Based on a set of Perturbed Physics Experiments (PPEs) using the Community Earth System Model (CESM) in which deep convective parameters were altered (Schiro et al. 2019), we conjecture that there are three major pathways through which deep convection and large-scale circulation changes can modify low cloud fraction (LCF) in a warmer climate: (1) the temperature-stability pathway, through which tropospheric temperature anomalies propagated by wave dynamics subsequently modify lower-tropospheric stability, (2) the moisture-mixing pathway that may depend on shallow ascent and subgrid-scale mixing of moisture between the free troposphere and the marine boundary layer (MBL), and (3) the radiation-stability pathway that involves longwave radiation mediated subsidence control on LCF. We hypothesize that the differences in deep convective parameterizations between climate models drive a significant fraction of inter-model spread in low cloud feedback and ECS through these pathways.

In this project, we will 1) characterize the representation of the three pathways in CMIP6 model simulations and determine the relative contribution of each pathway to the CMIP6 model spread in low cloud feedback and ECS; 2) use process-oriented diagnostics to evaluate CMIP6 model performance in capturing the observed cloud-circulation relation and deep convection characteristics including convective transition statistics and the bulk properties of mesoscale convective systems (MCSs); 3) conduct E3SM short-range hindcasts following the DOE Cloud-Associated Parameterizations Testbed (CAPT) protocol to pinpoint specific model parameters/processes that are crucial to the representation of deep convection, circulation, clouds and the pathways that connect them. Multiple satellite and ground-based observations and reanalyses will be applied throughout to assess CMIP6 model fidelity and constrain E3SM model physics. We will leverage existing metrics and benchmarking packages available at PCMDI and contribute new metrics and diagnostic tools to augment the diagnostic capabilities of RGMA. This work builds on the strong expertise of the proposal team in model-observation diagnostics and evaluation of climate models using observations especially for deep convection, circulation, clouds, and precipitation.