Abstract

A series of experiments, designed to study single level predictions with a barotropic and a primitive equation (PE) model, essentially following Krishnamurti et al. (1980), has been carried out for June to September 1979 utilizing the FGGE (First GARP Global Experiment) data sets. The level chosen for this study is the 700 mb. covering a domain extending from 100° W to 38° E and 25° S to 45° N. Most of the waves of 1979. identified by Frank and Clark (1980), were investigated. Because these waves were not always well defined in the analysis at 700 mb, a wave anomaly pattern, prepared from Reed et al.'s (1977) composite African wave model, was carefully inserted in the respective locations of the waves. Major improvement in the tracking and passage of waves resulted from the merging of the African and easterly wave models into the broadscale flow patterns defined by the FGGE data sets. This study includes an analysis of the prediction errors. In this context the root mean square vector wind errors and the anomaly correlations are presented. The results of some 52 experiments show that the simple model based on the conservation of vorticity, performs better than persistence up to 2 days, while the next model, which is based on the conservation of potential vorticity, performs better than persistence up to 4 days. In general, very useful forecasts up to 2 days are possible with the single level PE model. DOI: 10.1111/j.1600-0870.1983.tb00215.x

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call