From Marek.Szuba@cern.ch Sun Jan 13 20:20:12 2013 Date: Sun, 13 Jan 2013 20:19:58 +0100 From: Marek Szuba To: Bialkowska Helena , Gazdzicki Marek , Laszlo Andras , Seyboth Peter Subject: Fw: Azimuthal-correlation paper - comments on referee responses Dear everyone, As it seems that once again Peter has not received my e-mail, I am resending it to see if it comes through this time. Sorry if you end up getting it the second time as a result. Also, it would be great if you let me know that you've got this message. Thanks in advance! Begin forwarded message: Date: Thu, 3 Jan 2013 17:32:51 +0100 From: Marek Szuba To: Bialkowska Helena , Gazdzicki Marek , Laszlo Andras , Seyboth Peter Subject: Azimuthal-correlation paper - comments on referee responses Dear Colleagues, At last, please find below my comments on the CERN review of my paper draft: > Section I > 1. The abstract is somewhat vague and qualitative. The authors should > quote the "SPS energy range" (i.e. sqrt(s)=6.3-17.3 GeV) and the > "high-pT" values of the trigger particles used (pT=2.5-4 GeV/c). I have just quoted the exact energy range used. As for pT, I would like to discuss this first - while the 2.5-4 GeV range is indeed the main one we use for trigger particles, it is not the only one. This might cause some confusion, then again other ranges are used for cross-check purposes only and mostly go *below* this one. > 2. Jet modifications in heavy-ions are discussed, providing a citation > [1] which refers to p-p collisions. This citation is meant to introduce the subject of jets themselves, not their possible modification in heavy-ion collisions. That said, I suppose we could add another, heavy ion-specific citation there, possibly after the next sentence... The question is, which one - there are after all quite a few different postulated effects. > 3. "two-particle azimuthal correlations" and "correlations in the > plane perpendicular to the beam" are equated. I must say I am not quite sure why this is a problem. May the reviewer imply the latter term refers to particles travelling in said plane, i.e. possessing no longitudinal momentum? > 4. Terms like "soft" and "hard" are not defined in terms of a scale > (pT value). This can be added, albeit it will not be a definition per se - all that can be said about soft and hard processes in terms of pT is that products of hard scattering have got pT of at least 2-2.5 GeV/c whereas particles with pT above 4-4.5 GeV/c pretty much cannot originate from soft scatterings. > 5. It is claimed that hard-scattering amounts to 2% of sigma_inel "at > the SPS", whereas (i) this percentage will change rapidly from > sqrt(s)=6 to 17 GeV, (ii) it is not said how this percentage is > obtained, (iii) the authors likely refer to the cross section of > particles above pT~2 GeV/c but they seem to miss the fact that hard > scatterings among partons with pT>2 GeV/c will produce (fragmentation) > hadrons of potentially much lower momenta. Given I recall questions about this appearing in earlier discussions and that the source material is rather vague on the details behind this number, I feel tempted to replace this with a more qualitative statement ("even at the highest SPS energies, hard scattering is rare"). > 6. Statements such as "the nature of the transition between soft and > hard scatterings" are vague (which "transition" ?). This indeed might benefit from being reformulated, however it would be good to do so without using too much QCD terminology which would require explaining. I'll think about it, at the same time I would appreciate suggestions. > Section II > 7. It is interesting to know that NA49 has measured high-pT hadrons > from p-p collisions at sqrt(s) =17.3 GeV. The lack of an > experimentally measured baseline p-p reference to interpret high-pT > hadron production in heavy-ion collisions at the SPS has been > considered as a big handicap by the > nucleus-nucleus physics community since many years (see e.g. > http://arxiv.org/abs/nucl-ex/0403055). Yet, the authors do not provide > any details/references on the p-p measurement. Was this a run of > proton on liquid-Hydrogen (gas?) target? What was the integrated > luminosity? Why aren't the measured pT spectra compared to PYTHIA > and/or UrQMD? That would be certainly an interesting measurement. All this information can of course be provided, at the same time however we should take care not to exceed the scope of this work and wander into the territory of single-particle measurements made by Andras. How far do you think we should go here? In my opinion, comparison of pT spectra with models goes too far (as we don't show them for Pb+Pb either) but other questions can be answered in text. > Section III > 8. Nothing is said about the pT resolution of the measurement and the > associated effects (if any) due to possible bin migrations among the > different pT ranges defined (in the data and MC). Indeed. This should be discussed. My first impression here is that bin migration can safely be neglected. On one hand, assuming our pT resolution is in the same order as total-momentum resolution, in the region where we switch from trigger to associate particles in standard binning (i.e. around 2.5 GeV/c) the resolution effect would be mere 600 keV/c. On the other, we operate in the region where both soft and hard processes are at play so even with ideal pT resolution we couldn't completely remove soft particles from the trigger bin. > 9. I am confused by the plots in Fig. 1: (i) The 6.3 GeV data reaches > a higher pT than the 17.3 GeV measurements, and (ii) the 6.3 GeV > measurement show particles above the kinematical limit (pT=3.15 GeV/c > at midrapidity). How can this be? This... is a potentially big problem. These particles are almost certainly fake - and while in the past we decided to let them be on the grounds that pairs in which only one particle is fake, even if it is a trigger, do not significantly affect the shape of the correlation function, it certainly looks bad on acceptance plots. Therefore, unless we provide a good explanation for why such particles can be left in the analysis it looks like it's back to analysis level in order to determine how to get rid of them... Either way, I suggest we discuss this matter on Vidyo soon. > 10. In Line:160 it's claimed that the 2-source model is used only for > non-central collisions, whereas the authors show results for the 0-5% > most central collisions. Corrected the relevant sentence to "We employ this model in the present analysis only for the investigation of centrality dependence of azimuthal correlations; elsewhere, where only the most central collisions are studied, we use raw correlation functions instead." > Section IV > 11. The technical explanation of Lines:233-238 might be merged into > Lines:155-159. Done. > 12. The role of the possible different hadron composition (e.g. pions > vs kaons, protons) at RHIC and LHC energies and its influence (if any) > on the different measured azimuthal correlations is not discussed. Azimuthal correlations of identified particles are a rather different cup of tea from inclusive ones and I am not sure whether we want to get into this, especially given my attempts have shown we cannot really do this in NA49 due to limited statistics. Still, maybe it would make sense to meet the reviewer halfway and just provide some references to RHIC and LHC studies on the subject? > Section V > 13. It is argued that UrQMD reproduces well many processes at SPS > energies, but no control plot is shown that shows the level of > agreement of the model to the basic kinematical distributions of the > measured tracks: at least the comparison of detector-level > (uncorrected) data and MC pT,eta,phi distributions of the measured > single hadrons for the generated collision-systems, centralities, and > sqrt(s)'s should be mentioned. Here I would say again what I said commenting on point 7 - while we could provide such data, it feels like adding too much single-particle results to the manuscript. Is there by any chance some comprehensive comparison of NA49 spectra to UrQMD results we could simply cite here instead? > Section VI > 14. In the introduction the authors discuss the possible role of QGP > effects (parton energy loss), whereas is has just been claimed in the > previous section that UrQMD does not include such effects and can > reproduce well the measured two-particle azimuthal distributions. This > fact is only mentioned > > again later in Section IV.C. Slightly tricky. Initially I thought that maybe we could turn these statements around a bit, i.e. instead of talking about disagreement with QGP effects talk about agreement with triangular flow - but that would require the manuscript to actually contain any studies related to triangular flow, which it doesn't because I've performed no such analysis. Should we just say that disagreement with QGP effects has been expected but further studies are required to determine agreement with the triangular-flow hypothesis? > 15. Fig. 11: What are the systems plotted ? Does "heavy-ions" in the > caption mean Pb-Pb at the SPS (what about Si-Si?) and Au-Au at RHIC? Added specific species names in the figure captions. > 16. The role of the possible different hadron composition (e.g. pions > vs. kaons, protons) at RHIC and LHC energies and its influence (if > any) on the different measured azimuthal correlations might be > discussed. See my comment on point 12. Cheers, -- MS -- MS