1、Examining Supply-Side Options to Achieve 100%Clean Electricity by 2035Examining Supply-Side Options to Achieve 100%Clean Electricity by 2035Lead authors:Paul Denholm,Patrick Brown,Wesley Cole,Trieu Mai,Brian Sergi Contributing authors:Maxwell Brown,Paige Jadun,Jonathan Ho,Jack Mayernik,Colin McMilla

2、n,Ragini SreenathSUGGESTED CITATIONDenholm,Paul,Patrick Brown,Wesley Cole,et al.2022.Examining Supply-Side Options to Achieve 100%Clean Electricity by 2035.Golden,CO:National Renewable Energy Laboratory.NREL/TP-6A40-81644.https:/www.nrel.gov/docs/fy22osti/81644.pdfNOTICE STATEMENTThis work was autho

3、red in part by the National Renewable Energy Laboratory,operated by Alliance for Sustainable Energy,LLC.for the U.S.Department of Energy(DOE)under Contract No.DE-AC36-08GO28308.The views expressed in the article do not necessarily represent the views of the DOE or the U.S.Government.The U.S.Governme

4、nt retains and the publisher,by accepting the article for publication,acknowledges that the U.S.Government retains a nonexclusive,paid-up,irrevocable,worldwide license to publish or reproduce the published form of this work,or allow others to do so,for U.S.Government purposes.This report is availabl

5、e at no cost from the National Renewable Energy Laboratory at www.nrel.gov/publications.U.S.Department for Energy(DOE)reports produced after 1991 and a growing number of pre-1991 documents are available free via www.OSTI.gov.iv This report is available at no cost from the National Renewable Energy L

6、aboratory at www.nrel.gov/publications.Acknowledgments The authors would like to thank the following individuals for their contributions.Editing and other communications support was provided by Madeline Geocaris,Al Hicks,Mike Meshek,Devonie Oleson,and Andrea Wuorenmaa.Helpful review and comments wer

7、e provided by Doug Arent,Sam Baldwin,Jose Benitez,Michael Berube,Sam Bockenhauer,Lauren Boyd,Adria Brooks,Steve Capanna,Jaquelin Cochran,Joe Cresko,Paul Donohoo-Vallett,Janelle Eddins,Zach Eldredge,Jay Fitzgerald,Andrew Foss,Carla Frisch,Jian Fu,Sarah Garman,Jennifer Garson,Patrick Gilman,Tomas Gree

8、n,Courtney Grosvenor,Anna Hagstrom,Elke Hodson,Jared Langevin,Ookie Ma,Jason Marcinkoski,Marc Melaina,Julia Miller,Alejandro Moreno,Matteo Muratori,Ramachandran Narayanamurthy,David Palchak,Fernando Palma,Kara Podkaminer,Ben Polly,Gian Porro,Sean Porse,Amir Roth,Ian Rowe,Neha Rustagi,Nicole Ryan,Rob

9、 Sandoli,Avi Schultz,Ben Shrager,Carolyn Snyder,Paul Spitsen,Jason Tokey,Jeff Winick,Ryan Wiser,and Owen Zinaman This work was authored by the National Renewable Energy Laboratory,operated by Alliance for Sustainable Energy,LLC,for the U.S.Department of Energy under Contract No.DE-AC36-08GO28308.The

10、 views expressed in the article do not necessarily represent the views of the DOE or the U.S.Government.The U.S.Government retains a nonexclusive,paid-up,irrevocable,worldwide license to publish or reproduce the published form of this work,or allow others to do so,for U.S.Government purposes.v This

11、report is available at no cost from the National Renewable Energy Laboratory at www.nrel.gov/publications.List of Acronyms AC alternating current ACS American Cancer Society ADE accelerated demand electrification AEO Annual Energy Outlook B2B back-to-back BECCS bioenergy for carbon capture and stora

12、ge CCGT combined cycle natural gas turbine CCS carbon capture and storage CO2 carbon dioxide CREZ competitive renewable energy zones CSP concentrating solar power CT combustion turbine DAC direct air capture DC direct current dGen Distributed Generation Market Demand model DOE U.S.Department of Ener

13、gy EASIUR Estimating Air pollution Social Impact Using Regression EIA U.S.Energy Information Administration FERC Federal Energy Regulatory Commission GHG greenhouse gas GW gigawatt GWh gigawatt-hour HVDC high-voltage direct current IWG Interagency Working Group on the Social Cost of Greenhouse Gases

14、 kg kilogram kV kilovolt LCC line-commutated converter Li-ion lithium-ion LTS Long-Term Strategy of the United States MMBtu million British thermal unit MT million tonnes(tonne is a metric ton)MWh megawatt-hour NPV net present value NREL National Renewable Energy Laboratory PM2.5 fine particulate ma

15、tter PV photovoltaics quad unit of energy equal to 1 quadrillion(1015)British thermal units ReEDS Regional Energy Deployment System model SCC social cost of carbon SMR steam methane reforming(note that SMR also commonly refers to small modular reactors;we do not use that acronym in this report)TW te

16、rawatt TWh terawatt-hour TW-mi terawatt-mile VSC voltage source converter vi This report is available at no cost from the National Renewable Energy Laboratory at www.nrel.gov/publications.Executive Summary This study evaluates a variety of scenarios that achieve a 100%clean electricity system(define

17、d as zero net greenhouse gas emissions)in 2035 that could put the United States on a path to economywide net-zero emissions by 2050.These scenarios focus primarily on the supply of clean electricity,including technical requirements,challenges,and benefit and cost implications.The study results highl

18、ight multiple pathways to 100%clean electricity in which benefits exceed costs.The study does not comprehensively evaluate all options to achieve 100%clean electricity,and it focuses largely on supply-side options.This Report and the Inflation Reduction Act and the Bipartisan Infrastructure Law The

19、analysis presented in this report was conducted prior to the passage of the Bipartisan Infrastructure Law(BIL)of 2021 and the Inflation Reduction Act(IRA)of 2022,which include incentives for and investments in clean energy technologies along with other energy system modernization provisions.Initial

20、analyses estimate that the energy provisions of these new laws can help lower U.S.economy-wide greenhouse gas emissions by approximately 40%below 2005 levels by 2030.1.The impacts of these provisions are expected to be most pronounced for the power sector,with grid emissions initially estimated to d

21、ecline to 68-78%below 2005 levels by 2030 and the share of generation from clean electricity sources estimated to rise to 60-81%.Investments in end-use sector decarbonization measures,including efficiency and electrification,are also supported by the IRA provisions.While the longer-term implications

22、 of these new laws are more uncertain,they are unlikely to drive 100%grid decarbonization and the levels of electrification envisioned by 2035 in the primary scenarios analyzed in this report.More specifically,existing state and federal policies relevant to the power sector as of October 2021 are re

23、presented in the modeled scenarios;none of the scenarios presented in this report includes the energy provisions from the IRA or BIL,or other newer enacted federal or state policies or actions.As the addition of IRA and BIL provisions are not expected to enable the U.S.power system to reach 100%carb

24、on-free electricity by 2035,their inclusion is not expected to significantly alter the 100%systems explored in this study.As such,the studys qualitative findings for the implications of achieving 100%are expected to still apply.However,given the potential significant impact of these new laws,the inc

25、remental differences between the Reference and 100%scenarios are expected to be lower than estimated here.Including IRA and BIL provisions would likely lower emissions in the Reference scenarios,resulting in a smaller gap between them and the 100%scenarios.As a result,the incremental electricity sys

26、tem costs of the 100%scenarios are expected to be lower with the inclusion of the IRA and BIL provisions.Similarly,the climate and air quality benefits of the 100%scenarios(relative to the Reference scenarios)would also be reduced.These changes have not been quantified and it is important to note th

27、at the analysis in this report does not provide any estimates of the impacts of these new laws.100%Clean Electricity by 2035 Scenarios We evaluated four main 100%clean electricity scenarios,which were each compared to two reference scenarios:one with“current policy”electricity demand(Reference-AEO)2

28、 and a second with much higher load growth through accelerated demand electrification(Reference-ADE).The Reference-ADE case includes rapid replacement of fossil fuel use with low-carbon alternatives across all sectors,including electrified end uses and low-carbon fuels and feedstocks,1 Example analy

29、ses:https:/www.energy.gov/articles/doe-projects-monumental-emissions-reduction-inflation-reduction-act;https:/ 2 This refers to the projections in the Annual Energy Outlook(AEO)from the U.S.Energy Information Administration(EIA 2021a).vii This report is available at no cost from the National Renewab

30、le Energy Laboratory at www.nrel.gov/publications.resulting in annual electricity demand that is 66%higher than in the Reference-AEO case in 2035.The four core scenarios apply a carbon constraint to achieve 100%clean electricity by 2035 under accelerated demand electrification and reduce economywide

31、 energy-related emissions by 53%in 2030 and 62%in 2035 relative to 2005 levels.Table ES-1 summarizes the four primary scenarios evaluated,which represent a range of uncertainties and themes(e.g.,technology availability)and which are described below.In each scenario,assumptions common to all scenario

32、s are called“reference,”and details are provided in the main body and Appendix C.All Options is a scenario in which all technologies continue to see improved cost and performance consistent with the National Renewable Energy Laboratorys(NRELs)Annual Technology Baseline(NREL 2021).This scenario inclu

33、des the development and deployment of direct air capture(DAC)technology,while the other three main scenarios assume DAC does not achieve the cost and performance targets needed to be deployed at scale.3 Infrastructure Renaissance assumes improved transmission technologies as well as new permitting a

34、nd siting approaches that allow greater levels of transmission deployment with higher capacity.Constrained is a scenario where additional constraints to deployment of new generation capacity and transmission both limits the amount that can be deployed and increases costs to deploy certain technologi

35、es.No CCS assumes carbon capture and storage(CCS)technologies do not achieve the cost and performance needed for cost-competitive deployment.This scenario also acts as a point of comparison to demonstrate the potential benefits of achieving cost-competitive deployment of CCS at scale.This is the onl

36、y scenario that includes no fossil fuel capacity or generation in 2035,and therefore it is the only scenario that includes zero direct GHG emissions in the electric sector.3 Executive Order 14057 defines“carbon pollution-free electricity”as“electrical energy produced from resources that generate no

37、carbon emissions,including marine energy,solar,wind,hydrokinetic(including tidal,wave,current,and thermal),geothermal,hydroelectric,nuclear,renewably sourced hydrogen,and electrical energy generation from fossil resources to the extent there is active capture and storage of carbon dioxide emissions

38、that meets EPA requirements”.The inclusion of non-generation,negative emission technologies such as direct air capture is not consistent with the Administrations 2035 clean electricity goal but are considered in the studys All Options Scenarios because of their potential deployment,emissions,and cos

39、t impacts.https:/www.federalregister.gov/documents/2021/12/13/2021-27114/catalyzing-clean-energy-industries-and-jobs-through-federal-sustainability viii This report is available at no cost from the National Renewable Energy Laboratory at www.nrel.gov/publications.Table ES-1.100%Clean Electricity Sce

40、narios and Sensitivities Evaluated in This Study Scenario Demand Assumptions Generation Resource Assumptions Renewable Resources CCS Technologies Transmission Nuclear Other Infrastructure All Options ADE Reference All including DAC Reference interregional AC expansion Reference Reference Infrastruct

41、ure Renaissance No DAC HVDC macrogrid Lower-cost transport and storage for H2,CO2,biomass Constrained Reduced land available for wind,solar,and biomass Intraregional transmission only,higher(5x)costs Not allowed in regions with current legislative restrictions Higher-cost transport and storage for H

42、2,CO2,biomass No CCS Reference No CCS,bioenergy with CCS,or DAC Reference Reference Reference Sensitivities(applied to each of the four core scenarios)Annual Energy Outlook(AEO)and the U.S.Long-Term Strategy(LTS)demand cases Supply-side sensitivities include renewable energy costs,storage costs,nucl

43、ear costs,electrolyzer costs,CCS cost and performance,transmission constraints,new natural gas restriction,natural gas fuel costs,expanded biomass supply,low-cost geothermal,and allowing DAC in the Infrastructure Renaissance and Constrained cases.ix This report is available at no cost from the Natio

44、nal Renewable Energy Laboratory at www.nrel.gov/publications.Beyond the four core 100%scenarios,142 additional sensitivities were also analyzed to capture future uncertainties related to technology cost,performance,and availability.Of these 142 sensitivities,122 cases model 100%carbon-free electrici

45、ty by 2035.We also evaluated all scenarios with a sensitivity case using electricity demand from the Long-Term Strategy of the United States(LTS)(White House 2021a)to reflect an alternative demand-side pathway to reaching a net-zero emissions economy by 2050.The LTS reflects higher levels of energy

46、efficiency and demand-side flexibility,resulting in slower annual load growth of 1.8%/year(compared to 3.4%/year under ADE)and,importantly,lower demand peaks that occur predominantly in summer as compared to the sharp winter peaks assumed for our primary ADE scenarios.In addition to direct electrici

47、ty demand,both ADE and LTS assumptions include demand for clean hydrogen production for transportation and industrial applications,which may be produced from electrolysis or from natural gas with CCS depending on scenario.Non-power sector demand for hydrogen is an input to the analysis;however,hydro

48、gen demand for electricity generation(for seasonal storage)is also considered and is an outcome of the scenarios.Electricity generation and capacity needed to produce hydrogenfor both power and non-power applicationsare also considered in the modeling.Across these scenarios,this work uses NRELs Regi

49、onal Energy Deployment System(ReEDS)model to identify the resulting least-cost investment portfolios from a range of different generation,storage,and transmission technologies while considering the significant geographical variation in demand and resource availability,including the regional and temp

50、oral variations in the output of renewable resources.The geographical and temporal variability of various resources is evaluated by ReEDS,including additional transmission costs needed for remote resources and the need to maintain an adequate supply of energy during all hours of the year.A detailed

51、list of limitations of the modeling approach and key caveats regarding scope,and cost elements included is provided in the Key Caveats section(Section 2.4,page 17).Scenario Deployment Results Achieving a 100%clean electricity system requires significant clean energy deployment,and a summary of the r

52、esults from the 100%scenarios is provided in Figure ES-1,including generation capacity,annual generation,average annual installation rate,and transmission capacity.x This report is available at no cost from the National Renewable Energy Laboratory at www.nrel.gov/publications.Figure ES-1.Summary res

53、ults of the main scenarios show a large increase in renewable capacity,with wind and solar providing about 60%80%of electricity.Challenges to transmission and wind siting result in nuclear providing about 27%in the Constrained scenario.Differences in the capacity mix for the remaining resources are

54、driven largely by assumptions about technology availability,particularly those related to CCS and negative emissions technologies.4 Interregional transmission capacity grows by two to three times current capacity in three of the scenarios,allowing greater access to low-cost wind resources and provid

55、ing the benefits of spatial diversity.4 Imports are from Canada,largely hydropower imports into the Northeast.Bio/Geo=conventional biopower such as wood waste,landfill gas,and geothermal.BECCS=bioenergy with CCS.Natural gas includes combustion turbines(CT),combined cycle(CC),and older oil and gas-fi

56、red steam plants,which are tracked individually but xi This report is available at no cost from the National Renewable Energy Laboratory at www.nrel.gov/publications.Based on assumed growth in demand due to end-use electrification,and electric demand associated with hydrogen production(for direct us

57、e or for production of other clean fuels),total electricity generation grows by about 95%130%from 2020 to 2035.Total generation is shown for all end-use loads(dotted line in Figure ES-1Figure)plus the additional generation needed for transmission losses and generation used by the electric sector to

58、produce hydrogen for seasonal electricity storage.There are differences between scenarios in absolute amounts of generation based on differences in storage(and associated losses)and hydrogen production.The need for new generation capacity would be even higher without the energy efficiency and demand

59、-side flexibility measures assumed in the ADE trajectory.Results from the LTS sensitivity cases result in a 16%20%reduction in the need for new installed capacity compared to the ADE cases due,in part,to the higher levels of energy efficiency assumed in LTS.Wind and solar provide most(60%80%)of the

60、generation in the least-cost electricity mix in all the main scenarios.Nuclear capacity more than doubles in the Constrained scenario,reaching 27%of generation,while limited growth in the other three core scenarios results in a contribution of 9%12%,largely from the existing fleet.The overall genera

61、tion capacity grows to roughly three times the 2020 level by 2035,including a combined 2 TW of wind and solar.This would require growth rates in the range of 4390 GW/year for solar and 70145 GW/year for wind by the end of the decade,which would more than quadruple the current annual deployment level

62、s for each technology in many scenarios.Across the four core scenarios,58 GW of new hydropower is deployed by 2035 by adding capacity at unpowered dams and uprates at existing facilities,while geothermal capacity increases by about 35 GW by 2035.Differences in energy contribution among the four core

63、 scenarios are largely driven by constraints in transmission and renewable siting.In all scenarios,significant transmission is constructed in many locations,and significant amounts are deployed to deliver energy from wind-rich regions to major load centers in the eastern United States.Total transmis

64、sion capacity(which is a mix of AC and HVDC depending on scenario)in 2035 is 1.32.9 times current capacity.Beyond already planned additions,these total transmission builds would require 1,40010,100 miles of new high-capacity lines per year,assuming new construction began in 2026.5 The Infrastructure

65、 Renaissance scenario constructs the most transmission and wind,and it results in the lowest average system cost.6 The Constrained scenario limits both renewable energy and transmission deployment,resulting in higher costs.The higher costs of renewables makes new nuclear capacity more cost-competiti

66、ve,and in this scenario,the model builds about 200 GW of new nuclear capacity between 2030 and 2035.This scenario would require about 40 GW/year of new installation,or about four times the maximum historical rate in the United States.The core scenarios deploy 120350 GW of diurnal storage to help sup

67、port power system resource adequacy(ensuring demand for electricity is met during all hours of the year)and better align output of wind and solar with demand patterns.Storage in Figure ES-1 includes diurnal combined for reporting purposes.Solar includes all utility-scale and rooftop solar photovolta

68、ics(PV)and concentrating solar power(CSP).TWh=terawatt-hours.GW=gigawatts.TW-mi=terawatt-miles.5 From 2010 to 2020,the maximum annual installation in the United States was about 4,100 miles/yr.6 This includes capital and operating costs for the generation,storage,and transmission infrastructure.xii

69、This report is available at no cost from the National Renewable Energy Laboratory at www.nrel.gov/publications.storage with discharge capacity of 212 hours,which includes batteries and pumped storage hydropower but could also include a variety of technologies under various stages of development.Base

70、d on assumed cost declines of renewable energy technologies,the pathway to achieving roughly 90%clean electricity is fairly consistent across the scenarios,and wind and solar provide the most generation in three of the scenarios,supplemented by significant nuclear deployment in the Constrained scena

71、rio.The variation between the scenarios is largely focused on the specific technologies that can most cost-effectively meet peak demand and can contribute to the last 10%of clean generation.This is reflected largely in the differences in capacity contribution among the four scenarios,which are drive

72、n by multiple factors,including uncertainty about technology availability at scale in the coming decades.The main uncertainty in reaching 100%clean electricity is the mix of technologies that achieves this target at least costparticularly considering the need to meet peak demand periods or during pe

73、riods of low wind and solar output.The analysis demonstrates the potentially important role of several technologies that have not yet been deployed at scale,including seasonal storage and several CCS-related technologies.The mix of these technologies varies significantly across the scenarios evaluat

74、ed depending on technology cost and performance assumptions.Seasonal storage is represented in the modeling by clean hydrogen-fueled combustion turbines but could also include a variety of technologies under various stages of development assuming they achieve similar costs and performance.There is s

75、ignificant uncertainty about seasonal storage fuel pathways,which could include synthetic natural gas and ammonia,and the use of alternative conversion technologies such as fuel cells.Other technology pathways are also discussed in the report.Regardless of technology,achieving seasonal storage on th

76、e scale envisioned in these results requires substantial development of infrastructure,including fuel storage,transportation and pipeline networks,and additional generation capacity needed to produce clean fuels.In all scenarios,the 100%requirement is met on a net basismeaning gross emissions can be

77、 offset through negative emissions technologies that rely on carbon capture.In the No CCS scenario,the 100%requirement precludes any fossil generation and has the greatest use of seasonal storage.In the other cases,fossil generatorsfrom existing and new plantscontinue to contribute through 2035,but

78、their emissions must be offset by negative emissions technologies including DAC and bioenergy with carbon capture and storage(BECCS)to achieve net-zero emissions.Fossil plants with CCS must also have negative emissions offsets because capture rates are assumed to be 90%and upstream methane leakage f

79、rom natural gas production must also be offset.7 Across all scenarios,0%5%of 2035 generation is from fossil technologies(both with and without CCS),and the All Options scenario includes about 660 GW of fossil capacity of all types in 2035).Under All Options,which is the only primary scenario that al

80、lows DAC,190 million tons/year of CO2 are removed using DAC and BECCS.7 Higher capture rates are evaluated in sensitivity cases.xiii This report is available at no cost from the National Renewable Energy Laboratory at www.nrel.gov/publications.The Constrained scenario emphasizes the potential challe

81、nges with siting and land use associated with the required infrastructure to achieve a fully decarbonized grid,although these challenges are apparent across all scenarios.Figure ES-2(page xiv)shows the regional capacity of wind,solar,and transmission,demonstrating significant deployment in many regi

82、ons.The figure also illustrates the land use associated with wind,solar,and long-distance transmission in the main scenarios,along with several other historical land use activities,but the figure does not include land use associated with ongoing processes such as future fossil fuel extraction.Solid

83、boxes represent areas dedicated to a single primary use.For wind,this includes area physically occupied by wind turbine pads,roads,and other infrastructure.Boxes with dashed lines represent area that could have multiple uses.For wind,this area represents the spatial extent of entire wind farms,inclu

84、ding the space between turbines that is available for agriculture,grazing,and other uses.The total area physically occupied by wind and solar infrastructure(solid boxes)is about equal to that currently occupied by railroads.Although they are not modeled in detail here,several other demand-side mecha

85、nisms can reduce the supply-side infrastructure needs in the scenarios.These include geothermal heat pumps,which could reduce annual and peak load relative to other electric space heating options especially in cold climates;even greater demand-side flexibility than what we considered;and broader sec

86、tor-coupling opportunities,especially between heating and electricity and for clean fuels.The LTS sensitivity results highlight that increased energy efficiency and demand-side flexibility measures can potentially reduce the overall capacity required to meet the 100%target.On average,by 2035,the LTS

87、 sensitivities deploy about 20%less capacity and require about 25%less generation than the ADE scenarios(Figure ES-3).The reduced deployment driven by energy efficiency and demand-side flexibility could result in lower power system costs.However,the LTS data do not comprehensively capture demand-sid

88、e capital and implementation costs;therefore,a direct cost comparison is not presented in this study.xiv This report is available at no cost from the National Renewable Energy Laboratory at www.nrel.gov/publications.Figure ES-2.Regional capacity of wind,solar,and transmission(top)shows substantial t

89、ransmission additions into wind-rich regions of the United States in the 2035 ADE scenarios.Wind deployment is significant in all regions but is particularly concentrated in the Midwest.Total land use(bottom)includes area dedicated largely for a single use,including land physically occupied by wind

90、and solar infrastructure(filled boxes)and land used for multiple applications(open boxes).This is particularly important for wind development,where most occupied by the plant is available for other uses.8 8 Historically,disturbed land area for coal mining(34,000 km2)is larger than the currently dist

91、urbed land area shown here(17,000 km2)(Global Energy Monitor 2021).Oil and gas production on federal lands represents less than 25%.(https:/www.blm.gov/programs-energy-and-minerals-oil-and-gas-oil-and-gas-statistics).xv This report is available at no cost from the National Renewable Energy Laborator

92、y at www.nrel.gov/publications.Figure ES-3.Impact of LTS demand assumptions show a 23%26%reduction in annual generation(top)and 19%20%reduction in total capacity requirements(bottom).Benefits and Costs of 100%Clean Electricity Achieving 100%clean electricity produces benefits that,in most scenarios,

93、outweigh the additional direct costs relative to a reference scenario.Figure ES-4(top)shows the reduction in fossil fuel use.Compared to Reference-AEO,the electrification that occurs in the Reference-ADE scenario leads to substantial reductions in(1)petroleum use in transportation and(2)natural gas

94、in buildings and industry by 2035.Moving to the 100%clean electricity scenarios further reduces fossil fuel use in the power sector.9 This fossil fuel reduction leads to a 54%reduction in GHG emissions compared to 2020(bottom).Reduction of particulates,SO2,and other emissions in the electric sector

95、leads to an estimated 40,000130,000 avoided premature deaths between 2020 and 2035 due to improved air quality.9 The reduction in petroleum use in the 100%clean electricity scenarios relative to Reference-ADE is zero because both have the same level of transportation electrification.The additional b

96、enefits of electrification accrue to the electric sector,primarily in reduced natural gas use.ReferenceInfrastructureAll OptionsConstrainedNo CCSADELTSADELTSADELTSADELTSADELTS0250050007500Annual Generation TWhReferenceInfrastructureAll OptionsConstrainedNo CCSADELTSADELTSADELTSADELTSADELTS0123Instal

97、led Capacity TWxvi This report is available at no cost from the National Renewable Energy Laboratory at www.nrel.gov/publications.Figure ES-4.The 100%scenarios produce a substantial reduction in fossil fuel consumption(top)and a corresponding reduction in GHG emissions compared to 2020(bottom),with

98、electrification playing an important role in decarbonizing energy use.Figure ES-5 compares the system costs against a limited set of emissions-related benefits.System costs include capital,fixed,and variable costs associated with generation and transmission,but they do not include administrative cos

99、ts or costs of maintaining or upgrading the distribution system.The figure shows an estimate of the net present value of the evaluated costs and benefits from 2023 to 2035,presented as differences between the 100%scenarios and Reference-ADE.The left bar in each scenario represents the system costs w

100、ith a negative value(meaning additional cost)of$330 billion to$740 billion,which represents the additional costs of achieving 100%clean electricity compared to the Reference-ADE scenario.The Constrained and No CCS scenarios have the greatest increase in direct costs.In the Constrained scenario,limit

101、s to new transmission result in significant increases in overall system xvii This report is available at no cost from the National Renewable Energy Laboratory at www.nrel.gov/publications.costs(an additional$370 billion relative to All Options).This is because the amount of wind that can be delivere

102、d to population centers is constrained,which results in additional deployment of storage and nuclear generation.The increased costs in the No CCS scenario demonstrate the potential benefit of negative emissions technologies to offset continued use of fossil resources for meeting peak demand,as well

103、as the potential value 100%-capture CCS could provide.The next bar to the right of the system cost bar in Figure ES-5 shows the benefit of reduced human health impacts resulting from certain criteria air pollution at fossil-fueled power plants(this does not include emissions reductions from transpor

104、tation).The benefit is shown as a positive value on top of the negative value associated with the higher system costs.Premature mortality from these power plant emissions results in about$390 billion in estimated monetary impacts in 2020 and declines toward zero in the 100%scenarios.The 100%scenario

105、s provide health benefits from reduced power plant emissions of$390 billion to$400 billion,which suggests these benefits alone outweigh the cost of achieving 100%in three of the four scenarios.The third bar adds the value of avoided climate damages measured by the social cost of carbon(SCC)value use

106、d by the Interagency Working Group on the Social Cost of Greenhouse Gases(IWG)(IWG 2021),which is about$80/tonne in 2020 and increases to about$100/tonne in 2035.This SCC value adds a benefit of about$1,200 billion to$1,300 billion,resulting in a net benefit of$900 billion to$1,300 billion(solid hor

107、izonal arrows).Finally,the fourth bar also adds avoided climate damages,but with a higher(constant)value of$275/tonne from Pindyck(2019),producing net benefits exceeding$3,500 billion.xviii This report is available at no cost from the National Renewable Energy Laboratory at www.nrel.gov/publications

108、.Figure ES-5.Achieving 100%clean electricity results in significant net benefits over the study period.Additional power system costs(represented by negative values in the far-left bar of each chart)are$330 billion to$740 billion compared to the Reference-ADE cost,with the largest difference resultin

109、g from restrictions to new transmission and other infrastructure development in the Constrained scenario.This cost is offset by health benefits from improved air quality(positive value of$390 billion).The final two bars add the benefit of avoided SCC.Using the lower SCC value(third bar)produces a ne

110、t benefit of$900 billion to$1,300 billion,while the final bar shows the value of the higher SCC value,producing a net benefit of$3,400 billion to$3,900 billion.All the core scenarios(and sensitivities)produce benefits that exceed costs,even when using the lower SCC values.The avoided health and clim

111、ate damages are shown as positive values(additional benefits).The LTS sensitivity cases(with lower electricity demand)have lower system costs($307 billion to$506 billion)because of the lower amount of new generation and transmission required,but these costs do not include the investment in efficienc

112、y upgrades needed to achieve the reduction in generation.Furthermore,most generation across the core scenarios is derived from generators that have zero fuel costs(or costs have historically been relatively stable,in the case of nuclear)and therefore provide a predictable cost trajectory and reduce

113、potential price shocks,including price stability in electrified industrial and transportation applications.There are expected to be significant air quality and environmental justice benefits associated with the electrification of transportation,buildings,and other end uses;however,these benefits are

114、 not assessed in this work.xix This report is available at no cost from the National Renewable Energy Laboratory at www.nrel.gov/publications.Implications and Future Research The rapid reduction in the costs of renewable and several other clean energy technologies over the past two decades allows fo

115、r continued large-scale deployments that are expected to generate benefits that substantially outweigh the associated power system cost,assuming these technology cost declines continue in the coming decades.However,achieving the transformation of the U.S.energy system to 100%clean electricity as env

116、isioned in these scenarios requires four challenging actions to occur in the next decade:1.Dramatically accelerating electrification and increasing the efficiency of the demand sectors to get the country on the path to net-zero emissions by midcentury.Electrification will dramatically increase deman

117、d,which in turn may make it more difficult to decarbonize the electricity system due to the higher rate of generation and transmission capacity additions needed.However,electrification of end uses in buildings(with a critical parallel focus on efficiency of those end uses)and much of transportation

118、and industry is likely a key part of the most cost-effective pathway to achieving large-scale decarbonization across the economy.Furthermore,a parallel focus on efficiency and flexibility of end uses has the potential to greatly impact generation supply needed.More flexible operation could provide h

119、igher utilization of generation,transmission,and distribution assets,lowering the delivered cost of electricity.To achieve decarbonization of all energy sectors by 2050,further electrification,low-to zero-carbon fuel production,energy efficiency,and demand flexibility measures will be needed.2.Insta

120、lling new energy infrastructure rapidly throughout the country.This includes siting and interconnecting new renewable and storage plants at rates of three to six times recent levels,potentially doubling or tripling the capacity of the transmission system,upgrading the distribution system,building ne

121、w pipelines and storage for hydrogen and CO2,and/or deploying nuclear and carbon management technologies with low environmental disturbance and in an equitable fashion to all communities.3.Expanding clean technology manufacturing and the supply chain.The unprecedented deployment rates for clean elec

122、tricity technologies envisioned in the 100%scenarios requires a corresponding growth in raw materials supply,manufacturing facilities,and trained workforce throughout the supply chain.Further analysis is needed to understand how to achieve the scale-up of manufacturing as part of a just transition t

123、o a clean electricity system.This includes evaluating the economic and energy security benefits of increasing domestic manufacturing.4.Accelerating research,development,demonstration,and deployment to bring emerging technologies to the market.Technologies that are being deployed widely today can pro

124、vide most U.S.electricity by 2035 in a deeply decarbonized power sector.A 90%clean grid can be achieved at low incremental cost by relying primarily on new wind,solar,storage,advanced transmission,and other technologies already being deployed at scale today.However,the path from 90%to full decarboni

125、zation is less clear,as many of the technologies that could best aid full decarbonization,such as clean hydrogen and other low-carbon fuels,advanced nuclear,price-responsive demand response,CCS,and DAC,have not yet been deployed at large scale.A concerted research,development,demonstration,and deplo

126、yment effort is needed to reduce costs xx This report is available at no cost from the National Renewable Energy Laboratory at www.nrel.gov/publications.and improve performance to enable these technologies to be commercialized at scale and support a fully decarbonized grid.This ambitious list of tas

127、ks will require explicit support to be achieved in the coming decade.Failing to achieve any of these actions could increase the difficulty of realizing a 100%clean grid by 2035.However,damages from climate change are not binary,so even if emissions reductions fall short of those envisioned in the sc

128、enarios here,significant harm to human health,economies,and the ecological system can be avoided by making progress toward decarbonization.xxi This report is available at no cost from the National Renewable Energy Laboratory at www.nrel.gov/publications.Table of Contents 1 Introduction.1 2 Study Met

129、hods and Assumptions.3 2.1 Demand Assumptions.3 2.2 Scenario Framework.8 2.3 Costs and Benefits.15 2.4 Key Caveats.17 3 Scenario Deployment Results.21 3.1 Wind.26 3.2 Solar.28 3.3 Geothermal,Hydropower,and Biopower.29 3.4 Nuclear.30 3.5 Diurnal Energy Storage.31 3.6 Technologies That Help Address th

130、e Challenge of Seasonal Mismatch.33 3.6.1 Seasonal Storage.37 3.6.2 CCS and DAC.40 3.7 Transmission.43 3.8 Land Use.50 3.9 Deployment Results of the LTS Sensitivity.54 4 Costs and Benefits.57 4.1 Summary Results.57 4.2 Drivers of Costs and Benefits.59 4.3 Cost Results of the LTS Sensitivity.66 5 Con

131、clusions and Future Research.68 References.70 xxii This report is available at no cost from the National Renewable Energy Laboratory at www.nrel.gov/publications.List of Figures Figure ES-1.Summary results of the main scenarios show a large increase in renewable capacity,with wind and solar providin

132、g about 60%80%of electricity.x Figure ES-2.Regional capacity of wind,solar,and transmission(top)shows substantial transmission additions into wind-rich regions of the United States in the 2035 ADE scenarios.xiv Figure ES-3.Impact of LTS demand assumptions show a 23%26%reduction in annual generation(

133、top)and 19%20%reduction in total capacity requirements(bottom).xv Figure ES-4.The 100%scenarios produce a substantial reduction in fossil fuel consumption(top)and a corresponding reduction in GHG emissions compared to 2020(bottom),with electrification playing an important role in decarbonizing energ

134、y use.xvi Figure ES-5.Achieving 100%clean electricity results in significant net benefits over the study period.xviii Figure 1.The three-step process conducted for this study.3 Figure 2.Electricity demand projections show a substantial increase in the ADE assumptions due to electrification,offsettin

135、g energy demand in other sectors.5 Figure 3.Assumed clean hydrogen demand for industry and transportation.6 Figure 4.U.S.load profiles for seasonal peak days in the ADE scenario(top row)show how electrification increases peak demand and shifts the peak from summer to winter in 2035 due to electric s

136、pace heating.7 Figure 5.Electricity-sector emissions pathways for 100%clean electricity scenarios.14 Figure 6.Examples of the large range in estimates of SCC.17 Figure 7.Total annual generation for each of the four pathways exceeds end-use electricity demand because of transmission/distribution/stor

137、age losses and electricity use for hydrogen production and DAC.21 Figure 8.Energy and capacity contribution by resource in the main scenarios in 2020 and 2035(ADE demand case)demonstrates large growth in several clean energy technologies.22 Figure 9.Generation by resource type over time demonstrates

138、 the rapid transition to clean energy resources.23 Figure 10.Capacity and energy contribution by resource in 2035 show significant wind,solar,and storage deployments in all 122 100%clean electricity sensitivity cases.24 Figure 11.Annual growth in capacity of seven technology classes across all 122 c

139、lean electricity sensitivity cases.25 Figure 12.Annual growth in capacity in the main scenarios(ADE demand case).26 Figure 13.Growth in wind across the four main scenarios(ADE demand case)ranges from 60 to 160 GW/year by the end of the decade.27 Figure 14.Growth in solar across the four main scenari

140、os(ADE demand case)ranges from 25 to 120 GW/year by the end of the decade.29 Figure 15.Growth in geothermal includes about 8 GW of new capacity in all scenarios,with much greater growth in sensitivity cases with low-cost geothermal.30 Figure 16.Growth in nuclear is largest in the Constrained scenari

141、o,adding about 200 GW.31 Figure 17.Cumulative diurnal storage deployment(ADE demand case;measured by power capacity)shows large-scale growth,with significant contribution from capacity with 4-hour durations used to meet the summer peaks and absorb surplus solar generation in the late morning and ear

142、ly afternoon.32 Figure 18.Cumulative storage deployment(ADE demand case;measured by energy capacity)shows electric vehicles as the main driver for growth in manufacturing capacity and overall battery demand on the path to economywide decarbonization.33 Figure 19.Energy and capacity in the 2035 All O

143、ptions scenario(ADE demand case)show the significant dependence on remaining fossil-fueled capacity to provide peaking capacity and ensure resource adequacy during the clean energy transition.34 xxiii This report is available at no cost from the National Renewable Energy Laboratory at www.nrel.gov/p

144、ublications.Figure 20.The seasonal supply/demand balance for the contiguous United States in the All Options scenario(ADE demand case)in 2035 shows the seasonal mismatch challenge.35 Figure 21.Sources of firm capacity in the winter and summer in the 100%clean electricity scenarios in 2035(ADE demand

145、 case)show a much greater mix of resources compared to those providing energy.36 Figure 22.Sources of hydrogen fuel production(ADE demand case)shows the potentially important role of electrolysis.38 Figure 23.Capacity of hydrogen-fueled combustion turbine generators in 2035(ADE demand case)shows the

146、 importance of these technologies,particularly if CCS is not available.39 Figure 24.Capacity and energy contribution of fossil and CCS-related technologies(ADE demand case)shows the ability of negative emissions technologies to offset remaining fossil generation used for peaking capacity.42 Figure 2

147、5.Wind and solar resource maps of the United States show that many of the best resources are in locations remote from demand centers in the eastern part of the country,which require new transmission.44 Figure 26.Interregional transmission capacity grows substantially in three of the four scenarios(A

148、DE demand case).45 Figure 27.Maps of transmission capacity in 2020 and 2035(ADE demand case)show substantial additions into wind-rich regions of the United States.47 Figure 28.Wind capacity relative to transmission capacity in 2035(top left)shows the synergy between the technologies.48 Figure 29.Reg

149、ional capacity of wind,solar,and transmission shows the importance of codevelopment of resources in several regions(ADE demand case).50 Figure 30.Total area occupied by wind turbine and solar infrastructure(solid boxes)is about equal to the land occupied by railroads(ADE demand case).52 Figure 31.To

150、tal generation(top)and capacity(bottom)in 2035 in the ADE scenarios compared to the LTS sensitivity cases.55 Figure 32.Achieving 100%clean electricity results in significant net benefits over the study period.57 Figure 33.Net system costs adding electricity system,health damages,and climate damages

151、over time.58 Figure 34.Average system cost increases by$15$39/MWh(about 1.53.9 cents/kWh)in the main scenarios compared to the Reference-ADE case,and$834/MWh relative to 2020.60 Figure 35.Low technology cost assumptions reduce the cumulative costs of achieving 100%clean electricity by 2035(cumulativ

152、e net present value of system costs calculated using a 2.5%discount rate).61 Figure 36.The increase in electricity cost associated with 100%clean electricity is within the historical range of variations in retail costs.62 Figure 37.Avoided fossil fuel use in 2035 100%clean electricity scenarios rela

153、tive to reference cases by fuel type and scenario.63 Figure 38.Total sector CO2(e)emissions(top row)trajectories for the Reference-AEO,Reference-ADE,and 100%scenarios.64 Figure 39.Annual climate benefits of 100%clean electricity using different estimates of the SCC.65 Figure 40.Estimates of annual a

154、voided deaths and costs from 100%clean electricity scenarios compared to Reference-ADE.65 Figure 41.Cumulative net present value of costs and benefits produces a positive benefit-cost ratio for all sensitivities evaluated.66 Figure 42.Net present values of system costs in the ADE scenarios and LTS s

155、ensitivity cases show a reduction in investment costs associated with the lower LTS demand.67 xxiv This report is available at no cost from the National Renewable Energy Laboratory at www.nrel.gov/publications.List of Tables Table ES-1.100%Clean Electricity Scenarios and Sensitivities Evaluated in T

156、his Study.viii Table 1.100%Clean Electricity Scenarios and Sensitivities Evaluated in This Study.10 Table 2.Supply-Side Technologies Considered.12 Table 3.Total CO2 Captured and Stored(MT/year)in 2035(ADE Demand Case)in the Scenarios That Allow CCS.43 Table 4.Footprint Comparison(ADE Demand Case)(km

157、2).53 List of Text Boxes Text Box 1.Capacity-Related Terms Used in This Report.19 Text Box 2.The Three Rs:Resource Adequacy,Operational Reliability,and Resilience.20 Text Box 3.Delivering More with Less:Benefits of Enhanced Transmission Technologies.49 Text Box 4.The Role of Distributed Resources.54

158、 1 This report is available at no cost from the National Renewable Energy Laboratory at www.nrel.gov/publications.1 Introduction This study evaluates a variety of scenarios that achieve a 100%clean electricity system(defined as zero net greenhouse gas emissions)in 2035.The trajectories analyzed put

159、the United States on a path to economywide net-zero emissions by 2050,but pathways beyond 2035 are not evaluated in this work.10 This work builds on previous National Renewable Energy Laboratory(NREL)analyses,including the Renewable Electricity Futures Study(NREL 2022a),North American Renewable Inte

160、gration Study(Brinkman et al.2021),Electrification Futures Study(NREL 2022b),Storage Futures Study(NREL 2022c),and Solar Futures Study(Ardani et al.2021).This work also joins the growing body of work that examines 100%renewable and 100%clean electricity systems in the United States and international

161、ly.Recent examples include Brown and Botterud(2021),Larson et al.(2021),EPRI(2021)and DeAngelo(2021).There are also reviews and summaries of several studies provided by Azevado et al.(2021),Bistline(2021),and Breyer et al.(2022).Among the goals of this present study is to contribute to this larger b

162、ody of work,much of which examines various pathways and timelines.These previous studies illustrate several consistent themes,including:Maintaining resource adequacy while increasing contribution of renewable energy and other clean energy resources is technically feasible.Wind and solar,enabled by e

163、nergy storage,are typically the least-cost technologies used to meet most energy needs in the power sector.Transmission expansionespecially interregional expansionincreases access to low-cost generation resources and improves system flexibility,helping to achieve reliability at lowest cost.Demand gr

164、owth,particularly due to electrification,can dramatically increase overall generation requirements,potentially offset to various degrees by energy efficiency measures.Previous NREL studies mentioned looked at various levels of clean electricity deployment through 2050.In contrast,this study consider

165、s an accelerated time frame and complete decarbonization of the electric sector to achieve net-zero greenhouse gas(GHG)emissions by 2035.Though many of the trends seen in the previous studies are consistent with those seen in this study,this study also identifies the particular challenges of the sca

166、le and rate of deployment needed to reach 100%clean electricity by 2035.This study also illustrates the rapid accumulation of benefits associated with near-term reductions in both criteria pollutants and GHGs that result from achieving this target.The results highlight multiple pathways to 100%clean

167、 electricity in which benefits exceed costs,with the main uncertainty being the mix of technologies that achieves the 100%target at least cost.The uncertainty is particularly related to the challenges of moving from 90%to 100%clean electricity and providing reliable electricity during periods of pea

168、k demand.The study examines wide-ranging scenarios with different mixes of electricity supply-side optionsincluding 10 Throughout this report,we refer to the U.S.power system as a shorthand for the power system of the contiguous United States,which excludes Alaska and Hawaii.2 This report is availab

169、le at no cost from the National Renewable Energy Laboratory at www.nrel.gov/publications.renewable power,nuclear,carbon capture(and negative emissions technologies),energy storage,and transmissionthat could reach 100%clean electricity and meet demands from current and newly electrified loads.The mai

170、n scenarios assume widespread electrification to maximize the use of carbon-free electricity to support economy-wide decarbonization,but further study focused on the demand sectors,including the potential role of energy efficiency,demand-side flexibility,and low-carbon fuels,is needed to understand

171、the trade-offs between supply-side and demand-side solutions to decarbonization.All the scenarios evaluated achieve benefits that exceed the power system costs of achieving 100%clean electricity.The benefits analysis in this work includes reductions in CO2 and methane emissions in the electricity se

172、ctor11 and a high-level examination of the public health impacts of improved air quality resulting from electricity sector emissions reductions.It does not consider the additional health benefits from reduced emissions in buildings,industry,and transportation,or distributional aspects of these benef

173、its,including environmental justice,nor does it assess energy security and geopolitical implications.A detailed discussion of what is and is not considered in this work(including costs and benefits)is discussed in the Key Caveats section(Section 2.4,page 17).11 Note that when we report CO2 emissions

174、,we are actually reporting the combination of direct CO2 emissions and the CO2 equivalent emissions from methane leakage in the power sector,as discussed in the Methods section(Section 2,page 3).We estimate reductions in CO2 emissions in several other energy sectors,including transportation,building

175、s,and industry,due to electrification and use of clean fuels.We quantify these emission reductions benefits but do not account for them in the final cost/benefit analysis,as we do not estimate the cost of implementing these measures.3 This report is available at no cost from the National Renewable E

176、nergy Laboratory at www.nrel.gov/publications.2 Study Methods and Assumptions This study is designed to examine changes in the U.S.electric sector required to achieve 100%clean electricity by 2035,focusing on the mix of clean energy generation resources that must be deployed,along with enabling tech

177、nologies such as energy storage and transmission.It consists of three major steps,illustrated in Figure 1.The following section details how each step was performed.Figure 1.The three-step process conducted for this study.ReEDS refers to the Regional Energy Deployment System model.2.1 Demand Assumpti

178、ons Step 1 in Figure 1 requires estimating the changes in electricity demand that could occur by 2035.The demand scenarios are constructed to be on a path to economywide decarbonization in 2050.In 2020,the electric sector generated 2,408 TWh of electricity from fossil fuels,12 representing about 32%

179、of total CO2 emissions across all sectors(EIA 2021a),which means decarbonization of electricity alone is insufficient to meet climate goals.Four approaches are often considered to decarbonizing end-use energy sectors.One is electrificationthe conversion of devices currently burning fossil fuels to r

180、un on electricitywhile simultaneously decarbonizing the electric sector.The second approach involves taking energy efficiency measures to reduce demand.The third is using low-to zero-carbon liquid or gas fuels.Finally,the fourth is to capture and sequester CO2 to reduce emissions from fossil fuel us

181、e to offset emissions through CO2 removal.We assume all approaches will be deployed to achieve energy system decarbonization by 2050,with rapid 12 This includes 774 TWh from coal,1,617 TWh from natural gas,and 17 TWh from oil(EIA 2022a).4 This report is available at no cost from the National Renewab

182、le Energy Laboratory at www.nrel.gov/publications.electrification occurring in the near term and having the most direct impact on electricity supply-side changes.We evaluated three demand trajectories,all serving as inputs to the power sector modeling.The first is a business-as-usual electricity dem

183、and based on the Annual Energy Outlook(AEO)2021(EIA 2021a).The second is the accelerated demand electrification(ADE)case,which is used as our primary demand trajectory in the core scenarios;it assumes all measures to decarbonize end-use sectors but features substantial electrification.The third traj

184、ectory is used as a sensitivity and is based on a scenario from the Long-Term Strategy of the United States(LTS),which assumes greater energy efficiency adoption and less electrification than ADE(White House 2021a).The LTS report presents multiple pathways for the U.S.economy to achieve net-zero emi

185、ssions by 2050 and was based on diverse analytical tools and consultation from a wide range of stakeholders.The core scenarios use the ADE demand assumptions,as they reflect a high bound on future electricity supply needs demonstrating the potential need for large increases in electricity supply.How

186、ever,LTS sensitivities are highlighted throughout,given the broad uncertainties for end-use decarbonization pathways,and they illustrate the potential impact of further efficiency measures to reduce the need for supply-side resources.13 Figure 2 summarizes this studys projected growth in annual end-

187、use electricity demand from 2020 to 2035 by sector.The AEO 2021 reference trajectory shows a small annual increase in demand(0.9%/year from 2020 to 2035)associated with population and economic growth and minimal electrification in a“no new policy”scenario.The ADE trajectory assumes significant elect

188、ricity demand growth(3.4%/year),and the LTS sensitivity trajectory is between AEO and ADE(1.8%/year).In the ADE case,growth in demand is driven by aggressive electrification of end uses,including space and water heating,and deployment of electric vehicles.This electrification-driven demand growth as

189、sumption results in 6,200 TWh of annual direct electricity consumption in 2035,compared with 3,700 TWh in 2020.In contrast,2035 electricity demand is 21%(1,300 TWh)lower than ADE in the LTS case,largely through greater assumed energy efficiency.Appendix C provides a detailed summary of the assumptio

190、ns across various sectors for the ADE and LTS cases.13 Final energy consumption for 2035 is estimated to be 78 quads,71 quads,and 60 quads for the AEO,ADE,and LTS scenarios respectively.A quad is a unit of energy equal to 1 quadrillion(1015)British thermal units.Electricitys share of final energy in

191、 2035 is estimated to be 19%,30%,and 28%for the same respective scenarios.See Appendix C for details.5 This report is available at no cost from the National Renewable Energy Laboratory at www.nrel.gov/publications.Figure 2.Electricity demand projections show a substantial increase in the ADE assumpt

192、ions due to electrification,offsetting energy demand in other sectors.Figure 2 shows the assumed electricity demand inputs,but the total generation required will be larger due to electricity for clean fuel production losses,transmission and distribution losses,and energy storage losses.Direct use of

193、 multiple clean fuelsincluding hydrogen,biofuels,synthetic hydrocarbons,and ammoniais considered for decarbonization of the transportation and industrial sectors.Assumed electricity and hydrogen fuel input required for this clean fuel production is prescribed in Appendix C.The amount of electricity

194、needed for hydrogen production varies across the different scenarios based on two factors:the amount of hydrogen needed for direct and indirect use and the source of hydrogen production.Figure 3 shows the assumed demand for clean hydrogen required for processes where direct electrification can be mo

195、re challenging due to technical or economic reasonsfor example,long-haul freight,ships,aviation,and certain industrial processes.We assume 14.3 million tonnes(MT)of clean hydrogen demand in 2035for both direct use and as fuel input to other low-carbon fuelsfor these non-grid applications for scenari

196、os that use the ADE demand assumptions(see Appendix C);the LTS scenario assumes 7.5 MT of clean hydrogen demand by 2035.This demand for new clean hydrogen is distinct from the current demand for hydrogen(10 MT annually,largely derived from natural gas)to produce refined petroleum products and other

197、existing uses.14 In addition to providing direct end uses,we assume hydrogen can help meet electricity demand during periods of very low wind and solar output,providing a form of seasonal storage.This demand for additional clean hydrogen used by the grid is determined by the model used in this study

198、 and varies substantially depending on scenario;therefore,the total amount of hydrogen needed is an output of the model.We represent hydrogen production from natural gas steam methane reforming(SMR)with carbon capture and sequestration(CCS)and 14 Given the reduction in motor gasoline and diesel fuel

199、s in the transportation sector under the ADE and LTS scenarios,this demand would likely decline over time.Hydrogen production to serve these needs is also expected to transition to lower-emissions pathways given the overall economywide decarbonization trajectory modeled under the scenarios.These dyn

200、amics are not modeled explicitly in our analysis.AEO 2021 ReferenceADELTS2020202520302035202020252030203520202025203020350200040006000Electricity Consumption TWhTransportationCommercialResidentialIndustry6 This report is available at no cost from the National Renewable Energy Laboratory at www.nrel.

201、gov/publications.from water electrolysis using clean electricity.The selection of one of these two hydrogen production pathways is determined by the model based on the input technology cost and performance assumptions and scenario specifications(e.g.,SMR with CCS is not allowed in the No CCS scenari

202、os).Hydrogen production,biofuels,and other clean fuels are discussed further in Section 2.2(page 8)and Section 3.6(page 33).Figure 3.Assumed clean hydrogen demand for industry and transportation.Additional hydrogen,not shown here,is used for seasonal storage,so total hydrogen production is substanti

203、ally higher than shown due to use in the electric sector and is summarized in Section 3.6.Electrification of end uses changes both how much electricity is used and when it is used.Peak demand in most of the United States currently occurs during summer afternoons,driven by air conditioning,but electr

204、ification of heating may shift peak demand to the winter.Our scenarios assume significant deployment of air-source heat pumps,which enable electrification of space and water heating in buildings and are more energy efficient than fossil-fuel-based heating options.However,widespread electrification o

205、f space heating could substantially raise evening winter peaks and reduce the ability of solar energy to meet demand peaks.15 Figure 4 shows the demand profiles for the entire United States by season as modeled in 2035.The ADE case(top)has much greater electrification;hourly peaks are 35%higher in w

206、inter than in summer,and more than double the overall U.S.peak demand in 2020.The LTS sensitivity 15 The modeled electric demand profiles of heat pumps are based on air-source heat pumps(see Appendix C).Alternatively,significant adoption of geothermal heat pumps would reduce winter peaksand the corr

207、esponding amount of capacity needed to meet these peaks.Further research is needed to compare the relative benefits and costs of different end-use technologies,including different heating options.Note:demand does not includecurrent industrial H2 consumption(10 MT annually)05520302035Non-p

208、ower sector H2 demand MTHydrogen demandADELTS7 This report is available at no cost from the National Renewable Energy Laboratory at www.nrel.gov/publications.with lower buildings electrification is assumed to have similar load shapes to current patterns,with the most significant increase in demand b

209、eing electric vehicles,which do not show a strong seasonal pattern(Muratori and Mai 2020).Note that the demand profiles analyzed are based on a historical(2012)weather year only.16 Further research is needed on changing demand patterns resulting from electrification and climate change,and the associ

210、ated increase in frequency and severity of extreme weather events.Figure 4.U.S.load profiles for seasonal peak days in the ADE scenario(top row)show how electrification increases peak demand and shifts the peak from summer to winter in 2035 due to electric space heating.Panels show the projected hou

211、rly load for the day of each season in which the full U.S.hourly load reaches its seasonal peak(solid lines)and the average seasonal profile(dashed lines).The LTS scenario(bottom row)has lower electrification and increased efficiency,resulting in lower demand and a continuation of summer peaks.16 Th

212、e AEO 2021 scenario uses historical demand profiles for all years(based on 2012 weather,equipment stock,and behavior).8 This report is available at no cost from the National Renewable Energy Laboratory at www.nrel.gov/publications.The tremendous growth in electrification in the ADE demand trajectory

213、 provides opportunities for flexible demand to help integrate variable generation by shifting timing of electric vehicle charging,hydrogen production,and certain other loads.Power systems have long relied on demand-side resourcesincluding demand response,interruptible load,and demand flexibilityto m

214、eet some of these needs,as the cost of dropping noncritical loads can be less than that of building new generation capacity with low capacity factor.Policy mechanisms to incentivize demand flexibility include rate structures(e.g.,real-time or time-of-use pricing,demand charges)and direct incentives

215、to consumers.The default demand flexibility is derived from the Base Flexibility Case in NRELs Electrification Futures Study(NREL 2022b),which results in about 5%of annual demand being flexible in 2035.This is supplemented with a“super peak”demand response assumption that clips the peak of the top 1

216、%of hours,which reduces 2035 peak coincident demand for the contiguous United States by 246 GW in cases using ADE load profiles(this is captured in Figure 4 above,with unclipped profiles and additional details provided in Appendix C).Additional demand flexibility could potentially reduce the cost of

217、 achieving the high levels of reliability assumed here,and significant research is needed to assess the potential role of flexible loads for grid and energy-system decarbonization(Zhou and Mai 2021;Sun et al.2020).2.2 Scenario Framework The second step in the analysis process is to understand the su

218、pply-side(generation)resources needed to meet the demand profiles created in Step 1 in Figure 1.Given the significant uncertainty about the future cost and performance of low-and zero-carbon technologies,four primary technology evolution scenarios with varying assumptions regarding resource costs an

219、d technology availability were evaluated.Table 1 summarizes the four core scenarios,which are described below.In each scenario,assumptions common to all scenarios are called“reference,”and details are provided in Appendix C.All Options is a scenario in which all technologies continue to see improved

220、 cost and performance consistent with the National Renewable Energy Laboratorys(NRELs)Annual Technology Baseline(NREL 2021).This scenario includes the development and deployment of direct air capture(DAC)technology.The other three main scenarios assume DAC does not achieve the cost and performance t

221、argets needed to be deployed at scale,but they do consider availability of DAC in sensitivity cases.17 Infrastructure Renaissance assumes improved transmission technologies and new permitting and siting approaches that allow greater levels of transmission deployment with higher capacity.17 Executive

222、 Order 14057 defines“carbon pollution-free electricity”as“electrical energy produced from resources that generate no carbon emissions,including marine energy,solar,wind,hydrokinetic(including tidal,wave,current,and thermal),geothermal,hydroelectric,nuclear,renewably sourced hydrogen,and electrical e

223、nergy generation from fossil resources to the extent there is active capture and storage of carbon dioxide emissions that meets EPA requirements”.The inclusion of non-generation,negative emission technologies such as direct air capture may not be consistent with the Administrations 2035 clean electr

224、icity goal but are considered in the studys All Options Scenarios because of their potential deployment,emissions,and cost impacts.https:/www.federalregister.gov/documents/2021/12/13/2021-27114/catalyzing-clean-energy-industries-and-jobs-through-federal-sustainability 9 This report is available at n

225、o cost from the National Renewable Energy Laboratory at www.nrel.gov/publications.Constrained is a scenario where additional constraints to deployment of new generation capacity and transmission both limits the amount that can be deployed and increases costs to deploy certain technologies.No CCS sce

226、nario assumes CCS technologies do not achieve the cost and performance needed for cost-competitive deployment.This scenario also acts as a point of comparison to demonstrate the potential benefits of achieving cost-competitive deployment of CCS at scale.This is the only scenario that includes no fos

227、sil fuel capacity or generation in 2035,and therefore it is the only scenario that includes zero direct GHG emissions in the electric sector.The four primary technology evolution scenarios all reach 100%clean electricity by 2035 and use the ADE demand trajectory,with the LTS trajectory as a primary

228、sensitivity.An additional 122 sensitivities with the same 100%by 2035 constraint were also evaluated across all four core scenarios.Major sensitivity categories are shown at the bottom of Table 1 and described in detail in Appendix A.18 18 This amounts to roughly 44 sensitivities per core scenario.N

229、ote that not all core scenarios have all sensitivities applied to them;for instance,we do not test a“low-cost CCS”sensitivity on the No CCS scenario.10 This report is available at no cost from the National Renewable Energy Laboratory at www.nrel.gov/publications.Table 1.100%Clean Electricity Scenari

230、os and Sensitivities Evaluated in This Study Scenario Demand Assumptions Generation Resource Assumptions Renewable Resources CCS Technologies Transmission Nuclear Other Infrastructure All Options ADE Reference All including DAC Reference interregional AC expansion Reference Reference Infrastructure

231、Renaissance No DAC HVDC macrogrid Lower-cost transport and storage for H2,CO2,biomass Constrained Reduced land available for wind,solar,and biomass Intraregional transmission only,higher(5x)costs Not allowed in regions with current legislative restrictions Higher-cost transport and storage for H2,CO

232、2,biomass No CCS Reference No CCS,bioenergy with CCS,or DAC Reference Reference Reference Sensitivities(applied to each of the four core scenarios)AEO and LTS demand cases.Supply-side sensitivities include renewable energy costs,storage costs,nuclear costs,electrolyzer costs,CCS cost and performance

233、,transmission constraints,new natural gas restriction,natural gas fuel costs,expanded biomass supply,low-cost geothermal,and allowing DAC in the Infrastructure Renaissance and Constrained cases.In addition,“current policy”reference cases with both the AEO19 and ADE demand trajectories(referred to as

234、 Reference-AEO and Reference-ADE respectively)are also modeled under each of the four core scenarios from Table 1.20 19 The AEO demand is derived from the 2021 Annual Energy Outlook(EIA 2021a).20 This produces two reference cases for each of the four core scenarios.11 This report is available at no

235、cost from the National Renewable Energy Laboratory at www.nrel.gov/publications.The study uses NRELs Regional Energy Deployment System(ReEDS)model to simulate planning and operation of the U.S.21 electric power system(Ho et al.2020),while considering the significant geographical variation in demand

236、and generation resource availability.22 ReEDS examines the demand for electricity at various points on the grid,as well as the cost and performance of multiple generation technologies that can meet this demand,and selects the mix of technologies that produces the lowest power system cost under the v

237、arious assumptions applied in each scenario.The model performs resource adequacy calculations to ensure the total capacity(and availability)of all resources is sufficient to meet the peak demand for electricity during hot summer afternoons,when air conditioning demand is at its highest,or cold winte

238、r nights and mornings,when heating demandincreasingly served by electric heat pumpsis at its highest.These calculations require evaluating the availability(i.e.,capacity credit)of all resources using hourly resource and load data in the model,and how this capacity credit is impacted as the generatio

239、n mix evolves(see Text Box 1 and Text Box 2 at the end of this section for discussion of capacity credit and related terms).23 The model also enforces the need for operating reserves used to address random and unpredictable variations in supply or demand,including power plant outages.In this study,t

240、he model simulates the evolution of the power grid for each year from 2020 through 2035,using projections for electricity demand patterns in multiple regions.24 The model uses historical weather data,which essentially assumes future years will have similar weather to the recent past and therefore do

241、es not account for the potential impacts of climate change on demand or renewable energy availability profiles,including potential increased frequency and severity of extreme weather events.Changing land use patterns from climate change and other factors(which impact renewable energy supply)are also

242、 not included.The model can select from among 14 technology categories,summarized in Table 2,to produce a least-cost resource mix.All technologies except distributed grid-connected rooftop photovoltaics(PV)are considered as part of the ReEDS optimization.Rooftop PV deployments and corresponding gene

243、ration profiles are an input to the ReEDS model using values from the Distributed Generation Market Demand(dGen)model(Sigrin et al.2016)(additional 21 Throughout this report,we refer to the U.S.power system as a shorthand for the contiguous United States,which excludes Alaska and Hawaii.22 ReEDS use

244、s hourly data to represent contributions to resource adequacy,but it relies on simplified dispatch modeling given the long-term investment decision-making scope of the model.More detailed grid simulations were not executed for this study,but paired use of ReEDS and production cost and probabilistic

245、resource adequacy modeling has been employed in multiple prior studies that examined very high renewable energy systems with ReEDS(e.g.,the North American Renewable Integration Study Brinkman et al.2021,Solar Futures Study Ardani et al.2021).Results from these previous studies act as a partial valid

246、ation of the ReEDS resource adequacy calculations.However more detailed analysis will be needed to complete evaluate the adequacy and operational reliability of the modeled scenarios.23 A more thorough discussion of how ReEDS treats capacity credit is provided in the full model documentation(Ho et a

247、l.2020).24 The model is run through 2050,but we report results to 2035 for most cases.After 2035,the model is run using 5-year timesteps.12 This report is available at no cost from the National Renewable Energy Laboratory at www.nrel.gov/publications.discussion of distributed resources is provided i

248、n the Text Box 4 in Section 3.8).25 Unless otherwise noted,technology cost and performance data are derived from the 2021 Annual Technology Baselines Moderate case(NREL 2021).Appendix B provides details about the assumptions for the various technologies,including capital and variable costs,performan

249、ce,and data sources,which describe how the model considers the regional and temporal variations in the output of wind and solar.The model outputs the total capacity of each resource in each region over time,demonstrating the growth rates required for the various technologies and their annual energy

250、contribution in each year.Table 2.Supply-Side Technologies Considered Type Technology Notes Renewable Generation Utility-scale PV Ground-mount performance reflects one-axis tracking Rooftop PV Rooftop systems;quantity and locations externally generated from the dGen model using the deployment levels

251、 across all scenarios Hydropower Adding generators unpowered dams and uprates at existing facilities Land-based wind Range of turbines(hub heights)and rotor diameters depending on region and scenario Offshore wind Includes both fixed-bottom and floating technologies Nuclear Conventional light-water

252、reactors;can be deployed in all states except in the Constrained scenario;lower-cost advanced nuclear considered in sensitivity cases26 Geothermal Multiple geothermal technologies Concentrating solar power(CSP)Includes thermal storage,including options with 10 and 14 hours of capacity Fossil Generat

253、ion(without CCS)Natural gas combustion turbine Standard natural gas fueled technology Natural gas combined cycle(CCGT)Standard natural gas fueled technology;higher cost but more efficient than CT Conventional coal Standard pulverized coal technology Storage Hydrogen-fueled CT/seasonal storage New or

254、 retrofit of existing(CT and CCGT)plants;provides seasonal storage 25 Rooftop PV trajectories were identical for all ADES scenarios and derived from the Solar Futures Study(Ardani et al.2021).Trajectories for the AEO and LTS scenarios were derived from the Standard Scenarios 2021 Mid-case(Cole et al

255、.2021).26 Advanced nuclear represents an unspecified mix of next-generation technologies with reduced cost,as discussed in the main body and Appendix.Nuclear is used for production of electricity only.While its heat and electricity can be used for hydrogen generation in the overall grid,dedicated hy

256、drogen production via thermal cycles is not considered.13 This report is available at no cost from the National Renewable Energy Laboratory at www.nrel.gov/publications.Type Technology Notes Storage(2-to 10-hour duration)Cost and performance based on Li-ion batteries Storage(12-hour duration)Cost an

257、d performance based on pumped storage hydro Natural gas CC with carbon capture and storage(CCS)New plants with a 90%capture rate;retrofit of existing plants evaluated in a sensitivity case;higher capture rates(95%and 99%)evaluated in sensitivity cases Coal with CCS New or retrofit of existing plants

258、 with a 90%capture rate;higher capture rates(95%and 99%)evaluated in sensitivity cases Carbon Capture or Negative Emissions Bioenergy with CCS(BECCS)Uses woody biomass and produces net negative emissions of about-1.2 tonnes/MWh Direct air capture(DAC)Not a generation or storage technology,but it use

259、s electricity to capture and store carbon from the atmosphere,producing negative emissions.Use of fossil technologies without CCS and incomplete capture in CCS plants requires offsets with either DAC or BECCS.The CO2 constraint must be met in each year;there is no banking or borrowing of emissions b

260、etween years.Other Hydrogen production Hydrogen can be produced from either electrolysis(assuming 5155 kWh per kg)or via steam methane reforming with CCS(requiring 0.2 MMBtu natural gas and 2 kWh electricity per kg)Transmission Model can deploy conventional alternating-current(AC)lines in all scenar

261、ios.In the Infrastructure Renaissance scenario,the model can also choose from two direct-current(DC)technologies.The model includes representation of existing and new transmission,including the associated costs and benefits.New high-capacity alternating-current(AC)and high-voltage direct current(HVD

262、C)(where allowed)transmission may be added starting in 2026,reflecting lead time for siting,permitting,and construction.The cost of the types of new transmission allowed varies by scenario,as discussed in Section 3.7(page 43)and Appendix B.In addition to different transmission siting restrictions,th

263、e scenarios consider varying restrictions on the siting of new wind and solar to reflect development challenges or other factors that may impact deployment.The All Options,No CCS,and Infrastructure Renaissance scenarios use“reference”siting availability while the Constrained scenario uses“limited”av

264、ailability(Lopez et al.2021).The pathway to the 2035 clean electricity target is modeled via a national constraint on annual net CO2 emissions from the electric sector,starting in 2023.Figure 5 illustrates historical emissions(EIA 2022b)and the clean electricity trajectory met by the model.The core

265、100%scenarios all follow the same trajectory,shown in green,which reaches an 80%reduction in 2030 relative to 2005.This trajectory is not intended to represent a specific policy under 14 This report is available at no cost from the National Renewable Energy Laboratory at www.nrel.gov/publications.co

266、nsideration but instead considers a general decarbonization pathway,and it follows the general trend seen from 2010 to 2020.Existing policies,including state clean energy targets(including trajectories)and federal tax credits(including currently legislated ramp-downs and expirations),are included,bu

267、t no additional policies are evaluated.Figure 5.Electricity-sector emissions pathways for 100%clean electricity scenarios.Emissions reductions are defined relative to 2005(solid dot).Historical emissions measure only direct CO2 emissions from combustion,while modeled trajectories also include the CO

268、2 equivalent emissions from estimated methane leakage.We account for CO2 emissions in the electricity sector associated with direct combustion,as well as the global warming potential of upstream methane leakage in the natural gas production and delivery system for gas usage in electricity generation

269、 and hydrogen production.27 We assume a leakage rate of 2.3%(Alvarez et al.2018),which drops to 1.6%by 2030 to reflect a reduction in methane emissions of 30%by the end of the decade relative to their 2020 level(Mason and Alper 2021).We also account for incomplete capture in fossil CCS plants.CCS is

270、 allowed in new or retrofit applications starting in 2026 in all scenarios except the No CCS scenario.Scenarios with CCS assume a 90%capture rate from CCS plants(with additional sensitivities with higher capture rates),that captured CO2 is permanently stored in geologic formations,and that no leakag

271、e of CO2 occurs.However,the model does not account for all life cycle GHG emissions,including those from plant construction,operation and maintenance,and fuel processing and transport.Enhanced oil recovery or other applications that would impact the net carbon reduction benefits are not considered.A

272、ny net emissions from fossil plants without CCS or from less than 100%capture rates(and upstream methane leakage)must be offset by negative emissions technologies such as BECCS 27 CO2 equivalent emissions from upstream methane are sensitive to assumptions regarding leakage rate and the time horizon

273、for methane global warming potential.Other life cycle emissions(often with considerable uncertainty)are not included here,including methane from hydropower,biomass net emissions,CO2 leakage from CCS,and other emissions.Methane leakage is not included in emissions estimates for transportation or resi

274、dential/commercial/industrial end-use applications,or in historical(pre-2021)estimates of electricity sector emissions.15 This report is available at no cost from the National Renewable Energy Laboratory at www.nrel.gov/publications.and DAC.Multiple other negative emissions pathwayssuch as afforesta

275、tion,biomass carbon removal and storage,biorefining with CCS,enhanced mineralization,and ocean-based carbon dioxide removalcan also be used to offset emissions from fossil plants,but they are not modeled.BECCs results in a net negative emissions rate because carbon from the atmosphere is captured du

276、ring photosynthesis and then sequestered after combustion.28 The use of negative emissions technologies allows for continued operation of fossil plants without CCS even at 100%clean electricity because of the ability to achieve net-zero emissions.2.3 Costs and Benefits The final step in the analysis

277、 process is to evaluate the costs and benefits of the various 100%clean electricity scenarios.Three cost and benefit components were evaluated,and all monetary values are reported in 2021 U.S.dollars.In general,most comparisons are made between the 100%scenarios and the Reference-ADE scenario.Compar

278、isons between the 100%scenarios and the Reference-AEO scenario would likely include significant additional benefits associated with electrification,but since the costs of these electrification measures are not estimated,here would be limited context provided for these additional benefits.The first c

279、ost and benefit component is associated with bulk power system expenditures,including capital and operating costs for generation,storage,and transmission.29 We calculate the average system cost for each year,which is the annualized cost of these components divided by annual electricity demand.This i

280、s somewhat analogous to a wholesale cost of electricity and does not include several other costs(e.g.,distribution and administration)and therefore does not reflect the total cost seen by consumers and cannot be directly compared with retail electricity prices.30 Nonetheless,the reported average sys

281、tem cost provides an internally consistent measure to compare how estimated costs change over time and between scenarios,and these cost differences can be compared with historical changes in retail prices.The average system cost is calculated for each yearin real terms and on an undiscounted basis.S

282、eparately,cumulative net present values of costs and benefits between 2023 and 2035 are also calculated by summing and discounting the total annual costs,using a 2.5%real discount rate based on the Interagency Working Group on the Social Cost of Greenhouse Gases(IWG)(IWG 2021).Results using other di

283、scount rates are provided in Appendix A;changing the discount rate does not qualitatively change the findings.The second component calculated is associated with mortality from long-term exposure to fine particulate matter(PM2.5)from fossil fuel combustion in the electric sector.For each case,we 28 B

284、ecause we do not consider regional variations in BECCS fuel type,we assume a uniform fuel cost and emissions rate for all BECCS plants.29 It also includes the capital and operating costs for DAC and hydrogen production(e.g.,electrolyzers).The value of hydrogen used in non-grid applications is based

285、on the annual marginal cost of hydrogen production and is accounted for as a revenue for the electricity system cost metrics.30 The cost metric used reflects the average costs of electricity,in contrast to marginal costs reflected in wholesale prices from power markets,which also do not include all

286、grid services.Therefore,costs reported here cannot be directly compared with historical energy prices from restructured power markets.16 This report is available at no cost from the National Renewable Energy Laboratory at www.nrel.gov/publications.used estimates of the mortality risk per tonne of em

287、issions from three reduced complexity air quality models(AP2,EASIUR,and InMAP).31 Each of these models estimates PM2.5 formation associated with emissions of precursor pollutants(NOx and SO2).To generate annualized premature mortality,the models apply concentration response function models from two

288、studies which link exposure to PM2.5 to increased mortality risk.These are abbreviated ACS(from an American Cancer Society Study)and H6C from the Harvard Six-Cities study(Gilmore et al.2019).Total emissions are multiplied by the mortality risk per tonne of pollutant to get total mortality,with a ran

289、ge of results across the different combinations of air quality model and concentration response function.We show the mortality results for the entire set of models,but for the final benefit-cost analysis we use the most conservative model which combined ACS and EASIUR,which produces the lowest morta

290、lity rates and therefore the lowest benefit associated with cleaner air.Annual premature deaths from air pollution are translated into a monetary value by applying a value of a statistical life,using the U.S.Environmental Protection Agencys estimate of$7.4 million in 2006 dollars(EPA 2022)inflated t

291、o a present-day dollar value($9.9 million in 2021 dollars).These costs can then be translated into costs per unit of generation and cumulative(discounted)cost over the study period.Air quality benefits associated with electrification or in other sectors(such as transportation,buildings,and industry)

292、are also not considered.The third component is associated with damages from GHG emissions and is evaluated using the social cost of carbon(SCC).The SCC represents an estimate of the future damages of climate change caused by a marginal increase in GHG emissions.It is commonly measured in terms of co

293、st per unit of emissions(e.g.,$/tonne CO2).This estimate requires several major steps,each with significant uncertainty.First,it requires estimating the future emissions trajectory and the response of global climate to additional emissions.Then it requires estimating the impacts of various degrees o

294、f climate change on reduced agricultural output,human health(mortality/morbidity),economic growth,and other factors.Finally,these impacts must be translated into a total economic value,considering many components,such as the value of a statistical life and the degree to which future impacts should b

295、e discounted.The choice of discount rate is contentious,but it is generally agreed that intergenerational discount rates should be lower than discount rates used for shorter-term decision-making(IWG 2021).The IWG produced SCC estimates in 2013(IWG 2021);later values from 2016 and 2021 update the ori

296、ginal 2013 values based on inflation.Recent studies report higher SCC estimates than the IWG;updated estimates of human mortality(Bressler 2021)and uncertain but potentially catastrophic climate“tipping points”(Cai et al.2016)could increase the SCC further.Figure 6 shows examples of several SCC valu

297、es,demonstrating a large difference between the highest and lowest value for emissions occurring in 2020.Given the large uncertainty in SCC,we present costs using two values.The lower estimate value uses the IWG(2.5%discount rate)value,which is about$80/tonne in 2020 and increases to about$100/tonne

298、 in 2035(teal line).The higher estimate uses a constant value of$275/tonne(green line)from Pindyck(2019).Total emissions in each scenario are multiplied by the marginal SCC for a total cost impact in each year.As with 31 Additional description of the models and the data sources used for this study i

299、s provided at https:/www.caces.us/data.17 This report is available at no cost from the National Renewable Energy Laboratory at www.nrel.gov/publications.the other two cost components(direct electricity costs and health costs),SCC can then be translated into costs per unit of generation and cumulativ

300、e(discounted)cost over the study period.Figure 6.Examples of the large range in estimates of SCC.Values in parentheses indicate the SCC from the relevant source in 2035.Finally,we estimate the benefit-cost ratio over the study period.The benefits are calculated as the sum of(1)the difference in cumu

301、lative health impacts between the 100%clean electricity scenarios and the Reference-ADE scenarios and 2)the difference in cumulative SCC impacts between the 100%clean electricity scenarios and the Reference-ADE scenarios.The cost is the difference in cumulative power system costs between the 100%cle

302、an electricity scenarios and the Reference-ADE scenarios.Overall,a limited set of costs and benefits are evaluated in this work,as highlighted in the following section.2.4 Key Caveats Although ReEDS is designed to model many aspects of the power grid,the large scope of the model necessitates simplif

303、ications.One limitation of ReEDS is that its scope is limited to the bulk power system and the model does not directly consider an economywide optimization.The ADE trajectory assumes electrification plays a very large role,consistent with other recent literature.But there is uncertainty about the de

304、gree of electrification,which this study does not seek to resolve.The LTS scenario represents a demand pathway that does not use as much electrification and can be a proxy for any other non-electrification heavy scenario of decarbonization.But additional economy-wide analysis would be required to as

305、sess optimal portfolios across the entire economy.For example,the non-electricity costs and benefits associated with electrification and demand-side changes(e.g.,costs of electric vehicles and avoided gasoline expenditures)are outside the study scope.Similarly,we do not include analysis of the evolv

306、ing workforce needs for the transition described in this work,or how some clean 18 This report is available at no cost from the National Renewable Energy Laboratory at www.nrel.gov/publications.energy pathways may be more compatible with the existing workforce.We also do not consider repurposing exi

307、sting fossil infrastructure that could reduce system costs,beyond retrofits of existing generators to run on clean hydrogen.Other economy-wide impacts such as manufacturing requirements and trends,and international trade balance are not considered.Broader national and regional benefits,such as natio

308、nal security and many aspects of environmental justice,including distributional aspects of costs and benefits,are also not considered.Within the power sector,this study does not consider the costs or impacts related to changes that may occur on the distribution system;for this study,a single,predete

309、rmined projection is used to specify rooftop PV capacity by region and year in all ADE scenarios explored.Additional research is needed to understand the opportunities and operational impacts of widespread deployment of distributed generation in 100%scenarios.Electricity demand profiles and demand f

310、lexibility are also determined outside the model framework.There are many factors that could result in substantial changes in electricity demand patterns not considered here.These include the impacts of climate change and extreme weather,changing work patterns resulting from the COVID-19 pandemic,an

311、d other social and macroeconomic factors.Though ReEDS considers a large range of supply-side technologies,it does not represent all possible technologies that may be important in decarbonized energy systems.In particular,it represents a small subset of possible energy storage technologies and fuel p

312、roduction pathways,and it does not include all generation technologies that might be deployed by 2035.Therefore,results should be interpreted as representative but not determinate,as a variety of solutions may be cost-competitive.Though the model includes a variety of factors that can restrict devel

313、opment of individual technologies,including cost,siting restrictions,access to transmission,and contribution to resource adequacy,it does not consider limits to growth that could result from supply chain issues,financing,and local or regional factors.It also does not consider the interaction of reso

314、urce limits that could result from competition from other sectors,such as the supply of critical materials.Likewise,fossil fuel prices are based on AEO projections and do not consider the additional impact of significant reduction in the demand for these fuels in the 100%clean electricity scenarios.

315、Like all national-level models,ReEDS does not model specific transmission rights-of-way with detailed AC power flow simulation;transmission is modeled as aggregated regional transfer capacities with controllable flow.32 Additional intraregional network reinforcement would also be needed given the hi

316、gh degree of electrification that is assumed here,but this is not modeled in ReEDS.Though ReEDS performs detailed calculations to estimate the ability of the various scenarios to provide resource adequacy and operating reserves,it does not perform a comprehensive assessment of all aspects of reliabi

317、lity and resilience.Future analysis using detailed unit-commitment,economic dispatch,probabilistic outages,and contingency analysis will be needed to validate findings from this work.Lastly,ReEDS applies a systemwide least-cost planning approach across all technologies that may not fully reflect inv

318、estment decisions 32 This results in increased utilization of transmission assets compared to todays grid,which could be accomplished using flexible AC transmission components(see Text Box 3 in Section 3.7.)19 This report is available at no cost from the National Renewable Energy Laboratory at www.n

319、rel.gov/publications.made in response to competitive wholesale markets or regional,state,and local planning decisions.The ReEDS model is used to calculate a variety of costs and benefits represented in Step 3(Figure 1,page 3).The primary cost metric considers impacts on the bulk system,but because i

320、t does not evaluate the distribution network,total costs seen by end consumers are not calculated.Analysis of overall energy burden and electricity rate impacts,including analysis of new rate structures that could potentially unlock demand flexibility,will be an important component of future work wh

321、en assessing policy options.Benefits analysis includes direct energy-sector GHG emissions that result from fossil-fuel combustion and methane leakage but does not include other life cycle GHG emissions associated with electricity production or emissions from agriculture and other land use(except BEC

322、Cs).The benefits analysis associated with improved air quality is also limited to premature mortality from only the electric sector and therefore does not assess the benefits from emissions reductions in other sectorssuch as industry and transportationor include other benefits such as reduced morbid

323、ity,changes to hospitalizations,or ecosystem damage.Previous work has found that accounting for mortality results in the largest component of monetized benefits(EPA 1999;NRC 2010)and that PM2.5 exposure is the driver of 90%95%of all mortalities related to air pollution(Tessum,Hill,and Marshall 2017;

324、Tschofen,Azevedo,and Muller 2019).So,we likely capture most of the monetized benefits related to air quality improvements related to power sector decarbonization with this method,though other additional benefits not estimated here may have more salience in particular communities.In addition,while so

325、me aspects of land use changes are modeled,several other environmental impacts,such as potentially reduced water use or localized ecosystem changes,are not.Text Box 1.Capacity-Related Terms Used in This Report Capacity(also“nameplate capacity”or“peak capacity”)generally refers to the rated output of

326、 a power plant when operating at maximum output.The capacity of individual power plants is typically measured in kilowatts(kW)or megawatts(MW).The cumulative capacity of systems is often measured in gigawatts(GW)or terawatts(TW).Capacity of power plants is typically measured by their net AC rating,a

327、nd we use this standard in this report.Capacity factor(%)is a measure of how much energy is produced by a plant compared to its maximum output.It is calculated by dividing the total energy produced during some period by the amount of energy it would have produced if it ran at full output over that s

328、ame period.Capacity credit is a measure of the contribution of a power plant to resource adequacy,meaning the ability of a system to reliably meet demand during all hours of the year.It is measured either in terms of capacity(kW,MW)or as the fraction of its nameplate capacity(%),and it indicates the

329、 amount or portion of the nameplate capacity that is reliably available to meet load during times of highest system stresstypically the highest net-load hours of the year.Capacity credit may also be referred to as capacity value,but the latter term sometimes refers to the monetary value of physical

330、capacity(Mills and Wiser 2012).Firm capacity refers to generation capacity with high capacity credit.Note that capacity credit,and therefore the ability of a resource to provide firm capacity,may vary substantially as a function of location and amount of deployment.For example,the capacity credit of

331、 a PV plant may be high at low levels of deployment(meaning it can provide significant firm capacity)but drop as increased solar deployment results in demand peaks that shift to periods later in the day or across seasons.Likewise,the ability of storage to provide firm capacity can change substantial

332、ly depending on the mix of resources deployed.20 This report is available at no cost from the National Renewable Energy Laboratory at www.nrel.gov/publications.Text Box 2.The Three Rs:Resource Adequacy,Operational Reliability,and Resilience There are multiple aspects to maintaining a reliable power

333、grid.Resource adequacy represents planning for the systems ability to supply enough electricityat the right locationsto keep the lights on,even during extreme weather days and when“reasonable”outages occur.An adequate system has sufficient spare capacity to replace capacity that fails or is out of service for maintenance.Resource adequacy is measured by the probability of an outage over an extende