Uncertainty
Momentarily, I am engaged in several projects that relate to climate model uncertainty. Of great concern to me at the moment are the activities carried out by the MPI working group on uncertainty headed by J.-S. v. Storch, L. Tomassini and myself to better understand the effect of parameter perturbations on the model and climate sensitivity as simulated by the coupled ECHAM5/MPI-OM general circulation model.
In a first stage of our experiments we are systematically perturbing uncertain model parameters within ranges determined by experts in our modeling community. The perturbations are conducted in a statistically systematic fashion in order to achieve probability statements on the likelihood of certain model outcomes. This way a quantitative basis for climate model predictions can be
established.
A pilot study: Parametric uncertainty of the aerosol forcing
Aerosols affect climate via the direct and indirect radiative effects. The direct effect is the mechanism by which aerosols scatter and absorb shortwave and longwave radiation. The indirect effect is the mechanism by which aerosols modify the radiative forcing via microphysically induced effects on clouds. While the direct effect depends on the model radiation scheme, the indirect effect depends crucially on cloud parameterizations which are loaded with several uncertain parameters. Thus, the radiative forcing resulting from the aerosol indirect effect is intimately linked to parameter uncertainty.
Previous studies on aerosol radiative forcing have been focused on structural uncertainties, such as uncertainties in aerosol sources, in formulations used to represent aerosols in the model, in parameterizations that relate aerosols and cloud droplets to simulate the indirect aerosol effect, and in cloud schemes. In the present study, we assess another source of uncertainty, the parametric uncertainty resulting from model parameters, given a state-of-art atmospheric model with fixed cloud parameterization schemes.