Factorization for Jet Substructure. Andrew Larkoski Reed College

Transcription

1 Factorization for Jet Substructure Andrew Larkoski Reed College SCET 017, March 14, 017

2 Goal: Precision Calculations on Isolated Jets at the LHC

3 3 Perturbative Radiation

4 Perturbative Radiation Underlying Event 4

5 Perturbative Radiation Pile-up Underlying Event 5

6 Out-of-Jet perturbative radiation re-emission Perturbative Radiation Pile-up Underlying Event 6

7 Can eliminate these problems by grooming the jet! Butterworth, Davison, Rubin, Salam 008 Cacciari, Salam, Soyez 008 Krohn, Thaler, Wang 009 Ellis, Vermilion, Walsh 009 Soyez, Salam, Kim, Dutta, Cacciari 01 Dasgupta, Fregoso, Marzani, Salam 013 Krohn, Schwartz, Low, Wang 013 AJL, Marzani, Soyez, Thaler 014 Berta, Spousta, Miller, Leitner 014 Cacciari, Soyez, Salam 014 Bertolini, Harris, Low, Tran

8 Soft Drop Grooming Only one jet groomer removes contamination and eliminates NGLs fail fail pass Soft Drop: min[p Ti,p Tj ] p Ti + p Tj >z cut Rij R 8 Dasgupta, Fregoso, Marzani, Salam 013 AJL, Marzani, Soyez, Thaler 014

9 Soft Drop Grooming Soft Drop the hardest jet in pp Z + j events min[p Ti,p Tj ] Rij >z cut p Ti + p Tj R Measure of the soft dropped jet: m J ' X i<jj m J p Ti p Tj R ij Focus on the regime where: m J z cut p TJ p TJ All remaining particles in the jet must be collinear! soft, wide angle particle i 1) p Ti p TJ z cut m J z cut p TJ ) groomed away 9 p Ti p TJ m J p TJ

10 = β = + > = ( + ) (+ ) +α - - SCET 016: d resum dm J k=q, q,g includes pdfs, emissions sum over jet flavor that were groomed away, out-of-jet radiation,... = X m J z cut p TJ p TJ 10 Enables all-orders factorization of jet observables with no NGLs Presented NNLL+NLO predictions collinear-soft radiation Frye, AJL, Schwartz, Yan 016 See talks by: C Frye, K. Yan D k (p T,z cut,r)s C,k (z cut m J) J k (m J) hard collinear radiation

11 Goal: Discriminate between QCD jets and boosted hadronic decays of W/Z/H bosons 11

12 W/Z/H q/g Signal: Two-prong jet Characteristic angular size determined by mass Background: One-prong jet No intrinsic angular size 1

13 Optimal Observable: D ( ) e( ) 3 (e ( ) )3 (β) ( (β) (β) ) e ( ) 3 (e ( ) )3 e ( ) = 1 p TJ jet pt e ( ) 3 = 1 p 3 TJ X i<jj sum over distinct pairs of particles in the jet X i<j<kj p Ti p Tj R ij angle between i and j p Ti p Tj p Tk R ij R ik R jk (β) Note: e () ' m J p T 13 AJL, Moult, Neill 014

14 3 10 Events / ATLAS Preliminary s = 13 TeV, 3. fb -1 Trimmed anti-k t R=1.0 Dijet Selection > 50 GeV m calo Data 015 Pythia8 Herwig++ Stat. syst. uncert. Stat. uncert. Measurements in LHC Data Groomed (with Trimming) 40 0 Angular exponent: beta = Data/Pred ATLAS JETM β = 1 Large-R Jet D Calculations in e + e - at NLL No grooming; ignoring NGLs () + - [ ] = Angular exponent: beta = 14 () D ( =) AJL, Moult, Neill 015

15 Goal: Precision Soft Dropped D Predictions 15 AJL, Moult, Neill 17xx.sooon

16 Three Observations of Soft Dropped D: Kinematic Endpoint Fixed; Independent of Jet Properties Suppressed Non-Perturbative Corrections Process Universality 16

17 Kinematic Endpoint Fixed; Independent of Jet Properties: Ungroomed Case D ( ) = e( ) 3 (e ( ) )3 Maximum e3 for one hard prong and uniform soft radiation e ( ) 3 max = (e( ) ) D ( ) D () max max = 1 e ( ) ' p T m J Probability Density Z Boson vs. Gluon Jet HPythia8L R 0 =1, p T > 500 GeV, D H, L M GeV M GeV M GeV M GeV Endpoint drifts as mass cut is changed! D H, L

18 Kinematic Endpoint Fixed; Independent of Jet Properties: Soft Drop Groomed Case D ( ) = e( ) 3 (e ( ) )3 D ( ) max = 1 z cut Endpoint independent of mass cut! >z cut p T Probability Density Maximum e3 for one hard prong and uniform soft radiation e ( ) 3 max = (e( ) )3 z cut Z Boson vs. Gluon Jet HPythia8L Soft Drop, b= 0, z cut = 0.1 R 0 =1, p T > 500 GeV, D H, L M GeV M GeV M GeV M GeV Robust distribution D H, L

19 Suppressed Non-Perturbative Corrections Ungroomed Case D ( ) = e( ) 3 (e ( ) )3 Value of e3 dominated by a wide angle non-perturbative gluon: e ( ) 3 NP. QCD e ( ) p T D ( ) D () NP NP. QCD p T 1 (e ( ) ). QCDp 3 T m 4 J Severe sensitivity to non-perturbative effects as cuts are varied 19 () + - [ ] = () σ σ Ω =

20 Suppressed Non-Perturbative Corrections Soft Drop Groomed Case D ( ) = e( ) 3 (e ( ) )3 D ( ) NP. QCD 1 p T D () NP e ( ) 1. QCD m J z 3/ z 1 cut cut No non-perturbative sensitivity to the jet pt! 0 σ σ (β) Value of e3 dominated by a softcollinear non-perturbative gluon: >z cut p T e ( ) 3 NP. QCD p T = = = = (β) ( ) e ( ) z 1 cut [ ] 3 1

21 Process Universality Soft Drop Groomed Case p p > Z + j: p p > H + j: ~80% quark ~0% gluon ~50% quark ~50% gluon Soft Drop renders quark and gluon jet flavor IRC safe! Possible flavor ambiguities are groomed away D () p T = 500 GeV, Hadron Level m J [80, 100] GeV anti-quark quark σ σ (β) pp->h+j pp->z+j soft gluon (β) 1

22 Process Universality Soft Drop Groomed Case p p > Z + j: p p > H + j: ~80% quark ~0% gluon ~50% quark ~50% gluon Differences between quarks and gluons largely eliminated Robust, stable discrimination over huge range of pt! D () D () p T = 500 GeV, Hadron Level m J [80, 100] GeV pp (ZZ,Zj), Hadron Level m J [80, 100] GeV σ σ (β) pp->h+j pp->z+j σ σ (β) p T = 500 GeV p T = 1000 GeV p T = 1500 GeV p T = 000 GeV (β) (β)

23 Summary Soft Drop jet grooming can be used to eliminate NGLs in jet distributions Powerful techniques necessary to identify hadronic decays of W/Z/H Grooming improves robustness to process, cuts, and hadronization Predictions at NLL (and beyond!) soon 3 AJL, Moult, Neill 17xx.sooon