sPHENIX discussion of physics

[Anthony Frawley <afrawley AT fsu.edu>]

Hi All,

Sorry, I missed a page of my notes from the meeting. See the updated minutes here.

Tony

Particle Flow E/p study for single pi+ and e- tracks: Austin Rosypal.

------------------------------------------------------------------------------------

Slide 5: 

Jakub: Does "unmatched EM" mean an EMCal cluster with no matched track? Yes.

Slide 7: 

Jin: Is the high tail on E/p due to the radiative tail on the electron p measurement? Will look into it.

Jin: In simulations, can study cuts by using the subset of electrons that did not radiate. Select with p_reco/p_truth = 1.

Marzia: In real life there are many more pions than electrons. It would be better to use pion rejection, not electron purity.

Jin: Suggest that pion rejections vs electron efficiency is a good plot to characterize performance.

A rejection of x20 is quite low, would expect more like x100.

Jin: PF may not be specialized enough for pion/electron separation. More sophisticated approach uses

Slide 10:

Dennis: These are only EMCal clusters, right? Should be in the label.

Tony: How were they selected? No selection.

Jin: Pions that shower early in the EMCal give us trouble. Can study these by cutting on high E/p pions.

Requiring no energy in HCal will help.

There are rare charge exchange events (< 1%)  where charged pions become neutral pions and decay to photons, which shower. Note that pi_ and pi- have different charge exchange rates. So pi+/e+ and pi-/e- comparisons are different.

Slide 12:

Tony: Why +/- 3% cut? The momentum resolution depends strongly on momentum.

Jin: The histogram of E/p vs eta and phi should be smooth, not lumpy, for this many events. Check to see if the random seed was fixed. That would give a statistical distribution due to the events in one condor job only.

Slide 14:

Marzia: Why does E/p peak at 1.1? EMCal energy or momentum? Will check.

Discussion:

Antonio: The selections for electrons in PF will not substitute for more sophisticated ones, e.g. for Upsilon analysis.

PF does not (presently) allow for special purpose modules as plugins. Selections are hard-coded. May need development.

Jin: Could you apply position dependent corrections for the energy scale in the EMCal? Antonio: the container is hard coded, but it could be changed.

Jakub: You can take PF elements and apply additional selections.

Marzia: For jets it is not important if particle ID is wrong - just want to not double count energy. For HF it matters.

Rosi: We are focused on trying to improve how PF deals with electrons.

---

From: sPHENIX-upsilons-l <sphenix-upsilons-l-bounces AT lists.bnl.gov> on behalf of Anthony Frawley via sPHENIX-upsilons-l <sphenix-upsilons-l AT lists.bnl.gov>
Sent: Thursday, February 29, 2024 2:23 PM
To: sPHENIX-HF-jets-l <sphenix-hf-jets-l-bounces AT lists.bnl.gov>; Sphenix-upsilons-l AT lists.bnl.gov <Sphenix-upsilons-l AT lists.bnl.gov>; sphenix-physics-l AT lists.bnl.gov <sphenix-physics-l AT lists.bnl.gov>
Subject: Re: [Sphenix-upsilons-l] HF/Q meeting tomorrow, February 28, at 1:30 pm ET

 

Here are the notes from the meeting yesterday.

Tony

Particle Flow E/p study for single pi+ and e- tracks: Austin Rosypal.

------------------------------------------------------------------------------------

Slide 5: 

Jakub: Does "unmatched EM" mean an EMCal cluster with no matched track? Yes.

Slide 7: 

Jin: Is the high tail on E/p due to the radiative tail on the electron p measurement? Will look into it.

Jin: In simulations, can study cuts by using the subset of electrons that did not radiate. Select with p_reco/p_truth = 1.

Marzia: In real life there are many more pions than electrons. It would be better to use pion rejection, not electron purity.

Jin: Suggest that pion rejections vs electron efficiency is a good plot to characterize performance.

A rejection of x20 is quite low, would expect more like x100.

Slide 12:

Tony: Why +/- 3% cut? The momentum resolution depends strongly on momentum.

Jin: The histogram of E/p vs eta and phi should be smooth, not lumpy, for this many events. Check to see if the random seed was fixed. That would give a statistical distribution due to the events in one condor job only.

Slide 14:

Marzia: Why does E/p peak at 1.1? EMCal energy or momentum? Will check.

Discussion:

Antonio: The selections for electrons in PF will not substitute for more sophisticated ones, e.g. for Upsilon analysis.

PF does not (presently) allow for special purpose modules as plugins. Selections are hard-coded. May need development.

Jin: Could you apply position dependent corrections for the energy scale in the EMCal? Antonio: the container is hard coded, but it could be changed.

Jakub: You can take PF elements and apply additional selections.

Marzia: For jets it is not important if particle ID is wrong - just want to not double count energy. For HF it matters.

Rosi: We are focused on trying to improve how PF deals with electrons.

---

From: Anthony Frawley <afrawley AT fsu.edu>
Sent: Tuesday, February 27, 2024 3:40 PM
To: sPHENIX-HF-jets-l <sphenix-hf-jets-l-bounces AT lists.bnl.gov>; Sphenix-upsilons-l AT lists.bnl.gov <Sphenix-upsilons-l AT lists.bnl.gov>; sphenix-physics-l AT lists.bnl.gov <sphenix-physics-l AT lists.bnl.gov>
Subject: HF/Q meeting tomorrow, February 28, at 1:30 pm ET

 

Hello All,

We will have our next joint Heavy Flavor/Quarkonium TG meeting tomorrow 2/28/24 at 13:30pm ET 

Currently on the agenda at https://indico.bnl.gov/event/22555/ we have:

 

•      Introduction - Conveners

•  E/p discrimination for electrons and pions - Austin Rosypal

The zoom link is bnl.zoomgov.com/j/1613938007?pwd=QjYzUnpYQ2xNZWZLSy9qd3Z4dFpzZz09
 
We look forward to talking to you tomorrow!
Marzia, Cameron, Jin, and Tony