NLO Event Generation with Herwig++ Simon Pl atzer DESY on behalf - PowerPoint PPT Presentation

NLO Event Generation with Herwig++ Simon Pl¨ atzer DESY – on behalf of the Herwig++ collaboration – Simon Pl¨ atzer (DESY) NLO Event Generation with Herwig++ 1 / 16

b b b b b b b b b b b b Overview. Dedicated approaches to NLO matching, largely hand-made or semi-automated. Many processes available in current release, well established. Muon charge asymmetry in W decays A µ 0 . 3 ATLAS data Herwig++ LO 0 . 25 Azimuthal angle difference between the first and second jet Herwig++ Powheg d σ /d ∆ φ 12 [pb] 0 . 2 NLO LO+PS 0 . 15 NLO+PS 0 . 1 0 . 05 10 1 0 1 . 4 MC/data 1 . 2 1 0 0 . 5 1 1 . 5 2 2 . 5 3 0 . 8 ∆ φ 12 0 . 6 Z +jet 0 0 . 5 1 1 . 5 2 | η µ | pp → W [SP & S. Gieseke – Eur.Phys.J. C72 (2012) 2187] [K. Hamilton et al. – JHEP 0904 (2009) 116] Change in paradigm: Need for an automated, fully integrated framework. → Uncertainties and merging require full control of fixed-order input. Simon Pl¨ atzer (DESY) NLO Event Generation with Herwig++ 2 / 16

Dedicated NLO approaches: BSM Decay Chains. Powheg matching integrated with fexible and generic Herwig++ BSM infrastructure. [P. Richardson, A. Wilcock – Eur.Phys.J. C74 (2014) 2713] Jet pair mass in RS graviton decay. p ⊥ distribution in CMSSM squark decay. 10 − 5 d σ /d m jj [ fb/GeV ] d σ /d p T ,2 [ fb/GeV ] 0 . 07 Herwig++ PO Herwig++ PO Herwig++ LO Herwig++ LO 0 . 06 0 . 05 10 − 6 0 . 04 0 . 03 0 . 02 0 . 01 10 − 7 0 1 . 4 1 . 4 1 . 2 1 . 2 Ratio Ratio 1 1 0 . 8 0 . 8 0 . 6 0 . 6 1000 1200 1400 1600 1800 2000 2200 2400 0 50 100 150 200 250 300 350 400 m jj [ GeV ] p T ,2 [ GeV ] Simon Pl¨ atzer (DESY) NLO Event Generation with Herwig++ 3 / 16

Outline. – (N)LO Matrix Elements for Herwig++ with Matchbox – Matching Validation & Systematics – Shower & Matching Uncertainties – Further development: BSM, EW corrections, NLO merging – Summary & Outlook Simon Pl¨ atzer (DESY) NLO Event Generation with Herwig++ 4 / 16

Matchbox Overview. � |M n , 0 � � � �� � � � |M n , 0 � � � |M n , 0 � , |M n , 1 � � σ NLO = d σ LO + d σ V + d σ A |M ij n , 0 | 2 |M n , 0 | 2 2R e ( �M n , 0 |M n , 1 � ) n n 1 � � �� � � |M n , 0 � � P (˜ q ) , D ( p ⊥ ) + d σ PS − d σ A |M ij n , 0 | 2 R ME ( p ⊥ ) n +1 � |M n +1 , 0 � � � � � P (˜ q ) , D ( p ⊥ ) �� + d σ R − d σ PS |M n +1 , 0 | 2 R ME ( p ⊥ ) n +1 Interfaces at amplitude level Interfaces at squared amplitude level – Color bases provided, including interface to ColorFull . – Dedicated interfaces. [M. Sj¨ odahl, SP] [HEJ & SP] [nlojet++ & J. Kotanski, J. Katzy, SP] – Spinor helicity library and caching facilities. – BLHA2. – MadGraph5. [GoSam & J. Bellm, S. Gieseke, SP, C. Reuschle] [MadGraph & J. Bellm, S. Gieseke, SP, A. Wilcock] [NJet & SP] [OpenLoops & J. Bellm, S. Gieseke] – Some in-house calculations and parts of HJets++ . [VBFNLO & K. Arnold, S. Gieseke, SP] [F. Campanario, T. Figy, SP, M. Sj¨ odahl] Matchbox infrastructure Shower plugins based on [SP & S. Gieseke – Eur.Phys.J. C72 (2012) 2187] matching details & uncertainties [in preparation] – Dipole shower D ( p ⊥ ). – Process generation and bookkeeping, integration. – Angular ordered shower P (˜ q ). – Automated Catani-Seymour dipole subtraction. – ME correction R ME ( p ⊥ ), including adaptive sampling. – Diagram-based mutli-channel phase space. Simon Pl¨ atzer (DESY) NLO Event Generation with Herwig++ 5 / 16

Matchbox Validation. Extensive validation against e.g. MCFM [N. Fischer, D. Rauch, C. Reuschle] ∆ y between the WZ boson pair Antitop Quark Transverse Momentum p T ,¯ t d σ /d ∆ y [pb] d σ / dp T ,¯ t 10 3 MCFM MCFM 10 1 Herwig + GoSam H++ & GoSam 10 2 H++ & MadGraph 10 1 1 1 10 − 1 10 − 2 1 . 4 1 . 4 1 . 2 1 . 2 Ratio Ratio 1 1 0 . 8 0 . 8 0 . 6 0 . 6 0 1 2 3 4 5 0 100 200 300 400 500 600 700 800 900 ∆ y p T ,¯ t [GeV] Various internal cross checks: Subtraction checks, pole cancellation. u → e + e − gg Singularity cancellation in u ¯ 1/ ǫ pole cancellation in pp → 3 jets D / | M | 2 occurence in % 10 1 1 . 8 du → dug 1 . 6 gu → uug 1 . 4 1 1 . 2 1 0 . 8 10 − 1 0 . 6 0 . 4 0 . 2 10 − 2 0 10 − 2 10 − 1 10 1 1 - 16 - 14 - 12 - 10 - 8 - 6 √ s 14 log 10 ( ∆ ) pp → Z + jet (GoSam) pp → 3 jets (NJet) Simon Pl¨ atzer (DESY) NLO Event Generation with Herwig++ 6 / 16

NLO Calculations with Matchbox. Electroweak H +Jets production with HJets++ [F. Campanario, T. Figy, SP, M. Sj¨ odahl – PRL 111 (2013) 211802] – Employs all of Matchbox’s infrastructure for a hadron collider 2 → 4 process. – Hybrid interfaces of amplitude and squared amplitude infrastructure, internal cross checks possible. Transverse momentum of the third jet Transverse momentum of the third jet 10 2 d σ /d p ⊥ ,3 [fb/GeV] d σ /d p T ,3 [fb/GeV] 10 1 LO LO NLO NLO 10 1 1 10 − 1 1 10 − 2 10 − 1 10 − 3 10 − 2 HJets++ HJets++ 3 . 5 10 − 4 1 . 4 3 2 . 5 1 . 2 K 2 K 1 1 . 5 0 . 8 1 0 . 6 0 . 5 20 40 60 80 100 120 140 160 180 200 20 40 60 80 100 120 140 160 180 200 p ⊥ ,3 [GeV] p T ,3 [GeV] Inclusive cuts. VBF cuts. pp → H + 3 jets @ 14 TeV – inlcudes all VBF and Higgs-strahlung contributions Have pp → H + 2 jets available as well. [validated against Ciccolini, Denner, Dittmaier – Phys.Rev.Lett. 99 (2007) 161803] Simon Pl¨ atzer (DESY) NLO Event Generation with Herwig++ 7 / 16

Matching Validation & Systematics. Compare fixed order, unshowered S and H events, and full simulation. Z boson transverse momentum d σ /d p ⊥ ( Z ) GeV/pb 10 2 LO + Dipoles Prime validation: inclusive Z . NLO + Dipoles 10 1 NLO NLO + Dipoles H 1 Non-trivial application: Z plus jet 10 − 1 10 − 2 R separation of Z and first jet 10 − 3 d σ /d ∆ R ( Z , j ) 1/pb Herwig++ 10 4 Herwig++ LO + Dipoles OpenLoops / MadGraph / ColorFull OpenLoops / MadGraph / ColorFull 10 − 4 LO 10 3 1 . 4 NLO + Dipoles Ratio to NLO NLO 1 . 2 10 2 NLO + Dipoles H 1 NLO + Dipoles S 10 1 0 . 8 0 . 6 1 20 40 60 80 100 120 140 p ⊥ ( Z ) /GeV 10 − 1 Z boson rapidity 10 − 2 d σ /d y ( Z ) 1/pb 0 2 4 6 8 10 Herwig++ 180 ∆ R ( Z , j ) OpenLoops / MadGraph / ColorFull 160 Transverse momentum of first jet 140 10 1 d σ /d p ⊥ ( j ) GeV/pb 120 LO + QTilde LO + QTilde 100 1 LO LO NLO + QTilde 80 NLO + QTilde 10 − 1 NLO NLO + QTilde H 60 NLO + QTilde H NLO + QTilde S 10 − 2 40 NLO + QTilde S 20 10 − 3 0 1 . 4 10 − 4 Ratio to NLO 1 . 2 10 − 5 1 0 . 8 100 200 300 400 500 600 700 800 900 1000 p ⊥ ( j ) 0 . 6 - 6 - 4 - 2 0 2 4 6 y ( Z ) Simon Pl¨ atzer (DESY) NLO Event Generation with Herwig++ 8 / 16

Matching Validation & Systematics: Powheg-type. [A. Wilcock, P. Richardson, SP – work in progress] Powheg-type matching smoothly integrated into Matchbox – Adaptive sampling of ME correction Sudakov [SP – Eur.Phys.J. C72 (2012) 1929] – Various profile scale choices and uncertainty estimates – Can check impact of truncated showering Number events 10 6 Number events 10 5 Matchbox truncated shower POWHEGBOX no truncated shower Herwig++ 10 5 10 4 10 4 10 3 10 3 1 . 4 1 . 4 1 . 2 1 . 2 Ratio Ratio 1 1 0 . 8 0 . 8 0 . 6 0 . 6 0 20 40 60 80 100 120 140 160 180 0 20 40 60 80 100 p T , ll [ GeV ] p T , ll [ GeV ] Simon Pl¨ atzer (DESY) NLO Event Generation with Herwig++ 9 / 16

Shower & Matching Uncertainties. Shower uncertainties until now poorly understood. – Various scales in the game: µ R , µ F , µ Q . – Role of µ Q not a priory clear (no variable hard scale for a.o. showers, only p ⊥ veto) – µ R , µ F in hard process vs. in the shower? Matching is a way more complicated setting! – Some expectations confirmed in matched setups. – Surprises in uncertainties for higher jet multiplicities. – Need to profile hard emission to avoid NNLO jumps. Upshot: Cross-benchmark between different showers with and without matching. Hopefully more insight soon – needs close connection with resummation community. Simon Pl¨ atzer (DESY) NLO Event Generation with Herwig++ 10 / 16

µ Q variations and profile scales. Important to validate uncertainties at leading order : Matching may hide important details. Do we see what we expect? Transverse momentum of leading jet p d σ /d p ⊥ ( jet 1 ) [pb/GeV] 10 1 power shower natural shower natural shower hfact profile hfact profile resummation profile resummation profile 1 1 leading order 0 . 8 0 . 6 10 − 1 0 . 4 0 . 2 10 − 2 0 0 50 100 150 200 250 10 2 10 3 q / GeV p ⊥ ( jet 1 ) [GeV] log 10 ( k ⊥ jet resolution 1 → 2 [GeV]) Separation between Z boson and leading jet 10 3 d σ /d log 10 ( d 12 /GeV ) [pb] d σ /d ∆ R ( Z, 1 st jet ) [pb] power shower natural shower 10 2 hfact profile 10 2 resummation profile 10 1 10 1 power shower natural shower hfact profile 1 resummation profile 1 leading order 0 . 5 1 1 . 5 2 2 . 5 3 3 . 5 1 2 3 4 5 6 7 log 10 ( d 12 /GeV ) ∆ R ( Z, 1 st jet ) Full benchmark of uncertainties in progress – S. Gieseke & SP Simon Pl¨ atzer (DESY) NLO Event Generation with Herwig++ 11 / 16