Atmospheric Effects of Nuclear War Aerosols in General Circulation Model Simulations: Influence of Smoke Optical Properties

A global atmospheric general circulation model (GCM) is modified to include radiative transfer parameterizations for the absorption and scattering of solar radiation and the absorption of thermal infrared (IR} radiation by smoke aerosols. The solar scattering modifications include a parameterization for diagnosing smoke optical properties as a function of the timeand spacedependent smoke particle radii. The aerosol IR modifications allow for both the "grey" absorber approximation and a broadband approximation that resolves the aerosol absorption in four spectral intervals. We examine the sensitivity of some G CM-simulated atmospheric and climatic effects to the optical properties and radiative transfer parameterizations used in studies of massive injections of smoke. Specifically, we test the model response to solar scattering versus nonscattering smoke, variations in prescribed smoke single scattering albedo and IR specific absorption, and interactive versus fixed smoke optical properties. Hypothetical nuclear war created smoke scenarios assume the July injection of 60 or 180 Tg of smoke over portions of the mid-latitude land areas of the northern hemisphere. Atmospheric transport and scavenging of the smoke are included. Nonscattering smoke cases produce roughly 40 Wm -2 more Earth-atmosphere solar irradiance absorption over the northern hemisphere, when compared to scattering smoke cases having equivalent specific absorption efficiencies. Varying the elemental carbon content of smoke over a plausible range produces a 4ø-6øC change in average mid-latitude land surface temperature, and a variation of about 0.1 in zonal!y averaged planetary albedo in the northern hemisphere. The inclusion of IR absorption by smoke (IR specific absorption to visible specific extinction ratio of 0.1) produces mid-latitude July temperature decreases that are 4ø-6øC smaller in magnitude than produced by IR-transparent cases. Thus the smoke IR opacity effect can make a substantial relative change in land surface temperature estimates when compared to July mid-latitude land temperature decreases of 15ø-20øC found in IR-transparent cases.