Reliable and Valid Identification of a Small Number of Global Emission Scenarios

Numerous scenarios of global greenhouse gas emissions have been created, that are difficult to communicate to decision makers. To identify few significantly different and consistent scenarios is time consuming, requires deep understanding of the underlying driving forces, and may depend on the individual perspective of the scenario analyst. Developed from an expert based scenario technique a new method was developed, which in step 1 analyzes each given scenario with respect to the relations between its characteristics (e.g. parameters, state variables). This analysis may include a very large number of qualitative ('nominal'), logical, ordinal and metric characteristics. In step 2, a few consistent and significantly different scenarios are reliably determined according to modifications of three recently published scenario identification methods. The comparison of the different methods for scenario identification shows the convergent validity of the methodology. The presentation sketches the mathematical background of the analysis and of the identification and shows results of an application to the IPCC emission scenarios. It is concluded that the quantitative scenario selection procedure presented is very helpful for the communication of scenarios to decision makers. Because the mathematical methodology complies with approaches used in qualitative scenario techniques, in which experts estimate scenario consistency, a combination of qualitative and quantitative knowledge could be possible, but has not yet been investigated.