Changes in the Parasitism Rate and Parasitoid Community Structure of the Horse Chestnut Leafminer, Cameraria ohridella (Lepidoptera: Gracillariidae), in the Czech Republic

Round 1

Reviewer 1 Report

Review of MS -1684878 submitted to Forests- Changes in parasitism rate and parasitoid community structure of the horse chestnut leafminer, Cameraria ohridella (Lepidoptera: Gracillariidae) in the Czech Republic

The MS deals with a topic that affects a lepidopteran species such as horse chestnut leafminer, Cameraria ohridella. The topic is interesting as it helps to know the natural enemies able to control and limit this herbivore. For this species understanding the reasons why it cannot be effectively controlled by parasitoids is important.  Especially in reference to the horse chestnut used as an ornamental tree species, and therefore the moth control conditioned by an anthropized environment with less action of natural enemies.

The authors correctly present a multi-year study carried out in several sites (6) of moth presence and this certainly makes the results more comparable with other studies carried out on the same species.

 

Although this different aspects related to the MS are to be clarified. In several sections of the MS the information provided is lacking and this makes the MS deficient as a whole.

• No reference is made to the varietal sensitivity of the botanical species to herbivores. Which group was monitored considering that other Aesculus species are used for ornamental purposes or even grafted?
• Line 69-71. In this section authors need to be specific. The authors claim to start sampling only after adult flights and this varied from one location to another. It is advisable to specify how the first flights were evaluated and also it is necessary to insert a table with the sampling dates. The species usually overwinters on the fallen leaves on the ground within which the pupae of the moth overwinter. How were the flights of the species evaluated?;
• Was the wintering stadium also collected?;
• In table 1 enter the sampling start dates and all the dates of collection of the leaves. Also specify if these trees were isolated or close to natural areas or parks. This aspect improves understanding of the influence of habitat on the presence of natural enemies. They consider it appropriate that a geo-localized map of the sampled points can also be inserted;
• Line 79. It is necessary to specify better how the parasitization was evaluated. For example “parasitized or suspicious larvae” I don't understand the meaning of suspicious larvae. Are they larvae dead due to presumed parasitization but without parasitoid? In this case should they be counted among the dead larvae?;
• Line 106. Did the authors use the finite formula of SDI (Simpson's diversity index) to compare the set of data obtained for each location? Specify this choice better and also report bibliographical references on the application of this test in the study of parasitoids; Also I wonder why the choice to use this index and not another index;
• Line 111. Remove the sentence as it has already been reported in paragraph 2.5. Statistical analysis
• Line 112. In this paragraph it is not clear to me why the authors do not use the considerable amount of data they have. For example, they can also statistically compare the parasitization between the different stages of the host or relate the dynamics of appearance of the main natural enemies identified;
• Line 171. Figure 1 is superfluous the information is already reported in the table;
• Line 175-177. The relationship between the two species Pediobius saulius and Minotetrastichus frontalis seem not increase over time rather seems an inverse relationship.

Author Response

The MS deals with a topic that affects a lepidopteran species such as horse chestnut leafminer, Cameraria ohridella. The topic is interesting as it helps to know the natural enemies able to control and limit this herbivore. For this species understanding the reasons why it cannot be effectively controlled by parasitoids is important.  Especially in reference to the horse chestnut used as an ornamental tree species, and therefore the moth control conditioned by an anthropized environment with less action of natural enemies.

The authors correctly present a multi-year study carried out in several sites (6) of moth presence and this certainly makes the results more comparable with other studies carried out on the same species.

Response: We appreciate positive review and suggestions for improvement of our manuscript. 

Although this different aspects related to the MS are to be clarified. In several sections of the MS the information provided is lacking and this makes the MS deficient as a whole.

• No reference is made to the varietal sensitivity of the botanical species to herbivores. Which group was monitored considering that other Aesculus species are used for ornamental purposes or even grafted?

Response: Text on other host plants of C. ohridella was added to Introduction. In Methods we added tat only Aesculus hippocastanum trees were sampled.

• Line 69-71. In this section authors need to be specific. The authors claim to start sampling only after adult flights and this varied from one location to another. It is advisable to specify how the first flights were evaluated and also it is necessary to insert a table with the sampling dates. The species usually overwinters on the fallen leaves on the ground within which the pupae of the moth overwinter. How were the flights of the species evaluated?;

Response: Sampling dates were added into Table A1. We looked at adults emerged from green leaves (first generation infesting horse chestnut) as described in Material and Methods.

• Was the wintering stadium also collected?;

Response: No, we did not collect fallen leaves.

• In table 1 enter the sampling start dates and all the dates of collection of the leaves. Also specify if these trees were isolated or close to natural areas or parks. This aspect improves understanding of the influence of habitat on the presence of natural enemies. They consider it appropriate that a geo-localized map of the sampled points can also be inserted;

Response: Map as well as dates of sampling were included.

• Line 79. It is necessary to specify better how the parasitization was evaluated. For example “parasitized or suspicious larvae” I don't understand the meaning of suspicious larvae. Are they larvae dead due to presumed parasitization but without parasitoid? In this case should they be counted among the dead larvae?;

Response: We understand that “suspicious” term was confusing so we deleted it as these larvae were considered to be parasitized (by no or little activity or with different behaviour compared to healthy larvae).

• Line 106. Did the authors use the finite formula of SDI (Simpson's diversity index) to compare the set of data obtained for each location? Specify this choice better and also report bibliographical references on the application of this test in the study of parasitoids; Also I wonder why the choice to use this index and not another index;

Response: As suggested by reviewer 2 we removed this index from the manuscript. Our specialist on this statistics also did not recommend to use any of these indexes because our data are limited, particularly in certain developmental stages and numbers varies among years so indexes for summary data can not tell us much about changes in species diversity.

• Line 111. Remove the sentence as it has already been reported in paragraph 2.5. Statistical analysis

Response: It was removed.

• Line 112. In this paragraph it is not clear to me why the authors do not use the considerable amount of data they have. For example, they can also statistically compare the parasitization between the different stages of the host or relate the dynamics of appearance of the main natural enemies identified;

Response: New statistical analysis was applied to the data.

• Line 171. Figure 1 is superfluous the information is already reported in the table;

Response: We believe pie chart shows better proportion of individual species of summary data but we moved this figure into Appendix.

• Line 175-177. The relationship between the two species Pediobius saulius and Minotetrastichus frontalis seem not increase over time rather seems an inverse relationship.

Response: The text was modified to “the P. saulius/M. frontalis ratio increased over time” which we hope is clear now.

Reviewer 2 Report

The paper by Volter et al reports on the levels of parasitism of chestnut leafminer at 6 sites in the Czeck Republic between 2006 and 2010. The authors detected relatively low levels (aparently) of parasitism and a shift over time from parasitism by Minotetrastichus frontalis to Pediobius sauliu. The authors collected a great number of leaves and demonstrate a high level of expertise in identifying parasitoids. The results add to accumulating knowledge of this relatively new ornamental pest in Europe. Unfortunately, the results by themselves will not be of great interest to reserachers outside the Czeck Republic and some neighbouring states because they concern regional records, and the study is not rigerous enough to adequately test hypotheses concerning a possible competition between the two proncipal parasitoids or shifts in the parasitoid complex. This is compounded by the omission of some important data, for example, what were the densities of leafminers during sampling? At what altitudes were samples collected? Were sites sampled equally each year? What phenological stages of the plants were sampled on each occassion? What were the leafminer densities during sampling? etc.. For example, it is possible that one parasitoid is more prevalent during periods of low miner density and the other during high densities. These sort of questions should be addressed. Indeed, there is no discussion of density dependence in the paper, despite this being a key factor that determines the regulatory capacity of a parasitoid. The statistical analyses are also unconvincing. The authors have not tested hypotheses concerning the effects of sites, the effects of life stages or the effects of sampling periods on parasitism. Such tests require multiple factor analyses and at least a MANOVA design to handle non-independence of parasitism. Overall, the paper seems too simple to be of much interest to a wider readership as it currently stands. The authors might rectify this by putting their results into better context, for example, by rigerously reviewing parasitism from similar records (of which there are now many) and following trends over time and space. Because the results are now nearly 12 years old, the authors need to add some updates of trends also (possibly as part of a review). The discussion for example, introduces a number of factors that may affect parasitism at the different sites (altitude, raking dead leaves, etc.) which only adds to uncertainty as to the value of the results in depicting general trends. Overall therefore, the results are too minor to be of great interest without a more extensive review. I would encourage the authors to present the information in a more contextualized form that adds value to the study.

Other minor issues:

11 which = that

12 its first record = first recorded

14 a five-year survey

18 %

23 an autochtonous

52 etc..

Table 1 – better to include this information on a simple map so that the coverage of the sampling can be ascertained.

79 what does suspicious mean?

Table 2 be consistent with decimal places

Simpson’s index (this type of index tells us very little - I would recomend some rarefraction-related index)

175 spectrum = complex

174-177 we do not see these data/trends

Figure 2 legend: Evolution = composition

Author Response

The paper by Volter et al reports on the levels of parasitism of chestnut leafminer at 6 sites in the Czeck Republic between 2006 and 2010. The authors detected relatively low levels (aparently) of parasitism and a shift over time from parasitism by Minotetrastichus frontalis to Pediobius sauliu. The authors collected a great number of leaves and demonstrate a high level of expertise in identifying parasitoids. The results add to accumulating knowledge of this relatively new ornamental pest in Europe. Unfortunately, the results by themselves will not be of great interest to reserachers outside the Czeck Republic and some neighbouring states because they concern regional records, and the study is not rigerous enough to adequately test hypotheses concerning a possible competition between the two proncipal parasitoids or shifts in the parasitoid complex. This is compounded by the omission of some important data, for example, what were the densities of leafminers during sampling? At what altitudes were samples collected? Were sites sampled equally each year? What phenological stages of the plants were sampled on each occassion? What were the leafminer densities during sampling? etc..

Response: We appreciate very much your valuable comments and included all the above mentioned information into the manuscript.

For example, it is possible that one parasitoid is more prevalent during periods of low miner density and the other during high densities. These sort of questions should be addressed. Indeed, there is no discussion of density dependence in the paper, despite this being a key factor that determines the regulatory capacity of a parasitoid. The statistical analyses are also unconvincing. The authors have not tested hypotheses concerning the effects of sites, the effects of life stages or the effects of sampling periods on parasitism. Such tests require multiple factor analyses and at least a MANOVA design to handle non-independence of parasitism.

Response: Statistical analysis of data (GLM) was included.

Overall, the paper seems too simple to be of much interest to a wider readership as it currently stands. The authors might rectify this by putting their results into better context, for example, by rigerously reviewing parasitism from similar records (of which there are now many) and following trends over time and space. Because the results are now nearly 12 years old, the authors need to add some updates of trends also (possibly as part of a review). The discussion for example, introduces a number of factors that may affect parasitism at the different sites (altitude, raking dead leaves, etc.) which only adds to uncertainty as to the value of the results in depicting general trends. Overall therefore, the results are too minor to be of great interest without a more extensive review. I would encourage the authors to present the information in a more contextualized form that adds value to the study.

Response: Thank you for these comments, we tried to elaborate the discussion more also in context of new information added (altitude, C. ohridela density) and new data analysis.

Other minor issues:

11 which = that

12 its first record = first recorded

14 a five-year survey

18 %

23 an autochtonous

52 etc..

Response: All errors were corrected.

Table 1 – better to include this information on a simple map so that the coverage of the sampling can be ascertained.

Response: New figure (Fig. 1) with a simple map was inserted but Table 1 is kept as it contains additional data, e.g. altitude.

79 what does suspicious mean?

Response: This was removed as it was confusing because in the fact those larvae were considered as parasitized by endoparasitoids due to their low activity/change of behaviour compared to healthy larvae.

Table 2 be consistent with decimal places

Response: Decimal places corrected. Table is now in Appendix.

Simpson’s index (this type of index tells us very little - I would recomend some rarefraction-related index)

Response: As suggested by reviewer 3 we removed this index from the manuscript. Our specialist on this statistics also did not recommend to use any of these indexes because our data are limited, particularly in certain developmental stages and numbers varies among years so indexes for summary data can not tell us much about changes in species diversity.

175 spectrum = complex

Response: Corrected.

174-177 we do not see these data/trends

Response: This text was modified to reflect the data.

Figure 2 legend: Evolution = composition

Response: The text was modified to “Temporal changes of …”

Reviewer 3 Report

The topic of the manuscript (description of parasitoid community of the invasive species Camerata ochridella) is interesting, even if the specific aim of the manuscript (Lines 59-62) seems to be limited specific geographical area (Czech Republic). If this is the only aim of the manuscript, therefore it is probably of interest to a more local Journal; also because, as well reported by authors, the publications on this subject are numerous. However, in my opinion, the data collected by authors are very interesting for a more general discussion about parasitoids of C. ochridella and in general about adaptation of local parasitoids to invasive species. However in the present form this information can not be used and the manuscript need a major revision to be published. The main critics regard the results:

• Authors used a Spearman correlation separately for parasitism and other mortality and only year, but this does not allow to separate the possible effect of the different localities. A linear regression models using for example locations and years as predictors will be more adequate to separate the two effect.
• No analysis has been applied to characterize the differences between locations (if really present). In the discussion (Lines 190-203) authors stated that “Variability in parasitism rates between locations could be due to different climatic. ….”: but no results have been described about these differences except Tab.2 and 3. For example a multivariate approach to detect differences between locations could be useful to see how locations separate in years. Anyway, this part should be better analysed. For example there are any differences in parasitoid community according to altitude of locations or their locations in cities or more natural areas?

 

Introduction: this is well written; with only one exception (reference 29) the cited bibliography is exhaustive and complete; the main topic is well described and the aims are clear.

Materials and Methods: this part can be improved.

• Lines 65-66: from the Tab. 1 is difficult to see that locations represented all main regions of Czech Republic. A map with the locations is surely informative.
• Calculation of parasitism. The author should explain if the Equations have been taken from any publication. If so, which one? Authors must also justify their choice to use these formulas over others (probably less correct, but it should be explained).
• Calculation of species diversity: why author decide to quantify species diversity using Simpson’s index? In the results this value appear only in Tab 3 without any comment and/or discussion in the text. Was it really necessary?
• Line 111: here the statement is interrupted but probably it is a typo error because the same statement is repeated in the next paragraph.
• References [49] and [50] are both necessary?

  

Results: This part should be revised including a more in-depth statistical analysis of the data. Tab 2: is it really necessary? In my opinion it is very dispersive and should be placed in supplementary materials.

 

Discussion: the data collected by authors are compared with those available in the literature. However some statements are not supported by data and this part should be strongly revised with a more general aims in mind (the evolution of parasitoid community and differences between locations).

 

Bibliography: please check carefully the name of species.

 

I’m not a mother language, but I think that English should be check carefully.

Some additional observations are on the attached file.

Comments for author File: Comments.pdf

Author Response

We appreciate all comments and suggestions for improvement of the manuscript.

The topic of the manuscript (description of parasitoid community of the invasive species Camerata ochridella) is interesting, even if the specific aim of the manuscript (Lines 59-62) seems to be limited specific geographical area (Czech Republic). If this is the only aim of the manuscript, therefore it is probably of interest to a more local Journal; also because, as well reported by authors, the publications on this subject are numerous. However, in my opinion, the data collected by authors are very interesting for a more general discussion about parasitoids of C. ochridella and in general about adaptation of local parasitoids to invasive species. However in the present form this information can not be used and the manuscript need a major revision to be published. The main critics regard the results:

• Authors used a Spearman correlation separately for parasitism and other mortality and only year, but this does not allow to separate the possible effect of the different localities. A linear regression models using for example locations and years as predictors will be more adequate to separate the two effect.
• No analysis has been applied to characterize the differences between locations (if really present). In the discussion (Lines 190-203) authors stated that “Variability in parasitism rates between locations could be due to different climatic. ….”: but no results have been described about these differences except Tab.2 and 3. For example a multivariate approach to detect differences between locations could be useful to see how locations separate in years. Anyway, this part should be better analysed. For example there are any differences in parasitoid community according to altitude of locations or their locations in cities or more natural areas?

Response: We appreciate the above critics and revised the manuscript substantially, e.g. new information was included provided (altitude, population density of the host, etc.), generalized linear model was used for statistical analysis. Presentation of results as well as their interpretation was improved, too.

Introduction: this is well written; with only one exception (reference 29) the cited bibliography is exhaustive and complete; the main topic is well described and the aims are clear.

Response: Thank you for positive comment. Reference 29 was replaced by more general one and some text was enhanced.

Materials and Methods: this part can be improved.

• Lines 65-66: from the Tab. 1 is difficult to see that locations represented all main regions of Czech Republic. A map with the locations is surely informative.
• Calculation of parasitism. The author should explain if the Equations have been taken from any publication. If so, which one? Authors must also justify their choice to use these formulas over others (probably less correct, but it should be explained).
• Calculation of species diversity: why author decide to quantify species diversity using Simpson’s index? In the results this value appear only in Tab 3 without any comment and/or discussion in the text. Was it really necessary?
• Line 111: here the statement is interrupted but probably it is a typo error because the same statement is repeated in the next paragraph.
• References [49] and [50] are both necessary?

Response:  Thank you for all suggestions. Map with sites was added, reference to equations was given. We discussed the species diversity calculation with our ecologist and finally decided to removed this index from Table 3. Reference to SAS Studio user's guide was removed.

Results: This part should be revised including a more in-depth statistical analysis of the data. Tab 2: is it really necessary? In my opinion it is very dispersive and should be placed in supplementary materials.

Response:  Data were analysed with respect to geographical factors and other predictors using generalized linear model. Table 2 was moved to Appendix. In addition, Figure 1 was moved to Appendix, too and new Figure 2 on total parasitism and other mortality rates and leaf damage inflicted by C. ohridella larvae was added.

Discussion: the data collected by authors are compared with those available in the literature. However some statements are not supported by data and this part should be strongly revised with a more general aims in mind (the evolution of parasitoid community and differences between locations).

Response: Discussion was amended to incorporate the results of new statistical analyses. Some relevant and recent references were added.

Bibliography: please check carefully the name of species.

Response: Unfortunately, capitalization is made automatically by the journal style provided by MDPI publisher for Zotero. It can be fixed manually during copyediting and/or proofreading.

I’m not a mother language, but I think that English should be check carefully.

Response: We tried to improved the English but if it is necessary, we will send the manuscript to AJE for professional editing.

Some additional observations are on the attached file.

Response: Corrections in the manuscript were done according to commented file.

Round 2

Reviewer 1 Report

Review of MS -1684878R2 submitted to Forests- Changes in parasitism rate and parasitoid community structure of the horse chestnut leafminer, Cameraria ohridella (Lepidoptera: Gracillariidae) in the Czech Republic

 

The authors have revised their MS and in this new version the MS is improved. Only a few more suggestions are provided for this new version.

In figure 1 report latitude and longitude. To simplify, it is sufficient to report the values in the graph on the x axis (longitude) and on the y axis the latitude

In the tables and also throughout the text, round up to the third decimal place; for example P <0.001 and not P <0.0001;

Line 155-158. Since this is the use of multiple variables, it is necessary to specify main terms and covariates.

Author Response

The authors have revised their MS and in this new version the MS is improved. Only a few more suggestions are provided for this new version.

In figure 1 report latitude and longitude. To simplify, it is sufficient to report the values in the graph on the x axis (longitude) and on the y axis the latitude

Response: We appreciate the positive review on the revised version and suggestion to improve Figure 1. Grid indicating geographical coordinates (in blue color) was added to the figure.

In the tables and also throughout the text, round up to the third decimal place; for example P <0.001 and not P <0.0001;

Response: Numbers were rouded up as suggested.

Line 155-158. Since this is the use of multiple variables, it is necessary to specify main terms and covariates.

Response: All variables used were predictors and were continuous (not categorical) so there were no main terms and covariates (like in ANCOVA) in the model.

Reviewer 2 Report

I congratulate the authors on an excellent revision of the paper. The new analysis has added depth and purpose to the document and makes the study much more widely interesting. The authors should have the revised document checked for some language errors to complete their revision.

Author Response

Thank you for very much for a positive review. We are glad the revised version has been improved. As regards English, we will arrange language editing with MDPI or AJE.

Reviewer 3 Report

The manuscript has improved significantly over the previous version. The manuscript needs few corrections to be published:
1) the authors did not test whether the predictors are autocorrelated. This must be done necessarily because there may be factors that are related and therefore a subset of the variables must be selected.

2) Most predictors are clear, but some should be better explained how have been evaluated/calculated. In particular the greenery area surrounding the sampled trees and the greenery care. The first one is qualitative or quantitative (e.g. percentage of woods area in a xx radius). The second one should be better explained: how many mowing in a year? etc.

3) Lines 204-207: From what the authors write, it seems that the pupal stage is the most attacked, but there may be koinobiont parasitoids that attack the larva and flicker in the pupal stage. This part needs to be reformulated.

I suggest also a check of English by a mother language.

Author Response

The manuscript has improved significantly over the previous version. The manuscript needs few corrections to be published:

1) the authors did not test whether the predictors are autocorrelated. This must be done necessarily because there may be factors that are related and therefore a subset of the variables must be selected.

Response: Thank you for this suggestion. We reanalyzed the data using more appropriate GEE model. Both Methods and Results section were corrected.

2) Most predictors are clear, but some should be better explained how have been evaluated/calculated. In particular the greenery area surrounding the sampled trees and the greenery care. The first one is qualitative or quantitative (e.g. percentage of woods area in a xx radius). The second one should be better explained: how many mowing in a year? etc.

Response: As regards greenery area, it is quantitative variable (in hectares) - we measured size of the continuous area (in hectares) surrounding the sampled trees covered by lawn, shrubs and trees. The measurement was done on mapy.cz using area measure tool. The idea of percentage in some apriory set radius would be also interesting although would require some additional processing, e.g. image analysis.. As regards intensity of greenery care, it was as follows: 0 – no mowing or raking, 1 – mowing 1-2 times per season, 2 – mowing >2 times per season. We added these explanations into Table 1 footnotes.

3) Lines 204-207: From what the authors write, it seems that the pupal stage is the most attacked, but there may be koinobiont parasitoids that attack the larva and flicker in the pupal stage. This part needs to be reformulated.

Response: We reformulated this text and some explanation was added into Discussion “This species can develop as koinobiont, i.e. it can attack some younger stages of the host and finish its development in host pupa, in contrast to other parasitoids reared during this study and  considered to be idiobionts. Because P. saulius was also dominant species, the parasitism rates observed in individual stages of C. ohridella might thus be somehow biased.”

 

I suggest also a check of English by a mother language.

Response: We will arrange language editing with MDPI or American Journal Experts service.