3 Atmosphere | 3 Ambient air quality and other atmospheric issues | 3.1 State and trends of Australia's atmosphere

State of the Environment 2011 Committee. Australia state of the environment 2011.
Independent report to the Australian Government Minister for Sustainability, Environment, Water, Population and Communities.
Canberra: DSEWPaC, 2011.

3 Atmosphere

3.1 State and trends of Australia’s atmosphere

Assessing the state of Australia’s atmosphere in essence involves assessing the impact of a number of contaminants on three main areas: the stratospheric ozone layer, ambient (outdoor) air and indoor air.

3.1.1 Stratospheric ozone

The stratosphere is the layer of the atmosphere that begins at an altitude of around 10 kilometres above Earth’s surface and extends to approximately 50 kilometres. It is situated between the troposphere (near Earth’s surface) and the mesosphere.97 Stratospheric ozone limits the amount of harmful ultraviolet B (UVB) light (UVB wavelengths are 280–315 nanometres) passing through to lower layers of the atmosphere. The ozone layer, therefore, has a vital role in protecting life on Earth, as increased levels of UVB may result in damage to a range of biological systems, including human health. In humans, UVB—although necessary for the production of vitamin B—causes nonmelanoma skin cancer and is a significant factor in the development of malignant melanoma. In addition, it is associated with the development of cataracts.98-100 (However, it should be noted that, whereas ozone in the stratosphere is protective of human health, ozone near the ground, where it can be breathed in, is a pollutant and harmful to health. Section 3.1.2 further discusses ozone as a pollutant.)

Photosynthesis in many species of plants is impaired by UVB radiation, and overexposure can reduce yield and quality in some crop species, including varieties of rice, winter wheat, soybeans, corn and cotton. UVB radiation may also change the susceptibility of plants to insect and pathogen attack. In aquatic systems, photosynthesis in phytoplankton is more sensitive to UVB than in terrestrial plants, and short-term exposure to increased UVB levels can reduce productivity in such systems.101-103

The ozone layer was threatened by human-produced ozone depleting substances (ODSs), principally chlorofluorocarbons (CFCs) and halons, which were widely used in refrigerators, air conditioners, fire extinguishers and electronic equipment, as solvents for cleaning (including dry cleaning) and as agricultural fumigants. These substances are stable and long lived in the lower atmosphere, but slowly drift up to the stratosphere, where they are subject to breakdown through the action of UV radiation. This releases highly reactive molecules (chlorine and bromine) that react with ozone molecules and break them apart.

Since peaking in the mid-1990s, levels of stratospheric chlorine and bromine from CFCs and other ODSs have declined. The latest World Meteorological Organization (WMO) Scientific assessment of ozone depletion104 concludes that:

… the atmospheric abundances of nearly all major ODSs that were initially controlled [under the Montreal Protocol] are declining [Figure 3.19]. Nevertheless, ozone depletion will continue for many more decades because several key ODSs last a long time in the atmosphere after emissions end.

This has important implications for climate, since all ODSs (except methyl bromide) are powerful GHGs, and the gradual recovery of the ozone layer is expected to interact with climate change through a complex series of linkages. These relationships may, for example, reduce the capacity of the oceans to absorb carbon dioxide and delay the recovery of stratospheric ozone.105

Figure 3.19

CFC = chlorofluorocarbon; HCFC = hydrochlorofluorocarbon; ppb = parts per billion

Source: Krummel & Fraser,106 updated by P Krummel, Centre for Australian Weather and Climate Research, and Commonwealth Scientific and Industrial Research Organisation, unpublished data

Figure 3.19 Effect of the Montreal Protocol on levels of ozone depleting substances in the atmosphere

The ozone hole

The impact of ODSs on the stratospheric ozone layer has been observed at all latitudes, except in the tropics (i.e. 20°N and 20°S), where ozone depletion is negligible.104 However, by far the most pronounced ozone losses are associated with the Antarctic ozone hole, which occurs each year over Antarctica between August and December. The ozone hole reaches its maximum extent in spring, when 60% of the total ozone in the vertical air column is lost. The depleted ozone layer then breaks up and disperses over the areas surrounding Antarctica during the summer and autumn months (see also Chapter 7: Antarctic environment). The break-up of the ozone hole during summer is the cause of reductions in stratospheric ozone in the Southern Hemisphere, as parcels of ozone-depleted polar air move north and mix with mid-latitude air.107-109 There is also increased evidence that the Antarctic ozone hole affects the Southern Hemisphere’s climate, acting as the driver of changes in pressure, surface winds and rainfall at mid-to-high latitudes during summer.104

Images available from Environment Canada illustrate the break-up of the ozone hole and the dispersal of the depleted ozone layer across Tasmania and southern Australia (Figure 3.20).

Figure 3.20

Source: Environment Canada110

Figure 3.20 Example of the dispersal of the depleted ozone layer over areas surrounding Antarctica, 11 September 2006

This map represents total ozone deviations from the 1978–88 level estimated using total ozone mapping spectrometer data for all areas, except the Antarctic, and from the pre-1980 level estimated using Dobson data over the Antarctic. Although the map shows some reduction in ozone levels over parts of Australia relative to the base period, in absolute terms ozone levels would still be very high south of Australia.

Following a period of rapid growth from the late 1970s to the mid-1990s, the area of the ozone hole has remained relatively stable over the past 15 or so years (Figure 3.21), with October mean column ozone levels within the polar stratospheric vortex approximately 40% of 1980 values.104 The ozone holes of 2000 and 2006 were the most severe on record; the 2006 hole was the deepest and the 2000 hole the largest (in area). However, the hole can fluctuate markedly from year to year, with 2010 being one of the smallest on record in the past two decades.111-112

Figure 3.21

km2 = square kilometre

Source: Tully et al.,111 updated by P Krummel, Centre for Australian Weather and Climate Research, and Commonwealth Scientific and Industrial Research Organisation, unpublished data

Figure 3.21 Maximum 15-day average ozone hole area (million km2), 1979–2010

The figure is based on Total Ozone Mapping Spectrometer/Ozone Monitoring Instrument data. The error bars represent the range of the ozone hole area in the 15-day average window.

The relative stability of the ozone hole reflects the fact that there have been only moderate decreases in stratospheric chlorine and bromine in the past few years. Since around 1997, ODS levels have been nearly constant, and the depth and magnitude of the ozone hole have been controlled by variations in temperature and climate dynamics. Although summer ozone levels over Antarctica have yet to show any statistically significant increasing trend, recent simulations of the effect of reductions in ODSs that are projected to continue to flow from controls under the Montreal Protocol indicate a return to pre-1980 benchmark values late this century.104,113 Modelling results suggest that this recovery may be accelerated by climate change in the form of stratospheric cooling, linked to increases in GHGs.104

Ozone hole impacts

As noted above, the most recent WMO Scientific assessment of ozone depletion104 comments on the importance of the ozone hole as a driver of changes in Southern Hemisphere seasonal surface winds at mid-to-high latitudes. However, the influence of the hole extends to the whole of the hemisphere.114 Modelling by Son et al.,115 which incorporates stratospheric chemical interactions and takes into account the likely influence of recovering ozone levels, indicates that the anticipated recovery of the hole may result in a reversal of the current acceleration of these seasonal surface winds (summer tropospheric westerlies) on the poleward side. The authors concluded:

… our analyses suggest that stratospheric processes, and ozone recovery in particular, may be able to affect SH [Southern Hemisphere] climate in major ways and thus should be included in predictions of SH climate in the 21st century.

In addition to its influence on climate, the ozone hole has been of concern in relation to UVB effects on health. The progressively more rigorous controls established under the Montreal Protocol during the 1990s are expected to lead to the avoidance of a significant increase in cases of skin cancer that would otherwise have been associated with large reductions in global stratospheric ozone (Figure 3.22). This is of particular importance in Australia, where high levels of UVB radiation combine with outdoor lifestyles to produce one of the highest incidence rates of skin cancer in the world.116-117

Figure 3.22

ppb = parts per billions

Source: Climate Change Science Program118

Figure 3.22 Effect of the Montreal Protocol and its amendments on ozone depleting substances and excess skin cancer cases

The top panel gives a measure of the projected future abundance of ozone depleting substances in the stratosphere, without and with the Montreal Protocol and its various amendments. The bottom panel shows similar projections for how excess skin cancer cases might have increased.

Box 3.5 Ozone and UV radiation

Ultraviolet (UV) radiation levels at ground level are principally an inverse function of the amounts of ozone in the upper atmosphere, the concentration of aerosols and water vapour in the atmosphere, and the extent of cloud cover. Long-term trends in UV radiation levels are measured at Lauder in New Zealand’s South Island and are modelled at a number of sites in Australia. UV is expressed as an erythemal UV index—a measure that describes the strength of the skin-burning component of UV radiation.

Figure A shows total column ozone levels over Melbourne and erythemal index values for Lauder and three Australian capital cities. The top panel shows January mean total ozone values, measured by the Bureau of Meteorology’s Dobson network at locations around greater Melbourne from 1979 to 2011. The green line shows a five-year running mean. Although year-to-year variability is evident—as is a clear signal of the 11-year solar cycle (which peaked around 1980, 1991 and 2002)—the underlying negative trend in ozone ceased in the mid-1990s. The long-term ozone behaviour closely follows the concentration of ODSs measured in the global atmosphere, which peaked in the mid-1990s (see Figure 3.19)

Figure A

Source: M Tully, Leader Ozone Science Team, Bureau of Meteorology, pers. comm., August 2011; R McKenzie, Principal Scientist—Radiation, National Institute of Water and Atmospheric Research, Central Otago, New Zealand, pers. comm., July 2011; McKenzie et al.;119 Lemus-Deschamps et al.;120 Lemus-Deschamps & Makin121

Figure A Total column ozone for Melbourne, surface-measured erythemal index for Lauder (New Zealand) and satellite-derived clear-sky erythemal index for three Australian cities

The middle panel shows summer-time peak UV index values in 1990–2010, as measured by UV spectroradiometer at Lauder by New Zealand’s National Institute of Water and Atmospheric Research. Although UV radiation values are affected by factors other than just ozone, the underlying trend quite closely follows the expected inverse relation to the ozone timeseries, with highest UV index values in the late 1990s when ozone was lowest.

The bottom panel shows modelled clear-sky UV index values for three Australian cities (Sydney, Adelaide and Melbourne), illustrating the effect of location on the amount of UV radiation received. The values were calculated using summer satellite measurements of ozone and meteorological fields from the Bureau of Meteorology forecast model, as input to the UV radiation code.

Some differences in the detail of panels 1 and 3 are evident. However, the overall pattern of rising UV through the 1980s and 1990s, followed by a stabilisation, corresponds to the decline and subsequent stabilisation of ozone during the same period. The differences are primarily due to the use of satellite-measured ozone values rather than ground-based values, a slightly different averaging period (all of summer in panel 3 compared with just January in panel 1) and some missing periods of satellite data in the late 1990s, when ozone values were low.

From the early 1980s to the early 2000s, Australian skin cancer rates for both sexes showed a generally increasing trend, after which rates appear to have stabilised (Figure 3.23). Most recent data for melanoma show a decline in both male (7.1%) and female (10.7%) rates from 2005 to 2007.122-123 However, the period involved is too short to tell whether the reduction indicates a genuine decline or is the result of fluctuations in the data.

Figure 3.23

Source: Australian Institute of Health and Welfare122

Figure 3.23 Australian skin melanoma rates by gender, 1982–2007

3.1.2 Ambient air quality

Ambient air quality and health

Although air pollution can harm vegetation, erode the facades of historic stone buildings and limit visibility, the main focus of public concern over air pollution is its short-term and long-term effects on human health. Over the past decade, scientific studies have greatly expanded our understanding of the nature and extent of the effects of major air pollutants in our cities (e.g. Environment Protection Authority Victoria,124-125 Environment Protection and Heritage Council,126-127 Simpson et al.).128-129 On the basis of these and related studies, it is clear that urban air pollution is a significant cause of death and illness in the community. By one estimate,7 there were close to 3000 deaths due to urban air pollution in 2003. This was 2.3 % of all deaths and nearly twice the national road toll. Two-thirds of these deaths were attributable to long-term exposure to air pollutants, with the elderly most affected. The health burden associated with urban air pollution was shared about equally between males and females (53% to 47%). Such deaths occur from a range of medical causes (Figure 3.24).

Figure 3.24

Source: Begg et al.7

Figure 3.24 Deaths attributed to long-term exposure to urban air pollution, 2003

As shown in Table 3.4, a range of adverse health effects is associated with air pollution. The nature and severity of the effect are a function of the type and concentration of pollutant, the duration of exposure and the sensitivity of the individual. Individual sensitivity is influenced by factors such as age, general state of health and fitness, and prior illnesses.

Table 3.4 Health effects and populations at risk from certain air pollutants
Pollutant Health effects Population at risk
Carbon monoxide Mortality and increased hospital admission due to heart disease People with ischaemic heart conditions
Nitrogen dioxide Hospital admissions for respiratory diseases, decreases in lung function, cardiovascular disease Sufferers of respiratory disease, such as children with asthma; those with cardiovascular disease
Particulates Mortality due to cardiovascular and respiratory diseases; hospital admissions due to respiratory and cardiovascular disease; decreases in lung function Elderly people with respiratory and cardiovascular diseases; people with respiratory diseases, such as children with asthma
Ozone Mortality due to respiratory and cardiovascular diseases; hospital admissions due to respiratory diseases; decreases in lung function Elderly people; people with respiratory diseases

Source: Adapted from Environment Protection and Heritage Council127

National air quality standards

In 1998, the Australian and state and territory governments adopted a set of national air quality standards—the National Environment Protection (Ambient Air Quality) Measure (AAQ NEPM). The standards covered the six most common air pollutants (also referred to as ‘criteria pollutants’)—carbon monoxide, nitrogen dioxide, sulfur dioxide, ozone, lead and particulate matter smaller than 10 micrometres (PM10), which can be inhaled directly into the lungs (Table 3.5).

In 2003, the standards were amended to include advisory reporting standards for fine particulate matter smaller than 2.5 micrometres (PM2.5), reflecting growing concern over links between increases in PM2.5 levels and mortality and morbidity associated with respiratory and cardiovascular disease (Table 3.6). A number of studies have suggested that the impact of the PM2.5 size fraction may be most pronounced in relation to cardiovascular illness and mortality, whereas the coarse (PM10) fraction may be more important in worsening asthma and upper respiratory illnesses. The health effects of short-term exposure to elevated levels of PM2.5 is an important area for further research.6,135

Table 3.5 Standards and goals for pollutants other than PM2.5 particles
Pollutant Averaging period Maximum concentration Goal (within 10 years) for maximum allowable exceedences
Carbon monoxide 8 hours 9.0 ppm 1 day per year
Nitrogen dioxide 1 hour 0.12 ppm 1 day per year
1 year 0.03 ppm None
Sulfur dioxide 1 hour 0.20 ppm 1 day per year
1 day 0.08 ppm 1 day per year
1 year 0.02 ppm None
Photochemical oxidants (as ozone) 1 hour 0.10 ppm 1 day per year
4 hours 0.08 ppm 1 day per year
Lead 1 year 0.50 µg/m3 None
Particles (PM10) 1 day 50 µg/m3 5 days per year

PM10 = particulate matter smaller than 10 micrometres; ppm = parts per million; µg/m3 = micrograms per cubic metre

Source: Office of Legislative Drafting134

Box 3.6 Pollens—the forgotten air pollutants

Many types of flowering plant depend on the wind to distribute their pollen, and the small, light, dry pollen grains that these plants produce are easily breathed in by humans. When inhaled, proteins and glycoproteins associated with these pollens can interact with the immune systems of sensitive individuals to produce an allergic response in the form of hayfever or allergic asthma.130 Acting on its own or in combination with fine particles, airborne pollen is known to influence the incidence and severity of hayfever and asthma in a population.131 Traidl-Hoffman et al.130 report that the incidence of hayfever and allergic asthma in the community has more than doubled over the past 30 years, with 10–25% of the population experiencing symptoms.

Although intact pollen grains (with a diameter of around 10 micrometres) are too large to penetrate to the lower airways, research has shown that pollens can rupture under wet conditions and during thunderstorms, releasing allergen-carrying starch granules small enough to reach these airways. This can trigger severe asthma attacks in hayfever sufferers.132 In addition to the well-publicised links between pollen, hayfever and allergic asthma, a timeseries study in the Netherlands found a strong association between daily pollen levels and mortality due to cardiovascular disease, chronic obstructive pulmonary disease and pneumonia.133

The standards are measured at locations that are generally representative of the level of exposure of the broad population, rather than at ‘hot spots’ (such as near major point sources or roads). Authorities have agreed on standardised monitoring methods, to ensure national comparability of results.136

The standards are set at levels intended to protect human health. They reflect the evidence available in the mid-to-late 1990s on links between the various pollutants and human health. The AAQ NEPM is currently being reviewed in light of new evidence on health effects and international trends in air quality standards. Analysis of such evidence confirms that a number of the criteria pollutants (e.g. ozone and particles) do not have a threshold level below which there is no health effect. This means that sensitive individuals, such as asthmatics and people with respiratory or cardiovascular disease, may be affected even when air quality standards are met.6

As well as setting standards for the criteria pollutants, the AAQ NEPM established goals, expressed in terms of ‘maximum allowable exceedences’, to be achieved within 10 years (i.e. by 2008). These goals reflected a broadly based consensus on the extent of improvements in air quality that would be practicable over the period.

Table 3.6 Advisory reporting standards and goal for PM2.5 particles
Pollutant Averaging period Maximum concentration (µg/m3) Goal
Particles (PM2.5) 1 day 25 To gather sufficient national data to facilitate a review of the advisory reporting standards
1 year 8

PM2.5 = particulate matter smaller than 2.5 micrometres; µg/m3 = micrograms per cubic metre

Source: Office of Legislative Drafting134

Box 3.7 The air quality index

In a number of states, the agency responsible for monitoring air quality reports results at each station in its network in terms of an air quality index (AQI) for all or a subset of the pollutants covered by the National Environment Protection (Ambient Air Quality) Measure (AAQ NEPM), apart from PM2.5, which is not regularly reported. Often, the highest AQI is reported for each location. The AQI relates the observed level of a pollutant to the AAQ NEPM standard, expressed as a percentage. A given index level reflects the residual risks to public health.

Index = Pollutant concentration ÷ Pollutant standard level × 100

Five qualitative categories are used in public reporting of air quality. The categories and the AQI ranges that they represent are listed in Table A, together with a qualitative description of the associated health effects.

Table A Air quality index
Very poor 150+ Air quality is unhealthy, and everyone may begin to experience health effects. People in sensitive groups may experience more serious health effects.
Poor 100–149 Air quality is unhealthy for sensitive groups. The general population is not likely to be affected in this range.
Fair 67–99 Air quality is acceptable. However, there may be a health concern for very sensitive people.
Good 66–34 Air quality is considered good, and air pollution poses little or no risk.
Very good 0–33 Air quality is considered very good, and air pollution poses little or no risk.

Each category in the AQI corresponds to a different level of air quality and associated health risk.

Source: Australian Government Department of Sustainability, Environment, Water, Population and Communities6

Pollutant sources

Most pollutants (carbon monoxide, nitrogen dioxide, sulfur dioxide and PM2.5 particles) result from combustion (primary pollutants) (Table 3.7). Major sources include motor vehicles, industrial processes and domestic heating. Coarse particles (the PM10 fraction)—which include mineral dust, salt and soot—are also a form of primary pollutant, originating from both natural and human sources. Secondary pollutants (such as ozone) result from the action of complex photochemical processes on primary pollutants (oxides of nitrogen and volatile organic compounds), predominantly in the warmer months, forming photochemical smog.

Table 3.7 Major sources of criteria pollutants and fine particles (PM2.5)
Pollutant Major sources
Primary pollutants  
Nitrogen dioxide (NO2), together with nitric oxide (NO), generalised as NOx Combination of nitrogen and oxygen during high-temperature combustion of fossil fuels
Around 80% of urban NO2 is from motor vehicle exhaust
Other sources are petrol and metal refining, electricity generation from coal-fired power stations, other manufacturing industries and food processing
Sulfur dioxide Electricity generation from fossil fuels, metal smelting of sulfurous ores, including aluminium, copper, lead, zinc and iron
Carbon monoxide Combustion, including vegetation burning and wildfires, motor vehicles and metal manufacturing
Lead Road dust, metal manufacturing and metal ore mining
PM10 In nonurban areas: vegetation burning, wildfires, soot, windblown dust from agriculture and other land uses, road dust
In urban areas: predominantly motor vehicles and secondary particles
Other sources are solid (domestic) fuel burning in winter and mining
PM2.5 Combustion sources, secondary nitrates and sulfates, secondary organic aerosol and natural-origin dust
Secondary pollutants  
Ozone Atmospheric photochemical reactions of primary pollutants, NOx and hydrocarbons (volatile organic carbons ) from motor vehicles and industry
Naturally occurring ozone

PM2.5/10 = particulate matter smaller than 2.5 or 10 micrometres

Sources: Australian Government Department of Sustainability, Environment, Water, Population and Communities;6Environment Protection and Heritage Council;127Goldstein & Galbally;137Oltmans et al.138

Although standards and monitoring strategies necessarily focus on individual pollutants, it is becoming increasingly clear that many of the health effects of air pollution are not due to single pollutants acting in isolation. This is hardly surprising, given that most major sources of urban air pollution (such as motor vehicle exhausts and domestic combustion heaters) emit a complex mix of gaseous and solid pollutants, some of which act as the building blocks for secondary pollutants such as ozone and some forms of fine particulate pollution.

Ambient air quality trends

Air quality in Australia’s urban centres is generally good, with levels of all the criteria pollutants usually falling well below the national standards. The latest national state of the air report6 showed that the levels of these pollutants (other than ozone and particles) declined or remained stable from 1999 to 2008. In some cities, peak ozone levels occasionally approached or exceeded the NEPM standard, whereas peak particle levels in nearly all regions exceeded the standard on a number of days. If, as seems likely, the yet-to-be-released review of the AAQ NEPM leads to a tightening of the current four-hour ozone standard (readings averaged over four hours) and to the establishment of an eight-hour standard (readings averaged over eight hours), the frequency of ozone exceedences is likely to increase significantly, particularly in Sydney.

Urban air quality varies both with short-term meteorological conditions—such as temperature inversions, which can trap pollutants near ground level—and seasonally, with summer temperatures promoting the formation of ozone and other photochemical pollutants. Extreme weather events are often associated with ‘peak’ pollution levels in Australian cities, where peak refers to the top 5% of measurements—this is distinct from the maximum (or highest) measurement. In addition, the frequency and severity of pollution events are strongly influenced in centres such as Sydney and Melbourne (Box 3.8) by the regional topography and the presence of the sea, which affect the circulation of air in those airsheds, recirculating polluted air and promoting the formation of photochemical smog. (An airshed is a body of air, bounded by meteorology and topography, in which substance emissions are contained. For example, the Bunbury Regional Airshed study area in Western Australia covers an area of 165 kilometres [east to west] by 234 kilometres [north to south] and contains a population of 270 000 people.)139

Box 3.8 The Melbourne eddy recorded by a weather satellite in February 1985 (the eddy is made visible by low cloud)

The Melbourne eddy recorded by a weather satellite in February 1985 (the eddy is made visible by low cloud) including Melbourne CBD and Port Phillip Bay

Under a special set of meteorological conditions, air flowing from the north-east is funnelled by mountains to the north and east of Port Phillip Bay, creating a circular (clockwise), horizontal motion of air (about 100 kilometres in diameter). The eddy pushes air pollution out over the bay, initially taking it away from Melbourne before returning it in reacted form as photochemical smog.

Source: Satellite image originally processed by the Bureau of Meteorology from the polar-orbiting satellite NOAA-6, operated by the National Oceanographic and Atmospheric Administration (NOAA) and Bureau of Meteorology.

Ozone

The state of the air report shows ozone levels in the Sydney and the Illawarra regions to have been generally higher than in other Australian metropolitan and industrial regions, exceeding the one-hour and four-hour ozone standards in most years from 1999 to 2008. Expressed in terms of the air quality index (AQI), Sydney’s annual maximum four-hour ozone levels were generally classed as poor, whereas the 95th percentile levels were in the fair range. By comparison, annual maximum ozone levels in Melbourne, Brisbane, Perth, Adelaide and Canberra occasionally exceeded the standards, but generally rated as fair. Among these five cities, Melbourne’s annual peak ozone levels were the highest, exceeding the four-hour standard in some years, with a consequent AQI rating of poor (Figure 3.25). Median (50th percentile) levels in all regions were around 40% of the four-hour standard (in AQI terms, rating as good). Overall, across the 44 NEPM monitoring sites, the report discerned no trends in ozone levels.6

Figure 3.25 Melbourne, Sydney, Brisbane and Perth Concentration (ppm) between 1999 and 2008, indicating NEPM standard

µg/m3 = microgram per cubic metre; NEPM = National Environment Protection Measure; ppm = parts per million

Source: Australian Government Department of Sustainability, Environment, Water, Population and Communities6

Figure 3.25 For four major cities, the (a) average maximum four-hour average ozone concentrations and (b) average 95th percentile four-hour average ozone concentrations, 1999–2008

Particles (PM10)

Peak PM10 levels in both urban and nonurban areas tend to be seasonal. In summer, wildfires and dust storms associated with occasional extreme weather can lead to very high levels of particle pollution. In areas with a high dependence on solid fuel burning for domestic heating, the seasonal peak in particle levels usually occurs in winter. This is particularly the case in centres such as Launceston, where local topography can lead to a layer of cold polluted air being trapped near the ground by an overlying layer of warmer air (a situation referred to as a temperature inversion). In autumn, when most forest fuel-reduction burns occur, some areas, including metropolitan centres, can experience significant particulate pollution.

From 1999 to 2008, maximum PM10 levels in Australian capitals and in some regional centres often exceeded the AAQ NEPM 24-hour standard, with levels up to 4 times the standard in the capitals and up to 14 times in some regional centres. However, annual state and territory reports on NEPM implementation reveal that in the capital cities such exceedences were generally limited in number and mainly related to extreme events, on which government air quality improvement programs have very limited effect.

In all capitals (other than Canberra and Hobart), 95th percentile PM10 levels met the 24-hour standard (with levels falling in the fair or good AQI categories). In both Canberra and Hobart, the 95th percentile values exceeded the standard on one or two occasions early in the decade, subsequently declining and stabilising at levels comparable with the other capitals (Figure 3.26). These declines (along with a similar reduction in Launceston) largely reflect the success of programs to reduce wood smoke from domestic heaters. Setting aside these reductions, no trend is clear in the data for the other major cities.6

Figure 3.26a Melbourne, Sydney, Brisbane and Perth
Figure 3.26b Adelaide, Hobart, Darwin and Canberra

µg/m3 = microgram per cubic metre; NEPM = National Environment Protection Measure; PM10 = particulate matter smaller than 10 micrometres

Source: Australian Government Department of Sustainability, Environment, Water, Population and Communities6

Figure 3.26 Average 95th percentile 24-hour average PM10 concentrations in (a) Melbourne, Sydney, Brisbane and Perth, and (b) Adelaide, Hobart, Darwin and Canberra, 1999–2008

As the state of the air report notes, particle levels tend to be slightly higher in regional cities in south-eastern Australia than in the capital cities. The most likely explanation is their greater seasonal exposure to the effects of bushfires, dust storms, planned burning and the use of wood for domestic heating.6

Fine particles (PM2.5)

Fine particle levels are monitored at 18 sites around Australia, with Perth having the longest record, starting in 1994 (Figure 3.27). In 2008, the 24-hour advisory reporting standard (25 micrograms per cubic metre) was met at six sites (four in New South Wales, one in Queensland and one in South Australia). Peak levels are highly variable, being strongly influenced by extreme events such as bushfires and dust storms. This, together with a relatively short monitoring record at most sites, makes the identification of long-term trends problematic.6

Figure 3.27

µg/m3 = microgram per cubic metre; NEPM = National Environment Protection Measure; PM2.5 = particulate matter smaller than 2.5 micrometres

Source: Australian Government Department of Sustainability, Environment, Water, Population and Communities,6 National Air Quality Database

Figure 3.27 Capital cities’ highest daily average PM2.5 concentrations

Carbon monoxide

National Pollutant Inventory data show that, apart from vegetation burning and wildfires, motor vehicles are the main source of carbon monoxide. Over the past two decades, levels of carbon monoxide have declined significantly. This has been part of a broader improvement in air quality associated with strengthened vehicle standards that required the exhaust systems of new vehicles to be fitted with catalytic converters, and legislation requiring the phase-in of unleaded fuel. Current peak carbon monoxide levels fall into the very good AQI category in all regions and are less than one-third to one-fifth of the national standard.6,140

Nitrogen dioxide

Levels of nitrogen dioxide (both maximum and 95th percentile) are generally well below both the one-hour average and the annual average AAQ NEPM standards, rating good to very good (maximum) and very good (95th percentile) in terms of the AQI. Long-term records show a significant decline in maximum nitrogen dioxide levels, most notably in the 1990s. As is the case for carbon monoxide, the decrease was mainly driven by the introduction of tighter vehicle emission standards. During the past decade, although there has been a continued decline in nitrogen dioxide in some areas, levels have generally remained stable, despite an increase in vehicle numbers and distances travelled.6,141

Sulfur dioxide

Sulfur dioxide levels remain low in the capitals and most other urban areas. Between 1995 and 2000, there was an overall reduction in Australian sulfur dioxide emissions of almost one-third, due mainly to recovery of sulfur dioxide to produce sulfuric acid. However, over the past decade, levels across Australian cities have been relatively stable, despite the progressive tightening of standards for sulfur dioxide in fuel.6,142 Mount Isa and Port Pirie, with their large ore smelting operations, typically experience more than 20 exceedences of the one-hour sulfur dioxide NEPM standard, with Mount Isa recording more than three times and Port Pirie more than twice the standard—that is, in the very poor AQI category.6,107

Lead

Since the start of the national phase-out of leaded petrol in 1993, atmospheric lead levels in Australian cities have fallen markedly and are now below 10% of the national standard. The few exceptions are regional towns (such as Port Pirie) with large industrial point sources (Figure 3.28). This improvement is particularly welcome because lead (a persistent neurotoxin) is known to have adverse effects on children’s development of memory and motor abilities, even at low levels.143

Figure 3.28

µg/m3 = microgram per cubic metre

Source: Australian Government Department of Sustainability, Environment, Water, Population and Communities6

Figure 3.28 Lead levels at two locations in Port Pirie, South Australia

Volatile organic compounds

Volatile organic compounds (VOCs) are primary pollutants that react with nitrogen oxides in complex photochemical processes to generate a range of secondary pollutants (notably ozone). Biogenic sources (vegetation and soil) form about three-quarters of the total VOC emissions from natural and human sources.144 Although emissions from biogenic sources are not hazardous themselves, they add to the background level of VOCs and thus can contribute to the formation of ozone. Figure 3.29 shows the main sources of VOCs other than vegetation and soil.

Figure 3.29 - Motor vehicles 32%, Burning (fuel reduction/ regeneration/ agriculture and wildfires) 21%, Industry 13%, Domestic/ commercial solvents/ aerosols 11%, Solid fuel burning - domestic 9%, Architectural surface coatings 6%, Other area based 8%

Source: National Pollutant Inventory145

Figure 3.29 Total volatile organic compounds by source, excluding biogenics, 2009–10

Air toxics

Air toxics (also called hazardous air pollutants) are a broad group of pollutants found in ambient air, usually at relatively low levels. These hazardous air pollutants include known or suspected carcinogens and pollutants linked to other serious health impacts, including birth defects and developmental, respiratory and immune system problems.146 They include heavy metals and many types of volatile and semivolatile organic compounds. Some of these compounds (such as dioxins and furans) are highly persistent, tend to accumulate through food chains (bioaccumulation) and can be transported long distances through the atmosphere. These persistent organic pollutants are a focus of significant international concern and are controlled under the United Nations Stockholm Convention on Persistent Organic Pollutants, to which Australia is a signatory.147 Air toxics are formed as products of combustion (motor vehicles are a significant source), as volatile emissions from paints and adhesives, and from various industrial processes.

In 2004, the National Environment Protection Council agreed on an additional air quality NEPM to address air toxics. This measure deals with five priority air toxics: benzene, toluene, xylene (collectively referred to as BTX), formaldehyde and benzo(a)pyrene (as a marker for polycyclic aromatic hydrocarbons). The NEPM adopts a nationally consistent approach to monitoring this initial group of air toxics at sites likely to experience elevated levels (such as near major roads and industrial areas) and establishes a series of benchmarks (‘monitoring investigation levels’ [MILs]) that, if exceeded, require further investigation and evaluation (see Table 3.8.) A key aim of the NEPM is to develop a robust set of data on ambient levels of these priority air toxics to enable future ministerial councils to set national air quality standards that will protect human health.148

Table 3.8 Monitoring investigation levels for air toxics
Pollutant Averaging period Monitoring investigation level Goal
Benzene Annual averagea 0.003 ppm 8-year goal is to gather sufficient national data to facilitate development of a standard
Benzo(a)pyrene (as a marker for polycyclic aromatic hydrocarbons) Annual averagea 0.3 ng/m3
Formaldehyde 24 hoursb 0.04 ppm
Toluene 24 hoursb
Annual averagea
1 ppm
0.1 ppm
Xylenes (as total of ortho-, meta- and para-isomers) 24 hoursb
Annual averagea
0.25 ppm
0.2 ppm

ng/m3 = nanograms per cubic metre; ppm = parts per million

a For the purposes of this measure, the annual average concentrations in column 3 are the arithmetic mean concentrations of 24-hour monitoring results.

b For the purposes of this measure, monitoring over a 24-hour period is to be conducted from midnight to midnight.

Source: Adapted from Office of Legislative Drafting148

A mid-term review of the air toxics NEPM summarised the results of five years’ monitoring. Benzene levels at nearly all sites were at or below the MIL. However, some sites near heavily trafficked roads and one in a mixed industrial area (where non-NEPM monitoring methods were used) recorded levels close to or above the MIL. A clear time trend could not be discerned. Toluene and xylene levels were (with a small number of exceptions) well below the MILs, with some signs of a reducing trend for toluene. Most formaldehyde measurements were significantly below the MIL, although the review notes that the dataset is much more limited than for BTX. Due to the limited amount of monitoring, results for benzo(a)pyrene were inconclusive. Overall, the review noted the limited extent of data collection at most sites, expressed caution in relation to interpreting the limited results, and argued for additional monitoring.149

Assessing the condition of ambient air

The qualitative assessment summaries 3.43.7 (below), based on the method outlined in Box 3.9, are generalised across periods of up to 11 years. The fact that these assessments indicate that overall air quality is good or very good should not be allowed to obscure the fact that, on a number of days each year, all of these cities experience air quality that does not meet the national health-based standards (i.e. air quality is in the poor or very poor categories). As a result, air pollution in our capitals and major regional centres remains a significant cause of death and illness in the community, particularly affecting the health of sensitive individuals and groups.

Box 3.9 Applying a graded report-card approach to Australia’s urban air quality

As part of the State of the Environment reporting process, a qualitative assessment was made of ambient air quality in the eight state and territory capitals and a small number of major regional or industrial centres. Of the seven pollutants for which national health-based standards have been set, photochemical oxidants (as ozone) and particulate matter smaller than 10 micrometres (as PM10) were chosen as key pollutants potentially impacting on human health, reflecting the weight of scientific evidence.

The approach based the characterisation of an airshed from 1999 to 2009 on its worst performing monitoring station, rather than on the total number of exceedences across the airshed, since this is strongly influenced by the number of monitoring stations. Only data from monitoring stations established in accordance with an approved National Environmental Protection Measure (NEPM) monitoring plan were considered. (In some cases, less than 10 years of NEPM monitoring data were available, and in one case—Perth—11 years of data were available.) Most large regional cities have only one NEPM monitoring station, and most monitor particles, but not ozone, since they lack the scale of industry and traffic likely to give rise to ozone as a secondary pollutant. In each state, the regional cities selected for analysis of PM10 and ozone (where this was monitored) were the worst performing in the state.

It is recognised that the 10-year goals set in the Ambient Air Quality NEPM for ozone and particles allow for one exceedence per year for ozone and five exceedences per year for particles. Nevertheless, given the nature of the health-based standards, any exceedence may have a potentially adverse impact and should therefore be taken into consideration, even if the goal is met.

Ozone levels were evaluated against the four-hour exposure standard rather than the one-hour, as the four-hour standard is more likely to give a better indication of the impact on the general population, rather than on sensitive individuals who are likely to be affected by acute (i.e. shorter term) events.

Procedure

For each year, monitoring data for ozone and PM10 from each of the selected stations were converted into air quality index (AQI) values. These were used as the basis for calculating the percentage of observations that fell in each of the five AQI-based qualitative categories (very good, good, fair, poor and very poor) commonly used by Australian environment protection agencies to report air quality.

Each of these yearly percentage distributions for each pollutant at each station was then assessed against the criteria set out in Table A, to assign a general AQI score to each pollutant. The results across the period were represented graphically to assist in identifying any trends. It must be emphasised that the criteria set out in Table A are essentially subjective in nature. In almost all cases, their application resulted in the most frequently occurring AQI category being selected to generalise the year as a whole. In a small number of years, the AQI distribution was bimodal, with the result being borderline between the very good and good categories.

Overall qualitative AQI scores for ozone and PM10 were then assigned to each city, based on the most frequently occurring scores during the decade. A summary of the results is presented in assessment summaries 3.4 and 3.5, and the complete set of graphs is available on the State of the Environment website.a

Table A Criteria for assigning annual AQI-based qualitative scores
Overall category Very good (%) Good (%) Fair (%) Poor (%) Very poor (%)
Very good >50 >20      
Good >20 >30      
Fair     >30 >20  
Poor       >30 >20
Very poor       >20 >50
Supplementary rules

If the percentage very good is greater than 45 and is also greater than the percentage good, the assessment grade is very good.

If the percentage good is greater than 75, then the percentage very good can be as low as zero and assessment grade is good.

Assessment summary 3.4—metropolitan cities’ score card for ozone (four-hour) NEPM standard, based on analysis of air quality index values, 1999–2008
Component Summary Assessment grade Confidence
Very poor Poor Fair Good Very good in grade in trend
Adelaide Average percentage frequency distribution:
very good 46; good 53; fair 1; poor 0; very poor 0
       Stable    Adequate high-quality evidence and high level of consensus  Adequate high-quality evidence and high level of consensus
Brisbane Average percentage frequency distribution:
very good 31; good 64; fair 55; poor 0; very poor 0
       Stable    Adequate high-quality evidence and high level of consensus  Adequate high-quality evidence and high level of consensus
Canberra Average percentage frequency distribution:
very good 43; good 55; fair 2; poor 0; very poor 0
       Deteriorating    Adequate high-quality evidence and high level of consensus  Limited evidence or limited consensus
Melbourne Average percentage frequency distribution:
very good 34; good 63; fair 3; poor 0; very poor 0
       Stable    Adequate high-quality evidence and high level of consensus  Adequate high-quality evidence and high level of consensus
Perth Average percentage frequency distribution:
very good 16; good 79; fair 4; poor 0; very poor 0
       Stable    Adequate high-quality evidence and high level of consensus  Adequate high-quality evidence and high level of consensus
Sydney Average percentage frequency distribution:
very good 17; good 72; fair 9; poor 2; very poor 0
       Stable    Adequate high-quality evidence and high level of consensus  Adequate high-quality evidence and high level of consensus
Recent trends  Improving  Stable Confidence  Adequate high-quality evidence and high level of consensus
 Deteriorating  Unclear  Limited evidence or limited consensus
 Evidence and consensus too low to make an assessment
Grades  Very good: Air quality is considered very good, and air pollution poses little or no risk
 Good: Air quality is considered good, and air pollution poses little or no risk
 Fair: Air quality is acceptable. However, there may be a health concerns for very sensitive people
 Poor: Air quality is unhealthy for sensitive groups. The general population is not likely to be affected in this range
 Very poor: Air quality is unhealthy, and everyone may begin to experience health effects. People from sensitive groups may experience more serious health effects

NEPM = National Environment Protection Measure

Note: Melbourne assessment based on 2002–08; ozone is not regularly monitored in Darwin or Hobart.

Assessment summary 3.5—metropolitan cities’ score card for particles (PM10) NEPM 24-hour standard, based on analysis of air quality index values, 1999–2008
Component Summary Assessment grade Confidence
Very poor Poor Fair Good Very good in grade in trend
Adelaide Average percentage frequency distribution:
very good 42; good 51; fair 5; poor 1; very poor 1
       Stable    Adequate high-quality evidence and high level of consensus  Adequate high-quality evidence and high level of consensus
Brisbane Average percentage frequency distribution:
very good 63; good 34; fair 3; poor 0; very poor 0
         Stable  Adequate high-quality evidence and high level of consensus  Adequate high-quality evidence and high level of consensus
Canberra Average percentage frequency distribution:
very good 55; good 26; fair 11; poor 6; very poor 2
(Note: borderline very good – good)
       Improving  Adequate high-quality evidence and high level of consensus  Adequate high-quality evidence and high level of consensus
Darwin Average percentage frequency distribution:
very good 56; good 42; fair 2; poor 0; very poor 0
         Stable  Adequate high-quality evidence and high level of consensus  Limited evidence or limited consensus
Hobart Average percentage frequency distribution:
very good 73; good 26; fair 1; poor 0; very poor 0
         Unclear  Adequate high-quality evidence and high level of consensus  Limited evidence or limited consensus
Melbourne Average percentage frequency distribution:
very good 36; good 50; fair 11; poor 2; very poor 1
       Stable    Adequate high-quality evidence and high level of consensus  Adequate high-quality evidence and high level of consensus
Perth Average percentage frequency distribution:
very good 49; good 47; fair 3; poor 0; very poor 0
(Note: borderline very good – good)
       Stable  Adequate high-quality evidence and high level of consensus  Adequate high-quality evidence and high level of consensus
Sydney Average percentage frequency distribution:
very good 50; good 45; fair 4; poor 1; very poor 0
(Note: borderline very good – good)
       Stable  Adequate high-quality evidence and high level of consensus  Adequate high-quality evidence and high level of consensus
Recent trends  Improving  Stable Confidence  Adequate high-quality evidence and high level of consensus
 Deteriorating  Unclear  Limited evidence or limited consensus
 Evidence and consensus too low to make an assessment
Grades  Very good: Air quality is considered very good, and air pollution poses little or no risk
 Good: Air quality is considered good, and air pollution poses little or no risk
 Fair: Air quality is acceptable. However, there may be a health concerns for very sensitive people
 Poor: Air quality is unhealthy for sensitive groups. The general population is not likely to be affected in this range
 Very poor: Air quality is unhealthy, and everyone may begin to experience health effects. People from sensitive groups may experience more serious health effects

NEPM = National Environment Protection Measure; PM10 = particulate matter smaller than 10 micrometres

Note: Hobart assessment based on 2006–08; Melbourne assessment based on 2002–08; Perth assessment based on 2000–08

Assessment summary 3.6—regional cities’ score card for ozone (four-hour) NEPM standard, based on analysis of air quality index values, 1999–2008
Component Summary Assessment grade Confidence
Very poor Poor Fair Good Very good in grade in trend
New South Wales—Kembla Grange Average percentage frequency distribution:
very good 35; good 62; fair 2; poor 1; very poor 0
       Stable    Adequate high-quality evidence and high level of consensus  Adequate high-quality evidence and high level of consensus
Victoria—Moe Average percentage frequency distribution:
very good 68; good 31; fair 1; poor 0; very poor 0
         Unclear  Adequate high-quality evidence and high level of consensus  Limited evidence or limited consensus
Recent trends  Improving  Stable Confidence  Adequate high-quality evidence and high level of consensus
 Deteriorating  Unclear  Limited evidence or limited consensus
 Evidence and consensus too low to make an assessment
Grades  Very good: Air quality is considered very good, and air pollution poses little or no risk
 Good: Air quality is considered good, and air pollution poses little or no risk
 Fair: Air quality is acceptable. However, there may be a health concerns for very sensitive people
 Poor: Air quality is unhealthy for sensitive groups. The general population is not likely to be affected in this range
 Very poor: Air quality is unhealthy, and everyone may begin to experience health effects. People from sensitive groups may experience more serious health effects

NEPM = National Environment Protection Measure

Note: Ozone is regularly monitored in regional cities only in New South Wales, Queensland and Victoria. Of these states, only New South Wales and Victoria monitor ozone using both the NEPM one-hour and four-hour averaging periods. In Queensland, ozone is monitored at sites in Toowoomba and Townsville, but only using the one-hour averaging period. For this reason, only regional cities in New South Wales and Victoria were included in this assessment summary.

Assessment summary 3.7—regional cities’ score card for particles (PM10) NEPM 24-hour standard, based on analysis of air quality index values, 1999–2008
Component Summary Assessment grade Confidence
Very poor Poor Fair Good Very good in grade in trend
New South Wales—Wagga Wagga Average percentage frequency distribution:
very good 34; good 45; fair 13; poor 6; very poor 2
(Note: Although the overall assessment is good, the distribution has a significant ‘tail’ of 29 days on which the national standard was exceeded)
       Stable    Adequate high-quality evidence and high level of consensus  Adequate high-quality evidence and high level of consensus
Queensland—West Mackay Average percentage frequency distribution:
very good 30; good 61; fair 1; poor 0; very poor 2
       Stable    Adequate high-quality evidence and high level of consensus  Adequate high-quality evidence and high level of consensus
South Australia—Port Pirie Average percentage frequency distribution:
very good 55; good 34; fair 8; poor 2; very poor 1
         Improving  Adequate high-quality evidence and high level of consensus  Limited evidence or limited consensus
Tasmania—Launceston Ti Tree Bend Average percentage frequency distribution:
very good 65; good 30; fair 3; poor 2; very poor 1
         Unclear  Limited evidence or limited consensus  Limited evidence or limited consensus
Victoria—Geelong South Average percentage frequency distribution:
very good 39; good 49; fair 9; poor 2; very poor 1
       Stable    Adequate high-quality evidence and high level of consensus  Adequate high-quality evidence and high level of consensus
Western Australia—Bunbury Average percentage frequency distribution:
very good 46; good 52; fair 2; poor 0; very poor 0
       Stable    Adequate high-quality evidence and high level of consensus  Adequate high-quality evidence and high level of consensus
Recent trends  Improving  Stable Confidence  Adequate high-quality evidence and high level of consensus
 Deteriorating  Unclear  Limited evidence or limited consensus
 Evidence and consensus too low to make an assessment
Grades  Very good: Air quality is considered very good, and air pollution poses little or no risk
 Good: Air quality is considered good, and air pollution poses little or no risk
 Fair: Air quality is acceptable. However, there may be a health concerns for very sensitive people
 Poor: Air quality is unhealthy for sensitive groups. The general population is not likely to be affected in this range
 Very poor: Air quality is unhealthy, and everyone may begin to experience health effects. People from sensitive groups may experience more serious health effects

NEPM = National Environment Protection Measure; PM10 = particulate matter smaller than 10 micrometres

Note: Wagga Wagga assessment based on 2001–08; West Mackay based on 2000–08; Port Pirie based on 2003–10; Geelong South and Bunbury based on 1999–2008; Launceston based on 2006–08

3.1.3 Indoor air quality

Like citizens of other highly urbanised societies, most Australians spend more than 90% of their time indoors, leading to concern about the possible impacts of indoor air quality on our health. Such concern is heightened in situations where indoor pollutant concentrations equal or exceed outdoor levels and indoor exposure becomes the dominant form of exposure.150

Symptoms associated with poor indoor air quality can range from acute to chronic, and from mild and generally nonspecific (eye, nose and throat irritation, and headaches and dizziness) to severe (asthma, allergic responses and cancer risk).95,151-152 Despite the potentially significant health effects of indoor air, data on indoor air quality in Australia are limited, providing no firm basis upon which to form assessments of overall status and trend.

Until recently, there has been no comprehensive study of indoor air quality in typical Australian dwellings; previous studies tended to focus on situations with particular air quality issues, such as emissions from unflued gas heaters and gas cooking appliances. The release in 2010 of a two-part report by CSIRO and the Bureau of Meteorology of 40 typical homes in Melbourne has filled that gap, at least for temperate urban areas.153 The study found concentrations of indoor air pollutants to be either lower than or comparable with concentrations found in previous Australian studies. The study showed weekly average concentrations of carbon dioxide, carbon monoxide, nitrogen dioxide, formaldehyde, other carbonyls, BTEX (benzene, toluene, ethylbenzene and xylene) and total VOCs to be higher indoors than outdoors, whereas PM10, ozone and fungi concentrations were higher outdoors. Across the 40 dwellings, the ambient 24-hour NEPM advisory reporting standard for PM2.5 was equalled or exceeded on 3% of days. In dwellings that relied on gas appliances for cooking, levels of carbon dioxide, carbon monoxide, nitrogen dioxide, PM2.5, formaldehyde, benzene and total VOCs were significantly higher than in households that solely used electric cooking appliances. The effect of proximity to major roads on indoor air quality was limited to an increase in nitrogen dioxide levels (accounting for around 20% of indoor nitrogen dioxide in these situations).153 Unfortunately, although the study has significantly expanded our knowledge of indoor air quality in Australian homes, as the authors note in a separately published overview of the study, ‘the absence of specific guidelines for indoor air quality in Australia prevents an objective assessment of the quality of observed indoor air’.154