Atmosphere

Publications

Air toxics and indoor air quality in Australia

State of knowledge report
Environment Australia, 2001
ISBN 0 6425 4739 4

Part A: Air toxics (continued)

Air toxics: Australian studies (continued)

4.2 Review of studies conducted in Australia

4.2.1 Motor vehicle emissions

The Environment Protection Authority, New South Wales, conducted short-term roadside monitoring as part of the larger pilot air toxics project (see Section 4.2.1), and in conjunction with a broader study being undertaken by the NSW RTA (EPA NSW 1998). Ambient volatile organic canister samplers were located at two sites adjacent to busy roads, one a freeway (Ingleburn) with high-speed traffic (110 km/hour), the other on an arterial road (Rydalmere). Samplers were located 20 m and 40 m from the roads, and two consecutive 48-hour samples were collected at each site.

Table 4.19 presents the results of the roadside monitoring. Similar pollutant concentrations were measured at both sites, as at other urban sites, and there was a trend towards lower concentrations of pollutants associated with motor vehicle use (particularly benzene, toluene and xylene, or BTX) at weekends as compared with weekday samples.

Table 4.19: Roadside sampling undertaken by EPA NSW in conjunction with NSW RTA
Site Rydalmere Ingleburn
Start Date 17/10/95 20/10/95 Mean 14/09/95 16/09/95 Mean
Chemical (ppb)
Freon-12 0.6 0.9 0.8 0.6 0.6 0.6
Chloromethane 0.7 0.6 0.7 0.7 0.9 0.8
Freon-114            
Vinyl chloride            
Bromoethane            
Chloroethane            
Freon-11 0.2 0.3 0.3 0.3 0.3 0.3
1,1-Dichloroethylene            
Dichloromethane 0.4 0.2 0.3 0.2 0.2 0.2
Freon-113            
1,1-Dichloroethane            
cis-1,2-Dichloroethylene            
Chloroform            
1,2-Dichloroethane            
1,1,1-Trichloroethane            
Benzene 1.3 0.8 1.1 1.4 0.9 1.2
Carbon tetrachloride   0.2 0.2      
1,2-Dichloropropane            
Trichloroethylene            
cis-1,3-Dichloropropene            
trans-1,3-Dichloropropene            
1,1,2-Trichloroethane            
Toluene 3.7 1.6 2.7 2.6 1.3 2
1,2-Dibromoethane            
Tetrachloroethylene       0.8   0.5
Chlorobenzene            
Ethylebenzene 0.7 0.2 0.5 0.3 0.2 0.3
m- and p-Xylene 1.6 0.4 1 0.6 0.3 0.5
Styrene 0.4   0.3 0.2   0.2
1,1,2,2,-Tetrachloroethane            
o-Xylene 1.1 0.3 0.7 0.5 0.3 0.4
1,3,5-Trimethylbenzene            
1,2,4-Trimethylbenzene 0.4 0.2 0.3 0.5 0.2 0.4
1,3-Dichlorobenzene            
1,2-Dichlorobenzene            
1,4-Dichlorobenzene            
1,2,4-Trichlorobenzene            
Hexachlorobutadiene            
3-Chloropropene            
4-Ethyltoluene            
1,3-Butadiene 0.3   0.2 1.3 0.6 1

Note: Blank cell = not detected (<0.2 ppb)

Source: NSW EPA (1995).

Although this study is based upon a limited number of samples, some broad comparisons with other roadside monitoring studies are possible. Levels of freon-11, freon-12, ethylbenzene, and chloromethane are within the confidence limits of a study conducted by EPA Victoria in 1996 on the Westgate Freeway (EPA Victoria 1998b). However, benzene and toluene levels were lower than those found in the Victorian investigation, where the selected sites were located at a distance from the road that represented a worst-case exposure for the small section of the community living close to the Westgate Freeway (see 'Motor vehicle emissions of air toxics on freeways' in Section 4.1.2).

In-cabin motor vehicle study

The CSIRO Division of Coal and Energy Technology undertook a study of the concentrations of benzene, 1,3-butadiene and carbon monoxide inside the cabins of moving motor vehicles in Sydney (Duffy and Nelson 1996). Comparisons were made between cars equipped with catalysts and not equipped with catalysts, morning peak hour commuters and freeway driving (via the Sydney Harbour Tunnel). Samples of air were also collected at midday, about three hours after completion of the commuter trips. VOC concentrations were also measured in the ambient air to evaluate the relationship between vehicle VOC concentrations and those expected at background levels in the urban environment. Duplicate samples were collected from the back seat of the vehicles in stainless steel SUMMA passive canisters and analysed using the US EPA TO-14 method. The average concentrations of in-vehicle and ambient concentrations for benzene, 1,3-butadiene, carbon monoxide, and carbon dioxide are given Table 4.20.

Table 4.20: Mean in-vehicle concentrations for benzene, 1,3-butadiene, carbon monoxide and carbon dioxide using pre 1986 (not catalyst-equipped) and post-1986 (catalyst-equipped) vehicles
Test Benzene (ppb) 1,3-Butadiene (ppb) CO (ppm) CO2 (ppm)
Average in-vehicle (post-1986), morning peak hour 22.1 ± 4.1 5.5 ± 2.1 16 ± 2 595 ± 26
Average in-vehicle (pre-1986), morning peak hour 48 ± 6.9 11.5 ± 3.0 22 ± 3 659 ± 56
Average in-vehicle (post-1986), midday 5.4 ± 2.2 < 0.1    
Average in-vehicle (pre-1986), midday 32.0 ± 9.3 < 0.1    
Average ambient (post-1986) 2.0 ± 1.4 < 0.1    
Average ambient (pre-1986) 1.8 ± 1.1 < 0.1    

Notes:
ppb = parts per billion;
ppm = parts per million

Source: Duffy and Nelson (1996).

The presence of a catalyst had a marked influence on the levels of benzene and 1,3-butadiene. Concentrations observed in catalyst-equipped vehicles were more than double those in vehicles not equipped with catalysts. In-vehicle concentrations of benzene were about 11 and 24 times higher than the average ambient air concentration for post-1986 and pre-1986 vehicles, respectively. In the older vehicles tested, benzene concentrations dropped by a factor of two by midday, compared with a factor of four for catalyst-equipped vehicles. At midday, therefore, benzene levels were over 10 times higher in older vehicles than in ambient air.

Traffic density (morning peak hour) and vehicle speed (midday freeway travel) also influenced the concentrations of benzene and 1,3-butadiene. For both of these substances, the freeway travel levels were about an order of magnitude lower than those observed for the urban commuter trips.

1,3-Butadiene was observed only inside the cabins of moving vehicles during peak hour traffic. It was contended that high concentrations of this hydrocarbon would be observed only in fresh exhaust emissions, as the ambient air and midday samples were below the detection limit for 1,3-butadiene (0.1 ppb). The authors also proposed that for cars parked in underground or covered parking stations during the day there may be higher concentrations of benzene during evening trips than during morning trips. This is due to benzene tending to decay much more slowly than 1,3-butadiene.

The results of the Sydney study are directly comparable with a pilot study conducted by EPA Victoria in 1997 (Torre and Bardsley 1998). In the Victorian study, the levels of benzene and 1,3-butadiene during morning peak hour, within a 1991 model car, were 13.2 ± 2.6 ppb and 3.4 ± 0.7 ppb respectively. Both the benzene and 1,3-butadiene levels were slightly lower in the Melbourne study. The higher benzene levels in the Sydney study may be due to differences in driving conditions.

As an extension to the in-vehicle study, Duffey and Nelson (1997) measured the concentrations of benzene and 1,3-butadiene inside one bus with and one without airconditioning during morning and evening peak-hour periods. The results of this investigation are presented in Table 4.21.

Table 4.21: In-vehicle concentrations of benzene and 1,3-butadiene for a bus with airconditioning and one without airconditioning during driving at various times throughout the day and in different ventilation conditions
Type of bus Time Ventilation condition Sample Concentration (ppb)
Benzene 1,3-Butadiene
Non-air conditioned Morning peak-hour All windows closed Inside 9.4 2.6
Outside 10.3 2.8
Morning peak-hour Some windows open Inside 9.7 2.8
Outside 10.8 2.8
Midday Some windows open Inside 6.1 1.7
Outside 7.3 1.9
Evening peak-hour Some windows open Inside 10.1 3
Outside 11.9 3.6
Air conditioned Morning peak-hour Inside 7.2 2.1
Outside 7.9 2
Midday Inside 4.5 1.2
Outside 5.7 1.4
Evening peak-hour Inside 6 1.8
Outside 9.1 2.3

Note: ppb = parts per billion

Source: Duffy and Nelson (1997).

The concentrations of both benzene and 1,3-butadiene inside the bus were about 50% of the in-vehicle average for catalyst-equipped cars, and about 25% of that for cars not equipped with catalysts (see Table 4.20). This difference was attributed to:

The concentrations of both benzene and 1,3-butadiene inside the bus followed a similar trend to that shown in the car study, with midday levels being 30–40% lower than those measured during peak hour. In addition, the air conditioned bus had slightly lower levels than the bus without airconditioning. This was attributed to the intake of the airconditioning unit being on the roof of the bus and therefore having more vertical dilution.

4.2.2 Industrial emissions

Air toxics at a landfill site

The NSW EPA conducted a program in 1995 to identify and quantify a range of air pollutants that may be present at or released from the Castlereagh Waste Management Centre (CWMC) (Dean et al 1996). The CWMC has operated as a secure waste landfill since 1974; it is used for disposal of industrial liquid, sludge and solid wastes. The study involved ambient air sampling at neighbouring residences and in background areas. Residential sampling was directed at identifying and quantifying ambient levels of pollutants close to the site perimeter and at residences neighbouring the site. Ambient air was sampled continuously for 14 days at three locations near the site perimeter, and for two 48-hour periods at residences neighbouring the site.

Table 4.22: Summary of mean values (ppb) of AVOCS residential sampling
  Area 2 Area 8 Area 10c Blacktown Richmond Llandilo Road Northern Road Min Max
Freon-12 1.3 0.8 0.8 0.6 0.5 0.6 0.6 0.5 1.8
Chloromethane 1.6 0.9 0.9 0.6 1.6 0.8 0.9 0.6 3
Freon-14 0.1               0.2
Vinyl chloride 0.9 0.5             1.5
Methyl bromide 0.1               0.2
Chloroethane 0.3 0.5     0.4       0.8
Freon-11 0.4 0.3 0.3 0.3 0.2 0.3 0.3 0.2 0.5
Vinylidene chloride 0.1               0.2
Dichloromethane 0.5 0.8 0.3 0.2   0.2     1.5
Freon-113 0.2 0.1             0.2
1,1-Dichloroethane 0.2 0.5 0.1           0.9
cis-1,2-Dichloroethylene 0.6 0.5             1
Chloroform 0.1 0.1             0.2
1,2-Dichloroethane 0.1 0.1             0.2
Methyl chloroform 0.3 0.5 0.1           0.9
Benzene 1 1.5 0.6 1 0.6 0.5 0.4 0.3 2.3
Carbon tetrachloride 0.1               0.2
1,2-Dichloropropane                  
Trichloroethylene 0.4 0.4   0.3         0.6
cis-1,3-Dichloropropene                  
trans-1,3-Dichloropropene                  
1, 1,2-Trichloroethane                  
Toluene 3.1 3.1 1.4 2.4 1.7 0.8 0.8 0.5 5.5
1,2-Dibromoethane                  
Tetrachloroethylene 0.1 0.2             0.2
Chlorobenzene 0.1               0.2
Ethylbenzene 0.4 0.5 0.2 0.4 0.2       0.7
(p+m)-Xylene 1 1 0.7 1.1 0.7 0.2 0.2   1.9
Styrene 0.4 0.3 0.2 0.2 0.2       0.9
1, 1,2,2-Tetrachloroethane                  
o-Xylene 0.4 0.4 0.2 0.5 0.3 0.2     0.9
1,3,5-Trimethylbenzene 0.1 0.1   0.2         0.2
1,2,4-Trimethylbenzene 0.2 0.3 0.2 0.6 0.3 0.2     0.6
m-Dichlorobenzene 0.1     0.2         0.3
Benzyl chloride                  
o-Dichlorobenzene                  
p-Dichlorobenzene 0.1               0.2
1,2,4-Trichlorobenzene                  
Hexachlorobutadiene                  
3-Chloropropene                  
4-Ethyltoluene 0.2 0.1 0.1           0.3

Notes:
ppb = parts per billion;
AVOCS = ambient volatile organic canister samples;
empty cells mean substance not detected

Source: Dean et al (1996).

Background (sector sampling) canister samplers collected air samples at a constant rate while connected to a wind direction sensor to distinguish between pollutants originating from the landfill and pollutants originating from off-site sources. When the wind came from the area under investigation, a sample was directed into an 'in sector' canister; when wind came from other directions it was directed into an 'out sector' canister. In addition, a meteorological station was established on site for the duration of the study to record wind speed, wind direction and temperature.

A total of 28 ambient air samples, each of 48 or 72 hours duration, were collected and analysed for 40 compounds in the residential sampling component of the study. Table 4.22 shows that mean concentrations were similarly low across all sites. Internal sites showed slightly elevated concentrations of some pollutants in comparison to the four sites located outside the CWMC perimeter. Pollutant concentrations at residential sites neighbouring the CWMC were below concentrations measured at one or both background sites, suggesting that any pollutants emitted from the CWMC rapidly disperse and are not discernible from background concentrations by the time the air stream reaches nearby residential areas.

The sector sampling indicated that there were no pollutants that could be clearly identified as originating from the landfill site, as differences between 'in' and 'out' sector measurements were very small. Furthermore, mean concentrations of air toxics at residences neighbouring the site were similar to or lower than background levels. Further statistical analysis, however, indicated that eight compounds found in the samples were probably emitted from the site, including 1,1,1-trichloroethane, 1,2,4-trimethylbenzene, benzene, chloroethane, dichloromethane, chlorfluorocarbon-12 (freon-12), o-xylene and toluene. Sampling results from neighbouring residences suggest that these emissions rapidly disperse to undetectable levels as distance from the site boundary increases. It was, therefore, concluded that the CWMC site does not impact upon ambient air quality outside the site boundary, because at all locations concentrations of all pollutants measured were lower than at other urban monitoring sites.

Mining operations

Dust from mining operations has been identified as a potential source of air toxics. Mining operations that generate dust include coal mining operations, combustion facilities and gold extraction. Very little is known about the levels of air toxics emitted from these types of operations in Australia. An ambient air quality goal for PM10 has been set by the NEPC; however, there has been little investigation to date of the levels of PAHs, metals, PM10 and other pollutants in the dust fallout from these types of industries.

As part of a statement on environmental effects, Stawell Gold Mining Pty Ltd monitored the background levels of dust fallout in the Stawell area for five consecutive months at nine sites (Stawell Gold Mines Pty Ltd 1999). The collected deposited matter was analysed for a number of elements. The maximum background monthly average for arsenic in the Stawell region was 0.0004 g/m²/month. In all but one of the sites, arsenic levels were less than or equal to 0.0001 g/m²/month.

Yallourn Energy monitored PAH concentrations in ambient PM10 samples near existing coal mining operations as part of an assessment of the potential health impacts of a proposed Maryvale Coal Field development in the Latrobe Valley (Yallourn Energy Pty Ltd 1999). The limit of detection for individual PAHs was set at 20 ppb, as this was deemed sufficiently low to indicate that PAHs are not to expected to be of concern in the dust produced from the proposed mine. Within this detection limit, no PAHs were detected in the PM10 samples collected.

The ambient levels of PAHs near coke oven works at Corrimal, Coalcliff and Warrawong have been monitored by industry. The results were compared to background levels measured in the Latrobe Valley. The range of total PAHs and benzo[a]pyrene are compared to the Texas Natural Resources Center effects screening levels in Table 4.23.

Table 4.23: Summary of measured PAH concentrations near coal oven works in Corrimal, Coalcliff and Warrawong
Location/chemical Range of concentrations (µ g/m³, 24-hour averages) Texas Natural Resources Center effects screening level 1997 (µ g/m³, 24-hour averages)
Latrobe Valley (Victoria)
Benzo[a]pyrene equivalent up to 0.0046 0.03
Corrimal, Coalcliff and Warrawong (NSW)
Total PAH up to 0.047 0.5
Benzo[a]pyrene up to 0.006 0.03

Note: PAH = polycyclic aromatic hydrocarbon

Source: Yallourn Energy Pty Ltd (1999).

4.2.3 Wood smoke

Wood combustion contributes PM10 and PM2.5 emissions, CO, VOCs and a number of air toxics, such as PAHs and formaldehyde, to the atmosphere.

Researchers from the University of Washington in the United States of America (Larson et al 1994, cited in Lyons et al 1996) have estimated the chemical composition of wood smoke (Table 4.24). However, when released into the environment, many compounds in wood smoke are expected to undergo some degree of chemical transformation in the atmosphere. Factors such as fuel type, catalytic inclusions and degree of combustion also significantly influence the composition of the resulting smoke.

Table 4.24: Summary of the chemical composition of wood smoke
Species Level (g/kg wood) Physical state
Carbon monoxide 80–370 Vapour
Methane 14–25 Vapour
VOCs (C2–C7) 7.0–27 Vapour
Aldehydes 0.6–5.4 Vapour
Formaldehyde 0.1–0.7 Vapour
Acrolein 0.02–0.1 Vapour
Propionaldehyde 0.1–0.3 Vapour
Butyraldehyde 0.01–1.7 Vapour
Acetaldehyde 0.03–0.6 Vapour
Furfural 0.2–1.6 Vapour
Substituted furans 0.15–1.7 Vapour
Benzene 0.6–4.0 Vapour
Alkyl benzenes 1.0–6 Vapour
Toluene 0.15–1.0 Vapour
Acetic acid 1.8–2.4 Vapour
Formic Acid 0.06–0.08 Vapour
Oxides of nitrogen (N0 NO2) 0.2–0.9 Vapour
Sulfur dioxide 0.16–0.24 Vapour
Methyl chloride 0.01–0.04 Vapour
Naphthalene 0.24–1.6 Vapour
Substituted naphthalenes 0.3–2.1 Vapour and/or particle
Oxygenated monoaromatics 1.0–7 Vapour and/or particle
Guaiacol (and derivatives) 0.4–1.6 Vapour and/or particle
Phenol (and derivatives) 0.2–0.8 Vapour and/or particle
Syringol (and derivatives) 0.7–2.7 Vapour and/or particle
Catechol (and derivatives) 0.2–0.8 Vapour and/or particle
Total particle mass 7.0–30 particle
Particulate organic carbon 2.0–20 particle
Oxygenated PAHs 0.15–1.0 Vapour and/or particle
PAHs: fluorene, phenanthrene, anthracene, methylanthrancenes, fluoranthene, pyrene benzo[a]anthracene, chrysene, benzofluoranthenes, benzo[e]pyrene, benzo[a]pyrene, perylene, ideno [1,2,3,-cd] pyrene, benz[ghi]perylene, coronene, dibenzo[a,h]pyrene, retene, dibenz[a,h]anthracene   Vapours
Trace elements: Na, Mg, Al, Si, S, Cl, K, Ca, Ti, V, Cr, Mn, Fe, Ni, Cu, Zn, Br, Pb   Particle
Particulate elemental carbon 0.3–5.0 Particle

Note: PAH = polycyclic aromatic hydrocarbon

Source: Larson et al (1994), cited in Lyons et al (1996).

Tasmania

A pilot study found that winter air pollutants were a significant problem in the Launceston area. This was particularly the case for particle levels in the winter and the cooler periods of late autumn and early spring. Woodheaters were a major cause of air pollution in winter, with an estimated 600 tonnes of woodsmoke particles being produced at this time. A more coordinated study on air pollution, environmental health and respiratory diseases conducted in the Launceston area investigated PM10 at five sites and PM2.5 at one of these sites over three years (Lyons et al 1996). The investigation also included quantitative analyses of particle samples for the recommended 16 PAHs in the US EPA profile (Table 4.25).

Table 4.25: The 16 US EPA priority PAHs
Priority PAH Priority PAH
Naphthalene Benzo[a]anthracene
Acenaphthylene Chrysene
Acenaphthene Benzo[b]fluoranthene and benzo[k]fluoranthene
Fluorene Benzo[a]pyrene
Phenanthrene Indeno [1,2,3cd] pyrene
Anthracene Dibenz[ah]anthracene
Fluoranthene Benzo[ghi]perylene
Pyrene 4-Hydrox-3,5-dimethoxy benzaldehyde (syringaldehyde)

Source: Lyons et al (1996).

The study demonstrated that there was a sharp increase in particle levels in May, peaking in June and July and still high in August. The levels for one PAH, BaP, is shown in Table 4.26. BaP is regarded as an index for the carcinogenic potential of the PAH fraction of ambient air. Although there are numerous possible sources for PAHs, many slow-combustion woodheaters are known to produce large quantities of incomplete products of combustion. The results from the PAH profiles were consistent with the PAHs having originated from the pyrolysis of wood. Syringaldehyde and some related pollutants that were present are known to be chemical compounds that result specifically from the combustion of the lignin in wood.

Levels of BaP in the Launceston study peaked mainly in the winter months of June and July 1992. The mean and summer levels were significantly lower than the peak winter levels. The Launceston study concluded that BaP levels in the winter months were disturbing, because Launceston is a small city in a rural area. It also noted that, unless there was a major change in the community's domestic heating practices, the risk to health would increase with the increasing population of the area.

Table 4.26: Summary of benzo[a]pyrene concentrations (ng/m³) at in the Launceston area, 1991–93
Site Period Mean (ng/m³) Maximum(ng/m³) Minimum(ng/m³)
Ti Tree 24/07/91–28/09/91 2.23 12.2 0
04/05/92–27/12/92 2.11 19.47 0
02/01/93–03/09/93 1.24 7.46 0
Newham 24/07/91–28/09/91 2.87 12.76 0.06
04/05/92–27/12/92 2.33 33.07 0
02/01/93–03/09/93 1.29 39.67 0
East Launceston 24/07/91–28/09/91 2.8 15.62 0
04/05/92–27/12/92 3.09 34.28 0
02/01/93–03/09/93 1.66 8.95 0
Glen Dhu 24/07/91–28/09/91 3.79 15.99 0.09
04/05/92–27/12/92 2.18 20.39 0
02/01/93–03/09/93 1.18 7.15 0
Newstead 04/05/92–27/12/92 0.5 2.73 0
02/01/93–03/09/93 1 4.75 0

Source: Lyons et al (1996).

4.2.4 Agvet chemicals (pesticides)

In response to community concerns about the health effects of exposure to agricultural chemicals, ambient air was monitored for pesticides at four sites in Coffs Harbour during the peak agricultural spraying period of the summer of 1992–93 (Beard et al 1995). Six pesticides were detected (see Table 4.27).

Table 4.27: Pesticides detected in ambient air, Coffs Harbour, 1992–93
Site Heptachlor Chlorpyrifos
% of samples with detected levels Ambient level range (mg/m³) Mean ambient level (mg/m³) % of samples with detected levels Ambient level range (mg/m³) Mean ambient level (mg/m³)
Site 1 0 ND ND 5.2 ND–12.0 1.9
Site 2 2.2 ND–21.0 0.544 23.2 ND–208 0.57.444
Site 3 0 ND ND 8.5 ND–12.0 2.95
Site 4 21.2 ND–133 7.23 17.1 ND–25.0 3.9
All sites 7.1 ND–133 2.70 14.5 ND–208 3.6

Note: ND = not detected

Source: Beard et al (1995).

Chlorpyrifos was the most commonly detected pesticide. The other pesticides detected (chlordane, ethoprophos, diozinon and dieldrin) were found only rarely. Propconazole, the only pesticide applied by air in the district, was not detected. This was thought to be due to relatively low rates of drift and minimal evaporation after application. The study concluded that, even in a semirural town with nearby widespread use of agricultural chemicals, community exposure to pesticides in ambient air may largely relate to nonagricultural use.

Using international health guidelines as a yardstick, the study concluded that risks associated with exposure to pesticides in ambient air in Coffs Harbour (and, it was assumed, in other parts of Australia) are likely to be negligible. For instance, the estimated 24-hour inhalation exposures to heptachlor represented at most 3.8% of the acceptable daily intake set by the WHO; for chlorpyrifos the comparable figure was estimated to be 0.05%. However, the study cautioned that indoor air levels may be at least an order of magnitude higher if domestic usage was the predominant source of the pesticides.

Information on pesticide levels detected indoors can be found in 7.3.3.

4.2.5 Ambient air toxics studies

Polycyclic aromatic hydrocarbons – a review of Australian studies

The following text provides a summary of the State of Knowledge Report on PAHs in Australia, which Environment Australia commissioned the Department of Environmental Protection, Western Australia (WA DEP) to prepare. The full text of the report may be downloaded from the Living Cities – Air Toxics Program website.

PAHs are released into the atmosphere as a complex mixture of compounds resulting from incomplete combustion of organic matter. Although hundreds of PAHs have been identified in atmospheric particles, toxicological endpoint and/or exposure data are available for only 33 PAHs. They are widespread contaminants of the environment and a number of them are either known or suspected human carcinogens. Information on the fate and behaviour of PAHs in the air environment in Australia has been collected on an ad hoc basis and is located in a range of different organisations in different states and territories.

The WA DEP study on PAHs was conducted as a desktop study and reports PAH concentrations in ambient air in Australia. The study also compared sampling and analytical methods to provide information on the comparability of results from different jurisdictions.

There have been eight main studies of PAH concentrations in Australian cities since 1990. The studies were conducted in:

The average concentrations, and the range, of PAHs determined in Australian cities since 1990 are summarised in Table 4.28.

In most of the Australian studies, PAH emissions were attributed to domestic heating or other combustion activities prevalent during winter. The Launceston study, for example, recorded the highest benzo[a]pyrene concentration (34.3 ng/m³), which coincided with periods of high haze levels in winter.

At present, ambient PAH monitoring is being carried out by the NSW EPA. The ACT Government Analytical Laboratory has monitored PAHs in the past but is no longer doing so. The Environmental Aerosol Laboratory at the Queensland University of Technology is also engaged in PAH research.

The literature reviewed in the WA DEP State of Knowledge Report on PAHs in Australia suggests that the presence of chrysene and benzo[k]fluoranthene may be indicators for coal combustion emissions, whereas benzo[g,h,i]perylene, coronene and phenanthrene are indicators for motor vehicle emissions; phenanthrene, pyrene and fluoranthene are associated with incineration; and fluorene, fluoranthene and pyrene are associated with oil combustion. All these PAHs were identified in the Australian studies.

The WA DEP study found it difficult to compare PAH levels between different jurisdictions, because they were not determined by the same method. Several different methods were used for collecting samples, but more standardised methods were used for chemical analyses of PAHs. In addition, different suites of PAH compounds were studied.

The full WA DEP report details the sampling and analytical methods employed in each study, as well as providing comments regarding some of the problems faced and suggestions as to how these problems may be overcome. No standard or systematic measurement methods have been used in most of the studies across Australia. Since it is desirable to compare data across jurisdictions, the full report presents recommendations for the sampling and measurement of PAHs.

PAH concentrations in Australian cities were found to be comparable to those reported in the international literature. Winter PAH concentrations were two to five times higher than summer concentrations; this differential has also been reported in overseas studies.

There are no ambient air quality standards for PAHs in Australia. However, the annual average concentration of benzo[a]pyrene in Australian and European cities appear to be below available European guidelines. The Dutch National Institute of Public Health and the Environment (DNIPHE) has recently determined maximum permissible concentrations (MPCs) for toxic compounds. These MPCs represent risk limits. The highest annual average benzo[a]pyrene concentrations in Perth and Launceston were greater than the DNIPHE-calculated MPC value (1.0 ng/m³) in ambient air.

Table 4.28: Ambient polycyclic aromatic hydrocarbon concentrations (ng/m³) measured in Australian cities
PAH   Brisbanea Perthb NTc Launcd Melbourne Canberrah NSWi
MASe DPf Collingg
Number of samples   30 62 121 NA 7 51 12 42 708
Averaging period   2–9 days 8–56 h 24 h 24 h 8 h 24 h 24 h 24 h 24 h
    Vap. Part. Part. Part. Part. Part. Part. Part. Part. Part.
Total PAH (SPAHs)j Max 813 27.8 204 13.1 NA 121 12.06 21.5 1.20 136
Min 17.1 0.43 1.04 NA NA 28.0 0.08 1.68 0.048 Bdl
Ave 217 5.38 34.6 1.85 NA 63.2 1.91 6.43 0.360 9.0
Number of PAHs that SPAHs is based on   21 21 20 20   10 12 20 7 16
Acenaphthene Max 25.0 Bdl 0.198         1.59   1.0
Min 0.60 Bdl Bdl         Bdl   0.1
Ave 5.86 Bdl 0.030         1.48   0.6
Acenaphthylene Max 70.0 0.12 0.525         5.56   36.0
Min 0.62 Bdl Bdl         Bdl   0.1
Ave 10.9 0.03 0.081         0.17   3.3
Anthanthrene Max               0.11    
Min               Bdl    
Ave               0.093    
Anthracene Max 8.70 0.24 0.440     0.17 0.06 0.09   3.0
Min 0.15 Bdl Bdl     0.06 Bdl 0.02   Bdl
Ave 1.60 0.04 0.035     0.10 0.01 0.05   0.2
Anthracene, 2-methyl Max 16.00 0.58                
Min 0.37 Bdl                
Ave 4.23 0.12                
Benzo[a]anthracene Max 0.34 1.43 8.73     7.30 0.70 0.43 0.112 15.3
Min 0.01 0.01 Bdl     0.87 0.01 0.01 0.011 Bdl
Ave 0.09 0.21 0.715     3.88 0.10 0.17 0.017 1.1
Benzo[a]pyrene Max 0.18 1.95 19.7 >0.7 34.3 21.9 1.43 0.83 0.146 14.6
Min Bdl 0.01 Bdl Bdl Bdl 5.33 Bdl 0.02 0.004 Bdl
Ave 0.03 0.34 1.88 0.05 1.77 11.4 0.17 0.17 0.046 1.4
Benzo[b]fluoranthene Max 0.15k 2.31k 18.7       1.34 0.55 0.181 22.2
Min 0.01k 0.08k Bdl       0.01 0.04 0.003 Bdl
Ave 0.06k 0.55k 2.15       0.24 0.18 0.045 2.0
Benzo[e]pyrene Max 0.16 2.11 10.8     51.0   0.98    
Min Bdl 0.04 Bdl     11.19   Bdl    
Ave 0.03 0.35 1.26     27.68   0.53    
Benzo[g,h,i]perylene Max 0.01 6.60 38.7       4.26 3.38 0.578 16.3
Min Bdl 0.09 0.009       0.03 0.11 0.007 Bdl
Ave Bdl 1.39 3.18       0.58 0.86 0.172 1.3
Benzo[k]fluoranthene Max 0.06 3.04 16.0       0.70 0.19 0.081 16.4
Min Bdl 0.03 Bdl       Bdl 0.03 0.003 Bdl
Ave 0.02 0.57 1.68       0.16 0.069 0.045 1.4
Chrysene Max 0.86l 2.19l 9.79     15.1 0.65 0.47   14.6
Min 0.02l 0.02l Bdl     3.13 0.01 0.01   Bdl
Ave 0.26l 0.43l 0.801     8.16 4.70 0.22   1.2

Notes & sources:
aBrisbane data is presented separately for particulate bound (Part.) and vapour phase (Vap.) PAHs (Muller. et al 1998, 1996a, 1995a,b; Muller 1997)
bPerth (Gras 1996)
cNorthern Territory (Vanderzalm et al 1998)
dLaunceston (Expert Working Party 1996)
eMelbourne Aerosol Study (Gras et al 1992)
fDebney's Park at Flemington (VicRoads/EPA 1991)
gCollingwood (Panther et al 1999)
hCanberra (Ian Fox 1999)
iIndustrial self-monitoring data collected under conditions of NSW EPA discharge licences (NSW EPA 1996; 1997)
jSPAHs: sum of individual PAH concentrations
kReported as the sum of B[b]F and B[k]F, in some of the samples
lReported as the sum of chrysene and triphenylene
Bdl: below detectable limits;
NA = not available

Most of the data collected on the human health effects of PAH exposure arise from epidemiological studies conducted in occupational settings. There is evidence of a dose-response relationship between numerous PAHs and health endpoints, including lung cancer and depressed immune function, which were detected in coke-oven workers exposed to PAHs. Long-term exposure to PAHs can elevate the risk of various cancers and immunotoxic and respiratory problems.

Bioassay-directed chemical analysis has been used to determine PAH mutagenicities, with most studies employing bacterial assays as biological endpoints. Biological markers (biomarkers) such as PAH-DNA adducts (compounds formed by PAHs binding to DNA) have been used in molecular epidemiology studies to assess PAH-exposure and effect.

Little research has been done on the complexity of determining the health effects associated with PAH exposure in ambient air.

Western Australia

The WA DEP in Perth undertook a study during the period March 1997 to November 1998 to determine the composition and concentrations of VOCs in Perth's ambient air. A total of 157 samples were collected from six locations around Perth's airshed using the US EPA TO-14 method, involving continuous collection of air samples into canisters over 24-hour periods, on a 12-day cycle. Most samples were taken from four sites that were monitoring stations of the WA DEP. These sites were located at Duncraig (a suburb of Perth), Queen's Buildings (in the Perth CBD), Swanbourne (on the coast) and Hope Valley (downwind of the Kwinana industrial area). The remaining two sites were located at North Fremantle (downwind of petroleum storage tanks) and at Gooseberry Hill (a residential site on the Darling Scarp). Only a few samples were collected at each of these sites.

A total of 23 air toxics were detected in air samples at all of the sites. Benzene, toluene and xylenes (BTX) were the most abundant air toxic compounds at all of the sites; consequently BTX was expressed in very high maximum 24-hour concentrations over all of the sites (Table 4.29). Queen's Buildings in the CBD recorded the highest concentrations, while Duncraig recorded the second highest. Freon-12 and 1,2,4-trimethylbenzene were also present at significant levels, while the remaining compounds recorded low average 24-hour concentrations. The annual average 24-hour concentrations for Perth were calculated from 24-hour concentrations obtained from all of the sites (Table 4.29).

Table 4.29: Annual average, maximum, 90th percentile and minimum 24-hour concentrations of air toxics detected in Perth, 1997–98
Compound Concentration (ppb)
Annual average Maximum 90th percentile Minimuma
Freon-12 1.00 11.70 1.46 0.10
Methyl chloride 0.18 5.60 0.10 0.10
Freon-114 0.18 0.78 0.44 0.10
Freon-11 0.54 3.00 1.22 0.10
Dichloromethane 0.21 1.72 0.61 0.10
Freon-113 0.26 1.14 0.74 0.10
cis-1,2-Dichloroethylene 0.18 2.06 0.10 0.10
Methyl chloroform 0.19 0.84 0.33 0.10
Benzene 1.44 17.60 2.95 0.10
Carbon tetrachloride 0.26 1.06 0.67 0.10
Toluene 2.56 30.00 5.00 0.10
Tetrachloroethylene 0.22 1.57 0.35 0.10
Chlorobenzene 0.21 1.16 0.32 0.10
Ethylbenzene 0.44 4.84 0.82 0.10
m- and p-Xylene 1.46 16.70 2.81 0.10
Styrene 0.28 2.10 0.63 0.10
o-Xylene 0.63 5.86 1.25 0.10
1,3,5-Trimethylbenzene 0.27 1.92 0.60 0.10
1,2,4-Trimethylbenzene 0.63 5.02 1.65 0.10
1,3-Dichlorobenzene 0.21 1.06 0.37 0.10
1,4-Dichlorobenzene 0.24 1.34 0.52 0.10
1,2-Dichlorobenzene 0.19 1.23 0.36 0.10
1,2,4-Trichlorobenzene 0.22 2.10 0.66 0.10

Note: a Half the limit of detection.

Source: WA DEP (2000).

These values are weighted considerably towards data obtained from Duncraig and Queen's Buildings, as these sites comprised 69% (54 samples at each) of all of the samples. The annual average concentrations ranged from 0.18 to 2.56 ppb, while the maximum 24-hour concentrations ranged from 0.78 to 30.0 ppb. At all sites there were 15 air toxics that had annual average 24-hour concentrations well below 0.2 ppb; three substances were not detected at all (Table 4.30).

Table 4.30: Compounds not detected or rarely detected in Western Australia study
Vinyl chloridea Chloroform trans-1,3-Dichloropropene
Methyl bromide 1,2-Dichloroethane 1,1,2-Trichloroethane
Ethyl chloridea 1,2-Dichloropropane 1,2-Dibromoethane
Vinylidene chloride Trichloroethylene 1,1,2,2-Tetrachloroethane
1,1-Dichloroethane cis-1,3-Dichloropropene Hexachloro-1,3-butadiene

Note: a Not detected in any of the samples.

Source: WA DEP (2000).

Quarterly averages were calculated from 24-hour average concentrations. At Duncraig, the data depict strong seasonal variations in BTX concentrations. Statistically significant differences were found between ambient levels of benzene between winter and summer, and spring and autumn. After adjusting for season, temperature relative humidity, nitric oxide and carbon monoxide concentrations, these associations still remained significant. Toluene and xylene concentrations showed statistically significant positive associations with the season and with carbon monoxide or NO concentrations. However, at Queen's Building in the CBD no statistically significant associations were found for any of the adjusted variables and the ambient benzene levels. Toluene, on the other hand, was found to have a statistically significant positive association between ambient levels in winter and spring for NO levels. Although toluene levels were greater in winter than in other seasons, the differences were not statistically significant.

Trichlorobenzene concentrations peaked in spring at both Duncraig and Queen's Building sites. Methyl chloroform peaked only during the first spring season; concentrations remained constant for the following seasons. Neither compound showed the increase in concentration during the seventh quarter (ie winter 1998) that was observed for most of the other pollutants. Total VOC average concentrations were below 20 ppb and tended to peak during winter. The highest VOC concentrations were measured during the seventh quarter on days where meteorological conditions were stagnant. Lower wind speeds and a meandering wind direction lead to lower pollution dispersion.

Table 4.31 summarises the results of monitoring at North Fremantle and Gooseberry Hill. At Gooseberry Hill, all of the TO-14 compounds, with the exception of toluene, (m+p)-xylene, freon-12 and freon-11, had average 24-hour concentrations below the limit of detection. Additionally, the aromatic hydrocarbons were found to have concentrations similar to background levels. At the North Fremantle site, the average 24-hour toluene concentration was 0.66 ppb; for benzene it was 0.37 ppb. Both of these concentrations were above background levels and were measured downwind of the petroleum storage tanks. The high values of maximum 24-hour concentrations of aromatic hydrocarbons recorded at the North Fremantle site further coincide with emissions from the nearby petroleum storage tanks.

Table 4.31: Average 24-hour air toxics concentration at North Fremantle and Gooseberry Hill, 1997–98
Compound Concentrations (ppb)
North Fremantle Gooseberry Hill
Average Max Min Average Max Min
Freon-12 0.62 0.69 0.52 0.53 0.62 0.40
Methyl chloride 0.15 0.31 0.10 NA NA NA
Freon-11 0.32 0.38 0.25 0.23 0.27 0.20
Dichloromethane 0.11 0.15 0.10 0.19 0.29 0.10
Freon-113 NA NA NA 0.15 0.20 0.10
Methyl chloroform 0.11 0.13 0.10 0.16 0.22 0.10
Benzene 0.37 0.54 0.24 0.15 0.21 0.10
Carbon tetrachloride 0.12 0.15 0.10 0.10 0.10 0.10
Toluene 0.66 1.08 0.42 0.20 0.31 0.10
Chlorobenzene NA NA NA 0.10 0.10 0.10
Ethylbenzene 0.12 0.15 0.10 0.13 0.16 0.10
m- and p-Xylene 0.41 0.51 0.33 0.25 0.43 0.10
Styrene NA NA NA 0.10 0.11 0.10
o-Xylene 0.19 0.25 0.14 0.10 0.11 0.10
1,2,4-Trimethylbenzene 0.14 0.16 0.10 NA NA NA

Notes:
ppb = parts per billion;
NA = not available

Source: WA DEP (2000).

Table 4.32 compares the annual averages reported in Perth with guidelines set by the WHO (WHO 1987, 1994, 1996, 1997), the Swedish Government (OECD 1995) and the Expert Panel on Air Quality Standards in the United Kingdom (EPAQS 1994). All of the maximum 24-hour averages were below the WHO guidelines, with the exception of chloroform. The WHO guideline for chloroform is at the minimum limit of detection (LoD) for the TO-14 method (0.2), and most of the samples collected in Perth had chloroform concentrations below this level.

With the exception of benzene, all of the annual averages fell below the quoted guideline values from the Swedish Government. The annual average benzene concentration was below the 5 ppb standard recommended by the Expert Panel on Air Quality Standards in the United Kingdom. It is important to note that the recommended United Kingdom standard is a running average value that is calculated differently from the Perth annual average. Nevertheless, ambient air toxics concentrations in Perth are well below the guidelines currently accepted.

Table 4.32: Perth ambient air toxics concentrations compared to international guidelines
Air toxic Concentration (ppb)
Perth max 24-hour average   WHO 24-hour average guidelinesa WHO 1-week average guidelinesa Perth annual average Annual average guidelines
Swedenb United Kingdomc
Freon-12 11.7     1.0    
Methyl chloride 5.6     0.2    
Freon-114 0.8     0.2    
Freon-11 3.0     0.5    
Dichloromethane 1.7 792   0.2 100–250  
Freon-113 1.1     0.3    
cis-1,2-Dichloroethylene 2.1     0.2    
Chloroform 0.7 0.2   0.1    
1,2-Dichloroethane 0.3 159   0.1 100–150  
Methyl chloroform 0.8     0.2    
Benzene 17.6     1.4 0.4 5
Carbon tetrachloride 1.1     0.3    
Trichloroethylene 0.9     0.1 100–150  
1,1,2-Trichloroethane 1.2     0.1    
Toluene 30.0   63.2 2.6 10–100  
Tetrachloroethylene 1.6 33.8   0.2 100–200  
Chlorobenzene 1.2     0.2    
Ethylbenzene 4.8   4644 0.4    
Xylenes 22.6 1013   2.1    
Styrene 2.1   55.9 0.3 10  
Trimethylbenzenes 6.9     0.9    
1,3-Dichlorobenzene 1.1     0.2    
1,4-Dichlorobenzene 1.3     0.2    
1,2-Dichlorobenzene 1.2     0.2    
1,2,4-Trichlorobenzene 2.1     0.2    

Notes:
aWHO guidelines quoted are for 24-hour and 1-week averages as indicated.
bOECD (1995).
cEPAQS (1994).
Empty cells = not available at the time

Source: WA DEP (2000).

Air toxics concentrations in Perth were very similar to those in other cities in Australia and New Zealand. Table 4.33 shows the ambient concentrations of benzene, toluene, ethylbenzene and the xylenes from Perth, Sydney (NSW EPA 1998), Melbourne (Torre et al 1996) and Auckland (Graham and Narsey 1995). Each range quoted covers the average 24-hour concentrations computed for the various sampling locations in the cities.

Table 4.33: Comparison of Perth VOC data with other cities
Sitea Concentration (ppb)
Toluene Benzene Ethyl benzene m- and p-Xylene o-Xylene
Perth (6) 0.2–4.1 0.15–2.17 0.12–0.63 0.25–2.24 0.10–0.92
Sydney (18) 0.8–5.8 0.3–2.5 0.1–0.6 0.3–2.5 0.1–0.9
Melbourne (1)b 1.2–2.4 0.5–1.4 0.1–0.2 0.4–1.1 0.2–0.5
Auckland (2) 0.64–0.61 0.53–0.67 0.05–0.13 0.22–0.34c

Notes:
a Numbers in brackets refer to the number of sites.
b There were six samplers at this site.
c Combined concentration of all three isomers of xylene.

Source: WA DEP (2000).

Table 4.33 shows that VOC concentrations in Perth are higher than those in Melbourne and Auckland, but comparable to those in Sydney. It is important to note that the study in Melbourne obtained samples from one site in Dandenong that is an urban area containing an industrial zone. This study, therefore, does not represent the Melbourne metropolitan area and is better used as a comparison with samples obtained from the Hope Valley site, downwind of the Kwinana industrial area. The maximum 24-hour benzene and toluene concentrations recorded at Hope Valley were 1.05 ppb and 1.65 ppb respectively, which are below the corresponding values from the Dandenong area. However, due to the heavy traffic conditions in the Dandenong area, this is not unusual.

In the Auckland study, samples were collected by absorption onto a solid substrate, and later removed and analysed by gas chromatography–mass spectrometry. The TO-14 method was used to collect samples for all of the other studies. The variation in results between the Auckland study and Perth could therefore be explained by the use of different sampling techniques.

The results of United States and Canadian studies suggest that the benzene levels recorded in Perth are of concern given Perth's size and population. The 24-hour annual average benzene concentration in Perth was 1.44 ppb. This is comparable with the 2.5 ppb measured at a CBD site in Sydney. In Canada, the maximum 24-hour concentration was 32.2 ppb in Montreal and 5.39 ppb in Toronto. The 24-hour maximum recorded for Perth was 17.6 ppb, an alarming figure given the size and population of Perth compared with those of the Canadian cities.

The toluene/benzene ratios for both Duncraig and Queen's Buildings peak in the winter months after falling in the summer. Both sites are affected by motor vehicle emissions all year round, but Duncraig is severely affected by woodsmoke in winter. The toluene/benzene ratio at Duncraig was lower than the ratio at Queen's Buildings, indicating that wood-fire emissions contribute significantly to benzene levels. Table 4.34 shows the toluene/benzene ratios averaged for the warmer months (November to May) and cooler months (June to October). The table also shows the ratio for Perth air containing fresh motor vehicle emissions; these figures were obtained by Galbally et al (1995) from the Perth Photochemical Smog Study. The ratios were higher in the cooler months for both sites; this can be attributed to increased photochemical activity during the warmer months. During the summer, increased photochemical activity results in lower toluene concentrations because benzene is more stable than toluene with respect to reaction with the hydroxyl radical during smog formation. The Perth Photochemical Smog Study ratio was expected to be higher than those measured during the warmer months because the samples were obtained during peak smog periods.

Table 4.34: Table 4.34 Toluene/benzene ratios in Perth (all calculated from ppb concentration data), 1997­98
Site Toluene/benzene ratioa
Nov–May June–Oct Feb 1994
Duncraig 1.69 (0.76) 1.87 (0.47)  
Perth CBD (Queen's Buildings) 1.75 (0.35) 2.05 (0.38)  
Fresh motor vehicle emissions (PPSS)b     1.82 (0.17)

Notes:
CBD = central business district
a Figures in parenthesis represent the standard deviations.
b From the Perth Photochemical Smog Study (Galbally et al 1995).

Source: WA DEP (2000).

New South Wales

This section is based on the NSW EPA publication Technical Report Pilot Air Toxics Project, May 1998 (NSW EPA 1998). The results of a follow-up study in 1997–98 were not available for inclusion in the State of Knowledge Report.

During 1995–96, the NSW EPA conducted a pilot study of selected air toxics in the Sydney region (NSW EPA 1998). The study comprised a monitoring program for organic air toxics in ambient air and development of an emissions inventory, based on an assessment of air toxic emissions from stationary sources. Initially, monitoring was conducted at 18 sites around Sydney (16 EPA air-monitoring stations and two Federal Airports Corporation monitoring sites). Canister sampling was used, with a 24-hour averaging period; US EPA Toxic Organics Method TO-14 was used for analysis of samples.

In July 1996, the number of sites was reduced to six but monitoring frequency was increased to improve the statistical power of the results. A mobile gas chromatograph was used to make additional measurements (hourly average) at two field sites (Westmead in February, March and April 1995; and Rozelle in July 1995). These measurements provided detailed information about daily variations in levels of benzene, toluene and xylenes (BTX).

In total, 203 ambient air samples were collected using canister samplers at sites throughout Sydney between May 1995 and October 1996. These samples were subsequently analysed, initially by a laboratory in the United States, and subsequently by the NSW EPA laboratory following its commissioning in mid-May 1995.

The program targeted 40 compounds specified in US EPA method TO-14 and 1,3-butadiene (although this was not monitored regularly as part of the study). Of the pollutants targeted, 31 were detected in at least one sample and 22 of these are US EPA-listed HAPs. Most of the detected pollutants were present at very low concentrations and showed similar profiles across all sites. They were close to detection limits so the levels were considered to reflect background concentrations. Only benzene was found at levels that could be of significance to public health. The average level of benzene for all sites was 1.2 ppbv, with the highest average level (2.5 ppbv) recorded at the Sydney CBD site.

Samples taken by the mobile laboratory showed low average levels of BTX pollutants over the monitoring periods at both Westmead and Rozelle. These results were similar to those obtained from canister sampling of organic ambient air toxics at the same sites. Diurnal pollutant variations typical of traffic-affected areas were evident from averaged hourly concentrations for each pollutant over each day of monitoring data. Peaks could be related to traffic volumes, with peak overnight values accounted for by the build-up of pollutants within overnight surface temperature inversions.

A desktop emissions inventory was developed for the preliminary estimation of the quantities, types and sources of air toxics emissions throughout metropolitan Sydney. The inventory was compiled by drawing information from:

Extrapolations from the US TRI database indicated that approximately 45 000 tonnes of the 320 chemicals nominated in the US TRI program in 1988 may be released in the Sydney Statistical Region each year.

The chemical, petroleum and coal products industries were the largest contributors to air toxics emissions from the manufacturing sector. Toluene was the pollutant with the potential to be released in greatest quantity, followed by 1,1,1-trichloroethane, xylenes, methyl ethyl ketone, and dichloromethane. Potential high-risk industries include petroleum refineries, drycleaners, petroleum distribution centres (including service stations), and industries with chromium-containing cooling towers.

The inventory project was undertaken as a scoping exercise, so it cannot yield definitive data on sources and emissions of air toxics. However, it does attempt to give some indication of the likely extent of the problem so that the next phase of inventory development can be justified and planned.

Queensland

The Department of Environment, Queensland (Muller et al 1998), undertook a study into PAHs in the ambient air of the Brisbane metropolitan area as part of a large-scale investigation into vehicle emissions. Atmospheric concentrations of PAHs were determined in air samples collected from seven sites in different urban locations. Thirty ambient air samples were collected and analysed for PAHs with a molar mass smaller than 178g/mol (Table 4.35).

Table 4.35: Mean and range for total concentrations of all determined PAHs from the seven field sites
Compound Mean total concentration (F + A) (ng/m³)a
Site 1
(n=11)		 Site 2
(n=4)		 Site 3
(n=3)		 Site 4
(n=2)		 Site 5
(n=8)		 Site 6
(n=1)		 Site 7
(n=1)
Phenanthrene 15
(4.1–28)		 36
(16–60)		 36
(23–47)		 9.6
(8.3–11)		 17
(12–27)		 11 14
Anthracene 0.8
(<0.03–0.6)		 4.3
(1.2–8.8)		 2.6
(1.2–3.6)		 1.2
(1.0–1.4)		 1.6
(0.91–2.4)		 0.2 1
2-Methylanthracene 1.6
(<0.1–4.5)		 10
(4.8–16)		 11
(6.9–15)		 1.8
(1–2.5)		 2.8
(2.0–3.7)		 1.3 2.2
Fluoranthene 1.6
(0.6–2.7)		 7.6
(3.3–11)		 4.5
(2.3–5.8)		 1.8
(1.3–2.4)		 2.7
(2.0–4)		 1 2.2
Pyrene 2.4
(0.79–10)		 14
(3.9–31)		 6.1
(4.7–7.5)		 4.3
(4.0–4.5)		 4.0
(2.2–6.2)		 1.8 2.8
Benzo[a]anthracene 0.0052
(0.014–0.096)		 1.2
(1.0–1.4)		 0.17
(0.15–0.18)		 0.11
(0.09–0.13)		 0.23
(0.15–0.52)		 1 0.1
Chrysene + triphenylene 0.34
(0.14–1.1)		 2.0
(1.4–2.3)		 0.53
(0.45–0.66)		 0.38
(0.3–0.46)		 0.53
(0.39–0.91)		 0.2 0.3
Benzo[b]fluoranthene + Benzo[k]fluoranthene) 0.38
(0.11–0.9)		 4.1
(2.6–5.3)		 0.78
(0.58–1)		 0.39
(0.25–0.54)		 0.57
(0.26–1.18)		 0.3 1.2
Benzo[e]pyrene 0.15
(0.05–0.3)		 1.5
(1.2–2.1)		 0.47
(0.15–0.76)		 0.11
(0.06–0.15)		 0.16
(0.12–0.26)		 0.1 0.4
Benzo[a]pyrene 0.1
(<0.01–0.22)		 1.5
(1.2–2.0)		 0.27
(0.23–0.33)		 0.14
(0.05–0.23)		 0.21
(0.15–0.44)		 0 0
Perylene 0.006
(<0.01–0.032)		 0.19
(0.13–0.31)		 0.023
(<0.05–0.07)		 <0.1 0.013
(<0.01–0.04)		 <0.05 <0.05
Indeno[1,2,3-cd]pyrene 0.31
(<0.1–0.41)		 2.9
(1.7–4.3)		 0.75
(0.25–1.6)		 0.15
(<0.02–0.15)		 0.17
(<0.1–0.35)		 <0.1 <0.02
Benzo[ghi]perylene 0.22
(<0.1–0.45)		 4.9
(3.4–6.6)		 1.5
(0.62–2.7)		 0.31
(0.25–0.37)		 0.71
(0.43–1.3)		 <0.1 2.7
Coronene 0.28
(<0.1–0.28)		 NA 0.54
(<0.1–0.54)		 NA   NA NA

Notes:
a Numbers in brackets indicate range.
NA = not available;
F + A = vapour-phase and particle-associated PAHs.

Source: Muller et al (1998).

PAH concentrations were highest at the sites near major roads (sites 2, 3, 4, 5 and 7); traffic was identified as the major source of PAHs in Brisbane's air. At site 2, near a major intersection (up to 42 000 vehicles per day), levels of BaP exceeded 1 ng/m³. Good correlations were found between the levels of BaP and data for carbon monoxide and oxides of nitrogen. BaP may therefore be a good indicator compound when traffic is the main PAH source.

The Brisbane ambient PAH study showed strong variability in PAH levels across sites. Values obtained in winter were up to four times higher than in summer. The sample collected from site 2 had the highest overall PAH concentration and the highest proportion of high molecular mass PAHs. However, the authors (Muller et al 1998) emphasise that is it difficult to estimate the contributions of different pollution sources using ratios of chemicals. Differences in the ratios are likely to be due to differences in the mobility of different chemicals and/or persistence during transport than to major differences in source contribution.

From April 1998 to March 1999, the Queensland EPA monitored ambient air toxics in the Brisbane CBD using the DOAS Opsis system. The path length was 720 m, 10–20 m above ground level and crossed inner-city streets and the South East Freeway. The air toxics monitored were benzene, toluene and (from October 1998) p-xylene. The results are presented in Tables 4.36 and 4.37.

Table 4.36: Benzene, toluene and p-xylene levels in the Brisbane CBD: maximum monthly 24-hour average ambient concentrations, April 1998 to March 1999
Pollutant Maximum monthly 24-hour average ambient concentration (ppb)
Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar
Benzene 1.7 2.7 1.9 2.1 1.6 1.7 1.5 1.2 1.1 1.4 1.0 1.3
Toluene 7.3 10.2 9.8 8.2 7.3 5.7 5.0 3.9 3.0 1.9 1.7 2.3
p-Xylene No data No data No data No data No data No data 1.9 1.7 1.6 2.3 1.5 1.6

Source: Unpublished information from the Queensland EPA.

Table 4.37: Benzene, toluene and p-xylene levels in the Brisbane CBD: ambient concentrations, April 1998 to March 1999
Pollutant Max (ppb) Second highest (ppb) Percentiles Min (ppb) Annual average (ppb) Number of data values Data availability (%)
99 (ppb) 95 (ppb) 90 (ppb) 75 (ppb) 50 (ppb) 25 (ppb)
Benzene 2.7 2.7 1.9 1.4 1.3 1.0 0.8 0.6 0.1 0.9 357 97.8
Toluene 10.2 9.9 8.3 7.1 6.0 4.1 2.5 1.6 0.4 3.1 357 97.8
p-Xylene 2.3 2.1 2.1 1.7 1.6 1.3 1.0 0.7 0.3 1.1 166 45.5

Note: ppb = parts per billion

Source: Unpublished information from the Queensland EPA.

4.2.6 Summary – ambient air toxics in Australia

Air toxics exist in the atmosphere at low levels and are of significant concern to human and environmental health. There are currently no national ambient air quality standards specifically for air toxics in Australia. In the last decade, many studies have tried to obtain better information on the sources and levels of emissions and their impact on ambient air quality. However, our knowledge is still limited, and the levels of air toxics in the atmosphere have not yet been sufficiently monitored .

Studies performed in Australia have been reviewed to summarise the current state of knowledge of the levels of air toxics from industry, motor sources and domestic and commercial sources. This review will contribute to a national assessment of air toxics. The main sources of air toxics in Australia are motor vehicle emissions, wood combustion, and large industry.

Motor vehicles typically are the largest source of air pollutants in urban areas. Levels of benzene, 1,3-butadiene, formaldehyde, toluene and xylene are often found to be highest in areas of heavy traffic. Trends differ between capital cities but it is hard to know to what extent this is due to differences in sampling techniques, target air toxics and study sites. Comparisons are also difficult because of large variations within study sites and seasons, and low average levels of air toxics overall. However, motor vehicles were found to be the largest source of benzene, toluene, 1,3-butadiene and lead, and freeways were found to contribute around 60% of downwind benzene and toluene levels in the air. The seasonal variations in air toxics were generally associated with peaks in the cooler months.

Overall, the various investigations of ambient air toxics indicate that the levels of most of these substances are below international guidelines. However, some studies have demonstrated a need for further investigations, given the relative populations of Australian and overseas cities. The introduction of ADR 37/00 in 1986 saw the advance of vehicle emissions controls, including unleaded petrol and catalytic converters on new motor vehicles. Since that time, concentrations of air toxics measured in-traffic have been reduced significantly. Nonmethane hydrocarbons have been reduced by 61%. Furthermore, VOC concentration data show that exhaust composition has a smaller fraction of reactive aromatic species and a larger fraction of short-chain unreactive alkane species, which is consistent with the influence of catalytic converters. Consequently the concentrations of VOCs have decreased dramatically since the introduction of ADR 37.

Industrial sources are also a major contributor of air toxics emissions in the atmosphere, although little research has been done into these sources. An emissions inventory undertaken by EPA Victoria in 1995–96 showed that industry was the largest source of 50% of the air toxics monitored. In addition, a study conducted around the boundary of a major chemical complex in West Footscray found that levels of VOCs were highest at the downwind sites closest to the industry. Another study performed close to the boundary of an industrial chemical site in Altona found that peak levels of benzene coincided with southwesterly winds and were consistent with emissions from the site, suggesting it was the likely source. However, investigations into landfill and mining emissions indicate they have minimal influence on ambient levels of air toxics.

In conclusion, a combination of monitoring, and emissions inventory and modelling studies is needed to obtain more information on the ambient air levels of the high-priority compounds (see EPA Victoria 1999a for further information regarding the ranking of these compounds).

4.2.7 A review of studies of hazardous air pollutants performed in Australia and New Zealand

Despite the recognised adverse health impacts associated with air toxics, many countries have not regulated concentrations of air toxics in the atmosphere or monitored for their presence in a consistent manner.

In June 1999 EPA Victoria published a report Hazardous Air Pollutants – A Review of Studies Performed in Australia and New Zealand (EPA Victoria 1999a). This review was funded partially by ANZECC. The review acknowledged that air toxics have not been monitored consistently in Australia and New Zealand although a number of monitoring, emissions inventory and modelling studies have investigated the atmospheric concentrations of many of these substances. The review's foreword, by Dr Brian Robinson (Chairman, EPA Victoria), stated:

… in order to develop policies for the protection of human health and the environment and to implement appropriate control strategies for HAPs, further information on ambient air concentrations is required.

A critical feature of the EPA Victoria review is that it addresses the need to identify priority pollutants for further study. This is an essential element for the development of any air toxics strategy, since, given the number and diversity of compounds classified as air toxics, it would not be feasible to monitor them all.

The EPA Victoria review addressed this issue by providing details of a methodology that EPA Victoria had developed to rank air toxics. It also reported on the application of this methodology to a list of 106 potentially significant compounds to yield a list of 24 priority air toxic compounds (Table 4.38).

Table 4.38: Priority air toxics identified by EPA Victoria (listed in alphabetical order)
Acrylonitrile Mercury
Arsenic and compounds Methyl ethyl ketone
Benzene Methyl isobutyl ketone
1,3-Butadiene Nickel and compounds
Cadmium and compounds Polycyclic aromatic hydrocarbons (PAHs)
Chromium and compounds Styrene
1,4-Dichlorobenzene (paradichlorobenzene) Tetrachloroethylene
Dichloromethane Toluene
Dioxins and furans Toluene diisocyanate
Fluorides Trichloroethylene
Formaldehyde Vinyl chloride (monomer)
Manganese and compounds Xylenes

Source: EPA Victoria (1999a).

The ranking process relied on a methodology that assigned scores to compounds on the basis of their potential effects on human and environmental health, and their observed (or expected) levels in ambient air. Using this method, a selection of potentially significant compounds was ranked. The initial priority list can be refined as further information becomes available.

The Air Toxics Forum saw the approach taken by EPA Victoria as a sound starting point for the Living Cities – Air Toxics Program. After discussion, the forum supported using the list of 106 compounds (Table 4.39) identified by EPA Victoria as the basis for creating a national priority list of air toxics. This support was given on the understanding that chemical substances could be added or subtracted from the initial list with appropriate justification; and that there should be an opportunity for nongovernment stakeholders to provide input into the development of criteria for ranking the pollutants.

Table 4.39: Initial list of priority hazardous air pollutants (HAPs) to be ranked
HAP Source of listing HAP Source of listing
1,1,2,2-Tetrachloroethane HR, OECD Ethyl chloride C2
1,1,2-Trichloroethane C2 Ethylacrylate C2
1,2-Propylamine HR * Ethylbenzene C2
1,3-Butadiene HR, N, OECD, C2, NZ Ethylene OECD *
1,4-Dichlorobenzene N, OECD Ethylene dibromide HR, OECD
Acetaldehyde OECD, C2, NZ Ethylene dichloride OECD
Acrolein HR, OECD, C2 Ethylene oxide HR, OECD, C2
Acrylamide HR Fluorides N, OECD, C2 *
Acrylic acid HR, C2 Formaldehyde N, OECD, C2, NZ
Acrylonitrile HR N, OECD, C2, NZ Heptachlor HR
Ammonia OECD, C2 Hexachloro-cyclohexane OECD *
Aniline OECD, C2 Hexachloro-cyclopentadiene OECD
Arsenic and compounds HR N, OECD, NZ Hexachlorobenzene HR, OECD
Asbestos HR, OECD, C3 Hexachlorobutadiene OECD
Benzene HR, HP, N, OECD, C3, NZ Hexane (n-) C2
Benzidine HR, OECD Hydrogen chloride C2
Benzotrichloride HR Hydrogen sulfide OECD, C2
Benzylchloride OECD, C2 Lead and compounds HP, N, OECD, C1
Beryllium and compounds HR, OECD, C2 Maleic anhydride C2
Biphenyl C2 Manganese and compounds N, OECD
Bis(chloromethyl)-ether HR Mercury and compounds HR, HP, N, OECD, C3, NZ
Bromoform C2 Methanol C2
Cadmium and compounds HR, HP, N, OECD, NZ Methyl bromide NZ
Carbon disulfide OECD, C2 Methyl diphenyldiisocyanate HR, C3
Carbon tetrachloride HP, OECD, C2 Methyl ethyl ketone HP, N, C2
Chlordane HR Methyl isobutyl ketone HP, N, C2
Chlorinated ethers OECD * Methyl methacrylate OECD, C2
Chlorinated paraffins OECD * Methylisocyanate HR
Chlorine C2 Methyltertbutylether OECD *
Chloroanilines OECD * Nickel and compounds HP, N, OECD, C2, NZ
Chlorobenzenes OECD, C2 Nitrobenzene OECD, C2
Chloroform HP, OECD, C2 Phenols OECD, C2
Chloromethyl methyl ether HR Phosgene HR OECD, C2
Chlorophenols OECD * Phosphine N, C2
Chloroprene HR OECD Phthalic anhydride C2
Chlorostyrenes OECD * Platinum OECD *
Chlorotoluenes OECD * Polychlorinated biphenyls (PCBs) OECD
Chromium and compounds HR, HP, N, OECD, C2, NZ Polychlorinated terphenyls (PCTs) OECD *
Coke oven emissions HR, OECD Polycyclic aromatic hydrocarbons N, OECD, NZ
Copper and compounds OECD, C2 * Propylene oxide OECD, C2
Cresols (o,m,p-) OECD Phthalates OECD
Cumene C2 Silver and compounds OECD, C2 *
Cyanide and compounds HP, C2 Styrene N, OECD, C2, NZ
Dibenzofurans HR, OECD Tetrachloroethylene HP, N, OECD, C2, NZ
Dichlorobenzidine OECD* Toluene HP, N, C2
Dichloroethyl ether HR * Toluene diisocyanate HR, N, C3, NZ
Dichloromethane HP, N, OECD, C2, NZ Trichloroethylene HP, OECD, C2, NZ
Dichlorvos C2 Triethylamine C2
Dimethylaniline OECD * Vanadium OECD *
Dimethylnitrosamine OECD Vinyl chloride HR N, OECD, C3, NZ
Dinitrotoluene C2 Vinylidine chloride HR, OECD
Dioxins HR, N, OECD, NZ Xylenes HP, N, OECD, C2, NZ
Epichlorohydrin OECD, C2 Zinc OECD, C2 *

Notes:

The list is derived from:

* = compounds which are not on the US EPA list of 189 HAPs

The review provides results from a number of air toxics monitoring, emissions inventory and modelling programs from Australia and New Zealand. Review of this data led EPA Victoria to make the following general comments.

The compounds determined to be of highest priority in the general urban ambient environment were benzene, toxic metals, 1,3-butadiene and PAHs. Dioxins, dichloromethane, formaldehyde and styrene are also of high priority. The results of this ranking process were compared with the priorities assigned to compounds for the NPI reporting list. The results of the two scoring processes show fairly good agreement. Many of the compounds found to be of highest priority are also included in the list of compounds the OECD recommends should be monitored.

Further information on the ambient air levels of the high priority compounds should be obtained through a combination of monitoring, and emissions inventory and modelling studies.

The limited data from measurements of air toxics levels in indoor and in-vehicle environments indicate that these could make a significant contribution to individuals' total exposures. Therefore, it is also recommended that further studies of levels in these environments should be performed to allow total exposure assessments to be carried out.

Full copies of the EPA Victoria report Hazardous Air Pollutants – A Review of Studies Performed in Australia and New Zealand June 1999, Volumes 1 and 2 (publication numbers 650 and 651 respectively) can be ordered by calling (03) 9695 2722.

4.3 Health studies

Environment Australia has commissioned WA DEP to undertake a desktop review and develop a database of air toxics related health studies in Australia. Once finalised, this report will be included in the State of Knowledge Report and made available on the Air Toxics Program website.

The following is an indication of the nature of this review and its outcomes.

It is clear from the review that very little information is available on the potential exposure of the Australian population to air toxics in ambient air. There have been no rigorous exposure studies for air toxics.

Through the collation of existing information, the database provides the first step in developing a targeted and coordinated investigation of air toxics and human exposure in Australian cities. The small number of studies contained in the database suggests a need for more air toxics research in the interests of public health.

In this section

Before you download

Most publications are available as PDF files. Adobe Acrobat Reader  is required to view PDF files.

If you are unable to access a publication, please contact us to organise a suitable alternative format.

Key

   Links to another web site
   Opens a pop-up window