CMS-DP-2012/007
ECAL Detector Performance Plots: 2011 data

Abstract: Detector Performance Plots of the CMS Electromagnetic calorimeter, based on the 2011 pp data sample. Plots include: detector stability; timing and alignment calibration; response monitoring and corrections; energy calibration and resolution.

CDS entry

iCMS entry

Figure	Caption
pdf version pdf version pdf version	Temperature stability of the ECAL Barrel and Endcap detectors during the period April 2011 - October 2011 The temperature measurements are taken from 6114 calibrated thermistors in the Barrel (170 per SM) and 569 in the Endcap (one per supercrystal) which are measured with local ECAL Detector Control Units (DCUs). The RMS deviation of the temperature measurements is shown for each thermistor, for the time period specified in the plot and for periods with stable beams in the LHC. The temperature stability during 2011 is well within specifications (<0.05 °C for EB, <0.1 °C for EE)
pdf version pdf version pdf version	Trigger efficiency as a function of transverse energy for an electron/photon trigger path with a threshold of 15 GeV for the EB and EE The first plot shows the trigger efficiency for 2011 data, measured using a Tag and Probe method on Z→ee events. The table shows the transverse energy for which the trigger is 50%, 95% and 99% efficient in the EB and EE, and the efficiency at 100 GeV which corresponds roughly to the plateau efficiency The second plot shows the use of 2011 data to emulate the effect of response corrections that will be applied at the trigger level for 2012 data taking.
pdf version pdf version pdf version pdf version	Alignment Relative alignment of the ECAL crystals and the CMS tracker measured using electrons from Z→ee and (chiefly) W→eν events. The plots show the residual difference in η and φ between the position of the ECAL supercluster and the extrapolated track position, using the point of closest approach to the supercluster. The distributions of Δη and Δphi are shown for data before and after the ECAL alignment procedure has been carried out. The distributions for perfectly aligned Monte Carlo events are also shown. Technical details: Uses electrons from Z→ee and W→eν (4.98 fb-1). Selection cuts detailed in [1]. Uses “Golden” (low bremstrahlung) electron selection Alignment procedure involves translational offsets (x,y,z) • Conclusion: - The width of the Δη and Δφ distributions, after ECAL-tracker alignment, are 2.8 (5)x10−3 rad in Δφ in EB (EE) and 1(2)x10−3 units in Δη in EB (EE) . This meets the ECAL alignment goals for electron ID and di-photon resonance reconstruction, which are 20x10−3 rad in Δφ and 4x10−3 units in Δη. [1] CMS-DP-2010/008
pdf version pdf version pdf version pdf version	ECAL time synchronisation and resolution measured from Z→ee events in 2011 data The top two plots show the reconstructed times of the electron seed crystals; for electron pairs contained within the barrel and the endcaps respectively. The bottom two plots show the time difference between the seed crystals for the two electrons after removing the difference in time of flight between the two electrons due to the extension of the luminous region. The synchronization of the readout (0.38/0.36 ns in EB/EE) was stable in 2011 and ensured that there was no impact on the amplitude reconstruction with weights. The time resolution for a single ECAL crystal, for the energy range of electrons from Z decays, is found to be 0.19/0.28 ns in EB/EE
pdf version	Relative response to laser light (440 nm) measured by the ECAL laser monitoring system, averaged over all crystals in bins of pseudorapidity, for the 2011 data taking period The response change observed in the ECAL channels is consistent with expectations for the luminosities of 2011 data taking. The response change is of the order of a few percent in the barrel, while it reaches up to 15% in the most forward endcap regions used for electron and photon reconstruction. The response change is up to 40% in channels closest to the beam pipe. The plot shows all the measurements taken during 2011 data taking. These measurements are used to correct the physics data.
pdf version png version pdf version png version pdf version png version pdf version png version	The precision of channel inter-calibration, using energy deposits, as a function of pseudo-rapidity in the ECAL barrel and endcap detectors Top plots: the precision for measuring the inter-calibration constants from phi-symmetry,from π0→γγ and η→γγ decays, and from high energy isolated electrons (from W→enu decays), as a function of η in EB and EE, using 2011 data Bottom plots: the precision of the combination (weighted average) of the in-situ calibrations, as a function of η in EB and EE. nb: inter-calibration constants derived before LHC startup (test beam, cosmics, beam splash,and lab measurements) are used in the 2010 combination while, at present, the 2011 inter-calibration constants are derived from only in-situ methods. Inter-calibration precision at low eta in EB is ~0.5% and is better than 1% in all eta rings. EE inter-calibration precision is ~2% in the central part of EE and better then 4 % even up to the limit of electron and photon acceptance at η = 2.5
pdf version pdf version	History plot for 2011 data of the ratio of electron energy E, measured in the ECAL, to the electron momentum p, measured in the tracker The electrons are selected from W->enu decays. Each point in the plot is computed from 12000 selected W->enu events with the reconstructed electron located in the ECAL Barrel (top) and in the ECAL Endcaps (bottom). The E/p distribution for each point is fitted to a template E/p distribution measured from data (using the entire 2011 dataset) in order to provide a relative scale for the E/p measurement versus time. The history plots are shown before (red points) and after (green points) corrections to ECAL crystal response are applied. The magnitude of the average transparency correction for each point (averaged over all crystals in the reconstructed electromagnetic clusters) is indicated by the continuous blue line. A stable energy scale is achieved throughout the 2011 run after applying transparency corrections to the ECAL data. The average signal loss of 2.5% in the ECAL Barrel is corrected with an RMS stability of 0.12%. The average signal loss of 10% in the ECAL Endcaps is corrected with an RMS stability of 0.45%
pdf version pdf version pdf version	Stability of the energy scale as measured from the invariant di-photon mass from η→γγ decays as a function of time, before (red dots) and after (green dots) response corrections are applied. The energy scale is measured by fitting the invariant mass distribution of photon pairs in the mass range of the η meson. Each point is from a fit to approximately 20 minutes of data taking. The 3 plots show the full year history, a detail of the month of September and a 24-hour history. The energy scale is found to be stable to within 0.18% (0.1%) in the barrel over a time span of 6 (1) months after the corrections for the channel response changes are applied.
pdf version pdf version	Stability of the energy scale using the energy-momentum balance in W→enu decays as a function of the magnitude of the response change before (red dots) and after (green dots) laser monitoring (LM) corrections are applied. The relative scale is measured by comparing the energy reconstructed in the ECAL with the track momentum estimate from the tracker. Each point represents the mean value from 12000 events. The different shades correspond to different eta regions of the ECAL barrel, from module 1 (lightest shade) to module 4 (darkest shade). In the Endcaps, the different shades refer to four separate rings in eta. The response change measured by the laser monitoring system (R/R0) is scaled by a factor alpha to normalise to the response for electromagnetic showers (S/S0) according to the formula: S/S0=(R/R0)^alpha. In the barrel, a constant value of alpha is assumed. In the endcaps, an effective scaling of alpha has been introduced to optimize the resolution. The effective scaling is compatible with the change in alpha expected for large response changes. Good stability has been obtained for the E/p response, using the LM corrections and the particular values of α used in this analysis
pdf version pdf version	The mass resolution of the Z peak, reconstructed from its di-electron decay mode, as a function of time for the barrel and the endcaps. The width of the Z peak is fitted with a convolution of a Crystal Ball with a Breit-Wigner line shape. The gaussian width parameter of the crystal ball function is taken as a measure of the mass resolution. Towards the end of 2011 data taking the resolution worsens slightly in the barrel and more substantially in the endcaps before the laser monitoring (LM) corrections are applied, due to the increased LHC luminosity and increased response changes. After the LM corrections, a residual worsening of resolution remains in the endcaps, which gives a measure of the degradation which could be due to the need for further tuning of the laser corrections and pile-up effects. The relative mass resolution is stable to within 4% in both barrel and endcaps after the transparency corrections have been applied.
pdf version	Relative energy scale as measured from the energy-momentum balance in W→enu decays, as a function of pseudo rapidity, eta Each point is obtained by fitting the E/p distribution of electrons at a given eta to a reference E/p distribution obtained from the MC simulation. Because the E/p shape varies along eta, nine reference distributions are used. These regions (marked by the vertical dashed lines) are symmetric about η = 0 and correspond to the four modules in each ECAL half barrel, and to five additional regions in each endcap. Both MC and data E/p distributions are fitted to MC templates in order to study relative differences between data and MC. The momentum p is calibrated along eta using Z->ee events. The grey shaded regions between the barrel and endcaps are generally excluded from the acceptance in physics analyses using electrons and photons. The measured scale shows a larger variation as a function of eta than that observed for the MC. The scale in DATA is adjusted to that measured in the MC as an effective correction.
pdf version pdf version	Distribution of the ratio of the energy of the electron shower E measured in the ECAL and its associated track momentum p measured in the tracker for electrons from W→enu decays in the Barrel and Endcaps for 2011 data-taking. The selection requires exactly one electron, with ET > 30 GeV, which satisfies loose electron identification and isolation criteria. The combined isolation (the sum of track isolation, ECAL isolation and hadronic isolation in a cone of ∆R =0.03) relative to the electron pT is required to be lower than 0.04. The isolation is corrected on an event-by-event basis for contamination due to pile-up. The purity of the sample is >99%. The E/p distribution provides a measurement of the stability of ECAL at the level of ~0.1% with ~10 /pb in EB and of ~0.4% with ~10 /pb in EE
pdf version pdf version pdf version pdf version	Effect of single channel inter-calibration, transparency corrections and cluster corrections on the Z->ee invariant mass. The first two plots show the impact on the Z→ee energy scale and resolution that are obtained from applying energy scale corrections to account for the intrinsic spread in crystal and photo-detector response, and time-dependent corrections to compensate for channel response loss. The second two plots show the impact on the Z→ee energy scale and resolution from the incorporation of more sophisticated clustering and cluster correction algorithms.
pdf version pdf version pdf version pdf version pdf version pdf version	Invariant mass distribution of Z→ee events Distribution of Z→ee events for events with a) both electrons in EB and having R9>0.94, b) both electrons in EB and c) both electrons in EE. Data and Monte Carlo distributions are shown. Parameters listed are Δm(CB) - the difference (in GeV) between the Crystal Ball mean and the true Z mass, σ(CB) - the width of the gaussian term of the Crystal Ball function. The parameter R9 is a measure of the extent of electron bremsstrahlung. An unbinned likelihood fit is performed to the convolution of a Breit-Wigner (BW) and Crystal-Ball (CB) to extract the Z peak shape parameters. The BW mean and width are fixed to the PDG) values. The Crystal Ball cut-off parameter and tail parameter (α and n) are constrained to the values fitted on the MC distribution. The width of the gaussian term of the CB function in the barrel is 1.01 GeV and 1.56 GeV for high and low R9 and 2.57 GeV for the endcaps. The resolution in DATA is slightly worse than the MC in the barrel. In the endcaps, the DATA distribution is wider than in MC. The energy scale offset is consistent, between the fitted mass and the Z mass value from the PDG, within the errors quoted on the energy scale in CERN-CMS-DP-2011-008, of 0.6% and 1.5% for EB and EE respectively.
pdf version pdf version pdf version pdf version pdf version pdf version	Photon energy scale and resolution using Z->mumugamma events Invariant mass distribution of Z->mumugamma final states from 2011 DATA. The photon energy scale and resolution are extracted from de-convoluting the Z line shape in this final state. Plots are shown for DATA and MC for EB for R9>0.94, EB inclusive and EE inclusive categories. Z decays where the final state leptons radiate a photon, provide a source of clean photons, where the photon kinematics are constrained by the muon system. The photon ET is reconstructed from the muon kinematics and the Z mass constraint. The parameters for energy scale mu(ET) and resolution sigma(ET) are computed performing an unbinned maximum likelihood fit for M(mumugamma) using the sum of the signal and background models. Statistical uncertainties only are quoted in the plots. Systematic effects from the muon kinematics are not included. The photon energy scale agrees to within 0.7% between DATA and MC. The energy resolution for photons is 2.2% in the barrel (1.6% for high R9) and 4.8% in the endcaps. The energy scale and resolution are in agreement with the values measured for electrons from Z->ee decays.
pdf version png version pdf version png version pdf version png version pdf version png version pdf version png version pdf version png version	Z->ee resolution unfolding Relative electron energy resolution unfolded in bins of pseudorapidity eta for the barrel and in the endcap. Electrons from Z->ee decays are used. The resolution is shown separately for electrons having R9>0.94, R9<0.94 and for the inclusive sample. The resolution sigma(E) is extracted from an unbinned likelihood fit to Z->ee events, using a Voigtian (Landau convoluted with Gaussian) as the signal model. The resolution is plotted separately for data and MC events. Good agreement is obtained in the resolution extracted from the unfolding procedure and the fits to the Z invariant mass distribution, assuming a scaling of the mass resolution with sqrt(2) to obtain the equivalent energy resolution. The resolution in the barrel scales with the amount of material in front of the tracker, and is degraded in the vicinity of the ECAL supermodule boundaries, indicated by vertical lines in the plots. The resolution in the endcaps shows an eta dependence that is also correlated with the amount of material in front of the tracker