Circulation around the South Pacific anticyclone causes cold water to upwell along the South American coast and along the equator east of the date line. Normally the atmospheric circulation in the equatorial region is weakest in April, when the Northern Hemisphere near-equatorial convergence is closest to the equator.

Six times in the past 80 years, large positive sea surface temperature anomalies developed along the Peruvian coast and over a large part of the tropical Pacific and persisted for a year or more—the phenomenon now generally known as El Niño. In the past, data from isolated islands were used in attempts to explain the accompanying meteorological events; understandably a variety of hypotheses resulted.

In the intense 1972–73 El Niño (March 1972–March 1973) extensive weather satellite data have removed much of the observational uncertainty of earlier studies. The Northern Hemisphere near-equatorial convergence shifted about 3° closer to the equator near the Line Islands while the equatorial “doldrum” belt shifted from west of the date line to east of the date line. In the equatorial strip, the El Niño year was relatively wet in the central Pacific, but relatively dry in the western Pacific, while in the eastern Pacific no coherent pattern emerged. Certainly no positive correlation could be established between anomalies of sea surface temperature and of rainfall.

Our preliminary findings suggest that contrary to previous opinion, ocean-atmosphere feedback may not be important in maintaining El Niño, which could be a Pacific manifestation of a global phenomenon.

This content is only available as a PDF.


1 Expanded version of presentations: 1) on 5 April 1974 in the NASA Scientific Colloquia Series; 2) on 11 April 1974 at the 55th Annual Meeting of the American Geophysical Union. Contribution No. 74-4 of the Department of Meteorology, University of Hawaii.