DESY–15–253 IPPP/15/76 DCPT/15/152 MAN/HEP/2015/21 December2015 Summary of workshop on Future Physics with HERA Data A.Bacchetta1,J.Blümlein2,O.Behnke3,J.Dainton4,M.Diehl3,F.Hautmann5,6,A.Geiser3, H.Jung3,7,U.Karshon8,D.Kang9,P.Kroll10,C.Lee9,S.Levonian3,A.Levy11, E.Lohrmann3,12,S.Moch12,L.Motyka13,R.McNulty14,V.Myronenko3,E.R.Nocera6,15, S.Plätzer16,17,A.Rostomyan3,M.Ruspa18,M.Sauter19,G.Schnell20,21,S.Schmitt3, 6 1 H.Spiesberger22,23,I.Stewart24,O.Turkot3,A.Valkárová25,K.Wichmann3, 0 M.Wing26,3,12,A.F.Z˙arnecki27 2 n a J 1 UniversityofPaviaandINFN,Pavia,Italy,2 DESY,Zeuthen,Germany,3 DESY,Hamburg,Germany, 7 4 UniversityofLiverpool,UK,5 RutherfordAppletonLaboratory,Didcot,UK, ] 6 UniversityofOxford,Oxford,UK,7 UniversityofAntwerp,Antwerp,Belgium, x 8 WeizmannInstitute,Rehovot,Israel,9 LosAlamosNationalLaboratory,LosAlamos,NM,USA, e - 10 BergischeUniversitätWuppertal,Wuppertal,Germany,11 TelvAvivUniversity,TelAviv,Israel, p e 12 UniversitätHamburg,Hamburg,Germany,13 JagiellonianUniversity,Kraków,Poland, h 14 UniversityCollegeDublin,Dublin,Ireland,15 UniversityofGenovaandINFNGenova,Italy, [ 16 InstituteforParticlePhysicsPhenomenology,Durham,UK, 1 17 UniversityofManchester,Manchester,UK, v 18 UniversityofPiemonteOrientaleandINFNTorino,Italy, 9 9 19 UniversitätHeidelberg,Heidelberg,Germany, 4 20 UniversityoftheBasqueCountryandBasqueFoundationofScience,Bilbao,Spain, 1 0 21 GhentUniversity,Gent,Belgium,22 JohannesGutenberg-Universität,Mainz,Germany, . 23 UniversityofCapeTown,Rondebosch,SouthAfrica, 1 0 24 MassachusettsInstituteofTechnology,Cambridge,MA,USA, 6 25 CharlesUniversity,Praha,CzechRepublic,26 UniversityCollegeLondon,London,UK 1 27 UniversityofWarsaw,Warsaw,Poland : v i X r Abstract a Recent highlights from the HERA experiments, Hermes, H1 and ZEUS, are reviewed and ideas for future analyses to fully exploit this unique data set are proposed. This document is a summary of a workshop on future physics with HERA data held at DESY, Hamburg at the end of 2014. All areas of HERAphysicsarecoveredandcontributionsfrombothexperimentalistsandtheoristsareincluded. The document outlines areas where HERA physics can still make a significant contribution, principally in a deeper understanding of QCD, and its relevance to other facilities. Within the framework of the Data PreservationinHighEnergyPhysics,theHERAdatahavebeenpreservedforanalysestotakeplaceover atimescaleof10yearsandmore. Therefore,althoughanextensivelistofpossibilitiesispresentedhere, safestorageofthedataensuresthatitcanalsobeusedinthefarfutureshouldnewideasandanalysesbe proposed. 1 2 Contents 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 2 RecenthighlightsfromHERA. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 2.1 RecentHERAresultsonprotonstructure A.Levy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 2.2 Resultsonthehadronicfinalstateanddiffractioninepscattering A.Valkárová . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 3 OverviewofperspectivesonphysicswithHERAdata . . . . . . . . . . . . . . . . . . . . 9 3.1 Transversemomentumdependentpartondistributions H.Jung,F.Hautmann . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 3.2 PossiblefutureHERAanalyses A.Geiser . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 4 Partondensityfunctionsandelectroweakphysics . . . . . . . . . . . . . . . . . . . . . . 10 4.1 Three-loopheavyflavourcorrectionstodeep-inelasticscattering J.Blümlein . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 4.2 Precisepartondistributions S.Moch . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 4.3 ElectroweakphysicswithHERAdata H.Spiesberger . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 4.4 Futureelectroweakandcontactinteractionfits K.Wichmann,V.Myronenko,O.Turkot,A.F.Zarnecki . . . . . . . . . . . . . . 14 5 Jetsandhadronicfinalstates. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 5.1 Precisionjetphysicsindeepinelasticscattering D.Kang,C.Lee,I.W.Stewart . . . . . . . . . . . . . . . . . . . . . . . . . . 15 5.2 WhyandhowtosearchforcharmpentaquarkswiththeHERAdata U.Karshon . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 5.3 Anewsearchforinstantons E.Lohrmann . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 6 Physicstopicscommonwithotherexperiments . . . . . . . . . . . . . . . . . . . . . . . . 18 6.1 CommonphysicsbetweenHERAandLHCb R.McNulty . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 7 Diffractionandlow-xphysics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20 7.1 FutureprospectsfordiffractionatHERA M.Ruspa . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20 7.2 TwistdecompositioninDISandDDISinthedipoleapproach L.Motyka . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 7.3 Photoproductionofπ+π pairsinamodelwithtensor-pomeronandvector-odderonex- − change M.Sauter,A.Bolz,C.Ewerz,M.Maniatis,O.Nachtmann,A.Schöning . . . . 22 8 Spinphysics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 8.1 GPDsfromexclusivemesonleptoproduction P.Kroll . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 8.2 Inclusiveandsemi-inclusivespinphysics A.Bacchetta . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24 3 8.3 Thelongitudinalspinstructureofthenucleon E.Nocera . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25 9 MonteCarloprogrammesforHERAphysics . . . . . . . . . . . . . . . . . . . . . . . . . 27 9.1 PartonshowerMonteCarlogeneratorsbeyondcollinearapproximations F.Hautmann,H.Jung . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 9.2 Herwig++forep1 andRivetforep2 S.Plätzer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28 10 Summary: fromDirac’selectrontoDiracelectronsandquarks3 J.Dainton . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 10.1 The1992HERAperspective . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 10.2 HERAatthecloseof2014 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 10.3 Onwards? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 1WorkpresentedonbehalfoftheHerwig++collaboration. 2WorkinprogresswithHannesJung. 3Thepresentationonwhichthisveryshortsummaryisbasedisverymuchapersonalperspectiveof22yearsofHERA physics. ItisbasedonaninvitationtospeakattheendofthecolloquiumandworkshopatDESYinNovember2014. Itisnot inclusiveofthemultitudeofresultsandmeasurementsfromthefourHERAexperiments,H1ZEUS,HERMESandHERA- B.Itismadepossiblebygenerationsofcolleagues, bothontheexperimentsandontheHERAmachine, whosehardwork, dedication, innovativedeterminationandunswervingcommitmenthassecuredHERAaspivotalinthedevelopmentof20th andearly21stcenturyphysics,culminatingintheSMoftoday.Specificcontributionstothiscolloquium,whicharetobefound inotherpresentationsincludedwiththisshortsummary,containmoredetailsoflatestresultsinrespectofwhatiswrittendown here.Alongwrite-upisnearlycompletewhichwillbepublishedasaDESYpreprintshortly. 4 1 Introduction Thesefollowingproceedingsaresummariesfromindividualspeakersandotherswhocontributedtothe symposium and workshop on future physics with HERA data. In the symposium, some of the latest experimental results from the HERA collaborations are reviewed and a theoretical perspective is given (see Section 2). The workshop on future physics ideas was organised thematically into sessions which arethenorganisedintothefollowingsectionsinthiswrite-up: – OverviewofperspectivesonphysicswithHERAdata(seeSection3); – Partondensityfunctionsandelectroweakeffects(seeSection4); – Jetsandhadronicfinalstates(seeSection5); – Physicstopicscommonwithotherexperiments(seeSection6); – Diffractionandlow-xphysics(seeSection7); – Spinphysics(seeSection8); – MonteCarloprogrammesforHERAphysics(seeSection9); Additionally, a technical presentation was given [1] on the safe storage and maintenance of the HERAdataforfutureanalysesformanyyearstocome, aspartoftheglobalDPHEPeffort[2]. Finally asummaryoftheworkshopwasgiven(seeSection10). 2 RecenthighlightsfromHERA 2.1 RecentHERAresultsonprotonstructure A.Levy ThelatestresultsontheprotonstructurebythetwoHERAcollaborations,H1andZEUS,arepresented. They include new results since the last HERA Symposium and cover: the cross section data combi- nations; the new HERAPDF2.0 parton distribution functions; measurements of the charm and bottom structure functions, with the determination of the masses of these heavy quarks; energy dependence of D -meson production; high-Q2 measurements from the ZEUS collaboration in the high-Bjorken-x ∗ regionuptovaluesofx = 1;andlongitudinalstructurefunctionF frombothcollaborations. ∼ L Though the HERA collider stopped running in 2007, the data analysis has been going on with big efforts on finalising the inclusive cross section measurements. As both collaborations have done so [3,4], the way was paved for the ‘climax’ of an experiment; the data combination. This procedure, whichhadproducedprecisionmeasurementsalreadyfortheHERAIperiodofrunning[5],hasproduced text-bookresultsoftheHERAcrosssectionmeasurements. Inthenewcombination,therewere41data setsincluding2927datapointswhichwerecombinedto1307averagedmeasurementswith165sources ofcorrelatedsystematicuncertainties. Thedatasetswereconsistent,producingatotalχ2/ndf=1.04. Thescalingviolationplot,namelythecombinedneutralcurrent(NC)e+pande preducedcross − sectionsasafunctionofQ2forfixedvaluesofBjorkenx,forallHERAdataresultingfromanintegrated luminosityof1fb 1 isshowninFig.1. Theyarethemostprecisedatameasurementsinthiskinematic − region, reaching Q2 values of 50 000GeV2 and x values down to 0.00005 (in the present figure only values down to 0.02 are shown). The scaling violations are well described by the HERAPDF2.0 NLO predictions. Figure2showsthecombinedHERAchargedcurrent(CC)e preducedcrosssectionsasafunction − of x for some fixed Q2 values as indicated in the figure. These data are also well described by the HERAPDF2.0 NLO predictions. Figures 1 and 2 are part of the HERA legacy. They are the results of manyyearsofdedicatedworktoproducethebestqualitydatabythetwoHERAcollaborationsH1and ZEUS. 5 H1 and ZEUS preliminary H1 and ZEUS preliminary 2(x,Q)σr, NC110034 √Hs E=NNR 3ACC18 ( eepG+-prpee V 0l0..).45 ffbb--11 HH√sEE =RR 3AA1PPx8NN =DD G CC0FF.e0 22V2ee.. +- 00 p(p x((4pp7rr5)eell..)) NNLLOO,, QQ2m i n = 3.5 GeV2 2(x,Q)-σr, CC 12 Q2 = 300 GeV2 Q2 = 500 GeV2 Q2 = 1000 GeV2 Q2 = 1500 GeV2 x = 0.032 (x400) x = 0.05 (x270) 102 x = 0.08 (x170) 0 Q2 = 2000 GeV2 Q2 = 3000 GeV2 Q2 = 5000 GeV2 Q2 = 8000 GeV2 x = 0.13 (x80) 1 10 x = 0.18 (x20) 0.5 x = 0.25 (x6) 1 0 0.8 Q2 = 15000 GeV2 Q2 = 30000 GeV2 10-2 10-1 10-2 10-1 x 10-1 x = 0.40 (x2) 00..46 √s =H 3E1R8 AG CeVC e-p (prel.) 0.4 fb-1 HERAPDF2.0 (prel.) 10-2 x = 0.65 0.02 NNLLOO,, QQ2m i n = 3.5 GeV2 10-2 10-1 10-2 10-1 x 102 103 104 105 Q2/ GeV2 Fig. 2: The combined HERA inclusive CC e p Fig. 1: The combined HERA data for the inclusive − reduced cross sections as a function of x at se- NC e+p and e p reduced cross sections as a func- − lectedvaluesofQ2√s=318GeVwithoverlaid tion of Q2 for selected values of x at √s = 318GeV predictionsofthepreliminaryversionofHERA- withoverlaidpredictionsofthepreliminaryversionof PDF2.0NLO.Thebandsrepresentthetotalun- HERAPDF2.0NLO.Thebandsrepresentthetotalun- certaintiesonthepredictions. certaintiesofthepredictions. H1 and ZEUS preliminary dof The combined data are used to obtain the parton distribu- 2/χ1.24 HERAPDF2.0 (prel.) tion function (PDFs) in the proton, using the DGLAP [6] χ2/dof = 1386/1130 NLO 1.22 NNLO equations. Since perturbative QCD (pQCD) is expected to be valid only from a scale where partons can be resolved, 1.2 thepQCDanalysisusesdataabovethatscale. Alowervalue 1.18 forQ2 waschosenas3.5GeV2 andastudyoftheresulting cut 1.16 χ2/dof = 1156/1001 χ2/dofasafunctionofthiscutwascarriedoutandshownin Fig.3. Acleardependenceoftheχ2/dofonQ2 isobserved; 1.14 its value decreases as Q2 increases to 10GeV2 and starts 0 5 10 15 20 25 30 Q2 cut / GeV2 to increase again after 15GeV2. Follo∼wing this behaviour, Fig.3: Thedependenceofχ2/dofonQ2 two sets of PDFs have been determined, one with Q2 = cut min of the NLO and NNLO fits to the HERA 3.5GeV2 andanotherwithQ2 =10GeV2. min combinedinclusivedata. ThetwosetsofPDFsareshowninFigs.4and5. Althoughthereisnovisibledifferencebetween the two in the higher-x region, one sees that the gluon uncertainty is much larger in the low-x region for the Q2 = 10GeV2 case. This happens because leaving out data in the range Q2 = 3.5 – 10GeV2 min removes events which constrain the gluon density in the low-x region. Further details and final results onthecombinedinclusiveHERAdatacanbefoundintherecentpublication[7]. H1 and ZEUS combined their charm structure function data and extracted the running charm mass[8],resultinginm (m ) = 1.26 0.05(exp) 0.03(mod) 0.02(param) 0.02(α )GeV.The c c S ± ± ± ± ZEUScollaborationmeasuredthebottomstructurefunction[9]andextractedtheMSvalueofthebottom quark,m (m ) = 4.07 0.14(exp)+0.01(mod)+0.05(param)+0.08(theo)GeV. b b ± 0.07 0.00 0.05 − − − The ZEUS collaboration measured the energy dependence of the D cross section [10] in the ∗ 6 Fig.4: Thepartondistributionfunctionsxuv,xdv, Fig.5: Thepartondistributionfunctionsxuv,xdv, xS = 2x(U¯ +D¯) and xg of HERAPDF2.0 NLO xS = 2x(U¯ +D¯) and xg of HERAPDF2.0 NLO at µ2 = 3.5GeV2. The gluon and sea distributions at µ2 = 10GeV2. The gluon and sea distributions f f arescaleddownbyafactorof20. Theexperimen- arescaleddownbyafactorof20. Theexperimen- tal, model and parameterisation uncertainties are tal, model and parameterisation uncertainties are shown. shown. rangeof√s=225-320GeVandfoundthedependencetobeinagreementwithNLOQCDpredictions. It also used an improved reconstruction method of Bjorken x in the high-Q2 and high-x region and extracted [11] the inclusive NC cross section up to x 1. The fine binning in x with the extension of → kinematic coverage up to x = 1 make these data important input to fits constraining the PDFs in the ∼ valence-quarkdomain. The H1 collaboration measured [12] the F L structure function in the region 1.5 < Q2 < 800 GeV2 while the ZEUS collaboration measured it [13] in the kinematic range 9 < Q2 < 110 GeV2. The results are shown in Fig. 6. The uncertainties of the ZEUS results are larger than those of H1. The ZEUSresults,thoughconsistentlylowerthan those of H1, are consistent with them be- cause of the correlated uncertainties. The predictions shown by the shaded area are in reasonable agreement with both data sets. Fig. 6: F as a function of Q2 as measured by the H1 L One would hope to have a combined mea- and ZEUS collaborations. The shaded area are predic- surementwhichwillproducetheHERAFL. tions based on different PDF parameterisations, as indi- catedinthefigure. 7 H1 VFPS Data NLO H1 2006 Fit B × 0.83 ×( 1+δhadr) (Data/NLO) H1 Preliminary DIS (Data/NLO) γp (Data/NLO) γp (Data/NLO) DIS 0 0.5 1 1.5 Suppression factor Fig. 7: DIS and photoproduction integrated cross sections normalised to the NLO QCD theoretical cal- culationsareshownasawhiteline. ThedoubleratioofdataoverNLOQCDcalculationsforphotopro- ductiontodataoverNLOQCDcalculationsforDISispresentedasawhitelineinthelastrow. 2.2 Resultsonthehadronicfinalstateanddiffractioninepscattering A.Valkárová 2.2.1 QCDandhadronicfinalstate The ZEUS collaboration presented cross sections for events containing an isolated high-energy photon, with and without a jet, produced in photoproduction using the full HERA II data set [14]. These mea- surements were later on extended [15]. Events with isolated photons can provide a direct probe of the underlyingpartonicprocessinhigh-energycollisions,sincetheemissionofahigh-energyphotonisnot affected by hadronisation. Within the large theoretical uncertainties, the theoretical predictions agree withthedata. Inclusive jet, dijet and trijet differential cross sections were measured in neutral current deep- inelastic scattering (DIS) collisions using the H1 detector [16]. Theoretical QCD calculations at NLO, correctedforhadronisationandelectroweakeffects, provideagooddescriptionofthemeasuredsingle- and double-differential jet cross sections as a function of all studied variables. This measurement was confirmedalsobyaZEUSanalysismeasuringtrijetdifferentialcrosssections[17]. TheH1collaboration derivedthemostprecisevalueofthestrongcouplingconstantfromjetdataatNLOasmeasuredinone experiment,α (M ) = 0.1165(8) (38) . s Z exp pdf,theo A sophisticated analysis of H1 data aimed to find instantons in DIS ep interactions [18]. The observedupperlimitontheQCDinstantoncrosssectionofabout1.6pbdoesnotsupportthetheoretical prediction. Thisanalysisisstillcontinuing. 2.2.2 Diffraction It has been established by the ZEUS collaboration that the measurement of the shape of the azimuthal angular distributions of exclusive dijets in diffractive DIS prefers a 2-gluon exchange model of qq pro- ductionoverresolvedPomeronmodel[19]. TheH1collaborationpresentedthesingle-anddouble-differentialdijetcrosssectionsindiffractive DIS ep scattering using the large rapidity gap method to select diffractive events. Both shapes and normalisation of the single-differential cross sections are reproduced satisfactorily by the NLO QCD predictionswithintheexperimentalandtheoryuncertainties[20]. 8 Anewmeasurementofthediffractivedijetcrosssectionwithaleadingfinalstateprotondetected intheVeryForwardProtonSpectrometerwasperformedbytheH1collaborationinphotoproductionand DIS [21]. With the exception of the momentum transfer Q2, identical kinematic regions were used for DISandphotoproductiondijets. InbothphotoproductionandDIStheshapesofdifferentialdistributions arewelldescribedbyaNLOQCDcalculations. However,forphotoproductionthetheoreticalpredictions lie systematically above the data, corresponding to a data suppression factor of about 0.6. For DIS, the NLOQCDcalculationsagreewithintheoreticaluncertaintieswithdata. Integratedoverthemeasuredkinematicrange,thedoubleratioofdataoverNLOQCDcalculations forphotoproductiontodataoverNLOcalculationsforDISis: (DATA/NLO) γp = 0.55 0.10(data) 0.02(theo.) (1) (DATA/NLO) ± ± DIS and is shown in Fig. 7. This observation is in agreement with previous H1 measurements, where com- plementaryexperimentalmethodshavebeenused. Withinthetheoreticalframeworkapplied,thiscould hintatthebreakingofQCDfactorisationindiffractivedijetphotoproduction. 3 OverviewofperspectivesonphysicswithHERAdata 3.1 Transversemomentumdependentpartondistributions H.Jung,F.Hautmann Thecalculationofacrosssectioninepandppcollisionsproceedsviatheconvolutionofapartoniccross section with parton density functions (PDFs), which describe the probability density to find a parton in a hadron with a given momentum fraction x at a resolution scale µ. For inclusive single scale cross sections, like the inclusive DIS cross section, the parton densities are functions of x and the resolution scaleµonly(collinearfactorisation),whileformoreexclusiveanddifferentialmulti-scalecrosssections, thetransversemomentaoftheinteractingpartonsbecomeimportant,andso-calledtransversemomentum dependent(TMD)partondensityfunctionsareneeded. DIS is the process where TMD factorisation is proven. The precision determination of TMDs (for quarks and gluons) is extremely important especially in light of potential factorisation breaking effectsinhadron–hadroncollisions[22,23]. ThepresentstatusofTMDsissummarisedin[24];alibrary of available TMD fits and parameterisations can be found in [25]; an application of TMDs to W+jet productioninppisgivenin[26]. The TMDs, similarly to collinear PDFs, can be determined from fits to inclusive cross sec- tions [27], but the transverse momentum dependence is, at present, not well constrained. Dedicated measurementsareneededtoconstrainthesmallandlargek regionofquarkandgluonTMDs: T – the inclusive jet cross section from lowest p (0.5GeV) to large p to constrain the quark T T ∼ O andgluonTMDs – inclusiveparticle-andidentifiedparticle-productionasafunctionofx,Q2,η andp ofthetracks T toconstraintheflavourdependenceoftheTMDs In jet production in γp, but also in DIS, the region transverse in azimuth to the jet direction is usuallynotwelldescribedbystandardsimulationsincludingpartonshowers(underlyingeventmeasure- ments),andespeciallyforγp,multiplepartoninteractions(MPIs)mustbeincluded,processeswhichare alsoneededinpp. However,theseparationofmultipleinteractionsandhigherordercontributionsfrom a single interaction depends in general on the factorisation used and on the factorisation scheme. Espe- cially measurements in DIS are crucial, since there MPIs are not expected to contribute because of the point-like nature of the virtual photon. Such measurements would be important to constrain the contri- butionfromhigherorderQCDradiationinasinglepartonicinteraction. Withthis,acomparisonwithγp orppwouldallowasystematicestimateoftheMPIcontribution,inafactorisationschemeindependent way. 9 3.2 PossiblefutureHERAanalyses A.Geiser The purpose of this contribution is to give a (subjective) overview of a large variety of analyses which mightbepossibleinthefutureusingtheHERAdata. Itisbasedon – acomparisonofearlygoalswithwhathasactuallybeenachieved, – anextrapolationofhowalreadyexistingresultscanbeimprovedfurther,and – acollectionofthoughtsaboutanalysistopicswhichmightneverhavebeenattemptedbefore. Becausethereisobviouslylargeoverlap,manyoftheideaspresentedbyotherparticipantsofthework- shop are integrated as much as possible and need to be appropriately quoted and referenced. Doing this adequately is not possible within the limited space available for a single contribution to this docu- ment. Thereforethebulkoftheinformation,withthecorrespondingreferences,ispresentedinaseparate document[28],whichisonlybrieflysummarisedhere. AvarietyofpossiblefutureanalysesofHERAdatainthecontextoftheDPHEPdatapreservation programme [2,29] is collected, motivated, and commented. The focus is placed on possible future analyses of the existing ep collider data and their physics scope. Comparisons to the original scope of theHERAprogrammearemade,andcrossreferencestotopicsalsocoveredbyotherparticipantsofthe workshoparegiven. Thisincludestopicson – QCD, – protonstructure, – diffraction, – jets, – hadronicfinalstates, – heavyflavours, – electroweakphysics, andtheapplicationofrelatedtheoryandphenomenologytopicslike – NNLOQCDcalculations, – low-xrelatedmodels, – nonperturbativeQCDaspects,and – electroweakradiativecorrections. SynergieswithothercolliderprogrammeslikeLHCandEICarealsoaddressed. Insummary,therangeofphysicstopicswhichcanstillbeuniquelycoveredusingtheexistingdata is very broad and of considerable physics interest, often matching the interest of results from colliders currently in operation. Due to well-established data and MC sets, calibrations, and analysis procedures themanpowerandexpertiseneededforaparticularanalysisisoftenverymuchsmallerthanthatneeded foranongoingexperiment. Sincecentrallyfundedmanpowertocarryoutsuchanalysesisnotavailable anylonger,thiscontribution,andinparticularitsextendedversion[28],notonlyaddressesexperienced self-funded experimentalists, but also theorists and master-level students who might wish to carry out suchananalysis. 4 Partondensityfunctionsandelectroweakphysics 4.1 Three-loopheavyflavourcorrectionstodeep-inelasticscattering J.Blümlein Thepresentprecisionofthedeep-inelasticworlddatarequiresQCDanalysesandfitstotheheavyquark massesm andm at3-looporder[30–32]. Whilethemasslesscorrectionsto3-looporderareavailable c b 10