Why a GCM may overestimate the aerosol cloud lifetime effect: a comparison of CAM5 and CRM using ARM observations

 

Authors

Cheng Zhou — University of Michigan
Joyce E. Penner — University of Michigan

Category

Warm low clouds, including aerosol interactions

Description

The aerosol indirect effect, especially the cloud lifetime effect, has remained one of the biggest uncertainties in assessing the radiative forcing of past and future climate changes. Observation-based studies suggest that the response of cloud liquid water (LWP) to increased aerosol loading may have been overestimated in climate models. Here we use a global climate model (CAM5.3) and a cloud resolving model (GCE-CRM) to explore plausible causes. Shallow warm clouds on 05/27/2011 at the DOE ARM SGP site were simulated with both models. The LWP simulated by CAM increases substantially with aerosol loading while that in the CRM does not. The CAM LWP is more than doubled when the surface aerosol number is increased from 250 cm-3 to 4000 cm-3 while the LWP simulated by CRM decreases by ~5%. The large sensitivity of LWP to aerosol loading in CAM is shown to be caused by the large dependence of the autoconversion rate on droplet number. The increase of LWP can be reduced or eliminated when the dependence of the autoconversion rate on cloud droplet number is reduced. In the CRM, the autoconversion rate is reduced and the condensation rate is increased with increased aerosol loading, and these processes which act to increase the LWP are offset or even outweighed by the increased evaporation of cloud droplets. The net effect in the CRM is to increase the LWP in the growing stage of clouds but to decrease the LWP during the decaying stage. If the effect from changed condensation and evaporation rates (both are related to increased vertical movement) is excluded, the LWP simulated by the CRM can also increase by ~10%. Our results suggest that climate models need to include or parameterize the changes of evaporation/condensation of cloud droplets caused by aerosol perturbations and their effect on vertical velocity inside warm stratus clouds. They also suggest that when studying the aerosol lifetime effect one needs to distinguish between different stages of the cloud lifecycle.