Cloud feedbacks in ACCESS and CMIP5 models

Cloud feedbacks in ACCESS and CMIP5 models

Abstract

Cloud feedbacks provide the greatest source of uncertainty in projections under given greenhouse gas scenarios, and are currently the subject of intense international scrutiny.  Quantifying, understanding and evaluating cloud feedbacks therefore remain top-priority tasks for quantifying projection uncertainty, assessing confidence in models and narrowing the range of future projections.  Evaluating cloud feedbacks in models, however, presents particular challenges.  This study describes the results from implementation of a cloud feedback analysis package in the Australian Community Climate and Earth-System Simulator (ACCESS).

This package permits detailed analysis of the location and type of cloud changes contributing to both ‘rapid response’ of clouds to the CO2 forcing itself, and to the temperature related feedbacks.  Results from a version of ACCESS 1.3 (atmosphere only) will be presented.  They reveal a complex range of cloud responses, with strongly varying contributions in both short and long wave.

The sensitivity of cloud responses to some key convection and rainfall-linked parameters will also be examined using results from a set of perturbed physics experiments.

Finally, since cloud feedbacks vary so much between models, ACCESS model feedback will be put in the context of the range of CMIP5 model cloud feedbacks.  How can these feedbacks be understood and evaluated?