Characterizing and tracing the heavy metals' spatial distribution in karst surface river affected by acid mine drainage

Shanshan Zou; Pu Liu; Jingyi Wang; Hao Lu; Yingdan Zhang; Yidan Wang; Bo Li

doi:10.4491/eer.2024.180

Environ Eng Res > Volume 30(2); 2025 > Article

Zou, Liu, Wang, Lu, Zhang, Wang, and Li: Characterizing and tracing the heavy metals’ spatial distribution in karst surface river affected by acid mine drainage

Research

Environmental Engineering Research 2025; 30(2): 240180.

Published online: April 30, 2025

DOI: https://doi.org/10.4491/eer.2024.180

Characterizing and tracing the heavy metals’ spatial distribution in karst surface river affected by acid mine drainage

Shanshan Zou¹, Pu Liu^1,^2,^†, Jingyi Wang¹, Hao Lu¹, Yingdan Zhang¹, Yidan Wang¹, Bo Li¹

¹College of Resources and Environmental Engineering, Guizhou University, Guiyang Guizhou 550025, P.R. China

²Key Laboratory of Karst Georesources and Environment, Ministry of Education, Guizhou University, Guiyang Guizhou 550025, P.R. China

^†Corresponding author: E-mail: pliu6@gzu.edu.cn, Tel: +8613618501696 Fax: +8613618501696, ORCID: 0000-0002-8342-0013

Received March 25, 2024 Revised June 25, 2024 Accepted September 4, 2024

This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/3.0/) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

Abstract

A series of pollution problems have been caused by the drainage of closed coal mines. This study focuses on the XZ River in Guizhou Province, employing a comprehensive approach involving hydrochemical analysis, mathematical statistics, and one-dimensional water quality modeling to investigate the spatial evolution and source apportionment of heavy metals in karst surface rivers affected by AMD. The average pH, EC, and Eh values of groundwater and surface water in the XZ river basin were 6.27, 331 μs/cm, 230 mV and 3.86, 1671.57 μs/cm, 412.71 mV, respectively, indicating severe pollution of surface water by AMD. The study reveals that groundwater in the XZ River Basin is predominantly of the HCO₃-Ca-Mg type, shifting to HCO₃-SO₄-Ca-Mg type with the influx of acid mine wastewater. Utilizing statistical methods for source apportionment analysis indicated that downstream concentrations of Fe, Mn, Zn, Cu, As, and Sr most likely originate from tributaries. Further analysis through one-dimensional water quality simulation confirmed that downstream concentrations of Fe, Mn, Zn, Cu, As, and Sr are attributed to tributaries. This suggests that XZ River is affected by pollution from different coal mines, primarily from coal mine M4, followed by coal mine M1.

Keywords: Acid Mine drainage (AMD), Heavy metals, Karst surface rivers, Spatial evolution, Simulation prediction

Graphical Abstract

Keywords: Acid Mine drainage (AMD), Heavy metals, Karst surface rivers, Spatial evolution, Simulation prediction

Introduction

In recent years, with the depletion of shallow coal resources, coal mines of various scales and in different regions of China have been gradually facing closure and consolidation. The closure of these coal mines has resulted in a series of serious pollution issues. In the Xingren coal mining area of Southwest Qiannan Prefecture, Guizhou Province, AMD (acid mine drainage) has caused significant, widespread pollution to surface rivers and other water bodies, manifesting in orange-yellow or reddish-orange-colored waters. The region exhibits high ecological vulnerability. Guizhou Province is situated in the core area of the continuous karst distribution in Southwest China, where surface rivers, underground rivers, and karst springs are extensively distributed throughout the region. The increasing number of closed coal mines is inevitably impacting the water quality of the karst aquifer systems, which are highly sensitive and vulnerable. The influence of AMD on the environmental quality of karst water systems continues to expand in scope and severity, threatening regional water ecological security and causing adverse social impacts. Therefore, there is an urgent need to elucidate the evolution patterns of water environments and propose pollution warning technologies.

Similarly, many countries around the world are facing similar problems. For example, coal mining areas such as Rajrappa in India [1], Yatani in Japan [2], and Wiśniówka [3] in southern Poland have caused severe damage to local water resources and the water environment. AMD is formed in abandoned coal mines primarily by the exposure of minerals, mainly pyrite, to oxygen in the presence of water. This formation occurs through the catalytic action of Fe²⁺ and microorganisms. In the AMD environment, the primary source of heavy metals is the oxidation and dissolution of sulfide minerals. Ren et al. [4] found that the formation of secondary iron minerals during the mixing of acid mine drainage (AMD) with karst river water plays a crucial role in the migration of elements, Schwertmannite exhibits strong adsorption and co-precipitation capabilities for Mn, Cr, Cu, and As. The geochemical behavior of heavy metals includes processes such as oxidation, reduction, adsorption, co-precipitation, desorption, and dissolution [5]. In karst regions, the discharge of AMD from coal mines affects the hydro-chemical characteristics of rivers, with significant variations in heavy metal concentrations observed during different seasons [6]. The water sources contributing to AMD formation are complex and include underlying karst water, coal-bearing fracture water, overlying loose pore water, surface water, precipitation, etc. These various types of water may enter the goaf in different ways [7, 8]. Water environmental pollution in karst areas has been categorized based on pollution pathways into four types: direct discharge from coal mine openings, indirect discharge through karst conduits, pollution through fractures and faults, layered pollution, and combined pollution. Li et al. [9] discovered severe pollution persists near the closed mines in karst regions, including significant pollutants such as Fe, Mn, Cd, Al, F, and SO₄²⁻. Currently, further investigation is needed into the migration characteristics of heavy metals in surface rivers affected by acid mine drainage (AMD) in karst regions. Due to the frequent interchange between surface water and groundwater in karst areas, the concentration of heavy metals in contaminated karst rivers varies. The concentration of dissolved Zn and Cd peaks in the loss segment of karst groundwater, while Pb and Cu peak in the gain segment of the rivers. Qin et al. [10] found that the pH of karst aquifers contaminated by AMD did not increase; instead, there was an increase in trace element concentrations (Zn, Pb, Ni, Mn, Cu, and As), and elevated sulfate levels indicated the release of heavy metals into the karst aquifer. Hydraulic transport is a primary driving force for the migration of heavy metals. In areas with gentle river slopes, sensitive points with higher concentrations of heavy metals can be formed [11]. Under low-flow conditions, heavy metals may quickly stagnate in karst aquifers or precipitate in the sediment of rivers, including iron hydroxide and carbonate minerals. The migration of heavy metals is restricted in intermittent rivers with low or no flow [12]. Characterization through X-ray Absorption Spectroscopy (XAS) indicates that iron in suspended particulate matter (SPM) initially aggregates into small Fe (III) octahedral clusters, which then precipitate along the river as goethite [13]. In rivers contaminated by AMD, the initial mineral compositions of suspended particles are dominated by schwertmannite. As the river migrates, this gradually transforms into more stable minerals like aragonite and goethite, with these mineral compositions and transformations determining the distribution characteristics of heavy metals [14]. At pH < 3, secondary iron minerals such as jarosite and scorodite, which are strongly crystalline, dominate. Heavy metals in the sediment are mainly in the form of crystalline iron-bound and residual states, exhibiting strong stability. At pH > 3, secondary iron minerals include schwertmannite and ferrihydrite, which are less crystalline. In this case, heavy metals in the sediment are primarily in the form of surface adsorption and weakly crystalline iron-bound states, resulting in relatively lower stability of heavy metals [15]. Studies by Song [16] and others have found that the apparent structure and specific surface area of secondary iron minerals are the main factors influencing the concentration of heavy metals in water. Manganese (Mn) is present in acidic environments with lower pH, exhibiting autocatalytic properties and possessing high adsorption capabilities for metals such as Cr, Cu, Pb, and Zn [17]. Acidic conditions (pH = 3) facilitate the release and migration of Cd and Pb from minerals (pyrite, chalcopyrite), transforming them from residual states to non-residual states [18]. The precipitation of pyrite may restrict the mobility of Fe, As, and V, and to a lesser extent, limit the mobility of other trace metals, retaining them in waste heaps and riverbeds within mining sites [19]. Similar to aragonite and schwertmannite, goethite and jarosite have high specific surface areas and strong adsorption capacities, playing a crucial role in controlling the migration behavior of heavy metals [20]. During the neutralization process of AMD, heavy metals may be adsorbed by water iron minerals, and gypsum in suspended particles. These particles play a significant role in transporting heavy metals downstream, ultimately depositing on riverbeds, leading to an increase in heavy metal content in sediments and a decrease in waterborne heavy metal concentrations. Chen et al. [21] discovered that the dissolution of carbonate minerals and ion exchange are the primary hydrogeochemical processes leading to rapid decreases in heavy metals such as Sb and As in AMD in karst regions. The major ion chemistry of groundwater is largely determined by its carbonate background. Under these conditions, the acidification resulting from sulfide oxidation due to the presence of lead-zinc ores may be hindered by reactions with carbonate. Previous studies have primarily focused on AMD pollution from individual mines. This study will emphasize multiple mine pollution sources and discuss the characteristics of heavy metal migration and pollution in downstream rivers affected by these mines. Therefore, this research holds significant scientific and practical implications.

The study area is located in the karst region of Guizhou Province, China, with a total of 5 closed coal mines. The AMD generated from these mines is directly discharged into surface rivers, with 2 of the mines serving as headwaters, impacting the downstream areas. Therefore, research has been conducted on the discharge of AMD in the XZ river basin. The objectives of the study include: (1) Determining the impact of AMD on surrounding water bodies and the evolution process of groundwater; (2) Analyzing the spatial distribution characteristics of heavy metals in the study area; (3) Simulation analysis of downstream heavy metal sources.

Materials and Methodologies

2.1. Water Sampling

The research area is located in Guizhou Province, China (Fig. 1). The XZ River is a primary tributary of the Masha River and a secondary tributary of the Beipan River. It originates from Huangjiawanzi (longitude 105°11′, latitude 25°34′, elevation 1470 m), flowing initially northeastward, then turning southeastward, and subsequently changing direction northeastward again after passing through Dianmu. It flows through Mawo Valley to Tongzilin (longitude 105°20′, latitude 25°27′, elevation 835 m) on the left bank, where it joins the Boyang River and eventually enters the Beipan River. The source is 27 km from the river mouth, with a main river length of 23 km, a drop of 635 m, an average gradient of 2.76%, and a watershed area of 86.9 km². The research area experiences a subtropical humid monsoon climate with mild temperatures, abundant rainfall, moderate sunlight, and a simultaneous occurrence of rain and heat. However, due to significant elevation differences and complex topography, the local climate exhibits complexity and diversity. The climate undergoes noticeable vertical changes in the high plateau mountains and deep-cut river valleys, leading to variations in precipitation. The average annual temperature in the study area ranges between 12 to 18°C, with a maximum temperature of 34.6°C and a minimum temperature of −7.8°C. July is the hottest month with an average temperature of 22°C, while January is the coldest with an average temperature of 6.4°C, resulting in an annual average temperature of 15.2°C. Precipitation in the study area is abundant, with over 80% of the area receiving annual precipitation between 1100 to 1300 mm. Moreover, the relative variation in precipitation between years is small, making it one of the regions with relatively stable precipitation patterns nationwide. The average annual rainfall is 1327.9 mm, with the highest annual rainfall reaching 1887.6 mm and the lowest only 960.4 mm. The period from May to September experiences abundant rainfall, accounting for over 80% of the annual rainfall. The average annual evaporation is 1368.1 mm, with an average relative humidity of 81%. The wet season occurs from May to September, accounting for over 80% of the annual rainfall. The annual average evaporation is 1368.1 mm, and the annual average relative humidity is 81%. The region experiences a high frequency of rainy days, and concentrated precipitation from May to October, known as the rainy season. During this period, precipitation in various locations contributes to over 75% of the annual total, characterized by sporadic heavy rain with high intensity. From November to April of the following year, precipitation significantly decreases, mainly consisting of light rain. December to March has the least precipitation, accounting for only about 5% of the annual total, marking the dry season. The study area is located in a plateau mountainous region with well-developed valleys and significant terrain fluctuations, characterized as a karst area with abundant water sources. Groundwater primarily derives from atmospheric precipitation infiltrating through karst fissures, structural fractures, and interlayer fissures. The exposed strata consist of the Middle Permian Maokou Formation (P₂m), comprising gray medium-bedded bioclastic limestone, bioclastic limestone, and dolomitic banded limestone; and the Upper Permian Longtan Formation (P₃l), composed of gray to dark gray thin to medium-bedded sandstone, mudstone, interbedded chert, minor bioclastic limestone, and coal. The overlying strata belong to the Quaternary (Q) and sporadically consist of gray gravelly sand, sand, clay, subclay, and humus soil, with angular unconformities typical of karst landforms. The coal-bearing strata in the area belong to the Upper Permian Longtan Formation, consisting predominantly of medium to high sulfur anthracite. Mine waste mainly comprises crushed stone from vein and transport tunnel construction, characterized by high porosity and strong permeability, posing significant environmental hazards through slag and wastewater [22].

This study collected water samples in the XZ River basin of the research area in August 2022, comprising 5 groundwater samples and 14 surface water samples. Collect surface water and groundwater near the sampling area using 550 ml polyethylene bottles. Prior to sampling, rinse the polyethylene bottles with water from the sampling point. Collect two samples of 550 ml each. Conduct on-site analysis with one bottle and transport the other to the laboratory for processing and analysis after filtration. For ease of study, the research area was divided into upstream (S1~S7), tributaries (Z1~Z5), and downstream (X1~X7), with sampling points illustrated in Fig. 1. The surface water samples include two coal mine discharge points, S2 and Z1, as well as three points downstream of coal mines, S5, Z2, and X3. S2 and Z1 correspond to discharge points of coal mines M1 and M3, respectively. S5 and X3 are points downstream of coal mines M2 and M5, while Z2 is the point where coal mines M4 and M3 converge, representing a major source of AMD pollution.

2.2. Approaches

The collected water samples were processed step by step for on-site parameter testing and laboratory analysis (Table S1). Two 550ml polyethylene bottles were used to collect water samples from each sampling point in the study area after thorough cleaning. On-site parameters including temperature (T/°C), oxidation-reduction potential (Eh/mV), electrical conductivity (EC/μS/cm), and pH were tested using a portable water quality analyzer (WTW, multi2620, Germany). The bicarbonate ion (HCO₃⁻) was titrated on-site using the field titration method with methyl orange as an indicator, titrating with 0.1 mol/L dilute hydrochloric acid to determine the HCO₃⁻ content. Flow rate measurements were conducted using an LX-C (China) universal flow rate meter. The water samples were filtered through a 0.45 μm acid-washed membrane filter, then acidified and stored at a temperature below 4°C before being transported back to the laboratory for analysis of relevant ions. Initially, standard solutions of Ca²⁺, Mg²⁺, Na⁺, and K⁺ with certain concentration gradients were prepared, and the main cations Ca²⁺, Mg²⁺, Na⁺, and K⁺ in the water samples were tested using an inductively coupled plasma atomic absorption spectrometer (ICE-3500, USA). Standard solutions of Cl⁻, SO₄²⁻, NO₃⁻, and Fwere prepared using a mixed ratio method, and the main anions Cl⁻, SO₄²⁻, NO₃⁻, and F⁻ in the water samples were tested using an ion chromatograph (ICS-1100, USA). Finally, standard solutions of heavy metals such as Fe, Mn, Sr, and Cu were prepared using a mixed standard solution, and the heavy metal elements Fe, Mn, Sr, and Cu in the water samples were tested using inductively coupled plasma mass spectrometry (ICP-MS, USA). To quantitatively analyze the anions, cations, and heavy metals in aqueous solutions using (ICE-3500, USA), (ICS-1100, USA), and (ICP-MS, USA). Measurements were conducted only if the R² of the standard solution concentration curve prepared before testing exceeded 0.995. After completing all ion tests, statistical analysis was performed using SPSS 27. Sampling points and geological background maps were generated using ArcGIS 10.8. Processed data were plotted using Origin 2018.

Constructing 95% confidence intervals with mathematical statistics and noticing that, for a parameter confidence level of 1 − α, the confidence interval is a_i, b_i, there is p(a_i < b_i|x) = 1 − α. Create a t~distributed random variable

T = \frac{\bar{X} - μ}{\frac{S}{\sqrt{n}}}

with n − 1, T is the value of the heavy metal concentration containing the unknown parameter μ, obeying a T~t_n₋₁~distribution, and checking the t~distribution table yields the

\frac{α}{2}

~upper side quantile

t_{n - 1} (\frac{α}{2})

of the ^t-distribution with n − 1 degrees of freedom, such that:

(1)

P {\bar{X} - t_{n - 1} (\frac{α}{2}) \frac{S}{\sqrt{n}} ≪ μ ≪ \bar{X} + t_{n - 1} (\frac{α}{2}) \frac{S}{\sqrt{n}}} = 1 - α

where n is the number of randomizations, S is the standard deviation, and X̄ is the arithmetic mean. The 95% confidence interval for μ, assuming α = 0.05, is as follows:

(2)

P [| \frac{\bar{X} - μ}{\frac{S}{\sqrt{n}}} | \geq t_{α / 2} (n - 1)] = α

The study area was split into upstream-downstream and tributary-downstream combinations after confidence intervals were obtained. The 95% confidence interval predictions for these combinations were then found by applying a polynomial fit to the heavy metal concentrations.

A one-dimensional water quality model can describe the variation of pollutants in a river water body. XZ River belongs to a small-sized river, where pollutants can reach a state of nearly uniform mixing within a short period of time at cross sections. Therefore, a one-dimensional water quality model can be used to simulate the longitudinal migration of pollutants in the river. The equation for the one-dimensional water quality model is as follows:

(3)

\frac{\partial (A C)}{\partial t} + \frac{\partial (Q C)}{Q x} - \frac{\partial}{\partial x} (A E_{M} \frac{\partial C}{\partial x}) + A k_{1} = S_{m}

In the equation, C represents the pollutant concentration; A is the cross-sectional area of the river channel; Q is the cross~sectional flow discharge; E_M is the longitudinal dispersion coefficient of the river segment; k₁ is the degradation rate constant of the pollutant; S_m represents the source and sink terms; x,t represent distance and time, respectively.

Results

The water pH, electrical conductivity (EC), oxidation-reduction potential (Eh), and major ion concentrations serve as fundamental physicochemical parameters, reflecting the basic hydrochemical properties and characteristics of the water body. The results of different physical-chemical parameters and hydrochemical compositions of water samples in the study area are presented in Table 1.

3.1. Spatial Variability of pH, Eh, and EC

A low pH value can enhance the dissolution of minerals, determining the solubility of heavy metals. As shown in Fig. S1 (a), the pH value in the upstream at S1 is 7.42. However, the influence of acidic mine water discharged from mine M1 causes a significant decrease in pH at S2 to 2.17, indicating a pronounced change. As the water moves downstream, the pH gradually increases due to dilution effects. In the tributaries, the pH values for Z1 and Z2 are 6.47 and 2.49, respectively. Z1 represents water discharged from coal mine M3, while Z2 is a sample formed by the convergence of water from coal mine M4 with Z1. The sharp decrease in pH at Z2 indicates that coal mine M4 is the primary cause of the pH reduction in the tributary after the convergence with Z1. The average pH value in downstream water samples is 4.51, indicating an overall acidic condition. In the upstream, S7 and tributary Z5 have pH values of 4.15 and 2.54, respectively. The convergence point X1 has a pH of 2.88, and at the last downstream sampling point X7, the pH is 3.24, lower than the pH at upstream S7. This suggests that the overall acidity in the downstream is primarily influenced by contributions from the tributaries. In summary, the acidity of the XZ river basin is mainly attributed to the impact of AMD discharged from coal mine M4 in the tributary. Additionally, the acidic mine water discharged from coal mine M1 the upstream contributes, while the influence of coal mines M1, M2, and M5 is less pronounced. According to Fig. S1 (e), the relationship between pH and SO₄²⁻, the trend of pH values collected from water samples is inversely proportional to the concentration of SO₄²⁻. Lower pH values correspond to higher SO₄²⁻ concentrations, indicating that the pH of the water body is primarily influenced by acidic mine water.

The electrical conductivity (EC) reflects the composition and concentration of dissolved inorganic substances in the water. The presence of high concentrations of SO₄²⁻ and high electrical conductivity in acidic mine water is a primary factor leading to changes in the structure of microbial communities in sediment [23]. Water sample points S2, Z1, and Z2, significantly influenced by coal mines M1, M3, and M4, have electrical conductivities (EC) of 6270, 937, and 2710 μS/cm, respectively. As shown in Fig.S1 (b), the range of EC values in upstream water samples varies widely, ranging from 233 to 6270 μS/cm. The tributaries exhibit relatively smaller variations in EC, ranging from 278 to 2710 μS/cm. The downstream water samples, on the other hand, demonstrate relatively stable EC values within the range of 134.5 to 570.8 μS/cm. As shown in Fig. S1 (d), it is evident from the relationship between the pH and EC of the water samples that they are inversely proportional. This is because low pH facilitates the release of ions (such as metallic elements). This is evident in the substantial variation in EC values observed in upstream samples, reflecting the impact of pH on ion release. Compared to downstream water samples, the water samples influenced by coal mines exhibit significantly higher EC values, indicating a pronounced impact of AMD on the pH and EC of the water body. As depicted in Fig. S1 (c), the oxidation-reduction potential (Eh) values in the XZ river basin range between 63.4 and 647.6 mv. Following the influence of AMD, Eh in the water environment reflects the concentration trends of the redox potential Fe²⁺/Fe²⁺ and the presence forms of dissolved iron as the main component. Eh-pH is used to describe the stability of specific metals under different pH and potential conditions, demonstrating the possible oxidation states and dissolution forms of metals under various conditions [24, 25]. As shown in Fig. S1 (f), the samples are located in a relatively dispersed region, but all exist in the form of Fe²⁺. The distribution range of Eh ranges from 63.4 to 647.6 mv, all of which contain Fe²⁺ in the form of reducing agent, indicating the relatively stable water quality in the study area. It is worth noting that Fe exists in the form of Fe (OH)₃ in one of the samples, which may have been formed when iron ions were mixed with precipitants such as sodium hydroxide (NaOH) or ammonium hydroxide (NH₄OH).

3.2. The Hydrogeochemical Processes of The Watershed

As shown in Fig. 2, in the cationic triangle, most water sample points fall in the direction of Mg²⁺, with only a small number located in the central region of Ca²⁺. This indicates that the study area is primarily dominated by Ca²⁺ and Mg²⁺, and these ions may originate from the weathering and dissolution of silicates, carbonates, and evaporites, or the dissolution of minerals such as pyrite in small amounts. In the anionic triangle, most upstream and downstream points fall in the direction of CO₃²⁻+HCO₃⁻, with a small portion falling in the central region of SO₄²⁻. In contrast, most tributary water samples fall in the position of SO₄²⁻. Considering the type of analysis of water samples, those falling in the central region of CO₃²⁻+HCO₃⁻ are primarily groundwater, followed by surface water. One water sample located near the coal mine falls in the direction of CO₃²⁻+HCO₃⁻, while the majority of surface waterfalls are in the direction of SO₄²⁻. Through hydrochemical analysis, groundwater in the XZ river basin is primarily characterized as HCO₃-Ca-Mg type water. After merging with AMD, the groundwater transforms HCO₃-SO₄-Ca-Mg type water.

According to the law of conservation, relationships between elements in water can be deduced. Elemental ratios can reflect material sources and geochemical processes. Exploring the hydrochemical types, influencing factors, and formation mechanisms within the watershed involves analyzing the interrelationships between elements and the chemical equations of reactions. Table S2. presents the chemical equations corresponding to the equivalent ratios between different ions and the types of acids involved in the reactions. Ca²⁺, Mg²⁺, HCO₃⁻, and SO₄²⁻ are the predominant chemical constituents in the watershed’s water bodies, playing a crucial role in the evolution of water quality in the region. Analyzing the correlations between these components, as well as relationships with other parameters, provides a better understanding of the formation and evolution processes of water bodies in the study area.

The equivalent ratio of SO₄²⁻/HCO₃⁻ in groundwater is approximately 0, while the equivalent ratio of (Ca²⁺+Mg²⁺)/HCO₃⁻ is approximately 1. In the upstream, tributaries, and downstream, the equivalent ratios of (Ca²⁺+Mg²⁺)/HCO₃⁻ and SO₄²⁻/HCO₃⁻ continue to increase, indicating that the water-rock interaction processes in the watershed are mainly influenced by H₂CO₃ participating in the weathering of carbonate rocks. In the upstream, tributaries, and downstream, the dissolution of gypsum minerals is evident to balance the anionic and cationic species in the water, gradually becoming dominant. If the sulfate in the samples is primarily derived from the dissolution of gypsum, the molar concentration ratio of SO₄²⁻ to Ca²⁺ should be 1. The observed ratio of SO₄²⁻ to Ca²⁺ in the watershed is close to 1:1, suggesting that there is no other source of SO₄²⁻ in the groundwater samples, and it is solely derived from the dissolution of gypsum [26]. Unlike samples closer to downstream coal mines, the water samples fall on the right lower side of the line with an SO₄²⁻/HCO₃⁻ ratio of 1, indicating the presence of other sources of SO₄²⁻ in the water, possibly originating from the oxidation of minerals such as pyrite in coal mines. When both H₂CO₃ and H₂SO₄ participate in the dissolution of carbonate rocks, the ratio of (Ca²⁺+Mg²⁺) to (HCO₃⁻+SO₄²⁻) is 1:1. The groundwater in the study area satisfies this ratio, indicating that the evolution of hydrochemical types in the aquifers is primarily influenced by the dissolution of carbonate rocks and gypsum-bearing rocks [27]. As shown in Fig.S2, in the absence of AMD influence, the ratio of (Ca²⁺+Mg²⁺) to (HCO₃⁻+SO₄²⁻) in water is plotted on the 1:1 line [28]. The samples closer to coal mines are mainly distributed above the 1:1 line, while those in the upstream, tributaries, and downstream are mostly distributed below the 1:1 line. This suggests that the hydrochemical composition of water in the study area is influenced not only by the weathering and dissolution of carbonate rocks but also by AMD generated by coal mining activities.

Comparisons between different lithological units can help assess the impact of weathering of different rocks on solute concentrations in the watershed. The molar ratios of Mg/Na and Ca/Na, Mg/Ca, and Na/Ca are commonly used to differentiate the influences of solutes in water. Typically, groundwater in limestone aquifers exhibits a relatively high Mg/Ca ratio and a relatively low Na/Ca ratio. In groundwater from shale aquifers, both Mg/Ca and Na/Ca ratios tend to be lower. The Na/Ca ratio in groundwater from clastic rock aquifers is generally higher compared to limestone and shale aquifers.

As shown in Fig. S3 (ab), the upstream and downstream in the study watershed are predominantly distributed between carbonate rocks and clastic rocks, indicating that the control of the hydrochemical characteristics in the upstream and downstream is mainly influenced by the mixed weathering of carbonate rocks and clastic rocks. Meanwhile, tributaries are primarily distributed near the carbonate rock side, suggesting that tributaries are associated with certain carbonate rocks, and their hydrochemical characteristics are mainly influenced by carbonate rock dissolution, controlling the chemical composition of the tributaries. The ratios of Mg/Ca and Na/Ca can differentiate which type of carbonate rock is involved. From the diagram, it is evident that the hydrochemistry of both upstream and downstream in the watershed is simultaneously influenced by the weathering of limestone, dolomite, and silicate rocks. Tributaries, on the other hand, are distributed between limestone and dolomite and away from evaporites and clastic rock end-members, indicating that they are primarily influenced by the weathering of limestone and dolomite within the watershed. The Ca/Mg ratio can distinguish the primary sources of ions. When the Ca/Mg ratio is greater than 1, it indicates that calcite dissolution is predominant; when it equals 1, dolomite dissolution is predominant, while when it is less than 1, it may be influenced by other factors, such as gypsum. As shown in Fig. S3 (c), the Ca/Mg ratio is less than 1, indicating that gypsum dominates the dissolved minerals during the dissolution process.

3.3. Characteristics of Heavy Metal Evolution

The heavy metal elements present in the water samples collected within the XZ river basin include Fe, Mn, Cu, Zn, As, Sb, Pb, Sr, and Se. Statistical analysis was conducted on these heavy metals using indicators such as mean, minimum value, maximum value, coefficient of variation, etc., and the results are shown in Table 2.

In the study area, the overall concentrations of heavy metals were highest at sampling points S2 and Z2, aligning with their lower pH and higher SO₄²⁻ concentrations. This indicates that the heavy metal content in the water of the study area is primarily influenced by mining activities at upstream coal mines M1 and M4. The concentration of antimony (Sb), however, follows a different pattern, with the highest concentration observed at the upstream point S1 and gradually decreasing downstream, also exhibiting lower concentrations in tributaries. This suggests that the Sb content generated from coal mines M1 and M4 is relatively low.

In the Ficklin diagram (Fig. S4) [29], the water bodies in the study area are primarily classified into four types: high acidity high metal, low acidity low metal, acidic low metal, and near-neutral low metal. Due to the input of AMD, waters in the mining area exhibit characteristics of high acidity and high metal content. After mixing and precipitation, the ion concentrations rapidly decrease in the dispersion zone waters, accompanied by lower pH values. As the water flows downstream from upstream coal mines M1 and tributary coal mines M3 and M4, surface water transitions from acidic low metal to near-neutral low metal. During this downstream migration, the buffering effect leads to a gradual reduction in heavy metal concentrations and an increase in pH.

3.3.1. Spatial variation of heavy metals

In the study area, pollution sources are primarily located in the upstream and tributary headwaters, serving as the starting points for the migration of heavy metals. As water flows downstream, the spatial density of heavy metals changes, as illustrated in Fig. 3. The concentration distribution of the heavy metal Sr differs from other heavy metals. Its density exhibits irregular variations from upstream and tributaries to downstream, distributed across various regions. Higher Sr content is mainly concentrated at sampling points where groundwater discharges. On the other hand, the density variations of other heavy metals (Fe, Mn, Cu, Zn, As, Pb, Sb, and Se) are generally consistent. The closer the location is to coal mines M1 and M4, the denser the distribution of heavy metal concentrations, gradually decreasing as they migrate downstream. An increase in density is observed downstream, primarily influenced by coal mine M5, which releases some heavy metals into the river.

In the upstream water flow, the pH decreases from 7.42 to 2.17 as it moves from S1 to S2, accompanied by a significant increase in the concentrations of elements Fe, Mn, Cu, Zn, As, Se, and Sr. This is attributed to the supply of acidic mine wastewater from coal mine M1. The variation in heavy metal concentrations in the water is primarily influenced by dilution effects from tributaries and groundwater along the way, as well as the adsorption and co-precipitation effects of secondary mineral deposits [30]. As the water migrates downstream, pH increases, and at S3, there is a noticeable decrease in the concentrations of elements Fe, Mn, Cu, Zn, As, and Se. During this process, Fe is prone to hydrolysis, forming oxides or hydroxides of iron. AMD discharged contains a high concentration of SO₄²⁻, and during the dilution process, Fe precipitates as hydroxysulfate minerals, such as schwertmannite (Fe₈O₈(OH)₆SO₄·nH₂O) and jarosite (KFe₃(SO₄)₂(OH)₆) [31]. The rate differences in schwertmannite formation result in varying adsorption capacities for As. Slowly forming schwertmannite exhibits stronger adsorption capacity for As (V), positively correlated with mineral surface area [32]. The formed oxides, hydroxides, and sulfate secondary minerals of Fe tend to adsorb onto suspended particles or precipitate into riverbed sediments, leading to a decrease in the concentration of Fe in the water [33]. Similarly, the oxides and hydroxides of Fe exhibit certain adsorption capacities for other heavy metals. As the concentration of Fe decreases, it implies that other toxic elements are also adsorbed onto suspended particles or precipitated into sediments. Additionally, dilution from tributaries and groundwater further decreases the concentrations of metal ions in the water. The precipitation of sulfate secondary minerals with adsorption capabilities is accompanied by the co-precipitation of other heavy metals, resulting in a trend of decreasing concentrations of most heavy metals along the direction of water flow. In the presence of high concentrations of As and S, As can combine with the Fe-O-H structure and form stable insoluble precipitates with other toxic elements [34].

3.3.2. Variation of heavy metals along the flowpath

Based on Fig. 4 (a), Fe in the study area mainly originates from upstream coal mine M1 and tributary coal mine M4. Sampling points S2 and S4 have the highest Fe concentrations in the study area, measuring 446 μg/l and 48 μg/l, respectively. As the migration proceeds downstream, concentrations decrease continuously due to dilution from groundwater and streamflow, and processes such as adsorption and precipitation also contribute to reducing Fe concentrations in the water. This is attributed to the oxidation of pyrite (FeS₂) in the air, resulting in Fe²⁺, which further oxidizes to Fe³⁺. When the pH drops below 3.5, the formed Fe³⁺ undergoes hydrolysis, leading to the precipitation of ferric hydroxide [35]. In Fig. 4 (b, c, and d), Mn, Cu, and Zn show similar spatial variations. The highest concentrations are observed at S2 and Z2, originating primarily from coal mines M2 and M4. Concentrations gradually decrease downstream due to dilution from streams and groundwater. In Fig. 4 (e, f, and g), heavy metals As, Sb, Pb exhibit anomalous behavior in the study area. High concentrations are observed not only at S2 but also at groundwater sampling points S4 and S6, likely influenced by local groundwater and possibly geological formations along the flow path. The highest Sb concentration is at upstream point S1 near coal mine M1, indicating a pollution source upstream. Mixing with AMD at S2 reduces Sb concentrations, suggesting that Fe³⁺ in AMD inhibits Sb migration. In contrast, antimony (Sb) has the lowest concentrations among all elements, nearly absent in tributaries, likely due to low initial concentrations in AMD and further dilution by flowing water. Fig. 4 (h) shows Sr primarily originates from upstream coal mine M1 and tributary coal mine M4, with elevated concentrations also observed near coal mine M2 at S5, indicating influences from three coal mine pollution sources in the study area. Fig. 4 (i) illustrates Se concentrations in the study area differ from other heavy metals, showing elevated levels near coal mines, groundwater sampling points, and throughout the migration process, suggesting natural occurrence rather than originating from AMD. Spatially, Fe in Fig. 4 (a), Zn in Fig. 4 (d), As in Fig. 4 (e), and Sb in Fig. 4 (f) show sudden increases at S7 compared to sampling point S5, likely due to coal slag deposits along the riverbanks, contributing heavy metals possibly from the coal slag. Migration downstream involves processes such as precipitation, co-precipitation, and dilution, gradually decreasing concentrations of heavy metals in downstream waters. Consequently, Mn, As, Sb, Pb, Se, and Fe concentrations in downstream waters remain relatively low, indicating inputs from upstream and tributary coal mines and surrounding environmental sources.

In summary, the overall concentration of heavy metals in the study area gradually decreases along the water flow direction. The upstream and tributaries experience an increase in heavy metal concentration due to the influx of acid mine drainage (AMD) from coal mines M1 and M4, with the magnitude of the increase influenced by factors such as the flow rate of AMD and the concentration of heavy metals. However, the concentration subsequently decreases along the water flow direction.

3.4. Discussion of Downstream Heavy Metal Origins

In the study area, there are multiple coal mines, with the main sources of pollution being coal mines M1 and M4 located at the headwaters of tributaries and upstream. The pollution caused by their convergence is affecting the downstream. By tracing the sources of heavy metals upstream and tributary downstream, we explored the main sources of heavy metals downstream. As shown in Fig. 5, trace analysis of ion sources for Fe, Mn, Cu, Zn, As, and Sr in the upstream-downstream and tributary-downstream revealed that the R² values for Fe, Mn, Zn, As, and Sr were 0.95, 0.96, 0.94, 0.87, and 0.76, indicating relatively low dispersion in the tributary-downstream. On the other hand, the R² values for Fe, Mn, Zn, As, and Sr in the upstream downstream were 0.41, 0.19, 0.29, 0.37, and 0.27, indicating relatively high dispersion. The R² values in the tributary downstream were higher than the upstream downstream, indicating that the major sources of Fe, Mn, Zn, and As in the downstream come from the tributaries. Cu showed high dispersion in both upstream and tributary downstream, with R² values of 0.43 and 0.22, respectively, implying a poor fit of the curve. This suggests that Cu in the downstream does not come from the upstream or the coal mines in the tributary, but possibly originates from the surrounding environment near the river.

The analysis of one-dimensional water quality model simulations and measurements demonstrates a good agreement between the calculated results and the various experimental data, indicating that the established model can accurately simulate the diffusion process of pollutants. Using a one-dimensional water quality model to simulate the changes of solutes in water bodies, revealing the overall behavior during solute transport processes [36, 37]. Simulation of downstream sources of heavy metals through a one-dimensional water quality model, shown in Eq. (3). As shown in Fig. 6, the simulated values of Fe, Mn, Zn, As, and Sr in the tributary-downstream region exhibit a high degree of fitting, whereas the simulated values in the upstream-downstream region show a relatively lower degree of fitting. The R² values of the fitted curves for the simulated values of Fe, Mn, Zn, As, and Sr in the tributary-downstream region are 0.95, 0.95, 0.98, 0.95, and 0.97, respectively, while the R² values of the measured values are 0.91, 0.96, 0.94, 0.88, and 0.76. The close agreement between the simulated and measured values further validates that the downstream heavy metals, Fe, Mn, Zn, As, and Sr, originate from the tributaries. However, in both the upstream-downstream and tributary-downstream regions, the analysis of the measured and simulated values for Cu cannot determine its source. The measured results suggest that Cu has other sources, possibly related to the coal ash deposited along the riverbank.

Discussion

Due to dilution from stream water and precipitation during the migration of heavy metals, concentrations of elements such as Fe, Mn, Cu, Zn, and SO₄²⁻ decrease along the water flow direction, consistent with findings by Shin [38]. During Fe migration, Fe (II) oxidizes to form iron hydroxides, adsorbing or co-precipitating As, particularly As (V), facilitating its removal from acid mine drainage (AMD) to sediment [39], corroborating migration processes observed in this study where both Fe and As concentrations decrease. Eh gradually increases downstream, converting As from As (III) to As (V) in acidic mine water, stabilizing and depositing it in riverbeds, aligning with research by Sun et al [40]. The groundwater in the study area is predominantly HCO₃-Ca-Mg type, transitioning to HCO₃-SO₄-Ca-Mg type under the influence of acid mine water, consistent with findings by Kumar et al [41]. High sulfate content and heavy metals like Fe, Mn originate from oxidized pyrite in mine drainage, while Ca, Mg, Na, and bicarbonates primarily derive from carbonate and silicate weathering dissolution, similar to conclusions drawn by Zhu et al [42]. Buffering by carbonate dissolution and bacterial sulfate reduction stabilizes neutral pH, reducing dissolved metal concentrations in karst regions, consistent with gradual downstream reduction of heavy metals with groundwater and stream inputs observed in this study. Motsi et al. [43] found in their study on heavy metal adsorption that the adsorption rate is directly proportional to the solution pH. In strongly acidic solutions, adsorption decreases due to competition from hydrogen ions. This is consistent with the conclusion of this study where an increase in pH led to decreased adsorption of heavy metals in the water. Cai et al. [44], using a binary equilibrium kinetics model, studied the leaching kinetics of Zn and Cd, indicating that the rapid release phase is controlled by both equilibrium and kinetic sites, while the slow-release phase is predominantly controlled by kinetic sites. The study area is located in a karst mountainous region with significant elevation changes and fluctuating river cross-sections, causing varied hydrodynamic transport, which may also contribute to the variation in heavy metals in the XZ River. Surface water contaminated by acidic mine water threatens crops and aquaculture in the study area; irrigating contaminated surface water affects crop health, prompting conversion from rice paddies to dryland crops. Contaminated surface water with elevated heavy metal levels cannot be used for drinking, adversely affecting local residents’ health. Sampling in this study was not sufficiently systematic and will be improved in future research; water samples were limited, and subsequent studies will include sediment and rock samples.

Conclusions

(1) In the study area, multiple coal mines are distributed, resulting in groundwater being weakly acidic. Surface water exhibits acidity due to the influence of acidic mine water, with an overall pH lower than 3.5. The discharged acidic mine water contains high concentrations of heavy metal ions, causing severe pollution downstream, primarily from coal mines M1 and M4. Upstream and tributary heavy metals gradually decrease in concentration along the water flow direction due to dilution by streams and groundwater, as well as adsorption and precipitation during migration. The accumulation of coal gangue along the banks of XZ River is the main cause of the abnormally high levels of heavy metals.

(2) The primary type of groundwater in the study area is HCO₃-Ca-Mg water, with a minor presence of HCO₃-Ca water. After converging with the discharged AMD, there is a shift in the hydrochemical type, with HCO₃-SO₄-Ca-Mg water becoming predominant. The hydrochemical characteristics of the XZ River and groundwater are primarily influenced by the weathering of carbonate rocks and mixed weathering of clastic rocks. In contrast, the tributaries of the XZ River are mainly affected by the weathering of limestone and dolomite within the watershed.

(3) Using mathematical statistical methods, the source tracing of heavy metals downstream revealed that Fe, Mn, As, Zn, and Sr may originate from the tributaries, while Cu is associated with coal slag deposition along the river during the migration process. One-dimensional water quality simulation analysis showed that the R² values for the fitting curves of simulated Fe, Mn, As, Zn, and Sr values in the downstream were 0.95, 0.95, 0.98, 0.95, and 0.91, respectively. The corresponding R² values for measured values were 0.91, 0.96, 0.94, 0.88, and 0.76, further confirming that Fe, Mn, As, Zn, and Sr in the downstream originate from the tributaries, while Cu is not related to the pollution sources.

(4) According to the research findings, pollution in the downstream of XZ River primarily originates from upstream and tributary coal mines M1 and M4. The discharge of acid mine drainage (AMD) is identified as the main source of downstream pollution. This provides a solid basis for environmental pollution management, allowing effective resolution of downstream pollution issues at their source. Future studies will focus on investigating the temporal and spatial characteristics of pollution in XZ River, as well as conducting a systematic analysis of the migration patterns of heavy metals in karst regions.

Supplementary Information

eer-2024-180-supplementary.pdf

Acknowledgements

Thank you to the National Natural Science Foundation of China (Grants 42107080 and 42162022), Guizhou Provincial Science and Technology Fund (Grant QKHZC [2020] 4Y005), Beijing Natural Science Foundation (Grant 3234060), and Guizhou University Talent Introduction Project (Grant GDRJHZ [2018] 32) for their financial support.

Notes

Author Contributions

Z.S. (M.D.) conducted all the experiments and wrote the manuscript independently. L.P. (Associate Professor), J.W. (M.D.) and L.H (M.D.) participated in the coordination of the study and reviewed the manuscript. Y.Z. (M.D.), Y.W. (M.D.) and B.L. (Professor) helped analyze the results and assisted in the experiments. All authors read and approved the final manuscript.

Conflict of Interest Statement

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

References

1. Mohanty AK, Lingaswamy M, Rao VVSG, Sankaran S. Impact of acid mine drainage and hydrogeochemical studies in a part of Rajrappa coal mining area of Ramgarh District, Jharkhand State of India. Groundwater Sust. Dev. 2018;7:164–175. https://doi.org/10.1016/j.gsd.2018.05.005

2. Tomiyama S, Igarashi T, Tabelin CB, Tangviroon P, Ii H. Acid mine drainage sources and hydrogeochemistry at the Yatani mine, Yamagata, Japan: A geochemical and isotopic study. J. Contam. Hydrol. 2019;225:103502. https://doi.org/10.1016/j.jconhyd.2019.103502

3. Migaszewski ZM, Gałuszka A, Dołęgowska S. Arsenic in the Wiśniówka acid mine drainage area (south-central Poland) – Mineralogy, hydrogeochemistry, remediation. Chem. Geol. 2018;493:491–503. https://doi.org/10.1016/j.chemgeo.2018.06.027

4. Ren Y, Cao X, Wu P, Li L. Experimental insights into the formation of secondary minerals in acid mine drainage-polluted karst rivers and their effects on element migration. Sci. Total Environ. 2023;858:160076. https://doi.org/10.1016/j.scitotenv.2022.160076

5. Kun G, Qiaomei F, Wenjing Z, et al. Arsenic Pollution Mechanism and Associated Remediation Strategy in Acid Mine Drainage Environments. Environ. Sci. Technol. 2022;45:107–116. https://doi.org/10.19672/j.cnki.1003-6504.0719.22.338

6. Zhang J, Liang X, Liu Y, et al. CoKriging Method Based on Principal Components to Predict Spatial Distribution of Arsenic in Groundwater. J. Earth Sci. 2023;48:3820–3831. https://doi.org/https://link.cnki.net/urlid/42.1874.P.20211108.1758.008

7. Wang Z, Xu Y, Zhang Z, Zhang Y. Review: Acid Mine Drainage (AMD) in Abandoned Coal Mines of Shanxi, China. Water. 2020;13:8. https://doi.org/10.3390/w13010008

8. Zhang X, Guo J, Hu Q, et al. Effects of Fe-rich acid mine drainage on percolation features and pore structure in carbonate rocks. J. Hydrol. 2020;591:125571. https://doi.org/10.1016/j.jhydrol.2020.125571

9. Li X, Cai J, Chen D, Feng Q. Characteristics of water contamination in abandoned coal mines: a case study on Yudong River area, Kaili, Guizhou Province, China. Int. J. Coal Sci. Technol. 2021;8:1491–1503. https://doi.org/10.1007/s40789-021-00466-w

10. Qin W, Han D, Song X, Liu S. Sources and migration of heavy metals in a karst water system under the threats of an abandoned Pb–Zn mine, Southwest China. Environ Pollut. 2021;277. https://doi.org/10.1016/j.envpol.2021.116774

11. Chen M, Lu G, Wu J, et al. Migration and fate of metallic elements in a waste mud impoundment and affected river downstream: A case study in Dabaoshan Mine, South China. Ecotoxicol. Environ. Saf. 2018;164:474–483. https://doi.org/10.1016/j.ecoenv.2018.08.063

12. Bondu R, Casiot C, Pistre S, Batiot-Guilhe C. Impact of past mining activities on water quality in a karst area in the Cevennes region, Southern France. Sci. Total Environ. 2023;873:162274. https://doi.org/10.1016/j.scitotenv.2023.162274

13. Ye Z, Zhou J, Liao P, et al. Metal (Fe, Cu, and As) transformation and association within secondary minerals in neutralized acid mine drainage characterized using X-ray absorption spectroscopy. Appl. Geochem. 2022;139:105242. https://doi.org/10.1016/j.apgeochem.2022.105242

14. Huang J, Qin M, Ma W, et al. Characteristics analysis and ecological risk assessment of heavy metals contamination in suspended solids in a river affected by acid mine drainage. Environ. Chem. 2016;35:2315–2326. https://doi.org/10.7524/j.issn.0254-6108

15. Gao K, Zhang B, Liu C. Characteristics and Mechanisms of Heavy Metal Pollution in Acid Mine Drainage Environments. Environ. Sci. Technol. 2023;46:78–84. https://doi.org/10.19672/j.cnki.1003-6504.0216.23.338

16. Song Y, Wang H, Cao Y, Zhu Y, LZ . Effect of NH4+ on the formation of secondary iron minerals and the removal of heavy metals. J. Environ. Sci. (China). 2018;38:2116–2123. https://doi.org/10.19674/j.cnki.issn1000-6923.2018.0245

17. Acharya BS, Kharel G. Acid mine drainage from coal mining in the United States – An overview. J. Hydrol. 2020;588:125061. https://doi.org/10.1016/j.jhydrol.2020.125061

18. Hong-Bin W, Ming L, Lei X, Li-Si Z, Hui-Li Y. Analysis on the Distribution Characteristics and Influence Mechanism of Migration and Transformation of Heavy Metals in Mining Wasteland. Environ. Sci. 2023;44:3573–3584. https://doi.org/10.13227/j.hjkx.202206167

19. Moreno-Gonzalez R, Macias F, Olias M, Ruiz Canovas C. Temporal evolution of acid mine drainage (AMD) leachates from the abandoned tharsis mine (Iberian Pyrite Belt, Spain). Environ. Pollut. 2022;295:118697. https://doi.org/10.1016/j.envpol.2021.118697

20. Hajji S, Montes-Hernandez G, Sarret G, et al. Arsenite and chromate sequestration onto ferrihydrite, siderite and goethite nanostructured minerals: Isotherms from flow-through reactor experiments and XAS measurements. J. Hazard. Mater. 2019;362:358–367. https://doi.org/10.1016/j.jhazmat.2018.09.031

21. Chen W, Liu P, Luo Y, et al. Behavior of Sb and As in the hydrogeochemistry of adjacent karst underground river systems and the responses of such systems to mining activities. Sci. Total Environ. 2023;857:159411. https://doi.org/10.1016/j.scitotenv.2022.159411

22. Dai GC, Wang XH, Chen JS, et al. Regional Geology of Guizhou Province. 1rd edChina Geology. 2013. p. 339–448.

23. Fan L, Xiao-Hui Z, Song T, Mao W, Hong-Ling L. Effects of acid mine drainage on eukaryotic community in river sediments. J. Environ. Sci. (China). 2019;39:5285–5292. https://doi.org/10.19674/j.cnki.issn1000-6923.2019.0613

24. Pan Y, Chen J, Gao K, et al. Spatial and temporal variations of Cu and Cd mobility and their controlling factors in pore water of contaminated paddy soil under acid mine drainage: A laboratory column study. Sci. Total Environ. 2021;792:148523. https://doi.org/10.1016/j.scitotenv.2021.148523

25. Lin P, Yang X, Werner JM, Honaker RQ. Application of Eh-pH Diagrams on Acid Leaching Systems for the Recovery of REEs from Bastnaesite, Monazite and Xenotime. Metals. 2021;11:734–746. https://doi.org/10.3390/met11050734

26. Yan Z, Li Z, Li P, et al. Hydrochemical assessments and driving forces of water resources in coal mining areas: a case study of the Changhe River Basin, Shanxi. Environ Earth Sci. 2023;82. https://doi.org/10.1007/s12665-023-11146-0

27. Tao LC, Zou ZJ, Tu CL, et al. Hydrogeochemical Characteristics and Controlling Factors of the Karst Groundwater in the Area of Fuyuan, Estern Yunnan. Diqiu Yu Huanjing. 2024;52:253–265. https://doi.org/10.14050/j.cnki.1672-9250.2023.51.062

28. Qu S, Liang X, Liao F, et al. Geochemical fingerprint and spatial pattern of mine water quality in the Shaanxi-Inner Mongolia Coal Mine Base, Northwest China. Sci. Total Environ. 2023;854:158812. https://doi.org/10.1016/j.scitotenv.2022.158812

29. León R, Macías FR, Cánovas C, et al. Mine waters as a secondary source of rare earth elements worldwide: The case of the Iberian Pyrite Belt. J. Geochem. Explor. 2021;224:106742. https://doi.org/10.1016/j.gexplo.2021.106742

30. Windisch J, Gradwohl A, Gilbert BM, et al. Toxic Elements in Sediment and Water of the Crocodile River (West) System, South Africa, Following Acid Mine Drainage. Appl. Sci. 2022;12:10531. https://doi.org/10.3390/app122010531

31. Chen Q, Cohen DR, Andersen MS, Robertson AM, Jones DR. Stability and trace element composition of natural schwertmannite precipitated from acid mine drainage. Appl. Geochem. 2022;143:105370. https://doi.org/10.1016/j.apgeochem.2022.105370

32. Yue H, Xiaofei L, Yanping D, et al. Effect of hydrogen peroxide-driven ferrous oxidation on the formation and As (V) adsorption behaviors of schwertmannite. Huanjing Kexue Xuebao. 2023;43:125–135. https://doi.org/10.13671/j.hjkxxb.2022.0190

33. Ghezzi L, Buccianti A, Giannecchini R, Guidi M, Petrini R. Geochemistry of Mine Stream Sediments and the Control on Potentially Toxic Element Migration: A Case Study from the Baccatoio Basin (Tuscany, Italy). Mine Water Environ. 2021;40:722–735. https://doi.org/10.1007/s10230-021-00789-9

34. Tabelin CB, Uyama A, Tomiyama S, et al. Geochemical audit of a historical tailings storage facility in Japan: Acid mine drainage formation, zinc migration and mitigation strategies. J. Hazard. Mater. 2022;438:129453. https://doi.org/10.1016/j.jhazmat.2022.129453

35. Paikaray S. Arsenic Geochemistry of Acid Mine Drainage. Mine Water Environ. 2014;34:181–196. https://doi.org/10.1007/s10230-014-0286-4

36. Tu T, Ercan A, Kavvas ML. One-dimensional solute transport in open channel flow from a stochastic systematic perspective. Stoch. Environ. Res. Risk. Assess. 2019;33:1403–1418. https://doi.org/10.1007/s00477-019-01699-7

37. Ding Y, Li Z, Shi Y, Zhong D. Analytical solution to one-dimensional mathematical model of flow and morphological evolution in open channels. Sci. China Technol. Sci. 2020;63:2606–2616. https://doi.org/10.1007/s11431-020-1721-6

38. Shin W-J, Ryu J-S, Mayer B, Lee K-S, Lee SW. Natural and anthropogenic sources and processes affecting water chemistry in two South Korean streams. Sci Total Environ. 2014;485–486:270–280. https://doi.org/10.1016/j.scitotenv.2014.03.078

39. Zhu J, Zhang P, Yuan S, Tong M. Arsenic oxidation and immobilization in acid mine drainage in karst areas. Sci. Total Environ. 2020;727:138629. https://doi.org/10.1016/j.scitotenv.2020.138629

40. Sun J, Tang C, Wu P, Liu C, Zhang R. Migration of Cu, Zn, Cd and As in epikarst water affected by acid mine drainage at a coalfield basin, Xingren, Southwest China. Environ. Earth Sci. 2012;69:2623–2632. https://doi.org/10.1007/s12665-012-2083-3

41. Kumar V, Paul D, Kumar S. Acid mine drainage from coal mines in the eastern Himalayan sub-region: Hydrogeochemical processes, seasonal variations and insights from hydrogen and oxygen stable isotopes. Environ. Res. 2024;252:119086. https://doi.org/10.1016/j.envres.2024.119086

42. Zhu G, Wu X, Ge J, et al. Influence of mining activities on groundwater hydrochemistry and heavy metal migration using a self-organizing map (SOM). J. Cleaner Prod. 2020;257:120664. https://doi.org/10.1016/j.jclepro.2020.120664

43. Motsi T, Rowson NA, Simmons MJH. Adsorption of heavy metals from acid mine drainage by natural zeolite. Int. J. Miner. Process. 2009;92:42–48. https://doi.org/10.1016/j.minpro.2009.02.005

44. Cai Y, Xu N, Meng F, et al. Speciation and Release Kinetics Simulation of Zn and Cd from River Sediment Contaminated by Gold Mining. Water, Air, Soil Pollut. 2021;232:21. https://doi.org/10.1007/s11270-020-04948-6

Fig. 1

The distribution map of sampling points in the study area.

Fig. 2

Durov map indicates the research area’s primary bodies of water.

Fig. 3

Spatial variation of heavy metals in the XZ River Basin.

Fig. 4

Spatial distribution map of heavy metals.

Fig. 5

Tracing of heavy metal sources.

Fig. 6

Simulation of heavy metal sources.

Table 1

Water sample general ion parameters

parameters	Unites	upstream water sample (n=7)				sample of tributary water (n=5)				downstream water sample (n=7)

		Mean	Min	Max	variation coefficient	Mean	Min	Maxi	variation coefficient	Mean	Min	Max	variation coefficient
pH		4.96	2.17	7.42	0.36	4.12	2.43	6.69	0.34	4.63	2.88	7.80	0.45
EC	μs/cm	1739.00	233.00	6270.00	1.22	1297.86	278.00	2710.00	0.37	1015.29	307.00	1654.00	0.57
Eh	mv	292.26	63.40	440.00	0.46	846.86	230.00	2460.00	0.37	432.66	134.50	570.80	0.39
TDS	c/mg·L⁻¹	1130.35	151.45	4075.50	1.22	843.61	180.70	1761.50	0.37	659.94	199.55	1075.10	0.57
HCO₃⁻	c/mg·L⁻¹	35.04	0.00	86.40	1.00	48.69	0.00	180.01	0.44	33.47	0.00	122.04	1.51
SO₄²⁻	c/mg·L⁻¹	716.23	52.15	1737.80	0.93	1287.12	61.51	2249.37	0.44	780.30	45.35	1340.34	0.74
Cl⁻	c/mg·L⁻¹	7.90	5.10	14.67	0.44	7.27	5.12	10.40	0.24	6.71	3.67	11.03	0.36
F⁻	c/mg·L⁻¹	0.76	−0.90	3.03	1.88	9.93	−1.51	20.47	0.49	1.96	−0.04	4.34	0.79
NO₃⁻	c/mg·L⁻¹	25.86	9.12	67.01	0.93	23.37	11.53	42.90	0.34	21.01	12.35	52.79	0.67
Na⁺	c/mg·L⁻¹	49.17	5.31	132.64	1.06	14.24	3.27	19.36	0.35	35.77	2.90	66.14	0.79
K	c/mg·L⁻¹	6.87	4.37	9.92	0.30	5.33	3.89	6.87	0.18	5.29	4.27	6.16	0.14
Ca²⁺	c/mg·L⁻¹	153.30	26.37	370.84	0.90	261.05	41.70	415.17	0.34	186.59	49.37	277.41	0.51
Mg²⁺	c/mg·L⁻¹	61.89	4.10	231.03	1.30	53.78	7.37	89.73	0.38	29.21	3.49	53.63	0.75

Table 2

Heavy metal ions content in XZ River

parameters	Unites	upstream water sample (n=7)				sample of tributary water (n=5)				downstream water sample (n=7)

		Mean	Min	Max	variation coefficient	Mean	Min	Max	variation coefficient	Mean	Min	Max	variation coefficient
Fe	c/ug·L⁻¹	80.38	−0.42	446.01	2.02	19.38	−0.39	47.91	1.06	5.65	−0.27	13.98	0.99
Mn	c/ug·L⁻¹	3.91	−0.25	18.94	1.75	4.92	−0.12	9.59	0.94	2.54	−0.05	4.38	0.69
Cu	c/ug·L⁻¹	22.32	−4.53	121.98	2.01	39.43	−0.75	97.95	1.04	12.59	−2.74	25.52	0.97
Zn	c/ug·L⁻¹	287.58	36.14	1397.75	1.71	378.26	4.50	737.00	0.88	133.08	5.06	233.66	0.74
As	c/ug·L⁻¹	8.50	2.18	40.15	1.65	2.85	1.28	4.62	0.56	1.52	0.20	3.55	0.70
Sb	c/ug·L⁻¹	8.52	1.21	36.35	1.47	0.13	−0.04	0.30	1.03	0.32	−0.22	1.37	1.75
Pb	c/ug·L⁻¹	6.40	4.40	8.04	0.18	4.73	0.78	6.48	0.49	3.99	0.95	5.64	0.52
Sr	c/ug·L⁻¹	3721.77	−27.49	9911.91	1.19	3034.00	149.45	6247.47	0.85	1947.25	238.35	3582.24	0.82
Se	c/ug·L⁻¹	7.77	4.42	17.52	0.58	7.43	3.31	14.14	0.56	3.83	0.45	10.54	0.95