Individual versus averaged estimates of parameters used in large scale emissions inventories

Abstract

Development of emissions inventories for large geographic areas is currently based on the use of ‘mass balance’ equations to estimate the emissions from each source. The general equation used states that the emissions from a particular source for a specific pollutant in a specified time frame are equal to the product of the activity rate of the source in the selected time frame, an emission factor, which estimates the amount of pollutant emitted per unit activity, and additional parameters which vary according to the type of source whose emissions are being estimated. Values of the parameters used to estimate emissions are estimated in two different ways. ‘Individual’ estimates (an estimate for each source considered) are used for certain parameters; for other parameters, such as the emission factor, ‘averaged’ estimates (average values obtained by sampling a subset of the sources in a category) are used. The statistics resulting from the use of these two different methodologies are quite different. Derivation of the equations needed to calculate the mean square error in each case show that when the sources included in a category are homogeneous, then the averaged estimator may be better than individual estimates for estimating emissions from a particular source, but is worse for estimating sums. A numerical simulation was conducted to demonstrate these results. An example showing how to account for the effect of averaged estimators in a simple assessment is also included.