An Evaluation of the Performance of the ECMWF Operational System in Analyzing and Forecasting Easterly Wave Disturbances over Africa and the Tropical Atlantic

Abstract

Substantial changes were made to the ECMWF model in May 1985. The extensive revisions to the physical parameterizations were designed to improve the treatment of the large-scale flow in the tropics. In addition, the resolution was increased substantially to a triangular truncation at T106. The purpose of this paper is to evaluate the performance of the new forecasting system on the analysis and forecasting of easterly waves and their associated tropical storms over Africa and the tropical Atlantic. A wave history generated for the months of August and September 1985 with use of operational analyses and METEOSAT imagery provides the framework for evaluating the performance of the analysis system. The difficulties caused by lack of data are discussed. Shortcomings of the analysis system are illustrated using an example of a short-scale disturbance with a marked convergence line. On the other hand, examples are also presented demonstrating the ability of the analysis system to make sense of widely scattered observations. The maxima in the vorticity field provide a set of useful markers to track the easterly waves, both in the analyses and in the forecasts. The 48-h forecasts of the positions and intensities of the vorticity maxima are verified those cases for which there is sufficient observational data to have confidence in the low-level wind analysis. The verification results are quite encouraging. A particular feature of the paper is the series of synoptic studies of the four waves which gave rise to named storms (Danny, Elena, and Gloria) during the period.