Hydrochemical and Isotope (¹⁸O, ²H and ³H) Characteristics of Karst Water in Central Shandong Province: A Case Study of the Pingyi-Feixian Region

Abstract

Karst water serves as an important water supply source in northern China. Hydrochemical and isotope (¹⁸O, ²H, and ³H) characteristics are invaluable tools to identify water–rock interaction activities in karst water systems. In this study, the investigation of hydrogeological conditions, hydrogeochemistry, and hydrogen and oxygen isotopes of karst water revealed: (1) HCO₃-Ca-type water is widely distributed throughout the study area, while HCO₃-Ca·Mg-type water appears in the direct recharge areas and the discharge areas; karst water of the HCO₃⁻·Cl⁻, Cl⁻·HCO₃⁻, and Cl⁻ types is scattered in low-land areas; (2) karst water has high δ¹⁸O, which may be due to the dissolution and exchange of ¹⁸O from the surrounding carbonate rocks in the western discharge zone; and (3) the ³H concentration of karst water is strongly correlated with the content of major ions (Ca²⁺, Mg²⁺, HCO₃⁻, SO₄²⁻, and Cl⁻) and it increases along the flow path. It was also revealed that the karst water in the discharge areas is mixed with lateral recharge (infiltration recharge from surface water).

1. Introduction

Each karst water system has its own storage, transport, and regulatory functions and constantly exchanges water with other system components, resulting in changes in water quantity and quality as the entire system interacts with the external environment [1,2,3,4,5,6,7,8,9,10,11,12,13,14].

The hydrochemical characteristics of karst water, as well as the isotope (¹⁸O, ²H, and ³H), are direct indicators of spatial variations in water quantity and quality [15,16,17,18,19,20,21,22,23]. The hydrochemistry of karst water is generally focused on the distribution of major ions (K⁺, Na⁺, Ca²⁺, Mg⁺, HCO₃⁻, SO₄²⁻, and Cl⁻). ²H and ¹⁸O are stable isotopes that are widely used to retrieve information about the sources, areas, and times of karst water recharge. This information is acquired by comparing the isotope composition of karst water with that in global or regional water bodies, indicating the recharge source of karst water. ³H is a natural radioactive isotope of hydrogen with a half-life of 12.32 years. Due to atmospheric nuclear tests in the 1950s and in the early 1960s, the ³H concentration in precipitation increased sharply; it peaked at 1868 TU in 1963 [24] and declined every year until reaching 32 TU in 2001 (from the precipitation monitoring station of IAEA in Shijiazhuang), identifying ³H as a good indicator of whether the recharge sources of karst water formed before or after these nuclear tests (1952). In recent years, a large number of karst water studies have explored the structures, hydrodynamic processes, and hydrochemical evolution patterns of karst water systems based on analyses and measurements of major ions and isotopes (¹⁸O, ²H, and ³H), especially in the karst water areas of northern China [25,26,27,28,29,30,31,32,33,34,35,36,37,38,39].

In this case study, a hydrogeological survey, with a scale of 1/50,000, was carried out in the Pingyi–Feixian region at central Shandong Province, China. The aim of this study was to reveal the spatial distribution characteristics of major ions and isotopes (¹⁸O, ²H, and ³H) to provide insight into the hydrochemical evolution patterns of the karst water system.

2. Hydrogeological Setting of Karst Water in the Study Region

2.1. Geological Setting

The Pingyi–Feixian karst region in the central zone of Shandong Province, China, is situated at 117°30′–118°00′ E and 35°10′–35°30′00″ N, and it covers 596 km². According to monitoring data from the Feixian and the Pingyi meteorological stations, in Shandong Province from 1971 to 2013 the annual mean precipitation in the area is 798 mm and the annual mean evapotranspiration is 1536 mm. The precipitation from June to August accounts for approximately 65% of the annual precipitation. As shown in Figure 1, the region is characterized by hilly landforms, with a surface elevation of 100–500 m, and the strata generally dip toward the north and east, which is consistent with the topography.

The region has the following karstic formations composed of thick carbonate rocks: (1) the Mantou Formation (∈₂m) of the Changqing Group; (2) the Gushan Formation (∈_3–4g), the Zhangxia Formation (∈₃z), the Chaomidian Formation (∈₄O₁c), and the a, b, and c sections of the Sanshanzi Formation (O₁s^a, O₁s^b, and O₁s^c) of the Jiulong Group; and (3) the Dong Huangshan Formation (O₂d), Bei’anzhuang Formation (O₂b), Tuyu Formation (O₂t), Wuyangshan Formation (O₂w), Gezhuang Formation (O₂g), and Badou Formation (O_2–3b) of the Majiagou Group. Moreover, the region is covered by Quaternary deposits on both banks of the Junhe River, Zhutian River, and Wenliang River and in near low-lying valleys. Strata of Neogene, Cretaceous, and Carboniferous are distributed sporadically in the relatively flat zone in the northern part of the study area. Intrusive rocks and metamorphic rocks are mainly distributed in the low mountainous zones in the middle and southern regions of the study area. There are a series of fault structures in the southeast direction and four deep faults in the north and south directions.

2.2. Hydrogeological Characteristics of Karst Aquifers

Combining the previous research and the findings of this study, the karst region was divided into six zones according to hydrogeological conditions (Figure 2): two direct recharge zones, one indirect recharge zone, and three discharge zones [40].

The direct recharge zones consist of zones W and E, which are situated in the Changli–Dongyang area of Pingyi County and the Zhutian–Yuanwai area of Feixian County, respectively. The karst landscape is characterized by clustered hill valleys with hilly and low-mountain terrain in which seasonal rivers are distributed in a plumose pattern. The strata lithology is mainly limestone and dolomite intercalated with shale ranging from the Mantou Formation of the Changqing Group to the Sanshan Formation of the Jiulong Group. The strata are mostly bare, with a precipitation infiltration coefficient of 0.17–0.21, referring to the ratio of the amount of water infiltrated by precipitation to the amount of precipitation in the same period [40]. Surface karst landforms such as karst grooves and crevices are widespread, allowing a large amount of precipitation to directly infiltrate the underground karst water system as recharge.

The indirect recharge zone, which is situated in the Changli–Zhangli–Tongshi area in the middle of the study region, has a hilly and low-mountain terrain. The strata lithology is mainly metamorphic rocks such as gneiss, granite, and diorite porphyrite formed in the Upper Archean, as well as intrusive rocks formed at various stages. Some areas in the zone are partly covered with Cambrian carbonate rocks intercalated with mud shale. The precipitation infiltration coefficient in this region is 0.05–0.08 [40]. A small amount of precipitation infiltrates as groundwater recharge, forming shallow weathering fissure water, while most precipitation forms surface runoff that enters carbonatite-dominated areas, where the surface water infiltrates the karst aquifer as recharge.

The discharge zones consist of zones W, E, and S, which are situated in the Dongyang–Wenshui area of Pingyi County, the Difang–Chengbei area of Feixian County, and the urban area of Feixian County, respectively. The landforms are characterized by karst basins with loose surface Quaternary deposits and low-lying valleys. The karst strata are mainly limestone, dolomite, and argillaceous limestone (occasionally containing gypsum) of the Majiagou Group with well-developed karst crevices, pores, and caves in which the water yield of a single well exceeds 1000 m³/d.

2.3. Exploitation of Karst Water

Karst water is mainly used for agricultural irrigation in the Dongyang–Wenshui area of Pingyi County inside zone W and for urban residential consumption in the Difang–Chengbei area of Feixian County inside zone E. It is extracted from motor-pumped wells and the water yield of a single well is generally less than 300 m³/d, although it can reach as high as 2.4 × 10³ m³/d for domestic water plants in Feixian County.

The total exploitation volume of karst water in 2013 and in 2015 was 82.2 million m³ and 108.9 million m³, respectively, increasing by 18.9% to 144.3 million m³ in 2017. In this study, the whole area was divided into nine units (C1–C9) according to hydrological conditions and the extraction modulus (defined as the volume of karst water extracted per km² and year) was calculated for each unit. As shown in Figure 3, the extraction moduli of karst water in the units ranged from 2.48 × 10⁴ m³/(a·km²) to 28.07 × 10⁴ m³/(a·km²) in 2017. Notably, the extraction modulus exceeded 20 × 10⁴ m³/(a·km²) in both the Wenshui unit (C1) of Pingyi County and the Feixian County unit (C6), where artificial extraction is relatively concentrated, followed by 15.21 × 10⁴ m³/(a·km²) in C1 and less than 10 × 10⁴ m³/(a·km²) in other units.

3. Materials and Methods

In this study, 78 karst water samples were collected in the direct recharge zones: 22 karst water samples in the indirect recharge zone and 32 karst water samples in the discharge zones (Figure 4). The sampling period for 28 typical samples of major ions (K⁺, Na⁺, Ca²⁺, Mg²⁺, HCO₃⁻, SO₄²⁻, and Cl⁻) and isotopes (¹⁸O, ²H and ³H) was October–November 2013. The sampling depth was 0–370 m in boreholes (

Table 1. Typical hydrochemical and isotopic samples of precipitation and karst water.

Sample No.	Sampling Date	Sampling Location	Sampling Depth (m)	Notes
Fxj	November 2013	Feixian County		Precipitation
SW134	November 2013	Pingyi County	85–200	Indirect recharge zone
SW136	November 2013	Pingyi County	9–88
SW137	November 2013	Pingyi County	8–27
DW018	November 2013	Pingyi County	60–187	Direct recharge zone(W)
DW110	November 2013	Pingyi County	18–26
DW111	November 2013	Pingyi County	0–141
DW130	November 2013	Pingyi County	14–170
FK03-1	October 2013	Feixian County	4–32	Direct recharge zone(E)
FW281	November 2013	Feixian County	30–73
FW237A	October 2013	Pingyi County	80–100
FW332	October 2013	Feixian County	18–80
FW071	October 2013	Feixian County	50–120
FW073	October 2013	Feixian County	0–88
FW237	October 2013	Pingyi County	80–100
FW087	November 2013	Feixian County	40–170
FK03	October 2013	Feixian County	29–202
FK01	October 2013	Feixian County	80–325
FW232	November 2013	Feixian County	200–370
DW093	November 2013	Pingyi County	28–40	Discharge zone(W)
DW094	October 2013	Pingyi County	16–20
DK03A	October 2013	Pingyi County	0–301
DW007	October 2013	Pingyi County	40–150
DW100	November 2013	Pingyi County	180
DW102	November 2013	Pingyi County	150
DK03	October 2013	Pingyi County	0–301
SW018	October 2013	Feixian County	15–88	Discharge zone(E)
FW110	October 2013	Feixian County	20–200
SW216	October 2013	Feixian County	15–200

). The sampling sites were selected so the upstream–downstream relationship of karst water in the region would be revealed by subsequent analyses.

During sampling and testing of the karst water, three samples were collected at each sampling site, placed in airtight 1.0 L polyethylene plastic bottles without head space, and stored. ¹⁸O,²H and ³H isotope measurements were conducted at the Institute of Hydrogeology and Environmental Geology, Chinese Academy of Geological Sciences.

The detection of major ions in water adopted the standardized method of China’s geology and mineral industry. Flame emission spectrometry was used to determine K⁺ and Na⁺, ethylenediamine tetraacetic acid disodium titration was used for Ca²⁺ and Mg²⁺, argentometric titration for Cl⁻, Ethylenediamine tetraacetic acid disodium–barium titration for SO₄²⁻, and acidimetry acid titration for HCO₃⁻. Specifically, water δ¹⁸O was measured using the CO₂-H₂O equilibration method, while water δ²H was determined using the chromium reduction method. The analyses were conducted using a MAT-253 Gas Isotope Ratio Mass Spectrometer with a precision of σ ≤ 0.1‰ and σ ≤ 2‰, respectively. The water concentration of ³H was determined by the electrolytic concentration method using an ultra-low background 1220 Quantulus liquid scintillation spectrometer with a precision of σ ≤ 0.6 TU.

4. Results and Discussion

4.1. Hydrochemical Characteristics of Karst Water

Water samples were subjected to hydrochemical classification based on the values of the major ions (Schukalev classification), as shown in

Table 2. Chemical composition of the water samples.

Sample No.	pH	δ2H (‰)	δ18O (‰)	3H (TU)	TDS (mg/L)	K+ (mg/L)	K+ (mmol/L)	Na+ (mg/L)	Na+ (mmol/L)	Ca2+ (mg/L)	Ca2+ (mmol/L)	Mg2+ (mg/L)	Mg2+ (mmol/L)	HCO3 (mg/L)	HCO3 (mmol/L)	SO42 (mg/L)	SO42 (mmol/L)	Cl (mg/L)	Cl (mmol/L)	Notes
Fxj	/	−60	−9.2	12.6	/	/	/	/	/	/	/	/	/	/	/	/	/	/	/	Precipitation
SW134	7.6	−64	−8.3	7.2	678.2	0.89	0.023	16.7	0.72	120.4	3.01	26.5	1.10	288.7	4.73	73.1	0.76	36.5	1.03	Indirect recharge zone
SW136	7.2	−62	−8.0	8.9	1949.2	1.56	0.040	66.7	2.90	415.5	10.39	71.1	2.96	254.2	4.17	188.6	1.96	410.3	11.56
SW137	7.4	−56	−7.3	8.0	929.6	1.11	0.028	32.0	1.39	175.8	4.39	32.3	1.35	389.1	6.38	91.9	0.96	86.6	2.44
DW018	7.4	−60	−8.3	10.9	859.9	1.11	0.028	20.0	0.87	157.8	3.95	32.7	1.36	375.6	6.16	73.1	0.76	54.7	1.54	Direct recharge zone(W)
DW110	7.3	−62	−8.1	9.2	695.4	1.11	0.028	10.0	0.43	110.9	2.77	34.6	1.44	304.7	4.99	26.9	0.28	56.0	1.58
DW111	7.8	−59	−8.0	10.7	519.7	0.44	0.011	7.5	0.33	107.9	2.70	13.6	0.57	263.9	4.33	36.7	0.38	38.3	1.08
DW130	7.5	−62	−8.8	5.5	898.7	0.67	0.017	26.2	1.14	186.8	4.67	23.0	0.96	355.2	5.82	82.5	0.86	69.9	1.97
FK03–1	7.8	−58	−8.1	8.3	793.2	0.33	0.008	14.2	0.62	129.2	3.23	45.7	1.90	375.4	6.15	61.2	0.64	41.3	1.16	Direct recharge zone(E)
FW281	7.4	−61	−8.7	7.1	554.2	0.44	0.011	2.5	0.11	121.1	3.03	11.1	0.46	323.7	5.31	34.3	0.36	17.7	0.50
FW237A	7.6	−60	−8.5	7.8	964.2	0.33	0.008	20.0	0.87	199.4	4.99	22.8	0.95	424.4	6.96	97.9	1.02	59.0	1.66
FW332	7.9	−59	−8.4	7.1	1167.3	1.22	0.031	36.0	1.57	215.7	5.39	39.5	1.65	375.4	6.15	257.1	2.68	64.9	1.83
FW071	7.7	−60	−8.5	6.0	596.8	0.33	0.008	13.3	0.58	104.0	2.60	28.0	1.16	263.6	4.32	61.3	0.64	36.5	1.03
FW073	7.4	−62	−8.6	8.9	579.8	0.33	0.008	3.3	0.14	108.9	2.72	22.8	0.95	350.9	5.75	31.8	0.33	13.3	0.37
FW237	7.4	−61	−8.5	8.1	662.6	0.22	0.006	4.2	0.18	132.3	3.31	23.5	0.98	296.5	4.86	63.7	0.66	13.3	0.37
FW087	8.0	−61	−8.3	8.5	586.8	0.44	0.011	7.5	0.33	108.9	2.72	26.5	1.11	266.6	4.37	80.8	0.84	25.1	0.71
FK03	7.5	−59	−8.2	10.1	718.9	0.44	0.011	11.7	0.51	121.1	3.03	39.5	1.65	353.6	5.80	49.0	0.51	41.3	1.16
FK01	7.7	−61	−8.6	7.1	579.5	1.33	0.034	8.8	0.38	86.5	2.16	35.8	1.49	353.6	5.80	31.8	0.33	16.2	0.46
FW232	7.9	−62	−8.6	7.1	533.9	0.44	0.011	5.0	0.22	84.4	2.11	32.7	1.36	238.5	3.91	68.4	0.71	13.7	0.39
DW093	7.2	−70	−8.9	7.4	868.1	0.22	0.006	13.3	0.58	183.0	4.57	16.5	0.69	414.2	6.79	66.0	0.69	45.6	1.28	Discharge zone(W)
DW094	7.5	−59	−8.2	9.2	681.7	1.44	0.037	15.0	0.65	122.1	3.05	26.5	1.11	302.0	4.95	53.9	0.56	50.1	1.41
DK03A	7.4	−59	−8.3	8.4	725.6	0.44	0.011	10.8	0.47	143.5	3.59	22.2	0.93	402.6	6.60	46.5	0.48	35.4	1.00
DW007	7.6	−61	−8.5	12.2	714.3	0.44	0.011	7.9	0.34	147.5	3.69	21.0	0.87	315.6	5.17	61.2	0.64	32.4	0.91
DW100	7.7	−56	−7.6	9.6	656.0	0.44	0.011	11.7	0.51	130.2	3.26	18.5	0.77	302.0	4.95	46.5	0.48	38.3	1.08
DW102	7.5	−59	−8.3	6.7	880.5	0.56	0.014	15.0	0.65	176.9	4.42	32.3	1.35	379.7	6.22	47.1	0.49	92.7	2.61
DK03	7.9	−60	−8.4	8.1	500.1	0.33	0.008	5.8	0.25	93.6	2.34	20.4	0.85	236.7	3.88	39.2	0.41	26.5	0.75
SW018	7.6	−60	−8.3	10.9	530.5	0.56	0.014	9.2	0.40	99.7	2.49	16.7	0.69	269.3	4.41	39.2	0.41	25.1	0.71	Discharge zone(E)
FW110	7.4	−58	−8.0	8.1	606.4	1.7	0.044	15.0	0.65	108.5	2.71	21.9	0.91	315.5	5.17	60.8	0.63	31.9	0.90
SW216	7.6	−58	−8.0	9.0	535.7	1.22	0.031	9.2	0.40	100.7	2.52	21.0	0.87	269.3	4.41	46.5	0.48	29.5	0.83

and in Figure 5 and Figure 6a,b. Karst water types in the study area included the HCO₃-Ca type, HCO₃-SO₄-Ca type, and HCO₃-Ca-Mg type. The water type was found to mainly be HCO₃-Ca-type and HCO₃-SO₄-Ca type in the indirect recharge zone, while the main karst water types in the direct recharge and discharge zones were found to be HCO₃-Ca-Mg and HCO₃-Ca. Direct recharge zone E and discharge zone E, which are both in the eastern portion of the study region, were found to have a significantly higher proportion of HCO₃-Ca·Mg-type karst water than that in western direct recharge zone W and discharge zone W. Among the anions accounting for greater than 25% of the total, HCO₃⁻ was found to be widespread in the whole study area, with a combination of HCO₃⁻ and SO₄²⁻ mainly distributed in the central portion, which mainly features exposed metamorphic rocks of gneiss, granite, diorite porphyrite, and intrusive rocks, while Cl⁻ was found to be distributed sporadically in low-lying area. Cations were found to have a more obvious distribution pattern in the study region, with karst water abundant in both Ca²⁺ and Mg²⁺ primarily distributed in the main flow zones.

The distribution frequencies of TDS in karst water are shown in Figure 6c. The karst water samples of the indirect recharge zone presented TDS levels in the range of 248–1949 mg/L, of which samples with TDS levels of 400–500 and 500–600 mg/L accounted for 32% and 18%, respectively. The direct recharge zones had karst water TDS levels in the range of 323–899 mg/L: 400–500, 500–600, and 600–700 mg/L constituted the major intervals, especially for direct recharge zone E, where such water samples accounted for 35%, 31%, and 28% of the total samples, respectively. The karst water samples of the discharge zones had TDS levels in the range of 278–1059 mg/L, of which samples with TDS levels of 500–600 mg/L and 600–700 mg/L accounted for 49% and 32%, respectively. In particular, samples with TDS levels of 500–600 and 600–700 mg/L accounted for the highest proportions, with 55% and 48% in discharge zones W and E, respectively.

4.2. ²H and ¹⁸O Isotope Characteristics of Karst Water

As shown in Table 2, the measured δ²H and δ¹⁸O values of karst groundwater ranged, respectively, from −70‰ to −56‰, with an average of −60‰, and from −8.8‰ to −7.3‰, with an average of −8.3‰. As shown in Figure 7, the δ¹⁸O and δ²H values of karst water in the entire region were roughly distributed along the meteoric water lines, reflecting that the regional groundwater originated from precipitation. The δ¹⁸O-δ²H line of the study region was below the global meteoric water line [41], the China meteoric water line [27], and the meteoric water line of Shijiazhuang (Hebei Province), which is typical of northern China [42] and indicative of a continental effect. Specifically, the karst water in the study region underwent strong evaporation, thereby deviating from the global meteoric water line.

Linear regression of karst water δ²H against karst water δ¹⁸O for each zone revealed that the line fitting the indirect recharge zone was most similarly shaped to the meteoric water lines, followed by the lines fitting the discharge zones; while, the lines fitting the direct recharge zones had significantly smaller slopes than the meteoric water lines, reflecting that the water–rock interaction is weakest in the indirect recharge area and that the karst water in the direct recharge area features significant solution filtration in the runoff process, thereby increasing the degree of ¹⁸O isotope fractionation in the karst water.

The measured δ¹⁸O and δ²H values and the fitting lines for direct recharge zone E coincided well with those of discharge zone E, suggesting good runoff conditions in the two zones and, therefore, rapid karst water recharge. The δ¹⁸O values of discharge zone W were greater than those of direct recharge zone W, suggesting the karst water underwent strong evaporation after the recharge of precipitation in this zone in which the main outcrop is composed of limestone and dolomite intercalated with shale from the Mantou Formation of the Changqing Group to the Chaomidian Formation of the Jiulong Group. Moreover, the vadose zone is widely distributed in direct recharge zone W, where the vadose water undergoes evaporation and thus has a higher δ¹⁸O than precipitation. Such vadose water provides leakage recharge to deep karst water, thereby increasing the karst water δ¹⁸O. The slope of the fitting line of direct recharge zone E was close to the slopes of the meteoric water lines but significantly greater than the slope of direct recharge zone W, indicating a weaker evaporation effect on karst water in direct recharge zone E compared to direct recharge zone W. The δ¹⁸O–δ²H fit of discharge zone E coincided with that of discharge zone W, suggesting the karst water of discharge zone E features a similar ¹⁸O and ²H isotope composition to that of discharge zone W due to joint recharge from precipitation infiltration and river water leakage.

4.3. ³H Isotope Characteristics of Karst Water

As shown in Table 2, the ³H concentration in precipitation is 12.6 TU, which is significantly higher than that in karst water (by 5.5–12.2 TU), indicating the ³H concentration decreased naturally in the process of precipitation supplying karst water. Moreover, the ³H concentration of the karst water samples was observed to have the following major relationships with the five zones (Figure 8):

(1)

The ³H concentration generally increased with an increase in runoff distance. The lowest concentration was observed at the boundaries of karst watersheds (i.e., sampling sites DW130 and FW071). Sampling sites with ³H concentrations <7.5 TU were found to be mainly located at the front edge of the direct recharge zones (i.e., sampling sites FW281, FW232, FW332, and FK01). Sampling sites with ³H concentrations of 7.5–8.5 TU were located primarily inside the direct recharge zones (i.e., sampling sites FW237A, FW237, FK03-1, and FW087), while sampling sites with ³H concentrations of 9.0–12.5 TU were principally inside the discharge zones and at some near-boundary areas of the direct recharge zones relative to the discharge zones.

(2)

Discharge zones W and E both had significantly lower ³H concentrations distributed in a belt-like pattern. These low-value areas coincided with areas featuring the large-scale artificial extraction of karst water, indicating that karst water extraction induced old karst water that flowed from the aquifer to the extraction areas.

(3)

Changes in ³H concentrations were larger closer to large-scale extraction areas.

4.4. Hydrochemical Evolution of Karst Water

The dissolution of rocks and the precipitation of ions in the karst water during karst water flow determine the concentrations of major ions (Ca²⁺, Mg²⁺, HCO₃⁻, SO₄²⁻, and Cl⁻). However, the ³H in karst water is supplied by precipitation and surface water, and its concentration increases with an increase in cumulative recharge. Together, the dynamics of ³H and major ions represent the hydrodynamic characteristics of karst water.

4.4.1. Dissolution and Precipitation of Carbonate Rocks

The concentrations of Ca²⁺, Mg²⁺, and HCO₃⁻ are mainly subject to carbonate rock erosion by CO₂-rich karst water and the precipitation process. The relationship between the Ca²⁺ molar concentration and ³H concentration is shown in Figure 9a. Because of the large amount of surface water infiltration recharge to karst aquifers in direct recharge zone E and discharge zone E of the eastern study region, Ca²⁺ ions are mostly dissolved during karst water flow. Moreover, sampling sites FW237A and FW332 exhibited significantly higher concentrations of Ca²⁺ and SO₄²⁻ relative to the surrounding sampling sites, which suggests that, in addition to the dissolution of carbonate rocks, the two sampling sites were simultaneously influenced by the dissolution of CaSO₄ intercalated in the carbonate. However, the Ca²⁺ molar concentration was negatively correlated with the ³H concentration in direct recharge zone W and discharge zone W, indicating the karst aquifers in these zones were mainly recharged by deep lateral flow from outside the study region, while Ca²⁺ in the karst water underwent precipitation during the recharge process. The relationship of the Mg²⁺ molar concentration versus ³H concentration, which is shown in Figure 9b, suggests three different patterns: (1) the intensive dissolution and leaching of Mg²⁺ from carbonate rocks in direct recharge zone E; (2) the significant dilution of Mg²⁺ in discharge zone E by a large amount of karst water discharged from upstream; and (3) a relatively stable Mg²⁺ molar concentration in direct recharge zone W and discharge zone W, where the dilution effect of karst water recharge from surface water infiltration is offset by the dissolution and leaching of Mg²⁺ from the surrounding rocks.

Changes in the ratio of Ca²⁺ and Mg²⁺ molar concentrations (Figure 9c) further confirm that direct recharge zones E and W gradually experience a transition from carbonate mineral dissolution to calcium ions precipitation under the flow of karst water. Specifically, as the free CO₂ in karst water is gradually exhausted and the erosivity of karst water is weakened during the interactions of karst water with rocks, calcium ions precipitation gradually increases and the Ca²⁺/Mg²⁺ ratio decreases. The positive correlation between Ca²⁺/Mg²⁺ and the ³H concentration in discharge zones E and W suggests a large amount of CO₂-rich precipitation infiltrates as recharge to karst aquifers, thereby increasing the dissolution of Ca²⁺.

There is a positive correlation between the HCO₃⁻ molar concentration and ³H concentration (Figure 9d) in direct recharge zone E, where karst aquifers continue to receive a large amount of surface infiltration recharge as surface runoff flows carrying CO₂ and ³H into karst water. In contrast, there is a negative correlation between the HCO₃⁻ concentration and ³H concentration in direct recharge zone W and in discharge zones W and E. This result indicates that strong water–rock interactions occur in the lateral runoff of the deep karst aquifers such that the increased HCO₃⁻ concentration due to surface infiltration recharge fails to compensate for the consumed HCO₃⁻ concentration during lateral flow recharge, thereby decreasing the HCO₃⁻ concentration in karst water.

4.4.2. Dissolution of Sulfate Rocks

As shown in Figure 10a, there is a negative correlation between the SO₄²⁻ molar concentration and ³H concentration in direct recharge zones E and W, as well as in discharge zone E, indicating the SO₄²⁻ that dissolved and leached from rocks during the water–rock interaction was diluted by surface infiltration recharge. In contrast, the SO₄²⁻ molar concentration was found to be positively correlated with ³H in discharge zone W, suggesting the presence of higher strength water–rock interactions leading to the dissolution and leaching of SO₄²⁻ when compared with other zones. In addition, the scatter plot of SO₄²⁻ versus Ca²⁺ (Figure 10b) shows the data points reflecting the molar concentrations of SO₄²⁻ and Ca²⁺ in karst water were significantly below the gypsum dissolution line of slope 1:1. However, the scatterings in the eastern area of the direct recharge area are close to a 1:1 straight line, indicating the concentration of Ca²⁺ released from the gypsum is far less than that produced by the limestone in the study area and that gypsum dissolution is stronger in recharge zone E than in other areas.

4.4.3. Dissolution of Chlorides

Except for the abnormally high Cl⁻ concentration at SW136, the Cl⁻ concentration in the study area was found to range from 13.27 to 92.67 mg/L due to the dissolution of soluble halite rocks intercalated between mudstone and shale interbeds in the carbonate rocks. The karst water level is shallow and the evaporation is strong in the low-lying shallow discharge zones, leading to high Cl⁻ concentrations. Additionally, there is a decrease in Cl⁻ concentration with an increasing ³H concentration (Figure 11) in each zone, indicating the continuous dilution of Cl⁻ as a readily soluble and migratable component in the flow process of karst water.

5. Conclusions

Based on the hydrochemical and isotope (¹⁸O, ²H and ³H) investigation of karst water, the following conclusions can be provided:

(1)

Shallow karst water of the HCO₃-Ca type with a short flow path is ubiquitous in the study region, while karst water of the HCO₃-Ca·Mg type with a long flow path is present in both the direct recharge and discharge zones. The interactions of karst water with carbonate, sulfate, and halite result in a variety of hydrochemical processes such as dissolution, precipitation, mixing, and dilution. Overall, karst water has a slightly lower degree of mineralization in the direct recharge zones than in the discharge zones. In contrast, the indirect recharge zone suffers from a relatively closed condition with TDS up to 1949 mg/L, indicative of long-term water–rock interactions.

(2)

The δ¹⁸O-δ²H fitting line of karst water in the indirect recharge zone has the closest slope to the meteoric water lines, while the slopes of the fitting lines in the direct recharge zones are significantly lower than those of the meteoric water lines. The δ¹⁸O-δ²H fitting line of recharge zone E coincides with that of discharge zone E, indicating rapid karst water recharge. Moreover, δ¹⁸O of karst water is relatively high in discharge zone W, which may be attributed to the dissolution and leaching of ¹⁸O from carbonate rocks.

(3)

The ³H value of karst water increases along the flow path and it is generally higher than 9.0 TU in the discharge zones, where intensive exploitation of karst water has produced increased karst water flow in the surrounding areas, led to significant mixing between karst water receiving surface infiltration recharge and karst water receiving lateral recharge, and yielded larger changes in ³H concentrations.