27chAPTeR 3. Climate Change Summary

SteveMcNulty,JenniferMooreMyers, PeterCaldwell,andGeSun1

key FiNDiNGS

• Since1960,allbuttwosoutherncapitalcities(Montgomery,AL,andOklahomaCity,OK)haveexperiencedastatisticallysignificantincreaseinaverageannualtemperature(approximately0.016°C),butnonehasexperiencedsignificanttrendsinprecipitation.

•TheSouthisforecastedtoexperiencewarmertemperaturesforthedurationofthe21stcentury;forecastsaremixedforprecipitationchangesduringthesameperiod.

•Climatepredictionsrangefromwetandwarm (1167mm/19.06°C)tomoderateandwarm (1083mm/19.45°Cand1106mm/19.27°C)todryandhot(912mm/20.22°C).

iNTRoDucTioN

ThischaptersummarizestheclimatepredictionsthathavebeenusedthroughouttheSouthernForestFuturesProject(IPCC2007b).Fourdistinctcombinationsofgeneralcirculationmodels(GCMs)andspecialreportemissionsscenarioswereselectedasCornerstoneFutures.GCMsarecomplexmodelsthatprovidegeographicallyandphysicallyconsistentestimatesofregionalclimatechange(IPCC2009).Theemissionsscenariosareglobalstorylinesrepresentingalternativedemographic,socioeconomic,andenvironmentalfutures(Nakicenovic2000).

TheGCMsselectedfortheFuturesProjectweretheMK2andMK3.5fromtheAustralianCommonwealthScientificandIndustrialResearchOrganization(CSIRO),theHadCM3fromtheUnitedKingdomMeteorologicalCenter,andtheMIROC3.2fromtheJapaneseNationalInstituteforEnvironmentalStudies.

TwoemissionsscenarioswereselectedfortheFuturesProject.TheA1Bscenarioischaracterizedbylowpopulationgrowth,highenergyuse,andhigheconomicgrowth.TheB2scenarioischaracterizedbymediumpopulationgrowth,

1SteveMcNultyisSupervisoryEcologist,JenniferMooreMyersisaResourceInformationSpecialist,andPeterCaldwellandGeSunareResearchHydrologists,EasternForestEnvironmentalThreatAssessmentCenter,SouthernResearchStation,U.S.DepartmentofAgricultureForestService,Raleigh,NC27606.

mediumenergyuse,andmediumeconomicgrowth(IPCC2007b).Thesescenariosrepresenttwolevelsofglobalcarbondioxide(CO2)emissionsby2100:60gigatonsofCO2-equivalents(eq)(IPCC2007a)intheA1Bscenario(resultinginanatmosphericconcentrationofapproximately700ppm)(Solomonandothers2007)and65gigatonsofCO2-eq(IPCC2007a)intheB2scenario(resultinginanatmosphericconcentrationofapproximately600ppm)(Solomonandothers2007).TherelationshipbetweenCO2equivalentemissionsandatmosphericCO2concentrationisnotlinear,andtheestimatesfor2100areinfluencedbyemissionratesthroughoutthe21stcentury.TheA1Bscenariopeakshigheraround2050andtapersoff,whiletheB2scenarioincreasesmoreslowlyandsteadily.Forcomparison,carbondioxideemissionsfor2009wereestimatedat40gigatonsofCO2-eq(resultinginanatmosphericconcentrationof387partspermillion)(IPCC2007a,Tans2011).

TheFuturesProjectcombinesGCMsandemissionsscenariosintofourCornerstoneFutures—CSIROMK3.5+A1B,MIROC3.2+A1B,CSIROMK2+B2,andHadCM3+B2—whicharedescribedinthischapter.Althoughthischapterdoesnotdiscusssubregionalvariationsindetail,theGCMsummarydatahavebeenprovidedinbothtabularandgraphicformatstoallowthereadertoexamineclimatechangeimpactsforsubregionsofinterest.

DATA SouRceS AND meThoDS

BecausetheoriginalscaleoftheGCMswastoocoarseforregionalanalysis,theCornerstoneFuturesweredownscaledfromtheiroriginalresolutionofapproximately2degreesbytheWorldClimateResearchProgramme’sCoupledModelIntercomparisonProjectphase3(CMIP3)(Maurerandothers2007).EachGCMwasspatiallydownscaledtoone-twelfthdegree(5arcminute)usingANUSPLIN,ainterpolationmodelthatincorporatesfourdimensions(climaticvariable,latitude,longitude,andelevation)toproducegriddedsurfacesforbothmonthlyprecipitationandsurfaceairtemperature(Hutchinson2009).

TheCMIP3datawereobtainedandprocessedbyCoulsonandothers(2010)foruseinthe2010ResourcesPlanningAct(RPA)Assessment.Monthlyprecipitationand

Chapter 3. climate change Summary

28The Southern Forest Futures Project

temperaturedatafrom2000to2100werescaledtothecountylevelfortheconterminousUnitedStates.Allchaptersinthisassessmentusethecountylevelprecipitationandtemperaturedata.Allregionalandsubregionalaverageswerearea-weightedtoremovebiasthatwouldresultfromaveragingcountiesofdifferentareas.

Forthischapter,annualanddecadalaveragesweregeneratedfortheSouthandforitsfivesubregionsusingtheJMP8.0softwareapplication(SASInstituteInc.2010).Forahistoricalperspective,trendsinairtemperatureandprecipitationforthe13southerncapitalcitiesfrom1960to2007wereobtainedfromthePRISMClimateGroup(Gibsonandothers2002).MapsweregeneratedusingtheArcMapversion9.3.1softwareapplication(ESRI2010).Thedecadesselectedforthischapterwere2010,2020,2040,2060,and2090.Tocalculatethedecadalaverages,thetenyearssurroundingeachperiodweresummed,inthecaseofprecipitation,andthenaveraged.Thedecadalaveragefor2010includeddatafromtheyears2005–14,2020includeddatafrom2015–24,2040includeddatafrom2035–44,etc.TheresultssectiondescribesaveragesandanomaliesforeachofthefourCornerstones.

ReSulTS

Regional Forecasts

Table3.1summarizesprecipitationandtemperatureaveragesforecastedfortheSouththrough2100,withhistoricaldataforcomparison.Figures3.1through3.4presentgraphicandmapdisplaysofprecipitationdata,andfigures3.5through3.8presentgraphicandmapdisplaysoftemperaturedata.

Characterizedbylowpopulationgrowthandhighenergy-use/economic-growth(MIROC3.2+A1B),CornerstoneAisforecastedtobedryandhot,withaverageannualprecipitationof912mmandaverageannualtemperatureof20.22°C.Annualprecipitationexpectedforanysoutherncountyrangesfrom103to4999mm,andtemperaturerangesfrom-12.01to50.24°C.Averagemaximummonthlytemperatureswouldexceedthesingle-daysouthernmaximumof48.89°C,whichwassetinOklahomain1994(Burt2007).

Alsocharacterizedbylowpopulationgrowthandhighenergy-use/economic-growth(CSIROMK3.5+A1B),CornerstoneBisforecastedtobewetandwarm,withaverageannualprecipitationof1167mmandaveragetemperatureof19.06°C.Annualprecipitationexpectedforsoutherncountiesrangesfrom93to3912mm,andtemperaturerangesfrom-11.21to44.24°C.

Characterizedbymoderatepopulation/incomegrowthandenergyuse(CSIROMK2+B2),CornerstoneCisforecastedtobemoderateandwarm,withaverageannualprecipitationof1083mmandaverageannualtemperatureof19.45°C.Annualprecipitationexpectedforanysoutherncountyrangesfrom35to2641mm.Thatprecipitationminimumwouldbreakthe1956regionallowof42mminTexas(Burt2007).Temperatureisexpectedtorangefrom-19.73to45.39°C.

Alsocharacterizedbymoderatepopulation/incomegrowthandenergyuse(HadCM3+B2),CornerstoneDisalsoforecastedtobemoderateandwarm,withaverageannualprecipitationof1106mm(higherthanCornerstoneC)andaverageannualtemperatureof19.27°C(lowerthanCornerstoneC).Annualprecipitationexpectedfor

Table 3.1—Summary statistics for predicted (2010–2100) and historical (2001–09) annual precipitation and temperature forecasts for the Southern United States by four Cornerstone Futures A through D

cornerstonea

Precipitation (mm) Temperature (°c)

minimum maximum AverageStandard deviation minimum maximum Average

Standard deviation

A 733 1675 912 198 17.29 21.35 20.22 1.05B 627 1517 1167 138 17.98 23.93 19.06 1.33C 803 1369 1083 126 17.07 21.74 19.45 1.08D 724 1383 1106 121 16.76 22.36 19.27 1.10Average all Cornerstones NA NA 1066 NA NA NA 19.57 NA

Historical (2001 to 2009) 864 1552 1136 NA 16.97 19.45 17.87 NA

NA = not applicable.aEach Cornerstone represents a general circulation model paired with one of two emission scenarios (A1B represents low-population/high-economic growth, high energy use; B2 represents moderate growth and use): A is MIROC3.2+A1B, B is CSIROMK3.5+A1B, C is CSIROMK2+B2, and D is HadCM3+B2.Source: Intergovernmental Panel on Climate Change 2007b.

35chAPTeR 3. Climate Change Summary

anysoutherncountyrangesfrom102to2708mm,andtemperaturerangesfrom-18.68to48.01°C.

Subregional Forecasts

IntheSouthernUnitedStates,forecastedprecipitation(table3.2)andtemperatureaverages(table3.3)arenotexpectedtobeuniform,withsignificantvariationsacrossthefivesubregionsandbetweenseasons(table3.4).Figures3.9and3.10presentgraphicandmapdisplaysofprecipitationandtemperaturedata.

CornerstoneA’shighenergy-use/economic-growth(MIROC3.2+A1B)ispredictedtoresultintheleastdecadalprecipitationby2060,withanoverallaverageof810mmforallfivesouthernsubregionsandalowof525mmintheMid-South.Thistrendisexpectedtoabateonlyslightlyby2090toanaverageof858mmforallsubregionsand535mmfortheMid-South—stillmuchdrierthanthehistoricaloverallaverageof1136mm.

Althoughalsobasedonhighenergy-use/economic-growth,CornerstoneB(CSIROMK3.5+A1B)predictsmoredecadalprecipitationthantheotherCornerstonesby2060,withanoverallaverageof1156mm.Thistrendcontinuesinto2090,withanoverallaveragepredictedtobe1223mm.CornerstoneBalsopredictscoolerdecadaltemperaturesthantheotherCornerstonesby2060—withanoverallaverageof19.39°C—foreverysubregionexcepttheMid-South.Thistrendcontinuesinto2090,withCornerstoneB’soverallaverageof20.14°C,lowerthanalltheothersforallsubregions.

CornerstoneApredictswarmerdecadaltemperaturesthantheotherCornerstonesby2060,withanoverallaverageof20.83°Cforallfivesouthernsubregions.Thistrendcontinuesinto2090,withCornerstoneA’soverallaverageof21.84°Cleadingalltheothersforallsubregions.

Comparingthesepredictionswithhistoricaltrendsinairtemperatureandprecipitationforthe13southerncapitalcitiesfrom1960to2007showsastatisticallysignificantincrease(totalof0.705°C,averageof0.016°C)inairtemperaturebutnosignificantchangeinprecipitation(fig.3.11).Thesefindingsareconsistentwithatrendofsignificantincreasesintemperaturefrom1970to2008reportedbyKarlandothers(2009)(table3.4),butnotaftertheirdatafrom1901to1969wereincluded.

DiScuSSioN AND coNcluSioNS

GCMsprovidesomeindicationofhowclimatewillchangeacrosstheSouthincomingdecades.Eachhasbeen

independentlydeveloped,oftenforaspecificregion,andfrequentlycalibratedtorecreatehistoricalclimateontheassumptionthatsuccessfulmodelingofthepastincreasesthelikelihoodofaccuratelyforecastingthefuture.However,thesamecalibrationthatallowsanaccuraterecreationofhistoricalclimateforoneregioncanresultinover-orunder-predictingclimatechangeforothers.

Anexampleofpossibleover-predictingisCornerstoneA(MIROC3.2+A1B),whichassumeshighenergy-useandeconomic-growthandpredictsthewarmestconditions,withmonthlyaveragessometimesexceedingsingle-dayhistoricalhighs(fig.3.12).Similarly,CornerstoneA’saverageprecipitationisabout20percentlower(fig.3.13).Forthesereasons,itisconsideredthemostsevereoftheCornerstonesintermsofextremeeventsaswellasannualaverages.TheotherGCMsusedinthisanalysisalsopredictmaximummonthlyairtemperaturesinexcessofhistoricallyobservedconditions,butbyasmallermargin.Inparticular,CornerstoneB(CSIROMK3.5+A1B)predictsincreasesinaverageannualprecipitationcomparedtohistoricalaverages.

Anothercaveatisthataveragedorsummedmonthlyvaluesarelessabletoexpressclimatevariability(especiallyextremes)thandailyvalues.Monthlyaverageairtemperaturesareexpectedtobemuchlowerthansomeoftheindividualdailyhighs,andhigherthansomeoftheindividualdailylows.Forexample,ifamaximummonthlyairtemperatureispredictedtobe40°C,thenindividualdailyairtemperaturesarelikelytoexceed45oreven50°C.

Likewise,monthlyaverageprecipitationdoesnotfullyrepresentthenumberormagnitudeofindividualevents.AlthoughCornerstoneApredictsareductioninaverageprecipitation,manyofitsmonthlymaximumsexceedhistoricalhighs.Similarly,variationsamongmonthsmaynotbecapturedbymonthlyaveragesorannualsummaries.Forexample,1000mmofprecipitationduringa5-monthperiodinwinterandspringwouldproduceaverydifferentimpactthanifevenlydistributedthroughouttheyearorconcentratedduringgrowing-seasonmonths.Andformonthlylevelpredictions,a100-mmaveragewouldmaskthewaterqualityandfloodingimpactsthatwouldresultifprecipitationwereconcentratedinoneortwomajorevents.

TheGCMsalsohavelimitedspatialresolution.Theirone-twelfthdegreebyone-twelfthdegreeresolutionisasignificantimprovementonoldermodelforecasts,butstillcoarseforpredictingprecipitation,whichcanbehighlyvariablewithadjacentareasreceivingdrasticallydifferentprecipitationamountsfromasingleevent.Thisvariationisalsoimportantforlocalizedfloodforecastingandinestimatingwaterquality.

36The Southern Forest Futures Project

Table 3.2—Predicted average precipitation for subregions of the Southern United States as forecasted by four Cornerstone Futures A through D

Date Subregioncornerstonea prediction of average precipitation (mm)

A B c D

2010

Appalachian-Cumberland 1223 1419 1303 1390Coastal Plain 1216 1375 1268 1328Mid-South 721 812 663 784Mississippi Alluvial Valley 1351 1550 1358 1472Piedmont 1263 1484 1285 1379

2020

Appalachian-Cumberland 1257 1376 1371 1307Coastal Plain 1210 1313 1289 1257Mid-South 677 735 710 659Mississippi Alluvial Valley 1427 1397 1462 1348Piedmont 1285 1259 1326 1272

2040

Appalachian-Cumberland 1139 1448 1336 1298Coastal Plain 1174 1295 1307 1309Mid-South 579 837 713 725Mississippi Alluvial Valley 1261 1524 1392 1321Piedmont 1202 1273 1328 1331

2060

Appalachian-Cumberland 940 1444 1338 1362Coastal Plain 1037 1370 1309 1370Mid-South 525 729 650 717Mississippi Alluvial Valley 1024 1455 1371 1346Piedmont 1065 1345 1324 1371

2090

Appalachian-Cumberland 999 1434 1271 1417Coastal Plain 1109 1358 1195 1396Mid-South 536 884 666 743Mississippi Alluvial Valley 1110 1582 1303 1456Piedmont 1164 1395 1231 1388

aEach Cornerstone represents a general circulation model paired with one of two emission scenarios (A1B represents low-population/high-economic growth, high energy use; B2 represents moderate growth and use): A is MIROC3.2+A1B, B is CSIROMK3.5+A1B, C is CSIROMK2+B2, and D is HadCM3+B2.Source: Intergovernmental Panel on Climate Change 2007b.

37chAPTeR 3. Climate Change Summary

Table 3.3—Predicted average temperature (°C) for subregions of the Southern United States as forecasted by four Cornerstone Futures A through D

Date Subregioncornerstonea prediction of average temperature (°c) A B c D

2010

Appalachian-Cumberland 14.02 13.18 14.31 14.01Coastal Plain 19.36 18.89 19.49 19.45Mid-South 18.60 18.02 18.48 18.59Mississippi Alluvial Valley 19.01 18.54 19.36 19.15Piedmont 16.16 15.41 16.34 16.24

2020

Appalachian-Cumberland 14.57 13.99 14.67 13.91Coastal Plain 19.91 19.24 19.84 19.30Mid-South 19.15 18.40 19.01 19.01Mississippi Alluvial Valley 19.67 18.95 19.63 19.16Piedmont 16.73 16.02 16.72 16.05

2040

Appalachian-Cumberland 15.55 14.68 15.46 14.17Coastal Plain 20.61 19.98 20.27 19.80Mid-South 19.93 18.91 19.44 19.36Mississippi Alluvial Valley 20.38 19.63 20.04 19.75Piedmont 17.59 16.77 17.39 16.41

2060

Appalachian-Cumberland 16.87 15.03 15.91 15.16Coastal Plain 21.85 20.44 20.80 20.49Mid-South 21.34 20.11 19.97 19.97Mississippi Alluvial Valley 21.92 20.27 20.68 20.39Piedmont 18.79 17.05 17.84 17.26

2090

Appalachian-Cumberland 17.73 15.78 17.29 16.32Coastal Plain 22.78 21.30 21.96 21.50Mid-South 22.53 20.74 21.01 20.90Mississippi Alluvial Valley 22.73 20.94 21.87 21.34Piedmont 19.74 17.89 19.12 18.46

aEach Cornerstone represents a general circulation model paired with one of two emission scenarios (A1B represents low-population/high-economic growth, high energy use; B2 represents moderate growth and use): A is MIROC3.2+A1B, B is CSIROMK3.5+A1B, C is CSIROMK2+B2, and D is HadCM3+B2.Source: Intergovernmental Panel on Climate Change 2007b.

Table 3.4—Average change in temperature and precipitation in the Southeastern United States, as recreated from Karl and others (2009)

Temperature change (°F) Precipitation change (percent)1901-2008 1970-2008 1901-2008 1970-2008

Annual 0.3 1.6 Annual 6.0 -7.7Winter 0.2 2.7 Winter 1.2 -9.6Spring 0.4 1.2 Spring 1.7 -29.2Summer 0.4 1.6 Summer -4.0 3.6Autumn 0.2 1.1 Autumn 27.4 0.1

43chAPTeR 3. Climate Change Summary

However,gapsstillexist,bothinknowledgeanditsimplementation.Forexample,theGCMsfromthemostrecentassessmentincorporatechangesinalbedofrompolaricecapmelting(IPCC2007b),animprovementoverpreviousassessments(Winton2008)thatcanoffermoreaccuratesimulationsbutonlyifthisimportantfeedbackisincorporatedintonewmodelruns.Additionally,thepositivefeedbackbetweenpermafrostmeltingandsubsequentreleaseofcarbondioxideandmethaneaddsimportantgreenhousegasestotheatmospherethatmustbeincludedintheglobalwarmingpredictions(Walterandothers2006).

Justasweatherforecastscommonlypredictfrom7to10daysintothefuturewithdecreasingaccuracyovertime,climateforecastsbasedonexistinganddevelopingglobaloceanandatmosphericcirculationpatternscurrentlypredict6to12monthsintothefuture.Althoughadditionalimprovementintheaccuracyandforecastlengthoftheseseasonalpredictionsarelikely,accuratelypredictingspecificweathereventsorpatternsthatmayoccuryearsordecadesinthefutureisunlikelyanytimesoon.Thescienceneededtopredicttheimpactsofdoublingatmosphericcarbondioxideonglobalairtemperatureandprecipitationisverydifferentfromthescienceneededtopredictmonthlyairtemperatureforaspecificcityonaspecificdate.Giventheselimitations,landmanagerswillneedtorelyontheclimateenvelopes(rangesofclimaticconditionsforspecificplacesandtimes)astheydevelopclimatechangeimpactassessmentsandcopingstrategies.

AckNoWleDGmeNTS

Weacknowledgethemodelinggroups,theProgramforClimateModelDiagnosisandIntercomparisonandtheWorldClimateResearchProgramme(WCRP)WorkingGrouponCoupledModelingfortheirrolesinmakingavailabletheWCRPCMIP3multi-modeldataset.SupportofthisdatasetisprovidedbytheOfficeofScience,U.S.DepartmentofEnergy.WealsoacknowledgeJohnBuckleyandErikaCohen,SouthernResearchStation,U.S.DepartmentofAgriculture,ForestService,fortheirassistancewithreviewingclimatesummaries.

liTeRATuRe ciTeD

Burt,C.C.2007.Extremeweather:aguideandrecordbook.NewYork: W.W.NortonandCo.303p.

Coulson,D.P.;Joyce,L.A.;Price,D.T.[andothers].2010.ClimatescenariosfortheconterminousUnitedStatesatthecountyspatialscaleusingSRESscenariosA1BandA2andPRISMclimatology.FortCollins,CO:U.S.DepartmentofAgricultureForestService,RockyMountainResearchStation.http://www.fs.fed.us/rm/data_archive/dataaccess/US_ClimateScenarios_county_A1B_A2_PRISM.shtml.[Dateaccessed:January10,2011].

EnvironmentalSystemsResearchInstitute(ESRI).2010.ArcGIS.Version9.3.1.Redlands,CA:EnvironmentalSystemsResearchInstitute.

Gibson,W.;Daly,C.;Kittel,T.[andothers].2002.Developmentofa103-yearhigh-resolutionclimatedatasetfortheconterminousUnitedStates.In:Proceedingsof13thAmericanMeteorologicalSocietyConferenceonAppliedClimatology.Portland,OR:AmericanMeteorologicalSociety:181–183.

Hutchinson,M.F.2009.ANUSPLINVersion4.3.http://fennerschool.anu.edu.au/publications/software/anusplin.php.[Dateaccessed:March17,2011].

IntergovernmentalPanelonClimateChange(IPCC).2007a.ContributionofworkinggroupsI,II,andIIItothefourthassessmentreportoftheIntergovernmentalPanelonClimateChange.Geneva,Switzerland:IntergovernmentalPanelonClimateChange.104p.http://www.ipcc.ch/pdf/assessment-report/ar4/syr/ar4_syr.pdf.[Dateaccessed:March16,2011].

IntergovernmentalPanelonClimateChange(IPCC).2007b.IPCCfourthassessmentreport:climatechange2007(AR4).http://www.ipcc.ch/publications_and_data/publications_and_data_reports.htm.[Dateaccessed:July7,2010].

IntergovernmentalPanelonClimateChange(IPCC).2009.WhatisaGCM?http://www.ipcc-data.org/ddc_gcm_guide.html.[Dateaccessed:August18,2010].

Karl,T.R.;Melillo,J.M.;Peterson,T.C.2009.GlobalchangeimpactsintheUnitedStates.Cambridge,UK:CambridgeUniversityPress.188p.http://www.globalchange.gov/publications/reports/scientific-assessments/us-impacts/full-report.[Dateaccessed:August24,2010].

Maurer,E.P.;Brekke,L.;Pruitt,T.;Duffy,P.B.2007.Fine-resolutionclimateprojectionsenhanceregionalclimatechangeimpactstudies.Eos,TransactionsAmericanGeophysicalUnion.88(47):504.

NationalAssessmentSynthesisTeam(NAST).2001.ClimatechangeimpactsontheUnitedStates:thepotentialconsequencesofclimatevariabilityandchange.FoundationReport.ReportfortheU.S.globalchangeresearchprogram.Cambridge,UK:CambridgeUniversityPress.154p.

Nakicenovic,N.;Swart,R.,eds.2000.Specialreportonemissionsscenarios:aspecialreportofworkinggroupIIIoftheIntergovernmentalPanelonClimateChange.Cambridge,UK:CambridgeUniversityPress.599p.http://www.grida.no/climate/ipcc/emission/index.htm.[Dateaccessed:August26,2010].

SASInstituteInc.©2010.JMP.SAS/STATuser’sguide.Version8.Cary,NC:SASInstituteInc.[pagesunknown].

Solomon,S.;Qin,D.;Manning,M.;Chen,Z.[andothers],eds.2007.ContributionofworkinggroupItothefourthassessmentreportoftheIntergovernmentalPanelonClimateChange.Cambridge,UK:CambridgeUniversityPress.996p.http://www.ipcc.ch/publications_and_data/ar4/wg1/en/contents.html.[Dateaccessed:March15,2011].

Tans,P.2011.Trendsinatmosphericcarbondioxide.NOAA/ESRL.http://www.esrl.noaa.gov/gmd/ccgg/trends.[Dateaccessed:March15,2011].

Walter,K.M.;Zimov,S.A.;Chanton,J.P.[andothers].2006.MethanebubblingfromSiberianthawlakesasapositivefeedbacktoclimatewarming.Nature.443(7):71–75.

Winton,M.2008.SeaicealbedofeedbackandnonlinearArcticclimatechange.In:DeWeaver,E.;Bitz,C.M.;Tremblay,B.,eds.ArcticSeaIceDecline:Observations,Projections,Mechanisms,andImplications.GeophysicalMonographService.Washington,DC:111–131.Vol.180.