Freshwater Phenanthrene Removal by Three Emergent Wetland Plants

Reviewer #1 Comments

1)     The abstract reads as a descriptive summary rather than a concise synthesis of findings and implications. It lacks a clear statement of novelty - why this study is important or how it advances current understanding of plant-mediated phenanthrene remediation.

We have revised the abstract to be more of a concise synthesis of findings from this study. We have specifically revised the opening sentences of the abstract so that it addresses the importance and novelty of this work:

“The use of floating wetlands has been receiving increased attention as a minimally invasive method for oil spill remediation, but the species of vegetation incorporated in floating wetlands may influence the success of oil degradation. Therefore, a freshwater microcosm experiment was conducted at the IISD Experimental Lakes Area, Canada to assess the potential of common wetland plants Typha sp., Carex utriculata, and C. lasio-carpa, to remove phenanthrene, a polycyclic aromatic hydrocarbon ubiquitously found at oil spill sites.”

2)    The author’s rationale in the introduction for choosing phenanthrene as the model compound is partly justified by solubility and detection frequency. However, it lacks discussion of its representativeness for broader oil spill contexts or its degradation pathways in plant-associated systems. I think this can be improve to better justify the use of phenanthrene.

We are somewhat confused about why the reviewer feels our rationale for choosing phenanthrene was not sufficiently justified. Our original text included the following justification, which focuses on phenanthrene’s use as a general marker for oil contamination (lines 80-83 of the revised manuscript):

“The presence of phenanthrene in the environment is often used as an indication of oil contamination [6,32], and was a prevalent compound detected in previous in-lake, controlled oil spill research at the IISD-ELA [12,13,33].”

However, with reference to known degradation pathways in plant-associated systems, we have added the following statement, and 3 new references, to the Introduction section to further justify the choice of phenanthrene as a test compound (Lines 85-87 of the revised manuscript):

“Finally, several previous studies have demonstrated degradation of phenanthrene by single species of plants and their root biofilm consortia few [34-36] but compared that potential among different co-occurring species [22].

3)    In the experimental set up, the author use 10% Hoagland nutrient solution which creates a eutrophic to hypereutrophic conditions. This condition is not a representative of natural oligotrophic lake systems which compromises the ecological relevance of the findings of this study. A supporting explanation is needed in this part of the methodology.

We believe that we explained our use of Hoagland nutrient solutions and appropriately acknowledged that they are not representative of lake conditions in the original text:

“Final nutrient concentrations in one litre of media are indicated in Supplementary Table 1 and were selected to ensure that nutrient availability was not a limitation of the study, but that it was not representative of unrealistic conditions, as boreal lakes at the IISD-ELA are oligotrophic and have poor nutrient availability. However, noting that concentrations in 5% media are considered hypereutrophic based on Wetzel (1983) trophic classifications in Spalinger and Bouwens [35].”

However, to address the reviewer’s comment we have revised that section to more clearly outline our rationale (lines 106-112 of the revised manuscript):

“Nominal nutrient concentrations in one litre of media are indicated in Supplementary Table 1. Hoagland’s incubation media was selected to ensure that nutrient availability was not a limitation of the study, and was not intended to be representative of conditions in boreal lakes at the IISD-ELA, which tend to have poor nutrient availability and are categorized as oligotrophic. Concentrations of phosphorous available in 5% media would be considered hypereutrophic [38].”

4)    Although it is appropriate to use isotope dilution GC–MS/MS in this study. The paper lacks validation details such as calibration curve performance, procedural blanks, and recovery corrections for the reported concentrations. Please provide this in the revised manuscript.

In revised text, we have more clearly outlined that details regarding the isotopic dilution methods of phenanthrene detection are outlined in the cited reference Idowu et al. [43], including method performance, calibrations and appropriate blanks (lines 218-220 of the revised manuscript):

“Details regarding isotope dilution methods of phenanthrene detection, including method performance, calibrations and appropriate blanks are outlined in Idowu et al. [43].”

Should the editors require even more detail, we are prepared to include specific assay parameters from this study in SI.

5)    In this paper, biomass was not normalized to initial plant size, and no microbial community characterization was performed. I think without microbial or molecular data, claims regarding “plant-specific factors” remain speculative. This should also be addressed in the revised manuscript.

We did not normalize the study to initial plant biomass. Early season wild plants were collected from a donor wetland and allowed to establish in our microcosms for 5 weeks. At that stage it would have been disruptive to disentangle the fragile plants and their developed root systems from the stainless-steel wire support frames in the microcosm to obtain an initial weight prior to phenanthrene exposure. However, as noted in the text (lines 323-327, TableS3) we acknowledge this shortcoming of our study and normalized parameters to the final biomass in our analysis:

“relationships to final biomass were analyzed for some water quality parameters with a linear model using the stats package (v.4.2.2; [50]). In addition, biomass was not statistically analyzed to determine potential impact of phenanthrene as they were collected from a donor wetland and may have had different initial biomass. Supplementary Table S3 includes summary of final dry biomass by treatment.”

Regarding microbial community analysis, determining the microbial consortia colonizing the root systems of plants prior to phenanthrene exposure would have required dedicated plants in separate microcosms for destructive sampling; something that could not be accommodated within the limitations of space and analytical effort for our study. Furthermore, analysis of microbial consortia on root systems in some instances can verify shifts in community composition but may not directly correlate with degradative activity to remove specific compounds. However, to address the reviewer’s comment, we have added a paragraph in our Discussion that explains why microbial analysis was not our focus for this study, and directs the reader to  previous in-situ studies where we examined microbial community changes on floating wetlands during controlled oil exposures in mesocosms at the IISD-Experimental Lakes Area (lines 634-641 of the revised manuscript):

“We did not assess microbial communities in this study using molecular techniques but instead focused on gross indices of removal. We have previously shown that microbial communities on the roots of wetland plants grown on floating wetland platforms can change in response to oil exposure, and that some hydrocarbon degrading taxa can be stimulated [12].  However, attributing metabolism of a specific compound to one or more taxa is tenuous and it is likely that non-hydrocarbon degrading taxa may also support biodegradation, confounding interpretation of the removal with community change metrics  [78,79].”

6)    In the methodology, the authors didn’t include mass balance (volatilization, sorption to glassware, or root uptake quantification). Thus, “removal” cannot be equated with “biodegradation.” The authors should provide these additional data in the revised manuscript.

We did not determine volatilization of phenanthrene, nor did we extract glassware or plant roots to allow a mass balance accounting for the compound. However, we included appropriate reference microcosm replicates that were treated with phenanthrene but had no plants. These were included to account for sorption to glassware and volatilization processes. To acknowledge the importance of the processes outlined in the reviewer’s comment, and to provide context for our results, we have added a qualifying statement, and appropriate reference, regarding our use of the term “removal” (Lines 685-692 of the revised manuscript):

“Phenanthrene removal from water in the treated microcosms could have resulted from the collective processes of volatilization, sorption to glassware, adsorption or absorption to roots, biodegradation, or photolysis. Sorption of phenanthrene to glassware can be significant and is dependent on contact surface area of the test vessel [83]. We included reference microcosm replicates that were treated with phenanthrene but had no plants to account for sorption to glassware and volatilization processes. Additionally, all microcosms were wrapped in foil to reduce photodegradation effects.”

In conjunction with this addition, we have removed a redundant sentence at the end of the next paragraph of the original manuscript:

 “The decline in phenanthrene concentrations may have been caused by rapid metabolism, incomplete mixing, or the adsorption of phenanthrene onto plant roots or the glass microcosms.”

Finally, we have ensured that throughout the text we use the term “removal” and not “degradation” or “metabolism” to refer to the loss of phenanthrene from the microcosm water. Coupled with the text noted above, beginning on Line 685, we feel strongly that we have sufficiently qualified what we are referring to when we use the term “removal”.

7)    The authors interpret overall phenanthrene loss as “removal,” but does not distinguish between biodegradation, sorption, volatilization, or photolysis. Also, they didn’t provide mass balance or chemical fractionation data, making it inappropriate to infer biodegradation as the main pathway. The authors should address this in the revised manuscript.

We have addressed this in our response to the Reviewer’s comment #6, immediately preceding this one.

Further to our response above, we would like to point out that we have included several qualifiers throughout the manuscript that indicate “removal” of phenanthrene may have occurred via several mechanisms, and not exclusively via biodegradation. For example:

Lines 626-633 (bold text appropriately indicates uncertainty regarding microbial metabolism):

“It was hypothesized that available oxygen was consumed for aerobic metabolism of phenanthrene and the lack of vegetation or the small root system associated with Sedge B was not able to produce sufficient oxygen release into the rhizosphere, causing oxygen to decline. During this sample period, these microcosms were visibly cloudy which may indicate a microbial bloom. While this cannot be confirmed, the DT50 values identified from the phenanthrene exponential decay models were approximately 5 days for these two treatments (Table 1), so it is clear there was phenanthrene removal occurring, which may have included microbial aerobic metabolism.”

Lines 678-685: (bold text indicating uncertainty regarding the mechanism of removal and that “metabolism” is just of the potential mechanisms).

“Moreover, while we could not confirm that plants enhanced removal of phenanthrene compared to the control, it is clear that cattail treatments generally have faster removal rates, further supported by the DT50 (noting that all models had a different y0). In fact, all plant treatments had lower estimated y0 for the exponential decay model than Phenanthrene microcosms during the first round of sampling, suggesting that the plant treatments may have had fast removal of phenanthrene during the first ~4-7 hours of exposure (Table 1). The ANOVA analysis on day 0 indicated no statistically significant differences (p = 0.059) but was close to significant. ”  

Line 704-705 (indicating absorption as an alternative mechanism for phenanthrene removal):

“It is possible in this study that the phenanthrene quickly adsorbed to plant roots when introduced.”

Finally, the Discussion section from lines 685 to 748 are dedicated to competing mechanisms that contribute to phenanthrene removal. We certainly feel that we have presented a balanced discussion of mechanisms contributing to phenanthrene removal to the extent possible in the absence of actual measurements of microbially mediated phenanthrene metabolites.

8)    The discussion suggests that changes in dissolved oxygen and conductivity “mirror microbial activity.” However, microbial activity was never quantified. Without supporting evidence such as microbial activity assay, or 16S rRNA quantification, this remains conjectural.

We have considered all of the Discussion section and the only section that the reviewer could be referring to is (Line 626-633):

“It was hypothesized that available oxygen was consumed for aerobic metabolism of phenanthrene and the lack of vegetation or the small root system associated with Sedge B was not able to produce sufficient oxygen release into the rhizosphere, causing oxygen to decline. During this sample period, these microcosms were visibly cloudy which may indicate a microbial bloom. While this cannot be confirmed, the DT50 values identified from the phenanthrene exponential decay models were approximately 5 days for these two treatments (Table 1), so it is clear there was phenanthrene removal occurring, which may have included microbial aerobic metabolism.”

As noted above in our responses to comments #6 and #7, we have presented a balanced interpretation for potential mechanisms of phenanthrene removal and not relied solely on microbial activity as an explanatory factor.

9)    The discussion part of the manuscript never links microcosm behavior to real floating treatment wetlands (FTWs) or natural attenuation processes in situ. Given that the IISD-ELA context is mentioned in the introduction, the lack of scale translation is a major gap. The authors should address this in the revised manuscript.

In our response to Reviewer 1, comment 5, we added a paragraph, and relevant citations, to the Discussion outlining our previous work with microbial communities on roots of plants in floating treatment wetlands exposed to oil (lines 634-641 of the revised manuscript).  

10)     The authors suggest that microbial degradation explains phenanthrene loss in all treatments. However, they also argue that plant roots play a role - despite no difference between planted and unplanted systems. These statements are inconsistent and need reconciliation.

Again, we contend that an even more balanced approach is provided in the revised manuscript as a result of responses to Review #1 comments #6,7 and 8. Moreover, while there was some overlap of the removal rates among microcosms, the removal rate (DT50) was faster for some plants (cattail and Sedge A) than unplanted microcosms (Table 1). These are significant findings that we discuss thoroughly, and in a well-balanced manner throughout the manuscript.

11)     The discussion part of the manuscript lacks integration across parameters. Instead of synthesizing how pH, DO, and phenanthrene trends interact, the authors treat them separately.

I think a conceptual diagram or pathway model (showing potential abiotic and biotic routes of loss) would strengthen this section.

Integrative analysis of all parameters in a single model are not appropriate because water quality metrics were sampled on days 1,4,7,10,13,16,19 but phenanthrene concentrations were determined from samples collected on days 0,2,5,10,21.  Because water quality metrics fluctuated significantly from day to day, it  is not appropriate to explore linear relationships among these parameters.  However, the interactive effects of conductivity, pH, DO and phenanthrene exposure are highlighted in the Generalized Additive Mixed Models (GAMM) presented in Figures S4-S6. We also highlight the importance of the interactions of these parameters on potential phenanthrene removal throughout paragraphs 2-4 of our Discussion.  

Given the potential for multiple routes of phenanthrene removal, noted by the reviewers and acknowledged in our responses, we are reluctant to include a conceptual diagram. However, we can comply by modifying Figure 1 if the editors feel this is a necessary revision.

Reviewer #2 Comments

1)        Please include more data in the abstract to strengthen the overall summary.

As noted in our responses to Reviewer #1, comment 1, we have thoroughly revised the abstract to be more of a concise synthesis of findings from this study.

2)        Kindly add the missing references in the proper format (see line 133 - Error! Reference source not found.).

The text referred to by the reviewer was not a citation but instead was a link to a figure that was broken. It should have read: pipette (Figure S1).

We have corrected this error, and highlighted the changes, in the revised manuscript.

3)        Add appropriate references in the Materials and Methods

We are uncertain what references the Reviewer feels are missing but suspect that they are referring to Section 2.4.1 where details of phenanthrene extractions lack citation support. We have rectified this by adding the following (Lines 187-190 of the revised manuscript):

“Then 25 mL of water was collected, and 20 mL was filtered into a 40 mL glass amber vial with a PTFE lined cap for phenanthrene extraction, using modified methods from Stanley et al [12] for small volumes.”

4)        In the graphs, indicate statistical significance (significant and non-significant) that appears across all graphs - please correct it. Also, there is a spelling error in Figure 2. Minor revisions are required.

Data in panels A, C, and E of Figure 2 are presented to show trends and were not subjected to statistical analysis. However, those collective data are summarized in panels B, D and F of the same figure where we present statistical notation for each parameter based on the mean, and as described in section 2.7.1. We did perform GAMM to assess the interaction of time and treatment on each water quality parameter shown in this figure and those results are presented in SI.

Similarly, Figure 3 was intended to show “trends” of data and was not statistically analyzed for that figure. However, a first order exponential decay model was statistically derived for that data and is presented in Figure 4.  The model was developed by treatment, where microcosms serve as a random effect. As a results, the model outputs only one estimate and confidence interval for each treatment, rather than for each microcosm, so we cannot extract a p value for each model. However, the confidence intervals derived for each treatment can inform whether removal rates between treatments were different. Figure 4E shows the modelled removal rate with confidence interval. Overlap among the confidence intervals of each removal rate value indicates they aren’t different. The same type of rationale is evident in Figures 5 and 6, where statically derived models of plant height are presented over time in the top panels and growth rates and confidence intervals for each treatment are presented in respective the bottom panels of each figure.     

We have thoroughly checked the spelling in Figure 2. I and cannot find a spelling error.

5)        The quality of the figures can be improved for better clarity and readability.

All figures were developed according to criteria in Instruction to Authors. We feel that the figures are clear and understandable but will comply with editorial feedback if specific issues are unresolved.

6)      The Conclusion section is too general; please make it more specific, add numerical values and concise.

We have added the most relevant numerical values, removal rate expressed as the time to remove 50% of the original phenanthrene concentration (DT₅₀ expressed in hours) to the Conclusions section and highlighted those changes in the revised manuscript.  

7)   Although the author has incorporated many references, the overall discussion and critical analysis can be further improved to better highlight and support the results and findings. Add only suitable references at relevant points throughout the manuscript.

As noted in our responses to Reviewer 1, comments 2,5, and 6 we have added 8 additional citations, along with relevant supportive discussion.

Reviewer #3 Comments

1)   Phenanthrene specifically belongs to the group of polycyclic aromatic hydrocarbons. PAC is a broad term.

Phenanthrene is both a PAH and a PAC. The broader term, PAC, refers to PAHs that contain O,S and N, which phenanthrene does not. However, we use the broader PAC term throughout the manuscript because we discuss removal of PACs more generally and wish to avoid defining both PAHs and PACs. Finally, the term more inclusive term PACs is used in our previous work, which we have cited in this manuscript.

2) The introduction should explain why phenanthrene was selected for this study. There are also many other PAHs. Why specifically phenanthrene over other PAHs?

As noted in our response to Reviewer 1, comment #2 we have enhanced our rationale for selecting phenanthrene as a test compound and supported it with additional references.

3) Section 2.2 How are the plants selected? What is the range of height, size, and root mass?

As noted in our response to Reviewer 1, comment 5, early season wild plants were collected from a donor wetland and allowed to establish in our microcosms for 5 weeks. We did not determine initial height, size or root mass because the plants had only just started to emerge.  However, as noted in the text (lines 323 to 327, and Table S3) we acknowledge this shortcoming of our study and normalized parameters to the final biomass in our analysis:

“relationships to final biomass were analyzed for some water quality parameters with a linear model using the stats package (v.4.2.2; [47]). In addition, biomass was not statistically analyzed to determine potential impact of phenanthrene as they were collected from a donor wetland and may have had different initial biomass. Supplementary Table S3 includes summary of final dry biomass by treatment.”

4) Line 133. Recheck citation error “Error!     Reference source not found”

As noted in our response to Reviewer #2, comment 2, the text referred to by the reviewer was not a citation but instead was a link to a figure that was broken. It should have read: pipette (Supplementary Figure 1).

We have corrected this error, and highlighted the changes, in the revised manuscript.

5) Line 430-431. The authors mentioned rapid declines of phenanthrene concentrations within 4-7 hours. How much was it? The first value mentioned by the authors is on 21^st Looking at Fig., I am assuming that day 0 should start from 1.0 mg/L, however, most of the data points do not start from 1.0mg/L, so are these what the authors refer to as rapid decline? What could be the mechanism? Are these just due to the phenanthrene molecules adhering to the surface of the biomass and biofilm via adsorption? Is phenanthrene susceptible to degradation by light?

We have directed the reader to Table S7 where the data are compiled so that the actual decline can be assessed (line 462 of the revised manuscript). Furthermore, we had already specified (now appearing on lines 681-684 of the revised manuscript) in the text:

“In fact, all plant treatments had lower estimated y0 for the exponential decay model than Phenanthrene microcosms during the first round of sampling, suggesting that the plant treatments may have had fast removal of phenanthrene during the first ~4-7 hours of exposure (Table 1).”

With reference to the questions of binding of phenanthrene to biomass, glass, photodegradation, and other relevant removal mechanisms we have already address this thoroughly in response to Reviewer 1, comment 6.

6) It is difficult to understand Fig 3 and Fig 4E. Sedge B definitely have a faster removal rate; however, Fig 4E shows that it is about the same as the control.

As noted in our response to Reviewer 2, comment 4,  Figure 3 is the trend of phenanthrene concentration over time. Figure 4 is the modelled decline of phenanthrene concentrations.  The reviewer may be confused because the initial concentration (y0) of phenanthrene in the Sedge B treatment is lower than the control, suggesting “faster initial removal”. However, the model uses the first measured concentration, and not the nominal concentration of 1 mg/L, resulting in a similar rate calculated over 600 hours. This is thoroughly described in the Results section (lines 478-496 of the revised manuscript) and in the Discussion (lines 678-684 of the revised manuscript).   

7) Fig 5A. It is not clear what the bullet points of different shapes represent. There are diamonds, circles, and squares. I am assuming these are symbols for different plants. These should be mentioned clearly in the figure caption so that the readers do not have to guess.

Comparing figures 5A and 5B, why are plants of such large differences selected?

Each of the different shapes on Figure 5 A and B and Figures 6 A and B represent separate replicates of the same treatment. We have clarified that by adding the following statement to the Figure legends for both Figures, highlighted in the revised manuscript (lines 524 and 532 of ther revised manuscript):  

  “Different symbols represent separate microcosm replicates of the same treatment.”

As we noted in our earlier response to Reviewer 1, comment 5, we did not determine biomass of the newly emergent plants when they were collected during the first week of the growing season. Therefore, the heights shown in Figure 5A are somewhat divergent after 5w of growth.  As we noted in our response to Reviewer 1, comment 5, plant performance metrics and water quality results were normalized to final biomass to control for differences in plant sizes during the actual bioassay.

8) Fig 5 caption mentioned data for 25 days, however, data points stopped at day 19^th. Same comment for Fig 6.

The last data point presented in Figure 5 was collected on day 19, but this figure presents a model of the data for 25 days. This is clearly stated in the figure legend (lines 523-526 of the revised manuscript)

 “Logistic growth model (height) for Cattail (A) and Phenanthrene + Cattail (B) microcosms. Different symbols represent separate microcosm replicates of the same treatment. Estimated intrinsic growth rate (r) and 95% confidence interval whiskers are presented in C. Model trends (A & B) were based on daily predictions for 25 days”

and was further explained in an entire paragraph of Section 2.7.3 Plant Growth (lines 337-346).   

9) Line 744. It is confusing what the authors mean by as cited in 90? [Zazouli et al., 2014, as cited in 90]. Are the authors referring to reference #90? Also, this showed that there is a mix of citation style. Authors should consistently use square bracket number [90]

This confusing citation has now been clarified by simply citing reference [90] in the text.

“… ultimately the capacity to remediate contaminants [90]”.

10) Supplementary: The standard convention across most scientific journal to label a figure in the supplementary section is using a prefix “S” followed by the number, as in “Figure S1”, instead of “Supplementary Figure 1”. It is the same for the table. Authors should do the same for the in-text citations in the main text. See journal requirement https://www.mdpi.com/journal/water/instructions#suppmaterials

We have revised the Figure references and legends as suggested throughout the manuscript, Figure legends and SI document.