Effect of new jet substructure measurements on Pythia8 tunes
EE ff ect of new jet substructure measurements on Pythia8 tunes Deepak Kar a , Pratixan Sarmah b a School of Physics, University of WitwatersrandJohannesburg, South Africa.Email: [email protected] b Department of Physics, BITS PilaniRajasthan, India.Email: [email protected]
Abstract
This masters project used the recent ATLAS jet substructure measurements to see if any improvements can be madeto the commonly used Pythia8 Monash and A14 tunes.
Keywords:
Pythia8, jet substructure, FSR, tune
1. Introduction
The commonly used Pythia8 [1, 2] tunes, Monash [3] and A14 [4] are rather dated, and the latter was observed tohave some tension with LEP measurements, primarily due to its lower Final State Radiation (FSR) α s value. In lastcouple of years, a plethora of jet substructure [5, 6, 7, 8] measurements have been published by both ATLAS and CMScollaborations, utilising LHC Run 2 data. Here, we investigate the e ff ect of four such ATLAS measurements on parameterssensitive to jet substructure observables.
2. Tuning setup
The following ATLAS measurements were considered in this study (along with their Rivet identifiers): • Soft-Drop Jet Mass [9](ATLAS 2017 I1637587) • Jet substructure measurements in multijet events [10] (ATLAS 2019 I1724098) • Soft-drop observables [11](ATLAS 2019 I1772062) • Lund jet plane with charged particles [12] (ATLAS 2020 I1790256)The following parameters were considered in this tuning exercise, with the ranges stated in Table 1.Parameter Lower value Upper valueBeamRemnants:primordialKThard 1.25 3ColorReconncetion:range 1.25 3TimeShower:pTmin 0.5 1.5MultipartonInteractions:pT0Ref 1.5 3TimeShower:alphaSvalue 0.118 0.145
Table 1: Sampling range of the parameters considered
Weighted hardQCD events were generated with a
PThatMin of 300 GeV. 100 Sampling runs were performed, eachwith 100000 events. Rivet3 [13] and Professor tuning system [14] were used. The goodness of sampling and the weightfiles used can be found in Appenix 5.2 and in Appenix 5.3. 1 a r X i v : . [ h e p - e x ] J a n . Results The first step was to ascertain where we have a scope of improvement. While a detailed observable-by-observabledetermination was performed (see Appendix 5.1), here we highlight the most salient features: • For Lund Jet Plane (LJP) distributions, we observed that the hard-wide angle emissions part is better modelled bythe Monash tune whereas the region ranging from UE / MPI to Soft-collinear and Collinear limits are in generalbetter modelled by the A14 tune. However, this distributions also o ff er the biggest scope of improved modelling. • For the soft drop ρ and r g observables, in general Monash tune performs somewhat better than A14. One deviationfrom this trend is when the jet construction is Cluster based, in which case the A14 tune performs better over a largerange. • Both the Jet Substructure and Soft drop jet mass distributions are somewhat better modelled by the A14 tune.Table 2 lists the parameter values of A14 and Monash along with our tuned values. A separate tune for LJP wasperformed as this analysis had the largest discrepancy. The LJP tune column shows the parameter values correspondingto the best tune for LJP and the Common Tune column shows the values of the best tune for all the analyses considered.Figures 1 and 2 show the tuned distributions for the one dimensional vertical slices of the LJP. Figure 3 shows the tuneddistributions for the soft drop observables. Figure 4 shows the tuned distributions for soft drop mass. And lastly, Figure 5shows the tuned distributions for the jet substructure observables.Parameters A14 Monash LJP Tune Common TuneBeamRemnants:primordialKThard 1.88 1.8 2.288 2.065ColorReconnection:range 1.71 1.8 2.73 1.69TimeShower:pTmin 0.40 0.50 1.288 0.775MultipartonInteractions:pT0Ref 2.09 2.28 2.766 2.91TimeShower:alphaSvalue 0.127 0.1365 0.1308 0.1309
Table 2: Comparison of tuned values with Monash and A14 b b b b b b b b b b b b b b b b b bb DataA TuneMonash TuneLJP TuneCommon Tune . . Vertical slice, 0.67 < ln ( R / ∆ R ) < N j e t s d N e m i ss i o n s d l n ( R / ∆ R ) d l n ( / z ) b b b b b b b b b b b b b b b b b b b . . . . . . . . . ln ( z ) M C / D a t a b b b b b b b b b b b b b b b b b b bb DataA TuneMonash TuneLJP TuneCommon Tune . . Vertical slice, 1.00 < ln ( R / ∆ R ) < N j e t s d N e m i ss i o n s d l n ( R / ∆ R ) d l n ( / z ) b b b b b b b b b b b b b b b b b b b . . . . . . . . . ln ( z ) M C / D a t a b b b b b b b b b b b b b b b b b b bb DataA TuneMonash TuneLJP TuneCommon Tune . . Vertical slice, 1.33 < ln ( R / ∆ R ) < N j e t s d N e m i ss i o n s d l n ( R / ∆ R ) d l n ( / z ) b b b b b b b b b b b b b b b b b b b . . . . . . . . . ln ( z ) M C / D a t a b b b b b b b b b b b b b b b b b b bb DataA TuneMonash TuneLJP TuneCommon Tune . . Vertical slice, 1.67 < ln ( R / ∆ R ) < N j e t s d N e m i ss i o n s d l n ( R / ∆ R ) d l n ( / z ) b b b b b b b b b b b b b b b b b b b . . . . . . . . . ln ( z ) M C / D a t a Figure 1: Comparison of our tunes with A14 and Monash tunes for Lund Jet Plane distributions (vertical slices) b b b b b b b b b b b bb DataA TuneMonash TuneLJP TuneCommon Tune . . Horizontal slice, 2.08 < ln ( z ) < N j e t s d N e m i ss i o n s d l n ( R / ∆ R ) l n ( / z ) b b b b b b b b b b b b b . . . . . . . . . . . . . ln ( R / ∆ R ) M C / D a t a b b b b b b b b b b b b bb DataA TuneMonash TuneLJP TuneCommon Tune . . Horizontal slice, 2.36 < ln ( z ) < N j e t s d N e m i ss i o n s d l n ( R / ∆ R ) d l n ( / z ) b b b b b b b b b b b b b . . . . . . . . . . . . . ln ( R / ∆ R ) M C / D a t a b b b b b b b b b b b b bb DataA TuneMonash TuneLJP TuneCommon Tune . . Horizontal slice, 2.63 < ln ( z ) < N j e t s d N e m i ss i o n s d l n ( R / ∆ R ) d l n ( / z ) b b b b b b b b b b b b b . . . . . . . . . . . . . ln ( R / ∆ R ) M C / D a t a b b b b b b b b b b b b bb DataA TuneMonash TuneLJP TuneCommon Tune . . Horizontal slice, 2.91 < ln ( z ) < N j e t s d N e m i ss i o n s d l n ( R / ∆ R ) d l n ( / z ) b b b b b b b b b b b b b . . . . . . . . . . . . . ln ( R / ∆ R ) M C / D a t a Figure 2: Comparison of our tunes with A14 and Monash tunes for Lund Jet Plane distributions (horizontal slices) b b b b b b b b bb DataA TuneMonash TuneLJP TuneCommon Tune . . . . . . . . . Dijet selection, Central jet, Calorimeter Based, β = . , z cut = . ( / σ r e s u m ) d σ / d l o g [ ( m s o f t d r o p / p u n g r oo m e d T ) ] b b b b b b b b b b - . - - . - - . - - . - - . . . . . . . . . . log [( m softdrop / p ungroomed T ) ] M C / D a t a b b b b b b b b b bb DataA TuneMonash TuneLJP TuneCommon Tune . . . . . . . . Dijet selection, Central jet, Track Based, β = . , z cut = . ( / σ r e s u m ) d σ / d l o g [ ( m s o f t d r o p / p u n g r oo m e d T ) ] b b b b b b b b b b - . - - . - - . - - . - - . . . . . . . . . . log [( m softdrop / p ungroomed T ) ] M C / D a t a b b b b b b b b b bb DataA TuneMonash TuneLJP TuneCommon Tune . . . . . . . . . Dijet selection, Central jet, Cluster Based, β = . , z cut = . ( / σ r e s u m ) d σ / d l o g [ ( m s o f t d r o p / p u n g r oo m e d T ) ] b b b b b b b b b b - . - - . - - . - - . - - . . . . . . . . . . log [( m softdrop / p ungroomed T ) ] M C / D a t a b b b b b b b b b bb DataA TuneMonash TuneLJP TuneCommon Tune . . . . . . . . . Dijet selection, Central jet, Track Based, β = . , z cut = . ( / σ r e s u m ) d σ / d l o g [ ( m s o f t d r o p / p u n g r oo m e d T ) ] b b b b b b b b b b - . - - . - - . - - . - - . . . . . . . . . . log [( m softdrop / p ungroomed T ) ] M C / D a t a b b b b b b b b b bb DataA TuneMonash TuneLJP TuneCommon Tune . . . . . Dijet selection, Central jet, Cluster Based, β = . , z cut = . ( / σ r e s u m ) d σ / d l o g [ ( m s o f t d r o p / p u n g r oo m e d T ) ] b b b b b b b b b b - . - - . - - . - - . - - . . . . . . . . . . log [( m softdrop / p ungroomed T ) ] M C / D a t a b b b b b b b b b bb DataA TuneMonash TuneLJP TuneCommon Tune − − Dijet selection, Central jet, Track Based, Soft Drop β = . , z cut = . ( / σ r e s u m ) d σ / d l o g [ ( m s o f t d r o p / p u n g r oo m e d T ) ] b b b b b b b b b b - . - - . - - . - - . - - . . . . . . . . . . log [( m softdrop / p ungroomed T ) ] M C / D a t a Figure 3: Comparison of our tunes with A14 and Monash tunes for soft drop jet mass distributions b b b b b b b b bb DataA TuneMonash TuneLJP TuneCommon Tune . . . . . . . Soft drop β = z cut = p leadT >
600 GeV σ r e s u m d σ d l o g [ ( m s o f t d r o p / p u n g r oo m e d T ) ] b b b b b b b b b b - . - - . - - . - - . - - . . . . . . . . . . log [( m softdrop / p ungroomedT ) ] M C / D a t a b b b b b b b b b bb DataA TuneMonash TuneLJP TuneCommon Tune . . . . . . Soft drop β = z cut = p leadT >
600 GeV σ r e s u m d σ d l o g [ ( m s o f t d r o p / p u n g r oo m e d T ) ] b b b b b b b b b b - . - - . - - . - - . - - . . . . . . . . . . log [( m softdrop / p ungroomedT ) ] M C / D a t a b b b b b b b b b bb DataA TuneMonash TuneLJP TuneCommon Tune . . . . . . Soft drop β = z cut = p leadT >
600 GeV σ r e s u m d σ d l o g [ ( m s o f t d r o p / p u n g r oo m e d T ) ] b b b b b b b b b b - . - - . - - . - - . - - . . . . . . . . . . log [( m softdrop / p ungroomedT ) ] M C / D a t a Figure 4: Comparison of our tunes with A14 and Monash tunes for soft drop jet mass distributions b b b b DataA Tune [MODE=DJ]Monash Tune [MODE=DJ]LJP Tune [MODE=DJ]Common Tune [MODE=DJ] . . . . . . Dijet selection, soft-dropped σ d σ d N s u b j e t s b b b b . . . . . . . . . Nsubjets M C / D a t a b b b b b b b b b b b b b b b b DataA Tune [MODE=DJ]Monash Tune [MODE=DJ]LJP Tune [MODE=DJ]Common Tune [MODE=DJ] . . . Dijet selection, soft-dropped σ d σ d C b b b b b b b b b b b b b b b . . . . . . . . . . . . . . . . . . C M C / D a t a b b b b b b b b bb DataA Tune [MODE=DJ]Monash Tune [MODE=DJ]LJP Tune [MODE=DJ]Common Tune [MODE=DJ]
Dijet selection, soft-dropped σ d σ d D b b b b b b b b b . . . . . . . . . . . . . . D M C / D a t a b b b b b b DataA Tune [MODE=DJ]Monash Tune [MODE=DJ]LJP Tune [MODE=DJ]Common Tune [MODE=DJ] . . . . . . . . Dijet selection, soft-dropped σ d σ dL H A b b b b b . . . . . . . . . LHA M C / D a t a b b b b b b b b b b b b b bb DataA Tune [MODE=DJ]Monash Tune [MODE=DJ]LJP Tune [MODE=DJ]Common Tune [MODE=DJ] − Dijet selection, soft-dropped σ d σ dE C F n o r m b b b b b b b b b b b b b b .
05 0 . .
15 0 . .
25 0 . . . . . . . . . . . ECF norm M C / D a t a b b b b b b b b b b bb DataA Tune [MODE=DJ]Monash Tune [MODE=DJ]LJP Tune [MODE=DJ]Common Tune [MODE=DJ] − Dijet selection, soft-dropped σ d σ dE C F n o r m b b b b b b b b b b b .
005 0 .
01 0 .
015 0 .
02 0 .
025 0 .
03 0 .
035 0 . . . . . . . . . . ECF norm M C / D a t a Figure 5: Comparison of our tunes with A14 and Monash tunes for jet Sub structure observable distribution for dijet selection . Summary The results obtained show small improvements of roughly 5-10% in the distributions of the Lund Jet Plane and SoftDrop Mass distributions from the previous A14 and Monash Tunes. As in Table 2, it can be seen that the parameter valuesof the tunes obtained are pulled up from the A14 and Monash Tunes. In the case of the LJP , we see that the A14 andMonash Tunes deviate most from the data near the peaks of the distributions. This is the region where soft collinear e ff ectstransitions to UE / MPI e ff ects in the LJP. Since the tunes we obtained improve this region of the distributions, it can beinferred that higher values of these parameters facilitate more soft radiations in the final state.In the case of the soft drop observable distributions, there are regions that require generation of more mass to model thedata better. These compete with the LJP values and decreases values of parameters: BeamRemnants:primordialKThard from 2.288 to 2.065,
ColorReconnection:range from 2.73 to 1.69,
TimeShower:pTmin from 1.288 to 0.775. For the othertwo parameters,
MPI:pT0Ref and
TimeShower:alphaSvalue , the values increased slightly.
Acknowledgements
DK is funded by National Research Foundation (NRF), South Africa through Competitive Programme for RatedResearchers (CPRR), Grant No: 118515. We thank Andy Buckley and Holger Schulz for technical assistance withProfessor program, as well as for physics discussions.
References [1] T. Sjostrand, S. Mrenna, P. Z. Skands, A Brief Introduction to PYTHIA 8.1, Comput. Phys. Commun. 178 (2008)852–867. arXiv:0710.3820 , doi:10.1016/j.cpc.2008.01.036 .[2] T. Sj¨ostrand, S. Ask, J. R. Christiansen, R. Corke, N. Desai, P. Ilten, S. Mrenna, S. Prestel, C. O. Rasmussen,P. Z. Skands, An Introduction to PYTHIA 8.2, Comput. Phys. Commun. 191 (2015) 159–177. arXiv:1410.3012 , doi:10.1016/j.cpc.2015.01.024 .[3] P. Skands, S. Carrazza, J. Rojo, Tuning PYTHIA 8.1: the Monash 2013 Tune, Eur. Phys. J. C74 (8) (2014) 3024. arXiv:1404.5630 , doi:10.1140/epjc/s10052-014-3024-y .[4] ATLAS Collaboration, ATLAS Pythia 8 tunes to 7 TeV data, ATL-PHYS-PUB-2014-021 (2014).URL https://cds.cern.ch/record/1966419 [5] A. Altheimer, et al., Jet Substructure at the Tevatron and LHC: New results, new tools, new benchmarks, J. Phys.G39 (2012) 063001. arXiv:1201.0008 , doi:10.1088/0954-3899/39/6/063001 .[6] A. Altheimer, et al., Boosted objects and jet substructure at the LHC. Report of BOOST2012, held at IFIC Va-lencia, 23rd-27th of July 2012, Eur. Phys. J. C74 (3) (2014) 2792. arXiv:1311.2708 , doi:10.1140/epjc/s10052-014-2792-8 .[7] S. Marzani, G. Soyez, M. Spannowsky, Looking inside jets: an introduction to jet substructure and boosted-objectphenomenology, Vol. 958, Springer, 2019. arXiv:1901.10342 , doi:10.1007/978-3-030-15709-8 .[8] R. Kogler, et al., Jet Substructure at the Large Hadron Collider: Experimental Review, Rev. Mod. Phys. 91 (4) (2019)045003. arXiv:1803.06991 , doi:10.1103/RevModPhys.91.045003 .[9] ATLAS Collaboration, Measurement of the Soft-Drop Jet Mass in pp Collisions at √ s =
13 TeV with the ATLAS De-tector, Phys. Rev. Lett. 121 (9) (2018) 092001. arXiv:1711.08341 , doi:10.1103/PhysRevLett.121.092001 .[10] ATLAS Collaboration, Measurement of jet-substructure observables in top quark, W boson and light jet productionin proton-proton collisions at √ s =
13 TeV with the ATLAS detector, JHEP 08 (2019) 033. arXiv:1903.02942 , doi:10.1007/JHEP08(2019)033 .[11] ATLAS Collaboration, Measurement of soft-drop jet observables in pp collisions with the ATLAS detector at √ s =
13 TeV, Phys. Rev. D 101 (5) (2020) 052007. arXiv:1912.09837 , doi:10.1103/PhysRevD.101.052007 .[12] ATLAS Collaboration, Measurement of the Lund Jet Plane Using Charged Particles in 13 TeV Proton-ProtonCollisions with the ATLAS Detector, Phys. Rev. Lett. 124 (22) (2020) 222002. arXiv:2004.03540 , doi:10.1103/PhysRevLett.124.222002 . 813] C. Bierlich, et al., Robust Independent Validation of Experiment and Theory: Rivet version 3, SciPost Phys. 8 (2020)026. arXiv:1912.05451 , doi:10.21468/SciPostPhys.8.2.026 .[14] A. Buckley, H. Hoeth, H. Lacker, H. Schulz, J. E. von Seggern, Systematic event generator tuning for the LHC, Eur.Phys. J. C 65 (2010) 331–357. arXiv:0907.2973 , doi:10.1140/epjc/s10052-009-1196-7 .9 . Appendix Plots Observable ln( R / ∆ R ) or Better Tune Performance regions Better Tuneln(1 / z ) Slice A14 Monash (overall)d03-x01-y01 ln(1 / z ) 0.00-0.33 4.3-6 0.7-4.3 Monashd04-x01-y01 ln(1 / z ) 0.33-0.67 3.4-4.2 0.7-3.4 , 4.4-6 Monashd05-x01-y01 ln(1 / z ) 0.67-1.00 3-5 0.7-3 , 5-6 Monashd06-x01-y01 ln(1 / z) 1.00-1.33 3-4.2 0.7-3 , 4.2-6 Monashd07-x01-y01 ln(1 / z ) 1.33-1.67 2.4-4.2 0.7-2.4 , 4.3-6 Monashd08-x01-y01 ln(1 / z ) 1.67-2.00 2-4 , 5.2 , 5.8 0.7-2 , 4-5 , 5.5 -d09-x01-y01 ln(1 / z ) 2.00-2.33 1.8-4 , 4.7, 5.2-6 0.7-1.8, 4-4.4 A14d10-x01-y01 ln(1 / z ) 2.33-2.67 1.6-3, 4.4-5.8 0.7-1.4, 3.2-4.2 A14d11-x01-y01 ln(1 / z ) 2.67-3.00 1.4-5 0.7-1.2, 5.2-5.8 A14d12-x01-y01 ln(1 / z ) 3.00-3.33 0.8-1.4, 3-4, 4.7, 5.8 1.6-3, 3.3, 4-4.5, 5-5.6 -d13-x01-y01 ln(1 / z ) 3.33-3.67 0.7-3.4, 4.6-5 3.6-4.4, 5.1-6 A14d14-x01-y01 ln(1 / z ) 3.67-4.00 0-3.4 3.6-4.6 A14d15-x01-y01 ln(1 / z ) 4.00-4.33 0.7-2.6, 3.6, 4-4.6, 5, 5.57 2.7-3.4, 3.9, 4.7, 5.23 A14d16-x01-y01 ln( R / ∆ R ) 0.69-0.97 3-4.5 0-3 Monashd17-x01-y01 ln( R / ∆ R ) 0.97-1.25 3-4 0-3 Monashd18-x01-y01 ln( R / ∆ R ) 1.25-1.52 2.7-3.2, 3.5-4 0-2.6 Monashd19-x01-y01 ln( R / ∆ R ) 1.52-1.80 2.4-4.5 0.5-2.3 -d20-x01-y01 ln( R / ∆ R ) 1.80-2.08 2-3, 3.3-4.5 0-2 -d21-x01-y01 ln( R / ∆ R ) 2.08-2.36 1.6-4.5 0-1.6, 2.2, 3.4-4 -d22-x01-y01 ln( R / ∆ R ) 2.36-2.63 1.4-3, 3.6-4.5 0-1.3, 3-3.6 A14d23-x01-y01 ln( R / ∆ R ) 2.63-2.91 1.4-3, 3.5 0-1.3, 3-4.5 -d24-x01-y01 ln( R / ∆ R ) 2.91-3.19 0.6-4.5 0-0.5 A14d25-x01-y01 ln( R / ∆ R ) 3.19-3.47 0.6-2.4, 3.4-4.5 0-0.5, 2.4-3.3 A14d26-x01-y01 ln( R / ∆ R ) 3.47-3.74 0.5-4.5 0.2, 1.2, 2.5, 3.5 A14d27-x01-y01 ln( R / ∆ R ) 3.74-4.02 0.3-4.5 0.2 A14d28-x01-y01 ln( R / ∆ R ) 4.02-4.30 0-1.6, 3.7-4.5 1.7-3.6 -d29-x01-y01 ln( R / ∆ R ) 4.30-4.57 0.2, 0.5,1.2, 2.3-4.5, 0.8, 1-2.2, 3.2, 3.8 -d30-x01-y01 ln( R / ∆ R ) 4.57-4.85 0.2, 1.6-3.3, 3.5 0.3-1.5, 3.7-4.5 -d31-x01-y01 ln( R / ∆ R ) 4.85-5.13 0-3, 3.4-4.5 3.2 A14d32-x01-y01 ln( R / ∆ R ) 5.13-5.41 0.2, 1.7-4 0.4-1.7, 2.75 A14d33-x01-y01 ln( R / ∆ R ) 5.41-5.68 0.2, 2-3 0.5-2, 3.2-4 Monashd34-x01-y01 ln( R / ∆ R ) 5.68-5.96 1.7-2.3 0-1.7, 2.5-3 Monash Table 3: ATLAS 2020 I1790256(LJP) lots β z cut Observable
A14 Monash
Better Tuned01-x01-y01 Calorimeter based 0 0.1 ρ - all Monashd02-x01-y01 Track based 0 0.1 ρ - all Monashd03-x01-y01 Cluster based 1 0.1 ρ [-4.5,-3.7], [-3.5,-2.1], A14 / [-2,-1.3] [-1,-0.5] Monashd04-x01-y01 Track based 1 0.1 ρ [-4.5,-3.7] [-3.5,-0.5] Monashd05-x01-y01 Cluster based 2 0.1 ρ [-4.5,-1.1] -0.7 A14d06-x01-y01 Track based 2 0.1 ρ [-4.5,-3.7] [-3.5,-0.5] Monashd07-x01-y01 Track based 1 0.1 ρ [-4.5,-3.7] [-3.5,-0.5] Monashd16-x01-y01 Track based 1 0.1 r g - all Monashd17-x01-y01 Cluster based 2 0.1 r g [-1.2,-0.2] -0.15 A14d18-x01-y01 Track based 2 0.1 r g -1.1 [-1,-0.1] Monashd19-x01-y01 Central jet / Calorimeter 0 0.1 r g - all Monashd20-x01-y01 Central jet / Track 0 0.1 r g - all Monashd21-x01-y01 Central jet / Cluster 1 0.1 ρ [-4.5,-1] -0.7 A14d22-x01-y01 Central jet / Track 1 0.1 ρ [-4.5,-3.7] [-3.5,-0.5] Monashd23-x01-y01 Central jet / Cluster 2 0.1 ρ [-3.5,-0.9] -0.7 A14d24-x01-y01 Central jet / Track 2 0.1 ρ [-4.5,-3.7] [-3.5,-0.7] Monashd34-x01-y01 Central jet / Track 1 0.1 r g - all Monashd35-x01-y01 Central jet / Cluster 2 0.1 r g [-1.2,-0.4] -0.5,-0.15 A14d36-x01-y01 Central jet / Track 2 0.1 r g -1.1 [-1,-0.1] Monashd37-x01-y01 Forward jet / Calorimeter 0 0.1 r g - all Monashd38-x01-y01 Forward jet / Track 0 0.1 ρ - all Monashd39-x01-y01 Forward jet / Cluster 1 0.1 ρ -4.3 [-4,-0.5] Monashd40-x01-y01 Forward jet / Track 1 0.1 ρ [-4.5,-3.7] [-3.5,-0.5] Monashd41-x01-y01 Forward jet / Cluster 2 0.1 ρ -3.9,[-3.1,-1] -3.5,-0.7 A14d42-x01-y01 Forward jet / Track 2 0.1 ρ [-4.5,-3.7] [-3.5,-0.5] Monashd49-x01-y01 Forward jet / Track 0 0.1 r g all all -d51-x01-y01 Forward jet / Cluster 1 0.1 r g [-0.8,-0.2] [-1.2,-0.8],-0.1 -d52-x01-y01 Forward jet / Track 1 0.1 r g - all Monashd53-x01-y01 Forward jet / Cluster 2 0.1 r g all -0.15 A14d54-x01-y01 Forward jet / Track 2 0.1 r g -1.1 [-1,-0.1] Monash Table 4: ATLAS 2019 I1772062(Soft Drop Jet Observables)Plots Observable Better Tune Performance regions Better Tune
A14 Monash (overall)d01-x01-y01 Nsubjets 0-10 - A14d02-x01-y01 C D norm norm C D norm norm norm β Better Tune Performance regions Better Tune( p leadT >
600 GeV) ( z cut = . A14 Monash (overall)d01-x01-y01 log [( m soft drop / p ungroomed T ) ] 0 [-2.5,-0.5] [-4,-2.5] A14d02-x01-y01 log [( m soft drop / p ungroomed T ) ] 1 [-4.5,-0.5] - A14d03-x01-y01 log [( m soft drop / p ungroomed T ) ] 2 [-4.5,-0.5] - A14Table 6: ATLAS 2017 I1637587(Soft Drop Mass) .2. Envelope plots Having decided the range of values for each parameter, we visualise the region of the distributions to check its utilityand hence it is important before proceeding further. This can be done using the PROFESSOR tool by generating envelopesplots with the comand prof2-envelopes . The envelope plots show an area covering the distributions which indicatesthe bin values that the observables can take within the selected parameter ranges. These are shown in the following subsections. x = ATLAS_2017_I1637587_d01-x01-y010.100.150.200.250.30 f ( x ) MeanData 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 x = ATLAS_2017_I1637587_d02-x01-y010.000.050.100.150.200.25 f ( x ) MeanData4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 x = ATLAS_2017_I1637587_d03-x01-y010.000.050.100.150.200.250.30 f ( x ) MeanData
Figure 6: Envelope plots for ATLAS 2017 I1637587
As can be seen in Figure 6, The envelopes cover the reference data in almost every bin and hence we can say that therange selected for the parameters are appropriate.
As can be seen in Figure 7, the envelopes do not entirely cover the reference data. This is because we have reached alimit as to how much the distribution can be further fitted to the data with Pythia. Thus we consider this suitable for thepurpose of this report and proceed with our set of parameter ranges.12 x = ATLAS_2020_I1790256_d05-x01-y010.20.30.40.50.6 f ( x ) MeanData 1 2 3 4 5 6 x = ATLAS_2020_I1790256_d06-x01-y010.20.30.40.50.6 f ( x ) MeanData1 2 3 4 5 6 x = ATLAS_2020_I1790256_d07-x01-y010.20.30.40.50.6 f ( x ) MeanData 1 2 3 4 5 6 x = ATLAS_2020_I1790256_d08-x01-y010.10.20.30.40.50.6 f ( x ) MeanData
Figure 7: Envelope plots for Lund Jet Plane
In the Figure 8, we see that the envelopes cover the data points in almost all the bins of our distributions of interest i.esoft drop jet mass from the soft drop jet observables analysis. Thus the parameter ranges are suitable for proceeding totune the distributions. 13 .5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 x = ATLAS_2019_I1772062_d19-x01-y010.20.30.40.50.60.70.8 f ( x ) MeanData 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 x = ATLAS_2019_I1772062_d20-x01-y010.20.30.40.50.60.70.8 f ( x ) MeanData4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 x = ATLAS_2019_I1772062_d21-x01-y010.00.20.40.60.8 f ( x ) MeanData 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 x = ATLAS_2019_I1772062_d22-x01-y010.00.20.40.60.8 f ( x ) MeanData4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 x = ATLAS_2019_I1772062_d23-x01-y010.00.20.40.60.81.01.21.4 f ( x ) MeanData 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 x = ATLAS_2019_I1772062_d24-x01-y010.00.20.40.60.81.01.2 f ( x ) MeanData
Figure 8: Envelope plots for Soft Drop Mass observable .3. Weight file The weight file assigns weights to the distributions to be tuned and hence is manually changed depending on ourinterests. For this project, a total of 16 distributions were assigned weights greater than 1 and 6 distributions were givenno weight i.e 0 for obtaining the Common Tune. These are :
Analysis Distribution code WeightATLAS 2020 I1790256 d03-x01-y01 106ATLAS 2020 I1790256 d04-x01-y01 112ATLAS 2020 I1790256 d05-x01-y01 108ATLAS 2020 I1790256 d06-x01-y01 20ATLAS 2020 I1790256 d07-x01-y01 16.6ATLAS 2020 I1790256 d08-x01-y01 16ATLAS 2020 I1790256 d09-x01-y01 16ATLAS 2019 I1772062 d19-x01-y01 75ATLAS 2019 I1772062 d20-x01-y01 75ATLAS 2019 I1772062 d21-x01-y01 200ATLAS 2019 I1772062 d22-x01-y01 80ATLAS 2019 I1772062 d23-x01-y01 200ATLAS 2019 I1772062 d24-x01-y01 80ATLAS 2017 I1637587 d01-x01-y01 500ATLAS 2017 I1637587 d02-x01-y01 500ATLAS 2017 I1637587 d03-x01-y01 500ATLAS 2019 I1772062 d61-x01-y01 0ATLAS 2019 I1772062 d62-x01-y01 0ATLAS 2019 I1772062 d79-x01-y01 0ATLAS 2019 I1772062 d80-x01-y01 0ATLAS 2019 I1772062 d97-x01-y01 0ATLAS 2019 I1772062 d98-x01-y01 0