Passage One Just under a year ago, a sharp drop in equatorial Pacific sea-surface temperatures indicated the end of the 1997—1998 El Niño.Called by someone“the climate event of the century”,it was by several measures the strongest on record. Identifying why it was so strong challenges our understanding of the physical mechanisms responsible for El Niño.This is more than simply an academic question: the 1997—1998 El Ni ño severely disrupted global weather patterns and Pacific marine ecosystems,and by one estimate caused $ 33 billion in damage and cost 23,000 lives worldwide.There were warnings of a coming El Niño before it occurred.But although many computer forecast models predicted that 1997 would be warm in the tropical Pacific up to three seasons in advance,none predicted the rapid development or ultimate intensity of the event before it began.Clearly we have much to learn from this experience. El Niño,Spanish for“the child”(and specifically the Christ child),is the name Peruvian fishermen gave to coastal sea-temperature warmings that first appeared around Christmas time.Now El Niño more generally refers to a warming of the tropical Pacific basin that occurs roughly every three to seven years in association with a weakening of the trade winds.The opposite side of El Niño, La Niña, is characterized by stronger-than-normal trade winds and unusually cold sea-surface temperatures in the tropical Pacific.Both El Niño and La Niña are accompanied by swings in atmospheric pressure between the eastern and western Pacific.These swings are known as the Southern Oscillation.These phenomena are collectively referred to as ENSO or El Niño / Southern Oscillation.At the moment,a strong La Niña is evident in the tropical Pacific,with several (but not all) forecast models predicting a return to normal by the end of 1999. The general mechanisms underlying ENSO involve large-scale oceanatmosphere interactions and equatorial ocean dynamics.But each El Niño and La Ni ña is unique in the combination of its strength,duration and pattern of development.Irregularity in the ENSO cycle can be seen both in the record dating back to the middle of last century,and in other supporting data,such as lake sediments,coral growth rings and tree rings,going back hundreds or even thousands of years.So, in principle,it should not be surprising that an unusually strong El Niño occurs very so often. Nonetheless,the 1997—1998 El Niño was an unusual one.It developed so rapidly that every month between June and December 1997 set a new monthly record high for sea-surface temperatures in the eastern equatorial Pacific.Anomalies (that is,deviations from normal) in December 1997 were the highest ever recorded along the Equator in the eastern Pacific.Moreover,before 1997—1998,the previous record-setting El Niño occurred in 1982—1983.These two“super El Niños”were separated by only 15 years,compared with a typical 30—40 year gap between such events earlier this century. Several factors may have contributed to the strength of the 1997—1998 El Niño.One is chaos,which some theories invoke to account for the irregularity of the ENSO cycle.Nonlinear resonances involving ENSO and the seasonal cycle have received special attention,but other chaotic interactions may affect ENSO as well.In 1997—1998 events possibly acted together to produce an extraordinarily strong El Niño simply due to the underlying tendency toward chaos in the climate system.A related issue is that of weather “noise” .Weather phenomena,inherently unpredictable more than about two weeks in advance,are a source of random forcing in the climate system.In the tropical Pacific,weather events occurring at the right time,and on time and space scales to which the ocean is sensitive,can markedly alter the evolution of the ENSO cycle. One notable source of weather in the tropics is the Madden-Julian Oscillation(MJO),a wave-like disturbance in the atmosphere with a period of 30—60 days that originates over the Indian Ocean.It could have been that the ocean got a healthy kick from the MJO at just the right time to send it on a course toward record high temperatures.The tropical Pacific was preconditioned for the beginning of an El Niño by the build-up of excess heat in the western equatorial Pacific due to stronger than normal trade winds in 1995—1996.However,beginning in late 1996,the MJO was particularly energetic,and several cycles of the wave amplified through nonlinear ocean-atmosphere interactions as they passed over the western Pacific.This set in motion a series of positive feedbacks between the ocean and the atmosphere which reinforced initial MJO-induced warming. Another possibility is that the ENSO cycle may be interacting with the Pacific Decadal Oscillation (PDO) —which,as the name implies,is a naturally occurring oscillation of the coupled ocean-atmosphere system in the Pacific with a period of several decades.In association with the PDO, sea-surface temperatures have generally been higher in the tropical Pacific from the mid-1970s.Since then,there have been more El Niños than La Niñas.The early 1990s was a period of extended warmth in the tropical Pacific,and two super El Niños occurred.The PDO may be one of the reasons for the observed decadal changes of the ENSO cycle,because it affects the background conditions on which ENSO events develop.From that perspective,the strength of the 1997—1998 El Niño may be but one manifestation of a linkage between interannual and decadal climate variations in the Pacific. What motivated the author to say at the end of the 2nd paragraph,“Clearly we have much to learn from this experience” ?