An Integrated Approach to Characterising Sulphur Karst Springs: A Case Study of the Žvepovnik Spring in NE Slovenia

Round 1

Reviewer 1 Report

The article deals with the geochemistry of sulfur spring which is quite rare in the studied karst system in NE Slovenia. The authors present a detailed description of the geological location of the sulfur Żvepovnik spring and show an integrated approach to recognize the origin of spring groundwater as well as the origin of its chemical composition. A very detailed chemical and isotopic analyses were clearly presented in supplementary materials. A huge advantage of such integrated approach was providing a large set of isotopic analysis of water and dissolved compounds, as d¹⁸O and d²H of water, d¹⁸O and d³⁴S of dissolved sulfates, d¹³C DIC, and  ³H tritium. Based on the concentration and isotopic composition of dissolved sulfates the possible origin of sulfates in sulfur spring were determined.

The manuscript is interested and well written however needs more improvements. I found some elements which are not sufficiently clearly explained or missing in discussion. That is why I recommend the manuscript to major revision before publication.

My major remarks are attached in pdf file.

Comments for author File: Comments.pdf

Author Response

We thank the reviewer for valuable comments and suggestions that were carefully considered and taken into account. Answers to specific questions are listed below.

My major remarks are follows:

1. First of all it is not clear why the studied spring is called “sulfur spring” because provided chemical analysis do not support such term: a) according to Piper diagram the dominant ions are HCO3-Ca-Mg, and sulfates are not the dominant anion;

Answer: Thank you for remark. We tried to explain our decision of using this term. The term of “sulphur spring” we use to describe and emphasize the phenomenon of “special” type of spring in karst, the spring with odor of H₂S. You are right that sulphate is not the main anion, but its concentration is higher than generally in other (majority) of karst and fractured aquifers (Verbovšek and Kanduč, 2016). With the term used we want to emphasize the specificity of the source, not its detailed hydrogeochemical characteristics. In addition, the local name of the spring is Žvepovnik which derives from “žveplo” what means “sulphur” in Slovenian language.

1. b) in the text are use terms “sulfur spring”, “sulfate spring” or “sulfidic spring Żveplenica”, etc.. I recommend to organize this terms in all text and use only one term and explain clearly why such springs are called “sulfur springs” but not for example “sulfate spring” – which would be adequate in context of chemical composition.

Answer: Thank you for suggesting the solution. We unified the term throughout the text – “sulphur spring”.

2. There is missing information about bicarbonate content – i.e. HCO3- ions. There is no concentration of bicarbonates provided in the tables. However some value of HCO3- is taken to Piper diagram. Only alkalinity is provided. Adequate explanation should be provided concerning relationship between measured alkalinity and content of bicarbonate ions in studied spring water in section “2.3.4”. In case if bicarbonates were measured in other way – it should be described in the “Materials and methods”.

Answer: After collection, water samples were refrigerated and immediately transported to laboratory. We provided the alkalinity values as they were measured directly as soon the samples were transferred to the laboratory. We did not perform the HCO3- measurements along the other ions or species in the ActLabs, as the interval between the sampling and lab would be too long. However, we thank for the comment. It was calculated from the alkalinity in the program Phreeqc for Windows, which considers that HCO3- constitutes almost all of the total alkalinity (5.22 of total 5.4 mM) at the measured pH value (7.44). We have added the value of HCO3- (318 mg/L, calculated from 5.22 mM) in the table.

3. Fig. 6, line 513: there is more novel equation for GMWL : δ2H = 8.13 × δ18O + 10.8 after e.g.Rozanski, K.; Araguas-Araguas, L.; Gonfiantini, R. Isotopic patterns in modern global precipitation. In Climate Change in Continental Isotopic Records. Geophysical Monograph; Stewart, P.K., Lohmann, K.C., McKenzie, J.,Savin, S., Eds.; American Geophysical Union: Washington, DC, USA, 1993; pp. 1–36.

Answer: Thank you for remark. We appreciate the suggestion however we decided to keep the original GMWL model equation suggested by Craig as a reference to which we compare our data.

4. Line 563: “The deep (older) water of Żvepovnik spring…” – it is not clear why authors use such term concerning “old” deep water. Taking into account provided in this study O and H isotopic composition of spring water as well as tritium content there is no evidence for presence of some old or deep groundwaters in karst system of the studied spring. The authors have very nice data base of groundwaters from other springs or wells or boreholes in this karst system and should provide example of chemical and isotopic composition of deep waters in order to justify the mixing process. Having two end-members (e.g. deep waters and atmospheric precipitation) it is possible to calculate mixing proportions.

I recommend to reconsider mixing idea or explain in more detail the hipotetical influence of “deep (older)” water in the hydrological system of Żvepovnik spring.

Answer: Thank you for important remark and suggestion. The existing data of isotopic composition of groundwater and precipitation can unfortunately not be used for calculation of mixing proportions of the investigated sulphur spring. There are two main reasons: (1) based on the geological characteristic of local and more regional area we cannot assume a common karst system, and (2) due to the highly variable local precipitation regime, existing isotopic composition data do not correspond sufficiently to local conditions. To justify the mixing process with O and H data we would need appropriate information about the long-term isotopic composition of precipitation in the area. In addition, long-term (at least two years), systematic isotopic measurements of the spring are required. Therefore, the dynamics of the spring and mixing processes cannot yet be explained based on the available data that were limited due to limited financial resources for this study and any additional conclusions about the mixing based on available O and H isotope data would be too speculative. However, the holistic approach used in this study enabled the confirmation of the influence of “deep (older)” water in the hydrological system of Żvepovnik spring.

5. Explanation concerning origin of sulfates is not fully convincing because the authors avoided discussion concerning possible sulfate reduction processes. The isotopic composition of sulfates enriched in heavy S and O isotopes can be formed also during dissimilatory sulfate reduction processes; one of the direct evidence of the sulfate reduction process can be formation of reduced inorganic sulfur compounds (as e.g. sulfide minerals FeS2) which were found by the authors in geological profile. For the full “picture” of possible origin of sulfates in studied waters the possible presence (or not) of the sulfate reduction process should be discussed.

Technical remarks:

a) all values of isotopic composition should have units vs international standard – please make adequate complements in tables.

Answer: Thank you for suggestion. We added in Table S3 ‰ vs. international standard: VPDB (for C), VSMOW-SLAP (for O, H), VCDT (for S in sulphate) and VSMOW (for O in sulphate) as suggested by referee. We also included the data about the standards in the text (Material and Methods). In addition, sentences are added in chapters 2.3.5 and 2.3.7.:

2.3. 5: The reference material (Carlo Erba solution) was used to convert the analytical results to the VPDB (Vienna Pee Dee Belemnite) scale.

2.3.7: Standardization of the method is based on several other sulfide and sulfate materials by different laboratories, "and international standards OGS-1 and NBS123 that were used to convert the analytical results to the VCDT (Vienna-Canyon Diablo Troilite) for sulphur in sulphate and VSMOW (Vienna Standard Mean Ocean Water) for oxygen in sulphate."

b) Line: 67 – probably should be “residence times” – not “resistant..”

Answer: Thank you for remark. The text was corrected

c) line 168 – 172: it is not clear --- what preciptitation of BaCO3? Please explain in more detail how water samples were prepared to isotopic composition of sulfates?

Answer: Thank you for this remark. We corrected the text of water sampling and added the procedure of water treatment.

d) line 390-392: the presence of H2S gas is not an evidence of mixing processes of some two different waters; the hydrogen sulfide is a gas and may migrate in geological environment independently of flowing waters in the system. Please explain this or change adequately.

Answer: Based on re-reading the text, we found that the statement could be ambiguous. Consequently, the sentence about mixing was deleted. We completely agree that based on the H₂S gas we can not explaining the mixing of the waters. This topic is discussed later in relation to isotopes.

e) caption to figure 5 should be changed because actually it explains that dissolution of barium is connected with the increase of SO42- and decrease of Ba which is not the case. In fact, this plot simply shows the comparison of some water samples with different concentrations of sulfates and Ba.

Answer: We agree with the comment. The figure caption was corrected.

Reviewer 2 Report

1. The main remark to this manuscript is a small factual material that can characterize the seasonal dynamics of the studied indicators. For example, determination of ³H, δ³⁴S_SO4, δ¹⁸O_SO4 (Table S3) and hydrogeochemical analysis (Table S2) of spring water were performed only once.
2. Please explain why, with an increase in the oxygen concentration in the spring water by 2 times (Table S1, date 12.12.2017), the redox potential (ORP) of this sample remained at the same level as for the other samples?
3. Under the indicated values of the ORP (Table S1) the low concentrations of manganese (at the same level as Rb and Sc) and iron (below Zn) in the spring water (Table S2) seem somewhat strange.
4. Line 390 and Line 422: Replace Тable S3 with Table S4.

Author Response

We thank the reviewer for valuable comments and suggestions that were carefully considered and taken into account. Answers to specific questions are listed below.

1. The main remark to this manuscript is a small factual material that can characterize the seasonal dynamics of the studied indicators. For example, determination of H, δS , δO (Table S3) and hydrogeochemical analysis (Table S2) of spring water were performed only once.

Answer: Thank you for remark. We are aware that based on the existing number of data (presented in this study), which is also the result of very limited financial resources, it cannot provide a comprehensive interpretation of the dynamics of the spring. This will require longer-term (at least two years of observation) measurements of all endmembers. Our goal was to present the basic characterization of the sulphur spring, which with its characteristic smell of H2S present a specialty in the karst world. However, it would certainly make sense in the future to carry out additional research that would contribute to the understanding of the source.

2. Please explain why, with an increase in the oxygen concentration in the spring water by 2 times (Table S1, date 12.12.2017), the redox potential (ORP) of this sample remained at the same level as for the other samples?

Answer: Thank you very much for noticing this issue, for which however we do not have a suitable answer. We agree that the ORP value should change if the O2 changes two times. We have used the same equipment for all readings, and only though is that the ORP readings are incorrect (possible probe problems). As expected, values of electroconductivity and TDS are lower on the day with much higher discharge, and oxygen values also change, so these values are measured correctly. We are however not sure for the ORP, so we decided to remove this column.

3. Under the indicated values of the ORP (Table S1) the low concentrations of manganese (at the same level as Rb and Sc) and iron (below Zn) in the spring water (Table S2) seem somewhat strange.

Answer: See comments above. We find the measured ORP values problematic.

4. Line 390 and Line 422: Replace Тable S3 with Table S4.

Answer: Thank you for the remark. The labels of tables in the txt were corrected.

Reviewer 3 Report

General comments

Multi-disciplinary research that I am pleased to recommend for publication in Water. Revise the manuscript addressing all the specific comments before publications

Specific comments

Abstract

Line 22-23. The abstract is well written. But, please revise this sentence without starting with “of” a new sentence

Line 24. Specify the type of fault. Normal or strike-slip?

Introduction

Line 45. Exchanges within the geosphere are more important in karst environments due to the higher degree of vertical connectivity of these systems with the atmosphere and deeper aquifer-units through faulting. Please, specify this point and insert relevant papers:

- Allshorn, S.J.L., Bottrell, S.H., West, L.J. and Odling, N.E., 2007. Rapid karstic bypass flow in the unsaturated zone of the Yorkshire chalk aquifer and implications for contaminant transport. Geological Society, London, Special Publications, 279(1), pp.111-122.

- Maurice, L., Farrant, A.R., Mathewson, E. and Atkinson, T., 2021. Karst hydrogeology of the Chalk and implications for groundwater protection. Geological Society, London, Special Publications, 517.

- Keim, D.M., West, L.J. and Odling, N.E., 2012. Convergent flow in unsaturated fractured chalk. Vadose Zone Journal, 11(4).

The sentence finds no link with the scenario of your manuscript and sounds chunky, that’s why I have inserted a comment

Line 62-65. Lack of recent literature on the link between structural geology and carbonate aquifers, below recent reviews on the topic that includes new conceptual models of fault hydrogeology

- Medici, G., Smeraglia, L., Torabi, A. and Botter, C., 2021. Review of modeling approaches to groundwater flow in deformed carbonate aquifers. Groundwater, 59(3), pp.334-351.

- Medici, G. and West, L.J., 2021. Groundwater flow velocities in karst aquifers; importance of spatial observation scale and hydraulic testing for contaminant transport prediction. Environmental Science and Pollution Research, 28(32), pp.43050-43063.

Materials and methods

Line 117. Stream graben? Maybe, stream located in correspondence of a graben

Line 122. Crack below the slope? I cannot understand this definition and description. Revise the language

Lines 129-139. Great geological mapping, I did this type of work last time 9 years ago! Specify the mapping area in km²

Line 151-154. What about alkalinity? This is a typical parameter that a multi-parameter probe measures in the field. The value can vary if you wait too much time before analyzing the sample in the laboratory

Lines 198-199. Insert USGS reference for PHREEQC. Parkhurst and Appelo 1999?

Results and discussion

Line 298. Looking at the map in Figure 1 also another structural orientation appears, WNW-ESW. Correct? Direction of thrusts? At least five years that I don’t visit Alps!

Lines 293-364. I suggest a better description of the pre-carbonate stratigraphy that becomes important later in the manuscript. Permo-Triassic sandstones and evaporites?

Line 141. “Hypothesis” is it a verb?

Line 400. Ratio 1.4 you say that is high. Can you state the typical ratio in limestones with no or little contribution of dolomite dissolution?

Lines 451-454. This part on solubility of dolomite and calcite is un-clear. Please, provide further explanations

Line 601. Specify the type of permeable fault when you describe the conceptual model in Figure 9, strike-slip?

Conclusions

Lines 611-612. Revise the language for the first two lines

Line 645. Please, do not start a new sentence with “it”

References

Add the five relevant research and review papers in karst hydrogeology suggested

Figures and tables

Figures 6. Insert the equation of the line with R² value close to 1?

Figures 7. Karstic and fractured aquifers? Do you refer only to carbonates, or all hard bedrocks? I suggest revising the name of the box

Author Response

We thank the reviewer for valuable comments and suggestions that were carefully considered and taken into account. Answers to specific questions are listed below.

Specific comments

Abstract

Line 22-23. The abstract is well written. But, please revise this sentence without starting with “of” a new sentence

Answer: Thank you for remark. The text was corrected

Line 24. Specify the type of fault. Normal or strike-slip?

Answer: Poor outcrops do not allow precise determination of the fault nature. So, in abstract we would avoid exact determination. In addition, On Slovenia area exhibits very complex history of tectonic deformations and reactivation of faults is very common. For example, normal faults can turn into strike-slip or reverse, etc. So, it is often impossible to define major deformation. Anyway, within manuscript, we added a small discussion on displacement.

Introduction

Line 45. Exchanges within the geosphere are more important in karst environments due to the higher degree of vertical connectivity of these systems with the atmosphere and deeper aquifer-units through faulting. Please, specify this point and insert relevant papers:

- Allshorn, S.J.L., Bottrell, S.H., West, L.J. and Odling, N.E., 2007. Rapid karstic bypass flow in the unsaturated zone of the Yorkshire chalk aquifer and implications for contaminant transport. Geological Society, London, Special Publications, 279(1), pp.111-122.

- Maurice, L., Farrant, A.R., Mathewson, E. and Atkinson, T., 2021. Karst hydrogeology of the Chalk and implications for groundwater protection. Geological Society, London, Special Publications, 517.

- Keim, D.M., West, L.J. and Odling, N.E., 2012. Convergent flow in unsaturated fractured chalk. Vadose Zone Journal, 11(4).

The sentence finds no link with the scenario of your manuscript and sounds chunky, that’s why I have inserted a comment

Line 62-65. Lack of recent literature on the link between structural geology and carbonate aquifers, below recent reviews on the topic that includes new conceptual models of fault hydrogeology

- Medici, G., Smeraglia, L., Torabi, A. and Botter, C., 2021. Review of modeling approaches to groundwater flow in deformed carbonate aquifers. Groundwater, 59(3), pp.334-351.

- Medici, G. and West, L.J., 2021. Groundwater flow velocities in karst aquifers; importance of spatial observation scale and hydraulic testing for contaminant transport prediction. Environmental Science and Pollution Research, 28(32), pp.43050-43063.

Answer: Thank you for your suggestions. The references were included in the text.

Materials and methods

Line 117. Stream graben? Maybe, stream located in correspondence of a graben

Answer: Thank you for remark. The name of the spring is Rajserjev graben. The statement was corrected.

Line 122. Crack below the slope? I cannot understand this definition and description. Revise the language

Answer: Thank you for remark. The text was corrected

Lines 129-139. Great geological mapping, I did this type of work last time 9 years ago! Specify the mapping area in km

Answer: Thank you for suggestion. The size of detailed mapping area was added.

Line 151-154. What about alkalinity? This is a typical parameter that a multi-parameter probe measures in the field. The value can vary if you wait too much time before analyzing the sample in the laboratory.

Answer: Thank you for remark. Total alkalinity after Gran was measured as soon as we got off the field to JSI (Jožef Stefan Institute) laboratory.

Lines 198-199. Insert USGS reference for PHREEQC. Parkhurst and Appelo 1999?

Answer: The reference was included in the text.

Results and discussion

Line 298. Looking at the map in Figure 1 also another structural orientation appears, WNW-ESW. Correct? Direction of thrusts? At least five years that I don’t visit Alps!

Answer: Thank you for remark. We added the text: in the existing text we added a direction of Periadritic fault zone and Sava fault. We also added a sentence in which we explain that these tectonic directions are recognized also in the Žvepovnik spring area (represented in Figure 2).

Lines 293-364. I suggest a better description of the pre-carbonate stratigraphy that becomes important later in the manuscript. Permo-Triassic sandstones and evaporites?

Answer: In the contrast to major part of Alps, in this part of Southern Alps a pre-carbonate stratigraphy is Middle Permian Val Gardena Fm clastic rocks that do not outcrop in close vicinity of the spring. Everything above (including Upper Permian and Lower-Middle Triassic rocks that are mentioned in the discussion) is dominated by carbonates and is generally described within the existing text. Any further detailed descriptions (on formation level) would seriously expand the existing text, but would not add hydrogeologically important data. By our believe, it is better to keep the existing basic stratigraphic description.

Line 141. “Hypothesis” is it a verb?

Answer: We agree, the sentence was corrected.

Line 400. Ratio 1.4 you say that is high. Can you state the typical ratio in limestones with no or little contribution of dolomite dissolution?

Answer: The Ca²⁺/ Mg²⁺ molar ratio typical for dolomite is 1. The values higher than 1 indicates the contribution/dissolution of calcite (limestone) and/or gypsum/anhydrite minerals.

Lines 451-454. This part on solubility of dolomite and calcite is unclear. Please, provide further explanations

Answer: Thank you for response. We added the following in the MS: “The problematic precipitation rates of dolomite compared to calcite are referred to as ‘the dolomite problem’ and described in mostly sedimentological references (a nice review was given by Warren 2000). In comparison to the vast amounts of dolomite beds in former geological periods, recent dolomites are forming in very restricted areas. The reasons for this are many, but most authors point out the slow reaction kinetics (Morrow 1982; Usdowski 1994). Consequently, dolomite can be much more oversaturated than the calcite before precipitating.”

Line 601. Specify the type of permeable fault when you describe the conceptual model in Figure 9, strike-slip?

Answer: Thank you for remark. There is nothing said about faults in the Line 601. We believe that Reviewer was thinking about Line 618. The nature of the fault is hard to define as described in the lines 362 to 364 (see also the answer to the comment of Line 24). We added a sentence (section 3.1) that further explains a possible movement of the faulted blocks and we stated that a normal character of a fault seams more probable.

We used this decision also for the Line 618 by adding the supposed character of the fault.

Anyway, in Karst system the hydrogeological characteristics of the same fault can change along the fault plane in vertical as well as horizontal direction (explained in Čar, J. (2018). Geostructural mapping of karstified limestones. Geologija, 61(2), 133–162). Unfortunately, in the specific Žvepovnik situation, outcrops do not provide data necessary for further specification of hydrogeological fault characteristics.    

Conclusions

Lines 611-612. Revise the language for the first two lines

Answer: Corrected after English language corrections.

Line 645. Please, do not start a new sentence with “it”

Answer: Corrected after English language corrections.

References

Add the five relevant research and review papers in karst hydrogeology suggested

Answer: The references were updated.

Figures and tables

Figures 6. Insert the equation of the line with R value close to 1?

Answer: Thank you for the question. The line presented in the figure 6 is the Global Meteoric Water Line, GMWL as defined by the Craig (1961) and described in the figure caption. Therefore, no R value is added.

Figures 7. Karstic and fractured aquifers? Do you refer only to carbonates, or all hard bedrocks? I suggest revising the name of the box

Answer: We refer to the karstic and fractured dolomite aquifers of Permian and Triassic age. We rephrased this to specify rock types: in the text box: "Dolomite karstic and fractured aquifers of Permian and Triassic age".

Round 2

Reviewer 1 Report

Dear Editor,
Dear Authors,

I am satisfied with the improvements of the manuscript made by the authors and also I am satisfied with authors' answer to may comments and remarks. The most important remarks were applied in the text; other were full explained to me.

I recommend the manusctipt to further publication process.

Author Response

Dear reviewer,

We are pleased to have successfully answered all your questions. Thank you for give us the opportunity to complete the text, which we believe is even better because of your comments.

Best regards!