Greenhouse Gas Emissions from Spontaneous Combustion at Open Cut Coal Mines: An Investigation of Air Quality Modelling and Inverse Techniques

While spontaneous combustion of coal has been recognised by the Inter-Governmental Panel for Climate Change (IPCC) as a potential source of greenhouse gas emissions, it has been excluded from greenhouse gas inventories as it is considered that there is no acceptable method for estimating the emissions.

In recognition of this, ACARP has carried out two previous projects to explore methods for establishing greenhouse gas emissions from spontaneous combustion. The first project, C8059 (Carras et al, 2000) sought to provide methods, supported by direct measurement, to quantify the emissions. Measurements of emissions from spoil piles, coal rejects and tailings were conducted at 11 mines in the Hunter Valley in NSW and the Bowen Basin in Queensland using a chamber technique. While the project significantly advanced the knowledge of emissions from spontaneous combustion in open cut coal mines, there were practical problems in applying the results to estimate greenhouse gas emissions from operating mines.

The second project funded by ACARP, C9062 (Carras et al, 2002) used airborne infra-red thermography to investigate whether more accurate and cost-effective monitoring of the extent of spontaneous combustion in spoil piles and the associated greenhouse gas emissions could be achieved. The main conclusion drawn by Carras et al (2002) was that airborne infra-red data could be used to monitor the long term behaviour of spoil piles subject to spontaneous combustion. However, due to the complexity of the processes involved in producing heating and its surface manifestation and the associated emissions of greenhouse gases, the emissions estimates were still subject to significant uncertainty.

At the conclusion of projects C8059 and C9062 the focus of the research centred on reducing the uncertainty associated with the emission estimates.

The aim of the current study is to investigate a method for determining the emissions of CO2 from spontaneous combustion in spoil piles at open cut coal mines, through the use of air quality methods, including modelling and inverse techniques.

The approach used was to model the large sources of CO2 in the Hunter Valley using a computer based air quality model (TAPM) which has been widely used in air pollution studies in Australia. An investigation of CO2 sources in the Upper Hunter Valley has shown that spontaneous combustion and power stations can give rise to significant concentrations at ground level. However the impact of the power stations emissions are most pronounced during the day time hours while the impact of the spontaneous combustion emissions are most pronounced during the night time. This is because the former are elevated while the latter are ground level sources. This suggests that concentration measurements should focus on data during the night time period. While the emissions from road traffic and rail are significant their ground level concentration signature is not as pronounced as for the other two major sources.

Consideration of results of the air quality modelling suggests that monitoring sites for the inverse modelling should be sited such that;

• The location point should be sufficiently close to the spontaneous combustion sources to enable a large measurable signal.
• The site should be chosen on the basis of meteorology to best capture the likely CO2 spontaneous combustion signal
• The sites should be chosen to minimise the influence of other sources.

In order to examine some of the issues that would need to be taken into account in using air quality methods, concentrations were predicted for three chosen locations and examined in greater detail. The predicted concentration time series were used as surrogates for measured data and a simple ratio method used to produce results which highlighted the uncertainties associated with this approach.

In addition to these inherent uncertainties there will also be those due to the non ideal nature of the data and the fact that the air quality model still represents an approximation to reality. Supplementation of the CO2 concentration data with other micrometeorological data would constrain the data and reduce the uncertainty. In addition concurrent ground based field measurements of CO2 traverses across the plume coupled with the micrometeorological data and detailed plume modelling would provide the basis for a robust methodology to estimate greenhouse gas emissions from spontaneous combustion.