Magnetic Monitoring and Source Traceability of Heavy Metal Pollution in the Urban Topsoil of Xuzhou, China

Abstract

This study integrates environmental magnetism, geochemical analysis, and multivariate statistical methods to investigate heavy metal pollution in the urban topsoil of Xuzhou, China. A total of 142 samples were collected, and concentrations of Cr, Cu, Fe, Mn, Ni, Pb, Zn, and magnetic parameters (χ, χ_fd, SOFT, SIRM, χ_ARM) were measured. Results reveal elevated levels of Cu, Ni, Pb, and Zn in both 0–2 cm and 3–10 cm soil layers, with concentrations in the 0–2 cm layer (1.3–2.0 times background values) significantly exceeding those in the 3–10 cm layer, indicating anthropogenic inputs primarily accumulating at the soil surface. Magnetic parameters (χ, SOFT, SIRM, χ_ARM) correlated strongly with Pb and Zn, and the pollution load index (PLI), highlighting their potential as rapid indicators of pollution. Spatial analysis identified hotspots in city centers and industrial zones, aligning with moderate to severe PLI values. Principal component analysis and magnetic source tracing uncovered four pollution sources: transportation/fossil fuel combustion, industrial activities, natural soil formation, and other natural processes. By linking magnetic signatures to anthropogenic activities, the study provides a scientific basis for ecological restoration, safe land use, and pollution mitigation strategies in resource-based cities, directly contributing to urban environmental sustainability.

1. Introduction

Cities represent territorial complexes that are subject to the most intense influence from human activities. The unceasing progress of urbanization and industrialization has led to cities emerging as concentrated hubs of resource consumption and pollution emissions [1]. Anthropogenic activities introduce large amounts of pollutants, particularly heavy metals, into the urban environment. These heavy metals are transported to urban soils via atmospheric deposition and rainfall, rendering urban soils a crucial “sink” and “source” within urban ecosystems [2]. Heavy metals within soils are capable of infiltrating the human body via the hand-to-mouth route. Given their persistent nature, invisibility, irreversibility, and resistance to degradation, heavy metals pose a significant threat to human health [3,4]. Monitoring heavy metals in soil serves multiple vital functions. It can issue early warnings of soil pollution problems and offer a scientific basis for soil pollution prevention and control. Moreover, it helps ensure the quality and safety of agricultural products, maintains the health of ecosystems, and enables the rational use and management of land resources. Overall, such monitoring plays a crucial role in achieving sustainable environmental development.

A series of studies have been conducted on the concentration, spatial distribution characteristics, risk evaluation, and pollution sources of heavy metals in urban soils. Relevant studies have demonstrated that the levels of heavy metal contamination exhibit significant discrepancies across different urban functional areas. Moreover, remarkable variations are also evident in both the types and the degrees of severity of such contamination [5]. The distribution and accumulation of heavy metals in urban soils are closely related to the type of land use, with high contamination areas of heavy metals in soils mainly located in traffic areas, commercial areas, industrial areas and other areas where human activity is more intense [6,7,8]. Urban soil pollution can be attributed to two primary sources: natural and anthropogenic. The natural source is characterized by the weathering of bedrock and parent material, while the anthropogenic source is dominated by industrial emissions, fossil fuel combustion, pesticide and fertilizer use and traffic emissions. Multiple studies have demonstrated that industry and traffic represent the predominant sources of heavy metals in urban soil [9,10,11,12].

In recent years, there has been a significant focus on environmental magnetic studies of heavy metal contamination in urban soils. The magnetic enhancement of urban soils is closely related to human activities. The presence of magnetic particles, originating from human activities such as industrial production, fossil fuel combustion, and automobile exhaust emissions, has been observed to accumulate within the soil. The morphology and properties of these particles exhibit significant differences when compared to naturally occurring magnetic substances [13,14]. Magnetic minerals produced by natural soil formation are primarily superparamagnetic particles (SP), while those produced by anthropogenic activities are predominantly single-domain (SD) and multi-domain (MD). Soil affected by anthropogenic factors demonstrates reduced frequency susceptibility and elevated magnetic susceptibility. Magnetic susceptibility directly correlates with the degree of contamination, with higher levels indicating greater contamination. These magnetic characteristics offer a straightforward and efficient approach for rapid identification of contaminated soils [15,16,17].

In this research, the urban topsoil of Xuzhou City was selected as the study subject. A meticulous analysis was performed on the pollution characteristics of heavy metals in the urban topsoil. Concurrently, the correlation between heavy metals and magnetic parameters was explored. By using geochemical and environmental magnetic technology measurements, an in-depth examination of the magnetic response characteristics of heavy metals was made possible. Furthermore, a comprehensive investigation was carried out, employing multivariate statistical and environmental magnetic approaches to trace the origins of both heavy metals and magnetic minerals. This integrative study aims to provide a more profound understanding of the heavy metal pollution scenario in Xuzhou’s urban topsoil, its connection with magnetic properties, and the underlying sources contributing to this complex environmental issue. This study establishes a pollution monitoring system through environmental magnetic technology, offering low-cost and high-efficiency sustainable solutions for urban soil governance and facilitating the green transformation of resource-based cities. The findings not only provide theoretical support for constructing urban ecological security barriers but also offer an innovative paradigm for synergistic management of sustainable land resource utilization and pollution control through precise pollution source identification mechanisms, holding significant reference value for the cities facing similar heavy metal pollution challenges.

2. Materials and Methods

2.1. Sample Collection and Pretreatment

The sampling area is in Xuzhou City, Jiangsu Province, a historic coal-dependent industrial hub in eastern China. Xuzhou is crucial to this study because of its resource-intensive economy and role as a transitional ecological zone. As China’s largest coal production base, contributing over 80% of Jiangsu Province’s coal output, it has a long-standing history of mining and related industries like coal-fired power plants, coking, and heavy machinery manufacturing, which are known sources of heavy metal emissions (e.g., Cd, Pb, Cr) through atmospheric deposition and wastewater discharge. Geochemically, located at the junction of the Yellow River alluvial plain and the southern Shandong karst highlands, Xuzhou shows diverse soil properties (pH: 6.2–8.5; organic matter: 1.4–3.8%), enabling the study of heavy metal mobility in different soil matrices. These unique urban characteristics provide abundant samples and typical research scenarios for studying the impact of urbanization and industrialization processes on soil heavy-metal pollution and exploring the use of environmental magnetism for pollution monitoring.

The research area of this study is the central urban area of Xuzhou City, where the population, economy, transportation, etc., are highly concentrated. The area of this region is approximately 660 square kilometers. To ensure data accuracy, sample collection was carried out in the urban zone of Xuzhou City, where no rainfall occurred for at least a week before sampling. A systematic grid-based sampling strategy was used to select 71 sampling points (Figure 1). From these, 142 samples were collected at two depths (0–2 cm and 3–10 cm) per point to study the vertical distribution of heavy metals in soil. Manual sampling was performed with a chemically inert hard plastic shovel to avoid cross-contamination. The samples were subjected to multi-point homogenization and quartering for quality control. About 2.0 kg of each homogenized sample was sealed in pre-cleaned polyethylene bags. After air-drying at room temperature, impurities were removed, and the samples were sieved through a 2 mm nylon sieve. A part was further ground and sieved through a 200 mesh sieve for measuring heavy metal concentrations and magnetic parameters.

2.2. Determination of Heavy Metals

A total of 0.2 g of soil sample was accurately weighed and placed into a polytetrafluoroethylene (PTFE) beaker, followed by the addition of 10 mL of concentrated HCl. The mixture was then heated at 150~160 °C until the liquid was reduced to approximately 3 mL. Following a cooling period, 5 mL of concentrated HNO₃ was added and the mixture was heated at 220~260 °C until it acquired a jelly-like consistency. Next, 5 mL of HF and 3 mL of HClO₄ were added and the mixture was heated until there was no white smoke. After cooling, 1 mL of HNO₃ was added and dissolved in a warm water bath. The residue was then cooled and diluted to a volume of 50 mL. Finally, the heavy metal concentrations of Cr, Cu, Fe, Mn, Ni, Pb, and Zn were determined by inductively coupled plasma mass spectrometry (ICP-MS) [18].

The precision and accuracy of metal concentrations were evaluated by employing certified reference materials. For quality assurance and quality control purposes, two soil samples, namely ESS-3 (GB 15618-2018) [19] and ESS-4 (GB 15618-2018), along with a solution sample (GB 15618-2018), were utilized. The recoveries of the investigated heavy metals all fell within the 90–110% range.

2.3. Determination of Magnetic Parameters

Approximately 10 g of soil sample was weighed and placed in an 8 cm³ sample box. The low-frequency magnetic susceptibility (χ_lf, 0.47 kHz) and high-frequency magnetic susceptibility (χ_hf, 4.7 kHz) were determined using a MS2 magnetic susceptibility meter (Bartington Instruments, Witney, UK). The frequency susceptibility was calculated by the following formula:

χ_fd% = (χ_lf − χ_hf)/χ_lf × 100%

Thereafter, χ was referred to as the low-frequency magnetic susceptibility.

The anthysteretic remanent magnetization (ARM) was determined by a rotating magnetometer after applying an alternating magnetic field with a peak value of 100 mT and a DC field of 0.04 mT to the soil sample using an alternating demagnetizer, and the susceptibility of ARM (χ_ARM) was calculated by the formula χ_ARM = ARM/31.84.

The samples were exposed to magnetic fields of 20 mT, 100 mT, 300 mT, and 1000 mT, respectively, using a pulsed magnetization meter to obtain the remanence. Subsequently, the isothermal remanence (IRM) values of IRM20mT, IRM100mT, IRM300mT, and IRM1000mT were measured using a rotating magnetometer. The residual magnetism retained after magnetization in a 1 T field is called saturated isothermal remanence (SIRM), i.e., SIRM = IRM_1000mT [20].

2.4. Evaluation Method

To comprehensively evaluate the contamination status and ecological risks of heavy metals in soil, three widely recognized indices were employed: the geo-accumulation index (I_geo), pollution load index (PLI), and potential ecological risk index (RI).

2.4.1. Geo-Accumulation Index (I_geo)

The I_geo quantifies the enrichment of heavy metals relative to natural background levels, calculated as

$${\mathit{I}}_{\mathrm{geo}}={\log }_2({\displaystyle \frac{{\mathit{C}}_{\mathit{n}}}{1.5\times {\mathit{B}}_{\mathit{n}}}})$$

where C_n is the measured concentration of metal n, B_n represents the geochemical background value of metal n, in the study region, and the constant 1.5 accounts for lithogenic variability. The contamination degree was classified into seven categories [21]: I_geo ≤ 0 (unpolluted), 0–1 (unpolluted to moderately polluted), 1–2 (moderately polluted), 2–3 (moderately to heavily polluted), 3–4 (heavily polluted), 4–5 (heavily to extremely polluted), and >5 (extremely polluted).

2.4.2. Pollution Load Index (PLI)

PLI evaluates the integrated pollution level of multiple heavy metals. The calculation formula is as follows:

$$\mathrm{PLI}\hspace{0.17em}=\hspace{0.17em}\sqrt[n]{{\displaystyle \frac{{\mathit{C}}_1}{{\mathit{B}}_1}}\times {\displaystyle \frac{{\mathit{C}}_2}{{\mathit{B}}_2}}\times \dots \times {\displaystyle \frac{{\mathit{C}}_{\mathit{n}}}{{\mathit{B}}_{\mathit{n}}}}}$$

PLI values were interpreted as: PLI < 1 (no pollution), 1–2 (moderate pollution), 2–3 (significant pollution), and PLI > 3 (severe pollution) [22].

2.4.3. Potential Ecological Risk Index (RI)

The potential ecological risk index (RI), proposed by Hakanson [23], evaluates ecological risks by integrating the toxicity of heavy metals. Single-metal risk (E_r) and total risk (RI) were calculated as

$${\mathit{E}}_{\mathit{r}}\hspace{0.17em}=\hspace{0.17em}{\mathit{T}}_{\mathit{r}}\times {\displaystyle \frac{\mathit{C}}{\mathit{B}}},\mathrm{RI}\hspace{0.17em}=\hspace{0.17em}{\displaystyle \sum {\mathit{E}}_{\mathit{r}}}$$

where T_r is the metal-specific toxicity factor (Cr = 1, Cu = 5, Fe = 1, Mn = 1, Ni = 5, Pb = 5, Zn = 1). Risk levels:

• E_r: Low (<40), Moderate (40–80), Considerable (80–160), High (160–320), Extreme (≥320)
• RI: Low (<150), Moderate (150–300), High (300–600), Extreme (≥600).

2.5. Data Analysis Methods

SPSS 27.0 software was employed to conduct descriptive statistical analysis, correlation analysis, and principal component analysis on the heavy metal concentrations and magnetic parameters. ArcGIS 10.8 was utilized to generate the distribution map of sampling points, while Surfer 13.0 was applied to create the spatial distribution map. The remaining charts were plotted using Origin 2021 and Excel 2023.

3. Results and Discussion

3.1. Heavy Metal Concentrations and Spatial Distribution

3.1.1. Statistics of Heavy Metal Concentrations

The statistical results of heavy metal concentrations in the topsoil of the Xuzhou urban area are presented in Table 1. The mean concentrations of Cu, Ni, Pb, and Zn were 1.2 to 1.7 times higher than the soil background values, indicating that there may be additional heavy metal inputs brought about by human activities. The average concentrations of Cr, Fe, and Mn are comparable to or lower than the background values. This may be influenced by factors such as the nature of the parent material of the soil itself, the leaching effect of the soil, and the absorption by organisms. It is evident that the urban soil of Xuzhou City is contaminated with varying degrees of heavy metals, particularly Cu, Ni, and Zn, which exhibit the most pronounced levels of pollution.

In multiple cities, the heavy metal content in urban soils has been detected to exceed the local soil background values, which was consistent with the findings of this study. For example, in Shanghai, China, the topsoil contains elevated levels of Cd, Hg, Pb, Cu, Zn, and Ni compared to the background values [24]. Similarly, in Guangzhou, China, the urban topsoil exhibits Zn, Cr, Pb, Cu, and Ni concentrations far surpassing the background values [25]. In Lublin, Poland, the Cu and Zn concentrations in roadside soils are five-fold higher than the background value, and the Ni, Cr, and Zn concentrations in residential soils are twice the background value [26].

The concentrations of heavy metals in the 0–2 cm soils were significantly higher than those in the 3–10 cm soils, with the exception of Fe. This suggests that heavy metals deposited on the soil surface have migrated downward due to physical processes such as surface runoff [27]. These findings further indicate that the presence of these metals in the soil may be attributed to exogenous inputs.

The coefficient of variation (CV) has been demonstrated to reflect the average variation in heavy metal concentrations among sampling sites and the intensity of human activities. The larger the coefficient of variation is, the more significant the influence of human activities is. Generally, a CV of less than 15%, between 15% and 35%, and greater than 35% has been shown to represent low variability, moderate variability, and high variability, respectively [28]. In this study, the CVs of copper, nickel, lead, and zinc in the soil all indicate strong variability, suggesting that the concentrations of these heavy metals are significantly influenced by anthropogenic factors.

Table 1. Descriptive statistics of heavy metals concentrations in urban topsoil.

	0–2 cm			3–10 cm			Background Value [29]
	Range	Mean	CV/%	Range	Mean	CV/%
Cr	11.30–111.83	65.57	27.61	24.60–141.20	62.41	28.20	64.80
Cu	9.50–246.75	47.02	75.13	15.26–140.80	42.12	55.59	24.30
Fe	0.42–4.20	2.47	20.71	0.83–4.44	2.52	20.12	3.10
Mn	72.50–1272.10	525.39	26.63	153.40–975.20	515.55	22.76	633.00
Ni	10.25–547.30	60.00	126.95	19.07–489.00	52.13	127.63	30.70
Pb	7.25–82.80	27.70	53.37	4.75–162.80	26.19	80.14	21.30
Zn	21.00–389.80	141.61	49.89	62.90–413.00	126.93	49.92	74.10

Table 1. Descriptive statistics of heavy metals concentrations in urban topsoil.

	0–2 cm			3–10 cm			Background Value [29]
	Range	Mean	CV/%	Range	Mean	CV/%
Cr	11.30–111.83	65.57	27.61	24.60–141.20	62.41	28.20	64.80
Cu	9.50–246.75	47.02	75.13	15.26–140.80	42.12	55.59	24.30
Fe	0.42–4.20	2.47	20.71	0.83–4.44	2.52	20.12	3.10
Mn	72.50–1272.10	525.39	26.63	153.40–975.20	515.55	22.76	633.00
Ni	10.25–547.30	60.00	126.95	19.07–489.00	52.13	127.63	30.70
Pb	7.25–82.80	27.70	53.37	4.75–162.80	26.19	80.14	21.30
Zn	21.00–389.80	141.61	49.89	62.90–413.00	126.93	49.92	74.10

3.1.2. Evaluation of Heavy Metal Pollution

The geo-accumulation index (I_geo) is a critical method for assessing heavy metal contamination in soil and quantifying anthropogenic contributions, as it evaluates elemental enrichment relative to geological background values, thereby effectively reflecting human-induced disturbances in topsoil. In this study, the I_geo was employed to analyze heavy metal contamination characteristics in urban soils at depths of 0–2 cm (topsoil) and 3–10 cm (subsoil). The results revealed significantly higher contamination levels of Cr, Cu, Ni, Pb, and Zn in the 0–2 cm topsoil compared to the 3–10 cm subsoil (Figure 2). Cu and Zn exhibited the most widespread contamination, with 38.03% of Cu samples and 40.14% of Zn samples in the topsoil classified as slightly contaminated, while 9.86% and 10.56% reached moderately contaminated levels, respectively. Ni posed a moderate contamination risk, with 13.38% of samples slightly contaminated and 12.68% moderately contaminated. Pb showed a relatively high proportion of slight contamination (19.72%) but a low proportion of moderate contamination (2.82%). Cr contamination was minimal, with only 5.63% of samples slightly contaminated. Fe and Mn did not exceed background thresholds in any samples.

These findings indicate a pronounced enrichment of heavy metals in the topsoil, likely attributable to direct inputs from atmospheric deposition, traffic emissions, and other surficial processes affecting the 0–2 cm layer. The elevated contamination proportions of Cu and Zn may stem from urban-specific sources such as coating corrosion and tire wear, whereas Ni and Pb contamination warrants further investigation into potential industrial emissions or historical pollution legacies.

The pollution load index (PLI) is a widely utilized metric for assessing the degree of soil contamination by heavy metals. A PLI value less than or equal to 1 indicates non-contamination. Values in the range of 1–2 signify mild contamination, and those in the range of 2–3 signify moderate contamination. A PLI value greater than or equal to 3 reflects heavy contamination. As illustrated in Figure 3, the PLI for heavy metals in the 0–2 cm soils is higher than that in the 3–10 cm soils at most sampling points. The distribution of samples classified by pollution levels demonstrates that 3.5% of the samples exhibit heavy pollution, 18.3% display moderate pollution, and a substantial majority, accounting for 66.9%, are categorized as lightly polluted. Notably, only a small proportion (11.3%) of the samples are determined to be unpolluted.

The potential ecological risk index (RI) results indicated that all studied elements exhibited individual ecological risk factors (Ei) below 40, corresponding to a low ecological risk level. Only copper (Cu) and nickel (Ni) showed minor proportions of moderate risk, accounting for 0.7% and 2.8% of samples, respectively (

Table 2. Potential ecological risk index of heavy metals in urban topsoil.

Single Potential Ecological Risk Index	Element	Ei Range	Ei Mean	Percentage of Samples with Different Risk Levels/%
				Slight	Moderate	Strong and Above
Ei	Cr	0.35–3.50	1.96	100.00	0.00	0.00
	Cu	1.95–50.77	9.13	99.30	0.70	0.00
	Fe	0.30–1.29	0.80	100.00	0.00	0.00
	Mn	0.17–1.49	0.82	100.00	0.00	0.00
	Ni	1.67–47.00	8.57	97.20	2.80	0.00
	Pb	1.12–21.08	6.21	100.00	0.00	0.00
	Zn	0.28–4.77	1.80	100.00	0.00	0.00
Total Potential Ecological Risk Index		RI Range	RI Mean	Percentage of Samples with Different Risk Levels/%
				Slight	Moderate	Strong and Above
RI		6.42–91.38	29.29	100.00	0.00	0.00

). The comprehensive potential ecological risk index (RI) further confirmed that all samples were classified as low risk. Notably, the RI methodology integrates not only heavy metal concentrations but also their toxicity response coefficients. Despite detectable accumulations of Cu, Ni, lead (Pb), and zinc (Zn), the overall ecological risk remained low due to two key factors: One is the relatively low toxicity weights of these metals (e.g., Cu = 5, Ni = 5, Pb = 5, Zn = 1). The other is the dilution effect caused by high concentrations of low-toxicity metals such as iron (Fe) and manganese (Mn) (toxicity coefficients: Fe = 1, Mn = 1).

These findings suggest that, from an ecological risk perspective, the heavy metals in the studied soils pose minimal immediate threats to ecosystem health. However, the localized moderate risks associated with Cu and Ni (albeit at low frequencies) warrant attention to specific anthropogenic sources, such as industrial discharges or improper waste management.

3.1.3. Spatial Distribution of Heavy Metal Pollution

In order to visually depict the distribution of heavy metal pollution in the topsoil of the Xuzhou urban area, the PLI of heavy metals was spatially interpolated and analyzed. The results are shown in Figure 4. The spatial distribution map reveals that the metal pollution in 0–2 cm and 3–10 cm soil layers exhibit similar spatial distribution characteristics. The PLI exhibits elevated values in the central and north-eastern sections of the study area. The central part of the study area is the city center of Xuzhou, which is characterized by dense traffic. In contrast, the north-eastern part of the study area is characterized by the presence of numerous factories, including cement, machinery and chemical factories, all within a 10 km radius. It is hypothesized that the elevated PLI in this region is closely related to transportation and industrial production. The PLI in the southwestern part of the study area presents a low value distribution. This area is located in the vicinity of the natural scenic Yunlong Lake, and the presence of industrial pollution sources is absent in the surrounding area. Additionally, motor vehicles are prohibited from entering the area, thereby minimizing anthropogenic impacts. Consequently, the soil in this area is less affected by the input of heavy metals from anthropogenic sources.

3.2. Magnetic Properties

Magnetic parameters are instrumental in the characterization of the type, content, and size of magnetic minerals present within the soil. The magnetic minerals of general significance in nature are subferromagnetic minerals and incomplete antiferromagnetic minerals. The magnetic parameters χ, SOFT, SIRM, and χ_ARM have been demonstrated to be effective indicators of the type and concentration of magnetic minerals.

The magnetic properties of substances are significantly influenced by the grain size. Magnetic minerals with different grain sizes exhibit remarkably distinct magnetic characteristics. Grain size is typically expressed in terms of magnetic domains, and the classification of grains is typically based on their size. Grains can be classified as multi-domain (MD, >1 μm), pseudo-single-domain (PSD, 0.08–1 μm), single-domain (SD, 0.02–0.08 μm), and superparamagnetic (SP, <0.05 μm), among others [30,31,32]. The composition and differences in the grains in the samples can be discerned through the analysis of a variety of magnetic parameters, such as χ_fd and the ratio parameters χ_ARM/χ and χ_ARM/SIRM.

The results of the magnetic parameters measured are presented in

Table 3. Statistical results of the soil magnetic properties.

Depth	Magnetic Parameters	Mean	Range	CV/%
0–2 cm	χ/(10−8 m3·kg−1)	132.3	22.4–613.5	97.0
	χfd/%	3.5	0.3–11.2	66.5
	SOFT/(10−5 A·m2·kg−1)	719.4	89.4–2753.9	91.0
	SIRM/(10−5 A·m2·kg−1)	1530.2	199.8–6471.3	90.1
	χARM/(10−8 m3·kg−1)	285.0	3.3–787.2	58.4
	χARM/χ	2.7	0.1–9.5	45.0
	χARM/SIRM/(10−3 m·A−1)	0.2	0.01–0.5	37.9
3–10 cm	χ/(10−8 m3·kg−1)	99.6	11.4–823.5	113.4
	χfd/%	4.8	0.4–11.7	54.5
	SOFT/(10−5 A·m2·kg−1)	494.4	104.4–3208.0	91.9
	SIRM/(10−5 A·m2·kg−1)	1088.3	225.6–7049.2	87.1
	χARM/(10−8 m3·kg−1)	169.4	5.4–846.9	77.0
	χARM/χ	2.3	0.1–5.2	112.3
	χARM/SIRM	0.2	0.0–0.4	58.1

. It is evident that the values of each magnetic parameter are greater in soil from 0 to 2 cm depth than in soil from 3 to 10 cm, except for χ_fd.

The enhancement of soil magnetic susceptibility has been verified to be closely related to human activities [33,34]. Research findings have indicated the presence of a significant quantity of magnetic particles in industrial fly ash derived from fossil fuel combustion. These particles, when deposited on topsoil, resulted in an enhancement of the magnetic susceptibility of the soil [35,36]. The mean values of χ for 0–2 cm and 3–10 cm topsoil in this study were 4.4 and 3.3 times higher than the background value (30 × 10^–8 m³·kg⁻¹), respectively [37]. This finding suggested the presence of human inputs of magnetic minerals in the soil of Xuzhou City.

SIRM indicated the total concentration of ferrimagnetic minerals and incomplete antiferromagnetic minerals in the samples. SOFT was employed to determine the content of ferrimagnetic minerals, with a particular focus on the content of multidomain (MD) and pseudo-single-domain (PSD) particles. It was further demonstrated that both SIRM and SOFT exhibited a significant positive correlation with χ (Figure 5), thereby indicating that the magnetic minerals present in the topsoil of the study area are predominantly ferrimagnetic minerals characterized by MD and PSD.

The χ_fd can be utilized to estimate the relative content of SP particles in a given sample. When the χ_fd of the sample is found to be less than 2%, coarse grains are predominant; when χ_fd is greater than 10%, fine grains are dominant; and when χ_fd is less than 5%, the contribution of SP particles is quite weak [38]. The mean χ_fd value is 3.5% and 4.8% for the 0–2 cm and 3–10 cm topsoil, respectively, in this study, indicating that the magnetic minerals present in the topsoil of Xuzhou are predominantly coarse particles, specifically MD and PSD.

The ratio parameters χ_ARM/χ and χ_ARM/SIRM can be used to indicate the size of the subferromagnetic mineral particles. Studies have shown that these ratios decrease as particle size increases. Specifically, when χ_ARM/χ is less than 4 and χ_ARM/SIRM is below 0.6 × 10^–3 m·A⁻¹, it is reasonable to infer that the predominant magnetic mineral in the sample consists of PSD-MD grains [39].

In the study area, the mean values of χ_ARM/χ and χ_ARM/SIRM for the 0–2 cm topsoil were determined to be 2.7 and 2.3, respectively. For the 3–10 cm topsoil, the mean values of both χ_ARM/χ and χ_ARM/SIRM were 0.2. These results provide evidence that the magnetic minerals in the topsoil are predominantly coarse-grained particles of PSD and MD. Previous research by Lu et al. [40] has demonstrated that MD and PSD magnetic minerals are primarily sourced from anthropogenic activities, including traffic emissions and industrial production. The findings in this study, thus, further suggest the presence of anthropogenic inputs of magnetic particles within the study area.

3.3. Magnetic Response and Traceability of Heavy Metals

3.3.1. Correlation Analysis

Correlation analyses were performed between the magnetic parameters and heavy metal concentrations of the soils, and the results are presented in Figure 6. In the figure, the larger the circle, the greater the absolute value of the correlation coefficient. As depicted in the figure, the magnetic parameters that reflect the content of ferromagnetic minerals (χ, SOFT, SIRM, and χ_ARM) exhibit a strong correlation with the concentrations of Pb and Zn, with correlation coefficients ranging from 0.63 to 0.75. Additionally, these magnetic parameters are significantly correlated with Cr, Cu, and Fe, albeit with relatively weaker correlations, since the correlation coefficients are in the range of 0.23 to 0.42. Simultaneously, χ, SOFT, SIRM, χ_ARM, and the PLI demonstrate significant positive correlations, with correlation coefficients ranging from 0.58 to 0.68, and they are strongly intercorrelated. These findings indicate that the magnetic parameters can effectively serve as an indicator to reflect the contamination of urban topsoil in Xuzhou City, signifying the degree of heavy metal contamination in the soil.

Similar conclusions have been reached in several studies. For example, Lu et al. [41] found that the correlation coefficients between χ, Pb, and Zn in the urban soils of Beijing reached 0.65–0.71, indicating that magnetic minerals (such as magnetite and hematite) can serve as sensitive indicators of anthropogenic heavy metal pollution. Similarly, Yang et al. [42] found that in the industrial city of Wuhan, SOFT was closely related to Fe and Mn particles emitted by traffic. Due to its characteristics of being rapid, non-destructive and low-cost, the environmental magnetism method can screen for pollution over large areas, provide a scientific basis for soil protection and remediation, and contribute to environmental governance.

3.3.2. Principal Component Analysis (PCA)

To further elucidate the sources of heavy metal contamination, Principal Component Analysis (PCA) was performed on the above magnetic parameters and heavy metals. The results are presented in

Table 4. Principal component analysis of heavy metals and magnetic parameters.

Variable	PCA1	PCA2	PCA3	PCA4
Cr	0.599	0.502	0.062	0.239
Cu	0.427	0.004	0.740	0.286
Fe	0.601	0.693	−0.144	−0.180
Mn	0.245	0.842	−0.197	−0.286
Ni	0.180	0.140	0.830	−0.372
Pb	0.861	0.075	−0.069	0.228
Zn	0.900	0.136	0.059	0.075
χ	0.856	−0.315	−0.168	−0.051
χfd	−0.490	0.537	0.003	0.468
SOFT	0.912	−0.247	−0.120	−0.008
SIRM	0.915	−0.256	−0.096	−0.004
χARM	0.854	−0.170	−0.021	0.001
Eigenvalue	5.918	2.027	1.359	0.670
Variance contribution ratio/%	49.321	16.893	11.329	5.585
Cumulative contribution ratio/%	49.321	66.214	77.542	83.127

. The analysis shows that the Kaiser–Meyer–Olkin (KMO) value is 0.829, which exceeds 0.5, and the p-value is 0.000, less than 0.050. This indicates a strong correlation between the variables, rendering the data suitable for principal component analysis. Integrating the results of the correlation analysis, a total of four principal components were extracted, with a cumulative contribution rate of 83.13%. Based on this, the existence of four major sources of heavy metal and magnetic mineral contamination in the soil of the study area was inferred.

Based on the results of PCA, the first principal component (PCA1) accounts for 49.32% of the total variance. It exhibits strong loadings on Pb, Zn, and the magnetic parameters χ, SOFT, SIRM, and χ_ARM, with loading coefficients all exceeding 0.85. Additionally, PCA1 has relatively high loadings on Cr, Fe, and Cu. The correlation analysis further indicates significant intercorrelations among these variables, suggesting a close relationship. Since χ, SOFT, SIRM, and χ_ARM are parameters reflecting the content of ferrimagnetic minerals in the samples, it can be deduced that the ferrimagnetic minerals in the samples share the same origin as these metal elements.

Previous studies [43,44,45,46,47,48,49] have demonstrated that the accumulation of elements such as Pb, Zn, Cu, and Cr in urban soils is associated with factors including automobile exhaust emissions, tire wear, brake-pad friction, lubricant addition and fossil-fuel combustion. As the above-mentioned results have shown that the magnetic particles in the urban soils of Xuzhou City bear the characteristics of anthropogenic inputs, and that transportation and fossil-fuel combustion are the main anthropogenic sources of magnetic minerals [33,50,51], PCA1 is, thus, inferred to be a mixed source of transportation and fossil-fuel combustion.

The second principal component (PCA2) contributes 16.89% to the overall variance. It has high loadings on χ_fd, Cr, Fe, and Mn. The parameter χ_fd serves as an indicator of superparamagnetic (SP) particles that are associated with natural soil formation processes. Cr, Fe, and Mn are crucial elements involved in soil formation [52]. Significantly, the concentrations of these three elements do not exceed their respective background values. Given these facts, PCA2 can be attributed to the influence of natural soil-forming parent materials, representing a natural source for the soil composition in the studied area.

The third principal component (PCA3) accounts for 11.33% of the total variance, presenting high loadings on Cu and Ni. Besides natural origins, industrial procedures like electroplating, metal processing, and manufacturing operations can result in the enrichment of Cu and Ni in the soil [53,54]. Table 2 reveals that the concentrations of Cu and Ni in the soil of Xuzhou urban area are 1.65 and 1.95 times higher than their respective background levels. This finding strongly suggests the existence of human-induced inputs of Cu and Ni into the soil.

Analyzing the spatial distribution, high-value regions of Cu and Ni are predominantly concentrated in the north-eastern part of Xuzhou City. Notably, within a 10 km radius of the north-eastern part of the study area, there is a dense distribution of hardware and building materials manufacturing enterprises. Taking these factors into account, it is reasonable to assume that PCA3 can be attributed to an industrial production source.

The fourth principal component (PCA4) has a contribution rate of 5.58%. Given its high loading on χ_fd, it is hypothesized that PCA4 represents natural sources, including volcanic activity and biological processes, which are distinct from those involved in natural soil-formation mechanisms.

This conclusion is consistent with the results of several studies. Fazekašová et al. [55] carried out research to assess the origins of heavy metal contamination in the soil of agricultural areas in Slovakia. Their results showed remarkable exceedances of elements such as Ni, Cd, and Cu in this region, with industrial activities, mining, and transport identified as the main sources of pollution. Cao et al. [56] investigated the leaching potential of heavy metal elements in coal and the factors influencing it. Their work indicated that coal combustion and industrial activities play a crucial role as sources of heavy metal pollution. In summary, industrial and transport activities are identified as the dominant sources of heavy metal pollution in urban soils of many cities. This finding provides a scientific basis for formulating strategies to prevent and control heavy metal pollution in urban soils. A series of policies can be effective in alleviating heavy metal pollution. These policies encompass the rational planning of urban road networks to minimize traffic congestion, the energetic development of public transportation, the encouragement of residents to embrace green travel modes, the enhancement of supervision over industrial enterprises, and the promotion of the adoption of clean production technologies by enterprises.

4. Conclusions

In this study, the concentrations of Cu, Ni, Pb, and Zn all exceeded the soil background values. Specifically, the concentrations of Cu, Ni, and Zn were 1.7–1.9 times the soil background values, while those of Pb were 1.2–1.3 times the background values. Notably, the concentrations of Cr, Cu, Mn, Ni, Pb, and Zn in the 0–2 cm layer samples were higher than those in the 3–10 cm layer samples, suggesting that these metals in the soil are likely of exogenous origin.

The integrated assessment using the geo-accumulation index (I_geo), pollution load index (PLI), and potential ecological risk index (RI) revealed surface-enriched contamination of Cr, Cu, Ni, Pb, and Zn in the 0–2 cm topsoil (1.3–2.0× background values), with Cu and Zn exhibiting the highest contamination prevalence. Despite elevated concentrations, the overall ecological risk remained low, attributed to the low toxicity weights of dominant metals and dilution effects from Fe and Mn.

The magnetic parameters χ, SOFT, SIRM, and χ_ARM, which indicate the content of magnetic minerals, showed high values. Moreover, the magnetic minerals in the topsoil were mainly coarse-grained, belonging to the pseudo-single-domain (PSD) and multi-domain (MD) categories. A strong correlation was observed between these magnetic parameters and the elements Pb and Zn, as well as the PLI, indicating that these magnetic parameters can be used as surrogate indicators of soil heavy metal pollution.

Through multivariate statistical analysis, combined with environmental magnetic analysis, four major sources of heavy metal pollution have been identified, namely transportation and fossil fuel combustion, industrial production, natural soil formation, and other natural sources. Transportation and fossil fuel combustion are the primary sources of magnetic minerals, Pb and Zn. Natural soil-formation processes mainly contribute to Cr, Fe, and Mn. Industrial production is the origin of Cu and Ni. Some of the Cr, Cu, and Fe also come from transportation and fossil fuel combustion. These findings provide valuable insights into the sources and characteristics of soil pollution in the studied area.