CTEQ SEA-TAC: Three puzzles of large-x hadron scattering

Transcription

1 CTEQ SEA-TAC: Three puzzles of large-x hadron scattering Pavel Nadolsky Southern Methodist University Dallas, TX, U.S.A. in collaboration with M. Guzzi, J. Huston, H.-L. Lai, Z. Li, F. Olness, J. Pumplin, D. Stump, and C.-P. Yuan October 18, 2010 Pavel Nadolsky (SMU) INT-3 workshop, Seattle October 18,

2 Outline CT10 analysis of parton distributions (arxiv: , accepted to Phys. Rev. D) Three puzzles concerning the PDFs at x > 0.1 (may be resolved at a future ep/ea collider): d(x, Q)/u(x, Q) (arxiv: ) g(x, Q) (arxiv: ) hypothetical intrinsic charm c(x, Q) (M. Guzzi, P. N., F. Olness, in preparation) Pavel Nadolsky (SMU) INT-3 workshop, Seattle October 18,

3 CTEQ-Tung Et Al.: ongoing activities NLO general-purpose PDF fits CTEQ6.6 set (arxiv: ) CT09 (arxiv: , not released) CT10 (arxiv: ) Uncertainty in α s in the CTEQ PDF analysis (arxiv: ) PDFs for leading-order Monte-Carlos (arxiv: ) Benchmarking of heavy-quark contributions (2010 Les Houches Proceedings) Constraints on color-octet fermions (arxiv:1010.xxxx) Exploration of statistical methods and PDF parametrizations (Pumplin, arxiv: and ) Pavel Nadolsky (SMU) INT-3 workshop, Seattle October 18,

4 CT10 parton distribution functions (arxiv: ) Overview new experimental data, statistical methods, and parametrization forms combined HERA-1 data are included Version 2 (Aug. 20, 2010): extended discussion of the Tevatron Run-2 W asymmetry data search for deviations from NLO DGLAP evolution in the small-x HERA data Pavel Nadolsky (SMU) INT-3 workshop, Seattle October 18,

5 CT10 parton distribution functions (arxiv: ) New PDF sets in the LHAPDF library and on the CTEQ website CT10 and CT10W PDF best-fit sets and 44 eigenvector sets for α s (M Z ) = CT10AS PDFs for α s (M Z ) = sufficient for calculating the PDF+α s uncertainty (as explained in arxiv: ) CT10/CT10W PDFs with 3 and 4 active flavors Pavel Nadolsky (SMU) INT-3 workshop, Seattle October 18,

6 CT10 parton distribution functions (arxiv: ) Experimental data Combined HERA-1 neutral-current and charged-current DIS data with 114 correlated systematic effects replaces 11 separate HERA-1 sets used in the CTEQ6.6 fit CDF Run-2 and D0 Run-2 inclusive jet production Tevatron Run-2 Z rapidity distributions from both CDF and D0 W electron asymmetry from CDF II and D0 II; W muon asymmetry from D0 II (CT10W set) Other data sets inherited from CTEQ6.6 Pavel Nadolsky (SMU) INT-3 workshop, Seattle October 18,

7 CT10 parton distribution functions (arxiv: ) Developments in statistical techniques Experimental normalizations N i are treated on the same footing as other correlated systematic errors Minimum of χ 2 with respect to N i is found algebraically normalization shifts are automatically accounted for when producing the eigenvector sets Set data weights of 1 for all experiments do not prefer some experiments over the other experiments Exception: Run-2 W asymmetry data (see below) Pavel Nadolsky (SMU) INT-3 workshop, Seattle October 18,

8 CT10 parton distribution functions (arxiv: ) Revised functional forms at the input scale More data constraints more flexible (=less biased) parametrizations for g(x, Q 0 ), d(x, Q 0 ), and s(x, Q 0 ) R s = lim x 0 (s(x) + s(x)) / ( ū(x) + d(x) ) is not constrained by the data large uncertainty in s(x) at x 0 allow R s to vary in the fit, but softly constrain it by a penalty on χ 2 to satisfy 0.4 < R s < 1 Pavel Nadolsky (SMU) INT-3 workshop, Seattle October 18,

9 More flexible parametrizations CT10(green) vs. CT6.6(blue) g at Q 2 GeV CT66 CT6.6M,CT10 CT6.6M CT66 CT6.6M,CT10 CT6.6M x s at Q 2 GeV x g(x, Q): large uncertainty at x < 10 3, despite tighter constraints by the combined HERA data s(x, Q): wider uncertainty, covers both CTEQ6.6 and MSTW 08 Pavel Nadolsky (SMU) INT-3 workshop, Seattle October 18,

10 CT10: agreement between data sets Good overall agreement: χ 2 /d.o.f. = 3000/2750 = 1.1 Some tensions are also observed: D0 Run-2 W lepton asymmetry: apparently disagrees with constraints on d/u by the NMC and BCDMS data on F p,d 2 (x, Q) Tevatron single-inclusive jet production: Run-1 and Run-2 sets are moderately (in)compatible (arxiv: ) Two series of PDFs are produced: CT10: D0 Run-2 A l (y l ) are not included CT10W: with D0 Run-2 A l (y l ), with an extra χ 2 weight Theoretical cross sections: computed by ResBos; include the most important part of resummed NNLO corrections Pavel Nadolsky (SMU) INT-3 workshop, Seattle October 18,

11 Puzzle 1: the W lepton asymmetry Pavel Nadolsky (SMU) INT-3 workshop, Seattle October 18,

12 The puzzle of the W lepton asymmetry Rapidity asymmetry A l (y l ) of charged lepton l = e or µ in W boson decay provides important constraints on d(x, Q)/u(x, Q) at x > 0.1 A l (y l ) = dσ(p p (W + l + ν l )X)/dy l dσ(p p (W l ν l )X)/dy l dσ(p p (W + l + ν l )X)/dy l + dσ(p p (W l ν. l )X)/dy l A l (y l ) is related to the boson-level asymmetry, A W (y W ) = dσ(pp W + X)/dy W dσ(pp W X)/dy W dσ(pp W + X)/dy W + dσ(pp W X)/dy W, smeared by W ± decay effects Berger, Halzen, Kim, Willenbrock, PRD 40, 83 (1989); Martin, Roberts, Stirling, 1989 Pavel Nadolsky (SMU) INT-3 workshop, Seattle October 18,

13 The puzzle of the W lepton asymmetry NLO calculations based on recent PDFs fail in describing the most precise A e (y e ) and A µ (y µ ) from D0 Run-2 (2008) and prel. CDF Run-2 (2010) Agreement of Order PQCD with D0 A e (y e ) of α s χ 2 /npt Source CTEQ6.6 NLO 191/36=5.5 Our study CT10W NLO 78/36=2.2 ABKM 09 NNLO 540/24=22.5 Catani, Ferrera, Grazzini, MSTW 08 NNLO 205/24=8.6 JHEP 05, 006 (2010) JR09VF NNLO 113/24=4.7 The puzzle: why do these decent PDFs fail to describe A l (y l )? Pavel Nadolsky (SMU) INT-3 workshop, Seattle October 18,

14 The puzzle of the W lepton asymmetry Reconstructed boson-level asymmetry A ch (y W ) The CDF Run-2 reconstructed the boson-level A W (y W ) from A e (y e ) (PRL 102, (2009)) It agrees well with CTEQ6.1, despite the disagreement of CTEQ6.1 with A e (y e ) (???) A W (y W ) is not included in CT10 W Charge Asymmetry W Charge Asymmetry CDF 1 fb data(stat. + syst.) NLO Prediction(CTEQ6.1M) PDF uncertainty(cteq6.1m) -1 CDF 1 fb data(stat. + syst.) NNLO Prediction(MRST2006NNLO) PDF uncertainty(mrst2006nnlo) y W Pavel Nadolsky (SMU) INT-3 workshop, Seattle October 18,

15 The puzzle of the W lepton asymmetry Reconstructed boson-level asymmetry A ch (y W ) The CDF Run-2 reconstructed the boson-level A W (y W ) from A e (y e ) (PRL 102, (2009)) It agrees well with CTEQ6.1, despite the disagreement of CTEQ6.1 with A e (y e ) (???) Charge Asymmetry l E T > 25 GeV -1 CDF Run II Preliminary 1 fb (no systematic uncertainties) -1 DØ 0.75 fb data (e) -1 DØ Run II Preliminary 5 fb RESBOS with CTEQ6.6 data (e) data (µ) Lepton η A W (y W ) is not included in CT10 Pavel Nadolsky (SMU) INT-3 workshop, Seattle October 18,

16 How we calculate A l (y l ) Selection cuts on p l T of l emphasize sensitivity to d(x, µ)/u(x, µ) in different x ranges Default calculation: A l (y) at NNLL-NLO, using lookup tables for σ(p l T, y l) NNLL+NLO /σ(p l T, y l) LO from ResBos [Balazs, Yuan, PRD 56, 5558 (1997); Landry, Brock, P.N. Yuan, PRD67, (2003)]. Cross check: include NNLO corrections at Q T M W [Arnold & Reno, 1989]; A l (y l ) changes by a few percent at the highest y l and p T > 35 GeV magnitude of changes is comparable with full NNLO terms (Catani, Ferrera, and Grazzini, JHEP 05, 006 (2010)) changes are small compared to the experimental errors Pavel Nadolsky (SMU) INT-3 workshop, Seattle October 18,

17 Comparison of CT10 and CT10W fits CDF Run-1 and Run-2 (207 pb 1 ) A l (y l ) data agree with F p 2 (x, Q)/F 2 d (x, Q) Good fits to D0 Run-2 A e (y e ) are possible without NMC and BCDMS; and vice versa No acceptable fit to D0 Run-2 A e (y e ) and NMC/BCDMS data can be achieved, if they are included on the same footing Tension between Run-2 e asymmetry and D0 Run-2 µ asymmetry Reasonable agreement between Run-2 e W asymmetry data and Z y data Pavel Nadolsky (SMU) INT-3 workshop, Seattle October 18,

18 Comparison of CT10 and CT10W fits With special emphasis on D0 Run-2 A e (y e ) data (χ 2 weight>1), it is possible to obtain a reasonable agreement with A e (y e ) and one bin of A µ (y µ ) (χ 2 /d.o.f. = 1 2), with some remaining tension with NMC & BCDMS F2 d(x, Q) and F p 2 (x, Q) at x > 0.4 Pavel Nadolsky (SMU) INT-3 workshop, Seattle October 18,

19 CT10 and CT10W predictions for A e (y e ) (D0 Run-2) ) e (y A e ± pp W e p T ν E T e ± > 25 GeV > 25 GeV ν+x S=1.96 TeV ) e (y A e e p T ν E T ± pp W > 25 GeV > 25 GeV e ± ν+x S=1.96 TeV D0 electron data (0.75 fb ) CT10 (Solid band) CTEQ6.6 (Hatched band) D0 electron data (0.75 fb ) CT10W (Solid band) CTEQ6.6 (Hatched band) y e y e ) e (y A e e 25 GeV < p T ± pp W ν E T e ± > 25 GeV ν+x < 35 GeV S=1.96 TeV ) e (y A e e 25 GeV < p < 35 GeV T ν E T > 25 GeV ± pp W e ± ν+x S=1.96 TeV D0 electron data (0.75 fb ) CT10 (Solid band) CTEQ6.6 (Hatched band) y e D0 electron data (0.75 fb ) CT10W (Solid band) CTEQ6.6 (Hatched band) y e ) e (y A e e p T ν E T ± pp W > 35 GeV > 25 GeV e ± ν+x S=1.96 TeV ) e (y A e ± pp W e p T ν E T > 35 GeV > 25 GeV e ± ν+x S=1.96 TeV D0 electron data (0.75 fb ) CT10 (Solid band) CTEQ6.6 (Hatched band) D0 electron data (0.75 fb ) CT10W (Solid band) CTEQ6.6 (Hatched band) y e y e Pavel Nadolsky (SMU) INT-3 workshop, Seattle October 18,

20 CT10 and CT10W predictions for A µ (y µ ) (D0 Run-2) ) µ (y µ p T ν E T > 20 GeV > 20 GeV ) µ (y µ p T ν E T > 20 GeV > 20 GeV A µ -0.4 ± pp W µ ± ν+x S=1.96 TeV A µ -0.4 ± pp W µ ± ν+x S=1.96 TeV D0 muon data (4.9 fb ) CT10 (Solid band) CTEQ6.6 (Hatched band) D0 muon data (4.9 fb ) CT10W (Solid band) CTEQ6.6 (Hatched band) y µ y µ ) µ (y A µ µ 20 GeV < p T ± pp W ν E T > 20 GeV µ ± < 35 GeV ν+x S=1.96 TeV ) µ (y A µ µ 20 GeV < p T ± pp W ν E T > 20 GeV µ ± < 35 GeV ν+x S=1.96 TeV D0 muon data (4.9 fb ) CT10 (Solid band) CTEQ6.6 (Hatched band) D0 muon data (4.9 fb ) CT10W (Solid band) CTEQ6.6 (Hatched band) y µ y µ ) µ (y A µ µ p T ν E T ± pp W > 35 GeV > 20 GeV µ ± ν+x -1 D0 muon data (4.9 fb ) CT10 (Solid band) CTEQ6.6 (Hatched band) S=1.96 TeV y µ ) µ (y A µ ± pp W µ p T ν E T µ ± > 35 GeV > 20 GeV ν+x -1 D0 muon data (4.9 fb ) CT10W (Solid band) CTEQ6.6 (Hatched band) S=1.96 TeV y µ Pavel Nadolsky (SMU) INT-3 workshop, Seattle October 18,

21 A hint for the solution A l (y l ) is a sensitive probe of the slope s du of d(x, M W )/u(x, M W ) At LO: 1 A l (y l ) A l (y W ) x 1 x 2 [ d(x1, M W ) u(x 1, M W ) d(x ] 2, M W ) = s du u(x 2, M W ) x 1,2 = Q s e ±y W Small changes in s du cause significant variations in A l Berger, Halzen, Kim, Willenbrock, PRD 40, 83 (1989); Martin, Stirling, Roberts, MPLA 4, 1135 (1989); Phys. Rev. D50, 6734 (1994); Lai et al., PRD 51, 4763 (1995) Pavel Nadolsky (SMU) INT-3 workshop, Seattle October 18,

22 A hint, continued Small changes in s du cause significant variations in A l F2 d(x, Q)/F p 2 (x, Q) does not constrain s du in such direct way Lai et al., PRD 51, 4763 (1995) Pavel Nadolsky (SMU) INT-3 workshop, Seattle October 18,

23 d(x, Q)/u(x, Q) at Q = 85 GeV d/u at µ = 85 GeV d/u at µ = 85 GeV 1.3 Solid band: CTEQ6.6 uncertainty Hatched band: CT10 uncertainty 1.3 Solid band: CTEQ6.6 uncertainty Hatched band: CT10W uncertainty Ratio to CTEQ6.6M Ratio to CTEQ6.6M x CT10W prefers a larger slope of d/u, has smaller uncertainty than CTEQ6.6 or CT10 Pavel Nadolsky (SMU) INT-3 workshop, Seattle October 18, x

24 Theoretical uncertainties in large-x deuteron DIS Accardi, Christy, Keppel, Monaghan, Melnitchouk, Morfin, Owens, PRD 81, (2010) d(x)/u(x) from F p,d 2 (x, Q) at x > 0.1 and low Q is sensitive to nuclear, target-mass, and dynamic higher-twist corrections Pavel Nadolsky (SMU) INT-3 workshop, Seattle October 18,

25 CT10 & CT10W predictions for the LHC W ± and Z 0 total cross sections W ± & Z cross sections at the LHC 7 TeV NLO W ± & Z cross sections at the Tevatron Run 2 NNLL NLO ResBos 0.26 Σ tot pp Z 0 X nb CTEQ6.1 CT10W CTEQ6.6 CT10 Σ tot pp Z 0 X nb CTEQ6.6 CT10 CT10W Σ tot pp W ± Ν X nb Σ tot pp W ± Ν X nb Pavel Nadolsky (SMU) INT-3 workshop, Seattle October 18,

26 CT10 & CT10W predictions for the LHC & Tevatron W + and W total cross sections 4.3 W & W cross sections at the LHC 7 TeV W & W cross sections at the LHC 14 TeV Σ tot pp W Ν X nb CTEQ6.1 CT10W CT10 CTEQ6.6 Σ tot pp W Ν X nb CT10W CT10 CTEQ6.1 CTEQ Σ tot pp W ΝX nb Σ tot pp W Ν X nb Pavel Nadolsky (SMU) INT-3 workshop, Seattle October 18,

27 CT10 & CT10W predictions for the LHC σ(w + )/σ(w ) and σ(w ± )/σ(z 0 ) vs. y W/Z Ratio to CTEQ6.6M (dσ(w+)/dy)/(dσ(w )/dy), LHC at 7 TeV CT10 (Solid band) 0.9 CT10W (Skew hatched band) CTEQ6.6 (Vertical hatched band) y (dσ(w±)/dy)/(dσ(z)/dy), LHC at 7 TeV Ratio to CTEQ6.6M CT10 (Solid band) CT10W (Skew hatched band) 0.92 CTEQ6.6 (Vertical hatched band) y Pavel Nadolsky (SMU) INT-3 workshop, Seattle October 18,

28 Puzzle 2: gluon at x > The Tevatron data on p + p ( ) jet + X and p + p ( ) jet + jet + X allow g(x, Q) > u(x, Q) > d(x, Q) at x > 0.5. Is such valence-like gluon unphysical? 2. How large is the uncertainty on g(x, Q) at such x? Is a harder g(x, Q) preferred by post-cteq5 PDFs ruled out by the Tevatron Run-2 jet data? Constraints on CT09/CT10 PDFs by the Tevatron Run-2 jet data and current limitations are explored in arxiv: , arxiv: Pavel Nadolsky (SMU) INT-3 workshop, Seattle October 18,

29 At x > 0.1, g(x, Q) can exceed d(x, Q) and even u(x, Q) CT09 (arxiv: ): u(x, Q) (black), d(x, Q) (blue), g(x, Q) (red) Pavel Nadolsky (SMU) INT-3 workshop, Seattle October 18,

30 Inclusive jet production in the Tevatron Run-2 data / theory DØ Run II -1 L = 0.70 fb y < 0.4 R cone = 0.7 NLO scale uncertainty < y < NLO pqcd µ = µ = p R F T +non-perturbative corrections 0.4 < y < 0.8 CTEQ6.5M with uncertainties MRST < y < Data Systematic uncertainty 0.8 < y < < y < p (GeV) T D0 Coll., arxiv: (700 pb 1 ); CDF results (1.13 fb 1 ) (Almost) negligible statistical error MidCone/k T algorithm samples, corrected to parton level D0 paper: There is a tendency for the data to be lower than the central CTEQ prediction......but they lie mostly within the CTEQ uncertainty band Pavel Nadolsky (SMU) INT-3 workshop, Seattle October 18,

31 Dijet mass distributions from D0 Run fb 1, arxiv: The data appear to favor smaller cross sections than the CTEQ6.6 NLO prediction, for the selected theory parameters Pavel Nadolsky (SMU) INT-3 workshop, Seattle October 18,

32 However... Ratio of dσ/dm JJ Ratio of dσ/dm JJ Scale = 0.5 average p T of jets y < 0.4 max CTEQ6.6 / CT10.00 CT10 / CT10.00 CT10W / CT M JJ (GeV) Scale = 0.5 average p T of jets 1.6 < y max CTEQ6.6 / CT10.00 CT10 / CT10.00 CT10W / CT10.00 < M JJ (GeV) dσ/dp T depends strongly on the factorization scale CTEQ PDFs are fitted to the single-jet data with a scale µ = p jet T /2; MSTW PDFs with µ = p jet T DO compares their dijet data to NLO theory with µ p jet T only (incompatible with CTEQ) If µ = p jet T /2 is used, the agreement with all CTEQ PDFs improves Pavel Nadolsky (SMU) INT-3 workshop, Seattle October 18,

33 However... Ratio of dσ/dm JJ Ratio of dσ/dm JJ Scale = 0.5 average p T of jets y < 0.4 max CTEQ6.6 / CT10.00 CT10 / CT10.00 CT10W / CT M JJ (GeV) Scale = 0.5 average p T of jets 1.6 < y max CTEQ6.6 / CT10.00 CT10 / CT10.00 CT10W / CT10.00 < M JJ (GeV) Correlated experimental systematic shifts obtained in a PDF fit are different than those in the experimental analysis tend to reconcile the data with a wider range of PDFs than a standalone experimental analysis may suggest Pavel Nadolsky (SMU) INT-3 workshop, Seattle October 18,

34 Impact of Run-2 jet data on CT09 fit CT09 fit includes all four Run-1 and Run-2 jet data samples Excellent quality of the fit: χ 2 = 2756 for 2898 data points (Shifted CDF-Run 2 data)/ct09 Pavel Nadolsky (SMU) INT-3 workshop, Seattle October 18,

35 Impact of Run-2 jet data on CT09 fit CT09 fit includes all four Run-1 and Run-2 jet data samples Excellent quality of the fit: χ 2 = 2756 for 2898 data points (Shifted D0-Run 2 data)/ct09 Pavel Nadolsky (SMU) INT-3 workshop, Seattle October 18,

36 Tevatron Run-2 jet data: CT09 vs. CTEQ6.6 Gluon PDF: CT09 (red), CT66 (blue) CT09 PDF uncertainty is about the same as CT66 (compensation between the Run-2 jet constraints and more flexible g(x, µ)) Pavel Nadolsky (SMU) INT-3 workshop, Seattle October 18,

37 Constraints of Run-2 data on CT09 PDFs Black: g(x, µ) from CT09 fit Red: The same fit without Run-2 jet data Best-fit PDFs preferred by each jet experiment: CDF-1 (magenta), D0-1 (green), CDF-2 (blue), D0-2 (black) Pavel Nadolsky (SMU) INT-3 workshop, Seattle October 18,

38 Puzzle 3: Intrinsic charm production at future ep colliders Pavel Nadolsky, Frederick Olness and Marco Guzzi in progress Pavel Nadolsky (SMU) INT-3 workshop, Seattle October 18,

39 Intrinsic charm Many nonperturbative models predict an intrinsic charm (IC) component of the nucleon at µ m c (Brodsky, Hoyer, Peterson, Sakai; Vogt; Gutierrez and Vogt; Ingelman and Thunman;...) The IC component would participate fully in QCD dynamics and evolve along with the other partons as the energy scale increases (see,.e.g., J. Pumplin et. al. Phys.Rev.D75, (2007)). Pavel Nadolsky (SMU) INT-3 workshop, Seattle October 18,

40 IC production in future DIS experiments ep/ea colliders with luminosities of cm 2 sec 1 and s = GeV are well-suited for exploring charm production at large x and Q of a few GeV distinguishing between the IC models suggested by light-cone QCD (with IC concentrated at x > 0.1) and sea-like IC models (with IC contributing at all x) Pavel Nadolsky (SMU) INT-3 workshop, Seattle October 18,

41 ep e + c + X cross section ep ecx NC e p DIS at s = 12, 45, 105 GeV and Q =3 and 25 GeV. Reduced cross section: where Q 2 sxy. σ(x, Q 2 ) = Number of events: d2 σ dq 2 dx (1 + (1 y)2 ) 1 Q4 x 2πα 2, dn e dx = L dσ dx where dσ dx is the average cross section in a Q-bin of 0.15 GeV; L = 10 fb 1. Pavel Nadolsky (SMU) INT-3 workshop, Seattle October 18,

42 EIC is well-suited for searching for BHPS-like IC contributions c(x, µ = 2 GeV) in CTEQ6.5 set (blue) and BHPS IC (red) model. The long-dash (short-dash) curve corresponds to x c+ c = 0.57% (2.0%). Pavel Nadolsky (SMU) INT-3 workshop, Seattle October 18,

43 EIC is well-suited for searching for BHPS-like IC contributions Reduced cross section σ NC,er (x,q2 ) e - p DIS, s = 12 GeV, Q 2 = 9 GeV 2 CTEQ6.6 BHPS, mom. frac.= 0.01 BHPS, mom. frac.= sea-like, mom. frac.= 0.01 sea-like, mom. frac.= < y < 0.89 L dt = 10 fb -1 Tagging efficiency = dn e / dx x 10 3 Pavel Nadolsky (SMU) INT-3 workshop, Seattle October 18,

44 EIC is well-suited for searching for BHPS-like IC contributions Reduced cross section σ NC,er (x,q2 ) e - p DIS, s = 45 GeV, Q 2 = 9 GeV < y < 0.22 L dt = 10 fb -1 Tagging efficiency = 1 CTEQ6.6 BHPS, mom. frac.= 0.01 BHPS, mom. frac.= sea-like, mom. frac.= 0.01 sea-like, mom. frac.= x dn e / dx Pavel Nadolsky (SMU) INT-3 workshop, Seattle October 18,

45 EIC is well-suited for searching for BHPS-like IC contributions Reduced cross section σ NC,er (x,q2 ) e - p DIS, s = 105 GeV, Q 2 = 625 GeV < y < 0.97 L dt = 10 fb -1 Tagging efficiency = 1 CTEQ6.6 BHPS mom. frac.= 0.01 BHPS mom. frac.= sea-like mom. frac.= 0.01 sea-like mom. frac.= x dn e / dx Pavel Nadolsky (SMU) INT-3 workshop, Seattle October 18,

46 My wishlist for future ep/ea DIS experiments DIS on proton, neutron, nuclear targets at x > 0.1: determine d(x, Q)/u(x, Q) in the proton and its uncertainty; g(x, Q) test for the presence of valence-like gluons (Υ production?) Intrinsic charm and bottom production x < 10 2 : SIDIS, transition from DGLAP to BFKL/saturation dynamics All x: strangeness PDFs in (SI)DIS relevant for W mass measurement at the LHC All x: QED contributions to proton structure relevant for electroweak precision physics at the LHC isospin violation; f u/p (x, Q) f d/n (x, Q) photon PDFs in the proton Pavel Nadolsky (SMU) INT-3 workshop, Seattle October 18,

47 Backup slides Pavel Nadolsky (SMU) INT-3 workshop, Seattle October 18,

48 Confidence intervals in global PDF analyses CTEQ6 tolerance criterion (2001) acceptable values of PDF parameters must agree at 90% c.l. with all experiments included in the fit, for a plausible range of theoretical assumptions is realized by accepting all PDF fits with χ 2 < T modified in the CT10 fit 40 Eigenvector 4 40 Eigenvector BCDMSp BCDMSd H1a H1b ZEUS NMCp NMCr CCFR2 CCFR3 E605 CDFw E866 D0jet CDFjet BCDMSp BCDMSd H1a H1b ZEUS NMCp NMCr CCFR2 CCFR3 E605 CDFw E866 D0jet CDFjet distance 0 10 distance Pavel Nadolsky (SMU) INT-3 workshop, Seattle October 18,

49 Fisher s pseudo-gaussian distribution A variable S(N pt ) = 2χ 2 (N pt ) 2N pt + 1 is closely described by a normal Gaussian distribution ( S = 0, σ S = 1) for N pt 10 (R. A. Fisher, 1924) S(N (i) (i) pt ) from fully compatible experiments with N pt points each would follow a normal Gaussian distribution regardless of N (i) pt. In reality, the S-distribution is described by a Gaussian with σ S = 2 3 Pavel Nadolsky (SMU) INT-3 workshop, Seattle October 18,

50 Distribution of S from 29 experiments For CT10 PDFs For unpublished CT10 PDFs based on Chebyshev polynomials (70 parameters) Pavel Nadolsky (SMU) INT-3 workshop, Seattle October 18,

51 CT10 & CT10W predictions for the LHC & Tevatron Total cross sections + W Z + W (300) + W (600) Z (300) Z (600) tt t (s-channel) t (t-channel) gg H (120) gg H (160) gg H (250) VBF H (120) CT10 CTEQ6.6 CT10W Tevatron + W - W Z + W (300) + W (600) Z (300) Z (600) tt t (s-channel) t (t-channel) gg H (120) gg H (160) gg H (250) VBF H (120) CT10 CTEQ6.6 CT10W LHC 7 TeV VBF H (160) VBF H (160) VBF H (250) VBF H (250) HZ HZ + HW - HW + H + HW - HW + H Pavel Nadolsky (SMU) INT-3 workshop, Seattle October 18,

52 A cut fits to CT10 data Χ 2 data Npts Fit 1 witha cut 1.5 Fit 2 witha cut 1.5 CT10 fit Q 2GeV A gs = Q 2 x N pts A gs ranges If CT10 PDFs are fitted only to HERA data at A gs > A cut = 1.5, no significant disagreements with the DGLAP-connected data at A gs < A cut are observed, as a result of the increased PDF uncertainty at A gs < A cut Pavel Nadolsky (SMU) INT-3 workshop, Seattle October 18,

53 A cut fits to CT10 data Χ 2 data Npts Fit 1 witha cut 1.5 Fit 2 witha cut 1.5 CT10 fit Q 2GeV A gs = Q 2 x N pts A gs ranges We do not confirm the systematic disagreement of the A cut fits with the low-a gs data stated by NNPDF (Caola, Forte, Rojo, arxiv: , arxiv: ) Pavel Nadolsky (SMU) INT-3 workshop, Seattle October 18,