Improved decomposition of cloud feedback and corresponding pattern change under uniform surface ocean warming: I. Anomalous mean method

Abstract Clouds are the primary source of uncertainty in future climate projections, due to complex dynamical and radiative processes. They exert a significant positive radiative forcing, thereby amplifying greenhouse warming. Although cloud feedbacks due to property changes have been well characterized and categorized into cloud types, their relations to the spatial patterns of cloud changes require further investigation. This study investigates cloud change and radiative feedback using 16 climate models forced by a spatially uniform 4K warming of the surface ocean. To decompose cloud changes into amount, altitude, and optical depth, we develop a new methodology that is more concise with smaller total residual than two existing methods, termed anomalous mean method (AMM). Our findings identify that the excess residual arises from the over-participation of the proportional cloud change, resulting in an overestimation of the longwave cloud altitude and optical depth feedbacks. The AMM effectively corrects these biases and reduces the residual feedback for total and various types of clouds. Furthermore, the method reveals remarkable spatial correlations between changes and climatologies for low cloud amount, optical depth, and high cloud altitude. These findings demonstrate robust cloud feedbacks in the designated cloud regime, accompanied by large intermodel spreads.