Vegetation Survival in Green Roofs without Irrigation

Round 1

Reviewer 1 Report

Overall:

1) The paper requires extensive English editing – word choice, sentence structure, missing/improper words, misspelling, etc. It is almost unreadable as is and it detracts strongly from the perception of technical value.

2) It is important for the authors to accurately frame the paper for their technique can and cannot do, how it can be used, and with how much data, etc. The overall concept of predicting green roof plant survival is a good one and can add a little something to other types of green roof modeling/design efforts. Other efforts look at system performance when subjected to individual storms AND sequences of storms. Modeling sequences of storms is not novel, but looking at plant survival seems to be. The approach in this paper would not be used alone, because it can say nothing about green roof performance which is what a green roof is designed for. It is not designed to simply survive, it is designed to capture a given portion of a design storm and/or reduce peak runoff and/or delay peak flow. Only after that performance-based design is completed would a designer be interested in whether it would survive without an irrigation system and thus whether or not to either a) provide a moisture-sensor-activated irrigation system, b) increase the depth of the media to allow survival during a typical dry summer (as suggested in this paper), or c) include a storage feature in the design which helps with both volume retention and plant survival. However, I believe that the simplifying assumptions make the approach in this paper too simplified to really allow this. The approach to calculating drying via only a constant assumed ET rate is oversimplified and not realistic. It is not clear than any soil moisture info was used to verify the approach used. Normally, an expression that incorporates meteorological parameters such as temperature, humidity, etc. (e.g. Penman-Monteith or other) and a crop factor are used for ET. The effect of drainage and storage cannot be ignored in the water balance of the media, they are both critical. Normally, expressions for a soil moisture retention/drying curve (e.g. vanGenuchten) and hydraulic conductivity (e.g. Mualem or other) are used to model soil moisture content. These expressions are a strong function of the media type which does not appear to be addressed in this paper. The other problem is that the case studies do not make clear what is going on. Clearly there is a large data set of rainfall amounts and timing. However, what is being verified in the green roof is not clear. Nowhere does it talk about soil moisture readings or soil moisture predictions which is what determines plant survival-so it had to have been predicted.  How was that measured at the case study sites? at how many depths and different locations? Where are those predictions?  There is no description of the amount of data required to get a desired level of accuracy of prediction, and no quantitative measure of prediction/calibration accuracy. Without this, the value/accuracy of the approach is not understandable or convincing.

A few specific comments:

1) Abstract: Does not give any quantitative or specific findings, only general statements. Please make statements of actual results. The paper needs to explain, with an example, the steps of how this technique would be used for design of a green roof.

2) Introduction: The statement attributed to reference 13 does not seem correct and does not seem to agree with the rest of the Introduction. Is the utility of green roofs really in question?

The stated objective is to estimate the probability of vegetation to maintain a minimum water content based on a series of rainfall inputs. However, I cannot find where this minimum water content is identified; is it 0.05 or 0.01? The practical meaning of the average return interval (ARI) parameter should be explained.

3) Green Roof Hydrology: Define “plastic profiled elements”. Does this mean modular plastic trays? Green roof design is described as including selection of media depth and plant cover. However, the characteristics of the media are also critical to the water balance of the system. ET cannot be modeled as a separate process that is not affected my soil moisture retention and drainage functions. In other words, the amount of ET that will occur is not related to rainfall amount in a simple way.  Also, ET is not a simple process that can be correctly characterized as a constant, maximum value (potential ET in this paper). It is affected by several parameters, and does not occur during rain events. The assumption that ET is always equal to the maximum value seems like it introduces significant error. Can this error be quantified? Please explain the significance/meaning of the quantities: γ, β, βi and βi*

4) Case Study:  In English numeric notation, periods are used for commas. E.g. 150,70 should be 150.70

Model not explained well: what are the input data? How much data are needed to run the model? How is it calibrated to use for design? Is the output a time sequence of the average soil moisture content? Is that data available? How accurate is the prediction (a quantitative measure of prediction strength)?

A typical range of ET is mentioned (1-6 mm/d), so what value of ET was used?

The meaning of the results in Tables 4-5-6-9-10-etc is not explained well. What does it mean? Is ARI how long the plants will live if the media is the given depth? How this output is useful for design is unclear.

What value of minimum water content is used for plant survival (0.05, 0.01 or something else)? It is not explained how the model works. How does soil moisture relate to ET? Is it assumed that all moisture content, irrespective of depth is subject to ET withdrawal and irrespective of drainage functions?

5) Conclusion:  The proposed method (what data required, what is assumed, what would happen if storage was not ignored, effect of plant type, etc) and how it would be used for design is not understandable from this paper as written. The technique is fuzzy at best. The general conclusions sound good, but are not supported/explained in the paper. There must be a quantitative measure of prediction accuracy of the prediction technique. How much data is needed? Just predicting whether a green roof will survive (the definition of which is not clear) in a very simplified way that is not proven with any comparison of predicted and measured soil moisture in an green roof seems to add very little to the science of green roofs. It is not shown that this technique actually works.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

I would like to congratulate the authors for having obtained an equation to estimate the growing medium thickness required in the green roofs design. However, there is one main questions that to my understanding needs clarification: to prove the goodness of equations (7) and (8), in chapter 4 (case study) the authors refer to a continuous simulation analysis of recorded rainfalls, but the manuscript does not explain anything about that model. Is it explained in previous articles? Has that model been calibrated with experimental results obtained from monitoring the two green roofs cited in the case study? If all that information has been previously published, there is no need to include a long narrative about it, but some sentences clarifying it, would improve the understanding of this manuscript.

In my opinion, other aspects that would need further clarification, are as follows:

-Lines 161-162: The period of time analysed in each case study is very different. Have the authors any comment about it? Are the results equally significant? It would appear than results from Calabria could change depending on how representative is the selected year (was it a wet-medium-dry year?).

-Line 163: Has the value of 6 hours been calculated using ref [32]? Is it the same value for Milano and Calabria?

-Lines 185, 186: Have the soil moisture content at saturation of the growing media values been set based on results presented on the quoted references ([42], [39])

- Lines 224, 225: What are the bases to assume those values for Et?

Author Response

Please see the attachment

Author Response File: Author Response.docx

Reviewer 3 Report

Minor revision

The subject of the article is interesting, but the structure of the paper needs to be improved.

1- The literature review needs to be improved. In this regards, and to show better the meaning of vegetation survival in green roofs in summer, a table can be added to present the monthly water consumption by green roofs in different climates and the monthly precipitations in the summer.

2- The novelty of the paper is not clear. It is necessary to be added at the end of the introduction.

3- For improving the structure and sections:

• “ Green roofs hydrology” can be in “1. Introduction” section as “1.1. Green roofs hydrology”.
• “3. Probabilistic model” can be in “2. Materials and Methods” section as “2.1. Probabilistic model”.
• “4. Case study” can be in “2. Materials and Methods” section as “2.2. Case study”.
• The assumptions need to be added to the paper at the end of section 2 as “2.3. Assumptions”. For example, climate change is not considered the used date for Milan case study is 1971-2005, and for Calabria case study is 2015-2016. The building's type can affect the roof temperature and result in the evapotranspiration by the green roof, etc. These can be added as assumptions.
• From line “171”, “Table 3 contains the correlation … “ can be put in result as “3. Results and discussion”
• At the end of results, suggestions for future studies as “3.1. Recommendation for future works.”

4- More information about the case study needs to be added. For example, climate type of the two case studies, besides some information about the average temperature and rainfall, particularly in summer for both case studies. The type of vegetation in selected case studies are not mentioned and need to be added.

5- The conclusion needs to be extended, and the results should be presented in terms of parameters such as climate type, green roof type, plant or vegetation type, not the name of two case studies. The survival probability of which vegetation? In which climate? and in which type of green roof?

Author Response

Please see the attachment

Author Response File: Author Response.docx

Reviewer 4 Report

The manuscript reports on the implementation of green roofs as sustainable urban drainage systems. The probability of vegetation survival without irrigation has been investigated by means of their application to two case studies in northern and southern Italy.

The article is an original contribution and the topic is of interest for the readership of the Water journal.

English language is clear, and the presentation is adequate. The provided figures are necessary for the presentation and the understanding of the results. Anyway, I have detected some criticisms in the text that should be properly addressed. Further discussion is required on the obtained results and their implications. Some results/comments on the improvement of the analytical probabilistic model should be presented. Conversely, this part should be removed from the abstract.

The Authors can benefit from the comments below to improve their paper. These have to be accomplished before manuscript acceptance.

Abstract

The abstract is concise and reflects the content of the article.

Lines 10-11: please replace “average return interval of the water content at the end of the dry period” with “average return interval of the water content at the end of a dry period”.

Introduction

Aims of the study are clarified in the Introduction and supported by relevant references.

Lines 25-26: “… they constitute from about 30% to 50% of the urban impermeable surfaces”. This sentence should be circumscribed.

Green roofs hydrology

Lines 84-86: Please, discuss on the choice to neglect both interception by vegetation and drainage layer capacity.

Probabilistic model

This section is clear and adequately detailed.

Lines 118-119: “Rainfall variables … that is…” Please, check the text.

Lines 119-120: the assumption of independence between correlation between rainfall depth and rainfall duration does not correspond to reality. Please discuss on this assumption and its implications in the results.

Lines 121-122: Specify in this Section the adopted criterion for the assessment of minimum interevent time IETD.

Line 127: after [25] put a “.”.

Case Study

This Section is presented in a logical sequence, but further discussion is encouraged on the obtained results and their implications.

I suggest renaming the section “Case Studies”.

Lines 175-176: the correlation between rainfall depth and rainfall duration is high. The assumption of independence between rainfall depth and rainfall duration strongly affects the results. This should be specified in the “Conclusion” Section.

Line 186: I suggest to briefly discuss on the choice of the values of soil moisture content at saturation of the growing medium for the two case studies. Even if relevant references are included, it is useful to briefly specify the reason of the adopted values.

Lines 206-212: P and ARI are very different for the two case studies for given z_g. P and ARI increase more slowly with z_g for the case study of the University of Calabria. Discuss on this different trend, which is not expected while examining the average values of rainfall variables in table 1.

Lines 224-225: Discuss on the choice of E_t for the two case studies.

The improvement effects of the analytical probabilistic model are not presented.

Conclusion

Conclusions seem reasonable and are supported by the results.

References

The study is supported by relevant references. Based on my knowledge, no important reference is missing.

Author Response

Please see the attachment

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

I looked at the materials resubmitted. I do not believe that it is adequate. I still believe in my first analysis that the paper should be rejected outright. I do not wish to do a detailed review with comments, my first detailed comments were not addressed well.

Author Response

Dear Author,

we are sorry that your opinion has not changed despite our careful review.

Best regards and happy holidays!

Reviewer 2 Report

Dear authors,

Thank you for addressing previous revision comments. I would suggest to add an additional reference to lines 37-40 from a similar climate:

Andrés-Doménech, I.; Perales-Momparler, S.; Morales-Torres, A.; Escuder-Bueno, I. Hydrological Performance of Green Roofs at Building and City Scales under Mediterranean Conditions. Sustainability 2018, 10, 3105. http://dx.doi:10.3390/su10093105

Congratulations for your manuscript.

Author Response

Dear Reviewer, 

thank you!

Reference [14] was added to manuscript. In accordance with this add, subsequent references have been changed. Changes was highlighted in red in the manuscript.

We wish you a merry Christmas and a happy new year!

Reviewer 4 Report

The manuscript has been significantly improved following the recommendations of the Reviewers; all my concerns have been addressed and convincingly justified.

Author Response

Dear Reviewer,

thank you.

We wish you a merry Christmas and a happy new year!