In this paper, we review existing and alternative metrics of climate change, with particular emphasis on radiative forcing and global warming potentials (GWPs), in terms of their scientific performance. Radiative forcing is assessed in terms of questions such as the utility of the concept, uncertainties and sensitivity to key assumptions. The assessment of emission indices focuses on the climate and other resulting impacts (end points) against which emissions are weighted; the extent to which (and how) time dependence is included, with regard to both emission control and impact; how cost issues are dealt with; and the sensitivity of the metrics to various assumptions. It is concluded that the radiative forcing concept is a robust and useful metric of the potential climatic impact of various agents and that there are prospects for improvement by weighing different forcings according to their effectiveness. We also find that although the GWP concept is associated with serious shortcomings, it retains advantages over any of the proposed alternatives in terms of political feasibility. Alternative metrics, however, make a significant contribution to addressing important issues, and this contribution should be taken into account in the further development of refined metrics of climate change.