Distinguishing Algal Blooms from Aquatic Vegetation in Chinese Lakes Using Sentinel 2 Image

Round 1

Reviewer 1 Report

The authors failed to address some of the main concerns I identified in my initial review and these issues persist in the revised manuscript. The response to comments is fairly superficial and I don't think the revisions were as thorough as necessary to bring this manuscript closer to being publishable. I considered rejection initially and don't see any reason to revise that decision now.

Author Response

Based on the issues that you concerned, we respond them in the following four aspects:

1. Points of innovation. This paper extends the method of remote sensing monitoring of algal blooms in turbid inland waters, which does not rely on traditional measured data, and the model has good spatial portability, which provides strong methodological and theoretical support for future monitoring of algal blooms in turbid water bodies with vigorous aquatic vegetation (Page 2, Line 89-93). Furthermore, we have replotted some figures and revised the structure of the discussion section (Page 13, line 326-510). The discussion is now divided into four sections: The advantages of model for extracting vegetation information in turbid water, Validation with the absence of actual measurement data, Spatial transferability of the model, and Analysis of advantages and disadvantages.
2. Template and format. We apologize for the errors caused by carelessness and we have carefully checked and revised the formatting error.
3. Description of the method. We have rewritten the methods section, which was too long and complex to be easily understood, and have now modified it, such as the supplement of the morphometric information of the lakes in Table S1, removing some formulas for calculating the indices to the “supporting materials” (Page1, 5-60).
4. The wider application and significance of the method. To demonstrate the widespread use of the method, it has been validated and applied to two other lakes with good accuracy and results (Section 4.2, Section 4.3). And there are plans to apply it to more lakes for algal bloom monitoring in the future.

The article has been greatly improved in terms of objective description, innovative summary, method description, graphical presentation, structure and language of the discussion, etc. We hope that you will review it again carefully and give it the opportunity for further revision and publication.

Author Response File: Author Response.docx

Reviewer 2 Report

I can see some improvement of this article for the current version.

However, I still have several suggestions as follows:

1. The third objective of this study is clear. This content is much more like the method rather than an objective.
2. Figure 9 could be removed to Section 2.1, including a brief introduction. It’s not proper to have a map and study area introduction in the discussion part.
3. Section 2.3 could be simplified. The calculation of indices could be shown by citing some references.
4. Some abbreviations in Figures and Tables should be noted and double-checked. For example, NDWI3-5 in Figure 3. SA2VPF in Figure 5. Kappa in Table 1. If it’s new, the meaning should be added. Some abbreviations (Line 400-402) are not necessary for this paper and could be neglected.
5. The whole structure of the discussion are not clear now. I still suggest authors to modify the discussion part by focusing some major points about this topic. How about the performance of this method for turbid waters? How was the validation result with independent region? “et al. Section 4.3 "Comparison of results and error analysis" are too general”. Specific discussions are expected by focusing the main objective of this study.
6. I can't see Section 5 in the manuscript.

Author Response

I can see some improvement of this article for the current version. However, I still have several suggestions as follows:

1. The third objective of this study is clear. This content is much more like the method rather than an objective.

Response: Thank you for the constructive comment, and we totally agree with you. The objectives have been rewritten in the revised manuscript in Page 2, line84-89. “Specific objectives were: (1) realize the effective extraction of algal blooms and aquatic vegetation from the highly turbid water bodies; (2) establish a method to effectively distinguish algal blooms and aquatic vegetation with the absence of actual measurement data; (3) test the spatial transferability of the method in the independent lakes with the various optical properties and examine the spatial variations of algal blooms in these lakes.”

2. Figure 9 could be removed to Section 2.1, including a brief introduction. It’s not proper to have a map and study area introduction in the discussion part.

Response: Thanks to your suggestion, Figure 9 has been moved to Section 2.1, and merged with Figure 1 (Page 3, Figure 1).

Figure 1. Locations and morphology of lakes and corresponding climatic zones. The blue lakes are the five lakes used for modelling and the green ones represent the two lakes used for method validation.

3. Section 2.3 could be simplified. The calculation of indices could be shown by citing some references.

Response: We have simplified the section according to your suggestion. Some formulas for calculating the indices in this section have been removed to the “supporting materials”, and the references have been added. The specific steps of identifying algal blooms and aquatic vegetation are detailed in the “supporting materials”.

4. Some abbreviations in Figures and Tables should be noted and double-checked. For example, NDWI3-5 in Figure 3. SA2VPF in Figure 5. Kappa in Table 1. If it’s new, the meaning should be added. Some abbreviations (Line 400-402) are not necessary for this paper and could be neglected.

Response: Thank you for your suggestion. The abbreviations in the figure and table in the text have been rechecked and added the full meaning (Page 5, Figure 3; Page 6, Figure 4 and Page 10, Table 1). The abbreviations in lines 400-402 are not needed in this paper, have been deleted (Page 20, line 516-518).

5. The whole structure of the discussion are not clear now. I still suggest authors to modify the discussion part by focusing some major points about this topic. How about the performance of this method for turbid waters? How was the validation result with independent region? “et al. Section 4.3 "Comparison of results and error analysis" are too general”. Specific discussions are expected by focusing the main objective of this study.

Response: According to your suggestions, we have revised the discussion section (line 326-510). The discussion is now divided into four sections as follows.

4.1. The advantages of model for extracting vegetation information in turbid water

4.2. Validation with the absence of actual measurement data

4.3 Spatial transferability of the model

4.4. Analysis of advantages and disadvantages

6. I can't see Section 5 in the manuscript.

Response: The sixth section of the text should be the fifth, and we are sorry for the error caused by carelessness again.

Author Response File: Author Response.docx

Reviewer 3 Report

This paper deals with the important problem of the return of alga blooms in water bodies. The authors try to define a model allowing verification of such problem. In the introduction it is puzzling that there is no reference to the alternative steady state theory (e.g. Scheffer et al. 1993, Jeppesen. 1998). This is surprising as they describe a system with phytoplankton dominance (turbid water condition) vs. macrophyte dominance (clear water condition). Basing the ideas of the paper on well-known regularities is crucial and allows to better ground one's own research results in providing a new portion of knowledge. 

The goals of the work are far too ambitious. Suggest adjusting them to the data you have. The authors are not yet in a position to develop a progniostic tool analyzing habitats in different types of ecosystems. This requires a larger database. Besides, I do not know what the authors understand by the term "blooms"? This process concerns the activity of different groups of phytoplankton, but it is not always dangerous. In water reservoirs not subjected to human pressure these processes also occur. This aspect is not jansno presented. 

Some comments on the quality of the results presented:

The description of the lakes lacks basic morphometric information. Are these objects polymictic lakes, non-stratified ect. A long description of their features can undoubtedly be replaced by a table. 

Figure 5 nad 6 are illegible and besides, what do the letters mean? Think about how to present the results.

I am not convinced by the division into vegetation and water in the graphs and tables. The authors study vegetation and phytoplankton - water is only a medium at this point. 

Table 2 contains abbreviations that are not explained in the title. It should be self-explanatory. 

Figure 11 it is not clear what the letters on the maps mean. Some of them are used multiple times in the maps. This is largely unreadable. 

The search for new methods in environmental research should be appreciated. However, it is only an attempt, and the tools currently available cannot eliminate the need for field research. The authors completely ignore the taxonomic diversity of the blooms. A cyanobacterial bloom, a diatom or a green algal bloom different for reservoir management. If we could signal blooms of individual phytoplankton groups, it would be much more valuable. Consideration could also be given to the use of the proposed method in determining the displacement of stablized states in the context of restoration of degraded aquatic ecosystems.

Author Response

Reviewer 3:

1. This paper deals with the important problem of the return of alga blooms in water bodies. The authors try to define a model allowing verification of such problem. In the introduction it is puzzling that there is no reference to the alternative steady state theory (e.g. Scheffer et al. 1993, Jeppesen. 1998). This is surprising as they describe a system with phytoplankton dominance (turbid water condition) vs. macrophyte dominance (clear water condition). Basing the ideas of the paper on well-known regularities is crucial and allows to better ground one's own research results in providing a new portion of knowledge.

Response: We thank the reviewer for the positive comment, and we have carefully revised the manuscript according to the suggestion. As you mentioned, the references related to the alternative steady state theory (e.g. Scheffer et al. 1993, Jeppesen. 1998) are really important and crucial to the original intention of this research, and we have been cited them in the revised manuscript (Page 1, line 36).

2. The goals of the work are far too ambitious. Suggest adjusting them to the data you have. The authors are not yet in a position to develop a progniostic tool analyzing habitats in different types of ecosystems. This requires a larger database. Besides, I do not know what the authors understand by the term "blooms"? This process concerns the activity of different groups of phytoplankton, but it is not always dangerous. In water reservoirs not subjected to human pressure these processes also occur. This aspect is not jansno presented.

Response: The description of the target has also been modified in this paper (Page 2, line 84-89). The related text is “Specific objectives were: (1) realize the effective extraction of algal blooms and aquatic vegetation from the highly turbid water bodies; (2) establish a method to effectively distinguish algal blooms and aquatic vegetation with the absence of actual measurement data; (3) test the spatial transferability of the method in the independent lakes with the various optical properties and examine the spatial variations of algal blooms in these lakes.” As you mentioned, there are many different types of algal blooms with different algae filament, and not all of which are harmful to the environment, some of the language in this paper was inappropriate and has been revised. The FAI is designed to measure algae floating on the water surface when cyanobacteria form surface scum. The term 'algal bloom' in this paper therefore differs from the traditional sense of algal particles when suspended in the water column, and refers specifically to algal blooms that form scum on the water surface.

Some comments on the quality of the results presented:

• The description of the lakes lacks basic morphometric information. Are these objects polymictic lakes, non-stratified ect. A long description of their features can undoubtedly be replaced by a table.

Response: We have summarized the morphometric information on the lakes in Table S1 in the “supporting materials”.

Table S1. Basic information on seven lakes covered in the article

• Figure 5 and 6 are illegible and besides, what do the letters mean? Think about how to present the results.

Response: Thank you for your suggestion. We have replotted the figures and added the meaning of the letters and the content in the figure captions (Page 6, Figure 4 and Page 7, Figure 5).

Figure 4. VPF values during periods with large areas of stable aquatic vegetation for (a) Lake Hulun, (b) Lake Hongze, (c) Lake Chaohu, (d) Lake Taihu, and (e) Lake Dianchi in Sentinel 2 images. The redder the area in the figure the more frequently the vegetation signal has been present, the more probable the area is to be aquatic vegetation.

Figure 5. This figure shows the results of the identification of algal blooms and aquatic vegetation in the studied lakes on selected dates of algal outbreaks, using the modified VPF method. The red areas represent algal blooms and the green areas represent aquatic vegetation. (a)-(e) are the identification results of Lake Hulun, Hongze, Chaohu, Taihu and Dianchi, respectively. In the image HLH means Lake Hulun, HZH means Lake Hongze, CH means Lake Chaohu, TH means Lake Taihu and DC means Lake Dianchi. HLH-17 Jul 2019 means the image identification result of Lake Hulun on 17 July 2019. The numbers and letters following the abbreviations of the other lakes are the dates of the images.

• I am not convinced by the division into vegetation and water in the graphs and tables. The authors study vegetation and phytoplankton - water is only a medium at this point.

Response: The identification results for all lakes are validated for precision and compared with the results of other studies in section 4.2 and section 4.3. In general, the method was found to be accurate and effective in the identification of algal blooms and aquatic vegetation in the selected lakes.

• Table 2 contains abbreviations that are not explained in the title. It should be self-explanatory.

Response: The meaning of the abbreviations in Table 2 were added in the revised manuscript.

Table 2. Validation points number and the calculated accuracy results for the extent of aquatic vegetation and the extent of non-aquatic vegetation buffered outwards for lakes.

P_V represents the accuracy within the range of aquatic vegetation.

P_W represents precision within the buffer zone outside the aquatic vegetation range.

T represents the point verified as correct, i.e. he extraction of points within the aquatic vegetation range where it was verified that aquatic vegetation was growing or within the buffer zone outside the aquatic vegetation range where it was verified that no aquatic vegetation was growing.

F represents points verified as incorrect, i.e. the points within the aquatic vegetation range without significant vegetation growth in the area or within the buffer zone outside the aquatic vegetation range where the occurrence of aquatic vegetation was verified.

Pn represents the ac-curacy of aquatic vegetation extraction for lake n.

• Figure 11 it is not clear what the letters on the maps mean. Some of them are used multiple times in the maps. This is largely unreadable.

Response: Each of the lakes in Figure 11 has five letters from (a)-(e) in different sites, and the five small maps (a-e) matched with the letters in the lake are the images used for verification on Google Maps. We have corrected the titles and abbreviations of the figures and tables in the text to make them more readable.

Figure 11. Distribution of validation points and validation results in non-modeled Lakes, (1) TCP-Taipingchi Reservior, (2) CHH-Lake Chenghai. Five points have been selected in different areas of each lake and their positions in the lake are marked by the letters a-e. The corresponding Google Maps images used for the verification and the results of the verification are shown in the five sub-images (a)-(e). The date of the image used for verification can be seen in the bottom right corner of each sub-images. Aquatic vegetation-T/F are points within the range of aquatic vegetation. Water-T/F are points within the buffer zone outside the range of aquatic vegetation. T is the point verified as correct. F is points verified as incorrect.

• The search for new methods in environmental research should be appreciated. However, it is only an attempt, and the tools currently available cannot eliminate the need for field research. The authors completely ignore the taxonomic diversity of the blooms. A cyanobacterial bloom, a diatom or a green algal bloom different for reservoir management. If we could signal blooms of individual phytoplankton groups, it would be much more valuable. Consideration could also be given to the use of the proposed method in determining the displacement of stablized states in the context of restoration of degraded aquatic ecosystems.

Response: The algal blooms identified in this study included all algal outbreaks that formed scum on the surface and did not differentiate between different species of algal blooms. This allows the model to be applied on a larger scale.

Author Response File: Author Response.docx

Round 2

Reviewer 2 Report

Major revisions have been conducted for the manuscript for two times. We can see the structure are becoming much clearer. However, I’m sorry to see that there were not obvious revisions in the abstracts and in the conclusion. I think these two parts could be improved considering the revisions rather than listing some data. I recommend authors to check the whole manuscript seriously. There are still some obvious mistakes for the English writing. For example, “Each index was taked a strictly value to ensure 19 the correct classification of water surfaces with algal blooms” on line 19-20. “The image time series information was then combined ..….” on line 22-23.”  English expression could be improved.

Some concise and clear discussions about the performance of this VPF method in the revised version are expected. The main results of discussion are suggested be added in the conclusion part.

Author Response

Response: Thanks to your suggestion and comments. We have carefully rewritten the abstract and conclusion sections. We are very sorry for the poor English, and we consulted a professional language company to edit the English language of this article. The certificate of the language edit company is displayed here.

The related text of abstract and conclusion is in Page 1, Line 13-35 and Page 18, Line 503-516. “Abstract: Algal blooms frequently occur in numerous lakes in China, risking human health and the environment. In contrast, aquatic vegetation contributes to water purification. Due to the similar spectral characteristics shared by algal and aquatic vegetation, both are hardly distinguishable in remote sensing image, especially in turbid water bodies. To address this challenge, this study constructed a method to effectively extract algal blooms and aquatic vegetation from the turbid water bodies using Sentinel 2 image with high spatial resolution. Our results showed that the accuracy of the extraction of vegetation information could reach 96.1 %. Since this method combined the vegetation extraction results from multiple indices, it effectively tackled the mis-extraction when only the Floating Algae Index (FAI) or the Normalized Difference Vegetation Index (NDVI) is used in water with high turbidity. By combining the image time series information with the natural phenological characteristics of the aquatic vegetation and algal blooms, an im-proved Vegetation Presence Frequency (VPF) was developed. It effectively distinguished algal blooms and aquatic vegetation without actual measurement data. Based on the above method and process, the information of algal blooms and aquatic vegetation was sufficiently distinguished in five typical lakes in China (Lake Hulun, Lake Hongze, Lake Chaohu, Lake Taihu and Lake Dianchi), and the spatial distribution was reasonably mapped. The overall identification accuracy of aquatic vegetation and algal blooms using the improved VPF ranged 71.8%-84.3%. The spatial transfer-ability test of the method in the independent lakes with the various optical properties indicated the prospects of its application in other turbid water bodies. This study should provide strong methodological and theoretical support for future monitoring of algal blooms in turbid water bodies with vigorous aquatic vegetation, especially in the absence of actual measurement data. This should have practical relevance for water environment management and governance department.

5. Conclusions

Based on multiple indices and the modified vegetation presence frequency (VPF), this study accurately extracted and distinguished algal bloom and submerged vegetation us-ing Sentinel 2 images (without considering field measured data). This method effectively addressed the interference of turbid water bodies. The spatial transferability of the method was also verified in the other independent lakes with the satisfactory accuracy. This indi-cates the prospects of its general application to distinguish algal blooms from aquatic vegetation in turbid water bodies under stable water levels and adequate satellite images.

Overall, the method developed here can effectively differentiate algal blooms from aquatic vegetation with good stability, and it can avoid interference from thin clouds and other factors. Therefore, it might be feasible for large-scale classification and identification of aquatic vegetation and algal blooms. It might also provide a reference for distinguishing other features with similar spectral characteristics.

Author Response File: Author Response.docx

Reviewer 3 Report

Thank you very much for your fruitful cooperation. I would like to congratulate you on a good job. However, I would very much request you to verify the formatting of the literature according to the requirements of the journal.

Author Response

Response: Thanks to your suggestion. We have modified and adapted the reference formatting to the requirements of Remote Sensing by EndNote software. Now, we have guaranteed a uniform format of the references.

Author Response File: Author Response.docx

Round 3

Reviewer 2 Report

I think this manuscript could be considered for publication.

This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.

Round 1

Reviewer 1 Report

It’s an interesting topic to distinguish algal blooms from aquatic vegetation by using Sentinel 2 imagery with high spatial resolution. This study developed a method by combining multiple indices to provide accurate information on the extracted area with algal blooms. VPF modeling of vegetation presence frequency differences between algal blooms and aquatic vegetation were used for separating both. Advantage and disadvantage of this method was also discussed.

I think this manuscript could be improved with quantitative analysis and deep discussion by considering  some key points as the author claimed in the introduction. There are some minor mistakes which should be checked over the whole manuscript.

Some specific points are as follows:

1. Introduction: As claimed in Lines 57-59, existing methods are not universal enough. So how about the model in this study? Is it universal? Or is it related to the main object of this study? (Line 75-80)
2. Introduction: Line 61-Line 74. It’s not clear to see the main problem to distinguish aquatic vegetation from algal bloom according to existing studies. The meanings of some sentences are not easy to understand. I suggest authors to clarify this part.
3. Do the species of vegetation change over time in different lakes? Will it affect the threshold of VPF?
4. Line 298-299. Some of the extracted results for the lakes are shown in Fig.4? or Fig. 5? I suggest to go through the quotation of Figures and Tables for the whole manuscript
5. Line 323. “the resulting confusion matrix is shown in Table 3?” or Table 1?
6. Figure 6 are not clear enough. This detailed information is quite useful. Enlarged figures with clear notifications are expected. The mark’s color could be changed to distinguish them from the background.
7. Discussion 4.1. Superiority of the decision tree over the single FAI index was shown in this part by comparing lake Taihu and lake Hulun with day images. Turbidity was shown with false-color images rather than SPM data. This will make the discussion more superficial. I suggest a deep discussion about the effect of turbidity over the five lakes to confirm the advantage of the decision tree.
8. Discussion 4.2. In this part, the advantages and disadvantages were listed by comparing this study with that of Liu et al. (2015). The VPF thresholds are different in these two studies. Even though the authors stated that the method in this study could work effectively, we also want to know the different performances with two methods. Discussions with quantitative comparisons are expected here.

Some minor points:

1. Please specify the name of lakes, such as Lake Tai, or Lake Taihu?
2. Section 3.1.3, Line 283. Do you mean Figure 3 or Figure 4. VPF distribution was shown in Figure 4.
3. Line 359-362: What does Pu stand for?
4. Figure 7. What does CT mean?
5. English editing is suggested to improve the language.

Reviewer 2 Report

This manuscript presents an approach for distinguishing potentially harmful algal blooms from submerged aquatic vegetation, which is a challenging and important topic amenable to remote sensing. However, this paper offers very little innovation toward this end and is deficient in several respects. The writing quality is poor and sloppy in many places (a portion of the manuscript template is included in the manuscript, indicating that not even a cursory proofread was done). The use of visual interpretation of images as a means of accuracy assessment is subjective and unreliable. The methodology is not explained clearly, as the decision tree is not presented and the variable selection criteria are ambiguous. The accuracy assessment does not follow accepted practices and terminology and is thus difficult to understand. The results focus on five lakes in China but do not include any discussion of the broader applicability and significance of the approach. Due to these shortcomings, I'm afraid I must conclude that the paper is not suitable for publication in Remote Sensing.