Climate response to atmospheric changes brought about by human activity may depend strongly on the geographical and temporal pattern of radiative forcing <Taylor and Penner, 1994>. In the case of aerosols stemming from anthropogenic sulfur emissions, geographical and temporal variations are certainly caused by variations in local mass concentration <Charlson et al., 1991; Kiehl and Briegleb, 1993>, but could also arise from variations in the optical properties of sulfate aerosols. Since optical properties depend fundamentally on aerosol size and chemical form and since size and chemical form are features of the aerosol which are not likely to be modeled on the global scale in the near future, geographical and temporal variations in optical properties could represent a stumbling block to accurate climate change forecasts. While extensive measurements of aerosol optical properties are needed to fully assess this problem, a preliminary assessment can be gained by considering the sensitivity of climate forcing to realistic variations in sulfate aerosol size and chemical form. Within a plausible set of assumptions (sulfate aerosol resides in the accumulation mode size range and only interacts with water vapor and ammonia vapor), we show that this sensitivity is fairly small (¿20%).

This low sensitivity derives from a number of compensating factors linking the three optical parameters identified by Charlson et al. <1991>. By implication, these optical parameters, low RH scattering efficiency, the ratio of hemispheric backscatter to total scatter, and the RH dependence of scattering efficiency, should not be treated independently in either theoretical or experimental investigations of direct climate forcing. A suggested logical focus for such investigations is the backscatter efficiency at high RH. If borne out by future research, low sensitivity to sulfate aerosol size and chemistry would mean that direct sulfate climate forcing can be incorporated in global climate models with only a knowledge of sulfate mass concentration. We emphasize, therefore, the need to study the extent to which our assumptions break down, in particular, the fraction of anthropogenic sulfate that forms on coarse mode particles (i.e., those with diameters >1 μm) and the extent and effects of sulfate interactions with other accumulation mode components. Finally, we find that a significant fraction of direct aerosol forcing occurs in cloud-covered regions, according to a simple bulk parametrization. ¿ American Geophysical Union 1995