R.Blair, S.Chekanov, R.Norberg, M.Stockton, plus the Direct Photon group

Public results ATLAS-CONF-2013-02

Current CDS note: http://cdsweb.cern.ch/record/1453853

The current note is located here: https://svnweb.cern.ch/trac/atlasphys/browser/Physics/StandardModel/PromptPhotons/Notes/InclusivePhotons3fb How to get and work with it:

Outside CERN:

/usr/kerberos/bin/kinit   <username>@CERN.CH

(or just kinit <username>@CERN.CH). Then find the note:

export SVNROOT="svn+ssh://svn.cern.ch/reps/atlasphys/Physics/StandardModel/PromptPhotons/Notes/InclusivePhotons3fb"

Getting the note:

svn co $SVNROOT

This will create a directory “InclusivePhotons3fb”. “cd” into it and compile the package “make”. Always make sure that you keep the most recent directory “svn update”.

Modify a file, then do “svn ci” and put notes. To remove a file, use “svn remove file”. To add file, use “svn add file”.

Read here howto use SVN: https://twiki.cern.ch/twiki/bin/viewauth/Atlas/SoftwareDevelopmentWorkBookSVN

Old instruction:

The current note is located here: 
[[https://svnweb.cern.ch/trac/atlasusr/browser/chekanov/Physics/Photons/Notes/InclusivePhotons2011]]
How to get and work with it:
Outside CERN: 
/usr/kerberos/bin/kinit   <username>@CERN.CH
export SVNUSR="svn+ssh://svn.cern.ch/reps/atlasusr/chekanov"
Getting the note: 
svn co $SVNUSR/Physics//Photons/Notes/InclusivePhotons2011
This will create a directory "InclusivePhotons2011". 
"cd" into it and compile the package "make". Always make sure that you keep the most recent directory "svn update".
Modify a file, then do "svn ci" and put notes. To remove a file, use "svn remove file". To add file, use "svn add file".
Read here howto use SVN: [[https://twiki.cern.ch/twiki/bin/viewauth/Atlas/SoftwareDevelopmentWorkBookSVN]]

• final_gamma_pt.eps.png - pT of gamma in |Eta(gamma)|<1.37
• final_gamma_eta.eps.png - Eta of gamma in the central Eta region
• final_gamma_pt_F.eps.png - pT of gamma in the forward 1.52<|Eta|<2.37

Systematical and statistical errors added in guadrature.

Green bads shows the renormalisation scale uncertanty plus PDF uncertainties from 52 eigenvectors of CT10 PDF. Yellow-gray band shows the renormalisation uncertanty only.

Here is a plot with 2010 cross sections (extended to 25 GeV)

• final_gamma_pt_PUB.eps.png - pT of gamma in |Eta(gamma)|<1.37.

• final_gamma_pt_PUB_zoom.eps.png - same after zooming in

We follow the general recommendations described here

Current selection cuts:

2011 data:

iLumiCalc.exe –lumitag=OflLumi-7TeV-002 –livetrigger=L1_EM14 –trigger=None –xml=/tmp/lumifiles/795b77/data11_7TeV.periodAllYear_DetStatus-v18-pro08-05_CoolRunQuery-00-03-98_GammaJet.xml

1. Total Luminosity: 1.072 fb-1
2. NTUPLE_PHOTON sample
3. Event selection: 1 PV having at least 3 tracks and EF_g80_loose trigger. See: https://twiki.cern.ch/twiki/bin/view/AtlasProtected/CaloIsolationCorrections

Photon selection:

1. ph_goodOQ rejection
2. isEM & 0x800000 – AR cut (missing B-layer)
3. if ( !( (ph_OQ&34214)==0) )
4. if ( ( (ph_OQ&134217728)!=0 && (ph_reta>0.98||ph_rphi>1.0||(ph_OQ&67108864)!=0)))
5. pT(γ)>100 GeV and |η (γ)|<1.37  and  1.52<|η (γ)|<2.37
6. Isolation cut at 7 GeV. We use Etcone40_corrected that has both the leakage and ambient energy density pileup corrections applied.
7. Pile-up reweighting using the standard PileupReweighting tool

PYTHIA  simulation

1. Signal gamma+jet: 108082 (70 GeV),   108083 (140 GeV), 108084 (280 GeV), 119079 (500 GeV)
2. QCD  (JF sample): 105814 (70 GeV),    105815 (140 GeV),  105812 (240 GeV)

HERWIG simulation (only for systematics on efficiency)

1. Signal gamma+jet: 108092 (70 GeV), 108093 (140 GeV), 108094 (280 GeV), 119078 (500 GeV)

All samples are weighted according to their cross sections and generated efficiencies

General MC status is given here

MC simulations include pile-up as shown in this figure: det_nPV.eps.png

• det_gamma_iso.eps.png - isolation for pT>100 GeV

Isolation in bins of pT

• det_gamma_iso_pt1.eps.png pT>100 GeV
• det_gamma_iso_pt2.eps.png 300<pT<350 GeV
• det_gamma_iso_pt3.eps.png 600<pT<700 GeV

Isolations for signal and background

• det_gamma_iso_DP.eps.png - for signal
• det_gamma_iso_JF.eps.png - for background

Detector level pT in the region A and D

• det_gamma_pt_A.eps.png - pT of gamma in region A
• det_gamma_eta_A.eps.png - Eta of gamma in region A

Look also at the background region:

• det_gamma_pt_D.eps.png - pT of gamma in region D

Detector level pT after the side-band backgroud subtraction. We use a simple backgrown subtraction (non-correlated) due to small statistics for MC. We use correlated quadrant method as systematics.

• det_gamma_pt_corr.eps.png - pT after the subtraction using “Correlation” method
• det_gamma_pt_simple.eps.png - pT after the subtraction using side-bands without correlations
• det_gamma_pt_bob.eps.png - pT after the subtraction using side-bands using Bob's method

1. Isolation cut is changed by - 1 GeV
2. Isolation cut is changed by + 1 GeV
3. Photon energy scale is propogated for converted and non-converted photons (separately). *Down variation*
4. Photon energy scale is propogated for converted and non-converted photons (separately). *UP variation*
5. HERWIG for efficiency correction - next
6. Using a simple data-driven approach without assumptions on correlations between A,B,C,D regions
7. Using alternative pile-up reweighting based on comparing nPV with the data

The tables below show the contributions of each systematic variation in % to the total cross section:

• sys_gammaPT.txt - for pT in the central
• sys_gammaETA.txt - for pT in the forward
• sys_gammaETA.txt - Eta in the central

The total systematic uncertainty is calculated by addeding all uncertainties in quadrature.

Bin-by-bin correction factors are defined as C= N(true photons after isolation)/N(reconstructed in the region A). The efficiency is 1/C.

• corr_pt.eps.png - C for pT of gamma in the central Eta region
• corr_eta.eps.png - C for Eta of gamma in the central Eta region
• corr_pt_F.eps.png - C for pT of gamma in the forward Eta region

Purity is defined as N(after bkg subtraction)/N(region A).

• purity_gamma_pt_simple.eps.png - Purity as a function of pT

This measurement is based on the inclusive analysis above by adding a antiKT4 jet.

* det_gamma_jet_mass.eps.png - M(gamma+jet) using the quadrand A for photons

The JETPHOX predictions are based on the JETPHOX V1.3 (ROOT-based) generating ROOT ntuples using the pT ranges as for the final plots. This technique is descibed on the CERN Twiki

4 different JETPHOX ntuples were generated. First sample has 44 PDF sets, 2 other are used to estimate PDF and scale variations, changing one scale at the time. Additional sample was generated by changing the scale and PDF scales coherently.

PDF uncertainties are calculated by taking a max deviation from the central value.

Scale uncertanties were estimated by considering the largest deviation from the central value using different combinations of the scale and PDF uncertainties.

Here are the cuts:

• Isolation cut is 7 GeV. The cone size is 0.4
• PT(gamma) (varies). Max PT(gamma)=2 TeV
• PT(jet)=50 GeV Max PT(jet)=2 TeV

All figure are here: All figures

Root files and tables are available here: All root files

Jetphox prediction: here

Direct-photon group: https://twiki.cern.ch/twiki/bin/view/AtlasProtected/SMDirectPhotons

Our wiki: https://twiki.cern.ch/twiki/bin/view/AtlasProtected/PhotonJetAtANL

JetPhox wiki: https://twiki.cern.ch/twiki/bin/view/AtlasProtected/SMDirectPhotonsPhotonJetCS

Theory papers: https://twiki.cern.ch/twiki/bin/view/AtlasProtected/SMDirectPhotonsBiblio

— Sergei Chekanov 2011/09/02 22:07