Characterization of Heavy Metal Contamination in Groundwater of Typical Mining Area in Hunan Province

This paper mainly carries out the research work by analysing and testing the groundwater quality in typical mining areas, based on the hydrogeochemical investigation data of groundwater monitoring points, and analyses the statistical characteristics of different index factors in antimony mining areas by using Spss and Surfer software. Firstly, using the standard deviation feature analysis, it is found that mining is more sensitive to the impact of As and Sb, and the degree of its pollution is the most serious, and the exceeding rate reaches 100%; later, the results of the analysis through the principal component analysis method show that the �rst principal component a1 has a higher correlation with Sb and Se in the original variables, the second principal component a2 has a stronger positive correlation with Ba in the original variables, and a stronger negative correlation with Mo, and the third principal component a3 has a stronger positive correlation with Mn in the original variables, and the negative correlation between Co and the third principal component is stronger. Finally, the correlation between each heavy metal indicator factor and the other six indicator factors was calculated by Spss software analysis to �nd out the pattern of the order of strength of the correlation of the indicator factors, and it was concluded that the concentration of Se and Mo had a stronger correlation with the other heavy metals in the study area, whereas the correlation of Mn and Co with the other in�uencing factors was lower. This paper summarises and re�nes the statistical characteristics of groundwater hydrogeochemistry in the mining area to provide reference for the diagnosis, prevention and control of groundwater pollution risk in this and similar mining areas.


Introductory
The Problem of Groundwater Pollution has become a major concern for governments, businesses, and the public [1] .There is a growing awareness among people, and water quality safety has increasingly become an important issue for national sustainable development and ensuring public health [2][3][4][5] .
Therefore, conducting relevant research holds signi cant importance for the ecological environment construction and protection of regional groundwater resources [6] .Metal mining areas are one of the main sources of groundwater pollution, characterized by severe heavy metal contamination, which exhibits high toxicity, slow degradation, and ease of accumulation.Heavy metals such as Pb, Cr, Hg, Se, Mn, among others, possess strong toxicity.The cumulative effect of these heavy metal elements in the human body poses signi cant potential risks to the health of regional populations, causing severe harm to human organs [7][8][9] .The main source of these heavy metals is the mining activities that generate enormous value.The resulting tailings, dust, waste rocks, and wastewater from mining operations can lead to the diffusion and in ltration of heavy metals into soil and groundwater, causing severe pollution to groundwater bodies, thus resulting in irreparable harm to human health [10] .
The antimony mining area in the study region has a wide variety of mineral deposits, resulting in a diverse range of heavy metal pollutants.This research aims to provide relevant personnel with a better understanding of the groundwater pollution situation in the mining area.It serves as a scienti c basis for the prevention and control of groundwater pollution in the mining area and the management of public health risks.Consequently, effective measures can be formulated to reduce the health risks posed to residents.

Socioeconomic Overview of the Study Area
The investigation area is situated in the northwest of Lengshuijiang City, Hunan Province, encompassing the Mining Township, the Xikuangshan Administration, and the border area of Zhonglian Township.It lies approximately 13 kilometers south of Lengshuijiang City, with geographical coordinates ranging from 111°25′47″ to 111°31′22″ east longitude and 27°49′28″ to 27°43′05″ north latitude.The antimony mining industry in this region began production in 1897 and has a history of over a century.Mineral production and processing serve as the primary economic sources for the local government and residents, earning the area the titles of "World Capital of Antimony" and "Coal Sea of Jiangnan" [11][12] .Within this mining area, two large ore deposits, one medium-sized ore deposit, and three small-sized ore deposits have been identi ed.The antimony mineral eld in the northern Xikuangshan area is the concentration zone for antimony mineral production, with an accumulated con rmed reserve of 26.265 million tons (metal reserve of 855,202 tons).The reserve is exceptionally abundant, ranking rst in the nation and representing the world's largest antimony mineral eld [13][14] .The total population residing in the mining area is approximately 17,100, comprising approximately 15,000 urban residents and around 2,100 rural villagers [15,16] .
Over the past hundred years of mining operations in the Xikuangshan antimony mining area, the substantial generation of "three wastes"-waste gas, waste residue, and waste water-has caused severe environmental damage to local soil and groundwater.Consequently, it has posed varying degrees of health risks to the local residents [17,18] .As the subject of this investigation, the Xikuangshan mining area demonstrates typicality and representativeness, thus offering outcomes of considerable representational signi cance in the research.

Fault Zone Permeability and Aquifer Characteristics
The fault zones within the area, including the F75, F72, F3 of the North-Northeast group, and the F19, F17, F104 of the North-Northwest group, constitute signi cant and mechanically similar normal faults within the ore deposit.Due to the surrounding rocks being predominantly composed of argillaceous limestone, sandstone, and shale, the development of karsti cation is not favored.The fractured zones within the fault zones consist mainly of fractured mudstones, sandstones, and limestone blocks with a dense cementation, which impedes groundwater movement and storage.Drilling observations within these fault zones did not reveal any water seepage.Water injection tests conducted via boreholes demonstrated minimal water injection rates (0.0014L/S•m), particularly within the main F75 fault, where 14 cross veins in the midsections of tunnels 7, 9, 11, and 13 displayed dry fault surfaces without any indications of groundwater activity.These fault zones are entirely lled with fragmented rocks such as sandstone, limestone, and shale.However, smaller secondary fault zones and ssures were observed in the tunnel exposures, facilitating water storage within the ore deposit, primarily channeling fractured water from silici ed limestone and silici ed rock zones.These secondary faults trend in the northeast and northwest directions, with an inclination exceeding 60°, extending several meters to tens of meters in length, and possessing a maximum width of approximately 1.5 meters.The predominant lithology comprises silici ed limestone, followed by limestone, with ow rates ranging from 0.01 to 0.5L/S.Based on the above description, the fault zones within this mining area exhibit relatively poor water-bearing and waterconducting properties.

Groundwater Flow Status
The Xikuangshan mining area is situated at the hydrological watershed of the aquifer system.To the east, the Yunxi biotite granite forms a hydrological barrier, while the F75 fault acts as the western hydrological boundary.Consequently, it creates a north-south owing aquifer unit, effectively isolating the ore deposit from the regional groundwater.Moreover, a local hydrological watershed exists in the central part of the study area, positioned at Qilijiang, resulting in independent ow divisions between the northern and southern mines.Additionally, surface water bodies distributed within the ore deposit are all relatively small.As a result, the groundwater in the ore deposit primarily relies on in ltration from atmospheric precipitation for recharge.Natural features such as exposed rock cavities, fractures, and old workings become recharge zones for the aquifer.A few unclosed boreholes and surface silici ed limestone fractured zones also contribute to the aquifer recharge.Consequently, the overall recharge area of the aquifer is relatively limited.
The groundwater ow in the aquifer predominantly exists in the form of limestone ssures, karst caves, interlayer fractured zones, and fractured ssures in silici ed limestone.Generally, the groundwater ows from the Qilijiang watershed in the central part of the study area, discharging southwards and northwards from the study area.The runoff distance is relatively short.The primary discharge method is in the form of springs, with only a minimal portion discharged through faults.Extensive mining activities in the Xikuangshan area have led to the formation of arti cial discharge zones in deeper mining regions.
The dynamics of the ore deposit groundwater are primarily characterized by signi cant uctuations in water level and quantity.Since it is predominantly replenished by rainfall, variations in groundwater dynamics are closely related to precipitation.Seasonal changes are evident, with groundwater levels rapidly rising during the rainy season, leading to a sharp increase in ow rates.For instance, observation at Spring No. 7 recorded a ow rate of 2.129L/S before rainfall, surging to 14.13L/S within the initial 30 Sample Collection and Testing Methods

Principles of Sampling Point Layout
Based on the "Technical Speci cations for Groundwater Environmental Monitoring" (HJ164-2020), "Technical Guidelines for Site Environmental Monitoring" (HJ25.2-2014),"Sanitary Standards for Drinking Water" (GB5749-2006), "Groundwater Quality Standards" (DZ/T0290-2017), as well as the "Technical Regulations for National Integrated Water Resources Planning," the following principles for sample point layout were formulated: (1) The southern and northern mining areas belong to different hydrogeological units.Accordingly, sampling points were arranged based on the distinct hydrogeological conditions of these independent units.
(2) Considering the distribution of pollution sources such as leachate from mine waste, ore washing wastewater, and pit drainage, water quality monitoring points were set up at the groundwater recharge, ow, and discharge sections in each hydrogeological unit of the southern and northern areas.The spacing between points generally ranged from 150 to 200 meters.
(3) In addition to the existing monitoring wells, domestic wells, and springs within the mining area, additional groundwater quality monitoring points were selectively added and adjusted based on the actual exposure of groundwater and the conditions of recharge, ow, and discharge.The placement of monitoring points was adjusted according to the speci c conditions, avoiding uniform distribution.

Layout of Monitoring Network
(1) The controlled area of the groundwater quality pollution monitoring network within the mining area covers an area of 30km².
(2) A total of 26 groundwater quality pollution monitoring points were established within the mining area.
In this study, a total of 26 groundwater sampling points were collected within the research area.Their distribution is illustrated in Fig. 1. minutes of rainfall.The monthly amplitude of groundwater levels is generally around 2.40 meters, with an average annual amplitude of 7.6 meters.
Mining activities in the area have moderately impacted the underground aquifer due to groundwater extraction.However, water scarcity has not been observed, and there have been minimal occurrences of wells or springs running dry.Overall, the impact on regional groundwater balance remains relatively slight.

Principles for Selecting Heavy Metal Evaluation Indicators
Multiple characterization factors need to be considered to achieve the groundwater functional assessment's objective.However, the selection of characterization indicators varies widely, and it's not advisable to mechanically or indiscriminately include all factors in the evaluation index system.Doing so might dilute the role of dominant factors, affecting the accuracy of the assessment results.Therefore, it's essential to make an optimal selection of evaluation factors based on principles such as: the principle of dominance, measurability, operability, comprehensiveness, exibility, among others.
The mineral composition in the study area is relatively straightforward, with relatively simple chemical components.Apart from the primary element antimony (Sb), associated elements include As, Hg, Ag, Cu, Pb, Sn, V, Zn, Mo, Ga, and B. Following the classi cation system established by the International Agency for Research on Cancer (IARC) and the World Health Organization (WHO) for assessing the reliability of the carcinogenicity of chemical substances, essential trace elements for the human body in the associated elements within the study area include Mo, Co, Mn; harmful substances comprise As, Ba, Sb; carcinogenic substances include As, non-carcinogenic substances consist of Mn, Se, and potentially carcinogenic substances encompass Co, Sb.
Considering the aforementioned principles and the evaluation objectives of this study, the heavy metals Mo, As, Co, Mn, Ba, Sb, and Se were chosen as evaluation indicators.

Sampling and Testing Methods (1) Groundwater Sampling and Preservation
In accordance with the "Standard Examination Methods for Drinking Water" (GB5749-2006, GB/T5750.1-5750.13-2006), containers were acid-washed or alkali-washed as speci ed before sampling, followed by rinsing with distilled water.The sampling instruments were rinsed with the source water at least three times before sampling.Non-disposable groundwater sampling equipment was used, requiring pre-and post-sampling cleaning.Wastewater generated during the cleaning process should be collected and disposed of properly.For sample bottles without preservatives, they were rinsed with the water to be sampled two to three times before groundwater sampling.Samples containing heavy metals such as As, Mn, Sb were collected in 250ml polyethylene plastic bottles, adjusted to pH less than 2 with nitric acid, and analyzed within 10 days.The frequency of this sampling was once and the sampling period was two days.
(2) Groundwater Sample Testing Methods Adhering to the requirements of the "Standard Examination Methods for Drinking Water" (GB/T5750.6-2006)and "Water Quality-Determination of Barium" (HJ 602-2011) for various indicators, different instruments were equipped to ensure the quality of sample analysis based on the quality level of different analysis methods.Main testing instruments included gas chromatography-mass spectrometer, atomic absorption spectrophotometer, and atomic uorescence spectrometer, among others.The standard testing methods used for various detection components are detailed in Table 1.Through statistical characteristic analysis, one can grasp the apparent features of heavy metal pollution in the underground water of the study area, offering the necessary basis for evaluating the results [23] .In this section, SPSS and Excel software were primarily employed to analyze the experimental detection data of selected heavy metal evaluation indicators, namely Mo, As, Co, Mn, Ba, Sb, and Se.The analysis mainly encompasses parameters like maximum value, minimum value, mean, variance, kurtosis, skewness, and coe cient of variation.This translation could be used for an academic paper.
The minimum and maximum values represent extreme values in the data, indicating the degree of heterogeneity within the dataset.The median is a type of central tendency that describes the typical scenario within a set of data.The mean represents a measure of central tendency within a dataset and serves as an indicator re ecting the trend of the data set.In statistical work, the mean (average) and standard deviation are the two most important measures used to describe the trend and dispersion of data sets.Variance and standard deviation are used to measure the deviation of a random variable from its expected value and are often employed to study the deviation between mathematical variables and their mean in various scenarios.The coe cient of variation, similar to variance and standard deviation, primarily indicates the degree of dispersion within data, in uenced by both the dispersion level of variables and the average level of variable values.The statistical characteristic values of the indicator elements in the study area can be found in Table 2.This translation could be used in an academic paper.

Analysis of Standard Deviation Characteristics of Heavy Metal Evaluation Index Concentrations
From the data analysis results, the standard deviations for Mn, As, Ba, Co, Mo, Sb, and Se are 0.024081, 0.22651, 0.0166, 0.000152, 0.002301, 3.910359, and 0.006, respectively.Among these, As and Sb have relatively large standard deviations, with their maximum concentrations being 22.67 and 1921.25 times the minimum concentrations, respectively.These results indicate a signi cant level of dispersion in the concentrations of As and Sb within the sampled water, suggesting a strong correlation with anthropogenic factors in the surrounding environment, where human in uences predominantly affect these two index factors.In contrast, the variance for the other ve ions is almost zero, indicating that these ve ions are primarily in uenced by natural factors, or a combination of natural and anthropogenic in uences at relatively similar levels.

Analysis of Kurtosis and Skewness Characteristics of Heavy Metal Evaluation Index Concentrations
Under natural conditions, the concentration curves of heavy metal ions generally conform to a normal distribution.Kurtosis and skewness can re ect whether the concentration of certain ions conforms to or deviates from the normal distribution, and the degree of deviation can generally indicate the in uence of anthropogenic factors on these ions.Kurtosis [23] is a measure re ecting the shape of the distribution of a random variable.The kurtosis K of a random variable x is de ned as: When K > 3, it represents a leptokurtic curve, indicating that the variable values are densely distributed around the mode.
When K = 3, it matches the degree of sharpness of a normal distribution curve.
When K < 3, it represents a platykurtic curve, suggesting a relatively uniform dispersion of variable values around the mode.
Skewness measures the degree and direction of the distribution's asymmetry and is a dimensionless value.The skewness S of a random variable x is de ned as: When S > 0, the distribution of x is positively skewed.
When S = 0, it is symmetric around the mode.
When S < 0, the distribution of x is negatively skewed.
From Table 2, it can be observed that the kurtosis of Mn, Mo, and Sb are 12.278 > 3, 3.343 > 3, and 6.056 > 3, respectively, and the skewness values are 3.338 > 0, 1.84 > 0, and 2.54 > 0, respectively.Based on the analysis, it can be inferred that the concentrations of these three ions deviate from a normal distribution, indicating varying degrees of in uence from anthropogenic factors, especially notable for Mn and Sb in uenced by the surrounding anthropogenic environment.Meanwhile, As, Ba, Co, and Se exhibit kurtosis values of -1.591, -0.094, 1.395, and 1.662, respectively, with absolute values less than 3, suggesting platykurtic curves.The skewness values are 0.232, 0.987, 1.617, and 1.727, all greater than 0, indicating positive skewness.Similarly, these three factors are also in uenced to varying degrees by anthropogenic factors.

Analysis of Statistical Characteristics of Heavy Metal Evaluation Index Concentrations
From the perspective of the coe cient of variation, the degree of variability among the seven factors is in the following order: Sb > Mn > Se > Mo > Co > As > Ba.Particularly, the coe cients of variation for As, Se, and Sb exceed 100%, even reaching 200%.The pollution indices of the heavy metal indicator factors are as follows, in descending order: Sb > As > Se > Mn > Ba > Mo > Co. Notably, the pollution index for As reaches 2991%, while Sb is exceptionally high at 35293.2%.From the standpoint of the mean values, antimony and arsenic exhibit severely polluted states.The Table 2 shows that the exceedance rates for As and Sb are 100%, while for Se and Mn, they are 20% and 3.85%, respectively.The exceedance rate for the remaining three indicator factors is 0%.

Analysis of Pollution Characteristics of Heavy Metal Exceedances
To visually depict the pollution status of each sampling point's water samples in both groundwater and surface water for indicators showing exceedance rates, we have generated concentration status diagrams for the prevailing concentrations of Mn, As, Sb, and Se, which have non-zero exceedance rates.These diagrams illustrate a comparison between the concentrations of each exceedance factor against the standard values at each sampling point and depict the contrast among the sampling points, as shown in Fig. 2.
According to the Drinking Water Standard (GB5749-2006), the standard concentration for Mn is 0.1 mg/L.From Fig. 2, it's evident that only Spring 22 in Chuanshan Village of the mining area and surface water points DB1 and DB4 in Dongxia Village of the mining area and Shengli Village in Zhonglian Township respectively, exceed the standard concentration for Mn.Furthermore, it's notable that the Mn content in surface water is generally higher than in groundwater.Essentially, only point 22 exceeds the average Mn concentration in surface water, and only three groundwater sampling points exhibit Mn concentrations higher than those in surface water.
Figure 3 illustrates that both surface water and groundwater within the study area exhibit arsenic (As) concentrations higher than the standard value of 0.01 mg/L for drinking water.Notably, the surface water point DB2 in Tanjia Community, Minshan Town, records an alarming concentration of 11.93 mg/L.
Regarding antimony (Sb), after handling the outliers, spring water point 13 north of Minshan Hospital in Taotang Street, Minshan Town, was identi ed as an outlier and removed.Drainage from hospital facilities such as treatment rooms, laboratories, wards, laundry rooms, X-ray imaging, isotope therapy diagnostic rooms, and operating theaters intensi es the antimony pollution in the groundwater in that area, potentially compromising the accuracy of the water pollution assessment in the research area.This particular data point has been considered an outlier and excluded from the analysis.Figure 4 illustrates that the concentrations of Sb in all sampling points within the research area exceed the drinking water standard value of 0.005 mg/L, indicating severe antimony pollution.The minimum recorded concentration surpasses 0.0084 mg/L, more than 1.68 times the drinking water standard.Moreover, the Sb concentrations in four surface water sampling points across the research area are consistently higher than the median Sb concentration in groundwater.
The standard value for selenium (Se) in drinking water is 0.01 mg/L in Fig. 5, and the exceedance rate at the groundwater monitoring points in the surveyed area is 20%.Water from the Qilijiang Iron Mine (11), in direct contact with the mine area, exhibits signi cant pollution with a concentration of 0.047 mg/L, considered an outlier.After removing this outlier, the analysis reveals some sampling points with selenium concentrations at 0 mg/L, while the maximum concentration reaches 0.021 mg/L, twice the standard value, suggesting a distinctly regional pattern of pollution.Overall, the analysis indicates that arsenic and antimony pollution is most severe in the groundwater of the study area, followed by selenium and manganese, with minimal contamination from barium, cobalt, and molybdenum.

Principal Component Analysis of Heavy Metal Elements
The abundant variety of mineral resources in the study area leads to a multitude of factors affecting the groundwater quality in this region.This paper utilizes SPSS software to conduct a principal component analysis on selected indicator factors.This approach, while minimizing data loss, involves linear combinations and discarding minor details, replacing multidimensional variables with a few comprehensive ones.This method aims to reduce workload while ensuring an accurate assessment of the groundwater quality in the study area.

of Principal Component Analysis
Principal Component Analysis (PCA) [24][25][26][27] is a statistical method for dimensionality reduction, aiming to quantitatively study multiple-dimensional factors within the same system.The fundamental concept lies in the assumption that among numerous correlated factors, there are dominant common factors.By analyzing the internal structural relationships within the correlation matrix of the original variables, this method identi es representative and targeted indicators, facilitating their mutual comparison.

4.2
Where anm represents the eigenvectors corresponding to the eigenvalues of the covariance matrix of the original variable matrix X; ZX1, ZX2, ZXn are the standardized values of the original variable matrix X; λ1, λ2, ..., λp represent the eigenvalues of the ZX matrix; n denotes the number of factors; m represents the number of samples; p stands for the number of principal components.

Results and Analysis
The analysis results of heavy metal indicator factors in water samples at sampling points using SPSS are shown in the Table 3.The initial factor loading matrix displays the correlation coe cient values of each principal component with its corresponding variables.In the principal component matrix, dividing each column vector of the mth component by the square root of the mth eigenvalue provides the coe cients corresponding to each indicator for the mth principal component, i.e., the eigenvectors a1, a2, and a3, as shown in Table 5.The comprehensive assessment index was calculated from the principal component functions and their corresponding eigenvalues, as shown in Table 6.(1) According to Table 3, the eigenvalues of the rst, second, and third principal components are 2.902, 1.306, and 1.119 respectively.The contribution rates of the principal components are 41.463%, 18.650%, and 15.979% respectively, accumulating to 76.093%.This indicates that these three principal components essentially encompass the in uencing factors of heavy metal pollution in the groundwater of the research area.The rst principal component, in particular, contains more information and exhibits a stronger correlation with water quality.
(2) From Table 5, it can be observed that the rst principal component a1 has a higher correlation of 0.493 with the original variables Sb and Se.This suggests that these two factors are the primary contributors to this type of pollution, while the pollution contribution of the other ve indicators is relatively minor.The second principal component a2 shows a strong positive correlation of 0.656 with the original variable Ba and a signi cant negative correlation of -0.485 with Mo, indicating a relatively high contribution of Ba and Mo to this principal component.The third principal component a3 has a positive correlation of 0.579 with the original variable Mn and a strong negative correlation of -0.705 with Co, suggesting a signi cant contribution of Mn and Co to a3.
(3) Based on the comprehensive principal component score F, Table 6 is obtained.A higher F value indicates more severe heavy metal pollution.Among these, sampling points 11, 12, and15 rank in the top three, signifying the most severe heavy metal pollution at these locations.On the other hand, sampling points 25, 22, and 26 rank at the bottom three, indicating the least polluted water samples among the research sampling points.
(4) The distribution of heavy metal pollution shows that the less polluted monitoring points 26, 22 and 25 are located in the upstream area of the groundwater, and the most seriously polluted points are mainly distributed in the middle of the mining area or the downstream area of the groundwater.It means that the background value of heavy metal ions in the investigation area is small, and the heavy metal ions in the mining area and the surrounding groundwater mainly come from the mining of tin, antimony and iron ore, and the mining of minerals is the main source of heavy metal ions in the groundwater.

Conclusion
(1) Among the evaluation indicator factors of groundwater monitoring points in the research area, the standard deviation analysis re ects that As and Sb are more sensitive to the in uence of mining activities.Mn, Se, Mo, Co, and Ba are primarily in uenced by natural geological and hydrogeological conditions, with less impact from human activities.Skewness and kurtosis values of heavy metal evaluation indicators reveal varying degrees of anthropogenic environmental impact on the seven factors.The pollution sources of Mn, Mo, and Sb are relatively concentrated, while those of As, Ba, Co, and Se are more dispersed.The exceedance rate for As and Sb reaches 100%, with Se and Mn exceeding by 20% and 3.85%, respectively, while the exceedance rate for the remaining three indicator factors is 0.
(2) The severity of pollution in the research area's groundwater is highest for As and Sb, followed by Se and Mn.In contrast, Ba, Co, and Mo's pollution levels among these three heavy metals have not  The sampling point distribution in the study area

Figure 2 Current
Figure 2

Figure 3 Current
Figure 3

Table 1
Test method of heavy metal evaluation index factors for groundwater samples evaluating heavy metal pollution in underground water, it is essential to analyze the statistical characteristics and distribution patterns of indicator elements in the study area using statistical methods. Before

Table 2
Statistical Characteristics of Heavy Metal Indicator Factors

Table 3
Evaluation index system correlation coe cient matrix eigenvalues and contribution rate

Table 6
Principal Component Analysis Results.