Final Report - Environmental Protection Department

Loading...

Final Report

Hong Kong and the Pearl River Delta Pilot Air Monitoring Project: Pilot study on the use of atmospheric measurements to manage air quality in Hong Kong and the Pearl River Delta

Project 2: Fine Particulate Matter (PM2.5) in the Pearl River Delta

Prepared By: Michael Bergin, Georgia Institute of Technology Gayle Hagler, Georgia Institute of Technology Lynn Salmon, California Institute of Technology Mei Zheng, Georgia Institute of Technology William Chameides, Georgia Institute of Technology C.S. Kiang, Peking University James Schauer, University of Wisconsin at Madison Jian Yu, Hong Kong University of Science and Technology

August 13, 2004

Table of Contents 1.0 Project Summary 2.0 Introduction 3.0 Background 4.0 Field Measurements 4.1 Sampling Sites and Schedule 14 4.2 Sampling Methodology and Analyses 15 5.0 Results and Discussion 5.1 PM2.5 Mass Concentration and General Chemical Composition 16 5.1.1 Annual Mean PM2.5 Concentrations 16 5.1.2 Daily Concentrations During October, 2002 18 5.1.3 Daily Concentrations During December, 2002 19 5.1.4 Daily Concentrations During March, 2003 22 5.1.5 Daily Concentrations During June, 2003 24 5.2 Sub-regional Sources within the PRD 26 5.2.1 PM2.5 and Organic Carbon 26 5.2.2 Sulfate 27 5.2.3 Crustal Species 28 5.2.4 Mobile Sources 29 5.2.5 Biomass Burning 30 5.3 Concentrations of Organic Tracer Compounds and CMB Modeling 31 5.3.1 Organic Carbon and Elemental Carbon 32 5.3.2 Organic Tracers in Monthly Aggregate PM2.5 Samples 33 5.3.2.1 Hopane and Sterane (H+S) Concentrations 34 5.3.2.2 Levoglucosan Concentrations 35 5.3.2.3 PAH Concentrations 36 5.3.3 Organic Tracers in Episodic High Volume Samples 36 5.3.4 Source Apportionment of Primary Fine Particulate Organic Carbon 5.3.4.1 Biomass Burning 38 5.3.4.2 Meat Cooking 39 5.3.4.3 Diesel Exhaust 39 5.3.4.4 Gasoline Exhaust 40 5.3.4.5 Incomplete Combustion of Coal 41 5.3.4.6 Road Dust 41 5.3.4.7 Cigarette Smoke 42 5.3.4.8 Vegetative Detritus 42 5.3.4.9 Organic Carbon Not Accounted for by CMB and Water Soluble Organic Carbon (WSOC) 42 6.0 Acknowledgements 7.0 References

4 7 8 14 16

37

43 45

1

List of Tables Table 1. 24-hour sample completeness during field study Table 2. Intercomparison results for Caltech and ThermoAndersen samplers conducted at Shenzhen on September 9, 2002 Table 3. Intercomparison results for Hong Kong and Shenzhen ThermoAndersen samplers conducted on September 11, 2003 Table 4. Seasonal and annual mean concentrations across the PRD Table 5. Correlation of PM2.5 concentrations between sites Table 6. Correlation of organic carbon concentrations between sites Table 7. Correlation of sulfate concentrations between sites Table 8. Correlation of silica concentrations between sites Table 9. Correlation of iron concentrations between sites Table 10. Correlation of elemental carbon concentrations between sites Table 11. Correlation of lead concentrations between sites Table 12. Correlation of potassium concentrations between sites Table 13. Correlation of rubidium concentrations between sites Table 14. Mean annual concentrations of species at each site Table 15. Enrichment factors of elements in PM2.5 samples Table 16. Molecular markers for major sources Table 17. Concentrations of organic compounds, organic carbon and elemental carbon in monthly aggregated PM2.5 samples Table 18. Source contributions to organic carbon in PM2.5 Table 19. Percentage of source contributions to measured organic carbon in PM2.5

48 48 48 49 50 50 51 51 52 52 53 53 54 55 55 56 57 59 60

List of Figures Figure 1. Sampling sites in the Pearl River Delta Figure 2. Schematic diagrams of the ThermoAndersen and Caltech samplers Figure 3. Annual mean concentrations across the Pearl River Delta Figure 4. Monthly mean concentrations during October, 2002 Figure 5. Monthly mean concentrations during December, 2002 Figure 6. Monthly mean concentrations during March, 2003 Figure 7. Monthly mean concentrations during June, 2003 Figure 8. Daily concentrations during October, 2002 Figure 9. Daily concentrations during December, 2002 Figure 10. Daily concentrations during March, 2003 Figure 11. Daily concentrations during June, 2003 Figure 12. Wind speed, direction, and precipitation during October 1 – 2, 2002 Figure 13. Wind speed, direction, and precipitation during October 7 – 8, 2002 Figure 14. Wind speed, direction, and precipitation during October 13 – 14, 2002 Figure 15. Wind speed, direction, and precipitation during October 19 – 20, 2002 Figure 16. Wind speed, direction, and precipitation during October 25 – 26, 2002 Figure 17. Wind speed, direction, and precipitation during Nov. 30 – Dec. 1, 2002 Figure 18. Wind speed, direction, and precipitation during December 6 – 7, 2002 Figure 19. Wind speed, direction, and precipitation during December 12 – 13, 2002 Figure 20. Wind speed, direction, and precipitation during December 18 – 19, 2002 Figure 21. Wind speed, direction, and precipitation during December 24 – 25, 2002 Figure 22. Wind speed, direction, and precipitation during Feb. 28 – March 1, 2003

61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 2

Figure 23. Wind speed, direction, and precipitation during March 6 – 7, 2003 Figure 24. Wind speed, direction, and precipitation during March 12 – 13, 2003 Figure 25. Wind speed, direction, and precipitation during March 18 – 19, 2003 Figure 26. Wind speed, direction, and precipitation during March 24 – 25, 2003 Figure 27. Wind speed, direction, and precipitation during June 4 – 5, 2003 Figure 28. Wind speed, direction, and precipitation during June 10 – 11, 2003 Figure 29. Wind speed, direction, and precipitation during June 16 – 17, 2003 Figure 30. Wind speed, direction, and precipitation during June 22 – 23, 2003 Figure 31. Wind speed, direction, and precipitation during June 28 – 29, 2003 Figure 32. A matrix of PM2.5 measurements paired by site Figure 33. A matrix of organic carbon measurements paired by site Figure 34. A matrix of sulfate measurements paired by site Figure 35. A matrix of silica measurements paired by site Figure 36. A matrix of iron measurements paired by site Figure 37. A matrix of elemental carbon measurements paired by site Figure 38. A matrix of lead measurements paired by site Figure 39. A matrix of rubidium measurements paired by site Figure 40. A matrix of potassium measurements paired by site Figure 41. Measured source profiles for specific organic tracer compounds Figure 42. Seasonal and spatial distribution of elemental and organic carbon Figure 43. Seasonal and spatial distribution of the ratio of elemental carbon to organic carbon Figure 44. Ratios of elemental carbon to organic carbon from different source tests Figure 45. Seasonal and spatial distribution of hopanes and steranes Figure 46. Seasonal and spatial distribution of the ratios of hopanes and steranes to organic carbon and elemental carbon Figure 47. Seasonal and spatial distribution of levoglucosan Figure 48. Spatial and seasonal distribution of PAHs Figure 49. Seasonal and spatial distribution of normalized PAHs Figure 50. Source contributions to fine particulate organic carbon in PM2.5 Figure 51. Comparison of WSOC and CMB estimated other % carbon concentrations

83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 104 105 106 107 108 111 112 113 114 115

List of Appendices Appendix A: Site Selection and Sampling Methodology Appendix B: Standard Operating Procedures for Sample Preparation, Handling, and Analysis Appendix C: Mass and Carbon Data Appendix D: Ions Data Appendix E: Metals Data from X-Ray Fluorescence Appendix F: Characterization of Organic Compounds in PM2.5 Collected by High Volume Samplers at Tung Chung and Yuen Long Appendix G: Chemical Characterization of Water-Soluble Organic Aerosols in the Pearl River Delta Region

3

1.0 Project Summary As part of the Pilot Air Monitoring Project, sampling of fine particulate matter (PM2.5) took place at seven stations across the Pearl River Delta from October, 2002 to June, 2003. The sampling consisted of 3 urban locations (Shenzhen, Guangzhou, and Central and Western in Hong Kong), 2 background stations (Tap Mun and Conghua), and 2 receptor sites (Zhongshan and Tung Chung). Sampling involved the collection of filter samples over 24-hr integrated time periods during a month of each season. For a given month samples were collected every sixth day. The filters were used to determine the PM2.5 concentration as well as the concentrations of a variety of ions, elemental and organic carbon, specific elements, and solvent extractable organic compounds. The site-to-site variability of various elements and compounds related to specific sources (e.g. coal combustion, wind-blown dust, biomass burning etc.) is used to infer local and sub-regional influences within the PRD. In addition, the relative concentrations of specific organic compounds are used to infer sources and estimate source contributions via a chemical mass balance model (CMB). A summary of the findings is as follows: •

The highest annual mean PM2.5 concentration is observed in Guangzhou (71 µgm-3), followed by Shenzhen (47 µgm-3), Zhongshan (46 µgm-3), Conghua (37 µgm-3), Central and Western (34 µgm-3), Tung Chung (32 µgm-3), and Tap Mun (29 µgm-3).



The annual mean contributions to PM2.5 mass are dominated by organic compounds and sulfate, which are responsible for 24-35% and 21-32% of the fine particulate mass across the PRD region, respectively. Lesser contributions are from elemental carbon (3-8%), crustal species (3-13%), ammonium (6-8%), and nitrate (1-6%).



During the October sampling season surface wind patterns indicate many days are dominated by northerly flow. The highest observed monthly mean PM2.5 concentration of 66 µgm-3 occurred at Guangzhou, which is nearly two-fold greater than the values measured at Conghua and at the Hong Kong sites.



During December the highest PM2.5 monthly mean concentration is also measured in Guangzhou (72 µgm-3), although the second highest concentration is found at Zhongshan (59 µgm-3). On two of the five sampling days during December, Zhongshan has higher fine particulate concentrations than Guangzhou. On these days surface meteorological data indicate northerly flows throughout the PRD. Taken together, these results suggest that in addition to Guangzhou there is a significant source of PM2.5 in Guangdong between Guangzhou and the Zhongshan receptor site.

4



Relatively high fine particulate concentrations are observed throughout the PRD in March, with the two highest 24-hr PM2.5 concentrations measured during the entire study of ~ 120 µgm-3 occurring at Guangzhou. The monthly mean concentration at Conghua is 51 µgm-3, which is 10-15 µgm-3 greater than values observed at the Hong Kong sites. On many of the sampling days wind speeds are relatively low (< 4 ms-1) and shifting from both the north and south indicating more regional influences on local air quality.



During the summer sampling period in June monthly mean concentrations are relatively low at all sites (~25-35 µgm-3) with the exception of Guangzhou with PM2.5 of 64 µgm-3. Given the intensity of precipitation scavenging from heavy rains that occurred during the summer monsoon, it is likely that most locations were significantly influenced by local sources.



The square of the correlation coefficients (r2) among the Hong Kong stations and Shenzhen for PM2.5 are > 0.70, suggesting that these sites are influenced to a large extent by sources outside of their region and to a lesser extent by local sources. The Guangzhou and Conghua sites are not correlated with any other stations, indicating that they are influenced to a large degree by local sources.



Organic carbon has r2 values between the Hong Kong and Shenzhen stations that range from 0.61-0.75 suggesting that these sites are influenced both by source regions outside of their region as well as local and perhaps secondary organic aerosol formation.



Sulfate is highly correlated between the Hong Kong sites, Shenzhen and Zhongshan (r2 ranges from 0.82-0.93) indicating that these locations are influenced by the same source region. Given the relatively high concentrations of sulfate observed in Guangzhou city, these results suggest a significant source of sulfur, most likely the burning of coal, in the vicinity of Guangzhou.



The crustal species Silica and Iron have relatively high correlations (r2 values ranging from 0.76 to 0.97) and nearly identical concentrations at the Hong Kong sites indicating transport of dust into the region. The relatively high concentrations at Guangzhou as well as at Conghua suggest that both long range transport as well as sources in the vicinity of Guangzhou influence regional dust concentrations.



Comparison of elemental carbon measurements between the sites suggests that all sampling locations are significantly influenced by both local and regional sources of EC.



Lead concentrations are similar throughout Hong Kong and at Shenzhen (annual means ranging from 0.06 to 0.08 µgm-3) and highly correlated. The annual mean concentration in Guangzhou is nearly three times the concentrations measured in Hong Kong and not correlated with other sites, suggesting local sources dominate. Together, the results suggest a significant regional source of lead in Guangzhou, perhaps from gasoline combustion. 5



Concentrations of potassium, an element associated with biomass burning, are nearly identical (annual mean of 0.57 µgm-3) and highly correlated in Hong Kong. A high correlation also exists between the Hong Kong sampling sites and Shenzhen. The concentrations in Guangzhou are nearly three-fold the values in Hong Kong and Shenzhen and are not correlated with these sites. This suggests that Hong Kong and Shenzhen are influenced by similar source regions and that there are significant biomass burning sources throughout Guangdong.



The ratio of the concentrations of hopanes + steranes (H+S), compounds linked with mobile source emissions, to EC are generally highest in Guangzhou and lowest at the Hong Kong stations. These results when taken together with the relative concentrations of Pb at the sites suggest that Guangzhou is influenced to a greater extent by gasoline vehicles and Hong Kong by diesel vehicles.



Concentrations of Levoglucosan, a biomass burning tracer, are on average highest in Guangzhou (508 ngm-3), and Conghua (351 ngm-3) and somewhat uniform throughout Hong Kong (110-160 ngm-3). This indicates significant biomass burning sources in Guangdong, and is consistent with the observed concentrations of K.



Based on Chemical Mass Balance (CMB) modeling, on average across all seasonal samples and all stations the main sources of primary fine particulate OC are mobile sources that are estimated to account for 21±6% of the OC mass (with diesel vehicles contributing 13±6% and gasoline vehicles 8±4%) followed by 18±4% for biomass burning and 12±6% for meat cooking. Although, it should be noted that relative contributions vary with season and site.



Lesser annual contributions to particulate OC at all sites are attributed by the CMB model to coal combustion (7±4)% and road dust (6±1%). Although, it should be noted that the road dust contribution is an upper limit estimate as dust at many of the sites appears to be from regional and longer range transport and possibly not linked with road dust.



Caution should be used when interpreting CMB results since key source profiles may not be entirely representative of actual source emissions. This is particularly true for coal burning, which is evident by the high sulfate concentrations observed throughout the PRD. Mobile source profiles of diesel and gasoline vehicles may also vary throughout the PRD. In addition, the meat cooking and biomass burning profiles used may not be representative of the PRD and this again may influence the interpretation of CMB results. It is also possible that we do not account for all significant sources of primary OC (i.e. marine vessels). Hence, it is clear that future studies are needed to assess the emissions profiles of the major sources identified in this study throughout the PRD.



Despite lack of knowledge of emissions profiles used in the CMB model the results taken collectively suggest that the combined influence of fossil fuel combustion from mobile sources and coal as well as 6

biomass burning account for on average ~ 40-50% of the primary fine particulate organic carbon mass across the PRD in the samples analyzed during this study. 2.0 Introduction The Pearl River Delta and adjoining Hong Kong metropolitan area, like virtually all other major urbanindustrialized regions of the world, suffers from photochemical smog and the unhealthy concentrations of ozone and fine particles that it engenders. The characteristic of smog that is readily apparent to the unaided observer is the low visibility or haziness caused by the tiny particles or fine particulate matter suspended within the smog. Less readily apparent but equally ubiquitous in smog is the high concentration of ground level ozone (O3) generated from photochemical reactions between gaseous pollutants. Both fine particles and O3 are toxic and their presence in photochemical smog represents a health risk to a significant segment of the exposed population, especially the elderly, the young, and the infirm. To protect the health and to continue to improve the quality of life of its citizens and visitors, officials from the Pearl River Delta region recognize that it must improve its air quality by reducing the severity of photochemical smog. However, the control of photochemical smog is complex and depends on a multitude of factors, including meteorological conditions that foster smog formation (sea-breeze effects, winter monsoon circulations) and the transport and co-mingling of pollutants from distant and local sources. The subject Pilot Air Monitoring Project was initiated in May, 2002 with a duration of 2 years. The Environmental Protection Department (EPD) project AS:02-011 is part of this larger project being conducted by Civic Exchange. It is intended to build upon the useful work already done by the Hong Kong EPD, the universities in Hong Kong, and their counterparts in South China. The study is designed to explore the use of observation-based methods, analyses, and models based upon new measurements of NOx, VOC, ozone, and the chemical compositions of the atmospheric fine particles to: (1)

Fill in knowledge gaps in smog and visibility problems in Hong Kong and the PRD region;

(2)

Provide insight into policy relevant questions that could assist policy-makers in managing regional air quality in Hong Kong and South China;

7

(3)

Strengthen the ability of public sector agencies, private businesses, and the academic scientific community to develop policies to improve air quality management; and

(4)

Build long-term air quality management capacity in Hong Kong and Mainland China.

This report focuses on fine particulate (PM2.5) measurements made at seven stations across the PRD over a roughly one year period beginning during October, 2002 and ending during June, 2003. The fine particle mass concentrations a well as specific chemical composition are presented and discussed. The site-to-site variability of the concentrations of specific compounds and elements is used to infer sub-regional sources within the PRD. Also, the concentrations of solvent extractable organic compounds are used to estimate the contributions of a variety of sources to organic carbon concentrations. In addition, surface meteorological data is integrated into the study to gain a better understanding of the impact of meteorology on PM2.5 concentrations as well as additional insight into the sub-regional influences of source emissions on concentrations throughout the PRD. 3.0 Background Anthropogenic fine particles are a growing concern in urban regions worldwide for their immediate and long-term effects on human health and the environment. Fine particles, also known as PM2.5, are produced from both natural processes and human activities. The primary source of anthropogenic PM2.5 is combustion, which directly emits particles as well as compounds that eventually condense onto existing particles. With lifetimes of days to weeks, elevated levels of fine particles are both a local and regional problem for many urban areas worldwide. Fine particles have a significant impact on human mortality, associated specifically with short and longterm respiratory distress (13). Not only do the small sizes of the particles allow penetration to the inner depths of the lungs, but a high surface area allows urban pollution contaminants sorbed to the surface of the particles to cause damaging inflammation (4). A recent study by Pope et al. (2002) linked a long-term 10 µg m-3 increase in PM2.5 with a 4%, 6%, and 8% increased risk in all-cause, cardiopulmonary, and lung cancer mortality, respectively. Another important impact of fine particulate matter is its influence on climate. Aerosol particles affect the climate directly by scattering and absorbing solar radiation, and indirectly by serving as nuclei for cloud 8

droplet formation, causing increased cloud lifetimes and albedo (5-7). The net effect of fine particle climate forcing is believed to be a general regional cooling trend (6,7), although soot particles may locally warm the atmosphere through the absorption of solar radiation (32). Side effects of fine particle attenuation of solar radiation include observable reductions in visibility and possible negative impacts on photosynthesis and crop production (8). The Pearl River Delta region of Guangdong province in southern China, home to nearly 40 million inhabitants, is growing and industrializing at an incredible pace. In fact, the southern city of Shenzhen has grown from a small fishing village to a large city of about 4 million in only the past 20 years. The topography of the region is complex, with mountains bordering the north of the region and each flank of the delta. Hong Kong is located south of the Pearl River Delta, containing a population of nearly 7 million clustered in the small valley regions of its mountainous main peninsula and smaller islands. Particulate pollution is a recognized problem in the Hong Kong and Pearl River Delta region. While fine particles are currently still unregulated, a broader size class, respirable particulate matter (PM10), is limited in both regions. Daily and annual limits for PM10 in Hong Kong are 180 µgm-3 and 55 µg/m3, respectively. The Pearl River Delta area has regulations particular to two separate emission classes, nonindustrial regions (150 µgm-3 daily, 100 µgm-3 annual average) and industrial regions (250 µgm-3 daily, 150 µgm-3 annual average) (25). Several recent studies have shown that the inhabitants of both regions are regularly exposed to elevated levels of PM10. Lee and Chang (2000) monitored indoor and outdoor levels of PM10 for five classrooms in Hong Kong from November 1997 to January 1998. Indoor and outdoor average PM10 measurements for all five classrooms exceeded Hong Kong’s annual standard, with one site having a maximum outdoor concentration of 617 µgm-3. Although indoor concentrations are also influenced by sources within the measurement environment and not always by outdoor concentrations, such extremely high measurements were attributed to local emissions, such as construction, industrial processes, and vehicle emissions. Chan et al (2002) measured commuter exposure to PM10 for various public transportation options in Hong Kong, finding the highest levels of PM10 (147 µgm-3) for non-air-conditioned roadway transport and lowest levels (50 µgm-3) for railway transport. Finally, a recent PM10 monitoring study was carried out across the PRD region by the environmental engineering company CH2MHill from December 1999 to May 2000. PM10 concentrations were found to be highest in the cities of Guangzhou (267 µgm-3) and Shenzhen (114 µgm-3), with values ranging from 50 – 86 µgm-3 in Hong Kong. 9

Increasing interest in the fine particulate subset of PM10 has led to several recent studies of PM2.5 in the Hong Kong and Pearl River Delta region, with nearly all studies showing fine particle levels far exceeding the current U.S. National Ambient Air Quality Standard (NAAQS) of 15 µgm-3 (annual average) and 65 µgm-3 (24 hour average). In the Pearl River Delta region, Wei et al. (1999) measured ambient PM2.5 for two elementary schools in Guangzhou, one urban and one rural, collecting 15 samples per season for 1995 through 1996. The urban site was found to have an annual mean of 132 µgm-3 and 117 µgm-3 for 1995 and 1996, respectively. The suburban site had annual means of 68 µgm-3 (1995) and 54 µgm-3 (1996). The highest daily concentration reported was 506 µgm-3, measured at the urban site in 1995. A more specific study by Chan et al. (2002) measured exposure to PM2.5 from various transportation methods in Guangzhou, sampling during 5 consecutive weekdays for May and December 2001. The non-air-conditioned bus had the highest mean PM2.5 of 145 µgm-3, followed by non-airconditioned taxi (106 µgm-3), air-conditioned bus (101 µgm-3), air-conditioned taxi (73 µgm-3), and subway (44 µgm-3). Chan et al.(2001) measured PM2.5 at 11 roadside sites throughout Hong Kong from January 1999 to February 1999, classified as either urban industrial, new town, urban residential, or urban commercial. Mean PM2.5 levels ranged from 50 µgm-3 to 125 µgm-3. Chao and Wong (2002) compared indoor and outdoor exposure to PM2.5 by measuring fine particles at 34 homes in residential areas throughout the Hong Kong region from October 1999 to March 2000. Mean PM2.5 indoor and outdoor levels were 45 µgm-3 and 47 µgm-3, respectively. A wintertime measurement of PM2.5 in Hong Kong was conducted by Ho et al (2003), with samples taken at two urban sites (PolyU and Kwun Tong) and an urban background site (Hok Tsui) from November 2000 to February 2001. The average concentration ranged from 42 to 57 µgm-3, with a high of 117 µgm-3 at Kwun Tong. In addition, the Hong Kong Environmental Protection Department (HKEPD) led a recent long-term measurement campaign at three sites in Hong Kong, from November 2000 to October 2001 (30). Annual averages for Hok Tsui (rural), Mong Kok (urban roadside), and Tsuen Wan (urban), were 24, 58, and 34 µgm-3, respectively. Seasonal averages ranged from 15-32 µgm-3 (HT), 51-69 µgm-3 (MK), and 26-44 µgm-3 (TW). A recent broader field campaign measured particulate matter in both Hong Kong and the Pearl River Delta. Cao et al. (2003, 2004) measured both PM2.5 and PM10 at 8 sites in the Pearl River Delta region during the winter and summer of 2002. Measurement locations included the cities of Hong Kong, Guangzhou (PRD), Shenzhen (PRD), and Zhuhai (PRD). Peak concentrations were measured in the 10

winter samples, with Guangzhou having the highest average concentration of 106 µgm-3, followed by Shenzhen (61 µgm-3), Zhuhai (59 µgm-3), and Hong Kong (55 µgm-3). These results suggest that at all sites the annual mean concentrations will very likely exceed the U.S. annual NAAQS of 15 µgm-3. With detailed analysis of the carbonaceous fraction, Cao et al (2004) predicted a regional influence on concentrations of organic carbon and observed seasonality in organic carbon and elemental carbon concentrations. Sources of fine particles are currently poorly understood for this region, complicated by the mountainous topography, sea-land breezes (13-15), urban street canyon effect (16), and complex monsoon meteorology (17-19). Overall, Hong Kong displays maximum fine particulate levels in winter and minimum levels in the summer, usually explained by winter northeasterly winds and summer heavy rains (17-24). The most recent emissions inventory, carried out by CH2MHill, estimated a total PM10 production for the region of 258 kilo tonnes/year, with the PRD region producing 95%, and Hong Kong producing the remaining 5%. Specific contributions within each region were energy, industry, and transportation. In Hong Kong the leading source of PM10 was estimated to be transportation related sources, while in Guangdong industry is believed to make the largest contribution to PM10 (29). Specific emissions inventories for PM2.5 have yet to be determined. Chemical analyses of measured fine particulate matter have yielded important insights about their composition and sources. Several studies have found enrichment of pollutant elements in PM2.5 versus PM10, indicating that fine particles are more toxic and more heavily impacted by anthropogenic pollution, including compounds from incomplete combustion and toxics from industrial processes (12, 27, 30, 31, 36). In its year-long study of three sites in Hong Kong, the HKEPD found the most abundant constituents of PM2.5 to be ammonium sulfate, organic carbon (OC), and elemental carbon (EC), with crustal material only a small fraction (3-8%) of its total mass (30). At the urban Tsuen Wan site it was found that the contribution of carbonaceous aerosol (organic carbon plus elemental carbon) to seasonal PM2.5 concentrations ranged from 39%-50%. Similarly, the study by Cao et al. (2003) found that OC accounted for roughly 40% of the PM2.5 mass during the wintertime in the PRD region. Several early studies targeted particulate organic compounds in the Pearl River Delta region, gleaning information about local and regional sources through the specific organic constituents of the aerosol mass. In 1988, Simoneit (1991) analyzed solvent-extractable organic matter of aerosol samples taken at two sites in Guangzhou, with total extract yields of 5.0 µgm-3 and 26.1 µgm-3 at the suburban and urban site, 11

respectively. Of these extracts, 3.1 µgm-3 and 3.2 µgm-3 were characterized as n-alkanes, PAH, diterpenoids, n-alkanoic acids, n-alkanols, and n-alkanones. Zheng et al (1997) characterized TSP non-volatile solvent-extractable organics at six sites in Hong Kong for December 7-31, 1993. The total extract yield across the six sites ranged from 6.4 to 22.7 µgm-3, with compounds falling into four main categories of n-alkanes, n-fatty acids, n-alkanols, and PAH. The highest yields in PAH (54 ngm-3), UCM (unresolved complex mixture containing branched and cyclic hydrocarbons) (1919 ngm-3), alkanes (550 ngm-3), and alkanoic acids (2104 ngm-3) occurred at the roadside station, Mong Kok. By calculating parameters describing the speciated compounds, anthropogenic, microbial, and biogenic sources were evaluated. Petroleum residues were detected at the three urban sites while the rural sites had a bias towards biogenic and microbial components. Octadecanoic acid was discussed as both a marker for kitchen emissions and as a possible estimation of aged air versus recent biogenesis, as unsaturated fatty acids are known to degrade quickly in the environment. A more detailed study of polycyclic aromatic hydrocarbons (PAH) was performed by Zheng and Fang (1998), analyzing 34 TSP samples from 1993 to 1995 at the same six sites as their previous study. Seasonal variability was identified, with PAH values peaking in autumn and decreasing from winter to summer. Spatially, urban Mong Kok had the highest overall PAH levels and rural sites Hok Tsui and HKUST had the lowest. A unique study was performed by Fang et al in 1996, where two days of sampling characterized as a dust episode were compared with two non-episodic samples. TSP samples were taken at two sites – western residential Lantau Island and eastern rural HKUST. Organic analyses were performed on all samples and distinct trends were noted. The distribution of n-alkanes in the non-episodic samples showed significant petroleum residues, while the episodic samples showed elevated levels of terrestrial vascular plant wax input, increasing 43% at Lantau and 29% at HKUST. Also, the dust episode samples were determined to have low levels of unsaturated fatty acids, evidence that the air mass was aged and had been advected into the region. A few recent studies have focused on organic compounds in the fine mode of particulate matter. Xu and Lee (2000) sampled aerosols from May 18-30, 1998 in order to investigate the fraction of nitrated polycyclic aromatic hydrocarbons (NPAHs) adsorbed onto PM10 versus fine particles of diameter 0.5 µm or smaller. Although concentrations were within the pgm-3 range, fine particles were found to be enriched in NPAHs by 3 to 10 times. A more extensive coverage of solvent-extractable organics in the finer mode 12

was studied by Zheng et al (2000), with 18 24-hour PM2.5 samples taken at HKUST for over one year, from October 1996 to September 1997. With a total solvent extract yield of 56.4-233.6 ngm-3, important categories were fatty acids (46-80%), alkanes (10-34%), alkanols (4-21%), and PAHs (1-6%). Sources were determined to be combustion, biogenic, and microbial, with vehicular emissions determined to be characteristic of the Hong Kong region. On average, 79% of alkanes measured were from petroleum residue. The distribution of alkanes shifted seasonally, its peak moving towards lower carbon numbers in fall and winter suggesting a greater relative influence of diesel emissions during these times of the year. Another factor showing the vehicular trait was a strong correlation between total PAH and benzo(ghi)perylene, an organic species associated with mobile emissions. Fatty acids and alkanols both displayed seasonal variability due to microbial activity, with the contribution from microbial sources increasing during the summer season. The recent year-long PM2.5 measurement campaign led by the HKEPD included two types of organic analysis: solvent extractable organic analysis (SEOC) and water soluble organic analysis (WSOC). WSOC is a relatively unexplored region of PM2.5 analysis, an important aspect in that it involves compounds most likely to impact visibility due to water uptake. The SEOC analysis, carried out by the Hong Kong Government laboratory found the most abundant solvent extractable organics were UCM (3271%), fatty acids (23-56%), aliphatic hydrocarbons (5-21%), alkanols (0-5%), and PAHs (1-3%). Overall, SEOC was found to contain 2.4-5.4% of the total carbon and 3.1-10.6% of total organic carbon. Of the three sites measured, urban Mong Kok displayed the highest overall values but little seasonal variation. Oppositely, rural Hok Tsui had the lowest concentrations but displayed elevated levels in winter and low levels in the summer. Dominant sources were noted to be microbial activities, vascular plant wax, vehicular exhaust, and kitchen emissions. Water soluble organic compounds were measured to contribute 14-64% of the organic carbon mass. Analyses of water soluble organic compounds identified 12-29% as organic anions (4-14%), amino acids (3-8%), aliphatic amines in free and combined form (0.3-0.6%), and carbohydrates (2-7%). Aromatic acids, hydroxy acids, oxo-acids, and dicarbonyls together formed 0.7-3.2% of the WSOC mass. The annual mean ratio of WSOC to total aerosol carbon at remote Hok Tsui, Tsuen Wan, and Mong Kok was 40%, 20%, and 9%, respectively. Since the oxidation of organic compounds in the atmosphere generally leads to greater water solubility, these results suggest that secondary production of particulate organic compounds significantly influence fine particulate concentrations in Hong Kong.

13

Several studies indicate that secondary aerosols, particularly sulfate particles, are transported during winter to Hong Kong from mainland China (11, 18, 22). Ho et al. (2003) found high levels of non-seasalt sulfate at a rural coastal site (Hok Tsui), evidence that the sulfate species must have been advected into the region. The lack of variability of sulfate particles among a wide range of monitoring sites in Hong Kong also shows the regional nature of the sulfate aerosols (11, 18, 30). In one study, air masses moving to Hong Kong via continental China were correlated with increases in sulfate and ammonium concentrations by ~ 40-400% and 30-300% respectively (22). The presence of sulfate in Hong Kong particulate matter indicates that secondary formation of aerosols play an important role in PM2.5 levels in the region. Overall, past studies show that the concentrations of particulate matter across the Pearl River Delta are extremely high during the fall and winter seasons. Levels of PM2.5 have been found to often exceed the U.S. NAAQS 24 hour standard at many locations, as well as the annual average systematically throughout the region. Measurements of the chemical composition of PM2.5 indicates that sulfate, produced primarily from the burning of coal, as well as carbonaceous species which originate from a variety of combustion processes are the primary sources of fine particulate matter. Prior work also indicates that meteorology plays an important role in defining PM2.5 concentrations throughout the region. 4.0 Field Measurements 4.1 Sampling Sites and Schedule Sampling sites were carefully chosen to represent background, urban, and urban influenced sites throughout the region. The 7 sites dedicated to aerosol sampling are presented in Figure 1 and can be summarized as follows: •

2 Background sites located at Tap Mun and Conghua;



2 Receptor sites located at the cities of Zhongshan and at Tung Chung; and



3 Urban sites located at Shenzhen, Guangzhou, and Central and Western.

At each of these 7 sites, PM2.5 mass and chemical composition are measured on selected days (five 24hour integrated samples every 6th day) for one month during each season of the year using integrating filters and subsequent laboratory analysis. Sampling was conducted during the months of October and 14

December, 2002 and March and June of 2003. The criteria used in the selection of each of the sites, as well as descriptions of their characteristics, the general measurement strategy and project timetable are given in Appendix A. Table 1 shows the completeness of sampling over the 24-hr measurement periods at each location. In general, 133 24-hr samples were obtained on schedule (out of a possible 140 samples) yielding a sample completion of 95%. It should be pointed out that the stations operated by the Hong Kong EPD had a 100% sample success rate. In addition to the daily sampling during each season, Hi-volume samples collected by the Hong Kong EPD in coordination with Dr. Jianzhen Yu of HKUST at two locations in Hong Kong ( urban Yuen Long and Tung Chung) have been analyzed for solvent extractable organic compounds for four specific events. The filters were collected over 24-hours on the following dates: 10/20/02, 12/13/02, 12/19/02, 03/25/03. 4.2 Sampling Methodology and Analyses At 3 sites in Hong Kong and 2 sites in Guangdong ThermoAndersen RAAS PM2.5 chemical speciation samplers were used to collect aerosol samples. The samplers are designed with 4 separate channels. Two of the channels were operated at 16.7 lmin-1 with quartz filters that were analyzed for organic (OC) and elemental carbon (EC), and specific water soluble and solvent extractable organic compounds (SEOC) that are used as tracers for specific sources (i.e. mobile sources, biomass burning, coal combustion, etc.). The other two sampling lines collected aerosol particles on Teflon filters at a flowrate of 7.3 lmin-1 that were then used to determine the PM2.5 mass concentration as well as the concentrations of major ions and elements. At two additional sites in Guangdong 2 prototype samplers designed and fabricated at Caltech operated under similar conditions as described above. A detailed description of the samplers, analytical techniques, as was as quality assurance-quality control procedures (QA/QC) is given in Appendix B. In order to insure that the samplers as well as field and laboratory handling procedures did not add significant uncertainty to PM2.5 measurements at each of the sites, field intercomparisons were conducted as part of this study. Logistical considerations made it difficult to compare all samplers simultaneously at one location and hence two separate intercomparisons were conducted during the field study. The first intercomparison took place in Shenzhen between the four samplers that were subsequently deployed in Guangdong on September 9, 2002 in Shenzhen. An intercomparison was also conducted with the three Hong Kong samplers and the Shenzhen ThermoAndersen sampler at HKUST on September 11-12, 2003. In this way the Shenzhen sampler served as the reference sampler for PM2.5 mass concentrations. As 15

shown in Tables 2 and 3, all samplers are within ~5% of the reference sampler PM2.5 mass concentration. The results suggest that differences in samplers as well as the sample handling, shipping and analysis procedures did not significantly add to measurement uncertainties. In particular, slight differences in the cyclone cut-points due to the designs of the samplers and operational flowrates appear to have a negligible (5% or less) influence on fine particulate concentrations. 5.0 Results and Discussion 5.1 PM2.5 Mass Concentration and General Chemical Composition In the following section results are presented for the PM2.5 mass concentration as well as the concentrations of the major ions, elemental and organic carbon (EC and OC), and elements which are used to determine the crustal and trace elemental contributions to the PM2.5 mass. We begin by presenting annual mean concentrations for the field sampling study and follow with the daily concentrations during each month of sampling. The daily samples are discussed in terms of surface meteorological measurements (wind speed, direction, and precipitation). 5.1.1 Annual Mean PM2.5 Concentrations The annual mean concentrations of specific PM2.5 associated species are shown in Figure 3 as well as Table 4. It should be pointed out that the values do not necessarily represent the actual annual mean concentrations since samples were collected only for several days during a month of each season. The highest PM2.5 concentrations are observed in Guangzhou, which has an annual mean value of 71 µgm-3. The concentrations at all of the sites in Guangdong, including the background Conghua site, are greater than PM2.5 concentrations measured in Hong Kong. There is a roughly two-fold increase in PM2.5 concentrations between the Conghua and the urban Guangzhou site. The Hong Kong sites have PM2.5 concentrations that are within ~ 20% of one another, ranging from 29 – 34 µgm-3. This relative uniformity of concentrations across Hong Kong suggests that the city is influenced to some extent by more distant sources, outside of Hong Kong. Overall these general results suggest that Guangdong, and in particular Guangzhou city, is a significant source of PM2.5 in the PRD. The chemical composition throughout the PRD is dominated by organic compounds and sulfate which account for 24 - 35%, and 21 – 32% of the PM2.5 mass, respectively. Lesser contributions to PM2.5, as 16

presented in Table 4, are from crustal species (3 –13%), ammonium (6 – 8%), and nitrate (1 - 6%). Nitrate is generally low at all sites. This is due in part to the fact that the sulfate to ammonium equivalence ratio is nearly always greater than one for all daily samples at all sites. In fact, the sulfate to ammonium equivalence ratio for all samples is 1.43±0.29. This suggests that the aerosols are not neutralized by ammonium and hence the formation of aerosol nitrate is not favored. The portion of the pie charts labeled as ‘trace’ contains the contribution of other measured elements and ions not included as their own categories. A fraction of the ‘trace’ concentrations is due to seasalt aerosol. Generally speaking the ‘trace’ contribution is less than 4% of the fine particle mass at all sites. The annual mean contribution of seasalt aerosol (taken as the contribution from sodium and chloride) is less than 2% of the PM2.5 mass at all sites, indicating that seasalt, which is typically associated with the coarse mode, is not a significant fraction of the fine particulate mass. It should be pointed out that the unaccounted for PM2.5 mass, labeled as ‘other’, ranges from 10 – 21%. This fraction of the fine particulate mass may be due to several reasons including water mass associated with the deposited particulate matter, underestimation of the crustal mass from elemental concentrations, as well as the underestimation of the mass of organic compounds. We have used a multiplication factor of 1.4 times the measured organic carbon concentration to estimate the concentration of particulate organic compounds and it could be that this value represents a lower limit estimate, although this cannot be determined without knowledge of the full speciation of organic compounds. In the following sections (5.1.2 – 5.1.5) we present the daily measurements during each month, and discuss the influence of meteorology (local wind speed and direction as well as precipitation) on the relative concentrations at each of the sites. We also present and discuss the monthly mean concentrations at each of the sites and link these concentrations with observed surface meteorology. We do not have meteorological data for several of the specific aerosol sampling stations (CH, CW, and ZS) and have chosen stations that in general represent flow throughout the region. The location of the met stations is indicated on Figure 1. It should be noted that the surface meteorological data should be taken with some caution with respect to discussions on source regions since surface flow patterns are not always indicative of upper level winds that also influence air pollutant concentrations at the surface. Another important consideration is that the complex topographic structure in the Hong Kong region results in surface winds that are often not indicative more regional flow patterns. As discussed above we have attempted to choose meteorological stations that to a large extent are not influenced by local meteorological effects, such as surface topography, but represent more regional flow patterns. Although, it is apparent that the topography of the region results in a relatively complex relationship between surface winds and regional 17

transport. Nonetheless, we use the meteorological data for qualitative discussions on source regions and meteorological influences. The results clearly indicate that surface winds yield important insights into source regions within the PRD. 5.1.2 Daily Concentrations During October, 2002 Figure 8 shows the daily PM2.5 mass concentrations for the sampling days during October 2002. During the first sampling day of the month (October 2), Guangzhou (GZ) experienced the highest PM2.5 concentrations with lowest concentrations of nearly three-fold less occurring at the background Tap Mun (TM) and Conghua (CH) sites. The concentrations across the sites in Hong Kong are only ~ 30% greater than the concentrations at the background sites. Shenzhen experiences PM2.5 concentrations that are relatively high compared to the Hong Kong sites. Wind speeds and directions throughout the PRD region for October 1-2 (Figure 12) indicate that for the day prior to sampling, winds in the region are primarily northerly. During the day of sampling nearly all sites experienced relatively moderate wind speeds (2- 4 m s-1) with both northerly and southerly components suggesting a land-sea breeze effect throughout the region. During October 8th winds are moderate to high and from the north throughout the PRD region as shown in Figure 13. During this day Zhongshan (ZS) experiences the highest PM.2.5 concentrations, with a 24-hr average concentration of ~ 80 µg m-3 (Figure 8). The concentrations at TM, CW and TC are the highest observed in Hong Kong during the sampling month with values that are ~ 45 µg m-3. The fact that CH has the lowest observed PM2.5 concentration and that ZS and SZ experience higher concentrations than GZ during north-northeasterly flow conditions, combined with the uniformly high concentrations in Hong Kong indicate strong sources of PM2.5 within the Guangdong region most likely in the vicinity of Guangzhou. Furthermore, it does seem likely that there is a significant source of fine particulate matter located in Guangdong south of Guangzhou, north of Shenzhen and Zhongshan, since the concentrations increase as winds traverse south through the region. In contrast to the northerly flow conditions of October 8, October 14 had light winds (< 2 m s-1) with a southerly flow component in addition to a northerly component (Figure 14). On this day the concentrations throughout Hong Kong are uniformly low (~20 µg m-3) with the highest concentration of ~ 100 µg m-3 occurring at Guangzhou. The background CH site has a PM2.5 concentration that is more than a factor of 2 greater than the Hong Kong stations. These results also suggest that sources in and around 18

Guangzhou city are responsible for a significant fraction of the regional PM2.5 concentration, as demonstrated by the ~ 30 µg m-3 increase in fine particle mass from the Hong Kong stations to the background CH site. It should also be pointed out that the extremely high concentrations in GZ are also in part due to the low wind speeds that result in relatively lower mixing heights during the sampling day. On October 20, precipitation occurred at the Guangdong sampling sites as shown in Figure 15. On this day winds are moderate (2 – 4 m s-1) and from the south. The relative PM2.5 concentrations are similar to October 14, when surface flows are also from the south, with the Hong Kong sites having the lowest concentrations and GZ the highest PM2.5. The PM2.5 concentration at CH of 40 µg m-3 is nearly a factor of 4 greater than the HK sites. As with October 14, a ~ 30 µg m-3 increase in PM2.5 occurs as airmasses travel from Hong Kong across the PRD region to CH. These results are another indication that aerosol sources in and around Guangzhou city are responsible for a significant fraction of the PM2.5 in the PRD area. On October 26 surface meteorological data (Figure 16) from the Guangdong stations suggest northerly flow in Guangdong province. Guangzhou experiences the highest PM2.5 concentrations (~50 µg m-3). The concentration in Guangzhou is the lowest measured at that station during the month of October. This is very likely due to the high windspeeds (4-8 ms-1) in Guangzhou that occurred over much of the sampling day which resulted in greater atmospheric mixing and dilution of surface concentrations. As is typical of cases of northerly flow, the CH site has lowest PM2.5 concentration, with uniformly high (~35 µg m-3) concentrations throughout Hong Kong. The monthly mean PM2.5 chemical composition across the PRD is shown in Figure 4. Fine particulate mass is dominated by organic compounds and sulfate which account for from 30-41% and 21-32% of the fine particulate mass, respectively. The combined mass fraction of organic compounds and sulfate ranges from 57-67%. Crustal material also has a significant contribution to fine particle mass, accounting for 1015% of the PM2.5. Lesser mass contributions are from nitrate (1-3%), ammonium(5-8%), and trace elements(2-4%). 5.1.3 Daily Concentrations During December, 2002 Figure 9 shows PM2.5 concentrations during the month of December 2002. The first sampling day of the month (December 1) is characterized by moderate to low winds from the east-southeast throughout much of the PRD. Surface winds at Guangzhou are low and northerly (Figure 17). Surface meteorological data 19

indicates that on the prior day (November 30) wind speeds are low and generally from the north. On both days (November 30 and December 1) moderate precipitation occurred throughout the PRD region, with somewhat high amounts of rain (~30 mm) in northern Guangdong. On this day Guangzhou experiences the highest concentration (~100 µg m-3) of any of the PRD sites, and the highest concentration observed at any location during the month of December. This is probably due to the low wind speeds in Guangzhou resulting in less atmospheric mixing and less advection of local pollution out of the city. CH has the lowest concentrations, in part due to the relatively high precipitation observed in northern Guangdong. The concentrations at the Hong Kong sites and Shenzhen are relatively uniform at ~ 30 µg m-3. It is difficult to assess the source areas of the PM2.5 concentrations on December 1 due to the shifting winds prior to the sampling day, the relatively low wind speeds, and the precipitation occurring throughout the region. On December 6 and 7 winds in the PRD are generally moderate and from the south, as shown in Figure 18. Although, there is also a northerly component to airflow patterns, suggesting shifting winds at many of the meteorological sites. Light rain occurred at several of the stations. PM2.5 concentrations at GZ are nearly three-fold higher than at the other stations. Concentrations at the background TM station are lower than at CH, likely due to the southerly flow component bringing somewhat cleaner air to TM, and more polluted air from upwind source regions to CH. Concentrations are somewhat uniform at the other sampling sites, with the exception of TC that has the lowest observed concentrations. It is not clear why TC experienced the lowest concentrations, although it may be due to the spatial distribution of rainfall in Hong Kong. As with the case of December 1, shifting winds prevent a clear case to be made for source regions based on upwind-downwind relations. Nonetheless, it is apparent that the Guangzhou area is a dominant source of PM2.5 and related precursors. On December 13 the lowest PM2.5 concentration occurs at the background CH site, with the highest observed concentrations at ZS (Figure 9). The concentrations are uniformly high (~35 µgm-3) in Hong Kong. Figure 19 indicates that winds are moderate and from the north on the day prior to sampling, and predominantly from the north on the day of sampling. There are several hours at the end of the sampling day during which the flow shifted to the east for several of the sampling sites. The ~ 20 µgm-3 increase in PM2.5 concentration from GZ to ZS suggests significant sources of fine particulate matter south of Guangzhou city in Guangdong province.

20

The highest PM2.5 concentrations observed at the Hong Kong stations during the sampling month occur on December 19, with values ranging from 43 – 52 µgm-3. Guangzhou has the highest PM2.5 concentration of ~ 100 µgm-3, followed by Zhongshan with a value of ~ 80 µgm-3. Meteorological data indicates that surface winds are relatively low at most sites on the day of sampling with shifting winds from the north and south. The CH site experiences relatively high concentrations (~50 µgm-3) suggesting an impact from sources to the south. The elevated concentrations throughout the region are likely from the low wind speeds linked with minimal atmospheric mixing, as well as a shifting wind pattern distributing fine particulate mass from the sources throughout the region. On December 25, the lowest concentrations are observed at the background CH site. The concentrations are nearly two-fold lower at CH as compared to the Hong Kong sites and Shenzhen. Concentrations increase from CH to the background TM station by ~ 15 µgm-3. The PM2.5 concentration in Guangzhou is not significantly different than the values in Hong Kong and Shenzhen (~ 35 µgm-3). The highest PM2.5 concentration occurs at Zhongshan. The winds on December 25 are generally strong (> 4 ms-1) and from the north (Figure 21). As is often the case for northerly flow, the concentrations at CH are low with relatively high PM2.5 values in Hong Kong, Shenzhen, and Zhongshan. Once again for a northerly flow case, measurements suggest a significant source of fine particulate matter in Guangdong, between Guangzhou and Shenzhen. The surprisingly low concentrations in Guangzhou are probably due to the extremely high wind speeds, which reach 8 ms-1, resulting in greater than normal atmospheric mixing. Figure 5 presents the monthly mean PM2.5 mass and chemical concentrations for December, 2002. As shown in the figure, the lowest PM2.5 concentration is at CH (28 µgm-3), with the highest concentration of 72 µgm-3 at Guangzhou. The next highest concentration is observed at Zhongshan (59 µgm-3). The concentrations across Hong Kong range from 31 – 38 µgm-3. The tendency of surface winds on the sampling days, as discussed above, is from the north. Although, winds are often from the south for parts of the sampling days. Overall the spatial distribution of PM2.5 concentrations suggests that Guangzhou is a significant regional source of PM2.5. As also discussed above there would appear to be a significant source of PM2.5 and associated precursors between the Guangzhou and Zhongshan stations as is evidenced by the relatively high fine particulate concentrations observed at the Zhongshan receptor site. As with the month of December, sulfate and organic compounds are the dominant fine particulate chemical components accounting for 24-39% and 19-35% of the PM2.5 mass, respectively.

21

5.1.4 Daily Concentrations During March, 2003 Figure 10 shows the daily PM2.5 concentrations for the five 24-hr integrated filter samples during the month of March, 2003. Surface wind speeds and directions as well as precipitation for the month of March are given in Figures 22-26. Surface meteorological data (Figure 22) indicates relatively light winds with both a northerly and southerly component for the first half of the sampling day on March 1, with pronounced moderate winds from the south for the second half of the day. The concentrations in Hong Kong are relatively high (30 – 40 µgm-3), with the highest concentration occuring at Guangzhou (85 µgm-3). The second highest concentration observed throughout the sampling domain is at Conghua, which has a concentration of nearly 70 µgm-3. The concentration at CH is nearly two-fold higher than values measured in Hong Kong. Overall, the light and shifting winds suggest that the sites are influenced by both local pollution as well as regional sources. The location of the regional influences cannot be directly inferred due to the dynamic nature of the surface winds observed on the sampling day. Although, the dramatic increase in fine particulate concentrations between Hong Kong and CH suggests sources in and around Guangzhou contribute to PM2.5 in CH. On March 7, PM2.5 concentrations are relatively low throughout the entire PRD region as is seen in Figure 10. The fine particle concentration in Guangzhou is 40 µgm-3, which is nearly identical to values observed at SZ and ZS as well as at TC. Concentrations at the background CH and TM sites are the lowest values observed over the sampling domain and are ~ 35 µgm-3. The low concentrations are linked with the strong (> 4 ms-1) northerly flow that was observed throughout the region for both the entire sampling day and the day prior to sampling. In addition, light precipitation is observed at both Shenzhen and HKUST on March 6, indicating that precipitation scavenging may also play a role in the relatively low PM2.5 concentrations observed on March 7 at the sampling sites. The distribution of PM2.5 concentrations at the stations on March 13 is similar to the concentrations observed at the stations on March 1. Concentrations across the Hong Kong sites are uniform at 30 µgm-3, with a roughly 10 – 20 µgm-3 increase at SZ and ZS. The highest concentration of 120 µgm-3 is the second highest value observed at any site during the entire study. On this sampling day CH once again experiences the second highest concentration of any of the sampling sites (70 µgm-3). Surface winds (Figure 24) are low and from the north for the first part of the sampling day, and from the south-southeast 22

for the second half of the day. The increase in PM2.5 from the Hong Kong Stations to CH is nearly twofold. Combined with the extremely high concentrations observed in Guangzhou this suggests significant sources in and around GZ. It should be noted that the hypothesized sources of fine particulate matter south of Guangzhou in Guangdong that may be responsible for the high concentrations of PM2.5 observed in ZS during northerly flow conditions, may also be influencing Guangzhou PM2.5 concentrations on March 13. On the March 19 sampling day light precipitation is observed at both Fogang and HKUST (Figure 25). On March 18 several of the stations experience east-northeast flow, while all stations are under the influence moderate northerly flow on March 19. On this day, Zhongshan experiences the highest concentration of any site (~ 60 µgm-3). The fine particulate concentration in Guangzhou is uncharacteristically low at 35 µgm-3. Both TC and SZ have fine particulate matter concentrations that are higher than GZ. It is not apparent why this is the case, although it is possible that scattered precipitation plays a role in the regional fine particulate matter concentrations on March 19. On the March 25 sampling day all stations experience high PM2.5 concentrations. At GZ, the PM2.5 concentration is ~ 125 µgm-3 which is the highest measured value during the field study. Concentrations across Hong Kong are uniformly high and range from 40-50 µgm-3. Conghua has the lowest concentration of 30 µgm-3. As shown in Figure 26, wind speeds are relatively low (< 4ms-1) and variable on the day prior to and the day of sampling. In fact, at many of the sites winds are less than 1 ms-1 for a significant fraction of the sampling day. The meteorological data indicates relatively stagnant conditions in the PRD for much of the sampling period. Combined with the shifting nature of winds this suggests that there is not a great deal of vertical atmospheric mixing, and that advection of clean background air into the region is at a minimal. The monthly mean chemical composition of PM2.5 at each sampling site is presented in Figure 6. Once again organic compounds and sulfate dominate the fine particulate mass accounting for 20-33% and 2031% of the fine particle mass, respectively. Perhaps one of the most striking observations is the high monthly mean concentration at Conghua (51 µgm-3), which is ~10 µgm-3 greater than the concentrations observed across Hong Kong and similar to the concentrations at Shenzhen and Zhongshan. This is due to transport of air from south of CH, resulting from shifting winds from more southerly flow conditions as compared to the prior two sampling months. Guangzhou experiences the highest mean monthly concentration observed at any of the stations of ~ 80 µgm-3. The concentration in Guangzhou is roughly double the fine particle mass concentrations in Hong Kong. The high concentrations observed in 23

Guangzhou, and throughout the region, are due in part to the meteorology which was marked by low and shifting winds. The dramatic increase in PM2.5 from the stations south of Guangzhou to Guangzhou city suggests that local sources and very likely sources just south of the city impact concentrations. Another fact to point out is that the contribution of ‘other’, unidentified compounds to the total fine particulate mass concentration is on average 21±5% across all sites. This is in comparison to values of 8±3% and 11±4% for October and December. It is not entirely clear what is responsible for this roughly two-fold increase in unidentified fraction. The most likely reason for the change in the ‘other’ contribution for March is a shift in the OC correction factor. A factor of 1.4 is applied to all OC measurements to determine the organic compound concentrations, although it cannot be ruled out that the factor varies with season and location. In order to determine the actual correction factor it would be necessary to measure the concentrations of all organic compounds, of which there are several hundreds to thousands. It cannot be absolutely verified at this time that the assumption of a constant OC correction factor is responsible for the increase in the contribution of unidentified compounds to fine particulate mass during March (as well as June which will be discussed below). If it is indeed the reason then the OC concentrations presented during March can be considered as lower limit estimates. 5.1.5 Daily Concentrations During June, 2003 On the first day of the sampling month, June 5, concentrations are relatively high throughout the PRD (Figure 11). The lowest observed concentration is at the Hong Kong background site, Tap Mun, with the highest concentration being nearly two-fold (80 µgm-3) at Conghua. Surface winds are generally from the south across the PRD, although there are times of extremely low wind speeds (< 1ms-1) as well as periods of light northerly flow (Figure 27). Given the dramatic increase in concentration of roughly two-fold from TM to CH, it is not surprising that winds are primarily from the south during June 5. Given that the winds have both a northerly and southerly component does suggest that multiple source areas influence concentrations across the PRD during the sampling day. During June 10-11 strong rains occurred throughout the PRD, as shown in Figure 28. Surface wind patterns were generally from south, although several sites also had northerly flow components. As shown in Figure 11, the concentrations at all sites, with the exception of Guangzhou, are low due to precipitation scavenging of particulate matter. Once again the concentration at CH (30 µgm-3) is more than twice that at TM, which is linked with more polluted air from the south being advected into the region. The relatively 24

high concentration in Guangzhou is likely due to local sources as well as transport from the hypothesized source region in Guangdong to the south of Guangzhou. On June 17, The lowest fine particle concentrations are measured at the CH site (~30 µgm-3). The concentrations at all other sites, besides Guangzhou, are somewhat uniform ranging from 50-60 µgm-3. The PM2.5 concentration in Guangzhou is 100 µgm-3, which is the highest value measured at any of the sites for the month of June. In general the concentrations at the Hong Kong sites, Zhongshan and Shenzhen are ~ 30 µgm-3 higher than at CH. According to the surface meteorological data (Figure 29) precipitation occurred at several of the sites on the day prior to sampling. Surface wind patterns were primarily from the north on the day of sampling which is not surprising given the differences in concentrations between CH and the more southerly sampling sites. The last two sampling periods (June 23 and June 29) are characterized by very similar spatial distributions of fine particulate concentrations (Figure 11) as well as surface meteorology (Figures 30 and 31). On both days moderate southerly winds dominate surface meteorology, with rain throughout the PRD region. Concentrations are low and similar (10-15 µgm-3) at all Hong Kong sites as well as at Zhongshan and Shenzhen. The concentration at CH is nearly two times the values at these sites, with Guangzhou experiencing the highest fine particulate concentrations of any of the monitoring sites. Once again, results indicate a significant source of fine particulate mass in Guangdong between the sites south of Guangzhou and Conghua. The relatively low concentrations observed throughout the region are due to precipitation scavenging. Figure 7 presents the monthly mean PM2.5 mass and chemical composition for the month of June, 2003. The Tap Mun station has the lowest observed concentrations, with CH being ~12 µgm-3 higher. Concentrations at the other Hong Kong sites as well as Zhongshan and Shenzhen are somewhat similar ranging from 29-35 µgm-3. The monthly mean concentration in Guangzhou is nearly twice that of the more southerly stations. As with the previous sampling months, the chemical composition of fine particulate matter is dominated by organic compounds and sulfate which account for 19-30% and 2232% of the PM2.5 mass, respectively. Overall, the increase in fine particulate matter seen from the more southerly stations (CW, TC, TM, SZ, and ZS) to Guangzhou and Conghua once again suggests significant sources of fine particulate matter in the vicinity of Guangzhou.

25

5.2 Sub-regional Sources within the PRD In this section we explore the site-to-site relationships between the concentrations of specific compounds and elements from various sources in order to gain an understanding of how sub-regions within the PRD affect one another. This is accomplished by determining the squares of the correlation coefficients (r2) for a given tracer between all sites for the entire sampling period. Thus r2 values are based on linear regressions of 20 data points, representing five 24-hr samples collected during each of the four seasons. Based on r2 values, the relative contributions from local and distant sources is inferred. High r2 values (>0.9) between sites suggest that concentrations are not significantly influenced by local sources, while low values (<0.3) suggest local sources dominate. Values that are in the range from 0.5-0.9 indicate a combination of local and more regional sources for a specific compound. It should be pointed out that a site can have a relatively low correlation coefficient with another site and still be influenced by emissions from that site. This is due to the fact that the measurement site in the vicinity of a source will generally be influenced by local emissions while sites further away will be influenced by the particular source (when meteorological conditions are favorable) as well as local sources and other upwind sources when transport is not from the emission source area in question. Hence the combination of sites that have high correlation coefficients for specific species as well as those that have low r2 values, taken together with chemical concentrations and meteorological factors can be used to infer source regions within the sampling network. Table 14 shows the means and standard deviations for the compounds and elements used in the analyses below. Elemental enrichment factors (EF) are presented in Table 15, with values less than five indicating that soil is the main source of the particular element. This is done to rule out windblown dust as the main source of several trace elements. 5.2.1. PM2.5 and Organic Carbon Correlation coefficient values for PM2.5 between all sites are shown in Table 5. An alternative view of correlation between sites is a matrix of scatterplots, each plot linearly fitted via the least-squares method. It should be noted that the scatterplots have varying axis ranges to best show each pairing of sites. The PM2.5 scatterplot matrix is shown in Figure 32. The r2 values for PM2.5 among the Hong Kong stations range from 0.71 to 0.94. Within Hong Kong, Tung Chung is the site having the lowest correlation with the other sites. This suggests that TC is more influenced by local sources than the other Hong Kong sites. Nonetheless, the relatively high r2 values (>0.70) among the sites in Hong Kong is an indication that local PM2.5 has much less of a contribution to fine particle contributions as do sources outside of the region. 26

Shenzhen PM2.5 concentrations are also highly correlated with the Hong Kong stations (r2 values range from 0.74 to 0.83), suggesting that these sites have similar sources. Zhongshan r2 values with the Hong Kong sites and Shenzhen range from 0.56 to 0.77, suggesting that similar sources influence Zhongshan. It is interesting to point out that both Guangzhou and the background CH site have very low r2 values (0.22) with all other stations. This suggests that both of these stations are impacted by local sources. In the case of Guangzhou, which has the highest PM2.5 concentrations measured throughout the sampling network, it is perhaps obvious that local sources impact PM2.5 concentrations. For CH, the lack of strong correlations with other sites may be due to the fact that northerly flow often brings relatively clean background air to the station, with local sources adding to background concentrations. Although organic compounds make up a significant portion of the PM2.5 mass at each site, their correlation coefficients are lower than those calculated for the overall mass, indicating increased local contribution for this specific portion of mass. The OC correlations between sites is shown in Table 6 as well as more visually in Figure 33. Among the Hong Kong sites, r2 values range from 0.67 to 0.75, with the lowest correlation (r2 = 0.67) between Tung Chung and Central/Western. The pairing of Tap Mun and Central/Western has an r2 of 0.75, significantly lower than the r2 value of 0.94 for overall PM2.5 mass. Local contributions of organic compounds are expected to be the cause of this increased variability between the sites in Hong Kong. The formation of secondary organic aerosol also may contribute to decreasing the correlation between the stations. The Guangdong sites in close proximity to Hong Kong, Zhongshan and Shenzhen, also have lower correlation coefficients with the Hong Kong sites for organic compounds, in comparison with r2 values for fine particulate mass. Zhongshan’s pairing with the Hong Kong sites yields r2 values ranging from 0.36-0.46, while the pairing of Shenzhen with the Hong Kong sites results in r2 values of 0.61-0.72. Further north, sites Conghua and Guangzhou show very low correlation overall (average r2 = 0.05), indicating local sources and perhaps the formation of secondary organic aerosol are the dominating cause of variability at these sites. 5.2.2. Sulfate Sulfate is a species found in fine particles that is typically associated with the burning of coal and other high sulfur fuels. Although sulfur emissions inventories are not definitively known in the PRD, a recent study conducted by CH2M Hill (29) found that the combination of the energy and industrial sectors accounted for ~ 93% of all SO2 emissions. A relatively small contribution was found to be due to transportation related sources although it is not entirely clear what role marine vessels, potentially 27

important sources of SO2, play in the PRD. Taken on face value the CH2M Hill report suggests that marine sources are responsible for roughly a few percent of total SO2 emissions in the PRD. Among the sampling sites in this study sulfate contributes 21-32% of the fine particulate mass on an annual average. Correlations performed between sites in the Hong Kong region, shown in Table 7 and Figure 34, demonstrate low variability (r2 = 0.76-0.93). Zhongshan and Shenzhen also show high correlation with the Hong Kong sites and when paired with each other, with r2 values ranging between 0.82 and 0.90. This high sulfate r2 values for the southern five sites in the Pearl River Delta suggest that these locations are influenced by the same source region, with only a modest contributions from localized sources with respect to each station. In comparison, the two northernmost sites, Guangzhou and Conghua, have very low correlations with the five southern sites (average r2 = 0.16) while a higher correlation (r2 = 0.48) is seen between the Guangzhou and Conghua sites. This shows that local sources have strong influences on the sulfate concentrations at the two sites, and both sites at Conghua and Guangzhou may be receiving sulfate emissions from some of the same sources. These results are consistent with the interpretation that source regions within Guangdong, in particular the Guangzhou area, have a significant influence on air quality in southern Guangdong and Hong Kong. 5.2.3. Crustal Species Silica atoms are second only to oxygen in terms of average mass concentration in the Earth’s crust. Found in abundance across the globe, silica is commonly used to determine crustal contribution to fine particle concentrations. In Table 8 and scatterplot matrix 35, a very high correlation is seen among the three sites in Hong Kong, with r2 values ranging from 0.93 to 0.97. Such high r2 values among Tap Mun, Tung Chung, and Central/Western indicate a regional advection of crustal materials to the Hong Kong sites rather than contribution from local sources. Extending comparison to Shenzhen and Zhongshan, moderate correlation (average r2 = 0.77) is seen between the two and in pairing with the Hong Kong sites. It appears that Shenzhen and Zhongshan may have independent sources of crustal material influencing concentrations at their respective locations. This may be the reason that concentrations at these sites (Table 14) are nearly twice the values in Hong Kong. With the highest silica concentrations on nearly every sampling day, Guangzhou shows no correlation with the six other sampling sites (average r2 = 0.05). Such high variability indicates that local sources dominate crustal material found in fine particles at the Guangzhou site. The source of elevated silica in Guangzhou is not entirely clear, although possibilities include construction and changes in land use due to urban development. Though further north than Guangzhou, measured silica concentrations in Conghua have slightly higher correlation with the 28

Hong Kong sites, Shenzhen, and Zhongshan, with r2 values ranging from 0.36 to 0.49. It can be concluded that large-scale transport of crustal material may strongly contribute to background concentrations across the Pearl River Delta region. In looking at the daily concentrations of silica, sampling date October 8th is of note for a near quadrupling of measured silica levels at all sites except for Guangzhou. This date is shown to have no rainfall and moderately high winds from the north. Iron is a second mineral element that is commonly used as a tracer for crustal material in PM2.5. Table 15 indicates enrichment factors for both iron and silica that clearly indicate they are crustal in nature. Like silica, regressions performed on iron concentrations yielded very high r2 values among the three Hong Kong sites (r2 = 0.91 to 0.96), with lower correlation seen in combination with sites Shenzhen and Zhongshan (r2 = 0.53 to 0.79). Guangzhou is again seen to have no correlation with the Hong Kong sites, Shenzhen, and Zhongshan (average r2 = 0.02) though a low correlation (r2 = 0.23) is seen between Guangzhou and Conghua. Conghua has low correlations with the Hong Kong sites, though slightly higher correlations (average r2 = 0.23) with the four sites in Guangdong. As with silica, a spike in iron concentrations is seen again on October 8th. Again, the results suggest a source of crustal material in the vicinity of Guangzhou. 5.2.4. Mobile Sources The concentration of elemental carbon (EC) in fine particles is associated with combustion, in particular of diesel fuel. It should also be noted that EC in PM10 at specific sites in Hong Kong has also been attributed to harbor emissions from shipping (Yu et al., 2004). It can be seen in Table 10 and Figure 37 that elemental carbon varies significantly from one site to the next, with a peak correlation for the region of r2 = 0.66 for the pairing of Tung Chung and Central/Western. Correlation is moderate for the Hong Kong sites, Shenzhen, and Zhongshan, with an average r2 of 0.38. This signifies a strong local influence of mobile sources on elemental carbon at sites in the Hong Kong region. Conghua has virtually no correlation with any other site, with r2 values ranging from 0.00 to 0.06. Guangzhou, with nearly double the amount of elemental carbon compared with the other measured sites (Table 14), has a moderate correlation with Zhongshan (r2 = 0.39), though very low correlation with any other site. Overall, local sources appear to be the significant cause of concentrations of EC at each site. Though lead can be found naturally in the earth’s crust, the use of lead-containing gasoline can cause significantly higher lead concentrations in fine particles. While our crustal tracers, silica and iron, both 29

show a significant peak on the October 8th measuring day, the lead species does not. In addition, the enrichment factor for lead shown in Table 15 is ~6000. Combined with the observation of unnaturally high lead levels across the region, this indicates that anthropogenic sources, presumably combustion of leaded gasoline, dominate lead concentrations. Although, it is possible that other sources, such as lead smelting, may be the cause of the relatively high lead concentrations observed throughout the region. It is important to point out that the high concentrations of lead may not be only due to current lead emissions but also from re-suspension of Pb deposited to the surface. Given the relatively minor role that dust plays in fine particulate concentrations in the PRD it is not clear to what extent re-suspension of anthropogenic Pb contributes to observed lead concentrations. It is shown in Table 11 and Figure 38 that the correlation of lead among the seven sites is very different from the linear regression results of elemental carbon. Though EC has relatively high variability at each site, lead has quite low variability among the three sites in Hong Kong and Shenzhen, with a range of r2 values from 0.73 to 0.96. For these four sites, it is evident that regional transport of anthropogenic lead dominates the local concentrations. Although it should be noted that the annual concentration of Pb at Shenzhen is ~ 30% greater than at the Hong Kong sites indicating the Shenzhen may also contain sources of Pb. Zhongshan has moderate correlation with the Hong Kong sites and Shenzhen (average r2 = 0.45) and very low correlation with Guangzhou and Conghua (average r2 = 0.06), pointing towards a combination of local and regional sources at this site. Guangzhou and Conghua have very low correlations with all other sites (average r2 = 0.08), indicating that local sources strongly influence lead concentrations at each site. It is important to point out that the concentrations of lead at Guangzhou (Table 14) are nearly four times higher than at Hong Kong/Shenzhen indicating that this area is a significant source of lead throughout the region. 5.2.5. Biomass Burning Found in tissues of plants and animals, potassium is an element often used to trace emissions from biomass burning. Potassium is also found in the earth’s crust, but as with lead, has been measured to be at much higher levels than if the element was coming from crustal material alone (Table 15). High correlation is seen of potassium concentrations measured at the three Hong Kong sites and Shenzhen (average r2 = 0.88), shown in Table 12 and Figure 40. With such high correlation, it is expected that emissions of potassium from outside of the region encompassed by these four sites dominates the concentrations. Zhongshan shows lower r2 values when paired with the Hong Kong sites and Shenzhen (average r2 = 0.60) and no correlation with Guangzhou and Conghua (average r2 = 0.03). Guangzhou and Conghua have a lack of correlation with the other sites and with each other, yielding r2 values ranging 30

from 0.00 to 0.21. Local sources are expected to be partially influencing concentrations at Zhongshan and strongly influencing concentrations at Conghua and Guangzhou. It should be noted that the annual mean potassium concentrations at both Conghua and Guangzhou are similar at 1.6 µgm-3 (Table 14), roughly 23 times higher than the Hong Kong sites, suggesting that biomass burning in northern Guangdong is prevalent. Rubidium (Rb) is a second element that is also found in plant and animal tissues, but in much smaller concentrations in comparison with potassium. Due to detection limits in chemical analysis, rubidium has not been commonly used as a source tracer element. For the Pearl River Delta region, however, rubidium concentrations were high enough for detection and proved to be very highly correlated with potassium measurements, when the two were paired at a specific site (average r2 = 0.91). Although it cannot be verified at this time, it is possible that rubidium is also a tracer for biomass burning given that it is incorporated in plant tissues in a similar fashion as potassium. Rubidium may be present in soils in the PRD and/or present in fertilizer that also contains potassium and hence incorporated into plants along with potassium. Nonetheless, here we suggest that rubidium is a tracer for biomass burning and use it along with potassium concentrations to infer sub-regional source areas. As with potassium, Table 13 and Figure 39 show the high correlation of rubidium concentrations within the three Hong Kong sites and Shenzhen, with r2 values ranging from 0.79 to 0.93. Zhongshan shows lower r2 values when paired with the four previous sites, with r2 ranging from 0.39 to 0.46. Guangzhou and Conghua have very low correlation, with an average r2 value of 0.12. Again, the trend is seen of regional influence at Tap Mun, Tung Chung, Central/Western, and Shenzhen, partial local influence at Zhongshan, and strong local influences at Guangzhou and Conghua. 5.3. Concentrations of Organic Tracer Compounds and CMB Modeling In this section the concentrations of a variety of solvent extractable organic compounds are presented and discussed. The compounds are shown in Table 16 and represent a variety of sources including biomass burning, gasoline and diesel combustion, coal combustion, cooking, and road dust. Figure 41 shows measured source profiles of specific solvent extractable organic compounds, indicating that sources have unique organic tracers as well as compounds that have several sources in common. The concentrations presented and discussed below are monthly mean values based on composited daily samples for each month. The relative concentrations of specific compounds and classes of compounds are used to infer differences in sources of primary organic particulate matter across the PRD. Primary organic carbon 31

source apportionment is utilized through a chemical mass balance (CMB) model (37). In addition, Hivolume filter samples collected over 24-hour sampling times during 4 separate days at Yuen Long and Tung Chung are analyzed and discussed. More detailed information on sample preparation, handling and analyses can be found in Appendix B. More on the technique of identifying specific solvent extractable organic compounds, the sources of these compounds, and the application of CMB modeling is described by Schauer et al. (1996). 5.3.1 Organic Carbon and Elemental Carbon Before presenting the concentrations of solvent extractable organic compounds, the concentrations of OC and EC both seasonally and spatially in the PRD will be discussed. The seasonal and spatial distributions of OC and EC are shown in Figure 42, and the ratios of EC to OC are shown in Figure 43. As can be seen in Figure 42, Guangzhou has the highest average OC concentration (17.6 µgm-3), followed by ZS (10.6 µgm-3), SZ (10.1 µgm-3) and CH (9.3 µgm-3). The samples collected from the three Hong Kong sites had the lowest OC concentrations (CW 6.6 µgm-3, TC 6.3 µgm-3, and TM 4.9 µgm-3 on average). The highest EC annual average concentrations occur at GZ (4.4 µgm-3), followed by SZ (3.6 µgm-3), ZS (2.5 µgm-3), TC (2.0 µgm-3), and CW (1.9 µgm-3). The samples from the background sites CH (1.4 µgm-3) and TM (0.8 µgm-3) have the lowest EC concentrations. The relatively low concentrations of EC at the background sites, high concentrations at urban sites and general variability of EC throughout the sampling region supports the notion discussed previously that EC to a large extent is influenced by local sources. Generally speaking EC and OC concentrations are lowest in summer due to increased precipitation and scavenging of particulate matter. It is worthwhile to note that the EC concentrations are generally lower during the present study than measured during November 2000 to October 2001 as part of the ‘Twelve-month particulate matter study in Hong Kong’ funded by the Hong Kong EPD. As part of the pilot study annual mean concentrations for the rural Hok Tsui site were 1.3 µgm-3, for urban Tsuen Wan 5.9 µgm-3, and for the roadside Mong Kok site 19.7 µgm-3. There is not a site similar to Mong Kok in the present study so a direct comparison cannot be made. The Tap Mun site has an annual mean concentration of 0.8 µgm-3, significantly less than the value previously measured at Hok Tsui. Both CW and TC have EC concentrations in the present study more than a factor of two lower than measured at Tsuen Wan for the pilot study. There are several potential reasons for the apparent discrepancy. First, it is possible that the current stations are influenced by a different distribution of sources. Indeed the relatively high EC concentrations at Tsuen Wan suggest 32

the dominant influence of diesel vehicles. It is also possible that EC concentrations are different due to subtle differences in analytical techniques. Although, measurements were made using a similar thermal evolution technique for both studies. Another possibility is that EC is lower in Hong Kong during the present study, perhaps due to a decrease in diesel emissions. The EC/OC ratio provides information on the source of OC with higher ratios (>1) typically associated with diesel exhaust, although the burning of fuel oil also has relatively high EC/OC (Figure 44). As will be discussed later in the report, the EC/OC ratio for the coal source profile which represents residential coal burning in Beijing, may not be representative of coal burning emissions in the PRD. Hence, caution must be used when inferring general sources based on EC/OC ratios since it is possible that this ratio is either higher or lower for the coal burning taking place across the PRD. Generally speaking the EC/OC ratio is able to implicate the importance of diesel exhaust to the total organics in fine particulate matter only when the contribution from other EC sources (e.g. fuel oil, biomass burning and coal combustion) is insignificant. It should also be noted that the formation of secondary organic aerosol may also decrease the ratio through the formation of OC. The highest annual average EC/OC ratios are found at the SZ (0.35), TC (0.32), and CW (0.29) sites. The GZ and ZS samples have slightly lower EC/OC ratios (0.25 for GZ and 0.23 for ZS). The results generally suggest that the relative importance of diesel exhaust to fine OC is higher at the SZ, TC and CW than at GZ. 5.3.2 Organic Tracers in Monthly Aggregate PM2.5 Samples The organic tracers including n-alkanes, PAHs, steranes, hopanes, resin acids, and levoglucosan were identified and quantified in the PM2.5 samples collected form the HKEPD monitoring network including TM, TC, CW, CH, ZS, SZ, and GZ. The concentrations of these organic species are listed in Table 17. In the following section, results are presented for hopanes and steranes which are abundant in lubricating oils used in both diesel and gasoline powered vehicles (38). They can be used as molecular tracers for the fine particulate matter emitted from motor vehicles (39). Levoglucosan is a major constituent of the fine particle emissions from cellulose during biomass burning and has been widely accepted as a major tracer for biomass burning (40). PAHs are the products of incomplete combustion from variable sources, such as fossil fuel combustion and biomass burning. Important sources for PAHs include automobile emissions, coal combustion, biomass burning, and natural gas combustion (Figures 41a, c). Picene, a PAH with molecular weight of 278, is a unique emissions product from the burning of coal (Figure 41c) and thus has been selected as a tracer of coal combustion (41). 33

5.3.2.1 Hopane and Sterane (H+S) Concentrations The concentrations of H+S were found the highest in the samples collected from GZ (17.1-20.7 ngm-3), ZS (1.2-15.1 ngm-3), and SZ (4.8-8.7 ngm-3). Lower concentrations are observed at CH (1.1-2.5 ngm-3), TM (0.2-1.4 ngm-3), TC (1.0-3.6 ngm-3), and CW (1.4-4.5 ngm-3) (Figure 45). Concentrations of hopanes and steranes are lowest during the summer, which is not surprising given the frequency and amount of precipitation. It should be noted that the ZS fall sample showed extremely high concentration of H+S (15.1 ngm-3), most likely due to transport from the Guangzhou area that readily occurred during the fall month. At the urban Guangzhou site, hopanes and steranes were uniformly high throughout the year suggesting intense local mobile source emissions. Figure 46a shows the ratios of H+S to fine OC concentrations. Since H+S are mainly from motor vehicles, the normalization of H+S to OC can indicate the relative importance of motor vehicle emissions to OC concentrations, with high values indicating that mobile sources play a relatively important role. The highest annual average ratios are found at GZ (~1.0 e-3) and Shenzhen (0.7 e-3) as well as during October in Zhongshan (1.2 e-3). Thus results suggest that mobile sources significantly contribute to OC at GZ and SZ as well as during fall in ZS. The ratio of H+S to elemental carbon (EC) can be used to infer the relative contributions of diesel and gasoline emissions to OC at a particular location with diesel vehicles emitting relatively more EC and hence having lower H+S to EC ratios. Although, it should again be pointed out that coal burning also potentially emits both hopanes and steranes, and can thus influence H+S/EC ratios. An assumption in attributing hopanes and steranes to mobile sources in the present case assumes that contributions from coal burning are negligible. Given the current lack of knowledge of the coal burning source profiles in the PRD this assumption can not be verified. Figure 46b shows the previous emission test studies for H+S to EC ratios from four emission sources (42, 43, 44) . The (H+S)/EC ratios are 4.6 e-4 and 0.15 for the diesel- and gasoline-powered vehicle exhausts, respectively. In the present study, the average H+S to EC ratios (in ascending order) are 8 e-4 at TM (4 e-4 – 2 e-3), 1.2 e-3 at CH (9 e-4 - 2 e-3), 1.3 e-3 at TC (8 e-4 - 2 e-3), 2 e-3 at CW (1 e-3 – 2 e-3), at SZ 2 e-3, 3 e-3 at ZS (9 e-4 – 7 e-3), and 4.4 e-3 at GZ (3.7e-3 - 6.e-3) (Figure 46c). All of the H+S to EC ratios were smaller than 0.15 but larger than 4.6e-4. Overall, the results indicate that mobile source emissions of primary organic carbon in Guangzhou are very likely dominated

34

by gasoline vehicles as compared to diesel exhaust. This inference is also supported by the relatively high Pb concentrations observed in Guangzhou relative to the other sampling sites. 5.3.2.2 Levoglucosan Concentrations Figure 47 shows the distribution of levoglucosan concentrations among the seven sites. At five of the seven sites the highest concentrations of levoglucosan are found in the winter samples (from 207 to 813 ng m-3). At Tung Chung, the highest concentration is observed during the fall, while at CH the second highest concentration observed across all stations occurs in spring. It is unclear why the Tung Chung concentration is so high in the fall as compared to the other Hong Kong sites. With the exception of the fall levoglucosan concentration, the concentrations across Hong Kong are similar during each season. This is in agreement with the observed K and Rb concentrations at the sites, and generally supports the notion that biomass burning from outside of Hong Kong is responsible for biomass related PM in the area. The highest annual average concentration of levoglucosan occurs at GZ (508 ngm-3), followed by CH (351 ngm-3 on average), ZS (321 ngm-3) SZ (216 ngm-3), TC (159 ngm-3), CW (125 ngm-3), with the lowest concentration at Tap Mun (110 ngm-3). The relative relationship between the annual mean concentrations between the sites is generally similar to that observed for K and Rb suggesting that these elements are from the same sources as levoglucosan. The relatively high concentrations in both CH and GZ indicate that Guangdong province is a significant source of biomass burning aerosol. The annual mean concentrations of levoglucosan during the previous Hong Kong pilot study were 51 ngm-3, 81 ngm-3 and 213 ngm-3, at Hok Tsui, Tsuen Wan and Mong Kong, respectively. Values during the present study are within this range of values. Although it is worthwhile to note that concentrations at Mong Kok were significantly higher compared to the other sites during the Pilot Study, and the reason was attributed to possible local sources. Leaving out the roadside Mong Kok site, it would appear that concentrations of levoglucosan measured during this study are higher than values reported for the previous study, possibly indicating that biomass burning played a more significant role in fine particulate concentrations during the current study. It is not entirely clear why levoglucosan concentrations perhaps increased during the present study although it could be due to changes in meteorology as well as in emissions throughout the PRD.

35

5.3.2.3 PAH Concentrations Figure 48 shows the spatial and seasonal distributions of specific PAH concentrations. The samples collected from the sites in Guangdong (CH, ZS, SZ and GZ) have much higher PAH concentrations than the Hong Kong samples (TM, TC and CW). The highest annual mean concentration of total PAHs is at GZ (49 ngm-3), followed by ZS (42 ngm-3), SZ (30 ngm-3), CH (21 ngm-3), TC (9 ngm-3), CW (9 ngm-3), and TM (6.0 ngm-3). Generally speaking, at all seven sites PAH concentrations are lowest in summer, with winter samples typically having highest concentrations. An exception to this is the concentration in Zhongshan during the fall which has the highest measured PAH concentration of any of the measurements. The reason for this is unclear, but may include transport of air from the Guangzhou area as well as the influence from more localized sources of the specific PAHs. The PAH concentrations were grouped by their molecular weights and normalized to the total PAH concentration (Figure 49). Two groups of PAHs are further classified based as low molecular weight PAHs (PAHs with molecular weight less than 252) and high molecular weight PAHs (PAHs with molecular weight equal to or higher than 252). As can be seen in Figure 49, the samples have obvious regional characteristics with regards to relative PAH fractions. The proportion of the low molecular weight PAHs (e.g., PAHs with molecular weight 202, 226 & 228 and their methyl substitutions) is higher for the Hong Kong samples than for the PRD samples. Similarly, the high molecular weight PAHs (e.g., PAHs with molecular weight 252, 276 and 300) account for a smaller proportion of the total PAHs in the Hong Kong samples. The relative differences may be due to several reasons, including transformation of higher molecular weight compounds to lower molecular weights during transport as well as differences in the distribution of PAHs emitted from the sources impacting the concentrations at the sites. 5.3.3 Organic Tracers in Episodic High Volume Samples Four specific sampling events were chosen and high volume filter samples were obtained over 24-hour integrated sampling periods at Tung Chung and Yuen Long, an urban site in the northern new territory. The dates were chosen to represent time periods of relatively high particulate matter concentrations in Hong Kong (December 13 and 19, 2002; March 25, 2003) as well as a day having extremely low concentrations (October 20, 2002). The overall objective of the high volume sampling is to evaluate the extent to which daily samples represent monthly composite samples, and to determine the spatial variability of specific organic tracer compounds across Hong Kong for specific events. The high volume 36

samples have an advantage in that they collect enough particulate carbon mass during a 24-hr period to allow for the characterization of single event organic compound concentrations. Comparing side-by-side high volume and low volume samples at Tung Chung reveals high volume OC concentrations that are nearly 60% greater than low volume values. In addition, an extremely large organic compound spike is found in the high volume samples that is not present in the low volume samples. The compound appears to be impactor oil used in the high volume sampler impactor. The contaminant clearly interferes with several organic tracer compounds and further analyses are needed to determine the extent to which the contaminant influences the tracer compound concentrations identified in the present study. The concentrations of the high volume organic compounds as well as a report discussing the differences in the concentrations between the sites for the sampling days is presented in Appendix F. Although, it should be noted that at the present time it is not possible to draw conclusions from the data based on the presence of contamination. It should be pointed out that additional studies are needed to assess the source of the contaminant and extent to which the samples can be used to infer sources based on the analyses of organic compounds. 5.3.4 Source Apportionment of Primary Fine Particulate Organic Carbon The source profiles used for CMB model, including diesel exhaust, gasoline exhaust, vegetative detritus, biomass burning, cigarette smoke, road dust, meat cooking and natural gas combustion, were obtained from previous studies in North America and applied to this study (33, 42, 43, 45, 46, 47, 48) . The coal source profile was obtained from the analysis of fine particulate matter emitted from the burning of Datong coal in China (44) . For the biomass burning source, the source profile for wood combustion (33) is used in the present study. The eight source profiles used in the present study are shown in Figures 41ac. Given that source profiles have not been directly measured in Hong Kong, the model inputs served as the best estimates of the actual emission profiles based on available data. The results of the CMB source apportionment are shown in Figure 50 and in Table 18. Up to eight emission sources including diesel exhaust, gasoline exhaust, meat cooking, cigarette smoke, biomass burning, road dust, vegetative detritus, and coal combustion are identified. The contribution from cigarette smoke should be regarded as an upper limit estimate since the major tracers (iso-nonacosane, anteisotriacontane, and iso-hentriacontane) could be influenced by local vegetative detritus. As discussed in the previous section, significant sources of dust exist in Guangdong in addition to dust that is transported into the PRD region as well as generated locally due to the re-entrainment of road dust. It is not clear that the 37

dust measured at the site is associated with anthropogenic organic compounds as is the case with road dust. Despite this point, it is assumed that elemental tracers for dust are associated with organic compounds present in road dust. Therefore, estimates of the contributions of road dust to primary organic carbon can be considered as upper limit estimates. It should also be pointed out that there are potentially important sources of direct emissions of organic carbon that are not included in the CMB modeling due to lack of adequate source profile information. An example of one such source are marine sources related to shipping and transportation. Although it is not entirely clear that marine sources influence the concentrations of organic carbon in the PRD, the relatively heavy marine traffic in the area may be a source of both primary PM as well as compounds that contribute to secondary PM. 5.3.4.1 Biomass Burning The highest annual mean concentrations of primary fine particulate OC estimated by the CMB for biomass burning is at the GZ site (3.74 µgm-3), followed by CH (2.47 µgm-3), ZS (2.24 µgm-3), SZ (1.61 µgm-3), TC (1.18 µgm-3), CW (0.98 µgm-3), and TM (0.81 µgm-3). The high concentrations of particulate OC from biomass burning estimated at the CH site is in accordance with high levels of levoglucosan found in the CH samples. The elevated concentrations of biomass burning aerosol in Guangdong, which are in agreement with K and Rb concentrations relative to the other sampling sites as previously discussed, indicate that Guangdong is a significant source of these aerosols in the PRD. It should be pointed out that for the seven measured sites in the PRD, biomass burning is estimated to account for on average 18% of the measured fine particulate OC mass, ranging from 27% at Conghua to ~14% at Shenzhen and CW as shown in Table 19. At 5 of the 7 sites biomass burning is the most pronounced source of annual fine particulate OC based on CMB model results. It is apparent that biomass burning is estimated to be a much more significant fraction of primary fine particulate OC during the present study as compared to the Pilot Study. As pointed out earlier, with the exception of the Mong Kok roadside site, levoglucosan concentrations are higher during the present study. Although, it should be noted that it is difficult to make direct comparisons of the concentrations measured during the two separate studies since the sampling sites are not the same, indicating that they may be influenced by different sources. Nonetheless, given that the same source profiles are used for both studies, the results indicate an increase in the biomass burning contribution for the current study.

38

5.3.4.2 Meat Cooking Meat cooking exhaust was found the highest (annual mean) at the CH rural site (2.20 µgm-3), followed by GZ (2.14 µgm-3), SZ (1.25 µgm-3), CW (1.00 µgm-3), ZS (0.81 µgm-3), TC (0.64 µgm-3), and TM (0.38 µgm-3). The urban sites (GZ, SZ and CW) had generally higher meat cooking exhaust than the receptor (ZS and TC) and rural (TM) sites. CH had the highest contribution from meat cooking very likely due to local sources in the vicinity of the sampling site. The CMB results for meat cooking at CH correspond well with the concentrations of nonanal (the tracer used for meat cooking), with the highest mean value in the CH samples (11.38 ngm-3). The results are similar to values measured during the previous Pilot Study, with annual means of 1.45, 0.78 and 0.26 µgm-3 at MK, TW and HT, respectively. However, it should be noted that the style of meat cooking in the PRD is different than in the United States and therefore the source profile may not sufficiently estimate the actual contributions of cooking to primary OC mass. Given the fact that meat cooking contributes significantly to fine OC mass at several of the stations in the PRD suggests the importance of future studies to determine source profiles for the region. 5.3.4.3 Diesel Exhaust The highest yield for diesel exhaust is obtained from the SZ site (2.06 µgm-3), GZ (1.71 µgm-3), CW (1.36 µgm-3), TC (1.23 µgm-3), ZS (0.86 µgm-3), TM (0.43 µgm-3), and CH (0.31 µgm-3). The highest EC/OC ratios were also found in the SZ samples. Diesel exhaust at the SZ site is expected to be significant due to the large number of diesel-powered container trucks commuting between Hong Kong and Shenzhen. The relatively high variability between sites suggests local sources influence most of the sampling sites, as has been already inferred based on the spatial distribution of EC concentrations. The estimated contribution of diesel exhaust to annual mean measured OC concentrations ranges from 4% and 8% at the urban CH and TM stations to ~ 20% at TC, CW, and SZ. Across Hong Kong the annual mean contributions range from 9-20%. These values are much lower than mean contributions of 25% at the background Hok Tsui site to 60% at the urban Tsuen Wan location estimated for the Pilot Study. The primary reason for the difference is the much lower EC concentrations measured in Hong Kong during the current study, which are used in the CMB model to estimate the contribution of diesel exhaust. As discussed earlier this may be due to several factors including subtle differences in EC measurement techniques between the two studies (although this is unlikely), shifts in emissions, site differences, and changes in meteorology during the studies. Nonetheless, it should be pointed out that in the current study the EC concentrations were measured using the same laboratory and instrumental setup as for the source 39

profile samples used in the CMB model and therefore a bias should not exist in the present CMB estimates based on EC measurement techniques. Overall the results suggest that diesel emissions have a significant influence on fine particulate OC concentrations, particularly at the urban and receptor sites. 5.3.4.4 Gasoline Exhaust Unlike diesel exhaust, the contribution of gasoline exhaust to fine particulate OC is the highest at the GZ site (2.82 µgm-3), followed by ZS (1.30 µgm-3), SZ (0.83 µgm-3), CH (0.51 µgm-3), CW (0.48 µgm-3), TC (0.35 µgm-3), and TM (0.25 µgm-3). The sites in Guangdong appear to be more influenced by gasoline exhaust than those in Hong Kong. Even at the rural site in the PRD area (CH), the level of gasoline exhaust is as high as in urban Hong Kong (CW). On average gasoline exhaust accounts for 5-7% of the measured fine particulate organic carbon at the Hong Kong stations, which is a factor of 2-4 less than the % contribution from diesel exhaust. Hence the results suggest that among mobile sources, diesel emissions are more prominent than that of gasoline vehicles in Hong Kong. The contribution of gasoline vehicles at Guangzhou to annual measured OC concentrations is 16%, as compared to 10% for diesel vehicles. These results suggest that Guangzhou is a significant source of gasoline vehicle related pollutants. It is possible that the relative increase in the amount of OC attributed to gasoline vehicles is linked with differences in the profiles of mobile sources emissions in Guangzhou. For example, if diesel vehicles in Guangzhou emit a great proportion of H+S as compared to the Hong Kong sites a bias would exist in estimated that is not related to actual emissions. It should be pointed out that the CMB attribution of a significant fraction of primary OC to gasoline vehicles is in line with the relatively high concentrations of Pb measured in Guangzhou, and it is interesting to point out that Pb concentrations are roughly a factor of 5 higher in Guangzhou than in Central/Western, which is roughly the ratio of estimated gasoline vehicle organic carbon at the two sites. These results generally suggest that gasoline vehicle emissions in the Guangzhou area influence fine particle OC concentrations regionally throughout the PRD. The results of the current study are in general agreement with the Pilot Study that found relatively insignificant (< 5%) contributions of gasoline vehicles to fine particulate organic carbon at Hok Tsui and Tsuen Wan. At the roadside Mong Kok station the annual mean concentration of OC estimated to be from gasoline vehicles was 1.12 µgm-3, which accounted for 8% of the measured OC at that site. It is worthwhile to point out that the Guangzhou estimated annual mean concentration attributed to gasoline vehicles is more than a factor of 2 greater than that measured at Mong Kok. This is somewhat surprising 40

given that the site in Guangzhou is urban but not located directly at roadside. This underscores the impact of mobile sources, and in particular gasoline vehicles, on fine particulate OC in Guangzhou. 5.3.4.5 Incomplete Combustion of Coal The contribution of coal combustion to primary OC is estimated to be relatively minor (< 5%) in Hong Kong. In Guangdong the contribution of the incomplete combustion of coal is estimated to account for from 8% of the measured OC at Conghua to 12% at Zhongshan. It is interesting to note that the highest contribution of coal to primary OC at ZS may be due to the apparent source region of particulate matter that is suggested to exist in Guangdong between Guangzhou and the location. It was also estimated in the Pilot Study that coal combustion had a minor contribution (<5%) to measured fine particulate OC. As previously discussed, it is not clear that the coal source profile for Beijing coal well represents the actual burning of coal that takes place across the PRD. Hence, caution must be used in attributing coal combustion to primary organic carbon concentrations. It should be noted that an SO2 emission inventory has been generated for the PRD, and roughly 40% of SO2 emissions were attributed to electricity generation with ~60% due to industrial purposes. Based on the emissions inventories it not possible to infer a general source profile for the region given that the methodologies of coal combustion as well as type of coal being burned is not known. The lack of knowledge of the actual coal source profile influences not only the CMB estimate of the coal contribution, but also that of mobile sources since coal burning can generate hopanes and steranes as well as EC both of which are used as tracers for these sources. It should be pointed out that although coal does not appear to significantly influence OC concentrations it does indeed impact sulfate contributions, which are relatively high (annual mean values at monitoring sites ranging from ~ 9 µgm-3 at the Hong Kong sites to µgm-3 at Guangzhou). Therefore, the CMB results should not be interpreted to mean that coal burning does not contribute to overall fine particulate matter in the PRD, given that through sulfate, coal burning accounts for a significant fraction of the PRD’s PM2.5. 5.3.4.6 Road Dust As pointed out earlier, road dust contributions should be considered to be absolute upper limit estimates since it is not clear that the elements associated with dust are accompanied by organic compounds. Particularly given the fact that Guangzhou appears to be a tremendous source of dust related elements, 41

and that other locations are likely influenced by transport from the Guangzhou area and not necessarily locally emitted road dust. The estimated contributions of road dust to fine particulate organic carbon for the seven sites are estimated to be 0.91 (GZ), 0.75 (ZS), 0.64 (SZ), 0.53 (CH), 0.40 (TM), 0.38 (CW), and 0.33 µgm-3 (TC). The concentrations of road dust are uniform across Hong Kong, which relates to the similar uniformity in concentrations of the dust tracers Si and Fe measured in Hong Kong. Overall, road dust is estimated to account for on average an upper limit of 5 – 8% of fine particulate organic carbon in the PRD. 5.3.4.7 Cigarette Smoke The contribution of cigarette smoke to OC concentrations is highest at GZ (0.72 µgm-3), followed by ZS (0.24 µgm-3), SZ (0.23 µgm-3), CW (0.19 µgm-3), TC (0.14 µgm-3), CH (0.13 µgm-3), and TM (0.05 µgm3

). The contribution of cigarette smoke to fine particulate OC across all sites is on average 2% (with a

range of from 1-4%). As stated earlier, concentrations of cigarette smoke tracers can be influenced by the presence of vegetative detritus, which are found in samples at all sites, and therefore the contributions from cigarette smoking should be taken as upper limit estimates. Regardless, these results suggest that cigarette smoke has a negligible influence on OC concentrations measured at sampling sites within the PRD during the present study. 5.3.4.8 Vegetative Detritus The estimated contribution of contribution from vegetative detritus to fine OC is the lowest among the eight identified emission sources, with the highest annual mean (~0.21 µgm-3) located at both CH and GZ, followed by ZS (0.13 µgm-3), and SZ (0.09 µgm-3). Similar low levels of vegetative detritus are found at the three Hong Kong sites (0.07 µgm-3 each). Overall CMB results indicate that vegetative detritus plays a minor role in fine OC in the Pearl River Delta. 5.3.4.9 Organic Carbon Not Accounted For by CMB and Water Soluble Organic Carbon (WCOC) The difference between the measured organic carbon and the OC concentration estimated from the sum of the identified sources using CMB modeling is given in Figure 50 and classified as ‘other OC’. This other OC can be due to a variety of factors including unidentified sources, uncertainties in the CMB modeling, and potential artifacts in sampling carbonaceous particulate matter. Another possible explanation is 42

secondary organic aerosol, which is not attributed to primary emissions in the CMB model. Dr. Jian Yu of HKUST conducted a variety of analyses as part of the PRD project including water soluble organic carbon (WSOC) concentrations (Appendix G). In general, oxidation of organic compounds in the gasphase as well as heterogeneously on existing particulate matter will lead to more polar water soluble compounds. Figure 51 shows the relationship between % ‘other’ organic carbon defined as OC not identified by the CMB Model, and the% WSOC, both presented as a fraction of the measured OC concentration. The results are annual averages for all sampling sites. Interestingly, there is a clear relationship between other organic carbon and WSOC. In general the Tap Mun site has the highest fraction of both ‘other’ OC and WSOC, with lowest relative values at GZ. This may be due to a greater fraction of secondary organic aerosol present at the remote TM site, generated during transport from source regions, and less of a fraction at GZ due to the prominence of primary OC emissions. The differences seen for the sites could also be due to differences in the sources influencing the OC concentrations at each location. 6.0 Acknowledgements This report arose out of the Hong Kong and Pearl River Delta Pilot Air Monitoring Project (the Project). The Project was split into two focus areas with Project 1 gaining further understanding on the characteristics of Ozone in Hong Kong and Project 2 focusing on the characteristics of fine particulates in the Pear River Delta region. This report is the main publication arising from Project 2.

Organizational Collaborators The Project has required and enabled the collaboration between regional authorities, regulators, industry, civic organizations and the scientific community. It is the first time a cross-section of local and international stakeholders, led by a Hong Kong-based think tank, have teamed up to work on regional air quality issues. Listed alphabetically the Project collaborators include: Argonne National Laboratory USA, California Institute of Technology USA, Castle Peak Power Co. Ltd. HK, Civic Exchange HK, Conghua Environmental Monitoring Centre PRC, Georgia Institute of Technology USA, Guangdong Environmental Protection Bureau PRC, Guangzhou Environmental Monitoring Station PRC, Hong Kong Environmental Protection Department, Hong Kong Jockey Club Charities Trust, Hong Kong Observatory, 43

Hong Kong Polytechnic University, Hong Kong University of Science and Technology, NOAA Aeronomy Laboratory USA, Peking University PRC, Shell Hong Kong Ltd., Shenzhen Environmental Monitoring Centre PRC and Zhongshan Environmental Monitoring Centre PRC. We wish to acknowledge that many of the people and institutions involved contributed much more to the Project than there were resources in the budget to cover all their work, for which we are most grateful.

Funders One of the highlights of the Project is that it was largely funded by the private sector and in doing so shows the commitment of this sector to build the road for the development of sustainable solutions in regional air quality. Our special thanks go to the Hong Kong Jockey Club Charities Trust for their contribution of half the project funding and to Castle Peak Power Co. Ltd., for their contribution not only of funding but of scientific and management guidance. Similarly, recent minority funding and management guidance from Shell Hong Kong Ltd. has enabled the successful completion of the project. It is difficult to mention the Hong Kong Environmental Protection Department (EPD) only in its capacity as a financial funder as their contribution to this project was much more valuable than the funds provided. Open access to their history of relevant data, collection of Project 2 data through their monitoring stations, contributions to planning, implementation and management are a part of the resume which they compiled during this project. We also acknowledge the Hong Kong Observatory for the provision of the meteorological data used in this Project. Our special thanks are offered to Dr. Peter Louie of EPD for both his cooperation and steadfast determination.

Personnel The Project would not have been possible without the support of a large number of key people. C.S. Kiang of Peking University initiated and fostered a high level of partner cooperation throughout the entire project. Tao Wang of Hong Kong Polytechnic University, and Alexis Lau of the Hong Kong University of Science and Technology provided critical local knowledge. Professors Xiaoyan Tang and Yuanhang Zhang as well as Limin Zeng at Peking University were involved in both planning and field sampling. Many other dedicated people worked at the Chinese monitoring 44

stations and within Peking University including: Yingzi Cao, Jianjun Chen, Zuzhao Huang, Faping Jiang, Tao Liu, Cuiqin Luo, Yuwen Niu, Erfan Wu, Qiang Yang, Wendong Yang, Tingting Yao and Xiaoyun Zeng. Dr. Peter Louie from the Hong Kong EPD and Dr. Sjaak Slanina of Peking University spent a tremendous amount of time overseeing various aspects of QA/QC as we all many details associated with the science of the project. Others from EPD were involved in the sampling in Hong Kong including: Benny So, Roland Wong and Ka Chun Yeung. Dr. Slanina, Dr. Louie, Shaw Liu of Sinica (Taiwan) and Gayle Kendall of CLP Research Institute (Hong Kong) deepened the final report review process. The authors would also like to thank the assistance from Bo Wang for solvent extraction and GC/MS analysis and Dr. Lin Ke and Bo Wang for data analysis and report preparation. In addition we thank HKUST personnel Eric Wan for sampler maintenance and calibration, and Jinhui Xu for chemical analysis of water-soluble organic compounds. Christine Loh and Kylie Uebergang from Civic Exchange played key organizational roles and supplied wide-ranging logistics and management support to the Project. Most importantly, Christine through her long-term commitment to regional air quality, was an integral ingredient to make the Project possible. 7.0 References 1. Pope CA III, Burnett RT, Thun MJ, et al. Lung cancer, cardiopulmonary mortality, and long-term exposure to fine particulate air pollution. The Journal of the American Medical Association. 287:1132-1141, 2002. 2. Schwartz J, Dockery DW, Neas LM. Is daily mortality associated specifically with fine particles? Air & Waste Management Association. 46:927-939, 1996. 3. Dockery DW, Pope CA, Xu XP, et al. An association between air pollution and mortality in 6 United States cities. New England Journal of Medicine. 329(24):1753-1759, 1993. 4. Pozzi R, et al. Inflammatory mediators induced by coarse (PM2.5-10) and fine (PM2.5) urban air particles in RAW 264.7 cells. Toxicology. 183: 243-254, 2003. 5. Haywood JM, Shine KP. The effect of anthropogenic sulfate and soot aerosol on the clear sky planetary radiation budget. Geophysical Research Letters. 22,5:603-606, 1995. 6. Schwartz. The whitehouse effect – shortwave radiative forcing of climate by anthropogenic aerosols: an overview. Journal of Aerosol Science. 27,3:359-382, 1996. 7. Charlson RJ, Schwartz SE, Hales JM, et al. Climate forcing by anthropogenic aerosols. Science. 255:423-430, 1992. 8. Chameides WL, Yu H, Liu SC, et al. Case study of the effects of atmospheric aerosols and regional haze on agriculture: An opportunity to enhance crop yields in China through emission controls? Proceedings from the National Academy of Sciences. 96,24:13626-13633, 1999. 9. Cao JJ, Lee SC, Ho KF, et al. Characteristics of carbonaceous aerosol in Pear River Delta Region, China during 2001 winter period. Atmospheric Environment, 37:1451-1460, 2003. 10. Ho KF, Lee SC, Yu JC, et al. Carbonaceous characteristics of atmospheric particulate matter in Hong Kong. The Science of the Total Environment. 300:59-67, 2002. 45

11. Ho KF, Lee SC, Chan CK, et al. Characterization of chemical species in PM2.5 and PM10 aerosols in Hong Kong. Atmospheric Environment. 37:31-39, 2003. 12. Wei F, Teng E, Wu G, et al. Ambient concentrations and elemental compositions of PM10 and PM2.5 in four Chinese cities. Environmental Science & Technology. 33,23:4188-4193, 1999. 13. Liu HP, Chan JCL. Boundary layer dynamics associated with a severe air-pollution episode in Hong Kong. Atmospheric Environment. 36:2013-2025, 2002. 14. Liu HP, Chan JCL. An investigation of air-pollutant patterns under sea-land breezes during a severe air-pollution episode in Hong Kong. Atmospheric Environment. 36:591-601, 2002. 15. Liu HP, Chan JCL, Cheng AYS. Internal boundary layer stucture under sea-breeze conditions in Hong Kong. Atmospheric Environment. 35:683-692, 2001. 16. Chan LY, Kwok WS. Vertical dispersion of suspended particulates in urban area of Hong Kong. Atmospheric Environment. 34:4403-4412, 2000. 17. Man CK, Shih MY. Identification of sources of PM10 aerosols in Hong Kong by wind trajectory analysis. Journal of Aerosol Science. 32:1213-1223, 2001. 18. Qin Y, Chan CK, Chan LY. Characteristics of chemical compositions of atmospheric aerosols in Hong Kong: spatial and seasonal distributions. The Science of the Total Environment. 206:24-37, 1997. 19. Cheng SQ, Lam KC. An analysis of winds affection air pollution concentrations in Hong Kong. PII. 2559-2567, 1997. 20. Lee YL, Sequeira R. Visibility degradation across Hong Kong: its components and their relative contributions. Atmospheric Environment. 35:5861-5872, 2001. 21. Lee YL, Sequeira R. Water-soluble aerosol and visibility degradation in Hong Kong during autumn and early winter, 1998. Environmental Pollution. 116:225-233, 2002. 22. Pathak RK, Yao XH, Lau AKH, et al. Acidity and concentrations of ionic species of PM2.5 in Hong Kong. Atmospheric Environment. 37:1113-1124, 2003. 23. Kok GL, Lind JA. An airborne study of air quality around the Hong Kong Territory. Journal of Geophysical Research. 102,D15:19043-19057, 1997. 24. Cheng ZL, Lam KS, Chan LY, et al. Chemical characteristics of aerosols at coastal station in Hong Kong. I. Seasonal variation of major ions, halogens, and mineral dusts between 1995 and 1996. 25. Lee SC, Chang M. Indoor and outdoor air quality investigation at schools in Hong Kong. Chemosphere. 41:109-113, 2000. 26. Chan LY, Kwok WS, Lee SE, et al. Spatial variation of mass concentration of roadside suspended particulate matter in metropolitan Hong Kong. Atmospheric Environment. 35:3167-3176, 2001. 27. Chao CY, Wong KK. Residential indoor PM10 and PM2.5 in Hong Kong and the elemental composition. Atmospheric Environment. 36:265-277, 2002. 28. Chan LY, Lau WL, Zou SC, et al. Exposure level of carbon monoxide and respirable suspended particulate in public transportation modes while commuting in urban area of Guangzhou, China. Atmospheric Environment. 36:5831-5840, 2002. 29. CH2M HILL (China) Limited. Study of Air Quality in the Pearl River Delta Region. Commissioned by the Hong Kong Environmental Protection Department, Hong Kong Special Administrative Region Government. April 2002. 30. HKEPD. Twelve-month particulate matter study in Hong Kong. December 2002. 31. Zheng M, Fang M, Wang F, To KL. Characterization of the solvent extractable organic compounds in PM2.5 aerosols in Hong Kong. Atmospheric Environment. 34:2691-2702, 2000. 32. Chameides W.L, Bergin M.H. Soot takes center stage. Science, V. 297, 2214-2215, 2002. 33. Zheng M, Cass GR, Schauer JJ, Edgerton ES. Source apportionment of PM2.5 in the southeastern United States using solvent-extractable organic compounds as tracers. Environmental Science and Technology. 36:2361-2371, 2002. 34. Simoneit, BRT, Sheng GY, Chen XJ, et al. Molecular Marker Study of Extractable Organic Matter in Aerosols From Urban Areas of China. Atmospheric Environment. 25A:2111-2129, 1991. 46

35. Zheng M, Fang M. Particle-associated polycyclic aromatic hydrocarbons in the atmosphere of Hong Kong. Water, Air, and Soil Pollution. 117:175-189, 2000. 36. Xu JH, Lee FSC. Quantification of nitrated polynuclear aromatic hydrocarbons in atmospheric particulate matter. Analytica Chimica Acta. 416:111-115, 2000. 37. Watson, J.G., Robinson, N.F., Chow, J.C., Henry, R.C., Kim, B.M., Pace, T.G., Meyer, E.I., Nguyen, Q., The USEPA/DRI chemical mass balance receptor model, CMB 7.0. Environmental Software 5, 38-49, 1990. 38. Cass, G.R., 1998. Organic molecular tracers for particulate air pollution sources. Trends in Analytical Chemistry, 17, 356-366. 39. Schauer, J.J., Rogge, W.F., Hildemann, L.M., Mazurek, M.A., Simoneit, B.R.T., Source apportionment of airborne particulate matter using organic compounds as tracers. Atmospheric Environment, 30, 3837-3854, 1996. 40. Simoneit, B.R.T., Schauer, J.J., Nolte, C.G., Oros, D.R., Elias, V.O., Fraser, M.P., Rogge, W.F., Cass, G.R., 1999. Levoglucosan, a tracer for cellulose in biomass burning and atmospheric particles. Atmospheric Environment, 33, 173-182. 41. Oros, D.R., Simoneit, B.R.T., 2000. Identification and emission rates of molecular tracers in coal smoke particulate matter. Fuel, 79, 515-536. 42. Schauer, J.J., Kleeman, M.J., Cass, G.R., Simoneit, B.R.T., 1999b. Measurement of emissions from air pollution sources. 2. C-1 through C-30 organic compounds from medium duty diesel trucks. Environmental Science and Technology, 33, 1578-1587. 43. Schauer, J.J., Kleeman, M.J., Cass, G.R., Simoneit, B.R.T., 2002. Measurement of emissions from air pollution sources. 5. C-1-C-32 organic compounds from gasoline-powered motor vehicles. Environmental Science and Technology, 36, 1169-1180. 44. Zheng, M., Salmon, L.G., Schauer, J.J., Zeng L., Zhang Y., Kiang, C.S., Cass G.R. Source Apportionment of PM2.5 in Beijing, China by Chemical Mass Balance, manuscript in preparation. 45. Hildemann, L.M., Markowski, G.R., Cass, G.R., 1991. Chemical composition of emissions from urban sources of fine organic aerosol. Environmental Science and Technology, 25, 744-759. 46. Rogge, W.F., Hildemann, L.M., Mazurek, M.A., Cass, G.R., Simoneit, B.R.T., 1993a. Sources of fine organic aerosol. 4. Particulate abrasion products from leaf surfaces of urban plants. Environmental Science and Technology, 27, 2700-2711. 47. Schauer, J.J., Kleeman, M.J., Cass, G.R., Simoneit, B.R.T., 1999a. Measurement of emissions from air pollution sources. 2. C-1 through C-29 organic compounds from meat charbroiling. Environmental Science and Technology, 33, 1566-1577. 48. Schauer, J.J., 1998. Source contributions to atmospheric organic compound concentrations: Emissions measurement and model predictions. Ph.D. Dissertation, California Institute of Technology. Pasadena, CA. 49. Yu, J.Z., Tung, J.W.T., Wu, A.W.M., Lau, A.K.H., Louie, P.K.K., Fung, J.C.H., 2004. Abundance and seasonal characteristics of elemental and organic carbon in Hong Kong PM10, Atmos. Environ., 38, 1511-1521. 50. Cao JJ, Lee SC, Ho KF, Zou SC, Fung K, Li Y, Watson JG, Chow JC, 2004. Spatial and seasonal variations of atmospheric organic carbon and elemental carbon in Pearl River Delta Region, China, Atmospheric Environment, 38, 4447-4456.

47

Table 1. 24-hr Sample Completeness During Field Study Site

Complete

Partial

No Sample

Tap Mun (TM1)

20

--

--

Tung Chung (TC2)

20

--

--

C&W (CW3)

20

--

--

Shenzhen (SZ4)

17

3

--

Guangzhou (GZ5)

19

1

--

Conghua (CH6)

18

1

1

Zongshan (ZS7)

19

--

1

Table 2. Intercomparison results for Caltech (CT) and ThermoAndersen (TA) Samplers Conducted at Shenzhen on September 9, 2002 Sampler

Type

Shenzhen (SZ4) Guangzhou (GZ5) Conghua (CH6) Zhongshan (ZS7)

TA TA Caltech Caltech

PM2.5 (µg m-3) mean std. dev. 73.0 1.8 72.6 0.1 70.5 2.8 69.9 1.4

% Difference from SZ4 --0.5 -3.4 -4.2

Table 3. Intercomparison results for Hong Kong ThermoAndersen Samplers on Sept. 11, 12 2003 Sampler

Type

Shenzhen (SZ4) Tap Mun (TM1) Tung Chung (TC2) C & W (CW3)

TA TA TA TA

Shenzhen (SZ4) Tap Mun (TM1) Tung Chung (TC2) C & W (CW3)

TA TA TA TA

PM2.5 (µg m-3) mean std. dev. September 11, 2003 52.8 3.4 50.5 1.6 50.2 2.7 50.7 0.4 September 12, 2003 41.2 0.1 41.6 0.6 39.4 1.9 40.1 0.3

% Difference from SZ4 --4.4 -4.9 -4.0 -1.0 -4.4 -2.7

48

Table 4. Seasonal and annual mean concentrations across the PRD Mean Concentration (µg/m3)a Site PM2.5 OC EC Nitrate Sulfate Ammonia TM 25.8 (14.6) 8.0 (5.3) 0.9 (0.6) 0.4 (0.3) 8.2 (3.4) 2.1 (1.0) TC 29.1 (13.9) 10.6 (5.6) 2.0 (1.2) 0.5 (0.4) 7.9 (3.1) 2.3 (1.1) CW 30.1 (13.8) 9.0 (5.6) 2.0 (1.4) 0.7 (0.5) 8.2 (2.5) 2.2 (1.0) SZ 42.7 (21.2) 17.3 (10.5) 4.1 (1.4) 1.0 (0.9) 8.8 (3.3) 2.5 (1.3) GZ 66.2 (26.1) 28.3 (14.0) 4.6 (0.9) 1.9 (1.8) 14.2 (5.1) 3.6 (0.7) CH 35.3 (9.5) 12.4 (3.5) 1.5 (0.4) 0.3 (0.1) 9.8 (2.8) 2.2 (0.4) ZS 42.4 (27.7) 16.3 (11.9) 2.1 (1.0) 0.7 (0.6) 11.7 (5.4) 2.5 (0.7) TM 32.3 (7.8) 7.8 (2.5) 1.1 (0.5) 0.5 (0.2) 11.2 (3.5) 2.5 (0.8) TC 31.1 (9.4) 9.5 (4.3) 2.6 (0.8) 1.0 (0.5) 9.1 (3.8) 2.5 (0.9) CW 37.7 (7.8) 11.0 (1.8) 2.4 (0.7) 1.6 (0.6) 10.9 (3.1) 2.8 (0.8) SZ 54.7 (24.8) 18.1 (8.4) 4.9 (3.2) 3.5 (2.1) 11.8 (6.3) 4.0 (1.7) GZ 72.4 (30.7) 27.0 (13.0) 5.3 (1.9) 4.8 (3.4) 14.0 (5.4) 4.5 (2.4) CH 28.3 (11.7) 10.9 (5.1) 1.5 (0.5) 0.2 (0.1) 7.7 (3.6) 1.7 (0.8) ZS 58.5 (18.5) 21.2 (8.3) 4.0 (0.9) 2.9 (1.6) 13.7 (4.2) 4.0 (1.2) TM 32.9 (12.6) 6.7 (2.8) 0.7 (0.3) 0.7 (0.4) 10.1 (4.3) 2.9 (1.4) TC 39.9 (5.7) 9.6 (2.2) 2.1 (0.5) 1.4 (1.2) 10.6 (2.0) 3.1 (0.7) CW 39.4 (9.7) 10.5 (1.7) 1.9 (0.9) 1.6 (0.4) 9.9 (2.8) 2.9 (0.5) SZ 54.6 (11.8) 17.0 (3.3) 4.0 (1.5) 3.6 (0.6) 10.6 (3.3) 3.9 (0.5) GZ 79.4 (41.3) 23.8 (12.5) 4.5 (2.4) 7.1 (4.9) 16.3 (7.7) 5.8 (2.5) CH 50.6 (22.8) 16.9 (7.8) 1.5 (0.8) 0.6 (0.3) 14.6 (6.6) 3.1 (1.4) ZS 55.4 (15.7) 15.8 (4.7) 2.5 (1.0) 3.0 (2.1) 13.1 (2.4) 3.8 (0.9) TM 24.0 (19.9) 4.9 (4.9) 0.6 (0.5) 0.3 (0.1) 7.2 (7.2) 1.7 (1.6) TC 29.7 (25.0) 5.7 (5.6) 1.2 (1.0) 0.2 (0.1) 8.6 (9.6) 1.9 (2.0) CW 30.0 (22.4) 6.6 (6.1) 1.4 (0.8) 0.3 (0.1) 8.1 (8.1) 2.0 (2.0) SZ 35.0 (23.1) 9.3 (6.2) 2.5 (2.2) 0.9 (0.4) 8.3 (7.7) 2.2 (2.0) GZ 64.3 (28.9) 19.4 (7.9) 3.1 (1.2) 2.2 (2.1) 14.4 (7.5) 4.0 (2.5) CH 35.8 (25.2) 11.4 (8.4) 1.2 (0.5) 0.3 (0.2) 10.5 (7.1) 2.5 (1.8) ZS 28.8 (23.1) 5.8 (4.5) 1.4 (0.4) 0.3 (0.2) 9.1 (9.1) 2.5 (2.6) TM 28.7 (13.8) 6.9 (3.9) 0.8 (0.5) 0.5 (0.3) 9.2 (4.7) 2.3 (1.2) TC 32.5 (14.8) 8.8 (4.7) 2.0 (1.0) 0.8 (0.8) 9.0 (5.1) 2.5 (1.3) CW 34.3 (14.1) 9.3 (4.3) 1.9 (1.0) 1.0 (0.7) 9.3 (4.5) 2.5 (1.2) SZ 47.1 (21.3) 15.6 (7.9) 3.9 (2.3) 2.3 (1.8) 10.0 (5.3) 3.2 (1.6) GZ 70.6 (30.2) 24.6 (11.6) 4.4 (1.8) 4.0 (3.7) 14.7 (6.1) 4.5 (2.2) CH 36.8 (18.5) 12.7 (6.3) 1.4 (0.5) 0.3 (0.2) 10.4 (5.4) 2.4 (1.2) ZS 46.5 (23.0) 14.7 (9.1) 2.5 (1.2) 1.8 (1.8) 11.9 (5.7) 3.3 (1.6) a Values in parentheses represent standard deviation b OC (organic compounds) is calculated by multiplying the measured organic carbon by 1.4 Annual

June, 2003

March, 2003

December, 2002

October, 2002

b

Crustal 3.1 (2.8) 2.9 (2.6) 3.1 (2.7) 4.9 (3.1) 6.8 (3.4) 5.4 (1.6) 6.1 (6.7) 2.4 (1.2) 2.3 (1.2) 2.5 (1.1) 4.9 (2.4) 7.3 (3.6) 3.6 (1.5) 5.9 (2.5) 2.1 (0.9) 2.1 (0.7) 2.1 (0.7) 4.0 (0.8) 5.8 (2.9) 5.7 (3.8) 4.1 (0.3) 1.6 (1.4) 1.7 (1.5) 1.9 (1.5) 2.3 (1.5) 6.7 (3.5) 4.0 (2.2) 2.0 (1.4) 2.3 (1.7) 2.2 (1.6) 2.4 (1.6) 4.1 (2.3) 6.6 (3.1) 4.6 (2.3) 4.5 (3.5)

Trace 1.0 (0.6) 1.0 (0.3) 0.9 (0.8) 1.3 (0.6) 1.6 (0.7) 1.6 (0.5) 0.9 (0.8) 1.1 (0.1) 0.7 (0.4) 1.0 (1.9) 2.2 (1.5) 3.1 (0.2) 1.1 (0.5) 2.2 (0.7) 0.8 (0.7_ 1.4 (0.2) 1.1 (0.8) 1.6 (0.4) 2.7 (0.3) 1.8 (0.2) 1.8 (0.7) 0.7 (0.3) 0.9 (0.3) 1.0 (1.1) 1.1 (0.3) 2.4 (0.1) 1.0 (0.2) 0.7 (0.4) 0.9 (0.4) 1.0 (0.4) 1.0 (0.3) 1.6 (0.9) 2.5 (1.3) 1.4 (0.5) 1.4 (0.9)

49

Table 5. r2 values calculated from a least-squares fit of PM2.5 mass at a pair of sites.

PM2.5

TM1

TC2

CW3

SZ4

ZS7

GZ5

CH6

TM1

1.00

0.71

0.94

0.82

0.77

0.19

0.08

TC2

0.71

1.00

0.78

0.74

0.56

0.02

0.12

CW3

0.94

0.78

1.00

0.83

0.68

0.20

0.11

SZ4

0.82

0.74

0.83

1.00

0.77

0.18

0.15

ZS7

0.77

0.56

0.68

0.77

1.00

0.17

0.04

GZ5

0.19

0.02

0.20

0.18

0.17

1.00

0.22

CH6

0.08

0.12

0.11

0.15

0.04

0.22

1.00

Table 6. r2 values from a least-squares fit of organic carbon concentrations at a pair of sites. OC

TM1

TC2

CW3

SZ4

ZS7

GZ5

CH6

TM1

1.00

0.75

0.75

0.65

0.42

0.02

0.00

TC2

0.75

1.00

0.67

0.72

0.46

0.02

0.01

CW3

0.75

0.67

1.00

0.61

0.36

0.04

0.03

SZ4

0.65

0.72

0.61

1.00

0.50

0.10

0.03

ZS7

0.42

0.46

0.36

0.50

1.00

0.23

0.02

GZ5

0.02

0.02

0.04

0.10

0.23

1.00

0.08

CH6

0.00

0.01

0.03

0.03

0.02

0.08

1.00

50

Table 7. r2 values from a least-squares fit of sulfate concentrations at a pair of sites. Sulfate

TM1

TC2

CW3

SZ4

ZS7

GZ5

CH6

TM1

1.00

0.76

0.93

0.88

0.87

0.27

0.06

TC2

0.76

1.00

0.86

0.86

0.82

0.15

0.10

CW3

0.93

0.86

1.00

0.89

0.85

0.26

0.09

SZ4

0.88

0.86

0.89

1.00

0.90

0.29

0.12

ZS7

0.87

0.82

0.85

0.90

1.00

0.21

0.04

GZ5

0.27

0.15

0.26

0.29

0.21

1.00

0.48

CH6

0.06

0.10

0.09

0.12

0.04

0.48

1.00

Table 8. r2 values from a least-squares fit of silica concentrations at a pair of sites. Si

TM1

TC2

CW3

SZ4

ZS7

GZ5

CH6

TM1

1.00

0.93

0.97

0.83

0.75

0.03

0.28

TC2

0.93

1.00

0.94

0.76

0.73

0.01

0.26

CW3

0.97

0.94

1.00

0.81

0.73

0.04

0.27

SZ4

0.83

0.76

0.81

1.00

0.77

0.03

0.49

ZS7

0.75

0.73

0.73

0.77

1.00

0.04

0.31

GZ5

0.03

0.01

0.04

0.03

0.04

1.00

0.13

CH6

0.28

0.26

0.27

0.49

0.31

0.13

1.00

51

Table 9. r2 values from a least-squares fit of iron concentrations at a pair of sites.

Fe

TM1

TC2

CW3

SZ4

ZS7

GZ5

CH6

TM1

1.00

0.93

0.91

0.73

0.53

0.01

0.04

TC2

0.93

1.00

0.96

0.70

0.61

0.01

0.07

CW3

0.91

0.96

1.00

0.79

0.65

0.00

0.08

SZ4

0.73

0.70

0.79

1.00

0.59

0.05

0.26

ZS7

0.53

0.61

0.65

0.59

1.00

0.03

0.20

GZ5

0.01

0.01

0.00

0.05

0.03

1.00

0.23

CH6

0.04

0.07

0.08

0.26

0.20

0.23

1.00

Table 10. r2 values from a least-squares fit of elemental carbon concentrations at a pair of sites.

EC

TM1

TC2

CW3

SZ4

ZS7

GZ5

CH6

TM1

1.00

0.53

0.22

0.30

0.44

0.14

0.06

TC2

0.53

1.00

0.66

0.49

0.37

0.06

0.05

CW3

0.22

0.66

1.00

0.22

0.26

0.05

0.00

SZ4

0.30

0.49

0.22

1.00

0.27

0.19

0.00

ZS7

0.44

0.37

0.26

0.27

1.00

0.39

0.03

GZ5

0.14

0.06

0.05

0.19

0.39

1.00

0.01

CH6

0.06

0.05

0.00

0.00

0.03

0.01

1.00

52

Table 11. r2 values from a least-squares fit of lead concentrations at a pair of sites.

Pb

TM1

TC2

CW3

SZ4

ZS7

GZ5

CH6

TM1

1.00

0.73

0.96

0.81

0.40

0.05

0.01

TC2

0.73

1.00

0.79

0.80

0.44

0.18

0.04

CW3

0.96

0.79

1.00

0.79

0.40

0.05

0.01

SZ4

0.81

0.80

0.79

1.00

0.57

0.15

0.00

ZS7

0.40

0.44

0.40

0.57

1.00

0.09

0.03

GZ5

0.05

0.18

0.05

0.15

0.09

1.00

0.16

CH6

0.01

0.04

0.01

0.00

0.03

0.16

1.00

Table 12. r2 values from a least-squares fit of potassium concentrations at a pair of sites.

K

TM1

TC2

CW3

SZ4

ZS7

GZ5

CH6

TM1

1.00

0.86

0.95

0.92

0.64

0.02

0.10

TC2

0.86

1.00

0.86

0.82

0.58

0.02

0.08

CW3

0.95

0.86

1.00

0.85

0.58

0.02

0.08

SZ4

0.92

0.82

0.85

1.00

0.61

0.01

0.02

ZS7

0.64

0.58

0.58

0.61

1.00

0.00

0.05

GZ5

0.02

0.02

0.02

0.01

0.00

1.00

0.21

CH6

0.10

0.08

0.08

0.02

0.05

0.21

1.00

53

Table 13. r2 values from a least-squares fit of rubidium concentrations at a pair of sites.

Rb

TM1

TC2

CW3

SZ4

ZS7

GZ5

CH6

TM1

1.00

0.82

0.93

0.81

0.42

0.06

0.22

TC2

0.82

1.00

0.84

0.79

0.46

0.05

0.18

CW3

0.93

0.84

1.00

0.81

0.39

0.04

0.21

SZ4

0.81

0.79

0.81

1.00

0.41

0.10

0.13

ZS7

0.42

0.46

0.39

0.41

1.00

0.02

0.15

GZ5

0.06

0.05

0.04

0.10

0.02

1.00

0.12

CH6

0.22

0.18

0.21

0.13

0.15

0.12

1.00

54

Table 14. Mean annual concentrations of species at each site TM1 PM2.5

Mean annual concentrations at each site (µg/m3)a CW3 SZ4 GZ5

TC2

CH6

ZS7

28.7

(13.8)

32.5

(14.8)

34.3

(14.1)

47.1

(21.3)

70.6

(30.2)

36.8

(18.5)

46.5

(23.0)

OC

6.9

(3.9)

8.8

(4.7)

9.3

(4.3)

15.6

(7.9)

24.6

(11.6)

12.7

(6.3)

14.7

(9.1)

Sulfate

9.2

(4.7)

9.0

(5.1)

9.3

(4.5)

10.0

(5.3)

14.7

(6.1)

10.4

(5.4)

11.9

(5.7)

Si

0.35

(0.28)

0.31

(0.25)

0.35

(0.26)

0.61

(0.35)

0.88

(0.42)

0.51

(0.31)

0.68

(0.49)

Fe

0.14

(0.10)

0.16

(0.11)

0.17

(0.10)

0.29

(0.14)

0.53

(0.40)

0.27

(0.17)

0.30

(0.19)

EC

0.8

(0.5)

2.0

(1.0)

1.9

(1.0)

3.9

(2.3)

4.4

(1.8)

1.4

(0.5)

2.5

(1.2)

Pb

0.06

(0.05)

0.06

(0.05)

0.05

(0.04)

0.10

(0.08)

0.27

(0.11)

0.16

(0.10)

0.14

(0.12)

K

0.57

(0.43)

0.56

(0.41)

0.56

(0.42)

0.86

(0.55)

1.67

(0.88)

1.44

(0.82)

1.03

(1.01)

(0.003)

0.007

(0.005)

0.016

(0.009)

0.014

(0.009)

0.009

(0.010)

Rb 0.004 (0.004) 0.004 (0.003) 0.004 a Standard deviations are shown in parentheses

Table 15. Enrichment factors (EF) of elements in PM2.5 samples. Values represent a grouping of all measurements made at the seven sampling sites. Compound Silica Iron Lead Potassium Rubidium

Mean (µg/m3) 0.52 0.25 0.12 0.93 0.0081

S.D. (µg/m3) 0.38 0.17 0.11 0.77 0.0079

EF 1.2 3.1 5965.8 29.3 57.9

* Calculation of EF: EF = (CElement/CAluminum)air/(CElement/CAluminum)crust

55

Table 16. Molecular Markers for Major Sources Molecular Markers

Major Urban Sources

n-pentacosane b gasoline vehicles, diesel vehicles gasoline vehicles, diesel vehicles n-hexacosane b n-heptacosane b gasoline vehicles, diesel vehicles n-octacosane a gasoline vehicles, diesel vehicles vegetative detritus n-nonacosane b variety n-triacontane b vegetative detritus, cigarette smoke n-hentriacontane b a variety n-dotriacontane vegetative detritus, cigarette smoke n-triatriacontane c a gasoline vehicles, diesel vehicles 20 S&R-5α(H), 14β(H), 17β(H)-cholestanes gasoline vehicles, diesel vehicles 20R-5α(H), 14α(H), 17 α (H)-cholestane a gasoline vehicles, diesel vehicles 20 S&R-5α(H), 14β(H), 17β(H)-ergostanes a gasoline vehicles, diesel vehicles 20 S&R-5α(H), 14β(H), 17β(H)-sitostanes a gasoline vehicles, diesel vehicles 22,29,30-trisnorneohopane b b gasoline vehicles, diesel vehicles 17α(H), 21 β(H)-29-norhopane gasoline vehicles, diesel vehicles 17α(H), 21 β(H)-hopane a b gasoline vehicles, diesel vehicles 22S-17α(H), 21 β(H)-homohopane gasoline vehicles, diesel vehicles 22R-17α(H), 21 β(H)-homohopane b gasoline vehicles, diesel vehicles 22S-17α(H), 21 β(H)-bishomohopane b gasoline vehicles, diesel vehicles 22R-17α(H), 21 β(H)-bishomohopane b meat cooking cholesterol a a biomass burning levoglucosan softwood burning 8,15-pimaredienoic acid c a softwood burning pimaric acid softwood burning isopimaric acid a hardwood burning propionylsyringol c natural gas, coal combustion and gasoline vehicle benzo(b)fluoranthene a natural gas, coal combustion and gasoline vehicle benzo(k)fluoranthene a c natural gas, coal combustion and gasoline vehicle benzo(e)pyrene natural gas, coal combustion and gasoline vehicle indeno(1,2,3-cd)fluoranthene c c natural gas, coal combustion and gasoline vehicle indeno(1,2,3-cd)pyrene natural gas, coal combustion and gasoline vehicle benzo(ghi)perylene c a noncatalyst cars coronene coal combustion picene c cigarette smoke iso-nonacosane c c cigarette smoke anteiso-triacontane cigarette smoke iso-hentriacontane c elemental carbon diesel exhaust aluminum crustal material silicon crustal material a Identified and quantified by using authentic standard (PMSTD standard and internal standard 1, 2); b Identified by using mass spectra and quantified using authentic standard with similar structure and volatility; c Identified by using secondary standard (picene standard, wood smoke standard, wax extraction, etc) and quantified using authentic standard with similar structure and volatility.

56

Table 17. Concentrations of Organic Compounds, OC and EC in PM2.5 (ng /m3) Compounds heptadecane octadecane nonadecane eicosane heneicosane docosane tricosane tetracosane pentacosane hexacosane heptacosane octacosane nonacosane triacontane hentriacontane dotriacontane tritriacontane tetratriacontane pentatriacontane hexatriacontane iso-nonacosane anteiso-triacontane iso-hentriacontane tetradecanoic acid pentadecanoic acid hexadecanoic acid heptadecanoic acid octadecanoic acid nonadecanoic acid eicosanoic acid heneicosanoic acid docosanoic acid tricosanoic acid tetracosanoic acid pentacosanoic acid hexacosanoic acid heptacosanoic acid octacosanoic acid nonacosanoic acid triacontanoic acid 9-hexadecenoic acid 9,12-octadecanedienoic acid 9-octadecenoic acid 17α(H)-21β(H)-29-norhopane 17α(H)-21β(H)-hopane 22,29,30-trisnorneohopane 22,29,30-trisnorhopane 20S,R-5α(H),14β(H),17β(H)-cholestanes 20R-5α(H),14α(H),17α(H)-cholestane 20S,R-5α(H),14β(H),17β(H)-ergostanes 20S,R-5α(H),14β(H),17β(H)-sitostanes 22S,17α(H),21β(H)-homohopane 22R,17α(H),21β(H)-homohopane 22S,17α(H),21β(H)-bishomohopane 22R,17α(H),21β(H)-bishomohopane 22S,17α(H),21β(H)-trishomohopane

TM-03 TM06

0.06 0.31 0.32 0.62 0.88 1.52 1.68 1.49 1.09 1.78 0.80 2.67 0.69 0.91 0.39 0.35 0.22 0.09 0.08 0.11 2.01 1.49 10.91 0.47 3.53 0.21 0.92 0.32 1.54 1.04 3.09 0.72 1.96 0.46 1.76 0.35 1.45 1.75 0.31 1.78 0.18 0.41

0.07 0.08 0.01 0.08 0.20 0.19 0.35 0.37 0.64 0.37 0.96 0.23 0.76 0.18 0.27 0.10 0.21 0.10 0.12 0.02 0.42 0.41 2.04 0.12 0.43 0.09 0.37 0.17 1.06 0.43 1.39 0.31 0.76 0.21 0.75 0.14 0.47

0.03 0.04

TM10 0.08 0.08 0.13 0.21 0.28 0.33 0.66 0.85 1.64 2.04 2.57 1.70 2.76 0.99 3.41 0.82 1.41 0.34 0.36 0.18 0.55 0.15 0.41 0.44 0.19 2.81 0.21 0.06 0.17 0.85 0.47 2.21 1.54 3.64 0.82 1.80 0.69 3.77 0.78 5.82 0.26 0.06 0.20

0.08 0.01 0.18 0.11 0.13 0.11 0.14

0.02 0.03 0.02 0.02 0.02 0.02

0.12

TM12

0.05 0.11 0.13 0.26 0.30 0.48 0.77 1.33 1.42 2.16 1.70 2.64 1.04 3.38 1.06 1.35 0.41 0.34

TC03

0.33 0.88 1.31 2.96 3.96 4.77 3.68 2.70 1.60 2.23 0.95 3.01 0.88 1.06 0.76 0.46 0.21

TC06

TC10

TC12

0.10 0.37 0.46 0.32 0.29 0.44 0.44 0.63 0.44 0.76 0.37 1.36 0.26 1.08 0.25 0.41 0.10 0.26

0.22 0.07 0.32 0.41 0.53 0.62 1.08 0.96 1.53 1.74 2.75 1.84 2.89 1.21 4.63 1.31 2.29 0.61 0.42

0.25 0.25 0.33 0.43 0.45 0.74 1.86 3.45 5.04 4.68 4.55 3.43 4.37 1.84 4.61 1.77 1.82 0.93 0.75 0.39 1.07 0.55 1.26 0.64 0.23 12.38 0.39 4.10 0.21 1.00 0.43 1.96 1.16 3.02 0.63 1.63 0.62 2.95 0.54 4.34

0.46 0.36 0.16 0.06 2.47 0.16 0.15 0.83 0.43 2.01 1.33 3.41 0.66 1.67 0.61 3.15 0.64 4.94 0.33 0.08 0.30 0.14

0.42 0.55 1.92 1.14 18.44 0.52 6.81 0.25 1.06 0.33 1.58 1.01 2.75 0.64 1.72 0.42 1.67 0.31 1.24 0.83 0.69 2.69 0.49 0.61 0.12 0.12 0.30 0.18 0.20 0.29 0.29 0.26 0.19 0.16 0.11

0.12 0.40 0.30 3.31 0.14 1.36 0.10 0.44 0.19 1.18 0.49 1.41 0.32 0.82 0.24 0.82 0.14 0.50 0.05 0.01 0.14 0.17 0.04 0.06 0.05 0.05 0.04 0.10 0.10 0.07 0.07 0.04 0.05

0.85 0.19 0.50 0.57 0.25 6.34 0.30 2.71 0.20 0.98 0.47 2.30 1.39 3.32 0.74 1.94 0.82 4.58 0.92 7.94 0.32 0.12 0.32 0.50 0.27

0.20 0.09 0.49 0.75 0.38 0.18

0.14 0.16 0.26 0.23 0.25 0.36

0.05 0.20 0.34 0.32 0.26 0.27 0.16 0.22

CW- CW- CW- CW03 06 10 12

0.19 0.93 1.27 3.11 3.59 4.76 4.17 2.88 1.71 2.78 1.12 3.87 0.82 1.49 0.48 0.38 0.51 0.79 4.21 2.60 34.60 1.08 11.38 0.33 1.69 0.46 2.30 1.30 3.67 0.91 2.28 0.57 2.29 0.38 1.38 2.52 2.40 5.85 0.50 0.73 0.21 0.21 0.28 0.24 0.27 0.39 0.36 0.30 0.20 0.19 0.18

0.03 0.11 0.36 0.42 0.24 0.33 0.62 0.55 0.79 0.56 1.05 0.46 1.45 0.40 1.60 0.32 0.58 0.17 0.40

0.38 0.31 0.33 0.38 0.55 0.82 1.17 1.38 1.85 1.86 2.51 1.74 3.05 1.00 4.50 1.14 1.75 0.30 0.50

0.24 0.14 0.30 0.91 0.65 6.95 0.23 2.15 0.11 0.58 0.22 1.27 0.51 1.48 0.33 0.85 0.24 0.87 0.13 0.50 0.07

0.84 0.34 0.59 1.34 0.71 9.65 0.43 2.96 0.18 0.97 0.42 1.98 1.28 2.94 0.62 1.44 0.59 3.21 0.64 4.55 0.47 0.16 0.23 0.21 0.40 0.21

0.36 0.17 0.22 0.06 0.06 0.09 0.09 0.09 0.13 0.14 0.14 0.10 0.04 0.06

0.19 0.21 0.21 0.24 0.16 0.14

CH03

0.22 0.31 0.41 0.58 0.48 0.99 0.91 1.29 0.91 1.10 1.44 3.25 2.90 2.39 3.46 4.67 3.20 5.28 3.64 6.46 2.28 4.44 3.52 9.58 1.39 3.55 5.83 19.44 1.58 3.07 2.13 4.86 0.70 1.33 0.48 1.09 0.66 0.91 0.38 0.60 0.60 1.00 0.94 9.23 0.39 6.06 15.59 71.02 0.51 2.33 5.82 19.17 0.23 0.93 1.35 6.19 0.51 1.77 2.44 9.75 1.38 6.08 3.32 16.26 0.69 3.28 1.73 9.22 0.62 1.80 3.38 9.62 0.58 1.62 4.64 8.16 5.71 2.69 0.86 2.77 5.96 0.51 0.38 0.78 0.68 0.34 0.20 0.17 0.36 0.14 0.25 0.42 0.35 0.27 0.42 0.26 0.30 0.25 0.20 0.26 0.25 0.20

CH06

0.32 0.24 0.12 0.27 1.75 0.78 1.70 1.58 2.12 1.25 2.22 0.86 2.45 0.67 0.87 0.39 0.31 0.28 0.16 0.26 1.39 1.02 31.49 0.62 6.21 0.25 1.57 0.39 2.71 1.02 3.81 0.61 1.95 0.39 1.66 0.27 1.29 0.29 0.51 0.17 0.32

CH10

CH12

0.14 0.25 0.19 0.31 0.47 0.71 1.65 2.10 3.07 3.61 4.56 2.96 4.15 1.64 6.45 1.60 2.29 0.45 0.55

0.15 0.01 0.20 0.20 0.35 0.90 1.70 2.31 3.10 2.61 3.56 2.37 4.27 1.36 5.89 1.37 1.63 0.43

0.57 0.24 0.31 1.98 0.93 18.43 0.67 4.75 0.32 2.19 0.75 4.57 2.33 6.65 1.20 3.35 0.91 6.20 1.06 8.65

0.52 0.31 0.20 1.06 0.60 21.40 0.63 3.90 0.32 2.55 0.72 4.58 2.10 6.12 0.79 2.79 0.59 4.51 0.55 6.86

0.23 0.07 0.30 0.48 0.40

0.49 0.27 0.38 0.78 0.35 0.25

0.12

0.15

0.09

0.11 0.10 0.16 0.09 0.05

0.15 0.21 0.16

ZS-03

0.24 1.22 2.50 5.48 6.26 8.33 8.22 6.17 4.58 5.33 3.46 6.81 2.36 2.31 1.62 1.09 0.66 0.33 0.42 0.55 3.71 1.99 45.61 1.13 15.86 0.53 2.26 0.70 2.93 1.87 4.78 1.04 3.07 0.65 2.60 0.46 1.94 0.84 0.60 2.67 0.71 1.39 0.23 0.18 0.21 0.28 0.20 0.43 0.47 0.42 0.34 0.25 0.26

ZS06

0.33 0.40 0.18 0.26 0.75 0.96 1.53 1.63 2.02 1.48 2.08 0.88 1.58 0.55 0.61 0.31 0.54 0.31 0.14 0.96 0.64 9.31 0.25 2.04 0.17 0.57 0.25 1.31 0.57 1.75 0.36 1.04 0.24 0.84 0.13 0.50 0.03 0.18 0.36 0.07 0.05

0.13 0.13 0.11 0.08 0.07

ZS10

ZS12

0.31 0.32 0.57 1.66 1.94 3.00 6.02 10.60 14.20 13.81 13.62 10.11 10.07 5.92 11.80 5.25 4.84 3.01 2.02 1.49 2.22 1.76 1.73 1.98 1.18 47.71 1.51 20.70 0.57 2.75 0.91 4.55 2.43 6.33 1.16 3.60 0.88 5.34 0.85 7.76

0.16 0.33 0.41 1.22 0.63 1.60 2.91 5.11 7.66 8.01 7.72 6.11 6.60 3.33 7.07 2.88 3.32 1.33 1.19 0.65 1.09 0.64 1.15 1.46 0.84 21.60 0.83 7.02 0.35 1.67 0.66 3.16 1.90 4.57 0.98 2.64 1.12 4.23 0.97 6.29

2.87 4.77 1.84 3.94 0.74 0.54 0.59 0.95 0.69 1.09 1.24 0.94 0.83 0.74 0.59

0.64 1.24 0.80 1.77 0.61 0.33 0.43 0.26 0.48 0.61 0.64 0.48 0.37 0.28

SZ03

0.18 1.20 1.98 5.00 5.66 7.71 6.70 5.12 3.42 4.42 2.69 6.61 2.10 2.46 1.60 1.38 0.98 0.50 0.64 0.88 3.91 2.69 48.47 1.26 17.94 0.47 2.09 0.57 2.52 1.63 4.34 1.00 2.68 0.63 2.32 0.52 1.85 2.82 1.67 5.91 0.96 2.12 0.24 0.27 0.31 0.35 0.38 0.71 0.69 0.64 0.53 0.43 0.33

SZ06 0.06 0.14 0.59 0.75 0.69 0.80 2.37 1.03 1.33 1.42 1.77 1.15 2.19 0.79 2.06 0.56 0.72 0.40

SZ10

SZ12

0.20 0.34 0.23 0.40 0.54 0.64 0.64 0.85 1.02 1.32 1.39 2.40 2.62 4.49 3.61 5.92 5.53 6.65 5.57 5.46 6.18 5.17 4.63 3.50 5.45 4.66 2.88 2.30 7.40 7.18 2.93 2.69 3.42 2.70 1.71 1.55 1.76 0.87 0.87 0.58 1.33 1.22 0.58 0.60 1.07 1.13 1.28 1.42 0.58 0.65 25.40 28.14 0.69 0.74 11.45 9.23 0.36 0.33 1.84 1.90 0.66 0.55 3.21 3.22 1.92 1.61 4.69 4.46 1.03 0.81 2.47 2.45 0.95 0.69 4.49 4.01 0.94 0.75 5.96 5.76

0.35 0.31 0.47 2.16 1.48 22.12 0.61 6.81 0.21 1.00 0.31 1.74 0.75 2.25 0.46 1.21 0.29 1.03 0.18 0.64 0.93 0.07 1.34 1.10 3.31 0.52 1.01 1.07 1.90 0.18 0.52 0.27 0.34 0.18 0.30 0.16 0.25 0.19 0.26 0.39 0.74 0.42 0.71 0.35 0.63 0.35 0.48 0.27 0.41 0.23 0.36

2.40 4.54 1.09 1.88 0.46 0.44 0.35 0.22 0.39 0.61 0.70 0.59 0.61 0.46 0.50

GZ03

GZ06

GZ- GZ-12 10

0.96 2.04 3.77 5.54 11.32 11.20 13.80 11.35 9.40 6.12 8.09 4.65 14.96 3.98 5.17 3.11 2.04 1.29 0.66 1.70 1.87 6.12 4.49 78.06 2.27 27.55 0.87 4.30 1.10 5.56 3.25 8.66 1.84 5.08 1.17 4.87 0.94 4.04 4.22 3.58 11.42 2.05 4.46 0.46 0.52 0.62 0.75 1.05 1.23 1.51 1.35 1.07 0.82 0.73

0.15 0.40 0.93 1.18 1.71 2.31 6.17 3.94 5.43 6.02 6.26 4.75 6.40 3.84 7.46 2.80 3.23 2.02 1.61 1.01 0.62 1.01 1.79 4.66 3.64 50.39 1.53 17.37 0.52 2.71 0.68 4.02 1.75 4.94 1.05 2.63 0.66 2.08 0.42 1.56 3.31 3.27 9.07 2.03 4.37 0.46 0.62 0.55 0.72 1.14 1.59 1.72 1.55 1.20 0.93 0.81

0.21 0.36 0.61 0.83 1.14 1.95 3.64 4.74 7.59 7.75 9.52 6.90 9.99 4.61 14.76 5.44 7.54 3.19 1.92 1.32 2.23 1.36 2.70 1.39 0.78 30.85 1.08 15.87 0.50 3.16 1.04 5.63 2.91 7.50 1.56 3.92 1.35 7.05 1.56 10.71

0.50 0.75 1.30 2.27 3.02 6.08 11.69 15.05 17.11 12.36 11.96 7.26 9.17 4.96 17.03 5.88 7.87 4.46 2.57 0.90 2.61 2.49 3.22 1.61 0.98 58.43 1.56 21.50 0.56 3.68 0.93 5.47 2.44 6.99 1.19 3.37 0.89 5.68 1.02 8.09

1.47 18.78 2.67 23.86 1.85 2.61 4.45 5.25 0.80 0.88 0.46 0.82 0.47 0.85 0.50 0.81 0.67 1.03 1.28 1.83 1.68 1.65 1.49 1.40 1.13 1.15 0.94 1.03 0.87 0.88

57

Table 17. (Cont’d) Compounds 22R,17α(H),21β(H)-trishomohopane nonanal sinapyl aldehyde acetonylsyringol coniferyl aldehyde propionylsyringol benz(de)anthracen-7-one cholesterol levoglucosan 8,15-pimaredienoic acid pimaric acid sandaracopimaric acid isopimaric acid dehydroabietic acid abietic acid abieta-6,8,11,13,15-pentae-18-oic acid abieta-8,11,13,15-tetraen-18-oic acid 7-oxodehydroabietic acid 1,2-benzenedicarboxylic acid 1,4-benzenedicarboxylic acid 1,3-benzenedicarboxylic acid fluoranthene acephenanthrylene pyrene retene methyl substituted MW 202 PAH benzo(ghi)fluoranthene cyclopenta(cd)pyrene benz(a)anthracene chrysene/triphenylene methyl substituted MW 228 PAH methyl substituted MW 226 PAH benzo(b)fluoranthene benzo(k)fluoranthene benzo(j)fluoranthene benzo(e)pyrene benzo(a)pyrene perylene indeno(cd)fluoranthene indeno(cd)pyrene picene benzo(ghi)perylene coronene propanedioic acid methylpropanedioic acid butanedioic acid methylbutanedioic acid pentanedioic acid hexanedioic acid heptanedioic acid octanedioic acid nonanedioic acid OC EC

TM-03 TM06

TM10

TM12

TC03

TC06

TC10

TC12

CW- CW- CW- CW03 06 10 12

0.07 2.23

0.01 0.34

5.99

4.34

0.12 4.37

5.09

5.18

0.07 0.99

0.02 0.31

3.44

2.36

1.78

3.07 0.35 4.63 0.41 10.17 0.47 0.19 7 115

2.49

5.46 2.32 109 0.05 0.11 0.18 1.29 0.11 0.04 0.93 7.21 6.31 0.50 0.37 0.09 0.36 0.37 0.16 1.50 0.18 0.35 0.20 0.16 0.39 0.31 0.07 0.37 0.28 0.05 0.22 0.53 0.07 0.47 0.30 15.59 0.30 19.55 2.37 5.46 1.42 0.43 0.46 1.23 4847 698

0.05 0.11 0.01 0.11 1.74 0.90 0.14 0.05 0.01 0.04 0.06 0.02 0.11 0.02 0.05 0.03 0.02 0.04 0.03 0.04 0.03 0.02 0.06 0.01 0.06 0.04 2.50 0.06 2.85 0.29 0.96 0.25 0.10 0.11 0.29 3518 597

0.07 0.07 0.29 1.01 0.22 0.27 0.14 0.81 4.63 4.61 0.66 0.32 0.07 0.33 0.01 0.35 0.17 2.20 0.16 0.35 0.25 0.26 0.43 0.34 0.08 0.40 0.33 0.06 0.23 0.54 0.06 0.44 0.33 15.96 0.31 6.18 0.59 2.37 0.95 0.33 0.47 1.35 5710 890

4.81 10.33 9.62 0.58 1.03 207 75 0.07 0.08 0.28 0.99

10.64 5.61 0.74 14.58 16.58 6.20 1.55 0.31 0.92 1.99 10 299 250 92

0.07 0.07 0.14 0.12 0.18 0.04 0.17 0.61 2.02 0.19 1.05 1.75 1.96 0.10 0.13 0.35 0.09 0.01 0.27 0.36 0.11 0.15 0.03 0.13 0.17 0.03 0.81 0.91 0.11 0.70 0.84 1.00 11.16 3.98 2.26 6.38 9.70 6.52 7.38 4.09 1.32 7.86 6.85 6.27 0.81 0.49 0.22 0.68 0.93 0.86 0.45 0.47 0.09 0.38 0.49 0.41 0.08 0.10 0.01 0.09 0.10 0.08 0.44 0.52 0.09 0.40 0.49 0.43 0.02 0.02 0.01 0.03 0.03 0.02 0.36 0.45 0.10 0.42 0.48 0.43 0.20 0.32 0.04 0.21 0.28 0.22 2.18 2.28 0.22 2.89 2.87 1.55 0.18 0.30 0.04 0.19 0.26 0.22 0.38 0.61 0.09 0.44 0.57 0.46 0.41 0.41 0.05 0.40 0.55 0.35 0.22 0.24 0.04 0.25 0.35 0.18 0.42 0.62 0.06 0.66 0.76 0.51 0.42 0.63 0.06 0.58 0.76 0.39 0.08 0.11 0.01 0.13 0.13 0.06 0.49 0.62 0.07 0.66 0.77 0.43 0.31 0.50 0.05 0.48 0.54 0.31 0.08 0.07 0.01 0.11 0.13 0.03 0.30 0.27 0.02 0.27 0.37 0.19 0.61 0.75 0.08 0.85 0.84 0.47 0.04 0.10 0.01 0.08 0.06 0.06 0.55 0.75 0.08 0.80 0.89 0.52 0.35 0.54 0.05 0.56 0.54 0.31 20.16 5.60 4.11 18.57 16.82 4.92 0.34 0.25 0.26 0.15 15.83 7.75 3.65 6.61 10.38 15.14 1.44 1.13 0.36 0.56 1.03 1.88 4.51 3.08 0.87 1.81 2.80 5.56 1.72 1.01 0.33 1.19 1.43 1.71 0.47 0.40 0.13 0.41 0.39 0.67 0.49 0.48 0.18 0.61 0.68 1.01 1.49 2.31 0.78 2.38 3.32 5.56 5534 6873 4057 7504 6716 7479 1118 2137 1233 2023 2560 1854

0.04 0.67

1.00 1.22 24 0.02 0.05 0.12 0.40 0.02 0.01 0.01 0.19 1.37 1.29 0.27 0.10 0.01 0.10 0.11 0.05 0.22 0.05 0.11 0.09 0.03 0.08 0.06 0.01 0.07 0.04 0.01 0.04 0.08 0.01 0.09 0.06 3.95 0.42 1.24 0.38 0.17 0.32 1.47 4683 1346

8.01

4.95

3.23

5.19

5.09 7.76 7.69 12.77 1.25 2.81 129 255 0.10 0.08

0.15

1.07 1.38 0.09 0.26 0.30 0.12 0.12 0.59 0.84 4.28 10.28 5.02 8.01 0.69 1.00 0.35 0.46 0.07 0.08 0.34 0.49 0.25 0.15 1.51 0.15 0.37 0.28 0.13 0.43 0.38 0.06 0.45 0.33 0.07 0.19 0.45

0.51 0.25 2.50 0.22 0.48 0.49 0.25 0.72 0.58 0.11 0.70 0.46 0.12 0.31 0.62

0.47 0.24 17.60 0.27 5.65 0.48 1.39 1.13 0.32 0.63 3.12 6407 2001

0.64 0.34 18.78 0.31 12.53 1.04 3.31 1.67 0.63 0.98 5.29 7833 2432

CH03

CH06

CH10

CH12

12.53 5.51 11.37 16.12 8.01 9.39 2.28 14.69 23.98 0.62 15.35 17.87 24.57 6.48 15.07 27.09 3.75 0.42 0.47 0.38 543.1 67.6 322.7 470.1 0.40 0.23 0.06 0.16 0.49 0.44 0.15 0.12 0.25 1.47 8.97 2.20 2.48 5.19 0.24 0.05 0.13 0.35 0.49 0.09 0.40 0.66 0.16 0.04 0.21 0.35 5.20 1.60 1.35 2.24 15.51 3.93 5.92 7.64 17.38 5.51 5.08 6.06 1.21 0.37 0.58 0.50 1.60 0.19 0.48 0.79 0.23 0.05 0.11 0.15 1.58 0.21 0.50 0.85 0.06 0.01 0.04 1.16 0.33 0.45 0.76 0.63 0.15 0.25 0.44 5.21 1.42 2.99 5.28 0.65 0.14 0.27 0.46 2.11 0.28 0.60 1.29 0.83 0.22 0.53 0.99 0.57 0.19 0.46 0.63 2.99 0.45 1.17 3.22 1.75 0.37 0.77 2.08 0.25 0.08 0.16 0.35 2.45 0.46 1.27 3.34 1.37 0.30 0.70 1.43 0.20 0.06 0.17 0.31 0.94 0.31 0.43 0.91 2.86 0.70 1.23 2.35 0.46 0.10 0.10 0.21 2.70 0.80 1.28 2.62 1.92 0.70 0.66 1.14 14.90 3.14 24.65 12.43 0.43 0.09 0.34 0.25 24.91 4.45 8.43 10.18 4.80 0.76 0.90 1.36 7.17 1.33 2.33 2.98 3.43 1.29 2.40 1.82 1.81 0.64 0.78 0.57 3.16 1.13 1.20 1.12 12.27 5.06 5.07 5.50 12541 8119 8840 7762 1585 1163 1513 1481

ZS-03

ZS06

ZS10

ZS12

SZ03

SZ06

0.17 11.31

0.06 0.38

0.37 5.41

0.29 5.59

0.21 4.95

0.23 0.33 4.23 10.46

9.69

0.54 25.15 1.71 29.85 4.79 118.77 0.42 1.37 21 521 0.34 0.04 1.53 0.07 0.31 1.84 1.31 16.65 0.73 0.08 1.05 0.02 0.71 1.13 4.10 1.91 19.35 3.28 21.17 0.26 1.37 0.10 1.41 0.03 0.44 0.13 1.73 0.14 0.20 1.71 0.12 1.46 0.95 26.56 0.14 1.72 0.27 2.92 0.28 2.71 0.15 2.46 0.44 5.34 0.36 5.61 0.06 1.16 0.37 5.45 0.24 4.54 0.08 0.80 0.22 2.41 0.55 7.25 0.07 0.66 0.52 7.23 0.35 4.11 0.67 14.34 0.09 0.25 2.59 12.69 0.41 1.57 0.71 3.80 0.47 2.37 0.21 0.65 0.40 1.14 1.15 5.72 4149 12019 1401 2121

20.47

2.61

1.79 13.05 21.01

120.90 1.74 199 0.16 0.77 9.85 0.21 0.33 0.12 3.63 21.56 43.44 2.75 0.89 0.26 1.07 0.09 1.16 0.87 7.53 1.11 1.91 1.07 0.97 3.09 2.47 0.42 2.40 1.86 0.32 1.08 3.20 0.40 3.24 2.43 5.46 0.24 15.58 2.52 4.03 1.94 1.15 1.17 4.76 11293 2474

20.89 52.61 45.20 3.68 541 146 0.51 0.10 6.31 0.38 0.75 0.36 2.26 14.16 17.19 0.95 0.71 0.18 0.81 0.06 0.81 0.69 9.34 0.81 1.39 1.27 0.94 2.33 2.53 0.56 2.64 2.17 0.40 1.14 2.95 0.28 3.11 1.66 20.94 0.37 10.34 1.01 2.93 1.84 0.57 0.88 4.08 15116 3948

0.25 1.25 0.88 15.08 0.46 0.28 0.13 3.85 4.39 13.72 0.87 0.71 0.21 0.83 0.06 0.80 0.65 6.84 0.81 1.42 1.03 0.82 2.40 2.23 0.35 2.35 1.58 0.28 1.12 2.99 0.32 3.12 2.06 5.09 0.19 10.85 1.72 4.06 1.59 0.81 0.95 5.14 12155 3958

SZ10

SZ12

GZ03

0.37 0.49 5.28 11.68 6.55

GZ06

GZ10

GZ12

0.70 0.57 4.83 24.17

0.50 5.51

22.26 33.23 0.66 14.83 16.32 24.99 34.03 9.28 59.80 60.50 57.28 17.43 69.69 102.02 1.81 3.22 3.44 7.90 10.65 1.70 3.80 54 315 350 430 255 534 814 0.05 0.80 1.78 0.23 0.15 0.19 0.47 1.19 0.30 0.96 1.49 1.38 4.11 10.26 11.79 14.06 0.06 0.49 1.18 0.43 0.09 0.75 0.71 0.48 0.04 0.50 0.52 0.23 1.85 2.89 2.57 5.05 1.74 6.59 8.86 10.67 2.77 19.57 13.47 21.54 0.42 1.16 0.99 1.93 0.13 0.70 0.84 1.09 0.04 0.20 0.24 0.30 0.13 0.86 0.99 1.27 0.02 0.08 0.08 0.11 0.21 0.84 1.18 1.42 0.09 0.63 0.67 0.96 1.20 9.05 11.67 9.51 0.11 0.65 0.85 1.20 0.22 1.28 1.58 2.13 0.23 1.29 1.51 1.82 0.24 0.96 1.12 1.34 0.52 2.44 3.25 3.54 0.52 2.46 2.74 3.10 0.06 0.51 0.64 0.53 0.57 2.63 2.74 3.43 0.29 1.92 2.26 2.35 0.05 0.39 0.43 0.44 0.36 1.22 1.20 1.50 0.83 3.42 3.14 4.48 0.10 0.29 0.32 0.47 0.87 3.46 3.19 4.97 0.64 2.06 1.82 3.87 1.38 24.94 17.07 22.14 0.29 0.30 3.48 9.45 14.97 22.10 0.44 0.97 1.33 2.86 1.82 3.11 3.67 4.93 0.56 1.83 2.06 2.48 0.37 0.49 0.46 1.65 0.72 0.90 0.92 1.89 3.62 4.49 4.29 11.26 6640 12321 9390 16961 2493 4059 3743 4495

2.02

0.26 0.63 1.05 7.11 0.15 0.24 0.09 3.16 5.53 10.87 0.94 0.31 0.09 0.35 0.03 0.50 0.27 2.85 0.33 0.63 0.69 0.45 1.52 1.38 0.25 1.67 1.11 0.23 0.81 2.45 0.27 2.89 2.40 10.88 0.23 8.91 0.87 1.66 0.92 0.62 0.95 4.58 13789 3047

0.59 0.13 0.53 9.50 0.29 1.20 0.54 3.85 10.06 27.48 1.63 0.91 0.23 1.05 0.07 0.90 0.65 10.04 0.74 1.48 1.66 0.83 3.44 3.39 0.54 3.91 2.23 0.42 1.77 4.80 0.46 5.45 3.10 26.37 0.37 9.84 1.04 2.60 1.97 0.72 1.21 6.76 20446 4672

2.09 0.54 2.49 16.19 1.28 1.25 0.89 4.17 9.11 28.90 1.66 1.11 0.31 1.32 0.13 1.24 1.16 20.29 1.56 2.54 2.78 1.73 4.96 5.26 1.22 5.43 4.87 0.94 2.13 5.80 0.59 6.81 3.39 14.59 0.30 16.17 1.95 3.54 1.93 0.61 1.15 6.14 19235 5294

Note: TM: Tap Mun, TC: Tung Chung, CW: Central and Western, CH: Conghua, ZS: Zhongshan, SZ: Shenzhen, GZ: Guangzhou

58

Table 18. Source Contributions to Organic Carbon in PM2.5 (µg /m3) TM

Spring Summer Fall Winter Mean

Diesel Exhaust 0.38±0.13 0.29±0.04 0.48±0.15 0.58±0.15 0.43±0.13

TC

Spring Summer Fall Winter Mean

1.40±0.20 0.68±0.11 1.14±0.20 1.70±0.24 1.23±0.43

0.50±0.09 0.07±0.03 0.38±0.08 0.43±0.10 0.35±0.19

0.22±0.08 0.07±0.02 1.31±0.36 0.94±0.26 0.64±0.59

0.16±0.03 0.03±0.008 0.10±0.02 0.26±0.05 0.14±0.10

0.55±0.16 0.08±0.02 2.20±0.62 1.90±0.54 1.18±1.03

0.28±0.02 0.29±0.02 0.44±0.02 0.30±0.02 0.33±0.08

0.05±0.02 0.03±0.007 0.12±0.02 0.07±0.03 0.07±0.04

0.30±0.04 0.03±0.005 0.27±0.04 0.34±0.05 0.24±0.14

3.40 2.77 1.55 0.79 2.13

6.87±0.49 4.06±0.38 7.50±0.56 6.72±0.50 6.29±1.52

0.95 0.96 0.98 0.96 0.96

2.34 1.66 1.22 2.59 1.95

50 32 79 88 62

21 21 19 18 20

CW

Spring Summer Fall Winter Mean

1.29±0.20 0.77±0.12 1.36±0.20 2.03±0.38 1.36±0.52

0.78±0.10 0.15±0.04 0.32±0.09 0.68±0.13 0.48±0.30

0.89±0.20 0.14±0.04 1.87±0.50 1.10±0.30 1.00±0.71

0.21±0.04 0.07±0.01 0.15±0.03 0.34±0.09 0.19±0.11

0.69±0.20 0.18±0.05 1.04±0.29 2.01±0.57 0.98±0.77

0.31±0.02 0.34±0.01 0.51±0.02 0.35±0.02 0.38±0.09

0.06±0.02 0.03±0.009 0.11±0.02 0.09±0.04 0.07±0.04

0.19±0.03 0.03±0.005 0.06±0.09

3.06 2.97 1.04 1.23 2.08

7.48±0.53 4.68±0.43 6.40±0.50 7.83±0.55 6.60±1.42

0.97 0.97 0.98 0.97 0.97

1.58 1.66 2.51 2.25 1.99

59 37 84 84 66

22 22 13 14 18

CH

Spring Summer Fall Winter Mean

0.66±0.15 0.56±0.19 0.31±0.35

0.60±0.14 0.17±0.06 0.54±0.11 0.72±0.14 0.51±0.24

2.12±0.51 1.30±0.35 1.66±0.45 3.70±0.99 2.20±1.06

0.29±0.05 0.07±0.01 0.08±0.02 0.06±0.02 0.13±0.11

3.42±1.00 0.50±0.15 2.46±0.69 3.49±0.97 2.47±1.39

0.42±0.04 0.47±0.02 0.81±0.03 0.41±0.02 0.53±0.19

0.44±0.08 0.06±0.01 0.18±0.03 0.14±0.03 0.21±0.16

1.23±0.13 0.29±0.03 0.46±0.06 1.16±0.12 0.79±0.48

4.01 4.60 2.09 NV 2.68

12.54±0.85 8.12±0.64 8.84±0.63 7.76±0.58 9.32±2.20

0.95 0.98 0.97 0.96 0.97

2.01 2.2 2.7 2.32 2.31

68 43 76 125 78

19 17 18 20 19

ZS

Spring Summer Fall Winter Mean

0.64±0.25 1.08±0.13 1.73±0.33 0.86±0.73

0.77±0.11 3.16±0.34 1.27±0.19 1.30±1.35

0.85±0.24 0.11±0.03 1.11±0.32 1.18±0.33 0.81±0.49

0.15±0.03 0.03±0.009 0.51±0.11 0.27±0.06 0.24±0.20

1.27±0.36 0.20±0.06 3.37±1.05 4.11±1.17 2.24±1.81

0.68±0.02 0.37±0.01 0.96±0.04 0.98±0.03 0.75±0.29

0.13±0.03 0.04±0.01 0.20±0.07 0.15±0.04 0.13±0.07

1.48±0.16 0.25±0.03 2.30±0.22 1.25±0.15 1.32±0.84

5.32 2.07 0.03 4.19 2.90

11.29±0.76 4.15±0.37 11.65±0.86 15.12±1.00 10.56±4.61

0.98 0.98 0.98 0.99 0.98

1.78 1.96 2.38 1.83 1.99

53 50 100 72 69

18 16 20 18 18

SZ

Spring Summer Fall Winter Mean

2.13±0.34 1.48±0.22 1.96±0.34 2.68±0.47 2.06±0.50

0.94±0.15 0.42±0.08 1.00±0.13 0.95±0.17 0.83±0.27

0.70±0.21 0.90±0.25 2.28±0.63 1.11±0.31 1.25±0.71

0.24±0.05 0.16±0.02 0.25±0.05 0.28±0.06. 0.23±0.05

0.92±0.27 0.41±0.12 2.41±0.69 2.70±0.77 1.61±1.12

0.61±0.03 0.37±0.02 0.85±0.03 0.71±0.04 0.64±0.20

0.11±0.03 0.14±0.04 0.11±0.04 0.09±0.06

1.24±0.14 0.32±0.04 1.39±0.15 1.47±0.16 1.11±0.53

5.27 2.59 2.05 1.65 2.89

12.16±0.81 6.64±0.53 12.33±0.89 11.65±0.92 10.70±2.72

0.97 0.97 0.99 0.98 0.98

2.14 1.59 1.48 1.13 1.59

57 61 83 86 72

21 22 20 21 21

GZ

Spring Summer Fall Winter Mean

1.95±0.38 1.49±0.28 1.81±0.38 1.57±0.44 1.71±0.20

2.69±0.32 2.60±0.30 2.28±0.29 3.72±0.43 2.82±0.62

1.32±0.41 0.74±0.22 5.37±1.47 1.13±0.32 2.14±2.17

0.69±0.11 0.52±0.09 0.62±0.12 1.04±0.19 0.72±0.23

3.14±0.94 1.83±0.54 4.00±1.16 5.97±1.75 3.74±1.74

0.69±0.04 1.05±0.03 0.94±0.04 0.97±0.05 0.91±0.16

0.22±0.07 0.10±0.05 0.29±0.08 0.23±0.09 0.21±0.08

1.71±0.20 0.79±0.10 1.92±0.22 2.63±0.30 1.76±0.76

4.55 4.68 3.23 1.99 3.61

16.96±1.15 13.79±0.92 20.45±1.33 19.24±1.27 17.61±2.93

0.96 0.98 0.98 0.97 0.97

2.79 2.29 1.72 2.75 2.39

73 66 84 90 78

23 22 20 22 22

Site

Season

Gasoline Exhaust 0.37±0.09 0.25±0.10 0.38±0.10 0.25±0.18

Meat Cooking 0.22±0.06 0.07±0.02 0.74±0.21 0.50±0.14 0.38±0.30

Cigarette Smoke 0.03±0.006 0.08±0.02 0.08±0.02 0.05±0.04

Biomass Burning 0.78±0.23 0.05±0.02 0.86±0.25 1.53±0.43 0.81±0.61

Coal Combustion 0.19±0.03 0.02±0.004 0.20±0.03 0.18±0.03 0.15±0.09

Other OC 2.45 2.73 2.46 1.79 2.36

Measured Organic Carbon 4.85±0.39 3.52±0.35 5.71±0.47 5.50±0.42 4.90±0.99

R2

χ2

0.36±0.02 0.32±0.01 0.54±0.02 0.38±0.02 0.40±0.10

Vegetative Detritus 0.07±0.01 0.03±0.005 0.10±0.02 0.08±0.02 0.07±0.03

0.97 0.95 0.98 0.98 0.97

2.07 2.32 2.5 1.35 2.06

% Mass 50 23 57 68 49

Road Dust

Note: -, the contribution to OC is not significantly different from zero; NV, negative value.

59

DF 16 19 14 13 16

Table 19. Percentage of Source Contributions to measured OC in PM2.5 (%)

TM

Spring Summer Fall Winter Mean

Diesel Exhaust 7.8 8.2 8.4 10.5 8.8

TC

Spring Summer Fall Winter Mean

20.4 16.7 15.2 25.3 19.4

CW

Spring Summer Fall Winter Mean

CH

Gasoline Exhaust 7.6 4.4 6.9 4.7

4.5 2.0 13.0 9.1 7.1

Cigarette Smoke 0.6 1.4 1.5 0.9

7.3 1.7 5.1 6.4 5.1

3.2 1.7 17.5 14.0 9.1

2.3 0.7 1.3 3.9 2.1

17.2 16.5 21.3 25.9 20.2

10.4 3.2 5.0 8.7 6.8

11.9 3.0 29.2 14.0 14.5

Spring Summer Fall Winter Mean

8.1 6.3 3.6

4.8 2.1 6.1 9.3 5.6

ZS

Spring Summer Fall Winter Mean

5.7 26.0 11.4 10.8

SZ

Spring Summer Fall Winter Mean

GZ

Spring Summer Fall Winter Mean

Site

Season

Biomass Burning 16.1 1.4 15.1 27.8 15.1

7.4 9.1 9.5 6.9 8.2

Vegetative Detritus 1.4 0.9 1.8 1.5 1.4

8.0 2.0 29.3 28.3 16.9

4.1 7.1 5.9 4.5 5.4

0.7 0.7 1.6 1.0 1.0

4.4 0.7 3.6 5.1 3.4

49.5 68.2 20.7 11.8 37.5

2.8 1.5 2.3 4.3 2.7

9.2 3.8 16.3 25.7 13.7

4.1 7.3 8.0 4.5 6.0

0.8 0.6 1.7 1.1 1.1

2.5 0.6 0.8

40.9 63.5 16.3 15.7 34.1

16.9 16.0 18.8 47.7 24.8

2.3 0.9 0.9 0.8 1.2

27.3 6.2 27.8 45.0 26.6

3.3 5.8 9.2 5.3 5.9

3.5 0.7 2.0 1.8 2.0

9.8 3.6 5.2 14.9 8.4

32.0 56.7 23.6 NV 28.1

6.8 27.1 8.4 10.6

7.5 2.7 9.5 7.8 6.9

1.3 0.7 4.4 1.8 2.1

11.2 4.8 28.9 27.1 18.0

6.0 8.9 8.2 6.5 7.4

1.2 1.0 1.7 1.0 1.2

13.1 6.0 19.7 8.3 11.8

47.1 49.9 0.3 27.7 31.2

17.5 22.3 15.9 23.0 19.7

7.7 6.3 8.1 8.2 7.6

5.8 13.6 18.5 9.5 11.8

2.0 2.4 2.0 2.4 2.2

7.6 6.2 19.5 23.2 14.1

5.0 5.6 6.9 6.1 5.9

0.9 1.1 0.9 0.7

10.2 4.8 11.3 12.6 9.7

43.3 39.0 16.6 14.2 28.3

11.5 10.8 8.9 8.2 9.8

15.9 18.9 11.1 19.3 16.3

7.8 5.4 26.3 5.9 11.3

4.1 3.8 3.0 5.4 4.1

18.5 13.3 19.6 31.0 20.6

4.1 7.6 4.6 5.0 5.3

1.3 0.7 1.4 1.2 1.2

10.1 5.7 9.4 13.7 9.7

26.8 33.9 15.8 10.3 21.7

Meat Cooking

Road Dust

Coal Combustion

Other OC

3.9 0.6 3.5 3.3 2.8

50.5 77.6 43.1 32.5 50.9

Note: -, the contribution to OC is not significantly different from zero; NV, negative value.

60

Figure 1. Sampling and Meteorological Measurement Sites in the PRD

61

Figure 2. Schematic Diagrams of both the ThermoAndersen and Caltech samplers used to collect daily 24-hr filter samples

62

Figure 3. Annual Mean Concentrations Across the Pearl River Delta Region 63

Figure 4. Monthly Mean Concentrations During October 2002

64

Figure 5. Monthly Mean Concentrations During December 2002

65

Figure 6. Monthly Mean Concentrations During March 2003

66

Figure 7. Monthly Mean Concentrations During June 2003 67

Figure 8. Daily Concentrations During October 2002

68

Figure 9. Daily Concentrations During December 2002

69

Figure 10. Daily Concentrations During March 2003

70

Figure 11. Daily Concentrations During June 2003

71

HKUST: Oct 1-2

100

100

8 m/s (N)

90

90

70 60

0 m/s

50 40 30

4 m/s (S)

wind speed and direction

80

mm rain

4 m/s (N)

70 60

0 m/s

50 40 30

4 m/s (S)

20

20

10 8 m/s (S) 10/1/02 0:00

10/2/02 0:00

10 8 m/s (S)

0 10/3/02 0:00

10/1/02 0:00

date

Chek Lap Kok: October 1-2

Tap Mun: October 1-2 8 m/s (N)

4 m/s (N)

4 m/s (N)

wind speed and direction

wind speed and direction

0 10/3/02 0:00

date

8 m/s (N)

0 m/s

4 m/s (S)

8 m/s (S) 10/1/02 0:00

10/2/02 0:00

0 m/s

4 m/s (S)

8 m/s (S) 10/2/02 0:00

10/3/02 0:00

10/1/02 0:00

date

Shenzhen: Oct 1-2 8 m/s (N)

10/2/02 0:00

10/3/02 0:00

date

Guangzhou: Oct 1-2

100

100

8 m/s (N)

90

90

70 60

0 m/s

50 40 30

4 m/s (S)

wind speed and direction

80

mm rain

wind speed and direction

80 4 m/s (N)

4 m/s (N)

70 60

0 m/s

50 40 30

4 m/s (S)

20

20

10 8 m/s (S) 10/1/02 0:00

10/2/02 0:00

date

0 10/3/02 0:00

mm rain

wind speed and direction

80 4 m/s (N)

mm rain

Fogang: Oct 1-2 8 m/s (N)

10 8 m/s (S) 10/1/02 0:00

10/2/02 0:00

0 10/3/02 0:00

date

Figure 12. Wind Speed, Direction and Precipitation During October 1-2, 2002 72

HKUST: Oct 7-8

100

100

8 m/s (N)

90

90

70 60

0 m/s

50 40 30

4 m/s (S)

wind speed and direction

80

mm rain

4 m/s (N)

70 60

0 m/s

50 40 30

4 m/s (S)

20

20

10 8 m/s (S) 10/7/02 0:00

10/8/02 0:00

10 8 m/s (S)

0 10/9/02 0:00

10/7/02 0:00

date

Chek Lap Kok: October 7-8

Tap Mun: October 7-8 8 m/s (N)

4 m/s (N)

4 m/s (N)

wind speed and direction

wind speed and direction

0 10/9/02 0:00

date

8 m/s (N)

0 m/s

4 m/s (S)

8 m/s (S) 10/7/02 0:00

10/8/02 0:00

0 m/s

4 m/s (S)

8 m/s (S) 10/8/02 0:00

10/9/02 0:00

10/7/02 0:00

date

Shenzhen: Oct 7-8 8 m/s (N)

10/8/02 0:00

10/9/02 0:00

date

Guangzhou: Oct 7-8

100

100

8 m/s (N)

90

90

70 60

0 m/s

50 40 30

4 m/s (S)

wind speed and direction

80

mm rain

wind speed and direction

80 4 m/s (N)

4 m/s (N)

70 60

0 m/s

50 40 30

4 m/s (S)

20

20

10 8 m/s (S) 10/7/02 0:00

10/8/02 0:00

date

0 10/9/02 0:00

mm rain

wind speed and direction

80 4 m/s (N)

mm rain

Fogang: Oct 7-8 8 m/s (N)

10 8 m/s (S) 10/7/02 0:00

10/8/02 0:00

0 10/9/02 0:00

date

Figure 13. Wind Speed, Direction and Precipitation During October 7-8, 2002 73

HKUST: Oct 13-14

100

100

8 m/s (N)

90

90

70 60

0 m/s

50 40 30

4 m/s (S)

wind speed and direction

80

mm rain

4 m/s (N)

70 60

0 m/s

50 40 30

4 m/s (S)

20

20

10 8 m/s (S) 10/13/02 0:00

10/14/02 0:00

10 8 m/s (S)

0 10/15/02 0:00

10/13/02 0:00

date

Chek Lap Kok: October 13-14

Tap Mun: October 13-14 8 m/s (N)

4 m/s (N)

4 m/s (N)

wind speed and direction

wind speed and direction

0 10/15/02 0:00

date

8 m/s (N)

0 m/s

4 m/s (S)

8 m/s (S) 10/13/02 0:00

10/14/02 0:00

0 m/s

4 m/s (S)

8 m/s (S) 10/14/02 0:00

10/15/02 0:00

10/13/02 0:00

date

Shenzhen: Oct 13-14 8 m/s (N)

10/14/02 0:00

10/15/02 0:00

date

Guangzhou: Oct 13-14

100

100

8 m/s (N)

90

90

70 60

0 m/s

50 40 30

4 m/s (S)

wind speed and direction

80

mm rain

wind speed and direction

80 4 m/s (N)

4 m/s (N)

70 60

0 m/s

50 40 30

4 m/s (S)

20

20

10 8 m/s (S) 10/13/02 0:00

10/14/02 0:00

date

0 10/15/02 0:00

mm rain

wind speed and direction

80 4 m/s (N)

mm rain

Fogang: Oct 13-14 8 m/s (N)

10 8 m/s (S) 10/13/02 0:00

10/14/02 0:00

0 10/15/02 0:00

date

Figure 14. Wind Speed, Direction and Precipitation During October 13-14, 2002 74

HKUST: Oct 19-20

100

100

8 m/s (N)

90

90

70 60

0 m/s

50 40 30

4 m/s (S)

wind speed and direction

80

mm rain

4 m/s (N)

70 60

0 m/s

50 40 30

4 m/s (S)

20

20

10 8 m/s (S) 10/19/02 0:00

10/20/02 0:00

10 8 m/s (S)

0 10/21/02 0:00

10/19/02 0:00

date

Chek Lap Kok: October 19-20

Tap Mun: October 19-20 8 m/s (N)

4 m/s (N)

4 m/s (N)

wind speed and direction

wind speed and direction

0 10/21/02 0:00

date

8 m/s (N)

0 m/s

4 m/s (S)

8 m/s (S) 10/19/02 0:00

10/20/02 0:00

0 m/s

4 m/s (S)

8 m/s (S) 10/20/02 0:00

10/21/02 0:00

10/19/02 0:00

date

Shenzhen: Oct 19-20 8 m/s (N)

10/20/02 0:00

10/21/02 0:00

date

Guangzhou: Oct 19-20

100

100

8 m/s (N)

90

90

70 60

0 m/s

50 40 30

4 m/s (S)

wind speed and direction

80

mm rain

wind speed and direction

80 4 m/s (N)

4 m/s (N)

70 60

0 m/s

50 40 30

4 m/s (S)

20

20

10 8 m/s (S) 10/19/02 0:00

10/20/02 0:00

date

0 10/21/02 0:00

mm rain

wind speed and direction

80 4 m/s (N)

mm rain

Fogang: Oct 19-20 8 m/s (N)

10 8 m/s (S) 10/19/02 0:00

10/20/02 0:00

0 10/21/02 0:00

date

Figure 15. Wind Speed, Direction and Precipitation During October 19-20, 2002 75

HKUST: Oct 25-26

100

100

8 m/s (N)

90

90

70 60

0 m/s

50 40 30

4 m/s (S)

wind speed and direction

80

mm rain

4 m/s (N)

70 60

0 m/s

50 40 30

4 m/s (S)

20

20

10 8 m/s (S) 10/25/02 0:00

10/26/02 0:00

10 8 m/s (S)

0 10/27/02 0:00

10/25/02 0:00

date

Chek Lap Kok: October 25-26

Tap Mun: October 25-26 8 m/s (N)

4 m/s (N)

4 m/s (N)

wind speed and direction

wind speed and direction

0 10/27/02 0:00

date

8 m/s (N)

0 m/s

4 m/s (S)

8 m/s (S) 10/25/02 0:00

10/26/02 0:00

0 m/s

4 m/s (S)

8 m/s (S) 10/26/02 0:00

10/27/02 0:00

10/25/02 0:00

date

Shenzhen: Oct 25-26 8 m/s (N)

10/26/02 0:00

10/27/02 0:00

date

Guangzhou: Oct 25-26

100

100

8 m/s (N)

90

90

70 60

0 m/s

50 40 30

4 m/s (S)

wind speed and direction

80

mm rain

wind speed and direction

80 4 m/s (N)

4 m/s (N)

70 60

0 m/s

50 40 30

4 m/s (S)

20

20

10 8 m/s (S) 10/25/02 0:00

10/26/02 0:00

date

0 10/27/02 0:00

mm rain

wind speed and direction

80 4 m/s (N)

mm rain

Fogang: Oct 25-26 8 m/s (N)

10 8 m/s (S) 10/25/02 0:00

10/26/02 0:00

0 10/27/02 0:00

date

Figure 16. Wind Speed, Direction and Precipitation During October 25-26, 2002 76

HKUST: Nov 30 - Dec 1

100

100

8 m/s (N)

90

90

70 60

0 m/s

50 40 30

4 m/s (S)

wind speed and direction

80

mm rain

4 m/s (N)

70 60

0 m/s

50 40 30

4 m/s (S)

20

20

10 8 m/s (S) 11/30/02 0:00

12/1/02 0:00

10 8 m/s (S)

0 12/2/02 0:00

11/30/02 0:00

date

Chek Lap Kok: Nov 30 - Dec 1

Tap Mun: Nov 30 - Dec 1 8 m/s (N)

4 m/s (N)

4 m/s (N)

wind speed and direction

wind speed and direction

0 12/2/02 0:00

date

8 m/s (N)

0 m/s

4 m/s (S)

8 m/s (S) 11/30/02 0:00

12/1/02 0:00

0 m/s

4 m/s (S)

8 m/s (S) 12/1/02 0:00

12/2/02 0:00

11/30/02 0:00

date

Shenzhen: Nov 30 - Dec 1 8 m/s (N)

12/1/02 0:00

12/2/02 0:00

date

Guangzhou: Nov 30 - Dec 1

100

100

8 m/s (N)

90

90

70 60

0 m/s

50 40 30

4 m/s (S)

wind speed and direction

80

mm rain

wind speed and direction

80 4 m/s (N)

4 m/s (N)

70 60

0 m/s

50 40 30

4 m/s (S)

20

20

10 8 m/s (S) 11/30/02 0:00

12/1/02 0:00

date

0 12/2/02 0:00

mm rain

wind speed and direction

80 4 m/s (N)

mm rain

Fogang: Nov 30 - Dec 1 8 m/s (N)

10 8 m/s (S) 11/30/02 0:00

12/1/02 0:00

0 12/2/02 0:00

date

Figure 17. Wind Speed, Direction and Precipitation During Nov. 30 – Dec. 1, 2002 77

HKUST: Dec 6-7

100

100

8 m/s (N)

90

90

70 60

0 m/s

50 40 30

4 m/s (S)

wind speed and direction

80

mm rain

4 m/s (N)

70 60

0 m/s

50 40 30

4 m/s (S)

20

20

10 8 m/s (S) 12/6/02 0:00

12/7/02 0:00

10 8 m/s (S)

0 12/8/02 0:00

12/6/02 0:00

date

Chek Lap Kok: Dec 6-7

Tap Mun: Dec 6-7 8 m/s (N)

4 m/s (N)

4 m/s (N)

wind speed and direction

wind speed and direction

0 12/8/02 0:00

date

8 m/s (N)

0 m/s

4 m/s (S)

8 m/s (S) 12/6/02 0:00

12/7/02 0:00

0 m/s

4 m/s (S)

8 m/s (S) 12/7/02 0:00

12/8/02 0:00

12/6/02 0:00

date

Shenzhen: Dec 6-7 8 m/s (N)

12/7/02 0:00

12/8/02 0:00

date

Guangzhou: Dec 6-7

100

100

8 m/s (N)

90

90

70 60

0 m/s

50 40 30

4 m/s (S)

wind speed and direction

80

mm rain

wind speed and direction

80 4 m/s (N)

4 m/s (N)

70 60

0 m/s

50 40 30

4 m/s (S)

20

20

10 8 m/s (S) 12/6/02 0:00

12/7/02 0:00

date

0 12/8/02 0:00

mm rain

wind speed and direction

80 4 m/s (N)

mm rain

Fogang: Dec 6-7 8 m/s (N)

10 8 m/s (S) 12/6/02 0:00

12/7/02 0:00

0 12/8/02 0:00

date

Figure 18. Wind Speed, Direction and Precipitation During December 6-7, 2002 78

HKUST: Dec 12-13

100

100

8 m/s (N)

90

90

70 60

0 m/s

50 40 30

4 m/s (S)

wind speed and direction

80

mm rain

4 m/s (N)

70 60

0 m/s

50 40 30

4 m/s (S)

20

20

10 8 m/s (S) 12/12/02 0:00

12/13/02 0:00

10 8 m/s (S)

0 12/14/02 0:00

12/12/02 0:00

date

Chek Lap Kok: Dec 12-13

Tap Mun: Dec 12-13 8 m/s (N)

4 m/s (N)

4 m/s (N)

wind speed and direction

wind speed and direction

0 12/14/02 0:00

date

8 m/s (N)

0 m/s

4 m/s (S)

8 m/s (S) 12/12/02 0:00

12/13/02 0:00

0 m/s

4 m/s (S)

8 m/s (S) 12/13/02 0:00

12/14/02 0:00

12/12/02 0:00

date

Shenzhen: Dec 12-13 8 m/s (N)

12/13/02 0:00

12/14/02 0:00

date

Guangzhou: Dec 12-13

100

100

8 m/s (N)

90

90

70 60

0 m/s

50 40 30

4 m/s (S)

wind speed and direction

80

mm rain

wind speed and direction

80 4 m/s (N)

4 m/s (N)

70 60

0 m/s

50 40 30

4 m/s (S)

20

20

10 8 m/s (S) 12/12/02 0:00

12/13/02 0:00

date

0 12/14/02 0:00

mm rain

wind speed and direction

80 4 m/s (N)

mm rain

Fogang: Dec 12-13 8 m/s (N)

10 8 m/s (S) 12/12/02 0:00

12/13/02 0:00

0 12/14/02 0:00

date

Figure 19. Wind Speed, Direction and Precipitation During December 12-13, 2002 79

HKUST: Dec 18-19

100

100

8 m/s (N)

90

90

70 60

0 m/s

50 40 30

4 m/s (S)

wind speed and direction

80

mm rain

4 m/s (N)

70 60

0 m/s

50 40 30

4 m/s (S)

20

20

10 8 m/s (S) 12/18/02 0:00

12/19/02 0:00

10 8 m/s (S)

0 12/20/02 0:00

12/18/02 0:00

date

Chek Lap Kok: Dec 18-19

Tap Mun: Dec 18-19 8 m/s (N)

4 m/s (N)

4 m/s (N)

wind speed and direction

wind speed and direction

0 12/20/02 0:00

date

8 m/s (N)

0 m/s

4 m/s (S)

8 m/s (S) 12/18/02 0:00

12/19/02 0:00

0 m/s

4 m/s (S)

8 m/s (S) 12/19/02 0:00

12/20/02 0:00

12/18/02 0:00

date

Shenzhen: Dec 18-19 8 m/s (N)

12/19/02 0:00

12/20/02 0:00

date

Guangzhou: Dec 18-19

100

100

8 m/s (N)

90

90

70 60

0 m/s

50 40 30

4 m/s (S)

wind speed and direction

80

mm rain

wind speed and direction

80 4 m/s (N)

4 m/s (N)

70 60

0 m/s

50 40 30

4 m/s (S)

20

20

10 8 m/s (S) 12/18/02 0:00

12/19/02 0:00

date

0 12/20/02 0:00

mm rain

wind speed and direction

80 4 m/s (N)

mm rain

Fogang: Dec 18-19 8 m/s (N)

10 8 m/s (S) 12/18/02 0:00

12/19/02 0:00

0 12/20/02 0:00

date

Figure 20. Wind Speed, Direction and Precipitation During December 18-19, 2002 80

HKUST: Dec 24-25

100

100

8 m/s (N)

90

90

70 60

0 m/s

50 40 30

4 m/s (S)

wind speed and direction

80

mm rain

4 m/s (N)

70 60

0 m/s

50 40 30

4 m/s (S)

20

20

10 8 m/s (S) 12/24/02 0:00

12/25/02 0:00

10 8 m/s (S)

0 12/26/02 0:00

12/24/02 0:00

date

Chek Lap Kok: Dec 24-25

Tap Mun: Dec 24-25 8 m/s (N)

4 m/s (N)

4 m/s (N)

wind speed and direction

wind speed and direction

0 12/26/02 0:00

date

8 m/s (N)

0 m/s

4 m/s (S)

8 m/s (S) 12/24/02 0:00

12/25/02 0:00

0 m/s

4 m/s (S)

8 m/s (S) 12/25/02 0:00

12/26/02 0:00

12/24/02 0:00

date

Shenzhen: Dec 24-25 8 m/s (N)

12/25/02 0:00

12/26/02 0:00

date

Guangzhou: Dec 24-25

100

100

8 m/s (N)

90

90

70 60

0 m/s

50 40 30

4 m/s (S)

wind speed and direction

80

mm rain

wind speed and direction

80 4 m/s (N)

4 m/s (N)

70 60

0 m/s

50 40 30

4 m/s (S)

20

20

10 8 m/s (S) 12/24/02 0:00

12/25/02 0:00

date

0 12/26/02 0:00

mm rain

wind speed and direction

80 4 m/s (N)

mm rain

Fogang: Dec 24-25 8 m/s (N)

10 8 m/s (S) 12/24/02 0:00

12/25/02 0:00

0 12/26/02 0:00

date

Figure 21. Wind Speed, Direction and Precipitation During December 24-25, 2002 81

HKUST: Feb 28 - March 1

100

100

8 m/s (N)

90

90

70 60

0 m/s

50 40 30

4 m/s (S)

wind speed and direction

80

mm rain

4 m/s (N)

70 60

0 m/s

50 40 30

4 m/s (S)

20

20

10 8 m/s (S) 2/28/03 0:00

3/1/03 0:00

10 8 m/s (S)

0 3/2/03 0:00

2/28/03 0:00

date

Chek Lap Kok: Feb 28 - March 1

Tap Mun: Feb 28 - March 1 8 m/s (N)

4 m/s (N)

4 m/s (N)

wind speed and direction

wind speed and direction

0 3/2/03 0:00

date

8 m/s (N)

0 m/s

4 m/s (S)

8 m/s (S) 2/28/03 0:00

3/1/03 0:00

0 m/s

4 m/s (S)

8 m/s (S) 3/1/03 0:00

3/2/03 0:00

2/28/03 0:00

date

Shenzhen: Feb 28 - March 1 8 m/s (N)

3/1/03 0:00

3/2/03 0:00

date

Guangzhou: Feb 28 - March 1

100

100

8 m/s (N)

90

90

70 60

0 m/s

50 40 30

4 m/s (S)

wind speed and direction

80

mm rain

wind speed and direction

80 4 m/s (N)

4 m/s (N)

70 60

0 m/s

50 40 30

4 m/s (S)

20

20

10 8 m/s (S) 2/28/03 0:00

3/1/03 0:00

date

0 3/2/03 0:00

mm rain

wind speed and direction

80 4 m/s (N)

mm rain

Fogang: Feb 28 - March 1 8 m/s (N)

10 8 m/s (S) 2/28/03 0:00

3/1/03 0:00

0 3/2/03 0:00

date

Figure 22. Wind Speed, Direction and Precipitation During Feb. 28 – March 1, 2003 82

HKUST: March 6-7

100

100

8 m/s (N)

90

90

70 60

0 m/s

50 40 30

4 m/s (S)

wind speed and direction

80

mm rain

4 m/s (N)

70 60

0 m/s

50 40 30

4 m/s (S)

20

20

10 8 m/s (S) 3/6/03 0:00

3/7/03 0:00

10 8 m/s (S)

0 3/8/03 0:00

3/6/03 0:00

date

Chek Lap Kok: March 6-7

Tap Mun: March 6-7 8 m/s (N)

4 m/s (N)

4 m/s (N)

wind speed and direction

wind speed and direction

0 3/8/03 0:00

date

8 m/s (N)

0 m/s

4 m/s (S)

8 m/s (S) 3/6/03 0:00

3/7/03 0:00

0 m/s

4 m/s (S)

8 m/s (S) 3/7/03 0:00

3/8/03 0:00

3/6/03 0:00

date

Shenzhen: March 6-7 8 m/s (N)

3/7/03 0:00

3/8/03 0:00

date

Guangzhou: March 6-7

100

100

8 m/s (N)

90

90

70 60

0 m/s

50 40 30

4 m/s (S)

wind speed and direction

80

mm rain

wind speed and direction

80 4 m/s (N)

4 m/s (N)

70 60

0 m/s

50 40 30

4 m/s (S)

20

20

10 8 m/s (S) 3/6/03 0:00

3/7/03 0:00

date

0 3/8/03 0:00

mm rain

wind speed and direction

80 4 m/s (N)

mm rain

Fogang: March 6-7 8 m/s (N)

10 8 m/s (S) 3/6/03 0:00

3/7/03 0:00

0 3/8/03 0:00

date

Figure 23. Wind Speed, Direction and Precipitation During March 6 – March 7, 2003

83

HKUST: March 12-13

100

100

8 m/s (N)

90

90

70 60

0 m/s

50 40 30

4 m/s (S)

wind speed and direction

80

mm rain

4 m/s (N)

70 60

0 m/s

50 40 30

4 m/s (S)

20

20

10 8 m/s (S) 3/12/03 0:00

3/13/03 0:00

10 8 m/s (S)

0 3/14/03 0:00

3/12/03 0:00

date

Chek Lap Kok: March 12-13

Tap Mun: March 12-13 8 m/s (N)

4 m/s (N)

4 m/s (N)

wind speed and direction

wind speed and direction

0 3/14/03 0:00

date

8 m/s (N)

0 m/s

4 m/s (S)

8 m/s (S) 3/12/03 0:00

3/13/03 0:00

0 m/s

4 m/s (S)

8 m/s (S) 3/13/03 0:00

3/14/03 0:00

3/12/03 0:00

date

Shenzhen: March 12-13 8 m/s (N)

3/13/03 0:00

3/14/03 0:00

date

Guangzhou: March 12-13

100

100

8 m/s (N)

90

90

70 60

0 m/s

50 40 30

4 m/s (S)

wind speed and direction

80

mm rain

wind speed and direction

80 4 m/s (N)

4 m/s (N)

70 60

0 m/s

50 40 30

4 m/s (S)

20

20

10 8 m/s (S) 3/12/03 0:00

3/13/03 0:00

date

0 3/14/03 0:00

mm rain

wind speed and direction

80 4 m/s (N)

mm rain

Fogang: March 12-13 8 m/s (N)

10 8 m/s (S) 3/12/03 0:00

3/13/03 0:00

0 3/14/03 0:00

date

Figure 24. Wind Speed, Direction and Precipitation During March 12 – March 13, 2003 84

HKUST: March 18-19

100

100

8 m/s (N)

90

90

70 60

0 m/s

50 40 30

4 m/s (S)

wind speed and direction

80

mm rain

4 m/s (N)

70 60

0 m/s

50 40 30

4 m/s (S)

20

20

10 8 m/s (S) 3/18/03 0:00

3/19/03 0:00

10 8 m/s (S)

0 3/20/03 0:00

3/18/03 0:00

date

Chek Lap Kok: March 18-19

Tap Mun: March 18-19 8 m/s (N)

4 m/s (N)

4 m/s (N)

wind speed and direction

wind speed and direction

0 3/20/03 0:00

date

8 m/s (N)

0 m/s

4 m/s (S)

8 m/s (S) 3/18/03 0:00

3/19/03 0:00

0 m/s

4 m/s (S)

8 m/s (S) 3/19/03 0:00

3/20/03 0:00

3/18/03 0:00

date

Shenzhen: March 18-19 8 m/s (N)

3/19/03 0:00

3/20/03 0:00

date

Guangzhou: March 18-19

100

100

8 m/s (N)

90

90

70 60

0 m/s

50 40 30

4 m/s (S)

wind speed and direction

80

mm rain

wind speed and direction

80 4 m/s (N)

4 m/s (N)

70 60

0 m/s

50 40 30

4 m/s (S)

20

20

10 8 m/s (S) 3/18/03 0:00

3/19/03 0:00

date

0 3/20/03 0:00

mm rain

wind speed and direction

80 4 m/s (N)

mm rain

Fogang: March 18-19 8 m/s (N)

10 8 m/s (S) 3/18/03 0:00

3/19/03 0:00

0 3/20/03 0:00

date

Figure 25. Wind Speed, Direction and Precipitation During March 18 – March 19, 2003 85

HKUST: March 24-25

100

100

8 m/s (N)

90

90

70 60

0 m/s

50 40 30

4 m/s (S)

wind speed and direction

80

mm rain

4 m/s (N)

70 60

0 m/s

50 40 30

4 m/s (S)

20

20

10 8 m/s (S) 3/24/03 0:00

3/25/03 0:00

10 8 m/s (S)

0 3/26/03 0:00

3/24/03 0:00

date

Chek Lap Kok: March 24-25

Tap Mun: March 24-25 8 m/s (N)

4 m/s (N)

4 m/s (N)

wind speed and direction

wind speed and direction

0 3/26/03 0:00

date

8 m/s (N)

0 m/s

4 m/s (S)

8 m/s (S) 3/24/03 0:00

3/25/03 0:00

0 m/s

4 m/s (S)

8 m/s (S) 3/25/03 0:00

3/26/03 0:00

3/24/03 0:00

date

Shenzhen: March 24-25 8 m/s (N)

3/25/03 0:00

3/26/03 0:00

date

Guangzhou: March 24-25

100

100

8 m/s (N)

90

90

70 60

0 m/s

50 40 30

4 m/s (S)

wind speed and direction

80

mm rain

wind speed and direction

80 4 m/s (N)

4 m/s (N)

70 60

0 m/s

50 40 30

4 m/s (S)

20

20

10 8 m/s (S) 3/24/03 0:00

3/25/03 0:00

date

0 3/26/03 0:00

mm rain

wind speed and direction

80 4 m/s (N)

mm rain

Fogang: March 24-25 8 m/s (N)

10 8 m/s (S) 3/24/03 0:00

3/25/03 0:00

0 3/26/03 0:00

date

Figure 26. Wind Speed, Direction and Precipitation During March 24 – March 25, 2003 86

HKUST: June 4-5

100

100

8 m/s (N)

90

90

70 60

0 m/s

50 40 30

4 m/s (S)

wind speed and direction

80

mm rain

4 m/s (N)

70 60

0 m/s

50 40 30

4 m/s (S)

20

20

10 8 m/s (S) 6/4/03 0:00

6/5/03 0:00

10 8 m/s (S)

0 6/6/03 0:00

6/4/03 0:00

date

Chek Lap Kok: June 4-5

Tap Mun: June 4-5 8 m/s (N)

4 m/s (N)

4 m/s (N)

wind speed and direction

wind speed and direction

0 6/6/03 0:00

date

8 m/s (N)

0 m/s

4 m/s (S)

8 m/s (S) 6/4/03 0:00

6/5/03 0:00

0 m/s

4 m/s (S)

8 m/s (S) 6/5/03 0:00

6/6/03 0:00

6/4/03 0:00

date

Shenzhen: June 4-5 8 m/s (N)

6/5/03 0:00

6/6/03 0:00

date

Guangzhou: June 4-5

100

100

8 m/s (N)

90

90

70 60

0 m/s

50 40 30

4 m/s (S)

wind speed and direction

80

mm rain

wind speed and direction

80 4 m/s (N)

4 m/s (N)

70 60

0 m/s

50 40 30

4 m/s (S)

20

20

10 8 m/s (S) 6/4/03 0:00

6/5/03 0:00

date

0 6/6/03 0:00

mm rain

wind speed and direction

80 4 m/s (N)

mm rain

Fogang: June 4-5 8 m/s (N)

10 8 m/s (S) 6/4/03 0:00

6/5/03 0:00

0 6/6/03 0:00

date

Figure 27. Wind Speed, Direction and Precipitation During June 4 – June 5, 2003 87

HKUST: June 10-11

250

250

8 m/s (N)

225

225

200

150 0 m/s

125 100 75

4 m/s (S)

wind speed and direction

175

mm rain

4 m/s (N)

175 150

0 m/s

125 100 75

4 m/s (S)

50

50

25

25 8 m/s (S)

8 m/s (S) 6/10/03 0:00

6/11/03 0:00

6/10/03 0:00

0 6/12/03 0:00

Chek Lap Kok: June 10-11

Tap Mun: June 10-11

8 m/s (N)

8 m/s (N)

4 m/s (N)

4 m/s (N)

wind speed and direction

wind speed and direction

0 6/12/03 0:00

date

date

0 m/s

4 m/s (S)

8 m/s (S) 6/10/03 0:00

6/11/03 0:00

0 m/s

4 m/s (S)

8 m/s (S) 6/11/03 0:00

6/12/03 0:00

6/10/03 0:00

date

Shenzhen: June 10-11 8 m/s (N)

6/11/03 0:00

6/12/03 0:00

date

Guangzhou: June 10-11

250

250

8 m/s (N)

225

225

175 150

0 m/s

125 100 75

4 m/s (S)

wind speed and direction

200

mm rain

wind speed and direction

200 4 m/s (N)

4 m/s (N)

175 150

0 m/s

125 100 75

4 m/s (S)

50

50

25 8 m/s (S) 6/10/03 0:00

6/11/03 0:00

date

0 6/12/03 0:00

mm rain

wind speed and direction

200 4 m/s (N)

mm rain

Fogang: June 10-11 8 m/s (N)

25 8 m/s (S) 6/10/03 0:00

6/11/03 0:00

0 6/12/03 0:00

date

Figure 28. Wind Speed, Direction and Precipitation During June 10 – June 11, 2003 88

HKUST: June 16-17

100

100

8 m/s (N)

90

90

70 60

0 m/s

50 40 30

4 m/s (S)

wind speed and direction

80

mm rain

4 m/s (N)

70 60

0 m/s

50 40 30

4 m/s (S)

20

20

10 8 m/s (S) 6/16/03 0:00

6/17/03 0:00

10 8 m/s (S)

0 6/18/03 0:00

6/16/03 0:00

date

Chek Lap Kok: June 16-17

Tap Mun: June 16-17 8 m/s (N)

4 m/s (N)

4 m/s (N)

wind speed and direction

wind speed and direction

0 6/18/03 0:00

date

8 m/s (N)

0 m/s

4 m/s (S)

8 m/s (S) 6/16/03 0:00

6/17/03 0:00

0 m/s

4 m/s (S)

8 m/s (S) 6/17/03 0:00

6/18/03 0:00

6/16/03 0:00

date

Shenzhen: June 16-17 8 m/s (N)

6/17/03 0:00

6/18/03 0:00

date

Guangzhou: June 16-17

100

100

8 m/s (N)

90

90

70 60

0 m/s

50 40 30

4 m/s (S)

wind speed and direction

80

mm rain

wind speed and direction

80 4 m/s (N)

4 m/s (N)

70 60

0 m/s

50 40 30

4 m/s (S)

20

20

10 8 m/s (S) 6/16/03 0:00

6/17/03 0:00

date

0 6/18/03 0:00

mm rain

wind speed and direction

80 4 m/s (N)

mm rain

Fogang: June 16-17 8 m/s (N)

10 8 m/s (S) 6/16/03 0:00

6/17/03 0:00

0 6/18/03 0:00

date

Figure 29. Wind Speed, Direction and Precipitation During June 16 – June 17, 2003 89

HKUST: June 22-23

100

100

8 m/s (N)

90

90

70 60

0 m/s

50 40 30

4 m/s (S)

wind speed and direction

80

mm rain

4 m/s (N)

70 60

0 m/s

50 40 30

4 m/s (S)

20

20

10 8 m/s (S) 6/22/03 0:00

6/23/03 0:00

10 8 m/s (S)

0 6/24/03 0:00

6/22/03 0:00

date

Chek Lap Kok: June 22-23

Tap Mun: June 22-23 8 m/s (N)

4 m/s (N)

4 m/s (N)

wind speed and direction

wind speed and direction

0 6/24/03 0:00

date

8 m/s (N)

0 m/s

4 m/s (S)

8 m/s (S) 6/22/03 0:00

6/23/03 0:00

0 m/s

4 m/s (S)

8 m/s (S) 6/23/03 0:00

6/24/03 0:00

6/22/03 0:00

date

Shenzhen: June 22-23 8 m/s (N)

6/23/03 0:00

6/24/03 0:00

date

Guangzhou: June 22-23

100

100

8 m/s (N)

90

90

70 60

0 m/s

50 40 30

4 m/s (S)

wind speed and direction

80

mm rain

wind speed and direction

80 4 m/s (N)

4 m/s (N)

70 60

0 m/s

50 40 30

4 m/s (S)

20

20

10 8 m/s (S) 6/22/03 0:00

6/23/03 0:00

date

0 6/24/03 0:00

mm rain

wind speed and direction

80 4 m/s (N)

mm rain

Fogang: June 22-23 8 m/s (N)

10 8 m/s (S) 6/22/03 0:00

6/23/03 0:00

0 6/24/03 0:00

date

Figure 30. Wind Speed, Direction and Precipitation During June 22 – June 23, 2003 90

HKUST: June 28-29

100

100

8 m/s (N)

90

90

70 60

0 m/s

50 40 30

4 m/s (S)

wind speed and direction

80

mm rain

4 m/s (N)

70 60

0 m/s

50 40 30

4 m/s (S)

20

20

10 8 m/s (S) 6/28/03 0:00

6/29/03 0:00

10 8 m/s (S)

0 6/30/03 0:00

6/28/03 0:00

date

Chek Lap Kok: June 28-29

Tap Mun: June 28-29 8 m/s (N)

4 m/s (N)

4 m/s (N)

wind speed and direction

wind speed and direction

0 6/30/03 0:00

date

8 m/s (N)

0 m/s

4 m/s (S)

8 m/s (S) 6/28/03 0:00

6/29/03 0:00

0 m/s

4 m/s (S)

8 m/s (S) 6/29/03 0:00

6/30/03 0:00

6/28/03 0:00

date

Shenzhen: June 28-29 8 m/s (N)

6/29/03 0:00

6/30/03 0:00

date

Guangzhou: June 28-29

100

100

8 m/s (N)

90

90

70 60

0 m/s

50 40 30

4 m/s (S)

wind speed and direction

80

mm rain

wind speed and direction

80 4 m/s (N)

4 m/s (N)

70 60

0 m/s

50 40 30

4 m/s (S)

20

20

10 8 m/s (S) 6/28/03 0:00

6/29/03 0:00

date

0 6/30/03 0:00

mm rain

wind speed and direction

80 4 m/s (N)

mm rain

Fogang: June 28-29 8 m/s (N)

10 8 m/s (S) 6/28/03 0:00

6/29/03 0:00

0 6/30/03 0:00

date

Figure 31. Wind Speed, Direction and Precipitation During June 28 – June 29, 2003 91

Figure 32. A matrix of least-squares linear fit of PM2.5 measurements paired by site. Note that the axes for each pair of sites have been scaled to best show the data points.

92

Figure 33. A matrix of least-squares linear fit of organic carbon measurements paired by site. Note that the axes for each pair of sites have been scaled to best show the data points.

93

Figure 34. A matrix of least-squares linear fit of sulfate measurements paired by site. Note that the axes for each pair of sites have been scaled to best show the data points.

94

Figure 35. A matrix of least-squares linear fit of silica measurements paired by site. Note that the axes for each pair of sites have been scaled to best show the data points.

95

Figure 36. A matrix of least-squares linear fit of iron measurements paired by site. Note that the axes for each pair of sites have been scaled to best show the data points.

96

Figure 37 A matrix of least-squares linear fit of elemental carbon measurements paired by site. Note that the axes for each pair of sites have been scaled to best show the data points.

97

Figure 38 A matrix of least-squares linear fit of lead measurements paired by site. Note that the axes for each pair of sites have been scaled to best show the data points.

98

Figure 39 A matrix of least-squares linear fit of rubidium measurements paired by site. Note that the axes for each pair of sites have been scaled to best show the data points.

99

Figure 40 A matrix of least-squares linear fit of potassium measurements paired by site. Note that the axes for each pair of sites have been scaled to best show the data points.

100

1e-5 Tritriacontane

Indeno(cd)pyrene

Indeno(cd)fluoranthene

Picene

Benzo(e)pyrene

Benzo(b)fluoranthene

Benzo(k)fluoranthene

levoglucosan

Isopimaric acid

Pimaric acid

8,15-pimaredienoic acid

Nonanal

17α,21β-hopane

17α,21β-29-norhopane

22,29,30-trisnorneohopane

20S&R-αββ-sitostanes

20S&R-αββ-ergostanes

20R-ααα-cholestane

20S&R-αββ−cholestane

Tetratriacontane

1e+0

Dotriacontane

1e+0

Hentriacontane

g/g Organic

g/g Organic 1e+0

Triacontane

Nonacosane

Octacosane

Heptacosane

Hexacosane

Pentacosane

Silicon

Aluminum

Elemental carbon

g/g Organic 1e+1

Diesel Exhaust

1e-1

1e-2

1e-3

1e-4

1e-5

1e+1

1e-1

Gasoline Exhaust

1e-2

1e-3

1e-4

1e-5

1e+1

Vegitative Detritus

1e-1

1e-2

1e-3

1e-4

Figure 41a. Source Profiles for diesel exhaust, gasoline exhaust, and vegetative detritus (Rogge et al., 1993; Schauer et al., 1999; Schauer et al., 2002)

101

1e-5 Indeno(cd)pyrene

Indeno(cd)fluoranthene

Picene

Benzo(e)pyrene

Benzo(b)fluoranthene

Benzo(k)fluoranthene

levoglucosan

Isopimaric acid

Pimaric acid

8,15-pimaredienoic acid

Nonanal

17α,21β-hopane

17α,21β-29-norhopane

22,29,30-trisnorneohopane

20S&R-αββ-sitostanes

20R-ααα-cholestane 20S&R-αββ-ergostanes

20S&R-αββ−cholestane

1e+0

Tritriacontane

1e+0

Tetratriacontane

g/g OC

g/g OC 1e+0

Dotriacontane

Hentriacontane

Triacontane

Nonacosane

Octacosane

Heptacosane

Hexacosane

Pentacosane

Silicon

Aluminum

Elemental carbon

g/g OC 1e+1

Meat Cooking

1e-1

1e-2

1e-3

1e-4

1e-5

1e+1

Road Dust

1e-1

1e-2

1e-3

1e-4

1e-5

1e+1

Cigarette Smoke

1e-1

1e-2

1e-3

1e-4

Figure 41b. Source profiles for meat cooking, road dust, and cigarette smoke (Rogge et al., 1993; Schauer et al., 1998; Schauer et al., 1999)

102

1e-5 Indeno(cd)pyrene

Indeno(cd)fluoranthene

Picene

Benzo(e)pyrene

Benzo(b)fluoranthene

Benzo(k)fluoranthene

levoglucosan

Isopimaric acid

Pimaric acid

8,15-pimaredienoic acid

Nonanal

17α,21β-hopane

17α,21β-29-norhopane

22,29,30-trisnorneohopane

20S&R-αββ-sitostanes

20R-ααα-cholestane 20S&R-αββ-ergostanes

20S&R-αββ−cholestane

Tetratriacontane

Tritriacontane

1e+0

Dotriacontane

Hentriacontane

Triacontane

Nonacosane

Octacosane

Heptacosane

Hexacosane

Pentacosane

Silicon

Aluminum

Elemental carbon

g/g OC

g/g OC 1e+1

1e+0

Wood Combustion

1e-1

1e-2

1e-3

1e-4

1e-5

1e+1

Coal Combustion

1e-1

1e-2

1e-3

1e-4

Figure 41c. Source profiles for wood combustion and coal combustion (Zheng et al., 2002; Zheng et al., in preparation)

103

30 EC OC

3

Concentrations (µ g/m )

25

20

15

10

5

GZ-12

GZ-10

GZ-06

GZ-03

SZ-12

SZ-10

SZ-06

SZ-03

ZS-12

ZS-10

ZS-06

ZS-03

CH-12

CH-10

CH-06

CH-03

CW-12

CW-10

CW-06

CW-03

TC-12

TC-10

TC-06

TC-03

TM-12

TM-10

TM-06

TM-03

0

Samples

Figure 42. Seasonal and spatial distribution of elemental (EC) and organic carbon (OC) (TM: Tap Mun, TC: Tung Chung, CW: Central and Western, CH: Conghua, ZS: Zhongshan, SZ: Shenzhen, GZ: Guangzhou)

104

0.45 0.40 0.35

EC/OC

0.30 0.25 0.20 0.15 0.10 0.05

GZ-12

GZ-10

GZ-06

GZ-03

SZ-12

SZ-10

SZ-06

SZ-03

ZS-12

ZS-10

ZS-06

ZS-03

CH-12

CH-10

CH-06

CH-03

CW-12

CW-10

CW-06

CW-03

TC-12

TC-10

TC-06

TC-03

TM-12

TM-10

TM-06

TM-03

0.00

Samples

Figure 43. Seasonal and spatial distribution of the EC to OC ratios (TM: Tap Mun, TC: Tung Chung, CW: Central and Western, CH: Conghua, ZS: Zhongshan, SZ: Shenzhen, GZ: Guangzhou)

105

10

EC/OC Ratio

1

0.1

RDKWRE

RDBAKE

VEGETA

CATCAR

NONCAT

PINEWO

EUCALY

OAKWOO

BJCOAL

FUELOI

0.001

DIESEL

0.01

Sources

Figure 44. Ratios of elemental carbon (EC) to organic carbon (OC) from different source tests. DIESEL, diesel trucks; FUELOI, fuel oil combustion; BJCOAL, Beijing coal combustion; OAKWOO, oak wood combustion; EUCLAY, eucalyptus wood combustion; PINEWO, pine wood combustion; NONCAT, non-catalyst gasoline vehicles; CATCAR., catalyst gasoline vehicles; VEGETA, vegetative detritus; RDBAKE, Bakersfield road dust; RDKWRE, Kern Wildlife Refuge road dust

106

25 Total Hopanes Total Steranes

3

Concentrations (ng/m )

20

15

10

5

GZ-12

GZ-10

GZ-06

GZ-03

SZ-12

SZ-10

SZ-06

SZ-03

ZS-12

ZS-10

ZS-06

ZS-03

CH-12

CH-10

CH-06

CH-03

CW-12

CW-10

CW-06

CW-03

TC-12

TC-10

TC-06

TC-03

TM-12

TM-10

TM-06

TM-03

0

Samples

Figure 45. Seasonal and spatial distribution of hopanes and steranes (TM: Tap Mun, TC: Tung Chung, CW: Central and Western, CH: Conghua, ZS: Zhongshan, SZ: Shenzhen, GZ: Guangzhou)

107

1.6E-03 1.4E-03 1.2E-03

(H+S)/OC

1.0E-03 8.0E-04 6.0E-04 4.0E-04 2.0E-04

GZ-12

GZ-10

GZ-06

GZ-03

SZ-12

SZ-10

SZ-06

SZ-03

ZS-12

ZS-10

ZS-06

ZS-03

CH-12

CH-10

CH-06

CH-03

CW-12

CW-10

CW-06

CW-03

TC-12

TC-10

TC-06

TC-03

TM-12

TM-10

TM-06

TM-03

0.0E+00

Samples

Figure 46a. Seasonal and spatial distribution of the hopanes and steranes (H+S) to OC ratios (TM: Tap Mun, TC: Tung Chung, CW: Central and Western, CH: Conghua, ZS: Zhongshan, SZ: Shenzhen, GZ: Guangzhou)

108

1.6e-1

(H+S)/EC Ratio

1.5e-1

1.4e-1 1.0e-3

8.0e-4

6.0e-4

4.0e-4

2.0e-4

0.0 DIESEL

C*CARS

FUELOI

BJCOAL

Sources

Figure 46b. Ratios of hopanes (H) and steranes (S) to elemental carbon (EC) from different source tests. DIESEL, diesel trucks; C*CARS, gasoline vehicles; FUELOI, fuel oil combustion; BJCOAL, Beijing coal combustion.

109

8.0E-03 7.0E-03 6.0E-03

(H+S)/EC

5.0E-03 4.0E-03 3.0E-03 2.0E-03 1.0E-03

GZ-12

GZ-10

GZ-06

GZ-03

SZ-12

SZ-10

SZ-06

SZ-03

ZS-12

ZS-10

ZS-06

ZS-03

CH-12

CH-10

CH-06

CH-03

CW-12

CW-10

CW-06

CW-03

TC-12

TC-10

TC-06

TC-03

TM-12

TM-10

TM-06

TM-03

0.0E+00

Samples

Figure 46c. Seasonal and spatial distribution of the hopanes and steranes (H+S) to EC ratios (TM: Tap Mun, TC: Tung Chung, CW: Central and Western, CH: Conghua, ZS: Zhongshan, SZ: Shenzhen, GZ: Guangzhou)

110

900 800

3

Concentration (ng/m )

700 600 500 400 300 200 100

GZ-12

GZ-10

GZ-06

GZ-03

SZ-12

SZ-10

SZ-06

SZ-03

ZS-12

ZS-10

ZS-06

ZS-03

CH-12

CH-10

CH-06

CH-03

CW-12

CW-10

CW-06

CW-03

TC-12

TC-10

TC-06

TC-03

TM-12

TM-10

TM-06

TM-03

0

Samples

Figure 47. Seasonal and spatial distribution of levoglucosan (TM: Tap Mun, TC: Tung Chung, CW: Central and Western, CH: Conghua, ZS: Zhongshan, SZ: Shenzhen, GZ: Guangzhou)

111

PAH with MW202

Methyl substituted MW 202 PAH

PAH with MW226 and 228

Methyl Substituted PAH with MW 226 and 228

PAH with MW252

Retene(MW234)

PAH with MW276

Picene (MW278)

Coronene (MW300)

100 90

3

Concentration (ng/m )

80 70 60 50 40 30 20 10

GZ5-12

GZ5-10

GZ5-06

GZ5-03

SZ4-12

SZ4-10

SZ4-06

SZ4-03

ZS7-12

ZS7-10

ZS7-06

ZS7-03

CH6-12

CH6-10

CH6-06

CH6-03

CW3-12

CW3-10

CW3-06

CW3-03

TC2-12

TC2-10

TC2-06

TC2-03

TM1-12

TM1-10

TM1-06

TM1-03

0

Samples

Figure 48. Spatial and seasonal distribution of PAHs (TM: Tap Mun, TC: Tung Chung, CW: Central and Western, CH: Conghua, ZS: Zhongshan, SZ: Shenzhen, GZ: Guangzhou)

112

PAH with MW202 Methyl Substituted PAH with MW 226 and 228 PAH with MW276

Methyl substituted MW 202 PAH PAH with MW252 Picene (MW278)

PAH with MW226 and 228 Retene(MW234) Coronene (MW300)

100% 90% 80%

% Total PAHs

70% 60% 50% 40% 30% 20% 10%

GZ-12

GZ-10

GZ-06

GZ-03

ZS-12

ZS-10

ZS-06

ZS-03

SZ-12

SZ-10

SZ-06

SZ-03

CH-12

CH-10

CH-06

CH-03

CW-12

CW-10

CW-06

CW-03

TC-12

TC-10

TC-06

TC-03

TM-12

TM-10

TM-06

TM-03

0%

Samples

Figure 49. Seasonal and spatial distribution of percentage of PAH to Total PAHs (TM: Tap Mun, TC: Tung Chung, CW: Central and Western, CH: Conghua, ZS: Zhongshan, SZ: Shenzhen, GZ: Guangzhou)

113

Diesel Exhaust

Gasoline Exhaust

Meat Cooking

Cigarette Smoke

Road Dust

Vegetative Detritus

Coal Combustion

Other OC

Biomass burning

25

20

3

Concentrations (µg/m )

GZ

15

ZS

SZ

CH 10

TC

CW

TM 5

Winter

Fall

Summer

Spring

Winter

Fall

Summer

Spring

Winter

Fall

Summer

Spring

Winter

Fall

Summer

Spring

Winter

Fall

Summer

Spring

Winter

Fall

Summer

Spring

Winter

Fall

Summer

Spring

0

Seasons

Figure 50. Source contributions to fine particulate organic carbon in PM2.5. The error bars show the error propagated from the error associated with each source identified from the CMB model.

114

60 % WSOC/OC % OTHER OC

50

40

30

20

10

0 TM

TC

CW

CH

ZS

SZ

GZ

Figure 51. Comparison of measured WSOC/OC and CMB estimated % Other/ OC concentrations

115

Appendix A

Hong Kong and PRD Air Quality Study: Aerosol Project

I. Fine particulate matter (PM2.5) site selection for the PRD study II. Sampling Methodology III. Additional supporting measurements during the PRD study IV. PM2.5 sampling schedule and timetable

Report Prepared by:

Michael Bergin1,2 and William Chameides2 Georgia Institute of Technology School of Civil and Environmental Engineering1 School of Earth and Atmospheric Sciences2

July 11, 2002

I. Fine particulate matter (PM2.5) site selection for the PRD study

Introduction

The major policy-relevant, scientific objectives of the PRD PM2.5 study are to assess: 1. The relative contributions of various types of sources of pollution to the PM2.5 mass and chemical composition in the Pearl River Delta; 2. The relative contributions of sources from within and outside of the Pearl River Delta to PM2.5 in the Pearl River Delta; and 3. The relative contributions of pollution sources from various subregions of the Delta (i.e., Guangdong, Hong Kong) to PM2.5 in the Pearl River Delta .

In order to achieve these objectives, the Science Team will undertake an observation-based approach involving the detailed analysis of PM2.5 mass concentration and chemical composition (including major ions, organic and elemental carbon, and selected water soluble and solvent extractable organic tracer compounds) measured at 7 sites in the PRD region. Four of the sites are to be located in Guangdong province and three are to be located in Hong Kong. In this section we discuss the criteria adopted by the Science Team for selecting these seven sites and then the locations and characteristics of the seven sites chosen.

Site selection criteria

Five major criteria were used to select the locations of the seven sampling sites: 1. Overarching Logistical Criteria: Each site must have the infrastructure including power capability and security, as well as skilled personnel to conduct sampling; 2. Background Sites: At least one sampling site in Hong Kong and one sampling site in Guangdong should be located in a relatively remote outlying area so that the mass concentration and chemical composition of the so-called “background” PM2.5 in Hong Kong and Guangdong can be measured. Note that we use the term background to denote the portion of PM2.5 found in either Hong Kong or Guangdong that is transported into these regions from outside of the Pearl River Delta region and/or produced locally by natural processes and thus can not be affected by local pollution control measures. In general by combining these background measurements with meteorological information and the data we 2

gather from other sites, it should be possible to diagnose the fractions of this background that is natural and the fraction that is coming from outside of the Pearl River delta. In principle some fraction of the PM2.5 within the region can arise from pollution emanating from other regions in China and/or East Asia (or beyond) and so could be reduced by pollution control measures adopted outside of the Pearl River Delta region. 3. Receptor Sites: At least one site in both Hong Kong and in Guangdong should be located in areas where local pollution sources are negligible, but are downwind of major sources of pollution within the Pearl River Delta, so that the impact of these sources on regional PM2.5 can be assessed. 4. Urban Source Sites: The remaining sites need to be located within those areas in the Pearl River Delta where the major sources of PM2.5 pollution are found. The data from these sites will be used to assess: (i) the additional PM2.5 mass and chemical constituents added to the background PM2.5 from the individual source areas; and (ii) the so-called “chemical fingerprint” of the PM2.5 emanating from the individual source areas.

Sampling site Selection

In applying the above criteria to the selection of sampling sites, the Science Team had to be cognizant of the monetary and human resources available for the project which ultimately limited the number of sites to be operated to four in Guangdong and three in Hong Kong. (Resource limitations also placed constraints on the amount of sampling to be carried out at each site). To meet the overall objectives of the project with these constraints, the Science Team carefully prioritized sampling needs identified by the criteria listed above. After much discussion the Science Team elected to choose the sites as follows: ¾ Two Background Sites called for in Criteria #2; ¾ Two Receptor Sites called for in Criteria #3; and ¾ Three Urban Source Sites described in Criteria #4. The Science Team further prioritized the selection of these three Urban Source Sites to be located in the three areas believed to represent the major sources of urban air pollution to the region: namely (in alphabetical order) Guangzhou, urban Hong Kong, and Shenzhen, the three largest cities in the PRD.

With these criteria and corresponding prioritizations identified the Science Team then made the specific site selections. These selections were made on the basis of the analysis of available 3

PM2.5 data from several years of measurements by the Hong Kong EPD as well as a pilot study conducted in the PRD by CH2M Hill. A summary of the findings of this analysis were presented at the science team meeting (the presentation is available through Civic Exchange). Locations of Sampling Sites A map indicating the locations of the seven sampling sites selected for the PRD PM2.5 study is illustrated in Figure 1. Table 1 provides pertinent information on each site; i.e., location, organization responsible for operation of the site, and available PM concentration data at the sites.

Conghua

Tap Mun Central and Western Tung Chung

Figure 1. Locations of sampling sites selected for the PRD PM2.5 Study.

The 2 Background Sites are Tap Mun, which is located on a small, relatively unpopulated island in north-east Hong Kong, and Conghua which is situated ~ 80 km north east of Guangzhou in 4

rural Guangdong. Tap Mun is a routine monitoring station currently being operated by the Hong Kong EPD while the Conghua site is maintained by the local Conghua EPB. As shown in Table 1, both sites are known to have relatively low particulate matter concentrations. Table 1. Site Descriptions Site Location

Organization

PM Concentration

Site Type

Responsible For Operation Conghua

Conghua EPB

1

RSP = 38 µg m-3

Background

Tap Mun

Hong Kong EPD

2

PM2.5 = 29 µg m-3

Background

Shenzhen

Peking University

1

RSP = 114 µg m-3

Urban Source

Guangzhou

Guangzhou

1

RSP = 158 µg m-3

Urban Source

Hong Kong EPD

2

RSP = 50 µg m-3

Urban Source

Tung Chung

Hong Kong EPD

2

PM2.5 = 33 µg m-3

Receptor/Back.

Zongshan

Zhongshan EPB

3

RSP ~ 60 µg m-3

Receptor/Back.

Municipal EPD Central and Western

RSP is particles having diameters less than 10 µm 1 CH2M Hill Report, 2002; measurements made during Dec. 1999 – May 2000 2 Based on measurements made by HKEPD during ~1998 - 2002 3 Based on observed measurements on 5/29/02

As noted above, the three Urban Source Sites are located in Shenzhen, Guangzhou, and Hong Kong. In Hong Kong we plan to use the EPD Central and Western (CW) monitoring site located in a populated area of Hong Kong Island. The sampling site in Shenzhen will be situated at the Peking University Shenzhen Graduate School. The Guangzhou sampling station is located in the city at the Guangzhou Municipal EPD. As shown in Table 1, the particulate matter concentrations at the two urban sites in Guangdong are relatively high compared to the background sites; the urban site in Hong Kong also has elevated particulate matter concentrations, although not nearly as high as the two urban sites in Guangdong.

The two receptor sites will be located near the city of Zhongshan in Guangdong and at Tung Chung on Lantau Island in Hong Kong. The Zhongshan sampling station is maintained and 5

operated by the Zhongshan EPB and is very likely influenced by urban regions in both Hong Kong as well as Guangdong depending on the time of year and air flow patterns, and thus should work well as a Receptor Site (it is also possible that under some meteorological conditions, the Zhongshan site will provide data more representative of a Background Site). The Tung Chung site, under the control of the Hong Kong EPD, is likely most useful as a Receptor Site during the winter, when PM2.5 pollution is generally the worst, due to advection from both Hong Kong and Guangdong. During the summer the site generally experiences relatively low PM2.5 concentrations and likely acts as Background Site (M.H. Bergin, Science Team Meeting Presentation).

II. Sampling Methodology

At 3 sites in Hong Kong and 2 sites in Guangdong ThermoAndersen RAAS PM2.5 chemical speciation samplers (http://www.anderseninstruments.com/raas-400.htm) will be used to collect aerosol samples. The samplers are designed with 4 separate channels. Two of the channels will be operated at 16.7 lpm with quartz filters that will be analyzed for organic (OC) and elemental carbon (EC), and specific water soluble and solvent extractable organic compounds that will serve as tracers for various aerosol sources. The other two sampling lines will collect aerosol particles on Teflon filters at a flowrate of 7.3 lpm that will be used to determine the PM2.5 mass concentration as well as the concentrations of major ions, and elements. At two additional sites in Guangdong 2 prototype samplers designed and fabricated at Caltech will operate under similar conditions as described above.

The sampling will be conducted during one month of each season. The months of sampling will be October and December of 2002, and March and June of 2003 as shown in the project timeline in the Appendix. Sampling will be conducted every sixth day during each sampling month. The filter sample times will be 24-hr.

III. Additional supporting measurements during the PRD study Particulate Measurements To the extent possible additional measurements will used to add to our understanding of the spatial and temporal variability of PM concentrations in the study area. Within Hong Kong we 6

will have access to measurements being conducted by the Hong Kong EPD. Measurements of RSP (particulate matter having diameters less than 10 µm) will be available for ~14 locations, and PM2.5 at five locations. Of the five HKEPD PM2.5 measurements sites, three will be colocated with the ThermoAndersen samplers at Tap Mun, Tung Chung, and Central and Western. Both RSP and PM2.5 are being monitored in real-time at the HKEPD sites using Tapered Element Oscillating Microbalance’s (TEOM). At the sampling stations the TEOM data will fill in the gaps for PM2.5 concentrations when filter measurements are not being made. In addition, the TEOM data will be used to gain insights into the processes that influence the temporal and spatial variability of PM2.5 concentrations in Hong Kong. There are currently several stations that continuously measure RSP in the PRD, and it is anticipated that some of these measurements will be available during the study. Meteorological Measurements Meteorological measurements will be integrated with the PM2.5 data to better understand the influence of transport and meteorological conditions on aerosol concentrations. It is anticipated that basic meteorological measurements (temperature, pressure, relative humidity, wind speed and wind direction) will be available at all sampling locations. In addition, we will use measurements being made by the Hong Kong observatory at several locations around Hong Kong. It is also very likely that we will have meteorological data from each of the sampling locations in Guangdong as well as additional met data from the region

IV. PM2.5 sampling schedule and timetable

A timeline for tasks relevant to the PRD aerosol project is given in the Appendix. At this time the aerosol samplers, additional field equipment, and filters have been purchased. The ThermoAndersen RAAS PM2.5 chemical speciation samplers are currently being tested and prepared for field deployment at Georgia Tech. We are also in the process of developing protocols for field sampling. During the testing of the samplers, standard operations procedures will be developed for the field sampling. Four of the samplers will be shipped to HKUST for deployment in Hong Kong, and two will be shipped to Peking University for deployment in Guangdong province. It is anticipated that the samplers will arrive at both locations in early August. The field samplers will be deployed in Hong Kong and Guangdong during the last part

7

of August and beginning of September. The sampler deployments will be conducted jointly by members of the Georgia Tech research team, Peking University, HKUST, HKEPD, and sponsoring organizations in Guangdong. At each sampling location operators will be trained on standard field operations procedures by members of the Georgia Tech and Peking University research teams. Roughly 2-weeks prior to the first month of sampling in October a coordinated test run will be conducted for a 24-hr sampling period. This will be done in order to insure the samplers are operating properly and that the field technicians are adequately trained. Samples will be refrigerated and shipped directly to HKUST after each month of sampling. HKUST will remove samples being analyzed at that location, and ship the remaining samples to Georgia Tech. It is anticipated that a science team meeting will be held during December 2003. This date has tentatively been chosen since it occurs at a time when most of the chemical analyses will be completed and can be discussed during the meeting. A midterm progress report will be prepared during May-June of 2003, as well as a final project report from January-April 2004.

8

APPENDIX PRD Aerosol Project Timeline, Year 1 2002 Task Select sampling sites

May ____

June

July

Purchase Equipment

____

____

____

Test field sampling equipment and prepare protocols

____

Prepare field samples

____

Deploy field samplers and train operators

2003

Aug

Sept

Oct

____

____

____

____

Test run of samplers in field Aerosol Sampling at field sites

Nov

Dec

Jan

____

___

____

****

****

Feb

Mar

April

___

____

****

Laboratory analysis of field samples

___

____

****

____

___

____

____

PRD Aerosol Project Timeline: Year 2 2003

2004

Task Aerosol Sampling at field sites

May

June ****

July

Aug

Sept

Oct

Nov

Dec

Laboratory analysis of field samples

____

____

____

____

____

_____

____

____

Science team meeting Midterm progress report preparation Final report preparation

Jan

Feb

Mar

April

____

____

____

____

____ ____

____

9

Appendix B Prepared for the Hong Kong Civic Exchange

Standard Operating Procedures for Sample Preparation, Handling, and Analysis during the Pilot Study in Hong Kong and the Pearl River Delta

Lynn G. Salmon, Gayle S. Willis, Mei Zheng, and Michael H. Bergin

Environmental Quality Laboratory California Institute of Technology School of Civil and Environmental Engineering and Earth and Atmospheric Sciences Georgia Institute of Technology

June 24, 2004

Table of Contents

Table of Contents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-2 List of Figures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-7 List of Log Sheets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-8 1

INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-9

2

PREPARATION OF QUARTZ FIBER FILTERS . . . . . . . . . . . . . . . . B-11

3

2.1

Apparatus and Materials . . . . . . . . . . . . . . . . . . . . . . . . . . . B-11

2.2

General Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-12

2.3

Procedure for Baking Quartz Filters . . . . . . . . . . . . . . . . . . . . . B-12

2.4

Petri Dishes for Baked 47mm Quartz Filters . . . . . . . . . . . . . . . . . B-13

PROCEDURE FOR WEIGHING TEFLON FILTERS . . . . . . . . . . . . . B-15 3.1

Apparatus and Materials . . . . . . . . . . . . . . . . . . . . . . . . . . . B-15

3.2

Balance Facilities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-15

3.3

Environmental Requirements for Filter Weighing . . . . . . . . . . . . . . B-16

3.4

Electrostatic Charge Neutralization . . . . . . . . . . . . . . . . . . . . . . B-16

3.5

General Procedure for Weighing Teflon Filters . . . . . . . . . . . . . . . . B-16

3.6

3.5.1

Preparation for weighing teflon filters . . . . . . . . . . . . . . . . B-18

3.5.2

Precise weighing procedure . . . . . . . . . . . . . . . . . . . . . B-19

3.5.3

Precautions for weighing teflon filters . . . . . . . . . . . . . . . . B-20

Quality Control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-21 3.6.1

Teflon filter weighing lab blanks . . . . . . . . . . . . . . . . . . . B-21

3.6.2

Teflon filter field blanks . . . . . . . . . . . . . . . . . . . . . . . B-21

3.6.3

Mass error determination . . . . . . . . . . . . . . . . . . . . . . . B-21

3.6.4

General precautions . . . . . . . . . . . . . . . . . . . . . . . . . B-22

4 SAMPLE LOADING/UNLOADING IN THE FIELD . . . . . . . . . . . . . B-24 B-2

4.1

Filter samples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-24

4.2

Loading Filter Samples . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-28 4.2.1

4.3

4.4

4.5

Field blank . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-29

Andersen Samplers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-29 4.3.1

Filter loading for Andersen samplers . . . . . . . . . . . . . . . . . B-29

4.3.2

Leak check procedure . . . . . . . . . . . . . . . . . . . . . . . . B-33

4.3.3

Initial flow rate check . . . . . . . . . . . . . . . . . . . . . . . . . B-34

4.3.4

Programming an event to run . . . . . . . . . . . . . . . . . . . . . B-36

4.3.5

Middle flow check . . . . . . . . . . . . . . . . . . . . . . . . . . B-36

4.3.6

Unloading filter samples from Andersen samplers . . . . . . . . . . B-37

4.3.7

Steps for filter unloading . . . . . . . . . . . . . . . . . . . . . . . B-38

4.3.8

Downloading data with DataLink . . . . . . . . . . . . . . . . . . B-39

Caltech Gray Box Samplers . . . . . . . . . . . . . . . . . . . . . . . . . . B-39 4.4.1

Filter loading for Caltech samplers . . . . . . . . . . . . . . . . . . B-39

4.4.2

Flow measurement and programming a sampling event: . . . . . . B-40

4.4.3

Unloading filter samples from Caltech samplers . . . . . . . . . . . B-41

4.4.4

Steps for filter unloading . . . . . . . . . . . . . . . . . . . . . . . B-43

Trouble shooting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-43 4.5.1

Spare parts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-43

4.5.2

Problems that may occur during sample loading/initial check . . . . B-44

4.5.3

Problems that may occur during sample middle check . . . . . . . B-46

4.5.4

Problems that may occur during sample unloading . . . . . . . . . B-47

4.5.5

Reset Calibration (Andersen sampler) . . . . . . . . . . . . . . . . B-48

4.5.6

Make-up sampling event . . . . . . . . . . . . . . . . . . . . . . . B-49

5 CONTACT HELP LIST . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-49 5.1

For Field Operation or Sample Questions Contact: . . . . . . . . . . . . . . B-49

5.2

Monitoring Stations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-50 B-3

5.3

Principal Investigators: . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-50

6

SAMPLE TRANSPORTATION AND SHIPPING . . . . . . . . . . . . . . . B-51

7

SAMPLING SCHEDULE . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-52 7.1

Sample Dates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-52

7.2

Calibrating Samplers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-52

7.3 8

9

7.2.1

Calibration/verification of Andersen samplers . . . . . . . . . . . . B-54

7.2.2

Calibrating Caltech samplers . . . . . . . . . . . . . . . . . . . . . B-56

Sampler Maintenance . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-57

ORGANIC AND ELEMENTAL CARBON ANALYSIS . . . . . . . . . . . . B-57 8.1

Apparatus and Materials . . . . . . . . . . . . . . . . . . . . . . . . . . . B-57

8.2

Analysis Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-58

8.3

Daily Run . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-58

8.4

Quality Control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-60 8.4.1

Instrument specifications . . . . . . . . . . . . . . . . . . . . . . . B-60

8.4.2

Quality control standards . . . . . . . . . . . . . . . . . . . . . . . B-61

8.4.3

Freezer storage on samples . . . . . . . . . . . . . . . . . . . . . . B-61

GC/MS ANALYSIS OF QUARTZ FILTERS . . . . . . . . . . . . . . . . . . B-64 9.1

Apparatus and Materials . . . . . . . . . . . . . . . . . . . . . . . . . . . B-64

9.2

Analysis Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-65

9.3

9.2.1

Baking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-65

9.2.2

Extraction and concentration . . . . . . . . . . . . . . . . . . . . . B-66

9.2.3

Derivatization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-67

9.2.4

Quantification and GC/MS analysis . . . . . . . . . . . . . . . . . B-67

QA/QC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-70 9.3.1

Solvent blanks . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-70

9.3.2

Filter or procedural blanks . . . . . . . . . . . . . . . . . . . . . . B-70

9.3.3

Precision . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-70 B-4

10 ANALYSIS OF TEFLON FILTERS . . . . . . . . . . . . . . . . . . . . . . . B-71 10.1 Extracting Teflon Filters for Analysis . . . . . . . . . . . . . . . . . . . . . B-71 10.1.1 Apparatus and Materials for extraction . . . . . . . . . . . . . . . . B-71 10.1.2 Preparing teflon filter extracts . . . . . . . . . . . . . . . . . . . . B-72 10.2 Ion Chromatography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-72 10.2.1 Apparatus and Materials for IC Analysis . . . . . . . . . . . . . . . B-72 10.2.2 Advance preparations . . . . . . . . . . . . . . . . . . . . . . . . . B-73 10.2.3 Master standards . . . . . . . . . . . . . . . . . . . . . . . . . . . B-74 10.2.4 Eluent . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-74 10.2.5 Acid . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-75 10.2.6 Preparing to turn on the IC . . . . . . . . . . . . . . . . . . . . . . B-75 10.2.7 Turning on the instrument . . . . . . . . . . . . . . . . . . . . . . B-76 10.2.8 Daily run . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-77 10.2.9 Calibrating the integrator . . . . . . . . . . . . . . . . . . . . . . . B-78 10.2.10 Shutdown . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-79 10.2.11 Calculations

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-79

10.2.12 Cleaning the columns . . . . . . . . . . . . . . . . . . . . . . . . . B-79 10.2.13 Common problems . . . . . . . . . . . . . . . . . . . . . . . . . . B-80 10.2.14 Quality control standards . . . . . . . . . . . . . . . . . . . . . . . B-81 10.2.15 Instrument specifications . . . . . . . . . . . . . . . . . . . . . . . B-81 10.3 Alpkem Rapid Flow Analyzer for Ammonium Analysis . . . . . . . . . . . B-86 10.3.1 Apparatus and Materials for Ammonium Analysis . . . . . . . . . B-86 10.3.2 Paper work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-86 10.3.3 Set up . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-87 10.3.4 Turn on . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-88 10.3.5 Set up the photometer . . . . . . . . . . . . . . . . . . . . . . . . B-88 10.3.6 Typical days run . . . . . . . . . . . . . . . . . . . . . . . . . . . B-89 B-5

10.3.7 End of day . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-89 10.3.8 Calibration Curves . . . . . . . . . . . . . . . . . . . . . . . . . . B-90 10.3.9 Reagent Preparation for the Alpkem . . . . . . . . . . . . . . . . . B-90 10.3.10 Quality control standards . . . . . . . . . . . . . . . . . . . . . . . B-92 10.3.11 Instrument specifications . . . . . . . . . . . . . . . . . . . . . . . B-92 11 LAB WARE CLEANING . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-96 12 ERROR DETERMINATION FOR ATMOSPHERIC CONCENTRATIONS . . B-96 12.1 Obtaining Precision Data . . . . . . . . . . . . . . . . . . . . . . . . . . . B-97 12.2 Propagation of errors in formulas . . . . . . . . . . . . . . . . . . . . . . . B-97 12.3 Example: Error Analysis for Atmospheric Concentrations . . . . . . . . . . B-98 13 REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-99

B-6

List of Figures 1

Map Showing Air Quality Monitoring Sites . . . . . . . . . . . . . . . . . B-10

2

Historical Weighing of Calibration Filter . . . . . . . . . . . . . . . . . . . B-17

3

Schematic Diagram of (a.) Andersen Sampler and (b.) Caltech Sampler . . B-25

4

Sample Daily Event Log Sheet for Andersen Samplers . . . . . . . . . . . B-26

5

Sample Daily Event Log Sheet for Caltech Samplers . . . . . . . . . . . . B-27

6

Operator Training Checklist . . . . . . . . . . . . . . . . . . . . . . . . . B-30

7

Audit Checklist . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-31

8

Field Blank Log Sheet . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-32

9

Reference Flow Sheet for Andersen Samplers . . . . . . . . . . . . . . . . B-35

10

Reference Flow Sheet for Caltech Samplers . . . . . . . . . . . . . . . . . B-42

11

Sampling Schedule . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-53

B-7

List of Log Sheets Filter Preparation Log - Baked Quartz Fiber Filters . . . . . . . . . . .

B-14

Log Sheet for Teflon Filter Weights

. . . . . . . . . . . . . . . . .

B-23

Daily Event Log Sheet (Andersen Sampler) . . . . . . . . . . . . . .

B-26

Daily Event Log Sheet (Caltech Sampler) . . . . . . . . . . . . . . .

B-27

Operator Training Checklist

. . . . . . . . . . . . . . . . . . . .

B-30

. . . . . . . . . . . . . . . . . . . . . . . . .

B-31

Audit Checklist

Field Blank Log Sheet

. . . . . . . . . . . . . . . . . . . . . .

Reference Flow Sheet (Andersen Sampler)

B-32

. . . . . . . . . . . . . .

B-35

. . . . . . . . . . . . . . .

B-42

. . . . . . . . . . . . . . . . . . . .

B-63

. . . . . . . . . . . . . . . . . . . . .

B-83

Daily Standards Preparation Sheet . . . . . . . . . . . . . . . . . .

B-84

IC Sample Log Sheet . . . . . . . . . . . . . . . . . . . . . . .

B-85

Reagent Preparation Sheet - NH4 Analysis

B-93

Reference Flow Sheet (Caltech Sampler) OC/EC Analysis Log Sheet Eluent Preparation Sheet

. . . . . . . . . . . . . .

NH4 Analysis - Daily Standards Preparation Sheet

. . . . . . . . . . .

B-94

NH4 Sample Log Sheet . . . . . . . . . . . . . . . . . . . . . .

B-95

B-8

1

INTRODUCTION

Air monitoring will be conducted at seven locations in Hong Kong and the Pearl River Delta region of China beginning in October, 2002. The study will be conducted under the direction of Professor Michael Bergin at the Georgia Institute of Technology. The project’s collaboration partners include Peking University, Civic Exchange, Environmental Protection Department (Hong Kong), Georgia Institute of Technology, Guangzhou Research Section of Environmental Sciences, Hong Kong Polytechnic University and Hong Kong University of Science & Technology. The goal of this study is to make fine particle PM2.5 measurements to manage air quality in Hong Kong and the Pearl River Delta, China. The design, installation and preliminary operation of a fine particle monitoring network that spans the Hong Kong and Guangdong area will be undertaken. The network will be able to help define the pollutant concentrations in background air advected into the PRD region from upwind at various times of the year. Each monitoring station has been designated using a two letter identifying code and a site number that will be placed on all sample labels. See Figure 1. The station ID codes are: TM1 TC2 CW3 SZ4 GZ5 CH6 ZS7

Tap Mun Tung Chung Central and Western Shenzhen Guangzhou Conghua Zhongshan

Thermo Andersen PM2.5 speciation samplers will be used at the first five stations, and Caltech Gray Box PM2.5 ambient samplers (prototype models for the Thermo Andersen samplers) will be used at the remaining two sites. The Andersen samplers follow the monitoring requirements of the US PM2.5 national Ambient Air Quality Standards, 40 CFR 58, (EPA, 1997) and the Caltech samplers have had over 17 years of worldwide field experience (Solomon, et al. 1988; Salmon, et al. 1994, 1996). Both types of samplers work using pumps connected to a manifold with critical orifices for flow control. Andersen samplers also have mass flow meters to measure flows. Nominal flow rates in the samplers are:

SAMPLER TYPE Andersen Sampler Caltech Samplers

QUARTZ FILTERS 16.7 14

TEFLON FILTERS 7.3 10

During each 24-hour sampling event, fine particle samples will be collected on several filters operated in parallel which will be analyzed chemically to determine ionic B-9

112˚ 24˚

113˚

114˚

115˚ 24˚

Conghua

CHINA Guangdong Province

Guangzhou

23˚

23˚

Shenzhen Zhongshan

Tap Mun

Tung Chung

Central and Western

HONG KONG 22˚ 112˚

113˚

114˚

Figure 1: Map Showing Air Quality Monitoring Sites

B-10

22˚ 115˚

species (sulfate, nitrate, chloride) concentrations by ion chromatography, ammonium ion by colorimetric analysis, elemental carbon and organic carbon by thermal evolution and combustion analysis, and trace elements by X-ray fluorescence or inductively-coupled plasma mass spectrometry. From these measurements, a material balance on the chemical composition of the particles on each day will be created that generally indicates the overall nature of the local fine particle problem. For example, the relative importance of sulfates from sulfur-bearing fuel use vs. soot from incomplete combustion vs. soil dust from travel over dusty streets may be determined. Field sampling will be done on a 6 day schedule. This means that a 24-hour sampling event takes place once every 6 days following a predetermined schedule. Loading of filter samples will take place on the day before each sampling event and unloading will take place on the day after. Samplers are equipped with timers that will be set to take samples for the 24-hour period beginning at 12:00AM (midnight) on the day of sampling. The air monitoring network will be operated by taking samples for 24-hour periods at six-day intervals for one month in each quarter of the sampling year (e.g., October, 2002; December, 2002; March, 2003; and June, 2003). This sampling schedule was chosen to observe conditions that occur throughout the year as well as to permit determination of approximate annual average concentrations. This Standard Operating Procedures (SOP) for sample preparation, handling, and analysis is to provide guidance for handling and quality control for the 47mm diameter filters that will be used for atmospheric measurements during the Pilot Study in Hong Kong and the Pearl River Delta region. Details for substrate preparation, weighing procedures, loading and unloading filter samples in the field, as well as procedures for organic and elemental carbon analysis, and analysis of Teflon filters will be found in the following sections of this document.

2

PREPARATION OF QUARTZ FIBER FILTERS

2.1 Apparatus and Materials The following apparatus and materials are used to prepare quartz fiber filters for carbonaceous aerosol collection.

B-11

ITEM

SUPPLIER

MODEL or CATALOG NO.

Pall Gelman

2500QAO-UP

Furnace

Wheelco

292

Tweezers style AA

Ted Pella

5734

Gelman s.s. forceps

VWR

51147

Petri dishes

VWR

25388-606

8oz wide mouth jar

VWR

16195-088

Quartz fiber filters

Aluminum foil Disposable lab wipe Methanol HPLC Grade

Reynolds Kimwipes VWR

EM-MX0475-1

2.2 General Information Quartz fiber filters (type 2500QAO-UP), 47mm diameter, are available from Pall Gelman (formerly Pallflex). Before use the filters must be prepared and stored such that the organic carbon and elemental carbon blank values are low, and care must be used in handling the filters. It is desirable to prepare baked aluminum foil to use as a work surface on the lab bench and for field operations.

2.3 Procedure for Baking Quartz Filters 1. Glass jars should be cleaned and pre-baked in advance. Jars as received, sometimes have cardboard dust in them. They should be cleaned before baking using DI water obtained from the MilliQ dispenser. Rinse the jars 3-4 times using DI water and visually inspect them to make sure they look clean and are not cracked. Set the jars with the open end down on clean kimwipes until they dry. Also briefly rinse the plastic caps in DI water and set them on clean kimwipes to dry. The jar caps contain teflon lid liners. Rinse the teflon liners with methanol and dry on clean kimwipes. 2. For 47mm diameter quartz fiber filters, use 8 oz. annealed glass jars covered with baked aluminum foil to hold filters while baking. 3. Rolls of aluminum foil can be annealed by removing the cardboard core and baking at 550o C overnight. 4. Place approximately 50 quartz filters in each jar. Only clean, solvent-rinsed tweezers are used to touch the filters, and only on filter edges. Individual filters are examined, and filters that have any visible imperfections should be discarded. B-12

5. All of the filters, which come stuck together in their boxes, should be individually separated before baking. Stack the filters loosely in the jar then cover the top of the jar tightly with a double layer of annealed aluminum foil. Several jars of filters can be baked in the oven at one time. It is also possible to bake a jar or two containing quartz filters at the same time that empty glass jars are being baked. 6. Filters are baked at 550o C for approximately 16 hours (or overnight). This includes approximately 2 hours for the oven to heat to 550o C. After baking, the jars with the filters are cooled in the oven. After cooling, jars are capped with a jar lid containing a solvent washed teflon lid liner and stored in a freezer until they have been sealed into prepared petri dishes described below. 7. Each time a set of filters are baked the relevant information is entered onto the filter preparation log sheet. This information includes the lot number from the box of filters, the Caltech assigned batch number (e.g., Q-19), the date, time, and temperature baked. A new batch number is assigned for each batch of filters prepared. See Filter Preparation Log for Baked Quartz Fiber Filters. 8. After baking, allow the jars with filters to cool in the oven with the door ajar until they can be handled. As soon as the jars are cool enough to touch, remove them and replace the foil cover with a jar lid containing a solvent washed teflon lid liner. Cover the jars tightly and note the batch number on a label on the jar lid. Place the jars with baked filters in a freezer until they have been sealed into prepared petri dishes described below.

2.4

Petri Dishes for Baked 47mm Quartz Filters

Prepared petri dishes consist of annealed aluminum foil press-fitted into the top and bottom of petri dishes (VWR, cat. no. 25388-606). Foil liners are prepared by cutting circular pieces of aluminum foil using a 57mm diameter template, 2 foil circles per petri dish. The aluminum foil circles are annealed in the oven for approximately 16 hours at 550o C. After the aluminum foil circles have cooled, they are handled using only clean, solvent-rinsed tweezers. On a clean work area on a lab bench covered with pieces of annealed aluminum foil, the circular aluminum foil pieces are press fit, one at a time, into the tops and bottoms of petri dishes. This is done using a solvent-cleaned Lucite cylinder that was machined to be slightly smaller than the inside diameters of the petri dish tops and bottoms. One baked quartz filter is placed into each foil lined petri dish. The edges of each petri dish are sealed by stretching Teflon tape around the bottom edge of the petri dish sealing the gap. These are stored in a petri dish holder at room temperature in the lab or in a freezer until used for sampling.

B-13

FILTER PREPARATION LOG Baked Quartz Fiber Filters ======================================================================== | | | BAKED QUARTZ | | | | PROJECT __________________________________________ | | | | CALTECH BATCH NUMBER _____________________________ | | | | FILTER LOT NUMBER ________________________________ | | | | BAKED TEMP.____________ BAKED TIME ___________HRS | | | | DATE BAKED ____________ BY ______________________ | | | | DATE PACKAGED _________ PACKED BY _______________ | | | | BAKED QUARTZ | | | | PROJECT __________________________________________ | | | | CALTECH BATCH NUMBER _____________________________ | | | | FILTER LOT NUMBER ________________________________ | | | | BAKED TEMP.____________ BAKED TIME ___________HRS | | | | DATE BAKED ____________ BY ______________________ | | | | DATE PACKAGED _________ PACKED BY _______________ | | | | | ========================================================================

B-14

3

PROCEDURE FOR WEIGHING TEFLON FILTERS

3.1 Apparatus and Materials The following apparatus and materials are used to perform the mass weighing activities. ITEM

SUPPLIER

MODEL or CATALOG NO.

Mettler Instruments

M5-SA

Cahn

10-200 mg

Cole Parmer

8368-60

Markson

6612A1281

Sling psychrometer

Ertco

PS100C

210 Polonium strip

VWR

58580-041

Staticmaster positioner

VWR

58582-003

Petri dishes

VWR

25388-606

Microbalance Calibration Weights Hygrothermograph Digital thermohygrometer

Disposable lab wipe

Kimwipes

Tweezers style AA

Ted Pella

5734

VWR

21924-154

Pall Gelman

R2PJ047

McMaster-Carr

4591K12

Small Parts

D-HM-915

Lo Count Mat 18x45 47mm Teflon filter Teflon tape 1/2” wide Blower bulb Aluminum foil

Reynolds

3.2 Balance Facilities The balance room located at the California Institute of Technology, Pasadena, CA is completely isolated from the outside. It is located in the sub-basement of Keck Laboratories inside a separate room, 042I, within the specially designed vibration-free microscopy facility, room 042. The balance room is completely isolated from the outside and equipped with 24-hour air conditioning that maintains a constant temperature of 21o C. A Lo Count Mat is installed at the entrance to the balance room to remove loose objects from the bottom of shoes on entering personnel. Only authorized balance users who have been trained are allowed to operate the microbalance. A filter conditioning and weighing chamber is located next to the microbalance. Temperature and RH inside the chamber is monitored when filters are equilibrating. Pre-sampling filters are equilibrated separately from post-sampling filters. B-15

3.3 Environmental Requirements for Filter Weighing Before weighing, Teflon filters are equilibrated for at least 24 hours in a constant temperature, constant relative humidity environment. Temperature and relative humidity guidelines from EPA Quality Assurance Document 2.12 (EPA, 1998) are used. Temperature should be in the range 20-23o C and RH should be in the range 30-40%.

3.4 Electrostatic Charge Neutralization Staticmaster ionizing units (polonium 210 Po) are used to reduce electrostatic charges that could interfere with balance operations. A 210 Po strip is attached to a staticmaster positioner with a flex spring to allow easy positioning of the unit. For most effective results, the 210 Po strip should be approximately 1/2 to 3/4 inch away from the surface to be neutralized. Strips should be changed every 6 months. Old strips should be disposed of through the safety office.

3.5 General Procedure for Weighing Teflon Filters Teflon filters (Gelman, 2.0 µm pore size, R2PJ047, Teflon membrane with PMP support rings) are placed individually into clean petri dishes (VWR, cat. no. 25388-606). Each petri dish containing a Teflon filter is assigned a five digit number in addition to the sample label. This code is specific for the individual filter and used to identify it for mass determination. All filters are visually inspected for defects before the initial weighing. Any filters with defects are discarded. Defects include holes, discoloration, dirt, detached support rings, or severe curl in the filters. Filters are handled using only clean, non-serrated tweezers, and touched only on the support rings around the edge. The date, filter lot number, pore size, and any comments are recorded in a filter weight log book. There is also a master weight log book kept by the balance for recording calibration filter and calibration weight data along with temperature and relative humidity information. At the beginning of each weighing session a set of calibration filters is weighed. Calibration filter weights must agree with their historical weighings. See Figure 2, for example. Calibration weights are used as an additional diagnostic tool to verify correct balance operation. Periodically over the course of an experiment enough calibration weighings will be made to completely bracket the expected range of Teflon filter weights. Teflon filters tend to accumulate static charge. Therefore, each filter is passed under a static discharging strip before weighing. These discharging strips contain a mildly radioactive source 210 Po which neutralize electrostatic charges on items brought within one inch.

B-16

Figure 2: Historical Weighing of Calibration Filter

B-17

Mass (µg)

154600 1000

154650

154700

154750

154800

154850

154900

1500

2000 2500 day (1000=September 28, 1989)

Calibration Filter 2c

3000

3500

Each Teflon filter is weighed more than once, how many times depends on how repeatable the operator is finding the masses. Approximately half of each batch of filters are weighed subsequently on a different day to verify repeatability during initial weighings. After initial weighing, pre-weighed filters are stored in closed petri dishes at room temperature or in a freezer until they are used for sampling. Each petri dish is sealed with teflon tape. Sampled filters containing particles are stored upright in the freezer in petri dishes wrapped around the edges with teflon tape. 3.5.1 Preparation for weighing teflon filters 1. Set out unused filters in slightly open petri dishes for at least 24 hours to equilibrate in a constant temperature, constant relative humidity environment, currently established in room 042I Keck. Extra care should be taken with filters containing samples to avoid losing volatile species while the sample is out of the freezer. Sampled filters should be returned to the freezer as soon as possible after weighing. Equilibrating filters are kept in a chamber to avoid contamination by airborne dust during the equilibration period. 2. If the temperature and humidity of the room are not near the ideal range of 20-23o C and 30-40% RH, contact Physical Plant for HVAC service. Do not equilibrate or weigh filters. 3. The Mettler balance is a mechanical balance and does not need any warm up period. In fact, it is harmful to leave the Mettler balance on with no weights engaged for long periods of time, and this should be avoided. The key to successful use is to be PATIENT. Rotate all knobs and dials SLOWLY and GENTLY when interacting with the balance. 4. Always record the date, filter lot number, and any comments on the Log Sheet for Teflon Filter Weights. There is also a master weight log book kept by the balance for recording calibration filter and calibration weight data along with temperature and humidity information. It also contains pages at the front for assigning filter ID numbers to teflon filters. 5. Check the cleanliness of the balance area. If necessary, put out clean aluminum foil as a work surface near the balance. Use kimwipes to wipe surfaces. 6. At the beginning of each weighing session weigh a set of calibration filters. If calibration filter weights do not agree with their historical weighings, stop and determine if there is a problem with the balance. The balance can be sensitive to vibrations and activities when there is construction work in the building. Other potential problems come from static and air drafts. Calibration weights can be used as an additional diagnostic tool.

B-18

7. Periodically over the course of an experiment perform enough calibration weighings to completely bracket the expected range of Teflon filter weights. For 2.0 µm Teflon filters, which usually weigh between 130 and 170 mg, calibration weighings should be made for 130, 140, 150, 160, and 170 mg. The calibration weight data should be recorded BOTH in the master weight log book kept by the balance, and on the project specific Log Sheets for Teflon Filter Weights. See the instructions in the master log book to know which calibration weights to use. Always handle calibration weights with plastic tweezers kept in the box with them. 8. The balance zero should be checked approximately once every fifth weighing. Tiny adjustments (< 1mm) to the zero may be required but any significant change is a sign of trouble. Common zero problems include accidentally bumping the balance table. If the zero has drifted, reweigh all filters since the last zero check. 9. Always leave the balance ”off” when weighing is done. This means the front knob should be left in the 12 o’clock position with no weights engaged. 10. Tweezers should be stored in the plastic box labelled “tweezers, for weighing only”. These tweezers should not be used for any purpose other than weighing teflon filters. Inspect tweezers before use, if necessary, clean by rinsing with methanol and wiping with a kimwipe. 3.5.2

Precise weighing procedure

1. Blow off any dust on the weighing pan with a bulb syringe before beginning. Set all balance weights to zero. Then completely release the weighing pan by turning the knob on the front of the balance one-quarter turn to the left. The reading should be exactly zero. To adjust the zero, use the knob toward the top rear on the right side of the balance. 2. If the zero is too far out of alignment to adjust with the zero knob, check the balance is level by looking at the bubble spirit level inside the chamber. If the bubble is not centered, the three balance feet can be adjusted by rotating them. 3. When zero is adjusted, turn off the balance by moving the front knob back to the 12 o’clock position. 4. To access the weighing pan, rotate the knob on the lower right hand side of the balance counterclockwise (slowly). (The knob on the lower left hand side can also be used – it should be rotated clockwise to bring out the pan.) The pan will come out of the window in the front. 5. Caution: never move the weighing pan when the pan is fully released (front knob in the 9 o’clock position). This may damage the knife edges in the balance. 6. Place the item to be weighed on the weighing pan. Always handle the filters and calibration weights with tweezers. For a Teflon filter, first discharge it by passing it B-19

several times under a discharging strip. Caution: this is a mildly radioactive source Po210 . Handle with care, and have the Safety Office discard old dischargers . Place the discharged filter on the pan so that it will be centered over the weighing pan. Move the pan back into the weighing area by rotating the lower right hand knob slowly clockwise. 7. Pre-weigh the item by turning the knob on the front to the RIGHT (to about 2 o’clock). This partially releases the pan. Adjust the weights to add the most weight possible WITHOUT causing the light meter reading to go ”subzero”. 8. Turn the knob counterclockwise to 9 o’clock to fully release the pan. Then adjust the wheel controlling the µg digits of the weight to get a final reading. (Make sure the edges of the Teflon filter are not touching the pan lifting mechanism.) Record the mass on the Log Sheet for Teflon Filter Weighing. 9. Return the knob to 12 o’clock, then use the knob on the lower right side to take out the pan and remove the item being weighed. 10. Put the empty pan back in, reset the weights to zero, and recheck the zero before continuing. If the zero has drifted off, all items since the last zero check must be reweighed before continuing. 3.5.3

Precautions for weighing teflon filters

1. Teflon filters occasionally are somewhat curled, which can interfere with the position of the pan in the balance. To ensure that this does not result in bad weighings, filters should be weighed repeatedly, rotating the filter 90 degrees each time, until 2 sequential readings are within 15 µg of each other. Occasionally, a filter will not give reproducible results on subsequent weighings and should not be used. 2. Teflon filters tend to accumulate static charge, which is why each filter is passed under a static discharging strip before weighing. In addition to this procedure, it is helpful to place the filter under a discharging strip while it is still in the petri dish bottom for a few minutes before weighing. When weighing one filter, take the next one to be weighed, remove the petri dish lid, and put under a discharging strip. 3. Teflon filters are not cheap, so handle them carefully with tweezers. If you accidentally put a hole in a filter while pre-weighing it, set it aside. It can be used as an analytical blank, but it can’t be used to collect an aerosol sample. Also check for tiny existing pinholes in filters directly out of the box. As many as 5-10% of some filter lots may have pre-existing flaws.

B-20

3.6 Quality Control ITEM

MEASUREMENT FREQUENCY

ACCEPTANCE CRITERIA

Temperature

each weighing session

20-23o C (2o C sd)

Relative Humidity

each weighing session

30-40% (5% sd)

Calibration Filters

each weighing session

± 15µg

Calibration Weights

Bi-annually

± 10µg

Balance calibration

Annually or as needed

not applicable

Temp. & RH verification

Bi-annually

2o C;5%

Duplicate filter weighing

each filter

± 15µg

50% of filters

± 15µg

as needed

± 15µg

quarterly

± 30µg

Different day weighing Weighing Lab Blanks Field Blanks

3.6.1

Teflon filter weighing lab blanks

Include approximately 5-10% additional teflon filters with each monthly sample set to be used as weighing lab blanks. These filters will be weighed initially with the rest of the samples. However, after their initial weights have been determined, these filters will be kept in petri dishes sealed with teflon tape and will not leave the laboratory facility at Caltech. At the end of each sampling month when sampled filters are returned, the teflon filter weighing lab blanks will be re-weighed when the post-sampling filters final masses are determined. 3.6.2

Teflon filter field blanks

Field blanks are unsampled filters that are used to determine whether contamination is occurring during sample loading/unloading and sample transportation procedures. Weigh enough field blanks during pre-sampling weighing sessions to provide one single-use field blank for each sampler during the upcoming sampling month. Weigh the field blanks with the post-sampling filters. If the weight change in the field blank exceeds 30 µg, contamination during transportation or at the sampling site may be occurring. Take appropriate troubleshooting and corrective action in that event. 3.6.3

Mass error determination

For gravimetric mass determination, the reproducibility of the balance was determined by making a large number (n > 500) of replicate weighings over the course of an experiment. B-21

The precision for each weighing was found to be ± 11.2 µg per filter. The initial and final weighing errors are combined to obtain the precision for sample mass determination. Final error bound estimates are obtained by the statistical propagation of the sample, filter blank, and sampling volume precisions. 3.6.4

General precautions

1. Avoid weighing under unusual weather conditions, especially when it is raining or about to rain. Static charging of the balance is especially bad at these times, and interferes with weighing. 2. SIGNS OF STATIC CHARGING: The zero drifts badly from weighing to weighing, up and down without trend. Sometimes, the zero reading won’t stabilize – the light meter displays a fuzzy reading. Try using the static discharger gun and shoot it inside the weighing area. Squeezing the trigger releases positive ions; when the trigger is released, negative ions are released such that the static is neutralized. Squeeze and release the trigger slowly, otherwise no ions will be emitted from the crystal inside the gun. If static is not eliminated, turn off the balance and try another day. 3. DRIFTING FILTER WEIGHTS: Sometimes, the filter weights will drift during weighing. Drifting filter weights can be due to inadequate filter equilibration, or to curled filters interfering with the movement of the balance. If the filter has not equilibrated properly, return it to the conditioning chamber and weigh it on a future day. 4. Activities near the room can also interfere with the balance. If workmen are working in the area, vibrations from their activities can disturb the balance readings. Turn off the balance and come back another time.

B-22

LOG SHEET for

TEFLON FILTER WEIGHTS

page 1 of ___

DATE ___________________________

OPERATOR __________________________________

TIME EQUILIBRATED_______________

ROOM TEMP_____________

%RH________________

COMMENTS_____________________________________________________________________ :::::::CONTROL FILTERS-SPECIFY ID #::::::::::::: 2c

11c

2a

2b

1ST

__________

__________

__________

_________

2ND

__________

__________

__________

_________

CIRCLE:

INITIAL or FINAL WEIGHTS

LOT NUMBER________________________

============================================================================== |FILTER | SAMPLE ID. | FILTER WEIGHTS | | NO. | NO. | ********************************************* | |_______|_______________|____________|_____________|____________|____________| | | | | | | | |_______|_______________|____________|_____________|____________|____________| | | | | | | | |_______|_______________|____________|_____________|____________|____________| | | | | | | | |_______|_______________|____________|_____________|____________|____________| | | | | | | | |_______|_______________|____________|_____________|____________|____________| | | | | | | | |_______|_______________|____________|_____________|____________|____________| | | | | | | | |_______|_______________|____________|_____________|____________|____________| | | | | | | | |_______|_______________|____________|_____________|____________|____________| | | | | | | | |_______|_______________|____________|_____________|____________|____________| | | | | | | | |_______|_______________|____________|_____________|____________|____________| | | | | | | | |_______|_______________|____________|_____________|____________|____________| | | | | | | | |_______|_______________|____________|_____________|____________|____________|

B-23

4

SAMPLE LOADING/UNLOADING IN THE FIELD

The Andersen Instruments RAAS2.5-400 fine particulate chemical speciation sampler is one of two types of instrument that will be used for collecting aerosol samples during the Pilot Study. The fundamental design concept for the RAAS2.5-400 was developed at Caltech by Dr. Glen Cass. The Caltech gray box PM2.5 samplers, prototype for the Andersen instrument, will also be used at two of the PRD monitoring stations. Each sampler contains 4 filter holders for collecting fine particle samples located inside the cabinet. There are two quartz fiber filters for collecting organic particles. The other two filters are teflon membrane filters used for aerosol mass, ionic species (Cl− , SO4 = , NO3 − , NH4 + ), and trace metals determinations. These filter holders are labelled: Quartz 1, Teflon 2, Teflon 3, and Quartz 4. Samplers work using pumps connected to a manifold with critical orifices for flow control. Sample flow rates are calibrated during initial field placement and verified at least once at the start of each sampling month. See Figure 3 for a schematic diagram of the two sampler types. A notebook and tool box are provided with each sampler. The notebook contains a copy of the field operating procedures and Daily Event Log Sheets (one for each sampling event). Log sheets need to be filled out each time samples are loaded and unloaded. A Reference Flow Sheet is provided with each Log Notebook. The tool box contains the basic needs for filter changing and correcting minor problems with the sampler. Two pairs of tweezers are provided wrapped in annealed aluminum foil. Exercise care to keep the tweezers clean. Spare filter cassettes, o-rings, and a spare filter holder are provided for changing any that become soiled. There is also a cassette loader tool (Andersen only) for opening filter cassettes. In addition, the tool box contains an external flow measuring device and other useful items.

4.1 Filter samples The filters are pre-packaged in individual petri dishes which are stored in the laboratory before use. Collected samples are transported from the field in a cooler or ice chest with “blue ice” and stored in the freezer immediately upon return to the laboratory to minimize loss of volatile materials. Each sample is given a label in the form: TM1 − Q1 − 021002 where TM1 is the site label (TM1, TC2, CW3, SZ4, GZ5, CH6, or ZS7); Q1 is the filter holder label (Q1, T2, T3, or Q4); and 021002 is the sampling period given as year month day (YYMMDD). The corresponding dates for each sampling event are recorded on the log sheets.

B-24

INLET 48 LPM

Flow Splitter 16.7 LPM

7.3 LPM

7.3 LPM

24 LPM

24 LPM

Cyclone

QUARTZ FILTER

16.7 LPM

Cyclone

TEFLON FILTER

TEFLON FILTER

Critical Orifices

MASS

MASS = SO4 N03 + NH4

34 TRACE ELEMENTS

TO PUMP

QUARTZ FILTER ORGANIC CARBON ELEMENTAL CARBON

(a) Andersen Sampler

INLET 1

14 LPM

INLET 2

10 LPM

10 LPM

24 LPM

24 LPM

Cyclone

QUARTZ FILTER

TEFLON FILTER MASS

14 LPM

Cyclone

Critical Orifices

Critical Orifices

TO PUMP

TO PUMP

34 TRACE ELEMENTS

TEFLON FILTER MASS = SO4 N03 + NH4

QUARTZ FILTER ORGANIC CARBON ELEMENTAL CARBON

(b) Caltech Sampler Figure 3: Schematic Diagram of (a.) Andersen Sampler and (b.) Caltech Sampler

B-25

Daily Event Log Sheet Site: TM1 Tap Mun INITIAL

Filter Set ID:

Date:

Time:

Operator:

MIDDLE Date:

Time:

Operator:

END

Time:

Operator:

Date:

Rotameter ID#: Initial Filter Holder

Filter ID # (FID)

Middle

End

Initial Flow Rates Final Flow Rates Avg flow Vol Andersen Rotameter Andersen Rotameter (L/m) (m3) (L/m) (balls) (L/m) (balls) Q1a

black

black

silver

silver

black

black

silver

silver

black

silver

silver

black

black

black

silver

silver

Quartz 1 Q2a

Teflon 2 Q3a

Teflon 3 Q4a

Quartz 4

Weather Conditions: At end of sampling, record values below: Sampling Start: Date: Sampling Stop: Date: Duration (hours): Avg

Time: Time:

Min

Max

Comments:

Ambient Temp. Barometric Pres. Humidity Supplies Needed:

Figure 4: Sample Daily Event Log Sheet for Andersen Samplers

B-26

Daily Event Log Sheet Site: ZS7 Zhongshan

Filter Set ID:

INITIAL

Date:

Time:

Operator:

END

Date:

Time:

Operator:

Rotameter ID#:

Filter Holder

Filter ID # (FID)

Initial Flow Rates Rotameter (balls)

Quartz 1 Teflon 2 Teflon 3 Quartz 4

Final Flow Rates Rotameter (balls)

black

black

silver

silver

black

black

silver

silver

black

black

silver

silver

black

black

silver

silver

Weather Conditions: At end of sampling, record values below: Sampling Start: Date: Sampling Stop: Date: Duration (elapsed time):

Time: Time:

Supplies Needed: Comments:

Figure 5: Sample Daily Event Log Sheet for Caltech Samplers

B-27

One filter set includes 4 filters in petri dishes: (Q1) (T2) (T3) (Q4)

Quartz fiber filter, in foil lined petri dish Teflon membrane filter Teflon membrane filter Quartz fiber filter, in foil lined petri dish

When loading filters, the most important thing is to make sure the correct filter is loaded into the correct holder. Each filter has a designation (e.g., Q1) on the petri dish and the filter holders are labelled similarly (e.g., Quartz 1). Do not use any filter with a hole or other imperfection. If a filter is dropped on the ground while loading, do not use it. In either case, take a spare filter from the set of spares provided and write the sample label on the spare filter petri dish. Cross out the label on the original problem filter and write BAD on it to avoid mistaken use in the future. Bad filters should be kept and returned with the other samples. Each filter sample also has a unique 5 digit filter ID number (FID) on the petri dish. This number is specific for the individual filter. These ID numbers should be written on the daily event log sheets. When using a spare filter, make sure it remains with its original petri dish and write the sample label for the site and date of the sample on the spare filter petri dish. Don’t change the FID on the spare filter. All filters are loaded maintaining the same side up that is currently up in the petri dish. Filters are handled only using tweezers, never touched with hands or gloves. Operators should wear a clean pair of gloves (provided) when loading filter cassettes. A clean piece of aluminum foil should be placed on top of the sampler as a work surface. No smoking is allowed in the sampler area or when loading or unloading samples.

4.2

Loading Filter Samples

The filter holders are located downstream of a cyclone for collecting fine particles. The filter holders are left in the sampler between sampling periods with the empty filter cassettes inside. They should be completely assembled and connected so that the interior remains clean when not in use. The filter holders are connected to critical flow orifices on the manifold at the bottom of the sampler. A reference flow sheet with the calibrated flow rates for each site is provided. Operators will verify flows are in acceptable ranges when loading samples. Filters should be loaded as close to the sampling start time as possible to minimize contamination. Ideally, samples should be loaded no more than 12 hours in advance of sample start time. Special exceptions will be made for situations when the sampler cannot be accessed within this time frame. Please check with one of the contact help list people for permission to load a sample outside the 12 hour window.

B-28

Only authorized operators who have been trained may load/unload samples. The samplers and operators will be audited at least once to verify that proper procedures are being followed. See the Operator Training and Audit Checklists for details. 4.2.1 Field blank A field blank is a set of samples that are taken to the field and loaded normally and then immediately removed without running the sampler. The field blank should be taken on the last sampling day of each month after the final sample has finished. Field blank designations are: 0210FB - October, 0212FB - December, 0303FB - March, and 0306FB June. See the sample Field Blank Log Sheet.

4.3

Andersen Samplers

4.3.1 Filter loading for Andersen samplers 1. Set out a piece of clean annealed aluminum foil on top of the sampler as a work surface. Replace aluminum foil with fresh pieces as needed during the loading process. 2. Get items needed for filter load: tweezers, cassette loader tool, gloves, and 4 petri dishes with filters to be loaded. 3. Open the cabinet door which may be dirty, then put on a pair of clean disposable gloves. Do not touch anything dirty with gloves. If the gloves become dirty, throw them out and put on a new clean pair before continuing. 4. Beginning with the first filter holder, unscrew the black hand ring in the middle of the filter holder and remove the white filter cassette. Place the cassette on the clean foil work surface. 5. Loosely reconnect the filter holder so the interior remains clean while loading the filter into the cassette. Do not set anything down on a dirty surface. 6. Open the cassette using the aluminum cassette loader tool. Handle the cassette only on the edges, and take care not to touch the interior. 7. Using tweezers, take a filter from the correct petri dish and load it into the filter cassette. Only grasp the filter on the edge with tweezers. 8. Using the cassette loader to provide even pressure, carefully push the top of the filter cassette on to close. Inspect the loaded filter and make sure it is centered with no wrinkles. If the filter is wrinkled or uneven, reopen the cassette and press closed again so the filter is smooth and flat. A spare filter may be needed if the filter is unable to lay flat. B-29

Operator Training Checklist DATE:_______________ TRAINER:__________________ OPERATOR: _________________ Operator is aware of: YES

NO Smoking is not allowed near samplers. Sampler interior must be kept neat and clean. Fans must be kept unblocked. Supplies that are required for operation of the sampler, and how to obtain them if needed. That tweezers must be kept clean, wrapped in baked aluminum foil. Time frame for loading and unloading filters for a sampling event? That the sampled filters must be transported in coolers with blue ice after each event?

Loading/Unloading Filters: YES

NO Did operator wear clean gloves when touching the filter cassettes? Did operator use only tweezers to touch filters? Did operator keep aluminum cassette loader clean - placing it on aluminum foil during use and storing in a clean plastic bag? Did operator load filters properly into the cassettes? Did operator load cassettes properly into the filter holders? Did operator avoid placing any portion of the filter holders on a dirty surface during loading/unloading? Were filter holders hand-tightened prior to starting a sampling event? Did operator place filter into correct petri dish after event was finished? Did operator seal samples in petri dishes with Teflon tape?

Running an event and flow checks: YES

NO Did operator program an event correctly? Did operator perform the Rotameter and Andersen flow checks properly? Given a problem in operation, does the operator know who to contact for help?

Data acquisition and sample transportation: YES

NO Did the operator fill out the sample log sheets completely and accurately? Can the operator download data using the Datalink device?

Operator Authorization: YES

NO Did operator pass training program?

Figure 6: Operator Training Checklist

B-30

AUDIT CHECKLIST SITE: _______________ DATE:_______________ AUDITOR:__________________ OPERATOR: _________________ Sampler Appearance YES

NO Is the outside of the sampler undamaged? Is the inside of the sampler neat and clean? Are any fans blocked by items left in the sampler? Are there any incidents of operators smoking around the sampler? Are all necessary supplies in tool box? Is the power cord secure and undamaged?

Loading/Unloading Filters: YES

NO Are operators wearing clean gloves when touching the filter cassettes? Are tweezers kept clean, wrapped in clean foil? Are operators using only tweezers to touch filters? Are operators keeping aluminum cassette loader clean - placing it on aluminum foil during use and storing in a clean plastic bag? Are filters being loaded properly into the cassettes? Are cassettes being loaded properly into the filter holders? Are operators placing any portion of the filter holders on a dirty surface during loading/unloading? Are filter holders hand-tightened prior to starting a sampling event? Are operators taking care to place filter into correct petri dish after event is finished? Are all samples in petri dishes sealed with Teflon tape?

Running an event and flow checks: YES

NO Are filters loaded and unloaded within the set time-frame? Are operators programming an event correctly? Are the sampling events being done on the correct dates? Are operators doing the initial, middle, and final checks for each sampling event? Are operators performing the Rotameter and Andersen flow checks properly? Given a problem in operation, do the operators know who to contact for help?

Data acquisition and sample transportation: YES

NO Are the log sheets being filled out completely and accurately? Are operators obtaining data using the Datalink device for each event and bringing it to the appropriate person? Are the sampled filters transported in coolers?

Figure 7: Audit Checklist

B-31

Field Blank Log Sheet Site: ZS7 Filter Set ID: _________________ Zhongshan

Date: __________

Time: __________

Filter Holder

Operator: ___________________

Filter ID (FID)

Quartz 1

Teflon 2

Teflon 3

Quartz 4

Weather Conditions:______________________________________________________

Supplies needed:________________________________________________________

Comments:

Figure 8: Field Blank Log Sheet

B-32

9. Place the loaded filter cassette back into the correct filter holder and tighten. The filter cassette should be oriented with the stainless steel screen on the bottom and the filter facing up. 10. Repeat steps 3-9 for each of the other 3 filter holders. 11. Leave empty petri dishes in the sampler to collect filters after the run. Throw away used gloves. 12. Program the sampling event and record details such as the FID numbers on the log sheet as described below. Also note if any supplies are running low and in need of replacement. 4.3.2

Leak check procedure

LEAK CHECK PROCEDURE: To be done after filters are loaded into cassettes and filter holders are hand-tightened. After leak check is completed, the operator should do the initial flow rate check. 1. Remove the inlet tube rain shield by sliding it vertically off the inlet. 2. Take the leak check adapter out of the orange tool box (the piece is a small valve attached to a small metal column) and put it on top of the inlet tube, pushing it down into place. 3. Close the valve by rotating the lever so that it is horizontal. 4. On the sampler, select “Maintenance>”, then “QC>”, and then “Leak Check”. Press the “Enter” button for “Leak Check”, the leak check will automatically begin. The counter will display a 45 second zero-flow check and then the pump will automatically start. 5. When the pump is running, the following is on screen: FLOW 1 = (number) FLOW 1 = FLOW 2 = FLOW 2 = FLOW 3 = FLOW 3 = FLOW 4 = FLOW 4 =

B-33

Look at only the top value for each flow (the top value is corrected for system noise while the bottom value is a raw uncorrected number). If the value is at or below 0.15 L/min, the flow channel is working properly. If any of the top values for a single flow channel are above 0.15 L/min, hand-tighten the filter holders. If, after tightening, there is still a problem, stop the pump by pressing “Cancel”, open the filter cassettes and check the filters are loaded correctly. Repeat the leak check. If the problem still exists, get in touch with one of the contact people right away. 6. If the leak check fails, continue the normal sample unless instructed otherwise by the contact people. 7. After completing the leak check, open the leak check valve by slowly rotating the lever. Slide the leak check piece off and put the inlet tube rain shield back on. Continue with the initial flow rate check. 4.3.3

Initial flow rate check

Check the flow rates given by the electronic flow sensors in the Andersen sampler, and also check flow rates using the external rotameter provided. Verify the flows are in the correct ranges from the Reference Flow Sheet. 1. Press the Cancel button to get to the >>>>Main Menu<<<<. 2. Use arrow key to select “Maintenance>” and press enter. 3. Enter maintenance, use arrow key to select “Yes” and press enter. 4. Use arrow key to select “Diagnostics>” and press enter. 5. Use arrow key to select “Flow Valves” and press enter. 6. There are three columns of numbers. The last column has the flow rates. Wait approximately 15 seconds for the flow rates to stabilize. Record the initial values for Quartz 1 = Q1a, Teflon 2 = Q2a, Teflon 3 = Q3a, and Quartz 4 = Q4a on the event log sheet under Initial Flow Rates, Andersen (L/m). 7. An adapter is provided to measure flows using the external flowmeter/rotameter. To measure the flow with the external rotameter, disconnect each filter holder one-at-a-time by unscrewing the black plastic fitting located at the bottom of each filter holder. Attach the rotameter to the adapter and attach the adapter to the black plastic fitting. The flow through the orifice will be measured in this manner. Hold the rotameter vertically while reading the flow. There are two balls inside the glass tube on the rotameter (black and silver). Read the number nearest to the center of each ball and record the readings under Initial Flow Rates, Rotameter (balls). Also note the Rotameter ID# on the log sheet. 8. Stop pump running by pressing cancel. B-34

Reference Flow Sheet

Site: _________________ Calibration Date:______________ Calibrated with: Standard Dry Gas Meter (NIST Traceable-ID #C-0701) Technician signature: ______________________

Reference Values Filter holder

Andersen (L/min)

Rotameter ID#__________

Q1a Quartz 1

black ball ---------------------------------------------silver ball black ball

Q2a Teflon 2

---------------------------------------------silver ball black ball

Q3a Teflon 3

---------------------------------------------silver ball black ball

Q4a Quartz 4

---------------------------------------------silver ball

Leak Check: Flow 1:___________________ Flow 2:___________________

Flow 3:___________________

Flow 4:___________________

Temperature and Pressure Verification:

Figure 9: Reference Flow Sheet for Andersen Samplers

B-35

9. Compare the measured flow rates with the reference flow sheet. The Andersen values should be within ± 10% of the reference values. The rotameter values should be within ± 10% of the reference values. 10. If the the instrument is operating within specified ranges, close the cabinet door and leave the instrument to run overnight. If there are problems, refer to the trouble shooting section, or contact one of the people from the contact help list. 4.3.4

Programming an event to run

1. Go to the >>>>Main Menu<<<< on the programming screen. Use the Cancel key to return to the previous menu until >>>>Main Menu<<<< is reached. 2. Use the arrow key “V” to select “Events>” and press enter. 3. Use the arrow key to select “Add Event>” and press enter. 4. Use the arrow key to select “Manual Event” and press enter. 5. Event Start Date: change the date that appears automatically to be the current date that sampler should start on. Date is given in Month - Day - Year (MMDDYY) format. Press enter. 6. Event Start Time: enter the time sampling should begin (for sampling starting at 12:00am enter 0000). Press enter. 7. Event Duration 24:00 24 hours, press enter. 8. Enter Filter Set ID (SID). This will be the site number followed by the date as given on the top of the event log sheet. (e.g., 1-021002). 9. Enter Filter ID (FID1): 5-digit number from petri dish 10. Enter Filter ID (FID2): 5-digit number from petri dish 11. Enter Filter ID (FID3): 5-digit number from petri dish 12. Enter Filter ID (FID4): 5-digit number from petri dish 13. Record date, time of loading, and sign name as initial operator on the log sheet. 4.3.5

Middle flow check

Visit the site at least once during the 24 hours the sample is running to verify the sampler is operating and perform middle flow checks. 1. Press the Cancel button to get to the >>>>Main Menu<<<<. B-36

2. Use arrow key to select “Events>” and press enter. 3. Use arrow key to select “View Event>” and press enter. 4. Select the current event and press enter. 5. You will see a screen with the header “Status: Executing” 6. Use the arrow key to page through the status screens and verify the settings are correct. 7. Record the current values for Flow 1, Flow 2, Flow 3, and Flow 4 on the event log sheet under Initial Flow Rates, Andersen (L/m). 8. Use the adapter provided to measure flows using the external flowmeter/rotameter. To measure the flow with the external rotameter, disconnect each filter holder one-at-a-time by unscrewing the black fitting below the filter holder. Attach the flowmeter adapter to the top of the black fitting and hold the rotameter vertically while reading the flow. There are two balls inside the glass tube on the rotameter (black and silver). Read the number nearest to the center of each ball and record the readings under Middle Flow Rates, Rotameter (balls). 9. If the instrument is operating correctly, close the cabinet door and leave the instrument to complete the run. If there are problems, refer to the trouble shooting section, or contact one of the people from the contact help list for further instructions. 10. Record date, time of middle flow check and sign name as middle operator. Also, note down what weather conditions were like on the day the samplers were run (i.e., windy, rain, etc.) 4.3.6

Unloading filter samples from Andersen samplers

Filter samples should be unloaded as soon as possible after the end of sampling. Samples should remain in the field for no more than 16 hours after the sample end time. Special exceptions will be made for situations when the sites cannot be accessed on the day after operation. Contact one of the people on the contact help list for permission to unload a sample outside the 16 hour window. After the sample run has ended, record average values on the log sheet as follows: 1. Press the Cancel button to get to the >>>>Main Menu<<<<. 2. Use arrow key to select “Events>” and press enter. 3. Use arrow key to select “View Event>” and press enter. 4. Choose the desired event and press enter. 5. You will see a screen with the header “Status: Completed” B-37

6. Use the arrow key to page through the status screens and record information onto the log sheet. Use the Actual Information for START - Date - Time and STOP - Date Time. 7. Record the “Duration” of the run (should be 24 hours unless there was a problem). 8. Use the arrow key to page through status screens and record the Ambient Temp: Avg, Min and Max values. 9. Record the Barometric Pressure: Avg, Min and Max values. 10. Record the Humidity: Avg, Min and Max values. 11. For each flow record the Avg. Flow and volume of air sampled (Vol, m3) on the log sheet. 12. Record any problems that occurred while sampling in the comments section. Continue comments on the back of the log sheet or on addition pages if necessary. 13. Record date, time of unloading and sign name as End Operator on the log sheet. 4.3.7 Steps for filter unloading 1. Set out a piece of clean aluminum foil on top of the sampler as a work surface. 2. Get items needed for filter unloading: tweezers, cassette loader tool, and 4 empty petri dishes. 3. Open the cabinet door, then put on a pair of clean disposable gloves. Do not touch anything dirty with gloves. If the gloves become dirty, throw them out and put on a new clean pair before continuing. 4. Beginning with the first filter holder, unscrew the black hand ring in the middle of the filter holder and remove the white filter cassette. Place the cassette on the clean foil work surface. 5. Loosely reconnect the filter holder so the interior remains clean while loading the filter into the cassette. Do not set anything down on a dirty surface. 6. Open the cassette using the aluminum cassette loader tool. Handle the cassette only on the edges, and take care not to touch the interior. 7. Using tweezers, carefully remove the sampled filter from the cassette and place it into the correct petri dish. Touch the filter only with tweezers and only near the edges of the filter. 8. Replace the lid on the petri dish and set sample aside where it will not be disturbed.

B-38

9. Close the cassette and place the empty filter cassette back into the filter holder and tighten. 10. Repeat steps 4-9 to unload each of the other 3 filter holders. 11. Wrap Teflon tape around the edge of each petri dish, pulling it tight to seal. Make sure to hold the dish upright while doing this. 12. Put samples in cooler with “blue ice”. 13. Download data using the DataLink attachment (see below). 14. Remove DataLink and turn power to the instrument “OFF” between sampling events. 15. Bring samples, Datalink, and log sheets to the appropriate local office after each sampling event. See sample handling/shipping. 4.3.8

Downloading data with DataLink

1. Plug in the DataLink to the RS-232 port inside the instrument. The green light will illuminate. 2. Press the Cancel button to get to the >>>>Main Menu<<<<. 3. Use arrow key to select “Data Transfer>” and press enter. 4. Use arrow key to select “Transmit All” and press enter. The screen will read “Data Log Transmission in Progress” and the green light on the DataLink will flash. 5. This process will take a few minutes. Wait for the transfer to finish and remove the DataLink from the RS-232 port.

4.4 Caltech Gray Box Samplers Caltech Gray Box samplers have two pumps connected to manifolds with critical orifices for flow control. The pumps have been wired to run on 220V power, however, the fans and elapsed timers require 110V power, and a 240-110 step-down transformer has been provided inside each sampler. 4.4.1

Filter loading for Caltech samplers

1. Set out a piece of clean aluminum foil on top of the sampler as a work surface. Replace aluminum foil with fresh pieces as needed during the loading process. 2. Get items needed for filter load: tweezers, flow meter, gloves, and 4 petri dishes with filters to be loaded. B-39

3. Open the cabinet door which may be dirty, then put on a pair of clean disposable gloves. Do not touch anything dirty with gloves. If the gloves become dirty, throw them out and put on a new clean pair before continuing. 4. Begin with the first filter holder. Disconnect the swagelock fitting at the top of the filter holder to make it easier to handle. Leave the bottom connected to the accordion style Teflon tubing. 5. Unscrew the center ring on the holder which can be removed, then lift off the top cone-shaped part of the filter holder. Be careful not to tip the bottom of the filter holder over while the top is off because the two metal support screens inside are loose and may fall out. If the screens do fall out, replace them with clean spares and note that the smaller diameter screen needs to be oriented with its more curved side down, otherwise spots will appear on the filter where this screen touches the support screen on top and causes uneven air flow through the filter. Do not set anything down on a dirty surface. 6. Using tweezers, take a filter from the correct petri dish and place it centered on top of the stainless steel screen. Quartz fiber filters have a stainless steel gasket that is placed on top of the filter. Teflon filters do not use these gaskets. The gasket should be handled only using tweezers. 7. Next, the cone shaped part of the filter holder is placed on top of the loaded filter and the center ring is screwed tightly in place. 8. Loosely reconnect the top swagelock fitting. It will be tightened after all of the filters are loaded and flow rates have been measured. 9. Repeat steps 3-8 for each of the other 3 filter holders. 10. Leave empty petri dishes in the sampler to collect filters after the run. 11. Set the elapsed time meter to read 0. 12. Program the sampling event and record details like the FID numbers and flow rates on the log sheet as described below. 13. Record date, time of loading, and sign name as initial operator on the log sheet. 4.4.2 Flow measurement and programming a sampling event: 1. Before sample start time, turn the pump on briefly for flow measurements. The pump can be turned on by switching the Intermatic timer from “AUTO” to “MANUAL” mode (upper right corner). Then press the “ON/OFF” button to turn pump on and off. 2. The flows are measured at the top of the filter holders by attaching the rotameter to the swagelock fitting. Measure the filter holder flows one at a time and reconnect the swagelock fitting to the cyclone manifold after the flow is measured. B-40

3. Hold the rotameter vertically when taking a flow reading. There are two balls inside the glass tube on the rotameter (black and silver). Read the number nearest to the center of each ball and record the readings under Initial Flow Rates, Rotameter (balls) on the log sheet. Record both readings unless the top, black ball is off scale in which case the silver ball alone should be used. Also note the Rotameter ID# on the log sheet. 4. Compare the measured flow rates with the reference flow sheet. The rotameter values should be within ± 10% of the reference values. If flows are less than they should be, all fittings should be examined for leaks. It is also possible for critical flow orifices to clog with particles, and they may need to be cleaned or replaced. 5. Make sure all filter holder fittings are tightly connected to the cyclone. swagelock fittings should seal when finger tight.

The

6. Program the timer to start the sample event at 12:00am. Slide the switch in the upper right corner to “PROGRAM” mode. The switch next to it should be set to “OUTPUT”. Press the HOUR, MIN, and DAY buttons to program the first set point time desired (12:00am) and day of week. Day 1 is Monday, day 2 is Tuesday, etc. Press the ON/OFF button to select ON for the current day. Press ENTER and program the second set point to turn the sampler OFF at 12:00am on the following day. Slide the upper right corner switch to “AUTO” mode for program to run. (Additional detailed timer instructions are provided.) 7. If the instrument is operating correctly, close the cabinet door and leave the instrument to run overnight. If there are problems, refer to the trouble shooting section, or contact one of the people from the help list. 4.4.3

Unloading filter samples from Caltech samplers

Filter samples should be unloaded as soon as possible after the end of sampling. Samples should remain in the field for no more than 16 hours after the sample end time. Special exceptions will be made for situations when the sites cannot be accessed on the day after operation. Contact one of the people on the contact help list for permission to unload a sample outside the 16 hour window. 1. After sampling has ended, check the final flow rates before unloading samples. 2. Check flow rates using the external rotameter provided. The details for final flow rate checks are the same as in steps 1-3 above. Record black and silver ball readings on the log sheet under Final Flow Rates (balls). 3. Record any problems that occurred while sampling in the comments section. Continue comments on the back of the log sheet or on additional pages if necessary. Also, note what weather conditions were like on the day the samplers were run (i.e., windy, rain, etc.) 4. Record the elapsed time number on the log sheet under Duration. B-41

Reference Flow Sheet

Site: _________________ Calibration Date:______________ Calibrated with: Standard Dry Gas Meter (NIST Traceable-ID #C-0701) Technician signature: ______________________

Reference Values Filter holder

Rotameter ID#__________ black ball

Quartz 1

---------------------------------------------silver ball black ball

Teflon 2

---------------------------------------------silver ball black ball

Teflon 3

---------------------------------------------silver ball black ball

Quartz 4

---------------------------------------------silver ball

Figure 10: Reference Flow Sheet for Caltech Samplers

B-42

4.4.4

Steps for filter unloading

1. Set out a piece of clean aluminum foil on top of the sampler as a work surface. 2. Get items needed for filter unloading: tweezers, gloves, and 4 empty petri dishes. 3. Open the cabinet door, then put on a pair of clean disposable gloves. Do not touch anything dirty with gloves. If the gloves become dirty, throw them out and put on a new clean pair before continuing. 4. Beginning with the first filter holder. Disconnect the swagelock fitting at the top of the filter holder to make it easier to handle. Leave the bottom connected to the accordion style Teflon tubing. 5. Unscrew the center ring on the holder which can be removed, then lift off the top cone-shaped part of the filter holder. Be careful not to tip the bottom of the filter holder over while the top is off because the sample inside may fall out. Do not set anything down on a dirty surface, but use the foil work surface if necessary. 6. Using tweezers, remove a filter from the filter holder and place it in the correct petri dish. Touch the filter only with tweezers and only near the edges of the filter. 7. Reassemble the empty filter holder and reconnect it so that the interior will remain clean between sampling events. 8. Repeat steps 4-7 to unload each of the other 3 filter holders. 9. Wrap Teflon tape around the edge of each petri dish, pulling it tight to seal. Make sure to hold the dish upright while doing this. 10. Transport samples to the local office in a cooler with “blue ice” and store frozen until time for sample shipment at the end of the month.

4.5

Trouble shooting

During the course of the PRD project, operators may encounter a variety of problems. Some possible problems that may occur and recommendations for solving them are below. 4.5.1

Spare parts

If there are problems with either the filter cassettes or the filter holders, there are spares in the orange toolbox. There are five spare filter cassettes (Andersen samplers), a spare filter holder, and spare filter internal screens (Caltech samplers). When loading, visually inspect the filter cassettes to make sure the metal screens are not bent or misshapen - replace with a spare if there is even any minor damage.

B-43

4.5.2

Problems that may occur during sample loading/initial check

1. Sampler will not turn on (Andersen) (a) Operator should first check to make sure the sampler is plugged in and that there is no damage to the electrical cord. If the sampler is plugged in and the problem remains, the operator should check whether the problem is with the sampler or with the power source - either by using the pump/electronics by-pass cord to plug the pump-box into the outlet or by checking the outlet with another electrical device. (b) If the pump/electronics by-pass cord is able to turn on the pump, this means there is a problem with the electronics portion of the sampler and the sampler must be run in manual mode (see MANUAL MODE section) and people on the Contact Help List should be notified. (c) If there is simply no power to the site at all, operator should get in touch with whoever is in charge of power to the site. If power is not restored until after normal start time, samples should be loaded for a 24-hour sample as soon as possible after power is regained (a sample begun late does not have to start at midnight but should run for a full 24 hour period). 2. Timer not working (Caltech) - For Caltech samplers: If there is a problem with the timer in the sampler, samples can be taken in manual operation mode. For manual operation, disconnect the timer and plug the power strip directly into the wall outlet. In this mode, the sampler can be turned on and off by plugging/unplugging the power to the pump. 3. The electronics are malfunctioning (a) If the operator arrives at the site and finds the electronics have failed so that the operator cannot program an event or run initial flow checks, operator should run the sampler in MANUAL MODE or use the ”flow valves” command (used for initial flow checks) to manually turn the pump on for sampling. (b) For all sampling done in manual mode, operators may start the 24-hour sample during their normal loading time, instead of the normal midnight starting time. Unloading may be done 24 hours later, during what would have been the ”middle check” period. 4. MANUAL MODE (a) Operator should first check to see if it is possible to use the ”flow valves” command to start the pump. If this is possible, use ”flow valves” instead of the pump/electronics by-pass cord and proceed to step (c). Manual mode for Caltech samplers should begin with step (c). B-44

(b) Find the pump/electronics by-pass cord - either located within the sampler or in the orange toolbox. Unplug the interior power cord and plug the pump/electronics by-pass cord into the interior outlet. (c) Load all filters, following normal filter-loading procedures. (d) Once filters are loaded into the filter holders, begin sampling by plugging in the pump/electronics by-pass cord (or press ”flow valves” to begin Andersen pump if possible). For Caltech samplers plug in the power strip by-passing the timer. Note the time on the Daily Event Log Sheet. (e) Check flow rates using a rotameter, following the normal rotameter flow check process. (f) 24-hours later, return to the site. (g) Check flow rates using the rotameter. (h) Note the time and unplug the pump/electronics by-pass cord (or hit ”cancel” to exit ”flow valves” to stop the pump), ending the 24-hour sampling run. (i) Unload the filters as normal. (j) Take note of problems on the Daily Event Log Sheet and notify Contact Help List people. 5. Initial flow rates are too high/too low. (a) Case 1: The rotameter is reading flow rates that are in normal range, the Andersen readings are too high or too low: Operator should run sampling as normal and should notify Contact Help List people as soon as possible. (b) Case 2: The rotameter is reading flow rates that are too high/too low, the Andersen readings are in normal range: Operator should inspect rotameter to make sure the exhaust port is not blocked. If the rotameter has a valve, the operator should make sure the valve is open. If there is still a problem, operator should run the sampler as normal and notify Contact Help List people as soon as possible. (c) Case 3: Both the Andersen and rotameter flow checks read flows that are too high/too low: Operator should check to make sure the plastic accordion tubes are not pinched or blocked by an object, that the filters were loaded correctly, and that the filter holders were correctly closed and tightened. If the problem remains, call someone from the Contact Help List immediately for instructions. If operator is unable to get in touch with anyone from the Contact Help List, operator should unload all filters and reseal petri dishes with Teflon tape. Filters may be reused for a make-up sample. 6. The site is inaccessible due to dangerous conditions. - Operator should not go to site and should notify Contact Help List people.

B-45

7. Filters are damaged during loading. - If a filter is damaged by holes or tears which would prevent a proper seal in the filter cassettes, dirty from being dropped, or has become wet, a spare should be used in its place. When using a spare, be sure to put the FID of the spare filter on the log sheet in the place of the original FID. On the spare filter petri dish, change the label to match the label of the filter that is being replaced (for example, change from ”spare quartz filter” to ”TC2-Q1-021002”). On the petri dish of the damaged filter, write ”BAD” and keep the petri dish with the rest of the petri dishes for this event. If the operator has run out of spares, the operator should use a filter from a later event date. The operator should first use the field blanks for that month, followed by the filters for the latest date. If the operator has run out of spares and needs to use filters from later planned events, notify Lynn Salmon immediately so that replacements can be prepared and sent. 4.5.3

Problems that may occur during sample middle check

1. Sampler is not running, electronics appear to be fine. (a) Operator should first check for incorrect programming by going to ”View Event” and checking the date that was programmed. If there is an error in programming, operator should program a new event to begin a 24-hour run, starting immediately. Operator should delete the old event that had been incorrectly programmed by going to ”Events”, ”Delete Event”, selecting the event, and pressing ”Enter”. After the replacement event has begun, operator should complete the middle flow checks and return 24 hours later to unload the sample as normal. Operator should take note of problem in log sheet and notify contact people. (b) If there is no error in programming, operator should visually inspect the filters. If the filters are clean and blank - operator should switch to run the sampler in manual mode, (see MANUAL MODE). If filters are dirty - operator should unload filters following normal unloading and Datalink procedures, note problem in log sheet, and notify contact people. If possible, include the Datalink file with the notification email. 2. Operator finds electronics reading an error message or screen is blank. (a) If screen is blank, operator should rotate ”contrast” knob to make sure screen is at the proper contrast level. (knob is located directly to the right of the screen) (b) If error message/blank screen remains, operator should visually inspect the filters. If the filters are clean and blank - operator should switch to run the sampler in manual mode, (see MANUAL MODE). If filters are dirty - operator should unload filters following normal unloading procedures, note problem in log sheet, and notify contact people.

B-46

3. Flow rates are too high/too low - Operator should continue sampling as normal, note problem on Daily Event Log Sheet. 4. Power is out at the site - Operator should make note of the problem on Daily Event Log Sheet and should return the next day to unload the samples. If possible, operator should contact the person in charge of power supply to investigate the problem. Operator should notify contact people of the power problem as soon as possible. 4.5.4

Problems that may occur during sample unloading

1. Sampler has no power. (a) Operator should first unload all samples, following normal unloading procedures. (b) Operator should next do a brief check of the possible source of power failure, checking to see if the sampler is unplugged, if there is damage to the power cord, if there is information about power failure to the entire building. If possible, operator should use the pump/electronics by-pass cord to plug in the pump directly, checking to see if the problem is solely in the electronics. After doing a brief check, operator should notify contact people for troubleshooting help. (c) Operator should plan on returning at a later point to download data with the Datalink once power is resumed. 2. Datalink is malfunctioning (a) Unload samples following normal unloading procedures. (b) If Datalink is showing ”memory full” signal, the Datalink needs to be taken to a computer with the winDatalink program to have its memory cleared. (c) If Datalink is showing fatal/non-fatal errors, or if the data transfer function is not working, notify contact people for help. 3. Filters are damaged during unloading. - If damage occurs to filters during unloading, operator should unload as normal and describe damage on the Daily Event Log Sheet. Also, the operator should notify the contact people, describing the damage that occurred.

B-47

4.5.5

Reset Calibration (Andersen sampler)

If one of the Andersen samplers loses the stored calibration for any reason, it can be reset as follows. To reset calibration: 1. Turn on sampler 2. Go to “MAINTENANCE>”, choose “Yes” to enter. 3. Go to “Archive values>” 4. Go to “Restore Cal Values” and hit the “Enter” button. 5. Exit by hitting the “Cancel” button until you get to the main menu. To check to see if it works: 1. Go to “MAINTENANCE>”, choose “Yes” to enter. 2. Go to “Calibrate>”. 3. Go to “Flow 1>” 4. Press the “Cancel” button to proceed. 5. Select “Dry Gas Meter” and hit the “Enter” button. 6. Hit the “Enter” button to accept the Reference DGM Coefficient 7. You will now be at the calibration screen for Flow 1. Toward the lower-middle of the screen you will see the following:

OLD: NEW:

GAIN XXX (number) XXX

OFFSET XXX XXX

8. Write down the values for the gain and offset for Flow 1. 9. Repeat 1-8 for Flow 2, 3, 4. If there are any problems that occur and are not mentioned here, please notify the people on the Contact Help List.

B-48

4.5.6

Make-up sampling event

If problems prevented a sample to be obtained during the normal course of sampling, an additional “make-up” sample may be taken. For example, if a power failure or programming error caused an instrument to not sample, a “make-up” sample may be scheduled for the next day at that site only. If more serious problems, like a typhoon, affect multiple sites, a “make-up” sample may be scheduled for all sites later in the month. Contact one of the people on the contact help list to determine if a “make-up” sample is required and when to schedule it.

5

CONTACT HELP LIST

The Pilot Study is being conducted by Georgia Institute of Technology, Hong Kong University of Science & Technology, Peking University, and the California Institute of Technology. People to contact concerning the project are listed below.

5.1 For Field Operation or Sample Questions Contact: Lynn Salmon, Research Engineer Environmental Quality Laboratory California Institute of Technology 1200 E. California Blvd. (mail code 138-78) Pasadena, CA 91125 email: [email protected] or [email protected] Tel: 1(626)395-6420 fax: 1(626)395-2940 Gayle Willis, Graduate Student Georgia Tech student spending fall semester at HKUST email: [email protected] Tel: (852)2719-1352 HKUST Ph.D. students: Wan Chun Hong, Eric email: [email protected] Tel: 97846708

Wu Wai Shing email: [email protected] Tel: 91618958

Yang Qiang, Peking Univ. student email:[email protected] Tel: (86) 136 6117-6858

Zeng Limin, Research Engineer email: [email protected]

B-49

5.2 Monitoring Stations There are seven air monitoring stations in Hong Kong and the Pearl River Delta region of China. See Figure 1. • TM1 - Tap Mun - located on roof of police station on Tap Mun Island, Hong Kong. Lat: N 22o 28.277’ Lon: E 114o 21.642 • TC2 - Tung Chung - located on roof of shopping mall near Tung Chung MTR station, Hong Kong. Lat: N 22o 17.334’ Lon: E 113o 56.620 • CW3 - Central and Western - located on roof of Crime Wing Hong Kong Island Regional Headquarters, High Street. Lat: N 22o 17.122’ Lon: E 114o 08.713 • SZ4 - Shenzhen - located on roof of Peking University Shenzhen campus. Lat: N 22o 32.183’ Lon: E 113o 56.340 • GZ5 - Guangzhou - located on roof of Guangzhou EPB building. Lat: N 23o 07.969’ Lon: E 113o 15.585 • CH6 - Conghua - located on roof of Conghua Jing Xiu Hotel. Lat: N 23o 38.530’ Lon: E 113o 38.345 • ZS7 - Zhongshan - located on roof of building in amusement park. Lat: N 22o 30.660’ Lon: E 113o 24.429

5.3 Principal Investigators: Dr. Mike Bergin, Assistant Professor School of Civil and Environmental Engineering and School of Earth and Atmospheric Sciences Georgia Institute of Technology 200 Bobby Dodd Way Atlanta, GA, 30332-0512 email: [email protected] Tel: 404-894-9723 fax: 404-894-8266 Dr. Jian Yu, Assistant Professor Department of Chemistry Hong Kong University of Science & Technology Clear Water Bay, Kowloon, Hong Kong email: [email protected] Tel: 852-2358-7389, fax: 852-2358-1594

B-50

Professor Zhang Yuanhang, Director Center for Environmental Sciences Peking University Beijing 100871, China email: [email protected] Tel: 86-10-62756592 fax: 86-10-62751927

6

SAMPLE TRANSPORTATION AND SHIPPING

As previously discussed, all samples will be immediately put into coolers with blue ice each time samples are collected in the field. In Hong Kong, the coolers will be delivered to the HKEPD office in Wan Chai on the same day samples were collected. Operators will also bring the DataLink and log sheets to the HKEPD office. The DataLink information will be downloaded into a HKEPD computer and emailed to [email protected] and to [email protected] in a timely manner to allow review before the next sampling event. The HKEPD office will fax (or email scans) daily event log sheets to Lynn Salmon at 1(626)395-2940 after each sampling event. The original log sheet will be stored at HKEPD until the end of the month and forwarded to Jian Yu with the samples. Copies of the log sheets will be kept in the field notebooks. At HKEPD the samples will be stored under frozen (−4o C or lower) conditions. At the end of each month, Dr. Jian Yu will collect the filters from the HKEPD, remove the filters that will be analyzed at HKUST, and ship the remaining filters to Lynn Salmon at Caltech within 10 days. In the PRD, the samples will be delivered to the pertinent local office and will be placed in a freezer. Samples will be transported from the field to the local office in a cooler on the same day of sample collection. Operators will also bring the daily log sheets to the local office. Daily log sheets should be faxed to Lynn Salmon at 1(626)395-2940 after each sampling event. The original log sheet will be stored at the office until the end of the month and forwarded to Jian Yu with the samples. Copies of the log sheets will be kept in the field notebooks. At the end of each month, samples from each PRD site will be delivered (frozen) to the Guangzhou sampling site. The samples will be collected by a student from Prof. Jian Yu’s group at HKUST within one week of being received in Guangzhou. Dr. Yu will sort and ship the remaining samples to Caltech with the Hong Kong samples. Shipments will be made in insulated mailing boxes with blue ice packets. Fedex, or other fast delivery will be used.

B-51

7

SAMPLING SCHEDULE

7.1 Sample Dates Samples will be collected during four seasons beginning in October, 2002. Subsequent sampling will occur during December, 2002, March, 2003 and June, 2003. During each month there will be five 24-hour samples and one set of field blanks collected. Sampling dates will be (see calendar): October, 2002 Sampling Dates: 2, 8, 14, 20, 26 December, 2002 Sampling Dates: 1, 7, 13, 19, 25 March, 2003 Sampling Dates: 1, 7, 13, 19, 25 June, 2003 Sampling Dates: 5, 11, 17, 23, 29 Samples will be loaded on the day before sampling. Timers will be used to start the samplers and turn them off after 24 hours. After samples finish, operators will unload samples and place them immediately in a cooler with blue ice. Special arrangements will be made for situations when the sites cannot be accessed on the day before or after operation. Please check with one of the contact help list people for permission to load or unload a sample outside the normal time frame.

7.2 Calibrating Samplers Sampler flow rates will be calibrated using a dry gas meter prior to the start of sampling each month. Sampler calibrations will be verified by operators during each sampling event. It should only be necessary to recalibrate the instruments if problems arise and the flows deviate from the reference flow sheet. If there are any questions about the calibration, operators should contact one of the people on the contact help list and arrangements for sampler calibration/verification will be made. The following apparatus and materials are used to perform sampler calibration.

ITEM Dry Gas Meter Barometer

SUPPLIER

MODEL or CATALOG NO.

Schlumberger

Gallus 1000, No. 3600046

Oakton

s/n 51062

Thermometer

s/n 9266

Baked Quartz filters

Pall Gelman

2500QAO-UP

47mm Teflon filter

Pall Gelman

R2PJ047

Andersen

none

Flow check adapter

B-52

August Sunday July

2002

Monday 2002 September

Tuesday

Wednesday

2002

1 2 3 4 5 6 1 2 3 4 5 6 7 7 8 9 10 11 12 13 8 9 10 11 12 13 14 14 15 16 17 18 19 20 15 16 17 18 19 20 21 21 22 23 24 25 26 27 22 23 24 25 26 27 28 28 29 30 31 29 30

Thursday

Friday

December

Saturday

Sunday

Monday

2002 Tuesday

Wednesday

Thursday

Friday

April

Saturday

1

2

3

1

2

3

4

5

6

7

Sunday

2003 Monday

Tuesday

Wednesday

Thursday

Friday

Saturday

1

2

3

4

5

4

5

6

7

8

9

10

8

9

10

11

12

13

14

6

7

8

9

10

11

12

11

12

13

14

15

16

17

15

16

17

18

19

20

21

13

14

15

16

17

18

19

18

19

20

21

22

23

24

22

23

24

25

26

27

28

20

21

22

23

24

25

25

26

27

28

29

30

31

29

30

31

27

28

29

30

2002

January

September Sunday

Monday

Tuesday

Wednesday

Thursday

Friday

Saturday

Sunday

2

3

4

5

6

7

8

9

10

11

12

13

14

5

6

15

16

17

18

19

20

21

12

22

23

24

25

26

27

28

19

29

30

26

August

Sunday

1 2 3 1 2 3 4 5 4 5 6 7 8 9 10 6 7 8 9 10 11 12 11 12 13 14 15 16 17 13 14 15 16 17 18 19 18 19 20 21 22 23 24 20 21 22 23 24 25 26 25 26 27 28 29 30 31 27 28 29 30 31

2002

Monday 2002 November

2002

Tuesday 2002

1 2 3 4 5 6 7 1 2 8 9 10 11 12 13 14 3 4 5 6 7 8 9 15 16 17 18 19 20 21 10 11 12 13 14 15 16 22 23 24 25 26 27 28 17 18 19 20 21 22 23 29 30 24 25 26 27 28 29 30

Wednesday

Thursday

Friday

2

3

4

5

1 2 1 2 3 4 3 4 5 6 7 8 9 5 6 7 8 9 10 11 10 11 12 13 14 15 16 12 13 14 15 16 17 18 17 18 19 20 21 22 23 19 20 21 22 23 24 25 24 25 26 27 28 29 30 26 27 28 29 30 31

Tuesday

Wednesday

Thursday

Friday

Monday

Tuesday

4

7

8

9

10

11

4

5

6

13

14

15

16

17

18

11

12

20

21

22

23

24

25

18

19

27

28

29

30

31

25

26

Tuesday

Wednesday

Thursday

Friday

2003

1 2 3 4 1 5 6 7 8 9 10 11 2 3 4 5 6 7 8 12 13 14 15 16 17 18 9 10 11 12 13 14 15 19 20 21 22 23 24 25 16 17 18 19 20 21 22 26 27 28 29 30 31 23 24 25 26 27 28 29 30 31

2003 June

Thursday

Friday

Saturday

1

2

3

7

8

9

10

13

14

15

16

17

20

21

22

23

24

27

28

29

30

31

June

Saturday

2003

1 1 2 3 2 3 4 5 6 7 8 4 5 6 7 8 9 10 9 10 11 12 13 14 15 11 12 13 14 15 16 17 16 17 18 19 20 21 22 18 19 20 21 22 23 24 23 24 25 26 27 28 29 25 26 27 28 29 30 31 30 31

Wednesday

2003

1 2 3 4 5 1 2 3 4 5 6 7 6 7 8 9 10 11 12 8 9 10 11 12 13 14 13 14 15 16 17 18 19 15 16 17 18 19 20 21 20 21 22 23 24 25 26 22 23 24 25 26 27 28 27 28 29 30 29 30

2003

Monday 2003 March

April

26 2003 May

2003

Sunday

3

Sunday January

March

May

Saturday

2

2003

February

Saturday

1

2003

1

2002 February

1 2 3 4 5 6 7 1 8 9 10 11 12 13 14 2 3 4 5 6 7 8 15 16 17 18 19 20 21 9 10 11 12 13 14 15 22 23 24 25 26 27 28 16 17 18 19 20 21 22 29 30 31 23 24 25 26 27 28

2002 October

October September

December

2002 January

2003

Monday

1

November

2003

Sunday

Monday

Tuesday

Wednesday

Thursday

Friday

Saturday

1

1

2

3

4

5

6

7

6

7

8

9

10

11

12

2

3

4

5

6

7

8

8

9

10

11

12

13

14

13

14

15

16

17

18

19

9

10

11

12

13

14

15

15

16

17

18

19

20

21

20

21

22

23

24

25

26

16

17

18

19

20

21

22

22

23

24

25

26

27

28

29

30

31

23

24

25

26

27

28

29

30

November Sunday October

Monday 2002 December

2002 Tuesday

Wednesday

Thursday

2002

1 2 3 4 5 1 2 3 4 5 6 7 6 7 8 9 10 11 12 8 9 10 11 12 13 14 13 14 15 16 17 18 19 15 16 17 18 19 20 21 20 21 22 23 24 25 26 22 23 24 25 26 27 28 27 28 29 30 31 29 30 31

Friday

March

Saturday

Sunday

1

2

February

2003

Monday 2003 April

Tuesday

Wednesday

Thursday

Friday

2003

1 2 3 1 2 3 4 5 4 5 6 7 8 9 10 6 7 8 9 10 11 12 11 12 13 14 15 16 17 13 14 15 16 17 18 19 18 19 20 21 22 23 24 20 21 22 23 24 25 26 25 26 27 28 29 30 31 27 28 29 30 31

2003

Sunday June

28 2003 July

July

Saturday

1

2003

1 1 2 3 4 5 2 3 4 5 6 7 8 6 7 8 9 10 11 12 9 10 11 12 13 14 15 13 14 15 16 17 18 19 16 17 18 19 20 21 22 20 21 22 23 24 25 26 23 24 25 26 27 28 27 28 29 30

27 May

Monday 2003 August

Tuesday 2003

1 2 3 4 5 6 7 1 2 8 9 10 11 12 13 14 3 4 5 6 7 8 9 15 16 17 18 19 20 21 10 11 12 13 14 15 16 22 23 24 25 26 27 28 17 18 19 20 21 22 23 29 30 24 25 26 27 28 29 30 31

Wednesday

Thursday

Friday

Saturday

1

2

3

4

5

3

4

5

6

7

8

9

2

3

4

5

6

7

8

6

7

8

9

10

11

12

10

11

12

13

14

15

16

9

10

11

12

13

14

15

13

14

15

16

17

18

19

17

18

19

20

21

22

23

16

17

18

19

20

21

22

20

21

22

23

24

25

26

30

23/30

24/31

25

26

27

28

29

27

28

29

30

31

24

25

26

27

28

29

Hong Kong and Pearl River Delta Sampling Schedule October, 2002 Sampling Dates: 2, 8, 14, 20, 26, 0210FB December, 2002 Sampling Dates: 1, 7, 13, 19, 25, 0212FB March, 2003 Sampling Dates: 1, 7, 13, 19, 25, 0303FB June, 2003 Sampling Dates: 5, 11, 17, 23, 29, 0306FB ** September 19, 2002: Test Run all sites Figure 11: Sampling Schedule

B-53

7.2.1

Calibration/verification of Andersen samplers

According to the instrument manual (Thermo Andersen, 2001), the sampler needs to be calibrated every time it is moved to a new location. The frequency of calibration and verification will be user determined. Andersen recommends that the sampler should be verified or calibrated within 90 days of the previous calibration. Temperature verification: Sampler thermocouples were calibrated by the manufacturer, refer to the instrument manual for details (Thermo Andersen, 2001). A thermometer, which reads in degrees C, NIST traceable, and accurate to ±2C is used to verify that the sampler thermocouples are accurately measuring the ambient and manifold temperatures. 1. Hold reference thermometer close to the ambient thermocouple, located on the outside right side of the sampler under a white solar radiation shield. 2. On the sampler, go to “MAINTENANCE>”, press enter. 3. Select “QC>”, press enter. 4. Select “Ambient”, press enter. 5. The thermocouple temperature output is displayed on the screen. When the reference thermometer has come to rest on a value, press “Enter” and input the reference value when prompted. 6. The screen will display the difference between the reference and the sampler temperatures - this value should nominally be smaller than 2 degrees Celsius. 7. Press “Cancel” to exit the ambient temperature verification, select “Yes” to save the verification. 8. Repeat 1-7 for the manifold temperature, selecting “Manifold” from the “QC>” list and holding the reference thermometer near the manifold thermocouple (located on the upper right inside the sampler). Pressure verification: Sampler pressure was calibrated by the manufacturer. A pressure measurement device with a range of at least 500 to 900 mm Hg absolute is used to verify the accuracy of the sampler pressure sensor. The device should have a readability of 1 mm Hg and an accuracy of ±5mmHg. 1. On the sampler, go to “MAINTENANCE>”, press enter. 2. Select “QC>”, press enter. 3. Select “Barometer”, press enter. B-54

4. The sampler pressure reading is displayed on the screen. Press “Enter” to input the reference pressure reading. 5. The sampler will show the difference between the reference and the sampler. This value should nominally be smaller than 10 mm Hg. 6. Press “Cancel” to exit the pressure verification, select “Yes” to save the verification. Flow calibration: A NIST traceable reference dry gas meter is used to calibrate the flow sensors in the Andersen sampler. The dry gas meter should read in cubic meters and have a readability of at least 0.0002 cubic meters. 1. Load filters in cassettes into the four filter holders; quartz fiber filters in Q1 and Q4, teflon filters in T2 and T3. These filters should be the same type as the filters used during sampling. 2. Place the dry gas meter in a stable location and do not move it during calibration. 3. Disconnect the Q1 filter holder by unscrewing the white hand ring located above the filter holder. Attach the filter holder to the flow check adapter and attach the flow check adapter to the dry gas meter with accordion style teflon tubing. The teflon tubing has swagelocks on each end which will give a proper seal when finger-tightened. The Q1 filter holder may be laid down within the sampler, being sure not to block any fans or cause tight bending in the teflon tubing. 4. On the sampler computer, go to “MAINTENANCE>”. Select “Calibrate>”. Select “Flow 1”. 5. The screen displays current outputs of the pressure and temperature sensors, verify that the sensors are working properly by comparing them with a reference. Press “Cancel” to proceed. 6. Under “Select Flow Device”, choose “Dry Gas Meter” and press enter. The system will now ask for a Reference DGM Coefficient. This number is located on a label on the dry gas meter. Input this number into the computer and press enter. 7. The calibration screen for Flow 1 is displayed. Write down the old “gain” and “offset” in the field logbook. Next, press “Enter” to add a point. 8. The sampler will alternate measuring two types of points - zero flow points and full flow points. The first point will be a zero flow point which will take approximately one minute. 9. After the zero point finishes, press “Enter” to add another point. The system will now ask for the initial reading on the dry gas meter. Look on the dial of the dry gas meter and key this number into the computer, also write it down in the field logbook as Vo,DGM. After keying in the number, press enter. B-55

10. The pump will come on the system will display a screen showing the flow rate, flow variation (CV%), etc. Once the Volume parameter is equal or greater than 0.1000, a prompt at the bottom of the screen appears - “Hit Cancel to Enter Reference Data”. After this prompt appears, press enter. Write down the time in the field logbook. 11. The screen will prompt for the final volume reading on the dry gas meter. Write down the number on the screen in the field logbook as Vf,sampler. Write down the final dry gas meter value in the field logbook as Vf,DGM. Key in the final dry gas meter value, replacing the value on the screen. Press enter. 12. Repeat steps 8-11, taking two more zero points and two more flow points - totaling 6 calibration points for Flow 1. At the end, write down the new “gain” and offset” in the field notebook. 13. Press “Cancel” to exit calibration for Flow 1. Choose “Yes” to save the calibration. 14. Repeat steps 3-13 for Flow 2 (T2), Flow 3 (T3), and Flow 4 (Q4). 15. At the end of calibration, go to “MAINTENANCE>”, select “Archive Values”, and select “Archive Cal Values”. Press enter, choose “Yes” to overwrite past calibration values. This will save the updated calibrations which may be restored if the sampler loses the stored calibrations. 16. Calculate the flow rate in L/min by (Vf, DGM − Vo, DGM) × coefficient × 1000 time Take an average of the three results for each filter holder to obtain the calibrated orifice flow rate. 7.2.2

Calibrating Caltech samplers

A NIST traceable reference dry gas meter is used to calibrate the critical orifices used for flow control in the Caltech samplers. The dry gas meter should read in cubic meters and have a readability of at least 0.0002 cubic meters. 1. Load unused filters into the four filter holders; quartz fiber filters in Q1 and Q4, teflon filters in T2 and T3. These filters should be the same type as the filters used during sampling. 2. Place the dry gas meter in a stable location and do not move it during calibration. 3. Disconnect the Q1 filter holder by unscrewing the top swagelock fitting. Attach the filter holder to the dry gas meter with accordion style teflon tubing. 4. Write down the initial reading on the dry gas meter in a field notebook. Call this value Vo,DGM. B-56

5. Simultaneously start power to the pump and start timing with a stop watch. Run the sampler pump for several minutes until approximately 0.1000 m3 of air has been drawn through the filter holder. 6. Simultaneously stop the pump and stop the stop watch. 7. Write down the elapsed time and the final reading on the dry gas meter, Vf,DGM in the field logbook. Also note the reference DGM coefficient value from the dry gas meter label. 8. Calculate the flow rate in L/min by (Vf, DGM − Vo, DGM) × coefficient × 1000 time 9. Repeat the flow calibration measurement and take an average of the results to obtain the calibrated orifice flow rate. 10. Follow the same steps for filter holder T2, T3, and Q4.

7.3

Sampler Maintenance

Samplers were thoroughly cleaned with hexane and methanol before being brought to the field. It should not be necessary to disassemble and clean the equipment during the course of the year sampling study which will have 21 sampling days. Andersen recommends cleaning after 30 sampling days. However, care should be taken to keep the filter holders closed and all sampling lines connected between sampling months so that the parts remain as clean as possible. At the start of each sampling month, the large particle collection cup at the bottom of each cyclone should be emptied. This cup unscrews and any large particles collected can be discarded. There are two cyclones per sampler. Filter cassettes should be inspected for cleanliness before each sample load. Spares are provided should any be found dirty. Dirty cassettes can be rinsed in methanol, allowed to dry on clean kimwipes, and then returned to the tool box as spares. The filter cassette loader should be inspect for cleanliness before each trip to the field. It can be wiped with a clean kimwipe, or rinsed in methanol should it become soiled.

8 8.1

ORGANIC AND ELEMENTAL CARBON ANALYSIS Apparatus and Materials

The following apparatus and materials are used to analyze quartz fiber filters for organic (OC) and elemental (EC) carbon. B-57

ITEM

SUPPLIER

MODEL or CATALOG NO.

Sunset Laboratory

May, 1997

Ted Pella

5734

VWR

51147

Pall Gelman

2500QAO-UP

Petri dishes

VWR

25388-606

Sucrose, ACS reagent

Sigma

S5016

Hamilton

81265

Carbon Analyzer Tweezers style AA Gelman s.s. forceps Quartz filters

Syringe, 25 µl

8.2 Analysis Description Quartz fiber filters are analyzed for OC and EC content by the thermal evolution and combustion technique of Birch and Cary (1996). In the thermal-optical organic and elemental carbon concentration measurement method, 1.5 cm2 rectangular filter punches are first placed in a temperature and atmosphere-controlled oven. The analysis process occurs in two stages. In the first stage, the temperature is raised progressively to approximately 850o C in a pure helium atmosphere in order to volatilize organic carbon from the sample. The volatilized carbon is oxidized catalytically to CO2 , which is then reduced to CH4 . The CH4 is then quantified using a flame ionization detector (FID). During this process, the filter may darken somewhat due to production of artifact EC by pyrolysis. Elemental carbon measurement and correction for any artifact EC formed by pyrolysis during OC analysis are accomplished during stage two. At the beginning of this second stage of analysis, the oven temperature is initially reduced, a 4% oxygen, 96% helium mixture is introduced, and the temperature is then raised progressively to approximately 900o C. As the black elemental carbon is oxidized and CO2 is evolved, there is an increase in filter transmittance. In order to correct for any contribution to the elemental carbon present due to OC pyrolysis during stage 1 of the analysis, the amount of EC oxidation necessary to return the filter to its initial transmittance is first measured. The point at which the filter transmittance equals its initial value is taken to be the point at which the actual EC begins to evolve from the sample. Any EC oxidized prior to that point is assumed to be due to OC pyrolysis earlier in the analysis and thus is added to the OC concentration measured in the first step of this analysis. This analysis method is known as the NIOSH (National Institute of Occupational Safety and Health) protocol.

8.3 Daily Run 1. The carbon analyzer is generally kept in standby mode when not in use. A Windows-based personal computer is dedicated to instrument operation. If the B-58

instrument is not in standby mode, click the “ocecin55.exe” icon on the computer. Otherwise, begin operation by clicking on the continue option to leave standby mode. The status line on the computer screen will change from red to green when the instrument has warmed up and is ready for operation. 2. Check the parameter file being used by the program is correct, and change the output raw data file name following the convention of 3-4 letter project code followed by 4 digit current date. A maximum of 8 characters can be used for the saved file name which puts data into filename.txt. 3. Set analyzer and adjust gas flows according to manufacturer’s recommended settings which can be found in the OC/EC Analysis Log Book. 4. Light the flame in the FID and verify that all gas flows are in the correct ranges. 5. Pull down the computer menu under “Run” and choose “CLEAN OVEN”. This will do an 8 minute heating cycle to 800o C in the front sampling oven which will burn off any contamination. 6. When the oven cools down, “SAFE to put in a new sample” will appear on the status line of the computer screen and analysis may proceed. 7. Run at least one quality control standard to ensure that instrument calibration is in control. Quality control standards can be prepared at the beginning of the day, or made up to a week in advance and stored in the freezer. 8. If the quality control standard(s) looks fine and FID1 MAX is within proper range (1550-1950), proceed with running samples. If FID1 MAX is too high/too low, adjust gas flow settings and run additional quality control standards or blank filters until FID1 MAX is within range. To run a sample 1. Remove the teflon tape sealing a petri dish containing a 47mm quartz fiber filter. Filter samples are kept in the freezer until time for analysis. To avoid frequent opening of the freezer, remove 5-6 samples at one time and set them aside on the lab bench. Keep samples sealed with teflon tape until one is needed. 2. Using the punch tool, take a 1.5 cm2 punch from the sampled area of the filter. Leave the filter in the foil lined petri dish and do not touch the remainder of the sample that will be used in GC/MS analyses. Close and tape the petri dish with new teflon tape. Return the remainder of analyzed samples to the freezer when the next set of samples is retrieved for analysis. 3. Open the sample insertion port of the carbon analyzer, and using the curved tweezers that are dedicated for this purpose, pull out the quartz sample holder boat. Be very careful handling the sample holder boat which is very fragile. If it comes in contact with anything other than the clean tweezers, it can be returned to the oven and “CLEAN OVEN” run again to remove contaminants. B-59

4. Remove the 1.5 cm2 rectangular piece of quartz filter from the punch tool using the tweezers that have been dedicated for use with the carbon analyzer. Tweezers are solvent cleaned and wiped with a clean kimwipe. Take care not to disturb deposited material and avoid hand contact with the sample. 5. Place the sample punch on the quartz sample holder boat. There is a ridge on the sample holder boat indicating where the punch should be placed so that it lines up with the laser during analysis. 6. Push the sample holder boat back into position with the curved tweezers and close the sample insertion port. 7. Choose “Start Analysis” on the computer. 8. Each sample will take approximately 20 minutes for a complete run including a calibration injection at the end and a cool down phase. The computer screen will display “Safe to put in a new sample” when the next sample can be inserted. After the days run is complete 1. Put the carbon analyzer program into standby mode. 2. Turn all gas flows to zero except for helium. A small helium flow (approx. 20) should be kept running through the system while in standby mode. Turning hydrogen flow to zero will put out the flame in the FID. 3. A program to calculate carbon concentrations can be run by clicking on the “Calc60.exe” icon. Organic and elemental carbon are reported in terms of µg per cm2 of filter area. The total OC and EC on the filter are calculated by multiplying the reported values by the deposit area. A homogeneous sample deposit is assumed. 4. The calculation program will write calculated data into the file filename.csv which should be backed up to another computer storage medium.

8.4 8.4.1

Quality Control Instrument specifications

See NIOSH (1999) or Birch and Cary (1996) for additional details. • A calibration peak is achieved by methane injection included at the end of each sample run. Operator will verify the FID MAX for each sample run is in the correct range according to manufacturer’s recommendations. If any problems occurred during a sample run, a new punch will be taken and the sample rerun. • The estimated detection limit is 0.15 µg/cm2 C. B-60

• The ideal range for sample analysis is 1 to 105 µg per filter portion analyzed. For heavily loaded samples, a smaller sample punch tool can be used reduce the carbon load analyzed. • Instrument precision has been found to be 0.19 at 1 µg C and 0.01 at 10-72 µg C. • The split between organic-based carbon (OC) and EC may be inaccurate if the sample transmittance is too low. In general, the OC-EC split may be inaccurate when EC loadings are above 20 µg/cm2 of filter. The split should be reassigned manually (prior to EC peak) in such cases. 8.4.2

Quality control standards

1. Prepare clean filter punches. Use the sample punch tool to cut several 1.5 cm2 rectangles. These can be stored until needed in the box labelled “punches for standards, unbaked.” 2. Place 2-3 rectangular punches on the quartz sample boat and insert them in the sample oven. Run “CLEAN OVEN” to bake them. 3. Remove clean, baked punches and using tweezers, carefully place them on the pin rack used to hold them for spiking. 4. With a 25 µl syringe, apply aliquots of OC standard solution directly to the clean filter punches. Quality control standards are made containing 5 µl, 10 µl, 15 µl, and 20 µl of sucrose solution. 5. Sucrose (C12 H22 O11 ) solution is prepared by putting 5.848 g/l of sucrose yielding 2.46 µg/µl “C”. 6. Allow water to evaporate from punches. After drying, either run punches for analysis, or transfer them to petri dishes lined with annealed aluminum foil, seal with teflon tape, and store in the freezer for use later in the week. 8.4.3

Freezer storage on samples

The effect of long freezer storage time on OC/EC results has been investigated by our laboratory. Samples were originally analyzed in 1997 have since been stored frozen at -21o C. New punches were analyzed in 2001 for signs of sample degradation (See Table 8.1). There is no statistical evidence that results for OC/EC analysis have been affected by freezer storage to date. In order to account for oxygen and hydrogen associated with organic material, the OC concentrations were next multiplied by 1.2 to estimate the organic matter concentrations. Factors used to convert organic carbon into organic mass in material balance studies range from 1.2 to 1.4 (Gray et al., 1986). B-61

Table 8.1 Effect of freezer storage on OC/EC results of archived samples Sample Name Date Analyzed

Organic Carbon Elemental Carbon (ug cm−2 ) (ug cm−2 )

Total Carbon (ug cm−2 )

OG-C-970428 OG-C-970428

8-01-1997 4-16-2001

3.47 ± 0.37 3.17 ± 0.36

0.29 ± 0.21 0.21 ± 0.21

3.76 ± 0.49 3.37 ± 0.47

OG-C-970416 OG-C-970416

8-01-1997 4-20-2001

5.89 ± 0.49 5.53 ± 0.48

1.18 ± 0.29 1.14 ± 0.26

7.07 ± 0.65 6.66 ± 0.63

OG-C-970410 OG-C-970410

8-01-1997 4-16-2001

5.52 ± 0.48 5.04 ± 0.45

1.49 ± 0.27 1.13 ± 0.26

7.00 ± 0.65 6.18 ± 0.61

OG-C-970422 OG-C-970422

8-01-1997 4-16-2001

6.02 ± 0.50 5.32 ± 0.47

1.73 ± 0.29 1.48 ± 0.27

7.75 ± 0.69 6.81 ± 0.64

OG-C-970814 OG-C-970814

9-30-1997 4-16-2001

5.51 ± 0.48 5.19 ± 0.46

1.11 ± 0.25 0.62 ± 0.23

6.62 ± 0.63 5.81 ± 0.59

OG-C-971212 OG-C-971212

12-30-1997 4-16-2001

13.00 ± 0.85 13.04 ± 0.85

1.21 ± 0.26 1.27 ± 0.26

14.21 ± 1.01 14.31 ± 1.02

OG-C-971118 OG-C-971118

12-29-1997 4-16-2001

12.19 ± 0.81 12.49 ± 0.82

3.80 ± 0.39 3.50 ± 0.38

15.99 ± 1.10 15.99 ± 1.10

OCy-C-970430 OCy-C-970430

12-30-1997 4-20-2001

5.61 ± 0.48 5.30 ± 0.47

1.81 ± 0.29 1.31 ± 0.27

7.42 ± 0.67 6.61 ± 0.63

OCy-C-970709 OCy-C-970709

12-31-1997 4-20-2001

5.78 ± 0.49 5.43 ± 0.47

0.38 ± 0.22 0.20 ± 0.21

6.16 ± 0.61 5.63 ± 0.58

ICy-C-970727 ICy-C-970727

12-30-1997 4-20-2001

4.63 ± 0.43 4.15 ± 0.41

0.31 ± 0.22 0.25 ± 0.21

4.93 ± 0.55 4.41 ± 0.52

B-62

page 1 of ___ OCEC ANALYSIS LOG SHEET

Date:___________ Initials:_________

Sample Description___________________________________________________________ Parameter file used: ______________________ Gas tank pressures (psi): He ____/____ Air ____/____

H2 ____/____

Flow rates*: Cal____ He2____ H2____ (*black ball except for air)

Data file: _____________________

He+Ox ____/____

He+Ox____

He1____

Cal ____/____

He3____

Air____ silver

| Sample ID | FIDmax | Comments |_______________________|___________|_________________________________________ | | | |_______________________|___________|_________________________________________ | | | |_______________________|___________|_________________________________________ | | | |_______________________|___________|_________________________________________ | | | |_______________________|___________|_________________________________________ | | | |_______________________|___________|_________________________________________ | | | |_______________________|___________|_________________________________________ | | | |_______________________|___________|_________________________________________ | | | |_______________________|___________|_________________________________________ | | | |_______________________|___________|_________________________________________ | | | |_______________________|___________|_________________________________________ | | | |_______________________|___________|_________________________________________ | | | |_______________________|___________|_________________________________________ | | | |_______________________|___________|_________________________________________ | | | |_______________________|___________|_________________________________________

B-63

9

GC/MS ANALYSIS OF QUARTZ FILTERS

This section describes the resources and analytical procedures for the measurement of trace organic compounds in the atmospheric particulate matter.

9.1 Apparatus and Materials The following equipment and materials are required for analyzing quartz fiber filters for organic compounds.

ITEM

SUPPLIER

MODEL or CATALOG NO.

Agilent

HP 5973 MSD

Sonicator

Fisher

FS60

Solvent Evaporator

Fisher

24/40 R124A

Nitrogen Blow Down

Fisher

PI18870, PI 18780

Muffle Furnace

VWR

F30420C-80

Distillation

Fisher

K547110-3000

Flask

Fisher

09-559B

Test Tube

Fisher

14-958F

Sample Vial

Supelco

2-7083U

Diazomethane Generator

Aldrich

Z10159-1

Benzene

VWR

EMBX0212-6

Isopropyl Alcohol

Fisher

A464-4

Hexane

Fisher

H303-4

Methanol

Fisher

A454-1

Dichloromethane

Fisher

D151-4

Aldrich

30996-6

Fisher

Gas-tight syringe

Scientific Specialties

270808

Agilent

HP5 MS

Internal Standards

Wisconsin State Hygiene Lab

sealed in the ampule

Quantification Standards

Wisconsin State Hygiene Lab

sealed in the ampule

GC/MS

Ether Syringe Sample Jar GC Column

B-64

9.2 Analysis Description The important aspects for organic tracer analysis of aerosol samples include preventing any contamination from reagents and sample processing, and maintaining consistent robust quantification. Blank filters will be routinely analyzed along with the ambient samples. To ensure the quality of quantitative analysis by GC/MS, an amount of 1 milligram of organic carbon or higher in each sample is recommended. Quartz fiber filters are analyzed for organic compound content by the technique developed by Glen Cass and his colleagues using GC/MS (gas chromatography/mass spectrometry) (Schauer et al., 1996). It includes extraction, concentration, derivatization, and GC/MS analysis. 9.2.1

Baking

To avoid any possible contamination, all glassware and quartz fiber filters should be baked in the Muffle furnace prior to use. 9.2.1.1. Glassware Cleaning 1. Rinse glassware twice with distilled/deionized water. (For used glassware, it is first soaked in and cleaned in cleaning solution with detergent, and then rinsed with tap water until the detergent has been removed.) 2. Cover any openings with aluminum foil. 3. Bake glassware in the Muffle furnace at 550o C for 12 hours. 4. Place the baked glassware in the clean storage area. 5. Rinse two times with dichloromethane before use. 9.2.1.2. Gas-tight Syringe 1. Prior to use, take syringe apart. 2. Sonicate it in a beaker filled with dichloromethane for 30 min. 3. Sonicate in a beaker filled with methanol for 30 min. 4. Resemble syringe. 5. Using the same solvents as in sample, rinse syringe 15 times. Then the syringe is ready for use. 6. Rinse the syringe with the same solvents immediately after contact with a sample or standards for 15 times. 7. Repeat cleaning procedure from 1 to 4, and store syringe in the clean storage area. B-65

8. When measuring a volume with a syringe, the appropriate size syringe should be used. Take entire sample into syringe, turn syringe up-side-down, remove air bubbles by striking it gently, and read the volume. 9.2.1.3. Nitrogen Blow Down 1. Check to ensure enough N2 gas in the tank (above 200 psi). 2. Clean the needles in the blow down apparatus with dichloromethane twice and leave it on a piece of clean aluminum foil to air-dry. 3. Put the clean needles (or tips) in the blow down stand. The rest opening should be capped. 4. Open slowly and ensure that the head pressure of the flow to the blow down apparatus is less than 2 psi. 5. Let it flow for 5 min to equilibrate and clean the system. 6. The needles should be at least 1 inch above the solvent surface. 7. Increase the head pressure slowly until you see a dent in the meniscus. If the solvent is bubbling, then the head pressure is too high. 8. Blow down the samples and watch closely. 9.2.2 Extraction and concentration Before extraction, each sample containing at least 1 milligram of organic carbon is spiked with 250 µl of the injection concentration of internal standard #1 and 250 µl of the injection concentration of internal standard #2. The deuterated standards included in internal standards #1 and #2 include benzaldehyde-d6 , dodecane-d26 , decanoic acid-d19 , phthalic acid3,4,5,6-d4 , acenaphthene-d10 , levoglucosan-13 C6 (carbon-13 uniform-labeled compound), hexadecane-d34 , eicosane-d42 , heptadecanoic acid-d33 , 4,4’-dimethoxybenzophenone-d8 , chrysene-d12 , octacosane-d58 , ααα-20R-cholestane-d4 , cholesterol-2,2,3,4,4,6-d6 , dibenz (ah)anthracene-d14 , and hexatriacontane-d74 . The sample is then extracted twice with hexane (Fisher Optima Grade), followed by three successive extractions with distilled benzene/2-propanol (2:1 mixture) (benzene: E & M Scientific Ominosolv benzene; 2-propanol: Fisher Optima Grade) under mild sonication for 15 min for each extraction. The five extractions are combined, filtered, and reduced to 5-10 mL by a rotary evaporator. The extract is transferred to a test tube and the volume is further reduced to 1 mL by high purity nitrogen blow down. Then, the extract is split into two fractions with one fraction stored in the freezer and the other fraction derivatized with diazomethane to convert organic acids to their methyl analogues for GC/MS analysis. The deuterated internal standards, which are used for quantifying the target compounds after GC/MS analysis, were spiked before extraction. Therefore, the recovery B-66

is corrected for the quantification results reported. With this type of procedure, past work indicates that, for the filter analysis, the recovery was 81 ± 6% for the deuterated n-tetradecane internal standard, 65 ± 4% for deuterated n-pentadecane internal standard, and 69 ± 15% for deuterated n-alkanoic acid internal standard. The average detection limit is 0.01 ng m−3 and the average uncertainties of quantification of organic compounds were determined to be ±20% (Schauer et al., 2002). 9.2.3

Derivatization

The concentration sample should be derivatized with an equal volume of a diazomethane solution. After derivatization, the sample is further concentrated by high purity nitrogen gas blow down. The final volume is the same as the spike volume of internal standard #1 or #2 to ensure that the concentrations of internal standards in the final volume are the same as the injection standards. 9.2.4

Quantification and GC/MS analysis

Quantification of each target organic compound is made by reference to a deuterated internal standard, which has chemical characteristics and retention time similar to the target compound. Quantification standards include 6 PMSTD standard mixtures (PMSTD#1 #6). Relative response factor (RRF) for each organic compound is calculated from the GC/MS analysis of the quantification standards, which contain isotopically labeled internal standards. For those target compounds which are not present in the quantification standards PMSTD#1 - #6, secondary standards including Wood Smoke Secondary Standard, Motor Oil Secondary Standard, and Candle Wax Secondary Standard are used to assist the identification. The RRF of these compounds are estimated from the response factors of compounds having similar chemical structure and retention time.

B-67

9.2.4.1 Conditions for the Analysis GC CONDITIONS Column

Injector Temperature (o C) GC/MS Interface Temperature(o C) Initial Oven Temperature (o C) Initial Oven Hold Time (min) Oven Temperature Ramp Rate (o C/min) Final Oven Temperature (o C) Final Oven Temperature Hold Time (min) Carrier Gas Carrier Gas Flowrate (mL/min) Injection

HP5 MS 30 m length 0.25 mm diameter 0.25 µm film thickness 300 300 65 2 10 300 30 Ultrapure Helium 1.0 Splitless, Autosampler

MS CONDITIONS Solvent Delay (min) Data Collection Mode Scan Range (amu)

4.5 Scan 50-650

RINSE (before and after injection) Rinse with Hexane Rinse with Methanol Sample Pump Sample Rinse Post Rinse with Hexane Post Rinse with Methanol

4 times 4 times 5 times 1 time 4 times 4 times

9.2.4.2. QA/QC for GC/MS Analysis 1. Before running any sample or standards, run standard autotune to check the performance of the instrument and potential problems such as air leaks 2. Before running any sample or standards, run unconcentrated solvent blanks including all of the solvents in the analysis 3. Run PMSTD#1 1:10 and PMSTD#4 1:10. The sensitivity and contamination checks for daily PMSTD#1 1:10 and PMSTD#4 1:10 include the following: (a) PMSTD#1 (coronene 300 to pyrene 202 ion ratio): Min 25 percent B-68

(b) PMSTD#1 (coronene 300 abundance): Minimum area of 10000 (c) PMSTD#4 (cholestane-D4 221+222 to cholesterol 386 ion ratio): Min 3 times (d) PMSTD#4 (cholesterol 386 abundance): Minimum area of 1,000,000 4. At the beginning and the end of each sample set (or minimum of every 3 days), run the following PMSTD standards which contain the deuterated internal standards #1 and #2. • PMSTD#1 1:5 • PMSTD#2 1:5 • PMSTD#3 1:5 (derivatized with diazomethane before GC/MS analysis) • PMSTD#4 1:5 • PMSTD#5 1:5 • PMSTD#6 1:5 • Wood Smoke Secondary Standard • Motor Oil Secondary Standard • Candle Wax Secondary Standard 5. After each 6 samples, run 1 randomly selected 1:5 dilutions of the above PMSTD standards 6. Relative response factors (RRF) are calculated for target compounds from the PMSTD runs at the beginning and the end of each sample set. These RRF should not vary from historical records more than 50 percent 7. Daily single point RRF checks should not vary more than 10 percent (Target) and should be flagged if vary more than 20 percent (Tolerance) 8. After GC/MS injection, samples will be sealed with solid caps and Teflon tapes and stored in the freezer to minimize the time that samples reside on autosampler tray 9. Other maintenance for GC/MS: (a) Change septum every 20 runs (For Merlin injectors, it can last longer. The suggested number is 1000 injections.) (b) Check liner every column change or once a month (c) Check the ultrapure Helium tank pressure every day (> 500 psi)

B-69

9.3 QA/QC 9.3.1

Solvent blanks

Check solvents used in the sample processing by concentrating the same amount of solvent used for a sample and analyzing with GC/MS. All solvents except benzene are optima grade because solvents can be a major source of contamination. Benzene is distilled and tested prior to use. 9.3.2

Filter or procedural blanks

Check the whole procedure by analyzing the pre-baked quartz fiber filters to get an estimate of the background of the procedure before the analysis of the real samples. 9.3.3

Precision

Determine the precision of the analysis by analyzing seven samples prepared by the pre-baked filters with standard mixtures spiked onto each sample. Quantify recovery of these standards.

B-70

10

ANALYSIS OF TEFLON FILTERS

The concentrations of all chemical species analyzed by ion chromatography are determined relative to primary or secondary laboratory standards of known concentration. Aqueous daily standards are prepared from more concentrated solutions prepared from ACS grade analytical reagents. Standard log sheets are filled out each time standards or reagents are prepared. Teflon filters are placed in individual extraction cups and wetted with 0.2 ml of ethanol (100 percent) to reduce the hydrophobic nature of this material (Derrick and Moyers, 1981). A Teflon rod is placed on top of each filter to keep it submerged, the extraction cup is sealed with a tight-fitting lid, and each Teflon filter is extracted by shaking it in a known volume (10-20 ml) of distilled, deionized water for 3 hours or more. After extraction, the concentrations of the major water soluble particulate species (SO4 , NO3 − , and Cl− ) are determined using a Dionex ion chromatograph. The same Teflon filter extracts also are analyzed for particulate ammonium ion (NH4 + ) by an indophenol colorimetric procedure employing a rapid flow analyzer (RFA-300 TM, Alpkem Corp.) (Bolleter et al., 1961). −2

10.1

Extracting Teflon Filters for Analysis

10.1.1

Apparatus and Materials for extraction ITEM

SUPPLIER

MODEL or CATALOG NO.

Ohaus

E400D

Ted Pella

5734

VWR

51147

Teflon filters

Pall Gelman

R2PJ047

Sample cups

VWR

25384-148

Eppendorf

25 µl

VWR

53512-350

McMaster Carr

8546K12

Copier Labels

Avery

8460

Ethyl Alcohol

Quantum

200 proof

Rocking platform

VWR

100

Refrigerator

Sanyo

small

LabIndustries

none

Balance Tweezers style AA Gelman s.s. forceps

Pippettor Pipette tips Teflon round rod

Repipette Dispenser

B-71

10.1.2

Preparing teflon filter extracts

Instructions for preparing teflon filters to be analyzed by IC and Alpkem (Derrick and Moyers, 1981). 1. Prepare a set of sample ID labels on stickers (e.g., Avery Copier Labels). 2. Place a filter in a sample cup with top (dark) side up. Use tweezers at all times to handle filters. 3. Using Eppendorf pipette, add 200 µl of EtOH (100%) to each filter. Carefully place EtOH on filter sample drop by drop, tipping cup so that EtOH runs across filter to cover entire sample surface evenly. The ethanol is used to ”wet” the teflon surface, (i.e., it reduces the hydrophobic nature of this filter material.) 4. Add 10-30 ml of DI water from Repipette dispenser for typical 24-hour ambient samples. Obtain DI water from MilliQ dispenser in room 102. Use the Ohaus Balance to measure how much DI water is added to each sample cup and record on the sample label. 5. Place one long teflon rod on top of the filter in each sample cup to hold down the filter, otherwise they float. 6. Tightly cap the sample cup and make sure it has the appropriate ID label. 7. Keep the samples in order. Put them in a box and place on the rocking platform. Agitate the samples vigorously for approximately three (3) hours. After shaking, store samples in the refrigerator until analyzed.

10.2

Ion Chromatography

Ion chromatography employing the Dionex 2020i allows for chemical analysis of aqueous solutions for various ionic species depending on which IC column is used. The user is referred to the Dionex manual, Dionex application sheets (available from Dionex) and to ”A Handbook for Ion Chromatography,” by J. Weiss published by Dionex for additional information. The AS4 anion column with an AG4 guard column is used to detect strong acid anions (e.g., Cl− , NO2 − , NO3 − , SO4 = ). See IonPac AS4 spec. sheet for general information about this column. 10.2.1

Apparatus and Materials for IC Analysis

The following apparatus and materials are used to analyze teflon filters by ion chromatography (IC).

B-72

ITEM

SUPPLIER

MODEL or CATALOG NO.

Ion Chromatograph

Dionex

2020i

HPIC-AS4 anion separator column

Dionex

P/N 35311

HPIC-AG4A anion guard column

Dionex

P/N 37042

Ohaus

E400D

Ted Pella

5734

Balance Tweezers style AA Teflon filter extracts

see previous section

Acrodisc LC13 syringe filters

VWR

28143-989

B-D syringes

VWR

BD309585

Pipettor 1-5 ml

Labnet

E-24600-25

Electronic pipettor

Rainin

EP-2500

Pipette tips

VWR

53512-350

Pipette tips

Rainin

RC-2500

Glass pipettes, size 5 x 1/10 ml

VWR

53283-738

IC chloride standard

VWR

EM-19897-1

IC nitrate standard

VWR

EM-19811-1

IC sulfate standard

VWR

EM-19813-1

Sodium carbonate

Mallinckrodt

7527

Sodium Bicarbonate

Mallinckrodt

7412

JT Baker

JT9681-2

Spatula

VWR

57901-140

Weighing boats

VWR

12577-027

Kimax

various sizes

VWR

13915-985

Spectra Phys

4270

Sulfuric acid

Volumetric flasks diSpo beakers, 5ml 4270 integrator

10.2.2

Advance preparations

Daily paper work should be filled out to insure quality control. See attached copies of daily standard preparation and log sheets for examples. Blank forms are located in the plastic file holder next to the Dionex manuals. Make additional copies of any blank forms when supplies run low. B-73

10.2.3

Master standards

Refer to the master standard preparation sheets for the appropriate amount of each reagent. Prepare the master standards from analytical grade reagents which can be obtained from several vendors if new stocks are desired. When opened, reagents should be marked with the date so that their age can later be determined. Prepare new master standards at least every 30 days. First, prepare a concentrated master standard from which daily standards will later be prepared. The master standard should contain at least Cl− , NO3 − , and SO4 = which are the main species of interest found on Teflon filters used to collect particles. Bromide and phosphate can also be detected with this column but are rarely seen in any of our typical samples. Concentrated solutions of individual ionic species are available from VWR and are stored in the refrigerator. Verify expiration dates listed on the bottles and replace when needed. The amounts to use for preparing the master standard are listed at the top of the daily standard preparation log sheet. After preparation store this standard in the designated polypropylene bottle in the refrigerator labelled with date prepared and preparer’s initials. A concentrated mixed standard is available from Dionex and can be used as an independent test for quality control. This standard contains (Cl− , F− , NO3 − , and SO4 = ) and should be diluted to the concentration range of the daily standards. The dilute standard should be analyzed periodically along with the samples and standards. Analyzed concentrations of QA/QC standard should be within 5 to 10% of known concentrations. 10.2.4

Eluent

Eluent: (HCO3 − / CO3 = ) Refer to the eluent preparation sheet and have the eluent prepared at least a day in advance of intended use. Make up one liter of concentrated sodium carbonate/sodium bicarbonate eluent (100x) following the directions on the eluent preparation log sheet. Store it in the refrigerator in the designated polypropylene bottle. Label the bottle with the date prepared and preparer’s initials. Make a new concentrated eluent approximately every 30 days. To prepare the eluent for daily use, add 40 ml of the concentrate to 4 liters of DI water. There is a 4 liter plastic bottle designated for this purpose located in the cabinet beneath the IC and a graduated cylinder designated for measuring 40 ml of eluent marked with the label “eluent.” Allow the dilute eluent to equilibrate for 24 hours before using. The dilute eluent will later be transferred into the appropriate cubitainer located on top of the IC. These cubitainers hold slightly more than 4 liters of fluid and will last for approximately 14 hours of IC operation.

B-74

10.2.5

Acid

A pressurized 4 liter bottle of dilute acid is located on top of the IC. A full container of acid will last for approximately 9 hours of IC operation. Make 25 mN H2 SO4 by adding 2.8 ml conc. H2 SO4 to 4 liters DI water. Use designated bottles located beneath the IC to prepare acid and then top off the acid in the containers on top. Always get 4 liters of water first, then add acid to water under the fume hood using a disposable glass pipette. 10.2.6 Preparing to turn on the IC Check the sample loop size (50 µl, 100 µl, and 200 µl currently exist). Larger loop sizes will give improved detection limits by increasing the amount of sample injected into the column at one time but will also require more sample and additional rinse quantities. The Dionex 2020i with an AS4 column can efficiently use loop sizes up to 1000 µl. Typically, the 50 µl loop is used unless sample concentrations are expected to be very small. Once one size loop is chosen and the integrator calibrated, changing the loop will require a new calibration. This is true of changing the IC range also. Also note, the sample injection valve ports are very sensitive to being over tightened. Check that the correct column is in place and oriented correctly. Note: the flow will be from right to left as one faces the instrument. Check that eluent and acid levels are adequate. It is best to have the containers full when planning a long days run, rather than turning off the machine to refill them during the day. Important, do not let the fluid levels get too low or air may be pumped through the column and/or suppressor. Passing bubbles through the column may result in a loss of efficiency and resolution and could damage the instrument. Check pressures. Two regulators located behind the IC control the air input to the acid container and to the valves controlling sample injection. A third regulator is attached where the air comes from the wall supply and should be open full (approx. 80 psi). These are set where marked on the regulator and should not be adjusted unless they are far off the marks. The IC column pressure is displayed on the front panel. If this pressure exceeds 1400 psi, it may be necessary to change one or more of the inline filters (normal pressure is 1000 - 1200 psi). See the problem section for more information. Periodically check for leaks, especially around the column joints and valves behind the panel containing the columns. If the anion micro-membrane suppressor located below the columns develops a leak, it will need to be replaced. (model AMMS-1 P/N 038019). Side note: The sample injection valve has eight ports numbered with 1 and 5 above and below the hole and other ports are numbered sequentially clockwise from these. Tubing should be connected as follows: Sample port ⇒ port 7; Eluent in line ⇒ port 1; line to column ⇒ port 5; line to waste ⇒ port 3; sample loop, port 2 ⇒ port 6; other loop, port 4 ⇒ port 8. Also, yellow air line is at top, green air line is at bottom. B-75

The IC waste lines all empty into the sink located next to the instrument. Periodically check that these have not pulled free or become clogged. 10.2.7

Turning on the instrument

The IC and integrator power stay on full time - DO NOT TURN IT OFF. Turning off the integrator power will erase ALL calibrated files in the integrator. If the integrator ever looses power the date and time will need to be reset and the command ATTEN=1024 issued. Start the flow of acid by pushing Switch A ON. Switch A is located on the SYS 1 strip in the top left corner of the IC unit. Other switches in this strip should be left set on B OFF and LOCAL. The Load-inject button can be in either position. The eluent flow is controlled by the switches in the System 1 Analytical Pump module located below the door into the column compartment. There are two Pressure Limit Select switches: • x 10 psi (lower limit) •

00, (for turn on then raise to 80) This is the lower limit, below which the pump automatically turns off and the eluent flow stops.

• x 100 psi (upper limit) •

14, this will turn the pump off automatically if the column pressure exceeds 1400 psi.

Check settings: • LOCAL-REMOTE on LOCAL • FLOW: 2.0 • Choose eluent number: 1 = carbonate • START - turns eluent flow on The pressure will be 1000-1400 psi and gradually rise during weekly use. If above 1400 psi the column prefilters should be changed. The System 1 Conductivity Detector is located in the module on the top left beneath the eluent cubitainers. On start-up the Cell should be turned ON with the Auto offset switch left OFF until the instrument has warmed up. The range will probably be set (usually 3). Lower ranges may give increased sensitivity (lower detection limits) but more signal noise. The temperature compensation should be 1.7.

B-76

Allow the IC to warm up with acid and eluent flowing through the columns before injecting the first standard. (Approximately 15-30 minutes.) The conductivity should be 21-24 µs. Before running the first standard switch the Auto Offset on the conductivity detector to ON. This compensates for the conductivity of the eluent and will change the displayed conductivity to zero. If the conductivity increases above 25 µs, the columns may need to be cleaned or the eluent or DI water supply may be contaminated. Record pressure and conductivity data onto the Dionex 2020i IC log sign in sheet located next to the IC. 10.2.8

Daily run

Prepare and run a set of daily standards. (Refer to the daily standard prep sheet.) Note that dedicated volumetric flasks labelled IC-A, B, C, etc have been set aside for this purpose for the strong acid standards. If no calibration curves exist the data will have to be programmed into the integrator. Generally, the integrator will have 2 or more files programmed with calibration curves covering a range of standards. Choose an appropriate file. Press ’USE FILE’ on the integrator and then press the desired number (e.g., ’2’) followed by ’ENTER’. This sets the integrator file to which data will be entered. Do not use a file that contains data (unless it’s your own) or you might write over someone else’s data. The LOAD - INJECT button at the top of the IC must be in the LOAD position to load a sample into the sample loop. Install a new prefilter (located in front of the sample port) at the start of the day and change it periodically as it becomes clogged. Additional prefilters can be obtained from VWR, Cat No. 28143-989, Acro LC13 0.2 µm syringe filters. Use a 3 ml syringe to load samples (available from VWR cat. no. BD309585). Pour a small amount (3-5 ml) of a low standard into a disposable beaker. Rinse the syringe 2-3 times with DI water and then rinse with a small amount (approx. 1.5 ml) of standard. Fill the syringe with standard and load it into the SYS 1 sample port. The volume injected into the IC should be at least 5-10 times the volume of the sample loop. Discard any remaining standard and the beaker. Press INJECT on the IC and INJ A on the integrator simultaneously. The integrator will print out a graph of the sample and eluent conductivity over time. When the preset end run time is reached, the integrator will give a listing of the concentration of each species for which a calibration curve exists. A sample run takes approximately 8 min. To calculate the concentrations of a just run standard or sample with a calibration curve in another file, change the file and recalculate the sample values (e.g., press ’USE FILE’ ’1’ on the integrator and then press ’RECALC’). Run a few standards over the range expected. If standards are reproducible, start running samples. Low standards should be reproduced to within 15%, high standards to B-77

within 5% of the calculated concentrations. If the reproducibility of a standard is poor rerun it. If reproducibility is still unacceptable, remake the standard and analyze it again. Also, run other standards and make sure they are reproducible. If standards are not reproducible, it may be necessary to recalibrate the integrator. To run a sample, rinse a syringe several times with DI water. Pour a small amount of sample into a disposable beaker. Teflon filter samples are extracted in DI water and need to have 1% of concentrated (100x) eluent added to the portion of the sample in the disposable beaker and thoroughly mixed before injecting. This will eliminate the negative water dip which would occur at the beginning of the run. Load a sample and inject it in the same manner as the standards. Wait for the run to finish and check whether the integrator was in the file with the correct range. If not, enter the other file and recalculate the sample concentrations. Many samples may have one species in the range of one calibrated file, and other species in the range of the other calibrated file. If any species of interest is off scale, the sample will need to be diluted and rerun. Run a different standard every 10-12 samples during the day, and run a couple standards at the end of the day to insure that the IC is functioning properly and the calibration curve has not changed. A typical 8 hour day run will generally encompass approximately 30-40 samples. 10.2.9

Calibrating the integrator

Detailed instructions are given in the Spectra-Physics (SP4270) Operators Manual under the dialog function in section 3. Enter DIALOG and answer the questions as printed on the integrator output. When finished with DIALOG you are in calibrate mode. Now run each standard, in order, from the lowest concentration standard to the highest. The integrator will fit a calibration curve and retain this in its memory. An alternate method of calibrating is to analyze each standard 2-3 times and record the peak areas. Then in calibrate mode, instead of running a standard, enter the average area of each species for the lowest concentration. This is done by typing: •

PSR(1)=2340000

where (1) is the species and 2340000 is the peak area. Continue for species 2-N and then press ’RECALC’. Repeat this procedure for all species at the next concentration, and again until you reach the highest concentration standard in the calibration curve. This can be tricky because the N in PSR(N) does not necessarily follow 1, 2, 3 ... For example, if the last run before calibration had unidentified species appear in the printout, such as the eluent peak in the first position, then you would enter PSR(2) for the first species you wish calibrated, PSR(3) for the second, etc. B-78

To review what is contained in a file press ’PR FILE’ which prints the information stored in the integrator. See example PR FILE. One thing that might need to be changed is the peak threshold, PT. This can be changed outside of DIALOG by typing: •

PT=new #

The higher the PT value, the higher a peak needs to be above the background noise to be detected. A note about entering items into the integrator memory. A light on the top right corner of the integrator panel has a variety of blinks. A solid light allows the bottom key function to be entered. A steady blink allows the left symbols on keys (i.e., numbers) to be entered. An irregular blink allows the right symbols (i.e., letters) to be entered. The latter two may be switched between by pressing SHIFT once, and switch back to the solid light by pressing SHIFT twice quickly. 10.2.10

Shutdown

Turn off acid flow by pressing “A OFF.” Turn off eluent flow by pressing “STOP.” Making sure the fluid flows have stopped is the most important part of the shutdown procedure. Also turn the conductivity CELL - OFF and Auto Offset - OFF. Fold up the integrator paper and clean up the area around the IC. Close the air valve located behind the instrument. Sign time finished on the Dionex 2020i log sheet and note any problems on the log sheet. 10.2.11

Calculations

Concentrations in ppm of all species are printed on the integrator and should be copied to the daily sample log sheet or a lab notebook for more convenient reference. Also the integrator paper is thermal sensitive paper and degrades over time. Convert ppm to µg/filter by multiplying by the dilution factor. The dilution factor equals the ml of H2 O added for extraction divided by the fraction of filter used. (e.g., the DF of 1/2 filter extracted in 20 ml of H2 O is 20 divided by 1/2 = 40). Divide µg/filter values by the volume of air sampled in m3 to arrive at µg/m3 values for all species and samples. 10.2.12

Cleaning the columns

Temporary cleaning procedure: Prepare a strong solution of the regular eluent being used. It should be from 20 to 200 times as concentrated as the regular eluting ion. Using the regular sample injection procedure, B-79

inject this strong eluent and run it through the separation column. The guard column should be removed and flushed separately. Repeat this procedure several times. The suppressor column and conductivity detector should be disconnected during this procedure by disconnecting the end of the column and allowing it to drain into a beaker. Normal cleaning of a poisoned column: Note: avoid pumping strong eluent cleaning solutions through the suppressor column and conductivity detector. Also separate the guard and anion columns and flush separately to avoid flushing contaminants from the guard column into the separator column. Prepare a solution of 0.1 to 1 N Na2 CO3 and fill the designated eluent cubitainer. Pump the strong eluent through the guard and separator columns for 30-40 minutes each. Replace normal strength eluent cube and pump eluent through the separator and guard columns for approximately 30 minutes each. Reconnect columns and the suppressor column. Note: water is not normally used to rinse the separator columns. It is only used to remove the previous eluent from the separator column when a second eluent is incompatible with the first. Cleaning Dionex IC Lines: Prepare a solution of 21 gms NaHCO3 in 1 liter DI water plus 1 liter methanol. Put inlet eluent line into a beaker of cleaning solution. Remove the columns from the IC and connect the lines using unions. Pump the cleaning solution, at a high flow rate, through the lines for approximately 20 minutes. After 20 minutes, rinse the instrument with DI water for 30 minutes. 10.2.13 Common problems After a long period of non-use the IC usually requires more time to stabilize upon start-up. Column cleaning, and flushing the column for several hours may be required if the IC has been inactive for more than a month. No calibrated files exist on the integrator. The power has been turned off, perhaps in a routine building shutdown. Conductivity is too high. There is probably a clog in the suppressor flow. Check the flow rate of regenerant out of the waste line. It should be approximately 3-5 ml/min. The two most common places for a clog to develop are in the prefilter to the suppressor line inside the regenerant container or at the waste line constrictor fitting. Column pressure is too high. The prefilters to the guard column and/or the separator column may need to be changed or the inline filters may need changing. Each of the columns has a 10 µm filter inside the fitting at the the beginning of the column. Be very careful not to over tighten this fitting when changing the filters. The inline filters do not need changing very frequently, however, if changing the prefilters does not help, these too should be changed. The inline filters are located between the eluent reservoir and the guard column. There are two filters, a 35 µm followed by a 10 µm, both located back-to-back in the gray filter holder. See Dionex instruction sheets for more information on changing filters and refer to column maintenance sheets which are supplied with the columns. B-80

Drifting baseline. Allow more time for the IC to warm-up and stabilize after insuring that the acid and eluent containers have been connected properly. Sample won’t load. Check the LOAD/INJECT button is in the LOAD position, possibly cycle the button in and out to release trapped air in the valve. Change the injection prefilter if it becomes more difficult to make injections during an IC run. See Dionex manual for more trouble shooting information. 10.2.14

Quality control standards

Refer to the IC Daily Standards Preparation Sheet for details on how to prepare quality control standards for the IC. At least one set of quality control standards is run daily when the instrument is in use. Table 10.1 shows standard values, means, and percent differences of analyzed daily ion analysis standards. 10.2.15

Instrument specifications

IC detection limits were determined by running progressively more dilute standards until the species could not be reliably detected by the instrument. The estimated detection limits based on extraction of samples in 10µl of distilled, deionized water are: • sulfate 0.4 µg/filter • nitrate 0.4 µg/filter • chloride 0.6 µg/filter For heavily loaded samples, sample extracts are diluted and rerun to bring the concentration within the instrument’s calibrated analysis range. Sample precision based on replicate analyses is shown in Table 10.2.

B-81

Table 10.1 Ion Analysis Standards Standard Value Mean of Analyzed Difference (ppm) Standards (ppm) (ppm)

Percent Average Sample Difference % (ppm)

Sulfate Standards: B-(0.174) C-(0.349) D-(0.700) E-(0.877) F-(1.750) G-(3.500)

0.195 0.364 0.707 0.888 1.768 3.471

0.021 0.015 0.007 0.011 0.018 -0.029

12.0 4.2 1.0 1.3 1.0 0.8

2.34

Nitrate Standards: B-(0.150) C-(0.300) D-(0.600) E-(0.750) F-(1.500) G-(3.000)

.158 .311 .585 .745 1.465 3.021

.008 .011 -.015 -.005 -.035 .021

5.3 3.7 2.5 0.7 2.3 0.7

0.74

.064 .111 .199 .256 .485 .972

.014 .011 -.001 .006 -.015 -.028

28.0 11.0 0.5 2.4 3.0 2.8

0.16

Chloride Standards: B-(0.050) C-(0.100) D-(0.200) E-(0.250) F-(0.500) G-(1.000)

Table 10.2 PRD Sample Replicate Analysis Species

Sulfate Nitrate Chloride

No. of Samples

Precision (µg/filter)

Average Sample (µg/filter)

72 107 51

1.10 0.58 0.19

137 21 4

B-82

Sample Range (µg/filter) 13 - 346 1.3 - 147 0.2 - 35

Eluent Preparation Sheet Strong Acids

Date Prepared__________ Initials ______________

100X Concentrated Carbonate Eluent = _ CO / HCO 3 3 Dionex 2020i

Species #gm/1000ml Actual gm ______________________________________________________________ | | | | | | Na CO | 23.32 | 2 3 | | | | | | | | | | NaHCO | 23.52 | 3 | | | |

DILUTION: 40 ml concentrate to 4000 ml DI water - Dionex 2020i

B-83

IC Daily Standards Preparation Sheet Date:___________ Initials______ Master Std. Conc. Date Prepared ___________ Initials________ (Stock solutions from VWR are 1000 ppm) (10 ml) _ Cl ______

(20 ml) _ NO ______ 2

(30 ml) _ NO ______ 3

(35 ml) --> 1000 ml DI = SO ______ (ppm) 4

Pipet Calibrations _____________ Calibration Date ___________ Initials_____ (milliliters) 0.20 __________ 0.50 __________

2.00 __________

5.00 __________

0.25 __________ 1.00 __________

2.50 __________

10.0 __________

Eluent:

Standards

Date prepared

100x __________

1x __________

Prepared by

100x __________

1x __________

Dilution (master->vol) (milliliters)

A

0.50 -> 200

B

1.00 -> 200

C

2.00 -> 200

D

2.00 -> 100

E

2.50 -> 100

F

5.00 -> 100

G

5.00 ->

Retention Times

50

Calculated Concentrations (ppm) _______________________________________ | Cl | NO2 | | NO3 | SO4 | |_______|_______|_______|_______|_______| | | | | | | |_______|_______|_______|_______|_______| | | | | | | |_______|_______|_______|_______|_______| | | | | | | |_______|_______|_______|_______|_______| | | | | | | |_______|_______|_______|_______|_______| | | | | | | |_______|_______|_______|_______|_______| | | | | | | |_______|_______|_______|_______|_______| | | | | | | |_______|_______|_______|_______|_______| | | | | | | |_______|_______|_______|_______|_______| B-84

page 1 of ___ IC

SAMPLE LOG SHEET (Strong Acids)

Date: __________________ Initials: ______________

Sample Description___________________________________________________________ Leaching Solution Matrix ____________________________________________________ Date Leached ___________________

Shaker Table Time ________________________

IC Range _________ Sample loop size ___________ | Filter| Sample ID | Dil. | ppm (not blank corrected) |Comments | ## | | Fact. | | |-------|-------------------|-------|----------------------------|---------| |///////////////////| | Cl | NO2 | NO3 | SO4 | |-------|-------------------|-------|-------|------|------|------|---------|_______|___________________|_______|_______|______|______|______|__________ | | | | | | | | |_______|___________________|_______|_______|______|______|______|__________ | | | | | | | | |_______|___________________|_______|_______|______|______|______|__________ | | | | | | | | |_______|___________________|_______|_______|______|______|______|__________ | | | | | | | | |_______|___________________|_______|_______|______|______|______|__________ | | | | | | | | |_______|___________________|_______|_______|______|______|______|__________ | | | | | | | | |_______|___________________|_______|_______|______|______|______|__________ | | | | | | | | |_______|___________________|_______|_______|______|______|______|__________ | | | | | | | | |_______|___________________|_______|_______|______|______|______|__________ | | | | | | | | |_______|___________________|_______|_______|______|______|______|__________ | | | | | | | | |_______|___________________|_______|_______|______|______|______|__________ | | | | | | | | |_______|___________________|_______|_______|______|______|______|__________

B-85

10.3

Alpkem Rapid Flow Analyzer for Ammonium Analysis

We use the Model RFA-300 Alpkem Rapid Flow Analyzer for Ammonium ion (NH4 + ) analysis. There are a variety of other colorimetric methods, such as formaldehyde and S(IV) aldehyde adducts, which can also be run on the Alpkem. The user is referred to the Alpkem manual and application sheets. Ammonium ion in aqueous solutions reacts with alkaline phenol and sodium hypochlorite to form a green-blue complex. An aliquot of sample is mixed by the auto-analyzer with the complexing reagent, Na3 PO4 , alkaline phenol, nitroprusside and sodium hypochlorite. The color is allowed to develop, the absorbance is measured, and the NH4 + concentration is calculated from a standard curve. 10.3.1

Apparatus and Materials for Ammonium Analysis

The following apparatus and materials are used to analyze teflon filters for ammonium ion by Alpkem.

ITEM

SUPPLIER

MODEL or CATALOG NO.

Alpkem

RFA-300

Cole Parmer

E-59000-62

VWR

34120-309

Pump tube, 74 ul/min

Alpkem

A903-1015P01

Pump tube, 118 ul/min

Alpkem

A903-1020P01

Pump tube, 226 ul/min

Alpkem

A903-1030P01

Digital stopwatch

McMaster Carr

1314T11

Volumetric flasks

Kimax

various sizes

Cole Parmer

E-24600-25

Pipette tips

VWR

53512-350

Glass pipettes, size 5 x 1/10 ml

VWR

53283-738

Omniscribe

D5000

Alpkem Analyzer pH tester Ammonia standard 1000ppm

Pipettor 1-5 ml

Chart recorder

10.3.2

Paper work

Daily paper work should be filled out to insure quality control. See attached copies of standard preparation and log sheets for examples.

B-86

10.3.3

Set up

1. Prepare reagents as outlined on the reagent preparation sheet. A 1000 ppm ammonia master standard can be purchased from the VWR stockroom (cat. no. 34120-309). Prepare the master standard substock as indicated on the daily standards log sheet. Prepare master standard substock, complexing reagent, and nitroprusside solution fresh each month, and store them in glass bottles in the refrigerator. Alkaline phenol solution should also be prepared fresh each week, but stored in an amber bottle in the refrigerator. Prepare sodium hypochlorite solution fresh daily. Suppressed peak heights may indicate that the reagents or standard are old and should be remade. 2. Replace all reagent tubing before starting each day. Also check that air and sample tubing lines are in good condition. Change the tubing when tubes appear flat. ALPKEM TUBING

FOR NH4 ANALYSIS

A

-

complex

-

black

--

black

-

226 UL/min

S

-

sample tube

-

black

--

black

-

226 UL/min

B

-

phenolate

-

orange

--

yellow

-

118 UL/min

C

-

NaOCl

-

orange

--

yellow

-

118 UL/min

D

-

nitropruss

-

green

--

orange

-

74 UL/min

E

-

wash tube

-

yellow

--

yellow

-

642 UL/min

F

-

air

-

yellow

--

orange

-

118 UL/min

3. Make sure correct lens filters are installed in the photometer. We use 305-1640-0885 (two of them). Install with purple side toward the light beam (toward the right as one faces the instrument.) These are only changed if analyses other than NH4 + have been performed with the instrument. 4. Insure that the heat bath (located under panel in front of tubing) is connected. The connections are under the mixing coils. 5. Set up the strip chart recorder to record peak heights. Using the top dial on the right hand side set the recorder to 1 Volt FULL SCALE after checking the zero. It is convenient to position the zero line along the 1” ruling of the paper to allow for negative dips. Using the dial and switches on the left side, set the chart speed to 4 inch/min ÷ 10. (a) Note: An integrator can be used instead of the strip chart recorder. Settings include: PH = 1; MN = 0; PT = 5000 to 10000. B-87

(b) Time functions include: (c) TT(1)=0

PM 1

(d) TT(2)=655.35 ER

0 (Maximum setting)

6. Check the waste beaker is properly positioned under the drain tube behind the detector. Use caution in handling waste, WEAR GLOVES. Dispose of waste in an appropriate container located in the hood and have it removed by Safety Office. 10.3.4

Turn on

1. Place all reagent tubes in the appropriate solution and put the sample tube in a beaker of fresh DI water. 2. Turn on 304 power switch, heat bath, light source, and pump (302). All four platens must be latched for pump to start. 3. Run briefly (approx. 20-30 min) to stabilize. It takes 9 minutes for solution to go completely through the system. Verify a smooth and consistent bubble pattern through the manifold. Refer to the Alpkem manual for adjusting bubbles if the pattern is erratic. 4. Set up the photometer. See next section for appropriate settings. 5. Make up the daily standards while waiting. Refer to the daily standards preparation sheet for instructions. Note: the matrix of the standards should match the matrix of the leaching solution (e.g., water, or water + EtOH). 10.3.5

Set up the photometer

1. The two dials on the left should be set with absorbance=1, and damping=2. 2. Set the bottom part of the two dials on the right to 0.1; this adjusts the gain of the photomultiplier. If the light bulbs are weak (old), the settings below will not be obtained, and the gain may need to be set to 1.0. 3. Turn the center dial to check: (a) - reference beam (close to 5, slightly greater) (b) - sample (5.05 to 5.10) (c) - absorbance (0.02) (d) - Note: the reference and sample voltages will be off-scale for the strip chart, briefly power off the strip chart while performing the above checks.

B-88

4. Adjust reference, sample, and absorbance as necessary to fall within the above guidelines with the absorbance value being the most important. Adjust sample using top of right knob labeled sample, and adjust absorbance using top of right knob labeled reference. To view each setting while adjusting knobs the center dial must be set to the appropriate indicator. Allow 2-3 seconds between adjustments for photometer to up-date the reading. 10.3.6

Typical days run

1. Adjust the photometer settings again. This will be necessary since the instrument has warmed up and the reagents have had time to equilibrate. 2. Pour standards into cups and load them into the cup holders on the autosampler. The cup holders are numbered (1 - - - -); (6 - - - -), etc. The auto-analyzer sometimes skips the first cup, so either put in an empty cup or put an extra DI water at the beginning. Note: Currently the auto sampler is not working. Manual loading of samples is performed by dipping the sample tube needle into the sample for 20 seconds followed by 40 seconds in rinse water, repeat 20 seconds in sample; 40 seconds rinse water. 3. For the first set of standards (A through G) run 3 replicates for each. In later runs use 2 standard replicates each time. 4. Before running samples, make sure the standards are consistent and the peak heights are reasonable. Standard G should be approximately 9-10 on the strip chart. 5. Teflon samples are ready to run after extraction. (See separate section for filter extraction). 6. Each sample should be run in replicate. Place the sample needle into the solution in the sample cup for 20 seconds. Avoid touching the filter with the tip of the needle. Sample for 20 seconds, rinse water 40 seconds, re-sample 20 seconds, rinse for 40 seconds (or until ready for the next sample). Important: Keep careful note of each sample (or standard) on the log sheets indicating the order samples are run. If you get samples and the order mixed up you will have to start over. 7. Run approximately 10 samples followed by a standard (in replicate). Repeat this sequence until all samples are analyzed, then end the day with a full set of standards. 10.3.7 End of day 1. After the last standard of the day is run, and the peak comes out on the strip chart power off the Alpkem and the strip chart recorder. 2. Transfer all reagent tube ends into a small waste beaker. Cap the reagent bottles, and store them in the refrigerator.

B-89

3. Leave latches over rollers open so tubing is not left in a compressed state. 4. Empty waste solution into the waste bottle under the hood. 10.3.8

Calibration Curves

1. Transfer the peak height information from the strip chart onto the daily log sheets. 2. Examine the peak heights of all standards, and take an average of the entire day for each standard. Sometimes it may be necessary to break a days run up into smaller segments, for example, if there were problems with the machine and the peak heights of the standards changed significantly (due to a baseline shift, for example). 3. Fit a linear curve to the final peak ht. averages and known concentration (ppm) of standards. Using the calibration curve, calculate the concentration (ppm) of each standard. Calculated values should be within 10% for the lowest standards (A-C) and 5% for higher standards (F-G). 4. Use the final calibration curve(s) to calculate the ppm (NH4 + ) of each sample. 5. Convert calculated ppm to µg/filter by multiplying by the dilution factor. The dilution factor equals the ml of H2 O and ethanol added for extraction divided by the fraction of filter used. [i.e., the DF equals 40 if 1/2 of a filter is extracted in 20 ml (20 divided by 1/2 = 40)]. 6. Divide µg/filter values by the volume of air sampled in m3 to arrive at µg m−3 values for all filters. 10.3.9 Reagent Preparation for the Alpkem Follow the reagent preparation sheet for NH4 + analysis. Reagents should be prepared new at least every 30 days. Use only DI water from the MilliQ dispenser in room 102 for preparing standards and insure that all glassware is thoroughly cleaned to avoid contamination. Refer to lab ware cleaning description. A. Complexing Reagent 1. Weigh appropriate amounts of sodium tartrate and sodium citrate and add to DI water in an Erlenmeyer flask. ( 80% of final volume). Shake gently until dissolved. 2. Measure the pH of this solution (usually 8.6) and adjust pH to 5.0 by adding concentrated H2 SO4 with a disposable glass pipette (approx. 1-1.5 ml). 3. Transfer solution to a volumetric flask and fill almost to the line with DI water.

B-90

4. Add 1% Brij-35 (30%) at the end because it is a surfactant, then fill to final volume with DI water. 5. Transfer to a glass bottle and store in the refrigerator. Label all reagent bottles with preparer’s initials and the date prepared. B. Other reagents 1. Alkaline Phenol - Weigh out NaOH pellets and dissolve in 2/3 of the water in a volumetric flask. Add liquid phenol and dilute to volume with DI water. Store in an amber bottle in the refrigerator. Safety Note: Phenol is a high risk material. Wear plastic gloves when handling. Discard waste into a waste bottle designated for phenol or organics. DO NOT POUR DOWN THE SINK - HAVE SAFETY OFFICE REMOVE WASTE. 2. Sodium Nitroprusside - Weigh out Na2 Fe(CN)5 in 2/3 of the water in a volumetric flask. Shake gently to dissolve, then add DI water to dilution line. Transfer to a glass bottle and refrigerate. 3. Sodium Hypochlorite - Mix new every day. Only use Clorox and replace stock bottle every 30 days. C. Wash Solution 1. Use DI water in the wash beaker when analyzing samples. Get fresh water daily and fill the beaker.

B-91

10.3.10 Quality control standards Refer to the NH4 Analysis Daily Standards Preparation Sheet for details on how to prepare quality control standards for ammonium ion analysis. At least three sets of quality control standards are run daily when the instrument is in use. Table 10.3 shows standard values, means, and percent differences of analyzed daily ammonium analysis standards.

Table 10.3 Ammonium Analysis Standards Standard Value Mean of Analyzed Difference (ppm) Standards (ppm) (ppm) B-(0.102) C-(0.204) D-(0.408) E-(0.997) F-(1.49) G-(2.00)

.103 .211 .413 .981 1.498 1.992

.001 .007 .005 -.016 .008 -.008

Percent Average Sample Difference % (ppm) 1.0 3.4 1.2 1.6 0.5 0.4

1.49

10.3.11 Instrument specifications Ammonium detection limits were determined by running progressively more dilute standards until the species could not be reliably detected by the instrument. The estimated detection limit based on extraction of samples in 10µl of distilled, deionized water is 0.2 µg/filter. For heavily loaded samples, sample extracts are diluted and rerun to bring the concentration within the instrument’s calibrated analysis range.

B-92

Reagent Preparation Sheet -- NH4 Analysis Date___________________ Initials_______________ Amount Added Per Final Volume 250 ml 500 ml 1000 ml --------------- --------------- -------------A. Complexing Reagent 1. Sodium Tartrate *2H2O 2. Sodium Citrate 3. Conc. H2SO4

*

4. Brij-35 (30%)

6.83 gm ______

13.66 gm ______ 27.3 gm ______

6.00 gm ______

12.00 gm ______ 24.0 gm ______

*

______

0.25 ml ______

* 0.5

______ ml ______

*

______

1.0 ml ______

Dissolve 1. & 2. in DI water, * adjust pH to 5.0 with conc. H2SO4, dilute to volume. Add 4. B. Alkaline Phenol 1. Liquid Phenol (88%) 2. NaOH pellets

23.5 ml ______

47.0 ml ______

94.0 ml ______

9.0 gm ______

18.0 gm ______

36.0 gm ______

Dissolve NaOH in 2/3 water, add phenol, dilute to volume. Safety note: use caution when handling phenol. C. Sodium Hypochlorite 1. NaOCl (Chlorox)

50.0 ml ______

Mix new daily.

D. Sodium Nitroprusside 1. Na Fe(CN) 2 5

0.125 gm ______

0.25 gm ______

E. Wash Solution 1. DI water

B-93

0.50 gm ______

NH4 Analysis Daily Standards Preparation Sheet Date:___________ Initials_______ Master Std. Conc. __ 1000 __ (ppm) NH4 From VWR Stockroom:

Date Opened ________

Master Substock Conc. __________ (ppm) NH4 (20 ml of conc. std into 100 ml DI water) Date Prepared _________ Initials_______

Calib. Value 20 ml_________

Pipet Calibrations _______________ Calibration Date _______________ (milliliters) 0.05__________

0.20__________

0.50__________

1.00 __________

0.10__________

0.25__________

0.75__________

2.00 __________

Reagents prepared:

Date

Prepared By

NaOCl

___________

___________

Alkaline Phenol

___________

___________

Complexing Reagent

___________

___________

Nitroprusside

___________

___________

Standards

Dilution (substock->vol) (milliliters)

Dil. Fact.

Calculated Concentrations _________________________

___

____ -> ____

_____

_________

___

____ -> ____

_____

_________

A

0.05 -> 200

4000

_________

B

0.05 -> 100

2000

_________

C

0.05 ->

50

1000

_________

D

0.10 ->

50

500

_________

E

0.25 ->

50

200

_________

F

0.75 -> 100

133

_________

G

0.50 ->

50

100

_________

H

0.75 ->

50

67

_________

___

____ -> ____

_____

_________

B-94

page 1 of ____ NH4

SAMPLE LOG SHEET

Date: ___________________ Initials: _______________

Sample Description________________________________________________________ Leaching Solution Matrix _________________________________________________ Leaching Solution Volume _________________

Calibrated on________________

Leaching Time ____________________________

Date Leached_________________

| Sample ID |Alpkem | Dil. | Peak Height | ppm | Comments | |position| Fact. | | | |__________________|________|________|_____________|_______|______________ | | | | | | |------------------|--------|--------|-------------|-------|-------------|__________________|________|________|_____________|_______|______________ | | | | | | |------------------|--------|--------|-------------|-------|-------------|__________________|________|________|_____________|_______|______________ | | | | | | |------------------|--------|--------|-------------|-------|-------------|__________________|________|________|_____________|_______|______________ | | | | | | |------------------|--------|--------|-------------|-------|-------------|__________________|________|________|_____________|_______|______________ | | | | | | |------------------|--------|--------|-------------|-------|-------------|__________________|________|________|_____________|_______|______________ | | | | | | |------------------|--------|--------|-------------|-------|-------------|__________________|________|________|_____________|_______|______________ | | | | | | |------------------|--------|--------|-------------|-------|-------------|__________________|________|________|_____________|_______|______________ | | | | | | |------------------|--------|--------|-------------|-------|-------------|__________________|________|________|_____________|_______|______________ | | | | | | |------------------|--------|--------|-------------|-------|-------------|__________________|________|________|_____________|_______|______________ | | | | | | |------------------|--------|--------|-------------|-------|-------------|__________________|________|________|_____________|_______|______________

B-95

11

LAB WARE CLEANING

Lab ware (glass and plastic) is used in the chemical analysis of trace elements and compounds. Some elements are measured in the part per billion range (ng/ml). This level of analysis requires immaculately clean lab ware, therefore, special care must be taken to prevent contamination from previous analysis, dirt, and air dust. 1. A plastic soaking tub is kept next to the sink in the lab. It is filled with tap water and 2-5% Contrad brand cleaning solution is added, (80-200 ml of Contrad per 4000 ml of water). Keep the soaking tub covered with a piece of plastic at all times to minimize dust contamination. 2. Lab ware to be cleaned should first be rinsed in tap water to remove chemicals. Caution: Hazardous chemicals should not be discarded down the drain. They should be placed in labeled waste containers and be discarded by Caltech’s Safety Office. 3. Remove all tape, labels, and ink before soaking. Ink can be removed with acetone or ethanol. 4. Submerge lab ware in the soaking tub. Make sure each piece is filled with solution (i.e., no air bubbles) so all surfaces are in contact with soap solution. DO NOT soak anodized aluminum. The soap solution removes the anodized coating. 5. Soak lab ware for at least one hour, but not more than a few days. 6. Rinse each piece with tap water to remove all soap (i.e., until no bubbles remain). This usually takes about four rinses. 7. Rinse each piece with distilled/deionized water three or four times to remove all traces of tap water. 8. Place on clean kimwipe toweling to dry. Invert or cover pieces with clean kimwipes to prevent dust from settling on the lab ware. 9. When dry, store pieces in appropriate cabinets. Volumetrics should be stoppered. Beakers and other containers should be turned upside down or covered with pieces of aluminum foil to prevent contamination by dust.

12

ERROR DETERMINATION FOR ATMOSPHERIC CONCENTRATIONS

Error analysis is based on the statistical propagation of individual errors (or precisions) (Skoog and West, 1988). In this section we denote the absolute error in a quantity, y as σ0y and the relative error as, σ∗y . In general, these are related by σ∗y =

σ0y y

B-96

12.1

Obtaining Precision Data

Each measured quantity in an experimental protocol has an associated precision. Precision is typically obtained by making a number, N, of repeated (replicate) measurements of the same underlying quantity. (Typically N > 30 for good statistics.) 1. The estimate of the value of a quantity, x, is the mean of the replicated measurements, xi , i.e., x=

1 N ∑ xi N i=1

2. The absolute precision in the estimate of x is given by the standard deviation of the individual measurements: v u N u u ∑ (xi − x)2 t i=1 σ0x = N−1 When N = 2, i.e., when the same quantity has been measured exactly twice, the above formulas simplify to: x1 + x2 2 |x1 − x2 | √ = 2

x= σx 0

12.2

Propagation of errors in formulas

1. For a sum or difference (e.g., y = a + b - c ...), the absolute error is the square root of the sum of the squares of the individual absolute errors: q 0 σy = σ0a 2 + σ0b 2 + σ0c 2 ... q σ0a 2 + σ0b 2 + σ0c 2 ... ∗ σy = y 2. For a product or quotient (e.g., y = ab c ) the relative error is the square root of the sum of the squares of the individual relative errors: s  0 2  0 2  0 2 σb σa σc 0 σy = y × + + a b c σ∗y

=

q

σ∗a 2 + σ∗b 2 + σ∗c 2 ...

B-97

3. A general example For the case of blank correcting a measured concentration (A - B), multiplying by the dilution factor (C), and dividing by the volume (D), each of which has an associated absolute precision. y =

(A − B) C D

(a) Calculate the relative error of the difference in the numerator:

σ∗M

M = A−B q 0 σ0A 2 + σ0B 2 σM = = M M

(b) Then the result, y, is a product, whose absolute error is: y = s σ0y

=

y × σ∗y

= y



σ0M M

MC D 2

 +

σ0C C

2

 +

σ0D D

2

Report y ± σ0y .

12.3

Example: Error Analysis for Atmospheric Concentrations

Atmospheric concentration measurements typically use ’field blanks’, i.e., samples which are handled and stored according to the same protocols as ’normal’ samples, but which are not exposed to the actual measurement process. These blanks allow one to account for background processes that would otherwise lead to systematic errors. If the measurement of a filter blank is less than the instrument detection limit (IDL), no blank correction is necessary. If the filter blank is detectable (i.e., greater than the IDL), then blank correction is necessary and the filter detection limit is 2 times the error in the average blank (σb ). The fundamental measurements for the sample in this example consist of initial and final weights, wti and wtf , a blank correction, B, the sampling duration t, and the flow rate Q. Each of these quantities has an absolute precision, σ0x , obtained from a number of independent replicated measurements. The volume of air sampled is given by: v = t×Q with v u 2 u σ0 σ∗v = t t + t B-98

σ0Q Q

!2

Then the atmospheric concentration is given by: Catmos = where

∆M = (wtf − wti ) − B s σ0atmos = Catmos × σ∗atmos = Catmos

13

∆M v

(σ0wti )2 + (σ0wtf )2 + (σ0B )2 + (σ∗v )2 (∆M)2

REFERENCES

Birch, M.E.; and Cary, R.A. (1996) “Elemental carbon-based method for monitoring occupational exposures to particulate diesel exhaust,” Aerosol Science and Technology 25 221-241. Bolleter, W.T.; Bushman, C.T.; Tidell, P.W.; (1961) “Spectrophotometric determinations of ammonium as indophenol,” Analytical Chemistry 33 592. Derrick, M. and Moyers, J.; (1981) “Precise and Sensitive Water Soluble Ion Extraction Method for Aerosol Samples Collected on Polytetrafluorethylene Filters,” Anal. Lett. 14 1637-1652. EPA (1997) Ambient Air Monitoring Reference and Equivalent Methods, US Environmental Protection Agency, 40 CFR Part 58, July, 1997. EPA (1998) Quality Assurance Guidance Document 2.12, US Environmental Protection Agency, November 1998. Gray, H.A.; Cass, G.R.; Huntzicker, J.J.; Heyerdahl, E.K.; and Rau, J.A. (1986) “Characteristics of atmospheric organic and elemental carbon particle concentrations in Los Angeles,” Environmental Science and Technology 20 580-589. John, W.; and Reischl, G. (1980) “A cyclone for size-selective sampling of air,” Journal of the Air Pollution Control Association 30 872-876. NIOSH (1999) NIOSH Manual of Analytical Methods (NMAM), 4th edition. Salmon, L.G.; Christoforou, C.S.; and Cass, G.R. (1994) “Airborne Pollutants in the Buddhist Cave Temples at the Yungang Grottoes, China,” Environmental Science and Technology 28 805–811. B-99

Salmon, L.G.; Cass, G.R.; Kozlowski, R.; Hejda, A.; Spiker, E.C.; and Bates, A.L.; (1996) “Air Pollutant Intrusion into the Wieliczka Salt Mine,” Environmental Science and Technology 30 872–880. Schauer, J.J.; Rogge, W.F.; Hildemann, L.M.; Mazurek, M.A.; Cass, G.R.; and Simoneit, B.R.T. (1996) “Source apportionment of airborne particulate matter using organic compounds as tracers,” Atmospheric Environment 30 3837-3855. Schauer, J.J.; Kleeman, M.J.; Cass, G.R.; Simoneit, B.R.T.;(2002) “Measurement of emissions from air pollution sources. 4. C-1-C-27 organic compounds from cooking with seed oils,” Environmental Science and Technology 36 567–575. Skoog, D.A.; West, D.M.; and Holler, F.J. (1988) Fundamentals of Analytical Chemistry Saunders Publishing Company, pp. 70–79. Solomon, P.A.; Fall, T.; Salmon, L.G.; Lin P.; Vasquez, F.; and Cass, G.R. (1988) “Acquisition of Acid Vapor and Aerosol Concentration Data for Use in Dry Deposition Studies in the South Coast Air Basin,” Environmental Quality Laboratory Report #25, Final report, submitted to California Air Resources Board, March, 1988. Thermo Andersen (2001) RAAS Operator’s Manual: Model RAAS2.5-400 Revision 6, Thermo Andersen, Smyrna, GA 30082.

B-100

Appendix C – Mass and Carbon Data in µg m Site CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 GZ5 GZ5 GZ5 GZ5 GZ5

Date

Mass

20021002 25.57  0.66 20021008 35.19  0.88 20021014 50.19  0.94 20021020 37.00  0.78 20021026 28.76  0.69 20021201 18.10  0.88 20021207 32.27  0.97 20021213 26.60  0.93 20021219 46.27  1.10 20021225 18.16  0.88 20030301 68.47  1.26 20030307 26.05  1.03 20030313 71.44  1.30 20030319 no data 20030325 36.59  0.88 20030605 80.19  1.61 20030611 29.37  1.12 20030617 27.11  1.10 20030623 23.60  1.08 20030629 18.61  1.06 20021002 34.52  0.88 20021008 46.81  1.01 20021014 19.03  0.76 20021020 13.07  0.73 20021026 37.27  0.91 20021201 36.76  1.27 20021207 32.44  1.24 20021213 34.22  1.25 20021219 51.46  1.39 20021225 33.72  1.25 20030301 40.09  1.05 20030307 30.93  0.97 20030313 35.50  1.01 20030319 34.65  1.00 20030325 55.77  1.21 20030605 52.86  1.56 20030611 20.15  1.37 20030617 55.24  1.58 20030623 13.58  1.35 20030629 8.22  1.34 20021002 84.76  1.50 20021008 54.49  1.10 20021014 102.46  1.74 20021020 43.06  0.97 20021026 46.43  1.01

3

Organics Elemental Carbon (OC *1.4) 10.18  0.73 11.56  0.85 18.10  1.18 12.94  0.88 9.18  0.67 8.61  0.63 9.84  0.70 10.68  0.74 19.43  1.25 6.09  0.49 23.27  1.47 8.90  0.73 23.97  1.51 no data 11.51  0.81 26.18  1.62 8.27  0.62 9.60  0.69 7.25  0.56 5.47  0.46 10.12  0.71 16.35  1.06 4.97  0.42 2.18  0.27 11.44  0.78 8.40  0.62 11.31  0.78 10.63  0.74 13.55  0.91 11.17  0.77 11.38  0.78 11.12  0.77 8.90  0.64 8.55  0.63 12.38  0.84 9.62  0.69 3.86  0.36 15.78  1.02 2.04  0.27 1.50  0.24 41.71  2.49 21.53  1.34 44.94  2.67 14.25  0.94 18.90  1.20

C-1

1.16  0.20 1.40  0.25 1.93  0.23 1.93  0.23 1.12  0.19 1.68  0.21 1.59  0.21 1.35  0.20 2.02  0.23 0.77  0.17 2.18  0.25 0.58  0.24 2.15  0.25 no data 1.13  0.20 1.91  0.23 1.26  0.20 0.72  0.17 0.97  0.19 0.95  0.18 4.34  0.35 2.21  0.23 1.23  0.19 0.75  0.16 1.47  0.20 1.45  0.20 2.70  0.26 2.39  0.25 3.23  0.29 2.39  0.25 3.05  0.28 2.16  0.24 1.07  0.18 0.90  0.18 2.08  0.23 1.24  0.19 2.35  0.24 2.02  0.23 0.66  0.16 0.46  0.15 4.49  0.35 4.35  0.34 5.99  0.43 3.53  0.30 4.84  0.37

Site GZ5 GZ5 GZ5 GZ5 GZ5 GZ5 GZ5 GZ5 GZ5 GZ5 GZ5 GZ5 GZ5 GZ5 GZ5 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 TC2 TC2 TC2 TC2 TC2 TC2 TC2 TC2 TC2 TC2 TC2 TC2 TC2

Date

Mass

20021201 100.69  1.94 20021207 84.63  1.74 20021213 49.80  1.38 20021219 96.17  1.88 20021225 30.76  1.24 20030301 84.49  1.57 20030307 41.75  1.08 20030313 116.25  2.01 20030319 32.63  1.00 20030325 122.06  2.08 20030605 74.95  1.76 20030611 77.76  1.79 20030617 98.59  2.02 20030623 44.11  1.48 20030629 25.97  1.38 20021002 55.95  1.11 20021008 69.83  1.29 20021014 26.17  0.80 20021020 17.98  0.75 20021026 43.88  0.97 20021201 34.05  1.25 20021207 30.14  1.22 20021213 59.74  3.44 20021221 70.29  3.64 20021225 39.68  1.29 20030102 94.19  1.85 20030301 53.13  1.19 20030307 40.17  1.06 20030313 56.78  1.23 20030319 50.42  1.16 20030325 72.57  1.42 20030605 59.41  1.61 20030611 33.90  1.42 20030617 57.16  1.59 20030623 12.89  1.33 20030629 11.38  1.33 20021002 35.48  0.87 20021008 43.96  0.96 20021014 20.81  0.75 20021020 9.26  0.69 20021026 36.15  0.88 20021201 29.33  1.20 20021207 15.52  1.13 20021213 37.26  1.25 20021219 37.76  1.25 20021225 36.09  1.24 20030301 34.84  1.01 20030307 40.60  1.07 20030313 33.46  1.00

Organics Elemental Carbon (OC *1.4) 45.58  2.70 27.54  1.68 19.39  1.22 31.33  1.90 11.13  0.76 24.25  1.50 13.55  0.90 34.71  2.10 9.02  0.64 37.42  2.25 21.01  1.32 22.57  1.40 29.85  1.82 14.07  0.92 9.51  0.67 26.95  1.67 28.31  1.75 10.19  0.73 4.09  0.38 17.00  1.11 8.30  0.62 10.25  0.73 22.11  1.68 26.05  1.92 13.89  0.92 27.93  1.72 14.90  0.99 13.16  0.90 18.33  1.19 16.84  1.11 21.83  1.39 13.03  0.88 11.00  0.77 16.56  1.08 3.56  0.35 2.34  0.29 13.94  0.92 15.48  1.01 8.33  0.60 1.72  0.25 13.29  0.88 6.29  0.49 3.61  0.34 13.52  0.90 10.88  0.74 12.95  0.87 7.06  0.53 11.90  0.81 7.48  0.56 C-2

6.55  0.46 3.92  0.32 5.91  0.43 7.28  0.50 2.83  0.26 4.29  0.34 2.45  0.24 4.66  0.36 2.71  0.26 8.35  0.57 1.69  0.20 3.02  0.27 3.16  0.28 4.83  0.37 2.54  0.25 5.34  0.41 5.42  0.41 3.27  0.30 2.25  0.24 4.02  0.34 1.52  0.21 3.92  0.33 5.46  0.66 10.73  0.95 2.93  0.28 4.78  0.38 2.75  0.28 2.19  0.25 5.16  0.41 3.89  0.34 5.80  0.44 1.51  0.21 6.26  0.46 2.27  0.24 1.35  0.20 0.94  0.18 3.59  0.31 2.58  0.25 0.96  0.17 0.86  0.17 2.13  0.23 1.24  0.18 2.48  0.24 2.93  0.27 2.70  0.25 3.47  0.30 2.06  0.23 1.90  0.22 1.65  0.21

Site

Date

TC2 TC2 TC2 TC2 TC2 TC2 TC2 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7

20030319 20030325 20030605 20030611 20030617 20030623 20030629 20021002 20021008 20021014 20021020 20021026 20021201 20021207 20021213 20021219 20021225 20030301 20030307 20030313 20030319 20030325 20030605 20030611 20030617 20030623 20030629 20021002 20021008 20021014 20021020 20021026 20021201 20021207 20021213 20021219 20021225 20030301 20030307 20030313 20030319 20030325 20030605 20030611 20030617 20030623 20030629

Mass 45.87  1.12 44.84  1.11 58.34  1.61 13.63  1.35 55.75  1.59 11.89  1.34 8.92  1.34 24.78  0.81 48.63  1.03 15.78  0.75 10.85  0.73 29.06  0.84 31.65  1.23 23.12  1.19 31.49  1.23 44.65  1.33 30.62  1.23 29.94  0.99 20.40  0.93 32.88  1.01 27.30  0.97 53.94  1.21 44.13  1.49 11.67  1.34 47.00  1.52 10.75  1.34 6.18  1.33 no data 79.55  1.35 43.92  0.87 13.95  0.58 32.29  0.74 62.12  1.29 30.87  0.98 69.89  1.38 78.64  1.49 50.82  1.16 44.98  0.97 41.96  0.94 51.42  1.05 57.35  1.12 81.40  1.44 52.65  1.34 15.93  1.09 55.22  1.37 10.74  1.09 9.49  1.07

Organics (OC *1.4)

Elemental Carbon

10.63  0.74 11.09  0.76 9.39  0.66 2.67  0.29 13.75  0.91 1.40  0.22 1.19  0.21 8.55  0.63 15.72  1.04 2.87  0.32 3.37  0.34 9.70  0.69 6.45  0.50 4.00  0.38 8.95  0.64 10.00  0.70 9.34  0.67 3.95  0.38 5.85  0.48 5.63  0.46 6.71  0.53 11.41  0.78 7.31  0.56 2.28  0.28 12.39  0.84 1.30  0.24 1.36  0.24 no data 30.41  1.85 21.74  1.36 4.96  0.42 8.15  0.59 22.22  1.39 9.66  0.67 32.51  1.97 23.56  1.46 18.09  1.15 12.57  0.85 14.15  0.95 17.56  1.13 11.56  0.80 23.25  1.46 8.46  0.62 4.00  0.36 12.29  0.83 2.66  0.29 1.60  0.24

2.09  0.23 2.98  0.27 1.08  0.18 2.21  0.23 2.24  0.23 0.40  0.15 0.24  0.14 0.97  0.18 1.80  0.22 0.37  0.16 0.41  0.16 0.89  0.18 0.99  0.18 0.41  0.16 1.60  0.21 1.29  0.19 1.27  0.19 0.40  0.16 0.50  0.16 0.64  0.17 0.74  0.17 1.17  0.19 0.54  0.16 0.98  0.18 1.21  0.19 0.10  0.14 0.15  0.15 no data 3.21  0.28 2.51  0.25 1.72  0.21 0.97  0.17 3.85  0.32 2.56  0.25 4.46  0.35 4.87  0.37 3.99  0.32 1.71  0.22 1.65  0.22 3.34  0.31 1.92  0.23 3.76  0.33 1.29  0.19 2.12  0.23 1.43  0.20 1.18  0.19 0.97  0.17

C-3

Appendix D – Ions Data in µg m Site CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 GZ5 GZ5 GZ5 GZ5 GZ5 GZ5

Date 20021002 20021008 20021014 20021020 20021026 20021201 20021207 20021213 20021219 20021225 20030301 20030307 20030313 20030319 20030325 20030605 20030611 20030617 20030623 20030629 20021002 20021008 20021014 20021020 20021026 20021201 20021207 20021213 20021219 20021225 20030301 20030307 20030313 20030319 20030325 20030605 20030611 20030617 20030623 20030629 20021002 20021008 20021014 20021020 20021026 20021201

3

Chloride

Nitrate

Sulfate

Ammonium

0.02  0.04 0.20  0.05 0.05  0.10 0.01  0.04 0.06  0.04 0.05  0.04 0.05  0.04 0.02  0.04 0.01  0.04 0.00  0.03 0.12  0.05 0.01  0.05 0.04  0.05 no data 0.02  0.04 0.02  0.05 0.00  0.03 0.01  0.04 0.02  0.04 0.05  0.04 0.02  0.05 0.09  0.05 0.49  0.09 0.00  0.04 0.05  0.05 0.71  0.05 0.11  0.05 0.10  0.05 0.25  0.05 0.06  0.05 0.13  0.05 0.04  0.05 0.25  0.05 0.07  0.05 0.09  0.05 0.03  0.05 0.32  0.05 0.01  0.05 0.23  0.05 0.31  0.05 0.29  0.06 0.09  0.05 0.17  0.06 0.39  0.05 0.05  0.05 3.02  0.19

0.19  0.04 0.47  0.05 0.23  0.05 0.20  0.04 0.36  0.04 0.22  0.04 0.22  0.04 0.33  0.04 0.32  0.04 0.05  0.04 0.65  0.05 0.11  0.06 0.67  0.05 no data 0.88  0.05 0.44  0.05 0.13  0.04 0.06  0.04 0.32  0.04 0.37  0.04 0.42  0.05 0.75  0.06 1.50  0.10 0.18  0.05 0.65  0.06 1.80  0.07 1.42  0.07 1.09  0.07 2.44  0.09 1.00  0.07 1.23  0.06 1.19  0.07 1.82  0.08 1.54  0.08 2.02  0.08 0.22  0.05 0.33  0.05 0.25  0.05 0.30  0.05 0.32  0.05 4.83  0.17 0.97  0.06 2.55  0.11 0.76  0.06 0.49  0.06 4.10  0.13

6.95  0.22 7.92  0.35 13.31  1.01 12.11  1.02 8.66  0.30 2.34  0.18 10.84  0.92 6.86  0.22 11.28  0.36 7.27  0.70 22.11  1.57 9.01  0.27 18.21  1.47 no data 8.99  0.32 23.02  1.64 8.04  0.29 6.97  0.28 8.14  0.69 6.14  0.21 8.36  0.28 10.02  0.37 5.46  0.24 5.99  0.25 11.20  0.39 11.22  0.94 9.85  0.92 8.19  0.28 15.98  1.36 9.02  0.29 12.29  0.41 6.87  0.27 8.06  0.28 9.12  0.37 13.39  0.42 16.94  1.32 2.34  0.23 17.06  1.32 2.65  0.23 1.67  0.08 15.39  0.52 11.08  0.39 22.57  1.75 10.42  0.38 11.38  0.39 13.06  0.52

1.75  0.05 2.36  0.08 2.61  0.09 2.46  0.16 1.93  0.06 0.62  0.04 2.32  0.14 1.49  0.11 2.51  0.17 1.77  0.11 4.98  0.18 2.22  0.07 3.29  0.22 no data 2.00  0.07 5.63  0.26 1.82  0.06 1.81  0.06 1.67  0.08 1.52  0.05 1.58  0.06 2.47  0.08 1.16  0.05 1.98  0.06 3.87  0.11 3.19  0.15 2.45  0.14 1.96  0.14 4.04  0.21 2.24  0.14 2.97  0.14 2.41  0.07 2.75  0.09 2.64  0.09 3.69  0.15 3.85  0.15 0.56  0.05 4.33  0.16 0.64  0.05 0.39  0.04 4.36  0.13 2.91  0.09 4.21  0.27 2.96  0.09 3.75  0.10 4.73  0.29

D-1

Site GZ5 GZ5 GZ5 GZ5 GZ5 GZ5 GZ5 GZ5 GZ5 GZ5 GZ5 GZ5 GZ5 GZ5 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 TC2 TC2 TC2 TC2 TC2 TC2 TC2 TC2 TC2 TC2 TC2 TC2 TC2 TC2

Date 20021207 20021213 20021219 20021225 20030301 20030307 20030313 20030319 20030325 20030605 20030611 20030617 20030623 20030629 20021002 20021008 20021014 20021020 20021026 20021201 20021207 20021213 20021221 20021225 20030102 20030301 20030307 20030313 20030319 20030325 20030605 20030611 20030617 20030623 20030629 20021002 20021008 20021014 20021020 20021026 20021201 20021207 20021213 20021219 20021225 20030301 20030307 20030313 20030319

Chloride

Nitrate

Sulfate

Ammonium

1.57  0.07 0.31  0.05 2.77  0.19 0.08  0.05 0.91  0.06 0.18  0.05 1.06  0.07 0.50  0.05 0.51  0.07 0.03  0.06 1.08  0.06 1.06  0.06 0.39  0.05 0.12  0.05 0.68  0.18 0.23  0.06 0.03  0.05 0.03  0.05 0.08  0.05 0.58  0.05 0.30  0.05 0.35  0.14 1.05  0.14 0.22  0.05 1.08  0.06 0.28  0.05 0.26  0.05 0.55  0.06 0.95  0.06 0.40  0.06 0.21  0.05 0.21  0.05 0.01  0.05 0.21  0.05 0.30  0.04 0.00  0.03 0.09  0.05 0.13  0.05 0.00  0.03 0.02  0.05 0.12  0.05 0.09  0.05 0.11  0.05 0.02  0.05 0.08  0.05 0.00  0.04 0.12  0.05 0.05  0.05 0.19  0.05

7.43  0.22 2.36  0.09 9.00  0.31 1.02  0.07 5.80  0.20 1.54  0.09 13.43  0.40 4.17  0.15 10.70  0.29 0.81  0.06 3.86  0.12 5.01  0.16 0.87  0.06 0.41  0.05 2.44  0.18 1.34  0.07 0.48  0.05 0.36  0.05 0.54  0.05 2.07  0.10 1.19  0.07 4.25  0.22 4.71  0.24 1.97  0.08 6.78  0.24 3.42  0.16 2.87  0.10 3.43  0.14 3.87  0.12 4.49  0.14 1.39  0.08 1.23  0.06 0.85  0.06 0.45  0.05 0.44  0.05 0.23  0.05 0.74  0.05 1.10  0.07 0.18  0.05 0.26  0.05 0.75  0.05 0.54  0.05 1.51  0.08 0.65  0.05 1.72  0.07 0.37  0.05 2.74  0.11 1.09  0.07 2.64  0.10

19.00  0.60 9.22  0.37 20.24  1.89 8.50  0.29 23.24  0.81 10.98  0.52 21.30  0.61 5.48  0.25 20.47  0.64 20.91  1.79 14.13  0.43 22.58  1.31 9.04  0.36 5.20  0.24 9.68  0.49 13.17  0.49 5.58  0.45 5.29  0.24 10.16  0.37 11.05  0.94 7.92  0.28 9.84  0.71 8.18  0.74 9.30  0.32 24.53  0.84 15.47  0.58 8.43  0.36 9.30  0.38 7.37  0.27 12.23  0.38 16.76  0.97 3.82  0.24 16.67  0.96 2.46  0.23 1.74  0.07 8.92  0.35 9.82  0.36 5.33  0.24 4.00  0.23 11.24  0.39 11.39  0.92 3.61  0.16 7.93  0.33 13.60  1.19 8.91  0.29 13.23  0.43 10.16  0.38 7.99  0.24 9.66  0.38

5.94  0.30 2.31  0.20 7.62  0.32 2.09  0.19 6.51  0.23 3.76  0.11 7.46  0.39 2.73  0.08 8.76  0.57 6.97  0.23 3.70  0.19 6.03  0.22 2.02  0.08 1.33  0.05 2.35  0.08 4.21  0.12 0.88  0.05 1.71  0.06 3.25  0.10 3.25  0.12 2.71  0.08 3.98  0.16 4.07  0.17 2.60  0.16 7.17  0.39 4.22  0.16 3.39  0.10 3.66  0.10 3.63  0.10 4.65  0.13 4.46  0.16 1.02  0.07 4.34  0.16 0.68  0.05 0.43  0.04 2.45  0.08 3.04  0.09 1.12  0.05 1.29  0.05 3.55  0.10 3.47  0.12 1.12  0.05 2.45  0.19 3.20  0.19 2.42  0.14 3.28  0.10 3.58  0.10 2.00  0.06 3.66  0.10

D-2

Site

Date

Chloride

Nitrate

Sulfate

Ammonium

TC2 TC2 TC2 TC2 TC2 TC2 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7

20030325 20030605 20030611 20030617 20030623 20030629 20021002 20021008 20021014 20021020 20021026 20021201 20021207 20021213 20021219 20021225 20030301 20030307 20030313 20030319 20030325 20030605 20030611 20030617 20030623 20030629 20021002 20021008 20021014 20021020 20021026 20021201 20021207 20021213 20021219 20021225 20030301 20030307 20030313 20030319 20030325 20030605 20030611 20030617 20030623 20030629

0.02  0.05 0.00  0.04 0.16  0.05 0.00  0.05 0.26  0.05 0.34  0.05 0.00  0.04 0.03  0.05 0.55  0.09 0.01  0.05 0.02  0.05 0.19  0.05 0.02  0.05 0.09  0.05 0.03  0.05 0.02  0.05 0.02  0.05 0.02  0.05 0.15  0.05 0.05  0.05 0.01  0.05 0.01  0.05 0.19  0.05 0.02  0.05 0.16  0.04 0.28  0.05 no data 0.25  0.04 0.06  0.04 0.00  0.03 0.07  0.04 0.92  0.05 0.19  0.04 0.44  0.05 0.88  0.05 0.72  0.04 0.06  0.04 0.25  0.04 0.21  0.04 1.62  0.06 0.25  0.05 0.01  0.04 0.08  0.04 0.04  0.04 0.04  0.04 0.19  0.04

0.35  0.06 0.15  0.05 0.23  0.05 0.10  0.05 0.29  0.05 0.35  0.05 0.18  0.05 0.54  0.06 0.90  0.05 0.21  0.05 0.27  0.05 0.69  0.05 0.15  0.05 0.63  0.05 0.38  0.05 0.44  0.05 0.20  0.05 0.30  0.05 1.00  0.07 0.89  0.07 0.83  0.06 0.17  0.05 0.26  0.05 0.22  0.05 0.27  0.05 0.39  0.05 no data 1.46  0.08 0.51  0.04 0.17  0.04 0.59  0.04 2.28  0.09 0.70  0.04 4.44  0.15 4.50  0.15 2.45  0.10 0.67  0.05 2.22  0.08 1.81  0.07 4.78  0.16 5.67  0.15 0.12  0.04 0.35  0.04 0.15  0.04 0.44  0.04 0.58  0.04

11.84  0.34 19.15  0.53 1.05  0.07 19.07  1.65 2.27  0.23 1.58  0.08 7.96  0.28 12.01  0.41 5.40  0.25 4.47  0.24 11.22  0.93 11.51  0.94 9.52  0.30 8.47  0.29 17.15  1.23 9.37  0.30 13.19  1.00 6.77  0.28 7.64  0.28 6.90  0.27 16.03  1.38 15.13  0.56 1.41  0.08 15.08  1.32 2.60  0.17 1.73  0.08 no data 17.91  1.34 10.41  0.99 5.07  0.20 13.52  1.02 15.20  1.38 9.00  0.28 13.69  0.44 19.86  1.40 10.85  1.03 15.58  0.50 10.69  0.35 12.23  1.06 11.46  0.42 15.70  1.04 17.87  1.56 1.92  0.18 20.18  1.17 3.24  0.12 2.15  0.18

3.15  0.10 4.05  0.15 0.39  0.04 4.26  0.16 0.63  0.05 0.35  0.04 1.74  0.06 3.76  0.10 1.27  0.05 1.40  0.05 2.44  0.14 2.61  0.15 2.21  0.14 1.95  0.14 3.80  0.19 2.11  0.18 3.01  0.12 2.15  0.07 2.23  0.07 1.92  0.06 5.21  0.29 3.36  0.10 0.32  0.04 3.58  0.15 0.75  0.05 0.43  0.04 no data 3.15  0.20 2.22  0.11 1.76  0.05 3.04  0.16 4.38  0.22 2.31  0.14 3.63  0.22 5.51  0.23 4.09  0.22 3.58  0.13 2.97  0.16 3.23  0.09 4.16  0.23 5.10  0.24 5.88  0.16 0.60  0.04 4.71  0.14 0.98  0.04 0.39  0.04

D-3

Appendix E.1 – Metals Data from X-Ray Fluorescence in µg m Site

Date

Aluminum

CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 GZ5 GZ5 GZ5 GZ5 GZ5 GZ5

021002 021008 021014 021020 021026 021201 021207 021213 021219 021225 030301 030307 030313 030319 030325 030605 030611 030617 030623 030629 021002 021008 021014 021020 021026 021201 021207 021213 021219 021225 030301 030307 030313 030319 030325 030605 030611 030617 030623 030629 021002 021008 021014 021020 021026 021201

.1486  .0125 .4184  .0210 .1932  .0164 .2140  .0147 .1499  .0160 .0489  .0154 .1715  .0131 .1158  .0122 .1419  .0166 .0400  .0132 .2261  .0169 .0760  .0139 .2265  .0364 .0437  .0113 .0659  .0127 .2464  .0163 .1514  .0130 .0962  .0114 .1083  .0111 .0851  .0104 .1438  .0143 .3728  .0221 .0776  .0116 .0254  .0322 .1034  .0146 .1260  .0147 .0133  .0368 .0947  .0152 .0478  .0158 .1307  .0134 .0578  .0153 .0791  .0137 .1392  .0143 .0274  .0378 .0885  .0165 .1348  .0177 .1527  .0138 .1711  .0175 .0484  .0102 .0677  .0087 .1246  .0203 .3110  .0213 .4392  .0271 .1798  .0184 .1294  .0171 .2375  .0237

Silicon .4731  .0094 1.2628  .0164 .8062  .0125 .5947  .0109 .4633  .0098 .3321  .0079 .3859  .0090 .4731  .0093 .6394  .0108 .1867  .0066 .7740  .0129 .1999  .0085 1.0654  .0168 .0622  .0046 .3967  .0091 .7454  .0124 .5332  .0100 .3294  .0083 .2376  .0074 .1897  .0068 .3746  .0102 1.2282  .0170 .1811  .0074 .0880  .0062 .4358  .0108 .3878  .0102 .0843  .0067 .6158  .0124 .3519  .0108 .3101  .0092 .2675  .0107 .2558  .0096 .3275  .0106 .1945  .0090 .4535  .0127 .4019  .0125 .3804  .0106 .5748  .0136 .0459  .0055 .0295  .0046 .7483  .0144 1.1017  .0164 1.5859  .0204 .7159  .0136 .5262  .0123 1.7231  .0199

E-1

Sulfur 2.7330  .0145 3.1397  .0150 5.2043  .0213 4.7960  .0175 3.5746  .0153 .9795  .0087 4.2145  .0168 2.6126  .0132 4.7442  .0195 2.9521  .0129 7.5054  .0248 3.0822  .0161 7.5020  .0332 .7716  .0091 3.0393  .0175 8.3378  .0230 3.1424  .0142 2.4016  .0122 2.9166  .0129 2.1439  .0112 2.7897  .0149 3.3273  .0164 1.7948  .0116 2.0284  .0118 3.8564  .0172 3.8166  .0173 3.0187  .0145 2.7180  .0152 5.2222  .0192 3.0154  .0152 4.4951  .0205 2.4327  .0159 2.8241  .0167 3.2811  .0174 4.8150  .0220 6.5196  .0237 .9250  .0096 6.1889  .0233 .9198  .0096 .6023  .0080 4.9076  .0219 3.6290  .0176 7.2203  .0318 3.2560  .0206 3.7721  .0184 4.4609  .0311

Potassium 1.3265  .0097 1.2336  .0108 2.1338  .0123 1.5445  .0104 1.3152  .0098 .6886  .0069 1.5733  .0106 1.0616  .0086 1.6452  .0108 .5877  .0066 1.9270  .0120 .5079  .0077 3.8482  .0190 .5564  .0072 1.4956  .0109 2.9424  .0146 1.5912  .0106 .7732  .0077 .9713  .0085 1.0106  .0086 1.2643  .0112 1.3536  .0117 .1241  .0041 .1151  .0037 .9189  .0096 .6196  .0078 .1119  .0037 .9630  .0098 .5501  .0075 .9454  .0095 .3310  .0068 .5918  .0089 .2424  .0057 .2837  .0061 .8445  .0102 .7111  .0093 .1481  .0049 .9222  .0106 .0693  .0034 .0328  .0027 1.5931  .0128 1.4964  .0124 3.6537  .0190 .6502  .0081 1.2472  .0112 1.8709  .0134

3

Calcium .1462  .0075 .4391  .0085 .2957  .0117 .1633  .0086 .1987  .0077 .0688  .0043 .1217  .0086 .1248  .0063 .2131  .0092 .0569  .0039 .2740  .0107 .1163  .0046 .3482  .0204 .0350  .0039 .1457  .0085 .3041  .0157 .1509  .0088 .0985  .0050 .1947  .0062 .1244  .0061 .0998  .0074 .3962  .0090 .0629  .0029 .0274  .0025 .1359  .0061 .1114  .0048 .0855  .0032 .1694  .0065 .1545  .0048 .0865  .0059 .1286  .0045 .0960  .0051 .1351  .0045 .0774  .0040 .1124  .0061 .1307  .0057 .1006  .0038 .1549  .0066 .0340  .0030 .0236  .0028 .2551  .0096 .4762  .0099 .6511  .0199 .1795  .0054 .1967  .0078 .3828  .0112

Site

Date

Aluminum

GZ5 GZ5 GZ5 GZ5 GZ5 GZ5 GZ5 GZ5 GZ5 GZ5 GZ5 GZ5 GZ5 GZ5 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 TC2 TC2 TC2 TC2 TC2 TC2 TC2 TC2 TC2 TC2 TC2 TC2 TC2 TC2

021207 021213 021219 021225 030301 030307 030313 030319 030325 030605 030611 030617 030623 030629 021002 021008 021014 021020 021026 021201 021207 021213 021221 021225 030102 030301 030307 030313 030319 030325 030605 030611 030617 030623 030629 021002 021008 021014 021020 021026 021201 021207 021213 021219 021225 030301 030307 030313 030319

.3155  .0228 .0472  .0485 .1543  .0212 .1553  .0139 .3221  .0248 .0851  .0158 .1606  .0257 .0672  .0182 .2302  .0222 .1952  .0218 .5392  .0257 .5602  .0287 .1990  .0200 .1669  .0156 .2134  .0178 .4123  .0250 .1001  .0139 .2087  .0169 .1975  .0181 .0813  .0142 .1752  .0152 .2225  .0410 .3118  .0422 .0940  .0152 .2006  .0225 .2281  .0196 .0970  .0155 .2830  .0198 .0638  .0165 .2188  .0194 .1188  .0190 .0504  .0142 .1616  .0187 .0253  .0357 .0165  .0321 .0674  .0137 .3667  .0193 .0560  .0112 .0326  .0085 .0811  .0136 .0281  .0414 .0313  .0085 .0922  .0139 .0489  .0147 .0720  .0134 .1233  .0145 .1294  .0152 .0803  .0151 .0212  .0407

Silicon 1.0784  .0168 .6163  .0127 1.0012  .0161 .3592  .0089 .9767  .0183 .4352  .0123 .9773  .0189 .2063  .0095 .9482  .0180 .7553  .0161 1.2273  .0194 1.5177  .0221 .6159  .0145 .4147  .0119 .8022  .0140 1.6406  .0195 .3636  .0096 .5172  .0112 .6698  .0132 .3547  .0100 .3868  .0101 1.0646  .0307 1.0322  .0322 .4533  .0112 .8220  .0151 .5850  .0142 .4297  .0123 .7564  .0148 .4441  .0122 .7870  .0155 .4806  .0137 .3738  .0105 .5921  .0142 .1139  .0067 .0928  .0063 .3914  .0103 1.1652  .0158 .1260  .0066 .0568  .0048 .3593  .0099 .3049  .0094 .0785  .0051 .4775  .0108 .2795  .0096 .3730  .0103 .2178  .0096 .3752  .0116 .2571  .0103 .2163  .0093

E-2

Sulfur 5.6245  .0276 3.0175  .0174 6.1834  .0279 2.6693  .0164 7.8793  .0308 3.8104  .0216 7.1005  .0339 1.7195  .0146 7.0612  .0326 7.9344  .0270 5.0244  .0270 7.3795  .0351 3.2001  .0246 1.9394  .0172 3.1175  .0171 4.2942  .0199 1.8536  .0116 1.8109  .0112 3.3776  .0171 3.4527  .0162 2.4230  .0133 3.6887  .0346 3.2912  .0367 3.0868  .0163 7.3367  .0249 5.6226  .0231 2.9686  .0191 3.3267  .0185 2.6345  .0179 4.4336  .0224 6.5368  .0237 1.5938  .0124 6.4291  .0236 .9322  .0095 .6167  .0081 2.9214  .0148 3.1667  .0155 1.6738  .0109 1.2901  .0092 3.6639  .0163 3.8304  .0163 1.1554  .0091 2.4743  .0144 4.2934  .0166 3.2197  .0162 4.8587  .0207 3.3074  .0196 2.9152  .0164 3.3586  .0182

Potassium 2.6627  .0160 1.5383  .0121 1.7502  .0130 .9117  .0094 1.1305  .0119 1.0529  .0114 3.0627  .0196 .6463  .0090 2.2558  .0166 1.1596  .0117 1.7005  .0141 3.0925  .0192 1.5802  .0138 .3589  .0067 1.1761  .0108 1.7432  .0130 .3294  .0058 .1964  .0046 1.2287  .0110 .5476  .0073 .2809  .0056 1.6176  .0245 1.7452  .0254 1.0040  .0098 1.8153  .0132 .4934  .0081 .9836  .0111 .5995  .0087 .8165  .0102 1.1145  .0117 .7648  .0096 .3766  .0068 .9437  .0106 .1780  .0049 .1467  .0045 .8559  .0090 1.2956  .0111 .4364  .0065 .0800  .0034 .9328  .0094 .5245  .0071 .0650  .0032 1.1160  .0102 .4215  .0064 1.0877  .0102 .2861  .0062 .9069  .0107 .1958  .0056 .3352  .0067

Calcium .6379  .0153 .3739  .0096 .3052  .0104 .1302  .0060 .3058  .0084 .1773  .0074 .3335  .0171 .0747  .0051 .3461  .0132 .2563  .0081 .3146  .0106 .5728  .0177 .2500  .0098 .1648  .0049 .2157  .0076 .8209  .0118 .1195  .0039 .1772  .0040 .2039  .0078 .1196  .0045 .1510  .0041 .5394  .0161 .2954  .0151 .1328  .0064 .2936  .0106 .1985  .0055 .1665  .0070 .2413  .0061 .2137  .0067 .2478  .0080 .1900  .0062 .1259  .0046 .1939  .0069 .0549  .0034 .0698  .0035 .0998  .0056 .4082  .0087 .0456  .0036 .0153  .0022 .1022  .0059 .0888  .0042 .0598  .0027 .1211  .0068 .1393  .0043 .0901  .0065 .1069  .0043 .1253  .0065 .1248  .0043 .0719  .0041

Site

Date

Aluminum

TC2 TC2 TC2 TC2 TC2 TC2 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7

030325 030605 030611 030617 030623 030629 021002 021008 021014 021020 021026 021201 021207 021213 021219 021225 030301 030307 030313 030319 030325 030605 030611 030617 030623 030629 021008 021014 021020 021026 021201 021207 021213 021219 021225 030301 030307 030313 030319 030325 030605 030611 030617 030623 030629

.0844  .0150 .1096  .0182 .0673  .0136 .1629  .0183 .0427  .0094 .0346  .0099 .0786  .0136 .3087  .0222 .0636  .0109 .0259  .0304 .0960  .0162 .1659  .0135 .0513  .0099 .0649  .0152 .0769  .0162 .0980  .0134 .0289  .0446 .0283  .0406 .1021  .0149 .0433  .0136 .1169  .0179 .1452  .0165 .1045  .0138 .1046  .0181 .0502  .0098 .0279  .0282 .6394  .0261 .1813  .0155 .0807  .0105 .1277  .0139 .1181  .0166 .1513  .0137 .2852  .0183 .3770  .0204 .1859  .0189 .1990  .0149 .1751  .0157 .2081  .0159 .1057  .0162 .1845  .0150 .1757  .0145 .1081  .0114 .2025  .0157 .0596  .0086 .0369  .0077

Silicon .3238  .0113 .4193  .0134 .2782  .0093 .4987  .0135 .0151  .0154 .0438  .0050 .4031  .0104 1.3257  .0175 .1563  .0068 .0556  .0054 .4435  .0116 .3547  .0100 .0706  .0055 .5601  .0118 .4035  .0110 .3266  .0095 .2144  .0101 .2543  .0097 .3277  .0109 .2186  .0094 .5709  .0140 .4530  .0122 .3294  .0096 .4903  .0134 .0340  .0050 .0394  .0047 2.2478  .0207 .6156  .0112 .2185  .0071 .4727  .0100 .7346  .0124 .4176  .0097 1.2607  .0151 1.2545  .0154 .8584  .0130 .5454  .0108 .6346  .0114 .6710  .0117 .6486  .0116 .6473  .0113 .4207  .0100 .4115  .0085 .5871  .0113 .1407  .0056 .0679  .0046

E-3

Sulfur 4.7483  .0212 7.3606  .0253 .3859  .0067 6.4754  .0237 .7951  .0093 .6530  .0082 2.4690  .0142 3.9066  .0174 1.5535  .0109 1.4829  .0101 3.8527  .0171 3.6810  .0166 3.3172  .0146 2.8150  .0154 6.0108  .0201 3.3100  .0157 5.0713  .0212 2.5212  .0161 2.7073  .0162 2.5163  .0159 6.3593  .0256 5.6245  .0221 .5298  .0076 5.6094  .0222 .9366  .0097 .6780  .0083 6.3557  .0280 3.9676  .0155 2.0774  .0105 5.1936  .0166 6.1011  .0208 3.5935  .0151 4.7351  .0228 7.6061  .0239 4.3201  .0231 5.2831  .0177 3.6721  .0183 4.1224  .0161 3.8660  .0207 5.2221  .0193 5.5927  .0175 .7293  .0068 6.7480  .0198 1.1469  .0081 .6807  .0066

Potassium .6686  .0094 .7566  .0096 .1101  .0041 .9247  .0106 .0678  .0034 .0367  .0028 1.0158  .0100 1.4514  .0120 .1217  .0039 .1104  .0036 .9861  .0098 .5902  .0077 .0931  .0033 1.0314  .0099 .7048  .0084 1.0305  .0099 .3108  .0065 .5872  .0087 .2569  .0060 .4375  .0078 .9806  .0112 .5659  .0083 .1110  .0044 .8949  .0105 .0677  .0035 .0330  .0027 4.1689  .0178 1.0078  .0089 .2004  .0040 .6477  .0071 .9498  .0084 .5086  .0063 2.6532  .0139 1.4068  .0102 2.0836  .0123 .4484  .0087 .8667  .0082 .5583  .0068 1.2461  .0100 1.0810  .0092 .6544  .0072 .1664  .0037 .7773  .0079 .0913  .0029 .0483  .0024

Calcium .0758  .0052 .1806  .0062 .0955  .0038 .1404  .0065 .0235  .0028 .0316  .0029 .0970  .0063 .3764  .0093 .0495  .0029 .0277  .0025 .1199  .0063 .0984  .0045 .0220  .0023 .1284  .0065 .1492  .0053 .0730  .0062 .0739  .0040 .0483  .0047 .1281  .0045 .0458  .0041 .0780  .0064 .1003  .0049 .0706  .0035 .1329  .0065 .0213  .0028 .0225  .0028 .8429  .0222 .1410  .0062 .0572  .0025 .1421  .0047 .1345  .0058 .1084  .0039 .2839  .0142 .2994  .0086 .2410  .0114 .1843  .0044 .1635  .0057 .2017  .0048 .1845  .0074 .1270  .0065 .1207  .0047 .0891  .0029 .1271  .0052 .0538  .0023 .0375  .0022

Appendix E.2 – Metals Data from X-Ray Fluorescence in µg m Site

Date

CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 GZ5 GZ5 GZ5 GZ5 GZ5 GZ5

021002 021008 021014 021020 021026 021201 021207 021213 021219 021225 030301 030307 030313 030319 030325 030605 030611 030617 030623 030629 021002 021008 021014 021020 021026 021201 021207 021213 021219 021225 030301 030307 030313 030319 030325 030605 030611 030617 030623 030629 021002 021008 021014 021020 021026 021201

Titanium .0000  .0232 .0361  .0090 .0147  .0214 .0224  .0068 .0024  .0203 .0057  .0188 .0192  .0203 .0000  .0203 .0750  .0068 .0035  .0200 .0327  .0070 .0062  .0283 .0055  .0301 .0000  .0280 .0080  .0225 .0076  .0254 .0072  .0224 .0062  .0224 .0124  .0206 .0119  .0222 .0008  .0310 .0252  .0317 .0094  .0274 .0000  .0293 .0026  .0303 .0000  .0281 .0059  .0295 .0000  .0287 .0012  .0276 .0076  .0270 .0000  .0374 .0000  .0364 .0127  .0349 .0000  .0358 .0136  .0348 .0000  .0381 .0194  .0318 .0091  .0361 .0000  .0374 .0000  .0358 .0305  .0306 .0389  .0101 .0458  .0104 .0336  .0093 .0202  .0295 .0770  .0092

Vanadium

Chromium

.0000  .0136 .0037  .0115 .0022  .0092 .0234  .0030 .0000  .0087 .0005  .0081 .0205  .0030 .0000  .0087 .0160  .0031 .0020  .0086 .0311  .0030 .0020  .0115 .0000  .0170 .0000  .0161 .0014  .0091 .0218  .0050 .0158  .0043 .0004  .0090 .0224  .0029 .0232  .0031 .0073  .0189 .0000  .0193 .0062  .0118 .0037  .0126 .0068  .0131 .0025  .0121 .0381  .0044 .0002  .0124 .0102  .0120 .0066  .0116 .0080  .0218 .0000  .0148 .0084  .0142 .0078  .0146 .0099  .0142 .0000  .0225 .0068  .0129 .0066  .0207 .0000  .0216 .0068  .0146 .0335  .0045 .0197  .0044 .0582  .0047 .0387  .0042 .0138  .0043 .0399  .0042

.0023  .0037 .0094  .0008 .0045  .0006 .0060  .0007 .0054  .0006 .0100  .0006 .0282  .0010 .0071  .0006 .0183  .0008 .0127  .0007 .0146  .0009 .0232  .0010 .0143  .0017 .0016  .0046 .0237  .0009 .0087  .0015 .0065  .0013 .0137  .0007 .0079  .0007 .0069  .0007 .0005  .0051 .0000  .0053 .0010  .0020 .0000  .0020 .0010  .0023 .0023  .0007 .0002  .0026 .0021  .0022 .0015  .0021 .0019  .0021 .0008  .0062 .0016  .0026 .0021  .0025 .0015  .0025 .0047  .0009 .0000  .0064 .0017  .0022 .0012  .0058 .0000  .0060 .0018  .0024 .0016  .0027 .0017  .0024 .0051  .0011 .0054  .0010 .0026  .0008 .0075  .0012

E-4

Manganese

3

Iron

.0121  .0007 .1569  .0015 .0207  .0008 .4427  .0029 .0287  .0008 .3271  .0022 .0123  .0006 .2132  .0018 .0143  .0006 .2492  .0019 .0187  .0007 .1888  .0016 .0140  .0007 .3616  .0022 .0192  .0007 .2154  .0018 .0251  .0008 .4849  .0026 .0088  .0005 .1723  .0016 .0216  .0007 .3370  .0023 .0095  .0007 .1670  .0019 .0569  .0014 .8308  .0040 .0023  .0007 .0674  .0012 .0149  .0007 .2929  .0021 .0161  .0008 .3180  .0022 .0073  .0006 .1786  .0016 .0139  .0006 .2179  .0018 .0055  .0005 .1134  .0013 .0043  .0004 .0875  .0012 .0151  .0009 .1959  .0020 .0231  .0011 .4844  .0031 .0042  .0005 .0623  .0012 .0022  .0005 .0332  .0010 .0179  .0008 .1948  .0020 .0168  .0008 .1487  .0018 .0066  .0006 .1149  .0016 .0188  .0008 .2059  .0021 .0163  .0007 .1901  .0020 .0137  .0007 .1951  .0020 .0127  .0010 .2003  .0023 .0159  .0009 .2242  .0024 .0100  .0007 .1491  .0020 .0142  .0008 .1453  .0020 .0214  .0010 .2298  .0025 .0110  .0010 .1872  .0022 .0068  .0006 .1622  .0020 .0193  .0011 .2353  .0025 .0003  .0024 .0280  .0010 .0003  .0014 .0150  .0009 .0249  .0009 .3518  .0027 .0291  .0010 .4940  .0031 .0428  .0012 .6634  .0037 .0334  .0010 .5308  .0033 .0301  .0010 .3364  .0026 .2547  .0026 2.0712  .0065

Site

Date

GZ5 GZ5 GZ5 GZ5 GZ5 GZ5 GZ5 GZ5 GZ5 GZ5 GZ5 GZ5 GZ5 GZ5 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 TC2 TC2 TC2 TC2 TC2 TC2 TC2 TC2 TC2 TC2 TC2 TC2 TC2 TC2

021207 021213 021219 021225 030301 030307 030313 030319 030325 030605 030611 030617 030623 030629 021002 021008 021014 021020 021026 021201 021207 021213 021221 021225 030102 030301 030307 030313 030319 030325 030605 030611 030617 030623 030629 021002 021008 021014 021020 021026 021201 021207 021213 021219 021225 030301 030307 030313 030319

Titanium .0563  .0102 .0183  .0305 .0303  .0311 .0000  .0274 .0160  .0400 .0029  .0397 .0243  .0423 .0010  .0383 .0433  .0120 .0411  .0113 .0529  .0112 .0089  .0437 .0425  .0121 .0224  .0329 .0267  .0285 .0381  .0104 .0127  .0293 .0126  .0275 .0307  .0102 .0127  .0264 .0166  .0275 .0299  .1078 .0613  .1008 .0218  .0280 .0373  .0093 .0106  .0350 .0113  .0370 .0232  .0387 .0379  .0120 .0060  .0379 .0215  .0349 .0132  .0371 .0000  .0393 .0000  .0380 .0060  .0336 .0153  .0266 .0378  .0093 .0000  .0284 .0024  .0262 .0107  .0271 .0137  .0252 .0035  .0264 .0160  .0284 .0103  .0298 .0170  .0270 .0000  .0415 .0000  .0371 .0095  .0375 .0000  .0413

Vanadium

Chromium

Manganese

.0499  .0063 .0134  .0184 .0308  .0064 .0024  .0118 .0472  .0077 .0000  .0228 .0280  .0083 .0031  .0219 .0432  .0051 .0379  .0048 .0975  .0050 .0623  .0085 .0585  .0052 .0383  .0046 .0304  .0042 .0554  .0066 .0133  .0042 .0107  .0119 .0242  .0045 .0093  .0114 .0216  .0040 .0097  .0617 .0485  .0138 .0274  .0041 .0223  .0041 .0319  .0049 .0096  .0215 .0147  .0224 .0289  .0050 .0264  .0053 .0177  .0048 .0065  .0218 .0274  .0075 .0000  .0218 .0028  .0136 .0123  .0039 .0051  .0120 .0055  .0122 .0031  .0112 .0121  .0039 .0109  .0109 .0083  .0114 .0094  .0123 .0120  .0129 .0095  .0116 .0000  .0231 .0095  .0153 .0134  .0153 .0000  .0229

.0064  .0018 .0000  .0050 .0021  .0054 .0005  .0020 .0024  .0067 .0000  .0064 .0076  .0024 .0003  .0061 .0057  .0011 .0044  .0010 .0070  .0015 .0042  .0076 .0046  .0012 .0008  .0027 .0025  .0026 .0008  .0057 .0003  .0021 .0010  .0020 .0020  .0026 .0016  .0020 .0013  .0023 .0009  .0172 .0107  .0026 .0033  .0008 .0042  .0008 .0027  .0029 .0036  .0062 .0017  .0063 .0037  .0010 .0036  .0010 .0027  .0009 .0000  .0062 .0023  .0063 .0000  .0060 .0003  .0021 .0023  .0007 .0009  .0021 .0000  .0019 .0002  .0018 .0013  .0021 .0022  .0007 .0006  .0019 .0018  .0022 .0019  .0023 .0011  .0021 .0000  .0062 .0028  .0009 .0000  .0025 .0000  .0062

.0261  .0011 .0245  .0010 .0374  .0012 .0174  .0008 .0368  .0014 .0264  .0013 .0633  .0017 .0175  .0011 .0493  .0014 .0265  .0010 .0385  .0012 .0460  .0016 .0438  .0013 .0197  .0009 .0233  .0009 .0340  .0012 .0079  .0006 .0070  .0006 .0216  .0009 .0167  .0007 .0100  .0006 .0471  .0031 .0475  .0025 .0206  .0008 .0391  .0010 .0263  .0010 .0262  .0012 .0340  .0013 .0273  .0011 .0377  .0012 .0174  .0009 .0233  .0012 .0350  .0013 .0075  .0010 .0108  .0007 .0144  .0007 .0235  .0009 .0027  .0005 .0020  .0004 .0178  .0008 .0135  .0007 .0060  .0005 .0175  .0008 .0133  .0007 .0171  .0007 .0043  .0009 .0213  .0010 .0073  .0007 .0102  .0010

E-5

Iron .3956  .0028 .3193  .0025 .4669  .0030 .2169  .0021 .6447  .0040 .3486  .0030 .6188  .0040 .1756  .0022 .5710  .0038 .3628  .0030 .5164  .0035 .8291  .0045 .5435  .0036 .1756  .0021 .3326  .0026 .5919  .0034 .1176  .0016 .1663  .0018 .2719  .0024 .1540  .0018 .1658  .0018 .5019  .0062 .5114  .0063 .2099  .0021 .4174  .0028 .3422  .0029 .2620  .0026 .3817  .0031 .2613  .0026 .4292  .0033 .2693  .0026 .2691  .0026 .3460  .0029 .0851  .0016 .0938  .0016 .2015  .0020 .5068  .0031 .0509  .0011 .0207  .0008 .1864  .0019 .1288  .0016 .0923  .0014 .1999  .0020 .1626  .0018 .2129  .0020 .1537  .0021 .2685  .0027 .1369  .0020 .1751  .0022

Site

Date

TC2 TC2 TC2 TC2 TC2 TC2 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7

030325 030605 030611 030617 030623 030629 021002 021008 021014 021020 021026 021201 021207 021213 021219 021225 030301 030307 030313 030319 030325 030605 030611 030617 030623 030629 021008 021014 021020 021026 021201 021207 021213 021219 021225 030301 030307 030313 030319 030325 030605 030611 030617 030623 030629

Titanium .0124  .0364 .0000  .0359 .0132  .0357 .0000  .0377 .0002  .0332 .0000  .0364 .0107  .0294 .0400  .0093 .0036  .0290 .0000  .0287 .0157  .0306 .0109  .0265 .0030  .0271 .0000  .0279 .0127  .0267 .0098  .0260 .0000  .0378 .0013  .0360 .0000  .0416 .0113  .0348 .0000  .0370 .0000  .0332 .0125  .0344 .0078  .0392 .0010  .0346 .0013  .0328 .0912  .0079 .0343  .0078 .0133  .0214 .0158  .0215 .0121  .0238 .0098  .0224 .0615  .0077 .0219  .0254 .0339  .0076 .0000  .0277 .0326  .0079 .0204  .0254 .0241  .0074 .0191  .0252 .0222  .0234 .0156  .0227 .0000  .0248 .0011  .0226 .0000  .0247

Vanadium

Chromium

Manganese

.0221  .0050 .0059  .0146 .0022  .0145 .0000  .0217 .0022  .0134 .0000  .0208 .0139  .0043 .0249  .0041 .0054  .0125 .0004  .0124 .0270  .0045 .0050  .0114 .0206  .0039 .0123  .0040 .0141  .0039 .0178  .0038 .0124  .0216 .0051  .0146 .0000  .0236 .0350  .0049 .0000  .0216 .0095  .0136 .0049  .0140 .0071  .0226 .0040  .0140 .0057  .0133 .0677  .0038 .0447  .0035 .0137  .0031 .0115  .0031 .0745  .0049 .0436  .0047 .0619  .0036 .0440  .0053 .0308  .0034 .0355  .0052 .0212  .0046 .0436  .0050 .0301  .0031 .0479  .0036 .0259  .0033 .0114  .0031 .0266  .0047 .0029  .0131 .0007  .0142

.0018  .0027 .0011  .0025 .0012  .0023 .0000  .0061 .0010  .0021 .0000  .0058 .0004  .0022 .0020  .0024 .0000  .0020 .0005  .0020 .0006  .0025 .0017  .0020 .0008  .0021 .0011  .0021 .0025  .0007 .0005  .0021 .0001  .0060 .0017  .0025 .0000  .0065 .0010  .0028 .0000  .0061 .0015  .0024 .0010  .0023 .0000  .0063 .0005  .0022 .0014  .0023 .0223  .0012 .0283  .0011 .0036  .0006 .0012  .0017 .0088  .0016 .0036  .0042 .0082  .0010 .0035  .0047 .0104  .0009 .0858  .0020 .0056  .0013 .0152  .0016 .0251  .0010 .0080  .0009 .0098  .0008 .0022  .0006 .0054  .0014 .0012  .0037 .0079  .0014

.0159  .0009 .0126  .0008 .0038  .0006 .0183  .0011 .0010  .0014 .0003  .0023 .0123  .0007 .0234  .0009 .0021  .0005 .0024  .0005 .0160  .0008 .0157  .0007 .0014  .0004 .0160  .0007 .0155  .0007 .0124  .0007 .0065  .0009 .0130  .0008 .0031  .0010 .0099  .0007 .0180  .0011 .0098  .0007 .0038  .0006 .0176  .0011 .0009  .0014 .0007  .0015 .0444  .0011 .0181  .0008 .0056  .0004 .0143  .0006 .0291  .0009 .0075  .0006 .0315  .0009 .0399  .0011 .0251  .0008 .0212  .0013 .0193  .0008 .0216  .0009 .0195  .0008 .0238  .0008 .0111  .0006 .0060  .0005 .0141  .0008 .0027  .0005 .0016  .0017

E-6

Iron .1843  .0022 .1939  .0023 .1262  .0019 .2167  .0024 .0132  .0008 .0082  .0008 .1662  .0019 .4653  .0030 .0322  .0010 .0300  .0009 .2012  .0021 .1242  .0016 .0189  .0008 .1736  .0019 .1405  .0017 .1644  .0018 .0882  .0016 .1965  .0023 .1062  .0018 .1592  .0021 .2272  .0025 .1599  .0020 .1143  .0018 .1886  .0022 .0087  .0008 .0061  .0008 .8487  .0036 .3111  .0022 .0823  .0012 .1501  .0016 .3045  .0021 .1260  .0014 .3907  .0024 .4927  .0027 .3810  .0024 .5601  .0030 .2924  .0022 .3358  .0023 .3394  .0023 .2994  .0022 .2211  .0019 .1638  .0016 .2433  .0020 .0540  .0010 .0555  .0010

Appendix E.3 – Metals Data from X-Ray Fluorescence in µg m Site

Date

CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 GZ5 GZ5 GZ5 GZ5 GZ5 GZ5

021002 021008 021014 021020 021026 021201 021207 021213 021219 021225 030301 030307 030313 030319 030325 030605 030611 030617 030623 030629 021002 021008 021014 021020 021026 021201 021207 021213 021219 021225 030301 030307 030313 030319 030325 030605 030611 030617 030623 030629 021002 021008 021014 021020 021026 021201

Cobalt .0002  .0026 .0030  .0068 .0014  .0051 .0013  .0034 .0004  .0039 .0012  .0030 .0023  .0056 .0013  .0034 .0028  .0074 .0011  .0028 .0017  .0053 .0016  .0029 .0053  .0127 .0011  .0015 .0021  .0047 .0017  .0050 .0009  .0029 .0021  .0035 .0012  .0020 .0002  .0017 .0015  .0033 .0013  .0075 .0015  .0005 .0008  .0012 .0007  .0033 .0011  .0026 .0013  .0022 .0013  .0034 .0003  .0032 .0011  .0032 .0025  .0035 .0013  .0039 .0027  .0028 .0020  .0028 .0023  .0039 .0010  .0033 .0008  .0029 .0030  .0040 .0006  .0013 .0004  .0012 .0016  .0056 .0026  .0076 .0023  .0101 .0007  .0082 .0037  .0053 .0097  .0308

Nickle .0093  .0004 .0146  .0005 .0062  .0003 .0154  .0004 .0077  .0003 .0110  .0004 .0299  .0006 .0090  .0003 .0344  .0006 .0150  .0004 .0289  .0006 .0331  .0007 .0185  .0006 .0099  .0004 .0322  .0006 .0328  .0006 .0183  .0005 .0250  .0005 .0260  .0005 .0174  .0005 .0054  .0004 .0009  .0010 .0021  .0003 .0026  .0003 .0023  .0004 .0025  .0003 .0095  .0005 .0031  .0003 .0046  .0004 .0014  .0003 .0092  .0005 .0029  .0004 .0041  .0004 .0039  .0004 .0043  .0004 .0043  .0004 .0035  .0004 .0060  .0005 .0013  .0003 .0026  .0004 .0068  .0004 .0028  .0004 .0212  .0006 .0117  .0005 .0046  .0004 .0109  .0006

E-7

Copper .0697  .0008 .0146  .0005 .0464  .0007 .0148  .0004 .0442  .0006 .0592  .0007 .0164  .0005 .0374  .0006 .0697  .0008 .0847  .0008 .0224  .0005 .0463  .0008 .2358  .0017 .0713  .0009 .0884  .0009 .0222  .0005 .0161  .0004 .0264  .0006 .0085  .0004 .0065  .0004 .0262  .0006 .0081  .0005 .0055  .0004 .0044  .0004 .0087  .0005 .0155  .0005 .0134  .0005 .0131  .0005 .0149  .0005 .0083  .0005 .0157  .0006 .0111  .0006 .0054  .0005 .0080  .0005 .0164  .0006 .0136  .0006 .0084  .0005 .0171  .0006 .0061  .0005 .0053  .0005 .0273  .0007 .0220  .0006 .0426  .0008 .0233  .0007 .0607  .0009 .0919  .0011

Zinc .2381  .0013 .1251  .0011 .6062  .0021 .3600  .0016 .3214  .0015 .2269  .0012 .1725  .0011 .2741  .0014 .3479  .0015 .2018  .0012 .3696  .0016 .1204  .0012 .9591  .0030 .1814  .0013 .5147  .0019 .2930  .0014 .2234  .0013 .2768  .0014 .0933  .0008 .0830  .0008 .3124  .0018 .2714  .0016 .0419  .0007 .0297  .0007 .3570  .0019 .2560  .0016 .0780  .0009 .2897  .0017 .1849  .0014 .3323  .0018 .1046  .0012 .3324  .0020 .1396  .0014 .1521  .0014 .3182  .0020 .1614  .0014 .0769  .0010 .3098  .0019 .0259  .0007 .0113  .0006 .4161  .0020 .2029  .0014 .7493  .0027 1.1798  .0034 .2926  .0017 .7365  .0026

3

Gallium .0000  .0032 .0000  .0034 .0000  .0043 .0001  .0032 .0000  .0030 .0000  .0026 .0000  .0032 .0000  .0029 .0000  .0039 .0000  .0027 .0045  .0014 .0000  .0036 .0019  .0066 .0000  .0033 .0000  .0040 .0029  .0037 .0000  .0029 .0000  .0029 .0000  .0026 .0000  .0026 .0000  .0036 .0000  .0037 .0006  .0032 .0000  .0033 .0000  .0039 .0000  .0035 .0000  .0036 .0000  .0035 .0000  .0034 .0000  .0034 .0000  .0040 .0000  .0043 .0000  .0042 .0010  .0043 .0000  .0046 .0000  .0041 .0000  .0036 .0000  .0040 .0000  .0039 .0000  .0040 .0000  .0046 .0000  .0040 .0012  .0065 .0018  .0049 .0000  .0041 .0045  .0069

Site

Date

GZ5 GZ5 GZ5 GZ5 GZ5 GZ5 GZ5 GZ5 GZ5 GZ5 GZ5 GZ5 GZ5 GZ5 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 TC2 TC2 TC2 TC2 TC2 TC2 TC2 TC2 TC2 TC2 TC2 TC2 TC2 TC2

021207 021213 021219 021225 030301 030307 030313 030319 030325 030605 030611 030617 030623 030629 021002 021008 021014 021020 021026 021201 021207 021213 021221 021225 030102 030301 030307 030313 030319 030325 030605 030611 030617 030623 030629 021002 021008 021014 021020 021026 021201 021207 021213 021219 021225 030301 030307 030313 030319

Cobalt .0026  .0062 .0011  .0050 .0052  .0072 .0011  .0036 .0054  .0101 .0012  .0056 .0037  .0097 .0011  .0031 .0058  .0090 .0027  .0059 .0024  .0081 .0039  .0128 .0039  .0085 .0012  .0031 .0018  .0052 .0039  .0090 .0011  .0022 .0011  .0028 .0010  .0044 .0004  .0026 .0011  .0028 .0045  .0090 .0056  .0092 .0000  .0035 .0018  .0065 .0009  .0055 .0013  .0044 .0021  .0061 .0035  .0044 .0012  .0068 .0001  .0044 .0012  .0044 .0030  .0056 .0017  .0019 .0007  .0020 .0022  .0034 .0026  .0078 .0010  .0013 .0002  .0010 .0027  .0032 .0021  .0023 .0014  .0018 .0009  .0033 .0008  .0028 .0012  .0035 .0016  .0029 .0035  .0045 .0012  .0027 .0025  .0032

Nickle .0176  .0006 .0051  .0004 .0127  .0005 .0033  .0003 .0207  .0007 .0015  .0004 .0179  .0007 .0026  .0004 .0185  .0007 .0144  .0006 .0281  .0008 .0202  .0007 .0131  .0006 .0087  .0005 .0078  .0004 .0155  .0005 .0036  .0004 .0032  .0003 .0062  .0004 .0030  .0003 .0074  .0004 .0089  .0012 .0145  .0013 .0077  .0004 .0072  .0004 .0128  .0005 .0064  .0005 .0090  .0005 .0098  .0005 .0110  .0006 .0064  .0005 .0151  .0006 .0153  .0006 .0018  .0004 .0004  .0010 .0035  .0004 .0007  .0009 .0023  .0003 .0014  .0003 .0031  .0003 .0036  .0003 .0018  .0003 .0023  .0003 .0050  .0004 .0032  .0003 .0044  .0005 .0072  .0005 .0040  .0005 .0029  .0004

E-8

Copper .0320  .0007 .0242  .0006 .0447  .0008 .0624  .0009 .0416  .0009 .0399  .0009 .0930  .0013 .0776  .0011 .0446  .0009 .0179  .0007 .0308  .0008 .0560  .0010 .0199  .0007 .0133  .0006 .0293  .0007 .0184  .0006 .0065  .0005 .0028  .0004 .0126  .0006 .0116  .0005 .0103  .0005 .0241  .0016 .0275  .0016 .0067  .0005 .0267  .0006 .0196  .0007 .0085  .0006 .0122  .0006 .0165  .0007 .0196  .0007 .0076  .0005 .0085  .0005 .0127  .0006 .0023  .0004 .0013  .0004 .0235  .0006 .0120  .0005 .0097  .0005 .0131  .0005 .0238  .0006 .0138  .0005 .0153  .0005 .0151  .0005 .0159  .0006 .0135  .0005 .0197  .0007 .0148  .0007 .0115  .0006 .0167  .0006

Zinc .5322  .0022 .2766  .0016 .6318  .0024 .2486  .0015 .8373  .0032 .3183  .0020 .8010  .0031 .2678  .0018 .6055  .0027 .4979  .0024 1.2073  .0037 1.1344  .0036 .8393  .0031 .3296  .0020 .4364  .0020 .4933  .0022 .1068  .0011 .0780  .0009 .4652  .0021 .2041  .0014 .1689  .0013 .6778  .0049 .7306  .0052 .3621  .0018 .3728  .0019 .1847  .0015 .3960  .0022 .3210  .0020 .3481  .0021 .5532  .0026 .1718  .0015 .1048  .0012 .2890  .0019 .0253  .0007 .0112  .0005 .3115  .0017 .2838  .0016 .0303  .0006 .0135  .0005 .4084  .0019 .2101  .0014 .0255  .0006 .2735  .0016 .1016  .0010 .3921  .0019 .0572  .0009 .4213  .0023 .0955  .0012 .2080  .0016

Gallium .0009  .0057 .0000  .0039 .0000  .0057 .0000  .0039 .0039  .0055 .0000  .0049 .0003  .0069 .0025  .0045 .0032  .0064 .0000  .0046 .0152  .0019 .0000  .0073 .0000  .0060 .0000  .0047 .0000  .0040 .0000  .0041 .0000  .0034 .0000  .0032 .0000  .0041 .0029  .0033 .0000  .0033 .0022  .0115 .0101  .0125 .0000  .0038 .0000  .0042 .0000  .0041 .0054  .0015 .0005  .0042 .0000  .0044 .0031  .0050 .0009  .0041 .0028  .0039 .0000  .0043 .0020  .0040 .0007  .0038 .0000  .0035 .0000  .0036 .0000  .0033 .0000  .0030 .0000  .0036 .0030  .0032 .0000  .0031 .0000  .0037 .0000  .0036 .0000  .0036 .0000  .0041 .0000  .0046 .0000  .0044 .0000  .0044

Site

Date

TC2 TC2 TC2 TC2 TC2 TC2 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7

030325 030605 030611 030617 030623 030629 021002 021008 021014 021020 021026 021201 021207 021213 021219 021225 030301 030307 030313 030319 030325 030605 030611 030617 030623 030629 021008 021014 021020 021026 021201 021207 021213 021219 021225 030301 030307 030313 030319 030325 030605 030611 030617 030623 030629

Cobalt .0019  .0033 .0014  .0034 .0004  .0024 .0024  .0037 .0004  .0011 .0002  .0011 .0009  .0029 .0019  .0072 .0005  .0012 .0008  .0011 .0013  .0034 .0016  .0023 .0006  .0010 .0006  .0029 .0014  .0025 .0010  .0028 .0005  .0019 .0013  .0035 .0007  .0022 .0004  .0029 .0006  .0039 .0019  .0029 .0010  .0023 .0008  .0034 .0004  .0011 .0003  .0011 .0053  .0128 .0027  .0049 .0007  .0016 .0011  .0025 .0012  .0048 .0012  .0022 .0033  .0060 .0025  .0075 .0025  .0059 .0073  .0086 .0015  .0046 .0026  .0053 .0045  .0054 .0018  .0048 .0019  .0036 .0009  .0027 .0023  .0039 .0004  .0012 .0009  .0013

Nickle .0070  .0005 .0038  .0004 .0008  .0011 .0124  .0006 .0013  .0003 .0010  .0010 .0042  .0004 .0057  .0004 .0012  .0003 .0025  .0003 .0073  .0004 .0021  .0003 .0057  .0004 .0050  .0004 .0058  .0004 .0047  .0004 .0107  .0005 .0016  .0004 .0064  .0005 .0112  .0005 .0078  .0005 .0045  .0004 .0016  .0004 .0060  .0005 .0025  .0004 .0015  .0004 .0322  .0006 .0342  .0006 .0054  .0003 .0047  .0003 .0319  .0006 .0191  .0005 .0172  .0005 .0211  .0005 .0192  .0005 .1125  .0011 .0166  .0005 .0292  .0006 .0375  .0007 .0236  .0006 .0231  .0005 .0044  .0003 .0166  .0005 .0039  .0003 .0107  .0004

E-9

Copper .0193  .0007 .0174  .0006 .0116  .0006 .0199  .0007 .0084  .0005 .0082  .0005 .0128  .0005 .0047  .0004 .0000  .0011 .0016  .0004 .0042  .0005 .0097  .0005 .0017  .0003 .0090  .0005 .0077  .0004 .0039  .0004 .0069  .0005 .0035  .0005 .0033  .0005 .0066  .0005 .0160  .0006 .0062  .0005 .0014  .0004 .0064  .0006 .0009  .0012 .0017  .0004 .0279  .0006 .0110  .0004 .0043  .0003 .0061  .0004 .0390  .0007 .0128  .0004 .0241  .0006 .0394  .0007 .0281  .0006 .0192  .0006 .0491  .0007 .0180  .0005 .0452  .0007 .0194  .0005 .0096  .0004 .0031  .0003 .0169  .0005 .0030  .0003 .0024  .0003

Zinc .2210  .0017 .1626  .0014 .0104  .0006 .2587  .0018 .0021  .0004 .0054  .0005 .2958  .0017 .2778  .0016 .0194  .0006 .0168  .0005 .4250  .0020 .2282  .0015 .0193  .0005 .2623  .0016 .1265  .0011 .2990  .0017 .0571  .0009 .3862  .0022 .1236  .0013 .2752  .0019 .3810  .0022 .1205  .0012 .0162  .0006 .2617  .0018 .0022  .0004 .0014  .0005 .5330  .0020 .2071  .0013 .0478  .0006 .1443  .0011 .8571  .0025 .2224  .0013 .4981  .0019 1.0221  .0027 .5911  .0020 .1488  .0011 .3023  .0015 .2771  .0015 .4472  .0018 .3744  .0017 .1486  .0011 .0417  .0006 .4574  .0018 .0312  .0005 .0072  .0004

Gallium .0037  .0046 .0000  .0044 .0023  .0043 .0000  .0041 .0007  .0038 .0000  .0037 .0000  .0038 .0000  .0036 .0000  .0033 .0000  .0033 .0000  .0040 .0000  .0033 .0016  .0030 .0000  .0037 .0000  .0033 .0027  .0034 .0000  .0040 .0014  .0045 .0000  .0041 .0030  .0041 .0000  .0047 .0000  .0039 .0000  .0039 .0000  .0043 .0004  .0040 .0023  .0037 .0000  .0057 .0000  .0031 .0000  .0025 .0000  .0027 .0000  .0036 .0000  .0029 .0000  .0048 .0000  .0043 .0000  .0049 .0004  .0028 .0032  .0036 .0000  .0029 .0028  .0042 .0000  .0035 .0000  .0029 .0000  .0027 .0000  .0030 .0000  .0024 .0000  .0025

Appendix E.4 – Metals Data from X-Ray Fluorescence in µg m Site

Date

CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 GZ5 GZ5 GZ5 GZ5 GZ5 GZ5

021002 021008 021014 021020 021026 021201 021207 021213 021219 021225 030301 030307 030313 030319 030325 030605 030611 030617 030623 030629 021002 021008 021014 021020 021026 021201 021207 021213 021219 021225 030301 030307 030313 030319 030325 030605 030611 030617 030623 030629 021002 021008 021014 021020 021026 021201

Arsenic

Selenium

.0065  .0267 .0050  .0145 .0305  .0436 .0091  .0268 .0123  .0235 .0106  .0155 .0130  .0283 .0114  .0210 .0171  .0390 .0053  .0158 .0119  .0481 .0045  .0167 .0360  .0791 .0051  .0165 .0082  .0386 .0125  .0327 .0139  .0205 .0090  .0157 .0010  .0126 .0025  .0098 .0043  .0164 .0042  .0146 .0002  .0033 .0032  .0010 .0067  .0159 .0066  .0125 .0002  .0034 .0105  .0184 .0011  .0108 .0073  .0143 .0000  .0060 .0053  .0140 .0000  .0064 .0033  .0076 .0078  .0235 .0001  .0097 .0002  .0031 .0059  .0130 .0004  .0031 .0000  .0033 .0438  .0111 .0335  .0071 .0268  .0691 .0316  .0412 .0366  .0089 .0617  .0779

.0011  .0003 .0022  .0004 .0037  .0004 .0056  .0004 .0029  .0003 .0027  .0003 .0035  .0004 .0032  .0003 .0055  .0004 .0014  .0003 .0060  .0004 .0020  .0004 .0061  .0006 .0010  .0011 .0023  .0004 .0082  .0005 .0069  .0004 .0019  .0004 .0016  .0003 .0014  .0003 .0020  .0004 .0016  .0004 .0009  .0011 .0003  .0011 .0015  .0004 .0047  .0004 .0005  .0012 .0029  .0004 .0033  .0004 .0027  .0004 .0017  .0005 .0002  .0015 .0013  .0014 .0021  .0005 .0043  .0005 .0030  .0005 .0007  .0012 .0031  .0005 .0007  .0014 .0008  .0014 .0106  .0006 .0093  .0005 .0151  .0007 .0074  .0005 .0068  .0005 .0096  .0006

E-10

Bromine .0021  .0023 .0170  .0007 .0196  .0014 .0039  .0008 .0617  .0010 .0953  .0010 .0213  .0009 .0288  .0008 .0699  .0014 .0644  .0009 .0155  .0013 .0478  .0010 .2502  .0027 .0654  .0010 .0396  .0012 .0123  .0010 .0000  .0019 .0115  .0006 .0072  .0005 .0043  .0005 .0057  .0006 .0046  .0006 .0091  .0004 .0010  .0010 .0040  .0006 .0151  .0006 .0058  .0004 .0036  .0007 .0118  .0006 .0032  .0005 .0106  .0006 .0032  .0006 .0163  .0006 .0087  .0006 .0073  .0008 .0113  .0006 .0041  .0004 .0065  .0007 .0034  .0004 .0013  .0004 .0172  .0013 .0211  .0010 .0206  .0020 .0060  .0013 .0296  .0012 .0573  .0025

Rubidium .0110  .0004 .0063  .0005 .0197  .0006 .0148  .0004 .0110  .0009 .0076  .0012 .0161  .0005 .0104  .0005 .0163  .0010 .0043  .0009 .0195  .0006 .0044  .0008 .0361  .0032 .0035  .0009 .0160  .0007 .0323  .0006 .0165  .0005 .0072  .0005 .0099  .0004 .0098  .0004 .0072  .0005 .0063  .0005 .0006  .0013 .0016  .0004 .0078  .0005 .0039  .0005 .0013  .0013 .0090  .0005 .0040  .0005 .0077  .0005 .0010  .0016 .0055  .0006 .0009  .0017 .0016  .0017 .0066  .0006 .0031  .0006 .0001  .0014 .0045  .0005 .0008  .0016 .0006  .0016 .0148  .0006 .0099  .0006 .0370  .0008 .0113  .0005 .0100  .0007 .0212  .0010

3

Strontium .0050  .0004 .0018  .0005 .0003  .0012 .0022  .0004 .0006  .0011 .0005  .0010 .0010  .0011 .0009  .0011 .0004  .0011 .0003  .0011 .0046  .0004 .0006  .0014 .0022  .0005 .0003  .0013 .0000  .0012 .0039  .0004 .0012  .0004 .0005  .0013 .0014  .0004 .0013  .0004 .0131  .0006 .0021  .0005 .0008  .0014 .0000  .0016 .0000  .0017 .0013  .0015 .0008  .0017 .0010  .0015 .0017  .0005 .0000  .0014 .0025  .0006 .0002  .0019 .0018  .0018 .0004  .0019 .0020  .0006 .0011  .0018 .0019  .0005 .0022  .0006 .0000  .0018 .0008  .0019 .0005  .0017 .0017  .0017 .0052  .0006 .0025  .0005 .0000  .0016 .0024  .0005

Site

Date

GZ5 GZ5 GZ5 GZ5 GZ5 GZ5 GZ5 GZ5 GZ5 GZ5 GZ5 GZ5 GZ5 GZ5 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 TC2 TC2 TC2 TC2 TC2 TC2 TC2 TC2 TC2 TC2 TC2 TC2 TC2 TC2

021207 021213 021219 021225 030301 030307 030313 030319 030325 030605 030611 030617 030623 030629 021002 021008 021014 021020 021026 021201 021207 021213 021221 021225 030102 030301 030307 030313 030319 030325 030605 030611 030617 030623 030629 021002 021008 021014 021020 021026 021201 021207 021213 021219 021225 030301 030307 030313 030319

Arsenic .0372  .0580 .0183  .0289 .0422  .0557 .0158  .0274 .0172  .0434 .0333  .0109 .0553  .0675 .0193  .0197 .0345  .0638 .0110  .0250 .0427  .0521 .0618  .0767 .0344  .0561 .0244  .0338 .0128  .0251 .0140  .0287 .0036  .0055 .0000  .0037 .0119  .0223 .0040  .0112 .0024  .0076 .0146  .0376 .0183  .0546 .0113  .0208 .0197  .0307 .0007  .0099 .0134  .0269 .0040  .0120 .0073  .0200 .0112  .0280 .0052  .0094 .0024  .0063 .0070  .0147 .0015  .0034 .0000  .0032 .0054  .0174 .0077  .0150 .0016  .0046 .0000  .0027 .0067  .0171 .0034  .0110 .0018  .0025 .0112  .0184 .0019  .0072 .0067  .0201 .0007  .0050 .0080  .0257 .0000  .0055 .0005  .0105

Selenium .0070  .0006 .0054  .0005 .0115  .0006 .0034  .0004 .0086  .0007 .0055  .0006 .0102  .0008 .0037  .0005 .0138  .0008 .0084  .0006 .0285  .0008 .0494  .0011 .0083  .0007 .0053  .0005 .0041  .0005 .0054  .0005 .0012  .0004 .0015  .0004 .0030  .0005 .0039  .0004 .0008  .0011 .0063  .0014 .0078  .0014 .0017  .0004 .0070  .0005 .0028  .0005 .0034  .0005 .0018  .0005 .0022  .0005 .0053  .0006 .0028  .0005 .0005  .0013 .0036  .0005 .0000  .0013 .0001  .0013 .0031  .0004 .0029  .0004 .0006  .0011 .0007  .0010 .0022  .0004 .0033  .0004 .0000  .0010 .0024  .0004 .0016  .0004 .0024  .0004 .0006  .0014 .0030  .0005 .0001  .0015 .0014  .0014

E-11

Bromine .0189  .0018 .0154  .0010 .0348  .0018 .0498  .0011 .0195  .0014 .0069  .0012 .0881  .0023 .1071  .0015 .0146  .0019 .0115  .0009 .0176  .0017 .0100  .0023 .0196  .0018 .0013  .0033 .0227  .0009 .0393  .0011 .0095  .0005 .0019  .0003 .0084  .0008 .0120  .0006 .0066  .0005 .0104  .0016 .0644  .0025 .0090  .0008 .0652  .0013 .0192  .0007 .0059  .0009 .0210  .0008 .0248  .0009 .0168  .0010 .0116  .0007 .0090  .0005 .0087  .0007 .0025  .0004 .0017  .0004 .0103  .0007 .0050  .0006 .0085  .0005 .0009  .0003 .0043  .0007 .0133  .0006 .0034  .0003 .0034  .0007 .0091  .0005 .0035  .0007 .0077  .0005 .0037  .0009 .0128  .0006 .0071  .0006

Rubidium .0254  .0007 .0150  .0006 .0162  .0008 .0082  .0008 .0092  .0007 .0118  .0007 .0288  .0015 .0045  .0015 .0231  .0008 .0080  .0006 .0162  .0007 .0343  .0010 .0175  .0008 .0029  .0005 .0104  .0006 .0115  .0007 .0016  .0004 .0031  .0004 .0106  .0006 .0035  .0004 .0034  .0004 .0137  .0016 .0191  .0019 .0094  .0005 .0129  .0010 .0020  .0006 .0106  .0006 .0028  .0006 .0069  .0007 .0076  .0007 .0053  .0006 .0022  .0005 .0067  .0006 .0010  .0015 .0007  .0015 .0071  .0005 .0087  .0005 .0033  .0005 .0011  .0011 .0077  .0005 .0035  .0004 .0002  .0012 .0090  .0005 .0028  .0005 .0093  .0005 .0010  .0017 .0072  .0006 .0000  .0018 .0017  .0017

Strontium .0053  .0006 .0021  .0005 .0017  .0017 .0000  .0015 .0044  .0007 .0000  .0019 .0028  .0007 .0000  .0018 .0023  .0007 .0038  .0006 .0064  .0006 .0043  .0008 .0025  .0007 .0015  .0017 .0012  .0016 .0023  .0005 .0009  .0016 .0015  .0005 .0007  .0018 .0009  .0014 .0011  .0015 .0003  .0049 .0008  .0051 .0002  .0016 .0026  .0005 .0016  .0018 .0000  .0017 .0029  .0006 .0005  .0018 .0003  .0020 .0026  .0006 .0012  .0018 .0014  .0019 .0004  .0018 .0011  .0018 .0007  .0015 .0022  .0005 .0005  .0016 .0002  .0014 .0003  .0015 .0011  .0013 .0000  .0014 .0000  .0016 .0002  .0016 .0004  .0014 .0015  .0019 .0003  .0019 .0015  .0021 .0000  .0018

Site

Date

TC2 TC2 TC2 TC2 TC2 TC2 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7

030325 030605 030611 030617 030623 030629 021002 021008 021014 021020 021026 021201 021207 021213 021219 021225 030301 030307 030313 030319 030325 030605 030611 030617 030623 030629 021008 021014 021020 021026 021201 021207 021213 021219 021225 030301 030307 030313 030319 030325 030605 030611 030617 030623 030629

Arsenic .0067  .0181 .0060  .0105 .0009  .0034 .0055  .0133 .0000  .0030 .0013  .0030 .0084  .0185 .0041  .0169 .0017  .0030 .0027  .0029 .0065  .0187 .0026  .0127 .0000  .0039 .0088  .0198 .0068  .0121 .0056  .0161 .0007  .0057 .0076  .0146 .0023  .0065 .0033  .0120 .0123  .0309 .0044  .0084 .0007  .0032 .0038  .0132 .0000  .0031 .0000  .0030 .0182  .0645 .0069  .0161 .0008  .0035 .0040  .0099 .0200  .0302 .0091  .0188 .0261  .0507 .0225  .0387 .0214  .0532 .0074  .0025 .0146  .0340 .0046  .0121 .0148  .0438 .0089  .0265 .0003  .0084 .0004  .0037 .0083  .0199 .0000  .0025 .0004  .0022

Selenium .0022  .0005 .0024  .0005 .0004  .0014 .0046  .0005 .0001  .0012 .0004  .0013 .0024  .0004 .0030  .0004 .0005  .0011 .0004  .0011 .0019  .0005 .0036  .0004 .0012  .0003 .0031  .0004 .0034  .0004 .0021  .0004 .0014  .0005 .0021  .0005 .0018  .0005 .0029  .0005 .0055  .0005 .0016  .0005 .0000  .0013 .0033  .0005 .0001  .0013 .0000  .0012 .0112  .0006 .0041  .0004 .0013  .0003 .0047  .0004 .0139  .0005 .0080  .0004 .0116  .0005 .0128  .0005 .0072  .0005 .0020  .0003 .0063  .0004 .0031  .0004 .0064  .0005 .0043  .0004 .0026  .0004 .0010  .0003 .0231  .0006 .0007  .0008 .0000  .0009

E-12

Bromine .0067  .0007 .0089  .0006 .0024  .0004 .0055  .0007 .0017  .0004 .0013  .0004 .0038  .0007 .0052  .0006 .0072  .0004 .0008  .0010 .0043  .0007 .0132  .0006 .0028  .0003 .0040  .0007 .0105  .0006 .0026  .0006 .0077  .0005 .0024  .0007 .0147  .0006 .0154  .0007 .0068  .0010 .0118  .0006 .0029  .0005 .0069  .0006 .0023  .0004 .0008  .0011 .0224  .0018 .0150  .0006 .0022  .0003 .0111  .0005 .0385  .0011 .0108  .0007 .0217  .0015 .0338  .0013 .0586  .0017 .0142  .0005 .0555  .0012 .0246  .0007 .0589  .0014 .0221  .0009 .0081  .0005 .0029  .0004 .0062  .0007 .0020  .0003 .0021  .0003

Rubidium .0050  .0006 .0055  .0006 .0015  .0017 .0060  .0006 .0001  .0014 .0000  .0015 .0090  .0005 .0079  .0005 .0005  .0013 .0010  .0012 .0091  .0005 .0044  .0004 .0011  .0004 .0096  .0005 .0037  .0004 .0097  .0005 .0020  .0006 .0053  .0006 .0010  .0018 .0042  .0006 .0084  .0006 .0031  .0005 .0014  .0016 .0069  .0006 .0000  .0016 .0002  .0014 .0413  .0007 .0083  .0005 .0018  .0003 .0040  .0004 .0077  .0006 .0050  .0004 .0280  .0007 .0132  .0006 .0197  .0009 .0030  .0004 .0079  .0008 .0036  .0005 .0135  .0009 .0101  .0006 .0042  .0004 .0012  .0003 .0068  .0004 .0005  .0009 .0006  .0011

Strontium .0003  .0019 .0010  .0020 .0000  .0019 .0016  .0018 .0004  .0017 .0009  .0017 .0000  .0016 .0027  .0005 .0009  .0016 .0000  .0015 .0000  .0018 .0019  .0005 .0004  .0013 .0020  .0005 .0014  .0015 .0001  .0013 .0011  .0018 .0000  .0019 .0009  .0019 .0010  .0018 .0013  .0017 .0014  .0017 .0000  .0018 .0025  .0007 .0000  .0018 .0006  .0017 .0064  .0005 .0019  .0005 .0012  .0004 .0019  .0004 .0008  .0011 .0004  .0011 .0027  .0005 .0028  .0004 .0019  .0004 .0021  .0004 .0022  .0004 .0019  .0004 .0019  .0004 .0011  .0014 .0020  .0004 .0010  .0012 .0015  .0004 .0003  .0010 .0000  .0012

Appendix E.5 – Metals Data from X-Ray Fluorescence in µg m Site

Date

CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 GZ5 GZ5 GZ5 GZ5 GZ5 GZ5

021002 021008 021014 021020 021026 021201 021207 021213 021219 021225 030301 030307 030313 030319 030325 030605 030611 030617 030623 030629 021002 021008 021014 021020 021026 021201 021207 021213 021219 021225 030301 030307 030313 030319 030325 030605 030611 030617 030623 030629 021002 021008 021014 021020 021026 021201

Yttrium .0000  .0020 .0000  .0019 .0000  .0029 .0000  .0020 .0000  .0019 .0003  .0016 .0000  .0021 .0000  .0018 .0000  .0026 .0004  .0016 .0000  .0028 .0003  .0019 .0000  .0047 .0000  .0018 .0000  .0025 .0000  .0026 .0000  .0018 .0001  .0017 .0015  .0016 .0003  .0016 .0010  .0020 .0001  .0021 .0002  .0017 .0000  .0019 .0000  .0023 .0000  .0019 .0000  .0019 .0003  .0020 .0000  .0019 .0006  .0019 .0005  .0021 .0000  .0024 .0003  .0022 .0000  .0023 .0000  .0025 .0010  .0022 .0002  .0018 .0002  .0021 .0000  .0022 .0007  .0022 .0000  .0028 .0000  .0023 .0000  .0044 .0000  .0028 .0002  .0025 .0000  .0046

Zirconium

Molybdenum

Palladium

.0000  .0015 .0034  .0007 .0010  .0017 .0017  .0005 .0003  .0015 .0000  .0013 .0005  .0015 .0007  .0015 .0023  .0005 .0004  .0015 .0076  .0005 .0001  .0020 .0023  .0007 .0003  .0018 .0000  .0017 .0024  .0006 .0022  .0005 .0000  .0018 .0001  .0016 .0003  .0017 .0007  .0021 .0017  .0023 .0005  .0020 .0001  .0022 .0003  .0024 .0000  .0021 .0006  .0023 .0001  .0020 .0013  .0021 .0011  .0020 .0006  .0024 .0006  .0028 .0014  .0026 .0014  .0027 .0006  .0026 .0014  .0026 .0012  .0022 .0019  .0023 .0000  .0026 .0000  .0027 .0012  .0024 .0013  .0023 .0045  .0008 .0011  .0020 .0000  .0022 .0019  .0021

.0000  .0026 .0000  .0035 .0000  .0030 .0017  .0026 .0015  .0026 .0011  .0024 .0000  .0027 .0007  .0026 .0010  .0027 .0007  .0027 .0042  .0009 .0045  .0012 .0025  .0035 .0017  .0032 .0035  .0010 .0033  .0010 .0016  .0025 .0033  .0010 .0017  .0027 .0000  .0029 .0011  .0036 .0000  .0039 .0013  .0035 .0000  .0039 .0000  .0042 .0000  .0037 .0000  .0040 .0001  .0035 .0000  .0036 .0000  .0035 .0008  .0042 .0000  .0047 .0000  .0044 .0000  .0047 .0008  .0044 .0031  .0044 .0033  .0038 .0018  .0040 .0029  .0045 .0000  .0046 .0000  .0042 .0000  .0040 .0000  .0039 .0011  .0035 .0000  .0039 .0023  .0036

.0000  .0026 .0022  .0033 .0011  .0029 .0004  .0026 .0010  .0028 .0000  .0025 .0012  .0028 .0019  .0026 .0000  .0026 .0000  .0027 .0000  .0026 .0000  .0035 .0024  .0034 .0020  .0034 .0000  .0028 .0000  .0031 .0011  .0025 .0016  .0027 .0000  .0025 .0000  .0028 .0000  .0036 .0004  .0034 .0000  .0033 .0018  .0037 .0000  .0041 .0000  .0035 .0000  .0043 .0001  .0035 .0000  .0036 .0015  .0035 .0001  .0044 .0000  .0041 .0000  .0046 .0000  .0046 .0016  .0044 .0037  .0046 .0000  .0039 .0000  .0039 .0000  .0045 .0009  .0044 .0018  .0036 .0007  .0038 .0000  .0035 .0008  .0033 .0004  .0040 .0000  .0035

E-13

3

Silver .0440  .0019 .0827  .0029 .0390  .0019 .0409  .0019 .0930  .0026 .0561  .0021 .1457  .0031 .0425  .0019 .0962  .0026 .0803  .0024 .1008  .0028 .1247  .0036 .0853  .0030 .0446  .0023 .1844  .0036 .1387  .0031 .0466  .0020 .1006  .0027 .0993  .0027 .0455  .0020 .0009  .0047 .0000  .0049 .0023  .0047 .0000  .0046 .0000  .0050 .0000  .0047 .0000  .0051 .0000  .0047 .0000  .0046 .0012  .0043 .0008  .0052 .0022  .0058 .0016  .0055 .0000  .0058 .0000  .0056 .0038  .0056 .0027  .0057 .0000  .0050 .0000  .0055 .0054  .0056 .0001  .0049 .0008  .0047 .0030  .0047 .0007  .0045 .0000  .0049 .0020  .0048

Site

Date

GZ5 GZ5 GZ5 GZ5 GZ5 GZ5 GZ5 GZ5 GZ5 GZ5 GZ5 GZ5 GZ5 GZ5 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 TC2 TC2 TC2 TC2 TC2 TC2 TC2 TC2 TC2 TC2 TC2 TC2 TC2 TC2

021207 021213 021219 021225 030301 030307 030313 030319 030325 030605 030611 030617 030623 030629 021002 021008 021014 021020 021026 021201 021207 021213 021221 021225 030102 030301 030307 030313 030319 030325 030605 030611 030617 030623 030629 021002 021008 021014 021020 021026 021201 021207 021213 021219 021225 030301 030307 030313 030319

Yttrium .0002  .0037 .0000  .0024 .0000  .0036 .0002  .0023 .0000  .0032 .0000  .0029 .0011  .0045 .0001  .0025 .0000  .0040 .0000  .0025 .0012  .0034 .0000  .0048 .0012  .0038 .0008  .0026 .0000  .0023 .0000  .0025 .0000  .0019 .0002  .0017 .0000  .0025 .0000  .0017 .0000  .0018 .0020  .0063 .0000  .0069 .0000  .0022 .0014  .0026 .0011  .0022 .0013  .0026 .0000  .0023 .0002  .0024 .0000  .0028 .0006  .0023 .0012  .0021 .0000  .0024 .0004  .0021 .0000  .0021 .0000  .0020 .0000  .0020 .0000  .0019 .0003  .0017 .0000  .0020 .0000  .0017 .0000  .0017 .0000  .0022 .0000  .0020 .0004  .0021 .0001  .0023 .0002  .0026 .0000  .0024 .0016  .0023

Zirconium

Molybdenum

Palladium

.0031  .0007 .0004  .0020 .0027  .0008 .0011  .0021 .0043  .0009 .0006  .0028 .0011  .0030 .0000  .0026 .0029  .0009 .0022  .0025 .0070  .0008 .0068  .0011 .0069  .0010 .0028  .0008 .0008  .0022 .0015  .0021 .0000  .0022 .0008  .0020 .0005  .0025 .0000  .0019 .0012  .0021 .0028  .0070 .0070  .0074 .0003  .0022 .0000  .0021 .0006  .0025 .0005  .0024 .0000  .0026 .0018  .0025 .0023  .0028 .0006  .0026 .0011  .0025 .0000  .0027 .0010  .0025 .0009  .0025 .0013  .0020 .0012  .0021 .0000  .0022 .0008  .0020 .0000  .0020 .0000  .0018 .0000  .0020 .0012  .0022 .0000  .0023 .0008  .0020 .0012  .0027 .0023  .0027 .0000  .0029 .0020  .0026

.0004  .0037 .0000  .0035 .0000  .0039 .0000  .0036 .0004  .0046 .0000  .0047 .0000  .0052 .0000  .0045 .0000  .0046 .0020  .0043 .0051  .0014 .0000  .0053 .0000  .0048 .0037  .0041 .0000  .0038 .0000  .0037 .0000  .0038 .0000  .0036 .0000  .0043 .0000  .0033 .0000  .0036 .0076  .0122 .0059  .0127 .0000  .0038 .0000  .0037 .0028  .0043 .0024  .0042 .0009  .0045 .0028  .0044 .0005  .0048 .0026  .0045 .0023  .0043 .0000  .0046 .0007  .0044 .0031  .0043 .0002  .0036 .0000  .0037 .0000  .0038 .0000  .0034 .0000  .0036 .0002  .0032 .0000  .0036 .0000  .0038 .0000  .0040 .0014  .0035 .0000  .0046 .0018  .0046 .0000  .0050 .0027  .0045

.0006  .0037 .0000  .0034 .0015  .0036 .0001  .0034 .0000  .0048 .0000  .0045 .0000  .0046 .0000  .0043 .0000  .0045 .0000  .0046 .0025  .0042 .0000  .0046 .0000  .0045 .0000  .0040 .0012  .0036 .0000  .0036 .0009  .0034 .0002  .0036 .0014  .0038 .0031  .0033 .0006  .0036 .0000  .0117 .0000  .0125 .0000  .0037 .0008  .0037 .0000  .0045 .0000  .0040 .0010  .0045 .0000  .0048 .0000  .0045 .0000  .0042 .0026  .0041 .0000  .0043 .0037  .0043 .0000  .0044 .0000  .0035 .0000  .0035 .0000  .0036 .0001  .0031 .0000  .0034 .0000  .0034 .0000  .0032 .0018  .0038 .0000  .0039 .0000  .0033 .0004  .0042 .0011  .0043 .0000  .0045 .0034  .0048

E-14

Silver .0008  .0049 .0000  .0047 .0020  .0045 .0000  .0047 .0002  .0059 .0020  .0060 .0009  .0062 .0020  .0055 .0088  .0021 .0000  .0056 .0023  .0049 .0035  .0061 .0014  .0062 .0012  .0052 .0000  .0044 .0028  .0047 .0005  .0046 .0024  .0047 .0000  .0047 .0000  .0042 .0022  .0048 .0135  .0158 .0124  .0158 .0002  .0045 .0013  .0049 .0000  .0057 .0023  .0058 .0038  .0056 .0000  .0061 .0005  .0064 .0029  .0054 .0039  .0051 .0099  .0019 .0033  .0053 .0013  .0056 .0031  .0046 .0000  .0045 .0007  .0047 .0024  .0044 .0015  .0046 .0000  .0044 .0001  .0042 .0000  .0048 .0000  .0049 .0053  .0014 .0045  .0055 .0010  .0060 .0022  .0057 .0012  .0056

Site

Date

TC2 TC2 TC2 TC2 TC2 TC2 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7

030325 030605 030611 030617 030623 030629 021002 021008 021014 021020 021026 021201 021207 021213 021219 021225 030301 030307 030313 030319 030325 030605 030611 030617 030623 030629 021008 021014 021020 021026 021201 021207 021213 021219 021225 030301 030307 030313 030319 030325 030605 030611 030617 030623 030629

Yttrium .0000  .0025 .0001  .0024 .0009  .0023 .0000  .0022 .0006  .0020 .0010  .0020 .0000  .0022 .0000  .0021 .0000  .0019 .0006  .0019 .0000  .0024 .0001  .0018 .0002  .0016 .0000  .0021 .0000  .0019 .0000  .0019 .0011  .0021 .0000  .0024 .0015  .0023 .0000  .0022 .0000  .0025 .0017  .0020 .0007  .0022 .0002  .0024 .0000  .0021 .0000  .0020 .0000  .0041 .0000  .0018 .0001  .0013 .0000  .0015 .0003  .0021 .0000  .0016 .0000  .0033 .0000  .0026 .0000  .0033 .0004  .0015 .0016  .0022 .0001  .0016 .0000  .0027 .0006  .0022 .0005  .0017 .0005  .0014 .0000  .0017 .0005  .0012 .0000  .0015

Zirconium

Molybdenum

Palladium

.0011  .0027 .0010  .0028 .0014  .0028 .0006  .0024 .0009  .0024 .0007  .0024 .0005  .0022 .0012  .0021 .0000  .0022 .0022  .0022 .0000  .0025 .0000  .0019 .0012  .0019 .0011  .0021 .0006  .0020 .0000  .0018 .0003  .0025 .0013  .0027 .0000  .0026 .0010  .0025 .0007  .0023 .0000  .0023 .0011  .0026 .0008  .0027 .0007  .0026 .0000  .0024 .0033  .0006 .0002  .0018 .0002  .0016 .0011  .0016 .0021  .0005 .0023  .0005 .0035  .0006 .0047  .0006 .0016  .0018 .0010  .0017 .0020  .0005 .0015  .0018 .0019  .0006 .0007  .0019 .0000  .0019 .0014  .0017 .0025  .0005 .0006  .0014 .0006  .0017

.0000  .0047 .0000  .0048 .0000  .0048 .0025  .0042 .0001  .0042 .0000  .0042 .0005  .0040 .0000  .0037 .0000  .0039 .0000  .0038 .0000  .0043 .0000  .0034 .0000  .0032 .0000  .0036 .0000  .0035 .0000  .0032 .0000  .0042 .0025  .0047 .0005  .0046 .0026  .0044 .0021  .0040 .0016  .0040 .0000  .0045 .0000  .0047 .0000  .0045 .0024  .0041 .0022  .0032 .0000  .0032 .0000  .0027 .0004  .0028 .0013  .0027 .0000  .0026 .0000  .0031 .0000  .0031 .0000  .0031 .0128  .0010 .0016  .0027 .0016  .0031 .0060  .0010 .0000  .0033 .0000  .0032 .0022  .0029 .0017  .0027 .0002  .0025 .0000  .0030

.0000  .0045 .0014  .0046 .0023  .0043 .0013  .0038 .0018  .0041 .0000  .0043 .0000  .0037 .0000  .0034 .0000  .0036 .0000  .0035 .0000  .0041 .0000  .0031 .0000  .0036 .0000  .0036 .0000  .0033 .0019  .0033 .0000  .0043 .0000  .0047 .0004  .0048 .0000  .0048 .0000  .0041 .0000  .0038 .0004  .0043 .0000  .0045 .0000  .0043 .0000  .0045 .0008  .0032 .0000  .0028 .0000  .0025 .0000  .0027 .0030  .0009 .0001  .0025 .0000  .0028 .0000  .0030 .0000  .0029 .0000  .0034 .0000  .0026 .0000  .0027 .0000  .0030 .0000  .0029 .0018  .0030 .0001  .0028 .0000  .0028 .0014  .0026 .0000  .0028

E-15

Silver .0000  .0057 .0021  .0055 .0037  .0052 .0000  .0054 .0050  .0053 .0000  .0056 .0000  .0046 .0027  .0046 .0008  .0046 .0000  .0049 .0000  .0051 .0000  .0042 .0000  .0043 .0006  .0046 .0014  .0044 .0010  .0041 .0049  .0057 .0014  .0060 .0026  .0060 .0023  .0058 .0023  .0055 .0022  .0050 .0032  .0057 .0000  .0053 .0059  .0019 .0000  .0054 .0808  .0026 .0574  .0022 .0109  .0014 .0000  .0038 .0133  .0014 .0037  .0012 .0067  .0014 .0054  .0014 .0254  .0018 .4433  .0055 .0218  .0016 .0491  .0021 .1219  .0030 .0229  .0017 .0427  .0020 .0007  .0034 .0236  .0017 .0021  .0033 .0279  .0018

Appendix E.6 – Metals Data from X-Ray Fluorescence in µg m Site

Date

CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 GZ5 GZ5 GZ5 GZ5 GZ5 GZ5

021002 021008 021014 021020 021026 021201 021207 021213 021219 021225 030301 030307 030313 030319 030325 030605 030611 030617 030623 030629 021002 021008 021014 021020 021026 021201 021207 021213 021219 021225 030301 030307 030313 030319 030325 030605 030611 030617 030623 030629 021002 021008 021014 021020 021026 021201

Cadmium .0043  .0012 .0000  .0045 .0115  .0015 .0027  .0035 .0033  .0037 .0011  .0034 .0031  .0036 .0000  .0034 .0031  .0034 .0016  .0034 .0022  .0038 .0031  .0050 .0049  .0050 .0000  .0043 .0015  .0040 .0037  .0041 .0002  .0035 .0035  .0040 .0000  .0036 .0000  .0034 .0019  .0048 .0002  .0049 .0008  .0046 .0000  .0047 .0003  .0049 .0005  .0049 .0000  .0050 .0061  .0016 .0015  .0044 .0046  .0049 .0071  .0020 .0030  .0060 .0000  .0055 .0000  .0063 .0038  .0058 .0000  .0058 .0000  .0052 .0063  .0019 .0001  .0058 .0002  .0055 .0036  .0054 .0068  .0017 .0120  .0017 .0011  .0046 .0041  .0050 .0000  .0049

Indium .0000  .0040 .0000  .0059 .0015  .0047 .0000  .0040 .0000  .0040 .0000  .0041 .0000  .0043 .0010  .0040 .0021  .0044 .0000  .0042 .0000  .0044 .0000  .0057 .0033  .0060 .0000  .0050 .0000  .0046 .0000  .0047 .0000  .0041 .0000  .0046 .0002  .0044 .0000  .0045 .0000  .0058 .0036  .0062 .0025  .0057 .0000  .0062 .0016  .0065 .0000  .0060 .0000  .0060 .0048  .0059 .0023  .0056 .0000  .0056 .0026  .0071 .0037  .0073 .0000  .0071 .0000  .0073 .0012  .0071 .0000  .0075 .0047  .0067 .0003  .0067 .0000  .0070 .0000  .0069 .0041  .0061 .0035  .0060 .0016  .0061 .0000  .0054 .0000  .0058 .0000  .0056

E-16

Tin .0138  .0024 .0073  .0096 .0255  .0027 .0100  .0023 .0297  .0028 .0228  .0024 .0171  .0031 .0183  .0024 .0492  .0031 .0226  .0028 .0550  .0034 .0181  .0039 .0786  .0042 .0196  .0031 .0269  .0038 .0145  .0034 .0109  .0026 .0110  .0029 .0055  .0083 .0041  .0076 .0154  .0029 .0095  .0030 .0260  .0031 .0116  .0031 .0223  .0034 .0271  .0034 .0129  .0031 .0283  .0032 .0184  .0030 .0155  .0030 .0111  .0115 .0105  .0117 .0045  .0112 .0459  .0044 .0320  .0043 .0032  .0117 .0000  .0100 .0244  .0039 .0000  .0109 .0000  .0106 .0292  .0033 .0115  .0031 .0531  .0037 .0225  .0030 .0279  .0034 .0799  .0039

Antimony .0067  .0074 .0051  .0095 .0144  .0028 .0045  .0073 .0157  .0025 .0110  .0022 .0095  .0024 .0130  .0025 .0095  .0025 .0051  .0075 .0074  .0077 .0027  .0102 .0246  .0036 .0050  .0094 .0130  .0028 .0045  .0085 .0021  .0076 .0001  .0081 .0000  .0075 .0000  .0078 .0105  .0032 .0050  .0100 .0001  .0094 .0037  .0099 .0012  .0108 .0047  .0099 .0000  .0103 .0134  .0032 .0010  .0095 .0035  .0094 .0000  .0120 .0009  .0133 .0084  .0125 .0000  .0128 .0000  .0128 .0000  .0124 .0000  .0114 .0000  .0120 .0008  .0123 .0002  .0125 .0153  .0037 .0139  .0037 .0203  .0035 .0128  .0033 .0130  .0036 .0275  .0036

3

Barium .0393  .0109 .0000  .0432 .0000  .0344 .0000  .0328 .0177  .0325 .0000  .0306 .0000  .0322 .0155  .0327 .0000  .0320 .0000  .0325 .0179  .0343 .0000  .0473 .0309  .0448 .0272  .0420 .0000  .0370 .0372  .0122 .0123  .0338 .0092  .0372 .0077  .0343 .0000  .0368 .0240  .0454 .0042  .0465 .0000  .0445 .0216  .0472 .0200  .0489 .0215  .0452 .0000  .0473 .0427  .0449 .0207  .0444 .0000  .0431 .0217  .0564 .0326  .0600 .0118  .0581 .0261  .0593 .0000  .0579 .0287  .0569 .0000  .0530 .0047  .0548 .0142  .0568 .0000  .0594 .0031  .0495 .0000  .0481 .0509  .0158 .0000  .0446 .0066  .0478 .0000  .0439

Site

Date

GZ5 GZ5 GZ5 GZ5 GZ5 GZ5 GZ5 GZ5 GZ5 GZ5 GZ5 GZ5 GZ5 GZ5 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 TC2 TC2 TC2 TC2 TC2 TC2 TC2 TC2 TC2 TC2 TC2 TC2 TC2 TC2

021207 021213 021219 021225 030301 030307 030313 030319 030325 030605 030611 030617 030623 030629 021002 021008 021014 021020 021026 021201 021207 021213 021221 021225 030102 030301 030307 030313 030319 030325 030605 030611 030617 030623 030629 021002 021008 021014 021020 021026 021201 021207 021213 021219 021225 030301 030307 030313 030319

Cadmium .0037  .0051 .0000  .0050 .0000  .0050 .0016  .0048 .0034  .0063 .0068  .0022 .0000  .0070 .0003  .0058 .0128  .0021 .0018  .0054 .0000  .0058 .0122  .0022 .0018  .0061 .0000  .0055 .0091  .0017 .0043  .0050 .0000  .0048 .0000  .0043 .0028  .0053 .0017  .0044 .0049  .0016 .0000  .0160 .0064  .0159 .0050  .0016 .0018  .0049 .0024  .0057 .0095  .0022 .0015  .0059 .0027  .0061 .0073  .0022 .0004  .0058 .0047  .0055 .0006  .0054 .0005  .0054 .0000  .0057 .0000  .0046 .0027  .0049 .0000  .0046 .0024  .0045 .0062  .0016 .0000  .0043 .0038  .0044 .0000  .0048 .0000  .0050 .0000  .0046 .0000  .0057 .0084  .0022 .0067  .0020 .0042  .0060

Indium .0033  .0057 .0000  .0058 .0039  .0055 .0014  .0058 .0007  .0073 .0000  .0078 .0018  .0077 .0000  .0068 .0000  .0070 .0000  .0068 .0018  .0065 .0008  .0078 .0041  .0074 .0000  .0068 .0000  .0058 .0000  .0054 .0000  .0061 .0016  .0057 .0000  .0064 .0000  .0054 .0034  .0054 .0121  .0200 .0000  .0214 .0009  .0057 .0022  .0059 .0000  .0070 .0031  .0065 .0014  .0074 .0000  .0073 .0022  .0074 .0006  .0074 .0020  .0067 .0000  .0078 .0000  .0070 .0000  .0069 .0000  .0057 .0000  .0057 .0000  .0057 .0021  .0052 .0000  .0053 .0000  .0049 .0035  .0054 .0000  .0061 .0007  .0064 .0039  .0060 .0010  .0072 .0000  .0075 .0042  .0068 .0000  .0073

E-17

Tin .0287  .0030 .0183  .0030 .0302  .0032 .0487  .0034 .0229  .0040 .0070  .0122 .0470  .0048 .0231  .0040 .0406  .0044 .0140  .0039 .0121  .0038 .0362  .0044 .0116  .0120 .0000  .0109 .0216  .0032 .0210  .0030 .0100  .0030 .0131  .0029 .0274  .0033 .0140  .0030 .0181  .0029 .0017  .0305 .0287  .0350 .0150  .0030 .0340  .0033 .0112  .0037 .0051  .0107 .0119  .0122 .0107  .0124 .0414  .0044 .0093  .0117 .0047  .0111 .0123  .0125 .0000  .0109 .0000  .0111 .0129  .0029 .0028  .0084 .0089  .0029 .0075  .0079 .0172  .0029 .0152  .0027 .0000  .0080 .0118  .0031 .0165  .0030 .0215  .0030 .0025  .0114 .0011  .0116 .0112  .0113 .0094  .0124

Antimony .0321  .0036 .0143  .0032 .0269  .0036 .0100  .0032 .0160  .0045 .0131  .0139 .0200  .0051 .0070  .0127 .0068  .0132 .0000  .0123 .0203  .0044 .0156  .0047 .0000  .0133 .0000  .0126 .0126  .0035 .0120  .0034 .0068  .0100 .0060  .0095 .0100  .0109 .0019  .0094 .0043  .0096 .0000  .0355 .0324  .0369 .0000  .0099 .0114  .0033 .0000  .0123 .0055  .0124 .0002  .0128 .0000  .0135 .0081  .0134 .0000  .0126 .0000  .0118 .0000  .0131 .0000  .0118 .0007  .0126 .0081  .0095 .0000  .0097 .0055  .0097 .0034  .0090 .0013  .0094 .0071  .0088 .0016  .0095 .0039  .0100 .0043  .0101 .0040  .0094 .0060  .0129 .0124  .0133 .0000  .0130 .0000  .0128

Barium .0000  .0448 .0000  .0441 .0057  .0457 .0200  .0442 .0409  .0594 .0198  .0602 .0205  .0636 .0000  .0584 .0000  .0592 .0100  .0563 .0330  .0542 .0671  .0210 .0000  .0596 .0000  .0547 .0134  .0459 .0210  .0452 .0000  .0468 .0000  .0448 .0095  .0494 .0000  .0428 .0056  .0446 .0000  .1617 .0000  .1670 .0141  .0452 .0000  .0448 .0261  .0578 .0000  .0555 .0000  .0586 .0181  .0595 .0543  .0609 .0094  .0581 .0020  .0558 .0570  .0577 .0454  .0564 .0186  .0559 .0000  .0430 .0000  .0448 .0156  .0460 .0000  .0428 .0000  .0441 .0000  .0408 .0011  .0430 .0000  .0455 .0000  .0476 .0000  .0428 .0809  .0199 .0527  .0598 .0000  .0626 .0827  .0196

Site

Date

TC2 TC2 TC2 TC2 TC2 TC2 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7

030325 030605 030611 030617 030623 030629 021002 021008 021014 021020 021026 021201 021207 021213 021219 021225 030301 030307 030313 030319 030325 030605 030611 030617 030623 030629 021008 021014 021020 021026 021201 021207 021213 021219 021225 030301 030307 030313 030319 030325 030605 030611 030617 030623 030629

Cadmium .0017  .0059 .0040  .0062 .0018  .0060 .0019  .0061 .0012  .0054 .0000  .0056 .0025  .0048 .0000  .0049 .0000  .0048 .0000  .0049 .0014  .0053 .0000  .0046 .0027  .0043 .0038  .0045 .0005  .0044 .0070  .0015 .0023  .0055 .0000  .0062 .0021  .0062 .0058  .0059 .0081  .0017 .0000  .0053 .0043  .0058 .0000  .0059 .0033  .0055 .0000  .0053 .0061  .0014 .0024  .0040 .0016  .0034 .0104  .0015 .0059  .0013 .0000  .0036 .0000  .0041 .0041  .0042 .0033  .0042 .0000  .0041 .0073  .0013 .0023  .0039 .0000  .0037 .0047  .0014 .0015  .0039 .0000  .0038 .0167  .0016 .0008  .0033 .0014  .0034

Indium .0041  .0077 .0000  .0077 .0000  .0073 .0011  .0073 .0000  .0064 .0000  .0069 .0000  .0061 .0000  .0054 .0000  .0065 .0000  .0059 .0000  .0066 .0028  .0058 .0000  .0052 .0016  .0059 .0044  .0055 .0011  .0054 .0013  .0070 .0000  .0077 .0000  .0075 .0059  .0072 .0033  .0063 .0005  .0066 .0025  .0067 .0000  .0074 .0000  .0071 .0000  .0065 .0000  .0047 .0000  .0047 .0000  .0044 .0001  .0043 .0028  .0041 .0000  .0042 .0019  .0048 .0000  .0045 .0009  .0048 .0000  .0051 .0000  .0043 .0000  .0049 .0000  .0045 .0000  .0048 .0000  .0046 .0003  .0047 .0040  .0044 .0012  .0043 .0015  .0046

E-18

Tin .0231  .0042 .0047  .0117 .0000  .0116 .0247  .0039 .0000  .0102 .0000  .0103 .0146  .0032 .0026  .0083 .0012  .0090 .0049  .0089 .0236  .0034 .0143  .0029 .0108  .0028 .0193  .0030 .0257  .0030 .0171  .0029 .0093  .0114 .0043  .0121 .0010  .0115 .0150  .0040 .0670  .0045 .0000  .0100 .0014  .0107 .0149  .0042 .0000  .0109 .0000  .0104 .0255  .0030 .0280  .0029 .0000  .0066 .0062  .0068 .0262  .0026 .0098  .0022 .0296  .0028 .0383  .0027 .0244  .0026 .0000  .0187 .0185  .0027 .0168  .0028 .0264  .0033 .0406  .0031 .0118  .0027 .0000  .0070 .0221  .0026 .0002  .0063 .0034  .0076

Antimony .0000  .0134 .0000  .0128 .0000  .0130 .0041  .0125 .0047  .0115 .0026  .0119 .0000  .0101 .0054  .0094 .0038  .0102 .0000  .0099 .0024  .0105 .0069  .0090 .0000  .0091 .0075  .0096 .0000  .0097 .0026  .0091 .0000  .0122 .0007  .0132 .0000  .0129 .0026  .0127 .0012  .0118 .0000  .0113 .0000  .0122 .0000  .0130 .0000  .0121 .0000  .0117 .0115  .0028 .0029  .0082 .0020  .0074 .0017  .0077 .0080  .0081 .0049  .0071 .0170  .0029 .0282  .0029 .0075  .0081 .0016  .0086 .0057  .0078 .0072  .0084 .0076  .0082 .0064  .0087 .0000  .0085 .0000  .0081 .0094  .0027 .0000  .0073 .0000  .0081

Barium .0000  .0600 .0267  .0594 .0143  .0595 .0417  .0561 .0042  .0556 .0145  .0554 .0000  .0468 .0065  .0451 .0000  .0469 .0296  .0460 .0125  .0496 .0099  .0430 .0000  .0424 .0222  .0449 .0034  .0435 .0042  .0422 .0240  .0575 .0037  .0602 .0000  .0603 .0000  .0580 .0157  .0557 .0263  .0549 .0130  .0573 .0155  .0594 .0000  .0576 .0000  .0548 .0000  .0372 .0055  .0377 .0096  .0346 .0075  .0348 .0359  .0112 .0014  .0329 .0000  .0371 .0312  .0360 .0238  .0357 .0653  .0126 .0000  .0359 .0090  .0384 .0083  .0366 .0412  .0132 .0046  .0388 .0140  .0374 .0376  .0120 .0048  .0343 .0000  .0374

Appendix E.7 – Metals Data from X-Ray Fluorescence in µg m Site

Date

CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CH6 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 CW3 GZ5 GZ5 GZ5 GZ5 GZ5 GZ5

021002 021008 021014 021020 021026 021201 021207 021213 021219 021225 030301 030307 030313 030319 030325 030605 030611 030617 030623 030629 021002 021008 021014 021020 021026 021201 021207 021213 021219 021225 030301 030307 030313 030319 030325 030605 030611 030617 030623 030629 021002 021008 021014 021020 021026 021201

Gold .0000  .0079 .0000  .0054 .0000  .0199 .0000  .0116 .0000  .0106 .0001  .0083 .0000  .0062 .0000  .0095 .0000  .0116 .0000  .0070 .0003  .0123 .0011  .0053 .0000  .0309 .0012  .0067 .0000  .0173 .0000  .0100 .0000  .0078 .0000  .0091 .0000  .0042 .0000  .0039 .0000  .0110 .0000  .0094 .0020  .0037 .0000  .0037 .0000  .0121 .0000  .0090 .0000  .0046 .0000  .0102 .0000  .0069 .0000  .0111 .0054  .0055 .0000  .0120 .0008  .0064 .0000  .0067 .0000  .0114 .0000  .0069 .0000  .0047 .0031  .0107 .0000  .0042 .0000  .0042 .0000  .0145 .0000  .0075 .0000  .0242 .0000  .0365 .0000  .0101 .0000  .0235

Mercury .0000  .0015 .0000  .0019 .0000  .0018 .0004  .0015 .0000  .0014 .0011  .0014 .0008  .0016 .0009  .0015 .0011  .0017 .0000  .0015 .0011  .0017 .0008  .0020 .0012  .0024 .0013  .0018 .0000  .0018 .0012  .0018 .0000  .0014 .0005  .0017 .0000  .0015 .0002  .0016 .0000  .0019 .0001  .0021 .0000  .0019 .0003  .0021 .0000  .0023 .0000  .0020 .0000  .0021 .0005  .0019 .0000  .0019 .0000  .0019 .0002  .0024 .0000  .0026 .0000  .0023 .0000  .0025 .0000  .0025 .0000  .0023 .0000  .0021 .0005  .0022 .0004  .0024 .0005  .0025 .0000  .0024 .0000  .0022 .0009  .0025 .0004  .0021 .0000  .0022 .0017  .0025

E-19

Thallium .0030  .0042 .0018  .0028 .0026  .0066 .0015  .0042 .0017  .0038 .0026  .0027 .0025  .0044 .0014  .0034 .0021  .0059 .0010  .0027 .0089  .0024 .0000  .0032 .0040  .0114 .0000  .0030 .0026  .0058 .0065  .0017 .0012  .0033 .0021  .0028 .0024  .0008 .0021  .0007 .0032  .0032 .0003  .0029 .0003  .0017 .0000  .0018 .0012  .0032 .0017  .0027 .0000  .0019 .0002  .0033 .0016  .0024 .0015  .0029 .0010  .0024 .0023  .0032 .0000  .0024 .0003  .0026 .0028  .0043 .0011  .0028 .0000  .0020 .0000  .0030 .0007  .0021 .0001  .0021 .0045  .0056 .0027  .0039 .0080  .0104 .0021  .0065 .0017  .0046 .0040  .0117

Lead .1639  .0021 .0858  .0020 .2686  .0027 .1632  .0021 .1430  .0020 .0923  .0016 .1730  .0021 .1274  .0019 .2393  .0025 .0951  .0017 .2802  .0028 .0939  .0021 .4614  .0040 .0933  .0020 .2249  .0025 .1885  .0024 .1163  .0019 .0890  .0018 .0713  .0016 .0540  .0015 .0975  .0021 .0862  .0021 .0119  .0013 .0064  .0014 .0940  .0022 .0719  .0019 .0103  .0015 .1102  .0022 .0620  .0018 .0840  .0020 .0267  .0017 .0764  .0024 .0296  .0018 .0362  .0020 .1331  .0028 .0503  .0020 .0064  .0014 .0707  .0021 .0000  .0045 .0024  .0047 .2001  .0029 .1252  .0024 .4237  .0039 .2517  .0030 .1606  .0026 .4795  .0040

3

Uranium .0000  .0029 .0000  .0031 .0000  .0041 .0000  .0032 .0000  .0031 .0000  .0031 .0000  .0035 .0000  .0029 .0000  .0038 .0012  .0028 .0000  .0038 .0000  .0032 .0000  .0080 .0000  .0031 .0000  .0037 .0000  .0055 .0000  .0034 .0008  .0029 .0015  .0028 .0000  .0029 .0000  .0032 .0000  .0034 .0024  .0029 .0017  .0032 .0000  .0037 .0028  .0032 .0000  .0033 .0001  .0033 .0000  .0030 .0000  .0031 .0000  .0034 .0000  .0041 .0000  .0036 .0000  .0038 .0000  .0038 .0000  .0037 .0002  .0031 .0008  .0035 .0004  .0037 .0000  .0038 .0007  .0043 .0000  .0038 .0000  .0068 .0000  .0035 .0000  .0037 .0000  .0048

Site

Date

GZ5 GZ5 GZ5 GZ5 GZ5 GZ5 GZ5 GZ5 GZ5 GZ5 GZ5 GZ5 GZ5 GZ5 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 SZ4 TC2 TC2 TC2 TC2 TC2 TC2 TC2 TC2 TC2 TC2 TC2 TC2 TC2 TC2

021207 021213 021219 021225 030301 030307 030313 030319 030325 030605 030611 030617 030623 030629 021002 021008 021014 021020 021026 021201 021207 021213 021221 021225 030102 030301 030307 030313 030319 030325 030605 030611 030617 030623 030629 021002 021008 021014 021020 021026 021201 021207 021213 021219 021225 030301 030307 030313 030319

Gold .0000  .0174 .0020  .0094 .0000  .0204 .0000  .0086 .0000  .0263 .0000  .0116 .0000  .0261 .0000  .0097 .0000  .0196 .0026  .0165 .0000  .0382 .0000  .0351 .0000  .0276 .0000  .0118 .0000  .0144 .0000  .0164 .0000  .0051 .0000  .0044 .0000  .0154 .0016  .0074 .0000  .0064 .0029  .0255 .0053  .0280 .0000  .0125 .0000  .0122 .0031  .0077 .0000  .0134 .0000  .0116 .0009  .0119 .0000  .0184 .0000  .0073 .0000  .0055 .0000  .0103 .0000  .0041 .0041  .0041 .0000  .0107 .0000  .0101 .0000  .0036 .0000  .0032 .0000  .0140 .0021  .0075 .0000  .0034 .0000  .0099 .0000  .0050 .0000  .0132 .0009  .0048 .0000  .0145 .0000  .0056 .0010  .0080

Mercury .0019  .0024 .0006  .0020 .0007  .0024 .0000  .0020 .0015  .0028 .0000  .0027 .0000  .0030 .0000  .0025 .0000  .0028 .0012  .0025 .0020  .0026 .0000  .0032 .0000  .0028 .0000  .0023 .0000  .0021 .0000  .0020 .0000  .0020 .0000  .0019 .0000  .0023 .0005  .0018 .0000  .0019 .0000  .0066 .0000  .0069 .0008  .0021 .0000  .0020 .0000  .0024 .0009  .0024 .0014  .0026 .0000  .0024 .0000  .0026 .0018  .0025 .0004  .0023 .0000  .0025 .0000  .0023 .0000  .0024 .0000  .0019 .0000  .0020 .0000  .0020 .0000  .0018 .0002  .0020 .0003  .0018 .0002  .0019 .0000  .0020 .0001  .0022 .0000  .0019 .0000  .0025 .0010  .0026 .0004  .0027 .0000  .0024

E-20

Thallium .0049  .0088 .0038  .0048 .0046  .0085 .0035  .0046 .0036  .0069 .0033  .0055 .0080  .0102 .0020  .0039 .0059  .0096 .0000  .0044 .0034  .0081 .0093  .0113 .0072  .0086 .0042  .0056 .0025  .0042 .0014  .0047 .0000  .0020 .0003  .0018 .0000  .0039 .0001  .0025 .0030  .0007 .0012  .0085 .0047  .0107 .0011  .0037 .0031  .0050 .0002  .0028 .0000  .0046 .0011  .0030 .0023  .0039 .0015  .0049 .0019  .0028 .0000  .0023 .0000  .0032 .0005  .0022 .0004  .0020 .0008  .0031 .0010  .0029 .0014  .0020 .0010  .0017 .0021  .0032 .0015  .0024 .0011  .0017 .0010  .0034 .0014  .0022 .0023  .0036 .0013  .0023 .0029  .0045 .0023  .0026 .0000  .0029

Lead .3563  .0035 .1750  .0026 .3410  .0035 .1659  .0025 .2499  .0036 .1866  .0032 .3912  .0044 .1090  .0026 .3695  .0042 .1409  .0028 .2959  .0038 .4363  .0048 .3254  .0039 .1940  .0031 .1513  .0025 .1735  .0026 .0275  .0016 .0148  .0014 .1340  .0025 .0643  .0018 .0421  .0017 .2042  .0063 .3050  .0073 .1256  .0023 .1856  .0027 .0515  .0020 .1532  .0029 .0642  .0022 .1120  .0026 .1593  .0030 .0476  .0020 .0290  .0018 .0801  .0024 .0053  .0015 .0051  .0015 .1040  .0021 .0884  .0020 .0211  .0015 .0073  .0012 .1021  .0021 .0634  .0017 .0027  .0037 .1098  .0022 .0389  .0017 .1212  .0022 .0190  .0018 .1466  .0029 .0215  .0019 .0555  .0021

Uranium .0000  .0052 .0000  .0039 .0000  .0043 .0013  .0035 .0022  .0043 .0000  .0044 .0000  .0065 .0000  .0045 .0000  .0053 .0000  .0039 .0000  .0043 .0000  .0069 .0000  .0049 .0000  .0034 .0032  .0038 .0000  .0038 .0005  .0031 .0000  .0029 .0012  .0040 .0005  .0028 .0006  .0031 .0000  .0103 .0000  .0109 .0017  .0036 .0000  .0040 .0000  .0036 .0000  .0039 .0000  .0038 .0000  .0039 .0000  .0042 .0000  .0039 .0010  .0036 .0006  .0041 .0006  .0036 .0000  .0035 .0012  .0032 .0015  .0035 .0010  .0032 .0000  .0027 .0000  .0033 .0005  .0027 .0025  .0029 .0000  .0035 .0000  .0033 .0002  .0033 .0000  .0038 .0000  .0040 .0000  .0041 .0026  .0038

Site

Date

TC2 TC2 TC2 TC2 TC2 TC2 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 TM1 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7 ZS7

030325 030605 030611 030617 030623 030629 021002 021008 021014 021020 021026 021201 021207 021213 021219 021225 030301 030307 030313 030319 030325 030605 030611 030617 030623 030629 021008 021014 021020 021026 021201 021207 021213 021219 021225 030301 030307 030313 030319 030325 030605 030611 030617 030623 030629

Gold .0000  .0087 .0000  .0072 .0007  .0044 .0013  .0092 .0000  .0038 .0007  .0039 .0000  .0097 .0000  .0099 .0000  .0036 .0000  .0035 .0000  .0151 .0000  .0081 .0000  .0032 .0000  .0092 .0000  .0054 .0000  .0103 .0019  .0046 .0009  .0132 .0007  .0062 .0000  .0096 .0009  .0136 .0055  .0059 .0000  .0042 .0000  .0100 .0000  .0040 .0037  .0039 .0000  .0175 .0000  .0074 .0000  .0035 .0000  .0055 .0000  .0270 .0000  .0077 .0000  .0168 .0000  .0332 .0000  .0192 .0000  .0056 .0000  .0104 .0000  .0095 .0000  .0150 .0000  .0129 .0000  .0057 .0017  .0031 .0020  .0155 .0024  .0027 .0000  .0027

Mercury .0000  .0025 .0000  .0025 .0000  .0025 .0013  .0024 .0001  .0023 .0002  .0023 .0000  .0022 .0000  .0020 .0000  .0020 .0002  .0021 .0000  .0024 .0008  .0019 .0000  .0017 .0000  .0020 .0000  .0019 .0000  .0017 .0000  .0023 .0013  .0026 .0000  .0025 .0000  .0023 .0009  .0023 .0000  .0022 .0009  .0025 .0000  .0025 .0000  .0023 .0000  .0022 .0031  .0007 .0000  .0018 .0000  .0015 .0000  .0014 .0007  .0017 .0015  .0016 .0002  .0019 .0000  .0019 .0006  .0020 .0002  .0016 .0005  .0016 .0004  .0017 .0005  .0018 .0000  .0018 .0010  .0018 .0008  .0016 .0002  .0015 .0005  .0014 .0000  .0015

E-21

Thallium .0000  .0036 .0019  .0028 .0004  .0024 .0011  .0031 .0007  .0020 .0000  .0019 .0033  .0035 .0016  .0031 .0004  .0019 .0002  .0018 .0004  .0035 .0006  .0026 .0007  .0018 .0024  .0035 .0014  .0026 .0004  .0030 .0010  .0024 .0001  .0033 .0001  .0024 .0000  .0028 .0001  .0052 .0000  .0025 .0008  .0021 .0044  .0011 .0013  .0021 .0010  .0021 .0092  .0096 .0020  .0029 .0000  .0014 .0019  .0021 .0036  .0048 .0021  .0032 .0046  .0076 .0050  .0060 .0054  .0079 .0004  .0019 .0022  .0052 .0019  .0024 .0040  .0065 .0036  .0042 .0011  .0020 .0002  .0016 .0009  .0033 .0005  .0014 .0009  .0014

Lead .1010  .0026 .0542  .0022 .0012  .0048 .0719  .0022 .0024  .0043 .0002  .0042 .1108  .0023 .1004  .0021 .0066  .0014 .0054  .0013 .1115  .0024 .0741  .0019 .0176  .0013 .1190  .0022 .0703  .0019 .0962  .0020 .0249  .0017 .0795  .0024 .0294  .0019 .0638  .0022 .1767  .0030 .0424  .0019 .0020  .0046 .0713  .0023 .0000  .0045 .0034  .0042 .3968  .0033 .0961  .0019 .0164  .0011 .0571  .0015 .1815  .0023 .1126  .0019 .3105  .0029 .2354  .0026 .3269  .0029 .0384  .0014 .1966  .0024 .0669  .0017 .2545  .0027 .1523  .0023 .0450  .0015 .0157  .0012 .1087  .0020 .0083  .0009 .0023  .0031

Uranium .0000  .0039 .0000  .0041 .0009  .0040 .0000  .0036 .0008  .0035 .0015  .0035 .0013  .0036 .0011  .0034 .0027  .0032 .0000  .0030 .0008  .0039 .0000  .0029 .0004  .0026 .0017  .0035 .0000  .0030 .0000  .0031 .0023  .0036 .0000  .0040 .0004  .0038 .0000  .0037 .0000  .0037 .0000  .0034 .0003  .0037 .0000  .0041 .0027  .0037 .0002  .0034 .0000  .0071 .0013  .0031 .0005  .0023 .0006  .0024 .0000  .0028 .0000  .0023 .0000  .0052 .0000  .0035 .0000  .0043 .0000  .0026 .0000  .0029 .0000  .0027 .0000  .0034 .0000  .0032 .0000  .0028 .0000  .0024 .0003  .0026 .0000  .0020 .0000  .0025

Appendix F

Characterization of Organic Compounds in PM2.5 Collected by High Volume Samplers at the Hong Kong sites Tung Chung and Yuen Long

Mei Zheng1, James Schauer2 and Michael Bergin 1,3

1

School of Earth and Atmospheric Sciences, Georgia Institute of Technology 2

University of Wisconsin-Madison

3

School of Civil and Environmental Engineering, Georgia Institute of Technology

May 2004

Introduction

24-hour integrated PM2.5 samples were collected by high volume samplers from Tung Chung (TC) and Yuen Long (YL), located in the Hong Kong Special Administrative Region on October 20, 2002, December 13 and 19, 2002, and March 25, 2003 for the analysis of solvent extractable organic compounds. The TC sampling site is located in north Lantau Island, within a residential area adjacent to the highway and railway lines. The YL sampling site is located in an urban district of New Territory in Hong Kong (HK). The sampling days have been discussed in the main body of this report and can be summarized as follows: the samples collected on December 19, 2002 had the highest aerosol concentrations among the HK samples taken in December 2002. Surface winds were relatively light with both north and south components indicating relatively stagnant air with sources throughout the PRD influencing regional concentrations. Light precipitation also occurred in Hong Kong on December, 19. The samples obtained on December 13, 2002 can be characterized by northerly flow with highest observed PM2.5 concentrations observed at Zhongshan suggesting that sources from Guangdong influence fine particle concentrations in Hong Kong. The samples of March 25, 2003 were selected since during this day concentrations are high throughout the PRD, with the highest concentration of PM2.5 measured during the study occurring at Guangzhou (~130 µg/m3). This day was characterized by relatively stagnant airflow in the PRD. On October 20, 2002 concentrations of PM2.5 were extremely low throughout the PRD with southerly winds and precipitation throughout the PRD. Overall, the samples represent both periods of high and low PM2.5 concentrations in Hong Kong and the PRD. The overall objective of the high volume sampling is to evaluate the extent to which these samples represent monthly composite samples, and to determine the spatial variability of specific organic tracer compounds across Hong Kong for specific events. In this report we

2

compare specific compounds and classes of compounds between the sites and with monthly composites. One important point that should be noted is that samples were found to be highly contaminated with an unidentified organic compound, which was not present in appreciable amounts in the monthly composite samples. It is hypothesized that the compound is an oil associated with the impactors used in the high volume samplers. The compound not only overlapped with existing tracer compounds, but apparently had a significant impact on the overall organic carbon content of the high volume filters. It should be pointed out that it is possible that the contaminant also influenced the concentrations of other organic compounds such as n-alkanes and PAH’s. Nonetheless, we present concentrations of tracer compounds with the caveat that more work must be done to determine the source of the contaminant and possible influence on tracer compound concentrations.

Experimental The organic carbon (OC) mass collected on the high volume filter samples is relatively high ranging from 6.1 to 26.4 milligrams per filter. Therefore, a fraction of each of the high-volume filters, with OC masses ranging from 1.4 to 2.4 milligrams, was used for the extraction and the gas chromatography/mass spectrometry (GC/MS) analysis. Deuterated internal standard mixtures (IS) including 16 deuterated organic compounds (Table 1) were spiked into samples prior to extraction. The amounts of IS spikes were proportional to OC in the samples, with 250 µL of IS per one milligram of OC. The samples were extracted under mild sonication twice with hexane (Fisher Optima grade) and three times with distilled benzene/2-propanol (2:1 mixture, v/v, benzene: VWR Chromatographic grade; 2-propanol, Fisher Optima grade). The extracts were filtered, combined, and reduced in volume to about 5 mL by a rotary evaporator, then further concentrated to about 250 µL by nitrogen gas blowing down. The final extract was split into two fractions. One fraction was stored in the freezer,

3

and the other fraction was derivatized with diazomethane, followed by GC/MS analysis. More detailed information on sample handling, extraction and analysis is provided in Appendix B of this report.

Results and Discussion

Comparison of Elemental (EC) and Organic Carbon (OC) Concentrations in High Volume and Low Volume Samples at Tung Chung

In order to first determine the quality of the high volume samples their EC and OC concentrations are compared with the co-located ThermoAndersen sampler at Tung Chung used during the present study. Table 2 presents concentrations of solvent extractable organic tracer compounds in the high volume samples as well as monthly composite samples for comparison. The last four rows of Table 2 contain data for OC and EC concentrations measured for all of the high volume samples, as well as the ThermoAndersen sampler low volume concentrations measured at Tung Chung. The low volume (LV) samples at Tung Chung have OC concentrations over the four sampling events that are on average 6,634 ngm-3, as compared to 10,540 ngm-3 for the high volume (HV) samples. This discrepancy clearly indicates that the high volume samples contain significantly more organic carbon than the low volume samples. The average EC concentrations for the LV samples are 2375 ngm-3 as compared to 2482 ngm-3 for the high volume samples, indicating that EC concentrations are somewhat similar for the two samplers. The results generally indicate that the high volume filter samples are contaminated with an organic compound not present in appreciable amounts on the low volume samples. The following section discusses the contamination observed in the high volume filter samples.

4

Specific Organic Contaminants in High Volume Samples

Figures 4a and 4b show the spectra of the Tung Chung high volume sample collected on 10/20/02. An extremely large peak is observed with an extended retention time spanning 0.4 minutes from 25.4 to 25.8 minutes. The characteristic ion fragments are m/z 469, 313, and 197. The peak is observed in all high volume samples. The compound has never been found in such high abundance in our ambient samples. This compound is believed to be impactor grease (likely R&P WINS impactor oil). This oil was analyzed during the previous pilot study to insure that it did not contribute to contamination in low volume samples. Figure 5 shows the spectra of impactor oil previously analyzed in our lab. As shown in Figure 5, the spectrum is nearly identical to that observed in the current high volume samples. Although, it should be noted that the concentration in the high volume samples is nearly an order of magnitude greater than the concentration used in our analyses of the impactor oil. The organic compounds that clearly overlap the contaminant peak are shown in Table 3. Figure 6 is an example showing that the m/z 252 ion is so abundant in the high volume samples that it is impossible to correctly identify and quantify the compounds of benzo(e)pyrene, benzo(a)pyrene and perylene which using the same quantification ion of m/z 252. It is important to point out that it is also possible that the contaminant interferes with other compounds, in particular alkanes and PAHs, that are related to the combustion of fossil fuels. Below we present concentrations of solvent extractable organic compounds and it must be understood that some of the concentrations may be influenced by the contaminant. In addition, ratios of specific compounds to OC concentrations must also be used with caution since the contaminant had a significant influence on OC levels in the high volume samples. Before the actual impact of the contaminant on other organic concentrations can be determined more

5

detailed labwork is needed to definitively identify the compound, and to analyze the mass spectra of the compound at the concentration levels similar to that observed in samples. Nonetheless, we present results below with the knowledge that a contaminant, most likely impactor oil, may be influencing the concentrations of some of the organic tracer compounds.

In the following section concentrations of specific organic compounds are compared between the sites, as well as with the monthly composite concentrations measured based on low volume filter samples collected as part of this project. Again, it should be noted that the concentrations of the specific organic compounds for the high volume samples may be influenced by the contaminant present in the samples. Given that the contaminant is believed to be an oil, likely derived from petroleum products, the n-alkanes and PAHs may be particularly susceptible to contamination. Much of the discussion below is based on concentrations presented in Table 2.

Chemical characterization of the high volume samples at the TC site

March samples

We first begin by discussing the March 25th sampling day, which is characterized by extremely high fine particulate concentrations throughout the PRD. The concentrations of the total n-alkanes in the high volume and low volume TC samples are 20.51 and 32.73 ng/m3, respectively. The results suggest that the n-alkanes are present in roughly similar concentrations for the samples. The concentrations of the total fatty acids and hopanes + steranes, H+S, (Figure 2) also are very similar between the high volume and low volume samples, with values of 2.66 and 3.38 ng/m3 for H+S. Fatty acids are also in general

6

agreement with concentrations of 55.11 and 46.04 ng/m3, respectively. For the PAHs (Figure 3) and levoglucosan (Figure 1), however, lower concentrations are found in the high volume sample (4.58 and 39 ng/m3 for the total PAHs and levoglucosan, respectively) as compared with the low volume sample (10.69 and 75 ng/m3 for the total PAHs and levoglucosan, respectively). The values decrease for both PAHs and levoglucosan by a factor of around 2. Overall a general conclusion that can be drawn from the comparison is that many of the organic tracer compounds are within a factor of 2 between the high volume and low volume samples.

October samples

On October 20th low fine particulate concentrations are observed throughout the PRD with rain occurring throughout the region. The concentrations of the organic compounds in the high volume sample taken on October 20, 2002 (TC10202002) are much lower than that in the corresponding low volume monthly composite sample (TC-10). Some species such as C26-C30 n-alkanes, benzo(ghi)perylene, and coronene were even undetectable in the high volume sample. The concentrations of the total n-alkanes are 3.58 and 26.99 ng/m3 in the high volume and low volume samples, respectively. The fatty acids, PAHs, and H+S decrease also in concentration from 36.22 (fatty acids), 10.87 (PAHs), and 2.49 (H+S) ng/m3 in the low volume sample to 18.09 (fatty acids), 0.60 (PAHs) and 0.8 (H+S) ng/m3 in the high volume sample by factors of 2 – 18. The degree of decrease is even larger for levoglucosan (by a factor of ~40), from 299 ng/m3 in the low volume sample to 8 ng/m3 in the high volume sample. The lower concentrations in the high volume samples is not surprising given the abundance of rain throughout the region and generally low concentrations of fine particulate mass observed for the sampling day.

7

December samples

The high volume samples collected on December 13 and 19, 2002 (TC12132002 and TC12192002) represent meteorological cases of northerly flow and mixed north/south flow as discussed in the main report. The concentrations of the total n-alkanes and fatty acids are higher in the high volume samples than that in the low volume sample, with the highest values recorded in the TC12132002 sample (70.58 and 75.87 ng/m3 for n-alkanes and fatty acids, respectively). While for both of the total PAHs and levoglucosan, the highest concentrations are recorded in the low volume monthly composite sample (TC-12), with 12.27 and 250 ng/m3 for the total PAHs and levoglucosan, respectively. Somewhat lower concentrations of H+S are found in the high volume samples, as compared to the low volume samples. Although the sample taken on December 19, 2002 (TC12192002) is associated with relatively high particulate concentrations in Hong Kong, it does not correspond to the highest yields of organic compounds. The concentrations of n-alkanes and PAHs in that sample are the lowest among the three samples (TC12132002, TC12192002, and TC-12). The findings that the two high volume samples taken during December 2002 are characterized with similar levels of H+S, and lower levels of PAH’s and levoglucosan suggest that the events are perhaps not as influenced by biomass burning to the extent of the monthly composite sample.

Characterization of organic compounds at YL and comparison with TC

The concentrations of organic compounds including n-alkanes, branched alkanes, fatty acids, hopanes, steranes, resin acids, PAHs, alkanedioic acids, and other tracers such as levoglucosan are presented in Table 2. The comparison between the two sites will focus on

8

the compounds of n-alkanes, fatty acids, H+S, levoglucosan, and PAHs. H+S are used as molecular tracers for motor vehicles emission (Schauer et al., 1996), and they can be found in diesel fuel, and lubricating oil used by both diesel-powered and gasoline-powered motor vehicles (Cass, 1998). The ratio of H+S to EC is a useful indicator to determine some emission sources such as diesel-powered vehicles and gasoline-powered vehicles. From a previous studies (Schauer et al., 1999; Schauer et al., 2002; Zheng et al., in preparation), the (H+S)/EC ratio for diesel-powered vehicle exhaust (4.6e-4) is much lower than that for gasoline-powered vehicle exhausts (0.15). Levoglucosan is a major molecular tracer for biomass burning (Simoneit et al., 1999). PAHs come primarily from the incomplete combustion of organic material and the possible sources for PAHs include motor vehicles, biomass burning, power generation, and fossil fuels burning for heating purpose (Brun et al, 2004).

Different distribution patterns of organic compounds are found between the two sites. For the March samples, The YL sample (YL03252003) shows higher concentration of n-alkanes (53.40 ng/m3) and fatty acids (151.75 ng/m3) than the TC sample (TC03252003) (20.51 and 55.11 ng/m3 for n-alkanes and fatty acids, respectively), while the concentrations of levoglucosan, and H+S in the two samples are roughly the same. Concentrations of PAH’s at Yuen Long are more than two-fold greater than at TC. For the October high volume samples, the YL sample (YL10202002) has higher concentrations of PAHs than the TC sample (TC10202002), while the levels of levoglucosan and H+S are comparable between the two sites. For the December the YL samples generally have higher concentrations for most organic compounds than the corresponding TC samples (i.e., YL12132002 vs. TC12132002 and YL12192002 vs. TC12192002). For example, the concentrations of the total n-alkanes for YL12132002, YL12192002, TC12132002, and TC12192002 are 176.67, 163.96, 70.58,

9

and 21.45 ng/m3, respectively. Such patterns are also found for levoglucosan, PAHs, and fatty acids. The distribution pattern of H+S is an exception. The concentrations of H+S between the YL and TC samples collected on December 19, 2003 are comparable, while on December 13, 2002 the YL sample contains much more H+S than the TC sample (Figure 2). The relative similarities in the H+S concentrations between the two sites across all samples suggests similar mobile source contributions at the sites.

References

1. Brun, G.L., Vaidya, O.C., Leger, M.G., 2004. Atmospheric deposition of polycyclic aromatic hydrocarbons to Atlantic Canada: geographic and temporal distributions and trends 1980-2001. Environmental Science and Technology, 38, 1941-1948. 2. Cass, G.R., 1998. Organic molecular tracers for particulate air pollution sources. Trends in Analytical Chemistry, 17, 356-366. 3. http://www.lesker.com/newweb/Vacuum_Fluids/diffusion_pumps/KJL704.cfm?CFID= 327396&CFTOKEN=46003264§ion=diffsilicone&init=skip 4. Schauer, J.J., Rogge, W.F., Hildemann, L.M., Mazurek, M.A., Simoneit, B.R.T., 1996. Source apportionment of airborne particulate matter using organic compounds as tracers. Atmospheric Environment, 30, 3837-3854. 5. Schauer, J.J., Kleeman, M.J., Cass, G.R., Simoneit, B.R.T., 1999. Measurement of emission from air pollution sources. 2. C-1 through C-30 organic compounds from medium duty diesel trucks. Environmental Science and Technology, 33, 1578-1587. 6. Schauer, J.J., Kleeman, M.J., Cass, G.R., Simoneit, B.R.T., 2002. Measurement of emissions from air pollution sources. 5. C-1 – C-32 organic compounds from

10

gasoline-powered motor vehicles. Environmental Science and Technology, 36, 1169-1180. 7. Simoneit, B.R.T., Schauer, J.J., Nolte, C.G., Oros, D.R., Elias, V.O., Fraser, M.P., Rogge, W.F., Cass, G.R., 1999. Levoglucosan, a tracer for cellulose in biomass burning and atmospheric particles, 33, 173-182. 8. Zheng, M., Salmon, L.G., Schauer, J.J., Zeng, L., Zhang, Y., Kiang, C.S., Cass, G.R. Source apportionment of PM2.5 in Beijing, China by chemical mass balance (Manuscript in preparation).

11

Table 1. Concentrations of deuterated internal standards. Concentration Deuterated Isotope (ng/µL) dodecane-D26

437±21

hexadecane-D34

94.4±9.4

eicosane-D42

75.2±7.5

octacosane-D58

196±15

hexatriacontane-D74

406±20

decanoic acid-D19

180±13

heptadecanoic acid-D33

176±13

phthalic 3,4,5,6-D4 acid

184±13

benzaldehyde-D6

408±20

4,4’-dimethoxybenzophenone-D8

385±19

acenaphthene-D10

97.6±9.8

chrysene-D12

70.5±7.1

dibenz[ah]anthracene-D14

192±14

levoglucosan-U-13C6

1250±60

ααα-20R-cholestane-D4

15.0±0.8

cholesterol-2,2,3,4,4,6-D6

394±20

12

Table 2. Concentrations of organic compounds, EC, and OC in PM2.5 in Hong Kong (ng/m3). Compound heptadecane octadecane nonadecane eicosane heneicosane docosane tricosane tetracosane pentacosane hexacosane heptacosane octacosane nonacosane triacontane hentriacontane dotriacontane tritriacontane tetratriacontane pentatriacontane hexatriacontane iso-nonacosane anteiso-triacontane iso-hentriacontane tetradecanoic acid pentadecanoic acid hexadecanoic acid heptadecanoic acid octadecanoic acid nonadecanoic acid eicosanoic acid heneicosanoic acid docosanoic acid tricosanoic acid tetracosanoic acid pentacosanoic acid hexacosanoic acid heptacosanoic acid octacosanoic acid nonacosanoic acid triacontanoic acid 9-hexadecenoic acid 9,12-octadecanedienoic acid 9-octadecenoic acid 17α(H)-21β(H)-29-norhopane 17α(H)-21β(H)-hopane 22,29,30-trisnorneohopane 22,29,30-trisnorhopane 20S,R-5α(H),14β(H),17β(H)-cholestanes

TC 0325 2003

YL 0325 2003

TC03

TC 1020 2002

YL 1020 2002

D

D

M

D

D

TC10

TC 1213 2002

TC 1219 2002

YL 1213 2002

M

D

D

D

0.22 0.30

YL 1219 2002

TC12

D

M

0.15

0.25

0.02

0.07

0.53

1.51

0.25

0.26

0.32

2.65

2.39

0.33 0.43

0.70

1.23

0.33

0.71

0.66

0.41

0.77

0.72

3.58

3.85

0.51

0.72

0.88

0.51

0.60

0.53

0.28

0.44

2.55

2.97

0.45

0.81

1.22

1.31

0.66

0.62

0.60

0.58

3.10

3.09

0.74

0.93

2.50

2.96

0.46

0.74

1.08

1.28

0.82

6.75

4.99

1.86

1.08

3.41

3.96

0.98

0.96

3.37

0.92

10.35

7.51

3.45

1.97

7.15

4.77

0.35

1.11

1.53

8.86

1.69

25.36

15.12

5.04

2.52

8.41

3.68

1.21

1.74

11.81

2.26

26.85

19.57

4.68

1.63

3.92

2.70

0.77

2.75

8.71

1.80

12.85

13.71

4.55

1.33

2.72

1.60

1.53

1.84

7.03

1.78

8.53

10.08

3.43

1.89

5.44

2.23

18.35

18.60

4.37

0.91

2.14

0.95

1.84

3.09

5.91

3.01

0.83

1.66

0.88

1.09

2.66

1.06

0.52

1.23

0.76

2.89 0.32 0.25

0.72

1.21

4.98

1.02

7.08

8.71

1.85

4.63

7.77

3.42

19.93

22.38

4.61

0.67

1.31

4.32

1.12

6.71

7.07

1.77

1.06

2.29

3.99

1.85

9.37

9.34

1.82

0.66

0.61

2.06

0.70

4.19

4.86

0.93

0.42

2.05

0.92

1.44

1.59

0.75

0.78

0.89

0.39

2.18

2.32

0.55

0.53

0.46

0.18

0.36

0.21

0.09

1.14 0.85

0.32

0.83

0.42

0.39

1.04

0.55

1.35

3.54

1.92

0.66

1.07

0.30

0.19

0.94

0.82

0.30

0.50

0.63

0.59

3.55

3.26

1.26

3.27

0.57

0.37

0.28

10.31

8.13

0.64

0.85

3.31

1.14

0.27

2.98

0.25

0.13

0.08

6.30

6.43

0.23

20.49

66.72

18.44

8.76

43.11

6.34

28.33

12.63

298.31

170.29

12.38

0.68

1.91

0.52

0.34

1.17

0.30

0.72

0.47

8.55

5.66

0.39

7.38

23.78

6.81

3.77

10.17

2.71

15.01

6.19

133.76

71.02

4.10

0.28

0.78

0.25

0.11

0.28

0.20

0.50

0.28

3.51

2.49

0.21

1.14

3.13

1.06

0.28

0.93

0.98

2.04

1.07

16.40

11.06

1.00

0.40

1.04

0.33

0.07

0.33

0.47

0.83

0.40

5.58

4.53

0.43

1.94

4.97

1.58

0.26

1.27

2.30

3.32

1.90

24.54

19.88

1.96

1.46

3.29

1.01

0.18

0.80

1.39

2.49

1.67

16.11

17.07

1.16

1.09

3.98

8.77

2.75

1.12

2.17

0.64

2.13

3.32

6.50

3.74

41.00

37.79

3.02

0.47

0.74

1.38

1.22

9.16

10.94

0.63

2.83

6.07

1.72

0.97

1.94

1.99

1.54

23.77

25.75

1.63

1.16

2.16

0.42

0.38

0.82

1.09

0.95

7.10

9.85

0.62

3.84

7.09

1.67

1.62

4.58

4.48

3.43

28.98

32.73

2.95

0.88

1.77

0.31

0.32

0.92

1.15

1.04

5.75

8.75

0.54

3.59

6.46

1.24

5.19

4.44

31.49

38.47

4.34

1.35

3.65

0.83

1.13

0.41

0.65

0.69

0.48

2.69

0.42

0.43

0.59

0.64

0.33

1.30

7.94

1.04

0.32

0.38

0.44

0.12

0.47

1.82

0.49

0.25

0.09

0.32

0.39

0.61

0.29

0.15

0.50

0.59

0.12

0.27

0.12

6.60 0.33

0.34

1.45

2.82

2.27

3.68

0.20 0.09

0.55

0.88

0.35

0.49

0.64

1.38

0.56

0.75 0.38 0.18

0.30

0.14

13

Table 2. (Continued) Compound 20R-5α(H),14α(H),17α(H)-cholestane 20S,R-5α(H),14β(H),17β(H)-ergostanes 20S,R-5α(H),14β(H),17β(H)-sitostanes 22S,17α(H),21β(H)-homohopane 22R,17α(H),21β(H)-homohopane 22S,17α(H),21β(H)-bishomohopane 22R,17α(H),21β(H)-bishomohopane 22S,17α(H),21β(H)-trishomohopane 22R,17α(H),21β(H)-trishomohopane nonanal acetonylsyringol coniferyl aldehyde propionylsyringol benz(de)anthracen-7-one cholesterol levoglucosan 8,15-pimaredienoic acid pimaric acid sandaracopimaric acid isopimaric acid dehydroabietic acid abietic acid abieta-6,8,11,13,15-pentae-18-oic acid abieta-8,11,13,15-tetraen-18-oic acid 7-oxodehydroabietic acid 1,2-benzenedicarboxylic acid 1,4-benzenedicarboxylic acid 1,3-benzenedicarboxylic acid fluoranthene acephenanthrylene pyrene retene methyl substituted MW 202 PAH benzo(ghi)fluoranthene cyclopenta(cd)pyrene benz(a)anthracene chrysene/triphenylene methyl substituted MW 228 PAH methyl substituted MW 226 PAH benzo(b)fluoranthene benzo(k)fluoranthene benzo(j)fluoranthene benzo(e)pyrene benzo(a)pyrene

TC 0325 2003

YL 0325 2003

D

D

TC-03

TC 1020 2002

YL 1020 2002

TC10

TC 1213 2002

TC 1219 2002

YL 1213 2002

YL 1219 2002

M

D

D

M

D

D

D

D

0.18

TC -12 M 0.05

0.20

0.16

0.20

0.36

0.34

0.29

0.07

0.26

0.21

0.38

0.63

0.36

0.34

0.28

0.36

0.29

0.32

0.34

0.26

0.13

0.10

0.23

0.26

0.37

0.62

0.30

0.32

0.12

0.09

0.25

0.23

0.30

0.50

0.28

0.22

0.24

0.19

0.26

0.08

0.36

0.18

0.21

0.44

0.27

0.27

0.22

0.21

0.16

0.07

0.17

0.15

0.29

0.21

0.16

0.17

0.16

0.11

0.06

0.09

0.18

0.23

0.23

0.22

0.09

0.11

0.07

4.38

7.00

2.23

2.73

23.40

27.47

4.34

16.50

7.90

5.18

16.58

0.05 1.27

4.88

0.21

1.63

5.99

6.16

1.16

5.09

1.90

4.66

14.58

10.64 5.97

23.10

39

53

9.62 1.03

5.61 15.51

5.98

109.48

47.95

299

145

52

289

107

0.07

0.31

75

8

7

0.92 250

0.91 0.17

0.14

0.07

0.25

0.34

0.16

0.09

0.84

0.75

0.42

0.18

1.80

0.95

0.49

0.15

2.37

2.01

0.04

0.61

4.72

1.08

0.38

2.02

16.80

10.18

2.23

37.04

23.13

1.75

0.10

0.58

0.13

1.76

0.78

0.36 0.17

1.32 6.23

6.45

0.10

0.21

0.09

0.04

0.12

0.03

1.29

2.87

0.91

0.31

11.89

9.52

3.98

8.14

16.93

1.27

2.24

0.27

0.40

0.47

0.06

0.10

0.10

0.29

0.48

0.52

1.05

6.13 0.13

0.14

0.58

0.13

0.04

0.27

0.24

0.10

0.03

0.13

0.10

0.04

0.67

0.49

1.14

0.70

1.88

0.64

13.06

8.89

0.84

1.23

1.68

6.38

10.57

12.65

57.82

73.20

9.70

4.09

0.44

3.96

7.86

15.23

10.66

102.35

103.81

6.85

0.49

0.18

0.56

0.68

1.49

1.33

8.74

8.18

0.93

0.05

0.09

0.38

0.45

0.20

1.91

1.50

0.49

0.03

0.09

0.09

0.06

0.48

0.23

0.10

0.13

0.40

0.43

0.22

1.97

1.49

0.03

0.03

0.05

0.02

0.16

0.49 0.03

0.43

0.57

0.45

0.13

0.24

0.42

0.50

0.42

2.18

2.06

0.48

0.18

0.35

0.32

0.06

0.10

0.21

0.22

0.14

1.20

0.86

0.28

0.69

1.96

2.28

0.14

0.44

2.89

1.10

0.63

6.35

4.64

2.87

0.15

0.37

0.30

0.03

0.14

0.19

0.21

0.14

1.40

1.03

0.26

0.36

0.74

0.61

0.14

0.27

0.44

0.51

0.37

2.78

2.39

0.57

0.39

0.75

0.41

0.34

0.40

0.53

0.33

2.34

2.85

0.55

0.33

0.81

0.24

0.19

0.25

0.64

0.17

2.76

2.47

0.35

0.25

0.89

0.62

0.11

0.66

0.40

0.21

3.56

2.71

0.76

0.14

0.71

0.63

0.12

0.58

0.58

0.18

3.11

1.59

0.76

0.03

0.10

0.11

0.02

0.13

0.06

0.02

0.39

0.25

0.14

0.60

0.62

0.66

0.50

0.48

14

0.13 0.77

2.21

1.43

0.54

Table 2. (Continued) Compound

TC 0325 2003 D

perylene indeno(cd)fluoranthene indeno(cd)pyrene picene benzo(ghi)perylene coronene propanedioic acid methylpropanedioic acid butanedioic acid methylbutanedioic acid pentanedioic acid hexanedioic acid heptanedioic acid octanedioic acid nonanedioic acid OC EC OClowvol EClowvol PM2.5lowvol

YL 0325 2003

TC03

TC 1020 2002

YL 1020 2002

D

M

D

D

0.11

0.07

TC10

TC 1213 2002

TC 1219 2002

M

D

D

0.11

YL 1213 2002

YL 1219 2002

TC -12

D

D

M

0.40

0.22

0.13

0.11

0.31

0.27

0.08

0.27

0.25

0.09

1.52

0.91

0.37

0.26

0.92

0.75

0.13

0.85

0.74

0.24

4.26

2.32

0.84

0.03

0.09

0.10

0.02

0.08

0.08

0.03

0.47

0.16

0.06

0.25

0.96

0.75

0.18

0.80

0.69

0.25

4.47

2.71

0.89

0.21

0.66

0.54

0.11

0.56

0.48

0.16

3.14

2.09

0.54

5.60

18.57

16.82

0.25

0.26

16.36

17.42

7.75

1.20

2.25

6.61

14.98

14.36

77.61

75.50

10.38

1.80

1.66

1.13

0.22

0.47

0.56

1.32

1.41

3.11

2.71

3.08

0.21

0.36

1.81

2.93

2.38

7.00

8.93

1.03

15.13

16.88

1.98

1.72

1.01

0.25

0.48

1.19

1.60

1.20

9.03

2.80

6.54

1.43

0.85

1.36

0.40

0.38

0.41

1.24

0.83

6.38

4.68

0.39

1.07

1.58

0.48

0.42

0.79

0.61

1.62

0.68

9.22

4.95

0.68

5.08

9.88

2.31

1.15

2.83

2.38

11.65

3.95

70.17

27.66

3.32

10136

13559

6873

6932

3756

7504

13732

11360

16210

13068

6716

2391

2690

2137

1524

2786

2023

2458

3556

2746

3062

2560

9657

7730

7921

1229

2980

860

2930

2730

44840

9260

37260

37760

Notes: D, daily high volume sample; M, low volume monthly composite sample. TC, Tung Chung; YL, Yuen Long

15

Table 3. Summary of organic compounds that overlapped by ion peaks from 1,1,1-methylphenyl-2,2-dimethyl-3,3,3-methylphenyltrisiloxane.

Sample name

Nonacosane

Hexacosanoic acid

Benzo(e)pyrene

Benzo(a)pyrene

Perylene

Characteristic ion TC-10202002 TC-12132002 TC-12192002 TC-03252003 YL-10202002 YL-12132002 YL-12192002 YL-03252002

m/z 57,71,85 Overlap

m/z 74, 87 Overlap

m/z 252 Overlap Overlap Overlap Overlap Overlap Overlap Overlap Overlap

m/z 252 Overlap Overlap Overlap Overlap

m/z 252 Overlap Overlap Overlap Overlap

16

400 October, 2002 LowVol

300

HighVol

3

Concentration (ng/m)

December, 2002

LowVol

200 HighVol

March, 2003

HighVol

100

LowVol

HighVol

HighVol

YL10202002

HighVol

TC10202002

HighVo l

HighVol

TC-12

YL12192002

YL12132002

TC12192002

TC12132002

TC-10

TC-03

YL03252003

TC03252003

0

Figure 1. Distribution of levoglucosan in the high volume and low volume samples. The concentrations of levoglucosan for the low volume samples are from the corresponding low volume monthly composite samples. TC and YL stand for Tung Chung and Yuen Long sampling sites, respectively.

17

6 December, 2002 HighVo l

March, 2003

4

LowV LowVol

3

HighVol

October, 2002

HighVol

HighVol HighVol

LowVol HighVol

1

HighVol

HighVol

YL10202002

2

TC10202002

3

Concentration (ng/m)

5

TC-12

YL12192002

YL12132002

TC12192002

TC12132002

TC-10

TC-03

YL03252003

TC03252003

0

Figure 2. Distribution of the total hopanes and steranes (H+S) in the high volume and low volume samples. The concentrations of H+S for the low volume samples are from the corresponding low volume monthly composite samples. TC and YL stand for Tung Chung and Yuen Long sampling sites, respectively.

18

60 December, 2002 HighVol

3

Concentration (ng/m)

50

40 HighVol

30 March, 2003 October, 2002

20 HighVol

LowVol

LowVol

LowVol HighVol

10 HighVol HighVol

HighVol

HighVol

TC-12

YL12192002

YL12132002

TC12192002

TC12132002

TC-10

YL10202002

TC10202002

TC-03

YL03252003

TC03252003

0

Figure 3. Distribution of the total PAHs in the high volume samples and low volume samples. The concentrations of the total PAHs for the low volume samples are from the corresponding low volume monthly composite samples. TC and YL stand for Tung Chung and Yuen Long sampling sites, respectively.

19

Abundance TIC : TC 1020.D 1.1e+07

(a)

1.05e+07 1e+07 9500000 9000000

1

8500000 8000000 7500000 7000000 6500000 6000000 5500000 5000000 4500000 4000000 3500000 3000000 2500000 2000000 1500000 1000000 500000

6.00

8.00 10.0012.0014.0016.0018.0020.0022.0024.0026.0028.0030.0032.00

Tim e-->

Abundance Scan 3741 (25.460 min): TC1020.D 469 9000

(b)

8000 7000 6000 5000 313

197

4000 3000

149

2000

259

1000 0

51 50

77 105126 100

227

169 150

200

391 332 361

283

250

300

350

410 437 400

488

450

m/z--> Abundance #107755: 1,1,1-METHYLPHENYL-2,2-DIMETHYL-3,3,3-METHYLPHENY... 469 9000

(c)

8000 7000 6000

197

5000 4000 3000

149

313

2000 259 1000

392

227

0 50

100

150

200

250

300

350

400

450

m/z-->

Figure 4. a) Chromatogram of Tung Chung High Volume sample showing the contaminant Peak and b) Mass spectra of contaminant

20

Figure 5. a) Chromatogram and b) Mass spectra of R&P WINS Impactor oil

21

Abundance Ion 252.00 (251.70 to 252.70): TC1020.D 65000 60000 55000 50000 45000 40000 35000 30000 25000 20000 15000 10000 | | |

5000 0 24.80

25.00

25.20

25.40

| | | 25.60

25.80

26.00

26.20

Time-->

Figure 6. In TC10202002 sample, high m/z 252 ion peak from the contaminant covers all retention time intervals of Benzo(e)pyrene (BeP), Benzo(a)pyrene(BaP) and Perylene. The typical retention time of BeP, BaP, and Perylene is 25.49, 25.57, and 25.76 minutes respectively, and normally their peak width are 0.10, 0.09, and 0.07 minutes.

22

APPENDIX G Chemical Characterization of Water-Soluble Organic Aerosols in the Pearl River Delta Region Final Report Submitted by Dr. Jianzhen Yu Assistant Professor Department of Chemistry Hong Kong University of Science & Technology, Clear Water Bay, Kowloon, Hong Kong, Tel: 2358-7389, fax: 2358-1594.

August 2004 Version 3

Table of Contents List of Figures ................................................................................................................................ iii List of Tables ................................................................................................................................. iv Summary ..........................................................................................................................................v 1.

Introduction..............................................................................................................................1

2.

Extraction and analysis scheme ...............................................................................................3

2.1. Determination of aerosol total nitrogen .......................................................................... 3 2.2. Analysis of WSOC.......................................................................................................... 3 2.3. Analysis of cationic and anionic species ........................................................................ 3 2.4. Analysis of aliphatic amines and amino acids ................................................................ 3 3. Major inorganic ionic species ..................................................................................................4 3.1. Interlaboratory comparison............................................................................................. 4 3.2. Abundance, seasonal and spatial variation ..................................................................... 5 4. Total WSOC content................................................................................................................8 5.

Low molecular weigh organic acids ......................................................................................11

5.1. C2-C5 dicarboxylic acids.............................................................................................. 11 5.2. Formic, acetic, and lactic acid....................................................................................... 15 5.3. Methanesulfonic acid .................................................................................................... 15 6. Aerosol organic nitrogen........................................................................................................19 7.

Amino acids and aliphatic amines .........................................................................................23

8.

References..............................................................................................................................28

9.

Acknowledgements................................................................................................................30

List of Appendices .........................................................................................................................31

ii

List of Figures Figure 1. Location map of the sampling sites ................................................................................. 1 Figure 2. Derivatization reaction of –NH2 containing compounds to form fluorescent derivatives ................................................................................................................................................. 4 Figure 3. Comparison of three anions and one cation measurements made by UST and CIT ....... 6 Figure 4. Spatial variation of WSOC and OC concentrations. ..................................................... 10 Figure 5. Spatial variation of WSOC/OC and OC/EC.................................................................. 10 Figure 6. Correlation between WSOC and sulfate........................................................................ 11 Figure 7. Example anion IC chromatograms of the aerosol water extracts .................................. 12 Figure 8. Correlations between various dicarboxylic acid species ............................................... 14 Figure 9. Correlation of MSA and nitrate..................................................................................... 18 Figure 10. Time series of ON concentrations ............................................................................... 20 Figure 11. Correlation plot of sulfate versus ON.......................................................................... 23

iii

List of Tables Table 1. Site characteristics of the sampling locations ................................................................... 2 Table 2. Concentrations (µg/m3) of inorganic cations and anions in the aerosol samples ............. 7 Table 3. WSOC concentrations and relative abundance to other species....................................... 9 Table 4. Concentrations of low molecular weight organic anions (ng/m3) in the aerosol samples ............................................................................................................................................... 13 Table 5. Concentrations of MSA (ng/m3) in PM2.5 in two field studies...................................... 17 Table 6. Total ON Concentrations (µgN/m3)................................................................................ 19 Table 7. Statistics of ON/TN, ON/aerosol mass, and ON/OC..................................................... 21 Table 8. Correlation coefficient matrix of ON with other species in aerosols ............................. 22 Table 9. Concentrations of free amino acids and aliphatic amines (ng/m3) in the aerosol samples ............................................................................................................................................... 25 Table 10. Correlation matrix of various FAAs and aliphatic amines ........................................... 26 Table 11. Contribution of FAAs and aliphatic amines to organic nitrogen.................................. 27

iv

Summary As part of the overall effort to characterize the chemical properties of aerosols in the Pearl River Delta (PRD) region, we have analyzed water-soluble components of the aerosol samples collected in four sampling months spread over a year. A sampling network of seven sites across the PRD region was selected and sampling was carried out every six days in the months of October and December of 2002 and March and June of 2003. Individual daily filter samples were measured for total aerosol nitrogen (TN). The remaining filter materials in each month were combined together to yield water extracts of monthly composites. The water extract samples were analyzed for total water-soluble organic carbon (WSOC) content and individual inorganic and organic ions. This report describes our analysis results and provides preliminary data interpretation. The major observations and preliminary conclusions are summarized below. •

Aerosol organic nitrogen (ON) for individual samples was obtained as the difference of total aerosol nitrogen and the sum of two inorganic nitrogen species (NH4+ and NO3-). The ON concentration varies from non-detectable to 4 µgN/m3, with an average value of 1.2 µgN/m3. The ON fraction in TN was highly variable among individual samples, ranging from negligible to 60%. ON was found to be positively correlated with sulfate, suggesting that a significant fraction of ON was of secondary origin.



Our measurements of three inorganic anions (SO42-, NO3-, and Cl-) and one cation (NH4+) in the monthly composite samples were compared with those computed using the concentrations of individual daily samples as measured by Lynn Salmon at CIT. The two sets of measurements were in agreement within 15%.



The WSOC concentration varied from 1.7 to 5.2 µgC/m3, with an average value of 3.0 µgC/m3 for the region. This fraction accounted for 22-54% of OC, indicating that WSOC is a significant fraction of OC mass. A few lines of evidence are available from the measurement data to suggest that the WSOC fraction is mostly secondary organic carbon.



Nine low molecular weight organic acids were quantified. They include formic, acetic, lactic, methanesulfonic, oxalic, malonic, meleanic, succinic, and glutaric acid. The sum of these acids accounted for 3.5-9.1% of WSOC on a carbon mass basis. Oxalic acid was found to be the most abundant organic acid, ranging from 182 to 481 ng/m3. No discernable spatial contrast was found between rural and urban locations for the dicarboxylic acids, indicating the dominance of secondary sources.



Methanesulfonic acid (MSA) is a known oxidation product of dimethylsulfide, a compound emitted from marine water. Its concentration varied from 7.5 to 28.7 ng/m3. The principal controlling factor for the abundance of MSA in the region was found to be concentrations of atmospheric oxidants (OH and NO3). The temporal and spatial variation pattern in the atmospheric oxidants explains that higher MSA was observed in the winter than in the summer at all sites and that MSA was more abundant at two inland locations (Guangzhou and Zhongshan) than at the four coastal sites.

v



A few free amino acids and aliphatic amines were detected in the aerosol samples. The sum of these two groups of amine compounds had a concentration ranging from 29-79 ng/m3, accounting for a very minor fraction of WSOC (0.6%) or ON (1.2%).

Policy-relevant finding: The secondary nature of WSOC implies that the effective reduction of an important fraction of PM2.5 aerosol mass (11-24%) requires identification of their precursors and their formation mechanism and rates in the atmosphere.

vi

1. Introduction As part of the overall effort to characterize the chemical properties of aerosols in the Pearl River Delta (PRD) region, we analyzed water-soluble components of the aerosol samples collected in the project (called the PRD project hereafter). This report summarizes our analysis results and provides preliminary data interpretation. The background information and sampling details are provided in the report by Bergin et al. (2004). Prior to the PRD project, the Hong Kong Environmental Protection Department carried out an aerosol characterization project that involved three sampling locations in Hong Kong for a period of 12 months (called the 12-month HKEPD project hereafter). Our group did similar chemical characterization work for the 12month HKEPD project. Results of the 12-month HKEPD project are relevant to the understanding of the PRD project. As a result, we frequently cite information obtained from the first project in this report. A location map of the sampling sites in both projects is provided in Figure 1 and the site characteristics are described in Table 1.

Figure 1. Location map of the sampling sites (Solid circles are sampling sites in the PRD project; open circles represent the sites used in the 12-month HKEPD project. Credit: L. G. Salmon, G. S. Willis, M. Zheng, M. H. Bergin, Standard Operating Procedures for Sample Preparation, Handling, and Analysis during the Pilot Study in Hong Kong and the Pearl River Delta. )

1

Table 1. Site characteristics of the sampling locations Location

Abbreviation

Distance to

Site characteristics

coast/Pearl River Hong Kong 12-month project study Period: November 2000 to October 2001 Hok Tsui

HT

Within 1 km

rural area

Tsuen Wan

TW

within a few kilometers

urban area

Mong Kok

MK

within a few kilometers

Roadside in a downtown area

PRD project study Period: October 2002; December 2002; March 2003; and June 2003 Tap Mun

TM

within 1 km

rural area, on an island without local vehicular traffic

Tung Chung

TC

within a few kilometers

residential area on Lantau Island

Central Western

CW

within 1-2 km

urban area

Shenzhen

SZ

within a few kilometers

suburban area

Guangzhou

GZ

20-30 km

urban area

Conghua

CH

50-60 km

rural area with a small village

Zhongshan

ZS

20-30 km

urban area, inside a park

2

2. Extraction and analysis scheme We were allocated one quartz filter from each sampling day at each of the seven locations in the sampling network. A punch of 1x1 cm in size was removed from each filter for analysis of aerosol total nitrogen (TN). The filters in a same sampling month at each site were then combined before water extraction. Each monthly composite was extracted with 10.0 mL of ultra pure water in a sonication bath. This produced a total of 28 monthly water extracts, four extracts for each of the seven sites. The water extract samples were analyzed for the following chemical components: • Total water-soluble organic carbon (WSOC) content • Low molecular organic acids (formic, acetic, methanesulfonic acid, and C2-C5 dicarboxylic acids). • Major inorganic anions (sulfate, nitrate, and chloride) • Major inorganic cations (NH4+, Na+, K+, Ca2+, and Mg2+). 2.1. Determination of aerosol total nitrogen Aerosol TN was determined using an aerosol nitrogen analyzer developed by our group. The description of the instrument and characterization of the instrument performance are provided in the paper by Li and Yu [2004]. The instrument uses rapid thermoevolution of aerosol nitrogenous species at 800oC in a 2.5% O2/ 97.5% He carrier gas. Evolved nitrogen is oxidized to nitrogen oxides on a manganese dioxide catalyst, converted to nitrogen monoxide on a molybdenum catalyst, and quantified with a chemiluminescence detector. This system is able to provide fast (3 min per sample) and highly sensitive (a detection limit of 26 ng N) TN measurements for aerosol samples of a small size (~ 1×1 cm) without any pretreatment. 2.2. Analysis of WSOC The standard operating procedure for determination of WSOC is given in Appendix 1 and relevant discussion on the method can be found in the paper by Yang et al. [2003]. Below is a brief description. The WSOC content in each aerosol water extract was measured by a TOC analyzer under the normal catalyst mode. The injection volume was 35 µl. The TOC has a detection limit of 0.88 µgC/ml for an injection volume of 35ul. 2.3. Analysis of cationic and anionic species The cationic and anionic species were analyzed using ion chromatography (IC). Appendix 2 describes the detailed analysis conditions. The cationic species include NH4+, Na+, K+, Ca2+, and Mg2+. The anionic species encompasses both inorganic anions (sulfate, nitrate, and chloride) and organic anions (formate, acetate, methanesulfonate, oxalate, malonate, maleate, succinate, and glutarate). The IC method could not differentiate the free acid and its dissociated ion; consequently, the concentrations reported here represent the sum of the two forms for any acidic species. 2.4. Analysis of aliphatic amines and amino acids These two classes of compounds share a common functional group, -NH2. As a result, both can be analyzed using Waters’ AccQ-Tag method (Cohen et al., 1993). The amine group reacts with 6-aminoquinolyl-N-hydroxysuccinimidyl carbamate (AQC) to produce stable fluorescent derivatives that fluoresce strongly at 395nm (Figure 2). The resulting derivatives of various amino acids and aliphatic amines can be resolved on a HPLC column using gradient elution of three eluents: acetonitrile, water, and phosphate buffer. 3

O NH

N

O C

N

NH

R1

+ H

O

N

R2

N

O

6-aminoquinolylN-hydrozysuccinimidyl carbamate AQC

o

O

N

O

R2

This derivative fluoresces strongly at 395nm.

o

1 or 2 amine

O

R1

C

+

HO N O

Figure 2. Derivatization reaction of –NH2 containing compounds to form fluorescent derivatives Besides free amino acids, proteinaceous materials, which consist of two or more amino acid units, could be present in the ambient aerosols. For example, Likens et al (1983) detected this kind of materials in rain samples. The proteinaceous materials can be regarded as combined amino acids and measured as their compositional amino acids through conversion to the free amino acids by hydrolysis. The detailed procedure for analyzing the –NH2 containing compounds is given in Appendix 3. A brief description is given below. 500 µl of aerosol water extract is transferred to a derivatization tube and evaporated to dryness under a gentle stream of N2. To the derivatization tube, 20 µl of internal standard and 160 µl of buffer are added. N2 is bubbled into the mixture for 50 min to remove NH3. Otherwise NH3 peak would interfere the determination of Histamine and Arginine if an aerosol sample contains a large amount of NH4+. Then 40 µl of the derivatization agent is added. The mixture is mixed with a vortex mixer for 1 min followed by heating at 55oC for 10 min. The sample is transferred to a 150 µl autosampler vial insert. Finally 10 µl is injected into a HPLC system for analysis. 3. Major inorganic ionic species 3.1. Interlaboratory comparison Although the focus of our work is to characterize the organic fraction in the water-soluble aerosol, the water-soluble inorganic anions and cations were also measured for each monthly composite sample. Lynn Salmon at California Institute of Technology (CIT) measured a common set of three anions (SO42-, NO3-, and Cl-) and one cation (NH4+) in individual daily samples. She used an IC method for the determination of the three anions and a colorimetric method for the analysis of ammonium. The concentrations in the individual samples reported by Salmon were used to compute the concentrations in the monthly composite samples so that a comparison could be made with our measurements. Figure 3 plots the concentrations measured by us versus those computed from CIT’s measurements. The two sets of measurements are in agreement within 15%. In addition, the data scattering of sulfate, nitrate, and ammonium is small, reflected by the high correlation coefficients (r2>0.91). The reasonable agreement between the two labs adds to our confidence in the IC measurements from both our lab and CIT. It is noted that chloride measurements had a lower correlation coefficient (r2=0.83) than the other ions. The dominant source of aerosol chloride in this area is sea salt, which is known to mainly reside in coarse particles [Yao et al., 2002]. As a result, chloride concentrations in PM2.5 aerosols were present at much lower levels than either of nitrate, sulfate and 4

ammonium. The average chloride concentration was 0.26 µg/m3, in comparison with 1.50, 10.7, and 2.93 µg/m3 for nitrate, sulfate, and ammonium, respectively. Ten out of 139 valid samples had chloride concentration fall below detection limit. Consequently, the chloride measurements by CIT had higher measurement uncertainties. This explains the higher degree of scattering of the chloride comparison data. 3.2. Abundance, seasonal and spatial variation The concentrations of cationic species (NH4+, Na+, K+, Ca2+, and Mg2+) and anionic species (SO42-, NO3-, and Cl-) are tabulated in Table 2. The sum of these ions accounted for 28 to 52% of the PM2.5 mass. SO42- was the most abundant species, varying from 6.22 to 14.18 µg/m3. The concentration of NH4+ followed closely that of SO42- (r2 = 0.81). It varied from 2.00 to 6.16 µg/m3, with an average value of 3.64 µg/m3. The annual mean ratio of NH4+/ SO42- (in charge equivalent) is 1.00 for the region, with the ratio slightly smaller than unity at the sites in Hong Kong and larger than unity at the sites in Guangdong. Nitrate also had a significant presence in PM2.5. The annual mean of nitrate in the region was 1.23 µg/m3. The two most abundant ionic species, SO42- and NH4+, contributed to an average of 83% of the sum of all ionic mass. In comparison, Na+ and Cl- were less abundant, indicating the minor contribution of sea-salt particles to the overall PM2.5 aerosol mass in the region. The annual mean concentration of Na+ at the coastal sites and the inland locations was 0.32 and 0.29 µg/m3, respectively. The slightly lower mass concentrations of Na+ at the inland locations confirmed that sea salt was the dominant source of Na+ in this coastal area. Sulfate exhibited a consistent spatial gradient, declining from GZ, ZS to the coastal sites including SZ and the three HK sites in all seasons. Nitrate and ammonium had a similar spatial trend to that of sulfate. GZ generally had the highest sulfate, nitrate, and ammonium levels among all the seven sites. CH is a rural site further inland than GZ and ZS. The sulfate level at CH was similar to that at ZS, but the nitrate concentration was sometime more than a magnitude lower than those at ZS and GZ. The levels of sulfate, nitrate, and ammonium were similar at the three HK sites in each season, consistent with the secondary nature of these species. Sulfate concentrations in December were generally higher than those in June with the exception of the GZ and CH sites. Higher nitrate and ammonium concentrations were observed in December than in June at all sites except for CH. Lower concentrations of air pollutants in the summer were anticipated when various meteorological factors were considered. In the summer, more frequent removal by wet deposition, high mixing height, and favorable dilution by cleaner marine air all lead to lower pollutant concentrations. It is noted that nitrate measurements in the sampling set-up used in this study are subject to negative sampling artifacts due to volatilization of ammonium nitrate after sample collection. Such negative sampling artifacts for filter-based sampling are well documented in the literature (e.g., Kim et al., 1999; Slaninas, 2004). It is also known that the magnitude of the nitrate loss is temperature dependent and more loss is expected at higher ambient temperature. A larger nitrate loss in the summer also contributed to the lower concentrations observed in this season.

5

8

3

y = 0.86x + 0.52

15

UST nitrate, ug/m

UST sulfate, ug/m

3

20

2

R = 0.94 10

5

0

y = 0.86x - 0.09

6

2

R = 0.99

4

2

0 0

5

10

15

0

20

2

3

6

8

6

8

3

CIT nitrate, ug/m

CIT sulfate, ug/m

2

8 y = 1.10x + 0.42 2

3

UST ammonium, ug/m

3

y = 1.02x - 0.00 UST chloride, ug/m

4

2

R = 0.83

1

0

R = 0.91

6

4

2

0 0

1

2

0

3

2

4

3

CIT chloride, ug/m

CIT ammonium, ug/m

Figure 3. Comparison of three anions and one cation measurements made by UST and CIT

6

Table 2. Concentrations (µg/m3) of inorganic cations and anions in the aerosol samples Site

Sanpling month TM Oct. Dec. Mar. Jun. TC Oct. Dec. Mar. Jun. CW Oct. Dec. Mar. Jun. SZ Oct. Dec. Mar. Jun. GZ Oct. Dec. Mar. Jun. CH Oct. Dec. Mar. Jun. ZS Oct. Dec. Mar. Jun. Annual mean

SO42-

NO3-

Cl-

Na+

NH4+

K+

Mg2+

Ca2+

7.516 10.409 9.027 6.220 7.355 8.957 9.408 7.858 7.474 10.130 8.533 7.690 8.016 12.038 9.201 7.853 12.781 12.532 12.860 14.181 9.131 6.807 10.244 10.458 10.769 12.993 11.724 8.464 9.665

0.150 0.174 0.339 0.069 0.291 0.731 1.183 0.086 0.393 1.033 1.176 0.168 1.014 2.746 2.936 0.857 1.617 4.158 5.958 2.307 0.202 0.143 0.480 0.306 0.651 2.608 2.386 0.143 1.225

0.023 0.043 0.033 0.005 0.033 0.060 0.109 0.034 0.050 0.175 0.171 0.077 0.130 0.486 0.671 0.183 0.159 1.251 0.829 1.006 0.027 0.022 0.013 0.031 0.094 0.533 0.855 0.024 0.254

0.240 0.401 0.368 0.305 0.204 0.341 0.352 0.326 0.220 0.354 0.369 0.325 0.312 0.315 0.361 0.397 0.300 0.285 0.356 0.523 0.096 0.094 0.235 0.292 0.315 0.333 0.338 0.314 0.310

2.336 3.420 3.238 1.997 2.498 3.148 3.702 2.704 2.479 3.904 3.445 2.711 3.023 5.303 4.410 3.063 4.406 5.462 6.155 4.764 2.836 2.330 3.928 3.393 3.590 5.573 5.111 3.050 3.642

0.757 0.611 0.365 0.265 0.741 0.609 0.333 0.326 0.715 0.551 0.305 0.309 0.843 0.842 0.566 0.358 1.571 1.501 1.238 1.343 1.218 0.928 1.309 1.202 1.226 1.090 0.608 0.184 0.783

0.052 0.071 0.048 0.038 0.044 0.059 0.045 0.042 0.059 0.072 0.048 0.045 0.054 0.060 0.045 0.047 0.042 0.040 0.040 0.047 0.037 0.018 0.033 0.031 0.062 0.039 0.044 0.034 0.046

0.135 0.110 0.107 0.079 0.155 0.124 0.128 0.116 0.165 0.143 0.171 0.100 0.301 0.236 0.223 0.159 0.373 0.462 0.246 0.334 0.363 0.147 0.235 0.226 0.282 0.193 0.180 0.108 0.200

7

PM2.5 mass 23.77 29.18 34.39 24.89 30.76 33.08 41.41 30.64 28.15 35.67 40.85 30.95 38.68 45.98 56.09 35.88 68.27 75.01 80.92 65.21 34.25 27.88 45.79 36.50 60.01 81.06 56.59 29.56 43.62

Σions/mass 47% 52% 39% 36% 37% 42% 37% 38% 41% 46% 35% 37% 35% 48% 33% 36% 31% 34% 34% 38% 41% 38% 36% 44% 28% 29% 38% 42% 37%

4. Total WSOC content WSOC concentrations varied from 1.7 to 5.2 µgC/m3, with a mean value of 3.0 µg/m3for the region (Table 3). The WSOC fraction accounted for 22-54% of OC, indicating that WSOC is a significant fraction of OC mass. The water-soluble organic aerosol mass, calculated from the concentration of WSOC by multiplying by a factor of 2.1 [Kiss et al., 2002], ranged from 2.6 to 7.9 µg/m3, with a mean value of 4.4 µg/m3. The mean water-soluble organic aerosol mass was more than half of the sulfate aerosol mass, demonstrating that the role of organic components in affecting hygroscopicity of ambient aerosols can not be overlooked. A few lines of evidence are available to suggest that WSOC is mostly secondary in nature. First, the WSOC concentration levels were similar among all sites while an obvious contrast was seen in OC levels between rural and urban sites (Figure 4). Second, the WSOC/OC ratio had a similar spatial pattern to that of OC/EC, i.e., the ratios were lower in the urban locations and higher in the rural locations (Figure 5). The two rural sites, TM and CH, had a notably elevated ratio in WSOC/OC and OC/EC. The OC/EC ratio is known to increase with increased contribution of secondary OC. Third, WSOC and sulfate were positively correlated, with a correlation coefficient r of 0.75 (Figure 6). It has long been established that sulfate is a secondary aerosol component, formed from oxidation of precursor SO2 [e.g., Seinfeld and Pandis, 1996]. The secondary origin of WSOC means that water-soluble organic compounds are largely not directly emitted from sources; instead they are mainly acquired either from oxidation of gaseous precursors or pre-existing aerosol OC materials. The reduction in this significant portion of aerosol material (11-24% of aerosol mass) requires identification of their precursors and their formation mechanism and rates in the atmosphere. Chemical characterization of the WSOC fraction at a molecular level is an essential step towards unraveling the formation mechanism. As results in the subsequent sections indicate that only a small fraction of WSOC could be identified as known chemical identities, we are far from identifying the major sources (i.e., anthropogenic versus biogenic) for WSOC.

8

Table 3. WSOC concentrations and relative abundance to other species Site TM

TC

CW

SZ

GZ

CH

ZS

mean SD min max

Sampling month Oct. Dec. Mar. Jun. Oct. Dec. Mar. Jun. Oct. Dec. Mar. Jun. Oct. Dec. Mar. Jun. Oct. Dec. Mar. Jun. Oct. Dec. Mar. Jun. Oct. Dec. Mar. Jun.

WSOC (ugC/m3)

OC (ugC/m3)

2.76 3.00 2.23 1.71 3.29 2.70 2.48 1.95 2.80 2.13 2.48 2.07 3.79 2.80 2.96 2.26 5.04 4.21 4.29 3.63 3.06 2.25 4.48 2.82 2.74 5.24 3.27 1.96 3.01 0.94 1.71 5.24

5.74 5.53 4.79 3.52 7.54 6.75 6.88 4.06 6.44 7.87 7.48 4.69 12.36 11.69 12.15 6.64 20.19 19.28 16.96 13.86 8.88 7.81 12.54 8.12 11.65 15.15 11.30 4.15 9.43 4.60

WSOC/OC

OC/EC

2.1*WSOC/sulfate

2.1*WSOC/mass

48% 54% 46% 49% 44% 40% 36% 48% 44% 27% 33% 44% 31% 24% 24% 34% 25% 22% 25% 26% 34% 29% 36% 35% 23% 35% 29% 47% 32% 9% 22% 54%

6.47 4.98 6.95 5.91 3.72 2.63 3.22 3.29 3.22 3.23 4.04 3.48 3.05 2.95 3.07 2.69 4.35 3.64 3.78 4.55 5.87 5.27 7.91 6.98 5.54 3.84 4.56 2.97 4.36 1.45 2.63 7.91

0.71 0.56 0.46 0.50 0.88 0.62 0.49 0.47 0.72 0.41 0.52 0.54 0.91 0.47 0.59 0.57 0.75 0.63 0.55 0.53 0.65 0.61 0.63 0.57 0.49 0.80 0.52 0.45 0.59 0.12 0.41 0.91

22% 19% 14% 15% 24% 18% 13% 14% 20% 12% 13% 15% 19% 11% 11% 14% 16% 12% 11% 12% 18% 17% 18% 17% 14% 19% 12% 14% 15% 3% 11% 24%

9

3

WSOC or OC ( µgC/m )

25 WSOC

20

OC

15 10 5 0 TM

TC

CW

SZ

GZ

CH

ZS

Site

Figure 4. Spatial variation of WSOC and OC concentrations. (The error bars represent the maximum and minimum concentrations at a given site.)

60%

12 OC/EC

40%

8

20%

4

0%

0

OC/EC

WSOC/OC

WSOC/OC

TM

TC

CW

SZ

GZ

CH

ZS

Site

Figure 5. Spatial variation of WSOC/OC and OC/EC (The error bars represent the maximum and minimum ratios at a given site.)

10

6 y = 0.28x + 0.04 2

WSOC, µgC/m

3

R = 0.57 4

2

0 0

5

10

15

20

Sulfate, µg/m

3

Figure 6. Correlation between WSOC and sulfate 5. Low molecular weigh organic acids Nine low molecular weight organic acids were targeted for analysis by IC. They include lactic, formic, acetic, methanesulfonic acid (MSA), oxalic, malonic, meleanic, succinic, and glutaric acid. The IC method detects them in their ionic form. The IC chromatograms of two samples and one blank are shown in Figure 7 to illustrate their successful separation and detection by the IC method. Their concentrations in the aerosol samples are tabulated in Table 4. On a carbon mass basis, the sum of the organic ions accounted for 3.5-9.1% of WSOC. 5.1. C2-C5 dicarboxylic acids It is noted that the low molecular organic acids are semivolatile and therefore have sampling artifact problems common to any semivolatile organic compounds when a filter-based method is used for sample collection. The sampling problem aside, we found that oxalic acid was the most abundant organic acid species, accounting for approximately half of the sum of all the organic acids mass. Its levels were similar among all sites, ranging from 182 to 481 ng/m3. The lack of spatial contrast between urban and rural sites suggests that secondary sources dominated over primary sources. Oxalic acid had a moderate positive correlation with sulfate (Figure 8), further supporting the suggestion of the dominance of secondary formation pathways. Abundance of C3-C5 dicarboxylic acids was one-quarter to one-tenth that of oxalic acid. Each individual dicarboxylic acid also lacked a spatial gradient from urban to rural locations. As a result, they were most likely secondary in origin. Malonate had a moderate correlation with oxalate; glutarate had a weak correlation with oxalate; however, no correlation was seen between the two C4 dicarboxylic acids and oxalate. The weak to moderate correlation indicated that oxalate, malonate, and glutarate probably had common sources for their precursors. On the other hand, the lack of correlation between C4 dicaroxylic acids was an indication of different sources for their precursors.

11

Figure 7. Example anion IC chromatograms of the aerosol water extracts

12

Table 4. Concentrations of low molecular weight organic anions (ng/m3) in the aerosol samples Site TM

TC

CW

SZ

GZ

CH

ZS

mean SD min max a

Sampling month Oct. Dec. Mar. Jun. Oct. Dec. Mar. Jun. Oct. Dec. Mar. Jun. Oct. Dec. Mar. Jun. Oct. Dec. Mar. Jun. Oct. Dec. Mar. Jun. Oct. Dec. Mar. Jun.

oxalate

malonate

succinate

maleate

glutarate

MSA

Lactate a

formate

Acetate a

Σ(org ion) /WSOC a

349 285 290 225 336 302 286 232 316 320 283 234 347 387 380 290 481 270 326 365 327 182 316 236 385 346 425 345 317 65 182 481

53.5 50.4 57.9 42.6 57.9 65.7 62.2 41.6 62.9 63.2 60.9 43.3 72.3 89.5 72.1 44.5 99.9 62.6 65.3 74.5 86.0 0.0 66.7 61.5 45.3 43.0 59.6 44.5 58.9 18.7 0.0 99.9

47.7 54.7 75.0 62.4 44.4 38.2 76.0 74.4 56.5 31.4 62.0 7.1 34.3 45.2 59.5 56.8 67.0 39.6 68.6 82.9 20.0 0.0 69.9 59.7 29.8 0.0 63.4 53.3 49.3 22.6 0.0 82.9

12.6 20.9 19.8 16.7 14.5 29.6 42.2 20.1 8.8 20.2 16.9 6.7 13.2 19.8 44.1 7.3 33.9 198.4 0.0 6.9 38.4 51.1 53.4 64.6 28.9 87.4 35.3 22.4 33.4 37.8 0.0 198.4

38.7 48.9 44.5 10.5 36.9 39.2 35.0 11.4 35.0 41.9 30.4 10.5 39.5 49.6 31.9 11.5 37.0 40.8 43.6 10.9 40.6 25.1 28.6 14.4 0.0 46.4 35.5 14.6 30.5 14.2 0.0 49.6

13.3 20.6 9.1 9.2 15.3 18.6 13.5 8.0 14.1 21.2 14.8 7.5 16.3 19.8 17.6 9.6 22.2 27.9 16.9 14.7 13.7 15.9 10.7 8.5 23.0 28.7 16.9 7.9 15.6 5.8 7.5 28.7

41.1 0.0 0.0 1.4 48.7 19.1 0.0 1.9 43.6 23.0 0.6 0.0 177.8 54.1 6.3 0.0 74.5 0.0 0.0 0.0 55.7 7.4 6.7 2.4 111.7 24.1 1.6 3.7 25.2 41.1 0.0 177.8

40.0 39.2 31.7 21.5 49.3 38.5 29.5 27.4 39.1 30.0 27.2 19.2 58.2 47.6 36.6 32.8 63.7 37.3 34.5 39.6 39.4 13.0 34.4 31.2 17.7 14.2 37.3 10.5 33.6 12.5 10.5 63.7

43.5 0.0 37.0 8.5 60.2 44.7 10.6 8.6 71.4 58.8 8.6 8.5 82.9 87.9 3.3 9.1 104.1 0.0 31.1 0.0 70.4 2.9 20.0 9.2 5.9 0.0 4.0 0.1 28.3 31.9 0.0 104.1

7.2% 5.3% 8.1% 7.2% 6.4% 6.9% 7.1% 6.8% 7.4% 8.9% 6.3% 4.7% 7.3% 9.1% 6.7% 6.1% 6.2% 5.3% 4.2% 4.9% 7.2% 4.0% 4.3% 5.5% 7.2% 3.5% 6.3% 7.7% 6.3% 1.4% 3.5% 9.1%

Lactate and acetate partially coeluted; consequently their concentrations reported here are only for reference purpose. b On a carbon mass basis.

13

6

1.0 y = 0.12x + 0.15 y = 0.02x + 1.79

R = 0.32

3

Molonate, nmol/m

3

oxalate, nmol/m

2

2

R = 0.29 4

2

0.5

0.0

0 0

50

100

150

200

0

2

3

4

6

3

Sulfate, nmol/m

Oxalate, nmol/m

1.0

1.0 y = 0.01x + 0.40 2

3

Mealate, nmol/m

Succinate, nmol/m

3

R = 0.00

0.5

0.0

y = -0.05x + 0.49 2

R = 0.01

0.5

0.0 0

2

4

6

0

3

2

4 3

Oxalate, nmol/m

Oxalate, nmol/m

0.6 y = 0.06x + 0.04 2

glutarate, nmol/m

3

R = 0.17 0.4

0.2

0.0 0

2

4

6

3

Oxalate, nmol/m

Figure 8. Correlations between various dicarboxylic acid species

14

6

5.2. Formic, acetic, and lactic acid Lactate and acetate partially coeluted; a baseline separation of the two was usually not achieved (Figure 7). As a result, larger measurement uncertainties were associated with their quantification, which might explain the wide variation in their concentrations (from below detection to more than 100 ng/m3) (Table 4). Because of this analytical problem, we report here acetate and lactate only for reference purpose. No further attempt is made to explain their spatial and temporal variation. The concentration of formate ranged from 10 to 64 ng/m3. Similar to other acid species, formic acid also lacked a spatial variation among the sites, suggesting the dominance of secondary formation pathways over primary emission sources. It had a consistent seasonal variation pattern among the four coastal sites, i.e., Oct. > Dec. >Mar.> Jun. The seasonality in the inland locations was variable and different from that at the coastal sites. For example, the highest formate concentration at ZS was observed in the March composite, and the lowest formate concentration at CH was in the December composite. The underlying cause for the different seasonal behaviors at the inland locations, is not yet clear. 5.3. Methanesulfonic acid Table 5 lists the concentrations of MSA in the seasonal composites, along with MSA measurements made in the 12-month HKEPD project. The MSA level in the PRD project ranged from 7.5 to 28.7 ng/m3, with an average of 15.6 ng/m3. This level was similar to measurements made over Hong Kong by Arimoto et al. [1996]. Our measurement was lower than the average value of 29 ng/m3 (varying from 16 to 208 ng/m3) observed over the East China Sea, 36 ng/m3 at Qingdao, and 66 ng/m3 at Xiamen [Gao et al, 1996]. In the 12-month HKEPD project, the concentration levels of MSA were similar at all three sites. The seasonal trend was the same at all three sites, i.e, MSA concentration being winter > spring > summer > fall. The similar concentration levels at all three sites were consistent with its oceanic origin and photooxidation production pathway. In the PRD project, a spatial gradient was discernable between coastal and inland locations. Higher concentrations were observed at two of the inland sites (GZ and ZS) in the October and December monthly composite samples than at the four coastal sites (TM, TC, CW, and SZ). Among the four coastal sites, TM is located on an island with little local emission, while the other three sites (TC, CW, and SZ) are in urban environments. MSA concentrations at the three urban coastal sites were similar to those at the rural coastal site, consistent with the observation in the 12-month HKEPD project. This again reflects the marine source of its precursor and the secondary nature of MSA. The same seasonal trend was observed in the PRD project, i.e., higher MSA concentrations during the winter months than during the summer months. The MSA concentration in the winter is 1.7 times that in the summer in the 12-month HKEPD project and 2.3 times in the PRD project. Such a winter-high-summer-low seasonality contrasts most observations of seasonal variation reported in the literature. Most studies in the literature found a seasonal trend of aerosol MSA concentrations being the highest during summer and lowest in winter as a result of higher emission strength of dimethylsulfide (DMS) corresponding to increased primary productivity and enhanced photooxidation rate of DMS in the summer.

15

The MSA concentration in aerosols is determined by the relative strength of formation and removal processes, and gas-aerosol partitioning. The production rate of MSA is controlled by source strength of the DMS precursor and the formation rate from DMS. We examine below the various factors influencing the MSA aerosol concentrations in an effort to explain the seasonality. Measurements of DMS air concentrations in the Pearl River Estuary and the adjacent northern South China Sea by Ma et al. [2004] indicated similar DMS levels in May and in November, with the mean DMS concentration in May 30% lower than that in November. In the summer, the prevailing wind brings more marine air mass to the PRD region. Despite the ambiguous winter-summer variation in the input strength of DMS to the PRD, there are a number of factors that could enhance MSA abundance in aerosols in the winter. (1) MSA production rate is enhanced in the winter due to higher atmospheric oxidant concentrations and due to cold temperatures favoring the MSA production pathway over other oxidation pathways of DMS. MSA is one of the atmospheric oxidation productions of DMS by either OH radical or NO3 attack. A portion of DMS in the range of 5-50% is eventually oxidized to MSA [Saltzman et al., 1986]. An MSA yield of 46% from the reaction of NO3 with DMS was reported in a laboratory study [Jensen et al., 1991]. The oxidation of DMS by the OH radical proceeds by either addition or abstraction [Hynes et al., 1986; Allen et al., 1997; Seinfeld and Pandis, 1998]. Addition of the OH radical to DMS produces MSA as a major product. The abstraction route leads to formation of SO2 as a major product and MSA as a minor product [Yin et al., 1990]. The relative importance of the abstraction and the addition pathway is temperature dependent, with the abstraction pathway favored at higher temperature. According to the rates measured by Hynes et al. [1986], a temperature decrease from 25oC to 5oC should result in a factor of 3.8 increase in the branching ratio of the addition to abstraction. As a result, the colder the temperature, the more favored the path to MSA formation [Seinfeld and Pandis, 1998; Berresheim, 1987; Bates et al. 1992]. The oxidation rate of DMS is clearly proportional to the concentrations of the oxidants, the OH radical and NO3. The seasonal abundance variation of the two oxidants were reflected in that of nitrate. It is known that nitric acid is mainly formed through the reaction of OH and NO2 during daytime and a heterogeneous pathway involving NOx, O3, and NO3. The nitrate has been considered a tracer of anthropogenic pollution [Savoie et al., 1992]. A positive correlation (r2 = 0.44) was found between our measurements of aerosol nitrate and MSA (Figure 9), confirming that the production of nitrate and MSA is influenced by the same oxidants. Nitrate was consistently observed at all sites to be higher in the winter months than in the summer months (Table 2), suggesting a higher oxidant level in the winter. It is important to note that nitrate measurements in this study were subjected to negative sampling artifacts and the artifacts were more prominent in the summer months than in the winter months. The above suggestion of higher oxidant levels in the winter could be true only if the magnitude of the sampling artifacts was not so big as to mask the seasonal difference in nitrate concentrations. Further studies, in which special sampling set-up (e.g.,

16

denude followed by a nylon filter), is adopted to minimize the sampling artifacts, are needed to verify whether nitrate has a winter-high-summer-low seasonal pattern. (2) The removal of MSA through wet and dry deposition is reduced in the winter as a result of much lower rainfall in this season. Rainfall in this region has a stark seasonal contrast, with summer generally characterized as a wet season and winter as a dry season. The more frequent wet precipitation in the summer certainly facilitates the removal of MSA. (3) Lower mixing height in the winter also enhances MSA concentration in this season. (4) The lower temperatures in the winter favor the partitioning of MSA in the aerosol phase. The seasonal mean ambient temperature had a considerable variation during the sampling year from October 2002 to June 2003, varying from 17.8 to 27.9oC. The higher concentrations seen at the inland sites could be explained by the secondary nature of MSA. MSA, as a secondary pollutant, was expected to be at higher levels downwind of the marine emission source of its precursor DMS. DMS exists in the atmosphere for as long as 1 to 10 days [Berresheim et al., 1998; Chin et al., 1996; Putaud et al., 1999]. Its lifetime is sufficiently long for transport to these inland locations. Table 5. Concentrations of MSA (ng/m3) in PM2.5 in two field studies Conc. The 12-month HKEPD study Site Fall/2001 HT 13.8 TW 14.2 MK 11.1 season-mean 13.0 annual mean 20.3 min 11.1 max 33.4 The PRD study site Oct/2002 TM 13.3 TC 15.3 CW 14.1 SZ 16.3 GZ 22.2 CH 13.7 ZS 23.0 season-mean 16.8 annual mean 15.6 min 7.5 max 28.7

T (oC)

Conc.

T (oC)

Conc.

T (oC)

Conc.

T (oC)

26.1 26.9 27.7

winter/2000 27.4 23.2 33.4 28.0

18.8 19.3 20.1

Spring/2001 23.3 22.6 24.0 20.6 24.7 23.2 22.1

Summer/2001 27.8 17.9 29.3 18.6 30.0 17.8 18.1

25.0 26.7 26.4 25.7 25.2 -

Dec/2002 20.6 18.6 21.2 19.8 27.9 15.9 28.7 21.8

19.2 20.5 20.2 19.6 18 -

Mar/2003 9.1 13.5 14.8 17.6 16.9 10.7 16.9 14.2

Jun/2003 9.2 8.0 7.5 9.6 14.7 8.5 7.9 9.3

17

16.3 18.4 18.5 17.7 17.9 -

27.0 28.5 28.2 27.7 28.2 -

0.4

3

MSA (nmol/m )

0.3 2

R =0.44 0.2

0.1

TM

TC

CW

SZ

GZ

CH

ZS 0.0 0

20

40

60 3

Nitrate (nmol/m )

Figure 9. Correlation of MSA and nitrate

18

80

1. Aerosol organic nitrogen The aerosol TN was directly measured by the aerosol nitrogen analyzer. The difference between TN and the sum of the two inorganic nitrogen species (NO3- and NH4+) is taken to be the aerosol total organic nitrogen (ON) concentration. Table 6 lists the ON concentrations for the individual 24-hr samples. The ON concentration varied from non-detectable to 4 µgN/m3. The GD sites had higher mean ON concentrations than the three HK sites had. Among all the sites, ZS had the highest ON in the PM2.5 particles. No consistent spatial variation could be discerned among the three HK sites, despite their site characteristic varying from rural to urban. There appeared to be no discernable seasonable variation in the ON concentration, although the presence of ON in one-third of the samples collected in the summer month could not be verified due to the low concentrations. Figure 10 shows the time series of ON at each location. Table 6. Total ON Concentrations (µgN/m3) Sampling date 2002/10/02 2002/10/08 2002/10/14 2002/10/20 2002/10/26 2002/12/01 2002/12/07 2002/12/13 2002/12/19 2002/12/25 2003/03/01 2003/03/07 2003/03/13 2003/03/19 2003/03/25 2003/06/05 2003/06/11 2003/06/17 2003/06/23 2003/06/29

median mean SD max min

TM

TC

CW

SZ

GZ

CH

ZS

0.74 0.84 0.04 0.13 1.17 0.91 0.64 1.13 1.33 1.40 0.65 0.36 0.18 0.75 0.20 1.14 0.05 1.42 0.00 0.00 0.70 0.65 0.51 1.42 0.00

0.75 1.12 0.44 0.34 0.84 0.22 0.11 0.77 1.28 1.32 1.27 0.94 0.74 0.83 1.47 1.82 0.00 1.96 0.00 0.00 0.80 0.81 0.60 1.96 0.00

0.91 1.44 0.35 0.25 0.41 0.91 0.97 1.18 1.47 1.36 1.22 0.87 0.40 1.21 1.40 1.61 0.13 1.86 0.15 0.01 0.94 0.90 0.56 1.86 0.01

1.47 1.39 1.17 0.37 0.87 0.88 0.43 1.77 1.22 1.47 1.54 1.02 0.70 0.99 1.07 1.87 0.52 2.06 0.07 0.07 1.04 1.05 0.57 2.06 0.07

1.41 1.54 2.54 0.50 0.73 1.84 2.24 1.89 2.73 1.33 2.59 1.05 1.70 0.72 2.02 0.96 1.78 2.30 0.32 0.29 1.62 1.52 0.77 2.73 0.29

1.04 1.68 2.41 2.09 1.52 0.67 1.84 1.57 2.26 1.17 3.57 1.63 3.11 1.71 3.92 1.17 1.02 1.00 0.59 1.63 1.79 0.93 3.92 0.59

4.01 2.12 0.71 1.95 1.69 1.69 3.06 3.03 2.03 2.66 3.13 2.94 3.05 3.15 0.00 0.25 1.18 0.11 0.13 2.03 1.94 1.24 4.01 0.00

19

All sites

1.15 1.23 0.89 4.01 0.00

20 Jun-5

Jun-17 Jun-23 Jun-29

Jun-17

Jun-23

Jun-29

Mar-25

Mar-25

Jun-11

Mar-19

Mar-19

Jun-5

Mar-13

Mar-13

Jun-11

Mar-7

Dec-19

Dec-19

Mar-7

Dec-13

Dec-13

Mar-1

Dec-7

Dec-7

Dec-25

Dec-1

Dec-1

Mar-1

Oct-26

Oct-26

Dec-25

Oct-20

Oct-2

Oct-20

0

Oct-14

2

Oct-14

ZS Oct-8

4

Oct-8

Oct-2

3

ON (µgN/m )

3

ON (µgN/m )

Figure 10. Time series of ON concentrations Jun-29

Jun-23

Jun-17

Jun-11

Jun-5

Mar-25

Mar-19

Mar-13

Mar-7

Mar-1

Dec-25

Dec-19

Dec-13

Dec-7

Dec-1

Oct-26

Oct-20

Oct-14

Oct-8

Oct-2

3

ON (µgN/m )

Jun-29

Jun-23

Jun-17

Jun-11

Jun-5

Mar-25

Mar-19

Mar-13

Mar-7

Mar-1

Dec-25

Dec-19

Dec-13

Dec-7

Dec-1

Oct-26

Oct-20

Oct-14

Oct-8

Oct-2

3

ON (µgN/m )

Jun-29

Jun-23

Jun-17

Jun-11

Jun-5

Mar-25

Mar-19

Mar-13

Mar-7

Mar-1

Dec-25

Dec-19

Dec-13

Dec-7

Dec-1

Oct-26

Oct-20

Oct-14

Oct-8

Oct-2

3

ON (µgN/m )

Jun-29

Jun-23

Jun-17

Jun-11

Jun-5

Mar-25

Mar-19

Mar-13

Mar-7

Mar-1

Dec-25

Dec-19

Dec-13

Dec-7

Dec-1

Oct-26

Oct-20

Oct-14

Oct-8

Oct-2

3

ON (µgN/m )

Jun-29

Jun-23

Jun-17

Jun-11

Jun-5

Mar-25

Mar-19

Mar-13

Mar-7

Mar-1

Dec-25

Dec-19

Dec-13

Dec-7

Dec-1

Oct-26

Oct-20

Oct-14

Oct-8

Oct-2

3

ON (µgN/m ) 4

TM

2

0

4

TC

2

0

4

CW

2

0

4

SZ

2

0

4

GZ

2

0

4

CH

2

0

Table 7 lists the statistics of three ratios, ON/TN, ON/aerosol mass, and ON/OC. ON accounted for on average 30% of the aerosol TN. The ON fraction was highly variable among individual samples, ranging from negligible to 60%. ON was a small fraction (~3%) of the aerosol mass. The ON/OC mass ratio was on average 0.13, and the corresponding molar ratio was 0.15. That is, in an average organic compound molecule, there were 6.6 C atoms for every N atom. The actual ratio of N:C is expected to be higher in N-containing organic molecules, since many aerosol organics do not contain N atoms. The chemical nature of aerosol ON is little understood. A few classes of N-containing organic compounds have been reported to be present in aerosols. They include organic nitrates, nitro-PAHs, nitrophenols, urea, aliphatic amines, and amino acids [Mylonas et al., 1991; Bamford and Baker, 2003; Zhang and Anastasio, 2001; Zhang et al., 2002; Gorzelska and Galloway, 1990; Gorzelska et al., 1997; Cornell and Jickells, 1998]. The known organic N compounds have diverse sources. For example, organic nitrates and nitro-PAHs are mainly products of photoxidation of hydrocarbons and PAHs; urea and amino acids are most likely derived from biological sources. The correlation matrix of ON with other measured aerosol species reveals that among all the species, sulfate has the highest correlation with ON (r = 0.72) (Table 8). The correlation between sulfate and ON is also plotted in Figure 11. This moderate positive correlation appears to suggest that a significant fraction of ON is secondary in origin. K, Pb, and Si also show positive correlation with ON, with a correlation coefficient (r) greater than 0.6. Currently, we do not have a good explanation for such unexpected positive correlations. Table 7. Statistics of ON/TN, ON/aerosol mass, and ON/OC

ON/TN

ON/Mass

ON:OC

median mean max min median mean max min median mean max min

TM

TC

CW

SZ

GZ

CH

ZS

24% 22% 45% 0% 2.4% 2.0% 4.6% 0.0% 0.14 0.12 0.23 0.00

27% 23% 37% 0% 2.3% 2.2% 3.7% 0.0% 0.11 0.12 0.28 0.00

29% 27% 41% 3% 2.7% 2.4% 4.0% 0.1% 0.14 0.13 0.23 0.01

24% 27% 60% 10% 2.0% 2.3% 4.5% 0.5% 0.08 0.10 0.20 0.03

25% 26% 45% 15% 2.2% 2.2% 4.3% 0.7% 0.09 0.09 0.17 0.03

48% 45% 56% 0% 4.8% 4.6% 6.5% 0.0% 0.20 0.20 0.27 0.11

38% 37% 59% 0% 4.2% 3.8% 7.5% 0.0% 0.18 0.18 0.37 0.00

21

all sites 30% 30% 60% 0% 2.7% 2.8% 7.5% 0.0% 0.13 0.13 0.37 0.00

Table 8. Correlation coefficient matrix of ON with other species in aerosols 0 ON OC EC sulfate nitrate chloride NH4+ Al

ON 1

OC

EC

0.679

0.354

sulfate

nitrate

chloride

NH4+

0.725

0.393

0.273

0.585

Al

Si

Na

Mg

0.355

-0.094

-0.168

1

0.748

0.692

0.662

0.514

0.724

0.649

0.826

1

0.336

0.637

0.518

0.498

0.441

0.587

0.786

0.685

0.010

-0.247

0.803

0.754

0.645

0.321

0.764

0.824

0.578

0.544

0.176

-0.116

0.479

0.551

0.517

0.122

0.558

0.587

1

0.451

0.252

0.899

0.518

0.583

0.491

0.361

-0.070

-0.127

0.626

0.530

0.476

0.367

0.470

0.574

1

0.686

0.678

0.313

1

0.436

0.282

0.444

0.443

0.434

0.128

-0.033

0.452

0.418

0.445

0.146

0.605

0.580

0.435

0.585

0.652

0.099

-0.096

0.299

0.341

0.424

0.120

0.593

0.537

1

0.456

0.565

0.518

0.439

-0.012

-0.110

0.551

0.506

0.477

0.287

0.531

0.587

1

0.896

0.647

0.402

-0.032

-0.200

0.693

0.839

0.630

0.338

0.505

0.615

1

0.819

0.629

-0.067

-0.334

0.810

0.913

0.622

0.310

0.650

0.769

1

0.921

-0.064

-0.282

0.671

0.698

0.506

0.344

0.735

0.776

1

0.012

-0.204

0.491

0.482

0.381

0.156

0.675

0.663

1

0.410

-0.047

-0.019

0.043

-0.074

-0.016

0.024

1

-0.265

-0.296

-0.227

-0.110

-0.243

-0.260

1

0.785

0.495

0.334

0.712

0.886

1

0.543

0.272

0.578

0.686

1

0.478

0.528

0.589

1

0.212

0.336

1

0.867

0.551

Fe

0.634

Si Fe

Mn

0.520

Mn Na Mg K Ca V Ni As Pb

K

Ca

0.709

V

0.569

Ni

0.474

As

0.546

Pb

0.439

0.636

1

In bold, significant values (except diagonal) at the level of significance alpha=0.050 (two-tailed test).

22

30

y = 4.42x + 5.14 2

sulfate, µg/m

3

R = 0.53

20

10

0 0

1

2

3

4

5

ON, µgN/m

3

Figure 11. Correlation plot of sulfate versus ON 2. Amino acids and aliphatic amines A total of 20 amino acids and 3 aliphatic amines were targeted for analysis. Seven amino acids were detected in more than 50% of the samples. They are Asp, Ser, Gly, Thr, 2-Ala, Ala, and Phe. Among them, Gly, 2-Ala, and Ala were present in every samples. The three target aliphatic amines, methyl amine, ethyl amine, and ethanol amine, were detected in every sample. Their concentrations are listed in Table 9. Eight FAAs (Pro, 2-Aba, Cys, Tyr, Orn, Lys, Lle, and Leu) were detected in at least one sample but in less than 50% of the samples. Five FAAs (Gal, His, Arg, Val, and Met) were included in the calibration standards, but their presence in the samples could not be detected. The sum of all FAAs ranged from 18.9 to 62.5 ng/m3, with an average value of 32.5 ng/m3. The FAAs accounted for a small fraction (an average of 0.4%) of WSOC on a carbon mass basis. The most abundant FAA is glycine, ranging from 13.8 to 33.4 ng/m3. The average level of gly was roughly one order of magnitude larger than the next two most abundant AAs (Ala and 2-Ala). Gly alone accounted for three-quarter of the total FAA mass concentration. The sum of the three aliphatic amines ranged from 9.2 to 24.3 ng/m3, with an average value of 15.6 ng/m3. The contribution of the aliphatic amines to WSOC was also small, accounting for only 0.22% on a carbon mass basis. Methyl amine was the most abundant aliphatic amine. The other two aliphatic amines had similar abundance, about one-fifth that of methyl amine. No consistent seasonal variation pattern was seen across the sampling sites. The levels of the total FAAs and the total aliphatic amines were similar in different sampling months at each site. When measurements at all sampling sites pooled together, we note that positive correlation exist among some individual species. Table 10 tabulates the correlation (r) matrix for the FAAs and aliphatic amines. Glycine showed positive correlation with Ala, Phe, methyl amine, and ethyl amines. Methyl amine had a positive correlation with ethyl amine, but not ethanol amine. The correlations may indicate a dominant common source or common

23

sources for these species. Further inference to sources of the FAAs and aliphatic amines is not readily extractable other than that they are of biological origin. The contributions of FAAs and aliphatic amines to total aerosol ON were also computed and listed in Table 11. The two classes of amine compounds accounted for a very minor faction of ON, approximately 0.6% each. This is in line with the earlier suggestion that ON is mainly secondary in origin.

24

Table 9. Concentrations of free amino acids and aliphatic amines (ng/m3) in the aerosol samples Site TM

TC

CW

SZ

GZ

CH

ZS

a

mean SD min max

Sampling month Oct. Dec. Mar. Jun. Oct. Dec. Mar. Jun. Oct. Dec. Mar. Jun. Oct. Dec. Mar. Jun. Oct. Dec. Mar. Jun. Oct. Dec. Mar. Jun. Oct. Dec. Mar. Jun.

Asp

Ser

gly

Thr

2-Ala

Ala

Phe

0.0 0.0 0.0 2.6 0.0 1.4 0.0 2.5 1.6 0.0 1.7 4.5 1.3 0.0 1.4 2.5 1.6 0.0 0.0 1.1 1.6 0.0 0.0 2.0 2.0 2.1 0.0 1.6 1.13 1.17 0.00 4.53

1.2 1.3 0.3 0.0 1.2 1.1 0.0 0.3 1.1 1.2 0.4 6.6 1.5 1.6 0.0 0.2 1.4 1.1 0.1 0.0 1.4 1.6 0.0 0.3 1.9 2.1 1.1 0.0 1.03 1.27 0.00 6.57

17.1 19.9 20.8 13.8 22.7 17.9 27.4 26.6 19.9 22.4 23.7 33.4 19.9 24.5 21.5 23.2 28.0 26.7 32.3 28.8 25.9 14.7 30.8 24.2 21.1 27.3 26.2 23.8 23.7 4.9 13.8 33.4

0.0 0.5 0.5 0.4 0.0 0.6 0.5 0.3 0.0 0.0 0.4 2.2 0.0 0.0 0.7 0.4 0.0 0.0 0.7 0.5 0.0 0.0 0.6 0.6 0.0 1.0 1.0 0.4 0.40 0.47 0.00 2.16

1.4 1.2 1.4 1.9 1.6 1.1 1.1 2.2 1.5 1.5 1.3 2.0 1.9 1.5 1.3 1.8 1.6 1.4 1.3 1.9 2.1 1.1 1.6 1.8 2.2 1.4 1.5 2.1 1.60 0.34 1.11 2.24

0.8 0.7 1.3 1.6 1.2 1.1 1.9 1.7 1.2 1.7 2.6 4.6 1.5 1.6 2.5 2.0 3.1 3.3 4.5 3.5 1.8 0.9 2.6 2.4 1.5 2.1 4.1 1.7 2.13 1.07 0.74 4.57

0.0 0.3 0.4 0.4 1.1 0.9 0.7 0.6 0.8 1.5 0.9 1.5 1.2 0.8 1.3 1.1 2.1 2.1 2.0 2.1 1.1 0.6 0.7 1.4 0.8 1.0 1.2 0.5 1.04 0.56 0.00 2.11

other AAs a 0.3 0.0 0.0 0.8 0.0 0.0 0.0 0.4 0.3 0.0 0.2 7.7 1.2 0.8 0.8 0.8 3.4 6.7 3.1 4.1 0.0 0.0 3.3 0.6 1.9 2.1 2.7 0.0 1.48 2.04 0.00 7.74

ΣFAA 20.8 23.9 24.6 21.5 27.6 24.2 31.6 34.6 26.4 28.5 31.2 62.5 28.6 30.7 29.6 32.1 41.3 41.2 44.1 42.1 33.9 18.9 39.5 33.2 31.4 39.0 37.9 30.1 32.5 9.0 18.9 62.5

Ethanol amine 0.2 0.3 0.4 0.7 4.7 1.8 2.7 1.3 2.1 4.1 3.1 3.5 2.3 1.2 2.4 2.9 2.5 2.8 2.6 3.7 0.2 0.2 0.6 0.9 0.3 1.2 1.2 0.6 1.81 1.33 0.20 4.69

M. amine 8.6 11.7 11.6 7.8 9.2 10.5 10.1 10.7 9.7 11.7 10.7 10.4 9.5 9.9 14.2 10.1 18.1 15.6 18.4 16.4 9.8 8.2 12.5 13.4 14.0 13.7 13.8 9.3 11.76 2.86 7.83 18.36

E. amine 0.9 1.1 0.9 0.7 1.3 1.6 1.5 1.7 3.1 3.3 2.0 2.5 1.6 1.4 1.6 1.2 3.3 4.8 3.3 3.4 2.4 1.8 2.0 2.9 1.4 1.9 1.7 0.9 2.02 0.99 0.72 4.82

Σ(aliamine) 9.8 13.0 12.9 9.2 15.2 13.8 14.2 13.7 14.9 19.2 15.9 16.4 13.4 12.5 18.2 14.2 24.0 23.2 24.3 23.4 12.4 10.2 15.0 17.2 15.7 16.8 16.7 10.8 15.6 4.2 9.2 24.3

b

%WSOC FAA ali-amine 0.25% 0.14% 0.27% 0.17% 0.37% 0.23% 0.44% 0.22% 0.29% 0.19% 0.31% 0.21% 0.43% 0.23% 0.61% 0.29% 0.32% 0.22% 0.46% 0.37% 0.43% 0.26% 1.11% 0.33% 0.27% 0.14% 0.38% 0.18% 0.36% 0.25% 0.50% 0.25% 0.29% 0.19% 0.37% 0.23% 0.37% 0.23% 0.41% 0.26% 0.38% 0.17% 0.29% 0.19% 0.31% 0.14% 0.41% 0.25% 0.41% 0.23% 0.26% 0.13% 0.41% 0.21% 0.52% 0.22% 0.40% 0.22% 0.16% 0.06% 0.25% 0.13% 1.11% 0.37%

Other amines include eight minor FAAs (Pro, 2-aba, cys, Tyr, Orn, Lys, Lle, Leu) that have a detection rate less than 50%. b On a carbon mass basis. 25

Table 10. Correlation matrix of various FAAs and aliphatic amines other Ethanol M. ΣFAA AAs amine amine 0.434 0.458 0.634 0.400 Asp 1 0.135 0.199 0.077 0.253 0.072 -0.110 0.434 0.491 0.510 0.500 Ser 1 0.218 0.152 0.231 0.124 0.138 -0.110 0.493 0.787 0.590 0.647 0.904 0.577 gly 0.135 0.218 1 0.203 0.366 0.458 0.491 0.493 0.605 0.512 0.649 Thr 1 0.020 0.156 0.136 0.159 0.634 2-Ala 0.152 0.203 0.020 1 0.126 0.080 0.168 0.312 -0.062 -0.056 0.787 0.605 0.740 0.794 0.870 0.400 0.659 Ala 0.199 0.231 0.126 1 0.590 0.740 0.638 0.656 0.639 0.743 Phe 0.077 0.124 0.156 0.080 1 0.510 0.647 0.512 0.794 0.638 0.853 0.532 other AAs 0.253 0.168 1 0.313 0.400 0.500 0.904 0.649 0.870 0.656 0.853 0.392 0.536 ΣFAA 0.312 1 0.400 0.639 0.392 Ethanol amine 0.072 0.138 0.366 0.136 -0.062 0.313 1 0.221 0.577 0.659 0.743 0.532 0.536 M. amine -0.110 -0.110 0.159 -0.056 0.221 1 0.493 -0.011 -0.024 0.564 0.825 0.581 0.533 0.448 0.623 E. amine -0.020 0.120 0.631 0.715 0.911 0.604 0.620 0.578 0.906 Σ(ali-amine) -0.058 -0.003 0.150 -0.064 Note: Values in bold indicates statistically significant correlation (except diagonal) at the level of significance alpha=0.050 (two-tailed test). Asp

Ser

gly

Thr

2-Ala

Ala

Phe

26

E. amine -0.020 0.120 0.493 -0.011 -0.024 0.564 0.825 0.581 0.533 0.448 0.623 1 0.809

Σ(ali-amine) -0.058 -0.003 0.631 0.150 -0.064 0.715 0.911 0.604 0.620 0.578 0.906 0.809 1

Table 11. Contribution of FAAs and aliphatic amines to organic nitrogen Site TM

TC

CW

SZ

GZ

CH

ZS

mean SD min max

Sampling month Oct. Dec. Mar. Jun. Oct. Dec. Mar. Jun. Oct. Dec. Mar. Jun. Oct. Dec. Mar. Jun. Oct. Dec. Mar. Jun. Oct. Dec. Mar. Jun. Oct. Dec. Mar. Jun.

ΣFAA (ng N/m3)

Σ(ali-mine) (ng N/m3)

3.7 4.3 4.4 3.9 4.9 4.3 5.7 6.3 4.7 5.0 5.6 10.2 5.0 5.4 5.2 5.7 7.0 6.8 7.6 7.1 6.1 3.3 6.9 5.9 5.5 6.8 6.4 5.5 5.7 1.4 3.3 10.2

4.2 5.7 5.6 3.9 5.6 5.6 5.6 5.7 5.8 7.3 6.2 6.3 5.3 5.2 7.5 5.6 9.8 9.2 9.9 9.3 5.2 4.3 6.4 7.2 6.8 7.0 7.0 4.6 6.4 1.6 3.9 9.9

27

FAA 0.6% 0.4% 1.0% 0.7% 0.7% 0.6% 0.5% 0.8% 0.7% 0.4% 0.5% 1.4% 0.5% 0.4% 0.5% 0.6% 0.5% 0.3% 0.5% 0.6% 0.3% 0.2% 0.3% 0.4% 0.3% 0.3% 0.2% 1.6% 0.6% 0.3% 0.2% 1.6%

%ON ali-amines 0.7% 0.5% 1.3% 0.7% 0.8% 0.8% 0.5% 0.7% 0.9% 0.6% 0.6% 0.8% 0.5% 0.3% 0.7% 0.6% 0.7% 0.5% 0.6% 0.8% 0.3% 0.3% 0.2% 0.5% 0.3% 0.3% 0.2% 1.4% 0.6% 0.3% 0.2% 1.4%

total 1.4% 0.9% 2.4% 1.5% 1.5% 1.3% 1.1% 1.6% 1.6% 1.0% 1.2% 2.2% 1.0% 0.7% 1.2% 1.2% 1.3% 0.8% 1.1% 1.5% 0.6% 0.5% 0.5% 0.8% 0.6% 0.6% 0.5% 3.0% 1.2% 0.6% 0.5% 3.0%

3. References Allen, A. G., Dick, A. L. and Davison, B. M., Sources of atmospheric methanesulphonate, non-sea-salt sulphate, nitrate and related species over the temperate south Pacific, Atmos. Environ. 31, 1997, 191-205. Arimoto, R., Duce, R.A., Savoie, D.L., Prospero, J.M., Talbot, R., Cullen, J.D., Tomza, U., Lewis, N.F. and Ray, B.J., Relationships among aerosol constituents from Asia and the North Pacific during PEM-West A, J. Geophys. Res. 101, D1, 1996, 2011-2023. Bamford, H. A. and J. E. Baker, Nitro-polycyclic aromatic hydrocarbon concentrations and sources in urban and suburban atmospheres of the Mid-Atlantic region, Atmos. Environ. 37, 2003, 2077-2091. Bates, T. S., Calhoun, J. A. and Quinn, P. K., Variations in the methanesulfonate to sulfate molar ratio in submicrometer marine aerosol particles over the South Pacific Ocean, J. Geophys. Res. 97 (D9), 1992, 9859-9865. Bergin, M., Hagler, G., Salmon, L., Zheng, M., Chameides, W., Kiang, C. S., Schauer, J., Hong Kong and the Pearl River Delta Pilot Air Monitoring Project: Pilot study on the use of atmospheric measurements to manage air quality in Hong Kong and the Pearl River Delta, Project 2: Fine Particulate Matter (PM2.5) in the Pearl River Delta., Final Report, May 2004. Berresheim, H., Biogenic sulfur emissions from the Subantarctic and Antarctic Oceans, J. Geophys. Res. 92, 1987, 13245-13262. Berresheim, H., Huey, J.W., and Thorn, R.P., Measurements of dimethylsulfide, dimethyl sulfoxide, dimethyl sulfone, and aerosol ions at Palmer Station, Antarctica, J. Geophys. Res. 103(D1), 1998, 1629-1637. Chin, M., Jacob, D. J., Gardner, G. M., Foreman-fowler, M. S. and Spiro, P. A., A global three-dimensional model of tropospheric sulfate, J. Geophys. Res. 101 (D13), 1996, 1866718690. Cohen, S. A., Michaud, D. P., Synthesis of a fluorescent derivatizing reagent, 6aminoquinolyl-N-hydroxysuccinimidyl carbamate, and its application for the analysis of hydrolysate aminoacids via high performance liquid chromatography, Anal. Biochem., 211, 1993, 279-287. Cornell, S. E., Jickells, T. D., Thornton, C. A., Urea in Rainwater and Atmospheric Aerosol. Atmos. Environ. 32, 1998, 1903-1910. Gao, Y., Arimoto, R., Duce, R. A., Chen, L. Q., Zhou, M.Y. and Gu, D. Y., Atmospheric non-sea-salt sulfate, nitrate and methanesulfonate over the China Sea, J. Geophys. Res. 101, D7, 1996, 12601-12611. Gorzelska, K., Galloway, J. N., Amine Nitrogen in the Atmospheric Environment over the North Atlantic Ocean. Global Biogeochem. Cycles 4, 1990, 309-333. Gorzelska, K., Scudlark, J.R. and Keene, W.C., Dissolved organic nitrogen in the atmospheric environment. In Atmospheric Deposition of Contaminants to the Great Lakes and Coastal Waters; Baker, J. E., Ed.; 1997, SETAC Press: Pensacola, Florida, 379-392. Hynes, A. J., Wine, P. H. and Semmes, D.H., Kinetics and mechanisms of OH reactions with organic sulfides, J. Phys. Chem. 90, 1986, 4148-4156.

28

Jensen, N.R., Hjorth, J., Lohse, C., Skov, H. and Restelli, G., Products and mechanism of the reaction between NO3 and dimethylsulphide in air, Atmos. Environ. 25A, 1991, 1897-1904. Kim, B. M., Lester, J., Tisopulos, L., Zeldin, M. D., Nitrate artifacts during PM2.5 sampling in the South Coast Air Basin of California, J. Air & Waste Association, 49, 142-153 Sp. Iss. SI SEP 1999. Kiss, G. A., B. Varga, I. Galambos and I. Ganszky, Characterization of water-soluble organic matter isolated from atmospheric fine aerosol, J. Geophys. Res. 107 (D21), 2002, 8339, doi: 10.1029/2001JD000603. Li, Q. F. and J. Z. Yu, Determination of total aerosol nitrogen by thermal evolution, Aerosol Sci. Technol. 38, 2004, 382-390. Ma, Q. J., Hu, M., Zhu, T., Liu, L.L. and Dai, M.H., Seawater, atmospheric dimethylsulfide and aerosol ions in Pearl River Estuary and the adjacent northern South China Sea, manuscript submitted, 2004. Mylonas, D. T, D. T. Allen, S. H. Ehrman, S. E. Pratsinis, The sources and size distributions of organic nitrates in Los Angeles aerosol, Atmos. Environ. 25A, 1991, 2855-2861. Putaud, J.P., Davison, B.M., Watts, S.F., Mihalopoulos, N., Nguyen, B.C., and Hewitt, C.N., Dimethylsulfide and its oxidation products at two sites in Brittany, Atmos. Environ. 33, 1999, 647-659. Saltzman, E. S., D. L. Savoie, J. M. Prospero and R. G. Zika, Methanesulfonic acid and nonsea-salt sulfate in Pacific air: Regional and seasonal variations, J. Atmos. Chem. 4, 1986, 227240. Savoie, D.L., Prospero, J.M., Larsen, R.J. and Saltzman, E.S., Nitrogen and sulfur species in aerosols at Mawson, Antarctica, and their relationship to natural radionuclides, J. Atmos. Chem. 14, 1992, 181-204. Seinfeld, J. H. and S. M. Pandis, Atmospheric chemistry and Physics: from air pollution to Climate change, 1998, John Wiley, New York. Slaninas, S. Comments on this report., see Appendix 4. Yang, H., Q. F. Li and J. Z. Yu, Comparing of two methods for the determination of watersoluble organic carbon in atmospheric particles, Atmos. Environ. 37, 2003, 865-870. Yao, X., M. Fang and C. K. Chan, Size distributions and formation of dicarboxylic acids in atmospheric particles, Atmos. Environ. 36, 2002, 2099-2107. Yin, F., Grosjean, D. and Seinfeld, J. H., Photooxidation of dimethylsulfide and dimethyldisulfide, I, Mechanism development, J. Atmos. Chem. 11, 1990, 309-364. Zhang, Q. and Anastasio, C. Chemistry of Fog Waters in California's Central Valley - Part 3: Concentrations and Speciation of Organic and Inorganic Nitrogen. Atmos. Environ. 35, 2001, 5629-5643. Zhang, Q., Anastasio, C., Jimenez-Cruz, M., Water-soluble Organic Nitrogen in Atmospheric Fine Particles (PM2.5) from Northern California. J. Geophy. Res. 107, 2002, 10.1029/2001JD000870.

29

4. Acknowledgements I thank the PRD project management team and the sampling teams for the tremendous efforts put into the project to make the sampling successful. Eric Wan has helped a great deal in sampler maintenance and transport of samples. Jinhui Xu did all the laboratory work reported here. I also thank CCAR at HKUST for providing database on a wealth of meteorological parameters and air pollutants, and ATC at HKUST for letting us use its TOC instrument.

30

List of Appendices Appendix 1. SOP for extraction and analysis of water-soluble organic carbon in ambient aerosols. ...........................................................................................................................32 Appendix 2. Analysis of cationic and anionic species in aerosol water extracts.....................35 Appendix 3. SOP for analysis of amino acids and aliphatic amines in aerosol water extracts ..........................................................................................................................................39 Appendix 4 Responses to Review Comments .........................................................................43

31

Appendix 1. SOP for extraction and analysis of water-soluble organic carbon in ambient aerosols. NOTE: The volume of water used for extraction varies depending on the amount of filter materials to be extracted. In this project, 10 ml of water was used. Treatment of apparatus (e.g. forceps, Al foil) before doing the experiment: 1. Bake the apparatus for 12 hours at 550oC. 2. Rinse with Milli-Q water, methanol and hexane in sequence. (This step is skipped for Al foil) Treatment of glassware (e.g. beakers) before doing the experiment: 1. Wash the glassware with detergent. 2. Rinse with Milli-Q water. 3. Soak in hydrochloric acid for one day. 4. Rinse with Milli-Q water again. 5. Finally bake for 12 hours at 550oC. Extraction of filter samples: The seasonal composites consist of 5 filter pieces. The water extraction procedure is as follows: 1. Use a pair of tweezers to transfer the 5 pieces of filters from their storage cassettes to an 25-ml Conical flask. 2. Add 8 ml of UV oxidized pure water (Barnstead Ultrapure Water System, Dubuque, IA, USA) to the flask. (Water used for extraction in the following steps has to be the purified water without any exception.) 3. Place the flask in an ultrasonic bath for 20 min. Ice water is placed in the ultrasonic bath to keep the water temperature below 10oC. 4. After sonication, pass the supernatant solution through a 0.2 mm pore size Teflon membrane syringe filter. (Keep the syringe filter for the next water extract of the same group of filters). 5. Discharge the filtrate into a 10.0 ml volumetric flask. 6. Add 3 ml of pure water into the filter residue and repeat steps 2-5. The filtrate is placed in the same 10.0 ml volumetric flask. 7. Fill the volumetric flask with water to the 10-ml mark. (Approximately 1 ml of water was retained by the 5 pieces of filters.) 8. Discard the filter residue in the conical flask. 9. Transfer the 10-ml water extract to a brown glass bottle with Teflon-lined cap. 10. Store the water extract in the freezer until analysis. Analysis of WSOC using a TOC analyzer A TOC analyzer (Model TOC-5000A, Shimadzu, Kyoto, Japan) (located in ATC) is used for the analysis. In the TOC analyzer, an autosampler (ASI-5000A) introduces aqueous samples into a combustion tube filled with platinum catalyst on alumina spherical support. Carbonaceous materials are either combusted or decompose to become CO2. A high purity air carrier gas transports the combustion product from the combustion tube to a sample cell in a NDIR gas analyzer for the detection of CO2. The TOC analyzer allows a maximum injection volume of 100 µl to be introduced into the instrument when operated with a normal catalyst in place. A maximum injection volume of 2000 µl is possible when a high sensitivity catalyst

32

is used. The normal catalyst option was used in this study. An injection volume of 35 µl is typically used for our analysis. 1. Prepare three levels of potassium hydrogen phthalate solutions in the range of 0-20 µgC/ml to be used for instrument calibration. 2. Load no less than 3 mL of water extract into each autosampler tube. 3. Instrument condition: sparged by air for 2 min; 35 µl injection volume; a maximum of three injections for each sample. [For every sample, multiple consecutive injections are made until the peak areas of at least two injections achieved a standard deviation of less than a user-defined value or the specified maximum allowable number of injections, which we recommended to be 3, is reached.]

Sample pretreatment and analysis of WSOC using the Sunset Labs aerosol carbon analyzer (ACA): [This procedure was not used in the current study.] 1. Use one 5-ml and one 3-ml volumetric pipettes to withdraw 8ml water extract from its storage bottle and transfer to a 25-ml round bottom flask. 2. Connect the 25-ml flask with the rotary-evaporate. Evaporate the water extract to dryness. This step takes 20-30 min. 3. Add 500 µl water back to the 25-ml flask. Make sure to rotate the water so that the flask wall is thoroughly rinsed. 4. Use a syringe to withdraw 50 µl of the concentrated water extract and spike to a prebaked 1.45 cm2 quartz filter punch. 5. Let the spiked filter air-dry. This takes ~30 min. 6. Place the dried filter punch on the quartz boat for the carbon analyzer and start the C analysis. 7. The conditions for the carbon analysis are as follows: • Purge the combustion chamber with 1% O2:99% He with the blower off for 20 sec. • Raise the front oven temperature to 600oC and keep at this temperature until 600 sec. • Raise the front oven temperature to 870oC and keep at this temperature until 120 sec has passed since the temperature ramp. Method detection limits and precision The limit of detection (LOD) of the methods is calculated using the following equation: C L = kS b1 / S Where CL is the limit of detection; k is a constant related to the confidence level and is usually set to be 3 (Miller and Miller, 1993); Sb1 is the standard deviation of 5 measurements of blank water extracts; and S is the slope of the calibration curve. The ACA was determined to have a LOD of 0.48 µg carbon using this definition (Sb1= 0.16, S=0.999) whereas the TOC had a LOD of 0.88 µgC/ml for an injection volume of 35 µl (Sb1= 662, S=2242). These values translate to 4.8 µgC per sample in the ACA method (0.48 µgC in 50 µl of spiked water extract out of a 500 µl water concentrate) and 2.6 µgC per sample in the TOC method (0.88 µgC/ml in 35 µl of injected water extract out of a 3 ml solution). The ACA analyzer has a linear dynamic range up to approximately 900 times its LOD (Birch and Cary, 1996). The dynamic range of the TOC is up to 25,000 µg/ml, more than 1000 times its LOD. As a result, the lower detection limits of the analyzers are more of a

33

concern to the determination of aerosol WSOC concentrations rather than the upper limits of the linear range. For evaluation of the measurement precision of the two carbon analyzers, the relative standard deviation (RSD) is computed using the multiple consecutive measurement results for the 6 aerosol samples. When the WSOC loading levels are 20-40 times the LOD of the TOC method and 40-75 times the LOD of the ACA method, the RSD values are better than 3% in both methods.

References: Birch, M.E., Cary, R.A., 1996. Elemental carbon-based method for monitoring occupational exposures to particulate diesel exhaust. Aerosol Science and Technology 25, 221-241 Miller, J.C., Miller, J.N., 1993. Statistics for Analytical Chemistry, 3rd ed. Ellis Horwood and Prentice Hall, New York, pp115-117. Yang, H., Q. Li., J. Z. Yu, Comparison of two methods for the determination of water-soluble organic carbon in atmospheric particles, Atmos. Environ., 37, 2003, 865-870.

34

Appendix 2. Analysis of cationic and anionic species in aerosol water extracts Analysis of anionic species Instrument: A Dionex DX500 Ion Chromatograph with a conductivity detector (CD20) and a gradient pump GP40 operating in the anion analysis mode. Instrument analysis conditions: An IonPac AG11 guard column (4x50 mm) and an IonPac AS11 analytical column (4x250mm) are used to separate the analytes. An IonPac ATC-1 trap column (4mm, 9x24mm), and a Dionex ASRS-ultra 4mm anion self regenerating suppressor ensure the lowest possible background noise levels and detection limits. A sampling loop of 50ul is chosen for all the samples. The eluent is 0.2-5 mM NaOH (gradient) and the flow rate is 2 ml/min. The eluent gradient is as below: Time 0 min. 6 min. 16 min. 30 min. Conc. 0.2mM----0.2mM----5mM------5mM Anion analysis: Water extracts of aerosol samples are loaded into autosampler vials. The anions are separated by the anion exchange column. After separation, the anions pass through an electronic suppressor to eliminate background conductivity and are then detected with a conductivity cell. The resultant peaks are identified by their characteristic retention times and quantified by reference to external standards. Target analytes: Inorganic anions: Sulfate, nitrate, chloride, and nitrite. Organic anions: formate, acetate, lactate, methanesulfonate, oxalate, malonate, succinate, and glutarate. A typical IC chromatogram for anionic species in an aerosol water extract sample is given in Figure A-1. Method detection limits (MDLs): The analytical procedure MDL is defined as the amount of an analyte that generates a response equal to the sum of the mean blank signal (YB) plus three times the standard deviation (SDB) of the blank signal. YB is approximated with the intercept of a calibration line obtained from the analysis of a series of analytical standards in the vicinity of the blank response; and SDB is approximated with the standard error for the Y estimate, Y being the response (i.e., IC peak area in this application) (Miller and Miller, 1993; NIOSH, 1993). The MDLs for the target anions are listed in Table A3-1. Method precision: Replicate analyses indicate that the relative standard deviation is less than 10% for samples at a concentration level 10 times or greater than the MDL, and less than 20% for samples at a concentration level between 5 and 10 times the MDL. Traceability: Calibration standards of the inorganic anions are prepared from commercially available standards (Alltech), concentrations of which are traceable to NIST standard reference materials. Organic acid standards that are traceable to NIST are not commercially available. They are prepared from 90-95% pure chemicals.

35

Figure A-1. Example Anion Chromatogram of a Typical Aerosol Water Extract Sample.

Table A-1. Analytical Procedure Method Detection Limits of Anions Anion

Analytical procedure MDL µM

inorganic chloride

0.67

nitrite

0.12

nitrate

<2.5

sulfate

<11.7

lactate

0.47

acetate

0.74

formate

0.60

methanesulfonate

0.07

pyruvate

0.07

oxalate

0.98

malonate

0.56

maleate

0.49

succinate

0.28

glutarate

0.14

Organic

36

Analysis of cationic species Instrument: The cations are analyzed using a Dionex DX500 ion chromatograph with electrochemical detector ED 40 and gradient pump GP40. An IonPac CG12 guard column 4x50mm and an IonPac CS12 analytical column 4x250mm are used. Dionex CSRS-ultra 4mm is used. The sampling loop is 50ul. The eluent is 20mM methanesulfonic acid. Isocratic elution at a flow of 1 ml/min is used and each analysis lasts 12 min. Target analytes: NH4+, K+, Na+, Ca2+, and Mg2+. A typical IC chromatogram for cationic species in an aerosol water extract sample is given in Figure A-2. Sample analysis: Water extracts of aerosol samples are loaded into autosampler vials. Injection loop is 50 µL. The cations are separated by a cation exchange column. After separation, the cations pass through an electronic suppressor to eliminate background conductivity and are then detected with a conductivity cell. The resultant peaks are identified by their characteristic retention times and quantified by reference to external standards. Method detection limits (MDLs): The analytical procedure MDL is defined as the amount of an analyte that generates a response equal to the sum of the mean blank signal (YB) plus three times the standard deviation (SDB) of the blank signal. YB is approximated with the intercept of a calibration line obtained from the analysis of a series of analytical standards in the vicinity of the blank response; and SDB is approximated with the standard error for the Y estimate, Y being the response (i.e., IC peak area in this application) (Miller and Miller, 1993; NIOSH, 1993). The MDLs for the target cations are listed in Table A-2. Method precision: Replicate analyses indicate that the relative standard deviation is less than 10% for samples at a concentration level 10 times or greater than the MDL, and less than 20% for samples at a concentration level between 5 and 10 times the MDL. Traceability: Calibration standards of the inorganic cations are prepared from commercially available standards (Alltech), concentrations of which are traceable to NIST standard reference materials. Reference: Miller, J. C.; Miller, J. N. Statistics for Analytical Chemistry, Ellis Horwood: Chichester, 3rd edn; 1993, pp. 46-51 and 115-117. NIOSH (National Institute of Occupational Safety and Health), Method Evaluation Guidelines, 1993.

37

Figure A-2. Example Cation Chromatogram of a Typical Aerosol Water Extract Sample. Table A-2. Analytical Procedure Method Detection Limits of Cations Analytical procedure MDL uM +

1.3

Na

NH4

+

1.1

+

1.3

2+

1.3

2+

1.2

K

Ca

Mg

38

Appendix 3. SOP for analysis of amino acids and aliphatic amines in aerosol water extracts Treatment of apparatus (e.g. forceps) before doing the experiment: 3. Bake the apparatus for 12 hours at 550oC. 4. Rinse with Milli-Q water, methanol and hexane in sequence. Treatment of glassware (e.g. beakers) before doing the experiment: 6. Wash the glassware with detergent. 7. Rinse with Milli-Q water. 8. Soak in hydrochloric acid for one day. 9. Rinse with Milli-Q water again. 10. Finally bake for 12 hours at 550oC. Reconstituting AccQ·Fluor Reagent: 1. A heating block is preheated to 55oC. 2. Tape Vial 2A lightly before opening to ensure all AccQ·Fluor Reagent Powder is at the bottom of the vial. 3. Rinse a clean micropipette by drawing and discarding 1ml of AccQ·Fluor Reagent Diluent from vial 2B. 4. Transfer 1.0ml of AccQ·Fluor Reagent Diluent 2B to the AccQ·Fluor Reagent Powder in vial 2A. 5. Cap the vial tightly. 6. Vortex for 10 seconds. 7. Heat Vial 2A on top of the heating block at 55oC until the AccQ·Fluor Reagent Powder is dissolved. (Do not heat the reagent for longer than 10 minutes.) 8. Store the reconstituted AccQ·Fluor Reagent in desiccators at room temperature up to one week. Preparation for the internal standard solution (DL-Norleucine): Note: The procedure is the same if a different –NH2 containing compounds is selected as the internal standard. 1. Weigh 32.795mg DL-Norleucine into a 25ml volumetric flask and dissolve it using 20mM HCl. This is 10 µM DL-norleucine stock. 2. Pipette 1ml 10 µM DL-Norleucine to a 10ml volumetric flask and fill with 20mM HCl to the mark. This is 1 µM (1000 pmol/µl) DL-norleucine stock solution. 3. Pipette 100 µl of the 1000 pmol/µl stock solution to a 5ml volumetric flask and fill with 20 mM HCl to the mark. This is 20 pmol/µl internal standard working solution.

Internal Standard (DL-norleucine) Required concentration 10 µM 1000 pmol/µl 20 pmol/µl

Calculation Mass / Volume required 0.01M*0.025L*131.18g/mol 32.795mg 0.001M*10ml/0.01M 1ml 20pmol/μl*5ml/1000pmol/µl 100µl

Preparation of 2.5 mM Amino Acid standard mixture (AA Std mix-1 and AA std mix-2): 39

1.

Dissolve the following individual AA or aliphatic amine standards in a 25 ml volumetric flask to make 10 mM stock solution. 53.91mg D(+)-Galactosamine hydrochloride, 53.91mg D(+)-Glucosamine hydrochloride, 22.27mg Beta-Alanine, 24.39mg Ethanolamine hydrochloride, 16.88mg Methylamine hydrochloride, 25.78mg D(-)-2-Aminobutyric acid, 20.39mg Ethylamine hydrochloride 41.41mg L(+)-Ornithine hydrochloride

2. 3.

Pipette 1.25ml of the above 10mM standard mixture to a 5ml volumetric flask and fill with water to the mark. This is the 2.5 mM Amino Acid standard mixture (AA std mix-1). The commercially available AA standard mixture from Waters is AA std mix-2. Each AA standard in this mixture is supplied at a concentration of 2.5 mM except for cystein, which is supplied at 1.25 mM. The following 17 amino acids are included in this standard mixture: L-alanine L-leucine Ammonia L-lysine L-methionine L-arginine L-phenylalanine L-aspartic acid L-proline L-cystine L-serine L-glutamic acid L-threonine Glycine L-tyrosine L-histidine L-valine L-isoleucine

Preparation of the 100 pmol/ul standard mixture: 1. Add 40µl AA std mix-1 into a clean autosampler vial. (The remaining 2.5mM AA standard mixture-1 can be stored at –20oC for up to 3 months.) 2. Add 40µl AA std mix-2 (2.5mM Waters Amino Acid Hydrolysate Standard into the same autosampler vial. 3. Add 920µl Milli-Q water to the vial. This is 100 pmol/µl stock calibration standard mixture. It can be stored at –20 oC for up to one month. Preparation of 0.5, 2, 5, 20, 40 pmol/µl standards 1. 5, 20, 50, 200, and 400 µl of 100 pmol/ul stock calibration standard are added into five separate clean vials. 2. Add 995, 980, 950, 800, and 600 µl water, respectively, to make the final volume 1000ul. Standard dilution Concentration of diluted standards (pmol/µl)

Calculation

Volume of Volume of 100 pmol/µl water required standard required (µl) (µl)

0.5 2 5 20 40

1000µl*0.5pmol/µl / 100pmol/µl 1000µl*2pmol/µl / 100pmol/µl 1000µl*5pmol/µl / 100pmol/µl 1000µl*20pmol/µl / 100pmol/µl 1000µl*40pmol/µl / 100pmol/µl

Procedures for derivatization of the calibration standards:

40

5 20 50 200 400

995 980 950 800 600

1.

Pipette 10µl of calibration standard and 10µl of internal standard (20 pmol/µl) in a 6 x 50 mm sample tube. 2. Add 160µl AccQ-Tag borate buffer. 3. Vortex for 10 seconds. 4. Add 20µl of AQC reagent solution. 5. Vortex immediately for 10 seconds. 6. Wait for 1 minute. 7. Transfer the content of the sample tube to a low volume insert (LVI) and cap tightly with a silicon-lined septum. 8. Remove the bubbles in the sample tube. 9. Heat the sample vial in a heating block for 10 minutes at 55oC. 10. Inject 20µl of the solution into the HPLC for analysis. Note: The derivatives can be stored at room temperature for up to one week. Diluted standard concentration (pmol/µl)

0.5

2

5

20

40

Final standard concentration (pmol/µl)

0.025

0.1

0.25

1

2

Analysis of water soluble free amino acid in water extracts: 6. Prepare two sample tubes for each water extract sample. 7. Transfer 0.5ml water extract to each sample tube. 8. Evaporate the water extract to dryness under a gentle stream of argon gas or nitrogen gas (ultra high pure) at 40oC. This process takes about 2.5 – 3 hours. 9. Add 10µl internal standard solution (20pmol/µl) and 170µl borate derivatization buffer into the sample tube. 10. Vortex briefly the sample tube. 11. Gently purge the reconstituted sample solutions with ultra high purity argon gas or nitrogen gas for 50 minutes. The flow rate of gas must be controlled so that the volume loss of the sample solution won’t exceed 30%. The purpose is to lower the NH3 levels as a big NH3 peak would interfere the quantitation of the adjacent amino acid compounds. 12. Add 20 µl AccQ-fluorescence reagent solution into the sample tube and vortex immediately for 10 seconds. 13. After 1 minute of reaction, transfer the sample into a low volume insert (LVI) placed in an autosampler vial and cap tightly with a silicone-lined septum. The bubbles in the sample vial should be removed before heating. 14. Heat the sample vial to 55oC for 10 minutes prior to the HPLC analysis. Analysis of hydrolysable combined amino acid: 1. Prepare two all-Teflon hydrolysis containers for each sample. 2. Add 1ml of water extract and 1ml constant boiling HCl (6N) to each Teflon hydrolysis vial. 3. Vortex for 10 – 20 seconds. 4. Seal the surface of the solution with Ultra high purity nitrogen gas for three times each for 2 minutes. 5. Screw the cap for the hydrolysis vial and place the vial in an oven at 110o for 24 hrs.

41

6. 7. 8. 9.

Cool the sample to room temperature before opening the cap. Transfer 0.5 ml post-hydrolysis sample solution to a sample tube. Evaporate to dryness under a gentle stream of argon gas or nitrogen gas (ultra high pure) at 40oC. Follow instruction for step 4 and onwards in the free AA analysis procedure.

Calculation: Calibration curve: PR = m* CAA +b Where PR is the peak area ratio, CAA is the amino acid concentration in unit of pmol/µL. Amount of FAA & CAA in the water extract: CAA = (PR – b) / m Amount of FAA (nmol) in x ml water extract = CAA*200*(x/0.5)/1000 Amount of CAA (nmol) in x ml water extract = CAA*200*(2/0.5)*(5/1)/1000 Amount of FAA and CAA expressed as air concentration A filter 1 C (nmol / m 3 ) = C (nmol ) * * Awater V (m 3 ) Where V is the sampled air volume, Afilter is the entire filter area, and Awater is the filter area used in water extraction.

42

Appendix 4 Responses to Review Comments

Below is a point-by-point response to three reviewers’ comments on the report. For the convenience of discussion, review comments are provided and marked in a smaller font. Review I Comments: General: As far as I can judge (never having seen the original project description), the report answers in a very straightforward manner a number of policy relevant questions such as: what is the contribution of water soluble organic compounds, is it primary or secondary and, as a consequence of origin, can it be controlled. Data quality cannot always be judged, but as far as inorganic ions are concerned, seems to be all right indeed. It would have been nice if intercomparison data would have been available too for the organic species, as some of the analytical methodologies involve concentration steps, which could potentially introduce artifacts. The interpretation of the data is, with one exception (see specific remarks), straightforward and based on solid science. The report is written in a clear style. The conclusions are justified and relevant, again with one specific remark, see below. Specific remarks: 1.

Regarding section 2. Extraction and analysis scheme.

The methods are in general quite robust, though concentration by evaporation is always a bit dangerous. There is one notable exception though. The filter methods can easily miss a lot of the nitrate in PRD; see results obtained in Guangzhou by Zhang. 40 NO3

SJAC

-

Filter

µg/m3

30

20

10

0 6-25

6-26

6-27

6-28

6-29

Intercomparison between filter and SJAC results in Guangzhou, measured by Zhang. The error is dependent on many factors, but generally higher in summertime, as the volatilization of ammonium nitrate is strongly temperature dependent. So the nitrate data are really a lower estimate and this has other consequences, see below.

Response: The nitrate measurement artifact is a sampling issue, not an analytical problem. Nitrate loss occurs on the filter before water extraction takes place. In addition, I wish to clarify that concentration by rotary evaporation was not used in the current study. [We used rotary evaporation in our previous studies due to larger numbers of filter pieces that had to be extracted for each composite sample. Appendix 1 has been revised to clearly indicate that the procedure applied to the samples in this study did not involve concentration of water extracts.] I also revised the report text to indicate the sampling artifact issue for nitrate.

43

Comment: 2. The same problem is also relevant for semi-volatile organic compounds, e.g. oxalate. It would depend on the chemical form, e.g. the acid is semi-volatile, the sodium salt is not. So here too filter sampling methods could show a bias.

Response: Yes, I agree that there was a sampling artifact issue with semivolatile organic compounds (e.g., oxalic acid). The sampling set-up in this study was not designed to how to best characterize the aerosol phase concentrations of semi-volatile species. The report text on dicarboxylic acids is revised to note this sampling issue with semivolatile organic compounds. Comment: 3. Interpretation of MSA data, in 5.3. Methanesulfonic acid. Here some problems are encountered, in my view. The conclusion that OH and NO3 radical concentrations are highest in winter, based in nitrate data, seems questionable, especially if possible volatilization is kept in mind. The losses of nitrate will be much higher in summer, compared to winter. It is indeed true that competition is going on between different chemical pathways and that temperature is an important factor, as is pointed out in the report.

Response: The current nitrate measurements do not allow us to quantitatively assess the magnitude of nitrate loss during sampling. I agree that the nitrate data alone, due to the sampling artifact problem, could not lead to a definitive conclusion of higher oxidant levels in the winter. To answer this question, we need carry out addition experiments that are designed to minimize nitrate sampling artifacts. The text has been revised to indicate the uncertainty in oxidants levels as inferred from nitrate measurements. Comment: But there is perhaps another factor which contributes. The DMS emissions in the polluted

Fig. DMS, Chlorophyl a and salinity vs. latitude of (a) July 2000, (b) May 2001 (c) November 2002

44

Estuary region is much higher than in the open see, see data from Ma and Hu, Beida, in press. The result is that higher DMS concentrations are seen near the coast, the conversion to HMSA takes time and the end result is higher concentrations of HMSA in inland stations.

Response: In my original report, I have made an interpretation of the spatial variation pattern that is the same as expressed by the reviewer here. Below I quote what appears in my report. “The higher concentrations seen at the inland sites could be explained by the secondary nature of MSA. MSA, as a secondary pollutant, was expected to be at higher levels downwind of the marine emission source of its precursor DMS. DMS exists in the atmosphere for as long as 1 to 10 days [Berresheim et al., 1998; Chin et al., 1996; Putaud et al., 1999]. Its lifetime is sufficiently long for transport to these inland locations.”

Comment: Conclusions: This report is a valuable contribution to the PRD Pilot Project. The work is of good quality, the results are relevant and well presented. Discussion on the interpretation of some results does not really detract from the overall value of the report.

Response: None.

45

Review II Comment: This report provides comprehensive measurements of water-soluble species in the atmospheric aerosols in the PRD area. The measurements are very useful for an assessment of the environmental impact of aerosols. The author might want to revise the report by considering the following comments and suggestions: „

A complete description of the sampling method used in the work is needed.

Response: A detailed description of the sampling method is provided in the report by Bergin et al. (2004). This is now noted in the revised report. Comment: „

The extraction of aerosol species was made after the sampling seasons. Thus, the samples might be stored for several weeks before extraction. Given that all the target compounds (WSOC, organic acids, amines) analyzed in this work are highly biodegradable, the uncertainties and bias due to the storage processes should be carefully evaluated. This reviewer did not find any description on this point in the report.

Response: The filter samples were stored at 4oC before water extraction and subsequent analysis. Biodegradation of the WSOC species was likely limited under this low-temperature storage. We will conduct work in the near future to assess biodegradation of target WSOC species at the storage temperature. Comment: „

Sulfate was used as the indicator of secondary sources and nitrate was used as the proxy of oxidants in this work. There must be concurrent measurements of ozone, why not use it?

Response: Concurrent measurements of ozone are not available for the sites in Guangdong (SZ, GZ, ZS, and CH) since the focus of project 2 is on PM and limited resources do not permit measurements of all relevant species. Comment: „

Concentrations of MSA were plotted vs. nitrate concentration in Figure 9 and a linear regression line for MSA and nitrate with R2 of 0.44 was obtained. However, the line in Figure 9 seems dominated by the data points for GZ and SZ. The correlation between MSA and nitrate at the other sites was not as significant as the line showed. Examining the MSA-nitrate correlation site by site might be useful.

Response: Although there were 20 individual samples collected at each site during the project period, individual filters in the same sampling months were combined to generate one single seasonal composite to ensure sufficient material for chemical characterization. As a result, each site had only four seasonal composite samples. This limited number of samples does not permit to examine the MSA-nitrate correlation on a site by site basis. If the data points for GZ and SZ are removed, a positive correlation with r2 = 0.40 still exists.

46

Review III Comment: We have some minor comments for your consideration: 1.

Page vi. “Effective reduction in this portion of aerosol material requires identification of their precursors and their formation mechanism and rates in the atmosphere”. Since this part of the information is put forward as policy-relevant finding, relevant recommendations regarding sources identification and control options should be elaborated.

Response: Our current knowledge of the sources and formation mechanism is very limited, as reflected by the fact that only a small fraction of WSOC mass (less than 10% in this study) could be identified as known chemical identities. We have a long way to go before we could make recommendation regarding sources and control options. A few sentences are added into the WSOC section to note this. Comment: 2.

Page 8. “The policy relevant implication is that an important fraction of aerosol mass (11-24%) could not be reduced simply by cutting primary emissions.” This is an important finding from Jian’s study, in general, findings from this study should be considered together along with studies conducted under Project 2 should that a more holistic view can be formed.

Response: Integration of findings from WSOC chemical characterization work could be done once the scientific team decides that this should be the way to represent results from project 2.

47

Loading...

Final Report - Environmental Protection Department

Final Report Hong Kong and the Pearl River Delta Pilot Air Monitoring Project: Pilot study on the use of atmospheric measurements to manage air quali...

5MB Sizes 2 Downloads 0 Views

Recommend Documents

Import - Environmental Protection Department
control on import and export of waste in the Hong Kong Special Administrative Region. (HKSAR). Under the WDO, a permit .

Method Statement - Environmental Protection Department
Method Statement detailing the. Protective Measures on Declared Monument. Buildings at. Former Marine Police Headquarter

3. project description - Environmental Protection Department
The floor plans that show the preliminary design of the proposed Project are set out in Figure 3.1. The proposed Project

connecticut department of environmental protection - CT.gov
landmark publication, “Seepage and Pollutant Renovation Analysis for Land Treatment, Sewage. Disposal ... renovation o

Massachusetts Department of Environmental Protection | Mass.gov
MassDEP is responsible for ensuring clean air and water, safe management and recycling of solid and hazardous wastes, ti

Proposal for Site Investigation - Environmental Protection Department
3 Proposal for Site Investigation. 3.1 Sampling Strategy. As presented in Section 2, the potential risk for soil and gro

Subsurface Soil Investigation Report - Department of Environmental
Subsurface Soil Investigation Report for: '. Log # 08-93-0077. Colvin Oil Company, Port Orford. UST Facility # 6718. 113

Method Statement For Site Work - Environmental Protection Department
A “Method Statement for Gas Turbine Units Site Work” issued by CLP Power for OCGT units is given below. A1.1 ... sha

air protection branch - Environmental Protection Division
Apr 4, 2017 - Main Numbers. Land Branch - 404-362-2537. Solid Waste - 404-362-2692/2696. Lead & Asb - 404-362-7026. Scra

ENVIRONMENTAL REPORT
“SRI group is currently promoting global environment management as one of the most significant issue. Sumirubber Malay