Public Tau Trigger Plots for Collision Data

Introduction

Approved plots that can be shown by ATLAS speakers at conferences and similar events.
Please do not add figures on your own. Contact the responsible project leader in case of questions and/or suggestions.

Run2

Publications

• The ATLAS Tau Trigger in Run 2
  ATLAS-CONF-2017-061
• Performance of the ATLAS Trigger System in 2015
  TRIG-2016-01

Preliminary Plots:

Plots for ACAT 2021

Transverse energy resolution of hadronically-decaying tau leptons reconstructed in the ATLAS High Level Trigger (HLT). A requirement of ET ≥ 8 GeV is applied at Level 1, and no selection is applied at the HLT. Resolutions are shown for 1-prong and 3-prong tau leptons as a function of the dressed visible pT of the true tau lepton, for two different Tau Energy Scale (TES) calibrations. The baseline calibration (ATLAS-CONF-2017-061) consists of a pileup subtraction and a parametrized response correction. It was used throughout Run-2 data taking. The MVA calibration is based on a boosted regression tree that uses calorimeter energy and shower shape information, as well as the average number of pileup interactions. It was deployed online in July 2018. The resolution is computed from the ratio of the calibrated ET of HLT taus and the dressed visible pT of the corresponding generated tau leptons (which includes photons from final state radiation within a cone of radius 0.2 around the visible momentum). The resolution is defined as half of the 68th quantile of the distribution. Results are obtained with ɣ*(𝜏𝜏) Monte Carlo simulation using the multi-threaded Athena reconstruction framework developed by ATLAS for Run-3 data taking.

[png] [pdf] [eps]

Transverse energy resolution of hadronically-decaying tau leptons reconstructed in the ATLAS High Level Trigger (HLT). A requirement of ET ≥ 8 GeV is applied at Level 1, and no selection is applied at the HLT. Resolutions are shown for 1-prong and 3-prong tau leptons as a function of the average number of pileup interactions, for two different Tau Energy Scale (TES) calibrations. The baseline calibration (ATLAS-CONF-2017-061) consists of a pileup subtraction and a parametrized response correction. It was used throughout Run-2 data taking. The MVA calibration is based on a boosted regression tree that uses calorimeter energy and shower shape information, as well as the average number of pileup interactions. It was deployed online in July 2018. The resolution is computed from the ratio of the calibrated ET of HLT taus and the dressed visible pT of the corresponding generated tau leptons (which includes photons from final state radiation within a cone of radius 0.2 around the visible momentum). The resolution is defined as half of the 68th quantile of the distribution. Results are obtained with ɣ*(𝜏𝜏) Monte Carlo simulation using the multi-threaded Athena reconstruction framework developed by ATLAS for Run-3 data taking.

[png] [pdf] [eps]

Plots for Summer 2019

Average rate of 𝜏-lepton triggers as a function of the average number of pileup interactions, in pp collisions at √s = 13 TeV delivered by the LHC between July and October 2018, corresponding to an integrated luminosity of 36.8 fb-1. Rates are shown for a) the single-tau trigger with an ET threshold of 160 GeV at the High Level Trigger (HLT), and b) the di-tau trigger with a L1Topo and a jet requirement at Level 1, and ET thresholds of 35 and 25 GeV at the HLT. The medium BDT (1-3 fast tracks) trigger described in ATLAS-CONF-2017-061 applies a requirement on the number of tracks reconstructed by a Fast Track Finder algorithm at HLT preselection level, and uses a Boosted Decision Tree algorithm for 𝜏-lepton identification. The track preselection reduces CPU and rates but induces inefficiency for 3-prong 𝜏-leptons at high pileup. The medium BDT (1-3 precision tracks) trigger is identical, except that the track multiplicity requirement is applied to precision tracks reconstructed at a later HLT stage. Efficiency at high pileup is recovered at the expense of a slight rate increase. The medium RNN (0-3 precision tracks) trigger applies a requirement on the number of precision tracks, and uses a Recurrent Neural Network algorithm for 𝜏-lepton identification. The larger increase in the RNN di-tau trigger rate at high pileup comes from HLT tau candidates with no associated track, which were included in the RNN trigger to recover efficiency at low ET. The single-tau trigger does not exhibit this behaviour due to its high-ET threshold.

a) [png] [pdf] [eps]

b) [png] [pdf] [eps]

Efficiency of 𝜏-lepton triggers as a function of the offline 𝜏-lepton transverse momentum pT, for an ET threshold of 12 GeV and a medium isolation requirement at Level 1, and an ET threshold of 25 GeV at the High Level Trigger (HLT). The efficiencies are estimated from Monte Carlo simulation using prompt 𝜏-leptons produced in W(𝜏ν)+jets and tt events. Efficiencies are computed with respect to a) 1-prong and b) 3-prong offline 𝜏-leptons passing the medium BDT identification criteria. The three 𝜏-lepton trigger versions present in the 2018 trigger menu are shown. The HLT tau25 medium BDT (1-3 fast tracks) trigger described in ATLAS-CONF-2017-061 applies a requirement on the number of tracks reconstructed by a Fast Track Finder algorithm at HLT preselection level, and uses a Boosted Decision Tree algorithm for 𝜏-lepton identification. This version has run for the whole Run-2 period. The HLT tau25 medium BDT (1-3 precision tracks) trigger is identical, except that the track multiplicity requirement is only applied to tracks from precision tracking reconstructed at a later HLT stage; this version has run for the whole 2018 data taking. The HLT tau25 medium RNN (0-3 precision tracks) trigger uses a Boosted Regression Tree for energy calibration (instead of the pileup and response corrections applied in BDT triggers) and a Recurrent Neural Network algorithm for 𝜏-lepton identification; this version was deployed in July 2018. Only statistical uncertainties are displayed. The three HLT versions have comparable trigger rates.	a) [png] [pdf] [eps]
	b) [png] [pdf] [eps]
Efficiency of 𝜏-lepton triggers as a function of the average number of pileup interactions, for an ET threshold of 12 GeV and a medium isolation requirement at Level 1, and an ET threshold of 25 GeV at the High Level Trigger (HLT). The efficiencies are estimated from Monte Carlo simulation using prompt 𝜏-leptons produced in W(𝜏ν)+jets and tt events. Efficiencies are computed with respect to a) 1-prong and b) 3-prong offline 𝜏-leptons with pT > 30 GeV passing the medium BDT identification criteria. The three 𝜏-lepton trigger versions present in the 2018 trigger menu are shown. The HLT tau25 medium BDT (1-3 fast tracks) trigger described in ATLAS-CONF-2017-061 applies a requirement on the number of tracks reconstructed by a Fast Track Finder (FTF) algorithm at HLT preselection level, and uses a Boosted Decision Tree algorithm for 𝜏-lepton identification. This version has run for the whole Run-2 period. The inefficiency of the FTF track multiplicity requirement at high pileup is due to the larger number of fake tracks reconstructed from random hits alignment in the inner detector. The HLT tau25 medium BDT (1-3 precision tracks) trigger is identical, except that the track multiplicity requirement is only applied to tracks from precision tracking reconstructed at a later HLT stage; this version has run for the whole 2018 data taking. The HLT tau25 medium RNN (0-3 precision tracks) trigger uses a Boosted Regression Tree for energy calibration (instead of the pileup and response corrections applied in BDT triggers) and a Recurrent Neural Network algorithm for 𝜏-lepton identification; this version was deployed in July 2018. Only statistical uncertainties are displayed. The three HLT versions have comparable trigger rates.	a) [png] [pdf] [eps]
	b) [png] [pdf] [eps]

Plots for ICHEP 2016 (Obsolete) ATL-COM-DAQ-2016-088

Tau trigger efficiency measured in data and compared to simulation, with respect to offline reconstructed tau candidate with one or three tracks and passing the offline medium identification criteria, as function of the offline transverse momentum. The trigger efficiency is measured in a tag and probe analysis with 𝑍 → 𝜏𝜏 → 𝜇𝜏ℎ𝑎𝑑 event from the 2016 dataset in 13TeV collision (8.0fb-1). The corresponding online tau requirements are transverse momentum above 25 GeV, between one and three tracks and pass the online medium identification. The error bars correspond to statistical uncertainty.	[png] [pdf] [eps]
Tau trigger efficiency measured in data with respect to offline reconstructed tau candidate with one or three tracks and passing the offline medium identification criteria, as function of the offline transverse momentum. The trigger efficiency is measured in a tag and probe analysis with 𝑍 → 𝜏𝜏 → 𝜇𝜏ℎ𝑎𝑑 event from the 2016 dataset in 13TeV collision (8.0fb-1). The corresponding online tau requirements are transverse momentum above 12 GeV and pass the isolation criteria at L1 and above 25 GeV, between one and three tracks and pass the online medium identification at HLT. The error bars correspond to statistical uncertainty.	[png] [pdf] [eps]
Tau trigger efficiency measured in data with respect to offline reconstructed tau candidate with transverse momentum above 30 GeV, one or three tracks and passing the offline medium identification criteria, as function of the number of pileup. The trigger efficiency is measured in a tag and probe analysis with 𝑍 → 𝜏𝜏 → 𝜇𝜏ℎ𝑎𝑑 event from the 2016 dataset in 13TeV collision (8.0fb-1). The corresponding online tau requirements are transverse momentum above 12 GeV and pass the isolation criteria at L1 and above 25 GeV, between one and three tracks and pass the online medium identification at HLT. The error bars correspond to statistical uncertainty.	[png] [pdf] [eps]

Plots for LHCC 2015 (Obsolete) ATL-COM-PHYS-2015-1392

The BDT tau identification score for online tau candidates passing the HLT tau trigger with transverse momentum threshold at 25 GeV and online medium identification requirement. The HLT tau candidates are observed in an enriched sample of Z→ττ→μτ(had) events from the 2015 dataset in 13 TeV collisions, corresponding to an integrated luminosity of 3.3fb-1. These events are collected using a single muon trigger. The HLT tau candidates are matched with offline tau candidates passing the offline medium tau identification. Event selections and background estimations are described in ATL-PHYS-PUB-2015-025 and in Eur. Phys. J. C75 (2015) 303. Only statistical uncertainties are shown.

[png] [pdf] [eps]

Tau trigger efficiency measured in data and compared to simulation, with respect to offline reconstructed tau candidates with transverse momentum above 25 GeV, one or three tracks and passing the offline medium identification requirement. The corresponding online tau candidate is required to have a transverse momentum of at least 25 GeV, between one and three tracks and pass the online medium identification requirement. The trigger efficiency is measured in an enriched sample of Z→ττ→μτ(had) events from the 2015 dataset in 13 TeV collisions, corresponding to an integrated luminosity of 3.3fb-1. These events are collected using a single muon trigger. The efficiency is plotted as function of the transverse momentum of the offline tau candidate. Error bars are statistical uncertainties. Further details are described in ATL-PHYSPUB-2015-025 and in Eur. Phys. J. C75 (2015) 303. Only statistical uncertainties are shown.

[png] [pdf] [eps]

Plots for Lepton Photon 2015 (Obsolete) ATL-COM-DAQ-2015-125

The transverse momentum distribution of online tau candidates passing the HLT tau trigger with transverse momentum threshold at 25 GeV and online medium identification requirement. The HLT tau candidates are observed in an enriched sample of Z→ττ→μτ(had) events and are considered if matched to a tau candidate passing the offline medium tau identification criteria. Data has been recorded in the first 13 TeV collisions in 2015. The data are compared to a signal Z→ττ Monte Carlo and a combined background consisting of Z→ll, W+jets, and Top Monte Carlo and a data driven multijet estimate. Only statistical uncertainties are shown. The ratio of the observed data to the expected signal and background events is also shown, where the red band shows the statistical uncertainty on the total expectation. Event selections and background estimations are described in ATL-PHYS-PUB-2015-025 and in Eur. Phys. J. C75 (2015) 303.	[png] [pdf] [eps]
The pseudo-rapidity distribution of online tau candidates passing the HLT tau trigger with transverse momentum threshold at 25 GeV and online medium identification requirement. The HLT tau candidates are observed in an enriched sample of Z→ττ→μτ(had) events and are considered if matched to a tau candidate passing the offline medium tau identification criteria. Data has been recorded in the first 13 TeV collisions in 2015. The data are compared to a signal Z→ττ Monte Carlo and a combined background consisting of Z→ll, W+jets, and Top Monte Carlo and a data driven multijet estimate. Only statistical uncertainties are shown. The ratio of the observed data to the expected signal and background events is also shown, where the red band shows the statistical uncertainty on the total expectation. Event selections and background estimations are described in ATL-PHYS-PUB-2015-025 and in Eur. Phys. J. C75 (2015) 303.	[png] [pdf] [eps]
The track multiplicity distribution, for tracks within a ΔR<0.2 of the tau axis, of online tau candidates passing the HLT tau trigger with transverse momentum threshold at 25 GeV and online medium identification requirement. The HLT tau candidates are observed in an enriched sample of Z→ττ→μτ(had) events and are considered if matched to a tau candidate passing the offline medium tau identification criteria. Data has been recorded in the first 13 TeV collisions in 2015. The data are compared to a signal Z→ττ Monte Carlo and a combined background consisting of Z→ll, W+jets, and Top Monte Carlo and a data driven multijet estimate. Only statistical uncertainties are shown. The ratio of the observed data to the expected signal and background events is also shown, where the red band shows the statistical uncertainty on the total expectation. Event selections and background estimations are described in ATL-PHYS-PUB-2015-025 and in Eur. Phys. J. C75 (2015) 303.	[png] [pdf] [eps]
The track multiplicity distribution, for tracks within a 0.2<ΔR<0.4 of the tau axis, of online tau candidates passing the HLT tau trigger with transverse momentum threshold at 25 GeV and online medium identification requirement. The HLT tau candidates are observed in an enriched sample of Z→ττ→μτ(had) events and are considered if matched to a tau candidate passing the offline medium tau identification criteria. Data has been recorded in the first 13 TeV collisions in 2015. The data are compared to a signal Z→ττ Monte Carlo and a combined background consisting of Z→ll, W+jets, and Top Monte Carlo and a data driven multijet estimate. Only statistical uncertainties are shown. The ratio of the observed data to the expected signal and background events is also shown, where the red band shows the statistical uncertainty on the total expectation. Event selections and background estimations are described in ATL-PHYS-PUB-2015-025 and in Eur. Phys. J. C75 (2015) 303.	[png] [pdf] [eps]
The fraction of transverse energy deposited in a cone of ∆R < 0.1 to the energy deposited in a cone of ∆R < 0.2 with respect to the tau axis for online tau candidates passing the HLT tau trigger with transverse momentum threshold at 25 GeV and online medium identification requirement. The HLT tau candidates are observed in an enriched sample of Z→ττ→μτ(had) events and are considered if matched to a tau candidate passing the offline medium tau identification criteria. Data has been recorded in the first 13 TeV collisions in 2015. The data are compared to a signal Z→ττ Monte Carlo and a combined background consisting of Z→ll, W+jets, and Top Monte Carlo and a data driven multijet estimate. Only statistical uncertainties are shown. The ratio of the observed data to the expected signal and background events is also shown, where the red band shows the statistical uncertainty on the total expectation. Event selections and background estimations are described in ATL-PHYS-PUB-2015-025 and in Eur. Phys. J. C75 (2015) 303.	[png] [pdf] [eps]
The BDT tau identification score for online tau candidates passing the HLT tau trigger with transverse momentum threshold at 25 GeV and online medium identification requirement.The HLT tau candidates are observed in an enriched sample of Z→ττ→μτ(had) events and are considered if matched to a tau candidate passing the offline medium tau identification criteria. Data has been recorded in the first 13 TeV collisions in 2015. The data are compared to a signal Z→ττ Monte Carlo and a combined background consisting of Z→ll, W+jets, and Top Monte Carlo and a data driven multijet estimate. Only statistical uncertainties are shown. The ratio of the observed data to the expected signal and background events is also shown, where the red band shows the statistical uncertainty on the total expectation. Event selections and background estimations are described in ATL-PHYS-PUB-2015-025 and in Eur. Phys. J. C75 (2015) 303.	[png] [pdf] [eps]
The Level-1 tau trigger efficiency measured in data and compared to simulation, with respect to offline reconstructed tau candidates with transverse momentum above 20 GeV, one or three tracks and passing the offline medium identification criteria. The online tau candidates are reconstructed at Level-1 of the ATLAS trigger and are required to have a transverse energy of 12 GeV and pass the medium isolation criteria. The trigger efficiency is measured in an enriched sample of Z→ττ→μτ(had) events recorded in the first 13 TeV collisions in 2015. The efficiency is plotted as function of the transverse momentum of the offline tau candidate. Error bars are statistical uncertainties. Further details are described in ATL-PHYS-PUB-2015-025 and in Eur. Phys. J. C75 (2015) 303.	[png] [pdf] [eps]
The combined Level-1 and HLT tau trigger efficiency measured in data and compared to simulation, with respect to offline reconstructed tau candidates with transverse momentum above 20 GeV, one or three tracks and passing the offline medium identification criteria. The corresponding online tau candidate is required to have a transverse momentum of at least 25 GeV, between one and three tracks and pass the online medium identification. The trigger efficiency is measured in an enriched sample of Z→ττ→μτ(had) events recorded in the first 13 TeV collisions in 2015. The efficiency is plotted as function of the transverse momentum of the offline tau candidate. Error bars are statistical uncertainties. Further details are described in ATL-PHYS-PUB-2015-025 and in Eur. Phys. J. C75 (2015) 303.	[png] [pdf] [eps]
The transverse momentum distribution of online tau candidates passing the HLT tau trigger with transverse momentum threshold at 25 GeV and online medium identification requirement. The HLT tau candidates are observed in an enriched sample of W→μν events and are likely to be jets originating from quarks or gluons that are reconstructed as tau jets and accepted by the tau trigger. Such fake tau candidates represent the dominant source of background in the measurement of the performance of the tau trigger for real tau leptons in Z→ττ events. Data has been recorded in the first 13 TeV collisions in 2015. The data are compared to a signal W→μν+jets Monte Carlo and a combined background consisting of Z→ττ, Z→ll, and Top Monte Carlo and a data driven multijet estimate. Only statistical uncertainties are shown. The ratio of the observed data to the expected signal and background events is also shown, where the red band shows the statistical uncertainty on the total expectation. Event selections and background estimations are described in ATL-PHYS-PUB-2015-025 and in Eur. Phys. J. C75 (2015) 303.	[png] [pdf] [eps]
The pseudo-rapidity distribution of online tau candidates passing the HLT tau trigger with transverse momentum threshold at 25 GeV and online medium identification requirement.The HLT tau candidates are observed in an enriched sample of W→μν events and are likely to be jets originating from quarks or gluons that are reconstructed as tau jets and accepted by the tau trigger. Such fake tau candidates represent the dominant source of background in the measurement of the performance of the tau trigger for real tau leptons in Z→ττ events. Data has been recorded in the first 13 TeV collisions in 2015. The data are compared to a signal W→μν+jets Monte Carlo and a combined background consisting of Z→ττ, Z→ll, and Top Monte Carlo and a data driven multijet estimate. Only statistical uncertainties are shown. The ratio of the observed data to the expected signal and background events is also shown, where the red band shows the statistical uncertainty on the total expectation. Event selections and background estimations are described in ATL-PHYS-PUB-2015-025 and in Eur. Phys. J. C75 (2015) 303.	[png] [pdf] [eps]
The track multiplicity distribution, for tracks within a ΔR<0.2 of the tau axis, of online tau candidates passing the HLT tau trigger with transverse momentum threshold at 25 GeV and online medium identification requirement. The HLT tau candidates are observed in an enriched sample of W→μν events and are likely to be jets originating from quarks or gluons that are reconstructed as tau jets and accepted by the tau trigger. Such fake tau candidates represent the dominant source of background in the measurement of the performance of the tau trigger for real tau leptons in Z→ττ events. Data has been recorded in the first 13 TeV collisions in 2015. The data are compared to a signal W→μν+jets Monte Carlo and a combined background consisting of Z→ττ, Z→ll, and Top Monte Carlo and a data driven multijet estimate. Only statistical uncertainties are shown. The ratio of the observed data to the expected signal and background events is also shown, where the red band shows the statistical uncertainty on the total expectation. Event selections and background estimations are described in ATL-PHYS-PUB-2015-025 and in Eur. Phys. J. C75 (2015) 303.	[png] [pdf] [eps]
The track multiplicity distribution, for tracks within a 0.2<ΔR<0.4 of the tau axis, of online tau candidates passing the HLT tau trigger with transverse momentum threshold at 25 GeV and online medium identification requirement. The HLT tau candidates are observed in an enriched sample of W→μν events and are likely to be jets originating from quarks or gluons that are reconstructed as tau jets and accepted by the tau trigger. Such fake tau candidates represent the dominant source of background in the measurement of the performance of the tau trigger for real tau leptons in Z→ττ events. Data has been recorded in the first 13 TeV collisions in 2015. The data are compared to a signal W→μν+jets Monte Carlo and a combined background consisting of Z→ττ, Z→ll, and Top Monte Carlo and a data driven multijet estimate. Only statistical uncertainties are shown. The ratio of the observed data to the expected signal and background events is also shown, where the red band shows the statistical uncertainty on the total expectation. Event selections and background estimations are described in ATL-PHYS-PUB-2015-025 and in Eur. Phys. J. C75 (2015) 303.	[png] [pdf] [eps]
The fraction of transverse energy deposited in a cone of ∆R < 0.1 to the energy deposited in a cone of ∆R < 0.2 with respect to the tau axis for online tau candidates passing the HLT tau trigger with transverse momentum threshold at 25 GeV and online medium identification requirement. The HLT tau candidates are observed in an enriched sample of W→μν events and are likely to be jets originating from quarks or gluons that are reconstructed as tau jets and accepted by the tau trigger. Such fake tau candidates represent the dominant source of background in the measurement of the performance of the tau trigger for real tau leptons in Z→ττ events. Data has been recorded in the first 13 TeV collisions in 2015. The data are compared to a signal W→μν+jets Monte Carlo and a combined background consisting of Z→ττ,Z→ll, and Top Monte Carlo and a data driven multijet estimate. Only statistical uncertainties are shown. The ratio of the observed data to the expected signal and background events is also shown, where the red band shows the statistical uncertainty on the total expectation. Event selections and background estimations are described in ATL-PHYS-PUB-2015-025 and in Eur. Phys. J. C75 (2015) 303.	[png] [pdf] [eps]
The BDT tau identification score for online tau candidates passing the HLT tau trigger with transverse momentum threshold at 25 GeV and online medium identification requirement. The HLT tau candidates are observed in an enriched sample of W→μν events and are likely to be jets originating from quarks or gluons that are reconstructed as tau jets and accepted by the tau trigger. Such fake tau candidates represent the dominant source of background in the measurement of the performance of the tau trigger for real tau leptons in Z→ττ events. Data has been recorded in the first 13 TeV collisions in 2015. The data are compared to a signal W→μν+jets Monte Carlo and a combined background consisting of Z→ττ,Z→ll, and Top Monte Carlo and a data driven multijet estimate. Only statistical uncertainties are shown. The ratio of the observed data to the expected signal and background events is also shown, where the red band shows the statistical uncertainty on the total expectation. Event selections and background estimations are described in ATL-PHYS-PUB-2015-025 and in Eur. Phys. J. C75 (2015) 303	[png] [pdf] [eps]

Plots for EPS 2015 (Obsolete) ATL-COM-DAQ-2015-097

Level 1 rates before prescale versus the instantaneous luminosity measured by ATLAS for L1 single-tau and combined tau+X chains. ‘TAU’, ‘EM’, ‘J’ and ‘XE’ indicate the type of L1 object, tau, electron, jet or missing energy, respectively. The digit before these names indicates the object multiplicity, while the digits after correspond to the ET requirement, e.g. a L1 ET > 30 GeV for L1 TAU30. ‘IM’ and ‘HI’ indicate that isolation requirements are applied. The data have been collected at a center-of-mass energy of 13 TeV in 2015, from the 6th to the 12th of July.	[png][pdf][eps]
Level 1 rates before prescale versus the instantaneous luminosity measured by ATLAS for L1 di-tau chains. ‘TAU’ and ‘J’ indicate the type of L1 object, tau or jet, respectively. The digit before these names indicates the object multiplicity, while the digits after correspond to the ET requirement, e.g. a L1 ET > 20 GeV for L1 TAU20. ‘IM’ indicates that isolation requirements are applied. The data have been collected at a center-of-mass energy of 13 TeV in 2015, from the 6th to the 12th of July.	[png][pdf][eps]
Comparison of the ATLAS HLT expected single-tau trigger efficiency in Run-I and Run-II simulations. The efficiency is computed for offline reconstructed tau candidates with transverse momentum above 20 GeV, one or three tracks and passing the offline medium identification criteria [Eur.Phys.J.C75(2015)303] in simulated Z→ ττ events, where one tau lepton decays leptonically and the other hadronically. The correspondent online tau candidate is required to have a transverse momentum of at least 35 GeV, between one and three tracks and passing the online medium identification. The efficiency is plotted as function of the transverse momentum of the offline tau candidate. Error bars are statistical uncertainties. The overall improvement in the trigger efficiency expected in Run-II is due to the new high-level trigger strategy which includes a more precise energy calibration, a faster tracking and an online identification requirement closer to the one used in the offline tau reconstruction.	[png][pdf][eps]

Plots for EPS 2015 (Obsolete) ATL-COM-DAQ-2015-097

| Transverse momentum and pseudo-rapidity distributions of online tau candidates passing the HLT tau trigger with transverse momentum threshold at 35 GeV and online medium identification requirement. These tau candidates are observed in W→μν events and are likely to be jets originated from quarks or gluons that are reconstructed as tau jets and accepted by the tau trigger. Such fake tau candidates represent the dominant source of background in the measurement of the performance of the tau trigger for real tau leptons in Z→𝜏𝜏 events. Data has been recorded in the first 13 TeV collisions in 2015. Events have been select