Asbestos waste in Australia

Data quality limitations

Before presenting the findings of the analysis, it is worth reviewing the data quality limitations and considering how these constrain the analysis.

Intra-state tracking system / transport certificate data

Hazardous waste tracking systems are maintained primarily to enable checking of transport certificates and operators in the event of suspected illicit activity involving hazardous waste. The records are infrequently collated, and lack of data or even errors in particular fields may not be readily recognised or followed up. Some of the difficulties with the analysis of transport certificate data included that:

1. 
   There is some uncertainty about how wastes are recorded on transport certificates. In particular:
   

• 
  soil that is contaminated with asbestos could be recorded as ‘contaminated soil’ or as ‘asbestos’ or ‘asbestos containing material’.
  
• 
  demolition waste that is contaminated with asbestos could be recorded as either part asbestos or (for example, in NSW) all asbestos (or all asbestos containing material).
  

Guidance to waste generators on which of these to select under particular circumstances may not be consistent between or within jurisdictions. For example, the Qld tracking system covers asbestos but not contaminated soil, so mixed soil and asbestos would be recorded as asbestos in its tracking system. In other jurisdictions advice may vary, potentially leading to incompatible data across jurisdictions or years.

2. 
   Some transport certificates record asbestos as sent to ‘storage’. This material is likely to be subsequently taken to landfill under another transport certificate, leading to double counting. (This is likely to be a relatively small issue.)
   
3. 
   Some asbestos waste is not included. Jurisdictions vary in the extent to which they require tracking of small asbestos loads and domestic self-haul. Similarly there may be variation in the degree of asbestos contamination before a waste must be classified as asbestos. Asbestos waste that is illegally dumped will not be included in tracking system unless it is identified and collected and disposed to a licensed facility.
   
4. 
   Australian and New Zealand Standard Industrial Classification (ANZSIC) codes are not always well recorded, limiting our ability to analyse sources. One of the problems is that ANZSIC codes vary from 2-4 digits depending on the level of disaggregation of industry type, but tracking system software seems to require a consistent number of digits. Jurisdictions therefore design their own ‘ANZSIC-like’ coding systems.
   
5. 
   Analysis of the transport and fate of asbestos was limited by the fact names are often recorded or spelled differently, e.g. XYZ; X.Y.Z.; XYZ Pty Ltd; XYZ Demolitions. Reporting such information is also limited by commercial confidentiality requirements.
   
6. 
   Units of measurement are variable. Data may be recorded by weight or volume, requiring assumptions about density⁴.
   
7. 
   A small proportion of records contain clear errors, e.g. recording asbestos as liquid, or its treatment as ‘surface impoundment’ or ‘recycling’.
   

A review of jurisdictional tracking systems covering some of these issues is given in section 3.2.

Landfill data

Landfill data, for example from NSW, is self-reported and may not be audited at the material level.

Data discrepancies

Another data quality problem was discrepancies between overall tonnage data from different sources for the same jurisdiction and year. These were of two types:

Differences between data from landfill records and from transport certificates. Where data from these two sources was available for a jurisdiction and year, they were mostly within a few percentage points of each other. In one case, however, the difference was almost 60%. Possible explanations include:

• 
  Item 1 above: a landfill would be likely to record a load based on its management requirements at the landfill, not based on the transport certificate
  
• 
  small loads and domestic self-haul of asbestos may not subject to tracking requirements so would be included in one data set but not the other
  
• 
  waste from natural disasters may not always have been consistently recorded in the past⁵.
  

Differences between data submitted for Basel reporting purposes and data obtained for this research from either transport certificates or landfill records. The sources of each jurisdiction’s historical Basel submission data are not known. Some (e.g. Tas) appear to be incomplete. In some cases the values reported were very different from those collated by the project team from transport certificates, or reported by jurisdictions from landfill records or transport certificate data. In each of NSW, Qld and SA the differences were as high as 100% in at least one year. We are unable to determine the cause of these differences.

Where there were differences between the available data for a jurisdiction in a year, we selected data based on considerations of: consistency of source; consistency of scale with adjacent years; and age of the data (recent data sets are likely to be more accurate than older data sets). The sources of the tonnage data used are shown in Table 1.

Notwithstanding the data quality limitations, the data analysis revealed some interesting and useful traits as discussed below.

3. Tonnages and trends

Quantitative trends in the disposal of asbestos⁶ are shown in tonnes per year in Figure 1 and in kilograms per capita per year in Figure 2. The following observations can be made based on these graphs:

1. 
   Asbestos quantities vary significantly between years and jurisdictions. Spikes are often associated with particular large projects such as the Bradmill factory site redevelopment in Melbourne.
   
2. 
   On average, over the seven financial years from 2008 to 2014, Australians generated around 20kg of asbestos waste per capita⁷.
   
3. 
   NSW dominates asbestos waste generation on absolute and per capita measures. In all but one of the years for which NSW data is available, it records more asbestos waste than all the other jurisdictions combined. In NSW, waste with even the smallest proportion of asbestos contamination must all be classified as asbestos waste, so asbestos waste may include significant proportions of other demolition materials.
   
4. 
   In recent years (where the data quality is likely to be best), per capita rates of asbestos waste disposal in Qld, SA, Vic and WA have been similar.
   
5. 
   A trend of rising quantities of asbestos waste is apparent in several jurisdictions and nationally. Based on this data set, a recent study for the Australian Department of the Environment projected asbestos waste quantities to continue rising for the next 20 years at 2.8% per year, consistent with forecast long-term economic growth rate (Blue Environment et al. 2015).
   

Figure 1: Asbestos quantities disposed by jurisdiction and year

Notes: Qld and Vic tracking system data were corrected to remove ‘storage’, which was assumed to be double-counted. SA data may include some limited double-counting. In NSW, waste with even the smallest proportion of asbestos contamination must all be classified as asbestos waste, so asbestos waste may include significant proportions of other demolition materials.


Figure 2: Asbestos quantities disposed per person by jurisdiction and year

for some years, act, nt and tas. in nsw, waste with even the smallest proportion of asbestos contamination must all be classified as asbestos waste, so asbestos waste may include significant proportions of other demolition materials." align=bottom width=605 height=340 border=0>

* The data for Australia includes NSW, Qld, SA, Vic, WA and, for some years, ACT, NT and Tas. In NSW, waste with even the smallest proportion of asbestos contamination must all be classified as asbestos waste, so asbestos waste may include significant proportions of other demolition materials.

4.