A Remote Sensing Approach to Estimating Cropland Sustainability in the Lateritic Red Soil Region of China

Round 1

Reviewer 1 Report

1. The introduction is slightly confusing and does not center on the methodology proposed in the paper to explain the advantages and disadvantages of different indicators and different evaluation methods.

2. There are many elements of the study area that are unrelated to the manuscript.

3. The resolution of figures 1, 2, 3, 4 and 5 is too low, and the text in the figures is illegible, so it is recommended that high-resolution images be regenerated.

4. The paper does not indicate the innovative points compared to other studies, such as the specificity of the selection of indicators or the novelty of the methodology.

5. The paper should explain why red soil should be highlighted for sustainability research compared to other soils.

6. Simple formulas for R2, RMSE, MAE, UA, PA,OA, etc. do not need to be written out in the manuscript, but only briefly explained.

7. The manuscript does not describe the parameter settings for the different machine learning methods, making it difficult to believe their true accuracy.

8. After comparing the evaluation scores, it is not comparable to actual land sustainability, and other indicators are needed to assist in clarification.

9. Overall, the manuscript lacks innovation and the research methodology is common and cannot be described in detail, but is merely a listing of results. Therefore, I consider the manuscript unsuitable for publication in Remote Sensing journal.

English language needs further improvement.

Author Response

Comment 1:

The introduction is slightly confusing and does not center on the methodology proposed in the paper to explain the advantages and disadvantages of different indicators and different evaluation methods.

Response：

Thanks for your suggestion. We improved the relevant descriptions according to your comments in the revised manuscript. First, different indicators represent different aspects of cropland sustainability. Cropland sustainability is a comprehensive concept, and the evaluation system constructed by multiple indicators help us understand the evaluation results more accurately. Although existing studies have scientifically revealed several important factors influencing the sustainable use of cropland, such as natural conditions, soil fertility, and farmland productivity, a comprehensive assessment of cropland sustainability is still lacking. (Line 75-79). Based on the existing research, according to the characteristics of cropland in lateritic red soil region, this study selected ten typical evaluation indicators from three aspects of nature, management and food productivity, including soil organic matter, soil pH value, irrigation assurance capability, multiple cropping index and food productivity, etc. These indicators are representative and can be used to reflect the cropland sustainability in lateritic red soil region (Line 185-215). Second, we summarized the advantages and disadvantages of different evaluation methods, including comprehensive index method, fuzzy comprehensive evaluation model, pressure state response (PSR) model and remote sensing method (Line 80-90). Remote sensing technology is a cost-effective, rapid, and reproducible method to quantitatively assess the cropland sustainability and obtain its spatial information, especially at the large regional scale. Therefore, remote sensing technology was used to assess the cropland sustainability in the study area during 2010-2020.

Comment 2:

There are many elements of the study area that are unrelated to the manuscript.

Response：

Thank you so much for your suggestion. We have revised the descriptions of the study area. To give reviewers and readers a better understanding of the Lateritic red soil in the study area, we removed the unrelated content and supplemented essential information. Also, we added the spatial distributions, topographic characteristics, the potential for agricultural production of Lateritic red soil cropland, and the threats to cropland sustainability in the study area (Line 128-141).

Comment 3:

The resolution of figures 1, 2, 3, 4 and 5 is too low, and the text in the figures is illegible, so it is recommended that high-resolution images be regenerated.

Response：

Thank you so much for your reminding. We regenerated the high-resolution images and replaced them in the appropriate places in the revised manuscript (Line144, 180, 356, 367 and 385).

Comment 4:

The paper does not indicate the innovative points compared to other studies, such as the specificity of the selection of indicators or the novelty of the methodology.

Response：

Thank you for your suggestion. The innovation of this study has been emphasized in the abstract and introduction of the revised manuscript (Line 20-23; Line 115-119). The innovation of the evaluation indicators selected in this study is that it can effectively and comprehensively represent the typical characteristics of cropland in lateritic red soil region. First, lateritic red soil is characterized by poor fertility, acidity, high viscosity, and a thick soil layer (Tang et al., 2022; Li et al., 2021). Therefore, SOM, soil pH, soil texture, and soil layer thickness were selected as indicators of the natural capacity of the cropland (Line 185-187). Second, the cropland in lateritic red soil region is usually more fragmented, and small patches are often abandoned. Therefore, this study selected the centralized contiguity as one of the indicators to evaluate the cropland sustainability (Line 199-202).  The cropland in the lateritic red soil region is mainly paddy field. Thus, we selected the irrigation guarantee capability as one of typical indicators (Line 196-198). The cropland in the lateritic red soil region is overutilized because of the favorable hydrothermal conditions. Overuse of land is unsustainable. Therefore, multiple cropping index, which represents land use intensity, was chosen in this study (Line 207-209). Third, food productivity can reflect cropland sustainability directly. Considering the influence of natural disasters on grain production in lateritic red soil region, the high food productivity and stable food productivity were selected as the key indicators to evaluate cropland sustainability (Line 212-215).

In this study, remote sensing technology was used to evaluate the cropland sustainability at regional scale. By comparing the estimation accuracy of different models, the model with the highest accuracy was selected to obtain the spatial information of evaluation indicators (Line 309-314). The accuracy verification results showed that the cropland sustainability results based on remote sensing and machine learning models are accurate, which also highlighted the innovation of the evaluation method in this study.

1. Tang, X.; Hu, J.S.; Lu, Y.; Qiu, J.C.; Dong, Y.Q.; Li, B. Soil C, N, P stocks and stoichiometry as related to land use types and erosion conditions in lateritic red soil region, south China. Catena. 2022, 210, 105888. https://doi.org/10.1016/j.catena.2021.105888.
2. Li, Y.; Mo, Y.Q.; Are, K.S.; Huang, Z.G.; Guo, H.; Tang, C.; Abegunrin, T.P.; Qin, Z.H.; Kang, Z.W.; Wang, X. Sugarcane planting patterns control ephemeral gully erosion and associated nutrient losses: Evidence from hillslope observation. Agr Ecosyst Environ. 2021, 309, 107289. https://doi.org/10.1016/j.agee.2020.107289.

Comment 5:

The paper should explain why red soil should be highlighted for sustainability research compared to other soils.

Response：

Thank you for your good advice. Lateritic red soil covers an area of more than 400,000 square kilometers globally and is primarily concentrated in China (Shi et al., 2004). Guangdong Province has the largest area, accounting for 37% of the total lateritic red soil area in China. Lateritic red soil region is an important base for grain production in South China because of its superior hydrothermal conditions, affecting the survival and livelihoods of nearly 200 million people (Liao et al., 2015). Therefore, evaluating the cropland sustainability of lateritic red soil is crucial to increase farmers' incomes and ensure food security. We also listed the main reasons and importance of evaluating lateritic red soil (Line 104-113).

1. Shi, X.Z.; Yu, D.S.; Sun, W.X.; Wang, H.J.; Zhao, Q.G.; Gong, Z.T. Reference benchmarks relating to great groups of genetic soil classification of China with soil taxonomy. Chinese Sci Bull. 2004, 49, 1507-1511. https://doi.org/10.1360/03wd0476.
2. Liao, Q.; Hu, Y.M.; Hu, X.F.; Zhao, J.Y.; Wang, L. Spatial variability of topsoil nutrients in typical lateritic red soil areas of Guangdong Province. Bull of Soil and Water Conservation. 2015, 35, 322-328. (In Chinese). https://doi.org/10.13961/j.cnki.stbctb.2015.06.055.

Comment 6:

Simple formulas for R2, RMSE, MAE, UA, PA, OA, etc. do not need to be written out in the manuscript, but only briefly explained.

Response：

Thank you for your suggestion. In the revised manuscript, we deleted these commonly used and simple formulas and added relevant explanations (Line 297-299 and Line 304-305).

Comment 7:

The manuscript does not describe the parameter settings for the different machine learning methods, making it difficult to believe their true accuracy.

Response：

Thank you for your suggestion. We supplemented a description of the process for parameter tuning of machine learning models (Line 241-242), and added a new Table 4 to clearly explain the parameter settings of different machine learning models in the revised manuscript (Line 245).

Comment 8:

After comparing the evaluation scores, it is not comparable to actual land sustainability, and other indicators are needed to assist in clarification.

Response：

Thank you for your comment. We constructed a comprehensive evaluation indicator system from three aspects: natural capacity, management level and food productivity, including soil organic matter content, soil pH, irrigation assurance capability, land use intensity, food productivity and so on. These indicators are representative and closely related to cropland sustainability in lateritic red soil region. Moreover, by comparing with the field survey data, the evaluation results of indicator in this study were reliable (Line 309-314; Line 320-325; Line 329-336). Therefore, the results of the study can reflect the actual cropland sustainability to a certain extent. As we mentioned in the “Limitations and future work” section, in future studies, additional indicators should be further integrated to build a more comprehensive cropland sustainability evaluation system. For example, adding more indicators to the evaluation system, such as total nitrogen, available phosphorus, available potassium, ecological environment, and biodiversity, would enhance the evaluation of cropland sustainability.  (Line 506-510).

Comment 9:

Overall, the manuscript lacks innovation and the research methodology is common and cannot be described in detail, but is merely a listing of results. Therefore, I consider the manuscript unsuitable for publication in Remote Sensing journal.

Response：

Thank you for your comment. According to the opinions of reviewers and editors, we have carefully revised the manuscript. We hope the revised manuscript can get your approval. If there is still something that needs to be improved, please mention it, and we will do our best to improve it until it can meet the requirements of the journal. Thanks again for your careful review.

The innovation of this study is to propose a comprehensive indicator system for evaluating cropland sustainability in lateritic red soil region at regional scale, and we used remote sensing methods to reveal the spatiotemporal pattern of cropland sustainability in lateritic red soil region of Guangdong Province from 2010 to 2020. We have highlighted the innovations of this study in the abstract and introduction of the revised manuscript (Line 20-23; Line 115-119). According to the research results, we believed that the proposed cropland sustainability evaluation system and method can be used to reveal the temporal and spatial model of cropland sustainability in the lateritic red soil region. This is essential to promote sustainable development of cropland in the lateritic red soil region.

Comment 10:

English language needs further improvement.

Response：

Thank you for your suggestion. In order improve the readability of the manuscript, we invited professional native English speakers to polish the revised manuscript. The copy of certificate is in the word document (Detailed Responses to Reviewer-1).

Author Response File: Author Response.pdf

Reviewer 2 Report

In this study, a comprehensive evaluation indicator system for cropland sustainability was constructed from three aspects—natural capacity, management level. It was adopted to analyzing spatiotemporal patterns of cropland sustainability in lateritic red soil areas of Guangdong Province from 2010 to 2020.  This study provides a practical evaluation system and methods for dynamic monitoring of cropland sustainability, and the research results can provide a decision basis for cropland protection. This study provides a new method for the assessment of cultivated land sustainable development, and has a certain reference value for the development of this field. But there are some small problems that need to be corrected.

Minor editing of English language required

Author Response

Comment 1:

In this study, a comprehensive evaluation indicator system for cropland sustainability was constructed from three aspects—natural capacity, management level. It was adopted to analyzing spatiotemporal patterns of cropland sustainability in lateritic red soil region of Guangdong Province from 2010 to 2020.  This study provides a practical evaluation system and methods for dynamic monitoring of cropland sustainability, and the research results can provide a decision basis for cropland protection. This study provides a new method for the assessment of cultivated land sustainable development, and has a certain reference value for the development of this field. But there are some small problems that need to be corrected.

Response：

We really appreciate for your support and affirmation of this research. According to the suggestions of editor and other reviewers, we have tried our best to improve the manuscript. We believe that the small problems you mentioned have been corrected.

Comment 2:

Minor editing of English language required.

Response：

Thanks for your suggestion. In order improve the readability of the manuscript, we invited professional native English speakers to polish the revised manuscript. The copy of certificate is in the word document (Detailed Responses to Reviewer-2).

Author Response File: Author Response.pdf

Reviewer 3 Report

The introduction: you stated your literature information in good shape and order. Ultimately, you need to place your objectives clearly and precisely, allowing the reader to understand the study and ensure they match the results.

The Materials and Methods:

In Table 1, you showed different data sets in various resolutions and did not mention how you matched these data sets to run in a model. You need to explain the processes that you went through to prepare and run these data.

Results

In the materials and methods, you mentioned that you used land classification with an overall accuracy of 90%. Give a summary of your classified data, even by images, and the accuracy assessment of the two years.

Conclusion

You expressed the findings that you got from your study in the conclusion. However, you stated the limitations of your work and the need for future improvements in the discussion. You are encouraged to explain some details for future work. Therefore, elaborate on your conclusion to include some future recommended research and suggestions.

After reading the manuscript, I found some grammar mistakes in the text parts, including sentence structure, subject-verb agreement, and punctuation.

     For example,

In the introduction:

Therefore, there is an urgent need for a cropland (cropland) sustainability.

In Materials and Methods: This area covers 21 prefecture-level cities and (no and here) 76 counties and urban areas of the province. In this study, a higher SOM content, neutral soil pH, finer soil texture, deeper soil thickness, and lesser (a lesser) slope indicated a higher natural capacity of the cropland. In 2010 and 2020, the scores for the management level (levels) were high in CZ. Larger values of high (highly) stable food productivity indicate stronger overall food productivity of cropland.

In discussion: The results of this study were consistent with those of existing studies, indicating that the proposed cropland sustainability assessment system. Low (the low) nutrient content is a major feature of lateritic red soils, and (no and here) thus increasing the SOM is conducive to enhancing cropland sustainability.

Author Response

Comment 1:

The introduction: you stated your literature information in good shape and order. Ultimately, you need to place your objectives clearly and precisely, allowing the reader to understand the study and ensure they match the results.

Response：

Thank you for your suggestion. To make readers better understand, in the introduction part of the revised manuscript, we emphasized three main objectives of this study (Line 119-125). The purpose of this study was to: 1) propose a universal assessment indicator system and a remote sensing–based framework for cropland sustainability in lateritic red soil region; 2) reveal the spatial differences in and temporal dynamics of changes in cropland sustainability in the lateritic red soil region of Guangdong Province over the past decade; and 3) provide a scientific basis for the sustainable development of cropland in lateritic red soil region, ensuring the efficient use of cropland and increasing grain yield..

Comment 2:

The Materials and Methods: In Table 1, you showed different data sets in various resolutions and did not mention how you matched these data sets to run in a model. You need to explain the processes that you went through to prepare and run these data.

Response：

Thanks for your useful suggestion. We have supplemented the detailed descriptions of the data process in the revised manuscript (Line 164-175).

Comment 3:

Results: In the materials and methods, you mentioned that you used land classification with an overall accuracy of 90%. Give a summary of your classified data, even by images, and the accuracy assessment of the two years.

Response：

Thank you for your good question. In the revised manuscript, we supplemented the description of accuracy for classified data (Line 158-162). The China Land Use/Land Cover Remote Sensing Monitoring data (CNLUCC) dataset is based on Landsat TM/ETM and Landsat OLI remote sensing images that are generated by artificial visual interpretation with a spatial resolution of 30 meters. Based on the land cover classification system, the data set includes six first-level types of cultivated land, forest land, grassland, water area, construction land, and unused land. To ensure the classification accuracy of the dataset, a random sampling verification method and the Kappa coefficient were used (Ning et al., 2018). The dataset utilized in this research is known for its scientific classification system and high accuracy and has been widely used in scientific research. In the Resource and Environmental Science and Data Center (https://www.resdc.cn/), there are instructions for data production and accuracy verification.

1. Ning, J.; Liu, J.Y.; Kuang, W.H.; Xu, X.L.; Zhang, S.W.; Yan, C.Z.; Li, R.D.; Wu, S.X.; Hu, Y.F.; Du, G.M. Spatiotemporal patterns and characteristics of land-use change in China during 2010-2015. J Geogr Sci. 2018, 28, 547-562. https://doi.org/10.1007/s11442-018-1490-0.

Comment 4:

Conclusion: You expressed the findings that you got from your study in the conclusion. However, you stated the limitations of your work and the need for future improvements in the discussion. You are encouraged to explain some details for future work. Therefore, elaborate on your conclusion to include some future recommended research and suggestions.

Response：

Thank you for your suggestion. In the "Limitations and future work" section of the revised manuscript, we supplemented some details about future work (Line 515-520). In addition, based on the analysis of the research results, we improved the conclusion, and provided suggestions for future research and how to provide the cropland sustainability in lateritic red soil region (Line 537-539).

Comment 5:

Comments on the Quality of English Language: After reading the manuscript, I found some grammar mistakes in the text parts, including sentence structure, subject-verb agreement, and punctuation.

Response：

Thank you for your suggestion. We have corrected the grammar mistakes you mentioned in the manuscript. In addition, we invited professional native English speakers to polish the revised manuscript. The copy of certificate is in the word document (Detailed Responses to Reviewer-3).

Author Response File: Author Response.pdf