Descriptive statistics of heavy metals in mine soils
The summary statistics of the heavy metal concentrations in the soil samples are shown in Table 3. The mean concentrations of these heavy metals, in decreasing order, were Zn (39.00 mg·kg-1) > Cr (37.39 mg·kg-1) > Pb (17.45 mg·kg-1) > Ni (13.32 mg·kg-1) > Cu (10.93 mg·kg-1) > As (4.26 mg·kg-1) > Cd (0.09 mg·kg-1) > Hg (0.03 mg·kg-1). Compared with the risk screening values defined by the National Environmental Quality Standards for Soils in China (GB15618-2018) and the average values in China (Teng et al., 2014), the mean concentrations of these heavy metals were all lower than the reference values, indicating that the soil is relatively safe for agricultural production and human health. However, the maximum concentrations of Cd, Cu, Hg, Zn exceeded the risk screening values by factors of 1.30, 1.04, 2.78, and 2.03, respectively, showing that Cd, Cu, Hg, and Zn in the soil of the mining area were enriched to a certain extent.
The highest coefficient of variation (CV) occurred for Hg (4.56), reflecting a wider extent of variability concerning the mean. Zinc (0.97) also had high CV values. Cuprum (0.66), Ni (0.55), As (0.54), Cd (0.49), and Cr (0.45) showed moderate variation, whereas Pb (0.16) showed low variability concerning spatial distribution.The abnormal high CV values of Hg and Zn indicated that these metals were likely affected by anthropogenic factors (Lv, 2019; Turhun et al., 2022).
Table 3
Summary statistics of heavy metal concentrations (mg·kg-1) in top soil of the Shendong coal base (n=129)
|
As
|
Cd
|
Cr
|
Cu
|
Hg
|
Ni
|
Pb
|
Zn
|
Min
|
1.28
|
0.03
|
5.93
|
2.71
|
0.005
|
2.90
|
13.12
|
10.89
|
Max
|
15.20
|
0.39
|
92.93
|
51.97
|
1.39
|
41.97
|
36.48
|
405.73
|
Median
|
3.53
|
0.08
|
32.56
|
8.79
|
0.01
|
10.67
|
16.88
|
30.09
|
Average
|
4.26
|
0.09
|
37.39
|
10.93
|
0.03
|
13.32
|
17.45
|
39.00
|
CV
|
0.56
|
0.51
|
0.47
|
0.69
|
4.56
|
0.57
|
0.16
|
0.97
|
Background of Shanxi a
|
11.2
|
0.09
|
62.5
|
21.4
|
0.09
|
28.8
|
21.4
|
69.4
|
Average of China b
|
11
|
0.097
|
61
|
23
|
0.065
|
27
|
27
|
74
|
Risk screening values c
|
20
|
0.3
|
150
|
50
|
0.5
|
60
|
70
|
200
|
a Data from CEPA and TSCEM (1990).
b Data from Teng et al. (2014).
c Based on the lowest soil risk screening values in the National Environmental Quality Standards for soil in China (GB15618-2018).
Spatial analysis of the hevay metals in mine soils
We applied the GIS-based Ordinary Kriging interpolation method to map the spatial distribution of the heavy metal concentrations in mine soils. The spatial distribution of the concentrations of the eight heavy metals in the mining soils is substantially heterogeneous (Fig. 2). In the case of As, Cd, Cr, Cu, Ni and Pb, we find a zonal distribution pattern, with the most accumulation in the southern areas and least accumulation in the northern areas investigated in this study, and the concentrations of Hg and Zn decreased from the center to the southern and northern areas. We observe a “hot point” of Hg and Zn in the floodplain of the sewage outlet of the Daliuta Coal Washing Plant. Moreover, in addition to the enrichment of Hg and Zn, Cuprum (Cu), Cr, Cd and Pb are also enriched in the hot point, indicating that the heavy metals enriched in the gangue and/or chemical agents added to coal washing will leach into the wastewater during the coal washing process, and migrate to the soil with the sewage discharge.
Contamination assessment of heavy metals in mine soils
We categorize the degree of contamination of heavy metals identified in mine soils of Shendong coal base based on the values of the CF and the PLI. As shown in Fig. 3, the average CF of each heavy metal was below 1, indicating that the soils in the Shendong coal base were uncontaminated in general. Whereas, the proportions of each element with CF values greater than 1 were in decreasing order of Cd (32.56%) > Zn (8.53%) > Cr (8.50%) > Cu (7.75%) > Pb (6.98%) > Ni (4.65%) > Hg (3.10%) > As (0.78%), suggesting that the heavy metals in the soil of the mining area are enriched to a certain extent.
Fig.4 shows the samples with a CF value greater than 1. The slightly contaminated samples included 1 sample of As, 3 samples of Hg, 6 samples of Ni, 8 samples of Cu, 9 samples of Pb, 10 samples of Zn, 11 samples of Cr, and 38 samples of Cd. Moreover, the low contaminated samples included 2 samples of Cu and 3 samples of Cd. One sample of Cd falls into the moderately contaminated level. One sample of Hg and Zn falls into the heavily contaminated level. The areas with slight soil contamination are mainly distributed in the south areas, including the Huojitu trench, Zhugai trench, Dabantu trench and some coal gangue heaps. Before the coal mine integration, there were many small coal mines and small coal washing plants in these southern areas, and there were serious problems of private mining and random stacking of coal, which may be the main reason for the heavy metal pollution of soil in this area.
The areas with low, moderate and heavy soil contamination are mainly in the hot point in the floodplain of the sewage outlet of the Daliuta Coal Washing Plant, suggesting coal washing plant is the most severe threat to soil quality. In addition, some contaminated samples are located in remote villages, indicating that the polluted areas are not only affected by mining activities but also controlled by geological factors.
The pollution load index (PLI) of heavy metal elements in the mine soils varies between 0.45 and 11.72, with an average value of 0.85, which is at the slightly contaminated level. Heavy metals that belong to the higher contaminated groups are those that are more likely derived from anthropogenic sources. Because the proportion of Cd with a CF value greater than 1 is much higher than that of other elements, the critical situation of mine soils in the study area was primarily caused by Cd in comparison with the corresponding background values.
The potential ecological risk assessment
We use the ecological risk index method to estimate the potential ecological risks of heavy metals in mine soils in the Shendong coal base. As shown in Fig.5, on average, ecological risk of the individual ecological risk index (E) decreases as follows: Cd > Hg > Pb > Hg > As > Cu > Ni > Cr > Zn, with higher E values indicating higher ecological risk. Based on the classification standard of E, the investigated mine soils pose a low ecological risk. Specifically, Cd (E = 29.02), Hg (E = 11.90), Pb (E = 4.08), As (E = 3.81), Cu (E = 2.55), Ni (E = 2.31), Cr (E = 1.20), and Zn (E = 0.56). For Cd, a total of 112 samples fall into the low risk level, while 16 samples fall into the moderate risk, and 1 sample falls into the considerable risk. For Hg, a total of 128 samples fall into the low risk level, while 1 sample falls into the extremely high risk level. In addition, the mine soils present a low ecological risk based on the maximum E values for the other heavy metals (As, Cr, Cu, Ni, Pb, and Zn). Therefore, we find that Cd contributes most to the ecological risk levels of heavy metals in mine soils in the Shendong coal base. This result coincides with the previous research conducted by Chen et al. (2015), which indicates that Cd is the primary heavy metal that contributes to soil metal contamination risks in China.
The RI values of mine soils in the study area range from 6.24 to 637.03 with an average value of 18.86, which is at the low ecological risk level except for the sample in the floodplain of the sewage outlet of Daliuta Coal Washing Plant (RI=637.03). These results show that the current concentrations of heavy metals in mine soils of the Shendong coal base do not threaten the ecological security of this basin. However, maize, which is the main crop in the farmland around the mining area, has been shown to have a strong absorption and enrichment capacity for As, Pb and Cd in soil (Sun et al. 2021). Therefore, contamination of mine soils by Cd is of primary concern in the study area, due to the toxicity of this element and its bioaccumulation in plants and animals via the food chain (Taylor and Percival 2001). Based on our estimation results of contamination and ecological risk of heavy metals in mine soils, Cd poses the greatest biological risk to these corn fields.
We depict the spatial distribution patterns of the RI value of the eight heavy metals in the mine soils in Fig. 6. Here, we observe higher RI (>150) values of heavy metals distributed around the “hot point” of the Daliuta washing plant, whereas the RI values in other regions are less than the limits. This may indicate that anthropogenic factors in the coal washing process might be important sources of heavy metal contamination. It is precisely because of the abnormally high content of Hg in the wastewater of the coal washing plant, coupled with the very large toxic response factor of Hg (Ti=40), that the RI value of the hot point is very large. Mercury (Hg) has been ranked third among the most toxic heavy metals and is one of the most widespread environmental contaminants (Clifton, 2007; Chaudhary et al., 2014). Anthropogenic Hg emissions from human activities contribute more than 30% of the earth’s total Hg emissions (Frossard et al., 2018). Hg in contaminated soil has the potential to enter the food chain through plants and animals. Once in the food chain, Hg can be bioaccumulated in the body for a long time, causing adverse effects on human health (Chaudhary et al., 2014). Due to this, Hg was selected as the priority control metal in mine soils of the Shendong mining areas.
Identification of heavy metal source
Heavy metals in topsoils originate from either anthropogenic sources or natural sources. Correlation analysis and principal component analysis have been applied widely and successfully to identify the potential origins of heavy metals (Davis et al., 2009; Mohamed et al., 2016; Xie et al., 2018;). Fig.7 shows the correlation matrix for heavy metals in the soil in the Shendong coal base. The correlation between heavy metal elements is relatively strong, and the correlation coefficient generally exceeds 0.5 with significance at the 0.01 level. We find the strongest correlation between Cr and Ni, Hg and Zn, Cu and Ni, As and Ni and as well as Cd and Pb and the correlation coefficients are 0.92, 0.89, 0.87, 0.87 and 0.85, respectively. Positive correlations between the investigated heavy metals may indicate that they have common sources. In addition, the correlation between Hg and As, Ni and Cr is weak, and the correlation coefficients are 0.15, 0.21 and 0.23, respectively, indicating the different sources between Hg and As, Ni, Cr.
We also apply principal component analysis to distinguish the natural and anthropogenic inputs of the investigated heavy metals in the mine soils. Firstly, KMO and Bartley sphere test is carried out, and the KMO value of the test is 0.812, so the principal component analysis of eight heavy metal elements can be done. Tables 4 and Fig. 8 present the factor loadings matrix. Only the first two components extracted had eigenvalues > 1 and accounted for 89.12% of the total variance in the dataset. Varimax normalized rotation was applied to maximize the variance of the first two principal axes (Fu et al. 2014).
The Ni, As, Cr, and Cu dominate in the first PCA (PCA1) and explain the greatest total variance (61.96%). The maximum concentrations of Ni, As and Cr elements in mine soils in the study area are below the risk screening values and the correlation analysis also shows that Cr and Ni are the strongest positively correlated. Previous research results indicate that Cr, As and Ni are predominantly controlled by the soil’s parent materials through lithogenic and pedogenic soil processes (Ajigul et al. 2017), and we refer to this as the “lithological influence factor”.
The Hg and Zn elements dominate in the second PCA (PCA2) and explain 27.16% of the total variance. Again, the correlation analysis also shows that Hg and Zn elements are among the strongest positively correlated, with significance at the 0.01 level. The highly concentrated areas of these two elements are coincident in the distribution region in the floodplain of the sewage outlet of the Daliuta Coal Washing Plant. Previously, researchers have reported that Hg and Zn in soils are mainly anthropogenic sources (Li et al. 2018). The content of Hg and Zn is only particularly high in the coal washing plant, and the median values are much smaller than the local background value, and the high CV values of the two elements indicate their heterogeneous spatial distribution. Based on the results discussed above, we find that Hg and Zn in mine soils in the study area are from anthropogenic sources, especially from chemicals added in the coal washing process in the Shendong coal base. We refer to this as the “anthropogenic influence factor.
If a single element has considerable loadings on two or more different principal components, then it is more likely to be caused by two or more sources. The loadings of Cd and Pb on PCA1 are 0.77 and 0.69, the loadings on PCA2 are 0.53 and 0.58, respectively, and the spatial distribution of Cd and Pb partially coincides with the spot-like high values of Hg and Zn, indicating that Cd and Pb are probably influenced by both natural and anthropogenic inputs, and we refer to Cd and Pb as the “nature-human compound source”.
Table 4
Variances explained by the various factors
Factor
|
Initial eigenvalues
|
Extraction sums of squared loadings
|
Rotation sums of squared loadings
|
Total
|
% of variance
|
Cumulative (%)
|
Total
|
% of variance
|
Cumulative (%)
|
Total
|
% of variance
|
Cumulative (%)
|
1
|
5.68
|
70.98
|
70.98
|
5.68
|
70.98
|
70.98
|
4.96
|
61.96
|
61.96
|
2
|
1.45
|
18.14
|
89.12
|
1.45
|
18.14
|
89.12
|
2.17
|
27.16
|
89.12
|
3
|
0.32
|
4.02
|
93.14
|
|
|
|
|
|
|
4
|
0.27
|
3.34
|
96.48
|
|
|
|
|
|
|
5
|
0.13
|
1.60
|
98.07
|
|
|
|
|
|
|
6
|
0.09
|
1.09
|
99.16
|
|
|
|
|
|
|
7
|
0.04
|
0.54
|
99.70
|
|
|
|
|
|
|
8
|
0.02
|
0.30
|
100.00
|
|
|
|
|
|
|