Framework to Extract Extreme Phytoplankton Bloom Events with Remote Sensing Datasets: A Case Study

Round 1

Reviewer 1 Report

Review of “Framework to Extract Phytoplankton Bloom Events with Remote Sensing Datasets: A Case Study in the Northern South China Sea” by Lu et al., submitted to Remote Sensing – June 2022

This paper details the application of a previously established framework to characterize the frequency and geographical extent of marine heatwaves to extreme bloom events in an area of the South China Sea where winter blooms drive the variability in biogeochemical properties. The authors use reconstructed cloud-free satellite estimates of chlorophyll concentrations (Chl), also derived in previous work, to allow application of the method, since cloud conditions in the chosen region often prevent use of standard remote sensing products. The contribution of sub-mesoscale processes to driving the spatial-temporal variations in Chl distribution is discussed, providing an added forcing that must be recognized in future studies. I like the idea of applying a method developed for heat waves to assess extreme bloom events, and it is exciting that a reconstructed time series of data enables the use of satellite data in a region often excluded from these types of analysis due to cloud cover.

I think this manuscript could be improved by

1) a more careful explanation of what exactly the Chl signals are describing or characterizing. Chl is not a very reliable proxy for biomass. If there are previous work in the region linking the two, those should be mentioned. How does phytoplankton community structure change over the seasons in the region? Did the authors consider doing this analysis on other proxies for carbon concentration such as particle backscattering or reflectance data? I would imagine those would show less extreme variations, perhaps making it harder to depict blooms. Is that true? What would that say about what Chl is representing?

2) Although the general effects of mixed layer processes are touched upon, maps of mixed layer depth or mixed-layer integrated satellite values would be useful to help discard potential effects of Chl accumulation of dilution in the patterns observed. Those are not discussed at all.

3) Likewise, links between the observed patterns and climatic variables (long term trends in temperature, heat wave frequency, climatic forcings) are barely discussed – could those explain the increase in strength of extreme events that you describe?

4) What prevents applying this analysis to a larger geographical area (i.e. a larger portion of the South China Sea), and use the Luzon Strait bloom as the detailed case study? Applying (and showing) this analysis to only a 5x5 degree region does not do the method justice.  I think a broader spatial range would help make the paper more exciting for readers as they could get insights into areas not discussed in the text, etc. This is especially relevant since many readers (including myself) are not familiar with the geography of the North South China Sea.

5) In my opinion, the publication of a true framework should include the means for other scientists to apply the same analysis to different regions of the world (through sharing of code, data processing scheme, etc). Such presentation of data and methods in a way that can be easily reproduced by others would help make this paper stand out from others.  

Line-by-Line detailed comments:

Highlights section:

-       Point one: there is a mention of extreme bloom events, but title is more general and refers to bloom events. Maybe add the word “extreme” to title? Option to also remove the “(…) in the Northern South China Sea” to make the title more general and attract attention of a wider community.

-       Point two could be removed as it is the result of another publication, and could be substituted by a note on the importance of sub-mesoscale forcings to driving blooms in the region.

Line 17-20: I think sentences that strongly link Chl as a reliable proxy for biomass should be de-emphasized. Chl is a very imperfect proxy for biomass, and that point should be made somewhere in the text. If there are in situ studies that have looked at the relationship between satellite Chl and phytoplankton diversity or total particle biomass in the region, those should be cited. Have you considered doing this analysis on other proxies for biomass such as the backscattering coefficient or Rrs-derived particulate carbon? I wonder if the seasonal/within-bloom variations would be less extreme as those are not as strongly driven by changes in community structure and photoacclimation (independent of changes in biomass).

Lines 57 – 64 contradict each other. It is not clear to me what aspect of the blooms in the SCS are poorly understood. The Introduction tries to make the point that while some blooms have been discussed in the literature, others have not. Yet, this paper discusses some of the same events as Shang et al (2012), Lu et al (2015), and others. I think those sentences stating what is new and what has already been done need to be carefully edited. It is OK to discuss the same blooms if you are coming from a different perspective or if the Method allows looking at different features, but then I do not think so much emphasis should be added in the Introduction regarding how novel this analysis, in this region, is. Why not apply the method to a much larger geographic area and then discuss some specific blooms? I think that would potentially generate more insights to readers hoping to apply the method in different areas.

Line 154: please define HI again here.

Line 172 – not clear what MF3in5 is or where it came from.

Line 180 – “all these data sets were matched as a daily resolution” – not clear what this means. What data sets are you referring to? What spatial resolution were the data in Table 2 average/interpolated to and how?

Line 238 – point out Luzon Island in Fig. 5.

Line 257 – 258. See comment for Figures, but I think the maps need to encompass a larger geographical area to make it more easily placed in space. And the key geographical features mentioned in the text (strait, Luzon Island, Chinese coast, etc) need to explicitly shown in the map to guide the reader. Also, not sure what “offshore wing” means.

Line 276 – I think it is worth it here to link this to the statistics on marine heatwaves, or other parameters that may be linked to climate change – have those become fewer/higher intensity in the last few years as well? Do specific phases of climatic indices play a role in the timing of these events?

Line 317: is there a citation for the potential induction of nutrient entrainment?

Figure 1: what is the box in Figs. 2b-f for? If you choose not to apply the method to a broader area, could you “zoom out” to make the map easier to place in space? Then, you could also point out on the map Philippines, Luzon Island, China, and other relevant regions that are referenced in the text. 

Figure 2: caption reads H19.

Figure 3 (and Line 209): would it be possible to add little arrows to the times where the 12 extreme events were identified? I initially printed a black and white version of the paper and in that version it is very hard to differentiate the lines and see which events crossed the 90% threshold mark.

Table 1: is it possible to add a mixed-layer depth product? How does the mixed layer vary in the region? Is there a role for accumulation/dilution of Chl concentrations in the data you showed? What portion of the water column is the satellite data supposedly characterizing?

Author Response

Please see the detailed response from the attached file.

Author Response File: Author Response.docx

Reviewer 2 Report

The manuscript “Framework to Extract Phytoplankton Bloom Events with Remote Sensing Datasets: A Case Study in the Northern South China Sea” submitted by Wenfang Lu and co-authors presents an interesting results from the adaptation of the statistical framework for defining marine heatwaves to extract equivalent statistics of high chlorophyll concentration events off the Luzon Strait (LZB). 

The statistical approach and the demonstration of its application to the LZB region are of interest to the broad community working in marine ecology. However, the manuscript needs to be revised to improve its structure and better frame its objectives. The main concern refers to the “Discussion” section which, in its present form is a mix of methods, results and discussion stemming from a loosely defined scope for the paper. 

My recommendation is that the paper should be reconsidered for publication after a major revision. When preparing the revised manuscript, I suggest that the authors pay attention to my concerns and suggestions listed below. These are also included as comments in the PDF version of the manuscript that is part of this review.

Comments for author File: Comments.zip

Author Response

Please see the attachment for detailed responses.

Author Response File: Author Response.docx

Reviewer 3 Report

A brief summary.

The work is devoted to a detailed analysis of the spatial-temporal variability of heterogeneous satellite and model data on the parameters of the Northern South China Sea ecosystem in the context of the study of phytoplankton blooming events.
Undoubtedly, the article highlights the results of careful and time-consuming processing of large amounts of information. The structure of the article seems clear and logical.
Excessive self-citation was not detected (~10%).

General concept comments.

-The investigation applies the "H16" metrics (originally developed for the analysis of temperature data) to long time series of data on the concentration of chlorophyll-a, supplemented by the results of modeling. The authors explain that over 80% of the data are simulated, while a quantitative assessment of the accuracy of the simulation results is not given (it is only indicated that the applied model is the best available).  The article should explicitly indicate the accuracy characteristics of the data used, so that the reader could adequately evaluate the results of the study.
-The anomalies of chlorophyll-a concentration detected in the studied ecosystem are tied to the level of 0.1 mg/m3, which is generally not a high indicator. It can be recommended to explicitly present arguments in the article in favor of the fact that the recorded anomalies are caused by changes in the natural environment, and not by the self-system-noise of the set of materials and methods used.

Review and comments.

-The formulations of the highlights block is presented so generically that it can mislead the reader. Phrases of lines 38, 39, 40 can be interpreted ambiguously. It is recommended to avoid such phrases.
- In conclusion, it is stated that studies of the events of extreme algal blooms generally were overlooked (lines 370 371. It is recommended to correct this statement. To date, a large number of studies have been carried out on extreme and harmful algal blooms, both using remote sensing methods and field data. Such events attract the attention of researchers because they cause great damage to ecosystems and the economy. For example, significant events of recent years are the bloom in the Marmara Sea (2021) and the bloom near the Kamchatka Peninsula (2020). These events are covered in scientific periodicals.
-It is necessary to clarify the caption to Figure 1 (the figure shows data not only about chlorophyll-a).
- "H19" appears in line 198, apparently it should have been written "H16".

Thank you so much for your work!

 

Author Response

Please see the attached file for detailed responses to the comments.

Author Response File: Author Response.docx

Round 2

Reviewer 2 Report

No further comments