2.9.6 Spatial Patterns of Radiative Forcing and Surface Forcing
Figure 6.7 of Ramaswamy et al. (2001) presented examples of the spatial patterns for most of the RF agents discussed in this chapter; these examples still hold. Many of the features seen in Figure 6.7 of Ramaswamy et al. (2001) are generic. However, additional uncertainties exist for the spatial patterns compared to those for the global-mean RF. Spatial patterns of the aerosol RF exhibit some of the largest differences between models, depending on the specification of the aerosols and their properties, and whether or not indirect cloud albedo effects are included. The aerosol direct and cloud albedo effect RF also depend critically on the location of clouds, which differs between the GCMs. Figure 2.24 presents illustrative examples of the spatial pattern of the instantaneous RF between 1860 and present day, due to natural plus anthropogenic agents, from two GCMs. Volcanic aerosols play a negligible role in this calculation owing to the end years considered and their virtual absence during these years. The MIROC+SPRINTARS model includes an aerosol cloud albedo effect while the Geophysical Fluid Dynamics Laboratory Coupled Climate Model (GFDL CM2.1) (Delworth et al., 2005; Knutson et al., 2006) does not. Radiative forcing over most of the globe is positive and is dominated by the LLGHGs. This is more so for the SH than for the NH, owing to the pronounced aerosol presence in the mid-latitude NH (see also Figure 2.12), with the regions of substantial aerosol RF clearly manifest over the source-rich continental areas. There are quantitative differences between the two GCMs in the global mean RF, which are indicative of the uncertainties in the RF from the non-LLGHG agents, particularly aerosols (see Section 2.4 and Figure 2.12d). The direct effect of aerosols is seen in the total RF of the GFDL model over NH land regions, whereas the cloud albedo effect dominates the MIROC+SPRINTARS model in the stratocumulus low-latitude ocean regions. Note that the spatial pattern of the forcing is not indicative of the climate response pattern.
Wherever aerosol presence is considerable (namely the NH), the surface forcing is negative, relative to pre-industrial times (Figure 2.24). Because of the aerosol influence on the reduction of the shortwave radiation reaching the surface (see also Figure 2.12f), there is a net (sum of shortwave and longwave) negative surface forcing over a large part of the globe (see also Figure 2.23). In the absence of aerosols, LLGHGs increase the atmospheric longwave emission, with an accompanying increase in the longwave radiative flux reaching the surface. At high latitudes and in parts of the SH, there are fewer anthropogenic aerosols and thus the surface forcing has a positive value, owing to the LLGHGs.
These spatial patterns of RF and surface forcing imply different changes in the NH equator-to-pole gradients for the surface and tropopause. These, in turn, imply different changes in the amount of energy absorbed by the troposphere at low and high latitudes. The aerosol influences are also manifest in the difference between the NH and SH in both RF and surface forcing.
Figure 2.24. Instantaneous change in the spatial distribution of the net (solar plus longwave) radiative flux (W m–2) due to natural plus anthropogenic forcings between the years 1860 and 2000. Results here are intended to be illustrative examples of these quantities in two different climate models. (a) and (c) correspond to tropopause and surface results using the GFDL CM 2.1 model (adapted from Knutson et al., 2006). (b) and (d) correspond to tropopause and surface results using the MIROC+SPRINTARS model (adapted from Nozawa et al., 2005 and Takemura et al., 2005). Note that the MIROC+SPRINTARS model takes into account the aerosol cloud albedo effect while the CM 2.1 model does not.