I like how this paper is shaping up! I think its impact could be improved by some choices of tone in a couple of places. Generally, I would prefer a tone where we are exploring the parameter space of classes of models and declaring that our measurements are consistent or inconsistent with them. I think it's fair to point out when models are/are not favored by our theorist friends, but I think measurements are the final arbiter. In places where our measurements are not more constraining than models fit to other data, I would like for us to say that we are consistent (but not more constraining) with those models. Measurements have to paint a consistent picture to reach a scientific consensus, and so our limits, even if not particularly constraining, are another data point that should be put out there.
Here are some examples: "We present a semi-analytic framework for exploring the new measurements of the 21 cm power spectrum from the Donald C. Backer Precision Array for Probing the Epoch of Reionization (PAPER) which were presented in our companion paper." -> I think you should lead with "We present constraints on high-redshift prescriptions for X-ray luminosity and star formation rates implied by using a semi-analytic framework for ..." i.e. lead with the science/constraints, not the framework. "We also calculate the expected heating of the IGM" -> rather than saying "expected", I would prefer referring to currently favored/disfavored models. I don't think we want to weigh in on (or imply ownership of) what the theoretical expectation for heating is, just how our measurements constrain (or don't constrain) parameterizations of heating. This comes in again in the last sentence: "We also calculate an IGM heating history extrapolating to the fainter galaxies necessary for reionization, and find that the spin temperature should be above the CMB temperature, and the 21 cm signal in emission at z = 8.4." -> In this case, I prefer the phrasing that we are consistent with (but do not further constrain) the Tspin>Tcmb case that heating models that include ... would imply. The results and discussion sections look pretty good from this standpoint, but the conclusion could be strengthened by saying that we are consistent with the Robertson et al. model, but not any more constraining than CMB + optical. I also think that, while the Robertson et al. model is "physically motivated", saying so detracts from the previous examination of parameter space that is motivated by suggestions of Fialkov et al. How necessary is that statement? Finally, the "above the limits set by PAPER" is tricky, because it's easy for the casual reader to get confused about the direction of constraints (upward for Tspin, downward for Delta^2). Maybe the phrase "within the limits" would be better? Good paper. I'm really excited for this to come out. Aaron On Sat, Feb 21, 2015 at 1:27 PM, Jonathan Pober <[email protected]> wrote: > Yes, thanks, James -- the caption for Figure 4 has a number of errors. It > should read: [left] f_X = 0.2 (pink), f_X = 2.0 (green); [middle] f_abs = > 0.2 (pink), f_abs = 0.04 (green); [right] alpha = -3.6 (pink), alpha = > -10.9 (green). I tried to make the pink curve be the same fiducial model > in all three panels, and clearly messed up the legend in the process. > > Your suggestion for Figure 7 is a good one. Eventually the power law star > formation rate model becomes a bad approximation (the SFRD turns over), but > it's still a fine model at z = 7. > > On Fri, Feb 20, 2015 at 7:57 PM, James Aguirre <[email protected]> > wrote: > >> Jonnie - >> >> Can you clarify figure 4? The legends all say f_x, which I assume is >> just not propagating f_x, f_abs, alpha into the plots. However, the f_x >> plot is showing more heating for f_x = 0.2 than 2, which seems unlikely. >> >> Since some of your normalization is done at z = 7, it might be nice to >> plot Figures 4 and 6 down to z=7, and indicate where the PAPER constraint >> and the Planck and Robertson 50% x_i redshifts are. >> >> >> On 2/20/2015 10:17 PM, Jonathan Pober wrote: >> >>> Hi everyone, >>> >>> As many of have seen, Zaki got his paper submitted to astro-ph today, >>> while I unfortunately had to delay mine after some last minute sanity >>> checking caught an error in my implementation of the X-ray heating >>> equation. After correcting it, the predicted IGM temperatures have all >>> gone up, making the PAPER constraints... less impressive. Still, it's a >>> big improvement over being wrong. >>> >>> I attach an updated (and maybe final?) draft for your perusal. I will >>> be at a workshop in Ohio through Wednesday, but if you can get comments to >>> me by Thursday, maybe I can still get this submitted at the end of next >>> week. >>> >>> One point that still may change is in Section 5.3, where I calculate the >>> heating expected from the Robertson et al. 2015 star formation >>> rate/reionization model. I have the parameters necessary to reproduce the >>> best fit model, but due to correlations in the quoted parameter errors, I >>> haven't been able to reproduce the marginalized 1 sigma uncertainties on >>> the model. By eye, I don't think I can rule out anything even at the >>> lowest allowed regimes of the model, but I've emailed Brant for help >>> calculating the errors, and if anything changes I will update. >>> >> >> >> -- >> James Aguirre >> >> Assistant Professor >> Department of Physics and Astronomy >> University of Pennsylvania >> 209 South 33rd Street >> Philadelphia, PA 19104 >> (215) 898-9596 (office) >> (215) 898-9646 (lab) >> http://www.sas.upenn.edu/~jaguirre/ >> >> > -- Aaron Parsons 510-406-4322 (cell) Campbell Hall 523, UCB
