guidance on spatial wildland fire analysis - fs.fed.us · share a common foundation of procedures...

United StatesDepartmentof Agriculture

Forest Service

Rocky MountainResearch Station

General Technical Report RMRS-GTR-183

December 2006

Guidance on Spatial Wildland Fire Analysis:

Models, Tools, and Techniques

Richard D. Stratton

The Author ______________________________________Richard D. Stratton is a fire modeling analyst with Systems for Environmental Management (SEM), a nonprofit research and educational corporation in Missoula, Montana. His professional interests include fire modeling, wildland fire use, fuel treatment design, the wildland-urban fire problem, and wildland fire decision support systems. Rick received his MS at Utah State University in forestry (fire and GIS emphasis) and his BS in conservation biology (wildlife emphasis) from Brigham Young University. Previous to SEM he worked eight seasons with the U.S. Forest Service/Park Service in varying biological and fire suppression and monitoring capacities.

You may order additional copies of this publication by sending your mailing information in label form through one of the following media. Please specify the publication title and number.

Publishing Services

Telephone (970) 498-1392

FAX (970) 498-1122

E-mail [email protected]

Web site http://www.fs.fed.us/rm

Mailing Address Publications Distribution Rocky Mountain Research Station 240 West Prospect Road Fort Collins, CO 80526

Stratton, Richard D. 2006. Guidance on spatial wildland fire analysis: models, tools, and techniques. Gen. Tech. Rep. RMRS-GTR-183. Fort Collins, CO: U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station. 15 p.

Abstract ________________________________________ There is an increasing need for spatial wildland fire analysis in support of incident management, fuel treatment planning, wildland-urban assessment, and land management plan development. However, little guidance has been provided to the field in the form of training, support, or research examples. This paper provides guidance to fire managers, planners, specialists, and analysts in the use of “models” (FARSITE, FlamMap, RERAP-Term), tools/programs (KCFAST, RAWS, FireFamily Plus, WindWizard), and procedures for spatial fire analysis. The approach includes a brief discussion about models and their assumptions and limitations, historical fire and weather analysis, landscape file data acquisition and development, landscape file and model output critique, and model calibration.

Keywords: Calibration, FARSITE, FireFamily Plus, Fire History, Fire Modeling, Fire Weather, FlamMap, Fuel Treatment, Gridded Wind, Hazard/Risk Assessment, Landscape Analysis, RAWS, RERAP, WindWizard.

Acknowledgments ________________________________ This document was supported by the USDA Forest Service, Rocky Mountain Research Station, Missoula Fire Sciences Laboratory and the Fire Modeling Institute. I am grateful for the review comments of Mark Finney, Larry Bradshaw, Chuck McHugh, Robert Seli, Louisa Evers, and Pat Stephen and for discussions with Joe Scott and Elizabeth Reinhardt. I also express my thanks to Kristi Coughlon and Nancy Chadwick for editorial assistance and publication design.

�USDA Forest Service Gen. Tech. Rep. RMRS-GTR-�83. 2006

Introduction_____________________ Largefiresarealandscape-level phenomenonbecausetheyaffectmultiplestandsoveralargearea.Theyarealsoaspatialphenomenonbecausethemovementandbehavioroffiredependsonwhereandwhatthingsitencounters and their spatial arrangement toonean-other.Toadequatelyperformbotha landscape-levelandspatialfireanalysis,onemustconductamid-scaleassessmentofexpected firebehaviorandmovementacrossmultiplestandsunderspecificweatherandfuelmoistureconditions. ArnoandBrown(1989)stated,“Animpressivebodyof scientific evidence…makes it clear that much ofNorthAmericaisa‘fireenvironment’wherewildfireorasubstituterecyclingmechanismisinevitable.”Thisecological certainty has been exacerbated by three-quartersofacenturyof fireexclusion inmanypartsoftheUnitedStates,oftenresultinginuncharacteristiclargefiregrowth,behavior,anditsundesirableeffects(ArnoandBrown1991). Policymakersseedestructiveandcostlyfireseasonssuch as 2000 or 2002 and react with legislation andincreased scrutiny (for example, National Fire Plan,Healthy Forest Initiative, General Accounting OfficeReports).Accordingly,landmanagersandfirespecial-istsare required toadhere toFederalandstateman-datesandguidelines.Withcompliancecomesreports,manyofwhichcanbenefitfromorrequirespatialfireanalysis,suchaslandmanagementplans;wildlandfiresituationanalysisandimplementationplans;hazard/riskassessments;communityprotectionplans;andNationalEnvironmentalPolicyAct(NEPA)compliance,docu-mentation,andoutreach. Unfortunately,littletrainingorinstructionhasbeengiventothefieldtoadequatelyconductspatialwildlandfireanalyses.Irrespectiveoftheproject,plan,orassess-ment, all landscape-level spatial fire analyses shouldshareacommonfoundationofproceduresandanalysis

Guidance_on_Spatial__Wildland_Fire_Analysis:

Models,_Tools,_and_Techniques

Richard_D._Stratton

ofthefire environment (weather,fuel,topography).Thispaperdescribesandprovidesexamplesofseveralcommonmodels, tools, andmethods for spatial firegrowthandbehaviormodelingandweatherandfireanalysis for use in research, wildland fire decisionsupport,andlandmanagementplanning.

A_Note_About_Models_and_Their_Assumptions_and_Limitations_ _____ “Amodelisasimplificationorapproximationofrealityandhencewillnotreflectallofreality”(BurnhamandAnderson1998).GeorgeBox(1979)stated,“Allmodelsarewrong,butsomeareuseful.”Itisthetaskofthemodelertoselecttheappropriatemodel,produceusableoutput,andinterpretmodelfindingsgivenmodelassumptionsandlimitations.However,itistheclientthatultimatelydeterminestheusefulnessofthemodel.Modelingisanartaswellasascienceandone’sfieldexperienceenablestheartofthemodeler.Bemindfulthatamodelisadeci-sionsupporttool,notatoolthatmakesdecisions. Avarietyofprogramsandtoolssupportwildlandfiremanagement.Forexample,therearesystemstopredictfiregrowthandbehavior,toolsandinformationonfireeffects,andsmokemodelsfordispersionandemissionestimates.Firemodels,suchasFARSITE(Finney1998)andBehavePlus(Andrewsandothers2005)areactuallyfire modeling systemsthatlinkmultipleempiricalanddeterministicmodelsor set(s)ofmathematicalequa-tionstopredictfiregrowthandbehavior.Eachmodel(forexample,surfacespreadmodel[Rothermel1972]orspottingmodel[Albini1979])hasassumptionsandlimitations,andcanbeapplieddifferentlyinthemodel-ingsystems.Itisimportantthatusersunderstandmodellimitationsandassumptionsandknowhowthesemodelsareusedinthefiremodelingsystems. For example, among our greatest challenges in firescienceispredictingcrownfireoccurrenceandbehavior.TodothisweusemodelsbyVanWagner(1977,1993)

2 USDA Forest Service Gen. Tech. Rep. RMRS-GTR-�83. 2006

forcrownfireinitiationandRothermel(1991)forrateofspreadcalculations.However,modelingsystemssuchas NEXUS (Scott 1999), FVS/FFE (Reinhardt andCrookston2003),BehavePlus,andFARSITE/FlamMap(Finney 2006) are implemented differently and canproducedifferingresults.Thedifferentinterpretationsstemprimarilyfromourpoorunderstandingofcrownfire—afieldofresearchstillinitsinfancy.Thevariousmodelsofsurfaceandcrownfirespreadweredevelopedindependently,andthecurrentunderstandingofcrownfirebehaviorisinsufficienttosuggestonlyonewayofintegratingthem.Thisincongruityiscompoundedbymodelassumptionsandlimitations.Amajorassump-tionisthatofauniformtreecanopy—whichweknowisnotthecase,sinceacanopyiscomprisedofcrownsofindividualtreesseparatedbygaps.Likewise,crownbaseheightisnotauniformvalueforanareaofland,yetthemodelsassumeasinglevaluecanbeappliedtoanentirestand. Sowhatthenistheanalystlefttodo?Untilanimprovedcrownfiremodelisdevelopedanditisappliedconsis-tentlyamongfireprograms,usersmust(1)knowmodelassumptions,limitations,anddesign,(2)determinetheusefulnessofmodeloutputthroughadequatedatade-velopmentandcritique,and(3)calibratethemodeltobeconsistentwithfieldobservations.Remembermodelsare“…artfulapplicationsofexistingknowledge.Theydonotattempttoexplainthephysicsormechanics…Asthescienceofsurfaceandcrownfirebehavioradvances,so toowill our firemanagement applications” (Scott2006).

Modeling_Systems_and_Tools_for_Landscape_Wildland_Fire_Analysis___FARSITE_[1]

FARSITE (Fire Area Simulator) (version 4.1) is atwo-dimensionalprogramforspatiallyandtemporallysimulatingthespreadandbehavioroffiresunderhet-erogeneousconditions.FARSITEincorporatesexistingfirebehaviormodelsofsurfacefirespread(Albini1976,Rothermel1972),crownfirespread(Rothermel1991,VanWagner1977,1993),spotting(Albini1979),point-sourcefireacceleration(ForestryCanadaFireDangerGroup 1992), and fuel moisture (Nelson 2000) withspatialinformationonfuels,weather,andtopography(theFARSITElandscapefile[LCP]).Simulationoutputis in tabular,vector, and raster formats.FARSITE isoftenusedtohelpanswerthequestions:Wherewillthefirego?Howlargewillthefireget?Whenmightthefirereachaparticularlocation?

FlamMap_[2]

FlamMap (version 3.0) is a spatial fire behaviormapping and analysis program that requires an LCPandfuelmoistureandweatherdata.UnlikeFARSITE,FlamMapmakesindependentfirebehaviorcalculations(forexample,firelineintensity,flamelength)foreachlocationoftherasterlandscape(cell),independentofoneanother.Thatis,thereisnopredictoroffiremovementacrossthelandscape,andweatherandwindinformationareheldconstant.FlamMapoutputlendsitselfwelltolandscapecomparisons(forexample,pre-andpost-treat-menteffectiveness)andtoidentifyinghazardousfuelandtopographiccombinations,thusaidinginprioritizationandassessment(Stratton2004).

FireFamily_Plus_[3]

FireFamilyPlus (FFP) (BradshawandMcCormick2000)(version3.0.5)isafireclimatologyandoccur-renceprogramthatcombines thefunctionalityof thePCFIRDAT(Cohenandothers1994,Mainandothers1990),PCSEASON(Cohenandothers1994,Mainandothers1990),FIRES (AndrewsandBradshaw1997),and CLIMATOLOGY (Bradshaw and Fischer 1984)programsintoasinglepackagewithagraphicaluserinterface.Itallowstheusertosummarizeandanalyzeweatherobservations,associateweatherwithlocalfireoccurrencedata,andcomputefiredangerindicesbasedontheNationalFireDangerRatingSystem(NFDRS)(Deemingandothers1977)and theCanadianForestFireDangerRatingSystem(CanadianForestryService1987).

Rare_Event_Risk_Assessment_Process_[4]

Rare Event Risk Assessment Process (RERAP)(USDA 2000, Wiitala and Carlton 1993) (version 7.0)isaprogramusedtoestimatetheriskthatafirewillreachaparticularpointofconcernbeforeafire-endingeventoccurs(Term).Itincorporatesweather,fuels,topography,andRothermel’s surface spread (1972)andcrown fire(1991)modelswithtwowaiting-timedistributions—oneforthefire-endingeventandtheotherfor“critical”spreadevents—toproduceprobabilitiesalongastraight-linetran-sect.RERAPconsistsofthreemodules:Term,Spread,andRisk.Forthepurposeofthispaper,theTermModuleisusedtodevelopaWeibullwaiting-timedistributionforauserdefinedfire-andseason-endingevent.

[1/2/3]FARSITE,FlamMap,andFireFamilyPlus:www.fire.org[4]RERAP:www.fs.fed.us/fire/rerap/indexz.htm

3USDA Forest Service Gen. Tech. Rep. RMRS-GTR-�83. 2006

WindWizard_[5]

WindWizardisaninterfacetoacomputationalfluiddynamicsmodel (CFD) (FLUENT®) to simulate theeffectterrainhasonwind(Butlerandothers2004).CFDtechnologywasinitiallydevelopedbytheaerospaceandautomotiveindustriestosimulatefluidflowaroundandthroughaircraft,automobiles,pipes,etc.Outputiswindvelocityanddirectioninrasterandvectorformatwithadefault resolutionof100meters.Griddedwindhastwoprinciplefunctions:visualizationandmodelinput.Shapefiles canbeplotted tohelpmanagersvisualizethechannelingandcheckingeffecttopographyhasonwind flow foroperational, planning, andeducationalpurposes. Output can be imported into FARSITE/FlamMapandmayimprovefiregrowthandbehaviorsimulations, particularly at a finer scale (Butler andothers2004,2006).

The_Spatial_Wildland_Fire_Analysis_Process_________________________ Above I addressed the need for fire specialists toperform both landscape-level and spatial wildlandfireanalyses.Idefinedfrequentlyusedterms,brieflydiscussed model assumptions and limitations, andmentionedafewkeymodelingsystemsandtools.Thispaperwillnowshiftfromapurelyinformationalstyletoaninstructionalapproach.Eachsubsequenttopicisprefacedbyashort introductionfollowedbyspecificinstructionsintermingledwithafewthought-directingquestions.

Project_Design_and_the_Management_Question Aresearchormanagementquestion(s)shouldbeatthe core of any wildland fire analysis. Development,execution, and implementation of the project shouldbecenteredonanswering thequestion.Theformula-tion of a clear question will guide data acquisition,model selection, theanalysis process,application oftheresults,andreportingandarchival(the“fourAs”astaughtinS-492[RERAP]).Sometimesaprojectcanbestructuredasaseriesofquestionsandthefindingsoftheinitialinquiryareneededtoanswersubsequentquestions.Forexample,whereareourecologicaland

socio-economicvalues?Whichof thesevaluesareatriskfromawildfire?Whatcanbedonetomitigatethisrisk? It is important that thequestionbeanswerable,rolesandresponsibilities identified, fundingsecured,andareasonabletimelineprepared.

Historical_Fire_Analysis:_Understanding_Fire_Occurrence_and_Movement Historicalfirerecordscanbeusefulinunderstand-ing the size, behavior, frequency, seasonality, andspatialdistributionandorientationoffires.Theyhelpmanagersconceptualizethefireshed—thepossibleorexpectedareainfluencedbyasinglefirestartascon-strainedbynaturalbarriers,fuel,terrain,andweatherwithinagivenperiodoftime.Firerecordsareeitherhardcopyorelectronic—inaGeographicInformationSystem(GIS)/spreadsheetformatorinacentraldatabase(for example,National InteragencyFireManagementIntegratedDatabase[NIFMID][USDAForestService1993]). Information is typically categorized as a fireoccurrence or ignition history, or a fire area history(forexample,fireatlas).Afewareasmaycontainfirehistoriesreconstructedfromcrosssectionsoffire-scarredtreesorestablishmentdatesofpost-firetreecohorts[6].Inmostcases,thelongerthefirerecord,theeasieritistoascertainthefireshedandseepatternsonthelandscape. Fire ignition history—Common among federalandsomestateagenciesareGISpointignitionlayers.Federal fire occurrence information is available fordownloadat theNationalFireandAviationManage-ment Web Applications Site (FAMWEB) or KansasCityFireAccessSoftware (KCFAST) (USDAForestService1996)[7]—ForestServiceonly(Fire>StandardExtract)(.RAWfile).Ifelectronicdataarenotavailable,a GIS layer can be generated from local fire reportsor associated records—include attribute data such asfirename,date,size,andcause.Createadensitygridfromfirelocationstoidentifyhighfirefrequencyareas(fig.1).Wheredomostfiresoccur?Aretheyhumanornatural-caused?Displaythefiresbysize;wheredomostlargefiresoccur?Displaythefiresbydate;arethereany noticeable trends? This information should giveamanageranadequatefeelforwhereignitionscouldoccurandwhenthefireseasonbeginsandends.

[5]WindWizard:www.firelab.org [6]NOAAPaleoclimatology:www.ncdc.noaa.gov/paleo[7]FAMWEB(KCFASTandUSFederalWildlandFireManagementData):http://famweb.nwcg.gov/

� USDA Forest Service Gen. Tech. Rep. RMRS-GTR-�83. 2006

Figure_1—Fire ignition density, both human and lightning caused, on the Boise National Forest (ID) (�955 to 2002). One to three fires within the �-mile search radius are in blue, � to 6 fires (green), 7 fires (yellow), 8 fires (orange), and 9 to �0 fires (red).

An ignition density only indicates where fires havestarted,notwhereorunderwhat conditions fireswillspreadorwhatimpactswilloccur.Thus,fireoccurrencedataaloneareoflimitedvaluetoriskassessment.Thecomponentsoffireriskincludetheprobabilityofburn-ingandtheimpactsthatfireshavewhentheydoburn.Thisisdifferentthanhazard—thephysicalsetting(fireenvironment)adjacenttothevalue(BachmannandAll-gower2000).Mostareaisburnedbytherare,largefires(Finney2005)thatarenotnecessarilyrelatedtotheloca-tionsoffrequentignitions.Sousetheignitioninformationcautiouslyforwhereafiremayoriginate.

Fire area history—DevelopafireatlasorfireareahistoryusingGlobalPositioningSystem(GPS)perim-eters, remote sensing, aerial photographs, historicaldocuments, firescars,datesofpost-fire treecohorts,andsoforth(fig.2).Payparticularattentiontothelargespread events and consider the following: What wasthesize,behavior,andorientationoftheselargespreadevents?Whatdidthefuelcomplexlooklikewhentheselargefiresoccurred?Whatwastheweather,wind,andfuelmoisture?Whatwastheenergyreleasecomponent(ERC)orotherfiredangerindiceswhenthefiremadesignificantadvances?Toobtainthisinformation,talk


Figure_2—Large fire history (>�00 ac) on the Coconino National Forest, north of Flagstaff, AZ (�950 to 200�) (courtesy of Charles W. McHugh).

withpastfiremanagersandobtainfirepackagesandrelatedmaterialsfromthehostunit. Answeringtheabovequestionsaboutlargefiregrowthprovidestheframeworkforunderstandinganddevelopingthe problem fire—apossiblefireevent,asevidencedbythehistoricalfirerecord,whoseoccurrenceonthelandscape would be undesirable. Information from athoroughhistoricalfireanalysiswillenablemanagerstounderstandthefirethreatandvisualizethefireshed,resultinginanappropriatemanagementresponseandimprovedprojectdevelopmentandimplementation.

Environmental_Analysis:_RAWS,_FireFamily_Plus,_RERAP,_WRCC,_and_WindWizard FARSITE and FlamMap require weather and fuelmoisturevariablesformodeling(Finney1998).Historicalandcurrentweatherinformation—obtainedfromremoteautomated weather stations (RAWS) (NWCG 2000;Zachariassen and others 2003)—can be downloadedfromNIFMIDusingKCFAST,andimportedintoFFP.IfRAWSstationsareunavailableforyouranalysisarea,considerstationsadministeredbytheNationalWeather


Service,FederalAviationAdministration,andNaturalResourceConservationService(SNOTELsites).[8]

Remote Automated Weather Station retrieval—Locate thenearestRAWS(s) for yourproject area[9].ItisimportanttoselectaRAWSatasimilarelevationandtopographicsetting,keepinginmindtheinfluencethe terrain and vegetation may have on the station.Occasionally,localtopographicinfluencesmaskprevail-ingwindpatternsthatinfluencelargefiregrowth.Inthiscase,onestationmaybeusedforthewinddirectionandanotherfortheweather(Stratton2004). Use KCFAST Weather Information Retrieval[10] to(1)findtheWeatherInformationManagementSystem(WIMS)stationnumber(KCFAST:Weather>StationCatalog>StationInformation),(2)obtainayearcount(KCFAST:Weather>DataExtract>Utilities),(3)obtainthestationcatalog,and(4)retrievetherawhistoricalweatherdatafile(s).Thestationyearcountisimportantasitshowshowlongthestationhasbeenoperable.Gen-erally,thelongerthehistorythebetter.Thestationyearcountalsoshowsthenumberofobservationsperyear.Yearswithfewobservationsorlargegapsinthedata(KCFAST:Weather>DataExtract>Utilities>MissingData)shouldbeomitted.RegionandunitnumberscanbeobtainedusingFFP(Fire>Associations). Climatology and fires analysis in FireFamily Plus—Using FFP, import the station catalog (.TXT),historicalfire(.RAWor.FPL),andweatherfiles(1972dataformat[.FWX])(Data>Import).HourlydataarealsoavailablefromKCFAST(1998dataformat[.FW9]),butisonlyaccessiblefortheprevious18months.Viewthe import error log(s) after bringing in station andweatherinformation.Errorsinthesefilescansignifi-cantlyaffectyouranalyses.Listallweatherobserva-tions(Weather>ViewObservations>All).Sorteachweathervariableinascendingordescendingorder—rightmouseclick—lookingforerrorsincludingmissingdataandexcessivevalues(forexample,precipitation,wind,relativehumidity). Associatefiredatawiththeweatherstation(s)(Fires>Associations). Determine when fires start, peak, andend(FFP:Fires>Summary).WhatwastherangeofERCvaluesorotherfiredangerindiceswhenthelarge

firesoccurred(FFP:Fires>FireAnalysis)?Rememberthatfiresizeisassociatedwiththediscovery date,sobecarefulwheninterpretingthedata. Toobtainclimatologyandmoistureinformation,userscangenerateadailylistingofcustomizedvariables(forexample,relativehumidity,fuelmoisture)relativetothenumberoffiresorsizeofthefire(Weather>SeasonalReports>DailyListing).Thislistingcanbeimportedintoaspreadsheet.Or,userscangenerateapercentileweatherreport(Weather>SeasonalReports>PercentileWeather)usinganNFDRSindex(forexample,ERC)toproducetherequiredinputsformodeling.Usingthedailylistingfeatureispreferredinsomecasesbecausefiresresulted from these conditions and agency personneland the interestedpubliccanrememberandrelate tothatparticularevent. RERAP—Use RERAP, Term Module, to developwaiting-timedistributionsforfire-andseason-endingevents(fig.3).Afire-endingorfire-stoppingweathereventoccurswhensufficientmoistureresultsintermi-nationoffirespread.Aseason-endingweathereventconsistsofafire-endingeventfollowedbyapersistentcombinationofenvironmentalfactorsthatendthefireseason(USDA2000). Develop Term event criteria by combining localknowledgewithclimatologyandfiresdatainFFP.Afire-endingeventisusuallyaresultofacertainamount

[8] RAWS and other Station Information: http://raws.wrh.noaa.gov/roman/index.html[9]RAWSLocations:www.raws.dri.edu/index.html[10]KCFAST/FireWeather InformationRetrieval: http://famweb.nwcg.gov/kcfast/html/wxmenu.htm

Figure_3—Weibull waiting-time distribution from RERAP given the likelihood of a fire-stopping event (>2 inches of rain over a 3-day period) on the White Mountain National Forest (NH).


ofprecipitationoveraspecifiedtimeframe.Aseason-endingeventcombines thefire-stoppingeventwithapredictoroffirebehavior(forexample,ERC).InFFP,graphERCandoverlayaseriesofyearswithhistoricalfirestoidentifyathresholdERCvaluewhenfireoccur-renceandlargefiresceaseandresurgenceisnolongerathreat.UsetheEventLocatorinFFPtoidentifytheoccurrenceofthefire-stoppingeventsandthethresholdERCvalue.Thisinformation,alongwiththeseasonstartdate,isusedinRERAPtodeveloptheseason-endingevent.Itisnotuncommontohavemultipleseason-endingevents—abimodalfireseason.Terminformationisuse-fulforincidentmanagement,riskassessment,planningofprescribedfires,andfine-tuningseasonstartandenddates. Wind speed and direction from RAWS/FireFamily Plus—Windisoneoftheprimaryenvironmentalvari-ables influencing wildland fire growth and intensity(Catchpoleandothers1998,Rothermel1972).Forthisreason,itiscriticalthatrepresentativewindspeedanddirection information are used in any landscape fireanalysis.Whenobtainingandusingwindinformation,considerthefollowingquestions:Whatistheprevail-ingwinddirectionduringthefireseason?Isthewinddirectionconsistentwithwhatfiremanagersperceiveastheirpredominantwinddirection?Atwhatwindspeeddo fires torchandactivelycrown?What causeshighwindeventsandhowoftenandwhendotheyoccur? BecautiouswhenusingFWXfilesforwindspeedanddirectioninformation.First,mostRAWSwindspeedanddirectioninformationisa10-minuteaveragepriortodatatransmission,typically1300localstandardtime.Inthecaseofwindspeed,thisvalueisanaveragecomposedoflullsandgusts.Averagingthegustinessisproblematicwhenmodelingfirebehavior.Oftenthesegustsorpeakwindsaffectfiregrowth,intensity,andspotting(CrosbyandChandler2004).Byusingthe10-minuteaverage,evenatthe99thpercentile,thewindspeedisoftenunder-represented.Adjustthisvaluetobemorecharacteristicof target conditions. Consider using the maximum1-minutewindspeedfromthe“windgustestimatingtable”byCrosbyandChandler.Second,becauseFWXfilesareonlyasinglesnapshotintime,nootherwindinformationisprovided.Thismaybeofconcernformid-morningactivity,lateafternoonruns,andactiveburning in the evening. In FFP, wind direction in-formationcanbeobtainedbyawindspeedvs.winddirection report (Weather>Winds), as specifiedbyanalysisperiod(days)andsummarizedbydirection.InformationfromFFPcanalsobeexportedintoFAR-SITE/FlamMapformat.

Wind speed and direction from WRCC—Anexcel-lentsupplementtotheKCFASTdatabaseisweatherandwindinformationfromtheWesternRegionalClimateCenter(WRCC).WRCCstoreshistoricaldatafromallRAWSintheU.S.,includingtheCaribbeanIslands[11].WRCCdataishourly,mayincludegusts,andviewablevia their website or downloadable in FW9 or FWXformatforimportintoFFP(SelectyourRAWS[see9]/DataLister>DataFormat).Windspeedanddirectioninformationisreportedasa10-minuteaverage.Ausefulfeatureistheselectionofsubintervalsbymonth,day,andstartingandendinghours.Thisallowstheusertogetmoredetailedwindinformationforthefireseasonandburningperioddisplayedinawindrose(fig.4). WindWizard—WindWizard requires synoptic(generalorupperlevel)windspeedanddirection,andadigital elevationmodel (DEM) (USGS1990).Thisinput speed differs from the 20-ft wind speed. Usu-ally,thespeedneedstobetwotothreetimesthatoftheobserved20-ftridgetopwindtoproducethedesiredresults. WindWizard is not a forecast; it simulatestopographicallymodifiedflowforaninstantintime.MultiplesimulationsofwindspeedanddirectioncanbecreatedforuseinFARSITE/FlamMap. TheDEMmustincludeabufferareaofapproximately20 percent of the horizontal and vertical distance toincorporate surrounding terrain that may influencetheareaofinterest.IfDEMsareinaccessibleforyourareaof interest, theUnitedStatesGeologicalSurvey(USGS)FireDataOrderingSystem[12]providesaquickandefficientwaytoobtainaDEMandotherancillarylayers. Output from theWindWizard iswindvelocityanddirectioninvector(ArcVieworArcMap;onefile)(fig.5)and raster (ASCII/Raster; two files) format. To viewvectorformat,downloadtheinstructionalbriefsfromtheWindWizardwebsite[see5]. Import therasterdataforfiremodelinginFlamMapbyselectingNewRun>Inputs>WindGrid.InFARSITE,selectInput>ProjectInputs>SelectWindFile(.ATM).Theatmospherefile(ATM)isatextfilespecifyingmonth,day,hour,windspeed,direction,andcloudcover.ItmustbepreciselyformattedtofunctioninFARSITE(refertoFARSITEHelp).

[11] Western Regional Climate Center: www.wrcc.dri.edu/index.html[12]USGS,FireDataOrderingSystem:http://firedata.cr.usgs.gov/


Figure_4—Wind rose from the Western Regional Climate Center for Horton Peak RAWS (ID), July � to Oct. 3�, �985 to 200�. Observations are hourly from ��00 to �800.

Spatial_Data_Acquisition_and_Development:_The_FARSITE_Landscape_File Atthebeginningofanylandscapeproject,oneshoulddevelopaspatial needs analysis—alistofrequiredandancillaryGISinformationrequisiteforcompletionofthetask.Table1providesanextensivelistofGISdatafor landscape wildland fire analysis. Central to thistype of work is the accurate parameterization of thefireenvironment.ThemostcommonfilestructureforspatialfireanalysisistheFARSITE/FlamMaplandscapefileorLCP,asinglefileconsistingofelevation,slope,aspect,fuelmodel(Anderson1982,ScottandBurgan2005),andcanopycover(CC)—requiredinputs—aswellas optional themes of stand height (SH), crown baseheight(CBH),crownbulkdensity(CBD),duffloading,andcoarsewoodydebris(Finney1998).ThestandardpracticeistocreateanLCPusingalloftherequiredfilesandthethreecanopycharacteristicfiles.Develop-mentoftheoptionalthemesmaynotbenecessaryin

areaswherecrownfirespreadisoflittleconcern(forexample,hardwoodforests).EachGISlayerneedstobeexportedasanASCIIRastertocreatetheLCPfile. DatarequiredtocreateanLCPcomefromaDEM(elevation,slope,aspect)andsatelliteimageryorremotelysensed data that are usually classified by vegetationtype and attributed. DEMs are easily obtained fromtheUSGS;fuelmodel,canopycover,andcanopychar-acteristic information ismoreproblematic.Presently,severalFederallandshaveFARSITEdataaccessiblebycontactingthehostagency.LANDFIRE,anationwidefire, fuel,andvegetationmappingproject,willbeginrolloutofwesternproductsin2006,includingFARSITEdata,withfinalcoverageoftheUnitedStatesin2009(Alaska).[13]IrrespectiveofthesourceoftheFARSITEdata,LCPsshouldbethoroughlycritiquedandmappingassumptionsandlimitationsunderstoodtoensureproperuseandtoproducerealisticresults.

[13]LANDFIRE:www.landfire.gov


Critique_of_the_Landscape_File Errors in creating the LCP file—ThereareseveralcommonerrorsinthecreationoftheFARSITEland-scape file. To identify these errors, load the LCP inFARSITEorFlamMap.CritiquetheLCPbyqueryingthelandscapeandviewingthedialogueboxlookingforcommonoversights(listedbelow).ViewtheLCPfilegenerationbox looking forerrors including (1)a“0”latitude,(2)theloadingofincorrectASCIIRasterdata,and(3)distanceandASCIIRasterthemeunitsincorrectlyspecified.IfcreatingtheLCPfromindividualASCII

Rasterfiles,anadditionalcheckisdonebyopeningeachfileinWordPad©andcheckingtheextent,resolution,andgridvalues. Common oversights in ASCII Raster data—Thereare several common missteps in FARSITE layer de-velopment. Often, CC values are too high and CBDandCBHareinaccurate.CCrarelyexceeds70percenteveninso-calledclosed-canopyforests(seeScottandReinhardt[2005]).IfexcessiveCCvaluesexistwithinthemodelingdomain, rateof spreadwillbe reducedduetotheshelteringeffectofthetreecanopy.

Figure_5—Gridded wind (�00 m) derived from WindWizard based on 30+ mph ridgetop winds from the northeast. Wind vectors are colored by speed with 0 to 7 mph (blue), 8 to �� (green), �5 to 22 (yellow), 23 to 29 (orange), and 30+ (red). The Blossom Fire (��,908 acres) (Aug. �2, 2005; Agness, OR) perimeter is overlaid on a shaded relief with trails (black) and the Rogue River to the south (blue).

�0 USDA Forest Service Gen. Tech. Rep. RMRS-GTR-�83. 2006

Table_1—GIS data requirements and considerations for spatial wildland fire analysis.

FARSITE/FlamMap_Landscape_FileCanopy coverCrown base heightCrown bulk densityDigital elevation model (DEM)Fuel modelStand height

Wildland_Fire_HistoryFire atlas (perimeters)Fire occurrence (points)Fire progression layersFuel treatments and other stand modificationsPrescribed fire history

Ecological_ConsiderationsAreas of critical environmental concern (ACEC)Erosive soilsFire regime condition class (FRCC)Historical range of variability (HRV)Invasive weedsOld growthProper functioning condition (PFC)Research sites (for example, research natural areas [RNA])Rivers and streamsSensitive or critical wildlife habitat (for example, winter range, riparian)Threatened, endangered, and sensitive (TES) flora and fauna habitatTimber productsWater bodiesWatersheds of concern (6th code HUC particularly useful)

Socio-economic_ConsiderationsBridgesCampsCommunication sitesConcentrated use areas (CUA)Cultural resourcesGrazing allotmentsHistoric and recreational sitesMine and mineral claimsMunicipal water supplies and infrastructurePrimary and secondary residencesRailroadsRemote automated weather stations (RAWS)RoadsScenic integritySki areasStructures - other (for example, highway maintenance buildings, rest areas)Trails and trailheadsUtility corridorsWildland-urban interface areas

Other_Useful_Base_LayersAerial photosCoordinate system (for example, UTM, Lat/Long)Digital orthophoto quads (DOQ) or quarter quads (DOQQ)Digital raster graph (DRG)Ownership and jurisdictions (for example, private lands)Public land survey system (PLSS)Satellite imageryShaded/painted reliefVegetation or cover-type classification

��USDA Forest Service Gen. Tech. Rep. RMRS-GTR-�83. 2006

FieldmeasurementsofCBDrarelyexceed0.25kgm–3(forexample,spruce-fir)(ScottandReinhardt2005),buttoyieldrealisticcrownfireestimatesinFARSITE/FlamMap,theseuppervaluesoftenneedtobeincreasedto0.4+ifthedefaultcrownfirecalculationmethod(Finney1998)isused.IfCBDvaluesaregeneratedfromFVS/FFEorFMAPlus(Carlton2005),selecttheScottandReinhardt(2001)crownfirecalculationmethod.ParticularattentionshouldbegiventoCBH.Ifvaluesaretoohigh,particularlywhencoupledwithamodest fuelmodel (forexample,FM8),crownfireswillseldominitiate.

Critique_of_Fire_Behavior_Outputs AnexcellentwaytocritiquefirebehavioroutputsisbyusingFlamMap.Considersimulatinglow,moderate,andsevereconditionsusingbothcrownfirecalculationmethods(Finney1998,ScottandReinhardt2001).Com-pareoutputfromeachsimulationandperformsensitiv-ityanalysesbyaskingquestionssuchas:Istheflamelength,rateofspread,andintensityconsistentwithfieldobservations?Areareasofcrownfireinitiationappro-priategivenfuelandweatherconditions?Arepassive(torching)andactivecrownfireareasreasonable?Whenyouseethingsthatarenotquiteright,remember“Don’tBeDUM”—data,user,model.Weareoftentooquicktoaccusethemodelwhenthedataarethesourceoftheproblem(astaughtinS-493[FARSITE]).Ingeneral,ifmodificationsofthespatialdataareneeded,adjustthefuelmodellayerfirst,followedbyCBHandCC.TheFARSITElandscapecalculatorisanexcellentwaytomakechangestoanLCPwithoutGISassistance. Flame length and fireline intensity—Flamelengthiscalculatedfromfirelineintensity—theproductoftherateofspreadandheatgeneratedfromtheavailablefuelduringflamingcombustion(Byram1959).InFARSITE/FlamMap,Byram’sequationisusedtopredictsurfacefireflamelength;Thomas’(1963)equationisusedtopredictflamelengthforpassiveandactivecrownfire(Rothermel1991). Sinceflamelengthisdefinedandmeasuredforshallowsurfacefires,flamedimensionsincrownfiresordeepfuelbedsmayappear toanobserver tobegreater than thecalculatedvalues.Flamesgeneratedfromaerialfuelsareviewedrelativetothegroundbutoriginatewellabovethesurface.Sokeepinmindthat“flamelengthisanelusiveparameterthatexistsintheeyeofthebeholder.Itisapoorquantitytouseinascientificorengineeringsense,butitissoreadilyapparenttofirelinepersonnelandsoreadilyconveysasenseoffireintensitythatitisworthfeaturingasaprimaryfirevariable”(Rothermel1991).

Crown fire occurrence—Crown fire initiationis dependent on several factors, including surfacefireline intensity, canopy foliar moisture, and CBH(VanWagner 1977). CBH is used to determine iftorchingoccurs.CBDaffects transition toanactivecrown fire. If modeled output is inconsistent withfield observations (for example, an under-predictionofactivecrownfire)(Cruzandothers2003,Fuléandothers2001),modificationstofuelmodel,CBH,CBD,orwindspeedmaybenecessary.Also, lowering thefoliarmoisturecontent—alandscapewideadjustmentinFARSITE/FlamMap—willincreasetheoccurrenceofcrownfire.

Model_Calibration Calibrationiscriticaltoanylandscapeanalysis.Toproduce fire growth and behavior outputs consistentwithobservations,modelchecking,modifications,andcomparisonsaredonewithknownfireperimetersandweatherconditions(Finney2000).Ifthemodelhasnotbeenproperlycalibratedtolocalfires,howcananalystsormanagershaveanyconfidenceinitsoutput?Conversely,whenfirebehavioroutputshavebeencritiquedandthemodelcalibratedtopreviouslargefires,onecanhaveahigherdegreeofconfidenceinfuturesimulations.Forthepurposesofthispaper,calibrationoftheFARSITEmodelisdiscussed.However,asimilarapproachcanbeusedtocalibrateothermodelingsystems,suchasRERAP(SpreadModule)andFSPro(Finney,inpreparation). Rateofspreadisthemostcommonfirebehaviorvari-ablecalibrated.Whereappropriate,calibrationoffiretypemaybeimportantforfireeffectsmodelingandtoaccuratelydifferentiatebetweensurfacefirespreadrate(includestorching)andactivecrownfirespread.Fieldobservations,stillandvideophotographyfromaerialreconnaissance,andtheNationalParkService/USGSNationalBurnSeverityMappingProjectcanbeusedtocalibratefiretype[14].Considerselectingonefireorruntocalibratethemodelundermoderateconditionsandanotherforsevereconditions(fig.6)withandwith-outahighwindevent.In order of importance,criteriafundamentaltoanycalibrationexerciseinclude: • acarefullydevelopedandcritiquedLCP; • progression layer or detailed field observations

thatidentifythepositionandtimeofthefire(forexample, field notes, dispatch logs), including aprecisestartinglocation;

[14]BurnSeverityMappingProject:http://burnseverity.cr.usgs.gov

�2 USDA Forest Service Gen. Tech. Rep. RMRS-GTR-�83. 2006

Figure_6—Calibration of the FARSITE model (in red; �6,686 acres) to a �-day crown fire run on the northeast head of the Sanford Fire (June 8, 2002; Panguitch, UT). Overlaid on slope are trails (green), roads (black), and the fire perimeter (blue).

• a fireofa sufficient sizeand/or includesseveralburnperiods;

• ampleandrepresentativeweatherandwindinfor-mation(forexample,RAWS[s]nearby);

• afirethatburnedinseveraldifferentfuelmodelsandvaryingterrain;

• accuratefuelmoistureinformation; • a fire with minimal suppression or knowledge of

suppressiontactics; • a fire thatburnedunderavarietyofweatherand

windconditions;and • afirethatresultedinbothsurfaceandcrownfireruns.

Sufficient calibration takes time and patience. Donot expect perfection given model assumptions andlimitations, data inaccuracies, fire suppression,variability in the weather and wind, etc. Sometimes,FARSITEwilladequatelypredicttheshapeofthefire,butnotthetimingoffirearrival.EveryfireandLCPisdifferent,henceeverycalibrationexercisewillvary.Thefollowingcalibrationtipsusuallyholdtrue:

• Wind,weather,andfuelmoisturefilesarecrucialsogettheseinputsasclosetorealityaspossible.Spending timeherewill isolateseveralvariables

andstreamlineyourcalibrationprocess.Makesurethestationelevationiswithinthefireareaandfuelsareconditioned(ifneeded).

• Accuratelydefinetheburnperiodbythestartingandendingtimeofdailyfiremovement.

• Select the appropriate model parameters—usuallyacoarserresolutionatthestartofyoursimulationsmoving to a finer resolution as you get closer tocompletion.

• If spottingcontributed to thegrowthof the fire,enable spot fire growth early on, but at a lowfrequency(0.5to1percent).Iffirespreadispre-dominatelythroughspotting,adjustmentstoCBHwilllikelybenecessary.

• Makeonemodificationatatimeandthenrerunthesimulation.

• Donottrytocalibratetheentirefireatonce.Startwith the first fewhoursorburnperiodand thenbuildfromthere.Iftheinitialprogressionsareoff,itislikelytheentiresimulationwillfollow—aresultofcompoundingerror.

• Forlargefires,orwheretheoriginorprogressionsare lacking, use a reliable perimeter and beginyourcalibrationprocessthere—watchforerrorsinperimeterdatesandtimes.

• Asubstantialchange(~0.3to0.4)intheadjustmentfileindicatesadifferentfuelmodelmaybeneeded.Tryusingaconversion file,orasa last resort,acustomfuelmodel.

• Adjustforthelackofextinctionofthefireperimeterafternightfallorrain.Thisisimportantinlightfuels,wherethefirewillresumewhenthefuelmoisturedrops.

• Personnelonthefirecanbeausefulresourceasmultipleperspectivesleadtocorroborationofkeyevents.

• Keepadetailedlogthroughoutthecalibrationprocessofmodelsettings,parameters,adjustments,etc.

Summary_ ______________________ This paper has been written to provide directionto fire specialists in spatialwildland fire analysis. Itis a product of seven years of experience in supportof incidentmanagement, fuel treatmentprojects, riskassessments,andreviews/investigations.Figure7 isaflowchartthattiesthetoolsandprogramstotheanalysisprocess.Although“applications”arelistedontheright(theend),theresearchormanagementquestionshoulddictate model and tool selection. When a modeler isfamiliarwithmodelassumptions,limitations,anddesign;completesacomprehensivefireandweatheranalyses;

�3USDA Forest Service Gen. Tech. Rep. RMRS-GTR-�83. 2006

Figu

re_7

—Fl

ow c

hart

for s

patia

l wild

land

fire

ana

lysi

s: d

esig

n, to

ols,

and

app

licat

ion.

�� USDA Forest Service Gen. Tech. Rep. RMRS-GTR-�83. 2006

carefullycritiquestheLCPandresultantfirebehavioroutput;calibratesthemodel;andperformsthenecessaryanalyses,thenlineofficers,incidentcommanders,andfirespecialistshavecompellingandusefulanalysesforimproveddecisionmaking,documentation,andpublicoutreachandeducation.

References______________________Albini,F.A.1976.Estimatingwildfirebehaviorandeffects.Gen.

Tech.Rep.INT-30.Ogden,UT:U.S.DepartmentofAgriculture,Forest Service, Intermountain Forest and Range ExperimentStation.92p.

Albini,F.A.1979.Spotfiredistancefromburningtrees:apredictivemodel.Gen.Tech.Rep.INT-56.Ogden,UT:U.S.DepartmentofAgriculture,ForestService,IntermountainForestandRangeExperimentStation.73p.

Anderson,H.E.1982.Aidstodeterminingfuelmodelsforestimat-ing firebehavior.Gen.Tech.Rep. INT-122.Ogden,UT:U.S.DepartmentofAgriculture,ForestService,IntermountainForestandRangeExperimentStation.22p.

Andrews, P.L.; Bevins, C.D.; Seli, R.C. 2005. BehavePlus firemodeling system, version 3.0: User’s Guide. Gen. Tech. Rep.RMRS-GTR-106WWW.Ogden,UT:DepartmentofAgriculture,ForestService,RockyMountainResearchStation.142p.

Andrews, P.L.; Bradshaw, L.S. 1997. FIRES: Fire InformationRetrieval and Evaluation System - a program for fire dangerratinganalysis.Gen.Tech.Rep.INT-GTR-367.Ogden,UT:U.S.DepartmentofAgriculture,ForestService,IntermountainForestandRangeExperimentStation.64p.

Arno,S.F.;Brown,J.K.1989.Timeforanewinitiative.JournalofForestry.87(12):44-46.

Arno,S.F.;Brown,J.K.1991.Overcomingtheparadoxinmanagingwildlandfire.WesternWildlands.Spring:40-46.

Bachmann, A., Allgower, B., 2000. The need for a consistentwildfireriskterminology.In:Neuenschwander,L.F.,Ryan,K.C.,Gollberg,G.E.,Greer,J.D.(Eds.),ProceedingsoftheJointFireScience Conference and Workshop: Crossing the Millenium:IntegratingSpatialTechnologiesandEcologicalPrinciplesforaNewAgeinFireManagement,BoiseIdaho,June15-17,1999,UniversityofIdaho,pp.67-77.

Box,G.E.P.1979.Robustnessinthestrategyofscientificmodelbuilding. In: Robustness in Statistics. R.L. Launer and G.N.Wilkinson,Editors.AcademicPress:NewYork.

Bradshaw,L.S.;Fischer,W.C.,compilers.1984.ComputerprogramsforsummarizingclimaticdatastoredintheNationalFireWeatherDataLibrary.Gen.Tech.Rep.INT-164.Ogden,UT:U.S.Depart-mentofAgriculture,ForestService,IntermountainForestandRangeExperimentStation.39p.

Bradshaw,L.S.;McCormick,E.2000.FireFamilyPlususer’sguide,version2.0.Gen.Tech.Rep.RMRS-GTR-67WWW.Ogden,UT:U.S.DepartmentofAgriculture,ForestService,RockyMountainResearchStation.

Burnham,K.P.;Anderson,D.R.1998.ModelSelectionandInfer-ence.NewYork:Springer.

Butler, B.W.; Forthofer, J.M.; Finney, M.A.; Bradshaw, L.S.;Stratton,R.D.2004.In:Aguirre-Bravo,Celedonio,et.al.Eds.Monitoring Science and Technology Symposium: UnifyingKnowledgeforSustainabilityintheWesternHemisphere;2004September20-24;Denver,CO.ProceedingsRMRS-P-42CD.Ogden,UT:U.S.DepartmentofAgriculture,ForestService,RockyMountainResearchStation.

Butler,B.;Finney,M.;Bradshaw,L.;Forthofer,J.;McHugh,C.;Stratton,R.;Jimenez,D.2006.WindWizard:ANewToolfor

FireManagementDecisionSupport.In:Andrews,PatriciaL.;Butler, Bret W., comps. 2006. Fuels Management—How toMeasureSuccess:ConferenceProceedings.28-30March2006;Portland,OR.ProceedingsRMRS-P-41.FortCollins,CO:U.S.Department of Agriculture, Forest Service, Rocky MountainResearchStation.809p.

Byram,G.M.1959.Combustionofforestfuels.InK.P.Davis(ed.).Forest Fire: Control and Use. McGraw-Hill Book Company,NewYork.pp.61-89.

Canadian Forestry Service. 1987. Canadian Forest Fire DangerRatingSystem–Users’Guide.CanadianForestryServiceFireDangerGroup.Three-ringbinder(unnumberedpublication).

Carlton,D.W.2005.FuelsManagementAnalystSuiteUserGuide.FireProgramSolutions/AcaciaServices,Sandy,OR.112p.

Catchpole,W.R.;Catchpole,E.A.;Butler,B.W.;Rothermel,R.C.;Morris,G.A.;andLatham,D.J.1998.Rateofspreadoffree-burningfiresinwoodyfuelsinawindtunnel.Comb.Sci.Tech.131:1-37.

Cohen,E.;Nostrant,R.;Hawk,K.;Mitchell,W.1994.pcFIRDAT/pcSEASON User Guide: FireFamily for personal computers.Sacramento,CA:StateofCalifornia,CaliforniaDepartmentofForestryandFireProtection.46p.

Crosby,J.S.;Chandler,C.C.2004.FireManagementNotes.Getthemostfromyourwindspeedobservations.Vol.64,No.1,Winter2004.Reprintedfrom:FireControlNews27(4):12-13,1966.

Cruz,M.G.;Alexander,M.E.;Wakimoto,R.H.2003.Definitionofafirebehaviormodelevaluationprotocol:acasestudyapplica-tiontocrownfirebehaviormodels.In:Omi,PhilipN.;Joyce,LindaA.,technicaleditors.Fire,fueltreatments,andecologicalrestoration: Conference proceedings; 2002 16-18 April; FortCollins,CO.ProceedingsRMRS-P-29.FortCollins,CO:U.S.Department of Agriculture, Forest Service, Rocky MountainResearchStation.475p.

Deeming,J.E.;Burgan,R.E.;Cohen,J.D.1977.TheNationalFire-DangerRatingSystem-1978.U.S.DepartmentofAgriculture,ForestService,Gen.Tech.Rep.INT-39,IntermountainForestandRangeExperimentStation,Ogden,Utah.66p.

Finney,M.A.1998.FARSITE:FireAreaSimulator—modelde-velopmentandevaluation.Res.Pap.RMRS-RP-4.Ogden,UT:U.S.DepartmentofAgriculture,ForestService,RockyMountainResearchStation.47p.

Finney,M.A.2000.AspatialanalysisoffirebehaviorassociatedwithforestblowdownintheBoundaryWatersCanoeArea,Min-nesota.Duluth,MN:U.S.DepartmentofAgriculture,SuperiorNationalForest.

Finney,M.A.2005.Thechallengeofquantitativeriskassessmentforwildlandfire.For.Ecol.andMgt.211:97-108.

Finney,M.A.2006.AnoverviewofFlamMapmodelingcapabili-ties. In:Andrews,PatriciaL.;Butler,BretW.,comps.2006.Fuels Management—How to Measure Success: ConferenceProceedings.28-30March2006;Portland,OR.ProceedingsRMRS-P-41.FortCollins,CO:U.S.DepartmentofAgriculture,ForestService,RockyMountainResearchStation.809p.

Finney, M.A. In preparation. Fire Spread Probability (FSPro).Missoula,MT:USDAForestService,RockyMountainResearchStation,FireSciencesLaboratory.

ForestryCanadaFireDangerGroup.1992.Developmentandstruc-tureof theCanadianForestFireBehaviorPredictionSystem.Inf.Rep.ST-X-3.

Fulé,P.Z.;McHugh,C.;Heinlein,T.A.;Covington,W.W.2001.Potentialfirebehaviorisreducedfollowingforestrestorationtreatments. In: Vance, Regina K.; Edminster, Carleton B.;Covington, W. Wallace; Blake, Julie A., comps. Ponderosapine ecosystems restoration and conservation: steps towardstewardship: Conference proceedings; 2000 25-27 April;Flagstaff,AZ.ProceedingsRMRS-P-22.Flagstaff,AZ:U.S.DepartmentofAgriculture,ForestService,RockyMountainResearchStation.188p.

�5USDA Forest Service Gen. Tech. Rep. RMRS-GTR-�83. 2006

Main,W.A.;Paananen,D.M.;Burgan,R.E.1990.FIREFAMILY1988.Gen.Tech.Rep.NC-138.St.Paul,MN:U.S.DepartmentofAgriculture,ForestService,NorthCentralForestExperimentStation.

Nelson,R.M.,Jr.2000.Predictionofdiurnalchangein10-hfuelstick moisture content. Canadian Journal of Forest Research.30:1071-1087.

NWCG–NationalFireDangerRatingSystem,WeatherStationStandards.2000.Variousauthors.PublicationoftheNWCGFireWeatherWorkingTeam.PMS426-3,May2000.

Reinhardt, E.D.; Crookston, N.L. 2003. The Fire and FuelsExtension to the Forst Vegetation Simulator. Gen. Tech.Rep. RMRS-GTR-116. Fort Collins, CO: U.S. Departmentof Agriculture, Forest Service, Rocky Mountain ResearchStation.209p.

Rothermel,R.C.1972.Amathematicalmodelforpredictingfirespreadinwildlandfuels.Res.Pap.INT-115.Ogden,UT:U.S.DepartmentofAgriculture,ForestService,IntermountainForestandRangeExperimentStation.40p.

Rothermel,R.C.1991.PredictingBehaviorandSizeofCrownFiresintheNorthernRockyMountains.Res.Pap.INT-438.Ogden,UT:U.S.DepartmentofAgriculture,ForestService,IntermountainForestandRangeExperimentStation.46p.

Scott,J.H.1999.NEXUS:asystemforassessingcrownfirehazard.FireManagementNotes.59(2):20–24.

Scott,J.H.2006.Comparisonofcrownfiremodelingsystemsusedinthreefiremanagementapplications.Res.Pap.RMRS-RP-58.FortCollins,CO:U.S.DepartmentofAgriculture,ForestSer-vice,RockyMountainResearchStation.25p.

Scott,J.H.;Burgan,R.E.2005.Standardfirebehaviorfuelmodels:acomprehensivesetforusewithRothermel’ssurfacefirespreadmodel.Gen.Tech.Rep.RMRS-GTR-153.FortCollins,CO:U.S.Department of Agriculture, Forest Service, Rocky MountainResearchStation.72p.

Scott,J.H.;Reinhardt,E.D.2001.Assessingcrownfirepotentialbylinkingmodelsofsurfaceandcrownfirebehavior.Res.Pap.RMRS-RP-29.FortCollins,CO:U.S.DepartmentofAgriculture,ForestService,RockyMountainResearchStation.59p.

Scott,J.H.;Reinhardt,E.D.2005.Stereophotoguideforestimatingcanopyfuelcharacteristics inconiferstands.Gen.Tech.Rep.RMRS-GTR-145.FortCollins,CO:U.S.DepartmentofAgricul-ture,ForestService,RockyMountainResearchStation.49p.

Stratton,R.D.2004.Assessingtheeffectivenessoflandscapefueltreatments on fire growth and behavior. Journal of Forestry,Oct.-Nov.102(7):32-40.

Thomas, P.H. 1963. The size of flames from natural fires. In:Proceedingsninthsymposiumoncombustion.Pages844-859.Ithaca,NY.

U.S. Department of Interior, Geological Survey. 1990. Digitalelevationmodels.U.S.DepartmentofInterior,GeologicSurveyDataUser’sGuide.Reston,VA,Vol.5.51p.

U.S. Department of Agriculture, Forest Service. 1993. NationalInteragency Fire Management Integrated Database referencemanual. Washington, DC: U.S. Department of Agriculture,ForestService.

U.S.DepartmentofAgriculture,ForestService.1996.KCFAST:Kansas City Fire Access Software user’s guide. Washington,DC:U.S.DepartmentofAgriculture,ForestService,FireandAviationManagement.

U.S. Department of Agriculture, Forest Service. 2000. RERAPUser’sGuide(version5.03).NationalFireandAviationManage-mentInformationSystemsTeam.Boise,ID.154p.

VanWagner,C.E.1977.Conditionsforthestartandspreadofacrownfire.CanadianJournalofForestResearch.71(3):23-34.

Van Wagner, C.E. 1993. Prediction of crown fire behavior intwostandsofjackpine.CanadianJournalofForestResearch:442-449.

Wiitala, M.R.; Carlton, D.W. 1994. Assessing long-term firemovement risk in wilderness fire management. Proceedingsofthe12thconferenceonFireandForestMeteorology,26-28October,JekyllIsland,Georgia.SocietyofAmericanForesters:Bethesda,MD.

Zachariassen, J.; Zeller, K.F.; Nikolov, N.; and McClelland, T.2003.AreviewoftheForestServiceRemoteAutomatedWeatherStation (RAWS) network. Gen. Tech. Rep. RMRS-GTR-119.FortCollins,CO:U.S.DepartmentofAgriculture,ForestService,RockyMountainResearchStation.153p+CD.

Federal Recycling Program Printed on Recycled Paper

The U.S. Department of Agriculture (USDA) prohibits discrimination in all its pro-grams and activities on the basis of race, color, national origin, age, disability, and where applicable, sex, marital status, familial status, parental status, religion, sexual orientation, genetic information, political beliefs, reprisal, or because all or part of an individual’s income is derived from any public assistance program. (Not all prohibited bases apply to all programs.) Persons with disabilities who require alternative means for communication of program information (Braille, large print, audiotape, etc.) should contact USDA’s TARGET Center at (202) 720-2600 (voice and TDD). To file a complaint of discrimination, write to USDA, Director, Office of Civil Rights, 1400 Independence Avenue, S.W., Washington, DC 20250-9410, or call (800) 795-3272 (voice) or (202) 720-6382 (TDD). USDA is an equal opportunity provider and employer.

The Rocky Mountain Research Station develops scientific information and technology to improve management, protection, and use of the forests and rangelands. Research is designed to meet the needs of National Forest managers, Federal and State agencies, public and private organizations, academic institutions, industry, and individuals. Studies accelerate solutions to problems involving ecosystems, range, forests, water, recreation, fire, resource inventory, land reclamation, community sustainability, forest engineering technology, multiple use economics, wildlife and fish habitat, and forest insects and diseases. Studies are conducted cooperatively, and applications may be found worldwide.

Research LocationsFlagstaff, Arizona Reno, NevadaFort Collins, Colorado* Albuquerque, New MexicoBoise, Idaho Rapid City, South DakotaMoscow, Idaho Logan, UtahBozeman, Montana Ogden, UtahMissoula, Montana Provo, Utah

*Station Headquarters, Natural Resources Research Center,2150 Centre Avenue, Building A, Fort Collins, CO 80526

RMRSROCKY MOUNTAIN RESEARCH STATION

guidance on spatial wildland fire analysis - fs.fed.us · share a common foundation of procedures...

Documents