Characterizing Early-Season Nitrogen Fertilization Rate Effects on Growth, Yield, and Quality of Strawberry

Round 1

Reviewer 1 Report

This is well written and clear and concise.

line 87, change bare-root seedlings to bare root plants

line 101, change seedling, to bare root plants

My primary concern with this paper is the reliance on only 1 seasons data for conclusions on measures of plant size and fruit quality. I think a greater clarification of this being data that is from a single season would help to not over state the confidence of these results.

Some statements could be put in context that this study was conducted on winter produced strawberry in sandy soil, and that the daylength, crop timing (fruiting immediately after planting) and timing of repeated N application on a daily basis likely impacted observations of N  rate impacts on canopy growth over roots. For example line 311-313.

Author Response

I truly appreciate all the constructive comments and suggestions! And thank you so much for the opportunity to revise the paper!

Below I have listed my individual responses to your comments, describing how each reviewer concern and comment was addressed in the revision.

 

1. This is well written and clear and concise.

Thank you so much for your positive feedback!

 

2. line 87, change bare-root seedlings to bare root plants

I changed “bare-root seedlings” to “bare-root transplants”. I used “bare-root” with a dash because it was more commonly used in literature, and used “transplants” instead of “plants” to be more specific.

 

3. line 101, change seedling, to bare root plants

I changed “bare-root seedlings” to “bare-root transplants”. I used “bare-root” with a dash because it was more commonly used in literature, and used “transplants” instead of “plants” to be more specific.

 

4. My primary concern with this paper is the reliance on only 1 seasons data for conclusions on measures of plant size and fruit quality. I think a greater clarification of this being data that is from a single season would help to not over state the confidence of these results.

Yes, I agree that having two season data would make the conclusion more concrete! I added a new sentence in L400-403 to explain that the conclusion associated with growth data was made based on one season data. I also added a new sentence in 413-414 to emphasize the importance of future research on fruit quality attributes based on multiple-season data.

 

5. Some statements could be put in context that this study was conducted on winter produced strawberry in sandy soil, and that the daylength, crop timing (fruiting immediately after planting) and timing of repeated N application on a daily basis likely impacted observations of N rate impacts on canopy growth over roots. For example line 311-313.

Thank you so much for the suggestion! I revised the sentence in L319 and added a new sentence in L320-321 following your suggestion.

Author Response File: Author Response.docx

Reviewer 2 Report

The study is useful as it provided a strategy by increasing the early-season N fertilization rate to improve marketable yield and quality. However, some problems existed and need further improvement before considering accepting.

Some advice given for improving were listed.

 1) Line 104, why sprinkle irrigation instead of drip irrigation installed was used after transplanting?

2) Table 1 if  total N input rate of treatment 1 is 34.2kg.ha^-1, Treatment 2，treatment 3 and treatment 4 should be 51.3kg,68.4kg and 85.5kg, respectively proportionally. not 51.2kg,68.3kg and 85.4kg.

3) variance analysis of yield and quality index should be conducted to determine the difference significance between treatments, which is necessary to find optimum treatment.

4) Average yield, leaf area, and other parameters should be used for model analysis instead of values of each replication, otherwise, the correlation coefficient would be different.

The method just using a simple mathematical model for data fitting without in-depth analysis, made the paper scientific and technically weak.

5)   Line 221 the total season yield is marketable yield or the sum of marketable and unmarketable yields? the parameters should be specific.

6)   insect-damaged fruit yield of each treatment and cultivar should be listed and the effect of N rate on it should be described and(or) explained.

7)from line 301 to 305, effects of temperature were discussed, but not relative to the study.

8) conclusion was too simple and needs further description. For example， what are the optimal fertilization strategies for different cultivar? A possible explanation of the results?   

Author Response

Thank you so much for your constructive comments and suggestions! Your time and efforts are greatly appreciated.

Below I have listed my individual responses to your comments, describing how each reviewer concern and comment was addressed in the revision.

 

1. Line 104, why sprinkle irrigation instead of drip irrigation installed was used after transplanting?

Thank you so much for pointing that out! I added new sentences in L101-103 and L106-107 to explain why sprinkler irrigation was used during the establishment period. A new reference [16] was also added.

 

2. Table 1: If total N input rate of treatment 1 is 32.4 kg ha-1, treatment 2, treatment 3, and treatment 4 should be 51.3 kg, 68.4 kg, and 85.5 kg, respectively proportionally. Not 51.2 kg, 68.3 kg, and 85.4 kg.

Thank you so much for checking the information in the manuscript in detail! The application rate used for Treatment 1 is 34.2 kg/ha, not 32.4 kg/ha. I double-checked that the calculation was correct for all treatments, so I did not make changes in the table.  

 

3. Variance analysis of yield and quality index should be conducted to determine the difference significance between treatments, which is necessary to find optimum treatment.

Yes, I agree that ANOVA + mean comparison is one of data analysis options. However, because the main objective of this study was to characterize the dose-response instead of comparing means, model fitting analysis was used in this study. I added the discussion on the x and y values for the peak in a quadratic model to discuss the optimum rate for ‘Florida127’. The discussion on the optimum rate for ‘Florida radiance’ can be found in L358-360.

 

4. Average yield, leaf area, and other parameters should be used for model analysis instead of values of each replication, otherwise, the correlation coefficient would be different

To determine how much variability in Y (yield) is explained by X (growth variables), I believe individual data must be used. The use of average data would significantly increase r2, but r2 would no longer reflect the variability of the data, which we are interested in analyzing.

 

5. The method just using a simple mathematical model for data fitting without in-depth analysis, made the paper scientific and technically weak.

The models used in this study are commonly used to describe dose-responses in growth and biological activities. For example, sigmoidal model is used to identify the half maximal effective concentration for biological activities in various chemical tests. They are simple, yet their model coefficients provide practical interpretations. Because there is only one predictor variable (which is early season N rate), I believe the use of these simple models is a more appropriate approach than using complex models, which makes model interpretation very difficult.

 

6. Line 221 the total season yield is marketable yield or the sum of marketable and unmarketable yields? the parameters should be specific

Thank you for the comment! I changed “total season yield” to “total season marketable yield” to avoid the confusion.

 

7. Insect-damaged fruit yield of each treatment and cultivar should be listed and the effect of N rate on it should be described and(or) explained.

The potential cause of increased N rate to increase insect damage fruit is discussed in L374-378.

 

8. From line 301 to 305, effects of temperature were discussed, but not relative to the study.

Temperature effects were discussed here because they were relevant to the study by Deschamp et al, which demonstrated that cooling properties of reflective mulch promoted early season canopy growth and increased early season yield. This reference was used to emphasize the importance of early season growth for strawberry production in Florida. In this study, we found that increasing early season N rate can also promote early canopy growth and increase yield. A new sentence was added in L315-316 to discuss potential benefits of increasing leaf area to reduce heat stress during establishment.

 

9. Conclusion was too simple and needs further description. For example, what are the optimal fertilization strategies for different cultivars? A possible explanation of the result?

Thank you very much for the suggestion! I revised the conclusion in L406-414 to elaborate the discussion on the optimal fertilization strategy for each cultivar.

Author Response File: Author Response.docx

Reviewer 3 Report

Dear Authors, I have finished the review of the ms "Characterizing Early Season Nitrogen Fertilization Rate Effects 2 on Growth, Yield, and Quality of Strawberry" and found it of interest for Agronomy readers. I have a few suggestions and doubts I would like you to answer. I copy them below. In addition, I found some typos I edited in the file I attached.

M&M

Can you converse soil organic matter content to percentages? Readers (as I) can be more familiar with this measurement.

In crop management, were pollinators included? Misshapen fruit is due to poor irregular pollination? If so, how do you expect an influence of early N dose?

Can you confirm (and included in 2.4. Fertilization Treatments) that pre-planting N was not applied?

Please, check the Table columns format. I am not sure if the text should be centered.

I understand that if a split-plot design was followed, cultivars should be the main factors (Factor A with two levels (varieties), while fertilization treatments (included within the main factor) should be Factor B (depending on the year 4 or 5 levels (fertilization applied). I believe that if the specific value of the interaction is not sought, different cultivars can be treated as independent experiments and the differential response discussed. In fact, the results are presented considering the effects of the early N-dose for each cultivar independently.

I am not sure either about the correlation analyses performed. Two parameters are (co)related if the p-value of the correlation analysis is below 0.05 (being the strength of the relation determined by the coefficient). If the correlation is significant, then we may perform different analyses to determine the kind of relation (linear, quadratic, logistic…).

Results

L211 (and 273 too). A decline is expected to doses higher than those applied in the quadratic model (as seems to occur for 1.5 and 2 kg ha-1 d-1). Please, comment this circumstance of special interest.

Discussion

I would like you to develop a little more the hypothesis behind higher requirements for some varieties compared to previous cited works. Higher vigor, larger canopy? Can you address this point in your Discussion and give reader a hint, so future works might explore further this variation.

 

Comments for author File: Comments.pdf

Author Response

Thank you so much for reviewing the manuscript and for proving constructive comments and suggestions!

Below I have listed my individual responses to your comments, describing how each reviewer concern and comment was addressed in the revision.

 

1. Can you converse soil organic matter content to percentages? Readers (as I) can be more familiar with this measurement?

Thanks for point that out! The unit used for organic matter in the manuscript was actually incorrect. It was supposed to be g/kg. The unit was changed to % as suggested (L84).

 

2. In crop management, were pollinators included? Misshapen fruit is due to poor irregular pollination? If so, how do you expect an influence of early N dose.

No, pollinators were not used in this study. Nutrient deficiency, particularly Ca and B, is another cause of misshapen fruit. Increased fruit production could have increased nutrient demand and induced Ca or B deficiency. I added a new paragraph to discuss this in L371-380. Thank you so much for this comment!

 

3. Can you confirm (and included in 2.4. Fertilization Treatments) that pre-planting N was not applied.

Sure, that is a very good point! I made changes in L111-112 as suggested.

 

4. Please, check the Table columns format. I am not sure if the text should be centered

If the manuscript was accepted, I will check with the technical editor regarding the table format. Thank you so much for pointing that out!

 

5. I understand that if a split-plot design was followed, cultivars should be the main factors (Factor A with two levels (varieties), while fertilization treatments (included within the main factor) should be Factor B (depending on the year 4 or 5 levels (fertilization applied). I believe that if the specific value of the interaction is not sought, different cultivars can be treated as independent experiments and the differential response discussed. In fact, the results are presented considering the effects of the early N-dose for each cultivar independently.

I believe any factors can be assigned as either main or subplot factor. In this study, N rate was assigned as the main plot factor and cultivar was assigned as the subplot factor. Cultivar x N rate interaction can be assessed by 2-way ANOVA or by model fitting analysis used in this study. The former option is used in many studies, but it does help characterize how dose-responses differ. In this study, mode fitting analysis was used to demonstrate cultivar-dependent responses based on different dose-responses between the two cultivars. Model fitting also allows practical interpretations of models based on model coefficients.

 

6. I am not sure either about the correlation analyses performed. Two parameters are (co)related if the p-value of the correlation analysis is below 0.05 (being the strength of the relation determined by the coefficient). If the correlation is significant, then we may perform different analyses to determine the kind of relation (linear, quadratic, logistic…).

In linear correlation analysis, P values below 0.05 simply indicate that the slope is significantly different from zero. In Fig 3, linear correlation was found to be useful to discuss the impact of different growth variables on the variability in yield. In model fitting, correlation does not need to be tested before fitting other models. For example, if the data show exponential or logistic function, the linear correlation can be not significant. In this study, the best-fit model was selected based on AICC value, and the model coefficients were significant at P < 0.05.

 

7. L211 (and 273 too). A decline is expected to doses higher than those applied in the quadratic model (as seems to occur for 1.5 and 2 kg ha-1 d-1). Please, comment this circumstance of special interest.

I agree that this is an important discussion topic! New sentences were added in L? to elaborate the discussion.

 

8. I would like you to develop a little more the hypothesis behind higher requirements for some varieties compared to previous cited works. Higher vigor, larger canopy? Can you address this point in your Discussion and give reader a hint, so future works might explore further this variation.

Of course, and thank you so much for the suggestion! New sentences were added in L? to elaborate the discussion on the suggested topic.

Author Response File: Author Response.docx

Round 2

Reviewer 2 Report

I think my questiones were well answered and the papaer was also improved due to the  suggetions.