United States Department of Agriculture

Forest Service Fire Management

Volume 52, No.3 1991 Notes

.~ ~, I, '. \ United States Department of Fire Agriculture Management Forest service An international quarterly periodical Volume 52, NO.3 Notes devoted to forest fire management 1991

Contents 27 Federal Excess Personal Property Information: Where To Find It in Fire Management Notes 3 A Salute to Infrared Systems in Fire Detection and Francis R. Russ Mapping John R. Warren and Doris N. Celarier 30 Hallie Daggett: First Woman as Forest Service Fire Lookout 19 Selecting the "Right" Infrared System for a Firefighting JDb 31 Clark County Goes Face-To-Face With John R. Warren Wildland-Urban Interface Lane L. Jolly 21 Forest Fire Detection Systems Stanley N. Hirsch 38 Max Planck, Infrared, and Quantum Mechanics John R. Warren 2S Fire Mapping Using Airborne Global Positioning Philip L. Drake Special Features 28 Using the Global Positioning System in Firefighting on the Shorts Fire in the Okefenokee Swamp 16 Primer on Infrared Douglas Luepke John R ..Warren

35 Global Positioning System: Uses in Fire Management 32 The Pioneers (Some of Them) and Their Equipment on the Clearwater National Forest (a Little of It) in Forest Service Infrared Fire Mapping Byron J. Bonney and Detection Research and Operations

37 FIRE MOUSE TRAP Use in the Southern Region James P. Scott Fire Management Notesis published by the Forest service of the United States Departmentof Agriculture, WashIngton, DC. The Secretaryof Agricul1ure has determined 39 Evaluating the Hummers that the publication of this periodical is necessaryIn Ihe transaction of tM public business requiredby law of this Dlilpartment. Brian Hutchins Subscriptionsmay be obtainedfrom the Superintendentof Documents,U.S. GOvernment PrintingOffice, Washington, DC 20402.

41 Monthly Fire Weather Forecasts Send suggeltions and articles10Chief, Forest service (Ann: Fire ManagementNotes); Morris H. McCutchan, Bernard N. Meisner, P.O. Box 96090, U.S. Departmentof Agriculture,Washington,DC 20090-6090. Francis M. FUjioka, John W. Benoit, and Edward R. Madigan,Secretary Francis A. Russ U.S. Departmenlof Agriculture Benjamin Ly 13enerlll Manager F, DafeRobertson,Chief Coris N. ceaner FOfEISt Service Editor

l.A. Amicarella,Director Short Features F=lre and Aviation Milnagemenl U.S. Department of Agriculture programs, services, and Eimploymeni ar6 Op4ln toallon tlie basis of ment, withouf regard10 race, color, sex, religion, nationalorigin, age, or disability. 24 Minimizing the Risk of Wildfire: A Symposium on Any perSon who believeshe or she has been discriminatedagainst in any U.S. Departmentof Agriculture-telat~ activity should immediatelycontact the Secretary01 Wildfire Problems in the Wildland-Urban Interface Agrlculture.WaShington, DC 20250.

The usa 01trade, firm, or corporationnames in Ihis pubJJealion is lot Ihe jnlormalionlind convenience01 the reader.Such use dOlJS not constitute an official endorsementof any product or service by the U.S.Oepsrtmantof Atlriculture. Individual authorsare FtOnt CDWi': Foresl Servic& infraredtechnologydemonstrationallhe Discollery responsiblefot the technical accuracyof the material presentedIn Fire Mimar;8mertt FiIr, pan Of SouWl,CatoIina State Fair, Columbia, SC, October 1989. NoteS.

2 Fire Management Notes A Salute to Infrared Systems in - - Fire Detection and Mapping

John R. Warren and Doris N. Celarier USDA Forest Service, group leader, Advanced Electronics, Boise Interagency Fire Center, Boise.Il), and writer-editor, Public Affairs Office, Washington. DC

The USDA Forest Service has been Forest Service Infrared synchronous and signal conditioning using infrared (IR) systems since the Anniversaries--1991, the 25th in Fire circuits on a strip of film. The film is mid-196O's for fire detection and Operations and, 1992, the 30th in exposed, processed, and developed on mapping. In 1962, research in the Research board the aircraft so it is ready to use Forest Service began on IR systems, when delivered. and in 1966, the firstlR airborne line scanning system was moved from is now common, commercially What is IR? research to operations. The year 1991 is available technology was classified the "silver anniversary" ofForest information then. The late Bob It happens that all objects above Service operational use of IR detection Bjomsen served as leader ofa separate absolute zero (-273.16 "C) in tempera­ and mapping systems, and, 1992, the team to develop fire mapping opera­ ture radiate energy. At absolute zero, all 30th anniversary of Forest Service IR tional procedures (Bjomsen 1989). molecular activity ceases, and there is research. These are milestones. It is Later, Bjomsen, as the Forest Service no radiation. Since we do not encounter time to look at where the Forest Service Boise Interagency Fire Center (BIFC) objects at absolute zero, all objects are has been and where it is going in its use director, was instrumental in establish­ radiating energy called electromagnetic of infrared fire detection and mapping ing IR mapping and detection as a energy. However, we cannot see most and a time to-take stock of whatlR standard operational procedure. The of this energy, except energy radiated at advancements have been made and how operational reliabilily of the IR systems a narrow range (visible spectrum), or if those advancements are to be used. has been extremely high, due largely to the temperature becomes hot enough, the competence and diligence of the we can then feel the radiated energy, The Research Beginnings operationallR technicians. Now retired, such as from a hot stove. (See "Electro­ Bill Barrus, supervisory IR technician magnetic Radiation" and "Infrared For 14 yeai-s--1962to 1976­ for many years, was influential in this Radiation" in "Primer on Infrared" and infrared research was successfully success. in the accompanying boxes, 'The Ten conducted by Project F1RESCAN at the The Forest Service airborne IR line Most-Asked Questions on Infrared" and Intermountain Forest and Range scanningsystems have always used "Answers" and 'The Five Most Experiment Station with Stanley N. basic line scanner "front-end," rotating Misunderstood IR Notions" and Hirsch, electronicsengineer, as project mirror optics and dewar-detector "Discussion.") leader (Warren and Wilson 1981). assemblies from military systems, with Energy can also be generated and Working with him were Ralph Wilson, some unique, critical Forest Service radiated by other means. We are physicist; Forrest Madden, electronics modifications. The Forest Service IR constantly subjected to a large range of engineer; and Dale Gable, electronics systems are the only ones in the world people-made radiation fields. Consider technician. known to be designed and developed the electrical power fields. These fields Remarkably, during this period, the specifically for fire detection and are present all over the populated areas basic theory of fire mapping and mapping. (See "Forest Service Special of the world. In the United States, these detection with thennallR systems was IR Requirements.") Some of the are 60-hertz fields. (One hertz is developed. There are now new tech­ methods used seem fairly obvious now, equivalent to one cycle per second, nologies and methods available for use, but to develop them back then took a meaning the energy field is "changing but the basic tenets still apply. good deal of inspiration, and no doubt direction" 60 times per second.) The early research and development perspiration, from Hirsch, Wilson, and Radio Waves. AM (amplitude was carried out in cooperation with the others. modulation) radio waves range from Department of Defense (DOD) Office With the exception ofWilson's 550 to 1600 kilohertz (radio waves of Civil Defense and the Defense innovative, dual polaroid image system, change direction 550,000 to 1,600,000 Advanced Research Projects Agency. the Forest Service IR line scanners have times per second). People do not see Many of the systems and much of what produced the IR image via unique AM radio waves, or signals, nor do they

1991 Volume 52, Number 3 3 feel them or have any other natural way frequencies. (A megahertz-moving Microwaves and Infrared. Higher of knowing they are present. But a right along-is I million hertz or cycles in frequency, or shorter in wavelength, switched-on AM radio (electroacoustic per second.) People also cannot see, than the radio wave is the microwave. transducer) detects radio waves, hear, smell, or feel energy radiated by (Energy in the microwave band is processes them as electrical signals. and the television broadcast systems, but used in microwave ovens, satellite transduces those signals by a speaker so our television antennas can pick up television, and radar (radio detection people can hear them as sound. those signals, which are at a very low and ranging).) Just beyond the PM (frequency modulation) radio energy level, and pass them on to the microwave band is the millimeter operates from 88 to 108 megahertz, and circuits that amplify and process them. waveband and then the infrared. When television, on both sides of those Voila! Stereo sound and action in living a frequency is as high as 30 million color. megahertz, as it is for the thermallR

: i Forest Service Special Infrared Requirements : The Five Most Misunderstood I ! IR Notions I The Forest Service has special infrared following exposure to extended high requirements not available on military or temperatures of 600 °C or more. I 1. Thermal IR imagery is like black- I commercial systems: (Without DC coupling, the imagery is and-white photography. "washed out" around very hot areas, 2. IR imagery should be as good in • Dual-band detectors--Gives the precluding accurate location of the the daytime as it is at night. Forest Service a unique method with firefront position with respect to 3. IR imagery isjust like film­ about a tu-to-r advantage in the firebreaks, roads, and other identifi­ everything is right where you detection of small hot spots. The able features.) see it. target detection method and circuitry, • Target detection circuitry­ 4. One brand of IR "sees" through using a special combination of two Through a special circuitry mentioned moisture or clouds better than thermal bands, are not known to exist under "dual-band detectors," gives another brand. in any other systems. the capability of detecting very small S. (Converse of IR notion No.4.) All • Rectilinearization of the imagery­ hot spots (capability of detecting a hot IR systems are just alike; it Aids in the interpretation and spot of 600 'C as small as 0.0225 doesn't matter which type is transposing of fire features to maps. square meter (0.24 ft2) against ordered. (The interpretation task is still instantaneous terrain background performed manually and requires variations between 0 to 50°C from ! Discussion specially trained interpreters.) The 5,000 meters (16,404 ft) altitude. scale of imagery is different in both • Rapid processing of imagery on I. Thermal iR images resemble black­ the X and Y dimensions from the board aircraft-Produces good and-white photos or black-and­ scale of the maps, and the X scale of quality images to ensure the tire is white television, but the images imagery is constantly changing in a adequately covered and to minimize represent radiated energy levels in nonlinear manner. aircraft and crew operational time. the thermal IR part of the electro­ • Roll compensation and mileage (Fire staff needs to have imagery magnetic spectrum, not reflected markers-Assists in making images interpreted within I hour of flight energy levels in the visible part of more useful and easily interpreted. over active fire.) the spectrum. Some thermal lR e DC (direct current) response­ •A large total field of view-Covers images look somewhat like black­ coupled systems (rather than AC large areas in a short time for and-white pictures, but others look (alternating current) coupled detection missions and large fires in a completely different. systems-Prints terrain features of single frame of imagery, rather than a 2. At night, all the energy received in low thermal differences quickly, mosaic, for mapping missions. the thermal IR bands is radiated

4 Fire Management Notes band, it is customary to refer to this energy in terms ofequivalent wave­ lengths, such as 10 micrometers The thermal infrared bands are defined as 3 to 5 micrometers (microns) and 8 to 14 The Ten Most-Asked energy striking it; that is, it is a micrometers. (Wavelength has an Questions on Infrared perfect absorber, and also emitter of inverse relationship with frequency: energy. The higher the frequency, the shorter 5. Emissivity is the ratio ofthe radiant 1. What is infrared radiation? the wavelength.) energy emitted by an object at 2. Does an object have to be "hot" temperature T to the radiant energy Visible Spectrum. Eventually, as the to emit infrared radiation? emitted by a blackbody at tempera- frequency increases, or wavelength 3. Does the amount of infrared ture T. radiation vary with temperature? 6. All radiant energy striking an object 4. What is a blackbody? must be either absorbed, reflected, S. What is emissivity? or transmitted (on through the 6. What is the relationship between energy, giving a true indication of object). The sum of energy emissivity, reflectivity, transmis· temperature levels (when emissivity absorbed, or A, reflected, or R, and sivity, and absorptivity? is accounted for). In the day, in transmitted, or T, must equal 100 7. Are the FUR's and IR line addition to the radiated energy, percent of the total incident energy: scanners basically IR cameras? reflected solar energy and energy A+R+T=l. 8. Can we determine temperatures radiated from solar-heated rocks or 7. No. Thermal IR sensors are not with IR? bare spots may be confused with cameras. They do not directly 9. Does white indicDte hot or coo] fire spots. expose film to electromagnetic areas? 3. IR imagery from the line scanners energy and do not depend on 10. Can we "see" through clouds and always has terrain distortions. FLIR reflected energy for their usefulness. smoke? imagery also can contain similar They are fairly sophisticated distortions but usually to a much electro-optical instruments, smaller degree. When "looking" I displaying information from a part straight down or at a low angle, the ' Answers ofthe electromagnetic spectrum that distortion is much less. I we do not perceive without 4. IR systems, whether line scanner, 1. Infrared radiation is electromagnetic instruments. FLIR, or other type, are all subject radiation with wavelengths longer 8. Yes and no. Temperatures can be to the same laws of physics, than those of the visible part of the determined with IR under some including atmospheric attenuation spectrum and shorter than radio or conditions using special instruments of IR energy. For this condition, I microwave frequencies. usually called radiometers, but in various brands do not differ much, 2. No. All objects emit infrared the usual application of thermal IR as has been shown in side-by-side I I radiation, which is related to the for fire purposes temperature cannot demonstrations. temperature of the object. Only at bedetermined. We can, however, 5. Design concepts and designs of absolute zero (-273°C) will an distinguish relative temperature various types of IR systems differ I object cease to emit any radiation, differences. appreciably. Some are better suited including infrared radiation. 9. Either-with most forward-looking for certain applications and 3. Yes. Infrared radiation is emitted infrared (FUR) systems. Usually, conditions than others. The best over a wide band of wavelengths, the user selects the polarity of the choice can usually bemade by ! but the amount ofenergy emitted at video display so that either white or considering the situation and anyone wavelength increases as the black may indicate hot areas. conditions and choosing the type of temperature is increased. 10. Most thermal IR systems "see" IR system based on its capabilities 4. A blackbody is anobject that fairly well through smoke but not and limitations. completely absorbs all of the radiant well at all through clouds.

1991 Volume 52, Number 3 5 shortens, an interesting phenomenon spectrum. Detectors are usually is immediately clear: Is that spot hot? In occurs: People can see the radiated transducers (for example, the radio)­ a smoke-covered area, what is the fire's energy. Our eyes can reasonably detect devices that change one type ofenergy perimeter? Where are the hot spots? IR energy between about 0.37 and 0.75 into another. In IR systems, detectors systems can detect a hot area or spot, micrometers in wavelength. (Remern­ tum radiated energy at low levels into unable to be seen or accurately assessed ber, a micrometer is only I millionth of electrical e"ergy at low levels. These by the eye, that could take offinto a meter.) In fact, our eyes can sort these low-level electrical "signals" can then destructive fire. The !R systems can wavelengths out into various segments be amplified to more usable levels and also detect the land features where those and that is QPw we see color. Green, for processedin Various ways until they fires and hot spots are located, IR example, is the Part ofthe visible Can be used to expose film or be systems extend our capability to "see." segment ofthe electromagnetic displayed on a cathode-ray tube (CRT) (See "Interpretation Complexities" in spectrum falling between 0,4930 and or television scre~n in the visible accompanying box.) For the U.~. 0.5770 micrometers. Wavelengths spectrum. That's something like how military, IR systems are indispensible beyond 0.37 micrometers (X-rays, our television systems tum signals into fo~ such' operations as detecting gamma rays, and cosmic rays) are no imagespeople can set:? and-occasion­ activities carried on at night or under longer visible. Thus our natural ability ally--enjoy. camouflage. Electromagnetic energy to detect radiated energy is limited to a found in other bands of the EM small portion of the spectrum which we Why Use IR Imaging in spectrum also contributes similarly: call the "visible spectrum," although as .fire Detection? Consider what radar means to air mentioned earlier, we can feel heat in controllers and X-rays t') doctors, certain circumstances, The varying energy wavelengths, as dentists, and security checkers. The sources of energy in the visible defined within the electromagnetic S;lVing Money, Saving Lives. Most spectrum are the sun, stars, and artificial spectrum, have their own special incident commanders who have had an sources, such as light bulbs. In the characteristics-and limits. Within the opportunity to use IR will confirm that daytime, to see an object, our eyes visible spectrum, for instance, an it can save money. IR intelligence also detect energy generated by the sun and observer cannot teIl whether an object is saves lives, keeping firefighters out of reflected by the object our eyes are hot or cool, unless it is "red hot" jeopardy. Dick Montague, retired Forest focusing on. If that object only reflects (glowing or ablaze), nor see well Service regional fire and aviation energy in the green part of the visible through smoke. Within the IR band, IR director, tells many stories illustrating spectrum, then our eyes will see it as systems can detect thermal energy (how the value of IR: green. If it reflects energy all across the hot something is) and produce what is • A fire, thought to be cold, takes off visible spectrum, our eyes will see it as essentially a "thermogram" or a picture when winds come up and inflicts white-unless the object does not of the thermal energy content of a heavy damage on high-value reflect energy well-then it will be scene. They can do this through smoke, property. IR used for reconnaissance some shade ofgray. to areas of various size, and from a around the fire perimeter could have Even though we see energy only in the distance. Both visible and thermal IR found the hot areas and the damage visible spectrum and feel radiated heat energy bands are severely attenuated by avoided. at certain temperature levels in another moisture such as fog or clouds. (See • In another incident, IR was used to part ofthe spectrum, we know energy is "The Ten Most-Asked Questions on continn that there were no hot spots everywhere. Sir William Herschel Infrared" and "Answers" and "The Five or areas that might flare up. Crews found this out some time ago. (See Most Misunderstood IR Notions" and were confidently released, saving "How IR Was Discovered" in "Primer "Discussion" in the accompanying the costs of keeping crews in the on Infrared.") Physicists and engineers boxes.) area as a precaution in case of have developed materials and methods Extending Our Sight. The impor­ another outbreak. for detecting energy in most parts ofthe tance of IR systems to the fire manager

6 Fire Management Notes Interpretation Complexities located regardless of smoke cover, day implemented, primarily because of the or night, with IR mapping missions, cost and complexity of the ground­ How hot is that roof? Can the time of Knowing the exact location of the fire receiving stations. There ar~e only two day mislead interpretation of IR perimeter and hot spots is a prime mobile ground stations in the Forest imagery?. necessity in planning the suppression Service that can be used, . To know how hot something fairly and control actions. Forward-Looking Infrared simple, like a roof, is through IR Systems. Small handheld or turret­ imagery, more information than might What Kind of IR Instruments or mounted forward-looking infrared ~ expected-is needed. Consider these Systems Are Available! ' (FUR) systems became commercially .§ituations: ~ A metal.roof.with a high reflectivity available in 1979.(See "Imaging­ .' may appearvery cool jfit is reflecting Airborne II! Line Scanning Nonimaging" and "Line Scanning" in . the op,ter .~~mosphf;!y or may tlPpear Systems, Airbome IR line scanning "Primer on [nfrared,") The previous .•.. vep' 1)01ifi' h~P!'eDs to he reflecting systemshave been the primarysource FLIR systems were military, large fU!P so!~ el1§rgy-regardless,pf what i!§ ofForest Service IR information for the expensive, 'Since Fl,IR systems have actual temperaturemay be" at the tim!!_ last 25 years. (See tables I and 2 for become available commercially, they .~·A 'roofmade-ofmaterial. thatis a good chronologyof systems, fig. I, and "Line have been used increasingly for fire 'a~prber.m~Y - .. .-. 't appe}ll" warm, if on a, ' Scanning" in "Primer 01) Infrared.") detection and mapping and f'if a .house heated !!lat!s butpoorly . These systel1)s have been highly reliable number of nonfire applications as well. " insulated, Q!cold if the house is well with exceptional small hot spot They haye a much sJ11a1I\,r field of view insulated and the outside temperature i~ 0014. However", if the sun is shinir!g-. detection capability. They are able to than the airQorne line scanning system, squar~ I, the roof may have been warmed by scan large areas, over 1,000 and thus cover a smaller area, but they , solar energy and ap~arwarm even miles (2.590 km-) per hour from produce a clear IR image that can be with a-cold outside temperature. selected altitudes and produce excellent recorded on a standard portable • On a warm, sunny aftemcon, rocks or hard copy imagery. IR line scanning videocas;~tte recorder (VCR) for '" bart: terrain may ~ heated so that systems are installed on twin turboprop­ subsequent viewing if required. Used t4~y t~4jating ~fl.pugh are IR energy sized aircraft, flown (usually at night) with helicopters. they are a versatile .' . -" ',. '; to make it difficult to distinguish them by pilots, and operated by IR techni­ tactical information source. They may from fire-related hot spots and the cians specially trained in IR line be mounted ill a turret or held by hand warmer terrain severely limits IR scanning. (See fig. 2 for comparison of out the o~n door of a helicopter or detection capability During theday, water bodies will usually be cooler area scanned by various systerns.) small fixed-wing aircraft, (Helicopter than terrain and at night warmer than Two important limitations to airborne operations are usually restricted to terrain. Therefore, twice a day, IR line scanning systems ar~ connected daylight hours.) They can provide much usually early momingand evening, to timeliness: The length of lime it takes more detail of selected areas than the the"water and land temperatures will to deliver a film image product, usually line scanners. When kept near on the be about the same, and water features by laridingthe aircraft at the nearest incident command post (lCP), they can will he hard'0 distinguish. After a available air-strip,and 10 manually be ready for use on short notice. Several rain(both dayor nigh'), land interpret imagery to locale accurately contractors have added FLIR capability temperature differences will be small, the fire perimeter and hot spots on a 10 their helicopters. and roads and features hard to see. map. (See "The Five Most Misunder­ F1RE MOUSE TRAP (First· stood IR Notions.") The capability to Generation). FIRE MOUSE TRAP' transmit information from air to ground (FMT) (a name inspired by the then- Small fires previously unknown can exists. Starting in 1974,air-to-ground be found with IR detection missions, transmission of the imagery has been 'Flying InfraRed Enhanced Maneuverable and appropriate action taken. Large fires demonstrated in several ways. These Operational User Simple Electronic ra~tical with poorly defined perimeters can be systems, however, have not been widely Reconnaissance and Patrol.

1991 Volume 52, Number 3 7 Table I-A 30-year review ofachievements in infrared imaging and rhe equipment and aircraft used'

Year Aircraft Equipment Results

1962 Beechcraft AT-11 AAS/5 scanner First imagery through smoke. Preliminary detection of small fires under forest canopy. 1964 Beechcraft AT-11 AAS/5 (modified for Polaroid 16 flights over wildfires, imagery dropped to fire camp. readout) Data collected on detection probability versus scan angle in four coniferous types. 1964 Aero Commander 500B AAS/5, Polaroid 49 flights over wildfires, experience in use of imagery for fire control. Convair T-29 AAS/S, KD-14 rapid film processor No data due to equipment problems. 1965 Aero Commander 500B Reconofax XI scanner Preliminary evaluation, no data due to equipment problems. Convair T-29 RS-7 scanner, Litton CRT, KD-14, Data collected on detection probability versus scan angle in tape recorder three coniferous and three deciduous timber types. 1966 Aero Commander 500B Reconofax Xl, Dual Polaroid System delivered to Division of Fire Control for operation. Convair T-29 RS-7, Litton CRT, KD-14, tape Data collected on detection probability versus scan angle recorder, APN 81 Doppler in one coniferous and two deciduous timber types. First fire patrols. 1967 Aero Commander 500B Reconofax XI, Dual Polaroid Operational. Convair T-29 RS-7, Litton CRT, KD-14, target 21 fire detection patrols. discrimination module (TOM), Bendix DRA-12 Doppler 1968 Convair T-29 RS-7, Litton CRT, KD-14, TOM, Equipment modified for 2-color system and to reduce DRA-12 Doppler size and weight for Installation in smaller aircraft. 1969 Beechcraft King Air, RS-7, Litton CRT, KD-14, TDM, Testing and 25 regular fire detection patrols. Two B-90 ORA-12 Doppler, 2-color detector problems. temperature discriminator 1970 Beechcraft King Air, Same as 1969 Detector problems solved. Successful test. 41 regular B-90 detection patrols and 15 large forest fires mapped. 1974 Sweringen Merlin FFS-1 Forest Fire Surveillance Procured for National Forest Systems. (modified RS-7 to RS-25) 1983 Beechcraft King Air, FLAME unlf Built to Forest Service specifications of FIRESCAN B-90 Research. Forest Service supplied line scanner main frame and image recorder. Jet Propulsion Laboratory (JPL) updated electronics and assembled system. 1985 Beechcraft King Air, FLAME unit2 (Older Beechcraft Queen Air was retired from IR service, 200 reducing IR line scanner capability from 3 to 2 aircraft.) 1991 Sweringen Merlin FLAME2 and preliminary Firefly Initial airborne testing of Firefly electronics, coupled system equipment with FLAME IR system. 1992 Sweringen Merlin Firefly system No.1 and No.2 Acceptance testing of 2 operations. Firefly systems (scheduled).

'Updated table from John Warren's and Ralph A. Wilson's "Airborne Infrared Forest Fire Surveillance-a Chronology of USDA Forest Service Research and Development." U,S. Department of Agriculture. Forest Service, Intermountain Forest and Range Experiment Station, Gen. Tech. Rep. INT-115, August 1981. System and equipment development was conducted within Project FIRESCAN program trom 1962-1974.

~Fire Logistics and Mapping Equipment, latest of the traditional line scanners.

8 Fire Management Notes Table 2--Airborne line scanning systems Assistant Director ofFire Operations John Chambers when he compared the System moving of a mouse around the fire Texas perimeter on a computer screen to the Texas Instruments flying of a helicopter along the fire Instruments R5-25 perimeter) is a system using FUR. In HRB Singer R5-7 {Sweringen 1984, the FMT was conceived and, Item (Queen Air) (King Air) Merlin) FLAME' Fireflyl since then, developed into a useful fire Scanner mapping system (Dipert and Warren). (receiver) (See fig. 3.) The first operational use of Design age 1962 1962--159 1962-73 1962-81 1990 FMT was in 1985 with a system Date acquired 1964 1965 1974 1983 199 developed jointly by the Forest Service Operational 1966 1971 1974 1983 1993 and the Alaska State Division of date Forestry. That FMT system, named Source Purchased Purchased Modification Basic R8-7 Daedalus TROLL,' is described in detail in by Office of by Forest of R5-7 and scanner with Corp. Ronald G. Hanks' doctoral dissertation Civil Defense Service AA5-18to improved (1986). The FMThas been used and "given" Research, produce signal successfullyon manyfires in several to the Forest used about R5-25 processing regionseachyearsince its introduction. Service. 6 years,then transferred to To gather information, a FUR, National Forest mounted on the helicopter or small System. fixed-wing aircraft, is used to guide the Imageproducer pilot around the fire perimeter and to Type Polaroid KD-14 wet Electro Dry silver No hardcopy identity hot spots. On this flight, a chemical Mechanical image. A LORAN-C navigation receiver Research dry videomonitor determines the latitude and longitude silver. is on aircraft. Fire data are coordinates ofthe aircraft every 2 transmitted seconds, and sends data via an RS-232 to the ground interface to be stored in a laptop on a map. personal computer (PC). Design age 1965 Late 1960's 196&-1970 1981 1990 On completion of the flight, the PC is connected to a drafting plotter and the Laboratory One of 14 Note: a EDO-Western Source fire perimeter and hot spots are plotted prototype military later model built by prototypes dry silver to scale on a U.S. Geological Survey Northern built and image producer (USGS) quad-sheet size overlay in a Forest Fire then was procured matter of minutes. The plot is as Laboratory dropped. by NFFL for accurate as the pilot's ability to fly over {NFFL}. Went to dry FIRESCOPE the fire perimeter. minus navigation and silver, then in 1975. backto wet Modified for electronics error. This method of fire chemical. wet chemical in mapping does not depend on an 1975-76. intermediate, indirect step of producing ,Infrared systems can be operated inanyForestService infrared aircraft. a film strip image with a variable,

"Thermal Recorded Observation andLoran Locating.

1991 Volume 52, Number 3 9 Target discriminator Recorder Channel module 8-14\-1 B

Monitor

Pulse height and width K

Receiver Computer

Drift angle Off Track and x c::=J c::=J '~/1\ / / \ grl~r '*---"\1--+t , n II !! /

Figure l-The traditional infraredfire surveillance system used in airborne line scanning. nonlinear scale, which must be trans­ FMT capability to their helicopters. Clearwater National Forest are among posed to a map with a different scale. A There are also a number of FMT those who have successfully used GPS video ofthe FUR flight, which includes systems and trained operators in various (FMT style) for fire mapping (Drake the latitude and longitude location from Forest Service regions. and Luepke 1991; Bonney 1991). the LORAN-C, is made on a VCR for FIRE MOUSE TRAP (Second Others such as Gary Bergstrom of the subsequent viewing. A color television Generation). The FMT has developed Rogue River National Forest has used video, particularly useful for viewing to a stage that might reasonably be the method for nonfire area mapping surrounding background terrain and called a second generation of FMT, or (Bergstrom 1990). Tom Bobbe of the foliage, can be used as an alternative to FMT II. The FMT IR mapping system, Nationwide Forestry Applications the FUR video, but that video will not the "most modular" ofall the IR Program (NFAP) has demonstrated contain any thermal information. mapping systems, can use new compo­ GPS with real-time differential correc­ The data can be transmitted from the nents as they are developed or become tion (Bobbe 1992). Differential aircraft to the ground, via a common available commercially. The FMT II is correction with post processing of the Forest Service radio connected to the a menu-driven system, uses MS-DOS, GPS data has also been demonstrated. PC with a packet modem. Usually, and consists ofa GPS receiver, a Richard Myhre of the Methods when using a helicopter, the transmis­ notebook-sized PC, and a new, less Applications Group of Forest Pest sion capability will not be needed expensive plotter. Management, Washington Office, has because the helicopter can land near the Some regions have already success­ also successfully mapped pest-infested ICP and the notebook PC or a floppy fully demonstrated the FMT method areas using some of the GPS-FMT disk with the data is readily available using GPS. Douglas Luepke of the concept and equipment (Myhre 1990). for plotting the perimeter and hot spots. Southern Region, Philip Drake of the FMT "Super System."An FMT Some contractors have added the Northern Region, and Byron Bonney of "super system" has the capability of

10 Fire Management Notes transmitting a FLIR or color television similar to the super FMT systems, but specialist, and the Alaska Department video to an Iep or other location in without the live video transmission of Natural Resources, Division of real-time. Two super systems are capability. Dennis Pendleton, formerly Forestry, all actively supported the available in Pacific Southwest Region; ofthe Alaska Region and now in Fire development ofthe FMT capability one, in Los Angeles County; and one, Planning Cooperative Programs in the Pyroelectric Units. The pyroelectric on the Eldorado National Forest. John Washington Office, and Cathy Scofield, unit is an IR imaging system using a Bryant ofPacific Southwest Region Alaska Region cooperative fire different physical phenomenon for IR South Zone and Robert Woods ofthe Eldorado National Forest have been Airborne instrumental in establishing this capability. Both ofthese super systems work in conjunction with a mobile ...... receiving station that can be driven to an appropriate location near the lCP. Control Fire staff on the ground can observe, in unit real-time, the fire in progress as shown by RJR or color TV presentations. (It may not be national network, but it's not bad!) Alaska Region has a capability GPS receiver ...... c:::::::J

...... Notebook PC Ground

...... 1/2 inch VHS VCR ...... ~ Line scanner - (Covers 9,800 acres Notebook PC or 3,966 hal Firefly Plotter (Covers 4,900 acres 1 or 1,983 ha of fire area or 7,000 acres or 2,833 L D ha area of interest) USGS quad or W L- ather map or FUR (21 0 vertical ~ x transparency 28 0 horizontal) Monitor (Covers 150 acres or 61 hal

Figure 2-Area coverage (6,000 ft or 1,829 m) from the airborne line scanning, the forward­ Figure 3-FIRE MOUSE TRAP system (airborne and ground operations) using «forward-looking looking infrared (FUR), and the Firefly systems. infrared (FUR) system.

1991 Volume 52, Number 3 11 detection called the pyroelectric effect. direction that they are pointed. They That we have adapted to commonly The pyroelectric unit uses a pyroelectric can be helpful in examining previously available commercial units lets us take detector that depends on the rate of burned piles of material or lines to see if advantage of a much lower price than change of the detector temperature there is any latent hot spot that cannot what we would have to pay for spe­ rather than the actual value of the be seen visibly. (See "Imaging­ cially developed equipment or items temperature. The heating effect ofthe Nonimaging" in "Primer on Infrared.") procured to mil specifications. We can incident radiation causes a change in Sourees for Equipment. Forest take advantage ofdeclining costs, some physical or electrical property of Service IR systems seem to fall in that usually associated with large quantity the detector. (See "Infrared Detectors" never-never land between military and purchase. This also has the advantage of in "Primer on Infrared.") The instru­ commercial systems. Forest Service being able to adapt easily to new ments used in this system are called systems do not have the same perfor­ equipment as it comes along. The FMT pyroelectric vidicons (PEV's). The mance needs as the military, but some II, incorporating new technology and image quality ofthe PEV is not as good needs, such as for airborne line components as they become available, as that ofFLIR imaging systems, but is scanners, exceed those of the military. for example, does not use any of the adequate for many fire situations. Jim For instance, the Forest Service need to original FMT equipment, yet retains Scott ofthe Southern Region has used detect very small hot spots exceeds that compatibility, while taking advantage of PEV's effectively on fires ofvarious of the military. DC coupling is not cost, size, weight reduction, and sizes and conditions. Pyroelectric units usually needed by the U.S. military, but performance improvements. do not require liquid nitrogen like the the Forest Service needs this feature to Some contractors have added the line scanner or FLIR systems do; are fly over large fires and retain the ability FMT capability to their helicopters. handheld, portable, simple, and to see terrain details right next to the There are also a number of FMT straightforward to operate; and produce large hot areas, develop and review film systems and trained operators in various a video image that can be recorded on a on board the aircraft, and refly the fire if Forest Service regions. standard VCR. necessary. (The military might like that Hughes Probeye aud Nouimagiug capability, but not dare fly back over an The Latest and Future IR Systems Units. The "venerable" Hughes area too soou-they would probably get Probeye produces only a rudimentary shot down!) Airborne IR Line Scanuing image. It is a cryogenically cooled Forest Service systems have perfor­ Systems. The Firefly system is an instrument using pressurized argon gas mance needs beyond what's available in airborne IR line scanner system with as the coolant instead ofliquid nitrogen the commercial market of line scanners, substantial changes from the traditional or a mechanical refrigeration method. but the quantities we need are too low methods of processing and presenting The output ofthe scanned detectors is to get the attention ofany manufacturer. IR data. Firefly uses the GPS satellite­ displayed on an array of red light In the case ofthe FLIR and PEV, we are based navigation system for aircraft emitting diodes (LED's). The relative able to use the commercial product line location. In the future, Firefly will use brightness ofthe LED array produces a equipment as marketed. For FMT, we aeronautical satellite communications basic image, characteristically red­ have been able to use all commercially for the transmission of IR data (see tinted. The Probeye is portable and easy available equipment, integrating and fig. 4). to use. Its main disadvantage is the interfacing among the various sub­ In Firefly systems, IR technicians lower quality image and the need to systems so they all work together as process data on board the IR planes, handle and keep a supply of pressurized planned. For instance, the FLIR can be probably jet aircraft, and transmit this argon gas. interchanged with other FLIR's or data by way of satellite. Then, within 30 The nonimaging units only give an PEV's; the GPS's and the LORAN-Cs minutes or so of a completed flight, the indication by a light, noise, or meter Can be interchanged within kind and ICP receives fire and hot spot locatiou that there is something hot in the with each other; notebook computers information. While the data are being can be substituted for laptop. transmitted, the aircraft is able to go

12 Fire Management Notes Communications directly to the next fire. Firefly elimi­ satellite nates the time-consuming aircraft Aircraft EHB-f+m+, landing at the nearest available field for with global q..~ physical delivery ofthe IR image. positioning ~ Firefly locates all hot areas and hot ~ system unit spot positions and establishes an "intensity level" with each, but not the Aircraft with line scanner and burned perimeter. The entire image, it is global positioning anticipated, can be transmitted by system unit satellite eventually, along with the hot area and hot spot locations, so that relative positions can be readily Incident command post determined. Firefly will use commercial unit line scanners, not the military line scanners that have been used for 25 I " years. Figure 4(a)-The Firefly concept. Firefly has been under development for the last 4 years. Tests of the system will be completed by the end of A.ircraft Ground September 1992. Two operational Firefly systems are scheduled for delivery to the Forest Service in fiscal Video year 1992. Because the Department of recorder Defense imposes accuracy limitations on GPS and aero satellite communica­

Video tions are not yet reasonably available, monitor the Firefly system's potential may not be realized fully for a few years. FUR-FIRE MOUSE TRAP. The use ofFLIR-FMT for fire mapping and Data Processing Navigation Subsystem Subsystem detection will increase as fire managers recognize and understand its capability, GPS receiver I Computer handheld GPS units become more available, and people learn how to use Rate gyros Operatorterminal them. The advantage of using FLIR in conjunction with aerial retardant drops Heading gyro Data storage has been shown and documented by Charles George, Intermountain b%ii@§! Supplied by USDA Forest Service Research Station's project leader for ~ Commercial Daedalus infrared line scanner operational retardant evaluation (George 1989). That use is clearly Figure 4(b}-The Firefly system (airborne and ground operations.) needed and overdue.

1991 Volume 52, Number 3 13 Further Out in the Future It is entirely possible for all fires to the selection process. Adequate and have adequate infrared intelligence at appropriate communications systems The use ofvarious types of IR to all times-the implementation costs and equipment are taken for granted on provide adequate IR intelligence, for achieving this would be quickly all fires now, but IR intelligence coupled with satellite transmission of returned. gathering and USe has not kept pace. It imagery and data, geographical is entirely possible for all fires to have information systems (GIS)-rype IR adequate IR intelligence at all times­ layers compatible with Forest Service Infrared System for a Firefighting the implementation costs for achieving GIS (see fig. 5), INCINET (Incident Job.") Research efforts to confirm and this would be quickly returned. Network), and fire behavior worksta­ formalize IR selection and also to Age-Old Question. If over 25 years tions operating in current and projected establish and document equipment and ago, the Ranger space probes could future conditions, will greatly enhance procedures for retardant activities are send back color television pictures from fire management effectiveness. This needed now. A guide to selecting the the moon, why can't firereconnaissance will result in more fires burning under best, or at least adequate, type of IR find fire and communicate that to fire prescribed conditions, fewer escaped systems for the conditions being managers? The answer to such a fires, and less damage to all types of experienced has been explored, but as question is not technological capability, resources. yet not realized. Cost, system perfor­ but the time and money invested to Low-earth-orbit (LEO) remote mance, and information need are part of operate the IR intelligence. Just think sensing satellites may someday be used for fire mapping and detection. Current satellite capabilities have been studied Typical GIS Potential Ownership but are not used at this time. There is a "Fire GIS" Layers bit ofhyperbole about their use, and layers some fine looking images generated now and then, usually after such fires as the Greater Yellowstone Area or Oakland Hills, but high cost, lack of availability, and inflexible orbits (see "How Useful Are Satellites in Fire Detection and Mapping" in accompany­ Land cover ing box) are some of the reasons why we do not use satellites for IR intelli­ gence now. Base map More Effective Use of Infrared Systems Fuels Improve Selection. Judicious Fire perimeter selection ofthe best available IR system D '--, that can do an adequate job would ensure that the limited quantity ofIR mapping systems can be available on all fires when needed. (See article in this issue entitled "Selecting the 'Right' Figure 5--lnterpretation ofthe typical geographical information system (GIS) layers as potential ''fire'' GIS layers.

14 Fire Management Notes what would happen with half the time How Useful Are Satellites in Fire Detection and Mapping? and a small fraction ofthe money devoted to the lunar program! • Drawbacks • Detection of small hot spots, (less than 6 inches or .05 meter square) is Literature Cited • Spatial resolutions from orbital not possible with any known, altitudes are not as good as those nonmilitary satellite at this time. Bergstrom, Gary C. 1990. GPS in forest from aircraft altitudes (selected on •A satellite's pass over an area of management. GPS World. September­ the basis of fire size and conditions), interest is fixed in time, direction, October: 46-49. even with a fairly low resolution, 2,5 and orientation, (Aircraft normally Bjomsen, R.L. 1989. Infrared fire mapping: the milliradian airborne system, make several passes from different untold story. Fire Management Notes. 50(4): directions over a fire to obtain 13-15. • Orbiting satellites follow a fixed Bobbe, Tom. 1992. Real-time differential GPS: celestial track, oblivious to the fire several perspectives, an aid in the an aerial survey-remote sensing application. mapping needs of fire managers. interpretation process.) In: Proceedings of the fourth Forest Service (Aircraft can be scheduled by fire remote sensing applications conference; 1992 managers to support operations.) Advantages April 6-10; Orlando, FL. Bethesda, MD: Satellites in geostationary orbit, American Society for Photogrammetry and although available over a large given • Satellite imagery is useful in tracking Remote Sensing. [In press.l area, are too far out (22,300 mi or Bonney, Byron J. 1991. GPS uses in fire large fires in vast, remote areas such 35,887 km at the equator) for management on the Clearwater National as Alaska, where suppression actions adequate resolution with current Forest. Fire Management Notes. 52(3): may not be feasible or warranted. technology. 35-36· • The use of satellites for transmission Dipert, Duane; Warren. John R. 1988. Mapping • Receiving satellite imagery at the of lR data or imagery from the IR fires with the FIRE MOUSE TRAP, Fire incident command post (via direct or aircraft to the ground should be Management Notes. 49(2): 28-30. indirect means) is either too reasonably available in the next few Drake, Philip; Luepke, Douglas. 1991. GPS for expensive or time consuming. years. The use of satellites, as in the forest fire management and cleanup. GPS • Clouds may obscure area of interest World. September: 42-47. Global Positioning System for from orbital altitudes. (Aircraft can George, Charles w.; Ewart, Gerald, F.;and location ofIR aircraft or helicopters, often fly beneath clouds to obtain the Friauf, Walter C. 1989. FUR: a promising is in use now, tool for air-attack supervisors. Fire Manage­ needed imagery.) ment Notes. 50(4) 26-29. Hanks, Ronald G. 1986. The TROLL System­ Thermal Recorded Observation and Loran Locating System. San Rafael, CA: Columbia Pacific University. p. 196 plus appendixes. Ph.D. dissertation. Myhre, Richard; Graham, Lee A,; Sumpter, Carl W, 1990. Airborne video technology. In: Proceedings of the third Forest Service remote sensing applications conference. 1990 April 9-13; Tucson, AZ. Bethesda, MD: American Society for Photogrammetry and Remote Sensing: 293-298. Warren, John and Wilson, Ralph A. Airborne infrared forest fire surveillance-a chronol­ ogy of USDA Forest Service research and development. Gen. Tech, Rep. INT-115. Ogden, Utah: U.S. Department ofAgricul­ ture, Forest Service, Intermountain Forest and Range Experiment Station. 8 p.

1991 Volume 52, Number 3 15 Infrared Radiation the spectrum.He found that when he Primer on Infrared placed the thermometerbeyond red Electromagnetic Radiation Warm objects (an object or material where he could see no light it still gave at any temperatureabove absolute zero a reading. He had discovered a form of Electromagnetic radiation,or radiant (-273°C» send out radiant energy at radiant energy,able to be refracted by a energy, is the transmissionof energy various wavelengths and frequencies.If prism, but invisibleto the eye. through space in a wave motion, or wavelengthsof that radiated energy,fall Radiant Energy and the Laws of electromagnetic wave. Such a wave, within the 0.72 to 1,000micrometer Optics. Following up this discovery, consisting of an electric fieldand a band of the electromagneticspectrum, Herschel showed that this radiant magnetic field,occurs at a certain they are below ("infra") the red limit of energy, or radiation,is reflectedand frequency (the number of to-and-fro visible light. refracted according to the laws of motions per second) and wavelength Who First "Saw" the "Invisible" optics. He also showed that heat (the distance traveled during this World oflnfrared? In England in radiation from sources such as a candle, oscillation or from the top of one wave 1800, Sir WilliamHerschel, a scientist a fire, a red-hot poker, and even a to the next). The range of electromag­ and astronomer,using a prism and a domestic stove-with no visible netic waves, measured in micrometers thermometer,experimented with the radiationobeyed the same laws as solar (3 x 10'" micrometers to infinity),or heating power of the colors in a beam radiation. Herschelconjectured this hertz (1023 to 0 cycles per second), is of light. Herschelplaced a prism in a radiation was of the same nature as known as the electromagneticspectrum. window to catch a lightray,shooting light, differingfrom it only in The chief kinds of electromagnetic that ray with its color spectrum to a "momentum." He called it "invisible waves are (from the shortest to the table holding a mercury-in-glass light." longest) gamma rays, X-rays, ultravio­ thermometer (heat detector). In his let light, visible light, infraredrays, measuring of the heat of each color, he microwaves, and radio waves. (See was surprised to find thai the heating diagram of spectrum.) power increasedtoward the red end of

,Visible Gamma: I II x-rays Ultraviolet! : Infrared Radio rays I I UHF VHF HF MF LF VLF \

o.lA lA loA looA 0.11-1 11J 10jJ 01-1 O.1cm tom 10cm 1m 10m 100m lkm 10km 100km Wavelength 19 18 17 18 3x10 3x10 3xl0 3xl0 3x 15 3xl014 3xl013 3xl012 011 3x1010 3x109 3xl0a 3x107 3xl08 3x105 3)(104 3x103 Frequency, Hz

I : Visible Near infrared Middle Far Extreme v BGYOR infrared infrared infrared

0.4 0.6 0.8 1 1.5 2 3 4 6 8 10 15 20 30 Wavelength, IJ 25,000 10,000 5,000 2,500 1,000 500 Wavenumber, cm-t

16 Fire Management Notes I ISome Definitions ofEssentials in There is a direct interaction between the cannot be sure an area has been IInfrared Imaging incident photons, from the "heat adequately covered. source" and the electrons of the detector An imaging system such as a I Infrared Detectors, Infrared material. The change in voltage output forward-looking infrared system (FUR) i detection has come a long way since or in the conductivity of a quantum is made up of these main parts: Optics Herschel's prism and mercury-in-the­ detector is proportional to the number such as lenses or mirrors to image the glass thermometer experiment. A of photons ofenergy arriving from the rays, a detector to sense them, and a variety of detectors, available for heat source. Quantum detectors have means to communicate their messages. sensing radiant energy, are used in high sensitivity and fast response times Imaging systems provide an image (not infrared sensing equipment or systems. but usually require cryogenic, or low a photograph) ofthe scene by "scan­ " They can be classed into two main temperature, cooling, ning," or sweeping over a field of view Igroups: thermal detectors and quantum Ways of Classifying Infrared (back and forth 'and up and down) to ;Idetectors. Instruments or Systems. Infrared collect or map information. (See "Line 1 Thermal detectors absorb the radiant systems Can be classified in various Scanning.") From this information the energy causing a rise in the element's ways-for example-passive or active temperature of objects or areas within Itemperature. The rise in temperature or imaging or nonimaging, the scene is deduced by comparing the then causes a change in a physical or Passive-Active-Passive infrared "brightness" of those objects or areas to electrical property that can be mea­ sensors detect the natural emission of the background. In other words, hot sured. Some examples of thermal radiant energy that occurs in all objects objects are located based on their detectorsare thermometers, thermo­ and is directly related to the temperature relationship to the total scene, The couples, thermistors, and pyroelectric ofthe object. Fire detection and image of the covered area is shown on devices. In the mercury thermometer, a mapping infrared systems are passive the sensor's display, on a video monitor, Ithermal detector we all are familiar systems. An active sensor carries a on a hard copy printout, or some with, heating increases the volume of radiation source along with a receiver. combination ofthese. Video compatible mercury in a graduated glass tube, A common active sensor is a radar set, systems can also record the infrared Icalibrated to a determined scale. The usually operated in microwave bands, images with audio annotation. Ithermocouple, two different metals, the not in the infrared bands, that transmits , junctions of which are at different energy out and detects the energy ZJ.'=S;;r Swathwidth !I temperatures, proviides a way to reflected back to its receiver. /fII-.-.\---IFOV 1 measure heat. Thermal radiation lmaging-Nonimaging-Nonimaging One line

II changes its electrical characteristics. systems give an indication ofthe The thermocouple, for example, temperature of an object or area i triggers the furnace to tum on. In a averaged over the field of view of the \ pyroelectric device a change in sensor. This information is provided in : temperature creates a change in a qualitative (light or hom, for in­ Ipolarization or a change in electrical stance), quantitative (calibrated Line scanningfrom an airplane. I charge. A change in electrical charge temperature reading), or relative

! with time is electrical current. A (deflection of meter from an Line Scanning. A line scanner I pyroelectric detector produces current uncalibrated reference) manner. These (scanning mirrors, lens, detector, and I as it experiences a temperature change. are relatively inexpensive, low perfor­ electronic amplifier) detects, or views, I Quantum detectors or photon mance instruments useful for detecting energy (thermal energy in the case of Idetectors directly use the energy heat differences in nearby (20-to-30·ft­ thermal IR systems) from the area of I contained in the photons arriving to or 6.I-to-9.4-m-range) hot spots that one instantaneous field ofview (IFOV), cause electrons to go from a noncon­ might not otherwise be detected by an also referred to as a ground resolution ducting state to a conducting state. observer. With this system, the operator element, ground pixel, or simply pixel

1991 Volume 52, Number 3 17 ! Scene I(the basic unit or picture element scanned _ ,making up the image on a video screen) , at a given instant.The scanner sweeps •• •,, across these IFOV's horizontally , "scanning the line." ,, Visualize"scanning" in terms of a ,, Scanninq , mirrors grid superimposed on an image or , picture. Each square of the grid is a Detector Ipixel.One horizontalline of the grid is Electronic Field of view amplifier a line of pixels. The line scanner moves elevation Field of view along a line of our hypothetically azimuth (vertical) (horizontal) gridded area of the ground and then \returns to the starting side of the image i to sweep across the next line. i The line scannerconstantly changes !the thermalenergy present into an Ielectronic signal.The magnitude of the \ electronicsignal at any given instant is ianalogousto the amount of thermal Thej'orward-looking infrared system (FUR)-how it works and its main parts. energybeing emitted from the ground Ipixel.Ifthe signal is processed in this Ianalogousform throughout the system, locationscan be calculatedfrom the done at standard (NTSC) television we refer to it as an analog system.Ifthe !FOV area. the altitude,and the velocity rates (525 lines per image, 30 image signal is converted to a digitalelec­ of the aircraft. It is usually adjustableto framesper second, or about 15,750 tronicsignal, or group of bits, along the cover the range of altitudes and speeds lines per second). way, it would usually be called a digital that the aircraftcan cover.It is even Other Systems. Other systems,such I, system.There are several advantagesto given a name, VtH. which is velocity as linear arrays, use an array of idigital signal processing, one of which divided by height (above ground level). detectors in a line to output the total : is the ability to enter the digital signals Althoughcontiguous scanningof lines horizontalline signal without optical Iinto digital computers (most computers is the most "efficient," Forest Service scanning.Area arrays have one small Iare digital) for processing,storage,and IR line scannersintentionally overscan detectorelement for each IFOV, or otheroperations. The resultsare usually so that each "line area" is covered at square in the griddedarea, so that I, thenconverted back to an analog least twice.Overscanningbuilds in an neithervertical nor horizontalscanning ,displayfor viewing. effectivefalse alarm rejection method, is required.Althoughthese eliminateor Airborne Line Scanners. By the precludingrandom noise spikes from reduce the scanning,other characteris­ timethe sweep or scan of the first line is showingup as hot targetsor fires. tics of these systemsmust be consid­ completed.the aircraft has moved to the Forward-Looking Infrared. ered before they become part of a i start of the imaging area to scan the Forward-looking infrared(FUR) units system.• , second line, adjacent to the first. Now do not rely on aircraftmovementto 'i Ithereare two lines. This continues until "build up" the image.They still scan the John R, Warren, groupleader, I each line of the grid is scanned and a !FOV one line at a time, but do so AdvancedElectronics, USDA Forest I completethermal image of the total electronically, at a much faster rate, so Service, BoiseInteragency FireCenter, . groundarea completed.The speed that the entire imagecan be displayed Boise,ID. necessaryfor contiguous scanned line on a videomonitor. This is normally

18 Fire Management Notes Selecting the "Right" Infrared System for a Firefighting Job

John R. Warren Advanced electronics group leader, USDA Forest Service, Boise Interagency Fire Center, Boise, ID

Fire managers need to know the Judicious selection ofthe best specific details of a situation with its capabilities and limitations of the types available infrared system that can do close-in IR viewing, coupled with the of IR systems that are available for their an adequate job would ensure that visual observation and judgment of on­ use. Using the "right" or best-suited the limited quantity of infrared the-scene fire staff personnel. infrared system for a job is basic to mapping systems can be available on Asking cost questions are basic to an sound fire management. What are the all fires when needed. Incident Commander's selecting the system types? How do you choose? best IR system for a fire situation (See What do the "right" systems cost? table 2). Fire managers need to ask The types of infrared (IR) systems and general assessment of conditions these questions before making and how they perform in 18 conditions over large areas. The FLIR-FMT decisions: were reviewed and rated by two should be thought of as a tactical • What percentage of the total daily experienced Class I Incident Command­ system, useful for future planning suppression cost will infrared flights ers and the author. (See list ofcondi­ perhaps, but available for gathering be? tions in accompanying box and near-term or near real-time • What is potential loss and additional evaluation results in table 1.) The decisionmaking information and more costs if fire location is not known? results clearly show that relying on only one type ofIR system is not necessary or even reasonable. The main conclu­ sions are as follows: Conditions in Which the Usefulness of Infrared Systems • More that one type of IR system can Were Evaluated do a good job in similar conditions. • Line scanners are not-always the best type of system'for all jobs. Detection Other Fire Applications • IR intelligence can be available on 1. A reported new fire with 9. Retardant effectiveness verifica­ every fire when needed if fire approximate location (known within a tion. managers properly select and use mile or so). lO. Confirmation of fire behavior the IR system best suited for a job. 2. Multiple (3 or more) new fires projections made by fire behavior From this evaluation, for example, in over a large (1,200-to·l,300-acre or analyses. 486-to-526-ha)area. 11. Intelligence for containment or all 18 conditions, two or more system 3. Possible multiple new fires over a control. types rated "very good" or "good." In very large (3,()()()..acre or 1,214-ha or 12. Mopup activities. 14 of the 18 conditions, 3 or more more) area. 13. Documentation of fire activities system types rated "adequate or better." and locations for subsequent legal Forest Service line scanners rated Mapping needs. higher than the other system types in 4. First overall infrared view of a 14. Media releases to show that fire .' only 4 of the I8 conditions. FIRE fire (up to 1.200acres or 486 ha). managers do know exact location of MOUSE TRAP with a forward-looking 5. First overall infrared view of a fires. infrared (FUR) system rated higher large (over 1,200acres or 486 ha) fire. 15. Prescribedbum monitoring. than the other system types in 9 ofthe 6. Subsequent updates of fire 18 cases. perimeter after its location is estab­ Nonfire applications lished. Generally speaking, the Forest 16. Search and rescue. 7. Continnation of tire perimeter or 17. Law enforcement. Service line scanners, best used for hot spot location from other observe­ 18. Pest or disease or other visual large area detection and first-time large tions. mapping. fire mapping, should be considered, 8. Direct viewof fire during or soon usually, as strategic information after runs or flareups and other tactical systems, to be used for future planning information needs.

1991 Volume 52, Number 3 19 Table l-Results from infrared systems evaluation" • What money can be saved if suppression resources can be Forest Service FURwith IRline Contract IR FIRE MOUSE applied better? Condition scanners line scanner TRAP FLiR • What might be the loss if crews are " released too soon? Detection 1 B C A A • What might be the cost if crews are 2 A 8 C held after there is no more fire 3 A 8 danger? Mapping Knowing about these systems- 4 A 8 B what's available, how they function, 5 A 8 C 6 A 8 A C what their limitations and capabilities 7 A 8 A A are, how much they cost (to purchase 8 C C A 8 upfront and to operate}-will help fire Other managers make good decisions about A A 9 what system to use. • 10 8 C A 11 A 8 A 8 12 C C A A 13 A 8 A B 14 A 8 A A 15 8 C A A 16 A 8 17 8 C A 8 18 A 8 ",., "" very good; B =good; C =adequate; rIOletter"" rIOtsllitable. A remarks column was ind\lded in the evaluatiOn but not collated here,

Table 2-Hierarchy of infrared (IR) systems and equipment"

Systems and equipment Estimated cost Military airborne IR line scanners Forest Service Firefly airborne IR line scanners $2,000,000-3,000,000 Forest Service traditional airborne IR line scanners 400,000- 750,000 CommerciallR line scanners, general purpose 250,000- 500,000 Military FLiR Commercial FUR, turret mounted 55,000­ 120,000 Commercial FUR, handheld 30,000­ 60,000 Pyroelectric vidicons 20,000- 25,000 Nonimaging heat indicators 500­ 1,000 Additional equipment for FIRE MOUSE TRAP capabiiity 4,200- 8,200 ,i (used with any FUR or pyroelectic vidicons)

"Prices are approximate (not included for military systems). Note: Pertcrmanca depends on the conditions and what needs to be known. In many cases, a "lower" psrtormanca system may do a much better or just as good a job as a "higher" performance system. It is easy to find situations where a pyroelectric vidicon will do a better job than a top-of-the-list airborne line scanner. The list is not all-jnclusve. but is representative. Prices lnclude hard copy or video image production.

20 Fire Management Notes Forest Fire Detection Systems1

Stanley N. Hirsch

Electronics engineer, USDA Forest Service.Intermountain Forest and Range Experiment Station, Northern Forest Fire Laboratory. Missoula, MT

In 1962, the FIRESCAN project, The 30 years since the FlRESCAN ity to deploy a proper detection system research in infraredfire detection, research began have shown that the when the conditions indicate danger if a started at the Northern Forest Fire "ideal" fire detection analysis system fire were to start. In this version ofthe .' Laboratory. Stanley N. Hirsch, elec­ does not exist in isolation, hut must be "ideal" situation, fires would burn only tronics engineer, was the project leader. used together with otherfire prediction under valid prescribed conditions. A In a presentation to the Western Forest analysis systems. Along with knowing potentially dangerous fire would be Fire Committee, on December 1,1964, where the fires are is the important suppressed before it caUSesdamage, at Spokane, WA, he outlines the information aboutfire danger, accessi­ while a fire notpotentially dangerous attributes ofan idealfire detection bility to fire, and values at risk. Hirsch would be allowed to burn under system and evaluates those attributes in makes this point when he states an ideal prescription and surveillance. light of1964 technology and vision. system "would be able to distinguish -John R. Warren Hirsch's analysis is an enlightening potentially dangerous fires from those benchmark in the ongoing discussion of that would never concern fire control infrared detection systems. forces." What is needed is the capabil-

An ideal forest fire detection system essential attributes related to the varied depends on type and continuity of fuel, would detect fires the instant they start, kinds of detection and control prob­ rate of spread, and time elapsed from day or night, under any condition of lems, Ideally it would be able to detect ignition to detection and from detection visibility. Additionally, it could distin­ a fire the instant it started even though to initial attack. Under some circum­ guish potentially dangerous fires from such instant detection might not always stances, instant detection might be those that would not concern fire be critical. Second, it would be as desirable but not critical. The value of suppression forces. Although attainment effectively operable in darkness as in the area under protection would of such an ideal system now appears to daylight. Likewise it would be operable certainly influence the tolerability of be well beyond the reach of modem despite low visibility caused by smoke, delay; so would weather and fuel technology, detailed study of the fog, or dense timber cover. Fourth and conditioos at the point of ignition. If an requirements for an ideal system may finally, it would be able to distinguish interval of 2 to 4 hours between ignition well show where space age technology potentially dangerous fires from those and detection is tolerable, intermittent could be applied effectively. that would never concern fire control surveillance would be acceptable and an forces. Admittedly, this is a large order, airbome system could be used. Fre­ An Ideal Detection System and each of these attributes requires quently, though, a 2-to 4-hour delay some detailed explanation. would not be tolerable; then a fixed An ideal detection system-the sort location continuous lookout is required. ofequipment a fire control man would Probably there will always be need for dream about-would have four Hirsch's analysis is an enlightening some fixed lookouts. However, in many benchmark in the ongoing discussion situations a 2-to 4-hour delay would be " ofinfrared detection systems. tolerable, and the inherent advantages 'Purchased by the USDA Forest Service for of intermittent surveillance could be official use, this article is reprinted from a reprint exploited, Hence, it is evident that the of Stanley N. Hirsch's presentation originally Instant Detection first major question to be answered in a published in the proceedings of the December I, detection analysis must be; Does the 1964. annual meeting of the Western Forest Fire The primary consideration for any Committee, a permanent committee of the particular area under study require Western Forestry and Conservation Association, control operation is: How big is the fire continuous or only intermittent Portland OR. at the time of initial attack? This size surveillance?

1991 Volume 52, Number 3 21 Operability at Night detection of smoke. But there is great system rely upon some secondary promise for solving the problem of phenomenon produced by the fire, such The second attribute of an ideal smoke-filled attnosphere if the detection as the smoke column? It will be system is equal operability by day and device relies upon energy radiated virtually impossible to achieve continu­ night. Fires usually spread more slowly directly from the fire. The third major ous surveillance of an area if the during nighttime hours, and quite often decision in detection analysis must be detection system must be placed in they could justifiably be ignored until whether we can tolerate loss of detec­ some position far above the earth's daylight permitted detection by visual tion capability under conditions of poor surface. Likewise, it will be extremely means. However, ability to detect fires visibility. difficult to find many fixed locations during the hours ofdarkness would that afford a direct line of sight to large permit suppression forces to attack Distinguishing Dangerous from areas where fires might start. them at a time when control is easier to Harmless Fires The foregoing discussion is not achieve. Visual systems such as manned intended to imply that instant detection lookouts, manned aerial patrol, and The fourth attribute of an ideal can always be achieved from the fixed television require daylight for their system would be ability to distinguish locations. In many cases of interest, operation. Infrared, microwave, or potentially dangerous fires from those fires may bum for extended periods active optical systems work satisfacto­ that will not concern fire suppression before enough smoke rises to permit rily in either daylight or darkness, but forces. It would be desirable to be able detection. A detection system scanning infrared functions better in darkness. to distinguish between a camper's fire intermittently to detect the primary heat The second question to be answered is: and an incipient lightning-caused fire. It source may detect a fire before a Do we need a nighttime capability? would be desirable to distinguish continuous surveillance system would between a backyard barbecue that was detect a smoke column. Operability Despite Low Visibility preparing an evening meal from one that had "spilled" and was starting a Continuous Versus Intermittent A third attribute of an ideal system is major conflagration. However, these Surveillance operability under any condition of considerations hardly fall within the visibility. On many days during the fire scope offorest fire detection. The Within this framework I shall season, the atmosphere is clear and essential role of a fire detection system examine separately the problems of smoke can be detected by various should be to determine the existence of continuous surveillance and intermittent means. However, many of the days any combustion processes. The system surveillance. In beginning with the when visibility is obscured by smoke should be capable of distinguishing situation requiring continuous surveil­ are days when fire conditions are most between true combustion processes and lance, I assume that this precludes the dangerous. When there is no smoke such false indications as dust swirls and use of any aircraft, satellite, or other haze, no smog, or no cloud cover, mist-in the use ofvisual systems--or means of placing the detection system detection of small smoke columns is reflections from bright objects- in the at a very high vantage point above the easy. However, if the atmosphere is use of infrared systems. Any distinc­ surface ofthe earth. It would be cluttered with smog or smoke, the tions beyond this should properly fall impossible to find enough points on the detection of small smoke columns within the province of the dispatching earth's surface that would provide a "I becomes very difficult for either the process. direct optical line of sight between the visual system or other systems that Before evaluating these four at­ detection system and the primary heat would attempt to detect the smoke tributes of an ideal fire detection source-the fire itself. Therefore, column above a fire. Any attempt to system, we must ask one further wherever we require continuous develop systems capable ofoperating question: Is a direct optical line of sight surveillance, we must depend upon under all atmospheric conditions is available between the detection system observation of some secondary doomed to failure if it relies upon the and the fire to be detected, or must the phenomenon.

22 Fire Management Notes The manned lookout depends upon modem electronic technology can best have the angular resolution required in visual detection of the plume or smoke be applied. Fires radiate large amounts fire detection applications, but present column above a fire. If we are limited to of energy in portions of the electromag­ technology can construct infrared using visual systems---either manned, netic spectrum where the eye has no systems that meet this requirement. For television, or a combination ofboth­ sensitivity. Both infrared and micro­ this reason, the current research in fire we are limited to daylight operation wave devices are sensitive to these detection at the Northern Forest Fire unless we provide some active source portions of the spectrum and can be Laboratory under Project Fire Scan has of illumination. Optical radars, employ­ applied effectively to detect small fires been directed towards implementing ing lasers for illumination, may usefully both in daytime and at night. Theoreti­ airborne infrared systems. Although supplement manned lookouts by cally, both types ofdevices can detect microwave radiometry has the potential providing a capability for detecting small fires when smoke or smog for operation under conditions of small smoke columns at night, and by completely eliminates the use of visual complete cloud cover and infrared does providing a very high accuracy in range systems. not, the current state oftechnology and azimuth on smokes detected during dictates the choice of infrared in spite of either day or night. Infrared Versus Microwave this one major drawback. I have seen no other good scheme proposed for detecting any secondary In order to decide which of these Discussion fire phenomena. There has been systems to apply, we must look further considerable discussion about detecting at their similarities and differences. Any practical airborne surveillance carbon dioxide content, the thermal Microwave energy penetrates cloud system must be capable of covering at energy radiated from the smoke cover-infrared energy does not. least 5 miles to either side ofthe column, condensation nuclei, and other Microwave systems' perfonnance is not airborne vehicle. Coverage is deter­ phenomena; but even cursory examina­ degraded by the presence of strong mined by the altitude and scan angle tion of these allows one to eliminate sunlight-infrared performance is. used. Data accumulated during the past them rather rapidly for operational use. Infrared systems are inherently simpler 3 years indicate that timber obscuration From this, I conclude that continuous than microwave, but neither of them becomes a limiting factor beyond 60 surveillance and freedom from atmo­ approaches the simplicity of the human degrees from vertical. Because of this spheric limitations are not compatible. observer. The present state of the art in angular limitation, an operational A television system installed at a microwave technology does not permit altitude of 15,000feet above terrain is lookout could do approximately what construction of airborne devices that required to achieve lO-mile coverage. If the human lookout can do, but it does not appear to add anything in terms of Table I-Rating ofoptions available for use in fire detection systems capability. Ifreplacing a manned lookout with a television system Continuous surveillance Intermittent surveillance reduces cost, then perhaps such a Fixed location Airborne scheme has merit. If such replacement Operating Visual Optical Visual Infrared Mlcrowave cannot be justified on purely economic conditions passive radar passive passive passive grounds, I see no justification for Day, clear A B B C B considering it further. Day, smoke-smog 0 D D BB In situations where delays of 2 to 4 Night, clear X A X A B hours between ignition and detection Night, smoke-smog X D X A B can be tolerated, the detection system Cloud BB C D' B Simplicity A C A B C can be placed in an airborne vehicle to Status of technology A C A B D obtain a direct look at the primary heat 'Depends on degree, complete cloud cover would change this to X. A '" excellent. B '" good, C '" lair, D '" poor, source. These are the situations where X ""unusable.

1991 Volume 52, Number 3 23 the frequency ofcloud cover below this altitude is too great, infrared techniques must be abandoned in favor of micro­ wave. Figure I shows infrared imagery obtained in flight tests over small heat sources in a dense Engelmann spruce stand [not reproduced here]. Figure 2 shows a schematic of a methodfor using airborne infrared scanners in an operational system [not reproduced here]. Table I summarizes the options available in the design of fire detection systems with their attributes indicated inrelative order.

Conclusions

Studies conducted by Project Fire Scan show greatpromisefor improving Minimizing the Riskof Many Perspectives.Thesymposium detectioncapabilities where intermittent consisted of a series of presentations surveillance can be tolerated. If Wildfire: A Symposium to that described thewildland-urban issue continuoussurveillance is not required, Address Wildfire Problems from a variety of perspectives. Presenta­ then present infrared technology may in the Wildland-Urban tions were"madeby leadersin achieve a detection capability during Interface firefighting, planning, publiceducation, both daytime and nighttime hours and politics, insurance, and training. The under conditionswheresmoke and event also included workshops that smog eliminateconventional tech­ Jasper Park Lodge, Jasper, focused on developing solutionsto Alberta, September 26-30, 1992. problemssuch as publiceducation and niques. Although optical radars may Partners in Protection, a multiagency the role of the media, integrated have some application for continuous group formed in the Province of Alberta planning, cooperative training, and surveillance at night,no means are to address collectively wildland-urban influencing the politicalsystem. The availablefor obtaining continuous interface issues andconcerns, hosted program also included a fieldtrip, surveillance when smoke or smog limit \ a major symposiumon September 26- exhibitordisplays,social activities,and visual techniques. • I 30, 1992, at Jasper Park Lodge,Jasper, a program for spouses. AL. The symposium,"Minimizing the More Information. Forfurther Risk ofWildfire" gave an opportu- infonnation on the symposiumor the nity to develop an international andin­ Partners inProtection committee, please depth view of the problems that wildfire contact: Partners in Protection, Box 249, can present in the wildland-urban Onoway, Alberta,TOE IVa or call Ken interface. At least 250 representatives Saulit, fire chief for the County of from participating agenciesthroughout Parkland,Alberta,403-963-223\ or Canada andfrom the UnitedStatesand Kelly O'Shea, forestprotection officer others interested in fire subjects for the Bow/Crow Forest,403-297­ attended. 8800.•

24 Fire Management Notes Fire Mapping Using Airborne Global Positioning 1

Philip L. Drake

Regional land surveyor, USDA Forest Service, Northern Region, Missoula, MT

On September IS, 1990, Ray Evans, degradation of the GPS signals, use of The entire 35-mile perimeter of the Director of Fire and Aviation Manage­ multiple instruments improves positions fire, enclosing 14,200 acres, was ment, Rocky Mountain Region (Region from a 20-meter- (66-ft-) stated mapped in about 55 minutes offlight ,j 2), ordered the region's global position­ precision with one instrument to a 2­ time. ing units for use on the Swedlund Fire meter- (6.6-ft-) stated precision. The on the Black Hills National Forest. procedure of using a control GPS unit Blaine Cook, resource management in conjunction with remote or roving were noticed due to the main rotor; in forester, Black Hills National Forest; units is called "differential positioning." fact, 99 percent of the coordinate fixes Doug Williams, land surveyor, Black One ofthe GPS units was set on the were three-dimensional coordinates Hills National Forest; and Phil Drake, control station established at the Custer (latitude, longitude, and elevation). At regional land surveyor, formed a team Airport. The antenna ofanother GPS this project latitude and longitude, there to map the Swedlund Fire using unit was attached to a Jet Ranger are presently about 8 hours during a 24­ Trimble Navigation's Pathfinder, a helicopter. The Pathfinder's antenna hour period when three-dimensional navigation system and position-logging along with a foam backing was taped to coverage is available. This window is system based on the Global Positioning the tail boom of the helicopter, just continuing to expand as more of the System (GPS), a satellite network ahead of the horizontal stabilizer, and NavStar satellites are launched. It developed by the Department of about 12 feet (4 m) back from the cabin. should be noted that at least four Defense. Duct tape overlaid with filament satellites must be over 12 degrees strapping was wrapped around the tail above the horizon to receive three­ How the System Was Set Up boom to hold the antenna in place on dimensional coordinates. Two­ the boom. The entire length of the dimensional coordinates (latitude and To begin the mapping, the team first antenna cable outside the aircraft was longitude) can be received with three established a control station with taped to the aircraft skin.' Cook and available satellites. This extends the geodetic coordinates at the Custer Wi\\iarns acted as spotters while 1 window by about 4 hours; however, this Airport where it was convenient to operated the instruments. should be done cautiously. The error operate while mapping the fire. that the instrument is carrying in Presently, GPS units should be used in The Mapping Flights elevation while collecting data in a two­ multiple units with one positioned on a dimensional window will translate to at station with a known position. This is The mapping flight was flown during least that much error in the horizontal done primarily to resolve errors of up to a three-dimensional satellite window at position. This can beovercome by 100 meters (328 ft) imposed on the about 50 to 100 feet (15 to 31 m) above updating the elevation within the system by the Department of Defense. the treetops. At this height above the instrument with an altimeter, but it is These errors are imposed ostensibly as treetops, it was relatively easy to spot much more preferable to just plan the a security measure to hinder real-time the fireline. No reception problems data collection for three-dimensional positioning. Due to inherent errors windows. within the system, irrespective of the The entire 35-mile (56-km) perimeter Defense Department's deliberate 'Specific procedures for approval of mounting of the fire, enclosing 14,200 acres the GPS antenna have been outlined in a May 5, 1992. letter from the Director of Fire and (5,747 hal, was mapped in about 55 Aviation Management in the Washington Office minutes of flight time. The GPS I Philip Drake's and Douglas Luepke's ("Using to each Regional Forester. If the mounting instrument was configured to collect a the Global Positioning System in Firefighting on procedure described in this article is used, it position every one second, therefore the Shorts Fire in the Okefenokee Swamp" also must be submitted, with proper paperwork and approximately 3,000 coordinate published in this issue) use ofOPS was log book entries, to the Federal Aviation previously described in "GPS for Forest Fire Administration for approval. The FAA views positions were collected along the Management and Cleanup" in the September mounting the antenna with tape as a temporary fireline or about one position every 60 \991 issue ofGPS World. attachment procedure. feet (1.5 m). Within 2 hours oflanding,

1991 Volume 52, Number 3 25 the data had been postprocessed, using Forest, acted as the spotter to map the helicopter. However, there could be the controlstation data, and anoverlay high-intensity bum areas on the interior times when smoke would obscurethe plot at I:24000 scale ofthe entire ofthe fire. This was done to obtain the fireline visibility. At these times it may fireline, along with the acreage calcula­ acreages for reseedingandrehabilita­ be necessaryto carry out the operation tions, weredelivered to the Plans tion. Thirteen polygons ofhigh­ from a ground-based vehicle or by Section of the fire. intensity burn (5,300 acres or 2, 145 hal, walking the perimeter with the GPS The short turnaround time for the interior to the fire, were mapped in I unit. These methods would obviously finishedperimeter plot was impressive. hour and 20 minutes of flying time. As increasethe production time, and signal The GPS fire map nearly agreed with a polygon was closed with the aircraft reception breaks would be more the traditionally generated ground fire and GPS, an estimate was made of the common. But based on the experience map. However, because therewere percentage of intermingled green within we have had with other applications, some slight differences between the two the polygon, These estimates ranged this is a viable option. maps, the Plans Section sent the from 10 to 30 percent. The post­ An interesting sidelight to this airborne GPS fire mapping tearn out processing and -plotting took a couple experience: The helicopter pilot and the again to map the fire in the opposite of hours. Plots weremade at several mechanic were initiallyconcernedthat direction as a redundant check to verify scales for the rehabilitation, reseeding, the duct tape might damage the aircraft the accuracy of the GPS mapping and salvage sale plans as well as the paint, although this did not happen. The system. The second GPS fire map final fire reports. Trimble Navigation GPS units, in overlay matched almost exactly with Bill Ewin ofCEA Inc., Trimble addition to storingpositions for the first GPS map. The few differences Navigationmanufacturer's representa­ mapping, also provide real-time that occurred between the two flights tive fromTempe,AZ, arrived on the navigational information. Using the (+/- 50 ft or 15 m) arose, it seemed, foreston Monday as a volunteer. He GPS instrument, the pilot was navigated from the difficulty of flying an aircraft and Carol Thomas of Information to within 20 feet (6 m) of where he took over a sinuous line on the ground that Systems on the Black Hills National off and was given his trueground speed was difficult to distinguish in some Forest, converted the Pathfinder data to along the way. This impressed him to areas. The acreage calculations (which AutoCAD® (Computer Aided Drafting the extent that the GPS crew had a hard take about 2 minutes after the data is and Design) and MOSS (Map Overlay time gettingclose to the ship to remove processed) for the second flight was Statistical System) digital formats. The the instrument when the projectwas 14,100 acres (5,706 hal, less than a software that is supplied with the completed. l-percenr difference from the first. instruments makes this task relatively Althoughthere were some minor easy with some basic computer Other Applications and Advantages differences when the GPS fire map was knowledge. In this format, the polygons ofGPS compared with the traditionally (in digital form), mapped with the GPS generated ground fire map, it appeared instruments, will be used to update the The airborne application of GPS that the ground mappers had really done forestgeographical information systems technology for mapping or the updating an excellent job of photo and map (GIS) and resource inventory system of maps could be cost effective for interpretation. About one-third of the (RIS)3databases, silvicultural prescrip­ severalothertypes of projects wherea area of this fire was on land adminis­ tions, andotherfire-related manage­ largeareaneeds to be covered in a short tered by the State of South Dakota. A ment activities. amount of time and especially useful separate data file was created within the The entire GPS operation for this when time or visibility precludes GPS software for this area. Acreage was particular fire was completed from the traditional photogrammetric methods. then calculated for this area and will be Applications might include wildlife used to prorate the cost of suppression. counts and mapping, road and trail On the second day, Darwin Hoeft, "Rocky Mountain Regio~ timber standrecord inventory and mapping, wetland soil scientist on the Black Hills National system. acreagemeasurements, stream-length

26 Fire Management Notes surveys, timber surveys and mapping, Federal Excess Personal Subscription fire departments and FEPP, 51(1): 27 law enforcement activities, and rescue Property Information: operations, One of GPS's greatest assets Where to Find It in Civil Defense is its repeatability and the fact that it Fire Management Notes works on a common geographic base, Civil defense FEPP, 51(1): 32 ensuring spatial compatibility between Articles on Federal Excess Personal groups of data that are collected at Property (FEPP) written by Francis Russ, Inventory and Accountability different times and places. It is an property management specialist in the Fire understatement to say that this factor is and Aviation Management in the Washing­ Ten-percent rule for FEPP, 52(1): 36 important as we move further into the ton Office, and listed here, appeared in Fire Unrequested Federal Excess Personal electronic information age .• Management Notes during the past 2 1/2 Property, 50(2): 8 years. These articles focus on the basic Keeping Track of FEPP: internal control, requirements of the program-important 51(2): 6 i information for every FEPP user. If you need copies of any of the listed Items for Fire Use articles, please write Russ at the address on the inside from cover, call him at Acquisition guidelines for FEPP, 51(4): 13 (202) 205-089\ or DG:F.Russ:WO\c. You are also invited to submit articles on FEPP Management and Use use in fire protection. Every 2 years: FEPP inventory required, Acquisition 50(4): 41 Identifying Federal Excess Personal Acquisition guidelines for FEPP, 51(4): 13 Property, 50(1): 55 Cooperative agreements for the use of Keeping track of FEPP: internal control, FEPP, 51(2): 4 51(2):6 Olive-drab Federal property, 50(3): 33 Aircraft Subscription fire departments and FEPP. 51(1): 27 Forest Service aircraft on loan to State Ten-percent rule for FEPP, 52(1): 36 forestry agencies, 51(3): 22-24. Fonner military aircraft in fire protection, Marking FEPP 50(2): 28 Civil defense FEPP, 51(1): 32 Basics of the FEPP program Every 2 years: FEPP inventory required, ~ I Preserve the 50(4): 41 Cooperative agreements for the use of Olive-drab Federal property, 50(3): 33 wildlife. FEPP, 51 (2)4 Every year, more families are Forest Service aircraft on loan to State Francis R. Russ, property manage­ choosing to make their home closer forestry agencies, 5t(3): 22-24 ment specialist, USDA ForestService, to the forest. They're choosing to Fonner military aircraft in fire protection. Fire and Aviation Management, keep the home fires burning. Which 50(2): 28 Washington, DC, and chairmanofthe . they will. As long as y04 don't burn down their home. Remember. Only Identifying Federal Excess Personal FE?? Study Group you can prevent forest fires. Property, 50(1): 55

1991 Volume 52, Number 3 27 Using the Global Positioning System in Firefighting on the Shorts Fire in the Okefenokee Swamp1

Douglas Luepke

Cartographer, USDA Forest Service, Region 8, Atlanta, GA

On October 12, 1990, an overview of Flying the Perimeter and Collect­ (USGS) topographic 7.5-minute Trimble's Navigation's Pathfinder, a ing Data. As the flight began, the quadrangle map is accomplished by navigation and position-logging system Pathfinder was tumed on and began aligning the 2.5-minute tick marks based on Global Positioning System displaying latitude and longitude on the produced on the plot with the 2.5­ (GPS) capabilities, was presented to the polycorder. At the fire edge. the minute ticks on the USGS quad. fire management team in the regional Pathfinder was set in DATA LOGGING Locating Hot Spots and Plotting office of the Southem Region. The team mode to mark position data every Those Points on Fireline Map decided to mobilize the Pathfinder on second. The pilot flew slowly at treetop Overlay. The entire file was then the Shorts Fire, a fire in the Okefenokee level, following the sinuosity ofthe fire graphically displayed on a computer Swamp that had been buming for perimeter, while the Pathfinder logged a monitor. Using the Pathfinder software approximately 1 month. The fire was position per second. PFiNDER, the hot spots. as identified under control at this stage and mopup by the IR interpreter, were located in the was the principal activity-an file by matching the time tag of each excellent time to integrate GPS into the Although the activity ofthe tire was position with the times recorded by the operation. not intense, there was a real chance GPS operator. (As mentioned earlier. for this tire to take otT and run. the latitude and longitude of the hot Location of Hot Spots on Fire spots were recorded on a pad of paper Perimeter by the GPS operator.) The LOCATER As the helicopter slowly moved command allows the operator to move a Equipment and Personnel. The along the fire perimeter, the IR inter­ pointer around freely on the screen with GPS unit was first used on the fire on a preter identified a hot spot and the FMT a mouse, and as the pointer is moved, flight with an infrared (lR) detection operator pressed a computer function the latitude and longitude is displayed mission to locate hot spots along the fire key to mark or flag the location in the in the upper portion ofthe screen. The perimeter. The lR interpreter sat in the data file. At the same time, the GPS pointer was moved around until the back seat of the helicopter with a operator documented on a pad of paper exact latitude and longitude of each handheld forward-looking infrared the time displayed by the polycorder previously recorded hot spot appears on scanner (FURS) unit. The Pathfinder and, if time permitted, the latitude and the screen. As this was done, the pointer operator was in the copilot's seat with longitude. Recent hardware-software indicated on the screen where that hot the Pathfinder lying on the floor updates are now available for marking spot was located in relation to the fire between his feet and the polycorder or flagging hot spot locations in a file, perimeter. The hot spot point was then (recorder-display unit) in his lap. The making recording by hand unnecessary. visually transferred from the screen to antenna was fastened on top of the Downloading Data and Plotting the plotted I:24000 scale fireline map pilot's control panel with duct tape. In the Map. In 2 hours, using the methods overlay. With the hot spots located on the back seat ofthe helicopter with the described here, the east and south sides the overlay, it was detennined which IR interpreter was an operator ofthe of the fire had been flown and infonna­ spots were easily accessible by engine FIRE MOUSE TRAP, a system using tion collected on the fire perimeter. On and which required water bucket drops. the LORAN-C or GPS receiver to return to the camp, the data in the record a position every 2 seconds. polycorder were downloaded to a Taking Suppression Action laptop computer. The flight of the IR mission was then plotted using a During the detection flight, it was 'Douglas Luepke's and Philip Drake's ("Fire Hewlett Packard 7475 plotter at noted one hot spot was relatively small Mapping Using Airborne Global Positioning" I:24000 scale. The plot consisted of a and not smoking. It was near a firebreak also published in (his issue) use of GPS was recently constructed with bulldozers. previously described in "GPS for Forest Fire series of dots depicting points along the Management and Cleanup" in the September line of flight. Precise registration of the Considering the Pathfinder's navigating 1991 issue ofGPS World. plot with a U.S. Geological Survey capability, it was decided to try to locate

28 Fire Management Notes this spot with an engine crew using were collected as if this were a working sured, the calculation would likely have GPS. fire situation and information was been very close. First Step-s-Finding the Hot Spot. needed immediately. Due to antenna The first step in locating this hot spot location within the helicopter and The Gain, the Downside, the Future was to enter the recorded location into minimum satellite visibility, portions of the Pathfinder's waypoint (the hot spot) the perimeter had to be flown in This fire situation provided an file. The engine crew drove as close as different directions to get adequate excellent opportunity to introduce GPS they could to the spot navigating with satellite signals. The PFINDER to the firefighting procedures. Although the USGS quad map and the overlay software has the capability of doing the activity of the fire was not intense, plot. The Pathfinder was then used to area calculation if all data are flown in there was a real chance for this fire to navigate to the final location. With the the same direction. Flying portions of take off and run. The experience proved Pathfinder position at the engine the perimeter in opposite directions that GPS can be used effectively in a known, the NAVIGATE program was requires additional steps to calculate the helicopter. However, it was noted that executed. This program calculates the area. the location for the antenna within the azimuth and distance from where the Using the Pathfinder software, the helicopter is critical. In some cases the user is to the previously entered data acquired during the 3 days were loss of signal from one or more coordinates ofthe waypoint, or hot spot. merged into one file using the utility satellites created data gaps ranging from The polycorder displays the azimuth . program MULTISSF. This file was 50 to 200 feet (15 to 61 m). This was and distance from the current position graphically displayed on the computer not critical because the detail of the fire to the hot spot. A compass was used to monitor. All overlapping data from each perimeter was not lost. direct the crew toward the hot spot. As of the individual flights were then The process used to record hot spots, the engine crew approached the removed with the DELETE command. download the data, and locate these position, the distance displayed on the The final product was a single, clean spots on the map was a bit cumber­ polycorder continued to decrease. At file, depicting the fire perimeter of the some. Future versions of the software zero, indicating the location of the hot Shorts Fire. Next, the utility program will reportedly provide the capability to spot had been reached, there was no GISGENX was used to convert this tag points by pressing a function key visible evidence of a hot spot. This data into AutoCAD@ DXF format. similar to the function key used in the confirmed the observation during the An AutoCAD® drawing file was then FIRE MOUSE TRAP. detection flight that this spot was not created and the data collected by the One more valuable feature of the smoking. As the search for the hot spot GPS unit were added by using the GPS software is its ability to produce a narrowed, a pile of debris pushed up by AutoCAD@command, DFXIN. Using plot with 2.5-minute tick marks, which the dozers was noticed. Using the the POLYLINE command, a line was allows easy registration with the USGS "cold-touch" method, heat was detected drawn on a separate layer over the quads. This eliminates the need for coming from this pile. This was it-and existing GPS points. It took about 30 "rubber sheeting" (aligning or adjusting within to to 15 feet (3.1 to 4.6 m) of minutes to do this. Once this line was the map with known physical features) where the GPS unit had zeroed out. closed on itself, the AREA command the fire map to make it fit. Mission accomplished. was used to calculate the acreage. The The navigation ability of the GPS Calculating the Area ofthe Fire. total area measured was 20,079 acres unit can playa major role in Another task was to calculate the area (8,126 hal with a perimeter of 49.5 firefighting. The new, low-cost, of the fire. Because helicopter time was miles (80 km). This was less than the lightweight Pathfinder, Pathfinder in high demand, it took the next 3 days total fire acres of 20,779 (8,409 hal, Basic, now available seems to be an to complete flying over the remaining because the small bum areas outside the excellent tool to locate recorded hot portions of the fire perimeter. Although main bum on the southern perimeter spots. An individual in a spotter the flight, were made at times when were not included in the GPS measure­ helicopter can record hot spots and satellite visibility was not ideal, the data ments. If these areas had been mea- transfer the information by radio to a

1991 Volume 52, Number 3 29 ground crew for immediate response. Hallie Daggett: First Woman Hallie, "a wide-awake woman of 30 Imagine the spotter directing 10 or years," wrote McCarthy, "knows and as Forest Service Fire has traversed every trail in the Salmon more engine crews to hot spots and also Leokout' directing water drops all at the same River watershed, and is thoroughly familiar with every foot of the district." time using GPS. In 1913, the Klamath National Forest She supported the Forest Service and Several people suggested flying the Supervisor, W.B. Rider, had a big promised to stick with the job until she perimeter of the fire at a higher altitude. decision to make. The Eddy Gulch Fire was no longer needed. She was an At the higher altitude, a less-detailed Tower needed a new lookout, and excellent rifle marksman, rider, and but acceptable perimeter could be according to M.H. McCarthy, assistant trapper and was "not afraid of anything defined in less time and eliminate the fire ranger, in his letter to Forest that walks, creeps, or flies." additional steps and software used for Supervisor Rider, there were only three Hallie was hired at a salary of$840 a area calculations. However, the Shorts applicants to choose from: One had poor year and spent the next 15 years on the Fire experience provided an opportunity eyesight, one was "no gentleman," and job. Despite the rigors of such an to demonstrate another way to calculate the other was "also 'no gentleman' "but assignment, except for riding to and area and the flexibility of PFINDER in "has all the requisites of a first-class from the lookout at the start and end of Lookout." the fire season, Hallie continued to wear converting data into AutoCAD® format. With that rundown on the applicants, the popular ankle-length skirts and high­ One of the activities associated with assistant fire ranger McCarthy recom­ necked blouses fashionable in her day. the fire was the consideration of an mended Hallie Daggett as the best But "... it was a rare experience," existing facility for a combination qualified for this important job. As he writes Rosemary Holsinger in her incident base and incident command did so, he hoped the supervisor's "heart article, "A Novel Experiment: Hallie post. The Pathfinder was used to create was strong enough to stand the shock" Comes to Eddy's Gulch," "to catch her a quick, as-built plan of the existing of having a woman nominated since an without a revolver strapped to her belt," features including roads, parking areas, appointment of that sort was unheard of a weapon she readily used. During the and existing supply and motor pool at the time. fire seasonof 1915 for example, she locations. These data were collected, killed one bear, four wildcats, and three 'Adapted from a handout in the Museum in the downloaded, and a I-inch-equals-40­ coyotes. Schools Program jointly sponsored by the Why did she choose to do such feet (O.3-m = 12.9-m) plot provided to Siskiyou County Museum and the Klamath difficult work? As she said, "'Ilove it! the planning team in less than 2 National Forest 1991 Program. Rosemary And that's why I'm here:' • hours .• Holsinger's article on Hallie Daggett was published in Women in Forestry; 1983: 5(2) 21-25.

30 Fire Management Notes Clark County Goes near homes, electrical utilities, building new glow-in-the-dark signs for their Face-To-Face With construction, water supply, distance home address. They also handed out from a fire station, landscaping, outdoor recycling information. Wildland-Urban Interface burning, and firewood storage. Only three times were Sork and Sork and Engel visited with Engel asked to leave, but the residents The Problem and the Plan homeowners at their homes. They still took the information to rate their explained why they were there and own homes. Eight hundred homes were One cool, damp spring night in 1990, asked if they could walk around the visited in a 2-rnonth period. When the facing another fire season in Clark house to do a risk assessment. They horne visits were completed, they County, WA, one of fastest growing gave the homeowners a copy of outdoor followed up with a telephone survey, counties in the Nation, Fire Chiefs Steve burning regulations, a home protection calling 400 people. The results indicated Wrightson, Pat Humphries, and Tom guide, fire resistant Landscaping that 75 percent of the people surveyed McDowell and Fire Commissioners information, and the opportunity to get by telephone had made some type of Dave Campbell, Roy Bellcoff, and Fred change to lessen their risk. Pickering developed a plan to manage Homeowners made the following the increasing wildland fire interface changes: problem in Clark County. Forest lands • Firewood moved away from ofvarious ages and slash, combined buildings' with new homes, cause great concern • Lower limbs of trees removed for fire protection agencies. • Sprinkler systems installed The fire chiefs and commissioners • Retardant applied to shake roofs submitted an application to the • Shake roofs replaced with a fire Washington State Department of Natural resistant roofing material Resources for grant money from • Many glow-in-the-dark house Washington's Fire District Support numbering signs put up Program to finance visits to homeowners to assess risk and offer Benefits to the Fire District safety information. The Washington State legislature appropriated $200,000 The fire.district benefited in several in 1990 to be distributed to fire districts significant ways: that support the Department of Natural • The community became aware of the Resources in wildland fire protection i danger of fire and suppression. Clark County was • Homeowners took action-s-changed granted $18,000 to carry out its their behavior-e-tc reduce fire hazard proposal. I and enhance safety. I • Goodwill was built with the The Project homeowners-impressed that the fire district cared enough to come to their Two people, Eric Sork and Linda I homes. People who were not home Engel, were hired for 8 weeks to ! during the home visits called the fire conduct a door-to-door risk assessment I district to ask if they could be a part on each home in wildland-urban of the program.• interface areas with a high risk of fire. I The risk assessment considered road Lane L. Jolly, fire servicespecialist, access, topography-e-especially with Washington Departmentof Natural features increasing fire danger-fuels Resources, Olympia, WA

1991 Volume 52, Number 3 31 The Pioneers (Some of Them) and Their Equipment (a Little of It) in Forest · .. basic theory of fire mapping and Service Infrared Fire Mapping detection developed and Detection Research · .. a remarkable achievement and Operations · .. basic tenets still apply

From 1962-1976, through the research conducted by Project FIRES CAN under project leader Stanley N. Hirsch, electronics engineer, and physicist Ralph Wilson, electron­ ics engineer Forrest Madden, and electronics technician Dale Gable at the Intermountain Forest and Range Experi­ ment Station, the basic theory of fire mapping and detection with thermal infrared systems was developed. Under the leadership of Bob Bjornsen, as head of a separate group to develop procedures for fire mapping and as Forest Service director at Boise Interagency Fire Center, infrared mapping and detection was established as a standard part of fire operations. J.S. Barrows in the 1950's pointed out, after studying fires in the western zone of the Northern Region, the high number of fires escaping and acreage burned and the need for improved detection methods. Barrows envisioned a detection system that could efficiently cover areas where fires were likely and locate those fires while they were small. The system he foresaw would, in his words "detect Stan Butrym. pilot, Robert Miller, project the greatest number of fires in the fastest time at the lowest engineer, Captain D. Petersen, project officer, members ofthe U.s. Army Mohawk team, and cost." Robert L. Bjomsen, project head ofthe Many people contributed to our understanding of the Northern Forest Fire Laboratory forest fire operation and the usefulness of infrared forest fire detec­ mapping program, with aircraft prop jet ship tion. Some of the researchers, besides those already equipped with infrared sensing and recording mentioned, are P.H. Kourtz, Nonan Neste, Robert F. equipment. The U.S. Army team, temporarily assigned to the laboratory secured the image of Kruckeberg, and B. John Losensky; some of the technicians the forest fire, located it, and then fire and pilots, Robert A. Cook, John Voth, Ted Bishop, Eddie operationspersonnel mapped the fire. {photo: Downs, Bill Hodson, Nels Jensen, and John Holsman. Harold Pontecarvo, 1963)

Special thanks to Roberta Hartford of the Intermountain Fire Sciences Laboratory, Missoula, MT,for her help in tracking down photographs for this story. Thanks also to Richard Rothermel, Francis Russ, Brenda Breslin, John Warren, Ralph Wilson, Charles George, and Dale Gable for their efforts.

Photourapns: Courtesy of USDA Forest Service Fire surveillance Project aircraft crew, (left to right) Bill Hodson, copilot; Ted Bishoff, mechanic; and Intermountain Fire Sciences Laboratory in John Holsman. pilot, with Convair T-29. {Photo: Herman Wittman, 1968) MissoUla, MT.

32 Fire Management Notes "

I Ralph Wilson, physicist with the F1RESCAN Forrest Madden, F/RESCAN electronics engineer, points oul the small infrared scanner attached to group, examines a frame designed to hold the the light fixed-wing aircraft, which was developed hy personnel at the Northern Forest Fire Labora­ HRB Singer MS-5 scanner. tory (now known as the Intermountain Fire Sciences Laboratory. Missoula, MT) to detect small spot fires in small areas so the ground crew could locate exact fire position. (Photo: HermanWittman, 1968)

I i' ,, I

Stanley N. Hirsch, project leader ofFIRESCAN ofNorthern Forest Fire Laboratory, and Stan Butryn, pilot, with U.S. Army's Mohawk aircraft, cooperatively used with the Northern Forest Fire Laboratory's FIRESCAN programs. (Photo: Harold Pontecarvo, 1963)

1991 Volume 52, Number 3 33 John Voth, FlRESCAN infrared electronics technician, using the photographic Interior ofAero-Commander aircraft showing control panelfor film processor to record infrared imagery during test runs to detect small fires. infrared imagery detection equipment used in fire mapping (Photo: Vance Price, 1967) operations, (Photo: Herman Wittman, 1966)

Infrared electronic technicians.John Voth (left) and Dale Gahle (right) us~ng "". electronic infra~ed sensor within aircraft to locate smallforest fires, record the imagery, and Identify a ground location with the interpreter's map. (Photo: Herman Wittman, 1971)

34 Fire Management Notes Global Positioning System: Uses in Fire fO~E5T SEilVict Management on the Clearwater National Forest U~S Byron J. Bonney ~-IJIrENTOfAG~

Fire staffofficer, USDA Forest Service, Clearwater National Forest, Orofino, ID

The Clearwater National Forest had forest and using a 1/2 inch-to-the-mile­ in two or three dimensions. When " the opportunity in 1991 to test the (1.3 cm-to-1.6-km-) forest map for receiving in two dimensions, the fixed Trimble Navigation's Pathfinder, a plotting fire location. On many occa­ elevation at which the aircraft is flying navigation system and position­ sions, the lookouts would turn in a must be programmed into the GPS and logging system based on the Global legal description, and the aerial detec­ the aircraft needs to stay at that eleva­ Positioning System (GPS), in locating tion aircraft would give a different tion to obtain an accurate location. In initial attack fires on normal aerial fire legal description. The initial attack three dimensions, a fixed elevation is detection flights and in the mapping of crew would then refine the location not needed, as the elevation remains prescribed natural fires within the after finding the fire on the ground. variable within the program. There are Selway-Bitterroot Wilderness. Much time, energy, and money have also only certain times during the day Randy Doman, fuels specialist on been wasted in trying to obtain accu­ when the GPS will be able to receive the Clearwater National Forest, read rate legal descriptions to locate these in either two or three dimensions. A about the capabilities ofthe GPS in fires. timetable listing available satellite Engineering Field Notes. These capa­ access times is published daily. The bilities were brought to the attention of TheGPSTest cadastral engineer on the Clearwater the Clearwater fire coordinator, Chuck National Forest helped the forest ob­ The Equipment and its Mounting Petersen, who elected to try GPS in the tain these timetables. in the Aircraft. Wanting to raise the aerial fire detection aircraft. The level of accuracy in locating fires, fire Clearwater Engineering Cadastral From our estimates, the cost of managers decided to test the GPS in Survey staff had recently purchased GPS can be recovered after 2 to the aerial detection aircraft to see if it two Pathfinders and was gracious 3 years of operation and, de­ could improve performance. The sys­ enough to allow the Fire Management pending on the circumstances, tem-s-consisting of an antenna, a Systems staff to borrow one ofthese perhaps even after one fire. polycorder, and a battery pack-does units to test its application in fire de­ not take up much room in the aircraft. tection and mapping. Procedure and Use. The aerial The antenna was mounted on the dash observer detects a fire. The aircraft of a Cessna 185. It was determined Aerial Fire Detection then flies on a straight north-south line later that to access the satellites prop­ directly over the fire. Before flying The Clearwater National Fire Man­ erly, the antenna should be mounted over the fire, the polycorder is pro­ agement team saw the need to improve on top of the wing of the aircraft. The grammed to record location by latitude the aerial fire detection program. The problem with mounting on the dash of and longitude. As the aircraft flies over fire managers had a problem obtaining the aircraft is that the fuselage would the fire at a certain elevation, the aerial accurate legal descriptions on new block out some ofthe satellites and the observer presses the ENTER key on fires from aerial observers and fixed system would not respond with a cor­ I the polycorder. The latitude and longi­ ? lookouts. The LORAN-C in the aerial rect location. tude recorded are accurate within plus detection aircraft had not been the Satellite Use. The system can be or minus 10 meters (33 ft) of that fire. answer to this problem. as locations operated in two dimensions, using I This information would be transmitted " plotted by this instrument have been three satellites, or in three dimensions, to the district, where it would be plot­ up to 3 miles (5 km) off. At best, using four satellites. The system can be ted on a 7.5-minute U.S. Geological LORAN-C would plot the locations checked, as the aircraft is flying, to Survey topographic map and given to within 1/2 mile (0.8 km). Due to this ensure the proper number of satellites the initial attack crew. GPS was used inaccuracy, the forest had been con­ are being used for the operation under­ about 15 to 20 times in 1991 on flights tinuing with locating fires using the way. One mode within the program to locate initial attack fires. aerial observer's knowledge ofthe will tell you whether you are receiving

1991 Volume 52, Number 3 35 Under normal circumstances, with­ within the three-dimensional satellite system has the potential for many out GPS, the aerial detection aircraft window timeframe. An easily recog­ other applications on large fires. Inci­ might make three to six passes around nizable starting point was identified, dent management teams could utilize the fire to obtain accurate information and the GPS was programmed before the system in many different ways: on the fire location. In some cases reaching that point. As soon as the • Plot accurate fire perimeters daily. these locations might be offas much helicopter passed over that point, the • Locate certain points within or as I mile (1.6 km) from where the fire ENTER key was pressed and the GPS around the fire for rescue, helispots, is actually located. It is believed from began recording points along the path critical areas, and so on. the Clearwater experience that this ofthe helicopter at a rate ofone point • Gain accurate locations of hot spots technology will not only save precious per second. Different time intervals for when used in conjunction with time and money during detection but recording points can be programmed infrared imagery. also time and money for initial attack into the GPS. • Plot location of any new fire starts crews in locating fires on the ground. After flying around the fire perim­ or spotfires within the area of re­ eter and reaching the starting point, the sponsibility. Use in Prescribed Natural Fire GPS is then programmed to "log off." The Clearwater National Forest has If possible the fire perimeter should be completed its GPS test and is con­ The GPS was also used to map the flown with GPS another time in the vinced that this system needs to be perimeters of several prescribed natu­ opposite direction to obtain a double used as an operational tool for the ral fires within the Selway-Bitterroot check on the first perimeter and acre­ 1992 fire season. The system costs Wilderness. For these fires, the GPS age. approximately $8,400. Estimates indi­ was used from a helicopter, Bell Jet The perimeters of both the Two cate that this cost could be recovered Ranger 206C, on which foam-backed Mile and Blacktail prescribed natural after 2 to 3 years ofoperation and, antenna was mounted to the nose of fires were flown. In order to save flight depending on the circumstances, per­ the helicopter on the battery door with time, only one flight was used around haps even after one fire. The forest is duct tape, reinforced with filament each fire. The final maps were printed sold on GPS and would strongly rec­ tape.' The cable was then taped under and plotted on a 7.S-minute USGS ommend that other fire management the nose and sent up through the drain map. These maps would aid fire man­ teams consider this system. plug in the bottom of the bubble on the agers with interpretation of fire growth Ifyou have experiences to share or passenger side of the ship, for easy on these fires later in the season if any questions about the Clearwater access to the operator. these fires continued to be active. The National Forest's use ofthis system, The mapping flight was flown at SO actual fire perimeter has been difficult please give me a call at (208) 476­ to 100 feet (IS to 30.5 m) above the to obtain from fixed-wing observation 4541. • tree tops. The GPS was operated flights. The GPS answered many ques­ tions regarding fire perimeter location 'Specific procedures for approval of mounting on these two fires. the GPS antenna have been outlined in a May 5, 1992, letter from the Director of Fire and Did GPS Work for the Clearwater? Aviation Management in the Washington Office to each Regional Forester. If the mounting GPS has many uses in fire manage­ procedure described in this article is used, it ment. Systems such as the Pathfinder must be submitted, with proper paperwork and can save time and money and lead to log book entries, to the Federal Aviation our being more confident in our infor­ Administration for approval. The FAA views mounting the antenna with tape as a temporary mation on fire location-s-the start, hot attachmenl procedure. spots, and perimeter. Furthermore, this

, 36 Fire Management Notes " FIRE MOUSE TRAP Use in the Southern Region

James P. Scott Forest land surveyor, USDA Forest Service, Francis Marion-Sumter National Forests, Columbia, SC

The Startup The latitude and longitude points precisely where it belonged. This calculated by the LORAN-C are stored verification gave users confidence in the The FIRE MOUSE TRAP (FMT) in the laptop computer at selected system's ability to locate fireline in mapping system was introduced to the intervals, usually at 2 seconds, while the areas with little mapping definition. Southern Region of the USDA Forest helicopter flies the perimeter of the area Okefenokee National Wildlife Service in the spring of 1989, when to be mapped. Ifthe mapping is to be Refuge-the Shorts Fire. The system John Warren. electronics engineer, and done on foot or by all-terrain or other was used twice a day to map the Shorts Dale Gable, electronics technician, at vehicle, the recording interval can be Fire on the Okefenokee National Boise Interagency Fire Center (BIFe) set according to the speed over ground. Wildlife Refuge in south Georgia. Even gave a thorough 3-day training session When the project is completed, the though the quadrangle maps lacked in the operation and maintenance of the laptop computer is connected to the identifiable topographic features, the system in Columbia, Sc. The original plotter and the plotter scaled to the map FMT system successfully mapped the class consisted ofeight students: three or transparencies of the area covered by fireline. At one point on the Shorts Fire, from the Francis Marion-Sumter the LORAN-C traverse. The laptop will two incident bases needed good fire National Forests, two from the National then drive the plotter pen to plot to scale location maps. The decision was made Forests in Florida, one from the on the map the perimeter, hot spots, or to cut the mylar plot into eight inchwide National Forests in North Carolina, and other spots of interest. (2.5-cm) strips and transmit the plot to two from the Chattahoochee-Oconee the second base over a FAX machine. National Forests in Georgia. A second The results were satisfactory and training class was held in the spring of delivery time was immediate. During 1991 at the Witherbee Ranger District, The first use of the FIRE MOUSE the fire, more than 60 FMT maps were Francis Marion National Forest, where TRAP system in the Southern Region plotted, providing an excellent record of five more Southern Region Forest was on the Pinhook Fire on the fire growth. Service and four U.S. Department of Osceola National Forest. It was an The FMT system was used to fly the the Interior personnel (three U.S. Fish instant success. fire perimeter, mapping the numerous and Wildlife and one National Park helispots and dip tank sites. We then Service) received training. A third produced a perimeter map that located training class was held at Tallahassee, each helispot and dip tank (along with FL, during 1anuary 1992 where seven First Uses in the Southern Region latitude and longitude) on the fire more Forest Service personnel were perimeter, identifying each by name and trained. Osceola National Forest-Pinhook number. These maps were a great Fire. The first use of the FMT system in service to helicopter pilots who would System Components and Their the Southern Region was on the then load this data into their LORAN­ Functions Pinhook Fire on the Osceola National C's and fly directly to the location. At Forest. It was an instant success. The one point, several large areas of peat The Southern Region's FMT system fire was mostly in a low, swampy area bog were burning underground. The contains anARNAV R-15 LORAN-C with undefinable mapping features. refuge requested a map be made navigation receiver with an RS--232C Twenty minutes after the fire perimeter showing the location of the under­ output cable, a Hewlett-Packard flight, a fire map mylar overlay showing ground fire. A handheld infrared Portable Plus laptop computer and a the fire boundaries was in the hands of scanner was used to detect the under­ Houston Instruments DMP-56 plotter the plans chief. The accuracy ofthe ground burning with the infrared capable of being used on a full-size 7.5­ information was outstanding. Where the operator directing the helicopter pilot minute topographic quadrangle map, a topographic features were identifiable around the bogs while the FMT map specially created for the Forest on the 7.5-minute quadrangle, wewere recorded the line of flight. The final Service by the U.S. Geological Survey. able to verify that FMT put the fireline product was a map of the underground

1991 Volume 52, Number 3 37 fires related directly to the fire perimeter with the system. The first problem: The directly under the wires, but returned to map. effect of a high-powered military radar normal once we were 100 feet (30.1 m) station on fire mapping. While mapping from power lines. Other Fire Projects and Uses a fire in the Everglades National Park The Southern Region expects to use close to Homestead Air Force Base, the the FMT system for more activities in The Southem Region's FMT system base aircraft radar making a sweep in the future and awaits the arrival ofthe has now been used on six project fires the direction of the helicopter, destroyed new global positioning system (GPS) within the region and one project fire in the LORAN plot at that instant. The once the satellites needed for GPS the Pacific Southwest Region. We have second problem: The effect of high use are in place for 24-hour three­ also used it to locate bug infestations, voltage power lines on a ground dimensional coverage throughout the hiking and motorcycle trails, and forest mapping project. The plot was dis­ United States.• roads. rupted when the LORAN antennae was We have found that the system is a great tool when used with a handheld infrared scanner in mopping up a fire. Max Planck, Infrared, and Quantum Mechanics One person operates the infrared scanner, while a second operates the Max Planck, in 1900, working with Wein's Displacement Law and the Rayleigh­ FMT system. The infrared operator Jeans Law (both partially explaining infrared (IR) radiation), concluded that the laws spots a hot spot (many of which are not ofclassical physics were inadequate to describe the process taking place in the visible to the human eye) and calls out thermal IR bands at the atomic level. Planck agreed that oscillations could be "Hot spot!" to the mousetrap operator, allowed for all possible frequencies but introduced the idea that the amplitudes, and who then types a function key on the hence the energy, could increase only in discrete steps, differing by the quantity hv, computer, recording the geodetic now called a quantum of energy. Planck called h the quantum of action andlater position of the hot spot in the computer. experiments showed that it is a universal constant, now called Planck's constant. In less than 2 months, Planck developed the concept of quantum physics, nOW known After the retum to the helibase, a plot is as quantum mechanics, the basis for a whole area of modem physics. made on mylar from the system with the hot spots plotted in their true Integrating Planck's Law location on the fireline. This accurate plot becomes a valuable tool to the handcrews and helicopter water bucket operation, saving handcrews a lot of W :;: spectral radiant emittance, WCM-2~-1 time "cold trailing" along a fireline. We A :;: wavelength, ~ 34 have walked the fireline with the h =' Planck's constant e 6.6 X 10- , W sec' handheld infrared and the FMT map T =' absolute temp, K 8 arid found every hot spot located from C ::::: velocity of light, 3 X 10 m sec-' the air. This accurate information about K :;: Boltzman's constant 1.4 x lo-23Wsec K-' hot spots makes the handcrew's day a over wavelengths fromzero to infinity gives an expression for the radiant emittance little easier and gives the crew confi­ in a hemisphere above a blackbody I em" and is known as Stefan-Boltzrnan's law. dence in their work. Differentiating Planck's Law and solving for the maximum gives Wein's Displace­ ment Law. Stefan-Boltzman's Law describes. the area under the curve; Wein's Law, A Few Problems the position on the wavelength scale; and Planck's equation, the shape of the curve .• In its 2 years of use, the Southern John R. Warren, group leader, Advanced Electronics, USDA Forest Service, Boise Region has encountered a few problems Interagency Fire Center, Boise, ID

38 Fire Management Notes Evaluating The Hammer"

Brian Hutchins

Unit leader and mechanical engineer, Michigan Department ofNatural Resources, DN:i Michigan Forest Fire Experiment Station, Roscommon, MI

Need for Heavier Off-Road Vehicle In a national engine study, the cially. The MDNR purchase was one National Wildfire Coordinating step in this process. In June 1991, AM Group (NWCG) Fire Equipment General announced that it will eventu­ Like many wildfire agencies, the Working Team, surveying wildland ally sell the vehicle to the general Michigan Department of Natural engine user needs, documented the public. Resources (MDNR) has been searching disparity between those needs and for good vehicles for the rigors of off­ current truck design. Differences Between the Commercial road forest fire use. Recent truck design and Military HUMMER® has moved to meet the needs of the highway user. The newer vehicles determine the usefulness of the vehicle The commercial HUMMER@ has exhibit lighter frames, reduced ground before its availability through Federal several significant differences from the clearance, and larger profiles­ Excess Personal Property (FEPP), and military version. For instance, the characteristics that make the vehicle to develop a design concept for commercial vehicle has greater load less useful for wildland fire use. In a agencies interested in utilizing the capacity (a GVW of 10,300 Ib or 4,674 national engine study, the National vehicle. kg), "civilian" style seats, a 12-volt Wildfire Coordinating Group (NWCG) Up to now, the HUMMER" has 21 electrical system, and a hard cab top in Fire Equipment Working Team, military versions. These include troop, 2- and 4-person cab styles. Some of surveying wildland engine user needs, shelter and armament carriers, ambu­ these changes were made to meet documented the disparity between those lances, and light artillery prime movers. Federal safety and environmental needs and current truck design. Gross vehicle weight (GVW) ofthese standards. The MDNR unit is a From NWCG's survey, a set of vehicles ranges from 7,700 to 9,400 prototype ofthe commercial version, vehicle standards criteria was pounds (3,493 to 4,264 kg). In recent which includes upgraded load capacity, developed. The U.S. military's years, AM General began investigating seating, and safety items. multipurpose-wheeled vehicle marketing the HUMMER® commer- (HMMWV), produced by the AM General Corporation, meets many of these criteria for one size class. The "HUMMER"" as this vehicle is called is best known for its recent success in the Persian Gulf War.

Testing the HUMMER"

In January 1989, the MDNR began negotiating with AM General to purchase one vehicle for evaluation. In July 1990, the MDNR took delivery of " the first HUMMER· obtained through a nonmilitary procurement. Engineers and staff at the Michigan Forest Fire Experiment Station developed a fire package for initial testing of the vehicle. Project goals are to evaluate the effectiveness ofthe vehicle compared Michigan Department ofNatural Resources-AM General Corporation HUMMERoZl prototype with other wildfire control vehicles, with 300·gallon (I.J 36-L) engine.

1991 Volume 52, Number 3 39 Publishing the Field Test Results and grill protection, and an assessment of Designs for FEPP Users water capacities for the various military models. Potential commercial purchas­ The prototype vehicle is currently in ers will get an assessment of the field test. Information will be published HEAVY HUMMER®'s performance in in about 1 year in Roscommon Equip­ wildfire conditions. See tables I and 2 ment Center Report No. 56. For FEPP for information on some of the features users, the report will include a cab ofMDNR's HEAVY HUMMER® design for retrofitting soft cabs with a prototype.• hard version, improved bumper and

The rear view ofthe HUMMER®, showing the 5-gaflon (/9-L) foam tank, right hose reel, and Table i-Michigan Department of Narural Remurces-HUMMER® prototype unit features storage areas. Features from AM General Features from Michigan DNR HEAVY HUMMER® capacity Two-person hard cab (10,300 Ib or 4,672 kg GVWR) 300-gallon (1,136-L). 3-point suspended water tank Central tire inflation KK Products foam proportloner with 5-gallon (19-L) concentrate tank Radial tires Extended trent bumper and grill guard 12.000-pound (5,443 kg) electric Hose storage tray winch Cab doors Shovel and axe holder Automatic transmission Right and left hose reels with high-low and 4-wheel drive Detroit diesel 6.2 L engine Wajax·Pacific BB-4 pump fender nozzles and trigger-type cab gun

Table 2-Vehide weight and dimensional information ofMichigan Department ofNatural Leftside view ofthe HUMMER@withhydraulic Resources-HUMMER@prototype pump and left hose reel located in the rear occupant area. Specifications Data

Weight (300 gal nominal capacity) Front 3,880 Ib (1.760 kg) Rear 6.080 Ib (2,758 kg) Total 9,980 Ib (4,518 kg) Weight rating Front axle (FAWR) 4,100 Ib (1 ,861 kg) Rear axle (RAWR) 6,800 Ib (3,085 kg) Total 10,300 Ib (4.672 kg) Vertical center of gravity 35,3 inches (0.9 m) above ground Height(ground to top ot cab) 72 inches (1.8 m) Width 85 inches (2.2 m) Wheelbase 130 inches (3.9 m)

40 Fire Management Notes Monthly Fire Weather Forecasts

Morris H. McCutchan, Bernard N. Meisner, and Francis M. Fujioka and John W. Benoit and Benjamin Ly

USDA Forest Service. respectively, research meteorologists and computer programmers, Pacific Southwest Research Station, Riverside, CA

Fire Management Planning and the National Weather Service Extended Northern Hemisphere at an altitude Fire Weather Forecast Forecasts ofapproximately 10,000 feet (3,048 m) (where the air pressure is An important objective of fire Medium-Range Forecasts. The 700 millibars). The airflow at that management planning is to identify the NWS's medium-range (6- to lO-day) altitude reflects pattems in both the resources and activities needed for an forecasts produce daily data of the type overlying jet stream and the expected fire danger period. Improved used in the National Fire-Danger Rating temperature of the underlying air. fire danger prediction capabilities, in System (NFDRS) (Deeming and others • Then, maps of expected average which fire weather forecasting plays a 1977). The core ofthese medium-range temperature and precipitation totals vital part, can greatly aid fire managers weather forecasts is the output from are generated from equations, based in their planning. on the expected 700 millibar airflow Fire danger is predicted with the aid and persistence (the departures from of fire weather forecasts, typically Monthly fire weather forecasts may normal observed in the previous covering I to 3 days. Long-term fire not be the fire planner's magic month are expected to continue in planning has traditionally been based on bullets, but they do provide the current month). fire climatology and historical experi­ scientifically based long-range • The forecaster adjusts the tempera­ ence, since long-range fire weather forecasts. ture and precipitation outlooks and forecasts were not available. Although produces maps showing the prob­ the National Weather Service (NWS) ability that the temperature and routinely produces monthly and computer models of the atmosphere, precipitation will be either above or seasonal outlooks of average tempera­ using information gathered through a below normal during the month. ture and precipitation, those forecasts worldwide network of observations. Seasonal Outlooks. Prepared by were not designed to represent tire These models generate forecasts of CAC once each calendar month, the weather. future weather elements by applying seasonal outlooks are based solely on Now a system has been developed physical laws to the current state ofthe statistical techniques, since the accuracy specifically to forecast monthly fire atmosphere. The products that are ofcomputer model forecasts deterio­ weather. This monthly forecast differs typically produced from these forecasts rates rapidly over time. A group of fundamentally from the traditional daily and disseminated are manually modi­ forecasters prepares a consensus 700 fire weather forecast, which identifies fied maps that show much above, millibar airflow chart, considering the the trends and the expected worst above, near normal, below, and much following factors: Correlations in the condition in a 24-hour period. Rather, below average temperature and airflow from one season to the next; the monthly forecast represents the precipitation expected for the forecast correlations (teleconnection) between average fire weather over the month. It period. These categories are determined the airflow at one location and that at a does not fit the requirements ofthe by comparing the expected temperature distant site; extent of snow cover; sea National Fire-Danger Rating System. and precipitation with climatological surface temperature patterns, such as This paper describes the NWS norms. those associated with El Nino and La forecasts and the forecasting process Monthly Outlooks. The NWS Nina; and persistence. As with the briefly, examines the monthly fire Climate Analysis Center (CAe) issues monthly outlooks, a map of expected weather forecast in detail, and explains 30-day average weather outlooks at the temperatures consistent with the airflow how the monthly fire weather forecast beginning and middle of each month. chart and persistence is generated using may be accessed. Current limitations to These outlooks are prepared using both a set ofequations. Because seasonal the monthly forecasts are presented, computer model output and statistical precipitation pattems show little together with suggestions for dealing techniques: persistence, an objective routine with the limitations in decisionmaking. • The forecaster first prepares a chart identifies the 5 previous years with of the expected airflow over the temperature patterns most similar to

1991 Volume 52, Number 3 41 those forecast for the season of interest. The precipitation amounts that were observed during those years are used as .GTF 'N' ·"'50 guidance for the seasonal precipitation '"• FAR· 0 outlook. As with the monthly outlooks. '" SHR· 'l" the final products are categorical LN,EJ ~PR FSD

probability maps oftemperature and CYS. 0 'BF 0 '" 0 precipitation. sec ~n L?E~ .GLD ALS The Monthly Fire Weather Forecast 0

A~a The monthly fire weather forecasts myprepared by the USDA Forest Service at the Pacific Southwest ELP' SJTo Research Station 's Forest Fire Labora­ 'AT0 TE'A tory in Riverside, CA. Unlike the daily MeA weather forecast for NFDRS, which B~O describes the expected worst-case local conditions for a particular day, the F!g~re l-o--Lof-'afions ofNational Weather Service stationsfor which average afternoon (1300 !-ST) monthly fire weather forecast predicts surface temperature, dew point temperature, and windspeed are forecast: averggeconditions over a broad area for a month. The monthly fire weather forecast characterizes the weather-induced fire potential using equations developed for the contiguous United States and Alaska (Klein and Whistler 1991). These equations relate the expected 700 millibar airflow and persistence to 30­ day forecast percentiles ofthe following variables: Afternoon (l300 LST) surface temperature, relative humidity, Chandler Burning Index (CBI), and windspeed at 127 NWS stations (fig. I), and the precipitation frequency and amount over 60 climate areas (fig. 2) (Englehart and Douglas 1985). Our CBI is a modified version ofthe burning

index developed by Chandler and Figure 2--Locations ofclimate areas for whicn precipitationfrequency and amount are forecast. others (1983) and is given by: Approximately 15 reporting stations are in each climate area. Areas are numbered in alphabetical order hy state,

42 Fire Management Notes CBI = (110- 1.373(IU-!J)­ Figure 3---Maps plottingforecast percentilesfor June 1991: (a) average afternoon temperature, (b) 054(10.2 - T)) average afternoon modified Chandler B~r~jng Index. (c) overage afternoon windspeed, and (4) predpitationfrequen~y. (Note: For precipitation frequency. higher percentiles represent lower (124 x lO-ilO 142(RH))/60 (I) frequencies.)p where ~H = monthly mean afternoon relative humidity (percent) and

T = monthly mean afternoon surface temperature (0C)

The CBI is a weather-driven fire index that is highly correlated with fire activity on five offourteen nearby national forests examined ina previous study (McCutchan ~d Main' 1989). The forecasts are made in terms of percentilesbecause percentiles express how frequently the forecast value' actually occurred in the climatic data forany given location.Percentiles also translate readily into severity levels o so 75 90 97.5 100% which can be tied to appropriate preparedness actions (Chase 1991; Figure 3a-Average afternoon ~emperature. Mees and Chase 1991). The percentiles are determined completely by climatology. For example, a 90th percentile ofCBI would equal or exceed 90 percent of the CBI values in the climatological record. To have the high percentages signify high fire potential, the percentiles of precipitation frequency, precipitation amount andrelativehumidityare reversed. since low values of these parameters contribute to high fire danger. In other words, higher percen­ tiles represent lowerprecipitation frequencies.Thus, in ourforecasts. the 90th percentile of precipitation amount would be less than 90 percent ofthe amounts in the climatological record. o 50 75 90 97.5 100% Maps offorecast percentiles of Figure 3b-Average afternoon modified Chandler Burning Index. temperature, relative humidity, CBI, windspeed, and precipitation frequency are prepared. Examples for June 1991

1991 Volume 52, Number 3 43 are shown in figure 3. Thetemperature map (fig. 3a) shows one large area (extending from northern Arkansas to the eastern Great Lakes) and two small areas (in northern Minnesota and along the east Texas and Louisiana coast) that were forecast to have extremely high percentiles (~7.5 percent). A large area forecast to be cooler than normal «50 percent) covered the Western United States. The forecast values of the CBI (fig. 3b) were higher than normal, particularly for the eastern third ofthe country. The map depicts several centers ofextremely high percentiles extending from northern Mississippi to New England. The forecast windspeed o 50 75 90 97.5 100% map (fig. 3c) shows above-normal percentiles (<60 percent) over the Figure3c-Average afternoon windspeed. central and southern Rockies, the Great Basin, the Southwest desert, and scattered small areas elsewhere. The precipitation frequency map (fig. 3c) shows below-normal frequencies (~50 percent) over most of the area east of the Mississippi River. Very low precipitation frequencies (~O percent) were forecast for New England, the upper Ohio Valley, and the coast of North Carolina. Instructions for accessingthe monthly fire weather forecasts are printed in the screened box accompany­ ing this article.

Limitations of the Forecasts

o 50 75 90 97.5 100% In both the CAC monthly weather outlooks and our monthly fire weather Figure 3d-Precipirationfrequency. forecasts, each stationrepresents an area of about 30,000 square miles (77,770 km'), The average weather may vary greatly over relatively short distances not captured by the forecasts. In addition, the forecast values repre-

44 Fire Management Notes sent averages over the course ofa 30­ Pacific Sierra- Rocky day period; significant variations could NOrth c ..st , ,,,, ,, '::lilllMountalns ':lliJlOh'OV"", occur within that period. Work is 1:lllil ':lilll eo 'OOllillG""eo L,'" '''lilllG''''eo PI"" so currently underway at the Forest Fire 40 40 40 Laboratory to develop a technique for 20 20 20 '"ao "20 '"au c 0 0 o c c converting the monthly fire weather <50<75 <901!lO <50 <75 <90>90 <50 <75 -sceso <;0 <75 <00>90 <50<75<90>90 <50<75<90~O forecasts made for the NWS stations to forecasts for given forest locations. Experience with extended-range forecasts has shown that they tend to be most accurate in winter and least accurate during spring and fall, when

petoental1O~­ the day-to-day weather is most variable oll,me cat&ijory (Wagner 1989). Temperature is more au .- predictable than precipitation or relative : } .W~hin humidity, and monthly windspeeds '''00] 20 • Below are the most difficult to predict. Of o <50 <75<90>90 particular interest for fire managers and Foroc... ! porconti.. calogori .. Pacific Great Basin- Appalachians- Piei:lmontP Atlantic Gulf l~~lill1southwest l:OlilllUlhweSIDesert planners is the fact that summer and ':Soliill"h'"P"', "'''" ':lilllcoaSIS winter precipitation forecasts have 60 60 ':lllit comparable skill. 40 40 40 40 40 20 20 20 "o "0 o 0 0 Preliminary Verification of the <50<75<90200 0<50<75<90>90 <50<75<90>90 <50<75<00:>\10 <50<75<90~ Fire Weather Forecasts Figure 4-Relative accuracy ofthe fire weather temperature forecasts for various regions ofthe Forecasts can best be used when their United States for summer months for the period /973-83. For each ofthe four forecast percentile accuracy is known. The accuracy ofour categories, the bar represents the percentage ofthe time that the forecasts were correct. The line fire weather forecasts varies from one below the bar represents the percentage ofthe time that the observed temperatures were less than those forecast, and the line above the bar represents the percentage ofthe time that the observed parameter to another, from one season temperatures were greater than those forecast. to another, and from one location to another. A contingency table, or its graphical representation, provides a useful format for expressing the likelihood of success in forecasting. Figure 4 shows the relative accuracy of The diagram accompanying each region than the forecast category, and the line our afternoon temperature forecasts for shows the percentage ofthe time that above the bar represents the percentage different regions ofthe country for the the actual temperature was below, ofthe time that the observed tempera­ summer months for the period of 1973 within, or above the forecast category. ture was greater than the forecast to 1983. The forecasts were grouped For each category, the wide bar category. For example, in figure 4, into four percentile categories: less than represents the percentage of the time when the temperature in the Pacific 50 percent, 50-75 percent, 75-90 that the observed temperature was Northwest was forecast to lie between percent and greater than or equal to 90 within the forecast category-a correct the 50th and 75th percentiles, the percent. The stations were combined forecast. The line below the bar forecast was correct 25 percent of the into 11 regions that are both climato­ represents the percentage ofthe time time (bar). The observed temperature logically and statistically consistent. that the observed temperature was less was less than that forecast 63 percent of

1991 Volume 52, Number 3 45 the time (lower line) and was greater How To Access the.Monthly Fire Weather Forecasts than that forecast only 12 percent ofthe time (upper line). The set of five forecast percentile maps, two tables of the forecast values and their percentiles, and a narrative description of the CBI percentile map are combined into' Our fire weather forecasts of below a single computer file that is available shortly after the Ist and 15th of each month. normal temperatures «50 Percent) are Those with access to the Forest Service's Data General computer system can obtain generally the most accurate. Formost _these forecasts electronically; Those who do not have access to the Data General can regions, when slightly higher than access a file through the Automated Forest Fire Information Management and normal temperatures (50-75 percent) Retrieval System (AFFIRMS), a user-oriented interactive computer program are forecast, the actual conditions are as designed to permitentry of fire-weather observations and forecasts from field likely to be below normal as above locations. Thefile accessible from AFFfRMS includes the two tables of forecast values and the narrative description of the eEl percentile map. normal. Forecasts of the temperature . . . category that would contribute to moderate fire danger (75-90 percent) b~ta General, are least accurate, witli the observed temperatures more likely to be lower Go to the.Ig Main Menu: than higher, However, particularly in INFORMATION SYSTEM 6.42 . the Eastern States and along the Pacific Coast, our forecasts of extreme Select option 3. Utilities (Tapes, Dumpfiles, Import, Remote Access) (~O temperaiure percent) have been Select option 6. Ttansfer'(Information Transfer and DeC Access) accurate morethan half the time. Sele~t option I. Info transfer (Transfer Inforril~tion between FS offices)

Conclusions INFORMATION .TRANSFER UTILITY Screen 1 of2

Monthly fire weather forecasts that Transfer Type(1. Send,2. Retrieve): 2 can be used in fire management Local information structure planning are how available; medium­ range and seasonal fire weather Level(1. Public, 2. Staft): 2 Staff Name: (e.g.,MYSTAFF) DrawerName: (e.g.,MYDRAwER) [Respond here as appropriate forecasts are being developed. These FolderName:(e.g.,MYFOLDER) for your location] forecasts will not be the fire planner's File Name:CURMAPS.DMP magic bullets, but they do provide scientifically based long-range fore­ Local transferaction(YiN)? N casts. Because the fire weather forecast is inherently less accurate in the long­ range than the short-range, the user must consider the impact of variable be so severely degraded that the Literature Cited forecast accuracy. The accuracy of these forecast will provide no useful informa­ forecasts also varies with location, time tion. The intent ofproviding accuracy Chandler, Craig; Cheney, Phillip; Thomas, of year, and the parameter that is information, along with the forecast, is Phillip; Trabaud, Louis; Williams, Dave. forecast. to assist the user in determining when 1983. Fire in forestry. In: Forest fire behavior Planning activities require weather and effects, New York: Wiley & Sons. Vol. I. that point has been reached. • 450p. predictions to go further into the future. Chase, Richard A. 1991. Incorporating seasonal To maintain some level of accuracy, severity into annual fire program planning. In: forecasts are made for larger geographic Andrews, P., ed. Proceedings of the 11th areas and for longer periods of time. At Conference on Fire and Forest Meteorology; some point, however, the accuracy will

46 Fire Management Notes INFORMATION TRANSFERUTILITY . Screen 2 of 2

Host name: S27L05A Remote informationstructure Level (1. Public, 2. Stafl): 2 StaffNaine: MET Drawer Name: FORECAST Folder Name: MAPS . File Name: CURMAPS.DMP Remote Information (YIN)? N»" , Do you wantro omit CEO rriailnotification(YIN)? N Do you want-to submit transfer request NOW (YIN)? Y

Not~: To obtain the foreCasts for the previous period, substi~t1te·PRVMAps.DMPJor' CURMAPS.DMP.· .

Wait for mail message from IS_MGR: "Request: C.URMAPS.DMP queued to retrieve .. ; ."

To extract the files:At the ISj::Ll) prompt, type LOAD CUkMAPSDMf. This' extracts themaps, text,'aridmutative ftlesfrom the dumpfile, and they will.then _ appear in your directory. To print the maps: At the IS_CLI) prompt, type QPRIBINIQUE=3D JUN.91TEMP where, for example, 3D is a Data GenerafIaser printer.queuename and JUN91TEMP . Dotitlet bad is the map of forecast afternoon temperature percentilesfor June 1991. To print the text files:At the IS_CLI) prompt, type QPRIQUE"3D JUN9i FWTXT habits spread. where, for example,JUN9lFW.TX1: is the table of forecastfire weather variables at One little careless flick of a ciga­ the 127 stations. . rette butt can burn down an entire forest. So please be careful. Because AFFIRMS, once this bad habit starts, it's awfully hard to stop. Remember, The'A.FFIRMS file name of the current rn~ntl1's tables of forecastvalues and only you can ptevent forest fires.

'percentiles is MOFCU530,and, for the.previous month's forecast values, is f A PublicServiceof This Magazine MOFPV530. ,.' & TheAdverlising Council e ~ r-

1991 April 16-19; Missoula, MT. Bethesda, of Climate and Applied Meteorology. Fire and Forest Meteorology; 1989 April 17­ MD: Society ofAmerican Foresters: 34-41. 24(4).350-362. 21; Ottawa, ON. Ottawa, ON; Forestry Deeming, John E.; Burgan, Robert E.; Cohen, Klein, William H.; Whistler, Bruce T. 1991. Canada: 430-435. Jack D. 1977; The National Fire-Danger Specification of monthly mean anomalies of Mees, Romain; Chase, Richard. 1991. Relating Rating System-1978, Gen. Tech. Rep. fire-weather elements in the United States. burning index to wildfire workload over INT-39. Ogden, UT: U.S. Department of Agricultural and Forest Meteorology. broad geographic areas. International Journal Agriculture, Forest Service, Intermountain 560,2): 145-172. of Wildland Fire. 1(4): 211-214. Forest and Range Experiment Station. 63 p. McCutchan, Morris H.; Main, William A. 1989. Wagner,A. James. 1989. Medium- and long­ Englehart, Phillip J.; Douglas, Arthur V. 1985. A The relationship between mean monthly fire range forecasting. Weather and Forecasting. statistical analysis of precipitation frequency potential indices and monthly fire severity. In: 4(3):4 t 3-426. in the conterminous United States, including Maclver. D. C.; Auld, H.; Whitewood, R., comparisons with precipitation totals. Journal eds. Proceedings of the 10th Conference on

1991 Volume 52, Number 3 47 United States Department of Agriculture Washington, D.C. 20250

OFFICIAL BUSINESS Penalty for Private Use. $300

Superintendent of Documents Subscriptions Order Form Order Pmc_ing CodEr. Charge your order. IAiI.iIiiiIiiJ fIIII!!!IIl * 6131 It's Easyl ~~ o YES, enter my subscription as follows: To fax your orders (202) 512-2233 _ subscriptions to FIRE MANAGEMENT NOTES (FeN) for $650 per year ($8,15 foreign).

DNew o Renewal

The total cost of my order i~ $ . Price includes regular domestic postage and handling and is subject to change.

Please Choose Method of Payment: (Company or Personal Name) (Please type or print) D Check Payable to the Superintendent of Documents (Additional address/attention line) D GPO Deposit Account ITIIIIIJ- D D VISA or MasterCard Account (Street address)

(City, State, ZIP Code) Thank you for rrrn (Credit card expiration date) your order! (Daytime phone including area code) (Authorizing Signature) '019' (Purchase Order No.) YES NO Mail To: New Orders, Superintendent of Documents May we make your name/address available to other mailers? 0 0 P.O. Box371954,Pittsburgh, PA 15250-7954