Next Article in Journal
Heavy Metals Size Distribution in PM10 and Environmental-Sanitary Risk Analysis in Acerra (Italy)
Next Article in Special Issue
A Review of Airborne Particulate Matter Effects on Young Children’s Respiratory Symptoms and Diseases
Previous Article in Journal
Ship-Based Measurements of Atmospheric Mercury Concentrations over the Baltic Sea
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Personal Exposure to PM2.5 in the Megacity of Mexico: A Multi-Mode Transport Study

1
Centro de Ciencias de la Atmósfera, Universidad Nacional Autónoma de Mexico, Circuito de la Investigación Científica S/N, Ciudad Universitaria, Coyoacán, 04510 Ciudad de Mexico, Mexico
2
CONACYT—Consorcio CENTROMET, Camino a Los Olvera 44, Los Olvera, Corregidora 76904, Queretaro, Mexico
3
Arian International Projects, c/Baldiri Reixach, 4, 08028 Barcelona, Spain
4
Department of Respiratory Physiology, National Institute of Respiratory Diseases Ismael Cosío Villegas, Calzada de Tlalpan 4502, Col. Sección XVI, Tlalpan, 14080 Mexico City, Mexico
*
Author to whom correspondence should be addressed.
Atmosphere 2018, 9(2), 57; https://doi.org/10.3390/atmos9020057
Submission received: 17 November 2017 / Revised: 4 February 2018 / Accepted: 5 February 2018 / Published: 9 February 2018
(This article belongs to the Special Issue Impacts of Air Pollution on Human Health)

Abstract

:
Recurrent personal exposure to ambient PM2.5 is associated with adverse human health effects, in particular on the respiratory and cardiovascular systems. Here, we present an assessment of personal exposure and inhalation of PM2.5 for five modes of transport (walking, cycling, public bus (trolleybus and diesel bus), conventional car (CC) and hybrid-electric car (HEC)) and two routes of similar distance, along a major road in the Mexico City metropolitan area (MCMA). Arithmetic average exposure concentrations ranged from 16.5 ± 6.5 µg m−3 for walking to 81.7 ± 9.1 µg m−3 for cycling (henceforth shown as average ±1 SD), with no significant differences with geometric averages. The maximum exposure concentration of 110.9 µg m−3 was observed for the conventional car. The highest exposure concentrations depended on route and the mode of transport, being observed for cycling and walking. The PM2.5 measurements showed large spatial heterogeneity in the exposure levels for walking and cycling, while public buses and private transport showed less spatial heterogeneity. The greatest peaks in PM2.5 coincided with 4-way intersections for all modes of transport, being positively influenced by traffic density. The mass of PM2.5 inhaled depended mostly on the mode of transport, and ranged between 1.0 ± 0.3 and 30.1 ± 14.2 µg km−1 for the HEC and bicycle, respectively. Local area PM2.5 increments identified as ‘residuals’ after subtraction of data recorded at the closest fixed monitoring site from exposure concentrations along the studied road suggested that inhalation for bicycle and diesel buses is strongly influenced by vehicular emissions. Residuals estimated for the trolleybus, CC and HEC confirmed a lower inhalation than for the other modes of transport evaluated due to protection by the cabin.

1. Introduction

Immediate proximity to direct emissions of PM2.5 from motor vehicles during daily commutes represents a significant threat to human health. Although personal exposure in transport micro-environments represents only between 1–5% of the time spent in a day [1,2], in large urban areas this time may increase significantly. For example, within the Mexico City Metropolitan Area (MCMA), a typical single journey may take between 41 and 81 mins, while average journeys of 51 mins represent spending 16 h per week in transport micro-environments [3]. Therefore, longer commuting times may result in an increase in exposure to PM2.5 which can be significantly relevant for health effects [4]. For instance, Dons et al. [5] reported a difference of 2-fold in personal exposure to black carbon for commuters in Belgium compared with non-commuters. Within the Greater London Area, Smith et al. [2] reported that daily commutes contribute around 9% of total daily exposure to PM2.5. However, this can be further exacerbated during traffic rush hours, when exposure concentrations of PM2.5 may increase by between 5–20% [6,7].
Besides proximity to emission sources and commuting time, personal exposure to PM2.5 may depend on the mode of transport, being classified into active (cycling and walking) and passive (car, bus, underground) [1,8]. For instance, Okokon et al. [9] addressed exposure to PM2.5 when commuting in cars with open and closed windows, and by bicycle and bus in three European cities, observing mostly higher exposure for passive transport modes. Similarly, higher exposure to PM2.5 in Central London for passive as compared to active commute was reported by Adams et al. [1]. In the MCMA, studies have addressed personal exposure to carbon monoxide with the highest and lowest levels observed for private and public transport, respectively [10], and to volatile organic compounds with exposure levels to vehicular related compounds being a factor of 2 greater than observed indoors [11]. Vallejo et al. [12] addressed personal exposure to PM2.5 during transportation within the MCMA by underground, public bus and private car, reporting median exposure levels of 106.2, 101.7 and 62.4 µg m−3, respectively; however, their study design did not allow for comparison between micro-environment concentrations. This highlights the importance of assessing personal exposure to PM2.5 during daily commutes using comparable information of modes of transport and routes.
In this work, we describe personal exposure to PM2.5 of commuters using five modes of transport along a typical main road of the MCMA. We estimate commuter inhalation by taking into account physical activity to provide a better approximation of the effect of the transport micro-environments upon mass of PM2.5 inhaled. We compare commuters’ inhalation calculated from in-situ measurements of PM2.5 along the studied road with those from a fixed site to account for the effect of road proximity. Finally, we describe the traffic effect on the exposure to peaks in PM2.5 by showing spatial and temporal variability in recorded data.

2. Methodology

2.1. Study Site Location and Description

Personal exposure to local background and proximity to sources of PM2.5 as a function of the transport mode were evaluated along the Miguel Angel de Quevedo (MAQ) Road in South MCMA (Figure 1a). The MAQ Rd was selected due to the circulation of different modes of transport, proximity to a ground-based monitoring site, and confined lanes and a return route, which offered two routes for the same mode of transport. It is located in the Coyoacán municipality and is one of 10 main roads that cross from west to east in the southern region of the MCMA (Figure 1b). It runs through a densely populated residential area, with three lanes in each direction separated by a strip of trees and grass. The right extreme lanes are mostly designated for cyclists and public transport, consisting of diesel buses and trolleybuses. Measurements of personal exposure to ambient PM2.5 were made along 2 routes as shown in Figure 1: (i) From the MAQ roundabout to the Taxqueña (TAX) trolleybus stop (eastwards: MAQ-TAX) and (ii) from TAX to MAQ (westwards: TAX-MAQ). Table 1 lists the points of departure and arrival of each route and corresponding length. Figure 1a also shows the location of the ground-based Coyoacán (COY) monitoring site, which was used as reference of personal exposure in the local area. The COY site forms part of the Integral Atmospheric Monitoring System (SIMAT) of the Mexico City Government. PM2.5, wind speed (WS) and wind direction (WD) have been monitored continuously since 2003 at the COY site and the data sets were obtained from the SIMAT web site (Available online: http://www.aire.cdmx.gob.mx).

2.2. Study Design

The measurement campaign was carried out during 10–14 November 2014, between 11 a.m. and 2 p.m. CDT, when traffic loading allows capture of features relevant for personal exposure and inhalation, i.e., vehicular accumulation in 4-way intersections, traffic lights and different traffic composition [13]. One journey along two routes (MAQ-TAX and TAX-MAQ) was performed for walking and cycling, and one for trolleybus (MAQ-TAX), diesel bus, conventional car (CC) and hybrid-electric car (HEC) (TAX-MAQ). Although all journeys were made along the MAQ Rd, MAQ-TAX and TAX-MAQ, the journeys were analysed separately as differences in variance coefficients for journeys duration, which were used here as proxy for traffic conditions, and were ≤10% for all modes of transport apart from public bus (Supplementary Information S1.1, Table S1). This allows discarding the effect of traffic conditions on the personal exposure levels observed. Analysing the routes separately permits addressing the effect of traffic composition, time of day and meteorology. Walking exposure concentrations were measured by pedestrians moving at an average speed of 3 km h−1. Cyclists made journeys on 18 gear, full suspension mountain bicycles at variable speed between 8 and 10 km h−1 depending on traffic conditions. For the public bus’s eastward journey, volunteers travelled on a trolleybus (K route), whereas for the westward public bus journey, the volunteers travelled on a diesel-powered bus (1–29 route). Journeys by CC and HEC on the eastward route were made with closed windows and operating air conditioning, whereas westward journeys were made with open windows. Journeys between MAQ to TAX on the HEC and CC with closed windows and operating air-conditioning were discarded due to increases in relative humidity inside the vehicles (>85%), which affected instrumentation. Table 1 describes the experimental setup.

2.3. Instrumentation

The exposure concentrations to PM2.5 were measured using portable aerosol real-time photometric monitors (pDR-1500, Thermo Scientific, Franklin, MA, USA), operated at a frequency of 1 Hz, with stated precision for 30-days of ±2% (±0.005 mg m−3) at (2σ)2. During all journeys, 2 monitors were carried on the chest of volunteers and operated simultaneously. The monitors were operated at a flow rate of 2 L min−1 and calibrated with a commercial Agilent digital flow meter at a variance coefficient <5%. Calibrations to zero were carried out inside a closed chamber filled by a zero-air Teledyne generator 701, operating at a pressure of 5 psi. Overall, the monitors were placed inside the closed chamber for 15 min to obtain a stable reading of zero. For all modes of transport, there was a very good agreement with R2 > 0.9 between the two instruments, which suggests no significant differences in linearity between the two instruments over the exposure levels measured (Supplementary Information S1.2, Figure S1). Location of each volunteer along the MAQ Rd was recorded using a Hemisphere GPS receiver (60CSx, Garmin, Olathe, KS, USA).

2.4. Data Analyses

The personal exposure to PM2.5 for each mode of transport was determined by averaging the data of both pDR-1500. Exposure concentrations were allocated spatially by merging PM2.5 with GPS data. The data sets were analysed extensively using the R software [14], and tested for normality using the Shapiro-Wilk test for large samples (n > 1000), with descriptive statistics reported accordingly. Spatial variations in concentrations of PM2.5 were depicted using the Google Earth software 2017 [15]. The determined PM2.5 1-s averages concentration maps exhibit a picture of main PM2.5 high exposure areas along the MAQ Rd, where each point represents 1-s averaged data.

2.5. Residual Exposure Determination

We defined a ‘local area residual’ as the excess PM2.5 in measurements along MAQ Rd compared to the contemporary COY levels, i.e., COY 1-h averages are subtracted from PM2.5 exposure concentrations along MAQ Rd. The COY site is located upwind of MAQ Rd as shown in Figure 1.

2.6. PM2.5 Inhalation Calculation

We calculated the inhaled mass of PM2.5 for each mode of transport, considering the whole exposure windows experienced during the entire journeys. We defined an exposure window as the exposure concentrations experienced (recorded) over the duration of one period of inhalation-exhalation, here called “breath”. To allow a better interpretation, units for each calculated variable have been included. Thus, we first calculated duration by breath (b) as:
b   [ m i n   b r e a t h 1 ] =   1 R R   [ 1 b r e a t h   m i n 1 ]
where RR represents the respiratory rate (i.e., number of breaths per min). Therefore, the period for one breath (p) was calculated as
p   [ min ] = 1   [ b r e a t h ] × b   [ m i n   b r e a t h 1 ]
We assumed that the duration of each exposure window is equal to the period, and depends on the physical activity. Table 2 lists the duration of each exposure window, establishing lower and upper bounds by rounding p. Then, the number of exposure windows experienced during the whole journey (W) is given by
W = T p   [ m i n m i n ]
where T represents the journey duration for each mode of transport. Then, the volume of air inhaled (V) in each exposure window was obtained by dividing the volume of air inhaled per minute (VE) according to physical activities by the RR:
V   [ m 3 b r e a t h 1 ] =   V E R R   [ m 3   m i n 1 b r e a t h   m i n 1 ]
The exposure concentration in the i exposure window ( i [ 1 , W ] ) was obtained by averaging the concentrations recorded (n) during one breath:
C i ¯   [ b r e a t h   μ g   m 3 ] =   C i , 1 + +   C i , n n [ μ g   m 3 b r e a t h 1 ]
Finally, the total mass inhaled of PM2.5 (I) was calculated as the integrated inhalation for all exposure windows experienced during the whole journey [16,17] (Supplementary Information S1.3, Table S2), as:
I   [ μ g ] =   i = 1 W ( C i ¯ V )   [ μ g   m 3   b r e a t h 1   m 3 b r e a t h 1 ]
The RR and VE for a cross-sectional study of healthy adults and physical activity within the MCMA were obtained by the National Institute of Respiratory Diseases of Mexico. Table 2 lists average RR and VE were recalculated from data within the standard deviation (1σ) of the average calculated from the whole sample. Overall, Mexican men and women between 10 and 80 years of age, who were residents in the MCMA for more than 2 years were selected on the basis of pulmonary health and sedentary and low-moderate physical daily activity. High performance athletes, and patients with cardiopulmonary disease, obesity, physical limitations, and who smoked, were excluded from the cross-sectional study. Anthropometric evaluations, forced spirometry, maximum voluntary ventilation and an electrocardiogram in resting conditions were carried out for all volunteers. The analysis of the exhaled air was made using oronasal mask. A protocol of incremental exercise of 10 W min−1 symptom-limited was carried out using a cycle ergometer, monitoring cardiovascular, respiratory and metabolic responses. The cardiopulmonary tests were made with Jaeger Oxycon-Pro instruments.

3. Results

3.1. Time-Series of Personal Exposure to PM2.5

PM2.5 data were recorded for five transport modes from TAX to MAQ and for three modes from MAQ to TAX. No significant differences (p > 0.05) were observed in the travel times between routes for all modes of transport, which allows route traffic effects to be discarded, and also allows eastward and westward journeys to be compared separately. Travel times ranged between 0.30 h for the trolleybus to 1.07 h for walking. All PM2.5 time-series exhibited a log-normal distribution. Table 3 lists descriptive statistics for journeys from MAQ to TAX and from TAX to MAQ. Arithmetic averages (AAs) for PM2.5 recorded from TAX to MAQ ranged from 16.5 ± 6.5 µg m−3 for walking to 81.7 ± 9.1 µg m−3 for bicycle, while respective medians ranged between 14.7 and 81.8 µg m−3. From MAQ to TAX, AAs of PM2.5 ranged from 21.0 ± 7.1 µg m−3 for walking to 75.3 ± 9.3 µg m−3 for bicycle, whereas respective medians ranged from 20.7 to 78.1 µg m−3. No significant differences (p > 0.05) between geometric averages (GAs) and AAs were determined for all transport modes.
Figure 2 shows the box and whisker plots for PM2.5 exposure concentrations recorded from TAX to MAQ and from MAQ to TAX by mode of transport, and the corresponding 1-h averages in PM2.5 recorded simultaneously at the COY site. To account for the effect of PM2.5 emissions dispersion, WS and WD are also shown in Figure 2. Overall, no significant differences (p > 0.05) were observed among WS during the sampling campaign, while WD occurrence during the walking measurements was significantly different than the WD occurrence for the other modes of transport. The calculated PM2.5 residuals depend strongly on the mode of transport with the highest residuals of 31–39 µg m−3 determined for bicycle. A lower residual of 18 µg m−3 was determined for the diesel bus, while walking and the trolleybus exhibited negative residuals of 5–6 µg m−3. This suggests that the exposure levels for walking and trolleybus are similar to those observed within the local area, while exposure while cycling and taking the diesel bus is significantly higher than that in the local area. By contrast, the CC and HEC showed negative residuals of 14 and 36 µg m−3, respectively, which suggests a significant decrease in the personal exposure to PM2.5 compared to the other modes of transport studied.

3.2. Spatial Variations in Street-Level PM2.5 Concentrations

Figure 3 and Figure 4 show the spatial variations in PM2.5 observed for the eastwards and westwards routes along MAQ Rd, respectively, with peaks representing hotspots in personal exposure to PM2.5. Horizontal lines in each figure represent 1-h averages of PM2.5 recorded at COY and are shown as reference for the street-level measurements. Overall, walking and trolleybus exhibit exposure levels to PM2.5 below those recorded at COY for most of the journey, while the bicycle shows levels higher than at COY for the whole journey. The peaks in PM2.5 observed from MAQ to TAX for all transport modes show large spatial heterogeneity, with the largest peaks observed for walking and the smallest for trolleybus (Figure 3a). Bicycle and walking show higher PM2.5 concentration peaks from MAQ to TAX than the trolleybus, which coincided with 4-way intersections and was not observed for the trolleybus (Figure 3b). Such peaks in concentrations of PM2.5 could arise from combined emissions of PM2.5 from vehicles accumulating and waiting in front of traffic lights [18]. The greatest peak in PM2.5 observed coincided with a large number of vehicles queuing to cross Division del Norte Rd.
Figure 4a shows peaks in PM2.5 observed during journeys made from TAX to MAQ, which exhibited less spatial heterogeneity than those observed for MAQ to TAX journeys. Only close to the MAQ roundabout did all modes of transport, apart from the HEC, exhibit noticeable increases in PM2.5. However, as for the MAQ to TAX journeys, most of the peaks in PM2.5 were observed close to 4-way intersections (Figure 4b). The large increase in PM2.5 observed for the CC was originated by emissions from a large number of cars queuing to pick up children from a primary school close to the MAQ Rd and Virginia street corner. Clearly, the lowest variations of PM2.5 exposure concentrations are observed for the diesel bus, CC and HEC, while the highest ones correspond to bicycle and walking in good agreement with data recorded from MAQ to TAX. This suggests that commuting in vehicles may protect commuters to significant increases in the exposure to PM2.5.

3.3. PM2.5 Inhalation

Estimated inhalation of PM2.5 ranged from 6.3 ± 1.0 to 147.4 ± 69.4 µg for the trolleybus and bicycle, respectively, from MAQ to TAX, whereas from TAX to MAQ, it ranged from 4.7 ± 1.2 to 104.5 ± 49.1 µg m−3 for the HEC and bicycle, respectively (Table 4). Estimated inhalation of PM2.5 from ground-based measurements from MAQ to TAX ranged between 7.1 ± 1.1 µg for trolleybus and 91.3 ± 43.0 µg for bicycle. From TAX to MAQ, ground-based calculated inhalation ranged between 8.2 ± 1.3 µg for the diesel bus to 55.0 ± 25.9 µg for the bicycle. Higher inhalation of PM2.5 for bicycle and diesel bus was estimated from on-field measurements than from ground-based measurements. Table 4 shows µg km−1 of PM2.5 inhaled as function of the mode of transport, and relative to PM2.5 concentrations recorded at the COY site. For both routes, the highest inhalation was determined for bicycle (30.1 ± 14.2 to 39.0 ± 18.4 µg km−1), while the lowest one corresponded to the public buses (1.5 ± 0.2 to 3.0 ± 0.5 µg km−1). For all journeys, the lowest inhalation of 1.0 ± 0.3 µg km−1 corresponded to the HEC from TAX to MAQ.
The levels of inhalation determined from ground-based measurements were consistent with those estimated from on-field PM2.5 data. The highest inhalation of 15.9 ± 7.5 and 24.2 ± 11.4 µg km−1 was estimated for bicycle, while the lowest ones of 1.7 ± 0.3 and 1.9 ± 0.3 µg km−1 was determined for the public buses, with ratios highest-to-lowest inhalation ranging from eight times from MAQ to TAX to 14 times from TAX to MAQ. We observed the highest residual inhalation expressed as a percentage for bicycle in both routes (37.9 to 47.2%), followed by the diesel bus (36.7%). By contrast, the lowest residual inhalations corresponded to the CC (−50.0%) and HEC (−190.1%) from TAX to MAQ. Only walking showed similar inhalation residuals in both routes (−28.6 and −32.1%). Within the study area, active transport modes, but mostly the bicycle, are associated with the highest inhalation of PM2.5, while commuting by private transport may protect commuters from an increase in the exposure concentrations compared to those considered representative of local area levels.

4. Discussion

Personal exposure to PM2.5 has been typically assessed using air quality modelling and ground-based monitoring [2,18,19]. However, model performance can be affected by the quality of input data, i.e., significant variations in air pollutant levels recorded by ground-based monitoring, seasonality and proximity to large emissions sources. In particular, personal exposure to PM2.5 from vehicle emissions is highly distance-dependent and may vary significantly between few and hundreds of meters from roads [20,21]. The comparison between ground-based and on-field PM2.5 measurements can be used to improve personal exposure prediction and health effects. This study addressed personal exposure to PM2.5 in transport micro-environments using real-time measurements within the MCMA. Residual exposure and inhalation were defined to compare personal exposure to PM2.5 when commuting by two active modes (walking, bicycle) and three passive modes (public buses, CC and HEC).

4.1. Personal Exposure to PM2.5 in the MCMA

Assessments of personal exposure to PM2.5 in transport micro-environments have been conducted all over the world [22]. The median exposure levels reported here for public buses and CC are lower than those reported in the MCMA by Vallejo et al. [12] of 101.7 and 62.4 µg m−3, respectively. However, the average exposure levels reported here are within the range of those reported by Okokon et al. [9] (14 ± 5 to 85 ± 39 µg m−3) in three European cities for bicycle, public bus and car with open and closed windows. However, comparisons between on-field measurements and ground-based monitoring were not made. Average exposure levels of between 2.9 ± 2.3 µg m−3 and 11.0 ± 6.6 µg m−3, lower than those reported here, were observed by Quiros et al. [17] for cycling, walking and driving with open and closed windows in a residential area of Santa Monica, California. We estimate that average exposure to PM2.5 for cyclists in our study is between 7.8 to 15.4 times higher than the average for cyclists in Santa Monica. The range of average exposure for commuters reported in this study is higher than that observed in Sacramento, California, by Ham et al. [23], who reported exposure concentrations between 5 and 15 µg m−3 for journeys made in public bus, private car and bicycle. Such differences are likely due to vehicles fitted with cleaner technologies in the USA. For instance, in California, diesel technology is regulated under the EPA 2010 standards [24], while in Mexico analogous standards have not been introduced yet. However, consistent with our results, cyclists exhibited the highest exposure among the transport modes studied. By contrast, exposure concentrations for CC and HEC when travelling with open windows in our study are around 50% lower than those observed in St. Louis, MO, USA, by Leavey et al. [25].
Our results also contrast with those reported by Adams et al. [1] from a multi-mode transport study carried out in Central London, who reported the highest geometric average exposure for public transport (bus, 38.9 µg m−3), followed by CC (33.7 µg m−3) and the lowest ones for bicycle (23.5 µg m−3). We observed that for the same transport modes in the MCMA, the largest exposure occurs when cycling (81.1 µg m−3), followed by the diesel bus (49.5 µg m−3) and CC (27.7 µg m−3). The differences in the exposure levels to PM2.5 suggest that each mode of transport can be influenced by different factors. For instance, the highest exposure observed for cyclists along the MAQ Rd arises from proximity to the traffic flow, which consequently increases immediate exposure to vehicular emissions [23]. The lower exposure observed for walking is due to the distance between footpaths and road edges, which in the study area varies between 1 and 4 m [19]. The estimated exposure residuals confirmed that proximity to roads and, in particular, to vehicular emissions influence the personal exposure in transport micro-environments. In this study, no significant effect of difference in WD (wind direction) occurrence was observed on the exposure residuals. For instance, walking, CC and HEC exhibited negative values under different WD occurrence, while bicycle, public buses and CC and HEC exhibited positive and negative values under similar wind occurrence. However, further measurements are needed to support our results under more variable WD occurrence.
We observed that commuting in private transport may protect commuters from direct exposure despite the proximity to PM2.5 emissions sources. This contrasts with the findings of Okokon et al. [9], who observed a higher exposure to PM2.5 for the CC with open windows than for public bus and bicycle and with Ham et al. [23] who reported similar exposure for private cars and public buses. This could be due to differences in the traffic flow but mostly to vehicular fleet technology; while in Europe around 50% of total light duty vehicles are diesel-powered [26]; in the MCMA around 80% of the light duty vehicle fleet is petrol-powered [27]. The higher exposure to PM2.5 for the diesel bus reported here than for the trolleybus could be related to recirculation of exhaust gases inside the bus cabin due to the lack of diesel particulate filters fitted. Asmi et al. [28] reported higher concentrations of PM2.5 inside older buses than inside new buses related to differences in bus-to-bus emissions, which could explain the difference reported here. The lower exposure observed in private cars relative to public transport could be due to tailpipe diesel exhaust recirculating to the cabin, which was observed in the California school bus program [29]. Overall, the residuals estimated for the trolleybus, CC and HEC confirmed a lower exposure than that estimated for the other modes of transport evaluated here.

4.2. The Spatial Distribution of Peaks in PM2.5

Along MAQ Rd, the highest concentrations of PM2.5 recorded in each mode of transport were observed close to 4-way intersections. The most plausible explanation for such plumes of PM2.5 is large emissions from idling motor vehicles during the idling/acceleration cycle. Jazcilevich et al. [30] reported a net increase in PM10 emissions of 3.1% from diesel vehicles in the MCMA during the stop/acceleration cycle compared to vehicles travelling at constant speed. This was clearly noticed when walking from MAQ to TAX and, for the CC journey from TAX to MAQ when lines of >10 cars were observed close to the corner of Virginia St. and MAQ Rd. Such plume corresponded to the highest concentrations of PM2.5 recorded for HEC and CC. However, the link between plumes and traffic density could be improved significantly by performing vehicle counts or conducting visual recording. Leavey et al. [25] observed that concentrations of PM2.5 in-cabin followed closely with outdoor levels when travelling with open windows, which may explain the peak observed for the CC. Our results contrast with observations made in California by Quiros et al. [17], who reported an increase in the levels of ultrafine particles but not in PM2.5 when walking close to a diesel bus during the accelerating phase after a complete stop. Such difference could be related mostly to the number of vehicles passing by the area when the PM2.5 plumes were intercepted and cleaner diesel technology in California, while in the MCMA most of the diesel vehicles are not fitted with particulate filters.
We did not observe a noticeable effect of WD and WS on the spatial variability of the PM2.5 peaks between routes for each mode of transport. Venkatram et al. [21] addressed the effect of WD on air pollutants dispersion in the USA reporting three scenarios, (i) little variation of concentrations with WD at downwind receptors; (ii) the highest concentrations recorded when WD was perpendicular to the Rd; and (iii) no significant changes in concentrations for a WD angle of 90 degrees. The spatial distributions of PM2.5 observed during the sampling campaign suggest no significant changes for an angle of WD between 0–100 degrees, which is in good agreement with the results reported by Venkatram et al. [21] for such a scenario. Moreover, the similar behaviour in the PM2.5 peaks observed for cycling and walking despite the differences in different WD and WS occurrence may confirm their non-significant effect on the exposure to peaks in PM2.5. Our results agree well with those of modelling reported by Batterman et al. [19] in Detroit, U.S. of concentrations of pollutants decreasing with distance from the road, as observed here for walking, and hotspots in PM2.5 observed close to road intersections.

4.3. PM2.5 Mass Inhalation

Mass inhalation of PM2.5 by distance travelled (µg inhaled km−1) calculated in this study are compared with that estimated for European and USA urban areas to put the inhalation values estimated at the MCMA in context. Consistently with our results, Ham et al. [23] and Panis et al. [31] reported the highest inhalation of PM2.5 for commuters by bicycle compared to public bus and private car in three cities of Belgium and Sacramento, USA, respectively. However, inhalation for cyclists in our study exceeds between 14.3 to 18.5 times and 7.5 to 8.5 times that reported in Sacramento and in Belgium, respectively. Similarly, inhalation for car commuters in the MCMA was higher between 3.5 to 4.8 times that reported for Belgium [31], and 7.5 times that in Sacramento where commuting in public bus results in lower inhalation between 4.3 to 8.6 times than in MCMA [23]. Although such studies used a different approach to estimate inhalation based on average exposure levels, compared to the average levels by exposure windows used in this study, the results are consistent and confirm that cyclists exhibit the highest exposure risk followed by public buses and private cars. Public transport emissions per person are lower than from private cars [27]; nevertheless, highly pollutant buses expose other cleaner modes of transport, such as cycling, beyond acceptable limits, thus discouraging its use and popularisation. This suggests that to protect cyclist commuters, policy makers should consider the implementation of cycling lanes away from motorised traffic and along roads with low traffic, while inhalation reduction of public transport commuters could be achieved by the introduction of cleaner diesel technologies currently available in the market.

5. Conclusions

We assessed personal exposure to PM2.5 in transport micro-environments within the MCMA. Five modes of transport and two routes were assessed, with average personal exposure to PM2.5 during a single commute ranging between 16.5 µg m−3 for walking and 81.7 µg m−3 for cycling. The maximum exposure concentrations were observed for cyclists while the lowest ones corresponded to walking and CC. For commuters in motor vehicles, we observed a lower inhalation of PM2.5 per km travelled for private cars than for public buses. The higher inhalation of PM2.5 aboard diesel buses was likely due to recirculation of exhaust emissions inside the cabin due to open windows. This result disagrees with those reported in the USA and Europe where public transport often circulates with operating air conditioning and with closed windows, unlike in the MCMA. For active commuters, the highest inhalation of PM2.5 per km travelled was observed for cycling compared to walking. Estimated residuals revealed a significant increase in inhalation due to the interception of high PM2.5 concentration peaks. The use of residuals allowed comparison of local area and traffic derived exposure and inhalation, however, the definition of a baseline based on filtering techniques could improve such a comparison.
The results reported in this study provide information on PM2.5 exposure levels in transport micro-environments together with information on critical regions of high concentrations in PM2.5. This information can be used by local authorities, in particular, to design and modify lines for cyclists in an effort to secure distances away from diesel emissions. For instance, cyclist inhalation could be significantly reduced by installing bicycle lanes near footpaths and allowing a parking lane between those and traffic lanes. Another alternative could be to designate cycle lanes on adjacent streets to main roads. In addition, the introduction of hybrid and electric vehicles such as trolleybuses seems to represent a feasible opportunity not only to reduce motor vehicle emissions, but to decrease exposure risk for cyclists and pedestrians. The results presented in this study are consistent with existing studies that reported the highest exposure risk for cyclists circulating near or with the traffic stream. Our experiments were carried out along a representative road, where different modes of transport circulate. Continuous monitoring of exposure concentrations along other major roads would aid interpretation of the PM2.5 dynamics observed, especially on those with high traffic loading.

Supplementary Materials

The following are available online at https://www.mdpi.com/2073-4433/9/2/57/s1.

Acknowledgments

This research was supported by funds from the Secretaría de Ciencia, Tecnología e Innovación (SECITI) of the Mexico City. The data analyses were performed through grants SECITI/2016/103 and CDMX SECITI/056/16. Grant-aided support to Iván Y. Hernández Paniagua from the Coordinación de la Investigación Científica (CIC) of the Universidad Nacional Autónoma de México (UNAM) is gratefully acknowledged. We would like to thank Zulema L. Aranda-Ayala for conducting the oximetry tests.

Author Contributions

Iván Hernández analysed the data and drafted the original version of the manuscript. Gema Andraca and Aron Jazcilevich designed the experiments and performed the measurements. Diego Ayala calculated personal exposure and inhalation. Juan Zavala analysed and represented the data spatially. Luis Ruiz provided the closed chamber, air zero generator and flow meter, and traffic statistics for the study area. Silvia Cid, Luis Torre and Laura Gochicoa performed the cardiopulmonary tests. Irma Rosas provided funding to perform the data analyses. All co-authors re-viewed, revised and approved the final manuscript.

Conflicts of Interest

The authors declare no conflict of interests. The funding sponsors had no role in the design of the study; in the collection, analyses or interpretation of data; in the writing of the manuscript; nor in the decision to publish the results.

References

  1. Adams, H.S.; Nieuwenhuijsen, M.J.; Colvile, R.N. Determinants of fine particle (PM2.5) personal exposure levels in transport microenvironments, London, UK. Atmos. Environ. 2001, 35, 4557–4566. [Google Scholar] [CrossRef]
  2. Smith, J.D.; Mitsakou, C.; Kitwiroon, N.; Barratt, B.M.; Walton, H.A.; Taylor, J.G.; Anderson, H.R.; Kelly, F.J.; Beevers, S.D. London Hybrid exposure model: Improving human exposure estimates to NO2 and PM2.5 in an urban setting. Environ. Sci. Technol. 2016, 50, 11760–11768. [Google Scholar] [CrossRef] [PubMed]
  3. IGECEM. Encuesta Origen-Destino 2007. Principales Resultados. Available online: http://igecem.edomex.gob.mx/descargas/estadistica/ENCUESTADEORIGEN/EOD2007.pdf (accessed on 20 October 2017).
  4. Shields, K.N.; Cavallari, J.M.; Hunt, M.J.O.; Lazo, M.; Molina, M.; Molina, L.; Holguin, F. Traffic-related air pollution exposures and changes in heart rate variability in Mexico City: A panel study. Environ. Health 2013, 12, 7. [Google Scholar] [CrossRef] [PubMed]
  5. Dons, E.; Panis, L.I.; Van Poppel, M.; Theunis, J.; Willems, H.; Torfs, R.; Wets, G. Impact of time–activity patterns on personal exposure to black carbon. Atmos. Environ. 2011, 45, 3594–3602. [Google Scholar] [CrossRef]
  6. Hertel, O.; Hvidberg, M.; Ketzel, M.; Storm, L.; Stausgaard, L. A proper choice of route significantly reduces air pollution exposure—A study on bicycle and bus trips in urban streets. Sci. Total Environ. 2008, 389, 58–70. [Google Scholar] [CrossRef] [PubMed]
  7. Adams, M.D.; Requia, W.J. How private vehicle use increases ambient air pollution concentrations at schools during the morning drop-off of children. Atmos. Environ. 2017, 165, 264–273. [Google Scholar] [CrossRef]
  8. Wu, D.L.; Lin, M.; Chan, C.Y.; Li, W.Z.; Tao, J.; Li, Y.P.; Bu, C.W. Influences of commuting mode, air conditioning mode and meteorological parameters on fine particle (PM2.5) exposure levels in traffic microenvironments. Aerosol Air Qual. Res. 2013, 13, 709–720. [Google Scholar] [CrossRef]
  9. Okokon, E.O.; Yli-Tuomi, T.; Turunen, A.W.; Taimisto, P.; Pennanen, A.; Vouitsis, I.; Samaras, Z.; Voogt, M.; Keuken, M.; Lanki, T. Particulates and noise exposure during bicycle, bus and car commuting: A study in three European cities. Environ. Res. 2017, 154, 181–189. [Google Scholar] [CrossRef] [PubMed]
  10. Fernandez-Bremauntz, A.A.; Ashmore, M.R. Exposure of commuters to carbon monoxide in Mexico City—I. Measurement of in-vehicle concentrations. Atmos. Environ. 1995, 29, 525–532. [Google Scholar] [CrossRef]
  11. Serrano-Trespalacios, P.I.; Ryan, L.; Spengler, J.D. Ambient, indoor and personal exposure relationships of volatile organic compounds in Mexico City Metropolitan Area. J. Expo. Sci. Environ. Epidemiol. 2004, 14, S118–S132. [Google Scholar] [CrossRef] [PubMed]
  12. Vallejo, M.; Lerma, C.; Infante, O.; Hermosillo, A.G.; Riojas-Rodriguez, H.; Cárdenas, M. Personal exposure to particulate matter less than 2.5 μm in Mexico City: A pilot study. J. Expo. Sci. Environ. Epidemiol. 2004, 14, 323–329. [Google Scholar] [CrossRef] [PubMed]
  13. SEDEMA (Secretaría del Medio Ambiente de la Ciudad de México). Calidad del aire en la Ciudad de México, Informe 2015. Available online: http://www.aire.cdmx.gob.mx/descargas/publicaciones/flippingbook/informe_anual_calidad_aire_2015v3/files/downloads/Informe2015v3.pdf (accessed on 8 February 2018).
  14. R Core Team. R: A Language and Environment for Statistical Computing. Available online: http://www.Rproject.org/ (accessed on 25 September 2017).
  15. Google Earth Pro (Version 7.3.0.3832) [Software]; Google Inc.: Mountain View, CA, USA, 2017.
  16. USEPA (United States Environmental Protection Agency). Total Risk Integrated Methodology (TRIM) Air Pollutants Exposure Model Documentation (TRIM.Expo/APEX, Version 4). Volume II: Technical Support Document. Available online: http://www.epa.gov/fera/total-risk-integrated-methodology-trim-trimexpo (accessed on 30 October 2017).
  17. Quiros, D.C.; Lee, E.S.; Wang, R.; Zhu, Y. Ultrafine particle exposures while walking, cycling, and driving along an urban residential roadway. Atmos. Environ. 2013, 73, 185–194. [Google Scholar] [CrossRef]
  18. Charron, A.; Harrison, R.M. Fine (PM2.5) and coarse (PM2.5–10) particulate matter on a heavily trafficked London highway: Sources and processes. Environ. Sci. Technol. 2005, 39, 7768–7776. [Google Scholar] [CrossRef] [PubMed]
  19. Batterman, S.; Ganguly, R.; Harbin, P. High resolution spatial and temporal mapping of traffic-related air pollutants. Int. J. Environ. Res. Public Health 2015, 12, 3646–3666. [Google Scholar] [CrossRef] [PubMed]
  20. Gulliver, J.; Briggs, D.J. Personal exposure to particulate air pollution in transport microenvironments. Atmos. Environ. 2004, 38, 1–8. [Google Scholar] [CrossRef]
  21. Venkatram, A.; Snyder, M.; Isakov, V.; Kimbrough, S. Impact of wind direction on near-road pollutant concentrations. Atmos. Environ. 2013, 80, 248–258. [Google Scholar] [CrossRef]
  22. Cepeda, M.; Schoufour, J.; Freak-Poli, R.; Koolhaas, C.M.; Dhana, K.; Bramer, W.M.; Franco, O.H. Levels of ambient air pollution according to mode of transport: A systematic review. Lancet Public Health 2017, 2, e23–e34. [Google Scholar] [CrossRef]
  23. Ham, W.; Vijayan, A.; Schulte, N.; Herner, J.D. Commuter exposure to PM2.5, BC, and UFP in six common transport microenvironments in Sacramento, California. Atmos. Environ. 2017, 167, 335–345. [Google Scholar] [CrossRef]
  24. US EPA (Environmental Protection Agency). Diesel Fuel Standards and Rulemakings. Available online: https://www.epa.gov/diesel-fuel-standards/diesel-fuel-standards-and-rulemakings#overview (accessed on 17 January 2018).
  25. Leavey, A.; Reed, N.; Patel, S.; Bradley, K.; Kulkarni, P.; Biswas, P. Comparing on-road real-time simultaneous in-cabin and outdoor particulate and gaseous concentrations for a range of ventilation scenarios. Atmos. Environ. 2017, 166, 130–141. [Google Scholar] [CrossRef] [PubMed]
  26. EUROSTAT (Statistics Explained). Passenger Cars in the EU. Available online: http://ec.europa.eu/eurostat/statistics-explained/index.php/Passenger_cars_in_the_EU (accessed on 30 October 2017).
  27. SEDEMA (Secretaría del Medio Ambiente de la Ciudad de México). Inventario de Emisiones de la CDMX 2014 Contaminantes Criterio, Tóxicos y de Efecto Invernadero. Available online: http://www.aire.cdmx.gob.mx/descargas/publicaciones/flippingbook/inventario-emisiones-cdmx2014-2/IE-CDMX-2014.pdf (accessed on 8 February 2018).
  28. Asmi, E.; Antola, M.; Yli-Tuomi, T.; Jantunen, M.; Aarnio, P.; Mäkelä, T.; Hillamo, R.; Hämeri, K. Driver and passenger exposure to aerosol particles in buses and trams in Helsinki, Finland. Sci. Total Environ. 2009, 407, 2860–2867. [Google Scholar] [CrossRef] [PubMed]
  29. Adar, S.D.; D’Souza, J.; Sheppard, L.; Kaufman, J.D.; Hallstrand, T.S.; Davey, M.E.; Sullivan, J.R.; Jahnke, J.; Koenig, J.; Larson, T.V.; et al. Adopting clean fuels and technologies on school buses. Pollution and health impacts in children. Am. J. Respir. Crit. Care Med. 2015, 191, 1413–1421. [Google Scholar] [CrossRef] [PubMed]
  30. Jazcilevich, A.; Vázquez, J.M.M.; Ramírez, P.L.; Pérez, I.R. Economic-environmental analysis of traffic-calming devices. Transp. Res. D Transp. Environ. 2015, 36, 86–95. [Google Scholar] [CrossRef]
  31. Panis, L.I.; De Geus, B.; Vandenbulcke, G.; Willems, H.; Degraeuwe, B.; Bleux, N.; Mishra, V.; Thomas, I.; Meeusen, R. Exposure to particulate matter in traffic: A comparison of cyclists and car passengers. Atmos. Environ. 2010, 44, 2263–2270. [Google Scholar] [CrossRef]
Figure 1. (a) The Miguel Angel de Quevedo (MAQ) Rd and the Coyoacán (COY) monitoring site close to MAQ Rd in the local context; (b) The MAQ Rd in relation to the whole Mexico City Metropolitan Area (MCMA).
Figure 1. (a) The Miguel Angel de Quevedo (MAQ) Rd and the Coyoacán (COY) monitoring site close to MAQ Rd in the local context; (b) The MAQ Rd in relation to the whole Mexico City Metropolitan Area (MCMA).
Atmosphere 09 00057 g001
Figure 2. Boxplot of PM2.5 data recorded along the MAQ Rd in southern MCMA by mode of transport and route during the sampling campaign of November 2014. The eastward public bus journey was made on a trolleybus (yellow) and the westward journey on a diesel bus (orange). Red arrows show wind angle scaled to wind speed i.e., the longer the arrow, the higher the wind speed, as recorded at the COY site.
Figure 2. Boxplot of PM2.5 data recorded along the MAQ Rd in southern MCMA by mode of transport and route during the sampling campaign of November 2014. The eastward public bus journey was made on a trolleybus (yellow) and the westward journey on a diesel bus (orange). Red arrows show wind angle scaled to wind speed i.e., the longer the arrow, the higher the wind speed, as recorded at the COY site.
Atmosphere 09 00057 g002
Figure 3. (a) Spatial location of PM2.5 by mode of transport recorded during MAQ to TAX (Taxqueña) journeys; (b) PM2.5 time series per journey. Street names indicated in the X-axis correspond to 4-way intersections. The horizontal dashed lines show corresponding 1-h averages of PM2.5 recorded at COY used to calculate exposure residuals.
Figure 3. (a) Spatial location of PM2.5 by mode of transport recorded during MAQ to TAX (Taxqueña) journeys; (b) PM2.5 time series per journey. Street names indicated in the X-axis correspond to 4-way intersections. The horizontal dashed lines show corresponding 1-h averages of PM2.5 recorded at COY used to calculate exposure residuals.
Atmosphere 09 00057 g003
Figure 4. (a) Spatial location of PM2.5 by mode of transport recorded during TAX to MAQ journeys; (b) PM2.5 time series per journey. Street names indicated in the X-axis correspond to 4-way intersections. The horizontal dashed lines show corresponding 1-h averages of PM2.5 recorded at COY used to calculate exposure residuals.
Figure 4. (a) Spatial location of PM2.5 by mode of transport recorded during TAX to MAQ journeys; (b) PM2.5 time series per journey. Street names indicated in the X-axis correspond to 4-way intersections. The horizontal dashed lines show corresponding 1-h averages of PM2.5 recorded at COY used to calculate exposure residuals.
Atmosphere 09 00057 g004
Table 1. Transport modes and distances travelled. (MAQ: Miguel Angel de Quevedo; TAX: Taxqueña; CC: conventional car; HEC: hybrid-electric car).
Table 1. Transport modes and distances travelled. (MAQ: Miguel Angel de Quevedo; TAX: Taxqueña; CC: conventional car; HEC: hybrid-electric car).
Transport ModeMAQ-TAXTAX-MAQ
From:To:Distance (km)Duration (h)From:To:Distance (km)Duration (h)
WalkingMAQ roundaboutDivision del Norte Rd3.51.07Division del Norte RdMAQ roundabout3.31.02
CyclingMAQ roundaboutDivision del Norte Rd3.80.51Division del Norte RdMAQ roundabout3.50.49
TrolleybusMAQ roundaboutTAX trolleybus station4.20.30----
Diesel bus----Bus stationMAQ roundabout4.30.36
CCMAQ roundaboutDivision del Norte Rd 1,240.35Division del Norte RdMAQ roundabout 33.90.34
HECMAQ roundaboutDivision del Norte Rd 1,24.50.39Division del Norte RdMAQ roundabout 34.50.37
1 Trip with closed windows and operating air conditioning; 2 Data discarded due to increases in relative humidity inside the vehicles (>85%), which affected instrumentation; 3 Trip with open windows.
Table 2. Respiratory parameters used in the current study to estimate inhalation of PM2.5 by mode of transport.
Table 2. Respiratory parameters used in the current study to estimate inhalation of PM2.5 by mode of transport.
Transport ModeExposure Window Duration in s (p) *Respiratory Rate in min−1 (RR)VE in L min−1
Diesel bus[4,5]1311.4
Trolleybus[4,5]1311.4
CC and HEC[4,5]1514.3
Walking[3,4]1818.3
Bicycle[1,2]3442.1
* Lower and upper bounds were determined by rounding p. Note that for the reader convenience p is expressed in seconds.
Table 3. PM2.5 exposure concentrations in µg m−3 by transport mode measured along MAQ Rd.
Table 3. PM2.5 exposure concentrations in µg m−3 by transport mode measured along MAQ Rd.
Mode of Transport MAQ to TAX
Min *MedianArithmetic AverageGeometric AverageMax **COY 1-h Average
Walking11.020.721.0 ± 7.919.9112.227
Bicycle63.181.480.8 ± 9.380.5102.950
Trolleybus27.440.439.9 ± 6.439.454.745
CCDiscarded
HECDiscarded
Mode of TransportTAX to MAQ
Min *MedianArithmetic AverageGeometric AverageMax **COY 1-h Average
Walking9.114.716.5 ± 6.515.7104.022
Bicycle42.081.881.7 ± 9.181.1163.743
Diesel bus39.048.849.8 ± 5.349.571.332
CC9.027.128.3 ± 8.527.7111.942
HEC11.019.820.3 ± 4.219.833.357
* Minimum; ** Maximum.
Table 4. Distance, inhaled PM2.5 and residual inhalation by route and mode of transport along MAQ Rd.
Table 4. Distance, inhaled PM2.5 and residual inhalation by route and mode of transport along MAQ Rd.
Mode of TransportMAQ to TAX
On-Field MeasurementsCOY MeasurementsResidual Inhalation * (%)
µg Inhaledµg Inhaled km−1µg Inhaledµg Inhaled km−1
Walking12.1 ± 1.93.5 ± 0.615.6 ± 2.54.5 ± 0.7−28.6
Bicycle147.4 ± 69.439.0 ± 18.491.3 ± 43.024.2 ± 11.437.9
Trolleybus6.3 ± 1.01.5 ± 0.27.1 ± 1.11.7 ± 0.3−13.3
CC-----
HEC-----
Mode of TransportMAQ to TAX
On-Field MeasurementsCOY MeasurementsResidual Inhalation * (%)
µg Inhaledµg Inhaled km−1µg Inhaledµg Inhaled km−1
Walking17.0 ± 3.45.3 ± 1.122.6 ± 4.57.0 ± 1.4−32.1
Bicycle104.5 ± 49.130.1 ± 14.255.0 ± 25.915.9 ± 7.547.2
Diesel bus12.7 ± 1.93.0 ± 0.58.2 ± 1.31.9 ± 0.336.7
CC9.4 ± 2.42.4 ± 0.613.8 ± 3.53.6 ± 0.9−50.0
HEC4.7 ± 1.21.0 ± 0.313.2 ± 3.42.9 ± 0.7−190.1
* Estimated as on-field derived inhalation minus COY derived inhalation.

Share and Cite

MDPI and ACS Style

Hernández-Paniagua, I.Y.; Andraca-Ayala, G.L.; Diego-Ayala, U.; Ruiz-Suarez, L.G.; Zavala-Reyes, J.C.; Cid-Juárez, S.; Torre-Bouscoulet, L.; Gochicoa-Rangel, L.; Rosas-Pérez, I.; Jazcilevich, A. Personal Exposure to PM2.5 in the Megacity of Mexico: A Multi-Mode Transport Study. Atmosphere 2018, 9, 57. https://doi.org/10.3390/atmos9020057

AMA Style

Hernández-Paniagua IY, Andraca-Ayala GL, Diego-Ayala U, Ruiz-Suarez LG, Zavala-Reyes JC, Cid-Juárez S, Torre-Bouscoulet L, Gochicoa-Rangel L, Rosas-Pérez I, Jazcilevich A. Personal Exposure to PM2.5 in the Megacity of Mexico: A Multi-Mode Transport Study. Atmosphere. 2018; 9(2):57. https://doi.org/10.3390/atmos9020057

Chicago/Turabian Style

Hernández-Paniagua, Iván Y., Gema L. Andraca-Ayala, Ulises Diego-Ayala, Luis G. Ruiz-Suarez, Juan C. Zavala-Reyes, Silvia Cid-Juárez, Luis Torre-Bouscoulet, Laura Gochicoa-Rangel, Irma Rosas-Pérez, and Arón Jazcilevich. 2018. "Personal Exposure to PM2.5 in the Megacity of Mexico: A Multi-Mode Transport Study" Atmosphere 9, no. 2: 57. https://doi.org/10.3390/atmos9020057

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop