Radiative forcing continues to be a useful concept, providing a convenient first-order measure of the relative climatic importance of different agents (SAR; Shine and Forster, 1999). It is computationally much more efficient than a GCM calculation of the climate response to a specific forcing; the simplicity of the calculation allows for sophisticated, highly accurate radiation schemes, yielding accurate forcing estimates; the simplicity also allows for a relative ease in conducting model intercomparisons; it yields a first-order perspective that can then be used as a basis for more elaborate GCM investigations; it potentially bypasses the complex tasks of running and analysing equilibrium-response GCM integrations; it is useful for isolating errors and uncertainties due to radiative aspects of the problem