Hi, I'd like to solicit comments on the latest (last?) round with the dAu multiplicity paper. The latest version of the paper can be found, in various formats, at http://kunuc6.phsx.ukans.edu/~sanders/dndeta03.ps http://kunuc6.phsx.ukans.edu/~sanders/dndeta03.pdf http://kunuc6.phsx.ukans.edu/~sanders/dndeta03_2col.ps http://kunuc6.phsx.ukans.edu/~sanders/dndeta03_2col.pdf Briefly, the referee is unhappy with our need to use a multiplicity- derived centrality. It is correct that this definition of centrality will, in general, lead to a different event selection than one would have if Npart or the impact parameter could be measured. It is not clear to me from the referee's comment if the referee feels the paper should only be accepted if an Npart based centrality is presented--a requirement we will never be able to meet, or if the referee only wants us to address this issue forthright in the paper. My response attempts to do the latter. The changes are restricted to Fig. 2, an added paragraph describing the new histograms in Fig.2, and a change in the discussion of the NA35 comparison at mid-rapidity. (For BNL and paper committee collaborators, I have changed the text from what I had distributed earlier.) I realize you are all getting sick of this and are moving on to more interesting recent results, but I hope you can spend a few minutes looking this over and seeing if what I've done makes sense. Thanks, Steve ----Draft response to editor, followed by latest referee report.... Dear Dr. Wesselborg, We would like to resubmit our manuscript, reference LN9301, for your consideration. In the last round of comments by Referee B, the referee’s sole remaining objection appears based on the observation that the events selected in a light system on the basis of a multiplicity-derived centrality will not, in general, be the identical class of event that would have been selected had a measurement of impact parameter or the number of participant nuclei been possible. We agree with the observation and note that the BRAHMS collaboration made exactly this point in a preliminary discussion of our data back at the Fall 2003 DNP meeting (see http://www.sdcc.bnl.gov/brahms/presentations/2003/DNP2003_Hiro.ppt). However, in presenting our results we are careful to filter both the HIJING and AMPT model results through a realistic simulation of our experimental response before making comparisons to the data. Although a similar treatment was not done for the saturation model results since, in that case, we were working from a published figure rather than the output of an event generator, the model results are at such variance with the experimental distributions that the stated conclusions are firm. Referee B refers to an analysis presented by the PHOBOS collaboration on how a multiplicity-derived centrality based on particles emitted in the pseudorapidity window covered by the PHOBOS RING counters results in pseudorapidity density distributions that track well the HIJING model distributions as a function of centrality. While interesting, we note this is a model dependent claim and does not alter any of our observations or conclusions. Although we do not believe our previous submission was in any manner misleading, we do accept that a clear statement on how the empirical centrality selection based on multiplicity affects the results is useful. Consequently, we have added to Figure 2 HIJING calculated distributions where we assume a centrality based on impact parameter. The excellent agreement of HIJING with the experimental results lends credibility to these calculations. Supplying these distributions may also aid specialists who might want to compare the BRAHMS results to those obtained by other groups. We believe the discussion of the new curves fully addresses the final reservation stated by Referee B. We have also slightly revised our discussion comparing the BRAHMS mid-rapidity results with the lower energy data of the NA35 experiment to better recognize that differences in how centrality was determined in the two measurements can also influence the participant scaled ratio of pseudorapidity densities at the two energies. -----REFEREE REPORT---- Begin forwarded message: > From: Physical Review Letters <prl@ridge.aps.org> > Date: Mon Aug 16, 2004 4:52:06 PM America/Chicago > To: ssanders@ku.edu > Subject: Your_manuscript LN9301 Arsene > > Re: LN9301 > Centrality dependence of charged-particle pseudorapidity > distributions from d+Au collisions at $sqrt s sub {NN}$=200 GeV > by I. Arsene, I.G. Bearden, D. Beavis, C. Besliu, B. Budick, et al. > > Dr. S.J. Sanders > Physics and Astronomy > Malott Hall > 1251 Wescoe Hall Dr., Room 1082 > University of Kansas > Lawrence, KS 66045-7582 > > Dear Dr. Sanders, > > The above manuscript has been reviewed by one of our referees. > > The resulting report includes a critique which is sufficiently adverse > that we cannot accept your paper on the basis of material now at hand. > We enclose pertinent comments. > > If you feel that you can overcome or refute the criticism, you may > resubmit to Physical Review Letters. Please accompany any resubmittal > by a summary of the changes made, and a brief response to all > recommendations and criticisms. > > > Yours sincerely, > > Christopher Wesselborg > Senior Assistant Editor > Physical Review Letters > Email: prl@aps.org > Fax: 631-591-4141 > http://prl.aps.org/ > > P.S. Please note that we intend to conclude the editorial process > after the > next round of reviews. > > > ---------------------------------------------------------------------- > Third Report of Referee B -- LN9301/Arsene > ---------------------------------------------------------------------- > > I appreciate the additional efforts the BRAHMS collaboration has made > to > address the comments and requests made in the context of the original > draft. > > However, concerns which were discussed before have only become more > important now that there is additional peripheral data available. This > has made it easier to compare with other data both at lower energies > and > from other RHIC experiments. Unfortunately, these comparisons have made > it difficult to recommend releasing this result as it is, since it > exposes a substantial effect of the bias induced by the centrality > selection on the final physics result. In particular, this bias > affects > the shape of dN/deta, and not just the overall scale. > > My concerns are based on discussions shown in public talks from Quark > Matter 2004. In this conference, PHOBOS showed preliminary results > which were explicitly discussed by Miklos Gyulassy in a summary talk, > which referenced earlier data and models on this subject. It is > precisely this discussion which is in conflict with the BRAHMS data. > > Moreover, PHOBOS had additional slides which showed precisely the > effect > brought up recently in my first round of comments. In Rachid Nouicer's > talk, > > http://www-rnc.lbl.gov/qm2004/talks/parallel/Thursday01/RNouicer.pdf > > The ratio of reconstructed HIJING dN/deta with centrality selection > using (|eta|<3) and (3<|eta|<5.4) is shown. It is observed that there > is a substantial enhancement (about 25%) in the midrapidity density > when > using the former relative to the latter. In other words, the > centrality > bias is not just to the estimated number of participants, but to the > shape of the dN/deta distributions. > > The effect of this cut is seen most clearly in the comparison of the > BRAHMS data on R_0-30 (which is the ratio of the multiplicity in the > 0-30% centrality class over that measured in the 60-80% class) with > similar PHOBOS data, that Miklos Gyulassy showed in his QM2004 > concluding talk: > > http://www-rnc.lbl.gov/qm2004/talks/plenary/05Friday/MGyulassy.pdf > > The PHOBOS data shows a clear "triangle" shape in the ratio of central > d+A over p+p data (which I assume is comparable to the BRAHMS 60-80%). > This "triangle" shape had been seen before in comparisons of p+A and > p+p > data at lower energy and is clearly visible in the comparison of > central > d+Au PHOBOS data with p+p data. The fact that the BRAHMS data has a > pronounced peak for more central events is definitely some cause for > concern. In my opinion, it is not relevant to possibly argue that > HIJING > gets the same shape when applying the same cuts. This is probably for > the simple reason that HIJING has roughly the right pseudorapidity > similar correlations of multiplicity with centrality. In other words, > it does not represent the ratio one would get if one could in fact cut > on Npart directly. The PHOBOS discussion addresses this issue head-on > and a similar discussion needs to be part of this paper for me to > recommend publication. > > Thus, in its current form, I would not recommend publication of this > manuscript in Physical Review Letters. At the very least, this paper > would have to include studies which explain the effect of the > centrality > method (or various methods) on the shape of these ratios. Then they > must either be corrected, or sufficient caveats would have to be > included that would prevent others from misinterpreting these results > as > presenting an unbiased evolution of the centrality dependence. This, > however, would sufficiently weaken its physics importance for the > community and thus perhaps preclude publication in PRL. _______________________________________________ Brahms-l mailing list Brahms-l@lists.bnl.gov http://lists.bnl.gov/mailman/listinfo/brahms-lReceived on Sun Aug 29 16:56:43 2004
This archive was generated by hypermail 2.1.8 : Sun Aug 29 2004 - 16:57:07 EDT