Radiocarbon Dating and Stable Isotopes Content in the Assessment of Groundwater Recharge at Santiago Island, Republic of Cape Verde

Round 1

Reviewer 1 Report

The authors were courteous in responding to several requests for advice and revisions, and the revised paper is now very readable and of high scientific quality. I look forward to further research.

Author Response

Thank you for your comment.

English major revision was performed along the manuscript as suggested by Reviewer 3.

Reviewer 2 Report

Dear authors,

I'm satisfied from  the additions  related to my comments in your manucript  and generally from  the total revision. In my opinion it can be published in the present form in this journal.

Kind regards

Author Response

Thank you for your comment.

English major revision was performed along the manuscript as suggested by Reviewer 3.

Reviewer 3 Report

 

General comments

The manuscript presents some important grounwater data from Santiago Island – Republic of

Cape Verde. The data includes stable isotopes, radiocarbon and geochemical parameters and are used to understand the groundwater recharge, marine intrusion and estimatying residence times of groundwaters. The major weak points of the manuscript include very poor langauge, presentation, organization and unfocussed discussion throughout which make the manuscript very difficult to read and understand. The title and objectives are not matching, the discussion and interpretation are mostly out of context. The discussion/interpretation of radiocarbon and stable isotope data are very shallow. The authors should highlight what extra information they are providing in this manuscript that were not covered in Carreira et al. (2019), Advances in Science, Technology & Innovation. Springer, Cham. https://doi.org/10.1007/978-3-030-01572-5_40. The English langauage needs drastic improvement throughout the manuscript. I am sorry for my inability to recommend the manuscript for publication in its present form but suggets authors to restructure it with improved English and address all the issues raised here and resubmit.

 

Specific comments

 

Title: the title indicates that the study is related to assess groundwater recharge but objectives (line 105-110) indicate that the study is related to mineralization and residence time estimation. Please make it clear.

Some information of the groundwater level is required. Also mention the depths of the boreholes samples analysed in the present study

Line17-19: the authors started with a sentence stating pollution and degradation indicating that the study is realted to these topics but the actual contents of the manuscript are different, I suggest to change this sentence.

Line 80-87: please mention the total population of Republic of Cape Verde and Santiago Island

Line 267-272: these statements are very vague, please elaborate which location/samples have higher/lower EC and discuss the probable reasons.

Line 283-285: can you eaborate how marine aerosols can influence the groundwater chemical composition, if possible give some quantitative estimates of the marine aerosol formation, their transport pathways to the groundwater and extent of their influence on the groundwater chemistry. Also discuss the presence of marine aerosol in surface/rainwater as these waters ultimately infiltrate towards subsurface.

Section 4.2: the discussion is very shallow with some vague and traditional explanation. I suggest authors to discuss the isotopic data in details and make some quantitative estimates of percentage of water originated from precipitation and marine intrusion and what are the other processes such as evaporation that can modify the isotopic compositions. Also present the duterium excess data in a figure.

Figure 9: what is local trend? Is it Local Meteoric Water Line

Line 468-473: You have the age of these samples, can you discuss in details if these samples were really of LGM time and what are the expected isotopic ratios if these samples were infiltrated from the precipitation that occurred during LGM?

Line 492-512: the discussion is very general. Can you present the tritium data in a table with location of the samples, depth of the well, tritium content and estimated residence time? Can you show which samples were recharged before nuclear testing period and which after and is there any relation between residence times with depth or geography?

Equations 3 and 4: what are Co and C? Can you explain how change in d13C value from -23 per mil to -15 per mil changes the age from 0 to 3 – 5 kyr? I don’t have access of original paper of Salem et al., 1980, IAEA Vienna 165–179 article where this procedure is probably described.

 

Author Response

Please check the pdf file attached (WATER 1787735_Reviewer#3_reply.pdf)

Author Response File: Author Response.pdf

Round 2

Reviewer 3 Report

 

The manuscript is still very poor in English and presentation though authors implemented many of the suggestions made in the previous version. As per authors response, it is the third round of review and they are not ready to improve its presentation but I reviewed it first time and I pointed out whatever concerns I had. Not much improvement is noticed in the revised manuscript in terms of presentation and some important issues are overlooked. I regret for my inability to recommend it for publication unless authors seriously address the following issues:

1. Improve the presentation and English language throughout the manuscript

2. Some groundwater samples are depleted in 18O compared to the end-member assumed here, if the enriched samples are due to evaporation, what could be the reasons for lower d18O values? Please describe the mixing equations (with EC and d18O) that are used to estimate the fraction of seawater (Table 1) in the method section.

3. A quantitative explanation with actual calculation should be given showing how change in the carbon isotopic ratio from -23 per mil (C3) to -15 per mil (C4) vegetation changes the radiocarbon ages by 3 to 5 thousand years.

 

Author Response

Please check the pdf file

Author Response File: Author Response.pdf

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 discusses the behavior of groundwater on the island using radioactive and stable isotopes. Elucidating the origin of groundwater salinity and preventing it from rising is an important component of groundwater resource protection and management. This paper is well worth publishing, but I recommend that it be revised to take into account the following several points

 

Fig. 2

The geologic map is important for considering groundwater flow, and although several layers are color-coded in Fig. 2, the age and rock type of individual layers in each unit should be noted.

 

Table of analytical values

The table of analytical values should be included in the paper. Figures discussing chemical composition, such as Fig. 4, show that there is a lot of variability in the data. Whether this variation is true or due to poor analytical precision should be discussed by properly examining the charge balance.

The charge balance is expressed as Err% = ((Scat - San)/(Scat + San)) × 100, where the symbols Scat and San mean total cations and anions, respectively (Appelo and Postma, 1993). Charge balance is known to affect the total amount of dissolved constituents in groundwater (e.g., Willams et al., 2009), and it is important that groundwater, for example, with a total volume of 2 meql/L or less, be within Err% <10%. Analytical values that deviate from this should be excluded from the figure and discussion.

Ref.

Appelo, C.A.J., Postma, D. 1993. “Geochemistry, groundwater and pollution”. ed. by Appelo, C.A.J. and Postma, D., Rotterdam, Brookfield: Balkema. xvi + 536 pp.

Williams, A. J., Andersen, C. B., Lewis, G. P., 2009. Evaluating the effects of sample processing treatments on alkalinity measurements. J. Hydrol., 377, 455–464. doi.org/10.1016/j.jhydrol.2009.　09.007

 

Fig. 4a, b

It is unclear why only halite is discussed in Fig. 4. In addition to the dissolution line of halite minerals, the mixing line with seawater and meteoric water (Na/Cl=0.858) should be included in the Fig/.4 and discussion. The authors discuss seawater-meteoric water mixing in L349.

 

Fig 4c, d

It is commonly observed that dissolution of carbonate minerals causes Ca and HCO3 ions to rise in groundwater, but the molar ratio of Ca to HCO3 dissolved into groundwater varies depending on whether CO2 from the atmosphere is involved in the reaction, as shown in the reaction equation below. This should be discussed before discussing ion exchange.

CaCO₃ + H₂O = Ca²⁺ + HCO₃⁻ + OH⁻   (Ca²⁺/HCO3^- = 2 eq ratio)  (1)

CaCO₃ + H₂O + CO₂ = Ca²⁺ + 2HCO₃⁻  (Ca²⁺/HCO3^- = 1 eq ratio)  (2)

The carbonate dissolution line in Fig. 4 has an equivalence ratio of 1:1 and shows the reaction in Eq. (2). The equivalent ratio of 2:1 in eq. (1) should also be shown to indicate that the observed values are lower for Ca than the dissolution of the carbonate minerals. This is likely the result of ion-exchange reactions between the rock and Ca, as explained by the authors.

 

Fig.7

Figure 7 is very important in distinguishing water systems, but has not been discussed enough. The reader can determine that there is a δD-δ18O trend between the eastern region and the rest of the country, but we cannot identify which region is the eastern region. Figure 1a should be framed with regions that are considered eastern regions so that the reader can easily understand which regions are eastern regions.

 

L341

There is a large gap in the discussion between L340 and L341: up to L340, the isotopic composition indicates that the groundwater in the eastern region is derived from old age meteoric water, so the chemical composition from L341 is also discussed by distinguishing the eastern and other regions in the figure. If there is no difference, the overall salt concentration is better to discuss the mechanism.

 

 

 

 

 

Author Response

1. As suggested the rock type of the individual layers in each unit were added to the text.
2. As suggested a table with the analytical chemical data is now included in the manuscript, including the charge balance. Considering the errors associated to the chemical determinations (<10%) the authors followed the reviewer suggestion, and all points were considered in the graphical representations. A new sentence with the charge balance error was included in the manuscript and a new sentence with the results discussion were also added.

In most of the samples, we found a decrease in the anions sum that could be related with calcite precipitation thereby decreasing the HCO₃ concentration. It should be also mentioned that the Island is dominated by volcanic rocks contributing with silica that which is not considered in the imbalance calculations.

 

3. Figure 4 was replaced by a new one, as suggested, with the two carbonate dissolution (1:1) and (2:1) lines. In the Na vs Cl diagrams, the seawater-fresh water mixing line was included. New sentences were added in the discussion of Fig 4.

 

4. Fig 7 (now Figure 8) was replaced by a new one with the eastern group marked in the island map. A discussion concerning the isotopic composition variation and the possibility of seawater intrusion mechanism was added into the manuscript.

Reviewer 2 Report

Dear authors,

In my opinion your work is interesting and suitable for publication in this journal. Nevertheless some minor comments are given below to be  taken into account in this stage:

• Please give your photos in better resolution  in order to increase the readability.
• I think that some geochemical- mineralogical or petrographic features of related rocks  could be indicated in this paper.   
• Please rewrite your conclusions in a more briefly text or use bullets. 

Kind regards

 

Author Response

- The figures are with good quality / resolution, however in the PDF that is generated at the time of submission, the figures may appear to have poor resolution, but in the original submitted files have good quality.

- The geochemical-mineralogical features of the geological formations were added to the manuscript, as suggested.

- The conclusions were condensed as suggested.

Reviewer 3 Report

The authors present (graphically – no data table was provided) a large set of geochemical and isotope data for groundwater on San Antonio Island, Rep. of Cape Verde, and offer interpretations of the sources of water and salinity, and of water age. In particular, the roles of seawater intrusion and atmospheric deposition of sea-salt aerosol are considered.  My overall impression is that the interpretation of major ion data is better than that of the isotope data, although I have some suggestions for the former.    The interpretation of the isotope data must be greatly improved.  This will involve background reading, and is not likely to be accomplished with some cursory changes within the journal’s usual review response deadline.  My recommendation is therefore to reject for the present, but with encouragement to resubmit because I consider the data to be very interesting.

The English needs attention to grammar and idiom.  In certain cases, the meaning of the text is unclear because of non-standard idiom.

A general comment:     It would be a good idea, if possible, to distinguish samples from aquifers above sea level (and therefore beyond the reach of direct seawater intrusion) from samples drawn from below sea level.  Spring samples are clearly from aquifers above sea level.  In the case of wells, detailed knowledge of well depths is required.  Use of the base and middle aquifer classifications sheds little light on interpretation of several of the diagrams, and it seems a better idea to distinguish samples according to measured aquifer altitude.  This may provide a way of distinguishing mixing of fresh water and seawater from addition of marine salt in aerosol.

I hope that the following detailed comments, arranged by line or figure number, will be of assistance.

Figures

1. This (and Fig. 2) appear to have been drafted for other purposes. The green text in panel (a) is not relevant to the manuscript, and should be removed. In panel (b), make sure that all geographic names used in the text are shown, and eliminate any that are not needed.  This figure and other maps need larger font for the sample locations.
2. There is no point in showing a detailed geological legend without explanations of each unit. To make things easy for the reader, please draft this map with symbols only for the major units (with the possible exception of the conglomerate-breccia members of the AIEC).  Ages should be given if available, and the relationship between the major units and the aquifers named in the text.  Have the units been formally identified as geological formations?
3. and elsewhere: Designation of the aquifers as “base”, “middle” and then “recent” seems strange – shouldn’t “recent”  be “upper”  to complete the series?   See the general comment above.  All figures showing geochemical data should show ratios or concentrations (as appropriate) for seawater.
4. Please correct the spelling of “adsorption.”

Lines

110-115: This climatic information should appear in only one place (compare lines 75-76)

119-124:  The meaning is unclear.

132 onwards:  An acronym PA was introduced at line 130, and should be used thereafter, rather than repeating the name and acronym each time.  The same should apply for other acronyms.   Unless the PA + A acronym is fully explained, just PA would seem to be sufficient.

137:  What does “relatively high” mean?  Relative to what?

139-140:  Possibly clearer:  Rapid discharge of groundwater from these high-transmissivity aquifers is compensated by rapid recharge.

142-144: Flamengo and Conglomerate-breccia units (formations, members?) are shown as distinct from the AIEC in Fig. 2.

161: Does this mean that 13% of precipitation infiltrates the aquifer rather than the soil?  Water infiltrating soil can be lost to evapotranspiration.

165-166: How do these intermediate and deep zones relate to the base, middle etc. aquifers discussed elsewhere.  It will help the reader to have one consistent set of terms.

170:  Does this really mean salt deposits, by which I would understand halite beds, or is it intended to mean saline groundwater?  If halite, this should be described under Geology.

172-173:  Does this mean a permanent spatial depression in the interface beneath a freshwater lens, or an interface that has risen over time?

185-186: Use of the word “previous” is a problem here.  It leaves the relationship of the institutions unclear.

187-188:  When citing analytical precisions, here and below, say whether they represent 1σ  or 2σ.

190-192: Rather than name the authors, state explicitly what the method was, e.g. CO₂ was equilibrated with water at a temperature of…,    and so on.

195-196:  For tritium, give a detection limit, which is 0 + 2σ where σ is the analytical precision for low-tritium samples.  The statement “The error associated…..” is unclear.

197:  Say ….where 1 TU represents to a ratio of…..

198-200:  δ¹³C would not have been measured by AMS.  pMC is the customary abbreviation.

203-204:  Give methods for the major ion analyses, and analytical precisions.

212 onwards:  The temperatures are not based on the classification.  There is no statistical difference between temperature ranges in the three aquifers, so one range is sufficient.  There are too many figures – can temperatures really be measured reliably to 0.01 °C?

221-222: Where the extreme pH values occur may give some insight as to processes affecting major ion chemistry.  Simply say “for most samples.”

226-227: It would be more accurate to say that there is no distinction between the base and middle groups.  The recent samples are few in number, but it is possible that they are different from the others.

229-230:  The Piper diagram cannot show which samples are more concentrated.  It only shows ion ratios. 

230-231:  A better way of showing whether marine salt addition has occurred, and how it relates to salinity, might be to plot Na/Cl  and Cl/SO₄ vs. TDS (or Cl).  Cl/SO₄ is very distinctive for seawater, especially in a system where an alternative source of solutes is volcanic rock with leachable S >> Cl.  Incidentally, what is the S content of the carbonatites?  Note that Na/Cl (molar) is not 1 in seawater.

250-255: The carbonatites need to be mentioned under Geology.  How much carbonatite  is present, by volume?  Has its δ¹³C been measured?  Is the occurrence of carbonatite relevant to interpretation of ¹⁴C? While on the subject of carbonate sources, does soil carbonate (calcrete, caliche) form in the study area?  If so, what it its relevance to the interpretation of ¹⁴C?

260-1:  The two trends in the figure are addition of Ca at ~constant HCO₃,  and addition of HCO₃ with  ~constant or decreasing Ca.  How can these be explained?  Note, however, that two samples on the vertical trend plot close to the Ca/C (molar) ratio of seawater, which is ~5.

275-276:  Give references for ion exchange associates with seawater intrusion.

293 onwards:  This is discussion.  Results should come first.

315-317:  Give appropriate significant figures.  If the IAEA data are important, show them, or their field, in Fig. 7.  Where does the LMWL come from?  It appears to be a regression line fitted to some of the groundwater data, and this is not an LMWL because LMWLs must be estimated from measurements on precipitation only.   In Fig. 7, there are many points plotting to the right of the GMWL.  Therefore, isotope effects of evaporation may  be present. In the figure, the reader cannot tell which points are the basis of the Eastern Group trend line.  There is no way for the reader to substantiate the statement about isotope altitude effects.  Are such effects likely on the island?  What is the altitude range, and what are the likely corresponding shifts in δ¹⁸O?  Looking at this another way,  it is very unlikely that much recharge of low-elevation precipitation occurs, because low-elevation parts of the study area have very little rain.  If this is so, then altitude isotope effects will be diminished.

317-318:  These are not useful statistics because of the number of processes that may be modifying the isotopes (precipitation, evaporation and seawater intrusion).   Where does FT-109 plot on this diagram?

Some general comments on the interpretation of Fig. 7:  Please look at some suggestions for this figure in the file I’m sending separately.  I took a few data for regional seawater from the reference given at the end of these comments; the authors may be able to find more by increasing the depth range.  When these data are plotted, it becomes clear that seawater intrusion affects most of the samples (but cannot be distinguished from potential evaporation).   The Eastern Group trend, if it truly is a linear trend, must be a mixing trend with seawater as one end-member.  The  low- δ¹⁸O end member of the trend cannot simply be paleoprecipitation, because it plots far from the GMWL.   Is it possible that the low- δ¹⁸O end member is in fact an evaporated water, which would have had a pre-evaporation δ¹⁸O value of about -8.5‰, assuming an evaporation trend slope of 5?  Perhaps that could be paleoprecipitation.  If that is the case, the implications for water supply may be important, if the freshwater lens in one part of the island  preserves paleoprecipitation that has not been replaced by younger recharge.  The area in which the Eastern Group was sampled should be shown on Fig. 2.

I can’t see how Fig. 8 can help in identifying aerosol contributions in its present form, because all of the data appear to be included, even the samples with effects of seawater intrusion (Fig. 7).  Perhaps this could be useful  if only the samples plotting on or above the GMWL were used, or samples from high enough on the island to avoid seawater mixing.  It would be useful to know from what altitude each water sample came –  according to well depth, not to the altitude of the well collar.

364:  Instead of 0±0.6 TU, put <0.6 TU, which doesn’t imply negative tritium.

371:  Bamako

377:  The samples with tritium below detection in 2006 did not have residence times of 30-40 years in 2006.  Let us assume that average tritium in rain without bomb influence has been 3 TU prior to and since the bomb spike. Recharge from 1992 (when bomb  tritium had rained out in North America, and therefore probably globally)  would have contained ~ 1.5 TU by 2006, as would recharge from 1970 to 1992 (see Eastoe et al., 2012, for a discussion).   Recharge between the 1950s and 1970 had bomb-spike (>>3 TU) tritium and would have contained >1.5 TU in 2006.  Pre-1953 recharge with 3 TU now has tritium below detection.  Therefore, recharge with tritium below detection must be pre-bomb in age.  These calculations assume no mixing. 

408-409:  This discussion of the diagrams from Han et al.  needs an explicit statement about how the graphs are used.  For instance, do samples for which correction is possible plot along all of the diagonal lines, or in all of the smaller rectangles?  The authors do not seem to have used the Han diagrams in any case, because they have proceeded to correct all of the dates.

431-433:  FBE-201 is not on line Z in the diagram.  How do we know δ¹³C of soil gas in the study area?

443-444: Is fermentation, requiring anoxia, likely in high-transmissivity aquifers?

General comment about radiocarbon dating:  Too much text is devoted to this.  The procedure is of doubtful value in obtaining residence times.  Even if two-component mixing had occurred, we know too little about soil gas and rock carbon to make plausible corrections.  It might be possible to avoid the influence of agricultural C₃ plants by limiting consideration to the two samples on the mountain slopes, rather than on the coast (is there any agriculture in the watersheds of these samples?).  These may also be water from high above the influence of seawater intrusion.   Samples from near the coast are likely to be mixed with seawater (as indicated by O and H isotopes).  In the recent past, the pMC in seawater appears to have been about 6% less than the pMC in coexisting atmosphere (reference, below).    This mixing situation is too complex for the usual correction methods, which assume two-component mixing in the carbon.

The apparent ages assuming all C₄ carbon in soil gas (Table 1) appear to be incorrect, because they are greater than the uncorrected ages. Therefore Fig. 11 is of no use in its present form.

 

Useful references:

Eastoe, C.J., Watts, C.J., Ploughe, M.  and Wright, W.E., 2012, Future use of tritium in mapping pre-bomb groundwater volumes.  Ground Water,   50: 87-93.

António M Monge Soares1 • José M Matos Martins1,2 • João Luís Cardoso, 2011 MARINE RADIOCARBON RESERVOIR EFFECT OF COASTAL WATERS OFF CAPE VERDE ARCHIPELAGO RADIOCARBON, Vol 53, Nr 2, 2011, p 289–296

Seawater O,H isotopes from:  Data.GISS: Global Seawater Oxygen-18 Database (nasa.gov)

#	Lon.	Lat.	Depth.	θ	Sal.	δ18O	δD	Year	Month	Reference
1	-20.93	18.55	10	19.95	36.37	0.82	**	1991	2	(H+0.14) Pierre et al (1994)
2	-21.22	18.42	10	20.98	36.29	0.88	**	1991	2	(H+0.14) Pierre et al (1994)
3	-20.52	10.98	3	24.11	35.78	0.94	6.1	1973	3	GEOSECS Ostlund et al (1987)

 Note that the longitudes are negative.

Comments for author File: Comments.pdf

Author Response

Reviewer #3

General comment:

- With the available data, it is only possible to distinguish the groundwater samples from aquifers situated above the sea level from those that are situated in the most interior part of Santiago Island at higher altitude, and these points do not present high levels of mineralization.

 

- The total number of spring sampled in Santiago Island was 16; from those only 3 have electrical conductivity values higher than 1000 µS/cm (see Table 1 added to the manuscript). From these 3 springs the highest electrical conductivity measured was 2750 µS/cm. Considering our available parameters (isotopes and chemical data) and with these values of mineralization is impossible to identify the main source responsible for the groundwater mineralization, although due to their location most probably seawater intrusion mechanism can be present.

 

- The reviewer suggestion of use the wells depth to distinguish mixing seawater-fresh water would be most useful tool, but unfortunately the authors do not have this type of information. Most of the wells are private and this information is missing.

Figures

1. The suggestion to remove all the internal captions of Figure 2, however this suggestion was not followed, since new text concerning the geological formations was added by recommendation of #1.

 

2. The age of the formations were added to the manuscript.
3. The separation in 3 units was adopted since 1984 when BURGEAP - Engineering of Environment worked in Santiago Island. This company performed several hydrogeological studies in Cape Verde, and based on the volcanic sequence established by Serralheiro (1976) defined three hydrogeological units called Base Unit, Intermediate Unit and Recently Unit, and since then has been used in hydrological studies.

 

4. the word “adsorption” was corrected in Figure 5 (now figure 6).

 

 

Lines

110-115: lines 75-76 – were deleted and part of the text was move to the “section” 2.1 Climatology (lines 110-115) to avoid repetitions.

119-124 –the sentence was modified by: “As previous mentioned the rainfall over Cape Verde – Santiago Island is concentrated in 3 to 4 months and, is responsible for more than 75 % of the total precipitation in the island. Although, the Intercontinental Convergence Zone plays an important role in the atmospheric dynamic of the tropics and in the rainfall over West Africa [33]. According to these authors, the rainfall is mostly related with transient convective disturbances associated with frontal systems that cross the Atlantic and not with the Intercontinental Convergence Zone or from squall lines from the North Africa.”

132: the PA was deleted since is used onwards.

137: the sentence was modified by: “…the storage coefficient of Pico da Antónia and Eruptive Complex is relatively high when compared with the other hydrogeological formations in the island …”.

139-140: the sentence was modified as suggested.

142-144:  they correspond to different units with different geological ages. New text was added to the manuscript.

161: Even from the infiltrated water in the soils (13%), part of this can be lost by evapotranspiration, but up today from the water that infiltrates no quantification of this loses were performed.

165-166: these two zone were proposed by Gonçalves et al. (2018), and are based on the depth and mineralization of the groundwater using geophysical cross sections using electromagnetic methods (Transient ElectroMagnetic (TEM); Very Low Frequency (VLF) profiles and Magnetoteluric (MT) to confirm the TEM cross check data. The maximum depth reached in this research was - 100m.

170: The term “salt deposits” is used by Gonçalves et al. (2018) during geophysical survey at Santiago Island. These authors identified an increase of mineralization “about -50 m mls ….”, which they did not attribute to seawater intrusion but to “salt deposits” in Tarrafal area. The geological studies performed in the island no not mention outcrops of this type of “formations”.

172-173: The geophysical studies were performed in 2009 and published in 2018. Considering the agriculture and population increase in Santiago Island the depression in the interface is expected to growth.

185-186: the word “previous” was replace by “formerly designated by”.

187-186: the errors associated to the O-18, 2H, 3H and 13C, measurements is 1s. This information was added to the manuscript.

190-192: This type of methodology was not included in the manuscript, since it can be found in the references concerning the methods used in the laboratory. To mention that the CO2 is of mineral origin was equilibrated during 5 hours with 8 mm of water for example is not the aim. The authors believe that this type of information will not improve the manuscript, which is focused on hydrogeological research using chemical and isotopic parameters.

195-196: the sentence was modified.

197: the sentence was modified as suggested.

198-200: the sentence was modified as suggested.

203-204:  new sentence were added. As suggested with the methods. No analytical precisions were included due to missing information.

212: the values were change to 0.1 ^oC.

221-222: suggestion was accepted and the sentence change.

226-227: the suggestion was accepted and the sentence was modified.

229-230: the authors completely agree, from the piper diagram two samples with different chemical content but with the same ions in the same proportion can be placed in the same location when using the Piper Diagram. Through the Piper diagram is impossible to identify the samples mineralization.

230-231: A new figure was added to the manuscript (Figure 5) and a sentence was added.(see the figure attched  Figure 5NEW-Carreira etal.tif)

 

 

250-255: the percentage /volume of carbonatites formation in the island is not mention by Mata et al. (2017) and Martins et al. (2010), they only mentioned the number of samples that were analyzed. Concerning the isotopic determinations ¹³C in carbonatites, they were not carried out. The carbonates were mention in the manuscript, as a hypothesis to support the increase of calcium in the groundwater samples, and never used not used as a correction in the groundwater dating by radiocarbon.

260-261: New sentences were added to the manuscript in this section (carbonates dissolution) and new Figura 4 was included.

275-276: references were added to the manuscript:

[45] – Appelo CAJ; Postma D (1994) – Geochemistry, groundwater and pollution. Ed. Appelo, C.A.J. and Postma , D. Rotterdam, Netherland, Balkema

[51] - Pennisi, M.; Bianchini, G.; Muti, A.; Kloppmann, W.; Gonfiantini, R. (2006) Behaviour of boron and strontium isotopes in ground-water-aquifer interactions in Cornia Plain (Tuscany, Italy). Appl Geochem 21, 1169-1183]

293: This section starts with a brief introduction before start to present and discuss the results obtained.

315-317: A new Figure7 was place in the manuscript with the mean isotopic value of the IAEA dataset as suggested and the new L-MWL equation. The Local-Meteoric Water Line, the equation was calculated using the available isotopic composition of the groundwater samples with an electrical conductivity lower than 500 µS/cm, in order to reject the samples that could have influence of seawater mixture. This approach was used due to the absence of precipitation data “series”.

317-318: The sample FT-109 do not present any isotopic deviation δ¹⁸O = -4.05 ‰ and δ²H = -24.2 ‰. The sentence was not included to have a statistics approach but to help the reader to follow the author’s ideas.

General comments: the hypothesis presented by the reviewer is probable true; palaeowaters “should be “plotted along the GMWL and not in a parallel trend line or even parallel to the Local MWL. However, is peculiar that we found only this trend in the Eastside of the island. The authors consider that this unusual trend could be linked to palaeo-precipitation. Future hydrogeological works are need to confirm the results. Most of the groundwater samples are plotted

 

364: suggestion was accepted and the sentence change

371: Bamako was added.

377: The sentence was modified in order follow the reviewer comments. The lines that are delineated in the diagrams starts from 0.6 TU, which is the detection limit of the liquid scintillation counter, used in the groundwater measurements.

408-409: the carbon-14 content plotted in the “Han diagrams” was not corrected is the values in pMC obtained from the direct measurements in the total dissolved inorganic carbon. The author plot the carbon-14 and carbon-13 content without any correction and used the excel file provided by the IAEA – Isotope Hydrology Section.

431-433:  the sample plotted near the Z line is 59-24 not FBE-201. No isotopic determinations of the d13C of the soil were carried out in the region.

443-444: the hypothesis of methanogenesis processes involving organic matter is pointing as more reliable.

General comment on 14C:

TheC4 and C3 plants were considered to test which group of plants could be dominate in the carbon source in the groundwater samples. Based on the isotopic composition content the presence of old groundwater cannot be excluded, so the authors used the two values of d13C to check the most feasible water age. According to the work of Soares et al, 2011, the carbon13 measured in goat bones at S. Vicente Island Barlavento (windward) group and Santiago Island is part of Sotavento (leeward) group.  Part of the results obtained by Soares et al.2011, indicate that the  “d13C of the bone samples show an enrichment that may result from the animal (goat) feeding habits, essentially plants, in this case tropical plants (C4 plants) having a d13C content between –17 and –9‰”. This information was added to the manuscript. Regarding the possibility of these samples could represent a mixture with seawater, 5 of them are located near the coast at altitudes between 1 and 20 m, only the sample 59-24 is situated at higher altitudes. The possibility of mixing freshwater-seawater can occur, however, looking at the EC values we find that they change from 1364.5 µS/cm (FT-39) up to 2940 µS/cm (FT-78).

Author Response File: Author Response.docx

Round 2

Reviewer 3 Report

I have made a fairly rapid evaluation of the paper, mainly by looking at the diagrams and associated text.  I see little response to the important points I made in the first round of the review.   Details are given below.  The English still needs work, but it may e possible for the journal editors to tackle this.

Major points

1. The authors have added diagrams of Cl/SO4 and Na/Cl vs Cl. They do not indicate the units of the ratio scale (Cl/SO4 for seawater would be about 7 by weight, and about 20 by mols; I suggest not using meq here). The diagrams cannot be interpreted by the reader without such information.  The seawater-freshwater mixing line is problematic if it is based on the chemistry of the waters with lowest TDS, because such waters may also derive their salts from seawater and have the same Cl/SO4 ratio as seawater.     I suggest leaving the outlier samples off the diagram,  plotting a horizontal line corresponding the Cl/SO4  for seawater, and focusing on the non-outlying samples.  How many of them have Cl/SO4 or NaCl approaching the ratio of seawater, and is there a relationship with increasing Cl concentration?  The diagrams suggest that many dilute samples are likely have seawater ratios. 

How is it possible to generate Cl/SO4  ratios much higher than those of seawater?  One extreme case is FT-78. Two likely sources of solute are seawater and minerals dissolved from volcanic rock containing much more S than Cl.  Are these extreme data valid?

This brings up the mention of the “salt deposits” at line 205.  A reader like myself is left wondering how there could be evaporites on a volcanic island – but if there are, it needs to be stated explicitly.  Or were the salts  deposited from steam?  This is an important consideration for explaining  Cl/SO4 ratios higher than those in seawater.

Similar reasoning should be applied to the Na/Cl ratio plots.

2. The interpretation of the O-H isotope diagram still lacks adequate understanding. If a researcher is looking for seawater mixing, it would seem advisable to plot seawater on the diagram. A single seawater-freshwater mixing trend is not applicable – the data suggest multiple trends.  It would also be advisable to distinguish the Eastern Group samples with separate symbols so that the reader can see which data the trend line is based on.  The interpretation of Eastern Group samples as paleoprecipitation is oversimplified.  If seawater is plotted on the diagram, it becomes clear that the linear trend in question is a mixing trend.  One end-member is seawater, and the other an evaporated water that is more difficult to explain.  If it is paleoprecipitation (which is indeed possible), where would it have plotted on the GMWL prior to evaporation, and why would it be more evaporated than modern precipitation?

I repeat that the LMWL as shown is not a local meteoric water line.  It is clearly influenced by mixing with seawater.  A LMWL must be based on measurements of precipitation.  I don’t think it is necessary to have a LMWL on this diagram, but it would be interesting to distinguish the set of points with low EC and show that even they are influenced by seawater. 

3. A major suggestion in the first review was to use elevation rather than the present basal, middle units etc. as a basis for classifying the samples. The authors’ reason for not doing this is invalid.  The set of spring samples is adequate  for the task if ion ratios are used; whether they have high or low solute content is irrelevant.  What we are interested in determining is whether marine aerosol is present in spring waters of any solute concentration.

The present classification of samples as basal, middle units etc. is an unnecessary distraction in the paper  because it shows nothing.  It would be far more interesting to distinguish the samples as springs or wells.  The results should be interpretable, even if they turn out to be null.

 

Minor points:

Fig. 1 still contains irrelevant labelling and features.  What is the diagonal line in Fig. 1b? The font size in Fig. 1b is likely to be illegible in the formatted paper.

Fig. 2 still has unlabeled symbols.

If these two diagrams are copied from other publications without change, there will probably be copyright issues.

Figs. 4 and 5:  The symbol colors in the captions do not match the figures.

Fig. 9 shows little of interest.  It is unlikely that trends would be visible for all of the data plotted together, because there are different isotope trends in Fig. 8.  This might work if just the Eastern Group were plotted in Fig. 9.

Line 264:  My suggestion that there is no statistical distinction between the temperature ranges stands.  Therefore there should be no statement about a temperature trend among the groups of samples.

Line 329:  I cannot see two trend lines of data in this figure.

Line 330-1:  A surplus of Na seems more likely to correspond to seawater intrusion.  Or was it intended to say something like  “high ratios of Na/Cl”?

At this stage, I cannot recommend accepting the paper, even with major revisions.  I suggest that the authors go back to the comments from the first round of reviewing and do a thorough job of understanding and responding to them.