Quantitative Source Identification, Pollution Risk Assessment of Potentially Toxic Elements in Soils of a Diamond Mining Area

Abstract

Potentially toxic elements (PTEs) are the most important indicators of environmental pollution and represent a potential risk to the ecology and human health in industrial regions. Eight potentially toxic elements (Mn, Ni, Co, Cr, Pb, Zn, Cd, As) in soils formed on the territory of the industrial site of the Udachny Mining and Processing Division were considered in this study. The potential ecological risk index (RI) was calculated to determine environmental risks of soil contamination. The concentrations of PTEs decreased in the following order Mn > Ni > Zn > Co > Pb > Cr > As > Cd. In total, 19.51% of the sites in the study area exhibited a high potential ecological risk for Mn and Ni, while only 4.87% exhibited a low potential ecological risk for other PTEs. The greatest impacts on soil contamination are exerted by the areas of the Udachny and Zarnitsa pipes, tailings ponds, and the area’s highly mineralized water outlet. The results of correlation analysis (CA) and hierarchical cluster analysis (HCA) revealed that the same groups of elements were present: Co-Cr-Ni and Cd-Zn. The PMF findings demonstrate that the five main diverse sources of PTEs in this study area’s soils were natural, mining activities, transportation, and industrialization, as well as highly mineralized waters.

1. Introduction

Mineral resources are an important source of income for many countries. In recent years, the mining industry has caused serious environmental problems [1]. Mining operations, which encompass the extraction of minerals and ores both underground and on the surface, are thought to be accompanied by environmental degradation, contamination, and related diseases due to the release of certain trace elements into the environment [2]. Among the various chemical pollutants, trace elements, particularly those that are potentially toxic elements (PTEs), present a significant environmental risk due to their persistence, non-degradability, bioaccumulation, and high toxicity [3]. The concentration of PTEs in soils is influenced by two primary sources: the natural background and anthropogenic impacts [4,5]. The first source is primarily derived from the parent materials from which the soils were formed. The second is the result of intensive human activity. The most significant impact on the environment is caused by open-pit mining operations, such as diamond mining, which are accompanied by a number of geomechanical disturbances. These include the creation of excavations and the formation of dumps, which are caused by aerodynamic disturbances, as well as changes in the regime of water bodies, which are caused by hydrogeological disturbances [6,7]. The lithosphere is the most susceptible to negative impact, with the formation of quarries and dumps leading to the disruption of the terrain. Moreover, a number of processes influence environmental changes and result in anthropogenic geochemical anomalies [8,9,10]. Soil is a significant geochemical sink for a variety of pollutants, acting as a conduit for their transport to the atmosphere, hydrosphere, and biomass [11,12,13]. Consequently, a study is required to assess the ecological risk associated with PTEs in soils.

There are few studies on the qualitative identification of the source of PTE contamination from diamond mining, with even fewer studies on the quantitative distribution of the source. The sources of PTEs in soils are usually determined by using statistical methods including descriptive analysis, correlation analysis (CA), hierarchical cluster analysis (HCA), and the positive matrix factorization (PMF) model [14,15]. The results obtained from different methods may not be consistent from one method to another, which could add ambiguity to the conclusions. Thus, a source determined by one method requires confirmation by other methods in order to improve validity [16,17]. Therefore, in this study, we incorporated all the above-mentioned statistical methods as well as spatial analysis of contamination, namely the potential ecological risk index (RI), to investigate the contribution of an element in soil polyelemental contamination.

In this context, the primary objectives of this study are as follows: (1) To study the content and distribution of PTEs in the soils of the study area. (2) To assess the potential ecological risk of PTEs. (3) To determine the relationship between PTEs and possible sources using correlation analysis, hierarchical cluster analysis, and positive matrix factorization (PMF).

It is anticipated that the findings of this study will serve as the foundation for the formulation of policies aimed at reducing the contamination of PTEs in mining regions and mitigating the associated health risks to local communities.

2. Materials and Methods

2.1. Study Area

The study area is located within the Daldyn–Alakit diamond-bearing area, and specifically, the Daldyn kimberlite field. Figure 1 Shows the map of the study area where the Udachny Mining and Processing Division (Udachny MPD) is based (N 66°25′47″, E 112°24’07″) (Figure 1). At present, two pipes, Udachny and Zarnitsa, have industrial deposit status in this area [18].

The exploration of diamondiferous areas is accompanied by geological prospecting works of various levels of detail, where drilling, geophysical, and mining works are carried out, and a network of cross-cuts. In addition, the operation of the largest diamondiferous pipes is complicated by the inflow of brines from subpermafrost high-pressure aquifers; their safe disposal into underground storage facilities is the key objective of ensuring a favorable environmental setting [19]. Rocks brought up from the depths are accumulated in overburden dumps about 100 m in height, with concentration products being stored in extensive tailings ponds. Thus, significant amounts of chemical elements and their compounds are brought to the day surface from the depth of the Earth, which form anthropogenic geochemical anomalies in soils, bottom sediments, and surface waters.

The climate here is sharply continental: the average annual temperature is 12.7 °C and the average annual precipitation is 250 mm; it is surrounded by a zone of continuous distribution and close occurrence of permafrost [20]. The relief of the study area is characterized by a hilly–U-shaped valley relief with a height of 400–500 m and relative heights above the nearest watercourses of 100–250 m [21]. The major zonal soil type is Cryosols, the intrazonal type is Fluvisols. The subordinate types are Rendzic Leptosols and Umbric Gleysols. The vegetation cover of the study area is situated within the subzone of sparsely stable north taiga larch forests. Larch forests are the dominant vegetation type, occupying 80% of the area, and are represented by such species as Larix gmelinii [22,23].

Geological setting of the studied area, i.e., the presence of two levels of structure (crystalline basement and sedimentary cover), is determined by its location within the ancient craton, the Siberian platform, and specifically, in the junction zone of the northeastern side of the Tunguska syneclise with the southwestern slope of the Anabar anteclise [24].

The sedimentary cover of the study area is composed of the Paleozoic carbonate and terrigenous carbonate rocks of the Vendian, Cambrian, and Ordovician, as well as the Quaternary sediments of various genetic types. The total thickness of the sedimentary strata varies from 2200 to 2500 m. The structural stages are separated from each other by sedimentation breaks, and angular and stratigraphic unconformities. Lower Paleozoic rocks are partially overlapped by Triassic volcanogenic rocks and Quaternary sediments. Magmatic formations are not abundant and are represented by unexposed sills and dikes of dolerites that cross regional faults [25] (Figure 2).

2.2. Sampling, Analysis, and Quality Control

Considering the source of trace element emissions in the study area, soil samples were collected in the vicinity of the industrial zone of the Udachny Mining and Processing Division (near the Udachny and Zarnitsa kimberlite pipe quarries, waste rock dumps, tailings ponds, the processing plant, and other industrial facilities). A total of 60 soil sampling sites were determined. The sampling distribution is shown in Figure 1c.

Samples of soil were collected from the depths 0–20 cm. Sampling was performed using a stainless steel shovel from the ditches. Soil was packed in polyethylene bags labeled with notes with sample code, date, and sampling location. Soil samples were oven dried at room temperature and sieved through a <1.0 mm mesh. These samples were stored in a polyethylene bag for testing in the laboratory.

The samples were analyzed for pH by the potentiometrically method (ISO 10390:2005) [26], soil organic carbon (SOC) by the photoelectric colorimetric method (ISO 14235:1998) [27], total N by the spectrophotometric method (NF ISO 11261:1995) [28], and granulometric composition by sedimentation analysis using the pipette method [29]. The concentration of mobile forms of PTE: lead (Pb), nickel (Ni), manganese (Mn), cadmium (Cd), cobalt (Co), chromium (Cr), zinc (Zn), and arsenic (As), were determined by atomic absorption spectrophotometer (AAS) model MGA-1000 (GC Lumex, Saint Petersburg, Russia) according to the methodology M 03-07-2014 [30]. Extraction of mobile forms of elements from soils was conducted by a static method consisting of single shaking of the soil samples with a solution at 1:10 (1 mol/dm³ HNO₃) soil-to-solution ratio, for the extraction time of 2.5 h. Each sample was filtered through a 0.45 μm aqueous filter membrane, then sent for analysis.

To ensure and control quality, blinded duplicates and standard samples (SDPC-1,-2,-3 [31] and SSC-1,-2,-3 [32], Russian National Center for Standard Materials) were used. The standard deviation was <5% for all elements.

2.3. Statistical Analysis

The distribution of data was assessed using the Kolmogorov–Smirnov and Shapiro–Wilk tests to avoid data dispersion. When testing the data, the null hypothesis was generally rejected. Therefore, the data were transformed into “normal” distributions according to the principles of compositional data analysis (CoDa) [33,34,35] by means of the centered logarithmic ratio transformation (clr transformation). For the transfer of raw data to clr transformation data, CoDaPack software (Version 2.03.01, University of Girona, Girona, Spain) was used (http://www.compositionaldata.com/codapack.php, accessed on 16 November 2022) [36].

Statistical analysis comprised of the determination of the mean and geometric values, median, standard deviation, minimum, maximum, and coefficient of variation. Correlation analysis was utilized to assess the relationships between the studied variables with statistical significance set at the level of p ≤ 0.05 [2]. Elements with p ≤ 0.05 indicated that there was a significant correlation between the elements [37,38]. The hierarchical cluster analysis (HCA) is an alternative approach used to validate the results of correlation analysis. Cluster analysis is typically used to represent a group of variables that are comparable to each other at a particular sampling site, as opposed to parameters that exhibit particular variability [39,40]. Non-comparable sites are displayed in a separate cluster group to highlight specific sites corresponding to the level of contamination [41,42]. An identical site is mapped in one cluster group and a different site is mapped in a different cluster group. All statistical analyses were processed using Statistica (Version 13.0), IBM SPSS Statistics (Version 23.0) software and OriginPro 2024 (Version 10.1.0.178).

Spatial information of soil materials can be displayed directly through the application of geostatistical methods. The RI spatial distribution was performed using the standard Kriging method in the Surfer 2023 software (Version 25.2.259).

2.4. Calculations of Contamination Index

2.4.1. Potential Ecological Risk Index (RI)

To evaluate the toxicological effect of PTEs on the ecosystem, the potential ecological risk index (RI) was applied [43]. This method enables the direct reflection of the hazard posed by single or multiple PTEs. This approach has been widely employed to investigate the contamination of surrounding soil by PTEs in various mining regions, and to demonstrate the potential ecological risks posed by overall contamination [44]. The equations for this method are as follows:

$${ E}_r^i=T_r^i\times P_{i }$$

(1)

$${ P}_i={\displaystyle \frac{C_n}{B_n}}$$

(2)

$$RI=\sum _{i=1}^n{E}_r^i=\sum _{i=1}^n{T}_r^i\times P_i=\sum _{i=1}^n{T}_r^i\times {\displaystyle \frac{C_n}{B_n}}$$

(3)

where $E_r^i$ is the potential ecological risk factor of PTE I, $P_{i }$ is the pollution index of the i element (mg/kg), $C_n$ is the concentration of each i element in the soil, $B_n$ is a background value of each i element, $T_r^i$ is the toxicity response factor of PTE i, and the toxicity coefficient of each PTE is Cd = 30,  Ni = Pb = Co = 5, Zn = Mn = 1, Cr = 2, and As = 10 [45,46,47]. The potential ecological risk index (RI) of the soil can be divided into the following categories: RI ˂ 150—low risk; 150 ≤ RI ˂ 300—moderate risk; 300 ≤ RI ˂ 600—considerable risk; and RI > 600—very high risk [48].

2.4.2. PMF Model

To identify the sources of soil contamination, the PMF model was applied, based on factor analysis technology [49,50]. In this study, EPA PMF 5.0 was implemented to quantitatively analyze the sources of contamination. This method is based on the following Equation (4):

$$x_{ij}=\sum _{k=1}^p{g}_{ik}{f}_{kj}+e_{ij}$$

(4)

where x_ij is a measurement matrix of j chemical element in i number of samples; g_ik is a contribution matrix of the k source factor for i number of samples; f_kj is a source profile of j chemical element for the k source factor; and e_ij is the residual error matrix.

The residual error matrix e_ij is computed by minimizing the object function Q. Q is calculated by Equation (5):

$$Q=\sum _{i=1}^n\sum _{j=1}^m{\left({\displaystyle \frac{x_{ij}-\sum _k^p{g}_{ik}{f}_{kj}}{u_{ij}}}\right)}^2$$

(5)

where u_ij is the uncertainty in the j chemical element for sample i and is calculated by Equations (6) and (7):

$$Unc={\displaystyle \frac{5}{6}}\times MDL, (if c \le MDL)$$

(6)

$$Unc=\sqrt{(error fraction\times {c)}^2+{(0.5\times MDL)}^2}, \left(if c>MDL\right)$$

(7)

where c is the concentration values of soil samples, MDL is the species-specific method detection limit, and the error fraction represents the uncertainty measurement percentage. More detailed information about the software can be found in Reference [49].

3. Results

3.1. Descriptive Statistics of Soil Physical–Chemical Properties and PTE Concentrations

Results of the statistical analysis of soil physical–chemical properties are shown in

Table 1. Descriptive statistics of physical–chemical characteristics of soils in the study area.

Index	Physicochemical Properties					Granulometric Fractions, mm
	pH	Humus, %	SOC, %	TN, %	SOC/TN	1–0.25	0.25–0.05	0.05–0.01	0.01–0.005	0.005–0.001	<0.001	<0.01	>0.01
Mean	7.62	9.88	5.73	0.81	7.06	1.04	7.65	23.45	10.34	15.83	25.21	51.38	32.14
Geometric mean	7.56	7.01	4.07	0.62	6.56	0.44	6.66	22.57	10.31	15.28	24.12	51.24	31.43
Median	7.80	6.20	3.60	0.76	4.72	0.34	6.26	26.62	10.19	15.72	27.92	50.02	33.08
Minimum	4.32	1.10	0.64	0.03	13.80	0.05	3.00	15.61	9.55	11.67	12.70	46.55	22.14
Maximum	9.30	47.00	27.26	1.98	25.52	2.81	12.19	30.86	11.55	23.66	29.74	57.71	42.59
Variance	0.72	72.52	24.40	0.23	106.55	1.47	18.06	47.76	0.68	24.01	49.90	18.09	55.80
Standard deviation	0.85	8.52	4.94	0.48	10.32	1.21	4.25	6.91	0.83	4.90	7.06	4.25	7.47

Note: SOC—soil organic carbon, TN—total N.

. The pH values of soil samples were recorded as ranging from 4.3 to 8.7. The majority of the samples exhibited alkaline pH values, with 70% of the samples exhibiting a pH of 7.0 or above, 15% of the samples exhibiting neutral pH values, and 15% of the samples exhibiting slightly to moderately acidic pH values. The content of soil organic carbon in the studied territory is relatively high and exhibits medium spatial variability (CV = 24%). The proportion of physical clay is predominates, i.e., particles of heavier fractions in the studied soils (<0.01 mm).

The results of the PTEs’ content in the study area are shown in

Table 2. Descriptive statistics of PTEs’ mobile forms in the study area, mg/kg.

Element	Background Values	Mean	Geometric Mean	Median	Minimum	Maximum	Variance	Standard Deviation	Standard Error	Skewness	Kurtosis
Pb	1.79	2.21	1.80	1.96	0.33	6.60	1.584	1.41	0.22	1.13	1.08
Ni	3.12	15.39	3.43	2.40	1.00	163.80	295.4	40.97	6.40	3.17	8.82
Mn	189.0	311.63	199.98	184.90	34.76	1856.0	4242.9	415.5	64.88	3.00	8.49
Cd	0.11	0.14	0.11	0.10	0.03	1.12	0.035	0.18	0.03	4.56	24.00
Co	2.64	4.53	2.73	2.69	0.53	34.65	53.49	6.92	1.08	3.47	12.20
Cr	0.93	1.94	0.97	0.85	0.11	21.45	18.79	4.08	0.64	4.08	16.70
Zn	9.47	12.10	9.36	11.46	0.05	36.46	51.78	6.99	1.09	1.17	2.53
As	0.13	0.22	0.12	0.20	0.03	0.84	42.22	0.21	0.03	1.03	0.62

Note: Regional background values regarding the content of mobile forms of trace elements are based on the geometric mean values of soil samples (n = 212) of the natural/undisturbed landscapes outside the impact zone of the mining and processing division [51,52].

. The mean concentrations of Mn, Ni, Zn, Co, Pb, Cr, As, and Cd were 311.6, 15.4, 12.1, 4.53, 2.21, 1.94, 0.22, and 0.14 mg/kg, respectively. These results also exceeded the background values for Ni, Co, As, Mn, Zn, Pb, Cd, and Cr by 4.93, 1.72, 1.69, 1.65, 1.28, 1.23, 0.79, and 0.48 times, respectively. The coefficients of variation of PTEs in the study soils exhibited the following decreasing sequence: Mn > Ni > Zn > Co > Cr > Pb > As > Cd.

3.2. PTEs’ Contamination Levels of Soils

The Er values of each PTE were found to range from 1.65 to 33.0 for Pb, 5.0 to 819.0 for Ni, 34.8 to 1856.0 for Mn, 0.78 to 33.6 for Cd, 2.7 to 173.3 for Co, 0.22 to 42.9 for Cr, 0.05 to 36.5 for Zn, and 0.25 to 8.4 for As. The order of mean Er values for the investigated PTEs is as follows: Mn > Ni > Co > Zn > Pb > Cd > Cr > As. The results of the comprehensive assessment of potential ecological risk (RI) of the PTEs range from 92.0 to 2840.9, with a mean value of 485.8. The proportion of high environmental risk was 19.51%, while the proportions of considerable, moderate, and low environmental risk were 21.95%, 53.66%, and 4.88%, respectively (Figure 3).

3.3. Multivariate Statistical Analysis Results

The Kolmogorov–Smirnov and Shapiro–Wilk data analysis revealed that the concentrations of elements Mn, Ni, Cd, Co, and Cr in the soils within the study area exhibited a departure from the normal distribution. The observed distribution of Pb, Zn, and As is approximately normal. After undergoing a conformity check in accordance with the principles of normal distribution, the data underwent a transformation using the clr transformation. Upon application of the clr-transformed data, the corresponding standard deviation exhibited a notable decline, and the median and mean exhibited a tendency towards similarity. Thus, the data transformed by the clr transformation showed a normal distribution for most elements as a consequence of the diminished influence of outliers (

Table 3. Descriptive statistics for raw data and clr-transformed data of PTEs.

PTE	Raw Data						Clr-Transformed Data
	Shapiro–Wilk			Kolmogorov–Smirnov			Shapiro–Wilk			Kolmogorov–Smirnov
	Statistic	p-Value	Decision at Level (5%)	Statistic	p-Value	Decision at Level (5%)	Statistic	p-Value	Decision at Level (5%)	Statistic	p-Value	Decision at Level (5%)
Pb	0.91679	0.01	Reject normality	0.14706	0.38	Cannot reject normality	0.97565	0.60	Cannot reject normality	0.09596	0.97	Cannot reject normality
Ni	0.38995	<0.0001	Reject normality	0.46737	<0.0001	Reject normality	0.74921	<0.0001	Reject normality	0.25081	0.018	Reject normality
Mn	0.56651	<0.0001	Reject normality	0.29452	0.003	Reject normality	0.97585	0.60	Cannot reject normality	0.09874	0.93	Cannot reject normality
Cd	0.50433	<0.0001	Reject normality	0.30397	0.002	Reject normality	0.91881	0.011	Reject normality	0.13259	0.52	Cannot reject normality
Co	0.51187	<0.0001	Reject normality	0.3747	<0.0001	Reject normality	0.96699	0.35	Cannot reject normality	0.08672	1.00	Cannot reject normality
Cr	0.40102	<0.0001	Reject normality	0.40484	<0.0001	Reject normality	0.95914	0.20	Cannot reject normality	0.12286	0.62	Cannot reject normality
Zn	0.93232	0.02944	Reject normality	0.14168	0.43	Cannot reject normality	0.80372	<0.0001	Reject normality	0.20528	0.08	Cannot reject normality
Cu	0.8423	0.0001	Reject normality	0.19893	0.10	Cannot reject normality	0.98455	0.88	Cannot reject normality	0.07433	1.00	Cannot reject normality
As	0.86709	0.0005	Reject normality	0.18388	0.15	Cannot reject normality	0.93092	0.027	Reject normality	0.13363	0.51	Cannot reject normality

).

3.4. Correlation and Hierarchical Cluster Analysis

Correlation analysis is an important basis for determining the source of PTEs. The results of the correlation analysis of PTEs in the study area are presented in Figure 4.

The correlation study of the respective elements showed a significant high positive correlation between Cr, Ni, and Co at p ≤ 0.01. Another significant but weaker correlation was observed between Cd and Zn at p ≤ 0.05. They also showed negative correlation with the previous group—Ni, Co, and Cr. In addition, a significant negative correlation was observed between Pb and Ni (p ≤ 0.05), Pb and Co (p ≤ 0.01), and Mn and Cr (p ≤ 0.05).

Hierarchical cluster analysis can directly reflect the correlation among heavy metals and reveal the sources of soil PTEs. As a result of the present study, a heat map and a dendrogram were created. They were constructed using Ward’s method. The result of the construction is shown in Figure 5. There are three clusters for PTEs in the dendrogram. Zn and Cd are in cluster 1, Co, Cr, and Ni are in cluster 2, and Pb, Mn, and As are in cluster 3.

Clusters of sampling locations were also divided into three categories: Category 1 represents points in the impact zone of the Udachny pipe dumps, tailings pond, and drainage water injection area where soils are chemically transformed. Higher concentrations of Mn, Ni, Cr, Co, and Zn are observed here. Category 2 points are predominantly located near the dumps and pits of the Udachny and Zarnitsa pipes and roads. Category 3 points are predominantly associated with drainage water injection areas. In contrast to category 1 and 2 points, category 3 points are characterized by high As content.

3.5. Source Apportionment by PMF

The PMF model was launched using concentration data and uncertainty data files that cover errors such as sampling and analysis errors [53]. The signal-to-noise (S/N) ratio was defined as “strong” for Mn, Co, Zn, Cd, Ni, and Pb, and “weak” for Cr and As. Residual analysis revealed that the majority of PTE values were distributed primarily in the range of −3 to 3. The fit coefficients (r2) for Mn, Co, Zn, Cd, Ni, Pb, Cr, and As were 0.999, 0.994, 0.884, 0.673, 0.999, 0.689, 0.984, and 0.815, respectively, showing a significant degree of conformity. The PMF model distributed five sources; their factor profiles are shown in Figure 6.

For Factor 1, these were As (89.9%) and Pb (57.6%); for Factor 2, this was Ni (77.9%); for Factor 3, these were Cr (56.9%) and Co (44.3%); for Factor 4, these were Cd (65.5%) and Zn (71.1%); and for Factor 5, this was Mn (65.6%).

4. Discussion

The studied soils of the Udachny Mining and Processing Division are predominantly characterized by alkaline soil conditions. An alkaline environment contributes to the formation of hydrolyzed or complex forms of elements in soils, which can lead to the precipitation and accumulation of PTEs [12,54]. Therefore, when the pH of the soil substrate increases, the solubility of most trace elements decreases. According to the pH precipitation scale [12], the elements Mn, Co, Zn, Cd, and Ni tend to form rather stable compounds when the soil pH is above 7. Consequently, there is a high probability of the accumulation of PTEs, which can pose a significant threat to soil contamination. High soil organic matter values are more likely associated with low decomposition of plant residues, as indicated by the SOC/TN ratio data. The influence of industrial development of the territory is also not excluded, which has resulted in the suppression of natural soil formation processes. Therefore, these data may reflect not so much the actual humus content of the soils, but the total carbon content in them, in which the technogenic component (fuel hydrocarbons, lubricating oils, etc.) is significant [55]. The characteristics and size of soil particles determine the degree of sorption capacity, which contributes to the retention of PTEs by the soil. The more fine particles in the soil, the higher the sorption capacity of the soil. The studied soils have predominantly heavy fractions, which increases the probability of retaining the largest amount of PTEs. In early works, it was noted that the soils of the northern taiga zones have a heavier granulometric composition and high sorption capacity, which accumulate a fairly large amount of PTEs [56].

The soil coefficient of variation can be employed to assess the homogeneity and variability of heavy metal content in soil [57]. It is generally accepted that naturally occurring elements tend to exhibit low coefficients of variation (CVs), whereas elements associated with anthropogenic sources are characterized by high CVs and reflect a heterogeneous distribution of concentrations. This is evidenced by studies such as those conducted by Huang et al., 2009, and Li et al., 2013 [16,58]. The results indicated that Mn, Ni, Zn, Co, and Cr exhibit high variability, whereas the distribution of Pb, As, and Cd in the study area is more homogeneous. It can be postulated that the elements Mn, Ni, Zn, Co, and Cr in the study area are subject to external influences, which are largely attributable to human activities such as transportation and industrial operations. Exceeding the background values of PTEs in the studied soils indicates this influence. The areas with high ecological risk include the waste rock dump sites of the Udachny and Zarnitsa pipes, tailings ponds, and the zone of the highly mineralized water outlet. The majority of the study area was at or above the moderate environmental risk level, with Mn and Ni being the primary risk factors (Figure 7).

A combination of multivariate statistical analysis was used to analyze the sources of PTE. The correlation analysis and hierarchical cluster analysis revealed two identical groups: Ni-Co-Cr and Cd-Zn. The strong positive correlations between these PTEs indicate that the characteristics and emission sources for these elements might be similar [59,60].

The Cr-Ni-Co element group is the elements of the iron group; they usually have an affinity for siderophiles [60]. The close association of these elements reflects the geochemical specificity of kimberlite magmatism of the territory of the Daldyn kimberlite field, as they are elements typomorphic to kimberlites. Accordingly, significant contents of Cr-Ni-Co association are related to their secondary input to the soil surface combined in the form of suspended fine-dispersed dust, resulting from aerogenic dispersion from waste dumps, quarries, industrial sites, and others. A high correlation between these elements was also revealed in earlier studies [59,61].

The association of Cd-Zn makes it possible to suggest that these elements may have a similar origin. A review of the literature shows that mining enterprises emit dust into the environment, the Cd content of which significantly exceeds its level in the topsoil [13,62]. Consequently, the frequently observed accumulation of Cd in topsoil is often attributed to pollution [12]. Cd accumulation may also be influenced by high levels of poorly decomposed soil organic matter (SOM), which acts as a biogenic barrier and accumulates high Cd concentrations. This is indicated by the positive correlation between SOM content and Cd concentration observed in early studies [21].

However, Zn may have a lithogenic source because it forms a range of soluble salts (e.g., chlorides, sulfates, and nitrates) or insoluble salts (e.g., silicates, carbonates, phosphates, oxides, and sulfides) depending on the prevailing pedogenic processes [63]. High concentrations of Zn, though, are often attributed to anthropogenic activities, namely exposure to motor vehicles [16,64,65]. Similar results were obtained in other studies [66,67,68]. Thus, Cd is mostly supplied from the impact of industrial site activities, while Zn is more from vehicular traffic.

The dendrogram of the sampling points forms three groups. Group 1 includes the points (P-35, P-19/1, P-19/2, P-19, P-15) with the highest PTE concentrations, which are located directly in the impact zone of the industrial site facilities. This result illustrates the influence of anthropogenic activities on the variations in concentrations of the studied trace elements. In groups 2 and 3, the proportion of soils within the normal range is the same—43.9 and 43.9%, respectively. However, it is possible to distinguish several sites with enrichment of some elements Mn, Zn, Pb, Ni, Co, and Cr in group 2 and Mn, Zn, Pb, As, Ni, Co, and Cr in group 3.

The PMF model was applied to the further analysis of PTE sources.

The load of As and Pb elements in Factor 1 is the highest. Average geometric values of As and Pb in comparison with the background content were lower, which indicates the absence of significant contamination of the territory by these elements. Based on this, it can be assumed that As and Pb are of natural origin. In addition, previous researches exhibited that the chief source of As and Pb was parent material [69,70].

Ni was predominant in Factor 2. Ni belongs to elements of the iron group, easily binds to oxides in soil, and is closely associated with soil rock components [71]. The highest values are tracked near the Udachny kimberlite pipe dump and in the zone of the highly mineralized water outlet. High CV values and pollution levels confirm the impact of industrial activity.

Factor 3 was mainly loaded with Cr and Co. Cr and Co were usually representative elements of natural sources, and they were widely presented in the pedogenic process and soil parent material [72]. Most of the soil samples had Co and Cr contents identical to background values, pollution levels of the soil samples were predominantly uncontaminated, and the Cr and Co with high values found in the soils came from one source—the Udachny pipe dumps. Furthermore, as mentioned above, these elements are typomorphic elements, i.e., from the natural genesis. Consequently, this factor is classified as mixed—a natural geological origin and anthropogenic.

The main load elements of Factor 4 were Cd and Zn. The geometric mean content and contamination levels were low. In this case, single local points with significant concentrations are observed. The high value of Zn content was primarily distributed near roads, as well as densely populated and congested urban areas, and the high value of Cd content was observed in the area of the landfill of drainage brines and highly mineralized waters—“Oktyabrsky”. Thus, Factor 4 represents transportation and industrial activities.

The main characteristic element in Factor 5 was Mn. According to Goncalves et al. [73], Mn is the second most common metal and the twelfth most abundant element in the crust of the Earth. It is found in nature in various minerals, rocks, soils, and sediments in the form of manganese oxides. The occurrence and distribution of manganese in nature can vary depending on geological and environmental factors. However, the high CV value and very high pollution levels indicated that Mn was influenced by anthropogenic activities.

5. Conclusions

According to this study, the PTE content in the soils of the Udachny MPD industrial site is characterized by significant variability, with high spatial variability observed. Average concentrations of Mn, Ni, Zn, Co, Pb, and As exceeded background values. The results of potential ecological risk assessment showed that the mean RI value in the study area was 485.8, which indicates considerable ecological risk. In total, 19.51% of the sites had high ecological risk, and the main risk factors were Mn and Ni. Hot spots were detected at the impact areas of the dumps of Udachny and Zarnitsa pipes, tailings ponds, and the zone of the highly mineralized water outlet. The results of correlation analysis (CA) and hierarchical cluster analysis (HCA) revealed the same groups of elements: Co-Cr-Ni and Cd-Zn. Through PMF model analysis, five possible sources were identified. Among the PTEs investigated in the soils of the study area, Pb and As were mainly derived from parent materials and have natural origin. The concentrations of Co, Cr, and Ni were the result of a combination of anthropogenic impact and lithogenic nature. Cd, Zn, and Mn primarily originate from different anthropogenic and industry sources.

This study highlights the importance of considering current PTEs’ contamination in soils as an important factor to be considered when developing strategies to mitigate the effects of contamination in mining areas.