工业深度脱碳：建模方法和数据挑战（英文版）--未来能源研究所VIP专享VIP免费

下载本文档

阅读 700
格式 pdf
大小 7.46 MB
约82页
2023-09-05
收藏
点赞(0)
海报
举报

Industrial Deep Decarbonization: Modeling Approaches and Data Challenges A

Industrial Deep

Decarbonization: Modeling

Approaches and Data

Challenges

Elena Verdolini, Lorenzo Torreggiani, Sara Giarola, Massimo Tavoni,

Marc Hafstead, and Lillian Anderson

Report 23-10

August 2023

Resources for the Future i

About the Authors

Elena Verdolini is a climate economist. She is Professor of Political Economy

at the Law Department, University of Brescia, and Senior Scientist at the RFF-

CMCC European Institute on Economics and the Environment (EIEE) of the Euro-

Mediterranean Center on Climate Change, where she leads the research group on

“Sustainable Innovation and Digitalization”. She is the principal investigator of the

2D4D “Disruptive Digitalization for Decarbonization” project, funded by the European

Research Council through a Starting Grant, and the coordinator of the AdJUST

“Advancing the understanding of challenges, policy options and measures to achieve a

JUST EU energy transition” funded by Horizon Europe. She was a Lead Author of the

of the IPCC 6th Assessment Report, Working Group III. Elena holds a degree in Political

Science from the University of Pavia, a Master of Public Administration and a Master

of Arts in International Studies from the University of Washington, Seattle and a PhD

in Economics and Finance of the Public Administration from the Catholic University of

Milan. Her research interest includes the dynamics and drivers of innovation, adoption

and transfer of energy-eicient and climate-friendly technologies; the role of digital

technologies in the energy transition; and the economic and distributional implications

of climate and energy policy.

Lorenzo Torreggiani is a Post Degree researcher at EIEE working on renewables’ raw

data analysis. He is a student in Green Economy and Sustainability with a degree in

Banking and Finance at the University of Brescia. He is interested in climate change,

renewable energy, environmental economics, and statistics.

Sarah Giarola is the Marie Curie Research Fellow at the Polytechnic University of

Milan. She works on modelling the technological, epistemic, and socio-economic

uncertainty in climate economic models using machine learning techniques. She

received her PhD in Chemical Engineering from the University of Padova in 2012.

She joined Imperial College London in 2012, became a BG Research Fellow in 2014,

at the Sustainable Gas Institute, and in 2020, she joined the Chemical Engineering

Department. Her research interests lie in macro-energy systems with a focus on

the multiscale modelling to capture the nexus between technology diusion and

societal behaviour. In the area of energy systems modelling and optimisation, she

published 36 journal articles and 3 book chapters.

Industrial Deep Decarbonization: Modeling Approaches and Data Challenges ii

Massimo Tavoni is director of EIEE. He is also a full professor at the School

of Management of Politecnico di Milano. He coordinated the Climate Change

Mitigation programme at Fondazione Eni Enrico Mattei (FEEM) between 2015

and 2018. He has been a fellow at the Center for Advanced Studies in Behavioural

Sciences at Stanford University and a postdoctoral fellow at Princeton University

His research is about climate change mitigation policies, and has appeared in major

scientific journals. He is a lead author of the IPCC (5th and 6th assessment reports),

co-directs of the International Energy Workshop, and was deputy editor for the

journal Climatic Change. He is a recipient of a grant from the European Research

Council ERC. He has advised several international institutions on climate change

matters, including the OECD, the Asian Development Bank, the World Bank.

Marc Hafstead joined RFF in 2013 from Stanford University. He is an RFF fellow

and director of the Carbon Pricing Initiative and the Climate Finance and Financial

Risk Initiative. His research has primarily focused on the evaluation and design of

federal and state-level climate and energy policies using sophisticated multi-sector

models of the US economy. With Stanford Professor and RFF University Fellow

Lawrence H. Goulder, he wrote Confronting the Climate Challenge: US Policy

Options (Columbia University Press) to evaluate the environmental and economic

impacts of federal carbon taxes, cap-and-trade programs, clean energy standards,

and gasoline. His research has also analyzed the distributional and employment

impacts of carbon pricing and the design of tax adjustment mechanisms to reduce

the emissions uncertainty of carbon tax policies. His paper “Impacts of a Carbon

Tax across U.S. Household Income Groups: What Are the Equity-Eiciency

Trade-Os?” (with Larry Goulder, GyuRim Kim, and Xianling Long) won the Journal

of Public Economics 2021 Atkinson Award for best paper published in the journal

between 2018 and 2020. Hafstead has been cited in the popular press, including the

Wall Street Journal, the Washington Post, Axios, and CNNMoney.

Lillian Anderson is a research associate at RFF. She completed an undergraduate

degree in Economics and Mathematics and a master’s degree in Economic

Development Programming at the University of Southern California. Previously,

she worked with the Foresight and Policy Modelling team at the International Food

Policy Research Institution. Her focus is on working with, developing, and adapting

economic models.

/82

下载本文档

文本预览下载提示常见问题

IndustrialDeepDecarbonization:ModelingApproachesandDataChallengesAIndustrialDeepDecarbonization:ModelingApproachesandDataChallengesElenaVerdolini,LorenzoTorreggiani,SaraGiarola,MassimoTavoni,MarcHafstead,andLillianAndersonReport23-10August2023ResourcesfortheFutureiAbouttheAuthorsElenaVerdoliniisaclimateeconomist.SheisProfessorofPoliticalEconomyattheLawDepartment,UniversityofBrescia,andSeniorScientistattheRFF-CMCCEuropeanInstituteonEconomicsandtheEnvironment(EIEE)oftheEuro-MediterraneanCenteronClimateChange,wheresheleadstheresearchgroupon“SustainableInnovationandDigitalization”.Sheistheprincipalinvestigatorofthe2D4D“DisruptiveDigitalizationforDecarbonization”project,fundedbytheEuropeanResearchCouncilthroughaStartingGrant,andthecoordinatoroftheAdJUST“Advancingtheunderstandingofchallenges,policyoptionsandmeasurestoachieveaJUSTEUenergytransition”fundedbyHorizonEurope.ShewasaLeadAuthoroftheoftheIPCC6thAssessmentReport,WorkingGroupIII.ElenaholdsadegreeinPoliticalSciencefromtheUniversityofPavia,aMasterofPublicAdministrationandaMasterofArtsinInternationalStudiesfromtheUniversityofWashington,SeattleandaPhDinEconomicsandFinanceofthePublicAdministrationfromtheCatholicUniversityofMilan.Herresearchinterestincludesthedynamicsanddriversofinnovation,adoptionandtransferofenergy-efficientandclimate-friendlytechnologies;theroleofdigitaltechnologiesintheenergytransition;andtheeconomicanddistributionalimplicationsofclimateandenergypolicy.LorenzoTorreggianiisaPostDegreeresearcheratEIEEworkingonrenewables’rawdataanalysis.HeisastudentinGreenEconomyandSustainabilitywithadegreeinBankingandFinanceattheUniversityofBrescia.Heisinterestedinclimatechange,renewableenergy,environmentaleconomics,andstatistics.SarahGiarolaistheMarieCurieResearchFellowatthePolytechnicUniversityofMilan.Sheworksonmodellingthetechnological,epistemic,andsocio-economicuncertaintyinclimateeconomicmodelsusingmachinelearningtechniques.ShereceivedherPhDinChemicalEngineeringfromtheUniversityofPadovain2012.ShejoinedImperialCollegeLondonin2012,becameaBGResearchFellowin2014,attheSustainableGasInstitute,andin2020,shejoinedtheChemicalEngineeringDepartment.Herresearchinterestslieinmacro-energysystemswithafocusonthemultiscalemodellingtocapturethenexusbetweentechnologydiffusionandsocietalbehaviour.Intheareaofenergysystemsmodellingandoptimisation,shepublished36journalarticlesand3bookchapters.IndustrialDeepDecarbonization:ModelingApproachesandDataChallengesiiMassimoTavoniisdirectorofEIEE.HeisalsoafullprofessorattheSchoolofManagementofPolitecnicodiMilano.HecoordinatedtheClimateChangeMitigationprogrammeatFondazioneEniEnricoMattei(FEEM)between2015and2018.HehasbeenafellowattheCenterforAdvancedStudiesinBehaviouralSciencesatStanfordUniversityandapostdoctoralfellowatPrincetonUniversityHisresearchisaboutclimatechangemitigationpolicies,andhasappearedinmajorscientificjournals.HeisaleadauthoroftheIPCC(5thand6thassessmentreports),co-directsoftheInternationalEnergyWorkshop,andwasdeputyeditorforthejournalClimaticChange.HeisarecipientofagrantfromtheEuropeanResearchCouncilERC.Hehasadvisedseveralinternationalinstitutionsonclimatechangematters,includingtheOECD,theAsianDevelopmentBank,theWorldBank.MarcHafsteadjoinedRFFin2013fromStanfordUniversity.HeisanRFFfellowanddirectoroftheCarbonPricingInitiativeandtheClimateFinanceandFinancialRiskInitiative.Hisresearchhasprimarilyfocusedontheevaluationanddesignoffederalandstate-levelclimateandenergypoliciesusingsophisticatedmulti-sectormodelsoftheUSeconomy.WithStanfordProfessorandRFFUniversityFellowLawrenceH.Goulder,hewroteConfrontingtheClimateChallenge:USPolicyOptions(ColumbiaUniversityPress)toevaluatetheenvironmentalandeconomicimpactsoffederalcarbontaxes,cap-and-tradeprograms,cleanenergystandards,andgasoline.Hisresearchhasalsoanalyzedthedistributionalandemploymentimpactsofcarbonpricingandthedesignoftaxadjustmentmechanismstoreducetheemissionsuncertaintyofcarbontaxpolicies.Hispaper“ImpactsofaCarbonTaxacrossU.S.HouseholdIncomeGroups:WhatAretheEquity-EfficiencyTrade-Offs?”(withLarryGoulder,GyuRimKim,andXianlingLong)wontheJournalofPublicEconomics2021AtkinsonAwardforbestpaperpublishedinthejournalbetween2018and2020.Hafsteadhasbeencitedinthepopularpress,includingtheWallStreetJournal,theWashingtonPost,Axios,andCNNMoney.LillianAndersonisaresearchassociateatRFF.ShecompletedanundergraduatedegreeinEconomicsandMathematicsandamaster’sdegreeinEconomicDevelopmentProgrammingattheUniversityofSouthernCalifornia.Previously,sheworkedwiththeForesightandPolicyModellingteamattheInternationalFoodPolicyResearchInstitution.Herfocusisonworkingwith,developing,andadaptingeconomicmodels.IndustrialDeepDecarbonization:ModelingApproachesandDataChallengesiiiAcknowledgementsThisresearchwasconductedwithsupportfromBreakthroughEnergy.Theresultspresentedinthisreportreflecttheviewsoftheauthorsandnotnecessarilythoseofthesupportingorganization.ElenaVerdolinigratefullyacknowledgessupportfromtheEuropeanResearchCouncil(ERC)undertheEuropeanUnion’sHorizon2020researchandinnovationprogramme(Project“2D4D–DisruptiveDigitalizationforDecarbonization,”grantagreementNo853487).MassimoTavonigratefullyacknowledgessupportfromtheEnergyDemandchangesInducedbyTechnologicalandSocialinnovations(EDITS)projectfundedbytheMinistryofEconomy,Trade,andIndustry(METI),Japan.AboutRFFResourcesfortheFuture(RFF)isanindependent,nonprofitresearchinstitutioninWashington,DC.Itsmissionistoimproveenvironmental,energy,andnaturalresourcedecisionsthroughimpartialeconomicresearchandpolicyengagement.RFFiscommittedtobeingthemostwidelytrustedsourceofresearchinsightsandpolicysolutionsleadingtoahealthyenvironmentandathrivingeconomy.TheviewsexpressedherearethoseoftheindividualauthorsandmaydifferfromthoseofotherRFFexperts,itsofficers,oritsdirectors.SharingOurWorkOurworkisavailableforsharingandadaptationunderanAttribution-NonCommercial-NoDerivatives4.0International(CCBY-NC-ND4.0)license.Youcancopyandredistributeourmaterialinanymediumorformat;youmustgiveappropriatecredit,providealinktothelicense,andindicateifchangesweremade,andyoumaynotapplyadditionalrestrictions.Youmaydosoinanyreasonablemanner,butnotinanywaythatsuggeststhelicensorendorsesyouoryouruse.Youmaynotusethematerialforcommercialpurposes.Ifyouremix,transform,orbuilduponthematerial,youmaynotdistributethemodifiedmaterial.Formoreinformation,visithttps://creativecommons.org/licenses/by-nc-nd/4.0/.ResourcesfortheFutureivAbstractIndustrialenergyconsumptionrepresentsalmost40percentofcurrentglobaltotalfinalconsumptionandisstilldominatedbyfossilfuels.Inthispaper,wepresentkeydecarbonizationoptions—namelyfuelswitchingandelectrification,carbonefficiency,materialefficiency,carboncaptureandstorage,andcirculareconomypractices—andanalyzetheirpotentialfordecarbonizationinsixmainenergy-intensiveindustrialsectors:steel,cement,chemicals,lightmanufacturing,aluminum,andpulpandpaper.Wethendevelopaframeworktodistinguishamongthedifferentmodellingapproachestoindustrialenergydemandandemissions,withspecificfocusonthedatachallengesthatconstrainmodellingandthedifficultiesofmodellinginnovationandtechnologydiffusion.Wepresentthemostwidelyusedmodelsofindustrialenergydemandandemissionsandclassifythemalongthreekeydimensions:theanalyticalapproachunderlyingeachmodel,themethodologyusedtogeneratedecarbonizationpathways,andthegranularitywithwhichdifferentindustrialsectorsberepresented.Byhighlightingthestrengthsandweaknessesofavailabletoolsforindustrialemissionmodelling,wepointtonecessaryfuturemodeldevelopmenteffortsthatwouldgreatlyimprovetheabilitytodevelopdeepdecarbonizationpathwaysforindustry.IndustrialDeepDecarbonization:ModelingApproachesandDataChallengesvContents1.Introduction12.Energy-IntensiveIndustrySectors22.1.Steel22.2.Cement32.3.Chemicals32.4.LightManufacturing32.5.Aluminum32.6.PulpandPaper43.AvailableApproachesforIndustryDecarbonization43.1.FuelSwitchingandElectrification43.2.EnergyEfficiencyImprovementsinProductionProcess63.3.MaterialEfficiency73.4.CarbonCaptureandStorageTechnologies73.5.CircularEconomyApproaches84.DemandReductionandEnergyEfficiencyPotentialsofEnergy-IntensiveSectors104.1.Steel104.2.CementandConcrete124.3.Chemicals134.4.LightManufacturing154.5.AluminumandNonferrousMetals154.6.PulpandPaper165.DataonIndustrialEnergyDemandandEmissions175.1.LackofACommonDetailedStatisticsClassification175.2.LackofComprehensiveDataonEnergyDemandofDifferentEnergyCarriers175.3.LackofDetailedSectorialInformationonEnergyDemandattheLevelofDifferentProducts185.4.LimitedGeographicandTimeCoverageofDetailedDatabases185.5.DifficultyinLinkingDataonEmissionsandFuelInputs185.6.DifficultyinPredictingCostsandPerformanceofRadicallyNovelTechnologies195.7.LackofComprehensiveDataonMaterialandEnergyFlows19IndustrialDeepDecarbonization:ModelingApproachesandDataChallengesvi6.ApproachesforIndustrialEnergyDemandModeling206.1.AnalyticalApproach:Bottom-upandTop-downModels206.2.Methodology:SimulationandOptimizationApproaches216.3.GranularityofSectoralModeling227.TheModelingofInnovationandTechnologicalChangeandRelevantPolicies248.SpecificModelsforIndustryEnergyDemandandEmissions269.Conclusions27References29AppendixA.AvailableDatasetstoTrackIndustrialEnergyDemandandEmissions,andTheirLimitations39AppendixB.ClassificationofIndustrialSectors49B.1.TheInternationalStandardIndustrialClassificationofAllEconomicActivities(ISIC)50B.2.TheNorthAmericanIndustryClassificationSystem(NAICS)50B.3.TheStatisticalClassificationofEconomicActivitiesintheEuropeanCommunity(NACE)51AppendixC.ModelSummaries52AppendixD.ExamplesofApplicationsUsingDifferentModels72D.1.WorldEnergyModels72D.2.NEMS72D.3.GCAM72D.4.REMIND72D.5.MUSE72D.6.TIMES73D.7.IMAGE73D.8.MaterialEconomicsModelingFramework73D.9.E3ME73D.10.ISEEM-IS73D.11.U-ISIS74D.12.HYBTEP74D.13.FORECAST74ResourcesfortheFutureIndustrialDeepDecarbonization:ModelingApproachesandDataChallenges11.IntroductionIncreasingcarbonefficiencyandswitchingtocarbon-neutraltechnologiesforindustrialproductionareimperativetoachievedeepgreenhousegas(GHG)emissionsreductionsandtoaddressclimatechange,aswellastoeaseconcernsregardingenergysecurityandhigherenergyprices.Energyconsumptionbytheindustrialsectorrepresentsalmost40percentofcurrentglobaltotalfinalconsumptionandisstilldominatedbyfossilfuels,inparticularcoal.In2021,industrywasthesecond-largestemittingsector,afterpowergeneration,andwasdirectlyresponsibleforemitting9.4gigatonnes(Gt)ofCO2.Thisestimate,whichisequivalenttoaquarterofglobalemissions,doesnotincludeindirectemissionsfromelectricityusedforindustrialprocesses(IEA2022c).Industrialenergyandcarbonintensitiesvarysignificantlyacrosssectorsaswellaswithinsectorsacrossdifferentcountries,withsixsectorsemergingasparticularlyenergy-andemissions-intensive(seeSection2).Theaimofthispaperistodescribethemostcommonapproachestothemodelingofindustrialemissions,withaparticularfocusontheabilityofavailablemodelstodepictthedifferentmitigationoptionsrelevantforenergy-intensiveindustries.Theseoptionsrangefromincreasingenergyefficiencytodevelopinganddeployingnovelnegative-,zero-,orlow-emissionstechnologies.Importantly,producingquantitativeforecastsofindustrialenergydemandandemissionsisstronglydependentontheavailabilityofpastdataformodelcalibration.Furthermore,differentmodelingapproachesandmethodsarecharacterizedbythecapacitytoprovidemoreorlessdetailedscenariosintermsofgeographic,sectoral,andtechnologicaldetail.Understandingthestrengthsandweaknessesofavailableapproachesandtoolsforindustrialenergyandemissionsmodelingprovidesthebasisfordevelopingandinterpretingresultsthatcanbeusedtoinformenergy-andclimate-relatedpolicymaking.Wealsodescribethemostpromisingdeepdecarbonizationoptionsineachsectoranddiscusswhetherandhowtheseoptionsarerepresentedinindustrialenergymodels.Itisimportanttorecognizethatdeepemissionsreductionscannotbeachievedbypursuingasingledecarbonizationstrategyalone;rather,thesereductionsaremorelikelytobeachievedthroughacombinationofmanymitigationoptions,aswellasinvestmentinandsupportfordifferenttechnologiesandsubtechnologies.Conversely,ignoringsomeoftheseoptionsandpromotingonlyselectonesreducesthelikelihoodofachievingdeepdecarbonizationtargets.Therefore,ourassessmentoftherelevanceofvariousmitigationoptionsforindustrialdeepdecarbonizationshouldnotbeinterpretedassuggestingthatoneoptionshouldbechosenovertheothers.ResourcesfortheFuture2Thepaperisorganizedasfollows.Section2identifiesanddescribestheenergy-intensivesectorsthataccountforthemajorityofindustrialenergydemandandemissions,thefocusofthispaper.Section3reviewstheavailablestrategiesthroughwhichindustrialenergyemissionscanbereduced.Section4illustratesmoredetailedandspecifictechnologicaloptionsineachofthekeyenergy-intensivesectors.Section5providesanoverviewofthedataavailabletomeasureindustrialenergydemandandemissions,whicharecriticalinputsformodeldevelopmentandcalibration.Italsodiscussesthedifficultyofobtainingdatatomodelseveralofthedecarbonizationapproachesthatarerelevantinindustrialsectorsfordeepdecarbonization,duetolimiteddatacoverageanddetail.Section6presentsaframeworktodistinguishamongthedifferentmodelingapproachestoindustrialenergydemandandemissions,whileSection7summarizescommonapproachestomodelinginnovationandtechnologydiffusion.Section8looksatthekeymodelsthathavebeenusedintheliteraturetothisend,andSection9concludes.2.Energy-IntensiveIndustrySectorsWhileallindustrialsectorsrelyonfossilfuelsfortheirproductionactivities,energyintensityvariessignificantlyacrossthem.Themostenergy-intensivesectorsworldwidearesteel,cement,chemicals,lightmanufacturing(asdefinedinSection2.4),aluminum,andpulpandpaper.Historically,thesesectorsaccountedforabouthalfofallindustrialsectors’deliveredenergyuse(EIA2016).Theaimofthissectionistojustifyourfocusonalimitednumberofcarbon-intensivesectorsbydescribingtheircontributiontoeconomicgrowth(output),energydemand,andCO2emissions.Tothisend,werelyontheIPCCSixthAssessmentReport(IPCC2022)forsteel,cement,andchemicalsandIEAreports(IEA2022a,d,e)forlightmanufacturing,aluminum,andpulpandpaper.Itisimportanttonotethatinprovidingthisoverview,wetakeaglobalperspectiveandlargelyabstractfromnationalspecificitiesandintersectoralheterogeneity.2.1.SteelCrudesteelproductionrosegloballyby41percentbetween2008and2020.Worldwide,around40percentofsteelisusedinbuildings,20percentinindustrialequipment,18percentinconsumptiongoods,13percentininfrastructure,and10percentinvehicles.Emissionsassociatedwithsteelproduction1,primarilyfromtheuseofcokeovensandblastfurnaces,areestimatedat3.7to4.1gigatonnesofCO2equivalent(GtCO2e),accountingfor20percentofallworldwidedirectindustrialemissionsin2019(IPCC2022).1Percentagessumupto101percent.Thisisduetoroundingintheoriginalsourceofthedata.IndustrialDeepDecarbonization:ModelingApproachesandDataChallenges32.2.CementSincethemid-20thcentury,cementoutputhasgrownfasterthanworldpopulation,indicatinganincreaseintheuseofcementforinfrastructureandbuildings.In2019,directemissionsfromcementmanufacturingwereestimatedatbetween2.1and2.5GtCO2e,or14to17percentoftotalworldwidedirectindustrialGHGemissions,despitemajoradvancesinenergyefficiencyinthissectoroverthepasttwodecades.Typically,about40percentofthesedirectemissionsresultfromthecombustionoffuelstoproducethehightemperaturesneededinthemanufacturingprocess,andtheremaining60percentoccurduringthedecompositionofcalciumcarbonate(IPCC2022).2.3.ChemicalsChemicalproductsincludeplastic,rubber,fertilizers,solvents,andothersubstancessuchasfoodadditivesandpharmaceuticals.In2019,thechemicalsector’semissionswereestimatedat1.1to1.7GtCO2e,or10percentoftotalworldwidedirectindustrialemissions.Processesusedtoproducechemicalssuchasammonia(usedintheproductionofnitrogenfertilizers),methanol(usedintheproductionofadhesives,resins,andfuels),andolefinsandchlorine(essentialcomponentsofplasticsemissions)areveryenergy-intensive(IPCC2022).2.4.LightManufacturingIn2019,lightmanufacturingaccountedfor17percentoftotalindustrialemissions.Between2010and2020,overalloutputfromtheseindustriesincreased,althoughemissionsfellbyaround2.3percentoverthesameperiod(IEA2022d).AccordingtotheIEAdefinition,lightmanufacturingincludesadiversifiedsetofindustries:foodproduction(14percentoflightmanufacturingemissions),timber(1percent),machinery(8percent),vehicles(2percent),textiles(3percent)andotherconsumergoods(55percent),construction(9percent),andmining(8percent).2.5.AluminumIn2021,aluminum,acrucialinputinseveralcriticalenergytransitiontechnologies,accountedforapproximately3percentoftheworld’sdirectindustrialCO2emissions.Overthepast10years,theglobalaluminumindustry’sdirectemissionshavebeenrisinggraduallyasaresultofrisingproduction.Becauseofslightimprovementsinemissionsintensityovertimeandlevelingoutputin2019,emissionsdecreasedforthefirsttimein10years,butthistrendsubsequentlyreversed.Thealuminummanufacturingprocessusessignificantamountsofelectricity,andthesourceofelectricityisimportantindeterminingitsoverallemissionsprofile.In2021,directemissionsfromthesectorwere275megatonnes(Mt),butoverallemissionsincludingindirectemissionsfrompowerconsumptionweresubstantiallylarger,at1.1GtofCO2(IEA2022a).ResourcesfortheFuture42.6.PulpandPaperIn2021,pulpandpaperproductionreached190MtofCO2emissions,whichisahistorichighandaccountsfornearly2percentoftotalindustrialemissions.Paperoutputisexpectedtoexpanduntil2030,andthereforeimprovementsintheindustry’semissionsintensityarerequiredtoreducefutureemissionsfrompulpandpaper.Futurepaperoutputisexpectedtoriseslightlyasaresultofanincreaseinpackagingandsanitarypapergoods,particularlyindevelopingeconomies,whichmorethancompensatesforthedeclineinprinting-relatedpaperproduction(IEA2022e).3.AvailableApproachesforIndustryDecarbonizationWhileindustrialprocessesacrossandwithinsectorsdiffergreatly,fivehigh-levelstrategiescanbeidentifiedtoreduceindustrialenergyemissions:(1)fuelswitching,includingalternativefeedstocks,andelectrificationofindustrialproduction;(2)carbonefficiencyimprovementsthroughmoreefficientordigitaltechnologies(energyefficiency)orthroughzero-carbontechnologies;(3)improvementsinmaterialefficiency,includingthroughradicallynovelprocessesandbusinessmodels;(4)deploymentofcarboncaptureandstoragetechnologies;and(5)circulareconomypracticesbasedonthereduce,repair,refurbish,reuse,andrecycleparadigm.Someofthesestrategiesaremoreinlinewithdeepdecarbonizationtargets,whileothersrepresentmoremarginalimprovements.Inthissection,weprovideabroaddefinitionofeachoftheseapproaches.InSection4,wediscusstheextenttowhicheachapproachisrelevantforeachsectorandprovideconcreteandpromisingexamplesofspecifictechnologiesineachenergy-intensivesector.3.1.FuelSwitchingandElectrificationOnewaytoreduceenergyemissionsisthroughfuelswitchingandelectrification,whichrepresentamovefromacarbon-intensiveenergycarriertoonethathasloworzeroassociatedemissions.Fuelswitchingentailsashiftawayfromcoal,refinedoilproducts,andnaturalgastowardsustainableenergysourcessuchasbiofuel,solarheating,geothermal,sustainablehydrogenorammonia,nuclear,ornet-zerosynthesizedhydrocarbonfuels.Theabilityoffuelswitchingtoachievedrasticemissionsreductionsdependsonthenatureofthechosenfuelaswellasonthespecificindustrialsectorunderconsideration.Forinstance,biofuels—namely,fuelsfrombiogenicsources—areavailableinavarietyofforms,someofwhichhavepropertiessimilartothoseoffossilfuelsandthesameuses.Whilebiomethane,biomethanol,andbioethanolareavailabletodayatcostsgenerallycomparablewiththoseoffossilfuels(IPCC2022),theextenttowhichtheirusewillleadtodeepdecarbonizationisquestionedintheliterature:whiletheircarboncyclegoesintoandoutoftheatmosphere,theymaynotinfactbeGHG-neutralIndustrialDeepDecarbonization:ModelingApproachesandDataChallenges5becauseofthewaytheyareproduced,whichinvolveschangesinlanduse,soilcarbondepletion,andfertilizers(Hepburnetal.2019).However,mostbiofuel,chemical,andbiogasmanufacturingtechniquescreateconsiderablesideflowsofconcentratedCO2,whichcanbeeasilyabsorbedwhencombinedwithcarboncaptureandstorageandcarboncaptureandutilization(e.g.,bioenergywithcarboncaptureandstorage,orBECCS)andmightrepresentasourceofnegativeemissions(IPCC2022).Yettheuseofbiogeniccarbontoachievedeepdecarbonizationacrossallsectorsischallenging.CapturingcarbonduringaproductionprocessandusingitasafeedstockwouldrequirelargeamountsofhydrogentotransformtheCO2intoavarietyofchemicalsthroughareactionprocess(IPCC2022).Furthermore,thesustainablesupplyofbiomassfacessignificantchallenges,includingtheavailabilityoflandforbioenergycrops,wateruseneedsbybioenergycrops,thenecessitytoadaptbioenergycropstoachangingclimate,andtheabilitytotransportandstorelargequantitiesofcrops.Allthesechallengescurrentlyarenotfullymanaged,noraretheylikelytobeinthenearfuture(Harrisetal.2018:Buietal.2023).Switchingtosolarenergy,whichhasnoassociatedGHGemissions,ismoreinlinewithdeepdecarbonizationtargets.Directsolarheatinginindustry,forinstance,hasanacknowledgedpotential,particularlyincountrieswithhighsolarirradianceandindustrieswithmodestheatdemands,suchasfoodandbeverageproduction,textiles,andpulpandpaper(IPCC2022).Majorbarrierstoadoptionforthesetechnologiesarelocationandapplicationspecificity,theneedforenergystoragetechnologiestocompensateforintermittency(Asiabanetal.2021),highcapitalcosts,andtheabsenceofstandardizedmassproductionforequipment.Directelectrificationisaswitchfromdirectfuelusetowardelectricityandrepresentsanimportantoptiontoachievingindustrialcarbonneutrality(IPCC2022).Electricityisaversatileenergycarrierthatcanbeproducedfromavarietyofprimarysources,withsignificantpotentialforprocessimprovementsintermsofend-useefficiency(Eyre2021),qualityandprocesscontrollability,digitizability,andtheabsenceofdirectlocalairpollutants.Theemissionsreductionsthatcanbeachievedthroughelectrificationvarydependingonthespecificsectororsubsector.Forinstance,theyarehigherinmanufacturing,whichusesfossilfuelsasenergycarriers,butlowerinthechemicalsector,whichusesfossilfuelsasfeedstocksandnottogenerateenergyorheat.Furthermore,thepotentialforemissionsreductionsdependsonwhethertheelectricityisproducedusinglow-GHG-intensityprimarysources(wind,solar,hydro,advancedgeothermalnuclear,fossilfuelswithcaptureandstorage)(IPCC2022).Roelofsenetal.(2020)estimatethatalmosthalfofthefuelconsumedforenergycanbeelectrifiedwithtechnologythatisalreadyavailable.ResourcesfortheFuture6AsdiscussedinSection4,importantprogressisbeingmadeinallindustrialsectors;however,electrificationismosteasilyachievedinlightmanufacturingsectors.Forsectorswithlargeneedsofhigh-temperatureheat(e.g.,primarysteelproduction),significanttechnologicalbarriersstillhavetobeovercome.Directinductionandinfraredheatingareoptionsforhighertemperaturerequirements,whereassteamboilers,curing,drying,andsmall-scaleprocessheatingareeasilyelectrifiablefromatechnicalpointofview.Ofcourse,electrificationiseconomicallyattractiveonlyifelectricitypricesarecomparabletothoseoffossilfuel(IPCC2022).Becauseofavarietyoffactors,electricityandfossilfuelpricesvarybycountry,yettheEUaverageelectricitypriceperkilowatthour(kWh)hasbeenconsistentlylowerthanthatofgassince2008byafactorofthreetofive(Eurostat2022).3.2.EnergyEfficiencyImprovementsinProductionProcessIncreasingtheenergyefficiencyofproductionprocessesisasecondapproachtoreducingGHGemissionsfromtheindustrialsector.Energyefficiencyimprovementsnotresultingfromfuelswitchingareanimportant,yetincremental,mitigationstrategy,astheyoftendonotentailradicaltechnologicalchanges.Energyefficiencycanbeachievedthroughtwochannels:advancesinenergy-savingbestavailabletechnologies(BATs)andshiftingindustrialplants’specificenergyconsumptiontoamoreefficienttechnology,ideallyaBAT(IPCC2022).Whileimprovingenergyefficiencyinindustrialprocesseshasahighemissionsreductionpotential,energyefficiencyalonewillnotleadtodeepdecarbonizationinindustrialsectors.Forinstance,combustionproducesapproximately10percentofglobalGHGemissionsduetohigh-temperatureheatinbasicmaterialmanufacturingprocesses(Sandalowetal.2019),yetuntilrecently,effortsandinvestmentstoreducecarbonemissionsinheatgenerationwererelativelylimited;consequently,technologicalapproachesfordecarbonizingindustrialheatproductionarestillfarfrommaturity(ICEF2020).Thereisstillhighpotentialfortheuseofnon-high-temperaturewasteheat,particularlyifcoupledwithhigh-temperatureheatpumpstoincreasethetemperatureofthewasteheattotheneededlevel(Nandhinietal.2022).Anotherkeymethodtoimproveenergyefficiencyisthroughdigitalization.Thedevelopmentoftechnologyincludingsensors,communications,analytics,digitaltwins,machinelearning,virtualreality,andcomputingtechnologiesenablesfutureadvancesinprocesscontrolandoptimization.Smartenergysolutionswithreal-timetrackingenabletheoptimizationofnewtechnologies,energydemandresponsiveness,andenergysupply-and-demandbalance,includingpricing,productqualitycontrol,andforecastingandreducingunproductivetimeforhumansandrobots(IPCC2022).Significantinvestmentsindigitalsolutionsarebeingcarriedoutinmostindustrialsectors,alsoasaresultofpubliccommitmenttopromotethetransitiontowardIndustrialDeepDecarbonization:ModelingApproachesandDataChallenges7Industry4.0(see,e.g.,EU2020;ABIResearch2022;Verdolini2023).2Yetthepotentialfordigitaltechnologiestoreduceemissionsthroughincreasedefficiencyishigherinnon-energy-intensivesectorsandgenerallylimitedinenergy-intensivesectors(IPCC2022).Importantly,quantitativeevidenceisscarceregardingtheimpactofdigitaltechnologiesonindustrialenergydemandandassociatedemissions.Asaresult,itishardtoinformmodelsregardingthepotentialenablingroleofdigitaltechnologies.Moreover,appropriategovernanceofdigitalizationwillberequiredtoensurethatthebenefitsofdigitaltechnologiesareused“deliberate[ly]forthegood”(Creutzigetal.2022).3.3.MaterialEfficiencyMaterialefficiency,thesupplyofgoodsandserviceswithlessmaterial,isincreasinglyrecognizedasacriticalstrategytoloweringGHGemissionsintheindustry(IEA2019).Yet,similarlytoenergyefficiency,materialefficiencyisnotsufficienttoachievethedeepdecarbonizationofindustry.Optionsforimprovingmaterialefficiencyexistateverystageofamaterial’sorproduct’slifecycle,suchasbydesigninglightweightproducts,optimizingtopreserveend-of-lifeservicewhilereducingmaterialuse,anddevelopingcircularprinciples.Thepreciseopenphysicalmappingofcurrentmaterialsupplychainsallowsmaterialefficiencymeasurestobetrackeddowntowhereemissionsareemitted,andthesealternativesmaybecomparabletodecarbonizationandconventionalenergyefficiencytechniques(IPCC2022).Manymaterialefficiencyactions,suchasdesigninglightweightitems,resultindirectGHGemissionssavingsintheshortterm;othersalsohavelong-termemissionsreductioneffects.Forexample,developingaproductthatcanbereusedorhasalongerlifespanreducesemissionsbothtodayandinthefuture.Whilematerialefficiencyisgenerallynotwellrepresentedinclimate-energy-economymodels,theInternationalEnergyAgencydevelopedascenarioin2015thatprojectsa17percentdecreaseinindustrialenergydemandin2040(IEA2015)duetoincreasedmaterialefficiency.3.4.CarbonCaptureandStorageTechnologiesCarboncaptureandstorage(CCS)andcarboncaptureandutilization(CCU)representpotentialoptionstoachievedeepdecarbonization,butthesetechnologieshavenotyetbeenprovenatacommerciallevel,andthereareconcernsaboutthelong-termstorageofcarbon.Forinstance,themostdevelopedmethodforlong-termCO2storageinsubsurfaceporespaceissequestrationinsedimentaryformations.MajorpotentialhazardsfromCO2storageinsubterraneanporespaceincludeleakagefromwellboresornonsealedcracksinthecaprock,buildingofpressureinthereservoirthatmightresultincaprockhydraulicfracturing,andpollutionofdrinkingwater.InducedseismicitymayresultfromtheinjectionofCO2intosubterraneanreservoirs.Moststudiesbelievethatproducedearthquakescarryamodestriskofcausingfaultdisplacementsandcompromisingreservoirsecurity,butotherscontendthatevensmall-tomoderate-magnitudeearthquakescandamagethesealandunderminetheintegrityofsequestrationreservoirs(Kelemenetal.2019).2Industry4.0referstorapidimprovementsinindustrialsystemsandproductdesign,production,andmaintenanceasaresultsofwhattheliteraturehasdefinedasthefourthindustrialrevolution(hencethe4.0),mostlypromotedbythewidespreaddiffusionofdigitaltechnologies(EuropeanParliament2015).ResourcesfortheFuture8CCSandCCUusesimilarcapturetechnologiestocollectconcentratedflowsofCO2fromsmokestacks;themaindifferencebetweenthetwoiswhatoccurstotheCO2afteritiscaptured.CCSinvolvestherecoveryandstorageofCO2fromcombustion,gases,andambientairtothegeosphereforthousandsofyears(IPCC2022).CCUinvolvesthecaptureofcarbonfromoneprocessanditsreuseforanother,loweringemissionsfromthefirstprocess,yetpotentially,butnotinevitably,releasingcarbontotheenvironmentinsubsequentoperations(TanzerandRamirez2019).Thenetpotentialimpactofthesetechnologiesoncarbonemissionsisasourceofdebateintheliterature.Theircontributiontodeepdecarbonizationscenariosdependsontheinitialsourceofcarbon,fossilfuelorbiomass,aswellastheperiodofstorageorusage,whichcanrangefromdaystomillennia(IPCC2022).Accordingtorecentanalysis,CCScantechnicallyachievenear-zeroCO2emissionsinapplicationswheretheCO2canbecapturedduringtheproductionprocess,withhighlyvariablepartiallynegativeemissionsoverthelifecycleiftheoriginisbiogenicfuels.Brandletal.(2021),forinstance,arguethatcaptureratesupto98percentaretechnicallyfeasibleandresultinnegligibleincreasetotheoverallsystemcosts.However,achievingnetzerowouldrequiretheindirectcaptureofresidualemissionsbycomplementarycarbondioxideremovaltechnologies,suchasafforestation,BECCS,ordirectaircapturewithcarbonstorage(DACCS).3ArecentmodelinganalysisconcludesthatthecontributionofCCStoemissionsreductionsishighinthesectorsofsteel,cement,andrefineries,aswellasthepowersectortoalesserextent.TheemissionreductionpotentialofCCSvarieshoweveracrossdifferentcountriesoftheworldanddependingonthesocio-economicpathwaymodelled.(Turgutetal.2021).3.5.CircularEconomyApproachesFinally,industrialenergyemissionscanbeloweredthroughcirculareconomy(CE)approachestotheprovisionofgoodsandservices.Thiswouldentailstrategiestoreduce,repair,refurbish,reuse,andrecycle.CEpracticescancontributesignificantlytoemissionsreductions,thoughalonetheyarenotsufficienttoachievedeepdecarbonizationtargetsinindustrialsectors.Circularityentailsclosingmaterialandenergyflowloopsintheprovisionofgoodsandservicesbyimplementingpoliciesandproceduresformoreefficientenergy,materials,andusagewhilegeneratingtheleastpossibleamountofwastetotheenvironment(IPCC2022).Thismaybedonethrough,for3Whilenotamitigationoption,directaircapture(DAC)couldbeusedtooffsetbothfuelandprocessemissionsatindustrialplants.DACreferstochemicalprocessesthatseparateCO2fromtheambientair.TherequiredenergytocaptureCO2increasesastheconcentrationofCO2falls,sotheenergyrequirementtoremoveasingletonofCO2fromtheatmosphereisquitehigh.(TheconcentrationofCO2intheairisabout400partspermillion[ppm],comparedwithabout120,000ppminfluesatcoal-firedpowerplants.)ThecostsofDACdependonthesourceoftheenergy,andestimatedcostsvaryfrom$100toover$1000perton(NASEM2019).Todate,investmentinDACfacilitieshasbeenlimitedbecauseofthehighlyuncertainfutureofthistechnology,whichisforseveralreasons:DACisgenerallyconsideredtobeamongthemoreexpensivecarbondioxideremoval(CDR)pathways;mostCDRpathwaysofferbenefitsbesidesCO2removal,whereasDACprovidesnocobenefits;andDACrequireslargeamountsofenergypertonofCO2removed.Morerecently,fundingandincentiveshavetargetedthefurthertechnologicaldevelopmentofthismitigationoption,includingintheUnitedStatesandtheEuropeanUnion.IndustrialDeepDecarbonization:ModelingApproachesandDataChallenges9instance,theproductionofdurableitemsthatcaneasilybefixedandwhosepartscanbereused,refurbished,andrecycled,asopposedtoalinearproductionmode(Wiebeetal.2019).GiventhatCEencouragesreduction,reusing,andrecycling,aconsiderableproportionofenergy-andGHG-intensiverawmaterialdemandandassociatedprocessingmaybeeliminated,resultinginconsiderablecarbonemissionsreductions.Theextenttowhichcircularitypracticeswillresultinlowerenergydemand(andassociatedemissions)hasyettobedetermined.Inthecaseofrecyclingscrapmetal,theresultingmaterialisoftenoflowerquality,withpropertiesthatdifferfromtherawmetal,becausethescrapmetalisoftenmadeupofavarietyofmaterialsthatarehardtoseparate.Thisisreferredtoasdowncycling.Conversely,recyclingandupcyclingmaybeachieved,butatthecostofhighenergyuse(IEA2020).Circularitycanbeimplementedatthreelevels:micro(insideasinglefirm),meso(involvingthreeormoreenterprises),andmacro(cross-sectoralcollaboration).Eachlevelnecessitatesitsownsetofinstrumentsandstrategies,suchasincentivesandtaxpolicies(macrolevel)andeco-designlaws(micro-level)(IPCC2022).Atthemicrolevel,moreorganizations,particularlymultinationalcorporations,areimplementingCEpracticesasaresultoftheiradvantages(D’Amatoetal.2019).Industrialparks,fromameso-levelperspective,minimizeinfrastructurecostsbyclusteringindustrialoperationsinspecifiedregionsandareoftenestablishedaroundbigcorporations.Atthislevel,typicalCEtechniquesandstrategiesincludesustainablesupplychainsandindustrialsymbiosis,acollaborationamongdifferentindustrialactorstooptimizetheuseofresourcesandreducewastegeneration(IPCC2022).Theprimaryadvantageofindustrialsymbiosisisthereductionofbothvirginmaterialsandfinalwaste,aswellasavoideddeliverycostsfromexchangesamongfirms,whichcouldboostthecompetitivenessofsmallandmedium-sizeenterprises.Themacro-levelapproachaimstoexploitthepotentialCEsynergiesthatexistoutsidetheconfinesofasingleindustrialpark,expandingsymbiosistoentireurbansettingsthroughtheutilizationofwastefrommunicipalitiesasalternativeenergysources(Sunetal.2017).ResourcesfortheFuture104.DemandReductionandEnergyEfficiencyPotentialsofEnergy-IntensiveSectorsThissectiondiscussesspecificoptionsfordecarbonizationineachofthesixenergy-intensivesectors,providingdetailsonwhichspecifictechnologiesandpracticescouldbeadoptedtoachieveemissionsreductionsbypromotingfuelswitchingandelectrification,increasingenergyandmaterialefficiency,deployingCCSandCCU,orimplementingcirculareconomyapproaches.Table1providesavisualsummaryofthissection,highlightingtherelativeimportanceofeachstrategyineachofthesectors.Asclarifiedearlier,thissectiontakesaglobalperspectiveandlargelyabstractsfromnationalspecificitiesandheterogeneitywithinsectors.4.1.SteelSteelproductionmaybeclassifiedintotwocategories:primaryproductionfromironoreandsecondaryproductionfromsteelscrap.Theblastfurnacetobasicoxygenfurnace(BF-BOF)routeisthemostusedprimaryproductionrouteworldwide,whiletheelectricarcfurnace(EAF)isthepreferableprocedureforsecondaryproductionthroughmeltingandalloyingrecycledsteelwaste,asitrequireslessenergyandgeneratesfeweremissions(IPCC2022).Analternativebutlesscommonwaytoproducesteelisusingdirectreducediron(DRI)toreduceironore—thusreplacingBFs—whichisgenerallyfollowedbyanEAF(IPCC2022).In2019,approximately73percentofworldwidecrudesteeloutputwasproducedusingBF-BOFtechnologies,while26percentwasproducedusingtheEAFmethod.Ofthelatter,about5.6percentisderivedfromDRI(WorldSteelAssociation2021).Importantly,productionprocessesvarygeographically;forinstance,themajorityofUSsteelproductionisdonethroughEAFs.IndustrialDeepDecarbonization:ModelingApproachesandDataChallenges11ThereareseveralapproachestosignificantlyreduceGHGemissionsinthesteelsector(IPCC2022).First,potentialenergyefficiencyimprovementsofprimarysteelproduction—thatis,BF-BOFapproaches—areestimatedat15percent(IPCC2022).Second,circularpractices,promotingsecondaryratherthanprimaryproduction,areanothereffectivewaytoreduceemissions.Thisoption,however,stronglydependsontheaccessibilityofdomesticandforeignscrapsuppliesandnecessitatesmeticuloussortingandscrapmanagement,particularlytogetridofcoppercontamination(Daehnetal.2017).Oneimportanttechnologyincircularpracticesistheelectrowinningprocess,alow-temperatureelectrolysismethodforextractingsolid-stateelementalironfromironore.TheironisthenputintoanEAFtoproduceliquidcrudesteel,whichmayalsobemixedwithscrapsteel(Junjie2018).Third,fuelswitchingandelectrification,usingcarbon-freeenergyandfeedstocksourcesasinput,orcarboncaptureandstoragetechnologies(IPCC2022)couldpotentiallyreduceanestimated80percentofcurrentemissionsfromprimarysteelproductionwithtoday’sdominanttechnology,BF-BOF.TheextenttowhichBF-BOFscanberetrofittedforcaptureiscurrentlyamatterofdebateintheliterature.FanandFriedmann(2021),whoconsidernear-termoptionstorapidlyreduceGHGemissionsinsteelproductionbyexaminingtechnicaloptionsintermsofcost,viability,readiness,andabilitytoscale,arguethatitwouldbechallengingtoretrofitBF-BOFsbeyond50percentcapture;conversely,HughesandZoelle(2021),whoperformasensitivityanalysisonthecostofcapitalforironandsteelretrofit,assumethatretrofittingcanachieveupto99percentcapture.NotethatBF-BOFsmusthavetheirfurnacesrelinedTable1.SummaryofRelevanceofDifferentMitigationOptionsbySectorFuelswitchingandelectrificationEnergyefficiencyMaterialefficiencyCCS/CCU/DACCirculareconomySteelHighMediumHighMediumHighCementandconcreteLowLowHighMediumLowChemicalsHighLowMediumHighMediumLightmanufacturingHighHighAluminumandnon-ferrousmetalsHighLowMediumLowMediumPulpandpaperHighHighLowMediumLowResourcesfortheFuture12every15to25years(IPCC2022);thiscostsfrom80to100percentmorethanbuildinganewfacility.Forthisreason,itismoreeconomicallyviabletoconstructanewfacilitythatisbuiltfor90+percentcapturethantoretrofit.DRIwithCCSusingsyngasbasedonmethanecanalsobeusedtoreduceemissionsinthesteelindustry.Currently,themajorityofDRIplantsemploysyngasofH2andCObasedonmethaneasafuelandareductant.Furthermore,hydrogen-basedDRI(H-DRI)isbeingdevelopedonthewidelyusedDRItechniquebutusingonlyhydrogen.IronorereductionisfrequentlyfollowedbyanEAFforsmelting.ThissteelmanufacturingmethodmaybepracticallyCO2neutralifhydrogeniscreatedusingcarbon-freesources(Vogletal.2018).Moltenoxideelectrolysis(MOE)isanothermethodforextractingmetalfromitsoxidesource.Thebenefitsofliquidmetalproductionaretheeasewithwhichthemanufacturedmetalmaybecollectedandthecapabilitytooperatecontinuously.TheuseofelectricityformetalextractionincludesusingrenewableenergyandthedecouplingofmetalproductionfromCO2emissions.Asaresult,ifsuitableforindustrialscalesofproduction,thistechnologywillbeofsignificantinteresttothesteelindustry(Wienckeetal.2019).Fourth,emissionsinthesteelsectorcanbedramaticallyreducedbyincreasingmaterialefficiency(i.e.,lesssteelusagepervehicle)aswellasthroughcircularitypracticesanddemand-sideoptionsthatwouldlowerthedemandforsteelmanufacturingorincreasetheintensityofproductuse(i.e.,carsharing).Inparticular,theIEAestimatesthatstringentmeasurestargetingmaterialefficiencycouldrealisticallylowerthedemandforsteelby40percentby2060(IEA2019).4.2.CementandConcreteAvailableanalysessuggestthatthecementindustryhaslimitedmitigationoptions,yetsomeexist.Onestrategyisbasedonmaterialefficiency;makingstrongerconcreteviaimprovedmixing,aggregatesize,anddispersionisoneoftheeasiestandmostefficientmethodstominimizecementandconcreteemissions.Becausecementislowcost,corrosion-andwater-resistant,andeasytoworkwith,architects,engineers,andcontractorsfrequentlyoverbuildwithit.Buildingsandinfrastructurecanbeintentionallyplannedtolimittheuseofcementtoitsnecessaryapplicationsandsubstituteothermaterialsfornonessentialuses.Thismightcuttheuseofcementby20–30percent(IPCC2022).Indeed,whilefuelswitchingandelectrificationdonotappeartobeviableoptionstominimizeorreducetheCO2emissionsspecificallyassociatedwiththetypicalPortlandAcementmanufacturingprocess(IPCC2022),availableassessmentsindicatethatsomecountrieshavehighclinker-to-cementratios,withtheUnitedStateshavingthehighest.Decreasingtheseratioswouldreduceemissions(Pascaleetal.2021;IEA2022b;CTCN2016).Second,fuelswitchingandelectrificationhavethepotentialtoloweremissionsinspecificphasesoftheproductionprocess.Forinstance,theuseofbioenergysolids,liquids,orgases(IEA2018),hydrogen,orelectricityforproducingthehigh-temperatureheatneededatthecalcinercanalsominimizetheenergy-relatedemissionsofcementmanufacture.Becauseoftheirdifferentqualitiesofquickandslowcombustion,co-burninghydrogenandbioenergymaybebeneficialinthisrespect.IndustrialDeepDecarbonization:ModelingApproachesandDataChallenges13Third,givenlimitedothermitigationoptions,carboncaptureandstoragetechnologyisfrequentlycitedasapotentiallysignificantcomponentofanambitiousmitigationapproachinthecementindustry.CCStechnologycancaptureonlytheprocessemissionsorboththeenergyandprocessCO2emissions.SeveralCCStechniquesmaybeused,includingpost-combustiontechnologieslikemembrane-assistedCO2liquefactionandaminescrubbing,oxy-combustioninanatmospherewithlittletozeronitrogentogenerateaconcentratedCO2streamforcaptureanddisposal,andcalciumlooping(Deanetal.2011).Fourth,switchingtodifferentmaterialsandproductionprocessesmayberequiredtoachievedeepdecarbonizationinthissectorinthelongrunifmaterialefficiency,improvedmixingandaggregatesizing,andCCSwithextrabioenergyarenotviableinsomeplacesoratalltoreachnear-zeroemissions.AmongthesuccessfulstrategiestoreduceCO2emissionsintheglobalcementindustryistheuseofsupplementarycementitiousmaterials(SCMs)(Ayatietal.2022),includingfromindustrialandagriculturalwastes,toreplacepartoftheclinkerincement(Singh2022).However,implementingthisoptionatalargescalewillbechallengingduetothelimitedsuppliesofconventionalSCMs,unlessnewtypesofSCMsbecomeavailable(Scriveneretal.2018).Limestonecalcinedclaycement(LC3),whichismadebyblendingclinker,calcinedclay,limestone,andgypsum,hasbeengainingconsiderableattentionandinvestmentsasanimportantalternativetoPortlandcementbecauseitreducesCO2byupto40percent,useslow-graderawmaterials,andrequiresalowercalcinationtemperatureforclay.Itiscost-effectiveanddoesnotrequiremajormodificationincementplants.Inaddition,alkali-activatedmaterialsandgeopolymers,aluminatecements,magnesia-basedcements,andgypsum-basedmaterialsrepresentpromisingalternativebindersthatcanbeproducedwithlowercarbonfootprints(PengandUnluer2023).Indeed,someofthesealternativesolutionstolimestone-basedordinaryPortlandcementhavebeentestedandusedregionallyandhavegivenrisetopartialsavings(IPCC2022),buttheyremainhardertoimplementascomparedtoaccessinglimestoneresources(MaterialEconomics2019).4.3.ChemicalsAkeycharacteristicofthechemicalsectoristheheterogeneityofitsproducts,eachofwhichhasitsownproductionprocess,distinctsetoftechnologies,andmitigationoptions,makingitchallengingtomodeldecarbonizationinthissector.Thisisparticularlytrueforsomeofthechemicalproductionprocessesforwhichnotechnologicaloptionscurrentlyexistthatallowproductionofcertainchemicalstobedecoupledfromtheuseofacarbonsource(IPCC2022).Theprocessesthatconsumethemostenergyinthissectorarethosefortheproductionofhigh-valuechemicals(suchasethyleneorpropylene,typicallyprecursorstoplastics),fertilizers,methanol,andsomehalogenssuchaschlorine(Worrelletal.2000).Yettheliteraturesuggeststhatthechemicalsectoroffersseveralmajormitigationoptions.Thereareseveralpotentialroutestoloweremissionsinthechemicalsector.First,thissectorhasoneofthehighestpotentialsforelectrification,suggestingtheprospectofarapiddecreaseinassociatedemissions(Madedduetal.2020).Indeed,ResourcesfortheFuture14despitetheindustry’sconsistentimprovementsinenergyefficiencyoverthelastfewdecades,thedemandforheatandsteaminthemanufacturingofbasicchemicalsisresponsibleforalargeshareofemissions(BazzanellaandAusfelder2017).Mostofthisenergyisnowprovidedbyfossilfuels;theseintheorycanbereplacedwithbioenergy,hydrogen,orelectricitywithminimalornocarbonemissions(IPCC2022).Second,fuelswitchingcanalsosignificantlyloweremissions;thisisparticularlythecaseforammoniamanufacturing,whichaccountsforaround30percentofallCO2emissionsinthesector(IPCC2022).Nitrogenandhydrogenarecombinedtomakeammoniathroughacatalyticprocess,withhydrogenmostfrequentlyproducedthroughnaturalgasreforming(MaterialEconomics2019)or,insomeareas,coalgasification,whichhassignificantlyhigherrelatedCO2emissions.Futurelow-carbonalternativesforammoniaproductionincludemethanepyrolysis,whichconvertsmethaneintohydrogenandsolidcarbon(MaterialEconomics2019),hydrogenproducedbyelectrolysisusinglow-orzero-carbonenergysources,andnaturalgasreformingwithCCS.Comparedwithconventionalprocesses,electrifyingammoniawouldresultinareductionintheoverallamountofprimaryenergyused,althoughinnovativesynthesisproceduresstillhaveasubstantialopportunityforefficiencyimprovement(IPCC2022).Switchingtosyntheticfeedstockalsoplaysacrucialroleinthosechemicalproductionprocessesforwhichnocarbon-freeoptionsexist.Forinstance,apossiblestrategyissynergisticcombinationoflow-GHGhydrogenandcarbonobtainedbydirectaircaptureorfrompointsourcesforfurthervalorization(Kätelhönetal.2019).Toreplacethesteamcracker(IPCC2022)oraFischer-Tropschprocessthatmaymanufacturesynthetichydrocarbons,low-carbonmethanolcanbeproducedandusedinmaketoorder/maketoassembleprocessestoconvertmethanoltoolefinsandaromatics(IPCC2022).Anotherstrategyinvolvesemployingbiomassresources(IsikgorandBecer2015)orexistingresidualstreamstoprocesscarbonfromrenewablesourcesindefinedbiotechnologicalprocessesatthebeginningofaproduct’slifecycle(IPCC2022).Third,theliteratureidentifiesalargenumberofnewtechnologiesrelevantformitigationinthechemicalsectorthathaveexpecteddeploymentdatesrangingfromnowto2025.However,theirpotentialcontributiontoachievedeepdecarbonizationvaries.Amongthosewiththehighestpotentialinthisrespectarevariouscarboncapturemethodsandelectrolytichydrogengeneration(IPCC2022);conversely,methanepyrolysis,electrifiedsteamcracking,andbiomass-basedethanol-to-ethyleneandlignin-to-BTX(benzene,toluene,andxylenes)pathwaysareconsideredmediuminimportance.Whilemacro-levelcalculationsdemonstratethatlarge-scaleusageofcarboncirculationthroughCCUasamainapproachisfeasibleinthechemicalsindustry,itwouldbehighlyenergy-intensive,andtheclimaticeffectwouldbeheavilydependentonthesourceofCO2andprocedureforabsorbingit(Kätelhönetal.2019).CCSplaysaparticularlyimportantroleinthoseproductionprocessesfororganiccompoundsthatwillcontinuetorequireacarbonsourceasaninput(IPCC2022).Yetthelarge-scaledeploymentofcarbondioxideremovaltechnologieswouldrequireacompletereshapingofthechemicalsector,withsomeindustriesdedicatedtotheproductionofsorbents,necessarytooperatethetechnology.Sorbentproductioniscurrentlyonlyaby-productofprocessesinthechemicalsector.Upscalingitwouldimplyanincreaseinenergydemandfromthechemicalsector,astheseprocessesareveryenergyintensive(Realmonteetal.2019).IndustrialDeepDecarbonization:ModelingApproachesandDataChallenges15Fourth,circularitypracticesarealsorelevantinthechemicalsector.Forinstance,thepyrolysisofusedplasticscanproducebothgasandnaphthapyrolysisoil,aportionofwhichmightreplaceconventionalnaphthaasanenergysourceinthesteamcracker(Honusetal.2018).Catalyticcracking,hydrocracking,andpolymerselectivechemolysisareadditionalmethodsforchemicalrecycling(Ragaertetal.2017).Achievingnear-zeroemissionsmayrequirecombiningchemicalrecyclingwithCCStoreducecarbonlossesandprocessemissions.Finally,achievingdeepemissionsreductionsrequiresdemand-sidestrategiessuchasefficientenduse,materialefficiency,andreducingdemandgrowth,inadditiontorecyclingwheneverpossibletominimizetherequirementforprimaryproduction(IPCC2022).4.4.LightManufacturingAspreviouslydiscussed,lightmanufacturingcomprisesadiversifiedsetofindustries,eachofwhichhasspecificproductionprocessesandtechnologiesand,consequently,mitigationoptions.Processheatforapplicationssuchasdryingconsumesthemostenergyinthelightmanufacturingindustry,andapproximately90percentofthefossilfuelsemployedbythesectorareusedtogenerateprocessheat,whereaselectricityispredominantlyusedtopowermotor-drivensystems(IEA2022d).Asaresult,fuelswitchingandelectrificationrepresentanimportantmitigationoption,withhighrelevancefortheachievementofdeepdecarbonization.Forinstance,currentfossil-basedapproachesforheatinganddryingmaybereplacedbylow-orzero-GHGelectricityviadirectresistance,high-temperatureheatpumps,mechanicalvaporrecompression,induction,infrared,orotherelectrothermalprocesses.Directsolarheatingisfeasibleforlow-temperaturerequirements(100°C),whileconcentratedsolarheatingisaviableoptionforgreatertemperatures.Heatpumpsonthemarketcanprovide100°C–150°C,althoughtemperaturesofupto280°Carepossible.Wherehightemperatures(>1000°C)arenecessary,plasmatorchespoweredbyelectricitycanbeemployed,aswellashydrogenorbiogenicorsyntheticcombustiblehydrocarbons.Energyefficiencyalsoplaysanimportantrole,particularlyaimedatusingwasteheat,whichcanbetransferredfromplanttoplantatprogressivelylowertemperaturesordistributedaslow-gradesteamorhotwater,thenincreasedasneededviaheatpumpsanddirectheating(IPCC2022).Thesegeographicclustersalsocouldallowforreducedinfrastructurecostsforhydrogengenerationandstorage,aswellasCO2collection,transportation,anddisposal(IEA2022d).Materialefficiency,CCS,andcirculareconomypracticesarenotdiscussedforthissectorduetoitsheterogeneity.4.5.AluminumandNonferrousMetalsPrimaryaluminumisoftenproducedintwostages,generallyperformedinthesamelocation.Inthefirststage,theBayerhydrometallurgicalmethodisemployedtoseparatealuminumoxidefrombauxiteore;thisrequirestemperaturesofupto200°Cwhensodiumhydroxideisusedtoextractthealuminumoxideandupto1000°Cforkilning(IPCC2022).Inthesecondstage,thealuminumoxideisthenelectrolyticallyseparatedintooxygenandelementalaluminumusingtheHall-Héroultmethod.Thisisbyfarthemostenergy-intensivephaseofthealuminumproductionprocess.ResourcesfortheFuture16Inbothstages,electrificationhasthepotentialtosignificantlyloweremissionsiftheelectricityisfromlow-orzero-carbonsources.Electrificationalsohashighemissionsmitigationpotentialintheproductionprocessofothernonferrousmetals,suchasnickel,zinc,copper,magnesium,andtitanium,whichgeneratelowertotalemissions(IPCC2022).Thisisthecase,forinstance,fororeextractiontechnologiesusinglow-carbonelectricityratherthanpyrometallurgy,whichrequiresheattomeltandextractoreonceithasbeensmashed.Otherimportantmitigationoptionsfornonferrousmetalsincludehighermaterialefficiencyandcircularitypracticesaimedattherecyclingofexistingstock.Inthecaseofnonferrousmetals,manyofthesedecarbonizationoptionsareavailableandhavebeenusedonoccasioninthepast,buttheyhavenotbeenwidelyusedbecausetheyarecostlierthantraditionaltechniquesand,withlowfossilfuelprices,arenoteconomicallyattractive(IPCC2022).4.6.PulpandPaperPulpmills,integratedpulpandpapermills,andpapermillsthatusevirginpulpwoodandotherfibersources,wastesandcoproductsfromwoodproductmanufacture,andrecycledpaperasfeedstockareallpartofthepulpandpaperindustry(IPCC2022).Inchemicalpulpingoperations,pulpmillsoftenhaveaccesstobioenergy,whichcansupplymostoralloftheirheatandelectricityrequirements.Mechanicalpulpingismainlypoweredbyelectricity;hencedecarbonizationisdependentongridemissionsvariables.Excludingthelimekilninkraftpulpmills,temperaturerequirementsarenormallylessthanorequalto150°C–200°C,mostlyforheatinganddryingviasteam.Thisindicatesthatthisindustrymaybeeasilydecarbonizedbyimprovementsinenergyefficiencyaswellasfuelswitchingandelectrification,includingtheuseofhigh-temperatureheatpumps.Electrificationofpulpmillsmight,inthelongrun,makebioresiduesnowusedforenergyaccessibleasacarbonsourceforchemicals.Thepulpandpapersectorhasthecapability,resources,andknowledgetoundertakethesechanges.Inertiaindeployinglow-orzero-carbonproductionoptionsandtechnologiesisprimarilyinducedbyequipmentturnoverratesandrelativefuelandelectricityprices.Pulpmillshavebeenhighlightedasprospectivecandidatesforpostcombustioncarboncaptureandstorage,whichmayenablesomenetnegativeemissions(IPCC2022).IndustrialDeepDecarbonization:ModelingApproachesandDataChallenges175.DataonIndustrialEnergyDemandandEmissionsTheavailability,granularity.andcomparabilityofindustry-leveldataonenergydemandandemissionsacrossdifferentcountriesconstraintheabilitytosetupandcalibratemodelstoexplorepossiblepathwaysofindustrialenergyemissionsreductions.Thatistosay,therecurrentlyisnodetailedandcomprehensivedatasourcefromwhichmodelerscanobtaininformationonenergydemandandassociatedemissionsbyfueltypeorbytechnologyinthedifferentindustrialsectorsindifferentcountriesovertime.Severalmainchallengesarepresentinthisrespect,andtheyoftencompoundoneanother.Thesechallengesarediscussedinthissectionandcanbeobservedinthedetaileddescriptionofavailabledatabases,includingdataonindustrialenergydemand,presentedinAppendixA.Notethatthesedata-relatedchallengesareadditionaltoanydifficultylinkedwiththemodelingofdiversemitigationoptionsacrossdifferentsectorsanddifferentcountrieswiththesufficientlevelofdetailnecessarytogenerateindustry-leveldecarbonizationpathways(discussedinthenextsection).5.1.LackofACommonDetailedStatisticsClassificationThereiscurrentlynocommondetailedstatisticalclassificationofindustrialsectorsthatisusedworldwidetocollectenergyandemissionsstatistics.Asaresult,itisextremelychallengingtocollectdatafromdifferentnationalstatisticalofficesandcomparethem.ThethreemostwidelyusedapproachestoclassifyindustrialactivitiesaretheInternationalStandardIndustrialClassificationofAllEconomicActivities(ISIC),theNorthAmericanIndustryClassificationSystem(NAICS),andtheStatisticalClassificationofEconomicActivitiesintheEuropeanCommunity(NACE)(seeAppendixBfordetails).Theseclassificationsareusedbydifferentcountriesandinternationalinstitutionstogatherspecificdataonindustrialenergydemand.WhileNAICSisusedbytheUnitedStatesandCanada,NACEistheofficialclassificationoftheEU,andISICisoftenusedbyinternationalorganizationssuchastheIEA.5.2.LackofComprehensiveDataonEnergyDemandofDifferentEnergyCarriersInformationonenergydemandatthesectorallevelexistsforthemajorworldeconomiesbutisrarelyaccompaniedbydetailsregardingdifferentenergycarriers.Thisconstrainstheabilitytoexploretheroleoffuelswitchinginreducingenergyemissionsatthesectorallevel.Forinstance,theWorldInput-OutputDatabase(WIOD;Timmeretal.2015)November2016releaseconsistsofaseriesofdatabasescovering28EUcountriesand15othermajorcountriesintheworldfor2000–2014.Thedatabaseprovidesenergyandenvironmentalaccountsatthesectorallevelbutdoesnotincludedetailsonthetypesoffuelsused(e.g.,differenttypesofcoalversusgas).ResourcesfortheFuture185.3.LackofDetailedSectorialInformationonEnergyDemandattheLevelofDifferentProductsWhendetailedinformationisavailableattheenergycarrierlevel,informationonsectoraluseorallocationisoftenlackingordated.Intherarecaseswherethisinformationisavailable,itcannotbebrokendownacrossthedifferentproductsproducedinagivensector.Thisconstrainstheabilitytomodelsectoralspecificitiesandthustoproducesectoralscenariostodeepdecarbonization.Forinstance,theWorldEnergyBalancesdatabase(IEA2022f)containsdetailedcountry-levelstatisticsonallenergycarriersbutnosectoralbreakdown.5.4.LimitedGeographicandTimeCoverageofDetailedDatabasesDatabasesexistthatprovidedetailedinformationregardingenergydemandbyenergycarriersatthesectorallevel,buttheircoverageisnotcomprehensiveintimeorspace.Thislimitstheabilitytomodeldifferentsectorsacrossdifferentcountries—ortosimplyaccountforsectoraldynamicsinforeigncountriesthroughcalibration.Thisisparticularlyproblematicinseveralenergy-intensivesectorsthatarenotconcentratedgeographicallyandwhosedynamicsarenotnecessarilydrivenbyspecificcountries.5.5.DifficultyinLinkingDataonEmissionsandFuelInputsDetaileddatabasesprovidinginformationonGHGemissionsexist,someofwhichincludesectoraldetail,butemissionsdatacannoteasilybelinkedtodataonfuelinputs.Thismeansthatoverallemissionsreductionpathwaysatthesectorallevelmaybemodeled,butthedriversofemissionsreductionsatthesectorallevelcannotbeeasilycalibratedandstudied.Forinstance,onewouldhavetomakeassumptionsastowhetheremissionsreductionscomefromfuelswitchingorfrombroadertechnologicalchangedynamics.TheIEAAirEmissionAccountsprovidedataon16pollutantsforupto88industrialsectorsorsubsectorsfor40countries(nodatafortheUnitedStatesareavailable),buttheydonotincludeinformationontheenergycarriersassociatedwiththelevelsofemissions.Afewprivatelyowneddatabasesonemissionsbysectorandcountryexist,buttheirreliabilityisoftennotclear,astheydonotprovideinformationonthespecificenergyinputsordetailsonhowtheyarecompiled.IndustrialDeepDecarbonization:ModelingApproachesandDataChallenges195.6.DifficultyinPredictingCostsandPerformanceofRadicallyNovelTechnologiesItischallengingtopredictthecostandperformanceofradicallynoveltechnologies,aswellasthespeedoftheirdiffusiononcetheyhavereachedthemarket.ThismakesithardtoinformmodelsregardingtheprospectsoftechnologiessuchasCCU,CCS,andotherradicallynoveltechnologies,includingpromisingtechnologyoptionssuchasmoltenoxideelectrolysis,newcementchemistries,orelectrickilns.Inthisrespect,severalcontributionshaveusedexpertelicitationmethodstogenerateprobabilisticforecastsofenergytechnologycostsstartingfromthe1970s,focusingonnuclearand,morerecently,onasetoflow-carbontechnologiesincludingCCS(Verdolinietal.2018).Yetrecentcontributionsintheliteraturehaveshownthatexpertestimatesoffuturetechnologycoststendtobepessimistic,withrealizedcostdecreasessurpassingexpectationsforallthosetechnologiesforwhichwehavebothobservedandexpertelicitationdata(Mengetal.2021;Wiseretal.2021).Nevertheless,thegeneralityoftheseresultsforradicallynoveltechnologiessuchasCCUandCCScannotbetakenforgranted.Furthermore,projectionsoffutureenergytechnologycostsgeneratedbasedonlearningcurvesoftenfailtocapturethetruetrajectoryofsubsequentlyrealizedtechnologycostsforradicallynovelenergytechnologies(Mengetal.2021).5.7.LackofComprehensiveDataonMaterialandEnergyFlowsAlackofdetaileddataonmaterialandenergyflowslimitstheabilitytomodelcircularitystrategiesandassociatedemissionsreductionpotentials.Researchersstudyingindustrialmetabolismhavebeendevelopingmethodologicalapproachestothemodelingofcircularpracticesrelevantinthecontextoftheprovisionofgoodsandservices,butmuchworkremainstoproperlyintegratematerialefficiencymeasuresintoconventionalclimatechangemodels.Effortsarebeingmadetoendogenizematerialefficiencymethodswithinclimatechangemodeling,assessthesynergyeffectsandtrade-offsbetweenenergyefficiencyandmaterialefficiencyinitiatives,andcollectdataforcalculatingtheemissionssavedfromactualmaterialefficiencyactions.Thisnecessitatesanalystsworkinginmultidisciplinaryteamsandengagingwithstakeholdersthroughoutthewholematerialsupplychain.Afruitfulavenueoffutureresearchbeingpursuedbyseverallarge-scaleprojectsisthelinkingofmodelsfocusedonmaterialflowswithmodelsfortheintegratedassessmentofenergy,theeconomy,andcarbonemissions,withasufficientlevelofdetailforindustrialsectors(Haberletal.2019).Allknowndatabasesonindustrialenergyuseanddemandoremissionsareplaguedbyatleastoneoftheselimitations(seeAppendixA).Takentogether,theselimitationsalsoimplythatitisnotpossibletodescribepreciselyandcomprehensivelyallmitigationoptionsavailableinthedifferentsectors.Forinstance,scarcityofdataonmaterialgreatlylimitstheabilitytomodelsomemitigationoptions,suchascircularity.ResourcesfortheFuture20Overall,theinabilitytopreciselytrackovertimeandspacehowdifferentenergytechnologiesusedifferentfuelswithdifferentcarboncontentsillustratesthetrade-offsinmodelingindustrialenergydemand.Modelerscanfocusonasinglesectorinasinglecountryand,dependingonthecountryofinterest,potentiallyrelyondetailedfuelandemissionsdata,ortheycanfocusonmodelingseveralsectorsinoneormorecountries,withoutaccesstodetailsontechnologicaloptionsortypesoffuel.6.ApproachesforIndustrialEnergyDemandModelingTheliteratureincludesseveralapproachestomodelingclimatemitigation,basedonthespecificpurposesforwhichtheywereconceived.Althoughthecriteriaforclassifyingthemodelingapproachesmayvary,threekeydimensionsemergeasparticularlyusefulinthisrespect(Lopionetal.2018):theanalyticalapproachunderlyingeachmodel,themethodologyusedtogeneratedecarbonizationpathways,andthegranularitywithwhichdifferenteconomicsectors,includingindustry,canberepresented.Ineachofthesedimensions,modelerscanchooseamongdifferentapproaches,eachofwhichhasprosandcons.Importantly,withinlarge-scale,complexmodels,suchasintegratedassessmentmodelsoftheeconomy,energy,andclimatefeedbacks,differentapproachesmaybeadoptedindifferentmodulesorpartsofthemodel.Therefore,thethreedimensionsdiscussedhereshouldbeunderstoodasbroadguidingprinciplesbywhichtoclassifymodelsandunderstandtheirfoundingprinciplesandnotastrictclassificationmethod.6.1.AnalyticalApproach:Bottom-upandTop-downModelsModelsadopteitherabottom-uporatop-downanalyticalapproach.Accordingly,theenergysystemwouldbedescribedfromeitheranengineeringoraneconomicangle.Bottom-upmodelsintegrateahighleveloftechnicalinformationwithintheenergysystemmodeling,astheygivehighlypreciseimagesofenergydemandandsupplytechnologies,quantifyingthedemandofenergyandmassofeachtechnologycomponent(Herbstetal.2012).Theirkeystrengthisthecharacterizationoftheinterlinksbetweentechnologycomponentsbasedonthemutualdependenciesofenergyandmassflows.Thisfeatureenablesthemtoprovidein-depthanalysesofsectoralstrategies;however,italsodeterminestheirpartialequilibriumnature,asbottom-upmodelsignorerelationshipsbetweensectors.Disadvantagesofbottom-upmodelingarerelatedtodatarequirementsandtheexclusionofintersectoralfeedbacks.Modelersarelargelyreliantondataavailabilityandtrustworthinesstomodeltechnologydiffusion,investments,andoperationalcosts.Themaincriticismofbottom-upmodelingconcernsthefailuretoaccountforprogramcosts,thefeedbackofenergypolicy,andthelackofmacro-effectsoftheassumedtechnologyshiftongeneraleconomicactivity,structuralchanges,employment,andpricing(Herbstetal.2012).IndustrialDeepDecarbonization:ModelingApproachesandDataChallenges21Top-downmodelsdescribetechnologiesasasetoftechniquesbywhichinputssuchascapital,labor,andenergycanbetransferredintousefuloutputs;therefore,theygiveanaggregateddescriptionoftheenergysystembutlackdetailedtechnologicalvariability(IPCC2022).Theyhaveahighlevelofendogenousmodelingofsocialandeconomicbehavioralrelationships,suchasthoseamongwelfare,economicgrowth,andemployment.Thistypeofmodelingenablesafullunderstandingofenergypolicyeffectsontheeconomyofaregionorcountry(Herbstetal.2012).Adrawbackisthatbecauseoftheirinadequatetechnologicaldetail,theymaybeunabletoprovideanaccuratepictureoftechnicaladvances,nonmonetaryhurdlestoenergyefficiency,orregulationsforcertaintechnologies.Top-downmodelsarenotsuitabletoportraycredibletechnologyprospects,especiallyasdiscontinuitywouldapplyinthelongrun,whenintrasectoralstructuralchangewouldoccurasaresultoftechnologicaldevelopmentandsaturation(Herbstetal.2012).Moreover,becausetop-downmodelingapproachesarebasedonthetheoryofefficientlyallocatingmarkets,theytendtounderestimatethecomplexnatureofbarriersandtheirnonmonetaryforms,suchasalackofknowledge,insufficientdecisionroutines,orgroup-specificpreferencesoftechnologyproducersorwholesalers.Fourkeytop-downapproachesareinput-outputmodels,computablegeneralequilibrium(CGE)models,econometricmodels,andsystemdynamicsmodels.Becauseoftheirinherentdifferences,thesetwoapproachesareusefulinansweringradicallydifferentquestions.Bottom-upmodelsareoftenusedwhenthereisaspecificinterestinproducingdecarbonizationpathwaysforspecificsectors,forwhichassumptionsaremade(andimposedonthemodel)regardingallother,moreaggregatedriversofemissions.Top-downmodelsarenecessaryifthefocusoftheanalysisismodelingfeedbackloopsbetweenclimatepoliciesandwelfare,employment,andeconomicgrowth.Finally,becauseofthecomplementaritybetweenbottom-upandtop-downmodels,hybridapproacheshavebeenusedwherethemacroeconomyinteractswithanenergysystemmodule.6.2.Methodology:SimulationandOptimizationApproachesModelscanadopteitherasimulationoranoptimizationapproachtogeneratedecarbonizationpathways;often,bothapproachescanbeusedindifferentpartsofthesamemodel.Simulationallowsforreproducingasystemviainterpretingtheprinciplesofitsoperations.Thesemodelsrepresentthekeybehaviorsandcharacteristicsofagivenprocess(inourcase,industrialenergydemand),whilethesimulationallowsthemodeltorepresenthowtheprocessevolvesunderdifferentconditionsovertime.Simulationcanbestatic,ifitdescribesthecurrentsystemasasnapshot,ordynamic,ifthecurrentoutputisaffectedbyevolutionfrompreviousperiods.Simulationmodelscanshedlightontheendogenousrelationshipsbetweenvariablestoreproducereal-worldsystems,andalthoughthismayhappeninasimplifiedfashion,theycanalsoberathercomplex.Typicalapplicationsofsimulationmodelsareexploratoryanalyses,wherethemodelerstartswithrealisticvaluesforinputsandmodifiesthemwithinreasonablerangestodeterminewhathappenswiththeoutputs.BytweakingtheinitialResourcesfortheFuture22conditions,thebehaviorofthesimulatedsystemchangesandcanbeobserved.Aclassofsimulationmodelsexploresmultiagentapproaches,interpretingthedecisionmakingprocessesofkeyenergysystemplayers,whicharesuitedtointerpretmarketimperfectionsandconsumerandfirmheterogeneity(Hansenetal.2019).Optimizationisusedtodetermineoptimalsystemdesignoroptimalchoices.Unlikesimulationmodels,optimizationmodelsprovidethebestsolutionforagivenanswer.Theyconsistofthreeelements:theobjectivefunction,decisionvariables,andtheconstraints.Successfuloptimizationdependsonproperlyidentifyingtheconstraintsplacedonvariousparameters—forexample,themaximumlevelofenergyfromagivenenergysourceoremissionsfromagivensource.Dependingonthedescriptionoftheenergysysteminagivenmodel,optimizationandsimulationmodelscanbedemandingintermsofcomputingpowerandtimeandcanimplementsensitivityanalysestoexplorerobustnessofresultsatdifferentlevelsofcomplexity.Thedistinctionbetweenoptimizationandsimulationseemsparticularlyrelevantforthemodelingofmitigationoptionsthatarenovelanddisruptiveandaffectthetimingofwhennewtechnologiesmayenterthemarket.Optimizationmodelsapplytypicallyintertemporalapproaches,whichmeansthatavailabilityandcostsofallthetechnologiesinthefutureareknownfromthestartofthemodeling.Simulationmodelsaretypicallycoupledwithlimitedforesight,whichmeansthatinformationontechnologiesislimitedatacertaintimeinthesimulation.Thisiscriticalformodelingradicallynoveltechnologies,suchasradicaldeepdecarbonizationstrategies,wheresimulationmodelsdisplaymoreinertiatowardnoveltechnologydiffusionthanoptimizationmodels,whichshowanearliermarketuptake.Theintroductionofnoveltechnologiesis,however,linkedtotheleveloftechnologicaldetailofanenergysystem,aswellastoacomplexsetofparametersgoverningthetechnologycapacityandgrowthinthemarket.6.3.GranularityofSectoralModelingFinally,modelsdifferwithrespecttothedetailandgranularitywithwhichtheydepicttheenergysector,itsenergyandemissions.Modelscandepictasingleindustrialsector,aggregatingallproductionactivities(typicaloftop-downmodels)orspecifyingafewkeysectorsofinterestwhilelumpingallothereconomicactivitiesinthe“other”category,ortheycanbeverydetailedandincludealargenumberofsectorsandsubsectors(typicalofbottom-upmodels).Severaltrade-offsareassociatedwiththechoiceofaggregateversusdetailedrepresentation.Themoresectorallydetailedamodel,thebetteritcandepictdifferencesamongthevariousmitigationoptionswithinandacrosssectors.Thiscanbeachievedonlyifverydetaileddataareavailabletofeedthemodeandusuallyleadstolongcomputingtimestoachieveasolution.Thevariationofmodelsalongthesethreedimensionshasimportantimplicationsformodelingindustrialenergydemandandemissions.Bottom-up,sectorally-detailedmodelsaremorelikelytoallowthestudyofthechoicesamongthedifferentsector-specificmitigationoptionsdescribedinSection4.YettheyoftenabstractfromgeneralequilibriumeffectsandarebasedonstrongassumptionsregardingthedevelopmentIndustrialDeepDecarbonization:ModelingApproachesandDataChallenges23ofmacro-levelvariablesandindicators.Thislimitstheabilitytomodelcircularitypracticesthatgobeyondaspecificsectororgeographyandunderstandingoftheglobalimplicationsofchoosingdifferentmitigationoptionsindifferentsectorsorcountries.Theoppositeistruefortop-downaggregatedmodels.Giventhesetrade-offs,overtimealargernumberofmodelshaveadoptedahybridapproach,includingfeaturesofbothtop-downandbottom-upmodels.Sometimesmodelsalsocombineoptimizationandsimulationindifferentmodulesorportraycertainaspectsoftheeconomyatanaggregatedlevelwhiletheydetailspecificsectorsofinterest.Integratedassessmentmodels(IAMs),forinstance,oftenuseahybridapproachforenergydemandandemissions.AnIAMisaspecifictypeofmodelthatcombinesdatagatheredfromtwoorevenmoredisciplinesintoasingleframeworki.e.,theeconomy,theenergysystem,andtheenvironment.Researchersinphysical,biological,earth,economic,andsocialscienceshavetypicallyproducedelementsofthesemodelsautonomously.Theneedtostudyinterdisciplinaryinteractionsamongthesecomponents,asinthecaseofclimatechange,hasledtothedevelopmentofcohesiveandconsistentframeworksthatincludeseveralcomponentstoassessthestatusandimplicationsofenvironmentalchangemoreaccurately,aswellaspossiblesolutions(Bosetti2021).Someenergy-intensiveindustries,suchasironandsteelorcement,areincludedindividuallyinmosttop-downIAMs,butfewsector-specifictechnologiesareexpresslyincluded.Instead,advancesinenergyefficiencyintheindustrysectoranditssubsectorsarefrequentlydictatedbyexogenousassumptionsorareafunctionofenergycosts.Similarly,fuelswitchingismostlycausedbychangesinrelativefuelprices,whichareaffectedbyCO2pricetrends(Pauliuketal.2017).FuelswitchingcanberegulatedinIAMsthatincludespecifictechnologiesbasedonthefeaturesofthosetechnologies,butinIAMsthatlacktechnologicaldetail,moregenericlimitsoffuelswitchingintheindustrysectorareintegrated.MostIAMsemployaggregated,top-downindustrialsectormodelsthatarecalibratedfromlong-termhistoricaldata,suchastheintroductionofnewtechnologiesorfuels.Asaresult,thesemodelscanimplicitlyreflectreal-worldconstraintsintheentiresectorthatbottom-upapproachesmaynotcompletelyexplore.Theseconstraintsmayresultfromfactorssuchasinfrastructurebuildingdelaysormarketparticipants’insufficientunderstandingofnewtechnology.Furthermore,becauseIAMsmodeltheclimatesystem,thesemodelscanmainlyaccountfortheeffectsofclimatechangeonthegrowthandstructureofeconomies(Pauliuketal.2017).However,top-downmodelsareoftenlimitedintheirportrayalofspecifictechnologiesandprocessesintheindustrysector,especiallyoftechnology-drivenstructuralchange.Thislackoftechnologicalinformationrestrictsthemodels’utilityinanalyzingtechnology-andsector-specificmitigationmethodsandpolicies.Top-downmodelsalsofeatureahighlyaggregateddepictionofindustrialenergydemand,makingitdifficulttoevaluatedemand-sidemitigationtechniquessuchasrecycling,product-serviceefficiency,anddemandreductionchoices(Pauliuketal.2017).ResourcesfortheFuture247.TheModelingofInnovationandTechnologicalChangeandRelevantPoliciesThissectionprovidesabriefoverviewofthedifferentapproachestoportraytechnologicalchangeinlow-carbonemissionstechnologies,includingradicallynoveltechnologies,relevantforindustrialsectors.Asnotedintheprevioussection,thedifferentmethodologiestomodelinnovationmaycoexistinagivenmodel.Anotherrelatedkeypointregardstheabilityofmodelstomimichowspecificpoliciesandpolicyinstrumentsaffectthedifferentphasesoftheinnovationprocess.InnovationcomprisesstagesrangingfrombasicR&Dtoprototyping,demonstration,andlarger-scalemarketdiffusion.Howamodeldoesordoesnotaccountforallthesedifferentstagesaffectsthespeedanddepthofemissionsreductions(Blancoetal.2022).Inmostmodels,anoveltechnologyentersthemarketwhenitscostsdecreaseandbecomeequalto,orlowerthan,thecostsofthedominanttechnology(vanSluisveldetal.2020).Technologycostdecreasecanbecalibratedusinghistoricaldataorrelyingonexpertestimates.Theformerreliesonthelearningcurveapproach:itassumesthatcostsdecreaseeitherasafunctionofR&Dinvestment(learning-by-researching)orasafunctionofcumulativeproductionortime(Nagyetal.2013).Learningcurvemodelparametersarederivedfromhistoricaldataandthenusedtoprojectfuturecostdecreases.Thelatterapproach,whichreliesonexpertestimatesregardingthetrajectoryoffuturecosts,hasbeenusedforradicallynoveltechnologies,forwhichhistoricaldataarenotavailable(Verdolinietal.2018).Thiswasthecaseforthecostofnuclearpowerinthe1970sandthecostofCCSandotherradicallynoveltechnologiesmorerecently.Recentevidencehasshownthatbothlearningcurveandexpert-basedapproachesunderestimatethecostreductionsinseverallow-carbontechnologies,withexpert-basedmethodsemergingasparticularlypessimistic.Yettheseresultsmaynotnecessarilyextendtomoreradicaltechnologies.Moreover,noalternativeapproachexistsintermsofinformingthemodelingofinnovation(Mengetal.2021).Anotherimportantdistinctionbetweenmodelsiswhetherinnovationismodeledexogenouslyorendogenously.Intheformer,technologycostsareassumedtovaryovertimeatsomefixedrate,whichcanbederivedfromeitherhistoricaldataorexpertestimatesorbyrelyingonmarginalabatementcostcurvesderivedelsewhere.Inthelatter,technologycostdecreasesarecalibratedtohistoricalvalues,butcostsareachoicevariableinthemodel,andagentscandecidehowmuchtoinvestinit(Kreyetal.2019;Mercureetal.2016).Forexample,technologycostreductionscanbeassumedtofollowapredefined(historicallyobserved)patternorcanbemodeledasafunctionofR&Dinvestment,whichcanbechosen(asopposedtobeingimposed)inthemodel.Modelinginnovationexogenouslyratherthanendogenouslygenerallyunderestimatesfuturecostreductions;forinstance,itignorespolicy-inducedcarbon-savingtechnologicalchangeorspillovers.Inanycase,assumptionsregardingthespeedofinnovationandtechnologicalchangecanbetestedthroughsensitivityanalysis(Blancoetal.2022).IndustrialDeepDecarbonization:ModelingApproachesandDataChallenges25Finally,mostmodelsrelyontheassumptionthatwhenthecostofanewtechnologybecomescompetitive,thetechnologywillnaturallydiffusethroughtheeconomyfollowingacertainpattern,suchasanS-shapeddiffusionpattern(Hall2006).Thisoftenresultsinanoverestimationofthediffusionpotentialofmanynoveltechnologies,becausediffusionisdrivensolelybycostinthemodels,andnoconsiderationisgiventootherkeybarriersthatmayslowdowndeployment.Theseincludenoncost,nontechnologicalbarriersorenablersregardingbehaviors,societyandinstitutions(e.g.,pathdependenceorthecoevolutionoftechnologyclustersovertime),theriskaversionofusersandcapitalmarkets,personalpreferencesandperceptionsinaworldofheterogeneousagents,networkorinfrastructureexternalities,andalackofsupportiveinstitutionalframeworks(Iyeretal.2015;Bakeretal.2015;MarangoniandTavoni2014;vanSluisveldetal.2020;Nappetal.2017;Biresseliogluetal.2020).Ignoringthesebarriersgenerallyleadstoanoverestimationoftechnologydiffusioninsuchmodels.Toaddresstheseissues,modelscanimposeadhocrestrictionsoncertaintechnologies,suchasaceilingtopenetration.Inaddition,thesebarrierscanbeaccountedforthroughscenarionarratives,suchasthoseintheSharedSocioeconomicPathways(Riahietal.2017),inwhichassumptionsabouttechnologyadoptionspanaplausiblerangeofvalues.Theliteraturealsoindicatesthatmodelstendtounderestimatecostreductionpotentialsbuttooverestimatepenetrationrates.Carefulcalibrationandsensitivityanalysisarenecessarytotesttherobustnessofmodelresultsregardingtechnologyinnovationanddiffusion(Blancoetal.2022).Giventhekeyrolethatpoliciesplayintheinnovationprocess,itisparamounttounderstandhowtheycanbemodeledandaccountedfor.Alargenumberofcontributionshaveexploredhowdifferentpolicyinstrumentsinfluencetheavailabilityofnoveltechnologies,costdecreasesovertime,andtechnologydiffusioninthemarket.Policyinstrumentsaretraditionallycategorizedassupply-sidepolicies,whichtargettechnologyinnovationdirectlyintheformofR&Dinvestmentsorsubsidiesforresearch,anddemand-sidepolicies,whichincludebothcommand-and-controlpoliciessuchasemissionslimitsandmarket-basedpoliciessuchastaxesorpermits(IPCC2022).Ageneralresultemergingfromthisliteratureisthatlow-emissionsinnovationandtechnologydiffusioncanbeeffectivelysupportedthroughpolicypackagestailoredtonationalcontextsandtechnologicalcharacteristics.Yetlow-emissionsinnovationcanjointlyachieveenvironmental,social,andeconomicbenefitsonlyifenvironmentalpoliciesarepartofabroader,comprehensiveandtailoredpolicypackagethataddressespotentialnegativeimpactsandstrengthensgovernanceoftheinnovationsystem(Penascoetal.2021;IPCC2022).Noavailablemodelsareabletoaccountforsuchcomplexpolicyinstruments.Firstandforemost,modelsportrayinginnovationandtechnologydiffusionexogenouslycanaccountfortheimpactofdifferentpolicyinstrumentsoncostanddiffusiondynamicsonlythroughsensitivityanalysis.Second,evenmodelsthatrepresentinnovationasanendogenousprocessincludeonlyalimitedsetofpolicyinstruments:taxesormarketsforpermits,emissionslimits,andemissionsstandards.Thepotentialroleofinstrumentssuchaspublicprocurement,public-privatepartnership,voluntaryindustrystandards,orthenuancesofpolicydesignthataffecttheeffectivenessofthepolicyinstrumentcannotbestudiedindepth,andtheseinstrumentsarethusoverlookedorimplicitlyassumed.Finally,anumberofnonenvironmentalpolicies,suchasthosetargetinginflationortheeaseofaccessingcapital,caninfluenceinnovation.Theseoftenarenotappropriatelyaccountedforinmodelsofindustrialenergydemand.ResourcesfortheFuture268.SpecificModelsforIndustryEnergyDemandandEmissionsThissectiongivesanoverviewofthemostwidelyknownmodelsusedtoforecastindustrialenergydemand.DetailsareprovidedinAppendixC,whereweclassifyeachmodelbasedonthethreecriteriadiscussedintheprevioussection(top-downorbottom-up,simulationoroptimization,andlevelofgranularityoftheindustrymodule),describethemechanismsthroughwhichthespecificmodelrepresentsindustryenergydemandandthemainassumptionsmade,andnoteanapplicationintheliterature.Giventhenatureofthispaper,weexcludemodelsinwhichtheentireindustrialsectorisportrayedasasinglesector.ThemodelsanalyzedforthispaperandpresentedinAppendixCareasfollows:1.WorldEnergyModel,InternationalEnergyAgency2.NationalEnergyModelingSystem,USEnergyInformationAdministration3.GlobalChangeAssessmentModel,PacificNorthwestNationalLaboratory4.RegionalModelofInvestmentandDevelopment,PotsdamInstituteforClimateImpactResearch5.ModularEnergySystemSimulationEnvironment,ImperialCollegeLondon6.TheIntegratedMARKAL-EFOMSystem,ImperialCollegeLondon,GranthamInstitute7.IMAGE,PBLNetherlandsEnvironmentalAssessmentAgency8.MaterialEconomicsModellingFramework,MaterialEconomics9.Energy-Environment-EconomyGlobalMacro-Economic,CambridgeEconometrics10.IndustrialSectorEnergyEfficiencyModelforIronandSteel,LawrenceBerkeleyNationalLaboratory11.UniversalIndustrialSectorsIntegratedSolutions,USEnvironmentalProtectionAgency12.HybridTechnologicalEconomicPlatform,CENSEandCollegeofWilliamandMary13.FORECAST,FraunhoferInstituteforSystemsandInnovationResearchAmainresultofthisanalysisrelatestothegranularitywithwhichradicaltechnologiesaremodeled.ManyoftheenergysystemmodelsandtheintegratedassessmentmodeldescribedinAppendixCarenotdetailedenoughtomodelseparatelysomeofthespecifickeymitigationoptionsdescribedinSection4—suchastheelectrowinningprocessormoltenoxideelectrolysis—indifferentenergy-intensivesectors.Manyprovideonlyhigh-leveldetailsontheprocessofinnovationanditsdirection.Thisisalsotrueinthecaseofdetailedsectoralmodels,whichgenerallyarenottechnology-orsubtechnology-specific.Moreover,thesemodelsdonotdescribecircularitypractices,suchasthedifferencebetweenconventionalandnewtypesofSCMs,mostlybecausetheydonotmodelmaterialflowsbutrathermodelthecostofagiventechnologyoritsenergydemandandefficiency.Importantly,effortsarecurrentlyunderwaytointegratethemodelingofGHGemissionswiththemodelingofmaterialflows,yetthisprocesshasproventobechallenging,andnomodeliscurrentlyavailablethatincorporatessimultaneouslyattentiontoGHGemissionsandthematerialsideoftheeconomy.IndustrialDeepDecarbonization:ModelingApproachesandDataChallenges279.ConclusionsTheaimofthispaperistoimprovetheunderstandingofchallengeslinkedwiththemodelingofemissionsandenergydemandinkeyenergy-intensiveindustrialsectors,withaparticularfocusontherolethatnewlow-carbontechnologiescanplayinachievingdeepmitigationtargets.Withtheaimofhighlightingthesectoralpeculiaritiesofvariousemissionsreductionstrategies,wefirstdiscusstherelevanceofdifferentemissionsreductionapproaches—fuelswitchingandelectrification,carbonefficiency,materialefficiency,carboncaptureandstorage,andcirculareconomypractices—forsixhigh-energy-demandsectors:steel,cement,chemicals,lightmanufacturing,aluminum,andpulpandpaper.Tohighlightthelimitationsofmodelingindustrialenergydemandandassociatedemissions,wethendetailthedatalimitationsthatconstrainmodelcalibrationanddescribethemethodologiesthatcharacterizethemostwell-knownintegratedassessmentmodelsofindustrialenergydemand.Threekeyinsightsemergefromthisanalysis.Firstandforemost,modelsneedtobefurtherdevelopedtoappropriatelycaptureindustrialdecarbonizationoptionsandtheeffectsofpolicies.Noneofthewidelyusedindustrialenergyandemissionsmodelshavethecapacitytoportraytheadoptionanddiffusionofgranulartechnologicaloptionsforemissionsreductioninenergy-intensivesectors.Nomodelcanportrayheterogeneousinnovationandtechnologyadoptiondynamicsduetofirmcharacteristics(e.g.,sizeoraccesstocapital).Mostavailablemodels,includingthosewithrelativelyhighsectoraldetail,arenottechnology-orsub-technology-specificandrelyononlyahigh-levelrepresentationoftheinnovationprocess.Moreover,availableIAMsdonottrackmaterialflowsandconsequentlycannotdescribecircularitypracticesandtheirrelevanceformitigation.Whileeffortsinthisrespectareunderway,muchworkremainsaheadfortheresearchcommunity.Inthisrespect,thesoftorhardlinkingofavailablemodelswithothertechnology-specificmoredetailedmodels,includingagent-basedmodelsemergesasanimportantresearchavenueforthefuture.Dataavailabilityalsorepresentsakeybarrierformodeldevelopment.Datacollectioneffortsareinadequateandneedtobescaledup.However,markedfirmheterogeneitybothwithinandacrosssectorsisamainbarrierinthisrespect.Giventhelackofacoordinated,state-drivenefforttogatherstatisticsonindustrialenergydemandanduse,researchershavedifficultyobtainingthenecessarydata.Whenentrepreneursarewillingtoshare,dataareoftenlimitedandlackpaneldimensionorcannotbecomparedacrosscountriesandsectorsbecauseofalackofprecisecollectionstandards.Thissituationcouldbepartlyresolvedifpolicymakerscreateaframeworkaimedatfacilitatingthesharingofdataregardingtheenvironmentalperformanceofsmallandmedium-sizeenterprisesorentrustnationalstatisticalofficeswiththistask.Thiswouldallowresearcherstorelyondatawhoseconsistencyhasbeencertifiedbynumerousstudies,anddataharmonizationwouldrequirelesstimethanwhenperformingthesameoperationsmultipletimesonthesamedataset.ResourcesfortheFuture28Lastbutnotleast,whileaportfolioofdifferentoptionsisavailableinallindustrialsectorstoreduceenergydemand,existingtechnologiesarenotsufficienttoachievedeepdecarbonizationgoalsacrossindustrialsectors.Norisitpossibletoachievethesegoalsbyrelyingononlyonetechnologicaloption.ThusR&Donrelevantdecarbonizationtechnologiesneedstobespeduptoensurethefurtherdevelopmentofadditionallow-carbonindustrialtechnologiesinallenergy-intensivesectorsandmakeavailableotherdecarbonizationsolutionsthatarenotinuseyet.IndustrialDeepDecarbonization:ModelingApproachesandDataChallenges29ReferencesABIResearch.2022.“AutomotiveSectorLeadsIndustrialDigitalTransformationChargewithUS$238BillionInvestment,FollowedbyElectronics,Oil&Gas,andFMCG.”EMSNow.November2,2022.https://www.emsnow.com/automotive-sector-leads-industrial-digital-transformation-charge-with-us238-billion-investment-followed-by-electronics-oil-gas-and-fmcg/.Aguiar,Angel,MaksymChepeliev,ErwinL.Corong,andDominiquevanderMensbrugghe.2023.“TheGTAPDataBase:Version11.”JournalofGlobalEconomicAnalysis,7(2).https://doi.org/10.21642/JGEA.070201AFArora,Vipin,DavidDaniels,IanMead,andRusselTarver.2018.“EMF32ResultsfromNEMS:RevenueRecycling.”ClimateChangeEconomics9(1):1840014.https://doi.org/10.1142/s2010007818400146.Asiaban,Siavash,NezminKayedpour,ArashE.Samani,DimitarBozalakov,JeroenD.M.DeKooning,GuillaumeCrevecoeur,andLievenVandevelde.2021.“WindandSolarIntermittencyandtheAssociatedIntegrationChallenges:AComprehensiveReviewIncludingtheStatusintheBelgianPowerSystem.”MDPI,May4,2021.https://doi.org/10.3390/en14092630.Ayati,Bamdad,DarrylNewport,HongWong,andChristopherCheeseman.2022.“Low-CarbonCements:PotentialforLow-GradeCalcinedClaystoFormSupplementaryCementitiousMaterials.”CleanerMaterials5:100099.https://doi.org/10.1016/j.clema.2022.100099.Bachner,G.,J.Mayer,K.W.Steininger,A.Anger-Kraavi,A.Smith,andT.S.Barker.2020.“UncertaintiesinMacroeconomicAssessmentsofLow-CarbonTransitionPathways:TheCaseoftheEuropeanIronandSteelIndustry.”EcologicalEconomics172(June):106631.Baker,Erin,OlaitanOlaleye,andLaraAleluiaReis.2015.“DecisionFrameworksandtheInvestmentinR&D.”EnergyPolicy80(May):275–85.https://doi.org/10.1016/j.enpol.2015.01.027.Baumstark,Lavinia,NicoBauer,FalkBenke,ChristophBertram,StephenBi,ChenChrisGong,JanPhilippDietrich,etal.2021.“REMIND2.1:TransformationandInnovationDynamicsoftheEnergy-EconomicSystemwithinClimateandSustainabilityLimits.”GeoscientificModelDevelopment14(10):6571–6603.https://doi.org/10.5194/gmd-14-6571-2021.Bazzanella,Alexis,andFlorianAusfelder.2017.“LowCarbonEnergyAndFeedstockForTheEuropeanChemicalIndustry.”DECHEMA,GesellschaftfürChemischeTechnikundBiotechnologie.https://books.google.it/books?id=S8fetAEACAAJ.Bhander,GurbakhashandWojciechJozewicz.2017.“UniversalIndustrialSectorsIntegratedSolutionsModuleforthePulpandPaperIndustry.”NordicPulpandPaperResearchJournal32(3):375-385.https://cfpub.epa.gov/si/si_public_record_report.cfm?Lab=NRMRL&dirEntryId=338124Biresselioglu,MehmetEfe,MuhittinHakanDemir,MelikeDemirbagKaplan,andBerfuSolak.2020.“Individuals,Collectives,andEnergyTransition:AnalysingtheMotivatorsandBarriersofEuropeanDecarbonisation.”EnergyResearch&SocialScience66(August):101493.https://doi.org/10.1016/j.erss.2020.101493.ResourcesfortheFuture30Blanco,G.,H.deConinck,L.Agbemabiese,E.H.MbayeDiagne,L.DiazAnadon,Y.S.Lim,W.A.Pengue,etal.2022.Innovation,technologydevelopmentandtransfer.InClimateChange2022:MitigationofClimateChange.ContributionofWorkingGroupIIItotheSixthAssessmentReportoftheIntergovernmentalPanelonClimateChange,editedbyP.R.Shukla,J.Skea,R.Slade,A.AlKhourdajie,R.vanDiemen,D.McCollum,M.Pathak,etal.Cambridge:CambridgeUniversityPress.https://doi.org/10.1017/9781009157926.018.Bosetti,Valentina.2021.“IntegratedAssessmentModelsforClimateChange.”OxfordResearchEncyclopediaofEconomicsandFinance.January22,2021.https://doi.org/10.1093/acrefore/9780190625979.013.572.Brandl,Patrick,MaiBui,JasonP.Hallet,NiallMacDowell.2021.“Beyond90%Capture:Possible,butatWhatCost?”InternationalJournalofGreenhouseGasControl105:103239.https://doi.org/10.1016/j.ijggc.2020.103239.Brown,MarilynA.,andYoungsunBaek.2010.“TheForestProductsIndustryatanEnergy/ClimateCrossroads.”EnergyPolicy38(12):7665–75.https://doi.org/10.1016/j.enpol.2010.07.057.Budinis,Sara,JuliaSachs,SaraGiarola,andAdamHawkes.2020.“AnAgent-BasedModellingApproachtoSimulatetheInvestmentDecisionofIndustrialEnterprises.”JournalofCleanerProduction267(September):121835.https://doi.org/10.1016/j.jclepro.2020.121835.DucBui,Viet,HoangPhuongVu,HoangPhuongNguyen,XuanQuangDuong,DinhTuyenNguyen,MinhTuanPham,andPhuocQuyPhongNguyen.“Techno-EconomicAssessmentandLogisticsManagementofBiomassintheConversionProgresstoBioenergy.”SustainableEnergyTechnologiesandAssessments55(February2023):102991.https://doi.org/10.1016/j.seta.2022.102991.CambridgeEconometrics.2019.E3METechnicalManualv6.1.https://www.e3me.com/wp-content/uploads/sites/3/2019/09/E3ME-Technical-Manual-v6.1-onlineSML.pdf.Chen,Hsing-Hsuan,AndriesF.Hof,VassilisDaioglou,HarmenSytzedeBoer,OreaneY.Edelenbosch,MaartenvandenBerg,Kaj-IvarvanderWijst,andDetlefP.vanVuuren.2021.“UsingDecompositionAnalysistoDeterminetheMainContributingFactorstoCarbonNeutralityacrossSectors.”Energies15(1):132.https://doi.org/10.3390/en15010132.Creutzig,Felix,DaronAcemoglu,XuemeiBai,PaulN.Edwards,MarieJosefineHintz,LynnH.Kaack,SiirKilkis,etal.2022.“DigitalizationandtheAnthropocene.”AnnualReviewofEnvironmentandResources47:479–509.https://ssrn.com/abstract=4252383orhttp://dx.doi.org/10.1146/annurev-environ-120920-100056CTCN(ClimateTechnologyCentre&Network).2016.“ClinkerReplacement.”UNClimateTechnologyCentre&Network.https://www.ctc-n.org/technologies/clinker-replacement.Daehn,KatrinE.,AndréCabreraSerrenho,andJulianM.Allwood.2017.“HowWillCopperContaminationConstrainFutureGlobalSteelRecycling?”EnvironmentalScience&Technology51(11):6599–6606.https://doi.org/10.1021/acs.est.7b00997.D’Amato,Dalia,JouniKorhonen,andAnneToppinen.2019.“Circular,Green,andBioEconomy:HowDoCompaniesinLand-UseIntensiveSectorsAlignWithSustainabilityConcepts?”EcologicalEconomics158:116–33.https://doi.org/10.1016/j.ecolecon.2018.12.026.IndustrialDeepDecarbonization:ModelingApproachesandDataChallenges31Dean,Charles,JohnKokot-Blamey,NickFlorin,MohamadJ.Al-Jeboori,andPaulFennell.2011.“TheCalciumLoopingCycleforCO2CapturefromPowerGeneration,CementManufactureandHydrogenProduction.”ChemicalEngineeringResearchandDesign89(6):836–55.https://doi.org/10.1016/j.cherd.2010.10.013.EDGARV7.0.2022.EDGAR:EmissionsDatabaseforGlobalAtmosphericResearch.https://edgar.jrc.ec.europa.eu/dataset_ghg70.EIA(USEnergyInformationAdministration).2016.InternationalEnergyOutlook2016.https://www.eia.gov/outlooks/ieo/pdf/industrial.pdf.———.2019.TheNationalEnergyModelingSystem.https://www.eia.gov/outlooks/aeo/nems/overview/pdf/0581(2018).pdf———.2022.IndustrialDemandModule:NEMSDocumentation.https://www.eia.gov/analysis/pdfpages/m064index.php.EPA(USEnvironmentalProtectionAgency).2013.UniversalIndustrialSectorsIntegratedSolutions(U-ISIS)ModelforthePortlandCementManufacturingIndustry.EPA/600/R-14/089.Washington,DC:EPA.EU(EuropeanUnion).2020.ShapingEurope’sDigitalFuture.PublicationsOfficeoftheEuropeanUnion.https://ec.europa.eu/info/sites/default/files/communication-shaping-europes-digital-future-feb2020_en_4.pdf.EuropeanCommission.2018.ACleanPlanetforAll:AEuropeanStrategicLong-termVisionforaProsperous,Modern,CompetitiveandClimateNeutralEconomy.CommunicationfromtheCommissiontotheEuropeanParliament,theEuropeanCouncil,theCouncil,TheEuropeanEconomicandSocialCommittee,theCommitteeoftheRegionsandtheEuropeanInvestmentBankCOM(2018)773final.Brussels,28.11.2018.https://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=CELEX:52018DC0773EuropeanParliament.2015.Industry4.0.https://www.europarl.europa.eu/RegData/etudes/BRIE/2015/568337/EPRS_BRI(2015)568337_EN.pdf.Eurostat.2008.NACERev.2:StatisticalClassificationofEconomicActivitiesintheEuropeanCommunity.Luxembourg:OfficeforOfficialPublicationsoftheEuropeanCommunities.https://ec.europa.eu/eurostat/documents/3859598/5902521/KS-RA-07-015-EN.PDF.———.2022.“ElectricityandGasPricesintheFirstHalfof2022.”ProductsEurostatNews.https://ec.europa.eu/eurostat/web/products-eurostat-news/-/ddn-20221031-1.Eyre,Nick.2021.“FromUsingHeattoUsingWork:ReconceptualisingtheZeroCarbonEnergyTransition.”EnergyEfficiency14(77).https://doi.org/10.1007/s12053-021-09982-9.Fais,Birgit,NagoreSabio,andNeilStrachan.2016.“TheCriticalRoleoftheIndustrialSectorinReachingLong-TermEmissionReduction,EnergyEfficiencyandRenewableTargets.”AppliedEnergy162(January):699–712.https://doi.org/10.1016/j.apenergy.2015.10.112.Fan,Zhiyuan,andS.JulioFriedmann.“Low-CarbonProductionofIronandSteel:TechnologyOptions,EconomicAssessment,andPolicy.”Joule5,no.4(April2021):829–62.https://doi.org/10.1016/j.joule.2021.02.018.Fleiter,T.,M.Rehfeldt,A.Herbst,R.Elsland,A.Klingler,P.Manz,andS.Eidelloth.2018.“AMethodologyforBottom-UpModellingofEnergyTransitionsintheIndustrySector:TheFORECASTModel.”EnergyStrategyReviews22:237–54.https://doi.org/10.1016/j.esr.2018.09.005.ResourcesfortheFuture32Fleiter,T.,Rehfeldt,M.,andPfluger,B.(Eds.)(2016)AtransitionpathwayforGermany’sindustry:whichroleforenergyefficiency?ECEEEIndustrialSummerStudy,September12-14,Berlin.Fortes,Patrícia,RuiPereira,AlfredoPereira,andJúliaSeixas.“IntegratedTechnological-EconomicModelingPlatformforEnergyandClimatePolicyAnalysis.”Energy73(August2014):716–30.https://doi.org/10.1016/j.energy.2014.06.075.GCIMS(GlobalChangeIntersectoralModelingSystem).GCAM:GlobalChangeAnalysisModel.https://gcims.pnnl.gov/modeling/gcam-global-change-analysis-modelGiarola,Sara,JuliaSachs,Mayeuld’Avezac,AlexanderKell,andAdamHawkes.2022.“MUSE:AnOpen-SourceAgent-BasedIntegratedAssessmentModellingFramework.”https://doi.org/10.21203/rs.3.rs-1450486/v1.Gramkow,Camila,andAnnelaAnger-Kraavi.2019.“DevelopingGreen:ACasefortheBrazilianManufacturingIndustry.”Sustainability11(23):6783.https://doi.org/10.3390/su11236783.Haberl,Helmut,DominikWiedenhofer,StefanPauliuk,FridolinKrausmann,DanielB.Müller,andMarinaFischer-Kowalski.2019.“ContributionsofSocio-metabolicResearchtoSustainabilityScience.”NatureSustainability2(3):173–84.https://doi.org/10.1038/s41893-019-0225-2.Hall,BronwynH.2006.“InnovationandDiffusion.”InTheOxfordHandbookofInnovation,editedbyJanFagerbergandDavidC.Mowery,459–84.Oxford:OxfordUniversityPress.https://doi.org/10.1093/oxfordhb/9780199286805.003.0017.Hansen,Paula,XinLiu,andGregoryM.Morrison.2019.“Agent-BasedModellingandSocio-TechnicalEnergyTransitions:ASystematicLiteratureReview.”EnergyResearch&SocialScience49(March):41–52.https://doi.org/10.1016/j.erss.2018.10.021.Harris,ZoeM.,SuzanneMilner,andGailTaylor.2018.“BiogenicCarbon—CaptureandSequestration.”InGreenhouseGasesBalancesofBioenergySystems,editedbyPatriciaThornleyandPaulAdams,55–76.London:Elsevier.https://doi.org/10.1016/b978-0-08-101036-5.00005-7.Hepburn,Cameron,EllaAdlen,JohnBeddington,EmilyA.Carter,SabineFuss,NiallMacDowell,JanC.Minx,PeteSmith,andCharlotteK.Williams.2019.“TheTechnologicalandEconomicProspectsforCO2UtilizationandRemoval.”Nature575(7781):87–97.https://doi.org/10.1038/s41586-019-1681-6.Herbst,Andrea,FelipeToro,FelixReitze,andEberhardJochem.2012.“IntroductiontoEnergySystemsModelling.”SwissJournalofEconomicsandStatistics148(2):111–35.https://doi.org/10.1007/bf03399363.Honus,Stanislav,ShogoKumagai,GabrielFedorko,VieroslavMolnár,andToshiakiYoshioka.2018.“PyrolysisGasesProducedfromIndividualandMixedPE,PP,PS,PVC,andPET—PartI:ProductionandPhysicalProperties.”Fuel221(June):346–60.https://doi.org/10.1016/j.fuel.2018.02.074.Huang,Kaixin,andMatthewJ.Eckelman.2020.“AppendingMaterialFlowstotheNationalEnergyModelingSystem(NEMS)forProjectingthePhysicalEconomyoftheUnitedStates.”JournalofIndustrialEcology26(1):294–308.https://doi.org/10.1111/jiec.13053.———.2021.“EstimatingFutureIndustrialEmissionsofHazardousAirPollutantsintheUnitedStatesUsingtheNationalEnergyModelingSystem(NEMS).”Resources,ConservationandRecycling169(June):105465.https://doi.org/10.1016/j.resconrec.2021.105465.IndustrialDeepDecarbonization:ModelingApproachesandDataChallenges33Huang,Yun-Hsun,Yi-LinChang,andTobiasFleiter.2016.“ACriticalAnalysisofEnergyEfficiencyImprovementPotentialsinTaiwan’sCementIndustry.”EnergyPolicy96:14–26.https://doi.org/10.1016/j.enpol.2016.05.025.Hughes,Sydney,andAlexanderZoelle,“CostofCapturingCO2fromIndustrialSources,”NationalEnergyTechnologyLaboratory,Pittsburgh,July15,2022.https://netl.doe.gov/projects/files/CostofCapturingCO2fromIndustrialSources_071522.pdfICEF(InnovationforCoolEarthForum).2020.BiomassCarbonRemovalandStorage(BiCRS)Roadmap.https://www.icef.go.jp/pdf/summary/roadmap/icef2020_roadmap.pdf.IEA(InternationalEnergyAgency).2015.WorldEnergyOutlook2015.Paris:IEA.https://www.iea.org/reports/world-energy-outlook-2015———.2018.TechnologyRoadmap:Low-CarbonTransitionintheCementIndustry.Paris:IEA.https://www.iea.org/reports/technology-roadmap-low-carbon-transition-in-the-cement-industry.———.2019.MaterialEfficiencyinCleanEnergyTransitions.Paris:IEA.https://www.iea.org/reports/material-efficiency-in-clean-energy-transitions.———.2020.IronandSteelTechnologyRoadmap.Paris:IEA.https://www.iea.org/reports/iron-and-steel-technology-roadmap.———.2021.WorldEnergyModelDocumentation.Paris:IEA.https://iea.blob.core.windows.net/assets/932ea201-0972-4231-8d81-356300e9fc43/WEM_Documentation_WEO2021.pdf———.2022a.Aluminium.Paris:IEA.https://www.iea.org/reports/aluminium.———.2022b.Cement.Paris:IEA.https://www.iea.org/reports/cement.———.2022c.Industry.Paris:IEA.https://www.iea.org/reports/industry.———.2022d.LightIndustry.Paris:IEA.https://www.iea.org/reports/light-industry.———.2022e.PulpandPaper.Paris:IEA.https://www.iea.org/reports/pulp-and-paper.———.2022f.WorldEnergyOutlookDatabase2022.Paris:IEA.https://www.iea.org/data-and-statistics/data-product/world-energy-balances.IPCC(IntergovernmentalPanelonClimateChange).2022.ClimateChange2022:MitigationofClimateChange.ContributionofWorkingGroupIIItotheSixthAssessmentReportoftheIntergovernmentalPanelonClimateChange,editedbyP.R.Shukla,J.Skea,R.Slade,A.AlKhourdajie,R.vanDiemen,D.McCollum,M.Pathak,etal.Cambridge:CambridgeUniversityPress.https://doi.org/10.1017/9781009157926.Isikgor,FurkanH.,andC.RemziBecer.2015.“LignocellulosicBiomass:ASustainablePlatformfortheProductionofBio-BasedChemicalsandPolymers.”PolymerChemistry6(25):4497–559.https://doi.org/10.1039/c5py00263j.Iyer,Gokul,NathanHultman,JiyongEom,HaewonMcJeon,PralitPatel,andLeonClarke.2015.“DiffusionofLow-CarbonTechnologiesandtheFeasibilityofLong-TermClimateTargets.”TechnologicalForecastingandSocialChange90(January):103–18.https://doi.org/10.1016/j.techfore.2013.08.025.Junjie,Yan.2018.“ProgressandFutureofBreakthroughLow-CarbonSteelmakingTechnology(ULCOS)ofEU.”InternationalJournalofMineralProcessingandExtractiveMetallurgy3(2):15–22.https://doi.org/10.11648/j.ijmpem.20180302.11.ResourcesfortheFuture34Karali,Nihan,XuTenfang,andJayantSathaye.2013.IndustrialSectorEnergyEfficiencyModelling(ISEEM)FrameworkDocumentation.LBNL-6078E.ErnestOrlandoLawrenceBerkeleyNationalLaboratory.https://www.osti.gov/servlets/purl/1172249.———.2013.GreenhouseGasMitigationOptionsinISEEMGlobalEnergyModel:2010–2050ScenarioAnalysisforLeast-CostCarbonReductioninIronandSteelSector.LBNL-65550E.ErnestOrlandoLawrenceBerkeleyNationalLaboratory.https://www.osti.gov/servlets/purl/1129569.Kätelhön,Arne,RaoulMeys,SarahDeutz,SangwonSuh,andAndréBardow.2019.“ClimateChangeMitigationPotentialofCarbonCaptureandUtilizationintheChemicalIndustry.”ProceedingsoftheNationalAcademyofSciences116(23):11187–94.https://doi.org/10.1073/pnas.1821029116.Kelemen,Peter,SallyM.Benson,HélènePilorgé,PeterPsarras,andJenniferWilcox.2019.“AnOverviewoftheStatusandChallengesofCO2StorageinMineralsandGeologicalFormations.”FrontiersinClimate1(November).https://doi.org/10.3389/fclim.2019.00009.Kermeli,Katerina,OreaneY.Edelenbosch,WinaCrijns-Graus,BasJ.vanRuijven,SilvanaMima,DetlefP.vanVuuren,andErnstWorrell.2019.“TheScopeforBetterIndustryRepresentationinLong-TermEnergyModels:ModelingtheCementIndustry.”AppliedEnergy240(April):964–85.https://doi.org/10.1016/j.apenergy.2019.01.252.Krey,Volker,FeiGuo,PeterKolp,WenjiZhou,RobertoSchaeffer,AayushiAwasthy,ChristophBertram,etal.2019.“LookingundertheHood:AComparisonofTechno-EconomicAssumptionsacrossNationalandGlobalIntegratedAssessmentModels.”Energy172(April):1254–67.https://doi.org/10.1016/j.energy.2018.12.131.Kuramochi,Takeshi,AndreaRamírez,WimTurkenburg,andAndréFaaij.2012.“ComparativeAssessmentofCO2CaptureTechnologiesforCarbon-IntensiveIndustrialProcesses.”ProgressinEnergyandCombustionScience38(1):87–112.https://doi.org/10.1016/j.pecs.2011.05.001.Liu,Lu,MohamadHejazi,PralitPatel,PageKyle,EvanDavies,YuyuZhou,LeonClarke,andJamesEdmonds.2015.“WaterDemandsforElectricityGenerationintheU.S.:ModelingDifferentScenariosfortheWater–EnergyNexus.”TechnologicalForecastingandSocialChange94(May):318–34.https://doi.org/10.1016/j.techfore.2014.11.004.Lopion,Peter,PeterMarkewitz,MartinRobinius,andDetlefStolten.2018.“AReviewofCurrentChallengesandTrendsinEnergySystemsModeling.”RenewableandSustainableEnergyReviews96(November):156–66.https://doi.org/10.1016/j.rser.2018.07.045.Loulou,Richard,andMaryseLabriet.2008.“ETSAP-TIAM:TheTIMESIntegratedAssessmentModelPartI:ModelStructure.”ComputationalManagementScience5(1–2):7–40.https://doi.org/10.1007/s10287-007-0046-z.Luderer,Gunnar,ValentinaBosetti,MichaelJakob,MarianLeimbach,JanC.Steckel,HenriWaisman,andOttmarEdenhofer.2012.“TheEconomicsofDecarbonizingtheEnergySystem:ResultsandInsightsfromtheRECIPEModelIntercomparison.”ClimaticChange114(1):9–37.https://doi.org/10.1007/s10584-011-0105-x.Madeddu,Silvia,FalkoUeckerdt,MichajaPehl,JuergenPeterseim,MichaelLord,KarthikAjithKumar,ChristophKrüger,andGunnarLuderer.2020.“TheCO2ReductionPotentialfortheEuropeanIndustryviaDirectElectrificationofHeatSupply(Power-to-Heat).”EnvironmentalResearchLetters15(12):124004.https://doi.org/10.1088/1748-9326/abbd02.IndustrialDeepDecarbonization:ModelingApproachesandDataChallenges35Marangoni,Giacomo,andMassimoTavoni.2014.“TheCleanEnergyR&DStrategyfor2°C.”ClimateChangeEconomics5(1):1440003.https://doi.org/10.1142/s201000781440003xMaterialEconomics.2019.IndustrialTransformation2050:PathwaystoNet-ZeroEmissionsfromEUHeavyIndustry.https://materialeconomics.com/publications/industrial-transformation-2050.Meng,Jing,RupertWay,ElenaVerdolini,andLauraDiazAnadon.2021.“ComparingExpertElicitationandModel-BasedProbabilisticTechnologyCostForecastsfortheEnergyTransition.”ProceedingsoftheNationalAcademyofSciences118(27).https://doi.org/10.1073/pnas.1917165118.Mercure,Jean-François,FlorianKnobloch,HectorPollitt,RichardLewney,K.W.J.F.Rademakers,LisaEichler,JeroenvanderLaan,andLeonidasParoussos.2016.Policy-InducedEnergyTechnologicalInnovationandFinanceforLow-CarbonEconomicGrowth:StudyontheMacroeconomicsofEnergyandClimatePolicies.EuropeanCommission,Directorate-GeneralforEnergy.Nagy,Béla,J.DoyneFarmer,QuanM.Bui,andJessikaE.Trancik.StatisticalBasisforPredictingTechnologicalProgress.EditedbyLuísA.NunesAmaral.PLoSONE8,no.2(February28,2013):e52669.https://doi.org/10.1371/journal.pone.0052669.NAICS(NorthAmericanIndustryClassificationSystem).2017.ExecutiveOfficeofthePresident,OfficeofManagementandBudget.USA,2017.https://www.census.gov/naics/reference_files_tools/2017_NAICS_Manual.pdfNandhini,Rajendran,BaskaranSivaprakash,andNatarajanRajamohan.2022.“WasteHeatRecoveryatLowTemperaturefromHeatPumps,PowerCyclesandIntegratedSystems:ReviewonSystemPerformanceandEnvironmentalPerspectives.”SustainableEnergyTechnologiesandAssessments52(August):102214.https://doi.org/10.1016/j.seta.2022.102214.Napp,Tamaryn,DanBernie,RebeccaThomas,JasonLowe,AdamHawkes,andAjayGambhir.2017.“ExploringtheFeasibilityofLow-CarbonScenariosUsingHistoricalEnergyTransitionsAnalysis.”Energies10(1):116.https://doi.org/10.3390/en10010116.Napp,T.A.,S.Few,A.Sood,D.Bernie,A.Hawkes,andA.Gambhir.2019.“TheRoleofAdvancedDemand-SectorTechnologiesandEnergyDemandReductioninAchievingAmbitiousCarbonBudgets.”AppliedEnergy238(March):351–67.https://doi.org/10.1016/j.apenergy.2019.01.033.NASEM(NationalAcademiesofSciences,Engineering,andMedicine).2019.NegativeEmissionsTechnologiesandReliableSequestration:AResearchAgenda.Washington,DC:NationalAcademiesPress.http://dx.doi.org/10.17226/25259.OMB(OfficeofManagementandBudget).2022.NorthAmericanIndustryClassificationSystem.Washington,DC:ExecutiveOfficeofthePresident,OMB.https://www.census.gov/naics/reference_files_tools/2022_NAICS_Manual.pdf.Pascale,Andrew,ChrisGreig,andClaireWayner.2021.Princeton’sNet-ZeroAmericaStudy,AnnexK:CementIndustryTransition.Princeton,NJ:PrincetonUniversity.https://netzeroamerica.princeton.edu/img/NZA%20Annex%20K%20-%20Cement%20industry.pdf.Pauliuk,Stefan,AndersArvesen,KonstantinStadler,andEdgarG.Hertwich.2017.“IndustrialEcologyinIntegratedAssessmentModels.”NatureClimateChange7(1):13–20.https://doi.org/10.1038/nclimate3148.ResourcesfortheFuture36PBL(PBLNetherlandsEnvironmentalAssessmentAgency).2021.WelcometoIMAGE3.2Documentation.LastmodifiedDecember1,2021.https://models.pbl.nl/image/index.php/Welcome_to_IMAGE_3.2_Documentation.Peñasco,Cristina,LauraDíazAnadón,andElenaVerdolini.“SystematicReviewoftheOutcomesandTrade-OffsofTenTypesofDecarbonizationPolicyInstruments.”NatureClimateChange11,no.3(January18,2021):257–65.https://doi.org/10.1038/s41558-020-00971-x.Peng,Wei,GokulIyer,MatthewBinsted,JenniferMarlon,LeonClarke,JamesA.Edmonds,andDavidG.Victor.2021.“ToAchieveDeepCutsinUSEmissions,State-DrivenPolicyIsOnlySlightlyMoreExpensiveThanNationallyUniformPolicy.”NatureClimateChange11(11):911–12.https://doi.org/10.1038/s41558-021-01193-5.Peng,Yiming,andCiseUnluer.2023.“DevelopmentofAlternativeCementitiousBindersfor3DPrintingApplications:ACriticalReviewofProgress,AdvantagesandChallenges.”CompositesPartB:Engineering252(March):110492.https://doi.org/10.1016/j.compositesb.2022.110492.Qiu,Yang,PatrickLamers,VassilisDaioglou,NoahMcQueen,Harmen-SytzedeBoer,MathijsHarmsen,JenniferWilcox,AndréBardow,andSangwonSuh.2022.“EnvironmentalTrade-offsofDirectAirCaptureTechnologiesinClimateChangeMitigationtoward2100.”NatureCommunications13(1).https://doi.org/10.1038/s41467-022-31146-1.Ragaert,Kim,LaurensDelva,andKevinVanGeem.2017.“MechanicalandChemicalRecyclingofSolidPlasticWaste.”WasteManagement69(November):24–58.https://doi.org/10.1016/j.wasman.2017.07.044.Realmonte,Giulia,LaurentDrouet,AjayGambhir,JamesGlynn,AdamHawkes,AlexandreC.Köberle,andMassimoTavoni.2019.“AnInter-ModelAssessmentoftheRoleofDirectAirCaptureinDeepMitigationPathways.”NatureCommunications10(1).https://doi.org/10.1038/s41467-019-10842-5.Riahi,Keywan,DetlefP.vanVuuren,ElmarKriegler,JaeEdmonds,BrianC.O’Neill,ShinichiroFujimori,NicoBauer,etal.2017.“TheSharedSocioeconomicPathwaysandTheirEnergy,LandUse,andGreenhouseGasEmissionsImplications:AnOverview.”GlobalEnvironmentalChange42(January):153–68.https://doi.org/10.1016/j.gloenvcha.2016.05.009.Roelofsen,O.,KenSomers,EvelineSpeelman,MaaikeWitteveen.2020.“PluggingIn:WhatElectrificationCanDoforIndustry.”EnergyInsightsPractice,May28,2020.McKinseyandCompany.https://www.mckinsey.com/industries/electric-power-and-natural-gas/our-insights/plugging-in-what-electrification-can-do-for-industry.Ruth,Matthias,BrynhildurDavidsdottir,andSkipLaitner.2000.“ImpactsofMarket-BasedClimateChangePoliciesontheUSPulpandPaperIndustry.”EnergyPolicy28(4):259–70.https://doi.org/10.1016/s0301-4215(00)00009-4.Sandalow,David,JulioFriedmann,RogerAines,ColinMcCormick,SeanMcCoy,andJoshuahStolaroff.2019.“ICEFIndustrialHeatDecarbonizationRoadmap.”CDRLaw.https://cdrlaw.org/resources/icef-industrial-heat-decarbonization-roadmap/.Scrivener,Karen,FernandoMartirena,ShashankBishnoi,andSoumenMaity.2018.“CalcinedClayLimestoneCements(LC3).”CementandConcreteResearch114(December):49–56.https://doi.org/10.1016/j.cemconres.2017.08.017.IndustrialDeepDecarbonization:ModelingApproachesandDataChallenges37Sharmina,M.,O.Y.Edelenbosch,C.Wilson,R.Freeman,D.E.H.J.Gernaat,P.Gilbert,A.Larkin,etal.2020.“DecarbonisingtheCriticalSectorsofAviation,Shipping,RoadFreightandIndustrytoLimitWarmingto1.5–2°C.”ClimatePolicy21(4):455–74.https://doi.org/10.1080/14693062.2020.1831430.Singh,NakshatraBahadur.2022.“ClaysandClayMineralsintheConstructionIndustry.”Minerals12(3):301.https://doi.org/10.3390/min12030301.Sun,Lu,HongLi,LiangDong,KaiFang,JingzhengRen,YongGeng,MinoruFujii,WeiZhang,NingZhang,andZheLiu.2017.“Eco-benefitsAssessmentonUrbanIndustrialSymbiosisBasedonMaterialFlowsAnalysisandEmergyEvaluationApproach:ACaseofLiuzhouCity,China.”Resources,ConservationandRecycling119:78–88.https://doi.org/10.1016/j.resconrec.2016.06.007.Tanzer,SamanthaEleanor,andAndreaRamírez.2019.“WhenAreNegativeEmissionsNegativeEmissions?”Energy&EnvironmentalScience12(4):1210–18.https://doi.org/10.1039/c8ee03338b.Timmer,MarcelP.,ErikDietzenbacher,BartLos,RobertStehrer,andGaaitzenJ.DeVries.2015.“AnIllustratedUserGuidetotheWorldInput-OutputDatabase:TheCaseofGlobalAutomotiveProduction.”ReviewofInternationalEconomics23(3):575–605.https://doi.org/10.1111/roie.12178.Turgut,Ozgu,VegardSkonsengBjerketvedt,AsgeirTomasgard,andSimonRoussanaly.2021.“AnIntegratedAnalysisofCarbonCaptureandStorageStrategiesforPowerandIndustryinEurope.”JournalofCleanerProduction329(December):129427.https://doi.org/10.1016/j.jclepro.2021.129427.UN(UnitedNations).2008.InternationalStandardIndustrialClassificationofAllEconomicActivities(ISIC),Rev.4.StatisticalPapers(Ser.M).NewYork:UN.https://unstats.un.org/unsd/publication/seriesm/seriesm_4rev4e.pdf.vanSluisveld,MariësseA.E.,AndriesF.Hof,SamuelCarrara,FrankW.Geels,MånsNilsson,KarolineRogge,BrunoTurnheim,andDetlefP.vanVuuren.2020.“AligningIntegratedAssessmentModellingwithSocio-TechnicalTransitionInsights:AnApplicationtoLow-CarbonEnergyScenarioAnalysisinEurope.”TechnologicalForecastingandSocialChange151(February):119177.https://doi.org/10.1016/j.techfore.2017.10.024.Verdolini,Elena.2023.InterlinkagesbetweentheJustEcologicalTransitionsandtheDigitalTransformation.EuropeanTradeUnionInstituteWorkingpaper,Brussels:ETUIaisbl.ISSNPDF1994-4454.https://www.etui.org/publications/interlinkages-between-just-ecological-transition-and-digital-transformation.Verdolini,Elena,LauraDìazAnadòn,ErinBaker,ValentinaBosetti,andLauraA.Reis.2018.“FutureProspectsforEnergyTechnologies:InsightsfromExpertElicitations.”ReviewofEnvironmentalEconomicsandPolicy12(1):133–53.https://www.journals.uchicago.edu/doi/abs/10.1093/reep/rex028.Vogl,Valentin,MaxÅhman,andLarsJ.Nilsson.2018.“AssessmentofHydrogenDirectReductionforFossil-FreeSteelmaking.”JournalofCleanerProduction203:736–45.https://doi.org/10.1016/j.jclepro.2018.08.279.Wiebe,KirstenS.,MarekHarsdorff,GuillermoMontt,MoanaS.Simas,andRichardWood.2019.“GlobalCircularEconomyScenarioinaMultiregionalInput-OutputFramework.”EnvironmentalScience&Technology53(11):6362-6373.https://doi.org/10.1021/acs.est.9b01208.ResourcesfortheFuture38Wiencke,Jan,HervéLavelaine,Pierre-JeanPanteix,CarinePetitjean,andChristopheRapin.2019.“TheImpactofIronOxideConcentrationonthePerformanceofMoltenOxideElectrolytesfortheProductionofLiquidIronMetal.”MetallurgicalandMaterialsTransactionsB51(1):365–76.https://doi.org/10.1007/s11663-019-01737-3.Wiser,Ryan,JosephRand,JoachimSeel,PhilippBeiter,ErinBaker,EricLantz,andPatrickGilman.2021.“ExpertElicitationSurveyPredicts37%to49%DeclinesinWindEnergyCostsby2050.”NatureEnergy6(5):555–65.https://doi.org/10.1038/s41560-021-00810-z.Worrell,Ernst,DianPhylipsen,DanEinstein,andNathanMartin.2000.EnergyUseandEnergyIntensityoftheU.S.ChemicalIndustry.Berkeley,CA:LawrenceBerkeleyNationalLab.https://doi.org/10.2172/773773.WorldSteelAssociation.2021.WorldSteelinFigures2021.WorldSteelAssociation:Brussels.https://worldsteel.org/wp-content/uploads/2021-World-Steel-in-Figures.pdfIndustrialDeepDecarbonization:ModelingApproachesandDataChallenges39AppendixA.AvailableDatasetstoTrackIndustrialEnergyDemandandEmissions,andTheirLimitationsTableA.1.GTAP11andGTAP-ECharacteristicDefinitionContentGTAP11describesglobalbilateraltradepatterns,internationaltransportmargins,andprotectionmatricesthatlinkindividualcountriesorregions.Foreachcountryorregion,thedatabasepresentsvaluesofproductionaswellasintermediateandfinalconsumptionofcommoditiesandservicesinmillionsofUS$.GTAP-Eprovidescarbondioxide(CO2)emissionsdata,distinguishedbyfuelanduser.SpatialcoverageBothGTAP11andGTAP-Ecover141countriesand19aggregateregionsoftheworldforeachreferenceyear.TemporalcoverageBothGTAP11andGTAP-Eprovideatimeseriesofsnapshotsoftheglobaleconomyforeachoffivereferenceyears:2004,2007,2011,2014,and2017.IndustrialsectorsBothGTAP11andGTAP-Erepresent65productsandservices.IndustryclassificationBothGTAP11andGTAP-EsectorsarederivablefromtheInternationalStandardIndustryClassification(ISIC).StrengthsForGTAP11highspatialcoverage;forGTAP-EbothhighspatialcoverageanddistinctionofCO2emissionsbyfuelsandbyuser.WeaknessesGTAPusesasectorclassificationderivablefromtheISICclassification.Theresultisaproduct/serviceclassificationandnotanindustrysectorclassification.[Section5.1]Dataaredatedandpresentedasasnapshotforfiveyears.[Section5.4]Source:Aguiaretal.(2023)ResourcesfortheFuture40TableA.2.GTAP10DatabaseSectors#CodeSector1pdrPaddyrice2whtWheat3groCerealgrainsnec4v_fVegetables,fruit,nuts5osdOilseeds6c_bSugarcane,sugarbeet7pfbPlant-basedfibers8ocrCropsnec9ctlBovinecattle,sheepandgoats,horses10oapAnimalproductsnec11rmkRawmilk12wolWool,silk-wormcocoons13frsForestry14fshFishing15coaCoal16oilOil17gasGas18oxtOtherextraction(formerlyomn);mineralsnec19cmtBovinemeatproducts20omtMeatproductsnec21volVegetableoilsandfats22milDairyproducts23pcrProcessedrice24sgrSugar25ofdFoodproductsnec26b_tBeveragesandtobaccoproducts27texTextiles28wapWearingapparel29leaLeatherproducts30lumWoodproductsIndustrialDeepDecarbonization:ModelingApproachesandDataChallenges41#CodeSector30lumWoodproducts31pppPaperproducts,publishing32p_cPetroleum,coalproducts33chmChemicalproducts34bphBasicpharmaceuticalproducts35rppRubberandplasticproducts36nmmMineralproductsnec37i_sFerrousmetals38nfmMetalsnec39fmpMetalproducts40eleComputer,electronicandopticalproducts41eeqElectricalequipment42omeMachineryandequipmentnec43mvhMotorvehiclesandparts44otnTransportequipmentnec45omfManufacturesnec46elyElectricity47gdtGasmanufacture,distribution48wtrWater49cnsConstruction50trdTrade51afsAccommodation,foodandserviceactivities52otpTransportnec53wtpWatertransport54atpAirtransport55whsWarehousingandsupportactivities56cmnCommunication57ofiFinancialservicesnec58insInsurance(formerlyisr)59rsaRealestateactivities60obsBusinessservicesnec61rosRecreationalandotherservices62osgPublicAdministrationanddefense63eduEducation64hhtHumanhealthandsocialworkactivities65dweDwellingsResourcesfortheFuture42TableA.3.EDGARV7.0CharacteristicDescriptionContentProvidesemissionsofthethreemaingreenhousegases(CO2,CH4,N2O)andfluorinatedgasesinkilotons(Kt)persectorandcountry.Spatialcoverage210countries.TemporalcoverageTimeseriesfor1970–2021.Industrialsectors26sectors.IndustryclassificationEDGARsectorclassification,derivablefromtheIPCCGuidelinessectorclassification.StrengthsHighspatialandtemporalcoverage;dataonthreemainGHGs.WeaknessesEDGARusesasectorclassificationderivablefrom2006IPCCGuidelinessectorclassification[Section5.1]Nodataonenergydemand.[Section5.2]Energycarrierisidentifiedonlybybio/fossil.[Section5.3].Nodataonmaterialandenergyflows.[Section5.7]IndustrialDeepDecarbonization:ModelingApproachesandDataChallenges43TableA.4.EDGARDatabaseSectors#CodeSector1AGSAgriculturalsoils2AWBAgriculturalwasteburning3CHEChemicalprocesses4ENEPowerindustry5ENFEntericfermentation6FFFFossilfuelfires7IDEIndirectemissionsfromNOxandNH38INDCombustionformanufacturing9IROIronandsteelproduction10MNMManuremanagement11N2OIndirectN2Oemissionsfromagriculture12PRO_COALFuelexploitationCOAL13PRO_GASFuelexploitationGAS14PRO_OILFuelexploitationOIL15PRU_SOLSolventsandproductsuse16RCOEnergyforbuildings17REF_TRFOilrefineriesandtransformationindustry18SWD_INCSolidwasteincineration19SWD_LDFSolidwastelandfills20TNR_Aviation_CDSAviationclimbing&descent21TNR_Aviation_CRSAviationcruise22TNR_Aviation_LTOAviationlanding&takeoff23TNR_OtherRailways,pipelines,off-roadtransport24TNR_ShipShipping25TRORoadtransportation26WWTWastewaterhandlingResourcesfortheFuture44TableA.5.WorldInput-OutputDatabase(WIOD)2016CharacteristicDescriptionEnvironmentalaccountsContentWorldinput-outputtables(WIOTs)incurrentprices,denotedinmillionsofUS$.AWIOTprovidesacomprehensivesummaryofalltransactionsintheglobaleconomybetweenindustriesandfinalusersacrosscountries.Grossenergyuse(TJ),emissions-relevantenergyuse(TJ),andCO2(kt).Spatialcoverage43countries41countriesTemporalcoverageTimeseriesfor2000–2014Timeseriesfor2000–2016Industrialsectors58sectors64sectorsIndustryclassificationISICRev.4NACERev.2StrengthsHightemporalcoverage;presentsbothemissionsandenergyuse;twelveenergycarriersforgrossenergyuse;emissions-relevantenergyuse.WeaknessesDataaredated;thetimeseriesendedin2014.[Section5.4]ThereisnoinformationaboutfuelinputsrelatedtoCO2emissions.[Section5.5]Thefirstandsecondlevels(sectionsanddivisions)ofISICRev.4(UN2008)arethesameasthoseofNACERev.2(Eurostat2008).NACERev.2dividesthethirdandfourthlevels(groupsandclasses)ofISICRev.4inaccordancewithEuropeanstandards.However,NACERev.2groupsandclassescanalwaysbecombinedwiththeISICRev.4groupsandclassesfromwhichtheywerederived.TheadditionaldivisionsinNACERev.2overISICRev.4areintendedtocreateacategorizationthatisbetteradaptedtotheeconomicsystemsofEurope.IndustrialDeepDecarbonization:ModelingApproachesandDataChallenges45TableA.6.NACERev.2#CodeNACERev.21vA01Cropandanimalproduction;huntingandrelatedserviceactivities2vA02Forestryandlogging3vA03Fishingandaquaculture4vBMiningandquarrying5vC10_12Manufactureoffoodproducts,beverages,andtobaccoproducts6vC13_15Manufactureoftextiles,wearingapparel,andleatherproducts7vC16Manufactureofwoodandofwoodandcorkproducts,exceptfurniture;manufactureofarticlesofstrawandplaitingmaterials8vC17Manufactureofpaperandpaperproducts9vC18Printingandreproductionofrecordedmedia10vC19Manufactureofcokeandrefinedpetroleumproducts11vC20Manufactureofchemicalsandchemicalproducts12vC21Manufactureofbasicpharmaceuticalproductsandpharmaceuticalpreparations13vC22Manufactureofrubberandplasticproducts14vC23Manufactureofothernonmetallicmineralproducts15vC24Manufactureofbasicmetals16vC25Manufactureoffabricatedmetalproducts,exceptmachineryandequipment17vC26Manufactureofcomputer,electronicandopticalproducts18vC27Manufactureofelectricalequipment19vC28Manufactureofmachineryandequipmentnec20vC29Manufactureofmotorvehicles,trailersandsemi-trailers21vC30Manufactureofothertransportequipment22vC31_32Manufactureoffurniture;othermanufacturing23vC33Repairandinstallationofmachineryandequipment24vD35Electricity,gas,steam,andair-conditioningsupply25vE36Watercollection,treatment,andsupply26vE37_39Sewerage;wastecollection,treatment,anddisposalactivities;materialsrecovery;remediationactivities;andotherwastemanagementservices27vFConstruction28vG45Wholesaleandretailtradeandrepairofmotorvehiclesandmotorcycles29vG46Wholesaletrade,exceptofmotorvehiclesandmotorcyclesResourcesfortheFuture46#CodeNACERev.230vG47Retailtrade,exceptofmotorvehiclesandmotorcycles31vH49Landtransportandtransportviapipelines32vH50Watertransport33vH51Airtransport34vH52Warehousingandsupportactivitiesfortransportation35vH53Postalandcourieractivities36vIAccommodationandfoodserviceactivities37vJ58Publishingactivities38vJ59_60Motionpicture,video,andtelevisionprogramproduction;soundrecordingandmusicpublishingactivities;programmingandbroadcastingactivities39vJ61Telecommunications40vJ62_63Computerprogramming,consultancy,andrelatedactivities;informationserviceactivities41vK64Financialserviceactivities,exceptinsuranceandpensionfunding42vK65Insurance,reinsurance,andpensionfunding,exceptcompulsorysocialsecurity43vK66Activitiesauxiliarytofinancialservicesandinsuranceactivities44vL68Realestateactivities45vM69_70Legalandaccountingactivities;activitiesofheadoffices;managementconsultancyactivities46vM71Architecturalandengineeringactivities;technicaltestingandanalysis47vM72Scientificresearchanddevelopment48vM73Advertisingandmarketresearch49vM74_75Otherprofessional,scientificandtechnicalactivities;veterinaryactivities50vNAdministrativeandsupportserviceactivities51vO84Publicadministrationanddefense;compulsorysocialsecurity52vP85Education53vQHumanhealthandsocialworkactivities54vR_SOtherserviceactivities55vTActivitiesofhouseholdsasemployers;undifferentiatedgoods-andservices-producingactivitiesofhouseholdsforownuse56vUActivitiesofextraterritorialorganizationsandbodies57vTOTII_newTotalindustrialactivities58vCONS_h_newFinalconsumptionexpenditurebyhouseholdsIndustrialDeepDecarbonization:ModelingApproachesandDataChallenges47TableA.7.Eora26CharacteristicDescriptionContentGlobalmultiregioninput-outputtable(MRIO)documentingintersectoraltransfers,environmentalindicatorscoveringGHGsemissions,laborinputs,airpollution,energyuse,waterrequirements,landoccupation,NandPemissions,andprimaryinputstoagriculture.Spatialcoverage190countries.TemporalcoverageTimeseriesfor1990–2021.Industrialsectors26sectors.IndustryclassificationTheEora26sectorclassificationisbasedoncommonsectorclassifications,butnoconcordancematrixtoorfromotherclassificationsisavailable.StrengthsHighspatialandtemporalresolutions;highnumberofenvironmentalindicators.WeaknessesEora26classificationisnotderivablefromotherstandardclassifications.[Section5.1]Nodataonenergycarriersforenergyuse.[Section5.2]NodataonenergycarriersforGHGemissions.[Section5.5]ResourcesfortheFuture48TableA.8.Eora26Classifications#Classification1Agriculture2Fishing3Miningandquarrying4Foodandbeverages5Textilesandwearingapparel6Woodandpaper7Petroleum,chemical,andnonmetallicmineralproducts8Metalproducts9Electricalandmachinery10Transportequipment11Othermanufacturing12Recycling13Electricity,gas,andwater14Construction15Maintenanceandrepair16Wholesaletrade17Retailtrade18Hotelsandrestraurants19Transport20Postandtelecommunications21Financialintermediationandbusinessactivities22Publicadministration23Education,health,andotherservices24Privatehouseholds25Others26Re-exportandre-importIndustrialDeepDecarbonization:ModelingApproachesandDataChallenges49AppendixB.ClassificationofIndustrialSectorsTounderstandthedifferentapproachestoindustrialmodeling,itisnecessarytounderstandindustryclassificationsandtheavailabilityofsector-leveldataonenergydemandandemissions.TableB.1summarizesthecharacteristicsofthedifferentclassificationmethods.TableB.1.AvailableClassificationsofIndustrialSectorsStructureCodeISICRev.4NACERev.2SectionLetter-based21sections21sectionsDivision2-digit88divisions88divisionsGroup3-digit238groups272groupsClass4-digit419classes615classesStructureCodeNAICS(OMB2022)Sector2-digit20sectorsSubsector3-digit99subsectorsIndustrygroup4-digit311industrygroupsNAICSindustry5-digit709NAICSindustriesNationalindustry6-digit1,057nationalindustriesRepresentsthelowestlevelofcompatibilityamongtheUnitedStates,Canada,andMexico.ResourcesfortheFuture50B.1.TheInternationalStandardIndustrialClassificationofAllEconomicActivities(ISIC)ISICistheUnitedNations’systemforclassifyingeconomicactivitiesbasedonacollectionofideas,definitions,guidingprinciples,andclassificationcriteriathathavebeenuniversallyaccepted(UN2008).Itprovidesathoroughframeworkforcollectingandreportingeconomicdataintendedforuseineconomicanalysis,decisionmaking,andpolicymaking.ISICclassifiesproductiveactivitiesintofourlevelsofdistincthierarchicalcategoriestomakedatacollection,display,andanalysisatspecificeconomiclevelseasierandmoreuniformworldwide.Sectionsarethehighestlevelofclassification,withallformsofproductiveactivitydividedintomajorcategoriesdesignatedbyletters.Thesearesubdividedintoevermorespecificcategoriesdesignatedbynumbers:two-digitdivisions,three-digitgroups,andfour-digitclasses.Thecriteriausedtodefinethecategoriesarebasedon“theinputsofgoods,services,andfactorsofproduction,theprocessandtechnologyofproduction,thecharacteristicsofoutputs,andtheusetowhichtheoutputsareput”(UN2008).TheISICcategorieshavebeenusedtoorganizeeconomicactivitiesthatmeettheserequirements.Theeconomicactivities’processandtechnologyhavebeenprioritizedfordefiningspecificISICclassesatthemostgranularlevelofcategorization,especiallyinthecategoriesrelatedtoservices.Datamaybeusedtoinvestigatespecificindustriesorindustrialgroupsandevaluatetheeconomybydisaggregatingthedataintovariousdegreesofdetail.Numerousfactors,includingtheintendeduseoftheclassification,theaccessibilityofdata,andtheamountofaggregationconsideredinfluencethecriteriausedtodefinethecategoriesateverylevel.Similaritiesbetweenmanufacturingprocessesarealwaysconsideredwhenclassifyingactivitiesatthesectorlevelathighlevelofdetail;undoubtedly,themoreaggregatethesector,themoreheterogenousaretheprocessesbundledintheclassification.B.2.TheNorthAmericanIndustryClassificationSystem(NAICS)NAICSisasystemforclassifyingorganizationsbasedontheireconomicactivities.Itsgoalsaretomakeiteasiertogather,tabulate,display,andanalyzedataonbusinesses,aswellastoimproveuniformityandconsistencyintheanalysisandpresentationofstatisticaldataabouttheUSA,Canadian,andMexicaneconomy.ByworkingtogetheronNAICS,theInstitutoNacionaldeEstadísticayGeografíaofMexico,StatisticsCanada,andtheUSOfficeofManagementandBudgethavecreatedauniformframeworkthatmakestheindustrystatisticsprovidedbythethreenationscomparable.NAICSisusedbyfederalstatisticsorganizationsandpolicyanalyststogatheranddisseminatedatabyindustry,andstateagencies,academiaandresearchers,thebusinesscommunity,andthegeneralpublicalsomakeextensiveuseofthesystem.IndustrialDeepDecarbonization:ModelingApproachesandDataChallenges51NAICSisahierarchicalsysteminwhichenterprisesarecategorizedfromthebroadesttothemostgranularlevelsandisthefirstindustrialcategorizationsystemcreatedusingasingleaggregationbasis.Itreflectsrecenttechnologicaladvancesaswellastheincreaseanddiversityofservices.NAICS(2017)classificationishighlycomparabletothemostcurrentversionofISIC.FourprinciplesinfluenceNAICSdevelopment:(1)NAICShasaproduct-orientedframework,groupingtogethermanufacturingunitsthatemploythesameorcomparableproductionprocesses.(2)Thesystemfocusesonbuildingproduction-orientedcategoriesfornewandemergingsectors,theservicesectoringeneral,andindustriesinvolvedinhightechnologyproduction.(3)Time-seriesconsistencyispreservedtothehighestdegreepossible.(4)Thesystemseekstwo-digitcompliancewiththeISICclassification.B.3.TheStatisticalClassificationofEconomicActivitiesintheEuropeanCommunity(NACE)NACEistheEuropeanUnion’sacceptedsystemofcategorizingproductiveeconomicactivity,inwhichacodeisassignedtoastatisticalunitforeachactivity.TheacronymNACE(NomenclaturestatistiquedesActivitéséconomiquesdanslaCommunautéEuropéenne)comesfromtheFrenchnameforthesystemandreferstothemultiplestatisticalcategoriesofeconomicactivitiesintheEuropeanUnionsince1970.NACEprovidesaframeworkforgatheringanddisplayingawiderangeofstatisticaldatabasedoneconomicactivities.Aneconomicactivityisdefinedbytheinputofresources,themanufacturingprocess,andtheoutputofgoods.Whenresourcessuchaslabor,capitalgoods,manufacturingprocesses,orintermediaryitemsarecombinedtocreatecertaincommoditiesorservices,thatisconsideredaneconomicactivity.Anactivitymightbeasingle,straightforwardprocess,oritmightconsistofavarietyofsmalleractivitiesthatareeachclassifiedinadistinctcategory.NACEoriginatedfromISICbutismoredetailed.ISICandNACEincludeidenticalelementsatthehighestlevels,withNACEbeingmoregranularatlowerlevels.NACEconsistsofahierarchicalstructureasdefinedbytheNACEregulations,introductoryinstructions,andexplanatorynotes(Eurostat2008).StatisticsbasedonNACEarecomparableattheEuropeanandgloballevels,andtheusageofthisclassificationsystemwasmademandatoryintheEUbymemberstates,theEuropeanCommission,andtheEuropeanStatisticalSystem.Toguaranteeworldwidecomparability,thecriteriaandrulesdefinedfortheuseofNACEinsidetheEUarethesameasthoseforISIC.StatisticscollectedbytheEUmemberstatesregardingeconomicactivitycategorizationmustbeproducedbyNACEoranationalcategorizationderivedfromit(Eurostat2008).TheNACEregulationspermitmemberstatestouseanationalversionofthesystemfordomesticreasons.SuchnationalversionsmustcomplywithNACE’shierarchicalstructure.Manymemberstateshavecreatedtheirownversions,generallybyaddingafifthdigittorepresentnationalneeds.ResourcesfortheFuture52AppendixC.ModelSummariesTheInternationalEnergyAgency’sWEMisalarge-scalesimulationmodelthatisupdatedannually;ithasbeendevelopedovermanyyearsandfocusesonenergyusein26regionsuntil2050(IEA2021).Foreachregion,itincludesthreemainmodules:finalenergyconsumption(coveringresidential,services,agriculture,industry,transportation,andnonenergyuse),energytransformation,andothertransformation.Themodel’soutputsincludeenergyflowsbyfuel,investmentrequirementsandcosts,CO2emissions,andend-userpricing.WEM’sindustrialsectorisdividedintosixsubsectors:nonferrousmetals(includingaluminum),ironandsteel,chemicalsandpetrochemicals,nonmetallicminerals(includingcement),pulpandpaper,andotherindustries(transportequipment,machinery,miningandquarrying,foodandtobacco,woodandwoodproducts,construction,textileandleather,andnonspecified).Energyconsumptionintheindustrialsectorisdrivenbythemanufactureofgoodsintheenergy-demandingindustriesandbyvalueaddedinthenonspecifiedindustrysectors.Ineachsubsector,energyconsumptioniscomputedastheproductofproductionforecastsandmanufacturingprocessenergyintensity.Theenergyintensityofnewcapacityisdependentontheuseofenergy-efficienttechnologyandthelevelofenergycosts,whiletheenergyuseperunitofproductionforexistinginfrastructureisratherstable.Eachproductionmethodforaluminum,iron,steel,fiveprimaryproductcategoriesinchemicalsandpetrochemicals,cement,pulpandpaper,andcross-cuttingtechnologiesinnon-energy-demandingindustriesofferschancesfortechnologicalefficiency.Energy-efficienttechnologiesareadoptedbasedontheirprospectivepenetrationrateandpaybacktimeframe,bothofwhichvarydependingonthecircumstance.Inadditiontosingle-equipmentefficiency,theindustrysectorTableC.1.WorldEnergyModel(WEM)TypeSimulationIndustrysectorsNonferrousmetals(aluminum)IronandsteelChemicalandpetrochemicalNonmetallicminerals(cement)PulpandpaperOtherindustryApproachHybridSpatialresolutionGlobal(26regions)TemporalresolutionYearlyto2050IndustrialDeepDecarbonization:ModelingApproachesandDataChallenges53modelalsoincludeschoicesforsystemoptimizationandprocessimprovements.TheWEMindustrialmodelmayreflectmaterialefficiencytechniquesinadditiontoenergyefficiencytechnologyandmeasures,offeringfurtherenergysavings.Energy-intensiveindustriesoftenhavemorerestrictedopportunitiestoimproveenergyefficiencythanlessenergy-intensiveindustriesbecauseenergycostscontributesignificantlytoproductioncosts.Thisstrategy’suseisrestrictedtothematerialandenergyneedsofthecorrespondingindustrialsectors.WEMdoesnotexaminetheeffectsonenergyuseupstream,duringminingormaterialtransportation,ortheimpactonconsumptiondownstream,nordoesitincludethepossibleenergysavingsfromsubstitutematerials.WEMisusedonlytoproducetheIEA’sWorldEnergyOutlook(WEO)and,tothebestofourknowledge,isnotavailableforuseoutsideoftheagency.FortheWEO,WEMusesascenarioapproachtolookatpotentialchangesintheenergysectorbasedonthemodel.Fourscenarios—theAnnouncedPledges,NetZeroEmissionsby2050,StatedPolicies,andSustainableDevelopmentScenarios—weresimulatedindepthfortheWorldEnergyOutlook2021.Thescenariosincludethemostrecentenergydata,policystatements,investmentpatterns,andtechnologicaladvancesandarebasedonmodelingandanalysis.Whenexaminingfiguredevelopments,theWEOscenariostakeintoconsiderationthecompleterangeofexistingnationalconditions,resources,technology,andprospectivepolicyoptions.WEMenablestheassessmentoftheimpactofcertainpoliciesandinitiativesonenergyconsumption,production,trade,investmentrequirements,supplyprices,andemissions.PoliciesandmeasuresarederivedbytheWEOpolicydatabaseandincludeinitiativesaddressingrenewableenergy,energyefficiency,andclimatechange.TableC.2.NationalEnergyModelingSystem(NEMS)TypeOptimization/simulationPartialequilibriumIndustrysectorsFoodPaperChemicalsGlassCementandlimeIronandsteelAluminumApproachHybridSpatialresolutionRegional/national(UnitedStates)TemporalresolutionYearlyto2050ResourcesfortheFuture54NEMSisacomputer-basedenergy-economymodelingsystemfortheUnitedStatesdevelopedbytheUSEnergyInformationAdministration(EIA2019).NEMSforecastsenergyproduction,imports,conversion,consumption,andpricingbasedonassumptionsaboutmacroeconomicandfinancialaspects,globalenergymarkets,resourceavailabilityandcosts,behavioralandtechnicalchoicecriteria,costandperformancecharacteristicsofenergytechnologies,anddemography.Thus,itprovidesastandardizedframeworktodescribetheinteractionsoftheUSenergyinfrastructureanditsreactiontoawiderangeofdifferentassumptions,regulations,andpolicyinitiatives,aswellastomeasuretheeffectofnewenergyinitiativesandregulations.Theforecasttimeframeisaround30years.NEMScanbeusedtoassesstheenergy,economic,environmental,andsecurityeffectsofexistingandproposedgovernmentlawsandregulationspertainingtoenergyproductionanduseontheUSenergysystemandthepotentialimpactofadvancedandinnovativeenergyproduction,conversion,andconsumptiontechnologies.Inaddition,NEMScanassesstheeffectandcostofgreenhousegascontrol,theeffectofincreaseduseofrenewableenergysources,thecostsandbenefitsfromincreasedenergyefficiency,andtheimpactofregulationsrelatingtotheuseofinnovativeorreformulatedenergysources.Sinceenergysuppliesandcosts,demandforspecializedenergyservices,andotherenergymarketfeaturesvarygreatlyacrosstheUnitedStates,regionalversionsofthemodelcanbeusedtoaccountfordifferentgeologiesandothercharacteristicsandfocusonkeyareasmostrelevantforpolicyanalysis.Unitenergyconsumption(UEC)isestimatedforeachNAICSsectorintheindustrialdemandmodule.Thequantityofenergynecessarytogenerateonedollar’svalueofshipmentsoroneunitofphysicaloutputisspecifiedasUEC.Technologicalchangeinthemanufacturingprocessallowsforlowerenergyintensity;thisisassumedtocomeaboutthroughalearning-by-doingprocess—thatis,asexperienceisgainedinthetechnology,productioncostsdecrease.Industrialinnovationscanbeselectedanddeployedbecauseofavarietyofreasonsotherthantheirenergyconsumptioncharacteristics,suchasprocessimprovementstoenhanceproductquality,changestoincreaseproductivity,orchangesinreactiontothecompetitiveenvironment.Futurereductionsinunitenergyconsumptionarecalculatedusingtechnologypossibilitycurves,andfutureenergysavingsareestimatedforbothnewandexistingprocessesandfacilities.Inolderfacilities,energygainsarisefromprogressiveimprovementsdrivenbyenergy-savingmeasures,retrofitsofcertaintechnologies,andtheclosureofinefficientolderfacilities.EstimatedUECvaluesforstate-of-the-art(SOA)andadvancedtechnologiesarealsoprovided.Themostrecentandproventechnologiesavailableatthetimeofinvestmentdecision—thatis,theSOA—arechosen.EstimatedresultsarethencomparedwithUECsofthebaseyeartocomputeanindexofrelativeenergyintensity,calculatedastheenergyuseofanewprocessrelativetobase-yearenergyuse.TheefficiencybenefitfornewfacilitiesdependsontheinstallationoftheSOAtechnologiesappropriateforthatsector.Whennewtechnologiesareaccessibleforaspecificprocess,asecond,occasionallymoresignificant,roundofconsiderableimprovementsmayoccur.Industrialenergyconsumptionisaffectedbyincreasedenergyefficiencyinnewandoldfacilities,theindustry’sgrowthrate,andtheretirementrateforoldplants(EIA2022).IndustrialDeepDecarbonization:ModelingApproachesandDataChallenges55NEMShasbeenwidelyused.Forinstance,BrownandBaek(2010)studyenergyefficiencyimpactsofaportfolioofenergyandclimatepoliciestomitigateelectricityandbiomasspriceincreaseswhileincreasingenergysecurityandloweringCO2emissions.Theyevaluatethreedifferentnationalpolicyscenariosfocusingontheforestproductssector:arenewableelectricitystandard,acarbonpolicy,andincentivesforindustrialenergyefficiency.Theyalsoexplorethepotentialimpactsofsuchmarketshiftsonbiomassandelectricitycosts,energyuse,andCO2emissions.DevelopedbythePacificNorthwestNationalLaboratory,GCAMisamultisectorintegratedmodelthatanalyzesbothhumanandEarthsystemdynamics(GCIMS).Themodeladoptsasetofassumptionsandthenanalyzesthemtoprovideacompletepictureofpricing,energy,commodity,andotherflowsacrossregionsandinthelongterm.Itrepresentsfivedistinctinteractingandinterrelatedsystems:macroeconomics,energy,agricultureandland,water,andphysicalearth.Theenergy-economysystemrunsin32regionsglobally,thelandsystemissplitintomorethan300subregions,andtheEarthsystemmoduleisglobal.MarketequilibriumisthemainguidingconceptofGCAM,inwhichrepresentativeagentsdecidehowtodistributeresourcesbasedonpricingandotherpotentiallyrelevantinformation.Theserepresentativeagentscommunicatewithoneanotherthroughmarkets.Toensurethatsuppliesanddemandsarebalancedthroughoutallthesemarkets,GCAMsolvesforarangeofmarketprices.TheGCAMsolutionprocessentailsoptimizingmarketpricinguntilthisequilibriumisachieved.GiventhatGCAMisadynamicrecursivemodel,choicesmadetodaydonotconsiderfutureevents.TheenergysystemmoduleinGCAMmodelsninedetailedindustrialsectors:sixmanufacturingsectors(ironandsteel,chemicalsandpetrochemicals,aluminum,cement,fertilizers,andotherindustry)andthreenonmanufacturingsectors(construction,miningenergyuse,andagriculturalenergyuse).InternationaltradeTableC.3.GlobalChangeAssessmentModel(GCAM)TypeOptimization/simulationPartialequilibriumIndustrysectorsIronandsteelChemicalsAluminumCementFertilizersOtherindustryApproachBottom-upSpatialresolutionGlobal(32regions,300subregions)Temporalresolution5yearsto2100ResourcesfortheFuture56isnotmodeled.Physicaloutputs(Mt)andgeneralterminologyareusedtoexpresstheoutputofthedetailedindustrysectors.Theremainingindustrialsectorsarerepresentedas“otherindustries”andareconsumersofgenericenergyservicesandfeedstocks.Theironandsteelsectorisdividedintothreesubsectors:basicoxygenfurnace(BOF),electricarcfurnacewithscrap(EAF),andEAFwithdirectreducediron(DRI).Eachsubsectorhasseveralcompetingtechnologies,includingfossilfuelswithandwithoutCCS,electricity,hydrogen,andbiomass.Thechemicalsandpetrochemicalssectorisdividedintotwoparts:chemicalenergyusageandfeedstocks.Thealuminummanufacturingprocessconsistsoftwomajorsteps:aluminarefining,whichinvolvesrefiningbauxiteoreintoalumina,andaluminumsmelting,convertingaluminatoaluminum.Thereareseveralcompetingmethodsforaluminarefining,includingcoal,refinedliquids,gas,andbiomasswithandwithoutCCS.GCAMcontainsamodelforcementmanufacturingthatmeasuresbothfuelandlimestone-derivedCO2emissions.Mostofthekeyfossilfuelandlow-carbontechnologiesthatareexpectedtobeavailableatleastuntil2050arerepresentedinthemodel.InGCAM,mitigationismodeledasamovefromhigh-tolow-carbontechnologiesbasedonrelativecosts,emissionsconstraints,andcarbonprices.Inarecentapplication,Pengetal.(2021)investigatetwelvemitigationscenariosthatdifferalongtwodimensions:nationalmitigationeffortandsubnationalpolicyapproach.ThefirstismeasuredbyfournationalUStotalgreenhousegas(GHG)emissionsobjectivesfor2050.Thesecondisrepresentedbythreelevelsofvariabilityintheconstraintsofstate-levelclimatepolicy,representedasauniformcarbonpricecalculatedbyequalizingthemarginalabatementcost(MAC)acrossstatesinGCAM-USA.TheMACindirectlyevaluatesindustries’andhouseholds’willingnesstopaytoreducecarbonemissionsinagivenstate.AgreaterMACmeansahighercarbonpriceand,asaresult,stricterclimatepolicyactions.Pengandcolleaguesvalidatethewidespreadconclusionintheliteraturethatdeepdecarbonizationoftennecessitatesdecarbonizingtheelectricalsectorfirstbeforemovingontoharder-to-abatesectorssuchasindustry,residential,andtransport.IndustrialDeepDecarbonization:ModelingApproachesandDataChallenges57REMINDisamultiregionalmodeldevelopedbyPotsdamInstituteforClimateImpactResearch.ItisbasedontheGeneralAlgebraicModelingSystem(GAMS),whichincorporatestheeconomyandadetaileddescriptionoftheenergysector.Itemploysnonlinearoptimizationtocreatewelfare-optimalregionaltransformationpathsoftheenergy-economicsystemunderclimateandsustainabilityconstraintsfor2005–2100.Underthepresumptionsofperfectagentforesightandinternalizationofexternaleffects,thesolutionconformstothedecentralizedmarketoutcome.Withaspecificfocusonthescalingupofinnovativetechnologies,suchasrenewables,andtheirintegrationintoenergymarkets,REMINDcanbeusedtoanalyzetechnologicalpossibilitiesanddifferentpolicyapproachesforclimatechangemitigation(Baumstarketal.2021).Theindustrymodulesimulatesthetotalfinalenergydemandandemissionsoftheindustrysectoranditssubsectors:cement,chemicals,ironandsteel,andallremainingindustryenergydemand(“otherindustry”).Thesesharesarefixedat2005levelsforeachregion.Theconstantelasticityofsubstitutionproductionfunctionisusedtodeterminewhetherfuelswitchingisattractivedependingonfinalenergypricesandthefinalenergycarriers’substitutionelasticities.ThreeMACcurvesforCCSinthecement,chemicals,andironandsteelsectorshavebeengeneratedfromtheliteratureandusedinallindustrialmodulerealizations(Kuramochietal.2012).TocomputeindustryCO2emissionsandcapturelevels,sector-specificMACcurvesforCCSareappliedtoemissionscalculatedfromenergyuseandemissionsfactorsbasedontheendogenousCO2price.ProcessemissionsfromtheproductionofcementarecountedinthecementemissionsforwhichCCSisrelevantandarebasedonaneconometricestimateofcementproduction.TheintegralbelowtheMACcostcurveequalstheindustryCCScostsbysubsector(Baumstarketal.2021).TableC.4.RegionalModelofInvestmentandDevelopment(REMIND)TypeNonlinearprogrammingOptimizationIndustrysectorsCementChemicalsIronandsteelOtherindustryApproachHybridSpatialresolutionGlobal(11regions)Temporalresolution5yearsuntil2060,10until2110,20until2150ResourcesfortheFuture58MUSE,developedbyImperialCollegeLondon,modelsapartialequilibriumofthewholeenergysystem,whichincludestheextractionofresourceslikeoil,biomass,orrenewables(insupplysectors),theresourcetransformationintoenergyvectors(inpowersystemsandrefineries),andtheconsumptionofenergyvectorsforfulfillingsocietyneeds(indemandsectors)(Giarolaetal.2022).Withitsagent-basedstructure,themodelisusedtodescribedecisiongoalsandstrategiesofkeyplayersineachsector,andthusitshowsthatbusinessandconsumerdecisionmakingcanproducemacro-levelinefficienciesintheenergysystem.Fromthetechnologydescription,themodelisbottom-upandtechnology-rich;itmodelseachtechnologyperformance,costs,andemissionsandrecordstechnologystock,investments,operatingcosts,andenergyuse.Theleveloftechnologicaldetailnecessarytocharacterizethetechnologiesandagentsrequiresaccesstolargedatasetsoftechnoeconomicandsocioeconomicinformation,whichmaynotalwaysbepubliclyavailableforallcountries.MUSEdemandsectorsincludeindustry,agriculture,buildings,andtransport.Inadditiontoironandsteel,industryalsocoversnonmetallicminerals,nonferrousmetals,pulpandpaper,andchemicalandpetrochemicalproducts.Theseothersubsectorsaregroupedtogetheras“otherindustry.”Inthehard-to-abateindustrysectors,existingtechnologiesrepresentthebaseyearstock,whichmodelsexistingfacilitiesinstallation,operatingproductionlevels,energyconsumption,andemissionsaggregatedataregionalscale.Theexistingcapacityislinearlydecommissioned;this,alongsidethecapacityneededtomeetfuturedemand,isthenreplacedbynewassets,takingintoconsiderationevolvinglegislation,fuelandcarbonpricesandtheavailabilityofadvancedtechnologies.Themodelfirstforecastsfuturedemandforindustrialcommoditiesonaworldwidescalewitharegionaldisaggregationusinghistoricalpatterns.Thentheproductionofmaterialgoodsismodeledusingamerit-ordermethodbasedontheinvestor’sdecisiongoal,whichcouldbenetpresentvaluemaximizationorcostminimization.Asaresult,untilthedemandformaterialgoodsissatisfied,theprocesseswiththeTableC.5.ModularEnergySystemSimulationEnvironment(MUSE)TypeSImulationPartialequilibriumIndustrysectorsAluminumIronandsteelChemicalCementPulpandpaperOtherindustryApproachBottom-upSpatialresolutionGlobal(28regions)Temporalresolution5–10yearsto2050–2100IndustrialDeepDecarbonization:ModelingApproachesandDataChallenges59highestprofitsareusedtoprimarilycoverthedemand.Thequantityneededofeachdifferentfueliscalculatedbasedonthecombinationoftechnologiesbeingemployed.Thedecarbonizationofindustrysectorsmayoccurthrough(1)energyefficiency;(2)electrificationintheironandsteelindustry;(3)fuelswitching,withbiomethaneandbiomassoptionsavailableforallindustries,andhydrogenavailableforcementandtheironandsteelindustry;or(4)advancedtechnologies,whichincludeelectricfurnaceandthesmeltingprocessintheironandsteelsector,calciumloopinginthecementsector,andtheintegrationofCCSinallthestandardsectortechnologieswithahigherinstallationcostandgreaterenergyconsumption.OnerelevantapplicationoftheindustrymodelispresentedinBudinisetal.(2020),whichmodelsthedecarbonizationoftheChineseammoniaindustry.ThismayoccurviatheintegrationofCCSinadditiontofuelswitching,investigatingthepotentialuptakeofnegativeemissionsobtainedfromthecombinationofbioenergywithCCS.Themethodologyishasanagent-basedformulation,whichaimstosimulaterealinvestmentstrategiesintheenergysectormadebyagent-investorsinChina.Byclassifyingthemarketshareintosmallversuslargeenterprisesandinternationalversusstate-ownedcompanies,theworkshowsthatdespitetheavailabilityofCCS,therearebarrierspreventingtechnologyinnovation,whichcanberelatedtothedecisionmakingprocessandthecapitalaccessofindustries.TheTIMESmodel,developedbyImperialCollegeLondonandtheGranthamInstitute,isamodelgeneratorthatcombinestechnicalengineeringdetailandeconomicsinenergymodeling(LoulouandLabriet2008).AlthoughtheTIMESmodelgeneratorhasdifferentrealizationsatanationallevel(suchastheUKTM,theTIMESmodelfortheUK)andataglobalscale(suchasTIAM-GranthamorTIAM-UCL),themodelingapproachisaleast-costintertemporaloptimization.TIMESassumesperfectforesight,whichmeansthatallinvestmentchoicesineachmilestoneyearareoptimizedwiththeassumptionofcompleteknowledgeoffutureoccurrences.TableC.6.TheIntegratedMARKAL-EFOMSystem(TIMES)TypeOptimizationIndustrysectorsAluminumIronandsteelChemicalCementPulpandpaperOtherindustryApproachBottom-upSpatialresolutionGlobal(15regions)Temporalresolution10yearsto2100IndustrialDeepDecarbonization:ModelingApproachesandDataChallenges60WedescribeheretheenergysystemgranularityofTIAM-Granthamasreference.Themodelincorporatesallthephasesfromrawresourcesintothedeliveryofenergyservicesrequestedbyenergyconsumersviathechainofactivitiesthattransform,transport,distribute,andconvertenergy.Onthesupplyside,itincludesfuelmining,primaryandsecondaryproduction,andexternalimportsandexports.Energyisgiventothedemandsideviavariousenergycarriers,whicharesegmentedintoresidential,commercial,agricultural,transportation,andindustrialsectors.Energyefficiency,electrification,andfuelswitchingaretheavailabledecarbonizationoptionsinTIAM.Amongtheadvancedtechnologies,themodelincludesthefollowing:•intheironandsteelsector,blastfurnacewithdirectcoalinjection,blastfurnacewithtop-gasrecycling,blastfurnacewithtop-gasrecyclingandCCS,blastfurnacewithCCS,Corexsmeltingprocess,CorexwithCCS,DRIwithCCS,DRIwithhydrogen,on-sitepowergenerationwithrecycledgases,on-sitepowergenerationwithCCS•inthenonmetallicmineralsector,cementprecalcinerwithCCS,cementwholeplantwithCCS•inthechemicalandpetrochemicalsector,chemicalproductionwithCCS,ethyleneprocessinchemicalsectorwithCCS,hydrogenforammoniaproductionwithCCS,ethyleneandpropyleneprocesswithCCS•inthepulpandpapersector,steamgenerationinpulpandpaper(coal-orgas-fired)withCCS,processheatinpulpandpaper(coal-orgas-fired)withCCS•inalltheenergy-intensiveindustrysectors,combinedheatandpower(usingcoal,gas,orrecycledgases)withCCS•inotherindustries,processheatwithCCSOneoftherecentmodelapplicationsispresentedbyNappetal.(2019),whofocusonthechallengesposedbydecarbonizationinindustry.Theanalysisconcludesthatsubstantialinvestmentsinadvancedtechnologiesarerequiredtoachieveascenariothatisconsistentwithlimitingglobalwarmingto1.5°C.Keyadvancedtechnologiesintheindustrialsectorincludehydrogen-basedsteel,electrification,andCCSfromcementproduction.IndustrialDeepDecarbonization:ModelingApproachesandDataChallenges61TableC.7.IMAGETypeSimulationIndustrysectorsAluminumIronandsteelChemicalCementPulpandpaperOtherindustryApproachBottom-upSpatialresolutionGlobal(26regions)TemporalresolutionYearlyto2100IMAGEisanintegratedframeworkdevelopedbyPBLNetherlandsEnvironmentalAssessmentAgencyformodelinginteractionsbetweenhumanandnaturalsystems(PBL2021).Themodelincludestwoprimarysystems:theHumansystemandtheEarthsystem.ThesocioeconomicorHumansystemillustrateshowhumanactivitiesthatareimportantforsustainabledevelopmenthaveevolvedovertime.EnvironmentalchangesaredescribedbytheEarthsystem.TheeffectsofhumanactivityontheEarthsystemandtheeffectsofenvironmentalchangeintheEarthsystemontheHumansystemmakethetwosystemsinterdependent.Thespatialresolutionforsocioeconomicprocessesiscomposedof26regionschosenfortheirimportancetoglobalenvironmentalanddevelopmentchallenges,aswellasthehighlevelofconsistencywithintheseregions.Themodelframeworkiswellsuitedtolarge-scale(global)andlong-term(until2100)assessmentsofhuman-environmentinteractions.Theimpactsofhumanactivitiesonnaturalsystemsandnaturalresourcesareassessed,aswellashowsuchimpactsinhibittheavailabilityofecosystemservicestosustainhumandevelopment.FocusingontheinteractionbetweentheHumanandEarthsystems,IMAGEdefinesemissionsasafunctionofactivitylevelsintheenergysystem,industry,agriculture,andlandcoverandlandusechanges,aswellasprojectedabatementmeasures.Themodelrepresentskeygreenhousegasemissionsandavarietyofairpollutants.Itiscalibratedtocurrentglobalemissionsinventories,withitssettingsadjustedtoreproducethestateoftheworldfrom1970toafinalbaseyear.Changesinemissionsfactorsovertimearecalculatedbasedonthestoryline,andthemodelmayassumethatemissionsfactorsremainconstantordeclineovertimeinparallelwitheconomicprogress.Theenergymodelfortheindustrysectorcontainsthreecategories:cement,steel,andotherindustrialactivities.IMAGEincludesextensivedemandmodelsforcementandsteel.Exogenouslyspecifiedemissionsfactorsaremultipliedbyactivitylevelstocomputeemissions.Otherindustrialactivities,suchascopperproductionandsolventmanufacture,haveactivitylevelsthatareformulatedasageographicfunctionofindustryvalueadded.EmissionsarecomputedbymultiplyingactivitylevelsbyResourcesfortheFuture62emissionsfactors.Forthesteelandcementindustries,theheavyindustrysubmoduleisincluded.Theactivityinthegenericstructureofenergydemandisdescribedintermsofmetrictonsofcementandsteel,bothofwhichcanbetraded.Tradedemandcanbemetfromproductionthatcombinesavarietyoftechnologies.Costsandenergyuseperunitofproductionaretwocharacteristicsofeachtechnology,andbothgraduallydecreaseovertime.Themultinomiallogitequationusedtodeterminetheactualmixoftechnologiesinsteelandcementproductionleadstoalargermarketshareforthelowest-costtechnologies.Energyefficiencyincreasesbecauseofthesetechnologies’autonomousdevelopment.Thetechnologyselectionrepresentstheprice-inducedimprovementinenergyefficiency.Priceplaysaroleinfuelsubstitution,buttechnologytypealsoplaysarolebecausesometechnologiescanonlyusecertaintypesofenergycarriers.Sharminaetal.(2020)comparesector-specificanalysesoffourkeysectorsthatarechallengingtodecarbonizewitheconomy-widemodelingof1.5°Cand2°Cscenarios:aviation,shipping,roadfreighttransport,andindustry.Toanalyzeandmonitortheprogressofmitigationinthesesectors,theauthorscreateandimplementanovelframework.Theyfindthatinthe1.5°Cand2°CscenariosoftheIMAGEmodel,emissionsreductionsresultfromsignificantreductionsinCO2intensitiesandlowerenergyintensities,withrelativelyslightdemandreductionintheactivityofthesesectors.Severaladditionalactionsandpolicyleversthatcouldsignificantlyreduceemissionsareidentifiedbutnotexplicitlyincludedinthemodeledscenarios.Theseoptionsfordemandreductionincludemovingtheindustrytowardacirculareconomy.Cementandconcrete,plastics,steel,andammoniaareallcoveredundertheMaterialEconomicsModellingFramework(MaterialEconomics2019).Themodelingapproachstartsbyestimatingfutureactivitylevels.Abaselinescenariofor2050demandispredictedusingavarietyofmodels.Theprimarytoolforsteelisadynamicmaterialflowanalysis,togetherwithestimatesaboutfuturesaturationlevelsforsteelstockinvariousend-usesectors.Activitylevelsforplastics,cement,andammoniaarebasedonexpectedbuilding,mobility,foodproduction,andotheractivityscenarios.Thebaselinescenarioassumesnosignificantchangesinmaterialintensityorindustrialstructure.NochangeinnetimportsisenvisagedbecausethegoalistodefineanEUnet-zeroCO2industrialsystem.TableC.8.MaterialEconomicsModellingFrameworkTypeSimulationIndustrysectorsSteelCementandconcretePlasticAmmoniaApproachBottom-upSpatialresolutionEuropeanUnionTemporalresolution5yearsto2050IndustrialDeepDecarbonization:ModelingApproachesandDataChallenges63Thesecondstageistoestablishavarietyoflow-CO2manufacturingapproaches.Theanalysisdescribeseachprocess’stechnicalmaturity,investmentneeds,energyandfeedstockinputs,otheroperationalexpenses,massbalance,andCO2emissions.Energyinputcostsarecalculatedforcommonlyusedenergy-economicscenariosdevelopedbytheInternationalEnergyAgencyandotherorganizations.CO2emissionsincludenotjustemissionsfrompowergeneration,butalsocarboncontainedinitemsthatmaybedischargedasCO2attheendoftheirusefullife.Alongwiththeproductionside,theassessmentemploysavarietyofmodelstoinvestigateprospectsforthecirculareconomy:enhancedmaterialefficiencyandincreasedmaterialcircularity.Apackagingmodelcharacterizes35typesandcalculatesprospectsforreducedmaterialconsumptionandreplacementwithothermaterials.Thethirdcomponentisadescriptionofend-of-lifematerialflowsandmanufacturingpathwaysthatusethemasinputsforthemanufactureofnewmaterials.Adynamicmaterialsflowmodelisemployedforsteeltoanticipatethefutureavailabilityofsteelscrap.Forplastics,avarietyofend-of-lifeflowsareanticipatedbasedonstocklevelsandproductlifetimes;theseareevaluatedforrecyclingandrecoverysuitability,includinginfluencesonyields,quality,andtheconsequenteffectivesubstitutionofnewproduction.Chemicalrecyclingisdefinedasanewplasticsproductionmethod,withanemphasisonhigh-carbonmassbalanceapproaches.Plasticend-of-lifeincinerationisalsomodeled,andCO2emissionsarecalculated.Thepotentialforcementrecyclingofconcreteparticlesandunhydratedcementrecoveryiscalculated.Ascenarioanalysiscombinesthesethreecomponents.Allscenariosaredesignedtoachievenear-zeroCO2emissionsfromtheindustrialoutputby2050.Backcastingisusedtobuildfive-yearpathsthataccountforcapitalstockturnover,progressiveincreasesintechnicalmaturity,buildingleadtimes,andotherconstraints.MaterialEconomics(2019)investigatesvariousstrategiestomaintainEUsteel,plastic,ammonia,andcementoutputwhileachievingnet-zeroemissions.Itestimatesthepossibleeffectsofvarioussolutionsanddeterminesthatemissionsfromthoseindustriesmaybedecreasedtozeroby2050,supportingtheconclusionsofthepathsoutlinedinEuropeanCommission(2018).ResourcesfortheFuture64TableC.9.Energy-Environment-EconomyGlobalMacro-Economic(E3ME)TypeMacro-econometricIndustrysectorsNACE2-digitApproachTop-downSpatialresolutionGlobal(61regions)TemporalresolutionYearlyto2050E3MEisaglobal,macro-econometricmodeldevelopedbyCambridgeEconometricstoaddresstheworld’skeyeconomic,social,andenvironmentalissues(CambridgeEconometrics2019).Themodelhasahighlevelofdisaggregation,allowingforanexhaustiveanalysisofsectoralandcountry-leveleffectsfromavarietyofscenarios.Socialimpactsaresignificantmodeloutputs.Anotherdistinguishingfeatureistheeconometricspecification,whichsolvesproblemswithmacroeconomicmodelsandoffersasolidempiricalfoundationforresearch.Themodelcanaccuratelyexaminebothshort-andlong-termimplicationsandisnotconstrainedbymanyoftherestrictiveassumptionsthataretypicalincomputablegeneralequilibriummodels.Analysesoftheworld’seconomies,energysystems,emissions,andmaterialdemandsarealsoincludedinE3ME.ThismakesitpossibleforE3MEtorepresentthesecomponentsinanon-linearinteractionwithtwo-waysfeedbacks..E3MEincludes61worldwideregions,withfullsectoralbreakdownsineach,andforecastsannuallythrough2050.Itiscommonlyusedatthenational,European,andgloballevels,aswellasforbroaderEuropeanandglobalpolicyanalyses.E3MEisbasedontheESA95systemofnationalaccounts,alongwithbalancesforenergyandmaterialdemands,aswellasenvironmentalemissionsflows.Italsoincludesdetailedhistoricaldatasets,timeseriesthatspantheperiodsince1970,andsectoraldisaggregationbasedontheNACEclassificationofeconomicactivitiesatthetwo-digitlevel.E3MEiscomposedofthreemodules:economy,environment,andenergy.Themodel’seconomicmoduleissolvedforeachregion.Mosteconomicvariablesareaddressedatthesectorallevel.Althoughsingle-countrysolutionsarefeasible,theentiresystemisaddressedsimultaneouslyforallindustriesandareas.Unlesstherearerestrictionsonavailablesupply,demanddeterminesproductionandemployment.Thekeyexplanatoryvariablesforaggregateenergydemandareeconomicactivityineachoftheenergyusers,averagerealenergypricesforeachenergyuser,andtechnologicalvariables,whicharerepresentedbyinvestment,R&Dspending,andspilloversinmajorindustriesthatmanufactureenergy-consumingmachineryandvehicles.Foreachoftheenergyconsumersinthemodel,emissionsdataforCO2fromenergyuseareaccessible,andcoefficientsarecalculatedfromhistoricaldata.Thisestablishestheconnectionbetweenenergyuseandemissions.ProcessCO2emissions,suchasthosefromtheIndustrialDeepDecarbonization:ModelingApproachesandDataChallenges65cementandchemicalsindustries,areexplicitlyincludedinthemodelbutaretiedtoproductionfromthoseindustriesratherthanenergyuse.Otheremissionsaretreatedinalessdetailedway,andfindingsareoftennotbrokendownbyindustry.GramkowandAnger-Kraavi(2019)analyzeachangeinBrazil’seconomywhilecontributingtotheParistargets,usingthemanufacturingsectorsasanexample.E3MEisusedtoprojectBrazil’sgrowthoutlookupto2030withandwithoutaportfoliooffiscalpoliciesthatpromotelow-carboninvestments.TheresearchshowsthattherightcombinationofstrategiescanassistinmodernizinganddecarbonizingtheBrazilianmanufacturingsectorsandenablethenation’seconomytogrowmorequicklywhilereducingCO2emissions.ISEEM-IS,developedbytheLawrenceBerkeleyNationalLaboratory,isabottom-upoptimizationenergymodelingframeworkforrepresentingimpactsofenergypoliciesonUSironandsteelproduction(FigureC.1).Itisalinearprogrammingoptimizationmodelthatminimizesthecostsofproductionoverasetofpredefinedindustrialconstraintsacrossasetofplantsdefinedbytechnologies.ThemodelingframeworkanalyzestheuseandpotentialimprovementsoftechnologiesintheUSironandsteelsectorwithrespecttoreducingcarbonandGHGemissions,aswellaswidereconomicimplicationsofsuchenergyandenvironmentalpolicies.Itincorporatesinternationaltrade,specificallyfromIndiaandChina;energyandemissionspolicies;andthekeyassumptionthatproductiontechnologieschangeandimprovegraduallyovertime.ISEEM-IS’smainstrengthisitstechnologicaldetailandtheendogenousrepresentationofinvestmentanddeploymentofneworrefinedtechnologiesofproductionandsupply.Eachtechnologyhasitsownsetofconstraints,requirements,parameters,andratesofgrowthorchange.Themodeldividesproductiontechnologiesintothreecategories:current,advanced,andenergy-efficient.CurrentproductionTableC.10.IndustrialSectorEnergyEfficiencyModelforIronandSteel(ISEEM-IS)TypeLinearprogrammingOptimizationIndustrysectorsIronandsteelApproachBottom-upSpatialresolutionUnitedStates,China,IndiaTemporalresolution2010–50ResourcesfortheFuture66technologiesarethosethatarecurrentlyinuseintheindustry,suchasbasicoxygenfurnaceandelectricarcfurnace;advancedproductiontechnologiesrepresentautonomouslyimprovedversionsofcurrentproductiontechnologies;andenergy-efficienttechnologiesarethosethatimproveenergyefficiencyofcurrentproductiontechnologiesbutareassociatedwithextracosts.ThefulllistofproductiontechnologiesincludedinISEEM-IScanbefoundinKaralietal.(2013,AppendixesAandC).Despitethetechnologicaldetail,theenergy-efficienttechnologiesarelimitedtomethodsandprocessesofproductioninironandsteelanddonotincludeotheremissionsreductiontechnologiesthatareproduction-adjacent,suchasCCS.TheISEEMmodelstructurecanbeappliedtootherindustrialsectorsbutrequireshugeamountsofdatathatmightnotalwaysbeavailable.Karalietal.(2013)usethemodeltoinvestigatetheimpactofcarbonreductionoptionsontheUSironandsteelsectorunderasetofspecificscenarios.Theyalsoexaminehowlocalpoliciesandemissionsreductionstrategieswouldaffecttradewith,productionin,andCO2emissionsfromChina’sandIndia’sironandsteelsectors.Thepolicyscenariosimposereductionsinglobalemissionsof10,20,and30percentunderthreevariationsofthemodel:(1)withoutcommodityorcarbon(allowance)trading(ER),(2)withcommoditytradingbutwithoutcarbontrading(ET),and(3)withbothcommodityandcarbontrading(EC).OneofthekeyfindingsofKaralietal.(2013)washowcarbonemissionsreductionpoliciesinteractwithnotjustdomesticproductionbutalsointernationalproductionandtradeflowsbothofthecommodityproduced(steel)andofcarbon.Forexample,underthetradingscenarios,theauthorsfindthatproductionpartiallyshiftedoutoftheUnitedStatestoChinaandIndia,asitwaslessexpensivetoimportthantoinvestinenergy-efficienttechnologiesdomestically.Withrespecttothebaselinein2030undertheEC-andET-30scenarios,USimportsincreasedbyabout55percent,andUSproductiondecreasedbyabout25percent.Furthermore,underthetradingscenarios,whileChina’sandIndia’senergyintensitylevelsaremuchreduced,thoseoftheUnitedStatesremainedunchangedfromthebasescenariolevels,astherewaslittletonoincentivetoswitchtoenergy-efficienttechnologiesintheUnitedStates.IndustrialDeepDecarbonization:ModelingApproachesandDataChallenges67FigureC.1.ProductionFlowDiagramoftheIronandSteelSectorintheISEEM-ISModelSource:Karalietal.(2013)TableC.11.UniversalIndustrialSectorsIntegratedSolutions(U-ISIS)TypeLinearprogrammingOptimizationIndustrysectorsCementPulpandpaperApproachBottom-upSpatialresolutionUnitedStateswithdemandandsupplyregionsTemporalresolution2010–50ResourcesfortheFuture68U-ISISisabottom-upsector-baseddynamiclinearprogrammingoptimizationmodelingframeworkdevelopedbytheUSEnvironmentalProtectionAgency(EPA).Themodeloptimizestotalsurpluswithrespecttovariousconstraints,includingsupply,demand,costsofproduction,technologiesavailable,andenergypoliciesinstitutedbytheUSgovernment.Themodelincorporatesmultipleindustrieswithinamultimarket,multiproduct,multipollutant,andmultiregionemissionstradingframework.Itanalyzesoptimalsectoroperationstomeetdemandandpollutionreductionrequirementsoveraspecifiedperiodoftime,whiletakingintoaccountplant-leveleconomicandtechnicalfactorsandcosts,includingproductioncapacitychanges,fixedandvariableproductioncosts,transportationcosts,importcosts,emissionscosts,andenergyintensityandefficiency.ThemainstrengthofU-ISISisitsabilitytomodelindustrypollutiongenerationpathwaysandmethodsforabatingemissionsresultingfromthosepathwaysthroughbothmitigationandprevention.Themodelincludesmethodsfortrackingmultiplepollutantstreams—aswellasmultiplepollutants—associatedwithcontrolledanduncontrolledemissions,pollutionpreventionmeasures,andothercontrol-relatedeffects.Forthecementindustry,pollutantsincludecriteriapollutants,hazardousairpollutants(HAPs)suchasmercuryorhydrochloricacid,CO2,nitrogenoxides(NOX),sulfurdioxide(SO2),andparticulatematter.Thegenerationpathwaysincludecementmanufacturing,quarryingoperationsforrawmaterials,kilnoperations,andfuelcombustion.Themodelalreadyincludesinitsbasescenariotheextantregulatoryrequirementswithrespecttopollutantsandcanrunscenarioanalysesforavarietyofpolicyscenariostoaddresstheseemissionspathwaysandabatementoptions,includingemissionslimits,capandtrade,andemissionstaxesunderlong-andshort-termhorizons(decadesandannual)andregionalornationalrequirements.U-ISIShasanincredibleamountofplant-leveldetail—butthatcomesatthecostoflengthyandregularcalibrationandrecalibrationofthemodelingframework.Furthermore,themodel’sexistinguserinterfacerestrictsthekindofscenariothatcanberun.Itisunclearfromthedocumentationhowflexibletheinterfaceis,suchashowdifficultitwouldbetoaddatechnologyorscenarionotalreadywithintheinterface,orhownecessaryitistorunthemodel.TheU-ISISmodelwasfirstusedbyEPAtoexaminetheUSPortlandcementindustry,withafocusonhowpoliciesforemissionsreductionaffecttheindustry(EPA2013).EPAlateradaptedthemodeltoexaminetheUSpulpandpaperindustry(BhanderandJozewicz,2017).Forthecementindustry,itaddressedthedevelopmentofefficientandeffectivepolicyoptionsformanagingemissionsandairqualityresultingfromallstepsofcementproduction.Themodeldemonstrationforpulpandpaperfocusedonidentifyingoptimalindustryoperationthroughtheselectionofcost-effectivecontrolstomeetdemandwhilecomplyingwithemissionsreductionrequirementsandcalibratingabaselinebusiness-as-usualscenario.ThedemonstrationalsoshowcasedthreescenariosforNOXemissionsreductions:fuelsubstitution,installationofcontrols,andimplementationofenergyefficiencymeasures.IndustrialDeepDecarbonization:ModelingApproachesandDataChallenges69TableC.12.HybridTechnologicalEconomicPlatform(HYBTEP)TypeOptimizationandsimulationIndustrysectorsAluminumIronandsteelChemicalCementPulpandpaperOtherindustryApproachBottom-upandtop-downSpatialresolutionMultinationalorsinglecountryTemporalresolution2010–50HYBTEPisacombinationofthebottom-upTIMESmodel(seeAppendixC.6)andthetop-downCGEGeneralEquilibriumModelforEconomy,Energy,andEnvironment(GEM-E3)(Fortesetal.2014).GEM-E3isadynamicrecursiveCGEmodelthatsolvesfortheequilibriumpriceofgoods,services,labor,andcapitaltoclearallmarketsandoptimizebehaviorofeconomicagentssimultaneously.Bothmodelsaremultinationalbutcanbeadjustedtomodelasinglecountry.TheHYBTEPframeworkcreatesasoftlinkbetweenthetwosystems,solvingthemconcurrentlyandexchanginginformationbetweenthem.HYBTEP’smainstrengthisthatitssoft-linkmethodologyallowsuserstoperformintegratedassessmentsofclimateandenergypolicyinstrumentswithdetailedtechnologyprofilesfortheenergysector,whichissomethingthatneithertop-downCGEmodelsnorbottom-upmodelscandoontheirown.Softlinkingmaintainsthestructureandindividualstrengthsofeachmodelwhileeliminatingmanyofthedrawbacks.Itincorporatesanextensivegroupoftechnologiesandeconomicresponses,allowingforgreaterunderstandingoftheimpactsandeffectsofvariousenergyandclimatepolicies.Despitethesestrengths,HYBTEPdoeshaveafewlimitations,mostlyinheritedfromthetwomodelsitlinks—inparticular,theassumptionofperfectlycompetitivemarketsandoptimisticviewsofdeploymentoffuturetechnologies.TheplatformwasusedtoprovideinsightsintotheadvantagesofthehybridsystemthroughexaminingthemacroeconomiceffectsofvariousclimatepoliciesinPortugal.Thesepolicyscenarioswererunrelativetoacalibrationscenario(whichwasnotabusiness-as-usualscenario,giventhewayTIMESoptimizestheenergysystem)andincludedacurrentpolicyregulationscenario(CPR),aCO2pricescenario(TAX),andarenewableenergysupportscenario(RES).TheCPRscenarioreflectedthecurrentstateofenergyandclimatepolicies,includingreductionsinGHGemissions,increasesinrenewableenergyconsumption,andimprovementinenergyefficiency.TheTAXscenarioaddedtotheCPRscenarioadomesticcarbontax(setatthehighestlevelResourcesfortheFuture70indicatedinanEUroadmapfortransitioningtoacompetitivelow-carboneconomy)onGHGemissionsfromenergyconsumptioninsteadoftheemissionscapsfromtheEmissionsTradingSystem(ETS)andnon-ETSsources.TheRESscenarioaddedamonetaryincentiveforrenewableenergy(biofuels,solarandbiomassconsumption,renewableelectricity)totheCPRassumptions.TheFORECASTmodel,developedbytheFraunhoferInstituteforSystemsandInnovationResearch,isabottom-upsimulationmodelforthedevelopmentoflong-termscenariosforfutureenergydemandintheindustrialsectors,services,andhouseholds(Fleiteretal.2018).Itisamultinationalmodelthatcanbeadjustedforasinglecountryandcandisaggregateresultsdowntothedistrictlevelifdesired.Themodelincorporatesplant-leveldataandincludesbothenergy-intensiveindustrialsectorsandlessenergy-intensivesubsectorsandapplications.ItsmainoutputisatimeseriesoffinalandusefulenergydemandandtherelatedGHGemissionsunderahighlevelofdisaggregationregardingenergycarriers,subsectors,enduses,andtechnologies,presentedbycountryandscenario.OneofthecorestrengthsoftheFORECASTmodelisthatitincorporatesahighleveloftechnologicaldetail,policyparameters,andtransitionpathsandcostsinanintegratedapproach.FORECASThassixsubmodels:macro,energy-intensiveprocesses,spaceheatingandcooling,electricmotorsandlighting,furnaces,andsteamandhotwater.Itsindustrialsubsectorsincludepaperandprinting,nonmetallicmineralproducts,nonferrousmetals,ironandsteel,chemicalsindustry,foodanddrinkandtobacco,engineeringandothermetal,andothernonclassified.Thesimulationsareconductedattheindividualsubsectorlevel(e.g.,ironandsteel),whereenergy-intensiveprocessesareconsideredexplicitly,andothertechnologiesandenergy-usingequipmentaremodeledsimilarlyacrossallsubsectors.Further,themodelcancalculateTableC.13.FORECASTTypeSimulationIndustrysectorsPaperandprintingNonmetallicmineralproductsNonferrousmetalsIronandsteelChemicalsOtherindustryApproachBottom-upSpatialresolutionMultinationalorsinglecountryTemporalresolution2010–50IndustrialDeepDecarbonization:ModelingApproachesandDataChallenges71comprehensivetransition-decarbonizationscenariosforanindividualcountry’sentireindustrysectorthroughabroadscopeofmitigationoptions.However,thediversemethodsusedtogainsectoralandtechnologicaldetailreducetransparencyandrendertheinterpretationofresultsmoredifficult.FORECASTiswritteninVisualBasic.FORECASTcanaddressnumerousquestionsrelatedtoenergydemandandGHGemissions,specificallyinthecontextoftechnicalchange.Thesequestionsincludescenariosforfuturedemandofindividualenergycarriers,calculationsofenergysavingspotentialsandtheirimpactsonGHGemissions,abatementcostcurves,exanteandexpostpolicyimpactassessments,andlow-carbontransitionscenarios.Themitigationoptionsincludeenergyefficiency(incrementalandradicalchange),fuelswitching(torenewableandlow-carbonenergycarriers),CCS,circulareconomyandrecycling,andmaterialefficiencyandsubstitution.TwoexamplesofFORECAST’suseareastudyofthecementindustryinTaiwanandanalysisofenergyefficiencyandananalysisofadecarbonizationpathwayforGermany’sindustrialsectors.IntheTaiwancase,Huangetal.(2016)findthatadoptionofenergy-efficienttechnologycanresultin25percentsavingsforelectricityand9percentsavingsforfuels,ofwhich91percentcouldbeimplementedcost-effectivelyunderanassumeddiscountrateof10percent.IntheGermanycase,Fleiteretal.(2016)considertwoscenarios:areferencescenarioreflectingcurrentpolicies,economic,andtechnologicaltrendsandatransitionscenarioachievingaGHGemissionsreductioninindustryof83percentby2050(withaGHGemissionsreductionintheentireeconomyof80percent).Underthetransitionscenario,electricitydemandisreducedby16percentandfueldemandby32percentby2050;biomassuseincreases;coaluseisphasedoutinallindustrysectorsexceptironandsteel;useofalternativematerialsincreasesinthepaper,cement,glass,andaluminumsectors;andCCSmitigatesabout35metrictonsofCO2ein2050,withtotalemissionsreducedfrom140metrictonsin2010(thebaseyear)to75metrictonsofCO2eby2050versus110metrictonsofCO2eunderthereferencescenario.ResourcesfortheFuture72AppendixD.ExamplesofApplicationsUsingDifferentModelsD.1.WorldEnergyModelsFortheWorldEnergyOutlook,WEMusesascenarioapproachtolookatpotentialchangesintheenergysector,modelingtheworldfor26regions.FourscenariosweresimulatedindepthfortheWorldEnergyOutlook2021(seeAppendixC.1).D.2.NEMSHuangandEckelman(2020)modelmaterialfloweconomyintheUnitedStates.HuangandEckelman(2021)estimatepollutantsfromindustryintheUnitedStates.BrownandBaek(2010)estimaterenewablefuelstandardsimpactsonUSindustry.Aroraetal.(2018)estimatetaxationrecyclingimpactsonUSindustry.Ruthetal.(2000)estimateimpactsofmarket-basedclimatechangepoliciesontheUSpulpandpaperindustry.D.3.GCAMLiuetal.(2015)usetheGCAM-USAversiontoexplorethewater-energynexus(multisectoranalysis).TheUnitedStatesisrepresentedatastatelevel.Pengetal.(2021)estimatetheeffectsofstateandnationalcarbonpolicies(multisectoranalysis).TheUnitedStatesisrepresentedatastatelevel.D.4.REMINDLudereretal.(2012)employaversionoftheREMINDmodeltoexamineAsia’sparticipationintheglobalefforttomitigateclimatechange.D.5.MUSEBudinisetal.(2020)modelthedecarbonizationoftheChineseammoniaindustry.IndustrialDeepDecarbonization:ModelingApproachesandDataChallenges73D.6.TIMESNappetal.(2019)usetheTIAMversionandfocusonadvancesinenergydemandsectors,includingbestavailabletechnologiesinindustry.ThisstudymodelstheworldandrepresentstheUnitedStatesasoneregion.Faisetal.(2016)focusondecarbonizationoptionsforUKindustry.D.7.IMAGEQuietal.(2022)focusondirectaircaptureimplicationsforthepowersector.ThisstudymodelstheUnitedStatesasoneregion.Chenetal.(2021)analyzemultisectorcarbonneutrality;industrydiscussedaggregated.ThisstudymodelstheworldandrepresentstheUnitedStatesasoneregion.Sharminaetal.(2020)includehard-toabatesectorsandcirculareconomyforindustry.ThisstudymodelstheworldandrepresentstheUnitedStatesasoneregion.Kermelietal.(2019)focusonthecementindustry.ThisstudymodelstheworldandrepresentstheUnitedStatesasoneregion.D.8.MaterialEconomicsModelingFrameworkMaterialEconomics(2019)usesthisframeworktoinvestigatevariousstrategiestomaintainEUsteel,plastic,ammonia,andcementoutputwhileachievingnet-zeroemissions.D.9.E3MEBachneretal.(2020)modeldecarbonizationofironandsteelindustryinEurope.GramkowandAnger-Kraavi(2019)modelinvestmentsinBrazilianindustrydecarbonization.D.10.ISEEM-ISKaralietal.(2013)investigatetheimpactofcarbonreductionoptionsontheUSironandsteelsector.ResourcesfortheFuture74D.11.U-ISISEPA(2013)exmainestheUSPortlandcementindustryusingU-ISIS.BhandarandJozewicz(2017)applythemodeltothepulpandpapersector.D.12.HYBTEPFortesetal.(2014)applyHYBTEPtoanalyzethreeclimateandenergypolicyscenariosinPortugal.D.13.FORECASTFleiteretal.(2016)conductsananalysisofadecarbonizationpathwayforGermany’sindustrialsectors.Huangetal.(2016)studiesthecementindustryinTaiwan.IndustrialDeepDecarbonization:ModelingApproachesandDataChallenges75

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

碎片内容

碳中和的最新文档