• E-mail
• Print
• Comment
• Font Size
• Digg
• del.icio.us
• Discuss article

Use of Stationary and Mobile Measurements to Study Power Plant Emissions

Posted on: Sunday, 26 February 2006, 03:02 CST

By Yao, Xiaohong; Lau, Ngai T; Fang, Ming; Chan, Chak K

ABSTRACT

This paper presents a technique to study air pollution by combining high spatial resolution data obtained by a mobile platform and those measured by conventional stationary stations. Conventional stations provide time-series point data but cannot yield information that is distant from the sites. This can be complemented or supplemented by mobile measurements in the vicinity of the conventional sites. Together, the combined dataset yields a clearer and more precise picture of the dispersion and the transformation of pollutants in the atmosphere in a fixed time frame. Several experiments were conducted in the years 2002-2003 to track the impact of power plant plumes on ground receptors in the immediate vicinity (within a radius of 30 km) of the plants, using a combined mobile and stationary dataset. The mobile data allowed the identification of emissions from coal-fired and gas-fired power plants. Coal-fired power plants were the major source of sulfur dioxide (SO^sub 2^), whereas nitrogen oxides (NO^sub x^) emitted from the gas-fired power plant played an important role in the formation of ozone (O^sub 3^) at ground level. The mobile data showed that two particle size distribution regimes were detected: one had a dominant accumulation mode at 0.40-0.65 m and the other at 0.65-1 m. The existence of particles characterized by their mode at 0.65-1 m and formed by in-cloud processes suggests that vehicular emissions were not the important source. Other local sources, such as power plants (elevated emission), were the likely sources, because Hong Kong does not have much manufacturing industry.

INTRODUCTION

Conventional stationary air monitoring stations have served atmospheric researchers well for decades but can no longer satisfy the ever-demanding needs to learn more about air pollution problems, because conventional stations can only provide point information. This disadvantage can be solved by installing more and closer stations (network). There are many economic, facility, and operational problems associated with such a network. An alternative solution is needed. Mobile laboratories have been reported to reveal new issues on urban air quality.1,2 These laboratories can be modified to make two-dimensional measurements. Undoubtedly, two- dimensional continuous ground-level data, if available, will provide much more insight to the microstructures (physics, chemistry, movement, and transformation of air pollutants) of the atmosphere. Such a mobile platform can also be used to track pollution sources, and the dynamics of the dispersion of the pollutants can be revealed. It can be easily used for cross-border studies, and it can yield data at difficult-to-access yet very important places, such as tunnels and large traffic exchanges. It is clear that the combined stationary and mobile measurements should provide a more complete dataset to allow researchers to better understand the chemistry and physics of pollutants in the atmosphere.

Despite undertaking a series of air pollution control measures since 1990 in Hong Kong, after some initial improvement in air quality, more and more low-visibility days have been experienced in recent years. These measures include low-sulfur diesel fuel, no- lead gasoline, installation of three-way catalytic converters, and the conversion of diesel taxis and vans to liquefied petroleum gas engines. What and where are the sources of these air pollution problems? Can the power plants give a significant contribution to local air quality? Power plant plumes have not been considered to be a source of local pollution because of their tall stacks (215-250 m). In this paper, the authors report a series of experiments in which the plumes of local power plants were tracked by analyzing a combination of conventional stationary data and data measured by a mobile platform in an attempt to answer these questions. Data under specific wind conditions, when most of the territory was downwind of the power plants, were analyzed to study the impact of their plumes on ground receptors.

Monitoring Sites and Data Sources

The stationary air dataset used in this study is provided by the Hong Kong Environmental Protection Department (HKEPD), which operates a network of 14 air quality monitoring stations, including three roadside stations, to measure RSP (defined as particles ≤10 m, including total mass, metals, elemental carbon and organic carbon, and ionic concentrations), sulfur dioxide (SO^sub 2^), nitric oxide (NO^sub 2^), ozone (O^sub 3^), and carbon monoxide (CO).3 Mean hourly RSP, SO^sub 2^, NO^sub 2^, and O^sub 3^ concentrations at six general HKEPD air monitoring stations for the years 2002 and 2003 during episodes were used in this study. HKEPD follows U.S. Environmental Protection Agency protocol in quality control and assurance in their air monitoring data and reports a 10% accuracy and precision for their air quality monitoring network.

Figure 1 is a map showing the locations of the HKEPD air monitoring stations. The Tap Mun (TM) station is located on a remote island in northeast Hong Kong and is very close to the border. This site is designed to reflect marine background and transported air plumes from outside of the territory. The Tung Chung station (TC) is located on the opposite side of Hong Kong on the southwestern island of Lantau near the airport. The bulk of the emission sources, mostly vehicular, are located in Kowloon Peninsula and the northern part of Hong Kong island, roughly halfway between these two stations. The rest of the stations are located in the various parts of the territory.

Electric low-pressure impactor (ELPI) measurements were continuously conducted by the mobile real-time air monitoring platform (MAP), developed and built by the Hong Kong University of Science and Technology (HKUST), as it drove on the highways and urban streets. A comprehensive comparison of fine particulate (PM^sub 2.5^) measured by ELPI and tapered element oscillating microbalance (TEOM; R&P) has been conducted in different sites in the United States, and a good correlation between the two measurements was obtained.4 PM^sub 2.5^ concentrations measured by ELPI are higher than TEOM, because ELPI measures ambient particles, and TEOM measures dried particles. Detailed information on MAP has been reported elsewhere.5 The instruments used on MAP are briefly described as follows: (1) the platform was a gasoline-powered 4.6-t Euro III van; (2) an isokinetic unidirectional probe based on Liu and Pui6 and tested in a wind tunnel was used for the sampling of aerosols at speeds up to 50 km/hr^sup -1^; (3) a global positioning system and an on-board geographical information system were used for the real-time display of the collected data; (4) an automatic weather station provided instant ambient meteorological data; (5) an ELPI (Dekati, Ltd.) was used to measure particle size distribution (the data rate was one 12-cut scan per second); (6) an aethalometer (Magee Scientific) was used for black carbon (BC) analysis; and (7) gas analyzers by API, Inc., were used for CO, SO^sub 2^, nitrogen oxides (NO^sub x^), and O^sub 3^ detection (10% accuracy and precision was for gas analyzers and ELPI measurements, whereas 15% was for the aethalometer).

Figure 1. Location of HKEPD air quality monitoring stations, power plants, and incinerator. HKEPD monitoring sites: 1. Tap Mun, 2. Yeun Long, 3. TsiPo, 4. Tseum Wan, 5. Kwai Chung, 6. Shatin, 7. Shamshipo, 8. Mong Kok (roadside), 9. Kwun long, 10. Central/ Western, 11. Central (roadside), 12. Causeway Bay (roadside), 13. Eastern, 14. Tung Chung, 15. Hok Tsui. Powerplants and incinerator: a. CLP (Black Point), b. CLP (Castle Peak), c. Hazardous Waste Incineration plant, d. CLP (Penny's Bay), e. HEC. HKIA = Hong Kong International Airport.

Also shown in Figure 1 are two large power plants (Figure 1 locations a and b; CLP) located on the west side of Kowloon and a small oil-fired plant at Penny's Bay (Figure 1 location d; PB) on Lantau Island, all owned by CLP Power Hong Kong Limited. CLPa at Black Point is gas-fired, whereas a few kilometers south at Castle Peak is the coal-fired plant (CLPb). The stack temperature and velocity of the two larger plants are 110 C and 17 m/sec. On Lamma Island in the south is the coal-fired power plant of Hong Kong Electric (Figure 1 location e; HEC). The stack temperature and velocity are ~110-120 C and 15 m/sec. The exhaust from these stacks (215 and 250 m in height) can reach 500-600 m in altitude, depending on the ambient temperature and wind speed.7,8 There is a hazardous waste incinerator (Figure 1 location c; HWI) on Tsing Yi Island. Most of the manufacturing industry in Hong Kong moved across the border since the Chinese Mainland opened up for economic development almost 30 years ago. Power plants and vehicular emission are the two major local sources of air pollutants in Hong Kong.9 Power plants, on-road motor vehicles, marine traffic, airport, and other area sources account for 57, 20, 16, 3, and 3% of the NO^sub x^ emitted, respectively, and for SO^sub 2^, they account for 92, 0.2, 4, 0.2, and 4%, respectively. SO^sub 2^ emitted \from power plants overwhelms other sources in Hong Kong. Marine traffic affects mainly Kwai Chung (the container port) and exerts a minor influence on other stations.5

Tracking SO^sub 2^ Sources Using Stationary and Mobile Data

Analysis of Stationary Data. Data obtained on August 31, 2002, and January 13, 2003, were used to demonstrate the usefulness of the technique in tracking power plant plumes. On August 31, 2002, the prevailing wind at the Hong Kong International Airport (HKIA) was from the ocean (southwest and south), bringing in clean marine air, whereas on January 13, 2003, the dominant wind was from the northwest from ~9:00 a.m. to 4:00 p.m., and the prevailing wind was from southeast at other times, as shown in Figure 2. The meteorological data at HKIA are considered to be indicative of the large-scale background wind in this discussion because the 5th- Generation NCAR/Penn State Mesoscale Model (MMS) simulated surface wind fields are generally consistent with the observed wind data at HKIA.

Figure 2. Wind direction and wind speed: (a) August 31, 2002; (b) January 13, 2003.

The spatial patterns of air pollutants (hourly averages) based on stationary data are shown in Figures 3 and 4. Because of the use of very low-sulfur (<150 ppm for gasoline and <50 ppm for diesel) automotive fuels, local vehicular emissions are not likely to be the influential SO^sub 2^ source. The annual average SO^sub 2^ at Kwai Chung (near the container port) and Central (roadside station in business district) in 2002 was only 11 and 7 ppb, respectively, indicating low vehicular SO^sub 2^ emission. Therefore, any high SO^sub 2^ readings suggest contribution from other sources. SO^sub 2^ peaks were detected at various stations on these days. More interestingly, the SO^sub 2^ peak propagated from station to station in the downwind directions.

The August 31, 2002, data will be used as an illustration. Wind direction (south and southwest) data ruled out the transport of SO^sub 2^ from the north and east. Likewise, the average SO^sub 2^ daily concentration at the northeastern background station (TM) at 9 ppb was lower than the other sites, which also suggests contribution from local sources. The prevailing wind put the power plant on Lamma Island (Figure 1 location e) upwind of many of the HKEPD stations. Table 1 summarizes the average concentrations of the downwind stations according to wind directions. It can be seen that the highest SO^sub 2^ concentration point was either CW or the next station in the path, Shamshuipo (SP). CW, being very close to the power plant, is more sensitive to minor changes in wind direction. The SO^sub 2^ peak appeared at SP at 1:00 a.m. (41 ppb) and at 4:00 a.m. (61 ppb). Before that, SO^sub 2^ at all of the stations on the north was smaller than that at SP. Moreover, SO^sub 2^ at SP was approximately three times that at Central (roadside station, data not shown) before 4:00 a.m., whereas NO^sub 2^ showed a reverse trend. This suggests that vehicle emission was not a major source of SO^sub 2^.

Figure 3. Diurnal variations of air pollutants at various HKEPD stations on August 31, 2002 (RSP in g/m^sup 3^ and gases in ppb. Blanks are data not available.).

Figure 4. Diurnal variations of air pollutants at various HKEPD stations on January 13, 2003 (RSP in g/m^sup 3^ and gases in ppb. Blanks are data not available.).

Figure 5. MAP air pollution measurement track on August 31, 2002: (a) O^sub 3^ concentrations at tracks; (b) SO^sub 2^, NO^sub x^, and NO^sub 2^ + O^sub 3^ concentrations at tracks.

From 8:00 a.m. to 9:00 a.m., the wind direction at the CLP power plant, 10 km north of TC, changed to ~330 at ~2 m/sec. An SO^sub 2^ peak (60 ppb) appeared at TC at 9:00 a.m. Between CLP and TC is mostly flat terrain (sea); therefore, the Gaussian plume equation can be used to estimate SO^sub 2^ concentrations downwind of the CLP coal-fired power plant. At 10 km downwind, it was estimated to be ~120 ppb at the plume center line at ground-level based on the CLP emission parameters. This is within the ballpark value of that measured at TC because the calculated ground-level concentration is dependent on the distance between the receptor and center line.8 The question is to what extent do the plumes affect downwind receptors? A track of SO^sub 2^ concentration data downwind together with stationary site data are needed to gain insight to this problem.

Table 1. Average pollutant concentrations at downwind stations on August 31, 2002.

A similar analysis was also conducted on January 13, 2003 and, likewise, SO^sub 2^ peaks were observed at stations downwind of the power plants. The details of the analysis will not be elaborated here.

Use of MAP Data to Track SO^sub 2^ Sources. MAP measurements taken on the same days were used as an independent dataset to substantiate the contribution of the pollutants by local power plants. The platform originated from the campus of HKUST on the east side of Kowloon in the morning on August 31, 2002. The route took MAP in a clockwise (south-west-north-east) direction on the highways along the perimeter of Kowloon. The route is shown in Figure 5a. Analysis was performed on five transects on the northeastern part of the track, downwind of the power plants. There were no stationary sites in this area. The variations of pollutant concentration along these transects are presented in Table 2 and the bar chart in Figure 5b.

Table 2. Temporal and spatial distributions of pollutants measured at the northeastern track on August 31, 2002.

There was a bell-shaped SO^sub 2^ distribution in the 28-min (<23 km) track. The maximum value of 66 ppb at transect 3 was 11 times higher than that at transect 5 (HKUST campus, rural) at only 5.7 ppb and two times higher than that at Tuen Mun (transect a, urban) at 30 ppb. NO^sub x^ and BC concentrations between transects 1 and 3 were almost constant, suggesting that vehicular contribution at these transects was almost the same. Because the distance between these two transects was very close, and the sampling times were only 9 min apart, any large-scale transport influencing these transects should be similar. Therefore, it can be deduced that the excess SO^sub 2^ observed must be from a source other than vehicular.

Recall that the wind was from the southwest, making the CLP coal- fired power plant on the west side upwind of these transects. Referring to Figure 5a and using the 240 wind direction (in Figure 2a) as a rough reference, the bell-shaped SO^sub 2^ distribution seems to be in the path of the dispersion of a plume from a point source. However, the complex terrain made it impossible to estimate the ground-level concentrations of the power plant plume using the Gaussian plume equation.

Therefore, both stationary and mobile data on August 31, 2002, show that much of Hong Kong was affected by the power plants depending on the wind direction. HEC affected SP, CW, and ST, whereas TC and TP were probably impacted by CLP.

Similar results were obtained by analyzing MAP data taken on a winter day (January 13, 2003). As shown in Table 3, the upwind transects had SO^sub 2^ concentrations lower than the downwind transects, and this could be because of the intrusion of the CLP power plant plumes. The SO^sub 2^ concentrations were not as large as August 31, 2002, because of the strong interaction of cloud processes with the power plant plumes. This will be discussed later.

Table 3. Temporal and spatial distributions of pollutants measured at tracks upwind and downwind of power plants on January 13, 2003.

Photochemical Formation of O^sub 3^ Associated with Power Plant Plume

Analysis of Stationary Data. The highest O^sub 3^ concentration usually occurred at either TM (background) or TC at the two opposite ends of the territory, and the highest average concentration on August 31, 2002, was at TM (Table 1). The maximum hourly average value at 131 ppb was two to three times the background concentration (30-50 ppb) in East Asia.10 This suggests significant anthropogenic O^sub 3^ production. Low O^sub 3^ was observed at the urban sites because of the titration of O^sub 3^ by nitrous oxide (NO) according to the NO + O^sub 3^ [arrow right] NO^sub 2^ reaction. In the absence of volatile organic compounds or solar radiation, such as at night, the titration reaction conserves the sum of O^sub 3^ and NO^sub 2^ in the atmosphere, although the O^sub 3^ concentration varies. For this particular set of data, this sum was spatially uniform at several stations before 6:00 2. m.. The sum at CW was close to SP and at ST was close to TP; however, the former two stations were higher than the latter two. This difference could be caused by the NO^sub 2^ emitted directly by vehicles, and it can be significant in urban areas.11 After 6:00 a.m., the O^sub 3^ and (NO^sub 2^ + O^sub 3^) at all of the stations rapidly increased probably because of the photochemical formation of O^sub 3^ and/or an increase in the mixing depth of the boundary layer together with the transport of O^sub 3^-rich air mass from aloft to the ground.12 The small spatial variation of (NO^sub 2^ + O^sub 3^) observed during 6:00 a.m. to 11:00 a.m. suggests the transport of O^sub 3^- rich air mass from aloft to the ground. From 12:00 p.m. to 6:00 p.m., photochemical formation of O^sub 3^ became dominant, leading to a large O^sub 3^ peak and a large spatial variation in (NO^sub 2^ + O^sub 3^). Two peaks of O^sub 3^ and NO^sub 2^+O^sub 3^ were observed at TM and TC from 12:00 p.m. to 6:00 p.m., and NO^sub 2^ + O^sub 3^ at these two sites was higher than the other urban sites located in between them. This suggests contribution from intermittent point sources rather than area sources. The analysis in the previous section on SO^sub 2^ suggests that the plume from the power plant on Lamma Island could affect these urban stations. Why was NO^sub 2^ + O^sub 3^ at urban stations lower than those a\t TC and TM? O^sub 3^ and NO^sub 2^ + O^sub 3^ data obtained by MAP in the gap between the stations will be used to answer this question later.

Figure 6. Time-series particle concentrations (in m^sup 3^/ cm^sup 3^) and correlations of particles with NO^sub x^ and black carbon (BC): (a) time-series of particle concentrations; (b) correlations of particles with NO^sub x^; (c) correlations of particles with BC.

Similar analysis was conducted on January 13, 2003, when the highest O^sub 3^ (~79 ppb) was at 5:00 p.m. at TC and TM (Figure 4). The NO^sub 2^ at TC (97 ppb) was much higher than that at TM (5 ppb). At 5:00 p.m., the highest NO^sub 2^ + O^sub 3^ (173 ppb) was at TC, whereas the lowest NO^sub 2^ + O^sub 3^ (80 ppb) was at TM. Vehicles emit mostly NO and little NO^sub 2^.11 If only NO was considered, NO lowered O^sub 3^ concentration by reacting with O^sub 3^ to form NO^sub 2^, but NO^sub 2^ + O^sub 3^ was conserved. A small amount of direct NO^sub 2^ emissions increased NO^sub 2^ + O^sub 3^ where the vehicles were, that is, at the urban sites. This has been observed at most urban sites in this study. Therefore, vehicular emission cannot be responsible for TC having high O^sub 3^ and NO^sub 2^ + O^sub 3^ at the same time. An additional source(s) is needed. The intrusion of power plant plumes into vehicular emissions could explain this, because the NO + O^sub 2^ [arrow right] NO^sub 2^ reaction is important in concentrated power plant plumes according to Seinfeld and Pandis8 and Luria et al.13

Use of MAP Data to Track Precursors of O^sub 3^. O^sub 3^ and NO^sub 2^ + O^sub 3^ for the transects on August 31, 2002, can be found in Figure 5b. As demonstrated in the SO^sub 2^ analysis, the northeastern transects were under the influence of the power plants. The skewed NO^sub 2^ + O^sub 3^ distribution at these transects reveals another story. The CLP Black Point Plant on the north is gas fired, whereas the Castle Peak Plant, which is on the south of CLP, is coal fired. Under a prevailing southwest wind, theoretically, when these plumes eventually reached ground level, ground receptors on the north would experience high NO^sub 2^ + O^sub 3^, whereas those further south would have high SO^sub 2^. The NO^sub 2^ + O^sub 3^ distribution at the northeastern transects seems to the fit this. The directly emitted NO^sub 2^ from vehicles probably played only a minor role in transect 1. Although the total NO^sub x^ at transect 2 (720 ppb) was higher than that at transect 1 (NO^sub x^ = 400 ppb), the NO^sub 2^ + O^sub 3^ at transect 2 (182 ppb) was lower than that at transect 1 (215 ppb). It seems that the intrusion of the gas- fired power plant plume at ground level increased NO^sub 2^ + O^sub 3^. The increase of NO^sub 2^ + O^sub 3^ at these transects and the increase in NO^sub 2^ + O^sub 3^ at the northeastern background station (TM) may be because of the same mechanism, because both were downwind of the gas-fired power plant plume. Kok et al.7 observed ~50 ppb of SO^sub 2^ at 610-m altitude close to the these power plants using flight measurements in Hong Kong, and they also found that SO^sub 2^ rapidly decreased to <5 ppb after the flight was ~4 km away from the power plant plumes. They also observed high O^sub 3^ (~100 ppb) at 610-m altitude, and Luria et al.13 reported high O^sub 3^ (~100 ppb) at ~20 km downwind of power plants in Tennessee. A similar analysis using MAP data on January 13, 2003, was conducted, and both higher O^sub 3^ and higher NO^sub 2^ + O^sub 3^ at TC (transect E) were found to be from the gas-fired CLP power plant (see Table 3).

Figure 7. Size distributions of atmospheric aerosols (in dV/dLog Dp, m^sup 3^/cm^sup 3^): (a) size distribution on January 13, 2003; (b) size distribution on August 31, 2002 (ambient air is defined by when the measured NO^sub x^ was <50 ppb).

Identification of Power Plant Plumes Using MAP Particle Size Distribution Data

Now that it has been shown that power plant plumes can influence ground receptors in their proximity, the presence of secondary aerosols at these receptors would provide additional support to the these observations and analyses. In particular, the authors are interested in the particles formed by cloud processes. Cloud processes are known to accelerate the transformation of primary pollutants to secondary ones. When power plant plumes pass through clouds, SO^sub 2^ in the plumes decreases rapidly, while particles increase. The accumulation mode at 0.7 m has been used as a characteristic in identifying in-cloud formation atmospheric particles.14-16 Particle size distribution data obtained by MAP were used to provide evidence for the presence of power plant plumes in an urban plume. On January 13, 2003, the accumulation mode dominated at 0.65-1 m, whereas the dominant mode was at 0.40-0.65 m on August 31, 2002, as shown in Figure 6a and Figures 7a and b. Moreover, as shown in Figures 6b and c, poor correlations of particles at 0.65-1 m with NO^sub x^ and BC on January 13 suggested that the primary emissions associated with combustion processes were likely not the major sources of the particles, and secondary formation of the particles was probably important. On the other hand, the fair correlations of particles at 0.17-0.26 m with NO^sub x^ and BC on January 13 suggested that the particles probably originated from the primary sources, such as vehicle emissions. Moreover, the particle modes of crustal and sea-salt species in Hong Kong distribute at the supermicron particle size.17 The presence of the dominant 0.65-1 m particle mode at the transects on January 13, 2003, suggests that in- cloud processes had occurred. Cloud coverage varied from 20 to 40% from 10:00 a.m. to 4:00 p.m. on that day (from Hong Kong Observatory). Low clouds (<500 m) frequently occur in Hong Kong18 because of the hilly terrain and humid tropical coastal atmosphere. Based on the meteorological conditions (northerly winds), at 1:00 p.m., the center line ground-level SO^sub 2^ concentration at 10 km downwind of the CLP coal-fired power plant was ~80 ppb using the Gaussian plume equation. However, the observed SO^sub 2^ concentrations at transects D and E were low at only ~23 ppb. The estimated SO^sub 2^ was four times the observed values, suggesting that rapid transformation of SO^sub 2^, such as in-cloud processes, may have occurred.

Figure 8. MAP air pollution measurement track on January 13, 2003 (PM^sub 2.5^ volume concentrations in m^sup 3^/cm^sup 3^).

High PM^sub 2.5^ volume concentrations were observed at transects C-E located downwind of the power plant in Figure 8. At the same time, the wind at Macau, ~50 km west of Hong Kong across the Pearl River, was from the north to northeast with speed <1.5 m/sec^sup 1^ (data not shown). This suggests a negligible contribution of pollutants from the west bank of the river. Transport of PM^sub 2.5^ from the north, if it existed, would lead to higher concentrations of PM^sub 2.5^ at transect B instead of transect E, because transect B was closer to the north, but the PM^sub 2.5^ concentration at transect E was ~50% higher. Low NO^sub x^ and BC concentrations suggest that vehicular emission was not responsible for the high PM^sub 2.5^ concentrations at transects D and E. PM^sub 2.5^ concentration at transect E was ~50% higher than that at transect G (street canyon in town), yet NO^sub x^ and BC concentrations at the latter were seven and two times those at transect E. Thus, it can be inferred that the power plant plumes may have passed through clouds before arriving at transects C-E.

HKEPD stationary data also supported the presence of in-cloud formation particles on January 13, 2003, whereas in-cloud process played a minor role on August 31, 2002. The maximum hourly average RSP and SO^sub 2^ concentrations on January 13 at TC (3:00 p.m.) were 227 and 84 g/m^sup 3^ (RSP/SO^sub 2^ = 2.7), whereas on August 31 (9:00 a.m.), they were 138 and 158 g/m^sup 3^ (RSP/SO^sub 2^ = 0.9). The high RSP:SO^sub 2^ ratio on January 13 strongly suggests the presence of cloud processes in rapid transforming SO^sub 2^ into sulfate.

CONCLUSIONS

A combined stationary and mobile air pollution dataset was used to study the transport of SO^sub 2^ and the formation of O^sub 3^ and particles in the atmosphere. Stationary data provided fixed- point continuous monitoring, whereas mobile data filled the missing gaps between stations during fixed periods. The combined dataset allowed the identification of the intrusion of power plant plumes in urban plumes at ground level. The combined dataset also allowed the distinction between gas-fired and coal-fired power plant plumes at ground level when NO^sub 2^ + O^sub 3^ concentration was analyzed at the receptors.

High-intensity and high-resolution particle size distribution data obtained by the mobile platform allowed the identification of in-cloud process. This study demonstrates the usefulness of combining mobile and stationary data in studying the dispersion and transformation of air pollutants in the atmosphere.

ACKNOWLEDGMENTS

This work is supported solely by The Hong Kong Jockey Club Charities Trust. The authors thank Dr. Alexis K.H. Lau and David W. Yeung for providing the wind fields calculated using MM5. The collective effort of the MAP Team in operating and collecting mobile data is greatly appreciated. The authors thank HKEPD for making their stationary data available for study.

IMPLICATIONS

Conventional stationary air monitoring stations have served atmospheric researchers well for decades but can no longer satisfy the ever-demanding needs to learn more about air pollution problems. This paper demonstrates the usefulness of combining mobile and stationary data in studying the dispersion and transformation of air pollutants in the atmosphere. The analysis may provide useful information for air pollution professionals to improve technology for monitoring air quality.

RE\FERENCES

1. Bukowiecki, N.; Kittelson, D.B.; Watts, W.F.; Burtsher, H.; Weingartner, E.; Baltensperger, U. Real-Time Characterization of Ultrafine and Accumulation Mode Particles in Ambient Combustion Aerosols; J. Aerosol Sci. 2002, 33, 1139-1154.

2. Kittelson, D.B.; Watts, W.F.; Johnson, J.P. Nanoparticle Emissions on Minnesota Highways; Atmos. Environ. 2004, 38, 9-19.

3. Available on Hong Kong Environmental Protection Agency Web site, http://www.epd-asg.gov.hk/english/report/files/agr03e.pdf (accessed December 15, 2005).

4. Cowen, K.; Kelly, T.; Cotant, B.; Riggs, K. Environmental Technology Verification Report: Electrical Low Pressure Impactor (ELPI) Particle Monitor. Battelle: Columbus, OH, 2001.

5. Lau, A.P.S; Chan, C.K.; Fang, M. Mobile Real-Time Air Monitoring Platform (MAP) Volume-I, IESD Report 01-03; Hong Kong University of Science and Technology: Hong Kong, 2003; Chapters 4- 5.

6. Liu, B.Y.H.; Pui, D.Y.H. A. Universal Aerosol Sampling Probe for High Efficiency Particle Sampling From Flowing Gas Streams; Particle Characteristics 1985, 2, 125-132.

7. Kok, G.L.; Lind, J.A.; Fang, M. An Airborne Study of Air Quality around the Hong Kong Territory; J. Geophys. Res. 1997; 102, 19, 043-19057.

8. Seinfeld, J.H.; Pandis, S.N. Atmospheric Chemistry and Physics from Air Pollution to Climate Change; John Wiley & Sons: New York, 1998; pp 234-330.

9. Available on Hong Kong Environmental Protection Agency Web site, http://www.epd.gov.hk/epd/english/environmentinhk/air/data/ emission_inve.html (accessed December 15, 2005).

10. Sunwoo, Y.; Carmichael, G.R.; Ueda, H. Characteristics of Background Surface Ozone in Japan; Atmos. Environ. 1994, 28, 25-37.

11. Soltic, P.; Weilenmann, M. NO^sub 2^/NO Emissions of Gasoline Passenger Cars and Light-Duty Trucks with Euro-2 Emission Standard; Atmos. Environ. 2003, 37, 5207-5216.

12. Chan, L.Y.; Chan, C.Y.; Yin, Y. Surface Ozone Pattern in Hong Kong; J. Applied Metmrol. 1998, 37, 1153-1165.

13. Luria, M.; Valente, RJ.; Tanner, R.L.; Gillani, N.V.; Imhoff, R.E.; Mueller, S.F.; Olszyna, KJ.; Meagher, J.F. The Evolution of Photochemical Smog in a Power Plant Plume; Atmos. Environ. 1999, 33, 3023-3036.

14. Meng, Z.; Seinfeld, J.H. On the Source of the Submicrometer Droplet Mode of Urban and Regional Aerosols; Aerosol Sci. Teclmol. 1994, 20, 253-265.

15. Hering, S.; Eldering, A.; Seinfeld, J.H. Bimodal Character of Accumulation Mode Aerosol Mass Distributions in Southern California; Atmos. Environ. 1997, 31, 1-11.

16. Ondov, J.M.; Wexler, A.S. Where Do Particulate Toxins Reside? An Improved Paradigm for the Structure and Dynamics of the Urban Mid- Atlantic Aerosol; Environ. Sd. Teclmol. 1998, 32, 2547-2555.

17. Zhuang, H.; Chan, C.K.; Fang, M.; Wexler, A.S. Formation of Nitrate and Non-Sea-Salt Sulfate on Coarse Particles; Atmos. Environ. 1999, 33, 42234233.

18. Zhuang, H.; Chan, C.K.; Fang, M.; Wexler, A.S. Size Distributions of Particulate Sulfate, Nitrate and Ammonium at a Coastal Site in Hong Kong; Atmos. Environ. 1999, 33, 843-853.

Xiaohong Yao, Ngai T. Lau, and Ming Fang

Institute for Environment and Sustainable Development, Hong Kong University of Science and Technology, Clear Water Bay, Hong Kong, China

Chak K. Chan

Department of Chemical Engineering, Hong Kong University of Science and Technology, Clear Water Bay, Hong Kong, China

About the Authors

Xiaohong Yao, Ngai T. Lau, and Ming Fang are with the Institute for Environment and Sustainable Development, Hong Kong University of Science and Technology, Clear Water Bay, Hong Kong. Chak K. Chan is with the Department of Chemical Engineering, Hong Kong University of Science and Technology, Clear Water Bay, Hong Kong. Address correspondence to: Ming Fang, Institute for Environment and Sustainable Development, Hong Kong University of Science and Technology, Clear Water Bay, Hong Kong; fax: 852-2358-1528; e-mail: fangming@ust.hk.

Copyright Air and Waste Management Association Feb 2006

Source: Journal of the Air & Waste Management Association

More News in this Category