A Borehole-Based Approach for Seawater Intrusion in Heterogeneous Coastal Aquifers, Eastern Part of Jeju Island, Korea
Round 1
Reviewer 1 Report
The paper deals with the characterization of the shape of the salt water wedge in the volcanic aquifer of the Jeju island Korea, mostly from temperature monitoring, in a set of boreholes.
The topic of the paper is of interest, is rather original, and is worth to be published, provided some improvements that are detailed below. I then recommend a major revision of the paper.
This paper integrates in the frame of the current researches dealing with the characterization of island basaltic (volcanic) aquifers. These researches were numerous during the last decades, on various volcanic islands (Hawaii, Canary, Reunion, Azores, Jeju…). But this new paper deals with a higher resolution approach in such aquifers that is needed both (i) to better understand the structure and functioning of such aquifers, and (ii) to propose appropriate management schemes; this is a quite recent research trend, that gets from the scale of the island to the one of the local aquifers (see notably Lachassagne et al., 2014 who justify the need for high-resolution research in volcanic aquifers). In the Introduction, it would be worth to also replace this new paper in this research frame.
I have a first (minor) concern about a few vocabulary issues:
-> the term “basalt aquifer”, or “basalt” that is used in several places in the paper is too vague to my opinion. I think that it must be replaced at least with two different other terms:
when it is used to generally describe aquifers from basaltic islands (for instance in the abstract, line 8), with something like “basaltic aquifers”; when it is used to describe some structures of basaltic aquifers, the description must be much more precise. For instance, at line 11 in the abstract: “to the unique geological features of basalts” (with a “s”). In some other occurrences (for instance at lines 48-49), it must be much more precise to explain that basalts are mostly composed of basaltic lava flows, that comprise pervious layers (mostly clinker layers in between 2 lava flows and cooling fractures in the massive part of the lava flow), and may also comprise impervious layers in the unfractured (cooling fractures) massive part of the lava flows (see notably Lachassagne et al., 2014 who describe that in details). It may also be important to explain if, in the studied area, the lava flows are thick orf are a succession of thin (a few meters) lava flows. This could be briefly addressed in the introduction (line 42; current description is too vague), and with more details in section 2. (after line 109) and/or in the results (section 3.) when the borehole geological logs are described. A few months ago, I saw some cores from Jeju, and in the cored area, the lava flows were very thick.
Then, each occurrence of the “basalt” term in the paper should be checked and revised if required.
-> I have a similar concern with the term “seawater intrusion”, but to a much lesser extent. This term is largely used in the literature (see for instance the references of the paper), but often indiscriminately to describe the “natural” contact/equilibrium between fresh and seawater in aquifers or to describe the impact of pumping in an aquifer that provokes a dynamic intrusion of seawater in the aquifer. The authors may briefly explain (in the introduction) that they deal with the characterization of the fresh/seawater interface whatever it is natural of provoked by pumping. They may also replace the term “seawater intrusion” by another term that is more descriptive such as “fresh/seawater interface”, or “salt water wedge” or any other appropriate term. However, this proposal for changing this term is only a proposal from myself as, as said above, several authors are using this “seawater intrusion” vocabulary.
Additionally, highly weathered or “metamorphosed” (clogged with zeolites for instance) basalts such as the old basalts from the Deccan traps (line 55), or from the ocean floor, must be distinguished from those of recent volcanic islands like Jeju as the first ones are mostly of very low permeability, and the second ones are highly permeable. This is also explained in Lachassagne et al. (2014) and other papers of course (this paper from Lachassagne et al. also provides an extensive review).
Then, the introduction must be significantly revised to provide a more comprehensive and precise overview of basaltic aquifers and to well introduce the framework of the current research in Jeju.
My main technical concern about the paper, that justifies, to my opinion, a major revision, deals with the need for more details on the descriptions, and notably on the geological and technical logs of the studied wells to better support the interpretations.
In section 2.1, it would be necessary to refer to Figure 3 to explain where the sensors (EC and T°) are located in the well. From table 1, if I understand it well (see my remark below in the “minor comments”), I can post the sensors on the geological logs from Fig 3 (elevations in m amsl). This should be done on Fig. 3. In fact, temporal changes in EC may depend a lot from the location of the sensor, and from the piezometric level changes.
In section 3, first of all, it must be explained if the EC/T loggings where performed in wells that were pumped during the logging or not (section 3.1, line 234 and following). I understand that they were not pumped. In fact, in unpumped wells, local processes in the well can strongly impact the EC curve, such as vertical “ambient” flow in the well between various aquifers, diffusion processes in immobile water in the well facing unscreened parts of the well, or impervious rocks, etc..
Then, it is very important to provide data about:
- screened/unscreened, grouted/ungrouted parts of the well = technical log of the wells. I hypothesized that all wells are totally screened (or open) to follow the proposed interpretations;
- permeability of the rocks surrounding the well, at least qualitatively. In other words, where are the aquifers (pervious “layers”, or “intervals”), where are the impervious “layers”. Such a description (or interpretation) is lacking in the paper and this is preventing a good interpretation of the data, and the trust in the final conceptual model.
This description of pervious/impervious “layers” can either result from previous observations (such as detailed geological core description enabling to locate clinker layers, cooling fractures, massive part of the lava flows, measurements performed in the wells, observations performed during drilling, etc.) or can be a result from this paper, notably from the joint interpretation of these previous data and the EC and T profiles.
Without such data, EC (and T) profiles performed without pumping in the well (or without any measurement of the ambient flow) are very difficult to surely interpret. Then, it is difficult to trust on the interpretations provided in section 3.1. What is interpreted as transition zone (well SS-1 for instance) is not sure. If the SE formation is not an aquifer, the salted water could for instance come from a deeper aquifer, and what is interpreted as a transition zone is a transition zone in the well, but may not be a transition zone in the surrounding aquifer.
Then, the temperature data enable to complete the EC ones (from line 250). I agree that cooler water infiltrated at a higher elevation and flowing in the aquifer explains the lower temperature; this is a process that is commonly observed in volcanic aquifers – see for instance Toulier et al. (2019) – that should be introduced before in the paper in order to explain the process to the reader. Then, the explanation provided at lines 314-316 must also be revised.
However, the interpretation is not unique for all wells. For instance, in well SS-3, we may alternately interpret the sharp interface as:
- a sharp fresh/salt water interface in the aquifer if all the acicular basalt is an aquifer. But, then, one should explain that fresh water is flowing downwards, and salt water is flowing upwards (or is immobile) in the aquifer to explain the strong temperature change (thermally, due to diffusion processes, and possibly convection, I am not sure that this interpretation really works);
- or that this interface is at the limit between an aquifer (above -59 m amsl) and an impervious layer (deeper).
Some other hypotheses may surely be invoked.
In well SS-4, it seems that the aquifer extends within the whole saturated zone
A semi quantitative interpretation of temperature data could also be provided. In most presumed aquifers (SS-2 (0 - -40 m), 3 (0 - -60m), 4 (0 - -80m)) the temperature is about 15°C (Fig. 3). But such a “low” temperature is not reached at well SS-1 (only about 15.8°C but only between 20 and 30 m).
So, at the stage of section 3.1, unless other data are available, I would suggest to the authors:
to be much more cautious in their interpretations (notably lines 263 – 265, 294 - 300), and to provide some first graphs that summarize the most probable hypotheses about the hydrogeological structure and groundwater flow for each well.
I have the same concern, for the same reasons, in section 3.2., notably from line 340. Additionally, Fig. 6 is very difficult to read (even after printing on a A3 page, what I did!).
Is the tide provided measured in the well or I the sea? It should be in the well to ensure an adequate comparison between tide and temperature, without any delay.
I think that what the authors explain may be right, but it is very difficult to follow the explanations. The authors must find a way to better explain, step by step, the proposed process to be able to convince the reader. For instance, it is difficult to understand what means “during period A”, and to combine with the logs: period A lasts several hours during which the water level is rising-up; a log is instantaneous. Is the log measured during or after A, at the beginning or at the end? Moreover, for instance, Fig. 6b says that it is during period A, but the legend says A, B, C, D.
As said above, these data may help locating the aquifers. In SS-1 it suggests that it is in the 0 - -20 m interval (to my first and fast understanding). As said above, this should be compared with other data to validate this hypothesis.
Then, finally, the conceptual model must clearly present:
where are the aquifers and aquicludes; what is the salinity of the water in the aquifers. There is no interest to show salinity in aquicludes.
Lachassagne, P., Aunay, B., Frissant, N., Guilbert, M., Malard, A. (2014). High-resolution conceptual hydrogeological model of complex basaltic volcanic islands. A Mayotte, Comoros, case study Terra Nova Vol. 26, N°4, PP. 307-321 DOI 10.1111/ter.12102
Toulier, A., Baud, B., de Montety, V., Lachassagne, P., Leonardi, V., Pistre, S., Dautria, J.M., Hendrayana, H., Miftakhul Fajar, M.H., Satrya Muhammad, A., Beon, O., Jourde, H. (2019). Multidisciplinary study with quantitative analysis of isotopic data for the assessment of recharge and functioning of volcanic aquifers: Case of Bromo-Tengger volcano, Indonesia Journal of Hydrology: Regional Studies 26 (2019) 100634 https://doi.org/10.1016/j.ejrh.2019.100634
Some more minor comments:
- line 37: basaltic aquifers
- line 85: “defined” instead of “known”?
- Lines 90-91: I don’t understand well this sentence. Which equilibrium? The one of the fresh/seawater interface? Reaching such an equilibrium is very fast (years, or tens of years) compared with geological processes that shape a coastline. Is this sentence necessary? Or rephrase.
- Line 118-120: unclear sentence, please revise
- paragraph beginning at line 150. Only tidal influence. No other influence such as earth tides? There seem to be some irregularities even on the SS-4 curve
- Table 1: it is not very clear what means EL and GL. I understand that EL is elevation above mean sea level (amsl), but then, the term “sensor depth’ is not clear. Please revise
- Figure 2: the legend is very difficult to read. How to explain the temperature increase in SS-3?
- Table 2: the water classes are a bit arbitrary. The authors must explain that, and that this choice was made for sake of simplicity in the explanations of the paper.
- explanations at lines 228-231 are not very clear. If this issue may impact the results of the paper, it must be better explained.
- Fig. 3: please indicate in the caption that N° are explained in the text.
- Line 240 – 241: this sentence must be positioned in section 2. (Material and methods).
- section 3.1: what means EL that appears several time. I understand that it is “elevation” or depth below groundwater level. It would be better to describe as an elevation (– 18 m amsl), and maybe to explain at the first occurrence in the next that it means 18 m below the zero amsl elevation.
- Table 4 (and Fig. 3): as said in the text, the geological logs should be completed. Changes in lava mineralogy is of interest as they surely point out the limit between 2 lava flows. But they must be completed with data linking geology to permeability, such as clinker layers, cooling fractures, massive parts of the lava flows, etc..
- line 313-324 is unclear.
- line 322: replace “into” by “down to”?
- table 5: what does “frequency” mean? It is possible to have a unit for this parameter?
- line 338: add Tide level fluctuation “in the sea”?
Author Response
We appreciate reviewers’ detailed and kind comments/suggestions to provide improvements of article. Almost all of the proposed comments are considered in the article to enhance readers’ understanding. All figures are replaced with high resolution, and some manuscripts are revised as reviewers recommended. Additionally, the comments of the reviewers were carefully compiled, more literatures are added to support our results.
Author Response File: 
Author Response.docx
Reviewer 2 Report
The concept of the paper is well explained, methods used are innovative and authors did a good job of researching their subject. The conceptual model of seawater intrusion could be applied worldwide for the same/similar geological and hydrogeological conditions. The goal of the work is well stated, results are clear, however the paper needs some minor changes before publication. I have some presentation concerns. Please see comment comments below.
Please show on the map 1 four districts (please see line 139). All figures in papers are too small and resolution of figures is low. I suggest having a greater number of figures (e.g. figure 2 can be divided in two parts – the first one including results for SS-1 and SS-2 and the second one including results for SS-3 and SS-4 etc.) or at least authors needs to increase fonts and marks (legends). Figure 1 – instead of grey color for: “Automatic Weather System” please use black color; please increase symbols for Tide stations; please increase a small map showing state boundary and clearly mark investigated area on it. Figure 2 – legend is missing, please put it. Figure 3 – please mark clearly “outstanding variations of electrical conductivity” – line 246 Figure 5, table 5 and lines from 320-326 should be part of the chapter 2.3.
Author Response
We appreciate reviewers’ detailed and kind comments/suggestions to provide improvements of article. Almost all of the proposed comments are considered in the article to enhance readers’ understanding. All figures are replaced with high resolution, and some manuscripts are revised as reviewers recommended. Additionally, the comments of the reviewers were carefully compiled, more literatures are added to support our results.
Author Response File: Author Response.docx
Round 2
Reviewer 1 Report
The paper deals with the characterization of the shape of the salt water wedge in the volcanic aquifer of the Jeju island Korea, mostly from EC and temperature monitoring, in a set of boreholes.
I already reviewed this paper and I recommended a major revision. This review is thus dealing with the first revision of the paper.
Most of my proposals were considered by the authors. However, there are still a few issues. I summarized the most important below, and the minor ones afterwards.
My main concerns are (i) that the conclusive explanations, that are summarized on Fig. 8 are still a bit unclear for me (and then surely for the future readers), and (ii) that I am not convinced that the proposed interpretation/conceptual model may not be revised in the future.
Then, I would recommend:
- To address my concern N°(ii), the authors may explain in a few lines in the “Discussion” section that the interpretations presented in this paper are preliminary and may be improved/revised in the future (this may also be briefly stated in the abstract and conclusion). The authors may particularly stress on the fact that in the current conditions of ambient monitoring, natural vertical flows can occur in the wells and alter the real natural thermal and EC of the rocks, and notably of the aquifers (I already wrote a similar sentence in my previous review). Future improvements of the research may deal with:
- a much precise characterization of the location (and measurements of hydrodynamic parameters) of the aquifers/impervious layers (for instance with flowmeter measurements during pumping),
- absolute or relative measurements of the piezometric head in each of the identified aquifers, once they will have been precisely identified. This could be done with packers (absolute measurements) or with ambient conditions flowmeter (relative measurements). This should consider the salinity of the water (density effects, correction of piezometric head accordingly).
- To address my concern N°(i), the authors should revise Fig. 8 (and also some explanations in the text accordingly):
- It is not clear from figure 8 (and also notably from section 3.1.) if the aquifers are only the thin yellow layers or are much larger. To my understanding of volcanic aquifers, aquifers can correspond to thin layers (notably interfaces between lava flows), but here, on the figure, the black arrows are confusing as they are not corresponding to the yellow layers;
- On Fig. 10, the U formation is considered as an aquiclude. I agree with that. On Fig. 8, it is however indicated that it is filed with freshwater, which suggests that it is an aquifer. This is not consistent and must be revised. Moreover, this aquifer is not yellow on Fig. 8;
- 8 should indicate the salinity of the water only in aquifers (aquiclude were not sampled, and are not important for the purpose of this paper), and the 3 types of water (fresh, transition, salt) must be indicated. If the authors consider that several “yellow” layers constitute an unique aquifer, the figure could be adapted accordingly, and this could be explained in the text;
- Maybe a proposal would be to present 2 figures: one with aquifers and aquicludes, and the second one with the salinity of water (fresh, transition, salt) and the arrows that indicate the direction of the groundwater flow (explain in the caption of Fig. 8 that the arrows “denote the direction of the costal aquifer groundwater flow…”). To avoid any confusion between what is observed in the wells (cf. Fig. 3), and what is the reality in the aquifer (which is the objective of Fig. 8), colors for salinity could be drawn between the wells.
To summarize, I think that the conceptual model is not yet totally clarified after this paper, even after these last improvements; that’s why I suggest to write my first recommendation in the discussion.
Another important point: after this second reading of the paper, I don’t see at all how the age the rocks (line 488 – 489) can be directly related to the current functioning of the aquifer. I then suggest to suppress the entire sentence.
Other comments:
- line 123: I would suggest rewriting line 123 as follows: “interflow structures (notably clinker layers) that are permeable and with a high porosity [46]”
- table 1: drilling depth must be written with a minus. For instance, line 1: -159.62
- caption of Fig. 2: “temporal variation of ocean tide level
- lines 265-266: “Borehole SS-1 showed…”
- Fig. 3: explain in the caption what mean the black and red arrows
Author Response
We appreciate reviewer’s detailed and kind 2nd comments/suggestions to provide improvements of article. Almost all of the proposed comments are considered in the article to enhance readers’ understanding. Some manuscripts and figures are revised as reviewers recommended.
We would like to briefly introduce the research currently being conducted in Jeju Island. In early 2000, drilling began to recover drilling cores, and the drilling cores are used to study the scientific understanding of the geology of Jeju Island. Water cycle studies using isotopes, dating rock cores, and packer tests to characterize aquifers at high altitudes are also underway. As advised by the reviewers, we recognize the need for various in situ and laboratory test in order to accurately characterize aquifers in Jeju Island.
Author Response File: Author Response.docx
