Thursday, 5 August 2010: 5:15 PM
Torrey's Peak I&II (Keystone Resort)
The representation of stratocumulus is investigated using two versions of the Community Atmosphere Model (CAM). The versions are the publicly available CAM4 and CAM5. We investigate southeast Pacific stratocumulus using the Cloud-Associated Parameterization Testbed (CAPT) framework, wherein short-term forecasts are produced by initializing the models with realistic conditions from ECMWF analyses. The two models both produce stratocumulus decks, but they differ from each other significantly, and both have biases compared to observations. By the end of the second day of the forecasts, the models already evince characteristics of the long-term biases found in climate integrations. We investigate the balance of processes that maintain the stratocumulus and also the evolution of the simulations to describe how the models transition from the realistic initial conditions to their preferred states. Along with the initial conditions, the models also use the same finite difference dynamical core and the same horizontal resolution, so differences are associated with parameterized processes. The importance of the treatment of the atmospheric boundary layer and shallow convection are highlighted as key players in the differences between the models, with the implication that further improvements to these parameterizations are likely to improve the depiction of stratocumulus in climate simulations.
- Indicates paper has been withdrawn from meeting
- Indicates an Award Winner