Effects of Dissolved Organic Matter (DOM) on Perfluorooctanoic Acid (PFOA) in a Seagoing River—A Case Study of the Wanggang River Flowing into the East China Sea

Round 1

Reviewer 1 Report

Review of “Effects of Dissolved Organic Matter….” By Lai et al.

Based on the abstract, this manuscript describes a study within the Wanggang River Basin (China) assessing PFOA and DOM concentration(s) and distribution(s). The authors aim to describe DOM moieties (via optical properties) and link these to PFOA presence. The linkage(s) are described as causal (rather than only correlated, e.g.: lines 20-21, “The PFOA concentration and distribution in the Wanggang River Basin were affected by the DOM compositions.”). The methods seem consistent with current standard(s) and data processing (DOM Fluor toolbox, Student’s t-test and correlation via SPSS) all seem appropriate. The graphs and table seem appropriate (more below) although Fig. 4 appears to be a raster image and seems a bit “blurry”. Perhaps a larger screenshot or conversion to a higher-resolution format would be prudent.

Weaknesses: In terms of PFOA measurement (section 2.2), there doesn’t appear to be information addressing system blank(s). PFOA is relatively ubiquitous – were blanks systematically run? Were they subtracted from real samples? Filters, glassware, etc may all contain trace PFAS materials.

Overall – the main issue with the manuscript is that the wording and conclusions in the abstract are only weakly supported by the experimental data. The abstract states, “The PFOA concentration and distribution in the Wanggang River Basin were affected by the DOM compositions.” Looking at Table 3, only the C1 fluorescent component within DOM is weakly correlated with PFOA concentration(s). Considering that there was no correlation between bulk DOM nor the other 3 fluorescent components and PFOA, this conclusion isn’t well supported. “Affected” implies a casual relationship. The weak correlation doesn’t imply causality in my opinion. I think the abstract should be re-written in a much more “cautious” way in order to convey the very weak and non-existent correlation between bulk and fluorescent DOM and PFOA in the watershed. Thus stated, follow-on studies might be better defined – perhaps looking at aliphatic (non-fluorescent) industrial inputs or other un-screened organic contaminants as proxies for PFAS input (fuels, oils, other industrial chemicals, etc).

I thus rate this manuscript in need of major revision. The disconnect between the data and the abstract must be fixed.

More specific comments:

Ln 48: “….by degradation.” This is unclear. Do you mean by degrading the above list? I think you mean that the above list produce DOM. Some of the processes are related to degradation (e.g. microbes and fungi). Others (algal exudates for instance) are not by degradation but by synthesis. If that’s what’s meant to be conveyed – I would make two sentences listing the processes.

Ln 51-52: “…proteins reduce the concentration of PFASs” Odd wording? Does this mean that proteinaceous material (DOM?) sorbs PFAS making the aqueous concentration lower? Can this sentence be more clear?

Ln 140-142: I think this is boilerplate from a template? Probably remove?

Author Response

List of Responses

 

Dear Editor and Reviewers:

 

Thank you for your letter and for the reviewers’ comments concerning our manuscript entitled “Effects of Dissolved Organic Matter (DOM) on Perfluorooc-tanoic Acid (PFOA) in a Seagoing River—A Case Study of the Wanggang River Flowing into the East China Sea”. Those comments are all valuable and very helpful for revising and improving our paper, as well as the important guiding significance to our research. Taking account of reviewers’ comments, we have revised and improved the manuscript. We hope our revisions meet with approval. Revised portion is marked up using the “Track Changes” function in the paper. The main corrections in the paper and the responses to the reviewers’ comments are as follows.

 

Reviewer 1

 

Comments and Suggestions for Authors:

Review of “Effects of Dissolved Organic Matter….” By Lai et al.

 

Based on the abstract, this manuscript describes a study within the Wanggang River Basin (China) assessing PFOA and DOM concentration(s) and distribution(s). The authors aim to describe DOM moieties (via optical properties) and link these to PFOA presence. The linkage(s) are described as causal (rather than only correlated, e.g.: lines 20-21, “The PFOA concentration and distribution in the Wanggang River Basin were affected by the DOM compositions.”). The methods seem consistent with current standard(s) and data processing (DOM Fluor toolbox, Student’s t-test and correlation via SPSS) all seem appropriate. The graphs and table seem appropriate (more below) although Fig. 4 appears to be a raster image and seems a bit “blurry”. Perhaps a larger screenshot or conversion to a higher-resolution format would be prudent.

 

Weaknesses: In terms of PFOA measurement (section 2.2), there doesn’t appear to be information addressing system blank(s). PFOA is relatively ubiquitous – were blanks systematically run? Were they subtracted from real samples? Filters, glassware, etc may all contain trace PFAS materials.

 

More specific comments:

Ln 48: “….by degradation.” This is unclear. Do you mean by degrading the above list? I think you mean that the above list produce DOM. Some of the processes are related to degradation (e.g. microbes and fungi). Others (algal exudates for instance) are not by degradation but by synthesis. If that’s what’s meant to be conveyed – I would make two sentences listing the processes.

Ln 51-52: “…proteins reduce the concentration of PFASs” Odd wording? Does this mean that proteinaceous material (DOM?) sorbs PFAS making the aqueous concentration lower? Can this sentence be more clear?

Ln 140-142: I think this is boilerplate from a template? Probably remove?

 

Response: Thanks for the reviewer’s suggestion. We have revised the manuscript according to the review comments, please see below for details.

 

 

Point 1: Fig. 4 appears to be a raster image and seems a bit “blurry”. Perhaps a larger screenshot or conversion to a higher-resolution format would be prudent.

 

Response 1: Thanks for the reviewer’s suggestion. We have replaced the figure with a clearer picture.

 

Point 2: Weaknesses: In terms of PFOA measurement (section 2.2), there doesn’t appear to be information addressing system blank(s). PFOA is relatively ubiquitous – were blanks systematically run? Were they subtracted from real samples? Filters, glassware, etc may all contain trace PFAS materials.

 

Response 2: Thanks for the reviewer’s suggestion. We have increased more statement about the quality assurance and quality control of the study on page 4, line 118-131 of the revised version.

(1) Quality assurance and quality control were performed throughout the entire pro-cess of sample collection and transportation, PFOA extraction, and on-machine testing. In the sample collection, we avoided using PTFE or other fluoropolymer materials during sampling to prevent background contamination. The sampling bottles were rinsed three times with ultrapure water before collecting water samples. During the transportation of samples, container blank and transportation blank were set and ul-trapure water was used as the monitoring medium.

(2) Background contamination and cross-contamination should also be avoided dur-ing the extraction of PFOA. We avoided the use of PTFE or other fluoropolymer mate-rials in extraction to ensure that the analysis results were not affected. PFOA extrac-tion, enrichment, purification, and other processes were monitored in order to prevent background contamination and cross-contamination between samples. In addition, replicates were set up every 10 samples to maintain the reproducibility of the analysis.

 

Point 3: Ln 48: “….by degradation.” This is unclear. Do you mean by degrading the above list? I think you mean that the above list produce DOM. Some of the processes are related to degradation (e.g. microbes and fungi). Others (algal exudates for instance) are not by degradation but by synthesis. If that’s what’s meant to be conveyed – I would make two sentences listing the processes.

 

Response 3: Thanks for the reviewer’s suggestion. We are very sorry for our incorrect wroding and the misunderstanding we caused. It is rectified on page 2, line 48-49 of the revised version.

 

Point 4: Ln 51-52: “…proteins reduce the concentration of PFASs” Odd wording? Does this mean that proteinaceous material (DOM?) sorbs PFAS making the aqueous concentration lower? Can this sentence be more clear?

 

Response 4: Thanks for the reviewer’s suggestion. Again, we are very sorry for our negligence of the explanation. We have modified the content on page 2, line 53-54 of the revised version.

(1) Studies have shown that the molecular weight of the DOM can alter the binding ca-pacity of PFASs to the DOM, and proteins inhibit the bioconcentration of PFASs in Chironomus plumosus larvae.

 

Point 5: Ln 140-142: I think this is boilerplate from a template? Probably remove?

 

Response 5: Thanks for the reviewer’s suggestion. We have removed the sentence.

 

Point 6: Overall – the main issue with the manuscript is that the wording and conclusions in the abstract are only weakly supported by the experimental data. The abstract states, “The PFOA concentration and distribution in the Wanggang River Basin were affected by the DOM compositions.” Looking at Table 3, only the C1 fluorescent component within DOM is weakly correlated with PFOA concentration(s). Considering that there was no correlation between bulk DOM nor the other 3 fluorescent components and PFOA, this conclusion isn’t well supported. “Affected” implies a casual relationship. The weak correlation doesn’t imply causality in my opinion. I think the abstract should be re-written in a much more “cautious” way in order to convey the very weak and non-existent correlation between bulk and fluorescent DOM and PFOA in the watershed. Thus stated, follow-on studies might be better defined – perhaps looking at aliphatic (non-fluorescent) industrial inputs or other un-screened organic contaminants as proxies for PFAS input (fuels, oils, other industrial chemicals, etc).

 

Response 6: We appreciate it very much for this good suggestion, and have done it according to your ideas. We have modified our major conclusion about the correlation between DOM and PFOA on  page 1, line 20-21, page 9, line 297-299, page 10, line 315-316 of the revised version. We have also increased statement of outlook on further studies on page 9, line 300-306 of the revised version.

(1) Taking consideration of the strong correlation of PFOA with post-flood C1 and the non-existent correlation with other components, humus-like substances were one of the factors affecting PFOA distribution and concentration after the flood season.

(2) In consideration of the complex pollutant sources in Wanggang River Basin, de-tection of PFOA is not sufficient to explore the interaction between DOM and emerging organic pollutants. In future research, the monitoring of the 17 PFAS homologues might address the limitation. On this basis, research into the cause of PFAS concentra-tion variation across reaches and seasons, as well as the correlation between PFAS and DOM, is required. And the connection of DOM with PFAS might be more evident to discuss.

 

 

Author Response File: Author Response.pdf

Reviewer 2 Report

It is a very simple study. Data from 30 sampling sites were collected during pre- and post- monsoon periods to understand the PFOA concentration and the effect of DOM on the PFOA concentration. The methods used for analysis are not clear and properly justified. I was about to reject the paper. However, I believe there is a scope to improve the paper. I suggest addressing all comments properly before further consideration of the paper for publication.

Major comments:

1.       The authors just mentioned the software names used for data processing, for example, Microsoft Excel, SPSS, maPlot, etc. But not clearly justified the method used. They used a mixture of parametric and non-parametric methods. For example, parametric Student's t-test to determine if the means were significantly different, while non-parametric Spearman correlation to check the association. There is no justification for using the methods. How is the distribution of your data? The sample size is only 30. I believe you should use a non-parametric method for all cases.

2.       You can not tell PFOA concentration is higher or lower without a statistical test. The values presented in the text (The PFOA concentration ranged from 3.2 to 52.9 ng·L-1, and the average was 27.1± 13.4 ng·L-1. The pre-flood PFOA concentrations ranged from 4.7 to 52.9 ng·L-1 from 28.4± 13.8 ng·L-1) and also the data presented in Figure 2, it is very clear that there is no statistical difference in PFAO concentration before and after flood. Therefore, the major result mentioned in the Abstract “Differences were observed in the DOM components before and after the flood season” is not correct. I also disagree with the discussion made in Lines 171-188, because “higher PFOA concentration before than after the flood season” is not true.

3.       The results are discussed unclearly. Lines 172: “Previous studies have shown that the Wanggang River is slightly polluted, …” – which study it is. What is the difference between that study with your study? The author has not discussed anything about previous studies of pollution in the river. They also did not justify how their study differed from the other studies, the new issue they addressed, or the new knowledge they gathered. Just collecting some samples and showing results somehow can not be a scientific article.

4.       Why is ECMWF rainfall used instead of observed data? The quality of ECMWF in the basin may not be good. Have you tried to check the quality of this rainfall in the basin?

5.       “The DOC concentration was used as a proxy of the DOM concentration.” – How viable is this assumption in your river basin?

6.       Line 241: “A weak correlation” – what do you mean by weak correlation? If a correlation is not significant, then there is no correlation. Please change the wording. Revise the discussion accordingly.

7.       The major conclusion of the article “This finding indicated that the PFOA content and distribution in the surface water of the Wanggang River Basin were significantly dependent on the DOM composition.” – Is it a new finding? What about other seagoing rivers in China? The author mentioned similar studies in a different river basin in Table 2. I suggest discussing the comparative analysis of the results with other similar rivers in China and trying to tell what new or interesting they found in this study.

Author Response

List of Responses

 

Dear Editor and Reviewers:

 

Thank you for your letter and for the reviewers’ comments concerning our manuscript entitled “Effects of Dissolved Organic Matter (DOM) on Perfluorooc-tanoic Acid (PFOA) in a Seagoing River—A Case Study of the Wanggang River Flowing into the East China Sea”. Those comments are all valuable and very helpful for revising and improving our paper, as well as the important guiding significance to our research. Taking account of reviewers’ comments, we have revised and improved the manuscript. We hope our revisions meet with approval. Revised portion is marked up using the “Track Changes” function in the paper. The main corrections in the paper and the responses to the reviewers’ comments are as follows.

Reviewer 2

 

Comments and Suggestions for Authors:

It is a very simple study. Data from 30 sampling sites were collected during pre- and post- monsoon periods to understand the PFOA concentration and the effect of DOM on the PFOA concentration. The methods used for analysis are not clear and properly justified. I was about to reject the paper. However, I believe there is a scope to improve the paper. I suggest addressing all comments properly before further consideration of the paper for publication.

 

Point 1: The authors just mentioned the software names used for data processing, for example, Microsoft Excel, SPSS, maPlot, etc. But not clearly justified the method used. They used a mixture of parametric and non-parametric methods. For example, parametric Student's t-test to determine if the means were significantly different, while non-parametric Spearman correlation to check the association. There is no justification for using the methods. How is the distribution of your data? The sample size is only 30. I believe you should use a non-parametric method for all cases.

 

Response 1: Thanks for the reviewer’s suggestion. We used a Shapiro-Wilk test to determine the distribution of our data. And Student’ t-test was chosen because the data followed a normal distribution.We have added more explanation on page 4, line 150-152 of the revised version.

(1) SPSS 24.0 software was used to conduct a Shapiro-Wilk test to determine if the data followed a normal distribution. Student’ t-test was chosen when data followed a normal distribution.

 

Point 2: You can not tell PFOA concentration is higher or lower without a statistical test. The values presented in the text (The PFOA concentration ranged from 3.2 to 52.9 ng·L-1, and the average was 27.1± 13.4 ng·L-1. The pre-flood PFOA concentrations ranged from 4.7 to 52.9 ng·L-1 from 28.4± 13.8 ng·L-1) and also the data presented in Figure 2, it is very clear that there is no statistical difference in PFAO concentration before and after flood. Therefore, the major result mentioned in the Abstract “Differences were observed in the DOM components before and after the flood season” is not correct. I also disagree with the discussion made in Lines 171-188, because “higher PFOA concentration before than after the flood season” is not true.

 

Response 2: Thanks for the reviewer’s suggestion and we have done it according to your ideas. We conducted a comparison of PFOA between May and September in different reaches and increased the results on Fig. 3 of the revised version.We have also edited the discussion on page 6, line 189-191.

As for the DOM，we agree with you that a statistical test is needed when analyzing data. However, the components fo DOM before the flood season differed from after. So we used the differences of DOM as a major result.

(1) According to the comparison in Fig 3, there was a higher PFOA concentration before than after the flood season in downstream, implying contaminants in this reach was affected by the rain.

 

Point 3: The results are discussed unclearly. Lines 172: “Previous studies have shown that the Wanggang River is slightly polluted, …” – which study it is. What is the difference between that study with your study? The author has not discussed anything about previous studies of pollution in the river. They also did not justify how their study differed from the other studies, the new issue they addressed, or the new knowledge they gathered. Just collecting some samples and showing results somehow can not be a scientific article.

 

Response 3: Thanks for the reviewer’s suggestion. We are very sorry for our incorrect wroding and the misunderstanding we caused. The “previous studies” was a water quality report that investigated sources of pollution from local government. We have modified the sentence on page 6, line 191-193.

 

 

Point 4: Why is ECMWF rainfall used instead of observed data? The quality of ECMWF in the basin may not be good. Have you tried to check the quality of this rainfall in the basin?

 

Response 4: Thanks for the reviewer’s suggestion. We have obtained the precipitation data from the China Meteorological Data Service Center (CMDC) and remade the chart accroding to the data.

 

Point 5: “The DOC concentration was used as a proxy of the DOM concentration.” – How viable is this assumption in your river basin?

 

Response 5: Thanks for the reviewer’s suggestion. We have added more explanation on the relationship between DOC and DOM and altered our statement on page 7, line 212-214 of the revised version.

(1) By element composition, the DOM is mainly composed of C, O, H, N, P and S. The DOC accounts for more than 50% of DOM. Therefore, the DOM content in the water was characterized by DOC concentration.

 

Point 6: Line 241: “A weak correlation” – what do you mean by weak correlation? If a correlation is not significant, then there is no correlation. Please change the wording. Revise the discussion accordingly.

 

Response 6: Thanks for the reviewer’s suggestion. We have edited statement on page 8, line 265-268 of the revised version.

(1) There is no correlation occurred between the humus-like components C1 and pro-tein-like substances, indicating that the sources of different elements of DOM in the Wanggang River Basin might be different before the flood season.

 

Point 7: The major conclusion of the article “This finding indicated that the PFOA content and distribution in the surface water of the Wanggang River Basin were significantly dependent on the DOM composition.” – Is it a new finding? What about other seagoing rivers in China? The author mentioned similar studies in a different river basin in Table 2. I suggest discussing the comparative analysis of the results with other similar rivers in China and trying to tell what new or interesting they found in this study.

 

Response 7: Thanks for the reviewer’s suggestion. Our study focused on the relationship between PFOA and DOM in a seagoing river. On one hand, seagoing rivers are one of the main sources of water pollution in the coastal sea areas.Our research can guide the reduction of land-based sources to control PFOA concentrations to some extent. On the other hand, there are few analysis of the relationship between PFOA and DOM in river and lake of China.The study can be used as a reference for the prevention and control of pollutants in seaging rivers of other regions.According to your suggestion, we have modified our major conclusion about the correlation between DOM and PFOA on page 1, line 20-21, page 9, line 297-299, page 10, line 315-316 of the revised version. We have also added description of a similar study on page 9, line 300-306.

(1) Taking consideration of the strong correlation of PFOA with post-flood C1 and the non-existent correlation with other components, humus-like substances were one of the factors affecting PFOA distribution and concentration after the flood season.

(2) The research performed in western Lake Chaohu and its inflow rivers found a positive correlation between PFAS and DOM and no significant correlation between PFAS and DOC. They investigated not only PFOA but also other 10 PFAS homologues and revealed that the compositions of DOM could influence the distributions of PFAS in the aqueous phase in the lake.

 

 

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

The revised manuscript addresses all of my initial concerns. Any linkage between PFAS materials, natural organic matter and seasonality will be of societal value.  

Reviewer 2 Report

-