Classification of Conservation Tillage Using Enhanced Spatial and Temporal Adaptive Reflectance Fusion Model
Round 1
Reviewer 1 Report
The paper presents a novel and original approach to remote estimation of ground conditions representing crop cover status and soil management practices associated to soil conservation. The methodology proposed was highly performant (and better than other approaches also tested) when applied to the selected study area, where background information provided the due matching of remote-borne estimates with actual ground data. The time and space resolution of estimates is actually compatible with the requirements of a detailed monitoring of the actual implementation of soil conservation practices in cropland. The research outcomes may help improving ongoing monitoring schemes carried out by the responsible bodies in what regards adoption of soil conservation practices in agricultural areas.
The paper is very well written and structured, providing the necessary literature background. An extensive Methodology section provides an insight on the extensive and robust research work performed. Results are clearly presented and discussed in depth, and the conclusions are adequately drawn from results obtained.
In formal, as well as in substantial terms, the paper does not require any further improvement or comment to be considered by the authors.
As so, the manuscript is recommended for publication in its present form.
Author Response
Dear editor and reviewer,    
Re: Manuscript reference: Remote Sensing-2081692
Classification of conservation tillage using enhanced spatial and temporal adaptive reflectance fusion model.
Firstly, we would like to thank the reviewer and editor for reviewing and evaluating our manuscript. We appreciate the opportunity to revise our manuscript. We would like to thank the reviewer and editor for their contribution, they have clearly put considerable effort into reviewing our work and their input has undoubtedly resulted in a stronger paper.
Please find attached a revised version of our manuscript “Classification of conservation tillage using enhanced spatial and temporal adaptive reflectance fusion model”, which we would like to resubmit for publication as a research article in the Remote sensing.
Your comments and those of the reviewers were highly insightful and enabled us to improve the quality of our manuscript greatly. In the following pages are our point-by-point responses to each of the comments of the reviewers as well as your own comments.
Revisions in the text are shown in revision mode. In accordance with the suggestions of the editor and reviewer, the authors have revised and added to this manuscript. The authors emphasize that the study was on maize and redefine conservation tillage according to the proceedings of the 1st World Congress on Conservation Agriculture held in Madrid in 2001. The authors explain the meaning of the abbreviation "NT." The authors have corrected or deleted inaccurate statements in the manuscript. The diagrams in the article were fixed. Details missing from the article were added. The applicability of the tillage pattern discriminant method chosen in this study was explained. In the conclusion section of the manuscript, the following research directions for this study are outlined. More convincing references are added to some of the ideas presented in this article. Finally, formatting adjustments were made to the references section of this article.
We shall look forward to hearing from you at your earliest convenience.
Yours sincerely,
Authors
Responses to the comments of Reviewer #1
Revisions in the text are shown in revision mode. In accordance with the suggestions of the editor and reviewer, the authors have revised and added to this manuscript. The authors emphasize that the study was on maize and redefine conservation tillage according to the proceedings of the 1st World Congress on Conservation Agriculture held in Madrid in 2001. The authors explain the meaning of the abbreviation "NT." The authors have corrected or deleted inaccurate statements in the manuscript. The diagrams in the article were fixed. Details missing from the article were added. The applicability of the tillage pattern discriminant method chosen in this study was explained. In the conclusion section of the manuscript, the following research directions for this study are outlined. More convincing references are added to some of the ideas presented in this article. Finally, formatting adjustments were made to the references section of this article. We hope that the revisions in the manuscript and our accompanying responses will be sufficient to make our manuscript suitable for publication in the remote sensing.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
The paper reports a methodology for the recognition of soils cultivated with conservation agriculture techniques through remote sensing. This methodology is based on the comparison between the crop residue cover (CRC) detected and the CRC estimated through the NDTI index predicted by ESTARFM with the fusion of spectral bands obtained from Sentinel 2 and MODIS. The model is applied in a large area of Jilin Province in China.
The paper is current and the topic has a certain scientific importance because the discrimination between traditional agriculture and conservation agriculture implemented with remote sensing is crucial above all to verify whether farmers have adopted the techniques required by government regulations regarding soil protection.
General remarks
Conservation agriculture techniques are based on machines that perform, before sowing, a reduced tillage (RT) or a non tillage (NT) which leave a quantity of crop residues (CRC) > 30% on the surface. Consequently, the best time to obtain these images is immediately after sowing and before the crop develops because, in this case, the soil cover is due to the crop and not the crop residue left by the previous crop. For maize, the CRC does not seem to have been detected in the optimal period (June 12) and the NDTI was obtained from May to June in a too large region that included not only maize, but also soybeans and rice.
Some statements about conservation agriculture need to be revised.
The observations made by the Authors in the conclusions confirm that the work needs further major modifications.
Particular remarks
Abstract: specify somewhere that it is maize
Lines 36-37: conservation practices do not always maintain or improve soil fertility, which would be better defined as soil organic matter. Better to enter "often".
Lines 41-41: the definition of Conservation Agriculture is not the one reported by the Authors. I recommend deleting the sentence and replacing it with the definition reported in the proceedings of the 1st Word Congress on Conservation agriculture held in Madrid 2001.
Line 61: What is NT? It is never explained by the Authors. If NT stands for No-Tillage then it cannot be said that Conservation Tillage coincides with NT because there are other techniques that leave >30% residue on the surface without being NT (Strip Till, Ridge till, etc.)
Lines 182-183: the sentence is not clear
Table 1: specify the units of measurement in "Spatial resolution"
Lines 239-241: fix the sentence like this: “Currently, the NDTI using Sentinel-2 images can be calculated as follows:” eliminating: “based on the spectral band calculation can well estimate the CRC.”
Line 255: Delete the sentence "This method is highly accurate."
Line 264: Specify whether y is NTDI and x is CRC or vice versa
Line 290: In my opinion chapter 2.4.5 could be dropped
Figure 3: It is the same as Figure 4 on the right
Table 4: it is not clear what “cord” means
Line 451: see line 61
Conclusions: the reported sentences highlight that the paper could be much more useful if it were improved on some aspects: (a) more precise localization of the ESTARFM algorithm; (b) apply a CRC estimation model in spring; (c) study methods for monitoring conservation tillage over long time series.
Author Response
Dear editor and reviewer,    
Re: Manuscript reference: Remote Sensing-2081692
Classification of conservation tillage using enhanced spatial and temporal adaptive reflectance fusion model.
Firstly, we would like to thank the reviewer and editor for reviewing and evaluating our manuscript. We appreciate the opportunity to revise our manuscript. We would like to thank the reviewer and editor for their contribution, they have clearly put considerable effort into reviewing our work and their input has undoubtedly resulted in a stronger paper.
Please find attached a revised version of our manuscript “Classification of conservation tillage using enhanced spatial and temporal adaptive reflectance fusion model”, which we would like to resubmit for publication as a research article in the Remote sensing.
Your comments and those of the reviewers were highly insightful and enabled us to improve the quality of our manuscript greatly. In the following pages are our point-by-point responses to each of the comments of the reviewers as well as your own comments.
Revisions in the text are shown in revision mode. In accordance with the suggestions of the editor and reviewer, the authors have revised and added to this manuscript. The authors emphasize that the study was on maize and redefine conservation tillage according to the proceedings of the 1st World Congress on Conservation Agriculture held in Madrid in 2001. The authors explain the meaning of the abbreviation "NT." The authors have corrected or deleted inaccurate statements in the manuscript. The diagrams in the article were fixed. Details missing from the article were added. The applicability of the tillage pattern discriminant method chosen in this study was explained. In the conclusion section of the manuscript, the following research directions for this study are outlined. More convincing references are added to some of the ideas presented in this article. Finally, formatting adjustments were made to the references section of this article.
We shall look forward to hearing from you at your earliest convenience.
Yours sincerely,
Authors
Reviewer #2: Some statements about conservation agriculture need to be revised.The observations made by the Authors in the conclusions confirm that the work needs further major modifications.
1)Abstract: specify somewhere that it is maize
Response: Thanks for the comments. As suggested, the author modifies the crop residue cover (CRC) to maize residue cover (MRC) and all CRC in the text to MRC.
2)Lines 36-37: conservation practices do not always maintain or improve soil fertility, which would be better defined as soil organic matter. Better to enter "often".
Response: Thanks for the comments. As suggested, The author revised lines 40-42 as "Consequently, soil erosion has been reduced, soil organic matter has been often maintained and even increased, and black soil degradation has been effectively curbed. "
3)Lines 41-41: the definition of Conservation Agriculture is not the one reported by the Authors. I recommend deleting the sentence and replacing it with the definition reported in the proceedings of the 1st Word Congress on Conservation agriculture held in Madrid 2001.
Response: Thanks for the comments. As suggested, the author refers to the proceedings of the 1st Word Congress on Conservation agriculture held in Madrid in 2001 and changes Lines 42-46 to"There is currently no uniform understanding of conservation tillage. However, from the proceedings of the 1st World Congress on Conservation Agriculture held in Madrid in 2001, one authoritative definition of conservation tillage acknowledges that permanent soil coverage, crop rotations, and the reduction or elimination of tillage are essential(Vaneph S,2001)."
The references we mentioned are as listed below.
[Vaneph S, Benites J. First world congress on Conservation Agriculture A world-wide challenge. J. 2001.]
4)Line 61: What is NT? It is never explained by the Authors. If NT stands for No-Tillage then it cannot be said that Conservation Tillage coincides with NT because there are other techniques that leave >30% residue on the surface without being NT (Strip Till, Ridge till, etc.)
Response:Thanks for the comments. As suggested, the authors explain NT in Lines 65-87, which stands for No-tillage.
No-tillage: soil undisturbed with no primary or secondary tillage. Crop seeded/planted directly into the soil(Carter, 2005). Strip tillage is any row–crop cultural practice that restricts soil and residue disturbance to less than 30% of the field area (Morrison, 2002). For example, for crop rows spaced at 76 cm, tillage strips should be less than 19 cm in width. The definition presumes that the intervening 75% or more of the field surface area will be at least partially covered by residues, and the soil will not be intensely disturbed. Ridge tillage is defined as follows: Primary tillage is confined to forming raised ridges or beds in rows, often on a contour. Planting occurs on the ridges. Ridge tillage is consistent with strip-tillage features in remote sensing identification (Carter, 2005). Identifying the three tillage methods using remote sensing has yet to be reported. Zhou (2018) used a UAV with multispectral and infrared thermal imaging sensors to remotely monitor farmland where strip tillage was implemented. Sullivan (2008) evaluated the sensitivity of a remotely derived crop residue cover index for depicting conventional tillage (CT), strip tillage (ST), and no-tillage (NT) systems in a cotton-corn-peanut rotation in the southeastern Coastal Plain. However, due to the need for more spatial resolution of high-medium spatial resolution satellite images, research on identifying No-tillage, strip-tillage, and ridge-tillage has yet to be reported. Therefore, this study could not accurately distinguish the three tillage methods: No-tillage, Strip-tillage, and Ridge tillage.
The references we mentioned are as listed below.
Carter M. CONSERVATION TILLAGE. J. Encyc of Soils in the Environ, 2005,306-311.
Morrison J. Strip tillage for “no–till” row crop production. J. Applied engineering in agriculture, 2002 ,18,3, 277.
Zhou J.;Khot L.; Boydston R et al. Low altitude remote sensing technologies for crop stress monitoring: A case study on spatial and temporal monitoring of irrigated pinto bean. J. Preci agric, 2018, 19,3, 555-569.
Sullivan D.; Lee D.; Beasley J et al. Evaluating a crop residue cover index for determining tillage regime in a cotton-corn-peanut rotation. J. J of soil and water conserva, 2008, 63,1, 28-36.
5)Lines 182-183: the sentence is not clear
Response: Thanks for the comments. As suggested, the authors has modified lines
207-209 as "All Sentinel-2 remote sensing images used in the study were derived from the European Space Agency (ESA) Sentinel Scientific Data Hub (https://scihub.copernicus.eu/dhus/#/home)."
6)Table 1: specify the units of measurement in "Spatial resolution"
Response: Thanks for the comments. As suggested, the authors marked in Table 1 that Spatial resolution is m
7)Lines 239-241: fix the sentence like this: “Currently, the NDTI using Sentinel-2 images can be calculated as follows:” eliminating: “based on the spectral band calculation can well estimate the CRC.”
Response: Thanks for the comments.As suggested, the authors have modified Lines 269-270 to"Currently, the NDTI using Sentinel-2 images can be calculated as follows:" eliminating: "based on the spectral band calculation can well estimate the CRC."
8)Line 255: Delete the sentence "This method is highly accurate."
Response:Thanks for the comments.As suggested,the authors have deleted the sentence "This method is highly accurate."
9)Line 264: Specify whether y is NTDI and x is CRC or vice versa
Response:Thanks for the comments.As suggested, the authors have specified y as CRC and X as NDTI in Line 293
10)Line 290: In my opinion chapter 2.4.5 could be dropped
Response:Thanks for the comments. the authors have been deleted 2.4.5
11)Figure 3: It is the same as Figure 4 on the right
Response:Thanks for the comments.As suggested, the authors have deleted Figure 3.
12)Table 4: it is not clear what “cord” means
Response:Thanks for the comments. As suggested, "Core" is the measuring rope used for measuring CRC. It has been described in Table 4.
13)Line451: see line 61
Response:Thanks for the comments. As suggested, the authors have explained in 3).
14)Conclusions: the reported sentences highlight that the paper could be much more useful if it were improved on some aspects: (a) more precise localization of the ESTARFM algorithm; (b) apply a CRC estimation model in spring; (c) study methods for monitoring conservation tillage over long time series.
Response:Thanks for the comments. As suggested, the authors have added, "the paper could be much more useful if it were improved on some aspects in the conclusion: (a) The original ESTARFM model was improved to find best model parameters that are more suitable for the study area, making the model more suitable for the prediction work of this study. (b) MRC sampling in the study area was conducted to construct a more accurate spring MRC estimation model using the measured MRC data collected in the spring. (c) Further exploration of remote sensing monitoring methods for conservation tillage on a larger scale and more extended time series.
Author Response File: Author Response.pdf
Reviewer 3 Report
Acquiring available high spatial and high temporal resolution remote sensing data is crucial for conservation tillage remote sensing monitoring. This study used high spatial resolution Sentinel-2 images, high temporal resolution MODIS remote sensing images, and ESTARFM to predict the maize NDTI for a specific date. Then the predicted NDTI is used to estimate the MRC in the study area through the MRC estimation model. Finally, the authors used the estimated MRC to identify the tillage practice in the study area. On this basis, the results of IB and BI prediction methods are compared. Overall, it is a valuable and well-constructed paper. Some specific questions are as follows:
(1) The part of the abstract lacks the concise viewpoint.
(2) Line13-14:Mark the abbreviation of Index then Blend and Blend then Index in the introduction.
(3) Line114:Change "5D" to “5 days”.
(4) Line140:Change "Sentinel-2 and MODIS data"to"Sentinel-2 and MODIS images".
(5) Line263:Please cite the reference following the guide of the author of remote sensing. Please revise the similar mistakes in your manuscript..
(6) Line300:Please adjust the formula font size to follow the guide of author of remote sensing.
(7) Line348:Please mark the value range of NDTI in Figure 2
(8) Please mark the value of RMSE in all Statistical relation Figures and revise the similar mistakes in your manuscript.
(9) Line386:Please add units for each parameters in Table 4.
(10) Line419:Please change "[45] [33]" to "[45, 33]",revise the similar mistakes in your manuscript.
(11) Line454:Please explain why the authors use this tillage practices discrimination in this study.
(12) Line471:Please provide relevant data or references to support your statement.
(13) The references need to be in right format. Please follow the guide of author of remote sensing.Check all the references in this manuscript.
Author Response
Dear editor and reviewer,    
Re: Manuscript reference: Remote Sensing-2081692
Classification of conservation tillage using enhanced spatial and temporal adaptive reflectance fusion model.
Firstly, we would like to thank the reviewer and editor for reviewing and evaluating our manuscript. We appreciate the opportunity to revise our manuscript. We would like to thank the reviewer and editor for their contribution, they have clearly put considerable effort into reviewing our work and their input has undoubtedly resulted in a stronger paper.
Please find attached a revised version of our manuscript “Classification of conservation tillage using enhanced spatial and temporal adaptive reflectance fusion model”, which we would like to resubmit for publication as a research article in the Remote sensing.
Your comments and those of the reviewers were highly insightful and enabled us to improve the quality of our manuscript greatly. In the following pages are our point-by-point responses to each of the comments of the reviewers as well as your own comments.
Revisions in the text are shown in revision mode. In accordance with the suggestions of the editor and reviewer, the authors have revised and added to this manuscript. The authors emphasize that the study was on maize and redefine conservation tillage according to the proceedings of the 1st World Congress on Conservation Agriculture held in Madrid in 2001. The authors explain the meaning of the abbreviation "NT." The authors have corrected or deleted inaccurate statements in the manuscript. The diagrams in the article were fixed. Details missing from the article were added. The applicability of the tillage pattern discriminant method chosen in this study was explained. In the conclusion section of the manuscript, the following research directions for this study are outlined. More convincing references are added to some of the ideas presented in this article. Finally, formatting adjustments were made to the references section of this article.
We shall look forward to hearing from you at your earliest convenience. 
Yours sincerely,
Authors
Responses to the comments of Reviewer #3
Reviewer #3:Acquiring available high spatial and high temporal resolution remote sensing data is crucial for conservation tillage remote sensing monitoring. This study used high spatial resolution Sentinel-2 images, high temporal resolution MODIS remote sensing images, and ESTARFM to predict the maize NDTI for a specific date. Then the predicted NDTI is used to estimate the MRC in the study area through the MRC estimation model. Finally, the authors used the estimated MRC to identify the tillage practice in the study area. On this basis, the results of IB and BI prediction methods are compared. Overall, it is a valuable and well-constructed paper.
1)The part of the abstract lacks the concise viewpoint.
Response: Thanks for the comments. The author added, "After comparison, the IB scheme is better than the BI scheme in prediction results and prediction efficiency." to the abstract.
2)Lines13-14:Mark the abbreviation of Index then Blend and Blend then Index in the introduction
Response: Thanks for the comments. As suggested, the authors marked "the Index then Blend(IB) and Blend then Index(BI)" in the abstract.
3)Line114:Change "5D" to “5 days”.
Response:Thanks for the comments. As suggested,the authors modified "5D" to "5 days" in Line 139.
4)Line140:Change "Sentinel-2 and MODIS data"to"Sentinel-2 and MODIS images".
Response:Thanks for the comments. As suggested,the authors changed "Sentinel-2 and MODIS data" to "Sentinel-2 and MODIS images" in Line 165.
5)Line263:Please cite the reference following the guide of the author of remote sensing. Please revise the similar mistakes in your manuscript.
Response:Thanks for the comments. As suggested,the authors have revised according to the requirements of remote sensing.
6)Line300:Please adjust the formula font size to follow the guide of author of remote sensing.
Response:Thanks for the comments. As suggested,the authors have deleted this section.
7)Line348:Please mark the value range of NDTI in Figure 2.
Response:Thanks for the comments. As suggested,the authors have marked the value range of NDTI in Figure 2.
8)Please mark the value of RMSE in all Statistical relation Figures and revise the similar mistakes in your manuscript.
Response:Thanks for the comments. As suggested,the authors have marked the RMSE of all Statistical relation Figures.
9)Line386:Please add units for each parameters in Table 4.
Response:Thanks for the comments. As suggested,the authors have marked the unit of Table 4, the unit is" %" .
10)Line419:Please change "[45] [33]" to "[45, 33]",revise the similar mistakes in your manuscript.
Response:Thanks for the comments. As suggested,the authors have changed "[45] [33]" to "[51, 39]" in line 424.
11)Line454:Please explain why the authors use this tillage practices discrimination in this study.
Response:Thanks for the comments. As suggested, the authors explained how this criterion for identifying tillage patterns can be used in Lines 459-462. Using the identification criteria for tillage patterns can provide a more detailed description of the spatial distribution of the various tillage patterns in the study area and is a sound basis for conservation tillage promotion and monitoring by the agricultural department.
12)Line471:Please provide relevant data or references to support your statement.
Response:Thanks for the comments. As suggested, the authors have marked the appropriate references in Line 481.
The references we mentioned are as listed below.
Liu, X. Study on the Problems and Countermeasures of Water Resources in Lishu County. J. Intelli. 2011, 25, 304-304.
13)The references need to be in right format. Please follow the guide of author of remote sensing.Check all the references in this manuscript.
Response:Thanks for the comments. As suggested, the authors have revised all references according to the remote sensing standard.
Author Response File: Author Response.pdf
Round 2
Reviewer 2 Report
The Authors responded comprehensively to all comments
