Workshop Program
Register now
Workshop Program

Accounting for covariance in emissions' inventories
by Prof. Maurice Cox, Dr. Andrea Merlone, Dr. Tom Gardiner


The comprehensive approach adopted in the Kyoto Protocol expresses total emissions in terms of CO2e (CO2 equivalent) based on a sum of contributions of the form (Global Warming Potential) times (Activity) times (Emissions’ Factor) in which 'Activity' is the activity for a specific source category, 'Emissions' factor' relates to emissions of a given pollutant from that category, 'Global Warming Potential' is the factor (GWP100) converting emissions for a particular gas to ‘CO2e’ on a 100 year basis. For UK emissions’ submissions from the three largest GHG contributors (CO2, CH4 and N2O), most of the uncertainty in the inventory comes from the Agriculture, Land Use and Waste sectors, in all contributing 86 % of the uncertainty in the total inventory emissions. Examination of the Agriculture and Waste sectors shows that certain activities take identical values, implying they have the same origin and hence are common to the according emissions' contributions. Calculations were carried out in two cases: accounting for correlation due to this commonality and ignoring it. The results differed by no more than 2 %. Although this increase is small, the principle of accounting for correlation stands and in other areas is expected to be more significant. Although the Intergovernmental Panel on Climate Change (IPCC) has stated that GWP values have significant uncertainties, which should be considered in the overall assessment of total equivalent emissions, seemingly such account is rarely made. Accordingly, we made some analyses of this effect, taking Waste Incineration (IPCC category 5C) as a case in hand. The Fourth Assessment Report of the IPCC in 2007 stated the expanded uncertainties of these GWPs are 35 % (90 % confidence). For CO2, for instance, the same GWP factor is (correctly) applied to emissions from chemical and clinical waste. If this factor were applied independently, natural if different parties were involved, an estimate of total CO2 emissions would be obtained, and an associated uncertainty evaluated. By considering the commonality of the GWP, factor, accounting for covariance, the uncertainty is 24 % greater than neglecting it. This considerable effect should not be ignored. Since the model used is non-linear, the method recommended in the Guide to the expression of uncertainty in measurement (GUM) is approximate. We thus applied a Monte Carlo method, giving agreement of results for the two approaches modulo the uncertainties involved.

Video presentation

Open in a new tab


Open in a new tab

Topic : Theme 2: State of play in integrated approaches for advanced GHG emission estimates and the way forward to operational services.
Reference : T2-B9

Back to the list of submissions
Previous submission · Next submission


Comments are only accessible to participants. If you are a participant, please log in to see them.