Feedbacks and Convection in the Gap Between GCMs and CRMs

Abstract:

Radiative Convective Equilibrium (RCE) has proven to be a useful paradigm for studies of convection and the characteristics of idealized model configurations. Historically there has been little direct overlap between studies made with General Circulation Models (GCMs) and Cloud Resolving Models (CRMs). GCMs use global domains with resolutions ranging from 300 km to 20 km, while CRMs normally use domains roughly equivalent to a single GCM grid cell and resolutions ranging from 5km to 0.1km. Here we study the atmosphere of a GCM with boundary conditions that approximate RCE across a range of domain sizes (from 40 % of the surface of the Earth to 500x500 km^2) and resolutions of 20 km and 10 km. To that aim we use the newly developed nonhydrostatic, primitive equation model named ICON with two distinct physics parameterizations. Comparison of the simulations leads to a better understanding of the mechanisms and assumptions underlying the parameterizations of GCMs.

Climate sensitivity and convective organization are the two physical processes we focus on with this study. Convective organization strongly influences the mean state of the atmosphere and the various feedback responses to a given perturbation. We compare changes in the top of the atmosphere radiative imbalance which result from prescribed changes in the surface temperature. This provides a direct calculation of feedback processes. The magnitude of the influence these feedbacks have on our system is diagnosed with the climate sensitivity. Initial results indicate that ICON provides a convenient modeling framework that will allow for a unified approach to the study of the inevitable future overlap between CRMs and experiments with parameterized physics at relatively high resolution. Preliminary results indicate sensitive dependence of the atmospheric state on details of the model configuration.