Modeling the moist‐convective atmosphere with a Quasi‐3‐D Multiscale Modeling Framework (Q3D MMF)

The Q3D MMF (Quasi-Three-Dimensional Multiscale Modeling Framework) is a new generation of MMF that replaces the conventional subgrid-scale parameterizations in general circulation models (GCMs) with explicit simulations of cloud and associated processes by cloud-resolving models (CRMs). In the Q3D MMF, 3-D CRMs are applied to the channel domains that extend over GCM grid cells. To avoid “double counting” of the large-scale effects, only the eddy effects simulated by the CRMs are implemented into the GCM as far as the transports are concerned, while the total effects are implemented for diabatic processes. The CRMs recognize the large-scale horizontal inhomogeneity through the lateral boundary conditions obtained from the GCM through interpolation. To maintain compatibility between the GCM and CRMs, the averages of CRM variables over the GCM grid spacing are relaxed to the corresponding GCM variables with the advective time scale. To evaluate the Q3D MMF, a transition from a wave to strong vortices is simulated in an idealized horizontal domain. Comparison with a fully 3-D benchmark simulation shows that the Q3D MMF successfully predicts the evolution of the vortices. It also captures important statistics such as the domain-averaged surface precipitation rate, turbulent fluxes and subgrid-scale (co)variances. From tests with 3-D and 2-D CRMs, respectively, it is concluded that the ability to recognize large-scale inhomogeneities is primarily responsible for the successful performance of the Q3D MMF. It is also demonstrated that the use of two perpendicular sets of CRMs has positive impacts on the simulation.