A comparison of personal exposure to air pollutants in ...

Science of the Total Environment 619–620 (2018) 155–164

Contents lists available at ScienceDirect

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

A comparison of personal exposure to air pollutants in different travel modes on national highways in India Soma Sekhara Rao Kolluru a, Aditya Kumar Patra b,⁎, Satya Prakash Sahu b a b

School of Environmental Science and Engineering, Indian Institute of Technology Kharagpur, India Department of Mining Engineering, Indian Institute of Technology Kharagpur, India

H I G H L I G H T S

G R A P H I C A L

A B S T R A C T

• Personal pollutant exposure on a national highway was measured in car and bus. • Highest PM2.5 and CO exposures were obtained in car and car (ac), respectively. • Travel modes and ventilation settings significantly influenced the exposure levels. • In-city mass exposure levels were up to 1.3–2.2 times that of the open highways. • Exposure studies on highways for passenger health risk assessment are emphasized.

a r t i c l e

i n f o

Article history: Received 20 August 2017 Received in revised form 25 October 2017 Accepted 8 November 2017 Available online xxxx Editor: P. Kassomenos Keywords: Concentrations exposure PM2.5 CO Mass exposures Ventilation setting

a b s t r a c t People often travel a long distance on highways to the nearest city for professional/business activities. However, relatively few publications on passenger exposure to pollutants on highways in India or elsewhere are available. The aim of this study was to examine the contribution of different travel modes to passengers' pollutant exposure for a long distance travel on a national highway in India. We measured PM2.5 and CO exposure levels of the passengers over 200 km on a national highway using two portable air monitors, EVM-7 and EPAM-5000. Personal concentration exposures and per min-, per hour-, per trip- and round trip mass exposures for three travel modes were calculated for 9 trips. Association between pollutants and weather variables were evaluated using levels Spearman correlation. ANOVA was carried out to evaluate the influence of travel mode, the timing of trips, temperature and RH on personal exposures. On an average, PM2.5 personal concentration exposure levels were highest in the car (85.41 ± 61.85 μg m−3), followed by the bus (75.08 ± 55.39 μg m−3) and lowest in the car (ac) (54.43 ± 34.09 μg m−3). In contrast, CO personal exposure was highest in the car (ac) (1.81 ± 1.3 ppm). Travel mode explained the highest variability for CO (18.1%), CO2 (9.9%), PM2.5 (1.2%) exposures. In-city mass exposures were higher than trip averages; PM2.5:1.21–1.22, 1.13–1.19 and 1.03–1.28 times; CO: 1.20–1.57, 1.37– 2.10 and 1.76–2.22 times for bus, car and car (ac) respectively. Traveling by car (ac) results in the lowest PM2.5 exposures, although it exposes the passenger to high CO level. Avoiding national highways passing through cities can reduce up to 25% PM2.5 and 50% CO mass exposures. This information can be useful for increasing environmental awareness among the passengers and for framing better pollution control strategies on highways. © 2017 Elsevier B.V. All rights reserved.

Abbreviations: PM2.5, particulate matter of aerodynamic diameter ≤ 2.5 μm; CO, carbon monoxide; CO2, carbon dioxide; RH, relative humidity; ANOVA, analysis of variance; BCM, Bhadrachalam; KTDM, Kothagudem; TVR, Tiruvur; IBP, Ibrahimpatnam; VJA, Vijayawada; Car (ac), car using air-condition; K-S test, Kolmogorov-Smirnov test; Et, mass exposure; Ci, personal concentration level in ith minute; IR, inhalation rate; n, exposure duration. ⁎ Corresponding author. E-mail address: [email protected] (A.K. Patra).

https://doi.org/10.1016/j.scitotenv.2017.11.086 0048-9697/© 2017 Elsevier B.V. All rights reserved.

1. Introduction Road transport sector ranks among the top anthropogenic sources for PM2.5 (particulate matter of aerodynamic diameter ≤ 2.5 μm) and CO (carbon monoxide.) Tailpipe emissions, brake, and tire wear emissions, windblown particles from open trucks and re-suspended road

156

S.S.R. Kolluru et al. / Science of the Total Environment 619–620 (2018) 155–164

dust are the major traffic-related sources of PM2.5 (Laden et al., 2000). It was estimated that road traffic-related sources are responsible for N 50% of the total PM2.5 emissions in highly developed countries. Studies in six Indian cities consisting of Bangalore, Chennai, Delhi, Kanpur, Mumbai, and Pune revealed that 30–50% of the ambient particulate matter originates from the vehicular exhaust and re-suspended road dust (CPCB, 2010). Long-term exposure to PM2.5 can induce adverse health effects and can cause cardiovascular mortality via accelerated atherosclerosis, altered cardiac autonomic function, pulmonary inflammation, dysrhythmias, ischemic heart disease, heart failure, and cardiac arrest (Miller et al., 2007; Pope et al., 2009). Increased hospital admission rates for diagnosis of chronic obstructive pulmonary diseases and ischemic heart diseases were associated with short-term PM2.5 exposures (Dominici et al., 2006). Studies found 0.98% increase in all mortality, 8% increase in mortality due to lung cancer and diabetes, 1.68% increase in respiratory deaths and 6% increase in cardiopulmonary mortality for PM2.5 exposure increment of 10 μg m−3 (Pope et al., 2002; Ostro et al., 2006; Zanobetti and Schwartz, 2009). CO is a toxic gas which is the by-product of incomplete combustion of fossil fuels which is characterized by no color, no odor, and no taste. Road transportation sector is a significant source of anthropogenic CO (Flachsbart, 1999). At low concentrations, health effects range from subtle cardiovascular and neurobehavioral effects to unconsciousness and death after acute or chronic exposure to higher levels (Raub et al., 2000). CO when combined with hemoglobin, reduces the capacity to transport oxygen thus causing impaired concentration and confusion (Kampa and Castanas, 2008). Health effects such as exacerbating myocardial ischemia, worsen cardiovascular disease and mortality are caused by inhaling CO (Riojas-Rodríguez et al., 2006; Min et al., 2009). The exposure in onroad micro-environments is frequently higher than the limit values prescribed in air quality standards (Kaur et al., 2007; Cepeda et al., 2017). Exposure levels in transport microenvironments are influenced by various parameters such as mode of transport (Kingham et al., 1998; Adams et al., 2001; Gulliver and Briggs, 2004; Kaur et al., 2005; Briggs et al., 2008; McNabola et al., 2008; Kaur and Nieuwenhuijsen, 2009), condition of ventilation (Chan et al., 2002; Chan and Chung, 2003; Geiss et al., 2010; Zuurbier et al., 2010; Onat and Stakeeva, 2013; Wu et al., 2013), routes of the study (Adams et al., 2001; Adar et al., 2007; McNabola et al., 2008; Ramos et al., 2016), meteorology of the location (Wu et al., 2013) and self-pollution inside vehicle cabins (Marshall and Behrentz, 2005; Sabin et al., 2005; Zhang and Zhu, 2010; Abi-Esber and El-Fadel, 2013; Hudda and Fruin, 2013). In India, fast industrialization and urbanization over the last two decades led to the growth of road transportation sector. Around 12.42 million workers in India commuted to urban locations each day during 2009–10 (Sharma and Chandrasekhar, 2014). Road networks that include national highways, state highways, urban and other roads increased from 3,99,942 km in 1951 to 54,72,144 km in 2015, at a compound annual growth rate of 4.2% (MoRTH, 2016). National highways (NH) are the major road networks which connect various parts of the country. As of June 2016, there are approximately 620 NH in India with a total length of 100,087 km. These roads are only 1.7% of the whole road network, but carries 40% of the total road traffic in India (NHAI, 2017). In cities, the traffic moves at a slower speed due to higher traffic density and frequent halting at traffic signals. These conditions result in more exhaust emission per kilometer length of the road. Also, the presence of high rise buildings in cities does not allow for dilution of emissions and this can lead to increased PM2.5 and CO concentration in city road networks. On the other hand, highways are characterized by free flowing traffic, fewer traffic signals and higher speeds of the vehicle. The presence of open space along the highway surroundings enables greater dispersion of PM2.5 and CO emitted. Thus the exposure pattern in highways differs from exposure in cities. While many studies have been conducted on personal concentrations from traffic in Indian cities (Apte et al., 2011; Sabapathy et al., 2012; Namdeo et al., 2014; Goel et

al., 2015; Kumar and Gupta, 2016; Pant et al., 2017), limited studies are available for passengers exposure on highways (Hsu and Huang, 2009; Huang and Hsu, 2009). The aim of the study was to evaluate and compare the personal concentrations of PM2.5 and CO in different traffic microenvironments over a length of 400 km on a national highway in India. Effect of timing of trips and meteorological parameters was investigated. Additionally, pollutants mass exposures were calculated for different travel modes and trip timings. To our knowledge, this is the first long distance personal exposure study in India, which has been conducted on national highways. 2. Materials and methods 2.1. Study route The study was carried out on a busy NH (national highway) of the length of 200 km, connecting Bhadrachalam (17°40′00.42″N, 80°53′ 06.90″E) in Telangana State with Vijayawada (16°29′58.77″N, 80°38′ 54.21″E) in Andhra Pradesh State, India. The highway segment under study connects two major pilgrimage centers of South India and important business locations. The study route consists of 170 km in NH 30, from Bhadrachalam (BCM) to Ibrahimpatnam (IBP) and 30 km in NH 65, from Ibrahimpatnam to Vijayawada (VJA). Bhadrachalam is an important Hindu pilgrimage town in the Telangana State with a population of 89,048 (Census, 2011). Indian Tobacco Corporation operates one of its paper industries in this town. Telangana State Road Transport Corporation governs a bus terminal in Bhadrachalam from where buses, one in every 15–20 min, depart to Vijayawada traveling on the study route. Vijayawada is the major trading and pilgrimage city of the Andhra Pradesh State with the population of 1,328,988 (Census, 2011). The bus terminal in Vijayawada is the fourth largest and busiest in India. Andhra Pradesh Generation Corporation operates one of its biggest thermal power plant of capacity 1760 MW in this city. Nearly 70 major industrial units and many small to medium scale industrial units are located around the city. The city is characterized by heavy traffic, similar to many urban locations. This study route passes through three major towns: Kothagudem (KTDM) – Tiruvur (TVR) – Ibrahimpatnam (IBP) (Fig. 1). At all the three locations, state owned bus companies operate public buses which depart to Vijayawada at frequent intervals. Apart from public buses, trucks, bikes, cars, and auto-rickshaws also use the highway. Kothagudem is a district headquarters which is located at a distance of ~40 km from Bhadrachalam in Telangana State. The Singareni Collieries Company Limited, Nava Bharat Ferro Alloy plant, Nava Bharat Power Plant, sponge iron plant of National Mineral Development Corporation and Kothagudem Thermal Power Station of Telangana State Generation Corporation with a capacity of 1700 MW are established in the town. Tiruvur is located at ~ 60 km from the Kothagudem in the Andhra Pradesh State. No significant industries are located in and around this township. Ibrahimpatnam is located at a distance of ~ 70 km from Tiruvur in the Andhra Pradesh State. Several industries are located around this town. At Ibrahimpatnam, NH 30 joins NH 65. The study route bypasses the towns, except at Kothagudem where it passes through the town. Passengers traveling regime on this NH usually comprise of travel from Bhadrachalam to Vijayawada and return on the same day, one-way trip from Bhadrachalam to Vijayawada or from Vijayawada to Bhadrachalam and travel between the three major towns. During the sampling period, road repairing and extension works were undergoing throughout the highway. Sand and stones for road construction were seen littered on either sides of the highway. 2.2. Measurement of PM2.5 and CO An Environment Particulate Air Monitor (Model: EPAM – 5000; Environment Devices Corporation) was used to measure PM2.5 exposures


157

Fig. 1. Study route. (Source: Google Earth.)

inside bus/car. This instrument uses the near-forward light scattering of infrared radiation. The equipment was calibrated with gravimetric reference NIST Traceable - SAE fine test dust - ISO12103-1 (International Organization for Standardization, Geneva, Switzerland). It has higher correlations with TEOM (tapered element oscillating microbalance) and MiniVol. EPAM-5000 complements both Environmental Protection Agency (EPA) and Occupational Safety and Health Administration (OSHA) reference methods. It has the precision of ±0.003 mg m−3 with the sampling time interval range 1 s–30 min. It can run for 22 h on a single battery charge and can store data up to 15 months. During the study, EPAM-5000 was operated at a flow rate of 4 l per minute (range: 1–5 l per minute). Data were retrieved using DustComm Pro software. CO, CO2, temperature and relative humidity were measured along the route using portable multi-parameter Environmental Monitor (Model: EVM-7; Quest Technology–3M, USA). It uses the non-dispersive infrared sensor for measuring CO2, junction diode and capacitive sensor to measure temperature and relative humidity. 2.3. Study design and quality control In-vehicle concentrations were measured in the bus and the car, two most used travel modes on the study route. The buses (make: Ashok Leyland; model: between 2012 and 2013) were 55 seaters and naturally ventilated (adjustable windows). The four seater car (make: Tata Indigo; model: 2007) was used in two ventilation settings: no air-conditioning with windows open; air-conditioning in recirculation mode with windows closed (car (ac)). All the buses and car were diesel powered. Smoking was prohibited in the public buses. During the regular bus services, two non-smoking researchers took two consecutive middle window seats on the driver side. Instruments were placed on the laps of the researchers and windows were opened ~10–15 cm wide. Researchers had no control on remaining window positions in the bus. The front

door of the bus was always kept open for passengers' alighting and boarding. Researchers behaved like regular passengers in the bus. In the car, the instruments were placed on the cardboard box which was kept on the rear seat to mimic breathing position of travelers (see Fig. S1 in Supplementary information). On the sampling days, the exposure measurements were carried out from Bhadrachalam to Vijayawada during the morning hours and Vijayawada to Bhadrachalam during the evening hours. Morning trips started at 8:00 h and evening trips started at 15:00 h. Speed levels of 50–70 km h− 1 were maintained for the bus and the car. Typically, it took ~ 5 h for the one-way trip. The combination of one morning trip and one evening trip on a sampling day is termed as a round trip. In total, 9 round trips were made with 6 trips in each mode (bus, car, and car (ac)) during July–August 2016. Although these months represent the Indian monsoon season, we encountered no rain on our study days. The bus exposure study also included the times the bus halted at Kothagudem and Tiruvur bus terminals (~ 10 min) and at road side stoppages all along the length of highway. The buses drove for some time on local streets of the towns before entering and after exiting the bus terminals. However, during the car trips, study route was strictly followed. At different intermediate stoppages, passengers alighted and boarded the bus. On an average, nearly 30 passengers remain seated in buses at any time throughout the trips. Approximately 60 min traveling time were spent on local streets for the one-way trip in the bus. During the morning trips, the sampling was started at Bhadrachalam bus terminal and terminated at Vijayawada bus terminal and vice versa, in the evening trips. Battery power was checked, and manual zero setting was performed before sampling. One-minute interval readings were recorded by the instruments and data were retrieved into PC in MS Excel format on the same day after completion of the round trip. Hand watches were synchronized with the time setting in the instruments to determine entry and exit times of towns, bus terminals, and

158


to observe obvious/visible pollution sources at different locations. Passenger counts in the buses at each bus terminal were also noted. 2.4. Data analysis Due to heavy road construction activity during a second-morning trip in the car, vehicles were diverted to another route. Therefore, data from this trip was not considered for analysis. There were no missing data for the remaining trips. Descriptive statistics were calculated for the datasets for different travel modes and timing of the trips. K-S (Kolmogorov-Smirnov) test was used for testing the normality of the data. Mann-Whitney U test was used to assess variation in personal concentrations exposure between morning trips and evening trips in a travel mode. Kruskal-Wallis H test was used to check the exposure variation in different travel modes during morning and evening trips. Spearman's correlation analysis was conducted to find out associations among temperature, relative humidity, and exposure levels of PM2.5, CO, and CO2. Additionally, ANOVA analysis was performed to quantify variations in personal exposures due to temperature, relative humidity, travel mode and timing of trips. Tests were considered significant if a level of significance, p b 0.05. Analyses were conducted using statistical software package IBM SPSS Statistics 22.0 (IBM, Armonk, NY, USA). Additionally, passenger mass exposures were also estimated. Mass of the pollutants inhaled (mass exposure) depends on the inhalation rate of the passenger, personal concentration levels inside vehicles and travel time. For each travel mode, the mass exposures were calculated using the following equation: n

Et ¼ ∑i¼1 C i T IR

where, Et = mass exposure (PM2.5 in μg and CO in mg) Ci = personal concentration level in ith minute (PM2.5 in μg m−3 and CO in mg m−3) T = 1 min (exposure logging interval) IR = inhalation rate (1.48 × 10−2 m3 min−1) n = exposure duration (min) EPA has evaluated different inhalation rates for the different activity levels ranging from nap or sleep to high-intensity. Traveling by the car and the bus comes under light intensity activity (O'Donoghue et al., 2007; Huang et al., 2012). Inhalation rates for people above 20 years of age for light intensity activity are approximately 1.48 × 10−2 m3 min−1 (USEPA, 2011). 3. Results

Fig. 2. Temperature and RH (average of nine round trips).

3.2. Personal concentrations exposure Personal concentrations exposure profile of PM2.5 during three trips in each travel mode followed an identical pattern, although exposure level varied across the trips (see Fig. S2 in Supplementary information). The exposure level was highest in the car and lowest in the car (ac) (Table 2). During morning trips, mean personal exposures to PM2.5 concentrations for the bus, the car, and the car (ac) were 75.08 ± 55.39 μg m−3, 85.41 ± 61.85 μg m−3, and 54.43 ± 34.09 μg m−3 and during evening trips were 67.46 ± 44.84 μg m−3, 92.25 ± 69.35 μg m−3, and 53.59 ± 31.61 μg m−3 respectively. Exposure levels in the car were ~1.7 times of exposures in the car (ac) and ~1.4 times of the exposure levels in the bus during morning trips and evening trips. CO concentration levels were the highest in the car (ac), with values of 1.81 ± 1.3 ppm for morning trips and 1.29 ± 1.14 ppm for evening trips. Lowest CO concentrations were observed in the bus for both the trips. CO2 concentration levels were the highest in the car (ac) for both trips and lowest in the car. K-S test showed that the datasets in three travel modes followed the non-normal distribution. Kruskal-Wallis H test was performed to evaluate the role of travel modes on personal concentration levels irrespective of the timing of trips. It showed that travel mode is statistically a significant parameter that influences PM2.5, CO, and CO2 concentration levels both in morning trips and evening trips separately. For different timing of trips, the Mann-Whitney U test showed that PM2.5 exposure levels varied significantly in bus and car, CO varied significantly across all three travel modes and CO2 varied significantly in the bus and the car (ac). Fig. 3 shows the differences in concentration levels for PM2.5 for different travel modes during morning and evening trips.

3.1. Weather parameters The temperature inside the travel modes increased in morning trips until 13:00 h and then decreased in evening trips. As expected, relative humidity gradually decreased in morning trips and subsequently increased during evening trips (Fig. 2). The weather was relatively stable on sampling days with temperature and RH varying in the range of 20.00–41.70 °C and 32.40–91.40% respectively. RH inside the bus was higher than the RH inside car and car (ac). The reverse pattern has occurred for the temperature. The possible reason might be the heat radiation from the roof of a vehicle. The roof of the car is much closer to the sampling instrument than that of the bus, and this has resulted in higher temperature inside the car than inside the bus. Weather parameters did not vary significantly across different timing of trips, but the variations were significant across different travel modes (Table 1).

3.3. Concentration exposure variation during a typical trip The pattern of concentration variability in a typical trip (in this case, first morning trip in the car from BCM to VJA) is displayed in Fig. 4. The local background level of PM2.5 and CO throughout the trip had a range of 20–50 μg m−3 and 1 ppm respectively. As the towns (Kothagudem and Vijayawada) approached, the concentration levels increased and remained high until the car leaves the town limits, primarily because of the rise in traffic density. This increase is also corroborated by the higher CO level (CO is the tracer for vehicular emissions) when the car is passing through the towns. The increase in PM2.5 and CO concentrations is more pronounced at Vijayawada than other places. Tiruvur is the exception because majority of the traffic bypasses the city. Between


159

Table 1 Descriptives of weather parameters, (AM ± SD). Parameters

Travel mode

Temperature (°C) RH (%)

Timing of trips

Bus

Car

Car (ac)

Morning

Evening

30.13 ± 2.46⁎ 75.25 ± 7.91⁎

33.05 ± 2.65⁎ 60.50 ± 9.17⁎

29.90 ± 3.30⁎ 58.05 ± 12.19⁎

31.04 ± 3.10 64.85 ± 10.97

30.89 ± 3.22 64.68 ± 13.93

⁎ Temperature and RH vary across three travel modes with statistically significant difference (Kruskal-Wallis H test with a level of significance p b 0.05); there was no statistically significant difference in two trip timings (Mann-Whitney U test with a level of significance p b 0.05).

two towns, the concentration levels were lower because of the free flowing traffic and open fields on either sides of the highway resulted in easy dispersion of pollutants emitted. Several short-interval spikes in PM2.5 concentrations, which are not accompanied by proportionate CO rise, might be the signatures of re-suspended road dust. All along the route, road repair/relaying works were in progress during the study period. A similar trend observed for all the trips in three modes.

ANOVA was carried to find out the variability (%) of the travel mode, time of the trip, temperature and RH measured in the study on the pollutants (Table 4). Travel mode explained the highest variability for the pollutants: CO (18.1%), CO2 (9.9%) and PM2.5 (1.2%). Timing of the trips explained very low variability among all the factors. However, it was significant for CO (0.3%) and CO2 (0.3%). Trip timings did not add significant influence on PM2.5 personal exposures. Temperature explained significant variability for all pollutants with higher values for CO (8.8%) and CO2 (5.3%). RH explained higher variability for CO (16%). It did not show any effect on PM2.5.

passenger travel through the two bigger locations (Kothagudem and Vijayawada) on the study route. It indicates that in-route towns result in higher mass exposure than the free-flowing section of the highways. In the case of CO, the mass exposure ranges were 0.26–1.09 mg, 0.82– 1.99 mg, 0.83–3.18 mg and exposure levels in Kothagudem and Vijayawada were 1.20–1.57, 1.37–2.10 and 1.76–2.22 times the average exposure levels in the bus, the car and the car (ac), respectively (Fig. 5). A passenger traveling from Bhadrachalam to Vijayawada inhales on average 1.06, 1.09, and 1.25 times the mass exposure of PM2.5 that he will inhale while he travels back in the same route by bus, car, and car (ac) respectively (Table 5). Despite the car taking ~50 min less time to travel the same distance, the PM2.5 round trip mass exposures in the car are ~ 20% higher than the mass exposure during the bus round trips. As expected, the round trip PM2.5 mass exposures are lowest in the car (ac) mode. A passenger traveling from BCM to VJA inhales on average 1.12, 1.28, 1.60 times CO that he will inhale while he travels back in the same route by bus, car, and car (ac) respectively. However, for CO, round trip exposures are highest in the car (ac) (14.32, 13.24–15.41), which are 14% higher than mass exposure in the car. As mentioned earlier, not all the passengers on this route travelled the entire distance from Bhadrachalam to Vijayawada in all three modes. Therefore, PM2.5 and CO mass exposures of these passengers depend on the time they spent in a particular travel mode. We, therefore, calculated the average per minute mass exposures of PM2.5 and CO in all the modes so that short trip exposures are estimated. The average PM2.5 exposures were 2.30 μg min−1, 2.76 μg min−1 and 1.50 μg min−1 for bus, car, and car (ac) respectively. The estimates show that per minute PM2.5 mass exposures of the car were 20% more that of the bus. The average CO exposures were 0.02 mg min−1, 0.05 mg min− 1and 0.06 mg min− 1 for bus, car, and car (ac) respectively. The estimates show that per minute CO exposure in car (ac) was 20% higher than in car.

3.6. Mass exposures

4. Discussion

Hourly PM2.5 mass exposures varied in the range of 45.73–92.13 μg, 49.42–93.71 μg, and 33.35–72.21 μg for the bus, the car and the car (ac), respectively. The mass exposures were higher by factors of 1.21–1.22, 1.13–1.19 and 1.03–1.28 of the average exposure levels while a

PM2.5 concentration exposures were the highest for car users and the lowest for the car (ac) users. Car (ac) users experienced the highest CO exposure. CO2 concentrations exposures were the highest in the car (ac) and lowest in the car. Exposure levels exhibited statistically

3.4. Correlation between pollutants and weather variables Spearman correlations were used to evaluate the relationships of pollutants concentration levels with the weather variables. Very weak associations were observed among the pollutants (Table 3). These indicate that the pollutants measured were originating from many emission sources on the highway with different proportions, e.g., CO originates exclusively from the vehicles. PM2.5 originates from exhaust emissions of the vehicles, road dust resuspension and industrial emissions. Temperature and RH exhibited a moderate association with CO and CO2. 3.5. Variability in personal concentration exposure levels by different parameters

Table 2 Descriptives of pollutant concentrations, (AM ± SD). Travel modes

Bus Car Car (ac) a b c d e f i g

Morning trips

Evening trips

PM2.5 (μg m−3)

CO (ppm)

CO2 (ppm)

PM2.5 (μg m−3)

CO (ppm)

CO2 (ppm)

75.08 ± 55.39a,b 85.41 ± 61.85a,c 54.43 ± 34.09a

0.80 ± 0.61a,d 1.34 ± 1.03a,e 1.81 ± 1.3a,f

557.22 ± 188.90a,i 483.73 ± 31.38a 857.74 ± 268.8a,g

67.46 ± 44.84a,b 92.25 ± 69.35a,c 53.59 ± 31.61a

0.59 ± 0.84a,d 1.20 ± 1.20a,e 1.29 ± 1.14a,f

583.10 ± 254.80i 487.24 ± 37.8a 826.76 ± 238.51a,g

All pollutants vary across the travel modes with statistical significance difference in morning trips and evening trips (Kruskal-Wallis H test, p b 0.05). PM2.5 varies with a statistically significant difference during morning and evening trips in the bus (Mann-Whitney U test, p b 0.05). PM2.5 varies with a statistically significant difference during morning and evening trips in the car (Mann-Whitney U test, p b 0.05). CO varies with a statistically significant difference during morning and evening trips in the bus (Mann-Whitney U test, p b 0.05). CO varies with a statistically significant difference during morning and evening trips in open window car (Mann-Whitney U test, p b 0.05). CO varies with a statistically significant difference during morning and evening trips in the air-conditioned car (Mann-Whitney U test, p b 0.05). CO2 varies with a statistically significant difference during morning and evening trips in the bus (Mann-Whitney U test, p b 0.05). CO2 varies with a statistically significant difference during morning and evening trips in the air-conditioned car (Mann-Whitney U test, p b 0.05).

160


Fig. 3. Boxplots of PM2.5 personal concentration levels in (a) morning trips and (b) evening trips segregated by travel mode. Boxes represent data between 25th and 75th percentile, the dot inside the box is the mean, the central line represents the median, whiskers represent data under 1.5 times the interquartile range.

significant difference for pollutants across the travel modes. Exposure levels in the bus and the car showed statistically significant difference between the timing of the trips (morning vs. evening). For car (ac) mode, there was no significant difference in exposure level for the timing of the trips. Travel modes explained the largest variability for CO. As per our understanding, this is the first study on the personal exposure measurement that has considered longer study route on the NH in which three travel modes were used. A comparison of our study with other traffic exposure studies in cities that reported PM2.5 and CO exposures levels are tabulated in Tables S1 and S2 in Supplementary information. 4.1. Personal concentrations variation in travel modes In personal exposure studies, there was no substantial evidence to state that a particular travel mode will always experience the highest or lowest exposure levels (Kaur et al., 2007). In our study, passengers

were exposed to the PM2.5 concentrations levels exceeding the permissible limits of WHO (25 μg m−3, (WHO and UNAIDS, 2006)) and NAAQS (60 μg m−3, (CPCB, 2009)), India. Due to traffic congestion and industrial emissions around Kothagudem and Vijayawada, higher exposure levels were observed in all the travel modes. In addition to vehicular sources, re-suspended dust due to road repairing and extension works on the entire stretch of the highway at the time of study might have contributed to the exposure level. The highest average PM2.5 personal concentrations were recorded in the car followed by the bus and lowest in the car (ac). Our finding agrees with the earlier studies where personal concentrations were measured across different travel modes. A study in Barcelona concluded that open window car users were exposed to higher PM2.5 concentration than other modes (de Nazelle et al., 2012). Study in Delhi showed that open window car users were mostly exposed to alveolic particles (Kumar and Gupta, 2016). The possible reasons for higher exposure levels in the car are as follows: (1) Due to its smaller size, a car often penetrates into the center of the traffic congestion where the pollutant concentration is high. (2) The vertical distance between sitting position and the tailpipe/road is lesser in the car in comparison to that of the bus. It results in infiltration of the exhaust emissions of surrounding vehicles and re-suspended dust before getting appreciably diluted. (3) The higher speed of the car allows more inflow of polluted air into the cabin through open windows. Therefore, despite shorter travel time, concentrations levels were higher in open window car. PM2.5 exposure levels in buses were usually lower than exposures in the car because of lower speed, higher cabin height and more ventilation (due to large volume) inside the bus. Some scenarios that resulted Table 3 Spearman correlation between pollutants and weather variables.

PM2.5 CO CO2 Temperature RH Fig. 4. PM2.5 and CO personal concentrations during a typical car morning trip.

⁎ p b 0.05. ⁎⁎ p b 0.01.

PM2.5

CO

CO2

Temperature

RH

1 −0.4⁎ −0.099⁎⁎ 0.026 −0.002

−1 –0.013 0.349⁎⁎ −0.478⁎⁎

1 −0.22⁎⁎ −0.066⁎⁎

1 −0.601⁎⁎

1

S.S.R. Kolluru et al. / Science of the Total Environment 619–620 (2018) 155–164 Table 4 Explained variability (%) of travel mode, the timing of trips, temperature and RH on personal concentration levels. Pollutants

Parameters Mode

PM2.5 CO CO2

1.2⁎

18.1⁎ 9.9⁎

Timing of trips

Temperature

RH

0.1 0.3⁎ 0.3⁎

0.6⁎ 8.8⁎ 5.3⁎

0.0 16.0⁎ 0.2⁎

⁎ p b 0.05.

in an increased passenger exposure in the bus include the following: (1) When a bus entered a town, there was an additional movement of passengers inside the bus for alighting and boarding the bus. This resulted in re-suspension of dust from the bus floor (visible dust load on the bus floor observed during the study). (2) At the entrance to a bus terminal, buses queued up very close to each other with the engine idling. Unlike in case of moving traffic, there was no action of vehicle-induced turbulence in surrounding air for dispersion of pollutants. This may have caused an increase in the exposure levels (Richmond-Bryant et al., 2009). (3) The front engine hoods of the buses were loosely latched. During idling condition, emissions from crank case leaked into the cabin and dilution was poor due to less inflow of fresh air (Zhang and Zhu, 2010). Lowest passenger exposures were obtained in the car (ac). The closing of the windows barred the infiltration of outside air and exhaust emissions into car cabin. Additionally, recirculation mode of air conditioning prevented the entry of outside air into the cabin. A study in Barcelona concluded that ventilation system combined with air conditioning unit improve air quality inside vehicle cabins (Querol et al., 2012). Earlier studies conducted in cities had reported that car operated with ac recorded lower PM2.5 concentrations than the bus (Huang et al., 2012). Our findings show that exposure in the highways also follows the similar relative trend. We observed increased exposure levels when the car (ac) approached and passed through the town/city (Kothagudem and Vijayawada), although the increase was not prominent as in the case of open window car. This could be due to leaks and cracks in the car body (the car used for this study was ten years old), which allowed the outside emissions to enter the cabin. An entirely different trend was observed for CO exposure. During the morning and evening trips, highest mean CO exposures were recorded

161

in the car (ac) followed by the car and lowest in the bus. Several studies observed higher CO concentrations in the car (ac) mode. A study in Guangzhou found the highest CO concentrations in ac taxi when compared with other modes (Chan et al., 2002). In Tel Aviv, the highest CO concentration was recorded in the car with closed windows followed by the car with open windows and lowest in buses (Potchter et al., 2014). Mean CO concentrations in these two studies were higher than our finding because these studies were conducted in cities. Fewer traffic congestions on the highways, when compared to the cities, might be the reason for low CO exposure in our case. The operating conditions in the car (ac) have prohibited the entry of fresh air. Due to the aging of the vehicle, the CO from the engine crankcase leaks into the cabin. Limited airexchange through gaps and cracks in a car body is insufficient to dilute it. Further, lower cabin height might have provided easy access of CO released from tail pipes of the vehicles surrounding it to enter the cabin through the structural faults in the body. In contrary, fresh air exchange through open windows in the car and more volume of the bus cabin enabled dilution of the CO level, both from internal and external sources. This shows that choosing wisely the travel mode for long journeys on an NH can significantly reduce the personal concentrations exposures.

4.2. Influence of ventilation setting on personal concentrations The effect of ventilation on exposure level is shown in Fig. 6. For the PM2.5 during morning trips, up to 50% of the cumulative exposure levels were below 50–60 μg m−3 for car and the car (ac) and in evening trips, these were below 50–75 μg m−3 for the car and the car (ac). Therefore, the lower half of the exposure levels recorded in these two modes were similar. The difference between the car and the car (ac) exposure levels was more diverse for cumulative distributions N50%. For example, at 90% cumulative exposures, the difference in PM2.5 exposure levels between the car and the car (ac) was 80 μg m−3. At 100% cumulative exposures, PM2.5 in the car were twice the exposure levels in the car (ac). Statistical analysis also showed a significant difference in PM2.5 exposure levels between the car and the car (ac) modes. Therefore, high exposure levels can be effectively reduced by closing the windows and running the ac. The opposite occurred in the case of CO because emissions leakage from the engine into the cabin (self-pollution) is the primary source of CO in the car (ac) and it gets accumulated in the car cabin due to poor air exchange. Although in our case, the difference in CO exposure levels

Fig. 5. Hourly mass exposures: (a) PM2.5, (b) CO.

162


Table 5 Mass exposures of pollutants. Air pollutant

Travel modes

Timing

Number of trips

Mean concentration exposure during a tripa

Trip durationb

Inhalation ratec

Mass exposure per tripd

Mass exposure per round tripd

Mass exposure per minuted

PM2.5

Bus

Morning Evening Morning Evening Morning Evening Morning Evening Morning Evening Morning Evening

3 3 3 3 3 3 3 3 3 3 3 3

75.08 67.46 85.41 92.25 53.75 53.59 0.80 0.59 1.34 1.20 1.81 1.29

287 293 275 230 250 210 287 293 275 230 250 210

1.48 × 10−2 1.48 × 10−2 1.48 × 10−2 1.48 × 10−2 1.48 × 10−2 1.48 × 10−2 1.48 × 10−2 1.48 × 10−2 1.48 × 10−2 1.48 × 10−2 1.48 × 10−2 1.48 × 10−2

304.47 284.71 346.7 315.80 192.02 153.59 3.38 3.01 6.70 5.20 8.83 5.49

564.45

2.30

685.44

2.76

345.61

1.50

6.39

0.02

12.47

0.05

14.32

0.06

Car Car (ac) CO

Bus Car Car (ac)

a b c d

Average concentration for PM2.5 is in μg m−3 and for CO is in ppm. Duration per trip in minutes. Inhalation rate in m3 min−1. Round trip exposure for PM2.5 is in μg and for CO is in mg.

between two ventilation modes are very less (2–3 ppm), it was found to be statistically significant. Previous studies also suggest that ventilation setting can influence the in-cabin pollutant exposure levels (Karanasiou

et al., 2014; Zuurbier et al., 2010). In the case of a bus, the opening of windows allowed ambient air into the bus cabins. Other than the windows under study, remaining window positions were changing,

Fig. 6. Cumulative probability distributions for (a) PM2.5 concentrations in morning trips, (b) PM2.5 concentrations in evening trips, (c) CO concentrations in morning trips, and (d) CO concentrations in evening trips.


because passengers used to change their window positions as per their comfort level. However, the effect of changing window positions on exposure levels was not considered in this study. 4.3. Parameters influencing variability on personal concentrations Among the four parameters, travel mode explained the highest variability for each pollutant with the highest value for CO, followed by CO2 and the lowest for PM2.5. Our findings are similar to the results obtained in the earlier traffic studies conducted in the cities (Kaur and Nieuwenhuijsen, 2009; de Nazelle et al., 2012). CO is used as a tracer for vehicle exhaust emission. It is produced due to incomplete combustion of fuel in engines. Higher CO concentrations found in the areas having high traffic flow and congestion (Han and Naeher, 2006). CO2 is a product of vehicle combustion. In addition, CO2 is also generated from human exhalation. In our study, the average CO2 concentrations in the car (ac) and the bus mode were 77% and 15% higher than concentrations inside car due to restricted dilution of CO2 in the car (ac) mode and contribution of human breathing in the bus. Unlike CO and CO2, PM2.5 has several emission sources in addition to the exhaust emission, such as road dust re-suspension, industrial emissions, biomass burning and stone crushers that influenced its concentrations in our study. Therefore, the travel mode explained the lowest variability for PM2.5 concentrations. Concentration levels, in general, are found to be statistically different between morning and evening trips for all three travel modes. However, ANOVA analysis showed a feeble influence (~0.3%) of time of the journey on personal concentrations. Temperature and RH affect pollutants concentrations levels in traffic studies. RH was negatively correlated with the concentrations. In our study, temperature and relative humidity contributed to 8.8% and 16% CO variability. This relation is in contrast to the findings of exposure studies in cities (de Nazelle et al., 2012). This contrast is possibly due to the difference in wind flow pattern, which affects the in-cabin temperature and relative humidity in traffic micro-environments, between canyon type road network in cities and open road networks in highways. Wind speed that has been found to influence the pollutant concentration in traffic microenvironments significantly (Adams et al., 2001; Kaur et al., 2007; Knibbs et al., 2011) was not measured in our study. 4.4. Differences in mass exposures The health effects of mass exposure to pollutants not only depend on the pollutant concentration level to which a passenger is exposed but also depends on the duration of the exposure. Therefore, it is more realistic to evaluate the cumulative inhalation dose of the pollutants during the entire trip. In our case, we have estimated it as mass exposures. There were several personal exposure studies which focused on mass exposure to pollutants (McNabola et al., 2008; Zuurbier et al., 2010; Huang et al., 2012). These studies mainly focused on fixed time interval for mass exposures. In contrary, we have considered variable time intervals (per trip, per round trip, per hour and per minute) for in depth analysis of mass exposures to passengers. Our study revealed that PM2.5 exposures in the car and CO mass exposures in the car (ac) were higher in the morning trips than in the evening trips, primarily because morning trips have taken longer duration than evening trips. On the other hand, morning trips in the bus resulted in the higher mass exposure of PM2.5 and CO despite morning trips being shorter than evening trips, due to lower personal concentrations during the evening trips. A passenger usually prefers one travel mode on a highway during onward and return trip. Therefore, the round trip exposure is likely to give a better estimate to interpret the health hazard potential of a passenger exposed to pollutants while traveling, an observation also made by earlier studies (Huang et al., 2012). In our study, a round trip in the car (ac) took 460 min, and in the car, it took 505 min, about 10% higher than the car (ac). However, the round trip PM2.5 mass

163

exposure in the car (~685 μg) is nearly two times the round trip mass exposure in the car (ac) (~345 μg). It demonstrates that exposure estimate based on only pollutant concentration without accounting for the trip duration may not reflect the passengers' accumulated exposure in a traffic microenvironment. A similar comparison between the bus and the car shows that while the round trip exposure duration in the bus is 1.14 times that of the car, the round trip PM2.5 mass exposure is less in the bus, about 0.79 times the exposure level in the car. If the journey time is long and the study route is characterized by different road types (open highways vs. in city highway), an analysis of hourly mass exposures in these road networks can provide a better insight of the changing mass exposure pattern. In our study, the journey time for the round trip was around 8–10 h, and we calculated the hourly mass exposure rate for each trip. We found that the hourly PM2.5 and CO mass exposures when the highway passes through towns/cities are ~1.5 and ~2 times respectively, of the mass exposures on the highway away from towns/cities. This may be another reason to avoid traffic congestion and the national highways to bypass the cities. Per minute PM2.5 and CO exposures followed similar trend as per trip and round trip mass exposures. 5. Conclusion Personal exposure studies are crucial for the better assessment of the health risks posed by fine particulate matter and carbon monoxide in the traffic microenvironment and for taking appropriate measures to minimize the health risks. We observed that vehicle exhaust and road dust resuspension for PM2.5, and self-pollution of the car for CO are the major polluting sources on national highway. Selection of a travel mode for long distance journey on a highway can significantly influence the passenger exposure level to the pollutants. However, travel mode choice is of personal interest, which depends on many factors such as affordability, traveling distance, the time required to reach the destination, and so forth. A combination of public awareness and regulatory mechanism should guide the public for choosing an optimal travel mode. Road repairing and extension works significantly increased the local and average exposure levels. The government agencies should therefore focus on the highway repairing and extension works to be completed within a minimum time. Mass exposures to pollutants for passengers in the car was highest, despite car taking less time than bus to travel same distance. Our study indicated that for all travel modes, mass exposures in cities were higher when compared to open highway. It makes the argument stronger to bypass the towns and cities while planning the layout of the highways. In many countries, people use highways as the primary road networks to travel different places on regular basis. However, nearly all exposure studies in recent past were focused in the cities and therefore exposure assessments on the highways are limited. More studies on the highways need to be carried out for better understanding of the exposure pattern in the highways and to devise control strategies to safeguard the public health. Funding This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. Appendix A. Supplementary data Supplementary data to this article can be found online at https://doi. org/10.1016/j.scitotenv.2017.11.086. References Abi-Esber, L., El-Fadel, M., 2013. Indoor to outdoor air quality associations with self-pollution implications inside passenger car cabins. Atmos. Environ. 81, 450–463.

164


Adams, H.S., Nieuwenhuijsen, M.J., Colvile, R.N., 2001. Determinants of fine particle (PM2.5) personal exposure levels in transport microenvironments, London, UK. Atmos. Environ. 35, 4557–4566. Adar, S.D., Gold, D.R., Coull, B.A., Schwartz, J., Stone, P.H., Suh, H., 2007. Focused exposures to airborne traffic particles and heart rate variability in the elderly. Epidemiology 18, 95–103. Apte, J.S., Kirchstetter, T.W., Reich, A.H., Deshpande, S.J., Kaushik, G., Chel, A., Marshall, J.D., Nazaroff, W.W., 2011. Concentrations of fine, ultrafine, and black carbon particles in auto-rickshaws in New Delhi, India. Atmos. Environ. 45, 4470–4480. Briggs, D.J., de Hoogh, K., Morris, C., Gulliver, J., 2008. Effects of travel mode on exposures to particulate air pollution. Environ. Int. 34, 12–22. Cepeda, M., Schoufour, J., Freak-Poli, R., Koolhaas, C.M., Dhana, K., Bramer, W.M., Franco, O.H., 2017. Levels of ambient air pollution according to mode of transport: a systematic review. Lancet Public Health 2, 23–34. Chan, A.T., Chung, M.W, 2003. Indoor–outdoor air quality relationships in vehicle: effect of driving environment and ventilation modes. Atmos. Environ. 37, 3795–3808. Chan, L.Y., Lau, W.L., Zou, S.C., Cao, Z.X., Lai, S.C., 2002. Exposure level of carbon monoxide and respirable suspended particulate in public transportation modes while commuting in urban area of Guangzhou, China. Atmos. Environ. 36, 5831–5840. CPCB, 2009. Central Pollution Control Board. National Ambient Air Quality Standards. http://cpcb.nic.in/National_Ambient_Air_Quality_Standards.php, Accessed date: 15 August 2017. CPCB, 2010. Air Quality Monitoring, Emission Inventory and Source Apportionment Study for Indian Cities. Central Pollution Control Board. Dominici, F., Peng, R.D., Bell, M.L., Pham, L., McDermott, A., Zeger, S.L., Samet, J.M., 2006. Fine particulate air pollution and hospital admission for cardiovascular and respiratory diseases. JAMA 295, 1127–1134. Flachsbart, P.G., 1999. Human exposure to carbon monoxide from mobile sources. Chemosphere Global Change Sci. 1, 301–329. Geiss, O., Barrero-Moreno, J., Tirendi, S., Kotzias, D., 2010. Exposure to particulate matter in vehicle cabins of private cars. Aerosol Air Qual. Res. 10, 581–588. Goel, R., Gani, S., Guttikunda, S.K., Wilson, D., Tiwari, G., 2015. On-road PM2.5 pollution exposure in multiple transport microenvironments in Delhi. Atmos. Environ. 123, 129–138. Gulliver, J., Briggs, D.J., 2004. Personal exposure to particulate air pollution in transport microenvironments. Atmos. Environ. 38, 1–8. Han, X., Naeher, L.P., 2006. A review of traffic-related air pollution exposure assessment studies in the developing world. Environ. Int. 32, 106–120. Hsu, D.J., Huang, H.L., 2009. Concentrations of volatile organic compounds, carbon monoxide, carbon dioxide and particulate matter in buses on highways in Taiwan. Atmos. Environ. 43, 5723–5730. Huang, H.L., Hsu, D.J., 2009. Exposure levels of particulate matter in long-distance buses in Taiwan. Indoor Air 19, 234–242. Huang, J., Deng, F., Wu, S., Guo, X., 2012. Comparisons of personal exposure to PM2.5 and CO by different commuting modes in Beijing, China. Sci. Total Environ. 425, 52–59. Hudda, N., Fruin, S.A., 2013. Models for predicting the ratio of particulate pollutant concentrations inside vehicles to roadways. Environ. Sci. Technol. 47, 11048–11055. Kampa, M., Castanas, E., 2008. Human health effects of air pollution. Environ. Pollut. 151, 362–367. Karanasiou, A., Viana, M., Querol, X., Moreno, T., de Leeuw, F., 2014. Assessment of personal exposure to particulate air pollution during commuting in European cities—recommendations and policy implications. Sci. Total Environ. 490, 785–797. Kaur, S., Nieuwenhuijsen, M.J., 2009. Determinants of personal exposure to PM2.5, ultrafine particle counts, and CO in a transport microenvironment. Environ. Sci. Technol. 43, 4737–4743. Kaur, S., Nieuwenhuijsen, M.J., Colvile, R.N., 2007. Fine particulate matter and carbon monoxide exposure concentrations in urban street transport microenvironments. Atmos. Environ 41, 4781–4810. Kaur, S., Nieuwenhuijsen, M., Colvile, R., 2005. Personal exposure of street canyon intersection users to PM2.5, ultrafine particle counts and carbon monoxide in Central London, UK. Atmos. Environ. 39, 3629–3641. Kingham, S., Meaton, J., Sheard, A., Lawrenson, O., 1998. Assessment of exposure to trafficrelated fumes during the journey to work. Transp. Res. D 3, 271–274. Knibbs, L.D., Cole-Hunter, T., Morawska, L., 2011. A review of commuter exposure to ultrafine particles and its health effects. Atmos. Environ. 45, 2611–2622. Kumar, P., Gupta, N.C., 2016. Commuter exposure to inhalable, thoracic and alveolic particles in various transportation modes in Delhi. Sci. Total Environ. 541, 535–541. Laden, F., Neas, L.M., Dockery, D.W., Schwartz, J., 2000. Association of fine particulate matter from different sources with daily mortality in six US cities. Environ. Health Perspect. 108, 941–947. Marshall, J.D., Behrentz, E., 2005. Vehicle self-pollution intake fraction: children's exposure to school bus emissions. Environ. Sci. Technol. 39, 2559–2563. McNabola, A., Broderick, B.M., Gill, L.W., 2008. Relative exposure to fine particulate matter and VOCs between transport microenvironments in Dublin: personal exposure and uptake. Atmos. Environ. 42, 6496–6512.

Miller, K.A., Siscovick, D.S., Sheppard, L., Shepherd, K., Sullivan, J.H., Anderson, G.L., Kaufman, J.D., 2007. Long-term exposure to air pollution and incidence of cardiovascular events in women. N. Engl. J. Med. 356, 447–458. Min, J.Y., Paek, D., Cho, S.I., Min, K.B., 2009. Exposure to environmental carbon monoxide may have a greater negative effect on cardiac autonomic function in people with metabolic syndrome. Sci. Total Environ. 407, 4807–4811. MoRTH, 2016. Basic Road Statistics of India 2013–14 and 2014–15. http://morth.nic.in/ showfile.asp?lid=2445, Accessed date: 17 August 2017. Namdeo, A., Ballare, S., Job, H., Namdeo, D., 2014. Commuter exposure to air pollution in Newcastle, UK, and Mumbai, India. J. Hazard. Toxic Radioact. Waste 20, 4014004. de Nazelle, A., Fruin, S., Westerdahl, D., Martinez, D., Ripoll, A., Kubesch, N., Nieuwenhuijsen, M., 2012. A travel mode comparison of commuters' exposures to air pollutants in Barcelona. Atmos. Environ. 59, 151–159. NHAI, 2017. Indian Road Network. http://www.nhai.org/roadnetwork.htm, Accessed date: 17 August 2017. O'Donoghue, R.T., Gill, L.W., McKevitt, R.J., Broderick, B., 2007. Exposure to hydrocarbon concentrations while commuting or exercising in Dublin. Environ. Int. 33, 1–8. Onat, B., Stakeeva, B., 2013. Personal exposure of commuters in public transport to PM2.5 and fine particle counts. Atmos. Pollut. Res. 4, 329–335. Ostro, B., Broadwin, R., Green, S., Feng, W.Y., Lipsett, M., 2006. Fine particulate air pollution and mortality in nine California counties: results from CALFINE. Environ. Health Perspect. 114, 29–33. Pant, P., Habib, G., Marshall, J.D., Peltier, R.E., 2017. PM2.5 exposure in highly polluted cities: a case study from New Delhi, India. Environ. Res. 156, 167–174. Pope III, C.A., Burnett, R.T., Thun, M.J., Calle, E.E., Krewski, D., Ito, K., Thurston, G.D., 2002. Lung cancer, cardiopulmonary mortality, and long-term exposure to fine particulate air pollution. JAMA 287, 1132–1141. Pope, C.A., Burnett, R.T., Krewski, D., Jerrett, M., Shi, Y., Calle, E.E., Thun, M.J., 2009. Cardiovascular mortality and exposure to airborne fine particulate matter and cigarette smoke. Circulation 120, 941–948. Potchter, O., Oz, M., Brenner, S., Yaakov, Y., Schnell, I., 2014. Exposure of motorcycle, car and bus commuters to carbon monoxide on a main road in the Tel Aviv metropolitan area, Israel. Environ. Monit. Assess. 186, 8413–8424. Querol, X., Moreno, T., Karanasiou, A., Reche, C., Alastuey, A., Viana, M., Font, O., Gil, J., Miguel, E.D., Capdevila, M., 2012. Variability of levels and composition of PM10 and PM2.5 in the Barcelona metro system. Atmos. Chem. Phys. 12, 5055–5076. Ramos, C.A., Wolterbeek, H.T., Almeida, S.M., 2016. Air pollutant exposure and inhaled dose during urban commuting: a comparison between cycling and motorized modes. Air Qual. Atmos. Health 9, 867–879. Raub, J.A., Mathieu-Nolf, M., Hampson, N.B., Thom, S.R., 2000. Carbon monoxide poisoning—a public health perspective. Toxicology 145, 1–14. Richmond-Bryant, J., Saganich, C., Bukiewicz, L., Kalin, R., 2009. Associations of PM2.5 and black carbon concentrations with traffic, idling, background pollution, and meteorology during school dismissals. Sci. Total Environ. 407, 3357–3364. Riojas-Rodríguez, H., Escamilla-Cejudo, J.A., González-Hermosillo, J.A., Téllez-Rojo, M.M., Vallejo, M., Santos-Burgoa, C., Rojas-Bracho, L., 2006. Personal PM2.5 and CO exposures and heart rate variability in subjects with known ischemic heart disease in Mexico City. J. Expo. Sci. Environ. Epidemiol. 16, 131–137. Sabapathy, A., Ragavan, K.S., Saksena, S., 2012. An assessment of two-wheeler CO and PM10 exposures along arterial main roads in Bangalore city, India. Open Atmos. Sci. J. 6, 71–77. Sabin, L.D., Kozawa, K., Behrentz, E., Winer, A.M., Fitz, D.R., Pankratz, D.V., Colome, S.D., Fruin, S.A., 2005. Analysis of real-time variables affecting children's exposure to diesel-related pollutants during school bus commutes in Los Angeles. Atmos. Environ. 39 (29), 5243–5254. Sharma, A., Chandrasekhar, S., 2014. Growth of the urban shadow, spatial distribution of economic activities, and commuting by workers in rural and urban India. World Dev. 61, 154–166. USEPA, 2011. Exposure Factors Handbook: 2011 Edition (Final Report). Washington, DC (doi:EPA/600/R-09/052F). WHO, UNAIDS, 2006. Air Quality Guidelines: Global Update 2005. World Health Organization. Wu, D.L., Lin, M., Chan, C.Y., Li, W.Z., Tao, J., Li, Y.P., Sang, X.F., Bu, C.W., 2013. Influences of commuting mode, air conditioning mode and meteorological parameters on fine particle (PM2.5) exposure levels in traffic microenvironments. Aerosol Air Qual. Res. 13, 709–720. Zanobetti, A., Schwartz, J., 2009. The effect of fine and coarse particulate air pollution on mortality: a national analysis. Environ. Health Perspect. 117, 898–903. Zhang, Q., Zhu, Y., 2010. Measurements of ultrafine particles and other vehicular pollutants inside school buses in South Texas. Atmos. Environ. 44, 253–261. Zuurbier, M., Hoek, G., Oldenwening, M., Lenters, V., Meliefste, K., van den Hazel, P., Brunekreef, B., 2010. Commuters' exposure to particulate matter air pollution is affected by mode of transport, fuel type, and route. Environ. Health Perspect. 118, 783–787.