3.1. Hydrochemical characteristics
Table 1 shows the chemical composition of karst groundwater, pore groundwater and surface water in the study area. The order of cation concentration in the karst groundwater during the wet and dry seasons is Ca2+>Mg2+>Na+>K+ and Ca2+>Na+>Mg2+>K+, and there is little difference between Na+ and Mg2+. The order of anion concentration in the wet and dry seasons is HCO3−>SO42−>Cl−>NO3−>F−. The order of cation concentrations in the pore groundwater during the wet and dry seasons is Ca2+>Na+>Mg2+>K+, which is consistent with karst groundwater. The order of cation concentrations in the groundwater in the study area is Ca2+>Na+>Mg2+>K+, and that of the anion concentrations is HCO3−>SO42−>Cl−>NO3−>F−.
Table 1
Statistics of hydrochemical parameters of groundwater and surface water (mgl− 1)
Season
|
Type
|
Parameter
|
TDS
|
K+
|
Na+
|
Ca2+
|
Mg2+
|
Cl−
|
SO42−
|
HCO3−
|
NO3−
|
F−
|
TH
|
Wet
|
Karst groundwater
(n = 21)
|
Mean
|
643.80
|
2.53
|
25.68
|
138.58
|
32.04
|
67.46
|
162.88
|
299.08
|
44.46
|
0.40
|
480.87
|
Maximum
|
868.96
|
15.00
|
63.53
|
178.74
|
57.07
|
163.63
|
276.79
|
412.57
|
150.82
|
0.90
|
618.46
|
Minimum
|
467.62
|
0.65
|
6.04
|
87.24
|
8.01
|
17.54
|
53.04
|
209.56
|
3.40
|
0.04
|
329.42
|
Pore groundwater
(n = 12)
|
Mean
|
890.02
|
1.73
|
86.51
|
162.51
|
45.64
|
114.49
|
209.28
|
420.01
|
79.16
|
1.15
|
613.07
|
Maximum
|
1796.22
|
10.00
|
257.76
|
255.03
|
72.37
|
252.25
|
706.89
|
520.26
|
197.99
|
3.24
|
934.93
|
Minimum
|
507.66
|
0.50
|
21.18
|
53.20
|
10.32
|
31.72
|
51.00
|
181.41
|
0.12
|
0.18
|
308.17
|
Surface water
(n = 3)
|
Mean
|
1012.39
|
9.67
|
157.85
|
110.27
|
45.77
|
161.26
|
383.24
|
235.19
|
14.99
|
1.16
|
463.84
|
Maximum
|
1225.08
|
11.11
|
205.91
|
153.98
|
55.77
|
188.17
|
486.95
|
259.57
|
34.76
|
1.30
|
614.13
|
Minimum
|
901.14
|
8.69
|
117.65
|
68.93
|
38.77
|
117.82
|
279.84
|
196.96
|
0.19
|
0.97
|
331.80
|
Dry
|
Karst groundwater
(n = 21)
|
Mean
|
532.71
|
1.72
|
31.87
|
122.22
|
26.96
|
72.64
|
156.02
|
218.94
|
46.62
|
0.47
|
416.21
|
Maximum
|
713.01
|
8.75
|
142.50
|
163.17
|
82.46
|
180.83
|
336.23
|
376.99
|
159.02
|
0.90
|
711.39
|
Minimum
|
315.52
|
0.40
|
4.49
|
86.16
|
0.52
|
17.14
|
77.59
|
114.36
|
10.93
|
0.10
|
217.30
|
Pore groundwater
(n = 12)
|
Mean
|
721.54
|
1.11
|
60.38
|
125.84
|
33.78
|
113.12
|
144.55
|
256.80
|
67.13
|
0.88
|
453.32
|
Maximum
|
1265.56
|
6.25
|
141.81
|
215.83
|
71.99
|
242.08
|
455.96
|
392.59
|
207.87
|
3.41
|
751.07
|
Minimum
|
399.60
|
0.18
|
17.69
|
34.53
|
1.05
|
31.45
|
43.07
|
127.36
|
5.63
|
0.10
|
213.08
|
Surface water
(n = 3)
|
Mean
|
844.87
|
6.74
|
167.52
|
77.80
|
36.75
|
170.55
|
271.70
|
204.08
|
3.86
|
0.83
|
345.59
|
Maximum
|
895.35
|
8.00
|
192.00
|
116.55
|
38.26
|
180.83
|
335.20
|
259.99
|
7.65
|
0.90
|
441.92
|
Minimum
|
769.08
|
5.97
|
142.50
|
56.12
|
35.34
|
153.31
|
226.43
|
171.60
|
1.04
|
0.70
|
285.63
|
During wet and dry seasons, the average concentrations of TDS, Na+, Mg2+ and Cl− are the highest in surface water, followed by pore groundwater, and are lowest in the karst groundwater. The average concentrations of Ca2+, HCO3−, NO3− and TH are the highest in pore groundwater, followed by karst groundwater, and then surface water. Average concentration of K+ is the highest in surface water, followed by karst groundwater, and then pore groundwater. Average concentration of SO42− is the largest in surface water, followed by pore groundwater and then karst groundwater in wet season, and followed by karst groundwater and then pore groundwater in dry season. Average concentration of F− is the lowest in karst groundwater, followed by surface water and then pore groundwater in wet season, and followed by pore groundwater and then surface water in the dry season. In general, karst groundwater quality is better than pore groundwater, and surface water has the worst quality.
The average concentrations of TDS, K+, Ca2+, Mg2+, SO42−, HCO3−, and TH in the karst groundwater during the wet season are higher than those in the dry season, while other indicators are in the opposite way. The average concentration of all indicators in the pore groundwater during the wet season is higher than that in the dry season. The average concentrations of all indicators in the surface water are higher in the wet season than in the dry season except for Na+ and Cl−, which show the contrary. On the whole, the quality of karst groundwater, pore groundwater, and surface water in the dry season is better than that in the wet season, which may be related to the fact that precipitation carries various ion components into the groundwater during the wet season.
Along karst groundwater flow from the mountainous area in the south to the piedmont plain in the north and northwest, the ion content of karst groundwater shows different changes (Fig. 2). The concentrations of Na+, Mg2+, Cl−, SO42−, TDS, and TH in karst groundwater during the wet and dry seasons showed a continuous increase trend from mountain to piedmont plain area, while Ca2+, HCO3− and NO3− showed a continuous decrease trend. Karst groundwater constantly dissolves the ionic components in the rock formations during its migration to the piedmont plains after receiving precipitation in the southern mountainous area. Therefore, the concentrations of Na+, Cl−, SO42−, TDS and TH continue to increase. The steady decrease of Ca2+ may be related to cation exchange and the reduction of HCO3− may be related to the greater buried depth of aquifers far away from the mountainous area, because less corrosive CO2 can lead to less CaCO3 dissolution in limestone or dolomite. The continuous decrease of NO3− may be due to the increase of karst aquifer buried depth, karst aquifers are not easily affected by human activities on the surface.
In the vertical direction, the variation range of the ion concentration of the upper pore groundwater during the wet and dry seasons is significantly larger than that of the lower karst groundwater, while the ion concentration of the karst groundwater increases slightly with wellbore depth (Fig. 3). The large range of ion changes in the pore groundwater may be because it is more susceptible to human activities. The increase of ion content in karst groundwater with wellbore depth may be caused by the slower runoff due to the greater depth.
Figure 4 shows that there are 7 types of karst groundwater in the dry season, mainly HCO3-Ca and HCO3•SO4-Ca•Mg type. There are 5 types of water in the wet season, mainly HCO3•SO4-Ca•Mg and HCO3•SO4-Ca type. Compared with the dry season, the HCO3 and SO4 type in the karst groundwater during the wet season increased by 14% and 24%, respectively. The significant changes in karst groundwater chemical types between the wet and dry seasons indicate that the groundwater in the southern mountainous area can quickly drain to the whole area after receiving the precipitation recharge, and the karst groundwater circulation is faster. Compared with karst groundwater, the chemical types of pore groundwater are more complex, with 10 types in both dry and wet periods. Compared with the dry season, the HCO3 type in the pore groundwater during the wet season increased, and the Cl type and Mg type decreased. The inconsistency of groundwater chemistry types between wet and dry seasons may be related to precipitation recharge. The water chemistry type of surface water is quite different from that of pore groundwater and karst groundwater. There are 2 and 3 types of water chemistry in dry season and wet season, with Cl type and Na type being dominant.
Comparing the hydrochemical types of karst groundwater, pore groundwater and surface water in the study area, Na-type water rarely appears in karst groundwater, and only occurs during the dry season of hole J23. However, 33% of pore groundwater and 100% of surface water samples are Na type water. In addition, the hydrochemical types of karst groundwater also show clear spatial distribution characteristics. The southern mountainous area and its intermountain areas are the direct replenishment areas of karst groundwater, where the groundwater chemistry types are relatively simple, generally HCO3-Ca and HCO3•SO4-Ca types. The northern part is the drainage area of karst groundwater, where the groundwater chemistry types are complex and changeable, showing HCO3•SO4-Ca•Mg, HCO3-Ca, SO4•Cl-Ca, HCO3•SO4-Ca and other types. The complex groundwater chemistry types indicate that the water chemistry characteristics of pore groundwater and karst groundwater may be affected by many factors.
3.3. Formation mechanism controlling groundwater chemistry
3.3.1. Water-rock interaction
The Gibbs graphic method (Ronald J. Gibbs. 1970) can be used to analyze the role of precipitation, water-rock interaction and evaporation in the chemical evolution of groundwater(Brindha et al. 2011;Wang et al. 2014༛Liu et al. 2015; Bouzourra H et al. 2015; Li et al. 2015;Li et al. 2016; Adimalla N 2018). The karst groundwater, pore groundwater and surface water in the study area are all located in the dominant area of water-rock interaction in the Gibbs diagram (Fig. 5), indicating that they are all mainly affected by water-rock interaction. Figure 5a shows that Na/(Na + Ca) has the smallest value in karst groundwater, followed by pore groundwater, and surface water has the largest value. Some pore groundwater and surface water are close to the evaporation dominant area and may also be affected by evaporation to a certain extent.
Karst groundwater mainly occurs in the Ordovician Majiagou Group strata. The lithology of this strata is dominated by carbonate rocks such as limestone (CaCO3) and dolomite (CaMg (CO3)). If HCO3−, Ca2+, and Mg2+ are from the dissolution of carbonate rocks, then HCO3− and Ca2++Mg2+ should be basically equal, but Fig. 6b shows that the karst groundwater points are all located above the 1:1 line, indicating that the Ca2+ or Mg2+ should have other sources. According to the drilling core data in the study area, there is a gypsum layer in the area, that is, the dissolution of the gypsum will also increase the Ca2+ in the karst groundwater. Figure 6c shows that the karst groundwater sample points are all distributed near the 1:1 line, further confirming that the HCO3−, SO42−, Ca2+, and Mg2+ in the karst groundwater come from the dissolution of limestone, dolomite and gypsum.
Most of the karst groundwater samples in the wet season are located on the upper right side of the dry season (Fig. 6b and 6c), indicating that the content of HCO3−, Ca2+ and Mg2+ in the wet season is higher than that in the dry season. The reason may be related to the rainfall carrying CO2 into the karst groundwater during the wet season, which promotes the dissolution of carbonate rocks.
3.3.2. Cation exchange
The correlation between (Na+–Cl−) and [(Ca2++Mg2+)–(HCO3−+SO42−)] and the chlor-alkali index (CAI-1, CAI-2) can be used to explain the cation exchange effect (Marghade D et al. 2012; Rajesh R et al. 2015; Salem ZE et al. 2015; Li et al. 2016;Li et al. 2018). When cation exchange is the main factor affecting the formation of groundwater chemistry, the relationship between (Na+–Cl−) and [(Ca2++Mg2+)–(HCO3−+SO42−)] is linear, and the slope is close to -1. The correlation equations of karst groundwater in wet and dry seasons are y=-0.523x + 0.9399 and y=-1.3976x + 1.3393, R2 is 0.2897 and 0.1877 respectively; the correlation equations of pore groundwater in wet and dry seasons are y=-1.0329x + 1.1758 and y=-1.3511x + 1.0595, R2 is 0.921 and 0.946 (Fig. 7a) respectively. Comparing and analyzing the correlation equations and correlation coefficients of karst groundwater and pore groundwater, pore groundwater appears to be more affected by cation exchange than karst groundwater.
The expression of the chlor-alkali index is shown in formulas (1) and (2). When the chlor-alkali index is positive, the exchange of Na+ and Ca2+ occurs according to formula (3), that is, when Ca2+ enters the groundwater, Na+ is adsorbed by solid particles in the positive direction; when the chlor-alkali index is negative, Na+ and Ca2+ exchanges in the opposite direction (4). The greater the absolute value of the chlor-alkali index is, the stronger the cation exchange effect gets. Figure 7b shows that the chlor-alkali index of most karst groundwater is positive (only two samples are negative in the dry season), that is to say, the cation exchange of karst groundwater is mainly in the positive direction, and the maximum value does not exceed 0.8, indicating that the exchange is not strong. The negative and positive values of the chlor-alkali index of pore groundwater are about half, which means both the positive and negative directions of cation exchange are existing. The reason may be related to the pore groundwater's susceptibility to evaporation and human activities. The maximum value of CAI-1 in pore groundwater can reach − 5.20, indicating that cation exchange in pore groundwater is stronger than karst groundwater.
………….(1)
………….(2)
2Na++CaX2 = Ca2++2NaX………….(3)
Ca2++2NaX = 2Na++CaX2……….(4)
3.3.3. Human Activities
The human activities that can affect the hydrochemical characteristics of groundwater in the study area include agricultural activities, karst groundwater extraction and industrial production, which have direct or indirect effects on the formation of groundwater hydrochemistry. Fertilizers and pesticides used in agriculture enter groundwater directly with precipitation infiltration, directly affecting the quality of groundwater. The average concentration of NO3− and Cl− in pore groundwater in the study area is obviously greater than that in karst groundwater (Fig. 5). The NO3− concentration in karst groundwater gradually decreases from mountainous areas to piedmont plains (Fig. 4f), that is, points with higher NO3− concentrations are located in the groundwater replenishment area in front of the mountain which means that agricultural activities have a certain impact on karst groundwater chemistry. The urban and rural residents of Zoucheng City and the production of major industrial enterprises in Taiping Industrial zone mainly use karst groundwater as a source of water supply. The large-scale mining of karst groundwater promotes water-rock interaction and cation exchange, which indirectly affects the hydrochemical characteristics of groundwater. Taiping Town in the northern part of the study area is a high-tech industrial zone in Zoucheng City with many industries, such as chemical industry, pharmaceuticals, and paper making. On one hand, they extract a large amount of karst groundwater; on the other hand, the discharge of sewage and wastewater from industrial enterprises will affect the quality of pore and karst groundwater.
3.4. Principal Component Analysis
Table 4 shows the principal component analysis results of K+, Na+, Ca2+, Mg2+, Cl−, SO42−, HCO3−, F−, NO3−, TDS and TH during the dry and wet seasons of karst groundwater. For the karst groundwater in the dry season, two principal components were extracted based on the characteristic value greater than 1, which explained 82.27% of the original variables. Among them, the principal components RC1 and RC2 explained 58.36% and 23.91% of the variance, respectively. The principal component RC1 has high correlation with Na+, Mg2+, Cl−, SO42−, HCO3−, F−, TDS and TH, and the correlation coefficients are 0.897, 0.971, 0.918, 0.894, 0.609, 0.802, 0.945 and 0.915, respectively. This represents the dissolution of salt rock, carbonate rock and gypsum, indicating the occurrence of water-rock interaction. The main component RC2 is closely related to K+, Ca2+ and NO3−, and the correlation coefficients are 0.748, 0.824 and 0.939, respectively, representing human activities (such as pesticides, fertilizer use and industrial wastewater discharge). For the karst groundwater in the wet season, three principal components were extracted based on the characteristic value greater than 1, which explained 84.85% of the variance, of which the principal component RC1 and principal component RC2 explained 51.13% and 24.33% of the variance, respectively. The principal component RC1 has a good correlation with Na+, Ca2+, Mg2+, Cl−, SO42−, TDS and TH, and the correlation coefficients are 0.851, 0.649, 0.812, 0.903, 0.801, 0.977 and 0.965, respectively. The principal component RC2 has a good correlation with K+ and NO3−, and the correlation coefficients are 0.896 and 0.939, respectively. It is easy to see that RC1 and RC2 in the wet season have the same interpretation as the RC1 and RC2 in the dry season, which are water-rock interaction and human activity influence respectively. The principal component RC3, explaining 9.39% of the variance, has a high correlation with HCO3− and F−. It is speculated that CO2 in precipitation promotes the dissolution of carbonate rocks and precipitation carries industrial wastewater into the aquifer. Therefore, the RC3 indicates the influence of atmospheric precipitation.
Table 4
The principal component analysis results of karst groundwater
Parameter
|
Dry season
|
Wet season
|
RC1
|
RC2
|
RC1
|
RC2
|
RC3
|
K+
|
0.091
|
0.748
|
0.175
|
0.896
|
0.076
|
Na+
|
0.897
|
0.137
|
0.851
|
-0.035
|
-0.058
|
Ca2+
|
0.355
|
0.824
|
0.649
|
0.473
|
0.272
|
Mg2+
|
0.971
|
-0.091
|
0.812
|
-0.378
|
-0.156
|
Cl−
|
0.918
|
0.250
|
0.903
|
0.028
|
0.104
|
SO42−
|
0.894
|
-0.229
|
0.801
|
-0.231
|
-0.456
|
HCO3−
|
0.609
|
0.257
|
0.490
|
0.047
|
0.779
|
F−
|
0.802
|
-0.465
|
0.404
|
-0.251
|
-0.728
|
NO3−
|
-0.123
|
0.939
|
-0.198
|
0.939
|
0.150
|
TDS
|
0.945
|
0.315
|
0.977
|
0.193
|
0.018
|
TH
|
0.915
|
0.355
|
0.965
|
0.092
|
0.091
|
Eigenvalues
|
6.42
|
2.63
|
5.62
|
2.68
|
1.03
|
Variance explained (%)
|
58.36
|
23.91
|
51.13
|
24.33
|
9.39
|
Cum. var. explained (%)
|
58.36
|
82.27
|
51.13
|
75.46
|
84.85
|
Table 5 shows the principal component analysis results of K+, Na+, Ca2+, Mg2+, Cl−, SO42−, HCO3−, F−, NO3−, TDS and TH during the dry and wet seasons of pore groundwater. For the pore groundwater in the dry season, three principal components were extracted based on the characteristic value greater than 1, which explained 83.72% of the original variables. Among them, the principal components RC1, RC2 and RC3 explained 45.79%, 27.07% and 10.86% of the variance, respectively. The main component RC1 has a high correlation with Na+, Ca2+, Mg2+, Cl−, SO42−, TDS and TH, and the correlation coefficients are 0.647, 0.636, 0.828, 0.919, 0.835, 0.989 and 0.916 respectively. This represents the dissolution of chlorine rock, magnesium rock and gypsum, that is, water-rock interaction. The main component RC2 is closely related to Na+, Ca2+, Mg2+, HCO3−, F− and NO3−, and the correlation coefficients are 0.551, -0.647, 0.500, 0.814, 0.682 and − 0.760 respectively, indicating that they are affected by human activities such as industrial sewage and wastewater discharge, domestic sewage, agricultural activities, and mining. The principal component RC3 is only closely related to K+, with a correlation coefficient of 0.892, which may be due to the dissolution of silicate. For the pore groundwater in the wet season, three principal components were extracted based on the characteristic value greater than 1, which explained 84.54% of the variance. The principal components RC1, RC2 and RC3 explained 51.71%, 22.30% and 10.53% of the variance, respectively. The principal component RC1 also has a high correlation with Na+, Ca2+, Mg2+, Cl−, SO42−, TDS and TH, and the correlation coefficients are 0.768, 0.826, 0.601, 0.927, 0.955, 0.964 and 0.890 respectively, indicating that the principal component RC1 is similar to that of the dry season. The main component RC2 has a good correlation with K+, Ca2+, F− and NO3−, and the correlation coefficients are 0.724, 0.534, 0.640, and 0.721 respectively. It also shows that it is affected by human activities such as industrial wastewater discharge, domestic sewage, agricultural activities and mining activities. The main component RC3 has a high correlation with Mg2+, HCO3− and F−, and the correlation coefficients are 0.718, 0.882, and 0.632 respectively, which may be affected by precipitation during the wet season.
Table 5
The principal component analysis results of pore groundwater
Parameter
|
Dry season
|
Wet season
|
RC1
|
RC2
|
RC3
|
RC1
|
RC2
|
RC3
|
K+
|
0.010
|
0.072
|
0.892
|
-0.059
|
0.724
|
0.231
|
Na+
|
0.647
|
0.551
|
-0.362
|
0.768
|
-0.398
|
0.277
|
Ca2+
|
0.636
|
-0.647
|
0.366
|
0.826
|
0.534
|
-0.087
|
Mg2+
|
0.828
|
0.500
|
-0.100
|
0.601
|
-0.167
|
0.718
|
Cl−
|
0.919
|
-0.253
|
0.139
|
0.927
|
0.009
|
0.156
|
SO42−
|
0.835
|
-0.016
|
-0.153
|
0.955
|
-0.080
|
0.030
|
HCO3−
|
0.066
|
0.814
|
0.224
|
0.256
|
0.100
|
0.882
|
F−
|
-0.074
|
0.682
|
-0.487
|
-0.205
|
-0.640
|
0.632
|
NO3−
|
0.223
|
-0.760
|
-0.018
|
0.064
|
0.721
|
-0.196
|
TDS
|
0.989
|
-0.128
|
0.035
|
0.964
|
0.113
|
0.201
|
TH
|
0.916
|
-0.249
|
0.234
|
0.890
|
0.337
|
0.240
|
Eigenvalues
|
5.04
|
2.98
|
1.19
|
5.69
|
2.45
|
1.16
|
Variance explained (%)
|
45.79
|
27.07
|
10.86
|
51.71
|
22.30
|
10.53
|
Cum. var. explained (%)
|
45.79
|
72.86
|
83.72
|
51.71
|
74.01
|
84.54
|