concept is inappropriate to predict the sign or the magnitude of the global mean precipitation changes due to both scattering and absorbing aerosols, which affect precipitation differently in summer and winter.
Another limitation of the traditional radiative forcing concept is that it does not adequately characterize the regional response. Regional radiative forcings from atmospheric aerosols, tropospheric ozone, or land-use and land-cover changes can be much larger than global mean values. A regionally concentrated forcing may lead to climate responses in the region, in another region via teleconnections, or globally—or may even have no climate response.
Yet another limitation of the concept is that the assumption of a constant, linear relationship between changes in global mean surface temperature and global mean TOA radiative forcing does not always hold. This linear relationship breaks down for absorbing aerosols, which may have small TOA forcing, but disproportionately larger surface forcing due to absorption of solar radiation (Lohmann and Feichter, 2001; Ramanathan et al., 2001a). This motivated the introduction of the concept of efficacies of different forcing agents (Joshi et al., 2003; Hansen and Nazarenko, 2004). “Efficacy” is defined as the ratio of the climate sensitivity parameter λi for a given forcing agent to λ for a doubling of carbon dioxide (CO2) . The efficacy E is then used to define an effective forcing fe = f E. Table 4-2 summarizes the forcings, responses, efficacies, and effective forcings of different forcing agents from several models. Efficacies greater than 1, such as for black carbon impacts on snow and ice albedo, correspond to a larger effective forcing than that of 2 × CO2 (Table 4-2). On the other hand, scattering sulfate aerosols are less efficient than greenhouse gases in changing the surface temperature for a given forcing.
Overall, after weighing the strengths and limitations of the traditional radiative forcing concept, the committee finds that its strengths warrant continued use in scientific investigations, climate change assessments, and policy applications. The concept is relatively easy to use, particularly in enabling efficient comparisons between different forcing agents, forcing scenarios, and climate models. Further, it has clear applications within the climate policy community. Nonetheless, the limitations call for broadening the concept to account for nonradiative forcing, spatial and temporal heterogeneity of forcing, and nonlinearities. This chapter presents specific approaches to address these limitations.
Hansen et al. (2002) introduced the concept of fixed sea surface tem-