portant not just for absorbing aerosols but also for the gaseous species. Traditionally, the notion has been that it is enough to give the tropopause forcing for the well-mixed gases in order to obtain an estimate of the surface temperature response. However, to assess climate response beyond surface temperature change (e.g., changes in precipitation, latent heat release from surface, or in the surface heat and moisture balance), it becomes necessary to understand the surface radiative forcing for all forcings. Further, to understand the difference in the zonal-mean hydrologic response between different forcings, it is necessary to look at the surface terms (e.g., Chen and Ramaswamy, 1996).
Forcings with significant spatial variability can have regional magnitudes much greater than their global averages. Aerosols, and to a lesser extent tropospheric ozone, have shorter lifetimes than the well-mixed greenhouse gases, and therefore their concentrations are higher in source regions and downwind (e.g., Charlson et al., 1991; Kiehl and Briegleb, 1993; Mickley et al., 1999). Forcing due to land-use and land-cover changes also has significant spatial heterogeneity, leading to spatial variability in the associated climate response. The traditional global mean radiative forcing provides no information about this regional structure, so many researchers have begun to present estimates of radiative forcing on a regional scale as derived from models or observational campaigns.
A large number of modeling studies have been carried out to characterize the spatial variability in aerosol forcing due to direct, indirect, and semidirect effects (IPCC, 2001). Regional effects of aerosol forcing are large; regional mean values of anthropogenic aerosol radiative forcing can be factors of 5 to 10 higher than the global mean values of 0.5 to 1.5 W m−2 (IPCC, 2001). Comparisons with satellite radiation budget data can be used to constrain model results. For example, the calculations of Haywood et al. (1999) showed that the clear-sky outgoing flux at the TOA over oceans yields excellent agreement with Earth Radiation Budget Experiment (ERBE) observations when aerosol species are considered. This is a useful test of the chemical transport model (CTM)-derived concentrations of aerosols and assumptions about their sizes, at least in terms of their collective reflective ability. More recent computations from the National Center for Atmospheric Research and Geophysical Fluid Dynamics Laboratory models bear this out with updated CTM simulations. Soden and Ramaswamy (1998) inferred the existence of spatial aerosol effects in satellite datasets. Observations over source regions and downwind show very large forcings (see Box 4-1). High regional concentrations of scattering aerosols can completely offset the positive forcing due to increases in