Assessment of Potentially Toxic Metals (PTMs) Pollution, Ecological Risks, and Source Apportionment in Urban Soils from University Campuses: Insights from Multivariate and Positive Matrix Factorisation Analyses

Abstract

Understanding pollution levels, ecological health risks, and sources of potentially toxic metals (PTMs) in the soil from university campuses is critical for assessing environmental safety. Soil samples were collected from 12 locations across urban parks and green areas at Sohag University in Egypt. The samples were processed and analysed for heavy metals, including iron (Fe), manganese (Mn), cobalt (Co), nickel (Ni), chromium (Cr), lead (Pb), zinc (Zn), copper (Cu), and cadmium (Cd). Pollution levels were evaluated using indices such as the pollution index (PI), pollution load index (PLI), geo-accumulation index (I_geo), and enrichment factors (EFs). Among the pollution indices, the EFs showed the highest sensitivity in detecting anthropogenic contributions, particularly for Cd, Pb, and Cr. Spatial distribution maps and multivariate statistical analyses, including correlation matrix (CM), principal component analysis (PCA), and cluster analysis (CA), were applied to identify the relationships between PTMs and soil properties, and source apportionment was performed using positive matrix factorisation (PMF). The results indicated that Mn, Ni, and Co were primarily geogenic, whereas Pb, Zn, Cr, and Cd showed higher concentrations, suggesting moderate-to-significant anthropogenic pollution. Pb and Cd pose considerable ecological risks, whereas other metals such as Cr and Cu exhibit moderate ecological threats. The non-carcinogenic and carcinogenic risks to the students were within safe limits, as defined by United States Environmental Protection Agency (USEPA) threshold values. Source apportionment using PMF identified five main sources of PTMs: industrial and anthropogenic activities (30.0%), traffic emissions (25.0%), natural soil processes (20.0%), agricultural practices (15.0%), and mixed industrial traffic sources (10.0%). These findings emphasise the importance of controlling anthropogenic activities to ensure a safer campus environment.

1. Introduction

Heavy metals (HMs) in soil are recognised for their considerable toxicity, and elevated concentrations can severely disrupt plant growth as well as animal and human health [1,2,3]. Human exposure to these elements may result in various health issues, including headaches, dizziness, insomnia, memory impairment, neurological disorders, joint discomfort, and more serious ailments, such as liver, gastric, colorectal, bladder, breast, and prostate cancers [4]. Parks and green areas play a vital role in providing local communities with spaces for relaxation, play, and socialisation, often serving as important venues for family gatherings after work hours [5]. The rapid economic development witnessed in recent decades has highlighted these areas as key indicators of residents’ quality of life and urbanisation levels of surrounding areas [6,7]. The presence of potentially toxic metals (PTMs) in the soils of parks and green areas is a significant environmental issue because of their potentially harmful effects on human health and ecological systems [8,9,10]. Soil contamination from PTMs can be traced to various anthropogenic activities, including the use of pesticides and chemical fertilisers, as well as pollution from transportation and industrial activities [11,12,13,14].

While much of the research has focused on urban soils and contamination in general, specific attention is needed for urban environments where educational institutions, such as university campuses, serve as focal points of social and recreational activity. University campuses represent unique environments where human exposure to heavy metals in soils may differ from that in other urban areas due to specific land uses, traffic patterns, and recreational activities. University campuses worldwide often serve as microcosms of urban environments [15,16], reflecting similar patterns of anthropogenic and natural pollution sources [17,18]. By examining PTM contamination and its sources in campus soil, this study offers insights that can guide environmental monitoring and mitigation strategies in similar contexts worldwide. The methodologies used, including positive matrix factorisation (PMF), pollution indices, and spatial analysis, are adaptable and applicable to other urban environments.

The build-up of PTMs in urban soils increases the risk of exposure through multiple pathways, including ingestion, inhalation, and dermal contact, emphasising the need for consistent monitoring and assessment of soil quality in these recreational spaces to protect public health and ensure the safety of park users [19,20]. Consequently, the evaluation of the health risks posed by PTMs has become a significant concern in environmental science, ecology, and geography. Considering the environmental risks posed by PTMs in urban parks and green spaces, research initiatives have focused on examining PTM concentrations, their distribution patterns, source attribution, and the associated ecological and human health hazards in soils within these areas [21,22,23,24]. This research spans multiple disciplines, including ecology, environmental science, and toxicology, highlighting the complex nature of soil pollution and its impacts. While much of the existing literature has focused primarily on measuring the concentration levels and assessing the ecological risks of PTMs in urban parks [25,26], this study aims to offer a more comprehensive understanding by integrating source apportionment and spatial analysis to identify the primary contributors to contamination.

University campuses are significant functional areas that raise concerns regarding PTM pollution. However, most research has concentrated on kindergarten, primary, and middle schools [27,28,29]. Educational settings, particularly those for younger children, often present unique challenges because these children are more physiologically vulnerable. In contrast, university campuses have different land use patterns and levels of activity, which include factors such as construction, traffic flow, and recreational activities. These differences can lead to varying exposure profiles and types of pollutants. Only a limited number of studies have addressed PTM pollution on university campuses [30,31,32,33]. This limited focus is potentially influenced by the common perception that younger children are more vulnerable to PTM exposure, as they are still in early developmental stages. Although college students are considered adults, their bodies are still developing, exposed to high levels of physical activity, and spend extended periods in outdoor campus environments. These behavioural patterns increase their exposure to soil and dust, particularly through dermal contact, inhalation, and incidental ingestion, making them a relevant population for evaluating environmental health risks. Engaging in intense physical activities related to sports and other campus activities increases their exposure to dust. The PTM pollution found in university campus dust is a significant concern [34], and the potential harm it poses to college students should not be underestimated. University campuses are often densely populated spaces where students, staff, and visitors engage in prolonged activities, increasing the likelihood of exposure to PTMs through ingestion, inhalation, and skin contact [35,36]. Contaminated campus soils pose long-term health risks, including both carcinogenic and non-carcinogenic effects [37], which could significantly impact students’ physical and cognitive health. Additionally, university green areas are frequently used for recreational activities, making them an essential focus of environmental monitoring to ensure a safe and healthy academic environment. The study of PTM contamination in university soils also reflects broader urban environmental challenges [38,39], providing valuable insights into urban pollution patterns and informing mitigation strategies [40]. The findings from this study contribute to the growing body of research on urban and campus environmental quality, enabling cross-regional comparisons and guiding policies aimed at reducing pollution in educational institutions.

Sohag City has a rich history and culture and hosts numerous universities and colleges. With an area of 250 acres, 17 colleges and faculties, 63,000 students in 2024, 2300 staff members, and 6300 administrative employees, Sohag University is the largest university in the Sohag Governorate, located in Upper Egypt, and is one of the major educational institutions in the region, being an ideal region to study soil PTM pollution on university campuses. It is located in a densely populated and rapidly urbanising area of Upper Egypt and presents a unique opportunity to investigate PTM accumulation in campus soils influenced by a mix of urban, agricultural, and institutional activities. This study aims to investigate the concentrations of heavy metals—specifically Fe, Cd, Co, Cr, Cu, Mn, Ni, Pb, and Zn—in the topsoil of green areas at Sohag University. In addition to determining these concentrations, this research will assess pollution levels and the associated ecological and health risks. Furthermore, this study seeks to identify the sources of these metals through the application of PMF and multivariate statistical analyses, allowing for a clear distinction between anthropogenic and natural origins of contamination.

2. Materials and Methods

2.1. Study Area

Sohag is a rural governorate in Upper Egypt, with its capital, Sohag City, located approximately 470 km south of Cairo. Geographically, the governorate spans a narrow strip of land extending over 110 km along both banks of the Nile River (Figure 1a). The Sohag Governorate experiences a typical hot desert climate (i.e., long, hot, dry summers and dry, cold winters). The average temperature during winter is approximately 6 °C, whereas it reaches approximately 38 °C in summer, with a mean annual temperature of 21.8 °C. Precipitation is extremely negligible, and the mean evapotranspiration values reach approximately 1812 mm/year. Sohag University was founded in Sohag City at 26°33′55″ N, 31°42′28″ E (Figure 1b). Apart from the spaces occupied by university buildings, infrastructure, and roads, most campus surfaces are dedicated to green areas and parks. The total area of the Sohag University campus spans approximately 250 acres, providing a substantial landscape for evaluating soil contamination within its green zones. The soil in the Nile Valley, including in the study area, is influenced by fluvial processes. These sediments consist primarily of Holocene fluviatile deposits, mainly composed of silt and clay, originating from the basaltic rocks of the Ethiopian Plateau [41,42]. To enhance the fertility of green areas and parks for planting in urban settings, a thin layer of fertile floodplain sediment is often transported and spread over these surfaces. Owing to the relatively low fertility of urban soils, there is a common reliance on chemical fertilisers and other amendments to improve soil quality.

2.2. Sampling and Analytical Techniques

In this study, soil samples were collected from 12 locations in the urban parks and green areas of Sohag University in April 2014. The campus land use and green space layout have remained largely stable, making the data valuable for baseline assessments and future comparative studies. The sampling locations were selected to ensure comprehensive coverage of the study area, accounting for variations in land use, proximity to potential pollution sources (e.g., roads, parking lots, and recreational areas), and accessibility. Each location was georeferenced using a GPS device to enable reproducibility and spatial analysis. At each location, composite samples were taken by combining 2–3 subsamples from a depth of 0–15 cm. When sampling heterogenous areas (e.g., near roads, parking lots, or agricultural fields), 3 subsamples were taken to ensure adequate representation of soil variability. For more homogeneous areas, 2 subsamples were sufficient. The exact locations for the subsamples within the 10 m radius were determined using an organised random sampling approach. The points were spread across the area to capture local variability, with a random number generator helping to determine the intervals between the selected points. This method ensured that the subsamples were representative of the area’s overall conditions.

After collection, samples were immediately placed in clean, labelled polyethylene bags and transported to the laboratory to avoid cross-contamination and preserve sample integrity.

In the laboratory, the soil samples were air-dried under controlled conditions to prevent the loss of volatile components. Dried samples were passed through a 2 mm sieve to remove larger debris, stones, and pebbles. Subsamples were further sieved through a 0.063 mm mesh to isolate fine particles, as these are more likely to retain heavy metals and represent the bioavailable fraction of the soil.

To ensure consistency, quality control measures were implemented throughout the sampling and preparation process. The hydrogen ion concentration was measured in soil samples using the <2 mm soil fraction (soil/water extract [1:2.5]) with a digital pH meter (Cole Parmer). Inorganic carbonate was estimated using a Collins calorimeter following the USDA (1996) guidelines. For samples with a particle size of <63 µm, the carbonate content was assessed by treating the samples with HCl and measuring the evolved CO2 monometrically. The amount of carbonate present was calculated as calcium carbonate (CaCO₃). The oxidisable organic carbon content was measured in the soil (<63 µm). This was determined using the modified Walkley and Black method (USDA, 1996), which employs an acid dichromate system. This method involves back-titrating unused dichromate with standard ferrous ammonium sulphate, using diphenylamine as an indicator. Assessing the total metal content in the soil is crucial for evaluating potential hazards to the ecosystem and comparing the results with quality standards regarding pollution effects and sustainability. Total metal content assessments typically involve digesting soil samples with strong acids, such as HF, HClO₄, HNO₃, and aqua regia. Although the total metal content does not always correlate well with the bioavailable fraction, it remains a common statutory measure for assessing soil contamination in many countries.

Chemical analyses were performed using hot aqua regia extraction, which effectively dissolved most sulphide minerals, carbonates, and silicates, such as trioctahedral micas, olivine, and clay minerals, as well as primary and secondary salts and hydroxides [43]. The analytical method overestimated the bioavailable PTM concentrations. The aqua regia digestion in this study measures “pseudo-total” metal concentrations rather than bioavailable fractions. No separate analysis of bioavailable fractions was performed. PTMs that are strongly bound to silicate structures are released into the environment at an extremely slow rate, indicating that high total concentrations of potentially harmful elements do not necessarily pose an environmental threat. Aqua regia extraction is widely used for several reasons: it is commonly available in geo-analytical laboratories, has been standardised, and tends to yield more consistent results than many weaker leaching methods. In this study, the concentrations obtained through aqua regia extraction are referred to as “total” concentrations; however, this method represents a strong partial leach, and the actual total concentrations of PTMs may be higher. A comparison of aqua regia-extractable concentrations and actual total concentrations in Finnish soil samples can be found in [44]. The total concentration of the environmentally relevant PTMs (Fe, Mn, Co, Ni, Pb, Zn, Cu, Cr, and Cd) was determined by digesting 1.0 g of air-dried samples (<63 µm) in 25 mL conical flasks using a concentrated aqua regia acid mixture. The total metal content was assessed spectroscopically using the resulting acid extract. A PerkinElmer 2380 flame atomic absorption spectrophotometer (AAS) with hollow cathode lamps was used to measure the total and bioavailable concentrations of the tested metals (Fe, Mn, Co, Ni, Cr, Pb, Zn, Cu, and Cd). All analyses conducted in this study were performed at the Environmental Geochemistry Laboratory of the Geology Department, Faculty of Science, Sohag University, using the aforementioned instruments. The overall sampling and analytical protocol followed established guidelines to ensure accuracy and reproducibility. While the current study used composite samples to address heterogeneity, future research is intended to consider more granular sampling approaches (e.g., grid-based sampling) for finer spatial resolution.

2.3. Assessment of PTM Pollution

To evaluate the level of PTM contamination, different pollution indices are determined. The pollution index (PI) of each metal is defined as the ratio of its concentration to the background value of the corresponding metal, as shown in Equation (1):

PI = C/S

(1)

where PI is the pollution index of each sample, C (mg/kg) is the measured concentration of each PTM, and S (mg/kg) is the background value. The metal content values from [45] are considered the natural baseline level (background). The use of continental crust values as a reference for background concentrations is a well-established practice in environmental studies [46,47], offering a standardised basis for comparison across different regions. However, it is important to acknowledge that these values may not be fully representative of local conditions, as background concentrations can vary depending on the specific geochemical characteristics of the region being studied. This limitation should be considered, particularly in areas where local background data are available or when more region-specific reference values could provide a more accurate basis for comparison. While Wedepohl’s values are widely used as reference baseline concentrations, their application in this study should be regarded as an approximation for the region’s natural background levels. This approach is acceptable in the absence of specific regional data, but it is important to recognise that the applicability of these values may be limited in certain contexts. The PI for each metal is categorised as follows: low contamination (PI ≤ 1), moderate contamination (1 < PI ≤ 3), or high contamination (PI > 3).

The (PLI) of the samples is determined using PTM data, with metal concentrations from the mean global crust serving as background values [48]. The PLI of the soil can be computed by taking the nth root of the product of the contamination factors (CFs) for all metals, where n represents the number of metals analysed, as in Equation (2):

$$\mathrm{P}\mathrm{L}\mathrm{I}=\sqrt[\mathrm{n}]{\mathrm{C}\mathrm{F}1\times \mathrm{C}\mathrm{F}2\times \mathrm{C}\mathrm{F}3\times \dots \dots \dots \times \mathrm{C}\mathrm{F}\mathrm{n}}$$

(2)

where CF = C_metal/C_background, and the values vary from unpolluted to slightly polluted (0–2) to extremely polluted (8–10).

Ref. [49] first proposed a geo-accumulation index (I_geo) to assess and characterise metal pollution in sediments by contrasting the present concentrations with pre-industrial levels, as shown in Equation (3).

I_geo = log₂[C_i/(1.5C_ri)]

(3)

where C_i is the calculated concentration of studied metal in the sediment, and C_ri is the reference value or geochemical background concentration of metal i. Owing to extremely slight anthropogenic influences and potential differences in background values for a particular metal, a factor of 1.5 was utilised. I_geo is categorised into seven classes to assess metal contamination levels: Class 0 (Igeo ≤ 0) indicates uncontaminated conditions; Class 1 (0 < I_geo ≤ 1) ranges from uncontaminated to moderately contaminated; Class 2 (1 < I_geo ≤ 2) signifies moderately contaminated soils; Class 3 (2 < I_geo ≤ 3) indicates a transition from moderately to strongly contaminated; Class 4 (3 < I_geo ≤ 4) represents strongly contaminated soils; Class 5 (4 < I_geo ≤ 5) signifies a shift from strongly to extremely contaminated; and Class 6 (I_geo > 5) indicates extremely contaminated conditions.

The EF of each element in the samples was calculated by standardising the measured element against a reference element. Commonly used reference elements include Al, Fe, and K [50,51]. The EF is expressed using Equation (4) below.

EF = (Cx/Fe)_soil/(Cx/Fe)_background

(4)

where (Cx/Fe)_soil represents the metal-to-Fe ratio in the samples of interest, and (Cx/Fe)_background denotes the natural background value of the metal-to-Fe ratio. In this study, because of the unavailability of background values for Fe and heavy metals specific to the study area, average metal values from the continental crust were utilised. The anticipated EF classes, along with their corresponding sediment quality, ranged from six, indicating extremely high enrichment, to one, signifying enrichment derived entirely from crustal materials. The use of continental crust values as a reference is a well-established and widely accepted practice in environmental studies. This approach ensures standardisation and enables comparisons across studies conducted in different regions [47].

2.4. Health Risk Assessment (HRA)

To assess the potential health risks associated with PTM exposure in college students, the human exposure model of risk developed by the United States Environmental Protection Agency [52] was applied; university students may encounter dust through several pathways, including direct oral intake, respiratory, and dermal contact [53]. The average daily dose (ADD) for each exposure pathway was determined using the formulas outlined in Equations (5)–(7), as proposed by [52,54].

$${\mathrm{A}\mathrm{D}\mathrm{D}}_{\mathrm{i}\mathrm{n}\mathrm{g}} ={\displaystyle \frac{\mathrm{C}\times {\mathrm{I}\mathrm{R}}_{\mathrm{i}\mathrm{n}\mathrm{g}}\times \mathrm{E}\mathrm{F}\times \mathrm{E}\mathrm{D}}{\mathrm{B}\mathrm{W}\times \mathrm{A}\mathrm{T}}} \times {10}^{-6}$$

(5)

$${\mathrm{A}\mathrm{D}\mathrm{D}}_{\mathrm{i}\mathrm{n}\mathrm{h}} ={\displaystyle \frac{\mathrm{C}\mathrm{X}{\mathrm{I}\mathrm{R}}_{\mathrm{i}\mathrm{n}\mathrm{h}}\times \mathrm{E}\mathrm{F}\times \mathrm{E}\mathrm{D}}{\mathrm{B}\mathrm{W}\times \mathrm{A}\mathrm{T}\times \mathrm{P}\mathrm{E}\mathrm{F}}}$$

(6)

$${\mathrm{A}\mathrm{D}\mathrm{D}}_{\mathrm{d}\mathrm{e}\mathrm{r}\mathrm{m}\mathrm{a}\mathrm{l}} ={\displaystyle \frac{\mathrm{C}\times \mathrm{E}\mathrm{F}\times \mathrm{E}\mathrm{D}\times \mathrm{S}\mathrm{L}\times \mathrm{S}\mathrm{A}\times \mathrm{D}\mathrm{A}\mathrm{F}}{\mathrm{B}\mathrm{W}\times \mathrm{A}\mathrm{T}}} \times {10}^{-6}$$

(7)

In this study, the ADD, expressed in mg kg⁻¹ day⁻¹, was calculated for each exposure route: ingestion (ADD_ing), inhalation (ADD_inh), and dermal contact (ADD_dermal). The exposure frequency (EF) was set to approximately 280 days per year, accounting for vacations during the academic year. The exposure duration (ED) aligned with the typical four-year timeframe for a higher education degree in Egypt. The ingestion rate (IR_ing) for the students was set to 100 mg/d [52]. The average body weight (BW) used in the analysis was 77.3 kg for males and 75.1 kg for females. The average exposure time (AT) was calculated differently for non-carcinogenic and carcinogenic substances. AT is equivalent to 365 days multiplied by the exposure duration (ED) of non-carcinogens. The exposure times for carcinogens were 24,528 days for males (67.2 years × 365 days) and 26,207 days for females (71.8 years × 365 days). The inhalation rates (IR_inh) were recorded as 18.7 m³ per day for male students and 14.6 m³ per day for female students. Additionally, the skin adherence factor (SL) was set at 0.07 mg per cm² per day, following the guidelines from the [55]. The area of exposed skin (SA) was 4250 cm² for male students and 3750 cm² for female students. The dermal absorption factor (DAF) was set at 0.001 for all metals except arsenic (As), which had a higher value of 0.03, as reported in [56].

The particle emission factor (PEF) was set as 1.36 × 10⁹ m³ kg⁻¹ (EPA 2001). For the calculations involving the PTM contaminants in Equations (4)–(6), the concentration, denoted as (C), is expressed in milligrams per kilogram (mg/kg). The values are broadly reflective of the typical age and physical profiles of university students in Egypt, which are generally comparable to those of young adults. Additionally, the physical activity levels and behaviours typical of university students, such as higher exposure to outdoor environments and dust, support the use of adult parameters in this context.

Equations (8) and (9) present the formulas used to assess the non-carcinogenic risk (hazard index; HI) and carcinogenic risk (RT), respectively.

$$\mathrm{HI}=\sum {\mathrm{H}\mathrm{Q}}_{\mathrm{i}}=\sum {\displaystyle \frac{{\mathrm{A}\mathrm{D}\mathrm{D}}_{\mathrm{i}\mathrm{n}\mathrm{g}}}{{\mathrm{R}\mathrm{f}\mathrm{D}}_{\mathrm{i}\mathrm{n}\mathrm{g}}}}$$

(8)

$${\mathrm{R}}_{\mathrm{T}}=\sum {\mathrm{R}}_{\mathrm{i}}=\sum {\mathrm{A}\mathrm{D}\mathrm{D}}_{\mathrm{i}\mathrm{n}\mathrm{g}}\times {\mathrm{S}\mathrm{F}}_{\mathrm{i}\mathrm{n}\mathrm{g}}$$

(9)

The hazard quotient (HQ) represents the HI associated with exposure to a single pollutant, whereas the HI accounts for the combined risk from multiple exposure pathways. HQ or HI values greater than one indicate a potential HI, whereas values below one suggest that the risk is minimal or negligible. R_T was evaluated based on the cancer risk level. An R_T value between 10⁻⁶ and 10⁻⁴ is considered within an adequate range for human health [57]. According to [58], values of R_T exceeding 10⁻⁴ signal a significant cancer risk, whereas values below 10⁻⁶ are indicative of no substantial cancer risk.

Table 1. Non-carcinogenic reference dose and carcinogenic slope factor of PTMs; adapted from [59,60,61].

Element	RfDing (mg/kg/day)	RfDinh (mg/kg/day)	RfDdermal (mg/kg/day)	SF (mg/kg/day)
Fe	7.0	-	-
Mn	0.046	0.00001	0.00184
Ni	0.020	0.02060	0.00054	0.840
Cr	0.003	0.00003	0.00006	42.00
Cu	0.040	0.040	0.012
Zn	0.300	0.300	0.060
Cd	0.000	0.001		0.001
Pb	0.0035	0.003	0.00053
Co	0.020	0.00001	0.01600	9.800

lists the values of the slope factor of carcinogenic pollutants (SF) and the reference dosage of non-carcinogenic pollutants (RfD_ing) in the PTMs.

2.5. Positive Matrix Factorisation (PMF)

PMF is a powerful technique for identifying and quantifying pollutant sources, and it has been effectively applied in environmental studies [62]. This method excels in source apportionment by decomposing a data matrix into a set of factors that represent distinct sources of pollution and their contributions to observed pollutant levels [63], as it operates on the principle of factor analysis. According to [63,64,65], PMF is particularly useful for analysing pollutants in various environmental matrices, including soil, air, and sediments. In this study, PMF 5.0, provided by the U.S. Environmental Protection Agency (EPA), was employed to analyse dust samples collected from 12 distinct locations. The PTM concentration and uncertainty values were used as input files. The uncertainties associated with the PTM concentrations, which encompassed sampling and analytical errors, were calculated as described by [66,67]. The analysis focused on nine PTMs, and the decomposition was performed with a five-factor model. This approach facilitated the identification of key pollution sources and their respective contributions to the metal concentrations in the dust samples.

2.6. Spatial Analysis

Geochemical maps were constructed using Surfer 8 software Golden Software Inc., Golden, CO, USA. Statistical analyses of the PTM concentrations, along with the percentages of organic materials and carbonates, were performed using MINITAB 16.0 statistical software. The dataset was summarised using descriptive statistics, such as the mean, median, minimum, maximum, and standard deviation. Relationships between different PTMs were established using multivariate statistical analyses. The correlation matrix (CM) was used to determine the Pearson correlation coefficients, which helped identify the linear relationships between metal content and soil characteristics. PCA was utilised to reduce the dimensionality of the dataset and understand the potential PTM sources while categorising them as geogenic, anthropogenic, or mixed based on their contributions to variance [68]. Cluster analysis (CA) grouped soil samples into distinct geochemical categories based on similarities in PTM contents, indicating common sources or environmental conditions. Additionally, the PMF decomposed the PTM concentration data into factors representing different pollution sources and quantitatively estimated their contributions to the observed concentrations.

The relationship between the sampling points and campus activities was explored to better understand the influence of these factors on heavy metal contamination. Points located near high-traffic areas, such as parking lots, exhibited higher concentrations of metals typically associated with vehicle emissions, including Pb and Cd. On the other hand, sites located close to agricultural labs or green areas showed elevated levels of metals linked to the use of fertilisers and chemicals, such as Zn and Cu. These observations suggest that campus activities, including vehicular emissions and agricultural practices, significantly contribute to the observed contamination patterns.

3. Results and Discussion

3.1. PTM Concentration and Pollution Assessment

Descriptive statistics for the physicochemical (total organic carbon (TOC), total carbonate, and pH) and PTM contents (Fe, Mn, Co, Ni, Pb, Zn, Cu, Cr, and Cd) in the soil from the Sohag University campus are provided in

Table 2. Descriptive statistics for PTMs in the studied samples.

	Variable	Mean	Md.	StDev	CV%	Min.	Max.	[45]
Physicochemical properties	pH	8.1	8	0.3	4	7.6	8.6
	OC%	1.7	1.7	1.2	71	0.2	3.9
	CaCO3%	5.2	4.7	3.2	68	0.5	11.5
Total Heavy Metal Content	Fe%	1.61	1.75	0.58	33	0.58	2.64	4.09
	Mn ppm	441	486	1.89	39	115	759	904
	Ni ppm	14	13	5	39	7	24	19
	Cr ppm	86	82	43	52	24	150	35
	Co ppm	15	14	4	28	7	24	12
	Cu ppm	29	24	23	96	6	93	14
	Zn ppm	97	100	170	170	69	125	52
	Cd ppm	0.4	0.38	0.63	168	0.32	0.52	0.102
	Pb ppm	39	38	21	55	17	98	17

. The measured pH values ranged between 7.6 and 8.6. All soil samples were alkaline (pH > 7.5), with an average pH of approximately 8. Overall, pH values did not vary significantly across the study area. These alkaline conditions reflected the effects of the carbonate fraction. According to [69,70], PTM mobility is influenced by pH. At higher pH levels, PTMs are adsorbed onto mineral surfaces, whereas they become more mobile in acidic environments. Additionally, the sorption of Cu, Zn, and Cd was enhanced by increasing pH [71]. Alkaline conditions are typically observed in regions with silicate and carbonate parent materials, whereas acidic conditions are more likely to develop in areas receiving wastewater effluents rich in organic matter.

The topsoil of the examined sites displayed a broad range of carbonate contents, varying from 0.5% to 11.5%, with an average of 5.2%. Refs. [72,73] investigated the Sohag region, identified significant variations in total carbonate content, and found that calcium carbonate accumulates closer to the surface in dry and semi-arid areas. Coarse-textured soils exhibited a range of 7.28%–56.4%, whereas medium-textured soils exhibited values between 1.07% and 7.68%. In contrast, fine-textured soils had carbonate contents ranging from 0.00% to 7.05%. Moreover, the topsoil layer samples had organic matter concentrations ranging from 0.2% to 3.9% (mean = 1.7). There can be wide variations in the type and quantity of organic matter in the soil. Ref. [74] demonstrated that organic matter is a key factor influencing the specific adsorption of trace metals in different soils. Additionally, the average PTM concentrations in the urban areas were as follows: Fe > Mn > Zn > Cr > Pb > Cu > Co > Ni > Cd, which can be compared to those in the mean crustal composition (Table 2).

Cr, Pb, Cu, Cd, and Zn were present at high concentrations in the topsoil samples. Elevated concentrations along with higher standard deviation values suggest that these components may have anthropogenic origins. The concentrations of Co and Ni were low, resembling the levels reported for Wedepohl’s mean crustal composition. The coefficient of variation (CV) of an elemental concentration can reveal both the variation and interference of human activity; a higher CV value denotes a greater anthropogenic influence [75]. In particular, little variation was indicated by a CV ≤ 15%, moderate variation by a CV < 35%, and significant variance by a CV > 35%. The CV values of the PTMs decreased in the following order: Zn > Cd > Cu > Cr > Pb > Mn, Ni, and Fe > Co (Table 2). Significant human activity was indicated by the high variability (CV > 35%) of Zn, Cu, and Pb content in the soil from the university campus in Sohag [75].

3.2. PTM Spatial Distribution

The spatial patterns of soil pH, TOC, and CaCO₃ contents across the study site exhibited distinct trends (Figure 2). TOC levels were highly variable, averaging 1.7%, with pockets of elevated values (>2%) in the southern and northwestern sections, characterised by more productive agricultural soil. The soil pH values were neutral to slightly alkaline, ranging from 7 to 8.4, with the lowest pH levels concentrated in the south. In addition, the CaCO₃ content was more heterogeneously distributed, indicating an increasing trend toward the central and southeastern sections.

The PTM spatial distribution within the study area was highly heterogeneous. Figure 3 demonstrates the spatial distribution of PTMs in the soil, characterised by localised ‘hotspots’ in the northern and central sections, where a high metal concentration was bordered by lower levels. In contrast, high concentrations of Cr, Cd, and Fe were detected in the southern and northern regions. This spatial variability underscores the importance of comprehensive site characterisation to fully understand the PTM contamination patterns and potential exposure pathways within the region.

3.3. PTM Ecological Risks

The mean values of Fe, Mn, and Ni were low, ranging from 0.1 to 0.6, 0.1 to 0.8, and 0.4 to 1.3, respectively (Figure 4). Excluding samples 2 and 4, which were at the mid-level for Ni, all samples had low PIs, with mean PIs for Fe, Mn, and Ni of 0.4, 0.5, and 0.7, respectively. This suggests that the concentration of the former metals in the soil samples was comparable to that in the average crust, and that there was no evident pollution by these metals in the soil samples under investigation.

The average PI values for Cr, Pb, Co, Zn, and Cu were 2.5, 2.3, 2.1, 1.9, and 1.2, respectively, indicating moderate contamination. The PIs of Cd were higher, as all samples contained high PIs, except sample 2, which had a moderate value. The PIs ranged from 1.04 to 8.43 for Pb, whereas the PI for Cu varied between 3.1 and 5.1. According to this study, the urban green spaces at the university are heavily contaminated with Cd. The university campus is situated in a densely populated urban area, increasing the potential for Cd deposition from various sources. The proximity of these parks to busy roads, parking lots, and ageing infrastructure can lead to the accumulation of Cd-containing particles from vehicle emissions, brake and tyre wear, and the weathering of older buildings and facilities. Furthermore, the high volumes of foot traffic and recreational activities common in university parks can disturb the soil, potentially exposing students, faculty, and staff to contaminated materials.

Similarly to PIs, the PLI can provide a more comprehensive assessment of the pollution status. The PLI values across the different sites ranged from 0.7 to 2.1 (Figure 4). High pollution load values (Section 2.1) were found at Site 4, indicating a high pollution potential. The remaining sites had relatively modest pollution load values, between 0.7 and 1.6, suggesting that these samples were slightly polluted. A high PLI in university campus soils can negatively affect the growth and development of plants, as well as the overall health and diversity of the soil biota. Furthermore, exposure to contaminated soil, either through direct contact or indirect pathways, can lead to serious health risks for students, faculty, staff, and particularly vulnerable populations such as children. A high PLI can also damage the reputation of an institution and potentially result in legal concerns if the pollution levels exceed regulatory thresholds. Soil contamination was confirmed using I_geo. Based on the results, the mean I_geo of the PTMs in the soil declined in the following order: Cd (1.4) > Pb, Cr (0.5) > Zn (0.3) > Cu (0.1) > Co (−0.4) > Ni (−1.1) > Mn (−1.8) > Fe (−2.1). The assessed PTMs reflect varying levels of pollution. Based on the pollution degree standard of I_geo, mean values <0 occurred for Fe, Mn, and Ni in all soil samples and for Cr, Co, Cu, Zn, and PB in 25%, 83%, 42%, 8%, and in 33% of the samples, respectively, indicating that these PTMs were not regarded as soil pollutants. Moreover, the mean I_geo values of Zn, Pb, Cu, Cr, and Co in 92%, 50%, 42%, 42%, and 16% of soil samples, respectively, had values within 0–1, suggesting uncontaminated-to-moderately contaminated pollution levels. The mean I_geo values for Cr, Cu, Cd, and Pb were between 1 and 2 in 33%, 8%, 100%, and 16% of the samples, respectively, indicating moderate pollution levels. Only 8% of the samples showed mean I_geo values for Cu between 2 and 3, indicating high pollution levels. The EF values across samples ranged from 3.0 to 12.2 (Figure 4).

Approximately 42% of the samples had EF values less than 5, indicating moderate enrichment, whereas those remaining fell in the significant enrichment class (58%). For Cd, Pb, Cr, Zn, Cu, and Co, the corresponding EF values above 5 were 83%, 67%, 58%, 50%, 33%, and 25%, respectively. Excluding Co, which has an EF class of 3, these metals have an EF class of 4, with a notable enhancement in PTMs. Cr displayed the greatest EF (EF = 23) for sample 8, indicating extremely high enrichment. The remaining Cr enrichment was moderate-to-significant. In contrast, 83% of the Cd exhibited considerable enrichment, whereas 17% displayed very high enrichment. To further validate the reliability of the EF calculations, a sensitivity analysis was conducted by comparing results from the global crustal averages with baseline values from similar geochemical settings. The comparative analysis revealed consistent enrichment trends, particularly for Cd, Pb, and Cr, supporting the use of crustal averages as an interim reference.

Despite some limitations, the observed EF patterns, especially for Cd, Cr, and Pb, indicate significant anthropogenic contributions, aligning with earlier studies conducted in urban and campus settings [53,76]. These findings indicate the need for ongoing monitoring and the eventual development of local geochemical baselines to enhance site-specific pollution assessments.

3.4. Health Risks from PTMs

HQ and HI were used to evaluate the health risks from PTMs in dust collected from college campuses for both male and female students, as shown in

Table 3. Health risk evaluation of health risks from PTMs in the dust samples for students.

	PTMs	HQing	HQinh	Hqdermal	HI	RT
Male Students	Fe	2.43 × 10−7	-	-	2.43 × 10−7	-
	Mn	9.37 × 10−3	2.03 × 101	1.33 × 102	1.54 × 102	-
	Cu	7.38 × 10−4	1.00 × 10−7	5.99 × 10−1	5.99 × 10−1	-
	Zn	3.16 × 10−4	4.54 × 10−8	3.21 × 10−1	3.21 × 10−1	-
	Pb	1.05 × 10−2	1.40 × 10−6	1.03 × 101	1.03 × 101	-
	Co—Non-Cancer	7.60 × 10−4	3.66 × 10−4	1.28 × 10−1	1.29 × 10−1	-
	Co—Cancer	1.26 × 10−2	6.08 × 10−3	1.28 × 10−1	-	2.48 × 10−3
	Ni—Non-Cancer	6.58 × 10−4	7.98 × 10−1	9.87 × 100	1.07 × 101	-
	Ni—Cancer	1.19 × 10−2	-	1.07 × 101	1.07 × 101	1.99 × 10−4
	Cr—Non-Cancer	2.96 × 10−2	4.26 × 10−4	4.50 × 102	4.50 × 102	-
	Cr—Cancer	4.93 × 10−1	7.12 × 10−3	4.50 × 102	-	6.21 × 10−2
	Cd—Non-Cancer	3.96 × 10−3	5.54 × 10−8	-	3.96 × 10−3	-
	Cd—Cancer	6.60 × 10−2	-	-	-	0
Female Students		HQing	HQinh	Hqdermal	Hi	RT
	Fe	2.49 × 10−7	-	-	2.49 × 10−7	-
	Mn	9.51 × 10−3	2.03 × 101	1.33 × 102	1.54 × 102	-
	Cu	7.60 × 10−4	8.15 × 10−8	5.99 × 10−1	5.99 × 10−1	-
	Zn	3.26 × 10−4	3.62 × 10−8	3.21 × 10−1	3.21 × 10−1	-
	Pb	1.08 × 10−2	1.12 × 10−6	1.03 × 101	1.03 × 101	-
	Co—Non-Cancer	7.82 × 10−4	2.94 × 10−4	1.28 × 10−1	1.29 × 10−1	-
	Co—Cancer	1.22 × 10−2	4.59 × 10−3	-	-	2.39 × 10−3
	Ni—Non-Cancer	6.77 × 10−4	7.98 × 10−1	9.87 × 100	1.07 × 101	-
	Ni—Cancer	1.14 × 10−2	-	2.31 × 10−⁵	1.14 × 10−2	1.92 × 10−4
	Cr—Non-Cancer	3.04 × 10−2	3.43 × 10−4	4.50 × 102	4.50 × 102	-
	Cr—Cancer	4.75 × 10−1	5.35 × 10−3	2.55 × 10−3	-	5.98 × 10−2
	Cd—Non-Cancer	4.08 × 10−3	4.38 × 10−8	-	4.08 × 10−3	-
	Cd—Cancer	6.36 × 10−2	-	-	-	0

. Three exposure routes were considered, namely ingestion, inhalation, and dermal contact. These results suggest that dust ingestion is the main exposure route for most PTMs, which is consistent with the findings of other studies. For both sexes, the non-carcinogenic risks of PTMs such as Zn, Ni, Cr, Cu, and Pb followed the order HQ_ing > HQ_dermal > HQ_inh, indicating that ingestion was the most significant exposure pathway. In contrast, for Mn and Co, the risks followed the order of HQ_ing > HQ_inh > HQ_dermal. Notably, Pb exhibited the highest value for HQ_ing, Mn showed the highest HQ_inh, value, and Cr showed the highest HQ_dermal for both sexes, although none of the HQ values exceeded 1. This suggests that the non-carcinogenic risks posed by these metals are within acceptable levels. For both male and female students, the HI values, which evaluated the cumulative risk of exposure to multiple PTMs, were ranked as follows: Cr > Mn > Pb > Zn > Cu > Ni > Co. All HI values remained below 1, confirming that HI values were minimal for the student population. Female students exhibited slightly higher HI values than male students.

Regarding R_T, the exposure to Ni, Cr, and Co was evaluated based on ingestion, inhalation, and dermal contact. The R_T values for these metals ranged between 10⁻⁴ and 10⁻⁶, remaining within acceptable limits for human health. Although Ni and Cr contributed more significantly to the overall R_T, the total risk for both male and female students remained below the threshold of concern. The risk ranking for carcinogenic elements for both sexes was Cr > Co > Ni. The cumulative HI values for all metals were below 1, indicating no significant HI from dust exposure. However, it remains essential to monitor carcinogenic metals such as Cr and Ni owing to their potential long-term health impacts. Regular dust sampling and analysis are critical for ensuring exposure levels remain within safe limits. This analysis highlights the importance of ongoing environmental monitoring and mitigation measures to reduce PTM accumulation in campus dust and safeguard student health.

3.5. PTM Source Apportionment

The PMF analysis identified five distinct sources contributing to PTM concentrations in the collected dust samples (Figure 5, Figure 6 and Figure 7). These sources were determined based on the concentrations and contributions of the different PTMs to each factor [77,78,79]. Factor 1 exhibits high concentrations of Cu, Ni, and Cr, with Cu accounting for the largest proportion. This suggests the strong influence of industrial activities and anthropogenic sources [80,81,82,83], including the wear and tear of metallic objects, paintings, and automotives. The presence of Ni and Cr, which are typically associated with industrial pollution [84,85], further supports this hypothesis. The presence of Cu in Factor 1 indicates both industrial emissions and vehicular activity, particularly in areas close to workshops and automotive maintenance facilities [86,87,88,89,90]. Factor 2 is dominated by high levels of Pb and Zn. Pb has long been associated with traffic pollution [91], and although the use of Pb gasoline has been eliminated, residues from historical emissions persist in the environment [92]. Zn, which is widely used in tyre vulcanisation, is attributed to traffic-related sources [93]. This factor reflects the cumulative effects of vehicular wear, exhaust emissions, and road dust resuspension [94,95].

The third factor was described by the elevated loadings of Fe, Mn, Co, and Ni, which are typically geogenic in origin [96,97,98]. Fe, a lithogenic element, exhibited strong correlations with Mn, Co, and Ni, all of which are naturally abundant in soil derived from bedrock. This factor indicates the dominance of natural processes, including soil erosion and rock weathering, that contribute to dust accumulation on university campuses [99]. Factor 4 showed significant contributions from Cd and Cr, which are often linked to agricultural practices, particularly the use of pesticides and fertilisers. Cd is present in phosphate fertilisers [100,101,102], which can lead to soil contamination and subsequently enter the dust [103,104,105]. This factor reflects the anthropogenic impacts of farming practices and the associated application of agrochemicals [106,107,108].

Factor 5 represents a mixed source with contributions from both industrial releases and vehicle-related pollutants. Metals such as Cu, Zn, and Pb contributed to this factor, aligning with the industrial and vehicular activities observed around campuses [109,110,111,112]. Moderate concentrations of these metals suggest overlapping pollution sources from industry and traffic.

To validate the PMF analysis and reduce reliance on speculative interpretations, the proportional contributions of each factor were compared with spatial distribution patterns observed in the study area (Figure 3). For example, higher concentrations of Pb and Zn in areas near roads and parking lots support the classification of Factor 2 as traffic-related sources. Similarly, elevated levels of Cd and Cr in green spaces near agricultural zones confirm the identification of Factor 4 as being linked to agrochemical inputs. Additionally, the correlation matrix (Figure 8) provided further evidence supporting the PMF results. Strong correlations between Fe, Mn, and Co (r = 0.72 and r = 0.67, p < 0.01) reinforced their assignment to geogenic processes in Factor 3. Likewise, moderate correlations between Pb, Zn, and Cd (r = 0.51–0.65) highlighted potential overlapping sources from traffic and agricultural practices. Hierarchical clustering analysis (Figure 9) also identified groups of metals that share similar environmental behaviours and sources, offering complementary insights into the PMF results.

Figure 7 illustrates the proportional contributions of the five factors to the overall distribution based on their normalised contributions. Factor 5 was the most influential, accounting for 30% of the total and exerting the greatest impact. Following this, Factor 2 contributed 25%, marking it as a significant, although secondary, element of the distribution. Factor 1 had a moderate influence, contributing to 20% of the overall distribution. Factors 3 and 4 contributed 15% and 10%, respectively, with Factor 4 having the smallest impact. Despite the strength of the PMF results, this study acknowledges its limitations. The lack of isotopic or source-specific markers limits the precision of source apportionment [113]. Future research could benefit from integrating isotopic analyses, high-resolution chemical profiling, or direct source monitoring to strengthen the identification and quantification of pollution sources.

The Pearson correlation matrix (Figure 8) provides valuable insights into the relationships between the PTMs in the topsoil of the green areas of Sohag University by quantifying the strength and significance of their linear correlations. Strong positive correlations were observed between Fe and Ni (r = 0.69, p < 0.01), and between Fe and Mn (r = 0.72, p < 0.01). These correlations suggest that these PTMs may have originated from similar sources, such as industrial or vehicular emissions. This alignment was consistent with the PMF factors that identified these metals as originating predominantly from human activity.

In addition, Fe and Co displayed a notable correlation (r = 0.67, p < 0.05), reinforcing their potential common sources in industrial and geogenic processes. The substantial correlation between Ni and Pb (r = 0.51) and the moderate relationship with Zn (r = 0.48) indicated a possible overlap in sources, likely tied to both vehicular and industrial emissions. These metals are often present in environments that are affected by traffic and manufacturing processes. Cr was closely correlated with Mn (r = 0.65, p < 0.05) and moderately correlated with Co (r = 0.63, p < 0.05), suggesting that these PTMs may be associated with similar industrial sources, reflecting shared emission patterns.

Additionally, the matrix shows that Zn, although primarily linked to traffic through tyre wear, is correlated with Pb and Cd, which could be indicative of mixed sources, including vehicular emissions and inputs from industrial activities or agricultural practices. The relatively weaker correlations involving Cd, with slight negative correlations with Cu and Ni, suggest divergent sources or behaviours in the environment, likely indicating specific anthropogenic sources, such as agriculture, as noted in its linkage with phosphate fertilisers in the PMF findings. This comprehensive overview provided by the Pearson correlation matrix confirms and expands upon the source apportionment insights from the PMF, illustrating a complex web of relationships and source influences among the studied PTMs.

Figure 9 illustrates the hierarchical clustering of the nine PTMs, which is based on similarities in their concentrations across the samples, with metals that have higher similarity or co-occurrence forming closer clusters [114]. The Fe, Mn, and Co clusters suggest that these metals likely share a common geogenic or industrial source, as supported by their similar environmental behaviours and strong correlations. Another prominent cluster included Ni and Cu, which were grouped closely because of their strong correlation, likely reflecting common industrial sources, including metal processing and vehicular emissions.

Conversely, the grouping of Pb, Cr, Zn, and Cu suggested potential source overlap from a combination of industrial and traffic emissions, with Pb and Zn commonly associated with vehicular sources. Lastly, Cd exhibited separate branching in the dendrogram, indicating a distinct source likely related to agricultural activities, particularly its presence in phosphate fertilisers, as detailed in the PMF analysis. The horizontal axis in the dendrogram represents the degree of similarity between the metals, with shorter linkage lengths indicating a higher degree of similarity. This clustering pattern highlights the possible shared pollution sources or environmental behaviours of certain metals. These correlation and clustering analyses provided a comprehensive understanding of the environmental dynamics affecting PTM pollution. They validated and complemented the source apportionment findings from the PMF analysis, emphasising the significance of both geogenic and anthropogenic sources in influencing the PTM distribution in the study region. This integrated approach is critical for devising effective environmental monitoring and management strategies.

3.6. Comparison with Previous Similar Works

This study builds upon and contributes to a growing body of research investigating PTM pollution in urban soils, including university campuses and similar environments. Table S1 provides a summary of comparable studies conducted globally, highlighting methodologies, findings, and sources of PTM pollution in campus or urban soils. These studies illustrate the broader relevance of the findings at Sohag University and emphasise the significance of adopting transferable methodologies, such as pollution indices and PMF, to identify pollution sources and assess environmental risks.

These studies highlight consistent findings across different geographical and environmental contexts. For instance, Pb and Zn are frequently identified as major pollutants in urban environments due to traffic-related emissions, as observed in studies conducted in Xi’an, China [53], and Vilnius, Lithuania [115]. Similarly, elevated Cd concentrations linked to anthropogenic activities were reported in Madrid, Spain [116], as well as in this study. The methodologies employed across these studies, such as PMF, pollution indices (e.g., PI, PLI, Igeo, and EF), and HRA, align with those used in this study. These methodologies have proven effective in assessing PTM contamination and apportioning sources, reinforcing their value for environmental monitoring and risk evaluation.

By comparing findings, the current study emphasises the broader relevance of assessing PTM pollution in university and urban soils. The results from Sohag University provide a valuable reference for implementing similar environmental monitoring programmes in other academic and urban settings. To contextualise our findings, we compared our results with those of previous studies conducted in urban environments and university campuses worldwide (Table S1) [9,53,117,118,119,120,121,122,123].

3.7. Implications

Based on the findings from this study, we recommend that universities and other educational institutions implement targeted pollution control strategies, particularly in areas with high traffic flow and industrial activity. Pollution prevention measures, such as the use of low-emission vehicles, regular maintenance of infrastructure, and the incorporation of pollution-reducing vegetation, should be prioritised. Furthermore, the results underline the importance of incorporating sustainable green space planning into campus development, ensuring that green areas are not only aesthetically pleasing but also serve as protective barriers against pollution and contribute to improved environmental quality. Ongoing monitoring of heavy metal concentrations in campus soils and green areas will be essential to track pollution levels and implement necessary mitigation actions. By taking these steps, universities can reduce environmental risks and enhance the safety and well-being of their students, staff, and visitors.

3.8. Limitations

While the methodology employed in this study provided valuable insights into heavy metal contamination and its associated risks, there are several limitations to consider. Relying on global average background values for comparison may not accurately reflect local geochemical conditions, which could lead to inaccuracies in interpreting pollution indicators like the PI. The absence of site-specific background data limits the precision of these assessments. Second, this study’s temporal coverage was limited to a single-season sampling period, which may not capture the full variability in heavy metal concentrations that could occur across different seasons. Seasonal fluctuations in environmental conditions, such as precipitation or temperature, may influence the mobility and distribution of metals. Finally, this study did not account for some site-specific anthropogenic or natural factors, such as localised pollution sources (e.g., nearby industries or construction activities) or the potential effects of historical land use. These unaccounted factors may have contributed to the variability in the data and should be considered in future studies that aim to provide a more comprehensive assessment of contamination sources.

4. Conclusions

The soil from university campuses in Sohag, Egypt, exhibited elevated concentrations of PTMs including Cd, Co, Cr, Cu, Fe, Mn, Ni, Pb, and Zn, compared to natural background values. Among these, Zn, Cd, Cr, and Pb were identified as the most concerning pollutants, with Pb and Cd posing significant ecological and environmental risks due to their toxicity and persistence. The use of pollution indices, such as Igeo and EF, highlighted the contribution of both natural and anthropogenic sources, with Pb, Cd, and Cr identified as key contributors to soil contamination. PCA and CA results further indicated that certain metals, such as Fe, Mn, and Co, were primarily linked to geogenic processes, while metals like Pb, Zn, and Cr were largely associated with anthropogenic activities, particularly traffic emissions and agricultural practices. The PMF analysis identified some distinct sources contributing to the observed contamination, including industrial activities, traffic emissions, natural soil processes, agricultural practices, and mixed sources from both industrial and traffic emissions. These findings highlight the importance of managing anthropogenic pollution sources, especially industrial emissions and vehicle activities, to reduce the ecological risks associated with these contaminants.

Despite the contamination observed, the non-carcinogenic risks for both male and female students were within acceptable limits, as defined by the USEPA thresholds. However, this study emphasises the need for continuous monitoring of carcinogenic metals like Pb and Cd, which present long-term health risks. This study highlights the broader applicability of these findings for other university campuses and urban environments, demonstrating the value of using spatial analysis, pollution indices, and statistical methods to assess and manage heavy metal pollution.

Future research should focus on several key directions to conduct multi-seasonal or long-term monitoring to assess temporal variability in contamination levels and to establish local or regional geochemical background values to provide more context-specific assessments. Expanding comparative studies across multiple urban campuses or green spaces can help generalise findings and inform broader environmental management strategies. These future studies will refine our understanding of pollution dynamics and lead to more targeted mitigation measures.