![11
Including the ATLAS 2011 jet data
Juan Rojo PDF4LHC, CERN, 27/10/2015
The theory/data
comparison depends
strongly on assumptions
in the construction of
experimental covariance
matrix!
!
ATLAS 2011 jets lead to
reduction on large-x
gluon PDF uncertainties,
but need input about
correct treatment of
systematics
x
0.1 0.2 0.3 0.4 0.5 0.6 0.7
)[ref]2
)/g(x,Q2
g(x,Q
0.8
0.9
1
1.1
1.2
1.3
)=0.118Z
(MS
,2
GeV4
=102
NNPDF3.0 NLO, Q
Baseline
+ ATLAS11 jets (uncorr)
+ ATLAS11 jets (corr)
)=0.118Z
(MS
,2
GeV4
=102
NNPDF3.0 NLO, Q
!
ATLAS inclusive jet production from 2011 recently available from HepForge!
NLO theory from NLOjet++/APPLgrid, with pT
jet as central scale!
Information about how to treat systematic errors not available: explore two assumptions about the
experimental covariance matrix: treat all systematics either as fully correlated or as fully uncorrelated!
Non-perturbative and electroweak corrections included in the theory calculation](https://image.slidesharecdn.com/1510-pdf4lhc-nnpdf-151027083615-lva1-app6892/85/News-from-NNPDF-new-data-and-fits-with-intrinsic-charm-11-320.jpg)
News from NNPDF: new data and fits with intrinsic charm
- 1. ! News from Juan Rojo! STFC Rutherford Fellow! Rudolf Peierls Center for Theoretical Physics! University of Oxford! ! ! PDF4LHC Working Group Meeting ! CERN, 27/10/2015 Juan Rojo PDF4LHC, CERN, 27/10/2015 NNPDF
- 2. ! ! ! ! ! ! Juan Rojo PDF4LHC, CERN, 27/10/2015 Recent NNPDF studies (and related work) ! Inclusion of the legacy HERA combined dataset ! Inclusion of new measurements from ATLAS, CMS and LHCb! NNPDF3.0 fits with intrinsic charm, and associated theory calculations! NNPDF3.0 fits with threshold resummation ,and implications for high-mass SUSY xsecs! Impact of LHCb charm data in small-x gluon, and implications for neutrino astronomy! Parton Distributions in a doped VFN scheme! Specialised Minimal PDF sets ! Since the last PDF4LHC meeting …. JR, arXiv:1508.07731, ICHEP15 proceedings NNPDF, in preparation NNPDF, in preparation + Ball, Bertone, Bonvini, Forte, Groth-Merrild, JR, Rottoli, arXiv:1510.00009 Gauld, JR, Rottoli, Talbert, arXiv:1506.08025, + Sarkar, in preparation Bertone, Carrazza, JR, arXiv:1509.04022 Carrazza, Forte, Kassabov, JR, in preparation Bonvini, Marzani, JR, Rottoli, Ubiali, Ball, Bertone, Carrazza, Hartland, arXiv:1507.01006
- 3. ! ! ! ! ! ! Juan Rojo PDF4LHC, CERN, 27/10/2015 Recent NNPDF studies (and related work) ! Inclusion of the legacy HERA combined dataset ! Inclusion of new measurements from ATLAS, CMS and LHCb! NNPDF3.0 fits with intrinsic charm, and associated theory calculations! NNPDF3.0 fits with threshold resummation ,and implications for high-mass SUSY xsecs! Impact of LHCb charm data in small-x gluon, and implications for neutrino astronomy! Parton Distributions in a doped VFN scheme! Specialised Minimal PDF sets ! Since the last PDF4LHC meeting …. JR, arXiv:1508.07731, ICHEP15 proceedings NNPDF, in preparation NNPDF, in preparation + Ball, Bertone, Bonvini, Forte, Groth-Merrild, JR, Rottoli, arXiv:1510.00009 Gauld, JR, Rottoli, Talbert, arXiv:1506.08025, + Sarkar, in preparation Bertone, Carrazza, JR, arXiv:1509.04022 Carrazza, Forte, Kassabov, JR, in preparation Bonvini, Marzani, JR, Rottoli, Ubiali, Ball, Bertone, Carrazza, Hartland, arXiv:1507.01006 This talk JR, Later today Zahari’s talk in the afternoon
- 4. ! ! ! ! ! ! Towards NNPDF3.1:! New experimental data Juan Rojo PDF4LHC, CERN, 27/10/2015
- 5. 5 New LHC experimental data ATLAS 2011 jets: Implemented! Covariance matrix definition? Legacy Tevatron electroweak measurements! Impact studied in recent HERAfitter analysis! Implementation in progress Legacy HERA data: Implemented Juan Rojo PDF4LHC, CERN, 27/10/2015 New LHCb forward DY data: Implemented ATLAS low mass DY data: Implemented ATLAS Z pt: Implemented! NNLO calculation now available Work in progress, in various degrees of progress,! for all experiments in this list
- 6. 6 Final HERA legacy dataset ! The legacy HERA inclusive combination has been added to NNPDF3.0 in various ways! NNPDF3.0 already included all published HERA-II measurements from H1 and ZEUS,! When replacing individual HERA-II data with combined dataset, we find very minor impact on PDFs Juan Rojo PDF4LHC, CERN, 27/10/2015
- 7. 7 Final HERA legacy dataset ! The legacy HERA inclusive combination has been added to NNPDF3.0 in various ways! If we compare a NNPDF3.0 fit without any HERA-II data with the corresponding fit with the legacy combination, find good consistency and a moderate reduction of PDF uncertainties Juan Rojo PDF4LHC, CERN, 27/10/2015 ! Including HERA-II measurements useful to reduce PDF uncertainties in various flavours! Results consistent with the corresponding MMHT study arXiv:1508.0661
- 8. 8 Final HERA legacy dataset ! The legacy HERA inclusive combination has been added to NNPDF3.0 in various ways! If we compare a NNPDF HERA-I-only fit with the HERA-I+II-only fit, the reduction of PDF uncertainties is even more marked (same conclusions as in the HERAPDF2.0 analysis) Juan Rojo PDF4LHC, CERN, 27/10/2015 ! Including HERA-II measurements useful to reduce PDF uncertainties in various flavours
- 9. 9 Tension at small-x?! Fit quality somewhat improves if low-x, low-Q2 data removed, specially at NNLO! However for Q2 cut ≥ 7.5 GeV2 the NNLO quality always better than NLO! Related to BFKL-like effects at small-x? To investigate using small-x (high-energy) resummed fits Juan Rojo PDF4LHC, CERN, 27/10/2015
- 10. 10 Including the ATLAS 2011 jet data Juan Rojo PDF4LHC, CERN, 27/10/2015 ! ATLAS inclusive jet production from 2011 recently available from HepForge! NLO theory from NLOjet++/APPLgrid, with pT jet as central scale! Information about how to treat systematic errors not available: explore two assumptions about the experimental covariance matrix: treat all systematics either as fully correlated or as fully uncorrelated! Non-perturbative and electroweak corrections included in the theory calculation ! ATLAS1JET11 all systematic uncertainties uncorrelated ! ATLAS1JET11 all systematic uncertainties fully correlated The theory/data comparison depends strongly on assumptions in the construction of experimental covariance matrix! ! ATLAS 2011 jets lead to reduction on large-x gluon PDF uncertainties, but need input about correct treatment of systematics
- 11. 11 Including the ATLAS 2011 jet data Juan Rojo PDF4LHC, CERN, 27/10/2015 The theory/data comparison depends strongly on assumptions in the construction of experimental covariance matrix! ! ATLAS 2011 jets lead to reduction on large-x gluon PDF uncertainties, but need input about correct treatment of systematics x 0.1 0.2 0.3 0.4 0.5 0.6 0.7 )[ref]2 )/g(x,Q2 g(x,Q 0.8 0.9 1 1.1 1.2 1.3 )=0.118Z (MS ,2 GeV4 =102 NNPDF3.0 NLO, Q Baseline + ATLAS11 jets (uncorr) + ATLAS11 jets (corr) )=0.118Z (MS ,2 GeV4 =102 NNPDF3.0 NLO, Q ! ATLAS inclusive jet production from 2011 recently available from HepForge! NLO theory from NLOjet++/APPLgrid, with pT jet as central scale! Information about how to treat systematic errors not available: explore two assumptions about the experimental covariance matrix: treat all systematics either as fully correlated or as fully uncorrelated! Non-perturbative and electroweak corrections included in the theory calculation
- 12. ! ! ! ! ! ! NNPDF fits with ! intrinsic charm Juan Rojo PDF4LHC, CERN, 27/10/2015 ! Ball, Bertone, Bonvini, Forte, Groth-Merrild, JR, Rottoli, arXiv:1510.01009! Ball, Bonvini, Rottoli, arXiv:1510.02491! NNPDF, in preparation
- 13. 13 ! The motivation to fit a charm PDF is two-fold:! Stabilise the dependence of GM-VFN calculations with respect to value of the charm mass! Quantify the possible non-perturbative charm component in the proton! ! Not enough to only add a new fitted PDF at the input scale: FFN and GM-VFN scheme calculations need to be modified to account for genuinely new contributions: massive charm- initiated processes FONLL with a fitted charm PDF Juan Rojo PDF4LHC, CERN, 27/10/2015 ! Coefficient functions for NC and CC charm-initiated contributions in the massive scheme up to NLO have been computed, but NNLO not available yet! ! Hoffmann and Moore 83! Kretzer and Schienbein 98
- 14. 14 ! The original FONLL structure functions can be modified to account for massive charm-initiated contributions! ! ! The new piece to be added to the FONLL charm structure functions is FONLL with a fitted charm PDF Juan Rojo PDF4LHC, CERN, 27/10/2015 ! This correction vanishes at large Q2 (since massless scheme unaffected by new contributions) and is numerically tiny for a perturbative generated charm (use equations of motion)!
- 15. ! Use of generalized GM-VFN scheme crucial for any realistic fit with charm PDFs 15 FONLL with a fitted charm PDF Juan Rojo PDF4LHC, CERN, 27/10/2015 ! Perturbatively generated charm PDF ! Charm PDF from BHPS model, 0.5% mom fract ! ``This work” ! ``FLNR” ! For a dynamically generated charm, the new contributions have tiny numerical effects! For BHPS-like fitted charm, substantial differences close to threshold at low scales
- 16. 16 Previous global PDF fits with IC Juan Rojo PDF4LHC, CERN, 27/10/2015 CT10 IC fit, arXiv:1309.0025 JLAB IC fit, arXiv:1408.1708 ! Fitted charm with standard S-ACOT structure functions, without massive charm-initiated terms! Sizable charm still allowed in global fit, though results depend on the specific choice of model for fitted charm, as well as value of tolerance, ∆χ2=100 ! Fitted charm in the FFN scheme, with massive charm-initiated terms! Claims very stringent bounds on IC, also depends on choice of tolerance (in this case ∆χ2=1)
- 17. 17 NNPDF3.0 fits with intrinsic charm Juan Rojo PDF4LHC, CERN, 27/10/2015 ! Currently exploring how NNPDF3.0 fits are modified if c+(x,Q0) is also fitted! Initial exploratory study: use FLNR FONLL expressions for the DIS structure functions but also fit a charm PDF (as done in the CTEQ-IC fits)! Very preliminary results for fitted charm indicate that there might be room in the global fit for a relatively large charm contribution, though with sizeable dependence on the choice of fitted dataset x 3− 10 2− 10 1− 10 )2 (x,Q+ xc 0.04− 0.02− 0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 IC no neut IC global global )=0.118 Z (MS α,2 =1.69 GeV2 NNPDF3.0 NNLO, Q <xc+> = ( 1.3 +- 1.4 )% <xc+> = ( 2.6 +- 1.3 )% ! For any robust conclusion, need to implement modified FONLL structure functions with massive charm-initiated contributions! ! Work in progress PRELIMINARY
- 18. 18 Summary and outlook Juan Rojo PDF4LHC, CERN, 27/10/2015 ! Included the legacy HERA combined dataset ! Included new measurements from ATLAS, CMS and LHCb, and studied their constraints on NNPDF3.0! NNPDF3.0 fits with intrinsic charm should become available very soon. Study their implications for LHC phenomenology! Not discussed here: NNPDF fits with running heavy quark masses, and the associated determination of mc(mc)! Also, ongoing work towards NNPDF fits with high energy resummation. Relation with the tension with the low-Q2 HERA combined data? ! Working in different directions to improve the NNPDF global analysis: