Using a Statistical Description of Cloud Field Geometry for Radiative Transfer Calculations
Abstract
Clouds play an important role in the Earth's climate system, and are accredited as one of the largest sources of uncertainty when attempting to model the Earth's radiative budget. Most cloud parameterizations take the form of a cloud fraction in each vertical model layer, and an overlap assumption of some kind. While cloud fraction is a useful tool for performing radiative transfer calculations, it is not physically realistic. A modified version of the Byrne et al. (1996) stochastic model is used to statistically represent partially cloudy atmospheres. Shortwave radiation fields and atmospheric heating rates are calculated using the stochastic model and compared to results from a traditional plane-parallel parameterization run in a Single-Column model (SCM). A coupled version of the SCM and stochastic model is used to incorporate the effects of feedback mechanisms. Cluster analysis is used to determine in what situations a statistical treatment of a cloudy atmosphere is preferable to the traditional plane-parallel approach. Preliminary results suggest the stochastic model performs well when mid-level broken cloud fields are present, but tends to underpredict the domain averaged downwelling shortwave radiation during times when the sky is overcast with deep cumulus or high cirrus clouds.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2006
- Bibcode:
- 2006AGUFM.A13B0920F
- Keywords:
-
- 0321 Cloud/radiation interaction