10
2 Phenomenology of Jet Substructure
can be efficiently reconstructed as tracks, neutral particles only develop showers in
the calorimeters and the possibility to resolve two separate showers depends on the
granularity of the calorimeter and the lateral shower development. Hence, it becomes
more difficult to separate two adjacent particles in dense environments, such as high
momentum jets, and the situation is aggravated by the presence of hadronic showers
from charged hadrons. Often it is impossible to build one calorimeter cluster per
neutral particle. A way to improve the angular resolution in jet substructure analyses
is to combine measurements from the tracking detectors and calorimeters. Using
combined detector measurements as input to jet algorithms, for example using a
particle flow approach [71–75], results in improved resolutions of jet substructure
observables, compared to using only tracks or only calorimeter clusters.
An important aspect of experimental analyses at the LHC is the calibration of jets,
necessitated by the non-compensating nature of hadron calorimeters, suppression of
electronic noise, tracking inefficiencies, dead material in front of calorimeters, the
influence of pileup and other effects. While the calibration of the total jet energy scale
is an important aspect in all analyses using jets, the precise knowledge of the jet mass
scale and the detector response to jet substructure observables and jet tagging algorithms is specific to jet substructure analyses. Calibrating the jet energy scale results
in a change of the magnitude of the jet’s four-momentum, where the jet mass scale
comprises an additional degree of freedom that can not be constrained by the typical methods of balancing a jet with a well-calibrated reference object. The jet mass
scale is usually calibrated using jets from fully-merged, highly boosted W → qq
decays, facilitating a calibration of the peak position in the jet mass distribution.
Measurements of the jet mass distribution from light quark and gluon jets, as well
as from fully-hadronic highly-boosted W , Z and t decays allow for precise tests of
the modelling of perturbative and non-perturbative effects in jet production. Similar
measurements can also be used to study the detector response to jet substructure
observables and their modelling in simulation. A mis-modelling of variables used
for tagging, either in the detector simulation or on the level of the underlying physics,
can result in a wrong estimation of the tagging efficiency or the misidentification rate,
with important consequences for measurements. In order to overcome this limitation, measurements of tagging efficiencies and misidentification rates are performed
in samples enriched with the particle decays in question. While these measurements
do not help to understand the cause of the mis-modelling or to improve the description of jet substructure distributions, they can be used to correct the efficiencies in
simulation. It is these measurements that have enabled the use of jet substructure taggers in numerous physics analyses since the beginning of data taking at the LHC. The
increased statistical precision from a data sample corresponding to about 150 fb
−1
per experiment at a centre-of-mass energy of 13 TeV can now be used to improve our
understanding of the detector response to jet substructure algorithms, the underlying
physics and the performance differences of taggers.
2 Phenomenology of Jet Substructure
can be efficiently reconstructed as tracks, neutral particles only develop showers in
the calorimeters and the possibility to resolve two separate showers depends on the
granularity of the calorimeter and the lateral shower development. Hence, it becomes
more difficult to separate two adjacent particles in dense environments, such as high
momentum jets, and the situation is aggravated by the presence of hadronic showers
from charged hadrons. Often it is impossible to build one calorimeter cluster per
neutral particle. A way to improve the angular resolution in jet substructure analyses
is to combine measurements from the tracking detectors and calorimeters. Using
combined detector measurements as input to jet algorithms, for example using a
particle flow approach [71–75], results in improved resolutions of jet substructure
observables, compared to using only tracks or only calorimeter clusters.
An important aspect of experimental analyses at the LHC is the calibration of jets,
necessitated by the non-compensating nature of hadron calorimeters, suppression of
electronic noise, tracking inefficiencies, dead material in front of calorimeters, the
influence of pileup and other effects. While the calibration of the total jet energy scale
is an important aspect in all analyses using jets, the precise knowledge of the jet mass
scale and the detector response to jet substructure observables and jet tagging algorithms is specific to jet substructure analyses. Calibrating the jet energy scale results
in a change of the magnitude of the jet’s four-momentum, where the jet mass scale
comprises an additional degree of freedom that can not be constrained by the typical methods of balancing a jet with a well-calibrated reference object. The jet mass
scale is usually calibrated using jets from fully-merged, highly boosted W → qq
decays, facilitating a calibration of the peak position in the jet mass distribution.
Measurements of the jet mass distribution from light quark and gluon jets, as well
as from fully-hadronic highly-boosted W , Z and t decays allow for precise tests of
the modelling of perturbative and non-perturbative effects in jet production. Similar
measurements can also be used to study the detector response to jet substructure
observables and their modelling in simulation. A mis-modelling of variables used
for tagging, either in the detector simulation or on the level of the underlying physics,
can result in a wrong estimation of the tagging efficiency or the misidentification rate,
with important consequences for measurements. In order to overcome this limitation, measurements of tagging efficiencies and misidentification rates are performed
in samples enriched with the particle decays in question. While these measurements
do not help to understand the cause of the mis-modelling or to improve the description of jet substructure distributions, they can be used to correct the efficiencies in
simulation. It is these measurements that have enabled the use of jet substructure taggers in numerous physics analyses since the beginning of data taking at the LHC. The
increased statistical precision from a data sample corresponding to about 150 fb
−1
per experiment at a centre-of-mass energy of 13 TeV can now be used to improve our
understanding of the detector response to jet substructure algorithms, the underlying
physics and the performance differences of taggers.