Influence of Image Compositing and Multisource Data Fusion on Multitemporal Land Cover Mapping of Two Philippine Watersheds

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Thank you for submitting your paper to the Remote Sensing journal.

 

Random Forest is a well-known method for classifying data, and it proves to be a useful tool, particularly in the classification of satellite data. Publishing new research in this field is highly recommended.

 

The approach in this paper appears sound. The author utilized various satellite imagery and examined the results using ten Random Forest models. However, I believe this research should emphasize its originality and novelty more prominently. Given the extensive existing literature on the use of Random Forest for extracting land use changes and combining imagery sources to enhance land use mapping, it is crucial to articulate the unique contributions of this study. Therefore, I suggest elaborating on the research's originality in lines 145-150.

 

The methodological application of Random Forest could serve as a key highlight of the research. If there are novel findings or discussions resulting from its application, I recommend comparing your method with previous research to reveal its significance.

 

 

The most critical aspect to address is the results presented in Figure 7. While the analysis and statistical findings appear good, there are concerns regarding the trends observed, particularly in the PLW region. It seems counterintuitive that the built area would decrease. And traditionally, the previous research indicates that forest areas remain relatively stable, but in your results, the trends showed a pendulum-like pattern.

 

I am recommending you to focus on improving these results. While the statistical analyses may be sound, if the data trends do not align with expectations, it raises questions about the model's efficacy in mapping land use. Therefore, I recommend revisiting the trends for the built area and forest, striving to enhance their accuracy. If improving the trends proves challenging, it is essential to discuss why these discrepancies exist and elucidate this matter in the discussion section. By acknowledging and addressing these limitations, you can provide a more comprehensive understanding of your research's outcomes.

 

I look forward to receiving your revised manuscript with improved results.

Author Response

Dear Reviewer,

Thank you very much for taking the time to review this manuscript. Please find the detailed responses and the corresponding revisions (in red font) in the attachment below. These revisions – along with the revisions suggested by the other reviewers – are also reflected in the tracked changes of the resubmitted manuscript.

Thank you again for your valuable feedback.

Best regards,

 

Corresponding Author

Author Response File: Author Response.docx

Reviewer 2 Report

Comments and Suggestions for Authors

The manuscript titled "Influence of Image Compositing and Multisource Data Fusion 2 on Multitemporal Land Cover Mapping of Two Philippine Watersheds" has interesting topic and applicable in land cover mapping however manuscript well writing but after reading some minor comments and questions are as follow:

- Does the 12 classes of FAO meet the land cover diversity in study area and why not use other classification types?

- How many training data produced for 12 classes?

- What dataset used for validation?

- Is better to present some example of real photo of land cover types.

- Concussion not well support the result and not highlighted the main input of research output.

- Is this methods applicable for other region in Asia?

 

 

Comments on the Quality of English Language

English is good some minor check required

Author Response

Dear Reviewer,

Thank you very much for taking the time to review this manuscript. Please find the detailed responses and the corresponding revisions (in red font) in the attachment below. These revisions – along with the revisions suggested by the other reviewers – are also reflected in the tracked changes of the resubmitted manuscript.

Thank you again for your valuable feedback.

Best regards,

 

Corresponding Author

Author Response File: Author Response.docx

Reviewer 3 Report

Comments and Suggestions for Authors

With your manuscript you provided a pilot study into the effects of different sets of input data on the quality of cloud-based land use mapping results in two Philippine watersheds. Scrutinizing your presentation, I had only a few, rather formal remarks, which I indicated directly in the PDF attached.

Comments for author File: Comments.pdf

Comments on the Quality of English Language

With the exception of a small Americanism (L157: ...indices that were included), English is fine.

Author Response

Dear Reviewer,

Thank you very much for taking the time to review this manuscript. Please find the detailed responses and the corresponding revisions (in red font) in the attachment below. These revisions – along with the revisions suggested by the other reviewers – are also reflected in the tracked changes of the resubmitted manuscript.

Thank you again for your valuable feedback.

Best regards,

 

Corresponding Author

Author Response File: Author Response.docx

Reviewer 4 Report

Comments and Suggestions for Authors

Revision points:

1. Question:

1) More information about the natural environment and economic conditions of the two study sites should be included in order to compare and emphasize their unique characteristics and differences in greater depth. It was suggested later that adding terrain elements can increase the classification accuracy, therefore this should contain the description of terrain features, particularly the distribution of land types in vertical height.

2) I hadn't noticed any conversations regarding Formula (15). The text that followed made no mention of it.

3) In section 3.1, regarding the comparison of model performance, although the manuscript mentioned that the data used to generate the land cover map model covers the period from 2000 to 2020, it did not specify which year of land use data was involved in the model comparison. In order to provide much clearer information, it is recommended to supplement the specific year involved in the comparison of model.

4) Only the findings from the study and comparison of ten random forest models for the PLW basin was presented in section 3.1. Where is the BW basin results?

5) The 'coconut producing areas' stated in P.11 lines 324 and 327 was not specifically mentioned above. It was unclear whether "coconut production areas" were synonymous with "perennial crops" as mentioned in Figure 5. The authors need to explain clearly.

 

2. Tables and Figures:

1) It was recommended that Figure 2's latitude and longitude grid be included in order to improve the map's positioning accuracy and intuitiveness, making it easier for readers to recognize and correctly refer to two important watersheds.

2) It was suggested that the General Variable in Tables 2 and 3 be arranged in a manner that aligns with the formulas mentioned above. Thereby making the manuscript easier to read.

3) Figures 5 and 7's land cover types should be changed because they were inconsistent.

 

4) It is advised to combine Figure 5 and Figure 6 into a single figure.

Comments for author File: Comments.pdf

Comments on the Quality of English Language

Minor editing of English language required

Author Response

Dear Reviewer,

Thank you very much for taking the time to review this manuscript. Please find the detailed responses and the corresponding revisions (in red font) in the attachment below. These revisions – along with the revisions suggested by the other reviewers – are also reflected in the tracked changes of the resubmitted manuscript.

Thank you again for your valuable feedback.

Best regards,

 

Corresponding Author

Author Response File: Author Response.docx