High-Resolution Ratoon Rice Monitoring under Cloudy Conditions with Fused Time-Series Optical Dataset and Threshold Model

Round 1

Reviewer 1 Report

It is extremely meaningful to monitor the ratoon rice using remote sensing have information on the planting of seed corn. The authors propose a new method for ratoon rice extraction. It is well written, concise, scientifically sound, with great potential to support decision-making processes. The Materials and Methods section describes both the data used and the methodology used comprehensively. The results are significant and a discussion with the published articles shows their importance and validity.

However, there are some questions in the paper, as follows:

1.      The curves in Figures 4 and 5 do not reflect the differences with other crops. What are the specific characteristics of the differences with other crops?

 

2.      Chongqing is a mountainous area and the crop planting plots are very fragmented. The extraction of crop cultivation area in mountainous areas has been a difficult problem. Can the crop area be effectively monitored using ten-meter images in this paper?

English language quality can be improved with minor modifications

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

I guess this is an exciting paper. They challenge two big targets: a unique rice cropping (ratoon rice) and a problematic climate (very cloudy). However, reading this paper was not easy or comfortable because of small grammar mistakes (making precise understanding difficult) and the lack of details in the explanation (especially the figure captions). I mostly stopped reading the paper in the middle because I did not want to continue further "guess what" games in the review. I like the authors carefully review the paper and make it more readable. Then I am happy to review it again!

The followings are my detailed comments:

More explanation would be necessary for each figure.

equivalent to 116 million tons -> equivalent to 116 million tons per year

that combines low-cost labor ... "low cost labor" is using cheap workers. Not appropriate, perhaps. Something like "cost-saving"?

"Since the phenological changes of ratoon rice are similar to those of traditional double season rice" ... I don't think so. According to your data, the second rice stage is much shorter in the case of ratoon rice.

"The accuracy is 0.76" ... What is the accuracy?

L110-L120: Which study are you mentioning here? Your study or somebody's study?

"The main planting system in Yongchuan District is ”middle rice-ratoon rice”, rendering the region conducive for investigating the spatial distribution of ratoon rice." ... Not understandable.

Fig 1: What happens in June F, July F, July M, etc? There are no illustrations on the bottom.

L210-L212: Something wrong.

The cultivation of ratoon rice was relatively scattered. ... Is it about North China?

Is T1 un L280 the same as t1 in equation (1)?

L285 "reduce" ... not understandable.

L330~331: "variance" ... perhaps a standard deviation.

"9 conditions are then matched with each other" ... not understandable.

L338: "Compared the threshold with the phenological calendar." ... not understandable in the context.

As shown in Figure 5, the image in the key period when NDVI reaches its peak is MOD09GA, which is the data fused through the FSDAF model. ... not easy to understand. Please add more explanation to Figure 5.

Figure 5: How did you make a single line for each NDVI and LSWI? It represents the entire target area, right?

Figure 6: What is r? Correlation coefficient? Between what?

Figure 8 and 9: What is the meaning of the colors?

Figure 11: Not easy to tell which line is which. Please use multiple identifiers (such as line patterns) to help readers distinguish the lines.

Sorry for being late.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Round 2

Reviewer 2 Report

The new manuscript was much easier to read and understand than the first one. But still, I got many questions that exhausted me and finally stopped me keep reading. I again request the authors write more precisely and concisely, helping the readers keep reading until the end. The followings are the questions I came up with again:

1. (Abstract) with high r values ... What is "r value"? If it is correlation coefficient, it is better to write so, or use italic r.
2. (Abstract) with high r values and low RMSE values. ... Between what and what?
3. Specifically, the identified area of ratoon rice in the study area was 194.17 km2, which was close to the official data (158-180 km2), with an overall accuracy and a kappa coefficient of 90.73% and 0.81, respectively. ... OA and kappa of area estimation (194.17 km2)? It sounds strange.
4. “mid-season rice-ratoon rice” ... Does it mean “mid-season rice" and "ratoon rice”? Or “mid-season" and "ratoon rice”? By simply thinking, it appears the latter case. However, if we consider the context, the former may be the case.
5. MODIS Repro-jection Tool -> MODIS Re-projection Tool
6. band synthesis ...? What do you mean? Do you mean combining Band 3 and Band 4 into a single virtual band, for example?
7. Please mention about the projection (Coordinate Reference System = CRS) you used.
5. there was a relatively scattered situation of ratoon rice planting ... What is scattered? Rice stubbles with ratoon in a rice field? Or rice ratoon planting fields in the Yongchuan District?
6. determine the threshold of ratoon rice, including flooding signals and key dates of ratoon rice growth period. ... I cannot understand what the "threshold method" means. I know a threshold. But "threshold method" does not give me a clear idea about how you use a threshold.
7. The cloud pixel values were transformed into binary form, aligning with the indicators provided in the user manual, in order to detect and label cloud pixels. ... Maybe "The QA pixel values were ...".
8. This model was deployed to identify ratoon rice fields and area within regions featuring complex crop cultivation distributions. ... I have been puzzled about the meaning of "area" because the word has two meanings. Which do you mean, a size (something quatified in square meters) or region (something described by delineating a map)?
9. harvesting reserved stubble stage ... I cannot understand.
10. The NDVI and LSWI curves ... Not curves but time series?
11. the increase in ratoon rice biomass gradually covered the flooding signal ... I do not understand "cover". Do you mean obscure? Is "ratoon rice biomass" true in this sentence? I wonder because you are discussing here what is happening in the first rice, not ratoon rice stage.
12. “Double peaks” and “double troughs” between the heading stage and the harvest of the ratoon rice were captured. ... In what signal? NDVI? LSWI? Both?
13. Equation 2: Is this condition "All" or "Any"? I mean, do you require it in the entire period of DOY 90~110? Or do you require at least only one time (not all) in that period?
14. the peaks and troughs and their dates of occurrence ... Peaks and troughs of what? The latter sentence probably means NDVI only. If it's true, please help readers know it is about NDVI only, not LSWI.
15. T1 and T2 are the lower and upper limits of the LBP, ... Is it the same or different from T1 and T2 in the explanation of the FSDAF model? Perhaps different. Then it's better to declare so or use different symbols so as to avoid collisions.
16. Eq. (7)~(10): You said variance (not standard deviation). But it is strange because variance has a dimension of day^2. It is unreasonable to add it to LTP, which has a dimension of day. Besides, sigma is conventionally used for describing standard deviation rather than variance. Using sigma for variance is misleading.
17. Eq. (11) and (12): What is a unit for these values? Days? If so, please describe. You may think it is clear, but you cannot guarantee it if you skip it.
18. Thus, the application of the FSDAF model is warranted to enhance the spatial resolution. ... It is not straightforward to agree with this. Maybe "not warranted"?
19. The above is the designed ratoon rice threshold model. ... I do not understand.
20. which were and overlaid with the ... Strange. What is the meaning of "and"?
(I could not keep reading further.)

You may think you have answered to some of them already, but it is not reflected enough in the revised manuscript.

Author Response

Please see the attachment.

Author Response File: Author Response.docx