Introduction

The purpose of this study is to check the prediction of the conversion method in the QCD monte carlo, checking the prediction for the number of photons in a cone of R=0.3 which pass the charged particle veto (CPV) cut, to the number found in the GEANT record.

Sample

I'm using the MIT filtered QCD 2->2 monte carlo samples defined here:

http://drupal.star.bnl.gov/STAR/comp/grid/infrastructure/sites/mit/mit-simulation-productions

The samples are weighted and combined together, using a  little class written by Michael, which he kindly shared.

Event Selection

(1)Select standard candidates with the StGammaMaker.

(2)Candidate pt > 5 GeV.

(3)Ratio of pt in 1X1 cluster, corrected for leakage,  to pt in a cone of radius 0.3 > 0.9.  (Isolation cut).

(4)Energy in the preshower cone of radius 0.3 around the candidate is 0. (CPV).

(5)Energy in the second preshower cone of radius 0.3 around the candidate is greater than 0 (analysis cut).

The efficiency, e (because I can't find Greek letters) , is given by the ratio of the number of events passing stage 5 to the number of events passing stage 4.

The prediciton of the conversion method is that:

(1-e_bg)= (1-e_sig)^N where e_sig = 1.0-e^{-0.678/cos(theta)} and N is the number of photons in your cone of radius 0.3.

Method

I implemented the event selection.  I divided the sample into 3 regions of pseudo-rapidity and found the efficiency of the analysis cut in each region.  Then I use the equation above to fit for N.

The parameter in this fit is for the number of photons.  I fit 1.3.

Then I look in the GEANT record and find how many photons can be found in a cone of radius 0.3 around a candidate which passes the CPV cut.  I find, on average, 1.7 photons.

Comments

So for the first pass the numbers don't agree.  However, I'm not totally convinced of my answer.  First, the way the GEANT photons end up saved in the gamma trees is a little confusing - I should at least be sure there is no double counting.  Second, I haven't convinced myself that I shouldn't be applying some cut when I count GEANT photons in the cone.   I thought, for example, maybe the same pT cut should be applied to the GEANT photons.  If I require their pT>5.0, however, I match only 1 photon within the cone.  I'll have to think about this.