Rapid advancement of modern cities has led to the continuous discharge of trace toxic elements (TTEs) into the environment, causing urban environmental pollution1,2,3,4. TTEs contamination levels in urban environments can be measured by TTEs in road dust5. Persistent, toxic, and refractory TTEs accumulate on surface resuspended dust (SRD) easier than larger dust particles6,7,8,9. SRD is composed of tiny particles (particle size < 100 μm) with a considerable specific surface area and light weight7. Owing to meteorological changes and human activities, SRD can spread further and stay in the air longer, thereby posing serious environmental risks7. TTEs in SRD can enter the human body through various exposure pathways, causing permanent damage. Therefore, exploring the concentration levels and risks of TTEs in SRD is of great significance for monitoring urban TTEs pollution and safeguarding the safety of ecological environment and human health.

In order to clarify the degree of impact of TTEs in dust on the environment and human health, researchers used different approaches to quantify the degree of ecological health risks and contamination level of TTEs10,11,12,13,14,15,16. However, fixed parameters and concentrations used to assess risks may lead to uncertainty or even errors in the assessment results17. In this study, we aim to introduce Monte Carlo simulation (MCS) to reduce the uncertainty of risk assessment and improve the accuracy and reliability of assessment results18. In addition, content-oriented risk assessment cannot identify pollution sources that pose greater harm to ecology and health. Considering these challenges, in this study, we aimed to combine ecological and health risks simulated by MCS with positive matrix factorization (PMF)19, identify and quantify the leading source of anthropogenic contamination in risk assessment, and determine the priority control TTEs to establish a scientific basis for the precise management and prevention of TTEs in dust.

From the literature, it can be seen that the content, pollution level, and harm caused by urban surface dust TTEs vary across cities and functional areas owing to the repercussions of people’s behavior20,21. Youth are more susceptible to TTEs and are more affected by TTEs in dust than adults2. The university campus is the main places for college students to study and live. Due to outdoor activities such as physical education classes and sports events, students have many opportunities to be exposed to TTEs in campus surface dust. In addition, dust adhering to clothes, shoes, and socks is brought into the room, further increasing the exposure time22. Existing research has found that TTEs exhibited non-negligible pollution in university campus dust2. However, there is a lack of detailed research on the degree of contamination and health hazards of TTEs in SRD.

Xi’an is the capital of Shaanxi Province and an important political, cultural, and economic center in the northwest China. At present, Xi’an has 63 general higher education institutions with nearly 1.5 million college students, making it one of the cities in China with the highest number of educated people and the highest density of universities23. It is an ideal site for studying TTEs pollution in university campus SRD. The main purposes of this study were to (1) investigate the level of TTEs content in SRD of university campuses in Xi’an; (2) evaluate the probabilistic pollution and ecological health risks of TTEs in the SRD based on MCS; (3) judge the sources of TTEs in the SRD using PMF; and (4) determine the priority control sources and pollutants on basis of source-oriented probabilistic ecological health risks of TTEs. Our research findings are expected to provide scientific basis for improving the environmental quality and risk management of university campuses.

Materials and methods

Study area

Xi’an is located in the central part of the Guanzhong Basin in the Yellow River watershed of China. It borders the Weihe River to the north, Qinling Mountains to the south, Linghe River and the Bayuan upland to the east, and Taibai Mountains and the Qinghua loess tableland to the west. The built-up area of Xi’an City is 700.69 km2, with a permanent population of 10.2035 million and an urbanization rate of 74.61%23. The climate of Xi’an is warm-temperate semi-humid continental monsoon with four distinct seasons, and northeast winds are prevalent throughout the year. The mean yearly temperature ranges from 13.1 to 14.3 °C, and the yearly rainfall in urban areas is 648.4 mm. By 2020, Xi’an’s gross domestic product has exceeded 1 trillion yuan, with over 3.8 million private cars and over 9000 buses. The length of urban roads is 5283.95 km, with a road area of 123,452,100 m2, and the length of subway operating lines is 259 km23. The 35 university campuses surveyed in this study are located south of 1st subway, mainly concentrated near the South 2nd Ring Road (Figure S1), with nearly 10,000 students per campus24. This area is also the most prosperous and developed commercial and tourism area in Xi’an City. The traffic is busy and there is a large flow of people.

Sampling and analytical methods

In December 2018, dust samples were collected from 35 campuses of 25 universities in Xi’an under a week of continuous sunny weather conditions (Figure S1). Detailed sampling and sample pretreatment methods have been previously described elsewhere2,24. Take 100 g from each campus dust sample and sieve through a 100-µm sieve to obtain the SRD samples6,25. Using a vibrating grinder, all SRD samples were further grinded to a particle size of less than 75 μm. A total of 4.5 g of the ground dust sample was weighed and pressed into a disc using a semi-automatic tablet press2. An X-ray fluorescence spectrometer (S8 Tiger, Bruker, Germany) was used to determine the amount of TTEs (Mn: manganese, Zn: zinc, Cr: chromium, As: arsenic, Co: cobalt, Ba: barium, Cu: copper, Pb: lead, V: vanadium, and Ni: nickel) present in the SRD samples. In the experimental analysis, quality control was carried out using standard and duplicate samples to achieve an error of less than 5% for the analyzed elements2.

Probabilistic contamination assessment

Enrichment factor (EF) measures the enrichment or reduction of TTEs relative to a specific source, and its environmental footprint analysis helps distinguish between human-generated and natural sources, and assess the degree of impact of anthropogenic activities on TTEs26,27. This method is frequently used to assess the extent to which TTEs contamination is found in dust, soil, and sediments28,29. Although the EF can accurately determine the anthropogenic pollution of individual elements, it is cannot comprehensively portray the research area’s overall TTEs pollution level. Due to the presence of multiple elements in a region, to better assess the comprehensive pollution level of TTEs in SRD of Xi’an college campuses, this study introduces the Nemerow integrated enrichment factor (NIEF)30, which is an evaluation index based on the Nemerow integrated pollution index31. EF and NIEFare calculated using the following Eqs. (1) and (2)11,12,30,32,33

$$\:\begin{array}{c}EF=\frac{{\left(\frac{{C}_{i}}{{C}_{ref}}\right)}_{sample}}{{\left(\frac{{C}_{i}}{{C}_{ref}}\right)}_{background}}\end{array}$$
(1)
$$\:\begin{array}{c}NIEF=\sqrt{\frac{{{(EF}_{max})}^{2}+{\left({EF}_{ave}\right)}^{2}}{2}}\end{array}$$
(2)

where Ci and Cref are the concentrations of TTE i and the reference element, respectively. EFmax and EFavg represent the maximal and mean amounts of the enrichment factor, respectively.

In the calculation, the background values of elements in Shaanxi soil published by the China National Environmental Monitoring Center34 was used, and Al was used as the reference element due to its stability and minimal variation in the SRD (the average content was 10.2% and the coefficient of variation was 4.7%). The content of Al in the SRD was measured using an X-ray fluorescence spectrometer (S8 Tiger, Bruker, Germany). Table S1 lists the classifications of EF and NIEF. In this study, we combined the EF and NIEF with MCS to evaluate the probability contamination level of both single and combined pollutions of TTEs in the SRD. Table S2 displays the probability density functions (PDFs) of the concentration of TTEs in the SRD.

PMF

PMF was performed to investigate the sources of TTEs in the SRD. The fundamental formula is as Eq. (3)2311,28,3536373839:

$$\:{X}_{ik}={\sum\:}_{j=1}^{p}{G}_{ij}\times\:{F}_{jk}+{E}_{ik}$$
(3)

where Xik represents the quantity of TTE j in an SRD sample k, Gij is the source j’s contribution to the SRD sample i, Fjk represents the quantity of TTE j from source k, Eik is the residual matrix, and p represents the quantity of potential sources. The detail method of factor selection in the PMF is provided in the supplementary materials (Text S1).

Probabilistic ecological risk assessment

In this study, an extension modified ecological risk index (emRI) was used to evaluate the ecological potential hazards of TTEs in SRD from Xi’an college campuses. The calculation formula is as Eq. (6)30:

$$\:emRI=\frac{{\sum\:}_{i=1}^{n}{T}_{i}\times\:{EF}_{i}}{{\sum\:}_{i=1}^{n}{T}_{i}}$$
(4)

where Ti is the toxicity factor for TTE i2, and EFi is the enrichment factor of TTE i. The classification criteria for emRI are listed in Table S1. We combined emRI with MCS to evaluate the probabilistic ecological risk of TTEs in the SRD.

Probabilistic health risk evaluation

The health hazards of TTEs in SRD from Xi’an college campuses to students were evaluated using the health risk model40,41,42,43. According to the health risk model, carcinogenic and non-carcinogenic risks can be assessed through the TCR and HI, respectively44,45, as follows:

$$\:TCR=\sum\:CR=\sum\:ADD\times\:SF$$
(5)
$$\:HI=\sum\:HQ=\sum\:\frac{ADD}{RfD}$$
(6)

where TCR represents the total cancer risk, SF is the slope factor, HI is the total non-carcinogenic risk, ADD is the exposure dose, and RfD is the corresponding reference dose. The detail calculation methods of ADD are provided in the supplementary materials (Text S2). Table S3 lists the values of SF and RfD for different exposure pathways of the TTEs. We combined TCR and HI with MCS to assess the probabilistic health risks of TTEs to college students exposed to the SRD in Xi’an college campuses.

Software

Plots were created using Origin 2022 (OriginLab, USA) software. Statistical analyses were conducted in SPSS 26 (IBM, USA). PMF was conducted using the US EPA PMF V5.0 software, and the MCS using the Crystal Ball V11.1.24 software (Oracle, USA). In MCS, by defining assumptions and predictions, 10,000 simulations were performed for the levels of pollution, ecological risks, and health risks of TTEs to obtain stable results.

Results and discussion

TTE concentration

Table 1lists the descriptive statistics of TTE concentration in SRD samples from Xi’an university campuses, as well as Shaanxi’s soil background values and TTE content in university campus bulk dust2and resuspended road dust25. The mean concentrations of all TTEs analyzed in the SRD of Xi’an college campuses were higher than the corresponding soil background values34, especially Cr, Ba, Cu, Pb, and Zn. CV represents the coefficient of variation, which can be used to reflect how human activities affect TTEs in the environment. The higher the CV value, the stronger the impact of human factors on TTEs46. Based on CV categorization result2, the values of Zn, Pb, Cu, and Ba were > 35%, indicating high variability; those of Cr, Co, and As were between 15% and 35%, indicating medium variability; and those of Ni, Mn, and V were < 15%, indicating low variability. Among the investigated TTEs, Zn, Pb, Cu, and Ba were significantly influenced by anthropogenic factors; Cr, As, and Co were influenced to some extent by anthropogenic factors; and human activities had limited impact on Ni, Mn, and V. The kurtosis values were greater than zero for all the TTEs, except for V, Cr, and As, indicating that the peaks of these TTEs were steeper than those of the normal distribution2. The skewness values of Cu, Zn, Pb, and Ba > 1 indicate a positive bias towards lower concentrations2, which is supported by the higher mean concentration compared to their median counterparts.

Compared with the resuspended road dust in Xi’an City25, the SRD of Xi’an university campuses had higher contents of Cu, Mn, Ni, Pb, Zn, and V, especially the anthropogenic elements Pb, Zn, and Cu, indicating that the environmental situation of Xi’an universities is not optimistic and worrying. Compared with the bulk dust samples of Xi’an university campuses2, the content of TTEs in SRD of Xi’an university campuses was similar that in the bulk dust, mainly because the SRD of Xi’an university campuses accounted for 75–95% of bulk dust, and the amount of TTEs in SRD determines the amount of TTEs in bulk dust.

Table 1 Descriptive statistics of TTE content (mg kg−1) in the SRD of Xi’an college campuses, and the background value of Shaanxi soil and other studies in Xi’an.
Full size table

Contamination level of TTEs

The contamination assessment results of TTEs in the SRD of Xi’an university campuses are displayed in Fig. 1. The simulated average EF values for all TTEs were Zn (12.9) > Pb (9.5) > Cu (6.3) > Cr (3.1) > Ba (2.1) > As (1.7) = Ni (1.7) = Co (1.7) > V (1.6) > Mn (1.5). The simulated EF values of all TTEs obtained from MCS were slightly higher than the corresponding EF values calculated from the samples’ measured contents, indicating that the conventional assessment results calculated directly from the elements’ content in the samples may underestimate the contamination level. The EF values of V, Co, As, Ni, and Mn in 62.7%, 62.7%, 64.5%, 67.1%, and 71.3% of the samples, respectively, were between 1 and 2 (Fig. 1), indicating minimal enrichment. Cr exhibited moderate enrichment and above in 84% of the SRD samples and minimal enrichment in 16% of the SRD samples. Ba exhibited low enrichment in 49.9% of the SRD samples and moderate to extremely high enrichment in 48.6% of the SRD samples. Regarding Cu, the mean EF (6.3) and 57.6% of the EF were between 5 and 20, indicating significant enrichment, while 39.6% of the EF was between 2 and 5, indicating medium enrichment. Regarding Zn, 74.1% of the EF and the mean EF (12.9) ranged from 5 to 20, indicating significant enrichment, while 10.5%, 13.1%, and 2.1% of the EF ranged from 2 to 5, 20–40, and > 40, respectively, indicating moderate to extremely high enrichment. Regarding Pb, the mean EF (9.5) and 83.1% of the EF values ranged from 5 to 20, presenting considerable enrichment, while 13.2% and 3.6% of the EF values were between 2 and 5 and 20 and 40, respectively, indicating moderate and very high enrichment.

Fig. 1
figure 1

EF and NIEF results of TTEs based on MCS.

Full size image

In summary, the SRD was highly contaminated with Pb, Cu, and Zn, indicating that anthropogenic factors mainly influenced them. Moreover, Cr and Ba were moderately enriched, demonstrating that human activities have a significant impact on them. In contrast, Co, As, Ni, V, and Mn were primarily impacted by natural factors.

The MCS results of NIEF showed that 5.0, 49.3, and 45.7% of the NIEF values were between 3 and 5, between 5 and 10, and > 10 (Fig. 1k), indicating significant, very high, and extremely high enrichment, respectively. The simulated mean NIEF value (11.1) was larger than the detected mean NIEF (9.1), indicating that the conventional assessment result calculated directly using the elements’ content in the samples may underestimate the comprehensive contamination level. Sensitivity analysis revealed that Zn and Pb were the two most sensitive TTEs to NIEF, with contribution rates of 79.7% and 18.9%, respectively (Figure S2).

TTE source analysis

The PMF analysis results of TTEs determined in the SRD of Xi’an university campuses are shown in Fig. 2 and Figure S3. Factor 1 was mostly loaded with Pb and Cr, contributing 76.9% and 40.5% to Pb and Cr, respectively (Fig. 2). The above content analysis results demonstrated that Pb and Cr in the SRD of Xi’an university campuses were mostly influenced by human factors. Correlation analysis results further confirmed these results. Pb and Cr were positively correlated, and both were negatively correlated with Al representing natural sources2 (Fig. 3), indicating that Pb and Cr in the SRD mainly originate from human sources rather than natural sources. Existing studies have indicated that traffic emissions lead to the continuous entry of Cr and Pb into the environment47. The vehicular exhaust emissions generated by the combustion of fuel and lubricant oil are the emission sources of Pb and Cr28. Pb in the roadside soil and dust mainly originates from the combustion emissions of leaded gasoline (banned in 2000)2,29,48. In this study, the sampling sites S8, S21, and S35 had fairly high contamination levels of Pb and Cr (Figure S1). S8 was located near the South Second Ring Road in Xi’an, S21 was located near several shopping centers and a subway crossing, and S35 was located near a large agricultural and sideline product trading market and a motor vehicle driver examination field. They are all close to areas with heavy traffic. Considering the previous analysis, Factor 1 is considered to represent the source of vehicle exhaust.

Fig. 2
figure 2

Contribution share of TTEs from different sources in the SRD.

Full size image

Factor 2 was loaded mainly on Zn and Ba, contributing 94.7% and 34.3% of Zn and Ba, respectively (Fig. 2). The aforementioned discussion and evaluation results showed that Zn and Ba in the SRD were principally impacted by human factors. The significant positive correlation between Zn and Ba and their negative correlation with Al (Fig. 3) further indicated that Zn and Ba in the SRD primarily came from similar anthropogenic sources. Research data indicated that Zn and Ba in SRD mainly came from traffic sources and were typical traffic non-exhaust emission TTEs2,49. Zn can be used to prepare alloys or plated on surfaces of metals and other materials to provide corrosion resistance and aesthetics50. It is also widely used in vehicles and is gradually released into the environment with the use and corrosion of vehicles2. Galvanized car parts and volcanized tires by ZnO are also potential sources of Zn, released from cylinder head gasket and tire vehicles, respectively11. In addition, Zn and Ba are widely used in vehicle tires and can enter the environment through the friction between the tires and the ground11,28,51. Meanwhile, Ba is a component in brake discs28,52. Therefore, the frequent use of brake pads continuously introduces Ba into the environment, especially in heavy traffic areas. Among the SRD samples, S17 had the highest levels of Zn and Ba. A large hospital, school, and shopping center near S17 result in a complex traffic environment, which is prone to traffic congestion and severe and continuous slowdowns or stops. Consequently, Zn and Ba continuously enter the environment and cause pollution. These findings and observations support that Factor 2 is a traffic non-exhaust source.

Factor 3 was loaded primarily on Cu, contributing 59.0% of Cu in the SRD (Fig. 2). Cu was not related to other TTEs (Fig. 3), indicating a different source. The Cu content in the SRD of Xi’an university campuses exhibited a high variability, with an average content of more than 4.5 times the local soil background value, and a significant enrichment, indicating that it is mostly affected by human activities. Cu is mostly used as an additive to coatings, such as pigments and paints2, and is utilized in lubricants and radiators for automobiles53. High Cu concentrations were found in sampling sites S15 and S33. S15 is an art college with high levels of Cu on campus, attributable to the extensive use of pigments by students. S33 is located near a traffic police station and auto repair shops. Vehicle damage can lead to lubricant leakage, and the repair process can introduce Cu into the surroundings1,24. Therefore, Factor 3 was considered a mixed source of traffic and paint.

Factor 4 was loaded mainly on V, Mn, Co, Ni, and As, contributing 48.7% of V, 48.9% of Mn, 46.8% of Co, 45.5% of Ni, and 43.1% of As (Fig. 2). Mn is a characteristic element from natural sources, and V is associated with the soil parent, both of which are believed to originate from natural sources14. The content analysis and enrichment factor evaluation results showed that these five TTEs in the SRD from Xi’an university campuses were mainly affected by natural sources. The positive correlation between these TTEs and Al (Fig. 3) further confirms that they mainly originate from nature. Therefore, Factor 4 represents natural sources.

Fig. 3
figure 3

Pearson correlation coefficient of TTEs.

Full size image

In summary, the main sources of TTEs in the SRD of Xi’an university campuses were traffic exhaust, non-traffic exhaust, mixed, and natural sources, with contribution rates of 19.1%, 43.4%, 11.2%, and 26.3% to the total TTEs, respectively (Fig. 2). Therefore, traffic-related sources are the most important anthropogenic sources, accounting for over 60% of the total TTEs in the SRD. Recently, the blowout growth in the number of private cars in Xi’an support these results23,54,55.

Content-based probabilistic ecological risk

Figure 4a displays the content-based ecological risk assessment results of TTEs in the SRD of Xi’an university campuses. The simulated (3.75) and detected (3.00) mean values of the emRI were between 2.3 and 4.5, indicating a considerable risk. Moreover, 2.81% (emRI = 1.1–2.3), 76.90% (emRI = 2.3–4.5), and 20.29% (emRI > 4.5) of the samples were moderate to very high risk, indicating a quite serious ecological risk of TTEs. Sensitivity analysis (Fig. 4a) revealed that Pb had the highest impact on ecological risk (63.7%), followed by Cu (22.9%), Zn (8.7%), and As (3.3%). These results indicate that the ecological risk of TTEs in the SRD in Xi’an university campuses is mainly influenced by Pb, which is the priority control TTE.

Fig. 4
figure 4

Cumulative probability and sensitivity analyses of ecological risk based on content (a) and sources (b)

Full size image

Source-based probabilistic ecological risk

Figure 4b shows the source-based probabilistic ecological risk evaluation results of TTEs in the SRD of Xi’an university campuses. The emRI of TTEs from traffic exhaust emission source, traffic non-exhaust emission source, mixed source, and natural source was 1.29, 0.48, 0.87, and 1.20, respectively (Fig. 4b). The ecological risk value of TTEs from traffic exhaust emission source and natural source was slightly higher than 1.1, indicating moderate pollution. Traffic non-exhaust emission source and mixed source exhibited a low ecological risk (< 1.1). Therefore, the ecological risk of TTEs in the SRD of Xi’an college campuses mainly come from traffic exhaust emissions and natural sources. This speculation is confirmed by their contributions to the ecological risk (64.9%). The sensitive elements in traffic exhaust emission were mainly Pb and Cu, accounting for 83.82%. The sensitive elements in natural sources were mainly Pb, Cu, As, and Cr, accounting for 93.88% (Fig. 4b). In summary, traffic exhaust source is the priority source for controlling ecological risks, and Pb is the priority control TTE owing to its high content and toxicity.

Content-based probabilistic health risk

Table S4 summarizes the content-based MCS results of HQ, HI, CR, and TCR of TTEs in the SRD samples from Xi’an university campuses. Regarding the non-carcinogenic risk, the mean of the HQ for the 10 TTEs and HI for both populations were below the threshold of 1 (Fig. 5). According to the cumulative probability distribution of HI, only 1.04% of female and 0.49% of male students’ value exceeded 1 (Fig. 5). Therefore, the non-carcinogenic risk for college students in the study area can be neglected.

Fig. 5
figure 5

Cumulative probability of non-carcinogenic risk based on content.

Full size image

In terms of carcinogenic risk, the CR of the five carcinogenic TTEs for university students was as follows: Cr (male students: 4.22E-6, female students: 4.78E-6) > As (male students: 2.03E-6, female students: 2.28E-6) > Ni (male students: 8.42E-8, female students: 9.31E-8) > Pb (male students: 7.08E-8, female students: 8.23E-8) > Co (male students: 1.85E-9, female students: 1.66E-9). The CR (Cr) values of 79.51% (male students) and 81.86% (female students) were > 1.0E-6, while 72.01% and 76.82% of CR (As) values were > 1.0E-6 (Fig. 6), indicating the carcinogenic risk of Cr and As to college students cannot be ignored. The carcinogenic risk of the five carcinogenic TTEs for female students was greater than for male students, indicating that the cancer risk of TTEs in the SRD for female students should be given more attention.

Fig. 6
figure 6

Carcinogenic risk cumulative probability of each carcinogenic TTE based on content.

Full size image

The simulated mean TCR values of two groups (Fig. 7a) were female students (7.24E-6) > male students (6.41E-6), with 97.13% of female students and 96.27% of male students having TCR > 1.0E-6, and nearly 0.1% of TCR (male students and female students) were > 1.0E-4. This indicates that the carcinogenic risk of TTEs in the SRD of university campuses cannot be ignored. In summary, the HQ and HI for two groups of people is negligible, whereas the carcinogenic risk is not. Therefore, we conducted sensitivity analysis to evaluate the total carcinogenic risk.

Sensitivity analysis can determine the impact of TTE content and exposure parameters on health risks56. In this study, the sensitivity analysis results showed that the largest contributor to male CR was IngR (76.96%), followed by ED (9.86%), SL (3.78%), and Cr (3.52%) (Figure S4a). For female CR, the contribution of IngR was the highest (77.02%), followed by ED, Cr, and SL, with sensitivities of 9.21%, 4.41%, and 3.30%, respectively (Figure S4b). These results indicate that the higher CR among students is mainly related to ingestion, time spent outdoors, skin contact area, and Cr concentration.

Fig. 7
figure 7

Cumulative probability of TCR based on content (a) and sources (b, c).

Full size image

Source-based probabilistic health risk

The source-based health risk evaluation results of TTEs in the SRD in Xi’an university campuses to college students are shown in Fig. 7b, c. These results indicated that the TCRs of the two groups from four sources exhibited the same characteristics. In other words, the contributions of the four sources to the TCRs of the two groups of people were as follows: natural source > traffic exhaust source > traffic non-exhaust source > mixed source. Among them, the mean values of TCR for natural source and traffic exhaust source were > 1.0E-6. The health risk evaluation results of natural source showed that nearly 87.65% (female) and 93.72% (male) of TCR value were > 1.0E-06, while those of health risk evaluation for traffic exhaust source showed that 47.73% (female) and 89.03% (male) of TCR value were > 1.0E-06, indicating that natural and traffic exhaust source are the key causes of cancer risk among college students. According to the sensitivity analysis of natural source CR, IngR was the main influencing factor, followed by As, SL, and ED (Figure S4). IngR was also the main factor affecting the health risk of traffic exhaust source, with other influencing factors such as Cr, ED in order (Figure S4).

According to the source-oriented health risk assessment results, traffic exhaust source is the priority source for TTE pollution control in the SRD in Xi’an university campuses, with Cr as the priority pollutant.

Conclusion

The concentrations Zn, Pb, Cr, Cu, and Ba in the SRD in Xi’an university campuses were significantly higher than the local soil background values. The comprehensive pollution level of TTEs in the SRD was high, mainly caused by Zn and Pb. Traffic exhaust source, traffic non-exhaust source, mixed source, and natural source were the main sources of TTEs in the SRD of Xi’an university campuses. The ecological risk of TTEs in the SRD of university campuses was quite serious, mainly caused by Pb from traffic exhaust source. TTEs in the SRD have a certain cancer risk for college students, and traffic exhaust source was the main contributor of the anthropogenic sources to the cancer risk of TTEs in the SRD.

These findings suggest that the environmental impact of traffic activities and their health risks to college students should be given careful attention. According to the source-oriented ecological-health risk assessment results, traffic exhaust source is the priority control source, and Pb and Cr are the priority control TTEs. Relevant management departments should strengthen the monitoring and management of traffic exhaust emissions.