Comparisons of emissions. Comparison of emission inventories

An important aspect of quality considerations is the comparison with independent sets of data. Ideally, this would be ambient measurement data, as such ambient data also allow for a closure of cause (emissions) to effects (pollution). Ambient measurement data in sufficient quantity and quality are however only available for very specific campaigns. But there are also cases when emission inventories can be compared. Especially this is possible for inventories that have been assessed independently for the same area. A number of methods have been developed to describe the similarities and differences between such inventories. If discrepancies are flagged, it is possible to discuss and potentially decide upon an ideal approach. Even if such a decision is not possible, a comparison allows for an understanding of the reliability and the uncertainty of the input used to make up the inventory. A necessary precondition for this concept is that the two inventories are independent which are compared in such a way. It is usually quite difficult to approaches and consequently also similar input data. The largest difference in concept is between bottom-up inventories (made up from information about a multitude of individual sources) and top-down inventories which combine large scale data (as national totals) with surrogate information for disaggregation. Here it is possible to identify conceptual discrepancies, but not the detailed methodologies. The latter only can be performed when a closer concept is being compared, and when additionally the input information is made available for the comparison. There is several levels at which emission inventories can be compared. The following sections will give a more detailed account of this. Basically, comparisons are possible along the following paths: - Emission total (by chemical species) - Sector emission total (by chemical species) - Spatial distribution - Temporal distribution The respective pathways of differentiation reflect the diverging needs of users of emission inventories. While modellers are first of all interested in emission totals and the spatial and temporal resolution of these emissions, which all are basic input parameters, policy makers require information on individual sources and source groups and therefore also on the confidence of the respective information. The attribution of emissions to different source groups, however, is a common cause of discrepancy between two inventories. While this is not relevant for atmospheric chemistry, policy decisions are based upon such information and consequently the topic is extremely relevant.