Evaluating mean state relationships to high cloud feedbacks and climate sensitivity in CMIP model ensembles and E3SM

PI: Kathleen A. Schiro, University of Virgina; DOE Lab PI: Mark D. Zelinka,  Lawrence Livermore National Laboratory

Cloud feedbacks drive inter-model spread in climate sensitivity – the equilibrium warming of the climate in response to a doubling of atmospheric carbon dioxide (CO2) concentration. The cloud feedback is uniquely challenging not only due to the diversity of cloud types in the climate system, each with different radiative properties, and each governed by a unique set of meteorological and microphysical controls, but also because cloud formation, maintenance, and dissipation processes occur at scales that are typically not resolved by coarse climate models. We seek in this work to clearly quantify the causes of inter-model spread in high cloud feedbacks and delineate the role of inter-model differences in mean-state cloud properties from differences in the response of cloud properties to warming.

Our project objectives are guided by three main research questions:

1. How does the inter-model spread in the response of high clouds to warming relate to climate sensitivity?
2. What determines the spread in high cloud feedbacks, and what portion is related to inter-model differences in the mean-state?
3. How might we use observations to constrain high cloud feedbacks?

First, we propose to systematically evaluate the strength of the tropical high cloud optical depth, amount, and altitude feedbacks across global climate models taking part in phases 5 and 6 of the Coupled Model Intercomparison Project (CMIP5/6) and examine their relationships to the intermodel spread in climate sensitivity. We will examine models that implemented satellite cloud simulators, allowing for a detailed diagnosis of cloud feedbacks due to individual cloud types. We plan to pinpoint regions contributing most to model uncertainty in high cloud feedback estimates, which will help to identify sources of uncertainty.

Next, we will adopt a “product of scalars” framework to separately assess the contributions of the CMIP5/6 inter-model spread in mean-state cloud properties (amount, altitude, optical depth) and the spread in changing high cloud properties to the inter-model spread in the high cloud feedbacks. An existing perturbed physics ensemble (PPE) in the Department of Energy's Energy Exascale Earth System Model (E3SMv2), augmented with additional members in which deep convective parameters are perturbed, will then be used to systematically explore sources of spread in the high cloud feedbacks.

Finally, relationships between tropical high cloud feedbacks and modeled tropical high cloud climatology, if significant, will be used to observationally constrain the high cloud feedbacks. Changes in meteorological controls on high cloud properties in response to warming (e.g., tropopause height, radiatively driven divergence, tropospheric static stability) will also be examined for use as potential constraints.

Our Overarching Hypothesis is that high cloud feedbacks depend on mean-state cloud properties, permitting robust observational constraints and decreased uncertainty in model projections of climate sensitivity (Main Objective).