Changes in Rio Tinto's total emissions per tonne of product can result from changes in both the balance of our product mix and in the level of emissions from our production processes. For example, in 2000 there was a significant reduction in our overall emissions per tonne of product as a result of the acquisition of bulk mining iron ore operations. These operations have very low levels of emissions associated with each tonne of product and as such, they lowered the average emissions per tonne of product that we produced. We refer to this product mix related change as an intensity effect. On the other hand, Lihir is currently partially replacing its oil fired power station with a geothermal power station and this will reduce the emissions associated with its gold production. We call this change efficiency related. Rio Tinto considers efficiency changes, not intensity changes to be better measures of underlying performance. Therefore, the information presented in this section is based on changes in efficiency only.

Between 1998 and 2001 we had a target based upon our efficiency per tonne of product in terms of on site greenhouse gas emissions. During this period our emissions per tonne of product decreased by 4.8 per cent. We now believe, however, that total greenhouse gas emissions provide a better measure of performance against which to set targets. Using total emissions as the measure, our emissions per tonne of product increased by 4.2 per cent between 2002 and 2003, offsetting earlier improvements.

Reasons for the increase include:

• there were production decreases at Palabora, Richards Bay Minerals and Kennecott Utah Copper, however, total emissions remained at similar levels. In addition, more paraffin was used in the drying process at Richards Bay Minerals; and
• poor anode quality during a process excursion at Boyne Smelter resulted in increased process emissions and increases in the amount of natural gas used to bake replacement anodes. Additionally, there was an increase in the emission factor for purchased electricity and more diesel was used due to the use of mobile air compressors.

Over the past five years there has been a 0.4 per cent increase in total greenhouse gas emissions per tonne of product. This reinforces the need for quantitative and specific targets to drive improvement. Our target is a four per cent reduction in total greenhouse gas emissions per tonne of product between 2003 and 2008, based on efficiency improvements. A four per cent reduction due to improved efficiency is equivalent to approximately 1.3 million tonnes carbon dioxide equivalent (CO₂-e).