Interaction between Tourism Carrying Capacity and Coastal Squeeze in Mazatlan, Mexico
Round 1
Reviewer 1 Report
In my first review of this paper, I described several concerns, and I appreciate that the authors made edits to meet those recommendations. For example, I found the new organization much easier to follow. However, while the authors increased the explanations of their data processing and methods (another one of my suggestions), unfortunately, the increased detail in the methods section shows that when digitizing the extent of urban area, the methods were not scientifically rigorous enough to warrant publication. This must be improved before the paper can be published. My instinct is that the authors’ conclusions and suggested actions may be correct – but because the urban area is not properly calculated, the data cannot be said to scientifically support these conclusions. It is possible, however, that after recalculating the urban area extent, the data will lead to similar conclusions.
Below, I describe the mandatory changes under ‘major concern,’ and a smaller section details other, minor corrections. Some typos and other edits are also annotated on the uploaded PDF of the paper.
Major concern
Unfortunately, the authors have subjectively hand digitized the extent of the urban area from imagery, and that is not the scientific standard for mapping urban land cover change. There is a large field of research and dense literature dedicated to classifying imagery into urban and other land cover types. The subjective nature of hand-digitizing (as the authors have done here) leads to inconsistencies and invalid measurements.
Figure 11 illustrates this problem. In the northwest quadrant of the city, there is a large region that is classified outside the urban development in 2004, 2010, and 2015. In 2019, this area is now included in the urban area, and would account for a substantial portion of the increase in total urban area calculated for that year. However, as is apparent in the image, that area is still mostly green space. This is why land use and land cover (LULC) change analysis uses the spectral information in the pixels of an image to classify the land type. If, as I suspect, defining the urban extent with the current standard of pixel classification results in a smaller increase in the 2019 urban area, this will also lower the slope of all their predictions (2059 and 2100) for the future extent and related pressure on the environment. Because all other analysis and discussion hinge on this one value (the area of the urban environment) it must be defined according the modern scientific standard.
I understand that image classification may not be the authors’ usual field of research, but there is a well-tested and proven standard that they must adhere to. Because all the analysis and results of this paper are contingent upon how other variables (TLC, climate changes, etc.) are correlated to the increasing urban expansion, this is a critical error in the paper. Fortunately, I believe the authors have the data, tools, and skill to fix this misstep, and will end up with an excellent paper after (1) updating their method for delineating the urban extent, then (2) recalculating the correlations between the updated urban area values and their other variables (which do not need to be recalculated) – and there are several ways they can approach it.
First: quickly review the basics of this process.
Land cover classification is an extremely common process and therefore is often the topic of lab assignments in college remote sensing courses. This means that many free, step-by-step tutorials are available via the internet, and if the authors are unfamiliar with image processing and/or this specific technique, they will find plenty of guides to aid them in this process. (I’ve listed several below.)
- [begin at slide 11] https://appliedsciences.nasa.gov/sites/default/files/week2_final.pdf
- https://wrirosscities.org/sites/default/files/mapping-urban-land-use-india-mexico-using-remote-sensing-machine-learning.pdf
- http://step.esa.int/docs/tutorials/S1TBX%20Landcover%20classification%20with%20Sentinel-1%20GRD.pdf
- https://pdfs.semanticscholar.org/0a68/77bb53c896d0fd2a37cdb3d22b32e4ec746c.pdf
Second: determine which software programs they have access to/are most comfortable with, and use that program to perform a classification that delineates the boundary of the urban area.
Options for classifying the images:
Option 1: Add a collaborator
The authors may know someone who is already working in this field, and has software or self-written codes to perform the image classification. In that case, the fastest solution might be to ask that person to perform this task (short work for an experienced remote sensing or GIS professional or researcher). They can then acknowledge that person in the acknowledgements section of the publication, or add them as a last author if they feel that the contribution warrants authorship.
Option 2: Classify the urban land cover with Google Earth Engine
The authors describe using Google Earth Pro for some of their analyses. Google Earth Engine is a sister program (also free to use) with increased analytical capabilities, including the ability to classify image pixels into urban and non-urban types. There are step-by-step tutorials for using Google Earth Engine to perform either a supervised and unsupervised classifications that will result in a digitized urban area. While the program is free to use, you do have to apply for access – but you will get a response within one week notifying you if your application has been accepted and you are granted access to the program.
- https://signup.earthengine.google.com/#!/
- Unsupervised classification method - https://developers.google.com/earth-engine/guides/clustering
- Supervised classification method - https://developers.google.com/earth-engine/guides/classification
- There are also pre-existing land cover datasets, although I don’t know if that includes high-resolution data over Mazatlan: https://developers.google.com/earth-engine/datasets/tags/landcover
Option 3: Classify the urban land cover with QGIS
QGIS is another, even more robust mapping and image processing software program that is also open-source and therefore free to download, install and use. There are many tutorials for performing land cover classification with QGIS (I’ve listed a few below, but the authors may be able to find better ones). This program has a GUI format for the classification tools, which the authors may find easier to use if they are not enjoying using Google Earth Engine (or cannot access it).
- https://pdfs.semanticscholar.org/0a68/77bb53c896d0fd2a37cdb3d22b32e4ec746c.pdf
- https://un-spider.org/advisory-support/recommended-practices/recommended-practice-land-cover-change/step-by-step
- https://www.youtube.com/watch?v=Ceyhm3DlZNY
- https://www.youtube.com/watch?v=TMbtLC4gKzE
Option 4:
I loathe to admit it, but a final option is to simply cut discussion and analysis of the links between the expansion of the urban landcover and the other variables. The authors could then mention this is a future research direction in the conclusions of the paper. This would make for a faster path to publication, but a weaker paper overall. Because I think the authors have the tools available to choose from options 1 through 3 above, and because I think the paper needs no other major edits, I sincerely hope they do not choose this final option, although I must acknowledge it exists.
Finally: Once the scientific urban area values have been scientifically calculated, the authors can rerun their analysis between that new, accurate area value and their other datasets (as described in sections XX – XX). They may find that the values don’t change that much – but regardless, the urban area must be classified this way and not with a subjective, hand digitized boundary.
I see no other major edits needed, and found minimal typos or other corrections, and I even suspect the results may not change drastically and should still support the authors’ conclusions.
Minor corrections & other suggestions
- Add to both the caption for figure 11 and the text that discusses it: a brief summary (reminder) of which equations from section 2.2 and 2.3 are used to produce the values in table 11
- e.: using the values from table XX to run equations 1 and 2 produces the values in the bottom row of table 11
- the authors do a great job of this in lines 388-393 when explaining table 13
- Lines 377 – 379 report that the number of extreme events was not predicted to increase, but this is in direct contrast with the present predictions of the IPCC that extreme events will increase in frequency and magnitude by the year 2100
- And the authors reference this themselves, for example in lines 415-417
- Lines 381-382 appear to be missing words; these sentences are incomplete/don’t make sense.
- Check with journal standard: if possible, I would suggest adding a line above the last rows of tables 11 and 13, which makes it more obvious that these values are derived from the other values in that table listed above
- I recall that the previous version of the paper made many suggestions regarding the ecosystems in the area, despite not having collected ecosystem or biological diversity data. I appreciate that the authors have removed those suggestions from this new version of the paper.
- Suggest changing section 4 to first include (current) lines 454-485; then follow with section 4.1 possible responses, and finally a new section 5 ‘Conclusion’ consisting of (current) lines 486-499
- The movement of several tables to appendices was a good decision! Much easier to read, but there to reference if needed. I especially appreciate the detailed Appendix B.
- Some notes are also included in the attached PDF of the paper
Comments for author File: 
Comments.pdf
Author Response
Please see attached file.
Author Response File: Author Response.pdf
Reviewer 2 Report
Thank you for the opportunity to re-review this manuscript. The authors present modelled estimates of the coastal squeeze and tourism capacity for Mazatlán, Mexico, at four different time steps ranging from 1999 to 2100. Their methods are based on previously-published work, and draw on a wide range of interdisciplinary data types to quantify the ability of Mazatlán to accommodate development and tourism through the end of the 21st century.
The authors addressed all of my comments from the first review, and I am satisfied with the quality of the manuscript in its present form. I have no further comments or recommendations, and suggest the editors move forward with publication. Thank you again.
Author Response
Please see attached file.
Author Response File: Author Response.pdf
Reviewer 3 Report
Title: Interaction between the tourism carrying capacity and coastal squeeze in Mazatlan, Mexico
Abstract: I find this section well written as it encapsulates the study nicely.
Introduction:
Line 47 – 49: “Anthropogenic factors contributing to coastal squeeze include global and local climate change (sea-level rise and the increased frequency and intensity of storms)” Could you please justify this statement? How anthropogenic is climate change?
Line 55 – 56: “Notable works include [10-12] who developed methodologies and spatial models to quantify habitat loss.” Could you please rephrase this statement by mentioning the authors in question for [10-12].
Please do same throughout the entire manuscript e.g. Line 58: "Another example is [13]"
In general, I find the manuscript interesting to read as it seeks to tackle a reality that a lot of cities along the coast face or will be facing in the future. However, the paper is poorly discussed and not well concluded. I suggest these sections are thoroughly revised with recommendations
Also please check the English especially the proper use of punctuations.
Author Response
Please see attached file.
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
I am impressed with how quickly the authors were able to make the adjustments to the paper. Thank you for taking the time to update the way in which the extent of the urban area was defined, and for rerunning analyses with these new vales. From the new Figure 11 and updated values, I can tell that the spectral classification generated slightly smaller values for each year of imagery, which resulted in a slightly lower prediction for 2100. This was the decrease I expected, but I am pleased to see that the decrease was not so large that it drastically altered the final conclusions of the paper.
All of my previous comments have been addressed and I see no further edits needed to this manuscript.
Reviewer 3 Report
The author has modified the manuscript as per my suggestion. I suggest accepting the manuscript in its present form.
This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.
Round 1
Reviewer 1 Report
Please see the attached.
Comments for author File: Comments.pdf
Reviewer 2 Report
There are very few spelling or grammar mistakes in this paper. Instead, my concerns can be described in two main categories: (1) the paper structure and (2) additional material needed.
First: Structure
The paper has a dense introduction, followed by 12 pages of figures and data that are all included under the heading ‘2. The DES cycle.’ I am not familiar with this paper structure and found it very difficult. Instead, I recommend the authors reorganize to clearly show what data was (and was not) studied, and the results of those investigations. Lines 110-170, which describe the study site and its geography, are in the right location. However, the paper then clearly moves into methods, which should be its own section. Data sources and their processing are described throughout sections 2.1.1 through 2.3 (the bulk of the paper, from line 171 through 313). At present, each of these sections begins with a brief description of the data source and processing, then explains findings. A better structure would be to include an entire section “3. Methods” which is totally dedicated to explain the source and analysis of all data. All results (section 4) should then be summarized separately in a following section. This change will also help the authors address the edits suggested below (additional material needed). Finally, beginning at line 314 (currently section 2.4) the authors begin to intertwine discussion as well; this should also be its own separate section, or perhaps the authors would prefer to write a combined “results & discussion” section before finally listing their conclusions as the list of recommendations. At present, the methods, results, and discussion are intertwined, which makes it difficult to (a) imagine how one would complete a repeat study, (b) determine where specific results and findings are reported, or to (c) follow the connections between results observed and recommendations made. Reorganizing the paper would solve all three issues.
Second: additional material needed
All methods in this paper need to be explained in more detail. I have serious concerns about the scientific validity of some observations (was the urban extent of the region subjectively hand digitized?). Hopefully, the authors can add material explaining, for example, how the shoreline position was determined, how the boundaries of the urban area were delineated, and how certain values were chosen for the variables in the TLC equation. For nearly every type of data examined in this study, a better explanation is needed. I hope that improved detail regarding the collection and processing of data will an adequate revision, but I do fear that in some cases, the additional explanation may prove that the methods used were invalid, and the authors will have to re-do several analyses. My specific concerns are marked within the attached PDF.
Comments for author File: Comments.pdf
Reviewer 3 Report
Thank you for the opportunity to review this manuscript. The authors have written an interesting manuscript about an interesting (and relevant) subject to the readers of Land, and I recommend it for publication following minor revisions. My line-by-line comments are attached here.
Comments for author File: 
Comments.docx
