P YTHIA 8 Richard Corke Department of Astronomy and Theoretical Physics Lund University June 2010 Torbj¨ orn Sj¨ ostrand, Stefan Ask, Stephen Mrenna, Peter Skands, Lisa Carloni Richard Corke (Lund University) TOOLS 2010, Winchester, UK June 2010 1 / 43
Overview P YTHIA 1 Physics overview 2 BSM Physics 3 Running P YTHIA 8 4 Conclusions 5 Richard Corke (Lund University) TOOLS 2010, Winchester, UK June 2010 2 / 43
P YTHIA ◮ General purpose Monte Carlo event generator ◮ Combine pQCD and models to provide link from theory (quarks, gluons) to experiment (mesons, baryons) ◮ Full problem “factorised” into different components ◮ Hard process ◮ Resonance decays ◮ Parton showers ◮ Underlying event ◮ Hadronisation ◮ Hadron decays ◮ Different parts may be handled by other external programs (e.g. Tauola) ◮ Or (with P YTHIA 8) through plugins (e.g. V INCIA ) ◮ Outputs exclusive hadronic events ◮ Analyse (e.g. FastJet) ◮ Pass to detector simulator (e.g. GEANT) ◮ ... Richard Corke (Lund University) TOOLS 2010, Winchester, UK June 2010 3 / 43
P YTHIA Richard Corke (Lund University) Richard Corke (Lund University) TOOLS 2010, Winchester, UK P YTHIA 8 April 2009 June 2010 3 / 14 4 / 43
Physics overview Further reading ◮ Latest downloads and news: ◮ http://home.thep.lu.se/˜torbjorn/Pythia.html ◮ “P YTHIA 6.4 Physics and Manual” T. Sj¨ ostrand, S. Mrenna and P . Skands, JHEP 0605:026,2006, [hep-ph/0603175]. ◮ “A Brief Introduction to P YTHIA 8.1” T. Sj¨ ostrand, S. Mrenna and P . Skands, Comput. Phys. Comm. 178 (2008) 852 [arXiv:0710.3820]. ◮ And references therein Richard Corke (Lund University) TOOLS 2010, Winchester, UK June 2010 5 / 43
Physics overview Beams and hard processes ◮ Beams ◮ Incoming beams: pp , p¯ p , e + e − , µ + µ − ◮ P YTHIA 8: no ep , γ p or γγ beam configurations ◮ Built in parton distribution function (PDF) sets ◮ GRV94L, CTEQ5L ◮ MSTW2008 (LO and NLO), MRST LO** ◮ CTEQ6L, CTEQ6L1, CTEQ6.6, CT09MC1, CT09MC2, CT09MCS ◮ Easy to link to LHAPDF for many more W + u g d p p / p ◮ Hard Processes ◮ Built-in library of many leading-order processes ◮ SM: almost all 2 → 1 and 2 → 2, some 2 → 3 ◮ BSM: a bit of everything (more to come) ◮ External input through Les Houches Accord (LHA) and Les Houches Event Files (LHEF) Richard Corke (Lund University) TOOLS 2010, Winchester, UK June 2010 6 / 43
Physics overview Parton showers ◮ Regions of phase space where higher-order terms are enhanced ◮ Full matrix element calculation not feasible ◮ DGLAP evolution equations; leading log approximation of QCD ◮ Sudakov form factor; shower evolution as a probabilistic process + + FSR ISR ◮ Initial state radiation performed through backwards evolution ◮ Pick a hard 2 → 2 process ◮ What is the probability that incoming parton b came from a splitting a → bc ? ◮ PDF factors enter the evolution ◮ Iterate to build up event c s W + u g d p p / p Richard Corke (Lund University) TOOLS 2010, Winchester, UK June 2010 7 / 43
Physics overview Parton showers ◮ Still choices to make! ◮ Ordering ◮ Transverse-momentum-ordered showers ⊥ p 2 p 2 p 2 ⊥ ⊥ ⊥ y y y 2 ◮ Recoil strategy ◮ Dipole approach to recoil ◮ Each radiator parton has a recoiler partner ◮ Kinematics constructed directly after each branching ◮ All unevolved partons on mass shell Richard Corke (Lund University) TOOLS 2010, Winchester, UK June 2010 8 / 43
Physics overview Parton showers ◮ Matching to ME for first emission in many processes ◮ Aim to provide better shower behaviour at large p ⊥ ◮ Dampen shower tail in coloured final states ◮ Also examine interfacing of POWHEG NLO generators to P YTHIA ◮ RC & T. Sj¨ ostrand, arXiv:1003.2384 [hep-ph] (b) 10 -2 POWHEG Pythia Default (Power) 10 -3 Pythia Damp, k = 2 ] Pythia Damp, k = 1 [GeV -1 Pythia Wimpy 10 -4 dP / dp ⊥ 10 -5 10 -6 10 -7 100 200 300 400 500 600 700 800 900 1000 p ⊥ [GeV] 2 2 ◮ Implementation of CKKW-L in progress (Stefan Prestel) Richard Corke (Lund University) TOOLS 2010, Winchester, UK June 2010 9 / 43
Physics overview Underlying event ◮ Multiple parton-parton interactions ◮ QCD 2 → 2, prompt photon production, Drell Yan, Charmonium & Bottomonium ◮ Impact parameter dependence ◮ Dampened cross section in p ⊥ → 0 limit ⊥ → r r r r ∝ α 2 S ( p 2 → α 2 S ( p 2 ⊥ 0 + p 2 ⊥ ) d ˆ σ ⊥ ) d p 2 p 4 ( p 2 ⊥ 0 + p 2 ⊥ ) 2 d d ⊥ ⊥ λ ∼ 1 /p ⊥ resolved screened ◮ Interleaved p ⊥ evolution with ISR and FSR ◮ ISR and MI “compete” for beam ◮ Flavour dependent PDF effects ◮ Showering from all interactions Richard Corke (Lund University) TOOLS 2010, Winchester, UK June 2010 10 / 43
Physics overview Underlying event ◮ Picture now a lot more messy c s W + u g d p p / p ◮ Rescattering: scattered parton allowed to interact again ◮ Same order in α s , but one PDF weight less ◮ Large background → will be tough to find direct evidence ◮ RC & T. Sj¨ ostrand, JHEP 01 (2010) 035 [arXiv:0911.1909] → → Richard Corke (Lund University) TOOLS 2010, Winchester, UK June 2010 11 / 43 New source of 3-jet topologies; visible effects after tuning?
Physics overview Hadronisation � String fragmentation - “The Lund Model” � d E � � d p z � � d E � � d p z � � � � � � � � � � = � = � = � = κ � � � � � � � � d z d z d t d t � � � � g ( rb ) q ( r ) t q q z q ( b ) � String breaking modelled by tunnelling q � q � q q � q � q q q d = m ⊥ q / κ m ⊥ q � = 0 m ⊥ q � > 0 � Particle decays, usually isotropic Richard Corke (Lund University) TOOLS 2010, Winchester, UK June 2010 12 / 43
Physics overview Hadronisation ◮ Everything connected by colour confinement strings ◮ Strings fragment to produce primary hadrons ◮ Unstable hadrons decay further Everything is connected by colour confinement strings Richard Corke (Lund University) TOOLS 2010, Winchester, UK June 2010 13 / 43
Physics overview Hadronisation ◮ Everything connected by colour confinement strings ◮ Strings fragment to produce primary hadrons ◮ Unstable hadrons decay further The strings fragment to produce primary hadrons Richard Corke (Lund University) TOOLS 2010, Winchester, UK June 2010 13 / 43
Physics overview Hadronisation ◮ Everything connected by colour confinement strings ◮ Strings fragment to produce primary hadrons ◮ Unstable hadrons decay further Many hadrons are unstable and decay further Richard Corke (Lund University) TOOLS 2010, Winchester, UK June 2010 13 / 43
b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b Physics overview Current status ◮ Tuning with Rivet + Professor A. Buckley et al. [hep-ph/1003.0694, hep-ph/0907.2973] ◮ Tuning to e + e − data looks okay ◮ Hard physics distributions also okay ◮ But problems describing the underlying event? Transverse region charged ∑ p ⊥ density Transverse region charged particle density � N ch � /d η d φ � /d η d φ / GeV 2 CDF data 1 ’ 6142 -default’ ’ 8135 -default’ 0 . 8 1 . 5 � ∑ p track CDF data 0 . 6 T ’ 6142 -default’ 1 ’ 8135 -default’ 0 . 4 0 . 5 0 . 2 0 0 MC/data MC/data 1 . 4 1 . 4 1 . 2 1 . 2 1 1 0 . 8 0 . 8 0 . 6 0 . 6 0 50 100 150 200 250 300 350 400 0 50 100 150 200 250 300 350 400 p T ( leading jet ) / GeV p T ( leading jet ) / GeV ◮ Possible causes ◮ Final-state dipoles with initial-state recoil ◮ Azimuthal asymmetry of radiation Richard Corke (Lund University) TOOLS 2010, Winchester, UK June 2010 14 / 43
b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b Physics overview Current status ◮ Initial results promising, but still much to be checked Transverse region charged particle density Transverse region charged ∑ p ⊥ density � N ch � /d η d φ � /d η d φ / GeV 2 CDF data 1 ’ 6142 -default’ ’ 8135 -default’ 0 . 8 1 . 5 ’ 8140 -default-asym’ � ∑ p track CDF data 0 . 6 T ’ 6142 -default’ 1 ’ 8135 -default’ 0 . 4 ’ 8140 -default-asym’ 0 . 5 0 . 2 0 0 MC/data MC/data 1 . 4 1 . 4 1 . 2 1 . 2 1 1 0 . 8 0 . 8 0 . 6 0 . 6 0 50 100 150 200 250 300 350 400 0 50 100 150 200 250 300 350 400 p T ( leading jet ) / GeV p T ( leading jet ) / GeV ◮ Go further? ◮ Compare first parton shower emission to 2 → 3 matrix elements ◮ How does 2 → 2 ⊗ PS fill the phase space? ◮ Work ongoing! Richard Corke (Lund University) TOOLS 2010, Winchester, UK June 2010 15 / 43
BSM Physics Overview ◮ Much early focus on SM physics ◮ Emphasis on providing solid links to external programs ◮ Les Houches Accord (LHA) and Les Houches Events Files (LHEF) can be used to read in parton-level events for showering and hadronisation ◮ Easy to use P YTHIA to simulate a wide range of BSM processes in this way ◮ Important to understand what choices need to be made and what P YTHIA can and can’t do ◮ But also complemented by a library of common BSM processes Richard Corke (Lund University) TOOLS 2010, Winchester, UK June 2010 16 / 43
Recommend
More recommend