Seminar

Abstract

Global climate models (GCMs) have had a rough decade. In recent community consensus assessments of aerosol forcing and climate sensitivity, little weight was given to global models. This is understandable considering the justified doubts in these models' ability to represent the relevant cloud processes, which occur at scales far smaller than the model resolution. But discounting GCMs is a waste of a line of evidence that could be put to better use as a crosscheck on the assumptions underlying the other lines of evidence (process modeling and observations).

What would a world look like where global models contributed to physical understanding on an even footing with process modeling and observations? I will present an example from aerosol–cloud interactions. Cloud-top entrainment of warm, dry air into stratocumulus clouds increases as aerosol concentration increases. Climatically, this reduces cloud water enough under present-day (PD) compared to preindustrial (PI) aerosol conditions to cancel and probably overcome the effect of precipitation suppression by aerosol. The characteristic scales of cloud-top turbulence are O(1 m), so even large eddy simulations struggle to resolve entrainment; in global models, it will be impossible to resolve for the foreseeable future. Thus, a commonly held belief is that global models give an intrinsically biased answer for global aerosol forcing. To my great surprise, however, some global models are actually perfectly capable of reproducing the present-day satellite correlations that underpin our expectations of reduced cloud water. Nevertheless, their estimate of PD compared to PI is still one of cloud water increase due to anthropogenic aerosols. To conclude the presentation, I will try to get the different lines of evidence to engage in Hegelian dialectic to disentangle what this means for climate projections.

---

Johannes Muelmenstaedt's main research interest is the behavior of clouds in the multiscale climate system, one of the main uncertainties in our understanding of the climate system's response to human climate perturbations. In his current projects, Johannes aims to use observations of process variables, rather than state variables, to evaluate and eventually improve general-circulation and cloud-resolving global climate models. He is also interested in brute-forcing multiscale classical physics problems with quantum computers. Before moving to atmospheric science, Johannes received a PhD and MA in particle physics from the University of California, Berkeley, and a BS in physics from MIT.