Data-Driven Calibration Algorithm and Pre-Launch Performance Simulations for the SWOT Mission

Round 1

Reviewer 1 Report

In this paper, authors describes in detail the main sources of error in the SWOT task and the data-driven approach to error correction. Overall, the article gives a clear roadmap of almost all the factors and technical solutions that need to be considered in data correction. The multiple scenarios offered in this paper have important implications in surface water and ocean topography and will provide exciting new views of the dynamic pressure field in the upper ocean with unprecedented spatial resolution and coverage. 

Although, it may not be appropriate to submit your paper as a "technical review" rather than as an "article", due to the subject matter of your paper may not being suitable for the current format. In the article, based on the simulation data generated by themselves (described at the beginning of line 159), the authors are supposed to have carried out simulations and validations, but the authors' work is mixed with the work and results of others, and a clearer delineation of paragraphs and sections may be needed to achieve a distinction between the authors' work and findings and the general understanding 

The use of high-resolution in situ and satellite data and models to explore the overlapping dynamics of small-scale ocean processes, including internal waves and tides, is now developing rapidly and will occupy much of the public's attention in the coming years. There are many semantic and formatting proper names in this article, including variable names, formats, etc. The language needs to be simpler, which makes it difficult for readers in the broader field to read. The clarity of some of the images in the article may not meet the requirements of the journal and authors are requested to provide a clear and readable version. And a more detailed description of the salient meaning of the SWOT task may need to be developed in section 1.1.

1. Line 59: Discuss whether "systematic errors" are equivalent to a linear superposition of other errors. 

2. Figure 3.: What is “media error”? 

3. Line 73: Explain “K-2 power law”. 

4. Line 280: "best case", is it optimal in every error component. 

5. Line 343: Explain “11-year solar cycle” more. 

6. Figure 10 a): Explain “unknown”. 

7. Line 620: Please give a clearer explanation or formula for the "Gaussian kernel-based interpolator". 

8. Line 705-706: Explain “benefits” more. 

9. Line 801: How to calculate the weights for "weighted least squares. 

10. Line 1491: Add the application of in situ observations for validation. 

11. Figure 15 c): Please explain the reasons for the large errors in Africa and the small errors in other continents. 

 

12. Line 1555: Please give a clearer conclusion as to what algorithm should be better chosen for different research needs.

Author Response

Response in attached docx document.

Author Response File: Author Response.docx

Reviewer 2 Report

A prerequisite of the exploitation of the satellite remote sensing products is the activities of calibration and validation, especially regarding the mission in preparation like Surface Water and Ocean Topography (SWOT).

In the submitted manuscript, the authors developed the data-driven calibration algorithms dedicated to the pre-launched SWOT mission, and provided  a summary of the overall error budget of this mission via performance simulations.

In general, the research topic of the manuscript is new and interesting, definitely timely. The goals formulated by the authors are clear. The methodology is mostly determined but appropriate and acceptable, suitable for achieving the set goals.

The explanation of the topic is logically well structured, sufficiently detailed, which requires quite special knowledge. References used are both quantitatively and qualitatively relevant and can be found in the given links or contact routes.

Hence, I suggest the paper could be published in RS after minor revisions.

I suggest that the authors should pay attention to the figures. For example, some of them (Fig. 6, Fig 15 for instance and others) are blurred. In addition, the units are missing for the colorbar in some figures (like Fig 10 a).

 

Author Response

Response in attached docx document.

Author Response File: Author Response.docx

Reviewer 3 Report

This paper presents a very detailed look at the systematic error sources for satellite (SWOT) generated topography. This is an area that I'm not very familiar with and so I will only be able to offer my impressions of reading the paper as an interested outsider or generally interested reader of the journal.

1) Initially I felt that the paper is overly long with 1640 lines of actual text, which does limit its overall appear the more generally interested readers. However, this paper is clearly framed as a detailed walk through all aspects including development and validation/testing of the algorithm. I'm honestly not sure how to restructure this better w/o a lot of work. Maybe delegate large parts of the really dense technical aspects to an appendix or supplemental section? Ironically, the use of the word chapters (l.224) does make this paper more sound like material from a book and maybe that would be a more productive approach. I would also suggest to rather use the work section instead of chapter.

2) Presentation of figures, tables and images is generally good enough for a review but for a final version the quality of some images/plots needs to be of better resolution. I suggest using vector-based graph formats (like svg) as much as possible. I especially like the use of colors and comic bubbles in diagrams like fig. 3 and 8!

3) Although limitations are appropriately mentioned in the text where applicable I would have liked to see a dedicated part for limitation at the end, maybe as part of the conclusion sections. This is likely the only section that somebody like me would fully read and I think clearly spelling out methodology limitations, together with what your research results enable the field to now do better (i.e. contributions) would be appreciated.  In this context I was also a bit surprised that there is no Discussion section.

4) English quality is varies throughout the paper (I assume based on which author wrote it) from good to very good! However, even the merely "good" are generally easy to understand. The only improvement I would suggest is to try to be as short and concise as possible. This, obviously, is generally good advice, but with this long a paper such shortening would really have a positive impact.

5) My only real gripe is with the reference resolution errors, such as figures references showing  Error! Reference source not found ! In addition you use things like section 0 (l. 85, 176, 185, etc.), which is clearly wrong.

 

Author Response

Response in attached docx document.

Author Response File: Author Response.docx

Reviewer 4 Report

Comments on ‘Data-driven calibration algorithm and pre-launch performance simulations for the SWOT mission’ by Dibarboure et al.

 

This work developed empirical calibration algorithms to reduce the systematic errors in the SWOT mission. Two complete algorithms sequences were presented for various purposes. One of the algorithms was implemented in the ground segment to control the main source of error of SWOT’s hydrology products., and another was developed to improve the accuracy of ocean products. In general, the manuscript is well written. Compared to previous studies, it provides summarized information about the errors in SWOT. The proposed algorithms maybe used in the future for operational SWOT data processing. It could be published with the improvements of the figure’s resolution.

Author Response

Response in attached docx document.

Author Response File: Author Response.docx