Baseline Wildland Fires and Emissions for the Western United States

By Todd J. Hawbaker and Zhiliang Zhu

Chapter 3 ofBaseline and Projected Future Carbon Storage and Greenhouse-Gas Fluxes in Ecosystems of the Western United StatesEdited by Zhiliang Zhu and Bradley C. Reed

Professional Paper 1797

U.S. Department of the InteriorU.S. Geological Survey

U.S. Department of the InteriorKEN SALAZAR, Secretary

U.S. Geological SurveyMarcia K. McNutt, Director

U.S. Geological Survey, Reston, Virginia: 2012

For more information on the USGS—the Federal source for science about the Earth, its natural and living resources, natural hazards, and the environment, visit http://www.usgs.gov or call 1–888–ASK–USGS.

For an overview of USGS information products, including maps, imagery, and publications, visit http://www.usgs.gov/pubprod

To order this and other USGS information products, visit http://store.usgs.gov

Any use of trade, product, or firm names is for descriptive purposes only and does not imply endorsement by the U.S. Government.

Although this report is in the public domain, permission must be secured from the individual copyright owners to reproduce any copyrighted materials contained within this report.

Suggested citation:Hawbaker, T.J., and Zhu, Zhiliang, 2012, Baseline wildland fires and emissions for the Western United States, chap. 3 of Zhu, Zhiliang, and Reed, B.C., eds., Baseline and projected future carbon storage and greenhouse-gas fluxes in ecosystems of the Western United States: U.S. Geological Survey Professional Paper 1797, 10 p. (Also available at http://pubs.usgs/gov/pp/1797.)

iii

Contents

3.1. Highlights..................................................................................................................................................13.2. Introduction..............................................................................................................................................13.3. Input Data and Methods ........................................................................................................................23.4. Results ......................................................................................................................................................43.5. Discussion ................................................................................................................................................73.6. Limitations and Uncertainties ...............................................................................................................73.7. Implications for Management and Mitigation ...................................................................................8

Figures 3.1. Flowchart showing the process for calculating baseline estimates of

greenhouse-gas emissions from wildland fires ………………………………… 3 3.2. Graphs showing the annual number of wildland fires, area burned, and

emissions for the baseline time period (2001–2008) in the Western United States …………………………………………………………………………… 5

3.3. Graphs showing the annual number of wildland fires, area burned, and emissions for the baseline time period (2001–2008) for level II ecoregions in the Western United States ……………………………………………………… 6

Tables 3.1. Summary statistics for the number of wildland fires, area burned, and

emissions, by EPA level II ecoregion and for the entire Western United States region between 2001 and 2008 ………………………………………………… 4

This page intentionally blank

Chapter 3. Baseline Wildland Fires and Emissions for the Western United States

By Todd J. Hawbaker1 and Zhiliang Zhu2

3.1. Highlights

• Wildlandfiresburnedanannualaverageof13,173km2between2001and2008intheWesternUnitedStates.

• Theinterannualvariabilityintheareathatwasburnedbetween2001and2008washigh;asfewas3,345km2burnedin2004andasmuchas25,206km2burnedin2007.

• Theannualaverageemissionsfromwildlandfiresfrom2001to2008were36.7TgCO2-eq/yr(10.0TgC/yr),withamedianvalueof41.0TgCO2-eq/yr(11.2TgC/yr),andarangefrom6.8TgCO2-eq(1.9TgC/yr)in2004to75.3TgCO2-eq(20.6TgC/yr)in2007.

• Theminimum,average,andmaximumemissionsfromwildlandfiresintheWesternUnitedStatesfrom2001to2008wereequivalentto7.9percent,11.6percent,and87.0percent,respectively,ofthemeanterrestrialcarbonsequestrationestimatedfortheWesternUnitedStatesinthisstudy.

• Theminimum,average,andmaximumemissionsfromannualwildlandfiresintheWesternUnitedStatesfrom2001to2008wereequivalentto0.1percent,0.7percent,and1.3percent,respectively,ofthe2010fossil-fuelemissionsfortheentireUnitedStates.

• TheWesternCordilleraecoregionproduced77percentofallemissionsintheWesternUnitedStatesduringthebaselineperiod.

3.2. IntroductionThemethodologyforthisassessmentexplicitlyaddressed

ecosystemdisturbances,includinghuman-andnaturallycausedwildlandfires,asrequiredbytheEISAlegislation(U.S.Congress,2007;Zhuandothers,2010).Asindicatedbyfigure1.2inchapter1ofthisreport,theestimatesforthebaselinebiomasscombustionemissionsfromwildlandfiresarepresentedinthischapter.Theprojectedfuturepotentialwildlandfireemissionsaredescribedinchapter8.Thebaselineburnedareasandtheprojectedfuturepotentialburnedareasforwildlandfiresandtheirseverity,describedinthetwochapters,wereusedasinputintotheassessmentofbaselineandprojectedfuturepotentialterrestrialcarbonandgreenhouse-gasfluxes(chapters5and9ofthisreport,respectively).

Wildlandfiresareacriticalcomponentoftheglobalcarboncyclebecausetheyproduceanimmediatereleaseofgreenhousegases—carbonmonoxide(CO),carbondioxide(CO2),andmethane(CH4)—whenbiomassisconsumedthroughcombustion(SeilerandCrutzen,1980).Wildlandfiresalsohavelong-termeffectsonecosystemcarboncyclingbyinfluencingtherateofcarbonsequestrationaftercombustion,boththroughthedecompositionofdeadvegetationandthroughphotosynthesis,whichhelpsestablishnewvegetation;becauseofthoseeffects,yearstodecadescanpassbeforecarbonstocksreturntopre-fireconditions(M.G.Turnerandothers,1998;Clearyandothers,2010;HurteauandBrooks,2011;Kashianandothers,inpress).Iffireregimesarestable,thelong-termeffectsofwildlandfiresoncarbon

1U.S.GeologicalSurvey,Denver,Colo.2U.S.GeologicalSurvey,Reston,Va.

2 Baseline and Projected Future Carbon Storage and Greenhouse-Gas Fluxes in Ecosystems of the Western United States

cyclingaretypicallynegligiblebecausecarbonsequestrationthroughvegetationgrowthandcarbonlossthroughwildlandfireemissionscancelouteachotheroverlongtimeperiods(Balshi,McGuire,Duffy,Flanigan,Kicklighter,andMelillo,2009;Flanniganandothers,2009).Ifafireregimechanges,however,thevulnerabilityforcarbonstorageishighbecausetheamountofcarbonstoredintheecosystemcanbealteredorlostthroughemissions.Substantialevidenceisavailabletodocumentthatfireregimesarenotstatic.Forexample,thefrequencyofwildlandfireshasbeengreatlyreducedsincesettlementoftheUnitedStatesbegan,mainlyduetoland-usechangesandthesuccessoffiresuppressioninthelastcentury(Clelandandothers,2004);however,thefrequencyofwildlandfireshasbeenincreasingsincethe1990sbecauseofanincreasinglyearliersnowmelt(Westerlingandothers,2006).Therefore,anycredibleassessmentofcarbonstorageandfluxesinecosystemsthroughtimemustaccountforthepotentialchangesinwildland-fireoccurrenceandemissions.

IntheconterminousUnitedStates(CONUS),thenetcarbonfluxinecosystemsreportedbytheEnvironmentalProtectionAgency(EPA)was1,075TgCO2-eq/yr(293TgC/yr)in2010,themajorityofwhichwassequesteredwithinforests(EPA,2012).TheestimatesofemissionsfromwildlandfiresintheUnitedStateswerehighlyvariable;afterconvertingthereportedemissionstocarbondioxideequivalent,theywereasfollows:(1)from15to73TgCO2-eq/yr,(2001–2008;GlobalFireEmissionsDatabase(GFED),OakRidgeNationalLaboratoryDistributedActiveArchiveCenter,2012;seealsoGiglioandothers,2010;vanderWerfandothers,2010);(2)from29to199TgCO2-eq/yr(2001–2008;Frenchandothers,2011;MichiganTechResearchInstitute,2012),and(3)from157to283TgCO2-eq/yr(2002–2006;WiedinmyerandNeff,2007).Whencomparedwiththe2010netcarbonfluxofecosystemsintheCONUS(EPA,2012),theannualemissionsfromwildlandfireswereequivalentto1to26percentoftheecosystem’stotalannualflux.Incontrast,from2001to2008,thecombustionoffossilfuelsproduced5,642TgCO2-eq/yr(EPA,2012)andfluxincreasedatarateof1percentperyear(Pacalaandothers,2007).Basedontheserates,theannualemissionsfromwildlandfireswereequivalentto0.3to5.1percentoftheemissionsfromfossil-fuelconsumption.

Thedifferencesamongtheaccuracyandqualityofthesedata,theirspatialandtemporalresolution,andassumptionsaboutvariationsincombustionefficiencyweretheprimarysourcesofuncertaintiesinwildland-fireemissionsestimates(Larkinandothers,2009;Frenchandothers,2011).Theassumptionsabouttheproportionofabovegroundbiomassconsumedbywildlandfire,especiallyabovegroundwoodybiomassinforests,canhaveasubstantialinfluenceonemissionestimates(Campbellandothers,2007;Meigsandothers,2009).Themethodsusedforcalculatingemissionsreliedonestimatesoftheareathatwasburned,fuelloads(availableliveanddeadbiomassforburning),combustionefficiency,andemissionfactors(SeilerandCrutzen,1980;

Albiniandothers,1995;WiedinmyerandNeff,2007;Ottmarandothers,2008).Forinstance,theGFED(Giglioandothers,2010;vanderWerfandothers,2010)estimatesbiomassconsumptionandemissionatfirelocationsdetectedbyMODIS(Royandothers,2002;Giglioandothers,2003)onthebasisofland-covertypes,combustioncompleteness,soilmoisture,andland-coverspecificemissionfactors.GFEDalsoincorporateschangesinfuelloadsusingtheCarnegieAmesStanfordApproachtocharacterizebiomassproduction(Potterandothers,1993,2012).WiedinmyerandNeff(2007)alsousedactivewildlandfireobservationsfromNASA’sMODIS(Giglioandothers,2003),butcalculatedtheemissionsbasedonstaticland-covertypes,percentlandcover,andbiomassat1-kmresolution.Frenchandothers(2011)usedtheCONSUMEmodel(Ottmarandothers,2008),whichcalculatedfuelconsumptionandemissionusingfuelloadsderivedfromtheFuelCharacterizationClassificationSystem(FCCS;Ottmarandothers,2007)andfuelmoisturesderivedfromweather-stationdata.

Theseexistingstudiesprovidedanestimateoftheeffectsofwildlandfiresonanationalscalebutlackedtheregionaldetailrequiredbythisassessment.Furthermore,therewerefewprojectionsoffuturepotentialwildland-fireemissionsthatwereconsistentwiththeexistingbaselineemissionestimates.Therefore,asetofbaselineemissions(thischapter)andprojectedfuturepotentialemissions(chapter8)wasdevelopedtoensureconsistencythroughoutthisandotherregionalassessments.Thischapterfocusesonthebaselineestimatesofwildland-fireoccurrenceandemissionsfortheWesternUnitedStatesandstrivestoanswertwoprimaryquestions:(1)Whatwerethepatternsofwildland-fireoccurrenceandemissionsintheWesternUnitedStates?,and(2)HowdidwildlandfiresvarytemporallyandspatiallyamongtheecoregionsoftheWesternUnitedStates?Theresultsfromthischapterwerealsousedintheassessmentofbaselinecarbonstorage,sequestration,andgreenhouse-gasfluxesofterrestrialecosystems(chapter5).

3.3. Input Data and MethodsThebaselineestimatesforthenumberofwildfires,the

areaburned,andemissionswerederivedfromtheMonitoringTrendsinBurnSeverity(MTBS)database(Eidenshinkandothers,2007)andtheFirstOrderFireEffectsModel(FOFEM;Reinhardtandothers,1997)forthemajorGHGs:carbondioxide(CO2),carbonmonoxide(CO),andmethane(CH4).ThismethodwasappliedtoeachwildlandfireintheregionthatwasintheMTBSdatabasetoproduceestimatesofCO2,CO,andCH4emissions(convertedtoCO2equivalent).TheresultswereaggregatedtoproduceanestimateofemissionsfortheWesternUnitedStatesasawholeandforeachlevelIIecoregionwithinit.

Chapter 3 3

Thelocationsofwildland-firescarsweretakenfromtheMTBSdatabase,whichincludesinformationaboutwildlandfiresintheWesternUnitedStatesthatoccurredbetween1984and2008andthatweregreaterthan404ha(1,000acres).TheMTBSdatawereselectedbecauseofthehighdegreeofconfidenceinthedata(firesizeandseveritywerederivedmanually).Otherdatasourceswereconsideredbutultimatelynotused,includingthevariousversionsofFederalwildland-firedatabases,becauseofthespatialinaccuraciesandduplicaterecordsthatwouldhaveintroduceduncertainties(T.J.Brownandothers,2002).TheMODISactivefire(Giglioandothers,2003)andburnedarea(Royandothers,2002)productswerealsoconsideredbutnotusedbecausetheycontainednoinformationaboutthecausesofthewildlandfiresandhadacoarsespatialresolutionthatcomplicatedthecalculationsofemissionsandthemodelingoffuturetrends.

Wildland-fireemissionswerecalculatedforeachburnedpixelintheMTBSdatabaseusingtheFOFEM,whichusedfuelloadsalongwithfuelmoisturestoestimatetheamountofforestlitteranddowneddeadwoodthatwasconsumed(Albiniandothers,1995;AlbiniandReinhardt,1995,1997).Theconsumptionofduff(decayingforestlitter),trees,plants,andshrubswasestimatedasafunctionoftheregion,season,fuelmoistures,andfuelloads.CanopyfuelconsumptionwasestimatedasafunctionoftheburnseverityprovidedbytheMTBSdata.TheemissionsofCO2,CO,andCH4werethencalculatedonthebasisoftheamountoffuelconsumed,theorganic-mattercontentofthefuel,andhowefficientlyitburned.TherequiredinputdatafortheFOFEMincludedfuelloads,burnseverity,anddeadandlivefuelmoistures(fig.3.1).

Fuel-loaddataprovidedanestimateoftheamountofbiomassthatwasavailableforconsumptionandwerederivedfromtheLANDFIREproject’sFuelLoadingModelsdatalayer(FLM;Lutesandothers,2009).Thesefuel-loaddatawerecategorizedby1-,10-,100-,and1,000-hourfuelclasses(Lutesandothers,2009)fordeadbiomass,decayingbiomass(duffandforestlitter),andlivebiomass(grass,shrubs,andtreecanopy).IntheFOFEM,theamountoftreecanopythatwasconsumedwasadirectfunctionofburn-severityvaluesfromtheMTBSdata.Theamountofcanopyfoliagethatwasconsumedinthehigh,moderate,andlowburn-severitycategorieswasassumedtobe100,60,and20percent,respectively.Similarly,theconsumptionofthecanopy’sbranchwoodwassetat50,30,and10percentforthehigh,moderate,andlowburn-severitycategories,respectively.Thesevalueswerebasedonpreviouslypublishedestimates(Spracklenandothers,2009;Zhuandothers,2010)andonacomparisonoftheFOFEMemissionswithpreviouslypublishedresultsforselectedwildlandfires.PriortoprocessingwiththeFOFEM,theLANDFIREFLMandMTBSrasterdatawereaggregatedto250mresolutionusinganearest-neighbormethodtomatchtheresolutionofother

MTBS LANDFIRE Weather

Fire effects(FOFEM)

Burnedarea

Emissions

Figure 3–1.

Figure 3.1. Flowchart showing the process for calculating baseline estimates of greenhouse-gas emissions from wildland fires.

rasterdatabeingusedinthisassessment.Theemissionswerecalculatedforwildlandfiresoccurringbetween2001and2008,whichisthebaselinetimeperiodforthiscomponentoftheassessment.Wildlandfiresbefore2001wereexcludedbecausetheLANDFIREfuelsdatawerederivedfromcirca2001Landsatimagery.Theyear-to-yearvariabilityintheamountofburnedareawashigh.Therefore,dataonallwildlandfiresthatoccurredafter2001wereincludedtohelpreducetheinfluenceofextremewildlandfireyears(eitherhighorlowoccurrence)onthebaselinestatistics.

FuelmoistureswereestimatedbyapplyingtheNationalFireDangerRatingSystem(NFDRS;Bradshawandothers,1983)algorithmstoa1/8°griddeddailyweatherdatasetthatspannedtheconterminousUnitedStates(Maurerandothers,2002).Thesedataspanned1950to2010andincludedtheminimumandmaximumdailytemperatureanddailyprecipitation.TheMountainClimateSimulator(MT–CLIM)algorithms(GlassyandRunning,1994)wereusedtocalculaterelativehumiditybasedonminimumandafternoondailytemperatures(Kimballandothers,1997).DeadandlivefuelmoistureswerecalculatedbyusingtheNFDRS(Deemingandothers,1977;Bradshawandothers,1983;Burgan,1988).TheNFDRSalgorithmsrequiredinformationaboutthebeginningofbothspring(“green-up”)andfall(“brown-down”)toestimatelivefuelmoistures.Togeneratethegreen-upandbrown-downdates,atechniquewasimplementedthatdeterminedthedatesofseasonalchangesinlivefuelonthebasisofthedailyamountofexposuretolight,minimumtemperature,andthevapor-pressuredeficit(Jollyandothers,2005).

4 Baseline and Projected Future Carbon Storage and Greenhouse-Gas Fluxes in Ecosystems of the Western United States

3.4. ResultsInthefiveecoregionsofthisassessment,thenumberof

wildlandfiresbetween2001and2008averaged303peryear,butwasashighas467in2006andaslowas131in2004(table3.1andfig.3.2).Theareaburnedbywildlandfiresaveraged13,173km2/yror0.49percentoftheWesternUnitedStates,whichhasatotalareaofapproximately2.66millionkm2.Theareaburnedrangedfrom3,345km2(0.13percentoftotalarea)in2004to25,206km2(0.94percentoftotalarea)in2007.Theemissionsfromwildlandfiresandtheirinterannualvariabilityfollowedthepatternsoftheburnedareaandaveraged36.7TgCO2-eq/yr,rangingfrom6.8TgCO2-eqin2004to75.3TgCO2-eqin2007.Whennormalizedfortheareaburned,theannualemissionsaveraged2.9kgCO2-eq/m2/yr,butrangedfrom1.7to3.9kgCO2-eq/m2/yr(or0.8kgC/m2/yr,rangingfrom0.4to1.1kgC/m2/yr).

TheWesternCordilleraecoregionhadthemostwildlandfiresandthehighestemissionsamongalltheWesternUnitedStatesecoregions(table3.1andfig.3.3A).Thenumberofwildlandfiresbetween2001and2008averaged123/yrandrangedfrom61in2004to173in2003.Theburnedareaandtheemissionsaveraged5,708km2/yrand28.2TgCO2-eq/ yr,respectively.Theinterannualvariabilitywashighandtheareaburnedandtheemissionsrangedbetween1,856km2and4.7TgCO2-eqin2004and10,449km2and64.1TgCO2-eqin2007,respectively.TheWesternCordilleracoversapproximately872,000km2.Afternormalizingtheareaburnedandtheemissionsfortheecoregion’sarea,theannualareaburnedrangedbetween0.21and1.20percentandtheannualemissionsrangedbetween2.6and6.2kgCO2-eq/m2/yrfortheentireWesternCordillera.

Table 3.1. Summary statistics for the number of wildland fires, area burned, and emissions, by EPA level II ecoregion and for the entire Western United States region between 2001 and 2008.

[km2,squarekilometers;TgCO2-eq,teragramsofcarbondioxideequivalent]

Western Cordillera

Marine West Coast Forest

Cold Deserts

Warm Deserts

Mediterranean California

Western United States

Number of wildfires per year

Mean 123 5 121 25 31 303Standarddeviation 34.1 5.5 72.4 20.0 7.8 97.1Minimum 61 1 36 7 15 131Median 117 3 115 19 32 29095thquantile 168 12 225 57 39 434Maximum 173 13 245 66 39 467

Area burned per year (km2)

Mean 5,708 79 5,056 785 1,585 13,173Standarddeviation 2,966 87 4,212 702 1,198 6,736Minimum 1,856 9 583 198 301 3,345Median 5,371 58 3,962 544 1,220 12,13695thquantile 9,926 178 10,926 1,920 3,292 23,261Maximum 10,449 191 11,237 2,217 3,304 25,206

Emissions per year (TgCO2-eq)

Mean 28.2 0.3 4.1 2.4 1.8 36.7Standarddeviation 18.9 0.3 2.6 2.2 1.4 21.3Minimum 4.7 0.0 0.8 0.5 0.3 6.8Median 28.8 0.2 4.0 2.1 1.3 41.095thquantile 54.7 0.6 7.1 6.0 3.6 65.0Maximum 64.1 0.7 7.1 7.4 3.7 75.3

Chapter 3 5

2001 2002 2003 2004

Year2005 2006 2007 2008

Num

ber o

f wild

fires

0

200

400

600

Area

bur

ned,

in s

quar

e ki

lom

eter

s

0

10,000

30,000

20,000

0

20

40

60

80

Figure 3–2.

Emis

sion

s, in

tera

gram

s of

carb

on d

ioxi

de e

quiv

alen

t

Figure 3.2. Graphs showing the annual number of wildland fires, area burned, and emissions for the baseline time period (2001–2008) in the Western United States.

OfalltheecoregionsintheWesternUnitedStates,theMarineWestCoastForestecoregionhadthefewestwildlandfires,thesmallestareaburned,andthelowestemissions(table3.1andfig.3.3B).Thenumberofwildlandfiresaveraged5/ yrandrangedfrom1to13.Theareaburnedrangedfrom9to200km2andaveraged79km2/yr.TheemissionsfromtheMarineWestCoastForestrangedfrom

0to0.7TgCO2-eq/yrandaveraged0.3TgCO2-eq/yr.TheareaoftheMarineWestCoastForestecoregionisapproximately98,000km2.Afternormalizingthewildland-fire-occurrencestatisticsforarea,from0.01to0.13percentoftheecoregionburnedeachyearandemissionsrangedfrom2.6to4.8kgCO2-eq/m2/yr.

TheColdDesertsecoregionisthelargestecoregionintheWesternUnitedStatesat1millionkm2andhadnearlyasmuchwildland-fireactivityastheWesternCordillera(table3.1andfig.3.3C).Onaverage,therewere121wildlandfiresperyear,buttheinterannualvariabilitywashigh;asfewas36wildlandfireswereobservedin2004andasmanyas245wereobservedin2006.TheamountofareaburnedeachyearintheColdDesertsecoregionwassimilartothatoftheWesternCordilleraandaveraged5,056km2/yr,rangingfrom583km2in2004to11,237km2in2007.Thisrangeisequivalentto0.06to1.07percentoftheareaofentireecoregion;however,becausethevegetationintheColdDesertsispredominantlygrassandshrubs,emissionswerelower,averagingonly4.1TgCO2-eq/yrandrangingfrom0.8to7.1TgCO2-eqin2004and2007,respectively.

IntheWarmDesertsecoregion,wildlandfireswereinfrequentwithanaverageof25wildfiresperyear,butasmanyas66in2005andasfewas7in2004(table3.1andfig.3.3D).Theamountofareaburnedwasalsosmallandrangedfrom198km2in2004to2,217km2in2005withanaverageof785km2/yr.Thisrangeequatedto0.16to0.46percentoftheecoregionarea,whichwas478,000km2.Theemissionsweregenerallylow,withanaverageof2.4TgCO2-eq/yr;thevariabilityinemissionswashigh,however,andvariedfrom0.5TgCO2-eqin2004to7.4TgCO2-eqin2008.Whennormalizedfortheareaburned,theemissionsrangedfrom1.0to3.5kgCO2-eq/m2/yr.

TheMediterraneanCaliforniaecoregionisthesmallestintheWesternUnitedStates(173,000km2)butstillhadasubstantialamountofwildland-fireactivity.Onaverage,therewere31wildlandfiresperyearwitharangefrom15to39(table3.1andfig.3.3E).Theareaburnedaveraged1,585km2/yrandhadarangeof301to3,304km2in2001and2003,respectively.Theemissionsaveraged1.8TgCO2-eq/ yrbutrangedfrom0.3to3.7TgCO2-eqin2005and2008,respectively.Theamountofareaburnedandemissionswerelargerelativetothetotalareaoftheecoregion.Theareaburnedrangedfrom0.17to1.91percentoftheentireecoregion(thehighestpercentareaintheWesternUnitedStates),andtheemissionsrangedfrom0.6to1.2kgCO2-eq/ m2/yr.

6 Baseline and Projected Future Carbon Storage and Greenhouse-Gas Fluxes in Ecosystems of the Western United States

2001 2002 2003 2004 2005Year Year Year

2006 2007 2008 2001 2002 2003 2004 2005 2006 2007 2008 2001 2002 2003 2004 2005 2006 2007 2008

Number of wildfires Area burned,in square kilometers

0

50

100

150

200A. Western Cordillera

0

4,000

8,000

12,000

0

20

40

60

80

0

4

8

12

14B. Marine West Coast Forest

0

50

100

150

200

250

0

0.2

0.4

0.6

0.8

0

100

200

300C. Cold Deserts

0

4,000

12,000

8,000

0

2

4

6

8

0

20

40

60

80D. Warm Deserts

1,000

1,500

500

0

2,000

2,500

0

2

4

6

8

0

10

20

30

40

50E. Mediterranean California

0

1,000

3,000

4,000

2,000

0

1

2

3

4

Figure 3–3.

Emissions, in teragrams ofcarbon dioxide equivalent

Figure 3.3. Graphs showing the annual number of wildland fires, area burned, and emissions for the baseline time period (2001–2008) for level II ecoregions in the Western United States (EPA, 1999). The vertical scales are not constant among ecoregions.

Chapter 3 7

3.5. DiscussionFrom2001to2008,wildland-fireactivityinthe

WesternUnitedStateswassubstantial;largewildlandfiresnumberedbetween131and467peryearandburnedfrom3,345to25,206km2eachyear.Thesewesternwildlandfiresrepresented58percentofallthewildlandfiresthatoccurrednationwidefrom2001to2008andaremappedintheMTBSdatabase,andaccountedfor75percentoftheallareaburned.Theannualemissionsaveraged36.7TgCO2-eq/yr,whichwasequivalentto0.7percentofthenationwidefossil-fuelemissionsin2010(5,594TgCO2-eq/yr;EPA,2012).Theinterannualvariabilityinemissionswashighandrangedfrom6.8to75.3TgCO2-eqintheWesternUnitedStates,whichwasequivalentto0.1to1.3percentofthenationwidefossil-fuelemissions,respectively.Thus,therelativecontributionofwildlandfiresoftheWesternUnitedStatestonationwidegreenhouse-gasemissionswassmallwhencomparedtothecontributionofanthropogenicallygeneratedemissions.

Therewasalargeamountofyear-to-yearvariabilityinwildland-fireoccurrences,areaburned,andemissionsamongthefiveecoregionsintheWesternUnitedStates(table3.1andfig.3.3),whichwasrelatedtodifferencesinvegetationandfireregimes.TheMarineWestCoastForestandWesternCordilleraecoregionsaredominatedbyconiferousforests,buthavequitedifferentfireregimes.PrecipitationishighintheMarineWestCoastsForest,whichresultsinhighlyproductivevegetation,buttheinfrequentwildlandfirescanbesevere(Agee,1993).TheforestsintheWesternCordilleratendtoexistinadrierclimatewithmorefrequentwildlandfires,buttheseverityismixedanddependsondroughtandvegetationconditions(Schoennagelandothers,2004).InboththeMarineWestCoastForestandWesternCordilleraecoregions,wildland-fireemissionscanbehighundersevereconditionsbecauselargeamountsofcanopyfuelscanbeconsumedbyacrownfire.TheMarineWestCoastForestdidnotproduceasubstantialamountofemissionsinthisanalysis,butthatmaybeinpartbecausethewildland-firefrequencywaslowinthisecoregionandthe25-yearspancoveredbytheMTBSdidnotincludeasizeableamountoffireactivitythere.Theresultsofthisassessment,however,dohighlighttheimportanceofwildland-fireemissionsintheWesternCordilleraatbotharegionalandnationalscalebecausethisecoregionproduced77percentofallemissionsintheWesternUnitedStatesduringthebaselineperiod.

IncontrasttotheMarineWestCoastForestandtheWesternCordillera,theColdDeserts,WarmDeserts,andMediterraneanCaliforniaecoregionsaredominatedbythegrasslands/shrublandsecosystemandotherecosystems(primarilydeserts).WildlandfireswerefrequentinallthreeecoregionsbutweremorecommonintheColdDesertsandMediterraneanCaliforniathanintheWarmDeserts.Thefire

regimesinthedesertecoregionswererelatedtovegetationproductivityandclimate(Mensingandothers,2006;BrooksandChambers,2011).ThesesamedriversinfluencedwildlandfiresintheMediterraneanCaliforniaecoregion,buthumaninfluencesandextremewindsalsoplayedalargerole(Syphardandothers,2007;Moritzandothers,2010).Invasivespecieswerealsopresentinallthreeecoregions,oftendisplacingnativevegetationandthusincreasingwildland-firefrequency(D’AntonioandVitousek,1992;Brooksandothers,2004;Keeley,2006).Theseresultssuggestthatanextensiveamountofareacanburnintheseregions,buttheemissionsaregenerallylowerthanthoseoftheforestedecoregionssimplybecauseofthedifferenceintheamountofavailablefuel.

3.6. Limitations and UncertaintiesTheMTBSdatausedinthisassessmentdidnotinclude

smallwildlandfires,buttheystillcapturedthemajorityoftheareaburnedbecausetheyincludedthelargestwildlandfireswhichcontributedmosttotheamountofareaburned(Straussandothers,1989;Stocksandothers,2002).AcomparisonoftheMTBSdatawiththeFederalwildland-fire-occurrencedatabase(U.SDepartmentoftheInterior,2012)showedthattheMTBSlistedonly2percentofallwildlandfiresbutthat2percentaccountedfor80percentoftheareaburnedinthefiveecoregionscoveredbythisassessment.Therefore,theresultsofthisassessmentcapturedthegeneralpatternsandtrendsofwesternwildlandfires,butprovidedaslightunderestimateofwildland-fireemissions.

Inthisassessment,theestimatesofareaburnedandofemissionsalsodidnotincludetheinfluenceofprescribedandagriculturalfires(forexample,burningcropresidues);however,theemissionsproducedbythosetypesoffiresweresuspectedtobelowrelativetothewildland-fireemissions.Theinfluenceofprescribedfiresonemissionswasdifficulttoassessbecausethedatacharacterizingprescribedfiresweregenerallypoorbasedoninconsistentreportingaboutthemacrossthecountry.Theexistingestimatesofemissionsfromprescribedfiressuggestedthattheyproducedonly10percentoftheemissionsfromwildlandfires(Y.Liu,2004),inpartbecauseprescribedfiresusuallyburnunderlessextrememeteorologicalconditionsthanwildlandfires.Theinfluenceofagriculturalfireswasalsoestimatedtobeabout10percentofthewildland-fireemissionsintheGFEDdatabase.IntheWesternUnitedStates,therelativeamountofemissionsproducedbyprescribedandagriculturalfireswaslikelytobeevenlower,becauseagriculturalfiresweremorecommonintheGreatPlainsandtheEasternUnitedStates(Korontziandothers,2006;Tulbureandothers,2011)andprescribedfiresintheWesternUnitedStatesonlyaccountedfor22percentoftheareaburnedbyprescribedfiresnationwide(NationalInteragencyFireCenter,2012).

8 Baseline and Projected Future Carbon Storage and Greenhouse-Gas Fluxes in Ecosystems of the Western United States

TheemissionsresultsgeneratedforthisassessmentdifferedandweregenerallylowerthanpastestimatesofemissionsfortheWesternUnitedStates.ThehighestemissionsestimateswerefromWiedinmyerandNeff(2007),whousedtheactivewildlandfiredatafromtheMODISfrom2002to2006andestimatedthemeanandstandarddeviationofannualemissionsat105.0and42.0TgCO2-eq,respectively,forthefollowingStates:Arizona,California,Colorado,Idaho,Montana,NewMexico,Nevada,Oregon,Utah,Washington,andWyoming;thefiveecoregionsinthisassessmentdonotcovertheexactsamearea.Frenchandothers(2011)calculatedemissionsusingtheMTBSdataandCONSUMEmodel(Ottmarandothers,2008)forwildlandfiresoccurringfrom2001to2008;theirresultsareavailableonlineasecoregion-levelsummaries(MichiganTechResearchInstitute,2012).WhentheirresultsweresummarizedacrosstheWesternUnitedStates,theaverageandstandarddeviationofannualemissionswas78.4and43.3TgCO2-eq,respectively,whichisnearlytwicetheamountoftheemissionsestimatedinthisassessment.Theemissionestimatesfrombothoftheseanalysesweresubstantiallygreaterthantheestimatesgeneratedforthisassessment.Thedifferencesaremostlikelyduetodifferencesinmethods,data,andtheresolutionofthedataused.WiedinmyerandNeff(2007)reliedon1-kmresolution,active-wildland-firedatafromMODISandfuelsdatafromtheFuelsCharacterizationClassificationSystem(FCCS;USDAForestService,2012d).Theyalsoassumedthatalloftheavailablebiomasscouldpotentiallyburn,andthatisoftennotthecase,especiallyforwoodyfuels(Campbellandothers2007;Meigsandothers,2009).Frenchandothers(2011)alsousedthe1-km-resolutionFCCSfuelsdataandaggregated1-km-resolutionMTBSdata.Thefuelsdataintheirreportdifferedfromthedatalayerusedinthisassessmentbothintermsofinformationandresolution.Themethodsusedinthisassessmentmadeuseoffuelmoistures,whicharebasedongriddeddailyweatherdata.ThemethodsusedbyFrenchandothers(2011)alsomadeuseoffuelmoistures,butrecommended10percentlevelsfor1,000-houravailabilityandduffmoistures,whichareveryfavorableconditionsforcombustion.Thefulleffectsofthedifferencesinfuelmapsandmoisturelevelsonthisassessmentweredifficulttoassess,butthesecomparisonssuggestthattheresultsinthisassessmentaremoreconservativethanpreviouslypublishedestimatesofwildland-fireemissions.

3.7. Implications for Management and Mitigation

Forthisassessment,theeffectsofdifferentstrategiesforwildland-firemanagementandmitigationonfireemmissionswerenotexplicitlyaddressed,butthereisagrowingbodyofliteraturefromwhichtodrawsomeconclusions.Increasingtheeffectivenessoffiresuppression(whichincludesfirefighting,prevention,andeducation)isapopularfirstchoicebutunlikelytoworktoreducefireemmissionsgiventhatfiresuppressionoverthepast100yearshashadmixedsuccess(Rollinsandothers,2001;Keaneandothers,2002;StephensandRuth,2005).Insomeecoregions,reducingtheareathatcouldpotentiallybeburnedmaybepossibleandcriticaltomaintainecosystemhealth,especiallyinareaswherewildland-firefrequencyissuspectedtobeunnaturallyhighbecauseoftheincreaseinhumaninfluenceduetoarson,accidentalignitions,ortheaccidentalordeliberateintroductionofinvasivespecies,suchasinSouthernCalifornia(Keeleyandothers,1999)andinthegrasslands,shrublands,anddesertsoftheSouthwesternUnitedStateswherefire-adaptedinvasivespeciesarealteringwildland-firecycles(D’AntonioandVitousek,1992;Brooksandothers,2004;Keeley,2006).

Amoreeffectivewaytoreducewildland-fireemissionsfromforestsmaybetoimplementmanagementstrategiesdesignedtoreducewildland-fireseverity,whichisdirectlyrelatedtotheamountofbiomassconsumed.InmanypartsoftheWesternUnitedStates,firesuppressionhasresultedinunnaturallyhighfuelloadsproducingwildlandfiresthatareofuncharacteristicallyhighseverity(Stephens,1998;Keaneandothers,2002;Agee,2003;StephensandRuth,2005).Fueltreatments,includingmechanicalforestthinningandprescribedfires,aredesignedtoreducefuelloadssothatwildlandfiresarelessintenseandeasiertomanageiftheydooccur(AgeeandSkinner,2005;Reinhardtandothers,2008).Mostoftheevidencesuggeststhatfueltreatmentscanreducecarbonlossthroughwildland-fireemissionsoverthelongtermeventhoughthereisashort-termlossincarbonstorageduetobiomassremoval(Stephensandothers,2009;ReinhardtandHolsinger,2010;WiedinmyerandHurteau,2010;NorthandHurteau,2011).Suchtreatmentsaremosteffectiveinforestswherefuelloadsareuncharacteristicallyhighandmay

Chapter 3 9

notbeecologicallyappropriateinotherforestandvegetationtypes(Siboldandothers,2006;Mitchellandothers,2009).Toproduceanoticeableeffectataregionalscale,between20and40percentoftheWesternUnitedStates’forestsneedtobetreated(Finney,2001,2007).Thepotencyoffueltreatmentscanbeshort-lived,ontheorderof10to20years(Collinsandothers,2011);therefore,1to4percentoftheforestedlandscapewouldneedtobetreatedannually(Finneyandothers,2007).Giventhatwildlandfiresarerareevents,theproportionofthelandscapethatmustbetreatedismuchlargerthantheproportionofthelandscapethatburnsandbecauseofthatrelation,theamountofcarbonremovedfrombiomasspoolsinfueltreatmentsmaybegreaterthantheamountofcarbonprotectedfromfire(Campbellandothers,2012).

InsomenonforestedecosystemsintheWesternUnitedStates(especiallysouthernCalifornia,theGreatBasin,andtheSonoranandMojaveDeserts),thefrequenciesofwildlandfiresareuncharacteristicallyhighandaredrivenbyhumanignitionofsomefiresandbyinvasivespecieswhichprovideextrafuel(D’AntonioandVitousek,1992;Keeleyandothers,1999;Brooksandothers,2004;Keeley,2006).Wildland-fireemissionsinthesenonforestedecosystemsarelikelytobegreaterandcarbonstockslowerthaninhistorictimebecausenativewoodyvegetationhasbeenreplacedbyinvasive

grasses(Bradleyandothers,2006).Eventhoughwildland-fireemissionsfromthesenonforestedecosystemsarelowrelativetothosefromforestedecosystems,reducingthewildland-firefrequencytothehistoricalrangeofvariabilitymayresultinonlyslightlyreducedoverallwildland-fireemissions.InothernonforestedecosystemsintheWesternUnitedStates(suchasgrasslands/shrublands),theoppositehashappened.Grazingpressurehasreducedthegrasscoverandthefrequencyofwildlandfires,thusallowingwoodyvegetationtoexpanditsrangeandincreaseincover(VanAuken,2000).Theexpansionofwoodyvegetationusuallyresultsinanincreaseincarbonstocks(Asnerandothers,2003);however,itisuncertainiftheincreasedcarbonstockswillpersistoverlongtimeperiodsgivenchangesinclimateandfireregimes(Bargerandothers,2011).

Muchuncertaintyremainsabouttheshort-andlong-termeffectsofwildland-firemanagementoncarbonbudgetsinmanyecosystemsintheWesternUnitedStates.Anycarbon-managementstrategyfocusedonincreasingecologicalcarbonstocksorsequestrationratesshouldcarefullyconsiderthebenefitsandrisksofwildland-firemanagementoverbothshortandlongtimeperiods,aswellastheeffectsonotherecosystemcharacteristicsandservices(Jacksonandothers,2005;McKinleyandothers,2011;Olanderandothers,2012).

This page intentionally left blank.