Characterising Recycled Organic and Mineral Materials for Use as Filter Media in Biofiltration Systems
Round 1
Reviewer 1 Report
This manuscript is well written and well revised. I just have one comment about the Figure.
Figure 1 and Figure 2: Can you please describe how did you distribute the x axis of each graph? The graph shows +8, -8+4, -4+2, -2+1 and so on. Here does the particle size actually ranges from 8 to 4mm, 4 to 2mm, 2 to 1mm? Can you show it in the graph more clearly instead of using +/- sign?
Author Response
Graphs re-done to use "<" and ">" .
Reviewer 2 Report
The manuscript has been greatly improved since its initial version. The author’s have done a good job taking into account more of the relevant literature and have improved the structure by distinguishing more clearly between the methodology and results sections. They have also added important experimental details a much-needed description of how the model works (though I have some comments about improving the description).
I do have some remaining concerns about the manuscript, which must be addressed before it can be published.
-The manuscript was corrected remarkly fast, which is great, but there are quite a few grammar mistakes. It should be read over for grammar once the content is finalized.
-The citation numbers in the text do not seem to agree with the citation numbers in the References section (example line 52 refs 6 and 11 are not reviews on the use of compost as biofiltration media, line 227 « Performance & Validation Standards for Organic Bio-Filtration Media » is cited as references 22 but in the reference section both reference 1 and 25). These are not the only examples – the problem seems to be almost systematic. Have the authors used an automatic citation tool like Zotero ? If not, this would be helpful – if so, there seems to be a problem with updating the fields.
-Figure 1 : The X-axis is difficult to read (text too small) and understand (does +8 mean >8) ? Please use a more conventional system. I think it would also be more conventional to go from small particles to larger particles.
-Lines 127-132 : I still disagree about the 420 mm rain event. A rain event is the amount of rain producing the amount of water being treated. I agree that 1 L on your column produces 420 mm of water which goes through your system, but I don’t agree that this can be referred to a rainfall event (which would mean that it was the runoff produced by 420 mm of rainfall). This is not true because in the field, there would be a surface draining to the biofilter. In your modeling study, you consider that a 0.28 Ha (2800 m²) catchment is drained to a 100 m² rain garden. In this case (if we assumed a runoff coefficient of 1 for this surface, for example), the biofilter is receiving 29x the volume of rainfall falling on it directly. Therefore 420 mm on your biofilter corresponds to a 14.5 mm rain event. I think we actually agree on this but it is not clear in the way things are written in the paper. Please write something like « 420 mm of equivalent rainfall, corresponding to approximately 14.5 mm of rainfall assuming the biofilter covers 3.4% of its catchment ». (you can adjust the hypotheses here, of course – a runoff coefficient of 1 is certainly questionable, but it is important to underline that you are talking about an equivalent event and to say what that corresponds to in real rainfall).
-In lines 291-294 this issue comes up again. Here you talk about 3000 mm of equivalent rainfall and state that this is equivalent to 5 years of rainfall in Mebourne (600 mm/yr). The details of the volume you passed through your columns for these experiments were not provided in the methods section (please add this), but assuming that this was about 7 L through your 0.00238 column, then the 3000 mm (equivalent rainfall) needs to be divided by 29 again (or multiplied by 0.034, the ratio of the biofilter surface to the catchment surface) to get to real rainfall. Therefore, this is not 5 yrs of rainfall, it is more like 2 months of rainfall.
-I am not completely sure about this (the editors or journal managers will be more informed), but which it was interesting to include Appendix G of the Music Manual in the supplementary data, there may be copyright issues in publishing this ; the same could be true for Figures 4 and 5 if they are screen shots from Music (again, not sure on this).
-While the model description is important to include, lines 176-186 are an exact quote from the Music manual (and are not in quotations !!). In addition, these lines describe the use of k and C* in Music in general, so the description is not necessarily pertinent to bioretention systems. I believe that k and C* are used to represent a first order reaction kinetic in continuously stirred reactors – if their values depend on things like density and particle size distribution, they are describing settling, which is important in stormwater ponds, for example, but not in biofilters. Also, I believe the line ‘ambient concentrations in waterbodies receiving stormwater’ refers more to stormwater ponds (in the case of bioretention systems, the ambient concentration that could change between events would be that of pore water, in particular the water in the submerged zone). You also do not explained here how the look-up tables are taken into account, which is really important to understanding your point later. = Please focus only on how bioretention treatment nodes are evaluated and describe this thoroughly in your own words.
-Lines 230-232 : Please explain what the ANZECC guidelines are and what a trigger value is (the target range or the value that shows that there is a problem ?)
-Lines 250-258 : This should be in the methods section, not here (some of it is already in the methods section – move any new information to the methods section and cut the rest. Remember this is
-Line 268 : TKN is organic nitrogen (is this what you mean by bound ?) + ammoniacal nitrogen. I assume TON is the sum of nitrate and nitrite ? Ammonia/ammoniac are also soluble, so TON cannot be soluble nitrogen. Please correct this.
-Line 301 (Comparison to guidelines) : Please begin this section by explaining what you did in the response to reviewers : in Australia media needs to be equivalent to the FAWB to be used (is this a legal requirement or do municipalities just tend to follow it ?)
-Lines 370-375 : I appreciate that you have noted that short-term column experiments may not be representative of long-term behavior in the field. I think this is a major point that needs to be made in the conclusions – your experiments show that it does not leach that much under controlled conditions, so there really is a reason to believe that it might not leach nutrients in the field but they do not prove that it won’t leach nutrients in the field because you were not able to reproduce all of the field conditions (wetting/drying/further degradation of organic matter, etc). But modeling in Music is not going to fill this knowledge gap because as you show later, Music is based on empirical data and field studies haven’t been done with this type of filter media in order to be able to predict what might happen.
In my opinion, an empirical model like Music cannot be used to « validate the use of the alternative filter media », because you are getting out of its zone of validity. In addition, you make the point that the model cannot accurately represent the filter media you have studied – there is a logical contradiction in saying this and also saying that it validates the use of the filter media. Therefore, I think it is best to present the modeling in Music not as a validation of the filter media, but as a test of a tool which is commonly used in for dimensioning biofilters in Australia (again, not sure if this is a legal requirement or just a common practice – it would be interesting to include this information as well as a brief description of how it is used – do design guidelines require demonstrating that a certain load reduction would be achieved for given pollutants? If this is the case, would the results of the model mean that a larger biofilter would be required if a compost filter media were used ? If this is the case, it is certainly a deterrant for municipalities to used the compost filter media and if it based on false assumptions, it is quite important to correct this). This second point that alternative filter media cannot accurately be represented by Music is interesting and important to make, and it is important to show that further field studies of other types of filter media can make Music more reliable for other types of filter media.
This suggestion would make your conclusions less strong – you have not completely proven that your filter media will work well in the field – you have demonstrated that it has a potential of working well in the field and that Music tends to have assumptions that are not justified by your column tests, which may discourage use of filter media outside of the FAWB recommendations (this is in itself a very interesting conclusion). Both of these underline the importance of future work studying the performance of biofilters with compost media in the field in order to confirm the column observations and furnish further data to improve the representation in Music.
-Lines 395-403 : This discussion of choosing k and C* again seems to be based on the selection of values for systems with settling (discussion of specific gravity, which wouldn’t really come into play in filtration). Is it relevant to the bioretention nodes ?
-Line 416 : is this 16.9 and 19% or kg/yr (it corresponds to values in Table 12 of kg/yr). If values in Table 12 are all in % reduction, this should be stated clearly.
-Lines 412-please clearly state what output you were looking at for your senstivity analysis (outlet loads for your different pollutants ? Reduction in outlet loads for your pollutants ?).
-Conclusions : As I mentioned before, I don’t think you should talk about validating the use of the filter media wiht music.
-You talk about the same recent guidelines in lines 459-460 and lines 470-471. I would cut this out in lines 460-461.
Author Response
See attached.
Author Response File: 
Author Response.docx
Round 2
Reviewer 2 Report
I am happy with the way the authors have taken into account my remarks. They should be commended for all the work they have done on the manuscript during the review phase. In particular, I think the additions which explain the Australian context better add a lot to the significance of the paper as it really gets to the heart of questions that the industry can have about design strategies: is it better to impose a filter media that you know works well or to allow more freedom and innovation? The modifications to the modeling section are satisfactory as they are essentially used to show the effects the model would have on design rather than trying to draw any conclusions from the model itself.
One small comment: in line 348, the 7 L of runoff (which I realize was a response to my comment about the equivalent rainfall) now seems a little problematic as sounds like 7 L poured on a full-sized rain garden would have the same effet. Maybe it could be present in L/m² of rain garden or even in mm, but refer to it is as something like "water passing through the filter".
Besides this, there are a few grammatical issues. Now that the content is fixed, it would be useful to really proof-read for grammar.
Author Response
The final amendments have been made as per reviewer suggestions and the authors very much appreciated the review process to vastly improve this paper. Thanks.
Steve
This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.
Round 1
Reviewer 1 Report
This paper is well organized and scientifically sound. I would like to point out some minor changes as follows:
Keywords: First 3 keywords are already in the title, please use different keywords.
Page 1, line 37: Two “that that”
Line 44: Can you please elaborate “FAWB”?
Line 63: What do DM1 and DMS stand for?
Line 51: What is M165?
Figure 4: Legends are difficult to read.
Line 127: Typo “ECEC”
Table 3: If the guideline for organic carbon and total carbon is not available then instead of “?” please use “not available”
Table 6: “% Reduction” is not aligned.
Comments for author File: Comments.docx
Reviewer 2 Report
A comprehensive study titled "Characterising Recycled Organic and Mineral Materials for Use as Reactive Filter Media (RFM) in Biofiltration Systems" was conducted. The paper is well prepared and organized. The topic is important and good for the field of water.
The novelty and significance of the study should be addressed in the introduction part.
More recent studies with references must be provided.
It is recommended that raw data collected from experiments should be provided.
What is implementation of the results?
Reviewer 3 Report
The article “Characterizing recycled organic and mineral materials for use as reactive filter media(RFM) in biofiltration systems” has some interesting part but is not suitable for publication due to lack of organization of paper and novelty.
In order to increase the quality of information, I suggest some points
section of Introduction should be rewritten.
- add research reason, need, and goal.
-describe some information such as reactive filter media and biofiltration system compared with other research study.
-line 41, “results from this study indicate” move to results and discussion.
-
section of Materials and methods should be rewritten
From Line 77 to 153, some is experiment and some is results. The authors reorganize each content
in Table 2, what is the space between CaCO3 and Biochar?
Line 63, what is Figures 2-8 and 2-9?
Based on Table 2showing DM1 and DMS were consist of RO, Sand, M165 as volume, I think the value of table 3 has to be consist with Table 2. For example, in table 3, the value of DMS for Chloride soluble (362*1.0*1100) were sum of RO (2810+3030)/2*0.5*550 + 4.6*0.35*1520+1585*0.15*210. Why is the value different?
from line 144 to 153, authors mentioned “--- also demonstrated the contaminant---“. I did not find any contents related to nitrogen, phosphorous or nutrients before this paragraph. I think the authors did not demonstrate before.
In results and discussion, all section 3.1 needs to move to supplementary section
In table 4 (pollutant removal where stormwater used as influent to column experiments), I am confused to understand the information. Based on Table 4, I think the number is concentration of each pollutant. but the table title is “pollutant removal”
In table 4, the order of x-axis is “ storm water1, ro fine, ro medium, storm water 2, m165, coarse sand DM1, and DMS” Please add detail in the section of experiment.
where is Table 5?
In table 6 (metal removal from stormwater. % reduction), 162, 0.4, and 138 are concentration of influent. These are not %. What is the meaning of 4.2, 7.2, and 6.4? is it removal efficiency?
Reviewer 4 Report
This article presents laboratory tests and modeling done using using recycled materials as filter media in biofiltration systems. This work is interesting as it calls into question common practices for constructing stormwater biofilters in the Australian context; there are two major points that are particularly interesting: that the commonly recommended FAWB filter media may not be a sustainable resource and that existing design recommendations, in particular the use of the model MUSIC, basically have entrenched use of this filter media.
Unfortunately, this manuscript has major flaws in the way the work is presented and needs to undergo some major restructuration to be of a quality meriting publication in a scientific journal. In particular,
Ø The introduction does a good job of presenting why it is important to consider using recycled materials as filter media. However, only the FAWB filter media is discussed and no attempt is made to consider whether recycled materials have been used before elsewhere (spoiler alert: they have). Literature on bioretention devices in North America should be taken into account – many bioretention guidelines recommend using compost in filter media but they have sometimes had problems with nutrient leeching – I believe that some now recommend using specific types of compost to avoid these problems. There are several reviews, which would be a good place to start (Davis et al., 2009; Liu et al., 2014; Roy-Poirier et al., 2010). This doesn’t mean that your work is not interesting - you can use the fact that this has been used elsewhere to support that the FAWB media is not the only possible solution. The work needs to be justified within this larger context to really show how it is original. The term “reactive filter media” should also be defined clearly.
Ø The biggest issue with this paper is the presentation of the methodology, which is often lacking. In addition, the separation between “Materials and Methods” and “Results” is very strange – some results are in methods, some methods are in results.
o There need to be sub-headings so that this section is easier to follow
o The X-axis of Figure 1 is really hard to understand, and it is unclear why it is in reverse order.
o The recycled organics need to be described in more detail. Is this compost? What sorts of things go into it (food, grass clippings, leaves, wood…)? Could it be a source of pollutants that you haven’t studied (i.e. organic micropollutants)? How has it been prepared? This is an important point because not all organic matter will behave in the same way.
o Table 3 – Several parameters are presented several times in different units – don’t do this. Choose one set of units and stick with it.
o Lines 77-80 : this is true, but what is it doing here in the method section?
o Parameters of column experiments need to be justified – depth of column, depth of rainfall. 420 mm is extremely large for a rain event! Except that this calculation is not correct because you didn’t consider didn’t consider that a biofilter will also receive water from its catchment.
o It is unclear why the EC curve (Figure 4) has been presented. – Are you considering that EC is representative of something else? If so, this needs to be stated. Otherwise, there is no reason to present this. Some of the interpretations here are also doubtful or at least underexplained – and they are also presented as definitive rather than conditional statements, when they are hypotheses not fact. There is no clear logical connection between Ksat and the shape of the curve – in addition, opposite behaviors are explained (wide peak for biochar and higher peack for fine RO) by low Ksat – this doesn’t make sense. I would recommend cutting out this section (lines 109-137). If it is not cut out, it should be in the results not the method and it should clearly be justified how it tells us whether or not the material makes a good filter media. Same for the pH.
o All descriptions of soil parameters should also be moved to the results section
o Lines 144-153: Several problems here. Here you says Laughlan – in your references at the end it is McLaughlan. The method section is not the place to present information from the bibliography like this. It could either go in the intro or in the results to help you interpret your findings. In addition, these results appear to come from some kind of internal report, and no attempt has been made to find any published scientific work to support these claims – other people have worked on this before so there is no reason to rely only on unpublished work when you present this information.
o There needs to be a subsection that presents the methods used for the second and third sets of column tests. Also, water collected in an urban creek is not “stormwater runoff” (which would be collected in storm drain), it is water from an urban creek. Water from the urban creek can be used as a surrogate for stormwater runoff (which you can justify by the fact that only urban surfaces drain into it if that is the case), but this should be stated clearly.
o There needs to be a subsection that explicitly presents the hypotheses used to develop your life-span calculator, so that readers (and reviewers) can consider them critically. I think there are some hypotheses here that I do not agree with, but I would like them to be spelled out rather than having to guess what they are.
o The presentation of the modeling work done in Music needs to be presented in the Method section. A description of the model’s basic structure and its hypotheses must be included here. How are k and C* used in the model? How is the lookup table used? It is important for the user to have this information, especially since the model is being called into question.
Ø Results section:
o All of the results (and none of the methods) should be presented in this section. If the previous comments are accounted for, the results section will change substantially and it will be more possible to evaluate the results critically, so I will limit my comments.
o Remember that the question is not really whether recycled media can meet the FAWB guidelines but rather it can provide a good performance. It is important to get at why different recommendations are made. Some of the FAWB recommendations may not be optimal.
o Do not report numbers with lots of significant figures when you are working with rough estimations (Figure 6) or results from modeling that contain a lot of uncertainty (Table 10) – it just doesn’t make since to present a 97.6% TSS load reduction when this has been evaluated using a conceptual model like MUSIC.
Ø Conclusion
o The conclusions are interesting. The strong point of this paper is that it does have a critical view on current practices and that it points out some of the downsides of standardization in LID devices. I think some parts of conclusions probably need to be more nuanced – they need to be reevaluated once the methodology has been presented more clearly.
