SEARCH

SEARCH BY CITATION

Keywords:

  • aerosols;
  • biomass burning;
  • megacity;
  • urban;
  • vehicular traffic

Abstract

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Measurements
  5. 3. Measurement Site and Emissions Sources
  6. 4. General Circulation and Variations in Meteorological Parameters
  7. 5. Temporal Variations of BC and OC Aerosols
  8. 6. Emission Ratios of BC, CO, and OC
  9. 7. Weekend Effect on the Concentrations of BC and OC
  10. 8. Comparison With Other Urban Areas
  11. 9. Summary and Conclusions
  12. Acknowledgments
  13. References
  14. Supporting Information

[1] Measurements of black carbon (BC) and organic carbon (OC) were conducted in Bangkok during 2007–2008. Annual trends of BC and OC show strong seasonality with lower and higher concentrations during wet and dry seasons, respectively. Flow of cleaner air, wet removal, and negligible biomass burning resulted in the lowest concentrations of aerosols in the wet season. In addition to anthropogenic sources, long-range transport and biomass burning caused higher concentrations in the dry and hot seasons, respectively. Despite extensive biomass burning in the hot season, moderate levels of aerosols were due to the mixing with air masses from the Pacific Ocean. Diurnal distributions exhibit peaks during rush hour marked by minima in the OC/BC ratio and stagnant wind flow. The lowest concentrations in the afternoon hours could be due to deeper planetary boundary layer and reduced traffic. Overall, the concentrations of both BC and OC decrease with the increase in wind speed. The weekend effects, due to reduced emission during weekends, in the concentrations of both BC and OC were significant. Therefore, stricter abatement in vehicular emissions could substantially reduce pollution. A slope of ΔBC/ΔCO of 9.8 ngm−3 ppbv−1 for the wet season represents the emission ratio from vehicular sources. The highest of ΔOC/ΔBC (3 μg μg−1) in the hot season was due to the predominant influence of biomass burning and significant formation of secondary OC. The levels of BC and OC in Bangkok fall within the ranges of their concentrations measured in the major cities of East Asia.

1. Introduction

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Measurements
  5. 3. Measurement Site and Emissions Sources
  6. 4. General Circulation and Variations in Meteorological Parameters
  7. 5. Temporal Variations of BC and OC Aerosols
  8. 6. Emission Ratios of BC, CO, and OC
  9. 7. Weekend Effect on the Concentrations of BC and OC
  10. 8. Comparison With Other Urban Areas
  11. 9. Summary and Conclusions
  12. Acknowledgments
  13. References
  14. Supporting Information

[2] Anthropogenic emissions of aerosols and trace gases in Asia are increasing because of rapid economic growth and urban development [Ohara et al., 2007; Zhang et al., 2009]. Black carbon (BC) is a refractory component of carbonaceous particles emitted primarily by incomplete combustion [Antony Chen et al., 2001; Penner et al., 1993; Cooke and Wilson, 1996; Liousse et al., 1996]. BC particles strongly absorb radiation in the visible, near UV, and near IR due to their graphitic structure [Rosen et al., 1978]. Model estimates suggest that the strongly absorbing aerosols have large impacts on regional climate and the hydrological cycle [Menon et al., 2002; Ramanathan and Carmichael, 2008]. The atmospheric abundance of BC comprises mainly fine particles, including ∼90% fraction of PM2.5 size (particles up to 2.5 μm in aerodynamic diameter), which can be harmful to human health in polluted regions [Lighty et al., 2000]. The major source sectors of carbonaceous aerosols include emissions from transportation, heating, power generation, industrial processes, biofuels, and biomass burning [e.g., Streets et al., 2003; Bond et al., 2004]. In the global budget of BC, the contributions of emissions from fossil fuel, biofuel, and open biomass burning have been estimated to be ∼38%, 20%, and 42%, respectively [Bond et al., 2004].

[3] Organic carbon (OC) which has both primary and secondary sources constitutes a major fraction of the mass of carbonaceous aerosols in the fine mode [e.g., Turpin et al., 2000, and references therein]. OC aerosols can impact climate directly by scattering solar radiation, and they can also act as cloud condensation nuclei (CCN) [Novakov and Penner, 1993; Saxena et al., 1995] and hence influence the climate indirectly by forming cloud droplets. Primary organic carbon (POC) is emitted directly in particulate form by combustion processes, whereas secondary OC is formed via gas-to-particle conversion of oxidized products of volatile organic compounds (VOCs) in the atmosphere [Pankow, 1994]. In the global budget of OC, the contributions of fossil fuel, biofuel, and open biomass burning have been estimated to be ∼7%, 19%, and 74%, respectively [Bond et al., 2004].

[4] In the tropical regions of Asia, forest fires and biomass burning are widespread [Christopher et al., 1996; Folkins et al., 1997]. The emissions from anthropogenic, forest fires and open biomass burning sources in Southeast Asia (SEA) contribute substantially to the inventories of carbonaceous aerosols in Asia [Streets et al., 2003; Bond et al., 2004]. According to the estimates presented by Streets et al. [2003] for the year 2000, the total anthropogenic emissions of BC and OC aerosols in SEA were about 0.77 Tg and 3.68 Tg, respectively. In spite of the large emissions of various aerosols and trace gases from SEA, studies of the spatiotemporal variations of these species are very limited compared to other regions of the world where biomass burning is an important source of aerosols and trace gases [Folkins et al., 1997; Christopher et al., 1998].

[5] The growing emissions of various aerosols including carbonaceous species from the megacities of Asia can impact local air quality and climate on regional and global scales. In most of the megacities of the world, the emissions of carbonaceous aerosols are mainly due to the use of fossil fuels in automotive engines and industry. The annual emission map of BC from anthropogenic sources in parts of Asia for the year 2000 is shown in Figure 1. The major sources can be located mostly in the eastern part of China, South Korea and Japan where emissions from urban areas are highest. In the SEA region, the emissions of BC from anthropogenic sources are relatively less; however, the emissions from biomass burning and forest fire sources make major contributions.

image

Figure 1. Distribution of anthropogenic emissions of BC for the year 2000 at 0.5° × 0.5° resolution [Streets et al., 2003]. The sites of BC measurements discussed in the present study are also shown.

Download figure to PowerPoint

[6] Thus far, detailed studies characterizing the variations of carbonaceous aerosols near the major source regions in SEA are rare mainly due to the lack of continuous observations [See et al., 2006, and references therein]. The present study is based on continuous observations of PM2.5 carbonaceous aerosols in Bangkok, Thailand during the years 2007–2008. The characteristics of temporal variations of BC and OC aerosols have been discussed in view of the short-term changes in local meteorology, seasonality of the long-range transport and strength of local emissions. The observations of aerosols in Bangkok represent the activities of distinct emission sources than those in typical urban areas of Asia, for example, Beijing and Tokyo in East Asia [Han et al., 2009; Kondo et al., 2006]. The key features of variations observed in the concentrations of BC and OC aerosols are also compared with the measurements reported for other urban areas of Asia.

2. Measurements

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Measurements
  5. 3. Measurement Site and Emissions Sources
  6. 4. General Circulation and Variations in Meteorological Parameters
  7. 5. Temporal Variations of BC and OC Aerosols
  8. 6. Emission Ratios of BC, CO, and OC
  9. 7. Weekend Effect on the Concentrations of BC and OC
  10. 8. Comparison With Other Urban Areas
  11. 9. Summary and Conclusions
  12. Acknowledgments
  13. References
  14. Supporting Information

[7] The mass concentrations of carbonaceous aerosols were measured using a semicontinuous EC-OC analyzer (RT3006) manufactured by Sunset Laboratory, Inc. (Beaverton, Oregon, USA) [Bae et al., 2004; Kondo et al., 2006; Miyazaki et al., 2006], with 1 h time resolution. Air samples were drawn from an inlet fitted with PM2.5 cyclone (cutoff diameter of 2.5 μm) at a flow rate of 16.7 L min−1 to discard coarse particles. Samples were collected on a 1.13 cm2 quartz filter for 40 min, and then the mass concentrations of EC and OC were quantified by the thermal optical transmittance (TOT) method based on the National Institute for Occupational Safety and Health (NIOSH) protocol [Birch and Cary, 1996]. The zero level concentration of EC was measured by a particle filter placed upstream of the denuder. The average zero level was 0.07 μg m−3 during the measurement periods and the detection limit defined as ±3σ of a blank sample was 0.03 μg m−3. Further details of this analyzer and protocols used for the analyses of EC and OC aerosols were presented by Han et al. [2009].

[8] The optical measurements of BC were performed by the Continuous Soot Monitoring System (COSMOS, Kanomax, Osaka, Japan), for which the sampling and optical detection parts are based on the combination of those of a Particle Soot Absorption Photometer (PSAP) and Aethalometer [Miyazaki et al., 2008]. The operating wavelength of 565 nm in the COSMOS is the same as used in the PSAP. Air samples were drawn in by an internally mounted pump, and aerosols are collected on a quartz fiber filter at a flow rate of 0.7 L min−1. The sample collecting spot area of COSMOS was 18.1 mm2, which is smaller compared to the PSAP (19.6 mm2). The filter material, quartz fiber filter (Pallflex E70-2075W) used in COSMOS is the same as those used in the PSAP and Aethalometer. The COSMOS system automatically advances a roll of filter tape (40 mm wide) for 20 mm depending on a preset value of filter transmittance. The criterion of transmittance was usually set to 0.7 for the present study. This is one of the most significant advantages of the COSMOS compared to the PSAP, as it enables unattended operation. A stainless steel (SS) tube with an outer diameter of 3/8 inch and wall thickness of 0.049 inch was used for sample inlet. The inlet line was heated to 400°C in order to effectively volatilize the nonrefractory components of aerosol [Kondo et al., 2006]. The COSMOS measures the absorption coefficient (b0) which is determined by the following equation:

  • equation image

where A is the area of the sample spot, V is the volume of air sample during a period of Δt and It−Δt and It are the average transmittances. The b0 is corrected for the effects of multiple scattering as

  • equation image

where ffil is the correction factor for absorption by multiple scattering in the filter medium [Bond et al., 1999; Virkkula et al., 2005]. Based on the simultaneous measurements of the optical absorption using the heated inlet line and EC the estimated mass absorption cross section (Cabs) was stable between 9.6 m2 g−1 in the wet season and 10.6 m2 g−1 in the dry season in Bangkok [Kondo et al., 2009]. In the present study a Cabs value of 10 m2 g−1 was used to calculate the mass concentration of BC.

[9] For ambient observations of BC in Tokyo, Kondo et al. [2011] compared the measurements obtained using laser incandescence (single particle soot photometer, (SP2)), refraction (refractory mass method (RMM)), thermal optical transmittance (TOT), and light absorption (COSMOS). The excellent agreement between RMM and TOT (MTOT = 0.96 Mref + 0.11 (μgC m−3), r2 = 0.88) and COSMOS versus SP2 (MCOSMOS = 0.99 MSP2 − 0.02 (μgC m−3), r2 = 0.97) was reported. Based on the observations at 6 different sites in Asia, including the present data at Bangkok, Kondo et al. [2009, 2011] have reported very stable and good agreement between a heated COSMOS and EC. The good agreement between the thermal and optical methods has also been reported in several of previous studies [e.g., Venkatachari et al., 2006; Sahu et al., 2009; Miyazaki et al., 2008]. Therefore, for the present study, we have used BC instead of EC throughout the manuscript.

[10] A nondispersive infrared (NDIR) gas analyzer (Model 48C, Thermo Environmental Instruments, USA) was used for the measurement of carbon monoxide (CO) with an integration time of 1 min. The ambient air samples were dried using an electric cooler to reduce interference from water vapor. The background signal (zero level) was routinely measured every 2 h by supplying purified air into the sample line. The zero air was generated by passing ambient air through a column filled with hopcalite (mixture of manganese dioxide and copper oxide). The calibration of the analyzer was performed by supplying a standard of 5 ppmv of CO in air (manufactured by the Nissan-Tanaka Corporation, Japan). The overall precision and accuracy were estimated to be 4 ppbv and 20 ppbv, respectively, at a mixing ratio of 400 ppbv using the data integrated for 1 min. The measurements of CO could not be performed continuously due to technical reasons, therefore we have a limited database for discussing the temporal variation of CO.

[11] The measurements of meteorological parameters (ambient pressure, temperature, relative humidity (RH), wind speed, wind direction, precipitation, solar radiation, etc.) were performed using automated sensors (WXT-510, Vaisala, Finland) at the observation site [Miyazaki et al., 2009]. The data were recorded at a time resolution of 10 min but statistics calculated for 1 h and longer periods are used in the present study.

3. Measurement Site and Emissions Sources

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Measurements
  5. 3. Measurement Site and Emissions Sources
  6. 4. General Circulation and Variations in Meteorological Parameters
  7. 5. Temporal Variations of BC and OC Aerosols
  8. 6. Emission Ratios of BC, CO, and OC
  9. 7. Weekend Effect on the Concentrations of BC and OC
  10. 8. Comparison With Other Urban Areas
  11. 9. Summary and Conclusions
  12. Acknowledgments
  13. References
  14. Supporting Information

[12] Bangkok, the capital city of Thailand, is situated in the central part of Thailand, in the Chao Phraya River Basin of immediate proximity to the Gulf of Thailand. Economically, Bangkok is one of the most important cities in Southeast Asia (SEA). The Bangkok Metropolitan Region (BMR) covers an area of 7,761.5 km2 and had a registered population of 11,971,000 as of January 2008. The city often faces serious traffic congestion due to both public and private vehicles there are estimated to be more than 5.4 million vehicles running on the roads of Bangkok city [Department of Land Transport (DLT), 2008]. New vehicles registrations in the city have increased by 39% compared to that in the year 2002 (details given at http://apps.dlt.go.th/statistics_web/statistics.html).

[13] The measurement site on the campus of the Asian Institute of Technology (AIT, 14.08°N, 100.62°E), which is in the Pathumthani and is located ∼40 km north from the Bangkok city center along the west side of the Phaholyothin road (see Figure 2). This road is important for the northeastern sector connected to Bangkok city. The traffic consists of both gasoline- and diesel-fuelled vehicles. The traffic volume along the Phaholyothin highway does not show significant day-to-day variability during the weekday however, it is less on the weekend [Leong et al., 2002]. The other major sources of pollutants are small industrial estates located ∼6 km north of the site, intensive construction activities about 500 m to the north of the site and other day-to-day activities in this academic center. Additional information about the measurement site and traffic in Bangkok can be found elsewhere [Kim Oanh et al., 2000; Leong et al., 2002].

image

Figure 2. Map of Bangkok and network of roads (taken from Wikimapia http://wikimapia.org); the triangle shows the measurement site at AIT in Bangkok.

Download figure to PowerPoint

[14] The emissions from vehicular traffic contribute up to 80% of NOx (=NO + NO2), 75% of CO, 54% of particulate matter, and most of the volatile organic compounds (VOCs) in the BMR [Bangkok Metropolitan Administration, 2001]. Among the countries in SEA, the number of registered vehicles was highest in Thailand, while the amount of fuel and biomass burned was second after Indonesia [Streets et al., 2003]. The open biomass burning, mainly agricultural waste field burning, surrounding the site is significant source of air pollution in the dry season [Chuersuwan et al., 2008]. Pathumthani, where the site is located, is one of the largest rice growing provinces in Thailand [Office of Agriculture Economics, 2007] with an estimate emission from the agroresidue field burning, mainly rice straw, in tones per year in 2007 of 1470 PM2.5, 90 BC and 530 OC [Kanabkaew and Kim Oanh, 2011].

[15] During the dry season, most of harvested paddy (>90%) is burned and the burning were particularly extensive during December-February period [Tipayarom and Kim Oanh, 2007]. The open burning of rice straw in the Pathumthani region has important implications for the air quality of Bangkok as these sources are located upwind of the BMR. The emissions from vehicular traffic and industrial activity can be assumed to be fairly constant throughout the year, as they do not depend on season. On the other hand, the activities of biomass burning and forest fire exhibit strong seasonality.

4. General Circulation and Variations in Meteorological Parameters

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Measurements
  5. 3. Measurement Site and Emissions Sources
  6. 4. General Circulation and Variations in Meteorological Parameters
  7. 5. Temporal Variations of BC and OC Aerosols
  8. 6. Emission Ratios of BC, CO, and OC
  9. 7. Weekend Effect on the Concentrations of BC and OC
  10. 8. Comparison With Other Urban Areas
  11. 9. Summary and Conclusions
  12. Acknowledgments
  13. References
  14. Supporting Information

[16] There are three main seasons, namely wet (May-October), dry (November-February), and hot (March-April), in Thailand [Pochanart et al., 2003]. The wet season prevails due to southwesterly (SW) wind flow, and the dry season is due to northeasterly (NE) wind flow. The SW wind flow is associated with the northward movement of the intertropical convergence zone (ITCZ) across Thailand which brings cleaner marine air from the southern Indian Ocean. During the dry season, the long-range transport of continental air from different regions of East Asia takes place due to the southward movement of the ITCZ. The observation site is influenced by mixed air masses (marine and continental) during the hot season.

[17] The time series variations of surface level wind speed, wind direction, RH, pressure, and temperature based on daily data and total rainfall data observed between April 2007 and March 2008 are shown in Figure 3. During the months of May to October, episodes of rainfall and high RH (>60%) were very frequent. Therefore this period is also known as the wet season in Thailand. The wind speed varied between 0.5 m s−1 and 2.0 m s−1, including several episodes of stronger winds, particularly in the month of August. From the months of November to February rainfall was rare therefore, this period is also known as the dry season. The time series variation of surface temperature shows an increasing trend from March to April therefore, this period is known as the hot season. The measurements during the dry and hot seasons were influenced by a fairly stable wind flow of ∼1 m s−1 from the east-southeast (E-SE) direction. The surface level pressure varied between the ranges of 1000–1010 hPa and 1010–1015 hPa during the wet and dry seasons, respectively. The day-to-day variations in temperature and pressure were anticorrelated, which can be clearly seen during the episodes of strong winds from NE direction.

image

Figure 3. Time series plots of meteorological parameters and fire count data during April 2007 to March 2008.

Download figure to PowerPoint

[18] Almost all the meteorological parameters show systematic diurnal dependencies, with stronger amplitudes of variation in the dry season and weaker during the wet season. Generally, the levels of RH were elevated during the nighttime, while lower values were observed during the daytime. The diurnal variations of RH were in the ranges of 43–83%, 53–83%, and 40–76% during the hot, wet, and dry seasons, respectively. Variations in the wind speed show cycles opposite to that of RH, i.e., lower values during the nighttime and higher values during the daytime. The diurnal ranges of wind speed were 0.5–1.1 m s−1, 0.5–1.3 m s−1 and 0.5–1.1 m s−1 during the hot, wet, and dry seasons, respectively. The surface level pressure shows a peak between 0900 and 1200 LT and a minimum between 1700 and 1900 LT. The diurnal variation of surface temperature was similar to that of wind speed, exhibiting a minimum during 0600–0900 LT and a maximum during 1400–1700 LT. Since the observation site is not very far from the Gulf of Thailand, the systematic diurnal changes observed in the meteorological parameters could partly be attributed to the presence of land-sea breeze circulation. The relations of the concentrations of BC and OC aerosols with the meteorological parameters are also discussed in section 5.4. Based on the analysis of meteorological data reordered at different locations in the BMR during January 2002 to December 2004, the daily average depths of the planetary boundary layer (PBL) were 860 m, 990 m and 940 m for the wet, dry and hot seasons, respectively [Pongkiatkul and Kim Oanh, 2007].

5. Temporal Variations of BC and OC Aerosols

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Measurements
  5. 3. Measurement Site and Emissions Sources
  6. 4. General Circulation and Variations in Meteorological Parameters
  7. 5. Temporal Variations of BC and OC Aerosols
  8. 6. Emission Ratios of BC, CO, and OC
  9. 7. Weekend Effect on the Concentrations of BC and OC
  10. 8. Comparison With Other Urban Areas
  11. 9. Summary and Conclusions
  12. Acknowledgments
  13. References
  14. Supporting Information

5.1. Seasonal Variations

[19] As shown in Figure 4, time series plots of the concentrations of BC and OC and the OC/BC ratio show significant day-to-day variations. However, not shown in Figure 4, the hourly data of BC, OC, and OC/BC ratio were in the ranges of 0.06–28 μg m−3, 0.06–50 μg m−3, and 0.2–27 μg μg−1, respectively. Such large variations in the hourly data were not just random but constitute very systematic diurnal variations in Bangkok. The diurnal features observed in the concentrations of aerosols and impacts of the meteorological parameters are discussed in section 5.2. The daily averaged concentrations of BC and OC and the OC/BC ratio fall in the ranges of 0.9–13.7 μg m−3, 2.01–32.5 μg m−3, and 0.85–5.4 μg μg−1, respectively. Overall, these short-term variations in the concentrations of aerosols were significant throughout the year, and these were particularly strong during the dry and hot seasons.

image

Figure 4. The daily means and monthly variations of concentrations of (a) BC, (b) OC, and (c) OC/BC ratio. In the box-whisker plots the horizontal solid lines represent 10th, 25th, 50th, 75th, and 90th percentiles while dashed lines give the means obtained using hourly data.

Download figure to PowerPoint

[20] Box-whisker plots representing the monthly statistics of BC, OC, and OC/BC ratio are also shown in Figure 4. The concentrations of both BC and OC show clear seasonality, as their levels were low during the wet season and high during the dry season. The average concentrations of BC and OC were 3.0 ± 1.2 μg m−3 and 5.3 ± 2.0 μg m−3 in the wet season, while these were 4.3 ± 1.3 μg m−3 and 13.1 ± 5.8 μg m−3, respectively, during the dry season. The ratio of OC/BC shows a slightly different seasonality, with a lowest value of 1.9 ± 1.3 μg μg−1 in the wet season and highest of 3.3 ± 1.5 μg μg−1 during the hot season. In the hot season, though the activities of biomass burning were highest near the observation site, yet the levels of aerosols were moderate (see Table 1). The concentrations of both BC and OC aerosols show minima in the month of August and maxima in January. The monthly average OC/BC ratio shows a minimum of 1.7 μg μg−1 in the month of August and a maximum of 3.9 μg μg−1 in March. The highest value of OC/BC observed in March coincides with the largest numbers of fire count (hot spot) over SEA and regions surrounding the observation site (see Figure 3). Agreeing to this observation, higher SOC formation has been reported as also highest ozone (O3) is observed in the BMR during March-April period [Nghiem and Kim Oanh, 2008]. In section 5.3, the back trajectory and fire count data have been analyzed to explain the major causes of seasonality observed in the concentrations of aerosols.

Table 1. Monthly Statistics of BC, OC, and OC/BC Ratio Measured at AIT, Bangkok in Thailanda
MonthSeasonBC (μg m−3)OC (μg m−3)OC/BC (μg/μg)
  • a

    Statistics are median values with mean plus or minus standard deviation given in parentheses.

Apr 2007Hot3.7 (3.8 ± 1.2)8.0 (9.1 ± 2.8)2.9 (3.0 ± 0.8)
May 2007Wet2.8 (3.1 ± 1.1)4.1 (4.5 ± 1.5)1.6 (2.0 ± 1.4)
Jun 2007Wet3.2 (3.3 ± 1.2)4.1 (4.2 ± 1.1)1.7 (1.7 ± 0.4)
Jul 2007Wet2.9 (3.0 ± 1.6)4.1 (4.5 ± 1.5)1.9 (2.4 ± 1.5)
Aug 2007Wet2.4 (2.7 ± 1.2)3.7 (3.9 ± 0.8)2.0 (1.9 ± 0.4)
Sep 2007Wet2.8 (3.0 ± 1.1)4.3 (4.7 ± 1.5)2.0 (2.0 ± 0.6)
Oct 2007Wet3.2 (3.2 ± 1.0)4.8 (5.2 ± 1.9)1.9 (2.0 ± 0.6)
Nov 2007Dry3.9 (4.5 ± 1.8)7.2 (8.4 ± 2.8)2.0 (2.1 ± 0.3)
Dec 2007Dry5.1 (5.2 ± 1.6)10.8 (11 ± 3.0)2.4 (2.6 ± 0.6)
Jan 2008Dry5.6 (5.7 ± 2.3)17.4 (17.2 ± 6.4)3.6 (3.5 ± 0.7)
Feb 2008Dry4.3 (4.7 ± 1.3)14.6 (14.7 ± 5.1)3.6 (3.6 ± 0.9)
Mar 2008Hot3.0 (3.9 ± 2.4)9.4 (13.1 ± 7.5)3.8 (3.9 ± 0.6)

5.2. Diurnal Variations

[21] The diurnal variations of BC and OC for the different seasons are shown in Figure 5. The concentration of BC shows significant diurnal variability throughout the year. There are two prominent peaks of BC concentration a primary peak at around 0700 LT and a secondary peak at 2000 LT. The concentration of BC was observed to be lowest in the afternoon hours. The variations of BC were within the ranges of 1.4–5.3 μg m−3, 2.3–8.3 μg m−3, and 2.3–6.5 μg m−3 during the wet, dry, and hot seasons, respectively. The concentration of OC exhibits somewhat similar diurnal variation to that of BC in the wet and dry seasons. On the other hand, the concentration of OC shows less pronounced variation in the hot season. The diurnal variations of OC were within the ranges of 3.5–6.3 μg m−3, 8.5–17.5 μg m−3, and 9.5–12 μg m−3 during the wet, dry, and hot seasons, respectively.

image

Figure 5. Diurnal variations of concentrations of BC and OC, OC/BC ratio, and meteorological parameters during the wet (May-September), dry (November-February), and hot (March-April) seasons.

Download figure to PowerPoint

[22] The local time dependencies of emission and meteorological parameters can be important factors controlling the diurnal distribution of primary pollutants in urban areas. However, the contributions of different factors cannot be separated in a strict sense. The traffic volume on the Phaholyothin road peaked during the morning (0700–0900 LT) and evening (1600–1800 LT) hours, while it was moderate in the afternoon and lowest during the midnight and early morning [Kim Oanh et al., 2000; Leong et al., 2002]. The morning rush hour coincided with the period of stagnant airflow (<0.5 m s−1) favoring the accumulation of primary pollutants. The lower concentrations observed in the afternoon hours can be attributed primarily to the dilution of aerosols associated with the deeper PBL depth, higher wind speed, and slight reductions in the traffic volume. Incidentally, the minima in the concentrations of BC and OC observed at around 1300 LT coincide with a minimum in the daytime traffic on the roads of Bangkok [Kim Oanh et al., 2008]. The formation of a nocturnal boundary layer and stagnant flow would favor the accumulation of pollutants causing elevated concentrations of aerosols during the night and early morning hours even though the traffic volume was observed to be lowest during these hours. Emissions from biomass burning sources do not show a clear dependence on local time the systematic and smooth diurnal patterns of both BC and OC suggest the importance of local meteorology in controlling diurnal variations throughout the year.

[23] The differences between the concentrations of BC and OC were significantly higher during the daytime than nighttime. For example, in the wet season, the peak concentrations of BC and OC were ∼5.3 μg m−3, but in the afternoon hours the decrease in BC was higher by a factor of two compared to that in the concentration of OC. Such differences in the diurnal patterns of BC and OC highlight the presence of secondary OC, which partially compensates for the impact of dilution due to dispersion in the daytime. On the other hand, it would be difficult to predict that either SOC or POC is dominant in the observed OC.

[24] The diurnal plots of OC/BC ratio shows a minimum at around 0700 LT which coincides with the primary peak in BC concentration suggesting the influence of fresh emissions from automobiles due to the morning rush in traffic. The increasing trend starts immediately after this minimum and peaks at around 1300 LT. The plateau between 1300 LT and 1600 LT can be attributed to enhanced solar radiation leading to secondary formation of OC. The diurnal variations of OC/BC ratio were within the ranges of 1–3 μg μg−1, 1.9–4.1 μg μg−1, and 1.8–5.5 μg μg−1 during the wet, dry, and hot seasons, respectively. In the absence of secondary production of OC, no clear diurnal dependence of the OC/BC ratio can be expected.

5.3. Impacts of Long-Range Transport and Biomass Burning Emissions

[25] Back trajectory model data have been widely used to track the origins and transport pathways of air parcels. In the present study, isentropic back trajectories were calculated using the National Institute of Polar Research (NIPR) trajectory model [Tomikawa and Sato, 2005]. For this calculation, European Centre for Medium-Range Weather Forecast (ECMWF) wind data were used. The trajectories were calculated for a total run time of 240 h (10 days) at 950 hPa level with a time step of 20 min. The origins and pathways of trajectories arriving at the observation site show very clear and systematic seasonality (Figure 6). In the wet season, the site is influenced by cleaner air originating over the southern Indian Ocean due to transport from the SW direction (SW monsoon). The ambient pressure along the trajectories suggests that the air masses were confined within the marine boundary layer (MBL) of the Indian Ocean during transport. The trajectories in the month of October represent the transition in the transport pattern from SW to NE, indicating the onset of the dry season. In the dry season, the trajectories were traced to the free troposphere over the eastern continental regions of China. In the hot season, the trajectories originated over the Pacific Ocean and arrived at the observations site from the SE direction.

image

Figure 6. Back trajectories and hot spot map for different seasons of observations during April 2007 to March 2008.

Download figure to PowerPoint

[26] The fire count data detected by the Along Track Scanning Radiometer (ATSR) which represent the activities of biomass burning and forest fires are also plotted in Figure 6. Details of the detection methodologies used for the ATSR hot spot data can be found elsewhere [Buongiorno et al., 1997; Arino et al., 2001]. In addition to the negligible activities of biomass burning, the flow of cleaner marine air and wet removal could be the major cause for the lower concentrations of aerosols observed in the wet season. Still, the average values of BC and OC were significantly higher than their background levels in the wet season (see Table 2). The definition of “background” is not very strict and varies from study to study, and we have chosen to define the background as the 1.25 percentile level for urban measurements [Kondo et al., 2006; Verma et al., 2010]. In the dry season, the use of fossil fuels in automobiles is a major source of carbonaceous aerosols in the study area. While a few hot spots were detected near the site many spots can be seen in the far upwind regions particularly over the southeast of China. Therefore, the highest concentrations of carbonaceous aerosols observed in the dry season could be attributed to both local emissions and long-range transport of pollutants mainly from China. The activities of biomass burning and forest fires were very extensive in SEA in the hot season (see Figure 3). In this season, however unexpectedly, the concentrations of both BC and OC were lower compared to those measured during the dry season (Table 1). The analyses of hot spot data and back trajectories suggest that the impact of strong emissions in the hot season were subdued by the dilution and likely scavenging of aerosols due to mixing with cleaner and humid air from Pacific Ocean. In other words, the wind flow from the SE direction (near the site) diluted the local emissions significantly otherwise, we anticipate that we would have observed the highest concentrations of carbonaceous aerosols due to intense biomass burning.

Table 2. Background Concentrations of BC and OC and the OC/BC Ratio Estimated for Different Seasons at AIT, Bangkok in Thailand
 [BC]0 (μg m−3)[OC]0 (μg m−3)[OC/BC]0 (μg μg−1)
Wet season0.541.020.46
Dry season0.822.210.88
Hot season0.942.781.02

5.4. Relationship Between Aerosols and Local Meteorology

[27] The time series concentrations of both BC and OC show anticorrelated variations with wind speed throughout the year however, such relations can be seen more clearly during episodes of strong wind flow. For instance, during two episodes of high wind speed in the month of August, the level of BC decreased from ∼8.0 μg m−3 to ∼2.0 μg m−3 at the same time the change in OC was from ∼10.0 μg m−3 to ∼4.0 μg m−3. In another example, the highest concentrations of ∼14 μg m−3 of BC and ∼32 μg m−3 of OC coincided with episodes of stagnant flow (<0.5 m s−1) during the dry and hot seasons. In the wet season, the lesser impact of wind speed in the concentration of OC suggests the significant and widespread presence of secondary OC. The day-to-day variation in the concentration of BC seems to be controlled mainly by dispersion, while for OC secondary production of OC is also important. Consequently, the ratio of OC/BC and wind speed show distinct relations depending on the season. The sharp declines in the concentrations of BC and OC were also observed during the episodes of rainfall in Bangkok, which play an important role in the wet scavenging of aerosols during the wet season.

[28] The dependencies of concentrations of BC and OC and the OC/BC ratio on the wind speed (averaged for a bin of 0.25 m s−1) for different seasons are shown in Figure 7. In the lower wind speed regime (<1.0 m s−1), the concentrations of both BC and OC decrease significantly with the increase in wind speed during the dry and hot seasons, while in the higher wind flow regimes (1.0–2.5 m s−1) the concentrations of aerosols show weaker dependencies. The concentration of BC seems to be more sensitive than that of OC to wind speed. The lesser dependence of OC on wind speed, particularly in the wet season, could be due to the significant presence of secondary OC in Bangkok and surrounding regions. Consequently, the ratio of OC/BC tends to increase with the increase in wind speed (see Figure 7c).

image

Figure 7. Dependences of (a) BC, (b) OC, and (c) OC/BC ratio on wind speed (calculated for each 0.25 m s−1 of wind speed) during the wet (May-September), dry (November-February), and hot (March-April) seasons.

Download figure to PowerPoint

[29] In the dry season, long-range transport brings continental air masses with elevated backgrounds of both BC and OC aerosols. Therefore, dilution is not so effective even if the wind flow is stronger at the measurements site. On the other hand, the measurements site is influenced by the transport of cleaner air from the Pacific Ocean in the hot season. Consequently, the dilution due to mixing with air masses having lower backgrounds resulted in sharp decreases in BC and OC aerosols. In other words, the relationship between the concentrations of aerosols and wind speed varies with season. Similar relations of BC with wind speed have also been reported for other urban locations of Asia, for example, in Tokyo [Kondo et al., 2006], Beijing [Han et al., 2009], and Guangzhou [Verma et al., 2010].

6. Emission Ratios of BC, CO, and OC

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Measurements
  5. 3. Measurement Site and Emissions Sources
  6. 4. General Circulation and Variations in Meteorological Parameters
  7. 5. Temporal Variations of BC and OC Aerosols
  8. 6. Emission Ratios of BC, CO, and OC
  9. 7. Weekend Effect on the Concentrations of BC and OC
  10. 8. Comparison With Other Urban Areas
  11. 9. Summary and Conclusions
  12. Acknowledgments
  13. References
  14. Supporting Information

[30] The concentration of BC is expected to be correlated with the mixing ratio of CO due to their coemission from the incomplete combustion of fuels and biomass burning [Baumgardner et al., 2002; Park et al., 2005; Kondo et al., 2006]. However, the emission ratio of BC/CO (EBC/ECO) depends on the emission sector, even if they are colocated [Bond et al., 2004]. Scatterplots of BC and CO data observed during the morning and evening rush hours for all seasons are shown in Figure 8. In the wet season, the BC-CO plot shows a good correlation (r2 = 0.68) and the slope ΔBC/ΔCO of 9.8 ng m−3 ppbv−1 was estimated using the least square linear fit regression method. For the dry season, the slope of ΔBC/ΔCO is 7.9 ng m−3 ppbv−1. The slope of ΔBC/ΔCO in the wet season represents the emission ratio of BC from local sources, while the slope estimated for the dry season also accounts for the impact of long-range transport and rice straw burning. The slope of ΔBC/ΔCO of 9.8 ng m−3 ppbv−1 in the wet season represents the EBC/ECO mostly from vehicular traffic which is slightly higher compared to the values reported for the urban regions of Tokyo (5.7 ± 0.9 ng m−3 ppbv−1) and Nagoya (6.3 ± 0.5 ng m−3 ppbv−1) in Japan [Kondo et al., 2006, and references therein] Guangzhou (5.7 ± 0.9 ng m−3 ppbv−1) and Beijing (3.5–5.8 ng m−3 ppbv−1) in China [Verma et al., 2010, and references therein]. The slope of ΔBC/ΔCO for the wet season can be used to calculate the emission of BC from anthropogenic sources provided the emission of CO is known. Apart from the validation of the existing inventories the study of the BC-CO relation can be useful to investigate the impacts of local emissions and transport efficiency of BC from remote sources. The slope of ΔBC/ΔCO of 9.2 ng m−3 ppbv−1 was estimated for the hot season. The similarity in the slopes of ΔBC/ΔCO estimated for the wet and hot seasons is unexpected due to the major emissions from characteristically different sources. In the hot season, dilution and likely scavenging of aerosols due to the mixing with cleaner and humid air from the Pacific Ocean could have resulted in a lower ΔBC/ΔCO slope than the values expected in fresh biomass plume.

image

Figure 8. Correlations between the concentrations of BC and CO based on the data measured during the rush hours of wet (May-September), dry (November-February), and hot (March-April) seasons.

Download figure to PowerPoint

[31] The relationship between the concentrations of BC and OC depends mainly on the types of emission sources and the level of photochemical oxidation. Separate OC-BC scatterplots for the nighttime and daytime observations for the different seasons are shown in Figure 9. Overall, the moderate correlations indicate the presence of both POC and secondary OC. The better correlations observed in the nighttime data suggest the dominance of POC compared to the measurements during the daytime. The slopes of ΔOC/ΔBC estimated from the linear fit regression were within the ranges of 0.49–0.7 μg μg−1 and 2.08–3.08 μg μg−1 during the wet and hot seasons, respectively. The slope of ΔOC/ΔBC estimated for the wet season agrees reasonably with a range of 0.5–0.9 μg μg−1 from automotive emissions [Gillies et al., 2001; Kirchstetter et al., 2004]. On the other hand, the higher values of ΔOC/ΔBC observed in the hot season indicate a major contribution from biomass burning sources [Andreae and Merlet, 2001]. Previous studies in the BMR region [Kim Oanh et al., 2010a, 2010b] have also reported lower OC/BC ratio of ∼0.5 from diesel emission and higher values of ∼5.6 near the rice straw burning sources. The intercept of the OC-BC scatterplot can be interpreted as the contributions of OC from noncombustion sources and secondary production [Cao et al., 2003]. The slopes of ΔOC/ΔBC estimated for many cities in East Asia also show similar seasonality [Cao et al., 2003, and references therein]. To understand the relative importance of the scattering and absorption characteristics of carbonaceous aerosols the slope of ΔOC/ΔBC can be used as a direct reference. The results presented in this study can be used for the radiative forcing studies over SEA, as climate modelers still assume a constant OC/BC ratio for such estimates [e.g., Koch, 2001; Bond et al., 2004].

image

Figure 9. Correlations between the concentrations of BC and OC for the nighttime and daytime measurements during different seasons at AIT in Bangkok.

Download figure to PowerPoint

7. Weekend Effect on the Concentrations of BC and OC

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Measurements
  5. 3. Measurement Site and Emissions Sources
  6. 4. General Circulation and Variations in Meteorological Parameters
  7. 5. Temporal Variations of BC and OC Aerosols
  8. 6. Emission Ratios of BC, CO, and OC
  9. 7. Weekend Effect on the Concentrations of BC and OC
  10. 8. Comparison With Other Urban Areas
  11. 9. Summary and Conclusions
  12. Acknowledgments
  13. References
  14. Supporting Information

[32] The “weekend effect” is characterized by the reduction in the levels of primary pollutants due to the decrease in emissions from anthropogenic activities during the weekend [e.g., Cerveny and Balling, 1998]. The impact of decreased emissions, mostly due to vehicular exhausts in the ambient concentrations of various pollutants during the weekend has been a topic of research interest since the 1970s [e.g., Lebron, 1975; Elkus and Wilson, 1977]. To study the effect of the weekend in Bangkok, we have separately analyzed BC and OC data for the weekdays (Monday to Friday) and the weekend (Sunday).

[33] The average diurnal plots of BC and OC for the weekdays and weekends of different seasons are shown in Figure 10. In the wet and dry seasons, the local time dependence of the weekday-weekend difference in BC is somewhat similar to the diurnal pattern of BC mass concentration. The weekday-weekend differences were highest for the measurements during the rush hour and lowest during the afternoon hours. In the hot season, the diurnal features of weekday-weekend differences in both BC and OC were slightly different than those seen for the other seasons. In this season, the interpretation of the weekend effect in the concentrations of both BC and OC may not be straightforward due to emissions also from biomass burning. Unlike the weekend effect in BC concentration, the diurnal pattern of weekday-weekend difference in OC concentration changes significantly with the season, mainly due to the presence of a secondary source of OC. In addition to the seasonal changes in the meteorological parameters, the seasonality in the effects of the weekend can be attributed to the presence of additional sources like the long range transport and local biomass burning in the dry and hot seasons, respectively. The impacts of reduced anthropogenic activity in the ambient concentrations suggest that the emissions from vehicular exhaust play an important role in the distributions of aerosols all year around. The changes observed in the levels of BC and OC during the weekend were consistent with the reductions in the traffic volume in Bangkok during the weekend. The hourly traffic volume on Sunday was about 20–30% lower compared to weekday traffic, while on Saturday the traffic was lower only by 5–10% [Kim Oanh et al., 2008].

image

Figure 10. Diurnal variations of BC and OC for weekday and weekend observations at AIT in Bangkok during the wet (May-September), dry (November-February), and hot (March-April) seasons.

Download figure to PowerPoint

[34] Leong et al. [2002] have also reported substantial reductions in the concentrations of PM10, NO2, SO2 and benzene at a site in Bangkok that is influenced by vehicular traffic along the Phaholyothin road. Seasonal variations in the magnitude of the weekend effect have been observed in other urban locations, for example, the San Francisco Bay area [Kirchstetter et al., 2008], Rochester, Philadelphia [Jeong et al., 2004], etc. Interestingly, measurements in Beijing do not indicate the impacts of the weekend in the concentrations of BC, CO, or CO2 due to significant emissions also from nonvehicular sources throughout the year [Han et al., 2009].

8. Comparison With Other Urban Areas

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Measurements
  5. 3. Measurement Site and Emissions Sources
  6. 4. General Circulation and Variations in Meteorological Parameters
  7. 5. Temporal Variations of BC and OC Aerosols
  8. 6. Emission Ratios of BC, CO, and OC
  9. 7. Weekend Effect on the Concentrations of BC and OC
  10. 8. Comparison With Other Urban Areas
  11. 9. Summary and Conclusions
  12. Acknowledgments
  13. References
  14. Supporting Information

[35] Recent studies [Kondo et al., 2006; Han et al., 2009; Verma et al., 2010] have presented key features in the distributions of BC in major cities of East Asia such as Tokyo in Japan and Guangzhou and Beijing in China, where emissions of BC from anthropogenic activity are very high (see Figure 1). The present work is the first study characterizing the diurnal and seasonal variations of carbonaceous aerosols using well time-resolved measurements in an urban region of SEA. The vehicular traffic in Bangkok consists of motorcycle (gasoline, 40% of total), three-wheeler or “tuktuk” (liquefied petroleum gas (LPG), 0.2%), private car (gasoline/LPG/compressed natural gas (CNG), 37%), taxi (LPG/CNG/gasoline, 1.6%), van and pickup (diesel/CNG, 18%), bus (diesel/CNG, 0.6%), and truck (diesel/CNG, 1.8%) [DLT, 2010]. The nontraffic sources in the street include the use of LPG for cooking and charcoal by food vendor stalls in Bangkok [Kim Oanh et al., 2008]. The emissions of BC and OC aerosols from biomass burning sources in cities like Tokyo, Nagoya, and Guangzhou contribute less than in Bangkok. In the Tokyo metropolis, the average fractions of traffic volume from passenger vehicles (cars and two-wheeler), buses, light duty vehicles (LDVs) and heavy duty vehicles (HDVs) were 61%, 1%, 18.6%, and 19% respectively for the year 2005 [Ministry of Land Infrastructure and Transport, 2006]. In Guangzhou city, the average traffic of motorcycles, HDVs, medium duty vehicled (MDVs) and LDVs accounted for about 43%, 6%, 9%, and 42% respectively, for the year 2003 [Xie et al., 2003].

[36] The comparison of diurnal variations of BC in Bangkok with the measurements reported for major cities like Guangzhou, Beijing, Tokyo, etc. in East Asia are shown in Figure 11. The level and pattern of variability of BC differ between the cities. The peak of BC concentration during the morning rush hour in Bangkok was observed to be ∼2 h earlier than those in urban sites of China. Although the concentration of BC shows weaker diurnal dependences in Tokyo, however, it exhibits the opposite pattern compared to other cities. We can see a slightly higher level of BC during the daytime and lower during the night and early morning hours in Tokyo. The diurnal amplitudes of BC in Bangkok were significantly higher than the variations observed in Beijing, Guangzhou and Tokyo. One of the reasons for such large variations in Bangkok could be the large diurnal variation in PBL depth due to the tropical climate, as other sites are located in the higher latitudes. However, to investigate the roles of other factors, such as the local time dependence of emissions, detailed study is needed.

image

Figure 11. Comparison of the diurnal variations of BC at AIT in Bangkok and observations reported for major urban areas of East Asia.

Download figure to PowerPoint

[37] In Table 3, the average concentrations of BC and OC and the OC/BC ratio have been compared with the measurements reported for other urban areas of the world. The concentrations of both BC and OC at Bangkok fall within the range of values for many urban locations. The ratios of OC/BC in the wet season in present study are close to the values reported for urban areas of North America, South Korea, and Japan where fossil fuel combustion is a major source of carbonaceous species. On the other hand, the higher values of OC/BC during the dry and hot seasons are close to the values reported for several urban areas of China [Novakov et al., 2005]. This similarity could be due to the emissions also from non-fossil fuel sources in urban regions of China and SEA [Verma et al., 2010]. In the present study, our observations showing significant seasonality of the OC/BC ratio contradicts the suggestion that the OC/BC ratio does not change significantly with season for a given region [Novakov et al., 2005]. The OC/BC ratio of 7.13 derived from the inventory of biomass emissions in “other Asian countries” [Bond et al., 2004] is higher by a factor of two or more than the values observed in the present study in Bangkok.

Table 3. Comparison of Concentrations of BC and OC and the OC/BC Ratio in Bangkok With the Data Reported for Other Urban Areas of the World
LocationSampling DatesParticles Size Cut/Analytical MethodaBC (μg m−3)OC (μg m−3)OC/BC RatioReferences
  • a

    MNO method uses MnO2 as the oxygen donor for carbon; TOT, thermal optical transmittance.

BeijingAnnual 1999–20002.5/TOT25.309.402.69He et al. [2001]
BeijingAnnual 2005–20062.5/TOT6.915.02.6Han et al. [2009] and Lin et al. [2009]
ShanghaiAnnual 1999–20002.5/TOT14.346.212.31Ye et al. [2003]
Hong KongNov-Feb 2000–20012.5/MNO8.384.072.06Ho et al. [2003]
GuangzhouJan-Feb 20022.5/TOT22.608.302.72Cao et al. [2003]
GuangzhouJul 20062.5/TOT4.7  Verma et al. [2010]
SeoulNov-Dec 19992.5/MNO15.207.002.17Park et al. [2002]
TokyoAnnual 20042.5/TOT4.982.561.93Miyazaki et al. [2006]
Urban CaliforniaAnnual 1988–19892.5/TOT5.193.671.63Chow et al. [1993a, 1993b]
MexicoFeb-Mar 19972.5/TOT9.835.761.91Chow et al. [2002]
14 cities in ChinaWinter 20032.5/TOT38.109.903.85Cao et al. [2007]
14 cities in ChinaSummer 20032.5/TOT13.803.603.83Cao et al. [2007]
BangkokWet season 20072.5/TOT5.33.01.9this study
BangkokDry season 2007–20082.5/TOT13.14.33.05this study
BangkokHot season 20082.5/TOT8.73.353.3this study

9. Summary and Conclusions

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Measurements
  5. 3. Measurement Site and Emissions Sources
  6. 4. General Circulation and Variations in Meteorological Parameters
  7. 5. Temporal Variations of BC and OC Aerosols
  8. 6. Emission Ratios of BC, CO, and OC
  9. 7. Weekend Effect on the Concentrations of BC and OC
  10. 8. Comparison With Other Urban Areas
  11. 9. Summary and Conclusions
  12. Acknowledgments
  13. References
  14. Supporting Information

[38] Concentrations of BC and OC aerosols were measured in Bangkok during April 2007 to March 2008. Emission from automobile vehicles was a major anthropogenic source of aerosols throughout the year, while biomass burning contributed significantly during the hot season. The observation site was influenced by the flow of cleaner oceanic air due to the prevailing southwest monsoon during the wet season. On the other hand, the northeast flow transported from the continental polluted air mainly from China passing through SEA continent territory during the dry season.

[39] The annual trends of both BC and OC aerosols show clear seasonality, with the lowest concentrations in the wet season and highest in the dry season. The average concentrations of BC were 3.0 ± 1.2 μg m−3 and 4.3 ± 1.3 μg m−3 in the wet and dry seasons, respectively. The concentrations of OC were 5.3 ± 2.0 μg m−3 and 13.1 ± 5.8 μg m−3 in the wet and dry seasons, respectively. In the wet season, the concentrations of aerosols emitted from local sources were diluted due to the flow of cleaner air from the Indian Ocean and wet removal. In the dry season, the long-range transport of pollutants mainly from China added to the local emissions, resulting in the highest concentrations of BC and OC aerosols. In spite of the highest activities of biomass burning in the hot season, moderate levels of BC of 3.8 μg m−3 and OC of 11 μg m−3 were due to the mixing with cleaner Pacific air masses. However, the highest average ratio of OC/BC of 3.8 μg μg−1 observed in the hot season confirms the predominant contributions from biomass burning sources.

[40] The diurnal distributions of BC and OC showed elevated concentrations during evening to morning hours and the lowest in the afternoon hours. Concentrations of both BC and OC aerosols exhibited two prominent peaks during the rush hours, which also coincided with stagnant airflow (<0.5 m s−1). The minimum of OC/BC ratio (for example ∼1.0 μg μg−1 in the wet season) also coincided with the peaks in BC, suggesting the influence of fresh emissions from vehicular exhausts during the rush hour. The diurnal variation in PBL depth controlled the diurnal patterns of both BC and OC, while the elevated level of OC/BC ratio during the daytime suggests the secondary production of OC.

[41] The concentrations of both BC and OC decreased with the increase in wind speed and were particularly sensitive in the lower wind flow regime (<1.0 m s−1). The seasonality in the relationship between the concentrations of aerosols and wind speed could be due to the transport of air masses with different background levels of aerosols. The sharp declines in the concentrations of BC and OC were observed during rainfall in Bangkok due to wet scavenging of aerosols in the wet season.

[42] The slope of ΔBC/ΔCO of ∼9.8 ng m−3 ppbv−1 estimated for the measurements in the wet season represents the emission ratio (EBC/ECO) from local anthropogenic sources. In the dry season, a lower value of ∼7.9 ng m−3 ppbv−1 was due to the combined contributions of long-range transport of continental air and emissions from local sources. The slope of ΔOC/ΔBC also shows a clear seasonality with lower values of 0.5–0.7 μg μg−1 in the wet season and higher values of 2.08–3.08 μg μg−1 in the hot season characterizing the emissions from fossil fuels and biomass burning, respectively.

[43] The weekday-weekend differences in the distributions of BC and OC were significant. Overall, the decrease by ∼20–30% in the ambient concentration of BC was consistent with the reductions in traffic volume during the weekend. This study suggests that the implementation of stricter abatement rules in vehicular emissions could substantially reduce the levels of various anthropogenic pollutants in Bangkok.

Acknowledgments

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Measurements
  5. 3. Measurement Site and Emissions Sources
  6. 4. General Circulation and Variations in Meteorological Parameters
  7. 5. Temporal Variations of BC and OC Aerosols
  8. 6. Emission Ratios of BC, CO, and OC
  9. 7. Weekend Effect on the Concentrations of BC and OC
  10. 8. Comparison With Other Urban Areas
  11. 9. Summary and Conclusions
  12. Acknowledgments
  13. References
  14. Supporting Information

[44] The authors thank S. Han and Yu Morino (National Institute of Environmental Studies, Japan) for their support for the experiments. Do Thi Thanh Canh and colleagues in Asian Institute of Technology are thanked for their contribution in the day-to-day operation of the equipment and data collection. This work was supported by the Asia Pacific Network; the Ministry of Education, Culture, Sports, Science, and Technology; the strategic international cooperative program of the Japan Science and Technology Agency; and the global environment research fund of the Japanese Ministry of the Environment (A-083).

References

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Measurements
  5. 3. Measurement Site and Emissions Sources
  6. 4. General Circulation and Variations in Meteorological Parameters
  7. 5. Temporal Variations of BC and OC Aerosols
  8. 6. Emission Ratios of BC, CO, and OC
  9. 7. Weekend Effect on the Concentrations of BC and OC
  10. 8. Comparison With Other Urban Areas
  11. 9. Summary and Conclusions
  12. Acknowledgments
  13. References
  14. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Measurements
  5. 3. Measurement Site and Emissions Sources
  6. 4. General Circulation and Variations in Meteorological Parameters
  7. 5. Temporal Variations of BC and OC Aerosols
  8. 6. Emission Ratios of BC, CO, and OC
  9. 7. Weekend Effect on the Concentrations of BC and OC
  10. 8. Comparison With Other Urban Areas
  11. 9. Summary and Conclusions
  12. Acknowledgments
  13. References
  14. Supporting Information
FilenameFormatSizeDescription
jgrd17093-sup-0001-t01.txtplain text document1KTab-delimited Table 1.
jgrd17093-sup-0002-t02.txtplain text document0KTab-delimited Table 2.
jgrd17093-sup-0003-t03.txtplain text document1KTab-delimited Table 3.

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.