Scale-aware integral constraints on autoconversion and accretion in regional and global climate models

Abstract

We describe a novel method to constrain autoconversion and accretion rates in climate models. The method offers two desirable features that might ultimately help to effectively eliminate autoconversion efficiency as a free tuning parameter in climate models. Firstly, it can be constrained based on concurrent observations of column integrated cloud properties, namely cloud droplet number concentration, liquid water path, and precipitation rate. Secondly, the scale-dependency of autoconversion can be incorporated and resolved explicitly. The new method was implemented into the University of Hawaii's regional climate model iRAM. A series of test integrations were performed at horizontal resolutions ranging from 0.25° × 0.25° to 2° × 2°. The constrained approach was compared with a conventional approach commonly found in current climate models. Comparisons with an observational climatology of cloud liquid water path reveal significant improvements, in particular a better consistency between different model resolutions. Copyright 2011 by the American Geophysical Union.