100
4 Standard Model Measurements
4.1.4 Jet Substructure in Light Flavour and Gluon Jets
Beyond measurements of the jet mass, more complex jet substructure observables
can be used to study the energy flow inside jets, localised spots of high energy density,
angular particle-particle correlations or particle multiplicities. These measurements
help to improve our understanding of jet formation, as observables have different
sensitivities to contributions from perturbative and non-perturbative effects. These
measurements can also be used to study the approximations made in parton shower
and hadronisation models. Once unfolded at the particle level, distributions of jet
substructure observables can be used to constrain the free parameters in event generators, or for the determination of fundamental parameters of the SM, like the value
of the strong coupling [569, 570].
The first measurement of jet substructure at the LHC has been a measurement of
differential and integrated jet shapes from ATLAS using 3 pb
−1 of 7 TeV data [571].
While this measurement is limited in statistical precision, it shows the power of jet
substructure observables for the determination of the quark and gluon fractions in
jets. Unfolded measurements of observables related to tagging studies have also been
performed by ATLAS and CMS on 7 TeV data recorded in 2010 [551, 572, 573].
These measurements have been carried out on inclusive jet data, dominated by dijet
production with a large fraction of gluon jets at low p T . Distributions of k T splitting
scales, N -subjettiness, jet width, transverse jet size, eccentricity, planar flow and
angularity have been measured. The event generators Pythia and Herwig are in
good agreement with the data, which is a striking achievement considering that the
simulations available in 2011 have been tuned mostly to LEP, SPS and Tevatron
data [291, 574], with only charged particle multiplicities entering the tunes [575–
577]. These measurements gave confidence in the use of jet substructure at the LHC
and have preceded further developments of substructure techniques.
Measurements of charged particle multiplicities n ch are important for several
aspects of physics at the LHC. Multiplicity distributions are affected by the details of
particle production, from quark/gluon prodution to perturbative effects in the parton
shower, to hadronisation and the decay chains of short-lived hadrons. The underlying
event and pileup interactions also have a large impact on multiplicity distributions.
Charged particle multiplicities constitute an important input to a number of substructure taggers, most notably q/g discrimination (see Sect. 3.5.1). ATLAS and CMS
have measured n ch distributions in jets using 7 TeV data [573, 578]. ATLAS has also
performed measurements using 8 [579] and 13 TeV [580] data. The challenge of these
measurements lies in a precise understanding of the track reconstruction efficiencies
over a large range of p T and |η|. Also, the n ch distribution depends on the q/g content
of the sample, on jet p T and η, and on the track p T . Dominant experimental effects
include the track reconstruction efficiency, the rate of fake and secondary tracks, the
track momentum scale, and density effects from pixel and strip cluster merging for
high p T jets [415–418]. Nevertheless, precise measurements with uncertainties at the
few percent level have been performed, enabled by detailed studies of these effects.
The measured average n ch as a function of jet p T is shown in Fig. 4.5 (left) using
4 Standard Model Measurements
4.1.4 Jet Substructure in Light Flavour and Gluon Jets
Beyond measurements of the jet mass, more complex jet substructure observables
can be used to study the energy flow inside jets, localised spots of high energy density,
angular particle-particle correlations or particle multiplicities. These measurements
help to improve our understanding of jet formation, as observables have different
sensitivities to contributions from perturbative and non-perturbative effects. These
measurements can also be used to study the approximations made in parton shower
and hadronisation models. Once unfolded at the particle level, distributions of jet
substructure observables can be used to constrain the free parameters in event generators, or for the determination of fundamental parameters of the SM, like the value
of the strong coupling [569, 570].
The first measurement of jet substructure at the LHC has been a measurement of
differential and integrated jet shapes from ATLAS using 3 pb
−1 of 7 TeV data [571].
While this measurement is limited in statistical precision, it shows the power of jet
substructure observables for the determination of the quark and gluon fractions in
jets. Unfolded measurements of observables related to tagging studies have also been
performed by ATLAS and CMS on 7 TeV data recorded in 2010 [551, 572, 573].
These measurements have been carried out on inclusive jet data, dominated by dijet
production with a large fraction of gluon jets at low p T . Distributions of k T splitting
scales, N -subjettiness, jet width, transverse jet size, eccentricity, planar flow and
angularity have been measured. The event generators Pythia and Herwig are in
good agreement with the data, which is a striking achievement considering that the
simulations available in 2011 have been tuned mostly to LEP, SPS and Tevatron
data [291, 574], with only charged particle multiplicities entering the tunes [575–
577]. These measurements gave confidence in the use of jet substructure at the LHC
and have preceded further developments of substructure techniques.
Measurements of charged particle multiplicities n ch are important for several
aspects of physics at the LHC. Multiplicity distributions are affected by the details of
particle production, from quark/gluon prodution to perturbative effects in the parton
shower, to hadronisation and the decay chains of short-lived hadrons. The underlying
event and pileup interactions also have a large impact on multiplicity distributions.
Charged particle multiplicities constitute an important input to a number of substructure taggers, most notably q/g discrimination (see Sect. 3.5.1). ATLAS and CMS
have measured n ch distributions in jets using 7 TeV data [573, 578]. ATLAS has also
performed measurements using 8 [579] and 13 TeV [580] data. The challenge of these
measurements lies in a precise understanding of the track reconstruction efficiencies
over a large range of p T and |η|. Also, the n ch distribution depends on the q/g content
of the sample, on jet p T and η, and on the track p T . Dominant experimental effects
include the track reconstruction efficiency, the rate of fake and secondary tracks, the
track momentum scale, and density effects from pixel and strip cluster merging for
high p T jets [415–418]. Nevertheless, precise measurements with uncertainties at the
few percent level have been performed, enabled by detailed studies of these effects.
The measured average n ch as a function of jet p T is shown in Fig. 4.5 (left) using