Airborne particulate matter in Tehran’s ambient air

Abstract

In recent decades, particulate matter (PM) concentrations in Tehran have exceeded the World Health Organization’s (WHO) guideline on most days. In this study, a search protocol was defined by identifying the keywords, to carry out a systematic review of the concentrations and composition of PM in Tehran’s ambient air. For this purpose, searches were done in Scopus, PubMed, and Web of Science in 2019. Among the founded articles (197 in Scopus, 61 in PubMed, and 153 in Web of Science). The results show that in Tehran, the annual average PM₁₀ exceeded the WHO guidelines and for more than 50.0% of the days, the PM_2.5 concentration was more than WHO 24-h guidance value. The PM concentration in Tehran has two seasonal peaks due to poorer dispersion and suspension from dry land, respectively. Tehran has two daily PM peaks due to traffic and changes in boundary-layer heights; one just after midnight and the other during morning rush hour. Indoor concentrations of PM10 and PM2.5 in Tehran were 10.6 and 21.8 times higher than the corresponding values in ambient air. Tehran represents a unique case of problems of controlling PM because of its geographical setting, emission sources, and land use. This review provided a comprehensive assessment for decision makers to assist them in making appropriate policy decisions to improve the air quality. Considering factors such as diversity of resources, temporal and spatial variations, and urban location is essential in developing control plans. Also future studies should focus more on PM reduction plans.

Introduction

In many megacities, rapid economic development coupled with population growth led to serious air pollution [1], and its deleterious effects on human health, visibility, climate, and the urban ecosystem [2–5]. Epidemiological studies have shown strong associations between air pollution and morbidity and mortality rates [6] such that air pollution is the fourth overall risk factor for human health. Globally, more than 80% of urban residents are exposed to air pollution level exceeding World Health Organization (WHO) guidelines [5]. These exposures lead to effects ranging from sub-clinical to premature mortality [3]. For example, 6.50 million premature deaths per year occur worldwide due to air pollution [5]. However, the effects of air pollution are dependent on pollution composition, duration of exposure, frequency of exposure, exposure concentrations, and toxic effects of specific constituents [3].

Particulate Matter (PM) has the highest impacts of any air pollutant on human health, and is associated with exacerbation of respiratory diseases (asthma and Chronic obstructive pulmonary disease (COPD)), allergies, respiratory infections, and cardiovascular diseases [1, 2, 5, 7–11]. PM with aerodynamic diameters ≤2.5 μm (PM_2.5) are strongly correlated with cardiovascular and respiratory effects while PM with diameters ≤10 μm (PM₁₀) have been generally associated with respiratory hospital admissions. Smaller PM such as PM1 and ultrafine particles (UFP) are more capable to penetrate pulmonary system, and therefore, more likely to induce serious health effects [12]. Chemical composition also effects in its toxicity [13–16]. Annually 3.10 million premature deaths in the world have been related to PM_2.5 exposure [7]. The Global Burden of Disease (GBD) in 2018 estimated 4.58 deaths and 143 million disability-adjusted life-years (DALYs) due to long-term exposures to ambient PM_2.5 and reported that long-term exposure to ambient PM_2.5 is globally the fifth highest risk factor for premature death [17].

Air pollution in Tehran also had negative social and economic impacts such that Tehran’s air pollution in recent decades is one of the reasons for the suggesting the need to change the capital of Iran [5]. Air pollution in Tehran has led to closures of schools on multiple days [7]. PM is the main cause of critical air pollution episodes in Tehran [2]. A report in 2012 mentioned that 4500 premature deaths in Tehran were associated with air pollution. Tehran has high estimated mortality due to long-term exposure to fine particular matter [7].

High indoor air pollution is also an important risk factor because 80% of one’s lifetime is spent in indoor environments resulting in extended exposures to indoor air pollutants [14]. Indoor air quality studies in Tehran were performed in a school, a retirement home, and a subway station, and confirmed the presence of PM contamination in these environments [14, 18–21]. Although the subway system is one of the most important facilities in Tehran, the concentration of PM in subway stations is a risk to the passengers’ health [22]. On the other hand, due to tobacco consumption in Tehran’s cafes, the concentration of PM_2.5 is at least 4 times than of the outdoor, and this ratio is at least triple for PM₁₀ [20].

This study has systematically reviewed reports of ambient PM measurements in Tehran considering their quantitative and qualitative aspects and temporal and spatial variations. Providing complete information on Tehran’s airborne particles will be useful to decision making, and highlight future challenges and research needs.

Methods

Study area

Tehran, Iran’s capital is a 73-km wide city located on the southern slope of the Alborz Mountains (at 1190 m, 35.7°N, 51.5°E). Its population is ~8.7 million people in 22 urban districts (Fig. 1). In recent decades, changes in land-use and land-cover, population growth, and the significant increase in motor vehicles has led to increasing air pollution causing significant risks to human health and the ecosystem [7, 23]. Air pollution is known as the most significant environmental problem in the Tehran metropolitan [24]. The most important pollution sources in Tehran include traffic, fossil-fueled power generation, industry, and re-suspension of soil. Motor vehicles were reported to contribute 85% of the particulate pollution [25]. Tehran’s geographic situation (surrounded with the Alborz Mountains) limits natural ventilation, enhances inversions in winter and the effects of Middle Eastern dust storm such that Tehran is one of the most polluted metropolises in the world [26]. The location of most industries to the west of Tehran leads to the transport of their emissions to the city since prevailing wind direction are southwesterly with speeds of 0.4 to 3.3 m/s [23].

Fig. 1.

Geographical location of Tehran and its land use

Search of literature

To find for related studies, in which the study of PM in Tehran’s ambient or indoor air had been investigated, three databases: Scopus, PubMed, and Web of knowledge were searched. Meanwhile, the search strings were used as follows: (“Tehran”) AND (“air”) AND (“particular matter” OR “particulate matter” OR “PM” OR “TSP” OR “suspended solids” OR “atmospheric aerosol”) OR “suspended particles” OR “PM₁₀” OR “PM₅” OR “PM_2.5” OR “PM₁” OR “PM_0.1”) (Table 1) This search protocol was used to identify publications in 2019.

Table 1.

Search protocol and the number of articles found in each database

Search protocol	Database	Articles
(TITLE-ABS-KEY (tehran)) AND (TITLE-ABS-KEY (air)) AND (((TITLE-ABS-KEY (“particular matter”) OR TITLE-ABS-KEY (“particulate matter”) OR TITLE-ABS-KEY (pm) OR TITLE-ABS-KEY (tsp) OR TITLE-ABS-KEY (“suspended solids”) OR TITLE-ABS-KEY (“atmospheric aerosol”) OR TITLE-ABS-KEY (PM10) OR TITLE-ABS-KEY (PM5) OR TITLE-ABS-KEY (PM2.5) OR TITLE-ABS-KEY (PM1) OR TITLE-ABS-KEY (pm0.1))) OR (TITLE-ABS-KEY (“suspended particles”)))	Scopus	197
	Web of science	153
	PubMed	61

Eligibility criteria

The final papers were chosen on the basis of the eligibility criteria; in this research, on those studies that had investigated PM in Tehran’s ambient or indoor air were exclusively focused. Hence, all studies that (1) mentioned PM concentrations, (2) investigated temporal or spatial variation of PM, (3) investigated health effects of PM, (4) investigated chemical speciation and source apportionment of PM were selected.

Study selection

The literature was screened independently by all the authors, based on the criteria mentioned above. After the initial screening of the titles, 132 studies were selected (Fig. 1). In the case of conflicting decisions over the initial screening, the respective study was included in the next step of screening. Finally, the contents of the articles were studied, and 56 articles were selected because in which containing any information on the quantity, composition, and effects of ambient or indoor PM have been included (Fig. 2).

Fig. 2.

Outlines of the screening process to finally select the 56 papers used in this study

Literature review

The different studies of Tehran’s air across the years reported amounts for multiple size ranges of PM and are presented in Table 2.

Table 2.

Concentrations of PM reported in the studies

location	TSP	PM2.5	PM10	year	Ref.
		Min-max (Mean)	Min-max (Mean)	Min-max (Mean)
Outdoor	–	33–147 (98)	13–73 (43.1)	2018	[20]
	–	–	38.37))	2015	[1]
	–	(90)	(33)	2015	[2]
	–	–	(32.2)	Winter 2015	[4]
	–	–	(26.8)	Summer 2015	[4]
	–	201–236	–	12–15 November 2016	[55]
	–	(348)	–	June 62,016	[55]
	–	106–560	–	2005	[23]
	(124)	–	–	Winter 2008	[24]
	(78)	–	–	Summer 2007	[24]
	<100 - >900	–	–	2009	[25]
	–	2.5–222 (65)	–	2007	[30]
	–	64–123 (79)	14–45 (19)	Summer 2010	[35]
	–	134–266 (194)	(71)	Winter 2011	[35]
	–	<40 - >100	–	Jan-Nov 2008	[40]
	–	<40 - >120	–	Jan-Nov 2009	[40]
	–	(89.5)	–	Winter 2014	[44]
	–	(75.4)	–	Summer 2013	[44]
	–	(140)	–	2016	[50]
	–	(284) dusty days	–	2016	[50]
	–	(200)inversion days	–	2016	[50]
	–	(123) regular days	–	2016	[50]
	–	(102)	–	8/2012–4/2013	[51]
	–	(336)	(210)	Winter 2007	[24]
	–	(61)	(95)	2017	[56]
	2016	(348) dusty days	–	–	[55]
	2016	(220)inversion days	–	–	[55]
	2010	–	(43)	–	[57]
	2017	–	(31)	–	[57]
	2013	–	–	8496 ton/year	[31]
	Mar 2017- Mar 2018	–	23–49	–	[58]
	Mar 2013- Mar 2014	–	(39)	–	[59]
	Mar 2014- Mar 2015	–	(36)	–	[59]
	Mar 2015- Mar 2016	–	(31)	–	[59]
	2012	46–200 (87)		–	[60]
	Jan 2010- Jan 2011	91	–	–	[61]
	2005	(122.1)	(24.3)	(239.8)	[29]
	2009	71.2))	–	–	[62]
	2010	(75.2)	–	–	[62]
	2011	(61.1)	–	–	[62]
	2012	(76.3)	(37.5)	–	[7]
	2010	(100.8)	–	–	[6]
	2011	60–140	–	–	[63]
	2002	65–370	–	–	[42]
	2017	27.2–244.9 (74.8)	8.4–77.9 (34)	–	[34]
	2008	70–100	–	–	[28]
	2007	21.5–187.8 (90.5)	–	72.7–271.6 (151.6)	[39]
Waterpipe café	2018	334–1167 (627.8)	130–560 (249)	–	[20]
Waterpipe and cigarette café	2018	843–1345 (1124)	434–780 (627.4)	–	[20]
Cigarette café	2018	123–305 (198.4)	67–174 (104)	–	[20]
Non-smoking café	2018	27–142 (66.8)	12–67 (29.2)	–	[20]
Underground subway station	2016	33–102	40–98	–	[22]
Elementary Schools	2009	185–366 (274)	33–50 (42)	–	[19]
Retirement home	May 2012–May 2013	8–72	4–27	–	[45]
school dormitory	May 2012–May 2013	7–126	5–38	–	[45]
Retirement home	May 2012- Jun 2013	–	18–87 (31)	–	[64]
school dormitory	May 2012- Jun 2013	–	10–118 (33)	–	[64]
School dormitory	2012	8.5–145 (53.7)	5–52.7 (19)	–	[14]
Subway system	2012	–	10–103 (49)	–	[54]
Underground subway station	2011	32.5–126.2 (94.4)	23.7–85.3 (52.3)	–	[48]
Surface subway station	2011	43.2–131 (87.6)	15.2–86.6 (41.3)	–	[48]

Shirazi and Harding (2001) reported that TSP showed average monthly concentrations rose from 74 μg/m³ with an uptrend to 140 μg/m³ between 1988 and 1993 [27]. However, they were noted that TSP concentrations oscillated during each year. In other periods, the PM concentrations declined. Faridi et al. (2018) reported days with PM _2.5 less than 25 μg/m³ increased from less than 20% in 2011 to more than 40% in 2015 [3]. Amini and his colleagues in 2010 stated that the whole population of Tehran was exposed to annual average PM₁₀ exceeding the WHO Air Quality Guideline (AQG) of 20 μg/m³ and also 89% of children under 5 years lived in areas exceeding the WHO interim target 1 (70 μg/m³) [6]. Faridi and colleagues also reported that PM_2.5 concentrations in Tehran between 2006 and 2015 exceeded 25 μg/m³ on >50% of the days.

Naddafi et al. (2012a) showed that PM₁₀ exceeded the WHO guidelines more than 200 times in 2009 (50 μg/m³ as 24-h mean). However, PM₁₀ only exceeded the national ambient air quality standards (NAAQS) of Iran (150 μg/m³ as 24-h mean PM₁₀) 14 times [28]. While Faridi et al. (2018) showed that the annual average PM_2.5 concentration was at least twice the US EPA and WHO standards for every year between 2006 and 2015 [3]. In 2005 the daily concentrations of PM₁₀ and TSP exceeded the NAAQS standard for 25.7% and 38.9% of samples, respectively [29]. The data for a station in the center of Tehran in 2007 showed an annual average of PM₁₀ of 64.9 μg/m³, while minimum and maximum values were 2.5 μg/m³ and 222.5 μg/m³ respectively, and a standard deviation of 31.1 [30].

Result and discussion

PM concentrations

These results indicate that PM concentrations have varied over time. However, as Table 2 shows that all of the reported mean PM concentrations were higher than the WHO guidelines. The number of days since 2011 with concentrations <25 μg/m³ increased, indicating improving conditions in recent years. They showed that the ratio of days with PM_2.5 > 25 μg/m³ in 2006 to 2010 increase with time, and the decreased from 2011 to 2015 although the ratio never was <50% [3]. For most days of the year, the PM concentrations in Tehran exceed the WHO guideline. However, the national ambient air quality standards (NAAQS) of Iran are only exceeded for few days a year.

The PM concentrations in Tehran varied widely across time and location. Results indicate that, in 2007, in Tehran’s central region, the concentration of PM₁₀ for the best condition, worst condition, and annual average were 10 times less, 10 times more, and 3 times more than the WHO guidelines, respectively [30].

The PM emission sources are mainly traffic, industries, power plants, and residential space heating for fine PM [7, 24]. An important source of coarse PM is suspension of soil grains from dry lands [7]. The share from mobile sources is higher than from other sources with 70% of PM in Tehran in 2013 reportedly related to mobile sources. Annually, more than 400 tons of mobile source PM were emitted [5]. Shahbazi et al. (2019) also estimated that in 2013, 8496 tons of PM were emitted in Tehran of which 69.8% were from motor vehicles with residential-commercial and industrial sources contributing 2.2% and 7.2%, respectively [5]. They also reported that buses and trucks were the largest emitters of PM with 25,896 and 1920 tons of PM, respectively. Taxis and pickup trucks emitted 23 and 41 tons per year, respectively [31]. In another study, the share from personal cars, taxis, trucks, motorcycles, municipal buses, and minibuses to mobile source PM emissions were reported to be 2.5%, 0.7%, 25%, 20%, 21% And 5%, respectively [5]. Therefore, near road areas in Tehran have higher PM concentrations than other areas. Yazdi et al. (2015) found the concentration of PM₁₀ along the Hemmat highway under heavy traffic conditions was up to 120 μg/m³, and even with light traffic was approximately 100 μg/m³. They showed the PM concentration declined by 50% at a distance of 150 m from the highway and this decline continued with a lower slope of up to 300 m. At 300 m, the PM concentration had been reduced to 40 μg/m³, which they referred to “background concentration”. A similar trend was observed for PM_2.5, except that for high traffic conditions, the maximum concentration was 13 μg/m³, which declined by 30% at 100-m from the highway. After 200-m, the PM_2.5 concentration was approximately constant with a background concentration of 8 μg/m³ at 300-m from highway [32]. Driving with closed windows provides reduced exposure to PM. A study on the Tehran Resalat Highway showed that the average in-cabin exposure to particle number and PM₁₀ for open windows was seven times higher when compared to closed windows [33]. Tunnels have higher PM than the highways [33]. The in-cabin PM when driving through the Resalat Tunnel was 30% higher than driving on the Resalat Highway.

Temporal variations

Alizadeh-Choobari et al. (2016) reported peaks in PM concentration in summer and winter. The summer peal was caused by dryness and resulting PM suspension from nearby dry land. In winter, near ground emissions accumulate due to thermal inversions and the naturally low ventilation caused by the geographical characterizations [7]. Arhami et al. (2017) showed that the average PM₁₀ concentration in the western part of Tehran in the summer was 16.3 unit higher than the winter, while the average PM_2.5 concentration in this region in the winter was higher than in the summer [2]. This difference was attributed coarse particle dust (PM₁₀) being suspended in summer from dry soils while during winter, PM_2.5 increased by accumulation of near ground urban sources such as vehicles given the lower boundary-layer heights and wind speeds in winter [2]. Similar results were reported for the winter of 2008 compared to the summer of 2007 [24]. In the warm season, due to the intensity of the emissions from nearby dry lands and middle east dust storms transported to Tehran, the PM₁₀ concentration increased [3, 34] with increasing coarse PM/fine PM ratios, with this ratio in some cases being up to 4 [34].

The seasonal variation of PM concentrations changes the ratio of PM_2.5 to PM₁₀. The results in the western and central regions of Tehran in the summer of 2010 and winter of 2011 found that the mean PM₁₀ concentration in winter (193.86 μg/m³) was greater than the summer mean (95.72 μg/m³). The proportions of PM_2.5 and PM₁ in winter were 36.5% and 25.9%, respectively while in summer, they were 23.7% and 13.7%, respectively indicating an increased fraction of fine PM in the cold season [35]. Jaafari et al. (2018) found that the average PM₁₀ concentrations in Tehran in 2017 on dust storm days, in the winter, and during summer were 222.6, 79.6, and 62 μg/m³, respectively. The corresponding mean PM_2.5 values were 58, 424, and 20.6 μg/m³, respectively. During the cold season, PM emissions from dust sources decrease, but thermal inversion and lower wind speeds lead to accumulation of PM from stationary sources and traffic such that PM concentrations increases with the PM fine/PM coarse ratio decreasing, and in some cases, reported to be ~47% [34].

Meteorological and environmental parameters affect PM concentrations and these effects were evaluated by Hajiloo et al. (2018) and Halek et al. (2004). Hajiloo et al. (2018) reported that PM_2.5 concentration had a negative relationship with vegetation [4]. Increases in temperature in both summer and winter season were positively related to the PM_2.5 concentration. This result for winter is unusual since most studies find higher PM concentrations at lower temperatures when inversions tend to be stronger, Increase wind speed resulted in a decrease in the concentration of PM in both summer and winter. In winter increased wind would provide dilution and lower concentrations. However, in summer, increased wind speeds would be expected to increase dust suspension [36] since the suspension rate is approximately proportional to the cubic power of the wind speed [37, 38]. Rainfall of more than two millimeters reduced PM_2.5 [4]. Leili et al. (2007), showed that with an increase in the average monthly temperature, TSP concentrations increased because of increased photochemical activity and secondary particle formation [39]. In general, variations in boundary-layer heights in different seasons, variations in PM emissions from dry lands, variations in vehicle traffic, and variations in rainfall rates affect the temporal variations of PM [7].

Different PM concentrations were observed in different months. A 10-year study of PM_2.5 in Tehran between 2006 and 2015 showed that in December and January and in hot months including July, PM concentrations were highest. The minimum PM concentrations were measured in March due to the Iranian New Year holiday. Monthly average PM_2.5 concentrations were reported to be >20 μg/m³ for all months [3]. During the Iranian New Year holidays in late March to early April, traffic in Tehran drastically decreases. Dust storms rarely occur during these months. These conditions result in lower PM concentrations in this period relative to other months.

Sowlat et al. found that average monthly PM₁₀ concentrations ranged 60 μg/m³ to 80 μg/m³ for selected stations in March 2008. However, in March 2009, they ranged from 60 μg/m³ to 130 μg/m³. The PM trend for five stations differed between May to September 2008 (declined at two stations and rose at three stations). During this period, maximum PM concentrations were reported for all stations in March 2009 contradicting previous studies [40]. A study of PM₁₀ in Tehran in 2004–2014 indicated that the annual average PM₁₀ concentration was between 67 μg/m³ and 93 μg/m³, and the concentration variations did not follow a monotonic trend. In 50% of years, the warm season average was higher than cold season. The maximum concentration in the cold season was reported in 2009 (213 μg/m³) while the maximum concentration of the warm season was reported in 2008 (758 μg/m³) [41].

Halek et al. (2004) concluded that the PM concentrations in Tehran had two peaks; one in the afternoon and early evening (15:00 to 20:00) despite the expansion of the boundary layer at these hours, and in the morning (06:00 to 09:30) due to increased traffic volume [42]. Hourly variations in the PM concentrations were subsequently investigated by Faridi et al. (2018) in a PM_2.5 study between 2006 and 2015. They found that PM_2.5 has a diel pattern with two maxima. The smaller peak occurs around 07:00 to 08:00 suggesting the influence of morning rush hour traffic [3]. The larger maxima occurred around midnight to 02:00 likely because of the lowering mixed layer height and wind speed while there were still sufficient emissions to affect the ambient concentrations. Hourly variation in the concentration of PM is also affected by traffic condition and vehicle types traffic as well as atmospheric conditions. As the results of the studies show, the PM concentration rises in the evening until 02:00 due to the end of the traffic limitations in the evening for heavy vehicles in Tehran and the decline of the mixing layer height and decrease in wind speeds during these hours. In the morning hours, increasing PM concentrations was observed due to increased traffic. By midday, traffic has reduced and the rising mixing layer height reduces the PM concentration [3, 7, 28, 42]. Highest concentrations were observed on Mondays and Tuesdays with lowest values on Fridays (Muslim Sabbath). Naddafi et al. (2012b) showed that the increasing trend in the concentration of PM₁₀ between 22:00 and 04:00 was caused by increased movement of heavy vehicles because of the end of restrictions on them [28]. considering that the weekend in Iran is Thursday and Friday, the concentration of PM in Tehran’s air on these days was lower than other days, while middle days of the week (Monday and Tuesday), have a more concentration of PM than other days [3], which can be due to traffic condition in these days.

Spatial variations

Tehran has varying population densities, land use, and official and business centers, which produce different PM concentrations across the city. The results of a 2005 study showed significant differences in PM concentrations between northern and eastern Tehran, while the PM concentrations in central, southern, and western Tehran were approximately equal but more than the eastern are and less than northern Tehran [3]. However, a more recent study indicated the PM concentrations in southern Tehran were higher compared with the center and the center being higher than the northern area. Western Tehran had higher PM concentrations than eastern Tehran [7]. One reason for the differences in reported spatial variations may be the effect of different temporal patterns in different locations. For example, at midnight the PM concentration in northern Tehran is higher, while at midday the concentration of PM in southern Tehran is higher [7]. This difference was caused by differences in emission sources. Traffic is the main source of PM in northern Tehran and increases in the evening, while industries, especially the sand industry, in southern Tehran are active during the day [6]. However, PM concentrations at midday in southern Tehran were higher than at midnight due to resuspension from nearby dry land and local industrial activities such as a sand industry. A study in 2010 categorized the PM₁₀ concentration into 8 groups and found that an increasing density of streets, proximity to the bus station and the airport or air cargo facilities increased the PM₁₀ concentrations. This study showed PM₁₀ concentrations in eastern, south-eastern, and central Tehran were higher than in other areas with northern and western regions having minimum concentrations of PM₁₀, also PM concentrations increased with increasing population density in Tehran’s different urban areas [6].

Chemical speciation and source apportionment

There have been some PM chemical speciation studies in Tehran. The first was performed on TSP samples collected from March 1994 to February 1995. Concentrations of 21 elements were reported among which the highest concentrations for Ca, Al, K, and Na were reported by 6.96 μg/m³, 3.27 μg/m³, 2.15 μg/m³, and 1.68 μg/m³ respectively [43]. In 2007, the concentrations of lead, cadmium, and chromium associated with PM₁₀ in the center of Tehran were reported to be 39.6–485.3 ng/m³ (mean: 132.5 ng/m³), 3.3–10.6 ng/m³ (mean: 6.8 ng/m³), and 5.1–15.8 ng/m³ (mean: 9.1 ng/m³), respectively (Leili et al., 2008). Increasing concentrations of Al, Si, K, Ca, Ti, Mn, and Fe in were found during the warm season because of the dust source. Increasing concentrations of Cu, Zn, and Pb in the cold season are related by traffic and the declining dispersion conditions [25].

Toxic compounds such as PAHs associated with PM can increase health risks from exposure to PM. A study in 2005 at 15 locations in Tehran reported that the average cumulative PAH concentration in PM was 380 ng/m³ (13 to 526 ng/m³), while the mean concentration of seven carcinogenic PAHs was 43 ng/m³ [23]. Three and four-ring compounds were the major PAH species with acenaphtene, acenaphtylene, naphthalene, fluoranthene, fluorene, anthracene, and phenanthrene being predominant [23]. However, Mohseni Bandpi et al. (2017) found that in 2015, two-ring and three-ring compounds were generally the main PAHs in PM_2.5, but in traffic areas, six-ring and three-ring compounds were the most abundant PAHs, respectively. Low molecular weight PAHs including naphthalene, acenaphtylene, acenaphtene, fluorene, phenanthrene, and anthracene were 17.3% greater than the high molecular weight PAHs in summer 2013 and winter 2014 (76.44 and 65.13 ng/m³, respectively) [44]. Motor vehicles were the most important source of PAHs in PM [45]. Fluorene, naphthalene, benzo[k]fluoranthene, pyrene, and phenanthrene had the highest concentrations in PM [44]. The mean total PAH concentration in winter (148.4 ng/m³) was higher than summer (83 ng/m³) due to increased fossil fuel combustion for heating, reduced photochemical degradation of PAHs in winter, and poorer dispersion in winter [44]. In 2015, the average PAH concentrations in PM_2.5 measured in traffic and urban areas were 43.4 ng/m³ and 50.5 ng/m³ respectively. In the urban areas, phenanthrene, acenaphthylene, and benzo(b)fluoranthene had average concentrations of 4.92, 4.38, and 4.30 ng/m³, respectively, and represented the most abundant PAHs. The highest PAHs in high traffic areas were acenaphthylene, chrysene, and pyrene with average concentration ​​of 4.08, 3.82 and 3.68 ng/m³, respectively [46].

Fossil fuel combustion, traffic, and dust from nearby dry land were the main sources of PM in Tehran’s ambient air [24]. One method for determining PM source apportionment uses the chemical speciation of particles. In 2015, Al and Fe were the most abundant metals associated with PM_2.5 [47]. The Fe concentration in traffic and urban areas were reported as 1206 and 1032 ng/m³, respectively, while Al in the traffic and urban areas were 1606 and 1984 ng/m³, respectively. It was determined that these two metals in urban and traffic areas respective 79.2 and 85.7% of the total metals associated with PM_2.5 [47]. Heavy metals associated with PM_2.5 were due to sources such as industries in a 20 km radius of Tehran, motor vehicles, and resuspension of metals such as iron and aluminum in soil and road dust [46, 47]. High proportions (>70%) of organic and elemental carbon were measured in PM_2.5 in the cold season. Also, the significant increase in iron in PM measured at a subway station platform showed that the main source of PM here was the steel wheels of the train ablating the steel tracks [21].

Anthropogenic urban sources (i.e. vehicles) was reported as the main source of PM in Tehran. Dust was estimated to be 56% of PM_2.5 in the warm season due to suspension from nearby dray lands as its main source [2]. They reported that the most important heavy metals in PM_2.5 were zinc and lead (0.23 and 0.19 μg/m³, respectively) and concluded the origin of these metals were brake abrasion, tire wear, fuel combustion (mainly heavy oil), small industries, and lubricating oil [2]. Their study showed that the annual average of chemical speciation of PM_2.5 in Tehran from February 2014 to February 2015 consisted of organic matter (35%), dust (25%), non-sea salt sulfate (11%), EC (9%), ammonium (5%), nitrate (2%), and unidentified matter (11%) [48]. Esmaili et al. (2014) noted that Al, Si, K, Ca, Ti, Mn, and Fe came from dust sources with increased concentrations during the warm season. They identified the relationships of Cu, Zn, and Pb with traffic, and concluded increased concentrations of these elements in winter were due to increased PM from fuel consumption and poorer dispersion [25]. Mohseni Bandpi et al. (2017) showed that in 2015, the sum of SO₂⁻⁴ and NO₃⁻ in traffic areas and urban areas make up 34% (3.58 μg/m³ SO₂⁻⁴ + 2.12 μg/m³ NO₃⁻) and 28% (3.69 μg/m³ SO₂⁻⁴ + 2.77 μg/m³ NO⁻³) of ionic species of PM_2.5, respectively, and were related to motor vehicle emissions [47]. A 2019 study conducted by Tehran Air Quality Control Company on gasoline and diesel quality shows both fuels sold in Tehran contained sulfur in excess of the standards with diesel fuel S at 80 ppm and gasoline at 180 ppm (https://financialtribune.com/articles/environment/99944/gasoline-diesel-sulfur-content-larming). Internal combustion engines release SO₃ because of the cooling during the power stroke and thus, are sources of primary sulfate [49]. However, Rezaei et al. (2018) found Na⁺, Mg²⁺, Ca²⁺, Al, Si, Fe, and Ti dominant for PM₁₀ [50]. Kholdebarin et al. (2015) examined PM₁₀ in 4 areas by analyzing metallic elements including 15 elements to apportion sources in Tehran’s PM. They found average concentrations for Pb, Mn, Ni, As, and V of 44.2, 10.0, 7.7, 1.36, and 1.3 ng/m³ respectively, with all elements below the WHO health guidelines (WHO 2000) [51]. The metallic element based apportionment of PM₁₀ in four urban areas [2, 20, 28, 29] were road dust (tire wear, asphalt pavements, and brake linings), motor vehicle tailpipe emissions, and industries, with proportion of 95.4, 4.05 and 0.4% respectively [51]. Taghvaee et al. (2018) used positive matrix factorization (PMF) to identify vehicles, industries, secondary aerosol, biomass burning, soil, and road dust (including tire and brake wear particles) as sources of PM_2.5 in two areas in Tehran. Motor vehicles contributed 49% while secondary aerosol contributed 24% and 28% in the two locations. Less than 1% of PM_2.5 was from road dust [52]. Halek et al. (2010a) studied the chemical speciation of PM in four areas in northwestern Tehran in winter 2007 and showed that Zn, Mg, and Fe were the elements most commonly associated with PM₁₀ and PM_2.5. They concluded these elements were from industries within a forty km radius of the studied area, zinc and iron in local soil, and magnesium in the carbonate compounds of nearby areas [53].

Indoor PM

In tobacco cafes in Tehran due to smoking hookah and cigarettes, the concentration of indoor PM is greater than the ambient values. A 2019 study showed that the PM_2.5 concentrations in a water pipe café, a cigarette café, and a water pipe/cigarette café were 8.6, 3.6, and 21 times higher than the contemporaneous outdoor values, respectively. The PM₁₀ concentrations were 6.6, 2.2, and 12.8 times higher than the corresponding outdoor concentrations, respectively [20]. However, in a non-smoking café, the indoor PM_2.5 and PM₁₀ concentrations were lower than ambient by 4% and 11%, respectively [20]. The number of active smokers, tobacco types, and ventilation rates were important factors in the indoor PM quality. However, the number of active smokers in the café was the most important factor in the concentration of indoor PM based on the analysis of modulus standardized effect sizes (MSES) [20].

Measurement of the metallic elements in PM outside and inside a school dormitory and a retirement home between May 2012 and May 2013 found higher ambient concentrations of PM₁, PM_2.5, and PM₁₀ than the corresponding indoor values. Ambient Si (78.57–771.58 ng/m³), Fe (60.11–695.93 ng/m³), Zn (57.29–608.47 ng/m³), Al (14.76–265.57 ng/m³), and Pb (41.89–120.92 ng/m³) were the highest concentration species among the twenty studied metals. The indoor total metal(loid) (sum of the measured species) concentrations associated with PM₁₀, PM_2.5, and PM₁ measured at the two sampling sites ranged from 0.4 to 2.2, 0.2 to 1.1, and 0.1 to 0.6 μg/m³, respectively. The corresponding outdoor values ranged from 0.8 to 4.2, 0.4 to 1.8, and 0.2 to 1.1 μg/m³. The retirement home had lower metal concentrations than the school dormitory [45]. Halek et al.(2013) studied five primary schools in Tehran in the winter of 2009, and found PM₁ concentrations in the classrooms were lower than outdoor. However, PM_2.5 and PM₁₀ concentrations were higher than ambient. They concluded that the important parameters affecting the indoor PM concentrations were the outdoor PM values and the physical activity of the students. The temperature gradient between indoors and outdoors, the type of construction, and the class size had little effect on the indoor PM values [19].

Subway stations in Tehran have higher PM than in the ambient. Bolourchi et al. (2018) measured concentration PM₁₀ and PM_2.5 at the in Imam Khomeini station and found stated the concentrations of indoor PM₁₀ and PM_2.5 were 2.5 and 2.9 times higher than the respective outdoor values. PM₁₀ and PM_2.5 in the subway station were affected by the passing trains, such as on the underground platforms, in underground tunnels and on elevated tracks [22]. Kamani et al. (2014) studied PM in Tehran subway stations and found Al, Ca, Fe, V, and Ti in PM₁₀, while Ba, Cr, Cu, Pb, and Zn were a larger fraction of the PM_2.5 [48]. These results as well as other studies are presented in Table 3.

Table 3.

The outdoor/indoor ratio in different places in Tehran

location	year	Outdoor to indoor ratio (%)			Ref.
		PM1	PM2.5	PM10
Elementary Schools	2009	116	90.5	51.1	[19]
waterpipe café	2018	–	15.9	19.9	[20]
Waterpipe and cigarette café	2018	–	6.7	9.5	[20]
cigarette café	2018	–	36.9	56.1	[20]
Non-smoking café	2018	–	147	147	[20])
School dormitory	2013	232	173	139	[14]
Retirement home	2013	191	171.2	200	[14]
Underground subway station	2015	–	40	34.5	[22]
Underground subway station	2011	–	66.9	86.6	[48]
Surface subway station	2011	–	72.6	84.4	[48]
Underground subway station	2014	73.5	–	–	[21]
Surface subway station	2014	82.8	–	–	[21]

Various indoor environments have been categorize into three groups. The first group includes places that have sources of PM emissions and resulting high PM concentrations of PM. In these locations, concentrations of indoor PM are much higher than the ambient air. Tehran’s cafes, especially water pipe or cigarette cafes are the most important examples of this group (Fig. 3).

Fig. 3.

Concentration of indoor PM and outdoor PM in Tehran cafes in weekends. Adapted from [20]

The second group includes locations that have the potential for PM emissions, but not at the rates of the cafes in the first group. Examples of this group are subway stations. PM emission arise from the contact between the trains and the rails [21]. The ventilation systems in these locations and their operating conditions are the driving factor in reducing the PM concentrations [21, 54]. The third group incorporates places that have very low indoor PM emissions. Schools and retirement homes were two examples of this group that were studied in Tehran. The PM concentrations PM in this group were primarily affected by the outdoor PM concentrations since the typical activities in these places do not have significant potentials for PM emissions [19].

An important aspect in policy-making for PM control in Tehran is attention to vehicles as the main PM source. Trucks, buses, motorbikes, and minibuses contribute the highest PM emissions to Tehran. Thus, policies that drive renovation or replacement of these vehicles with lower emitting vehicles has the best potential for effective PM control. Shahbazi et al. (2019) concluded that 70% (650 tons a year) of the decrease in PM emissions were related to the renovation of minibuses and municipal buses. Their results also showed that other policies such as traffic restrictions in different parts of the city had less impact on PM reductions [5].

Strengths and limitations

The present study has some strengths and limitations. This is the first study investigating the spatial and temporal variations, possible sources, and indoor and outdoor concentrations of PM in Tehran as a megacity in the Middle East. The systematic nature of this review allowed us to collect and assess all the relevant studies. Still there are some limitations that should be acknowledged. First, other types of studies conducted in Tehran on other aspects of PM such as epidemiology of health effects, burden of diseases, and toxicity of PM could have been included in this study. Also Some databases, such as Embase and Medline could have been included in this study. Need to quality assessment of the selected papers and statistical approach are other limitations that should be acknowledged. However, these topics require a separate comprehensive study and we preferred to focus only on the concentrations and composition of particulate matter. In addition, the insufficiency in the number and quality of conducted studies could have affected our interpretation of the current situation. We were aware of this limitation and tried to propose the future direction of studies to overcome this limitation in the future.

Conclusion

This study systematically reviewed all the published articles about the concentrations and composition of PM in Tehran’s ambient air. The study showed that Tehran’s ambient air is severely polluted by PM that is emitted from vehicles (mainly in the cold seasons) and dust sources (mainly in the warm seasons), leading to PM concentrations that continuously exceed the WHO guidelines, causing remarkable burden to the public health. Central, southern, and western parts of the city have higher concentrations of PM comparing to the other areas. In addition to seasonal and monthly temporal variations, special hourly patterns were observed. These seasonal, monthly, and hourly variations of PM can be used by the authorities to control the peaks of air pollution in Tehran. Fossil fuels and traffic are among the major sources of PM in this city. Proper strategies and plans can be set and designed based on our results to reduce the overall concentrations of particulate matter, and also to decrease the population exposure, especially in areas with higher concentrations. Indoor spaces in Tehran can be classified into three groups including the high, medium, and low PM concentrations. Tehran is a useful example of air pollution and PM issues. Prior researchers have focused on PM concentrations and in some cases on the identification of PM compositions. However, there has yet to be a comprehensive source apportionment study and thus, uncertainty remains with respect to the importance of sources other than motor vehicles and suspended dust. This could be a critical area for future studies. Another direction of studies could be modelling of PM concentrations across the city in higher resolutions. Such studies would determine the hot spots of air pollution more precisely. In addition, there is a lack of studies investigating the impact of different air pollution control policies and strategies on the air quality. These studies can assess the benefits and drawbacks of the possible intervention scenarios, and can serve as a basis for practical solutions.

Funding

This study was funded by Iran University of Medical Sciences (Grant Number: 98–4–61-16815).

Compliance with ethical standards

Declarations of interest

None.