Herbicide Uptake (Flashback) in Non-Target Hardwood Species within South Carolina Floodplains

Round 1

Reviewer 1 Report (Previous Reviewer 3)

Comments and Suggestions for Authors

Please see the attachment.

Comments for author File: Comments.pdf

Author Response

Lines 166-182 (cut and moved paragraphs on triclopyr degradation to Material and Methods section (lines 177-179 ). Added absorption in organic matter soils with citation (Helling 2005) and expected lower half life in lower organic matter soils. (lines 82-83; 532-533)

Line 123: moved last scientific question to first

Lines 123-127: added numeric listing prior to all scientific questions

Lines 135-138: moved the South Carolina climate conditions paragraph to the beginning of the Study Area subsection

Various locations: removed duplicated scientific tree names

Line 218: changed to 2.54 cm depth

Line 263: capitalized Table 1

Line 297: added 2.54 cm but also kept one inch

Lines 264: changed dbh in title and column heading

Line 289: It is our opinion that the legend be included in case the reader “misses” the categories beneath the figures

Line 292: removed or added spaces; deleted extra comma

Line 301: changed “one unit” to 2.54 cm

Figure 3 title: I am not sure what the reviewer is referring to? I do not see a deviation from the abbreviations.

Line 330: deleted heading 4. Treated Trees and renumbered remaining headings

Line 239: added “Figure 4”

Line 376: changed feet value to meters

Line 403: changed feet value to meters

Line 451: added Figure 8 to depict the underplanted oak seedlings

Line 484: added Figure 6&7 to depict detrimental effects to residual trees

Line 141: moved figure 9 (now figure 1) to Materials and Methods after mention of the individual sites; renumbered all subsequent figures. The authors respectfully believe that the pictures representing crown damage/mortality are fine following the Conclusion section.

Literature Cited: deleted commas and added semicolons between authors; removed spaces between numeric volumes/pages; removed “and” for two author citations.

Author Response File: Author Response.pdf

Reviewer 2 Report (Previous Reviewer 2)

Comments and Suggestions for Authors

L33-L35: I believe it is entirely inappropriate to substantiate that the effect of triclopyr on non-target vegetation was not shown in a study that is more than 20 years old. Science has advanced significantly since then, and numerous studies demonstrate the effect of triclopyr on non-target vegetation/trees: Oberweger et al. 2023; Graziano et al. 2022 (cited by the authors); Isbister et al. 2017; ...). It would be appropriate to include in the introduction what is already known on the topic and then clearly delineate this manuscript against previous literature - what is innovative and how does it investigate things differently? Defining this in just one sentence is insufficient.

L50-L60: Here, the impact of triclopyr (or rather the lack of impact) on non-target trees is discussed in more detail after the topic has already been introduced earlier, and after a digression to other herbicides, the authors return to it. In my opinion, the introduction could be better organized. For example, first discuss that herbicides strongly affect non-target trees. Then state that triclopyr was long thought to have no negative impacts. Nowadays, studies are emerging that indicate it does have negative consequences. Therefore, the authors decided to explore this topic in-depth... An introduction structured in this way would be consistent with the narrative style used in science.

Lines L64-L72: Does ring girdling have a direct connection with the study's topic? I understand it might relate to the authors' second objective, but an uninformed reader would not know this. If ring girdling is to remain in the introduction, it should be more thoroughly discussed that combining this method with herbicide use might have different impacts on non-target trees than herbicide application without ring girdling. It should be noted that this has been observed with other herbicides, but nothing is known about this effect with triclopyr. This would clearly explain why this paragraph should interest the reader. Therefore, it is questionable whether this belongs in the introduction, as the direct connection as currently written is missing. If the authors wish to discuss it, mentioning it in the discussion might be sufficient.

L67: I understand what probably happened here - the reviewer likely pointed out that inches are not an SI unit. However, an uninformed reader might find it odd why such a precise measure is used here. The best approach would be to state that in the USA, it is defined as 6 inches, which is approximately 15.24 cm in Europe.

L73-L86: This is another example of poorly managed structure in the introduction - triclopyr has already been mentioned three times in the manuscript before it is officially introduced here. The subtopic of triclopyr should have been opened with the information provided in this section - starting with what it is, then stating that its effects on non-target trees were not previously observed, but later they were, which is why the authors are studying it...

L138-L142: This paragraph was probably added at the suggestion of the second reviewer. Although it contains essential information, it seems more general than the information provided earlier, and I think it should be better integrated into the rest of the text. As it is, it feels like the authors tried to meet the request in the simplest way possible and pasted the text in the first convenient place without attempting deeper integration.

L147-L149: Unclear and difficult to understand

L149-L150: Codominant or better? Again, this text is difficult to understand - did the authors mean codominant and more abundant (up to eudominant)? If so, they should better define what percentage representation "codominant" means.

L173-L192: Principally, I like the data analysis and do not find anything seriously wrong with it, except for the concerns I mentioned previously. These concerns were related to my worries that some variables might be correlated with each other (the size of the non-target tree and surrounding trees - see my previous review, the amount of herbicide used and the size of the trees - as the authors themselves defined in the methodology). I think it would be appropriate to describe and display these correlations. Conducting a correlation analysis and presenting the results would clarify whether these suspected correlations exist. If no significant correlations are found, it would dispel any doubts about the independence of the variables. Conversely, if correlations do exist, the authors should account for them in their analysis, possibly through multicollinearity diagnostics or adjusted models.

L214-L221: This might belong more appropriately in the methodology section (where it is already mentioned).

Figure 2: If there are partial correlations (or more precisely, collinearities) among the explanatory variables, then depicting the influence of these variables on raw data might not be the most appropriate approach. This is because some effects could be obscured by other variables. The estimates of the model for each variable are provided separately below (presumably accounting for the influence of the other variables), and as mentioned in my previous review, I would suggest that Figure 2 may be redundant (as it only serves as exploratory data analysis) in the presence of Figure 3... However, completely dismissing Figure 2 might overlook its exploratory value. A balanced approach could involve retaining Figure 2 but clearly explaining its exploratory nature and the limitations due to potential collinearity, and emphasizing the importance of Figure 3 for adjusted insights.

L229-L252: These two paragraphs largely overlap. Both focus on the influence of tree diameter and the number of treated individuals on the probability of herbicide effect on untreated leave trees. Specific details and figures are provided in both paragraphs, though they are phrased slightly differently. I would recommend that the authors consider merging them to avoid redundancy and make the text more consistent and comprehensible.

Figure 5: In this figure, it would be appropriate for the authors to emphasize the red dots and lines (mean and CI) to make them more visible in the graph. Additionally, I would like to reiterate my suggestion from the previous review that adding some jitter to the black dots would be highly beneficial. As it stands, we see only six precisely placed dots, which is completely uninformative.

 

Discussion: I recommend that the authors consider including a discussion on the potential distance over which triclopyr can influence non-target trees. Even without specific statistical data, they could provide qualitative observations from the study. For instance, they might discuss any notable patterns observed at the edges of the study areas or any apparent trends in tree health relative to distance from treated individuals. Including such information would enhance the comprehensiveness of the discussion and provide readers with a clearer understanding of triclopyr's spatial impact.

Author Response

L33-L35: I believe it is entirely inappropriate to substantiate that the effect of triclopyr on non-target vegetation was not shown in a study that is more than 20 years old. Science has advanced significantly since then, and numerous studies demonstrate the effect of triclopyr on non-target vegetation/trees: Oberweger et al. 2023; Graziano et al. 2022 (cited by the authors); Isbister et al. 2017; ...). It would be appropriate to include in the introduction what is already known on the topic and then clearly delineate this manuscript against previous literature - what is innovative and how does it investigate things differently? Defining this in just one sentence is insufficient.

Lines 79-85: Added Oberweger et al. study on triclopyr acid and Lines 88-93:observations from this study using less herbicide (cut stem vs. basal bark). Triclopyr acid is a newer formulation whereas amine and ester formulations have been commercially available for a longer time.

L50-L60: Here, the impact of triclopyr (or rather the lack of impact) on non-target trees is discussed in more detail after the topic has already been introduced earlier, and after a digression to other herbicides, the authors return to it. In my opinion, the introduction could be better organized. For example, first discuss that herbicides strongly affect non-target trees. Then state that triclopyr was long thought to have no negative impacts. Nowadays, studies are emerging that indicate it does have negative consequences. Therefore, the authors decided to explore this topic in-depth... An introduction structured in this way would be consistent with the narrative style used in science.

Lines 33-44: introduce the method of deadening and flashback potential; Lines 45-53: added narrative on studies involving flashback; Lines 54-67: flashback by soil active herbicides; Lines 68-95: flashback potential by triclopyr herbicides and relation to this study; Lines 122-127: scientific questions involving triclopyr’s effect

Lines L64-L72: Does ring girdling have a direct connection with the study's topic? I understand it might relate to the authors' second objective, but an uninformed reader would not know this. If ring girdling is to remain in the introduction, it should be more thoroughly discussed that combining this method with herbicide use might have different impacts on non-target trees than herbicide application without ring girdling. It should be noted that this has been observed with other herbicides, but nothing is known about this effect with triclopyr. This would clearly explain why this paragraph should interest the reader. Therefore, it is questionable whether this belongs in the introduction, as the direct connection as currently written is missing. If the authors wish to discuss it, mentioning it in the discussion might be sufficient.

Lines 107-109: added statement on type of herbicide and increased amount needed for ring girdling.

L67: I understand what probably happened here - the reviewer likely pointed out that inches are not an SI unit. However, an uninformed reader might find it odd why such a precise measure is used here. The best approach would be to state that in the USA, it is defined as 6 inches, which is approximately 15.24 cm in Europe.

We have added SI units but also incorporated English units throughout document

L73-L86: This is another example of poorly managed structure in the introduction - triclopyr has already been mentioned three times in the manuscript before it is officially introduced here. The subtopic of triclopyr should have been opened with the information provided in this section - starting with what it is, then stating that its effects on non-target trees were not previously observed, but later they were, which is why the authors are studying it...

The introduction section has been reordered. In my opinion, this has been adequately addressed.

L138-L142: This paragraph was probably added at the suggestion of the second reviewer. Although it contains essential information, it seems more general than the information provided earlier, and I think it should be better integrated into the rest of the text. As it is, it feels like the authors tried to meet the request in the simplest way possible and pasted the text in the first convenient place without attempting deeper integration.

Lines 144-169: added some more detail on procedure; the description is adequate for reader comprehension; pithy

L147-L149: Unclear and difficult to understand

Percent solution and application rate of targeted stems are given…not sure what is difficult to understand???

L149-L150: Codominant or better? Again, this text is difficult to understand - did the authors mean codominant and more abundant (up to eudominant)? If so, they should better define what percentage representation "codominant" means.

Lines 163-167: better definition of codominant trees vs dominant trees. Individual leave trees were not classified during data collection.

L173-L192: Principally, I like the data analysis and do not find anything seriously wrong with it, except for the concerns I mentioned previously. These concerns were related to my worries that some variables might be correlated with each other (the size of the non-target tree and surrounding trees - see my previous review, the amount of herbicide used and the size of the trees - as the authors themselves defined in the methodology). I think it would be appropriate to describe and display these correlations. Conducting a correlation analysis and presenting the results would clarify whether these suspected correlations exist. If no significant correlations are found, it would dispel any doubts about the independence of the variables. Conversely, if correlations do exist, the authors should account for them in their analysis, possibly through multicollinearity diagnostics or adjusted models.

We have addressed this comment with statistics in our previous review. To address the reviewer’s concern, we have added the following line in section 2.3.Data Analysis

“To ensure that there is no multicollinearity issue, we looked at the correlation coefficient between our independent variables and the variance inflation factors (VIFs). Only NumTreated and NHack had higher correlation (-0.74); all other correlations were less than +/- 0.58. The VIFs were all less than 2.8; therefore, we concluded that we did not have the multicollinearity issue.”

L214-L221: This might belong more appropriately in the methodology section (where it is already mentioned).

Thank you for the comment. We deleted these sentences from the result section.

Figure 2: If there are partial correlations (or more precisely, collinearities) among the explanatory variables, then depicting the influence of these variables on raw data might not be the most appropriate approach. This is because some effects could be obscured by other variables. The estimates of the model for each variable are provided separately below (presumably accounting for the influence of the other variables), and as mentioned in my previous review, I would suggest that Figure 2 may be redundant (as it only serves as exploratory data analysis) in the presence of Figure 3... However, completely dismissing Figure 2 might overlook its exploratory value. A balanced approach could involve retaining Figure 2 but clearly explaining its exploratory nature and the limitations due to potential collinearity, and emphasizing the importance of Figure 3 for adjusted insights.

In the text, we acknowledge that Figure 2 is exploratory data analysis and the trend we see in Figure 2 is further supported by logistic regression.

L229-L252: These two paragraphs largely overlap. Both focus on the influence of tree diameter and the number of treated individuals on the probability of herbicide effect on untreated leave trees. Specific details and figures are provided in both paragraphs, though they are phrased slightly differently. I would recommend that the authors consider merging them to avoid redundancy and make the text more consistent and comprehensible.

We modified and rewrote the paragraph to remove the overlap.

Figure 5: In this figure, it would be appropriate for the authors to emphasize the red dots and lines (mean and CI) to make them more visible in the graph. Additionally, I would like to reiterate my suggestion from the previous review that adding some jitter to the black dots would be highly beneficial. As it stands, we see only six precisely placed dots, which is completely uninformative.

 We changed Figure 5 according to the suggestions. We emphasized the red dots and lines and added jitter to the black dots

Discussion: I recommend that the authors consider including a discussion on the potential distance over which triclopyr can influence non-target trees. Even without specific statistical data, they could provide qualitative observations from the study. For instance, they might discuss any notable patterns observed at the edges of the study areas or any apparent trends in tree health relative to distance from treated individuals. Including such information would enhance the comprehensiveness of the discussion and provide readers with a clearer understanding of triclopyr's spatial impact.

Lines 42-43; lines 372-373: damage zone limited to contact of untreated trees’ roots adjacent to targeted trees.

Round 2

Reviewer 2 Report (Previous Reviewer 2)

Comments and Suggestions for Authors

The authors have appropriately incorporated most of my comments; I find the manuscript publishable, with only two remaining comments.

Perhaps it is because I am not a native speaker, but the sentence 'There were initially two targeted residual, live stand (residual) basal areas approximately 4.6 m² and 9.19 m² of basal area per hectare on each location. Residual “leave” trees were stems considered codominant or better.' seems to have unnecessarily high cognitive load, as I previously suggested. Does it mean the same as 'The initial goal was to maintain residual stands with basal areas of approximately 4.6 m² and 9.19 m² per hectare at each location. The trees that were left standing were those considered to be codominant or better.'? ... I also kindly ask the authors to consider the unambiguous comprehensibility of their statements, or better yet, to have the text, which they personally find fully understandable, read by an independent party before using inappropriate triple question marks in their response.

I couldn't find this part in the text anywhere, allegedly inserted by the authors: 'To ensure that there is no multicollinearity issue, we looked at the correlation coefficient...

Author Response

Dear Editor

We have left everything in track changes so that you can see the changes we made. We addressed the reviewer's comment in addition to some editorial changes. Below we explain in detail.

Reviewers Comments

Perhaps it is because I am not a native speaker, but the sentence 'There were initially two targeted residual, live stand (residual) basal areas approximately 4.6 m² and 9.19 m² of basal area per hectare on each location. Residual “leave” trees were stems considered codominant or better.' seems to have unnecessarily high cognitive load, as I previously suggested. Does it mean the same as 'The initial goal was to maintain residual stands with basal areas of approximately 4.6 m² and 9.19 m² per hectare at each location. The trees that were left standing were those considered to be codominant or better.'? ... I also kindly ask the authors to consider the unambiguous comprehensibility of their statements, or better yet, to have the text, which they personally find fully understandable, read by an independent party before using inappropriate triple question marks in their response.

We changed the text in Line 214-219 according to the reviewer’s suggestion.

I couldn't find this part in the text anywhere, allegedly inserted by the authors: 'To ensure that there is no multicollinearity issue, we looked at the correlation coefficient...

Sorry, we forgot to insert this in the text earlier. We added this to Data Analysis, at the end of the first paragraph (lines 266-271 and Lines 278-279).

 

During the review process, we added things to the manuscript to satisfy the reviewers. Therefore, some information was repeated in several places. We cleaned redundant information and moved paragraphs (in methods) to be coherent. Additional things we edited besides the reviewers' comments without changing the manuscript are given below.

Deleted lines line 92-96 from Introduction.

Moved lines 99-109 to “Treatments” section lines 237-239 to complete ring girdling information presented at one location.

Reshuffled paragraphs in Methods section to be coherent, and removed redundant information, without changing any meaning and additional information) (Lines 141-150, Lines 193-209)

Discussion and Conclusion

Lines 471-496: added potential explanation of central mycorrhizal networks (CMNs) contributing to herbicide transference; reduction with natural mortality/less connectedness of larger trees

Lines 540-542: added CMNs info into Conclusion

Lines 623-624: Added Karst et al. citation

Author Response File: Author Response.pdf

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

Comments and Suggestions for Authors

1. Please include structure activity relationship of triclopyr acid for translocation in plants.

2. Correlation of herbicide with various soil bio-chemical properties may be mentioned.

Reviewer 2 Report

Comments and Suggestions for Authors

L08: The second author is only listed in the affiliations; he/she should probably be named in the line where the authors are enumerated.

L15: The term "potential trees" sounds unclear – whether it refers to existing trees that could be damaged or trees that have not yet been planted. A better formulation might specify "non-target trees" (trees not aimed at by the herbicide) or "surrounding vegetation" to more clearly delineate which trees are being discussed.

L16 and elsewhere: Number ranges 14 – 20 → 14–20

L34: Authors should certainly find support in the literature for the given statement, not only in their study.

L41: Citations should not look like this (not "and others" but rather "et al."), and for Forests, citation numbering is used. Authors should consult the guidelines for authors on this.

L71: The study's explicitly defined goals and hypotheses are missing.

L104: The abbreviation should be introduced in the text once - and DBH has already been introduced by the authors. Moreover, there's inconsistency – earlier as dbh, here as DBH; the authors should unify this.

L105: Once the abbreviation is introduced, it should be consistently used; otherwise, its introduction was unnecessary.

L158: For similar graphs, it would be good to include error bars.

L161-L166: This belongs in the methodology section.

L167 and onwards: In the results section (and consequently, in the methodology of data analysis), I miss a description of:

A)     How the effects were tested - using Type I ANOVA (Incorporating the effect of each subsequent variable after accounting for the preceding ones) or Type II ANOVA (Incorporating the effect of each variable after subtracting the effects of all others).

B)     I can imagine collinearity between some variables – in such a case, using stepwise selection based on AIC might be somewhat contentious – it would be appropriate to first display potential correlations among explanatory variables, test them, and perhaps exclude some in further phases.

C)     There was also the possibility for a candidate model containing all variables to measure and evaluate the Variance Inflation Factor (VIF).

L171: Figure 2: I don't understand the significance of this graph – predictions of the proportions of 0 and 1 in the graph below are more important; this seems more like exploratory data analysis. Also, I don't understand why the authors didn't proceed symmetrically for both of their models – for the first one, they present both this exploratory graph and the model estimates, but for the second one, they provide no graph.

L193: At least for these two variables, I expect strong collinearity – larger trees are likely to be surrounded by larger trees (if forest management in the USA is similar to that in Europe) – what about some correlation or VIF?

L202: Figure 3: Are these graphs individually displayed after accounting for the influence of other variables? It would be appropriate to accompany such graphs with a display of the original points (for example, with some jitter, if it concerns 0 and 1).

L222: Why aren't the same types of graphs shown for the second model as for the first model (even though here there might be just one)?

L286: Redundant use of words expressing time ("more readily" and "more quickly").

 

L346-L348: While these images are nice, they have no place in the manuscript without any caption.

Reviewer 3 Report

Comments and Suggestions for Authors

Dear Authors!

The manuscript (Manuscript ID: forests-2953445) titled "Herbicide uptake (flashback) in non-target hardwood species within South Carolina floodplains" explores the effect of triclopyr acid herbicide on non-target tree species in three different forest stands in South Carolina.

The study is interesting and the results are both significant and valuable for forest management. The findings enable us to predict the effects of the herbicide triclopyr on selected tree species. Therefore, the study holds practical value as it aids in better planning for the use of triclopyr in similar forests in the future. Nevertheless, the study does have some shortcomings, which I will address below.

Major issues:

·        The entire manuscript needs careful proofreading and should adhere to the journal guidelines for authors.

·        SI units should be consistently used throughout.

·        It is recommended to use scientific names instead of common names for tree species.

·        Consistency in discussing topics across all chapters (introduction, methods, results, and discussion) would enhance the study's readability and comprehension.

 

·        The introduction requires improvement to better introduce the topic and articulate specific study aims (hypotheses). Additionally, it lacks an adequate number of references.

·        The methods section needs significant enhancement, particularly in describing the treatments with greater precision.

Specific comments are in the attached pdf file.

Comments for author File: Comments.pdf

Reviewer 4 Report

Comments and Suggestions for Authors

The study contains results that support current information in terms of subject and research.

 

It is very difficult to determine the effect of control studies of perennial and woody plants within 2 years. Therefore, more years of work are required.

 

Title of the article:

 

Recommended: ‘Evaluation of the phytotoxicity effects of herbicide containing triclopyr active substance on non-target trees’’

 

Abstract:

 

Briefly, herbicide dosage and application methods should also be mentioned.

 

Materials and Methods:

 

It should be emphasized that the effective substance triclopyr is an auxin-type herbicide that is absorbed by the roots and leaves, migrates throughout the plant and accumulates in the meristematic tissues of the root and shoot, and its degradation time may vary depending on soil characteristics.

 

Conclusion

 

Finally, conclusion must be given.

 

Specific comments:

 

1. An attempt was made to determine the effectiveness of applications (50% triclopyr acid (Trycera®) and 50% water solution) to reduce or eliminate the phytotoxicity of herbicide applications to non-target targets.

2. Determination of the effects of reduced doses of herbicide and doses mixed with 50% water on trees of different sizes presented an original result.

3. Provides validation of other published results and applicable method.

4. In the methodology, a richer alternative can be offered to reduce the damage to non-terget trees.

5. The results are consistent with the goal and purpose of the trial.

6. References are appropriate to the article.

7. Tables and figures support the research.