Update on CTEQ-TEA PDFs Carl Schmidt Michigan State University On - PowerPoint PPT Presentation

Update on CTEQ-TEA PDFs Carl Schmidt Michigan State University On behalf of the CTEQ-TEA group July 31, 2107 DPF 2017 Fermilab, Batavia, IL, USA

CTEQ-TEA group • CTEQ – Tung et al. (TEA) in memory of Prof. Wu-Ki Tung, who established CTEQ Collaboration in early 90’s • Current members of CTEQ-TEA group: Sayipjamal Dulat (Xinjiang U.), Tie-Jiun Hou, Pavel Nadolsky, Bo-Ting Wang, Keping Xie (Southern Methodist U.), Jun Gao (Shanghai Jiaotong U), Marco Guzzi (U. of Manchester & Kennesaw State), Joey Huston, Jon Pumplin, Dan Stump, CS, Jan Winter, C.-P.Yuan (Michigan State U.)

CT14 Parton Distributions • 2015 major release of general purpose PDFs CT14NNLO/NLO, including α s u series and n f =3,4,5 [1506.07443] d g • Combined HERA charm, H1 FL data in NC DIS s • Early LHC Run 1 data on W/Z charged lepton rapidity and asymmetry • Inclusive jet production from ATLAS & CMS • More flexible parametrization: gluon, d/u at large- x, d/u & dbar/ubar at small-x, 28 eigenvectors compared to 25 for CT10 • http://hep.pa.msu.edu/cteq/public/index.html

Beyond standard CT14 • CT14QEDinc: constraints on photon PDFs in the nucleon [1509.02905] • CT14MC: MC replicas for certain applications [1607.06066] • CT14HERA2: effects of combined HERA1+2 data [1609.07968] • CT14IC: intrinsic/fitted charm component [1707.00657] • CT17: preliminary fits (CT17p)

Outline • CT14IC (intrinsic charm) • CT17 preliminary fits and impact of new data • LHC ttbar distributions (ATLAS 8 TeV) • Conclusion

Fitted Charm and CT14IC • Charm PDF in CT14 is generated perturbatively from c ( x , Q 0 = m c )=0 (nonzero matching at NNLO) • However, nonzero c ( x , Q 0 = m c ) is possible at % level of total mom. fraction • “fitted charm” = “nonperturbative charm” + other higher order terms in α s or Λ 2 / m c 2 • Dominant higher twist terms evolve in Q just as twist-2 terms • Perturbative charm cancels at Q ~ m c up to higher order in α s • Assume factorization holds .

Parametrizations of c(x,Q 0 ) 1. “Valence-like” intrinsic charm: BHPS1 and BHPS2 (Brodsky et al PLB 1980)  1 � c ( x ) = 1 2 Ax 2 3(1 − x ) (1 + 10 x + x 2 ) − 2 x (1 + x ) ln (1 /x ) c ( x ) = ¯ 2. “BHPS3 model”: compute numerical solutions to BHPS model with physical masses, and include uubar, ddbar, ccbar intrinsic components => gives physical behavior of c/ubar, c/dbar at large x 3. “Sea-like” models SEA1, SEA2 ⇥ ¯ ⇤ c ( x ) = ¯ c ( x ) = A d ( x, Q 0 ) + ¯ u ( x, Q 0 )

Models of IC in the proton IC models in the CT14IC fit, Q = 1.3 GeV, N f = 4 IC models in the CT14IC fit, Q = 2 GeV, N f = 4 0.02 0.02 1000*CT14-pert. CT14-pert. BHPS1 BHPS1 BHPS2 BHPS2 BHPS3 BHPS3 0.015 0.015 SEA1 SEA1 SEA2 SEA2 xc(x,Q) xc(x,Q) 0.01 0.01 0.005 0.005 0 0 0.001 0.01 0.1 1 0.001 0.01 0.1 1 x x There is a sizable perturbative contribution, • Valence-like models: BHPS1, BHPS2, BHPS3 just from evolving from 1.3 to 2.0 GeV. • Sea-like models: SEA1, SEA2 • Perturbative: No IC contribution

<x> IC for various models 120 120 CT14 Q 0 =1.3GeV CT14HERA2 Q 0 =1.3GeV 100 100 BHPS BHPS BHPS + Tier-2 BHPS + Tier-2 80 80 SEA SEA 60 SEA + Tier-2 60 SEA + Tier-2 ∆χ 2 ∆χ 2 40 40 SEA2 BHPS2 20 20 SEA1 0 0 BHPS1 -20 -20 BHPS3 -40 -40 0 0.01 0.02 0.03 0 0.01 0.02 0.03 < x > IC < x > IC 1 • Defining: at Q 0 =1.3 GeV ∫ x IC = x c ( x , Q c ) + c ( x , Q c ) [ ] dx 0 • 90% CL limits for CT14 and CT14HERA2 fits, using charm pole mass m c =1.3 GeV:

Dependence on charm mass 120 120 CT14, SEA Q 0 =1.0 GeV CT14, BHPS Q 0 =1.0 GeV 100 100 m c =1.5 GeV m c =1.5 GeV m c =1.4 GeV m c =1.4 GeV 80 80 m c =1.3 GeV m c =1.3 GeV 60 60 m c =1.2 GeV m c =1.2 GeV m c =1.1 GeV m c =1.1 GeV ∆χ 2 ∆χ 2 40 40 20 20 0 0 -20 -20 -40 -40 0 0.01 0.02 0.03 0 0.01 0.02 0.03 < x > IC < x > IC • CT14NNLO (no IC) fits prefer light MSbar charm mass m c ( m c ) ~ 1.2-1.3 GeV • Exact value depends slightly on gluon parametrization. • BHPS models are roughly independent of m c . • SEA models with larger m c allow associated larger < x > IC .

Impact of IC on PDFs 5.0 2.0 c(x,Q) at Q =2.0 GeV 90% C.L. c(x,Q) at Q =100.0 GeV 90% C.L. CT14NNLO CT14NNLO PDF Ratio to CT14NNLO PDF Ratio to CT14NNLO 4.0 BHPS1 BHPS1 BHPS2 BHPS2 1.5 BHPS3 BHPS3 3.0 SEA1 SEA1 SEA2 SEA2 2.0 1.0 1.0 0.0 0.5 10 -4 10 -3 10 -1 0.2 10 -4 10 -3 10 -1 0.2 10 -2 10 -2 0.5 0.9 0.5 0.9 x x • At low Q : • SEA models uniformly above CT14 • BHPS models dominant at high x • At high Q , perturbative contribution dominates at low x

Impact of IC on gg luminosity L gg at E cm =8.0 TeV, 68% C.L. L gg at E cm =13.0 TeV, 68% C.L. 1.2 1.2 CT14NNLO CT14NNLO BHPS1 BHPS1 BHPS2 BHPS2 Ratio to CT14NNLO Ratio to CT14NNLO 1.1 1.1 BHPS3 BHPS3 SEA1 SEA1 SEA2 SEA2 1.0 1.0 0.9 0.9 0.8 0.8 10 1 10 2 10 3 10 1 10 2 10 3 [GeV] M [GeV] M X X • Sea models: gg luminosity suppressed at low M X , enhanced at high M X • BHPS models: suppressed at high M X • Impact on Higgs cross section small. • Sizable impact for high mass gg luminosities, but still within uncertainties.

Impact on LHC observables 9.5 50 BHPS1 BHPS1 LHC 13 TeV LHC 13 TeV BHPS2 BHPS2 CT14 NNLO 90% C.L., m pole c =1.3 GeV BHPS3 BHPS3 H(gluon fustion) [pb] SEA1 SEA1 9.0 SEA2 SEA2 45 W - [nb] 8.5 40 1603.09222 ATLAS-CONF-2016-081 CT14 NNLO 90% C.L., m pole c =1.3 GeV 1603.09222 8.0 11.0 11.5 12.0 1850 1900 1950 2000 2050 W + [nb] Z [pb] • Impact on inclusive Higgs, Z, W + , W - cross sections is mild.

m c and <x> IC dependence 45 9.0 LHC 13 TeV LHC 13 TeV CT14 NNLO 90% C.L., m pole CT14 NNLO 90% C.L., m pole c =1.3 GeV c =1.3 GeV 8.8 H(gluon fustion) [pb] 44 <x> IC =3% 8.6 W - [nb] BHPS m pole 43 c =1.1 m pole c =1.5 8.4 m pole <x> IC =0% m pole c =1.5 c =1.1 CT14IC BHPS, Q 0 =1.0 GeV 42 8.2 • LHC 13 TeV CT14IC BHPS, Q 0 =1.0 GeV • CT14NNLO 41 8.0 1850 1900 1950 11.0 11.2 11.4 11.6 11.8 12.0 W + [nb] @ 90%CL Z [pb] 45 9.0 • m c = 1.1-1.5 GeV LHC 13 TeV LHC 13 TeV CT14 NNLO 90% C.L., m pole CT14 NNLO 90% C.L., m pole c =1.3 GeV c =1.3 GeV • < x > IC = 0 - 3% 8.8 H(gluon fustion) [pb] 44 8.6 CT14IC SEA, Q 0 =1.0 GeV W - [nb] m pole SEA c =1.1 43 <x> IC =0% m pole m pole m pole c =1.5 c =1.1 c =1.5 8.4 42 8.2 CT14IC SEA, Q 0 =1.0 GeV <x> IC =3% 41 8.0 11.0 11.2 11.4 11.6 11.8 12.0 1850 1900 1950 W + [nb] Z [pb]

Z+c @ LHC 13 TeV, NLO Fixed order Parton Shower with Sherpa Z Sherpa CT14nnlo 1 − d /dp (pb/GeV) σ 10 − 1 10 Sherpa BHPS2 T Sherpa BHPS3 − 2 LHC 13 TeV 10 2 − Sherpa SEA1 10 Sherpa SEA2 MCFM CT14nnlo 3 − 3 10 − 10 CT14nnlo BHPS2 4 − 10 4 − 10 Z BHPS3 d /dp (pb/GeV) σ T 5 SEA1 − 10 − 5 10 LHC 13 TeV SEA2 2 1.5 Ratio to Sherpa CT14nnlo CT14nnlo PDF unc. 1.8 1.4 1.6 Ratio to CT14nnlo 1.3 1.4 1.2 1.2 1.1 1 1 0 100 200 300 400 500 600 0 100 200 300 400 500 600 Z p (GeV) Z p (GeV) T T • High pT excess of BHPS models is dampened by parton showers.

CT17p : new data to be included • Combined HERA 1+2 DIS update ✔ ︎ • LHCb 7 TeV Z,W muon rapidity distribution update ✔ ︎ • LHCb 8 TeV Z,W muon rapidity distribution update ✔ ︎ • ATLAS 7 TeV inclusive jet update ✔ ︎ • CMS 7 TeV inclusive jet (extended y range) update ✔ ︎ • ATLAS 7 TeV Z pT distribution NEW ✔ ︎ • LHCb 13 TeV Z rapidity distribution update • CMS 8 TeV Z pT and rapidity distribution (double diff.) NEW • CMS 8 TeV W muon rapidity distribution update • ATLAS 7 TeV W/Z lepton rapidity distribution update • CMS 7,8 TeV tT differential distribution NEW • ATLAS 7,8 TeV tT differential distribution NEW

ATLAS 8 TeV Data Data: Eur. Phys. J. C76 (2016) 538, arXiv: 1511.04716 Measurements of top-quark pair di ff erential cross-sections in the lepton + jets ATLAS 8 TEV ttbar Data channel in pp collisions at 8 TeV using the ATLAS detector • Eur. Phys. J. C76 (2016) 538, arXiv: 1511.04716 ATLAS 8 TeV Data • M ttbar , p t T , |y t |, |y ttbar | Data: Eur. Phys. J. C76 (2016) 538, arXiv: 1511.04716 Measurements of top-quark pair di ff erential cross-sections in the lepton + jets Including �� distribution data channel in pp collisions at 8 TeV using the ATLAS detector • Micha� Cza���, David Hey�es a�d A�exa�der Mit�v re�ease fastNLO tab�es with NNLO QCD top-quark pair production at 8TeV arXiv: 1704.08551 • Theory: Michal Czakon, David Heymes and Alexander Mitov • fastNLO tables with NNLO QCD, arXiv:1704.08551 |� �� | Shows the least agreement with NNLO prediction, with MRST2008nnlo PDFs. �� • Micha� Cza���, David Hey�es a�d A�exa�der Mit�v re�ease fastNLO tab�es with NNLO QCD top-quark pair production at 8TeV arXiv: 1704.08551

Correlation with g-PDF @ 100 GeV Correlation of -PDF @ 100 GeV � (t) �� ��� X X X Strong correlation of g-PDF found in large x region. Strong correlation at large x

Estimating impact of new data ePump: error PDF updating method package - a package for (very quickly) estimating impact of new data using Hessian eigenvector PDFs - C.-P. Yuan, J. Pumplin, CS: arXiv: 1708.xxxxx - based on ideas of H. Paukkunen and P. Zurita [1402.6623]