An Adaptive Approach to Derive Sea Ice Draft from Upward-Looking Acoustic Doppler Current Profilers (ADCPs), Validated by Upward-Looking Sonar (ULS) Data

Round 1

Reviewer 1 Report

My comments and questions are presented in the attached file.

Comments for author File: Comments.pdf

Author Response

Please see the attachment for our point-by-point response to your comments.

Author Response File: Author Response.pdf

Reviewer 2 Report

The authors present an interesting method to use ADCP  information for other analysis than what was intended. In addition it may complement the limited data in the such a-difficult-to-access area. In their case, their ADCP dit not have
pressure sensor but they could compute ice draft. The result is very promising. However, various questions may need to be addressed:

 

This technique might work only at the place where the ADCP is located. How can it be applied to another area?

The calculation requires a precise knowledge of the density. How can the density at the beginning of the measurement be applied to all period when salinity will vary when ice is present?
Although "a more comprehensive sound speed correction... is not possible" (Line 436-437) an estimate of the potential error introduced by changing density would provide a valuable information.

The ADCP-measured values can either overestimate or underestimate the ULS-measured values. When do we know whether it over- or underestimates the real values?

How can one ensure that the ADCP and the ULS measure the same ice pack? An idea of the currents information (speed and direction) might provide some idea whether the moving ice would be first measured at one instrument then at the other.
The authors state that the "general ice conditions in those area ... are comparable on a daily scale". The authors may need to confirm that statement with some data or analysis of the ice drift.

Could the authors explain why the ADCP calculation "measured" ice draft in June-July while the supposedely more reliable ULS instrument did not measure any ice.

The scatterplot on Fig. 4 shows underestimation by the 1893 ADCP for 2014-15 and overestimation by the ADCP for 2013-14. For Taymyr, underestimation by the ADCP in 2013-14 seems to happen. These need to be investigated and explained a little
further so that the technique can be used at any location where only ADCP (without pressure sensor) data are available.

Although the boxplot in Fig. 7 shows statistics of almost similar mean, median, and quartiles, Fig. 6 shows that the percentage deviations are within a range of +-25% from the average deviation calculated by the author (that is 0 to 50% of deviation). This is relatively large to allow to state that the two data sets show the same information. With the provided statistic values, I would think the comparison is good not for the daily average for but rather long term average.

I am surprised by the comparison being better when assuming constant depth of the ADCP compared to time-varying depth. If that is the case, some other parameters are much more important in computing the distance between the ADCP head and the bottom of the ice than just tilting.

The authors aslo make a final statement about using remote sensing data (CryoSat-2, SMOS, ENVISAT) to validate their data. I suggest such validation is done in order to confirm the validity of their technique.

In conclusion, I think the work and result provided in this paper is very valuable and is of interest to numerous ADCP users and users with the necessity to get ice draft data. However, I would suggest the authors address the above comments, in particular the comparison with remote sensing data, should the paper be published in Remote Sensing Journal.

Author Response

Please see the attachment for our point-by-point response to your comments

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

The authors' response is detailed and informative. The changes made to the manuscript look balanced. Hope the authors will continue the method development and improvement in their future studies.

Author Response

Thank you very much for your review. We are happy that we have met your concerns from round one adequately and we will definitely work on further improving the method and finding additional data sets for further validation.

All the best,

the authors

Reviewer 2 Report

The authors addressed most of the comments with responses and/or changes which make sense. One point that they did not address as I would have expected is about comment #2: The authors did not make any changes in the text to address the uncertainties resulting from temperature and salinity; however, I would suggest the authors to just include a simple calculation of the uncertainties should temperature and/or salinity change. This will give some  information on what uncertainties to expect in the result related to the variation of temperature and salinity  happening between seasons.

Author Response

Dear Reviewer #2,

following your comment we have revised and edited section 4.3 to include a simple estimation of the effect of changing salinities on the travel time of the ADCPs acoustic signal. We focused on salinity since our presented sound speed correction takes the temporal changes in temperature into account already (see our answer to your comment 2 in round one of this review). We base the included estimates on measurements from that region that have been published recently (we have added the references to the respective publications in section 4.3 as well). We hope you agree that the changes conducted provide a better understanding of potential seasonal changes in sound speed and therefore ranges measured and give the reader this important additional information.

Please find our revised manuscript attached and the changes mentioned above between lines 430 and 466.

We would like to thank you again for reviewing our manuscript and hope that you are as satisfied with the revised manuscript as we are. 

Author Response File: Author Response.pdf