The occurrence of ENSO events is quasi-cyclic, with at least two rhythms beating against each other. One involves a period of 2-3 years, though it is apparently unconnected with the stratospheric Quasi-Biennial Oscillation (Section 12.3). The other rhythm is more important and takes 3-5 years for each cycle. Despite this, the SOI was continuously negative for more than five years between January 1990 and September 1995. No two El Niños are the same.
There might perhaps be a link to sunspots, since 'strong' ENSO events since 1710 at least have been more frequent in years with fewer sunspots. Another clue comes from the similarity of the dates of El Niños and those of Nile floods in recent decades which suggests the possibility of using the long record of the floods to examine ENSO frequency since AD 622; it seems they were less common during the Medieval Warm Period, between AD 900-1200 (Chapter 15).
In view of the connection between SOI and rainfalls, it is not surprising that ENSO events also coincide with variations of precipitation. Examination of layers of ice on top of Quelccaya (Section 3.2) shows that 30 per cent less was deposited from the easterly winds during five recent ENSO events. Trujillo (at 8°S in South America) normally receives less than 25 mm of rain annually but 400 mm fell in 1925 on account of an El Niño. However, twelve of the driest twenty years between 1875 and 1978 in southeast Africa were El Niño years, and twenty out of twenty-six El Niño episodes in the last 120 years were associated with droughts in Australia. The relationships are not simple. Connections between El Niño and droughts (or abnormal rainfalls in the case of Santiago, for instance) are shown in Figure 10.16 to be only approximate, and the strength of the connection between rainfalls in eastern Australia and the SOI seems to change from decade to decade.
In summary, we now know enough about the progress of an ENSO episode to permit the following few months' events to be predicted with modest confidence (Note 12.N) once it has started, but we still cannot foretell its onset. One problem is that so many feedbacks complicate the process, and another is our ignorance of the interaction of the ocean with relatively weak winds.
It is not only the ENSO phenomenon that is becoming better understood. Recent decades have also seen enormous advances in explaining the synoptic-scale events which constitute weather. These are dealt with in the next chapter.
Was this article helpful?