the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 10 Feb 2025
Managed aquifer recharge and exploitation impacts on dynamics of groundwater level and quality in northern China karst area: Quantitative research by multi-methods
Abstract. Managed aquifer recharge (MAR) is an effective way to counter groundwater level decline and spring depletion caused by excessive groundwater exploitation in karst areas. However, the unique characteristics of karst groundwater systems make the groundwater quantity and quality more susceptible to human activities, posing challenges for MAR research. This research employed multi-methods including numerical simulations, isotope analysis, infiltration tests, flow monitoring and tracer tests to quantitatively analyze the impacts of MAR and groundwater exploitation on the dynamics of groundwater level and quality in a typical northern China karst area, the Baotu Spring area in Jinan City. First, the percentage of surface water recharge in karst groundwater was calculated using isotope data with the improved two-end-member mixing model. Next, the quantitative relationship between volume of released water and actual recharge was established with data from infiltration tests and flow monitoring. Then, the actual groundwater flow velocity and effective porosity of the karst aquifers were calculated with former tracer test isochrone maps. Finally, the impacts of MAR and groundwater exploitation on dynamics of groundwater level and quality were quantitatively analyzed with a groundwater flow-solute transport model for the area. The results indicate that the MAR and groundwater exploitation in the Baotu Spring area have significantly impacted karst groundwater levels and quality. These complementary methods enhance the accuracy of decisions in MAR and groundwater exploitation.
- Preprint
(8095 KB) - Metadata XML
- BibTeX
- EndNote
Status: final response (author comments only)
-
CC1: 'Comment on egusphere-2025-281', Giacomo Medici, 16 Feb 2025
General comments
Very good modelling research in the field of karst hydrology. Please, follow my guidance to improve the manuscript.
Specific comments
Lines 74-76. “In karst regions with low development, such as the karst areas in northern China, groundwater flow is predominantly laminar, largely complying with Darcy's law”. Insert supporting references for dominance of darcian flow in poorly karstified carbonate rocks in regions outside China:
- Agbotui P.Y., West L.J., Bottrell S.H. 2020. Characterisation of fractured carbonate aquifers using ambient borehole dilution tests. Journal of Hydrology, 589, 125191
- Medici G., Munn J.D., Parker B.L. 2024. Delineating aquitard characteristics within a Silurian dolostone
aquifer using high-density hydraulic head and fracture datasets. Hydrogeology Journal, 32(6), 1663-1691.
Line 117. Specify the 3 to 4 specific objectives of your research by using numbers (e.g., i, ii, and iii).
Lines 120-150. Insert information on the presence of faults that can either influence the groundwater flow, or represent preferential pathways for the recharge.
Line 216. Specify that you are representing the advective flow velocities for the transport.
Lines 487-642. Please, insert the two relevant papers on darcian flow in poorly karstified and fractured carbonates suggested above.
Lines 435-465. The objectives of your modelling research appear 4 by reading your conclusions. See my comment above in the introduction.
Figures and tables
Figure 1. I can see faults in your geological cross-section. They look normal faults. But, please explain this point in detail in study area section.
Figure 5a. Add the spatial scales. The vertical one cannot be detected.
Figure 6. Insert regression equations with the R2 values.
Figure 9. Make the graphs larger and improve the graphical resolution of the figure.
Figure 11. Make letters and numbers larger.
Citation: https://doi.org/10.5194/egusphere-2025-281-CC1 -
AC1: 'Reply on CC1', Weihong Dong, 19 Feb 2025
Thank you very much for your valuable suggestions on this study, I have made the following changes to this paper as per your comments:
Comment 1: Lines 74-76. “In karst regions with low development, such as the karst areas in northern China, groundwater flow is predominantly laminar, largely complying with Darcy's law”. Insert supporting references for dominance of darcian flow in poorly karstified carbonate rocks in regions outside China:
- Agbotui P.Y., West L.J., Bottrell S.H. 2020. Characterisation of fractured carbonate aquifers using ambient borehole dilution tests. Journal of Hydrology, 589, 125191
- Medici G., Munn J.D., Parker B.L. 2024. Delineating aquitard characteristics within a Silurian dolostone
aquifer using high-density hydraulic head and fracture datasets. Hydrogeology Journal, 32(6), 1663-1691.
Answer 1: The above two supporting references have been inserted in the manuscript in line 74.
Comment 2: Line 117. Specify the 3 to 4 specific objectives of your research by using numbers (e.g., i, ii, and iii).
Answer 2: The specific objectives of this research have been specified in lines 118-123.
Comment 3: Lines 120-150. Insert information on the presence of faults that can either influence the groundwater flow, or represent preferential pathways for the recharge.
Answer 3: The information on the presence of faults has been inserted in lines 145-148.
Comment 4: Line 216. Specify that you are representing the advective flow velocities for the transport.
Answer 4: It was mentioned in lines 227-228.
Comment 5: Lines 487-642. Please, insert the two relevant papers on darcian flow in poorly karstified and fractured carbonates suggested above.
Answer 5: The two relevant papers have been inserted in the manuscript in line 74.
Comment 6: Lines 435-465. The objectives of your modelling research appear 4 by reading your conclusions. See my comment above in the introduction.
Answer 6: Three research objectives have been summarized and specified in lines 118-123.
Comment 7: Figure 1. I can see faults in your geological cross-section. They look normal faults. But, please explain this point in detail in study area section.
Answer 7: The information on the faults have been added in lines 145-148.
Comment 8: Figure 5a. Add the spatial scales. The vertical one cannot be detected.
Answer 8: The MAR wells, groundwater exploitation wells and groundwater level monitoring wells are all in the aquifers from the Majiagou Formation of the Ordovician to the Chaomidian Formation of the upper Cambrian. There are hydraulic connections of the aquifers and none of the wells are stratified.
Comment 9: Figure 6. Insert regression equations with the R2 values.
Answer 9: Regression equation and Pearson’s R have been inserted in figure 6.
Comment 10: Figure 9. Make the graphs larger and improve the graphical resolution of the figure.
Answer 10: Figure 9 has been improved.
Comment 11: Figure 11. Make letters and numbers larger.
Answer 11: Figure 11 has been improved.
Citation: https://doi.org/10.5194/egusphere-2025-281-AC1
-
AC1: 'Reply on CC1', Weihong Dong, 19 Feb 2025
-
CC2: 'Comment on egusphere-2025-281', Nima Zafarmomen, 25 Feb 2025
This paper makes a significant contribution to the field by seamlessly integrating multiple research methods—ranging from numerical simulations to isotope analyses and field tracer tests—to provide a nuanced, quantitative assessment of MAR impacts on karst groundwater systems. Its innovative approach, particularly the improved two‐end‐member mixing model and the rigorous evaluation of effective porosity, enhances the accuracy and reliability of the findings. Moreover, by addressing both groundwater quantity and quality in a real-world setting, the work not only advances scientific understanding but also offers practical insights for sustainable water resource management in similar karst environments.
-
How do the authors quantify and address uncertainties in the improved two‐end‐member isotope mixing model, particularly given potential variations in isotopic signatures and possible contamination sources?
-
The study relies on an effective porosity value derived from tracer tests; how sensitive are the solute transport simulation results to variations in effective porosity and hydraulic conductivity assumptions, and how is this uncertainty evaluated?
-
What are the limitations of the groundwater flow model in capturing the spatial and temporal heterogeneities of the karst system, and how robust is the model calibration and validation against observed data?
-
The analysis of infiltration efficiency uses empirical relationships (e.g., formulas (7) and (8)) to relate released water volume to actual recharge; under what conditions might these relationships fail, and how do variable hydrological conditions affect their applicability?
- Considering the impact of MAR water quality on long-term groundwater sustainability, how do the authors define acceptable thresholds for contaminant levels, and what management strategies are proposed to mitigate the risk of groundwater quality deterioration over time?
I strongly recommend to cite below paper:
"Assimilation of sentinel‐based leaf area index for modeling surface‐ground water interactions in irrigation districts"
Citation: https://doi.org/10.5194/egusphere-2025-281-CC2 -
AC2: 'Reply on CC2', Weihong Dong, 27 Feb 2025
Thank you very much for the valuable questions and suggestions you have raised regarding this paper. I would like to offer the following responses to your comments:
Comment 1: How do the authors quantify and address uncertainties in the improved two‐end‐member isotope mixing model, particularly given potential variations in isotopic signatures and possible contamination sources?
Answer 1: As you have pointed out, potential variations in isotopic signatures and possible contamination sources do indeed exist objectively. Firstly, our research group collected multiple rounds of isotopic samples from groundwater between 2021 and 2023, and the values of the two-end-member isotopes showed little variation, remaining consistent with the values mentioned in the paper. Furthermore, based on our preliminary investigations and research, we believe that the amount of water discharged from potential pollution sources is relatively small. While it might affect the nitrogen and sulfur isotopes of the groundwater, it is unlikely to have a significant impact on the hydrogen and oxygen stable isotopes. Regarding the issue you raised, we will add an explanation in the paper.
Comment 2: The study relies on an effective porosity value derived from tracer tests; how sensitive are the solute transport simulation results to variations in effective porosity and hydraulic conductivity assumptions, and how is this uncertainty evaluated?
Comment 3: What are the limitations of the groundwater flow model in capturing the spatial and temporal heterogeneities of the karst system, and how robust is the model calibration and validation against observed data?
Answer 2&3: Effective porosity and hydraulic conductivity are important factors that influence the results of solute transport simulations. In this paper, hydraulic conductivity is assigned to different zones in order to characterize the spatial heterogeneity of the karst system, and calibration has been performed using observational data. However, there is a lack of a quantitative description of the calibration results, and this will be addressed in future revisions soon. We did not subdivide effective porosity into zones but instead took the average value from tracer test results, which is a simplified approach. This decision was primarily based on the inherent uncertainty in hydraulic conductivity, one of the key inputs for calculating effective porosity. Moreover, subdividing effective porosity would lack a calibration basis. We acknowledge this limitation and will consider how to improve it in future work.
Comment 4: The analysis of infiltration efficiency uses empirical relationships (e.g., formulas (7) and (8)) to relate released water volume to actual recharge; under what conditions might these relationships fail, and how do variable hydrological conditions affect their applicability?
Answer 4: The empirical relationship between released water volume and actual recharge presented in this paper is derived from actual flow monitoring data. Therefore, when the released water volume exceeds the range of the monitoring data, this empirical relationship may no longer be applicable. This point will be further elaborated upon in the paper.
Comment 5: Considering the impact of MAR water quality on long-term groundwater sustainability, how do the authors define acceptable thresholds for contaminant levels, and what management strategies are proposed to mitigate the risk of groundwater quality deterioration over time?
Answer 5: The acceptable thresholds for contaminant levels in this paper are based on China's groundwater quality standards. Groundwater that does not meet the Class III water standard is generally considered unacceptable. In practical applications, however, this acceptable threshold may also be determined based on the requirements set by decision-makers, typically water resource management authorities, regarding groundwater quality.
Comment 6: I strongly recommend to cite below paper: "Assimilation of sentinel‐based leaf area index for modeling surface‐ground water interactions in irrigation districts"
Answer 6: The paper you mentioned conducts in-depth and meaningful research on the numerical model of surface water-groundwater interaction. We will reference it in the first chapter of our paper.
Citation: https://doi.org/10.5194/egusphere-2025-281-AC2
-
-
RC1: 'Comment on egusphere-2025-281', Anonymous Referee #1, 06 Mar 2025
General remarks
The study described in this paper was necessary and has been done correctly.
However, the discussion and results are focused only on this case study. There is no reference of the results to the karst aquifers environment in general, so that this methodology could be adopted in other areas worldwide and the results compared. Karst areas have their own specificity, so certain principles and processes are often common in most of them. However, there are no connections in manuscript to these general features of karst aquifers.
The second shortcoming is the lack of discussion (and possibly conclusions) regarding the suitability of the adopted research methods for studies of this kind.
Detailed remarks
Fig. 1b. - this is a hydrogeological cross-section, not "profile". It is necessary to enlarge this cross-section, because it is important for understanding the hydrogeological conditions. Precipitation is everywhere along the entire cross-section. In the zone described as "Precipitation" it is the aquifer "Recharge".
Fig. 2. is unnecessary, because everything is repeated in Fig. 7 - where it is all better visible.
Fig. 5a should be rotated by 180 degrees or 90 degrees, to be consistent with the cross-section (Fig. 1b) and to show the geological layers indicated in the lithology explanations. Now there are no visible. In addition, on Fig. 5a it will be very useful to overlay the rivers, as well as mark the MAR zone (if it is graphically possible).
Fig. 5b – not needed, because it repeats what is in Figures 1a and 2. Recharge coefficient values can be added (e.g. in brackets) next to the lithology explanation on Fig.1a.
Fig. 6. - it is necessary to add the numbers/names of wells, boreholes, springs, etc., corresponding to the numbers in Fig. 7. If adding all of them is not possible graphically, then at least most of them.
Fig. 10 - maps in panels "a" and "b" are illegible, because they are too small. It is necessary to enlarge them significantly - maybe even 2x (?). No indication of units for parameters shown on these maps. Panels c, d, e, f can be moved to Supplement Material, because they do not contribute anything important enough to be in the main text.
Fig. 12 - Maps are too small. Only four of them are enough to compare the results, i.e. for 150 and 350 mg/L, for 2 and 18 months, respectively. The rest of these maps can possibly be in Supplement, if necessary.
Lines 444-448 – this is an important problem, but there is no attempt to explain its cause. Either the infiltration tests were inaccurate, because they significantly overestimated the intensity of infiltration, or there is some other cause. But, there is no attempt to answer what could be the reason and what could be the way to counteract this issue.
Lines 458-461 – an accurate remark, but there is no determination of minimum standards for MAR water quality in this case study. It seems that this is necessary because thanks to this, this study will have an additional application effect.
Table 1 can be moved to the Supplement because it does not contribute anything important enough to be in the main text.
Citation: https://doi.org/10.5194/egusphere-2025-281-RC1 -
AC3: 'Reply on RC1', Weihong Dong, 08 Mar 2025
I sincerely appreciate your valuable suggestions for this study. I would like to provide the following responses to your comments:
Response to General Remarks:
As a representative region, the study area in this paper (the Baotu Spring Basin in Jinan City) reflects the general development characteristics of karst aquifers and groundwater flow patterns in northern China’s karst regions. Additionally, similar karst formations can be found in certain countries and regions in northern Europe. Therefore, the research findings and methods presented in this study may also be applicable to those areas. In response to your suggestion, we will incorporate relevant references in Chapter 1 and expand our discussion on this topic. Furthermore, we will add a discussion on the applicability of these research methods to similar studies.
Response to Detailed Remarks:
Your detailed remarks are very insightful, and we will optimize the figures and tables accordingly based on your suggestions. Additionally, we would like to provide the following clarifications:
- Regarding Figure 2 and Figure 7: Figure 2 contains more information on wells and sampling sites compared to Figure 7. We will consider removing Figure 2 and merging its information into Figure 7.
- Regarding Figure 5b: Since the recharge coefficient for rainfall infiltration is influenced by different land use types, its zonation does not completely align with the lithological zonation. We believe that Figure 5b remains necessary, but we will revise it to avoid redundant information with Figures 1a and 2.
- Regarding the issue mentioned in Lines 444–448:
- First, we believe that the two research results are not contradictory. When the discharge exceeds 21.6 × 10⁴ m³/d, the infiltration recharge will be lower than the discharge; however, they remain positively correlated. Theoretically, the maximum recharge capacity of Yufu River can reach 114.9 × 10⁴ m³/d.
- Nevertheless, we acknowledge certain limitations in our study that may affect the accuracy of this result. For instance, during the flow monitoring period, the maximum flow in Yufu River was 36.72 × 10⁴ m³/d, which is still significantly lower than 114.9 × 10⁴ m³/d. This means that the linear relationship expressed in Equation (7) lacks sufficient empirical support when discharge exceeds 36.72 × 10⁴ m³/d.
- Additionally, this maximum recharge capacity was estimated under the assumption of a hydraulic gradient of 1. However, substantial recharge may lead to groundwater level rise, potentially invalidating this assumption. We will clarify these limitations in the paper.
- Regarding the issue mentioned in Lines 458–461: We will determine the minimum water quality standard for MAR based on China’s groundwater quality standards and the findings of this study, ensuring greater applicability of our research.
Citation: https://doi.org/10.5194/egusphere-2025-281-AC3
-
AC3: 'Reply on RC1', Weihong Dong, 08 Mar 2025
-
RC2: 'Comment on egusphere-2025-281', Anonymous Referee #2, 02 Apr 2025
The manuscript “Managed aquifer recharge and exploitation impacts on dynamics of groundwater level and quality in northern China karst area: Quantitative research by multi-methods” presents a comprehensive analysis of groundwater dynamics in terms of head and quality in a karst area of China. It is well-structured, well-written, and employs established methodologies.
In my opinion, the manuscript lacks novelty and generalization. I do not see it as a research article but rather as an outstanding case study. All the methods employed in the study (i.e., isotope analysis, infiltration tests, river flow monitoring, tracer tests, and groundwater flow and transport simulations) have been previously introduced in the literature and extensively applied in other studies. Specifically, I do not see any significant advancement in any of these methodologies within this work. Moreover, the conclusions drawn from the study cannot be generalized and apply only to the analyzed system.
That said, I want to emphasize that the work is of very high quality, and its results are indeed relevant to the management of the Baotu Spring area. However, I believe this is not a research article in the sense that it does not produce novel knowledge that can be generalized to other systems or develop new methodologies.
Some elements that should be considered and could potentially improve the quality of the work include:
- Key studies that I see as relevant to this research are not cited, such as Mudarra & Hartmann (2019) and Hartmann et al. (2015).
- Some references mentioned in the article are missing or not properly cited in the reference list, making it difficult to review the manuscript (e.g., Chuanlei Li et al., 2022, and J. Li et al., 2023).
- Be more precise when stating that simulating karst systems using the Equivalent Porous Medium (EPM) assumption is valid in karst. As written, it suggests that this approach is applicable to any non-developed system. However, some of the references you cite indicate that this assumption is valid only in specific scenarios and for particular applications (e.g., Scanlon et al., 2003). Additionally, it is difficult to support the claim that flow within karst systems is "predominantly laminar," as these systems typically contain sinkholes and caves where turbulence surely occurs.
- Typo in figure 1 (Reservoirs)
- In Section 2.2., it is unclear why the ratios calculated via Eqs. (3) and (4) are not appropriate for your analysis, how Eq. (5) was derived, and to what extent Eq. (5) improves upon Eqs. (3) and (4).
- The preprocessing steps for the streamflow data are not clearly explained. Additionally, it is unclear whether these data are representative of other years for these rivers. More details on the approach used in the analysis should be included.
- How did you construct Figure 7? Is this any sort of kriging?
- How did you carry out the infiltration experiments? Did you perform the experiment in the riverbed or in areas close to the riverbed?
- Why the Figure 8b indicates that in almost the entire period the infiltrated water coincides with the released water and in figure 9b the actual recharge differs from the released water volume?
Citation: https://doi.org/10.5194/egusphere-2025-281-RC2 -
AC4: 'Reply on RC2', Weihong Dong, 04 Apr 2025
Thank you very much for your valuable comments on our manuscript. Below are our responses to your suggestions:
Although all the methods used in this study are well-established and widely applied, their combined application serves to mutually validate the results. The Baotu Spring karst aquifer is a representative example of karst systems in northern China, and research on groundwater in this area has long been a focal point. However, despite the abundance of existing research data, previous studies often lacked comprehensive analysis and utilization of these datasets. Particularly in the past decade, several Managed Aquifer Recharge (MAR) projects have been implemented in this region, leading to significant changes in groundwater balance and flow patterns. Therefore, this study was primarily conducted from the perspective of addressing practical engineering needs in the area. We will supplement relevant references and provide a more detailed discussion on the generalizability of our findings.
Responses to specific comments:
- We will cite the key research papers you mentioned.
- The degree of karst development in the Baotu Spring aquifer is relatively low. Based on extensive borehole data, the deep karst aquifer media mainly consist of karst fissures and pores, with only a few locations showing centimeter-scale dissolution cavities. Numerous groundwater numerical modeling studies on the Baotu Spring aquifer have employed the Equivalent Porous Media (EPM) approach, although these studies did not elaborate on this point in detail. Furthermore, the MODFLOW-based model used in this study successfully reproduces the observed groundwater level dynamics. Following your suggestion, we will further clarify the validity of the EPM model in our study area.
- We will correct the typographical errors and conduct a thorough review of the entire manuscript.
- Due to space limitations, the description of this section was overly simplified. We will supplement the explanation of the meaning and derivation process of Equation (5).
- We will provide a more detailed explanation of this section. The primary purpose of analyzing this dataset was to investigate the riverbed's leakage capacity, which remains consistent across different years (as no large-scale construction has been conducted on the riverbed in recent years).
- Figure 7 was generated using the natural neighbor interpolation method, with appropriate extrapolation and smoothing applied to the initial interpolation curve.
- The infiltration tests were conducted on the riverbed, but at its edges (as there was still minimal flow in the center during the dry season). We will supplement the manuscript with field photos.
- During the flow monitoring period of the Xingji River (2014), the discharge was mostly below 20,000 m³/day, except in the final month when it exceeded this value (at which point the infiltration volume significantly differed from the discharge). However, during the simulation period (2020–2022), the discharge from the Xingji River was much higher, often exceeding 20,000 m³/day.
Citation: https://doi.org/10.5194/egusphere-2025-281-AC4
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 298 | 73 | 25 | 396 | 11 | 26 |
- HTML: 298
- PDF: 73
- XML: 25
- Total: 396
- BibTeX: 11
- EndNote: 26
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1