美国岸基风电市场报告2023（英文版）-美国能源部VIP专享VIP免费

下载本文档

阅读 902
格式 pdf
大小 4.1 MB
约96页
2023-10-20
收藏
点赞(0)
海报
举报

Land-Based Wind Market Report: 2023 Edition

Disclaimer

This report is being disseminated by the U.S. Department of Energy (DOE). As such, this document was prepared

in compliance with Section 515 of the Treasury and General Government Appropriations Act for fiscal year

2001 (public law 106-554) and information quality guidelines issued by DOE. Though this report does not

constitute “influential” information, as that term is defined in DOE’s information quality guidelines or the Office

of Management and Budget’s Information Quality Bulletin for Peer Review, the study was reviewed both

internally and externally prior to publication. For purposes of review, the study benefited from the advice and

comments of 19 industry stakeholders, U.S. Government employees, and national laboratory staff.

NOTICE

This report was prepared as an account of work sponsored by an agency of the United States government. Neither

the United States government nor any agency thereof, nor any of their employees, makes any warranty, express

or implied, or assumes any legal liability or responsibility for the accuracy, completeness, or usefulness of any

information, apparatus, product, or process disclosed, or represents that its use would not infringe privately

owned rights. Reference herein to any specific commercial product, process, or service by trade name, trademark,

manufacturer, or otherwise does not necessarily constitute or imply its endorsement, recommendation, or

favoring by the United States government or any agency thereof. The views and opinions of authors expressed

herein do not necessarily state or reflect those of the United States government or any agency thereof.

Available electronically at SciTech Connect: http://www.osti.gov/scitech

Available for a processing fee to U.S. Department of Energy

and its contractors, in paper, from:

U.S. Department of Energy

Office of Scientific and Technical Information

P.O. Box 62

Oak Ridge, TN 37831-0062

OSTI: http://www.osti.gov

Phone: 865.576.8401

Fax: 865.576.5728

Email: reports@osti.gov

Available for sale to the public, in paper, from:

U.S. Department of Commerce

National Technical Information Service

5301 Shawnee Road

Alexandria, VA 22312

NTIS: http://www.ntis.gov

Phone: 800.553.6847 or 703.605.6000

Fax: 703.605.6900

Email: orders@ntis.gov

Land-Based Wind Market Report: 2023 Edition

iii

Preparation and Authorship

This report was prepared by Lawrence Berkeley National Laboratory for the Wind Energy Technologies

Office of the U.S. Department of Energy’s Office of Energy Efficiency and Renewable Energy.

Corresponding authors of the report are Ryan Wiser and Mark Bolinger, Lawrence Berkeley National

Laboratory. The full author list includes: Ryan Wiser, Mark Bolinger, Ben Hoen, Dev Millstein, Joe Rand,

Galen Barbose, Naïm Darghouth, Will Gorman, Seongeun Jeong, Eric O'Shaughnessy, and Ben Paulos.

/96

下载本文档

文本预览下载提示常见问题

Land-BasedWindMarketReport:2023EditionLand-BasedWindMarketReport:2023EditionDisclaimerThisreportisbeingdisseminatedbytheU.S.DepartmentofEnergy(DOE).Assuch,thisdocumentwaspreparedincompliancewithSection515oftheTreasuryandGeneralGovernmentAppropriationsActforfiscalyear2001(publiclaw106-554)andinformationqualityguidelinesissuedbyDOE.Thoughthisreportdoesnotconstitute“influential”information,asthattermisdefinedinDOE’sinformationqualityguidelinesortheOfficeofManagementandBudget’sInformationQualityBulletinforPeerReview,thestudywasreviewedbothinternallyandexternallypriortopublication.Forpurposesofreview,thestudybenefitedfromtheadviceandcommentsof19industrystakeholders,U.S.Governmentemployees,andnationallaboratorystaff.NOTICEThisreportwaspreparedasanaccountofworksponsoredbyanagencyoftheUnitedStatesgovernment.NeithertheUnitedStatesgovernmentnoranyagencythereof,noranyoftheiremployees,makesanywarranty,expressorimplied,orassumesanylegalliabilityorresponsibilityfortheaccuracy,completeness,orusefulnessofanyinformation,apparatus,product,orprocessdisclosed,orrepresentsthatitsusewouldnotinfringeprivatelyownedrights.Referencehereintoanyspecificcommercialproduct,process,orservicebytradename,trademark,manufacturer,orotherwisedoesnotnecessarilyconstituteorimplyitsendorsement,recommendation,orfavoringbytheUnitedStatesgovernmentoranyagencythereof.TheviewsandopinionsofauthorsexpressedhereindonotnecessarilystateorreflectthoseoftheUnitedStatesgovernmentoranyagencythereof.AvailableelectronicallyatSciTechConnect:http://www.osti.gov/scitechAvailableforaprocessingfeetoU.S.DepartmentofEnergyanditscontractors,inpaper,from:U.S.DepartmentofEnergyOfficeofScientificandTechnicalInformationP.O.Box62OakRidge,TN37831-0062OSTI:http://www.osti.govPhone:865.576.8401Fax:865.576.5728Email:reports@osti.govAvailableforsaletothepublic,inpaper,from:U.S.DepartmentofCommerceNationalTechnicalInformationService5301ShawneeRoadAlexandria,VA22312NTIS:http://www.ntis.govPhone:800.553.6847or703.605.6000Fax:703.605.6900Email:orders@ntis.goviiLand-BasedWindMarketReport:2023EditionPreparationandAuthorshipThisreportwaspreparedbyLawrenceBerkeleyNationalLaboratoryfortheWindEnergyTechnologiesOfficeoftheU.S.DepartmentofEnergy’sOfficeofEnergyEfficiencyandRenewableEnergy.CorrespondingauthorsofthereportareRyanWiserandMarkBolinger,LawrenceBerkeleyNationalLaboratory.Thefullauthorlistincludes:RyanWiser,MarkBolinger,BenHoen,DevMillstein,JoeRand,GalenBarbose,NaïmDarghouth,WillGorman,SeongeunJeong,EricO'Shaughnessy,andBenPaulos.iiiLand-BasedWindMarketReport:2023EditionAcknowledgmentsFortheirsupportofthisongoingreportseries,theauthorsthanktheentireU.S.DepartmentofEnergy(DOE)WindEnergyTechnologiesOfficeteam.Inparticular,weacknowledgeGageReberandPatrickGilman.Forreviewingelementsofthisreportorprovidingkeyinput,wealsothank:RichardBowers(U.S.EnergyInformationAdministration);CharlieSmith(EnergySystemsIntegrationGroup);FengZhao(GlobalWindEnergyCouncil);DixieDowning(U.S.InternationalTradeCommission);OwenRoberts(NationalRenewableEnergyLaboratory,NREL);AndrewDavid(Silverado);DavidMilborrow(consultant);JohnHensley(AmericanCleanPowerAssociation);MattoxHall(Vestas);EdgarDeMeo(consultant);MattMcCabe(ArcLight);JustinSabrsula,ElizabethChu,andAllisonHolly(Pattern);LawrenceWilley(consultant);GeoffreyKlise(SandiaNationalLaboratories);andPatrickGilman,GageReber,andLizHartman(DOE).Forprovidingdatathatunderlieaspectsofthereport,wethanktheU.S.EnergyInformationAdministration,BloombergNEF,WoodMackenzie,GlobalWindEnergyCouncil,andtheAmericanCleanPowerAssociation.ThanksalsotoDonnaHeimiller(NREL)forassistanceinmappingwindresourcequality;andtoPardeepToorandAlexsandraLemke(NREL),andLizHartmanandWendellGrinton,Jr.(DOE)forassistancewithlayout,formatting,production,andcommunications.LawrenceBerkeleyNationalLaboratory’scontributionstothisreportwerefundedbytheWindEnergyTechnologiesOffice,OfficeofEnergyEfficiencyandRenewableEnergyoftheDOEunderContractNo.DE-AC02-05CH11231.Theauthorsaresolelyresponsibleforanyomissionsorerrorscontainedherein.ivLand-BasedWindMarketReport:2023EditionListofAcronymsACPAmericanCleanPowerAssociationBPABonnevillePowerAdministrationCAISOCaliforniaIndependentSystemOperatorCODcommercialoperationdateCCAcommunitychoiceaggregatorCREZcompetitiverenewableenergyzonesDOEU.S.DepartmentofEnergyEIAU.S.EnergyInformationAdministrationERCOTElectricReliabilityCouncilofTexasFAAFederalAviationAdministrationFERCFederalEnergyRegulatoryCommissionGEGeneralElectricCorporationGWgigawattHTSHarmonizedTariffScheduleIOUinvestor-ownedutilityIPPindependentpowerproducerISOindependentsystemoperatorISO-NENewEnglandIndependentSystemOperatorITCinvestmenttaxcreditkVkilovoltkWkilowattkWhkilowatt-hourLCOElevelizedcostofenergym2squaremeterMISOMidcontinentIndependentSystemOperatorMWmegawattMWhmegawatt-hourNRELNationalRenewableEnergyLaboratoryNYISONewYorkIndependentSystemOperatorO&MoperationsandmaintenanceOEMoriginalequipmentmanufacturerPJMPJMInterconnectionPOUPublicly-ownedutilityPPApowerpurchaseagreementPTCproductiontaxcreditvLand-BasedWindMarketReport:2023EditionPVphotovoltaicsRECrenewableenergycertificateRPSrenewablesportfoliostandardRTOregionaltransmissionorganizationSGRESiemensGamesaRenewableEnergySPPSouthwestPowerPoolWwattWAPAWesternAreaPowerAdministrationWECCWesternElectricityCoordinatingCouncilviLand-BasedWindMarketReport:2023EditionExecutiveSummaryWindpoweradditionsintheUnitedStatestotaled8.5gigawatts(GW)in2022.1Windpowergrowthhashistoricallybeensupportedbytheindustry’sprimaryfederalincentive—theproductiontaxcredit(PTC)—aswellasmyriadstate-levelpolicies.Long-termimprovementsinthecostandperformanceofwindpowertechnologieshavealsobeenkeydriversforwindadditions.Nonetheless,2022wasarelativelyslowyearintermsofnewwindpowerdeployment—thelowestsince2018—dueinparttoongoingsupplychainpressures,higherinterestrates,andinterconnectionandsitingchallenges,butalsothereductioninthevalueofthePTCthatwasinplaceupuntilthepassageoftheInflationReductionAct(IRA)inAugust2022.PassageofIRApromisesnewmarketdynamicsforwindpowerdeploymentandsupplychaininvestmentsintheyearsahead.IRAcontainsalong-termextensionofthePTCatfullvalue(assumingthatnewwageandapprenticeshipstandardsaremet)alongwithopportunitiesforwindplantstoearntwo10percentbonuscreditsthataddtothePTCformeetingdomesticcontentrequirementsandforlocatingprojectsinenergycommunities.Amongmanyotherprovisions,IRAalsoincludesnewproduction-basedandinvestment-basedtaxcreditstosupportthebuild-outofdomesticcleanenergymanufacturing.ThoughitistooearlytoseethefullimpactsofIRAinhistoricaldata,IRAhasalreadyimpactedanalystforecastsforfuturewindpowercapacityadditionsandwindindustrysupply-chainannouncements.Keyfindingsfromthisyear’sLand-BasedWindMarketReport—whichprimarilyfocusesonland-based,utility-scalewind—include:InstallationTrends•TheU.S.added8.5GWofwindpowercapacityin2022,totaling$12billionofinvestment.DevelopmentwasconcentratedintheElectricReliabilityCouncilofTexas(ERCOT)andtheSouthwestPowerPool(SPP).2Cumulativewindcapacitygrewtomorethan144gigawatts(GW)bytheendof2022.Inaddition,1.7GWofexistingwindplantswerepartiallyrepoweredin2022(thefinal,repoweredcapacityoftheseplantsis1.8GW),mostlybyupgradingrotors(blades)andnacellecomponentslikegearboxesandgenerators.•WindpowerrepresentedthesecondlargestsourceofU.S.electric-powercapacityadditionsin2022,at22%,behindsolar’s49%.Windpowerconstituted22%ofallgenerationandstoragecapacityadditionsin2022.Overthelastdecade,windrepresented27%oftotalcapacityadditions,andalargerfractionofnewcapacityinSPP(85%),ERCOT(49%),theMidcontinentIndependentSystemOperator(MISO)(47%),andthenon-ISOWest(30%).•Globally,theUnitedStatesagainrankedsecondinannualwindcapacitybutremainedwellbehindthemarketleadersinwindenergypenetration.Globalwindadditionstotaledover77GWin2022,yieldingacumulative906GW.TheUnitedStatesremainedthesecond-leadingmarketintermsofannualandcumulativecapacity,behindChina.Manycountrieshaveachievedhighwindelectricityshares,withwindsupplying57%ofDenmark’stotalelectricitygenerationin2022andmorethan20%inatotalofeightcountries.IntheUnitedStates,windsuppliedabout10%oftotalgeneration.•Texasonceagaininstalledthemostwindcapacityofanystatein2022(4,028MW),followedbyOklahoma(1,607MW);twelvestatesexceeded20%windenergypenetration.Texasalsoremainedtheleaderonacumulativecapacitybasis,withmorethan40GW.Notably,thewindcapacityinstalledinIowasupplied62%ofallin-stateelectricitygenerationin2022,followedbySouthDakota(55%),1NotethatthisreportseekstoalignwithAmericanCleanPower(ACP)forannualwindcapacityadditionsandproject-levelspecifics,wherepossible.DifferencesinreportingexistbetweenACPandtheEnergyInformationAdministration.2ThenineregionsmostusedinthisreportaretheSouthwestPowerPool(SPP),ElectricReliabilityCouncilofTexas(ERCOT),MidcontinentIndependentSystemOperator(MISO),CaliforniaIndependentSystemOperator(CAISO),ISONewEngland(ISO-NE),PJMInterconnection(PJM),andNewYorkIndependentSystemOperator(NYISO),andthenon-ISOWestandSoutheast.viiLand-BasedWindMarketReport:2023EditionKansas(47%),Oklahoma(44%),NorthDakota(37%),NewMexico(35%),andNebraska(31%).Withinindependentsystemoperators(ISOs),windelectricityshares(expressedasapercentageofload)were37.9%inSPP,24.8%inERCOT,14.5%inMISO,8.7%inCaliforniaIndependentSystemOperator(CAISO),4.0%inPJMInterconnection(PJM),3.2%inISONewEngland(ISO-NE),and3.1%inNewYorkIndependentSystemOperator(NYISO).•Hybridwindplantsthatpairwindwithstorageandotherresourcessawlimitedgrowthin2022,withjustonenewprojectcompleted.Therewere41hybridwindpowerplantsinoperationattheendof2022,representing2.6GWofwindand0.8GWofco-locatedgenerationorstorageassets.Themostcommonwindhybridprojectcombineswindandstoragetechnology,where1.4GWofwindhasbeenpairedwith0.2GWofbatterystorage.Theaveragestoragedurationoftheseprojectsis0.6hours,suggestingafocusonancillaryservicesandlimitedcapacitytoshiftlargeamountsofenergyacrosstime.Whileonlyonenewwindhybrid—combiningwind,solarphotovoltaics(PV),andstorage—wascommissionedin2022,solarhybridscontinuetoexpandrapidlywith59newPV+storageprojectscomingonlinein2022.•Arecord-high300GWofwindpowercapacitynowexistsintransmissioninterconnectionqueues,butsolarandstoragearegrowingatamuchmorerapidpace.Attheendof2022,therewere300GWofwindcapacityseekingtransmissioninterconnection,including113GWofoffshorewindand24GWofhybridprojects(inthelattercase,mostlywindpairedwithstorage).NYISO,thenon-ISOWest,andPJMhadthegreatestquantityofwindintheirqueuesattheendof2022.In2022,90GWofwindcapacityenteredinterconnectionqueues,41%ofwhichwasforoffshorewindplants.Storageandsolarinterconnectionrequestshaveincreasedrapidlyinrecentyears,oftentimespairingsolarwithstorage.IndustryTrends•Justfourturbinemanufacturers,ledbyGE,suppliedalltheU.S.utility-scalewindpowercapacityinstalledin2022.In2022,GEcaptured58%ofthemarketforturbineinstallations,followedbyVestaswith24%,Nordexwith10%,andSiemens-GamesaRenewableEnergy(SGRE)with8%.3•Thedomesticwindindustrysupplychainbegan2022indecline,butpassageoftheInflationReductionActhascreatedrenewedoptimismaboutsupply-chainexpansion.Thenumberofwindturbinetowersandnacelles(whichsitontopofthetowerandhousethegearboxandgenerator)thatwecanmanufacturedomesticallyintheUnitedStateshasheldsteadyorincreasedoverthelastseveralyears.Attheendof2022,domesticcapacitywas15GWperyearfornacelleassemblyand11GWperyearfortowermanufacturing.Blademanufacturingcontinueditsdeclinein2022,withunder4GWperyearofcapabilitybytheendoftheyear.Morebroadly,manyturbinemanufacturerscontinuedtofacedecliningandevennegativeprofitmarginsin2022.Nonetheless,wind-relatedjobtotalsincreasedby4.5%in2022,to125,580full-timeworkers.Moreover,passageoftheInflationReductionActholdspromiseforaddressingrecentdomesticsupply-chainchallengesandfuelingexpansion:atleastelevennew,re-opened,orexpandedmanufacturingfacilitieshavebeenannouncedinrecentmonthstoservetheland-basedwindmarket,totalingmorethan3,000newjobs.•Domesticmanufacturingcontentisstrongforsomewindturbinecomponents,buttheU.S.windindustryremainsreliantonimports.TheUnitedStatesimportswindequipmentfrommanycountries,includingmostprominentlyin2022:Mexico,India,andSpain.Nonetheless,forwindprojectsinstalledin2022,over85%ofnacelleassemblyand70%–85%oftowermanufacturingoccurredintheUnitedStates;inthecaseoftowers,benefittingfromimporttariffs.Forblades,domesticcontentwasjust5–25%in2022,havingplummetedinrecentyears.HowthesetrendschangeafterpassageoftheInflationReductionActremainstobeseen,thoughsupply-chainannouncementsinrecentmonthssuggestaresurgenceindomesticmanufacturing.3Numericalvaluespresentedhereandelsewheremaynotaddto100%,duetorounding.viiiLand-BasedWindMarketReport:2023Edition•Independentpowerproducersownmostwindassetsbuiltin2022,extendinghistoricaltrends.Independentpowerproducers(IPPs)own84%ofthenewwindcapacityinstalledintheUnitedStatesin2022,withtheremainingassets(16%)ownedbyinvestor-ownedutilities.•Forthefirsttime,non-utilitybuyersenteredintomorecontractstopurchasewindthandidutilitiesin2022.Directretailpurchasersofwind—includingcorporateofftakers—buyelectricityfromatleast44%ofthenewwindcapacityinstalledin2022.This~44%shareexceeds,forthefirsttime,thatofelectricutilities,whoeitherown(16%)orbuyelectricityfrom(17%)windprojectsthat,intotal,represent33%ofthenewcapacityinstalledin2022.Merchant/quasi-merchantprojectsandpowermarketersmakeupatleastanother3%and6%,respectively,whiletheremainder(14%)ispresentlyundisclosed.TechnologyTrends•Turbinecapacity,rotordiameter,andhubheighthaveallincreasedsignificantlyoverthelongterm.Tooptimizeprojectcostandperformance,turbinescontinuetogrowinsize.Theaveragerated(nameplate)capacityofnewlyinstalledwindturbinesintheUnitedStatesin2022was3.2MW,up7%fromthepreviousyearand350%since1998−1999.Theaveragerotordiameterofnewlyinstalledturbineswas131.6meters,a3%increaseover2021and173%over1998−1999,whiletheaveragehubheightwas98.1meters,up4%from2021and73%since1998−1999.•Turbinesoriginallydesignedforlowerwindspeedsitesdominatethemarket,butthetrendtowardslowerspecificpowerhasreversedinrecentyears.Withgrowthinsweptrotorareaoutpacinggrowthinnameplatecapacity,therehasbeenadeclineintheaverage“specificpower”4(inW/m2),from393W/m2amongprojectsinstalledin1998–1999to233W/m2amongprojectsinstalledin2022—thoughspecificpowerhasmodestlyincreasedoverthelastthreeyears.Turbineswithlowspecificpowerwereoriginallydesignedforlowerwindspeedsitesbutarenowbeingusedatmanysitesasthemostattractivetechnology.•Windturbinesweredeployedinhigherwind-speedsitesin2022thaninrecentyears.Windturbinesinstalledin2022werelocatedinsiteswithanaverageestimatedlong-termwindspeedof8.3meterspersecondataheightof100metersabovetheground—thehighestsite-averagewindspeedsince2014.FederalAviationAdministration(FAA)andindustrydataonprojectsthatareeitherunderconstructionorindevelopmentsuggestthatthesiteslikelytobebuiltoutoverthenextfewyearswill,onaverage,haveloweraveragewindspeeds.Increasinghubheightswillhelptopartiallyoffsetthistrend,however,enablingturbinestoaccesshigherwindspeedsthanotherwisepossiblewithshortertowers.•Low-specific-powerturbinesaredeployedonawidespreadbasis;tallertowersareseeingincreaseduseinawidervarietyofsites.Lowspecificpowerturbinescontinuetobedeployedinallregions,andatbothlowerandhigherwindspeedsites.Thetallesttowers(i.e.,thoseabove100meters)arefoundingreaterrelativefrequencyintheupperMidwestandNortheasternregions.•Windprojectsplannedforthenearfuturearepoisedtocontinuethetrendofever-tallerturbines.Theaverage“tipheight”(fromgroundtobladetipextendeddirectlyoverhead)amongprojectsthatcameonlinein2022is164meters.FAAdatasuggestthatfutureprojectswilldeployeventallerturbines.Among“proposed”turbinesintheFAApermittingprocess,theaveragetipheightreaches195meters.•In2022,thirteenwindprojectswerepartiallyrepowered,mostofwhichnowfeaturesignificantlylargerrotorsandlowerspecificpowerratings.Partiallyrepoweredprojectsin2022totaled1.7GWpriortorepowering(1.8GWafter),aslightincreasefromthe1.6GWofprojectspartiallyrepoweredin2021.Ofthechangesmadetotheturbines,largerrotorsdominated,reducingspecificpowerfrom300to4Awindturbine’sspecificpoweristheratioofitsnameplatecapacityratingtoitsrotor-sweptarea.Allelseequal,adeclineinspecificpowershouldleadtoanincreaseincapacityfactor.ixLand-BasedWindMarketReport:2023Edition220W/m2.Theprimarymotivationsforpartialrepoweringhavebeentore-qualifyforthePTC,whileatthesametimeincreasingenergyproductionandextendingtheusefullifeoftheprojects.PerformanceTrends•Theaveragecapacityfactorin2022was36%onafleet-widebasisand37%amongwindplantsbuiltin2021.Theaverage2022capacityfactoramongprojectsbuiltfrom2013to2021was40%,comparedtoanaverageof31%amongallprojectsbuiltfrom2004to2012,and23%amongallprojectsbuiltfrom1998to2003.Thishaspushedthecumulativefleet-widecapacityfactorhigherovertime,to36%in2022.Theaverage2022capacityfactorforprojectsbuiltin2021was37%,somewhatlowerthanforprojectsbuiltfrom2014to2020.•Stateandregionalvariationsincapacityfactorsreflectthestrengthofthewindresource;capacityfactorsarehighestinthecentralpartofthecountry.Basedonprojectsbuiltfrom2017to2021,averagecapacityfactorsin2022werehighestincentralstatesandlowerclosertothecoasts.Notsurprisingly,therelativestateandregionalcapacityfactorsareroughlyconsistentwiththerelativequalityofthewindresourceineachregion.•Turbinedesignandsitecharacteristicsinfluenceperformance,withdecliningspecificpowerleadingtosizableincreasesincapacityfactoroverthelongterm.Thedeclineinspecificpoweroverthelasttwodecadeshasbeenamajorcontributortohighercapacityfactors,buthasbeenoffsetinpartbyatendencytowardbuildingprojectsatsiteswithlowerannualaveragewindspeeds.Asaresult,averagecapacityfactorshavebeenrelativelystableamongprojectsbuiltoverthelastnineyears,withsomeevidenceofmodestdeclinesamongpost-2018vintageprojectsasspecificpowerhasdriftedupwardsinthemostrecentseveralyearsandsitequalityhasdecreasedsomewhat.•Windpowercurtailmentin2022acrosssevenregionsaveraged5.3%,upfromalowof2.1%in2016.AcrossallISOs,windenergycurtailmentin2022stoodat5.3%—generallyrisingoverthelastsixyears.Thisaveragemasksvariationacrossregionsandprojects:SPP(9.2%),ERCOT(4.7%),MISO(4.4%),andNYISO(3.2%)experiencedthehighestratesofwindcurtailmentin2022,whiletheotherthreeISOswereeachatlessthan2%.•2022wasanabove-averagewindresourceyearacrossmostofthecountry.Thestrengthofthewindresourcevariesfromyeartoyear;moreover,thedegreeofinter-annualvariationdiffersfromsitetosite(and,hence,alsoregiontoregion).Thistemporalandspatialvariationimpactsprojectperformancefromyeartoyear.In2022,thenationalwindindexstoodat1.06,itshighestlevelsince2014,asmostregionsexperiencedanabove-averagewindyear(thenon-ISOWestexcepted).•Windprojectperformancedegradationalsoexplainswhyolderprojectsdidnotperformaswellin2022.Capacityfactordatasuggestperformancedeclinewithprojectage,thoughperhapsmostlyonceprojectsagebeyond10years.Theapparentdeclineincapacityfactorsasprojectsprogressintotheirseconddecadepartiallyexplainswhyolderprojects—e.g.,thosebuiltfrom1998to2003—didnotperformaswellasnewerprojectsin2022.CostTrends•Windturbinepricescontinuedtoincreasein2022,reachingroughly$1,000/kW.Windturbinepricesdeclinedby50%between2008and2020.However,recentsupplychainpressuresandelevatedcommoditypriceshaveledtoincreasedturbineprices.Dataindicaterecentaveragepricingintherangeof$900/kWto$1,200/kW5,alevelroughlysimilartothatlastseenin2017and2018andupfromarangeof$800-$1,000/kWfor2019–2021.•Surprisingly,averageinstalledprojectcostsamongoursmallsampleof2022projectsdidnotfollowturbinepriceshigher.Afterfouryearsofrelativelystablecostsof~$1600/kWfrom20185Allcostfigurespresentedinthereportaredenominatedinreal2022dollars.xLand-BasedWindMarketReport:2023Editionthrough2021,thesurprisingdropinthecapacity-weightedaverageinstalledcostin2022—to$1,370/kW—ispartlyattributabletotheoutsizedinfluenceofasinglelargeprojectinourrelativelysmall2022plantsampleandtotheconcentrationofwinddeploymentin2022inthelow-costregionsofSPPandERCOT.The2022capacity-weightedaveragemaychangeasmoredatabecomeavailableovertime.•Recentinstalledcostsdifferbyregion.Thelowest-costprojectsinrecentyearshavebeeninERCOT(averaging$1360/kW)andSPP($1470/kW),whileMISOprojectshaveaveraged$1730/kW.Again,samplesizein2022(and,toalesserextent,in2021)isabnormallylow,andtheseaveragesmaychangeasmoredatabecomeavailable.•Installedcosts(permegawatt)generallydeclinewithprojectsize;arelowestforprojectsover200MW.Installedcostsexhibiteconomiesofscale,withcostsdecliningasprojectcapacityincreases.•Operationsandmaintenancecostsvariedbyprojectageandcommercialoperationsdate.Despitelimiteddata,projectsinstalledoverthepast16yearshave,onaverage,incurredloweroperationsandmaintenance(O&M)coststhanolderprojects.ThedataalsosuggestthatO&Mcoststendtoincreaseasprojectsage,atleastfortheolderprojectsinthesample.PowerSalesPriceandLevelizedCostTrends•Windpowerpurchaseagreementpriceshavebeendriftinghighersinceabout2018,witharecentrangefrombelow$20/MWhtomorethan$40/MWh.ThecombinationofdecliningcapitalandoperatingcostsandimprovedperformancedrovewindPPApricestoall-timelowsthrough2018,thoughpriceshavesincestabilizedandthenincreased—inpartduetosupply-chainandotherinflationarypressures.Thoughoursamplesizeinthelastyearortwoisrelativelysmall,recentpricingappearstobearound$20/MWhintheCentralregionofthecountry,abithigherintheWest(rangingfrom$20/MWhto$40/MWh),andhigherstillintheEast(~$50/MWh).•LevelTenEnergy’sPPApriceindicesconfirmrisingPPApricesandregionalvariation.IncontrasttothePPAssummarizedabove,whichprincipallyinvolveutilitypurchasers,LevelTenEnergyprovidesanindexofPPAoffersmadetolarge,end-usecustomers.ThesedataalsoshowthatpriceshaverisenoverthelastcoupleofyearsandvarybyISO.Amongregionsreportingdata,CAISOfeaturesthehighestpricing(~$60/MWhinthethirdquarterof2022onceconvertedtolevelized2022dollarterms);thelowestpricesarefoundinSPPandERCOT(~$33/MWhin2022dollars).Inrealdollarterms,LevelTen’sreportedpricetrendssince2018aresimilartothereal-dollardenominatedPPAtrendsdescribedinthepriorsection.•Amongarelativelysmallsampleofprojectsbuiltin2022,the(unsubsidized)averagelevelizedcostofwindenergyhasfallentoaround$32/MWh.Trendsinthelevelizedcostofenergy(LCOE)followPPAtrends,atleastoverthelongterm.Wind’sLCOEdecreasedfrom1998to2005,rosethrough2009-2011,declinedthrough2018,buthasremainedsteadyoverthelastseveralyears.ThenationalaverageLCOEamongasmallsampleofprojectsbuiltin2022—excludingthePTC—was$32/MWh.Thisaverageisimpactedbytheconcentrationofprojectsinstalledin2022inthewindy,low-costregionsofERCOTandSPP.Asmoredatabecomeavailable,theaverageLCOEamong2022(and2021)windplantscouldberevised.•Levelizedcostsvarybyregion,withthelowestcostsinSPPandERCOT.ThelowestaverageLCOEsforprojectsbuiltin2021and2022—onlyconsideringregionswithatleasttwoplantsinthesample—arefoundinSPPandERCOT(both~$33/MWhonaverage),withPJMaveragingthehighestat$46/MWh.CostandValueComparisons•DespiterelativelylowPPAprices,windfacescompetitionfromsolarandgas.Theonce-widegapbetweenwindandsolarPPApriceshasnarrowed,assolarpriceshavefallenmorerapidlythanwindxiLand-BasedWindMarketReport:2023Editionpricesoverthelastdecade.Withthesupportoffederaltaxincentives,bothwindandsolarPPApricesareonparwithorbelowtheprojectedcostofburningnaturalgasingas-firedcombinedcycleunits.•Thegrid-systemmarketvalueofwindsurgedin2022acrossmanyregionsandwasoftenhigherthanrecentwindPPAprices.Followingthesharpdropinwholesaleelectricityprices(and,hence,windenergymarketvalue)in2009,averagewindPPApricestendedtoexceedthewholesalemarketvalueofwindthrough2012.ContinueddeclinesinwindPPApricesbroughtthosepricesbackinlinewiththemarketvalueofwindin2013,andwindhasgenerallyremainedcompetitiveinsubsequentyears.In2022,windenergyvalueremainedatelevatedlevelsafterhavingreboundedin2021fromthelowassociatedwiththepandemic.Thenationalaveragemarketvalueofwindin2022was$32/MWh.Withlowernaturalgaspricessofarin2023,wind’saveragemarketvaluemaydeclinethisyear.•Thegrid-systemmarketvalueofwindin2022variedstronglybyprojectlocation,fromanaverageof$18/MWhinSPPto$83/MWhinISO-NE.Regionally,windmarketvaluein2022waslowestinSPP(averageof$18/MWh)andhighestinISO-NEandCAISO($83/MWhand$76/MWh).ThemarketvalueacrossallwindprojectslocatedinISOsspanned$12/MWhto$77/MWhin2022(10th–90thpercentilerange).Withinaregion,transmissioncongestioncannoticeablyreducethegridvalueofwindplants.•Thegrid-systemmarketvalueofwindtendstodeclinewithwindpenetration,impactedbygenerationprofile,transmissioncongestion,andcurtailment.Theregionswiththehighestwindpenetrations(SPPat38%,ERCOTat25%,andMISOat14%)havegenerallyexperiencedthelargestreductioninwind’svaluerelativetoaveragewholesaleprices.In2022,wind’svaluewasroughly40%,50%,50%,and60%,lowerthanaveragewholesalepricesinNYISO,MISO,ERCOT,andSPP,respectively;butwasonlyroughly10%lowerinISO-NEand~20%lowerinCAISOandPJM.Thesevaluereductionswereprimarilycausedbyacombinationoftransmissioncongestionandhourlywindgenerationthatwasnegativelycorrelatedwithwholesaleprices.Curtailmenthadonlyaminimalimpact.•Thehealthandclimatebenefitsofwindarelargerthanitsgrid-systemvalue,andthecombinationofallthreefarexceedsthelevelizedcostofwind.Windreducesemissionsofcarbondioxide,nitrogenoxides,andsulfurdioxide,providingpublichealthandclimatebenefits.Nationallyandconsideringallwindplants,thesehealthandclimatebenefitscanbequantifiedinmonetaryterms,averaging$135perMWhofwindin2022(basedonupdatedmethodsanddamageassumptions—seethefullreportandAppendix).ThesebenefitswerelargestintheCentral($200/MWh),Midwest($133/MWh),Texas($111/MWh),andWestern($109/MWh)regions,andwerelowestinNewYork($58/MWh),NewEngland($83/MWh),andtheMid-Atlantic($89/MWh).Combined,thenationalaverageclimate,health,andgrid-systemvaluesumstofivetimestheaverageLCOEofplantsbuiltin2022.Specifically,climate,health,andgridvalueaveraged$99/MWh,$37/MWh,and$32/MWh,respectively,comparedtoanaverageLCOEof$32/MWh.FutureOutlook•Energyanalystsprojectgrowingwinddeployment,spurredbyincentivesintheInflationReductionAct.Expectedcapacityadditionsrangefrom7.1GWto12GWin2023.Expectedadditionsthenincreaserapidly,supportedbyexpandedincentivesintheInflationReductionActaswellasanticipatedgrowthinoffshorewind.By2027,expectedadditionsrangefrom18.4GWto22.7GW.TheinfluenceoftheIRA—mostimportantly,itslong-termextensionofthePTCalongwithopportunitiesforwindplantstoearnbonuscreditsifmeetingdomesticcontentrequirementsand/orlocatedinanenergycommunity—dominatesanalystforecasts.Forexample,theaveragedeploymentforecastfor2026is18GW,comparedto11GWoneyearago,pre-IRA.Butheadwindsremain:inflation,higherinterestrates,limitedtransmissioninfrastructure,interconnectioncostsandtimeframes,sitingandpermittingchallenges,andcompetitionfromsolarmaydampengrowth,asmayanycontinuingsupplychainpressures.xiiLand-BasedWindMarketReport:2023Edition•Longerterm,theprospectsforwindenergywillbeinfluencedbytheInflationReductionActandbythesector’sabilitytocontinuetoimproveitseconomicposition.TheprospectsforwindenergyinthelongertermwillbeinfluencedbytheimplementationoftheInflationReductionAct,whichnotonlyprovidesextensionsandexpansionsofdeployment-orientedtaxcreditsbutalsonewincentivesforthebuildoutofdomesticsupplychains.Thespeedwithwhichsupplychainconstraintsareaddressedwillimpactdeploymentvolumes.Changingmacroeconomicconditions,corporatedemandforcleanenergy,andstate-levelpolicieswillalsocontinuetoimpactwindgrowth,aswillthebuildoutoftransmissioninfrastructure,resolutionofsiting,permittingandinterconnectionconstraints,andthefutureuncertaincostofnaturalgas.xiiiLand-BasedWindMarketReport:2023EditionTableofContentsExecutiveSummary.........................................................................................................................................vii1Introduction..............................................................................................................................................12InstallationTrends..................................................................................................................................43IndustryTrends.....................................................................................................................................164TechnologyTrends...............................................................................................................................275PerformanceTrends............................................................................................................................366CostTrends...........................................................................................................................................437PowerSalesPriceandLevelizedCostTrends.................................................................................498CostandValueComparisons.............................................................................................................559FutureOutlook......................................................................................................................................66References.......................................................................................................................................................68Appendix:SourcesofDataPresentedinthisReport................................................................................71xivLand-BasedWindMarketReport:2023EditionListofFiguresFigure1.Regionalboundariesoverlaidonamapofaverageannualwindspeedat100meters.......2Figure2.AnnualandcumulativegrowthinU.S.windpowercapacity......................................................4Figure3.RelativecontributionofgenerationtypesandstoragetoU.S.annualcapacityadditions....5Figure4.Generationandstoragecapacityadditionsbyregionoverlasttenyears...............................6Figure5.Windelectricityshareinsubsetoftopglobalwindmarkets.....................................................7Figure6.LocationofwindpowerdevelopmentintheUnitedStates........................................................8Figure7.Wind(leftpanel)andcombinedwind&solar(rightpanel)generationasaproportionofloadbyindependentsystemoperatorregions................................................................................10Figure8.LocationandcapacityofhybridwindplantsintheUnitedStates.........................................11Figure9.DesigncharacteristicsofhybridpowerplantsoperatingintheUnitedStates,forasubsetofconfigurations..................................................................................................................................11Figure10.Generationcapacityininterconnectionqueuesfrom2014to2022,byresourcetype..12Figure11.Windpowercapacityininterconnectionqueuesatendof2022,byregion......................13Figure12.Generationcapacityininterconnectionqueues,includinghybridpowerplants...............14Figure13.Hybridwindpowerplantsininterconnectionqueuesattheendof2022..........................15Figure14.AnnualU.S.marketshareofwindturbinemanufacturersbyMW,2005–2022..............16Figure15.Locationofturbineandcomponentmanufacturingfacilities..............................................17Figure16.Domesticwindmanufacturingcapabilityvs.U.S.windpowercapacityinstallations......18Figure17.TurbineOEMglobalprofitability................................................................................................19Figure18.Importsofwind-relatedequipmentthatcanbetrackedwithtradecodes........................21Figure19.Summarymapoftrackedwind-specificimportsin2022:top-10countriesoforiginandstatesofentry.......................................................................................................................................22Figure20.OriginsofU.S.importsofselectedwindturbineequipmentin2022.................................23Figure21.Approximatedomesticcontentofmajorcomponentsin2022...........................................24Figure22.Cumulativeand2022windpowercapacitycategorizedbyownertype.............................25Figure23.Cumulativeand2022windpowercapacitycategorizedbypowerofftakearrangement26Figure24.Averageturbinenameplatecapacity,hubheight,androtordiameterforland-basedwindprojects..................................................................................................................................................27Figure25.Trendsinturbinenameplatecapacity,hubheight,androtordiameter.............................28Figure26.Trendsinwindturbinespecificpower......................................................................................29Figure27.Windresourcequalitybyyearofinstallationat100metersandatturbinehubheight.30xvLand-BasedWindMarketReport:2023EditionFigure28:Locationoflowspecificpowerturbineinstallations:allU.S.windplants..........................31Figure29:Locationoftalltowerturbineinstallations:allU.S.windplants..........................................32Figure30.TotalturbineheightsproposedinFAAapplications,bydevelopmentstatus....................33Figure31.TotalturbineheightsproposedinFAAapplications,bylocation.........................................33Figure32.Annualamountofpartiallyrepoweredwindpowercapacityandnumberofturbines....34Figure33.Changeinaveragephysicalspecificationsofallturbinesthatwerepartiallyrepoweredin2022.......................................................................................................................................................35Figure34.Calendaryear2022capacityfactorsbycommercialoperationdate.................................37Figure35.Averagewindcapacityfactorincalendaryear2022bystate.............................................38Figure36.2022capacityfactorsandvariousdriversbycommercialoperationdate........................38Figure37.Calendaryear2022capacityfactorsbywindresourcequalityandspecificpower:2014–2021projects.......................................................................................................................................39Figure38.WindcurtailmentandpenetrationratesbyISO.....................................................................40Figure39.Inter-annualvariabilityinthewindresourcebyregionandnationally................................41Figure40.Changesinproject-levelcapacityfactorsasprojectsage.....................................................42Figure41.Reportedwindturbinetransactionpricesovertime..............................................................43Figure42.Installedwindpowerprojectcostsovertime..........................................................................44Figure43.Installedcostof2021and2022windpowerprojectsbyregion........................................45Figure44.Installedwindpowerprojectcostsbyprojectsize:2021and2022projects....................46Figure45.AverageO&Mcostsforavailabledatayearsfrom2000to2022,bycommercialoperationdate......................................................................................................................................47Figure46.MedianannualO&Mcostsbyprojectageandcommercialoperationdate......................48Figure47.LevelizedwindPPApricesbyPPAexecutiondateandregion(fullsample)......................50Figure48.Generation-weightedaveragelevelizedwindPPApricesbyPPAexecutiondateandregion.....................................................................................................................................................51Figure49.LevelTenEnergywindPPApriceindexbyquarterofoffer....................................................52Figure50.Estimatedlevelizedcostofwindenergybycommercialoperationdate...........................53Figure51.Estimatedlevelizedcostofwindenergy,byregion................................................................53Figure52.LevelizedwindandsolarPPApricesandlevelizedgaspriceprojections..........................55Figure53.WindPPApricesandnaturalgasfuelcostprojectionsbycalendaryearovertime.........56Figure54.Regionalwholesalemarketvalueofwindandaveragelevelizedlong-termwindPPApricesovertime....................................................................................................................................58Figure55.Regionalwholesalemarketvalueofwindin2022,byregion.............................................59xviLand-BasedWindMarketReport:2023EditionFigure56.Project-levelwholesalemarketvalueofwindin2022..........................................................60Figure57.Trendsinwindvaluefactoraswindpenetrationsincrease..................................................61Figure58.Impactoftransmissioncongestion,outputprofile,andcurtailmentonwindenergymarketvaluein2022..........................................................................................................................62Figure59.Marginalhealthandclimatebenefitsfromallwindgenerationbyregionin2022.........63Figure60.Marginalhealth,climate,andgrid-valuebenefitsfromnewwindplantsversusLCOEin2022.......................................................................................................................................................64Figure61.Windpowercapacityadditions:historicalinstallationsandprojectedgrowth..................66ListofTablesTable1.InternationalRankingsofTotalWindPowerCapacity.................................................................7Table2.U.S.WindPowerRankings:TheTop20States..............................................................................9TableA1.HarmonizedTariffSchedule(HTS)CodesandCategoriesUsedinWindImportAnalysis.72xvii1IntroductionWindpoweradditionsintheUnitedStatestotaled8.5gigawatts(GW)ofcapacityin2022.Windpowergrowthhashistoricallybeensupportedbytheindustry’sprimaryfederalincentive—theproductiontaxcredit(PTC)—aswellasmyriadstate-levelpolicies.Long-termimprovementsinthecostandperformanceofwindpowertechnologieshavealsobeenkeydriversforwindadditions,yieldinglow-pricedwindenergyforutility,corporate,andotherpowerpurchasers.Nonetheless,2022wasarelativelyslowyearintermsofnewwindpowerdeployment—thelowestsince2018—dueinparttoongoingsupplychainpressures,increasedinterestrates,andinterconnectionandsitingchallenges,butalsothereductioninthevalueofthePTCthatwasinplaceupuntilthepassageoftheInflationReductionAct(IRA)inAugust2022.PassageofIRApromisesnewmarketdynamicsforwindpowerdeploymentandsupplychaininvestmentsintheyearsahead(U.S.DOE2023a).IRAcontainsalong-termextensionofthePTCatfullvalue(assumingthatnewwageandapprenticeshipstandardsaremet)alongwithopportunitiesforwindplantstoearntwo10percentbonuscreditsthataddtothePTCformeetingdomesticcontentrequirementsandforbeinglocatedinenergycommunities.6Amongmanyotherprovisions,IRAalsoincludesnewproduction-basedandinvestment-basedtaxcreditstosupportthebuild-outofdomesticcleanenergymanufacturing.ThoughitistooearlytoseethefullimpactsofIRAinhistoricaldata,IRAhasalreadyimpactedanalystforecastsforfuturewindpowercapacityadditionsandwindindustrysupply-chainannouncements.Thisannualreport—nowinitsseventeenthyear—providesanoverviewoftrendsintheU.S.windpowermarket,withaparticularfocusontheyear2022.•Thereportbegins(Chapter2)withanoverviewofinstallation-relatedtrends:U.S.windpowercapacitygrowth;howthatgrowthcomparestoothercountriesandgenerationsources;theamountandpercentageofwindenergyinindividualU.S.states;hybridprojectsthatcouplewindwithstorageandothersourcesofgeneration;andthequantityofproposedwindpowercapacityininterconnectionqueuesintheUnitedStates.•InChapter3,thereportcoversanarrayofwindindustrytrends:developmentsinturbinemanufacturermarketshare;manufacturingandsupply-chaindevelopments;windturbineandcomponentimportsintotheUnitedStates;projectfinancingdevelopments;andtrendsamongwindpowerprojectownersandpowerpurchasers.•Chapter4summarizeswindturbinetechnologytrends:turbinecapacity,hubheight,rotordiameter,andspecificpower,aswellaschangesinsite-averagewindspeedandrecentrepoweringactivity.•Chapter5discusseswindplantperformance.•Chapter6discussesthecostandpricingofU.S.windenergy.Indoingso,itdescribestrendsincapacityfactors,windturbineprices,installedprojectcosts,andoperationsandmaintenance(O&M)expenses.•Chapter7reportsonlevelizedcosts,calculatedbasedontheinputparametersfromearlierchapters.Thereportalsoreviewsthepricespaidforwindpowerthroughpowerpurchaseagreements(PPAs)andhowthosepricescomparetothevalueofwindgenerationinwholesaleenergymarkets,forecastsoffuturenaturalgasprices,andsalespricesforsolarpower.6Formoreonenergycommunities,see:https://energycommunities.gov/energy-community-tax-credit-bonus/.Foradditionaldetailsonthedomesticcontentbonusandothertaxprovisions,see:https://www.irs.gov/inflation-reduction-act-of-2022.1Land-BasedWindMarketReport:2023Edition•Chapter8assessesthelevelizedcostofwindenergyrelativetoitssocietalvalue,definedsomewhatnarrowlyheretoincludethegrid-systemvalueofwindalongwithitshealthandclimatebenefits.•Thereportconcludes(Chapter9)withapreviewofpossiblenear-termmarketdevelopmentsbasedonthefindingsofotheranalysts.Manyofthesetrendsvarybystateorregion,dependinginpartonthestrengthofthelocalwindresource.Tothatend,Figure1superimposestheboundariesofnineregions,sevenofwhichalignwithorganizedwholesalepowermarkets(i.e.,independentsystemoperators),7onamapofaverageannualU.S.windspeedat100metersabovetheground.Thesenineregionswillbereferencedonmanyoccasionsthroughoutthisreport.Sources:AWSTruepower,NationalRenewableEnergyLaboratory(NREL)Figure1.Regionalboundariesoverlaidonamapofaverageannualwindspeedat100metersThiseditionoftheannualreportupdatesdatapresentedinpreviouseditionswhilehighlightingrecenttrendsandnewdevelopments.Thereportconcentratesonlarger,utility-scalewindturbines,definedhereasindividualturbinesthatexceed100kWinsize.8TheU.S.windpowersectorismultifaceted,andincludessmaller,customer-sitedwindturbinesusedtopowerresidences,farms,andbusinesses.Furtherinformationondistributedwindpower,whichincludessmallerwindturbinesaswellastheuseoflargerturbinesindistributedapplications,isavailablethroughaseparateannualreportfundedbytheU.S.DepartmentofEnergy(DOE)—theDistributedWindMarketReport.InChapters2,3,and9—whereitissometimesdifficultto7Thesevenindependentsystemoperators(ISOs)includetheSouthwestPowerPool(SPP),ElectricReliabilityCouncilofTexas(ERCOT),MidcontinentIndependentSystemOperator(MISO),CaliforniaIndependentSystemOperator(CAISO),ISONewEngland(ISO-NE),PJMInterconnection(PJM),andNewYorkIndependentSystemOperator(NYISO).8This100-kWthresholdbetween“smaller”and“larger”windturbinesisappliedstartingwith2011projectstobettermatchtheAmericanCleanPowerAssociation’shistoricalmethodology,andisalsojustifiedbythefactthattheU.S.taxcodemakesasimilardistinction.Inyearspriorto2011,differentcut-offsareusedtobettermatchACP’sreportedcapacitynumbersandtoensurethatolderutility-scalewindpowerprojectsinCaliforniaarenotexcludedfromthesample.2Land-BasedWindMarketReport:2023Editionseparateoffshoreandland-basedwind—thisreportcoversland-basedandoffshorewind,incombination.Otherchaptersexclusivelyfocusonland-basedwind.AcompanionstudyfundedbyDOEthatfocusesexclusivelyonoffshorewindpowerisalsoavailable—theOffshoreWindMarketReport.MuchofthedataincludedinthisreportwerecompiledbyDOE’sLawrenceBerkeleyNationalLaboratory(BerkeleyLab)fromavarietyofsources,includingtheU.S.EnergyInformationAdministration(EIA),theFederalEnergyRegulatoryCommission(FERC),andtheAmericanCleanPowerAssociation(ACP—alongwithitspredecessor,theAmericanWindEnergyAssociation).TheAppendixprovidesasummaryofthemanydatasources.Insomecases,thedatashownrepresentonlyasampleofactualwindpowerprojectsinstalledintheUnitedStates;furthermore,thedatavaryinquality.Emphasisshouldthereforebeplacedonoveralltrends,ratherthanonindividualdatapoints.Finally,eachsectionofthisreportprimarilyfocusesonhistoricalandrecentdata.Withsomelimitedexceptions—includingthelastsectionofthereport—thereportdoesnotseektoforecastwindenergytrends.3Land-BasedWindMarketReport:2023Edition2InstallationTrendsTheU.S.added8.5GWofwindpowercapacityin2022,totaling$12billionofinvestmentU.S.windcapacityadditionstotaled8.5GWin2022,bringingcumulativewindcapacitytomorethan144GWattheendoftheyear(Figure2).9Thisgrowthrepresentednearly$12billionofinvestmentinnewwindpowerplantsin2022,foracumulativeinvestmentofmorethan$300billionsincethebeginningofthe1980s.10,11Nearly77%ofthenewwindcapacityinstalledin2022islocatedinERCOT(39%)andSPP(37%),withtheremaindermostlyinMISOandthenon-ISOWest(eachwith11%).Inadditiontothenewlyinstalledcapacityreportedabove,1.7GWofexistingwindplantswere“partiallyrepowered”in2022(thefinal,repoweredcapacityoftheseplantsis1.8GW).12Partialrepowering,inwhichmajorcomponentsofturbinesarereplaced(mostoftenresultinginincreasedrotordiametersandupgradestomajornacellecomponents),providesaccesstofavorabletaxincentives,increasesenergyproductionwithmore-advancedturbinetechnology,andextendsprojectlife.SeeChapter4formoredetailsonpartialrepowering.AnnualRegionalCapacity(GW)CumulativeTotalCapacity(GW)16020Noncontiguous120CumulativeTotalSoutheast(non-ISO)80ISO-NE4015NYISOCAISOPJMWest(non-ISO)10MISOSPPERCOT5001998200020022004200620082010201220142016201820202022Source:ACPFigure2.AnnualandcumulativegrowthinU.S.windpowercapacityThesefiguresdepictarelativelyslowyearintermsofnewwindpowerdeploymentin2022—asteepdeclinefromthehighin2020andthelowestsince2018.Thisdownwardtrendwasdriveninpartbythestep-downin9The144.2GWofcapacityincludesthe30MWBlockIslandoffshorewindplantandthe12MWCoastalVirginiaOffshoreWindpilotproject.Whenreportingannualcapacityadditions,thisreportfocusesongrossadditions,anddoesnotconsiderpartialrepowering.Thenetincreaseincapacityeachyearcanbesomewhatlower,reflectingturbinedecommissioning,orhigher,reflectingpartialrepoweringthatincreasesturbinecapacity.Fullrepowering,ontheotherhand,isconsideredanewprojectandsoisincludedinannualadditions.Cumulativecapacity(‘Total’inFigure2)includesbothdecommissioningandrepowering.10Allcostandpricedataarereportedinreal2022dollars.11Theseinvestmentfiguresarebasedonanextrapolationoftheaverageproject-levelcapitalcostsreportedlaterinthisreportanddonotincludeinvestmentsinmanufacturingfacilities,researchanddevelopmentexpenditures,orO&Mcosts;nordotheyincludeinvestmentstopartiallyrepoweredplants.12AnychangeincapacityfrompartialrepoweringisincludedinthecumulativedatabutnottheannualdatareportedinFigure2.4Land-BasedWindMarketReport:2023EditionthefederalproductiontaxcreditpriortothepassageoftheIRA,andechoedsimilarboom/bustcyclesassociatedwithpreviousPTCexpirationdatesthatcanbeseeninFigure2in2002,2010,and2013.Theindustryalsocontendedwithcontinuedheadwindsin2022,relatedtosupplychainpressures,interconnectionbacklogs,limitedtransmissioninfrastructure,sitingandpermittingchallenges,andcompetitionwithsolar.PushingintheotherdirectionandsupportingdeploymentwasthecontinuedavailabilityofthePTC(evenifatareducedlevel),staterenewablesportfoliostandards(RPS),andcorporatedemandforrenewableenergy.Meanwhile,theabilityofpartiallyrepoweredwindprojectstoaccessthePTChasbeentheprimarymotivatorforthegrowthinpartialrepoweringinrecentyears.Long-termimprovementsinthecostandperformanceofwindpowertechnologieshavealsobeenkeydriversforwindadditions,yieldinglow-pricedwindenergyforutility,corporate,andotherpowerpurchasersevenassupplychainconstraintsandincreasedcommoditycostsandinterestrateshavepushedrecentcostshigher.WindpowerrepresentedthesecondlargestsourceofU.S.electric-powercapacityadditionsin2022,at22%,behindsolar’s49%Windpoweragaincontributedasizableshareoftotalgenerationandstoragecapacityadditions.In2022,itconstituted22%ofallU.S.generationandstoragecapacityadditions,secondonlytosolarpowerat49%(Figure3).13Naturalgasandothernon-renewablecapacityadditionswereroughlythesameastheyearprior,whichwastheirlowestlevelinmorethan20years.AnnualCapacityAdditions(GW)Othernon-RE50Coal4030GasOtherRE20Storage10Solar6%24%38%26%24%20%32%42%31%22%Wind02013201420152016201720182019202020212022Sources:Hitachi,ACP,EIA,BerkeleyLabFigure3.RelativecontributionofgenerationtypesandstoragetoU.S.annualcapacityadditionsOverthelastdecade,windpowerrepresented27%oftotalU.S.generationandstoragecapacityadditions,andanevenlargerfractionofnewcapacityinSPP(85%),ERCOT(49%),MISO(47%),andthenon-ISOWest(30%)(Figure4;seeFigure1forregionaldefinitions).Windpower’scontributiontocapacitygrowthoverthelastdecadeissmallerinPJM(9%),NYISO(7%),ISO-NE(7%),CAISO(4%),andtheSoutheast(1%).13Datapresentedherearebasedongrosscapacityadditions,notconsideringretirementsorpartialrepowering.Forsolar,bothutility-scaleanddistributedapplicationsareincluded.Dataincludeonlythe50U.S.states,notU.S.territories.5Land-BasedWindMarketReport:2023EditionPercentofCapacityAdditions:2013–2022Othernon-RE100%80%Coal60%Gas40%OtherREStorage20%Solar85%49%47%30%9%7%7%4%1%27%WindISO-NE0%U.S.TotalSPPERCOTMISOWestPJMNYISOCAISOSoutheast(non-ISO)(non-ISO)U.S.TotalalsoincludesAKandHI,inadditiontotheregionslistedSources:Hitachi,ACP,EIA,BerkeleyLabFigure4.GenerationandstoragecapacityadditionsbyregionoverlasttenyearsGlobally,theUnitedStatesagainrankedsecondinannualwindcapacitybutremainedwellbehindthemarketleadersinwindenergypenetrationGlobalwindadditionstotaledover77GWin2022(includingbothland-basedandoffshorewind).Withits8.5GWrepresenting11%ofnewglobalinstalledcapacityin2022,theUnitedStatescontinuedtomaintainitssecond-placepositionbehindChina(Table1).Cumulativeglobalwindcapacitytotaled906GWattheendoftheyear(GWEC2023),14withtheUnitedStatesaccountingfor16%—alsoadistantsecondtoChina.14YearlyandcumulativeinstalledwindpowercapacityintheUnitedStatesarefromthepresentreport,whileglobalwindpowercapacitycomesfromGWEC(2023)butareupdated,wherenecessary,withtheU.S.datapresentedhere.6Land-BasedWindMarketReport:2023EditionTable1.InternationalRankingsofTotalWindPowerCapacityAnnualCapacityCumulativeCapacity(2022,GW)(endof2022,GW)China37.6China365UnitedStates8.5UnitedStates144Brazil4.1Germany67Germany2.7India42Sweden2.4Spain30Finland2.4UnitedKingdom28France2.1Brazil26India1.8France21UnitedKingdom1.7Canada15Spain1.7Sweden15RestofWorld12.4RestofWorld153TOTAL77.5TOTAL906Sources:GWEC(2023);ACPforU.S.Manycountrieshaveachievedhigherwind-electricitymarketshares(i.e.,windgenerationasapercentageoftotalgeneration)thantheUnitedStates.Figure5presentsdataonasubsetofcountries.ThewindelectricitysharewashighestinDenmark,at57%,andwasover20%insevenothercountries.IntheUnitedStates,windsuppliedabout10%oftotalelectricitygenerationin2022.WindasPercentageofTotalGenerationin202260%50%40%30%20%10%0%DenmarkIrelandLithuaniaPortugalU.K.GermanyGreeceSpainSwedenNetherlandsFinlandCroatiaEURomaniaBelgiumBrazilAustraliaPolandTurkeyEstoniaAustriaNorwayUnitedStatesChinaFranceItalyMexicoCanadaIndiaSource:ACPFigure5.Windelectricityshareinsubsetoftopglobalwindmarkets7Land-BasedWindMarketReport:2023EditionTexasonceagaininstalledthemostwindcapacityofanystatein2022(4,028MW),followedbyOklahoma(1,607MW);twelvestatesexceeded20%windenergypenetrationNewutility-scalewindturbineswereinstalledin14statesin2022.Texasonceagaininstalledthemostnewcapacityofanystate,adding4,028MW.AsshowninFigure6andinTable2,otherleadingstates—intermsofnewcapacityaddedin2022—includedOklahoma(1,607MW),Nebraska(602MW),andIowa(484MW).Onacumulativebasis,Texasremainedtheclearleader,withmorethan40GWinstalledattheendof2022—morethanthreetimesasmuchasthenext-higheststate(Iowa).Infact,Texashasmorewindcapacitythanallbutfourcountries(Table1).StatesdistantlyfollowingTexasincumulativeinstalledcapacityincludeIowaandOklahoma(both>12GW),Kansas(>8GW),andIllinois(>7GW).Thirty-fivestates,plusPuertoRico,hadmorethan100MWofwindcapacityattheendof2022,with23oftheseabove1GW,19above2GW,and17above3GW.Sources:ACP,BerkeleyLabFigure6.LocationofwindpowerdevelopmentintheUnitedStatesSomestateshavereachedhighwindelectricityshares.TherighthalfofTable2liststhetop20statesbasedonactualwindelectricitygenerationin2022dividedbytotalin-stateelectricitygenerationandbyin-stateelectricitysalesin2022.Electrictransmissionnetworksenablemoststatestobothimportandexportpowerinrealtime,andstatesdosoinvaryingamounts.Denominatingin-statewindgenerationasbothaproportionofin-stategenerationandasaproportionofin-statesalesisrelevant,butbothshouldbeviewedwithsomecautiongivenvaryingamountsofimportsandexports.Asafractionofin-stategeneration,Iowaleadsthelist,with62%ofelectricitygeneratedinthestatecomingfromwind,followedbySouthDakota,Kansas,Oklahoma,andNorthDakota.Asafractionofin-statesales,8Land-BasedWindMarketReport:2023EditionIowaonceagainleads,withnearly82%oftheelectricitysoldinthestatebeingmetbywind,followedbySouthDakota(~77%),Kansas,NorthDakota,andWyoming(allthreeover60%),andthenOklahomaandNewMexico(bothover50%).Twelvestateshaveachievedwindpenetrationlevelsof20%orhigherwhenexpressedasapercentageofgeneration(thirteenexceed20%asapercentageofsales).Table2.U.S.WindPowerRankings:TheTop20StatesInstalledCapacity(MW)2022WindGenerationasaPercentageof:Annual(2022)Cumulative(endof2022)In-StateGenerationIn-StateSales40,151Texas4,028Texas12,783Iowa62.4%Iowa81.9%12,222Oklahoma1,607Iowa8,240SouthDakota54.8%SouthDakota76.9%7,129Nebraska602Oklahoma6,118Kansas47.0%Kansas69.9%5,194Iowa484Kansas4,749Oklahoma43.5%NorthDakota65.5%4,327Montana366Illinois4,302NorthDakota36.7%Wyoming60.4%4,055SouthDakota304California3,519NewMexico34.9%Oklahoma54.0%3,468Minnesota245Colorado3,407Nebraska31.0%NewMexico52.6%3,231NewMexico235Minnesota3,219Colorado28.0%Nebraska37.7%3,176Oregon210NewMexico2,435Minnesota23.5%Colorado29.2%2,192Colorado145NorthDakota1,487Maine22.8%Montana25.9%8,769Illinois120OregonWyoming21.8%Texas25.3%144,173Michigan72NebraskaTexas21.6%Maine23.3%California72IndianaVermont18.2%Minnesota21.5%Maine20WashingtonIdaho16.6%Oregon17.1%MichiganMontana14.8%Illinois16.9%SouthDakotaOregon14.3%Idaho11.1%WyomingIllinois12.1%Washington10.1%MissouriIndiana9.9%Indiana9.7%NewYorkMissouri9.4%Missouri9.3%MontanaMichigan7.8%Michigan9.1%RestofU.S.0RestofU.S.RestofU.S.1.7%RestofU.S.1.5%Total8,511TotalTotal10.1%Total11.2%Note:Basedon2022windandtotalgenerationandretailsalesbystatefromEIA’sElectricPowerMonthly(2023b).Sources:ACP,EIAGiventheabilitytotradepoweracrossstateboundaries,windelectricityshareswithinentiremulti-statemarketsoperatedbythemajorindependentsystemoperators(ISOs)arealsorelevant.In2022,wind-electricitymarketshares(expressedasapercentageofcustomerloadinclusiveofbehind-the-metersolargeneration)were37.9%inSPP,24.8%inERCOT,14.5%inMISO,8.7%inCAISO,4.0%inPJM,3.2%inISO-NE,and3.1%inNYISO(Figure7).Asalsoshowninthefigure,combinedsolarandwindsharesexceedstheselevels,especiallyinCAISO,ISO-NE,andERCOT.9Land-BasedWindMarketReport:2023EditionWindMarketShare(%)Wind+SolarMarketShare(%)40%40%SPP35%CAISOSPP35%30%ERCOT30%ERCOT25%25%20%MISO20%MISO15%15%10%CAISO10%ISO-NE5%PJM0%PJM5%NYISO20122020202220142016NYISOISO-NE0%2014201620182018202020222012Sources:SPP,ERCOT,MISO,CAISO,PJM,ISO-NE,NYISOFigure7.Wind(leftpanel)andcombinedwind&solar(rightpanel)generationasaproportionofloadbyindependentsystemoperatorregionsHybridwindplantsthatpairwindwithstorageandotherresourcessawlimitedgrowthin2022,withjustonenewprojectcompletedThoughonlyonenewwindhybridprojectwascommissionedin2022,therewere41hybridwindpowerplantsinoperationattheendof2022,representing2.6GWofwindand0.8GWofco-locatedassets(storage,PV,orfossil-fueledgenerators).Someoftheserepresentfullhybridswhere,forexample,windandstorageareco-locatedandthedesign,configuration,andoperationoftheconstituenttechnologiesarefullyintegrated.Inothercases,plantsareco-located,sharingapointofinterconnection,butaredesigned,configured,andoperatedmoreindependently(e.g.,hybridsthatpairwindandgasplants).Themostcommontypeofwindhybridprojectcombineswindandstoragetechnology,where1.4GWofwindhasbeenpairedwith0.2GWofbatterystorageacross14plants.However,nonewprojectscombiningjustwindandstoragewereinstalledin2022.OthercombinationsincludewindandPV;wind,PV,andstorage;windandgas;andmore(Figure8).TheWheatridgeprojectinOregon,theonlynew2022windhybrid,incorporateswind,PV,andstoragetechnologies.TheERCOTregionhoststhelargestamountofwindcapacityinhybridplants(0.86GW),followedbyPJM(0.77GW)andthenon-ISOWest(0.63GW).Windcapacitytendstobelargerforwind+storagehybridsthanforotherhybridconfigurations.10Land-BasedWindMarketReport:2023EditionSources:EIA-8602022EarlyRelease,BerkeleyLabFigure8.LocationandcapacityofhybridwindplantsintheUnitedStatesFigure9displaysdesigncharacteristicsforasubsetofthemore-commonhybridplantconfigurations,includingthosethatdonotincorporatewind.Wind+storagehybridshavea14%storage-to-generatorratiowithanaveragestoragedurationofjust0.6hours,suggestingafocusonprovidingancillaryservicesandonlylimitedcapacitytoshiftlargeamountsofenergyacrosstime.Fossil+storagehybridshavesimilarstorage-to-generatorratios(16%)butlongerbatterydurations(2.3hours).PV+storagehybridshavesignificantlyhigheraveragestorage-to-generatorratios(49%)andbatterydurations(3.1hours).#projectsTotalcapacity(MW)StorageratioDuration(hrs)02,0004,0006,0008,00010,00012,000WindPVFossilStoragePV+Storage2138,193.94,018.449%3.1WindWind+Storage141,425.3198.1PV14%0.668.8FossilWind+PV+Storage5525.776.0Storage11%2.0Fossil+Storage266,575.41,042.916%2.3Wind+PV8590.3267.50.0n/an/aNotes:Notincludedinthefigurearemanyotherhybridprojectswithotherconfigurations.Storageratiodefinedastotalstoragecapacitydividedbytotalgeneratorcapacityforagivenprojecttype.Sources:EIA-8602022EarlyRelease,BerkeleyLabFigure9.DesigncharacteristicsofhybridpowerplantsoperatingintheUnitedStates,forasubsetofconfigurations11Land-BasedWindMarketReport:2023EditionThetrendtoco-locatewindwithotherassetshasprogressedataslowpacesince2006,withonlyonenewwindhybridcommencingoperationin2022.Incontrast,commercialinterestinsolarhybridshasexpandedrapidly,with59newPV+storageprojectscomingonlinein2022.Arecord-high300GWofwindpowercapacitynowexistsintransmissioninterconnectionqueues,butsolarandstoragearegrowingatamuchmorerapidpaceOnetestamenttotheamountofdeveloperandpurchaserinterestinwindenergyistheamountofwindpowercapacityworkingitswaythroughthemajortransmissioninterconnectionqueuesacrossthecountry.Figure10providesthisinformationoverthelastnineyearsforwindpowerandotherresourcesaggregatedacrossmorethan40differentinterconnectionqueuesadministeredbyISOsandutilities.15Thesedatashouldbeinterpretedwithcaution:placingaprojectintheinterconnectionqueueisanecessarystepinprojectdevelopment,butbeinginthequeuedoesnotguaranteethataprojectwillbebuilt.Recentanalysisfoundanoverallaveragecompletionrateof21%forprojectsofalltypesproposedfrom2000to2017(Randetal.2023).Someprojectsareexploratoryinnature,andduplicateprojectsalsocomplicateinterpretation.HatchedareasarehybridprojectsTealareasareoffshoreNotes:Hybridstoragecapacityisestimatedusingstorage:generatorratiosfromprojectsthatprovideseparatecapacitydata;storagecapacityinhybridswasnotestimatedforyearspriorto2020;offshorewindwasnotseparatelyidentifiedpriorto2020.Source:BerkeleyLabreviewofinterconnectionqueuesFigure10.Generationcapacityininterconnectionqueuesfrom2014to2022,byresourcetypeEvenwiththisimportantcaveat,theamountofwindcapacityinthenation’sinterconnectionqueuesstillprovidesanindicationofdeveloperinterest.Attheendof2022,therewere300GWofwindcapacityinthequeuesreviewedforthisreport—amarkedincreasefromthe247GWinthequeuesthepreviousyearandsupportedbycontinuedgrowthinoffshorewindinthequeues.In2022,90GWofnewwindcapacityenteredthequeues,11GWofwhichwereinhybridconfigurationsand37GWofwhichwereforoffshorewind.Solaradditionstointerconnectionqueuesfaroutpacedwindin2022,with351GWadded.Storageadditionstothe15ThequeuessurveyedincludePJM,MISO,NYISO,ISO-NE,CAISO,ERCOT,SPP,WesternAreaPowerAdministration(WAPA),BonnevillePowerAdministration(BPA),TennesseeValleyAuthority(TVA),andalargenumberofotherindividualutilities.Toprovideasenseofsamplesizeandcoverage,theISOs,RTOs,andutilitieswhosequeuesareincludedherehaveanaggregatednon-coincident(balancingauthority)peakdemandofover85%oftheU.S.total.Thefiguresinthissectiononlyincludeprojectsthatwereactiveinthequeuesatthetimesspecifiedbutthathadnotyetbeenbuilt;suspendedprojectsarenotincluded.12Land-BasedWindMarketReport:2023Editionqueueshaveincreasedmuchmorerapidlythanwindinrecentyearsaswell,bothforstandaloneplantsandhybridizedwithsolarorwind.Overall,windrepresented15%ofallactivecapacityinthequeuesattheendof2022,comparedto46%forsolar,33%forstorage,andjust4%fornaturalgas.Thecombinedcapacityofwindandsolarnowactiveinthequeues(1,250GW)approximatelyequalsthetotalinstalledU.S.electricgeneratingcapacityin2022.Concerningly,thesubsetofproposedplantsthatworktheirwaythroughtheinterconnectionprocessandcomeonlinearetakinglongertodoso:themedianwindprojectreachingcommercialoperationin2022submittedaninterconnectionrequestnearly6yearsprior(Randetal.2023).16ThetotalwindcapacityintheinterconnectionqueuesisspreadacrosstheUnitedStates,asshowninFigure11(leftimage),withthelargestamountsinNYISO(22%),theWest(non-ISO)(21%),andPJM(16%).SmalleramountsarefoundinSPP(12%),MISO(11%),CAISO(6%),ERCOT(6%),ISO-NE(5%),andtheSoutheast(non-ISO)(1%).Nearlyhalf(48%)ofactivewindcapacityinthequeueshasrequestedtocomeonlinebytheendof2025,and15%ofwindcapacityhasafullyexecutedinterconnectionagreement.Focusingjustonwindpoweradditionstothequeuesin2022(Figure11,rightimage),NYISO,theWest(non-ISO),andMISOexperiencedespeciallylargeannualadditions(>17GWeach),withNYISO’sadditionsbeingalmostentirelyforoffshorewind.Acrossallqueues,38%(113GW)ofallwindcapacityinthequeuesattheendof2022wasoffshore,and41%(37GW)ofthewindaddedtoqueuesin2022wasoffshore.NewoffshorewindcapacitywasaddedontheEastCoastin2022(NYISO,PJM,ISO-NE),butnottheWestCoastduetoCAISOdelayingtheirnextinterconnectionapplicationwindowuntil2023.Note:Offshoreareasreflecttheamountofoffshorewindintheinterconnectionqueuesofeachregion.Source:BerkeleyLabreviewofinterconnectionqueuesFigure11.Windpowercapacityininterconnectionqueuesatendof2022,byregionAsshowninFigure12,48%ofthesolarcapacityininterconnectionqueuesattheendof2022hasbeenproposedashybridplants,whereasonly8%ofthewindcapacityispairedwithstorageoranothergenerationresource.Inpartthisisduetopolicydesign—untilthepassageoftheInflationReductionAct,theinvestmenttaxcreditforsolarcouldbeusedforpairedstorage,whereastheproductiontaxcreditregularlyusedbywindplantshadnosuchstorageallowance.Ofthe24GWofproposedwindcapacityinhybridconfigurations,themajority(19GW)ispairedwithstorage,withtherestprimarilypairedwithsolar(1GW)orbothsolarandstorage(4GW).16TheU.S.DepartmentofEnergyisengagingwithinterconnectionstakeholdersviatheInterconnectionInnovatione-Xchange.Formore,see:https://www.energy.gov/eere/i2x/interconnection-innovation-e-xchange13Land-BasedWindMarketReport:2023EditionNote:Eachbarreflectsthelistedresourcetype.Asolar+storagehybridwillhaveitssolarcapacityinthe‘solar’columnanditsstoragecapacityinthe‘storage’columnHybridstoragecapacityisestimatedusingstorage:generatorratiosfromprojectsthatprovideseparatecapacitydata.Source:BerkeleyLabreviewofinterconnectionqueuesFigure12.Generationcapacityininterconnectionqueues,includinghybridpowerplantsAsshowninFigure13,commercialinterestinwindhybridplantsishighestinCaliforniaandtheWest(non-ISO).Infact,45%ofthewindinCAISO’squeuesisproposedasahybrid,asis17%ofthewindintheWest.14Land-BasedWindMarketReport:2023EditionSource:BerkeleyLabreviewofinterconnectionqueuesFigure13.Hybridwindpowerplantsininterconnectionqueuesattheendof202215Land-BasedWindMarketReport:2023Edition3IndustryTrendsJustfourturbinemanufacturers,ledbyGE,suppliedalltheU.S.utility-scalewindpowercapacityinstalledin2022Ofthe8.5GWofwindinstalledintheUnitedStatesin2022,GEWindsupplied58%,followedbyVestas(24%),Nordex(10%)andSiemensGamesaRenewableEnergy(SGRE,8%).17GEandVestashavedominatedtheU.S.marketforsometime,withSGREandNordexvyingforthird(Figure14).U.S.MarketSharebyMW100%Other80%GoldwindAcciona(pre-2016)60%Nordex(pre-2016)NordexAcciona40%Gamesa(pre-2017)Siemens(pre-2017)20%SGREVestas0%GEWind200520072009201120132015201720192021Source:ACPFigure14.AnnualU.S.marketshareofwindturbinemanufacturersbyMW,2005–2022Thedomesticwindindustrysupplychainbegan2022indecline,butpassageoftheInflationReductionActhascreatedrenewedoptimismaboutsupply-chainexpansionFigure15identifiesthemanywindturbinecomponentmanufacturing,assembly,andothersupplychainfacilitiesoperatingintheUnitedStatesattheendof2022.ThreeofthefourmajorturbineOEMsthatservetheU.S.windindustry—GE,Vestas,andSGRE—arerepresentedwithinthistotal,eachhavingoneormoreoperatingmanufacturingfacility.Alsoincludedinthefigureareelevenplannednew,re-openedorexpandedfacilitiesintendedtoservetheland-basedwindindustry,allannouncedsincepassageoftheInflationReductionAct.Ingeneral,Figure15highlightsthegeographicbreadthofthesupplychain.17MarketshareisreportedinMWtermsandisbasedonprojectinstallationsintheyearinquestion.16Land-BasedWindMarketReport:2023EditionSource:ACPandBerkeleyLabFigure15.LocationofturbineandcomponentmanufacturingfacilitiesDomesticturbinenacelleassembly18capabilityisdefinedhereasthemaximumGWcapacityofnacellesthatcanbeassembledannuallyatU.S.plantsoperatingatfullutilization.Thisvaluegrewfromlessthan1.5GWin2006tomorethan13GWin2012,felltoroughly10GWin2015,andthenroseto15GWin2018andhasheldlargelysteadyatthatlevelsince(Figure16).From2012through2020,domesticbladeandtowermanufacturingcapabilitywaslargelystableorgrowing,ineachcaseincreasingfromaround7to8GW/yearin2012toaround10GW/yearin2020.Inthecaseoftowers,domesticcapabilitycontinuedtoincrease,reachingover11GWin2022,supportedinpartbyimporttariffs.In2021,however,domesticblademanufacturingplummeted—adeclinethatcontinuedinto2022,withunder4GWofbladeproductioncapabilityattheendoftheyear.Competitionfromforeignsuppliers,growingbladelengthsthatwouldrequireretoolingofmanufacturingequipment,anduncertain(pre-IRA)future18Nacelleassemblyisdefinedastheprocessofcombiningthemultitudeofcomponentsincludedinaturbinenacelle,suchasthegearboxandgenerator,toproduceacompleteturbinenacelleunit.17Land-BasedWindMarketReport:2023Editiondeploymentprospectsforland-basedwindintheUnitedStatescombinedtoweakendomesticwindmanufacturingcapabilities.Figure16contraststhisequipmentmanufacturingcapabilitywithpastU.S.windadditionsaswellasnear-termforecastsoffuturenewinstallations(seeChapter9,“FutureOutlook”).Itdemonstratesthatdomesticmanufacturingcapabilityfortowersandnacelleassemblyremainsreasonablywellbalancedwithnear-termprojectedwindadditionsintheUnitedStates,butthatblademanufacturingcapabilityhasfallenwellbelownear-termwindadditionsasinternationalsuppliersoutcompetedomesticones.Notethatmanufacturingfacilitiesdonottypicallyoperateatmaximumcapability;seethenextsectionofthereportforestimatesofdomesticmanufacturingcontent.AnnualCapacity(GW)ActualwindcapacityadditionsAverageforecastcapacityadditions25NacellemanufacturingcapacityTowerproductioncapacity20Bladeproductioncapacity151050200820102012201420162018202020222024e2026e2006Sources:ACP,independentanalystprojections,BerkeleyLabFigure16.Domesticwindmanufacturingcapabilityvs.U.S.windpowercapacityinstallationsMoregenerally,fiercecompetitionamongmanufacturersand,insomecases,technicalfailuresresultinginincreasedwarrantyclaims,hasgenerallyreducedturbineOEMprofitabilityoverthelastseveralyears.HighcommodityandtransportationcostsalongwithCOVID-19restrictionshavealsolimitedmanufacturerprofitability.Figure17illustratesthedeclining(andnegative)profitmarginsofseveralmajorinternationalturbinemanufacturersin2022.1919AlthoughitisoneofthelargestturbinesuppliersintheU.S.market,GEisnotincludedbecauseitisamulti-nationalconglomeratethatdoesnotreportsegmentedfinancialdataforitswindturbinedivision.18Land-BasedWindMarketReport:2023EditionProfitMargin(EBITDA)20%GoldwindGamesa/SGRE15%10%VestasNordex5%0%-5%-10%-15%200820092010201120122013201420152016201720182019202020212022Note:EBITDA=EarningsBeforeInterest,Taxes,DepreciationandAmortizationSources:OEMannualreportsandfinancialstatementsFigure17.TurbineOEMglobalprofitabilityDespitethesesupply-chainchallenges,wind-relatedjobtotalsintheUnitedStatesincreasedby4.5%in2022,to125,580full-timeworkers—benefittingfromcontinueddeployment(U.S.DOE2023b).Thesejobsinclude,amongothers,thoseinconstruction(45,088)andmanufacturing(23,543).Moreover,whiletheabovestorylinesaredecidedlymixedfor2022,passageoftheInflationReductionActholdspromiseforaddressingrecentchallengesandsupportingsupply-chainexpansion.TheIRAcontains,forthefirsttime,production-basedtaxcreditsfordomesticmanufacturingofkeywindturbinecomponents,includingnacelles,blades,andtowers(U.S.DOE2023a).ItalsoextendsthePTCforwindpowerdeployment,inclusiveofanew10%bonusontopofthefull-valuePTCforwindprojectsthatmeetdomesticcontentrequirements(aseparate10%bonusisavailableforprojectslocatedinenergycommunities).Consequently,asshownearlierinFigure15,sinceIRApassedtherehavebeenatleastelevenannouncementsofdomesticmanufacturingfacilitiesthatplantoopen,re-open,orexpandtoservetheland-basedwindindustry.Thisincludes:•TowerfacilitiesinNewMexico(Arcosa,newfacility),Colorado(CSWind,expansion),andSouthDakota(Marmen,expansion)•BladefacilitiesinIowa(TPICompositesandSGRE,re-openings)andColorado(Vestas,expansion)•GearboxmanufacturinginIllinois(FlenderCorporation,expansion)•Nacelleandturbinecomponentassemblyand/ormanufacturinginFlorida(GEVernova,expansion),NewYork(GEVernova,expansion),Kansas(SGRE,re-opening),andColorado(Vestas,expansion)19Land-BasedWindMarketReport:2023EditionIntotal,theseelevenplannedfacilitiesandexpansionsanticipatemorethan3,000newjobs.Additionally,KeystoneTowersbegancommercialproductionofitsfirstspiral-weldedtowersin2022,beforeIRAbecamelaw,fromanewmanufacturingfacilityinPampa,Texas.Domesticmanufacturingcontentisstrongforsomewindturbinecomponents,buttheU.S.windindustryremainsreliantonimportsDespitethebreadthofthedomesticwindindustrysupplychain,theU.S.windsectorisreliantonimportsofwindequipment.Thelevelofdependencevariesbycomponent:somecomponentshavearelativelyhighdomesticshare,whereasothersremainlargelyimported.Thesetrendsarerevealed,inpart,bydataonwindequipmenttradefromtheU.S.DepartmentofCommerce.20Figure18presentsdataonthedollarvalueofestimatedimportstotheUnitedStatesofwind-relatedequipmentthatcanbetrackedthroughtradecodes.Thefigureshowsimportsofwind-poweredgeneratingsetsandparts,includingnacelles(i.e.,nacelleswithblades,nacelleswithoutblades,and,insomecases,otherturbinecomponentsinternaltothenacelle)aswellasimportsofotherselectturbinecomponentsshippedseparatelyfromthegeneratingsetsandnacelles.21Theturbinecomponentsincludedinthefigureconsistonlyofthosethatcanbetrackedthroughtradecodes:towers,generators(aswellasgeneratorparts),andbladesandhubs.2220SeetheAppendixforfurtherdetailsondatasourcesandmethodsusedinthissection,includingthespecifictradecodesconsidered.21Windturbinecomponentssuchasblades,towers,andgeneratorsareincludedinthedataonwind-poweredgeneratingsetsandnacellesifshippedinthesametransaction.Otherwise,thesecomponentimportsarereportedseparately.22Thoughalltheimportestimatesinthefiguresince2020arespecifictowindequipment,importtrendsshouldbeviewedwithcautionbecausetheunderlyingdatafromearlieryearsarebasedontradecategoriesthatarenotallexclusivetowind.Someoftheseearlier-yearestimatesthereforerequiredassumptionsaboutthefractionoflargertradecategorieslikelytoberepresentedbywindturbinecomponents.Notealsothatthetradecodefortowersisnotexclusivetowind,butisbelievedtobedominatedbywindsince2011—weassumethat100%ofimportsfromthistradecategory,since2011,representwindequipment.20Land-BasedWindMarketReport:2023EditionImports(Billion2022$)Otherwind-relatedequipment6WindgeneratorsandgeneratorpartsWindbladesandhubsWindtowersWind-poweredgeneratingsetsandparts,includingnacelles42020062007200820092010201120122013201420152016201720182019202020212022Note:Wind-relatedtradecodesanddefinitionsarenotconsistentoverthefulltimeperiod.Source:BerkeleyLabanalysisofdatafromUSATradeOnline,https://usatrade.census.govFigure18.Importsofwind-relatedequipmentthatcanbetrackedwithtradecodesTheestimatedimportsoftrackedwind-relatedequipmentintotheUnitedStatesincreasedsubstantiallyfrom2006to2008,beforefallingthrough2010,increasingsomewhatin2011and2012,andthenplummetingin2013withthesimultaneousdropinU.S.windinstallations.From2014through2022,importsofwind-relatedturbineequipmentgenerallyfollowedU.S.windinstallationtrends,bouncingbackfromthelowof2013andthenwithamarkeddeclinein2021and2022aswindplantinstallationsalsodeclined.Interpretingtimetrendsinthesedataischallenginggivenchangesinannualwindadditionsfromyeartoyear,timelagsbetweenequipmentimportandinstallation,andfluctuationsinwindturbineandequipmentpricing.Also,becauseimportsofcomponentpartsoccurinadditional,broadtradecategoriesdifferentfromthoseincludedinFigure18,thedatapresentedhereunderstatetheaggregateamountofwindequipmentimports.Nonetheless,focusingonthesubsetoftradecategoriesshowninFigure18andnormalizingbywindturbinepricesandtimelags,overallturbine-levelimportsharesareestimatedtohaveincreasedfromroughly20%in2015toover35%in2022.ThissuggeststhattheU.S.hasbecomemorereliantonimportsoverthisperiod.Figure19showsthetotalvalueoftrackedwind-specificimportstotheUnitedStatesin2022,bycountryoforigin,aswellasstatesofentry.MajorcountriesfromwhichtheUnitedStatesimportswindequipmentincludeMexico,India,andSpain,whichtogetheraccountfor$1.4billioninwind-specificexportstotheU.S.in2022.Texasremainedthedominantentrypointin2022,withnearly$1.4billionofwind-specificequipmentflowingthroughitlastyear,followeddistantlybyNewYork,Michigan,Florida,andOhio.21Land-BasedWindMarketReport:2023EditionNote:Linewidthsareproportionaltoimportamountbycountry.Figuredoesnotintendtodepictthedestinationoftheseimports,bystate.Source:BerkeleyLabanalysisofdatafromUSATradeOnline,https://usatrade.census.govFigure19.Summarymapoftrackedwind-specificimportsin2022:top-10countriesoforiginandstatesofentryLookingbehindthesedata,India,followedbyDenmark,Spain,Belgium,andSweden,weretheprimarysourcecountriesforwind-poweredgeneratingsetsandparts,includingnacelles,in2022(Figure20).Towerimportscamefromamixofcountriesnearandfar—SouthKorea,Canada,Mexico,Argentina,andMalaysia.Forbladesandhubs,MexicoandIndiaaccountedforalmost70%ofimports,withSpain,China,andCanadathenextlargestsourcecountriesin2022.Finally,almost80%ofwind-relatedgeneratorsandgeneratorpartsin2022camefromVietnam,Germany,andSpain,therestprimarilycomingfromSerbiaandAustria.22Land-BasedWindMarketReport:2023EditionWind-poweredgeneratingsetsandWindgeneratorsandpartsparts,includingnacellesTotal2022imports:Total2022imports:Asia$632million$146millionEuropeNorthAmericaTopcountries:Topcountries:SouthAmericaIndia(42%)Vietnam(44%)OtherGermany(19%)Denmark(18%)Spain(16%)Spain(16%)Serbia(5%)Belgium(11%)Austria(5%)Sweden(7%)WindbladesandhubsWindtowersTotal2022imports:Total2022imports:$1,237million$178millionTopcountries:Topcountries:Mexico(49%)S.Korea(28%)India(20%)Canada(23%)Spain(10%)Mexico(13%)China(6%)Argentina(11%)Canada(4%)Malaysia(6%)Source:BerkeleyLabanalysisofdatafromUSATradeOnline,https://usatrade.census.govFigure20.OriginsofU.S.importsofselectedwindturbineequipmentin2022Figure21presentsroughestimatesofthedomesticcontentforasmallsubsetofthemajorwindturbinecomponentsusedinnew(andrepowered)U.S.windprojectsin2022.Asshown,forwindprojectsinstalledin2022,over85%ofnacelleassemblyand70%–85%oftowermanufacturingoccurredintheUnitedStates.Inthecaseoftowers,tariffsonsomeimportsinfluencethehighlevelofdomesticcontent.Thedomesticmanufacturingcontentofbladesandhubs,ontheotherhand,hasdeclinedprecipitouslyinrecentyears,tojust5%–25%in2022.Morebroadly,thesefiguresmayunderstatethewindindustry’srelianceonforeignsuppliers,becausesignificantwind-relatedimportsoccurundertradecategoriesnotcapturedinthisfigure.HowthesetrendschangeafterpassageoftheInflationReductionActremainstobeseen,thoughsupply-chainannouncementsinrecentmonthssuggestaresurgenceindomesticwindmanufacturing.23Land-BasedWindMarketReport:2023EditionNacelleAssembly>85%WindTowers70−85%BladesandHubs5−25%0%20%40%60%80%100%Source:BerkeleyLabanalysisDomesticContentFigure21.Approximatedomesticcontentofmajorcomponentsin2022Independentpowerproducersownmostwindassetsbuiltin2022,extendinghistoricaltrendsIndependentpowerproducers(IPPs)own7,116MWor84%ofthe8.5GWofnewwindcapacityinstalledintheUnitedStatesin2022(Figure22,rightpiechart).Investor-ownedutilities(IOUs)—mostnotablythePublicServiceCompanyofOklahoma(996MW),butalsoincludingNorthernStatesPowerCompany(326MW)andDTEEnergy(72MW)—owntheremaining1,395MW(16%).Ofthecumulativeinstalledwindpowercapacityattheendof2022(Figure22,leftchart),IPPsown81%andutilitiesown18%(17%IOUand1%publicly-ownedutility,orPOU),withtheremaining1%fallingintothe“other”categoryofprojectsownedbyneitherIPPsnorutilities(e.g.,ownedbytowns,schools,businesses,farmers,etc.).2323Manyofthe“other”projects,alongwithsomeIPP-andPOU-ownedprojects,mightalsobeconsidered“communitywind”projectsthatareownedbyorbenefitoneormoremembersofthelocalcommunitytoagreaterextentthantypicallyoccurswithacommercialwindproject.Notethatanychangestoownershiporofftakebeyondthecommercialoperationdataarenottrackedinthisorthefollowingsection.24Land-BasedWindMarketReport:2023Edition%ofCumulativeInstalledCapacityOther2022Capacityby100%OwnerTypeInvestor-OwnedUtility(IOU)IPP:7,116MW80%PubliclyOwnedUtility(POU)IOU:60%1,395MW40%IndependentPowerProducer(IPP)20%0%1998200020022004200620082010201220142016201820202022Source:BerkeleyLabestimatesbasedonACPFigure22.Cumulativeand2022windpowercapacitycategorizedbyownertypeForthefirsttime,non-utilitybuyersenteredintomorecontractstopurchasewindthandidutilitiesin2022Whereasthepriorsectionanalyzeswindplantownership,thissectionfocusesonwhousesorbuysthewindgenerationfromthoseplants.Electricutilitieseitherown(16%)orbuytheelectricityfrom(17%)windprojectsthat,intotal,represent33%ofthenewcapacityinstalledlastyear(withthe33%splitbetween29%IOUand5%POU—Figure23,rightpiechart).Onacumulativebasis,utilitiesown(18%)orbuy(40%)powerfrom58%ofallwindpowercapacityinstalledintheUnitedStates(withthe58%splitbetween41%IOUand17%POU,withthePOUcategoryincludingcommunitychoiceaggregators(CCAs)).Directretailpurchasersofwindpower,includingadiverseandgrowingsetofcorporateandnon-corporateofftakers,supportedatleast44%ofthenewwindpowercapacityinstalledintheUnitedStatesin2022(and15%ofcumulativewindpowercapacity).Suchpurchasershistoricallyhavespannedawiderangeoforganizations,fromtechnologycompanies(e.g.,Microsoft,Google),retailers(e.g.,Walmart,Lowe’s,Gap),finance(e.g.,WellingtonManagement,JPMorganChase),andtelecommunicationfirms(e.g.,AT&T,Verizon,Sprint)togovernments(e.g.,MarylandDepartmentofGeneralServices)anduniversities(e.g.,BostonUniversity).Merchant/quasi-merchantprojectsaccountedforatleast3%ofallnew2022capacityand19%ofcumulativecapacity.24Finally,powermarketers—definedheretoincludecommercialintermediaries24Merchant/quasi-merchantprojectsarethosewhoseelectricitysalesrevenueistiedtoshort-termcontractsand/orwholesalespotelectricitymarketprices(withtheresultingpriceriskcommonlyhedgedovera10-to12-yearperiod),ratherthanbeinglockedinthroughalong-termPPA.MostoftheseprojectsarelocatedwithinERCOT,thoughtherearesomemerchant/quasi-merchantprojectswithinothermarkets,includingPJM,MISO,SPP,andNYISO.Associatedhedgesareoftenstructuredasa“fixed-for-floating”powerpriceswap—apurelyfinancialarrangementwherebythewindpowerprojectswapsthe“floating”revenuestreamthatitearnsfromspotpowersalesfora“fixed”revenuestreambasedonanagreed-uponstrikepricewiththeswapcounterparty.Notethatanychangestoownershiporofftakebeyondthecommercialoperationdataarenottrackedhere.25Land-BasedWindMarketReport:2023Editionthatpurchasepowerundercontractandthenresellthatpowertoothers25—boughtatleasttheremaining6%ofnew2022windcapacityand5%ofcumulativecapacity.Wequalifythelevelofsupportfromthesenon-utilityofftakersas“atleast”becauseitislikelythatmuchofthe1.2GWof2022capacitythathasnotyetdisclosedanofftakerisbeingsoldtocorporatebuyers,powermarketers,orintomerchantarrangements,ratherthantoutilities.%ofCumulativeInstalledCapacityUndisclosed2022Capacityby100%DirectRetailOfftakeTypeMerchant/Quasi-Merchant80%PowerMarketerMerchant:263MW60%POURetail:IOU:40%3,776MW2,443MW20%IOUPOU:0%392MW1998PowerMarketer:2000481MWUndisclosed:20021,155MW2004200620082010201220142016201820202022Source:BerkeleyLabestimatesbasedonACPFigure23.Cumulativeand2022windpowercapacitycategorizedbypowerofftakearrangement25TheseintermediariesincludethewholesalemarketingaffiliatesoflargeIOUs,whichmaybuywindonbehalfoftheirload-servingaffiliates.26Land-BasedWindMarketReport:2023Edition4TechnologyTrendsTurbinecapacity,rotordiameter,andhubheighthaveallincreasedsignificantlyoverthelongtermTheaveragenameplatecapacityofnewlyinstalledwindturbinesintheUnitedStatesin2022was3.2MW,7%largerthanin2021andup350%since1998–1999(Figure24).26Theaveragehubheightofturbinesinstalledin2022was98.1meters,4%largerthanin2021andup73%since1998–1999.Theaveragerotordiameterin2022was131.6meters,3%largerthanin2021andup173%since1998–1999.Thetrends,inturn,impacttheproject-levelcapacityfactorshighlightedlaterinthisreport.Capacity(MW)Height&Diameter(m)3.51403.0Rotordiameter1202.51002.0Hubheight801.5601.0400.5200.020062008Nameplatecapacity201820200'98−99'02−0320102012201420162022Sources:ACP,BerkeleyLabFigure24.Averageturbinenameplatecapacity,hubheight,androtordiameterforland-basedwindprojectsFigure25presentsthesesametrendssince2012,butwithadditionaldetailontherelativedistributionofturbineswithdifferentcapacities,hubheights,androtordiameters.Forexample,2022sawanincreaseintheproportionofturbinesinstalledinthe2.75–3.5MWrange,whiletheproportionofturbinesat3.5MWorlargeralsoincreased.Thepercentageofturbineswithhubheightslargerthan100metersincreasedin2022,to43%—upfrom27%in2021andjust2%in2018.Finally,thesteadyprogressiontowardlargerrotorscontinued.In2012,only1%ofturbinesemployedrotorsthatwere115metersindiameterorlarger,while98%ofnewlyinstalledturbinesfeaturedsuchrotorsin2022(and29%ofthosewereatleast130meters).26Figure24andanumberoftheotherfiguresandtablesincludedinthisreportcombinedataintobothone-andtwo-yearperiodsinordertoavoiddistortionsrelatedtosmallsamplesizeinthePTClapseyearsof2000,2002,and2004;althoughnotaPTClapseyear,1998isgroupedwith1999duetothesmallsampleof1998projects.Though2013wasaslowyearforwindadditions,itisshownseparatelyheredespitethesmallsamplesize.27Land-BasedWindMarketReport:2023Edition%ofturbinesAverageCapacity(M%Wo)fTurbinesAverageHubheight%(mof)TurbinesAverageRotordiameter(m)100%130.50%10100%014080%average38.0%8905%12060%26.50%6900%average10040%≥3.5MW24.0%average≥100m4805%≥130m802.75−3.5MW90−100m115−130m20%2.0−2.75MW12.50%2800%80−90m100−115m600%<2.0MW1.0%<80m705%<100m40201220142016201820202022201220142016201820202022201220142016201820202022Sources:ACP,BerkeleyLabFigure25.Trendsinturbinenameplatecapacity,hubheight,androtordiameterTurbinesoriginallydesignedforlowerwindspeedsitesdominatethemarket,butthetrendtowardslowerspecificpowerhasreversedinrecentyearsAswindturbinebladelengthhasincreasedovertime,theamountofareathebladescoverwhenspinning,knownastherotorsweptarea(inm2),hasgrownrapidlyoverthelasttwodecades.Rotorsweptareahasgrownfasterthantheincreaseinaveragenameplatecapacityofwindturbinesovertime.Thishasresultedinadeclineintheaverage“specificpower”amongtheU.S.turbinefleetovertime,whichiscalculatedbydividingthenameplatecapacity(inwatts[W])bytherotorsweptarea(m2).Thisvaluehasdeclinedfrom393W/m2amongprojectsinstalledin1998–1999to233W/m2amongprojectsinstalledin2022.However,asshowninFigure26,thelong-termdeclineinspecificpowerhasreversedinrecentyears,withspecificpowerrisingslightlysincethelowpointin2019asturbineswithaspecificpowerintherangeof180–200W/m2havebecomelesspopularoravailableaswindturbinecapacitieshaveincreasedsignificantlyoverthistimeframe.Allelseequal,alowerspecificpowerwillboostcapacityfactors,becausethereismoresweptrotorareaavailable(resultingingreaterenergycapture)foreachwattofratedturbinecapacity.Thismeansthatthegeneratorislikelytorunclosertooratitsratedcapacitymoreoften.Ingeneral,turbineswithlowspecificpowerwereoriginallydesignedforlowerwindspeedsites,intendedtomaximizeenergycaptureinareaswherelarge-rotormachineswouldnotbeplacedunderexcessivephysicalstressduetohighorturbulentwinds.AssuggestedinFigure26andasdetailedlater,however,suchturbinesareinwidespreaduseintheUnitedStates—eveninsiteswithhighwindspeeds.Theimpactoflowerspecific-powerturbinesonproject-levelcapacityfactorsisdiscussedinmoredetailinChapter5.28Land-BasedWindMarketReport:2023Edition%ofTurbinesAverageSpecificpower(W/m2)100%40080%35060%average300≥350≥300−35040%≥250−300250≥200−250180−20020%2000%2008201020122014201620182020150'98−99'02−0320062022Sources:ACP,BerkeleyLabFigure26.TrendsinwindturbinespecificpowerWindturbinesweredeployedinhigherwind-speedsitesin2022thaninrecentyearsFigure27showsthelong-termaveragewindresourceatwindprojectsites,bycommercialoperationdate.Thefiguredepictsthesite-averagewindspeed(inmeterspersecond,ontherightaxis)bothat100metersandatthehubheightsforprojectsinstalledineachyear.Windresourcequalityat100meters(bluebars)ismeasuredontheleftaxis.27Windprojectsthatcameonlinein2022arelocated—onaverage—atsiteswithanestimatedlong-termaverage100-meterwindspeedof8.3meterspersecond(m/s,or18.6milesperhour).Giventhattheaveragehubheightamong2022windplantswasnearly100meters,thesame8.3m/swindspeedlargelyholdsathubheightaswell.Measuredat100meters,thisisthehighestsite-averagewindspeedsince2014.Measuredataveragehubheight,itisthehighestsinceatleast1998–1999.Thedifferenttrendsat100meters(shownbytheblueline,withanoveralldeclinesince1998-1999)andathubheight(shownbytheredline,withanoverallincreasesince1998-1999)illustratethevalueofincreasinglytallertowersinboostingrealizedaveragewindspeedsathubheight.Meanwhile,FederalAviationAdministration(FAA)andindustrydataonprojectsthatare“underconstruction,”in“advanceddevelopment,”“pending,”or“proposed”suggestthatprojectswillbebuiltinlesswindysites.28Trendsinthewindresourcequalityindex—whichrepresentsestimatesofthegross27Thewindresourcequalityindexisbasedonsiteestimatesofgrosscapacityfactorat100metersbyAWSTruepower.Asingle,commonwindturbinepowercurveisusedacrossallsitesandtimeframesinthiscase,andnolossesareassumed.Thevaluesareindexedtoprojectsbuiltin1998—1999.FurtherdetailsarefoundintheAppendix.Abenefitofthiswindresourcequalityindexisthatchangesintheindexvaluewillbetterapproximateexpectedchangesinactualwindprojectperformancethanwillchangesinaverageannualwindspeed.28“Underconstruction”turbinesarepartofaprojectwhereconstructionhasbegun,buttheprojecthasnotyetbeencommissioned.Turbinesin“advanceddevelopment”haveoneofthefollowinginplace:asignedPPA(orsimilarlong-termcontract),afirmturbineorder,oranannouncementtoproceedunderutilityownership,indicatingahighlikelihoodthattheywillbebuilt.“Pending”turbinesarethosethathavereceiveda“NoHazard”determinationbytheFAAandarenotsettoexpirefor29Land-BasedWindMarketReport:2023Editioncapacityfactorforeachturbinelocation,indexedtothe1998–1999installations—arebroadlysimilartoaveragewindspeedestimatesat100meters.Windresourcequality(index,100m)Windspeed(m/s,100m&hubheight)1008.895PastFuture8.68.490WindSpeed8.285(100m)8.0807.875(WhuinbdhSepigehetd)7.6UnderConst.Adv.Dev.7.4Pending70IndexofwindresourcequalityProposed7.2(1998−1999=100)657.0'98-99'02-03200620082010201220142016201820202022Sources:ACP,BerkeleyLab,AWSTruepower,FAAObstacleEvaluation/AirportAirspaceAnalysisfilesFigure27.Windresourcequalitybyyearofinstallationat100metersandatturbinehubheightSeveralfactorscouldhavedriventheobservedlong-termtrendsinaveragesitequalityandwindspeeds.First,theavailabilityoflow-wind-speedturbinesthatfeaturelowerspecificpowerhasenabledtheeconomicbuild-outoflower-wind-speedsites;thesameistruewithtallertowers.Second,transmissionconstraints(orothersitingconstraints,orevenjustregionallydifferentiatedwholesaleelectricityprices)mayhave,overtime,increasinglyfocuseddeveloperattentiononthoseprojectsintheirpipelinethathaveaccesstotransmission(orhigher-pricedmarkets,orreadilyavailablesiteswithoutlongpermittingtimes),eveniflocatedinsomewhatlowerwindresourcesites.Thesefactorsmaypartiallyexplainwhyaverageresourcequalityandwindspeedsdroppedfromthelate1990sto2012andagaintendedtodeclinefrom2014through2021.Thebuild-outofnewtransmission(forexample,thecompletionofmajortransmissionadditionsinWestTexasin2013),however,mayattimeshaveofferedthechancetoinstallnewprojectsinmoreenergeticsites.Otherformsoffederaland/orstatepolicycouldalsoplayarole.Forexample,windprojectsbuiltinthefour-yearperiodfrom2009through2012wereabletoaccessa30%cashgrant(orITC)inlieuofthePTC.Manyprojectsavailedthemselvesofthisincentiveand,becausethedollaramountofthegrant(orITC)wasnotdependentonhowmuchelectricityaprojectgenerates,itispossiblethatdevelopersalsoseizedthislimitedopportunitytobuildouttheless-energeticsitesintheirdevelopmentpipelines.Statepoliciescanalsosometimesmotivatein-stateorin-regionwinddevelopmentinlowerwindresourceregimes.Finally,thesizableincreaseinsite-averagewindresourcequalityin2022maybeduetotherelativelyslowpaceofnewprojectinstallationsin2022,partiallyaconsequenceofthedecliningvalueofanduncertaintyintheproductiontaxcreditpriortotheInflationReductionAct.Thesefactorstendedtoconcentratedeveloperattentionin2022onthehighest-qualityandlowest-costwindsitesinSPPandERCOT,leadingtoabuildoutinhighwind-speedareas.another18months,while“proposed”turbineshavenotyetreceivedanydetermination.Pendingandproposedturbinesmaynotallultimatelybebuilt.However,analysisofpastdatasuggeststhatFAApendingandproposedturbinesofferareasonableproxyforturbinesbuiltinsubsequentyears.30Land-BasedWindMarketReport:2023EditionLow-specific-powerturbinesaredeployedonawidespreadbasis;tallertowersareseeingincreaseduseinawidervarietyofsitesOnemightexpectthattheincreasingmarketshareoflow-specific-powerturbines(definedhereasturbineswithspecificpower<250W/m2)wouldbeduetoamovementbydeveloperstodeployturbinesinlowerwindspeedsites.Thereissomeevidenceofthismovementhistorically(seeFigure27),butitisclearinFigure28(whichshowsallU.S.windprojects)thatlow-specific-powerturbineshaveestablishedastrongfootholdacrossthenationandoverawiderangeofwindspeeds.Sources:ACP,U.S.WindTurbineDatabase,AWSTruepower,BerkeleyLabFigure28:Locationoflowspecificpowerturbineinstallations:allU.S.windplantsLikewise,tallertowersarealsobeingdeployedacrossawidearrayofsites(Figure29).Thatsaid,verytalltowers(>100m)stilltendtobemostconcentratedwithintheupperMidwestandNortheastregions,tworegionsknowntohavehigher-than-averagewindshear(i.e.,greaterincreasesinwindspeedwithheight),whichmakestallertowersmoreeconomical.31Land-BasedWindMarketReport:2023EditionSources:ACP,U.S.WindTurbineDatabase,AWSTruepower,BerkeleyLabFigure29:Locationoftalltowerturbineinstallations:allU.S.windplantsWindprojectsplannedforthenearfuturearepoisedtocontinuethetrendofever-tallerturbinesFAAdataontotalproposedturbineheights(fromgroundtobladetipextendeddirectlyoverhead)inpermitapplicationsarereportedinFigure30.Notethatthesedatarepresenttotalturbineheightor“tipheight”—nothubheight—andincludethecombinedeffectofboththetowerandhalftherotordiameter.Figure30showstheaverageFAAtipheight,alongwiththedistribution,for2022installationsaswellasturbinesunderconstruction,inadvanceddevelopment,pending,andproposed.29Averagetipheightsforprojectsthatcameonlinein2022are164meters,upfrom158metersfor2021projects,andseemdestinedtoclimbhigherinthenextfewyears,reachinganaverageof195metersamongthe“proposed”turbines.Thetallestturbinesinthepermittingprocessareover225meters.Turbinesofatleast200metersappearlikelytobeinstalledinnearlyeveryregionoftheUnitedStates,apartfromtheSoutheast(non-ISO)region(Figure31).29TurbineheightsreportedinFAApermitapplicationsrepresentthemaximumheightandcandifferfromwhatisultimatelyinstalled.Historically,however,theFAApermitdatasetshavestronglyconformedtosubsequentactualinstallationsonaverage.32Land-BasedWindMarketReport:2023Edition%ofturbinesAverageTotalheight(meters)100%22580%200>225m60%40%200-225m20%175-200m175150-175m<150m150Average:rightaxis0%Adv.dev.Pending1252022projectsUnderconst.ProposedSources:ACP,FAAfiles,BerkeleyLabFigure30.TotalturbineheightsproposedinFAAapplications,bydevelopmentstatusNote:FigureincludesFAAdataonunder-construction,advanceddevelopment,pending,andproposedturbinesSources:FAAObstacleEvaluation/AirportAirspaceAnalysisfiles,AWSTruepower,ACP,BerkeleyLabFigure31.TotalturbineheightsproposedinFAAapplications,bylocation33Land-BasedWindMarketReport:2023EditionIn2022,thirteenwindprojectswerepartiallyrepowered,mostofwhichnowfeaturesignificantlylargerrotorsandlowerspecificpowerratingsThetrendofpartialwindprojectrepoweringcontinuedin2022,albeitataslowerpacethanin2019-2020,andinvolvedreplacingmajorcomponentsofturbineswithmore-advancedtechnologytoincreaseenergyproduction,extendprojectlife,andaccesstaxincentives.In2022,13projectswerepartiallyrepowered,involving838turbinesthattotaled1.7GWpriortorepowering.Retrofittedturbinesrangedinagefrom10to15yearsold;themedianwas11years.The1.7GWofretrofittedturbinesin2022isaslightincreasefromthe1.6GWretrofittedin2021,butadeclinefrom2019and2020,when3GWwereretrofittedeachyear(Figure32).ProjectCapacity(MW)andNumberofTurbines(#)Originalcapacity(MW)4,000Addedcapacity(MW)Numberofturbines(#)3,0002,0001,0000201820192020202120222017Sources:ACP,BerkeleyLab,turbinemanufacturersFigure32.AnnualamountofpartiallyrepoweredwindpowercapacityandnumberofturbinesThemostcommonretrofitin2022wasthereplacementofshorterwithlongerblades,butchangesinturbinenameplatecapacitywerealsocommon.Overall,theaverageturbinenameplatecapacityoftheretrofittedprojectsincreasedmodestly(thefinalrepoweredcapacityoftheseplantsis1.8GW),butrotordiametersstronglyincreased(Figure33).Noneoftheturbinesretrofittedin2022sawachangeinhubheight.Withtherelativelysmallchangeincapacitybutthelargerchangeinrotordiameter,theseretrofitsdroveasignificantdecreaseinaveragespecificpower,from300to220W/m2.34Land-BasedWindMarketReport:2023EditionHubheight&rotordiameter(m)Capacity(MW)Specificpower(W/m2)350140100←leftaxis2.81208092.82.43008101.20.380.03001002.032.152.02508092.6801.622020080.080.060401.2150400.8100200.45020000.00OriginalRetrofittedOriginalROeritgroinfiattledRetOrorifgititneadlORriegtirnoafilttedRetrofittedOriginalRetrofittedHubHeightRotorDiametHeurbHeightCapacRityotorDiameterSpecificPowerSources:ACP,BerkeleyLab,turbinemanufacturersFigure33.Changeinaveragephysicalspecificationsofallturbinesthatwerepartiallyrepoweredin202235Land-BasedWindMarketReport:2023Edition5PerformanceTrendsTheaveragecapacityfactorin2022was36%onafleet-widebasisand37%amongwindplantsbuiltin2021Followingthepreviousdiscussionoftechnologytrends,thischapterpresentsdatafromacompilationofproject-levelcapacityfactors.30Thefulldatasampleconsistsof1,160windprojectsbuiltbetween1998and2021andtotaling128.7GW.Excludedfromthisassessmentareolderprojectsinstalledpriorto1998.Inaddition,projectsthateitherpartiallyorfullyrepoweredin2022areexcludedfromthe2022capacityfactorsample,giventhattheywereatleastpartlyofflineduringaportionoftheyear.Unlessotherwisenoted,allcapacityfactorsinthischapterarereportedonanas-observedandunadjustedbasis(i.e.,afteranylossesfromcurtailment,less-than-fullavailability,wakeeffects,iceorsoilonblades,etc.).Whenlookingatperformancedegradationovertime,however,adjustmentsaremadeforinter-annualvariabilityinthewindresource(asdescribedintheAppendix).Tostart,Figure34showsbothindividualprojectandaveragecapacityfactorsin2022,brokenoutbycommercialoperationdate.31Projectsbuiltin2022areexcluded,asfull-yearperformancedataarenotyetavailableforthoseprojects.Fromlefttoright,Figure34showsanincreaseinweighted-average2022capacityfactorswhenmovingfromprojectsinstalledinthe1998–2003periodtothoseinstalledinthe2004–2005period.Subsequentprojectvintagesthrough2012showlittleifanyimprovementinaveragecapacityfactorsrecordedin2022.Thispatternofstagnationisbrokenbyprojectsinstalledin2013–2021.Theaverage2022capacityfactoramongprojectsbuiltfrom2013to2021was40%,comparedtoanaverageof31%amongallprojectsbuiltfrom2004to2012,and23%amongallprojectsbuiltfrom1998to2003.Cumulative,fleet-wideperformancehasalsoincreasedovertime,growingfromunder27%in1999(notshown)to36%in2022(showninFigure34).Theseoveralltrendsareimpactedbyseveralfactorsthatareexploredlater,includingprojectlocationandthequalityofthewindresourceateachsite,turbinescalinganddesign,andperformancedegradationovertime.The2022capacityfactorforprojectsbuiltmostrecently,in2021,was37%,lowerthanthe41%averageamongprojectsbuiltfrom2014to2020andcontinuingacapacityfactordeclinethatbeganwithwindprojectsbuiltin2019,followingapeakaveragecapacityfactorof44%amongprojectsthatcameonlinein2018.3230Capacityfactorisameasureoftheactualenergygeneratedbyaprojectoveragiventimeframe(typicallyannually)relativetothemaximumpossibleamountofenergythatcouldhavebeengeneratedoverthatsametimeframeiftheprojecthadbeenoperatingatfullcapacitytheentiretime.31Focusingoncapacityfactorsinasingleyear,2022,controls(atleastloosely)forfactorsthatcanimpactperformancefromoneyeartothenextbutthatareunrelatedtotechnologychange,forexample,thedegreeofwindpowercurtailmentorinter-annualvariabilityinthestrengthofthewindresource.ButitalsomeansthattheabsolutecapacityfactorsshowninFigure34maynotberepresentativeoverlongertermsif2022wasnotarepresentativeyearintermsofcurtailmentorthestrengthofthewindresource(asnotedlater,2022wasanabove-averagewindyearoverall).32The2022capacityfactorofprojectsthatwerebuiltin2021maybebiasedlow,duetopossiblefirst-year“teething”issues,asprojectsmaytakeafewmonthstoachievenormal,steady-stateproductionafterfirstachievingcommercialoperations.36Land-BasedWindMarketReport:2023EditionCapacityFactorin2022Individualprojects(byCODyear)60%Fleet-wideaverage50%(acrossallCODyears)Generation-weightedaverage(byCODyear)40%30%20%10%0%1998-992000-012002-032004-052006200720082009201020112012201320142015201620172018201920202021CommercialOperationYear(CODYear)Sources:EIA,FERC,BerkeleyLabFigure34.Calendaryear2022capacityfactorsbycommercialoperationdateStateandregionalvariationsincapacityfactorsreflectthestrengthofthewindresource;capacityfactorsarehighestinthecentralpartofthecountryTheproject-levelspreadincapacityfactorsshowninFigure34isenormous,withcapacityfactorsin2022rangingfromaminimumof9%toamaximumof53%amongthoseprojectsbuiltin2021.Someofthespread—forprojectsbuiltin2021andearlier—isattributabletoregionalvariationsinaveragewindresourcequality.Figure35showsaveragestate-levelcapacityfactorsin2022forthefullsampleofprojectsbuiltfrom1998through2021(left)andasubsetofnewerprojectsbuiltfrom2017through2021(right).Theoverallrangerunsfrom21%–48%,withconsiderablyhighercapacityfactorsintheinteriorofthecountry.ConsistentwithFigure34,thesubsetofnewerprojectsshownintheright-handmapgenerallydemonstratehigherstate-averagecapacityfactorsthanthoseamongthefullsampleshownintheleft-handmap.Figure35.Averagewindcapacityfactorincalendaryear2022bystateNote:Statesshadedinwhitehavenoprojectsinfullsample(left)orinnewersample(right)Sources:EIA,FERC,BerkeleyLab37Land-BasedWindMarketReport:2023EditionTurbinedesignandsitecharacteristicsinfluenceperformance,withdecliningspecificpowerleadingtosizableincreasesincapacityfactoroverthelongtermThetrendsinaveragecapacityfactorbycommercialoperationdateseeninFigure34canlargelybeexplainedbyseveralunderlyinginfluencesdescribedinChapter4andshownagaininFigure36.First,asdocumentedinChapter4,therehasbeenalong-termtrendtowardlowerspecificpowerandhigherhubheights.ThesetwodriversareshownagaininFigure36inindexform,relativetoprojectsbuiltin1998–1999(withspecificpowershownintheinverse,tocorrelatewithcapacityfactormovements).Allelseequal,alowerspecificpowerwillboostcapacityfactors,becausethereismoresweptrotorareaavailable(resultingingreaterenergycapture)foreachwattofratedturbinecapacity.Meanwhile,increasingturbinehubheightshelpstherotoraccesshigherwindspeeds.Second,counterbalancingthesedrivershasbeenthetendencytobuildnewwindprojectsinareasthatfeatureloweraveragewindspeeds,33especiallyamongprojectsinstalledfrom2009through2012asshownbythewindresourcequalityindexinFigure36.Thistrendreversedcoursein2013and2014,butthendriftedloweronceagainthrough2021(thesewindresourcetrendsareeasiertoseeinFigure27,wherethey-axisscaleislessexpansive).Finally,asshownlater,twootherdriversmightincludeprojectage(giventhepossibledegradationinperformanceamongolderprojects)andincreasingcurtailmentoverthepastfewyears(curtailmentisbakedintothecapacityfactorsshownthroughoutthischapter).AverageCapacityFactorin2022IndexofCapacityFactorInfluences(1998−99=100)50%20040%InverseofbuiltspecificpowerCapacityfactor17530%150Builtturbinehubheight20%12510%Builtwindresourcequalityat100m1000%1998-99752000-012002-032004-0520062007200820092010201120122013201420152016201720182019202020212022CommercialOperationYear(CODYear)Note:Tohaveallthreeindicesbedirectionallyconsistentwiththeirinfluenceoncapacityfactor,thisfigureindexestheinverseofspecificpower(i.e.,adeclineinspecificpowercausestheindextoincreaseratherthandecrease).Sources:EIA,FERC,BerkeleyLabFigure36.2022capacityfactorsandvariousdriversbycommercialoperationdate33AsdescribedearlierrelatingtoSources:ACP,BerkeleyLab,AWSTruepower,FAAObstacleEvaluation/AirportAirspaceAnalysisfilesFigure27(withfurtherdetailsintheAppendix),estimatesofwindresourcequalityarebasedonsiteestimatesofgrosscapacityfactorat100meters,asderivedfromnationwidewindresourcemapscreatedforNRELbyAWSTruepower.Thosesiteestimatesareindexedtoprojectsbuiltin1998–1999.38Land-BasedWindMarketReport:2023EditionInFigure36,thesignificantimprovementinaverage2022capacityfactorsfromamongthoseprojectsbuiltin1998–2001tothosebuiltin2004–2005isdrivenbybothanincreaseinhubheightandadeclineinspecificpower,despiteashifttowardsomewhatlower-qualitywindresourcesites.Thestagnationinaveragecapacityfactorsthatsubsequentlypersiststhrough2011-vintageprojectsreflectsrelativelyflattrendsinbothhubheightandspecificpower,coupledwithanongoingdeclineinwindresourcequalityatbuiltsites.Thesharpincreaseinaveragecapacityfactorsamongprojectsbuiltfrom2012to2018isdrivenbyasteepreductioninaveragespecificpoweroverthatentireperiod,coupledwithamarkedimprovementinthequalityofwindresourcesitesinthefirstfewyearsandanincreaseinaveragehubheightinthelastfewyearsofthatperiod.Finally,projectsbuiltafter2018hadloweraveragecapacityfactorsin2022,drivenbyaslightriseinspecificpowerandacontinuingmovetowardslower-qualitywindresourcesites.Inaddition,projectsthatcameonlinein2021mayhavealsoexperiencedteethingissuesthatoftenconfrontprojectsintheirfirstyears.Lookingaheadto2023,projectswithcommercialoperationdatesin2022couldrecordhighercapacityfactorsonaveragethanthosebuiltin2021,consideringstrongincreasesinbothaveragehubheightandsitequality(despiteslightlyhigheraveragespecificpower).Tohelpdisentangletheprimaryandsometimescompetinginfluencesofturbinedesignevolutionandwindresourcequalityoncapacityfactor,Figure37controlsforeach.Acrossthex-axis,projectsbuiltfrom2014to2021aregroupedintofourdifferentcategories,dependingonthewindresourcequalityestimatedforeachsite.Withineachwindresourcecategory,projectsarefurtherdifferentiatedbytheirspecificpower.Aswouldbeexpected,projectssitedinhigherwindspeedareasgenerallyrealizedhighercapacityfactorsin2022thanthoseinlowerwindspeedareas,regardlessofspecificpower.Likewise,projectsthatfallintoalowerspecificpowerrangetypicallyrealizedhighercapacityfactorsin2022thanthoseinahigherspecificpowerrange.Interestingly,thisisnottrueforthelowest(<200W/m2)specificpowerturbines;itisunclearwhatisdrivingthisspecificresult.AverageCapacityFactorin2022(projectsbuiltfrom2014to2021)50%SpecificPowerbins(W/m2):40%-224600-222400-220200<20030%≥26020%10%0%MediumHigherHighestLowerEstimatedWindResourceQualityatSiteNote:TheAppendixprovidesdetailsonhowthewindresourcequalityateachindividualprojectsiteisestimated.Sources:EIA,FERC,BerkeleyLabFigure37.Calendaryear2022capacityfactorsbywindresourcequalityandspecificpower:2014–2021projects39Land-BasedWindMarketReport:2023EditionWindpowercurtailmentin2022acrosssevenregionsaveraged5.3%,upfromalowof2.1%in2016Curtailmentofwindprojectoutputresultsfromtransmissioninadequacyandotherformsofgridandgeneratorinflexibilityinconcertwithwindover-supply.Forexample,over-generationcanoccurwhenwindgenerationishighbuttransmissioncapacityisinsufficienttomoveexcessgenerationtootherloadcenters,orthermalgeneratorscannotfeasiblyrampdownanyfurtherorquicklyenough.Thiscanpushlocalwholesalepowerpricesnegative,therebypotentiallytriggeringwindcurtailment,especiallyamongprojectsnotearningthePTC.Curtailmentisgenerallyexpectedtoincreaseaswindenergy’smarketsharegrows,and—asshowninFigure38—thathascertainlybeenthecaseinSPP,wherecurtailmentrosefromjust1.3%in2018to9.2%in2022,atthesametimeasthepercentageofelectricityfromwindexpandedfrom~24%to~38%ofload.Thiscorrelationbetweenmarketshareandcurtailmentdoesnotalwayshold,though.Particularlyinareaswherecurtailmenthasbeenacuteinthepast,stepstakentoaddresstheissuehaveoftenbornefruit.Forexample,Figure38showsthatjust0.5%ofpotentialwindenergygenerationwithinERCOTwascurtailedin2014,downsharplyfrom17%in2009.ThisdeclineinERCOTcurtailmentcorrespondstoasignificantbuild-outofnewtransmissionservingWestTexas,mostofwhichwascompletedbytheendof2013.Since2014,however,wind’smarketsharehascontinuedtoincreaseinERCOT,andsotoohaswindcurtailment,whichhashoveredaround5%forthepastthreeyears.MISO,withthethird-highestwindmarketshare(behindSPPandERCOT),alsohadthethird-highestrateofwindcurtailmentin2022,atjustover4%.WindCurtailmentRateWindPenetration(asa%ofload)20%40%2022Curtailment10%16%8%32%12%WindCurtailment6%8%WindPenetration4%24%2%0%16%PJMCAISOISO-NENYISOMISOERCOTSPP4%8%0%0%2014-222007-222009-222015-222012-222014-222012-222007-22SPPERCOTMISOCAISONYISOISO-NEPJMTotalSources:ERCOT,MISO,CAISO,NYISO,PJM,ISO-NE,SPPFigure38.WindcurtailmentandpenetrationratesbyISOCurtailmentratesintheotherfourISO/RTOregionswererelativelylowin2022:3.2%inNYISO,1.3%inISO-NE,0.5%inCAISO,andatleast0.1%inPJM(thePJMdatashownherelikelyreflectonlyaportionofoverallwindcurtailment,whichtheRTOdoesnotregularlyreport).Theoverallwindpowercurtailmentratein2022acrossallsevenregionswas5.3%,upfromalowof2.1%in2016.40Land-BasedWindMarketReport:2023Edition2022wasanabove-averagewindresourceyearacrossmostofthecountryThestrengthofthewindresourcevariesfromyeartoyear;moreover,thedegreeofinter-annualvariationdiffersfromsitetosite(and,hence,alsoregiontoregion).Thistemporalandspatialvariation,inturn,impactsprojectperformancefromyeartoyear.Figure39showsnationalandregionalindicesofthehistoricalinter-annualvariabilityinthewindresourceamongtheU.S.fleetovertime.34Thoughinter-annualvariationhas,attimes,exceeded+/-20%attheregionallevel(i.e.,0.8and1.2inthegraphic),geographicalaveraginghasenablednationwidevariationtoremainwithin+/-10%.In2022,thenationalwindindexstoodat1.06,itshighestlevelsince2014,asmostregionsexperiencedanabove-averagewindyear(exceptforthenon-ISOWest).AverageAnnualWindResourceIndices(Long-TermAverage=1.0)1.21.11.00.90.8WestSPPMISOERCOTCAISONational0.7PJM2004NYISOISO-NESoutheast20202022200020022006200820102012201420162018Sources:ERA,BerkeleyLab;methodologybehindtheindexofinter-annualvariabilityisexplainedintheAppendixFigure39.Inter-annualvariabilityinthewindresourcebyregionandnationallyWindprojectperformancedegradationalsoexplainswhyolderprojectsdidnotperformaswellin2022Afinalvariablethatcouldinfluencethevariationinproject-levelcapacityfactorsin2022isprojectage.Ifwindturbine(andproject)performancetendstodegradeovertime,thenolderprojects—e.g.,thosebuiltfrom1998to2001—mayhaveperformedworsein2022thanmorerecentprojectssimplyduetotheirrelativeage.Figure40exploresthisquestionbygraphingmedian(and25thto75thpercentileranges)“weather-normalized”(i.e.,correctingforinter-annualvariabilityinthestrengthofthewindresource)capacityfactorsovertime.Here,timeisdefinedasthenumberoffullcalendaryearsaftereachindividualproject’scommercialoperationdate,andeachproject’scapacityfactorisindexedto100%inyeartwotofocussolelyonchangesincapacityfactorovertime,ratherthanonabsolutecapacityfactorvalues.Yeartwoischosenastheindexbase34Theseindicesestimatechangesinthestrengthoftheaverageregion-orfleet-widewindresourcefromyeartoyear(seetheAppendixformoredetails).Notethattheseindicesofinter-annualvariabilitydifferfromtheAWSTruepowerwindresourcequalitydatapresentedelsewhere,inthattheformershowvariabilityfromyeartoyearacrosstheentireregionorfleet,whilethelatterfocusonthemulti-yearlong-termaveragewindresourceatspecificwindprojectsites.41Land-BasedWindMarketReport:2023Editiontoreflecttheinitialproductionramp-upperiodcommonlyexperiencedbywindprojectsastheiroperatorsworkthroughandresolveinitial“teething”issuesduringthefirstyearofoperations.Figure40suggestssomeamountofperformancedecline,especiallyinlateryearsandamongolderprojectsbuiltbefore2008.Projectsbuiltin2008andlaterappear,onaverage,tohaveexperiencedonlyamodestdeclineincapacityfactorduringtheirfirstdecade,followedbyaturnfortheworseinthefewyearsthereafter—perhapsreflectingachangeinhowprojectsareoperatedoncetheyagebeyondthe10-yearPTCwindow.Hamiltonetal.(2020)exploretheseperformancetrendsinmoredepth.Importantly,thewindprojectsampleforFigure40excludesanyprojectsthathavebeenpartiallyrepowered(e.g.,refurbishedwithlongerblades)inrecentyears;theperformanceofsuchprojectstypicallyimprovespost-refurbishment.IndexedCapacityFactor(Year2=100%)Older(pre-2008)projects110%(with25thand75thprecentilerange)100%90%80%70%Newer(post-2007)projects60%(with25thand75thpercentilerange)50%234567891011121314151617181920212223YearofOperationsSources:EIA,FERC,BerkeleyLabFigure40.Changesinproject-levelcapacityfactorsasprojectsageTakentogether,Figure34throughFigure40suggestthat,inordertounderstandtrendsinempiricalcapacityfactors,oneneedstoconsider(andideallycontrolfor)avarietyofparameters.Theseincludenotonlywindpowercurtailmentandtheevolutioninturbinedesign,butalsoavarietyofspatialandtemporalwindresourceconsiderations—suchasthequalityofthewindresourcewhereprojectsarelocated,inter-yearwindresourcevariability,andevenprojectage.42Land-BasedWindMarketReport:2023Edition6CostTrendsWindturbinepricescontinuedtoincreasein2022,reachingroughly$1,000/kWWindturbineprices(in$/kW)havedroppedsince2008,despitecontinuedtechnologicaladvancementsthathaveyieldedincreasesinhubheightsandespeciallyrotordiameters.However,withsupplychainpressuresandelevatedmaterialsprices,turbinepricescontinuedtotrendhigherin2022.Figure41depictswindturbinetransactionpricesfromavarietyofsources:(1)Vestas,SGRE,andNordex,onthosecompanies’globalaverageturbinepricing,asreportedincorporatefinancialreports;(2)BloombergNEF(2022a)andWoodMackenzie(2023a),onthosecompanies’turbinepriceindicesbycontractsigningdate;and(3)121U.S.windturbinetransactionsannouncedfrom1997through2016,aspreviouslycollectedbyBerkeleyLab.Windturbinetransactionscandifferintheservicesincluded(e.g.,whethertowersareprovided,thelengthoftheserviceagreement,etc.),turbinecharacteristics(andthereforeperformance),andthetimingoffutureturbinedelivery.Thesedifferencesdrivesomeoftheobservedintra-yearvariabilityintransactionprices.Mostofthepricesandtransactionsreportedinthefigureareinclusiveoftowersanddeliverytothesite.TurbinePrice(2022$/kW)2,500U.S.turbineorders1,200Vestas2,000BloombergNEFglobalindex800SGRE1,500WoodMackenzieU.S.index400NordexNordexglobalaverage0WoodMacSGREglobalaverageBNEFVestasglobalaverage1,000500019971999200120032005200720092011201320152017201920212023Sources:BerkeleyLab,annualfinancialreports,forecastprovidersFigure41.ReportedwindturbinetransactionpricesovertimeAfterhittinganinitiallowofroughly$1,000/kW,onaverage,from2000to2002,windturbinepricesroughlydoubled,risingtoanaverageofaround$2,000/kWin2008.Thisincreaseinturbinepriceswascausedbyseveralfactors,includingadeclineinthevalueoftheU.S.dollarrelativetotheEuro;increasedmaterials,energy,andlaborinputprices;ageneralincreaseinturbinemanufacturerprofitability;andincreasedcostsforturbinewarrantyprovisions(Monéetal.2017).Windturbinepriceshavedeclinedby50%since2008,inpartreflectingareversalofsomeofthepreviouslymentionedunderlyingtrendsthathadearlierpushedpriceshigheraswellassignificantcost-cuttingmeasuresonthepartofturbineandcomponentsuppliers.Nonetheless,recentsupply-chainpressuresandelevated43Land-BasedWindMarketReport:2023Editioncommoditypriceshaveledtoincreasedturbinepricessince2020.Dataindicatesrecentaveragepricingintherangeof$900/kWto$1,200/kW,alevelroughlysimilartothatlastseenin2017and2018.Surprisingly,averageinstalledprojectcostsamongoursmallsampleof2022projectsdidnotfollowturbinepriceshigherBerkeleyLabalsocompilesavailabledataonthetotalinstalledcostofwindprojectsintheUnitedStates,includingdataon13projectscompletedin2022andtotaling3.3GW—just39%ofthewindpowercapacityinstalledinthatyear.Inaggregate,thedatasetincludes1,206completedwindpowerprojectsinthecontinentalUnitedStatestotaling121.1GWandequaling84%ofallwindpowercapacityinstalledasoftheendof2022.Ingeneral,reportedprojectcostsreflectturbinepurchaseandinstallation,balanceofplant,andanysubstationand/orinterconnectionexpenses.Datasourcesarediverse,however,andarenotallofequalcredibility,soemphasisshouldbeplacedonoveralltrendsinthedataratherthanonindividualproject-levelestimates.AsshowninFigure42,theaverageinstalledcostsofprojectsdeclinedfromthebeginningoftheU.S.windindustryinthe1980sthroughtheearly2000s,andthenincreased—reflectingturbinepricechanges—throughthelatterpartofthatdecadebeforepeakingin2009–2010.Project-levelcostshavesincedeclinedbacktolevelsseenintheearly2000s.Afterfouryearsofrelativestabilityfrom2018to2021,thesurprisingdropinthecapacity-weightedaverageinstalledcostin2022—to$1,370/kW—ispartlyattributabletotheoutsizedinfluenceofasinglelargeprojectthataccountsforalmostone-thirdofthetotalcapacityinourrelativelysmall2022plantsample.Additionally,totalwindcapacityinstallationin2022isdominatedbyprojectsinSPPandERCOT—thetwolowest-costregions.Finally,thesourcesforsomeofourother2022installedcostestimatesdatebackto2020,perhapspre-datinganysubsequentcostincreasesthatmayhaveresultedfromthesupplychainchallengesandinflationarypressuresthathavecharacterizedthelasttwoyears.Itis,therefore,possiblethatthe2022capacity-weightedaveragewillcreepupwardsasmoredatabecomeavailableovertime.InstalledProjectCost(2022$/kW)6,0005,000Capacity-weightedaverage4,000(orangemeans<50%samplesize)3,0002,0001,000019901995200020052010201520201985CommercialOperationDateNote:Areaof“bubble”isproportionaltoprojectcapacitySources:BerkeleyLab,EIA(somedatapointssuppressedtoprotectconfidentiality)Figure42.Installedwindpowerprojectcostsovertime44Land-BasedWindMarketReport:2023EditionRecentinstalledcostsdifferbyregionRegionaldifferencesinaverageprojectcostsarealsoapparentandmayoccurduetovariationsinlaborcosts,developmentcosts,transportationcosts,sitingandpermittingrequirementsandtimeframes,andotherbalance-of-plantandconstructionexpenditures—aswellasvariationsinaverageprojectsizeandtheturbinesdeployedindifferentregions(e.g.,useoflow-wind-speedtechnologyinregionswithlesserwindresources,ortallertowersinareaswithhigherwindshear).Becausesamplesizeforboth2021and2022islimited,Figure43combinesdatafrombothyears.(Evenaftercombiningyears,fiveregions—CAISO,PJM,NYISO,ISO-NE,andtheSoutheast—stilldonothaveenoughsampletowarrantinclusion.)Asshown,thelowest-costprojectsinrecentyearshavebeeninERCOT(averaging$1360/kW)andSPP(averaging$1470/kW).Again,samplesizeinthesetwoyearsisabnormallylow,andtheseaveragesmaychangeasmoredatabecomeavailable.InstalledCostof2021and2022Projects(2022$/kW)2,0001,5001,000500$1,363$1,468$1,511$1,7250SPPWest(non-ISO)MISOERCOTNote:Sizeofbubblereflectsprojectcapacity.Otherregionslackadequatedataforinclusion.Source:BerkeleyLabFigure43.Installedcostof2021and2022windpowerprojectsbyregionInstalledcosts(permegawatt)generallydeclinewithprojectsize;arelowestforprojectsover200MWInstalledcostsexhibiteconomiesofscale,whichisperhapstheprimaryreasonsmallprojectsareincreasinglyrare.Amongasampleofprojectsinstalledin2021and2022(Figure44),thereisnotenoughsamplesizetocalculateaveragecostsforthelowest-capacitybin,buteconomiesofscaleareevidentwhenmovingfromsmallerprojects(5–20MW)tolargerprojects>50MW.45Land-BasedWindMarketReport:2023EditionIndividualprojectsInstalledProjectCost(2022$/kW)3,0002,0001,000Capacity-weightedaverage05-20MW20-50MW50-100MW100-200MW>200MW≤5MWProjectCapacitySource:BerkeleyLabFigure44.Installedwindpowerprojectcostsbyprojectsize:2021and2022projectsOperationsandmaintenancecostsvariedbyprojectageandcommercialoperationsdateOperationsandmaintenance(O&M)costsareakeycomponentoftheoverallcostofwindenergyandcanvaryamongprojects.Unfortunately,publiclyavailabledataonactualproject-levelO&Mcostsarenotwidelyavailable.Evenwheredataareavailable,caremustbetakeninextrapolatinghistoricalO&Mcostsgiventhechangesinwindturbinetechnologythathaveoccurredovertime(seeChapter4).BerkeleyLabhascompiledlimitedO&Mcostdatafor209installedwindpowerprojects,totaling25,083MWandwithcommercialoperationdatesof1982through2021.35ThesedatacoverfacilitiesownedbybothIPPsandutilities,althoughdatasince2004areexclusivelyfromutility-ownedprojectsandsomaynotbebroadlyrepresentative.AfulltimeseriesofO&Mcostdata,byyear,isavailableforonlyasmallnumberofprojects;inallothercases,O&Mdataareavailableforjustasubsetofyearsofprojectoperations.AlthoughnotalldatasourcesclearlydefinewhatitemsareincludedinO&Mcosts,inmostcasesthereportedvaluesincludethecostsofwagesandmaterialsassociatedwithoperatingandmaintainingthewindproject,aswellasrent.36Otherongoingexpenses,includinggeneralandadministrativeexpenses,taxes,propertyinsurance,depreciation,andworkers’compensationinsurancearegenerallynotincluded.Assuch,Figure45andFigure46arenotrepresentativeoftotaloperatingexpensesforwindpowerprojects.Figure45showsO&Mcostsbycommercialoperationdate.Here,eachproject’sO&McostsaredepictedasaverageannualO&Mcostsfrom2000through2022,basedonhowevermanyyearsofdataareavailablefor35Forprojectsinstalledinmultiplephases,thecommercialoperationdateofthelargestphaseisused.Forrepoweredprojects,thedateatwhichrepoweringwascompletedisused.Nodataforprojectsinstalledin2022areincluded,assuchprojectswouldnothaveafullyearofO&Mdataavailablebytheendof2022.36MostoftherecentdataderivefromFERCForm1,whichusestheUniformSystemofAccountstodefinewhatshouldbereportedunder“operatingexpenses”—namely,thoseoperationalcostsassociatedwithsupervisionandengineering,maintenance,rents,andtraining.Thoughnotentirelyclear,theredoesappeartobesomeleewaywithintheUniformSystemofAccountsforprojectownerstocapitalizecertainreplacementcostsforturbinesandturbinecomponentsandreportthemunder“electricplant”accountsratherthanmaintenanceaccounts.46Land-BasedWindMarketReport:2023Editionthatperiod.Forexample,forprojectsthatreachedcommercialoperationin2021,only2022dataareavailable,andthatiswhatisshown.Manyotherprojectsonlyhavedataforasubsetofyears,soeachdatapointinthechartmayrepresentadifferentaveragingperiodwithintheoverall2000–2022period.Thechartshowsthe118projects,totaling21,034MW,forwhich2022O&Mcostdatawereavailable;thoseprojectshaveeitherbeenupdatedoraddedtothechartsincethepreviouseditionofthisreport.AverageAnnualO&MCost,2000−2022(2022$/kW-yr)Projectswithno2022O&Mdata160Projectswith2022O&Mdata120804001984198819921996200020042008201220162020CommercialOperationDateSource:BerkeleyLab;somedatapointssuppressedtoprotectconfidentialityFigure45.AverageO&Mcostsforavailabledatayearsfrom2000to2022,bycommercialoperationdateThedatademonstratethatO&Mcostsarefarfromuniformacrossprojects.Figure45alsosuggeststhatprojectsinstalledinthepastdecadehave,onaverage,incurredlowerO&Mcoststhanthoseinstalledearlier.Specifically,capacity-weightedaverage2000–2022O&Mcostsforthe24projectsinthesampleconstructedinthe1980sequal$72/kW-year,droppingto$60/kW-yearforthe37projectsinstalledinthe1990s,to$31/kW-yearforthe65projectsinstalledinthe2000s,and$20/kW-yearforthe83projectsinstalledsince2010.Thisdeclinemaybeduetoatleasttwofactors:(1)O&Mcostsgenerallyincreaseasturbinesageandcomponentfailuresbecomemorecommon;and(2)projectsinstalledmorerecently,withlargerandmorematureturbinesandmoresophisticatedO&Mpractices,mayexperienceloweroverallO&Mcosts.Limitationsintheunderlyingdatadonotpermittheinfluenceofthesetwofactorstobeclearlydistinguished.Nonetheless,tohelpillustratekeytrends,Figure46showsmedianannualO&Mcostsovertime,basedonprojectage(i.e.,thenumberofyearssincethecommercialoperationdate)andsegmentedintothreeproject-vintagegroupings.Thoughsamplesizeislimited,thedatashowageneralupwardtrendinproject-levelO&Mcostsasprojectsage,atleastamongtheoldestprojectsinthesample.Figure46alsoshowsthatprojectsinstalledoverthelast16yearshavehad,ingeneral,lowerO&Mcoststhanthoseinstalledintheearlieryearsof1998–2005,atleastforthefirst16yearsofoperation.47Land-BasedWindMarketReport:2023EditionMedianAnnualO&MCost(2022$/kW-year)Commercial80OperationDate:601998-20052006-20132014-2021402001234567891011121314151617181920ProjectAge(NumberofYearsSinceCommercialOperationDate)Source:BerkeleyLab;mediansshownonlyforgroupsoftwoormoreprojects,andonlyprojects>5MWareincludedFigure46.MedianannualO&McostsbyprojectageandcommercialoperationdateAsindicatedpreviously,thesedataincludeonlyasubsetoftotaloperatingexpenses.AU.S.windindustrysurveyoftotaloperatingcostsshowsthattheseexpensesforrecentlyinstalledprojectsareanticipatedtoaveragebetween$33/kW-yearand$59/kW-year,withamid-pointof~$44/kW-year(Wiseretal.2019).ThedisparitybetweentheseestimatesoftotaloperatingcostsandthecostsreportedinFigure45andFigure46reflects,inlargepart,differencesinthescopeofexpensesreported;thesurveynotedthatturbineO&Misexpectedtoconstitutelessthanhalfoftotaloperatingcosts—otherongoingexpensesincludepropertytaxes,insurance,assetmanagement,andmore(Wiseretal.2019).48Land-BasedWindMarketReport:2023Edition7PowerSalesPriceandLevelizedCostTrendsWindpowerpurchaseagreementpriceshavebeendriftinghighersinceabout2018,witharecentrangefrombelow$20/MWhtomorethan$40/MWhEarlierchaptersdocumentedtrendsincapacityfactors,installedprojectcosts,O&Mcosts,andprojectfinancing—allofwhicharedeterminantsofthewindpowerpurchaseagreement(PPA)pricesandlevelizedcostofenergy(LCOE)estimatespresentedinthischapter.BerkeleyLabcollectsdataonwindPPAprices,resultinginadatasetthatincludes548PPAstotalingmorethan56GWfromwindprojectsthathaveeitherbeenbuiltorareplannedforinstallationlaterin2023orbeyond.AllofthesePPAsbundletogetherthesaleofelectricity,capacity,andrenewableenergycertificates(RECs;alatertextboxhighlightsRECprices),andmostofthemhaveautilityasthecounterparty.37Exceptwherenoted,PPApricesareexpressedonalevelizedbasisoverthefulltermofeachcontractandarereportedinreal2022dollars.38WheneverindividualPPApricesareaveragedtogether,theaverageisgeneration-weighted.Whenevertheyarebrokenoutbytime,thedateon(oryearin)whichthePPAwasexecutedisused.BecausePPApricesarereducedbythereceiptofstateandfederalincentivesandareinfluencedbyvariouslocalpoliciesandmarketcharacteristics,theydonotdirectlyrepresentwindenergygenerationcosts.Accordingly,attheendofthischapter,thedatapresentedearlierinthisreportareleveragedtoestimateproject-levelandaveragewindLCOEforalargesampleofU.S.windprojects.Figure47plotscontract-levellevelizedwindPPApricesbycontractexecutiondate,showingacleardeclineinPPApricessince2009–2010,bothoverallandbyregion.Asaresultofthelowaverageprojectcostsandhighaveragecapacityfactorsshownearlierinthisreport,ERCOTandSPPtendtobethelowest-pricedregions.Ofnote,PPApriceshavenotsmoothlydeclinedovertime.Instead,pricesdeclinedthrough2003,thenrosethough2009withtheincreasedturbineandinstalledcostspresentedearlieraswellaswithgeneralpriceincreasesduringthisperiodinthepowerandnaturalgasmarkets.Followingthatrisewasasteepreductionand,morerecently,stabilizationandthenanincreaseinPPAprices—partlyduetosupplychainpressures,includinghighermaterialpricesandtransportationcosts.Thesesamesupplychainandinflationarypressuresmayhaveledtosomerenegotiationsofpreviouslyagreed-uponPPApricesamongplantsnotyetbuilt.37ThoughsomePPAswithcorporateofftakersareincludedinthesample,inmanycasessuchPPAsaresyntheticorfinancialarrangementsinwhichtheprojectsponsorentersa“contractfordifferences”withthecorporateofftakeraroundanagreed-uponstrikeprice.Becausethestrikepriceisnotdirectlylinkedtothesaleofelectricity,itisrarelydisclosed(atleastthroughtraditionalsources,likeregulatoryfilings).DatafromLevelTenEnergypresentedlaterinthischapter,however,shedsmorelightontrendsincorporatePPAprices.38Havingfull-termpricedata(i.e.,pricingdataforthefulldurationofeachPPA,ratherthanjusthistoricalPPAprices)enablesthesePPApricestobepresentedonalevelizedbasis(levelizedoverthefullcontractterm),whichprovidesacompletepictureofwindpowerpricing(e.g.,bycapturinganyescalationoverthedurationofthecontract).Contracttermsrangefrom5to35years,with20yearsbeingbyfarthemostcommon(at54%ofthesample;87%ofcontractsinthesamplearefortermsrangingfrom15to25years).Pricesarelevelizedusinga4%realdiscountrate.49Land-BasedWindMarketReport:2023EditionLevelizedPPAPrice(2022$/MWh)CAISOWestMISO140SPPERCOTPJMNYISOSoutheastISO-NE12010080604020020062008201020122014201620182020202219961998200020022004PPAExecutionDateNote:Sizeofbubblereflectscontractcapacity.Source:BerkeleyLab,FERCFigure47.LevelizedwindPPApricesbyPPAexecutiondateandregion(fullsample)Figure48providesasmootherlookatthetimetrendnationwideandregionallybyaveragingtheindividuallevelizedPPApricesshowninFigure47,andconsolidatingtheregionalbreakdownintojustthreecategories:West,Central,andEast.Aftertoppingoutnear$80/MWhforPPAsexecutedin2009,thenationalaveragelevelizedpriceofwindPPAswithintheBerkeleyLabsampledroppedtobelow$20/MWhforPPAsexecutedin2018.Sincethen,priceshavebeendriftinghigher.Thoughoursamplesizeinthelastyearortwoissmall,recentpricingappearstobearound$20/MWhintheCentralregionofthecountry,abithigherintheWest(rangingfrom$20-$40/MWh),andhigherstillintheEast(~$50/MWh).50Land-BasedWindMarketReport:2023EditionAverageLevelizedPPAPrice(2022$/MWh)West100East80604020CentralNationwide01996-992000-012002-032004-052006200720082009201020112012201320142015201620172018201920202021-22PPAExecutionYearNote:West=CAISO,West(non-ISO);Central=MISO,SPP,ERCOT;East=PJM,NYISO,ISO-NE,Southeast(non-ISO)Source:BerkeleyLab,FERCFigure48.Generation-weightedaveragelevelizedwindPPApricesbyPPAexecutiondateandregionLevelTenEnergy’sPPApriceindicesconfirmrisingPPApricesandregionalvariationIncontrasttothePPAssummarizedabove,whichprincipallyinvolveutilitypurchasers,LevelTenEnergy(2023)providesanindexofwindPPAoffersmadetolarge,end-usecustomers.Eachquarter,theLevelTenEnergyPPAPriceIndexreportsthepricesthatwindandsolardevelopershaveofferedforPPAsavailableontheLevelTenMarketplace.Contracttermstendtorangefrom10to15years,reflectiveoftheshortertermstypicallypursuedbyend-usecustomersthatpurchasewindenergyrelativetotheutilityPPAssummarizedearlier.Pricedataareaggregatedandreportedinnominaldollarsona‘P25’basis,referringtothemostcompetitive25thpercentileofofferprices.AsshowninFigure49,priceshaverisenoverthelastcoupleyears,andvarybyISO;here,LevelTendataareconvertedtoreal,levelized2022$toenhancecomparabilitywithdatapresentedelsewhereinthisreport.Amongregionsreportingdata,CAISOfeaturesthehighestwindPPApricing(~$60/MWhinthethirdquarterof2022whenconvertedtolevelizedrealdollarterms),whereasthelowestpricesareinSPPandERCOT(~$33/MWhinthesecondquarterof2023).Inrealdollarterms,LevelTen’sreportedpricetrendssince2018arebroadlysimilartothosedescribedinthepriorsection.51Land-BasedWindMarketReport:2023EditionLevel10PPAPriceIndex(2022$/MWh,25thpercentileofoffers)CAISO70PJMMISO60ERCOT50SPP403020100Q3Q4Q1Q2Q3Q4Q1Q2Q3Q4Q1Q2Q3Q4Q1Q2Q3Q4Q1Q2201820192020202120222023Source:LevelTenEnergyFigure49.LevelTenEnergywindPPApriceindexbyquarterofofferAmongarelativelysmallsampleofprojectsbuiltin2022,the(unsubsidized)averagelevelizedcostofwindenergyhasfallentoaround$32/MWhInacompetitivemarket,long-termPPApricescanbethoughtofasreflectingtheLCOEreducedbythevalueofanyincentivesreceived(e.g.,thePTC).Hence,asafirst-orderapproximation,LCOEcanbeestimatedsimplybyaddingthelevelizedvalueofincentivesreceivedtothelevelizedPPAprices.LCOEcanalsobeestimatedmoredirectlyfromitscomponents,andBerkeleyLabhasdataonboththeinstalledcostandcapacityfactorof120GWofwindpowerprojectsinstalledfrom1998through2022,representing83%ofallcapacitybuiltoverthatperiod.Here,thosedataareused,inconjunctionwithestimatesofoperationalcosts,financingcosts,projectlifeandotherfactors,toestimateLCOEinreal2022dollars(seetheAppendixfordetailsonthedataandcalculations).Onebenefitofthis“bottomup”approachtoestimatingLCOEisthatitreliesonalargesampleofproject-levelinstalledcostandperformancedata,coveringmoreprojectsthanthePPAsample.Figure50depictstheresultingaverageLCOEvaluesovertimeonanationalbasis.Asshown,averagewindLCOEdeclinedfrom$114/MWhin1988−1999to$76/MWhin2004−2005,beforerisingto>$100/MWhin2009-2011.Subsequently,averageLCOEdeclinedrapidlythrough2018,to$36/MWh.ThenationalaverageLCOEofnewlybuiltwindprojectshaslargelyheldsteadysince2018,butdeclinedto$32/MWhamongarelativelysmallsampleof2022plants.Thedeclinein2022isdue,inpart,tothestrongconcentrationof2022projectsinSPPandERCOT,bothlow-costandhigh-resourceregions.Itisalsoinfluencedbyasingle,verylargeprojectthatcameonlinein2022,whichhasasignificantimpactontheaveragevalue.Finally,asnotedearlier,theprojectsampleforwhichdataareavailableislimitedin2021and2022.Asmoredatabecomeavailableovertime,theestimatedaverageLCOEfor2021-and2022-vintageplantscouldchange.52Land-BasedWindMarketReport:2023EditionAverageandPlant-LevelLCOE(2022$/MWh)160140OrangeshadingmeansthatLCOEcanbeestimatedfor<50%ofprojectsinyear.12010080604020114969076898998105102102846556514644363837383201998-992000-012002-032004-0520062007200820092010201120122013201420152016201720182019202020212022CommercialOperationYearNote:Sizeofbubblereflectsprojectcapacity.Source:BerkeleyLabFigure50.EstimatedlevelizedcostofwindenergybycommercialoperationdateLevelizedcostsvarybyregion,withthelowestcostsinSPPandERCOTBecauseofthesmallsamplesizeamong2021and2022windplants,Figure51combinesbothyears(andeventhenonlyhasenoughdatatoshowfiveofthenineregions).ThelowestaverageLCOEsforprojectsbuiltin2021and2022—onlyconsideringregionswithatleasttwoplantsinthesample—arefoundinSPPandERCOT(both~$33/MWhonaverage),withPJMaveragingthehighestat$46/MWh.LCOEof2021and2022Plants(2022$/MWh)70605040302010$32.7$33.3$39.3$40.3$46.10ERCOTMISOWest(non-ISO)PJMSPPNote:Sizeofbubblereflectsprojectcapacity.Someindividualoutliersmaybeexcluded.Otherregionslackadequatedataforinclusion.Source:BerkeleyLabFigure51.Estimatedlevelizedcostofwindenergy,byregion53Land-BasedWindMarketReport:2023EditionRenewableEnergyCertificate(REC)PricesWindpowersalespricespresentedinthisreportreflectbundledsalesofbothelectricityandRECs.ProjectsthatsellRECsseparatelyfromelectricity,therebygeneratingtwosourcesofrevenue,areexcluded.RECmarketsarefragmented,butconsistoftwodistinctsegments:compliancemarkets,inwhichRECsarepurchasedtomeetstateRPSobligations,andgreenpowermarkets,inwhichRECsarepurchasedonavoluntarybasis.MandatoryRPSprogramsexistin29statesandWashington,D.C.Inrecentyears,roughlyone-thirdofthesestateshaveincreasedtheirRPStargets,inmanycasestolevelsrangingfrom50%to100%ofretailelectricitysales.Voluntarymarketsforrenewableenergyhavealsogrown.Thefigurebelowpresentsindicativedataofspot-marketRECpricesinbothcomplianceandvoluntarymarkets.SpotRECpriceshavevaried,bothovertimeandacrossstates,thoughpricesacrossstateswithincommonregionalpowermarkets(NewEnglandandPJM)arelinkedtovaryingdegrees(consequently,severalofthelinesinthefigureoverlap).InNewEngland,RECpricesin2022(outsideofME)fellmodestlyfrom$40/MWhatthebeginningoftheyeartoroughly$35/MWhbyyear-end.Thesepricesarejustbelowthealternativecompliancepaymentratesinthesestates,suggestingatightbutsufficientRPSsupply.InPJM,RECpricesinmanystatescontinuedtheirupwardtrajectoryfromthepastseveralyears,reflectingagradualtighteningofsupplies.WithinthepremiummarketsofDE,NJ,andPA,pricesmovedtogether,andallendedtheyearatnearly$30/MWh,anall-timehighforthosestates.PricesforRECsofferedinthenationalvoluntarymarketandforRPScomplianceinTexas,whichtrackeachothercloselyandarewellbelowRECpricesinmostcompliancemarkets,felltoroughly$2/MWhover2022,followingtheirspiketheyearbefore.Notes:Dataforcompliancemarketsfocuson“ClassI”or“TierI”RPSrequirements;thesearetherequirementsformore-preferredresourcetypesorvintagesandarethereforethemarketsinwhichwindwouldtypicallyparticipate.PlottedvaluesarethemonthlyaveragesofdailyclosingpricesforRECvintagesfromthecurrentornearestfutureyeartraded.RECpricestradeatsimilarlevelsinanumberofmarketssuchthatsomeofthelinesintheabovegraphicoverlap.Source:MarexSpectron54Land-BasedWindMarketReport:2023Edition8CostandValueComparisonsDespiterelativelylowPPAprices,windfacescompetitionfromsolarandgasFigure52plotswindPPApricesagainstutility-scalesolarPPApricesonalevelizedbasissince2009(theblueandgoldlinesshowthegeneration-weightedaveragewindandsolarPPApricesineachyear,respectively).AlthoughthegapbetweenwindandsolarPPApriceswasquitewideadecadeago,thatgaphasnarrowed,assolarpricesfellmorerapidlythanwindprices.39ThefigurealsoshowsthatwindPPAprices—and,morerecently,utility-scalesolarPPAprices—have,inmanycases,beencompetitivewiththeprojectedfuelcostsofgas-firedcombinedcyclegenerators.Specifically,theblackdashmarkersshowthe20-yearlevelizedfuelcosts—convertedfromnaturalgastopowertermsatanassumedheatrateof7.5millionBritishThermalUnits(MMBtu)perMWh—fromthen-currentEIAprojectionsofnaturalgaspricesdeliveredtoelectricitygenerators.40Supportedbyfederaltaxincentives,theaveragelevelizedwindandsolarPPApriceswithinthiscontractsamplehave,forseveralyearsnow,beenbelowtheprojectedlevelizedcostofburningnaturalgasinexistinggas-firedcombinedcycleunits.LevelizedPPAandGasPrice(2022$/MWh)180PVPPAprices160140120Levelized20-yearEIAgaspriceprojections1008060402020132014201520162017201820192020202120222023WindPPApricesPPAExecutionDateandGasProjectionYear02009201020112012Note:Smallestbubblesizesreflectsmallest-volumePPAs(<5MW),whereaslargestreflectlargest-volumePPAs(400MW)Sources:BerkeleyLab,FERC,EIAFigure52.LevelizedwindandsolarPPApricesandlevelizedgaspriceprojectionsRatherthanlevelizingthewindPPApricesandgaspriceprojections,Figure53plotsthefuturestreamofwindPPAprices(the10th,50th,and90thpercentilepricesareshown)fromPPAsexecutedin2020–2022againsttheEIA’slatestprojectionsofjustthefuelcostsofnaturalgas-firedgeneration.41Asshown,the10th-90th39ThesolarPPApricesaresourcedfromBerkeleyLab’s“Utility-ScaleSolar”dataseries.40Forexample,theblackdashmarkerin2009showsthe20-yearlevelizedgaspriceprojectionfromAnnualEnergyOutlook2009,whiletheblackdashin2023showsthesamefromAnnualEnergyOutlook2023(bothconvertedto$/MWhtermsataconstantheatrateof7.5MMBtu/MWh).41ThefuelcostprojectionscomefromtheEIA’sAnnualEnergyOutlook2023publication.Theupperandlowerboundsofthefuelcostrangereflectthelow(andhigh,respectively)oilandgasresourceandtechnologycases.Allfuelpricesareconvertedfrom55Land-BasedWindMarketReport:2023Editionpercentilerangeofwindpricesisquitewide,dueinparttotherelativelysmallsampleof27contracts.ThemedianwindPPApricehoversjustbelow$30/MWhthrough2040,andovermostofthatperiodfallssquarelywithintherangeoffuelcostprojections.2022$/MWh6050MedianwindPPAprice(with10thand90thpercentilerange)4030GasWind2010AEO2023referencecasenaturalgasfuelcostprojection(withfullrangeofprojections)020232025203020352040Sources:BerkeleyLab,FERC,EIAFigure53.WindPPApricesandnaturalgasfuelcostprojectionsbycalendaryearovertimeFigure53alsohintsatthelong-termvaluethatwindpowermightprovideasa“hedge”againstrisingand/oruncertainnaturalgasprices.ThewindPPApricesthatareshownhavebeencontractuallylockedin,whereasthefuelcostprojectionstowhichtheyarecomparedarehighlyuncertain.Actualfuelcostscouldbelowerorhigher.Eitherway,asevidencedbythewideningrangeoffuelcostprojectionsovertime,itbecomesincreasinglydifficulttoforecastfuelcostswithanyaccuracyasthetermoftheforecastincreases.Thegrid-systemmarketvalueofwindsurgedin2022acrossmanyregionsandwasoftenhigherthanrecentwindPPApricesInmanyregionsofthecountry,windprojectsparticipateinorganizedwholesaleelectricitymarkets.Insomecases,windprojectsdirectlybidintothosemarkets,andearntheprevailingmarketprice.Inothercases—especiallywhenaPPAisinplace—thewindpurchaserwillschedulethewindenergyintothemarket,payingthewindprojectownerthepre-negotiatedPPApricebutearningrevenuefromtheprevailingwholesalemarketprice.PPAsbetweenwindgeneratorsandcommercialcustomersareoftenahybridofthesetwomodels.Inallthesecases,therevenueearned(orthatcouldhavebeenearned)fromthesaleofwindintowholesalemarketsisreflectiveofthemarketvalueofthatgenerationfromtheperspectiveoftheelectricitysystem.Inthecaseofmerchantwindprojects,thelinkisdirectandaffectstherevenueoftheplant.InthecaseofwindprojectssoldunderaPPA,ontheotherhand,thepre-negotiatedPPApriceestablishesplantrevenueand,dependingonthespecificsofthePPA,pricingmayormaynotbelinkedtowholesalemarketprices.Inthislattercase,however,therevenueearnedorthatwouldhavebeenearnedbythesaleofwindinthewholesalemarketstillreflectstheunderlyingmarketvalueofthatwind—butinthisinstance,forthepurchaser,inthe$/MMBtuinto$/MWhusingtheheatratesimpliedbythemodelingoutput(whichstartat7.6MMBtu/MWhin2023andrangefrom7.5-8.0MMBtu/MWhin2040,dependingonthescenario).56Land-BasedWindMarketReport:2023Editionformofanavoidedcost.Thisisbecausewholesaleelectricitypricesreflectthetimingofwhenenergyischeaporexpensiveandembedthecostoftransmissioncongestionandlosses.Apurchasercould,intheory,obtainpowerfromthewholesalemarketinsteadoffromawindproject.Awindproject’sestimatedrevenueparticipatinginthewholesalemarketthereforereflectscostsavoidedbythepurchaserofwindunderaPPA.This(potential)revenue—orvalue—canbesegmentedinto“energy”marketvalueand,wherecapacitymarketsorrequirementsexist,“capacity”value.Wholesaleenergypricesvaryovertimeandbylocation.Theyarestronglyinfluencedbythecostofnaturalgas.Becausewindpowerdeploymentissometimesconcentratedinareaswithlimitedtransmissioncapacity,wholesaleenergypricesatthelocalpricingnodestowhichwindplantsinterconnectareoftensuppressedandtherelationshiptothecostofnaturalgasisdiminished.Evenabsenttransmissionconstraints,windplantspushwholesaleenergypriceslowerwhenwindoutputishigh.Moregenerally,thetemporalprofileofwindoutputisnotalwayswell-alignedwithcustomerloadandsystemneeds,potentiallyfurtherreducingtheenergymarketvalueofwindgeneration.Someofthesetendenciesalsoapplytowind’scapacityvalue,whichisimpactedbythecostofcapacitybutalsobyregionalrulesthatdefinethecreditthatwindreceivesforprovidingcapacity.Figure54showstheestimatedhistoricalwholesaleenergyandcapacitymarketvalueofwindacrossdifferentregionsofthecountry.Specifically,theenergymarketvalueofwindisestimatedusingplant-levelhourlywindoutputprofilesandreal-timehourlywholesaleenergypricingpatternsatthenearestpricingnode(i.e.,locationalmarginalprices,LMPs).Plant-levelcapacityvaluesareestimatedbasedontherelevantcapacitypriceorcostfortheregioninquestion,andlocalrulesforwind’scapacitycredit.42Energyandcapacityvaluesaresummedforeachplant,andplant-leveltotalvalueestimatesarethenaveragedtoestimateregionalvalues.Asaresult,theanalysisconsiderstheoutputprofileofwind,thelocationofwind,andhowthosecharacteristicsinteractwithlocalwholesaleenergyandcapacitypricesandrules,yieldinganestimateoftherevenuethatwouldhavebeenearnedhadwindsolditsoutputatthehourlyLMPandconsideringanypossiblecapacity-basedrevenue.Thefigurethencontraststhosewholesalemarketvalueestimatesforwindwithnationwidegeneration-weightedaveragelevelizedwindPPAprices(witherrorbarsdenotingthe10thand90thpercentiles)basedontheyearsinwhichthePPAswereexecuted.ThecomparisonbetweenmarketvalueestimatesandPPApricesisrelevantinasmuchasPPApricesreflectthecostofwindtothepurchaser,whereaswholesalemarketvaluereflectsaportionofthevalueofthatwindgeneration.Theseestimatesshowthatthewholesalemarketvalueofwindvariesstronglybyregion.Themarketvalueofwindgenerallydeclinedthrough2020buthasincreasedsince.Withthesharpdropinwholesalepricesandthereforemarketvalueofwindin2009,averagewindPPApricestendedtowellexceedthewholesalemarketvalueofwindfrom2009to2012.WithcontinueddeclinesinPPAprices,however,thosepricesreconnectedwiththemarketvalueofwindin2013andhaveremainedincompetitiveterritoryinsubsequentyears.Thissuggeststhat—withthehelpofthePTC,whichreducesPPAprices—winddevelopersandofftakersaresuccessfullycontractingatlevelsthataregenerallycomparableintermsofbothcostandvalue.In2020,naturalgasandwholesaleelectricitypriceshitnewlows,inpartbecauseoftheeconomicimpactsofthepandemic.Naturalgaspricesthenrosein2021andagainin2022;in2022,annualaveragenaturalgaspriceswerehigherthaninanyyearsince2008(inrealdollarterms,basedontheHenryHubspotprice).Withtheincreaseinnaturalgasandelectricityprices,2022windmarketvaluesrosetolevelslastseenin2014inseveralregionsandarelargerthanrecentPPApricesinmanylocations.However,thehighmarketvaluesforwindmayeasein2023asnaturalgaspriceshavedeclineddramaticallyfrom2022’shighlevels.42TheAppendixprovidesadditionaldetailsonthemethodsusedtoestimatethewholesaleenergyandcapacityvalueofwind.57Land-BasedWindMarketReport:2023EditionNote:Hourlywindoutputprofilesandwholesalepricesarenotavailableforallhistoricalyearsforallregions.Sources:BerkeleyLab,Hitachi,ISOsFigure54.Regionalwholesalemarketvalueofwindandaveragelevelizedlong-termwindPPApricesovertimeImportantNoteonPriceandValueComparisonsNotwithstandingthecomparisonsmadeinthischapter,neitherthewindpricesnorwholesalemarketvalueestimates(norfuelcostprojections)reflectthefullsocialcostsofpowergenerationanddelivery.Amongthevariousshortcomingsofcomparingwind(andsolar)PPApriceswithwholesalevalueandnatural-gascostestimatesinthismannerarethefollowing:•Wind(andsolar)PPApricesarereducedbyfederalandstateincentives.Similarly,wholesaleelectricityprices(orfuelcostprojections)arereducedbyanyfinancialincentivesprovidedtothermalgenerationanditsfuelproduction.Wholesalepricesmayalsonotfullyaccountforthehealthandenvironmentalcostsofvariousgenerationtechnologies(thoughalatersectionwithinthischapterassessesthehealthandclimatebenefitsofwind),andforothersocietalconcernssuchasfueldiversityandresilience.•Wind(andsolar)PPApricesdonotfullyreflectintegration,resourceadequacy,ortransmissioncosts,whilewholesaleelectricityprices(orfuelcostprojections)alsodonotfullyreflecttransmissioncostsandmaynotfullyreflectcapitalandfixedoperatingcosts.•WindandsolarPPAprices—onceestablished—arefixedandknown.Theestimatedwholesalemarketvalueofwindrepresentshistoricalvalueswhereasfuturenaturalgaspricesareuncertain.Saidanotherway,levelizedwind(andsolar)PPApricesrepresentafuturestreamofpricesthathasbeenlockedin(andthatoftenextendsfor20yearsorlonger),whereasthewholesalevalueestimatesarepertinenttojustthespecifichistoricalyearsevaluated,andfuturenaturalgaspricesreflectuncertainforecasts.Inshort,comparinglevelizedlong-termwindPPApriceswitheitheryearlyestimatesofthewholesalemarketvalueofwindorforecastsofthefuelcostsofnaturalgas-firedgenerationisnotappropriateifone’sgoalistoaccountfullyforthecostsandbenefitsofwindenergyrelativetoothergenerationsources.Nonetheless,thesecomparisonsstillprovidesomesensefortheshort-termcompetitiveenvironmentfacingwindenergyandconveyhowthoseconditionshaveshiftedovertime.58Land-BasedWindMarketReport:2023EditionThegrid-systemmarketvalueofwindin2022variedstronglybyprojectlocation,fromanaverageof$18/MWhinSPPto$83/MWhinISO-NEFigure55presentsestimatesofwind’swholesalemarketvalue,byregion,butonlyforthelatestyear—2022.Thefigurealsodisaggregatesthemarketvalueestimatesintotheirconstituentparts:energyandcapacity.Infiveofthesevenregionsshown(ERCOTandSPPexcepted),wholesalemarketvaluewassignificantlyhigherin2022thanithadbeenin2021,thankstohigherenergyvaluethatwasdrivenbyhigherwholesaleelectricitypricesingeneral.Higher-valuemarketswereISO-NE($83/MWh),CAISO($76/MWh),PJM($58/MWh),andNYISO($45/MWh).Theaveragemarketvalueofwindin2022wasthelowestinSPP($18/MWh).WindmarketvalueinMISO($29/MWh)andERCOT($29/MWh)fellinthemiddle.Inallregions,energyvaluerepresentedthelargestshareofthetotalvalue,withcapacityvaluevaryingwidelyregionallyandbeinglowerinabsolutemagnitude.Sources:BerkeleyLab,Hitachi,ISOsFigure55.Regionalwholesalemarketvalueofwindin2022,byregionFigure56presentsthe2022windpowermarketvalueestimatesataprojectlevel.Theseestimatesspanawiderangein2022,withthe10th,50th,and90thpercentilevaluesequaling$12,$31,and$77perMWh,respectively.Thefigureshowsvariabilityinmarketvaluewithineachregion,especiallyinMISO,SPP,andERCOT,withareasfacingtransmissioncongestionandhighwindpenetrationsgenerallyexperiencinglowermarketvalue.Highermarketvalueestimatesarefoundinuncongestedareas,areaswithhigheraveragewholesaleprices,andareaswherewindoutputprofilesaremorecorrelatedwithelectricitydemand.(Developmentsrelatedtonewtransmissionandwindenergyarediscussedinanaccompanyingtextbox).59Land-BasedWindMarketReport:2023EditionSources:BerkeleyLab,Hitachi,ISOsFigure56.Project-levelwholesalemarketvalueofwindin2022Thegrid-systemmarketvalueofwindtendstodeclinewithwindpenetration,impactedbygenerationprofile,transmissioncongestion,andcurtailmentTheregionswiththehighestwindpenetrations(SPPat38%,ERCOTat25%,andMISOat14%)haveexperiencedthelargestreductioninwind’svaluerelativetotheregionalaveragevalueofa24x7flat-profilegenerator.The“valuefactor”ofwindgenerationin2022wasroughly0.4,0.5,and0.5ineachofthesehigh-penetrationregions,respectively.Valuefactoriscalculatedseparatelyineachregionandrepresentstheratiooftheaveragevalueofwindgenerationtotheaveragevalueofa24x7flatprofileatallgeneratorlocations.The2022windvaluefactorinNYISOwas0.6butwashigherinISO-NE(0.9),CAISO(0.8),andPJM(0.8).Theprogressionofeachregion’svaluefactorwithwindpenetrationcanbeseeninFigure57.Whilethereisaloosecorrelationbetweenpenetrationlevelandvaluefactor,eachregion’svaluefactorprogressedalongaconvolutedpathaspenetrationincreased.Millsteinetal.(2021)showthatdifferencesbetweentheregions’transmissioninfrastructure,andupgradestothatinfrastructure,areoneoftheprimaryreasonsvaluefactorsdonotcorrelatemorecloselywithpenetrationlevel.AninterestingfeatureisthepathofwindvaluefactorinERCOT,whichstartedatcloseto0.5butincreasedwiththecompletionoftheCompetitiveRenewableEnergyZone(CREZ)transmissionlinesandthendeclinedovertimewithcontinuingwindpenetration.In2021,thevaluefactordroppedto0.2duetoconditionsassociatedwithextremeweather,butthenreboundedin2022to0.5.60Land-BasedWindMarketReport:2023EditionSources:BerkeleyLab,Hitachi,ISOsFigure57.TrendsinwindvaluefactoraswindpenetrationsincreaseUsingmethodsfurtherdescribedinMillsteinetal.(2021),Figure58showstheimpactofthreeseparatecausesofreductiontothevalueofwindgenerationin2022.Asusedhere,thetermvaluereductionistheoppositeofvaluefactor:atotalvaluereductionof40%wouldindicateavaluefactorof0.6.Thethreecausesofvaluereductionare:(1)profilevaluereductions:causedbythetemporalcorrelationofwindgenerationwithlowmarketprices,(2)congestionvaluereductions:causedbytheinabilitytoservethemostvaluablelocationsinaregionduetotransmissioncongestion,and(3)curtailmentvaluereductions:causedbycurtailmentofoutput,typicallyduetowindplantoperatorresponsetolow(usuallynegative)localprices.Thecausesofwindvaluereductionsvarybyregion.SPPandERCOTvaluereductionsin2022weresplitbetweenprofile-basedvaluereductionsandcongestionvaluereductions.InSPPandERCOT,2022profilevaluereductionswere34%and25%,respectively,alittlelargerthanthe24%valuereductionfromcongestionseeninbothregions.MISOandNYISOfacedlargecongestionvaluereductionsin2022of37%and32%,respectively.Curtailmentvaluereductionsdidnotreachabove3%inanyregion.The2022profileofwindoutputinISO-NEwasmildlymorevaluablethanaflatoutputprofile,providingasmallvalueboostof2%versusaflatprofile(butthisbenefitwascanceledoutbythecongestionvaluereductionof7%intheregion).Thevaluereductionsassociatedwithcongestioncouldpotentiallybeaddressedwithnewwithin-regiontransmissioninfrastructure.Conversely,mitigatingtheprofilevaluereductionssuchasthosefoundinSPPandERCOTin2022wouldrequirestrategiesbeyondexpansionofwithin-regionaltransmission.Millsteinetal.(2021)discussesarangeofstrategiestoaddressprofilevaluereductions,includingcross-regionaltransmissionandstoragedeployment,newdemandsources(e.g.,coordinatedelectricvehiclecharging),andregulatoryandratechangessupportingresponsiveload.Kempetal.(2023)furtherexploretherelativevaluetowind(andsolar)plantsofaddingenergystorageversusthevalueoflocaltransmissionexpansion,findingthatthevalueofincreasedregionaltransmissionislargerforwindplantsthanforsolarplants,butthatbothtypesofplantsseesimilarproportionalvalueincreasesforaddingenergystorage.61Land-BasedWindMarketReport:2023EditionSources:BerkeleyLab,Hitachi,ISOsNote:InISO-NE,thetemporalprofileofwindprovidesaslightpremiumvalueoveraflatoutputprofile(+2%).Thecolorshowsastealbecausethenegativecongestionpenalty(-7%)islayeredontopofthepositiveprofilepremium.Figure58.Impactoftransmissioncongestion,outputprofile,andcurtailmentonwindenergymarketvaluein2022Thehealthandclimatebenefitsofwindarelargerthanitsgrid-systemvalue,andthecombinationofallthreefarexceedsthelevelizedcostofwindThebenefitsofwindinreducinghealthandclimateburdensfrompollutingenergysourcesarenotincludedintheearlierestimatesofgrid-systemvalueandthecomparisonsofthatvaluewithPPAprices.Windgenerationreducespower-sectoremissionsofcarbondioxide(CO2),nitrogenoxides(NOx),andsulfurdioxide(SO2).Thesereductions,inturn,providepublichealthandclimatebenefits(Millsteinetal.2017).Inthissection,thehealthandclimatebenefitsofwindpowerareestimatedandcompared,alongwithgrid-systemvalue,totheunsubsidizedlevelizedcostofnewwindplantsbuiltin2022.43UsingmethodsdescribedintheAppendix,44Figure59presentsthehealthandclimatebenefitsfromwindbyregionintheyear2022,consideringalmostallwindplantsinthecontiguousUnitedStates.Notethatthevaluescalculatedherearebasedonamethodologythatiscurrentlyundergoingpeerreview;itisanticipatedthatvaluespublishedafterpeer-reviewmayvaryfromthesereportedvaluesbutthattheoverallconclusionsof43Thegoalwastocomparethemostimportantcostandbenefitcomponentsfromasocietalperspective,butthiscomparisonisnotexhaustive.Notincludedareconsiderationsofemployment;localenvironmental,ecological,land-use,andcommunityimpacts;wateruse;mercuryandprimaryparticulatematter;andtransmissionorgrid-integrationcostsnotcoveredbygrid-valueestimates.44Briefly,theper-MWhhealthandclimatebenefitsofwindwereestimatedthroughatwo-stepprocess:first,determinethemarginalavoidedemissionrate;second,multiplyavoidedemissionsbyaregionaldamagerate(i.e.,healthorclimateimpactspertonofpollutantemitted).Marginalavoidedemissionratesarederivedusinganapproachbasedon,butupdatedfrom,FellandJohnson(2021).DamageratesforCO2emissionsaresettoequalthesocialcostofcarbon(Rennertetal.(2022);2.0%discountrate),andhealthdamageratesforSO2andNOxcomefromEPA(2023)andmodelscompiledinCACES(2023),InMAP(Tessemetal.2017),EASIUR(Heoetal.2016),andAP2(Muller2014).Healthdamageratesvarybytheregioninwhichtheemissionsoccurred.62Land-BasedWindMarketReport:2023Editiontheanalysisareunlikelytochange.Nationally,healthandclimatebenefitstogetheraveraged$135/MWh-wind;thisestimateisupsharplyfromlastyear’sestimateof$80/MWhin2021,dueinlargeparttoanupwardrevisioninthesocialcostofcarbon,basedonRennertetal.(2022).BenefitswerelargestintheCentral($200/MWh),Midwest($133/MWh),Texas($111/MWh),andWestern($109/MWh)regions.ValueswerelowestinNewYork($58/MWh),NewEngland($83/MWh),andtheMid-Atlantic($89/MWh).Inthehighestvalueregions,windoffsetsmore-pollutingpowerplantsthaninotherregions.HealthandclimatebenefitsarenotreportedintheSoutheastduetothesmallnumberofwindplantsinthatregion.Regionalandnationalvaluespresentedhereincludebothin-regionemissionimpactsaswellascross-regionimpactsduetoelectricitytradeacrossregionalboundaries.Californiaandthenorthwestandsouthwestregionswerecombinedtoasinglelargeregionduetothemagnitudeoftradeacrossthoselocations.Note:EstimatesnotprovidedforSoutheastduetosmallnumberofwindplantsinthatregion.Sources:BerkeleyLab,FormEIA-930Figure59.Marginalhealthandclimatebenefitsfromallwindgenerationbyregionin2022Thenationalaverageclimate,health,andgrid-systemvaluesumstofivetimestheaverageLCOEofwindplantsthatcameonlinein2022(seeFigure60).Onecaveathereisthateachnationalestimateisbasedonaslightlydifferentregionalweightingofplants–LCOEbasedonasetofrecentplants,healthandclimatebenefitsbasedontheaveragenationalvaluefromallplants,andgrid-systemvaluebasedonallplantsinthesevenISO/RTOs.ThesedifferencesarenotlargeenoughtomeaningfullyimpactthesizabledisparitybetweentheLCOEandvalueestimates.63Land-BasedWindMarketReport:2023EditionCostsandBenefits(2022$/MWh)180160140120Climate1008060Health4020LCOEGridValue0CostSources:BerkeleyLab,EIAForm930Figure60.Marginalhealth,climate,andgrid-valuebenefitsfromnewwindplantsversusLCOEin2022Forsimplicity,singlevaluesforhealthandclimatebenefitsarepresented.However,thesevaluesrepresentcentralestimatesfromarangeofplausiblevalues.Thecentralhealthandclimatevaluespresentedherearederivedfrommethodsdetailedintheappendixconsideringnumerousuncertainties.Lowandhighnationalhealthandclimatebenefitsestimatesrangefrom$61/MWhto$254/MWh,andrepresentthe5%to95%rangeconsideringthesameuncertainties.TheclimatebenefitsusearepresentativesocialcostofcarbonfromRennertetal.(2022),butarangeofestimatesexistintheliterature.FurtherdiscussionontherangeofhealthimpactscanbefoundinMillsteinetal.(2017),EPA(2023),andGilmoreetal.(2019).Likewise,furtherdiscussionoftherangeofsocialcostofcarbonestimatescanbefoundinRennertetal.(2022).64Land-BasedWindMarketReport:2023EditionTransmissionInvestmentsandWindPowerTheareaswiththegreatestwindspeedsareoftendistantfromelectricityloadcenters,makingwinddependentontransmissioninfrastructure.Related,thelowgrid-systemmarketvalueofwindinsomeareasofthecountryis,inpart,drivenbylimitedtransmissionandtheresultinggridcongestion.Transmissionadditionsreachedanewlowin2022,withjust675milesofnewtransmissionlinescomingonlineaccordingtotheFederalEnergyRegulatoryCommission(seefigurebelow).Thedeclinesincethepeakin2013ispartlyduetothecompletionofthetransmissionbuildoutinWestTexasin2013,aswellasasignificantbuildoutoflarger-scaletransmissioninSPPandMISOinthatsametimeframe.Sincethattime,muchofthetransmissionbuildoutintheUnitesStateshasfocusedonlocalreliabilityprojects,andnotthelarge-scale,longdistancenewtransmissionintendedinparttoaccesswindresources.CompletedTransmission(miles/year)500kV5,000345kV≤230kV4,0003,0002,0001,00002017201820192020202120222013201420152016Source:FERCmonthlyinfrastructurereportsCompilationofproposedtransmissionprojectsbytheNorthAmericanElectricReliabilityCorporationshowssimilartrends.Proposalsforfuturecircuitmilesdroppedfrom3,400miles/yearforthe2008–2014reportingyears(20%motivatedbyvariablerenewableintegrationvs.55%forreliability)to2,400miles/yearforthe2015–2022reportingyears(8%forrenewableintegrationvs.66%forreliability).45Dataoninterconnectionqueuesandtransmissioncongestionprovidefurtherevidenceofwind’srelianceonandchallengeswithtransmission.Asreportedearlier,themedianwindprojectreachingcommercialoperationin2022submittedaninterconnectionrequestnearly6yearsprior(Randetal.2023).Otherrecentresearchhasfoundthatinterconnectioncostsareontheriseacrossmanyregionsofthecountry,andthatwindtypicallyfaceshigherinterconnectioncoststhannewnatural-gaspowerplants(Seeletal.2023).Turningtotransmissioncongestion,theanalysispresentedinthischapterfindsthatwithin-regiontransmissioncongestionreducedthegrid-systemmarketvalueofwindbyanaverageof~$15/MWhin2022—aclearsignalofthevalueofnewtransmissionforwindpower.Millsteinetal.(2023)furtherfindwidespreadtransmissioncongestionacrosstheUnitedStates.Thevalueofpotentialnewintra-andinter-regionaltransmissioninprovidingcongestionreliefwashigherin2022thanatanypointinthelastdecade.Finally,asreportedearlier,windenergycurtailmentaveraged5.3%in2022,upfrom2.1%in2016andyetanothersignaloftransmissionconstraintsandtheirimpactonthewindpowersector.45Dataarecompiledfrom:https://www.nerc.com/pa/RAPA/ESD/Pages/default.aspx.Dataincludeproposedtransmissionlinesoverthefollowing10-yearperiod(e.g.the2008datasetreportstransmissionlineproposalsfor2009-2018).65Land-BasedWindMarketReport:2023Edition9FutureOutlookEnergyanalystsprojectgrowingwinddeployment,spurredbyincentivesintheInflationReductionActEnergyanalystsprojectthatannualwindadditionswillgrowinthecomingyears(BloombergNEF2023,WoodMackenzie2023b,GWEC2023,EIA2023c,IEA2022,2023).AmongtheforecastsforthedomesticmarketpresentedinFigure61,expectedcapacityadditionsrangefrom7.1GWto12GWin2023.Subsequentexpectedannualadditionsthenrampupsteadilythrough2027,supportedbyexpandedincentivesintheInflationReductionAct(U.S.DOE2023a)aswellasanticipatedgrowthinoffshorewind;allforecastsreportedhereincludebothland-basedandoffshorewind.By2027,expectedadditionsrangefrom18.4GWto22.7GW.TheseprojectedtrendsaredriveninpartbythepassageoftheInflationReductionActin2022.Asnotedearlier,IRAcontainsalong-termextensionofthePTCatfullvalue(assumingthatnewwageandapprenticeshipstandardsaremet)alongwithopportunitiesforwindplantstoearntwo10percentbonuscreditsthataddtothePTCformeetingdomesticcontentrequirementsandforbeinglocatedenergycommunities.AnalystsforecastgrowingimpactsofIRAovertime,partlyreflectingthefactthatwindprojectdevelopment,siting,andinterconnectioncantakeanumberofyears.Near-termadditionsarealsoinfluencedbythecostandperformanceofwindtechnologies,corporatewindenergypurchases,andstate-levelrenewableenergypolicies.Inflation,higherinterestrates,limitedtransmissioninfrastructure,interconnectioncostsandtimeframes,sitingandpermittingchallenges,andcompetitionfromsolarmaydampengrowth,asmightanycontinuingsupplychainpressures.Ingeneral,however,theinfluenceoftheInflationReductionActdominatesforecasts.Forexample,theaveragedeploymentforecastfor2026is18GW,comparedto11GWoneyearago,pre-IRA.AnnualCapacity(GW)Analystprojections25(bar=average)2015105019982000200220042006200820102012201420162018202020222024e2026eSources:ACP,BloombergNEF(2023),WoodMackenzie(2023b),GWEC(2023),EIA(2023c),IEA(2022)Figure61.Windpowercapacityadditions:historicalinstallationsandprojectedgrowth66Land-BasedWindMarketReport:2023EditionLongerterm,theprospectsforwindenergywillbeinfluencedbytheInflationReductionActandbythesector’sabilitytocontinuetoimproveitseconomicpositionTheprospectsforwindenergyinthelongertermwillbeinfluencedbytheimplementationoftheInflationReductionAct,whichnotonlyprovidesextensionsandexpansionsofdeployment-orientedtaxcreditsbutalsonewincentivesforthebuildoutofdomesticsupplychains.Alsoinfluencingdeploymentwillbethesector’sabilitytocontinuetoimproveitseconomicpositioneveninthefaceofchallengingcompetitionfromothergenerationresources,suchassolarandnaturalgas.Thespeedwithwhichsupplychainconstraintsareaddressedwillimpactdeploymentvolumes.Finally,changingmacroeconomicconditions,corporatedemandforcleanenergy,andstate-levelpolicieswillalsocontinuetoimpactwindpowerdeployment,aswillthebuildoutoftransmissioninfrastructure,resolutionofsiting,permittingandinterconnectionconstraints,andthefutureuncertaincostofnaturalgas.67Land-BasedWindMarketReport:2023EditionReferencesAmericanCleanPowerAssociation(ACP).2023.CleanPowerAnnualMarketReport2022.Washington,D.C.:AmericanCleanPowerAssociation.https://cleanpower.org/market-report-2022/BloombergNEF.2022a.2H2022WindTurbinePriceIndex(WTPI):SignsofCooling.December2022.BloombergNEF.2022b.2H2022WindO&MPriceIndex:InflationPressuresPersist.December2022.BloombergNEF.2023.1H2023U.S.CleanEnergyMarketOutlook.April2023.CenterforAir,ClimateandEnergySolutions(CACES).2023.DamageestimatesusingAP2/EASIUR/InMAPmodels.https://www.caces.usDavidson,M.R.andD.Millstein.2022.“Limitationsofreanalysisdataforwindpowerapplications.”WindEnergy.https://doi.org/10.1002/we.2759.EnergyInformationAdministration(EIA).2023a.AnnualEnergyOutlook2023.WashingtonD.C.:EnergyInformationAdministration.https://www.eia.gov/outlooks/aeo/EnergyInformationAdministration(EIA).2023b.ElectricPowerMonthly,withDataforMarch2023.WashingtonD.C.:EnergyInformationAdministration.https://www.eia.gov/electricity/monthly/EnergyInformationAdministration(EIA).2023c.Short-TermEnergyOutlook,June2023.WashingtonD.C.:EnergyInformationAdministration.https://www.eia.gov/outlooks/steo/FederalEnergyRegulatoryCommission(FERC).2023.EnergyInfrastructureUpdateforMarch2023(andpreviouseditions).Washington,D.C.:FederalEnergyRegulatoryCommission.https://www.ferc.gov/staff-reports-and-papersFell,H.andJ.X.Johnson.2021.“Regionaldisparitiesinemissionsreductionandnettradefromrenewables.”NatureSustainability,4:358-365.Gilmore,E.A.,Heo,J.,Muller,N.Z.,Tessum,C.W.,Hill,J.D.,Marshall,J.D.andP.J.Adams.2019.“Aninter-comparisonofthesocialcostsofairqualityfromreduced-complexitymodels.”EnvironmentalResearchLetters,14:074016.GlobalWindEnergyCouncil(GWEC).2023.GlobalWindReport2023.Brussels,Belgium:GlobalWindEnergyCouncil.https://gwec.net/globalwindreport2023/Hamilton,S.,D.Millstein,M.Bolinger,andR.Wiser.2020.“HowDoesWindProjectPerformanceChangewithAgeintheUnitedStates?”Joule4:1004-1020.Heo,J.,Adams,P.J.andH.O.Gao.2016.“Reduced-formmodelingofpublichealthimpactsofinorganicPM2.5andprecursoremissions.”AtmosphericEnvironment,137:80-89.Hitachi.2023.VelocitySuiteDataProduct.AccessedJune2023.IEA.2022.Renewables2022:Analysisandforecastto2027.Paris,France:InternationalEnergyAgency.https://www.iea.org/reports/renewables-2022IEA2023.RenewableEnergyMarketUpdate:Outlookfor2023and2024.Paris,France:InternationalEnergyAgency.https://www.iea.org/reports/renewable-energy-market-update-june-2023/68Land-BasedWindMarketReport:2023EditionKemp,J.M.,D.Millstein,J.H.Kim,andR.Wiser,2023.“Interactionsbetweenhybridpowerplantdevelopmentandlocaltransmissionincongestedregions.”AdvancesinAppliedEnergy,10:100133.LevelTenEnergy.2023.Q22023PPAPriceIndex:ExecutiveSummary,NorthAmerica.Seattle,Washington:LevelTenEnergy.https://www.leveltenenergy.com/resourcesMillstein,D.,R.Wiser,S.JeongandJ.M.Kemp.2023.TheLatestMarketDataShowthatthePotentialSavingsofNewElectricTransmissionwasHigherLastYearthanatAnyPointintheLastDecade.Berkeley,California:LawrenceBerkeleyNationalLaboratory.https://emp.lbl.gov/publications/latest-market-data-show-potentialMillstein,D.,R.Wiser,A.D.Mills,M.Bolinger,J.Seel,S.Jeong.2021.“SolarandwindgridsystemvalueintheUnitedStates:Theeffectoftransmissioncongestion,generationprofiles,andcurtailment.”Joule5:1-27.Millstein,D.,R.Wiser,M.Bolinger,andG.Barbose.2017.“Theclimateandair-qualitybenefitsofwindandsolarpowerintheUnitedStates.”NatureEnergy,2:17134.Moné,C.,M.Hand,M.Bolinger,J.Rand,D.Heimiller,andJ.Ho.2017.2015CostofWindEnergyReview.Golden,Colorado:NationalRenewableEnergyLaboratory.https://www.nrel.gov/docs/fy17osti/66861.pdfMuller,N.Z.2014.“BoostingGDPgrowthbyaccountingfortheenvironment.”Science,345:873-874.Rand,J.,R.Strauss,W.Gorman,K.Seel,J.Kemp,S.Jeong,D.Robson,andR.Wiser.2023.QueuedUp:CharacteristicsofPowerPlantsSeekingTransmissionInterconnectionAsoftheEndof2022.Berkeley,California:LawrenceBerkeleyNationalLaboratory.https://emp.lbl.gov/queuesRennert,K.,Errickson,F.,Prest,B.C.,Rennels,L.,Newell,R.G.,Pizer,W.,Kingdon,C.,Wingenroth,J.,Cooke,R.,Parthum,B.andD.Smith.2022.“ComprehensiveevidenceimpliesahighersocialcostofCO2.”Nature,610:687-692.Seel,J.,J.M.Kemp,J.Rand,W.Gorman,D.Millstein,F.KahrlandR.Wiser.2023.GeneratorInterconnectionCoststotheTransmissionSystem.Berkeley,California:LawrenceBerkeleyNationalLaboratory.https://emp.lbl.gov/publications/generator-interconnection-costsTessum,C.W.,Hill,J.D.andJ.D.Marshall.2017.“InMAP:Amodelforairpollutioninterventions.”PloSone,12:e0176131.UnitedStatesWindTurbineDatabase(USWTDB).U.S.GeologicalSurvey,AmericanWindEnergyAssociation,andLawrenceBerkeleyNationalLaboratorydatarelease.https://eerscmap.usgs.gov/uswtdbU.S.DepartmentofEnergy(U.S.DOE).2023a.AdvancingtheGrowthoftheU.S.WindIndustry:FederalIncentives,Funding,andPartnershipOpportunities.Washington,DC:U.S.DepartmentofEnergy.https://www.energy.gov/eere/wind/articles/us-wind-industry-federal-incentives-funding-and-partnership-opportunities-factU.S.DepartmentofEnergy(U.S.DOE).2023b.UnitedStatesEnergy&EmploymentReport2023.Washington,DC:U.S.DepartmentofEnergy.https://www.energy.gov/us-energy-employment-jobs-report-useerU.S.EnvironmentalProtectionAgency(EPA).2023.RegulatoryImpactAnalysisfortheProposedNationalEmissionStandardsforHazardousAirPollutants:Coal-andOil-FiredElectricUtilitySteamGeneratingUnitsReviewoftheResidualRiskandTechnologyReview.EPA-452/R-23-002.ResearchTrianglePark,North69Land-BasedWindMarketReport:2023EditionCarolina:U.S.EnvironmentalProtectionAgency.https://www.epa.gov/system/files/documents/2023-04/MATS%20RTR%20Proposal%20RIA%20Formatted.pdfWiser,R.,M.Bolinger,andE.Lantz.2019.“AssessingwindpoweroperatingcostsintheUnitedStates:Resultsfromasurveyofwindindustryexperts.”RenewableEnergyFocus30:46-57.Wiser,R.,andM.Bolinger.2019.BenchmarkingAnticipatedWindProjectLifetimes:ResultsfromaSurveyofU.S.WindIndustryProfessionals.Berkeley,California:LawrenceBerkeleyNationalLaboratory.https://eta-publications.lbl.gov/sites/default/files/wind_useful_life_report.pdfWoodMackenzie.2023a.GlobalWindTurbineOrderAnalysis:Q12023.February2023.WoodMackenzie.2023b.GlobalWindPowerMarketOutlookUpdate:Q12023.March2023.70Land-BasedWindMarketReport:2023EditionAppendix:SourcesofDataPresentedinthisReportInstallationTrendsDataonwindpoweradditionsandrepoweringintheUnitedStates(aswellascertaindetailsontheunderlyingwindpowerprojects)aresourcedlargelyfromACP(2023).Annualwindpowercapitalinvestmentestimatesderivefrommultiplyingwindpowercapacitydatabyweighted-averagecapitalcostdata(providedelsewhereinthereport).Dataonnon-windelectriccapacityadditionscomefromEIAandHitachi’sVelocitySuitedatabase.Globalcumulative(and2022annual)windpowercapacitydataaresourcedfromGWEC(2023)butarerevised,asnecessary,toincludetheU.S.windpowercapacityusedinthepresentreport.Country-levelwindenergypenetrationiscompiledbyACP(2023).ThewindprojectinstallationmapwascreatedbasedonACP’sprojectdatabase.WindenergyasapercentagecontributiontostatewideelectricitygenerationandconsumptionisbasedonEIAdataforwindgenerationdividedbyin-statetotalelectricitygenerationorconsumptionin2022.DataononlinehybridpowerplantscomeslargelyfromEIA(updatedwhenerroneousdataarediscovered).Thewindhybrid/co-locateddataarecompiledfromthe2022earlyreleaseEIA860dataset.Projectsareidentifiedashybridswithtwoapproaches.Thefirstapproachinvolvesidentifyingdistinctpowerplants(e.g.windandstorage)thatsharethesameEIAID.Thisapproachidentifiesmostofthehybriddatasummarizedinthereport.ThesecondapproachinvolvescompilingdatafromHitachi’sVelocitySuiteandmatchingpowerplantsthathavethesameHitachiPlantIDbutdifferentfueltypes.TheseplantswerethenfoundintheEIAdatasetandcross-checkedagainstlatitudeandlongitudeinformationtoconfirmco-location.DataonwindpowercapacityinvariousinterconnectionqueuescomefromareviewofpubliclyavailabledataprovidedbyeachISOorutility.FormoreinformationseeRandetal.(2023).IndustryTrendsTurbinemanufacturermarketsharedataarederivedfromtheACPprojectdatabase.DataonrecentU.S.nacelleassemblycapabilitycomefromACP(2023),asdodataonU.S.towerandblademanufacturingcapability.Manufacturerprofitabilitydatacomefromcorporatefinancialreports.DataonU.S.importsofselectedwindturbineequipmentcomefromtheDepartmentofCommerce,accessedthroughtheU.S.CensusBureau,andobtainedfromtheU.S.Census’sUSATradeOnlinedatatool(https://usatrade.census.gov/).Theanalysisofthetradedatareliesonthe“customsvalue”ofimportsasopposedtothe“landedvalue”andhencedoesnotincludecostsrelatingtoshippingorduties.Thetablebelowliststhespecifictradecodesusedintheanalysispresentedinthisreport.Alltradecodesusedtotrackwindequipmentimportedin2020andlaterareexclusivetowind.Insomepreviousyears,somecodesareexclusivetowind,whereasothersarenot.Assumptionsaremadefortheproportionofwind-relatedequipmentineachofthenon-wind-specificHTStradecategories.Theseassumptionsarebasedon:ananalysisoftradedatawhereseparate,wind-specifictradecategoriesexist;areviewofthecountriesoforiginfortheimports;personalcommunicationswithU.S.InternationalTradeCommissionandwindindustryexperts;U.S.InternationalTradeCommissiontradecases;andimportpatternsinthelargerHTStradecategories.71Land-BasedWindMarketReport:2023EditionTableA1.HarmonizedTariffSchedule(HTS)CodesandCategoriesUsedinWindImportAnalysisHTSCodeDescriptionYearsapplicableNotes8502.31.00007308.20.0000wind-poweredgeneratingsets2005–2022includesbothutility-scaleand7308.20.00202006–2010smallwindturbines8501.64.0020towersandlatticemasts2011–2022notexclusivetowindturbine8501.64.00212006–2011components8501.64.0121towers-tubular2012–20218412.90.9080ACgenerators(alternators)from750tomostlyforwindturbines8412.90.908110,000kVA20228503.00.9545ACgenerators(alternators)from750to2006–2011notexclusivetowindturbine8503.00.954610,000kVAforwind-poweredgeneratingsets2012–2022componentsACgenerators(alternators)from750to2006–20118503.00.956010,000kVAforwind-poweredgeneratingsets2012–2022exclusivetowindturbineotherpartsofenginesandmotorscomponents8503.00.95702014–2019windturbinebladesandhubsexclusivetowindturbinepartsofgenerators(otherthancommutators,2020–2022componentsstators,androtors)notexclusivetowindturbinepartsofgeneratorsforwind-poweredcomponentsgeneratingsetsexclusivetowindturbinecomponentsmachinerypartssuitableforvariousmachinerynotexclusivetowindturbine(includingwind-poweredgeneratingsets)componentsmachinerypartsforwind-poweredgeneratingexclusivetowindturbinesetscomponentsnotexclusivetowindturbinecomponents;nacelleswhenshippedwithoutbladescanbeincludedinthiscategory46exclusivetowindturbinecomponents;nacelleswhenshippedwithoutbladesareincludedinthiscategoryWindprojectownershipandpowerpurchasertrendsarebasedonaBerkeleyLabanalysisofACP’sprojectdatabase.TechnologyTrendsInformationonturbinenameplatecapacity,hubheight,rotordiameter,andspecificpowerwascompiledbyBerkeleyLabwithintheU.S.WindTurbineDatabasebasedoninformationprovidedbyACP,turbinemanufacturers,standardturbinespecifications,theFAA,websearches,andothersources.Thedataincludeprojectswithturbinesgreaterthanorequalto100kWthatbeganoperationin1998through2022.Estimatesofthequalityofthewindresourceinwhichturbinesarelocatedweregeneratedasdiscussedbelow.FAA“ObstacleEvaluation/AirportAirspaceAnalysis”datacontainingprospectiveturbinelocationsandtotalproposedheights,incombinationwithACPdataonnear-terminstallations,wereusedtoestimatefuturetechnologytrends.AnydatawithexpirationdatesbetweenNovember7,2022andJune6,2024werecategorizedaseither“pending”turbines(forthosethatalreadyhadreceivedanevaluationof“nohazard”)or“proposed”turbines(forthosethatwerestillbeingevaluated).AportionofthoseturbinesarecategorizedbyBerkeleyLab,withinputfromACPdataandHitachi’sVelocitySuitedata,aseither“underconstruction”orin46Theexplicitinclusionofnacelleswithoutbladeswaseffectivein2014becauseofCustomsandBorderProtectionrulingnumberHQH148455(April4,2014).Thatrulingstatedthatnacellesalonedonotconstitutewind-poweredgeneratingsets,astheydonotincludeblades—whichareessentialtowind-poweredgeneratingsetsasdefinedintheHTS.72Land-BasedWindMarketReport:2023Edition“advanceddevelopment.”Theformerareprojectsthathavebeenpartiallyorfullyconstructedbuthavenotbeenfullycommissioned.Thelatterarenotunderconstructionbutarehighlylikelytobeinthenextfewyearsandhaveoneofthefollowinginplace:asignedPPA(orsimilarlong-termcontract),afirmturbineorder,oranannouncementtoproceedunderutilityownership.PerformanceTrendsWindprojectperformancedatawerecompiledoverwhelminglyfromtwomainsources:FERC’sElectronicQuarterlyReportsandEIAForm923.AdditionaldatacomefromFERCForm1filingsand,inseveralinstances,othersources.Wherediscrepanciesexistamongthedatasources,thosediscrepanciesarehandledbasedonthejudgmentofBerkeleyLabstaff.DataoncurtailmentarefromERCOT,MISO,PJM,NYISO,SPP,ISO-NE,andCAISO.Thefollowingprocedurewasusedtoestimatethequalityofthewindresourceinwhichwindprojectsare(orareplannedtobe)located.First,withintheU.S.WindTurbineDatabase,thelocationofindividualwindturbinesandtheyearinwhichthoseturbineswere(orareplannedtobe)installedwereidentifiedusingFAADigitalObstacle(i.e.,obstruction)filesandFAAObstacleEvaluation/AirportAirspaceAnalysisfiles,combinedwithBerkeleyLabandACPdataonindividualwindprojects.Second,NRELused200-meterresolutiondatafromAWSTruepower—specifically,grosscapacityfactorestimates—toestimatethequalityofthewindresourceforeachofthoseturbinelocations.Thesegrosscapacityfactorsarederivedfromtheaveragemapped100-meterwindspeedestimates,windspeeddistributionestimates,andsiteelevationdata,allofwhicharerunthroughastandardwindturbinepowercurve(commontoallsites)andassumingnolosses.Tocreateanindexofwindresourcequality,theresultantaveragewindresourcequality(i.e.,grosscapacityfactor)estimateforturbinesinstalledinthe1998–1999periodisusedasthebenchmark,withanindexvalueof100%.Comparativepercentagechangesinaveragewindresourcequalityforturbinesinstalledafter1998–1999arecalculatedbasedonthat1998–1999benchmarkyear.Whensegmentingwindresourcequalityintocategories,thefollowingAWSTruepowergrosscapacityfactorsareused:the“lower”category,whichincludesallprojectsorturbineswithanestimatedgrosscapacityfactoroflessthan42%;the“medium”category,whichcorrespondsto≥42%–48%;the“higher”category,whichcorrespondsto≥48%–54%;andthe“highest”category,whichcorrespondsto≥54%.Separatefromwindresourcequality,alsoreportedareAWSTruepowerestimatesofsite-averagelong-termwindspeed,bothat100metersandathubheight.Hub-heightlong-termwindspeedestimatesaredevelopedbylinearlyinterpolatingbetweenAWCTruepowerestimatesfor80and100meters.NotallturbinescouldbemappedbyBerkeleyLabforthesepurposes;thefinalsampleincluded69,178turbinesofthe69,612installedfrom1998through2022inthecontinentalUnitedStates(i.e.,over99%).Mostoftheturbinesthatarenotmappedaremorethanadecadeold.Separatefromtheabove,therelativestrengthoftheaverage“fleet-wide”windresourcefromyeartoyearisestimatedbasedonweightingeachoperationalproject-levelwindresource(or“windindex”)byitsshareofthetotaloperationalfleet-widecapacityfortheparticularyear.Foreachindividualwindplant,anannualwindindexiscalculatedastheratioofaparticularyear’spredictedcapacityfactortothelong-termaveragepredictedcapacityfactor(withthelong-termaveragecalculatedfrom1998-2022).Site-levelavailablewindresourcesarecalculatedforeachhourofeachyearbasedonERA5reanalysiswindspeeddataforeachplant’slocation.ERA5hasahorizontalresolutionof~30km×30km.Site-specificestimatedwindspeeds(withthegeographicresolutionpreviouslynoted)areinterpolatedbetweenERA5modelheightstothecorrespondingrepresentativehub-heightforeachwindproject.Hourlywindspeedsateachprojectarethenconvertedtowindpowerbyapplyingproject-specificpowercurves.Inthiscase,powercurvesarebasedonthesetofturbine-specificpowercurvesderivedfromNREL’sSystemAdvisorModel,v2020.11.29andvarybasedonaplant’saveragespecificpower(averagedacrossallturbinesintheplant).Thisuseofpowercurvesisasimplification,butonethatdoesaccountfortheshiftinwindplantdesigntowardlowerspecificpowerturbines.Thewindindicesarecalculatedwithoutaccountingforwake,electrical,orotherlosses,orcurtailment,andarebasedonlyontheERA5windspeeds.TheseindicesareusedtorepresentchangesinthewindresourcefromoneyeartothenextandreflecttheERA5-basedstrengthofthetotalpotentialwindresourcegiventhetypesof73Land-BasedWindMarketReport:2023Editionturbinesthataredeployedateachsite.Notethatthesedataandindicesareusedtocharacterizeyear-to-yearvariationsinthestrengthofthewindresource,whereasAWSTruepowerestimatesareusedtocharacterizethestrengthofthesite-specificlong-termannualaveragewindresource.ThereportusesAWSTruepowerestimatesforthelatterneedduetotheirhighergeographicresolution.CostTrendsHistoricalU.S.windturbinetransactionpriceswere,inpart,compiledbyBerkeleyLab.Sourcesoftransactionpricedatavary,butmostderivefrompressreleases,pressreports,andSecuritiesandExchangeCommissionandotherregulatoryfilings.AdditionalandmorerecentdatacomefromVestas,SGREandNordexcorporatereports,BloombergNEF,andWoodMackenzie.BerkeleyLabusedavarietyofpublicandsomeprivatesourcesofdatatocompilecapitalcostdataforalargenumberofU.S.windprojects.Datasourcesrangefrompre-installationcorporatepressreleasestoverifiedpost-constructioncostdata.SpecificsourcesofdataincludeEIAForm412,EIAForm860,FERCForm1,variousSecuritiesandExchangeCommissionfilings,filingswithstatepublicutilitiescommissions,WindpowerMonthlymagazine,AWEA’sWindEnergyWeekly,theDOEandElectricPowerResearchInstituteTurbineVerificationProgram,ProjectFinancemagazine,variousanalyticcasestudies,andgeneralwebsearchesfornewsstories,presentations,orinformationfromprojectdevelopers.For2009–2012projects,datafromtheSection1603TreasuryGrantprogramwereusedextensively;forprojectsinstalledfrom2013through2020,EIAForm860dataareusedextensively.Somedatapointsaresuppressedinthefigurestoprotectdataconfidentiality.Becausethedatasourcesarenotallequallycredible,lessemphasisshouldbeplacedonindividualproject-leveldata;instead,thetrendsinthoseunderlyingdataoffergreaterinsight.Onlycostdatafromthecontiguouslower-48statesareincluded.WindprojectO&Mcostscomeprimarilyfromtwosources:EIAForm412datafrom2001to2003forprivatepowerprojectsandprojectsownedbyPOUs,andFERCForm1dataforIOU-ownedprojects.Asmallnumberofdatapointsaresuppressedinthefigurestoprotectdataconfidentiality.SalesPriceandLevelizedCostTrendsWindPPApricedataarebasedonmultiplesources,includingpricesreportedinFERC’sElectronicQuarterlyReports,FERCForm1,avoided-costdatafiledbyutilities,pre-offeringresearchconductedbybondratingagencies,andaBerkeleyLabcollectionofPPAs.SupplementaldatafromLevelTenEnergyarealsoreported,inbothnominal(asreported—seeassociateddatafile)andreal2022dollars.The2022dollarconversionassumesthatLevelTen’sreportedpricesineachquarterarefor12-year,flat-priced(innominaldollars)PPAsthatcommenceinthefollowingcalendaryear.Ineachquarter,wedeflatethe12-yearnominaldollarpriceseriesto2022dollarsusingtheGDPdeflator(actualdeflatorshistorically,alongwithprojectedfuturedeflatorsfromtheEIA’sAnnualEnergyOutlook2023),andthenlevelizetheresulting12-yearreal-dollarpriceseriesusinga4%realdiscountrate.RECpricedatawerecompiledbyBerkeleyLabbasedoninformationprovidedbyMarexSpectron.TheanalysiscalculatestheLCOEofwindbasedonLCOEinputdatacollected,inlargepart,byBerkeleyLabandpresentedelsewhereinthisreport—andassessedasexpectedLCOEasofthelistedcommercialoperationdates.Theseinputsincludecapitalcosts,capacityfactors,operationalexpenses,financingcosts,andassumptionsaboutusefullife.Specifically:•Forcapacityfactors,project-leveldataarelevelizedovertheassumedusefullifeofeachplant,applyingdegradationassumptionsfromHamiltonetal.(2020)asappropriate.Forprojectsbuiltin2022(thathavenotyetbeenoperatingforafullyear),capacityfactorsareassumedtomatchtheaveragecapacityfactorofprojectsbuiltinthesameregionsfrom2017to2020.•BasedonWiseretal.(2019),totaloperationalexpensesareassumedtofallfromalevelizedcostof$94/kW-yearin1998(expressedin2022dollars)to$71/kW-yearby2003,$60/kW-yearby2010,and74Land-BasedWindMarketReport:2023Edition$50/kW-yearby2018(andareinterpolatedlinearlybetweentheseyears).Projectsbuiltfrom2019-2022areindexedtothe2018valuebutvarybyCODyearbasedonBloombergNEF’sNorthAmericanwindO&Mpriceindex(BloombergNEF2022b).Notethattheseareprojectedfuturecosts;actualoperationalexpenditurescoulddivergefromindustryexpectations,astheyhaveinthepast.•Theweightedaveragecostofcapitalassumesa70%:30%debt-to-equitysplit(possibleintheabsenceofthePTC),withthecostofdebtvaryingovertimebasedonhistoricalchangesinthe20-and30-yearswapratesandbankspread,whilethecostofequitydeclinesfrom15%in1998to8%in2022.FinancingcostsareestimatedasifthePTCwerenotavailable.Theseareassumptionsforfuturereturns;actualreturnscoulddifferdependingonhowperformance,operationalexpendituresandprojectlifetimestrackexpectations.•Projectlifeisassumedtoincreaselinearlyfrom20yearsforprojectsbuiltin1998to30yearsforprojectsbuiltin2020andafter,basedonindustryexpectations(seeWiserandBolinger2019).•A35%corporatetaxrateisassumedfrom1998–2017and21%thereafter,withaconstant5%statetaxrateovertheentireperiod.Inflationexpectationsrangefrom1.9%to3.1%.Five-yearaccelerateddepreciationisappliedforallvintagesofwindprojects.CostandValueComparisonsTocomparethepriceofwindtothecostoffuturenaturalgas-firedgeneration,therangeoffuelcostprojectionsfromtheEIA’sAnnualEnergyOutlook2023isconvertedfrom$/MMBtuinto$/MWhusingheatratesderivedfromthemodelingoutput.Tocalculatethehistoricalwholesaleenergymarketvalueofwind,estimatedhourlywindgenerationprofilesarematchedtohourlynodalreal-timewholesaleprices.Thecapacityvalueateachplantisalsocalculated,basedonthemodeledwindprofilesandISO-specificrulesforwind’scapacitycreditandISO-zone-specificcapacityprices.Theresultingestimatesreflecttheaverage$/MWhenergyandcapacityvalueforeachplantandyear.ISO-levelaveragevaluesareestimatedbyweightingplant-levelvalueestimatesbyplantoutput.Tocalculatetheaverageenergyandcapacityvaluein$/MWh,thenumeratorisbasedonmodeledhourlygenerationaftercurtailment,butthedenominatorisbasedonthetotalgenerationwithoutcurtailment.Curtailmentisaccountedforonlyinthenumeratorsothatincreasedlevelsofcurtailmentwillreducetheaverage$/MWhvalue.TheMWh,inthiscase,reflectspotentialwindgenerationbeforecurtailment.Notethatpublicdatadonotbroadlyexistforhourlywindoutputprofilesattheplantlevel.Consequently,themodeledwindgenerationestimatesdescribedearlierareleveraged,albeitadjustedforcurtailmentandcorrectedforbias.Formodeledhourlyprofilesweuseadifferentinputmeteorologicalmodelthanwasusedforthewindindexcalculationdescribedearlier.InsteadofERA5weuseNOAA’sHigh-ResolutionRapidRefresh(HRRR)dataset.ComparedtoERA5,HRRRreducesbiasesandincreaseshourlycorrelationtorecordedgeneration(DavidsonandMillstein2022).WearenotabletouseHRRRforthelong-termwindindexcalculationbecausetheHRRRrecordsbeginin2014(andHRRRmethodologyisupdatedovertime).Byapplyingabiascorrectionprocesstothegenerationestimateswecanincorporatepubliclyavailableinformationonactualgenerationaswellassite-specificHRRRmodeledwindspeeds.OneexceptiontothisprocessisforplantslocatedinERCOT.ERCOTprovidedhightime-resolutionrecordsofplantlevelgenerationandcurtailmentgoingbackto2013,and,whereavailable,thosereportedvaluesareutilized.TotalcurtailmentisreportedbyeachISOforeithereachhouroreachmonth.TocorrectHRRRoutputestimatesforcurtailment,plantsaredividedintothreegroups:plantsreceivingthePTC,plantsthathaveagedoutofthePTC,andplantsthatelectedthe1603TreasuryGrantinsteadofthePTC.NotethatwecountplantsthathavebeenrepoweredaswithinthePTCgroup(assumingithasbeenlessthan10yearssincetherepowering).TotalreportedhourlycurtailmentisdistributedevenlyacrossallplantswithinaparticularISOthatfacelocalhourlypricesbelowathresholddefinedforeachgroup(initially,–$23/MWhforPTCplantsand$0/MWhfortheothertwogroups).Asimilarprocessisusedtodistributemonthlycurtailmentdata.75Land-BasedWindMarketReport:2023EditionBiascorrectioninvolvesaniterativelinearscalingapproachsothat:(1)thesumofestimatedgenerationacrossallplantswithineachISOmatchesthetotalwindgenerationreportedbyeachISOineachhourand(2)theannualtotalgenerationfromeachindividualplantmatchesitsexpectedannualoutput.Theexpectedannualoutputisbasedonthemodeledannualoutputadjustedforage-relatedperformancedecline(Hamiltonetal.2020)andcurtailment.Also,aregion-wideannualcorrectionfactorwasappliedbasedonEIAreportedplant-levelgenerationfromtheprioryear.Theseregion-widecorrectionfactorsweregenerallysmall,forexampleinMISO,SPP,ISO-NE,andPJMcorrectionfactorswerelessthan3%.ButHRRRgenerationestimateswerebiasedhighinsomeregions;CAISOandNYISOcorrectionfactorswere1.32and1.16.(NobiascorrectionwasneededforERCOTasweuseactualreportedplantgenerationprofiles).Overall,thedebiasingprocessensuresthatboththehourlydistributionofgenerationandthetotalannualgenerationmatchesbothmodeledandrecordedISO-leveldata.Hourlynodalreal-timewholesaleelectricitypricesandhourlyregionalwindoutputprofilesarefromHitachi’sVelocitySuitedatabase.CurtailmentdataaredownloadeddirectlyfromeachISO,orinsomecases,fromHitachi’sVelocitySuitedatabase.Foreachwindpowerplant,thenearestormost-representativepricingnodeisidentified,whichallowsrepresentativepricestobematchedtoeachplant.Forsomeregions,hourlywindoutputprofilesareonlyavailableforasubsetoftherelevantyearsoftheanalysis;assuch,estimatesofthewholesaleenergyvalueofwindarenotavailableforallyearsforallregions.Capacityvalueisestimatedforeachplantbasedonthebias-corrected,modeledwindprofilesandISOandISO-zonespecificcapacitypricesorcosts,aswellasrelevantregionalrulesforwind’scapacitycredit.AseparatecapacityvalueisnotcalculatedforERCOT,becauseERCOTrunsanenergy-onlymarketthatdoesnotrequireload-servingentitiestomeetaresourceadequacyobligation.InERCOT,however,hourlyOperatingReserveDemandCurvepricesareaddedtonodalenergyprices.CapacityvalueinERCOTisessentiallyincorporatedintotheenergymarkets.Asforcapacitypricesandcosts,manyregionshaveorganizedcapacitymarkets.Inthosecases,theanalysisusesmarket-clearingpricesfromcapacitymarketauctionsinconcertwithISO-rulesorestimatesforthecapacitycreditofwind.Forregionswhereload-servingentitieshavearesourceadequacyobligationbutlackorganizedcapacitymarkets,theanalysisusesdatafromregulatorybodiestoapproximatecapacitycostsandregionalestimatesorrulesforwind’scapacitycredit.Theanalysiscalculatesthedifferencebetweenwindvalueandflat-profilevalue(called“valuereduction”)andthenfurtherdecomposesthevaluereductionintothreeseparatecauses:profile,congestion,andcurtailment.Flatprofilevalueiscalculatedintwosteps.First,theaveragevalueofflat(“always-on”)generationiscalculatedatallpowerplantpricingnodesinaregion(bothwindandothertypesofpowerplants).Theregionalflatvalueisthencalculatedbytakingtheweighted-averagevalueacrossallthesepowerplantswithweightsbasedonrecordedenergyoutputateachplant.Theprofilevalueofwindiscalculatedinasimilarmannertotheregionalflatvalue,butinsteadofusingaflatprofile,awindplantoutputprofileisappliedtoallpowerplantsinaregion(bothwindandothertypes)andtheregionalweightedaveragevalueiscalculated.Thisprocessisrepeatedfortheprofilesforallwindplantsinaregiontodeveloptheregionalaveragewindplantprofilevalue.Thereductioninwindvalueduetoitsprofileisthencalculatedasthedifferencebetweentheregionalwindprofilevalueandtheregionalflatvalue.Next,thevalueofwindgenerationateachwindplantiscalculatedgivenitsoutputprofile,andtheregionalaveragevalueiscalculatedacrossallwindplants.Thisprovidesavalueofwindprofilesatwindplants—ineffect,thevalueofwindgeneration(notyetadjustedforcurtailment).Theprofilevaluecalculationfindsthevalueofwindoutputatallgeneratorlocationsandthewindgenerationvaluefindswindvalueonlyatwindgenerators,sothedifferencerepresentstheimpactoftransmissioncongestion.Finally,thevalueofwindisadjustedforcurtailmentbyincreasingthetotalenergyoverwhichenergyandcapacityrevenuearenormalized.Thisfinaladjustmentprovidestheoverallvalueofwindateachplant.ThesemethodsaredescribedinfurtherdetailinMillsteinetal.(2021).Turningtohealthandclimatebenefits,asmentionedinthemaintext,thevaluescalculatedherearebasedonamethodologythatiscurrentlyundergoingpeerreviewandshouldthereforebeconsideredpreliminary.Itis76Land-BasedWindMarketReport:2023Editionanticipatedthatvaluespublishedafterpeerreviewmayvaryfromthesereportedvaluesbutthattheoverallconclusionsandimplicationsoftheanalysisareunlikelytochange.Themarginalrateofhealthandclimatebenefitsisestimatedbasedonatwo-stepprocess.First,themarginalrateofavoidedemissionsforwindiscalculatedbasedonanupdateoftheapproachlaidoutbyFellandJohnson(2021).Fulldocumentationonthemethodologicalupdateswillbeavailableinaforthcomingarticle.Asummaryisincludedhere.First,California,theNorthwest,andtheSouthwestregionsarecombinedintoasingleregion,the‘West,’meaningthatimpactsarecalculatedacrosssevenseparateUnitedStatesregions,ratherthanthenineincludedbyFellandJohnson(2021).FellandJohnson(2021)estimatesarebasedonregressionsthatuseddataovertheperiodJuly2018throughMarch2020.Intheapproachhere,avoidedemissionsareestimatedbasedonlyongenerationprofilesfrom2022.Animportantchangeisthattheunderlyingregressionsareusedtofindtheimpactofhourlywindgenerationonhourlycoalandnaturalgasgeneration,ratherthanontotalhourlyregionalemissions.Avoidedemissionsarethencalculatedbyapplyingregionalaverageemissionratesbypowerplanttypetotheavoidedgenerationtotals(basedontheEPA’seGrid2021data).LikeFellandJohnson(2021),theapproachincludesanestimateoftheimpactofgenerationonneighboringregions,asimpliedbythechangetonet-exportsassociatedwithwindgeneration.TheanalysishereincludesnetexportimpactestimatesforNewYorkandNewEngland,regionsforwhichFellandJohnson(2021)donotassessexportimpacts.Theapproachalsoincludesanestimateofwindgeneration’sinstantaneousimpactonhydropower,andthesubsequentdelayedimpacttoemissions(fromshiftinghydropowerintime).Adifferenceinapproachisthatnetexportandhydropowerimpactsongenerationiscalculatedinseparateregressionequations,andwind’simpactinthesecasesisthencalculatedastheproductoftworegressionterms.Importantly,relativetoFellandJohnson(2021),thesetofchangesdescribedaboveledtoreducedestimatesofavoidedemissionsfromwindgeneration,particularlySO2,thoughthechangevariedbyregion.Inthisrespect,thechangeshereareconservative.HadavoidedemissionratesasestimatedbyFellandJohnson(2021)beenapplied,nationalbenefitsestimateswouldhavebeenlargerthanthepresentestimates.Asuiteofreduced-orderhealthimpactsmodelsisthenusedtoestimatethevalueoftheavoidedemissionsfromwind.Reduced-orderhealthimpactsmodelsusetheresultsoffullmeteorologicalandairqualitymodelstoprovidemoregeneralizedestimatesofthemarginalimpactsofemissionsfromspecificregions.ThisanalysisusesestimatesdevelopedinEPA(2023),andestimatescompiledinCACES(2023)representingthemodelsInMAP(Tessemetal.2017),EASIUR(Heoetal.2016),andAP2(Muller2014),whichcontainmarginalimpactestimates(asdollarsofhealthdamagepertonofemittedSO2andNOxemissionsbyregion)forpower-sectoremissions.Marginalimpactestimateswereadjustedforinflationtoa2022dollaryear.Eachreduced-ordermodelcontainsahighandlowestimateforthemarginaldamagerate,basedondifferingepidemiologicalstudies.FortheEPAestimatestheanalysiswasbasedona3%discountrate.NotethatonlytheEPAdataincludedanestimateofthebenefitsofreducedozoneexposure,whiletheestimatescompiledinCACEScontainedonlybenefitsestimatedfromreducedparticleexposure.AllCACESNOxbenefitestimateswerethereforepairedwiththeestimateofozonebenefitsfromEPAbasedonthehealthbenefitsofreducinglongtermexposureimpactsfromozone.TheproductofthesebenefitestimateswiththemarginalemissionrateprovidesamonetizedmarginalbenefitperMWhofwindgeneration.TheEPAestimatedhealthbenefits,butnottheCACESbenefits,includereducedhospitalizationsandreducedwork-daysmissed,buttheEPAmonetizationisdominatedbythecostofprematuremortalityduetopopulationexposuretoairpollution.ThevalueofavoidedCO2emissionsduetowindgenerationwascalculatedinacomparablemanner.Specifically,avoidedCO2emissionsweremultipliedbythesocialcostofcarbonfromRennertetal(2022),usingthe2.0%discountratecase,andwereadjustedforinflation(to2022$)toderiveamonetizedper-MWhbenefitforwindgenerationbyregion.Estimatesofhealthandclimatebenefitsaresubjecttouncertainty.WeuseMonteCarlosimulationtoestimateuncertainty.CentralpointinputsareusedasthecentralvaluesinnormalorskewednormaldistributionsforthepurposesoftheMonteCarlosimulations.Wepresentresultsforthe5th-95thpercentilesoftheMonteCarlosimulations.Inputparameteruncertainty(i.e.,standarddeviationsinthesimulations)isdetermineddirectly77Land-BasedWindMarketReport:2023Editionthroughtheregressionresultsinthecaseofavoidedcoalandnaturalgasgeneration.Uncertaintyintheemissionratesofcoalandgasplantsisrepresentedbythespreadofemissionratesacrossindividualplantsineachregion,weightedbygeneration.Uncertaintyinthereduced-orderhealthimpactmodelsisrepresentedbythespreadofestimatesacrossthesetofmodels,anduncertaintyinthesocialcostofcarbonisreporteddirectlyinRennertetal(2022).Benefitsineachregionarecalculatedindependentlyfromeachother.78Land-BasedWindMarketReport:2023EditionFormoreinformationvisit,energy.gov/eere/windDOE/GO-102023-6055•August2023Coverdetails:AwindFarmonthenorthshoreofOahu.ItisoperatedbyHECO.PhotobyDennisSchroeder,NREL57714

1、当您付费下载文档后，您只拥有了使用权限，并不意味着购买了版权，文档只能用于自身使用，不得用于其他商业用途（如 [转卖]进行直接盈利或[编辑后售卖]进行间接盈利）。
2、本站所有内容均由合作方或网友上传，本站不对文档的完整性、权威性及其观点立场正确性做任何保证或承诺！文档内容仅供研究参考，付费前请自行鉴别。
3、如文档内容存在违规，或者侵犯商业秘密、侵犯著作权等，请点击“违规举报”。

碎片内容

碳中和的最新文档