Shoreline Change from Optical and Sar Satellite Imagery at Macro-Tidal Estuarine, Cliffed Open-Coast and Gravel Pocket-Beach Environments

Round 1

Reviewer 1 Report

Authors explain the method of measuring shoreline change through optical satellite imagery and SAR satellite imagery, especially from Sentinel-2 and Sentinel-1, respectively. Waterline and shoreline are extracted through the image process and the other processes that consider DEM and datum. Several minor corrections both in content and figures are required.

Comments for author File: Comments.pdf

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

This paper presents techniques of high interest to the scientific community related to the detection of shorelines in macrotidal environments. Although this is relevant work, in my opinion, the current manuscript needs to be revised. Both the structure and the presentation of results and discussion must be improved before publication

The following are some essential aspects:

Line 37 PCA  is about  Principal Component Analysis?

Overall, the abstract should be written more clearly and simply. It has a lot of acronyms and technical words so that it would be difficult to understand for a reader who is not specialized in satellite imagery

Lines 118-131 Shorten this part of the text if possible.

Line 151 Insert a Reference here: “A deepening of  the area fronting the cliff is caused by the erosion of the seabed.”

Figures 2, 3 , 4 and 5 This is part of the methodology, but how the information is displayed is mistaken for results. The method should be detailed with some other table or graph, which summarises this information.

Lines 392-394- Include a Reference here: “used Transect and Baseline aAnalysis (TBA) to both compare the products with datum-based ground truth tidelines and to calculate the different metrics of coastline change.”

Figure 6 Include a Legend in the map if possible. Cyan for site Site #1 appears orange, Revise also colours and caption: “ Datum-based tidelines extracted from SurfZone DEM at Site #1 (cyan), Site #2 (green) and 416 Site #3 (fuchsia).” 

Lines 476-480 Please detail better the outputs obtained after running the methods. This sentence should be rewritten or split into 2 more precise sentences.

Lines 484-484 It's  methodology, Please delete or move for the method section

Figs 8 and 9 are in vertical position. Authors should improve these figures and include them horizontally if possible.  Also, Fig 8 is very difficult to read legend for 4-20, 20-40 ...in greyscale, i cannot be differentiated from each other.

Figure 11 Concerning WL Confidence Level, the legend of the following values: 0-1, 10-11, 11-12 there are no differences between them. Can the line colours for this range of values be improved?

Figure 12. Add legend on the map on the top for lines of Elevation and position differences between the datum-based tidelines and S2-WL for each 611 Site the datum-based tideline (fuchsia) and S2-WL (blue).

Lines 638-640 Some of the elevation differences between the datum based tideline and the S2-WL might be due to the uncertainty derived from how the  datum-based tideline has been extracted from the DEM?  Can you please explain how it has been extracted from DEM and why these differences in elevation?

 Line 787 The Section 3.4. Extraction shoreline variability via empirical orthogonal function analysis it seems that methodology and results are mixed up, the structure of this section must be revised.

Line 937 Regarding  The Section 4. Discussion and conclusions: This work is a very comprehensive study that presents many results not discussed with other authors or in previous research. Thus, it lacks a more elaborated discussion and then a concise conclusion of the advances made by this work, highlighting its importance for the scientific community.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

I would like to congratulate the authors for their manuscript. Please see the annotated pdf for some minor corrections.

Comments for author File: Comments.pdf

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

The authors have made most of the suggestions and corrections that were submitted in the first revision. Therefore the article could be published,  nevertheless, the Conclusion is still very poor. It should include at least one more paragraph highlighting the main findings of this research.

Author Response

 We thanks the reviewer for his comments. 

We have expanded the conclusion section as shown below in the revised manuscript.

"The present study contributes to our understanding of a poorly known aspect of using coastlines derived from publicly available MSI and SAR satellite missions. It outlines a quantitative approach to assess their mapping accuracy with a new non-foreshore method. The proposed non-foreshore method takes advantage of locations along the coast where the foreshore is minimum or non-existing. These non-foreshore locations are ideal places to assess the accuracy of the waterlines because its location is minimally affected by water level changes even in macro-tidal environments. The number of non foreshore locations varies with the slope of the nearshore as shown for two of the study sites. For places like Start Bay, with macro-tidal range but steep rocky outcrops along the coast, there are thousands of suitable non foreshore locations on a single Sentinel-2 tile (100 km²). For places like Spurn Head, also macro-tidal but with very gentle tidal flats, the number of non-foreshore points is often limited to places were man-made coastal infrastructure (e.g. harbours, levees, jetties, etc…) exists. This work also illustrates how the resolution of the satellite imagery can be explicitly included when assessing the metrics of shoreline change using the transect and baseline approach. We have used the new Open Digital Shoreline Analysis System (ODSAS) to explicitly include the resolution and provided a guideline with detail step by step instructions for anyone interested on using this approach elsewhere.. We have also shown how other scales of shoreline change can be extracted using the Principal Component Analysis. This is particularly relevant for SAR derived waterlines which potentially contains a richer amount of information on the different scales of change than waterlines derived from MSI imagery which image usability might be constrained due to cloud coverage. While the challenge of assessing the quality of the extracted waterlines from both MSI and SAR imagery is still on early stages of development, the mapping accuracy and skills detecting change using publicly available data is comparable to the accuracy of ordnance survey mapping. "