Predictability and Spectral Truncation

View More View Less
  • 1 Dept. of Atmospheric and Oceanic Science, The University of Michigan, Ann Arbor 49104
  • 2 Dept. of Meteorology, Massachusetts Institute of Technology, Cambridge 02108
© Get Permissions
Full access

Abstract

A quasi-geostrophic, two-level baroclinic model with a fixed heating function and simple friction represented in spectral form was integrated with climatological initial conditions for 30 days. Three separate integrations were performed changing only the spectral truncation. Two of the experiments had 10 degrees of freedom in latitude with 16 and 20 planetary waves, while the third had 11 degrees of latitudinal freedom and 18 planetary waves. Comparison of the integration results indicates that the increase in latitudinal resolution caused pronounced changes in the predicted variables whereas the increase in the number of planetary waves had a negligible effect on the distribution of variables over the integration period.

Abstract

A quasi-geostrophic, two-level baroclinic model with a fixed heating function and simple friction represented in spectral form was integrated with climatological initial conditions for 30 days. Three separate integrations were performed changing only the spectral truncation. Two of the experiments had 10 degrees of freedom in latitude with 16 and 20 planetary waves, while the third had 11 degrees of latitudinal freedom and 18 planetary waves. Comparison of the integration results indicates that the increase in latitudinal resolution caused pronounced changes in the predicted variables whereas the increase in the number of planetary waves had a negligible effect on the distribution of variables over the integration period.

Save