In this study, the authors have developed an algorithm to derive river ice phenology and thickness from satellite altimetry more specifically using the
backscatter coefficient. The paper intends to provide complementary methods to estimate ice phenology and thickness as these parameters are important for ice road safety and climate studies. The results were compared to in situ data. Though the algorithm to predict the date of ice onset and melt is premature, the overall result has potential and adds value. Before the paper potentially can be accepted I recommend some changes. Please, see the comments below
1) The methodology is only vaguely described, this needs to be improved. In section 4.1 a flow diagram could help to show the algorithm and then adding a more mathematical description would also help. Please, also provide more detail regarding the manual approached which seems to work better. In 4.2 It is not clear how it was decided that expression (1) was the best choice. If possible please add a figure that demonstrates how the relationship between the accumulated backscatter and the in situ ice thickness was established and what it looks like.
2) The method sections should be expanded with a section describing how the results are validated and all the additional analysis performed, which currently is described in the result section.
3) Currently, the result section is a mixture of results, methods, and an interpretation of the results. Though, a matter of style, this in my opinion makes the paper difficult to read and understand. In this case, it is more difficult to separate the actual result obtained from data, methods, and statistics from the author's interpretation. I, therefore, recommend rearranging this section, so the result section only objectively presents the results.
4) Section 3.2, I do not understand why this section is in the data section, since you do not use the waveform info in the algorithm, or am I missing something. I would not expect that you will find two peaks (related to the ice/air and ice/water surfaces) in a Jason waveform due to the bin distance of 46 cm and an expected ice thickness of approximately 1 m, at least not often. The two peaks are not clearly seen in figure 4. Maybe this section should be moved to a discussion section regarding improving the algorithm.
5) In the discussion section I lack some comments on the strengths and weaknesses of the method and its use. e.g. One limitation is that you need in situ data to establish the thickness relationship, but if the relationship can be applied for another river it will add value.
6) The paper contains many language errors and should be proofread by a native English speaker or a proofreading service.
Title: consider changing it to "River ice phenology and thickness from satellite altimetry. Potential for climate studies and ice bridge road operation
L 61-62: How does this work, maybe add a sentence.
L 63: "High resolution" please define.
L 114 "installation of .." -> "formation of "
L 122-123: Do not write "significant" unless you have performed a test
L 171. "In our previous studies (Kouraev et al., 2005; Zakharova et al., 2019, 2020), we noted that over" -> Previous studies (Kouraev et al., 2005; Zakharova et al., 2019, 2020), showed ..."
L 322 What do you mean by an "acceptable accuracy"?
L 335 "A significant difference" rephrase if you have not performed a test
L 350: "Error" is an imprecise formulation please clarify; RMSE, sd, ...
+ several languages errors not specified
Figure 1: Please add lines to indicates the branches of the Ob River. Color code the VS used as training and test.
Figure 3: Are the red lines based on in situ data? please explain in the figure text.
Figure 6: put the three figures on one line