Yield and Quality of Romaine Lettuce at Different Daily Light Integral in an Indoor Controlled Environment
, Ewa Ropelewska 2,*, Anna Wrzodak 2, Artur Kowalski 3 and Stanisław Kaniszewski 3Round 1
Reviewer 1 Report
Please refer to the attached document for more information on how to improve the manuscript.
Comments for author File: 
Comments.pdf
Author Response
The authors tested two lettuce cultivars exposed to different light conditions and examined how the growth and overall quality of the lettuce is affected. As the market requires more food in an all-season fashion, enhancing the production of the main aliments, is of great interest. Indoors growth conditions are affecting the quality of the plants and adjusting the parameters so that the biomass is increased, is a challenge took by the authors. The manuscript is well written and the results are comprehensive. There are several indications presented bellow, to help improve the message that the authors want to send to the readers and scientific community. Once these minor setbacks are addressed, the manuscript can be accepted for publication.
Answer: Thank you very much for your careful reading and reviewing of the manuscript. All the suggestions have been included in the revised manuscript to further improve it.
Title – the title is too long and hard to follow. We recommend you reduce it to comprise the essential.
Answer: The title has been changed into: Yield and Quality of Romaine Lettuce at Different Daily Light Integral in an Indoor Controlled Environment
Abstract.
Please explain the abbreviations upon first appearance in the abstract text (i.e. PPFD)
Answer: The abbreviations have been explained.
Introduction
Emphasize more on the novelty. It is known that light influences the growth and overall quality of plants, as they rely on photosynthesis, so why did you do all these experiments? What is the overall purpose?
Answer:
The novelty has been indicated as follows:
“The objective of this study was to determine the optimal daily light integral (DLI) for romaine lettuce grown in an indoor vertical cultivation system by investigating biomass, morphology and quality features important for consumers, such as the concentration of nitrates in leaves and sensory value. A novelty in our experiment was evaluating the differentiation of the samples, a discrimination analysis using texture features of leaf images and machine learning algorithms.”
“A comprehensive approach to evaluate the effect of different Daily Light Integral in an Indoor Controlled Environment on morphology, biomass production, sensory quality and image textures of romaine lettuce is original and was not reported in the available literature. The use of different machine learning algorithms for development of innovative discriminative models based on image textures of lettuce leaves subject-ed to different photoperiod and PPFD can also be considered as a great novelty of this study.”
Line 69 – remove dash between 14 and h, same at line 259, 261 etc.
Answer:
It has been removed in the manuscript.
Mat & Met
You can choose to abbreviate cultivar varieties with ‘c.v.’, it is an international accepted abbreviation, such as ‘var.’ (e.g. Lactuca sativa var. longifolium c.v. ‘Casual’). It might be easier to follow this way.
Answer: It has been corrected according to the Reviewer’s comment.
Line 122 – write six instead of 6
Answer: It has been corrected.
Discussions
As indicated for the introduction section, emphasize on the importance of your experiments and the novelty brought. As it is, the discussions section lacks this information.
Answer: It has been corrected as follows:
“A comprehensive approach to evaluate the effect of different Daily Light Integral in an Indoor Controlled Environment on morphology, biomass production, sensory quality and image textures of romaine lettuce is original and was not reported in the available literature. The use of different machine learning algorithms for development of innovative discriminative models based on image textures of lettuce leaves subject-ed to different photoperiod and PPFD can also be considered as a great novelty of this study.”
Additionally, the Discussion section has been supplemented with other information (please see the manuscript).
Conclusions
Too long and too many technical data. It is a rephrase of the abstract. Reduce this section to the main and overall findings, why you did the experiments and who will benefit from these results.
Answer:
The conclusions section has been partially rewritten. Some details have been removed. The importance of the research has been emphasized. Future research is indicated. Please see the manuscript.
Reviewer 2 Report
The MS by Matysiak et al explores the impact of different light regimes on growth and other key characteristics of two lettuce cultivars. The approach is interesting and such studies are needed for the horticultural sector where research is still limited. However, the MS as it stands is quite descriptive and it would be good to see more context and importance assigned to the finding and approaches. For example, the abstract states “The obtained results can be used in practice to improve romaine lettuce production in an Indoor Controlled Environment” – this needs more context. Also the authors use imaging features and machine learning algorithms, however it is unclear what was achieved with this and what the value was. The MS is also on the long side and there are several sections that are not clear and need further clarification.
Line 62 – introduce sensory quality to the readers not familiar with these.
Lone 65-67 – repetitive text.
Introduce the problem of nitrate to the readers
Line 72 – a lot of facts in this section but the flow of information is limited and not clear.
Line 93 – AI – needs more explanation and clarification.
Design and replication not clear
Section on QDA is not clear, needs more information to guide the readers.
Image processes section is not clear – more information and clarification is required for the reader – “…place in a black box..”? What were the textual features?
Line 203 – same as M&M – no need to repeat
Section 2.7 should come earlier in the M&M?
Line 225 – information about PCA etc is not needed?
Line 284 – Quantum yield at which light levels? The QY would be expected to decline at a higher light level as it is a function of light.
The authors calculate ETR from the QY – however, it is not clear what values were used for partitioning between the two photosystems and it appears that light absorption as not measured – and this would be influenced by day length and light intensity and therefore these were not taken into account. ETR should be removed and the authors should use the PSII efficiency to determine differences.
How the sensory information was evaluated is not clear and needs clarification.
Line 340 – some of this information will not be familiar for readers not working on machine learning approaches and needs more information.
In general I found this section difficult to follow although this is not my area. However, clarification would broaden the readership.
Line 395 – remove very – not scientific.
Line 407 – the authors did not measure thermal dissipation or protection – and therefore this is not supported by the results. This should be removed or rephrased to not be misleading.
Line 414 “ However, high ETR was associated with a lower quantum yield of PSII, so the challenge for achieving increased efficiency of conversion of electrical energy into electron transport will be to find ways to minimize thermal dissipation of energy as well and photoinhibition” This does not make sense?
The values of the image processing work is not clear in the discussion.
In general the discussion is very repetitious of the results in places and needs to be tightened so that the overall take home messages and value are you clear. The same applies to the conclusion.
Author Response
The MS by Matysiak et al explores the impact of different light regimes on growth and other key characteristics of two lettuce cultivars. The approach is interesting and such studies are needed for the horticultural sector where research is still limited. However, the MS as it stands is quite descriptive and it would be good to see more context and importance assigned to the finding and approaches. For example, the abstract states “The obtained results can be used in practice to improve romaine lettuce production in an Indoor Controlled Environment” – this needs more context. Also the authors use imaging features and machine learning algorithms, however it is unclear what was achieved with this and what the value was. The MS is also on the long side and there are several sections that are not clear and need further clarification.
Answer: Thank you for these valuable comments. All the suggestions have been included in the revised manuscript to further improve it. Each section has been improved. Detailed information is available in the responses to the comments below.
Line 62 – introduce sensory quality to the readers not familiar with these.
Answer: It is introduced as follows:
“However, little information is available on the influence of treatment of light factors and photoperiod of romaine lettuce on the sensory quality. The sensory quality of food products can be determined using the attributes related to color, texture, smell and taste.”
Lone 65-67 – repetitive text.
Answer: It has been changed into: “Few studies relating to light requirements in an indoor plant production system have concerned head-forming as romaine lettuce Lactuca sativa var. longifolia”
Introduce the problem of nitrate to the readers
Answer:
It has been introduced as follows: “The nitrate concentration in lettuce grown in hydroponics can approach a level considered hazardous for human health, therefore, it is essential for food safety to limit nitrate accumulation in leaves.”
Line 72 – a lot of facts in this section but the flow of information is limited and not clear.
Answer:
This paragraph has been slightly rewritten.
Line 93 – AI – needs more explanation and clarification.
Answer:
It has been specified as follows: “Besides the measurements of morphological traits, biomass production, or sensory attributes, the effect of different growing conditions on lettuce growth and quality may also be evaluated using artificial intelligence approaches including machine learning [27, 28, 29]. It is desirable to increase the intelligence of machines. Machines can be trained using the available data for processing and analyzing visual data and making predictions without human intervention [30, 31]. The application of artificial intelligence can be an important component of the agricultural revolution and can allow, e.g., predicting crop yield or evaluating the plant quality using image processing [32].
Design and replication not clear
Answer: It has been corrected as follows:
“The experiment used a two-factorial design of PPFD × photoperiod. In the case of both cultivars of romaine lettuce, three containers for each of the four light treatments were three and thus 12 containers in total for each of the cultivars. Six plants of one cultivar with three replications, and four treatments were selected. A total of 144 plants were used for the experiment.”
Section on QDA is not clear, needs more information to guide the readers.
Answer: Additional details have been added. It has been corrected as follows:
“For sensory evaluation, the method of Quantitative Description Analysis (QDA), i.e., sensory profiling, was used in accordance with the procedure included in the standard Sensory Profiling ISO 13299:2016… The brainstorming session was run to select attributes. During the analysis, each person was in the individual evaluation box equipped with the computer and specialized software (ANALSENS ver. 7) designed for the preparation of tests, recording of individual assessments and processing of the results. Lettuce leaf samples were brought to the stands... The intensity of each descriptor was assessed on a graphical scale, corresponding to 0 (low intensity) - 10 (high intensity) conventional units, with marginal markings. The evaluation was carried out in two sessions. In the case of each attribute, the mean was calculated.”
Image processes section is not clear – more information and clarification is required for the reader – “…place in a black box..”? What were the textual features?
Answer: It has been corrected as follows:
“The digital camera and light source were placed in a box of dimensions 1m x 1m x 1m with black internal walls. Color calibration of the digital camera was performed. The upper surface of each lettuce leaf was imaged separately. The leaves were placed on a black background.”
“The images were converted to individual color channels L, a, b, R, G, B, U, V, S, X, Y, Z. The L* is the lightness component, a* - green (negative values) or red (positive values), b* - blue (negative values) or yellow (positive values), R - red, B - blue, G - green, U and V determine the color itself (chromaticity), S - Saturation, Y - lightness, and X and Z components are color information”
“The image texture was a function of the spatial variation of the pixel brightness intensity. Textures can give information about the object structure and their quantitative analysis can provide insights into object quality [39, 40]. Textures were computed based on the co-occurrence matrix (132 textures), run-length matrix (20 textures), Haar wavelet transform (10 textures), histogram (9 textures), gradient map (5 textures), and autoregressive model (5 textures)”
Line 203 – same as M&M – no need to repeat
Answer: It has been deleted.
Section 2.7 should come earlier in the M&M?
Answer:
It has been split into several parts and included before.
Line 225 – information about PCA etc is not needed?
Answer: It has been corrected as follows:
“Principal component analysis (PCA) is a widely used multivariate analytical statistical technique. It was performed to synthetically determine the similarities and differences in sensory quality of romaine lettuce using the STATISTICA software ver.13.1. PCA was applied to QDA data to reduce the set of dependent variables (i.e., attributes) to a smaller set of underlying variables (called factors) based on patterns of correlation among the original variables [35]. When performing PCA, lettuce smell, color, crispness, juiciness, lettuce taste, sweet taste, bitter taste, grassy taste and overall quality were included.”
Line 284 – Quantum yield at which light levels? The QY would be expected to decline at a higher light level as it is a function of light.
Answer:
Those sentences have been deleted.
The authors calculate ETR from the QY – however, it is not clear what values were used for partitioning between the two photosystems and it appears that light absorption as not measured – and this would be influenced by day length and light intensity and therefore these were not taken into account. ETR should be removed and the authors should use the PSII efficiency to determine differences.
Answer:
ETR has been removed.
How the sensory information was evaluated is not clear and needs clarification.
Answer:
Analysis of sensory information has been specified as follows:
“Principal component analysis (PCA) is a widely used multivariate analytical statistical technique. It was performed to synthetically determine the similarities and differences in sensory quality of romaine lettuce using the STATISTICA software ver.13.1. PCA was applied to QDA data to reduce the set of dependent variables (i.e., attributes) to a smaller set of underlying variables (called factors) based on patterns of correlation among the original variables [35]. When performing PCA, lettuce smell, color, crispness, juiciness, lettuce taste, sweet taste, bitter taste, grassy taste and overall quality were included.”
Line 340 – some of this information will not be familiar for readers not working on machine learning approaches and needs more information.
Answer: More information has been added as follows:
“Multilayer Perceptron is a type of neural network using the back-propagation method. A supervised learning technique was used”
“The accuracy can range from 0 to 100%, and the TPR - True Positive Rate, FPR - False Positive Rate, Precision, F-Measure, ROC Area - Receiver Operating Characteristic Area, PRC Area - Precision-Recall Area can be in the range of 0.000–1.000. The higher performance metrics such as accuracy, TPR, Precision, F-Measure, ROC Area, PRC Ar-ea, and the lower FPR, the more effective the model is”
In general I found this section difficult to follow although this is not my area. However, clarification would broaden the readership.
Answer: It has been corrected. Clarification of the interpretation of the results has been indicated as follows:
“The accuracy can range from 0 to 100%, and the TPR - True Positive Rate, FPR - False Positive Rate, Precision, F-Measure, ROC Area - Receiver Operating Characteristic Area, PRC Area - Precision-Recall Area can be in the range of 0.000–1.000. The higher performance metrics such as accuracy, TPR, Precision, F-Measure, ROC Area, PRC Area, and the lower FPR, the more effective the model is”
Line 395 – remove very – not scientific.
Answer:
It has been removed.
Line 407 – the authors did not measure thermal dissipation or protection – and therefore this is not supported by the results. This should be removed or rephrased to not be misleading.
Answer: It has been corrected as follows:
“The decrease in the yield of PSII can suggest increased thermal dissipation of absorbed light energy as a result of photoprotective processes and a lower amount of absorbed energy driving photochemistry [47] or alternative electron absorption such as reduction of nitrates”
Line 414 “ However, high ETR was associated with a lower quantum yield of PSII, so the challenge for achieving increased efficiency of conversion of electrical energy into electron transport will be to find ways to minimize thermal dissipation of energy as well and photoinhibition” This does not make sense?
Answer:
It has been corrected as follows:
“The decrease in the yield of PSII can suggest increased thermal dissipation of absorbed light energy as a result of photoprotective processes and a lower amount of absorbed energy driving photochemistry [47] or alternative electron absorption such as reduction of nitrates”
The values of the image processing work is not clear in the discussion.
Answer:
It has been corrected as follows:
“The use of different machine learning algorithms for development of innovative discriminative models based on image textures of lettuce leaves subjected to different photoperiod and PPFD can also be considered as a great novelty of this study.”
“Some samples were distinguished from others with an accuracy of up to 100% as in the case of 16 h/160 µmol m-2 s-1, 16 h/240 µmol m-2 s-1 and 20 h/240 µmol m-2 s-1 for cv. ‘Casual’ lettuce. Additionally, for 16 h/160 µmol m-2 s-1 and 16 h/240 µmol m-2 s-1 samples, the values of True Positive Rate, Precision, F-Measure, Receiver Operating Characteristic Area, and Precision-Recall Area reached 1.000 and False Positive Rate was equal to 0.000. The accuracy of 100%, low False Positive Rate and high other metrics indicated an effective model. These results are very promising.”
“Machine learning may be promising due to better computational power than conventional techniques of data processing. The application of machine learning can extract more necessary information for the evaluation of the plant quality [34]. Machine vision can have many advantages such as high accuracy, high repeatability, and low costs. Therefore, inspection systems using machine vision can be applied in modern manufacturers, e.g., in food processing for quality control. Such systems can support making a decision besides techniques and methods involving, e.g., human experts, spectroscopy, or molecular markers, which may be time-consuming, subjective, or more expensive [64].”
In general the discussion is very repetitious of the results in places and needs to be tightened so that the overall take home messages and value are you clear. The same applies to the conclusion.
Answer:
The details have been added as follows:
“A comprehensive approach to evaluate the effect of different Daily Light Integral in an Indoor Controlled Environment on morphology, biomass production, sensory quality and image textures of romaine lettuce is original and was not reported in the available literature. The use of different machine learning algorithms for development of innovative discriminative models based on image textures of lettuce leaves subjected to different photoperiod and PPFD can also be considered as a great novelty of this study.”
“Some samples were distinguished from others with an accuracy of up to 100% as in the case of 16 h/160 µmol m-2 s-1, 16 h/240 µmol m-2 s-1 and 20 h/240 µmol m-2 s-1 for cv. ‘Casual’ lettuce. Additionally, for 16 h/160 µmol m-2 s-1 and 16 h/240 µmol m-2 s-1 samples, the values of True Positive Rate, Precision, F-Measure, Receiver Operating Characteristic Area, and Precision-Recall Area reached 1.000 and False Positive Rate was equal to 0.000. The accuracy of 100%, low False Positive Rate and high other metrics indicated an effective model. These results are very promising.”
“Machine learning may be promising due to better computational power than conventional techniques of data processing. The application of machine learning can extract more necessary information for the evaluation of the plant quality [34]. Machine vision can have many advantages such as high accuracy, high repeatability, and low costs. Therefore, inspection systems using machine vision can be applied in modern manufacturers, e.g., in food processing for quality control. Such systems can support making a decision besides techniques and methods involving, e.g., human experts, spectroscopy, or molecular markers, which may be time-consuming, subjective, or more expensive [64].”
The Conclusions section has been rewritten. Some information has been removed or added (please see the manuscript).
