STAR Correlations and Fluctuations PWG

[Rutik Manikandhan <manikandhan.rutik AT gmail.com>]

Hello Sooraj,

Thank you for your feedback.

//In the correlation analysis, our lower cut off was 0.20 GeV/c and we changed it +/- 0.02 GeV/c. We also changed our eta window for different acceptance and accounted for all this in our systematic uncertainty, but for the correlator analysis these uncertainties are quite small.

For UrQMD we checked the correlator with different eta acceptance windows in COM and Collider mode and all of them gave us zero. 

So, we performed an asymmetric acceptance cut at 11.5 GeV (where the correlator was non zero). I'm attaching the plot of that study here, we did not do a lower momentum cutoff study yet and we plan to do that in future.//

What is the COM here? Do you mean FXT set up? If you look at the acceptance plots for various particles at 3 GeV, you can see that even when you use a 0.5 rapidity window around mid-rapidity, the acceptance is very asymmetric. The low pT reach is changing with the rapidity and differently for different particles. So you are effectively changing the particle composition and kinematic reach in pT at the FXT energies. This for example, can enhance your fluctuations or covariances. Also would make it not directly comparable with collider energy measurements. One way to check would be to impose the acceptance constraints for various particles in UrQMD and see the impact. Not sure with the UrQMD study above you were checking for this. 

I meant COM for 3.0 GeV, UrQMD @ 3.0 GeV was generated in collider mode, so we boosted the frame to FXT and studied the correlator with varying acceptances.

It's understandable that varying particle acceptances would affect the correlations.

This is also why on slide 20, we do the analysis with varying pseudorapidity acceptance, and here we see at we get closer to mid-rapidity, the value saturates. This varying of eta acceptance has been taken into account for our systematic uncertainties as well. To be backwards compatible we compare the +/- 0.5 point with other energies, but I could compare it with +/- 0.2 and still make the same comparisons, and +/- 0.2 we will have uniform acceptance.

Our plan for the rest of the BES-II FXT energies is to study this effect precisely too.

Best,

Rutik Manikandhan

PhD Candidate, Experimental Nuclear High Energy Group

Physics Dept.

University of Houston

On Mon, Feb 12, 2024 at 11:51 AM Sooraj Radhakrishnan <skradhakrishnan AT lbl.gov> wrote:

Hi Rutik,

   Thanks for the reponses

//In the correlation analysis, our lower cut off was 0.20 GeV/c and we changed it +/- 0.02 GeV/c. We also changed our eta window for different acceptance and accounted for all this in our systematic uncertainty, but for the correlator analysis these uncertainties are quite small.

For UrQMD we checked the correlator with different eta acceptance windows in COM and Collider mode and all of them gave us zero. 

So, we performed an asymmetric acceptance cut at 11.5 GeV (where the correlator was non zero). I'm attaching the plot of that study here, we did not do a lower momentum cutoff study yet and we plan to do that in future.//

What is the COM here? Do you mean FXT set up? If you look at the acceptance plots for various particles at 3 GeV, you can see that even when you use a 0.5 rapidity window around mid-rapidity, the acceptance is very asymmetric. The low pT reach is changing with the rapidity and differently for different particles. So you are effectively changing the particle composition and kinematic reach in pT at the FXT energies. This for example, can enhance your fluctuations or covariances. Also would make it not directly comparable with collider energy measurements. One way to check would be to impose the acceptance constraints for various particles in UrQMD and see the impact. Not sure with the UrQMD study above you were checking for this. 

Best,

Sooraj

On Sun, Feb 11, 2024 at 3:33 PM Rutik Manikandhan <manikandhan.rutik AT gmail.com> wrote:

Hello Sooraj,

Here are my responses: 

//We use primary charged particles for the analysis//

This will only reduce contribution from weak decays. But there are others. The simple number fluctuations and independent emission assumption from mixed-events do not take these into account

Noted, maybe we can quantify the effect of these contributions from some models, but we haven't done that for now.

//Yes, exactly as you described, for instance, if I have an event of 100 particles, for the corresponding mixed event, I use 100 events and take a particle each.//

This can be indicated in the slide, would make it easier for those looking at your slides to follow

Done

//The lower cut for pT was changed systematically to account for systematic uncertainties//

In what range did you vary the low pT cut-off? The acceptance is strongly asymmetric w.r.t mid-rapidity. When you quote a -1,1 window, it is very different from that in the collider mode 

So, for the fluctuation analysis, the lower cut off was 0.15 GeV/c, which was changed to 0.13 and 0.17, I'd assume that this difference you mentioned is why our systematic uncertainties were large, hence the need for the correlation analysis.

// S16: How does the varying acceptance with rapidity affect the correlations here? Can the 3 GeV measurements be compared to those at higher energies?

        To understand this effect, we varied the acceptance region and saw that it saturates around midrapidity and then hits zero since we dont have enough particles to correlate.//

Since you are using a window around mid-rapidity and you go to very low pT, you are changing the particle composition and pT range, just from the acceptance, compared to those at higher energies. Since you are looking at fluctuations and correlations, these could have an impact. These can be checked in simulations like UrQMD with and without acceptance constraints as in data. Was any such study done?

In the correlation analysis, our lower cut off was 0.20 GeV/c and we changed it +/- 0.02 GeV/c. We also changed our eta window for different acceptance and accounted for all this in our systematic uncertainty, but for the correlator analysis these uncertainties are quite small.

For UrQMD we checked the correlator with different eta acceptance windows in COM and Collider mode and all of them gave us zero. 

So, we performed an asymmetric acceptance cut at 11.5 GeV (where the correlator was non zero). I'm attaching the plot of that study here, we did not do a lower momentum cutoff study yet and we plan to do that in future.

Here is the link for the updated version : https://drupal.star.bnl.gov/STAR/system/files/WWND_2024_ver_7.pdf

Best,

Rutik Manikandhan

PhD Candidate, Experimental Nuclear High Energy Group

Physics Dept.

University of Houston

On Sun, Feb 11, 2024 at 5:08 PM Sooraj Radhakrishnan <skradhakrishnan AT lbl.gov> wrote:

Hi Rutik,

   Thanks for the updated version and the responses. Please see a few further comments below

//We use primary charged particles for the analysis//

This will only reduce contribution from weak decays. But there are others. The simple number fluctuations and independent emission assumption from mixed-events do not take these into account

//Yes, exactly as you described, for instance, if I have an event of 100 particles, for the corresponding mixed event, I use 100 events and take a particle each.//

This can be indicated in the slide, would make it easier for those looking at your slides to follow

//The lower cut for pT was changed systematically to account for systematic uncertainties//

In what range did you vary the low pT cut-off? The acceptance is strongly asymmetric w.r.t mid-rapidity. When you quote a -1,1 window, it is very different from that in the collider mode 

// S16: How does the varying acceptance with rapidity affect the correlations here? Can the 3 GeV measurements be compared to those at higher energies?

        To understand this effect, we varied the acceptance region and see that it saturates around midrapidity and then hits zero since we dont have enough particles to correlate.//

Since you are using a window around mid-rapidity and you go to very low pT, you are changing the particle composition and pT range, just from the acceptance, compared to those at higher energies. Since you are looking at fluctuations and correlations, these could have an impact. These can be checked in simulations like UrQMD with and without acceptance constraints as in data. Was any such study done?

Best,

Sooraj

On Sun, Feb 11, 2024 at 1:09 PM Rutik Manikandhan <manikandhan.rutik AT gmail.com> wrote:

Hello Sooraj,

Thank you for your email.

Here are the replies to your comments:

S3: BES-II collider program ... 

       Done

S3: above --> >=

       Done

S3: and the BES-II FXT program ...

       Done

S4: Why 150 M? Previous analyses had 260 M good events. May be change to #of events used 

       This was from the fluctuation analysis list, but the change has been made

S5: TPC and TOF to reject out of time pileup events?

       This pile-up rejection is done using the StRefMultCorr package, so  yes it accounts for out of time

S7: How do correlated emission from resonance decays etc affect the <p_T> distributions? Mixed-events won't capture these correlations 

       We use primary charged particles for the analysis

S8: How are mixed-events constructed here? Not clear from the cartoon. Do you select for an event with N tracks, one track each from N different events to make the mixed event?

       Yes, exactly as you described, for instance, if I have an event of 100 particles, for the corresponding mixed event, I use 100 events and take a particle each.

S9: The reference article number seems not correct, could you check? Also, why is the distribution of mean not Gaussian?

       The distribution is not gaussian as we have a convolution of poissonian (as expected for thermal distribution) and Gaussian (due to detector effects) hence its a  

       Gamma distribution. I have corrected the reference now.

S10: What is eta [-1,1] here? Also, do we have uniform acceptance down to pT = 0.15 across the rapidity window at 3 GeV? How does this affect the dynamical fluctuations? For example, see the acceptance plots here https://www.sciencedirect.com/science/article/pii/S037026932200137X

          Since we are quoting 3.0 GeV as COM energy, Nu suggested using the acceptances in COM frame as well. The lower cut for pT was changed systematically to 

          account for systematic uncertainties

S10: Is the <p_T> expected to be higher at 3 GeV? Or is this an acceptance effect?

         It should be expected as the system is dominated with protons now, and <pT> has contributions from Teff and mass of the particle, this drives it up. 

         While analysing different energies as checks we also noticed that mean starts to rise up at 7.7 collider energy as well.

S12: How do you apply the efficiency correction? Do you do unfolding or subtract the difference of the mean from the simulation?

         I do 1D Bayesian unfolding with the Response matrix, I have added that line now.

S15: Why does the ALICE data point differ from the trend at top RHIC energies?

         In our paper we explain that difference due to an <Npart> contribution, since ALICE has higher <Npart>, the value becomes lower.

S16: How does the varying acceptance with rapidity affect the correlations here? Can the 3 GeV measurements be compared to those at higher energies?

        To understand this effect, we varied the acceptance region and see that it saturates around midrapidity and then hits zero since we dont have enough particles to  

        correlate.

S16: The value of the correlator at 3 GeV is closer to the value at LHC energy, and very different from that from UrQMD? What do we conclude here? 

         We did multiple checks, by boosting our frame in UrQMD, particle identifying, but the correlator gave us zero in all these scenarios, but from the correlator vs centrality plot (Slide 17), I believe we can infer that, at the central collisions, UrQMD doesnt have the sensitivity to obtain the small number the correlator is. But again we dont fully understand it as of now.

In the backup I have the same plot for 40-50 % centrality, there we see that UrQMD underpredicts the data with almost the same difference at all energies except 200 GeV.

S20: This is the impact of Delta eta between the pairs, but doesnt reflect the detector acceptance effects in the FXT set up  

         Here, Delta eta is our acceptance region around midrapidity, its not a pairwise cut, for instance Delta =1 is an acceptance window of [-1.55,-0.55], so +/- 0.5                   around midrapidity. To avoid the confusion, I have added an asterix remark.

S21: Do we understand the energy dependence? May be the last statement can be made more broad that we are working to understand the energy dependence as well?

        After accounting for it not being an acceptance effect, relating pt fluctuations to temperature fluctuations, naively we can say that this is to be expected,  as we get closer to a CP the temperature should get fixed (due to latent heat) and hence reduce the fluctuations on it. (This is my naive interpretation)

But if needed I can add a broader line that the dependence on energy is yet to be understood.

Here is the link with the updated version, do let me know if any more changes are needed.

https://drupal.star.bnl.gov/STAR/system/files/WWND_2024_ver_7_0.pdf

Best,

Rutik Manikandhan

PhD Candidate, Experimental Nuclear High Energy Group

Physics Dept.

University of Houston

On Sun, Feb 11, 2024 at 2:18 PM Sooraj Radhakrishnan <skradhakrishnan AT lbl.gov> wrote:

Hi Rutik,

   Please find some comments from me on your slides below 

S3: BES-II collider program ... 

S3: above --> >=

S3: and the BES-II FXT program ...

S4: Why 150 M? Previous analyses had 260 M good events. May be change to #of events used 

S5: TPC and TOF to reject out of time pileup events?

S7: How do correlated emission from resonance decays etc affect the <p_T> distributions? Mixed-events wont capture these correlations 

S8: How are mixed-events constructed here? Not clear from the cartoon. Do you select for an event with N tracks, one track each from N different events to make the mixed event?

S9: The reference article number seems not correct, could you check? Also, why is the distribution of mean not Gaussian?

S10: What is eta [-1,1] here? Also, do we have uniform acceptance down to pT = 0.15 across the rapidity window at 3 GeV? How does this affect the dynamical fluctuations? For example, see the acceptance plots here https://www.sciencedirect.com/science/article/pii/S037026932200137X

S10: Is the <p_T> expected to be higher at 3 GeV? Or is this an acceptance effect?

S12: How do you apply the efficiency correction? Do you do unfolding or subtract the difference of the mean from the simulation?

S15: Why does the ALICE data point differ from the trend at top RHIC energies?

S16: How does the varying acceptance with rapidity affect the correlations here? Can the 3 GeV measurements be compared to those at higher energies?

S16: The value of the correlator at 3 GeV is closer to the value at LHC energy, and very different from that from UrQMD? What do we conclude here? 

S20: This is the impact of Delta eta between the pairs, but doesnt reflect the detector acceptance effects in the FXT set up

S21: Do we understand the energy dependence? May be the last statement can be made more broad that we are working to understand the energy dependence as well?

Best,

Sooraj

On Sat, Feb 10, 2024 at 4:18 PM Rutik Manikandhan via Star-cf-l <star-cf-l AT lists.bnl.gov> wrote:

Hello Nu,

The idea was that a smaller transverse flow would keep the system together longer 

and hence increase the interactions, but the experimental evidence as you say would imply otherwise,

so I've removed that line.

Here is the updated version : https://drupal.star.bnl.gov/STAR/system/files/WWND_2024_ver_6_0.pdf

Thank you!

Best,

Rutik Manikandhan

PhD Candidate, Experimental Nuclear High Energy Group

Physics Dept.

University of Houston

On Sat, Feb 10, 2024 at 1:07 PM Nu Xu <nxu AT lbl.gov> wrote:

Dear Rutik and All,

slide 10: What is the experimental evidence to claim “The system remains in a regime of elastic interactions for a longer period of time as
compared to collider energies”? In fact, from pion HBT measurements, one observes that the duration of the pion source becomes longer at higher collision energy.

Best regards,
Nu

> On Feb 10, 2024, at 11:01 AM, Rutik Manikandhan <manikandhan.rutik AT gmail.com> wrote:
>
> Hello Nu,
> Thank you for your email.
> Here are my replies
>
> 1) slide 10 - last bullet: I do not see the connections with other points. I suggest remove it;
>     I still do not follow what you are talking about. For example, what do you mean by ‘longer time’? Longer compared to high energy collisions or what?
>  Yes, as compared to collider energies, I have added that now.
> 2) Plots on slides 21, 16, 13, 10: add unit for collision energy;
>    Added
> 3) slide 11: STAR QM2022 proceedings: add the Proceedings publication information.
>  Added
>
> Here is the updated version : https://drupal.star.bnl.gov/STAR/system/files/WWND_2024_ver_6.pdf
>
> Best,
> Rutik Manikandhan
> PhD Candidate, Experimental Nuclear High Energy Group
> Physics Dept.
> University of Houston
>
>
> On Sat, Feb 10, 2024 at 12:40 PM Nu Xu <nxu AT lbl.gov> wrote:
> Dear Rutik and All,
>
> Thank you for the updated draft talk. Here are my suggestions:
> 1) slide 10 - last bullet: I do not see the connections with other points. I suggest remove it;
>     I still do not follow what you are talking about. For example, what do you mean by ‘longer time’? Longer compared to high energy collisions or what?
> 2) Plots on slides 21, 16, 13, 10: add unit for collision energy;
> 3) slide 11: STAR QM2022 proceedings: add the Proceedings publication information.
>
> With these implemented, I sign off.
>
> Nu
>
>
> > On Feb 9, 2024, at 12:22 PM, Rutik Manikandhan <manikandhan.rutik AT gmail.com> wrote:
> >
> > Hello Nu,
> >
> > I have made the changes as you've asked for.
> > Link : https://drupal.star.bnl.gov/STAR/system/files/WWND__2024_ver_5.pdf
> > Here are our comments for your responses:
> >  1) slide 10 - last bullet: I do not see the connections with other points. I suggest remove it;
> >     The elastic interactions would possibly widen the distribution for <pT> as now more particles gain momentum in either direction.
> > 2) slide 11: It is not clear from the texts if the efficiency is extracted from embedding or other methods. Please make it clear;
> >     The plot has embedding efficiency on it, and I've added a QM reference
> > 3) slide 13: (i) again, the notion for eta is for different frame, please make it consistent. Otherwise it will lead to confusion; (ii) plot: for y-axis, I suggest change the lower limit to something like -0.25;
> >    Changed accordingly
> > 4) slide 16: (i) 2nd-bullet: the sentence is incomplete. Should be something like “Calculations from transport model …”; (ii) The meaning of bullet 3,4,5 is unclear. Not sure what do you want to say;
> >     Changed accordingly and added a relation between Temp and <pT>, we are trying to establish a thermal fluctuations scenario
> > 5) slide 18 - 2nd bullet: the sentence is incomplete. Something like “STAR data from 200 GeV Au+Au collisions …”. The same for the 4th bullet;
> >     Changed accordingly
> > 6) slide 19: It is redundant with respect to slide 18. I suggest remove it.
> >     This slide shows that UrQMD also follows an increase with decreasing centrality and that we might not have sensitivity at most central collisions, so we would like to keep it
> > 7) slide 21 - last bullet: at the end, add “especially at the 3GeV Au+Au collisions. The effect is not yet fully understood” or something like this. This means that we are still thinking about the new result. This is th eke slide of the talk. The UrQMD model is a hadronic transport model. Supposedly, it works better for low energy collisions where partonic medium may not be dominant, see Ref. 5 and 6. However, here as one can see that the model result is totally failed while the model calculations show a reasonable energy dependence of the pT-correlation at higher collision energies. I do not understand why so.
> >    Changed accordingly
> > 8) slide 21 - References: do we need 5 and 6?
> >     Changed accordingly
> >
> > Best,
> > Rutik Manikandhan
> > PhD Candidate, Experimental Nuclear High Energy Group
> > Physics Dept.
> > University of Houston
> >
> >
> > On Fri, Feb 9, 2024 at 12:05 PM Nu Xu <nxu AT lbl.gov> wrote:
> > Dear Rutik and All,
> >
> > Here are my suggestions for the draft:
> > 1) slide 10 - last bullet: I do not see the connections with other points. I suggest remove it;
> >
> > 2) slide 11: It is not clear from the texts if the efficiency is extracted from embedding or other methods. Please make it clear;
> >
> > 3) slide 13: (i) again, the notion for eta is for different frame, please make it consistent. Otherwise it will lead to confusion; (ii) plot: for y-axis, I suggest change the lower limit to something like -0.25;
> >
> > 4) slide 16: (i) 2nd-bullet: the sentence is incomplete. Should be something like “Calculations from transport model …”; (ii) The meaning of bullet 3,4,5 is unclear. Not sure what do you want to say;
> >
> > 5) slide 18 - 2nd bullet: the sentence is incomplete. Something like “STAR data from 200 GeV Au+Au collisions …”. The same for the 4th bullet;
> >
> > 6) slide 19: It is redundant with respect to slide 18. I suggest remove it.
> >
> > 7) slide 21 - last bullet: at the end, add “especially at the 3GeV Au+Au collisions. The effect is not yet fully understood” or something like this. This means that we are still thinking about the new result. This is th eke slide of the talk. The UrQMD model is a hadronic transport model. Supposedly, it works better for low energy collisions where partonic medium may not be dominant, see Ref. 5 and 6. However, here as one can see that the model result is totally failed while the model calculations show a reasonable energy dependence of the pT-correlation at higher collision energies. I do not understand why so.
> >
> > 8) slide 21 - References: do we need 5 and 6?
> >
> > That is all for now.
> >
> > Best regards,
> >
> > Nu
> >
> >
> >
> > > On Feb 9, 2024, at 8:51 AM, Rutik Manikandhan <manikandhan.rutik AT gmail.com> wrote:
> > >
> > > Hello Nu,
> > >
> > > The version is v4 now, and this is the link:
> > > https://drupal.star.bnl.gov/STAR/system/files/WWND_2024_ver4.pdf
> > >
> > > Do let me know if this works for you.
> > > Thank you.
> > >
> > > Best,
> > > Rutik Manikandhan
> > > PhD Candidate, Experimental High Energy Group
> > > Physics Dept.
> > > University of Houston
> > >
> > >
> > > On Fri, Feb 9, 2024 at 10:47 AM Nu Xu <nxu AT lbl.gov> wrote:
> > > Dear Rutik,
> > > Somehow the link to drupal does not work.
> > > Please fix it asap.
> > >
> > > Thank you,
> > > Nu
> >
>

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

--

Sooraj Radhakrishnan

Research Scientist,

Department of Physics

Kent State University

Kent, OH 44243

Physicist Postdoctoral Affiliate

Nuclear Science Division
Lawrence Berkeley National Lab
MS70R0319, One Cyclotron Road

Berkeley, CA 94720

Ph: 510-495-2473

Email: skradhakrishnan AT lbl.gov

--

Sooraj Radhakrishnan

Research Scientist,

Department of Physics

Kent State University

Kent, OH 44243

Physicist Postdoctoral Affiliate

Nuclear Science Division
Lawrence Berkeley National Lab
MS70R0319, One Cyclotron Road

Berkeley, CA 94720

Ph: 510-495-2473

Email: skradhakrishnan AT lbl.gov

--

Sooraj Radhakrishnan

Research Scientist,

Department of Physics

Kent State University

Kent, OH 44243

Physicist Postdoctoral Affiliate

Nuclear Science Division
Lawrence Berkeley National Lab
MS70R0319, One Cyclotron Road

Berkeley, CA 94720

Ph: 510-495-2473

Email: skradhakrishnan AT lbl.gov