters who are fur information available, such as the personnel should be trained to use nished daily with an atmospheric computerized lifted index prognos it. ■ stability forecast can plan ahead tic charts, the forestry meterologist and use their manpower and equip can project the stability into the ment better.
Russo and Schoemaker (1989) Decision Trap 5— Shortsighted Shortcuts: Relying inappropriately on “rules of thumb,” such as implicitly trusting the most readily available information or anchoring too much on convenient facts.* * See page 9.
Fire Management Today 58 DOWNBURSTS AND WILDLAND FIRES: A DANGEROUS COMBINATION*
Donald A. Haines
On June 8, 1981, a wildland fire on Merritt Island, FL, suddenly A downburst is a downdraft associated with a changed directions, killing two fire thunderstorm or other well-developed cumulus fighters. On August 2, 1985, Delta clouds that induces an outburst of damaging Flight 191 crashed and burned while attempting to land at Dallas- winds on or near the ground. Fort Worth Airport. These two events had one common theme, the downdraft is almost always strong thunderstorm down- Definition associated with moderate to heavy bursts.** The odds are high that the weather event described in the introduction rain), usually experience wet down- bursts. The wet downburst pro t happens to most firefighters was a downburst. A downburst is a duces a core of rain that is visible, sooner or later if they have been downdraft associated with a thun although it may be obscured by on the job long enough. Every derstorm or other well-developed I associated weather. thing along the fireline seems fairly cumulus clouds that induces an well controlled. But then, unex outburst of damaging winds on or Dry downbursts occur in more arid pectedly, the wind shifts and near the ground. When the burst is places, like Colorado, when cloud becomes erratic. Wind speed picks small (0.4 to 4 km or 0.25 to 2.5 bases are higher and precipitation up dramatically for 5 to 15 minutes miles in diameter), it is a micro- evaporates before the downdraft and then decreases. burst; larger ones (more than 6.5 km or 2.5 miles in diameter) are reaches the ground (Monastersky 1987). The dry downburst might Another factor is added to the high macrobursts. Not all downdrafts are not be seen easily by either radar or winds: Precipitation ranging from downbursts. Fujita (1978) stated observers in such cases. Both very light to very heavy. It may fall that horizontal wind speeds gener cumulonimbus clouds as well as so hard during a thunderstorm that ally exceed 40 miles per hour (64 less fully developed rain clouds can it puts out the fire, or it may evapo km/h) on the ground in a true produce them. rate before it hits the ground. downburst. Although Schroeder and Buck (1970) discussed down- During a study of microbursts in With a change in weather comes a drafts in their handbook Fire the Denver area, Fujita and change in fire behavior—this time Weather, recent research has great Wakimoto (1983) found that 81 for the worse. The fire changes ly increased our knowledge of percent were the dry type. Little or direction, previously controlled downburst occurrence and struc no rain fell to the surface with lines are lost, and a routine opera ture. Because a downburst can them. In contrast, during an tion becomes life threatening. What cause dramatic and dangerous fire Oklahoma study, Eilts and Doviak happened? behavior, firefighters should under stand this phenomenon. (1987) found that the macrobursts When this article was originally published, detected on their radar were imbed Donald A. Haines was a principal research Downbursts are classed as either ded in intense convective storms meteorologist, USDA Forest Service, North and had large, heavy rain cores. Central Forest Experiment Station, East dry or wet. Most investigators Lansing, MI. believe that both types require rain But, these differences in detection drops as an initial condition may be the result of scanning strategies used with the different * The article is reprinted from Fire Management Notes because evaporation of these drops 49(3) [Summer 1988]: 8–10. radar units Eilts and Doviak (1987). ** Editor’s note: This article, unlike any others in the cools the air, which then falls as it Summer 1988 issue of Fire Management Notes, con gets heavier. Humid areas, like the In particular, the Oklahoma radar tains a short summary paragraph. Except for a brief may have missed lighter rain cores. period from 1969 to 1972, this practice was highly Southeastern United States (where unusual in the journal.
Volume 64 • No. 1 • Winter 2004 59 Sherman (1987) concluded that Although wet downbursts are difficult to forecast, with the falling dense air in a downbursts in a dry environment can be predicted downburst, the flow behaves like a toroidal vortex. In other words, as from morning upper-air soundings. the vortex at the head of a down- burst approaches the ground, each flank of the Ransom Road Fire. A and the flames overtook the two element of the falling vortex moves thunderstorm developed and winds men. A tower with a recording downward and outward along a abruptly changed from south to anemometer to the northeast of the roughly hyperbolic path. Near the west. In response, the head of the fire area showed wind speeds cloud base, winds and rain con fire changed from north to east, increasing from an average of 7 to verge around the descending air, feeding into it. A sharp observer might be able to spot the develop ing downburst if it is outlined by rain because the precipitation falls rapidly, reaching a downward veloc ity of 65 miles per hour (105 km/h).
When flying directly beneath a microburst, a pilot in a spotter plane will find that the difference between the headwind and the tail wind is typically 60 miles per hour (97 km/h) as the winds spill out horizontally to either side of the parent cloud. Fujita (1978) showed that in one case this difference exceeded 172 miles per hour (279 km/h).
Several researchers have found a relationship between an observed temperature drop at surface and the increased wind speed. The larg er the temperature change, the more severe the wind gusts. The leading edge of the horizontal movement of the wind gusts is called a gust front. As it spreads horizontally, the gust front may develop as an expanding fluid struc ture many miles long, depending on the strength of the downburst (fig. 1). A Tragic Example The weather that occurred with the 1981 Florida wildland fire seems to have been a classical downburst (USDA FWS 1981). Two men oper Figure 1—The four stages of a thunderstorm downburst and gust front. The precipitation ating a dozer and plow attempted roll is a horizontal roll vortex formed by airflow that is deflected upward by the ground. Note the changes in wind direction as the gust front passes a point and moves on containment along the eastern (Wakimoto 1982).
Fire Management Today 60 25 miles per hour (11–41 km/h), Conclusions References with gusts to 52 miles per hour (84 Even though research is taking the Caracena, F.; Maier, M.W. 1987. Analysis of km/h). Within 10 minutes, the tem surprise out of the dry downburst, a microburst in the FACE meteorological perature fell from 82 °F (28 °C) to mesonetwork in southern Florida. forecasting the wet downburst will Monthly Weather Review. 115(5): 60º F (16 °C). The tower readings be a difficult problem for some 969–985. ended at that point as lightning hit time to come. Predicting the Eilts, M.D.; Doviak, R.J. 1987. Oklahoma it. downbursts and their asymmetry. Journal of Climate and Applied Meteorology. 26(1): 69–78. This then was a true wet-core Fujita, T.T. 1978. Manual of downburst downburst as “a heavy rainstorm, The Board suggested identification for project NIMROD. Res. accompanied by thunder and light that crews pull back Pap. 156. Chicago, IL: University of Chicago. ning, descended on the fire area, during impending Fujita, T.T.; Wakimoto, R.M. 1983. lasting for 15 to 20 minutes and thunderstorms in areas Microbursts in JAWS depicted by Doppler just about completely extinguished radar, PAM, and aerial photographs. In: with fuels that burn with Preprint of Proceedings, 21st Conference the wildfire” (USDA FWS 1981). on Radar Meteorology; 1982 September high intensity and rate 19–23; Edmonton, AB. Boston, MA: Forecast Possibilities American Meteorological Society: of spread. 638–645. Although wet downbursts are diffi Monastersky, R. 1987. Mastering the cult to forecast, downbursts in a microburst. Science News. 131(12): 185–187. dry environment can be predicted impressive winds that accompany Schroeder, M.J.; Buck, C.C. 1970. Fire from morning upper-air soundings. these downbursts remains an elu weather … A guide for application of According to Caracena and Maier sive goal. Accordingly, the Board of meteorological information to forest fire (1987), “inroads have already been control operations. Agric. Handb. 360. Inquiry for the Ransom Road Fire Washington, DC: U.S. Department of made into the microburst forecast aimed recommendations at the Agriculture. problem in understanding the dry field level. The Board felt that an Sherman, D.J. 1987. The passage of a weak end of the convective spectrum observer in a spotter plane in direct thunderstorm downburst over an instru mented tower. Monthly Weather Review. where the concept of severe weath contact with the line crews could 115(6): 1193–1205. er is extended to conditions that have anticipated the weather condi USDA FWS (U.S. Department of the favor strong downdrafts from high tions and, hence, fire behavior Interior, U.S. Fish and Wildlife Service). 1981. Report of Board of Inquiry on base cumulonimbi.” They believe changes. This could have allowed Merritt Island National Wildlife Refuge that to be able to forecast down- directions for an escape route. The wildfire fatalities. Kennedy Space Center, bursts in all parts of the United Board also suggested that crews Titusville, FL. States, meteorologists must first Wakimoto, R.M. 1982. The life cycle of pull back from the fire during thunderstorm gust fronts as viewed with understand how nature generates impending thunderstorms in areas Doppler radar and radiosonde data. them over the entire range from with fuels that burn with high Monthly Weather Review. 112(8): ■ wet to dry extremes. Forecasters intensity and rate of spread, as in 1060–1082. then could diagnose typical down- the Ransom Road Fire. burst conditions from the daily upper-air data.
Russo and Schoemaker (1989) Decision Trap 6—Shooting From the Hip: Believing you can keep straight in your head all the information you’ve discovered, and therefore “winging it” rather than following a systematic procedure when making the final choice.* * See page 9.
Volume 64 • No. 1 • Winter 2004 61 ESTIMATING SLOPE FOR PREDICTING FIRE BEHAVIOR*
Patricia L. Andrews
hen predicting fire behavior in the field, it is desirable to Those monitoring fires, such as prescribed fires in W be able to obtain the wilderness, will find this method especially useful. required input information with a minimum of special equipment. This article tells how to estimate with adequate precision using the Calculations were done using slope (percent) using materials in a method described here. BEHAVE (Andrews 1986). A resolu belt weather kit. This method can tion of less than 5 percent is clearly be used on wildfires by fire behav Figure 1 illustrates the effect of not necessary, especially when all of ior analysts, field observers, and slope on predicted flame length for the other uncertainties involved in strike team leaders. Those who are four fuel models: fire behavior prediction are taken monitoring fires that are not into account. On the other hand, receiving full suppression action, •4 (chaparral); the value for percent slope has such as prescribed fires in wilder • 13 (heavy logging slash); enough influence that a poor esti ness, will find it especially useful. •2 (timber litter and understory); mate might lead to a significant and over- or underprediction. Importance of Slope •9 (hardwood litter). To predict fire behavior, a fire spe Estimating Slope cialist must supply values for fuel In this example, there is no wind, The lines in figure 2 represent model, fuel moisture, windspeed, dead fuel moisture is 6 percent, and slope percentages from 0 to 100. and slope. Calculations can be done live fuel moisture is 100 percent. Using a sheet of adhesive acetate, using tables, nomograms, calcula tors, or computer programs Figure 1—The (Andrews 1986). As described by influence of slope on Rothermel (1983), fuels are classi calculated flame fied as a particular fuel model by length is shown for observation (Anderson 1982); wind- four fuel models under constant wind and fuel speed is measured; live fuel mois moisture conditions. ture is estimated by the state of curing; dead fuel moisture is deter mined by an estimate of shade and measurements of temperature and relative humidity; and slope is determined from a topographic map, estimated, or measured with an instrument such as a clinome ter. Slope can also be estimated
When this article was originally published, Patricia Andrews was a mathematician for the USDA Forest Service, Intermountain Research Station, Fire Sciences Laboratory, Missoula, MT.
* The article is reprinted from Fire Management Notes 49(3) [Summer 1988]: 16–18.
Fire Management Today 62 Figure 2—Diagram for use as directed to estimate slope to within 5 percent. The slight distortion caused by photocopying the diagram is unimportant. attach a copy of figure 2 to the References board in a belt weather kit. Notch Anderson, H.E. 1982. Aids to determining the board where the lines converge. fuel models for estimating fire behavior. Hang the compass by its neckstring Gen. Tech. Rep. INT–122. Ogden, UT: USDA Forest Service, Intermountain at the notch to serve as a plumb. Forest and Range Experiment Station. Sight along the board parallel to Andrews, P.L. 1986. BEHAVE: Fire behavior the slope, as shown in figure 3. prdiction and fuel modeling system— BURN subsystem, part 1.Gen. Tech. Rep. Noting where the string lies, read INT–194. Ogden, UT: USDA Forest the slope to the nearest 5 percent. Service, Intermountain Forest and Range Experiment Station. Andrews, P.L. 1986. Methods of calculating This method of estimating slope is fire behavior—You do have a choice. Fire a simple, no-cost alternative to eye Management Notes. 47(2): 6–10. ball estimates, which are notorious Rothermel, R.C. 1983. How to predict the ly poor, and to instruments such as spread and intensity of forest and range fires. Gen. Tech. Rep. INT–143. Ogden, clinometers, which are expensive UT: USDA Forest Service, Intermountain and give a level of resolution not Figure 3—The influence of slope on Forest and Range Experiment Station. ■ calculated flame length is shown for required for fire behavior predic four fuel models under constant wind tion. and fuel moisture conditions.
Volume 64 • No. 1 • Winter 2004 63 AIR TANKER VORTEX TURBULENCE— REVISITED*
Donald A. Haines
xtreme drought had a devastat ing impact on wildland fire The crown fire was described as a “waterfall,” a E activity over much of the “breaking wave,” a “curl,” and a “wave curl”—in Central and Western United States other words, a horizontal roll vortex of some type. during the summer and autumn of 1988. State and Federal suppression forces in Michigan’s Upper Penin The trees were 3 to 6 inches (8–15 What happened? Of equal interest, sula confronted fire behavior rarely cm) in diameter and 25 to 30 feet why did it happen only along this experienced in early summer, typi (7.5–9 m) high. Compared with section of the line? cally a period of low fire occur other sectors, this was a quiet area. rence. The operators plowed 50 feet (15 Possibilities Rejected m) from a backing fire with 2-foot The Stockyard Fire None of the more typical causes (0.6-m) flame lengths. Aided by a can explain the unexpected changes The Stockyard Fire, near Escanaba, firing-out crew well behind the in fire behavior. There were no MI, proved especially troublesome tractor operators, the fire burned to heavy fuel concentrations. Fuels because of unexpected fire behavior. the line, leaving a wide black area. were relatively uniform in a typical Among other features, 100-foot jack pine plantation. Also, the area long (30-m) sheets of flame moved Winds were light and then became was relatively flat with no unusual horizontally, undulating like waves calm. The low flames suddenly topographic features. on a water surface. Fire, brands began to “climb” up a few trees into moving with the sheets caused spot the crowns. Within a minute or two There were no apparent immediate fires that quickly turned into 15- to the flames became a high wall. The weather concerns. The weather 30-foot-high (4.5- to 9-m) fire wall changed into a crown fire, charts showed that the region was whirlwinds. Even though the moving directly toward the tractor covered by a large, flat, high-pres Burning Index (National Fire crew. Flame tilt had shifted from sure cell. Although the fire Danger Rating System) was 27 with slightly eastward to vertical and occurred near one of the Great fuel model E, burning was so then to westward. Lakes, the land/sea breeze circula intense along some sectors of the tion did not change at that time. fire that escaping gases did not The resultant crown fire was Also there was no apparent change ignite until well above the fire. In described as a “waterfall,” a “break in the vertical structure of the those cases, the gases exploded as ing wave,” a “curl,” and a “wave atmosphere over the fire. large bubbles high in the air. curl.” In other words, it was a hori zontal roll vortex of some type. Burnout operations upstream of But the most interesting behavior Witnesses also stated that this wave the site had no effect on down occurred along a 300-foot (90-m) (vortex) moved along the fire line at stream activity, nor did anyone see length of the right flank. Here about 15 miles per hour (24 km/h). the formation of a large vertical fire three tractor-plow operators built The vortex rotation threw foot-long whirl or other suspicious fire-initi line within a jack pine plantation. (0.3-m) fire brands westward, 100 ated features. feet (30 m) away from the flank, When this article was originally published, into unburned fuels. Flame heights Lessons Relearned Don Haines was a research meteorologist increased to 150 to 250 feet (45–76 for the USDA Forest Service, North Central However, one interesting incident Forest Experiment Station, East Lansing, m). Luckily no one was killed, MI. although one of the tractor opera did occur in this sector only min tors was badly injured and spent utes before the sudden, violent * The article is reprinted from Fire Management Notes increase in fire activity. A DC-4 air 50(2) [Spring 1989]: 14–15. weeks in a medical bum center.
Fire Management Today 64 tanker carrying 2,000 gallons Davis and Chandler warned that under special (7,571 L) of retardant flew along circumstances, vortex wakes may cause fire the fire line, circled, then came back and dropped the retardant just behavior to change dramatically. south of this sector as the fire intensified. The tanker was flying at through the air. They usually roll For a more complete description of less than 400 feet (120 m) and at apart as they hit the surface of the the action of these vortices, please perhaps 140 miles per hour (225 ground. This vortex phenomenon read Davis and Chandler (1965)* km/h). was discovered when it caused the and also Chandler and others crash of several light aircraft (1983). Almost a quarter of a century ago, caught in the wakes of large air Davis and Chandler (1965) pub planes. Be Aware lished an article in Fire Control Notes, “Vortex turbulence—its Today’s fire crews and air tanker Ordinarily, aircraft vortex turbu pilots would be wise to heed the effect on fire behavior.” In it they lence does not endanger fire con warned about aircraft vortex turbu warnings offered by Davis and trol forces. But Davis and Chandler Chandler. Fire crews should be lence, a sheet of turbulent air left warned that under special circum in the wake of all aircraft. It rolls alert for trouble in these circum stances, vortex wakes may cause stances: up into a strong vortex pair—two fire behavior to change dramatically. compact, fast-spinning funnels of • The air is still and calm. air (fig. 1). Unfortunately, this vor Vortex severity and persistence vary tex pair is usually invisible. Under • The fire is burning in open land with several factors. Most impor or in scattered or low timber. certain conditions, the two vortices tant are the type, size, speed, and may stay close together, sometimes • The air tanker is large or heavily altitude of the aircraft and the pre loaded. undulating slightly as they stretch vailing atmospheric conditions. rearward. The interaction between • The air tanker is flying low and Other factors being equal, the slowly. them tends to make them stay strongest vortex pair is produced by together as they move downward a large, slow-flying aircraft with a Air tanker pilots should be aware of high wingspan loading. The speed the problem the aircraft can cause is most important before landing or and take these precautions: after takeoff. It is also a factor when an air tanker slows down for an • Do not fly parallel to the fire line accurate airdrop. more than necessary. • Keep high except when making Aircraft altitude is important the actual drop. because vortices weaken rapidly • Ensure that ground crews are with time. Under typical wind alert to the presence of an air speeds, the vortex pair may lose its tanker and the intended flight potential impact in less than a path. minute. But the pair tends to per sist in calm air. At high altitude, References the two vortices remain separated Chandler, C.; Cheney, P.; Thomas, P.; by a distance slightly less than the Trabaud, L.; Williams, D. 1983. Fire in aircraft’s wingspan. However, the forestry. Vol. II: Forest fire management interaction of the vortices causes and organization. New York: Wiley & Sons: 240–241. them to drop at a rate of 300 to 500 Davis, J.B.; Chandler, C. l965. Vortex turbu feet per minute (90–150 m/min) lence—Its effect on fire behavior. Fire depending on various factors. Control Notes. 26(2): 4–6.
Figure 1—Low-flying spray plane. Note funneling effect on spray trailing each * Editor’s note: Reprinted in this issue of Fire wing. Management Today.
Volume 64 • No. 1 • Winter 2004 65 A TREND ANALYSIS OF FIRELINE “WATCH OUT” SITUATIONS IN SEVEN FIRE-SUPPRESSION FATALITY ACCIDENTS*
Gene A. Morse
nder the auspices of the National Wildfire Coordinating In each of seven fatality events, a single UGroup’s (NWCG’s) Fireline overlooked “Watch Out” Situation appeared to be Safety Committee, seven events the major contributing factor. resulting in nine firefighter fatali ties were analyzed. Common to all the fatalities was the use of tractor- these events. The Standards for In analyzing these events, it was plow units. The tractor plow is the Survival focus on the 18 “Watch apparent that, in each instance, a primary equipment used for forest Out” Situations and the Standard single overlooked “Watch Out” and wildland fire suppression activ Fire Orders. The “Watch Out” Situation appeared to be the major ities in the South and the East. Situations and Fire Orders have contributing factor. Simply follow gained widespread use as aids to ing that reasoning process a step The events were well documented safety among forest and wildland further leads to the conclusion that with extensive details, photographs, fire suppression agencies. if the dominant positive action— and maps. They provided an ade Fire Order—to counteract that quate background of the events and Practical application of the negative situation had been imme factors leading to the deaths of the Standards for Survival training on diately observed, then a tragic situ nine firefighters. A careful analysis, an incident centers around identify ation may have been avoided. it was believed, might reveal a pat ing a potentially dangerous fireline tern of unsafe actions that could be event, linking it to a “Watch Out” Perhaps some readers might say changed in the future to avoid a Situation on the Survival Checklist, that the method used in this analy recurrence of these tragic events. and then taking a positive action sis is too simplistic—that overlook (observing the appropriate Fire ing common threats to safety is too A first reading of the fatality reports Order) to eliminate or minimize basic to be neglected. In this indicated no common factors in the possibility of firefighter injury response lies a pitfall: The “Watch fuels or topography. Some similari or death. One response from one of Out” Situations—commonly occur ties were noted in weather patterns, the fatality reports of the seven ring during a fire event—are haz but it was difficult to draw defini events illustrates how a “Watch ardous situations. It is hard, when a tive conclusions based on the Out” Situation was identified but fire seems routine, to believe that it weather factor alone. the Fire Order was not observed: could become threatening. But a fire event has the potential to Approach • Potentially hazardous event: “It develop a “Watch Out” Situation The decision was made to apply the had been jumping our lines … quickly. Danger is inherent in a fire process developed in the NWCG the thing [fire] had already event. To develop “scotoma” in Standards for Survival training pro jumped a 60-foot canal….” regard to these dangers is a major gram as the criteria for analyzing • “Watch Out” Situation (#16): contributing factor to many fireline Getting frequent spot fires across fatalities. line. When this article was originally published, Gene Morse was a division training and • Fire Order not observed (#1): What is scotoma and how does it safety officer for the Florida Division of Initiate all action based on cur apply to fireline fatalities? Scotoma, Forestry, Tallahassee, FL. rent and expected fire behavior. a medical term, has direct rele vance to this analysis. Scotoma is, * The article is reprinted from Fire Management Notes literally, a blind spot. In a psycho 51(2) [Spring 1990]: 8–12.
Fire Management Today 66 logical sense, it is that condition Figure 1—A number which occurs when a person tends of “Watch out” Situations were pres to block out from his or her con ent when a fire sciousness anything considered not tragedy occurred in important—or critical—to survival. this mountainous region, resulting in two firefighter deaths. The significance of scotoma in fire- The familiar state line suppression operations is dra ment, “personnel on the fire considered matically emphasized by this state the situation to be ment found in the fatality reports: routine until fire blew “Personnel on the fire considered it up,” was contained in to be routine … until the fire blew the fire report. Note victims’ location on up” (figs. 1 and 2). Although it was the windward side of phrased differently in several the ridge, adjacent to reports, this same type of comment a draw. Mild drought conditions existed, surfaced repeatedly. The meaning is with 30-mile-per clear: It was “just another fire” to hour (48-km/h) the firefighters. winds.
Scotoma had taken hold and blocked out sensitivity to hazardous events or conditions present in the fire environment.
The prevalence of this attitude or Figure 2—This mindset was best expressed by a sketch of a fatality veteran firefighter recently during a scene, prepared by a fireline safety training session when fire behavior analyst, illustrates a danger he commented, “I know those ous fireline condi things [“Watch Out” Situations] are tion. It shows the fire out there on the fire, but I’ve seen with three separate heads, burning in them so many times I’m not really three subdivision aware of them now.” blocks, part of a huge, largely unpop ulated subdivision Trends and Conclusions with heavy fuel load This analysis—to identify haz ing. Mild drought conditions existed. ardous conditions or events in the Note that the fire fatality reports and then link them fighter’s tractor is to the NWCG Survival Checklist— located in a “pocket,” aimed at determining significant with the fire heads on either side trends. The findings established advancing more rap that there were 84 separate haz idly than the fire in ardous conditions or events in the Block No. 2. Person nel on this fire con fatality reports. Some specific sidered it to be “rou examples drawn directly from the tine”—until it blew reports, linked to the Survival up. Checklist, and the appropriate dom inant Standard Fire Order are out lined in table 1.
An analysis of the 84 hazardous conditions or events, when linked
Volume 64 • No. 1 • Winter 2004 67 Table 1—Nine examples of hazardous conditions listed on the Survival Checklist for which there is a Standard Fire Order. Hazardous condition or event Survival Checklist Fire Order Initial instructions to the firefighter: “Grab the first No. 6: Instructions and E: Ensure instructions piece of fire you come to—and do the best you can.” assignments not clear. are given and understood.
“[The fire] looked like one of those waves in Hawaii, No. 4: Unfamiliar with I: Initiate all action based like when you shoot the waves on a surfboard. The weather and local factors on current and expected smoke was going up; it looked like an explosion.” influencing fire behavior. fire behavior.
Q: “What radio traffic did you get after XXX offloaded No. 7: No communication R: Remain in communi and started plowing?” link with crew members cation with crew mem A: “None … I never heard any.” or supervisor. bers, your supervisor, and adjoining forces.
Q: “Had there been any briefings? Weather briefings? No. 5: Uninformed on R: Retain control at all Fire behavior briefings? Safety briefings? strategy, tactics, and haz times. A: “Not to my knowledge “ ards.
“There was no apparent briefing with the crew on a No. 3: Safety zones and D: Determine safety zones plan of attack and escape, if necessary.” escape routes not identi and escape routes. fied.
“Heavy palmetto growth prohibited penetration to No. 17: Terrain and fuels D: Determine safety zones safety only 60 feet [18 m] away.” make escape to safety and escape routes. zones difficult.
“[He] began initial attack by plowing lines across the No. 10: Attempting F: Fight fire aggressively head of the fire.” frontal assault on fire. but provide for safety first.
“[He] noticed a space 50 to 100 feet [15–30 m] long No. 11: Unburned fuel O: Obtain current infor on the line that was not tied together.” between you and the fire. mation on fire status.
“It [the wind] blew from the east, southeast, south, No. 15: Wind increases R: Recognize current southwest, west, and then back again without warn and/or changes direction. weather conditions and ing.” obtain forecasts. to the Survival Checklist, revealed the following trends: Scotoma—blindness to danger perceived as routine—had taken hold and blocked out sensitivity •Twenty-two were tied directly to to hazardous events or conditions present in the Survival Checklist Situation No. 4 (Unfamiliar with weather and fire environment. local factors influencing fire behavior). • Thirteen were linked closely to •Twelve were connected directly to • Eleven were linked to Survival Survival Checklist Situation No. 7 Survival Checklist Situation No. Checklist Situation No. 16 (No communication link with 15 (Wind increases or changes (Getting frequent spot fires crew member or supervisor). direction). across line).
Fire Management Today 68 What does this analysis of the The relationship is clearly established between deaths of nine firefighters estab fireline fatalities and a lack of awareness or lish? With 13 conditions or events associated with communication, it sensitivity to significant changes in fire behavior. is obvious that poor or nonexistent communication placed the fire may wish to consider: to determine if participants had fighters in a vulnerable position. No made the right tactical decisions. one can question the paramount • Besides the established national •Determine a fuel condition necessity of maintaining close, courses in fire behavior threshold (possibly fuel moisture) effective communication with other (Introduction to Fire Behavior; for their local area in which personnel in the hostile fire envi Intermediate Fire Behavior; and going beyond a certain level ronment. Advanced Fire Behavior), develop would signal the mandatory more localized fire behavior establishment of a safe anchor But it is even more revealing to training focused on individual point, posted lookout, and desig note that more than half of the State or regional fuel types. nated escape routes and safety hazardous conditions or events •Teach firefighters about fire sci zones to ensure safe tactical oper identified in the analysis relate to ence—the relationship between ations in the event of unexpected some aspect of fire behavior. fuels, weather, and topography changes in weather and fire Specifically, the relationship is and fire—and how to transfer fire behavior. clearly established between fireline behavior knowledge into the • Give high priority to fireline safe fatalities and a lack of awareness or most prudent application of tac ty training, such as the NWCG sensitivity to significant changes in tics that will get the fire suppres Standards for Survival course. fire behavior. sion job done without compro mising firefighter safety. Follow Agencies with few materials avail Recommendations To up classroom instruction in fire able for fireline safety training Improve Safety behavior training courses with should obtain a copy of the recently What recommendations can be simulated fire exercises in the prepared “Fireline Safety and made on the basis of this trend field, where firefighters would be Health Resources.” This publication analysis to reduce scotoma on the required to demonstrate safe, was developed by the NWCG fireline and ensure firefighter safe effective firefighting tactics in dif Fireline Safety Committee listing ty? Listed below are some specific ferent fuel, weather, and topogra materials available for sharing by action items that NWCG agencies phy conditions. Evaluate critically all NWCG agencies. ■
Russo and Schoemaker (1989) Decision Trap 7—Group Failure: Assuming that with many smart people involved, good choices will follow automatically, and therefore failing to manage the group decisionmaking process.* * See page 9.
Volume 64 • No. 1 • Winter 2004 69 LCES—A KEY TO SAFETY IN THE WILDLAND FIRE ENVIRONMENT*
Paul Gleason
LCES—A System for L—Lookout(s) • The LCES system must be con Operational Safety tinuously reevaluated as fire con C—Communication(s) ditions change. n the wildland fire environment, where four basic safety hazards E—Escape routes How To Make LCES I confront the firefighter—light S—Safety zone(s) ning, fire-weakened timber, rolling Work rocks, and entrapment by running •Train lookouts to observe the fires—LCES is key to safe proce Key Guidelines wildland fire environment and to dure for firefighters. LCES stands LCES is built on two basic guide recognize and anticipate fire for “lookout(s),” “communica lines: behavior changes. tion(s),” “escape routes,” and “safe • Position lookout or lookouts ty zone(s)”—an interconnection • Before safety is threatened, each where both the hazard and the each firefighter must know. firefighter must be informed how firefighters can be seen. (Each Together, the elements of LCES the LCES system will be used. situation—the terrain, cover, and form a safety system used by fire fighters to protect themselves. This safety procedure is put in place The LCES system approach to fireline safety is an before fighting the fire: Select a outgrowth of my analysis of fatalities and near- lookout or lookouts, set up a com munication system, choose escape misses for over 20 years of active fireline routes, and select safety zone or suppression duties. zones (fig. 1).
In operation, LCES functions sequentially—it’s a self-triggering mechanism: Lookouts assess—and reassess—the fire environment and communicate to each firefighter threats to safety; firefighters use escape routes and move to safety zones. Actually, all firefighters should be alert to changes in the fire environment and have the authority to initiate communica tion.
When this article was first published, Paul Gleason was the North Roosevelt fire man agement officer, USDA Forest Service, Arapaho and Roosevelt National Forests, Redfeather Ranger District, Fort Collins, CO.
* The article is reprinted from Fire Management Notes 52(4) [Fall 1991]: 9. Figure 1—LCES components.
Fire Management Today 70 fire size—determines the number LCES simply refocuses on the essential elements of lookouts that are needed. As of the standard Fire Orders. Its use should be stated before, every firefighter has both the authority and responsi automatic in fireline operations. bility to warn others of threats to safety.) escape route was cut off by the not imply that a shelter should • Set up communication system— advancing fire.) not be deployed if needed, only radio, voice, or both—by which • Reestablish escape routes as their that if there is a deployment, the the lookout or lookouts warn effectiveness decreases. (As a fire safety zone location was not truly firefighters promptly and clearly fighter works along the fire a safety zone.) of approaching threat. (Most perimeter, fatigue and distance often the lookout initiates a increase the time required to A Final Word warning that is subsequently reach a safety zone.) The LCES system approach to fire- passed down to each firefighter • Establish safety zones—locations by word of mouth. It is para line safety is an outgrowth of my where the threatened firefighter analysis of fatalities and near-miss mount that every firefighter may find adequate refuge from receive the correct message in a es for over 20 years of active fire- the danger. (Fireline intensity, air line suppression duties. LCES sim timely manner.) flow, and topographic location • Establish the escape routes (at ply refocuses on the essential ele determine a safety zone’s effec ments of the standard Fire Orders. least two)—the paths the fire tiveness. Shelter deployment sites fighters take from threatened Its use should be automatic in fire- have sometimes been termed, line operations. All firefighters position to area free from dan improperly and unfortunately, ger—and make them known. (In should know LCES, the “safety zones.” Safety zones Lookout–Communication–Escape the Battlement Creek 1976 fire, should be conceptualized and three firefighters lost their lives routes–Safety zone interconnec planned as a location where no tion. ■ after retreat along their only shelter will be needed. This does
The author’s personalized license plate. Paul Gleason developed the LCES concept while serving as superin tendent for the Zigzag Interagency Hotshot Crew. The photo was taken at the request of Gleason’s wife Karen after he passed away.* Photo: Mike Goodman, Lake Estes, CO, 2003.
* For more on Gleason and LCES, see the Summer 2003 issue of Fire Management Today, a special issue dedicated to Paul Gleason. In particular, see Paul Keller, “‘Gleason Complex’ Puts Up Huge ‘Plume’: A Tribute to Paul Gleason” (Fire Management Today 63[3]: 85–90); and Jim Cook and Angela Tom, “Inteview With Paul Gleason” (Fire Management Today 63[3]: 91–94).
Volume 64 • No. 1 • Winter 2004 71 HOW IC’S CAN GET MAXIMUM USE OF WEATHER INFORMATION*
Christopher J. Cuoco and James K. Barnett
uring initial and extended attack, up-to-date weather The two most critical factors in acquiring weather D information is critical to suc forecasts during an incident are communications cessful and safe wildland firefight ing. Unfortunately, obtaining and and time. evaluating fire weather forecasts can be a challenge. With the few weather conditions. With it, the IC the designated individuals below basic weather concepts plus the two will be able to evaluate the accura assume the responsibilities follow user-friendly field aids provided cy of a fire weather forecast and ing their job titles: here, Incident Commanders (IC’s) determine the effect of current and can get maximum use of weather forecasted weather conditions on IC’s: information. fire behavior and firefighting opera • Develop fire weather and fire tions. behavior interpretation skills. The first of the reproducible field • Practice taking observations aids, the Supplemental Observation using techniques recommended Sheet (see sample on page 73), can Planning for Efficient in the Intermediate Wildland assist in using the “Mobile Fire- Communications Fire Behavior course (S–290). Weather Observer’s Record” provid The two most critical factors in • Become familiar with Remote ed in every field belt weather kit. acquiring weather forecasts during Automated Weather Stations The Supplemental Observation an incident are communications (RAWS) and other real-time Sheet can prompt a fire weather and time. Typically, dispatchers and weather information sources in observer to take notice of impor IC’s communicate via radio. their area and become proficient tant weather phenomena that may However, radio frequencies often in the means to obtain the data. affect fire behavior. This informa become overloaded and subse They should seek out this infor tion can be recorded in the quently slow down or eliminate mation when fighting fire out “Characteristics and Comments” requests for updated weather infor side their home territory.** section of the observation form and mation. In addition, taking a passed on to the IC and the fire weather observation, relaying the Dispatchers: weather forecaster. data, and preparing and transmit ting a fire weather forecast all take • Become sufficiently trained to understand and communicate The second field aid—the Weather valuable time. Evaluation Sheet (see sample on ** The local NWS Fire Weather Operating Plan To make communications more (OPLAN) is a good resource for weather observations. page 73)—will lead an IC through a The OPLAN will list RAWS sites, locations, elevations, series of questions designed to efficient and effective, we suggest and ID numbers. increase understanding of current
Original editor’s note: Chris Cuoco was the National Weather Service When this article was originally published, (NWS) Colorado Fire Weather program manager throughout the severe Chris Cuoco was a warning coordination meteorologist for the U.S. Department of fire season of 1994. The U.S. Department of Commerce recently pre Commerce, National Weather Service, sented him the Silver Medal Award for the fire weather forecasts and Flagstaff, AZ; and Jim Barnett was the Red Flag Warnings he issued before, during, and after the tragic South regional dispatcher for the USDA Forest Service, Rocky Mountain Area Interagency Canyon Fire on July 6, 1994. He accepted the award in the names and Fire Coordination Center; Broomfield, CO. memory of the 14 firefighters who died while fighting the South Canyon Fire. It is his hope that the information presented here will in * The article is reprinted from Fire Management Notes some way help prevent such a tragedy from ever happening again. 56(1) [Winter 1996]: 20–24.
Fire Management Today 72 Supplemental Observation Sheet In addition to the items specifically requested on the Spot Weather Observation Form found in the belt weather kit, the following should be observed. Circle or fill in appropriate items and communicate this information to the weather forecaster.
Cloud Observations Cloud cover percentage Cumulus development Key cloud indicators Possible consequences Clear (0-10% cover) Small cumulus Towering cumulus* Erratic winds Scattered (11-50% cover) Towering cumulus* Cumulonimbus* Erratic winds/thunderstorms Broken (51-90% cover) Cumulonimbus (anvil)* Horsetail cirrus Frontal approach (24-72 h) Overcast (91-100% cover) Direction(s)______Milky sky Frontal approach (24-72 h) Fog Distance______Lenticular clouds Increasing winds
Other Important Weather Observations Local Terrain Factors Inversion break Time______Fuel types______General wind shift Time______New direction______Canyons (chimneys, chutes) Upslope/downslope wind shift Time______New direction______Steep slopes Upvalley/downvalley wind shift Time______New direction______Large body of water nearby or snowpack Smoke dispersal: Rapid* Moderate Slow Direction______Dust devils* Additional comments
*May indicate instability which may cause erratic fire behavior.
Weather Evaluation Sheet 1. Do you have the current Fire Weather Zone Forecast for your area?
NO > Call dispatch. Request a forecast. YES > Evaluate forecast for your area and current weather conditions. Call for Spot Forecast > If information is incomplete or if the zone forecast is not representative of conditions on the incident.
2. Evaluating the Spot Forecast: Answer the questions in the first two columns. Use the third column to relate fire weather and fire behavior to firefighting strategy and tactics. Note that one weather parameter out of criteria may not require an updated forecast; it could be offset by other weather measurements or fuel conditions.
Instability Winds, temps, RH Relating weather to fire behavior 1. Cumulus cloud development Cloud cover compared to forecast. 1. How will the observed and forecasted ____ More development than forecasted ____More ____Same ____Less weather affect fire behavior? (more unstable)? Wind speed within 5 mph of forecast? 2. Are current strategy and tactics ____ Less development than forecasted ____ YES ____ NO > Consider new forecast. appropriate for observed and predicted (more stable)? fire behavior? Does observed wind direction fit the terrain? 2. Smoke column characteristics ____ YES ____ NO > Consider new forecast. 3. Do we need to change strategy and ____ Higher column than expected (more tactics to fight this fire safely? unstable)? Is the wind direction as forecast? ____ Lower column than expected? ____ YES ____ NO > Consider new forecast. (more stable)? Temp within 5 degrees of forecast? Request a new Spot Forecast if you believe 3. Conditions appear more unstable than ____ YES ____ NO > Consider new forecast. fire weather and fire behavior conditions forecasted? require a change in tactics. RH within 5% of forecast? ____ NO ____ YES ____ NO > Consider new forecast. ____ YES > Consider new forecast.
Volume 64 • No. 1 • Winter 2004 73 weather information as rapidly as IC’s should try to obtain a copy of the entire fire possible. weather zone forecast package or have the Fire Management Officers dispatcher read the applicable zone forecast over (FMO’s): the radio. • Develop coaching and prompting techniques to assist less experi enced field personnel. • Discusses the weather situation information by radio, IC’s should and general forecasts for geo ask the dispatcher to read all head FMO’s and Dispatchers: graphic and topographic zones in lines in their zone and in the dis • Establish primary and backup the issuing office’s area. cussion section of the forecast radio frequencies early each fire • Includes predictions of upper package. After reading or hearing season. level winds and smoke dispersal, the zone forecast, the IC should ask • Establish a rapid process for pass and provides extended weather these questions: ing weather information between outlooks. the field and the forecaster (e.g., •Provides an overall understand • Do I have a complete picture of with radio, phone, cellular ing of forecasted weather and the the weather situation? phone,* fax, computer). meteorological features causing • Do I feel comfortable with my • Develop guidelines for broadcast the weather. knowledge about the general ing fire weather forecasts, Fire weather pattern (i.e., pressure Weather Watches, Red Flag Note: The zone forecast may be too systems, cold fronts, general wind Warnings, pertinent special general to apply to some initial and patterns)? weather announcements, and key extended attack scenarios. • Do I understand the predicted National Fire-Danger Rating fire weather for my area? System (NFDRS) data. Warning and Watch • Do the predicted conditions make • Develop a “confirmation of Headlines sense for my incident? receipt” process for routine fire Red Flag Warnings and Fire weather forecasts and for critical Weather Watches are “highlighted” If the IC discovers that the infor fire weather information. with headlines preceding the fore mation is incomplete or if the zone • Establish a fire-danger tracking cast discussion and each applicable forecast is not representative of system for each dominant fuel zone forecast. (The conditions war conditions on the incident, the IC type in the area. (Such a system ranting a Red Flag Warning or Fire should consider requesting a Spot will aid in determining trends Weather Watch are explained in Forecast. and danger levels.) detail within the weather discus sion.) These headlines: Information To Provide Evaluating the Fire the Forecaster Weather Zone • Announce critical fire weather During initial or extended attack, Forecast conditions that need to be com detailed site-specific weather obser municated to the field complete vations can greatly improve weath NWS fire weather offices produce ly and accurately in all wildland er forecast accuracy. To enhance fire weather zone forecasts twice a firefighting situations. the information provided to the day and update as needed. The zone • Highlight significant weather weather forecaster, we recommend forecast provides weather informa conditions that do not meet the observations be taken: tion for relatively large areas. While warning or watch criteria but the most important purpose of the may require the IC’s heightened • At the same times each day. zone forecast is to issue and explain awareness. (These will reveal trends of tem Fire Weather Watches and Red Flag perature, humidity, and winds on Warnings, it also: IC’s should try to obtain a copy of the incident.) the entire fire weather zone fore •Across the range of elevations cast package or have the dispatcher and aspects of the incident, if
* Cellular phones can be especially useful because they read the applicable zone forecast possible. allow direct communication from the field to the weath over the radio. If receiving the er forecaster.
Fire Management Today 74 • At key (local) times of day: this onsite information, the IC can The forecaster will let the IC know – 0600–0800 for lowest tempera compare the observed weather to if more information is being ture and highest relative the weather forecast and then requested than the forecaster’s humidity (RH). develop a fire behavior prediction. workload will allow or if the – 1500–1700 for high tempera The key consideration for the IC: request is beyond the limits of the ture, low humidity, and always make the connection science of weather forecasting. strongest diurnal winds. between observed and forecasted weather and observed and forecast- Monitoring the The IC should also provide the fore ed fire behavior. Weather and caster with observations at various Evaluating a Forecast times of day to report such other Optimizing the IC’s can evaluate a forecast and data as: Spot Forecast decide when a new forecast is need The requestor has plenty of input ed by monitoring—through meas • The time the morning inversion into the Spot Forecast provided by urement and visual indicators—the broke. the fire weather forecaster. IC’s atmospheric instability, winds, tem • Diurnal wind shifts and the time perature, and RH. they occurred. • Cumulus cloud growth and thun During an extended Monitoring Instability. A highly derstorm development. unstable atmosphere is a primary •Precipitation. attack, appointing a cause of radical fire behavior. • Cloud cover. dedicated weather Strong instability can create erratic lookout or field observer winds and can greatly enhance fire During an extended attack, appoint to take observations growth. Cumulus cloud develop ing a dedicated weather lookout or ment and smoke column character field observer to take observations each hour is ideal. istics can be used as visual indica each hour is ideal. Observations tors of atmospheric instability. The from one well-trained individual should attempt to anticipate the fire weather forecast should provide will be consistent and will ensure kinds of information they will need IC’s with the predicted cumulus that quality weather observations and request that information. The development and instability condi are provided to the IC and the typical Spot Forecast includes: tions from which smoke column weather forecaster throughout the behavior can be estimated. course of the incident. •A weather discussion, • Forecasts of sky condition, Atmospheric conditions are more What Should Be Done • The chance of precipitation, unstable than predicted when: With Weather • High and low temperature and Observations? RH, 1. Cumulus clouds develop sooner The IC should pass all fire weather •Winds at eye level or 20 feet (6.1 and to greater heights than observations to the fire weather m) above ground, and expected.* forecaster. An observation from the • Smoke dispersal. 2. The smoke column rises faster fire site should be included with and to greater heights than every Spot Forecast request. If the IC’s can request more detailed expected. firefighting effort continues into a information when needed such as: second or third burning period, we Conditions are more stable than recommend all observations taken •A forecast of temperature, humid predicted when: during the previous burning period ity, and wind at 2- to 3-hour be included with the next Spot intervals. 1. Cumulus clouds develop later Forecast request. •A forecast for a single element, and/or to lesser heights than such as the 20-foot (6.1-m) wind, expected. A quality weather observation pro at 2- to 3-hour intervals. * Cumulus clouds may be more developed or cover a gram will also provide the IC with •A prediction of the time of high larger area if there is more moisture available in the critical information for input into est temperature and lowest RH. atmosphere, but the instability may not differ from the forecast. Fewer cumulus clouds or less vertical develop tactical firefighting decisions. With ment may mean drier conditions than expected.
Volume 64 • No. 1 • Winter 2004 75 2. The smoke column does not rise stronger wind speeds and gustier determine how accurate the fore as rapidly or as high as expected, conditions possible. cast is by comparing the forecasted or it does not develop at all. and observed data for that hour. The IC should consider requesting When evaluating atmospheric a new Spot Forecast if the shift to Temperature and RH are strongly instability, the IC should ask these upslope and upvalley winds is influenced by cloud cover. Often, questions: delayed by more than 1 hour or if small differences between observed the wind speed varies from the and forecasted temperature and RH • Does the atmosphere appear forecast by 5 mph (8 km/h) or can be accounted for by observing more unstable than expected? more. cloudiness. A 30-percent difference • If so, do we need to relay this in cloud cover may lead to a 1- to information to the weather fore When considering the wind direc 3-degree Fahrenheit (about 1 caster and ask for a new Spot tion, the IC should always be suspi degree Celsius) difference in tem Forecast? cious of any wind from a different perature and a 2- to 4-percent dif • How will greater instability affect ference in RH. The questions to fire behavior? ask: The key consideration When IC’s believe the observed for the IC: always make • Is the observed temperature instability conditions may signifi the connection between within 5 degrees of the forecast- cantly increase fire behavior, they ed temperature? should strongly consider requesting observed and • Is the observed RH within 5 per a new Spot Forecast. forecasted weather and cent of the forecasted RH? observed and Monitoring the Winds. Wind forecasted fire behavior. The IC should consider requesting observations taken every hour will a new Spot Forecast if the actual yield important information about temperature differs from the fore daily wind shifts and the strength direction than the terrain would be cast by 5 degrees or more and/or of valley breezes at differing eleva expected to produce. The question the actual RH differs from the fore tions. Accurate wind observations to ask: Does the wind direction fit cast by more than 5 percent. will record the true character of this terrain? If winds run counter local slope and valley breezes. Many to the normal slope and valley Note: When comparing observed factors can influence the develop breezes and these winds were not and forecasted temperature and ment of local winds, but cloudiness predicted, there may have been a humidity, be certain to take into is one of the most important and drastic change in weather condi account the effect that aspect, easiest to evaluate. Cloudiness over tions. The IC should consider cloud cover, sheltering, and eleva a site will affect surface heating and requesting a new Spot Forecast. tion will have on the observed val the shift in slope and valley breezes. ues. The ideal for comparative pur When examining the cloudiness at Monitoring Temperature and RH. poses would be to take observations the fire site, the IC should ask this If an observer is available, we rec from the same location exactly question: Is there more or less ommend monitoring the tempera throughout the course of the inci cloud cover than forecasted? Based ture and RH by plotting the fore dent. on the answer, the IC can draw cast temperatures and RH on graph some general conclusions: paper every 2 to.3 hours, then com Conclusion paring these plots to the observed As we have stressed, throughout • More clouds than predicted will data. (This procedure assumes the the incident, the IC should commu delay the shift to upslope and IC requested predictions of temper nicate as much information as pos upvalley winds and often result in ature and RH every 2 to 3 hours.) sible to the fire weather forecaster. lower wind speeds. An alternative would be to request As time permits, the IC should give • Less cloud cover than predicted a temperature and humidity fore the forecaster quality feedback on will cause an earlier shift to ups cast for a key decisionmaking time, forecast accuracy, observed weather lope and upvalley winds, with i.e., 1200 or 1300. The IC would conditions, and fire behavior.
Fire Management Today 76 We have summarized the recom- When IC’s believe the observed instability mendations presented here in the conditions may significantly increase fire behavior, Supplemental Observation Sheet and the Weather Evaluation Sheet. they should strongly consider requesting a new We recommend these two field aids Spot Forecast. be reproduced and carried to the field to be used with the “Mobile Fire-Weather Observer’s Record.” overall weather conditions will have fire ecologist, Arapaho-Roosevelt on fire behavior and firefighting National Forest; Bill Pies, initial When using the evaluation sheet, tactics. attack dispatcher, Deschutes please keep in mind that a single National Forest; Mark Rogers, weather element determined to be Acknowledgments superintendent, Wyoming Hot outside the criteria mentioned The authors want to thank the fol Shots; Alan Farnsworth, prescribed above may not require a request for lowing for their editorial and tech fire manager, Peaks Ranger a new Spot Forecast. A weather ele nical recommendations: Dave District, Coconino National Forest; ment outside the stated criteria Aldrich, national safety officer, Steve Keighton, science and opera may be offset by fuel conditions or Forest Service; Paul Broyles, chief, tions officer, NWS. other weather measurements. The Fire Training and Safety, National IC needs to consider what effect the Park Service; Rick Ochoa, NIFC Dedication: This article is dedicat ■ staff meteorologist; Paul Gleason, ed to the Storm King 14.
Russo and Schoemaker (1989) Decision Trap 8— Fooling Yourself About Feedback: Failing to interpret the evidence from past outcomes for what it really says, either because you are protecting your ego or because you are tricked by hindsight.* * See page 9.
Volume 64 • No. 1 • Winter 2004 77 BEYOND THE SAFETY ZONE: CREATING A MARGIN OF SAFETY*
Mark Beighley
ildland firefighting is fraught with hazards. When The safety zone and escape route network must W firefighters encounter those be an integral part of tactical fireline operations, hazards, they are at risk—risk of not added as an afterthought or after a fireline is injury, risk of death. To guarantee safety while wildfires are sup constructed. pressed, humans would have to stop being involved in firefighting. selecting a course of action. They The safety zone and escape route In most cases, this is not an option. may have as little as a few minutes network must be an integral part of We need firefighters to save lives, to conduct this risk analysis on tactical fireline operations, not protect communities, and reduce fast-spreading fires. On fires that added as an afterthought or after a damage to natural resources. Yet have not developed to their full, fireline is constructed. All fireline the question remains—how can explosive fury, firefighters may have construction should start from a firefighters suppress wildfires effi as much as several hours to analyze safe anchor point. As fireline con ciently without jeopardizing their their risk and decide what to do to struction proceeds from that safe own lives? maximize suppression effectiveness. point, safety zones are identified or constructed along the way. Any Firefighters Have No matter what course of action time firefighters venture beyond Alternatives firefighters choose, their decisions the safety zones, they are at risk. As Firefighters must consider current are not usually final because they the distance between the firefighter and future weather and burning must base their decisions on infor and the safety zone increases, so conditions and the effect they have mation that is incomplete. In addi does the risk of entrapment or on how, what, and where the fire is tion, information deteriorates burnover. expected to burn before making quickly with time. decisions about the best suppres Risk Threshold sion strategy to use. For any given Safety Zones At some distance from the safety suppression operation, firefighters A basic element of fire suppression zone, firefighters begin to feel can choose from a variety of strate safety is a safety zone, a place uncomfortable about their position. gic and tactical alternatives. Some where firefighters are free from This discomfort may result from alternatives maximize the effective danger, risk, or injury. It is vital increased fire activity or the threat ness of the suppression effort, and that firefighters know where and of increased fire activity. They real some maximize firefighter safety. how to get to areas that provide a ize that there may be insufficient Sometimes the most effective sup safe refuge when they analyze risk. time to successfully retreat to the pression action is also the safest, In any given tactical operation, fire safety zone should the need arise. but generally there is a tradeoff fighters must identify or create They have reached their risk between the two. Firefighters must safety zones and “escape routes” threshold—that point at which the always evaluate the risks before that provide access to them. For level of risk is too high. To reduce operational assignments that the level of risk, firefighters must When this article was originally published, require extensive and lengthy fire- then reduce the distance to a previ Mark Beighley was a team leader for line construction, firefighters must ous safety zone or locate or create a Resource Operations, USDA Forest Service, develop a network of safety zones new safety zone. Deschutes National Forest, Bend and Fort Rock Ranger Districts, Bend, OR. and escape routes. How is this net work designed? What factors should The risk threshold for all firefight
* The article is reprinted from Fire Management Notes be considered? ers is different. Every firefighter 55(4) [Fall 1995]: 21–24. possesses a different combination of
Fire Management Today 78 knowledge and experience with Even if firefighters have developed accurate risk which to evaluate the relative safety thresholds, they always have a degree of of the current situation. Fire- fighters may also have different uncertainty because of inadequate or deteriorating information regarding local factors information. that might affect fire behavior.
There is an assumption that veter sophical break-even point is the The second is the distance between an firefighters have well-defined, line between safe and unsafe fire- the firefighter and the closest safety accurate risk thresholds. Also, it is line operations. The firefighter zone and the time (T2) it would assumed that these risk thresholds must constantly evaluate where take for the firefighter to retreat to can be depended upon to provide a that line is and how close he or she it. Knowing these two times will consistent and appropriate assess is to it, given the current situation. allow the firefighter to determine ment of safety for any given tactical Uncertainty is always present. Risk where the line between a safe and fireline operation. But even if fire threshold is not measurable, there unsafe operation exists. For exam fighters have developed accurate fore not quantifiable. Firefighters ple, in figure 1, the firefighters esti risk thresholds, they always have a cannot measure how close they are mate that it will take 18 minutes degree of uncertainty because of to an unsafe situation. Only the fire (T1) for the fire to reach the safety inadequate or deteriorating infor can provide feedback to the accura zone and 12 minutes (T2) for them mation. Because conditions on a cy of their risk threshold. to reach the zone. fire seldom stay constant for more than a few hours and can change Quantifying Fireline Creating a Margin of quite rapidly, a constant supply of Safety Safety information is an important facet of Without the ability to measure the A margin of safety can be described the risk assessment process. safety of their position, firefighters as a cushion of time in excess of will not consistently know when a the time needed by the firefighters When Safe Becomes safe situation becomes unsafe. to get to the safety zone before the Unsafe What is safe in the morning could fire gets to them. It is the positive Risk threshold applications are, for become unsafe in the afternoon. difference of T1 – T2. In figure 1, tunately, rarely tested. Even when What is safe about their current the difference is 6 minutes (18 firefighters are uncomfortable with position could become unsafe as minutes – 12 minutes), so the fire their position, the fire does not they continue to build fireline. fighters are in a safe position. If T1 always test the situation. Feedback = T2 as in figure 2, the difference is on risk threshold is infrequent; In order to assure safe fireline oper 0 and the fire and firefighters arrive therefore the accuracy of a fire ations, firefighters need processes at the safety zone at approximately fighter’s risk threshold is often to evaluate fireline safety that are the same time. Obviously, this situ unknown. Even under the best of measurable, consistent, and trans ation would not benefit the fire circumstances, the most experi ferable. When they can measure fighters; the fire may block their enced and knowledgeable firefight how safe they are, firefighters can planned escape route. At best, they ers are plagued with imperfections repeat that safety measurement and would experience a very close call, inherent in the human condition. communicate it to others. They will so they need to evaluate their mar Inattention, distraction, fatigue, be able to describe what is safe and gin of safety for escape or build a attitude, boredom, information unsafe and evaluate the safety of new safety zone. overload, mind set, and carbon their current and planned actions. monoxide poisoning can all work to If the difference is less than 0 as in erode the judgment of the most Two distance and time relationships figure 3 (T1 is 12 minutes and T2 is vigilant of firefighters. must be evaluated by firefighters 15 minutes equaling –3 minutes), before they can determine how safe then it is likely that the fire will Safe becomes unsafe when the fire they are. The first is the distance reach the firefighters before they has the potential to get to the fire between the fire and the safety zone get to the safety zone. While we fighter before the firefighter can and the time (T1) it would take the would hope that firefighters would get to a safety zone. That philo fire to spread to the safety zone. deploy fire shelters and survive the
Volume 64 • No. 1 • Winter 2004 79 In order to assure safe fireline operations, firefighters need processes to evaluate fireline safety that are measurable, consistent, and transferable.
fire, for a margin of safety, firefight ers must arrive at the safety zone before the fire. T2 must be less than T1. In this example, the fire fighters need to locate or construct a closer safety zone, abandon their suppression effort, or the fire behavior characteristics need to change. In short, the greater the positive difference between T1 and T2, the greater the margin of safety.
Figure 1—T1 is estimated at 18 minutes—the time it would take a fire to reach safety Firefighters should increase their zone 3 (SZ3). T2—the time it would take a firefighter to reach SZ3—is tested at 12 min margin of safety when there is an utes. A 6-minute margin of safety exists, and firefighters are in a safe position. increase in uncertainty. Uncertainty can come from many situations. Firefighters can be uncertain about future weather conditions, a specif ic fire location, expected fire behav ior in local fuel types, their own and others’ physical ability, and the effectiveness of control actions on adjacent divisions or other fires in the immediate area. Firefighters must consider these variables when managing a margin of safety. There should never be any uncertainty about the location of safety zones and escape routes, the adequacy of communications, or the posting of lookouts. Knowing When “Safe” Becomes “Unsafe” Firefighters can use the T1 and T2 concept to provide a measurable, consistent, and transferable process to assess their margin of safety. This will enhance the value of Figure 2—It is estimated that the fire will reach safety zone 4 (SZ4) 20 minutes after it L.C.E.S. applications—Lookouts, begins to run—T1 on the map. The time it would take a firefighter to reach SZ4 is the Communications, Escape Routes, same (T2 = 20 minutes). There is no margin of safety. and Safety Zones. Firefighters will
Fire Management Today 80 There should never be any uncertainty about the location of safety zones and escape routes, the adequacy of communications, or the posting of lookouts.
be able to identify when “safe” will become “unsafe” and communicate that to all affected personnel. They will know when to look for new safety zones and when escape route travel times are too long.
For large fire operational planning, this assessment can be conducted prior to committing firefighters to a fireline assignment. Safety zone and escape route requirements can be identified in the Incident Action Figure 3—T1—the time for the fire to reach safety zone 5 (SZ5)—is estimated at 12 min utes after the fire run begins. T2—the time it would take a firefighter to reach SZ5—is Plan. A network of safety zones and tested at 15 minutes, an unsafe situation. escape routes can then be devel oped in conjunction with fireline construction. Firefighters will be able to create and maintain a mar gin of safety when they are beyond the safety zone. ■
Russo and Schoemaker (1989) Decision Trap 9—Not Keeping Track: Assuming that experience will make its lessons available automatically, and therefore failing to keep systematic records to track the results of your decisions and failing to analyze these results in ways that reveal their key lessons.* * See page 9.
Volume 64 • No. 1 • Winter 2004 81 FIREFIGHTER SAFETY ZONES: HOW BIG IS BIG ENOUGH?*
Bret W. Butler and Jack D. Cohen
ll wildland firefighters working is worn, second degree burns will on or near the fireline must be How much heat can occur after 90 seconds when a fire able to identify a safety zone. humans endure before fighter is subjected to radiant fluxes A 2 2 Furthermore, they need to know injury occurs? greater than 0.6 Btu/ft /s (7 kW/m ). how “big” is “big enough.” The Nomex shirts and trousers cur Beighley (1995) defined a safety define more clearly the relationship rently used by wildland firefighters zone as “an area distinguished by between convective heating and have fabric weights of 5.7 and 8.5 2 2 characteristics that provide free safety zone size in future work. oz/yd (190 and 280 g/m ), respec dom from danger, risk, or injury.” tively. Few studies, however, have The National Wildfire Coordinating What Do We Know? explored relationships between Group proposed that a safety zone flame height and the safety zone be defined as “a preplanned area of Two questions are important when size necessary to prevent burn sufficient size and suitable location specifying safety zone size: injury. that is expected to prevent injury to fire personnel from known hazards 1. What is the radiant energy distri Theory Versus Reality bution in front of a flame? and without using fire shelters” We formulated a theoretical model (USDA/USDI 1995). 2. How much heat can humans endure before injury occurs? to predict the net radiant energy arriving at the firefighter wearing In our study of wildland firefighter Concerning the first question, Nomex clothing as a function of safety zones, we focused on radiant flame height and distance from the heating only. In “real” wildland Fogarty (1996) and Tassios and Packham (1984) related the energy flame (Butler and Cohen 1998). fires, convective energy transport in Figure 1 displays the results. the form of gusts, fire whirls, or received by a firefighter to fireline turbulence could contribute signifi intensity and distance from the flame front. Green and Schimke The amount of radiant energy cantly to the total energy received arriving at the firefighter depends by a firefighter. However, convec (1971) presented very specific infor mation about fuel break construc both on the distance between the tion is subject to buoyant forces firefighter and the flame and on the and turbulent mixing, both of tion on slopes and ridges in the Sierra Nevada mixed-conifer forest flame height. The information which suggest that convective heat shown suggests that in most cases ing is important only when a fire type. Others have discussed the per formance of fire shelters under dif safety zones must be relatively fighter is relatively close to the fire. large to prevent burn injury. One reason that firefighters in ferent heating regimes (for exam ple, King and Walker 1964; Jukkala potential entrapment situations are We compared safety zone sizes pre told to lie face down on the ground and Putnam 1986; Knight 1988). As one would expect, there is not dicted by our model against those is to minimize their exposure to reported on four wildfires: the convective heating. We hope to much information related to the second question. The available Mann Gulch Fire, the Battlement information suggests that 0.2 Creek Fire, the Butte Fire, and the When this article was originally published, 2 2 South Canyon Fire. Bret Butler and Jack Cohen were research Btu/ft /s (2.3 kW/m ) is the upper scientists for the USDA Forest Service, Fire limit that can be sustained without Behavior Research Unit, Rocky Mountain injury for a short time (Stoll and The Mann Gulch Fire overran 16 Research Station, Intermountain Fire firefighters on August 5, 1949. Wag Sciences Laboratory, Missoula, MT. Greene 1959; Behnke 1982). Studies by Braun and others (1980) Dodge, one of only three survivors, suggest that when a single layer of lit a fire and then lay face down in * The article is reprinted from Fire Management Notes 2 2 the burned-out area as the main 58(1) [Winter 1998]: 13–16. 6.3 oz/yd (210 g/m ) Nomex cloth
Fire Management Today 82 fire burned around him. The Mann The amount of radiant energy arriving at the Gulch Fire occurred in an open firefighter depends both on the distance between stand of scattered, mature pon derosa pine (60 to 100+ years old) the firefighter and the flame and on the flame with a grass understory. Flame height. heights of 10 to 40 feet (3–12 m) were estimated to have occurred at the time of entrapment. Rothermel The Battlement Creek Fire oc his body. Figure 1 suggests that for (1993) indicates that Dodge’s fire curred in western Colorado during this fire, the safety zone should burned about 300 feet (92 m) July of 1976 (USDI 1976). The fire have been large enough to separate before the main fire overran it. burned on steep slopes covered firefighters from flames by 150 feet Assuming an elliptical shape for the with 6- to 12-foot- (2- to 4-m-) (46 m). Clearly, the 25-foot- (8-m-) burned area, with its width approxi high Gambel oak. Flames were esti wide clearing did not qualify as a mately half the length, the safety mated at 20 to 30 feet (6–9 m) safety zone. zone created by Dodge’s escaped above the canopy. Four firefighters fire would have been about 150 feet were cut off from their designated Flame heights were reported to be (46 m) wide. Figure 1 indicates that safety zone. When the fire overran 200 to 300 feet (62 to 92 m) high the safety zone needed to be large them, they were lying face down on on the Butte Fire that burned on enough to separate the firefighters the ground without fire shelters in steep slopes covered with mature and flames by 90 to 150 feet (27 to a 25-foot- (8-m-) wide clearing near lodgepole pine and Douglas-fir dur 46 m) or approximately the same the top of a ridge. Tragically, only ing August of 1985 (Mutch and width as the area created by one of the four survived, and he Rothermel 1986). Figure 1 indi Dodge’s fire. suffered severe burns over most of cates that a cleared area greater than 1,200 feet (370 m) across would have been needed to prevent injury to the firefighters standing in its center. In fact, safety zones 300 to 400 feet (92 to 123 m) in diameter were prepared (Mutch and Rothermel 1986). This diameter was not sufficiently large enough to meet the definition of a safety zone, as indicated by the fact that 73 fire fighters had to deploy in fire shel ters to escape the radiant heat. As the fire burned around the edges of the deployment zone, the intense heat forced the firefighters to crawl while inside their shelters to the opposite side of the clearing.
On July 2, 1994, the South Canyon Fire was ignited by a lightning strike to a ridgetop in western Colorado. During the afternoon of July 6, the South Canyon Fire “blew up,” burning across the pre dominately Gambel-oak-covered slopes with 50- to 90-foot- (15- to Figure 1—Lines represent predicted radiant energy arriving at the firefighter as a func 28-m-) tall flames (South Canyon tion of flame height and distance from the flame. It is assumed that the firefighter is Fire Accident Investigation Team wearing fire-retardant clothing and protective head and neck equipment. The heavy shad ed line represents the burn injury threshold of 0.6 Btu/ft2/s (7 kW/m2). The heavy solid 1994). Tragically, 14 firefighters black line indicates the rule of thumb for the size of the safety zone. were overrun by the fire and died
Volume 64 • No. 1 • Winter 2004 83 while attempting to deploy their A safety zone should be dictions shown in figure 2 suggest fire shelters. Twelve of the firefight large enough that the that heat levels remain below the ers died along a 10- to 12-foot- (3 injury limits for deployment zones to 4-m-) wide fireline on a 55-per distance between the wider than 50 feet (15 m), even cent slope, the other two in a steep firefighters and flames with 300-foot- (92-m-) tall flames. narrow gully. Eight other firefight is at least four times However, this model does not. ers deployed their fire shelters in a the maximum flame account for convective heating that burned out area approximately 150 height. could significantly increase the feet (46 m) wide. They remained in total energy transfer to shelters their shelters during three separate deployed within a few flame lengths crown fire runs that occurred 450 zone, meaning the safety zone of the fire. feet (138 m) away from them; none diameter should be twice the value of these eight firefighters was indicated. Conclusions injured (Petrilli 1996). One fire Radiant energy travels in the same fighter estimates that air tempera What About Fire form as visible light, that is, in the tures inside the shelters reached Shelters? line of sight. Therefore, locating 115 °F (46 °C) and remembers We calculated the net radiant ener safety zones in areas that minimize smoke and glowing embers enter firefighters’ exposure to flames will ing the fire shelters during the gy transferred through a fire shel ter like those used by firefighters in reduce the required safety zone crown fire runs. Survivors felt they size. For example, topographical were far enough from the flames the USDA Forest Service. The fire shelter is based on the concept that features that act as radiative shields that survival with minor injuries are the lee side of rocky outcrop would have been possible without the surface will reflect the majority of the incoming radiant energy. An pings, ridges and the tops of ridges, the protection of a fire shelter or peaks containing little or no (Petrilli 1996). A firefighter who did average emissivity for the alu minum-foil exterior of a fire shelter flammable vegetation. Safety zone not deploy in a shelter but size is proportional to flame height. remained on a narrow ridge below is 0.07, indicating that approxi mately 93 percent of the energy Therefore, any feature or action the eight firefighters during the that reduces flame height will have “blowup” experienced no injuries incident on a fire shelter is reflect ed away (Putnam 1991). Model pre a corresponding effect on the (South Canyon Fire Accident required safety zone size. Some Investigation Team 1994). Figure 1 suggests that in this situation, the safety zone must be large enough to separate the firefighters and flames by 250 to 350 feet (77–115 m).
A general rule of thumb can be derived from figure 1 by approxi mating the injury limit with a straight line. After doing so, it appears that a safety zone should be large enough that the distance between the firefighters and flames is at least four times the maximum flame height. In some instances such as the Mann Gulch, Battle ment Creek, and Butte Fires, the fire may burn completely around the safety zone. In such fires, the separation distance suggested in Figure 2—Predicted radiant energy on a fire shelter as a function of distance between the figure 1 is the radius of the safety fire shelter and flames, and flame height. The heavy shaded line represents the burn injury threshold for a firefighter inside a deployed fire shelter.
Fire Management Today 84 examples are burnout operations Locating safety zones in areas that minimize that leave large “black” areas, thin firefighters’ exposure to flames will reduce the ning operations that reduce fuel load, and retardant drops that required safety zone size. decrease flame temperatures. Beighley, M. 1995. Beyond the safety zone: Petrilli, A. May 21, 1996. [Personal commu We emphasize that while this study Creating a margin of safety. Fire nication with B. Butler]. Missoula, MT. addresses the effects of radiant Management Notes. 55(4): 22–24. Putnam, T. 1991. Your fire shelter: Pub. Braun, E.; Cobb, D.; Cobble, VB.; Krasny, NFES 1570. Boise, ID: National energy transfer, convection is not J.F.; Peacock, R.D. 1980. Measurement of Interagency Fire Center, National Wildfire addressed. Convective energy trans the protective value of apparel fabrics in a Coordinating Group, National Fire fer from gusts, fire whirls, or tur fire environment. Journal of Consumer Equipment System, 18 p. bulence could significantly increase Product Flammability. 7: 15–25. Rothermel, R.C. 1993. Mann Gulch Fire: A Butler, B.W.; Cohen, J.D. 1998. Firefighter race that couldn’t be won. GTR INT–299. the total heat transfer to the fire safety zones: A theoretical model. Ogden, UT: U.S. Department of fighter and thus the required safety International Journal of Wildland Fire. Agriculture, Forest Service, zone size. Further work in this area 8(2): 73–77. Intermountain Research Station. 10 p. Fogarty, L.G. 1996. Two rural/urban inter South Canyon Fire Accident Investigation is needed. face fires in the Wellington suburb of Team. 1994. Report of the South Canyon Karori: Assessment of associated burning Fire Accident Investigation Team. Atlanta, Acknowledgments conditions and fire control strategies. FRI GA: U.S. Department of Agriculture Bulletin No. 197, Forest and Rural Fire Forest Service, Southern Region. 39 p. The United States Department of Scientific and Technical Series, Rep. No. plus appendices. the Interior’s Fire Coordinating 1. Rotorua and Wellington, NZ: New Stoll, Alice M.; Greene, Leon C. 1959. Zealand Forest Research Institute in asso Relationship between pain and tissue Committee, Boise, ID, provided ciation with the National Rural Fire damage due to thermal radiation. Journal financial assistance for a portion of Authority. 16 p. of Applied Physiology. 14(3): 373–382. this study. Ted Putnam of the Green, L.R.; Schimke, H.E. 1971. Guides Tassios, S.; Packham, D. 1984. An investiga Forest Service’s Missoula Tech for fuel-breaks in the Sierra Nevada tion of some thermal properties of four mixed-conifer type. Berkeley, CA: U.S. fabrics suitable for use in rural firefight nology and Development Center, Department of Agriculture, Forest ing. National Center for Rural Fire Missoula, MT, provided valuable Service, Pacific Southwest Forest and Research. Tech. pap. No. 1. Canberra, information and advice on the Range Experiment Station. 14 p. ACT, Australia: Forest Research Institute, Jukkala, A.; Putnam, T. 1986. Forest fire Forestry and Timber Bureau. 13 p. effects of heat on human tissue. shelter saves lives. Fire Management U.S. Department of the Interior. Notes. 47(2): 3–5. [Unpublished July 17, 1976, report]. References King, A.R.; Walker, I.S. 1964. Protection of Accident report of Battlement Creek Fire forest firefighters. Unasylva. 18(1): 29–32. fatalities and injury. U.S. Department of Behnke, W.P. 1982. Predicting flash fire Knight, Ian. 1988. What intensity of fire the Interior. 125 p. protection of clothing from laboratory can a fire fighter survive in a reflective USDA/USDI. [Review copy, 10/31/951. tests using second degree burn to rate shelter? Fire Technology. 24(4): 312–332. Glossary of wildland fire terminology. performance. London: International Mutch, R.W; Rothermel, R.C. 1986. 73 fire Boise, ID: National Interagency Fire Conference on Flammability; 30 p. fighters survive in shelters. Fire Center, National Fire and Aviation Command. 53(3): 30–32, 48. Support Group. 160 p. ■
Russo and Schoemaker (1989) Decision Trap 10— Failure to Audit Your Decision Process: Failing to create an organized approach to understanding your own decisionmaking, so that you remain constantly exposed to all the above mistakes.* * See page 9.
Volume 64 • No. 1 • Winter 2004 85 SAFETY ALERT: WATCH OUT FOR AIRCRAFT TURBULENCE!*
Billy Bennett
ircraft play a vital role in today’s fire control operations, Aircraft turbulence should be one of the unwritten A carrying out such crucial mis Watch Out Situations for firefighters to keep in sions as water and fire retardant mind on the fireline. drops. Yet turbulence from aircraft can sometimes contribute to erratic fire behavior, potentially endanger When the helicopter approached •A sudden wind shift had already ing firefighters. As the National the area near the dozer, the rotor caused the fire to jump contain Wildfire Coordinating Group notes downwash caused the fire to behave ment lines. in a training publication for fire erratically, encircling the immedi Watch Out Situations: fighters, “The blasts of air from low ate area around the dozer and #15 Wind increases and/or flying helicopters and air tankers dozer boss with fire. The only changes direction. have been known to cause flare escape was to push through the #16 Getting frequent spot fires ups” (NWCG 1992). Those on the active fire into the safety zone of across line. fireline should keep this potential the black. As the dozer operator • The incident occurred in a some hazard in mind, mentally adding it bladed through the fire, the dozer what narrow part of the canyon, to their list of 18 Watch Out boss followed close behind, using where topography might have Situations. the dozer as a heat shield. They influenced fire behavior. managed to escape unharmed. • When the helicopter pilot ap Incident Within an proached the slopover, he could Incident Contributing Factors not make radio contact with fire A case in point occurred on July 11, Several factors contributed to this fighters on the ground. This 1996, on the Broad Canyon Fire in near-tragic incident, including cir caused a delay, because the pilot central Utah. At about 3 p.m., a cumstances clearly identifiable as did not know specifically where wind shift caused the fire to jump Watch Out Situations: to make the drop. containment lines during a burn Watch Out Situations: out operation. A Cat D–7 dozer and •Available fuels were very dry and #5 Uninformed on strategy, dozer boss began constructing line extremely volatile. tactics, and hazards. around the slopover, which was burning in brush and 15-foot (4.6 m) juniper.
A type 2 helicopter using a bucket with a 35-foot (10.7-m) line began making drops along the fire edge.
When this article was first published, Billy Bennett was a law enforcement officer and fire management officer for the South Carolina Forestry Commission, Piedmont Region, Spartanburg, SC. In July 1996, he was the Staging/Initial Attack Safety Officer for the USDA Forest Service and USDI Bureau of Land Management in cen tral Utah. Resources assembling for the initial attack on the Broad Canyon Fire in central Utah, * The article is reprinted from Fire Management Notes July 1996. Photo: Billy Bennett, South Carolina Forestry Commission, Spartanburg, SC, 58(4) [Fall 1998]: 20–21. 1996.
Fire Management Today 86 #6 Instructions and assign ments not clear. #7 No communication link with crew members/super visor. • The airspeed of the helicopter as it approached the scene was about 46 miles per hour (74 km/h), and altitude was less than 200 feet (61 m) above ground level. Firefighters on the ground believe that this was too low under the conditions, and the pilot now concurs. • The helicopter was large enough to cause substantial rotor down- wash (the larger the helicopter, the more rotor downwash to expect).
If any of these contributing factors had been removed, the incident likely would not have occurred. However, rotor downwash was probably the final contributing fac tor to the erratic fire behavior and resulting entrapment. The firefight ers were operating within accept able risk limits before the helicop ter arrived, having to some extent compromised only a minimum number of Watch Out Situations. Not until the helicopter arrived did acceptable risk escalate into unac ceptable risk within just a matter of seconds.
Unwritten Watch Out Fire behavior in brush–juniper fuels on the Broad Canyon Fire in central Utah, July 11, 1996. Fuels were extremely dry and volatile. Photo: Billy Bennett, South Carolina Situation Forestry Commission, Spartanburg, SC, 1996. One of the most important func tions of fire managers on the fire- Watch Out Situation or Standard always cause erratic fire behavior. line is to recognize when Watch Fire Order specifically addresses However, it does suggest that air Out Situations and Standard Fire how aircraft turbulence affects fire craft turbulence should be one of Orders are excessively compro behavior. Pilots and firefighters the unwritten Watch Out mised, and to take immediate cor should keep in mind that low or Situations for firefighters to keep in rective action to ensure firefighter moderate hazards, under certain mind on the fireline. safety. Pilots will most likely not conditions, can quickly become know how close firefighters on the high or extreme hazards due to Literature Cited: ground are to this point of unac unexpected aircraft turbulence. NWCG (National Wildfire Coordinating ceptable risk. When air operations Group). 1992. Common denominators of are in progress, pilots and firefight This incident in no way suggests fire behavior on tragedy and near-miss forest fires. NFES 2225. Boise, ID: ers alike must remember that no that turbulence from aircraft will National Interagency Fire Center. ■
Volume 64 • No. 1 • Winter 2004 87 THE CONSUMPTION STRATEGY: INCREASING SAFETY DURING MOPUP*
Tom Leuschen and Ken Frederick
or many years, the wildland fire community has known that The consumption strategy for mopup exploits a F mopping up a fire can be just as fire’s natural tendency to consume its fuels and dangerous as containing and con burn itself out. trolling it. Unfortunately, we have not always done the best job in mit igating the hazards that firefighters 1. Mopup strategy and standards sidered for lengthy operational are exposed to during this impor flow from a determination made assignments that could place tant phase of fire suppression. about the fire’s potential to crews in harm’s way. escape across firelines after it is 4. Sections of the fire avoided due A new approach is now available for declared contained. to gravity-related hazards are assessing the need for, and accom 2. Sections of the fire that show a still patrolled or otherwise moni plishing, mopup on wildland fires. high potential for escape receive tored. “Patrolling” means that Known as the consumption strate the normal mopup treatment. crews or scouts hike along fire- gy, the new approach departs from 3. Sections of the fire that do not lines in the avoided areas (stay traditional thinking by using the show a high potential for escape ing alert for falling or rolling natural tendency of a fire to burn and that contain significant grav material) to check for escapes of itself out by consuming its fuel. ity-related hazards are not con- the fire across firelines but not The consumption strategy realisti cally compares the risks and conse quences associated with an escaped fire to the risks and consequences associated with the hazards fire fighters typically face during mop- up, which tend to be related to gravity (falling snags, rolling mate rials, and tripping and falling). The strategy is designed to improve fire fighter safety while still suppressing a fire.
The consumption strategy is planned during containment and implemented during control or mopup. It includes these steps (fig. 1):
When this article was first published, Tom Leuschen was a fire and fuels specialist for the USDA Forest Service, Okanogan National Forest, Okanogan, WA; and Ken Frederick was an information assistant for the Forest Service, Wenatchee National Forest, Chelan Ranger District, Chelan, WA.
* The article is reprinted from Fire Management Notes Figure 1—Consumption strategy decision tree, for application separately to each 59(4) [Fall 1999]: 30–34. section of the fire.
Fire Management Today 88 The consumption strategy reduces gravity-related risks to firefighters during mopup, such as falling trees, slippery slopes, and rolling rocks, logs, and stumps.
to extinguish flames or embers hazardous areas to do mopup when really needed had nearly cost a life. within the firelines. there was minimal risk of the fire 5. Operational assignments in escaping.” By the third day of the The Gold Creek incident made it avoided areas can include, in Gold Creek Fire, Leuschen had increasingly obvious that we need a addition to patrolling, tasks such hiked the perimeter of the fire and strategy for assessing risk to reduce as blacklining (burning fuels determined that the blaze posed lit firefighters’ exposure to hazards adjacent to firelines), flush-cut tle threat of escaping. However, the during mopup. Since the South ting stobs, trimming tree operations and plans sections of the Canyon tragedy in 1994, risk branches immediately inside the type 2 team managing the fire were assessment has focused primarily lines, and gridding (searching still trying to control the fire on avoiding fire entrapments. In systematically along gridlines) according to standards agreed to by recent years, the wildland fire com for spot fires well outside of the the local line officer and the inci munity has paid more attention to lines. Firelines can be strength dent management team—and that mitigating risk during containment ened, as long as crews maintain included risky mopup work inside and control (constructing and good lookouts and do not linger the black. securing firelines) than during in dangerous spots. mopup. We need to rethink what After the accident, Leuschen and mopup is. Are we out there trying Origins of the the district ranger walked out to to physically put out every flame Consumption Strategy the lines with the incident com and ember, or are we trying to pre The consumption strategy originat mander, safety officer, and opera vent the fire from escaping control ed in response to a near tragedy tions section chief to take a sober lines while those flames and during the 1997 fire season. The look at the work. Although discus embers burn out? Depending on season was relatively quiet in east sion continued to focus on how the situation, we currently do both; ern Washington. In fact, the only firefighters could work safely inside but we should remember to distin project fire on the Wenatchee the lines, Leuschen questioned guish between the two and to National Forest was the Gold Creek whether firefighters needed to work choose the approach that best pro Fire on the Naches Ranger District inside the black at all. Areas where tects our crews. in August 1997, which burned firefighters had completed several about 480 acres (190 ha) of pon shifts of mopup showed little differ Managers’ perceptions of the risks derosa pine and Douglas-fir near ence in the kinds and amounts of to firefighters must change with Cliffdell, WA. During mopup on the smoldering debris from similar changes in a given fire. At a certain incident, a Washington Department areas where no mopup had oc point in a fire, the primary danger of Natural Resources crewmember curred. Residual interior smokes facing firefighters is no longer the was struck and seriously injured by were not a threat to the lines. fire itself, but rather falling or a snag being felled by a sawyer. Furthermore, a large percentage of rolling objects (fig. 2). As the fire Ironically, the accident occurred the fire perimeter consisted of sec nears containment, entrapment when areas inside the fireline were tions where the fire had backed risk decreases but gravity-related being “snagged” for firefighter safe downhill; in order to escape in risk increases. Trees, both live and ty. these areas, the fire would have to dead, with fire in their bases jump the lines and aggressively become increasingly unstable; Tom Leuschen, the fire and fuels spread downhill, a highly unlikely stumps roll as they lose the old, dry specialist for Washington’s eventuality. “As a result of our roots that have held them on the Okanogan National Forest, was on observations,” Leuschen said, “we slope; and firefighter fatigue accu the Gold Creek Fire as a fire behav recommended a change in mopup mulates, reducing energy and alert ior analyst. “It occurred to me,” standards to the line officer.” The ness and causing more tripping and Leuschen recalled, “that we were group had learned a lesson: Per falling on steep terrain. asking the firefighters to work in forming mopup where it wasn’t
Volume 64 • No. 1 • Winter 2004 89 Entrapment during mopup obvi inaccessible by road, convention- 3. Reduced cost. Assisted by the ously remains a serious risk that al mopup was likely to involve consumption of available fuels, overhead and crews must never for lots of crews, long hoselays, and mopup would cost less than tra get. However, we must elevate our significant helicopter use. ditional, labor-intensive mopup. awareness of the risks to firefight ers from gravity-related hazards during mopup. Operational Success In August 1998, the 8,500-acre (3,400-ha) North 25 Fire on the Wenatchee National Forest’s Chelan Ranger District in Washington pro vided the first opportunity to implement the consumption strate gy. A number of factors coincided to make testing possible under actual field conditions. First, Tom Leuschen was detailed to the dis trict as the fire management officer for the summer. Second, the Central Washington Area Incident Command Team, the same team that had handled the Gold Creek Fire, was assigned to manage the Figure 2—Consumption strategy risk assessment on a fire in coniferous forest that is North 25 Fire when it escaped ini contained after 3 days. As the fire nears containment, gravity-related risks (such as falling tial attack. With the Gold Creek trees, slippery slopes, and rolling rocks and stumps) exceed risks from an escaped fire. In sections of the fire where gravity-related risks exceed the risk of fire escape (the no-work experience still fresh in their zone), mopup should be avoided. minds, the team’s leaders were will ing to consider a new approach. Third, District Ranger Al Murphy and Forest Supervisor Sonny O’Neal were both willing to accept the possibility of a longer lasting or larger fire if the consumption strat egy were implemented. Finally, the North 25 Fire had the topographi cal and fuel conditions necessary for applying the new approach (fig. 3).
Implementing the consumption strategy on the North 25 Fire offered several immediate benefits:
1. Reduced risk of firefighter injury due to falling and rolling materi als on steep, rocky slopes. Figure 3—An Erickson S–64 helicopter drops water on an inaccessible spot fire, part of the North 25 Fire, Chelan Ranger District, Wenatchee National Forest, WA, in August 2. Reduced need for resources and 1998. The steep terrain and poor accessibility of the site called for applying the consump labor. Because much of the tion strategy, which succeeded in controlling the fire while minimizing the risks to fire North 25 Fire’s perimeter was fighters from gravity-related hazards such as falling snags and rolling logs. Photo: Paige Houston, USDA Forest Service, Okanogan National Forest, Tonasket Ranger District, Tonasket, WA, 1998.
Fire Management Today 90 The consumption strategy saves labor and than 7 miles (11.2 km) of the 9.5 reduces costs, freeing resources for use on other miles (15.2 km) of accessible perimeter were allowed to smolder incidents. under the watchful eyes of daily patrols. There were no accidents 4. Reduced spread of noxious noted. “I had a number of superin during mopup and no significant weeds, particularly the diffuse tendents come up to me and thank escapes. Because almost no hose knapweed (Centaurea diffusa). us for using this approach.” was laid and operations were much Ranger Murphy saw that tilling Twenty-two interagency hotshot less labor intensive than under the less soil would reduce the crews from the Pacific Northwest conventional mopup approach, amount of prepared seedbed for and California were on the North seven crews could be freed right weed propagation. “The North 25 25 Fire. away for fire assignments else Fire burned on both sides of one where. of the busiest roads on this dis The consumption strategy succeed trict,” he said. “The less ground ed. About a quarter of the fire Lessons Learned we dig up, the more we prevent perimeter was never considered for Several lessons can be learned from weeds from spreading outside of direct attack, let alone mopup, our experience with the consump the road corridor.” because it was on an extremely tion strategy on the North 25 Fire: steep, rock-strewn slope overlook The incident management team ing Lake Chelan (fig. 4). Around •Firefighters should mop up in carefully briefed all operational per the remainder of the fire, the oper areas of high gravity-related haz sonnel on why and how the new ations section chiefs opted for ard only when necessary. Too mopup standards were to be imple intensive mopup on only 22 per often we approach mopup based mented on the fire. Even after sev cent of the firelines, based on the on tradition and habit. Especially eral briefings, however, some crews prevalence of unburned fuels next in an age of increasingly large still had trouble accepting the idea to the lines. For 3 to 5 days, more fires across the West, the same of merely patrolling firelines for 3 to 5 days while allowing the fire to consume fuels just inside the lines. “This approach is a cultural shift in how we manage fires,” said Incident Commander Jim Furlong. “We are used to being aggressive in extin guishing fires, so being patient like this feels a little unnatural.” Some crews modified their line patrol assignments by scavenging a 20 foot (6.2-meter) strip of ground just inside the lines for fuel and then constructing and burning numer ous small handpiles. The result was a cleanly burned and very secure blackline.
According to Furlong, many crews understood that the incident man agement team was looking out for Figure 4—The North 25 Fire burns deep in Box Canyon on the south shore of Lake firefighter safety in using the con Chelan, Chelan Ranger District, Wenatchee National Forest, WA, in August 1998. About a sumption strategy. “The crews that quarter of the fire perimeter was never considered for direct attack, let alone mopup, picked up on what we were doing because it was on an extremely steep, rock-strewn slope overlooking the lake. The con were the hotshot crews,” Furlong sumption strategy is well suited for consideration on such sites. Photo: Paige Houston, USDA Forest Service, Okanogan National Forest, Tonasket Ranger District, Tonasket, WA, 1998.
Volume 64 • No. 1 • Winter 2004 91 safety mindset should prevail for less risk to firefighter safety. The mopup is making sure the fire mopup as for line construction. risk of escape is often only mar does not cross control lines. Sometimes it might be safer and ginally higher under the con Making this subtle distinction more sensible to be vigilantly sumption strategy. will help fire managers and fire patient for a few days while a fire •Fire behavior analysts should fighters avoid the potentially high consumes its fuels than to measure the potential for escape costs of doing what the fire will aggressively put it out. on each section of line as it is likely do by itself—given just a • Line officers and fire managers completed. Each section must little time. on project fires should reflect also be evaluated for gravity- upon what might be a false sense related hazards. These data must Safety must always be our first pri of insecurity regarding how thor then be presented to the line offi ority in suppressing wildland fires. oughly a fire should be extin cer for determining mopup stan Applied correctly, the consumption guished before the local adminis dards. strategy offers a safer, more cost- trative unit reassumes responsi • Although perceiving mopup as effective means of achieving the bility for the fire. Line officers putting out the fire is often same objective—wildland fire sup should consider accepting more appropriate, sometimes a more pression. ■ risk of fire escape in exchange for reasonable interpretation of
Wildland Fire Research’s Raison D’etre* One basic presupposition seems just one justification for existence, measure necessary for the suc to be essential, and to demand just one function, just one objec cessful practice of forestry is pro- full agreement and understand tive. That is: To aid the present and tection from forest fires.” Fire ing…. This is the premise that future administrators of fire con- research is therefore intended to all of our experiment station trol, Federal, State, and private. We serve as directly as possible the divisions of fire research have are not doing research for fire-control men who must first research’s sake. We have a definite, be successful before any of the
* From H.T. Gisborne “Review of Problems and decidedly practical goal, and it is other arts or artists of forestry Accomplishments in Fire Control and Fire still the basic, over-all goal that can function with safety. Research (Fire Control Notes 6[2] [April 1942]: 47–63). Graves stated in 1910: “The first
Fire Management Today 92 GUIDELINES FOR CONTRIBUTORS Editorial Policy Paper Copy. Type or word-process the manu Clearly label all photos and illustrations (figure Fire Management Today (FMT) is an interna script on white paper (double-spaced) on one 1, 2, 3, etc.; photograph A, B, C, etc.). At the end tional quarterly magazine for the wildland fire side. Include the complete name(s), title(s), affil of the manuscript, include clear, thorough fig community. FMT welcomes unsolicited manu iation(s), and address(es) of the author(s), as ure and photo captions labeled in the same way scripts from readers on any subject related to well as telephone and fax numbers and e-mail as the corresponding material (figure 1, 2, 3; fire management. Because space is a considera information. If the same or a similar manuscript photograph A, B, C; etc.). Captions should make tion, long manuscripts might be abridged by the is being submitted elsewhere, include that infor photos and illustrations understandable without editor, subject to approval by the author; FMT mation also. Authors who are affiliated should reading the text. For photos, indicate the name does print short pieces of interest to readers. submit a camera-ready logo for their agency, and affiliation of the photographer and the year institution, or organization. the photo was taken. Submission Guidelines Submit manuscripts to either the general man Style. Authors are responsible for using wild- Electronic Files. See special mailing instruc ager or the managing editor at: land fire terminology that conforms to the latest tions above. Please label all disks carefully with standards set by the National Wildfire name(s) of file(s) and system(s) used. If the USDA Forest Service Coordinating Group under the National manuscript is word-processed, please submit a Attn: April J. Baily, F&AM Staff Interagency Incident Management System. FMT 3-1/2 inch, IBM-compatible disk together with Mail Stop 1107 uses the spelling, capitalization, hyphenation, the paper copy (see above) as an electronic file 1400 Independence Avenue, SW and other styles recommended in the United in one of these formats: WordPerfect 5.1 for Washington, DC 20250-1107 States Government Printing Office Style DOS; WordPerfect 7.0 or earlier for Windows 95; tel. 202-205-0891, fax 202-205-1272 Manual, as required by the U.S. Department of Microsoft Word 6.0 or earlier for Windows 95; e-mail: [email protected] Agriculture. Authors should use the U.S. system Rich Text format; or ASCII. Digital photos may of weight and measure, with equivalent values in be submitted but must be at least 300 dpi and USDA Forest Service the metric system. Try to keep titles concise and accompanied by a high-resolution (preferably Attn: Hutch Brown, Office of Communication descriptive; subheadings and bulleted material laser) printout for editorial review and quality Mail Stop 1111 are useful and help readability. As a general rule control during the printing process. Do not 1400 Independence Avenue, SW of clear writing, use the active voice (e.g., write, embed illustrations (such as maps, charts, and Washington, DC 20250-1111 “Fire managers know…” and not, “It is graphs) in the electronic file for the manuscript. tel. 202-205-1028, fax 202-205-0885 known…”). Provide spellouts for all abbrevia Instead, submit each illustration at 1,200 dpi in e-mail: [email protected] tions. Consult recent issues (on the World Wide a separate file using a standard interchange for Web at mat such as EPS, TIFF, or JPEG, accompanied Mailing Disks. Do not mail disks with elec
Contributors Wanted We need your fire-related articles and photographs for Fire Management Today! Feature articles should be up to about 2,000 words in length. We also need short items of up to 200 words. Subjects of articles published in Fire Management Today include: Aviation Firefighting experiences Communication Incident management Cooperation Information management (including systems) Ecosystem management Personnel Equipment/Technology Planning (including budgeting) Fire behavior Preparedness Fire ecology Prevention/Education Fire effects Safety Fire history Suppression Fire science Training Fire use (including prescribed fire) Weather Fuels management Wildland–urban interface To help prepare your submission, see “Guidelines for Contributors” in this issue.
Volume 64 • No. 1 • Winter 2004 93 AUTHOR INDEX—VOLUME 63
Alexander, Martin E. Technology transfer Cook, Jim; Tom, Angela. Interview with Hughes, Joseph. New Jersey, April 1963: and wildland fire management/research. Paul Gleason. 63(3): 91–94. Can it happen again? 63(4): 40–44. 63(2): 41. Cornwall, Mike. Dome Peak Fire: Keeley, Jon E. Fire and invasive plants in Alexander, Martin E.; Stam, Joseph C. Witnessing the extreme. 63(1): 16–18. California ecosystems. 63(2): 18–19. Safety alert for wildland firefighters: Fuel Davis, Rollo T.; Ogden, Richard M. Black Keeley, Jon E.; Fotheringham, C.J. conditions in spruce-beetle-killed forests Wednesday in Arkansas and Oklahoma— Historical fire regime in southern of Alaska. 63(2): 25. 1971. 63(4): 15–16. California. 63(1): 8–9. Alexander, M.E.; Thomas, D.A. Wildland fire Dell, John D. The fire behavior team in Keller, Paul. “The Bison and the Wildfire.” behavior case studies and analyses: Other action: The Coyote Fire, 1964. 63(3): 63(2): 54. examples, methods, reporting standards, 81–84. Keller, Paul. “Gleason Complex” puts up and some practical advice. 63(4): 4–12. DeJong, Lisa. Improving fire hazard assess huge “plume”: A tribute to Paul Gleason. Alexander, M.E.; Thomas, D.A. Wildland fire ment in South Lake Tahoe, CA. 63(2): 63(3): 85–90. behavior case studies and analyses: Value, 35–40. Mack, Cheryl A. A burning issue: American approaches, and practical uses. 63(3): Dieterich, John H. Jet stream influence on Indian fire use on the Mt. Rainier Forest 4–8. the Willow Fire. 63(4): 17–19. Reserve. 63(2): 20–24. Argow, Keith A. The Carolina Blowup. Dillon, Madelyn. Fire Management Today Morris, William G. Rate of spread on a 63(4): 13–14. announces winners of 2003 photo con Washington Fern Fire. 63(3): 56–58. Arno, Stephen F.; Allison-Bunnell, Steven. test. 63(4): 85–89. Olson, C.F. An analysis of the Honey Fire. Managing fire-prone forests: Roots of our Division of Fire Control. Blackwater Fire on 63(3): 29–41. dilemma. 63(2): 12–16. the Shoshone. 3(3): 9–10. Patterson, Sara. A cooperative fire preven Baily, April J. Franklin Awards for achieve Editor. National Fire Plan at work. 63(1): tion adventure. 63(1): 14–15. ments in cooperative fire protection. 23. Powell, Gordon. A fire-whirlwind in 63(2): 45–49. Editor. Websites on fire. 63(1): 9. Alabama. 63(3): 71–73. Banks, Wayne G.; Little, Silas. The forest Editor. Websites on fire. 63(2): 11. Pyne, Stephen J. Firestop II. 63(2): 17. fires of April 1963 in New Jersey point Editor. Websites on fire. 63(3): 80. Rothermel, Richard C.; Brown, Hutch. A the way to better protection and manage Editor. Websites on fire. 63(4): 84. race that couldn’t be won. 63(4): 75–76. ment. 63(3): 74–78. Gebert, Krista M.; Schuster, Ervin G.; Rothermel, Richard C.; Mutch, Robert W. Bates, Robert W. A key to blowup condi Hesseln, Hayley. How would a 24-hour Behavior of the life-threatening Butte tions in the Southwest? 63(3): 68–70. pay system affect suppression cost? 63(2): Fire: August 27–29, 1985. 63(4): 31–39. Beighley, Mark; Bishop, Jim. Fire behavior 31–34. Simard, Albert J. The Mack Lake Fire. in high-elevation timber. 63(4): 56–62. General Land Office. “It is not understood 63(4): 29–30. Bosworth, Dale. Fires and forest health: why forest fires should get away.” 63(1): Small, R.T. Relationship of weather factors Our future is at stake. 63(2): 4–11. 37. to the rate of spread of the Robie Creek Brotak, Edward A. The Bass River Fire: Graham, B.J. The Harrogate Fire—March Fire. 63(3): 63–67. Weather conditions associated with a 15, 1964. 63(3): 79–80. Summerfelt, Paul. The wildland/urban fatal fire. 63(4): 25–28. Graham, Howard E. A firewhirl of tornadic interface: What’s really at risk? 63(1): Brotak, Edward A. Low-level weather condi violence. 63(3): 54–55. 4–7. tions preceding major wildfires. 63(4): Graham, Howard E. Fire-whirlwind forma Thomas, Leon R. The Bower Cave Fire. 67–71. tion favored by topography and upper 63(3): 42–45. Brotak, Edward A.; Reifsnyder, William E. winds. 63(3): 59–62. Thorburn, W.R.; MacMillan, A.; Alexander, Predicting major wildland fire occur Greelee, Jason; Greenlee, Dawn. Trigger M.E.; Nimchuk, N.; Frederick, K.W.; Van rence. 63(4): 20–24. points and the rules of disengagement. Nest, T.A. “Principles of Fire Behavior”: A Brown, A.A. The factors and circumstances 63(1): 10–13. CD-ROM-based interactive multimedia that led to the Blackwater Fire tragedy. Haines, Donald A. Horizontal vortices and training course. 63(2): 43–44. 63(3): 11–14. the New Miner Fire. 63(4): 45–47. Thorpe, Dan. Injuries and fatalities during Brown, A.A. Lessons of the McVey Fire, Haines, Donald A.; Lyon, L. Jack. nighttime firefighting operations. 63(2): Black Hills National Forest. 63(3): 25–28. Horizontal roll vortices in complex ter 26–30. Brown, Hutch. Book review: Ghosts of the rain. 63(4): 54–55. Thorpe, Dan. Those really bad fire days: Fireground. 63(2): 52–53. Headley, Roy. Lessons from larger fires on What makes them so dangerous? 63(4): Brown, Hutch. Photo contest 2001. 63(1): national forests, 1938. 63(3): 15–22. 72–74. 24–32. Headley, Roy. Lessons from larger fires on Werth, Paul; Ochoa, Richard. The Haines Butler, Bret W.; Bartlette, Roberta A.; national forests, 1939. 63(3): 23–24. Index and Idaho wildfire growth. 63(4): Bradshaw, Larry S.; Cohen, Jack D.; Hester, Dwight A. The pinyon–juniper fuel 63–66. Andrews, Patricia L.; Putnam, Ted; type can really burn. 63(3): 52–53. Vittoria, Stephen. “Keeper of the Flame”: A Mangan, Richard J.; Brown, Hutch. The Hirsch, Kelvin G. Documenting wildfire journey to the heart of fire. 63(2): 50–51. South Canyon Fire revisited: Lessons in behavior: The 1988 Brereton Lake Fire, Williams, Gerald W. Big Ed Pulaski and the fire behavior. 63(4): 77–84. Manitoba. 63(4): 50–53. Big Blowup. 63(1): 19–21. Byram, George M.; Nelson, Ralph M. The Hirsch, Kelvin G. An Overview of the 1987 Williams, Gerald W. Inventing the pulaski. possible relation of air turbulence to Wallace Lake Fire, Manitoba. 63(4): 63(1): 22–23. ■ erratic fire behavior in the Southeast. 48–49. 63(3): 46–51.
Fire Management Today 94 SUBJECT INDEX—VOLUME 63
American Indians The Carolina Blowup. Keith A. Argow. Predicting major wildland fire occurrence. A burning issue: American Indian fire use 63(4): 13–14. Edward A. Brotak; William E. Reifsnyder. on the Mt. Rainier Forest Reserve. Cheryl Documenting wildfire behavior: The 1988 63(4): 20–24. A. Mack. 63(2): 20–24. Brereton Lake Fire, Manitoba. Kelvin G. “Principles of Fire Behavior”: A CD-ROM Hirsch. 63(4): 50–53. based interactive multimedia training Art Dome Peak Fire: Witnessing the extreme. course. W.R. Thorburn; A. MacMillan; “The Bison and the Wildfire.” Paul Keller. Mike Cornwall. 63(1): 16–18. M.E. Alexander; N. Nimchuk; K.W. 63(2): 54. The factors and circumstances that led to Frederick; T.A. Van Nest. 63(2): 43–44. Fire Management Today announces win the Blackwater Fire tragedy. A.A. Brown. A race that couldn’t be won. Richard C. ners of 2003 photo contest. Madelyn 63(3): 11–14. Rothermel; Hutch Brown. 63(4): 75–76. Dillon.63(4): 85–89. Fire behavior in high-elevation timber. Rate of spread on a Washington Fern Fire. “Keeper of the Flame”: A journey to the Mark Beighley; Jim Bishop. 63(4): 56–62. William G. Morris. 63(3): 56–58. heart of fire. Stephen Vittoria. 63(2): The fire behavior team in action: The Relationship of weather factors to the rate 50–51. Coyote Fire, 1964. Dell, John D. 63(3): of spread of the Robie Creek Fire. R.T. Photo contest 2001. Hutch Brown. 63(1): 81–84. Small. 63(3): 63–67. 24–32. A firewhirl of tornadic violence. Howard E. The South Canyon Fire revisited: Lessons Graham. 63(3): 54–55. in fire behavior. Bret W. Butler; Roberta Burned Area Rehabilitation Fire-whirlwind formation favored by topog A. Bartlette; Larry S. Bradshaw; Jack D. National Fire Plan at work. Editor. 63(1): raphy and upper winds. Howard E. Cohen; Patricia L. Andrews; Ted Putnam; 23. Graham. 63(3): 59–62. Richard J. Mangan; Hutch Brown. 63(4): A fire-whirlwind in Alabama. Gordon 77–84. Cooperation Powell. 63(3): 71–73. Those really bad fire days: What makes The Bower Cave Fire. Leon R. Thomas. The forest fires of April 1963 in New Jersey them so dangerous? Dan Thorpe. 63(4): 63(3): 42–45. point the way to better protection and 72–74. A cooperative fire prevention adventure. management. Wayne G. Banks; Silas Wildland fire behavior case studies and Sara Patterson. 63(1): 14–15. Little. 63(3): 74–78. analyses: Value, approaches, and practical Franklin Awards for achievements in coop The Haines Index and Idaho wildfire uses. M.E. Alexander; D.A. Thomas. erative fire protection. April J. Baily. growth. Paul Werth; Richard Ochoa. 63(3): 4–8. 63(2): 45–49. 63(4): 63–66. Wildland fire behavior case studies and The Harrogate Fire—March 15, 1964. analyses: Other examples, methods, Costs Graham, B.J. 63(3): 79–80. reporting standards, and some practical How would a 24-hour pay system affect Horizontal roll vortices in complex terrain. advice. M.E. Alexander; D.A. Thomas. suppression cost? Krista M. Gebert; Ervin Donald A. Haines; Jack L. Lyon. 63(4): 63(4): 4–12. G. Schuster; Hayley Hesseln. 63(2): 54–55. 31–34. Horizontal vortices and the New Miner Fire Ecology Fire. Donald A. Haines. 63(4): 45–47. Fire and invasive plants in California Crews Jet stream influence on the Willow Fire. ecosystems. Jon E. Keeley. 63(2): 18–19. “Gleason Complex” puts up huge “plume”: John H. Dieterich. 63(4): 17–19. Historical fire regime in southern A tribute to Paul Gleason. Paul Keller. A key to blowup conditions in the California. Jon E. Keeley; C.J. 63(3): 85–90. Southwest? Robert W. Bates. 63(3): Fotheringham. 63(1): 8–9. Interview with Paul Gleason. Jim Cook; 68–70. “Keeper of the Flame”: A journey to the Angela Tom. 63(3): 91–94. Lessons from larger fires on national heart of fire. Stephen Vittoria. 63(2): forests, 1938. Roy Headley. 63(3): 15–22. 50–51. Equipment and Engineering Lessons from larger fires on national Managing fire-prone forests: Roots of our Inventing the pulaski. Gerald W. Williams. forests, 1939. Roy Headley. 63(3): 23–24. dilemma. Stephen F. Arno; Steven 63(1): 22–23. Lessons of the McVey Fire, Black Hills Allison-Bunnell. 63(2): 12–16. National Forest. A.A. Brown. 63(3): Fire Behavior 25–28. Fire Effects An analysis of the Honey Fire. C.F. Olson. Low-level weather conditions preceding Fire and invasive plants in California 63(3): 29–41. major wildfires. Edward A. Brotak. 63(4): ecosystems. Jon E. Keeley. 63(2): 18–19. The Bass River Fire: Weather conditions 67–71. Historical fire regime in southern associated with a fatal fire. E.A. Brotak. The Mack Lake Fire. Albert J. Simard. California. Jon E. Keeley; C.J. 63(4): 25–28. 63(4): 29–30. Fotheringham. 63(1): 8–9. Behavior of the life-threatening Butte Fire: New Jersey, April 1963: Can it happen Managing fire-prone forests: Roots of our August 27–29, 1985. Richard C. again? Joseph Hughes. 63(4): 40–44. dilemma. Stephen F. Arno; Steven Rothermel; Robert W. Mutch. 63(4): An Overview of the 1987 Wallace Lake Fire, Allison-Bunnell. 63(2): 12–16. 31–39. Manitoba. Kelvin G. Hirsch. 63(4): The pinyon–juniper fuel type can really Black Wednesday in Arkansas and 48–49. burn. Dwight A. Hester. 63(3): 52–53. Oklahoma—1971. Rollo T. Davis; Richard The pinyon–juniper fuel type can really M. Ogden. 63(4): 15–16. burn. Dwight A. Hester. 63(3): 52–53. Fire History Blackwater Fire on the Shoshone. Division The possible relation of air turbulence to Behavior of the life-threatening Butte Fire: of Fire Control. 3(3): 9–10. erratic fire behavior in the Southeast. August 27–29, 1985. Richard C. The Bower Cave Fire. Leon R. Thomas. George M. Byram; Ralph M. Nelson. Rothermel; Robert W. Mutch. 63(4): 63(3): 42–45. 63(3): 46–51. 31–39.
Volume 64 • No. 1 • Winter 2004 95 Big Ed Pulaski and the Big Blowup. Gerald Fuels/Fuels Management Managing fire-prone forests: Roots of our W. Williams. 63(1): 19–21. Fire behavior in high-elevation timber. dilemma. Stephen F. Arno; Steven Black Wednesday in Arkansas and Mark Beighley; Jim Bishop. 63(4): 56–62. Allison-Bunnell. 63(2): 12–16. Oklahoma—1971. Rollo T. Davis; Richard Fire and invasive plants in California M. Ogden. 63(4): 15–16. ecosystems. Jon E. Keeley. 63(2): 18–19. Preparedness Blackwater Fire on the Shoshone. Division New Jersey, April 1963: Can it happen The forest fires of April 1963 in New Jersey of Fire Control. 3(3): 9–10. again? Joseph Hughes. 63(4): 40–44. point the way to better protection and The Bower Cave Fire. Leon R. Thomas. The pinyon–juniper fuel type can really management. Wayne G. Banks; Silas 63(3): 42–45. burn. Dwight A. Hester. 63(3): 52–53. Little. 63(3): 74–78. A burning issue: American Indian fire use Rate of spread on a Washington Fern Fire. Improving fire hazard assessment in South on the Mt. Rainier Forest Reserve. Cheryl William G. Morris. 63(3): 56–58. Lake Tahoe, CA. Lisa DeJong. 63(2): A. Mack. 63(2): 20–24. Safety alert for wildland firefighters: Fuel 35–40. The Carolina Blowup. Keith A. Argow. conditions in spruce-beetle-killed forests Predicting major wildland fire occurrence. 63(4): 13–14. of Alaska. Martin E. Alexander; Joseph C. Edward A. Brotak; William E. Reifsnyder. Dome Peak Fire: Witnessing the extreme. Stam. 63(2): 25. 63(4): 20–24. Mike Cornwall. 63(1): 16–18. Websites on fire. Editor. 63(3): 80 Those really bad fire days: What makes The factors and circumstances that led to them so dangerous? Dan Thorpe. 63(4): the Blackwater Fire tragedy. A.A. Brown. Hazard/Risk Assessment 72–74. 63(3): 11–14. Improving fire hazard assessment in South The forest fires of April 1963 in New Jersey Lake Tahoe, CA. Lisa DeJong. 63(2): Research/Technology point the way to better protection and 35–40. The fire behavior team in action: The management. Wayne G. Banks; Silas Those really bad fire days: What makes Coyote Fire, 1964. Dell, John D. 63(3): Little. 63(3): 74–78. them so dangerous? Dan Thorpe. 63(4): 81–84. The Harrogate Fire—March 15, 1964. 72–74. Improving fire hazard assessment in South Graham, B.J. 63(3): 79–80. The wildland/urban interface: What’s really Lake Tahoe, CA. Lisa DeJong. 63(2): Interview with Paul Gleason. Jim Cook; at risk? Paul Summerfelt. 63(1): 4–7. 35–40. Angela Tom. 63(3): 91–94. Technology transfer and wildland fire man Lessons from larger fires on national Incident Management agement/research. Martin E. Alexander. forests, 1938. Roy Headley. 63(3): 15–22. Behavior of the life-threatening Butte Fire: 63(2): 41. Lessons from larger fires on national August 27–29, 1985. Richard C. Wildland fire behavior case studies and forests, 1939. Roy Headley. 63(3): 23–24. Rothermel; Robert W. Mutch. 63(4): analyses: Value, approaches, and practical Lessons of the McVey Fire, Black Hills 31–39. uses. M.E. Alexander; D.A. Thomas. National Forest. A.A. Brown. 63(3): “Gleason Complex” puts up huge “plume”: 63(3): 4–8. 25–28. A tribute to Paul Gleason. Paul Keller. Wildland fire behavior case studies and The Mack Lake Fire. Albert J. Simard. 63(3): 85–90. analyses: Other examples, methods, 63(4): 29–30. Interview with Paul Gleason. Jim Cook; reporting standards, and some practical Managing fire-prone forests: Roots of our Angela Tom. 63(3): 91–94. advice. M.E. Alexander; D.A. Thomas. dilemma. Stephen F. Arno; Steven Injuries and fatalities during nighttime fire 63(4): 4–12. Allison-Bunnell. 63(2): 12–16. fighting operations. Dan Thorpe. 63(2): A race that couldn’t be won. Richard C. 26–30. Safety Rothermel; Hutch Brown. 63(4): 75–76. A race that couldn’t be won. Richard C. Book review: Ghosts of the Fireground. The South Canyon Fire revisited: Lessons Rothermel; Hutch Brown. 63(4): 75–76. Hutch Brown. 63(2): 52–53. in fire behavior. Bret W. Butler; Roberta The South Canyon Fire revisited: Lessons “Gleason Complex” puts up huge “plume”: A. Bartlette; Larry S. Bradshaw; Jack D. in fire behavior. Bret W. Butler; Roberta A tribute to Paul Gleason. Paul Keller. Cohen; Patricia L. Andrews; Ted Putnam; A. Bartlette; Larry S. Bradshaw; Jack D. 63(3): 85–90. Richard J. Mangan; Hutch Brown. 63(4): Cohen; Patricia L. Andrews; Ted Putnam; Injuries and fatalities during nighttime fire 77–84. Richard J. Mangan; Hutch Brown. 63(4): fighting operations. Dan Thorpe. 63(2): Wildland fire behavior case studies and 77–84. 26–30. analyses: Value, approaches, and practical Trigger points and the rules of disengage Interview with Paul Gleason. Jim Cook; uses. M.E. Alexander; D.A. Thomas. ment. Jason Greelee; Dawn Greenlee. Angela Tom. 63(3): 91–94. 63(3): 4–8. 63(1): 10–13. Safety alert for wildland firefighters: Fuel Wildland fire behavior case studies and conditions in spruce-beetle-killed forests analyses: Other examples, methods, Personnel of Alaska. Martin E. Alexander; Joseph C. reporting standards, and some practical How would a 24-hour pay system affect Stam. 63(2): 25. advice. M.E. Alexander; D.A. Thomas. suppression cost? Krista M. Gebert; Ervin Trigger points and the rules of disengage 63(4): 4–12. G. Schuster; Hayley Hesseln. 63(2): ment. Jason Greelee; Dawn Greenlee. 31–34. 63(1): 10–13. Fire Prevention Websites on fire. Editor. 63(1): 9. A cooperative fire prevention adventure. Policy Sara Patterson. 63(1): 14–15. Fires and forest health: Our future is at Suppression stake. Dale Bosworth. 63(2): 4–11. An analysis of the Honey Fire. C.F. Olson. Fire Use Firestop II. Stephen J. Pyne. 63(2): 17. 63(3): 29–41. A burning issue: American Indian fire use Interview with Paul Gleason. Jim Cook; Big Ed Pulaski and the Big Blowup. Gerald on the Mt. Rainier Forest Reserve. Cheryl Angela Tom. 63(3): 91–94. W. Williams. 63(1): 19–21. A. Mack. 63(2): 20–24. “It is not understood why forest fires Black Wednesday in Arkansas and The forest fires of April 1963 in New Jersey should get away.” General Land Office. Oklahoma—1971. Rollo T. Davis; Richard point the way to better protection and 63(1): 37. M. Ogden. 63(4): 15–16. management. Wayne G. Banks; Silas “Keeper of the Flame”: A journey to the Blackwater Fire on the Shoshone. Division Little. 63(3): 74–78. heart of fire. Stephen Vittoria. 63(2): of Fire Control. 3(3): 9–10. 50–51. Fire Management Today 96 Book review: Ghosts of the Fireground. The South Canyon Fire revisited: Lessons Horizontal roll vortices in complex terrain. Hutch Brown. 63(2): 52–53. in fire behavior. Bret W. Butler; Roberta Donald A. Haines; Jack L. Lyon. 63(4): The Bower Cave Fire. Leon R. Thomas. A. Bartlette; Larry S. Bradshaw; Jack D. 54–55. 63(3): 42–45. Cohen; Patricia L. Andrews; Ted Putnam; Horizontal vortices and the New Miner The Carolina Blowup. Keith A. Argow. Richard J. Mangan; Hutch Brown. 63(4): Fire. Donald A. Haines. 63(4): 45–47. 63(4): 13–14. 77–84. Jet stream influence on the Willow Fire. Documenting wildfire behavior: The 1988 John H. Dieterich. 63(4): 17–19. Brereton Lake Fire, Manitoba. Kelvin G. Training A key to blowup conditions in the Hirsch. 63(4): 50–53. “Gleason Complex” puts up huge “plume”: Southwest? Robert W. Bates. 63(3): Dome Peak Fire: Witnessing the extreme. A tribute to Paul Gleason. Paul Keller. 68–70. Mike Cornwall. 63(1): 16–18. 63(3): 85–90. Low-level weather conditions preceding The forest fires of April 1963 in New Jersey Websites on fire. Editor. 63(4): 84. major wildfires. Edward A. Brotak. 63(4): point the way to better protection and Interview with Paul Gleason. Jim Cook; 67–71. management. Wayne G. Banks; Silas Angela Tom. 63(3): 91–94. The possible relation of air turbulence to Little. 63(3): 74–78. Predicting major wildland fire occurrence. erratic fire behavior in the Southeast. The factors and circumstances that led to Edward A. Brotak; William E. Reifsnyder. George M. Byram; Ralph M. Nelson. the Blackwater Fire tragedy. A.A. Brown. 63(4): 20–24. 63(3): 46–51. 63(3): 11–14. “Principles of Fire Behavior”: A CD-ROM Predicting major wildland fire occurrence. The Harrogate Fire—March 15, 1964. based interactive multimedia training Edward A. Brotak; William E. Reifsnyder. Graham, B.J. 63(3): 79–80. course. W.R. Thorburn; A. MacMillan; 63(4): 20–24. Lessons from larger fires on national M.E. Alexander; N. Nimchuk; K.W. Rate of spread on a Washington Fern Fire. forests, 1938. Roy Headley. 63(3): 15–22. Frederick; T.A. Van Nest. 63(2): 43–44. William G. Morris. 63(3): 56–58. Lessons from larger fires on national Relationship of weather factors to the rate forests, 1939. Roy Headley. 63(3): 23–24. Weather of spread of the Robie Creek Fire. R.T. Lessons of the McVey Fire, Black Hills The Bass River Fire: Weather conditions Small. 63(3): 63–67. National Forest. A.A. Brown. 63(3): associated with a fatal fire. E.A. Brotak. Those really bad fire days: What makes 25–28. 63(4): 25–28. them so dangerous? Dan Thorpe. 63(4): New Jersey, April 1963: Can it happen A firewhirl of tornadic violence. Howard E. 72–74. again? Joseph Hughes. 63(4): 40–44. Graham. 63(3): 54–55. An Overview of the 1987 Wallace Lake Fire, Fire-whirlwind formation favored by topog Wildland/Urban Interface Manitoba. Kelvin G. Hirsch. 63(4): raphy and upper winds. Howard E. Improving fire hazard assessment in South 48–49. Graham. 63(3): 59–62. Lake Tahoe, CA. Lisa DeJong. 63(2): The pinyon–juniper fuel type can really A fire-whirlwind in Alabama. Gordon 35–40. burn. Dwight A. Hester. 63(3): 52–53. Powell. 63(3): 71–73. The wildland/urban interface: What’s really The Mack Lake Fire. Albert J. Simard. The Haines Index and Idaho wildfire at risk? Paul Summerfelt. 63(1): 4–7. 63(4): 29–30. growth. Paul Werth; Richard Ochoa. Websites on fire. Editor. 63(2): 11. ■ 63(4): 63–66.
Volume 64 • No. 1 • Winter 2004 97 PHOTO CONTEST ANNOUNCEMENT Fire Management Today invites Rules job descriptions; and descriptions you to submit your best fire-related • The contest is open to everyone. of any vegetation and/or wildlife). photos to be judged in our annual You may submit an unlimited •You must complete and sign a competition. Judging begins after number of entries from any place statement granting rights to use the first Friday in March of each or time; but for each photo, you your photo(s) to the USDA Forest year. must indicate only one competi Service (see sample statement tion category. To ensure fair eval below). Include your full name, Awards uation, we reserve the right to agency or institutional affiliation All contestants will receive a change the competition category (if any), address, and telephone CD–ROM with all photos not elimi for your photo. number. nated from competition. Winning • An original color slide is pre • Photos are eliminated from com photos will appear in a future issue ferred; however, we will accept petition if they have date stamps; of Fire Management Today.In high-quality color prints, as long show unsafe firefighting practices addition, winners in each category as they are accompanied by nega (unless that is their express pur will receive: tives. Digitally shot slides (pre pose); or are of low technical ferred) or prints will be accepted quality (for example, have soft • 1st place—Camera equipment if they are scanned at 300 lines focus or show camera move worth $300 and a 16- by 20-inch per inch or equivalent. Digital ment). (Duplicates—including framed copy of your photo. images will be accepted if you most overlays and other compos • 2nd place—An 11- by 14-inch used a camera with at least 2.5 ites—have soft focus and will be framed copy of your photo. megapixels and the image is shot eliminated.) •3rd place—An 8- by 10-inch at the highest resolution or in a • Photos are judged by a photogra framed copy of your photo. TIFF format. phy professional whose decision •You must have the right to grant is final. the Forest Service unlimited use Categories Postmark Deadline •Wildland fire of the image, and you must agree •Prescribed fire that the image will become pub First Friday in March •Wildland/urban interface fire lic domain. Moreover, the image • Aerial resources must not have been previously Send submissions to: •Ground resources published. Madelyn Dillon •Miscellaneous (fire effects; fire • For every photo you submit, you CAT Publishing Arts weather; fire-dependent commu must give a detailed caption 2150 Centre Avenue nities or species; etc.) (including, for example, name, Building A, Suite 361 location, and date of the fire; Fort Collins, CO 80526 names of any people and/or their
Sample Photo Release Statement
Enclosed is/are (number) slide(s) for publication by the USDA Forest Service. For each slide submitted, the contest category is indicated and a detailed caption is enclosed. I have the right to grant the Forest Service unlimited use of the image, and I agree that the image will become public domain. Moreover, the image has not been previously published.
Signature Date
Fire Management Today 98 First of Its Kind: A Historical Perspective on Wildland Fire Behavior Training M.E. Alexander and D.A. Thomas
In 1957, the Chief of the USDA Forest Service appointed a task force to study ways of preventing firefighter fatalities in the future. A review of 16 fatality fires found that the associated fire behavior in all but one case was unexpect ed by those entrapped or over run. One of the task force’s major recommendations was an intensified program of fire behavior training.*
The recommendation led to the first National Fire Behavior Training School. Trainees assem bled at the Smokejumper Center Students and instructors at the first National Fire Behavior Training School, held in in Missoula, MT, for a course that spring 1958. Front row (left to right): A. Brackebusch (INT), E. DeSilvia (R-1), J. lasted from March 31 to May 1, Philbrick (R-6), E. Marshall (R-6), M. Lowden (WO), E. Williams (R-8), J. Coleman (R 1958. Bacon (1958) has written a 9), E. Bacon (WO), and W. Moore (R-1). Middle row (left to right): F. Brauer (R-1), K. Knutson (R-2), K. Wilson (R-2), J. Koen (R-8), J. Kilodragovich (R-1), C. Phillips (CDF), good account of the 5-week D. Pomerening (R-8), B. Emerson (R-9), H. Reinecker (CDF), and J. Dieterich (INT). course. Back row (left to right): L. Biddson (R-5), C. Fox (R-4), S. Moore (R-6), K. Scholz (R 2), J. Davis (RMF), K. Thompson (R-2), B. Rasmussen (R-4), J. Keetch (R-7), T. Schlapfer (R-5), L. Kelley (R-7), T. Koskella (R-4), W. Murray (R-4), K. Weiesenbam (R The 28 trainees came from all 3), F. Mass (R-1), J. Franks (BLM), C. Hardy (INT), W. Krumm (WB), and J. Barrows regions of the Forest Service, (INT). Abbreviations: BLM = U.S. Department of the Interior, Bureau of Land various forestry schools, the U.S. Management; CDF = California Division of Forestry; INT = USDA Forest Service, Intermountain Forest and Range Experiment Station; R-1 = Forest Service, Northern Department of the Interior, and Region; R-2 = Forest Service, Rocky Moutain Region; R-3 = Forest Service, the National Association of State Southwestern Region; R-4 = Forest Service, Intermountain Region; R-5 = Forest Foresters. The instructors came Service, Pacific Southwest Region; R-6 = Forest Service, Pacific Northwest Region; R-8 from the Forest Service, the U.S. = Forest Service, Southern Region; R-9 = Forest Service, Eastern Region; RMF = Forest Service, Rocky Mountain Forest and Range Experiment Station; WB = U.S. Weather Bureau, Yale University, Weather Bureau; and WO = Forest Service, Washington Office.
Marty Alexander is a senior fire behav and the Munitalp Foundation. help with naming the individuals ior research officer with the Canadian Trainees and some instructors are shown in the photo. Forest Service at the Northern Forestry shown in the group photo below Centre, Edmonton, Alberta; and Dave Thomas is regional fuels specialist for (from Bacon 1958). Reference the USDA Forest Service, Intermountain Bacon, E.M. 1958. Training in forest fire Region, Ogden, UT. Acknowledgment behavior. American Forests. 64(7): 24-25, 47-49. * The task force’s full report is on the World Wide The authors wish to thank Mike Web at