Dear authors,

after reading the revised version of the manuscript and the answers to the reviewers I still have some questions.

A general question/concern for me is the suitability of X-band for the melt pond application. Could the results be used to argue for future X-band missions? Or specific modes? Or should space agencies better invest in C/L-band, perhaps conically scanning to avoid the effect of incidence angle dependency and reduce the wind speed dependency?

How was the significance calculated? I found no explanation of the method. Do you consider the independence of the data? Smoothing reduces the number of degrees of freedom and increases autocorrelation. How was this taken into consideration?

A specific reviewer comment was: "Give also pixel size of SAR images, only window size in pixels does not tell geometric size on the area.". Could you please add the geometrical sizes of the windows?

I wonder how the original SAR image looks like. Usually it is very instructive to have a look at the original texture and I suggest to show the unfiltered images in comparison to the smoothed versions.

"Scale sensitivity was tested by using a range of different smoothing window sizes (13 × 13 to 51 × 51 pixels) in the f M P estimation". This sentence comes without a result or conclusion.


The unit of f_MP is confusing, given as a value beteen 0 and 1 _and_ as a percentage. I would expect values between 0 and 100 for the percentage.

I was also confused about the definition of "empirical" and "global". A regression is an empirical technique. Perhaps use different terms?

Fig. 6 looks a bit strange. The rectangles seem to have a border color? There are obvious blocky/stripe artefacts from the smoothing. Have you tried to use different window functions? I assume that you used a simple boxcar weighting but I did not found the information explicitly in the manuscript.

Fig. 10 \sigma

Fig. 11: What is the meaning of negative melt pond fraction?

Eg. 16 use \sin

20 "Despite this, our findings demonstrate suggest new possibilities in melt pond fraction estimation from SAR, opening for expanded monitoring of melt ponds during melt season. " What are the new possibilities?

21 "In the next step, melt pond estimation from SAR may supplement surveillance from optical satellites, providing melt pond information to climate applications during cloudy conditions." Isn't this a bit too optimistic? If the method would be a feasible all weather approach (still to be proven), we would need long time series and large coverage for climate applications. The application needs at the same time high SAR resolution. This could be maybe in several years ahead but certainly not for the "next step"

519 "Scharien et al. (2014b) finds that the Bragg criterion is exceeded for melt ponds at wind speeds above U 10 =∼ 5 m/s in X-band, reducing the expected correlation between R V V /HH and f M P above this wind speed." Sharien et al. deal with C-band. How do you justify this statement.

623 "Such an algorithm could supplement optical methods, and be used as a tool in climate applications, both as input in climate models and in studies of melt pond evolution mechanisms." Not clear to me how to use the algorithm as "input" for climate models.

635 "In the future, use of X-band scenes can possibly increase the total amount of SAR data accessible for f M P retrieval, despite their limitations compared to C-band scenes." Do you argue for future new X-band SAR systems? Or would C-band (or L-band) be a better choice for the application. This is not clear.

644 "..and introduced uncertainties in our results, possibly causing artificially low correlation values" Why artificially? Fig. 2 suggests that the geo-location is not too bad?

Dear authors,

I am pleased to accept your manuscript and thank you for sufficiently addressing all comments and questions.

I have some additional minor corrections which you can introduce in the final version.

Thank you for submitting your work to The Cryosphere!

Sincerely,

Lars Kaleschke


Minor comments/suggestions:

Abstract: A smoothing window of 51 × 51 pixels
Pixel size is not yet introduced in the abstract.

Equations: maybe insert commas for transposed vectors?

275 sliding window: add boxcar

277 italics for unit

Consider to thank the reviewers in the acknowledgements.