Source dynamics and environmental risk of street dust as a vector of human exposure to potentially toxic elements in Istanbul, Türkiye

Abstract

Urban street dust acts as both a sink and a secondary source of potentially toxic elements (PTEs), contributing to environmental contamination and air quality degradation. Using geochemical and statistical methods, this study aimed to investigate the concentrations, spatial distribution, ecological risks, sources, and associated health risks of selected PTEs (Al, As, Cd, Co, Cr, Cu, Fe, Mn, Ni, Pb, Zn) in < 63 μm street dust samples collected from 29 locations across Istanbul, Türkiye. Elemental concentrations were determined using ICP-MS, and contamination was evaluated using geo-accumulation (I_geo), enrichment (EF), contamination (CF), potential ecological risk (PERI), and Nemerow’s pollution indices (NPI). Positive Matrix Factorization (PMF) identified three dominant sources: industrial runoff (39.4%), traffic emissions (31.3%), and natural/soil inputs (29.4%). Health risk assessments indicated ingestion as the primary exposure pathway. Monte Carlo simulation revealed that the 95th percentiles of THI (3.57) and TCR (2.61 × 10⁻⁴) exceeded recommended thresholds for children, indicating potential non-carcinogenic and carcinogenic risks, while adult risks largely remained within acceptable limits. Traffic-related elements such as Pb, Cu, and Zn were the major contributors to non-carcinogenic risks, with additional implications for inhalation exposure through dust resuspension. Although the Air Quality Index (AQI) remained below 50, suggesting generally good atmospheric conditions during the study period, localized dust contamination was found to pose significant health risks. These findings emphasize the need for integrated mitigation strategies, including traffic emission controls, dust suppression, and urban greening, to minimize PTE exposure and enhance urban environmental health.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-025-11472-2.

Introduction

The rapid growth of urban populations and infrastructure has intensified environmental pressures in metropolitan areas, where restricted air circulation due to dense building layouts contributes to pollutant accumulation. These conditions have led to elevated human exposure to airborne contaminants, reflected in increasing levels of traffic congestion, noise pollution, and declining air quality, thereby posing significant risks to public health and urban sustainability¹. In other words, urban environments are increasingly challenged by contamination with potentially toxic elements (PTEs), such as arsenic, cadmium, lead, chromium, and nickel These elements originate from both natural sources and anthropogenic activities, including rapid urbanization, industrial processes, and vehicular emissions². Among various environmental matrices, street dust acts as a major reservoir and transport medium for these contaminants. It also represents a critical source of particulate matter (PM) in urban atmospheres, significantly influencing pollution dynamics and posing substantial public health risks³.

Human exposure to PTEs, including arsenic, lead, nickel, chromium, and cadmium, primarily arises through ingestion, inhalation, and dermal penetration⁴. These contaminants, introduced into urban environments via industrial discharges, construction activities, vehicular emissions, and atmospheric deposition, often accumulate in urban street dust. Once deposited, PTEs can be resuspended into the air or migrate into water and soil systems, thereby amplifying exposure risks⁵. Arsenic, a recognized carcinogen, is associated with bladder, skin, and lung cancers, as well as developmental disorders and cardiovascular problems^6,7. Chromium, particularly in its hexavalent form, exhibits high carcinogenicity and mutagenic toxicity, while nickel exposure is linked to respiratory diseases and allergic reactions⁸. Lead primarily affect neurological and renal systems^9,10. Cadmium toxicity is a serious health concern, primarily damaging the kidneys, lungs, and bones, and is classified as a human carcinogen with severe acute and chronic effects dependent on exposure¹¹. Prolonged exposure to these elements has been further correlated with increased risks of cardiovascular disease and cancer¹². Children represent a particularly vulnerable group due to their frequent ground-level activities and greater physiological sensitivity¹³. The co-existence of multiple PTEs in street dust exacerbates health hazards, underscoring the urgent need for specific actions designed to limit exposure and protect community health. Implementing effective strategies is essential to minimize the adverse health and ecological impacts of urban street dust contamination^14,15.

Air quality refers to the degree to which the ambient atmosphere is clean or polluted, typically measured by the concentrations of key pollutants such as particulate matter (PM₁₀ and PM_2.5), ozone (O₃), nitrogen dioxide (NO₂), sulfur dioxide (SO₂), and carbon monoxide (CO), and is often expressed using composite indices like the Air Quality Index (AQI)¹⁶. Therefore, air quality plays a critical role in safeguarding human health and promoting environmental sustainability. Comprehensive understanding of pollutant concentrations and their impacts provides essential insights for regulatory compliance and public health protection^17–19. Istanbul, a transcontinental megacity characterized by a combination of historical heritage and rapid urbanization, faces a complex and evolving pollution landscape. High population density, extensive transportation networks, significant industrial activities, and ongoing urban development projects contribute substantially to the contamination of the city’s environmental media²⁰. Meteorological dynamics, including the interaction of local wind patterns and seasonal precipitation, further influence the transport, dispersion, and deposition of street dust contaminants, emphasizing the necessity for integrated assessments of pollution sources and environmental drivers²¹. Although previous studies have addressed soil and water contamination in Istanbul^22,23 comprehensive investigations focusing specifically on the source identification, spatial distribution, and health risks of PTEs in urban street dust, along with their relationship to meteorological factors, remain limited.

PM not only acts as a vector for potentially toxic elements in urban environments but also plays a crucial role in atmospheric processes related to climate change. Fine particles can influence radiative forcing, urban heat island effects, and cloud formation dynamics²⁴. Additionally, the increasing frequency of extreme weather events and prolonged dry periods associated with climate change may enhance the resuspension and deposition of contaminated dust, thereby intensifying human exposure in densely populated areas²⁵. The AQI is a standardized system used to assess air pollution levels based on key atmospheric pollutants. In this study, the monitored parameters included PM_2.5, PM₁₀, NO₂, and SO₂, each with distinct environmental and health implications²⁶. PM_2.5 (particulate matter ≤ 2.5 μm) consists of fine particles primarily emitted from combustion sources, such as vehicle exhaust, industrial activities, and biomass burning. Due to their small size, PM_2.5 particles can penetrate deep into the respiratory tract and are associated with cardiovascular and pulmonary health risks²⁷. PM₁₀ (particulate matter ≤ 10 μm) is coarser, originating from sources such as road dust, construction activities, and industrial emissions. Although less harmful than PM_2.5, PM₁₀ can cause respiratory irritation and contribute to overall reductions in air quality²⁸. NO₂, a gaseous pollutant generated mainly from vehicular emissions and industrial processes, can exacerbate respiratory diseases, particularly in vulnerable groups such as individuals with asthma²⁹. SO₂ is primarily emitted from fossil fuel combustion, industrial processes, and volcanic eruptions and is known to cause respiratory ailments while contributing to acid rain formation²⁶. Both the United States Environmental Protection Agency Air Quality Index Scale¹⁹ and the World Health Organization Air Quality Guidelines¹⁸ provide standardized frameworks for evaluating air quality and associated health risks. These guidelines set critical thresholds for pollutants such as PM₁₀, PM_2.5, Pb, and As, which are essential for contextualizing exposure risks in urban environments.

Thus, analyzing urban street dust’s geochemical composition and associated human health risks is crucial for sustainable urban planning and environmental management, aligning with UN Sustainable Development Goals 3 (Good Health and Well-Being) and 11 (Sustainable Cities and Communities) by identifying pollution sources and evaluating exposure risks to minimize environmental degradation and protect public health in metropolitan areas. Recent studies have applied machine learning and optimization methods to environmental pollution modeling^30,31. However, this study aims to address these knowledge gaps by systematically investigating street dust contamination in Istanbul, with an emphasis on the quantification, source identification, and health risk assessment of PTEs. The specific objectives of the study are as follows: (i) to quantify the concentrations of key PTEs, including arsenic, chromium, lead, cadmium, and nickel, in urban street dust collected from diverse districts of Istanbul and to analyze their spatial distribution patterns; (ii) to identify the major sources of PTE contamination through advanced statistical techniques, including Positive Matrix Factorization (PMF) and hierarchical cluster analysis (HCA), while evaluating the influence of meteorological factors such as wind patterns, precipitation, and temperature on contaminant dispersion and deposition; (iii) to assess ecological risks using multiple contamination indices, namely the geo-accumulation index (I_geo), contamination factor (CF), enrichment factor (EF), ecological risk index (E_rⁱ), pollution load index (PLI), potential ecological risk index (PERI), and Nemerow’s pollution index (NPI); (iv) to evaluate non-carcinogenic and carcinogenic health risks associated with PTE exposure, refined through Monte Carlo Simulation to enhance the accuracy and robustness of risk estimations; and (v) to propose targeted mitigation strategies aimed at reducing PTE contamination and associated health risks, with particular consideration of the role of meteorological dynamics in shaping urban contamination patterns.

Materials and methods

Study area

Istanbul, the largest city in Türkiye, is home to over 15 million people as of 2025, accounting for approximately 19% of the country’s total population. This transcontinental metropolis spans Europe and Asia, connected by the Bosphorus Strait, making it a vital cultural, economic, and industrial hub³². The city’s population density varies significantly, with central urban districts such as Kadıköy, Beşiktaş, and Şişli being highly concentrated, while suburban and peripheral areas are less densely populated but experiencing rapid growth. Urban expansion fueled by migration and economic development has heightened urban stressors, leading to increased construction, industrial activity, and vehicular traffic, which significantly contribute to street dust pollution. Furthermore, Istanbul experiences a transitional climate between the Mediterranean and oceanic types, with some humid subtropical characteristics in specific areas. Winters are mild, with average temperatures ranging from 4 °C to 8 °C in January, while summers are warm to hot, with temperatures averaging between 23 °C and 30 °C in July. The city receives moderate annual precipitation of 850–1,200 mm, predominantly during autumn and winter, while occasional snowfall occurs in winter. Humidity levels are generally high, ranging from 70 to 85%, due to the influence of the surrounding Black Sea and Marmara Sea. Istanbul enjoys approximately 2,500 h of sunshine annually, primarily in summer³³. Street dust samples were collected in February 2024 from 29 locations across Istanbul (Table S1). At each site, five samples were obtained to represent industrial zones, residential areas, and high-traffic regions, ensuring comprehensive coverage of spatial variability. Figure 1, created using ArcGIS Desktop version 10.8.2 (Esri, Redlands, CA, USA), illustrates the spatial distribution of the sampling sites and key geographic features within the study area. All spatial data were processed and visualized following standard cartographic principles to ensure clarity and accuracy.

Fig. 1.

Spatial distribution of the 29 street dust sampling locations across the European and Asian sides of Istanbul, Türkiye.

Sampling protocols and analytical approaches

Once collected, the dust samples were initially treated at 40 °C for 24 h to ensure proper drying. Afterward, they were sieved through a 63 μm mesh to separate finer particles from larger debris. The fine fractions were then stored at 4 °C to maintain their integrity and prevent contamination until further analysis. Sample preparation involved microwave-assisted acid digestion. For each sample, 100.00 mg of dust was placed in Teflon^® vessels and treated with 12 mL of an acidic mixture consisting of concentrated nitric acid (HNO₃) and hydrochloric acid (HCl) in a 3:1 ratio. The digestion was performed under controlled conditions to ensure the complete dissolution of target elements. Following digestion, the solution was diluted to a final volume of 50 mL using ultrapure water, which was used exclusively throughout the process to ensure consistency and eliminate potential contamination³⁴.

The levels of PTEs in street dust subjects were determined using a Shimadzu ICP-MS 2050 (Kyoto, Japan), renowned for its sensitivity and precision in trace metal detection. To achieve reliable and valid analytical outcomes, stringent quality assurance and quality control (QA/QC) protocols were implemented in accordance with ISO/IEC 17,025 standards for laboratory competence and analytical accuracy. Certified reference materials (CRMs) were used to validate the analytical method, and triplicate analyses of each sample were performed to assess precision and reproducibility. The ICP-MS method was confirmed by analyzing NIST 2702 (Gaithersburg, Maryland, USA) six times, evaluating parameters such as the precision, accuracy, recovery and limit of detection. Blank samples and spiked recoveries were included in the workflow to monitor potential contamination and assess method recovery efficiency. Internal standards were also employed to track instrumental consistency and correct for potential signal drift during analysis. Herewith, the elements such as⁷Li (low mass)⁸⁹, Y (medium mass), and ²⁰⁵Tl (high mass) were used, providing optimal signal intensities across the mass range during the ICP-MS analysis. Calibration was performed using multi-element standard solutions, with calibration curves achieving high correlation coefficients (R² > 0.99) throughout the analytical process, ensuring robust and reliable quantification of potentially toxic elements. The results demonstrated satisfactory analytical performance across all evaluated parameters. Recoveries for the quantified PTEs ranged from 96.34% (As) to 100.50% (Cu), reflecting high analytical accuracy and method reliability. The relative bias remained within ± 5% for all elements, indicating minimal systematic error. Moreover, the method exhibited excellent sensitivity, as evidenced by the low limits of detection (LOD) and limits of quantification (LOQ) across all analytes. The calculated uncertainties were within acceptable ranges, and precision was confirmed by low %RSD values (< 2%), demonstrating robust repeatability. Collectively, these results confirm that the applied method is sufficiently accurate, precise, and sensitive for the determination of PTEs in complex environmental matrices such as street dust. Hence, the validation results presented in Table S2 confirm that the analytical procedure is precise and accurate, ensuring the reliability and robustness of the measurements.

Meteorological and air quality data

Meteorological and air pollutant data in Istanbul during the study period (10–20 February 2024) were obtained from the Turkish State Meteorological Service (TSMS) and the National Air Quality Monitoring Network (NAQMN)³⁵. During this period, temperatures ranged from 6 °C to 10 °C, with wind speeds of 4–6 m/s predominantly from the north-northeast, and no precipitation was recorded. These stable conditions likely contributed to the persistence of street dust and reduced variability in pollutant deposition.

As detailed in Table S3, the air quality categories defined by the USEPA Air Quality Index Scale¹⁹ and WHO Air Quality Guidelines¹⁸ are presented. The AQI is a standardized measure used to evaluate the concentration of key air pollutants and their potential impact on human health. The AQI is calculated using the formula (1) established by the U.S. Environmental Protection Agency¹⁹:

1

where AQI_p represents the computed AQI for a specific pollutant, C is the measured pollutant concentration (in µg/m³), C_L and C_H are the lower and upper concentration breakpoints respectively, and I_L and I_H are the corresponding AQI values.

Ecological indices employed for street dust

In this research, certain ecological indices (I_geo, CF, EF, E_rⁱ, PLI, PERI, and NPI) were applied to assess contamination levels, pollution sources, and ecological risks linked with PTEs in urban street dust. The indices were calculated based on the concentrations of selected metals, including Pb, Cd, Zn, Cu, Ni, Cr, and As, using standardized formulas provided in Table S4.

The I_geo was employed to assess the extent of metal enrichment relative to background values by comparing current metal concentrations to pre-industrial reference levels. It classifies contamination into different categories, ranging from uncontaminated to extremely contaminated, helping to determine the severity of metal accumulation. The CF and EF were applied to distinguish anthropogenic influences from natural sources. CF is computed as the ratio of the measured metal concentration to its background value, providing insights into the degree of contamination, while EF normalizes metal concentrations against a reference element (such as Al or Fe) to assess the extent of enrichment due to human activities. The E_rⁱ was conducted to estimate the potential risks posed by individual elements as it quantifies the potential ecological hazard by incorporating metal toxicity factors, offering a risk classification ranging from low to very high ecological risk³⁶.

To assess overall contamination and ecological risks, the PLI, PERI, E_rⁱ, and NPI were computed. PLI provides an integrated assessment of pollution across multiple elements, indicating whether an area is polluted or unpolluted. PERI is a comprehensive index that aggregates individual E_rⁱ values to predict the overall ecological risk of heavy metals in a given environment. NPI standardizes contamination levels across different elements to facilitate comparison³⁷.

Health risk assessment

The health risk assessment followed the guidelines of the U.S. Environmental Protection Agency³⁸ and previous studies^39–44. The assessment estimated non-carcinogenic risks using the hazard quotient (HQ) and hazard index (HI) and evaluated total cancer risk (TCR) for carcinogenic elements. Separate assessments were conducted for children and adults based on age-specific exposure factors. The hazard index (HI) was calculated by summing the HQ values across all metals. The hazard index (HI), which represents the cumulative non-carcinogenic risk for a given metal across all exposure pathways, was calculated using the formulas (2–4):

2

3

4

where ADD is the average daily dose (mg/kg-day) and RfD is the reference dose (mg/kg-day). The following formulas (5, 6) were used to calculate ADD_ingestion, ADD_inhalation, and ADD_dermal.

5

where C represents the metal concentration in dust (mg/kg), IR_ing is the dust ingestion rate (200 mg/day for children and 100 mg/day for adults), EF is the exposure frequency (350 days/year), ED is the exposure duration (6 years for children and 24 years for adults), BW is the body weight (15 kg for children and 70 kg for adults), and AT is the average time for non-cancer risk (2190 days for children and 8760 days for adults). For inhalation exposure, the ADD was calculated using the formula (6) below:

6

where IR_inh is the dust inhalation rate (7.6 m³/day for children and 20 m³/day for adults), and PEF is the particle emission factor (1.36 × 10⁹ m³/kg). For dermal contact, the ADD was determined using formula (7) below:

7

where SA is the exposed skin area (2800 cm² for children and 5700 cm² for adults), AF is the skin adherence factor (0.2 mg/cm²-day for children and 0.07 mg/cm²-day for adults), and ABS is the dermal absorption factor (0.001, unitless).

Carcinogenic risk was assessed using the cancer risk (CR) formula (8):

8

where CSF is the cancer slope factor (kg-day/mg). The total cancer risk (TCR) was calculated by summing the cancer risks from ingestion, inhalation, and dermal contact:

9

The average time for cancer risk assessment (AT_cancer) was set at 25,550 days for both children and adults⁴⁵. The exposure parameters and their probability distributions are provided in Table S5, while Table S6 presents the reference dose (RfD) and cancer slope factors (CSF) used in the assessment. The metal concentrations were assumed to follow a lognormal distribution, while exposure factors followed triangular, uniform, normal, and beta-PERT distributions. This methodology provides a comprehensive risk evaluation, allowing for the assessment of potential health hazards due to heavy metal exposure in urban dust.

Source identification and statistical analysis

To analyze the relationships among PTEs in street dust collected from Istanbul, Türkiye, statistical evaluations were performed using OriginPro 2022 and SPSS^® version 27. Pearson correlation coefficients PCC helped determine the strength of pairwise associations between elements, while HCA was applied to group elements exhibiting similar behaviors or origins. Additionally, PCA was utilized to identify key influencing factors and distinguish possible contamination sources. To further investigate the origins of PTEs in street dust, PMF analysis was employed using the US EPA PMF 5.0 software. The analysis was based on elemental concentration data, and the extracted factor profiles provided insights into contamination sources across the study area. In addition, a Monte Carlo Simulation (MCS) was carried out using Crystal Ball software (Oracle, USA) to estimate the potential health risks linked with human exposure to PTEs in street dust. Model performance and reliability were validated through diagnostic tools integrated within the software, applying a 95% confidence interval and 10,000 iterations. The health risks linked to PTEs exposure via ingestion, inhalation, and dermal routes of administration were assessed using a Monte Carlo Simulation conducted with Crystal Ball software (Oracle, USA). A comprehensive overview of the distribution settings values and input parameters used in the MCS-based health risk assessment is presented in Table S5.

Results and discussion

PTE concentrations

This study investigated the levels of PTEs in urban street dust samples collected from 29 locations across Istanbul, providing insights into their spatial distribution and potential environmental and health risks (Table S7). The descriptive statistics for the analyzed elements are summarized in Table S8. The elemental concentrations (mg/kg) were quantified in the following increasing order: Cd (0.9) < Co (10.9) < As (11.8) < Ni (63.1) < Pb (67.8) < Cr (135.1) < Cu (333.3) < Mn (459.2) < Zn (477.2) < Fe (27,252) < Al (29,916).

In this study, the mean concentrations of Al and Fe were the highest among all elements, reflecting their natural abundance in the Earth’s crust. These elements exhibited relatively low variability, suggesting a dominant geogenic origin influenced by local soil composition⁴⁶. In contrast, elements such as Cr, Cu, Zn, and Pb displayed considerable variability, indicating mixed sources likely from anthropogenic activities, including industrial emissions, vehicular wear, and urban pollution⁴⁷. Similarly, Co and Ni exhibited localized concentration elevations, indicating potential industrial contributions, particularly from activities such as metal processing, electroplating, and combustion-related emissions⁴⁸. The presence of As and Cd at moderate levels necessitates further investigation due to their toxicological significance.

The spatial distribution maps of PTEs across Istanbul reveal significant heterogeneity, reflecting diverse pollution sources and environmental influences (Fig. 2). Al and Fe exhibit relatively uniform distributions, with slightly higher concentrations in northern and central areas, consistent with their geogenic origin⁴⁶. Elevated concentrations of Ni and Cr were observed in the northern European side of the study region and along major transportation corridors, indicating contributions from industrial activities and vehicular sources. Existing literature supports the finding that vehicular traffic is a significant source of these metals in urban environments. A study in Toronto found that Ni levels were higher in sweepings from major roads than in urban soils, indicating a vehicular emission source. Additionally, Cr and Ni concentrations increased with road hierarchy, peaking on high-traffic routes⁴⁹. Manganese exhibited a more dispersed spatial distribution, suggesting the influence of both natural soil composition and anthropogenic contributions. Mn is naturally abundant in the Earth’s crust, yet its dispersed distribution in the study area suggests mixed geogenic and anthropogenic origins. Industrial emissions, vehicular traffic, and Mn-based fuel additives contribute to elevated levels in urban environments, particularly through steel manufacturing and atmospheric deposition⁵⁰. Co displayed localized hotspots on the European side, potentially indicating site-specific industrial activities or areas affected by historical contamination sources, such as past metallurgical operations or waste disposal practices⁵¹. Cu and Zn exhibited elevated concentrations in the central and northern urban areas, closely associated with traffic-related emissions, including brake and tire wear, as well as urban runoff. These patterns are consistent with their widespread use in automotive components and their known mobility in urban environments⁵². As and Cd exhibited moderate concentrations with localized peaks, indicating the presence of point sources, potentially related to industrial discharges, waste disposal sites, or legacy contamination from metallurgical activities and urban landfills^14,53. Pb distribution reveals pronounced hotspots in densely populated urban areas, indicating the impact of historical pollution from formerly used leaded gasoline, along with ongoing contributions from industrial emissions and lead-containing construction materials. These legacy sources have resulted in long-term accumulation of lead in urban soils despite regulatory phase-outs⁵⁴. Hg exhibited significant spatial variability, with localized hotspots likely associated with industrial sources, such as combustion processes, waste incineration, or historical contamination. These patterns stress the importance of implementing targeted pollution management strategies to mitigate mercury-related environmental and health risks⁵⁵.

Fig. 2.

Spatial distribution maps of PTEs across sampling stations in Istanbul, Türkiye.

As can be evidenced in Table 1, the content of PTEs in Istanbul’s street dust are generally moderate and comparable to those reported in other global urban areas. However, certain elements, particularly Cu (333.3 mg/kg), Zn (477.2 mg/kg), and Cr (135.1 mg/kg), exhibited elevated concentrations, suggesting the influence of anthropogenic activities such as traffic emissions and industrial processes. Copper and zinc, which are strongly associated with vehicular brake and tire wear, showed notably higher concentrations in Istanbul compared to cities like Abbottabad (Cu: 50.0, Zn: 139.0 mg/kg)⁵⁶ and Ankara (Cu: 35.7, Zn: 0.9 mg/kg)⁵⁷. Istanbul’s Cu level is comparable to Moscow (380.0 mg/kg)¹⁴ and exceeded values reported in highly urbanized regions such as Jiaozuo, China (Cu: 49.9 mg/kg)⁵⁸. The Zn concentration in Istanbul is also comparable to Taiwan (500.4 mg/kg)⁶² and Jinan (492.8 mg/kg)⁶⁸, indicating a significant contribution from traffic-related sources. Chromium concentrations in Istanbul were relatively high (135.1 mg/kg), aligning closely with levels in Jiaozuo (112.1 mg/kg) and Xiangtan (115.6 mg/kg), but lower than those reported in Kermanshah (181 mg/kg)³ and Chhattisgarh (833 mg/kg)⁶⁶, where heavy industrial activity is prevalent. The elevated Cr levels may be attributed to metal alloy wear, industrial discharges, or construction-related activities. Lead (Pb: 67.8 mg/kg) showed moderate enrichment, consistent with findings from Al-Hillah (64.6 mg/kg)² and higher than those in Warsaw (17.1 mg/kg)⁶⁹, suggesting residual contamination from historical leaded fuel use and urban infrastructure. Nickel levels (63.1 mg/kg) in Istanbul also suggest mixed geogenic and anthropogenic origins, with values similar to those in Taiwan (64.1 mg/kg)⁶² and slightly lower than in Al-Hillah (75.4 mg/kg)², yet elevated compared to Ankara (25.9 mg/kg)⁵⁷ and Abbottabad (10.3 mg/kg)⁵⁶. This distribution pattern implies possible influence from fossil fuel combustion, industrial sources, and soil background. Cadmium (0.9 mg/kg) and arsenic (11.8 mg/kg) concentrations were within ranges reported globally, with Cd comparable to levels found in Jiaozuo (1.3 mg/kg)⁵⁸ and Xiangtan (0.4 mg/kg)¹⁵, and As matching those reported in Al-Hillah (11.8 mg/kg)² and Kerman (11.0 mg/kg)⁷². These findings indicate localized enrichment, possibly from point sources such as waste disposal or industrial emissions.

Table 1.

Comparison of the results obtained in this study with previously reported values in the global literature.

Location	As	Cd	Cr	Cu	Mn	Fe	Co	Ni	Pb	Zn	References
Istanbul, Türkiye	11.8	0.9	135.1	333.3	459.2	27,252	10.9	63.1	67.8	477.2	This study
Abbottabad, Pakistan	-	0.2	13	50.0	304	15,540	6.7	10.3	21.5	139	56
Ankara, Türkiye	-	3.8	-	35.7	-	-	18	25.9	48.1	0.9	57
Jiaozuo, China	23.1	1.3	112.1	49.9	373.8	-	17.9	51.7	55.3	374.3	58
Khulna City, Bangladesh	3.0	-	67.5	-	386	25,648	8.9	-	-	98	5
Xiangtan, China	-	0.4	115.6	91.3	-	-	-	-	549.6	536.8	15
Moscow, Russia	1.8	1.1	91	380	945	46,506	22	61	133	1389	14
Zhengzhou, China	-	0.6	41.9	21	515	-	2.8	8.3	44.2	151.0	59
Czeck Republic	11	6.0	173	-	914	36,692	13	182	-	1023	60
Ilam, Iran	5	0.4	46	63	-	-	-	44	52	213	61
Kermanshah, Iran	2.4	0.1	181	4.62	12.9	-	4.62	87.3	64.9	134	3
Taiwan	-	-	67.7	182.1	304.8	20,586	8.1	64.1	57.4	500.4	62
Recife, Brazil	-	1.3	56.5	111.1	315.6	-	-	12.1	37.4	154.4	63
Al-Hillah, Iraq	11.8	0.2	300	57.3	480	25,600	12.1	75.4	64.6	140	2
Lublin, Poland	-	5.5	112.0	120.6	-	-	-	17.1	46.6	296.2	64
Monterrey, Mexico	-	0.4	-	-	-	-	-	-	653.0	52.5	65
Chhattisgarh, India	-	1.3	833.0	152.0	274.0	-	-	571.0	98.6	231.0	66
Bosnia & Herzegovina	-	3.2	33.2	30.0	236.0	-	-	73.0	52.5	81.7	67
Jinan, China	-	1.1	114.1	87.7	517.0	-	8.2	30.3	80.3	492.8	68
Warsaw, Poland	-	0.2	20.3	184.3	110.1	-	1.8	17.0	17.1	149.9	69
Jiaozuo, China	23.1	1.3	112.1	49.9	473.8	-	25.3	51.7	55.3	374.3	58
Izmir, Türkiye	13.9	0.2	39.4	22.4	274.1	13,303	5.2	19.4	26.9	61.6	70
Spain	-	0.3	29.0	173.3	-	-	-	41.3	109.0	258.3	71
Kerman, Iran	11.0	0.3	28.7	71.1	417.0	8.57	-	20.6	45.6	170.0	72

Collectively, the elemental profiles observed in Istanbul’s street dust indicate a multifactorial origin, shaped by a combination of traffic-related emissions, industrial activities, residual historical contamination, and natural geochemical background levels. The spatial distribution and elevated concentrations of key elements, particularly Cu, Zn, and Cr, highlight the necessity for sustained environmental monitoring, as their levels frequently exceeded or closely approached the upper ranges reported in comparable urban environments worldwide. In comparison with other metropolitan areas, Istanbul exhibits a distinctly heterogeneous pattern of PTE distribution, reflecting the concurrent influence of both geogenic and anthropogenic sources.

Air quality

Air pollution is a significant environmental concern affecting public health, ecosystems, and urban sustainability. Monitoring key air pollutants provides crucial insights into their concentration levels, sources, and potential impacts⁷³. This study evaluates the air quality in Istanbul during February 2024, using key atmospheric pollutants: NO, NO₂, NOx, PM_2.5, PM₁₀, and SO₂. Descriptive statistics due to these parameters are provided in Table 2.

Table 2.

Descriptive statistics of air quality parameters in istanbul, Türkiye.

Parameter	Min	Max	Median	Mean	Std Dev
NO	0.20	263.30	22.61	23.21	15.00
NO2	2.50	204.80	34.89	37.49	12.98
NOX	7.88	934.61	73.01	83.81	46.56
PM 2.5	0.00	64.00	9.61	9.61	4.52
PM10	0.00	176.91	21.28	24.19	9.25
SO2	0.08	15.62	3.91	3.49	1.59

Values are provided in µg/m³.

The interpretation of the air quality data for February 2024 in Istanbul reveals notable trends among the invesitigated pollutants. Nitrogen oxides (NO, NO₂, NOx) exhibited significant variability, with NOx reaching a maximum concentration of 934.61 µg/m³ and a mean of 83.81 µg/m³, suggesting substantial contributions from traffic emissions and industrial activities. The high standard deviation further indicates fluctuations due to varying traffic density and meteorological influences. Particulate matter (PM_2.5 and PM₁₀) levels remained relatively moderate, with PM_2.5 showing a low mean concentration of 9.61 µg/m³, while PM₁₀ displayed greater variability, peaking at 176.91 µg/m³, likely due to localized dust resuspension and construction activities. SO₂ concentrations remained low, with a mean of 3.49 µg/m³, suggesting minimal influence from coal combustion or heavy industrial sources. These findings indicate that while air quality was generally within acceptable limits, elevated NOx levels may pose a concern for respiratory health and ozone formation, particularly in high-traffic areas. Additionally, occasional PM₁₀ peaks warrant further investigation to mitigate exposure risks, especially for vulnerable populations. Overall, targeted measures, including improved traffic emission controls and dust management strategies, could further enhance Istanbul’s air quality.

As evidenced in Table 3, the overall AQI value calculated for Istanbul is 40, which falls under the “Good” (0–50) category according to the U.S. Environmental Protection Agency¹⁹ and World Health Organization guidelines¹⁸. This suggests that the air quality is satisfactory and that air pollution poses minimal or no risk to the general population⁷⁴. Among the individual pollutants, PM₂.₅ (AQI = 40) was identified as the primary contributing pollutant, which aligns with global concerns over fine particulate matter as a major air quality indicator. Despite being within the “Good” category, PM_2.5 should be continuously monitored due to its long-term health effects. PM₁₀ (AQI = 22) remains well below hazardous levels, indicating that coarse particulate pollution is not a significant issue in this assessment. NO₂ (AQI = 35), primarily from traffic emissions, is also within the acceptable range¹⁹ implying that vehicular pollution is not currently at dangerous levels. SO₂ (AQI = 5) is extremely low, suggesting minimal industrial or fossil fuel-related emissions affecting the air quality.

Table 3.

AQI results with USEPA Thresholds.

Pollutant	Measured AQI	USEPA Category	EPA description
PM2.5	40	Good	Air quality is deemed acceptable, with pollutant levels posing minimal or no health risk19.
PM10	22	Good
NO2	35	Good
SO2	5	Good
Overall AQI	40	Good

Although air quality data were obtained from 10 different monitoring stations of the NAQMN, most of which are located near the street dust sampling points, no direct air sampling was conducted at the same locations. Nevertheless, the AQI values, particularly those associated with PM₁₀ and PM_2.5, provide context for atmospheric conditions that may influence the deposition of PTEs in street dust. The moderate levels of PM₁₀, with occasional peaks, suggest that particulate matter from traffic and industrial sources could contribute to PTE accumulation in dust through atmospheric deposition. This potential relationship highlights the interaction between air quality and urban dust contamination, warranting further integrated investigations. These findings indicate that Istanbul’s air quality in the study period was safe, with PM_2.5 as the most notable pollutant. Given its potential impact on human health, efforts to reduce emissions from combustion sources (e.g., vehicle regulations, industrial emission controls) should continue to ensure sustained air quality improvements. Moreover, long-term exposure assessments should be conducted to determine seasonal and spatial variations in pollution levels⁷⁵.

It should be noted that AQI data in this study was used solely as a complementary aid to contextualize ambient air conditions during the dust sampling period. Due to the spatial separation between monitoring stations and dust sampling sites, direct correlation analysis or spatial co-mapping was not feasible. However, a general consistency was observed between higher PM₁₀ levels and traffic-heavy zones with elevated dust concentrations of Cu, Zn, and Pb, suggesting a potential link through common emission sources. The interpretation of air quality and inhalation-related exposure risks in this study is informed by both WHO (2021)¹⁸ and USEPA (2016)¹⁹ thresholds, as summarized in Table S3.

Ecological investigation

Ecological health was evaluated employing a comprehensive set of indices, including I_geo, CF, EF, E_rⁱ, PLI, PERI, and NPI.

The geo-accumulation index is a widely used tool to evaluate the contamination levels of PTEs in environmental samples, such as street dust and sediments. It compares current concentrations to pre-industrial baseline values, accounting for natural variations³⁴. In this study, I_geo analysis revealed significant spatial variability of PTEs across 29 sampling sites in Istanbul. Cu exhibited the highest contamination levels, particularly at sites S15 and S19, indicating moderate to heavy pollution likely originating from anthropogenic sources, including traffic-related activities and industrial runoff. Zn and Pb also exhibited moderate contamination, reflecting urban influences. Conversely, Al, Fe, Mn, and Cr consistently showed low I_geo values, suggesting natural background levels. As evidenced in Fig. 3, The violin plots further highlighted site-specific variability, with wider distributions in S15 and S19 indicating localized contamination events, while narrower distributions in S3, S4, and S26 suggested more uniform contamination patterns.

Fig. 3.

Heatmap and violin plot illustrating the spatial distribution of I_geo across sampling stations.

The CF is used to evaluate the extent of contamination by comparing the concentration of PTEs in environmental samples to background values⁴⁴. Herewith, CF analysis revealed significant spatial variability in street dust across 29 sampling sites in Istanbul. Zn showed the highest CF values, particularly at sites S14 and S19, indicating considerable to very high contamination, likely originating from human activities such as vehicular traffic and industrial emissions. In contrast, Pb, Cd, As, Cu, Ni, Co, Fe, Mn, Cr, and Al exhibited consistently low CF values, suggesting natural background levels with minimal anthropogenic influence. As illustrated in Fig. 4, The violin plots further illustrated site-specific variability, with wider distributions at S14 and S19 indicating localized contamination events, while narrower distributions at S3, S4, S26, and S29 suggested more uniform contamination patterns.

Fig. 4.

Heatmap and violin plot illustrating the spatial distribution of CF across sampling stations.

The enrichment factor is applied to evaluate the degree of anthropogenic influence on PTEs by normalizing the concentration of each element against a reference element, typically Al or Fe. EF values are categorized as minimal enrichment (EF < 2), moderate enrichment (2 ≤ EF < 5), significant enrichment (5 ≤ EF < 20), very high enrichment (20 ≤ EF < 40), and extremely high enrichment (EF ≥ 40)⁷⁶. In this study, EF analysis revealed that Zn exhibited the highest enrichment, particularly at sites S22, S24, and S26, indicating significant to very high enrichment levels likely due to anthropogenic sources such as vehicular traffic and industrial runoff. In contrast, Pb, Cd, As, Cu, Ni, Co, Fe, Mn, Cr, and Al showed minimal enrichment, suggesting natural background levels with negligible anthropogenic influence. As can be seen in Fig. 5, The violin plots illustrated site-specific variability, with wider distributions at S22, S24, and S26 indicating localized contamination events, whereas narrower distributions at other sites suggested more uniform patterns.

Fig. 5.

Heatmap and violin plot illustrating the spatial distribution of EF across sampling stations.

The ecological risk index (E_rⁱ) is used to assess the potential ecological risks posed by PTEs by integrating contamination levels with toxic response factors⁷⁷. In this study, E_rⁱ values for Cr, Mn, Ni, Cu, Zn, As, Cd, and Pb were evaluated across 29 sampling sites in Istanbul (Fig. 6). The results revealed that Cd consistently exhibited the highest E_rⁱ values across all sites, particularly at S25, S22, and S16, indicating considerable to high ecological risks likely due to industrial and vehicular emissions. Cu also showed elevated E_rⁱ values, particularly at sites S22 and S23, reflecting localized contamination likely linked to urban activities. In contrast, Mn, Zn, Cr, and Ni displayed relatively lower E_rⁱ values, suggesting minimal ecological risks.

Fig. 6.

Heatmap and violin plot illustrating the spatial distribution of E_rⁱ across sampling stations.

Thus, the findings derived from I_geo, CF, EF, and E_rⁱ indicate dominant influence of anthropogenic emissions to Zn and Cd contamination, emphasizing the need for targeted pollution control in high-enrichment urban hotspots to mitigate associated ecological and health risks.

In addition, the PLI, PERI, NPI, and NRI are comprehensive indices used to evaluate the environmental pollution status and ecological risks of PTEs⁷⁸. In this study, spatial distribution maps of these indices were generated for the study area in Istanbul (Figure S1). The results indicate that the western region exhibits higher pollution and ecological risks, particularly around industrial and urban areas. High PLI and NPI values suggest significant pollution loads, mainly due to human activities known as vehicular traffic and industrial emissions⁷⁹.

The elevated PERI values in these areas highlight potential ecological risks, emphasizing the need for targeted pollution management strategies. Conversely, the eastern region shows relatively lower pollution levels, reflecting a more natural background state. These findings provide crucial insights into spatial pollution patterns and can guide environmental protection and management policies.

Source identification

The identification of sources contributing to the contamination of street dust in Istanbul was performed using PMF, PCA, PCC, polar HCA. These methods collectively provide insights into the possible sources of PTEs, allowing for a more comprehensive understanding of the major contributors⁴.

The performance of the PMF model was evaluated using standard diagnostic tools, including the Q-robust (15,328.2) and Q-true (15,742.5) statistics, as well as the distribution of residuals, which were symmetrically centered around zero for most metals, indicating acceptable model fit. The total R² between observed and predicted concentrations exceeded 0.85 for key elements such as Cu, Zn, Pb, and Cr, supporting the validity of the extracted factor profiles. As can be seen in Fig. 7a, The PMF model identified three major sources contributing to PTE contamination in street dust: industrial runoff (39.4%), traffic emissions (31.3%), and natural and soil sources (29.4%). As can be seen in Fig. 7b, the source profiles derived from PMF indicate that traffic emissions are characterized by high loadings of Cu (76.8%), Zn (62.6%), and Pb (58.2%), which are typically associated with vehicular sources such as brake and tire wear, engine wear, and fossil fuel combustion⁸⁰. Industrial runoff shows a significant contribution from Cr (81.7%), Ni (84.3%), and Co (41.4%), suggesting emissions from industrial activities, particularly metal processing, electroplating, and manufacturing⁸¹. In addition, percent contribution of each metal is illustrated in Fig. 7c. As a result, natural and soil sources are dominated by Fe (44.5%), Al (54.7%), Mn (46.4%), and As (47.0%), indicating contributions from geological background sources and soil resuspension⁸². These findings suggest that while industrial activities remain a significant contributor to PTE contamination, vehicular emissions also play a substantial role in metal accumulation in street dust. As shown in Figure S2, The rotated component matrix from PCA provides further insights into the association of PTEs with potential sources. PC1, explaining 52.4% of the total variance, shows strong loadings of Cr (0.959), Ni (0.956), Co (0.732), and As (0.726), confirming an industrial origin, particularly from metal-processing industries. PC2, accounting for 18.1% of the variance, has high correlations for Cu (0.904), Zn (0.903), and Pb (0.890), indicating vehicular emissions as the dominant contributor. PC3, explaining 10.5% of the variance, includes Mn (0.829), Al (0.683), and Fe (0.593), suggesting contributions from natural sources such as soil and rock weathering⁸³. The correlation heatmap highlights significant relationships between specific PTEs, further confirming the source identification. Strong correlations between Cr, Ni, and Co (r > 0.9, p < 0.01) suggest a common industrial source⁸⁴ while high positive correlations between Zn, Cu, and Pb (r > 0.8, p < 0.01) reinforce their association with traffic-related emissions. Fe, Al, and Mn show moderate correlations, supporting their natural geogenic origin⁸⁵.

Fig. 7.

Graphical demonstration of PMF results (a), percent of contribution of factors (b).

Cluster analysis further validates the identified sources by grouping similar elements together. The results indicate that Cr, Ni, and Co are primarily associated with industrial emissions, Zn, Cu, and Pb are closely linked to vehicular emissions, and Al, Mn, Fe, and As are indicative of natural soil-derived sources (Fig. 8). The dominance of industrial runoff as the primary contributor to PTE contamination suggests the need for stricter environmental regulations targeting industrial waste management⁸⁶. The substantial influence of traffic emissions stresses the necessity for improved urban air quality measures, such as promoting public transportation and cleaner vehicle technologies. Additionally, the contribution of natural sources highlights the importance of controlling dust resuspension through urban greening initiatives⁸⁷. The integrated approach utilizing PMF, PCA, PCC, and cluster analysis has provided a comprehensive understanding of the sources of PTEs in Istanbul’s street dust. The findings emphasize the need for targeted mitigation strategies addressing industrial, vehicular, and natural contributions to urban metal contamination.

Fig. 8.

Graphical demonstration of PCC (a), and polar heatmap HCA (b).

Additionaly, the spatial variability of air quality parameters, especially PM₁₀, aligns with regions showing higher concentrations of traffic-related metals such as Cu, Zn, and Pb. The proximity of air quality monitoring stations to dust sampling sites and the moderate AQI levels observed suggest that airborne particulates may play a supplementary role in the deposition of these elements. Although the overall AQI was classified as ‘Good’, particulate matter remains a critical vector for metal transport. Previous studies²⁹ have shown that road transport significantly contributes to particulate matter and NO₂ emissions in urban environments, further supporting the notion that vehicular activities influence both air quality and metal accumulation in street dust.

Health risk assessment

The noncarcinogenic health risks linked with exposure to PTEs in street dust were evaluated using HQ and the HI. HQ values were computed for ingestion, dermal contact, and inhalation pathways, with the HI representing the sum of these exposures for each element⁸⁸. The results indicate that for adults, the THI is 0.345, which remains below the acceptable risk threshold of 1.0, suggesting that noncarcinogenic risks are not significant for this group. However, for children, the THI is 3.13, exceeding the acceptable limit, indicating potential adverse health effects due to exposure to street dust (Table 4).

Table 4.

Estimated non-carcinogenic risk values for adults and children based on exposure ptes.

	Noncarcinogenic risks for adult				Noncarcinogenic risks for child
	HQ ing	HQderm	HQ inh	HI	HQ ing	HQderm	HQ inh	HI
Al	4.09E-02	8.16E-04	1.20E-03	4.29E-02	3.82E-01	5.35E-03	2.13E-03	3.89E-01
Cr	6.11E-02	9.76E-03	2.70E-04	7.12E-02	5.71E-01	6.39E-02	4.78E-04	6.35E-01
Mn	2.63E-02	2.62E-03	1.86E-03	3.08E-02	2.45E-01	1.72E-02	3.29E-03	2.66E-01
Co	5.00E-02	9.98E-04	3.68E-04	5.14E-02	4.67E-01	6.54E-03	6.52E-04	4.74E-01
Ni	4.28E-03	4.27E-04	1.40E-04	4.85E-03	4.00E-02	2.80E-03	2.48E-04	4.30E-02
Cu	1.13E-02	2.25E-04	1.65E-06	1.15E-02	1.05E-01	1.47E-03	2.92E-06	1.07E-01
Zn	2.16E-03	4.30E-05	3.17E-07	2.20E-03	2.01E-02	2.82E-04	5.62E-07	2.04E-02
As	5.42E-02	6.83E-03	1.60E-04	6.12E-02	5.06E-01	4.48E-02	2.83E-04	5.51E-01
Cd	2.55E-03	2.04E-04	1.88E-05	2.77E-03	2.38E-02	1.33E-03	3.33E-05	2.52E-02
Pb	6.57E-02	8.73E-04	3.86E-06	6.65E-02	6.13E-01	5.72E-03	6.85E-06	6.18E-01
			THI	3.45E-01			THI	3.13E + 00

Among the analyzed elements, Pb presents the highest noncarcinogenic risk for children, with an HI value of 0.618, followed by As (0.551) and Cr (0.635). These elements are known to cause neurological, developmental, and systemic health effects, particularly in children, who are more vulnerable due to higher ingestion rates and lower body weight⁹. For adults, Pb and As also show the highest HI values at 0.0665 and 0.0612, respectively, but remain below the risk threshold. The elevated HQ_ing values for Pb, Cr, and As indicate that ingestion is the primary exposure route contributing to health risks, which is consistent with previous studies highlighting the ingestion of contaminated dust as a major pathway for metal exposure^3,5,15,59. Nickel, zinc, and cadmium exhibit the lowest HI values for both age groups, suggesting minimal noncarcinogenic risks from these elements. However, manganese, cobalt, and aluminum contribute moderately to the total HI, particularly in children.

The Monte Carlo simulation results indicate a considerable non-carcinogenic health risk associated with exposure to PTEs in street dust, particularly for children. The HI values for certain metals, including Al, Mn, Cr, Ni, Cd, Cu, and Pb, exceed the safe threshold (HI > 1), suggesting potential adverse health effects. Children exhibit significantly higher HI values than adults, which is expected due to their increased dust ingestion rates, lower body weight, and heightened physiological susceptibility⁸⁹. Notably, Cr and Mn demonstrate the highest HI values, with their 95th percentile levels exceeding the acceptable range, indicating potential neurotoxic and systemic health risks^90,91. Furthermore, the cumulative non-carcinogenic hazard index further emphasizes that a portion of the exposed population may experience health effects, warranting immediate attention (Fig. 9). The moderate temperatures and wind conditions, along with the absence of rainfall during the sampling period, favored the accumulation and limited dispersion of street dust contaminants. The consistent meteorological conditions likely enhanced the representativeness of the sampled PTE concentrations, as dust resuspension was not heavily influenced by sudden weather shifts. The elevated PM₁₀ values in some locations further support the role of dry conditions in particulate mobilization.

Fig. 9.

Visualization of human health risk distributions derived from Monte Carlo simulation.

The carcinogenic risks related with exposure to PTEs in street dust were evaluated for ingestion (CR_ing), dermal contact (CR_derm), and inhalation (CR_inh) exposure pathways. The total carcinogenic risk (TCR) represents the cumulative risk from all pathways for each element. The results indicate that for adults, the TCR is 9.34 × 10⁻⁵, while for children, the TCR is 2.16 × 10⁻⁴. Both values exceed the acceptable carcinogenic risk threshold of 1 × 10⁻⁶, indicating potential long-term health concerns by reason of chronic exposure to carcinogenic metals in street dust. The risk level for children was found to be more than twice that of adults, reflecting their greater vulnerability stemming from higher ingestion rates, frequent hand-to-mouth activity, and lower body weight⁹² which collectively increase their susceptibility to PTE exposure (Table 5).

Table 5.

Estimated carcinogenic risk values for adults and children based on exposure ptes.

	Carcinogenic risks for adult				Carcinogenic risks for child
	CR ing	CR derm	CR inh	CR	CR ing	CR derm	CR inh	CR
Cr	3.14E-05	5.02E-07	3.88E-08	3.20E-05	7.34E-05	8.22E-07	1.72E-08	7.42E-05
Ni	4.99E-05	4.98E-07	3.63E-09	5.04E-05	1.17E-04	8.16E-07	1.61E-09	1.17E-04
As	8.37E-06	2.44E-06	1.24E-09	1.08E-05	1.95E-05	4.00E-06	5.49E-10	2.35E-05
Cd	1.66E-07	1.10E-08	4.05E-10	1.78E-07	3.88E-07	1.80E-10	1.80E-10	4.06E-07
Pb	2.68E-08	5.28E-09	1.95E-10	3.23E-08	6.25E-08	8.65E-09	8.63E-11	7.12E-08
			TCR	9.34E-05			TCR	2.16E-04

Among the analyzed elements, nickel and chromium contribute the most to carcinogenic risk, highlighting their potential threat to long-term human health. For adults, the highest carcinogenic risk is associated with Ni (5.04 × 10⁻⁵), followed by Cr (3.20 × 10⁻⁵) and As (1.08 × 10⁻⁵). A similar trend is observed in children, where Ni (1.17 × 10⁻⁴) and Cr (7.42 × 10⁻⁵) present the highest risks, with As (2.35 × 10⁻⁵) contributing moderately. Although Ni and Cr contributed most to the estimated TCR values, it is critical to recognize that the carcinogenic risk related with nickel is primarily linked to inhalation exposure. According to recent EFSA evaluations⁹³ there is insufficient evidence to classify nickel as carcinogenic via the oral route. Therefore, while the modeled TCR values for Ni appear elevated, they do not necessarily represent a realistic cancer risk through ingestion. The elevated risk levels for these elements generally align with their known carcinogenic profiles, with Ni and Cr linked to lung and nasal cancers⁸and As associated with bladder, lung, and skin cancers^94,95. The ingestion pathway (CR_ing) is the predominant exposure route for carcinogenic risk, suggesting that accidental ingestion of contaminated dust poses the greatest health concern.

The carcinogenic risks associated with cadmium and lead are relatively low in comparison but still contribute to the overall risk profile. In other words, the health risks are negligible³⁷ with TCR values well below the acceptable threshold of 1 × 10⁻⁶. Specifically, Cd poses a minimal risk, with TCRs of 1.78 × 10⁻⁷ for adults and 4.06 × 10⁻⁷ for children, while Pb shows even lower values of 3.23 × 10⁻⁸ and 7.12 × 10⁻⁸, respectively. Despite the low risk estimates, the toxicological relevance of these metals should not be overlooked. Cadmium is a recognized nephrotoxin and carcinogen, with a long biological half-life and bioaccumulative potential⁹⁶ while lead is associated with neurological, haematological, and renal toxicity, particularly in children⁹⁷. These results support the continued need for environmental surveillance and exposure minimization plans in urban ecosystems.

To more accurately characterize uncertainties in health risk estimates, MCS was applied. This stochastic modeling approach is commonly used in environmental health risk assessments to account for the uncertainty and variability inherent in input parameters². In contrast to deterministic approaches that yield single-point estimates, MCS generates a probabilistic distribution of potential outcomes by repeatedly sampling from defined probability distributions assigned to each variable, such as metal concentrations, exposure duration, ingestion rate, and body weight⁹⁸. This probabilistic framework provides a more comprehensive and realistic evaluation of health risks under variable exposure scenarios, thereby enhancing the robustness of risk characterization⁹⁹. Further refinement of risk assessment has been conducted through Monte Carlo simulation. Herewith, the probabilistic analysis of carcinogenic risk highlights significant concerns regarding exposure to arsenic, chromium, cadmium, and lead through street dust (Fig. 9). While the mean TCR remains within the acceptable limit of 1 × 10⁻⁴ for both adults and children, the 95th percentile estimates indicate that a fraction of the population, particularly children, is exposed to risk levels exceeding this threshold. Among the analyzed elements, Cr poses the highest carcinogenic threat, followed by As, with their upper-bound risk estimates exceeding the acceptable range for children. The high mobility and bioavailability of these metals in urban environments further exacerbate their potential impact¹⁰⁰. The cumulative nature of these exposures suggests that prolonged contact with contaminated street dust may contribute to long-term health implications⁶³. To mitigate these risks, implementing strict emission control measures, enhancing street cleaning practices, and enforcing stringent industrial regulations are crucial steps toward reducing human exposure to carcinogenic metals in urban environments.

Source contribution to health risks: linking pollution origins to human exposure

Sankey diagrams serve as effective visualization tools for representing the proportional contributions of pollution sources to environmental contaminants and their associated health risks. In urban environments, they facilitate an extensive characterization of the relationships between emission sources and the distribution of PTEs in various environmental media¹⁰¹.

Herewith, the Sankey diagram (Fig. 10) illustrates the proportional contributions of traffic emissions, natural/soil sources, and industrial runoff to the existence of PTEs in urban street dust and their subsequent impact on human health risks. Traffic emissions are identified as the dominant contributor to chromium (81.7%), copper (76.8%), lead (58.2%), and zinc (62.6%), highlighting vehicular activities such as fuel combustion, brake wear, and tire abrasion as key sources of metal contamination⁴⁹. Natural and soil sources contribute significantly to arsenic (47%), aluminium (29.7%), and nickel (24.4%), suggesting that geological composition and soil resuspension play a major role in metal dispersion⁸². Meanwhile, industrial runoff is a notable source of cadmium (39.9%), manganese (46.4%), cobalt (41.4%), and iron (28.2%), reflecting the influence of industrial discharges and wastewater contamination⁸¹.

Fig. 10.

Sankey diagram demonstrating the flow from pollution sources to metal contaminants and their corresponding health risks.

From a health risk perspective, the THI results of this research indicate that traffic emissions play a substantial role in cumulative exposure, with children exhibiting slightly higher risk levels (20.29%) compared to adults (19.26%). This trend aligns with children’s increased susceptibility because of higher dust ingestion rates and lower body weight¹⁰². The TCR further highlights the severity of industrial runoff, particularly in its contribution to cadmium and arsenic exposure. Notably, children face a significantly higher cancer risk (54.44%) compared to adults (34.23%), emphasizing their heightened vulnerability to long-term exposure effects⁴². Given these findings, targeted mitigation strategies are essential, including stricter vehicular emission controls, improved industrial waste management, and enhanced urban dust suppression measures. Addressing these pollution sources through policy interventions and public health initiatives is critical for reducing environmental health risks and safeguarding vulnerable populations, particularly children.

Policy implications

The findings of this study have important implications for urban environmental management and public health protection in Istanbul and similar megacities. In line with the WHO Air Quality Guidelines reported in 2021¹⁰³ and the European Union’s urban pollution report¹⁰⁴ our results highlight elevated non-carcinogenic and carcinogenic risks for children due to exposure to toxic elements such as Pb, Cr, and As through ingestion of street dust. The source apportionment results, particularly those derived from PMF, PCA, and cluster analysis, indicate that traffic emissions and industrial runoff are the primary contributors to PTE contamination in urban dust. These results support the need for targeted mitigation measures. In particular, the implementation of green or smart road cleaning technologies could help reduce dust resuspension and airborne exposure⁶⁶. Restrictions on heavy-duty vehicles in densely populated or high-risk districts, alongside enhanced urban greening efforts, could further limit the circulation of contaminated particles in the environment¹⁰⁵. Additionally, land use policies and zoning regulations may be strengthened to limit industrial emissions near residential zones. Establishing regular monitoring programs focused on street dust composition could also provide an early warning system and help evaluate the effectiveness of interventions^106,107. These locally tailored measures would align with broader sustainability and public health goals while reinforcing Türkiye’s commitment to harmonizing urban environmental policies with EU standards. These findings not only inform local pollution control but also support Istanbul’s progress toward sustainable urban development and health-focused environmental governance.

Conclusion

This study provides a detailed evaluation of PTEs in urban street dust from Istanbul, integrating geochemical analysis, source apportionment, ecological risk indices, and health risk assessment. The findings reveal that street dust serves not only as a contaminant reservoir but also as a potential airborne vector for toxic metal exposure in urban settings. Through the application of Positive Matrix Factorization, Monte Carlo simulation, and multivariate statistics, the spatial patterns, sources, and health implications of heavy metal contamination were effectively characterized. Accordingly, the key findings include:

• Based on the PMF results, supported by PCA, HCA, and PCC analyses, industrial runoff (39.4%) was identified as the dominant source of PTEs, followed by traffic emissions (31.3%) and natural/soil sources (29.4%).

• Contamination indices (I_geo, EF, CF, PERI) indicated moderate to very high ecological risks at several locations, particularly for Cr, Cu, Zn, and Pb.

• MCS-enhanced health risk assessments showed that ingestion is the primary exposure route, with children facing a higher total hazard index (THI = 3.13) compared to adults (2.32).

• Although the TCR values numerically exceeded the USEPA threshold, the outcome was predominantly influenced by nickel ingestion, which is not classified as carcinogenic via the oral route. Therefore, the elevated TCR does not represent a significant health concern in this context.

• Traffic-related metals (Pb, Cu, Zn) were major contributors to health risks and also pose a concern for air quality due to their resuspension potential.

• Although the overall air quality during the study period was classified as ‘Good’, the findings suggest that localized dust contamination can still present significant health risks through inhalation and ingestion pathways.

• Stable meteorological conditions favored the persistence of dust-borne contaminants, which can contribute to long-term human exposure.

• Sankey diagram visualization confirmed the dominance of vehicular and industrial emissions in driving both ecological and human health burdens.

Given these outcomes, a multifaceted mitigation approach is necessary. Regulatory control of industrial discharges, expansion of low-emission transport systems, street cleaning and dust suppression strategies, and urban vegetation initiatives are strongly recommended to reduce human exposure and environmental accumulation of PTEs.

Importantly, the integration of Monte Carlo simulation enhanced the robustness of the health risk assessment by accounting for variability and uncertainty in exposure parameters, reinforcing the conclusion that no significant health risk exists under current exposure scenarios. Future studies should incorporate bioaccessibility and metal speciation analyses, and consider real-time monitoring of airborne particulate-bound metals to better evaluate inhalation risks. In conclusion, this work highlights the urgency of implementing integrated environmental and public health strategies in rapidly urbanizing areas to safeguard both ecological integrity and human well-being.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Author contributions

T. Ö: Sampling, Laboratory analysis; M. M. Y: Formal analysis, Visualization, Software (PMF, PCA, MCS), Writing – review & editing; F. U: GIS mapping, Spatial interpretation, Ecological indices, Writing – review & editing; E. H: Air Quality Data interpretation, Validation, Literature review; B. Y: Writing – original draft; Supervision, PMF, Methodology design, Health risk modeling, Final manuscript preparation, Project administration,

Funding

This work was supported by Giresun University Scientific Research Projects Coordination Unit under award number FEN-BAP-A-290224-43.

Data availability

All data generated or analysed during this study are included in this published article [and its supplementary information files].

Declarations

Competing interests

The authors declare no competing interests.