A Deep Learning Model to Predict Evapotranspiration and Relative Humidity for Moisture Control in Tomato Greenhouses
Round 1
Reviewer 1 Report
Some comments and suggestions provided as follows:
L79 Zou, comma is not necessary.
L107 An internal sensor module to… (what brand?)
L108 ..(data required).. (?)
L124 The title is not compete and check the range of VWC ?
L127 add [8].
L130 add [11].
L131 Which growing stage did you start to measure the LAI? Sample size? How to get the LAI measurement?
L124 The title is incorrect and check the range of VWC ?
L136 Table 2. Define eo and U. Where is I=15 cm (5-20 cm) ?
L139 ..(data required).. (?)
L156 should be rectified linear unit (ReLu)
L172-173 Why did you include the weight of the medium? How to calculation the dew point temperature? The range? What did the root-zone sensing data include?
L179 delete (Jung et al., 2020) and add [4]
L211 should be Equation 4 and 5.
L217 20-day dataset. (Which growing stage?)
L227-228 Cannot find any explanation about Figure 4.1.
L235-236 Cannot find these results from Figure 5.
L258 Which growing stage for the test set?
L307 For the data-based modeling, you should use more data to test the training result; otherwise the performances of these results may be qestioned.
Author Response
Dear Reviewer
Thank you for your suggestions regarding our manuscript and the opportunity to address the comments made by the referees. We are grateful for the comments provided by referees and editors. We have addressed all the comments raised by reviewers and made appropriate revisions. In what follows, we detail our responses to the comments.
Comments and Suggestions for Authors
Some comments and suggestions provided as follows:
L79 Zou, comma is not necessary.
:Thanks for the reviewer comments. The citation method of this paper has been modified to fit the journal format.
L107 An internal sensor module to… (what brand?)
:Thanks for the reviewer comments. The sensor module information is entered in line number 117.
L108 ..(data required).. (?)
: The sensor module information is entered in line number 117.
L124 The title is not complete and check the range of VWC ?
: The name of Volumetric water content has been accurately described. Thanks for the reviewer's point.
L127 add [8]. L130 add [11].
:Thanks for the reviewer comments. The citation method of this paper has been modified to fit the journal format. Thanks again.
L131 Which growing stage did you start to measure the LAI? Sample size? How to get the LAI measurement?
:Thank you for the reviewer comments. LAI indicators have been described in detail. Line number 145~151. Individual leaf length and maximum width were manually measured weekly on 10 randomly selected plants. LAI was determined by multiplying the maximum width and leaf length and a reduction coefficient of 0.64. LAI started measuring after the first flowering flower cluster was created, and the data set used in this study was carried out at the harvest time of the third flower cluster, about 12 to 15 weeks after planting.
L124 The title is incorrect and check the range of VWC ?
Thanks for the reviewer pointing out that number was a typo and corrected to 5.53–38.5%.
L136 Table 2. Define eo and U. Where is I=15 cm (5-20 cm) ?
: Thanks to the reviewer for pointing out, e0 is, which is Saturation vapor pressure at mean air temperature, u is the friction velocity (m·s−1). And the incorrect entry in the table has been corrected. The I=15 cm (5-20 cm) part was also deleted due to a typo.
L139 ..(data required).. (?)
: We have added citations of references in the relevant section.
L156 should be rectified linear unit (ReLu)
: Thanks for the reviewer's suggestion. We have corrected the term to rectified linear unit (ReLu).
L172-173 Why did you include the weight of the medium? How to calculation the dew point temperature? The range? What did the root-zone sensing data include?
: Thank you for the reviewer's comments. The text has been modified as follows: Soil volumetric water content (VWC) and soil temperature were measured with a root zone sensor. In addition, the dew point values were determined through the humidity sensor near the crop and the leaf temperature sensor. During the experiment, the dew points ranged from 8.4 to 20.1 °C
L179 delete (Jung et al., 2020) and add [4]
: Thanks for the reviewer comments. The citation method of this paper has been modified to fit the journal format.
L211 should be Equation 4 and 5.
: Thank you for the reviewer comments. Modified with formula numbers [4] and [5].
L217 20-day dataset. (Which growing stage?)
: Thank you for the reviewer comments. Added a reference to the duration of the experiment. Line number 133-134. A description of the growth stages has been added to lines 148-149.
L227-228 Cannot find any explanation for Figure 4.1.
: Thanks to the reviewer for pointing out, I added an explanation for Figure 4 on line 263.
L235-236 Cannot find these results from Figure 5.
:Since the SEP is not shown in Figure 5, as the reviewer pointed out, the text is left alone.
L258 Which growing stage for the test set?
: Thank you for the reviewer comments. The testing period of the model is also the same as the experimental period, and the description of this growth stage has been added to lines numbers 148-149.
L307 For the data-based modeling, you should use more data to test the training result; otherwise the performances of these results may be qestioned.
: We agree with the reviewer's opinion. There are several practical problems with this approach of data-driven modeling of greenhouse environments to facilitate sharing without tuning the trained model. This will require more learning and training results. However, it was mentioned in the last paragraph that additional studies are needed under various greenhouse conditions in the future with the same approach as in this study.
Author Response File: 
Author Response.docx
Reviewer 2 Report
Dear Authors,
I revised the manuscript "A deep learning model to predict evapotranspiration and relative humidity for moisture control in tomato greenhouses" submitted to the Agronomy journal. The manuscript is interesting. However, I have some concerns which need to be addressed before considering for final publication.
General comments:
1. Moderate editing of English language and style required.
2. References should be numbered in order of appearance and indicated by a numeral or numerals in square brackets - e.g., [1] or [2,3], or [4–6]. Do not use the publication year of the paper.
Specific comments:
1. Section "1. Introduction". In this section, more examples related to the application of deep learning methods in greenhouses should be cited.
2. Subsection "2.1. Greenhouse and sensor description". In this subsection, add information about the date of the study conducted. This is important information that also allows you to verify the data in Table 1.
3. Line 208-2013. I suggest adding another error measure, MAPE, which is often used in this type of analysis. You can use it instead of %SEP. You can find it in the paper https://doi.org/10.3390/agronomy11050885
4. Section "4. Discussion". This section is worth adding a few new references since the number of them in the entire manuscript is only 20.
Author Response
Dear Authors,
I revised the manuscript "A deep learning model to predict evapotranspiration and relative humidity for moisture control in tomato greenhouses" submitted to the Agronomy journal. The manuscript is interesting. However, I have some concerns which need to be addressed before considering for final publication.
:Thank you for your letter regarding our manuscript and the opportunity to address the comments made by the referees. We are grateful for the comments provided by referees and editors. We have addressed all the comments raised by reviewers and made appropriate revisions. In what follows, we detail our responses to the comments.
General comments:
Moderate editing of English language and style required.
:Thanks for the reviewer comments. We reviewed this manuscript in its entirety again, examining English language expressions and editing styles.
References should be numbered in order of appearance and indicated by a numeral or numerals in square brackets - e.g., [1] or [2,3], or [4–6]. Do not use the publication year of the paper.
: Thanks for the reviewer comments. The overall citation method has been modified to fit the citation format of this journal.
Specific comments:
Section "1. Introduction". In this section, more examples related to the application of deep learning methods in greenhouses should be cited.
: Reflecting the opinion that there are few deep learning researchers in the reviewer's introduction, we have additionally cited 4 related documents in line numbers 83-91.
Subsection "2.1. Greenhouse and sensor description". In this subsection, add information about the date of the study conducted. This is important information that also allows you to verify the data in Table 1.
:Thanks for the reviewer comments. Added a reference to the duration of the experiment. Line number 133-134. Description of growth stages added to lines 148-149
Line 208-2013. I suggest adding another error measure, MAPE, which is often used in this type of analysis. You can use it instead of %SEP. You can find it in the paper https://doi.org/10.3390/agronomy11050885
:Thanks for the reviewer comments. MAPE is a similar index to SEP used in this study, and was used to compare values with previous studies [4]. We will apply MAPE in future studies. thanks for the comments.
Section "4. Discussion". This section is worth adding a few new references since the number of them in the entire manuscript is only 20.
:Thank you for the reviewer's comments. We cited 14 additional research literature in the introduction and discussion sections.
Author Response File: Author Response.docx
