Journal Archive

As results of PM measurement in Seoul subway lines, which are mostly consisted of underground tunnels, the daily average PM₁₀, PM_2.5, and PM₁ concentrations for the entire sections were 101.3±38.4, 81.5±30.2, and 59.7±19.9 μg/m³, respectively. As shown in Table 3, the daily average PM₁₀, PM_2.5 and PM₁ concentrations in winter were 106.7±40.1, 86.6±32.6, 64.2±20.9 μg/m³, respectively; however, in summer were 98.0±37.4, 78.4±28.7, 56.9±19.2 μg/m³, respectively. The PM levels in winter were about 1.2 times higher than those in summer. Especially during the summer period, the highest average concentrations of PM₁₀ and PM_2.5 were observed respectively by 136.0 μg/m³ and 101.7 μg/m³ in the Line-1 and further that of PM₁ was 77.9 μg/m³ in the Line-9. When comparing among subway lines, the seasonal pattern showed that the behavior of Line-2 was quite different from other lines. In other words, the range of average concentration was widest between two seasons since very low levels were observed in summer, but very high levels in winter. Generally in the ambient air, the PM concentration in Korea is highest during the winter heating season, but lowest during the summer rainy season. Thus the subway Line-2 might be easily influenced by the ambient air when comparing other lines since the Line-2 has 11 above-ground sections out of total 44 sections as shown in Table 1. On the other hands, seasonal differences in terms of PM levels were very small since other lines have a few or none ground sections.

The Korean MOE has established the PM₁₀ indoor air quality (IAQ) standard of 150 μg/m³ for subway platform, concourse, and passage way, while the standard for tunnel air does not exist yet (Ministry of Environment, 2014). According to the IAQ recommendation for urban transit units, PM₁₀ concentration in train cabins should not be exceeded 200μg/m³. Furthermore, the Korean ambient air quality standards (AAQS) of 24-hr basis PM₁₀ and PM_2.5 were set by 100 μg/m³ and 50 μg/m³, respectively.

Fig. 3 showed seasonal cumulative distributions of PM concentrations monitored on the Line-2, 3, 6, 8 and 9. The distributions were obtained by integrating total concentration range with respect to the frequency distribution of the PM concentration in a certain section. The cumulative distribution analysis was performed using an Excel program. The respective median concentrations of PM₁₀ on the Line-2, 3, 6, 8 and 9 were 122.6, 76.7, 91.1, 89.6, and 109.7 μg/m³ in winter [see Fig. 3(a)] and 66.2, 76.3, 72.6, 82.5, and 110.0 μg/m³ in summer [see Fig. 3(b)]. In addition, the respective median concentrations of PM_2.5 on the Line-2, 3, 6, 8 and 9 were 99.1, 58.3, 78.4, 85.2, and 87.8 μg/m³ in winter [see Fig. 3(c)] and 57.6, 61.1, 61.3, 73.9, and 97.0 μg/m³ in summer [see Fig. 3(d)].

Fig. 3. 
Seasonal cumulative distributions of PM_2.5 and PM₁₀ concentrations monitored on the selected subway lines in tunnels, where the solid line A in figure boxes indicates the Korean IAQ standard of PM₁₀ (150 μg/m³), the solid line B as the urban transit units’ recommendation of PM₁₀ (200 μg/m³), and the solid line C as the Korean AAQS of PM_2.5 (50 μg/m³ for 24-hr basis).

The PM₁₀ distribution in Fig. 3(a) and (b) showed that 36% of Line-2 did not meet the IAQ standard of 150 μg/m³ in winter, while 30% of Line-6 did not meet the standard in summer. In addition, the PM_2.5 distributions in Fig. 3(c) and (d) showed that 99% of Line-8 did not meet the Korean AAQS of 50 μg/m³ in winter, while 95% of Line-2 as well as Line-8 did not meet the standard in summer.

Particle size as one of physical parameters is important to determine the properties, effects, and fate of airborne particles. Also aerosol deposition rate as well as residence time directly affecting human health and wealth are strongly related with aerodynamic diameter (EPA, 1999). To examine the aerosol behaviors and its possible sources in the subway environment, we initially calculated the size-oriented PM mass ratios. Table 4 shows the average PM mass ratios for PM_2.5/PM₁₀, PM₁/PM₁₀, and PM₁/PM_2.5 which were monitored on all the ground and underground sections on the selected Seoul subway lines.

When comparing results among subway sections, the average mass ratio of PM_2.5/PM₁₀ was calculated by 0.78 on ground and 0.82 on underground sections in winter and that was also calculated by 0.84 on ground and 0.86 on underground sections in summer. The range of mass ratios for PM_2.5/PM₁₀ was 0.67-0.78 in Taipei subway stations (Cheng et al., 2008), 0.79 in Guangzhou (Chan et al., 2002a), 0.72-0.78 in Hong Kong subway stations (Chan et al., 2002b) and 0.76 on the ground stations and 0.73 on underground stations in Los Angeles (Kam et al., 2011). Further the respective mass ratios of PM₁/PM₁₀ and PM_2.5/PM₁₀ were 0.49-1.00 and 0.56-1.00 in a Shanghai subway tunnel (Qiao et al., 2015a). The result of LA was relatively close to our results obtained on the Line-4 during summer survey. It might be natural that the mass ratios of PM₁/PM₁₀ and/or PM_2.5/PM₁₀ were observed to be somewhat different in every subway environment due to various control facilities, power and brake systems, ventilation systems, and operation conditions. Above all things, the in/out air quality near each station should be a significant influence factor on the PM mass ratios.

There are many internal and external PM sources worsen the subway air quality. Internal PM is mainly produced from mechanical wear during the period of trains running or braking, maintenance activities as well as construction works in tunnels, etc. (Kim et al., 2008). It is also noted that most of the internal PM is considered as coarse particles. According to a research paper (Qiao et al., 2015a), the mass ratios of PM₁/PM₁₀ and PM_2.5/PM₁₀ were higher when subway trains went out of service since the emission activities generating coarse particles were completely ceased. On the other hands, external PM is penetrated from various outdoor sources like traffic emissions surrounding the subway (Qiao et al., 2015b), where the PM penetration into the subway is significantly assisted by mechanical and natural type ventilations. It seems that the amount of ventilation presumably affects the PM mass ratios on the Seoul subway lines. During our study period, the amount of subway ventilation was usually decreased to keep warm air temperature from inflowing cold ground air during winter time; however, that was increased to reduce hot air temperature and to enhance air quality by inflowing fresh outdoor air during the summer time. Thus it seems that internal emissions tend to be accumulated in subway tunnels during winter, while external emissions tend to easily flow into the tunnels during the other seasons.

In our study, it was observed that the fine proportions (i.e. the mass ratios of PM₁/PM₁₀ and PM_2.5/PM₁₀) were generally higher in summer than those in winter and also each fine proportion was elevated on ground sections in summer as well as on underground sections in winter. For example, the respective PM₁/PM₁₀ ratios on underground and ground sections were 0.61 and 0.62 in winter and 0.67 and 0.70 in summer. It is presumably because the fine secondary aerosols, mostly PM₁, are generated by photochemical reaction during summer. It is also worth noting in this study that the PM₁/PM_2.5 ratio was always higher on ground sections; however, the PM_2.5/PM₁₀ ratio was always higher on underground sections. It is partly because a proportion of fine particles including PM₁ and PM_2.5 was increased on ground sections due to various anthropogenic emissions such as on-road mobile source, fossil fuel combustion source, secondary aerosol source, and so on. In an earlier study, the mass ratio of PM_2.5/PM₁₀ for on-road vehicles was almost 1 (Jin et al., 2012).

It is necessary to figure out the behaviors of internal and external emission sources in order to improve subway air quality. According to a result from PMF (positive matrix factorization) receptor modeling study in Seoul subway tunnels, a total of 3 sources was identified and then average contribution to PM₁₀ mass was determined by 56.9% from brake wear-related source, 13.8% from iron-related source, and 17.9% from secondary aerosol sources (Park, 2013). Since a mass proportion of fine particle is considered to be significant in subway tunnels, it is needed to extend receptor modeling study to determine quantitative contribution to each PM₁ and PM_2.5 mass from various emission sources.

Fig. 4 shows mass concentrations of PM₁, PM_2.5, and PM₁₀ measured on the subway Line-2, 3, 6, 8, and 9 during the study periods. On the basis of PM₁₀ mass concentration, the sum of fine mass proportions including PM₁ and PM_2.5 occupied a minimum of 0.75 in winter on the Line-3 and a maximum of 0.88 in summer on the Line-8, while the range of PM₁ mass proportion occupied 0.51-0.71 in PM₁₀.

Fig. 4. 
Proportions of PM mass concentration observed during winter and summer on the Seoul subway lines.

The MOE has been operating automatic monitoring networks to monitor criteria air pollutants including PM_2.5 and PM₁₀ on 259 monitoring sites nationwide including 27 sites in Seoul City (NIER, 2016). For our study on comparing between subway tunnel air and outdoor air, we obtained ambient PM_2.5 as well as PM₁₀ hour-data from the MOE. Initially we had to select each MOE monitoring site locating near the subway lines, and then we synchronized ambient MOE data with our subway data one by one basis. Fig. 5 shows a comparative result studied on both tunnel air and outdoor air for each subway line. The mass ratio range for PM_2.5/PM₁₀ in tunnel air was 0.76-0.89, while that in outdoor air was 0.48-0.68. In general, the ratio of PM_2.5/PM₁₀ was much higher in tunnel air than that in outdoor air. Especially the ratio of PM_2.5/PM₁₀ in outdoor air was always higher in summer than that in winter.

Fig. 5. 
The PM_2.5/PM₁₀ mass ratios in subway tunnel air and in outdoor air near each tunnel.

Fig. 6 shows the average proportions of PM₁, PM_2.5, and PM₁₀ mass observed in tunnel sections, passenger cabins and outdoors on the Line-8. The outdoor PM_2.5 and PM₁₀ data were obtained from the MOE as mentioned earlier. The PM_2.5/PM₁₀ ratio in cabin air was observed to be slightly higher than that in tunnel, but the ratio was much higher than in outdoor air. It seemed from the result that fine particles generated in tunnels intruded into cabins after PSD installation more significantly than before so that tunnel air directly affected cabin air. Thus it is necessary to improve ventilation systems to block fine particle inflows from tunnels or to enhance tunnel air quality by proper controls.

Fig. 6. 
Proportions of PM mass concentration monitored on tunnel, cabin, and near the area of the Seoul subway line-8 in summer.