Forest Management - US Forest Service · Fire Management Notes . An international quarterly...

38
United States Department of Agriculture Forest Service Volume 47, NO.3 1986 Fire Management Notes ".

Transcript of Forest Management - US Forest Service · Fire Management Notes . An international quarterly...

United StatesDepartment ofAgriculture

Forest Service

Volume 47, NO.31986

FireManagementNotes

".

FireManagementNotes .An international quarterly periodical devoted to

forest fire management

Contents3 Danger Zone: The Wildland/Urban Interface

James B. Davis

6 Chilean Fire CourseKen Dittmer

7 Florida NIIMS Resource Inventory ProgramJ.P. Greene and James Brenner

8 The Plastic Sphere Dispenser Aerial Ignition SystemJames D. Lunsford

10 Fine Tuning the Incident Command SystemJames R. Abbott

12 Fire Division Under ICSDavid L. Hanson

13 ICS QualificationsMarvin Newell, Bernie Erickson, and Jim Schneider

15 Wildfire Prevention: New Perspectives on an Old ProblemLinda R. Donoghue

19 The True Story of the Pulaski Fire ToolJames B. Davis

22 Evaluating Arson-Caused Forest Fires in Wisconsin, 1982--1l5Earl Meyer

26 Wildland Firefighters Personal Protection GearArt Jukkala and Ted Putnam

3I Using Interactive Videodisc Technology in Wildland Fire BehaviorTraining

M.L. Jenkins and K. Y. Matsumoto-Gran

Cover: Wildland fll'efighter equipped with personal protection geardeveloped by

the Missoula Equipment Development Center. See story beginning on p. 26.

2

United StatesDepartment ofAgriculture

Forest Service

Volume47, No.31986

Fire Management Notes is published bythe Forest Service of the untteo StalesDepartment of Agriculture, Washington,D.C. The Secretary of Agriculture hasoetermtneo tnat the publication of thisperiodical is necessary in the traneac­lion of the public business required bylaw 01 Il'\is Department. Use of funds forprinting this periodical has been ap­proved by the Director 01 the Office ofManagement and Budget throughSeptember 30, 1984.

Subscriptions may be obtained from theSuperintendent of Documents, U.S.Government Printing 011 ice,Washington, D.C. 20402.

NOTE-The use of trade, firm, or cor­poration names in this pubiication is forthe information and convenience of thereader. Such use does not constitute anofficial endorsement 01 any product orservice by the U,S. Department ofAgriculture.

Disclaimer: Individual authors are responsi­ble for the technical accuracy of the materialpresented in Fire Management Notes

Send suggestions and arncres 10 ChJef,Forest Service (Altn: Fire ManagementNotes), P.O. Box 2417, U.S. Departmentof Agriculture, Washington, DC 20013.

Richard E. LyngU.S. Department 01 Agriculture

R. Max Peterson, ChiefForest Servi.ce

L.A. Amicarella, DirectorFire and Aviation Management

Francis A. Russ,General Manager

Fire Management Notes

Danger Zone: The Wildland/UrbanInterfaceJames B. Davis

Research forester, USDA Forest Service, Forest Fireand Atmospheric Sciences Research, Washington, DC

.,

,,

In October 1871, a fire in Pesh­tigo. WI. killed more than 1,500people and burned 1.2 million acres.The Peshtigo Fire occurred in thesame month as the infamous Chi­cago Fire. But while everyone hasheard of Mrs. O'Leary's cow, fewhave heard of the Peshtigo fire inwhich four times as many peopledied.

Ancient history? Not so!The 1985 wildland fire season

was the most severe of this century.By the end of the fiscal year. over83,000 wildfires had burned almost3 million acres, destroyed or dam­aged in excess of 1,400 structuresand dwellings, caused the deathsof 44 civilians and firefighters. andcost the Federal, State. and localfire agencies and private industryover 400 million dollars in firefight­ing costs. Damage estimates to natu­ral resources and property are notavailable, but probably run into thehundreds of millions of dollars.

The Southern States east of theAppalachian Mountains from Floridato Virginia, parts of New England.Idaho. Nevada. and central Califor­nia were especially hard hit by wild­fires. National mobilization wasneeded in the Western States andin the South to cope with wildfires.During the first week of July. atotal fire mobilization of over20.000 Federal and State firefighterswas committed to fires in 13 West­ern States, including massive fires inCalifornia, Idaho, Oregon, and Ne-

1986 Volume 47, Number 3

vada.The loss of property was the

worst since 1871 when the PeshtigoFire destroyed entire communities.Major losses of buildings occurredin Florida, North Carolina, and Cal­ifornia, but reports of structurelosses have also come from Wash­ington. South Carolina, Oregon,and New England. The number ofstrnctures saved by wildland fire­fighters is not known, but wildfirereports routinely listed "structuresthreatened" in daily status reports.Because protection of property andlives took priority, natural resourcelosses increased when fire forceswere diverted to protect structures.

Last year's loss in lives and prop­erty is part of a developing trend.A major population shift from urbanto suburban living in the years sinceWorld War II has greatly expandedwhat is now called the urban/wild­land or woodland home environ­ment-the zone where people arein contact with the wildlands forreasons not related to timber orother traditional forest uses. Al­though this trend has increased thegeneral population's appreciationfor the amenity values of forests.it has also greatly increased thenumber of primary residences, sec­ond homes, and retirement homeslocated in forests and brushlands.Vast areas of the United States con­tain high-value properties inter­mingled with flammable native veg­etation.

Structural fire losses are increas­ing dramatically as more peoplebuild and live in proximity to flam­mable plant communities. Majorloss of life is possible-in fact, in­evitable. The problem is not, asis often believed, one uniqueto southern California. The extensionof residential and commercial de­velopment into areas with high firerisk has been noted throughout theNation-from the Georgia Piedmontand the sand plains of central Michi­gan to the Rocky Mountain foothillsnear Denver to northern New Eng­land.

Although current fire managementpractices make it unlikely that fireswill ever again reach the huge pro­portions of those in 19th-centuryAmerica, the risks to life, property,natural resources, and economicwelfare are much higher today thenever before. Huge fires are not re­quired for catastrophic losses in themodern wildland/urban interface.Even small fires can be killers--­three homeowners died when an8-acre fire swept their BaldwinHills, CA. subdivision. Fire manage­ment must change in order to betterprevent and suppress smaller, fast­moving single and multiple fires asa wildland/urban interface continuesto expand. This change must occurnationally.

The task of protecting lives andproperty from wildfires in the wild­land/urban interface poses one ofthe most critical and elusive prob-

3

Most forest fire suppression personnel areinadequately prepared for fighting structuralfires. . . .

Losses of property and life as a result of fires in the wildland/urban interface have been increas­ing dramaucaliy in recent years.

lems faced by wildfire protectionagencies. Wildfire protection agen­cies have broken the problem downinto several parts:

• Fire managers are unable toreliably predict erratic fire behaviorin mixture of structures, ornamentalvegetation, and wildland fuels char­acterizing the interface. Physicalfuel properties and moisture relationsin these areas are not well under­stood, as they are governed by bothnatural and human-caused phe­nomena. Possible relationshipsamong building and landscaping lo­cation, design, and construction,with respect to terrain and otherstructures, add to the complexityof fire behavior. For example, spot­ting (fires starting from flying em-

4

bers) is especially difficult to fore­cast due to the diversity of firebrandmaterials and unusually complexwindflow patterns. Yet, spotting isthe chief cause of structural fireignitions in wildland/urban areas.

• Use of prescribed fire for haz­ard reduction (fires purposely setto remove undesirable vegetation)is made difficult by legal, political,and environmental concerns. Lia­bility for damages to intermingledprivate holdings is a significant de­terrent. In many cases, the veryreason for living in the interfaceprecludes the use of fire. Nonethe­less, means must be found to man­age fire hazards in the interface,while maintaining or enhancing de­sired environmental and economic

values.• Many property owners are un­

aware of the wildfire threat, andfire safety ordinances and buildingcodes are frequently inadequate,unenforced, or disregarded. A quin­tessential example is the insistenceon flammable roof materials in thechaparral area of southern Califor­nia, but similar..JlUitudes are ex­hibited throughout the world: Thedesign of subdivisions, also',' con­tinues to defy principles of firesafety. Many areas include narrow,winding, or dead-end roads withinadequate water systems. Lots arefrequently too narrow to permiteffective vegetation removal. With­out strong motivation to change,homeowners and developerswill continue to produce and main­tain these dangerous communities.

• Most forest fire suppressionpersonnel are inadequately preparedfor fighting structural fires, whereasmunicipal fire departments are notalways fully trained or equipped forwildland fire suppression. Althoughrelatively new organizational systemsfor integrating a variety of fire pro­tection resources and personnel haveproven effective, the special de­mands of fires in the wildland/urbaninterface often force firefightingpersonnel to perform unfamiliar tasks. The need to meld struc­tural and vegetation fire expertiseon interface fires remains a formid­able challenge.

The following actions and im­provements are needed:

FireManagement Notes

",

• Effective techniques and strat­egies to assess and manage fire haz­ards in the wildland/urban interface.

• Aids for planning, budgeting,and training for increased involve­ment in the residential/wildland in­terface to ensure a balanced capa­bility in conducting structural andwildland fire suppression activities.

• Effective ways to educate prop­erty owners, land developers, in­surance carriers, and local plannersabout vegetation fire problems andsolutions.

• Fundamental knowledge aboutthe physics of fire spotting andcrowning in the wildland/urban inter­face.

• Knowledge about relationshipsof building design, materials, and

1986 Volume 47, Number 3

landscaping with fire hazard and be­havior.

• Improved understanding of whypeople build fire-prone homes inhighly some flammable areas andhow they respond to various motiva­tional tactics to reduce vulnerability.

Who is responsible for the solu­tion, the fire protection agency,the homeowner, the county planner?The responsibility for fire protec­tion cannot be relegated to a singleelement of society. It calls for acombined effort. Just a few of thegroups that share in the respon­sibility include:

• Homeowners• Fire protection agencies• Local and regional planners

• Media and communication ex­perts

• Insurance carriers• Builders, contractors, and archi­

tects• Training and motivational ex­

pertsA trnly integrated approach to

the problem would greatly reduceits impact. We all must take a handin solving the problem. We muststrive to avoid a 20th-century Pesh­tigo fire. There is no justificationfor continuation of such a serioushazard to life and property.•

5

Chilean Fire CourseKen Dittmer

Director, National Advanced Resource TechnologyCenter. Marana, AZ

Smokey Bear and his Chi/ean counterpart, Forestin the beaver, share the fire prevention messagewith Chilean schoolchildren.

For the third time in as manyyears, firefighters from Central andSouth America, Spain, and Portugalgathered for training in various as­pects of fire suppression, The mostrecent session, in late Novemberand early December, 1985, was heldnear Santiago, Chile, rather thanat the National Advanced ResourceTechnology Center (NARTC), Mar­ana, AZ, where the two previouscourses had been held. Through thecombined efforts of 17 fire expertsfrom Spain, Argentina, Chile, andthe United States, 56 firemen repre­senting 15 countries were trainedin the latest techniques and philoso­phies of fire suppression. It is hopedthat these individuals will he ableto train others upon returning home.

During the 2 weeks of classroomtraining, the students were givencourses in effective training tech­niques, fire behavior, preventionand detection, fuels management,organization for suppression,line construction, and air operations.A week of field exercises in earlyDecember, the start of the fire sea­son south of the equator, affordedthe students an opportunity to applywhat had been presented in theclassroom and to participate in a

6

prescribed burn. Field trips to na­tional and private fire facilities andforest operations exposed thetrainees to a fine example of organi­zation and cooperation in an en­vironment similar to what they areaccustomed to at home.

Funding for Tercer Curso Interna­tional Avanzado, El Combate deIncendios Forestales was made avail­able through the Agency for Inter­national Development (AID); Corop-racion Nacional Forestales(CONAF), the national land manag­ing agency of Chile; and the U.S.

Department of Agriculture ForestService.

Smokey Bear made his first tripoutside the United States, joininghis Chilean counterpart, Forestin, abeaver, in promoting fire prevention.Smokey and Forestin visited severalschools to pass out fire preventionmaterial and appeared on nationaltelevision.

As part of the continuing effortto attempt to share experience andtechnology among countries, Mexicoplans to host a similar course in1987.•

FireManagement Notes

'\

."

Florida NIIMS Resource InventoryProgramJ.P. Greene and James Brenner

Fire management specialist and fire prevention coor­dinator, respectively, Florida Division of Forestry,Tallahassee, FL

When the Florida National Inter­agency Incident Management System(NIIMS) Task Force was formedin 1982, one of its first actions wasto form a logistics working teamwith the mission of developing aworkable wildland fire suppressionand support resources inventoryfor the State.

In order to inventory resources,it is first necessary to define theresources in such a manner as toensure that the inventory is uniforrnand accurate. When consideringfire suppression resources, however,enough flexibility must be built intothe system to allow for variationsin equipment while ensuring that theresource will do the job desired.Using the southern California defini­tions as a model, the team generateda list of resources adapted to theSoutheast in general and Florida inparticular. As outlined in nationaldoctrine, resources are typed by ca­pability, with lower numbers havinggenerally greater capabilities.

After adoption of the resourcedefinitions, an inventory method wasdevised. An inventory form wasdeveloped and distributed to the 17Florida Division of Forestry districtsfor completion. Fire departments,forest industry cooperators, andFederal, State, and local agencieswere surveyed by county unit. The

1986 Volume 47, Number 3

resulting individual forms were thenentered into a computer program.

The NIIMS resource computerprogram requires an IBM PC orcompatible unit with 256 K mem­ory. The software used is RBASE4000 or 5000, published by Micro­rim, Inc., and the NIIMS programdeveloped by the Florida Division ofForestry. The data-base size is lim­ited only by the number of bytes perdisk. The Florida data is currentlystored on two double-density, two­sided floppy disks with approx­imately 364 K on each.

The two disks contain 1,268 ad­dress records, 1,813 radio recordsand 3,197 equipment records. Thisdoes not mean that there are 3,197pieces of equipment recorded onthose two disks; rather, it means thatthere are 3,197 lines or recordsavailable. The actual numbers ofequipment recorded are 3,055of type I, 2,529 of type 2, 1,620of type 3, and 1,099 of type 4,for a grand total of 8,303 piecesof equipment on record.

The data-base is divided into threerelations-addresses and relateddata, radios and related data, andequipment types and amounts. Forthose using RBASE 5000, it maybe desirable to rewrite the reports sothat all three relations can be ac­cessed at the same time.

To use the data, RBASE is loadedinto the PC, the appropriate data

disk is selected for the area in­volved, the data base is called up,and the program is run followingthe menu and instructions presentedon the computer monitor. Editingof the data follows much the sameprocess.

Through the menu, the programwill:

I. Display contact data (address,telephone number, etc.) for agenciesby cooperator, county, district, orstatewide.

2. Display equipment data.3. Display radio equipment.4. Display all equipment and radio

data for a selected cooperator,county, or district.

5. Search for specific kinds ofequipment by county, district, orstatewide.

6. Search for specific radio fre­quencies by county, district, orstatewide.

Through the use of RBASE Ex­tended Report Writer software, thedata may be manipulated in variousways to produce mailing lists andother products.

For further information on theFlorida NIIMS resource inventory,contact J.P. Greene or Jim Brennerat:

Florida Division of ForestryFire Control Bureau3125 Conner Blvd.Tallahassee, FL 32301

7

The Plastic Sphere Dispenser AerialIgnition SystemJames D. Lunsford

Prescribed fire, fire behavior, and smoke managementspecialist, USDA Forest Service, Atlanta, GA

Fire Management Notes

Figure I-The plastic sphere disp~;;;;'-;;o~n-r;d ;;~-helicopter, is an ejjectjve-~;gnitio;' ;~u--;:;for prescribed burns.

.y! ,'1+\~'/'';;W

glycol, an exothermic reaction oc­curs in about 20 seconds, causing aflame sufficient to ignite fine fuelssuch as grass and pine needles. Thesystem has also been used to ig­nite logging slash in clearcut areasand windrows of logging slash.

Use of Dispenser

Successful operations were con­ducted in February 1986 on theAppalachicola National Forest inFlorida and the Kisatchie NationalForest in Louisiana for fuels reduc­tion burning. The England Air ForceBase uses a portion of the Kis­atchie National Forest as a firingrange for A-10 aircraft. The Forestmaintains the fuels on the area to

second. At aircraft speed of 50miles per hour. ignitions can beplaced from 25 to 200 feet apart,as desired. Spacing can be varied bychanging dispenser speed and/oraircraft speed.

The machine, a Premo-Mark IIIdispenser, is manufactured by PremoPlastics Engineering, Ltd., of Vic­toria, BC, Canada, and has provento be a very effective aerial ignitionsystem for both prescribed fire andburnout/backfire operations on wild­fires.

About the size of Ping-Pong balls(33 mrn), the plastic spheres contain3 grams of potassium permanganate(KM"04)' When the spheres areinjected with I milliliter of ethylene

them from an aircraft at a rate ofone each 2.6 seconds to three per

8

(common antifreeze) and dispensing

How the Dispenser Works

The dispenser is portable (about100 pounds) and can be mountedin a helicopter or fixed-wing­aircraft. The dispenser performs thefunction of systematically injectingthe spheres with ethylene glycol

During the early 1960's, Aus­tralian foresters developed an aerialignition system to ignite spot firesin large blocks of eucalyptus to con­sume litter and reduce a fire hazard(1). The system has been used inthe United States since the early1970's by private industry and someState divisions of forestry but wasnot approved for use by the ForestService until 1986.

After safety testing by the Mis­soula Equipment Development Cen­ter (ED&T 4E42 P14, May 1985),the Southern Region (R-8) wasasked to develop a user's guide onoperating procedures prior to ap­proval of the system by the ForestService.

The operating guide was com­pleted in February 1986, followinga workshop held in Tallahassee,FL, and attended by representativesfrom each Forest Service Regionand the Washington Office. ForestService regulations require eachoperator to attend an 8-hour work­shop to obtain certification to oper­ate the dispenser.

Figure 2-Plastic spheres containing potassium pennanganate are injected with ethylene glycol,causing a flame sufficient to ignite fine fuels. '

prevent escaped wildfire ignited byAir Force use. This year, the ForestService was' assigned a 2-hourperiod to conduct a prescribed firebetween strafing runs. The 3:500acres were burned in 1 hour usingthe plastic sphere dispenser anda Bell 206-B helicopter. Asthe smoke cleared, the A-lO's wereflying again. Under these conditions,

1986 Volume 47, Number 3

it is imperative to' complete theburning operation on time (the AirForce does use real bullets!). Inanother operation, 11,285 acres wereburned in 9.4 hours of flight timeover a 3-day period.

Cost of ignition is about $1 peracre. Plastic spheres are about13 cents each, and antifreeze is in­expensive. One ignition spot peracre gives satisfactory results in the

piney woods of the South. Thedispenser holds I gallon of anti­freeze, which will inject 7,000spheres. Thus, the dispenser hasthe capability of burning 7,000 acreswithout the helicopter returning tothe helibase. If the bum boss is onboard the helicopter, several areascan be burned during a single flight.

Linear spacing of ignition withthe wind causes excessive heat asflank fires merge. A grid patternwith shorter flanks seems to produceless intensity (2).

Holding crews are needed oneach bum unit but may progressto another unit because ofshort bum out time.

Efficiency and simplicity of theplastic <sphere dispenser make ita desirable alternative ignition sourcefor prescribed fire and wildfireburnout.

Literature Cited

l. Baxter, I.R.; Packham, D.R.; Peet, G.B.Control burning from aircraft'. Melbourne,Australia: CSIRO Chemical ResearchLaboratory; 1966. 26 p.

2. Johansen, R.W. Prescribed burning withspot fires. Georgia For. Res. Pap. 49.Macon, GA: Georgia Forestry Commission;1984. 7 p.

9

Fine Tuning the Incident CommandSystem

James R. Abbott

ICS working team chairman, forest supervisor, LincolnNational Forest, Alamogordo, NM

The Incident Command System(ICS) received an effective test on aninteragency basis during the 1985 fireseason. Although some units had littletime to train and prepare, their firstuse of the system on actual fires wentquite smoothly. This learning experi­ence provided a practical means ofbridging gaps between fire manage­ment systems and promoting an un­derstanding of the differences in ter­minology and organization. It alsoprovided an opportunity for experi­enced and knowledgeable people tocritique the system and begin to sug­gest improvements that should bemade. In addition, adaptations to lo­cal needs were considered.

As all agencies gain experience inusing ICS there will be a natural tend­ency to propose modifications to meetspecific local concerns. ICS is in­tended to be flexible enough to meetlocal needs; however, uncoordinated,independent changes could affect thesystem's utility as an interagencytool. It is also probably true that somechanges are suggested only becausenew users do not fully understand

10

how a particular problem is meant tobe handled.

Other organizations such as law en­forcement agencies and various emer­gency services are also adopting thesystem for their use. Although thisexpansion has many positive aspectsit could produce more variations inapplication that mayor may not beapplicable to wildland fire. Thus,wildland fire agencies must carefullymanage system changes if the in­tended interagency management bene­fits are to be realized.

The National Wildfire CoordinatingGroup (NWCG), through the ICSworking team, provides the mecha­nism to coordinate the maintenanceand revision of ICS with appropriateinput and assistance from involvedand interested agencies. Proposalssuggested through team members, ifaccepted by the NWCG, will be in­corporated into the system on a peri­odic scheduled basis. Without this co­operation, different agencyapproaches could soon create prob­lems in applying the system in intera­gency situations.

Controlling system change throughthe NWCG has a high payoff for theagencies and States involved. Each

. idea and need can be examined byusers in a way that features which en­hance the system can in tum beadopted. Fine tuning in this way mayrequire more time but the alternativeof each agency modifying the systemon a continual basis would seriouslydetract from the usefulness of thesystem.

The ICS working team has a setprocess for considering all sugges­tions. The process used for consider­ing all suggestions is shown in figure1.

What can you do if you have a sug­gestion for improving the IncidentCommand System?• Analyze and document the benefits,

costs, and consequences of theproposal.

• Review the proposal with agenciesthat you work with locally to obtaintheir ideas.

• Submit the proposal through agencychannels or to the author or an ICSworking team member. •

Fire Management Notes

NWCG submits to appropriate entityfor implementation 14

NWCG returnstolCSWTwithinstructions 17

NWCGdisapproves >--------1 No 1.------<

NWCGapprovesproposal

ta

12

Submit proposalto NWCG withrecommendedaction

Submit10review

ICSWT makesassignmentfor furtherstudy 84

Return \0sender withexplanationor request

Proposal or ICSWT Outsidesuggestion

ICSWT prepares reviewc,

(Input) submittedevaluates proposal in neededinput final form

to JCSWT 1 2 9

10

,

Figure I-Process used by the ICS working team (lCSWT) for considering suggestions for modifying the system.

1986 Volume 47, Number 3 11

Fire Division Under ICSDavid L. Hanson

lCS working team member, assistant fire chief, LosAngeles County Fire Department, Los Angeles, CA

As it is becoming more widelyused across the Nation, the IncidentCommand System (ICS) has leftsome of us in the wildland fire pro­tection agencies with questions abouthow fire incidents are intended tobe broken down into manageable geographic areas.

Divisions. as they are describedin ICS, are intended to representa geographical area for the purposeof organizing and directing fire sup­pression resources on wildland fireincidents. Some of the concerns thathave been expressed about the useof ICS divisions include a belief thatthe division may encompass toolarge a geographical area and thatsome additional smaller geographicalarea designator may be needed.There has also been a reluctanceto use ICS divisions on initial attackoperations or on small multiopera­tional period fires because of thelack of readily available qualified di­vision supervisors.

Some of these concerns can beaddressed by accepting the ideathat using the ICS term "division"to refer to geographical areas onyour fire incident does not necessar­ily indicate that the division coversa large area or has a large num-ber of resources or even that a per­son qualified as a division supervisoris commanding that division. The

12

Initial Attack Incident Commanderwho separates an incident into twoor more .divisions, regardless ofthe size of the incident, has madean important step towards imple­menting ICS. If the incident esca­lates in size and complexity theincident commander is in a positionto expand the organization accord­ingly.

Even on a relatively small inci­dent, the Initial Attack IncidentCommander, with one crew, cantake one side of the fire and call itdivision A, assigning a second crewto take the other side of the fire,division B. There is no real need toassign someone qualified as a divi­sion supervisor to command a di­vision until or unless the personneland resources required by that di­vision become extensive enough tomake such an assignment necessary.

Another concern in our adjustmentto the use of divisions in ICS isthe loss of the lesser geographicalarea designator, the sector, thatmany of us were accustomed to inour previously used incident­management system, the Large FireOrganization (LFO). Perhaps itwould be helpful to consider thatin the old LFO, even when a sectorwas used within a division, often­times further informal unstructuredgeographical separations were-

employed within the sector. For ex­ample, the sector boss would haveone crew boss cover from pointA to point B and another crew bosscover from point B to point C andso on.

lt is also perfectly acceptablewithin ICS to use divisions that in­corporate other lesser geographicalareas as designated by the divisionsupervisor. For example, the divi­sion supervisor on a major incidentmay decide that he needs help insupervising fire control line­construction in part of the division.He delegates that task to a personqualified as task force leader fromamong the division's assigned singleresource bosses.

lt is important to realize that ICSshould not be seen as limitingwhat we do in dividing or organiz­ing our fires. Instead, ICS shouldbe seen as a basic framework for or­ganization that can be applied toinitial attack and, if necessary, sys­tematically expanded to handle ex­tended attack or even major inci­dents.

Sharing experiences, problems,and solutions among all the users ofICS still offers the promise thatall of us with a fire protection mis­sion can improve our own opera­tional abilities as well as our abilityto work mare effectively together.

Fire Management Notes

les Qualifications

Marvin Newell, Bernie Erickson,and Jim Schneider

Staff assistant and computer systems programmer, respec­tively, USDA Forest Service; and forester, USDI Bureauof Indian Affairs; Boise Interagency Fire Center, Boise,ID

In 1984 the National WildfireCoordinating Group's (NWCG) quali­fication and certification workingteam, now the Incident CommandSystem (ICS) working team, pub­lished the Wildland Fire QualificationGuide (I). This guide includes theNWCG requirements for minimumtraining, experience, and physical fit­ness for positions in the ICS. The re­sult of these common standards hasbeen the formation of a nationwidepool of wildland firefighting person­nel who are qualified to perform firesuppression jobs fur which they havebeen certified. Some agencies'havesupplemented the guide with agencyrequirements for their own use.

After the production of the qualifi­cation guide, an obvious next stepwas to assist managers by developinga data management system to recordand track information necessary forqualifications. The ICS working teamdescribed the needed criteria, and theForest Service agreed to develop, test,and recommend such a system to theNWCG. The result was an automatedrecordkeeping system for wildlandfire qualifications that has been ac­cepted by NWCG (fig. I). The pro­gram resides at the Ft. Collins Com­puter Center (FCCC).

Basically the program is a datamanagement system for personnel in­volved in wildland firefighting. Quali­fications are determined by agencymanagers rather than by computeranalysis. The initial transition to ICSis accomplished by entering the posi­tion(s) that the individual has been

1986 Volume 47, Number 3

qualified to perform in, based upon alocal management review of his or hertraining and experience. Target posi­tions for future qualification can beentered as well as subsequent trainingand experience. Planned training, spe­cial skills, or other information can beentered in the remarks section. Theprogram has the capability to createshort history files suitable fordownloading to microcomputers foruse in locally developed applications.

The information is transferred fromthe data entry form (form 320) to acomputer data file format by whatever

data entry system is. available to theuser. Forest Service users can useForms Entry Systems (FES) on DataGeneral equipment. Other userswould have to find a means on theirlocal system that could communicatethe resulting data file to FCCC forprocessing by the program or use dataentry facilities available at FCCC.

Agencies or units interested in thesystem should order the needed mate­rials from the Boise Interagency FireCenter warehouse. There are fouritems available: User's Guide for theADP System, catalog #1516; PMS

13

Form 320 for manually recording andstoring data, # 15J7; red cards withthe headings preprinted for typewriteror manual use, #1518; and blank redcards for computer printer use,# 1519. The Wildland Fire Qualifica­tions Guide, PMS 310- J, which isnecessary for determining nationallevel qualifications, is #1414.

General Services AdministrationSupports Firefighters

The General Services Administra­tion (GSA) has been a member ofthe wildlife suppression communitysince 1956. Through its FederalSupply Service (FSS), GSA con­tracts for wildfire suppression equip­ment and supplies needed to helpprotect the Nation's forest and grass­land resources from the ravagesof fire. FSS considers its wildfiresuppression equipment program sec­ond only to national security inorder of priority.

An FSS catalog, "Wildfire Sup­pression Equipment and Supplies,"is the basic supply publication foruse by the United States ForestService, Bureau of Land Manage­ment, Bureau of Indian Affairs,National Park Service, and otherForest Service certified participantsin the Governmentwide program.Because GSA complies with all reg­ulations governing the competitivepurchasing of items, customer agen-des may, in compliance witI-! theirinternal agency procedures, order di­rectly from GSA.

14

This program is one method ofmanaging rcs data currently availableto any user. Individual agencies orunits may have or develop othermethods to better meet their particularneeds. Local agency managers maywant to contact their supervisor to de­termine the best method available tothem.•

Easy acquisition of firefightingequipment and supplies stocked inGSA warehouses is the first stepin GSA's assistance to firefighters.In the event there is any problemwith a GSA stock item order, thenext important form of assistanceis the followthrough provided--that'swhere GSA's Discrepancy ReportsCenter comes into play.

At the time catalog orders arereceived, the merchandise shouldimmediately be checked for visibleirregularities such as overages, short­ages, or damage. Any carrier abnor­malities should be stated on thecarrier's delivery form with the car­rier's representative acknowledgingthe same. A report of discrepanciesshould be sent immediately onStandard Form 361 for carrier dis­crepancies and on Standard Form364 for shipper discrepancies (GSAwarehouse or vendor) to:

Discrepancy Reports Center!500 E. Bannister RoadKansas City, MO 64131

Quality complaints are handled

Literature Cited

I. Whitson, Jim; Hanson, Dave. How NIIMSKeeps Qualification Standards From Becom­ing a Barrier to Interagency Cooperation.Fire Management Notes. 45(4): 18.

by the GSA Quality ComplaintHotline on FTS 557-1368 (703)577-1368. Regulatory informationregarding discrepancy reporting isdetailed in FPMR 101-26.8, "Dis-.crepancies or Deficiencies in GSAor DoD Shipments, Material, orBillings. "

Additionally, customer serviceofficers in the GSA regional officesacross the country are available toassist firefighters in obtaining wild­fire suppression equipment and sup­plies (telephone ndmbers and ad­dresses are in the "GSA SupplyCatalog"). Discrepancy ReportsCenter specialists are on hand toassist with order discrepancies onFTS 926-7447 or Autovon 465­7447.

The natural environment is pre­cious, and GSA's Federal SupplyService stands ready to help main­tain it.

Louise NylandGeneral Services AdministrationWashington. DC

Fire Management Notes

Wildfire Prevention: NewPerspectives on an Old ProblemLinda R. Donoghue

Research forester, USDA Forest Service North CentralForest Experiment Station, East Lansing, Ml

Table l-elassijication ofdump fires in the Northeastern United States

gories (general cause, specific cause,and class-of-people) were addedto pinpoint both the cause and whostarted the fire.

Although to our credit we'vemaintained consistency in our fire­cause reporting system, unfortunatelywe've also created categories thatare repetitious and not mutually ex­clusive. As a result, we can easilyhave multiple classification schemesfor a given fire cause, making itdifftcult to determine from fire-re­port data the actual cause and personresponsible for a particular wildfire.For example, a dump fire set by oneor more persons is reported in theNortheastern United States usingseven different cause and class-of­people combinations (table I). Whatseems simple to record tums outto be complex!

Compounding the problem ofoverlapping or non-mutually exclu­sive categories is the broad andrepetitious nature of our fire-causereporting system.

Our system also lacks any wayfor reporters to indicate the certainty

Because of certain social trendsoccurring in the United States andalso because of some relatively newadministrative requirements, ourneeds for fire prevention informationare changing. I will describe ourcurrent system of information gather­ing, pointing out some of thestrengths and weaknesses of the sys­tem, and then discuss the new de­mands for fire prevention informa­tion growing out of social changeand administrative mandates in theUnited States.

Most of our current fire preven­tion information, like yours, comesfrom our individual fire reports.The key elements from a preventionstandpoint are: statistical cause,general cause, specific cause, andclass of people. Statisticalcauses, which are used by all pro­tection agencies in the United States,are reported annually in our "Wild­fire Statistics." General cause isdefined as the general type of landuse activity responsible for a fire,and specific cause as the specific ac­tivity or ignition source causingthe fire. This cause classificationsystem, though amended occasion­ally, has been used by the ForestService for 80 years and appears onState and Federal fire reports invarying formats. When combinedwith location, size class, andacreage burned information, it notonly gives us an idea of whatour problems are but where and howbig they are. Based on this infor­mation, our land managers select ap-

Statistical General Specificcause cause cause

1. 'ncandial)' incenciary Burning dump2. Incendiary Incendiary Grudge3. Debris burn Other Trash burning4. Debris bum Resident Burning dump5. Debris burn Incendiary Trash burning6. Debris burn Other Burning dump7. Miscellaneous Resident Burning dump

propriate fire prevention programsand allocate resources to solve theirwildfire problems. By monitoringtrends in fire causes, they also usethe information to evaluate the im­pacts of their programs.

Although managers find this proc­ess workable, it still can be im­proved. No matter how much sta­tistical wizardry we scientistsperform on the data, we can't makeit any better. Although we take whatis on the forms and transform itinto impressive graphs and chartsfor our user groups, what comes inmust go out. Therefore, if thedata coming in are inaccurate whatgoes out is inaccurate. One of thereasons for inaccurate data is the de­sign of the reporting system; theother part is due to our assessmentof fire causes at the fire scene.

First, let's look at the reportingsystem itself. Ours follows a­progression of "add on" informa­tion. Eighty years ago, westarted with eight basiccauses. These apparently were inade­quate for over the years new cate-

Class-of­people

Local permanentlocal permanentVisitorLocal permanentlocal permanentlocal permanentlocal permanent

1986 Volume 47, Number 3 15

of their fire-cause data. That is,once entered on a fire report form,reliable causes are indistinguishablefrom unreliable ones. With no "un­known" category available, an un­known cause is typicallyclassified as a smoking. incendiary,or miscellaneous wildfire. Becausereporters cannot indicate the re­liability of fire causes, the accuracyof fire-cause data is often ques­tionable, leading to erroneous as­sumptions about the fire preventionproblems.

Even if our reporting system wereperfect, we would still have thedifficulty of assessing the cause ofa wildfire at the fire scene. Whatit boils down to is that weoften don't do a good job of fire­cause investigation. And, if We

don't do a good job investigating,what goes on the fire reports andinto our data files isn't very goodeither.

So 'what are we doing about allof this? It wasn't too long ago thatthe building block system was pro­posed to replace our traditionalsystem. This new system includedthe:

• Form of heat energy that igni­ted the fire: for example, flames,sparks, and hot surfaces from a vati­ety of sources.

• Reporter's certainty of heatof ignition.

• Equipment involved in startingthe fire: e.g., cooking equipment,electrical equipment, woodlandequipment.

16

- -----

• Material first ignited: grass,leaves, paper, hay, etc.

• Ignition factor: how the formof heat energy and the material firstignited combined to start a wild­fire-s-for example, incendiarism, misuse of equipment, me­chanical failure, or design defi­ciency.

This system also included thetype of person responsible for flres,the person's age and sex, and theprimary activity the person was in­volved in when the flre started.Although the idea was good,it wasn't widely accepted becauseit was too cumbersome to use (25pages of categories to choose froml)and broke too much with tradition.

The Forest Service decided tocompromise. Our new fire reportcontains only statistical causes; de­bris burning, campfires, incendiary,railroads, smoking, equipmentuse, children, miscellaneous, andlightning. We're back to where westarted in 1905! These arereported uniformly across the Na­tion. The system is obviously simpleto use and historically compatiblewith previous fire report data.If, however, managers want moredetailed fire prevention informationfor their administrative units, theycan also use the building block sys­tem contained on a supplementalfire report form.

Finally, to improve the qualityof the data, fire-cause investigarionis being emphasized nationwide.This is, in part, a reaction to the in-

crease in arson in the UnitedStates. Nationally, it's the numberone cause of wildfires. To get abetter handle on the magnitude ofthis problem, the wildland fire COm­munity has decided to improve itsinvestigative skills. With better in­vestigative work comes better wild­fire cause data.

What I've just described is ourcurrent system of reporting datathat is useful to the fire preventioncommunity. And it's a traditionalsystem. We use the information, aswe have for decades, to defineour fire prevention problemsin terms of number of fires, theircauses, acreage burned, and soforth. And, if we're asked to dem­onstrate our achievements. we pointto the increase or decrease in hu­man-caused fires over time. But inour day and age, this is nolonger adequate.

With tightening budgets, cutbacksin our management programs andpersonnel, and mandates from Con­gress to demonstrate the cost ef­fectiveness of our fire managementactivities, the Forest Service hasimplemented a national system foranalyzing, from an economic stand­point, fire management programs.

The system provides a formalprocess for evaluating the efficiencyand effectiveness of fire programsat the national and regional levels.

When I say "fire programs,"1 mean total fire management pro­grams which include wildfire­prevention. In the past, we've

FireManagement Notes

IIIIIIIIIIIIIII,,:(A)

:1500IIIIIIIIII

500 1000

1125

375

750

(E)

All of this information can thenbe combined into one map or setof nomograms to show. for a par­ticular administrative unit, whatour most efficient level of programoperation is and what the conse­quences of that program might be.Figure I is an example. We'vedeveloped this diagram for the State

9

8

7

6

54

3

21

1000 500

Enforcementcost Arson wildfires(in thousands .--+--'-t-'--t-="""Y---+--f---i-+--...

of dollars)

Enforcementeffort

Figure I-Nomogram for the State of Arkansas depicting the relationship between the law en­forcement program, the number of arson wildfires, and the economic impact of those fires.

C+NVC(in millionsof dollars)

2. How much of each of thoseactivities we perform.

3. How much these activities cost,both fixed and variable costs.

In addition to our suppressioncosts, we also need to know the netvalue change in our wildland re­sources resulting from human-causedwildfires.

tended to separate prevention fromthe rest of the fire picture. It wasalways an isolated event outside ofthe mainstream. Now. we're lookingat the complete picture, includingprevention, presuppression, and sup­pression activities. This brings upa number of questions about theeconomics of fire prevention pro­grams. For instance:

• How much do our fire preven­tion programs cost us and whatdo they "buy" us in terms of num­ber of wildfires?

• What is an efficient programlevel?

• What is an efficient budgetlevel?

• How many fires do these effi­cient program and budget levelsgenerate and what's the cost plusnet value change of these fires?

• How does fife prevention inter­act with the rest of fire manage­ment? If we reduce the number offires, how does that affect our sup­pression forces?

These are some of the questionsthat those of us in fire preventionare now asking and are required toask. We're no longer looking atjust an increase or decrease in thenumber of fires but, in addition,the costs and effects of thesechanges.

So, what kinds of additional pre­vention information do we need?

Initially we need to document:1. The kinds of educational, engi­

neering. and law enforcement ac­tivities we perform.

,~

1986 Volume 47, Number 3 17

of Arkansas. This figure depictsonly one aspect of their fire preven­tion program-law enforcement ac­tivities targeted at arson wildfires.

The axis pointing downward re­flects the quantity or amount oflaw enforcement activity (in thiscase the number of prosecutions,convictions, and settlements) inArkansas; the axis pointing to theleft depicts the cost of these en­forcement activities. The axis point­ing to the right indicates the numberof arson wildfires that can occurin Arkansas, the axis pointing up­ward shows the suppressioncosts (C) plus net value changes(NYC) due to these fires.

The lines of this graph representrelationships. The diagonal line inthe lower left quadrant, for example,represents the relations betweenlaw enforcement activities and thecosts of those activities. This lineshows that as law enforcement ac­tivities increase, the costs of thoseactivities will also increase.We're assuming for the purposesof this exercise that each unit ofenforcement, i.e., each prosecution,conviction, or settlement, costs Ar­kansas $500. The curved line in thelower right quadrant indicates thatat low levels of law enforcementeffort, Arkansas can expect a highnumber of arson fires, but as theState increases its enforcement ef­forts, it can expect a gradual de­crease in these fires. Finally, thestraight line in the upper right quad­rant indicates that as the number

18

of arson fires increases, the C +NYC will increase correspondingly.The curved line in this quadrantreflects Arkansas' total fire manage­ment costs, including not only theC + NYC caused by arson wildfiresbut also the State's prevention (lawenforcement) costs.

So, how would a manager inArkansas use this diagram? Let'sassume that the manager is aimingto undertake 1,000 prosecutions,convictions, and settlements againstarsonists next year. That level oflaw enforcement effort (E), couldbe expected to cost about $500,000(C). (Follow the dashed line fromletter E to letter C.) Given those1,000 units of enforcement (E),1,300 arson wildfires (A) could beexpected. (Follow the dashed linefrom letter E to letter A.) The totalfire management costs (T), includingprevention (law enforcement) costs,would be about $8.7 million-themost efficient level of operation (thelowest point of the curve). (Followthe dashed line from letter A to let­ter T and over to the C + NYCaxis.)

What if the manager's preventionbudget were cut in half, from$500,000 to $250,000? What effectswould that have? With a $250,000prevention budget (Enforcement Costaxis), the manager could affordabout 500 units of enforcement (i.e.,500 prosecutions, convictions, andsettlements). With this reduced levelof prevention activity, about 1,400arson wildfires could be expected.

The C + NYC due to these ad­ditional fires would be $8.5 million,and the total fire management costwould add up to $8.8 million (C +NYC + enforcement costs).

There's a lot more that we cando with this diagram from an anal­ysis standpoint, but I believe youcan get an idea of where we areheaded with our analyses of fire pre­vention programs. Although thisis a simplistic model, we're hopingto improve it greatly by collectingcost and prevention activity data foreach component of fire preventionso that we can more adequately re­flect what's happening in the realworld. If we can do that, then wecan turn this into an operationallyuseful system.

Of course, no matter how sophis­ticated we get, we can't do ourjob well unless we have good fireprevention information. Some ofthat valuable information will comefrom better reporting systems andsome from better fire cause inves­tigation. And with improved qualityof information, we scientists willbe able to give our clients betterproducts to work with. •

Fire Management Notes

..

The True Story of thePulaski Fire ToolJames B. Davis

Research forester, USDA Forest Service, Forest Fire andAtmospheric Sciences Research, Washington. DC

The nickel-plated pulaski looks asgood as new in its glass-fronted Col­lins Tool Company display case at theSmithsonian Museum of Arts and In­dustry in Washington, DC. Sur­rounded by equally shiny cutting toolsof all description, the pulaski was firstput on display at the Nation's Centen­nial Exhibit in Philadelphia in 1876.

Conventional wisdom holds that thepulaski fire tool was invented byEdward C. "Big Ed" Pulaski in thesecond decade of the 20th century. EdPulaski, a descendant of AmericanRevolution hero Casimir Pulaski, wasa hero of the Great Idaho Fire of1910, leading his crew to safety whenthey became imperiled. He was alsoone of a group of ranger tinkererswho struggled to solve the equipmentproblems of the budding forestry pro­fession. However, the pulaski tool ondisplay at the Smithsonian must havebeen made when Big Ed was no morethan 6 years old!

In the early days of forestry in thiscountry, fire tools were whatever hap­pened to be available. The earliestmethods of firefighting were confinedmostly to "knocking down" or "beat­ing out" the flames, and the toolsused in the job were simple and prim­itive. The beating out, when such anapproach was possible, was often ac­complished with a coat, slicker, wetsack, or even a saddle blanket. Acommonly used tool was a pine boughcut on arrival at the fire edge (4).

Soon farming and logging tools,available at general and hardwarestores, came into use. These included

1986 Volume 47, Number 3

the shovel, ax, hoe, and rake-all ba­sic hand tools developed over centu­ries of manual labor. Even afterfirefighting became an importantfunction of forestry agencies, thesetools were accepted as they were,wherever they could be picked up,and little thought was given to size,weight, and balance. There appears tobe no record of the use of the CollinsTool Company pulaski for fire con­trol. Most likely, it was sold to farm­ers for land clearing and may havebeen forgotten by the late 1800's (2).

With the advent of the USDA For­est Service and State forestry organi­zations, a generation of "ranger in­ventors" and tinkerers began toemerge. It became apparent that care­ful selection and modification was es­sential for efficient work and laborconservation. In the early days whenalmost everybody and everything hadto travel by horseback transportationwas a particular problem. For yearsforesters worked on the idea of com­bination tools. Most of the attemptswere built in home workshops, andmost "went with the wind." Two im­portant survivors, now in general use,are the Mcleod tool, a sturdy combi­nation of rake and hoe, and the com­bination of ax and mattock. TheMcLeod was probably the first firetool to be developed. It was designedin 1905 by Ranger Malcolm McLeodof the Sierra National Forest.

Who first invented the ax-hoe com­bination and used it for firefighting isa matter of minor dispute. Earle P.

Dudly claims to have had a pulaski­like tool made by having a light­weight mining pick modified by a lo­cal blacksmith. He says he used thetool for firefighting in the USDA For­est Service's Northern Rocky Moun­tain Region in 1907. Dudly was wellacquainted with Ed Pulaski, and thetwo had discussed fire tools.

Another account of the origin of thepulaski is that William G. Weigle, su­pervisor of the Coeur d' Alene Na­tional Forest, thought of the idea-butnot for firefighting (5). Rangers EdPulaski and Joe Halm worked underhim <all three became heroes of theGreat Idaho Fire) at Wallace, ro, thenheadquarters for the Coeur d'AleneNational Forest. At that time, planswere being made for some experimen­tal reforestation, including the plant­ing, pine seedlings. As SupervisorWeigle planned the job, he decided anew tool was needed to help with theplanting as well as other forestrywork. He decided on a combinationof ax, mattock, and shovel. One dayin late 1910 or 1911, Weigle sentRangers Joe Halm and Ed Holcomb toPulaski's home blacksmith shop totum out a combination tool that mightreplace the mattock that was then incommon use for tree planting. Halm,with Holcomb helping, cut one bladeoff a double-bitted ax, then welded amattock hoe on at right angles to theformer blade position. He then drilleda hole in an old shovel and attached itto the ax-mattock piece by means of awing bolt, placing it so the user couldsink the shovel into the earth by ap-

19

Figure I-Firefighter using pulaski on Plume Creek fire in the Kaniksu National Forest in Idaho.

plying foot pressure to the mattockblade.

The rather awkward device was nota success as a planting tool. Probablythe whole idea would have been aban­doned had not Ranger Pulaski beenfascinated with the possibilities of thetool. He kept using it, experimentingwith it, and improving it. He soondiscovered that the bolted-on shovelwas awkward and unsatisfactory. Heabandoned the shovel part and alsolengthened and reshaped the ax andmattock blades. It is too bad Pulaskidid not know about the Collins Toolpulaski-it would have saved him alot of time. Nevertheless, by 1913Pulaski had succeeded in making awell-balanced tool with a sharp ax onone side and a mattock or grubbingblade on the other.

Pulaski use now spread throughoutthe Rocky Mountain region. How­ever, it was used not for tree plantingbut for fire control. By 1920 the de­mand was so great that a commercialtool company was asked to handleproduction.

Although the pulaski went intowidespread use in the Rockies in the1920's, it saw little or no use in otherareas. Prior to 1931 the USDA ForestService had no good internal methodfor handling equipment developmentand promotion. Most new equipmentideas were introduced and discussedat the regular Western Forestry andConservation Association meetings(3, 7).

By the mid i930' s, with the adventof the CCC, fire tools began to prolif-

20

erate, and the USDA Forest Servicesought to standardize tools rather thandevelop new ones. It was at an equip­ment standardization conference atSpokane in 1936 that the pulaski toolwas proposed for national distribu­tion. The conference instructed theUSDA Forest Service's Region I todevelop and further test a prototypesuitable for servicewide use (6, 8).

Since "Big Ed's" day the pulaski,as well as other fire tools, hasundergone continual improvement.Pulaski development is an ongoing ef-

-,-----Ii.'!'

1t"",'. ~. ~, .•.

~" .

fort at the USDA Forest Service'sMissoula Equipment DevelopmentCenter. Careful engineering study,design, and testing have resulted instandards of shape, weight, balance,and quality (fig. I).

Although Ed Pulaski may not haveinvented the first fire tool put intogeneral use or even first thought ofthe tool that bears his name, he diddevelop, improve, and popularize thepulaski. The General Services Admin­istration now puts out bids for morethan 35,000 new pulaskis each

Fire Management Notes

'I

I

li'

.,1

year-a long way from the prototypeso laboriously made in RangerPulaski's home blacksmith shop (1).

ii

Literature Cited

I. Daffern, Jerry. Fire suppression equipmentfrom GSA. Fire Management. 36(2):3--4-;1975.

2. Gisbom, H.T. Forest fire-a mother of in­vention. American Forests and Forest Life.32(389);265-268; 1926.

1986 Volume 47, Number 3

3. Goodwin, David P. The evolution offirefighting equipment. American Forests.45(4);205-207; 1939.

4. Graves, Henry S. Protecting the forest fromfire. Forest Service Bulletin. 82; 1910. 48p.

5. Huh, Ruby E. How the pulaski became apopular tool. The Regional ForestRanger-Part Four. Article on file: U.S.Department of Agriculture, Forest Service,Washington Office, History Section.

6. Lowden, Merle S. Equipment developmentand fire research. Fire Control Notes.19(4);127-129; 1958.

7. Osborne, W.B., Jr. Fire protection equip­ment accomplishments and needs. Journalof Forestry. 29(8): 1195-1201; 1931.

8. Pyne, Stephen J. Fire in America.Princeton, NJ: Princeton University Press;1982. 654 p.

21

Evaluating Arson-Caused Forest Firesin Wisconsin, 1982-85EarlMeyeeer

Staff forest ranger, Wisconsin Department of NaturalResources, Madison. WI

,•

20

15

demographic data were gleaned fromtheir records.

Of the 361 records examined,201 or 55 percent indicated that thefires were set within 6 miles ofthe responding fire station. Almost19 percent of the arsonists set fireswithin I mile of the station (fig. I).Since most of our ranger stationsare located in rural communities thisinformation may indicate that thefire setters live in or near town.

The majority of the arson fires(5I.7 percent) were discovered bylocal residents. This fmding is con­sistent with the percentage of allfires discovered by the local citi-

Figure I-Distance from fire station of arson-causedforest fires in Wisconsin. 1982- 85.>

2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20Milesfrom fire station

Percent30,-----------------------,

25

3. Less probable-Form of ignitionis established by weak circum­stantial evidence, by processof elimination, or by fire historyof the area and experienced judg­ment of the investigator.

4. Undetermined-No definite cluesor several probable forms of igni­tion or fire not investigated.Only fires that were coded certain

or reasonably certain were selectedfor the study. Three hundredsixty- one records (64 percent) werecoded certain or reasonably certainby the investigating officers. Thepersons responsible for 54 arsonfires were also identified, and some

Forest fire arson is a problemof some concern for forest fire con­trol people in Wisconsin. In thepast 4 years, 567 forest fires wereintentionally set. Wisconsin forestryofficials consider this number tobe entirely too high, especially con­sidering that the figure covers onlythose lands protected by the Depart­ment of Natural Resources (DNR),or roughly one-half of the State landarea. Fire responsibility for the re­mainder of the State lies with eitherthe township or municipal govern­ment.

Forest fire arson, for purposesof this article, consists of fires will­fully set by anyone to burn vege­tation or property not owned orcontrolled by the fire setter andwithout the consent of the owneror owner's agent (I).

Wisconsin forest rangers reported567 arson wildfires in the periodfrom 1982 through 1985. These fireswere sorted as to degree of certaintyof fire cause according to the fol­lowing criteria (2):1. Certain-Form of cause is estab­

lished by admission, statementof reliable witness, or physicalevidence. This category is in­tended to cover cases where causeis established beyond doubt.

2. Reasonably certain----{:ause isestablished by strong circumstan­tial evidence. This categorycovers cases where cause is rea­sonably certain, but witness state­ments or physical evidence maynot be conclusive.

22 Fire Management Notes

II)

Figure 2--A.rson-caused fires in Wisconsin according to month. 1982-85.

13.6%14.8%

18.4%

14.8%

23

21.2%

9.7%7.5%

Saturdays, peak on Sunday, andthen generally decrease to the lowof 7.5 percent on Friday. In muchof the DNR's jurisdiction, the burn­ing of trash on Sundays is pro­hibited, following the premise thatcareless campers, hikers, sightseers,and other recreationists would keepthe frrefighting forces busy. It ap­pears that, with the demonstratedpeak of arson activity on Sundays,the banning of trash burning maybe even more justified.

The pattern for time of day forfire starts is interesting. Fire settingactivity seems to begin at about1200 hours, peaks at 1500 hours,decreases to a low point at 1700hours, and then rises again to alesser peak at 1900 hours. Fire set­ting then decreases to a low pointat 2200 hours, at which time over

Figure 3-Arson-caused fires JIJ Wisconsinaccording to day of the week, 1982-85.

quent fire runs by the fire agencies,ease of ignition of forest fuels atthis time of year, and the balminessof the weather, combine to stimulatefire selling behavior.

The day of the week that themost fires occurred was somewhatsurprising. Sunday ranked highest,with an average of 21 percent occur­ring on that day; 35 percent oc­curred on Saturday or Sunday.Forty- eight percent of all the firesoccurred on the first 3 days of theweek-Monday (14.8), Tuesday(18.4), and Wednesday (14.8).Thursday and Friday ranked thelowest at 9.7 and 7.5 percent, re­spectively (fig. 3).

Arson fires seem to pick up on

zenry as opposed to lookout towers,aircraft, or other means.

Almost 60 percent of the incendi­ary fires occurred in April, and bythe end of May, more than 75 per­cent of the State's incendiary fireshad been set (fig. 2). There appearto be two rather significant peaks.One is an II-day period from April24 through May 4 when 27 percentof this type arson fires were set.Another peak occurred during a9-day period from April 9 to April17, when 19 percent of the fireswere set.

April and May are Wisconsin'sbusiest forest fire months both innumbers of fires and acres lost. Per­haps several factors, such as fre-

1986 Volume47, Number 3

i(I

I Percent

January .03

February 0

March 3.3

April 58.9

May 17.2

June 4.6

July 4.4

August

September 1.1

October

November .3

December .3

Time of day (24-hour clock)1

23456789

1011

Noon 121314151617181920212223

Midnight 24 l-----'-.l__---L_---Jo 5 10 15

Percent

Figure 4-Arson-caused fires in Wisconsinaccording to rime of day, /982-85.

90 percent of the incendiary fireshave been set (fig. 4). Perhaps thedip in the late afternoon reflectssome chores that have to be accom­plished and time out for the eveningmeal. The increased early eveningactivity may indicate increased idletime after the evening meal or adesire to re-create the emo-tional"high" attained by the afternoon ar­son or perhaps both. This daytime

24

fire setting pattern is contrary to thegeneral belief the public has thatarson is a crime of the night. Ourexperience shows that wildland arsonoccurs in broad daylight during themost dangerous burning time, whenour firefighting forces are busiestwith accidental fires and when theset fires could potentially do themost destruction.

Another myth was brought downwith this study. It had generallybeen felt that State-ownedlands were targets for the arsonists.This proved not to be the case. Pri­vate landowners suffered the worstof the attacks-more than 57 percentof the arson fires occurred on pri­vate land. Federal lands weresingled out next with 15.3 percent,then State lands with 7.4 percent.The remaining fires were on county,municipal, and various highwayrights-of-way (fig. 5).

Of the 361 records that were stud­ied, persons responsible for 54 fireswere identified, The demographicsdidn't produce any surprises: 97 per­cent were male and 63 percent were25 years of age and under. How­ever, 24 percent were between 38and 56 years old. There seem tobe two main age groups involvedin fire setting: 25 and under andthe late 30's to early 50's (fig. 6and 7).

The people that set fires are gen­erally not transients, vacationers,or summer people. Eighty percentwere identified as permanent res­idents, the group most likely to be

57.2%

15.3%

Figure 5--Ownership of land where arson­caused forest fires originated, 1982-.'15.

Female %

Figure 6----Sex of known forest fire arsonistsin Wisconsin, 1982-85.

injured from a forest fire gone awry(fig. 8).

In conclusion, in Wisconsin arsonforest fires are more likely to occurduring two short periods in April

FireManagement Notes

.'

Ii

35%

28%

10-18 19-25 26-35 36-45 46-56Age (years)

F1gure 7-Age distribution of known forestfire arsonists in Wisconsin, 1982-85.

and May, During these two periodsincreased vigilance may well bewarranted and productive, Fires tendto be set during the day in earlyafternoon, on private lands within6 miles of the ranger station, Theperpetrator will probably bemale, either less than 25 or between38 and 55 years old, His activitywill be most likely on Sundays orSaturdays,

Perhaps the information providedby this study, along with personalknowledge of local conditions, willaid in selling up patrol routes andsurveillance procedures that will re­sult in fewer arson fires and theapprehension of arson suspects, •

2%

Literature Cited1. U.S. Department of Agriculture, Forest

Service. Glossary of terms used in forestfire control. Ag. Handb. 104. Wash­ington, OC: U.S. Department of Agricul­ture; 1956, 24 p.

2. Wisconsin Department of Natural Re­sources. Individual forest fire report hand­book. No. 4305.1. Madison. WI: Wis­consin Department of Natural Resources;1981. p. 40-7,

Figure 8-Residency status of known forestfire arsonists in Wisconsin, /982-85.

1986 Volume 47, Number 3 25

Wildland Firefighters PersonalProtection GearArt Jukkala and Ted Putnam

Forester and equipment specialist, respectively, USDAForest Service, Missoula Equipment DevelopmentCenter, Missoula, MT

'I

Properly equipped wildland firefighter in the process oj deploying fire shelter.

Introduction

Wildland fires trapped an esti­mated 200 firefighters during the1985 fire season. These entrapments,and the role fire shelters and otherpersonal protective equipment playedin saving lives and preventing seri­ous bum injuries, have stimulatedgreat interest. The purpose of thisarticle is to provide field peoplewith additional information on thisvital safety equipment.

Fire shelters, special clothing,and other protective equipment existtoday hecause 28 years ago, afteran analysis of casualty-causing fires,the Forest Service committed itselfto developing effective personal pro­tective gear for wildland firefighters.The Missoula Equipment Develop­ment Center (MEDC) has spear­headed this effort over the yearsand continues to improve the itemsthat make up the firefighter's per­sonal protective system.

Personal Protective System

The basic components of the sys-tem include:

• Fire shelter• Hardhat• Goggles• Flame-resistant shirts and jeans• Leather boots and wool socksThe fire shelter provides protec-

tion during fire entrapment and pro­tects vulnerable airways and lungs,which the other components ofthe system cannot do. We wroteabout the fire shelter in the last

26

issue of Fire Management Notes,so here we will focus on the other

items of personal protective equip­ment (PPE).

Fire Management Notes

,\

)

1I

In

Together, these items protect fire­fighters from thermal and other jobhazards while they work. Severalkey requirements guide the MissoulaCenter's development and test workon PPE. Each item must protectthe wearer and not contribute to fa­tigue or injury by premature failure.In addition, it must be lightweight,functional, durable, comfortable, andeconomical. With regard to comfort,heat stress is a prime concern. Whenwe speak of economical equipment,we don't necessarily mean thelowest in price. We look at anitem's life-cycle cost, not just ac­quisition cost. A good example isflame-resistant clothing. A garmentthat costs twice as much as anothermay still be more economical ifit outwears the less expensive gar­ment by a factor of more than twoto one.

Hardhat

With the possible exception ofthe fire shelter, the hardhat is themost important piece of wildlandfuefighting safety equipment. Hard­hats have saved many lives andprevented serious injuries by protect­ing the wearer against falling trees,limbs, and rolling rocks.

The literature indicates that ap­proximately 15 percent of the body'sheat is lost through the head, sohardhats, which are cooler andlighter in weight, are preferred overhelmets designed for structural fire­fighting. Special clips are addedto attach goggles and night firefight-

1986 Volume 47, Number 3

ing headlamps. As wildland fire­fighters confront more fires in areaswhere there are structures and as­sociated electrical hazards, class Bplastic hardhats, which provide elec­trical hazard protection, are pre­ferred.

Goggles

A study conducted from 1967through 1971 showed that eye inju­ries accounted for about 7 percentof all fire suppression injuries. Dust,smoke, brush and branches, andhot substances cause most eye inju­ries.

Thus goggles are an importantcomponent of the PPE system. Fire­fighters are dissatisfied with thegoggles now available. A recentsurvey indicated comfortable gogglesthat won't fog or scratch are oneof the firefighter's greatest needs.We hope to begin a project aimedat finding better fog-free gogglesfor forest workers that will meetfirefighting needs.

Development of Flame-ResistantClothing

Work on flame-resistant clothingbegan in 1958 concurrently withfire shelter development. The objec­tive was to design garments thatwould offer firefighters protection(1) against flames, falling embers,and coals; (2) when dashing forsafety to avoid entrapment; (3) dur­ing entrapment without a fire shelterand added protection within a shel-

ter; and (4) if involved in an aircraftaccident.

Extensive use of flame-resistantshirts began in 1962 with the intro­duction of flame-retardant treated(FRT) cotton shirts. FRT cottontrousers were field evaluated alongwith shirts. But trousers are subjectto greater wear, and FRT cottonhad poor durability, so flame-resist­ant trousers were impractical untilNomex" fabrics became available inthe early 1970's.

After using orange FRT cottonshirts for several years, we switchedto yellow shirts in the late 1960's.Studies on PPE for urban fire­fighters showed yellow to bemore visible in dark and smokeyenvironments. Also, there were sev­eral incidents in which the orangeshirts were mistaken for flames andaircraft dropped fire retardant oncrews on the line.

Firefighters wore FRT cottonshirts for about 10 years. Then in1973, after many years of field eval­uation, we adopted flame-resistantNomex" for shirts. Nomex" iswoven from a nylon-type fiber de­signed to withstand high tempera­tures. Like most fabrics, Nomex"bums if exposed to flame, but un­like the others, it stops burningwhen removed from the flame. In­stead of melting or burning to ash,it forms a char that continues tohelp protect the skin. In 1974,Nomex'" pants became available.We've worked to improve thesegarments over the years.

27

Flame-Resistant Shirt

MEDC specialists worked withthe producer of Nomex" fiber andseveral fabric weavers for 3 yearsto develop a more comfortableflame-resistant shirt fabric. The newfabric has a more open, thickerweave. Weight and fire resistancyremain the same. Forest Servicefield evaluators said the new shirtsare cooler on hot days and warmeron cool nights. The openness ofthe weave improves evaporativecooling while its thickness traps airnext to the skin to keep the wearerwarmer at night.

The fabric feels softer, too, eval­uators said. The softer fabric shouldreduce chafing problems some peo­ple experienced with the previousNomex" shirts.

To insure a better fit for morepeople, the shirts are being madein a new extra-small size.

Key features of the earlier designhave been retained. Sleeves are cutfull for ventilation plus free move­ment of tbe arms. This reduces fab­ric stress at the elbows and improvesdurability of the shirt. For addedprotection against radiant heat andfalling embers, the collar can beturned up. Sleeves can be drawnsnugly around the wrists with quick­fastening Velcro" closures. Buttonsare oversized for easy fastening.Two large pockets with flaps holdmaps, note pads, and similar items.The shirt can be worn inside thetrousers, or outside of the jacket.Wearing the shirt inside offers the

28

most thermal protection.

Flame-Resistant JeansThe firefighters pants were re­

designed into a basic jean cut. Forgreater comfort, they have a littlefuller cut in the crotch, upper leg,and seat. The new jean design fea­tures western scoop front pocketsand a return to hook and pile tabson the back pockets. These pocketsremain large for carrying extra per­sonal items and to increase thermalprotection with a double layer offabric.

The zipper fly was retained, butthe waistband is now closed witha traditional metal jean rivet button.The button is embossed with theletters FSS for Federal Supply Sys­tem.

These jeans also come in twonew waist sizes: 26 inch and 40inch. And there is a choice of twoinseam sizes, 3D and 34 inches.For those who need 31- or 32-inchinseams. it is easier to let out the3D-inch inseam than to cut and re­hem the 34 inch. The 34-inch in­seam can be let out to inseams of3S to 36 inches.

What To Wear with Nomex"

The other clothes worn withNome,," shirts and jeans affect theirprotective qualities. Underwear ofa polyester cotton blend is accept­able, but we recommend T-shirtsand undershorts of 100 percent cot­ton. AU-synthetic underwearshould never be worn. Two layers

of clothing, that is Nome,," plusunderwear, provide better thermalprotection. But don't wear otherwork clothing under or over yourNomexe garments, Doing so in­creases body heat and puts an addedload on the heart.

For colder weather and nights,jackets should be all wool, all cot­ton, or wool blends of at least 85percent wool.

Coveralls or Jumpsuits

Firefighters in some agencies wearNomex lltl or FRT cotton coverallsor jumpsuits, and we are often askedwhy the Forest Service doesn't usethem.

We favor shirts and pants anddiscourage coveralls and jumpsuitsfor these reasons: (I) Coveralls areusually worn over other garments.This restricts body cooling,which results in higher heart ratesand increased heat stress. (2) Cover­alls are made from a single-weightfabric, but ideally different weightfabrics are needed for the upper andlower parts of the body; for exam­ple, a heavier, more durable fabricto protect legs from brush and gen­eral abrasion, and a lighter-weightfabric with a more open weave tohelp dissipate heat ftom arms, chest,and back. One-piece garments ofa heavy trouser-weight fabric, likework coveralls, restrict cooling.They also chafe parts of the upperbody if worn only over T-shirts.Jumpsuits are made from shirt­weight fabric and do not provide

Fire Management Notes

'I

I

",I

.h

Ic.

adequate leg protection. (3) Fire­fighter's clothing is produced ina limited number of sizes to keepcosts down. With such a limitedsize distribution, one-piece garmentsseldom fit many people properlybecause the distance between crotchand shoulders varies so much. Ina work situation this causes chafingand discomfort. (4) One-piece gar­ments made from flame-resistantfabrics are expensive. Unlike shirtsand jeans, if part of the garmentbecomes unserviceable, the entiregarment must be replaced.

Flame-Resistant Gloves

Specially designed gloves areessential in protecting the fire­fighter's hands against blisters,scratches, small cuts, and minorbums during routine frrefighting.But they also play a major fire pro­tection role in event of an aircraftaccident or fireline entrapment.

Reviews of past fire entrapmentsshow that those trapped burnedwhile attempting to flee entrapmentsometimes lost fingers if theywere not wearing gloves or if theirconventional oil-tanned work glovesburned or shrunk from intense heat.Additionally, reports from peopleentrapped in shelters emphasize theimportance of gloves in holdingdown hot shelter material withoutgetting burned.

After numerous material and de­sign changes, MEDC's late-st modelforest workers' gloves appear tobe meeting firefighting and general

1986 Volume 47, Number 3

forestry requirements very well.The new gloves are full-grain,

chrome-tanned leather; previousmodels were split-grain leather. Full­grain leather is stronger, so the newgloves wear better and last longer.Full-grain leather requires narrowerseam margins to hold stitches, whichreduces bulk. Compared with oiltanning, chrome tanning providesexcellent protection against heatand shrinkage. These gloves alsoresist stretching and retain theirshape better wet or dry because theleather is water and oil resistant.

A drawstring wrist closure keepsembers and debris from getting intothe gloves.

Leather Boots and Wool Socks

Past fire entrapment investigationsfound that good quality leather bootstraditionally worn for wildland fire­fighting provide adequate foot pro­tection. All-wool or mostly woolsocks offer added thermal protection.Wool wicks moisture from the skin.This helps keep feet cooler anddryer, reducing the chance of blis­ters, a common firefighting injury.

The Forest Service requires thefirefighter to wear a lace-type leatherboot with at least an 8-inch top.Skid-resistant soles are required,with a lug-type sole preferred. Slipsand falls account for many fire­fighting injuries. One study overa lO-year period indicates that 17percent of injuries result from slipsand falls. So the importance of goodskid-resistant soles cannot be over-

emphasized.

Chain Saw Chaps

One last piece of PPE that we'dlike to review is chain saw chaps.Since not all firefighters wear them,we don't include chaps as a com­ponent of the basic PPE system.However, due to the extensive useof chain saws in firefighting, wefeel a brief review of this item isappropriate.

Chain saw chaps were redesignedin 1982. The redesign focused onincreasing sawyer protection andcomfort while minimizing costs.

MEDC tested many combinationsof materials to find the best pro­tection with the least weight. Thenew outer shell is II-ounce Cor­dura" nylon, replacing the bulky 15­ounce cotton canvas. Cordura'"cleans easily, resists tears and abra­sions, and keeps the protective padfree of oil better than cotton canvas.The protective pads combine twolayers of woven Kevlar" with twolayers of Kevlar" felt. Kevlar" isan aramid fiber like Nomex", butwith more flame resistance. More­over, because of its cut resistance, itcan slow and quickly jam the chainbefore cutters penetrate the leg. At 2pounds, the new chaps weigh 40percent less than the old style andoffer 50 percent more protection.

Quick-release buckles make thechaps easy to put on and take off.This is a nice feature for firefightingbecause the chaps can be easilyremoved to reduce heat stress when

29

sawing is completed. Leg strapshave been relocated to improve fitand minimize snagging, and a toolpouch was added to hold a roundfile, file guide, flat file, spark plugwrench, and screwdriver. A pamph­let titled "Inspecting and RepairingYour Chain Saw Chaps" is availablefrom the MEDC.

Respiratory Protection

Even though the personal protec­tive system doesn't include any kindof breathing device, we want todiscuss briefly respiratory protection.

As we pointed out above, thefire shelter protects the firefighter'sairways and lungs during fire entrap­ment. But what about protectingthe working firefighter's respiratorysystem from smoke, dust, and hotgases?

Periodically, MEDC has beenfunded to search for effective, prac­tical respiratory protection that

30

does this. Self-contained breathingapparatus used by municipal andindustrial firefighters are effectivebut impractical for wildland fire­fighting. The Center has testedvarious types of respirators, andall restrict breathing and don't filtergases well, exposing the wearerto high concentrations of poisonouscarbon monoxide and other toxicgases. For this reason, we haven'tbeen ahle to recommend respirators.

For limited smoke and dust pro­tection, firefighters have worn ban­dannas for years and are beginningto use disposable dust filters.Bandannas or dust filters shouldbe kept dry. In intense heat, aswhen working against a hot flamefront, there is a possibility thatbreathing hot, moist air through awet dust filter or bandanna candamage the respiratory system.

Furthermore, firefighters shouldbe cautioned against covering too

much of the face. Cheeks and earsare excellent heat sensors. Coveringthem can lead a person to worktoo long in a hot situation. The re­sult can be dehydration, heat stress,and prolonged elevated heart rateleading to premature fatigue, orworse.

Conclusion

Improvement of personal protec­tive equipment is a dynamic process.Materials and production technologychange constantly. And newknowledge, particularly in the arenaof human physiology, is expandingrapidly. The staff at the MissoulaEquipment Development Centerkeeps up with these changes so asto continue to improve personalprotective equipment for the wild­land firefighter. We always welcomecomments on product improve­ments.•

Fire Management Notes

.-~

I

i

\.~

Using Interactive VideodiscTechnology in Wildland Fire BehaviorTrainingM.J. Jenkins and K.Y. Matsnmoto-Grah

Assistant professor and researchassistant, respectively, Departmentof Forest Resources, Utah StateUniversity, Logan, UT

,I' .

In recent years, fire research hasdeveloped a large body of complexinformation. Proper wildland firemanagement and use requires a thor­ough understanding of combustion,fire behavior, fire ecology andhistory, fire economics, fire effects,and prescribed fire practices. Largeareas of valuable resources andhuman lives are often at stake dur­ing prescribed fire and fire suppres­sion activities, making thorough,up-to-date professional training es­sentiaL Because safety is such anoverriding concern, adequate trainingis mandatory to reduce risk.

Most fire management organiza­tions require approximately 40 hoursof formal training as a basic prere­quisite to any entry level suppressionposition. The curriculum includesfundamentals of fire behavior, firesuppression, fire organization, tooland equipment use, and safety. Firebehavior is central to this curricu­lum, and a subject taught by all firemanagement agencies. A complete,multilevel instructional packagedeveloped by the National WildfireCoordinating Group (NWCG) isthe training standard in fire behaviorfor all NWCG agencies.

Our project involves using exist­ing fire behavior curricula in aninteractive Videodisc/computer train­ing system. This disc will supple­ment existing training courses inwildland fire behavior and will in­clude materials produced by NWCGon introductory, intermediate, and

1986 Volume 47, Number 3

advanced fire behavior. In addition,elements from existing courses onfuels inventories, debris prediction,and prescribed fire will be includedon the program (9).

The objective of this project isto demonstrate the application ofinteractive videodisc instruction towildland fire management byproducing a program on wildlandfire behavior. This training packagewill not replace existing instructionalmaterials, but will act as a supple­ment to current training materials.The project is funded by NWCGand is being administered by Bureauof Land Management, Division ofTraining, Boise Interagency FireCenter.

Interactive Training andInstruction

Basic principles of interactivedesign must be clearly understoodin order to realize the benefits whichcan occur through the use of aninteractive training medium. In orderto provide useful interaction betweenthe learner and the instructionalprogram, intellectual options shouldbe provided that allow users to ac­tively make decisions and be sub­jected to their consequences (1). Forexample, an incorrect calculationof a fire's rate of spread in a givensimulation may result in inappro­priate tactical response to the inci­dent. Learners make choices, andthe system responds, sometimes in

surprising ways. An interactive vid­eodisc lesson should allow learnersto "create" their own training expe­rience, to interview and make fre­quem decisions about the content ofthe lesson and the way it isdelivered.

The instructional system respondsinstantly to learner input by pro­viding relevant, previously designedinstructional cues, reinforcement,and feedback segments composedof computer text and graphics, stilland motion video, with or withoutaudio. The scope, sequence, rate,style, duration, level, and mediumof instruction is, to a large degree,determined by a dialog betweenthe learner and the system.

The linear, presequenced formatof most traditional instructional ma­terials such as texts, workbooks,films, or slide/tape presentationstreats all learners alike and does notallow the learner to actively takepart in the choice of sequence orcontent of a lesson (5).

Interactive video is a powerfulmedium, bringing together the emo­tional impact of video and film andthe interactive capabilities of thecomputer. Development of interac­tive videodisc instruction has, toa large degree, mirrored computer­assisted instruction (CAl). Like CAl,interactive videodisc instruction canbe used as a tutorial, for drill andpractice and in simulations, as wellas for information storage.

31

Figure I-Linear versus interactive instruction. A linear program has a predetermined format,whereas an interactive program allows the learner to make choices.

EndSummary

FireTriangle

Combustion

there have been disruptive weak­nesses in the quality of video imag­ery and search time for sequences.The introduction of videodisc tech­nology has been an obvious andpractical solution to these problems.

Flowcbarting. The branchingsystem developed for the instructionmust be graphically represented asa flowchart. The flowchart depictsall instructional events that will beincluded in the wildland fire be­havior training program for thevarious specified audiences. Theflowchart shows each instructional

InformationIntroduction

is made possible through a processcalled branching. A branching pro­gram has alternate tracks for therapid and slow learner, can followup certain responses in detail, allowlearners to see only the materialsthey need or want to see, test learn­ers on the comprehension of thematerials, allow learners to repeatmaterial or to have remedial workwhen needed, and effectively pose aquestion without demanding oneunique, correct answer. Althoughbranching has been extensively usedin computer-assisted instruction,

Interactive

Unit

linearDesigning Interactive VideodiscInstruction

The process of designing, writing,directing, and producing an interac­tive videodisc/computer lesson inwildland fire behavior is very similarto the basic design and productionof an instructional film, However,the major challenge is the integrationof various media, designing thedynamic relationship between userand the system, and working with amodular, rather than a linear me­dium (4).

Instructional Matrix. A contentmatrix for the instructional packagein wildland fire behavior was ini­tially developed from a careful anal­ysis of program goals and objec­tives, existing course materials,learner characteristics, and the vari­ous forms of media used withinthe disc to be selected. This matrixlists the appropriate instructionalsegments classified by topic and in­tended learner group in a matrixstructure. The matrix thus acts asa "map" to guide the designer increating the dynamic flowchart thatrepresents the interactive branch­ing training program.

Interactive instruction is designedto be modular and nonlinear. withhundreds of preplanned optionsavailable to learners based on priorknowledge of their unique interests,abilities, feedback preferences, pri­mary language ability, and leamingstyles (fig. 1).

Branching, Videodisc interactivity

32 Fire Management Notes

']

'"IJ

I

IJ

frame, both still frames and motionsequences, all branching options,menu driven and under programcontrol, all "help" routines, all testitems, and dedicated jump forwardand review options.

The procedures used in flowchart­ing interactive instruction are similarto those used in any flowchartingoperation. Symbols are very nearlythe same, with a few variationsto allow for easy identification ofthe nature of the instructional frame,whether it is a still of freeze frame,or motion sequence. A sample flow­chart is shown in figure 2 to demon­strate the complexity involved inthe development of an interactivelesson in wildland fire behavior.

Script, Narrative Writing, andGraphics. After the major elementsof flowchart construction are com­pleted, scripts and narratives mustbe written, including all text, com­puter and quiz screens, and testitems. Special terms are used tohelp integrate the various media,and keep track of hundreds of in­structional segments that will beused to individualize instruction fora variety of learners. Scripts andstoryboards are then drafted and re­viewed for content, style, and feasi­bility; graphics, animation, andspecial effects are also planned atthis time. Formative evaluation andrevision are also appropriate atthis stage of design.

Preproduction. The preproductionphase of an interactive videodisc

1986 Volume 47, Number 3 .

is very similar to that of a film orvideotape. However, there are somesteps unique to the videodisc me­dium, including the mapping of the"geographical" layout of the disc,writing of the accompanying com­puter program, and locating the sim­ple time code numbers for the videoframes from the master tape. Thesesteps must be completed beforethe final production and the master­ing of the videodisc (4, 11).

Production details will not bediscussed, as basic production fora videodisc is almost identical tothat of other media such as vid­eotape or film. Once a master vid­eotape of the training materials isedited and contains all instructionalsequences, the tape is sent to oneof five mastering houses, where thevisual images and audio are trans­ferred onto a videodisc.

Hardware and Software Consid­erations

Intaractive videodisc training isorientad toward the developmentof learning stations for individualand small group learning. Interactiveinstruction can take place in anylocation where an individual orsmall group can have access to acomputer that controls several pe­ripherals, including a videodiscplayer and monitor (fig. 3 and 4).Input to the computer can bethrough a touch screen, light pen,or keyboard. The video output from

the computer, as well as video/audiofrom the. disc, appears on a singlescreen or two separate screens. Thedelivery system chosen for thisproject uses one screen, allowingboth video from the videodisc, andtext over the video, which is gener­ated by the computer. "Windows"of video can be opened to illustratea point described by computer text,or lettering can be written over avideo image.

The hardware configuration se­lected for this project includes anIBM PC computer, Pioneer LDV-lOOO videodisc player, and Zenithcolor monitor. Programming is beingdone using an authoring systemcalled "Quest."

Expected Benefits

Interactive videodisc technologyhas proven to be a powerful educa­tional tool, using computer-basedlearning systems that can dis-play high-quality video imagery,provide rapid access to images, andutilize quality audio. The individu­alized, self-paced format representedby videodisc instruction focuseson learner needs, rather than a--pre­determined pace and sequence. Alearner is able to choose amonga variety of instructional options,and learning is directed accordingto performance and measurements ofunderstanding. Interactive videodiscallows learners to have remedial

33

t

(Still Frame)

20M 1,433-2,984IntroductionToWildland Fire

Behavior

(Motion Segment)

What) I (Level •,

Incorrect

Correct

-~

!!iilf

Iaz

i

Figure 2-A sample flowchnrt showing the complexity of the development of an interactive.lesson.

......, &

'.\

Because of the huge storage capac­ity of a videodisc, beginning, inter­mediate, and advanced fire behaviortraining courses can be placed onone disc, allowing the learner toseek remedial help when necessaryor go on to advance material whendesired. Fire behavior simulationexercises can also be included, al­lowing learners to apply their knowl­edge of fire behavior in lifelikescenarios requiring management de­cisions. Such simulations are moreversatile and "transportable" thanpreviously developed fire simulators,and can be used in small or remotecenters or for individual use forreview or training. The incorporationof tutorial lessons, as well as drilland practice, test items, and infor­mation storage make interactivevideodisc an efficient, space-savingmedium for instruction and a poten­tially powerful training tool for firemanagers and personnel.

Monitor

Microcomputer

'~lflJ ----.

D

o

Player

Figure 3-Hardware components of an interactive videodisc/microcomputer system. Literature Cited

IJ

work if needed or skip ahead ifprior mastery is demonstrated.

Other benefits of using interactivesystems include; the use of programsby either individuals or smallgroups, flexible scheduling of train­ing, ease of transport of instructionto remote training sites, and trainingcosts much lower than for traditionaltraining methods.

Recent studies i.ll the comparisonof videodisc instruction to traditionalmethods of instruction have shown

1986 Volume 47, Number 3

numerous advantages from theuse of interactive videodisc. Suchfactors include; savings in actuallearning time, higher mastery rate,and favorable response to the me­dium. Other cost-related benefits areless time spent away from the job,reduced travellliving expenses fortraining, reduced need for classroominstructors, and use in remote train­ing stations (6, 7).

Fire behavior training naturallylends itself to the videodisc medium.

1. Bosco, J.1. Interactive video: Educationaltool or toy? Educational Technology.24: 13-19; 1984.

2. Clark. J.D. How do interactive videodiscsrate against othermedia? InstructionInnovations. 29: 1-16; 1984.

3. Cunier, R.L. Interactive videodisc learn­ing systems. High Technology. 3: 51­59; 1983.

4. Daynes, R. The videodisc interfacingprimer. Byte. 7: 48-59: 1982.

5. DeBloois, M.; Bennion, J.L.; Bork,A.; Brandy, R.C.; Knapp, B.ff.; Willis,B.D.; Wood, R.K.; Woolley, R.D.Videodisc/Microcomputer courseware

35

Figure 4---Videodisc training systems can be used either by individuals or small groups workingtogether.

·36

design. Englewood Cliffs, NJ; EducationalTechnology Publications; 1982. 178 p.

6. Ebner, D.G.; Danaher, B.G.; Mahoney,J.V.; Lippert, H.T.; Balson, P.M.Current issues in interactive videodiscand computer-based instruction. Instruc­tion Innovations. 29: 24-29; 1984.

7. Glenn, A.D.; Kozen, N.A.; Pollak.R.A. Teaching economics: research find­ings from a microcomputer/videodiscproject. Educational Technology. 24: 30­34; 1984.

8. Hon, D. Interactive training in cardio­pulmonary resuscitation. Byte. 7: 108­138; 1982.

9. Jenkins, M.J.; DeBloois, M.; Matsumoto­Grab, K.Y. Future frrefighting-by bookor by screen? Training andDevelopment Journal. 39(7): 36-39;1986.

10. Kindleberger. C.P. Whither the interactivevideodisc? Educational TV. 9: 60-65;1982.

11. Nugent, R.; Christie, K. Using videodisctechnology. Video Systems. 2: 16-21;1982.

12. 51. Lawrence, J. The interactive vid­eodisc: Here at last. Electronic Learning.4: 49-54; 1984.

Fire Management Notes

I!

Florida State Forester John Bethea accepts the Golden Smokey Awardfrom Regional Forester Jack Alcock. Mr. Bethea received this award inrecognition of his total commitment to protecting America's forests from wildfire.

1986 Volume 47, Number 3 37

United StatesDepartment of Agriculture

Washington, D.C.20250

OFFICIAL BUSINESSPenalty for Private Use. $300

III II IU.S.MAlL

oc

Postage and Fees Paid

United StatesDepartment of Agriculture

AQR·,01

~,

J.

I'

I

ORDER FORM To: Superintendent of Documents, U.S. Government Printing Office, Washington, D.C. 20402

~Credit Card Order. Only

Enclosed is $ 0 check, VISA· Total charges $ Fill in the boxes below.o money order, or charge to myDeposit Account No. Credit r-r-r-r-r-r-r-',-,-,-,-,-,-,-,-,-,-,-.-..,--

ITIITITI-D [Ee'lwic..n:t 1 Card No.tt IwcCord Expiration Date ITIIJ

Order No. Month/Year

Please enter rn_y subscription to Fire Management Notes at $13.00 domestic, $16.25 foreign.(Prices are subject to change without nottce.)Make checks payable to Superintendent 01 Documents.

Company Of personal name

PLEASE PAINT OR TYPE

For Office Use Only.

Quantity Charges

Enclosed

To be mailedSubscriptions

PostageForeign handling

MMOBOPNR

UPNS

Discount

Refund

1II..I'