|Please note that line numbers refer to the copy of the manuscript with differences marked up.|
The reviewers have mostly addressed my comments from the first review. The revised text includes a few typos and is clumsy in places.
Main points to be addressed prior to publication:
Can you be sure that the lack of small leads in the observations explains all of the difference between model and observations up to 40% lead fraction? It is also possible that the model is biased at lead fractions below 40%. I don’t think the statements at L230 that
• ‘agreement between model and reality is probably better than a first-order interpretation of figure 3 would suggest’; and
• ‘Perhaps a fairer comparison between the model simulation and observations would therefore consider only lead fractions larger than 40%’
are justified based on what has been presented in the manuscript.
In a similar vein, there is no justification presented for claiming that there is ‘noise’ in the observations (Fig. 3 caption).
The authors conclude that, ‘The model reproduces the PDF and scaling of observed lead fraction in the Central Arctic very well.’ (L383) This seems too strong, and I think it should be rephrased to something like ‘shows agreement with observations for lead fractions over 40%’
The following points aren’t necessary to address, but the authors may wish to consider:
Regarding my previous comment on the thin ice threshold-
It seems somehow artificial to me to compute the thin ice threshold based by tuning the model output to observations, when a model-observation comparison is being performed. In the end, it doesn’t really matter because the value is more or less the same as the value of 10 cm noted in the citation for the observational product, but I don’t really understand why the authors didn’t just use the 10 cm value.
The authors have added some broader motivation in the Introduction, but I think the manuscript would still benefit from some discussion of broader implications in the Conclusions/Summary (which is rather abrupt).
Line-by-line comments (should be addressed)
L21: remove ‘in the centre of the Arctic Basin’ after ‘Leads’
L40: When referring to ‘larger’ and ‘smaller’ leads, what is the approximate length scale?
L45: ‘model results are lacking in several aspects.’ Such as?
L58: ‘he’ should be ‘the’
L62: ‘the statistical properties of leads in large-scale sea-ice models have not yet been shown to be robustly reproduced’ I think this means to say that the models have not yet been shown to reproduce observations
L64: Add comma after deforms
L110: Typo, ‘comparable’
L115: Typo, ‘results’
L124: You might add that the cohesion parameter appears in the Mohr-Coulomb criterion
L126: Are the Bouillon & Rampal 2015a and 2015b references listed the right way round? I think they might be mixed up
L169: Mentions snapshots – but text above states that only daily means are used. Remove ‘snapshots’?
L178: Perhaps say, ‘as a a function of spatial scale’
L179: Remove ‘the’, before ‘log-log space’
L186: Suggest removing the first part of the sentence, and simply say ‘B(q) is the structure function, which ’…
L190: This paragraph is a bit clumsy, consider rewording
Fig. 1 caption: The last sentence should be in the main text rather than in the caption. How do you know that the observational product displays ‘noise’ rather than real leads? This needs to be justified.
L208: suggest rewording to ‘Figure 2 shows the PDF of observed and simulated lead fraction in the Central Arctic on a log-log scale. On this scale, both the observed and the simulated PDF’…
Fig. 3 caption: should include brief description of the error bars
L271: I don’t think differences of 20% and 30% should be described as ‘nearly the same’
L292: ‘This fractal character is inherited from the multifractal character of the ice deformation rates, as discussed below.’ This sentence is unnecessary and can be removed.
L369: replace ‘should otherwise’ with ‘would otherwise’