Introduction

Asbestos fibers were widely used in commercial, construction, and industrial applications due to their high strength and resistance to heat and corrosion before their hazardous health effects were fully recognized1. Over time, asbestos-containing materials degrade under environmental conditions such as wind, rain, humidity, and temperature changes, releasing fine fibers2. These fibers are airborne, and their small, odorless, and tasteless nature increases the risk of unintentional inhalation3,4,5.

Inhalation of asbestos fibers is the main exposure route and has been linked to severe lung diseases such as pulmonary fibrosis (asbestosis), lung cancer, and mesothelioma5,6,7,8. The World Health Organization (WHO) classifies asbestos as a Group 1 carcinogen for humans, based on extensive epidemiological and toxicological evidence. A long latency period of 10 to 30 years is a key feature of asbestos-related diseases9.

Iran banned the use and importation of white asbestos on July 20, 2007; however, recent studies have identified high concentrations of these fibers in urban areas10. Table 1 summarizes the mentioned results. For instance, Maleki et al. (2020)11 reported asbestos fiber levels in Tehran ranging from 1.46 × 10−3 to 2.31 × 10−3. Similarly, Moteallemi et al.‘s study in Mashhad indicated levels above the standard3. The presence of asbestos fibers in urban air stems primarily from construction activities and the use of asbestos-containing brake and clutch pads in vehicles. Building materials like thermal insulation, floor tiles, pipe insulation, fire retardants, ventilation ducts, and chimneys commonly contain asbestos12. Demolition of old commercial or residential buildings also releases asbestos fibers into the air7,13. According to Iranian census data from 2016, approximately 300,000 people reside in Shiraz’s old urban fabric, which covers about 1800 hectares. This makes the demolition of old buildings a significant potential source of asbestos fiber release14. Furthermore, according to the Shiraz Traffic Organization’s annual transportation report (2014), the number of registered vehicles in the city increased from 250,000 in 2007 to 700,000 by 2014 a 180% increase in just seven years15. Brake linings are a potential source for asbestos exposure, as they historically contained 33–73% asbestos. A study by Jacko and Ducharme (1975) estimated that approximately 32 million kilograms of asbestos fibers are released annually into the environment in the United States due to brake lining abrasion16. There is limited data available on urban airborne asbestos levels in Iran. Given the country’s four-decade history of asbestos use, it is essential to measure current airborne asbestos concentrations in order to develop effective control strategies17. Continuous measurement and evaluation of asbestos concentrations, particularly in polluted cities like Shiraz, Iran’s fifth-most populous city, are essential. The findings of this study can serve as a valuable database for controlling asbestos emissions and safeguarding public health. Therefore, the objective of this study was to quantify asbestos fiber concentrations in the ambient air of Shiraz City, analyze their spatiotemporal variations, and assess associated health risks. In addition, we aimed to explore potential correlations between asbestos concentrations and factors such as road network density, population density, number of bus stations, and green space.

Table 1 Mean values of asbestos levels in different cities of Iran.
Full size table

Materials and methods

Study area

Shiraz, located in Fars Province in the southwest of Iran, is the fifth-largest city in Iran and comprises 11 districts. It lies between 29°36′36″N and 52°32′33″E, covering a total area of 340 km2. This city is ranked fifth in Iran by population (1.86 million people in 2016). It has a cold semi-arid climate with distinct seasons. The weather is mild in spring and autumn, hot and dry in summer, and cold and moderate in winter. In Shiraz, July is the hottest month with an average temperature of 38.8 °C, and January is the coldest with an average temperature of − 5 °C. Most of the annual precipitations (around 300 mm) occur in winter. The prevailing winds in Shiraz are from the northwest and west, with an average speed of 7 m/s. Shiraz city is among the most polluted cities in Iran.

Sampling methods and sample analysis

Sampling location

The sampling stations were set up on the rooftops of low-rise buildings (mostly single-story), at designated locations in Shiraz. The sampling inlets were placed at a height of approximately 3 m above ground level. The height was selected to represent the human breathing zone in urban areas and to measure the asbestos concentration that would be expected in normal human exposure situations. Erecting the samplers on the rooftops also helped to avoid interferences from local sources like vehicle exhaust at ground level and, importantly, kept the equipment secured and protected. The height we selected is consistent with many similar practices in air pollution studies, which suggest sample heights between 2 and 4 m above ground level17,28. 18 sampling sites were selected from different areas of Shiraz city. Representative sampling sites were selected to cover the industrial sector, traffic zone, and residential area. Figure 1 displays the location of the sampling sites selected in the city.

Fig. 1
figure 1

The study area locations and urban land use (road types, bus stops, green spaces, and parking) of Shiraz. This map was created by the authors using ArcGIS® software by Esri (ArcMap version 10.8.1; https://www.esri.com).

Full size image

Sampling process

8-hour air samples were collected from ~ 8:00 am every day within a year of the sampling period from 12 May 2019 to 2 February 2020, using a low-volume peripheral pump (SKC, Leland Legacy, USA) with a cassette holder. Airflow passed through a cellulose ester filter (MCE, diameter 25 mm, pore size 0.45 μm), with a flow rate of 6 L/min. The parameters (time and flow) were set before the sampler was operated. An entire 72 samples (18 stations × 4 seasons) were collected from the whole process. On the whole, the samples were collected under no-rain weather conditions, then were put into a covered storage box and stored at 4 °C until analysis.

Sampling period

Samples were collected from the stations during spring, summer, autumn, and winter seasons as per the following schedule:

  1. a.

     Spring: From May to June 2019.

  2. b.

     Summer: From August to September 2019.

  3. c.

     Autumn: From November to December 2019.

  4. d.

    Winter: From January to February 2020.

Sampling analysis

PCM analysis

In our present study, the NIOSH-7400 method was incorporated to determine the asbestos concentrations29. First, the sample or control filters were clarified separately on the glass slide using acetone vapor. Concerning the instructions, the amount of the acetone injected for clarification via the acetone vaporizing device could be in the range of 200–350 µl. The asbestos fibers were then tested using phase-contrast microscopy (PCM) (green filter with diaphragm adjusting mechanism, Walton-Bechette graticule (type G-22), magnification × 400). Asbestos fibers are well-defined as visible particles having a length greater than 5 micrometers and an aspect ratio greater than or equal to 3:1. Then, after counting the fibers, their levels in the air volume sampled were calculated using formulas (1) and (2):

$$\:E=\:\frac{(\frac{F}{nf}-\frac{B}{nb})}{Af}$$
(1)
$$\:C=\frac{Ac\:\times\:E}{V}$$
(2)

Where E is defined as the density of fiber on the filter surface, F/nf and B/nb are the numbers of fibers counted at each graticule field respectively in the sample and the control filter, Af is the graticule field, C is the fiber concentration (F/L), Ac is the applicable surface area of the filter (mm2, and V is the air volume sampled (L).

Spatial and statistical analysis of asbestos

The asbestos spatial distribution in Shiraz city was analyzed and mapped by ArcGIS software, version 10.2. The resulting data were analyzed via SPSS v.23.0. Furthermore, the Spearman correlation test was used to determine whether any correlation could be identified among the concentrations of asbestos fibers and meteorological data (wind speed, temperature, and relative humidity).

Health risks assessment

In this study, a model developed by the EPA was used to estimate people’s exposure to asbestos. Researchers have used this model to assess the health risks of asbestos fibers. To do so, the excess lifetime cancer risk (ELCR) was estimated in the city’s air due to asbestos fiber inhalation. The ELCR was calculated using Eqs. 3 and 430.

$$\:EC=\frac{C\:\times\:ET\:\times\:EF\times\:ED}{AT}$$
(3)
$$\:ECLR=EC\:\times\:URF$$
(4)

Where, EC is defined as chronic exposure concentration (average value during a 70-year lifetime) (f/cm3), C is the asbestos level stated in fibers per cubic centimeter of air (f/cm3), ET is the time of exposure stated in hours per day, EF is the exposure frequency stated in days per year, ED is the exposure duration stated in years, AT is the averaging time, and URF is the unit risk factor for inhalation of asbestos fibers ((f/cm3) −1). In this study, the URF was considered 0.2331. For a better understanding of the uncertainty about assumptions made in the exposure model, a sensitivity analysis was performed. It was done to explore the likely variability of ELCR values based on different exposure conditions. There were two exposure scenarios modeled: a conservative scenario with continuous exposure (ET = 24 h/day, EF = 365 days/year for ED = 70 years) and a representative scenario modeling average urban exposures (8 h/day, 180 days/year for 30 years). Moreover, the Thiessen Polygon method was applied to determine each sampling station’s area. Based on this method, the sampling stations were divided into 18 polygons. Using this approach, the study area was divided into 18 polygons, which fully covered the city. The Excess Lifetime Cancer Risk (ELCR) was then calculated for each polygon. Additionally, to determine the most vulnerable areas, a population block map of adults was created using 2016 census data from the Statistical Center of Iran, and the adult population within each polygon was estimated. Vulnerable areas were identified by multiplying the ELCR by the adult population in each polygon. Identifying the most vulnerable areas as a statistical technique enables better consideration of the carcinogenic effects of asbestos in more densely populated zones during analysis and risk assessments.

Spatial analysis

The land use map of the study area was prepared, and the layers related to bus stations, population, and roads were extracted. Moreover, satellite images were used to provide the green land map. To this end, Normalized Differential Vegetation Index (NDVI) images of the study area were prepared, and the vegetation layer was developed according to the method presented by Khalili et al.32. The corresponding density maps were created using the Kernel Density Estimation (KDE) method to explore the effect of each land use on the asbestos concentration. We computed the average density of land use within three distances: 300, 500, and 700 m around each sampling station (Figs. 2 and 3). Then, we examined their correlation with asbestos fiber concentrations.

To create a road density map of the study area, roads were classified into four categories based on traffic density: 1st-class for arterial roads and highways, 2nd-class for county-standard city roads, 3rd-class for alley roads, and 4th-class for other types of roads. Each road class was assigned a weight (1 for 1st-class, 2 for 2nd-class, 3 for 3rd-class, and 4 for 4th-class), reflecting their impact on density proportional to their size and traffic volume. Population density (individuals/km²) at each station was calculated using 2016 census data. Additionally, the study area was divided into 2,831 units of 300 m × 300 m (Fig. 1), and average densities for each land use were computed. Statistical significance of high and low density areas for each land use was determined through Hot Spot analysis. Also, we used multiple linear regression (MLR) to predict the relationship between asbestos and land use (roads, bus stations, population, and green space). Critical assumptions of the model, including normality of residuals, no multicollinearity of independent variables, and independence of observations, were verified before analysis. Multicollinearity was verified with the Variance Inflation Factor (VIF) and Tolerance measures. All the VIF scores were below 10, while tolerance measures were greater than 0.2, which did not indicate any problem of multicollinearity among the predictors. Statistical analysis was conducted using SPSS Version 26.0.

Hot spot analysis (Getis-Ord Gi*)

We implemented Hot Spot analysis to identify hot and cold spot areas with 90%, 95%, and 99% confidence levels. Spatial autocorrelation, necessary for determining hot spots, was analyzed using Moran’s I index, including both local and global Moran indices (Srikanth & Srikanth, 2020; Zhao et al., 2020). The local Moran index detects the spatial distribution of hot and cold spots and compares them with adjacent samples. The Getis-Ord Gi* statistic identifies spatial clusters with statistically significant high-value (hot spot) and low-value (cold spot) areas. The statistical formula for Gi* is given by Eq. 533.

$$\:{G}_{i}^{*}=\frac{\sum\:_{j=1}^{n}{w}_{ij}{x}_{j}-\stackrel{-}{X}\sum\:_{j=1}^{n}{w}_{ij}}{S\sqrt{\frac{[n\sum\:_{j=1}^{n}{w}_{ij}^{2}-{\left(\sum\:_{j=1}^{n}{w}_{ij}\right)}^{2}}{n-1}}}$$
(5)

Where wij is defined as the spatial weight between attributes i and j, xj is the value observed for attribute j, n is the number of attributes, and S is the sum of all spatial weights. Moreover, Eq. 6 was used to calculate the S factor.

$$\:S=\sqrt{\frac{\sum\:_{j=1}^{n}{x}_{j}^{2}}{n}}-\:{\left(\stackrel{-}{X}\:\right)}^{2}$$
(6)
$$\:\stackrel{-}{X}=\frac{\sum\:_{j=1}^{n}{x}_{j}}{n}$$
(7)

The outputs of the Gi* statistic include the P-value and z-score, where the latter is normalized based on the mean and variance of the sample Gi. High positive z-scores are known as hot spots. In addition, a smaller negative score of z means low values known as cold spots; near-zero z indicates a random distribution.

Output maps

The Hot Spot Analysis tool results are classified by default into seven distinct categories for each land use, covering a wide range of scenarios. Additionally, measuring stations were categorized based on pollutant concentrations: high, medium, and low concentrations. To this end, the K-means cluster analysis was used to classify the stations based on the measured concentrations.

Results and discussion

The concentrations of asbestos fibers

Figure 2 presents the levels of asbestos fibers in air samples based on PCM analysis. The mean asbestos concentration across all stations was 1.25 F/L, with values ranging from0.92–1.55 F/L. The highest concentrations were observed at Stations 7 and 14 during autumn (2.56–2.64 F/L), while the lowest was recorded at Station 2 in summer (0.03 F/L). As shown in Fig. 2, some stations exceeded the permissible limit of 2 F/L34. Notably, stations 7 and 14 consistently showed the highest asbestos levels across all seasons, identifying them as the most polluted areas in Shiraz. These elevated levels are likely due to the high density of commercial activities, heavy traffic, and proximity to Highway23. Figure 2 also illustrates the spatial distribution of asbestos concentrations across Shiraz: 10% of sites (2 stations) showed low levels (< 0.5 F/L), 20% had moderate levels (0.5–1 F/L), 60% had high levels (1–2 F/L), and 10% recorded very high levels (> 2 F/L). Given the acceptable threshold of 2 F/L, it can be concluded that asbestos concentrations exceeded this limit at 10% of the sampling sites, while the remaining stations reported levels below the threshold. Studies in Iran suggest that environmental asbestos pollution mainly stems from emissions related to vehicle brake pads and clutch plates18,19. Similar findings have been reported in Tehran and Isfahan11,18. In comparison, the asbestos concentrations found in our study were significantly higher than those reported in Zahedan city (1.2 F/L)25. However, the mean values in our samples were similar to those observed in Urmia (0.0018 F/mL)24. Various characteristics of sampling stations and different measurement strategies may contribute to the differences in interpretation of the results between this study and others. Corn et al. (1994)35 emphasized that interpreting asbestos measurements based on phase-contrast microscopy (PCM) can differ substantially depending on whether samples are collected in occupational versus non-occupational sites.

Fig. 2
figure 2

Temporal trend of asbestos fibers levels (F/L) at the sampling stations.

Full size image

Asbestos fiber concentrations by season and correlations with meteorological data

The Meteorological Department of Fars Province provided us with meteorological parameters such as temperature, humidity, and wind speed during the sampling period, which we summarized in Table 2. We used the Pearson correlation test to examine the relationship between airborne asbestos fiber levels and meteorological data. The results showed no significant relationship between asbestos levels and wind speed (P = 0.177, n = 72, r = −0.333), humidity (P = 0.730, n = 72, r = −0.087), or temperature (P = 0.730, n = 72, r = −0.087). This lack of correlation may be due to the special properties of asbestos fibers, like being water-repellent and settling slowly, which make them less affected by short-term weather changes than other air pollutants. Also, Shiraz’s crowded city structure with narrow streets and tall buildings probably blocks the wind and causes asbestos to build up in certain areas, even if the wind or humidity changes36. These findings are consistent with previous studies conducted in other Iranian cities3,17,19. However, the results of previous studies have shown contradictory findings. For example, Ghorbani et al. (2020)18 reported a significant negative correlation between asbestos fiber concentrations and temperature and wind speed. In contrast, Johnston et al. (1982)37 found no significant relationship between wind speed and asbestos fiber concentrations. These differences may be due to local conditions, like how cities are built, what the main asbestos sources are, or whether asbestos use has stopped in those areas. In Shiraz, asbestos mainly comes from brakes and old buildings, so things like how much traffic there is might matter more than the weather. According to Table 2, winter (1.55 F/L) and summer (0.92 F/L) recorded the highest and lowest levels of asbestos fibers, respectively (see Fig. 2). Asbestos levels were much higher in the colder seasons than in the warmer ones (P = 0.000, n = 72, r = 0.848) (Table 3). This difference between seasons might be because there’s less traffic in hot weather and cooler temperatures in winter change how air pollution spreads21.

Table 2 The mean values of meteorological data during different seasons of the study period (one year) in Shiraz city.
Full size table
Table 3 Correlation analysis of the asbestos fiber concentration between cold and warm seasons.
Full size table

Density of each land use

The density of each land use is illustrated in Fig. 3. Regarding bus stations, the central areas of Shiraz had a density of more than 10 N/km2. In contrast, the southern regions had a density of fewer than 1 N/km2. The results for green space density indicated that the highest density (298,000 m²/km²) was found in the northern part of Shiraz. Furthermore, road density exceeded 26.47 km/km² in the east and northwest, while it was less than 9.9 km/km² in the city’s southwest. The city center recorded the highest population density, with 15,000 persons/km2. These findings showed that Shiraz’s center, northwest, and east had the highest land use densities.

Fig. 3
figure 3

Density distribution of urban land-use indicators in Shiraz city: (A) Population, (B) Road, (C) Green spaces, (D) Bus stations. This map was created by the authors using ArcGIS® software by Esri (ArcMap version 10.8.1; https://www.esri.com).

Full size image

We also performed a hot spot analysis to determine the maximum density of each land use in the study area. Figure 4 shows the pattern of cold and hot spots, categorized into seven groups based on z-scores. This pattern is similar to the land use density map. The analysis revealed clear differences between the north, south, and center of Shiraz. The main distribution of hot spots for road networks and populations was at s11, s14, and s3. Additionally, the majority of green space hotspots were associated with the northern areas of the city, primarily concentrated in stations s2 and s4. The bus station’s hot spots were also in the city’s central areas, as well as a small area in the northwest. Cold spot areas for population, bus stations, and road networks were mostly in the south (e.g., station s9) and suburban zones, while there were no significant cold spots for green space. Moreover, other regions showed no significant clustering for either high or low values.

Fig. 4
figure 4

Hot spot analysis results for density of land uses: (A) Population, (B) Road, (C) Green spaces, (D) Bus stations. The map was created by the authors using ArcMap version 10.8.1 (https://www.esri.com). This map was created by the authors using ArcGIS® software by Esri (ArcMap version 10.8.1; https://www.esri.com).

Full size image

According to the results obtained from measuring land use density, it was found that density varied significantly among different stations throughout the city, with the highest densities concentrated in the central areas. This created notable differences in land use across Shiraz. Hot spots were identified in the center and northwest of Shiraz, specifically at stations s11, s3, s10, and s14, while the northern and southern parts of the city were classified as cold spots, having significantly lower densities.

Furthermore, Fig. 4 shows several asbestos hot spots in the city. The northwest and southeast, which have the densest roads and limited green space, showed the highest asbestos levels, compared to the northern areas.The hot spot analysis also revealed that higher road density was linked with more asbestos, and lower density with less. This means traffic has a key role in asbestos distribution. This finding aligns with a study by Awad et al. (2011)38which reported the highest concentrations of asbestos fibers near road networks. Consequently, the main points of concern were located along the highways.

Also, the two main roads in the northwest and east of Shiraz were areas with the highest asbestos levels. These results show that traffic is a major factor in asbestos spread. Asbestos levels were moderate to low in areas with more green space. However, stations 13 and 18, which had low asbestos levels, were located in areas with little green space—showing that green space may not strongly affect fiber levels. The city’s center, east, and northwest areas had the highest land use density, according to the results. As a result, it can be concluded that these are also the busiest parts of the city, with heavier traffic. In contrast, the north, west, and south had both lower land use density and lower asbestos levels than other areas.

Relationship between asbestos fibers and land use type

Table 4 reports the correlation coefficients between asbestos fiber concentrations and each type of land use. According to the results presented in Table 4, a significant relationship existed between asbestos fibers and the road network (P < 0.01). The Spearman’s correlation coefficients for the asbestos fiber concentration and the road network were found to be 0.610, 0.792, and 0.820 for the 300, 500, and 700 m radius, respectively (p < 0.01). As the density of the road network increased, so did the concentration of asbestos fibers.

The multiple linear regression model (adjusted R² = 0.773, p < 0.001) revealed significant associations between land-use variables and asbestos fiber concentrations (Table 5). Road density exhibited the strongest positive influence (β = 0.679, p < 0.001), with a 1-unit increase in road density corresponding to a 0.093-unit rise in asbestos concentration (B = 0.093). Similarly, population density (β = 0.441, p 0.015) and bus stops (β = 0.361, p = 0.026) were positively affected with statistical significance, which means that areas with more people and urban infrastructure tend to have higher asbestos levels. In contrast, green space density was negatively correlated with asbestos concentrations (B = −0.000001, β = 0.136), although the correlation was not significant (p = 0.301). This small effect might be due to limited green space differences in the city and needs more study with a larger dataset.

The model had a high predictive power (R² = 0.773), showing it can well estimate asbestos levels. The beta values in Table 5 illustrate the impact of the independent variables on the dependent variable. Notably, the road density parameter exerts the most significant influence on asbestos fiber concentration among the variables studied. The negative beta value for the green space parameter confirms its inverse relationship with asbestos fiber concentration.

Also, as seen in Fig. 4, asbestos levels were higher in the northwest and east, where road density was also high. As shown in Figs. 3 and 4, the road network density in the south and southwest areas was lower than in other parts of the city, resulting in a significant variation in fiber concentration among the stations. With more roads and traffic, car-related emissions like asbestos increase because drivers use brakes more often. Consequently, pushing down the car’s brake causes the asbestos-containing brake pads to release and disperse asbestos dust as a friction product in the urban environment, leading to an increase in the concentration of asbestos fibers (Khadem, Somea, Hassankhani, & Heravizadeh, 2018). According to a study carried out in the United States, the asbestos level in the vicinity of expressways was about 3.3 F/L39.

We also evaluated the relationship between the effect of urban green space and the concentration of asbestos fibers; it is commonly believed that green spaces help reduce air pollution40. As reported in previous studies, urban green spaces can prevent the spread of pollutants such as volatile organic compounds and other pollutants41. They can trap, filter, and absorb air pollutants to some extent42. Accordingly, urban areas have more green spaces due to greater exposure to air pollutants. However, this study found no positive relationship between urban green space and asbestos fibers. Also, the results showed no strong link between the number of bus stations and asbestos levels, meaning these factors didn’t clearly reduce or raise asbestos concentration.

Table 4 Correlation coefficient between different land-use types and asbestos fibers.
Full size table
Table 5 Model coefficients assessed by multiple linear regression for asbestos fibers.
Full size table

Risk assessment of exposure to asbestos fibers

There are only a few published reports associated with ambient asbestos exposure, unlike occupational exposure studies. Thus, it is arduous to compare our findings with those in other studies. In addition, minimal research has used the outdoor concentrations of airborne asbestos to evaluate the effects of exposure to these fibers. However, most studies have focused on how asbestos causes cancer, not its other possible health effects. Therefore, the present study investigated the carcinogenic risk of exposure to asbestos-contaminated air. We conducted an asbestos risk assessment using data from the PCM method. Generally, how much someone is exposed depends on the fiber levels in the air and the type of exposure. In this regard, two scenarios for asbestos risk assessment are defined as follows: (1) Occupational exposure and (2) Public exposure; the main difference between these is how long people are exposed 8 h for workers and 24 h for the public. Because our purpose was to examine public exposure, the duration of exposure was considered to be 24 h. Table 6 summarizes the exposure concentration (EC) values.

Table 6 provides the excess lifetime cancer risk (ELCR). Based on the results, the ELCR value was in the range of 8.3 × 10−5−5.22 × 10−4. Furthermore, the mean value of ELCR calculated was equal to 2.89 × 10−4. This means that for every 10,000 people in Shiraz, about 2.89 may develop cancer due to asbestos exposure. These values fall near or above the upper limit of the EPA’s acceptable risk range of 10⁻⁶ to 10⁻⁴, suggesting that asbestos exposure in certain areas of the city could pose a significant public health concern.

To better understand how reliable the cancer risk estimates are, we performed a sensitivity analysis to account for possible changes in exposure assumptions. Because default exposure assumptions (i.e., 24 h/day, 365 days/year, 70 years) may not reflect actual population behaviors, two exposure scenarios were tested. The analysis showed substantial decreases in ELCR values under more realistic conditions, with reductions of over 70% in some cases. These results show that cancer risk estimates can change a lot depending on the exposure assumptions, so it’s important to consider such uncertainties.

Consequently, this study’s health risk assessment analysis showed that the ELCR value obtained was higher than the recommended value. In a similar study in Tehran, ELCR (3.56 × 10⁻⁴) was even higher than what we found11. Also, the result obtained from research in Shahriar (ELCR = 2.2 × 10−4) showed a higher value than that in the present investigation19. We used the Thiessen Polygons method to divide the city into areas based on each sampling station (Fig. 5A). We divided Shiraz city into 18 polygons based on the number of sampling stations. We then estimated the population of each polygon using the population map (Fig. 5B).

As shown in Fig. 5B, Station s12 had the highest population, followed by Stations 3 and 10, with higher populations than the other stations. Furthermore, stations 13, 18, and 4 had the lowest populations. Figure 5C shows the cancer risk calculated for each polygon using the ELCR formula. According to Fig. 5C, the highest health risks were at stations 14 and 7. One of the reasons for the increase in station 7 was its proximity to the busy urban area and the Delgosha Highway.

Furthermore, station 14 was situated near the Shiraz Ring Expressway, which is the city’s largest highway. This highway connects many main and side roads and carries a mix of light and heavy traffic, including passing and local vehicles all of which increase car numbers and pollution. However, stations 13 and 2 exhibited the lowest health risk. Afterward, we multiplied the population of each area by the calculated risk amount (Fig. 5D) to determine the areas with the highest vulnerability. Based on this figure, stations 14 and 8 had the highest population-level health risk, while stations 13, 18, 4, and 2 had the lowest. The Thiessen Polygons method revealed that the west (near the Ring Expressway) and southeast (near the highway and high-traffic area) of the city had the highest ELCR values. Therefore, some areas of Shiraz with more asbestos fibers in the air may pose a cancer risk. To reduce this threat, we need to take steps to control asbestos pollution and protect public health.

Table 6 The lifetime excess cancer risk (ELCR) of asbestos fibers under different exposure scenarios.
Full size table
Fig. 5
figure 5

Spatial analysis of population distribution, cancer risk, and vulnerability in Shiraz; (A) Area map of each sampling station (Thiessen Polygons method), (B) population block map of Shiraz city, (C) the excess lifetime cancer risk (ELCR) map, and (D) the city’s most vulnerable areas map (ELCR × Population in each area).

Full size image

Conclusions

This study aimed to investigate the one-year asbestos concentrations in the Shiraz metropolis, assess the resulting health risks, and examine the relationship between fiber concentrations and the density of various land uses (road networks, population, number of bus stations, and green spaces). According to the results, the mean ELCR for asbestos inhalation was 1.44 × 10−4, indicating a cancer risk for 1.44 out of every 10,000 people. The highest ELCR values were found in the west (close to the Shiraz Ring Expressway) and the southeast (near areas with heavy traffic). Additionally, there was a positive and significant relationship between asbestos pollutants and road networks. However, no significant correlation was observed between fiber concentrations and population density, bus stations, or green spaces. The study highlighted the significant role of road networks and seasonal changes in increasing the concentration of pollutants in Shiraz, emphasizing the impact of traffic activities on the city’s air pollution. Based on these findings, we recommend strategies that could include enhanced regulation and the phased removal of asbestos-containing brake materials, especially in high-traffic areas. Also, establishing regular air quality monitoring in the identified regions of risk and integrating asbestos fiber monitoring in urban environmental health initiatives is imperative. Although this study did not test it directly, urban planning efforts aimed at reducing traffic congestion and encouraging low-emission transportation methods could help lower long-term exposure to asbestos fibers.