General Formulation for Representing Cloud-to-Rain Transition in Atmospheric Models

Liu, Y., Brookhaven National Laboratory

General Circulation and Single Column Models/Parameterizations

Aerosol, Cloud Modeling, Cloud Properties

Liu, Y., P. H. Daum, R. McGraw, M. Miller, and S. Niu, 2007: Theoretical formulation for autoconversion rate of cloud droplet concentration. Geophys. Res. Lett., 34, L116821, doi:10.1029/2007GL030389

Figure 1. The typical drop radius r* as a function of the volume-mean radius r3 derived from the new theoretical formulation. Note that a constant r* corresponds to the commonly used assumption that the autoconversion rate for droplet concentration is linearly proportional to that for liquid water content, which is evidently incorrect.

The transition from cloud droplets to embryonic raindrops (called autoconversion) is an important process that needs to be accurately represented in atmospheric and climate models. The so-called one-moment scheme, which only predicts the cloud liquid water content, has dominated the field since the 1960s.

Despite its essential role in the development of atmospheric models, especially for quantitative forecasting of precipitation, such a simple scheme needs to be replaced for further progress, especially in evaluating indirect influences of aerosols on climate. Multi-moment schemes that carry more quantities (e.g., cloud droplets number concentration in addition to the liquid water content) are a pressing need. However, the few existing multi-moment parameterizations are empirical in nature, and are not adequate for these purposes. For example, it has been commonly assumed that the autoconversion rate for droplet concentration is proportional to that for liquid water content.

In a paper published in the August 24 issue of the Geophysical Research Letters, Liu, Daum, McGraw, Miller, and Niu present a general theoretical formulation of the autoconversion process. This new formulation is applicable to any desirable quantities, including liquid water content, droplet concentration and radar reflectivity as special cases and can be readily used to improve the representation of the autoconversion process. This work also reveals that the assumption commonly used to obtain the autoconversion rate for the droplet concentration is incorrect (see figure).