Thursday, 23 August 2007
Holladay (DoubleTree by Hilton Portland)
The NCEP's Climate Forecast System (CFS) is designed for monthly and seasonal forecast. The atmospheric model is the same as used in the NCEP's Global Forecast System (GFS) for short-term forecast. In CFS the atmospheric model is coupled with an ocean model and a land model. Numerous forecast experiments have been carried out. Previous analyses mostly focused on the lower atmosphere, and the model showed reasonable results. However, how well the CFS does in the stratosphere has not been examined. In this study CFS forecast results of the stratospheric basic states, including geopotential height, temperature, zonal flow and ozone, will be assessed. In particular, two sets of CFS experiments will be analyzed and compared with observations: one is an ensemble experiment with a low resolution (T126L64) model, and the other is a single experiment with a high resolution (T382L64) model. In the first experiment, the impact of different model physics, forcing and coupling on stratospheric forecast will be addressed. In the second experiment, all runs for each year from 1981 to 2006 have observed values of CO2, solar cycle and volcanic aerosols in their initial conditions. Monthly and seasonal climatology of stratospheric states are calculated from the 26 years of CFS run. They are compared with the NCEP Reanalysis-2 and Climate Prediction Center (CPC)'s analysis data. Attribution of forecast errors and bias will be discussed.
- Indicates paper has been withdrawn from meeting
- Indicates an Award Winner