Environ Geochem Health
DOI 10.1007/s10653-016-9844-y
ORIGINAL PAPER
Characteristics of PM2.5, CO2 and particle-number
concentration in mass transit railway carriages in Hong
Kong
Hai-Long Zheng . Wen-Jing Deng .
Yan Cheng . Wei Guo
Received: 5 February 2016 / Accepted: 14 June 2016
Ó Springer Science+Business Media Dordrecht 2016
Abstract Fine particulate matter (PM2.5) levels,
carbon dioxide (CO2) levels and particle-number
concentrations (PNC) were monitored in train carriages on seven routes of the mass transit railway in
Hong Kong between March and May 2014, using realtime monitoring instruments. The 8-h average PM2.5
levels in carriages on the seven routes ranged from
24.1 to 49.8 lg/m3, higher than levels in Finland and
similar to those in New York, and in most cases
exceeding the standard set by the World Health
Organisation (25 lg/m3). The CO2 concentration
ranged from 714 to 1801 ppm on four of the routes,
generally exceeding indoor air quality guidelines
(1000 ppm over 8 h) and reaching levels as high as
those in Beijing. PNC ranged from 1506 to 11,570
particles/cm3, lower than readings in Sydney and
higher than readings in Taipei. Correlation analysis
indicated that the number of passengers in a given
carriage did not affect the PM2.5 concentration or PNC
in the carriage. However, a significant positive correlation (p \ 0.001, R2 = 0.834) was observed between
passenger numbers and CO2 levels, with each
H.-L. Zheng W.-J. Deng (&)
Department of Science and Environmental Studies,
The Education University of Hong Kong, Tai Po,
Hong Kong, China
e-mail: [email protected]
Y. Cheng W. Guo
School of Human Settlements and Civil Engineering,
Xi’an Jiaotong University, Xi’an, China
passenger contributing approximately 7.7–9.8 ppm
of CO2. The real-time measurements of PM2.5 and
PNC varied considerably, rising when carriage doors
opened on arrival at a station and when passengers
inside the carriage were more active. This suggests
that air pollutants outside the train and passenger
movements may contribute to PM2.5 levels and PNC.
Assessment of the risk associated with PM2.5 exposure
revealed that children are most severely affected by
PM2.5 pollution, followed in order by juveniles, adults
and the elderly. In addition, females were found to be
more vulnerable to PM2.5 pollution than males
(p \ 0.001), and different subway lines were associated with different levels of risk.
Keywords CO2 PM2.5 Particle-number
concentration (PNC) Mass transit railway (MTR) Hong Kong
Introduction
Hong Kong’s mass transit railway (MTR) resembles
other major mass-transportation systems in mid-sized
and large cities. It covers most of the area inside the
city and is considered to be the most convenient mode
of transport for commuters. Approximately 4.04
million passengers use the MTR per day. Most users
of rapid-transit systems (metros), such as those in
Toronto (Havas et al. 2004) and Taipei (Cheng et al.
123
Environ Geochem Health
2012), travel by metro for more than 2 h every day.
Metro train systems are often believed to be the
‘cleanest’ mode of transportation, for the following
reasons: (a) they are electrically operated, reducing
their direct emission of air pollutants; (b) they have a
large capacity; and (c) they operate underground and
thus ease traffic congestion. However, some researchers have found that air quality in metro systems is
adversely affected by high levels of particulate matter
(PM), carbon dioxide (CO2) and heavy metals (Querol
et al. 2012; Nasir and Colbeck 2009; Gustafsson et al.
2012; Knibbs and deDear 2010). PM is widely
recognised as a risk factor in pulmonary and cardiovascular diseases and cancer (Pope III et al. 2004;
Alfaro-Moreno et al. 2007). Boldo et al. (2006) found
that decreasing the level of fine PM (PM2.5) may
reduce the number of cases of early death and increase
life expectancy. PM2.5 can be transported by the blood
to organs such as the liver within 4–24 h of exposure
(Oberdörster et al. 2002). Studies have shown that
exposure to particulate air pollution is closely linked to
daily mortality and number of hospital admissions
(Linn et al. 2000; Halonen et al. 2009). The concentration of PM2.5 in the air inside metro carriages has
important implications for passenger health. High
levels of PM and CO2 and high particle-number
concentrations (PNC) have been detected in masstransportation systems in London (Seaton et al. 2005),
Beijing (Li et al. 2007), Taiwan (Hsu and Huang 2009;
Cheng and Yan 2011), Helsinki (Aarnio et al. 2005),
Seoul (Park and Ha 2008; Kwon et al. 2008) and
Mexico City (Mugica-Álvarez et al. 2012). However,
little information is available on indoor air quality
(IAQ) in Hong Kong’s metro system. The differences
in pollutant levels between subway lines have not been
examined, and although passengers spend far more
time in subway carriages than on the platform,
carriage IAQ has received far less systematic investigation than platform IAQ. Most researchers investigating PM2.5 levels have focused on roadside ambient
air quality or taken measurements only on metro
platforms (Chan et al. 2002a, b; Lau and Chan 2003;
Raut et al. 2009); none have compared PM2.5 levels
across the various lines of the MTR. Due to the serious
health hazard associated with air pollution, the large
number of passengers on the MTR and commuters’
long travelling times, the air quality in MTR carriages
should be evaluated. The objectives of this study were
to monitor the levels of PM2.5, PNC, CO2 and the net
123
number of passengers between consecutive stations on
MTR routes; to compare the results between lines or
stations; and to examine the relationships between
pollutant levels and passenger health.
Materials and methods
MTR lines
Hong Kong’s MTR currently consists of 11 lines: East
Rail (East), West Rail (West), Kwun Tong (KT),
Tseung Kwan O (TKO), Hong Kong Island (Island),
Tsuen Wan (TW), Disney and Resort, Ma On Shan,
Tung Chung, Airport Express and Light Rail. The
combined length of these lines is approximately
210 km, and the MTR is used by an average of 5.2
million passengers each weekday (MTR 2014). Six
lines and one multiple-line route were investigated:
East, West, KT, TKO, Island, TW and the route from
Sha Tin to Tsim Sha Tsui (ST–TST) (Fig. 1). Passengers travelling from ST to TST must change trains
twice: once at Kowloon Tong Station and once at
Mong Kok Station. These lines connect three districts
of Hong Kong: Hong Kong Island, Kowloon Peninsula
and the New Territories. PM2.5 level, CO2 level, PNC,
number of passengers, relative humidity and temperature were measured in carriages on each route for two
8-h periods per day (8:00–12:00 and 13:00–17:00)
between March and May 2014.
Monitoring procedures
A Q-Trak IAQ monitor (TSI Model 7575) with a
portable probe (Model 982) was used to measure the
CO2 concentrations in carriages travelling on the
above seven routes of the MTR. The real-time PM2.5
level was measured at a sampling rate of 1.7 L/min,
using a Dust-Trak sampler (TSI Model 8532). The
data obtained using both the Dust-Trak and the Q-Trak
were averaged over 10-s intervals, and all of the
readings were recorded in a data logger. A TSI 3007
condensation particle counter (CPC) was used to
measure total PNC in the range 0.01 lm to [1 lm,
although the overwhelming majority of particles
recorded in urban areas fall within the ultrafine
particle size range (Morawska et al. 2008). The CPC
is capable of detecting particle concentrations up to
1 9 105 p cm-3. In a comparable experiment, we
Environ Geochem Health
Fig. 1 Seven MTR routes
sampled
found that the TSI 3007 is capable of producing
readings up to 3 9 105 p cm-3. The air samplers were
calibrated before sampling. The Dust-Trak and TSI
3007 samplers were checked pre- and post-zero by
attaching a high-efficiency particulate arrestance filter
to their inlets before every survey. The Q-Trak, DustTrak and TSI 3007 CPC were placed roughly
1.0–1.5 m above the floor of the MTR carriages,
following Cheng et al. (2008). The number of
passengers inside the carriages was monitored after
departure from each station for the whole sampling
period (8 h).
Quality assurance
Real-time instruments can help to identify the changing patterns of indoor air pollutant levels that
contribute significantly to the risk associated with
exposure. However, real-time measurements do not
always reflect the true particle mass concentration,
even when the instruments are calibrated and recalibrated for accuracy. As the TSI Dust-Trak typically
overestimates PM concentration (Yanosky et al. 2002;
McNamara et al. 2011), a portable Tas-Minivol air
sampler (Airmetics, USA) with a 47-mm quartz filter
(Pure Quartz Filter, 47 mm, Whatman Inc., USA) was
used simultaneously to obtain a correction factor.
Statistical analysis revealed the following linear
relationship between the average PM2.5 levels measured using the Dust-Trak and Minivol air samplers:
y = 0.50x ? 0.12 (R2 = 0.99). The experimental
results suggested that the TSI Dust-Trak overestimated the PM2.5 concentration in the MTR by
approximately 2.0 times. This calibration factor is
similar to that obtained by Morawska et al. (2008) and
Chan et al. (2002a, b).
Data analysis
SPSS Version 19.0 and OriginPro 8.0 were used to
analyse the data. A t test was conducted to compare the
PM2.5 and CO2 concentrations detected in the underground and overground trains. Correlation analysis
was performed to examine the matrix of relationships
between PM2.5 concentration, CO2 concentration and
PNC.
Results and discussion
PM2.5 concentration in MTR
The PM2.5 readings obtained in MTR carriages on the
above seven routes are summarised in Table 1. The
overall average concentration of PM2.5 ranged from
17.6 to 63.0 lg/m3. The highest PM2.5 level was
123
Environ Geochem Health
Table 1 Levels of PM2.5 in MTR carriages on selected routes (lg/m3)
Line
8 h mean
KwunTong line (KT)
42.9
SDa
5.26
12.5
Rangeb
8 h mean above the
WHO standard (%)
38.8–53.9
71.6
East rail line (East)
45.3
22.9–60.0
81.2
West rail line (West)
25.6
5.59
17.9–34.9
2.4
Island line (Island)
24.1
4.11
17.6–30.8
Tseung Kwan O line (TKO)
49.8
7.19
38.2–63.0
99.2
Tsuen Wan line (TW)
35.2
3.25
29.9–40.0
40.8
Sha Tin to Tsim Sha Tsui (ST–TST)
42.0
7.60
32.4–54.9
68.0
a
Standard deviation
b
Minimal value–maximal value
Table 2 Comparison of in-train PM2.5 concentration data with
results of other studies (lg/m3)
Study
City
Mean
Range
Chan et al. (2002a, b)
Hong Kong
33
21–48
Gómez-Perales et al. (2004)
Mexico
61
31–99
Aarnio et al. (2005)
Helsinki
21
17–26
Seaton et al. (2005)
London
170
130–200
Li et al. (2006)
Kim et al. (2008)
Beijing
Seoul
37
126
13–111
115–136
Huang and Hsu (2009)
Taiwan
24
Wang and Gao (2011)
New York
39
34–44
This study
Hong Kong
38
18–63
4–120
observed in carriages on the TKO line, located inland,
and the lowest PM2.5 was found in carriages on the
Island line, in a coastal region of Hong Kong. The 8-h
average PM2.5 concentration was listed in Table 1,
ranged from 24.1 to 49.8 lg/m3. Both the PM2.5
concentration and the ratio of PM2.5 concentration to
calibrated real-time PM2.5 concentration were higher
in carriages on the KT and TKO lines than in carriages
on the other five routes. A lower ratio was observed in
carriages on the West and Island lines, which exhibited lower PM2.5 concentrations. On average, the
PM2.5 concentration was higher in inland areas than in
coastal areas, and the mean PM2.5 levels detected on
all of the routes except the Island line were higher than
the IAQ standard set by the World Health Organisation (WHO) (25.0 lg/m3).
Table 2 shows the results of similar studies conducted in other cities. The PM2.5 concentration
123
measured in Hong Kong’s MTR (38 lg/m3) was
noticeably higher than that measured in mass-transit
systems in Helsinki, Finland (21 lg/m3) (Aarnio et al.
2005) and Taiwan (24 lg/m3) (Huang and Hsu 2009);
lower than readings taken in Mexico (61 lg/m3)
(Gómez-Perales et al. 2004), London (170 lg/m3)
(Seaton et al. 2005) and Seoul (126 lg/m3) (Kim et al.
2008), and similar to readings taken in Beijing (37 lg/
m3) (Li et al. 2006) and New York (40 lg/m3) (Wang
and Gao 2011). PM2.5 concentrations have been found
to differ substantially between metro systems worldwide, probably due to variation in time and place,
meteorological factors and ambient urban pollution
levels (Kim et al. 2008). These factors may also
explain the differences in PM2.5 concentrations
between the seven routes sampled in this study. For
example, higher PM2.5 levels were found in carriages
travelling between Shatin and Kowloon Tong using
the ST–TST route (holiday) than in carriages on the
East line (weekday). Passenger numbers were not
responsible for this difference. Carriages on the Island
line held 1.38 times more people (mean: 77 people per
carriage) than those on the West line (mean: 56 people
per carriage), but the PM2.5 concentration was lower in
carriages on the Island line than those on the West line,
as shown in Table 1. Similarly, Kwon et al. (2008)
found that the number of passengers per carriage was
highly correlated with the carriage CO2 level, but not
the PM2.5 level. It is worth noting that the concentration of PM2.5 increased by 5–6 times to 167–255 lg/
m3 in Beacon Hill Tunnel on the East line, between Tai
Wai and Kowloon Tong. All of the trains on the East
line are electrified; the high PM level in Beacon Hill
Tunnel was mainly due to the re-suspension of road
Environ Geochem Health
Fig. 2 Real-time variation in PM2.5 levels in trains on three lines
and air dust inside the tunnel. When a train entered the
tunnel at high speed, the settled PM in the tunnel resuspended and entered the train through the central airconditioning system, significantly increasing the
PM2.5 concentration inside the carriages (Chan et al.
2002a, b). In addition, the real-time readings of PM2.5
increased when carriage doors opened on arrival at a
station, but remained relatively stable while the train
was in motion. The real-time variation in PM2.5 levels
on three lines of the MTR is shown in Fig. 2. It is clear
from this figure that the real-time PM2.5 concentration
varied considerably along each route, rising when the
123
Environ Geochem Health
(1000 ppm over 8 h), and the level of CO2 in carriages
on the West line approached this standard. High CO2
levels have also been reported in other studies of metro
systems. Park and Ha (2008) detected a 1800-ppm
concentration of CO2 in Seoul subway trains; Kam
et al. (2011) reported a 1200-ppm CO2 concentration
in Los Angeles metro trains; and Li et al. (2006) found
CO2 levels in Beijing’s overground trains to reach
1110 and 1270 ppm in the summer and winter,
respectively. Many studies have investigated the
factors that increase CO2 concentration, such as
environmental conditions (Cheng et al. 2012), meteorological variation (Fang and Chang 2010) and
ambient urban pollution (Tsai et al. 2008).
In the current study, a significant positive relationship was consistently observed between the number of
passengers in a carriage and the carriage’s CO2 level
(all p \ 0.001, R2 = 0.834). This correlation was
most significant on the West line (p \ 0.001,
R2 = 0.951). Figure 3 shows the relationship between
passenger numbers and CO2 levels inside metro trains
on selected routes. Each passenger was responsible for
approximately 7.7–9.8 ppm of the carriage’s CO2
level. A similar result was obtained by Cheng et al.
(2012) in a study of the Taipei metropolitan area and
Table 3 Levels of CO2 (ppm) in MTR carriages and numbers
of passengers per carriage on East, KT, West and Island lines
Line
CO2
Mean ± SD
ppm
Range
Average
number of
passengers
East rail line (East)
1110 ± 169
877–1437
85
Kwun Tong line (KT)
1139 ± 246
862–1801
79
West rail line (West)
992 ± 271
714–1497
56
1083 ± 140
925–1252
77
Island line (Island)
trains arrived at stations and their doors opened. It was
also noted that carriages with more active passengers
had a greater indoor PM2.5 mass concentration. Air
pollutants outside the trains may be a source of PM2.5
in MTR carriages, as in the case of Beacon Hill Tunnel
described above (Chan et al. 2002a, b).
Factors influencing CO2 levels
Table 3 presents the CO2 measurements taken inside
the metro trains on selected routes. The mean CO2
levels detected in carriages on the KT, East and Island
lines were higher than the IAQ standard set by the
Hong Kong Environmental Protection Department
ppm
1500
East
1400
2
CO2
CO 2
Y=8.36X+392.95
1200
Adj.R =0.7254
ppm
West
Y=9.37X+440.33
1200
2
Adj.R =0.8956
900
1000
800
600
60
80
100
30
120
60
90
Number of passengers
Number of passengers
2000
2000
ppm
ppm
KT
Island
Y=7.93X+340.27
1500
2
Adj.R =0.66347
2
Adj.R =0.81338
CO 2
CO 2
Y=7.72X+521.09
1500
1000
1000
50
100
150
50
Number of passengers
Fig. 3 Correlation between passenger number and CO2 level in metro carriages
123
100
150
Number of passengers
200
Environ Geochem Health
Table 4 Levels of PNC in MTR carriages on selected routes
(particles/cm3)
Line
Sha Tin to Tsim Sha
Tsui (ST–TST)
Tseung Kwan O line (TKO)
West rail line (West)
Mean
8894
SD
Range
1254
7047–11,570
11,282
2277
7424–11,272
3098
1671
1506–5504
the Taipei metro system: each passenger was found to
be responsible for 6–9 ppm of the CO2 concentration.
Huang and Hsu (2009) and Li et al. (2006) also
reported a significant positive relationship between
CO2 levels and the number of passengers in buses and
trains.
Differences in PNC between stations and lines
Table 4 presents the PNC measurements obtained on
three routes (the TKO line, the West line and the ST–
TST route). The results of a t test indicated that the
PNC in carriages on the West line was significantly
lower than those in carriages on the TKO line
(p \ 0.001) and the ST–TST route (p \ 0.001).
During the sampling period, the mean PNC on the
TKO line and the East line was approximately 3.64
and 2.87 times greater, respectively, than that on the
West line.
The PNC detected in the Hong Kong MTR was
lower than that reported in certain other studies of
mass-transit systems. For example, Knibbs and
deDear (2010) used a TSI 3007 (particle diameter
0.01 to [1 lm) to measure PNC in Sydney subway
trains and obtained a mean PNC of nearly 4.6 9 104
particles/cm3. Using a P-Trak (particle diameter 0.02
to [1 lm), Aarnio et al. (2005) detected an average
PNC of 3.1 9 104 particles/cm3 in subway trains in
Helsinki, Finland. However, some researchers have
obtained PNC readings similar to or lower than those
reported in the current study. For example, Cheng
et al. (2009) detected a PNC of 1.1 9 104 particles/
cm3 in Taipei subway trains, using a P-Trak; and
Klepczyńska-Nyström et al. (2012) reported that the
PNC in Stockholm subway trains (again measured
using a P-Trak) was 9330 particles/cm3. The field
measurements of PNC obtained using the P-Trak are
within 20 % of those obtained using a TSI 3007, which
is recognised as a precision instrument (Matson et al.
2004).
Real-time PNC varied considerably in carriages on
the ST–TST route over the 8-h sampling period
(Fig. 4a). Based on passengers’ environment and the
need to change trains during the journey, the route is
divided into five parts: Sha Tin to Kowloon Tong
(ground level), Kowloon Tong Station (first change),
Kowloon Tong to Mong Kok (underground), Mong
Kok Station (second change) and Mong Kok to Tsim
Sha Tsui (underground). Passengers must walk for
5 min, moving underground from ground level, to
change trains at Kowloon Tong Station (an underground station), and for 2 min at Mong Kok Station
(another underground station) to make the second
change. Figure 4b shows the single-trip real-time
fluctuation in PNC and the corresponding passenger
activities from Sha Tin to Tsim Sha Tsui. The mean
PNC was highest during the Mong Kok Station change
(11,470 particles/cm3), followed by the journey from
Sha Tin to Kowloon Tong (11,350 particles/cm3), the
journey from Mong Kok to Tsim Sha Tsui (9077
particles/cm3), the journey from Kowloon Tong to
Mong Kok (8164 particles/cm3) and the Kowloon
Tong Station change (8104 particles/cm3). This can be
explained by the increase in commuter number and
human activity, such as walking, which increased the
proportion of traffic-related fine fraction aerosols. Son
et al. (2013) found 1.0–10 lm of suspended PM in
Seoul’s subway system, which accounted for 78.7 and
83 % of total PM under two sets of operating
conditions: fan at half speed and fan off, respectively.
PM2.5 (1.0–10 lm in diameter) increases significantly
as a result of human indoor activities, such as cleaning
(vacuuming, dusting or sweeping), walking, field
sampling and children’s games (Abt et al. 2000).
Long et al. (2000) reported that indoor activities such
as dusting, vacuuming and vigorously walking cause
settled particles to re-suspend and that this effect is
especially pronounced for particles between 2.5 and
10 lm in size. In the current study, large crowds of
pedestrians increased the PNC at Mong Kok Station.
In addition, some studies have shown that PM levels in
metro systems are positively correlated with PM levels
in the outdoor environment (Kim et al. 2008; Jung
et al. 2010). Mong Kok Station has 14 entrances and is
surrounded by bustling commercial activities and
heavy traffic, allowing a large volume of fine particles
to enter the station. The journey from Sha Tin to
Kowloon Tong (ground level) occurs at a higher
altitude than other parts of the route (underground).
123
Environ Geochem Health
Fig. 4 Real-time variation in PNC in trains. a Daily real-time PNC variation; b Real-time PNC and corresponding human activities
(single trip: ST–TST)
Therefore, the effect of outdoor ambient air on station
PNC is likely to be greater than that of human activity.
Other MTR lines, such as the TKO line and the West
line, showed similar trends.
Health-risk assessment
In general, three main methods are used to quantify
personal exposure to pollutants: (1) direct personal
exposure measurements (Jantunen et al. 2002), (2)
123
measurements of micro-environmental concentrations and time spent in environment (USEPA
2004), and (3) personal-activity information (Meng
et al. 2005). An individual’s micro-environment is
made up of areas such as indoor environment at
home, working environment, leisure locations (e.g.
pubs and cafés), transport settings (e.g. cars, buses
and trains) and shops and parks (dog walking)
(Harrison et al. 2002). In this study, the second
method was used to measure health risk. The two
Environ Geochem Health
Fig. 5 CDI of PM for
females (a) and males (b) on
selected lines
exposure-evaluation methods used in the study are
described in Eq. (1), as follows:
X
Csðt2 t1 Þ
ð1Þ
Eðt1 ; t2 Þ ¼
where t1 and t2 denote the times at which sampling
begins and ends, respectively; E(t1, t2) is the exposure
during this period, and Cs is the average pollutant
concentration between t1 and t2. The unit of measurement is (pollutant concentration * time). Equation (1)
shows that the higher the pollutant concentration in
MTR carriages and the more time spent on the MTR,
the greater the health risk. The data on air pollutants in
the MTR carriages shown above indicate that passengers on the KT line and the TKO line were at particular
risk. Daily PM2.5 exposure on the KT and TKO lines
was 85.8 and 99.6 lg/m3h, respectively, (mean time
spent on trains on these lines: 2 h). Certainly, passengers’ health responses to the same level of exposure
may differ. Risk varies between individuals according
to various factors such as age, body weight, duration of
exposure and health condition. To take these factors
into account, chronic daily intake (CDI) was assessed
as follows:
CDI ¼
C IR IEF
BW
ð2Þ
where C is the concentration of pollutant (mass/
volume), IR is the inhalation rate, BW is body
weight and IEF is the exposure factor (percentage of
staying in MTR account for whole day, 1/12). In the
present study, the contribution of PM2.5 on train is
only a part of daily exposure to PM2.5 (2 h in the
MTR in a whole day), and thus the CDI calculated
in the present study is due to transportation in MTR
only. IR is not fixed: it varies with age, sex, physical
health and labour intensity. During a life, IR first
123
Environ Geochem Health
increases and then decreases gradually, reaching its
maximum point between the ages of 16.5 and 25.
Males have a higher IR than females in every age
group (males: 20.12 m3/day; females: 16.21 m3/day)
(Brochu et al. 2011). The passengers were divided
into four groups according to their age. We obtained
data on IR and BW in different age groups from
Brochu et al. (2011) and the 2013 Key Statistics
Report on Women and Men in Hong Kong (CSDHK
2015) to assess the health risk associated with PM
exposure (male children: 7.35 m3/day and 15.3 kg;
juvenile: 15.60 m3/day and 43.5 kg; adult:
20.12 m3/day and 70.3 kg; elderly: 15.25 m3/day
and 68.9 kg; female: children: 6.90 m3/day and
14.4 kg; juvenile: 13.32 m3/day and 45.2 kg; adult:
16.21 m3/day and 58.9 kg; elderly: 11.51 m3/day
and 57.2 kg). As shown in Fig. 5, children (3 years
old), adolescents (16 years old), adults (35 years
old) and the elderly (65 years old) were found to
experience different levels of risk due to PM
exposure. The order of risk was as follows for all
of
the
sampled
lines:
children [ adolescents [ adults [ the elderly. Females were at
greater risk than males on all of the lines
(p \ 0.001). The subway lines were associated with
different levels of risk, and these differences were
more pronounced for children. CDI was highest on
the TKO line for all four age groups, both male and
female (male children: 1.99 lg/kg-day; adolescent
males: 1.49 lg/kg-day; adult males: 1.18 lg/kg-day;
elderly males: 0.91 lg/kg-day; female children:
1.98 lg/kg-day; adolescent females: 1.22 lg/kgday; adult females: 1.14 lg/kg-day; elderly females:
0.83 lg/kg-day). The lowest CDI was found on the
Island line (male children: 0.96 lg/kg-day; adolescent males: 0.72 lg/kg-day; adult males: 0.57 lg/
kg-day; elderly males: 0.44 lg/kg-day; female children: 0.96 lg/kg-day; adolescent females: 0.59 lg/
kg-day; adult females: 0.55 lg/kg-day; elderly
females: 0.40 lg/kg-day). The risk associated with
exposure decreased with age. However, the data did
not indicate that the elderly were healthier than the
adolescents and the adults. This inconsistency may
be due to the low resistance of the elderly. Certainly,
children were found to be affected most severely by
PM pollution due to their greater exposure to
pollutants and their lower resistance to the damage
caused by environmental pollutants.
123
Conclusion
In this study, commuters’ exposure to CO2 and PM2.5
in Hong Kong’s MTR carriages, along with the PNC
of the system, was measured. The passengers’ in-train
exposure to CO2 and PM2.5 was found to exceed IAQ
and/or WHO standards on most of the lines under
study. A significant positive relationship was found
between the number of passengers in a carriage and the
carriage’s CO2 level. However, no correlation was
found between the number of passengers and either the
PM2.5 level or the PNC. The concentration of PM2.5
was highest in carriages on the TKO line (49.8 lg/m3),
located inland, and lowest in carriages on the Island
line (24.1 lg/m3), in a coastal area of Hong Kong.
Carriage-door opening at stations and human movement were found to increase PNC and PM2.5. In
addition, risk assessment revealed that females are
more severely affected by exposure than males and
that children are the most vulnerable group of
passengers.
Acknowledgments This study was supported by Dean’s
research fund (Ref no. REG-3) and Internal Research Support
(Ref no. R3679) of the Hong Kong Institute of Education.
References
Aarnio, P., Tarja, Y. T., Anu, K., Timo, M., Anne, H., Kaarle,
H., et al. (2005). The concentrations and composition of
and exposure to fine particles (PM2.5) in the Helsinki
subway system. Atmospheric Environment, 39(28),
5059–5066.
Abt, E., Suh, H. H., Catalano, P., & Koutrakis, P. (2000). Relative contribution of outdoor and indoor particle sources to
indoor concentrations. Environmental Science and Technology, 34(17), 3579–3587.
Alfaro-Moreno, E., Nawrot, T. S., Nemmar, A., & Nemery, B.
(2007). Particulate matter in the environment: pulmonary
and cardiovascular effects. Current Opinion in Pulmonary
Medicine, 13(2), 98–106.
Boldo, E., Medina, S., LeTertre, A., Hurley, F., Mücke, H. G.,
Ballester, F., & Aguilera, I. (2006). Apheis: Health impact
assessment of long-term exposure to PM2.5 in 23 European
cities. European Journal of Epidemiology, 21(6), 449–458.
Brochu, P., Brodeur, J., & Krishnan, K. (2011). Derivation of
physiological inhalation rates in children, adults, and
elderly based on nighttime and daytime respiratory
parameters. Inhalation Toxicology, 23(2), 74–94.
Chan, L. Y., Lau, W. L., Lee, S. C., & Chan, C. Y. (2002a).
Commuter exposure to particulate matter in public transportation modes in Hong Kong. Atmospheric Environment,
36(21), 3363–3373.
Environ Geochem Health
Chan, L. Y., Lau, W. L., Zou, S. C., Cao, Z. X., & Lai, S. C.
(2002b). Exposure level of carbon monoxide and respirable
suspended particulate in public transportation modes while
commuting in urban area of Guangzhou, China. Atmospheric Environment, 36(38), 5831–5840.
Cheng, Y. H., Lin, Y. L., & Liu, C. C. (2008). Levels of PM10
and PM2.5 in Taipei rapid transit system. Atmospheric
Environment, 42, 7242–7249.
Cheng, Y. H., Liu, C. C., & Lin, Y. L. (2009). Levels of ultrafine
particles in the Taipei rapid transit system. Transportation
Research Part D: Transport and Environment, 14(7),
479–486.
Cheng, Y. H., Liu, Z. S., & Yan, J. W. (2012). Comparisons of
PM10, PM2.5, particle number and CO2 levels inside Metro
trains traveling in underground tunnels and on elevated
tracks. Aerosol and Air Quality Research, 12(5), 879–891.
Cheng, Y. H., & Yan, J. W. (2011). Comparisons of particulate
matter, CO, and CO2 levels in underground and groundlevel stations in the Taipei mass rapid transit system. Atmospheric Environment, 45(28), 4882–4891.
CSDHK (Census and Statistics Department of Hong Kong).
(2015). Women and Men in Hong Kong: Key Statistics (Statistical
Reports). http://www.statistics.gov.hk/
pub/B11303032015AN15B0100.pdf.
Fang, G. C., & Chang, S. C. (2010). Atmospheric particulate
(PM10 and PM2.5) mass concentration and seasonal variation study in the Taiwan area during 2000–2008. Atmospheric Research, 98(2–4), 368–377.
Gómez-Perales, J. E., Colvile, R. N., Nieuwenhuijsen, M. J.,
Fernández-Bremauntz, A., Gutiérrez-Avedoy, V. J.,
Páramo-Figueroa, V. H., et al. (2004). Commuters’ exposure to PM2.5, CO, and benzene in public transport in the
metropolitan area of Mexico City. Atmospheric Environment, 38(8), 1219–1229.
Gustafsson, M., Blomqvist, G., Swietlicki, E., Dahl, A., &
Gudmundsson, A. (2012). Inhalable railroad particles at
ground level and subterranean stations-Physical and
chemical properties and relation to train traffic. Transportation Research Part D, 17(3), 277–285.
Halonen, J. I., Lanki, T., Yli-Tuomi, T., Tiittanen, P., Kulmala,
M., & Pekkanen, J. (2009). Particulate air pollution and
acute cardiorespiratory hospital admissions and mortality
among the elderly. Epidemiology, 20(1), 143–153.
Harrison, R. M., Thornton, C. A., Lawrence, R. G., Mark, D.,
Kinnersley, R. P., & Ayres, J. G. (2002). Personal exposure
monitoring of particulate matter, nitrogen dioxide, and
carbon monoxide, including susceptible groups. Occupational and Environmental Medicine, 59(10), 671–679.
Havas, M., Shum, S., Dhalla, R. (2004). Passenger exposure
to magnetic fields on go trains and on buses, streetcars,
and subways run by the Toronto transit commission,
Toronto, Canada, Biological Effects of EMFs, 3rd
International Workshop, Kos, Greece 4–8 October 2004,
1065–1071.
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. Atmospheric Environment, 43(36), 5723–5730.
Huang, H. L., & Hsu, D. J. (2009). Exposure levels of particulate
matter in long-distance buses in Taiwan. Indoor Air, 19(3),
234–242.
Jantunen, M., Hänninen, O., Koistinen, K., & Hashim, J. H.
(2002). Fine PM measurements: Personal and indoor air
monitoring. Chemosphere, 49(9), 993–1007.
Jung, H. J., Kim, B. W., Ryu, J. Y., Maskey, S., Kim, J. C., Sohn,
J., & Ro, C. U. (2010). Source identification of particulate
matter collected at underground subway stations in Seoul,
Korea using quantitative single-particle analysis. Atmospheric Environment, 44(19), 2287–2293.
Kam, W., Cheung, K., Daher, N., & Sioutas, C. (2011). Particulate matter (PM) concentrations in underground and
ground-level rail systems of the Los Angeles Metro. Atmospheric Environment, 45(8), 1506–1516.
Kim, K. Y., Kim, Y. S., Roh, Y. M., Lee, C. M., & Kim, C. N.
(2008). Spatial distribution of particulate matter (PM10 and
PM2.5) in Seoul metropolitan subway stations. Journal of
Hazardous Materials, 154(1–3), 440–443.
Klepczyńska-Nyström, A., Larsson, B. M., Grunewald, J.,
Pousette, C., Lundin, A., Eklund, A., & Svartengren, M.
(2012). Health effects of a subway environment in mild
asthmatic volunteers. Respiratory Medicine, 106(1),
25–33.
Knibbs, L. D., & deDear, R. J. (2010). Exposure to ultrafine
particles and PM2.5 in four Sydney transport modes. Atmospheric Environment, 44(26), 3224–3227.
Kwon, S. B., Cho, Y., Park, D., & Park, E. Y. (2008). Study on
the indoor air quality of Seoul metropolitan subway during
the rush hour. Indoor and Built Environment, 17(4),
361–369.
Lau, W. L., & Chan, L. Y. (2003). Commuter exposure to aromatic VOCs in public transportation modes in Hong Kong.
Science of the Total Environment, 308(1–3), 143–155.
Li, T. T., Bai, Y. H., Liu, Z. R., & Li, J. L. (2007). In-train air
quality assessment of the railway transit system in Beijing:
A note. Transportation Research Part D, 12(1), 64–67.
Li, T. T., Bai, Y. H., Liu, Z. R., Liu, J. F., Zhang, G. S., & Li, J.
L. (2006). Air quality in passenger cars of the ground
railway transit system in Beijing, China. Science of the
Total Environment, 367(1), 89–95.
Linn, W. S., Szlachcic, Y., Gong, H., Kinney, P. L., & Berhane,
K. T. (2000). Air pollution and daily hospital admissions in
metropolitan Los Angeles. Environmental Health Perspectives, 108(5), 427–434.
Long, C. M., Suh, H. H., & Koutrakis, P. (2000). Characterization of indoor particle sources using continuous mass and
size monitors. Journal of the Air and Waste Management
Association, 50(7), 1236–1250.
Matson, U., Ekberg, L. E., & Afshari, A. (2004). Measurement
of ultrafine particles: A comparison of two handheld condensation particle counters. Aerosol Science and Technology, 38(5), 487–495.
McNamara, M. L., Noonan, C. W., & Ward, T. J. (2011). Correction factor for continuous monitoring of wood smoke
fine particulate matter. Aerosol and Air Quality Research,
11(3), 315–322.
Meng, Q. Y., Turpin, B. J., Korn, L., Weisel, C. P., Morandi, M.,
Colome, S., et al. (2005). Influence of ambient (outdoor)
sources on residential indoor and personal PM2.5 concentrations: analyses of RIOPA data. Journal of Exposure
Analysis and Environmental Epidemiology, 15, 17–28.
Morawska, L., Keogh, D. U., Thomas, S. B., & Mengersen, K.
(2008). Modality in ambient particle size distributions and
123
Environ Geochem Health
its potential as a basis for developing air quality regulation.
Atmospheric Environment, 42(7), 1617–1628.
MTR Corporation Limited. (2014). http://www.mtr.com.hk/
eng/overview/profile_index.html.
Mugica-Álvarez, V., Figueroa-Lara, J., Romero-Romo, M.,
Sepúlveda-Sánchez, J., & López-Moreno, T. (2012).
Concentrations and properties of airborne particles in the
Mexico City subway system. Atmospheric Environment,
49, 284–293.
Nasir, Z. A., & Colbeck, I. (2009). Particulate air pollution in
transport micro-environments. Journal of Environmental
Monitoring, 11(6), 1140–1146.
Oberdörster, G., Sharp, Z., Atudorei, V., Elder, A., Gelein, R.,
Lunts, A., et al. (2002). Extrapulmonary translocation of
ultrafine carbon particles following whole-body inhalation
exposure of rats. Journal of Toxicology and Environmental
Health Part A, 65(20), 1531–1543.
Park, D. U., & Ha, K. C. (2008). Characteristics of PM10, PM2.5,
CO2 and CO monitored in interiors and platforms of subway train in Seoul, Korea. Environment International,
34(5), 629–634.
Pope III, C. A., Burnett, R. T., Thurston, G. D., Thun, M. J.,
Calle, E. E., Krewski, D., & Godleski, J. J. (2004). Cardiovascular mortality and long-term exposure to particulate air pollution: epidemiological evidence of general
pathophysiological pathways of disease. Circulation, 109,
171–177.
Querol, X., Moreno, T., Karanasiou, A., Reche, C., Alastuey, A.,
Viana, M., et al. (2012). Variability of levels and composition of PM10 and PM2.5 in the Barcelona metro system.
Atmospheric Chemistry and Physics, 12(11), 5055–5076.
123
Raut, J. C., Chazette, P., & Fortain, A. (2009). Link between
aerosol optical, microphysical and chemical measurements
in an underground railway station in Paris. Atmospheric
Environment, 43(4), 860–868.
Seaton, A., Cherrie, J., Dennekamp, M., Donaldson, K., Hurley,
J. F., & Tran, C. L. (2005). The London Underground: Dust
and hazards to health. Occupational and Environmental
Medicine, 62(6), 355–362.
Son, Y. S., Salama, A., Jeong, H. S., Kim, S., Jeong, J. H., Lee,
J., et al. (2013). The effect of platform screen doors on
PM10 levels in a subway station and a trial to reduce PM10
in tunnels. Asian Journal of Atmospheric Environment,
7(1), 38–47.
Tsai, D. H., Wu, Y. H., & Chan, C. C. (2008). Comparisons of
commuter’s exposure to particulate matters while using
different transportation modes. Science of the Total Environment, 405(1–3), 71–77.
USEPA. (2004). Air quality criteria for particulate matter (Final Report, Oct 2004), EPA 600/P-99/002aF-bF. Washington, DC: US Environmental Protection Agency.
USEPA. (2005). Guidelines for carcinogen risk assessment,
EPA/630/P-03/001F. Washington, DC: US Environmental
Protection Agency.
Wang, X., & Gao, H. O. (2011). Exposure to fine particle mass
and number concentrations in urban transportation environments of New York City. Transportation Research Part
D, 16(5), 384–391.
Yanosky, J. D., Williams, P. L., & Maclntosh, D. L. (2002). A
comparison of two direct-reading aerosol monitors with the
federal reference method for PM2.5 in indoor air. Atmospheric Environment, 36(1), 107–113.