Probabilistic temperature change projections and energy system implications of greenhouse gas emission scenarios

This paper explores the implications for global average temperature change of a set of reference and mitigation scenarios in a probabilistic framework. First, we use published probability density functions for climate sensitivity to investigate the likelihood of achieving targets expressed as levels or rates of global average temperature change. We find, for example, that limiting warming to 3 C above pre-industrial levels with at least a medium likelihood requires cumulative emissions reductions on the order of 30-60% below one unmitigated reference scenario by 2 100, while a more favorable baseline scenario requires no reductions at all to achieve this outcome with the same likelihood. We further conclude that the rate of temperature change may prove to be more difficult to control, especially if most of the mitigation effort is postponed until later in the century. Rate of change targets of 0.1-0.2 degrees C/decade are unlikely to be achieved by a target for the long-term level of climate change alone. Second, we quantify relationships between mitigation costs and the likelihood of achieving various targets and show how this depends strongly on the reference scenario. Third, we explore relationships between medium-term achievements and long-term climate change outcomes. Our results suggest that atmospheric concentrations and the share of zero-carbon energy in the middle of the 21st century are key indicators of the likelihood of meeting long-term climate change goals cost-effectively. They also suggest that interim targets could be an effective means of keeping long-term target options open. Our analysis shows that least-cost mitigation strategies for reaching low climate change targets include a wide portfolio of reduction measures. In particular, fundamental long-term structural changes in the energy system in these scenarios are a necessary but not sufficient condition to achieve high likelihoods for low temperature targets. The cost-effective portfolio of emissions reductions must also address demand-side measures and include mitigation options in the industry, agriculture, and the forest sector. (c) 2006 Elsevier Inc. All rights reserved.