Mean state and wave disturbances during phases I, II, and III of GATE based on ERA-40

Abstract

Using ECMWF's second-generation reanalysis, ERA-40, the large-scale mean state and synoptic-scale features associated with African easterly wave disturbances (AEWs) are examined over West Africa and the adjacent eastern Atlantic Ocean during the three 21-day observing periods of the Global Atmospheric Research Program (GARP) Atlantic Tropical Experiment (GATE) in 1974 (Phase I, 26 June-16 July; Phase II, 28 July-17 August; Phase III, 30 August-19 September). Results are partitioned into four geographical boxes, in order to highlight differences among the AEW vortices as they propagate westward along two tracks (northern and southern) over West Africa (land) and the adjacent eastern Atlantic Ocean (water). This marks the first time that a detailed diagnosis of the northerly track AEWs has been conducted. Results are also compared to previous GATE studies and a 30-yr climatology is extracted from ERA-40. In general, the subjectively analyzed wind fields presented in earlier studies compare favorably with the ERA-40 horizontal wind fields. The vertical motion field is one of the parameters that shows the largest differences to previously published results. In the area of the GATE A-B-scale ship array in the eastern Atlantic Ocean, low-level ascent during GATE is twice as large as in the ERA-40 climatology, most likely due to the dense upper-air network that allowed for an exceptionally good analysis of the divergent wind field. The midtropospheric outflow layer found over the ship array is absent in the ERA-40 climatology. Detrimental to the ERA-40 analyses of the upper-level easterly jet over the central Gulf of Guinea and along parts of the Guinea coast, were the assimilation of erroneous aircraft data. Using a recently developed tracking method of midtropospheric African easterly waves, a complete tracking history of northerly and southerly AEW vortices is presented and discussed for all three phases of GATE. One important result is that the activity of the northern waves at about 20°N was, in contrast to the southern waves at about 9°N, already quite strong during Phase I. At the same time, the low-level monsoonal flow, the heat low, and the upward motion in the northern desert zone were strongest. In contrast, the midtropospheric African easterly jet (AEJ) and the related horizontal shear instabilities were strongest during Phase III. The AEJ is also found at the lowest altitude over land during Phase III and it extends out to the Atlantic Ocean without changing its height and strength. These factors are associated with the well-known peak in the activity of AEWs in the southern wet zone during Phase III. In contrast to earlier findings, no reduction of AEW energy, by lifting of anomalously cool low-level air along the southern moist AEW track, could be observed over land. © 2004 American Meteorological Society.