Identification of Ecological Restoration Approaches and Effects Based on the OO-CCDC Algorithm in an Ecologically Fragile Region

Round 1

Reviewer 1 Report

A brief summary

This paper addresses an important issue, identification of ecological restoration approaches and effects. The article is well written and addresses a wide audience, but needs some improvement.

General concept comments

1. Please, rewrite the abstract to make it clearer (too much abbreviations).

2.Please develop the introduction further, give a general background of what has been done so far, as it is difficult to determine what the article adds to science. Also show and highlight the research gap. Add more references to other studies on this issue. Maybe a deeper literature review would help? It is worth taking a broader view that these algorithms such as XGBoost also work well for other applications eg. Bartold, M.; Kluczek, M. 2023. A Machine Learning Approach for Mapping Chlorophyll Fluorescence at Inland Wetlands. (https://doi.org/10.3390/rs15092392) and Sharma, R.C. Dominant Species-Physiognomy-Ecological (DSPE) System for the Classification of Plant Ecological Communities from Remote Sensing Images. Ecologies 2022, 3, 323-335. https://doi.org/10.3390/ecologies3030025.

3. Also work on the discussion, show what results other researchers have achieved and what you have achieved and why they differ.

Author Response

Please see the attachment

Author Response File: Author Response.docx

Reviewer 2 Report

Even though this is an interesting study, it could be useful for understanding the ecological restoration effects. However, the article needs some revisions. Please address the comments to improve the quality of your article.

1. The abstract should be revised and possibly condensed slightly. 

2. How did you select the boundary of your study area? Is it determined by the river basin? Please clarify. In addition, Figures 1 require revision. Probably, remove panel b. On the panel c, authors may include river networks on top of DEM. Please review the manual for ArcGIS or another professional software in order to generate publication-quality figures. 

3. The OO-CCDC algorithm requires a thorough explanation with equations. Only figure 2 lacks clarity. Moreover, the resolution of figure 2 is obscure to me! Consider reorganizing the figures and text size to make the written material more professional. 

4. Please describe NDVI, NDPI, GLCM, and RF models with a thorough discussion and equations, accompanied by citations. For example, this reference describes NDVI: Sarker, Shiblu, "Investigating Topologic and Geometric Properties of Synthetic and Natural River Networks under Changing Climate" (2021). Electronic Theses and Dissertations, 2020-. 965. https://stars.library.ucf.edu/etd2020/965

5. Figures 6, 7 and 13 require extensive revision. Authors may use Python or Matlab to generate publishable figures. Reduce the space between subplots. Please review this python toolbox. https://timcera.bitbucket.io/plottoolbox/docs/command_line.html#time 

6. What is the importance of this study? Please explain the implications of this study in the context of environmental protection. Please describe in a distinct section (prior to the conclusion) the potential implications of this study. Please review the following literature to address this issue. Gao et al. (2022), Analyzing the critical locations in response of constructed and planned dams on the Mekong River Basin for environmental integrity, Environmental Research Communications, https://iopscience.iop.org/article/10.1088/2515-7620/ac9459

P. ANDERSON, Ecological restoration and creation: a review, Biological Journal of the Linnean Society, Volume 56, Issue suppl_1, December 1995, Pages 187–211, https://doi.org/10.1111/j.1095-8312.1995.tb01133.x

Perhaps it would be beneficial for the authors to revise their compositions, particularly the sentence structure.

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Reviewer 3 Report

The article appears aimed at identifying the approaches and effects of ecological restoration based on the OO-CCDC algorithm in the ecologically fragile region of Ninfgia.

The abstract appears to be well structured, rich in information and with a focus on the applied methodologies and the results achieved.

The survey methodology seems appropriate and scientifically sound.

A good percentage of current bibliographic references (last 5 years) is present. The manuscript appears to be scientifically sound, and the experimental design and applied methodology, however, is not expounded with much clarity. If the authors are interested in further investigation modalities akin to the one carried out, the following reference is suggested: Fiorentino, C.; et. al. New Technique for Monitoring High Nature Value Farmland (HNVF) in Basilicata. Sustainability 2023, 15, 8377. https://doi.org/10.3390/su15108377

The conclusions appear to be consistent with the evidence and arguments presented; however, they should be supplemented with more experimental information. The paper appears to be syntactically-grammatically correct.

Minimal revision is required, in relation to bibliographical references, experimental design, proposed methodology, better exposition of conclusions.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 4 Report

We do want to comment that some improvements should be made. For example, more details on the choice of the Random Forest classification method and the proposed OO-CCDC method should be provided. Also, in the Table 3 of the section on Multiscale segmentation, it would be better to provide visualizations of the segmentation results rather than using descriptive texts. How are the results of Table 3 used in the study?

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 5 Report

I reviewed the paper entitled "Identification of ecological restoration approaches and effects based on the OO-CCDC algorithm in an ecologically fragile region" by Wei et al. The paper has serious flaws, making it unsuitable for publication in Remote Sensing. First, the novelty of the paper is incremental and very limited and does not meet the standards of Remote Sensing. Second, the paper suffers from a lack of suitable references and detailed explanations for reproducibility. It also includes several errors in different parts. Some specific comments are provided below:

(Section 1)

1- There is a lack of references in the Introduction section:

- Line 40; Which "other studies"? Please cite relevant studies.

- Lines 62-64; Please cite papers relevant to the corresponding statement.

- Line 79; Which "few studies"? Please cite relevant studies.

2- What is the contribution of this paper when compared with previous publications that evaluated the ecological effects of multiple restoration approaches?

(Section 3) 

3- The authors combined two well-known algorithms of multi-scale segmentation and CCDC, and the proposed methodology includes very limited novelty. It seems to be a new strategy than a new algorithm/method.

4-  I understand that the data preparation was performed within the GEE, and the segmentation process was done using the eCognition software. So, which platform was used for data integration and trend analysis?

5- Line 118; What was the utility of the land use map of 2022? Complete information (e.g., accuracy, data source, etc.) of this map should have been stated in the paper. Is it the land cover map that you generated according to section 3.2? Did you use the land cover map of 2022 to extract the land cover types for your analysis? If so, why only 2022 was used for the purpose? As already observable in the results section, the study area has experienced land cover changes, so why multi-year land cover maps was not used?

6- Did you use any scene cloud cover percentage to exclude highly clouded scenes?

7- Lines 133-134; The authors mentioned the lowest and highest number of images were 70 and 253; however, these values do not match with provided values in Figure 3. Why?

8- Lines 141-142; Why the median synthesis was used for seasonal data production?

9- Lines 149-150; The authors mentioned that 2,288 field samples were collected; however, the sum of samples for each class (e.g., 623, 687, and 532) does not match 2,288. This is also observable in Table 1.

10- Lines 191-193; Please provide the tunning parameters of the RF model. How did you tune the required parameters?

11- In Figure 5b/c, there are two graphs (lines) in each plot. One for pixel-based and one for object-based. The value of the object is the median value of all corresponding pixels, and it is clear. So, which pixel has been chosen for comparison as the pixel-based graphs? For instance, an object contains 100 pixels, and the object-based graph is calculated as its median value. Which one of these 100 pixels was used as the pixel-based graph?

12- Line 231; Please correct the citation of Zhu et al.

(Section 4)

13- Tables 5 and 6; The reference samples as described in section 3.1.2, are point-based samples? How did you use point-based samples to evaluate the object-based method? This is critical to show the reliability of the evaluation process. A thorough description, along with its uncertainties, is required to make it acceptable.

14- How did you calculate the accuracies for each year? Do you check the status of all reference samples each year using Google Earth and GF1-2? Why are there so many fluctuations in the accuracies of both methods? No explanation is provided.

15- The provided description of Figure 9 is not enough. I was not able to get the correct message of the Figure. A Figure with high-quality and clear description is required.

16- Line 308; How did you justify that the achieved accuracies met the change detection and trend analysis requirements?

(Section 5)

17- Line 388; Which "few studies"? Please cite and discuss these studies. What is the difference between your paper with the previously published studies?

18- Line 391-392; These statements are not acceptable. For instance, CCDC can provide information about the evolution and temporal behavior of a time series. So, what do you mean? Please clarify.

19- Lines 404-405; The authors mentioned the effect of the optimal segmentation scale on the detection accuracy; however, no statistical values were given. It is suggested to provide the statistical values for justification and clarification.

In light of these, I can not recommend this paper for publication in Remote Sensing.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

Thanks for the revision. 

Looks good but Check it Again. 

Reviewer 5 Report

I appreciate the authors' efforts to address all comments in detail and reasonably. The current version of the paper is suitable for publication in Remote Sensing.