Treatment of Uncertainties for National Estimates of Greenhouse Gas Emissions
Global Atmosphere Division within the Department of the Environment, Transport and the Regions (DETR) has the UK policy lead on climate change. Annual updates of the UK Greenhouse Gas Inventory are reported to the UN Framework Convention on Climate Change (UNFCCC). The parties are encouraged to assess uncertainties in the reported estimates and makes some suggestions, but no definitive guidelines are given.
This study has two main objectives:
- To review of methods of assessing uncertainty - in the context of their applicability to the UK emissions inventory
- To estimate as far as is practical the uncertainty in estimates - both of emissions in 1990 and in projected emissions for the year 2010, and in the projected percentage change from 1990 to 2010
An important issue to be explored in this study related to the role of uncertainty in assessing compliance with the UK's international obligations on climate change.
This study examines the UK emissions and sinks of the six greenhouse gases that are limited by the Kyoto agreement:
Nitrous oxide (N2O)
Sulphur hexafluoride (SF6)
In order to provide a firm basis for the uncertainty analysis, information relating to UK inventories of these greenhouse gases was first reviewed. In parallel, the features, advantages and disadvantages of the main methods for assessing uncertainties were studied. These methods range from qualitative and semi-quantitative techniques designed in the main to qualify the quality of the data used to estimate the Inventory, and hence the Inventory estimates themselves, to quantitative methods able to transmit numerical uncertainties in the input parameters to the resulting Inventory estimate.
No assessment was made in this study of the uncertainty of any projected fuel consumption or economic data. This study is limited to the uncertainties that arise from the emission inventory calculation methodology itself.
Following the initial appraisal of the UK situation, sufficient information was available or could be uncovered readily, or alternatively, sufficiently robust judgements could be made, to justify the use of the quantitative methods. The transmission of parameter uncertainties to those in the Inventory was accomplished using the software tool @RISK. The numerical analysis was therefore conducted for each of the individual gases listed above for 1990 and 2010, and for the global warming potential-weighted total.
The total warming potential of the net emissions of the six greenhouse gases considered in this study are predicted to fall on average by about 6% between 1990 and 2010. This is much lower than the predicted uncertainties in the overall emissions in these years (just under 20%), although the reduction is still statistically significant. The predicted fall in total global warming potential does not arise primarily from a decrease in emissions of any single gas. The results show that the distribution of possible change in the total emissions ranges from around -2% to about -10%.
The results for the range of gases studied vary. For methane, nitrous oxide and the PFCs, the best estimate emissions fall significantly, and their attendant uncertainties do not change this conclusion. For SF6 and the HFCs, emissions are predicted to rise significantly between 1990 and 2010 (with respect to their uncertainties). This is caused by changes in their uses, together with a shift in the stages in the life cycles of most of the substances of concern (i.e. phasing out of certain substances from common use, recycling of refrigerators, disposal of existing inventories). For carbon dioxide, no overall significant change is anticipated between 1990 and 2010, although increases in emissions from such sources as the burning of natural gas in power stations and the increase in transport is partly countered by decreases in most other sectors.
Owing to the large number of contributions to the overall uncertainty in emissions and sinks of individual greenhouse gases in the UK, the statistical procedure employed was only able to identify the largest contributions in this regard. No single parameter (or factor) dominated the uncertainties in the emissions of carbon dioxide, methane, nitrous oxide or the aggregate total significantly. Many sources of uncertainty contribute to the overall effect for these gases, and none of them dominate as their individual effects are diluted. For HFCs, PFCs and SF6 there are fewer sources (and hence model input parameters), particularly in 1990, and so some of them turn out to be significant sources of the uncertainties in their respective totals.