Analyzing the Losses and Gains of a Land Category: Insights from the Total Operating Characteristic

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

 

This is an interesting manuscript that applies the TOC method to analyze the spatiotemporal characteristics of marshland changes in a specific area. The results provide more information for validity testing and comparative analysis, which has inspired me to explore the use of this tool. However, what puzzles me is that the writing purpose of this manuscript seems unclear. The authors appear to have created an "official" case using the TOC Curve Generator tool, attempting to teach readers how to use the tool and interpret the results. Although the author(s) are developers of this software, this approach does not completely meet my expectations for an academic article, as I have repeatedly read it without clearly understanding the scientific questions addressed and what was done. Given that this article focuses on the application of the TOC method, it should concentrate on discussing the unique role of this method. Here are my proposed revisions:

1. In the introduction, authors can provide more specific details on why the TOC method is used in the analysis of spatiotemporal changes in land use, its prominent advantages compared to other methods.
2. In the methodology section, authors can provide a more detailed description of the input data requirements and summarize the general principles of data processing. They should also include more details and formulas about the TOC algorithm to help readers better understand its implementation process and the potential output results.
3. I think the explanations for Figures 6 and 7 are vague, with terms such as "steep part" being unclear, and the meaning of the grey areas in the figures is not well-defined. Authors should delve deeper into discussing the practical significance of the results, going beyond just data.
4. As a reader, I not only want to see the application of methodology but also hope for authors to provide insights into the potential of using the method in this field and its integration with other methods, including a discussion on the limitations of the method.

Comments on the Quality of English Language

The manuscript needs to maintain consistency in professional terminology and provide more straightforward and popular explanations for important fig or charts.

Author Response

Thank your for your feedback. We have attached our response.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

This is my first review of the manuscript titled "Analyzing the Losses and Gains of a Land Category: Insights from the Total Operating Characteristic". In it, the authors demonstrate a practical example of the use of the Total Operating Characteristic for the analysis of land use/cover changes along time, and particularly for the evaluation of the possible associations between categorical or numeric independent variables and the observed changes. Overall, the paper is well presented, the methodology is sound and the references are adequate, making it a valuable addition to the bibliography on methodologies for land use/cover analysis. I would only point out several issues that would need attention from the authors:

Lines 123-127. Why we would need "insight" in order to rank the water and upland categories seems a bit difficult to understand at first reading. This is cleared later on, but it would be probably a good idea to reorder the explanation a little bit so it can be easily understood: if I got it right, we need to estimate the intensity of change in each category so we can rank them in order of increasing intensity.

Lines 182-186. It could be argued that gain intensity in the first period, from water and from upland, is quite similar. Or, at least, that the difference is more marked in the second period. This is also suggested by the AUC value being quite close (or indeed, closer than in the second period) to 0.5. Maybe this could be introduced in the explanation to facilitate the interpretation of Fig. 7.


Line 193-194: "Figure 6b demonstrate that marsh loses most intensively where marsh previously lost, which is a phenomenon called Alternation". This is contradictory to what is stated in lines 171-173. I can suggest that "distance to recent change" labels in the figures are changed to "distance to recent gain" or "distance to recent loss" (or, even better, "distance to gain/loss in the previous period") in order to reduce the likelihood of confusion.

Lines 254-258 outline the relevance of masking areas where change is not possible for the purposes of producing useful AUC values. But, as it is explained in lines 206-217, the AUC is certainly not the only relevant value and may not even be relevant at all, depending on the research question. Thus, masking those areas may also not be completely necessary if AUC values are not to be used in a particular analysis. Accordingly, I wonder if the statement in lines 263-265 ("We gain no insight by inclusion of thresholds to the left of the origin or to the upper-right of where the curve first arrives at the upper bound.") could be also seen in a different light: precisely, the analysis of those thresholds (whether they are included in plots or not), allows to discern areas where change is not possible (which may be, in itself, relevant information).



Typos:

Line 126: Repeated "Pontius [13]"

L. 134: "This step is relevant for TOC curves a gaining land cover class..."

L. 283. "Similarly, water and upland have 8% and 6% intensities... (*during the second interval)?"

Reference 19, relative to the TOC curve generator should probably be the same as reference 10, according to the github site of the project.

Author Response

Thank your for your feedback. We have attached our response.

Author Response File: Author Response.pdf