Study on the Erosion and Deposition Changes of Tidal Flat in Jiangsu Province Using ICESat-2 and Sentinel-2 Data

Round 1

Reviewer 1 Report

 

1. Authors should conisder adding  Sentinel 2 to the list of keywords

2. The methodology lacks fundamental details to allow a replication of the experiments, such is the case of software, reference to used ML packages, the 908 special geodesy surveys.

 

Specific comments

Line 29: The term "subject" should read "subjected"

Line 30: The sentence ending with "...coastal processes" is so general that it doesn't need a citation.

Line 32: I would rather start the sentence with: "The Jiangsu Province in China HOSTS the most...". Please revise accordingly.

Lines 33-35: The association of longest shoreline and largest tidal flat is not straight-forward, as suggested here and needs to be corrected. 

Line 36: The verb "lead" should refer to the past. Consider using "...having led to..." or "...which has led to...".

Lines 37-38: There is an attempt to add "sea level rise" to the list of natural calamities. That is unfounded since 'calamity' often refers to a sudden event and not a rather slow process such as SLR. 

Lines 38-43: The sentence in Line 38 starting with "In recent years..." until end-of-page needs to be revised. I was unable to understand the intended message.

Lines 148-149: What was he resolution before it was adjusted to 10m, and please elaborate further on the adjustment itself (was it an interpolation, extrapolation, or some filtering of any sort?). How did it affect the data quality?

Lines 155-160: This section states the improvements of ICESat-2 in a repetitive manner. Please revise and avoid repeating the ideas.

Line 169: How does the national 1985 datum compares to the global 1984 datum? What do we gain using the 1985 datum?

Line 171: Does ALT08 and ATL08 refer to the same product? Same for ALT03 and ATL03 terminology that are used in the manuscript. Please revise accordingly.

Line 176: What are the 908 special surveys referred to in the text?

Lines 243-244: That is okay, but then, isn't the choice of a particular model or combination of models something to explore? The higher accuracy of outputs is not only because models A+B were chosen instead of A+C? The model strengths and weaknesses (in terms of conceptual and mathematical formulation) need to be evaluated to avoid ambiguity in the choice of model combinations. 

Lines 194-227: The section is part of the "general methodology", and present sa very modest summary of the tools employed. However, it lacks a reference to the literature where these tools are actually introduced and/or evaluated. It seems like the different tools were used as 'black boxes' only, which is not a bad thing if authors acknowledge that.

Line 255: "As shown in Figure 4"? The so-called Fig4 does not add value to the information given in the text, it gives no further clarification to the message being conveyed to potential  readers. And just like the "general methodology" given in lines 194-227 this Fig4 could be moved to a section that contains 'additional information' or 'annexes', depending on what's allowed by the journal.

Line 270: In fact, 'Table 2' presents stat parameters (RMSE, MAE, R, etc) used to assess the validation of model accuracy, but not the validation nor the accuracy themselves (as stated in the table's legend). Authors need to elaborate more on why R-squared is used to define the 'best-performing' model instead of other available parameters. Since the ultimate goal is the long-term shoreline monitoring, should not the temporal stability be as important as having a small MAE or RMSE? Stability is not assessed/discussed in the manuscript. The choice of what is considered valid or not should include a compromise with the intended use of the overall approach.

Lines 282-285: The information in this section seems to be repeated elsewhere, or it should have been moved to the Methodogy chapter, and simply present the results here. 

Lines 288-289: Figure 4 refers to a step of 0.001, and it suggests that 1000 results were obtained for each scene. The text in line 248 states otherwise: step-size of 0.0001 and 10000 results. Table 2 presents numbers with 0.001 precision. Lines 276 and 279 present RMSE values with 0.01 and 0.001 precisions, respetively. In lines 288 and 289 the precision is also 0.001. This calls for an harmonization of the way values of the same parameter are presented. Also, the values in line 288 suggest that satellite elevation is estimated with 0.001 precision, which is a rather strong claim. Moreover, Figure 6 doesn't allow to read such high precision values for elevation.

Line 297: Can this challenge be further portrayed in the profiles of Figure 6? Is it possible to indicate what parts of the profile are affected or not by the impaired penetetration of laser signals?  

Line 305 Figure(6): Please use the same precision for all RMSE values. Explain in the legend what is the point where Dist=0m in each plot? Add a mention to the difference in X-axis limit. Is it fair to assume that Y=0m is the Mean Sea Level in the 1985 datum?

Line 314: The expression "can be attributed" should be replaced to "could be attributed" since it is a mere speculation. A proper investigation should be made to reach conclusive results. 

Line 318: Instead of "-0.5 to 2.45m" use "-0.50 to 2.45m" for consistency.

Lines 349-356: It should be noted that changes in elevation could be attributed to processes other than erosion and deposition, for instance land subsidence. 

Line 394: Sediment supply dominates over what? Would it be it be sediment removal?

Line 396: It is necessary to include references to these earlier research whose findings are being used. This applies to the entire chapter 4 of the manuscript.

Line 401 and 420: the term "environment is" should instead read "conditions are".

Line 403: "topography" is not a valid subject for the action of blocking sediment transport> Instead consider "sand bars" or any other feature that arises from sediment accretion. Please, revise the terminology employed.

Line 416-417: Groins, breakwater etc cannot be considered as "contributing factors" because they were specifically, purposedely installed to modify the tidal flat evolution. 

Line 419: Kindly note that "hydrodynamic data" was not presented or discussed here. So, all references to hydrodynamic contribution are a mere inference if not speculation. It  would look good to see some of the hydrodynamics disscued in this chapter, even if only references to earlier studies were used.

Lines 425-426: Again, one could say that the results presented here only show erosion at the shoreline, and this erosion can be used to make inferences about the sediment supply, but not the other way around, as the authors intend to do throughout the Discussion chapter.

Line 428-431: I could not agree more to this statement. I was saving those words for my final comments, thank you. But out of curiousity, what authors are being referred to here?

Line 436: Instead of the term "various" the authors should list the ML methods discussed in the manuscript. Some users may only be interested in reading this chapter of the manuscript, and still they should comprehend the paper's essence.

Line 438 onwards: This is a repetition the previous lines... revise accordingly

Lines 446-447: I completely disagree with the final statement: The paper was not about evaluating the use of Sentinel and ICESat data for elevation retrieval, but rather an application of AI/ML tools.

Line 455: The attribution to human activites is a mere speculation, and as discussed earlier, it could be a misinterpretation, and should not be stressed out in Conclusions. 

Lines 467-471: I was hoping to see what the contribution of K.W. was, only H.L. and N.Z. are mentioned (along with some other authors K.Z.? ).

 

I stopped checking for grammar mistakes in line 43. And authors should consider submitting the manuscript for some professional check of teh language

Author Response

Please see the attachment ‘Response to Reviewer 1 Comments’.

Author Response File: Author Response.pdf

Reviewer 2 Report

The tidal flats exhibit remarkable ecological significance with regard to their role in coastal protection and blue carbon sequestration. This manuscript centers on the scientific inquiry into alterations in sedimentation and erosion patterns within tidal flats. Leveraging ICESat-2 laser data in conjunction with Sentinel-2 multispectral remote sensing imagery, a composite model for remote sensing inversion is established. This novel approach enables the development of a large-scale remote sensing inversion technique for assessing tidal flat topography. By comparing variations in intertidal zone topography across different historical periods, the study examines the characteristics of sedimentation and erosion changes specifically in Jiangsu Province. The manuscript showcases a well-defined argument, abundant data, rigorous theoretical summaries, inductions, and demonstrations. The manuscript exemplifies a substantial amount of research with a coherent structure. Overall, it carries substantial theoretical significance and practical implications.

 

Here are some concerns and suggestions for modification：

 

(1) Can additional remote sensing data from other time periods be included in the paper, which used data from 2008 and 2021?  

 

(2) When building models, why divides the study area into four areas for model building? What is the basis for this division?

 

(3) The elevation datum of ICESat-2 data is the height of the WGS 84, and the article uses the 1985 national elevation datum, how the elevation datum is transformed.

 

(4) According to the research results of this article, the vicinity of the Sheyang River Estuary is silted, but some studies have found that this area is scoured through field observations. Is there any data to support the research results of this article?

 

(5) The font size in Figure 2 is too small and needs to be enlarged. Ensuring that all elements within the figure, including text and labels, are clearly legible will enhance the visual presentation of the results.

 

(6) The concerns about the font size and clarity of the text in the technical roadmap of Figure 4 are valid. To address this issue, the manuscript should increase the font size and resolution of the figure to ensure that the details of the technical roadmap are easily discernible. By improving the visibility of the text, readers will be able to better understand and appreciate the methodology outlined in the figure.

 

(7) The discrepancy between the caption of Figure 7, which states "North Region," and the representation of the entire Jiangsu region within the figure needs to be rectified. The caption should accurately reflect the content of the figure to avoid any confusion. It is recommended to revise the caption to match the scope of the depicted region.

 

Proofreading the entire manuscript for any grammatical or typographical errors would help maintain the overall quality of the paper.

Author Response

Please see the attachment ’Response to Reviewer 2 Comments.pdf ‘.

Author Response File: Author Response.pdf