Dear Jennifer, Cathy and David,

Overall, this paper is much improved upon the first version submitted. It is much clearer now which physical phenomena you are investigating. Thank you for cleaning up the terminology in the paper referring to “sudden changes”.

You have added much to the paper. The addition of the comparison between methods of calculating dispersion and deformation is very interesting, and could actually result in further discussion on later papers. Unfortunately the shear volume of description of results in the paper does distract from what your main points might be.

Your addition of the ratio of divergence to shear rates in response to reviewer 1’s comments is potentially interesting. I do have some specific comments, below, regarding this. In particular pay attention to the fact that you are only considering strain-rate. Without information about the forcing or material properties you can not comment on yield stress. I am pretty sure reviewer 1 understands this, I would just make it clear in the text that you have to make assumptions in your interpretation of this.


While the paper is much clearer, and I think I am understanding there are interesting findings, you need to make this much clearer in your text.

Please consider this comment:
“I also urge the authors to streamline the paper, making it more palatable and
understandable.”
On reading the paper I see that there is still a need to streamline the paper. My suggestion is that there are very interesting phenomena you show, yet this is not clear in your results and discussion sections. Perhaps choose the two most interesting phenomena and focus the writing on those. Or focus the writing on the phenomena that best make the point that your methodology is useful.

I agree with reviewer 1 that the dataset does provide information that could be important. This just needs to be made clear.

“I understand the importance of this topic
and find that it may be useful to understand the dynamical features of sea ice near the
shore on a daily time scale and on a sub-grid spatial scale. And that must be what the
author aims at in this paper. I see that the authors attempted to make the most of available
several buoy data to reveal them, and their efforts should be appreciated. However, I feel
the manuscript is a bit more descriptive and the conclusion does not necessarily seem
clear. “
I believe you can still hone the paper to write towards your conclusions more concisely. If you can do this it will be much easier to get a positive review.


I would like to send this paper out to review in a form that will showcase this unique study’s most interesting findings. There are some useful concepts in here that could improve upon previous methodologies investigating large scale ice mechanics, so I do feel there is potential for this paper to have impact.

Some specific suggestions would be to reconsider your motivation for the paper, you need to introduce why your questions are of interest. Consider your conclusions and the new methodology you have developed. Perhaps rethink how you introduce these. So you find dispersion and strain-rate varies with distance from coast. This is not a new finding, however the context you bring of considering the changing dispersion behavior between particular events related to changes in wind forcing is useful.



Specific Comments:

Line 10: a conjunction is missing between implications and ice hazard.

page 2 Line 23-24: Can you be a little more precise in your description of the force balance? I assume by sea ice motion you mean the inertial term?

page 3 line 9-10: This is the first place you have introduced elliptic and hyperbolic in this paper. It would be easier for a reader to follow if you first describe the mechanical properties of the ice pack, such as homogeneous and discrete, and then introduce the mathematical concepts through a one line discussion of the fact that the equations describing each behavior are elliptic or hyperbolic. I do see that you introduce this concept nicely near the bottom of page 3.

Page 7, line 20: consider grammer here. I think there is a missing “they are”

Page 12, line 10: you need to more explicitly state what this angle tells you about ice stress. It is only a signature of ice stress if you assume a constant flow rule and yield behavior. It would help a casual reader if you can explain ‘rheology’. You actually do not need to invoke any discussion of rheology to make the last sentence in this paragraph.


Rheology discussion on page 21: Strain-rate can not constrain ice stress. What you do show is that in convergence ice failure is followed by opening. Does theta change as well? This is a nice little diagnostic to identify when failure happens for particular modes of failure.

page 21 line 8: use greek symbol for theta.

page 23 Line 2: You need to be precise when you say rheological characterizations. You are not actually assessing strength, but you are assessing the kinematics and hence can identify particular modes of failure. The ratio of shear to divergence gives you a non-conclusive indication of where on a yield curve of arbitrary shape you lie (you can not determine the shape without a lot of data), and you have to assume no dilation in the flow rule or a constant predictable flow rule. I am not saying this is not a useful piece of evidence, it just has to be taken in context of the forcing.

I am not convinced you addressed the comment about accuracy of strain-rate reducing as the triangles become more elongated. It would help to show the math in the response to reviewer. Essentially as the triangles become skewed, their area decreases and therefore they do not represent the regional deformation. Reviewer 1’s point, which you reiterate is that the smaller deforming triangles have greater deformation rates, so it is difficult to compare to larger triangles and hard to interpret the time series of deformation events.

Figure 2: identify the bars in the bottom panel.
Figure 4a: labels are too small to read. This is true for several figures. A rule of thumb: make the labels the same scale as the text in the caption.
Figure 6: Why are the event numbers coloured black and red? In subsequent figures you place the arrows for event numbers, but no labels.
Figure 11: it is figure 6 that shows the same color bars, not 7.
Figure 15: This is rather confusing. What are the four rows in the figure? I am not sure I can pull out what you are trying to use this figure for from the paper.


Some more general comments:

While you have nicely addressed many of the reviewers comments, this paper has become exceptionally long. Do they need all the figures? Can figure 1a and 1b be combined into one panel for example. I am not sure you need Figure 8 for the discussion in the paper. It is acceptable to describe the deployment location and ice conditions in text as you did. I will admit, I got lost in the discussion of particular events in relation to the dispersion and deformation. I need help drawing between this overly detailed discussion and the conclusions.

It is much clearer in this revision why you are considering deformation events. These do present a way to study response of the ice pack to changes in forcing. It might actually be simpler to introduce your study in this more basic way, as I think this is what reviewer one and others actually see as the strength of splitting the time series into events.

Here are some highlights of the study in my mind: You have inter-compared three ways of characterizing dynamics of the ice pack. You are presenting a methodology to find changes in forcing in the dynamic response of the ice pack, and how this relates to dispersion characteristics. You have related changes in dispersion as differing response to wind forcing and the confinement on the ice. This is useful because it provides tools with which to identify timing of shock events, such as failure, and the transitions in dispersion related to these. Looking at figure 12, the large range in DKPs is indicative of arrays that have very small area. The time series are difficult to interpret because they are swamped by large values when the array is highly sheared. It is actually quite hard to determine how the direction change events line up in this time series with the one and two particle dispersion time series.


My main concern is that it is very hard for the reader to follow your findings, and understand what the particular observations of different ice drift direction changes are telling them. I would urge you to sharpen the manuscript further. Remove unnecessary descriptions and speak clearly to the finding that the enhanced shear near the coast happens during particular deformation events, that you here describe as direction changes. I want to see your work presented in the best light, so I hope you will revise the manuscript before it is sent out for reconsideration by the reviewers. I also would urge you to have someone less familiar with the paper proof read it.


With best regards,
Jenny

Dear Jennifer and Cathy,

Both reviewers were very happy with your improvements to the paper and recommend publishing. I am in agreement.

Please do proof read the paper throughly.

Page 3, Line 22: Consider not starting the sentance with 'And so,', perhaps just drop this. Not proper English grammar.

Page 3, Line 32: Do you mean "Two-particle dispersion also explains". It it more appropriate to say "describes"?.

Page 4, line 2: Ditto for "multi-particle analysis explains".

Page 8, do you mean to have a prime on y in the equation? Number equations.

Page 19, line 20: "shear-to-divergence"

Take this opportunity to make sure your main findings are clearly described in the conclusions.

Thank you, it has been a pleasure working with you,
Jenny