- Email: [email protected]
Differential Z + multi-jet cross section measurements at CMS
Available online at www.sciencedirect.com
Nuclear and Particle Physics Proceedings 273–275 (2016) 2808–2810 www.elsevier.com/locate/nppp
Differential Z + multi-jet cross section measurements at CMS Eric H. Takasugi1 , on behalf of the CMS Collaboration 1 University
of California, Los Angeles
Abstract We present a measurement of the cross section of Z/γ∗ plus jet events in proton-proton collisions as a function of the √ vector boson transverse momentum. The data were collected with the CMS detector at s = 8 TeV, corresponding to an integrated luminosity of 19.7 fb−1 . The measurement is performed for vector bosons that have a transverse momentum larger than pT > 40 GeV and for different jet multiplicities. We consider MadGraph[1] , Sherpa[2] and BlackHat + Sherpa[3] for Monte Carlo simulation (MC). MadGraph and Sherpa are multi-leg, leading-order matrix elements with parton showering programmes while BlackHat is an NLO parton-level MC. We also study the variables pZT /HT and log10 (pZT /pTj1 ) which are suitable to challenge the validity of NLO predictions. In hadronic searches for new physics, pZT /HT would correspond to a E/T /HT ratio. Keywords: CMS, Standard Model, QCD, Z physics
1. Introduction The associated production of a Z/γ∗ together with one or more jets in the final state has been extensively studied as these processes are background to many searches for new physics. Precise measurements also provide important tests for the standard model and of parton densities in the proton. They also provide a handle to check and tune models used in the MC simulation. The variables pZT /HT and log10 (pZT /pTj1 ) allow us to validate the NLO predictions given by BlackHat as these distributions are quantities in which NLO predictions might reach their calculational limit due to large logarithms or where missing higher-order effects could play a larger role. Here, we use data collected with the CMS detec√ tor at s = 8T eV and an integrated luminosity of 19.7 fb−1 [5]. 2. pTZ Spectrum and cross section ratio The Z/γ∗ +jets pZT spectrum and cross section ratio is briefly presented here [figs 1, 2]. We require that the lepton pT > 20 GeV, |η| < 2.4 and a Z mass window http://dx.doi.org/10.1016/j.nuclphysbps.2015.10.071 2405-6014/© 2016 Published by Elsevier B.V.
of 71 < mll < 111 GeV for same flavour, opposite sign dileptons. The uncertainty bands for MadGraph and Sherpa are statistical. The uncertainty bands for BlackHat are from PDF variation (MSTW) and from varying the renormalization and factorization scales up and down by a factor of 2. The systematic uncertainties considered on data are jet energy scale, lepton scale, pileup, background, jet energy resolution, lepton resolution, scale factor and luminosity uncertainties. MadGraph and Sherpa (scaled to NNLO) overestimate data in the high pZT end. BlackHat underestimates the cross section in the lower end tail. As seen in the good agreement in the 2-jet over 1-jet case, the offset between MadGraph and data is the same in both cases. As seen in the offset between BlackHat and data in the ratio plot, BlackHat agrees with the pZT spectrum in the 2-jet case. j1
3. log10 ( pTZ / pT ) Spectrum The log10 (pZT /pTj1 ) cross section ratio is briefly presented here [fig 3]. MadGraph and Sherpa perform
Preliminary
+-
Z/ γ → l l , Njets ≥ 1
≥2) dσ(N
data stat+syst
T
102
CMS
2809
19.7fb-1 (8TeV) Preliminary
0.7 0.6
jets dpZ T
Sherpa k
NNLO
MadGraph k
0.8
/
BlackHat(Z+1jet)
10
0.9
≥1)
19.7fb-1 (8TeV) *
jets dpZ T
CMS
103
dσ(N
dσ/dp [fb GeV-1]
E.H. Takasugi / Nuclear and Particle Physics Proceedings 273–275 (2016) 2808–2810
NNLO
1
0.5
data
0.4
stat+syst BlackHat(Z+1/2jet)
0.3
10-1
Sherpa
0.2 10-2
100
200
300
400
500
600
700
800
pZ [GeV]
0.1
MadGraph 100
200
300
400
500
600
700
T
BlackHat(Z+1/2jet)/Data
CMS
Preliminary
PDF
Scale
1.4 1.3
/Data
Pdf
Scale
1.1
1.0 0.9
0.8
0.8 0.7 0.6
Sherpa Stat. Err.
NNLO
MadGraph k
Preliminary
05 1.5 1.4 1.3 1.2 1.1 1.0 0.9 0.8 0.7 0.6 0.5
Sherpa Stat. Err.
MadGraph/Data
Sherpa k
1.1
1.0 0.9
1.4 1.3 1.2 1.1 1.0 0.9 0.8 0.7 0.6
CMS
1.3 1.2
1.2
0.7 0.6
19.7fb-1 (8TeV)
1.4
NNLO
/Data
1.4 1.3 1.2 1.1 1.0 0.9 0.8 0.7 0.6
1.5 1.4 1.3 1.2 1.1 1.0 0.9 0.8 0.7 0.6 1.5 0 5
Sherpa/Data
BlackHat/Data
19.7fb-1 (8TeV)
800
PZT
MadGraph Stat. Err. 100
200
300
400
500
600
700
800
pZ [GeV] T
Figure 1: The pZT spectrum in the njets ≥ 1 selection.
MadGraph Stat. Err. 100
200
300
400
500
600
700
800
PZT
Figure 2: Cross section ratio of njets ≥ 2 to njets ≥ 1.
E.H. Takasugi / Nuclear and Particle Physics Proceedings 273–275 (2016) 2808–2810
T
j1
CMS
T
In the pZT /HT case, MadGraph and Sherpa again perform well. BlackHat performs well in the low region, but analogous behavior to log10 (pZT /pTj1 ) was observed in the high end tail of the distribution, in the region where pZT /HT > 1. The disagreements occur in regions where the fixed-order calculation reaches its limit.
104
19.7fb-1 (8TeV) Preliminary
10
well. BlackHat performs well in the central region, but disagrees below at the low and high ends. The disagreement occurs in regions where BlackHat’s NLO njet calculation produces an extra jet with only LO accuracy, i.e. where non-leading order terms dominate in the BlackHat prediction which leads to larger uncertainties.
dσ/d log (pZ/p )
2810
*
+-
Z/ γ → l l , Njets ≥ 2, pll >40
103
T
data stat+syst 102
BlackHat(Z+2jet)
4. Conclusion
Sherpa k
NNLO
∗
10 -1
-0.8
-0.6
-0.4
-0.2
0
NNLO
0.2
0.4
0.6
0.8
j1
log (pZ/p ) 10
T
T
BlackHat/Data
19.7fb-1 (8TeV)
CMS
/Data
1.4 1.3 1.2 1.1 1.0 0.9 0.8 0.7 0.6
Preliminary
PDF
Scale
1.4 1.3
NNLO
We find that the ratio of MC to data for Z/γ + jets is not well reproduced by MadGraph and Sherpa at high vector boson pT . Furthermore, this discrepancy increases as function of Z pT . However, using NLO calculations, we find a reduction in the discrepancy between data and MC, indicating that this might be related to missing higher orders. Disagreements between between BlackHat and data occur exactly in those kinematic regions where NLO QCD is well understood to be unreliable and can be seen in the large scale errors. In these regions the LO + parton shower descriptions (MadGraph and Sherpa) are superior.
MadGraph k
1.1
1.0 0.9 0.8
Sherpa Stat. Err.
1.4 1.3 1.2 1.1 1.0 0.9 0.8 0.7 0.6
NNLO
/Data
0.7 0.6
MadGraph k
[1] J. Alwall, M. Herquet, F. Maltoni, O. Mattelaer, T. Stelzer, MadGraph 5 : Going Beyond, JHEP 1106 (2011) 128. arXiv:1106.0522, doi:10.1007/JHEP06(2011)128. [2] T. Gleisberg, S. Hoeche, F. Krauss, M. Schonherr, S. Schumann, et al., Event generation with SHERPA 1.1, JHEP 0902 (2009) 007. arXiv:0811.4622, doi:10.1088/1126-6708/2009/02/007. [3] Z. Bern, L. Dixon, F. Febres Cordero, S. Hche, H. Ita, et al., Ntuples for NLO Events at Hadron Colliders, Comput.Phys.Commun. 185 (2014) 1443–1460. arXiv:1310.7439, doi:10.1016/j.cpc.2014.01.011. ∗ [4] CMS Collaboration, Measurement of the ratio of the √ Z/γ +jets and photon+jets cross section in pp collisions at s = 8 TeV, CMS Physics Analysis Summary CMS-SMP-14-005 (2014). URL: http://cds.cern.ch/record/1740969. [5] CMS Collaboration, The CMS experiment at the CERN LHC, JINST 3:S08004, 2008.
Sherpa k
1.2
References
MadGraph Stat. Err. -1
-0.8
-0.6
-0.4
-0.2
0
0.2
0.4
0.6
0.8
j1
log (pZ /p ) 10
T
T
Figure 3: The log10 (pZT /pTj1 ) ratio in the njets ≥ 2 selection.
Nuclear and Particle Physics Proceedings 273–275 (2016) 2808–2810 www.elsevier.com/locate/nppp
Differential Z + multi-jet cross section measurements at CMS Eric H. Takasugi1 , on behalf of the CMS Collaboration 1 University
of California, Los Angeles
Abstract We present a measurement of the cross section of Z/γ∗ plus jet events in proton-proton collisions as a function of the √ vector boson transverse momentum. The data were collected with the CMS detector at s = 8 TeV, corresponding to an integrated luminosity of 19.7 fb−1 . The measurement is performed for vector bosons that have a transverse momentum larger than pT > 40 GeV and for different jet multiplicities. We consider MadGraph[1] , Sherpa[2] and BlackHat + Sherpa[3] for Monte Carlo simulation (MC). MadGraph and Sherpa are multi-leg, leading-order matrix elements with parton showering programmes while BlackHat is an NLO parton-level MC. We also study the variables pZT /HT and log10 (pZT /pTj1 ) which are suitable to challenge the validity of NLO predictions. In hadronic searches for new physics, pZT /HT would correspond to a E/T /HT ratio. Keywords: CMS, Standard Model, QCD, Z physics
1. Introduction The associated production of a Z/γ∗ together with one or more jets in the final state has been extensively studied as these processes are background to many searches for new physics. Precise measurements also provide important tests for the standard model and of parton densities in the proton. They also provide a handle to check and tune models used in the MC simulation. The variables pZT /HT and log10 (pZT /pTj1 ) allow us to validate the NLO predictions given by BlackHat as these distributions are quantities in which NLO predictions might reach their calculational limit due to large logarithms or where missing higher-order effects could play a larger role. Here, we use data collected with the CMS detec√ tor at s = 8T eV and an integrated luminosity of 19.7 fb−1 [5]. 2. pTZ Spectrum and cross section ratio The Z/γ∗ +jets pZT spectrum and cross section ratio is briefly presented here [figs 1, 2]. We require that the lepton pT > 20 GeV, |η| < 2.4 and a Z mass window http://dx.doi.org/10.1016/j.nuclphysbps.2015.10.071 2405-6014/© 2016 Published by Elsevier B.V.
of 71 < mll < 111 GeV for same flavour, opposite sign dileptons. The uncertainty bands for MadGraph and Sherpa are statistical. The uncertainty bands for BlackHat are from PDF variation (MSTW) and from varying the renormalization and factorization scales up and down by a factor of 2. The systematic uncertainties considered on data are jet energy scale, lepton scale, pileup, background, jet energy resolution, lepton resolution, scale factor and luminosity uncertainties. MadGraph and Sherpa (scaled to NNLO) overestimate data in the high pZT end. BlackHat underestimates the cross section in the lower end tail. As seen in the good agreement in the 2-jet over 1-jet case, the offset between MadGraph and data is the same in both cases. As seen in the offset between BlackHat and data in the ratio plot, BlackHat agrees with the pZT spectrum in the 2-jet case. j1
3. log10 ( pTZ / pT ) Spectrum The log10 (pZT /pTj1 ) cross section ratio is briefly presented here [fig 3]. MadGraph and Sherpa perform
Preliminary
+-
Z/ γ → l l , Njets ≥ 1
≥2) dσ(N
data stat+syst
T
102
CMS
2809
19.7fb-1 (8TeV) Preliminary
0.7 0.6
jets dpZ T
Sherpa k
NNLO
MadGraph k
0.8
/
BlackHat(Z+1jet)
10
0.9
≥1)
19.7fb-1 (8TeV) *
jets dpZ T
CMS
103
dσ(N
dσ/dp [fb GeV-1]
E.H. Takasugi / Nuclear and Particle Physics Proceedings 273–275 (2016) 2808–2810
NNLO
1
0.5
data
0.4
stat+syst BlackHat(Z+1/2jet)
0.3
10-1
Sherpa
0.2 10-2
100
200
300
400
500
600
700
800
pZ [GeV]
0.1
MadGraph 100
200
300
400
500
600
700
T
BlackHat(Z+1/2jet)/Data
CMS
Preliminary
Scale
1.4 1.3
/Data
Scale
1.1
1.0 0.9
0.8
0.8 0.7 0.6
Sherpa Stat. Err.
NNLO
MadGraph k
Preliminary
05 1.5 1.4 1.3 1.2 1.1 1.0 0.9 0.8 0.7 0.6 0.5
Sherpa Stat. Err.
MadGraph/Data
Sherpa k
1.1
1.0 0.9
1.4 1.3 1.2 1.1 1.0 0.9 0.8 0.7 0.6
CMS
1.3 1.2
1.2
0.7 0.6
19.7fb-1 (8TeV)
1.4
NNLO
/Data
1.4 1.3 1.2 1.1 1.0 0.9 0.8 0.7 0.6
1.5 1.4 1.3 1.2 1.1 1.0 0.9 0.8 0.7 0.6 1.5 0 5
Sherpa/Data
BlackHat/Data
19.7fb-1 (8TeV)
800
PZT
MadGraph Stat. Err. 100
200
300
400
500
600
700
800
pZ [GeV] T
Figure 1: The pZT spectrum in the njets ≥ 1 selection.
MadGraph Stat. Err. 100
200
300
400
500
600
700
800
PZT
Figure 2: Cross section ratio of njets ≥ 2 to njets ≥ 1.
E.H. Takasugi / Nuclear and Particle Physics Proceedings 273–275 (2016) 2808–2810
T
j1
CMS
T
In the pZT /HT case, MadGraph and Sherpa again perform well. BlackHat performs well in the low region, but analogous behavior to log10 (pZT /pTj1 ) was observed in the high end tail of the distribution, in the region where pZT /HT > 1. The disagreements occur in regions where the fixed-order calculation reaches its limit.
104
19.7fb-1 (8TeV) Preliminary
10
well. BlackHat performs well in the central region, but disagrees below at the low and high ends. The disagreement occurs in regions where BlackHat’s NLO njet calculation produces an extra jet with only LO accuracy, i.e. where non-leading order terms dominate in the BlackHat prediction which leads to larger uncertainties.
dσ/d log (pZ/p )
2810
*
+-
Z/ γ → l l , Njets ≥ 2, pll >40
103
T
data stat+syst 102
BlackHat(Z+2jet)
4. Conclusion
Sherpa k
NNLO
∗
10 -1
-0.8
-0.6
-0.4
-0.2
0
NNLO
0.2
0.4
0.6
0.8
j1
log (pZ/p ) 10
T
T
BlackHat/Data
19.7fb-1 (8TeV)
CMS
/Data
1.4 1.3 1.2 1.1 1.0 0.9 0.8 0.7 0.6
Preliminary
Scale
1.4 1.3
NNLO
We find that the ratio of MC to data for Z/γ + jets is not well reproduced by MadGraph and Sherpa at high vector boson pT . Furthermore, this discrepancy increases as function of Z pT . However, using NLO calculations, we find a reduction in the discrepancy between data and MC, indicating that this might be related to missing higher orders. Disagreements between between BlackHat and data occur exactly in those kinematic regions where NLO QCD is well understood to be unreliable and can be seen in the large scale errors. In these regions the LO + parton shower descriptions (MadGraph and Sherpa) are superior.
MadGraph k
1.1
1.0 0.9 0.8
Sherpa Stat. Err.
1.4 1.3 1.2 1.1 1.0 0.9 0.8 0.7 0.6
NNLO
/Data
0.7 0.6
MadGraph k
[1] J. Alwall, M. Herquet, F. Maltoni, O. Mattelaer, T. Stelzer, MadGraph 5 : Going Beyond, JHEP 1106 (2011) 128. arXiv:1106.0522, doi:10.1007/JHEP06(2011)128. [2] T. Gleisberg, S. Hoeche, F. Krauss, M. Schonherr, S. Schumann, et al., Event generation with SHERPA 1.1, JHEP 0902 (2009) 007. arXiv:0811.4622, doi:10.1088/1126-6708/2009/02/007. [3] Z. Bern, L. Dixon, F. Febres Cordero, S. Hche, H. Ita, et al., Ntuples for NLO Events at Hadron Colliders, Comput.Phys.Commun. 185 (2014) 1443–1460. arXiv:1310.7439, doi:10.1016/j.cpc.2014.01.011. ∗ [4] CMS Collaboration, Measurement of the ratio of the √ Z/γ +jets and photon+jets cross section in pp collisions at s = 8 TeV, CMS Physics Analysis Summary CMS-SMP-14-005 (2014). URL: http://cds.cern.ch/record/1740969. [5] CMS Collaboration, The CMS experiment at the CERN LHC, JINST 3:S08004, 2008.
Sherpa k
1.2
References
MadGraph Stat. Err. -1
-0.8
-0.6
-0.4
-0.2
0
0.2
0.4
0.6
0.8
j1
log (pZ /p ) 10
T
T
Figure 3: The log10 (pZT /pTj1 ) ratio in the njets ≥ 2 selection.