Improved Cropland Abandonment Detection with Deep Learning Vision Transformer (DL-ViT) and Multiple Vegetation Indices

Round 1

Reviewer 1 Report

Overall, this manuscript is well-organized and clearly written. The method is solid and conclusions are supported by the results. I think it's qualified for publication in Land journal as long as the authors make some minor changes:

1. Line 19-20, make sure you tell the audience your study area.

2. Simplify paragraph 3 in the Introduction section.

3. Lines 167-168, state the source of the DEM dataset.

4. Make sure you have Table names for each table.

5. Table 1: what is the cloud coverage? what were the dates for images you adopted?

6. The authors defined abandoned cropland as those "remains inactive for agricultural purposes for two consecutive years". Usually, the minimum temporal resolution is 5-year interval. If the authors decided to use 2-year due to the limitation of GF data availability, please cite proper references to support the choice of 2-year interval.

7. Line 356: "pre0valent", typo.

Overall the qualify is good.

Author Response

Dear Reviewer ,

Thank you for taking the time to review my manuscript, titled "Improved Cropland Abandonment Detection with Deep Learning Vision Transformer (DL-ViT) and Multiple Vegetation Indices". I appreciate the thoughtful and constructive feedback you provided. Your insights have been invaluable in refining and strengthening the quality of the paper.

Author Response File: Author Response.docx

Reviewer 2 Report

This paper presents a novel methodology integrating machine learning algorithms and VIs for the classification of abandoned croplands through Gaofen satellite data, taking also into account for the relevant analysis the associated socioeconomic factors of the study area. The use of English is excellent, although authors should apply a last check for minor typos (also listed below for the authors convenience). Literature analysis, novelty and design of methods and presentation and discussion of the results are of very high standards and quality, supported by the illustration of relevant maps, graphs, annexes and tables. Conclusions expressed highlight the findings of the study and demonstrate significant scientific importance and novelty, while all technical aspects considering the methods followed are clear and adequately described. In overall, it is suggested to accept this high-quality paper for publication, after minor additions listed below: 

 

1.    Lines 35-37: "Moreover, the consequences of abandoned lands extend beyond mere neglect, as they become sources of greenhouse gas emissions, thereby....".

According to relevant literature (e.g., see Zheng et al., 2023, https://www.nature.com/articles/s41467-023-41837-y ), abandoned cropland could serve also for the exact opposite purpose; restoration to natural habitats via natural growth (reforestation) for enhancing carbon sequestration. Thus, such a statement seems to be controversial, unless you specify that you are talking about abandoned croplands that remain barren or become built-up areas, for example.

 

2.    Line 122: "....leading to enhanced the accuracy...."

Change "enhanced" to "enhance".

 

3.    Lines 160-161: "...with panchromatic multispectral camera (PMC) and willed filed view camera (WFV)...."

Change "willed filed" to "wide field".

 

4.    Lines 167-168: "Additionally, incorporation of a digital elevation model (DEM) with a resolution of 15 meters allows for slope calculation."

Please, elaborate more on the DEM source and technical characteristics, it is not evident just by the spatial resolution provided to account for any errors during terrain correction routines.

 

5.    Page 5: Table 1 caption is missing

 

6.    Section 2.5 Vegetation Indices: Although VIs abbreviations are fully listed in the provided table 2, they should also be explained first time introduced in the main part of your manuscript.

 

7.    Line 241: ".....the SAVI employed to enhance the...."

Change "employed" to "was employed".

 

8.    Line 356: "....technique is a pre0valent approach for...".

Pleasy correct typo.

 

9.    Figure 10: You can add in the caption what blue and red colours indicate, as also done in caption of Figure 9.

Listed above

Author Response

Dear Reviewer ,

Thank you for taking the time to review my manuscript, titled "Improved Cropland Abandonment Detection with Deep Learning Vision Transformer (DL-ViT) and Multiple Vegetation Indices". I appreciate the thoughtful and constructive feedback you provided. Your insights have been invaluable in refining and strengthening the quality of the paper.

Author Response File: Author Response.docx