Australian inventory of dioxin emissions

Technical Report No. 3
K Bawden, R Ormerod, G Starke and K Zeise
Department of the Environment and Heritage, May 2004
ISBN 0 642 54995 8

Download

Executive summary

As part of the Australian Government’s National Dioxin Program (NDP), Pacific Air & Environment (PAE) was commissioned by the Australian Government Department of the Environment and Heritage (DEH) to compile an inventory of dioxin emissions to the Australian environment for the calendar year 2002. The inventory is an update to the dioxin emissions inventory, 'Sources of Dioxins and Furans in Australia: Air Emissions - Revised' (EA, 2002a).

The previous emissions inventory estimated emissions to air of dioxins and furans based on Australian specific data and internationally published dioxin emission factors. The updated inventory includes emissions to air, water and land of dioxins and furans and is based on a prescribed international protocol namely, the 'Standardized Toolkit for Identification and Quantification of Dioxin and Furan Releases' (UNEP, 2003) (hereafter referred to as the UNEP Toolkit).

Reservoir sources of dioxins and furans are estimated under a separate cover and are not included in this inventory.

Since the publication of the original inventory, dioxins have become a reportable substance under the National Pollutant Inventory (NPI), a national program designed to provide information on the types and quantities of pollutants emitted by anthropogenic activities throughout Australia. Where possible, data and information from the NPI have been incorporated into the inventory. However, dioxin emissions reported in response to NPI requirements are designated as kilograms per annum (kg/a) whereas under the UNEP protocol emissions are reported on a mass toxic equivalence per year (g TEQ/a). Any comparison between data generated by the NPI and data provided in the dioxins inventory would require the emissions to be converted from mass (kg) to mass toxic equivalence (g TEQ), or vice versa.

The methodology used to compile this inventory follows the UNEP Toolkit guidelines and was performed as follows:

  • Dioxin emission sources to air, water and land were identified in accordance with the classification provided in the UNEP Toolkit
  • Activity data were collected on a national basis (e.g. material throughput, production data) from publicly available information sources for each emission source category
  • Dioxin emission factors presented in the UNEP Toolkit judged suitable for Australian conditions were identified for each emission source category
  • Dioxin emissions were estimated based on default UNEP emission factors and publicly available, nationally based, source activity data (i.e. this inventory is not a facility based inventory. Estimates of dioxin emissions were generated on a national level)
  • Default emission estimates and report sections were supplied to industry peak bodies and government agencies for review and comment
  • Estimated dioxin emissions were updated based on the response from industry peak bodies and government agencies.

Where possible, an estimate has been made of the possible range of dioxin emissions from each emission source category. The estimate of the range of dioxin emissions from each emission source category is based on the following:

  1. In instances where emission testing data and specific process information are not available, the highest and lowest UNEP emission factors are used to estimate the range of dioxin emissions. The average between the high (H) and low (L) emission factor is used to generate a best estimate (BE) of dioxin emissions from the emission source category.
  2. In instances where dioxin measurement data are available, the highest and lowest measured dioxin concentration is used to derive the high and low emission estimate. The median dioxin concentration is used to estimate the best estimate of dioxin emissions from the emission source category.

The estimate of the range of dioxin emission are 1.4 g TEQ/annum to 1,780 g TEQ/annum for air, water and land.

The preparation of this inventory and sources of dioxin emissions are categorised as follows (as prescribed in the UNEP Toolkit):

  1. Waste incineration
  2. Ferrous and Non-Ferrous Metal Production
  3. Power Generation and Heating
  4. Mineral Products
  5. Transport
  6. Uncontrolled Combustion Processes
  7. Production of Chemicals and Consumer Goods
  8. Miscellaneous
  9. Disposal/Landfill.

Table ES1 presents summary dioxin emission estimates for each category included in the inventory.

The category with the highest estimated emission is uncontrolled combustion (i.e. biomass burning, waste burning and accidental fires). Uncontrolled combustion is estimated to contribute to nearly 70% of total emissions to air and over 80% of total emissions to land. Disposal and landfilling is estimated to be the largest source of dioxin emissions to water, contributing over 75% of total emissions.

Based on Table ES1 uncontrolled combustion processes, which are significantly influenced by emissions from grass fires, contribute approximately 75% of all emissions in Australia. This high estimate results from the combustion of a large mass under uncontrolled conditions associated with higher dioxin formation potential.

The other major emission categories, in order of decreasing emissions, are:

  1. Ferrous and non ferrous metal production
  2. Production of chemicals and consumer goods
  3. Power generation and heating
  4. Disposals/Landfilling
  5. Waste Incineration.

Table ES2 summarises the information from Table ES1 by subcategory. Only the top 25 emitters are included in Table ES2 as emissions produced by the remaining subcategories become insignificant (Total 0.9 g TEQ/year over 26 subcategories).

The limitations of the study are discussed below:

  • The most significant limitation is the lack of source test data for Australian sources, resulting in a heavy reliance on default UNEP Toolkit emission factors based on international dioxin measurement data. Emission factors as a tool for estimating emissions are inherently prone to uncertainties as they are typically based on limited testing of a source population. When applying these international emissions data to Australian sources this uncertainty is increased due to potential differences in process technologies, operating conditions and practices and pollution control equipment
  • In addition to the above point, some UNEP Toolkit emission factors were originally derived using assumed conversion factors to supplement data gaps. Examples of this include assumed fuel heat values, densities and flue gas conversion rates (i.e. m³/tonne) to derive emission factors. This adds further to the uncertainties in the original emission factors
  • The emission factors for a large number of sources span several orders of magnitude. This is indicative of the potentially large variations that are observed within a particular emission source category. With such large ranges it becomes difficult to identify significant dioxin contributors, particularly if the upper bound indicates that the source may be significant, while the lower bound indicates a minor contribution. As the estimated emission ranges indicate, a source thought to be significant may in fact be quite minor. Better source characterisation and source test data will enable greater confidence in smaller ranges
  • Some source categories may have such variable process technologies, operational conditions etc. that it may be difficult to reliably predict emissions from these sources using limited data. Better characterisation of industry will enable the identification of these industry types. Industries where little variation is encountered could perhaps base emission factors upon more limited test data. This type of characterisation would initially be important for sources considered potentially large emitters
  • Emission factors for many industries are based on test data taken during very short sample periods. The emission results are likely to be reflective of relatively good combustion and operational practice and therefore may not be indicative of likely emissions during process upsets and/or abnormal operation.

Considering these limitations, it is stressed that the emission estimates as determined by this study are INDICATIVE only of the likely dioxin releases by various sources in Australia. The estimates have been developed based on the best information currently available.