Electrons and photons for Higgs measurements with the
ATLAS detector
Christophe Goudet
Journées de Rencontre des Jeunes Chercheurs
Chédigny, November 17, 2015
Goudet (LAL)
Electrons & photons for Higgs measurements
JRJC, November 17, 2015
1 / 23
Content
1
Electrons & photons reconstruction
2
Electrons & photons calibration
3
Higgs boson in electromagnetic channels
Goudet (LAL)
Electrons & photons for Higgs measurements
JRJC, November 17, 2015
2 / 23
∑ weights / GeV
Introduction
∫ L dt = 4.5 fb , s = 7 TeV
-1
∫ L dt = 20.3 fb , s = 8 TeV
180
-1
160
ATLAS
Data
S/B weighted sum
Signal+background
Signal strength categories
140
Background
120
Goudet (LAL)
Electrons & photons for Higgs measurements
100
80
60
40
∑ weights - fitted bkg
20
0
10
5
0
-5
110
120
130
140
150
160
m γ γ [GeV]
Events / 2.5 GeV
Higgs discovered with electrons and
photons (and muons) .
mH = 125.09 ± 0.21(stat) ± 0.11(syst) GeV
(ATLAS+CMS) PhysRevLett.114.191803
Run 2 started in june 2015 at an increased
center of mass energy of 13 TeV.
Over the next three years, about 30 times
more Higgses are expected.
With reduced statistical uncertainties
→ need to reduce systematic
uncertainties.
Calibration is a important source of systematic.
Needs to be improved in Run 2.
Signal
m H = 125.4 GeV
35
30
25
ATLAS
Data
H → ZZ* → 4l
Signal (m = 125 GeV µ = 1.51)
H
∫ Ldt = 4.5 fb
s = 8 TeV ∫ Ldt = 20.3 fb
s = 7 TeV
-1
-1
Background ZZ*
Background Z+jets, t t
Systematic uncertainty
20
15
10
5
0
80 90 100 110 120 130 140 150 160 170
m4l [GeV]
JRJC, November 17, 2015
3 / 23
ATLAS experiment
Large acceptance
Radiation hard
Table 1.1: General performance goals of the ATLAS detector. Note that, for high-pT muons,
the muon-spectrometer performance is independent of the inner-detector system. The units for E
and pT are in GeV.
Performance goals of the ATLAS detector
Detector component
σ pT /pT = 0.05% pT ⊕1%
√
σE /E = 10%/ E ⊕ 0.7%
√
σE /E = 50%/ E ⊕ 3%
√
σE /E = 100%/ E ⊕ 10%
σ pT /pT =10% at pT = 1 TeV
η coverage
Measurement
Trigger
±2.5
±3.2
±2.5
±3.2
3.1 < |η| < 4.9
±2.7
±3.2
3.1 < |η| < 4.9
±2.4
The muon instrumentation includes, as a key component, trigger chambers with timing resolution
of the order of 1.5-4 ns. The muon spectrometer defines the overall dimensions of the ATLAS
detector.
The proton-proton interaction rate at the design luminosity of 1034 cm−2 s−1 is approximately
Goudet (LAL)
2008 JINS
Tracking
EM calorimetry
Hadronic calorimetry (jets)
barrel and end-cap
forward
Muon spectrometer
Required resolution
Silicon and TRT tracker in 2T
magnetic field
Measure position and momentum of
charged particles
Liquid argon electromagnetic
calorimeter (LAr)
Measure energy of electrons and
photons.
Scintillating tiles hadronic
calorimeter
Measure energy of jets
Muon chambers
Electrons & photons for Higgs measurements
JRJC, November 17, 2015
4 / 23
phénomène continue en cascade, donnant lieu à la création d’une gerbe
tique. Les caractéristiques de la gerbe (extensions latérales, longitudinales,
ent l’identification des particules avec une certaine efficacité/réjection.
Electromagnetic calorimeter (LAr)
1.4m < r < 2m
Sampling calorimeter :
- absorber : lead
- active material : Liquid Argon (88K)
Accordion geometry gives uniformity and
hermeticity along φ.
Longitudinally segmented for pion
– Schéma de principe du calorimètre à argon liquide pour un absorbeur de plomb.
discrimination
iquide est un gaz rare, inerte chimiquement par les couches électroniques
Cells in Layer 3
∆ϕ×∆η = 0.0245×0.05
Trigge
r Towe
∆η = 0 r
.1
2X0
47
0m
m
s complètes,
η = 0 ce qui l’empêche de capturer des électrons. La phase liquide
orte densité, donc 16X
de0 couvrir la gerbe électromagnétique avec une meilleure
hantillonnage et un bon rapport S/B. En conclusion, TTilriggaer les propriétés d’une
ow
∆ϕ = 0er
.0982
m
m
arfois génériquement
électrode par abus de langage dans la littérature.
4.3X0
15
00
1.7X0
∆ϕ=0.0
245x
36.8m 4
mx
=147.3 4
mm
ϕ
Square cells in
Layer 2
∆ϕ = 0
.0245
∆η = 0
.025
m/8 =
4
∆η = 0 .69 mm
.0031
Strip cells in Layer 1
37.5m
η
Goudet (LAL)
Electrons & photons for Higgs measurements
JRJC, November 17, 2015
5 / 23
Energy measurement in LAr
Signal drift time (∼ 600ns) too long for collisions
every 25ns (pile-up).
Analog signal pass through an bipolar filter to reduce
signal time. Shape optimize signal over pileup and
electronic noise.
APITRE 2. Le LHC et le détecteur
ATLAS
ADC sampling
every 25ns (4 points are kept).
Energy computed using calibration constants and
filtering
of : the samples.
L’énergie reconstruite dansoptimal
une cellule
s’écrit
n
samples
Ecell =
X
| i=1
ai (si − ped) ·GADC→DAC ·
{z
ADC
}
Mphys
Mcalib
−1
· FDAC→µA · FµA→M eV
faisant apparaı̂tre :
– l’énergie mesurée par la technique de filtrage optimal à partir du signal exp
Goudet (LAL)
Electrons & photons for Higgs measurements
JRJC, November 17, 2015
6 / 23
coups ADC.
Reconstruction & Identification
Reconstruction links the energy deposit in detector cells to a
physical particle and its properties.
Divide the central part (|η| = |ln(tan(θ/2))| < 2.47) into towers of size
∆η × ∆φ = 0.25 × 0.25
Sum energies from all cells and all layers of the tower
Sliding window (3 × 5 towers ) algorithm look for 2.5 GeV of transverse energy
Track matching and clustering :
no track → photon → 3 × 7 cluster
I track → electron → 3 × 7 cluster
I conversion vertex → converted photon → 3 × 7 cluster
I
Identification is to separate prompt electrons from both jets and other electrons
from either hadron decay or photon conversion.
A multivariate likelihood method using 23 variables
of energy deposit and tracking is used.
Goudet (LAL)
Electrons & photons for Higgs measurements
JRJC, November 17, 2015
7 / 23
Full calibration
To reach the physics analyses, data and simulated reconstructed events must pass a
calibration procedure. This procedures aim to correct the measured energy to
retrieve the true energy of the particle at the interaction point.
simulation!
1! training of !
5!
3!
Zàee #
resolution
smearing !
MC-based !
e/γ calibration!
EM !
cluster!
energy!
calibrated !
e/γ !
energy!
MC-based !
e/γ energy!
calibration!
data!
2!
longitudinal
layer intercalibration!
4!
uniformity
corrections!
5!
Zàee #
scale
calibration!
6!
J/ψàee Zàllγ#
data-driven scale validation!
Electrons and photons follow the same steps but with dedicated analyses.
Goudet (LAL)
Electrons & photons for Higgs measurements
JRJC, November 17, 2015
8 / 23
MVA calibration
MPV
Simulated events are passed through a full GEANT4 simulation of the ATLAS
detector.
Events are then categorized in η and pT bins, separately for electrons and
photons.
A multivariate analysis (MVA) is performed to compute the true
energy from detector observables.
Plot shows most probable value (MVP) of E corr /E true .
1.02
1.015
MVA uses :
Energies in all layers of the ECAL
EM shower shape variables
Barycenters of energy deposits
1.01
ATLAS Simulation
Electrons
25 GeV
50 GeV
100 GeV
200 GeV
1000 GeV
1.005
1
0.995
0.99
0
0.5
1
1.5
2
2.5
|η|
Goudet (LAL)
Electrons & photons for Higgs measurements
JRJC, November 17, 2015
9 / 23
# Events / 0.20
Energy scale factors
Data
MC
0.025
After MVA calibration, mass distribution
of Z → ee for data and MC still have
discrepancy.
A data-driven analysis is performed
to match data to MC distribution
(relative matching).
8TeV
0.02
0.015
0.01
0.005
0
80
82
84
86
88
90
92
94
96
98
100
m12
A correction, applied to both electrons of Z decay, is computed to shift the central
value of data distribution :
energy scale factor (α)
E corr = E meas (1 + α)
Goudet (LAL)
Electrons & photons for Higgs measurements
JRJC, November 17, 2015
10 / 23
Resolution constant term
b
σ
a
= √ ⊕ ⊕c
E
E E
a : sampling term ( 10%). Linked to the fluctuations of electromagnetic showers.
Can be simulated.
b/E : noise term ( 350cosh(η) MeV ). Measured in dedicated runs.
c : constant term (0.7%). Must be measured on data.
We observe that data distribution is larger than MC. An additional constant
term (C) is measure to enlarge MC up to the data width. Both MC electrons
undergo the correction :
Resolution constant term (C)
E corr = E meas (1 + N(0, 1) ∗ C )
N(0, 1) : a Gaussian distributed random number
Goudet (LAL)
Electrons & photons for Higgs measurements
JRJC, November 17, 2015
11 / 23
Template method
2
(αmin , χ2min )
χ2
χ2
140
120
0.12
Template : alpha=-11089
Template : alpha=-8758
0.1
0.08
0.06
0.04
0.02
0
80
82
84
86
88
90
92
94
96
98
100
M ee
0.014
0.012
0.01
0.008
0.006
α
fit χ = f (α) at constant C (lines) →
.
2
I fit χ
min = f (C ) → (C , ∆C )
I project C in αmin = f (C ), corresponding bin gives
(α, ∆α).
I
Data
σ
The template method is used to measure α and C
simultaneously.
Create distorded MC (templates) with test
values of α and C .
Compute χ2 between Z mass
distribution of data and template.
Fit the minimum of the χ2 distribution in
the (α, C ) plane.
Fit performed in 2 steps of 1D fits :
0.14
140
0.004
−0.0098
120
−0.0099
100
100
80
80
0.002
−0.01
−0.0101
0
60
60
−0.0102
−0.011
−0.0105
−0.01
−0.0095
40
40
−0.0103
20
−0.011
−0.0105
−0.01
−0.0095
−0.009
0
α
Goudet (LAL)
0.002
0.004
0.006
0.008
0.01
0.012
0.014
0
c
0.002
0.004
0.006
0.008
0.01
0.012
−0.009
α
0.014
c
Electrons & photons for Higgs measurements
JRJC, November 17, 2015
12 / 23
Detector splitting
Detector is not uniform along η.
To improve resolution,
calibration is performed in bin of
ηcalo .
68 and 24 bins are used respectively
for α and C .
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 1.1 1.2 1.285 1.37 1.42 1.47 1.51 1.55
1.59 1.63 1.6775 1.725 1.7625 1.8 1.9 2 2.05 2.1 2.2 2.3 2.35 2.4 2.435 2.47
Electrons are labelled by their η bin, hence Z are labeled by the combination
of electrons bins. Scales are computed for each combination.
Goudet (LAL)
Electrons & photons for Higgs measurements
JRJC, November 17, 2015
13 / 23
Inversion Procedure
Obtaining electron scales from Z scales need the minimizations of the following
χ2’s
2
P
(α
+α
−2α
)
i
j
ij
χ2 =
2
α
P (
2
χ =
i,j≤i
−0.01
(∆αij )
(1)
ci2+cj2
2
−c
)
ij
2
∆2cij
C
i,j≤i r
0.018
0.016
−0.015
0.014
0.012
−0.02
0.01
0.008
−0.025
0.006
0.004
−0.03
−2
−1.5
−1
−0.5
0
0.5
1
1.5
2
−2
−1.5
ETA_TRK
Goudet (LAL)
Electrons & photons for Higgs measurements
−1
−0.5
0
0.5
1
1.5
2
ETA_TRK
JRJC, November 17, 2015
14 / 23
Goudet (LAL)
Electrons & photons for Higgs measurements
0.04
-3
δ α (10 )
Effective constant term, c
∫
s=8 TeV, Ldt = 20.3 fb-1
ATLAS
0.02
0
-0.02
-0.04
4
3.5
3
2.5
2
1.5
1
0.5
0
Z resonance (total uncertainty)
Total
Stat.
-2
-1.5
-1
-0.5
0
0.5
0.06
0.05
1
1.5
2
η
∫
s=8 TeV, Ldt = 20.3 fb-1
ATLAS
Z resonance
Azimuthal non-uniformity
0.04
0.03
0.02
0.01
0
-3
δ c (10 )
Uncertainties are evaluated as the difference between
official scales and the ones measured with a changed
parameter. They include :
electron identification quality from medium to tight.
Z mass window
electron pT cut
uncertainties on efficiencies scale factors
energy loss through bremshtrahlung
background
pile-up
measurement method
Energy correction, α
Run 1 : results and uncertainties
12
10
8
6
4
2
0
Total
Stat.
-2
-1.5
-1
-0.5
0
0.5
1
1.5
JRJC, November 17, 2015
2
η
15 / 23
Run 2 pre-recommandations
Run 2 early analyses need scales factors for 13TeV but not enough data will be
available. Need to estimate run 2 scales from run 1 data.
Pre-recommandations are computed using 8 TeV data reprocessed with :
new detector geometry
new reconstruction algorithm
new calibration machine learning
0.035
2012
2012
2015 pre-recommandations
2015 pre-recommandations
0.06
0.03
0.04
0.025
0.02
0.02
0
0.015
−0.02
0.01
−0.04
0.005
−2
−1.5
−1
Goudet (LAL)
−0.5
0
0.5
1
1.5
2
0
−2
−1.5
Electrons & photons for Higgs measurements
−1
−0.5
0
0.5
1
1.5
JRJC, November 17, 2015
2
16 / 23
Run 2 pre-recommandations systematics
2012 systematics are used for the pre-recommandations.
Two more systematics are added in quadrature :
Increasing the number of bin for α shows sub-patterns. Systematic is defined as
difference between a bin value and the average of its sub-bins.
Pre-recommandations being computed with 8TeV datasets, one needs to evaluate
the impact of the center of mass energy. Systematic is defined as the scale
measured from 13 TeV MC on 8TeV templates.
C
0.045
α
0.045
Pre-recommandations, syst only
Pre-recommandations, syst only
8/13TeV systematic
0.04
0.04
8/13TeV systematic
68/34 bins systematic
0.035
0.035
0.03
0.03
0.025
0.025
0.02
0.02
0.015
0.015
0.01
0.01
0.005
0.005
0
−2
−1.5
−1
−0.5
0
0.5
1
1.5
2
0
η
−2
−1.5
−1
−0.5
0
0.5
1
1.5
calo
Goudet (LAL)
Electrons & photons for Higgs measurements
2
η
calo
JRJC, November 17, 2015
17 / 23
Run 2 results
α
C
Scales are measured with 13TeV data at 25ns :
Data are corrected with energy scales from pre-recommandations
MC is not smeared with pre-rec
ATLAS Work in progress
0.06
s = 13 TeV
Pre-recommandations, syst only
∫ Ldt = 2.220 fb
0.05
24Bins, with prerec
-1
ATLAS Work in progress
s = 13 TeV
0.04
Pre-recommandations, syst only
∫ Ldt = 2.220 fb
Pre-recommandations, stat only
-1
24Bins, with prerec
0.04
0.02
0.03
0
0.02
− 0.02
0.01
− 0.04
−2
−1.5
−1
−0.5
0
0.5
1
1.5
0
2
−2
−1.5
−1
−0.5
0
0.5
1
1.5
η
2
η
calo
calo
α discrepancies are below 0.1% out of the crack (1.37 < |η| < 1.55).
Goudet (LAL)
Electrons & photons for Higgs measurements
JRJC, November 17, 2015
18 / 23
Higgs boson predicted in 1964, discovered in
2012.
Gives mass to weak boson, and fermions
through Yukawa coupling.
Several production mode are available at
the LHC.
ggH : gg → H
I VBF : qq → Hjj
I VH : Z (W ) → Z (W )H
I ttH : t t̄ → t t̄H
σ(pp → H+X) [pb]
102
pp
→
s= 8 TeV
H(
NN
LO
+N
NL
LQ
CD
10
+N
LO
LHC HIGGS XS WG 2012
Higg boson at the LHC
EW
)
pp →
1
10-1
qqH
(NN
LO
QC
D+
NLO
ZH WH
EW
(N (N
)
NL NL
O
pp
O
Q
Q
→
CD CD
ttH
+N +
(N
LO
LO NL
O
QC
EW E
D)
) W)
pp
→
pp
→
10-2
I
80 100
200
300
1000
MH [GeV]
At a mass of 125 GeV, many decay modes
available :
H → b b̄ : dominant decay mode ( ∼ 57% ) but high
background in hadronic machines.
I H → 4l : low expected events, almost no background.
I H → γγ : low branching ratio (0.28%) but clean
signature. High but smooth background.
I
Goudet (LAL)
Electrons & photons for Higgs measurements
JRJC, November 17, 2015
19 / 23
Mass measurement
-2lnΛ
Higgs mass is the last unknown parameter of the standard model :
mH = 125.36 ± 0.37(stat) ± 0.18(syst)
Table 4: Principal systematic uncertainties on the combined mass. Each uncertainty is determined from the change in the 68%
when the corresponding nuisance parameter is removed (fixed to its best fit value), and is calculated by subtracting this reduc
the original uncertainty in quadrature.
7
Combined γ γ +4l
H → γγ
H → ZZ* → 4l
without systematics
ATLAS
6
s = 7 TeV ∫ Ldt = 4.5 fb-1
s = 8 TeV ∫ Ldt = 20.3 fb-1
5
2σ
4
3
2
1σ
1
0
123 123.5 124 124.5 125 125.5 126 126.5 127 127.5
Systematic
LAr syst on material before presampler (barrel)
LAr syst on material after presampler (barrel)
LAr cell non-linearity (layer 2)
LAr cell non-linearity (layer 1)
LAr layer calibration (barrel)
Lateral shower shape (conv)
Lateral shower shape (unconv)
Presampler energy scale (barrel)
ID material model (|η| < 1.1)
H → γγ background model (unconv rest low pTt )
Z → ee calibration
Primary vertex effect on mass scale
Muon momentum scale
Remaining systematic uncertainties
Total
Uncertainty on mH [MeV]
70
20
60
30
50
50
40
20
50
40
50
20
10
70
180
mH [GeV]
In order
Value of −2 ln Λ as a function of mH for the individual H → γγ and H→ ZZ ∗ → 4` channels and their combination,
whereto
theassess
signal the compatibility of the mass measurements from the two channels a dedicated t
µγγ and µ4` are allowed to vary independently. The dashed lines show the statistical component of the mass
measurements.
For the
takes
into account
correlations between the two measurements is used, as described in Sec. 6. A value
→ 4` channel, this is indistinguishable from the solid line that includes the systematic uncertainties.
Statistical uncertainties highly dominant.
∆m = 1.47 ± 0.67 (stat) ± 0.28 (syst) GeV
Run 2 will increase sensitivity to systematics.
= 1.47 ± 0.72 GeV
GeV))
H
4
ATLAS
(LAL)
s = Goudet
7 TeV Ldt
= 4.5 fb-1
is derived. From the value of −2 ln Λ at ∆mH = 0, a compatibility of 4.8%, equivalent to 1.98σ, is esti
asymptotic assumption. This probability was cross-checked using Monte Carlo ensemble tests. With
compatibility of 4.9% is obtained, corresponding to 1.97σ.
Combined γ γ +ZZ*
H →Electrons
γγ
& photons
for Higgscross-check,
measurements
JRJC, November
/ 23 sc
As an additional
some of the systematic uncertainties
related17,
to 2015
the photon20energy
TABLE XIV.
ATLAS
V
Main systematic uncertainties σµ
ategories sensitive to individual production modes, and
FIG. 20.
The two-dimensional
value value
of (µV
V
combined
strength
parameterbest-fit
µ. The
combined signalData
strength, for the combination
of the signal
for a Higgs boson
of determined
mass mH = 125.4
GeV
dec
γγ8 TeV data. The vertical hatched band
V and
indicates
group
of
uncertainties
are
by
subtr
H → γγ when fixing both µtH and µbb̄H to 1 a
68% confidence Signal+background
interval of the combined signal
strength. from
quadrature
uncertainty
the change
µγγ isdashed
a main
measure.
It is the
relatedingtoallthe
the
cross
section
(production
the total
other signal
strength
parameters.
Thei
Background
vertical
linevariable
at signal to
strength
1 indicates
CL contours
are shown with thenuisance
solid and das
CL range
µ when
the corresponding
p
expectation.
The
the on 95%
probability)
: vertical dashed red line indicates
Signal
The result is obtained for mH = 125.4
t below which the fitted signal plus background
mass
dis- to respectively.
are
fixed
their
best
fitof values.
The8 TeV
experimen
meas
the
combination
the
7
TeV
and
data.
m
=
125.4
GeV
H
ution for the combination
of BR)
the V H categories becomes
(σ ×
+0.12 (theory)
the yield+0.10
does(syst)
not include
the luminosity
= tainty
1.17 ±on
0.23(stat)
µmass
ative for some
γγ =in the fit range.
−0.08
−0.08
0
µ
measurement
(σ × BR)SM
µ
µ
tion, which is accounted
forthe
separately.
Compared with
measured tt̄H signal stre
Total
ttH
Stat.
ZH
Syst.
µ
µ
µ
WH
ATLAS
-1
∫Ldt = 4.5 fb , s = 7 TeV
-1
∫Ldt = 20.3 fb , s = 8 TeV
VBF
ggF
H → γ γ , m H = 125.4 GeV
µ
-1
0
140
1
2
150
3
4
5
6
7
8
160
Signal strength
+0.8
rameter µtt̄H = 1.3 +2.5
−1.7 (stat.) −0.4 (syst.) in R
µtt̄H measured in this analysis profits syst.
from the c
Uncertainty
group
σµsuch as V
tion
of tt̄H events
in other categories
and
V H one-lepton.
this meas
Theory
(yield) In addition, in0.09
the other contributions to the signal strength
Experimental
filed,
whereas they(yield)
are fixed at the0.02
SM predi
Luminosity
0.03
Ref.
[96].
As mentioned
order to
MC
statisticsin the introduction,<in0.01
production
through VBF and associated
product
Theory (migrations)
0.03
a W or Z boson or a tt̄ pair, independently of the
Experimental
(migrations)
0.02
branching
ratio, the
ratios µVBF /µggF
, µV H /µ
0.07 µtH a
µResolution
tt̄H /µggF are fitted separately by fixing
to
1 andscale
profiling the remaining signal
Mass
0.02strengt
measured
ratios shape
Background
0.02
m γ γ [GeV]
+0.8
−0.5 ,
µV H /µggF = 0.6 +1.1
−0.6 ,
2015 ,
µ ¯ JRJC,
/µ November
= 1.217,+2.2
µVBF /µggF = 0.6
If no improvements, calibration uncertainty will be dominant in run 2.
. 19. Measured signal strengths, for a Higgs boson of mass
Goudet (LAL)
Electrons & photons for Higgs measurements
21 / 23
α
Conclusion
ATLAS Work in progress
0.06
s = 13 TeV
∫
Pre-recommandations, syst only
24Bins, with prerec
-1
Ldt = 2.220 fb
0.04
0
− 0.02
− 0.04
−2
−1.5
−1
−0.5
0
0.5
1
1.5
2
η
calo
C
Higgs measurement uncertainties are
dominated by statistics :
new challenges ahead to keep it that
way with 30× more stat.
Calibration procedures are diverse and
complicated.
Calibration uncertainties are among the
dominant ones.
13 TeV data are in good agreement with
expectations.
0.02
0.05
ATLAS Work in progress
s = 13 TeV
∫
Pre-recommandations, syst only
Pre-recommandations, stat only
-1
Ldt = 2.220 fb
24Bins, with prerec
0.04
0.03
0.02
0.01
0
−2
−1.5
−1
−0.5
0
0.5
1
1.5
2
η
calo
Goudet (LAL)
Electrons & photons for Higgs measurements
JRJC, November 17, 2015
22 / 23
Electrons and photons for Higgs measurements with the
ATLAS detector
Christophe Goudet
Journées de Rencontre des Jeunes Chercheurs
Chédigny, November 17, 2015
Goudet (LAL)
Electrons & photons for Higgs measurements
JRJC, November 17, 2015
23 / 23
Identification variables
Table 1: Definition of electron discriminating variables.
Type
Description
Name
Hadronic leakage
Ratio of ET in the first layer of the hadronic calorimeter to ET of the EM cluster
RHad1
(used over the range |η| < 0.8 or |η| > 1.37)
Ratio of ET in the hadronic calorimeter to ET of the EM cluster
Back layer of
(used over the range 0.8 < |η| < 1.37)
Ratio of the energy in the back layer to the total energy in the EM accordion
EM calorimeter
calorimeter
Middle layer of
Lateral shower width,
EM calorimeter
energy and ηi is the pseudorapidity of cell i and the sum is calculated within
q
(ΣEi η2i )/(ΣEi ) − ((ΣEi ηi )/(ΣEi ))2 , where Ei is the
a window of 3 × 5 cells
Ratio of the energy in 3×3 cells over the energy in 3×7 cells centered at the
RHad
f3
Wη2
Rφ
electron cluster position
Strip layer of
EM calorimeter
Ratio of the energy in 3×7 cells over the energy in 7×7 cells centered at the
Rη
electron cluster position
p
Shower width, (ΣEi (i − imax )2 )/(ΣEi ), where i runs over all strips in a window
wstot
of ∆η × ∆φ ≈ 0.0625 × 0.2, corresponding typically to 20 strips in η, and
imax is the index of the highest-energy strip
Ratio of the energy difference between the largest and second largest energy
Eratio
deposits in the cluster over the sum of these energies
Ratio of the energy in the strip layer to the total energy in the EM accordion
f1
calorimeter
Track quality
Number of hits in the B-layer (discriminates against photon conversions)
nBlayer
Number of hits in the pixel detector
nPixel
Number of total hits in the pixel and SCT detectors
nSi
Transverse impact parameter
d0
Significance of transverse impact parameter defined as the ratio of d0
σd0
and its uncertainty
Momentum lost by the track between the perigee and the last
∆p/p
measurement point divided by the original momentum
TRT
Total number of hits in the TRT
nTRT
Ratio of the number of high-threshold hits to the total number of hits in the TRT
FHT
Track-cluster
∆η between the cluster position in the strip layer and the extrapolated track
∆η1
matching
∆φ between the cluster position in the middle layer and the extrapolated track
∆φ2
Defined as ∆φ2 , but the track momentum is rescaled to the cluster energy
∆φres
before extrapolating the track to the middle layer of the calorimeter
Conversions
Goudet (LAL)
Ratio of the cluster energy to the track momentum
E/p
Veto electron candidates matched to reconstructed photon conversions
isConv
Electrons & photons for Higgs measurements
JRJC, November 17, 2015
1/3
Reconstruction & Identification efficiencies
Not all electrons pass the reconstruction and identification criteria.
3 menus with increasing purity ( but deceasing efficiencies) are defined :
loose, medium, tight. The efficiency of these procedures is given as a function of
the pT and η = −ln(tan(θ/2)).
Goudet (LAL)
Electrons & photons for Higgs measurements
JRJC, November 17, 2015
2/3
Photon correction
0.025
0.02
0.015
∆α
∆α
Electrons scale factors are also applied to photons. An additionnal scale factor (∆α)
is measured from Z → ll γ.
Unconverted photons
0.025
0.02
Data
Calibration uncertainty
0.01
0.005
0.005
0
0
-0.005
-0.005
-0.01
-0.01
-0.02
0
ATLAS
0.2 0.4 0.6 0.8
s=8 TeV;
1
Data
Calibration uncertainty
0.015
0.01
-0.015
Unconverted photons
∫
-1
L dt = 20.3 fb
1.2 1.4 1.6 1.8
2
-0.015
-0.02
2.2
ATLAS
10
s=8 TeV;
20
30
40
∫
-1
L dt = 20.3 fb
50
|η|
Goudet (LAL)
Electrons & photons for Higgs measurements
60
ET
JRJC, November 17, 2015
3/3