STAR HardProbes PWG

["Ma, Rongrong" <marr AT bnl.gov>]

Hello Rosi

I agree that we should be careful not to over-interpret the results. My main concern is that the current jet v2 measurement potentially has very large non-flow contribution, and we should point that out. 

It looks like we mean different things when we say "non-flow". In a v2 measurement, there is signal, which could come from collective flow (low pt) or path-length dependent energy loss (high pt), and non-flow background. I do not think people call v2 signal from path-length dependent energy loss non-flow. The non-flow background usually refers to the influence of jet signal in event plane determination, and Shengli's result show that this influence increases with track pt, which should also affect Tristan's results. The effect might decrease at very high pt, but have we done any quantitive studies?

Thanks. 

Best

Rongrong

On Aug 29, 2023, at 1:35 PM, Rosi Reed <rosijreed AT lehigh.edu> wrote:

Hi All,

I think we need to be careful in not interpreting the results too far just yet.   Some thoughts:

>>What do you mean by "he is entirely looking at nonflow"? If I understand correctly, Tristan is looking at correlations between 

>>mid-rapidity jets and forward-rapidity event plane by EPD. If the EPD event plane is NOT influenced by say away-side jets, then his

>>v2 signal is one he is after. However, Shengli's results show that there is substantial non-flow contribution.

Yes, he is looking at mid-rapidity jets and forward rapidity event plane with a mid-rapidity centrality definition.  Maybe I am a little confused here, Shengli is showing that there is substantial non-flow contribution at mid-rapidity even when using the EPD for a rapidity gap, or am I mistaken?  (I  have been looking at: https://drupal.star.bnl.gov/STAR/system/files/OO-ShengliHuang-QM2023_1.pdf but I'm happy to be pointed to other things.)

I think the EPD EP can be influenced by away-side jets in general, however, when selecting events where you have a high momentum mid-rapidity jet, the probability decreases considerably - especially in OO where Nbin is so small.  Probably at a track pt of 5 GeV there is still some contamination, but much higher probably not so much because the awayside parton will have enough pt that it can't swing so much.

>> So I think Tristan's result is a mixture of flow and non-flow signals, and the contribution of latter might even be dominating, which

>> I think is important to point out.

I guess I'm still stuck here, he's analyzing jets which are non-flow....  So I would say his results are non-flow with some potential contamination from flow.  The reality is, we actually want to cut all the flow out - the question is whether there is some azimuthal assymmetry in the hard sector.

>> the away side and thus v2 (including using EP method) gets significant contribution from jet-like correlations and decays that

>> are not correlated with the EP and thus not v2.

Well, I would disagree that jet-like correlations aren't correlated with the EP. :-)  I would completely agree that when looking at a normal "flow" v2 analysis that you have to worry about contributions from non-flow (and you have to worry about them precisely because they are correlated wiht the EP - if they were completely uncorrelated it would be a resolution issue).  I also think scalar product and other methods are probably better for this reason.

>> For eg, you can see the near side ridge in the pPb measurements here 

https://journals.aps.org/prc/pdf/10.1103/PhysRevC.90.044906.

Yeah... but this is for events where one isn't selecting for mid-rapidity jets.  So I completely agree with you here, and  I think this is of interest to many people - the question of why hydrodynamics seems to make predictions for small systems is an important one.  But for me, I'm after the answer to a different question.  Why is there a non-zero jet v2 in small systems, since they are too small to have path length dependence and also there are no other signatures of jet quenching in the nuclear modification factor/jet structure.  As you can see in: https://arxiv.org/pdf/1910.13978.pdf - and in those events with jets the away-side contamination is very minimal.

I digress.  The reason I'm pushing back is I think there is something that is not understood here, not by us, not by Atlas, not really by anyone because the origin of jet v2 in small systems is IMO unknown.  So I think Tristan really needs to report what he did (i.e. find the EP at 2 < |eta|<5 and jets with |eta| < 1 - R), and not try to discuss flow vs non-flow, etc.  The high pt track v2 is mainly there to confirm that the jet-finding isn't adding something odd.  The comparison with Shengli with no additional subtraction to the v2 signal validates the procedure.  

Well - I'm being interrupted again by students so I will send this - essentially I'm happy with Rongrong's suggestion of v2^obs and I want to minimize what we say about it other than we see a non-zero v2 in OO that seems real in the hard sector.

Cheers,

Rosi

On Tue, Aug 29, 2023 at 12:54 PM Sooraj Radhakrishnan <skradhakrishnan AT lbl.gov> wrote:

Hi Rosi,

   To add to the discussion, the away side and thus v2 (including using EP method) gets significant contribution from jet-like correlations and decays that are not correlated with the EP and thus not v2. 

The smoking gun for an EP correlated v2 would be the near side ridge that extends over large pseudorapidity separations. Even in small collision systems like pPb, we can see this near side ridge at high pT, which needs some path length dependent effect to account for. For eg, you can see the near side ridge in the pPb measurements here 

https://journals.aps.org/prc/pdf/10.1103/PhysRevC.90.044906. We can quantify this in different ways, not necessarily v2, and can also attempt peripheral subtraction to quantify 

Perhaps something similar could be done with EPD, looking at pair correlations with jets in TPC. This would be integrated in pT for the associated particles, but can still be of use 

Best,

Sooraj 

On Tue, Aug 29, 2023 at 6:59 AM Ma, Rongrong via Star-hp-l <star-hp-l AT lists.bnl.gov> wrote:

Hello Rosi

I think we are converging. 

What do you mean by "he is entirely looking at nonflow"? If I understand correctly, Tristan is looking at correlations between mid-rapidity jets and forward-rapidity event plane by EPD. If the EPD event plane is NOT influenced by say away-side jets, then his v2 signal is one he is after. However, Shengli's results show that there is substantial non-flow contribution. So I think Tristan's result is a mixture of flow and non-flow signals, and the contribution of latter might even be dominating, which I think is important to point out. 

Thanks. 

Best

Rongrong

On Aug 29, 2023, at 9:39 AM, Rosi Reed <rosijreed AT lehigh.edu> wrote:

Hi Rongrong,

I agree that they are both looking at the correlation of the object of interest with the event plane.  How we construct v2 would be the same regardless, it's really just a Fourier coefficient.  I also agree that non-flow is the v2 signal originating from pQCD based geometry correlations - i.e. jet-initiated correlation.  So he is entirely looking at nonflow (well, at least at the higher pT).  We use the EPD in order to ensure that the presence of the jets don't bias the determination of the event plane - essentially one needs some sort of rapidity gap so that the jet is really being compared to the underlying geometry and not influenced by some jet-based autocorrelation.  (If you recall, Alice Ohlson did a study on this: https://arxiv.org/abs/1205.1172).  

When using the EPD to determined the event plane, there still will be a contribution at mid-rapidity from jets - it doesn't remove that, it only means you can measure the different quantities such that they aren't causally connected.

In any case, I think we agree that v2_obs works in both cases, and I do think understanding the results will be the work of the next few months.  (For instance, there's the question of what happens if we use the EPD for the centrality, since in a small system the mid-rapidity determination is more likely to be influenced by non-flow.)

Cheers,

Rosi

On Tue, Aug 29, 2023 at 9:19 AM Ma, Rongrong <marr AT bnl.gov> wrote:

Hello Rosi

Yes, Shengli and Tristan are looking at different physics, but they use the same observable, i.e. correlation of object of interest with event plane, not with jets. Shengli focuses on low pt, while Tristan focuses on high pt. I think non-flow refers to the v2 signal originating from jet-initiated correlation. What Tristan wants to measure is correlation of jet or high-pt track with event plane, not those generated by jets themselves. Isn't the reason to use EPD is to suppress non-flow? Otherwise, we can use TPC to reconstructed event plane, which has a higher resolution. Shengli's results show that non-flow contribution is still very significant at high pt even if one uses EPD, which could be due to away-side jets, I suppose. 

Best

Rongrong 

On Aug 29, 2023, at 9:08 AM, Rosi Reed <rosijreed AT lehigh.edu> wrote:

Hi Rongrong,

I'm a little confused, the greatest contribution to non-flow comes from jets, which is what Tristan is looking at.  So we wouldn't want to subtract non-flow from his results, basically the non-flow is his results.  There is perhaps a question of how much flow is contaminating his results at 5 GeV.  I think writing v_2^obs for the tracks is fine - the main point is that the the track analysis agrees with the jets, so there isn't something in the jet finding that's creating the high v2, and that the tracks agree with Shengli, so that Tristan's methodology is validated (as is Shengli's).  But they're looking at very different physics.

The large value of v2 for flow is also interesting in terms of what all this means.

Cheers,

Rosi

On Tue, Aug 29, 2023 at 8:58 AM Ma, Rongrong via Star-hp-l <star-hp-l AT lists.bnl.gov> wrote:

Hello Tristan

Thanks for looking into this quickly. 

Shengli's study shows that more than 50% of the v2 signal observed at 3-4 GeV/c is due to non-flow effect. After non-flow subtraction, the real v2 signal decreases with pT above 3 GeV/c, while your result shows a clear increasing trend. If I follow the trend of Shengli's result, the real v2 could be very small at high pT. So, I think we probably should at least label your plot as "v_2^obs", and state clearly that the non-flow contribution could be very large and has not been subtracted yet. This also raises the question of how much non-flow contamination is there in your Isobar results. 

Best

Rongrong

On Aug 29, 2023, at 12:03 AM, Tristan Protzman <tlp220 AT lehigh.edu> wrote:

Hi Rongrong,

I have managed to throw together a quick v2 on part of the data down to 0.5 GeV, see slide 16 of https://drupal.star.bnl.gov/STAR/system/files/preliminary_figure_request_OO_v2.pdf.  I believe the open blue squares in Shengli's results are the closest method to what I am using, where we see general agreement interpolating between the 20-40% and 40-60% with the non-subtracted EPD based method.  

Cheers,

Tristan 

On Aug 28, 2023, at 12:10 PM, Ma, Rongrong via Star-hp-l <star-hp-l AT lists.bnl.gov> wrote:

Hello Tristan

I would like to follow up on the comment I brought up during the PWG meeting about the single track v2, which if I remember correctly is about 0.15 at 4 GeV/c. On the other hand, slide 9 of https://drupal.star.bnl.gov/STAR/files/OO-ShengliHuang-QM2023_2.pdf shows that charged hadron v2 is about 0.075 at 2.5 GeV/c in 20-40% centrality. Have you looked at single track v2 down to say 2-3 GeV/c, and check if your results are consistent with Shengli's? Or at least we should put both results together and see if we have a consistent picture. 

Thanks. 

Best

Rongrong

On Aug 25, 2023, at 4:33 AM, webmaster--- via Star-hp-l <Star-hp-l AT lists.bnl.gov> wrote:

Dear Star-hp-l AT lists.bnl.gov members,

Tristan Protzman (tlp220 AT lehigh.edu) has submitted a material for a review,  
please have a look:
https://drupal.star.bnl.gov/STAR/node/64873

Deadline: 2023-09-03
---
If you have any problems with the review process, please contact  
webmaster AT www.star.bnl.gov
_______________________________________________
Star-hp-l mailing list
Star-hp-l AT lists.bnl.gov
https://lists.bnl.gov/mailman/listinfo/star-hp-l

_______________________________________________
Star-hp-l mailing list
Star-hp-l AT lists.bnl.gov
https://www.google.com/url?q=https://lists.bnl.gov/mailman/listinfo/star-hp-l&source=gmail-imap&ust=1693843848000000&usg=AOvVaw1sg__Mu9_CtkEOxQLmQVvV

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

-- 

Rosi Reed
Associate Professor, Physics Department
Lehigh University
(610)758-3907
16 Memorial Drive East Office 406
Bethlehem, PA 18015

she/her/hers

Click here to sign up for an appointment

-- 

Rosi Reed
Associate Professor, Physics Department
Lehigh University
(610)758-3907
16 Memorial Drive East Office 406
Bethlehem, PA 18015

she/her/hers

Click here to sign up for an appointment

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

--

Rosi Reed
Associate Professor, Physics Department
Lehigh University
(610)758-3907
16 Memorial Drive East Office 406
Bethlehem, PA 18015

she/her/hers

Click here to sign up for an appointment