Dear Chris and Co-authors,
Thank you for your detailed response to the reviewers.

Please could you consider the following points.

Line 30 page 4: "where shear stress determines sea ice shape". What shape do you refer to? floe shape? the shape of the ice between leads? Perhaps clarify.

Line 23 page 7: "aggrement" Spelling.

Rather than ask the reader to find and interpret the code in the supplimental section, can you provide a brief description of the operations you performed in the paper, referencing the supplimental section.

End of page 9: "We interpret this seasonal cycle in size over time as due to the formation of large first-year ice pans in October which are later fractured into smaller floes throughout the winter months"
- Personally I don't like speculation like this. In Autumn storm activity is high, pans of ice can get broken up. I get what you are implying here, that once sheet of ice form and the ocean is calm large pans with grow. Intuitively we would expect leads to crack the entire ice pack once it becomes consolidated to the coast and able to propagate stresses long distances, this is in December/January according to stress measurements taken by Jackie Richter-Menge in the Beaufort in 2001/2002 (Richter-Menge et al. 2003). So there may be some truth in what you say, but it surprises me. It would be best to clarify that further observations are required to understand your results.

line 9 page 14: "With colored cells those where" may read easier as "colored cells are those where"

line 20, page 14. Consider splitting into two sentences after "disagree". The hyphen confused me on the first read.

Section 5: Do you consider the larger cut off flow size of 900m that you implimented earlier? This is getting to a scale with relatively few floes in the MIZ though.
I agree with reviewer 2 that this section seams a little out of place in the paper. However you make a reasonable point that the method was developed so models and observations could be compared. Perhaps you can clarify this at the top of the section. It is fair to say that floe size distributions in models is a newly developing field and your method might facilitate testing models following this example. With the caviate of the minimum detectable floe sizing being rather high and further work required to understand seasonality and regional variability.

page 18 around line 5: Can you comment on whether your choice of geographic regions is appropriate to isolate ice regimes that may have power law scaling for floe size? You do show that there is no global power law scaling for the satellite product, and I feel you should be careful to not draw conclusions about local dynamics from this. This does not negate your point that models acounting for floe-scale processes are needed.

page 18 line 24: "This implies that the principles of isotropy, homogeneity and stationarity of the FCD, required to produce such a distribution, are invoked on the length scale of the CICE model grid and time scale of a month."
- define this length scale, because it is not set my CICE itself but by the modellor. Also, the model referenced does not need to be CICE, it can be any model that has a continuum or homogenous element assumption.

Please make sure you are consistent in your symbols throughout the manuscript, including appendicies. For example, did you change beta consistently throughout?

page 22, line 4: Check the brackets.

Equation C10 needs a comma behind it rather than a period.

Sincerely,
Jenny

Reference
Richter-Menge, J. A., Elder, B., Claffey, K., Overland, J., & Salo, S. (2002, December). In situ sea ice stresses in the western Arctic during the winter of 2001–2002. In Proceedings of the 16th IAHR International Symposium on Ice (Vol. 2, pp. 423-430). Dunedin, New Zealand: Univ. of Otago.

Dear Chris and co-authors,
Thank you for your response to my minor points.
Jenny