Round 1

Reviewer 1 Report

Review of Cartwright et al.

“Estimation of Chlorophyll-a Concentration with Remotely Sensed Data for the Nine Plateau Lakes in Yunnan Province”

Remote Sensing

September 26, 2022

 

This study introduces a methodology for retrieving chlorophyll and temperature using RF. Overall, this paper presented topics, objectives and results quite well. Authors also clearly addressed those research questions. As remote sensing techniques becomes more accurate and vast amount of data from various satellite platforms become available to cover even alpine lakes, studies of data analyses will certainly provide new insights and approaches that help improve decision making processes in more evidence-based ways. With that this manuscript will be a valuable input to the community. I recommend for publication of this manuscript after minor revisions.

 

Some minor comments and suggestion include:

 

1. Change “in situ” to italic font in the text;

2. L59-60: Is this correct citation? I don’t find any information about Gini Index in the literature;

3. L114: What is the criteria of selecting “cloud free” images? Any metrics used for this?;

4. L139: What is the criteria for removing outliers? How did you do QC (quality control)?;

5. L214: VIM was not spelled out before this abbreviation;

6. L184, L191, L198, L204: What are these equations 2,3,4,5 used for? Please, elaborate in the text about what authors intend to use these for. Or, you can refer to these equations at the L340 to remind readers of what these equations are used for;

7. I cannot quite follow what the Table 4 tells. What does “model” represent? How were they derived? From where? I think this table is crucial for readers to follow the methodology authors used. Please, provide reader-friendly information in the text;

8. L356: It is an incomplete sentence;

9. L413: has → have;

10. L434: The temperature dependent relationship and predictability works only for this temperature range. What is the annual temperature pattern for those studied lakes? Any implications of this limited temperature-chlorophyll relationship in the water quality managements? This can be added as further discussion.

Author Response

Response to reviewer 1 comments

Dear Reviewer,

We deeply appreciate the effort and time you’ve spent in reviewing our manuscript “Estimation of Chlorophyll-a Concentration with Remotely Sensed Data for the Nine Plateau Lakes in Yunnan Province” (ID: remotesensing-1929660). We studied the comments carefully and have made revisions which we hope meet with your approval. The outline of the revisions is listed as follows:

• We reorganize the ideas and focus the potential evaluation of lakes eutrophication by estimation of chlorophyll-a concentration with remotely sensed data. (Please check the Abstract section)
• We have revised unclear descriptions and corrects references to the literature, and added a description of the sixth part of the manuscript. (Please check the Introduction section)
• We have added indicators of lake eutrophication in China, deleted redundant Landsat-8 data presentations, and revised several ambiguous statements. (Please check Materials section)
• We have corrected the misstatements and spelled the abbreviations. (Please check Methods Section)
• We have improved the interpretation of Table 4 and improved the description of errors. (Please check Result section)
• We have added a discussion of the limited temperature-chlorophyll-a relationship. (Please check Discussion section)
• We have checked the experimental results of the whole manuscript carefully.
• We have checked and revised the image format of the entire manuscript carefully.
• We have revised the entire manuscript, revising some issues with terminology meticulously.

Comment #1

Change “in situ” to italic font in the text.

Response:

Thank you for helpful comment. We have changed "in situ" to italic font in the text. There are 11 "in situ" places in the manuscript. Please refer to line 17, 21, 68, 132, 136, 240, 339, 356, 472 and 510.

Comment #2

L59-60: Is this correct citation? I don’t find any information about Gini Index in the literature.

Response:

Thanks for your helpful comment. We have re-cited the correct references: An experimental study of the intrinsic stability of random forest variable importance measures. Doi: 10.1186/s12859-016-0900-5. Please refer to line 547-548 on page 18.

Comment #3

L114: What is the criteria of selecting “cloud free” images? Any metrics used for this?

Response:

Thanks for your helpful comment. The specific index for controlling the number of clouds is to select remote sensing images with cloud cover below 10%. The study area is Nine Plateau Lakes, and the area is not extensive, so remote sensing images with clear skies over the lakes and no cloud cover are selected as far as possible, which is very important for retrieving land surface temperature. We have added cloud number control metrics in the manuscript. Please refer to line 113-114 on page 3.

Comment #4

L139: What is the criteria for removing outliers? How did you do QC (quality control)?

Response:

Thanks for your helpful minor comment. We performed quality control on the measured data by calculating the mean and standard deviation of chlorophyll-a in a month and We consider outliers points that are three standard deviations from the mean and then remove it. We have added to the manuscript how to do quality control. Please refer to line 139-140 on page 4.

Comment #5

L214: VIM was not spelled out before this abbreviation.

Response:

Thanks for your helpful comment. We have spelled the full name of VIM when it first appears.

Please refer to line 216 on page 6.

Comment #6

L184, L191, L198, L204: What are these equations 2,3,4,5 used for? Please, elaborate in the text about what authors intend to use these for. Or, you can refer to these equations at the L340 to remind readers of what these equations are used for.

Response:

Thanks for your helpful comment. we refer to 2,3,4,5 equations in line 345 to remind readers of what these equations are used for. Please refer to line 344-345 on page 9.

Comment #7

I cannot quite follow what the Table 4 tells. What does “model” represent? How were they derived? From where? I think this table is crucial for readers to follow the methodology authors used. Please, provide reader-friendly information in the text.

Response:

Thanks for your helpful comment. The "Model" is the selected band and band combination, which we named "Index Model" in the manuscript. The Index Model is based on the spectral features of the Plateau Lakes, Because the reflection characteristics of Plateau Lakes are mainly located in the visible to near-infrared band multispectral band and the red-edge band of Sentinel-2 were selected for chlorophyll-a concentration retrieval in this study. We have changed Table 4 "Model" to "Index Model" and we have added a way to select the Index model. Please refer to line 339-345 on page 9.

Comment #8

L356: It is an incomplete sentence.

Response:

Thanks for your helpful comment. We have reworked the sentence. Please refer to line 362-363 on page 11.

Comment #9

L413: has → have.

Response:

Thanks for your helpful comment. We have changed the grammatically incorrect "has" to "have". Please refer to line 421 on page 14.

Comment #10

The temperature dependent relationship and predictability works only for this temperature range. What is the annual temperature pattern for those studied lakes? Any implications of this limited temperature-chlorophyll relationship in the water quality managements? This can be added as further discussion.

Response:

Thanks for your helpful comment. According to the relevant knowledge, the temperature of the plateau lake is greatly affected by the subtropical plateau monsoon climate, and the annual average temperature of the lake is about 16°C. However, I am sorry that we have not conducted research on the average monthly temperature of plateau lakes. Usually, summer begins when the average temperature stabilizes above 22℃. The temperature retrieval data selected in this study is June 2022, which is one of the highest summer temperature months in Yunnan Province. Limited by the time range of measured chlorophyll-a concentration data and remote sensing satellite imagery data, the results of this study can only be used during the period of summer temperature rise, which provides a little convenience to study the correlation between the two. We plan to obtain more chlorophyll-a data in future studies and explore the temperature pattern of plateau Lake in different months, and correlation between temperature and chlorophyll-a in different months. We have added a discussion of the limited temperature-chlorophyll-a relationship in the water quality managements in manuscript. Please refer to line 457-465 on page 16.

 

 

Author Response File: Author Response.pdf

Reviewer 2 Report

The authors presented  the selection method of retrieval models of chlorophyll a concentration for Plateau lakes in Yunnan Province and made a comparison of the different models. This study provides a feasible means for monitoring the concentration of chlorophyll-a in plateau lakes and contributes to the scientific and effective management of plateau lakes. 

In this paper, the Sentinel-2 data is used for chlorophyll inversion, and the thermal infrared band (band 10) of Landsat-8 satellite is used for surface temperature inversion. Therefore, in Section 2.2, there is no need to introduce the development history of landsat and its payload in detail, just briefly introduce the two thermal infrared bands of landsat-8. 

In Table 4, there is little difference between the feature importance of different bands or models. What is the sensitivity of model selection to the feature importance? Please give a more clear explanation.

Minor comments:

1) P.2, Line 77: "...and sixth part is summarized ". This sentence is not clear, please rewrite it. 

2) Line 114: Please give the full name of USGS when it first appears. 

3) Line 119: "reentry period of 10 days" should be replaced with "revisit period of 10 days".

4) Line 137: What does “defective” mean here？Is it missing or poor precision? Please clarify.

5) In Table 3, what does "Amount" mean here? Does it refer to the number of sampling points?

6) Line 154-155: "...was based on the near-infrared band replaced by shortwave infrared band,..." had better be replaced with "...was based on the shortwave infrared band instead of near-infrared band,..."

Author Response

Response to reviewer 2 comments

Dear Reviewer,

We deeply appreciate the effort and time you’ve spent in reviewing our manuscript “Estimation of Chlorophyll-a Concentration with Remotely Sensed Data for the Nine Plateau Lakes in Yunnan Province” (ID: remotesensing-1929660). We studied the comments carefully and have made revisions which we hope meet with your approval. The outline of the revisions is listed as follows:

• We reorganize the ideas and focus the potential evaluation of lakes eutrophication by estimation of chlorophyll-a concentration with remotely sensed data.  (Please check the Abstract section)
• We have revised unclear descriptions and corrects references to the literature, and added a description of the sixth part of the manuscript. (Please check the Introduction section)
• We have added indicators of lake eutrophication in China, deleted redundant Landsat-8 data presentations, and revised several ambiguous statements. (Please check Materials section)
• We have corrected the misstatements and spelled the abbreviations. (Please check Methods Section)
• We have improved the interpretation of Table 4 and improved the description of errors. (Please check Result section)
• We have added a discussion of the limited temperature-chlorophyll-a relationship. (Please check Discussion section)
• We have checked the experimental results of the whole manuscript carefully.
• We have checked and revised the image format of the entire manuscript carefully.
• We have revised the entire manuscript, revising some issues with terminology meticulously.

Comment #1

In Section 2.2, there is no need to introduce the development history of Landsat and its payload in detail, just briefly introduce the two thermal infrared bands of landsat-8.

Response:

Thank you for helpful comment. We have reedited the paragraph to remove the development history of Landsat and its payload and briefly introduce the two thermal infrared bands of Landsat-8. Please refer to lines 106 on page 3.

Comment #2

In Table 4, there is little difference between the feature importance of different bands or models. What is the sensitivity of model selection to the feature importance? Please give a more clear explanation.

Response:

Thanks for your helpful comment. We have added the reasons for the small difference in feature importance: We based on the spectral analysis (in Section 3.2) and screened the band combinations that may be strongly correlated with chlorophyll-a and selected these band combinations for experiments, thus the result of the experiment characteristics of important degree of difference is small. On this basis, four band combinations are selected for comparison, which can well select the optimal chlorophyll-a retrieval model. Sensitivity of all importance features, we have not yet conducted research, we think this is a very good idea, the next will be studied from this aspect, thank you very much for your comments. Please refer to line 344-347 on page 9.

Comment #3

P.2, Line 77: "...and sixth part is summarized ". This sentence is not clear, please rewrite it.

Response:

Thanks for your helpful comment. We have added the explanation of this sentence: The sixth part summarizes the selection of chlorophyll-a retrieval model; the retrieval results of chlorophyll-a concentration in nine plateau lakes; and the correlation be-tween lake surface temperature and chlorophyll-a concentration. Please refer to line 77-80 on page 2.

Comment #4

Line 114: Please give the full name of USGS when it first appears.

Response:

Thanks for your helpful minor comment. We have added the full name of USGS when it first appears. Please refer to line 113 on page 3.

Comment #5

Line 119: "reentry period of 10 days" should be replaced with "revisit period of 10 days".

Response:

Thanks for your helpful minor comment. We have "reentry period of 10 days" replaced with "revisit period of 10 days". Please refer to line 119 on page 3.

Comment #6

Line 137: What does “defective” mean here？Is it missing or poor precision? Please clarify.

Response:

Thanks for your helpful comment. We re-edited the sentence. "Defective" mean here is the in situ measured of chlorophyll-a concentration in some lakes were missing or the available data are less. Please refer to line 136-137 on page 4.

Comment #7

In Table 3, what does "Amount" mean here? Does it refer to the number of sampling points?

Response:

Thanks for your helpful comment. “Amount” indicates the number of sampling points. We corrected the mistake here. Please refer to Table 3 on page 4.

Comment #8

Line 154-155: "...was based on the near-infrared band replaced by shortwave infrared band..." had better be replaced with "...was based on the shortwave infrared band instead of near-infrared band..."

Response:

Thanks for your helpful comment. We are very grateful and have taken your advice to replace the sentence "...was based on the near-infrared band replaced by shortwave infrared band..." with "...was based on the shortwave infrared band instead of near-infrared band...". Please refer to line 156-157 on page 5.

Author Response File: Author Response.pdf

Reviewer 3 Report

Dear Authors,

Line 12 and Line 28-29: The evaluation of lakes eutrophication have relation to effective management, but the water quality monitoring have relation with pressures and drivers from external human uses.We suggest to reorganize the ideas and focus the potential evaluation of lakes eutrophication by estimation of chlorophyll-a concentration with remotely sensed data.

Line 40-42:  I strongly recommended changing the sentence ”the improvement of water quality parameter retrieval model, remote sensing images can accurately”, because, the water quality parameter represented with more accurately quality information.  

Line 91-94: It is important to mention the scales for water quality standards in China and uses per water quality standards. 

Line 135-140:The chlorophyll-a concentration was measured, but the table did not mention the units for the chlorophyll-a measurements. I strongly recommended including those units.

Line 154-157:We suggested that the authors explain why the MNDWI represent the improved characteristic of open waters, instead of improving the characteristic in The Lake waters.

Author Response

Response to reviewer 3 comments

Dear Reviewer,

We deeply appreciate the effort and time you’ve spent in reviewing our manuscript “Estimation of Chlorophyll-a Concentration with Remotely Sensed Data for the Nine Plateau Lakes in Yunnan Province” (ID: remotesensing-1929660). We studied the comments carefully and have made revisions which we hope meet with your approval. The outline of the revisions is listed as follows:

• We reorganize the ideas and focus the potential evaluation of lakes eutrophication by estimation of chlorophyll-a concentration with remotely sensed data.  (Please check the Abstract section)
• We have revised unclear descriptions and corrects references to the literature, and added a description of the sixth part of the manuscript. (Please check the Introduction section)
• We have added indicators of lake eutrophication in China, deleted redundant Landsat-8 data presentations, and revised several ambiguous statements. (Please check Materials section)
• We have corrected the misstatements and spelled the abbreviations. (Please check Methods Section)
• We have improved the interpretation of Table 4 and improved the description of errors. (Please check Result section)
• We have added a discussion of the limited temperature-chlorophyll-a relationship. (Please check Discussion section)
• We have checked the experimental results of the whole manuscript carefully.
• We have checked and revised the image format of the entire manuscript carefully.
• We have revised the entire manuscript, revising some issues with terminology meticulously.

Comment #1

Line 12 and Line 28-29: The evaluation of lakes eutrophication have relation to effective management, but the water quality monitoring have relation with pressures and drivers from external human uses. We suggest to reorganize the ideas and focus the potential evaluation of lakes eutrophication by estimation of chlorophyll-a concentration with remotely sensed data.    

Response:

Thank you for helpful comment. We have refocused our thinking to focus on the potential assessment of lake eutrophication by estimating chlorophyll-a concentrations from remote sensing data. We re-edited the Line12 to: The quantitative retrieval of the chlorophyll-a concentration is an important remote sensing method used to monitoring the nutritional status of water bodies; We re-edited the Line28-29 to: These results will provide support for remote sensing monitoring of eutrophication in plateau lakes.

Comment #2

Line 40-42: I strongly recommended changing the sentence” the improvement of water quality parameter retrieval model, remote sensing images can accurately”, because, the water quality parameter represented with more accurately quality information.

Response:

Thanks for your helpful comment. Chlorophyll-a is one of the water quality parameters, because the focus of this paper is chlorophyll-a concentration inversion and model construction We have described the water quality parameter inversion model as chlorophyll-a inversion model more precisely. Please refer to line 40-42 on page 1.

Comment #3

Line 91-94: It is important to mention the scales for water quality standards in China and uses per water quality standards.

Response:

Thanks for your helpful comment. The information released by the Department of Ecology and Environment of Yunnan Province does not make it clear whether the water quality rating is based on a certain indicator. We have changed this paragraph. We investigated the eutrophication status of plateau lakes in Yunnan Province and mention the evaluation criteria of lake eutrophication in China. Please refer to line 93-96 on page 2.

Comment #4

Line 135-140: The chlorophyll-a concentration was measured, but the table did not mention the units for the chlorophyll-a measurements. I strongly recommended including those units.

Response:

Thanks for your helpful comment. The in situ measured chlorophyll-a unit is mg/L We have added a unit of in situ measured chlorophyll-a to the Tables footer position. Please refer to Table 3 on page 4.

Comment #5

Line 154-157: We suggested that the authors explain why the MNDWI represent the improved characteristic of open waters, instead of improving the characteristic in The Lake waters.

Response:

Thanks for your helpful comment. The study area is the typical Lake waters area. The buildings, vegetation and soil information near the Lake waters area. MNDWI can effectively suppress the building and soil information near the lake, and greatly reduce the background noise. We made a mistake in editing the article, now The Open waters was changed to The Lake waters. Please refer to line 157-158 on page 5.

Author Response File: Author Response.pdf