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

4.1 Victorian case study

4.1.1 Monitoring methodologies applied in Victoria

Many different methods are used to monitor concentrations of air toxics in the atmosphere, each targeting a specific group of chemicals. The method used therefore depends on the compound that is to be measured. The most common methods used can be broadly categorised into the following two groups:

Canister sampling

Discrete sampling methods can be applied to measure a large number of air toxics simultaneously. However, care needs to be taken to prevent artefact formation and sample contamination during collection and storage. The most commonly used discrete sampling method is 'US EPA method TO-14 – Determination of Volatile Organic Compounds (VOCs) in Ambient Air Using SUMMA® Passivated Canister Sampling and Gas Chromatographic Analysis'. This method allows determination of 40 VOCs and 28 air toxics.

The TO-14 method involves the collection of whole air samples in canisters either by pressurised or subatmospheric sampling modes (ie above or below atmospheric pressure) (Winberry and Murphy 1988). Using either sampling mode requires a pre-evacuated canister and a pump-ventilated sample line. A sample of ambient air is drawn through a sampling train into the canister so that the rate and duration of sampling is regulated. The canister valve is then closed, ready to be transported to a laboratory for analysis.

The analysis process begins by reducing the water vapour in the gas stream using a Nafion® dryer. The VOCs are then concentrated by collecting them in a cryogenically cooled trap (cryogen) (Winberry and Murphy 1988). The cryogen is removed and the trap temperature raised so that the VOCs are revolatilised and can be separated on a GC column, then identified and quantified by high-resolution GC, along with one or more detectors.

The required sensitivity and specificity of the application determines whether one detector or combination of detectors are used. The detectors are either nonspecific or specific. Generally, nonspecific detectors are less expensive per analysis and in some cases are more sensitive; however, they vary in their specificity and sensitivity (Winberry and Murphy 1988). Some examples of commonly used detectors are listed below.

Nonspecific detectors Specific detectors
Nitrogen-phosphorus detector (NPD) Mass spectrometer (MS) (selected ion monitoring (SIM) mode or SCAN mode)
Flame ionisation detector (FID) Ion trap detector
Electron capture detector (ECD)  
Photoionisation detector (PID)  
Differential optical absorption spectroscopy

Differential optical absorption spectroscopy (DOAS) is an example of a continuous sampling method technique. A system based on DOAS (Opsis system) is used in a number of Australian States and Territories. DOAS is used to measure gases in both ambient and emission monitoring applications (Ecotech 1999). The system enables the concentrations of pollutants along a path to be measured and then averaged for the length of the path. A light source (emitter), receiver and an analyser are required for the process.

The emitter projects light along a path up to two kilometres long through the air to a receiver. Therefore, the positioning of the emitter and the receiver define the monitoring path. The emitter is equipped with a high-pressure xenon lamp that radiates an almost smooth spectrum ranging from approximately 200 nm to 500 nm (Ecotech 1995). This allows a number of gaseous substances to absorb light from known parts of the spectrum in the ultraviolet, visible and infrared wavelength ranges (Ecotech 1999).

The captured light at the receiver is then transferred to the analyser via a fibre-optic cable where absorption is recorded using a computer-controlled spectrometer. At the analyser, the light is refracted into its wavelength components so that each part of the spectrum can be projected onto a rapid scanning slit in front of a photomultiplier detector. This method allows all parts of the spectrum to be detected. The wavelength window can be optimised for a component with respect to parameters such as sensitivity and interfering pollutants (Ecotech 1995). At the conclusion of the data capture, some evaluation and correction procedures are required to produce a differential absorption spectrum. This contains the 'fingerprints' from all gases in the atmosphere along the light path (Ecotech 1995).

4.1.2 Results from monitoring performed in Victoria

Ambient levels of polycyclic aromatic hydrocarbons
Air toxics associated with filter based sampling for particles

The Melbourne Aerosol Study (MAS) was undertaken in 1990–91 and included PAH measurements of the fine component of aerosols under conditions of low visibility at two suburban sites in Melbourne – Alphington and Footscray (Gras et al 1992). The results show strong variability in PAH levels across sites (see Table 4.2). Values obtained at the two sites varied by up to a factor of four, with the higher values recorded at Alphington, probably due to the higher contribution from wood smoke to fine particle levels at Alphington.

Another study of PM10 and lead in air at Debney's Park Estate, Flemington, was performed by EPA Victoria and commissioned by VicRoads as part of a major study to assess the possible effects of the Western Bypass. It included measurement of the levels of PM10, lead and PAHs in the Debney's Park housing estate (about four kilometres from central Melbourne). Samples of PM10 for PAH analyses were collected every six days over 24 hours. Samples were collected on Pallflex teflon-coated glass fibre filters using an Ecotech Model 2000 high-volume sampler at a height of approximately 5 m. The samples were transferred to aluminium foil and stored in a freezer until analysed. Analysis was performed according to US EPA method 8310.

The sampling method used does not capture some of the lower molecular weight PAHs, which have a relatively high vapour pressure and pass through the filters in the gas phase. The 10 PAHs analysed were fluoranthene, pyrene, benzo[a]anthracene, chrysene, benzo[b]fluoranthene, benzo[k]fluoranthene, benzo[a]pyrene, benzo[g,h,i]perylene, dibenzo[a,h]anthracene, indeno[1,2,3-cd]pyrene. Of these, benzo[a]pyrene (BaP) is often used as an indicator for PAH pollution. BaP is the most potent carcinogen of the PAHs under consideration. The monthly average levels of BaP and total PAHs observed are shown in Figure 4.1, and the corresponding annual averages are summarised in Table 4.1.

Figure 4.1: Monthly average levels of total PAHs and benzo[a]pyrene, measured during 1990 and 1991 at Debney's Park, Flemington, Melbourne.

Figure 4.1: Monthly average levels of total PAHs and benzo[a]pyrene, measured during 1990 and 1991 at Debney's Park, Flemington, Melbourne.

Source: VicRoads/EPA (1991).

Table 4.1: Annual average levels of benzo[a]pyrene and total PAHs measured at Debney's Park, Flemington, between November 1990 and November 1991.
Location Benzo[a]pyrene (ng/m³) mean (range) Total PAHs (ng/m³) mean (range)
Debney's Park 0.17 (0.03–0.59) 2.09 (0.47–4.92)

Note: Air samples were collected over 24-hour periods

Source: VicRoads/EPA (1991).

Figure 4.1 shows that PAH levels at Debney's Park vary seasonally, with the highest levels during winter and the lowest levels in summer. The variation may be partly attributed to increased emissions of PAHs from combustion sources such as wood fires in winter. Other reasons for the seasonal variation of PAH levels are the more frequent occurrence of stable air conditions over the Melbourne area in winter, which decreases the rate at which PAHs are dispersed, and the slower rate at which PAHs are removed by photochemical processes in winter. The annual average BaP levels observed were below the interim goal set in the Netherlands of 0.5 ng/m³ as an annual average (Streeton 1990).

Table 4.2: Comparison of polycyclic aromatic hydrocarbon concentrations at sites in Melbourne (ng/m³)
PAHs Alphington Footscray Debney's Park
Phenanthrene 0.30 0.33 0.05 (0.09)
Anthracene 0.11 0.08 0.01 (0.02)
Fluoranthene 0.37 0.33 0.11 (0.16)
Pyrene 1.59 1.45 0.15 (0.21)
Benzo[a]anthracene 5.01 1.06 0.1 (0.18)
Chrysene 9.74 4.20 0.16 (0.22)
Benzo[k]fluoranthene     0.11 (0.18)
Benzo[e]pyrene 32.94 14.53  
Benzo[a]pyrene 13.33 6.44 0.17 (0.33)
Perylene 19.21 6.56  
Indeno[1,2,3-cd]pyrene     0.32 (0.46)
Dibenzo[a,h] anthracene     0.15 (0.29)
Benzo[g,h,i]perylene     0.58 (0.93)
Coronene 5.92 1.86  

Source: Gras et al (1992).

Motor vehicle emissions were the major source of airborne particle pollutants in the Debney's Park area. The results of the Debney's Park study and the MAS are summarised in Table 4.2. The values quoted for Debney's Park are annual averages, while data from Alphington and Footscray are averages of samples taken during autumn and winter.

Values obtained in the study conducted at Debney's Park (VicRoads/EPA Victoria 1991) are considerably lower than those observed at either Alphington or Footscray in the MAS. This is probably due to the fact that in the latter study (Gras et al 1992) samples were collected only on days when fine particle levels were elevated (FPM > 30 µg/m³) whereas the Debney's Park samples were collected routinely every six days.

Lead emissions

The SEPP (AAQ) (1999), which reflects the NEPM standard, states an annual average objective of 0.5 µg/m³ for lead concentrations in ambient air. Peak lead levels occur near heavy traffic areas and during periods of stable weather with light winds and poor dispersion in winter. In Melbourne, nearly all the airborne lead is from motor vehicles (EPA Victoria 1998b).

The use of unleaded petrol is the most effective long-term strategy for controlling lead in air (EPA Victoria 1997). Since the introduction of ADR 37/00 and of unleaded petrol in 1985, the lead content of petrol used by the motor vehicle fleet has progressively been reduced. The decline in lead concentrations in petrol has been reflected in the lead levels measured in the atmosphere. Table 4.3 summarises the lead concentrations measured at various sites around Melbourne between 1992 and 1997. Figures 4.2 and 4.3 clearly depict the decline in lead concentrations in the atmosphere, both in annual averages and annual maximums.

Table 4.3: Annual averages and annual maximums recorded for ambient lead concentrations in Melbourne, 1992–97
Concentration of lead (µg/m³)
Site 1992 1993 1994 1995 1996 1997
Annual average Max Annual average Max Annual average Max Annual average Max Annual average Max Annual average Max
Central business district 0.26 0.94 0.21 0.72 0.14 0.40 0.11 0.44 0.08 0.31 0.09 0.22
Alphington     0.23 1.03 0.15 0.41 0.10 0.50 0.08 0.38 0.08 0.26
Sackville     0.52 1.60 0.35 1.29 0.22 0.75 0.17 0.62 0.21 0.55
Lord 0.20 0.41 0.24 1.09 0.16 0.60 0.10 0.45 0.08 0.43 0.09 0.29
Paisley         0.10 0.44 0.07 0.37 0.05 0.29 0.06 0.18
Geelong                 0.04 0.18    

Figure 4.2: Annual averages of lead concentrations, 1992–97

Figure 4.2: Annual averages of lead concentrations, 1992-97

Note: CBD = central business district

Figure 4.3: Annual maximums of lead concentrations, 1992–97

Figure 4.3: Annual maximums of lead concentrations, 1992-97

Note: CBD = central business district

Canister sampling
Monitoring of air toxics from motor vehicle emissions

In Melbourne, motor vehicle emissions are the largest source of air pollution (EPA Victoria 1998b). Higher concentrations of motor vehicle pollutants are recorded within vehicles and in close proximity to roads. EPA Victoria and CSIRO have embarked on a number of studies to gain a better understanding of motor vehicle emissions and their impact on near-road air quality, as well as exposure of neighbouring residents and people travelling on roads by different modes of transport. A summary of the results found in these studies is presented here.

Vehicle emissions of air toxics

The former Commonwealth Environment Protection Agency commissioned a comprehensive National In-Service Vehicle Emissions Study that included a study of vehicle emissions of the air toxics benzene, toluene, xylene and 1,3-butadiene (Ye et al 1995, 1996). The aim of the study was to determine the emission performance of Australia's current passenger vehicle fleet. As a supplementary study, CSIRO (the Division of Atmospheric Research in collaboration with the Division of Coal and Energy Technology) undertook to measure air toxics emissions from a total of 50 vehicles chosen as a representative subset of the 500 vehicles selected for the larger study. The specific aims of the CSIRO study were:

Vehicle emissions were determined using the ADR 37 drive cycle test, including cold start transient, cold start stabilised and hot start transient phases. The fuels used were standard leaded and unleaded petrol supplied by Mobil. The 1,3-butadiene concentration in exhaust samples was found to decrease within a few hours of collection (probably due to reactions catalysed by oxides of nitrogen present in the sample). Corrections were applied to determine the true 1,3-butadiene concentration in exhausts. Previous emissions studies, where such corrections were not included, could be erroneous.

The greatest emissions of the air toxics studied were observed to occur during cold start conditions. This effect was most pronounced for pre-1986 vehicles and is attributed to the fact that the fuel-rich conditions that occur during start-up lead to a higher proportion of unburnt fuel in the exhaust. In addition, for newer vehicles, catalytic converters generally have poorer efficiencies during the first few minutes after ignition, before they have reached their optimum operating temperature. In general, the average per-vehicle emissions for the four air toxics from post-1986 vehicles are 30% or less of those from pre-1986 vehicles. For 1,3-butadiene, the value was 16%. Vehicle ageing and continued use are known to degrade catalyst efficiency, but this effect is less marked in the case of 1,3-butadiene.

The benzene/toluene ratio in exhausts from catalyst-equipped and noncatalyst-equipped vehicles was 0.67 and 0.59 respectively. A number of noncatalyst-equipped vehicles were tested using leaded and unleaded petrol. The emissions of air toxics were greatest when these vehicles were run on leaded petrol. The greatest difference occurred during cold start transient conditions.

The motor vehicle emission studies mentioned above focused on hydrocarbon emissions. A further study determined vehicle emission factors for the carbonyls formaldehyde, acetaldehyde, acrolein, acetone, propionaldehyde, crotonaldehyde, methyl ethyl ketone, benzaldehyde, iso-butyraldehyde, n-butyraldehyde, iso-pentanal, n-pentanal, hexanal and p-tolualdehyde (EPA Victoria 1990). Vehicles equipped with catalytic converters running on unleaded fuel emitted 2.5-17.9 mg/km of total carbonyls averaged over the complete driving cycle. In contrast, vehicles without catalytic converters running on leaded fuel emitted 110-130 mg/km. Formaldehyde, acetaldehyde, acetone and benzaldehyde jointly accounted for 80% by weight of the total carbonyls present in motor vehicle exhaust. Emissions of carbonyls were greatest immediately after starting vehicles from a cold condition.

Motor vehicle emissions of air toxics on freeways

VicRoads commissioned a study to assess particle and carbonyl pollutants close to the South Eastern Arterial Freeway (VicRoads/EPA Victoria 1994). This was conducted between 30 April and 29 July 1994 at the site of the VicRoads depot on the corner of the freeway and Tooronga Road, Malvern, approximately 100 m from the freeway. Sampling equipment was situated on the roof of the depot building at a height of approximately 5 m. The sampling system was activated manually on days of predicted poor air quality. Thus, the data obtained correspond to some of the worst-case situations. However, higher levels of these pollutants might occur due to photochemical reactions in summer. Carbonyls were sampled on Sep-Pak® dinitrophenylhydrazine-silica cartridges using a specially prepared system with low flow pump and gas meter. Samples were taken over periods of 19-27 hours (except for one sample taken over 65 hours starting on 24 April 1994). The carbonyl levels were determined by an HPLC method similar to US EPA method TO-11 – 'Determination of formaldehyde in ambient air using adsorbent cartridge followed by HPLC' (Winberry et al 1988). Altogether, samples were taken on 17 occasions. Many of the carbonyls monitored are listed as air toxics and are thus relevant to this study. These were formaldehyde, propionaldehyde, butyraldehyde and methyl ethyl ketone. The results obtained for these compounds are summarised in Table 4.4.

In an appendix to the report on the above study, results of a brief formaldehyde monitoring study conducted on the roof of the EPA Victoria building in Francis Street, Melbourne (CBD), are reported. Three samples were collected over periods of 20 to 26 hours and one over 6 hours, in March 1994. The levels measured were 2.1, 3.2, 4.2 and 7.6 ppb. These are similar to the levels measured near the South Eastern Arterial.

Table 4.4: Summary of carbonyl levels measured near the South Eastern Arterial Freeway, between 30 April and 29 July 1994.
Compound Concentration mean (range) (ppb)
Formaldehyde 3.26 (0.4–6.5)
Acetaldehyde 1.39 (0.4–2.6)
Acrolein 0.11 (<0.1–0.7)
Propionaldehyde 0.32 (<0.04–0.8)
Butyraldehyde 0.56 (<0.03–1.1)
Methyl ethyl ketone 0.34 (<0.2–0.7)

Air samples were collected over about 20 hours.
ppb = parts per billion

Source: VicRoads/EPA (1999).

In 1995, Shell Australia commissioned a short-term study in which benzene concentrations outside 1 Spring Street were monitored (Shell 1995). Daily duplicate samples were collected over 12-hour periods between 7 am and 7 pm. Seventeen samples were taken on Flinders Street 3 m from the roadway between 8 July and 24 July. Eleven samples were taken on Flinders Lane between 25 July and 4 August. The samples were collected on 400/200 mg charcoal tubes at flow rates of around 300 cm³/minute. The benzene was desorbed from the charcoal with carbon disulfide. Analysis was performed by split injection gas chromatography. Measured benzene levels ranged from 0.2 to 8.5 ppb, with a mean of 3.0 ppb. The levels measured in Flinders Street were higher than those measured in Flinders Lane, consistent with the greater volume of traffic on Flinders Street.

A further study was undertaken during the winter and autumn months of 1996 on the Westgate Freeway, Altona, to determine the levels of a number of pollutants emitted from motor vehicles and their contribution to local pollution (EPA Victoria 1998b). The selected sites for the study were located on either side of the freeway 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. A range of pollutants associated with motor vehicle traffic emissions were measured at both sites. VOCs, including benzene, were the air toxics targeted for measurement in this study. Evacuated 6-L SUMMA® canisters were used for collection of VOC samples that were manually activated in the field to collect samples during peak periods. Samples were collected in two 1-hour blocks for both morning (7 am–9 am) and afternoon (4 pm–6 pm) peak traffic periods, as well as a 24-hour sample for both sites.

The 1-hour average samples collected provide information on the variation of VOCs during peak traffic periods. The summary statistics, in particular the mean and standard deviation, derived from the data can be used as a worst-case short-term exposure situation. However, in this study they have been shown to be affected by statistical bias. This occurred because many of the VOCs measured in the study were often found to be below the reporting limit for the analytical procedure and were therefore not included in the estimates of the summary statistics.

Morning driving periods are characterised by more variable driving conditions, such as idling, accelerating and decelerating. These 'forced-flow' driving patterns influence the concentrations of VOCs measured as most fuel-derived compounds are emitted at high levels at very low speeds. Furthermore, congested traffic conditions decrease the dispersion effects of wind for vehicle emissions. Pollutants are therefore kept closer to the ground in and around the vehicles, and this is reflected in the higher concentrations observed during these driving conditions. Afternoon traffic periods, on the other hand, are generally high speed, free-flowing, with lower inbound traffic volumes than the morning period, and fewer bottlenecks in the outbound traffic.

Traffic volumes for the 7–8 am and 8–9 am periods were very similar, indicating that traffic volumes between the two sampling times would be unlikely to be a significant factor in any difference in VOCs (EPA Victoria 1998b). The mean benzene concentration for the morning period from both sites was 3.2±0.5 ppb, with a range from less than 0.2 to 17.2 ppb.

Table 4.5: Comparison of 1-hour mean results of morning and afternoon peak periods
Compound Mean and standard deviation (ppb) Difference
Morning Afternoon
Freon 12 (dichlorodifluoromethane) 0.7±0.2 0.7±0.2 0%
Chloromethane 0.6±0.1 0.6±0.1 0%
Freon 11 (trichlorofluoromethane) 0.3±0.1 0.2±0.1 50%
Benzene 3.2±0.5 2.4±0.4 25%
Toluene 7.7±1.3 5.4±0.9 30%
Ethylbenzene 1.2±0.2 0.5±0.1 58%
1,3- and 1,4-dimethylbenzene 3.2±0.7 2.0±0.4 39%
1,2-dimethylbenzene 1.0±0.2 0.6±0.1 39%

Note: ppb = parts per billion

Source: EPA Victoria (1998b).

In the afternoon sampling periods, the mean concentration levels of aromatic hydrocarbons were higher during the 5–6 pm period than the 4–5 pm period. However, this difference can be attributed to a 13% increase in traffic volume in the 5–6 pm period (EPA Victoria 1998b). Comparisons can be made when the morning and afternoon 1-hour concentrations are averaged. This reveals that the mean concentrations of aromatic hydrocarbons for the morning period are significantly higher than for the afternoon period (Table 4.5).

Predominantly stable atmospheric conditions and forced-flow traffic conditions are two factors accounting for the higher levels of VOCs experienced during the morning period, in comparison with the afternoon period. The shallower mixing depth of the atmosphere in the mornings results in reduced dispersion of pollutants. In addition, the characteristic traffic conditions in the morning described above increase the level of vehicle emissions and decrease the space between vehicles for dispersion.

The 24-hour sample can be used as an indicator of the average daily exposure of people living in close proximity to the freeway. Table 4.6 shows the 24-hour sampling results. Lower concentrations of nonchlorinated aromatic hydrocarbons were recorded in the 24-hour samples than in either the morning or afternoon peak driving hours. The 24-hour benzene concentration was 1.7±0.3 ppb, which is 46% lower than during the morning peak period (EPA Victoria 1998). The 24-hour toluene concentration was 4.4±0.7 ppb, and that of 1,4-dimethylbenzene was 1.3±0.3, which are 43% and 59% lower than the morning peak period, respectively. This occurs because the longer averaging period takes into account the low traffic flows experienced between 9 pm and 5 am and eliminates the effects of peak traffic periods.

Table 4.6: 24-hour sample period summary (8 am–8 am)
Compound Frequency of detectiona Maximum (ppb) Minimum (ppb) Mean and standard deviation (ppb) 24-hour WHOb level limit/ national guideline values (ppb)
Freon 12 (dichlorodifluoromethane) 85 4.9 <0.2 0.8±0.2 Not available
Chloromethane 87 1.1 0.5 0.6±0.1 Not available
Freon 11 (trichlorofluoromethane) 82 1.0 <0.2 0.2±0.1 Not available
Benzene 87 4.5 0.2 1.7±0.3 5c
Toluene 87 19.3 0.7 4.4±0.7 2089d
Ethylbenzene 52 1.8 <0.2 0.4±0.1 Not available
1,3- and 1,4-dimethylbenzene 65 7.6 <0.2 1.3±0.3 10–100e
1,2-dimethylbenzene 45 2.2 <0.2 0.4±0.1 10–100e

a Number of samples when the compound was above the detection limit of the analyser out of a total of 87.
b World Health Organization.
c U.K. annual limit.
d WHO 24-hour limit.
e Swedish guideline
ppb = parts per billion

Source: EPA Victoria (1998b).

The contribution of freeway emissions to the total downwind pollution was estimated by recording simultaneous measurements of air quality at the sampling sites on either side of the road. Pollutant levels measured at the upwind site were subtracted from the pollutant levels measured at the downwind site, when winds were perpendicular to the freeway. For morning peak periods during north winds greater than 2 m/second, it was estimated that 64% of downwind benzene levels, and 60% of downwind toluene levels, could be accounted for by freeway traffic (EPA Victoria 1998). However, this was highly variable because of the changing effects of wind speed and direction over the morning periods, and forced-flow driving conditions. During the afternoon periods, the average contribution of benzene to downwind levels was 75%; for downwind toluene levels it was 60%. Contribution levels were more variable in the morning periods; this can be explained largely by the generally more stable meteorological conditions and the more variable driving modes experienced at those times. Similarly, a study of different transport modes (Torre and Bardsley 1998) found that total VOC concentrations were highest on days when low wind speeds prevailed. Higher wind speeds are observed to be more effective in dispersing vehicle emissions than low wind speeds. Low wind speeds tend to dominate early morning meteorological conditions, so higher concentrations of vehicle emissions are observed during those periods.

VOCs measured for different modes of transport

EPA Victoria conducted a study to measure the concentrations of VOCs to which morning commuters are exposed when travelling (Torre and Bardsley 1998). Four different modes of transport were used to measure exposure to commuters – driving, bicycling, walking and tram travelling – and compared with a fixed roadside site. Table 4.7 contains the results of the mean VOCs measured for each mode along with the fixed roadside site. Figure 4.4 presents a graph of the mean concentrations of the major VOCs.

Table 4.7: VOC concentrations for four modes of commuting and fixed roadside monitoring
Compound Mean concentration (ppb) Compound Mean concentration (ppb)
R/side Walk Tram Bike Car R/side Walk Tram Bike Car
1,3-Butadiene 0.7 0.7 1.6 2.6 3.4 2,4-Dimethylpentane 0.1 <0.1 <0.1 0.2 0.3
Isobutane 6.9 5.9 9.4 9.6 9.7 Benzene 6.0 4.4 7.7 12.2 13.2
1-Butene 3.4 2.1 3.8 5.7 6.9 Cyclohexane 0.2 0.5 0.2 0.2 0.5
Butane 10.3 8.0 11.8 14.8 19.0 2-Methylhexane 0.7 0.7 0.8 1.1 1.5
trans-2-Butene 1.7 0.7 1.1 1.9 2.2 2,3-Dimethylpentane 0.3 0.2 0.2 0.5 0.6
cis-2-Butene 1.2 0.5 0.8 1.5 1.7 3-Methylhexane 0.6 0.7 0.9 1.4 2.0
3-Methyl-1-butene 0.2 0.1 0.2 0.2 0.3 Iso-octane 0.3 0.2 0.3 0.5 0.9
Isopentane 11.2 8.3 10.9 17.3 20.8 Heptane 0.4 1.4 0.6 0.8 1.4
1-Pentene 0.6 0.3 0.6 1.1 1.2 Methylcyclohexane 0.2 0.4 0.6 0.3 0.6
Pentane 5.8 4.5 6.1 9.5 10.6 2,3,4-Trimethylpentane <0.1 0.2 0.2 0.1 0.2
Isoprene 0.3 0.4 3.6 1.1 6.5 Toluene 8.9 7.8 9.3 20.8 22.3
trans-2-Pentene 0.9 0.5 0.8 1.3 1.6 2-Methylheptane 0.1 0.2 0.3 0.3 0.3
cis-2-Pentene 0.5 0.2 0.4 0.7 0.8 3-Methylheptane 0.2 0.2 0.4 0.3 0.4
3-Methyl-2-butene 1.1 0.6 1.0 1.6 2.0 Octane 0.2 0.2 0.2 0.2 0.3
2,2-Dimethylbutane 0.5 0.3 0.5 0.9 1.1 Ethylbenzene 0.9 0.9 1.5 2.4 2.7
Cyclopentane 0.3 0.2 0.3 0.5 0.6 p- and m-Xylene 3.8 4.1 6.7 9.8 10.7
2,3-Dimethylbutane 0.6 0.4 0.6 1.1 1.3 Styrene 0.3 0.2 0.7 0.8 0.8
2-Methylpentane 2.9 2.1 3.3 5.2 6.2 o-Xylene 1.3 1.3 2.2 3.4 4.3
3-Methylpentane 1.9 1.5 2.3 3.6 4.2 Nonane 0.1 0.1 0.2 0.1 0.7
2-Methyl-1-pentene 0.2 <0.1 <0.1 0.3 0.4 Cumene <0.1 <0.1 <0.1 <0.1 <0.1
Hexane 2.0 2.2 2.5 3.8 4.6 α - and β -Pinene <0.1 <0.1 <0.1 <0.1 <0.1
trans-2-Hexene 0.2 0.1 0.2 0.3 0.4 Propylbenzene 0.3 0.2 0.3 0.4 0.5
cis-2-Hexene 0.1 <0.1 <0.1 0.1 0.2 1,3,5-Trimethylbenzene 0.4 0.3 0.6 0.9 1.0
Methyldcylopentane 0.5 0.5 0.5 0.8 1.2 1,2,4-Trimethylbenzene 1.4 1.1 2.0 2.7 3.1

Note: ppb = parts per billion

Source: Torre and Bardsley (1998).

Results of the study showed that the top 10 VOCs detected on each mode of transport were the same, although concentrations differed between modes. The highest concentrations were recorded for driving, followed by bicycling, tram travel and walking. This pattern is to be expected, since close proximity to the source will result in higher concentrations. It was found that the commuter travelling by car or bicycle can be exposed to concentrations of VOCs up to 2.8 and 2.4 times higher than a pedestrian on the same route, and a tram traveller is exposed to levels 1.6 times higher than a pedestrian (Torre and Bardsley 1998). Compounds measured at levels below the reporting limit (0.1 ppb) are expressed as <0.1 ppb in the results. The reporting limit represents the smallest quantity of a detectable compound that can be reported. The following compounds were found to be below this limit: cumene and α- and β-pinene for all commuting modes; 2-methyl-1-pentene, cis-2-hexene and 2,4-dimethylpentane for walking and tram commuting; and 2,3,4-trimethylpentane for the roadside site.

Figure 4.4: Graph of mean concentration of major VOCs

Figure 4.4: Graph of mean concentration of major VOCs

Source: Torre and Bardsley (1998).

As mentioned earlier, total VOC concentrations were highest on low wind speed days. However, on other days when total VOC concentrations were high, increased traffic density and slower driving speeds were observed (Torre and Bardsley 1998). Increased traffic density contributes to source strength and reduces the distance between vehicles and dispersion of air around the vehicles. Similarly, lower car speeds increase VOC emissions and reduce roadway air turbulence, thus decreasing the dispersion of VOCs and increasing roadway concentrations. Therefore, although meteorological conditions have a large influence on the concentrations of vehicle emissions, traffic conditions can also have a significant impact.

Monitoring of air toxics from industry

In 1991, a brief benzene monitoring campaign was conducted at a residential site in Altona (Bardsley 1991). The site was approximately 1200 m south of the Altona chemical complex. The partially integrated petrochemical complex in Altona consists of a number of companies, the largest of which generates olefins for the production of resins, plastics and elastomers. The other companies in the complex also manufacture resins, as well as petrochemicals and vinyl chloride monomers. Benzene levels in the air were measured between 4 pm on 4 March and 7.30 am on 7 March using a portable gas chromatograph. Levels above the detection limit, 5 ppb, occurred between 12.30 am on 6 March and 2.30 am on 7 March. The concentrations during this time averaged about 8 ppb and ranged from less than 5 to 20 ppb.

During February and March 1996, EPA Victoria conducted a pilot study in which the atmospheric concentrations of a range of VOCs were measured at six sites in and around the Melbourne suburb of Dandenong (Torre et al 1996). The prime reason for choosing to conduct the study in this area was to assess the impact of the Dandenong Offensive Industries Zone (DOIZ). This 300-ha zone includes animal and food processing industries and waste treatment and incineration facilities. An additional reason for performing a study in this area was to allow comparisons with results of the NPI Dandenong study area. The six sites chosen for the VOC monitoring study were located as follows:

Time-integrated air samples were collected over 24 hours in 6 dm³ SUMMA® passivated canisters. The levels of VOCs in the air samples were then determined by GC/MS. Meteorological data obtained from the EPA Victoria air monitoring station at Dandenong showed that the prevailing winds during the sampling period were from the south to the east. The results of the study are summarised in Table 4.8.

Table 4.8: Levels of air toxics at each site in the study of the effects of the Dandenong Offensive Industries Zone (DOIZ) (arithmetic mean of all measurements)
Compound Site 1 DOIZ mean (range) (ppb) Site 2 DOIZ mean (range) (ppb) Site 3 DOIZ mean (range) (ppb) Site 4 DOIZ mean (range) (ppb) Site 5 residential mean (range) (ppb) Site 6 roadside mean (range) (ppb)
Benzene 0.6 (0.2–1.6) 0.5 (0.2–1.5) 0.7 (0.2–1.6) 0.7 (0.3–1.5) 1.1 (0.5–2.3) 1.4 (0.7–2.5)
Toluene 1.4 (0.4–3.3) 1.2 (0.3–4.3) 1.7 (0.5–3.6) 2.0 (0.8–3.6) 1.8 (0.9–3.1) 2.4 (1.2–4.8)
Ethylbenzene 0.1 (<0.2–0.2) 0.1 (<0.2–0.3) 0.2 (<0.2–0.3) 0.3 (<0.2–1.3) 0.2 (<0.2–0.2) 0.2 (<0.2–0.4)
o-Xylene 0.2 (<0.2–0.4) 0.2 (<0.2–0.5) 0.2 (<0.2–0.4) 0.5 (<0.2–0.8) 0.2 (<0.2–0.4) 0.3 (<0.2–0.6)
p- and m-Xylene 0.4 (<0.2–0.9) 0.4 (<0.2–1.2) 0.5 (0.2–1.1) 1.1 (0.2–4.5) 0.6 (0.4–1.0) 0.8 (0.4–1.7)
Styrene 0.1 (<0.2–0.3) 0.2 (<0.2–0.4) 0.2 (<0.2–0.5) 0.4 (<0.2–2.8) 0.1 (<0.2–0.3)
Carbon tetrachloride 0.1 (<0.2–0.2)
Chloromethane 0.7 (0.5–1.6) 0.6 (0.2–1.1) 0.8 (0.5–1.5) 0.7 (0.5–0.9) 0.6 (0.5–0.9) 0.6 (0.3–1.2)
Dichloromethane 0.7 (<0.2–2.1) 0.5 (<0.2–1.3) 0.7 (<0.2–1.5) 0.9 (0.3–3.6) 0.2 (<0.2–0.5) 0.4 (<0.2–0.9)
Trichloroethene 0.2 (<0.2–0.4) 0.1 (<0.2–0.4) 0.2 (<0.2–0.4) 0.2 (<0.2–0.5) 0.2 (<0.2–1.1) 0.1 (<0.2–0.3)
Freon 113 (1,1,2-trichloro-1,2,2-trifluoroethane) 0.1 (<0.2–0.2) 0.1 (<0.2–0.2) 0.1 (<0.2–0.2) 0.1 (<0.2–0.3)
Benzene/toluene 0.44 (0.21–0.75) 0.51 (0.24–0.80) 0.43 (0.29–0.67) 0.34 (0.17–0.44) 0.63 (0.46–0.82) 0.58 (0.46–0.72)

ppb = parts per billion
– = not measured

Source: Torre et al (1996).

There were no significant differences in pollutant levels between the six sites. The levels of benzene and toluene, both present in vehicle emissions, were highest at the roadside site and generally lowest at the sites in the region of the DOIZ. The benzene/toluene ratios at the roadside and residential sites were around 0.6, which is at the top end of the range typical of motor vehicle emissions. The corresponding ratios at the DOIZ sites were between about 0.3 and 0.5. This is still consistent with motor vehicle emissions but is lower than at the other two sites in the study and may indicate contributions from a source other than motor vehicles.

In general, similar concentrations of chlorinated hydrocarbons were found at all sites. The exceptions to this were dichloromethane and freon-12 (not included in Table 4.8), which occurred at slightly elevated levels at the DOIZ sites, possibly indicating an additional source of these compounds in that area.

The compounds included both in this pilot study and the NPI trial were xylene, benzene, styrene, toluene and dichloromethane. The results of the monitoring study are in the same order of magnitude but slightly higher than those obtained by modelling of estimated emissions during the NPI trial (Table 4.9) (See EPA Victoria 1999a for a comprehensive assessment of the NPI trial). It should be noted that the monitoring study involved 24-hour average measurements whereas the NPI results were annual averages.

Table 4.9: Annual average concentrations of air toxics predicted by the NPI trial for the sampling sites included in the DOIZ study
Compound Site 1 (DOIZ)
Site 2 (DOIZ)
Site 3 (DOIZ)
Site 4 (DOIZ)
Site 5 (residential)
Site 6 (roadside)
Xylenes 0.57 0.53 0.56 0.58 0.85 1.03
Benzene 0.49 0.46 0.49 0.51 0.72 0.86
Styrene 0.01 0.04 0.03 0.02 0.01 0.01
Toluene 0.73 0.68 0.72 0.75 1.07 1.29
Dichloromethane 0.24 0.25 0.28 0.38 0.19

ppb = parts per billion;
– = not assessed

Source: Torre et al (1996).

EPA Victoria conducted a study around the boundary of Huntsman Chemical Company in West Footscray during 1997. The study addressed both on-site and off-site air quality issues, but in the context of this report the most important was the canister sampling performed off-site to determine whether the chemical complex is a significant source of the target VOCs in the area. The VOCs selected for measurement were pentane, benzene, toluene, ethyl benzene, xylene, styrene and cumene. These compounds were chosen because they are used in significant quantities within the plant and because they may pose a health risk to the local community. In particular, benzene is of concern due to its carcinogenity (Bardsley 1997).

Samples were collected at all sites in evacuated 6-L SUMMA® passivated canisters at approximately 2 m above ground level (Bardsley 1997). Sites were chosen around all areas of the boundary so that no particular wind direction would bias the results. A further control site was selected away from the area so that the contribution of the chemical complex to local area pollution could be assessed.

Concentrations of styrene and cumene at the sites were targeted as the best indicators of emissions from the chemical complex, because they have low background concentrations in urban air (Bardsley 1997). The other VOCs are commonly found in urban air due to their presence in motor vehicle emissions. Table 4.10 lists the concentration range of 24-hour VOCs measured at all sites.

Table 4.10: Table 4.10 Concentration range of 24-hour target volatile organic compounds measured at all sampling sites
Site Pentane
Ethyl benzene
p and m-Xylene
Geelong Rd 0.1–16.4 0.8–8.1 0.9–8.8 0.1–0.7 0.1–2.7 0.1–0.8 <0.1–3.0 <0.1–1.1
Bunting Rd 1.5–20.9 1.3–4.9 1.9–18.3 0.2–0.7 0.1–1.9 <0.1–0.7 0.4–6.5 <0.1–3.0
Judge St 0.5–9.6 0.3–4.2 1.4–9.1 <0.1–0.9 0.1–2.8 0.1–5.0 <0.1–0.9 <0.1–1.2
Fourth Ave 0.5–3.7 0.7–3.7 2.5–9.2 0.2–0.8 0.8–3.0 0.3–0.9 <0.1–0.3 <0.1
Sunshine Rd 0.5–4.0 0.4–3.1 0.8–8.0 <0.1–0.6 0.1–2.6 <0.1–0.8 <0.1–0.8 <0.1–0.4
Indwe St 0.4–6.2 0.3–3.6 0.6–8.9 <0.1–1.0 <0.1–4.5 <0.1–1.4 <0.1–0.2 <0.1–0.1

Note: ppb = parts per billion

Source: Bardsley (1997).

The results of the 24-hour sampling generally show that the highest concentrations of the target VOCs occurred at downwind sites closest to the Huntsman complex. Concentrations of pentane, toluene, styrene and cumene were highest at Bunting Road, the closest site to the Huntsman, only 400 m south of the boundary (Bardsley 1997). The Geelong Road site recorded the highest benzene concentration for the study; however, it was later concluded that the main source was nearby motor vehicle emissions rather than Huntsman. The mean 24-hour averages of target VOCs are shown in Figure 4.5.

The low mean concentrations of all target VOCs at Sunshine Road and Fourth Avenue, the sites located north of Huntsman, are similar to those levels measured at the control site, Indwe Street. The high incidence of northerly winds during the sampling period would be expected to produce results such as these, as both sites are located upwind of the Huntsman complex. The above-background mean concentrations of styrene and cumene at the other sites indicate that the likely source is the Huntsman complex due to its proximity and the prevailing wind conditions (Bardsley 1997).

The SEPP for Air Quality Management specifies design ground level concentrations for many VOCs, based on three-minute averaging periods. The design ground level concentrations are not directly comparable with the measured concentrations of VOCs in this study as they are based on 24-hour sample periods. The measurements do, however, indicate that the highest concentrations detected during the study are likely to be below the SEPP guidelines.

Figure 4.5: Mean 24-hour concentrations of target organic compounds measured at all sampling locations

Figure 4.5: Mean 24-hour concentrations of target organic compounds measured at all sampling locations

VOC = volatile organic compound;
ppb = parts per billion

Source: Bardsley (1997).

Fleet emissions factors

EPA Victoria and CSIRO studied average fleet emission factors, measuring carbon monoxide, oxides of nitrogen, and VOC emissions from motor vehicles on arterial roads as an independent method for validating the 1995–96 EPA Victoria inventory for the Port Phillip Region (EPA Victoria 1998a). The method adopted for this study, developed by CSIRO, was to conduct sampling above the moving traffic on the roadway and in traffic so that the overall emission rate for the roadway was determined by calculating its ratio to a known total emission from the motor vehicles (EPA Victoria 1999b). This approach measures the integrated emissions from multiple vehicles using a particular roadway at the time of the sampling.

Measurements of the lower molecular weight PAHs – phenanthrene, fluoranthene and pyrene – were mutually correlated; these substances had much higher concentrations in the in-traffic samples than in the MAS (Gras et al 1992). A comparison of the results of the EPA/CSIRO study with the MAS indicates that the source of these species is primarily vehicular (EPA Victoria 1999b) (see Section 4.1.2 for a review of this study). Miguel et al (1998) observed that aerosol PAH concentrations in vehicular tunnels indicated that emission factors for fluoranthene, pyrene and benzo[a]anthracene from heavy-duty vehicles were, respectively, 60, 75 and 30 times those from light-duty vehicles, and concluded that diesel trucks were the main source of low molecular weight PAHs. This is consistent with the findings of the EPA/CSIRO study. Furthermore, the estimated emission factors for formaldehyde, acetaldehyde and butyraldehyde as derived from the filter samples for in-traffic and background samples also suggest that heavy-duty vehicles and light-duty vehicles with carburettors are the major contributors to carbonyl emissions in the on-road vehicle fleet in Melbourne (EPA Victoria 1999b). Table 4.11 presents the emission factor estimates for the three lower molecular weight PAHs identified as having a vehicular origin, along with results from Miguel et al (1998).

Table 4.11: Mean emission rates (ng/m³) for polycyclic aromatic hydrocarbon species in the aerosol phase for the EPA/CSIRO and Miguel et al studies
  CSIRO/EPA studya Miguel studyb
Emission rate Standard error Emission rate Emission rate
Aerosol phenanthrene 2.6 0.4
Aerosol fluoranthene 6.4 0.8 0.7 190
Aerosol pyrene 8.6 1.3 0.8 270

LDV = light-duty vehicle;
HDV = heavy-duty vehicle

a EPA Victoria (1999b).
b Miguel et al (1998)

Measurements undertaken to support the development of the 1995–96 Port Phillip region inventory may be compared with the findings of two earlier studies. In 1990, in developing the Port Phillip region emissions inventory, over 1200 inservice motor vehicles covering pre-1976 and post-1985 model years were measured to calculate past and future fleet average emission rates on arterial roads in Melbourne. An earlier study (1983–84) examined air toxics in the Melbourne CBD during morning peak hour traffic just before the introduction of vehicle emissions controls, including unleaded petrol and catalytic converters on new motor vehicles in 1986. The aim of the study was to test the effectiveness of the introduction of ADR 37.

Samples of air were collected in 1000 dm³ fluorinated ethylene polymer-teflon bags over 20 minutes while following a standard route through the CBD in peak traffic. The levels of C2-C10 hydrocarbons in the air samples were determined after cryogenic preconcentration on a Tenax trap by dual-column gas chromatography separation and flame ionisation detection (Evans et al 1985). Further measurements of vehicle exhaust pollutants in morning peak periods were conducted in 1990 (Galbally et al 1995), to determine the change in hydrocarbon concentration and composition of Melbourne's air after the introduction of ADR 37. The results of all of these studies are compiled in Table 4.12 (EPA Victoria 1999b).

It is evident from Table 4.12 that there has been a notable decline in the atmospheric concentrations of the individual aromatic compounds over the period. Total concentrations were observed to decline by 69%, which is in line with the change in car exhaust emission rates over this time. The relatively constant level of hydrocarbon composition in petrol over the period, in addition to the removal of aromatics in the engine exhaust by catalytic converters, has contributed to these reductions.

Table 4.12: Average concentrations (ppbv) and standard errors of the mean for some air toxic compounds measured in-traffic in the CBD of Melbourne
  1983–84 1990 1997 % Δ from 1983–84 to 1997
1,3-Butadiene NA NA 1.3 ± 0.1 NA
Benzene 21.8 ± 1.0 14.9 ± 0.9 7.9 ± 0.6 64%
Toluene 39.5 ± 1.8 20.2 ± 1.3 13.0 ± 0.9 67%
Ethylbenzene 5.5 ± 0.3 2.6 ± 0.2 1.5 ± 0.1 73%
m- and p-Xylene 22.0 ± 1.0 10.4 ± 0.7 6.0 ± 0.4 73%
o-Xylene 8.4 ± 0.4 4.0 ± 0.3 2.2 ± 0.2 74%
Total       69%

ppbv = parts per billon by volume;
CBD = central business district;
NA = not available

Source: EPA Victoria (1999b).

Evaluation of DOAS for ambient air monitoring

In 1995, EPA Victoria conducted a study to evaluate the performance of the DOAS (Opsis) open-path monitoring system and to provide a preliminary assessment of a range of air pollutants (Bardsley 1996). Two sites were selected for the study, one at a major industrial chemical site in Altona and the second at a heavily trafficked roadside location in Essendon. Monitoring was conducted from January to May 1995 at the Altona site, and from June to July 1995 at the Essendon site. However, instrumental problems in the first two to three months at the Altona site resulted in significant data loss; therefore the results presented here are for the period from March to May 1995.

Air toxics concentrations

The levels of measured pollutants at both sites were generally low and were well within the objectives for design ground-level concentrations as specified in the SEPP (AQM).

Benzene concentrations were generally low throughout the monitoring period, with an average of 2.7 ppb at the Altona site. Generally, benzene measurements at Altona were in the range of 1–2 ppb (56% of all measurements). These results were very similar to those recorded for Essendon, where the average was 2.2 ppb, with only two hourly periods exceeding an arbitrary value of 10 ppb. Of all measurements taken at Essendon, 44% were in the range of 2–2.5 ppb.

Hourly benzene averages at Altona exceeded an arbitrary value of 10 ppb on 14 days out of the three months of monitoring, while at Essendon only two hourly periods exceeded 10 ppb.

Table 4.13 shows the benzene maximum daily hourly events that exceeded 10 ppb. At Altona, peak concentrations of benzene did not appear to be related to any particular time events (such as peak hour traffic). Rather, they most often occurred during periods with southwesterly winds and were characterised by rapidly changing concentrations over relatively short periods. This pattern indicates that the source was nearby so that there was insufficient time for the pollutant to mix effectively (Bardsley 1996). Emissions from distant sources tend to be better dispersed, producing smoother broader peaks. This was further confirmed by the high benzene/toluene ratios observed during these events, identifying the Altona petrochemical complex as the most likely source of the emissions.

Table 4.13: Benzene levels at Altona and Essendon: maximum daily hourly events >10 ppb
Site Date Time Max hourly (ppb) Wind direction Benzene/ toluene
Altona 12/03/95 1700 16 SW ND
13/03/95 2300 17 SW ND
21/03/95 2200 12 SW ND
22/03/95 1900 15 SW ND
24/03/95 1900 10 SW ND
29/03/95 2100 17 SW ND
5/04/95 1000 15 SW ND
6/04/95 1900 30 SW ND
26/04/95 1900 12 SW 3.4
28/04/95 0600 14 NNE 1.4
1/05/95 1100 14 SW 2.7
5/05/95 0700 12 SW 3.9
16/05/95 0600 12 SW 1.2
19/05/95 0900 15 N 0.4
Essendon 22/6/95 0900 11 NE 0.3
30/7/95 0000 11 NNW 20

Note: ND = no data

Source: Bardsley (1996).

The toluene average at Altona was 2.2 ppb for April and 5.7 ppb for May, while at Essendon it averaged 4.2 ppb over the period from June to July 1995. Levels of toluene exceeded an arbitrary value of 20 ppb on four occasions at Altona and five at Essendon. At Altona these peaks were observed during northeasterly wind periods. The benzene/toluene ratio for the events at Altona suggest that motor vehicles are likely to be the main source of toluene emissions (Bardsley 1996). Table 4.14 shows the toluene maximum daily hourly events that exceeded 20 ppb.

Table 4.14: Toluene levels at Altona and Essendon: maximum daily hourly events >20 ppb
Site Date Time Max hourly (ppb) Wind direction Benzene/ toluene
Altona 30/04/95 1900 21 NE 0.3
8/05/95 0300 23 NE 0.2
9/05/95 1900 34 NE 0.2
19/05/95 1900 40 NE 0.3
Essendon 7/7/95 1900 27 S 0.3
12/7/95 1700 21 S 0.3
20/7/95 1700 21 S 0.4
22/7/95 2200 26 ENE 0.2
23/7/95 0000 35 SSE 0.1

Note: ppb = parts per billion

Source: Bardsley (1996).

Para-xylene and styrene levels were low, with no significant events recorded during the monitoring period.

Opsis evaluation

A number of problems with the Opsis equipment resulted in significant data loss for the first two to three months of operation. Most of these problems were resolved. However, measurements recorded before the corrections were made had to be omitted from the data. The problems were:

Once the problems were resolved, the system provided some significant advantages over conventional forms of air monitoring (point analysers), such as:

Comparison of DOAS and canister sampling techniques

A study of VOC levels by means of a discrete spot sampling technique was undertaken simultaneously with the DOAS (Opsis) monitoring program (Torre 1995). Time-integrated air samples were collected in SUMMA® passivated canisters over 12 hours. Gas chromatographic separation and mass spectroscopic analysis determined the concentrations of a range of VOCs in these samples. At Altona, canister sampling was performed at four sites, one at each end of the equipment's light beam, one about 1 km north and one about 2 km south of the complex. The samples were taken over 12-hour periods beginning between 8 am and 3 pm. At the Essendon site, canister sampling was performed at each end of the equipment's light beam. The samples were also taken over 12 hours, beginning between 1 pm and 4 pm. The results obtained by this canister sampling program for air toxics are summarised in Table 4.15.

Table 4.15: Summary of data obtained by canister sampling, GC/MS separation and analysis, at Altona and Essendon
Compound Altona, chemical complex Essendon, Tullamarine Freeway
mean (range) (ppb)
mean (range) (ppb)
mean (range) (ppb)
mean (range) (ppb)
mean (range) (ppb)
mean(range) (ppb)
Number of samples 19 19 10 8 11 3
Benzene 0.7 (<0.2–1.2) 0.7 (<0.2– 2.5) 0.8 (<0.2–1.6) 0.7 (0.3–1.6) 2.6 (0.8–4.3) 3.9 (3.8–4)
Toluene 2.0 (1.0–5.0) 1.8 (0.6–6.8) 3.3 (0.7–8.9) 2.3 (0.6–6.9) 3.1 (1.2–5.8) 4.8 (3.9–5.8)
Styrene 0.5 (<0.2–1.4) 0.7 <0.2–2.0) 0.3 (<0.2–0.6) 0.3 (<0.2–0.4) 0.3 (0.1–0.6) 0.4 (0.4–0.5)
Ethylbenzene 0.3 (<0.2–0.7) 0.3 (<0.2–0.7) 0.4 (<0.2–0.8) 0.4 (<0.2–0.9) 0.7 (0.2–4.2) 0.7 (0.7–0.8)
m-and p-Xylene 1.0 (0.2–2.5) 0.8 (0.2–2.6) 1.1 (0.2–2.8) 1.2 (0.2–3.2) 1.7 (0.6–3.5) 2.8 (2.5–3)
o-Xylene 0.3 (<0.2–0.7) 0.3 (<0.2–0.9) 0.5 (<0.2–0.9) 0.4 (<0.2–1.1) 0.6 (<0.2–1.2) 1.0 (0.9–1)
Benzene/toluene 0.8 (0.1–1.5) 0.4 (0.1–0.9) 0.3 (0.1–0.5) 0.3 (0.2–0.4) 1.0 (0.4–3.3) 0.8 (0.7–1.0)

Air samples were collected over 12-hour periods
ppb = parts per billion

Source: Torre (1995).

Using the canister sampling method, higher average concentrations of these VOCs were measured at the Essendon site than at the Altona site. This was particularly noticeable for benzene and toluene, which are present in high amounts in vehicle exhausts. The highest levels of these pollutants at the Essendon site were observed during light winds and are attributed to build-up of vehicle emissions from the Tullamarine Freeway.

In order to compare the results obtained using the Opsis monitoring method and the canister sampling method, the Opsis results were averaged over the same 12-hour periods during which canister samples were collected. The 12-hour averages of the Opsis results for benzene and toluene are shown, together with the corresponding canister sampling results, in Figures 4.6 and 4.7.

Figure 4.6: Comparison of results obtained for benzene at Altona by canister sampling at four sites (one at each end of the Opsis light beam (light and receiver ends), one 1 km north and one 2 km south of the Opsis) with those obtained by the Opsis averaged over the same 12-hour period.

Figure 4.6: Comparison of results obtained for benzene at Altona by canister sampling at four sites (one at each end of the Opsis light beam (light and receiver ends), one 1 km north and one 2 km south of the Opsis) with those obtained by the Opsis averaged over the same 12-hour period.

Source: Torre (1995).

Figure 4.7: Comparison of results obtained for toluene at Altona by canister sampling at four sites (one at each end of the Opsis light beam (light and receiver ends), one 1 km north and one 2 km south of the Opsis) with those obtained by the Opsis averaged over the same 12-hour period.

Figure 4.7: Comparison of results obtained for toluene at Altona by canister sampling at four sites (one at each end of the Opsis light beam (light and receiver ends), one 1 km north and one 2 km south of the Opsis) with those obtained by the Opsis averaged over the same 12-hour period.

Source: Torre (1995).

This comparison shows that, although the two measurement techniques displayed similar trends, the results obtained using the Opsis system generally were two to three times higher than those from the canister sampling and gas chromatograph analysis. The reasons for these differences are unknown. A similar comparison was not feasible for the results obtained at the Essendon site due to the small amount of Opsis data retrieved during canister sampling periods.

4.1.3 Victorian emissions inventory

An emissions inventory of the Port Phillip Region, spanning some 24 000 km², was undertaken for the 1995–96 financial year. Emissions of 33 pollutants were estimated for the inventory, including a number of air toxics. A list of the air toxics targeted in the inventory is included in Table 4.16. Table 4.17 shows a summary of the emissions from the sources discussed.

Table 4.16: The hazardous air pollutants targeted in the emissions inventory of Port Phillip Control Region in 1995–96
Primary pollutants Hazardous organic air pollutants
Lead Acrylonitrile
Total volatile organic compounds, excluding methane Benzene
Hazardous inorganic air pollutants 1,4-Dichlorobenzene
Ammonia Dichloromethane
Arsenic and compounds Dioxins and furans
Cadmium and compounds Diphenyl methane di-isocyanate
Chromium and compounds Formaldehyde
Manganese and compounds Methyl ethyl ketone
Mercury and compounds Methyl isobutyl ketone
Nickel and compounds PAHs
Fluorides Styrene
  Toluene 4,4-di-isocyanate
  Vinyl chloride
Industrial sources

Table 4.17 shows the industrial sources of air toxics. Figure 4.8 shows annual total emissions from industry sources for substances where industry is the largest source of the pollutant.

Contributions by industry type

The industry types with the highest levels of emissions of individual air toxics are:

Annual emission levels for all industry types and air toxics are listed in Table 4.18.

Table 4.17: Emissions contributions from all sources (tonnes per year)
Pollutant Motor vehicles Other mobile sources Industry Domestic/commercial/rural Total
1,3-Butadiene 450 11 86 32 580
Acrylonitrile     15 1.3 16
Ammonia     92 10 108 10 200
Arsenic and compounds   0.16 0.48 0.049 0.69
Benzene 3539 89 148 675 4451
Cadmium and compounds   0.016 0.43 0.11 0.55
Chromium and compounds 0.43 0.14 2.5 0.17 3.2
Dichloromethane     461 36 497
Dioxins     2.6E-09   2.55E-09
Diphenyl methane di-isocyanate     0.023   0.023
Fluoride compounds     493 0.47 494
Formaldehyde 823   48 409 1280
Lead and compounds 184 0.52 3.4 1.1 189
Manganese and compounds 1.110 0.002 1.134 0.58 2.8
Mercury and compounds   0.16 0.033 0.015 0.21
Methyl ethyl ketone     652 151 804
Methyl isobutyl ketone     123 3 126
Polyaromatic hydrocarbons 114 5.6 0.26 149 269
Styrene   1.6 68 5.6 75
Tetrachloroethylene     19 608 626
Toluene 5958 146 2403 1888 10 394
Toluene 2,4-di-isocyanate     0.01   0.01
Trichloroethylene     780 1.5 782
Vinyl chloride     10 1.5 11
Volatile organic compounds 63 315 2153 33 442 69 531 168 441
Xylenes 4369 152 1987 1106 7613

Note: Empty cells = no data

Source: EPA Victoria (1998a).

Figure 4.8: Air toxics for which industry is the largest contributor in the Port Phillip Region.

Figure 4.8: Air toxics for which industry is the largest contributor in the Port Phillip Region.

Note: *2.6E-09 = 2.6 x 10-9 tonnes per year = 0.0026 grams per year

Table 4.18: Emissions of air toxics by industry type (tonnes/year)
Pollutant Food, beverage and tobacco m'facturing Textiles, clothing, footwear and leather m'facturing Wood and paper product m'facturing Printing, publishing and recorded media Petroleum, coal, chemical and associated product m'facturing Nonmetallic mineral product m'facturing Metal product m'facturing Machinery and equipment m'facturing Electricity and gas supply Water supply, sewerage and drainage services Storage Health services Waste disposal Unclassified Total
1,3-Butadiene 0.0464 0.0647 0.0528 0.0107 85.8 0.109 0.148 0.174 0.00196   0.00446 1.95E-05 9.28E-04   86.4
Acrylonitrile         6.99       7.36   0.331       14.7
Ammonia                   92.0         92.0
Arsenic and compounds 0.0896 0.0500 4.86E-04 1.44E-05 0.174 0.0739 0.0258 0.0362 0.00197 1.22E-05 9.73E-05 0.0133 0.00288 0.00682 0.475
Benzene 0.354 0.486 0.395 0.0802 99.2 0.857 1.11 1.31 2.59 0.0221 42 0.00595 0.00698 0.0261 148
Cadmium and compounds 0.00381 0.00212   6.52E-09 0.170 0.125 0.0943 0.0265     2.68E-05 5.68E-04 0.00769   0.430
Carbon monoxide 151 319 135 23.0 1590 1396 3194 335 1492 199 6.90 10.3 1.33 1025 9879
Chromium and compounds 0.166 0.0923 0.00232 6.88E-05 0.340 0.297 0.472 0.981 0.00944 5.86E-05 1.20E-04 0.0245 0.0631 0.03259 2.48
Dichloromethane 0.00166 39.5     279 84.0   36.0 22.3     0.00124     461
Dioxins         1.36E-10 2.42E-09                 2.55E-09
Diphenyl methane di-isocyanate   0.0180     0.00440     1.41E-04             0.0225
Fluoride compounds         7.98 356 129.5 0.00533         0.00300   493
Formaldehyde 0.370 1.34 1.20 0.0165 7.20 4.19 26.4 0.395 1.54 0.0372 0.0484 0.0181 0.0327 5.18 47.9
Lead and compounds 5.53E-04 1.89E-04 7.81E-04 3.76E-05 0.808 0.819 0.939 0.602 0.00238 1.44E-05 0.038 2.24E-05 0.180 0.00804 3.40
Manganese and compounds 7.40E-04 2.51E-04 8.05E-04 2.38E-05 0.0422 0.310 0.484 0.280 0.00328 2.03E-05 2.13E-04 3.00E-05 4.67E-04 0.01130 1.13
Mercury and compounds       2.43E-09 8.03E-04 0.0114 5.21E-05 7.87E-06     7.78E-06 9.65E-08 0.0143 0.00671 0.0333
Methyl ethyl ketone 0.00224 14.0   184 151 1.64 7.37 284 8.10   2.17 0.00166     652
Methyl isobutyl ketone       1.40 10.8   15.7 92.6 2.24   0.0172       123
Nickel and compounds 0.107 0.0587 0.00763 2.26E-04 0.197 0.302 0.638 0.337 0.0310 1.92E-04 0.00555 0.0153 0.0145 0.1071 1.82
Oxides of nitrogen 320 183 280 17.3 4359 5567 942 293 3012 481 10.7 23.9 11.3 4021 19522
Particles (PM10) 42.9 38.1 23.7 1.06 767 5930 120 144 29.6 8.05 4.10 7.50 10.6 403.3 7529
Particles (PM2.5) 27.4 25.1 20.2 0.863 467 586 74 91.7 25.2 1.81 2.87 5.04 5.29 342.8 1675
Polyaromatic hydrocarbons 0.00590 0.00146 0.00760 3.12E-04 0.0609 0.0723 0.0696 0.00977 0.00449 0.00399 0.00120 0.00128 0.0126 0.00886 0.260
Styrene 0.00645 2.24 0.00718 0.00145 60.0 4.34 0.0202 0.0238 2.66E-04   1.20 1.09E-04 1.26E-04   67.9
Sulfur dioxide 71.8 71.3 3.25 0.228 6448 556 3629 15.0 18.0 3.85 4.58 3.69 0.587 15.46 10841
Tetrachloroethylene 2.46E-04 2.77     3.89     6.24 5.47   0.181 1.83E-04     18.6
Toluene 0.526 5.48 0.598 43.9 363 5.79 244 1626 53.6 0.00981 60 0.00147 0.0114 0.0735 2403
Toluene 2,4-di-isocyanate         0.0102           1.40E-04       0.0104
Trichloroethylene   0.412       3.23 24.0 750 2.57           780
Vinyl chloride         1.27     0.0345 8.45           9.75
VOCs 751 726 49.4 4058 13897 349 1276 8704 1437 12.7 2056 0.491 0.740 125.9 33442
Xylenes 0.510 5.68 0.609 0.117 155 1.52 125 1639 23.9 0.00675 35.5 3.95E-04 0.0102   1987

VOCs = volatile organic compounds;
empty cells indicate no emissions

Mobile sources

Emissions from mobile sources, in particular motor vehicles, are the largest contributing sources for a number of air toxics in the atmosphere. Most notably, 97% of annual lead emissions and 80% of benzene emissions are from motor vehicles. The highest levels of air toxics from mobile sources are illustrated in Figure 4.9.

Contributions of motor vehicle emissions

The emission factors for motor vehicles were estimated for three road categories to account for differences in traffic flow, vehicle fleet composition (age and type), and fuel type.

Figure 4.9: Highest concentrations of air toxics from motor vehicle sources (g/km)

Figure 4.9: Highest concentrations of air toxics from motor vehicle sources (g/km)

As indicated in Section 4.2.1, emissions of all major air toxics from motor vehicles has decreased since the introduction of ADR 37/00. This has resulted in an overall reduction in total emissions from the motor vehicle fleet, despite a 16% increase in vehicle kilometres travelled between 1990 (the year of the previous emissions inventory for the Port Phillip Region) and 1995–96. This can be attributed to the ADR 37 specification requiring new cars to be fitted with a catalytic converter to reduce the emission rate of gaseous pollutants. In addition, the turnover of the Australian fleet has resulted in fewer pre-1986 vehicles on-road, thus reducing the use of leaded petrol. Figure 4.10 illustrates the extent of the reduction in emissions from motor vehicles between the two inventory years.

Figure 4.10: Total motor vehicle emissions for 1990 and 1995-96

Figure 4.10: Total motor vehicle emissions for 1990 and 1995–96

Note: VOCs = volatile organic compounds

Contributions of other mobile sources

Although motor vehicles are the major contributor to emissions from mobile sources, other mobile sources, in total, also make a significant contribution. These sources include locomotives, aircraft, commercial boats and ships, and pleasure craft. In brief, other mobile sources were found to contribute at least 5% of total emissions of three heavy metal air toxics: arsenic, mercury and nickel. Commercial ships emitted 0.16 t/yr of mercury and 2.6 t/yr of nickel, while aircraft were responsible for 0.13 t/yr of arsenic emissions for the source group. Furthermore, pleasure craft emitted 0.26 t/yr of lead, as well as 870 t/yr of VOCs.

Domestic/commercial/rural sources

This source group consists of domestic sources such as lawn mowing, solid fuel combustion, waste incineration and barbecues. Commercial sources include both private and public sector commercial activities; rural sources include agriculture, forest fires, fertiliser and crops, and livestock.

This source group generally emits relatively small amounts of most of the air toxics targeted in the inventory, in comparison to industry and motor vehicle sources. However, it is the major contributing source group for ammonia, PAHs, tetrachloroethylene and total VOCs. Figure 4.11 illustrates the highest concentrations of emissions of air toxics from the source group.

The source types in this group with the largest emissions of air toxics are:

Figure 4.11: Highest concentration of air toxics recorded from domestic, commercial and rural sources

Figure 4.11: Highest concentration of air toxics recorded from domestic, commercial and rural sources

Seasonal differences in emission levels were most evident from domestic and commercial fuel combustion, lawn mowing, barbecues, use of pleasure craft and natural gas leakage. This is due to closures of industrial plants over the Christmas–New Year period, and lower human outdoor activities over winter months.

Inventory conclusions

The 1995–96 inventory confirmed that motor vehicles are a significant source of the common criteria pollutants, as well as total VOCs and lead. Although VOC emissions are highest from domestic sources, they are concentrated over the winter months when solid fuel combustion (heating) is highest and photochemical activity required for photochemical smog formation is low. Motor vehicles are also responsible for the highest emissions of benzene, 1,3-butadiene, formaldehyde, toluene and xylenes, while industry sources emit the highest concentrations of most of the other air toxics targeted in the inventory.


   Links to another web site
   Opens a pop-up window