AtlasPublicTopicHeader.png

JetEtmissApproved2013Pileup1

Distributions of the LCW-scale per-event median pT density ρ (based on topological clusters with |η| < 2) for four representative values of the reconstructed primary vertex multiplicity NPV. Events with the same number of vertices may have widely varying levels of pile-up activity in the calorimeter, as measured by ρ. The use of ρ instead of NPV in the pile-up correction therefore results in significantly reduced fluctuations, or improved energy resolution.
[eps]
RMS width of the distribution of pTreco − pTtrue for anti-kt R=0.6 jets matched to truth jets (20 < pTtrue < 30 GeV), with no jet energy scale correction, before and after two pile-up subtraction methods. The RMS is presented as a function of pseudorapidity for 25 ≤ <μ> < 35. Subtraction of ρ⋅Area results in more than 10% improvement in the RMS, as compared to the average pile-up subtraction parameterized by <μ> and NPV. However, a large fraction of the resolution degradation due to pile-up remains.
[eps]
RMS width of the distribution of pTreco − pTtrue for anti-kt R=0.6 jets matched to truth jets (20 < pTtrue < 30 GeV), with no jet energy scale correction, before and after two pile-up subtraction methods. The RMS is presented as a function of <μ> for |η| < 2.4. Subtraction of ρ⋅Area results in more than 10% improvement in the RMS, as compared to the average pile-up subtraction parameterized by <μ> and NPV. However, a large fraction of the resolution degradation due to pile-up remains.
[eps]
RMS width of the distribution of pTreco pTtrue for anti-kt R = 0.6 jets matched to truth jets (20 < pTtrue < 30 GeV), with no jet energy scale correction, before and after two pile-up subtraction methods. The RMS is presented as a function of pseudorapidity for 18 ≤ NPV < 23 (left) and as a function of NPV for |η| < 2.4 (right). While the average pile-up subtraction (parametrized by NPV and <μ>) does not improve the RMS in narrow bins of NPV, subtraction of ρ⋅Area does. However, a large fraction of the resolution degradation due to pile-up remains.
[eps]
RMS width of the distribution of pTreco pTtrue for anti-kt R = 0.6 jets matched to truth jets (20 < pTtrue < 30 GeV), with no jet energy scale correction, before and after two pile-up subtraction methods. The RMS is presented as a function of pseudorapidity for 18 ≤ NPV < 23 (left) and as a function of NPV for |η| < 2.4 (right). While the average pile-up subtraction (parametrized by NPV and <μ>) does not improve the RMS in narrow bins of NPV, subtraction of ρ⋅Area does. However, a large fraction of the resolution degradation due to pile-up remains.
[eps]
Dependence of the jet pT on in-time pile-up, characterized by NPV for fixed <μ>, for anti-kt R=0.4 LCW jets as a function of |η|. The dependence of in-time pile-up is much reduced by subtraction of ρ⋅Area, but the dependence on out-of-time pile-up is largely unaffected. After the application of a residual correction dependent on NPV and <μ>, the pile-up dependence of the jet pT is consistent with zero. The shaded bands show the 68% confidence interval of a piecewise linear fit which captures the |η| dependence observed in the sensitivity of jets to pile-up.
[eps]
Dependence of the jet pT on out-of-time pile-up, characterized by <μ> for fixed NPV, for anti-kt R=0.4 LCW jets as a function of |η|. The dependence of in-time pile-up is much reduced by subtraction of ρ⋅Area, but the dependence on out-of-time pile-up is largely unaffected. After the application of a residual correction dependent on NPV and <μ>, the pile-up dependence of the jet pT is consistent with zero. The shaded bands show the 68% confidence interval of a piecewise linear fit which captures the |η| dependence observed in the sensitivity of jets to pile-up.
[eps]
Mean number of pile-up jets (anti-kt R=0.4) per unit pseudorapidity, in central (|η| < 2.1), intermediate (2.1 < |η| < 3.2), and forward (3.2 < |η| < 4.5) regions, as a function of <μ>. Jets were required to have pT > 20 GeV after the full calibration including ρ⋅Area subtraction. The central region contains the highest density of pile-up jets, with 3 to 4 times as many as in the forward region.
[eps]
Mean jet multiplicity in Z → μμ events for anti-kt R=0.4 LCW jets with pT > 20 GeV before the JES correction and |η| < 2.1, as a function of the average number of interactions per bunch crossing <μ>. Before pile-up subtraction, the jet multiplicity is overestimated in simulation, due to mis-modeling of pile-up jets. Pile-up subtraction shifts most of the pile-up jets below the 20 GeV pT threshold, which results in improved modeling of the jet multiplicity. The remaining pile-up jets may be rejected by using a discriminant such as the jet vertex fraction (JVF).
[eps]
Mean jet multiplicity in Z→μμ events for anti-kt R=0.4 LCW+JES jets with pT > 20 GeV and |η| ≤ 2.1, as a function of the average number of interactions per bunch crossing <μ>. With no cut on the jet vertex fraction (JVF) applied, the jet multiplicity is overestimated in simulation, due to mis-modeling of pile-up jets. By rejecting pile-up jets with a cut of |JVF| > 0.25 or |JVF| > 0.5, this systematic bias is effectively removed.
[eps]
Mean jet multiplicity in Z→μμ events for anti-kt R=0.4 LCW+JES jets with pT > 20 GeV and |η| ≤ 2.1, as a function of the average number of interactions per bunch crossing <μ>. With no cut on the jet vertex fraction (JVF) applied, the jet multiplicity is overestimated in simulation, due to mis-modeling of pile-up jets. By rejecting pile-up jets with a cut of |JVF| > 0.25 or |JVF| > 0.5, this systematic bias is effectively removed.
[eps]
Mean jet multiplicity in Z → μμ events for anti-kt R=0.4 LCW+JES jets with pT > 50 GeV and |η| < 2.1, as a function of the average number of interactions per bunch crossing <μ>. There is no discernable slope in either data or simulation, which illustrates that a JVF cut is not necessary for jets with pT > 50 GeV.
[eps]
Mean jet multiplicity in Z → μμ events for anti-kt R=0.4 LCW+JES jets with pT > 50 GeV and |η| < 2.1, as a function of the average number of interactions per bunch crossing <μ>. JVF cuts at 0.25 and 0.5 are observed to have little effect, further illustrating that pile-up jets are negligible for pT > 50 GeV.
[eps]


Major updates:
-- MichaelBegel - 07-Mar-2013

Responsible: MichaelBegel
Subject: public

Edit | Attach | Watch | Print version | History: r4 < r3 < r2 < r1 | Backlinks | Raw View | WYSIWYG | More topic actions
Topic revision: r4 - 2013-04-09 - MichaelBegel
 
    • Cern Search Icon Cern Search
    • TWiki Search Icon TWiki Search
    • Google Search Icon Google Search

    Atlas All webs login

This site is powered by the TWiki collaboration platform Powered by PerlCopyright & 2008-2019 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding TWiki? Send feedback