Swimming under Pressure: The Sub-Lethal Effects of a Pesticide on the Behaviour of Native and Non-Native Cypriniformes Fish

Round 1

Reviewer 1 Report

The manuscript by Leite and co-authors entitled “Swimming under pressure:  sub-lethal effects of a pesticide on the behavior….” Documents the effect of a short -term exposure to the pyrethroid esfenvalerate on a native (to Portugal) and invasive fish species.  The authors focus on behavioral endpoints with immediate ecological relevance such as shoaling, swimming, and exploratory behaviors.  While the invasive fish species did not experience any alterations in the observed behaviors because of short-term sub-lethal esfenvalerate exposure, high concentrations, which were environmentally relevant, did disturb normal locomotor activities in the native fish species.  These findings are important as they suggest subtle, but ecologically important effect of short-term exposures that may exacerbate sustainability of native fish species under pressure from invasive fish and environmental pollution.

The manuscript is well written, extensively referenced, provides a clear description of its methodology, and provides an in-depth discussion of the findings that is consistent with the observed results.

The lack of confirmatory water chemistry is problematic, especially since the authors chose to use a commercial formula of esfenvalerate (Flower’s SUMIFIVE PLUS; Line 111).  As highlighted in one of the manuscript’s references: “Conversely, esfenvalerate is nearly insoluble in water at 0.006 mg L1 and has a Koc value of 215,000 (ibid.). With these listed properties a significant proportion of esfenvalerate is expected to remain bound to sediment (as well as most surfaces)….” (Reference [13] Brady et al., 2006).  Given these properties, the commercial formulations almost certainly contain inert ingredients that may be biologically active (for example emulsifiers or alkylphenolic substances).  In addition, the extreme hydrophobicity of esfenvalerate makes it likely that at least some of the compound was bound to the experimental surfaces and not bioavailable to the exposed fish, calling into question the concentrations calculated for this study.  Furthermore, since a commercial formulation was used for the two pyrethroid treatments, the control may be compromised as it consisted solely of water, not a water control similar to the inert ingredients found in the commercial formulation.  It would benefit the manuscript if the authors could expand on the total composition of this commercial formulation and provide some literature evidence that the inert ingredients are unlikely to be biologically active (it is my understanding that European pesticide manufacturers must declare inert ingredients in their labeling).

Another point of concern is the comparison of juvenile barbels with adult bleak (Lines 133-134).  The authors justify this approach in pointing out that there are (i) no overlapping life stages of similar size, and (ii) their “frequency of occurrence in Iberian streams.  I assume the first argument is based on concerns about uptake kinetics of esfenvalerate if the fish were of different sizes – this point (if this is indeed why the authors chose different life stages across the two species) should be elaborated on.  I am not sure if uptake kinematics are relevant in exposures lasting only two hours and immediately followed by the behavioral assessment. The second argument is unclear to me in its current phrasing: are the authors implying that there are roughly as many juvenile barbels as there were adult bleak in these rivers (and by logical extension, less adult barbels than adult bleak)?  If that is indeed the case, the lower survival of juvenile barbels in these rivers suggests that the findings of this study are indeed applicable to these waterways – this could be discussed further if the authors agree to the logical chain.

A last point of concern is that barbels were captured in a different river than the bleak.  While this may be logistically been necessary, the authors need to convince the readers that the different sources of the fish do not contribute to the different results for the two species.  What is known about water quality, pollutant burden, and especially esfenvalerate concentrations in these two rivers.  Please expand on this issue to put readers concerns to rest.

Despite these shortcomings, this manuscript has value in that it tests ecologically relevant endpoints in ecologically important species and explores the interplay between pollutant pressures in native and invasive species. Following revisions, the manuscript should be a valuable addition to our understanding of the impacts of pyrethroid pesticides on aquatic life.

Minor suggestions for edits and clarification are provided below.

Line 49:  the authors indicate esfenvalerate concentrations of up to “0.76 g/L [13-15]”. This seems to be extremely high (not microgram/l as in most of the reported studies but g/L) and almost impossible given the extreme hydrophobicity of this compound (A quick check lists a solubility limit of <0.3mg/L) – please double check this concentration and if it is accurate please explain where it was found and why it is so much higher than the majority of studies on this compounds occurrence in the environment.

Lines 185-189: if 4 fish were exposed to each treatment concentration at a time and each experimental replicate consists of 3 treatments (control, low, high), the 36 captured fish per species would result in three replicate exposures.  Is this how the statistical analysis was conducted (i.e., the four fish together in one exposure were averaged and the averages of the three trials were used for analysis) or was each fish handled as a separate individual.  If the latter, I assume a block-effect was included in the development of the generalized linear model (Line 235).  The authors should clarify the statistical methodology further.

Results:  I appreciate that the authors include b and z values.  However, p-values with three significant digits (for example p=0.002) overestimate the statistical power of this study (power of “2 in 1000) consisting of three replicates (or twelve replicates if each fish was treated as an independent replicate).  Just two significant digits (i.e., p<0.01) would more accurately reflect the statistical power of the experiments conducted.

Author Response

The manuscript by Leite and co-authors entitled “Swimming under pressure:  sub-lethal effects of a pesticide on the behavior….” Documents the effect of a short -term exposure to the pyrethroid esfenvalerate on a native (to Portugal) and invasive fish species.  The authors focus on behavioral endpoints with immediate ecological relevance such as shoaling, swimming, and exploratory behaviors.  While the invasive fish species did not experience any alterations in the observed behaviors because of short-term sub-lethal esfenvalerate exposure, high concentrations, which were environmentally relevant, did disturb normal locomotor activities in the native fish species.  These findings are important as they suggest subtle, but ecologically important effect of short-term exposures that may exacerbate sustainability of native fish species under pressure from invasive fish and environmental pollution.

The manuscript is well written, extensively referenced, provides a clear description of its methodology, and provides an in-depth discussion of the findings that is consistent with the observed results.

The lack of confirmatory water chemistry is problematic, especially since the authors chose to use a commercial formula of esfenvalerate (Flower’s SUMIFIVE PLUS; Line 111).  As highlighted in one of the manuscript’s references: “Conversely, esfenvalerate is nearly insoluble in water at 0.006 mg L1 and has a Koc value of 215,000 (ibid.). With these listed properties a significant proportion of esfenvalerate is expected to remain bound to sediment (as well as most surfaces)….” (Reference [13] Brady et al., 2006).  Given these properties, the commercial formulations almost certainly contain inert ingredients that may be biologically active (for example emulsifiers or alkylphenolic substances).  In addition, the extreme hydrophobicity of esfenvalerate makes it likely that at least some of the compound was bound to the experimental surfaces and not bioavailable to the exposed fish, calling into question the concentrations calculated for this study.  Furthermore, since a commercial formulation was used for the two pyrethroid treatments, the control may be compromised as it consisted solely of water, not a water control similar to the inert ingredients found in the commercial formulation.  It would benefit the manuscript if the authors could expand on the total composition of this commercial formulation and provide some literature evidence that the inert ingredients are unlikely to be biologically active (it is my understanding that European pesticide manufacturers must declare inert ingredients in their labeling).

R: Dear reviewer, thank you for your commentary. Regarding the commercial formula, the product we used “is formulated based on a combination of esfenvalerate 5% concentration, phenylethyl-xylene isomers and ethyl phenylethyl”, and “it dissolves easily” as described by the manufacturer (https://www.alchimiaweb.com/en/polyvalent-insecticide-sumifive-15cc-product-12726.php). This information has been added to the text (lines 111-114). This type of insecticide is widely used in Portugal by spraying on the cultures. Our objective was to test the effects of the compound esfenvalerate, considering the runoff from agriculture that leaches into freshwater systems, therefore the concentrations that stay bioavailable in the water.

Another point of concern is the comparison of juvenile barbels with adult bleak (Lines 133-134).  The authors justify this approach in pointing out that there are (i) no overlapping life stages of similar size, and (ii) their “frequency of occurrence in Iberian streams.  I assume the first argument is based on concerns about uptake kinetics of esfenvalerate if the fish were of different sizes – this point (if this is indeed why the authors chose different life stages across the two species) should be elaborated on.  I am not sure if uptake kinematics are relevant in exposures lasting only two hours and immediately followed by the behavioral assessment. The second argument is unclear to me in its current phrasing: are the authors implying that there are roughly as many juvenile barbels as there were adult bleak in these rivers (and by logical extension, less adult barbels than adult bleak)?  If that is indeed the case, the lower survival of juvenile barbels in these rivers suggests that the findings of this study are indeed applicable to these waterways – this could be discussed further if the authors agree to the logical chain.

R: Thank you for your commentary. To clarify, species were captured in different rivers - Rivers Lizandro (barbel) and Sor (bleak) [141-142]. Sampling was based on the known existing communities in these river systems, with stable populations of barbels (non-threatened species by the IUCN Red List) and was dependent on the moment/season of the capture, in this case May 2021. We are not implying the abundances of the species are comparable, however the sizes of the individuals are. Additionally, we also had to consider the scale of the experiments, i.e. dimensions of the flume structure and the proportions of the fish inside it. In this case, adult barbels would be too big to test in groups, affecting the results of the behavioural trials, as the available space in the mesocosm flumes would also be a variable affecting their responses.

A last point of concern is that barbels were captured in a different river than the bleak.  While this may be logistically been necessary, the authors need to convince the readers that the different sources of the fish do not contribute to the different results for the two species.  What is known about water quality, pollutant burden, and especially esfenvalerate concentrations in these two rivers.  Please expand on this issue to put readers concerns to rest.

R: Thank you for your comment. The two rivers from which we sampled the fish are very similar coastal central Portugal systems, with moderate water quality with some agricultural use areas, but pesticides concentration was not measured at sampling sites. The bleak is an exotic species in Portugal, and expanding throughout Iberia; maybe in the future we could test populations of both species from the same river to output more comparable results in terms of water quality and pollutants parameters, but the idea of the present study was to already give insights on the potential competitive advantages bleak may have in relation to barbel when in the presence of the pesticide.

Despite these shortcomings, this manuscript has value in that it tests ecologically relevant endpoints in ecologically important species and explores the interplay between pollutant pressures in native and invasive species. Following revisions, the manuscript should be a valuable addition to our understanding of the impacts of pyrethroid pesticides on aquatic life.

R: Thank you for your suggestions.

Minor suggestions for edits and clarification are provided below.

Line 49:  the authors indicate esfenvalerate concentrations of up to “0.76 g/L [13-15]”. This seems to be extremely high (not microgram/l as in most of the reported studies but g/L) and almost impossible given the extreme hydrophobicity of this compound (A quick check lists a solubility limit of <0.3mg/L) – please double check this concentration and if it is accurate please explain where it was found and why it is so much higher than the majority of studies on this compounds occurrence in the environment.

R: The correct value is indeed 0.76 μg/L. The text has been altered accordingly.

Lines 185-189: if 4 fish were exposed to each treatment concentration at a time and each experimental replicate consists of 3 treatments (control, low, high), the 36 captured fish per species would result in three replicate exposures.  Is this how the statistical analysis was conducted (i.e., the four fish together in one exposure were averaged and the averages of the three trials were used for analysis) or was each fish handled as a separate individual.  If the latter, I assume a block-effect was included in the development of the generalized linear model (Line 235).  The authors should clarify the statistical methodology further.

R: Thank for your comment. The statistical analyses (both GLM and Kruskal-Wallis) considered the fish group as a whole, based on the metrics adopted for each behavioural trait (activity, boldness and shoal cohesion), as mentioned in the Material and Methods section (lines 219-231). As such, three replicates were considered for each treatment. This information has been added in the Experimental set-up section, in lines 211-213 (“Replicates were conducted, being a school of 4 wild-caught fish of each species the experimental unit (number of replicates = 3 schools of 4 fish per treatment, totalling 12 fish per treatment, 72 fish overall”).

In addition, conducting statistical analyses on each fish individually (for activity and boldness), assuming a block-effect as suggested by the reviewer, would require a mixed model (GLMM) which, in turn, would require a higher number of replicates to be more accurate than the analysis conducted in this study (GLM), as the number of fish considered for the block-effect (4) would be higher than the number of replicates for each treatment (3). Shoal cohesion was a group measure and, as such, a block-effect approach would not apply. To clarify, the text “Replicates were conducted, being a school of 4 wild-caught fish of each species the experimental unit (number of replicates = 3 schools of 4 fish per treatment, totalling 12 fish per treatment, 72 fish overall)” in lines 213-215.

Results:  I appreciate that the authors include b and z values.  However, p-values with three significant digits (for example p=0.002) overestimate the statistical power of this study (power of “2 in 1000) consisting of three replicates (or twelve replicates if each fish was treated as an independent replicate).  Just two significant digits (i.e., p<0.01) would more accurately reflect the statistical power of the experiments conducted.

R: Thank you for your suggestions. P-values were changed accordingly.

Reviewer 2 Report

Dear Authors,

The authors investigated the effects of esfenvalerate (1.2 µg/L and 2 µg/L) 
on behavior of the native Iberian barbell (Luciobarbus bocagei) and the non-native invasive bleak (Alburnus alburnus). The results of the paper are potentially interesting for “Fishes” and it is well-aimed. However, there seem to be areas of scientific problems in the manuscript. Before decision on the manuscript, the author should give appropriate answers to the major comments at below.

MAJOR REMARKS

1. As is known, the sensitivity of each living thing to the target chemical is different. Even the response to the same chemical at different stages of development (embryo, larva, and adult) is different. Therefore, the responses of both barbel and bleak to esfenvalerate may differ. In this case, the lethal dose of both fish for esfenvalerate should be determined either by previous studies or by this study.

2. "sub-lethal effects of a pesticide on the behavior of native and non-native cypriniformes fish" In a study examining only the effect of esfenvalerate, the phrase "pesticide" in the title is very assertive. I wonder if it should be changed from pesticide to esfenvalerate?

3.  in the abstract: "The effect of these, namely of pyrethroids, the fourth major group of insecticides in use worldwide, on aquatic fauna, is still unknown." I don't think this statement is correct. Pyrethroids are a highly studied group and there are many studies showing their effects on fish. The authors even mentioned these studies in the manuscript.

4. Letters indicating statistical differences in figures should be indicated for each column.

5. The authors should add details such as chemical and device (model, company, city, and country). 

SOME MINOR COMMENTS

Page 3, line 128, 129, Although literature is usually given numerically, some are given by name and date. Please read the journal guide carefully.

Page 3, line 142, delete “-“ after 150

Page 5, (experimental set up), In the manuscript, dots were used in some numerical expressions and commas were used in others. Please, correct it.

Page 6, line 204, please correct “°C”

Page 8, line 274, 275, 277, This expression "mean ± SD" should be given in the statistical analysis section, its use in the text should be avoided.

Author Response

Dear Authors,

The authors investigated the effects of esfenvalerate (1.2 µg/L and 2 µg/L) 
on behavior of the native Iberian barbell (Luciobarbus bocagei) and the non-native invasive bleak (Alburnus alburnus). The results of the paper are potentially interesting for “Fishes” and it is well-aimed. However, there seem to be areas of scientific problems in the manuscript. Before decision on the manuscript, the author should give appropriate answers to the major comments at below.

R: Dear reviewer, thank you for your commentary.

Major remarks

1. As is known, the sensitivity of each living thing to the target chemical is different. Even the response to the same chemical at different stages of development (embryo, larva, and adult) is different. Therefore, the responses of both barbel and bleak to esfenvalerate may differ. In this case, the lethal dose of both fish for esfenvalerate should be determined either by previous studies or by this study.

R: Recent studies found that “Esfenvalerate has a 96 h-LC50 of 0.1962 μg/L for common carp” (Navruz et al. 2023). However, to our knowledge lethal levels of esfenvalerate have not been determined for L. bocagei or A. alburnus in previous studies. We did not test the lethal dose of pesticide on fish, since our type of licensing does not allow to test mortality of the fish and we intended to limit the stress of the animals, following the ethical standards of testing, and be able to return the native species to the wild after the experiments. Also, please note that the goal of the present study was to address the sub-lethal effects of a previous esfenvalerate exposure to fish behaviour as shown by the different metrics evaluated.

Navruz, F. Z., Acar, Ü., Yılmaz, S., & Kesbiç, O. S. (2023). Dose-dependent stress response of esfenvalerate insecticide on common carp (Cyprinus carpio): Evaluating blood parameters and gene expression. Comparative Biochemistry and Physiology Part C: Toxicology & Pharmacology, 272, 109711.

2. "sub-lethal effects of a pesticide on the behavior of native and non-native cypriniformes fish" In a study examining only the effect of esfenvalerate, the phrase "pesticide" in the title is very assertive. I wonder if it should be changed from pesticide to esfenvalerate?

R: Thank you for your suggestion. Based on previous studies we have referenced in this study – such as, Renick et al. 2015 – we prefer to keep “pesticide” in the title, since we immediately mention that we use esfenvalerate and give more details in the abstract section and keywords, and later generalize in the discussion, as we believe our results can be extrapolated to other pesticide pollution events and be relevant for other studies.

Renick, V. C., Anderson, T. W., Morgan, S. G., & Cherr, G. N. (2015). Interactive effects of pesticide exposure and habitat structure on behavior and predation of a marine larval fish. Ecotoxicology, 24, 391-400.

3. in the abstract: "The effect of these, namely of pyrethroids, the fourth major group of insecticides in use worldwide, on aquatic fauna, is still unknown." I don't think this statement is correct. Pyrethroids are a highly studied group and there are many studies showing their effects on fish. The authors even mentioned these studies in the manuscript.

R: Thank you for the suggestion. The text has been rewritten to ” Pyrethroids are the fourth major group of insecticides in use worldwide and have extremely negative effects on aquatic fauna.”

4. Letters indicating statistical differences in figures should be indicated for each column.

R: We followed the standard for graphical representation of statistical differences, where statistical differences are represented with a symbol or lettering above the respective data (as in some columns figures 2 and 3), and other results with no statistical differences don’t require any symbol or lettering.

5. The authors should add details such as chemical and device (model, company, city, and country).

R: Thank for your commentary. We verified that all equipment used and mentioned in the text has the corresponding information and reference. In addition, this information was added to legends of tables and figures when relevant.

Some minor comments

Page 3, line 128, 129, Although literature is usually given numerically, some are given by name and date. Please read the journal guide carefully.

Page 3, line 142, delete “-“ after 150

Page 5, (experimental set up), In the manuscript, dots were used in some numerical expressions and commas were used in others. Please, correct it.

Page 6, line 204, please correct “°C”

Page 8, line 274, 275, 277, This expression "mean ± SD" should be given in the statistical analysis section, its use in the text should be avoided.

R: Thank you for your notes. Minor revisions were corrected as suggested in the respective sections.

Reviewer 3 Report

The manuscript investigated the effect of sublethal level of esfenvalerate on spontaneous swimming behavior in one native and one non-native cyprinid fish species, which is measured in artificial flume channel mesocosm. This kind of pesticide are now widely used. It can results in impaired the aerobic metabolic capacity due to the uncoupled oxidative phosphorylation, thus may worsen the effect of globe warming effect. i think the manuscript can be accepted for publish after adequate revision. Below are my specific comments.

Line 74-75, the sentence need rewrite, what is the ‘other metabolic traits’ refer to?

Line 75, it seems to me, it is not the definition of boldness, but the variable or the method for measurement of boldness

Line 78-79, I suggest not use shoaling cohesion, but sociability, because boldness, activity, sociability, exploration and aggression are personality traits, the shoal cohesion is a variable of shoal behavior which can be used as an indicator for sociability.

Line 123 2.3. Target species and life-stages It is nice to provide the useful information for two species selected in the study. However, it seems that the information is not enough, especially for non-native bleak. Please provide more information which use this species, rather than other non-native species.

Line 184 I might miss some information here, or such information in elsewhere which I missed? How the concentration achieved? Did the concentration measured?

Table 1, did the authors consider use the parameter as covariate for statistical analysis? 

Line 220-232, when the behavior traits are measured during day time? In the morning, noon or afternoon, the weather might also matter during measurement.

Line 242 The results are kinds of simple. Why only use relative frequency? If use different variables, the effect of species and concentration can be statistically tested by two-way ANOVA.

Line 289, discussion

Line 306, it would be nice to investigate the effect of acclimation temperature on the impairment.

Line 322 yes, the results seems weird. With only one variable, I think the explanation in the next sentences is a little farfetched. I am agreed more with that in line 342-344.

Conclusion, it’s kinds of lengthy

Author Response

The manuscript investigated the effect of sublethal level of esfenvalerate on spontaneous swimming behavior in one native and one non-native cyprinid fish species, which is measured in artificial flume channel mesocosm. This kind of pesticide are now widely used. It can results in impaired the aerobic metabolic capacity due to the uncoupled oxidative phosphorylation, thus may worsen the effect of globe warming effect. i think the manuscript can be accepted for publish after adequate revision. Below are my specific comments.

R: Dear reviewer, thank you for your commentary.

Line 74-75, the sentence need rewrite, what is the ‘other metabolic traits’ refer to?

R: Text has been rewritten.

Line 75, it seems to me, it is not the definition of boldness, but the variable or the method for measurement of boldness.

R: Boldness has been defined as the tendency or willingness to explore unfamiliar environments by several authors over the years. Wilson et al. 1993 describe the shy-bold continuum as such: “Shy individuals react to unfamiliar situations by retreating or becoming quiet and vigilant. Bold individuals act normally or become actively exploratory in the same situations”. Therefore, we follow this definition in our observations for this study, and previous ones as well. However, observing the attempts to swim/jump up the ramp of the flume is just one of the variables we believe to be directly associated with bold behaviour. For example, in Leite et al. 2019 we used a different method to evaluate boldness, by placing barbels inside a shelter (box with a door) and observing the time they took to emerge from inside the chamber.

Wilson, D. S., Coleman, K., Clark, A. B., & Biederman, L. (1993). Shy-bold continuum in pumpkinseed sunfish (Lepomis gibbosus): An ecological study of a psychological trait. Journal of comparative psychology, 107(3), 250.

Leite, T., Santos, J. M., Ferreira, M. T., Canhoto, C., & Branco, P. (2019). Does short-term salinization of freshwater alter the behaviour of the Iberian barbel (Luciobarbus bocagei, Steindachner 1864)?. Science of The Total Environment, 651, 648-655.

Line 78-79, I suggest not use shoaling cohesion, but sociability, because boldness, activity, sociability, exploration and aggression are personality traits, the shoal cohesion is a variable of shoal behavior which can be used as an indicator for sociability.

R: We understand the reviewer’s suggestion, however we prefer to use shoaling cohesion, as it is a more specific concept, which can be better understood by the readers. In addition, we used the result of the shoal cohesion index based on Manek et al. 2014 work, as our variable, as such we think shoal cohesion is the preferred nomenclature for this particular study. We also did text alterations to reflect that the shoal cohesion can be viewed as a sociability indicator.

Manek, A. K., Ferrari, M. C., Niyogi, S., & Chivers, D. P. (2014). The interactive effects of multiple stressors on physiological stress responses and club cell investment in fathead minnows. Science of the Total Environment, 476, 90-97.

Line 123 2.3. Target species and life-stages It is nice to provide the useful information for two species selected in the study. However, it seems that the information is not enough, especially for non-native bleak. Please provide more information which use this species, rather than other non-native species.

R: The bleak is an invasive species in Portugal, and continuously spreading throughout Eurasia waters during recent decades (Latorre et al., 2023), therefore relevant for the type of comparative study considering native and non-native species. Sampling of the bleak was based on our knowledge of the populations of non-native species in the Sor river, and knowledge of this species overall behaviour in captivity from previous experiments using Alburnus alburnus.

Latorre, D., Masó, G., Cano-Barbacil, C., Zamora-Marin, J. M., Almeida, D., Vilizzi, L., ... & Ribeiro, F. (2023). A review and meta-analysis of the environmental biology of bleak Alburnus alburnus in its native and introduced ranges, with reflections on its invasiveness. Reviews in Fish Biology and Fisheries, 1-45.

Line 184 I might miss some information here, or such information in elsewhere which I missed? How the concentration achieved? Did the concentration measured?

R: The insecticide (Flower's SUMIFIVE ® PLUS as an oil in water emulsion (15 mL volume and 5% - i.e., 50 g/L – concentration)) was dissolved in the 50L holding tanks of each treatment, mentioned in Lines 111- 114 and 186-189.

Table 1, did the authors consider use the parameter as covariate for statistical analysis?

R: We did not consider any of the listed parameters because all parameters are comparable between species and trials. Care was taken when doing the experimental design and when conducting the experiments to maintain variations to a minimum to reduce possible experimental biases and uncertainties.

Line 220-232, when the behavior traits are measured during day time? In the morning, noon or afternoon, the weather might also matter during measurement.

R: Experiments were performed in the mornings under similar weather conditions, where the mesocosms facility is mostly shaded during those hours. This was now clarified in the manuscript, as it follows: “All behavioral trials were performed under similar weather conditions” in lines 235-236.

Line 242 The results are kinds of simple. Why only use relative frequency? If use different variables, the effect of species and concentration can be statistically tested by two-way ANOVA.

R: We agree that our results are simple, but, as in previous studies, we wanted to test the effects of one variable, i.e., esfenvalerate concentration, on the behaviour of two species. Therefore, our results directly answer the research question, and we believe the study would not benefit from an increased complexity of behavioural testing. This allow us to be clear in our message, and we think this is more approachable for end users, while maintaining scientific accurateness, thus increasing the real-life implications of the results.

Line 289, discussion

R: We don’t understand the reviewer comments since this line corresponds to the title of the section.

Line 306, it would be nice to investigate the effect of acclimation temperature on the impairment.

R: Thank for your suggestion. Temperature during acclimation did not vary in these experiments and was similar to the temperature in the holding tanks, nonetheless we have done similar experiments in the same flume structure by observing activity, boldness and shoal cohesion of barbels after exposure to high temperature for six consecutive days (Mameri et al. 2020).

Mameri, D., Branco, P., Ferreira, M. T., & Santos, J. M. (2020). Heatwave effects on the swimming behaviour of a Mediterranean freshwater fish, the Iberian barbel Luciobarbus bocagei. Science of the Total Environment, 730, 139152.

Line 322 yes, the results seems weird. With only one variable, I think the explanation in the next sentences is a little farfetched. I am agreed more with that in line 342-344.

R: We appreciate your commentary. Boldness results were distinct from our initial hypothesis [328-333], therefore interpreting these results regarding the intermediate disturbance hypothesis made sense for us. But to improve the overall clarity of the discussion the readers, we removed the lines “Instead, the present result seems to be more consistent with the intermediate disturbance hypothesis [90]. The hypothesis contends that low species richness is maintained by environmental stability at one extreme, or high disturbance at the other, with maximum richness occurring at some intermediate frequency of disturbance.”

Conclusion, it’s kinds of lengthy

R: Thank you for the suggestion. This section has been shortened as suggested.

Round 2

Reviewer 1 Report

The authors addressed all concerns raised by this peer-reviewer.  I deem the manuscript in its current form acceptable for publication.