Diversity of Thrips Species Associated with Soybean Grown in Different Plant Arrangements at Various Phenological Stages

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

 

This is an important article on the study of thrips on soybeans in Europe. Currently, thrips on soybean, including both phytophagous and predatory species, increase dramatically in number during flowering, but many phytophagous thrips species are extremely poorly studied, and information on patterns of occurrence is not available. In fact, T. flavus, T. nigropilosus, F. intonsa, these three are common soybean pests in East Asia. There have been major outbreaks in history and have also been research reports consistently. Due to language limitations, the authors are unable to access this information.

There are only some minor issues in the manuscript, which I hope the author can supplement or revise.

1. Introduce more info on the overview of thrips in Poland in the introduction.

2. Supplement the list and pictures of thrips investigated in this article in appropriate text. Figure S1 is not incomplete.

3. Supplement the occurrence dynamics of dominant thrips species in the results.

4. What is the vegetation around soybean fields and how can it influence the occurrence of thrips?

5. According to the survey results, which species other than O. loti are likely to become the main pests of Polish soybeans in the future, and what prevention and control measures are in place?

6. The occurrence of thrips is closely related to planting density. What suggestions can the survey results of the manuscript provide for soybean cultivation and breeding scientists?

7. The letters in the figures are small, please replace it.

 

Author Response

This is an important article on the study of thrips on soybeans in Europe. Currently, thrips on soybean, including both phytophagous and predatory species, increase dramatically in number during flowering, but many phytophagous thrips species are extremely poorly studied, and information on patterns of occurrence is not available. In fact, T. flavus, T. nigropilosus, F. intonsa, these three are common soybean pests in East Asia. There have been major outbreaks in history and have also been research reports consistently. Due to language limitations, the authors are unable to access this information.

There are only some minor issues in the manuscript, which I hope the author can supplement or revise.

1. Introduce more info on the overview of thrips in Poland in the introduction.

Response 1: More info about thrips in Poland was included in the introduction. It should also be noted that the majority of sources describing the harmfulness of thrips (lines 72-82) refer to studies conducted in our country and Eastern Europe.

2. Supplement the list and pictures of thrips investigated in this article in appropriate text. Figure S1 is not incomplete. 

Response 2: In the supplementary material photo of female Thrips fuscipennis was added. Despite our efforts, we were unable to find more suitable photos to include here. The photos were taken by a PhD student during his fieldwork, who did not provide us with any more.

3. Supplement the occurrence dynamics of dominant thrips species in the results.

Response 3: Thank you for the comment. We made the dynamics of the three most abundant species: Thrips tabaci, Aeolothrips intermedius, and Thrips fuscipennis. We also analyzed its abundance according to the GLM model

4. What is the vegetation around soybean fields and how can it influence the occurrence of thrips?

Response 4: The experiment was carried out in the vicinity of other soybean plots. The conditions of conducting the research were maintained from the other experiments to limit the influence of the neighbour environment. However, it is likely that other crops, mainly cereals, located at least 50 metres away could have been the source of Thysanoptera monocotylodonous species. The importance of migrating thrips from other crops was also noted in the discussion of the manuscript.

5. According to the survey results, which species other than O. loti are likely to become the main pests of Polish soybeans in the future, and what prevention and control measures are in place?

Response 5:  We suppose that polyphagous Thrips tabaci and T. fuscipennis could also be significant on soybean and it is possible to expect some damage caused by them in the near future. It was developed in the discussion. Our trial is one of the first studies in Poland, and such studies are necessary to determine the harmfulness threshold. So far, there are no registered insecticides to control thrips on soybean.

6. The occurrence of thrips is closely related to planting density. What suggestions can the survey results of the manuscript provide for soybean cultivation and breeding scientists?

Response 6: The new data analysis proposed by another reviewer did not show the significant effects of plant density. Therefore, we refrained from discussing this aspect in detail.

7. The letters in the figures are small, please replace it.

Response 7: We did a new data analysis as was suggested in another review, better quality of figures were incorporated.

Reviewer 2 Report

Comments and Suggestions for Authors

In this study, the authors evaluated agronomic management factors of soybean cultivation, such as row spacing, seeds/sown m2, phenological stage, and variety to determine their effect on the populations of different thrips species. The information reported is relevant in terms of important pest species, the presence of some natural enemies, and crop seasons when protection against thrips is most important. 

However, it is necessary to improve multiple aspects for the work to be publishable. Several concerns regarding statistical analysis must be addressed. Here are some important comments: 

 

Line 18 (abstract): The treatments must be described before the design is discussed.

 

Line 116 : Since different fertilization formulas were used in the two locations, please explain what were the criteria to define these formulas.

 

line 119: Explain why pre-emergence weed management was different in the two locations.

Table 1: This table has errors. As it stands there are only two repeated treatments. The combinations of factors are wrong.

Or else, if the two factors are not independent, there is no way to separate the effects in the ANOVA.

Line 141: Were these references used only for identification or also to establish dietary and host habits (table 4)?

Line 157: Insect abundance is a variable that usually presents overdispersion and the variance is not independent of the mean. Did the authors validate the assumption of normality of residuals? In many cases, this type of data can be better modeled using a generalized linear model (GLM) under negative binomial distribution, because ANOVA assumptions are not met. The authors must clarify this aspect.

 line 158: Since the authors repeatedly sampled the same plots, there is to some extent a correlation effect between measurements from the same plot over time, so it is not appropriate to include time in a factorial structure. In this case, it is appropriate to perform repeated measures analysis over time

Line 159: Observing the moments that were evaluated, it would be more appropriate to refer to "phenological stages", rather than "sampling dates".

Figure 1: Do the letters represent groups of different averages? In the methodology, it should be mentioned which comparison test was used and what significance is being used.

Table 1 Location 1: Why are there nine values ??of degrees of freedom if there are only seven sources of variation?

Table 1 Location 2: Degrees of freedom for Error and Corrected Total: This value does not seem correct, taking into account the number of treatments, sampling over time, and repetitions.

Table 4: Does this information come from previous work? An additional column should be placed with the citation

Figure 3: There is no detailed analysis or interpretation of results relevant to these graphs and this analysis. The authors must include this information, or otherwise, delete this analysis and its graph.

Line 250: In this work, the seeding rate was not evaluated as such, but rather the seeding density.

Line 262: Although this work (Krobb et al.) is cited as support for the results obtained in this study, the results of the cited work are opposite, and therefore an explanation for the discrepancy found is required.

Lines 276-277: Regardless of whether different authors propose that the effect of increasing sowing density is to increase or decrease pest insect populations, the important thing is to discuss the underlying causes of the phenomenon and propose which of them could be relevant in the two environments studied. This is something that should be expanded upon in the discussion.

There are additional comments in the attached file

Comments for author File: Comments.pdf

Comments on the Quality of English Language

There are several basic spelling errors. It is recommended that the text be reviewed by an English language editing service or a native speaker.

Author Response

In this study, the authors evaluated agronomic management factors of soybean cultivation, such as row spacing, seeds/sown m2, phenological stage, and variety to determine their effect on the populations of different thrips species. The information reported is relevant in terms of important pest species, the presence of some natural enemies, and crop seasons when protection against thrips is most important. 

However, it is necessary to improve multiple aspects for the work to be publishable. Several concerns regarding statistical analysis must be addressed. Here are some important comments: 

 Line 18 (abstract): The treatments must be described before the design is discussed.

 Response 1: Thank you for this comment. We made the changes in the abstract.

Line: 92: fawn soil of light clay on medium soil,

Response 2: The soil description was given in more detail on the selected literature. The soil type at the experimental site Wrocław-Pawłowice was typical brown Luvisols developed from light loam underlain by medium loam, suitable for wheat production [IUSS 2014].

Line 116 : Since different fertilization formulas were used in the two locations, please explain what were the criteria to define these formulas.

 Response 3: Thank you for your question. The differential fertilisation of soybean in both experimental locations depended on the content of macronutrients in the soil. Before soybean sowing, soil samples were collected from two locations (Wrocław-Pawłowice, Łosiów). Then the content of P, K, Ca, Mg, and S in the soil was determined. Soil analyzes were performed at the District Chemical and Agricultural Station in Wrocław using typical methods. Based on the results of the analysis, fertiliser doses were planned in both experimental locations. In addition, a starting dose (28-30 kg/ha N) was applied prior to the symbiosis of plants with Bradyrhizobium japonicum. The above explanation was marked in the methodology.

line 119: Explain why pre-emergence weed management was different in the two locations.

Response 5: Thank you for your comment. A year before starting the experiment, we observed weeds in two locations. We identified the main weed species and apply herbicides accordingly during the soybean growing season. We tried to use herbicides that best control the designated weed species. Therefore, pre-emergence weed management was different in the two locations. Also this explanation was shortly included in the methodology.

Table 1: This table has errors. As it stands there are only two repeated treatments. The combinations of factors are wrong.

Or else, if the two factors are not independent, there is no way to separate the effects in the ANOVA.

Response 5: Thanks for the comments according to the data analysis. We agree that the analysis of variance was not the best way here to analyse our data thus new analysis as was later suggested was conducted. 

​Line 141: Were these references used only for identification or also to establish dietary and host habits (table 4)?

Response 6: In this place, these references were used for thrips identification only. We have supplemented new references to the food and habitat preferences of the thrips in Table 6 (4 in previous version).

Line 157: Insect abundance is a variable that usually presents overdispersion and the variance is not independent of the mean. Did the authors validate the assumption of normality of residuals? In many cases, this type of data can be better modeled using a generalized linear model (GLM) under negative binomial distribution, because ANOVA assumptions are not met. The authors must clarify this aspect.

Response 7: Thank you for the comments regarding the data analysis. We agree that variance analysis is not the best way to analyse our data. In the previous analysis, we standardised the data to obtain the normal distribution. We are grateful for the suggestion of an appropriate analysis - Generalized Linear Model (GLM) with a negative binomial distribution.

The data (total number of thrips and the numbers of the most abundant species) from Location 1 and Location 2 were analyzed separately. The analyses were performed using a fit-ted Generalized Linear Model (GLM) with a negative binomial distribution in Python. The response variable was the thrips count, while the predictor variables were year, phenological stage, row spacing, number of seeds (only for Location 1), and soybean variety (only for Location 2). Therefore, the model for Location 1 was: Thrips number ~ Row spacing + Seed density + Year + Phenological stage, and for Location 2: Thrips number ~ Row spacing + Variety + Year + Date. The Log-Likelihood and Pseudo R-squared provided information on the goodness of fit of the model.  

 line 158: Since the authors repeatedly sampled the same plots, there is to some extent a correlation effect between measurements from the same plot over time, so it is not appropriate to include time in a factorial structure. In this case, it is appropriate to perform repeated measures analysis over time

Response 8: Agree, we applied the GLM model for data analysis.

Line 159: Observing the moments that were evaluated, it would be more appropriate to refer to "phenological stages", rather than "sampling dates".

Response 9: Thank you, fixed in the text

Figure 1: Do the letters represent groups of different averages? In the methodology, it should be mentioned which comparison test was used and what significance is being used.

Response 10: We have new results of the analysis. Comparative analyses were performed with a Tukey post-hoc test.

Table 1 Location 1: Why are there nine values ??of degrees of freedom if there are only seven sources of variation?

Response 11: We applied the GLM model for data analysis. The degrees of freedom are given in the tables

Table 1 Location 2: Degrees of freedom for Error and Corrected Total: This value does not seem to be correct when taking into account the number of treatments, sampling over time, and repetitions.

sources of variation?

Response 12: We applied GLM model for data analysis. 

Table 4: Does this information come from previous work? An additional column should be placed with the citation

Response 13: We have supplemented new references to the food and habitat preferences of the thrips in Table 6 (4 in previous version).

Figure 3: There is no detailed analysis or interpretation of results relevant to these graphs and this analysis. The authors must include this information, or otherwise, delete this analysis and its graph.

Response 14: We deleted those analyses. They did not bring much new information and constituted too much of the results

Line 250: In this work, the seeding rate was not evaluated as such, but rather the seeding density.

Response 15: We agree, corrected.

Line 262: Although this work (Krobb et al.) is cited as support for the results obtained in this study, the results of the cited work are opposite, and therefore an explanation for the discrepancy found is required.

Response 16: We agree with this comment and we are sorry for this error. We improved this text in the discussion.

Lines 276-277: Regardless of whether different authors propose that the effect of increasing the density is to increase or decrease the populations of pest insects, the important thing is to discuss the underlying causes of the phenomenon and propose which of them could be relevant in the two environments studied. This is something that should be expanded upon in the discussion.

Response 17: As we mentioned previously, the data analysis (GLM) did not show any significant differences in the sowing density effect. Therefore, we had to refrain from discussing this result in more detail. We focussed on developmental stages that had a significant impact on the number of thrips.

 

There are additional comments in the attached file

peer-review-38784985.v3.pdf

Response 18: Thank you for the specific comments. We fixed all the suggestions in the text.

Comments on the Quality of English Language

There are several basic spelling errors. It is recommended that the text be reviewed by an English language editing service or a native speaker.

Response 19: The English in the manuscript was carefully proofread after all corrections had been made.

Reviewer 3 Report

Comments and Suggestions for Authors

The authors targeted to survey thrips assemblegas (density, species) in two locations at different planting densities in different years.

The authors need to clarify, justify:

- L55-56, 65: "Thrips are one of the most important herbivores in soybean" ? in the USA, Brasil, Europe it is not true. Thips usually do not need control interventions.

- L128-135: Row spacing (10-15 cm) in extremely unusual, if at all exists, cropping density (dense row spacing is cc 24 cm, usual one is 48-48 cm and wide one is 75 cm) in practical farming. Please explain the relevance of the 1015 cm row spacing results for the practice.

Table 1 Location one row spacing (14 cm results in 50 seeds/m2, while 30 cm in 90). I assume it is not correct. Location 2 (row spacing 12 and 45 cm but no data on seed density but varieties are listed. Not consistent data demonstration.

L 200 species name should be in italics.

Table 4 column "Host" needs clarification (Plant ? pollen, monocotyledonous plants?). How does this comes to soybean? Please explain other plants/weeds on the soybean plots and their impact on thrips presence?

L 213-214: Please provide support for this statement.

L 224-225: harvest period of what?

L 268-271: Needs strong clarification: WCR adult density in X year, subsequent larval damage in next year (in maize after maize), what row spacing, when?

L 283-285: Consequences, implications thereof?

L 301: Pollen grains of what crop?

L 315-321: Is likely a bit too general statement.

Author Response

The authors targeted to survey thrips assemblegas (density, species) in two locations at different planting densities in different years.

The authors need to clarify, justify:

- L55-56, 65: "Thrips are one of the most important herbivores in soybean" ? in the USA, Brasil, Europe it is not true. Thips usually do not need control interventions.

 

Response 1: Yes, I agree. The indicated sentence in the introduction has been slightly changed. But I also think that we can point to their growing importance, not only because of their direct harmfulness, but also because of the transmission of viruses.The significant increase in the area of soybean cultivation in Europe and Poland will certainly favour herbivores.This topic was discussed in more detail in the discussion.

 

- L128-135: Row spacing (10-15 cm) in extremely unusual, if at all exists, cropping density (dense row spacing is cc 24 cm, usual one is 48-48 cm and wide one is 75 cm) in practical farming. Please explain the relevance of the 1015 cm row spacing results for the practice.

Response 2: I fully understand why you ask about row spacing. In Poland, farmers have been growing soybeans for about 15 years. Soybean is a new species whose agrotechnics are being tested under various conditions. That is why Polish farmers are most often using grain drills to sow soybeans. In these seeders, the row spacing is usually set at 15 or 30 cm for soybeans. Therefore, we used these two row spacings in our investigation. You are right, in other countries farmers use row spacing 48-75 cm. However, they use precision seeders dedicated to sowing soybeans or beets. In Poland, there are tests for sowing soybeans and other legumes using precision seeders. I think soybean will be grown in Poland in a few years with a wider spacing (e.g., 48 cm), but this depends on the availability of new precision seeders. According to the breeders' recommendations, we sown 50-90 seeds per 1 m² in Poland.

 

Table 1 Location one row spacing (14 cm results in 50 seeds/m2, while 30 cm in 90). I assume it is not correct. Location 2 (row spacing 12 and 45 cm but no data on seed density but varieties are listed. Not consistent data demonstration.

Response 3: Thank you for this comment, we corrected the data in the table.

 

L 200 species name should be in italics.

Response 4: It was corrected

 

Table 4 column "Host" needs clarification (Plant ? pollen, monocotyledonous plants?). How does this comes to soybean? Please explain other plants/weeds on the soybean plots and their impact on thrips presence?

 

Response 5: It was clarified in the Table 6 (4 in a previous version). These data are intended to indicate which species should be considered as a direct threat to soybean. Many species of thrips migrate considerable distances, actively or passively, and are found on soybean even if they do not cause damage. This issue was developed in the discussion. Recognition of species that may feed on Fabaceae indicates a potential threat to soybean.

 

 

L 213-214: Please provide support for this statement.

Response 6: We added the references to the thrips feeding preferences to table 6.

 

L 224-225: harvest period of what?

Response 7: We deleted those sentences from the results part

 

L 268-271: Needs strong clarification: WCR adult density in X year, subsequent larval damage in next year (in maize after maize), what row spacing, when?

 

Response 8: We corrected the text in the discussion

 

L 283-285: Consequences, implications thereof?

Response 9: We added some explanation to the text in the discussion

 

L 301: Pollen grains of what crop?

Response 10: “of leguminous plants”  it was added

 

L 315-321: Is likely a bit too general statement.

Response 11: Yes, it was really general. This part was modified and I think it is more specific now.

 

 

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

The authors made many important modifications to improve the quality of the work. There is still something that needs to be checked carefully. In Table 1, the treatment approach is still unclear. If there are two factors and they are independent (Row spacing and seed density), the four treatments in Table 1 would be the combinations of the different levels of the factors (1-Row spacing 15 x 90 seeds,2-Row spacing 15 x 50 seeds, 3-Row spacing 30 x 90 seeds, 4-Row spacing 30 x 50 seeds). However, in Table 1, although 4 treatments are mentioned, 1 and 2 are the same treatment (15x90), as are 3 and 4 (30x50). On the other hand, if these factors are not independent (Row distance=15 necessarily implies seed density=90), there is confusion of effects in the analysis, whether ANOVA or GLM and in this case, it does not make sense to include them as sources of variation separated. The authors must clarify this.

Legend of figures 1 and 2:Mention that significant differences are "according to post hoc Tukey test".

 

Author Response

Comments and Suggestions for Authors

The authors made many important modifications to improve the quality of the work. There is still something that needs to be checked carefully. In Table 1, the treatment approach is still unclear. If there are two factors and they are independent (Row spacing and seed density), the four treatments in Table 1 would be the combinations of the different levels of the factors (1-Row spacing 15 x 90 seeds,2-Row spacing 15 x 50 seeds, 3-Row spacing 30 x 90 seeds, 4-Row spacing 30 x 50 seeds). However, in Table 1, although 4 treatments are mentioned, 1 and 2 are the same treatment (15x90), as are 3 and 4 (30x50). On the other hand, if these factors are not independent (Row distance=15 necessarily implies seed density=90), there is confusion of effects in the analysis, whether ANOVA or GLM and in this case, it does not make sense to include them as sources of variation separated. The authors must clarify this.

Response 1: We would like to thank you for bringing this to our attention. This is our obvious error in the original description of the experimental treatments at Location 1. We confirm that there are two factors: Row Spacing and Seed Density (Number of seeds sown per 1 m²) and four distinct combinations, as listed in Table 1:

• Row Spacing: 15 cm, Seed Density: 50 seeds/m²
• Row Spacing: 15 cm, Seed Density: 90 seeds/m²
• Row Spacing: 30 cm, Seed Density: 50 seeds/m²
• Row Spacing: 30 cm, Seed Density: 90 seeds/m²

It was corrected in the manuscript.

Below is the detailed experimental scheme for the field experiment at Location 1, which is not included in the manuscript. This scheme further clarifies that the two factors: Row Spacing and Seed Density, were varied independently, allowing for the analysis of both their individual and interactive effects on the response variables.

Detailed experiment outline in Location 1 not included in the manuscript

Legend of figures 1 and 2 Mention that significant differences are "according to post hoc Tukey test".

Response 2: Thank you for this comment, we added this information to the figures’ legend

 

Reviewer 3 Report

Comments and Suggestions for Authors

Thanks for addressing the comments, considering and adjusting your MS accordingly.

 

Author Response

Thank you for your constructive comments that have improved our manuscript.