Microplastics in Freshwater: What Is the News from the World?

Round 1

Reviewer 1 Report

This is a review article about the presence of microplastics in freshwater ecosystems. Since the literature on the subject has not been reviewed yet, and it is largely fragmentary, the authors provided a first attempt to collect currently available knowledge. The authors focused on data regarding microplastics concentration in freshwater ecosystems, and identified the main polymers contaminating freshwater for better management planning. Finally, based on the results the authors suggested future research pathways to achieve a standardised protocol for monitoring, to identify sources and transport routes, and to investigate effects on biota.

The article is timely and relevant, since the topic of microplastics is becoming more prominent in management and policy, and due to the fact that scientific knowledge around this topic is still incomplete and fragmentary. The article has delivered a comprehensive database of existing knowledge that can form the basis for further developments in the field. Moreover, the authors have discussed very relevant issues to standardise processes and identified research needs, based on the gaps identified in the literature so far. The presentation and discussion of the results in satisfactory, but there is some room for improvement. In the following comments, I will present a few suggestions for the authors to consider for revising this article.

Comments:

1. When discussing the polymer identification in the different freshwater systems, I think you could add another layer in your analysis to strengthen the results and the discussion. You could use plastic production data and compare the locations of production and microplastic emergence. Then, you could identify whether the identified concentrations of polymers in a given location corresponds to levels of production of the polymer regionally. For example, if PE and PP are the types of polymers with the highest production rates in China, then it is expected that these polymers would be also identified in the water samples and sediments. However, if another type of polymer is found in the samples, then it could give an indication of mismanagement of a point source. Therefore, I believe it would strengthen your discussion if you analyse also relevant data on polymer production by region (e.g. China, USA, Europe, etc.)
2. It is very important to be able to determine whether microplastics are of primary or secondary source. However, there is a trend (at least in Europe) to regulate against the use of microbeads (primary) in production. Therefore, this type of identification might become less relevant in the future. However, within current affairs, there is no strong indication of worldwide bans on microbeads. Probably, it would be relevant to discuss in the respective section the possibility of regulatory banning of primary microplastics, so that the research can focus more on tracking studies of the secondary sources. You mention already the policy in Italy, but maybe you could expand on the potential of the expanded regulation (in Europe or globally).
3. You have presented the results in many graphs, which brings clarity to the contents of the article. However, there are certain discrepancies in the visual presentations. For example, Figures 8 and 9 are in colour, while all other graphs are in black&white. It would be best if you could be consistent and use either one or the other way of presentation. Another comments about Figures 5 and 6, is that some descriptive information is missing from the captions, which makes it harder to understand. The visual graphic shows circles and rectangles in relation to size, but certain areas have many co-centric circles stacked on each other. Obviously, this means several samples on the same site. However, this information is missing in the caption. Please make the captions a little more descriptive. Finally, Table 1 can be omitted. The table presents p values and coefficients found in another study. Firstly, the numbers in the table cannot be sufficiently understood without the context of the referred study. Secondly, the existence of the table does not add significantly to the clarity of the content. Just a simple description in the text and a reference to the source would suffice. By adding this table, it attracts the reader to search for empirical evidence in the article which does not exist directly in the article but on the referred source.
4. The introduction needs an additional proof-reading to make sure that the language is up to the standard of the rest of the article. Several grammar and syntax mistakes have been detected. For example, in line 39, you mention ‘contribute to the plastic widespread’. Here the word ‘widespread’ is an adjective. It would sound better if you write ‘contribute to the wider spread of plastics.’ In line 47, you mention ‘Anyway, plastics have detrimental effects on aquatic biota, as well.’ This is not good syntax. A simpler sentence would sound much better, for example ‘Plastics have also detrimental effects on aquatic biota.’ In line 50, you mention ‘in example’ while it should be ‘for example’. In lines 69-70, you write ‘In nature, MPs contaminates inland and marine waters. The assessments of the hazard due to MPs within inland water ecosystems focuses on different topics.’ The correct sentences would be ‘In nature, MPs contaminate inland and marine waters. The assessments of the hazard due to MPs within inland water ecosystems focus on different topics.’
5. In section 3.3 you have capitalised the word ‘Country’ throughout the text. There is no need for capital letter in the work country.
6. In line 434 you write ‘900. I guess you mean the 1900s ? Please correct this.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

The authors present a literature review of the occurrence and concentration of freshwater micro-plastics (fMPs) in water, sediment and biota.  Strengths of this work include coverage (publications through May 2020), and graphics showing distributions of measured concentrations worldwide.  I believe this review could be a valuable resource for other researchers.  However, I have several concerns. I sincerely hope that my comments and suggestions are helpful.

(1) My major concern is that the review draws conclusions about comparative levels of fMP concentrations based on uneven sampling designs. I believe these conclusions are unfounded and potentially misleading.  For example, Line 177 – I do not believe we can say where “the most polluted areas are” given that not every place has been sampled, or sampled equally.  I suggest rewording this statement.  In fact, I think “more polluted” (line 180) must be based on the level of pollutants relative to area sampled in randomized efforts.  I think what you show graphically are reported concentrations for fMPs, but these values cannot be compared among areas without standardizing for sampling effort.

Similarly, Sections 4.1 and 4.2 – Statements about what geographic areas are “most contaminated” are unsupportable by the data presented.  Lacking a fully randomized sampling effort across waterbodies worldwide, we cannot infer where contamination is relatively high or low.  Picking out values that high in particular countries without understanding the underlying sampling design is uninformative.  Attributing these high values to economic status (e.g., “developing countries”) is unscientific and perceivable as racist.

Lines 381-381 – similarly, the statement “Riverine waters have a significantly higher concentration of fMPs than lacustrine waters” is only valid if based on a randomized sampling design, and even then should be qualified with respect to the geographic areas sampled.  One might say that “Concentrations of fMPs reported for rivers are generally higher than those that have been reported for lakes”, and then explore reasons for that difference – sampling intensity, geographic scope for each, differing methodologies, actual difference in the processes that result in fMPs between lakes and rivers? Indeed, the remainder of this paragraph points to sampling geography may bias the comparison.

(2) Attributing studies to “lotic” or “lentic” systems is quite coarse in resolution. I believe it would be valuable to know how those studies are distributed across sizes of lakes, and among streams and rivers by drainage area.  That is, are there studies of fMPs from small lakes and streams, or are all or most from larger, more downstream systems?

(3) Lines 250-268, Section 3.7 – these ‘diachronic’ analyses all involve comparisons between 2 or 3 points in time. Although these could be meaningful, I do not think one can validly infer temporal trends without verifying that samples were taken in the same way in each time period, and that there are sufficient samples within each time period to account for sampling variability.  One could use repeated samples across systems to assess evidence for a temporal trend if samples were always taken in the same season, in the same way.  However, it’s not clear to me that the assorted values plotted in SuppMat 5 do not represent variability attributable to seasonal or sampling differences, as opposed to temporal trends.  It is not clear to me how the conclusion, e.g., there is a “downward trend of fMPs in lakes in recent years”, is supported by the data or analyses.  In fact, the Discussion (lines 427-428) highlights “seasonal and hydrologic conditions” as causes for temporal differences.

(4) Lines 284-289 and Table 1 – Please clarify what these correlation values represent. Are these based on data from studies that examined fMPs by type in water and sediment, or water and biota, or biota and sediment?  How many studies are included for each correlation value?  Why is a “p value” of 0.006 not considered “significant”?

(5) Lines 187-191 – Explanation is needed for these values. What do the values following “+” represent?  Standard errors?  These values are mostly large and, if they standard errors, they would imply negative counts on the lower 95% confidence interval bounds, which don’t make sense.  I suggest reformatting this information into a table that shows the number of studies, the mean value, the range across studies, the types of waterbodies, and the region or regions represented.

Additional editorial suggestions:

Line 29 – suggest “for human use” rather than “consumption”

Line 31 – suggest “Additives increase specific properties of plastic polymers; for example,…

Line 39 – suggest “contribute to the distribution of plastics”         

Line 42 – suggest omitting “all”                   

Line 47 – suggest “Plastics also have detrimental effects on aquatic biota.”

Line 50 – “For example,…”  (rather than “In example”)

Line 59 – “which originate from”

Line 69 – “MPs contaminate…”

Line 69 – “Assessment of the hazard…”

Line 103 – “information is collected”

Lines 140-141 – suggest “The number of articles varies among continents:…”

Line 141 – clarify – does “America” include only North America?

Line 143 – “the studies mostly focus on particular waterbodies. In particular, the most researched lakes are:…”

Line 157 – “researchers belonging to”

Lines 153-161 – suggest not capitalizing “Country”, “Countries” or “Research Institutions”

Line 167 – “literature focuses on”

Lines 192-198 – this information would be useful if presented prior to introducing Figures 5 and 6, to explain why two sets of units are presented in those figures.

Line 225 – the values “24 items/kg” and “47 items/kg” as median concentrations of fMPs in lakes and rivers do not appear to agree with the data presented in Figure 7.  Here the medians for “log(#items/kg)” appear to be about 5 (or exp(5)=148 items/kg) for both types of waterbodies.  Clarify?

Lines 230-233 – suggest replacing “fouling” with “colonization”

Line 236 – “most specimens”

Line 237 – “Moreover, ingestion is investigated…”

Line 238 – “freshwater bird specimens”

Line 239 – clarify, does this mean that fMPs were detected in digestive tracts of 11% of birds examined?  I suggest explaining how researchers test for fMPs in the bodies of animals. 

 

 

 

 

 

 

 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

This study presents a broad review about microplastics presence in freshwater, sediments and biota. It is very clear and pertinent.

Supplementary material presents the collected data from the different studies in a very transparent and verifiable way.

I believe some issues should be addressed by the authors:

Issue 1) The presented p-values are of dubious use, as sample randomness and independence if very difficult to assess in such datasets. Yet, they could be interesting as a "if considered independent and random" exercise... But eventually, I don't see how to statistically support their use. Simple descriptive statistics would be sufficient to characterize the database. Please consider this and possibly discuss this in the manuscript or remove p-values.

Issue 2) Authors mention normality Shapiro-Wilk test in methods but don't mention their results in the manuscript. Then, authors rely on non-parametric tests, passing the idea that samples where not  normally distributed (?). Please address or justify. 

Issue 3) Discussion should have in consideration that sample randomness and independence is unknown (thus sentences assuming it is representative ought to be avoided, and referred only as database own characteristics). Furthermore, there are some possible biases that could be further discussed. For instance: i) can the detected patterns of main polymers contaminating freshwater be related to cheaper/easier/readily available identification techniques in each country? ii) is the data spatially aggregated and/or spatially autocorrelated? This could bring further insight to the analyses.

Issue 4) Sometimes, standardized protocols are not followed by researchers as they are not sufficiently flexible and/or scalable. The authors proposal on keeping old ways of data recording is valuable. However, for a review article I feel the authors could be more ambitious in the standardization proposals, possibly defining scalable solutions depending on research budget (?). A table or diagram summarizing authors proposals would be certainly useful.

Additionally, I send the following comments and suggestions:

Page 7, line 203. I'm not sure the intended Italian meaning of "omologare" exists in English as "homologate". Please confirm. Maybe "transmformed" or "converted" is enough.

Page 7, line 217. Again "homologated". Please confirm.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

I believe the authors have satisfactorily responded to my review comments.  I also think this is an interesting and useful literature review of the state-of-knowledge regarding microplastics in freshwater systems.  I have just a few additional minor edits to suggest.

Line 72 – “Assessments …focus on several different topics (…)”. 

Line 80 – “In fact,…” or “Indeed,…”

Lines 88-89 – “polymers that contaminate”

Figure 7 – the lower of the two graphs is to be deleted?

Lines 252-253 – perhaps reword to clarify; does this mean that 11% of individuals, or of species, examined were found to contain fMPs?

Figure 9 – suggest “data were available…”

Line 403 – suggest “Three major types of information…”

Line 450 – “to the biota studied for risk assessment”?

Author Response

AU: we are pleased that Reviewer 2 considered the review scientifically interesting and we thank for the further suggestions.

Line 72- we add "different" as suggested.

Line 80- we opted for "Indeed,".

Lines 88-89- corrected gramatically.

Figure 7- yes, the lower of two graphs is to be deleted. It has been deleted in review mode, when the changes are accepted it is eliminated.  

Lines 252-253- we clarified that are 11% of individuals.

Figure 9- modified as suggested. 

Line 403- modified as suggested.

Line 450- we clarified the sentence changing it to "...the research on biota regarding...".