Abstract
Some barotropic 500 mb forecasts are compared in the following manners:
1 comparison of the differences between forecasts made from analyses with identical boundary regions but different inner parts
2 comparison of forecasts made from two independent analyses based on identical material, and
3 comparison between forecasts based on two analyses with identical inner region but different boundary regions.
The result of a computation of the standard deviation of the vorticity and the Jacobian J(η, φ) is given and discussed, the deviation being due entirely to the personal factor in upper air analysis.