STAR Correlations and Fluctuations PWG

[EsumiShinIchi <esumi.shinichi.gn AT u.tsukuba.ac.jp>]

Dear Bijun

Thank you very much for the answers. I think I have been fooled by the 

automatic z(color) scale in root (also in log-scale) of the 2D histograms between 

page 4 and 5, I’m very sorry about it. It could just be because of statistics as you 

say and might even be similar between the forward- and mid-rapidities. So could 

you please choose the same linear scale for more systematic comparisons (like 

average +/- 10% or +/-20% fixed in the linear z scale) for all the panels, which 

could also be helpful to have more bins in rapidity, in stead of two large bins, 

even just 3 eta bins could help understanding the trend as you cover 1.3 eta 

units in total. You could even try to include pion and proton for these tests in order 

to get statistics to determine the pair cuts. 

Could you also please help me to translate your rapidity scale [-1.0, -0.5] and 

[-0.5, -0.1] to the eta scale of [-1.45, -0.15] of your figure that you have attached 

in your E-mail? Are they same scale (just with rapidity or with eta)? or are they 

shifted or reflected by the mid-rapidity etc, as I see more yield in lower part of 

your figure below about -0.9, which sounds like the mid-rapidity? Are they all 

PIDed particles as kaon, which is identified by TPC dE/dx (nSigmaKaon) alone 

or including TOF mass-square as well? Is identification method changes at around 

-0.95 in eta as I see some different structure there? Or do you apply different PID 

methods depending on the absolute momentum? Could you also plot this for a 

few different pT selection? (and momentum selection to understand the structure)? 

The pT (and momentum) selection (2 or 3 pT bins) should be feasible at least for 

these single particle plots even with limited statistics, like this "nHit vs eta” plot 

(which is better to be reversed x-y; nHit as a function of eta, though). 

For about the “iterative procedure”, you make the correlation function Fm with CRAB 

including momentum resolution effect, and another one Ft with the same weighting 

factor but without momentum smearing, so that one can determine the correction factor 

C = Ft / Fm. This correction factor C could in principle depends on the signal (shape of 

the correlation function) itself, which is in your case quantum effect from the source 

geometry and expansion given by the event generator model you used with CRAB 

including the coulomb effect. The corrected results (in terms of correlation function) 

could again be used to calculate the new Fm and Ft, where Ft is the corrected result 

itself including the coulomb effect and Fm is the smeared version of Ft, we can then 

define C again with these two new Fm and Ft for the next step and so on. If Ct is quite 

steep function as a function of qinv compared to the momentum resolution (amount of 

momentum smearing), Cm should be different from Ct, so that the correction factor is 

different from 1. If C~1, you would not even need any correction, but if it’s != 1, we 

might need to make sure whether it depends on the Ct (signal) itself to not. Therefore 

we need an iteration, do you get the point? If the momentum resolution is very good 

compared to the shape of the resulting correction function, the correction would be 

minor/small, that might be the case for you as well. 

Best regards, ShinIchi

2023/05/31 20:44、樊碧君 <bjfan AT mails.ccnu.edu.cn>のメール:

Dear ShinIchi,

Thank you very much for your thoughtful comments.

The formula I used for calculating phi* is following,

phi* = phi + asin(sign*0.15*0.5*R/pT). 

As the statistics(we used the full statistics) of K+ in FXT energies is too low, it is difficult to see the dependence of different R. What's more, the statistics of forward rapidity is lower than mid-rapidity, it's more difficult to see that. The following 2D plot shows nHitsfit vs. eta after PID and track cuts.

<300AFAFF AT 8E98C959.3033776400000000.png>

As for the weight in CRAB, the "raw" I mentioned last email refers to "un-smeared". Sure, when we fill the qinv histograms for smeared correlation function, I use the smeared qinv to do that. But I couldn't understand what you said "You would also need to do this correction in an iterative way, since you do not know the final results to start with…" Could you please kindly tell me what's the "iterative way" means?

About the last suggestion you mentioned, due to the limited statistics of kaon, at FXT energies, it might be tough, but I'd like to do that in different pT regions in my 7.7 GeV analysis, as at this energy, there are more statistics.

Thanks and regards,

Bijun

------------------ Original ------------------

From:  "EsumiShinIchi"<esumi.shinichi.gn AT u.tsukuba.ac.jp>;

Date:  Wed, May 31, 2023 10:08 AM

To:  "star-cf-l"<star-cf-l AT lists.bnl.gov>;

Cc:  "樊碧君"<bjfan AT mails.ccnu.edu.cn>;

Subject:  Re: [Star-cf-l] CF PWG meeting, May 25th, 2023, new time: 9:00 (BNL time)

 

Dear Bijun

On the last part of the 1st paragraph, it might be worth to look at these

2D plots for a few different pT regions. Thank you and sorry for bothering

you several times with these E-mails…

Best regards, ShinIchi

2023/05/31 11:00、EsumiShinIchi <esumi.shinichi.gn AT u.tsukuba.ac.jp>のメール:

Dear Bijun

Thank you very much for very good additional studies. I would agree

with you that phi* and phi cuts are equally good based on your plots.

I would like to ask a few more questions, do you know why the size

of the hole in page 4 are almost independent of the radius choice of

phi*, which is counter intuitive. If the merging is happening inside the

TPC, I would expect some dependence according to the previous

studies. Could you remind me what is the exact formula for your

phi* = phi + (charge_sign) * Fun(pT, B-field, radius)?

Is it like this?  phi* = phi + sgn * asin(0.075*rad/pt);

Another question, do you know why we do not see a significant merging

effect in the forward rapidity case? I thought it could be shorter in terms

of the number of hit in the TPC for the forward tracks (although the actual

track length might be longer because of the theta angle), so that merging

and/or splitting effect might be even more severe in the forward region.

So could you please plot "nFitHit vs eta (or rapidity)” with your track

quality and PID selections?

My last question, that I did not fully understand your last sentence,

"When I calculate the weight of same events distribution using CRAB,

I use the raw qinv to calculate the weight.” What does it mean? We would

need to use the generated (un-smeared) momentum (or qinv) for the

weight calculation with CRAB, but we would need to use the reconstructed

(smeared) momentum (or qinv) for filling the histograms to get the

correlation function in order to include the effect from the momentum

resolution on the correlation including the coulomb effect. If you mean

“raw” as “un-smeared”, it sounds OK, but you still need to use the

“smeared" qinv for making the real and mixed pair distributions. You

would also need to do this correction in an iterative way, since you do

not know the final results to start with… Thanks again for your follow up.

Best regards, ShinIchi

2023/05/30 23:21、樊碧君 via Star-cf-l <star-cf-l AT lists.bnl.gov>のメール:

Hi Esumi, Grigory, Hanna, and all,

I have tried the phi star method to do the track merging effect at 3.2 GeV, and the details and the comparison of Corelation functions of delta phi and delta phi star are shown in the attachment, they are consistent within error; What's more, When I calculate the weight of same events distribution using CRAB, I use the raw qinv to calculate the weight. Thank you for your nice comments!

Best regards,

Bijun

 

------------------ Original ------------------

From:  "樊碧君"<bjfan AT mails.ccnu.edu.cn>;

Date:  Thu, May 25, 2023 05:31 PM

To:  "Hanna Paulina Zbroszczyk"<hanna.zbroszczyk AT pw.edu.pl>; "star-cf-l"<star-cf-l AT lists.bnl.gov>;

Subject:  Re:[Star-cf-l] CF PWG meeting, May 25th, 2023, new time: 9:00 (BNL time)

 

Dear All,
             Please find my slides for today meeting:
https://drupal.star.bnl.gov/STAR/system/files/KK_3p2GeV_CF_0525_pwg.pdf

Best and regards,
Bijun

 

 

------------------ Original ------------------

From:  "樊碧君"<bjfan AT mails.ccnu.edu.cn>;

Date:  Wed, May 24, 2023 08:48 AM

To:  "Hanna Paulina Zbroszczyk"<hanna.zbroszczyk AT pw.edu.pl>; "star-cf-l"<star-cf-l AT lists.bnl.gov>;

Subject:  Re:[Star-cf-l] CF PWG meeting, May 25th, 2023, new time: 9:00 (BNL time)

 

Dear Hanna, Toshihiro, and All,

If time permit, I would like to present an update on the k+ femtoscopy at 3.2 GeV.

Could you please add me to the agenda?  Thank you!

Best,

Bijun

 

------------------ Original ------------------

From:  "Hanna Paulina Zbroszczyk via Star-cf-l"<star-cf-l AT lists.bnl.gov>;

Date:  Tue, May 23, 2023 03:04 PM

To:  "star-cf-l"<star-cf-l AT lists.bnl.gov>;

Subject:  [Star-cf-l] CF PWG meeting, May 25th, 2023, new time: 9:00 (BNL time)

 

Dear CF Activists,

We will have our weekly CF PWG meeting on Thursday, May 25th, 9:00 AM 
(BNL time).

Please note that this week we start at a new time (30 minutes earlier than before).

If you would like to contribute, please let us know.

Please try to post your slides at your earliest convenient.
We will connect via Zoom. Please find the connection details under the
meeting webpage.

Thank you,
Best regards,
Toshihiro and Hanna

https://drupal.star.bnl.gov/STAR/blog/luoxf/correlations-and-fluctuations
https://drupal.star.bnl.gov/STAR/blog/gos/correlations-and-fluctuations-preliminary-plots 

Hanna Paulina Zbroszczyk

PhD DSc Eng, Professor WUT

E-mail: hanna.zbroszczyk AT pw.edu.pl

Tel: +48 22 234 5851 (office)

Address:

Warsaw University of Technology

Faculty of Physics

Nuclear Physics Division

Koszykowa 75

Office: 117b (via 115)

00-662 Warsaw, Poland

<deltaphistar_3p2_KK.pdf>_______________________________________________
Star-cf-l mailing list
Star-cf-l AT lists.bnl.gov
https://lists.bnl.gov/mailman/listinfo/star-cf-l