Shared Socio-Economic Pathways of the Energy Sector – Quantifying the Narratives
Nico Bauer
a,*, Katherine Calvin
b, Johannes Emmerling
c,d, Oliver Fricko
e,
Shinichiro Fujimori
f, Jérôme Hilaire
a,g, Jiyong Eom
b,h, Volker Krey
e, Elmar Kriegler
a, Ioanna Mouratiadou
a, Harmen Sytze de Boer
i, Maarten van den Berg
i, Samuel Carrara
c,d, Vassilis Daioglou
i, Laurent Drouet
c,d, James E. Edmonds
b, David Gernaat
i, Petr Havlik
e, Nils Johnson
e, David Klein
a, Page Kyle
b, Giacomo Marangoni
c,d, Toshihiko Masui
f, Robert C. Pietzcker
a, Manfred Strubegger
e, Marshall Wise
b, Keywan Riahi
b, Detlef P. van Vuuren
i,jaPotsdamInstituteforCIimateImpactResearch(PIK),Germany
bPacificNorthwestNationalLaboratory(PNNL),MD,UnitedStates
cFondazioneEniEnricoMattei(FEEM),Italy
dCentroEuro-MediterraneodeiCambiamentiClimatici(CMCC),Italy
eInternationalInstituteforAppliedSystemsAnalysis(IIASA),Austria
fNationalInstituteforEnvironmentalStudies(NIES),Japan
gMercatorResearchInstituteforGlobalCommonsandClimateChange(MCC),Germany
hGraduateSchoolofGreenGrowth,KAISTBusinessSchool,RepublicofKorea
iNetherlandsEnvironmentalAssessmentAgency(PBL),TheNetherlands
jCopernicusinstituteforsustainabledevelopment,UtrechtUniversity,TheNetherlands
ARTICLE INFO Articlehistory:
Received15December2015 Receivedinrevisedform20July2016 Accepted24July2016
Availableonline23August2016 Keywords:
SharedSocio-economicPathways(SSPs) IntegratedAssessmentModels(IAMs) Energysystem
Energydemand Energysupply Energyresources
ABSTRACT
Energyiscrucialforsupportingbasichumanneeds,developmentandwell-being.Thefutureevolutionof thescaleandcharacteroftheenergysystemwillbefundamentallyshapedbysocioeconomicconditions anddrivers,availableenergyresources,technologiesofenergysupplyandtransformation,andend-use energydemand.However,becauseenergy-relatedactivitiesaresignificantsourcesofgreenhousegas (GHG)emissionsandotherenvironmentalandsocialexternalities,energysystemdevelopmentwillalso beinfluencedbysocialacceptanceandstrategicpolicychoices.Alloftheseuncertaintieshaveimportant implicationsformanyaspectsofeconomicandenvironmentalsustainability,andclimatechangein particular.IntheShared-SocioeconomicPathway(SSP)frameworktheseuncertaintiesarestructured intofivenarratives,arrangedaccordingtothechallengestoclimatechangemitigationandadaptation.In this study we explore future energy sector developments across the five SSPs using Integrated AssessmentModels(IAMs),andwealsoprovidesummaryoutputandanalysisforselectedscenariosof globalemissionsmitigationpolicies.Themitigationchallengestronglycorrespondswithglobalbaseline energysectorgrowthoverthe21stcentury,whichvariesbetween40%and230%dependingonfinal energyconsumerbehavior,technologicalimprovements,resourceavailabilityandpolicies.Thefuture baselineCO2-emissionrange is even larger, asthe most energy-intensiveSSP also incorporates a comparativelyhighshareofcarbon-intensivefossilfuels,andviceversa.Inter-regionaldisparitiesinthe SSPsareconsistentwiththeunderlyingsocioeconomicassumptions;thesedifferencesareparticularly strongintheSSPswithlargeadaptationchallenges,whichhavelittleinter-regionalconvergenceinlong- termincomeandfinalenergydemandlevels.Thescenariospresenteddonotincludefeedbacksofclimate changeonenergysectordevelopment.TheenergysectorSSPswithandwithoutemissionsmitigation policiesareintroducedandanalyzedhereinordertocontributetofutureresearchinclimatesciences, mitigationanalysis,andstudiesonimpacts,adaptationandvulnerability.
ã2016TheAuthors.PublishedbyElsevierLtd.ThisisanopenaccessarticleundertheCCBYlicense (http://creativecommons.org/licenses/by/4.0/).
*Correspondingauthorat:P.O.Box601203,D-14412Potsdam,Germany.
E-mailaddress:Nico.Bauer@pik-potsdam.de(N.Bauer).
http://dx.doi.org/10.1016/j.gloenvcha.2016.07.006
0959-3780/ã2016TheAuthors.PublishedbyElsevierLtd.ThisisanopenaccessarticleundertheCCBYlicense(http://creativecommons.org/licenses/by/4.0/).
ContentslistsavailableatScienceDirect
Global Environmental Change
j o u r n a lh o m e p ag e :w w w . e l s e vi e r . c o m / l o c a t e / g l o en v c h a
1.Introduction
The transformation of the energy sector is important in addressingthechallengesofbothclimatechangemitigationand adaptation. On the one hand, it is the main contributor to GreenhouseGas(GHG)emissionsandairpollution(Blancoetal., 2014) resulting in much emphasis being put on emission mitigation(Clarkeetal.,2014).Ontheotherhand,globalenergy systemsarevulnerabletoclimatechangeandcanserveasmeans for adaptation to a changing climate (Bazilian et al., 2011;
ChandramowliandFelder,2014;CiscarandDowling,2014;Fricko etal.,2016,2017;IsaacandvanVuuren,2009).Theenergysector transformation also has important implications for social and environmentalsustainabilitygoals(vonStechowetal.,2015).This studyintroducesand discusses theenergysectorresultsof the IntegratedAssessment Models (IAMs) quantification of thefive SharedSocio-economicPathways(SSPs)forthebaselinesandtwo climatechangestabilizationlevels.TheSSPsprovideaframework for assessing socio-economic challenges to climate change mitigation and adaptation, as wellas analyzing broader social andenvironmentalsustainabilityissues.
Various energy sector challenges and the way they are addressedarecrucialinshapingfuturetransformationpathways withimportant implicationsformitigationand adaptation.Five key energy sector challenges, to support basic human needs, developmentandwell-beingare(i)energydemandgrowthandits couplingwithdemographicandeconomicdrivers,(ii)thephasing outoftraditionalformsofenergyuse,improvingenergyaccessand modernizationofenergyuseinthecontextofstructuraleconomic change,(iii) theexpansion of primary energysupplies, (iv) the futureofexistingandbuild-upofnewenergyinfrastructuresand technologies,and(v)theGHGandotherpollutantemissionsand their mitigation. These challenges are related to key scientific debates on global and long-term developments in the energy sector. The coupling between socio-economic development patternsandenergydemandhasbeenidentifiedasafundamental issueforunderstandingthescaleandstructureofenergydemand (CsereklyeiandStern,2015;Grübleretal.,2012;Jakobetal.,2012;
Schäfer,2005).Historicaltrendsshowthateconomicdevelopment iscorrelatedwithmodernizationoftheenergymixtowardshigher sharesofelectricityandgasesandlowersharesofsolidandliquid energycarriers(FouquetandPearson,2012;Grübleretal.,1999).
These shifts are related topreferences for alternative lifestyles expressed in consumer choices, transportation modes, etc.
Dedicatedenergyaccesspoliciesarealsodiscussedasmeansto enhance the modernization process in developing countries (Pachauri et al., 2013).The availability, trade and useof fossil fuels, and energy security concerns, are key energy sector challengesstronglyrelatedtomitigation(McCollumetal.,2014;
Bauer et al., 2015; McGlade and Ekins, 2015). The lock-in of incumbenttechnologiesandthediffusionofinnovativetechnolo- gies for energy demand and supply, much depend on socio- economic and political factors as well as the development of technology performance and costs (Goldemberg, 1998; Unruh, 2000). Overcoming limitations on up-scaling of innovative technologies and their wide diffusion are key energy sector challengesfor policymakers(vanSluisveld etal., 2015;Wilson et al., 2013).The corresponding scientific debates are far from providingfinalconclusions,butareopeninguptheperspectiveon multipleuncertaintiesthatarecrucialforclimatechallengesand broadersustainabilityissues.
Althoughtheenergysectordevelopmentsaresubjecttostrong near-terminertias,theirfundamentalfactorsanddrivingforces– suchasdemographicchange,economicgrowth,andtechnological change–becomefluidanduncertainastheperspectivestretches towardsthemiddle,oreventheendofthe21stcentury.Therefore,
theshapeoffutureenergysectorpathwaysisdeeplyuncertainas are the resulting social, economic, political and environmental consequences.Awayofaddressingtheuncertaintyistoformulate alternative sets of input assumptions for dynamic drivers, parametersandpolicysettingsthatgiverisetodifferentenergy transformationpathways.Toensureconsistencyoftheseassump- tions along multiple dimensions, they are often bundled into scenarios derived from broad narratives about socio-economic futures, as in the Special Report on Emission Scenarios (SRES, Nakicenovicetal.,2000).Theuseofscenariosisacommontool appliedtostudypossiblelong-termenergyfutures(Nakicenovic etal.,2000;Riahietal.,2012;Turkenburgetal.,2000),particularly forthosewithafocusonclimatechangestabilization(Bruckner etal.,2014;Clarkeetal.,2014).
TheSSPsarethenextgenerationofscenarios,succeedingthe SRES published in 2000, and they are intended to serve as referencescenariosforvariousassessmentsintheareaofclimate changechallenges,aswellasbroadersustainabilityissues(van Vuuren et al.,2014).The SSPscomplement the Representative ConcentrationPathways(RCPs,vanVuurenetal.,2011)byadding the underlying socio-economic narratives and quantitative pathways consistent with the challenges to mitigation and adaptation. TheSSPsincludefivevastlydifferentglobal futures (SSP1-5) thatstartat thenarrativeforalternativedevelopment pathways,andvary,dependingonhowtheenergychallenges(i- iv)areaddressed(O’Neilletal.,2017).TheSSPbaselinesrepresent purereferencecases thatexclude(i)climatechangemitigation policies (such as the Parisagreement) and (ii) feedbacks from climate change on socio-economic or natural systems. For example,usingtheSSPsasastartingpointformitigationpolicies (Kriegler et al.,2014),enables thedifferences tothereference baseline to be examined. This will be part of this study. The climate change mitigationcases describeenergy systempath- ways that reach forcing levels consistent with the RCPs. The energy sector pathways presented in this paper are part of broaderSSPscenariosthatalsocoverotherkeydimensions.The overviewincludingoverallGHGemissionsisgivenbyRiahietal.
(2017),whereasland-use andcompetition (incl.bio-energy)is analyzedbyPoppetal.(2017)andairpollutionimplicationsare exploredbyRaoetal.(2017).
Theremainderofthispaperisorganizedasfollows.InSection2 weintroducetheenergysectorSSPsataqualitativelevelandthe scenariosthathavebeencomputedforeachSSP.Section3presents the quantitative energy sector pathways. Finally, Section 4 summarizes the energy sector SSPs, discusses the results and indicatesdirectionsforfutureresearchusingtheSSPs.
2.Methods
SixleadingIAMshavecontributedtothequantificationofthe energy-land-use-emissionsoutcomesassociatedwiththefiveSSP narratives (see Table1 foran overviewand theSupplementary materialfordetailsontheIAMs).TheSSPshavebeenquantified usingIAMsthatintegrateeconomy,energy,land-use,andclimate, coveringallGHGsandairpollutants.ForeachSSP,theresultsofone IAM havebeen selectedas theMarkerthat best illustratesthe narrativeoftheSSP.Fortheresultspresentedhere,wefocusour discussion on the Marker scenarios and provide cross-model ranges. For the selection of the Marker scenarios see the SupplementarymaterialandRiahietal.(2017).
The SSPs have been implemented into the IAMs at various levels.FortheSSPdriver scenariosof populationandeconomic growth,quantitativeprojections,thataredevelopedinlinewith theSSPnarratives,havebeenadoptedbyallmodels(Dellinketal., 2017;KCandLutz,2017;bothinthisissue).Theseprojectionsare complemented by qualitative harmonization on energy sector
developmentalong thelinesofthebasicand theextendedSSP narratives(Section2.1).TheSSPshavebeenimplementedintothe IAMs by systematically varying the assumptions along various dimensions accordingto thebasic and the extended SSPs.The modelsderivedaBaselinescenarioandadditionallycalculateaset ofclimatepolicyscenarios toachievelong-termclimatechange stabilizationatvariousambitionlevels(Section2.2).
2.1.TheSSPnarrativesfortheenergysector
ThebasicSSPsnarratives(O’Neilletal.,2017)providetheoverall scenarioframingfor thevariousdimensionsthatdeterminethe challengestomitigationand adaptation. Thegeneral character- istics of the basic SSPs relevant for the energy sector are summarized in Fig. 1. These also relate to the energy sector challengesmentionedabove.TheextendedSSPsaredesigned to
interpretthebasicSSPsandservetoqualitativelyharmonizethe models providing more detail in three domains of the energy sector: (i) final energy demand development, (ii) energy conversiontechnologiesincludingspecificmitigationtechnologies and(iii)thefossilfuelsupply.Detailedinformationisprovidedin theSupplementarymaterial.TheextendedSSPsguidedmodeling teamstoderiveassumptionsfortheirmodelimplementations.No attemptwasmadetoprescribesharedquantitativeassumptions becausetheteamsfollowdifferentmodelingapproachesand,thus, differentparameterdefinitionsapply.
ThemainpointsoftheextendedSSPswithaperspectiveonthe energysectorareas
SSP1sustainability—takingthegreenroad
Economicvaluecreationdecouplesfrommaterialconsumption and final energy demand. This is combined with a strong
Fig.1.OverviewofbasicSSPs,theenergysectorelementsofthenarrativesandtheSPAspecifications(O’Neilletal.,2017).HICandMICabbreviationsforHighandMedium IncomeCountries,respectively.TheSharedClimatePolicyAssumptions(SPAs),coloredinyellow,arenotusedinthebaselinescenarios,butonlyinthemitigationscenarios introducedinSec.2.2.(Forinterpretationofthereferencestocolourinthisfigurelegend,thereaderisreferredtothewebversionofthisarticle.)
Table1
OverviewofSSPsandIAMs.
SSP Descriptor Markerteam(institution) MarkerpaperinthisSpecialIssue Alsocomputedby
SSP1 Sustainability IMAGE(PBL) VanVuurenetal.(2017) All
SSP2 Middle-of-the-Road MESSAGE-GLOBIOM(IIASA) Frickoetal.(2017) All
SSP3 RegionalRivalry AIM/CGE(NIES) Fujimorietal.(2017) IMAGE,GCAM,MESSAGE-GLOBIOM,WITCH-GLOBIOM
SSP4 Inequality GCAM4(PNNL) Calvinet al.(2017) AIM/CGE,WITCH-GLOBIOM
SSP5 Fossil-fueledDevelopment REMIND-MAgPIE(PIK) Kriegleretal.(2016) AIM/CGE,GCAM,WITCH-GLOBIOM
modernizationofenergyuseduetotechnologicaldevelopment, lifestyle changes and policies supporting energy efficiency improvements. Social acceptability is generally low for all technologies (particularly nuclear) except non-biomass renew- ables.Thelatterissubjecttorapidtechnologicalimprovements, buttheseareparticularlyslowinthefossilfuelsector.
SSP2middle-of-the-road
Energyintensity improvementscontinue at global historical growth rates with a medium degree of regional convergence.
Technologicalimprovementsaremediumforalltechnologiesand socialacceptancedoesnotshiftmarkedly.Thisresultsinmoderate growthoftheenergysector,noremarkableshiftsintheprimary energymixandcontinuedmodernizationofthefinalenergymix.
SSP3regionalrivalry—arockyroad
Fastpopulationgrowthin developing countriesiscombined with slow economic growth and income convergence. Slow technologicaldevelopment,materialintensivelifestylesandlittle environmental awareness maintain the strong link between economic activity and final energy demand. Modernization of finalenergy useisslowandtraditional bio-energyuseremains important.Concernsaboutenergysecurityandnationalpolicies supporttheuseofdomesticcoalandlimittradeinenergy.
SSP4inequality—aroaddivided
Final energy demand is moderately coupled to economic activity,whichresultsinlargedisparitiesinenergyconsumption becauseofslowincomeconvergence.Inpoorcountriestheuseof traditionalbio-energyremainsimportant.Technologicalimprove- mentsinconventionaloilandgasextractionarehigh,butpolicies arerestrictiveinhigh-incomecountriesbecauseoflocalpollution problems. There are significant technological improvements in nuclearpower.Investmentsareriskybecauseofgenerallyvolatile markets.
SSP5fossil-fueleddevelopment—takingthehighway
Energy demand growth is strongly coupled to economic growth,particularlyinthetransportationsectorduetomaterially intensivelifestyleswithastrongpreferenceforintensivematerial consumption patterns including high transportation demand.
Technological development in the fossil fuel sector, including CCSbasedmitigationtechnologies,israpidandsocialacceptanceis high.Non-biomassrenewables,however,aresubjecttolowsocial acceptance.
The five SSPs are similar to earlier scenario frameworks developed for various purposes (van Vuuren et al., 2012).The SRESscenariosproducedbytheIPCC,beingthepredecessorofthe SSPs,showsomesimilaritieswiththeSSPs(Vuurenand Carter, 2014).Therelationshiptothebroaderscenarioliterature isalso discussedbyO’Neilletal.(2017).
2.2.Baseline,climateforcingtargetsandsharedclimatepolicy assumptions
Bydesign,theSSPBaselinesinthisstudydonotaccountforany climatepoliciesthataimtoreduceemissionsandtheyalsodonot considerfeedbacksofclimatechangeimpactsontheeconomy,the energysectororthelandsystem(Mossetal.,2010).Thisapproach enablesthedevelopmentofreferencescenariosthatcanthenbe perturbedbymitigationpoliciesandfeedbacksfromtheclimate systemresultinginimpactsandadaptationmeasures.Implemen- tationoftheSSPnarrativesrequiredbaselinescenariostoconsider non-climatechangemitigationpoliciessuchasbio-fuelmandates thataffectenergypathways,butdonotvarywiththeimpositionof climatetargets(seeTableS12).
For climate change stabilization, the scenarios aim to obey specificradiativeforcinglevelssimilartothoseintheRCPs.Inthis paper, wefocusonamoderate target(similartoRCP4.5)anda stringenttarget(similartotheRCP2.6).Theformercasereaches 4.2W/m2by2100andincreasesto4.5W/m2inthelong-run.The lattercaseissimilartoRCP2.6thatreaches2.6W/m2in2100,but allowsforapeakanddeclineofforcingduringthiscentury.The policy implemented into the IAMs to achieve these targets is explicitGHGpricingthatisdeterminedbythelong-termtarget,as well as short-term ambition and regional participation. The explicit GHG prices can also be interpreted as comprehensive policy packages that vary in intensity such that the marginal abatementcostsimplicitlycorrespondwiththeGHGprice.
Near-termpoliciesaresubjecttoimplementationbarriersthat limittimingandregionalparticipationaswellasthetransitiontoa globally uniform carbon price. These Shared Climate Policy Assumptions(SPAs)areharmonizedacrossmodelingteamsand consistentlydefinedforeachSSP(Kriegleretal.,2014).Inthelong- run, GHGemissions fromallcountries, sources and sectors are pricedatauniformleveldeterminedbythestringencyofthelong- termforcingtarget.
ThebasicspecificationsoftheSPAsfortheenergysectorare included in Fig. 1. All SPAs assume moderate and regionally fragmentedcarbonpricingupto2020,reflectingtoalargeextent the Cancun pledges (Kriegler et al., 2015). There are three alternativetransitionsfromaregionallyfragmentedtoaglobally uniform GHG pricing regime to achieve the long term forcing target.InSSP1andSSP4withlowmitigationchallenges,uniform GHGemissionpricingisimplementedimmediatelyafter2020.In SSP2 and SSP5 with medium and high mitigation challenges, respectively, all countries transition from their moderate GHG pricelevelstotheglobalGHGpriceby2040asnecessaryinorderto reachtheforcingtarget.Finally,SSP3,anticipatinglittleinterna- tional cooperation and significant challenges to mitigation, assumesthat itis onlycountrieswithper-capitaincome above worldaveragethatwillstartthetransitionfrom2020until2040, whereastheotherregionsbeginafter2030andconvergetothe globaluniformcarbonpriceby2050.Thus,thelevel,shape,and regionalfragmentationoftheGHGpricingregimevarywithSSP andlongtermforcingtarget.
Theregionalresultsarepresentedinaggregatesoffivemega regions: (i) OECD, (ii) Reforming Economies (REF), (iii) Latin America(LAM),(iv)Asiaand(v)MiddleEastandAfrica(MAF).The relationshipbetweenaggregatesandnativemodelregionsisnot exactly thesame for all modelsand is reportedas partof the Supplementarymaterial.
3.Resultsofenergysectorpathways
Thetransformationoftheenergysectorisdrivenbythescale and structure of future final energy demand as described in Section3.1.Thisincludesregionalconvergencepatternsandthe modernizationoffinalenergyuse.Wethendiscussthecomple- mentary developments of the primary energy supply side (Section 3.2) with a particular focus on the fossil fuel sector.
Section3.3providesamoredetailedanalysisofelectricitysector pathways. Finally,wediscuss theenergy-relatedCO2 emissions from fossil fuel combustion and industry resulting from the interplayofenergysupplyanddemand(Section3.4).
3.1.Finalenergydemand 3.1.1.Totalfinalenergygrowth
Final energy demand is linked to the fundamental socio- economicdriversofpopulationdevelopment,economicgrowth, technologicalchangeandlifestyles.Historically,globalfinalenergy
demandfrom1980to2010grewannuallyonaverageby1.6%;for thedecade2000–2010itwasevenfaster(2%/yr),IEA(2012).
The SSP marker scenarios cover a range between 600 and 1200EJ/yrandfeatureverydifferenttimeprofiles.Theymostlylie, however,withintherangeofbaselineAR5scenarios(Fig.2).The SSP2baselinecontinuesa similartrajectorytohistoricalgrowth rates(1.4%/yruntil2050).ThisissimilarinSSP3andSSP4,which showa decelerating growth in the second half of the century mainlybecauseeconomicgrowthslowsdown.Thehigheconomic growthandmaterialintensivelifestyleintheSSP5scenarioleads tohighfinal energy growthrates (2.1%/yr until 2050) that are similartothehighgrowthphasebetween2000and2010.TheSSP1 decouples economic growth from final energy demand which leadstoapeakin2070.
A major findingof themitigationcasesisthat the2.6W/m2 targetisnotsolvableforSSP3byanymodelbecauseofweaknear- termpolicyambitionandinsufficientemissionmitigationresult- ingfromslowtechnologicalprogress.Fortheothercasesunder climatepolicies,thedemandreductionsaregreaterforSSPswith significantmitigationchallengesandtighterstabilizationtargets.
Incomparisonwiththeunmitigatedcases,thelevelandtherange offinalenergydemandacrosstheSSPsIslower.Inscenarioswith fastgrowthinfinalenergyuse,forinstanceSSP5,demandismore significantly reduced than in slow energy demand growth scenariossuchasSSP1.Finalenergyconsumption,however,does notreducebelow thelevel alreadyachieved in2010 ascanbe observedintheSSP1markercase.
3.1.2.Regionaltrendsinfinalenergydemand
The development of final energy demand is strongly correlatedwith economic growthand thereforeto patterns of income convergence. However, the degree of coupling is uncertain (e.g. Csereklyei and Stern, 2015). The SSPs span a broad range of scenarios for economic growth and regional income convergence (Dellink et al., 2017) as well as very differentpatternsofcouplingtofinalenergydemand.Thisleads to very diverse regional convergence patterns of per-capita incomeandper-capitafinalenergydemand.Moreover,basedon observed data, annual per-capita final energy consumption below30GJ/capitaiscorrelatedwithlowlevelsofdevelopment, whereasobservations around 100GJ/capitaare correlated with veryhigh levelsof development (LambandRao,2015; Steckel et al., 2013 andliterature therein).In future,the efficiency of
generatinghumandevelopmentfromfinalenergycouldincrease throughtechnologicalimprovements.
Panel(A)ofFig.3depictsper-capitafinalenergyconsumption against per-capita income ona double log-scaleof themarker baselinescenariosforthefivemacroregions.Panel(B)compares finalenergyper-capitausefordevelopedanddevelopingregionsin thebaselineandmitigationcasesforthemarkerandalsodepicts thecrossmodelranges.InSSP2thehistoriccouplingbetweenGDP andenergyisongoing,althoughGDPgrowssomewhatfasterthan energy.InOECDcountriestheper-capitaincometriplesby2100, whileper-capitafinalenergyconsumptionincreasesfrom140to 170GJ. Developing and emerging economies follow less energy intensive developmentpathways,while thecoupling withGDP growthisstrongerthanindevelopedregions.Theconvergencein energyuseremainsincomplete.
In SSP1 and SSP5 global GDP growth is stronger and convergence faster than in SSP2, which also leads to faster convergenceof energy useacrossregions,but atvery different levels.InSSP1globalconsumptionpatternsandlifestylesquickly shift to less material intensive modes, while more efficient technologiesdiffusequicklyandenergydemanddecouplesfrom economic growth for annual per-capita incomes beyond US
$30,000measuredinpurchasingpowerparity(PPP).Developing andemerging economiesfollow lessenergyintensive pathways because they are leapfrogging inefficient end-use technologies and, hence, human development is achieved more efficiently.
OECD countries show decreasingenergy use as more efficient technologies replace obsolete equipment. In SSP5 economic development is quickest; the coupling with energy demand is strongand rapideconomic convergencecoincides withconver- genceinenergyuseleadingtoaquadruplingindevelopingand emerging economies. The preference for energy intensive con- sumptionpatternsandlowenergypricesleadtostrongincome- drivenenergy-demandgrowth.
InSSP3andSSP4globaleconomicgrowthisweakandincome convergenceslow.Consequently,thelong-termdisparityinfinal energyuseisgreaterthaninSSP2withenergyusebeinghigherin developed countries and lower in emerging and developing countries.
The regional energy demand patterns change with the imposition of policies designed to achieve climate change stabilization (see panel (B)). In all SSPs the coupling between economicandenergydemandgrowthbecomesweakerandsome Fig.2. GlobalfinalenergypathwaysforthemarkerSSPsanddifferentclimateforcingtargets.HistoricaldatafromIEA(2012).ThegreyshadedareasshowtherangeoftheAR5 database(rangesofthe1/99-percentileinlightgreyand5/95-percentileindarkgrey);thethickdashedgreylineisthemedian.AllAR5scenarioswithoutclimatepolicies wereusedforthebaselinerange;thescenariosfromthecategoriesIV&VandcategoryIwereusedforthe4.5W/m2and2.6W/m2targets,respectively(ascategorizedinthe IPCCAR5WG3AnnexII.10.3).Thebarsontherighthandsideofthepanelsdepictthe2100rangesofallsixSSPmodelsforeachSSP.TableS13summarizestheresults.
scenariosevenfeaturedecoupling.InSSP2thedecouplingresults indecreasingfinalenergyper-capitauseintheOECD,ifthestrong stabilizationtargetisachieved.IntheSSP3scenariothemoderate stabilization target leads tofull decouplingin the low-income regions, and theper-capita energy demandin MAF remains at 30GJ/capita.Thisleadstoanincreaseintherelativeenergygap betweenhigh-andlow-incomecountries.InSSP5thecouplingis alsodampenedinresponsetoclimatepolicy,butafulldecoupling isnotachievedinanyregion.
3.1.3.Finalenergymix
Theshareofelectricityintheglobalfinalenergymixincreased from11%in1980to18%in2010.Thiscontrastswithadecreasein theshareofsolidsfrom14to10%andforliquidsfrom45%to41%
overthesametimehorizon(IEA,2012).Thefuturedevelopmentof thefinalenergymixacrossSSPsisshowninFig.4.
TheSSP2scenariofeaturesamoderatemodernizationoffinal energyuse.Theuseofliquidsincreasesbytwothirdsupto2050 and remains roughly constant thereafter. The picture is mixed Fig.3.Developmentofregionalfinalenergyusepercapita.ThetoppanelshowsthedevelopmentagainstpercapitaGDP(PPP)forthefivemacroregionsacrossSSPmarker baselines.Thevariationsin2010valuesbetweentheregionsareduetodifferentdefinitionsofnativemodelregionsbythemodelingteams.Thebottompanelshowstheper- capitafinalenergyuseinthedeveloping(MAF,ASIA,LAM)andthedevelopedregions(OECD,REF).TheverticallinesrepresentmodelrangesforeachSSP;thecoloredboxes indicaterangesacrossallSSPs.
whenlookingattraditionalandmodernenergycarriers.Onthe onehand,electricityconsumptionmorethandoublesfrom2010to 2050,andquintuplesby2100.Ontheotherhandthedirectuseof coaldoubles(triples)by2050(2100)tofuelindustrialdevelop- mentinAsiaandMAF.Intheclimatechangemitigationscenarios, energy sector modernization accelerates with higher shares of
electricity(incl.electrificationoftransport)andafasterphaseout ofsolidenergycarriers.
SSP1 and SSP5 show similarities in the trends in energy modernization,althoughthescaleoftotalfinalenergyconsump- tion is very different. Electrification is rapid, particularly in developing countries. Demand for gaseous fuels grows Fig.4.Panel(A)showsglobalfinalenergymixesacrossSSPsandclimatestabilizationcases.Electricityaccountsfortheconsumptionoffinalelectricityconsumersanddoes notincludelossesfortransmissionanddistribution.HistoricdataistakenfromIEA(2012).PleasenotethattheSSP4markerteam(GCAM)appliesadifferentaggregationof IEAenergybalancedatabetweenenergytransformationsectorsandend-usesectors,tothatofothermodelingteams.Consequently,thehistoricaldataisdifferent, particularlywithrespecttoindustrialfinalenergy.Panel(B)showstheelectricitysharesinDevelopedcountriesandEmergingandDevelopingcountries.Theboxesindicate thefullrangeacrossSSPsandmodelsfortheseregions.
substantiallyupuntil2040inbothscenarios, onlybeginningto deviate from 2040 onwards. The most remarkable difference, however,isinthetransportationsectorwherethereisdifferent growthofliquidfueldemand.Thisreflectsalternativepathwaysof futuremobilitywithastrongerfocusonatransformationtowards publictransportandelectricorhydrogencarsinSSP1comparedto a more conventional transport system with high demand for transportationservicesinSSP5.Thisisimportantforthemitigation challenge.InanSSP5worldthedecarbonizationofthetranspor- tation sector is the main bottleneck, which is addressed by decreasingdemandandincreasingtheuseofelectricandhydrogen vehicles and bio-fuels. In contrast, the low transport energy demand in the SSP1 baseline eases the mitigation challenge significantly and the necessary changes for achieving climate changestabilizationremainrelativelysmall(see Figs.S14–S15).
Moreover, with increasing stringency of mitigation policies electricity demand decreases in SSP1, whereas it increases in thelong-runinSSP5(seealsoFig.S12).
Thetwoscenarioswithslowgrowthandconvergence(SSP3and SSP4)featureslowermodernizationintheglobalfinalenergymix.
The electrification in developing regions is slow and does not catch-up with that of developed regions. Additionally, solids continue toplaya prominentrole inenergy use. Thereislittle accelerationof the slow modernization of SSP3 in the climate changemitigationcasesbecausethedevelopmentoftechnologyis stagnant.To achievetheclimatechangestabilizationtargets,in SSP3non-electricenergydemandisreduced;electricitydemandis only reduced in Asia and theMAF region.In contrast,there is strongerelectrificationthroughoutthecenturyinSSP4,duetothe strongertechnologydevelopmentintheend-usesector.Thishelps reduce non-electric energy use in climate change stabilization scenarios.However,thereisnomodernizationin SSP4forlarge partsofthepopulationinpooreconomiesinAsiaand theMAF region.Thesecountriescontinuetorelyontraditionalbiomassuse.
3.2.Primaryenergysupply 3.2.1.Primaryenergymix
From1980–2010,globalprimaryenergysupplygrewfrom300 to510EJ/yr.Fossilfuelsdominate,supplyingaround85%oftotal primaryenergy.Overthesameperiodtheincreaseinthesharesof natural gas (17%–22%) and coal (25%–29%) have been at the expense of oil (44% to 34%). The increase in coal use was concentrated in Asia, whereas natural gas has increased more evenly around the world. Oil has become more important in developing and emerging economies. Bioenergy (mostly tradi- tionalbio-energy)hasremainedstableataround10% whilethe contributionofnon-biomassrenewableshasdeclined(IEA,2012).
In Fig. 5 the SSP2 baseline scenario projects a substantial growthinprimaryenergyusewiththedominationoffossilfuels, whereasrenewables,suchaswindandsolar,increaseonlyslightly.
Oilsupplypeaksin2050,andgrowsagainattheendofthecentury withexpandingnon-conventionaloilproduction.Coalandnatural gasincreasecontinuouslythroughoutthecenturyandshow50%
and125%higherproductionlevels,respectively,from2010to2050.
Also,intheSSP3andSSP5baselines,fossilfuelsdominateprimary energy supply. In SSP5 this is represented byremarkably high sharesofmodernandcleannaturalgas,whereasconventionaland dirtiercoalexpandssignificantlyinSSP3.Thesmallchallengesto mitigationareassociated withdecreasingsharesof fossilfuels, which even peak around mid-century in the baseline, with renewablesexpandinginSSP1andSSP4alsorelyingtoasignificant degree on nuclear power. In both cases bio-energy plays a significantrole,butinSSP1 itisusedinmodernways,whereas in SSP4 it is used in traditional modes as a result of income inequalityandfailureofenergyaccesspolicies.
In the stabilization cases primary energy consumption is reduced andfossil fuelspeakbeforemid-century.By2050the fossil fuels share is, however, still significant. The most significantreductionisintheuseofcoal.Itsuseincombination withCCSishigherinthemoderatestabilization cases.Natural gasstillincreasesinstabilizationscenariosandthecombination withCCSbecomesmoresignificantfortheachievementof the 2.6W/m2target.FossilfuelswithCCSareimportantintheSSP2 and SSP5 scenario, but do not play a prominent role in the SSP1scenario. This isbecausepolicies prioritizesustainability and corresponding technological developments. The faster diffusion ofrenewable energytechnologiesis also determined by the extended SSP1 narratives regarding political and technological factors.The high shareof renewablesin SSP5is due to limitations in nuclear as well as high oil and gas consumptionthatcannotbecombinedwithCCS,forexamplein thetransportationsector.
Bio-energyisakeyoptionformitigationintheenergysector.In SSP1theneedforthecombinationwithCCSismoderate,whereas inSSP5thedemandforbiofuelsaswellasforcarbonoffsetsishigh due tohighenergy demandand theabundanceof cheapfossil fuels. Thedemandfor bio-energyin SSP5growsto480EJ/yrby 2100.TheSSP2showslessdeploymentofbio-energywithCCS,but inthisscenariocarbonoffsetsarealsogeneratedbyafforestation, which is notavailable in SSP5marker becausepoliciessupport engineeringbasedsolutions.Thedemandforagriculturalcropland decreasesafterglobalpopulationhaspeakedinSSP5(seePoppet al. (2017) for details). There is also significant bio-energy deployment in SSP4, including atmospheric carbon dioxide removalusing BECCS,becauseofstrongtechnological improve- ments developed by the innovative global elite. In SSP3 the potential for emission reductionsaswellas thedeployment of carbondioxideremovaltechnologiesisstronglylimitedbecauseof slowtechnologicaldevelopmentandhighlanddemandsfroma growing population. This meansthe2.6W/m2 targetcannot be achieved.Itisworthmentioningthatbio-energyincombination with CCS is mostly used to produce liquid fuel rather than electricity(Fig.S15),whichreconfirmsearlierfindings(Roseetal., 2013).
3.2.2.Fossilfueluse
Fig.6showsthecumulativefossilfuelextractionoverthe21st centuryandcomparesitwithreservesasreportedbyRogneretal.
(2012).ForcoalusetheintuitiverankingoftheSSPbaselinesisin accordancewiththechallengetomitigation.Itisworthnotingthat SSP3projectsveryhighcoalextractioninAsia,butmuchlesssoin theOECDandReformingEconomiescomparedwiththeSSP5.The energysecurityconcernsassumedinSSP3establishasignificant limit on trade between regions (Note: very high global coal extractioninSSP3isprojectedbyonlyonemodel,whichassumes relatively free energy trade). By contrast, in SSP5 the world economyismoreglobalizedandtradeismoreintegrated,which leadstosignificantexportsfromcoalrichregionsintheOECDand theReformingEconomiestofueldevelopmentinrapidlygrowing economies.InSSP1,however,cumulativecoaluseislessthanthe reserve that is considered available today. In the stabilization scenario,ofalltheSSPs,alargeportionofthecoalreservesarenot utilized,eveninthemoderate4.5W/m2case.
OilandgasarenotequallyrankedacrossSSPbaselinesmainly becauseofdifferencesinavailabilityandtradeoffossilfuelsacross SSPnarratives.SSP5hasthehighestconsumptionofgasbecauseit is (i) relatively clean, (ii) technological improvements increase supply, (iii) globalized markets allow for trade, and (iv) social acceptance for gas related infrastructure is high. In SSP3 gas extraction is low because technological progress and demand growthareslowandtradeissubjecttoenergysecurityconcerns.In
SSP1gasextractionisrelativelyhigh,becausecleangasreplaces relativelydirtycoalinthebaseline.Inthestabilizationcasesofall SSPs, gas consumption remains significant and exceeds the conventionalreserveestimate.
OilextractionintheSSPbaselinesexceedscurrentestimatesof conventional and unconventional reserves and also opens resources.Again,SSP3 rankslow becauseofslow technological
progress.SSP2isclosetoSSP5becauseSSP2featureslesscoal-to- liquidsproduction(seeSupplementarymaterial).Inthe4.5W/m2 mitigationcaseallscenarios,exceptSSP1,resultincumulativeoil consumption thatexceedscurrent reserveestimatesofconven- tionalandnon-conventionaloil.ForSSP5,thisevenholdsinthe 2.6W/m2case.Thesensitivityofoilextractiontoclimatechange mitigationpoliciesissmallerthanforgasandcoal.Itisparticularly Fig.5. Panel(A)showsglobalprimaryenergymixacrossSSPsandclimatepolicycases.Accountingforprimaryenergyfollowsthedirectequivalenceapproach.Historicaldata istakenfromIEA(2012).Panel(B)showsthefossilfuelshares.ThecoloredboxesindicatetherangesacrossallSSPsandmodelsforthetworegions.
Fig.6.Cumulativefossilfuelextractionbyregion.Thesefiguresincludeown-consumptionforoperatingextractionactivities.Thegreyhorizontallinesindicatetheglobal rangesacrossmodels;whitesquaresrepresentglobalvaluesofmarkermodelsandgreydotsrepresentnon-markermodels.Forthepurposeoforientationthedottedlines depictthelevelofconventionalreserves(provenandeconomicallyrecoverable)whereasthedashedlinesadditionallyincludeunconventionalrecoverablereservesforoil andgas(Rogneretal.,2012).Wedonotshowresourcesbecausereportedfiguresareoriginal-in-placequantities,whichrequireadditionalassumptionsonrecoveryfactors.
Thegreyverticallinesindicatetherangeacrossmodels.ThewhitesquareindicatestheSSPmarkermodel,whereastheblackdotsindicatethenon-markermodels.Note:the regionalallocationpatternisdifferentfornon-markermodelsthanformarkermodels.
Fig.7.GlobalpowergenerationbytechnologydifferentiatedbySSPsandpolicyscenario.Historicaldata1980–2010istakenfromIEA(2012).
smallfortheMAFregion,whichisendowedwithmostofthelow costoil(McGladeandEkins,2015).
3.3.Electricitysector
Theelectricitysectoriscriticalformitigatingclimatechange becausealthoughitsgenerationcausessignificantCO2emissions, thelargestnumberofdecarbonizationoptionsareavailableinthis sector(Bruckneretal.,2014).Italsohasthepotentialtodisplace fossilfuelconsumptioninothersectorsthroughelectrificationof energy end uses. The electricity sector is important for the challengestoadaptationbecausethermalgenerationandhydro- electriccapacities served nearly98% ofglobal powersupply in 2010; their sensitivity to ambient temperatures and water availabilitymakesthemvulnerabletoglobalwarming.
Inbaselinescenarios,themainuncertaintythroughthemiddle ofthecenturyrelatestotheoverallsizeoftheelectricitysectorand thedominantgenerationtechnologiesinthemix(Fig.7,seealso Fig.S12).SSP2projectsashifttowardsgas(80EJ/yrin2050)thatis alsofeaturedinSSP5(160EJ/yrin2050),butinafastergrowing electricitymarkettriggeredbyabundantgasavailability.Towards theend ofthecenturySSP5shifts tocoal(150EJ/yr)andnon- bioenergy renewables (180EJ/yr). Nuclear power is much less widelyused due to its unfavorableeconomics and issues with public acceptance. SSP3 projects electricity sector growth that mainlydepends oncoalfiredpowerstations(170EJ/yrin 2100) fromdomesticallyproducedcoal.Theshifttowardsgasislimited inSSP1becauserenewabletechnologies,suchaswindandsolar, improvequicklyandaresociallymoreacceptableSSP4issimilar buttheroleofnuclearpowerismuchmoreprominent(65EJ/yrin 2100), which reflects the differences with SSP1 in social acceptability and technological development. SSP1 and SSP4 baselinesfeaturesignificantlyincreasingsharesofnon-fossilbased powergenerationtechnologies,which reducesthe challengeto mitigation.
Thepowersectorreactsstronglyinscaleandstructureinthe stabilizationcases.AcrossallSSPs,theverysignificantreductionof coalfiredpowergenerationisarobustfeature.Besidesthis,the samestabilizationtargettheSSPsdifferverymuch inscaleand structure of the power sector. The use of CCS only slightly counteracts the first-order phase-out of coal from the power sector.TheCCSoptionismorerelevantforgasfiredpowerstations,
particularlyinthemoderatestabilizationcaseofSSP5.Thelarge deploymentofnuclearpowerinSSP2andSSP4(upto150EJ/yr)is possiblebecauseitisassumedthereishighsocialacceptancefor this technology. Limited social acceptance,however, dampens theexpansionof nuclearpowerin SSP1andalso inSSP5.The shareofelectricityfrombio-energywithCCSissmallinallSSPs, due to a combination of low conversion efficiency and the demandforbioenergytoproduceliquidsand/orhydrogen,which can also becombined withCCS (seeFig. S15). Thelarge-scale deployment of non-bioenergy renewables in the stringent stabilization caseof SSP5is duetotheextremelyhigh carbon prices thatexceed US$300/tCO2 post-2050. This leads to high costsfortheresidualemissionsfromfossilfuelswithCCS.The high shares of wind and solar lead to very high electricity prices,becausetheintegrationofthesevariablesourcesrequires substantialenergystorage.
3.4.Energysectoremissions
ThissectionfocusesonenergysectorCO2emissionsfromfossil fuelsandindustry;emissionsofCH4andF-gasesarediscussedin theSupplementarymaterial(seeFig.S10–S11).Thechallengeto mitigation is influenced by the cumulative residual emissions allocatedtotheenergysector.
CO2emissionsfromthecombustionoffossilfuelsaccountfora dominantshareof pastglobalanthropogenicgreenhousegases.
TheIPCCreportsthatcumulativeglobalCO2emissionsfromfossil fuel combustion and industry from 1750 to 2010 amount to 1350GtCO2,ofwhichcoalaccountedfor650GtCO2,oil470GtCO2
andnaturalgas180GtCO2(Blancoetal.,2014).Theannualaverage growth rate was 1.7%/yr between 1970 and 2010 accelerating duringthelastdecadeto2.2%/yr,basedonvanVuurenetal.(2011).
The SSP baseline scenarios span a broad range of possible emissionfutures(Fig.8)reflectingthelargeunderlyingdifferences inthedevelopmentoffinalenergydemandandprimaryenergy supplyacrossSSPs.TheSSPbaselinescovertheuncertaintyrange ofIPCCAR5baselineemissions.Therankingofemissions(incl.
markerandcrossmodelranges)isconsistentwiththemitigation challengesfortheSSPs.SSP2beginswithmoderategrowthrates (1.2%/yr for the period 2010–50) which accelerates during the secondhalfofthecenturyasmorecoalisusedtofueleconomic development.ThehighmitigationchallengeinSSP5withhighfinal
Fig.8.GlobalCO2emissionsfromfossilfuelsandindustryforbaselineandstabilizationat4.5W/m2and2.6W/m2.ThethinlinesshowtheRCPscenarios(vanVuurenetal., 2011).ThegreyshadedareasshowtherangeoftheAR5database(rangesofthe1/99-percentileinlightgreyand5/95-percentileindarkgrey);thethickgreydashedlineisthe median.AllAR5scenarioswithoutclimatepolicieswereusedforthebaselinerange;thescenariosfromthecategoriesIV&VandcategoryIwereusedforthe4.5W/m2and 2.6W/m2targets,respectively(ascategorizedintheIPCCAR5WG3AnnexII.10.3).Thebarsontherighthandsideofthepanelsdepictthe2100rangesofallSSPmodelsfor eachSSP.
energydemandandabundantfossilfuelsupplycorrespondswith high growth rates (2.3%/yr up to 2050) exceeding the RCP8.5 scenario. The SSP3 scenario is also subject to a substantial mitigationchallengebut sloweconomic growthimplies slower emissiongrowth.ThelowmitigationchallengeinSSP1andSSP4is associatedwithpeakingbaseline emissionsresultingfromslow energydemandgrowthandagradualshiftawayfromfossilfuels.
Stabilizinglong-termclimatechangeatacommontargetlevel acrossSSPsnarrowstherangeofemissionpathwaysconsiderably, but there are still remarkable differences; for the stabilization casesthevariousSSPsclusteraroundtheoriginalRCPs,but the rangein emissions isnotable. Themoderate forcing targetstill allowsforconsiderableemissionstowardstheendofthecentury.
The stringent 2.6W/m2 target requires early peaking and net negativeemissionsinallSSPs.FortheSSP3scenariothistarget level is not achievable because of the little near-term policy ambitionassumedinthecorrespondingSPA3,thesmalltechno- logicalcapacitytoreduceoroffsetCO2emissionsandthehighGHG emissionsfromtheland-usesectorduetopopulationgrowth.SSP5 shows a highand late emissions peak due to the difficultyof lockingabundantfossilfuelsoutofthesystemandaconsiderable volumeofnetnegativeemissionsbytheend of thecentury. In contrast,thesmallmitigationchallengeinSSP1correspondstoa relativelysmoothemissionprofile.
GHGemissionsfromtheland-usesectorareverydifficultto reduceinthestabilizationrunsofSSP3andSSP4(seePoppetal., 2017). Therefore, stronger CO2 emission reductions from fossil fuelsandindustryarerequired.Moreover,SSP5doesnotallowfor GHG emission off-sets from afforestation, in contrast with the otherSSPs.Thisputs a largermitigationburden ontheenergy sector.
4.Summary,discussionandfutureresearch
ThisstudydescribestheenergysectorpathwaysoftheSSPs focussingonthemarkerscenariosforeachSSP,whichillustrates theimplementationofthevaryingchallengesofclimatechange mitigationandadaptation.TheSSPimplementationwasbasedon quantitative assumptions of population and GDP as well as interpretations of thetechnology, lifestyle and policy elements of the SSP narratives. The SSP quantification applied detailed energy system models that are fully integrated with land-use modelsandmodelsofthemacroeconomy.Theyfullyrepresentall GHGandair pollutantemissions,andtheinterrelationshipwith bio-energymarketsthatareincompetitionwithotherecosystem services incl. food markets. The SSPs address common energy sectorchallenges in differentways and resulting energy sector developmentsspanabroadrangeofpossiblefuturesattheglobal level(seeTable2).Thesealsotakeaccountofregionaldifferences.
Theimplementationof SSPsinto IAMsdelivered scenarios that wellreflecttheSSPnarrativesandlocatethesetofSSPscenariosin thespaceofclimatechangemitigationandadaptationchallenges.
ComparedwiththebundleofscenariosusedinIPCCAR5theSSPs spanasimilarrange,asaresultofdifferencesoriginatingintheSSP narratives and corresponding implementations in the IAMs.
Consequently,thequantitativepathwaysareconsistentwiththe challengestomitigationandadaptation.ThisisreflectedintheSSP baselinescenarios(e.g.CO2emissions,finalenergydemand,fossil fuelrelianceetc.)aswellasbeingameansofreducingemissionsin thepolicyscenarios(e.g.demandreductions,decarbonizationof energysupplyetc.).
TheSSP2baselinedescribesamiddle-of-the-roadscenariowith mediumchallengestomitigationandadaptation.SSP2relieson mediumassumptionsforkeyinputparameterssuchaspopulation dynamicsandeconomicgrowth. TheimplementationintoIAMs projectsascenariothatcontinueshistoric trendsobservedover
recent decadesincluding thedominanceof fossilfuels, conver- genceofper-capitaenergyconsumption,gradualmodernizationof energyuseandgreaterenergyaccessandthereforeincreasingGHG emissions.
Challengestomitigationmainlydifferintermsofconsumption patterns,technologicalchange,fossilfuelavailabilityandefficien- cyimprovementsthatleadtothehighest(SSP5)andlowest(SSP1) emissions in the un-mitigated baseline cases. SSP1 assumes decoupling of economic growth and energy demand that is achieved by increasing energy efficiency and increased use of renewables.Alternatively,SSP5assumesstrongcouplingbetween GDPandenergydemandthatissupportedbyabundantfossilfuel supplies.TheglobaldecouplingofGDPgrowthandenergydemand assumed in SSP1 hasnot beenobserved (Csereklyei and Stern, 2015),butenergyefficiencyanddemandreductionpotentialsare considerable(e.g.Simsetal.,2014fortransport).Thetechnological lock-incontinuesinSSP5(Unruh,2000),butmobilizationoffossil fuelsisunprecedented(Aguileraetal.,2009;Baueretal.,2016).
Thedifferentenergysectordevelopments,combinedwithland- useemissionsresultinradiativeforcinginSSP5exceeding8.5W/
m2in2100,whereasitincreasesto5W/m2inSSP1.Consequently, mitigationpoliciesaimed atforcing levelsof 4.5W/m2 or even 2.6W/m2differinstrengthandimplyverydifferentchangesinthe scaleandstructureofenergysupplyanddemand,particularlyin thepowerandthetransportsectors.
Thetwoscenarioswithhighadaptationchallenges(SSP3and SSP4)initiallydiffer fromSSP1and SSP5withrespecttosocio- economicdrivers.Highadaptationchallengesareconsistentwith slowincomeconvergenceaswellasslowtechnologicalchange(in SSP3)anddiffusion(inSSP4).InSSP4thebusinesselitedevelops advancedtechnologiesintheenergysector,butbroaderdiffusion is slow and energy access is a pressing, yet unresolved, issue (Pachauri et al., 2013). The technological progress in SSP3, however,isgenerallyslowandenergysecurityisofgreatconcern in a worldof political fragmentation (Jewell etal., 2014).Slow regionalincomeconvergencetranslatesintoslowconvergencein per-capita final energy demand and slower modernization of energyuse.ThegrowthoftotalenergydemandandCO2emissions islessthan(SSP4)orsimilar(SSP3)toSSP2.Akeyresultisthat SSP3,despitehighpopulationgrowthandslowenergyintensity improvements,doesnotgenerateanincreaseinradiativeforcing to8.5W/m2until2100,becauseeconomicgrowthistooslowand energysecurityconcernslimitthetradabilityand,consequently, theuseofcoal.ThehighmitigationchallengeinSSP3however,is Table2
SummarystatisticsofbaselineSSPsatthegloballevelfortheyear2100indexedto 2010=100,exceptforforcingandfossilshare,whichisgivenin%oftotalprimary energyconsumption.
Challengetomitigation
Small Medium High
SSP1 SSP4 SSP2 SSP3 SSP5
Forcing[W/m2] 5.0 6.4 6.5 7.2 8.7
CO2Emissions 84 136 262 253 396
Kayafactors Population 101 135 132 183 107
Per-capitaIncome 821 390 607 227 1426
EnergyIntensity 17 34 34 52 21
CarbonIntensity 59 76 97 117 121
Other Indicators
FossilShare[%] 55 70 77 83 84
Electricity 392 313 515 349 654
Transport 143 228 275 218 450
Solidsc 63a 89 191b 217 16
aIncludelargeshareofmodernsolidbio-energyuseinindustryandhouseholds.
bIncludeslargeshareofdirectcoaluseinindustry.
cIncludesalsodirectuseofcoalintheindustrysector.
reflected in slow technological improvements of mitigation options and small near-term climate policy ambition, which makesthelong-termforcingtargetof2.6W/m2unachievable.This resultisrobustacrossallmodelsthataddressedtheSSP3scenario.
Eventhoughthenear-termpolicyambitionisstronger,onlytwo models(AIM/CGEandMESSAGE-GLOBIOM)couldfindasolution forthe2.6W/m2targetunderSSP3conditions.
OnthetechnologyleveltheSSPscoverabroadrangeofvastly different pathways that are subject to many uncertainties.
Comparedwiththeexisting scenarioliterature threepointsare worthhighlighting.First, theuseofbio-energyin the2.6W/m2 scenariovaries acrossSSPstoa greaterdegreethan intheAR5 database,becauseSSP5combinesveryhighenergydemandwith veryhighyield increasesin bio-energy supply.Moreover,as in previousstudies(Roseetal.,2013)theallocationofbio-energyis dominatedbyliquidfuelproductionratherthanpowergeneration.
Theoptiontogenerateelectricityfrombio-energyincombination with CCS is also applied. This contributes significantly to the carbondioxideremovalbecauseofthehighercapturerateofCO2
comparedwiththefuelalternative.Theshareofbio-energyinthe electricitymix,however,remainssmall.Thisfindingissubjectto techno-economicuncertaintiesofthesepre-commercialtechnol- ogies.Second,electrificationhasbeenidentifiedasanimportant component in mitigation pathways. Krey et al. (2014) and Sugiyama (2014) found electricity shares increased in stricter stabilizationcases,whereasabsoluteincreasesinelectricitywere onlyfoundinthelongerrun,infewmodels(Edmondsetal.,2006).
TheSSPsdepictamorediversepicture.SSP2,SSP4andSSP5show increasingshares ofelectricityand, inthelong-run,alsohigher absolute electricity production. SSP1, however, demonstrates decreasingshares ofelectricity in developed countriesasmore stringent stabilization targets are achieved. Finally, the use of nuclearenergyisgenerallylessthaninthehighendscenariosof SRESandAR5scenarios,whichismostlyduetothespecificSSP narratives.SSP3andSSP4arecandidatesforhighnuclearpower, buttheenergy demandgrowthisrelatively small,whereasthe SSP5withhighdemandassumesless social acceptancefor this technology.
TheSSPsserveasaframeworkforsystematicfutureresearchof climatechangemitigation,climateimpactsandadaptationaswell asbroadersustainabilityissuesaimingtointegratestudiesfroma greatdiversityofresearchfields.TheSSPsarenowfullyoperational andRiahietal.(2017)providesageneraldiscussionintotheiruse.
The energy sector SSPs are useful for future research in the followingfourdirections.
First, the SSPsdiffer strongly with respect toenergy sector challenges, such as technology diffusion and energy sector modernization,that aretightlyinterlinked withclimatechange mitigationandadaptationchallenges.TheSSPshelpresearchersto guideandclassifytheirscenarioswithinabroaderframeworkof the challenges space, which helps to communicate results, comparethemwithotherstudiesandclassifytheiruncertainties (Trutnevyteetal.,2016).Also,assessmentsofnationalandsectoral energysystems can benefitfrom guidanceand classification of assumptions on, for example, energy demand development or fossilfuelavailabilityandtrade.Regionalandsectoralextensionsof theSSPscouldbeformulatedtodeepenthescenarioframework.
Coordinationof research in this wayhelps to link global with regional, national and sectorial studies to improve mutual information flow and synthesis of various studies. Analysis of mitigationcouldalso beenriched by buildingbridgesto social sciences focusing on technology transformations (Geels et al., 2016).Moreover,theSSPscanserveasastartingpointtodiscuss climateengineeringoptionssuchassolarradiationmanagementin vastlydifferentcontexts.
Second,therobustnessofpoliciescanbetestedinvarioussocio- economic contexts. The climate change stabilization scenarios assumedacombinationofshort-termsecondbestandlong-term first best climate policies given the SPAs and the long-term stabilization targets. The long-term uniform carbon price is a highlyidealizedpolicyimplementationwithverystronginstitu- tional requirements. It induces relativelysynchronous transfor- mationdynamics,whichareindicatedbythesmalldifferencesin thefossilfuelsharesbetweenregionsshowninFig.5(B).Tobetter understand secondbest policies alternativeproposals could be implementedintovariousSSPs,whichwouldimplyverydifferent energysectorandmarketdynamicsacrossregions(Burkeetal., 2016).
Third,theSSPspresentedherearedesignedasreferencecases thatdonot–bydefinition–considertheimpactsofclimatechange on socio-economic development including the energy sector (Kriegleretal.,2012;Mossetal.,2010).ThecombinationofSSPs andRCPsareessentialpartsofabroaderresearchframeworkfor theassessmentofadaptationchallengesbecausephysicalimpacts derivedincollaborationwithclimatemodelingteams(Eyringetal., 2015)aresuperimposedonsocio-economicdevelopments.Future studiesonclimatechangeimpacts,adaptationandvulnerability,in which theenergy sectoris relevant,canderivedifferentsetsof consistentassumptionsabouttotalenergydemandandfuelmix fromtheSSPs.Forexample,ifacountrystudyisinterestedinthe vulnerabilityoftheelectricitysectortoclimatechange,theSSPs canguide thechoiceofassumptions aboutgenerationmixand electricity demand. This can also be donefor mitigation cases consideringchangesinthescaleandstructureof theelectricity sector combined with the changes in climate variables that correspondto radiative forcing levels. Evaluatingthe effects of climatechange correspondingto, forexample, 4.5W/m2 under differentpowersectorconfigurations(seemiddlerowofFig.7) establishesthelinkbetweenSSPsandRCPsinstudiesonimpacts, adaptationandvulnerabilitystudies.
Fourth,studiesonbroadersocialandenvironmental sustain- abilityissues canalsobeguidedbytheenergy sectorSSPs.For example,theuseofmaterialsandlandareimportantdriversof global and regional environmental change that are partly determinedbyenergysectordevelopmentsand partlybyother socio-economicandtechnologicaldrivers.Similarly,researchinto energyaccess,airpollutionandenergysecuritycangreatlybenefit fromenergysectorSSPs(Jewelletal.,2014;Pachaurietal.,2013).
TheenergysectorSSPspresentedhereaimtoprovidereference cases forfuture integration, deepening and expandingresearch into energy transformation pathways. They constitute a major milestonethatlinktheSSPnarrativesanddifferentlevelsofforcing stabilization as described by the RCPs with the quantitative developmentsoftheenergysector.Assuchtheycanserveasa basisformoreintegrativeassessmentsinthefuture.
Acknowledgements
N.B. and J.H. were supported by funding from the German Federal Ministryof Education and Research (BMBF) in theCall
“EconomicsofClimateChange”(fundingcode01LA11020B,Green Paradox).NIESisgratefulfortheresearchsupportofthe“Global EnvironmentalResearchFund”(2-1402)providedbytheMinistry oftheEnvironment,Japan.
AppendixA.Supplementarydata
Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.
gloenvcha.2016.07.006.
