Isotope-Based Early-Warning Model for Monitoring Groundwater–Leachate Contamination Phenomena: First Quantitative Assessments

Round 1

Reviewer 1 Report (Previous Reviewer 2)

This paper has been significantly improved compared to the original paper. However, there are still some issues:

1) I still don't know why the proposed method is called a "early-warning" model... What makes it an early-warning system? Compared to what? What are the standard warning systems currently put in place? And how does your proposed method compare to these?

2) I still don't understand why an F of 1.1 was chosen as the warning threshold.

3) If you were able to identify potential contamination by analyzing the groundwater for certain metals, what makes stable isotopic monitoring a better method or warning system?

4) I think there should be much more emphasis on isotopic characterization  of local groundwater that is most vulnerable to contamination, rather than characterization of the governing meteoric water line. While the isotopic composition of groundwater is usually related to that of precipitation, the isotopic composition of groundwater does not necessarily always plot on a meteoric water line. For example, it is common for mountain front recharge to undergo evaporation in mountain streams before entering the groundwater system, resulting in a significant decrease in the deuterium excess. Therefore, a subsequent increase in deuterium excess due to mixing with the leachate may not be recognizable. It appears that for your case study,  the isotopic composition of groundwater does plot along the meteoric water line, but this is not always the case. Another example is an aquifer with very old groundwater that is not receiving significant amounts modern recharge. In this case, the isotopic composition of modern precipitation does not influence the isotopic composition of groundwater. If groundwater does indeed plot on the local meteoric water line, it needs to be shown. 

It is MUCH better to directly characterize the groundwater, rather than to rely on assumed relationships between the isotopic composition of groundwater and that of precipitation. A sufficient time series data set for the isotopic composition of groundwater can be used to statistically characterize the isotopic composition and temporal variability for local groundwater. Then an outlier for deuterium excess can be used to set off this early warning system to do additional analyses  to assess specific contaminants. 

Author Response

Thank you and please see the attachment. 

Author Response File: Author Response.pdf

Reviewer 2 Report (New Reviewer)

This manuscript presents data from water isotopes in groundwater near a Landfill and in landfill leachate. The paper presents a novel model for detecting the presence of leachate in groundwater. This is based on deuterium excess in the analyzed waters due to the activity of methanogenic bacteria in the landfill. These drive excess deuterium data to higher values than do climatic events. Therefore, the authors present an innovative idea which is worth being published. The study adds data on trace elements which underline the theory of deuterium excess as a warning system and prove its applicability. That warning system is built on a very simple index F which will be easy to use in practice. The index F is applied in the field study and shows that leaking of leachates is present at certain seasons on that site. The reviewer found the study overall well written, clear, with good data. Considering a high novelty and originality, the paper can be published in Water, pending a few revisions as explained below.

L51: SMOW, not SNOW

L60: occurring under. conditions

L72-74: Sentence is not understandable in this context. Same sentence is on lines 109ff where it is ok.

I the pdf of the reviewer there were parts of text underlined with yellow colour - also this sentence. Remove that in revision.

L135 and elsewhere: better call the parameter d-ref theo for theoretical.

Figure 3 contains unreadable text in black rectangle, and also text in Italian that should be translated.

L181 Isotopes in title, not isotopies

198: isotope data

242 isotope compositions

L245ff: Deuterium enrichment is not clear: why are the negative numbers enrichments? Negative means depletion. The reader will be confused. Why not taking deuterium excess here? Or deuterium isotope ratios.

L314: Trace element concentrations

L386: Change spelling of isotopes in title

Figure 4: Axis labels can be printed in bigger font size.

Table 3: too big and to archival; better give that table as appendix or supplementary Material.

Alright, check for things like isotopes composition where the s has to be removed

Author Response

This manuscript presents data from water isotopes in groundwater near a Landfill and in landfill leachate. The paper presents a novel model for detecting the presence of leachate in groundwater. This is based on deuterium excess in the analyzed waters due to the activity of methanogenic bacteria in the landfill. These drive excess deuterium data to higher values than do climatic events. Therefore, the authors present an innovative idea which is worth being published. The study adds data on trace elements which underline the theory of deuterium excess as a warning system and prove its applicability. That warning system is built on a very simple index F which will be easy to use in practice. The index F is applied in the field study and shows that leaking of leachates is present at certain seasons on that site. The reviewer found the study overall well written, clear, with good data. Considering a high novelty and originality, the paper can be published in Water, pending a few revisions as explained below.

Thanks for your comments and suggestions to improve this paper. All changes in the paper are highlighted in yellow. Replies to comments are written in red.

L51: SMOW, not SNOW Ok, it’s done.

L60: occurring under. conditions Ok, it’s done.

L72-74: Sentence is not understandable in this context. Same sentence is on lines 109ff where it is ok. I the pdf of the reviewer there were parts of text underlined with yellow colour - also this sentence. Remove that in revision. The sentence on lines 72-74 has been deleted, as it is repeated on lines 109-111. All sentences in yellow, which were in the pdf version, referred to previous revisions. All new changes are underlined in yellow to make it easier for reviewers to correct.

L135 and elsewhere: better call the parameter d-ref theo for theoretical. Okay. All changes have been made to the text.

Figure 3 contains unreadable text in black rectangle, and also text in Italian that should be translated. Ok, the Figure 3 has been modified.

L181 Isotopes in title, not isotopies Ok, it’s done.

198: isotope data Ok, it’s done.

242 isotope compositions Ok, it’s done.

L245ff: Deuterium enrichment is not clear: why are the negative numbers enrichments? Negative means depletion. The reader will be confused. Why not taking deuterium excess here? Or deuterium isotope ratios. Thanks for the comment and your suggestion. In general, isotopic assesment, with respect to the SMOW reference standard, are indicated with negative numbers (see table 2). As is done generally for papers focused on the isotopic data assesment, in this part of the paragraph (lines 245-250), for the first elaboration, the discussion on the assesment of isotopic connotations was focused. This also aims to give even more soundness to the subsequent calculations for the F index. Generally, it is preferable to refer to the isotopic connotations (d² H and d¹⁸O) because they are measured data, instead deuterium excess is a calculated data. Therefore, we prefer to leave this part (lines 245-250) of the text in terms of isotopic connotations.

L314: Trace element concentrations Ok, it’s done.

L386: Change spelling of isotopes in title Ok, it’s done.

Figure 4: Axis labels can be printed in bigger font size. Ok, it’s done also for Figure 6 (ex Figure 5).

Table 3: too big and to archival; better give that table as appendix or supplementary Material. Ok, the data of Table 3 are shown in Appendix A (Table 1)

 

Comments on the Quality of English Language

Alright, check for things like isotopes composition where the s has to be removed

Ok, it’s done.

Reviewer 3 Report (New Reviewer)

See attached pdf file

Comments for author File: Comments.pdf

Author Response

COMMENTS ON THE MANUSCRIPT

Line 17-22. Unclear sentence

Thanks for your comments and suggestions to improve this paper. All changes in the paper are highlighted in azure. Replies to comments are written in red. Please refer to the new sentence formulation (rows 16-19 R1_version)

 

Line 44-46. The sentence “… correlate …. with precipita�on” is anon sense sentence

Thanks for your comments and suggestions to improve this paper. All changes in the paper are highlighted in yellow. Replies to comments are written in red. Please refer to the new sentence formulation (rows 80-89 R1_version)

Line 46-47… is correlate with the physical condi�ons (… evapora�on during precipita�on) …. Line 50. The delta values of deuterium and oxygen isotopes does not represent an “isotopic varia�on”; they are only “values”!

Thanks for your comments and suggestions. The correction was made, but we decided, as suggested, to shorten this paragraph. Please refer to the new short version of the Introduction.

Line 59-62. Surprising sentence: GMWT does not represent equilibrium condi�ons at 25 degrees!

Thanks for your comments and suggestions. Surely an excuse for the refusal. Please refer to the new short version of the Introduction.

Line 45 to 102. This chapter must be largely shortened. It generates confusion in the mind of the reader. Most cita�on and descrip�ons of meteoric water lines is useless.

Thanks for your comments and suggestions. Surely an excuse for the refusal. Please refer to the new short version of the Introduction.

More in detail:

Lines 49-50 and 66-67.

As it frequently occurs, also in this case there is incongruence of delta and d-surplus defini�on. The IUPAC (Interna�onal Union of Pure and Applied Chemistry) defini�on of delta is the following: IUPAC definition δiupac = (R/RST) -1 = [1000*(R/RST) -1]/ 1000 = X/1000 = X‰ (1a) where ‰ = 1/1000 and X = 1000 δiupac. The old ambiguous defini�on of delta is: Traditional old definition δold =1000* [(R/RST) -1] (1b) Thus δold = 1000*δiupac Dansgaard (1964) define the d-surplus in the following way: d-surplus = δ(2H)old – 8*δ(18O)old Thus: d-surplus =1000* [δ(2H)Iupac – 8*δ(18O)Iupac] Concluding: using the IUPAc definition you must write: d-surplus = 1000*[δ(2H)Iupac – 8*δ(18O)Iupac] δ(2H)Iupac = 8*δ(18O)Iupac + (A/1000) Using the old defini�on, you must write: d-surplus = δ(2H)old – 8*δ(18O)old δ(2H)old = 8*δ(18O)old + A I hope it is clear. Unlikely, the different authors mostly report the data in uncorrected way!

Thanks for your comments and suggestions. The correction was made, but we decided, as suggested, to shorten this paragraph. Please refer to the new short version of the Introduction.

Line 81. The primary reason of methanogenesis are not bacteria; the primary reason are the redox condi�ons of the system. Bacteria play only a cataly�c effect on thermodynamically spontaneous reac�ons!

Thanks for your comments and suggestions to improve this paper. All changes in the paper are highlighted in yellow. Replies to comments are written in red. Please refer to the new sentence formulation (rows 90-95 R1_version)

Line 89-97. Useless for this paper.

Thanks for your comments and suggestions to improve this paper. This sentence was deleted.

Line 109-111. Repe��on

Thanks for your comments and suggestions to improve this paper. This sentence was deleted.

Line 112 and following. I do not understand why the authors reported the East Mediterranean meteoric water line (MMWL) which is far from the water line which characterise Central Italy. All the discussion is useless. In addi�on: the defini�on of the d-surplus of Dansgaard is strictly valid only in case the local precipita�on gives the same regression line. The authors must use the local precipita�on regression line in case they suppose the interac�on occurs with precipita�on waters. The authors are lucky: they have this line for Pian dell’Elmo (see lines 232-233)

Thanks for your comments and suggestions to improve this paper. We elaborate the d-excess starting from the local meteoric water line for Pian dell’Elmo station as you suggested. The Pian dell’ Elmo (MC) station has been identified as a reference of precipitations isotope composition in this area and its isotopic data are provided by the International Atomic Energy Agency (IAEA). In fact, the equation (5) is defined by the isotopic data for 2002 year. Please refer to the new sentence formulation (refer to the updated version of the model, rows 123-163 R1_version)

! I think that the F parameter introduced by the authors is useless and, in some case misleading because it is referred to GMWL of Dansgard. I think that for the authors should be more useful to proceed as follows: Calculate the regression line δ(2H) on δ(18O) for Pian dell’Elmo (MWL): δ(2H)MWL = B * δ(18O)MWL + A and the related Standard error on regression s(yx). On the basis of a priori considera�on of opportunity, define a threshold value k*s(yx), where k is a coverage factor 2, 3 , … as used in geochemical prospec�ng. Consider now the generic sample Sp with determined values δ(2H)sp and δ(18O)sp. Thus calculate: Treshold = δ(2H)MWL(Sp) + k*s(yx), = B * δ(18O)MWL(sp) + A + k*s(yx), where δ(2H)MWL(Sp) is the calculated values of the sample Sp on the regression line using δ(18O)MWL(sp) Now consider the ra�o Contrast = δ(2H)sp/ threshold We have suspect of pollu�on in case is contrast > 1. Page 4

About the parameter F. In science, introduc�on of new parameter when similar already exist is a useless exercise. See comment above.

Thanks for your comments and suggestions to improve this paper. We elaborate the d-excess starting from the local meteoric water line for the Pian dell’Elmo station. Please refer to the new sentence formulation (rows 135-156 R1_version)

Lines 132-136. To perform a meteoric waterline, 4 (or 5 as reported in Fig.2) rain samples are not enough. Since rainfall must be collected throughout the year, it is worth sampling monthly and at different al�tudes if necessary. 4 samples represent the very least. If you really don't want to use the local waterline, use the Central Italy one (Longinelli and Selmo, 2003) with the threshold indicated above.

Thanks for your comments and suggestions to improve this paper. We elaborate the d-excess starting from the local meteoric water line for Pian dell’Elmo station. Please, refer to the updated version of the model, rows 123-163 R1_version)

Line 145-146. This subdivision may be of interest from a bureaucra�c point of view but has no scien�fic significance. There is something quirky, formalis�c, pedan�c about it.

Thanks for your comments and suggestions to improve this paper. This sentence was deleted.

Line 197. Table 2. If you use the IUPAC’s defini�on of delta, indicate 1000*δ( 2 H) and 1000*δ( 18O) in place of the very esoteric symbols δ2 ‰H and δ18‰O! Line 237 and following. The mixing processes described are only hypothe�cal. There is a permanent ambiguity on the use of the Mediterranean Water Line; this line is not relevant for the paper. See above.

Thanks for your comments and suggestions to improve this paper. We elaborate the d-excess starting from the local meteoric water line for the Pian dell’Elmo station. Please refer to the new Table formulation (R1_version).

Concluding. The paper cannot be accepted in the present form. The paper should be corrected for formalism used and largely shortened; moreover, the definiton of the parameter F is redundant: similar defini�on is already used in geochemical prospecting.

Thanks for your comments and suggestions to improve this paper. We elaborate the revised version of the manuscript accordingly.

Round 2

Reviewer 1 Report (Previous Reviewer 2)

This manuscript has been improved a lot. 

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 proposes a new assessment method for detecting the influence of landfill leachate on groundwater using d18O and dD stable isotopes. The idea presented in this paper is both interesting and promising for identifying groundwater contamination by only using the stable water isotopes. Therefore, I strongly recommend that the methods proposed in this paper should be verified thoroughly using other indicators, such as hydrochemical parameters, multiple isotopes, or microorganisms, to detect the impacts of leachate on groundwater. This paper should be resubmitted after the hypothesis has been verified.

Reviewer 2 Report

 While the subject of this paper is interesting and potentially important, this manuscript (and the project in general) needs quite a bit of work. The idea of having an “early warning” indicator of groundwater contamination is appealing. However, the authors use this term many times without defining it. To me, an “early warning” means that I will be warned about potential contamination in the future and that as of yet, significant contamination has not occurred. The effect of methanogenesis on the stable isotopic composition of water has been shown, but how sensitive is this method of detecting contamination? Can you observe a significant change in dD before most contaminants reach their MCLs? If not what is the advantage of using dD as an indicator of contamination versus other constituents such as chloride, nitrate, chloride – to - bromide ratio, methane, or a specific contaminant of concern? This is very important and it needs to be addressed. It would be great to be able to use this isotopic effect of methanogenesis to our advantage in this way, but you need to show that this method has an advantage over existing methods of detecting groundwater contamination at landfills.

The authors put in a lot of effort to describe meteorological controls on the stable isotopic composition on precipitation and the deuterium excess and attempt to establish a meteoric water line (global, Mediterranean, local) as a precise diagnostic boundary for assessing groundwater contamination. A meteoric water line is not a good diagnostic boundary because:

1)   Meteoric water lines are not precise lines that represent some “ideal conditions” They are basically linear regressions of multiple precipitation isotopic data. Actual isotopic compositions for precipitation group around the meteoric water line, and sometimes can show significant variability over space and time.

2)   Precipitation is not being evaluated for contamination… Groundwater is! Yes, the isotopic composition of groundwater is controlled by regional and/or local precipitation, but the relationship between the isotopic composition of groundwater and precipitation can be complex and depends on many different factors, such as local and regional climate, geographic location, proximity to recharge area, recharge mechanisms, local and regional hydrogeology, etc.

Therefore, while a short description of controls on the isotopic composition of precipitation and meteoric water lines is appropriate in the introduction and/or background description, you do not need to go into as much depth.

There are areas where the isotopic composition groundwater is not related to that of local precipitation but rather it is controlled by precipitation 50+ miles away in the mountains, where the isotopic composition of precipitation defines a different LMWL. I think the only way of defining a diagnostic boundary for assessing potential contamination is by characterizing the isotopic composition of groundwater in the study area. The spatial and/or temporal variability of the isotopic composition for groundwater can differ a lot depending on the factors listed above. Therefore, the general characterization of these factors and their effects on the isotopic composition of local groundwater will be necessary. Probably the best way of establishing a site-specific isotopic boundary for a potential contamination warning system would be to monitor the local groundwater on a regular basis (at least monthly). A pre-landfill chemical and isotopic characterization for groundwater is preferable, but at the very least, groundwater should be monitored on a regular basis while the landfill is in operation.  This ever-growing groundwater isotope dataset will allow you identify anomalous dD values for groundwater that sets off the alarm for potential contamination.

The author’s justification for the F>1.1 to enter the “alarm zone” does not make sense to me. I think you will need to do some actual experiments to determine where the alarm zone is.

For the case study in Italy, you need to include MUCH more information, including:

1)    A figure showing the layout of the landfill, including location of monitoring wells, drain system, etc.

2)   General climatic, geologic and hydrogeologic setting

3)   Discussion of groundwater chemical and isotopic characterization

While there remains a lot or work to be done, if the authors can show that this method has an advantage over other existing contamination warning systems, I think they should pursue this project and ultimately publish this paper.

 

 

Reviewer 3 Report

 I found the concept of the isotope based early-warning model to be very interesting. This is an important concept, and one that I think will be used by others.  There are, however, some significant problems in the design of the study (inadequate number of data), along with the lack of detail (methods, etc.) that make this manuscript lacking. It could have been so much better.

·      The methods section is incomplete, which is unacceptable. What instrumentation was used to determine the isotope compositions? How were the samples like leachate samples (that contain VOCs), pretreated or the data post-treated to correct for known interference that occurs with isotopic measurements? This is a potentially critical oversight by the authors. However, because the methods are not described at all, the reader cannot assess whether any of these known issues were taken into consideration. Here is one reference as an example by Pires et al. (2022): https://onlinelibrary.wiley.com/doi/full/10.1002/eco.2444?utm_sq=grq7sdtfar The Picarro company, also sells in-line equipment, but as the reader, I cannot tell if this was used. https://www.picarro.com/products/micro_combustion_module

·      The method seems like it should work. However, for a “novel” method, this experiment does not seem to be designed optimally. Why are there no traditional metrics reported for leachate determination, such as conductivity, ionic composition, nitrate, pH, etc.? The use of only 2 to 4 “alarm zone” samples to define an entirely new method, without tying them to traditional indicators seems weakly founded.

·      The data presented and some of the interpretations made are incredibly confusing to the reader. For example, lines 144-145 seem incorrect here to me. It states that “10% is the maximum variation we can have, usually for d2H and d18O contents in rainwater”. This is simply untrue – universally, worldwide.  First, nearly every meteoric water line constructed in the world shows drastic seasonal variation in the d2H and d18O contents of rainwater, certainly by more than 10%. This is also true for deuterium excess values.  A simple examination of the authors’ Table 2, however, shows that there are only 2 “AP” samples presented in the study that indicate rainwater (per the language for Table 2).  How can this be? Where are the precipitation data?  How are the rainwater samples so drastically different from the groundwater samples? Clearly there is something wrong with the data in this study, or the reader is not presented with enough data to clearly make sense of the results.  This is where a figure such as precipitation d-excess vs. season should be shown, but those data either don’t exist or are not shown to the reader. There needs to be a better list of background data on precipitation and groundwater. Individual seasonal events can vary drastically, but the authors’ claims (e.g., Line 144-145) seem to ignore this.

·      Lots of awkward writing, with way too many commas. About half of the commas are not grammatically correct. The writing could be more concise in many areas. For example, line 202 is much more complicated than it needs to be with phrases such as “the exceeding of a value”, etc. 

·      The warning zone seems quite narrow. First, there are not enough data in the study to be able to define such a narrow range with any statistical power.  Furthermore, this range is only 10%.  How does this compare to the additive errors one might encounter?

·      The “Alarm Zone” seems to be defined by only 4 samples, and it indicates where methanogenesis has occurred. Must all leachate have methanogenesis? If not, this method will not adequately screen all leachate.

 

 

Line 16: Awkwardly written.

Line 27: Awkward. Remove “in”?

Line 75: Unclear – There is a deuterium increase in what, the residual water? Be clear for the reader, and use the ratio.

Line 122: “line to refer to”

Line 255: Grammar.

Line 265: spacing

Line 267: “lactate” really?  Do the authors mean “leachate”?

Line 268-274: What other seasonal variations could cause this? Change in seasonal isotope composition is obvious, but what about water level changes in the aquifer/landfill that could yield such differences?

Line 301 and elsewhere: Take more care in typing in references. It’s sloppy.

Line 293 (Acknowledgments): No one is worthy of acknowledgment? How about the folks at the landfill who allowed sampling? Others?

 

Additional Comments:

All figures: The text on most figures (especially the axes) is way too small and is difficult to see.

The figure captions on nearly all figures are inadequate. They don’t explain to the reader that they’re looking at. The captions should explain the symbols (e.g., Figures 4 & 5). The colors are not always easy to distinguish (e.g., April, July, and October in Figure 4 – why not change colors and shapes so they are more easily distinguishable.

 Figure 2. This is somewhat confusing and could be better explained to a reader not intimately familiar with the process.