Assessment of personal exposure to particulate air pollution during commuting in European cities—Recommendations and policy implications

Science of the Total Environment 490 (2014) 785–797

Contents lists available at ScienceDirect

Science of the Total Environment journal homepage: www.elsevier.com/locate/scitotenv

Review

Assessment of personal exposure to particulate air pollution during commuting in European cities—Recommendations and policy implications Angeliki Karanasiou a,⁎, Mar Viana a, Xavier Querol a, Teresa Moreno a, Frank de Leeuw b a b

Institute of Environmental Assessment and Water Research (IDAEA-CSIC), Barcelona, Spain National Institute for Public health and the Environment (RIVM), Bilthoven, The Netherlands

H I G H L I G H T S • Car commuter’s exposure depends on traffic intensity and emissions by nearby vehicles • Cyclists are exposed to lower PM levels in comparison to those inside vehicles • Renovation of public vehicles will reduce commuter’s exposure

a r t i c l e

i n f o

a b s t r a c t

Article history: Received 17 December 2013 Received in revised form 28 April 2014 Accepted 11 May 2014 Available online xxxx

Commuting is considered as one of the high-exposure periods among various daily activities, especially in high vehicle-density metropolitan areas. There is a growing awareness of the need to change our transportation habits by reducing our use of cars and shifting instead to active transport, i.e. walking or cycling. A review was undertaken using the ISI web of knowledge database with the objective to better understand personal exposure during commuting by different modes of transport, and to suggest potential strategies to minimise exposure. The air pollutants studied include particulate matter, PM black carbon, BC and particle number concentration. We focused only in European studies in order to have comparable situation in terms of vehicle fleet and policy regulations applied. Studies on personal exposure to air pollutants during car commuting are more numerous than those dealing with other types of transport, and typically conclude by emphasising that travelling by car involves exposure to relatively high particulate matter, PM exposure concentrations. Thus, compared to other transport methods, travelling by car has been shown to involve exposure both to higher PM and BC as compared with cycling. Widespread dependence on private car transport has produced a significant daily health threat to the urban commuter. However, a forward-looking, integrated transport policy, involving the phased renovation of existing public vehicles and the withdrawal of the more polluting private vehicles, combined with incentives to use public transport and the encouragement of commuter physical exercise, would reduce commuters' exposure. © 2014 Elsevier B.V. All rights reserved.

Editor: P. Kassomenos Keywords: Ultrafine particles Traffic Particle number concentration Black carbon Cycling Commuter

Contents 1. 2.

3.

Introduction . . . . . . . . . . . . . Materials and methods . . . . . . . . 2.1. Study identification and selection 2.1.1. Air pollutants studied . Results and discussion . . . . . . . . 3.1. Exposure during cycling . . . . 3.2. Exposure in cars . . . . . . . . 3.3. Exposure in buses . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

⁎ Corresponding author. E-mail address: [email protected] (A. Karanasiou).

http://dx.doi.org/10.1016/j.scitotenv.2014.05.036 0048-9697/© 2014 Elsevier B.V. All rights reserved.

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

786 786 786 786 787 787 787 787

786

A. Karanasiou et al. / Science of the Total Environment 490 (2014) 785–797

3.4. 3.5. 3.6.

Exposure in underground train systems . . . . . Comparison studies . . . . . . . . . . . . . . Other parameters investigated than transport mode 3.6.1. Traffic conditions and route . . . . . . 3.6.2. Travel speed . . . . . . . . . . . . . 3.6.3. Between-vehicle distance . . . . . . . 3.6.4. Ventilation . . . . . . . . . . . . . . 3.6.5. Fuel type . . . . . . . . . . . . . . . 3.6.6. Meteorological conditions . . . . . . . 3.6.7. Other compounds studied . . . . . . . 4. Conclusions and recommendations . . . . . . . . . . . Acknowledgements . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

1. Introduction The association between traffic-related air pollution and health is becoming well established and documented from both epidemiological and toxicological studies (WHO, 2005, 2013). Exposure to particulate matter, PM can cause respiratory diseases, trigger cardiovascular morbidity and mortality after long-term and short-term periods (Anderson et al., 2012). Black carbon, BC is considered to be a better indicator of harmful particulate substances from combustion sources (especially traffic) than undifferentiated PM mass (Janssen et al., 2012) and is strongly associated with health outcomes in epidemiological studies (Heal et al., 2012). Another parameter that has drawn the attention of the research community due to its association with adverse health effects is the ultrafine particle UFP number concentration. Toxicological and laboratory studies have demonstrated cardiovascular and respiratory health effects of UFP, which likely have different and partly independent effects from larger particles, due to their small size, large surface area, different chemical composition and ability to penetrate deep into the alveolar system (Hoek et al., 2010). Commuting is considered as one of the high-exposure periods among various daily activities, especially in high vehicle-density metropolitan areas (Duci et al., 2003). The report from the World Health Organization on the health effects of traffic-related air pollution points out that people spend 1–1.5 h/day commuting in many countries (WHO, 2005). Furthermore the levels of most air pollutants are particularly high along busy roads, common in urban transport environments and their concentrations peak during morning commute hours (Morawska et al., 2008; Moreno et al., 2009). As pollutants concentration are often elevated in the traffic microenvironment, individuals may gain a significant contribution to their daily exposure when commuting in traffic even though such individuals usually travel for no more than 6–8% time of the day (Kaur et al., 2007). This is confirmed by many studies demonstrating that commuting accounts for high contributions in total personal exposure. Indeed, during their regular journeys commuters can receive up to 30% of their inhaled daily dose of BC, and approximately 12% of their daily PM2.5 personal exposure, (Dons et al., 2011, 2012; Fondelli et al., 2008). In addition to decreasing emissions and keep on the efforts to decrease concentrations, one potential solution would be to reduce personal exposure by managing the actual exposure. The primary aim of this paper is to review the studies performed to date in order to better understand exposure to key air pollutants (PM, BC and UFP) during commuting by different modes of transport, and to suggest potential strategies to minimise personal exposure. UFP typically constitute 90% or more of particle number concentrations in areas influenced by traffic emissions (Morawska et al., 2008) thus in this paper we use particle number concentrations to describe UFP. We focus only in European studies in order to have comparable situation in terms of vehicle fleet and policy regulations applied. For example large shifts to diesel fuels in European cities in the last decade are considered to be a cause of stable (not lower) PM10 levels in European cities and no decline in the health impacts of air pollution — despite the introduction of cleaner

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

793 793 794 794 794 795 795 795 795 795 795 796 796

diesel technologies (WHO, 2005). As most of the population in Europe lives in urban areas (73% according to the United Nations, World Urbanization Prospects, 2011 Revision) the studies examined have concentrated on the urban scale, although it is also important to take into account the exposure of the population living in more rural areas. The commuting modes that we selected include car, bus, bicycling, and subway. Exposure during walking was not examined as this transport mode is mainly used for short trips or is part of other transport modes e.g.: combination of public transport with walking to transit stations. We note that dose assessment which is a complementary yet distinct concept to that of exposure is not the focus of this review. In the present study we do not examine the inhaled dose of pollutants during commuting due to the lack of studies determining it and also its complexity of interacting factors such as breathing rate, ventilation and/or particle deposition to the respiratory system. 2. Materials and methods 2.1. Study identification and selection An electronic search on the ISI web of knowledge database and Google Scholar was conducted using various combinations: “air pollutants”, “black carbon”, “elemental carbon”, “ultrafine particle”, “transport mode”, “commuter”, “exposure” “public transport”, “microenvironment”, “vehicle”, “car”, “automobile”, “bus”, “cyclist”, “bicycle”, “underground system”, “metro”, “subway” without restrictions of publication type or publication date. The reference lists of studies identified by this method were reviewed for links to additional literature. In addition recent articles in relevant journals were collected. Only the studies concerning exposure measurements conducted in Europe are included in this paper. We present the results of the exposure studies performed across 4 transport modes: car, bicycle, bus, and subway focused solely on PM, EC or BC and particle number concentrations. 2.1.1. Air pollutants studied The main air pollutants that have been determined in different commuting environments include: • PM mass concentrations (Aarnio et al., 2005; Adams et al., 2001a, 2002; Alm et al., 1999; Asmi et al., 2009; Berghmans et al., 2009; Boogaard et al., 2009; Boudet et al., 1998; Braniš et al., 2006; Briggs et al., 2008; Colombi et al., 2013; de Nazelle et al., 2012; Dennekamp et al., 2002; Diapouli et al., 2008; Fondelli et al., 2008; Gee et al., 1999; Gee and Raper, 1999; Geiss et al., 2010; Gulliver and Briggs, 2007; Int Panis et al., 2010; Jacobs et al., 2010; Johansson and Johansson, 2003; Kingham et al., 1998; McNabola et al., 2008; Molle et al., 2013; Querol et al., 2012; Raut et al., 2009; Ripanucci et al., 2006; Salma et al., 2007; Seaton et al., 2005; Strak et al., 2010; Pfeifer et al., 1999; Rank et al., 2001; Zuurbier et al., 2010). • Black carbon, BC or elemental carbon, EC (Adams et al., 2002; de Nazelle et al., 2012; Dons et al., 2012, 2013; Fromme et al., 1998; Le Moulle et al., 1998).

A. Karanasiou et al. / Science of the Total Environment 490 (2014) 785–797

• Particle number concentrations, PNC (Aarnio et al., 2005; Asmi et al., 2009; Berghmans et al., 2009; Boogaard et al., 2009; Diapouli et al., 2008; Kaur et al., 2006; Molle et al., 2013; Ragettli et al., 2013). 3. Results and discussion Tables 1–4 summarise the studies reporting on PM, PNC and BC exposure during commuting by 4 different modes: bicycle, car, bus and metro. Different instruments have been used for the PM measurements including optical devices and gravimetric measurements making difficult the comparison. For example in the study of Zuurbier et al. (2010) PM10 was measured gravimetrically while PM2.5 by an optical monitor resulting in PM2.5 levels higher than PM10 concentrations. In some studies the optical instruments were compared using the gravimetric method (Molle et al., 2013; Int Panis et al., 2010) while some researchers provide the ambient concentrations in fixed monitoring stations (Asmi et al., 2009; Briggs et al., 2008; Kaur et al., 2005; Gulliver and Briggs 2004). Similarly for BC concentrations different metrics and techniques have been used by researchers; de Nazelle et al. (2012) and Dons et al. (2012) used the micro-aethalometer to measure BC concentration in PM2.5, Zuurbier et al. (2010) report on the absorption of soot particles determined by means of the reflectometer while Fromme et al. (1998) and Adams et al. (2002) report on the elemental carbon EC concentrations. Although the terms BC, soot and EC are somewhat different, for urban-traffic areas all three represent light absorbing particles emitted by traffic. Similarly for PNC the aerosol size range varies between the studies. When the condensation particle counter, CPC was used (e.g.: de Nazelle et al., 2012; Zuurbier et al., 2010) the studied size range was 0.01–1 μm, with the Particle counter, P-Trak was 0.02–1 μm (e.g.: Int Panis et al., 2010) while with the Diffusion Size Classifier (miniDiSC) (Ragettli et al., 2013) the size range was 0.01–0.3 μm. 3.1. Exposure during cycling A total of 20 European studies have been identified that calculated the exposure during cycling (Adams et al., 2001, 2002; Berghmans et al., 2009; Bevan et al., 1991; Boogaard et al., 2009; de Nazelle et al., 2012; Dons et al., 2011, 2012, 2013; Gee and Raper, 1999; Int Panis et al., 2010; Kaur et al., 2005, 2006; Kingham et al., 1998; McNabola et al., 2008; Rank et al., 2001; Ragettli et al., 2013; Strak et al., 2010; van Wijnen et al., 1995; Zuurbier et al., 2010). The majority of these studies have been performed in the United Kingdom (UK), the Netherlands and Belgium, Table 1. For PM2.5 the average exposure was found in the range 29–72 μg/m3 while for PM10 was in the range of 37–62 μg/m3. For Barcelona de Nazelle et al. (2012) report cyclist average exposure to PM2.5 of 35 μg/m3. Similarly Kaur et al. (2005) and Adams et al. (2001) observed similar exposure concentrations for cyclists equal to 34 μg/m3 in London urban area. In Dublin the exposure of the cyclist was higher and depended on the route taken. For two different routes the average exposures to PM2.5 for cyclists were 88 and 72 μg/m3 respectively (McNabola et al., 2008). On the other hand in the study of Zuurbier et al. (2010) cyclist's exposure showed no significant difference between the high traffic and low traffic routes. For BC the average exposure was in the range of 3–21 μg/m3 with the highest values recorded in London (Adams et al., 2002) and for high traffic routes (Strak et al., 2010; Kingham et al., 1998). Concerning particle number concentration for the London urban area Kaur et al. (2006) report an average value of 84,000 particles/cm3 while in Brussels and in 11 Dutch cities this was b 30,000 particles/cm3 (Int Panis et al., 2010; Boogaard et al., 2009). In Basel, Switzerland Ragettli et al. (2013) found average UFP concentrations equal to 22,660 particles/cm3 for cyclists. The exposure levels for cyclists depend on the detailed conditions of the daily trip and will therefore vary significantly among the studied areas. A cyclist riding in the middle of a busy road will be exposed to concentrations higher than those at the kerbside. By increasing the

787

distance to the main traffic flow, e.g. on a separate bicycle lane, the exposure might be substantially lower. However, exposure to air pollution while cycling in urban areas is generally considered high if taking into account that the minute ventilation (volume of air per minute) of cyclists is 1–5 times the minute ventilation of car and bus passengers (Zuurbier et al., 2009; Int Panis et al., 2010). Quantities of particles inhaled by cyclists can be between 4 and 7 times higher compared to car passengers on the same route, (Int Panis et al., 2010; O'Donoghue et al., 2007). In London, commuting to work by bicycle has been associated with increased long-term inhaled dose of BC (Nwokoro et al., 2012).

3.2. Exposure in cars Several studies exist on the personal exposure to air pollutants inside private vehicles (Dons et al., 2012; Zuurbier et al., 2010, 2011; Geiss et al., 2010, Boogaard et al., 2009; Cattaneo et al., 2009; McNabola et al., 2008; Alm et al., 1999; Kingham et al., 1998; Dennekamp et al., 2002), Table 2. Typical levels for commuter exposure in passenger cars were in the range of 36–76 μg/m3 for PM10 and 22–85 μg/m3 for PM2.5. Similar PM2.5 exposure values were obtained by Adams et al. (2001) and Kaur et al. (2005) equal to 38 and 37 μg/m3 respectively. High exposure values were determined for PM10 in Athens area especially in the heavy traffic route (Diapouli et al., 2008: PM10 exposure = 173 μg/m3). In the studies of Gulliver and Briggs (2007 and 2004) the personal exposure inside the car was 30% higher than the concentrations in the fixed monitoring station. BC exposure was determined in the range of 6–30 μg/m3 while the average PNC exposure was higher than 30,000 #/cm3. Exposure levels in diesel and gasoline-fuelled cars were comparable for PM, PNC and BC (Zuurbier et al., 2010). The main conclusion from all studies is that commuters' exposure is significantly influenced by the traffic intensity and the air pollutants concentration inside a private car mostly depend on the ambient air pollutant concentration and the choice of ventilation used inside the cars (Geiss et al., 2010; Cattaneo et al., 2009; Briggs et al., 2008; Diapouli et al., 2008; Rank et al., 2001). However, Jalava et al. (2012) found that pollutant concentrations inside the car depend highly on the type of fuel used.

3.3. Exposure in buses Exposure levels in buses are quite variable across the different countries, Table 3. Typical PM2.5 exposure values were found in the range 35–69 μg/m3 (Adams et al., 2001; Kaur et al., 2005; Dennekamp et al., 2002; Fondelli et al., 2008; Zuurbier et al., 2010). The lowest PM2.5 exposure levels (26 μg/m3) were observed in Barcelona probably due to the clean bus fleet of the city (de Nazelle et al., 2012). McNabola et al. (2008) found high PM2.5 exposures in Dublin (116 μg/m3) while Praml and Schierl (2000), report PM10 average exposure equal to 131 μg/m3 in Munich. In a recent study conducted in Paris, PM2.5 concentration inside buses was 59 μg/m3 with no significant differences in concentration within the cabin (front, middle, rear) while the outdoor concentration was 30% lower (Molle et al., 2013). Asmi et al. (2009) report PM2.5 average concentrations inside new buses equal to 12 μg/m3, 1.3 times increased when compared to the background air in Helsinki area (average value of 9 μg/m3). These low values were expected since the background mass concentrations and also the traffic density in Helsinki are lower than in other European cities like London. In the same study the passengers were exposed to somewhat higher concentrations than the bus driver due to the driver's compartment isolation while in newer buses, Euro3 concentrations were lower than in Euro2 buses; (Asmi et al., 2009). Similarly, Zuurbier et al. (2010) found that for all studied parameters PM, BC and PNC, exposures were higher in diesel than in electric fuelled buses and inside the cabin than in the driver's compartment.

788

Table 1 European studies reporting on exposure levels when commuting by bicycle. PM exposure

Parameter

Equipment

N of samples

Mean, μg/m3

Barcelona, (de Nazelle et al., 2012) 3 cities, Belgium (Int Panis et al., 2010) Utrecht, Netherlands (Strak et al., 2010)

PM2.5 PM10 PM10

“Adams” high volume sampler Aerosol monitor, DustTrak Harvard impactor

46 31 (persons) 14 (days)

35 62 44.0 (high traffic route) 45.7 (low traffic route)

Arnhem, Netherlands (Zuurbier et al., 2010)

PM10

Harvard impactor

15 (days)

PM2.5

Personal aerosol monitor, DataRAMa

16 (days)

Optical monitor, Grimm

7 sampling events

11 Dutch cities, (Boogaard et al., 2009) Dublin, Ireland (McNabola et al., 2008)

TSP PM10 PM2.5 PM1 PM2.5 PM2.5

London, UK (Kaur et al., 2005) London, UK (Adams et al., 2001) Copenhagen, Denmark (Rank et al., 2001) Manchester, UK (Gee and Raper, 1999) Southampton, UK (Bevan et al., 1991)

PM2.5 PM2.5 TSP PM4 PM N 3.5 μm

Aerosol Monitor, DustTrak High flow personal sampler (HFPS) High flow personal sampler High flow personal sampler Air sampling pump SKC sampling pump Impactor

126 (routes) 56 (route 1) 48 (route 2) 48 105 4 8 18

38.8 (high traffic route) 37.2 (low traffic route) 72.3 (high traffic route) 71.7 (low traffic route) 168 62.4 38.8 37.4 44.5 88.1 (route 1) 71.6 (route 2) 33.5 29 44.5 54 130

14.1 11.6 49.8 65.2 140 33.5 26.4 28.3 39.2 66.4 (route 1) 46.9 (route 2) 1.5 (GSD) 1.9 (GSD) 34

17–122 13–253

BC exposure

Parameter

Equipment

N of samples

Mean, μg/m3

SD

Min–max

Barcelona, (de Nazelle et al., 2012) Antwerpen (Dons et al., 2012) Utrecht, Netherlands (Strak et al., 2010)

BC BC Soot

Micro-aethalometer Model AE51 Micro-aethalometer Model AE51 Smokestain reflectometer

46 (trips) 1167 (5 min measurements) 16

Arnhem, Netherlands (Zuurbier et al., 2010) London, UK (Adams et al., 2002)

Soot

Smokestain reflectometer

16 (days)

EC

99

London, UK (Kingham et al., 1998)

Black smoke

EEL reflectometer calibrated with the Sunset OCEC analyzer Smokestain reflectometer

9.5 3.2 6.0 4.4 6.6 5.3 21

6 (road) 6 (path)

6.3 (road) 2.7 (path)

4.6 (road) 2.0 (path)

2.9–15.1 (road) 1.2–6.7 (path)

PNC exposure

Parameter

Equipment

N of samples

Mean, #/cm3

SD

Min–max

Basel, Switzerland (Ragettli et al., 2013) Barcelona, (de Nazelle et al., 2012)

Particle number concentration (0.01–0.3 μm) Particle number concentration (0.01–0.1 μm) Particle number concentration (0.02–1 μm) Particle number concentration (0.01–1 μm) Particle number concentration (0.01–1 μm)

Diffusion size Classifier (miniDiSC) Condensation particle counter, CPC Particle counter, P-Trak

51

22,660

16,000–41,000

46 (trips)

77,500

50,000–120,000

31 (number of test persons)

16,000

8500–50,000

Condensation particle counter, CPC Condensation particle counter, CPC

14 (days)

44,090 (high traffic route) 27,813 (high traffic route) 48,939 (high traffic route) 39,576 (low traffic route)

Particle number concentration (0.02–1 μm) Particle number concentration (0.01–1 μm) Particle number concentration (0.02–1 μm)

Particle counter, P-Trak

7

Condensation particle Counter, CPC Particle counter, P-Trak

Mol, Flanders Belgium (Berghmans et al., 2009)

Utrecht, Netherlands (Strak et al., 2010) Arnhem, Netherlands (Zuurbier et al., 2010)

Mol, Flanders Belgium (Berghmans et al., 2009) 11 Dutch cities, (Boogaard et al., 2009) London, UK (Kaur et al., 2006)

10−5/m (high traffic route) 10−5/m (low traffic route) 10−5/m (high traffic route) 10−5/m (low traffic route)

29.39 (high traffic route) 28.35 (low traffic route)

3.5 3.2 3.2 2.8 2.0 (GSD)

Min–max 16–59 40–92 16.7–118.7 (high traffic route) 14.2–109.3 (low traffic route)

27.2–1373 18.8–160 8.7–102 6.1–105 0–452

9.7–77.5 5.2–94.4

14–18 0–10 1.1–14.0 2.3–16.0

1.6–48.4

21,266

10,036 5919 19,039 (high traffic route) 18,178 (low traffic route) 13,795

5429–122,000

120 (number of routes)

24,441

18,020

5103–151,182

10

84,005

23,285

41,060–104,952

15 (days)

GSD: geometric standard deviation. a PM2.5 N PM10 because PM2.5 was measured by a photometric method and not gravimetrically as PM10 (Zuurbier et al., 2011).

28,443–58,409 18,047–38,796

A. Karanasiou et al. / Science of the Total Environment 490 (2014) 785–797

3 Belgium cities (Int Panis et al., 2010)

× × × ×

SD

Table 2 European studies reporting on exposure levels when commuting by car. Parameter

Equipment

N of samples

Mean, μg/m3

Barcelona, (de Nazelle et al., 2012) Ispra, Italy (Geiss et al., 2010)

PM2.5 PM10 PM2.5 PM1 PM10 PM10

“Adams” high volume sampler Optical particle counter, Grimm

34 (trips) 18 (number of cars tested)

Aerosol monitor, DustTrak, TSI Harvard impactor

9 (number of test persons) 14 (days)

PM2.5

Personal aerosol monitor, DataRAMa

16 (days)

Aerosol monitor, DustTrak Portable light scattering device

126 (number of routes) 46 (routes)

Athens, Greece (Diapouli et al., 2008)b

PM2.5 PM10 PM2.5 PM1 PM10

Aerosol monitor, DustTrak

Dublin, Ireland (McNabola et al., 2008)

PM2.5

Leicester, UK (Gulliver and Briggs 2007)

Portable light scattering device

2 routes, 73

London, UK (Kaur et al., 2005) Aberdeen, UK (Dennekamp et al., 2002) London, UK (Adams et al., 2001) Copenhagen, Denmark (Rank et al., 2001) Manchester, UK (Gee et al., 1999)

TSP-PM10 PM10–2.5 PM2.5–1 PM1 PM10 PM2.5 PM1 PM2.5 PM2.5 PM2.5 TSP PM4

High flow personal sampler (HFPS) Portable light scattering device

3 (heavy traffic routes) 2 (less traffic routes) 90

35.5 48.6 26.9 22.5 47 78.5 (diesel) 58.7 (gasoline) 101.3 (diesel) 114.8 (gasoline) 48.9 5.9 3.0 1.8 172.5 80 85.8

High flow personal sampler Aerosol monitor, DustTrak High flow personal sampler Air sampling pump SKC sampling pump

29 13 54 4

BC exposure

Parameter

Equipment

Barcelona, (de Nazelle et al., 2012) Antwerpen (Dons et al., 2012)

BC BC

Micro-aethalometer Model AE51 Micro-aethalometer Model AE51

Arnhem, Netherlands (Zuurbier et al., 2010)

Soot

Smokestain reflectometer

London, UK (Adams et al., 2002)

EC

London, UK (Kingham et al., 1998)

Black smoke

EEL reflectometer calibrated with the Sunset OCEC analyzer Smokestain reflectometer

PNC exposure

Parameter

Basel, Switzerland (Ragettli et al., 2013)

Particle number concentration (0.01–0.3 μm)

Barcelona, (de Nazelle et al., 2012) 3 Belgium cities (Int Panis et al., 2010) Arnhem, Netherlands (Zuurbier et al., 2010)

3 Belgium cities (Int Panis et al., 2010) Arnhem, Netherlands (Zuurbier et al., 2010)

Dutch cities, (Boogaard et al., 2009) London, UK (Briggs et al., 2008)

Min–max

30.7 15.2 11.4

24–48 0.9–263.7 0.9–94.4 0.8–82.9 11–93

101.5 (diesel) 34.6 (gasoline) 103.9 (diesel) 118.1 (gasoline) 75.2 3.1 1.1 1.1 90 42 78.3

5–806

42–1060 20–157 40.1

18.2 15.1 8.3 2.9 43.2 15.6 7.0 38.0 11 (median) 36.8 75 42

6.0 5.1 4.0 1.54 22.7 15.9 9.7 1.2 (GSD)

15.2–58.5

2.6 (GSD)

6.6–94.4

N of samples

Mean, μg/m3

SD

34 (trips) 3190b (car driver) 645b (car passenger) 16 (days) 33

19.5 6.4 5.6 8.2 × 10−5/m (diesel) 9.3 × 10−5/m (gasoline) 30.3

2.7 (diesel) 3.0 (gasoline) 2.1 (GSD)

1.4–88.8

6

7.6

4.4

3.5–14.7

Equipment

N of samples

Mean, #/cm3

SD

Min–max

84

31,784

25,255

6000–60,000

Particle number concentration (0.01–1 μm) Particle number concentration (0.02–1 μm) Particle number concentration (0.01–1 μm)

Diffusion size Classifier (miniDiSC) Condensation particle counter, CPC Particle counter, P-Trak Condensation particle counter, CPC

34 31 (number of test persons) 15 (days)

Milan, Italy (Cattaneo et al., 2009) Athens, Greece (Diapouli et al., 2008)c

Particle number concentration (0.01–1 μm) Particle number concentration (0.01–1 μm)

Condensation particle counter, CPC Condensation particle counter, CPC

London, UK (Kaur et al., 2006) London, UK (Kaur et al., 2005) Aberdeen, UK (Dennekamp et al., 2002) Kuoopia, Finland (Alm et al., 1999)

Particle number concentration (0.02–1 μm) Particle number concentration (0.02–1 μm) Particle number concentration (0.02–1 μm) Particle number concentration (0.3–1 μm)

Particle counter, P-Trak Particle counter, P-Trak Particle counter, P-Trak Laser particle counter

123,000 17,500 37,129 (diesel) 40,526 (gasoline) 107,000 128,000 63,000 36,821 99,736 25,000 (petrol car) 79.5

Northampton, UK (Gulliver and Briggs 2004)

b c

PM2.5 N PM10 because PM2.5 was measured by a photometric method and not gravimetrically as PM10 (Zuurbier et al., 2011). 5 min average measurements. 15-min average exposure.

1 (route in commercial areas) 1 (route in residential areas) 8 13 22

8.9–31.9 8.6–28.3 3.0–16.4 1.1–7.0

Min–max 5–44 0.5–19.0 1.0–14.0

51,000–260,000 5000–62,000 8262 (diesel) 10,142 (gasoline) 64,200 42,000 29,000 6299 1.4 (GSD)

5000–280,000 67,000–203,000 23,000–130,000 29,951–45,599 36,474–151,810 15–214

789

a

2 routes, 33

SD

A. Karanasiou et al. / Science of the Total Environment 490 (2014) 785–797

PM exposure

790

Table 3 European studies reporting on exposure levels when commuting by bus. PM exposure

Parameter

Equipment

N of samples

Mean, μg/m3

SD

Paris, France (Molle et al., 2013)

PM2.5 PM2.5 PM10

49 599 28 (trips) 11 (days)

74 69

Barcelona, (de Nazelle et al., 2012) Arnhem, Netherlands (Zuurbier et al., 2010)

Personal sampler, R&P/Thermo Optical aerosol monitor, TSI “Adams” high volume sampler Harvard impactor

PM2.5

Personal aerosol monitor, DataRAMa

10 (days)

PM2.5

Personal aerosol monitor, DataRam

2 days

Min–max

15 89

PM2.5 PM2.5 PM2.5 PM10 PM4

Portable Exposure Monitor High flow personal sampler (HFPS) High flow personal sampler Aerosol monitor, DustTrak High flow personal sampler Low volume sampler SKC sampling pump

55 59 26 68.5 (diesel) 53.2 (electric) 68.7 (diesel) 40.5 (electric) 17 (driver compartment in old bus) 22 (cabin in old bus) 10 (driver compartment in new bus) 13 (cabin in new bus) 56 116

42 14 68 137 17

34.5 38 (median) diesel bus 39 144 296

170 300

BC exposure

Parameter

Equipment

N of samples

Mean

SD

Barcelona, (de Nazelle et al., 2012) Antwerpen (Dons et al., 2012) Arnhem, Netherlands (Zuurbier et al., 2010)

BC BC Soot

Micro-aethalometer Model AE51 Micro-aethalometer Model AE51 Smokestain reflectometer

28 (trips) 190 (5 min measurements) 12 (days)

Helsinki, Finland (Asmi et al., 2009)

BC

Aethalometer

2 days

Florence, Italy (Fondelli et al., 2008) London, UK (Adams et al., 2002)

BS EC

15 46

London, UK (Kingham et al., 1998)

Black smoke

Smokestain reflectometer EEL Reflectometer calibrated with the Sunset OCEC analyzer Smokestain reflectometer

7.6 6.6 9.0 × 10−5/m (diesel) 5.1 × 10−5/m (electric) 7 (driver compartment in old bus) 3 (driver compartment in new bus) 11.5 22

6

5.3

3.0

2.3–10.7

PNC exposure

Parameter

Equipment

N of samples

Mean, #/cm3

SD

Min–max

Paris, France (Molle et al., 2013) Basel, Switzerland (Ragettli et al., 2013)

Particle number concentration (0.3–20 μm) Particle number concentration (0.01–0.3 μm)

416 53

1149 14,055

Barcelona, (de Nazelle et al., 2012) Arnhem, Netherlands (Zuurbier et al., 2010)

Particle number concentration (0.01–1 μm) Particle number concentration (0.01–1 μm)

Optical particle counter, Grimm Diffusion size Classifier (miniDiSC) Condensation particle counter, CPC Condensation particle counter, CPC

28 (trips) 13 (days)

Helsinki, Finland (Asmi et al., 2009)

Particle number concentration (0.01–1 μm)

Condensation particle counter, CPC

2 days

3728

55,200 43,235 (diesel) 28,602 (gasoline) 27,500 (driver compartment in old bus) 30,000 (cabin in old bus) 18,500 (driver compartment in new bus) 30,000 (cabin in old bus) 117,600

6 18 11

95,023 101,364 55,000 (diesel bus)

Helsinki, Finland (Asmi et al., 2009)

2 days PM2.5 PM2.5

London, UK (Kaur et al., 2005) Aberdeen, UK (Dennekamp et al., 2002) London, UK (Adams et al., 2001) Munich, Germany (Praml and Schierl, 2000) Manchester, UK (Gee and Raper, 1999)

2 days

Milan, Italy (Cattaneo et al., 2009)

Particle number concentration (0.01–1 μm)

London, UK (Kaur et al., 2006) London, UK (Kaur et al., 2005) Aberdeen, UK (Dennekamp et al., 2002)

Particle number concentration (0.02–1 μm) Particle number concentration (0.02–1 μm) Particle number concentration (0.02–1 μm)

Condensation particle counters, CPC and P-Trak Particle counter, P-Trak Particle counter, P-Trak Particle counter, P-Trak

GSD: geometric standard deviation. a PM2.5 N PM10 because PM2.5 was measured by a photometric method and not gravimetrically as PM10 (Zuurbier et al., 2011).

7–34 9–42

15 130

33–77

1.6 (GSD)

6.0–64.6 5.9–97.4 43–686

Min–max 4–10 1–17.5

7.2 (diesel) 2.1 (electric)

2.4

1.5–13.8 0.2–8.5 6.1–14.7 1.9–79.8

40,000–91,000 17,388 (diesel) 8399 (gasoline) 9000–47,000 13,000–49,000 6000–37,000 7500–44,000 59,700 25,176 1.3 (GSD)

63,778–121,285 64,463–158,685

A. Karanasiou et al. / Science of the Total Environment 490 (2014) 785–797

Florence, Italy (Fondelli et al., 2008) Dublin, Ireland (McNabola et al., 2008)

16–42 53.6 (diesel) 23.2 (electric) 91.7 (diesel) 32.1 (electric)

Table 4 European studies reporting on exposure levels when commuting by the subway. PM exposure

Parameter

Equipment

N of samples

Mean, μg/m3

Milan metro, Italy (Colombi et al., 2013) Barcelona metro, Spain (Querol et al., 2012)

PM10 PM10

Low volume sampler High volume samplers, MCV

3 lines tested 1 old line and 1 new line tested, 20 daysa

Aerosol monitors, GRIMM and DustTrak

3 days

52 SD

PM2.5 PM2.5

High flow personal sampler SKC Inc. pump

56 16 h

BC exposure

Parameter

Equipment

N of samples

Mean, μg/m3

Helsinki metro, Finland (Aarnio et al., 2005)

BC

Aethalometer

12 days

6.3 (inside the station during weekdays from 6 to 18 h)

PNC exposure

Parameter

Equipment

N of samples

Mean, #/cm3

Helsinki metro, Finland (Aarnio et al., 2005)

Particle number concentration (0.01–0.5 μm) Particle number concentration (0.02–1 μm)

Differential mobility particle sizer

12 days

Particle monitor, P-Trak

3 days

31,000 (inside the station 4 m from platform level) 20,667 (driver's cabin)

PM2.5 PM10 PM2.5 PM1 Budapest metro, Hungary (Salma et al., 2007) Paris metro, France (Raut et al., 2009)

PM10 PM10

Tapered element oscillating microbalance, TEOM Tapered element oscillating microbalance, TEOM

PM2.5 (Ripanucci et al., 2006) Prague, metro (Braniš et al., 2006)

PM10 PM10

Low volume sampler Aerosol monitor, DustTrak

108 journeys

Helsinki metro, Finland (Aarnio et al., 2005)

PM2.5

Optical monitor, Eberline

12 days

London metro, UK (Seaton et al., 2005)

PM2.5 PM10 PM2.5

Low volume sampler Aerosol monitor, DustTrak Aerosol monitor, DustTrak

3 days 3 days

PM10

Tapered element oscillating microbalance, TEOM

257 h

Stockholm metro, Sweden (Johansson and Johansson, 2003)

PM2.5

London metro, UK (Seaton et al., 2005) a

257 h

SD

Min–max 58–299

55

12–103 17–45

212–722 3.5–280 105–388 4–69 12.2–371.2

A. Karanasiou et al. / Science of the Total Environment 490 (2014) 785–797

London, UK (Adams et al., 2001) London, UK (Pfeifer et al., 1999)

103–184 (platform, weekdays) 367 (old line platform) 193 (new line platform) 133 (old line platform) 59 (new line platform) 79 (old line, inside train) 45 (new line, inside train) 25 (old line, inside train) 14 (new line, inside train) 24 (old line, inside train) 5 (new line, inside train) 155 183 (weekdays) 84 (weekends) 58 (weekdays) 29 (weekends) 381 (platform) 102 (vestibule) 113 (inside train) 52 (inside the station 4 m above platform level during weekdays from 6 to 18 h) 21 (inside train) 1037 (platform) 343 (platform) 170 (driver's cabin) 469 (weekdays) 336 (weekends) 258 (weekdays) 185 (weekends) 202.3 246

Min–max 2.5–16

SD

Min–max 8900–97,000

Measurements conducted from 8.30 to 20.30 h.

791

792

Table 5 European studies comparing PM, BC and PNC exposure levels in different commuting modes. Parameter

Equipment

Car

Bus

Bicycle

Taxi Subway

Barcelona, (de Nazelle et al., 2012) Belgian cities, (Int Panis et al., 2010) Arnhem, (Zuurbier et al., 2010)

PM2.5 PM10 PM2.5

“Adams” high volume sampler DustTrak Personal aerosol monitor, DataRAM

–

–

Dutch cities, (Boogaard et al., 2009) Dublin, (McNabola et al., 2008);

PM2.5 PM2.5

DustTrak HFPS high volume personal sampler

– –

– –

Florence, (Fondelli et al., 2008) London, (Kaur et al., 2005) Aberdeen, (Dennekamp et al., 2002) London, (Adams et al., 2001)

PM2.5 PM2.5 PM2.5 PM2.5

Cyclone Casella vortex ultraflow DustTrak Casella vortex ultraflow

39 42 – –

Manchester, (Gee and Raper, 1999) London, (Pfeifer et al., 1999)

PM4 TSP PM2.5

SKC sampling pump SKC sampling pump

35 50–73 72 (high traffic) 72 (low traffic) 39 (high traffic) 37 (low traffic) 6–112 88 (route 1) 72 (route 2) – 34 – 23.5 (winter) 34.5 (summer) 54 –

– – –

Harvard impactor

26 – 69 (diesel) 41 (electric) 69 (diesel) 53 (electric) – 128 (route 1) 104 (route 2) 56 35 38 38.9 (winter) 39.0 (summer) 296 –

– – –

PM10

36 38–74 101 (diesel) 115 (gasoline) 79 (diesel) 59 (gasoline) 14–122 83 (route 1) 89 (route 2) – 38 11 33.7 (winter) 37.7 (summer) – –

– – – 157.3 (winter) 247.2 (summer) – 669 246

BC exposure, μg/m3

Parameter

Equipment

Car

Bus

Bicycle

Barcelona (de Nazelle et al., 2012) Antwerpen (Dons et al., 2012) Arnhem, (Zuurbier et al., 2010)

BC BC Soot

Micro-aethalometer Model AE51 Micro-aethalometer Model AE51 Smoke stain reflectometer‡

Berlin (Fromme et al. (1998))

EC (PM10)

7.6 6 9 (diesel) 5 (electric) –

9.5 4 7 (high traffic) 5 (low traffic) –

London (Adams et al., 2002)

EC (PM2.5)

London, UK (Kingham et al., 1998)

PNC exposure, #/cm3

– 54 33

Subway – – –

25 (winter) 16 (summer) 19 (winter) 15 (summer)

–

6 (winter) 11 (summer) –

Black smoke

19.5 6 8 (diesel) 9 (gasoline) Gravikon sampler 8 (winter) EC was determined by coulometry 14 (summer) EEL reflectometer calibrated with the 34 (winter) 26 (summer) Sunset OCEC analyzer Smokestain reflectometer 7.6

5.3

6.3 (road) 2.7 (path)

–

–

Size

Equipment

–

5 –

Car

Bus

Bicycle

Basel, Switzerland (Ragettli et al., 2013)

31,784

14,055

22,660

–

–

43,235 (diesel) 28,602 (gasoline) 95,023

48,939 (high traffic route) – 39,576 (low traffic route) 84,005 –

–

London, UK (Kaur et al., 2006)

37,129 (diesel) 40,526 (gasoline) 36,821

Particle number concentration (0.01–0.3 μm) Diffusion size Classifier (miniDiSC) Arnhem, Netherlands (Zuurbier et al., 2010) Particle number concentration (0.01–1 μm) Condensation particle counter, CPC Particle number concentration (0.02–1 μm)

Particle counter, P-Trak

Subway

–

A. Karanasiou et al. / Science of the Total Environment 490 (2014) 785–797

PM exposure, μg/m3

A. Karanasiou et al. / Science of the Total Environment 490 (2014) 785–797

3.4. Exposure in underground train systems Previous studies have found high PM levels in the subway systems of Milan (Colombi et al., 2013), Barcelona (Querol et al., 2012), Berlin (Fromme et al., 1998), Budapest (Salma et al., 2007, 2009), Helsinki (Aarnio et al., 2005), London (Seaton et al., 2005; Adams et al., 2001; Pfeifer et al., 1999), Rome (Ripanucci et al., 2006), Paris (Raut et al., 2009), Prague (Branis, 2006) and Stockholm (Johansson and Johansson, 2003). In general PM levels were much elevated on the platforms, being around 3–4 times higher than inside trains (Querol et al., 2012). As it was expected, pollutant exposure was lower during weekends due to the lower circulation of trains (Raut et al., 2009). The mean PM10 levels on the platform ranged from 103 to 1030 μg/m3 while PM2.5 levels ranged from 59 to 375 μg/m3, with the highest concentrations found in London, Stockholm and Rome undergrounds. These variations in exposure levels among the metro systems could be explained by the abrasion of railways and catenary metal, and to braking systems (Salma et al., 2007). The lowest PM2.5 exposure during metro commuting was obtained in Barcelona in the new constructed lines (27 μg/m3: Querol et al., 2012). That same study concluded that the ventilation system and air conditioning inside the trains decisively improve air quality in the subway system.

3.5. Comparison studies In Table 5 we summarise the PM, BC and PNC exposure studies conducted in Europe that compare different commuting modes. It should be mentioned that the variety of techniques and methods used (e.g. Boogaard et al., 2009 determined PM2.5 by the DustTrak that uses light scattering and a factory calibration to express the measurement results in mass units while Adams et al., 2001 used a

793

high flow personal sampler) make difficult the comparison between different studies. Several studies (Boogaard et al., 2009; de Nazelle et al., 2012; Zuurbier et al., 2010; Kaur et al., 2005; Adams et al., 2001) report lower exposures to PM for cyclists than for car-passengers. In Netherlands the overall mean concentration of PM2.5 during driving was 11% higher than during cycling (Boogaard et al., 2009). However, some studies observed that exposure to fine particles is higher during cycling than during commuting with a private car (McNabola et al., 2008; Gee et al., 1999; Gee and Raper, 1999). Personal exposure to PM for bus passengers depends strongly on the type of bus used and consequently varies significantly among the studied areas. For example in Barcelona where the bus fleet is quite modern (a large proportion 41% is natural gas fuelled, 8% hybrid and the rest, 51% with a combined Selective Catalytic Reduction Trap and Continuously Regenerating Particulate Trap installed) the exposure levels in buses were lower than in cars and when commuting by bicycle (de Nazelle et al., 2012). On the other hand passengers in public bus (diesel-powered) in Dublin generally displayed the highest personal exposure to PM2.5 compared to other modes (McNabola et al., 2008). Commuting in the subway had the lowest exposure levels in Barcelona (de Nazelle et al., 2012; Querol et al., 2012), whereas mean exposure levels on the London underground rail system were 3–8 times higher than in the other transport modes (Adams et al., 2001). Kaur et al. (2005) concluded that PM2.5 exposure during commuting for each of the transport modes in London are up to three times higher than observed at the fixed measurement stations. Lower BC exposures were reported for cyclists and higher exposure during commuting by car (De Nazelle et al., 2012; Adams et al., 2002; Dons et al., 2012; Zuurbier et al., 2010). With reference to particle number counts exposures Zuurbier et al. (2010) found that these were highest inside diesel buses

Fig. 1. Particle number concentrations a) and BC levels b) for the two cars during stable conditions (engines off).

794

A. Karanasiou et al. / Science of the Total Environment 490 (2014) 785–797

Fig. 2. Particle number concentrations a) and BC levels b) for the two cars during driving in heavy traffic real-world conditions in an urban circuit.

(38,500 particles/cm3) and for cyclists along a high-traffic intensity route (46,600 particles/cm3) and lowest in electric buses (29,200 particles/cm3). On the contrary Ragettli et al. (2013) report that average UFP concentrations were higher in car (31,784 particles/cm3) and on bicycle (22,660 particles/cm3) compared to public transportation (14,055– 18,818 particles/cm3). Nevertheless, we should mention that the inhalation dose of commuters could be different from the exposure levels. Dons et al. (2012) found that exposure to BC during travel in motorised transport was clearly higher than exposure while walking or biking (6.3 μg/m3 versus 3.4 μg/m3), but when accounting for inhaled doses, this relationship was reversed. Cyclists seem to receive the highest inhaled PM2.5 dose followed by bus and car passengers (McNabola et al., 2008). However, in the present study we do not examine the inhaled dose of pollutants during commuting due to the lack of studies determining it and also its complexity of interacting factors such as breathing rate, ventilation and/or particle deposition to the respiratory system.

3.6. Other parameters investigated than transport mode

3.6.1. Traffic conditions and route Traffic emissions and consequently the selected route (low or high traffic) can influence the exposure levels during commuting. It has been proven that in a moving vehicle, the concentration of pollutants depends on the exhaust emissions of neighbouring vehicles (Dor et al., 1995). Indeed, Dons et al. (2012, 2013) found that on traffic peak hours with high traffic intensities in-car BC concentrations are about 2 μg/m3 higher than average. Higher exposures to particle number concentrations were also observed in heavy-traffic areas and during rush hours (Diapouli et al., 2008). Personal exposure for cyclists is substantially reduced by decreasing proximity to motorised traffic, and traffic intensity (Dons et al., 2013; Boogaard et al., 2009; Kaur et al., 2006; Adams et al., 2001b; Kinghams et al., 1998). The average particle number concentration can be 59% higher, while the average soot concentration 39% higher on hightraffic routes than on low-traffic routes. On the contrary exposure to PM10 is not significantly affected by the traffic intensity or the distance of the cycle path from the motorised traffic as PM10 dependence on primary vehicle emissions is lower than that of particle number and BC concentrations (Strak et al., 2010).

Except from the transport mode numerous variables might influence personal exposure in transport. Kaur et al. (2007) classified potential confounders in four categories: personal factors, mode of transport factors, road traffic factors and meteorological factors. Personal factors (e.g. breathing rates) are not discussed in this paper.

3.6.2. Travel speed Travel speed is also strongly related to personal exposure. The invehicle BC concentrations are elevated at lower speeds (up to 30 km/h) and at speeds above 80 km/h (Dons et al., 2013). The relevance of speed is two-fold: (a) because vehicle emissions differ at different speeds,

A. Karanasiou et al. / Science of the Total Environment 490 (2014) 785–797

and (b) because at lower speeds the influence of the emissions from surrounding vehicles is higher. The same was observed for fine particles and also CO exposure, found to be considerably higher in the slower than in the faster speeds (Alm et al., 1999). 3.6.3. Between-vehicle distance The between-vehicle distance can influence the exposure levels since the reduced distance between vehicles at lower speeds in urban traffic, or when congestion is present on highways, allows emissions to infiltrate into vehicles nearby, resulting in higher exposures. In the study of McNabola et al. (2009b) a significant drop in in-vehicle VOC and PM2.5 concentrations occurred within the first 2 m of their emission from the preceding vehicles exhaust. 3.6.4. Ventilation Ventilation rates, whether driven by fans, natural leakage or open windows describe how rapidly outdoor air is capable of entering passenger cabins (Knibbs et al., 2011). Evidence suggests that ventilation is a key determinant of in-cabin concentrations in cars and buses as a high ventilation rate allows outdoor pollutants to enter the cabin (Zuurbier et al., 2010). Another constraint is the filtration system of the vehicle that helps to prevent ingress of particles, so that the vehicle is insulated against much of air pollution present in the street (Briggs et al., 2008). Recently, Querol et al. (2012) concluded that ventilation system and the air conditioning inside the trains decisively improve air quality in the subway system. 3.6.5. Fuel type We identified only two studies where the fuel type was examined as an influencing parameter in personal exposure (Jalava et al., 2012; Zuurbier et al., 2010). Zuurbier et al. (2010) found the highest median particle number concentration and PM10 exposures in diesel cars and buses while the median soot exposure was highest in gasoline-fuelled cars. In that same study it was reported that in electric buses exposure levels were lower than in diesel buses although the difference for PM10 was small. However the observed differences were not statistically significant. Additionally, it is difficult to separate the effects of fuel type from those due to differences in ventilation under a standard setting between vehicles of different manufacturers (e.g. Knibbs et al., 2009). In the study of Jalava et al. (2012) the PM emissions from a heavyduty EURO IV diesel engine powered by three different fuels, a conventional diesel fuel and two biodiesels (methyl ester and hydrotreated vegetable oil) were evaluated and the toxicological properties of the emitted PM were investigated. The vegetable oil performed very well in emission reduction and in lowering the overall toxicity of emitted PM, but mixtures with the conventional diesel fuel were no better in this respect than the plain diesel fuel. Within the purposes of the present literature review we present the results of a short experiment that was conducted to investigate the influence of other parameters like the vehicle type (electric or conventional) on the exposure levels (IDAEA-CSIC, unpublished data). The vehicles under investigation were provided voluntarily. A gasoline vehicle (Saab 93) and an electric car (Tazzari Zero 100% electric, model CLASSIC 2010 edition, ENISOLA S.L.) were tested in terms of particle number and BC concentrations in a real-world urban circuit in Barcelona. Each vehicle was equipped with a DiscMini (Matter Aerosol) to monitor the particle number concentrations and a micro-aethalometer (Model AE51, Magee Scientific) to monitor BC levels. The instruments were placed in the car cabin approximately 1 h before driving. From the 1-h measurements during stable conditions (the two cars were parked in parallel with the engines switched off and no passengers inside) the difference in the air exchange rate between the two vehicles was tested. As illustrated in Fig. 1, both particle number and BC concentrations were higher inside the electric vehicle during stable conditions. Since no indoor emission sources were present in the two cars and indoor

795

concentrations were depending on the outdoor pollution levels this fact reveals the higher exchange rate of the electric car compared to the gasoline car. This result aimed to test the different infiltration rates in each car, which are only dependent on the vehicle structure and building materials. Afterwards the volunteers conducted a 45 minute trip during heavy traffic conditions with the windows closed and two passengers in each car. Air re-circulation and air conditioning were switched off in both vehicles. In the first 30 min the two cars were moving in parallel. Although the concentrations levels were constantly higher inside the electric vehicle (due to the higher air exchange rate) the time variation was similar in the two cars revealing the strong influence of the outdoor (traffic) emissions. In the last 15 min of the trip the gasoline car drove directly behind the electric one in order to observe any reduction in the particle number and BC levels inside the gasoline car due to zero exhaust emissions of the electric car. Again no significant reduction was observed probably due to the higher influence of the emissions of the surrounding conventional vehicles, Fig. 2. Both results suggest that exposure during commuting by car depends on the emissions from the surrounding vehicles, rather than on the actual vehicle the subject is commuting in. Therefore, the composition of the vehicle fleet (clean or high polluting vehicles) plays a key role in exposure by car in urban environments, as do traffic intensity. It is interesting to observe that these results do not coincide with other studies found in the literature (Jalava et al., 2012), which suggest that pollutant concentrations inside the car depend highly on the type of fuel used. At this point we are unable to explain these different results. 3.6.6. Meteorological conditions Personal exposure concentrations in non-motorised forms of transport (cycling) are influenced to a higher degree by wind speed, whereas personal exposure in motorised forms of transport is driven to a higher degree by traffic congestion (McNabola et al., 2008, 2009a). The analysis of the PM2.5 exposure in London showed that wind speed has a significant influence on personal exposure levels. The occurrence of higher wind speeds is strongly associated with a decrease in personal exposure levels (Adams et al., 2001b). Alm et al. (1999) found that the meteorological parameters that affect the passengers' exposures to fine particles and CO are wind speed (decreasing exposure) and relative humidity (increasing exposure). 3.6.7. Other compounds studied Although this review focuses solely on PM, BC and UFP we briefly mention some other compounds that have been studied simultaneously with particulate pollutants in exposure studies during commuting. These include: • Carbon monoxide, CO (Alm et al., 1999; Bevan et al., 1991; de Nazelle et al., 2012; Kaur et al., 2005, 2007; Le Moulle et al., 1998). • Volatile organic compounds, VOCs (Bevan et al., 1991; McNabola et al., 2008; Parra et al., 2008) • Nitrogen dioxide, NO2 (Molle et al., 2013) • PAHs (Fromme et al., 1998). Concerning other pollutants such as VOCs, again car passengers are exposed to higher levels when compared to other transport modes. McNabola et al. (2008) report that car commuters have the highest exposure to VOCs followed by cyclists and bus passengers. In Berlin a comparison between personal exposure in the subway and in the car showed significantly higher concentrations of PAHs in the subway train (Fromme et al., 1998). 4. Conclusions and recommendations Studies on personal exposure to air pollutants during car commuting are more numerous than those dealing with other types of transport, and typically conclude by emphasising that travelling by car involves

796

A. Karanasiou et al. / Science of the Total Environment 490 (2014) 785–797

exposure to relatively high PM exposure concentrations. Thus, compared to other transport methods, travelling by car has been shown to involve exposure both to higher PM (11%, according to Boogaard et al., 2009) and BC (Adams et al., 2002; de Nazelle et al., 2012) as compared with cycling. The pollutant exposure will however differ greatly depending on traffic intensity and speed and the type of ventilation inside the car. The need for reducing the number of vehicles and especially of older (more polluting) vehicles inside the cities is evident and an obvious recommendation for the improvement of the urban environment. Personal exposure to PM for bus commuters depends on several variables, varying greatly from city to city, depending on the traffic intensity of each place and the type of buses in use. Thus, as in the case of travelling by car, levels of PM and BC to which bus passengers are exposed will very much depend on the selected route, as highly busy streets will contain higher ambient levels of exhaust emissions from neighbouring vehicles. Similarly, the contamination emitted by buses will be small in those cities where the bus fleet is modern and ecologically friendly, fitted with natural gas, hybrid, or diesel with filter trap systems (Zuurbier et al., 2010). As with the case for reducing private cars circulating in urban areas, the argument for improving urban bus fleets to reduce ambient air pollution is encouraging the implementation of modernising programmes in all cities. Commuting by underground rail is a transport mode used daily by over one hundred million people worldwide, with more than 60 cities in Europe alone utilising rail subways to facilitate commuter movement. Although less information is available, studies on commuting in the subway show higher PM levels compared to other modes of transport (Nieuwenhuijsen et al., 2007). However, some subway systems present lower PM exposure levels than those shown by other transport modes (Chan et al., 2002a, 2002b). Of special concern is the fact that, when compared to outdoor air, subway air might be anomalously rich in metals, especially iron but also trace metals such as manganese, chromium, copper, nickel, zinc as well as the toxic metalloids antimony and arsenic (Querol et al., 2012). Key factors influencing subway PM concentrations include station depth, station design, type of ventilation (non-forced/forced), air-conditioning inside trains, types of brakes (electric/conventional brake pads) and wheels (rubber v. steel) used on the trains, train frequency, driving behaviour and more recently the presence or absence of platform screen door systems (Querol et al., 2012; Nieuwenhuijsen et al., 2007). These variables can lead to an increase of up to 8 times PM exposure when compared to other transport modes (Adams et al., 2001), especially on platforms rather than inside trains. In general, exposure studies published so far have revealed that cyclist commuters are exposed to lower particulate matter concentrations in comparison to those inside vehicles. However, when accounting for inhaled doses, this relationship is reversed, with cyclists having higher PM2.5 lung deposition due to their greater energy expenditure and consequent inhalation rate (McNabola et al., 2008). The fact that proximity to the pollutant sources has a significant impact on exposure levels experienced by cyclists and pedestrians is an important aspect that should be included in developing traffic circulation plans, for example, by creating separate bicycle lanes parallel to the arterial roads feeding large volumes of traffic in and out of the city or accommodate bike lanes in quieter parallel streets (Jacobs et al., 2010; Yang et al., 2010; Puchewr and Buehler, 2008). Even without employment of such avoidance strategies, however, the personal benefits of cycling are still significant. In this context, Rojas-Rueda et al. (2012) evaluated the health benefit of the bike sharing “Bicing” programme in Barcelona, including the effect of pollution exposure for the cyclists (although not the public benefit due to reduced vehicle emissions). Public bicycle sharing initiatives such as Bicing are increasingly common and have wider benefits than simply the reduction of pollutant (Rojas-Rueda et al., 2011; Woodcock et al., 2009; de Hartog et al., 2010; Rabl and de Nazelle, 2012). Foremost among these advantages, despite large uncertainties in the figures, is the health benefit gained by the individual due

to the physical activity involved. It is this personal benefit, rather than the actual decrease in pollutant emissions induced by the switch away from car use, which perhaps provides the strongest case for encouraging urban bicycle transport. All cities face the considerable challenge of adopting transport policies that reduce, or at least do not enhance, ambient levels of air pollutants. Widespread dependence on private car transport has produced a significant daily health offence to the urban commuter, with consequent health risks that are especially threatening to those already compromised by illness or old age. However, enough has already been published to demonstrate that a forward-looking, integrated transport policy, involving the phased renovation of existing public vehicles and the withdrawal of the more polluting (older) private vehicles, combined with incentives to use public transport and the encouragement of commuter physical exercise, will produce immediate results towards solving what is, a difficult but not intractable problem. Acknowledgements The authors would like to acknowledge the participation of ENISOLA S.L. (Soluciones de movilidad eléctrica y energía fotovoltaica Distribuidor oficial TAZZARI ZERO, www.enisola.com). This work was supported by the European Environment Agency (EEA) and the European Topic Centre on Air Quality and Climate Change Mitigation (ETC.-ACM). References Aarnio P, Yli-Tuomi T, Kousa A, Mäkelä T, Hirsikko A, Hämeri K, et al. The concentrations and composition of and exposure to fine particles (PM2.5) in the Helsinki Subway System. Atmos Environ 2005;39:5059–66. Adams HS, Nieuwenhuijsen MJ, Colvile RN, McMullen MA, Khandelwal P. Fine particle (PM2.5) personal exposure levels in transport microenvironments, London, UK. Sci Total Environ 2001a;279(1–3):29–44. Adams HS, Nieuwenhuijsen MJ, Colvile RN. Determinants of fine particle (PM2.5) personal exposure levels in transport microenvironments, London, UK. Atmospheric Environment 2001b;35:4557–66. Adams HS, Nieuwenhuijsen MJ, Colvile RN, Older MJ, Kendall M. Assessment of road users' elemental carbon personal exposure levels, London, UK. Atmos Environ 2002;36:5335–42. Alm S, Jantunen MJ, Vartiainen M. Urban commuter exposure to particle matter and carbon monoxide inside an automobile (Conference Paper). J Expo Anal Environ Epidemiol 1999;9(3):237–44. Anderson JO, Thundiyil JG, Stolbach A. Clearing the air: a review of the effects of particulate matter air pollution on human health. J Med Toxicol 2012;8(2):166–75. Asmi E, Antola M, Yli-Tuomi T, Jantunen M, Aarnio P, Mäkelä T, et al. Driver and passenger exposure to aerosol particles in buses and trams in Helsinki, Finland. Sci Total Environ 2009;407(8):2860–7. Berghmans P, Bleux N, Panis LI, Mishra VK, Torfs R, Van Poppel M. Exposure assessment of a cyclist to PM10 and ultrafine particles. Sci Total Environ 2009;407:1286–98. Bevan MAJ, Proctor CJ, Baker-Rogers J, Warren ND. Exposure to carbon monoxide, respirable suspended particulates, and volatile organic compounds while commuting by bicycle. Environ Sci Technol 1991;25(4):788–91. Boogaard H, Borgman F, Kamminga J, Hoek G. Exposure to ultrafine and fine particles and noise during cycling and driving in 11 Dutch cities. Atmos Environ 2009;43:4234–42. Boudet C, Déchenaux J, Balducci F, Masclet P, Zmirou D. Total personal exposure to fine particulate matter in an urban adult population: the role of ambient air and transport. Int J Veh Des 1998;20(1–4):30–8. Braniš M. The contribution of ambient source to particulate pollutionin spaces and trains of the Prague underground transport system. Atmospheric Environment 2006;40: 348e356. Branis M. The contribution of ambient source to particulate pollution in spaces and trains of the Prague underground transport system. Atmos Environ 2006;40:348–56. Briggs DJ, de Hoogh K, Morris C, Gulliver J. Effects of travel mode on exposures to particulate air pollution. Environ Int 2008;34(1):12–22. Cattaneo A, Garramone G, Taronna M, Peruzzo C, Cavallo DM. Personal exposure to airborne ultrafine particles in the urban area of Milan. J Phys Conf Ser 2009;151:012039. Chan LY, Lau WL, Lee SC, Chan CY. Commuter exposure to particulate matter in public transportation modes in Hong Kong. Atmos Environ 2002a;36:3363–73. Chan LY, Lau WL, Zou SC, Cao ZX, Lai SC. Exposure level of carbon monoxide and respirable suspended particulate in public transportation modes while commuting in urban area of Guangzhou, China. Atmos Environ 2002b;36:5831–40. Colombi C, Angius S, Gianelle V, Lazzarini M. Particulate matter concentrations, physical characteristics and elemental composition in the Milan underground transport system. Atmos Environ 2013;70:166–78. de Hartog J, Boogaard H, Nijland H, Hoek G. Do the health benefits of cycling outweigh the risks? Environ Health Perspect 2010;118(8):1109–16.

A. Karanasiou et al. / Science of the Total Environment 490 (2014) 785–797 de Nazelle A, Fruin S, Westerdahl D, Martinez D, Ripoll A, Kubesch N, et al. Travel mode comparison of commuters' exposures to air pollutants in Barcelona. Atmos Environ 2012;59:151–9. Dennekamp M, Mehenni O, Cherrie J, Seaton A. Exposure to ultrafine particles and PM2.5 in different microenvironments. Ann Occup Hyg 2002;46:412–4. Diapouli E, Grivas G, Chaloulakou A, Spyrellis N. PM10 and ultrafine particles counts invehicle and On-road in the Athens area. Water Air Soil Pollut Focus 2008;8(1):89–97. Dons E, Int Panis L, Van Poppel M, Theunis J, Willems H, Torfs R, et al. Impact of timeactivity patterns on personal exposure to black carbon. Atmos Environ 2011;45: 3594–602. Dons E, Int Panis L, Van Poppel M, Theunis J, Wets G. Personal exposure to Black Carbon in transport microenvironments. Atmos Environ 2012;55:392–8. Dons E, Temmerman P, Van Poppel M, Bellemans T, Wets G, Int Panis L. Street characteristics and traffic factors determining road users' exposure to black carbon. Sci Total Environ 2013;447:72–9. Dor F, Moullec YL, Festy B. Exposure of cityresidents to carbon monoxide and monocyclic aromatichydrocarbons during commuting trips in the Paris metropolitanarea. Journal of Air and Waste Management Association 1995;45:103–10. Duci A, Chaloulakou A, Spyrellis N. Exposure tocarbon monoxide in the Athens urban area duringcommuting. The Science of the Total Environment 2003;309:47–58. Fondelli MC, Chellini E, Yli-Tuomi T, Cenni I, Gasparrini A, Nava S, et al. Fine particle concentrations in buses and taxis in Florence, Italy. Atmos Environ 2008;42(35): 8185–93. Fromme H, Oddoy A, Piloty M, Krause M, Lahrz T. Polycyclic aromatic hydrocarbons (PAH) and diesel engine emission (elemental carbon) inside a car and a subway train. Sci Total Environ 1998;217(1–2):165–73. Gee IL, Raper DW. Commuter exposure to respirable particles inside buses and by bicycle (Conference Paper). Sci Total Environ 1999;235(1–3):403–5. Gee IL, Coleman B, Raper DW. Commuter exposure to respirable particles inside vehicles in Manchester, UK. In: Raw G, Aizlewood C, Warren P, editors. The 8th International Conference on Indoor Air and Climate, Indoor Air 99, 4. ; 1999. p. 1066–71. [Edinburgh, Scotland, 8–13 August 1999]. Geiss O, Barrero-Moreno J, Tirendi S, Kotzias D. Exposure to particulate matter in vehicle cabins of private cars. Aerosol Air Qual Res 2010;10:581–8. Gulliver J, Briggs DJ. Personal exposure to particulate air pollution in transport microenvironments. Atmospheric Environment 2004;38:1–8. Gulliver J, Briggs DJ. Journey-time exposure to particulate air pollution. Atmospheric Environment 2007;41:7195–207. Heal RM, Kumar P, Harrison RM. Particles, air quality, policy and health. Chem Soc Rev 2012;41:6606–30. Hoek G, Boogaard H, Knol A, De Hartog J, Slottje P, Ayres JG, et al. Concentration response functions for ultrafine particles and all-cause mortality and hospital admissions: results of a European expert panel elicitation. Environ Sci Technol 2010;44:476–82. Int Panis L, de Geus B, Vandenbulcke G, Willems H, Degraeuwe B, Bleux N, et al. Exposure to particulate matter in traffic: a comparison of cyclists and car passengers. Atmos Environ 2010;44(19):2263–70. Jacobs L, Nawrot TS, de Geus B, Meeusen R, Degraeuwe B, Bernard A, et al. Subclinical responses in healthy cyclists briefly exposed to traffic-related air pollution: an intervention study. Environ Heal 2010;9:64. Jalava PI, Aakko-Saks P, Murtonen T, Happo MS, Markkanen A, Yli-Pirilä P, et al. Toxicological properties of emission particles from heavy duty engines powered by conventional and bio-based diesel fuels and compressed natural gas. Part Fibre Toxicol 2012;9. [Article number 37]. Janssen NAH, Gerlofs-Nijland ME, Lanki T, Salonen RO, Cassee F, Hoek G, et al. Health effects of black carbon World Health Organization report. Copenhagen, Denmark: WHO Regional Office for Europe; 2012. Johansson C, Johansson P-A. Particulate matter in the underground of Stockholm. Atmos Environ 2003;37:3–9. Kaur S, Nieuwenhuijsen M, Colvile R. Personal exposure of street canyon intersection users to PM2.5, ultrafine particle counts and carbon monoxide in Central London, UK. Atmos Environ 2005;39(20):3629. Kaur S, Clark R, Walsh P, Arnold SJ, Colvile RN, Nieuwenhuijsen M. Exposure visualisation of ultrafine particle counts in a transport microenvironment. Atmos Environ 2006;40: 386–98. Kaur S, Nieuwenhuijsen MJ, Colvile RN. Fine particulate matter and carbon monoxide exposure concentrations in urban street transport microenvironments (Review). Atmos Environ 2007;41(23):4781–810. Kingham S, Meaton J, Sheard A, Lawrenson O. Assessment of exposure to traffic-related fumes during the journey to work. Transp Res Part D: Transp Environ 1998;3(4):271–4. Knibbs LD, de Dear RJ, Atkinson SE. Field study of air change and flow rate in six automobiles. Indoor Air 2009;19:303–13. Knibbs LK, Cole-Hunter T, Morawska L. A review of commuter exposure to ultrafine particles and its health effects. Atmospheric Environment 2011;45:2611–22. Le Moulle Y, Laurent AM, Person A. City-dwellers exposure to atmospheric pollutants originated by cars when commuting in Paris urban area. World Clean Air Congress, vol. 6. ; 1998. [17A-2.1-17A-2.6]. McNabola A, Broderick BM, Gill LW. Relative exposure to fine particulate matter and VOCs between transport microenvironments in Dublin. Personal exposure and uptake. Atmos Environ 2008;42(26):6496–512. McNabola A, Broderick BM, Gill LW. A principal components analysis of the factors effecting personal exposure to air pollution in urban commuters in Dublin, Ireland. J Environ Sci Health A Tox Hazard Subst Environ Eng 2009a;44(12):1219–26.

797

McNabola A, Broderick BM, Gill LW. The impacts of inter-vehicle spacing on in-vehicle air pollution concentrations in idling urban traffic conditions. Transp Res Part D: Transp Environ 2009b;14(8):567–75. Molle R, Mazoue S, Gehin E, Ionescu A. Indoor-outdoor relationships of airborne particles and nitrogen dioxide inside Parision buses. Atmos Environ 2013;69:240–8. Morawska L, Ristovski Z, Jayaratne ER, Keogh DU, Ling X. Ambient nano and ultrafine particles from motor vehicle emissions: characteristics, ambient processing and implications on human exposure. Atmos Environ 2008;42:8113–38. Moreno T, Querol X, Alastuey A, Viana M, Gibbons W. Profiling transient daytime peaks in urban air pollutants: city centre traffic hotspot versus urban background concentrations. J Environ Monit 2009;11:1535–42. Nieuwenhuijsen MJ, Gomez-Perales JE, Colvile RN. Levels of particulate air pollution, its elemental composition, determinants and health effects in metro systems. Atmos Environ 2007;41:7995–8006. Nwokoro C, Ewin C, Harrison C, Ibrahim M, Dundas I, Dickson I, et al. Cycling to work in London and inhaled dose of black carbon. Eur Respir J 2012;40(5):1091–7. O'Donoghue RT, Gill LW, McKevitt RJ, Broderick B. Exposure to hydrocarbon concentrations while commuting orexercising in Dublin. Environment International 2007;33: 1–8. Parra MA, Elustondo D, Bermejo R, Santamaría JM. Exposure to volatile organic compounds (VOC) in public buses of Pamplona, Northern Spain. Sci Total Environ 2008; 404(1):18–25. Pfeifer GD, Harrison RM, Lynam DR. Personal exposures to airborne metals in London taxi drivers and office workers in 1995 and 1996. Sci Total Environ 1999;235:253–60. Praml G, Schierl R. Dust exposure in Munich public transportation: a comprehensive 4 year survey in buses and trams. Int Arch Occup Environ Health 2000;72:209–14. Puchewr J, Buehler R. Making cycling irresistible: lessons from the Netherlands, Denmark and Germany. Transp Rev 2008;28:495–528. Querol X, Moreno T, Karanasiou A, Reche C, Alastuey A, Viana M, et al. Variability of levels and composition of PM10 and PM2.5 in the Barcelona metro system. Atmos Chem Phys 2012;12:5055–76. Rabl A, deNazelle A. Benefits of shift from car to active transport. Transp Policy 2012;19: 121–31. Ragettli MS, Corradi E, Braun-Fahrländer C, et al. Commuter exposure to ultrafine particles in different urban locations, transportation modes and routes. Atmos Environ 2013;77(1352–2310):376–84. Rank J, Folke J, Jespersen PH. Differences in cyclists and car drivers exposure to air pollution from traffic in the city of Copenhagen. Sci Total Environ 2001;279(1–3):131–6. Raut J-C, Chazette P, Fortain A. Link between aerosol optical, microphysical and chemical measurements in an underground railway station in Paris. Atmos Environ 2009;43: 860–8. Ripanucci G, Grana M, Vicentini L, Magrini A, Bergamaschi A. Dust in the underground Railway Tunnels of an Italian Town. J Occup Environ Hyg 2006;3:16–25. Rojas-Rueda D, de Nazelle A, Tainio M, Nieuwenhuijsen MJ. Bike sharing system (Bicing) in Barcelona, Spain: a description and health impacts assessment. Br Med J 2011; 343(7819). [Article number d4521]. Rojas-Rueda D, de Nazelle A, Teixido O, Nieuwenhuijsen MJ. Replacing car trips by increasing bike and public transport in the greater Barcelona metropolitan area: a health impact assessment study. Environ Int 2012;49:100–9. Salma I, Weidinger T, Maenhaut W. Time-resolved mass concentration, composition and sources of aerosol particles in a metropolitan underground railway station. Atmos Environ 2007;41:8391–405. Salma I, Pósfai M, Kovács K, Kuzmann E, Homonnay Z, Posta J. Properties and sources of individual particles and some chemical species in the aerosol of a metropolitan underground railway station. Atmos Environ 2009;43:3460–6. Seaton A, Cherrie J, Dennekamp M, Donaldson K, Hurley JF, Tran CL. The London underground: dust and hazards to health. Occup Environ Med 2005;62:355–62. Strak M, Boogaard H, Meliefste K, Oldenwening M, Zuurbier M, Brunekreef B, et al. Respiratory health effects of ultrafine and fine particle exposure in cyclists. Occup Environ Med 2010;67(2):118–24. van Wijnen JH, Verhoeff AP, Jans HWA, van Bruggen M. The exposure of cyclists, car drivers and pedestrians to traffic-related air pollutants. Int Arch Occup Environ Health 1995;67:187. WHO. Health effects of transport-related air pollution. Copenhagen: WHO Regional Office for Europe; 2005. WHO. Review of evidence on health aspects of air pollution — REVIHAAP. Copenhagen: WHO Regional Office for Europe; 2013. Woodcock J, Edwards P, Tonne C, Armstrong BG, Ashiru O, Banister D, et al. Public health benefits of strategies to reduce greenhouse-gas emissions: urban land transport. Lancet 2009;374(9705):1930–43. Yang L, Sahlqvist S, McMinn A, Griffin SJ, Ogilvie D. Interventions to promote cycling: systematic review. Br Med J 2010;341:5293–303. Zuurbier M, Hoek G, Hazel PVD, Brunekreef B. Minute ventilation of cyclists, car and bus passengers: an experimental study. Environ Health 2009;8(1):48. Zuurbier M, Hoek G, Oldenwening M, Lenters V, Meliefste K, van den Hazel P, et al. Commuters' exposure to particulate matter air pollution is affected by mode of transport, fuel type, and route. Environ Health Perspect 2010;118(6):783–9. Zuurbier M, Hoek G, Oldenwening M, Meliefste K, Krop E, van den Hazel P, et al. In-traffic air pollution exposure and CC16, blood coagulation, and inflammation markers in healthy adults. Environ Health Perspect 2011;119(10):1384–9.