Investigating BTEX Emissions in Greece: Spatiotemporal Distribution, Health Risk Assessment and Ozone Formation Potential

Abstract

This study investigates the atmospheric concentrations, spatiotemporal distribution, the associated health risks and the ozone formation potential of benzene, toluene, ethylbenzene and xylenes (BTEX) across 33 monitoring sites of Greece over a one-year period. Samples were collected using passive diffusive samplers and analyzed by gas chromatography–mass spectrometry (GC-MS). The highest BTEX concentrations were detected during winter and autumn, particularly in urban and industrial areas such as in the Attica and Thessaloniki regions, likely due to enhanced emissions from combustion-related activities and reduced atmospheric dispersion. Health risk assessment revealed that hazard quotient (HQ) values for all compounds were within the acceptable limits. However, lifetime cancer risk (LTCR) for benzene exceeded the recommended limits in multiple regions during the colder seasons, indicating notable public health concern. Source apportionment using diagnostic ratios suggested varying seasonal emission sources, with vehicular emissions prevailing in winter and marine or industrial emissions in summer. Xylenes and toluene exhibited the highest ozone formation potential (OFP), underscoring their role in secondary pollutant formation. These findings demonstrate the need for seasonally adaptive air quality strategies, especially in Mediterranean urban and semi-urban environments.

1. Introduction

Air pollution is among the most pressing environmental challenges worldwide, with profound impacts on human health and ecosystems. It is estimated to cause over 4 million premature deaths each year and is now considered as the leading environmental risk factor for mortality from all causes as well as the second most significant environmental health threat after climate change [1,2,3,4].

Among air pollutants, volatile organic compounds (VOCs) are of particular concern due to their diverse physicochemical properties, persistence and ability to disperse far from emission sources [5]. Within this group, benzene, toluene, ethylbenzene and xylenes (ortho-, meta- and para- isomers), also known as BTEX, are aromatic compounds predominantly emitted from traffic, combustion processes and industrial activities [6,7,8]. Urban areas often exhibit the highest BTEX concentrations, raising concerns about chronic exposure [6]. In addition to their outdoor presence, BTEX, as well as other VOCs, are commonly found indoors, where exposure may contribute to both carcinogenic and non-carcinogenic risks. Indoor concentrations can be elevated due to sources such as cleaning products, paints and building materials. Recent studies have shown relatively high concentrations, exceeding recommended exposure limits [9,10]. These findings underscore the importance of managing BTEX in the atmosphere to reduce human exposure.

Their carcinogenic, mutagenic and neurotoxic properties—such as the well-documented association between benzene and leukemia [11,12,13,14,15]—have attracted significant public health attention [6,7,8,9,16,17,18,19]. Benzene is classified by the International Agency for Research on Cancer (IARC) as a Group 1 agent, recognized as carcinogenic to humans [11,12] while ethylbenzene is classified as a Group 2B carcinogen [20]. Xylenes and toluene can irritate the skin, eyes and respiratory tract and affect central nervous and immune system function [21,22,23].

Given the widespread presence of BTEX compounds and the health risks associated with prolonged exposure, several parameters and computational models have been developed to assess both cancer and non-cancer risks from inhalation based on measured ambient concentrations [24,25]. These approaches provide essential tools for determining exposure limits and supporting effective air quality management strategies.

Beyond toxicological effects, BTEX compounds influence atmospheric chemistry. Through reactions with hydroxyl radicals (OH) and nitric oxide (NO), they contribute to the formation of tropospheric ozone (O₃), a pollutant harmful to humans and ecosystems [25,26]. Accurate knowledge of BTEX concentrations is therefore essential for assessing public health risks and developing air pollution mitigation approaches.

Although several studies have examined BTEX levels in Greece, most focused on individual cities or short monitoring periods [27]. Comprehensive, long-term assessments that integrate nationwide ambient measurements, health risk evaluation and emission source identification remain limited. In this light, this study addresses these gaps by conducting a year-long investigation of BTEX concentrations at 33 sites across Greece, covering areas with diverse urban and environmental characteristics. In addition to evaluating potential health risks from inhalation exposure, the study identifies possible emission sources, providing a holistic assessment not previously reported at the national scale. By combining long-term monitoring with risk and source analyses, this work offers novel insights to support evidence-based air quality management and policy decisions, contributing to improved emissions inventories and public health protection strategies.

2. Materials and Methods

2.1. Sampling Campaign Description

The spatiotemporal variations of BTEX compounds were studied at 33 monitoring sites across Greece, as illustrated in Figure 1, where the stations of the National Air Pollution Monitoring Programme are located. The selection of these locations was designed to ensure a representative coverage of population exposure on a national scale. For this purpose, sampling was conducted at the Greek principal urban agglomerations, including the Attica Region (Greater Athens Area), where nine different sampling points were set and the Thessaloniki Region with five different sampling sites. In addition to the metropolitan areas, sampling campaigns were also carried out in several other urban sites characterized by different emission profiles. These included industrially influenced regions such as Ptolemaida and Megalopolis; cities exhibiting mixed pollution sources from urban, industrial and harbor-related activities, including Heraklion, Patra1, Patra2 and Volos; and mixed-sized urban sites like Agrinio, Lamia, Thiba and Larisa, where air quality is affected by both urban and rural sources. Furthermore, locations with distinct climatic conditions such as Ioannina and Florina were also included [28]. Background sites such as Syros, Mytilene and Kerkira were incorporated as areas with theoretically lower atmospheric degradation potential. The network was completed with additional sites including Kavala, Alexandroupolis Kalamata and Chania in order to provide more comprehensive spatial coverage and a more robust characterization of BTEX variation levels. It should be noted that within the Attica Region (Greater Athens area) the monitoring network included the following sampling points: Aristotelous (center of Athens–urban) Aspropirgos (industrial–urban), Elefsina (industrial–urban), Kifisos (adjacent to the A1 motorway), L. Kifisias (the city’s busiest avenue), Nea Smirni (urban), Palaio Faliro (urban), Peristeri (urban) and Piraeus (urban, encompassing Port of Piraeus). Similarly, the Thessaloniki Region included five locations: Agia Sofia (urban), Kalamaria (urban), Kordelio (urban–industrial), Inner ring road (motorway-adjacent) and Plateia Dimokratias (center of Thessaloniki).

The sampling campaign was conducted over the period from 25 October 2014 to 31 August 2015. This period was subdivided into 4 sampling seasons, each season lasting for approximately 15–20 days. At each sampling site and during each season, two passive samplers were placed concurrently. As a result, the arithmetic means of the two measurements obtained per site and season was subsequently calculated and reported as the representative concentration value. Comprehensive details regarding the geographic coordinates of each site as well as the sampling timeframe are shown in Table S1. In total, 264 samples were collected.

2.2. Chemicals and Reagents

Air samples were collected using passive samplers. The sampling apparatus consisted of cartridge absorbents for sampling BTEX and VOCs, featuring a matrix of SS net (100 mesh, 5.8 mm diam.) activated with activated charcoal (30–50 mesh) (Radiello, Pavia, Italy). Throughout the sampling period, the absorbents were housed within white diffusive bodies (Radiello, Pavia, Italy) that were placed onto a triangular support plate (Radiello, Pavia, Italy). It should be noted that passive samplers provide time-integrated concentrations and may underestimate short-term peak events compared with active sampling methods. Nevertheless, their use is well suited for long-term monitoring across multiple sites, ensuring comparability and capturing spatial and seasonal trends, which were the primary objective of this work [29]. All samplers were installed at a height ranging between 2 and 4 m above ground level to approximate human breathing zone exposure while minimizing potential ground-level interferences. Upon collection of samples, the absorbents were removed from the white diffusive bodies and placed in their storage vials. The vials were, subsequently, covered with aluminum foil and maintained at 4 °C, until transported to the laboratory for analysis.

Sample pretreatment involved extraction with a toluene-d8 solution in carbon disulfide (Reagent grade, Low Benzene, Honeywell, Charlotte, (NC) USA) at a concentration of 25 ppm. Toluene-d8 solution (certified reference material, 2000 μg/mL, Supelco, Belefonte, (PA), USA was utilized as an internal standard. The desorption of the analytes is achieved by agitating 2 mL of the toluene-d8 solution using a vortex mixer (Scientific Industries, Inc., Bohemia, (NY), USA) for 30 min. Subsequently, a 1 mL aliquot of the extract was transferred into 2 mL autosampler vials for chromatographic analysis.

2.3. BTEX Determination

A gas chromatograph (6890N, Agilent, Santa Clara, (CA) USA) coupled with a mass spectrometer (MSD 5975B Agilent, Santa Clara, (CA) USA) was employed for BTEX determination. A 1 μL sample aliquot was injected at the gas chromatographic system operating in pulsed splitless mode, with injector’s temperature maintained at 280 °C. Separation of the analytes was achieved by a capillary HP-5MS column, (30 m × 0.25 mm × 0.25 μm), with the following temperature program: the initial temperature was set at 35 °C and held for 2 min, then ramped to 70 °C at rate of 5 °C min⁻¹ and subsequently raised to 200 °C at a rate of 10 °C min⁻¹, where it was held for 2 min. Helium was used as a carrier gas with a flow of 1.5 mL min⁻¹. Electron impact (EI) was the ionization source, with its temperature set at 230 °C.

MS data for BTEX determination is shown in Table S2. It should be noted that as m-xylene and p-xylene could not be separated, the sum of m+p-xylenes is calculated.

The method was fully validated. The linear range of all the analytes was 0.50–25 μg L⁻¹. R-squared values (R²) for each of the analytes was >0.996. Validation experiments of precision and accuracy were conducted at two calibration levels (5.0 and 25 μg L⁻¹). Repeatability RSD ranged from 0.13% (toluene 5.0 μg L⁻¹) to 0.90% (benzene 25.0 μg L⁻¹), while reproducibility RSD varied from 0.80% (o-xylene 5.0 ppm) to 2.5% (benzene 25.0 μg L⁻¹). Accuracy experiments were conducted by spiking the appropriate amount of the analytes to cartridge absorbents followed by the pretreatment procedure. Satisfactory results were obtained with accuracy values varying from 95.3% (o-xylene 25.0 μg L⁻¹) to 103% (ethylbenzene 25.0 μg L⁻¹). Limits of detection (LOD) were calculated from 0.028 μg L⁻¹ (m+p-xylene and o-xylene) to 0.087 (toluene) μg L⁻¹, which corresponds from 0.056 to 0.174 μg per sample.

Field blank and laboratory blank samples were treated as real samples and analyzed at each sampling campaign. The results were corrected if necessary.

2.4. BTEX Exposure Health Risk Assessment

For the assessment of the health risk posed by airborne BTEX to human health, various indicators have been proposed in the literature to calculate both the non-carcinogenic risk of all BTEX compounds and the lifetime cancer risk (LTCR) associated with benzene and ethylbenzene inhalation. The LTCR was determined based on the following equation:

LTCR = CDI × CSF

(1)

where CSF refers to the cancer slope factor (${\displaystyle \frac{\mathrm{m}\mathrm{g}}{\mathrm{k}\mathrm{g}\times \mathrm{d}\mathrm{a}\mathrm{y}}}$) and CDI is the chronic daily intake (${\displaystyle \frac{\mathrm{m}\mathrm{g}}{\mathrm{k}\mathrm{g}\times \mathrm{d}\mathrm{a}\mathrm{y}}}$). CDI can be calculated according to the following equation:

CDI = (C × IR × CF × ET × ED × EF)/(AT × BW)

(2)

where C is the concentration of the compound (μg m⁻³), CF is the conversion factor (0.001 mg μg⁻¹), IR is the inhalation rate (0.83 m³ h⁻¹), ET is the exposure time (24 h day⁻¹), ED is the exposure duration (24 years), EF is the exposure frequency (350 days year⁻¹), BW is the body weight (70 kg) and AT is the average lifetime (70 years × 365 days year⁻¹ for cancer risk assessment and ED in years × 365 days year⁻¹ for non-cancer risk assessment) [30,31,32,33]. CSF values for benzene and ethylbenzene are 0.0273 and 0.00385 (${\displaystyle \frac{\mathrm{m}\mathrm{g}}{\mathrm{k}\mathrm{g}\times \mathrm{d}\mathrm{a}\mathrm{y}}}$), respectively [34,35].

The assessment of non-carcinogenic risk for BTEX exposure is represented by the hazard parameter known as the Hazard Quotient (HQ). This parameter is expressed as the ratio of CDI to the reference dose (RfD), which is determined as the product of the reference concentration (RfC, mg m⁻³) × 20 (assumed respiratory rate in m³ day⁻¹) × BW⁻¹ (kg⁻¹). The HQ is calculated using the following equation:

HQ = CDI/RfD

(3)

Values of HQ > 1 indicate significant potential risk, while values of HQ ≤ 1 suggest that the exposure levels are within acceptable limits [36,37,38,39]. RfC values for benzene, toluene, xylenes and ethylbenzene are 0.03, 5, 0.1 and 1 mg m⁻³ respectively [31], while RfD values were 0.00855, 1.4, 0.029 and 0.286 mg kg⁻¹ day⁻¹, respectively. It should be noted that LTCR values for toluene and xylenes cannot be calculated as there are not CSF values for these compounds [36,38].

3. Results and Discussion

3.1. Spatiotemporal Variations

Spatial and seasonal variations of all the compounds and areas studied are presented in Tables S3–S7 and Figure 2, Figure 3, Figure 4, Figure 5 and Figure 6. A clear seasonal pattern was observed, with the highest BTEX concentrations occurring during winter, followed by autumn or spring depending on the location, while minimum levels occurred in summer. This behavior can be attributed to a combination of emission-related and meteorological factors. In winter, the increased use of domestic heating substantially contributes to emissions while at the same time the lower boundary layer height and weaker atmospheric mixing enhance pollutant accumulation near ground [27,40]. This pattern is particularly, evident in Florina (approximately 897 m ASL), where winter ΣBTEX reached 27.0 μg m⁻³, attributable to intense domestic heating emissions combined with the shallow boundary heights and stagnant meteorological conditions that characterize mountainous northern regions. By contrast, in summer, enhanced solar radiation and photochemical activity increase the degradation rates of BTEX, while deeper boundary layers favor dilution and dispersion, leading to notably lower concentrations [41,42]. Elevated O₃ levels during summer also accelerate the oxidative loss of reactive aromatics such as benzene [41].

Spatial differences across sites reflect both local emission sources and site-specific meteorological conditions. For example, the highest annual ΣBTEX levels were observed in the Attica Region (e.g., Palaio Faliro, Piraeus, L. Kifisias and Kifisos) and in Patra1, where dense traffic, harbor activities and restricted ventilation in street canyons amplify concentrations. Similarly, Agrinio exhibited high ΣBTEX concentrations during both winter and spring (18.1 and 18.8 μg m⁻³, respectively), reflecting the influence of mixed urban and rural sources, while elevated summer values were observed at harbor-adjacent sites (e.g., in Patra, Plateia Dimokratias in Thessaloniki and Piraeus) highlighting the role of harbor-related activities [43]. In contrast, Syros presented the lowest annual values (5.1 μg m⁻³), reflecting the combined effects of weaker local emissions, stronger coastal winds and enhanced atmospheric dispersion. Strong wind speeds contribute to dispersion; Sahu et al. in 2016 found VOC levels decreased exponentially with increasing speed [44]. In other studies, higher concentrations of BTEX were reported in Mexico, attributed to the different characteristics of each site including the variety of industries, the population density, vehicular traffic density, commercial establishments, etc. [45].

Regarding each individual BTEX, the highest concentration of benzene was observed in Agrinio (6.4 μg m⁻³) during spring, and in Alexandroupolis and Florina (6.4 μg m⁻³) during winter. On the contrary, in other sites such as Volos, benzene reached its maximum concentration during autumn (5.0 μg m⁻³). In general, the detected concentrations of benzene in this study are significantly lower than those reported in urban sites and/or traffic stations, in Kolkata, India (25.0–79.2 μg m⁻³) [46], and in Nairobi, Kenya (21.6 ± 7.09 μg m⁻³) [47], but higher that the concentrations observed in European sites including Berlin, Germany (0.82 ±0.54 μg m⁻³), Budapest, Hungary (0.89 ± 0.67 μg m⁻³), or Torino, Italy (0.63 ± 0.57 μg m⁻³) [48]. For comparison purposes, average benzene values for other studies are presented in Table 1. Annual benzene values, regarding mostly urban sites, are comparable to those observed in Düzce, Turkey (4.64 ± 2.42 μg m⁻³), during autumn [49] as well as those reported for Arad, Romania, which exhibited a mean concentration of 2.87 ± 0.58 μg m⁻³ [41]. Regarding the annual threshold values, the reported values at all locations remain below the established annual limit thresholds. The highest annual values were observed in areas of the Attica Region (Palaio Faliro and Peristeri, 3.5 and 3.4 μg m⁻³, respectively), with the lowest value recorded in the island of Syros (0.99 μg m⁻³).

Several cases of benzene exceedances above the EU annual limit of 5 μg m⁻³ were observed during specific periods, particularly in winter at nine locations (e.g., Alexandroupolis, Ioannina, Lamia, Patra1, Patra2, Florina, Kifisos, Palaio Faliro and Agia Sofia), as well as in autumn (Volos, Aristotelous and Palaio Faliro, and once in spring (Agrinio). These exceedances indicate that air quality degradation related to benzene is not restricted to the major urban centers of Athens, Thessaloniki and Patra, but also extends to medium-sized cities and even rural-urban environments.

Toluene was the most abundant among the studied BTEX compounds across all examined areas, with levels significantly lower than the WHO guideline limit of 260 μg m⁻³ [50]. Compared to other studies, lower concentrations have been detected in areas of Poland and Hungary (<1.39 μg m⁻³) [51] and Turkey where the average values for the January-February and July periods were 2.38 ± 1.27 and 0.45 ± 0.33 μg m⁻³, respectively [40]. On the other hand, significantly higher concentrations were reported in urban areas of Africa—44.5 ± 17.1 and 32.9 ± 8.27 μg m⁻³ in Nairobi and Lagos, respectively—as well as in Yazd, Iran (38 ± 42 μg m⁻³) [47,52]. In this study, the highest toluene levels were observed in location of northern Greece during the winter period such as Florina (7.69 μg m⁻³), Alexandroupolis (7.57 μg m⁻³) and Ioannina (7.04 μg m⁻³) as well as in locations in the networks of the Attica (i.e., Kifisos and Palaio Faliro) and Thessaloniki Regions (i.e., Agia Sofia). The strong influence of meteorology combined with emissions’ intensity on toluene and general BTEX seasonality has been highlighted in previous works, where wintertime accumulation under inversion conditions contrasted with summer depletion due to both dispersion and atmospheric chemistry [44,53].

Regarding m+p-xylenes, similar seasonal variations were observed, with higher values being detected during winter, followed by mainly autumn, while the lowest were noticed during summer. Locations with the highest detected values were in the Attica Region (Palaio Faliro) as well as in Patra1 and Ioannina. Higher m+p-xylene levels have been detected in Iran [36,54], while values over 10 μg m⁻³ were detected in urban sites of Africa [47]. On the contrary, in other studies conducted in Poland and Hungary, the detected concentrations were relatively lower than those of this study [51].

Ethylbenzene and o-xylene exhibited similar concentrations, being the least abundant BTEX species, with annual concentrations from 0.47 to 2.6 μg m⁻³. The highest concentrations of ethylbenzene and o-xylene were detected at Attica Region locations as well as in Patra1 and in Florina, whereas the lowest ones were noticed in the Island of Syros. In other studies, lower concentrations of o-xylene (0.44–1.13 μg m⁻³) and ethylbenzene (0.42–1.64 μg m⁻³) have been reported in Salvador, Brazil [55]. Significantly higher levels have been reported in Tehran, Iran, with annual mean concentrations of 4.53 μg m⁻³ for o-xylene and 3.63 μg m⁻³ for ethylbenzene [36]. Similarly, in Shiraz, Iran, average morning concentrations were reported as 23.38 ± 8.28 μg m⁻³ and 7.50 ± 2.50 μg m⁻³, for o-xylene and ethylbenzene, respectively, while nighttime levels were measured at 21.25 ± 7.13 μg m⁻³ and 6.58 ± 1.95 μg m⁻³, respectively [54].

The noticeable spatiotemporal variability in BTEX levels across different locations can be attributed to factors such as vehicular traffic density, fuel quality and vehicle technology, industrial emissions as well as the geographical and meteorological characteristics of the studied locations [54]. For example, the influence of meteorological dynamics on atmospheric dispersion and chemical reactivity, seasonal shifts in wind direction and speed underscores that meteorology acts as a key modulator of BTEX seasonality [27,42,44]. It should be noted that no normalization for traffic density of other activity indicators was applied, since the purpose of this study was to capture the real-world variability of BTEX concentrations across diverse environments.

Table 1. Benzene concentrations (μg m⁻³) at different locations and seasons.

Location, Time/Reference	Average ± SD or Range
Bolu, Turkey, January–February 2017 [40]	1.75 ± 1.06
Arad, Romania, January–December 2016 [41]	2.87 ± 0.58
Kolkata, India, March–June 2009 [46]	25.0–79.2
Nairobi, Kenya (Traffic station), June 2019 [47]	21.6 ± 7.09
Lagos, Nigeria (Traffic station), November–December 2019 [47]	24.2 ± 7.29
Nairobi, Kenya (background), June 2019 [47]	4.95 ± 1.74
Lagos, Nigeria (Background), November–December 2019 [47]	4.20 ± 2.20
Berlin, Germany, 2015–2017 [48]	0.82 ± 0.54
Budapest, Hungary, 2015–2017 [48]	0.89 ± 0.67
Mons, France, 2015–2017 [48]	0.57 ± 0.45
Torino, Italy, 2015–2017 [48]	0.63 ± 0.57
Düzce, Turkey, October–November 2014 [49]	4.64 ± 2.42
Düzce, Turkey, August 2015 [49]	0.74 ± 0.24
Gdansk, Poland, January–December 2012 [51]	0.75 ± 0.67
Gdynia, Poland, January–December 2012 [51]	0.66 ± 0.51
Sopot, Hungary, January–December 2012 [51]	0.63 ± 0.55
Yazd, Iran, July 2015 and January 2016 [52]	21 ± 18 (2–108)
Leon, Mexico, August and October 2018 [56]	1.96
Bari, Italy, April and October 2008 [57]	0.8–9
Haifa, Israel, August 2015–May 2021 [58]	1.18–3.19
Jerusalem, Israel, August 2015–May 2021 [58]	1.18–4.00
Tel-Aviv, August 2015–May 2021 [58]	1.18–3.66
Dongbaituo, Hebei, China January 2013 [59]	27.2 ± 16.0
Thissio, Greece, Winter 2017 [60]	2.6 ± 3.4
Thissio, Greece, Summer 2016 [60]	0.9 ± 0.8
Algiers, Algeria, Spring 2009 [61]	16.7

Table 1. Benzene concentrations (μg m⁻³) at different locations and seasons.

Location, Time/Reference	Average ± SD or Range
Bolu, Turkey, January–February 2017 [40]	1.75 ± 1.06
Arad, Romania, January–December 2016 [41]	2.87 ± 0.58
Kolkata, India, March–June 2009 [46]	25.0–79.2
Nairobi, Kenya (Traffic station), June 2019 [47]	21.6 ± 7.09
Lagos, Nigeria (Traffic station), November–December 2019 [47]	24.2 ± 7.29
Nairobi, Kenya (background), June 2019 [47]	4.95 ± 1.74
Lagos, Nigeria (Background), November–December 2019 [47]	4.20 ± 2.20
Berlin, Germany, 2015–2017 [48]	0.82 ± 0.54
Budapest, Hungary, 2015–2017 [48]	0.89 ± 0.67
Mons, France, 2015–2017 [48]	0.57 ± 0.45
Torino, Italy, 2015–2017 [48]	0.63 ± 0.57
Düzce, Turkey, October–November 2014 [49]	4.64 ± 2.42
Düzce, Turkey, August 2015 [49]	0.74 ± 0.24
Gdansk, Poland, January–December 2012 [51]	0.75 ± 0.67
Gdynia, Poland, January–December 2012 [51]	0.66 ± 0.51
Sopot, Hungary, January–December 2012 [51]	0.63 ± 0.55
Yazd, Iran, July 2015 and January 2016 [52]	21 ± 18 (2–108)
Leon, Mexico, August and October 2018 [56]	1.96
Bari, Italy, April and October 2008 [57]	0.8–9
Haifa, Israel, August 2015–May 2021 [58]	1.18–3.19
Jerusalem, Israel, August 2015–May 2021 [58]	1.18–4.00
Tel-Aviv, August 2015–May 2021 [58]	1.18–3.66
Dongbaituo, Hebei, China January 2013 [59]	27.2 ± 16.0
Thissio, Greece, Winter 2017 [60]	2.6 ± 3.4
Thissio, Greece, Summer 2016 [60]	0.9 ± 0.8
Algiers, Algeria, Spring 2009 [61]	16.7

3.2. Evaluation of BTEX Specific Ratios

The evaluation of BTEX emission sources was conducted using specific compound ratios, primarily the toluene/benzene (T/B), m+p-xylene/ethylbenzene (m+p X/E) and m+p-xylene/benzene (m+p X/B) ratios. The T/B ratio is widely recognized as an indicator of vehicle exhaust, industrial and stationary sources particularly in densely populated areas [30,36,62,63]. For example, T/B values from 1.3 to 3.0 are associated with vehicle exhaust emissions sources [51,64], with the same sources being indicated, in other studies, with ratio values from 1.5 to 4.3 [36,65].

Although toluene is typically emitted in larger quantities than benzene, its shorter atmospheric lifetime (2.1 days for toluene vs. 9.5 days for benzene) results in more rapid degradation. Consequently, benzene tends to persist longer in the atmosphere, accumulating in greater quantities compared to toluene in the vicinity of their emission sources. Such differences in atmospheric persistence often lead to lower T/B values [66].

As presented in Table S8, the T/B values for each sampling location in the Attica region ranged from 1.2 to 2.0 across all seasons, except for summer, during which values ranged from 3.1 to 96. This variation suggests a significant contribution of vehicle emissions throughout the year, except during the summer months. This phenomenon has also been observed in other European cities [41] and can be attributed to the reduction in the population density of highly urbanized areas, such as Athens, during the summer vacation period. However, in sampling locations adjacent to major roads such as Aristotelous and L.Kifisias as well as in highly densely populated areas including N. Smirni and Piraeus, the T/B ratio did not exhibit a significant increase in summer, highlighting the persistent contribution of vehicular emissions. On the contrary, in the Thessaloniki region, stationary sources dominated during autumn, while traffic emissions prevailed in winter and spring. For areas including Patra1 and Patra2, Agrinio and Florina, traffic emissions are identified as the predominant sources of pollution throughout the entire year. In addition, a notable increase in the ratio was calculated in areas such as Volos, Kalamata, Lamia, Ioannina and Megalopolis, indicating the significant contribution of stationary sources. These areas share a common characteristic of increased industrial activities which may explain the observed ratio variation. A similar increase in the T/B ratio was also observed, during the summer period, on the island of Syros and in Chania, which can likely be attributed to elevated marine traffic emissions, as marine traffic significantly increases during this period. In other studies, Mentese and Akca in 2020 reported a same increase in the T/B ratio at two Turkish harbors, during the summer, attributing the increase to marine traffic emissions, since the demand of domestic heating is minimal during this time [67]. A recent study conducted across seven European countries and twenty-two monitoring sites with various characteristics reported that the calculated ratio indicated pollution from local traffic emissions, followed by industry-related activities as well as from biomass burning [66]. In Iran, the T/B value was found 1.8 in summer and 1.3 in winter highlighting the variety of parameters influencing the ratio including photochemistry, meteorology, and emission aspects [52].

In addition to evaluating potential sources through the ratio calculations of the T/B ratio, other ratios can be calculated to estimate atmospheric aging, based on average oxidation time of BTEX by OH radicals [68,69]. Specifically, the ratio of m+p-xylenes/ethylbenene (m+p X/E) provides insights of the age of an air mass of a studied area [51,70]. This ratio acts as an indicator of photochemical reactivity, as the reaction rate constant of m+p-xylenes with OH radicals is higher than that of ethylbenzene and OH radicals [42].

In this study, m+p X/E ratios varied from 0.53 to 64 (Table S9) which aligns with findings from other studies that include areas with different characteristics [36]. Most of the studied sites presented similar variations, with consistent values during autumn, winter and spring, and a notable increase during summer, suggesting that air masses are relatively fresh, indicative of direct emission from their sources, whereas during summer, transportation of air masses appears to play a more significant role [30]. This is also in agreement with other studies which suggest m+p X/E ratios higher than 3 are indicative of fresher air masses [71]. Furthermore, the m+p X/E ratio exhibits an increasing trend with rising temperature and solar radiation intensity due to the enhancement of photochemical activity [42]. In several urban sites of Europe, the aforementioned ratio varied from 1.75 ± 0.91 to 3.68 ± 0.30, highlighting the variety in BTEX emission sources across monitoring sites, although in most of the studied sites, it was indicated that pollutants were mainly emitted from local sources [66].

Similarly, higher values of the m+p X/B ratio suggest fresher air masses, which means emission from local sources whereas lower values are indicative of transported air masses [72]. Other studies suggest that m+p X/B ratios below 1.8 are indicative of long-range transported air masses with predominant vehicular emission, whereas higher ones reveal emissions from local sources [73]. As shown in Table S10, the majority of the calculated ratios were found to be less than 1.8, indicating vehicle exhaust emissions as the major sources of pollution. However, local sources appear to be more significant for the Attica Region during summer, as well as for the Thessaloniki Region during autumn. High ratio values have been reported in the literature and are attributed not only to the different pollution sources but also to the specific climate and meteorological conditions of each area studied [30].

To further support the findings from compound ratios, the ratios of B:T:E were examined using a ternary diagram, as shown in Figure 7. The classification of emission sources followed the framework of Zhang et al., 2016, and Pinthong et al., 2022, where three main categories are identified: traffic emissions, industrial and solvent emissions and combustion emissions (including biomass, biofuel and coal) [53,74]. This approach provides a useful means of visualizing how different sources contribute simultaneously and how their relative importance shifts through time and space.

The ternary analysis revealed two notable patterns. First, traffic emissions appear to dominate across all seasons, with their contribution being particularly strong during summer. This finding aligns with receptor modeling studies in Greek urban environments, which show that vehicular sources are consistently significant during the warm season [75,76]

Second, during winter, the distribution of data points becomes less constrained to a single vertex and shows a more dispersed pattern. This suggests that wintertime air masses are often characterized by a mixed-source composition, reflecting the influence of combustion for residential heating, industrial activities and traffic emissions. The clustering of data points outside the well-defined clusters highlights the complexity of wintertime source apportionment. This pattern is consistent with previous source apportionment studies in Athens and Thessaloniki, which have demonstrated the strong influence of biomass burning for residential heating, combined with traffic and industrial activities, leading to a complex mixture of sources during cold periods [76,77].

A third aspect worth noticing concerns industrial and solvent emissions which appear to be significant during autumn. This seasonal increase is particularly evident in northern Greece such as Ptolemaida, Kavala, Alexandroupolis and Thessaloniki Region (Kalamaria, Kordelio, Agia Sofia) as well as in Thiba and Chania. Studies in western Macedonia and especially the lignite basin of Ptolemaida-Kozani have clearly demonstrated that lignite combustion and associated industrial activities contribute significantly to particulate matter and trace elements, leaving a distinctive chemical fingerprint in the region [78,79]. Such spatial differentiation points to the role of regional industrial activities and possibly seasonal changes in solvent use, for example, in agriculture and small-scale industries, that become more relevant during this period.

3.3. Ozone Formation Potential (OFP)

In the field of atmospheric chemistry, the ozone formation potential serves as a predictive metric for assessing the capacity of BTEX, and VOCs in general, to generate ozone [80]. OFP is estimated by the concentration of each individual compound combined with is reactivity in the air with the following equation [71]:

OFP (i) = C (i) × MIR (i)

(4)

where OFP (i) is the ozone formation potential of each individual compound (i) in μg m⁻³, C (i) is the concentration of the compound in μg m⁻³ and MIR (i) is the maximum incremental reactivity of each compound calculated initially by Carter and co-workers [81]. The MIR values used in this work for benzene, toluene, ethylbenzene and xylenes are respectively equal to 0.72, 4.00, 3.04, 7.74 [80].

OFP values for individual compounds and ΣBTEX are reported in Tables S11–S15 while Figure 8 and Figure 9 illustrate the spatiotemporal variability.

ΣBTEX OFP values ranged from 0.5 to 118.3 μg m⁻³ with the highest values being observed during winter in urbanized sites in the network of the Attica and Thessaloniki Regions (Kifisos, L. Kifisias, Palaio Faliro and Agia Sofia), with values over 100 μg m⁻³ and in sites influenced by mixed pollution sources such as Patra (100.9 and 99.3 μg m⁻³), Alexandroupolis (102.0 μg m⁻³) and Florina (118.3 μg m⁻³). A pronounced seasonal trend was evident, with total OFP levels peaking in winter and autumn and reaching their minimum in summer. Elevated winter OFP values likely reflect a combination of enhanced emissions from heating and traffic, lower atmospheric mixing and reduced photochemical degradation due to shorter daylight hours and lower temperatures. Conversely, OFP values were the lowest during summer, when higher temperatures and stronger solar radiation promote ozone photochemical cycling and increased atmospheric dispersion [66]. In other studies, however, OFP values did not show significant seasonal changes, suggesting different dominant sources across seasons [82]. Spatially, urban and port areas such as L. Kifisias, Peiraeus and Patra1 demonstrated consistently elevated OFP values, aligning with higher vehicular and/or harbor related emissions.

Across nearly all sites and seasons, xylenes dominated the OFP profile, suggesting their significant role in ozone precursor activity. Toluene was the second-largest contributor, with higher levels observed during summer at selected sites. These patterns are consistent with previous studies, which also reported xylenes and toluene as the main contributors [52,80]. On the other hand, benzene and ethylbenzene displayed much lower OFP values across the dataset. Similar studies in India have reported elevated ΣBTEX OFP values in both urban and rural sites [72,83], highlighting that strong anthropogenic emissions and/or regional transport of pollutants can result in high ozone precursor levels even outside major cities. Beyond BTEX compounds, other VOCs can play an important role in ozone formation. For example, in Yibin, ethylene and acetaldehyde were the highest OFP contributors in Yibin winter and summer, respectively [82]. Similarly, in a typical industrial Park in the Pearl Delta River, alkenes were the dominant OFP contributor group, with isoprene, trans-2-butene and 1-Hexene as the most significant contributors, followed by monoaromatic hydrocarbons such as xylenes and trimethylbenzenes [84].

The pronounced seasonal and spatial trends observed here underscore the importance of seasonally resolved source apportionment to understand ozone formation dynamics and design mitigation strategies. While BTEX compounds are significant OFP contributors in Greek urban environments, a comprehensive assessment of photochemical ozone formation must also account for other VOCs and their interactions, especially in atmospheres influenced by multiple anthropogenic and biogenic sources.

3.4. Lifetime Cancer Risk (LTCR)

The calculated LTCR values are summarized for benzene and ethylbenzene at Table S16 and Table S17, respectively. Moreover, Figure 10 and Figure 11 illustrate the spatiotemporal variations of benzene LTCR values across Greece, while Figure 12 presents the LTCR benzene values within the network of the Attica Region, the Thessaloniki Region and Patra. It should be mentioned that LTCR values of benzene between 1 × 10⁻⁶ and 1 × 10⁻⁵ are considered acceptable by the WHO, while USEPA suggests that values should be less than 1 × 10⁻⁶ [36,54,85].

The calculated benzene LTCR values illustrate the degradation of the air quality in the entire Greek territory mostly during the winter period, but also during autumn and spring. Specifically, during winter, benzene LTCR values have exceeded the threshold limits established by the WHO, not only in highly dense areas, such as within the network of the regions of Attica and Thessaloniki, but also in other areas such as Patra, Lamia, Larisa, Florina, etc. During autumn, in most of the areas of Attica Region, as well as in Volos and Patra, excess of the threshold was observed. Respective excesses were observed during spring in Piraeus and Heraklion. Nevertheless, during summer, the LTCR values are significantly reduced.

Regarding the benzene LTCR value thresholds established by USEPA, the degradation of the air quality in most of the studied locations in Greece is evident, as the corresponding values in these areas exceeded the USEPA thresholds during most of the seasons. During the period of the economic crisis in Greece, substantial air quality degradation has been observed in Athens and Thessaloniki, the two major urban centers, due to the significant emissions of biomass burning used to cover the domestic heating needs [86,87,88,89]. Similar threshold exceedances have been reported in the city of Arad, Romania, with benzene LTCR values being higher than the WHO limits during the colder months and decrease during the warmer months [41]. Even higher benzene LTCR values have been reported in certain regions of Iran reaching as high as 2.49 × 10⁻⁴ [54]. In other studies, regarding a European monitoring network, although the LCTR was calculated with a different approach, the benzene LTCR values especially in traffic and industrial sites were calculated above the permissible risk level suggesting moderate risk from benzene exposure [66]. In

Table 2. Benzene and ethylbenzene LTCR values from other studies.

Location/Reference	Benzene LTCR	Ethylbenzene LTCR
Kuala Lumpur, Malaisia [30]	1.59 × 10−5	1.63 × 10−6
Tehran, Iran [36]	3.93 × 10−7	not included
Arad, Romania [41]	4.6 × 10−5 (max) 6.13 × 10−6 (min)	not included
Shiraz, Iran [54]	1.96 × 10−4 (morning) 2.49 × 10−4 (afternoon)	not included
Dongbaituo, Hebei, China [59]	7.59 × 10−5	not included
Gorakhpur, India [63]	1.0 × 10−5	4.6 × 10−6
Changzhi, China [71]	3.8 × 10−5 (average)	2.9 × 10−6
Bangkok, Thailand [90]	4.37 × 10−6	1.47 × 10−6

, benzene and ethylbenzene LTCR values are presented, calculated from various other sites. Collectively, these results suggest that benzene-related health risks in Greece are not isolated to specific hotspots but rather extend to multiple environments with differing emission profiles.

By contrast, ethylbenzene LTCR values were considerably lower, yet non-negligible. The highest values occurred during winter and autumn, particularly in areas within the Attica Region, Patra, Florina, Ioannina, Alexandroupolis, Larisa, Mytilene and in Agia Sofia (Thessaloniki Region), where they exceeded the 10⁻⁶ threshold proposed by Zhang et al. [71]. In autumn, fewer excesses were observed mainly in the Attica Region and Volos, while during spring, only in Heraklion ethylbenzene LTCR value was calculated at 1.1 × 10⁻⁶. Compared with other studies worldwide, such as Gorakhur, India [63] and Bangkok, Thailand [90], where ethylbenzene LTCR values were of similar magnitude to benzene, the risk levels observed in Greece appear relatively moderate. However, the co-occurrence of benzene and ethylbenzene exposure raised concern about cumulative effects.

Overall, these results highlight that while benzene poses the most significant carcinogenic risk among BTEX compounds, ethylbenzene can also contribute to cumulative long-term health burdens, particularly during colder months.

3.5. Hazard Quotient (HQ)

HQ values for all studied compounds, areas and seasons were calculated to assess non-carcinogenic risk. As shown in

Table 3. Annual HQ values for each sampling location.

Location	Benzene	Toluene	Ethylbenzene	m+p-Xylene	o-Xylene
Agrinio	0.10	8.0 × 10−4	1.0 × 10−3	1.9 × 10−2	1.0 × 10−2
Alexandroupolis	8 × 10−2	7.1 × 10−4	1.4 × 10−3	1.9 × 10−2	1.2 × 10−2
Attica Region—(Greater Athens Area)
Aristotelous	8.0 × 10−2	7.6 × 10−4	2.5 × 10−3	2.6 × 10−2	1.8 × 10−2
Aspropirgos	5.0 × 10−2	6.4 × 10−4	1.9 × 10−3	2.4 × 10−2	1.4 × 10−2
Elefsina	5.0 × 10−2	5.4 × 10−4	1.1 × 10−3	1.7 × 10−2	1.2 × 10−2
Kifisos	0.10	8.3 × 10−4	2.0 × 10−3	3.4 × 10−2	1.7 × 10−2
L. Kifisias	0.11	8.7 × 10−4	1.9 × 10−3	3.3 × 10−2	1.7 × 10−2
Nea Smirni	7.0 × 10−2	6.1 × 10−4	1.3 × 10−3	2.3 × 10−2	1.3 × 10−2
Palaio Faliro	0.11	9.6 × 10−4	1.9 × 10−3	3.6 × 10−2	1.9 × 10−2
Peristeri	8.0 × 10−2	6.6 × 10−4	1.5 × 10−3	3.0 × 10−2	1.6 × 10−2
Piraeus	0.11	9.1 × 10−4	2.0 × 10−3	3.4 × 10−2	1.9 × 10−2
Chania	6.0 × 10−2	5.9 × 10−4	1.4 × 10−3	2.0 × 10−2	1.1 × 10−2
Florina	9.0 × 10−2	7.8 × 10−4	1.7 × 10−3	2.7 × 10−2	1.7 × 10−2
Heraklion	6.0 × 10−2	5.5 × 10−4	1.4 × 10−3	1.8 × 10−2	1.3 × 10−2
Ioannina	9.0 × 10−2	7.8 × 10−4	1.8 × 10−3	1.7 × 10−2	1.3 × 10−2
Kalamata	4.0 × 10−2	4.5 × 10−4	7.3 × 10−4	1.2 × 10−2	8.7 × 10−3
Kavala	3.0 × 10−2	5.4 × 10−4	8.5 × 10−4	1.3 × 10−2	7.9 × 10−3
Kerkira	8.0 × 10−2	7.2 × 10−4	1.4 × 10−3	1.8 × 10−2	1.3 × 10−2
Lamia	7.0 × 10−2	6.3 × 10−4	1.2 × 10−3	1.8 × 10−2	1.3 × 10−2
Larisa	0.10	8.4 × 10−4	1.9 × 10−3	1.9 × 10−2	1.4 × 10−2
Megalopolis	5.0 × 10−2	5.1 × 10−4	9.0 × 10−4	1.0 × 10−2	6.3 × 10−3
Mytilene	9.0 × 10−2	6.9 × 10−4	1.6 × 10−3	2.5 × 10−2	1.6 × 10−2
Patra1	0.11	8.7 × 10−4	1.9 × 10−3	3.6 × 10−2	2.2 × 10−2
Patra2	9.0 × 10−2	8.4 × 10−4	1.7 × 10−3	2.0 × 10−2	1.6 × 10−2
Ptolemaida	5.0 × 10−2	4.9 × 10−4	1.1 × 10−3	1.5 × 10−2	9.7 × 10−3
Syros	3.0 × 10−2	3.8 × 10−4	5.8 × 10−4	1.0 × 10−2	4.4 × 10−3
Thiba	5.0 × 10−2	4.7 × 10−4	9.0 × 10−4	1.4 × 10−2	1.1 × 10−2
Thessaloniki Region
Agia Sofia	7.0 × 10−2	7.1 × 10−4	1.2 × 10−3	2.4 × 10−2	1.5 × 10−2
Kalamaria	4.0 × 10−2	4.4 × 10−4	1.0 × 10−3	1.6 × 10−2	1.1 × 10−2
Kordelio	6.0 × 10−2	5.9 × 10−4	1.0 × 10−3	2.1 × 10−2	1.3 × 10−2
Inner ring road	4.0 × 10−2	6.3 × 10−4	5.7 × 10−4	1.8 × 10−2	1.3 × 10−2
Plateia Dimokratias	7.0 × 10−2	7.2 × 10−4	1.4 × 10−3	2.3 × 10−2	1.2 × 10−2
Volos	0.10	8.1 × 10−4	1.7 × 10−3	2.1 × 10−2	1.7 × 10−2

, none of the calculated HQ values exceeded the threshold of 1, suggesting that exposure to individual BTEX compounds is unlikely to pose significant non-carcinogenic risks under the observed concentration levels. Among the compounds, benzene consistently yielded the highest HQ values, reflecting its greater toxicological potency compared with the other species, followed by m+p-xylene, o-xylene, ethylbenzene and toluene. The spatial distribution of HQ values also highlights that urban and port areas, where traffic and industrial activity are more intense, tend to present higher exposure risk estimates compared with rural sites.

These findings align with several studies reporting HQ values below 1 in outdoor environments [30,36,39,41], although exceedances have been documented in other studies, such as o-xylene HQ values up to 2.33 [54], indicative of significant localized risks. Although HQ values for each BTEX compound were within acceptable limits, it is important to note that HQ values derived for single compounds may underestimate the overall health burden, as real-world exposure typically involves mixtures of VOCs. Cumulative or even synergistic effects may arise when BTEX co-occur with other toxic compounds such as aldehydes. For example, Nabizadeh et al. (2020) reported HQ values above 1 for formaldehyde and acetaldehyde in urban environments, highlighting the potential for elevated risks when multiple VOC groups are considered together [91]. Taken together, our results indicate that while individual BTEX exposures were below non-carcinogenic thresholds, benzene remains a compound of particular concern due to its consistently higher HQ values. Moreover, the presence of other VOC groups in urban air may enhance cumulative risks. Therefore, future risk assessments should not only include single-compound evaluations but incorporate mixture toxicity approaches to provide a more comprehensive understanding of population exposure in complex urban and/or industrial environments.

4. Conclusions

The nationwide year-long monitoring of BTEX compounds in Greece revealed pronounced spatial and seasonal variability, with elevated concentrations during colder months, driven by combustion-related activities and meteorological stagnation. Although the non-carcinogenic risk (HQ) for all compounds was within acceptable limits, the calculated benzene LTCR values frequently exceeded health-protective thresholds, especially during cold periods. Diagnostic ratios revealed the dynamic nature of emission sources, ranging from vehicular and industrial emissions to marine traffic. The elevated ozone formation potential of xylenes and toluene further illustrated their importance in urban photochemistry. These outcomes emphasize the critical need for source-specific air quality management policies that consider both direct toxicological effects and the secondary formation of pollutants, especially in urban environments where human exposure is continuous and multifaceted. Beyond its scientific contribution, this work also highlights the relevance of its findings for air-quality management in Greece. Although detailed policy design was beyond the scope of this study, the results underscore the importance of developing evidence-based strategies to reduce BTEX emissions and limit ozone formation, particularly in urban and port environments.