The reviewers have done a thorough job with this paper, and have extended considerable helpfulness in their responses. The authors are also to be commended on their diligent responses. The manuscript is much improved and is almost ready to be accepted.

There are a few remaining points, listed below, that I feel must be dealt with. If these are dealt with as I suggest then I will decide to publish. The citations I suggest would improve the paper, in my view, but, as Editor, I feel I cannot insist on them since I am a coauthor on these works.

I do not see the need to send this out to the reviewers again unless new points are raised.

Comments to address

1. In the edits in response to R1#2 dealing with the continuum assumption, you cite Overland et al 1998. This point is addressed head on in Feltham [2008], Ann Rev Fluid Mech, in supplemental Appendix C. It would be appropriate to cite this work.

2. In the response to R1#3 it is stated that the fracture angle provides “for the first time” a meaningful diagnostic that allows discrimination between different rheologies. I am not sure what is meant by “first time” in this context. Fracture angles and have been used to motivate sea ice rheology before [Erlingsson, 1988], and to constrain them [Wilchinsky and Feltham, J. Geophys Res, 2011; J. Phys. Oceanogr., 2012]. These latter two references probably do deserve to be cited, perhaps where you talk about initial conditions on page 14, where it is also relevant to cite/mention Wilchinsky, Feltham, and Hopkins [2011], which used a discrete element model to simulate ice fracture from an initial state of oriented cracks.

3. In response to R1#48, I agree with your point that the model may produce small floes that appear realistic but the manner of producing them may not be realistic. However your edit to the manuscript does not capture this point, where you say “creates smaller floes in a realistic manner”. I hope/assume this is just a subtlety of English that was not appreciated. A phrasing that more accurately captures your response to the reviewer is “creates smaller floes in a manner that appears realistic”.

4. In the edit in response to R2#11 you have stated “The scale invariance of the fracture process at the floe scale has not yet been shown, especially due to the lack of observations at both high spatial and temporal resolution.”
The clear implication of this is that (1) you believe such scale invariance is true; and (2) the only reason it has not been shown is because of lack of high resolution observations. Since neither you nor I have divine knowledge, and there are plenty of factors that the model, mathematical or conceptual, does not account for in relation to real sea ice, you should reword to a more modest, but defensible, statement. I suggest: “The scale invariance of the fracture process at the floe scale has not been shown. There is a lack of observations at both high spatial and temporal resolution that might show this.”