The UCD Nuclear Physics Group Radial Flow Paper Response to Comments

Comments from Chris Pinkenberg

Your introduction is more or less focussed only on qgp (at least it's mentioned 5 times in a single paragraph), but it doesn't make a connection between radial flow and qgp detection. The only connection brought up is via the EOS but at the end you mention the possibility that radial flow is just a temperature effect. It would be nice to bring your introduction more in tune with the conclusion. Radial flow is not really my area of expertise but isn't there something of a limiting Hagedorn temperature of 120 MeV which we hit in our beam energy range?
JED: We have met this concern to a reasonable extent. Perhaps others can improve on this.

This is just my opinion and I leave it completely up to any authors discretion, but I think in general E895 has a lot of really interesting physics to offer and we shouldn't restrict restrict ourselves to the message "we looked at that aspect and again we didn't find qgp" (if we find it that's a different story...)
JED: This is a reasonable concern. We feel that the QGP should be at least mentioned but not pushed. It has long been expected, and found by CERN, that various experiments must be combined for effective conclusions about the QGP, so not finding a smoking gun is not soo bad.

For the technical part: We didn't measure the kinetic energy in the middle of the target (that would be a real technical challenge), we measured the beam momentum in the TPC and after long discussions then came up with an agreement what our beam energy was. As far as I recall it was a mixture of energy loss calculations, measured beam momentum and the requirement that Hengs <px> should cross zero at midrapidity.
JED: Done

I am puzzled about the statement, that the pi+ contamination for the protons decreases with beam energy. Is that a typo or did I miss something?
JED: Yes a typo.

The TPC measures zero pt particles very well (actually better than pt=100-200MeV/c protons), the "almost zero transverse momentum" is not correct.
JED: We removed this.

Your fits and therefore the results are completely dominated by the He3 (actually at the 2/4 AGeV the pions don't really support a common origin) Just out of curiosity (if the effort is too large forget that), does the result depend on the pions at all?
JED: Yes He3 is very important and we've said more about it. The main purpose of including the pions is to show that another species (pions) is not incompatible with the other two.
DAC: I would not say that the fits are dominated by the helions. It is really the protons that dominate the fits. The helions and pions serve to stablize the results. Check out the pages were Mike takes a look at protons only, and then at protons and helions, without pions. You are correct, the pions from 2 and 4 AGeV do not seem to fit with the other species. This is probably telling us something about pion production at these energies.

How clean is the pid for He3? There is nothing mentioned in the manuscript. Do you assume anything in the z=2 band is He3, or is there some separation between the helions?
JED: Yes the entire z=2 band is used. However the Wang reference for EOS shows that those data are compatible with coalescence and the He4 are argueably smaller by more than an order of magnitude than the He3.
DAC: We have looked into where the He4 might contaminate the He3

You give an estimate on the corrections for the spectra, but how does that change the beta/T? Your error bars are very small when compared to EOS which supposedly had much less of a problem with pid.
JED: This is fairly well illuminated by what Mike is putting on the web. And the meaning of the error bars is now defined.

You don't have too many He3 at the high energies, do you have an estimate of the ghost track contribution(like cutting outside the pid band for He3) and its effects (mt of these ghost tracks?). The embedding of particles is our method of choice but it doesn't say anything about ghost tracks.
JED: We have seen the phrase "ghost tracks" mentioned, but are not sure what you mean. We're guessing that you mean tracks whose dE/dx are in lala land, in which case that has not been studied.

Comments From Paul Chung

In the paper you say that the pi+ contributions were removed from the proton spectra by estimating the pi+ contributions by extrapolating the pi+/pi- ratio. Is this the same as the UCD probabilistic PID ? My understanding was that Jenn Klay had put a lot of work into generating the probabilistic PID with the specific intent of deconvoluting the spectra in the region of overlap. So, was the UCD probabilistic PID used to deconvolute the proton/pi+ spectra ?
JED: The PID was not the standard UCD probabilistic PID on the web. Instead the more complex method you allude to was used. An essential feature of this is to make many narrow rapidity cuts before proceeding with the more conventional parts. This is a lengthy process, as you can imagine, and it has not been automated to the point that it can readily go on the web as a push button operation.

The paper says that the spectra were corrected for efficiency losses using the slow simulator. Can you show the efficiency vs (mt-m0) for protons, pi- and He for 6GeV ?
JED: Yes, this will be put on the web.

What cut was applied to generate the spectra: rapidity cut or angular cut ? Does a particular midrapidity cut, let's say |(y-ycm)/ycm| < 0.1 , correspond exactly to a certain angular cut ?
JED: Of course eta does not equal rapidity, but that seems not to be your question. The specific angle (eta) cut approach was used in the EOS paper on radial flow, and it was also used here. We did not see that it was a critical point.

The paper says that the 2GeV He3 spectra suffers from acceptance problems at high mt. But the high mt spectra seems to be higher than the tail of the fit. An acceptance loss at high mt ( due to particles curving out of the TPC ) would lead to a deficiency in the number of particles in that region. So, I wonder if your explanation is not in contradiction to the observation.
JED: I touched on that above, and you will see it in the web material to come.

Comments By Ajit

If the slow simulator can be trusted to obtain the efficiency, it should yield a proper description of the acceptance hole in px-py space observed in the data. My understanding is that the SS fails to do the needful in this regard.
JED: Recall that the TPC matrix T, above, is only for a region of eta very close to midrapidity. This already deemphasizes some of the lowest pt particles from the data, as well as from Siemens Rasmussen theory. However, the embedding should address this properly. We have not directly studied this hole however.

The change of slope in 2GeV He3 cannot be ascribed to acceptance loss. Can you display He3 spectra up to larger mt ?
JED: I think this is addressed in the paper and the web. If not, bring it up again.

Are there any model calulations to show effect of QGP formation on observed radial flow beta and T i.e. what is the level of signal we are looking for ?
JED: We are not aware of such calculations. I guess we'll have to leave this to the theorists.

Text : 1) Page 2 par. 3 : stating active volume of TPC in pixels seems odd. 2) Page 2 par. 3 "with 0.75 Tesla..." 3) Better if eqn. on pg 3 is in terms of plotted quantity.
JED: The first two are done. The third seems OK as is. That is, Eq (1) is in the form given by the authors. But the manuscript says that for midrapidity E = mt, thus making the connection with the label of the ordinate of Fig. 1.

---