Key points are not available for this paper at this time.
Trends in the large-scale circulation and seasonal characteristics of climate variability are investigated based on ship observations of sea level pressure, zonal and meridional wind components, sea surface temperature, and cloudiness and land station records, during 1948–83. Cluster analysis captures such features as the surface pressure dipoles associated with the Southern Oscillation, the subtropical region of influence of the North Atlantic Oscillation, preferred areas of surges in the North Atlantic trades and monsoon over the Indian Ocean, equatorial warm water regions with El Niño-type behavior in the eastern Pacific and Atlantic, and typical centers of interannual cloudiness variations associated with trade wind variations and the ITCZ. Long-term trends are examined in terms of internal consistency between elements, between ship and land records, and in regard to the possible effect of shifts in observational practices. Most prominent are the following trends over the 1948–83 period: (i) increasing prevalence of the negative Southern Oscillation phase (characterized by anomalously low/high pressure at Tahiti/Darwin) accompanying the increase of El Niño occurrences; (ii) in boreal summer a southward shift of the Atlantic near-equatorial low-pressure trough and embedded confluence zone and warming/cooling of surface waters to the south/north of the equator, paralleling the progressive aggravation of drought in sub-Saharan Africa; (iii) likewise in boreal summer a more easterly position of both the North and South Atlantic highs which together with (ii) connotes a reduced annual cycle of circulation in the tropical Atlantic sector, commensurate with the increasing prevalence of the negative SO mode; (iv) modulation of the annual cycle of circulation and rainfall regime over the western equatorial Atlantic, involving a somewhat earlier occurrence of the rainy season peak in Northeast Brazil; and (v) warming in the Indian Ocean.
Wolter et al. (Wed,) studied this question.