The Dependence of Aerosol–Cloud Indirect Effects on the Representation of the Autoconversion Process in CESM: Formulation and Sensitivity Experiments

Aerosol indirect effects constitute one of the most important uncertainties for climate simulations. Anthropogenic aerosols can act as cloud condensation nuclei (CCNs). An increase in the aerosol loading often leads to a higher number of smaller cloud droplets. Smaller cloud droplets are more reflective and result in an higher cloud albedo (first aerosol indirect effect). Also, smaller cloud droplets are expected to result in less efficient precipitation and an increase of the liquid water path (second aerosol indirect effect). The first and second aerosol indirect effects cool the Earth. However, there are large uncertainties on the magnitude of the cooling, with the possibility of buffering of indirect effects, for example, by cloud-top entrainment, which in some meteorological regimes could limit or even reverse aerosol-induced increases in liquid water path.

Recent studies argue that general circulation models (GCMs) are overestimating the second indirect effect because they overestimate the increase in cloud liquid water path (LWP) due to an increase in CCNs. One of the culprits is thought to be the autoconversion parameterization (i.e the process in which cloud droplets grow to a size sufficient to form precipitation). The autoconversion rate is typically parameterized as a function of the cloud droplet number concentration (Nc) and the cloud water content (qc). An increase of Nc leads to a reduced autoconversion rate, delaying precipitation and increasing LWP. Especially, changing the sensitivity to Nc has a strong impact on the indirect effect. Therefore, a careful evaluation of the autoconversion schemes is important for understanding the role of cloud-aerosol indirect effect processes in the overall climate sensitivity of climate models.

In this study, we examine the sensitivity of the second indirect effect to the autoconversion scheme in a development version the Community Earth System Model. We assess the second indirect effect in atmosphere-only simulations with observed SSTs and in fully coupled historical simulations. We evaluate the effect of changing sensitivity to Nc in the autoconversion scheme. Our results demonstrate the primary importance of autoconversion processes in the representation of the second indirect effect magnitude, and the potential impact on the skill of transient CESM simulations.