Geophysical Fluid Dynamics Laboratory general circulation model investigation of the indirect radiative effects of anthropogenic sulfate aerosol

Abstract

The Geophysical Fluid Dynamics Laboratory (GFDL) atmosphere general circulation model, with its new cloud scheme, is employed to study the indirect radiative effect of anthropogenic sulfate aerosol during the industrial period. The preindustrial and present‐day monthly mean aerosol climatologies are generated from running the Model for Ozone And Related chemical Tracers (MOZART) chemistry‐transport model. The respective global annual mean sulfate burdens are 0.22 and 0.81 Tg S. Cloud droplet number concentrations are related to sulfate mass concentrations using an empirical relationship (Boucher and Lohmann, 1995). A distinction is made between “forcing” and flux change at the top of the atmosphere in this study. The simulations, performed with prescribed sea surface temperature, show that the first indirect “forcing” (“Twomey” effect) amounts to an annual mean of −1.5 W m−2, concentrated largely over the oceans in the Northern Hemisphere (NH). The annual mean flux change owing to the response of the model to the first indirect effect is −1.4 W m−2, similar to the annual mean forcing. However, the model's response causes a rearrangement of cloud distribution as well as changes in longwave flux (smaller than solar flux changes). There is thus a differing geographical nature of the radiation field than for the forcing even though the global means are similar. The second indirect effect, which is necessarily an estimate made in terms of the model's response, amounts to −0.9 W m−2, but the statistical significance of the simulated geographical distribution of this effect is relatively low owing to the model's natural variability. Both the first and second effects are approximately linearly additive, giving rise to a combined annual mean flux change of −2.3 W m−2, with the NH responsible for 77% of the total flux change. Statistically significant model responses are obtained for the zonal mean total indirect effect in the entire NH and in the Southern Hemisphere low latitudes and midlatitudes (north of 45°S). The area of significance extends more than for the first and second effects considered separately. A comparison with a number of previous studies based on the same sulfate‐droplet relationship shows that, after distinguishing between forcing and flux change, the global mean change in watts per square meter for the total effect computed in this study is comparable to existing studies in spite of the differences in cloud schemes.