Reviewer #1

[ncclementi]

Reviewer 1 comments

This is a nicely written report that provides an important demonstration of high-quality V&V and R&R. The effort of the group to provide all materials and act as a positive example for the computational community is commendable. The report also benefits from a deep and informative background section; a positive aspect and a good fit for a journal such as Phil. Transactions A.

Minor comments:

• Pg 2 line 26 – trecode should be treecode
• Pg 2 line 36 – mode should be modes
• Pg 8 line 39 – there is repetition of ‘a a’
• Pg 9, and elsewhere, when discussing the features of the figures, in the case of Pg 9 it is figure 4, it would be useful to be more descriptive, for instance “One can see that the second peak in the E||b plot at 10.6 um is not present for the green curve”
• In the Figure 4 caption it would be useful to briefly explain the legend terms: trimesh, uniform and uni + round.
• Pg 12 line 27 – change “per Angstrom-squared” to use the Angstrom symbol and rewrite?
• Pg 19 line 19 – given he relationship

[ncclementi]

Modifications c5095c3

In abstract:
trecode -> treecode
position of some mode -> position of some modes

In results:
a a -> a

In discussion:
given he -> given the

[ncclementi]

Pg 9, and elsewhere, when discussing the features of the figures, in the case of Pg 9 it is figure 4, it would be useful to be more descriptive, for instance “One can see that the second peak in the E||b plot at 10.6 um is not present for the green curve"

I think the reviewer has a point, we should be more descriptive, but we should discuss the wording since we need to include what Rockstuhl pointed out in his email (see issue #3 )
(made a partial change which does not include Rockstuhl explanation yet, changes included in rev_one)

In the Figure 4 caption it would be useful to briefly explain the legend terms: trimesh, uniform and uni + round.

I think this add information and we can add it to the description of the figure (Have a possible incorporation in branch rev_one)
Add labels to the legend of Fig 1 as well, included in commit b06052

Pg 12 line 27 – change “per Angstrom-squared” to use the Angstrom symbol and rewrite?

If we change it for the symbol, we should write triangles/Ang^2, I don't mind changing it but I remember that Lorena explicitly said she liked the written version better, so let's discuss that.
I read the manuscript again, and we used before triangles per (Angstrom symbol) squared, so I modified the text to be consistent across the manuscript. 6f5dec