-- SeoKonKang - 21 Jan 2010


*** Discussion title: Heavy Ion Physics

Posted below and linked on the agenda are the minutes. Here are some 
links promised in the meeting:

Jet Algorithms meeting with slides on using internal anti-kT pileup 
subtraction on p+p pileup events: http://indico.cern.ch/conferenceDisplay.py?confId=74290

Overview analysis note on Jet Energy corrections: http://cms.cern.ch/iCMS/analysisadmin/cadi?ancode=JME-07-002
Other documents on Jet Energy Corrections: https://twiki.cern.ch/twiki/bin/view/CMS/JetEnergyScale
#Documents_and_Analysis

--------------------------------
[high_pt] meeting minutes, 19 Jan 2010
--------------------------------
agenda: http://indico.cern.ch/conferenceDisplay.py?confId=78341
--------------------------------
Irina: Jet Plus Track in Heavy Ions
Irina described the combination of the Pileup Jet finder with the Jet 
Plus Track algorithm for use in Heavy Ion events.
-This algorithm was implemented in ORCA, and is now implemented in CMSSW
.... Discussion during time Olga & Irina were offline...
-Discussion of going through a full analysis like particle flow, to 
match the tracks with the calorimeter and using that to see the energy 
of jets
-Question of why mean value of background subtracted
        -Answer is mostly for historical reasons
-Question about separating neutral from charged energy
        -Reminder about stopping in ECAL, and for instance, a 1 GeV track 
gets stopped and dumps all its energy into ECAL crystals. This makes 
response difficult for charged hadrons. Have to do an extrapolation to 
the calorimeter and guess what the track deposited, and assume a pion 
mass.
....
-Track response is from isolated tracks in p+p with a special Non Zero 
Suppressed data stream.
-Background tracks  is calculated in an event by event basis
-different algorithmic efficiency is observed between p+p and HI
     -Look at hypernews, Ed emailed Jorge about all the track 
collections created in HI track reconstruction, and consult the twiki: https://twiki.cern.ch/twiki/bin/view/CMS/SWGuideHeavyIonTrackReco
        ***suggest to look at pT, eta, and fakes, and consider pT cutoff of 
different track containers (discussion moved to mailing list)
-JPT + PU is implemented for iCone5, SisCone5, and Anti-Kt5
-Question if anyone has studied internal kT pileup subtraction
        -Gavin Salam has done studies in Heavy Ions.
        -Philipp Schieferdecker showed some studies: http://indico.cern.ch/conferenceDisplay.py?confId=74290
-Question of usefullness of JPT with low track reconstruction efficiency
        -In p+p, seen that tracking algorithmic efficiency correction is not 
needed
        -In Heavy Ions can apply this correction (dependent on centrality)
        -It's not a real problem
***Suggestion to look at energy scale and resolution in larger sample 
studies
-------------------------------------------------------

Marguerite: General News
***Suggestion that Marguerite sets the meeting agenda and request talks
-MC: Need to have official Heavy Ion production and have dataops 
produce more events
*** People need to look at samples
-Photons:
        1) Have old 3_3_X samples, and plan to compare data at 900 GeV and MC 
in p+p, look at the photon object in PAT
        2) Validate heavy ion variables, and study in p+p (data and MC), will 
show slides on 29th meeting
*** Send/post slides for HI-West meeting update by the 27th
*** iterativeCone5Pileup code has been reorganized, but not yet 
validated or in official CMSSW
*** Suggestion to test the algorithms we have on hand and see if they 
are good enough to extract first physics. It would be useful to start 
an analysis exercise on a largescale MinBias dataset and try to 
extract physics.
*** Suggestion to also closely follow p+p physics groups use of p+p 
data, and try our algorithms on recorded data.
-Discussion of current "paper" topics: Jet Shapes, Jet Fragmentation 
Function, and Photon R_AA
-Photon group will measure inclusive photon spectrum in QCD p+p group, 
and ensure the reference data is taken and covers a low enough pT 
range. Also, practice photon spectra measurement in a MinBias sample.
-Reminder that we will have difficulty in review process because there 
does not exist a good Monte Carlo for Heavy Ion data to compare to.
*** Alternate proposal: Look at particle flow type analysis, develop 
for heavy ions.
-----------------------------------------------------------

Yetkin: Monte Carlo requests
*** Need to understand if turning off pyquen collisional energy loss 
and changing dijet settings to match p+p is good for us AND other PiNGs
-Important to make our Pythia QCD dijet settings match the p+p group's 
work to take advantage of their efforts on jet energy scale, 
resolutions, and corrections.
*** Would like to see these plots with more statistics
*** Suggest to make presentation in a CMS HI meeting of collisional 
energy loss and full proposal to communicate to the rest of HI.
-----------------------------------------------------------

Yetkin: Jet Correction plans:
-Shows status as of November 2009
-Use same data samples, but concentrate on signal events, no heavy 
ions. Use pp track and jet reconstruction to understand what happens 
when jets are quenched and the jet fragmentation is modified, and how 
that effects the jet energy corrections.
***Need more statistics
        -And/or to fit distributions when looking at subtracting quenched-
unquenched
-Used own method for single particle response
-Need to figure out how to measure RMS, energy correlations, and test 
resolution (from data).
-Suggestion based on RHIC work looking at fluctuations in background, 
measuring random cones and what area is recovered.
-Concern that this is difficult to fold into the fragmentation 
function, considering nonlinearities in the calorimeter
-Need to be careful of how energy scale is sensitive to parton 
composition, and for instance, if you make a pythia pt-hat bin of 
100-150, and then plot 60 GeV jets, the distribution is biased towards 
gluon jets. Need to be careful combining samples and binning them.
*** slide 3 statistical error bars are probably not correct
*** Suggestion to carefully spend time checking distributions and 
errors and do this study carefully
-----------------------------------------------------------

Seo Kon: JetMET in pp samples
-Studying p+p samples in 3_3_1 with Jet Response analyzer and QCD flat 
pT
-slide 3: we assume the Trigger result (HLT content?) is ported from 
Reco content to the Jet Response Analyzer
-Comparisons are made between CaloJet and Reference jets which have 
been matched
-Comparison mades to corrected jets
*** Links to jet energy corrections notes:
Main twiki: https://twiki.cern.ch/twiki/bin/view/CMS/JetEnergyScale
Overview analysis note on Jet Energy corrections: http://cms.cern.ch/iCMS/analysisadmin/cadi?ancode=JME-07-002
Analysis notes that describe how corrections are derived in p+p 
(remember that things like Z+jet and photon+jet balance will be 
affected in Heavy Ions by jet quenching, we do want to try p+p 
corrections as is):
https://twiki.cern.ch/twiki/bin/view/CMS/
JetEnergyScale#Documents_and_Analysis

*** Suggestion: Take Monte Carlo, and see if you can derive jet energy 
corrections
*** Recommendation to concentrate on:
1) How to apply corrections
2) Make Corrections from a Monte Carlo file
3) Take these Jet Response Analyzer corrections and make a .txt (or 
other type) of file which can be implemented in CMSSW. For a specific 
tag/dataset, that can be compared to the official corrections.
-Request for update every meeting from Seo Kon
-----------------------------------------------------------

-Marguerite
Edit | Attach | Watch | Print version | History: r1 | Backlinks | Raw View | WYSIWYG | More topic actions
Topic revision: r1 - 2010-01-21 - SeoKonKang
 
KOREA-CMS
This site is powered by the TWiki collaboration platform Powered by Perl This site is powered by the TWiki collaboration platformCopyright © by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding KoreaCmsWiki? Send feedback