Air toxics and indoor air quality in Australia

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

Part A: Air toxics

Air toxics overview

1.1 Community concerns

Air pollution continues to be the environmental problem of greatest concern for Australians. In 1999, an Australian Bureau of Statistics (ABS) survey found that 29% of Australians reported air pollution as their major concern, consistent with previous surveys (32% response in 1998 and 31% in 1996). Overall, in 1999, 69% of people reported having environmental concerns, compared with 71% in 1998.

These findings should come as no surprise since we all need to breathe air 24 hours a day to survive. The other essentials for life – food and water – have an element of choice not found with air. If concerned about contaminated food or water, the consumer has the option to defer consumption or buy organic vegetables, water filters, etc. There is no such option when it comes to air quality.

Exposure to air toxics can affect health, with effects ranging from none, through mild and immediate (eg watery eyes), to more extreme (eg lung damage, nervous system damage or even birth defects and cancer). The extent to which these adverse effects present themselves depends on a number of factors such as the type of air toxic to which a person is exposed and the length and severity of the exposure.

1.2 Definitions of air toxics

The terms 'air toxics' and 'hazardous air pollutants' (HAPs) may be used interchangeably. While there is no universally accepted definition for air toxics, the more recognised definitions share a number of common elements. In general terms, air toxics are pollutants present in the atmosphere in low concentrations that are known or suspected to cause serious health or environmental problems.

1.2.1 OECD

The Organisation for Economic Co-operation and Development (OECD) has used the following definition for air toxics:

… gaseous, aerosol or particulate contaminants present in the ambient air in trace amounts with characteristics (toxicity, persistence) so as to be a hazard to human health, plant and animal life. (OECD 1995)

In establishing this definition, the OECD noted that a comprehensive approach to the air toxics problem would have to take into account many different types of pollutants in order to evaluate total exposure and risk. This would require consideration of chemical agents, physical agents like radionuclides, ionising and non-ionising radiation, and biological agents such as microorganisms. For practical purposes, the OECD definition narrowed its focus to chemical agents only.

The OECD grouped chemical pollutants as follows:

More information on OECD approaches to air toxics is given in Section 2.2.

1.2.2 United States EPA

In 1970, HAPs were formally introduced into the United States Environmental Protection Agency (US EPA) Clean Air Act. In the current version of the act, section 112(a)(6) defines HAPs by referring to a list of 1881 substances that have been identified as HAPs (see Table 2.4). However, a subsequent section of the act, section 112(b)(5)(2), concerns the requirements for adding substances to the HAPs list. It identifies HAPs as:

… pollutants which present, or may present, through inhalation or other routes of exposure, a threat of adverse human health effects (including, but not limited to, substances which are known to be, or may reasonably be anticipated to be, carcinogenic, mutagenic, teratogenic, neurotoxic, which cause reproductive dysfunction, or which are acutely or chronically toxic) or adverse environmental effects whether through ambient concentrations, bioaccumulation, deposition, or otherwise... as a result of emissions to the air.

Another part of section 112 of the Clean Air Act excludes consideration of the criteria pollutants as HAPs. The criteria pollutants are discussed further in Section 1.2.4.

A more straightforward definition is that used by the Californian Environmental Protection Agency (EPA) Air Resources Board which defines 'toxic air contaminant' as:

… an air pollutant which may contribute to an increase in mortality or in serious illness, or which may pose a present or potential hazard to human health.

The US EPA also aggregates HAPs into four general categories (which differ from the OECD classes), namely:

More information on the US EPA's approaches to management of HAPs is given in Section 2.3.

1.2.3 Australia

The Commonwealth's State of the Environment report (DEST 1996) defined air toxics as:

… pollutants present at very low concentrations, known to cause or suspected of causing long-term health effects in humans.

For the purposes of the ATP, the following definition will be used:

Air toxics are gaseous, aerosol or particulate pollutants (other than the six criteria pollutants) which are present in the air in low concentrations with characteristics such as toxicity and persistence so as to be a hazard to human, plant or animal life. The terms 'air toxics' and 'hazardous air pollutants' (HAPs) are used interchangeably.

The ATP definition for air toxics was originally considered by the Air Toxics Forum2 , and later agreed by the TAG and the SG (see Appendix C). It is based on the OECD definition but differs in that it specifically excludes the criteria pollutants, which are discussed in Section 1.2.4.

The use of mutually exclusive definitions for criteria pollutants and air toxics by regulators is a fairly common practice. The abundance and broad range of sources of the criteria pollutants require that they be given special attention by regulators to ensure good air quality. In Australia, the criteria pollutants are specifically addressed by the Ambient Air Quality NEPM. However, four of the six criteria pollutants (ie carbon monoxide, lead, nitrogen dioxide and particles) have particular significance to the indoor environment and will be considered with the air toxics in addressing indoor air quality issues (see Part B of this report).

1.2.4 Criteria pollutants

The US EPA (1970) defines criteria pollutants as those pollutants that:

… cause or contribute to air pollution that may reasonably be anticipated to endanger public health or welfare ... the presence of which in the ambient air results from numerous or diverse mobile or stationary sources.

In Australia, six criteria pollutants have been identified, namely carbon monoxide, lead, nitrogen dioxide, photochemical oxidants3 (measured as ozone), particles as PM10 (particles of 10 micrometres or less) and sulfur dioxide. The same set of six criteria pollutants are used by the World Health Organization (WHO), which refers to them as the 'classical' air pollutants, and US EPA. Under normal circumstances, the criteria pollutants are typically found in significantly higher concentrations than air toxics in ambient air.

The criteria air pollutants require special focus due to their abundance in the environment and their demonstrated adverse health effects. Thus, it is common practice for regulatory authorities to establish standards for ambient concentrations for the criteria pollutants, whereas guidelines, goals and targets are typically used for air toxics.

Having established national standards for the six criteria pollutants in ambient air, Australian environmental authorities have the opportunity to focus on the development of a national approach for addressing priority air toxics.

1.3 Broad categories of pollutants of concern

The following subsections provide a brief overview of a range of broad pollutant categories of relevance to air toxics. More detailed information on air toxics of significance to the indoor environment is given in Part B and on specific air toxics in Part C.

1.3.1 Metal compounds

Among metals, those belonging to the group 'heavy metals' are of particular concern due to their potential to harm human health and the environment. The Australian State of the Environment report (DEST 1996) defines heavy metals as:

… metallic elements with relatively high atomic mass (over 5.0 specific gravity), such as lead, cadmium, arsenic and mercury; generally toxic in relatively low concentrations to plant and animal life.

Heavy metals tend to be toxic at relatively low concentrations, persistent and bioaccumulative. Certain heavy metals, such as arsenic, can display both chronic and acute human toxicity and they may also be carcinogenic. Unlike most persistent organic pollutants (POPs, Section 1.3.5), metals occur naturally in the environment and as they are elements they do not degrade but can change form into various metal compounds. Figure 1.1 illustrates the factors that affect emissions of heavy metals and their ambient levels, the levels of human exposure to heavy metals and the hazards associated with heavy metals.

Figure 1.1: Factors affecting emissions, ambient levels, human exposure and hazards of heavy metals.4

Figure 1.1: Factors affecting emissions, ambient levels, human exposure and hazards of heavy metals.

When listing metals, environment regulators generally make no distinction between the elemental form of a metal and a metal compound. For example, the list of priority metals in the United Nations Economic Commission for Europe (UNECE) Convention on Long-Range Transboundary Air Pollution5 makes no distinction between metal species. Although detailing all compounds for a metal simplifies the listing process, it fails to recognise the wide differences in toxicity between different compounds of the same metal and the fact that many are non-toxic.

Knowledge of the species (compound or form) of a metal in a given environment is critical in assessing its toxicity as different species of the same element will have different toxicological properties. For example, bound or complexed metals are generally considered to be less toxic than their free ionic forms.

Although generally toxic at higher doses, traces of some metals are essential in the diet. Nickel, chromium, copper, zinc, selenium, cobalt and arsenic appear to be essential elements in many mammals. Lead, mercury, cadmium, thallium and uranium have no known physiological function. Similarly, some metals are essential to plants in low doses but are toxic at higher doses.

Heavy metals exist naturally in the environment and can be released in a number of ways, for example, by erosion, volcanic activity, or as windblown dust. Forest fires and sea spray are sources that can release small amounts of heavy metals. Anthropogenic (human-made) sources include mining and refining operations, and pulp mills. Most heavy metals are released into the atmosphere as particles, which can travel considerable distances before returning to the ground in rain and snow.

Once fully implemented, the NPI will require reporting on emissions of the following metals:

Not all of the metals listed above (eg selenium, organo-tin, and chromium III compounds) are a priority concern in ambient or indoor air. The list of priority pollutants developed in consultation with the TAG and the SG for the Air Toxics Program includes the following metals and metal compounds:

The final list and an overview of the ranking process are provided in Chapter 5. Details of the prioritisation process applied by the TAG are given in Chapter 5.

As part of the ATP, Environment Australia has recently commissioned WA DEP to undertake a review of data on the sources of heavy metal emissions. This project will collect and review existing studies and analyse any raw monitoring data held by State environment agencies. More information on this project is given in Section 1.4.9.

1.3.2 Pesticides

Pesticides are of potential interest to the ATP due to their physicochemical properties, particularly volatility, and the common methods used for the application of pesticides (eg ground and aerial spraying).

Pesticides, or more broadly agricultural and veterinary (agvet) chemicals, are used widely in Australia. They have brought long-term benefits to Australian agriculture, including farming, forestry, horticulture and aquaculture, and to the national economy. Their application has reduced the impact of weeds, pests and diseases on agricultural and forest production, leading to improved productivity, better quality produce and more competitive primary industries. Veterinary chemicals and drugs protect animals from suffering and death due to parasites and diseases, much like the pharmaceutical agents used in human medicine. Agvet chemicals also have a role in the wider community – for controlling pests and weeds in urban areas and home gardens, and in the management of similar problems in national parks and nature reserves. From time to time, they also help address serious threats to human health, such as their use for broad-scale control of insects that are vectors (transport agents) for diseases, for example, mosquitoes carrying malaria and Ross River fever (ARMCANZ 1998).

Agvet chemicals can have unwanted side effects, particularly if used improperly. The very properties that make them effective in treating pest infestations or noxious weeds, for example, may lead to agvet chemicals having harmful effects on useful crops or beneficial insects and invertebrates (eg bees and worms). Agvet chemicals may cause illness in either humans or animals if they are ingested, inhaled or otherwise absorbed. They may contaminate food, fibre or animal feed, rendering them unsaleable or unsuitable for human or animal use. The chemicals may also contaminate water or accumulate in the environment, with longer-term damaging effects (ARMCANZ 1998).

The best known and most studied mechanism for the unintended contamination (sometimes referred to as 'off-target contamination') of food and drinking water from pesticides is spray drift. The extent of chemical drift will depend on a number of variables such as formulation, prevailing weather conditions, type of spray nozzle and resultant droplet size. Also of particular significance is the application method, which can usually be divided into aerial, orchard and ground boom spraying. Of these, the aerial method has the greatest drift potential, particularly if small droplet sizes are used as is popular in Australia.

Once released to the environment, chemicals are generally not confined to the target. On release, chemicals will tend to partition into environmental compartments, namely, air, water, soils (including dust), sediments and biota. The extent of movement into an individual compartment depends on the chemical's properties. Transfer to the atmosphere tends to be a transitional phase but it provides a significant transport pathway for the distribution of chemicals. With pesticides, such transfer is normally facilitated by the initial application as a spray droplet. The pesticide then moves into the air in the form of vapour from either evaporation or sublimation from plants, water and soils. Pesticides may also be transported in dissolved forms (eg through fog and rain) and by adsorption to particles such as dusts.

Before agvet chemical products can be sold, supplied, distributed or used in Australia, they must be registered by a Commonwealth agency, the National Registration Authority for Agricultural and Veterinary Chemicals (NRA). Registration offers farmers, food producers, the chemical industry and the general public assurance that chemicals on the market are efficacious, suitably formulated and labelled, and safe for humans and animals when used as directed. It also aims to ensure that agvet chemicals do not have unintended adverse effects on the environment and do not leave unacceptable residues in produce marketed in Australia and overseas.

To gain access to the Australian market, agvet chemicals must go through the NRA's rigorous assessment process to ensure that they meet high standards of safety and effectiveness. Any changes to a product that is already on the market must also be referred to the NRA. The National Registration Scheme requires that companies supply the NRA with extensive data about new and existing chemical products. These chemicals are evaluated independently by the Department of the Environment, the Department of Health and Aged Care and the National Occupational Health and Safety Commission (NOHSC). Independent evaluation of agvet chemicals determines that products are safe for people, animals and the environment and will not pose any unacceptable risk to trade with other nations. If a product meets the NRA's standards it may be registered for use in Australia. The NRA also reviews products that have been on the market for many years to ensure that they meet contemporary standards.

The NRA manages a national compliance program to ensure that products supplied in Australia continue to meet the conditions of registration. The responsibility for control of use lies with the relevant agencies at the State and Territory level. State government legislation controlling and ensuring compliance in chemical use aims to promote correct usage; it covers factors such as spray drift and misuse. State governments have adopted different legislative approaches for controlling the use of chemicals. In addition, hazardous substances legislation has been introduced by State and Territory Governments to improve the provision of information on hazardous chemicals and to control their use in the workplace.

Where certain agvet chemicals pose special risks and/or require special skills and training in their handling, State governments have developed licensing schemes to cover specialist operators such as fumigators, pest controllers and aerial applicators. The aim is to minimise risk to operators and bystanders by ensuring that access to such chemicals is restricted to those with the necessary competencies and equipment.

A number of new initiatives for addressing agvet chemicals are summarised below.


ChemCollect, a one-off initiative of the Australia and New Zealand Environment and Conservation Council (ANZECC), is a nationally coordinated, free collection scheme to ensure that existing unwanted and deregistered agvet chemicals are safely collected from rural areas and destroyed in a socially and environmentally acceptable manner.

The program will run for approximately three years and is expected to be completed by the end of 2002 (completion dates will vary between jurisdictions). The program is particularly focused on the removal of persistent organochlorine pesticides.

Participating State and Territory governments will match the Commonwealth's contribution on a dollar for dollar basis. (The ACT will not be participating as it has already conducted a similar chemical collection.)

To ensure that stocks do not build up again, the agriculture industry has agreed to institute ChemClear – an ongoing program for regular collections of registered farm chemicals that are otherwise non-returnable. ChemClear will begin after ChemCollect has finished in each State. ChemClear is a joint initiative involving Avcare (National Association for Crop Protection and Animal Health), the Veterinary Manufacturers and Distributors Association (VMDA) and the National Farmers Federation (NFF). The program is an example of industry's increasing recognition of its 'cradle to grave' stewardship of its products.

Industry waste reduction scheme

ChemCollect and ChemClear are complemented by the agvet chemical 'Industry Waste Reduction Scheme'. This scheme concerns the estimated four million nonreturnable agricultural chemical containers sold every year to Australian farmers. It has two objectives:

The scheme aims to reduce container packaging by 32% and reduce the amount disposed of in landfill by 68% by 2001.


The national DrumMUSTER program aims to collect and recycle empty, cleaned, non-returnable agricultural chemical containers. Agsafe manages the program for the NFF, Avcare, the VMDA and the Australian Local Government Association. DrumMUSTER is funded from a levy on crop protection and on farm animal health products sold in nonreturnable chemical containers of over one kilogram in content. Containers will be refilled, recycled or shredded for landfill. The program aims to recover 66% of these chemical containers and to supply 50% of raw materials in recyclable or returnable packaging.

Collection of data on the use of agricultural and veterinary chemicals

In December 1999, ANZECC endorsed a plan to define and undertake a feasibility study into the collection of data on the use of agvet chemicals.

The Commonwealth's Bureau of Rural Sciences (BRS) is in the process of engaging a consultant to recommend a strategy and methodology to assess the quantity of agvet chemicals used in Australia. The BRS-commissioned report could form the basis for a pilot study to test proposed data collection methods.

A more detailed description of the management framework for agvet chemicals is given in Appendix A.

1.3.3 Polycyclic aromatic hydrocarbons (PAHs)

The following text has been extracted from the report State of Knowledge Report on PAHs in Australia prepared by WA DEP on behalf on Environment Australia (WA DEP 1999). The full text of this report is available at the air toxics website7.

Polycyclic aromatic hydrocarbons (PAHs), a complex class of organic compounds, consist of two or more fused aromatic rings, and contain only carbon and hydrogen atoms. The physical and chemical properties of PAHs are determined by their conjugated p-electron systems, which are dependent on the number of aromatic rings and the molecular mass. The smallest member of the PAH family is naphthalene, a two-ring compound, which is found in the vapour phase in the atmosphere. Three to five-ring PAHs can be found in both the vapour and particulate phases in air. PAHs consisting of five or more rings tend to be solids adsorbed onto other particles in the atmosphere.

The terms 'polycyclic aromatic compounds' or 'polycyclic organic matter' are used to include similar compounds with nitrogen, oxygen or sulfur substituents such as nitro-PAHs, hydroxy-PAHs and heterocyclic compounds. Sometimes the term 'polynuclear' is used in the literature instead of 'polycyclic' to describe these compounds. They are not included in the WA DEP report, which covers only PAHs.

Incomplete combustion of organic matter releases PAHs into the atmosphere as a complex mixture of compounds. Wood-burning heaters, agricultural waste burning, motor vehicle exhaust, cigarette smoke and asphalt road and roofing operations are all sources of PAHs. These compounds are widespread contaminants of the environment and a number of them are known or suspected carcinogens. Benzo[a]pyrene (BaP), a widely reported five-ring PAH, is known for its carcinogenic potency. Although hundreds of PAHs have been identified in atmospheric particles (Lao et al 1973; Lee et al 1976), toxicological endpoint and/or exposure data are available for only 33 PAHs.

Table 1.1: Types of PAH compounds identified in Australian cities from 1990 to the present
PAH Totala Brisb Perthc NTd Launce Melbourne Cbri NSWj
MASf DPg Collingh
Acenaphthene 4 Tick Tick M M     Tick   Tick
Acenaphthylene 4 Tick Tick M M     Tick   Tick
Anthanthrene 1     M       Tick    
Anthracene 6 Tick Tick M M Tick Tick Tick   Tick
Anthracene, 2-methyl 1 Tick                
Benzo[a]anthracene 7 Tick Tick M M Tick Tick Tick Tick Tick
Benzo[a]pyrene 9 Tick Tick Tick Tick Tick Tick Tick Tick Tick
Benzo[b]fluoranthene 6 Tick Tick M M   Tick Tick Tick Tick
Benzo[e]pyrene 4 Tick Tick M   Tick   Tick    
Benzo[g,h,i]perylene 6 Tick Tick M M   Tick Tick Tick Tick
Benzo[k]fluoranthene 6 Tick Tick M M   Tick Tick Tick Tick
Chrysene 6 Tick Tick M M Tick Tick Tick   Tick
Coronene 3 Tick   M   Tick   Tick    
Dibenz[a,h]anthracene 6 Tick Tick M M   Tick Tick Tick Tick
Fluoranthene 6 Tick Tick M M Tick Tick Tick   Tick
Fluorene 4 Tick Tick M M     Tick   Tick
Indeno[1,2,3-c,d]pyrene 5 Tick Tick M M   Tick Tick   Tick
Naphthalene 4 Tick Tick M M     Tick   Tick
Naphthalene, 2-methyl 2 Tick Tick              
Perylene 4 Tick Tick M   Tick   Tick    
Phenanthrene 6 Tick Tick M M Tick Tick Tick   Tick
Pyrene 7 Tick Tick M M Tick Tick Tick Tick Tick
Vanillan, 3-methoxy-4-hydroxybenzaldehyde 1   Tick              

a Total = number of studies reporting the concentration of the PAH compound.
b Brisbane (Muller JF et al 1998, 1996a, 1995a,b).
c Perth (Gras 1996).
d Northern Territory (Vanderzalm et al 1998).
e Launceston (Expert Working Party 1996).
f Melbourne Aerosol Study (Gras et al 1992).
g Debney's Park at Flemington (VicRoads/EPA 1991).
h Collingwood (Panther et al 1999).
i Canberra (Ian Fox 1999).
j Industrial self-monitoring data collected under conditions of NSW EPA discharge licences (NSW EPA 1996; 1997).
Tick = Data reported for compound;
M = Compound studied, concentration not reported;
PAH = polycyclic aromatic hydrocarbon

During the past 10 years, PAHs have been studied in various Australian cities. The studies involved identification and quantification of several PAH compounds in samples collected from the air environment.

The types of PAH compounds identified in Australian cities from 1990 to the present are listed in Table 1.1. The total number of 'Tick' entries indicates the number of times the concentration of any PAH has been reported. 'M' entries represent PAH studies in which concentrations were not reported.

1.3.4 Volatile organic compounds

The State of the Environment report (DEST 1996) defined VOCs as organic compounds with boiling points between 50°C and 260°C. The term encompasses a very large and diverse group of carbon-containing compounds, including aliphatic, aromatic and halogenated hydrocarbons; aldehydes; ethers; esters; acids; alcohols and ketones. Examples of VOCs include benzene, toluene, chlorofluorocarbons (CFCs), halons, carbon tetrachloride and some pesticides.

Burning fuels containing carbon (eg gasoline, oil, wood, coal, natural gas, etc), solvents, paints, glues and other products used at work or home releases VOCs. Motor vehicle emissions are also an important source of VOCs.

Environment authorities are particularly concerned about VOC emissions because of their potential to contribute to the formation of ground-level ozone and to global warming. The potential contribution that a VOC makes to ozone formation and global warming depends on its photochemical reactivity. Fortunately, most VOCs rapidly break down in the atmosphere and so have short residence times, thereby limiting their contribution to ozone formation.

The US EPA has recognised that some VOCs have such low reactivity that they do not have an appreciable impact on the formation of ozone. It defines a VOC as:

… any compound of carbon, excluding carbon monoxide, carbon dioxide, carbonic acid, metallic carbides or carbonates, and ammonium carbonate, which participates in atmospheric photochemical reactions.

This EPA definition is linked to an appendix that identifies organic compounds excluded from the definition of VOCs on the basis that the US EPA has found that they have negligible photochemical reactivity.

Australia's National Health and Medical Research Council (NHMRC) has accepted the WHO definition of VOCs:

… all organic compounds in the boiling range of 50–260 degrees C, excluding pesticides.

This definition is more encompassing than that of the US EPA as it does not exclude any compounds on the basis of their low photochemical reactivity.

The extent to which VOCs can cause health problems depends on their toxicity, concentration and the duration of personal exposure. Chronic low-level exposure to some VOCs has been associated with adverse health effects, an example being the link between benzene and leukaemia. VOCs are of particular concern because they are not only toxic to humans but also can contribute to the formation of ozone and photochemical smog when other chemicals are present in the air. Certain VOCs (eg ethylene) may also harm plants.

1.3.5 Persistent organic pollutants

POPs are toxic and environmentally persistent substances that can be transported between countries by the earth's oceans and atmosphere8. The substances bioaccumulate and have been traced in the fatty tissues of humans and other animals. Due to their presence in the atmosphere in low concentrations, POPs may be considered to be air toxics.

On 23 May 2001, Australia and 90 other countries signed the Stockholm Convention, an international agreement on POPs developed under the auspices of the United Nations Environment Programme (UNEP). The Stockholm Convention sets out a variety of control measures to reduce and eliminate POPs releases, including bans on production, import, export, and use of some POPs. The Convention also includes obligations for the use of best available techniques and practices for some industrial production processes so that emissions of by-product POPs are reduced and minimised9.

The control measures will apply to an initial list of twelve POP chemicals: aldrin, chlordane, dichlorodiphenyltrichloroethane (DDT), dieldrin, dioxins, furans, endrin, hexachlorobenzene (HCB), heptachlor, mirex, polychlorinated biphenyls (PCBs) and toxaphene. The Convention also includes a procedure for taking action against additional POPs in the future.

Australia is well advanced in meeting the potential obligations of the Convention because all of the commercially produced POPs have been banned domestically or are being phased out. In addition, Australia has well-established national plans in place under the Australia and New Zealand Environment Conservation Council (ANZECC) to manage POPs waste.

One POP, mirex is still used under licence in small quantities in tropical areas to control giant termites (Mastotermes darwiniensis) but research is well advanced in identifying a suitable alternative. Consistent with the POPs agreement, which allows for some limited exemptions, Australia has lodged a limited exemption for Mirex use.

In regard to the management of dioxins and other unwanted by-products, environmental agencies in Australia are working together to develop a national dioxin program for ANZECC to consider in June 2001. The Commonwealth Government has provided AUS$5 million over four years, commencing 2001–02, to manage the impacts of dioxins and related toxic combustion by-products.

1.3.6 Dioxins and furans

Of particular relevance to the ATP is the group of chemicals known as dioxins and furans. A report released by Environment Australia in 1998, Sources of dioxins and furans in Australia: Air Emissions10 defines a dioxin as 'any compound containing the dibenzo-p-dioxin nucleus' and a furan as 'any compound containing the dibenzofuran nucleus'.


Polychlorinated dibenzo-p-dioxins and polychlorinated dibenzofurans (PCDDs and PCDFs, often referred to as PCDD/Fs or simply dioxins) are not produced intentionally but are highly toxic, unintended by-products released in small quantities from some human activities (combustion, steel works, waste incineration, and certain types of chemical manufacture) or through such natural activities as bush fires and volcanic activity. Almost all of the possible 210 congeners11 are released from these sources. Due to chemical, physical, and biological stability and long-range transport, they are ubiquitous and have been detected in all environmental compartments in trace amounts. The persistence of the 2,3,7,8-substituted congeners and the lipophilicity (solubility in fats) of these compounds means that dioxins and furans accumulate in fatty tissues and in carbon-rich matrices such as soil and sediments.

Dioxins and furans exhibit the biological effects commonly associated with chlorinated organic chemicals. The observed health effects in humans are dependent on the level and duration of exposure as well as the susceptibility of the individual. In humans, acute dioxin exposure results in increased risk of severe skin lesions (a condition known as chloracne), altered liver function and lipid metabolism, depression of the immune system and endocrine and nervous system abnormalities. There is also evidence that exposure to these compounds at low levels may be linked with effects on the cardiovascular system (such as myocarditis, myocardial infarctions and rapidly progressive atherosclerosis), type II diabetes, developmental effects in children (such as growth retardation and cognitive and behavioural effects), endometriosis and cancer.

In 1997, the International Agency for Research on Cancer (IARC) classified the most toxic of the PCDD/PCDF congeners, 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD), as a Group 113 compound because of its carcinogenicity to humans.

In 1998, following a review of new epidemiological data, the WHO agreed to a downward revision of its tolerable daily intake (TDI) guideline to a level of 1–4 toxic equivalent (TEQ)12 picograms per kilogram (pg/kg) bodyweight (the previous 1990 WHO TDI guideline was 10 TEQ pg/kg). WHO experts recognised that in developed countries current background levels in the general population of 2–6 TEQ pg/kg bodyweight may already be having an adverse affect on health. Noting this, WHO recommended that every effort should be made to reduce exposure to the lowest possible level (WHO 1998).

A significant area of concern for human health is linked to the relatively high exposure of breast fed infants. Studies of nursing mothers in Europe have shown that, despite decreases in overall concentrations of dioxins and furans in breast milk since the late 1980s, infants are still receiving doses significantly over the TDI, noted above.

Biological uptake of dioxins

Human exposure to background contamination with dioxins and furans is possible via several routes:

The US EPA released a draft reassessment report on dioxins in September 200014. This long-awaited report is widely recognised as likely to be the definitive statement on dioxins once finalised. The report was open for public comment and has been the subject of a detailed peer review by the US EPA's Science Advisory Board (an independent panel of experts). It is anticipated that the final amended version of this report will be published in early 2002.

The draft US EPA report classifies for the first time the most potent form of dioxin, 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD), as a human carcinogen. More than 100 other dioxin-like compounds are classified as 'likely' human carcinogens. In the past, dioxins have been linked to several cancers in humans, including lymphomas and lung cancer. Now, the report links low-grade exposure to dioxins to a wide array of other health problems, including changes in hormone levels and developmental defects in babies and children.

While the major transport pathway of dioxins from their point of formation to the environment is through air emission, human exposure is via direct ingestion. The WHO estimates that up to 90% of human exposure to dioxins occurs through food consumption. Dioxins enter the food chain when animals eat contaminated plants (the plants being contaminated by atmospheric deposition of dioxins). Dioxins are then accumulated in the fat of mammals and fish.

The draft US EPA report indicates that, for a small segment of the population which eat large amounts of foods rich in fat (eg. meats and dairy products) that are relatively high in dioxins, the odds of developing cancer could be as high as 1 in 100. The report also concludes that the dioxin intake of children is proportionally much higher than that of adults because of the presence of the chemical in dairy products and breastmilk.

Identification of sources of dioxins entering environment

The cycling of dioxins through the environment is a complex process, involving multiple sources (natural and anthropogenic), flows, reservoirs (temporary storage) and sinks (long-term storage).

Key sources are:

International and national research and information gathering suggests that there is good agreement on the identification of the likely primary sources of anthropogenic emissions, both industrial and domestic.

A UNEP Dioxin and Furan Inventory, compiled in May 199915, categorised anthropogenic sources into nine major industry sectors:

The UNEP inventory showed that, whilst many countries, including Australia, have estimated dioxins emissions to air, few findings are available for emissions to water and soil.

Some of the emission controls installed in recent years have reduced formation of dioxins; others have removed dioxins from air and water emissions and transferred them to solid residues.

The major difficulty in obtaining accurate and widespread information regarding emissions of dioxins is quantification. The analysis required for quantification is expensive and because of the range of dioxins, a multiple sampling regime is needed to characterise the nature of emissions adequately. One example is the production of dioxins during incineration, which is highly dependent on the operating conditions. The levels of production at optimum conditions may not be representative of releases under other conditions.

Analysis of dioxins requires sophisticated methods that are available only in a limited number of laboratories around the world. According to the WHO factsheet on dioxins, only about 100 laboratories are able to analyse dioxins in environmental samples (eg air, ashes, soil, water) and in food. About 20 laboratories in the world are able to reliably measure dioxins in biological materials (eg human blood or breast milk). These laboratories are mostly in industrialised countries. Costs vary according to the type of sample, but range from US$1200 for the analysis of a single biological sample to in excess of US$10 000 for the assessment of release to air from a waste incinerator.

The Australian Government Analytical Laboratories (AGAL) can currently test for 2,3,7,8-TCDD contamination to 1 mg/kg (equivalent to 1 part per billion [1 ppb]) in soil and in food samples (eggs, cheese, chocolate, biscuits and meat fat). The method also gives an indication of the presence of other dioxins, furans and PCBs. Methods for testing other media (eg air) are currently being developed.

AGAL has also established an ultra-trace laboratory in Sydney. This facility uses extensive sample cleanup procedures and high resolution gas chromatography/mass spectrometry (GC/MS) to detect dioxins, furans and co-planar and mono-ortho-substituted PCBs at their native levels of 0.1–1 ng/kg (equivalent to 1 part per trillion [1 ppt]) level. This capacity is comparable to European, Asian and North American facilities. NATA accreditation has been obtained for the initial target matrices of soils, sludge, and water, and it is anticipated accreditation will soon be granted for food (milk, meat, fish, eggs and butter) and biota followed by air (emission and ambient) and blood. The facility takes part in international quality control programs to assure worldwide comparability of results.

The prohibitive cost of analytical methods has resulted in dependence on estimation of emissions using emission factors and there has been reluctance to confirm estimates with extensive testing programs.

The cost of testing and the need for repeated testing under a variety of conditions raise questions about monitoring for improvement. It may be useful to consider using proxies for human and environmental health such as body burdens in breast milk and serum.

Dioxins in Australia

The Environment Australia study Sources of Dioxins and Furans in Australia: Air Emissions (1998a) estimates that aggregate dioxin emissions in Australia range from a minimum of 150 g TEQ per year up to a maximum of 2300 g TEQ per year. Using the minimum value, Australia's emissions represent 1.4% of the total emissions from 15 OECD countries (UNEP 1999), increasing to around 10% if maximum figures are used. Australia's emission profile is different from other countries in terms of type of emissions and location of sources. For example, around 70% of OECD total emissions come from municipal waste incineration, whereas Australia currently has none from this sector (we currently use landfill for disposal of waste). Of emissions in this sector, Japan and the United States have a large majority share due to their looser emission standards compared to their European counterparts. The distribution of Australia's emission sources also appears to be different; unlike other countries, urban centres do not represent a large share of our emissions.

Prescribed burning, bushfires, residential wood fires, sinter production, coal and oil combustion, metal production, medical waste incinerators and cement production have been identified as the key Australian sources (Environment Australia 1998a). These findings, however, are only estimates. Lack of local data necessitates the use of overseas emission factors, so results are uncertain. For example, the Australian cement industry is conducting studies to confirm that its emissions are more in line with low United Kingdom emissions rather than the higher United States levels which had been used in developing the Australian emission inventory. Further work has also been commissioned in relation to the incinerator and metal processing industries.

In the future, more data will become available as a result of the NPI, which requires industrial facilities releasing chemicals on the NPI reporting list to estimate and report annually on their emissions of those substances to air, land and water. Reporting on dioxins will be required by 2002 for those industries that meet the NPI reporting threshold for dioxins (ie those burning 2000 tonnes or more of fuel/waste or consuming more than 60 000 megawatt hours of energy in a reporting period.

1.3.7 Asbestos

The summary of information on asbestos presented in this section has been based on the NICNAS report Chrysotile Asbestos – Priority Existing Chemical No. 9 – Full Public Report, published in February 1999. The NICNAS report focuses on chrysotile (white asbestos) as it continues to be imported into Australia in the raw mineral form and articles containing chrysotile are produced locally and imported. Full copies of this report are available from NICNAS.

Asbestos is defined as the fibrous form of mineral silicates belonging to the serpentine and amphibole groups of rock-forming minerals. The most common asbestos types are chrysotile, amosite (brown asbestos) and crocidolite (blue asbestos).

Chrysotile is unequivocally a human carcinogen; however the risk to the public associated with its continued use is dependent on the nature of the material to which the public is exposed and the level, frequency and duration of exposure.

The most prevalent chrysotile-induced disease is lung cancer. An assessment of the likely risk from this hazard to the public, from current sources of exposure, will provide a qualitative indication of the likely risks for mesothelioma (a form of lung cancer associated with asbestos). At the levels of exposure likely to be encountered by the public, the risk of asbestosis, which follows a dose-response relationship, is essentially zero (IPCS 1986).

In Australia, raw chrysotile is currently fabricated into friction materials and gasket sheeting in three locations and into epoxy-resin adhesive at another site. In general, manufacturing processes are essentially carried out in enclosed systems and dust levels are controlled by dust extraction with automatic plant shutdown when dust levels exceed prescribed limits. Monitoring of manufacturing and processing activities has revealed personal exposure levels generally of < 0.1 fibres per millilitre (f/mL), with most samples ≤ 0.05 f/mL. The national exposure standard for chrysotile (currently under review) is 1 f/mL (NOHSC 1988). Significant exposure of the public to chrysotile fibres from these manufacturing processes is, therefore, unlikely.

Automotive applications are likely to be the major source of public exposure to chrysotile dusts. Chrysotile friction materials (eg clutch plates, brake pads and brake shoes) contain 40–60% chrysotile. A proportion of the end-use products containing chrysotile may be sold directly to the public, particularly automotive friction products and gaskets.

Home mechanics are likely to wear little if any personal protective equipment when replacing brake pads and shoes, clutch plates or engine gaskets. In the case of gaskets, significant quantities of dust are unlikely to be generated as the chrysotile is bound into the matrix of the gasket. Similarly, dusts created from clutch facings tend to be enclosed in the transmission of the vehicle and most replacement clutch facings do not contain chrysotile. During the changing of brake pads and drum shoes, however, significant exposure is possible. In commercial operations, compressed air is generally no longer used to remove excess dust. Improved housekeeping practices have reduced exposure levels occupationally and, as a consequence, reduced the likelihood of public exposure from this source. The home mechanic, however, may have significant intermittent exposure during the changing of brake pads and shoes.

Generation of chrysotile dusts at busy traffic intersections, by braking vehicles, is a known source of public exposure. A study of the levels of chrysotile fibres at two busy (approximately 2000 vehicles/hr) London intersections found total asbestos levels of between 5.5 x 10-4 to 6.2 x 10-3 f/mL (Jaffrey 1990). Of the fibres detected, less than 10% had dimensions within the peak hazard range (< 5 µm long by < 3 µm wide) prescribed by WHO (Spurney 1995). Another study carried out in Australia found airborne asbestos levels to be very low (0.5 particles/mL) in the immediate vicinity of the intersection braking area of the southeast exit of the Tullamarine freeway (Alste et al 1976). The particles consisted of small bundles of fibres and the number of fibres in the bundle was not determined. The majority of fibres had a maximum dimension of ≤ 2 µm and the crystal structure of the fibres was unchanged. At a different location (30 m from the nearest traffic), levels were below the limits of detection (Alste et al 1976). Therefore, exposure from this source is likely to be highly localised and intermittent for most people.

The use of chrysotile in industrial gaskets, such as in petrochemical plants, is unlikely to yield significant public exposures. Similarly, the use of 'sag resistant' epoxy resins (containing approximately 2% chrysotile) in building and construction work, is unlikely to lead to significant public exposure as the chrysotile is bound into the adhesive matrix. These applications are generally limited to large industrial complexes or commercial buildings, use relatively small volumes of chrysotile at any one location, are likely to enclose the material between metal or masonry surfaces and would not generally be expected to produce significant quantities of free chrysotile fibres.

Waste chrysotile, the polyethylene bags in which it is supplied, and chrysotile containing materials from the manufacturing process are disposed to landfill by licensed disposal contractors. As chrysotile fibres are unlikely to be mobile in the soil or water table, landfilling is not inappropriate from a public health perspective.

As the major source of public exposure to chrysotile is that generated from brake linings in commercial and private vehicles, exposure is likely to be widespread and to occur through oral, dermal and inhalation routes, with inhalation and dermal contact predominating.

Levels of exposure will vary widely, with rural residents expected to have the lowest exposure levels and those living or working adjacent to busy intersections having the highest exposures. However, the levels of exposure at peak generation points such as traffic intersections remain low in absolute terms, and tail off rapidly as measurements are taken further from the point of generation, so that cumulative exposures for the bulk of the population are expected to be low.

The majority of new cars manufactured and imported into Australia no longer include asbestos-containing components. As a consequence, the generation of chrysotile-containing dusts from this source will gradually decline as the proportion of chrysotile-free vehicles increases. The general decline in use of chrysotile-containing parts is assisted by the trend away from drum brakes towards disc brakes. A high proportion of vehicles may be fitted with chrysotile-containing brake pads or shoes in the automotive aftermarket and in home car maintenance, due to the lower cost of chrysotile components. However, the overall use of chrysotile-containing brake parts is unlikely to rise, because non-asbestos brakes are increasingly used in new vehicles.

The International Program on Chemical Safety (IPCS), in assessing the risk to the public from asbestos exposure, concluded that 'the risks of mesothelioma and lung cancer cannot be quantified reliably and are probably undetectably low' and that 'the risk of asbestosis is virtually zero' (IPCS 1986).


Chrysotile is a known human carcinogen; however the risks associated with its use are dependent on the nature of the application and of the product utilised.

Based on the data available, the continued use of chrysotile for friction surfaces, gaskets and seals for critical industrial applications is not expected to present a significant hazard to public health. The use of chrysotile (including manufacturing) in the ways outlined in the NICNAS report will result in a low hazard to the environment. Thus, the NICNAS report raises no objections to the continued use of chrysotile in these applications. However, it recommends continued progress towards phasing out this material in favour of less hazardous materials (where this does not introduce greater risks through the lesser performance of substitute materials).

Follow-up actions

In October 1999, NOHSC agreed to support a recommendation by NICNAS to phase out the manufactured uses of chrysotile asbestos, subject to an impact analysis. As a result of this decision, NOHSC commissioned several projects to facilitate this phasing out.

1.4 Sources of air toxics

Air toxics can be released from many different sources, including natural sources such as bushfires. However, the major sources of air toxic emissions are related to human activities (anthropogenic sources). These include motor vehicles, industry, combustion of fossil fuels (by industry or in the home), cigarette smoking, household chemicals, releases from carpets or furniture and many others. Section 4 of this report provides an indication of the contribution that industry, motor vehicles, other mobile sources and domestic sources can make to the emissions of air toxics in selected regions. More comprehensive data on emission sources will become available as the NPI develops.

Generally, sources of air toxics can be divided into two main categories: point sources and diffuse sources. Point sources comprise industrial and other facilities that emit large amounts of air toxics in a localised area. Diffuse sources include mobile sources (eg motor vehicles and aircraft) and area-based sources (eg solid fuel combustion, dry-cleaning, building materials and use of paints and thinners).

The following subsections discuss point and diffuse sources that are, or may be, significant in terms of their contribution to emissions of air toxics in Australia. Any significant environmental effects associated with areas of land contaminated with toxic substances are usually localised, except where contamination of streams or groundwater is involved.

The classification of industry sources presented below is based on the Australian and New Zealand Standard Industrial Classification (1993) and is largely consistent with the classification adopted by the NPI. Major categories of diffuse sources are described separately following a discussion of industry sources.

1.4.1 Emissions from industry

Given the broad range of chemicals used by industry in the manufacture of products such as pesticides, paints and plastics and its energy requirements (largely sourced from the combustion of fossil fuels), there is an obvious potential for industry to emit significant levels of air toxics. Some industry sectors have a more direct potential for emissions of air toxics because of their manufacturing processes (eg clinical waste incineration and metals smelters) or practices (eg burning off for the purpose of land clearing or the use of pesticides in agriculture). While the effects of some activities are likely to be localised, others result in the increased presence of air toxics in the ambient air environment.

The industries listed below (Table 1.2) have the potential to contribute to emissions of air toxics in Australia. Where applicable, a reference has been made to the number of facilities that reported their emissions under the NPI. This emissions data is reported in the NPI database, which can be accessed at the NPI website16.

Detailed descriptions of individual industry sectors, including descriptions of the relevant technological processes and the pollutants likely to be emitted, are available in the NPI emission estimation technique manuals. To date, 78 manuals have been released. These manuals are available from the NPI website. Further manuals are being developed and will become available from the NPI website as soon as they are finalised.

The NPI database provides industry emission estimates for the following airsheds:

More information on the NPI is given in Section 3.2.

Table 1.2: Industry categories reporting to the NPI
Category (number reporting)a Category (number reporting) a
Petroleum product wholesaling (503) Synthetic resin manufacturing (3)
Mining – coal, iron ore, nickel ore, silver-lead-zinc ore and metallic mineral ore (110) Waste disposal services (includes waste incineration industry) (3)
Electricity supply (91) Chemical product manufacturing (3)
Ceramic product manufacturing (54) Copper, silver, lead and zinc smelting, refining (2)
Oil and gas extraction (27) Hospitals (except psychiatric hospitals) (2)
Paper and paper product manufacturing (26) Water transport terminals (2)
Organic industrial chemical manufacturing (25) Services to air transport (1)
Beer and malt manufacturing (14) Concrete product manufacturing (1)
Petroleum refining (13) Plywood and veneer manufacturing (1)
Iron and steel casting and forging (13) Wood product manufacturing (1)
Printing, publishing and services to printing (11) Pesticide manufacturing (1)
Paint and ink manufacturing (11) Plastic injection moulded product manufacturing (1)
Glass and glass product manufacturing (10) Nonmetallic mineral product manufacturing (1)
Medicinal and pharmaceutical product manufacturing (9) Basic iron and steel manufacturing (1)
Dry-cleaners and laundries (7) Aluminium rolling, drawing, extruding (1)
Aluminium smelting (6) Metal container manufacturing (1)
Inorganic industrial chemical manufacturing (6) Gas supply (1)
Sugar manufacturing (6) Road and bridge construction (1)
Alumina production (5) Port operators (1)
Nonferrous metal casting (5) Chemical wholesaling (1)
Petroleum and coal product manufacturing (4) Rail transport (1)
Basic nonferrous metal manufacturing (3) Agriculture, forestry and fishing

Note: a Number in brackets indicates the number of facilities that reported emissions data for the 1998–99 reporting year.

1.4.2 Motor vehicle emissions

Due to the extensive use of motor vehicles in our society, most people are exposed to potentially harmful air toxics from vehicle emissions. Some important examples of air toxics that are emitted by vehicles include benzene, 1,3-butadiene, formaldehyde, acetaldehyde, PAHs and particles. Table 1.3 below presents the substances found in motor vehicle emissions that are also included in Tables 1 and 2 of Schedule A to the NPI NEPM.

Table 1.3: NPI substances emitted from motor vehicles
Acetaldehyde n-Hexane
Acetone Lead and compounds
Benzene Manganese and compounds
1,3-Butadiene Nickel and compounds
Cadmium and compounds Oxides of nitrogen
Carbon monoxide Particulate matter ≤ 10 µm
Cyclohexane Polycyclic aromatic hydrocarbons
Chromium (III) compounds Styrene
Chromium (VI) compounds Sulfur dioxide
Cobalt and compounds Toluene
Copper and compounds Total volatile organic compounds
Ethylbenzene Xylenes
Formaldehyde Zinc and compounds

Most of the air toxics from cars arise from the by-products of the combustion process when fuel is burnt in the engine and then emitted via the exhaust system, and from evaporation of the fuel itself. Particles are also emitted from brakes and through tyre wear.

Various types of pollutants are produced in the combustion process. Formaldehyde and a range of VOCs, including acetaldehyde and 1,3-butadiene are produced because the fuel is not completely burnt (oxidised) during combustion. Oxides of nitrogen (NOx) result from the oxidation of nitrogen at high temperature and pressure in the combustion chamber. Carbon monoxide (CO) occurs when carbon in the fuel is partially oxidised rather than fully oxidised to carbon dioxide. Sulfur dioxide (SO2) and lead are derived from the sulfur and lead in fuels. Particles are produced from the incomplete combustion of fuels, additives in fuels and lubricants, and worn material that accumulates in the engine lubricant. These additives and worn materials also contain trace amounts of various metals and their compounds, which may be released as exhaust emissions.

Evaporative emissions come mainly from petrol fuel; diesel fuel has a much lower vapour pressure. Evaporative emissions from petrol consist of VOCs such as benzene, which is also a major component of exhaust emissions, and small amounts of lead. These emissions may occur in several ways:

Another source of evaporative emissions is the crankcase of early model (pre-1970) vehicles without positive crankcase ventilation systems. In such vehicles, losses occur directly from venting of the crankcase during engine operation. Evaporative emissions also occur from vehicle refuelling at service stations or from fuel tanker loading and unloading.

Another type of emission that arises from use of motor vehicles is dust emission from roads.

Factors affecting vehicle emissions

The principal factors affecting vehicle emissions are:

The emission control technologies employed by an in-service vehicle, the condition of its emission control equipment, and its state of maintenance and repair, have significant impacts on emissions. Emissions also vary significantly with vehicle and engine operation, which in turn are strongly related to road types (based on traffic flow conditions), and hence vehicle speeds and driving patterns. Reid vapour pressure (RVP), temperature and number of trips per day have important effects on evaporative emissions. Other factors affecting motor vehicle emissions include road conditions and grade, weather conditions, the proportions of hot and cold starts, and the use of air conditioners.

Vehicle emission standards

Australian Design Rules (ADRs) are mandatory standards for motor vehicle safety and emissions, made under the Commonwealth's Motor Vehicle Standards Act 1989. They apply to all vehicles prior to first registration in Australia, and are administered by the Commonwealth Department of Transport and Regional Services. Requirements for in-service vehicles may be subject to State regulation.

Exhaust and evaporative emissions from new petrol-fuelled passenger vehicles and light commercial vehicles up to 2.7 tonnes gross vehicle mass (GVM) are currently regulated by ADR 37/01. This rule was based on standards applying in the United States for 1981 and 1982 models, and was applied in Australia during 1997 and 1998. A summary of the principal ADRs for control of emissions of light-duty petrol-fuelled vehicles is presented in Table 1.4.

ADR 36/00 applies long-superseded United States emission standards to petrol engines used in heavy-duty vehicles greater than 2.7 tonnes GVM. Since there are very few of these vehicles in Australia, this ADR no longer has much significance.

ADR 70/00, which was implemented in 1995-96, sets emission limits for diesel engines used in heavy-duty vehicles. It provides manufacturers with the option of complying with one of three sets of emission standards – those current in Europe in 1995-96 (Euro1), the United States in 1991 and Japan in 1993–94. This ADR is based on engine dynamometer testing procedures, which are different in each country. As an indication only, the ADR 70/00 limits (expressed as Euro1) are shown in Table 1.5.

Table 1.4: ADR emission standards for light-duty petrol vehicles
Standard Year of application Equivalence Exhaust emission limits (g/km) Evaporative emission limit (g/test)
ADR 27A 1976 US 1973 24.2 2.1 1.9 2b
ADR 27B 1978   22 1.9 1.9 6c
ADR 27C a 1981   18.6 1.75 1.9 2c
ADR 37/00 1986 US 1975 9.3 0.93 1.93 2c
ADR 37/01 1997–99 US 1981 2.1 0.25 0.62 2c

a Applied in New South Wales only.
b As measured by the canister test.
c As measured by the more rigorous SHED (sealed housing for evaporative determination) test.
CO = carbon monoxide; HC = hydrocarbon;NOx = oxides of nitrogen

Table 1.5: Emission standards for heavy-duty diesel vehicles
Standard Years of application Exhaust emission limits (g/kW/hr)
ADR 70/00 (Euro1) 1997–99 4.5 1.1 8.0 0.36

Note: CO = carbon monoxide; HC = hydrocarbon; NOx = oxides of nitrogen; TSP = total suspended particles

ADR 30/00 was introduced in 1976 and sets visible smoke emission standards for diesel vehicle engines. It is consistent with standards used in Europe and the United States in the early 1970s.

Vehicle emission controls

Exhaust emissions from petrol-engine vehicles of more recent design (eg all passenger cars manufactured from 1986) are primarily controlled by catalytic converters. These catalysts convert hydrocarbons (HC) and carbon monoxide (CO) to carbon dioxide (CO2) and water, and (in the case of three-way converters) reduce oxides of nitrogen to nitrogen and oxygen. However, as lead is able to 'poison' or deactivate the catalyst, the more stringent exhaust emission standards introduced in 1986 necessitated the use of unleaded petrol. Leaded petrol will be phased out nationally by 1 January 2002 or earlier in some States. Western Australia phased out leaded petrol on 1 January 2000 and Queensland will do so on 1 March 2001. Sulfur also adversely affects catalyst performance and durability, and regulation of the sulfur content of petrol has led to lower emissions of sulfur dioxide.

Increasingly sophisticated emission control technologies have been employed progressively to meet emission standards, commencing with tighter control of air-fuel mixtures and exhaust gas recirculation in the mid-1970s. Current vehicle models are equipped with computer-operated engine management systems, oxygen sensors and three-way catalytic converters, enabling catalysts to operate at optimal conversion efficiency during different modes of engine operation.

Increased catalyst durability, improved control of evaporative emissions and computerised diagnostic systems that identify malfunctioning emission controls are further reducing vehicle emissions. General improvements in vehicle technology and fuel efficiency are also reducing emissions overall.

Evaporative emissions have been controlled primarily through design features in the fuel system. Reducing the volatility of petrol (especially in summer) has also reduced evaporative emissions of air toxics like benzene.

A number of initiatives concerning improved management of transport emissions and greenhouse gases were announced by the Commonwealth Government as part of the package 'A New Tax System' (ANTS). These initiatives are described in the Measures for a Better Environment (MBE) program (see Appendix B).

Several management options are available to facilitate the gradual decrease of air toxics emissions from motor vehicles:

Review of fuel quality requirements for Australian transport

In 1999, Environment Australia commissioned a review of fuel quality requirements for Australian transport. The review, which has subsequently been published (Environment Australia 2000), addressed the relationship between fuel quality and emission of greenhouse gases and air pollutants. It also collated background information on the existing policies for motor vehicle emission control technologies and studies relating fuel quality to emissions. The main conclusions of the fuel quality review are summarised below.

Australia will adopt vehicle emission standards equivalent to Euro3 for petrol vehicles in 2005/06 and Euro4 for diesel vehicles from 2006–07. These standards are more stringent than those currently used in Australia and represent substantial reductions to the allowable emissions standards for new vehicles. Fuel quality will need to be improved to support the technologies employed to deliver improved emissions standards. In particular, reduction of sulfur levels will be required: in petrol to a maximum of 150ppm by 2005 and in diesel to a maximum of 50ppm by 2006.

In general, emissions are relatively insensitive to changes in fuel quality. Improvements in the emission rates of future vehicle fleets will be due primarily to advances in pollution control technology. The major advantage of cleaner fuels is to enable new pollution control technologies.

Emissions of some pollutants tend to vary in proportion to the amount of the pollutant in the fuel (eg sulfur and lead). Such substances in fuel may undergo little or no transformation during the combustion process before being emitted. Fuel quality, therefore, tends to have a more direct effect on emissions for these substances.

Evaporative HC emissions are expected to contribute an increasing proportion of total HC emissions in the future and it is estimated that they will account for the majority of annual emissions by 2010. This is because stricter limits on evaporative emissions are not part of the new exhaust emission standards to be introduced. However, the modelling of evaporative emissions undertaken in this study has some uncertainties, particularly with regard to the magnitude of running and resting emissions for Australian vehicles.

Substantial reductions in emissions are predicted over the next 10 years for HCs, carbon monoxide, oxides of nitrogen, particles and air toxics including benzene.

Most of the benzene emissions over the next 10 years are likely to result from the exhaust component, although the evaporative component will gradually become more important. Stricter limits on evaporative emissions are not part of the new exhaust emission standards to be introduced, meaning that future evaporative emissions of benzene will depend upon benzene levels in fuel.

Fuel quality changes within the ranges considered have little impact on greenhouse emissions from the transport fleet. A range of scenarios predict that greenhouse emissions will increase by 2010 by between 20% and 27% from the 2000 level. These increases would result in Australian transport emissions exceeding the 1990 emissions by between 39% and 43% by the year 2010. This is substantially in excess of the Australian target of an 8% increase in greenhouse emissions as agreed under the Kyoto Protocol.

Production costs of the order of 1–2 cents per litre (c/L) would be required to meet the refiner capital and operating costs needed to deliver Euro4 standard petrol to the Australian public, with the corresponding cost for diesel in the vicinity of 1.5c/L. These costs correspond to approximately 2% of the current retail price of fuel and would result in a reduction of 0.19% in gross domestic product and an increase of 0.08% in the consumer price index. This assessment does not include an allowance for increased vehicle costs associated with improved emissions performance and the impacts of the goods and services tax (including reduced purchase price for new vehicles) and does not consider the economic benefits associated with reduced air emissions.

If higher fuel standards were to raise refining costs and these cost increases were passed on to the industry and consumers, then these effects would be expected to reduce national output. This does not mean that higher fuel standards would necessarily be detrimental to the economy overall, because benefits of higher fuel standards must be taken into account.

An analysis of economic benefits of improved fuel standards undertaken by the Environmental Economics Unit of Environment Australia indicates that there would be considerable health and environmental benefits from adopting cleaner fuels.

The US EPA has published the results of extensive studies on air toxics emitted by vehicles in the United States. The report is titled Motor Vehicle-Related Air Toxics Study (1993) and is available from the US EPA Mobile Source Air Toxic Emissions website17.

Alternative fuels

A report by the Commonwealth Scientific and Industrial Research Organisation (CSIRO) to the Australian Greenhouse Office (Beer et al 2000) has examined available information on low and ultralow sulfur diesel and alternative fuels for heavy vehicles in terms of their emissions of greenhouse gases and air pollutants. The fuels examined were low sulfur diesel (LSD), ultralow sulfur (ULS) diesel, (ULS), compressed natural gas (CNG), liquefied natural gas (LNG), liquefied petroleum gas (LPG), ethanol, diesohol, canola oil, biodiesel (BD) and waste oil.

For this report, a detailed search for available literature and data on the emissions of greenhouse gases and air pollutants arising from the use of alternative fuels was carried out. A lifecycle approach was adopted, which considered precombustion emissions of greenhouse gases and air pollutants as well as tailpipe emissions during combustion. Most of the information used in the study is from overseas as the data from Australia are limited.

The study used a risk-weighted scoring system based on estimates of human health risk to rank the fuels on their air pollutant emissions. On a lifecycle basis, the gaseous fuels (LPG, CNG, LNG) gave the lowest contribution on this criterion, followed by LSD and ULS (based on only one United Kingdom test of urban buses), and then ethanol. The use of waste oil as a diesel extender slightly reduced greenhouse gases but increased air pollution. Biodiesel scored poorly in relation to air quality because its production and use generate considerable amounts of particulate matter.

The report made the following recommendations of significance to air quality:

The next stage of this project will examine the emissions and other characteristics of alternative fuels using a full fuel lifecycle analysis. Generally, full fuel lifecycle emissions are emissions of a fuel product from when a raw material is extracted (for fossil fuels) or planted (for renewable fuels) to when the fuel is combusted. Significant health and environment related issues from the use of the fuels will also be examined. Special attention will be paid to a range of emissions, which are likely to include particles, benzene, 1,3-butadiene, formaldehyde, acetaldehyde and PAHs.

1.4.3 Emissions from diesel vehicles

The following information is from the Supplementary Toxic Study, commissioned by Environment Australia and undertaken by Parsons Australia Pty Ltd. The study tested toxic emissions from a range of diesel vehicles using commercial grade diesel and a variety of low sulfur fuels, and reviewed the existing information on emissions of air toxics from diesel vehicles.

Diesel vehicles are a major source of air pollutants, which are now widely acknowledged to have a direct impact both on human health and on overall emissions of greenhouse gases.

In Australia, diesel vehicles are an increasingly significant source of carbon dioxide and most of the criteria air pollutants (ie those included in the NEPM for Ambient Air Quality). These include ozone (O3), nitrogen dioxide (NO2), and suspended particulate matter (PM10); and to a lesser extent, carbon monoxide and sulfur dioxide. Diesel vehicles are also of increasing concern as a source of noncriteria air pollutants, including VOCs, metals, fine and ultrafine particulate matter (PM2.5 and PM1.0), and reactive organic compounds (ROCs).

The absolute contribution of diesel vehicles to total pollutant and greenhouse emissions has been growing due to their increasing numbers and total distance travelled. Their relative contribution to pollutant emissions has also been growing because until quite recently diesel emissions were largely unregulated, unlike pollutant emissions from petrol vehicles and industrial sources, where control programs have led to progressive reductions (see Appendix B).

A number of diesel emissions are of concern.

Some of these contaminants are emitted as gas or as particles; others are emitted as liquids, which may be adsorbed by particulate matter. Air toxics are generally emitted in very low concentrations but contribute significantly to the toxicity of ambient air. In 1999, the California Air Resources Board identified diesel exhaust as a toxic air contaminant.

Regulatory environment

Over the last 20 years, in many of the major world population centres, there has been growing concern about health and visibility problems attributable to emissions from motor vehicles, leading to progressively tightening emission standards for all new motor vehicles.

Initially, regulatory action was primarily directed at reducing emissions from petrol engine passenger and light commercial vehicles. However, as the emission standards for these classes of vehicles were progressively tightened it became apparent that more stringent action needed to be taken to reduce the emissions of all diesel vehicles.

Research into means to reduce the emissions from diesel vehicles has established that modification of the specification of commercial diesel fuel is necessary to implement and optimise the emission reductions possible from diesel engines. More stringent diesel fuel standards have been progressively introduced, primarily to assist in reducing the emissions of oxides of nitrogen, particulate matter, carbon dioxide and fuel consumption.

Anticipated benefits of changing specific fuel properties are given below.

World-Wide Spec fuel has been proposed by industry representatives from Europe, Japan and the United States for introduction as a future worldwide diesel fuel specification. The proposal to standardise fuel properties would form part of a package aimed at having common vehicle emission standards around the world.

Cetane number, density, T90 and T95, sulfur and aromatic emissions are progressively reduced as more stringent fuel standards (Euro2, Euro3, Euro4 and World-Wide Spec) are introduced.

The aim of these modifications to fuel properties is to assist primarily in reducing emissions of oxides of nitrogen and particles (including smoke). Reductions of HC, carbon dioxide, carbon monoxide and fuel consumption may also be achieved.


Vehicle exhaust standards apply to emissions of oxides of nitrogen, HC, carbon monoxide, total particulate matter and smoke opacity. Future European, United States and Japanese standards are tending to converge in response to growing globalisation of the automotive industry.

In Europe, new standards are being introduced in five progressively more stringent steps, Euro1 from 1992, Euro2 from 1995, Euro3 from 2000, Euro4 from 2005 and Euro5 from 2008. The technical requirements of these standards are being progressively adopted as UNECE standards, which are the basis for standards set in many non-European nations.

More stringent fuel specifications are required for compliance with most stages of emission control stringency.

In Australia, before 1995 diesel-vehicle engines only needed to be certified to ADR 30/00, which set limits for smoke opacity. All diesel engines (and most diesel vehicles) marketed in Australia are imported. Currently, there is little published information to indicate the gaseous and particulate emission standards these vehicles/engines were actually designed to meet.

ADR 70/00, phased in between 1995 and 1997, set emission limits for oxides of nitrogen, HC, carbon monoxide and particles by reference to European, United States and Japanese standards current in 1994 (See Table 1.5).

Recently, the Australian Government adopted new ADRs for emissions from light diesel vehicles (79/00 and 79/01) and heavy diesel vehicles (80/00 and 80/01). ADR 30/01 which applies to all diesel vehicles sets standards for smoke emissions. For diesel vehicles, these new ADRs adopt the technical requirements of Euro2/3 for implementation in 2002–03, and the technical requirements of Euro4 for implementation in 2006–07. For heavy vehicles, US 1998 is specified as alternate to Euro2/3, and US 2004 is specified as alternate to Euro4. Progressive implementation of Euro2, 3 and 4 fuels is part of the MBE package (see Appendix B).

These new ADRs are intended to be introduced as Trans Tasman Vehicle Standards, and will substantially reduce new vehicle emission levels. The standards US 1998 and Euro3 adopt new requirements for emissions durability and for onboard diagnostics, which are expected to significantly improve inservice compliance.

Australian emissions data

Design certification testing for heavy-duty diesel vehicles/engines supplied in Australia has in all cases been carried out overseas. These emissions-type certification tests have been based upon engine bench test procedures, which are difficult to relate to actual vehicle performance under 'real world' driving conditions. Currently, there are no applicable standard procedures for emissions testing of completed diesel vehicles (as opposed to engines) anywhere in the world.

Estimates of diesel vehicle emissions have so far been based on emission factors derived overseas and developed mainly from desk analysis of emission results from bench tests. For the Australian fleet, emissions estimates for inventory purposes have been derived mainly from US EPA emission factors. While these may be representative of vehicles in the United States, their applicability to vehicles in Australia is unclear. As a result, local inventories of diesel vehicle emissions are at best crude, providing doubtful guidance for emission control policy.

Over the last several years a number of researchers, particularly in North America and Europe, have studied inservice vehicle emissions performance, reflecting growing concern about the health and environmental impacts of diesel vehicle emissions. However, there remains almost a complete lack of representative data that can be extrapolated to the Australian context to provide reliable information on the performance of the Australian heavy-duty diesel fleet.

New Australian research

Environment Australia commissioned Parson Australia Pty Ltd to test toxic emissions from a range of diesel vehicles using commercial grade diesel and a variety of low-sulfur fuels. Commercial grade fuel was tested on 12 vehicles over the 'composite urban emissions drive cycle' (CUEDC). In addition, two vehicles were tested on the following low-sulfur fuels:

A full report on this study will soon be available on the internet18 (ATP Technical Paper Series).

The interim results from the study were highly variable across the range of vehicles tested, but overall the emission rates of the toxic species are consistent with those previously determined in United States and European studies:

The type of driving strongly influenced the emissions for all vehicles tested, with highest emissions measured during congested traffic conditions and lowest emissions recorded for high-speed freeway driving conditions.

Results indicate that reducing the sulfur and aromatic content of diesel fuel has no significant impact on emission of VOCs and aldehydes. This is as expected, because these species are not diesel fuel components but are formed in the combustion process from fuel fragments produced in the initial oxidative pyrolysis of the fuel. These fuel fragments are largely derived from the major constituents of the fuel (straight-chain aliphatic HCs and related species). Thus, their formation is largely determined by combustion conditions such as local stoichiometry and temperatures. Of the four target VOCs studied, benzene was generally the most abundant, followed by toluene and the xylenes, with 1,3-butadiene levels much lower and in many cases not detected. The most abundant aldehyde measured was formaldehyde, followed by acetaldehyde, with smaller amounts of acetone and acrolein and trace amounts of other carbonyls.

In contrast to VOCs and aldehydes, total emissions of PAHs are reduced with a decrease in the aromatic content of fuel. Certain PAHs are fuel components and their formation is closely related to carbonaceous particles formation.

This result is, perhaps, not unexpected. The monoaromatics and aldehydes are not diesel fuel components, but are formed in the combustion process from fuel fragments produced in the initial oxidative pyrolysis of the fuel. These fuel fragments are hydrocarbon radical species that are largely derived from the major constituents of the fuel (straight chain aliphatic hydrocarbons and related species). Hence, their formation is largely determined by combustion conditions such as local stoichiometry and temperatures.

The use of new fuel formulations was shown to reduce gaseous and particulate emissions from diesel vehicles.

Whilst a relationship between particulate matter emissions and fuel sulfur content was demonstrated, it is not possible, because of variations in fuel specifications, to attribute all of the benefits solely to the sulfur content reduction. The major benefit of reduced sulfur content of diesel fuel is that exhaust gas after-treatment devices can be fitted. The durability of these devices, particularly in the long term, is affected by the sulfur content of diesel fuel.

1.4.4 Solid fuel combustion

Wood is the main solid fuel in use in Australia, with coal and briquettes used in smaller amounts. Other solid fuel sources are unlikely to be significant contributors to aggregate emissions (ie emissions from non-industrial, area based, sources). However, domestic solid fuel combustion can contribute significantly to overall area-based emissions and is recognised as the cause of episodes of severe pollution in many Australian cities during winter when wood heaters are at peak use. Several studies have shown that woodsmoke poses a threat to human health.

Concern about possible health impacts of woodsmoke in Australia has been voiced since 1981 (Todd 1981). Recent links between respirable particle concentrations and human health have increased this concern. Robinson and Campbell (1998), concerned about woodsmoke concentrations in Armidale, New South Wales and elsewhere in Australia, reviewed articles that demonstrated health impacts of woodsmoke. They concluded that more government intervention is needed to reduce the health risks associated with woodsmoke.

In addition to particulate matter emissions, a significant range of priority air toxics has been associated with woodheaters and open fireplaces, including dioxins, PAHs, aldehydes and VOCs. The quantity and composition of emissions from domestic wood combustion are highly variable and are a function of the type of woodheater used, the characteristics of the wood being consumed and operating practices.

Table 1.6 lists the main substances that Table 2 of Schedule A of the NPI NEPM lists as being emitted during burning of wood and coal/briquettes.

The NPI based its woodheater emission estimates on three types of heaters commonly in use in Australia. Other stove types, such as catalytic stoves, pellet stoves and masonry heaters, are not currently popular in Australia and were not considered by the NPI, but their exclusion may need to be reconsidered if their use increases significantly. Seasonal and temporal variations were not considered, as the NPI NEPM requires reporting of annual emissions. Further information is given in the NPI Emissions estimation technique manual: Aggregated emissions from domestic solid fuel burning, November 1999.

Three main types of woodheater and stove are used in Australia – open fireplaces, conventional heaters and controlled combustion heaters. Open fireplaces are the least efficient and have the highest emissions. Controlled combustion heaters are the most efficient and have the least emissions.

Open fireplaces can be either integral to the structure of the building (typically masonry and built into the wall structure) or prefabricated (either freestanding or inserted into an existing masonry fireplace). These heaters warm by radiant heat.

Conventional heaters and stoves are enclosed and control burn time by limiting the amount of air that can be used for combustion. Conventional stoves do not have specific technology or design features for emission control. Examples of conventional stoves are pot-bellied stoves and older style slow combustion heaters.

Controlled combustion heaters employ emission reduction technology such as baffles or secondary combustion chambers. Wood stoves meeting AS4013 would fit into this category. AS4013 of 5.5 g/kg (grams of smoke measured as particulate matter emitted per kilogram of wood burnt) was released in 1992. Requirements for compliance with the standard have varied between jurisdictions.

Table 1.6: NPI priority substances emitted from solid fuel combustion
Acetaldehyde n-Hexane
Acetone Hydrochloric acid
Antimony and compounds Lead and compounds
Arsenic and compounds Manganese and compounds
Benzene Mercury and compounds
Beryllium and compounds Methyl ethyl ketone
1,3-Butadiene Nickel and compounds
Cadmium and compounds Oxides of nitrogen
Chromium (III) compounds Particulate matter ≤10µm (PM10)
Chromium (VI) compounds Phenol
Carbon disulfide Polycyclic aromatic hydrocarbons
Carbon monoxide Selenium and compounds
Cobalt and compounds Sulfur dioxide
Cyanide compounds Styrene
Dichloromethane Tetrachloroethylene
Ethylbenzene Toluene
Di-(2-ethylexyl) phthalate Total volatile organic compounds
Fluoride compounds Xylenes
Formaldehyde Zinc and compounds

Source: National Pollutant Inventory NEPM.

The Australian Standard for new woodheaters (to come into effect from 1 July 2001) has been tightened from 5.5 g/kg to 4.0 g/kg (AS4013:1999). A code of practice for woodheater installation that will address the optimal dispersion of particles from the chimney is currently being developed. Woodheaters that comply with AS4013 standards have been shown to have significantly lower emissions of carbon monoxide, PM10, and VOCs (Source: Air Emission Inventory, Port Phillip Region, EPA Victoria 1998a).

There are no guidelines for fireplaces and wood stoves, and the contribution of these appliances to woodsmoke is not clear.

It is possible to achieve overall reductions in emissions from a solid fuel-burning appliance in several ways. Improved combustion efficiency, through increased burn rate and flame intensity, will decrease PM10 and carbon monoxide emissions from fireplaces and wood stoves. This could be achieved through increased public awareness and education to promote better operation of existing heaters. The increased introduction of heaters compliant with AS4013 will also reduce emissions and can be achieved through appropriate promotion and/or jurisdictional control.

Different types of wood have different burning qualities. The density and rates of release of volatile gases vary significantly from one species of wood to another. It is generally accepted that softwoods (eg pine) produce higher PM10 emissions than hardwoods (eg eucalyptus). However, emission factors for different species of wood are not readily available. There is significant variation in the type of fuelwood used across Australia, partly related to increasing awareness of the impact on biodiversity of harvesting firewood. For instance, in the ACT region there are significant sources of plantation softwood and the ACT firewood strategy encourages the use of mixed loads of fuelwood. There have been recent proposals for the establishment of hardwood plantations specifically aimed at the firewood market.

A high level of moisture in fuelwood will decrease burning efficiency and increase emission rates. Regulations governing the quality of fuelwood for sale in Western Australia require a maximum moisture content of 20% and prohibit the sale of fuelwood that is painted, chemically treated or plastic coated. A code of practice in the ACT also specifies a maximum moisture content of 20%. This code operates through self-regulation of the firewood industry.

The following factors also influence emissions from solid fuel burning:

Emissions from solid fuel combustion are strongly seasonal, and can also vary with time of day, distance from coast, altitude, age of residence and economic factors. Future work could lead to development of emission factors or factor adjustments that address some of the effects discussed above. Environment Australia has recently commissioned a study to characterise the full range of air toxics emissions from range of domestic solid fuel burning appliances, using a variety of fuels under different operating conditions. A review of data on air toxics in woodsmoke emissions is also being prepared as part of this project. More information on this project is given in Section 1.4.9.

Firewood use trends in Australia

Figure 1.2: Proportion of Australian households using firewood as main heating source.

Figure 1.2: Proportion of Australian households using firewood as main heating source.

Figure 1.2 summarises the data set for Australia for the proportion of households using firewood as their main heating fuel, based on the above ABS publications. Fitting a fourth order polynomial to the data suggests a peak in popularity around 1992-93. The data indicate that the proportion of households using firewood as their main heating source has dropped from about 18.5% in 1992 to 15.5% in 2000. This is a significant fall (about 3 percentage points or16% in relative terms). But when growth in the number of households is included, the fall in the actual number of wood-burning households is much smaller, around 2%. The annual decrease in the number of households using firewood as their main heating fuel is only around 3000–4000.

Information on the quantity of firewood consumed for space heating has been collected in some surveys (eg Todd et al 1989), but is not as reliable as the data on the proportion of households using firewood. Table 1.7 provides the most recent estimates of firewood use for each state and territory.

The data presented in Figure 1.2 and Table 1.7 suggest firewood use of around 4 million tonnes per year (air-dry weight). The oven-dry weight (used for calculation of emission factors) would be around 3.4 million tonnes (assumes air-dried wood has a moisture content of about 16%). However, this estimate might still be high because a small survey conducted in Armidale, New South Wales, found that actual firewood weights were, on average, 23% less than the consumer had purchased (Wall 1997). If firewood supply is generally underweight, the actual quantity of firewood consumed annually might be closer to 2.5 million tonnes (oven-dry weight).

Firewood use has been decreasing by an estimated 6000 to 8000 t/yr (oven-dry weight) over the past seven years, a decrease of about 0.3%/yr.

Table 1.7: Estimates of the number of households using firewood as their main heat source and secondary heat source, and estimates of total firewood use
State Number of households as main heater Tonnesa/year/ household Number of households as second heater Tonnes a /year/ household Total tonnesa all households/year
NSW 351 800 2 117 267 1 820 867
Vic 240 900 4.324 180 675 1.932 1 390 716
Qld 128 700 2.9 32 175 1 405 405
SA 107 400 3 26 850 1 349 050
WA 176 400 2.53 44 100 1 490 392
Tas 104 700 4.8 12 564 2.2 530 201
NT 1800 3 450 1 5850
ACT 6700 3.7 5025 1.9 34 338
Australia 1 118 400   419 106   4 026 818

Note: a Firewood weight is expressed as air-dry tonnes (16% moisture wet-weight basis).

Sources: Number of households using firewood as main heat source: ABS (1999);
Number of households with secondary firewood use: Todd et al (1989) (adapted);
Firewood use per household: Todd et al (1989) (adapted).

Toxic components and health effects of woodsmoke

The gaseous and particulate components of woodsmoke have complex chemical structures. Primary emissions include volatiles driven from the wood as the wood is heated. Secondary chemicals are formed through reactions in the hot combustion chamber and the flames, and further transformations take place when the emissions enter the atmosphere. The presence of many toxic compounds in woodsmoke is now well established; however, the concentrations of individual compounds are highly variable so it is difficult to establish emission factors with any certainty. The US EPA publishes a list of selected emission factors (Table 1.8) with the warning that 'Data show a high degree of variability within the source population. Factors may not be accurate for individual sources'19.

Table 1.8: Emission factors for various compounds for woodheaters
Compound Emission factor (g/kg)a Australiab
Ethane 0.735  
Ethylene 2.245  
Acetylene 0.562  
Coronene   B
Propane 0.179  
Propene 0.622  
i-Butane 0.014  
n-Butane 0.028  
Butenes 0.596  
Pentenes 0.308  
Benzene 0.969 A
Toluene 0.365  
Furan 0.171  
Methyl ethyl ketone 0.145  
2-Methyl furan 0.328  
2,5-Dimethyl furan 0.081  
Furfural 0.243  
o-Xylene 0.101  
PAH total <0.250  
Acenaphthylene 0.016  
Anthracene 0.005 A
Benzo[a]pyrene 0.003 B
Benzo[b]fluoranthene 0.004 B
Benzo[k]fluoranthene <0.001 B
Benzo[g,h,i]perylene 0.028 B
Benzo[a]anthracene <0.001 B
Dibenzo[a,h]anthracene 0.004  
Biphenyl 0.011  
Chrysene 0.005 B
Fluoranthene 0.004 A,B
Fluorene 0.007  
Indeno[c,d]pyrene 0.020 B
Naphthalene 0.072 A
Phenol <0.001
Pyrene 0.004 A,B

a Noncatalytic or conventional woodheaters.
b Compounds found in Australian studies of woodsmoke.
A = identified in woodheater emissions in Australia (Chesterman 1984).
B = identified in open fireplace emissions in Australia (Freeman and Cattell 1990).

Source: selected from AP-42 Compilation of Air Pollutant Emission Factors. Vol 1. Stationary Point and Area Sources, available at .

Saloma et al (1985), in Finland, analysed the genotoxic potency of woodsmoke and compared it to cigarette smoke. Woodsmoke from a small, horizontal-baffle woodheater burning birch and spruce was shown to be in the same range as condensate from cigarette smoke (ie highly genotoxic).

The toxic nature of woodsmoke has also been demonstrated in some animal studies. For example, studies of woodsmoke inhalation by rats showed an immediate change in respiration (Kou et al 1997). The authors attributed this change to the increased presence of the hydroxyl radical in the woodsmoke.

Few measurements of the toxic components of woodsmoke have been made in Australia. The most extensive database relates to total particles emitted by woodheaters, but most of the data involve testing to the Australian Standard method (AS4013) under laboratory conditions. No in situ measurements of woodsmoke emissions in private homes have been made in Australia.


For all models tested for AS4013 compliance, the arithmetic mean emission factor is 3.3 g/kg (range 0.8–5.5 g/kg). If only models complying with the 1999 revision of the standard (ie limiting emission factors to a maximum of 4 g/kg) are considered, the arithmetic mean is 2.8 g/kg.

Testing also provides information on appliance efficiency (average 62%, range 48–81%) and heat output at high burn rate (average 10.5 kW, range 5.2–22 kW).

A limited amount of testing has compared particle emission factors for hardwood and softwood, and emission factors for wet and air-dry wood. Softwood particle emission factors are similar to hardwood at high and medium burn rates but softwood gives emission factors two to three times higher than hardwood at low burn rates (Todd 1991). Wet softwood (40% moisture) burnt at high burn rates yields emission factors about 50% higher than dry wood (16% moisture) (Todd 1991). Quraishi (1987) tested heaters with larger fuel loads (> 9 kg) and smaller loads (< 9 kg) and found larger loads doubled the emission factor.

The lack of field testing of particulate matter emission factors in Australia has led to various attempts to estimate average emission factors for the mix of appliances and operating factors. Todd (1997) estimated an average particle emission factor of 13.6 g/kg, gradually reducing as certified heaters replaced older appliances with higher emission factors.

Assuming an emission factor of 10–15 g/kg and annual firewood combustion of 3 million tonnes (oven-dry weight), around 30,000–45,000 tonnes of fine particles are emitted annually. Two factors influence the trend in emissions; one is the total firewood consumed (decreasing slowly at present) and the other is improvement in emission factors.

Woodheater sales in Australia have averaged around 40 000 per year since 1992 (AHHA 1999). This suggests that about one-third of woodheaters now in use have been manufactured after the publication of the Australian Standard for emissions (AS4013). However, not all states and territories require compliance with the standard. Assuming three-quarters of new heaters do comply with the standard (there are no data to confirm or reject this assumption), it is likely that only about 25% of heaters now in use comply with the standard.

A rough estimate of the reduction in emissions based on replacement of older, more polluting, woodheater models can be made. Assuming AS4013 heaters produce half the emissions of noncompliant heaters, each replacement results in a reduction of 20 kg of particle emissions per year (average 2.8 oven-dry tonnes burnt, with emission factor reducing from 14 to 7 g/kg). If there were 30 000 replacements per year, the total reduction in particles would be about 600 tonnes or about 1.5% of total particles.

The health impact of respirable particles (less than 10 microns aerodynamic diameter, PM10) has been the focus of numerous epidemiological studies over the past 10 years. These studies have consistently shown a link between PM10 concentrations and mortality and morbidity. The effects appear to be due to short-term raised particle concentrations rather than the chemical composition of the particles, which has not been identified as significant.

Woodsmoke consists of fine and ultrafine particles and so will contribute to PM10 concentrations. Several studies, including the NEPM for Ambient Air Quality (NEPC 1998) have identified woodsmoke as a significant source of PM10.

Polycyclic aromatic hydrocarbons

Australian measurements of PAHs from residential wood burning are limited. Freeman and Cattell (1990) analysed the PAH concentration of particulate matter from an open fireplace. Only two tests were conducted. PAH was expressed as concentration of the particulate matter, not as an emission factor. Table 1.8 indicates the compounds detected. If Freeman and Cattell's emissions are converted to emission factors, assuming an emission factor of 15 g/kg for the open fire, their results are the same order of magnitude as the US EPA figures (Table 1.8). For example, Freeman and Cattell measured 77 µg/g of particles for fluoranthene, which converts to 1.2 mg/kg of wood burnt. The US EPA figure is 4 mg/kg. For pyrene, Freeman and Cattell measured 1.3 mg/kg, compared to the US EPA's 7 mg/kg. The uncertainties in this comparison are too large to draw any firm conclusions.

Chesterman (1984) carried out measurements of organic compounds in woodsmoke from a woodheater. The range of compounds detected in the smoke was similar to that in United States studies.

Moller et al (1985) measured ambient concentrations of PAH in a small town with winter woodsmoke problems. Relatively high average PAH concentrations of 158 ng/m³ were measured in winter, with a maximum of 497 ng/m³.

Lee et al (1977) compared PAH emissions from burning coal, wood and kerosene. They found greater concentrations of alkylated PAH from coal than the other two fuels, but lower concentrations of high molecular weight species from coal than from wood and kerosene.

Kamens et al (1986) have demonstrated that PAH in woodsmoke decays in the presence of sunlight, but the decay rate is slower at low temperatures. At 20°C the half-life of various PAH species was 30–60 minutes, but it increased to many hours and even days at -7°C. The authors concluded that there will be some decay in mild winter regions depending on the available sunlight.

McCrillis et al (1992) found higher emissions of PAHs when burning softwood (southern yellow pine) than hardwood (oak). They found no correlation between the PAH emission factor (g PAH/kg wood burned) and burn rate. They also noted a decrease in PAH emissions at high altitude (825 m). Overall, they reported an emission factor for the sum of 16 PAHs of about 0.3 g/kg.

Knight et al (1982) conducted emission tests on five woodheaters burning hardwood (oak) under various burn conditions. The PAH emission factor for the five heaters ranged from 0.047 g/kg to 0.251 g/kg (overall average 0.148 g/kg). The PAH emission factors were higher when larger loads of fuel were burnt and lower in catalytic heaters than conventional heaters.


A Danish study of four woodheaters, including three commercial models and one experimental model, measured an average dioxin emission factor of 1.9 ng TEQ/kg (Vikelsoe et al 1994). This study measured dioxin using the TEQ (Nordic) method. The study investigated the four heaters operated under 'normal' (slow to medium burn rate) and 'optimal' (minimum carbon monoxide emissions) firing conditions with three firewood species. The authors concluded that there was a statistically different emission of dioxin as a function of heater model, wood species and operating condition. Interestingly, they did not find any correlation between dioxin emission and total HC emissions. Also of interest was the observation that one heater showed much higher dioxin emissions when operated in the optimal mode rather than normal.

Bumb et al (1980), in their pioneering work on dioxins, found TCDD in the soot/creosote collected from two residential open fireplaces where only wood had been burnt. The TCDD concentration in the soot was up to 0.4 ng/g.

The United States Inventory of Sources of Dioxin adopted an emission factor of 2 ng TEQ/kg, apparently largely on the basis of the above Danish study. Houck (1998) is critical of this figure, on the grounds of its uncertainty. He points out that the Danish study found significant variation in emission factors as a function of heater design and wood species and therefore argues that it is inappropriate to use this value for emissions in the United States.

Assessment of dioxin sources in Australia included residential use of firewood (Pacific Air and Environment 1998). The emission factors used were based on United Kingdom data. The study concluded that residential firewood use was the fourth largest category of dioxin emissions, behind prescribed burning, bushfires and cement production, but ahead of coal combustion. The assumed emission factors for 'clean' firewood were 1–3 µg/t for woodheaters and 1–29 µg/t for open fireplaces. The study also assumed emission factors of 5–50 µg/t and 100–500 µg/t for 'treated' wood burnt in woodheaters and open fireplaces respectively.

In view of the very limited scientific data on dioxin emissions from residential firewood use, these assumptions are probably the best available. However, in applying these to Australian firewood use, Pacific Air and Environment have assumed that 470 000 tonnes of 'treated' wood is burnt annually. There is no evidence that this is the case; in fact all firewood surveys in Australia indicate very little waste wood or treated wood is consumed. If it is assumed that all treated wood is replaced with clean firewood then residential firewood use drops to the seventh or eighth most significant source of dioxins in Australia.

Chemically treated wood

Several studies of firewood from chemically treated trees have been carried out in the United States (eg Bush et al 1987, Newton 1986, Woolston 1986). All authors concluded that the use of herbicides to kill trees did not present a health risk when the wood from the treated trees was burnt. Organic herbicides are degraded in the combustion process. Arsenic-based herbicides do not spread through the tree and Woolston (1986) concluded that arsenate was present in particulate matter emissions but that, if wood more than 0.3 m from the point of injection was burnt, the health risk was minimal.

These tests of treated wood do not apply to CCA (copper, chrome, arsenic) treated timber, where much higher concentrations of chemicals are present.


As part of the Integrated Air Cancer Project in the United States, the mutagenic properties of woodsmoke and motor vehicle emissions were investigated for ambient and indoor air. This project generated a number of the studies described below. Mutagenicity is usually measured as direct (without the metabolic activator S9) or indirect (with S9 activation, which causes the substance to metabolise). Two strains (TA98 and TA100) of Salmonella typhimurium bacteria are used in most of the reported testing, with mutagenicity reported in units of revertants/µg or revertants/m³ for air samples20.

McCrillis et al (1992) report on a series of woodheater tests conducted to investigate wood species, wood moisture, burn rate and fuel load on mutagenicity and PAH. Mutagenicity was measured using the Ames method. Based on 90 or 95% confidence intervals, the study concluded that:

The authors concluded that 'The results also show that there is wide variability due to small, uncontrollable differences from one fire to the next and to (heater) design.' (McCrillis et al 1992:691).

To summarise, the mutagenic potential of woodsmoke in ambient air is reported as 0.12–1.3 revertants/µg in the United States, with some higher results in Scandinavia of 0.2–4.4 revertants/µg (Lewis and Wallace 1988). The gas phase emissions may be more mutagenic than the particle phase. The presence of oxides of nitrogen significantly increases mutagenicity during ageing of the woodsmoke in the atmosphere. Use of an open fireplace (but not an enclosed woodheater) increased indoor mutagenicity to levels comparable to indoor cigarette smoking, although this result is from measurements in a single house.


Firewood use for residential heating in Australia is about 2.5 million t/yr (oven-dry weight). Firewood use is decreasing at about 7000 t/yr (oven-dry weight), equivalent to a decrease of only 0.3%/yr. New models of woodheaters, with lower emission factors, are replacing older models at a rate of about 30 000/yr. This contributes about 1.5%/yr reduction in total particles emitted in woodsmoke. Thus, technology improvements and changes in heating preferences are resulting in a 1–2% reduction of woodsmoke from residential space heating per year.

Total emission of particles from residential firewood combustion is estimated at 35 000 t/yr (there is a high degree of uncertainty in this figure). Total PAH emission (based on United States emission factors) is estimated at 625 t/yr. Total dioxin emissions are estimated at 5 g TEQ/yr, but this estimate is based on extremely few measurements, all conducted overseas.

Major gaps or uncertainties exist in Australian data, including:

The lack of data on toxic emissions from eucalypt firewood species introduces uncertainties in applying United States emission factors to Australian conditions. The current Air Toxics Measurement Program will reduce this uncertainty.

1.4.5 Use of lawnmowers

Four types of lawnmower are used in Australia: two-stroke engined, four-stroke engined, electric and push mowers. Only the first two types emit pollutants to the atmosphere at the point of use. Four-stroke mowers have lower emissions of VOCs, carbon monoxide and PM10 than two-stroke mowers, but higher oxides of nitrogen emissions. Fuel type (leaded or unleaded petrol) can also affect emissions, especially for lead and sulfur dioxide. High-pressure fuel injection two-stroke technology has been developed recently and will penetrate the Australian market in coming years. Emissions from these mowers are lower than from standard two-stroke mowers. At present there are no regulations or standards governing lawnmower engines.

Lawnmower use can vary across a region. Important factors affecting local equipment use include climate, land use, lot size, population demographics and the availability of water in more arid regions (Heiken et al 1997). Lawnmower use also varies both seasonally and with the day of the week. It can contribute significantly to overall area-based emissions, particularly for lead and some VOCs. Emissions from spills during fuel transfer can also be significant. Emissions from lawnmowers can be affected by combustion chamber temperature and air/fuel ratio (Priest 1996).

Table 1.9 below lists the main substances from Table 2 of Schedule A of the NPI NEPM that are emitted by lawnmowers.

Table 1.9: NPI substances emitted by domestic lawnmowers
Benzene Manganese and compounds
1,3-Butadiene (vinyl ethylene) Nickel and compounds
Carbon monoxide Oxides of nitrogen
Chromium (III) compounds Particulate matter ≤ 10 mm
Chromium (VI) compounds Polycyclic aromatic hydrocarbons
Cobalt and compounds Styrene
Copper and compounds Sulfur dioxide
Cyclohexane Toluene (methyl benzene)
Ethylbenzene Total volatile organic compounds
Formaldehyde (methyl aldehyde) Xylenes
n-Hexane Zinc and compounds
Lead and compounds  

Source: National Pollutant Inventory (NPI) NEPM

Changes in emission levels with increasing mower age were not specifically addressed in the emissions estimation technique developed for lawnmowers. However, this factor is (at least to some extent) incorporated into the emission factors used as they are based on tests of mowers in use. Seasonal variations were not taken into account for the purpose of estimating lawnmower emissions, as the NPI NEPM requires reports on annual emissions. Emissions from spills during fuel transfer and variations in emissions due to the combustion chamber temperature and air/fuel ratio changes (Priest 1996) were not specifically addressed in the emissions estimation techniques.

Further information is given in the NPI Emissions estimation technique manual: Aggregated emissions from domestic lawn mowing, November 1999, from which this information was largely derived.

1.4.6 Use of paints and thinners

Architectural surface coatings protect the substrates to which they are applied from corrosion, abrasion, decay and damage from ultraviolet light and water. Architectural surface coatings may also be applied to increase the aesthetic value of structures by changing the colour and texture of surfaces.

Architectural surface coatings generally have three main components:

The predominant emissions from architectural surface coatings during application and drying are VOCs contained in the architectural coatings themselves (ie paint, paint primer, varnish or lacquer) and in the solvents used as thinners and for cleaning up.

Architectural surface coatings are generally classified as solvent-based or water-based. Solvent-based coatings contain between 30 and 70% VOCs by weight and water-based coatings contain approximately 6%.

The substances listed by NPI may be emitted into the air through evaporation from architectural surface coatings and thinners. VOCs used as solvents in the coatings are emitted during the application of the coating and as the coating dries. The volumes of coatings used, the VOC contents of the coatings, and the weight fractions of NPI-listed substances within the VOCs are the key factors that primarily determine emissions from architectural surface coating activities. Emissions may also arise from solvents used to clean the application equipment and as reaction byproducts while the coatings dry and harden.

Since the use of organic solvents in architectural surface coatings is the primary source of emissions, control techniques involve either product substitution or product reformulation. Alternate formulations include low-solvent products, water-based coatings and powder coatings. The Australian surface coating market is currently in a state of flux as water-based product continues to capture market share from solvent-based coatings. More than 80% of architectural paints sold in Australia are now water-based (NPI 1999)

Recycling of unused coatings by manufacturers also serves to reduce emissions, although to date the opportunity of returning and recycling unused product is limited in Australia.

Structural maintenance practices also indirectly influence emissions by controlling the total coating consumption on a long-term basis. Regular inspection and maintenance programs can be used to reduce the need for entire surface recoating.

Table 1.10 below lists the main substances from Table 2 of Schedule A of the NPI NEPM that are typically emitted during application and drying of surface coatings.

Table 1.10: NPI substances typically emitted during application and drying of architectural surface coatings
Acetone Methanol
Benzene Methyl ethyl ketone
Cyclohexane Methyl isobutyl ketone
Dichloromethane n-Hexane
Ethanol Toluene
Ethylene glycol Total volatile organic compounds
2-Ethoxyethanol acetate Xylenes

Source: National Pollutant Inventory (NPI) NEPM.

Further information is given in the NPI Emissions estimation technique manual: Aggregated emissions from architectural surface coating, November 1999, from which this information was largely derived.

1.4.7 Use of agvet chemicals (pesticides)

The general information on agvet chemicals is provided in Section 1.3.2 of this report and a more detailed description of the management framework for agvet chemicals is given at Appendix A.

There is a limited body of information on the magnitude of use of agvet chemicals in Australia. According to National profile of chemicals management infrastructure in Australia (Environment Australia 1998), agvet chemicals sales in Australia in 1996 totalled $1 661 938 227. This figure indicates that Australian agriculture, including farming, forestry, horticulture, aquaculture and the livestock industry, continues to depend on agvet chemicals. These chemicals should be used under controlled conditions and are subject to stringent regulatory controls.

1.4.8 Environmental tobacco smoke

It is commonly recognised that environmental tobacco smoke contains many harmful substances, some of which are considered air toxics.

Environmental tobacco smoke contains, in a diluted form, many of the toxic agents and carcinogens that are present in mainstream smoke (MS), the smoke drawn through the tobacco and into the smoker's mouth (US DHHS 1986). The major source of environmental tobacco smoke is sidestream smoke – smoke generated by smouldering tobacco between puffs, and smoke diffusing through the cigarette paper and escaping from the burning cone during puffing. Sidestream smoke (SS) contains higher amounts of some toxic and carcinogenic agents than mainstream smoke when it is obtained in its undiluted form under laboratory conditions (US DHHS 1989). For example, the release of volatile N-nitrosamines and aromatic amines is higher in sidestream smoke than in mainstream smoke.

Table 1.11: Toxic and carcinogenic agents in undiluted cigarette sidestream smoke
Compound Type of toxicity Amount in SS (per cigarette) Ratio of SS/MS
Vapour phase
Carbon monoxidea T 26.8–61 mg 2.5–14.9
Carbonyl sulfide T 2–3 mg 0.03–0.13
Benzenea C 400–500 µg 8–10
Formaldehydea C 1,500 µg 50
3-Vinylpyridine SC 300–450 µg 24–34
Hydrogen cyanide T 14–110 µg 0.06–0.4
Hydrazine C 90 ng 3
Nitrogen oxidesa T 500–2000 µg 3.7–12.8
N-nitrosodimethylamine C 200–1040 ng 20–130
N-nitrosopyrrolidine C 30–390 ng 6–120
Particulate phase
Tar C 14–30 mg 1.1–15.7
Nicotine T 2.1–46 mg 1.3–21
Phenola TP 70–250 µg 1.3–3.0
Catechol CoC 58–290 µg 0.67–12.8
o-Toluidine C 3 µg 18.7
2-Naphthylamine C 70 ng 39
4-Aminobiphenyl C 140 ng 31
Benzo[a]anthracene C 40–200 ng 2–4
Benzo[a]pyrene C 40–70 ng 2.5–20
Quinoline C 15–20 µg 8–11
n'-Nitrosonornicotine C 0.15–1.7 µg 0.5–5.0
NNK (4-(methyl-nitrosamino)-(3-pyridyl)-1 -butanone) C 0.2–1.4 µg 1.0–22
n-Nitrosodiethanolamine C 43 ng 1.2
Cadmium a C 0.72 µg 7.2
Nickel a C 0.2–2.55 µg 13–30
Polonium-210 C 0.5–1.6 pCi 1.06–3.7

C = carcinogenic;
CoC = cocarcinogenic;
MS = mainstream smoke;
pCi = pico curie;
SC = suspected carcinogen;
SS = sidestream smoke;
T = toxic;
TP = tumour promoter.
a National Pollutant Inventory reporting substance. See Section 3.2.

Sources: US DHHS (1989); Hoffmann and Hecht (1989).

A major reason for the difference in concentrations of toxic and carcinogenic agents in undiluted sidestream smoke and mainstream smoke is that peak temperatures in the burning cone of a cigarette reach 800–900°C during puffing, but only 600°C between puffs. Thus, there is less complete combustion of tobacco during generation of sidestream smoke. In addition, most of the burning cone is oxygen deficient during smouldering and produces a strongly reducing environment (NRC 1986). Table 1.11 lists 26 toxic and carcinogenic agents identified in sidestream smoke and mainstream smoke.

Estimated emissions from tobacco for the Perth and southeast Queensland airsheds, from the NPI database21, are presented in Table 1.12.

For more information also see the United States National Institute of Health and Safety (NIOSH) report Environmental Tobacco Smoke in the Workplace – Lung Cancer and Other Health Effects 1991.

Table 1.12: Estimated emissions from cigarettes: southeast Queensland airshed, 1998–1999
Substance Perth airshed 1998–1999 amount (kg) Southeast Queensland airshed 1998–1999 amount (kg)
Acetaldehyde 5300 NR
Acetone 2800 1000
Acrylonitrile (2-propenenitrile) 220 85
Benzene 750 350
1,3 Butadiene (vinyl ethylene) 530 130
Cadmium and compounds 2.0 0.77
Carbon monoxide 130 000 57 000
Cyanide (inorganic) compounds NR 95
Formaldehyde (methyl aldehyde) 1000 1100
Lead and compounds 0.14 NR
Methyl ethyl ketone 540 250
Nickel and compounds 0.10 NR
Oxides of nitrogen 3900 1500
Particulate matter < 10.0mm NR 6900
Phenol 830 240
Styrene (ethenylbenzene) 250 130
Toluene (methylbenzene) 1300 560
Total volatile organic compounds NR 16 000
Xylenes (individual or mixed isomers) NR 310

Note: NR = not reported

Source: NPI Database (Data supplied by Department of Environment Protection WA and Queensland Environment Protection Agency).

1.4.9 Studies under the Living Cities – Air Toxics Program

Environment Australia has commissioned a number of studies under the ATP focusing on sources and levels of exposure to air toxics. These studies address priority project areas which were identified in consultation with community groups, industry bodies and government agencies. The studies focus on core information that is needed to develop targeted management responses for priority air toxics.

When the studies are complete, their findings will be published by Environment Australia in a technical paper series. The first of these is expected to be published in late 2001, and the remainder in 2002.

Emissions from diesel vehicles 22

Environment Australia commissioned Parsons Australia Pty Ltd, with the participation and assistance of the CSIRO Energy Technology Section, to conduct a study of toxic emissions from diesel vehicles. Emissions were measured from a range of current technology vehicles using standard commercial diesel fuel and new generation fuels.

The aim of this study was to:

  1. obtain a mass based emission result for a range of air toxics emitted from a set of 12 in-service diesel vehicles
  2. obtain an indication of the likely reductions in emissions of toxics, particulates, gaseous and metal emissions from current model diesel vehicles when operated on a range of diesel fuels with varying levels of sulphur content.

Background and results for this study are discussed in section 1.4.3 Emissions from diesel vehicles, and a full copy of the report can be found on the Air Toxics website.

Personal monitoring of selected VOCs: the contribution of different sources to exposure

This study is a collaborative project led by the Western Australia Department of Environmental Protection, and includes the Environment Protection Authority Victoria, NSW Environment Protection Authority, NSW Health, the SA Environment Protection Agency, Flinders, Monash and Murdoch Universities, the University of Western Australia and the CSIRO.

The principle objective of this personal exposure study is to identify significant non-occupational sources of the following VOCs; benzene, toluene, ethyl benzene, xylenes (collectively known as BTEX), and to determine their contribution to personal exposure.

The specific objectives of the study are to:

The study is being conducted in four city centres – Sydney, Melbourne, Perth and Adelaide. It aims to identify the significant sources of BTEX by sampling a range of recognised sources, including service stations, major roads, petrol stations, indoor air where woodsmoke is present, indoor drive-through areas, industrial sites close to residential areas and attached domestic garages. The project then aims to determine individual levels of exposure by conducting personal exposure monitoring. The study is engaging 200 volunteers (50 people in each city) and will provide data on the levels of exposure associated with particular sources and seasonal variation. Participants will be required to wear personal gas monitoring tubes for five days in two different seasons (winter and summer) and keep a diary of their daily activity to help identify where exposure to BTEX may occur.

Preliminary work to investigate sources of potential exposure to BTEX have identified underground car parks and fast food drive-through outlets as having elevated BTEX levels

Information on the actual levels of individual exposure to BTEX will assist the development of targeted management strategies.

Determination of emission factors from in-service vehicles

Environment Australia has engaged a consortium led by EPA Victoria to undertake a study to determine emission factors from local in-service vehicles. The study will utilise the emissions data from the Melbourne City Link Tunnels ventilation stack, which is linked to vehicle data derived from compulsory E-tag (electronic tolling) registrations. The E-tag data, combined with data from the VicRoads vehicle registration database, will provide accurate information on the number, type and age profile of vehicles using the tunnel at the time of testing. This will allow specific emission factors to be calculated for vehicles of certain types and age categories (through mathematical methods) and will provide a valuable comparison with emission factors derived using traditional methods.

Emission factors are critical for determining the impact of vehicle emissions on air quality. Determination of accurate emission factors will allow for accurate projections of the impact of the motor vehicle fleet on future air quality and will inform the development of new ADRs and policies to reduce the impact of motor vehicle emissions. The information generated will contribute to the ongoing process to improve the accuracy of NPI data. The criteria pollutants (carbon monoxide, NO, NO2, lead, PM10 and PM2.5) and a range of air toxics, including benzene, 1,3-butadiene, formaldehyde, toluene, xylenes and PAHs, will be monitored.

Characterisation of emissions from domestic solid fuel burning appliances (woodheaters, open fireplaces)

Environment Australia has commissioned a consortium led by CSIRO to characterise the full range of emissions from a number of domestic solid fuel burning appliances, using a variety of fuels under different operating conditions. Emissions from domestic woodheaters are acknowledged as an important source of air pollution and the cause of acute episodes of pollution in many Australian cities. Currently, there is no complete characterisation of emissions from Australian woodheaters and no quantitative data on how emissions are affected by heater design, operating conditions, and fuelwood quality.

The CSIRO project will characterise the full range of emissions from domestic solid fuel burning appliances including open fireplaces and old and new slow combustion heaters (ie before and after compliance with the new Australian Standard AS4013). The project will also consider a variety of fuelwood types (hardwood, softwood, different moisture content and age) and extremes of operating conditions (maximum and minimum air flow).

Characterisation of the full range of emissions from domestic solid fuel burning appliances will inform the development of national management strategies for air toxics.

Review of data on heavy metals in ambient air in Australia

Heavy metals are widely acknowledged as pollutants of concern in ambient air (see Section 1.3.1). In recognition of this, Environment Australia has commissioned the WA DEP to review current data on levels and sources of heavy metals in ambient air in Australia.

The project will collect and analyse existing studies and will analyse any raw monitoring data collected by State EPAs. It will examine data on the sources of heavy metal emissions, including point, area and mobile sources, and their comparative importance in terms of overall emissions and contribution to ambient air levels. The project will also identify and discuss any trends over time in data on ambient air levels for each heavy metal. It will allow an assessment on the need for further physical testing of heavy metals in ambient air and inform the development of national management strategies.

1.5 Seasonal and regional variation

This section incorporates information from the State of the Environment report (DEST 1996), National Pollutant Inventory Contextual Information (NPI 1999) and the independent inquiry into urban air pollution in Australia (AATSE 1997).

The occurrence of air toxics in Australia shows regional and seasonal variability. This variability is caused by regional and seasonal differences in emissions from human activity and natural sources, the influence of local topographical and meteorological factors, and the stability of the emitted substances and their interaction with atmospheric chemical processes. This may result in the transformation, dispersion and/or accumulation of the emissions in localised areas.

1.5.1 Variability in emissions

Variability in emissions is an important source of regional and seasonal variability in air toxics. As the major source of air toxic emissions are anthropogenic, the level of emissions will depend to a large degree on the presence and level of anthropogenic emission sources (eg motor vehicle use, presence of factory smokestacks, and industry and domestic combustion of fossil fuels and firewood.) These sources depend on factors such as population size and density, the location of certain industries, industry types, level of regulatory controls, level of vehicle use and season.

Pollutant emissions in several of Australia's capital cities show distinct spatial patterns associated with point emissions and emissions from motor vehicles (Newton 1997). Most of Australia's population is concentrated in a small area of the land within about 100 kilometres of the coast. Many anthropogenic emissions are therefore not distributed evenly across the land but concentrated in major metropolitan areas.

As well as areas of high population density, other locations have air quality concerns, either potential or actual. This can be due to rapid population growth or to specific local emission sources. An important source of mainly regional emissions is the metal processing industry in centres such as Mount Isa, Boolaroo and Port Pirie. Agricultural burning, fuel reduction burns and aerial spraying can all emit a range of pollutants and are generally associated with rural areas.

People's perception of air quality is strongly influenced by emissions of pollutants in their local areas. This can be the case even where local emissions make only a relatively minor contribution to the total atmospheric pollutant load. Important local emission sources include vehicle traffic, wood stoves, backyard incinerators, car repair workshops, industry, dry-cleaners, intensive agriculture such as chicken and pig farms, spray painting and even cooking. Local emissions are usually intermittent rather than continuous and the areas affected are often determined by wind direction and atmospheric stability.

The most significant direct impact on air quality from the residential sector comes from emissions from heating appliances. Use of heating is very climate dependent; thus it is inappropriate to try to predict a single emission contribution for the whole country. The availability of natural gas in suburban areas is another factor that influences choice of heating fuels.

Rural areas are known to have different energy mixes from urban areas. For example, the national fuelwood study (Todd and Sangline 1989) demonstrated that in 1988, the proportion of homes using firewood as their main heating fuel was 7.4% in the whole of Victoria, 15.2% in Ballarat city and 28% in the environs of Ballarat.

Air toxics can also be released from natural sources such as bushfires, which are influenced by seasonal and regional variations and decisions about controlled burns.

In major Australian metropolitan airsheds, pollutants are being emitted every day at much the same rate. However, air pollution is only serious (measured against health standards) on a small number of occasions. A combination of geographic setting and adverse weather conditions leads to high pollution conditions (Newton 1997). The important point to note is that identical amounts of an emission do not always have the same effect, because local conditions play such an important role.

The movement of pollutants following their release into the atmosphere depends on the nature of the original emissions, such as exhaust plumes, prevailing atmospheric conditions and the stability of the pollutant.

1.5.2 The nature of emissions

How and where a substance is emitted into the atmosphere may also affect its fate. The properties of the plume (thermal buoyancy and height of release) can affect the dispersion process. Emissions from tall stacks usually travel further and are better dispersed than those arising at ground level, resulting in lower local concentrations of pollutants. Chemical reactions can vary depending on the height within the atmosphere because factors such as temperature, radiation from the sun, moisture and other gas concentrations vary with height.

However, highly unstable conditions, such as those that may occur on hot sunny days, can sometimes cause normally well dispersed emissions from high stacks to come to ground in high concentrations, resulting in short-term high exposure to pollutants in their plumes.

1.5.3 The stability of emitted pollutants

Pollutants may undergo physical and chemical transformations following their emission. These changes are closely related to meteorological conditions such as air temperature, intensity of solar radiation (sunlight), availability of water vapour or droplets and the presence or absence of other atmospheric substances. The final rate of removal of pollutants by settling, scavenging, surface absorption and impaction (ie collision with water droplets or particles) is also closely related to atmospheric state. Thus, even if emissions in a given area remain fairly constant, air quality can vary widely (Newton 1997).

The capacity of the atmosphere to absorb pollutants is sometimes referred to as its assimilative capacity. This may vary in time and place due to a range of factors. For example, the assimilative capacity will be strongly influenced by the level of atmospheric emissions impacting on an airshed. However, it is possible to reduce the assimilative capacity of natural systems (eg through changing land use) and thereby reduce the capacity of sinks (processes or places that remove pollutants from the atmosphere).

Gaseous or particulate emissions can remain airborne for considerable periods. Some emitted substances may interact with each other. In this way, or through atmospheric chemical processes and the influence of sunlight, primary emissions may turn into new, secondary substances. Not all secondary substances are harmful – chemical reactions in the atmosphere and elsewhere can sometimes change emissions into forms that are less damaging to the environment or to human health. Thus, many emitted substances are degraded and then absorbed by natural systems. However, pollutants may be generated in greater volumes than natural systems can remove. In general, long-lived emissions are responsible for regional and global impacts, while those that are short-lived and reactive lead to local impacts.

Recently, new evidence has suggested that high voltage power lines may cause cancer by making particles of pollution 'stick' to people's lungs (Henshaw et al 1999a,b). Power lines carry electricity at high voltages and the electric field strength near each power line cable is often sufficient to ionise the air. This creates streams of ions (electrically charged particles), known as corona ions, which are emitted into the air. Research indicates that corona ions from overhead power lines can give an electrical charge to pollutant aerosols, such as car exhaust particles. The additional charge can make the particles stick to surfaces. This may result in increased lung deposition for people living close to power lines, in particular those downwind from them. Research suggests that these people can experience two or three times the average daily dose of potentially damaging pollutants.

1.5.4 Meteorological and topographical effects on dispersion

Meteorology and topography are two of the most important environmental factors controlling the movement of pollutants in the atmosphere.

The wind's speed, direction and prevalence affect the directional movement of pollutants and the distance transported. The greater the wind speed, the greater the dilution of emissions. Increased wind speeds also promote increased turbulence and greater vertical mixing. The wind direction determines the shape of a plume and the path it follows. When conditions are 'unstable', such as on warm to hot sunny days, pollutants in the atmosphere are quickly mixed and pollutants soon disperse. This vertical mixing is reduced under stable conditions.

High humidity and frequent rain can wash particles and soluble gases from the atmosphere, causing some cleaning of the air before long-range transport occurs. Air pollutants washed from the atmosphere may end up in dust, soil or water.

In some circumstances, the temperature of a layer of air can actually increase with height, rather than decrease as would normally happen; this is called an inversion. Ground level temperature inversions can occur when the ground cools rapidly after sunset under conditions of light winds and clear skies. The air above the ground is cooled and becomes stable or less dispersive (less turbulent). As very little vertical mixing of polluted air takes place in an inversion layer, released pollutants become trapped under or within the stable layer of air.

The influence of topography (the shape of the land) on meteorology is also important. Topography affects rainfall patterns and can affect the drainage flow of stable air during evening and night hours. Cold, stable air flows slowly down hills and mountains at night. It traps pollutants released into the drainage layers, typically extending 50–150 m above the ground and the concentration builds up to greater levels than are experienced from the same sources during the day. The reverse of a drainage flow is an updraft, which develops on the sides of steep mountains when the first heat of the sun falls on them during the morning. Sea and lake breezes are also important in transporting and dispersing pollutants.

A well-known example of the importance of topography and meteorology is the so-called morning 'drainage' or inversion flows across the Sydney basin from west to east. These trap pollutants emitted into the flow, transport this pollution back over the metropolitan area again by the late morning and are followed by afternoon sea breezes from the east. Thus, peak oxidant levels occur in southeastern parts of the metropolitan area due to emissions several hours earlier in the western or central parts of the basin.

1.5.5 Climate of Australia

Australia is a fairly flat, sparsely populated island continent located on the western rim of the Pacific in the largely oceanic southern hemisphere. Its geographic location and size mean that it experiences many climate zones. These range from tropical climates in Australia's northern third to temperate ones in Tasmania and the southern parts of the mainland (with a small alpine region occurring in the southeast of the continent and in central Tasmania) to Mediterranean in the southwest and southcentral areas.

A significant feature of the Australian climate is its large year-to-year variability, partly due to the El Niño-Southern Oscillation (ENSO). The most pronounced variability in the Australian region occurs over the eastern two-thirds of the continent, where ENSO accounts for 30–40% of the variance in rainfall. ENSO is also detectable to some extent in the variability of air quality measurements between years, particularly the measurements of ozone in the lower part of the atmosphere, where it is a pollutant.

The cycle of wet periods followed rapidly by dry ones increases the risk of fires caused by natural conditions (such as dry thunderstorms) and by human activities.

Australia's long coastline and the large temperature difference between land and sea can greatly influence the patterns of local air circulations and thus air quality. One of the features experienced to some degree along the entire coastal fringe is a phenomenon known as the sea-breeze/land-breeze regime. During daytime, and under conditions of light or favourable winds, a temperature contrast builds up between the land and the cooler ocean, resulting in a regular onshore sea breeze. These sea-breeze circulations usually extend about 500–1000 m vertically.

During the night, stable conditions, light winds and clear skies aid the rapid cooling of the land surfaces. Air in contact with the ground cools faster than the air in the free atmosphere, so cooler (and therefore denser) air near the surface flows down any slope towards lower ground, where it may combine with other cold air flows. Since circulations of this type are typically confined to the lowest 300–500 m, even relatively low-level terrain can produce such drainage air flows. The ways in which these flows interact is what determines the extent of an airshed.

In Australia, the regional airsheds around each major city are discrete and widely separated. Sometimes, areas along the coastal fringe experience a daily reversal in the direction of air flow. This usually happens under conditions of stable atmosphere and light winds. Under these circumstances, air carried inland by the afternoon sea breeze can recirculate in overnight drainage flows. This cycle can continue for some time – up to three days in summer. Pollutants released into the airsheds are trapped within the circulation and concentrations may build up in the lower levels.

Australia receives a lot of sunshine by world standards, with most of the country receiving, on average, about six hours a day, and some areas more than 10 hours. As a result, the potential for photochemical smog formation is high in all our cities in the summer, and remains so in the winter in northern cities such as Brisbane.

Recirculation of air within urban coastal airsheds is one of the most important events linked to the formation of photochemical smog and ozone episodes monitored over urban areas (Manin et al 1994). At certain times, major population centres, including the capital cities of Sydney, Brisbane, Perth and Melbourne, are subjected to recirculation and, with it, photochemical smog formation.

1.5.6 Recirculation in urban airsheds

The recirculation of pollutants over Australia's major capital cities is important for determining pollution levels. The build-up of pollutants over the major cities during recirculation episodes often exceeds the capacity of the atmosphere to assimilate air pollution. Photochemical smog formation under such conditions poses the biggest single threat to clean air in urban Australia.

The 'Melbourne eddy' is a deep, slow, clockwise wind circulation responsible for keeping polluted air over the Melbourne area on some summer days. On days with weak easterly winds and a low-level inversion, the eddy develops in the lee of the Victorian Alps (to the east of Melbourne), trapping urban emissions. Computer models simulating the Melbourne Eddy have shown that a similar pattern more frequently occurs due to the daily cycle of land and sea breezes.

On summer days with high pollution, common in Sydney, night-time airflows from the mountain slopes move cold air to the north towards Richmond in the lee of the Blue Mountains and to the east towards the centre of the city. The flow towards the centre of Sydney accumulates pollutants as it travels over industrial and dense urban areas in the western suburbs. The air then flows out to the ocean in the morning. With the arrival of the sea breeze, the same air can frequently be returned and travel westward, reaching the Hawkesbury Basin near Penrith or Campbelltown in the afternoon. Under some circumstances, air parcels enter the Sydney basin overnight from the Hunter Valley and are caught in the sea-breeze/land-breeze circulation. This may carry the air parcel out to sea and down the coast towards Wollongong the next day.

Perth lies on a coastal plain between the Indian Ocean and the 300 m high Darling Scarp rising to the east. On summer days of high smog, surface winds are generally easterly. Pollutants from Perth and the southern industrial areas are carried out to sea in the morning, but return as smog in the early afternoon with the onset of a strong sea breeze.

Mountain ranges rise to the west of Brisbane and extend well to the north and south. To the east, Moreton and North Stradbroke islands lie less than 20 km from the shore. At midday on a winter's day in Brisbane in light northerly winds, the local winds are also light and very variable in direction. The simulated trajectory of pollutant-laden air parcels near the surface is very confused. Computer models predict that the air parcel crosses and recrosses the Brisbane urban area several times in both the morning and the evening. By the end of the day the whole region will probably have been affected by increased smog levels in the form of ozone, nitrogen oxides (NOx) or aerosols as the parcels accumulate pollutants and these are changed photochemically to smog.

Adelaide can experience poor dispersion of pollutants when light northeasterly winds and warm temperatures occur. Winds in the early morning on the Adelaide plains typically flow off the land towards the sea. However, midmorning a sea breeze develops in the Gulf, leading to southwesterly winds in the coastal region. This means peak hour emissions from early in the morning are transported to the sea and can return to the city and suburbs during late morning.

Although the geographical spread and climatic variability across the continent may produce special features for each city, airsheds are likely to have many common characteristics due to their broadly similar history, growth patterns and emission regulations (Todd 1997).


1 Although 189 hazardous air pollutants (HAPs) were originally identified by the US EPA, caprolactam was removed from the list in June 1996; therefore, the US EPA currently defines 188 HAPs

2 The Air Toxics Forum (26 May 1999, Canberra) was sponsored by Environment Australia to discuss issues relevant to the development of the air toxics strategy. The Forum engaged Commonwealth, State and Territory Governments in dialogue with key stakeholders from industry, community and environment NGOs.

3 'Photochemical oxidants' is a term used to describe a complex mixture of chemicals produced in the atmosphere by the action of sunlight. It is commonly known as photochemical smog. The principal component of photochemical oxidants is ozone; also present are formaldehyde, other aldehydes, and peroxyacetyl nitrate. Measurements of photochemical oxidants (and standards relating to them) are usually referenced to ozone. Thus, measurements of ozone are taken to be a surrogate for photochemical oxidants, and ozone will generally be used in the discussion

4 Source: Adapted from Hart (1982)


6 As a criteria pollutant, lead is not an air toxic under the definition used by the ATP; however it is included as a priority air toxic metal pollutant for indoor air due to its significance in this context.


8 There is significant scientific evidence to demonstrate that all 12 of the POPs currently identified by UNEP will undergo long-range atmospheric transfer.



11 'Congener' is a technical term that refers to one variation of a chemical structure.

12 TEQ provides a way to represent the combined toxicity of a mixture of dioxin congeners. It relies on all congeners being referenced against the most toxic of the dioxins, 2,3,7,8-TCDD, which is given a score of 1.0.

13 'Group 1' is the term given to proven human carcinogens; this category includes chemicals or groups of chemicals for which there is sufficient evidence from epidemiology studies to support a causal association between exposure and cancer.






19 From AP-42 Compilation of Air Pollutant Emission Factors. Vol. 1 Stationary Point and Area Sources. US EPA, available on the internet at 

20 Testing of chemicals for mutagenicity in Salmonella typhimurium is based on the knowledge that a substance that is mutagenic in the bacterium is likely to be a carcinogen in laboratory animals, and thus, by extension, present a risk of cancer to humans. Although about three-fourths of chemicals that are positive in the salmonella test are found to be rodent carcinogens, not all substances that cause cancer in laboratory animals are mutagenic in this assay. Not all substances that cause cancer in laboratory animals will cause cancer in humans.


22 This project was an adjunct to Diesel NEPM Project 2.2 which measured the emissions performance of a representative sample of Australia's urban in-service diesel fleet under the conditions of the composite urban emissions drive cycles. Eighty vehicles from a range of age groups and vehicle classes were evaluated. Emissions measured were oxides of nitrogen, carbon monoxide, carbon dioxide, oxygen, total hydrocarbons, visible smoke and particles. The final report of Diesel NEPM Project.


   Links to another web site
   Opens a pop-up window