Summary

As the impacts of what may be prove to be the strongest El Niño event this century start to bite across the world, there is growing evidence that we may be beginning to witness the effects of human-induced climate change.

This report reviews the effect that El Niño on society and evidence of a distinct change in the behavior of El Niño dating back to the mid-1970s together with the possible explanations for this and it’s implications. A key finding is that while recent events could have occurred naturally, human-induced climate change may also be playing a part. Moreover, while scientific uncertainties remain over how El Niño’s response to climate change, it is clear that continuing to load the atmosphere with greenhouse gases we are gambling with the lives of millions and risking major ecological and economic disruption. Faced with the possibility that changes in El Niño would compound the already severe impacts of climate change, Greenpeace is advocating early and significant action to cut emissions of greenhouse gases.

El Niño and it’s Impacts

El Niño - a periodic warming of surface waters in the central and eastern Pacific Ocean - and associated changes in atmospheric conditions are a major cause of climatic disruptions around the world. The 1982-83 El Niño (the strongest event this century) claimed an estimated 1,500 lives and cost the world economy well over US$8 billion.

Such figures, however, tell little of the true extent of human misery that can accompany an El Niño event. Crop losses in poorer countries expose millions to the risk of famine and malnutrition, while changes in the weather are also associated with outbreaks of cholera, malaria and encephalitis in many areas. In addition, many suffer economic hardship due to weather effects on crops, fisheries and industrial production, while droughts, floods and fires can cause major social disruptions as people are compelled to leave their homes.

The ecological effects can also be devastating. In 1982-83, 85% of Peru’s sea bird population was lost as fish died or migrated and 70 to 95% of corals in Costa Rica, Panama and the Galapagos islands. On land, the most obvious impacts are through fires, but plant and animal life - like humans also suffer the effects of drought and floods.

The effects of El Niño are not all bad but the overall benefits are modest in comparison with the damage caused and the continuing ability of El Niño to play havoc with national economies and people’s lives is all to evident in the ongoing event.

El Niño 1997

Since March this year, sea surface temperatures in the central and eastern Pacific have risen rapidly and in August and September reached record levels for the time of year. By some measures, this already makes this latest El Niño the second strongest on record.

The effects of this years El Niño are also becoming severe. In Papua New Guinea, drought has already closed industries and put over 80,000 people at risk of starvation. Other parts of South East Asia have been doubly hit by failing rice crops and the effects of forest fires prolonged by the drought in Indonesia. Already, the fires have laid waste to an estimated 300,000 hectares of forest, while the toxic smog arising from the fires have affected perhaps 70 million people in six countries in South East Asia.

Grain production has been cut throughout the southern hemisphere and prices for some commodities are expected to spiral. Large areas of Australia and southern Africa have suffered from drought, while many countries in Latin America are also reporting massive damage due to drought or flooding. Costa Rica and Bolivia have declared a state of emergency, while Peruvian officials are anticipating a $300 million fall in exports due to El Niño-related impacts. In California, insurance claims for damages are mounting, while wine producers of the Napa Valley and in Chile are already anticipating lower production.

Unusual Times

The occurrence of the current El Niño event is all the more disquieting as it the latest example of a distinct change in the behaviour of El Niño dating back to the mid-1970s. This period is distinguished by:
 

This change in the behaviour of El Niño is, as the Intergovernmental Panel on Climate Change (IPCC) has remarked, "unusual in the context of the last 120 years". This observation is reinforced by the emergence of another strong El Niño this year.

Quite how unusual recent events are in a longer term context is less clear. Some analysts calculate that the likelihood of a persistent event like 1990-95 is just once is every 1,500 to 3,000 years while others estimate that such an event could be expected every 150 to 200 years. Records of past climates, on the other hand, suggest that while recent events may not be unprecedented, they are certainly rare.

Act of Man or Act of Nature?

The unusual behaviour of El Niño in recent times has prompted considerable speculation over what lies behind recent events and, in particular, over the possible role of climate change induced by rising concentrations of greenhouse gases due to human activities.

Some argue that the chances of recent events occurring are so remote that they are unlikely to be due to natural variability. Others question the statistical basis for this assertion and cite evidence for apparently comparable events in the past. To some extent such arguments are a distraction. Even if, as is suggested, recent events fall within the realms of what could occur naturally, this does not rule out the possibility of other influences.

Another possible explanation for recent events is a fundamental change in background conditions. The unusual behaviour of El Niño since the mid-1970s coincides with a sharp rise in global air temperatures which, in turn, is linked with a rapid rise in sea surface temperatures in the tropical Pacific. This background warming could have contributed to both the apparent strength of recent El Niño events and, possibly, to the observed increase in frequency.

The observed change in background conditions may itself be attributable, in part, to human-induced climate change. In 1996, the IPCC concluded that, "the balance of evidence suggests a discernible human influence on global climate". Similarly, a new study from the Geophysical Fluid Dynamics Laboratory in Princeton finds that the observed trends in Pacific sea surface temperatures could be attributable both to natural variability and to greenhouse gas forcing.

Given the sensitivity of El Niño to changes in background conditions, it is reasonable to suppose that El Niño may be being influenced by human-induced climate change. The question is how? Some indication of whether or not recent events are symptomatic of a human-influence can be gleaned from, in particular, computer modelling studies of how El Niño may respond to climate change.

While scientific confidence in the results of individual experiments is low, the results of a number of studies suggest:
 

In direct contrast to these findings, a few studies suggest that La Niña conditions could be prevalent in a warmer world. Even in this case, however, one study suggests that those El Niño events which do occur could be more intense.

Given the contradictory nature of these findings, it is impossible to gauge the "human component" to recent events with any certainty. Nevertheless, aspects are consistent with the observed background warming which may itself be partly attributable to human-induced climate change.

Implications

To the extent that El Niño is a natural phenomenon, it is clearly something we must learn to live with. But it is also clear that human-induced climate change brings with it the possibility of more intense and more frequent El Niño events.

Since the 1982-83 event, responses have been guided by the principle "forewarned with forearmed", with huge investments in the development of early warning systems to assist people in developing appropriate response strategies. But, for all this, the potential effects of human-induced climate change on El Niño cannot be ignored.

As climate changes, the reliability of forecasts could go down, tempering their usefulness in designing responses, while more intense and/or more frequent El Niño events would increase the likelihood of catastrophic damage and the cost of response measures. Still more fundamentally, high levels of consumption, population growth, land degradation and pollution mean that many areas are becoming more - rather than less - vulnerable to climate extremes.

In this context, it is clear that by continuing to pump greenhouse gases into the atmosphere, we are gambling with the lives of millions of people around the world and risking major ecological and economic disruption.

A Precautionary Approach

Faced with the possibility of catastrophic impacts, Greenpeace is advocating a precautionary approach to the risk of climate change. Central to this is the need to cut emissions of greenhouse gases to the levels required to stabilise atmospheric composition. An important next step towards this goal would be international agreement at Kyoto in December for developed countries to cut their carbon dioxide emissions by 20 per cent on 1990 levels by 2005.

Introduction

As the impacts of what may prove to the strongest El Niño event this century begin to bite, attention has been drawn to the possibility that the behaviour of El Niño may be changing under the influence of human-induced climate change.

El Niño - a periodic warming of the central and eastern Pacific Ocean - is renowned for it’s association with climatic disruptions both locally and around the world. The strongest El Niño event this century, which peaked in the northern hemisphere winter (December to February) of 1982-83, caused widespread human suffering and ecological damage and is estimated to have cost the world economy well over US$8 billion.

The current event is the latest example of a distinct change in the behaviour of El Niño dating back to the mid-1970s. While these events could have occurred naturally, there is growing evidence that we may be starting to witness the effects of climate change induced by rising concentrations of greenhouse gases. Given the widespread damage associated with El Niño events, this is a risk that can not be ignored.

El Niño - a Force for Change

The term El Niño - Spanish for "the boy child" or the "Christ Child" - was originally coined by fishermen to describe a warm current that typically appeared off the coast of Peru in December. Now, scientists use the term to refer to the recurrence of anomalously warm conditions across the central and eastern Pacific.

El Niño is closely linked with atmospheric conditions associated with one extreme of the Southern Oscillation - a see-saw in atmospheric pressure between the eastern equatorial Pacific and Indo-Australian areas. The state of the Southern Oscillation both directly affects tropical weather around the globe and profoundly influences climate in many other regions.

The intimate connection between El Niño and atmospheric conditions means that scientists often connect the two phenomenon and refer to the El Niño-Southern Oscillation (ENSO). In these terms, an El Niño event thus describes both the anomalously warm oceanic conditions and associated changes in atmospheric circulation. The occurrence of anomalously cold conditions is known as an La Niña event.

What goes on during an El Niño event is fairly well understood. Under "normal" conditions, easterly trade winds draw warm surface water away from the coast of Peru and Chile westwards. This allows cold water from the ocean depths to well up in the eastern Pacific while causing warm surface water to pile up in the west around Australia and the Philippines. The resulting warm pool is several degrees hotter and about a half metre higher than the eastern ocean. El Niño is prompted by a slackening of the trade winds which allows warm water to slosh back across the ocean, lowering the temperature of the warm pool and increasing the temperature of the central and eastern Pacific.

Once started, the process is kept going by an intimate connection between oceanic and atmospheric conditions. The strength of the easterly winds is related to differences in atmospheric pressure across the Pacific ocean which, in turn, are related to differences in temperature in the underlying water. As the central and eastern Pacific warm, the air above it also heats up and becomes more similar in temperature to the air in the west. This causes the trade winds to weaken still further, allowing the continued influx of warm water.

El Niño conditions cease once the slightly cooler warm pool starts to heat up and the trade winds start to strengthen. This sets the stage for transition to La Niña - or cold - conditions in the central and eastern Pacific, during which the process in reversed.

The climatic repercussions of this intricate interplay of the ocean and atmosphere are felt around the globe. As warm water moves eastwards, so does the rain, causing droughts in Indonesia, the Philippines and north-eastern Australia. Conversely, the normally dry western edge of South America experiences heavy rain.

Unusual weather patterns most consistently associated with El Niño conditions during the northern hemisphere winter (December to February).
Source: Ropelewski, 1992 1 .

The movement of the tropical storms also disrupts the flow of the jet streams - strong, fast winds high in the atmosphere - which drive the world’s weather systems. El Niño events have thus been linked with dry conditions in places as far afield as north-east Brazil, India, southern Africa and the Iberian Peninsula, with unseasonal warmth in western Canada and with a good chance of floods from Texas to Florida in North America.

During El Niño events, the areas afflicted by tropical cyclones also changes. The presence of warmer water further east causes tropical cyclones to form in the central Pacific from where they are more likely to strike central Pacific Islands such as French Polynesia, rather than the Australian Queensland coast.

El Niño events tend to recur every three to seven years. What triggers individual events remains uncertain, but El Niños typically develop in the northern spring when the easterly trade winds are at their weakest and they can last for 12 to 18 months. No two El Niño events are exactly alike, varying in manner of development, strength, duration and their impacts on society.

Impacts of El Niño

The effects of El Niño can be devastating. During the 1877/78 event, between 9 and 13 million people died from hunger, disease and violence as crops failed in four provinces in northern China alone 2 . A decade later, famine and epidemics resulted in 1.5 million deaths in India and the death of one-third of the population of Ethiopia 3 .

Economic and political developments since then mean the impacts tend to be measured in terms of economic losses rather than lives lost. The 1982-83 El Niño claimed an estimated 1,500 lives, but the cost to the world economy was estimated at the time to be US$8 billion through lost output and damage to infrastructure 4 . More recent estimates put the cost of that El Niño at two to three times this amount 5 . One study values US crop losses alone, as US$10-12 billion 6 .

Such figures, however, tell little of true extent of human misery that can accompany an El Niño event. In poorer countries, crop losses through drought all too readily translate into severe food shortages and malnutrition. During the widespread droughts accompanying the 1991-92 El Niño event more than 30 million people were thus affected in southern Africa alone 7 .

Effects of the 1982-1983 ENSO
 

More pernicious is the effect that changes in the weather can have on the incidence of disease. El Niño events have been variously linked with outbreaks of cholera in Peru, malaria in Columbia, Costa Rica, Pakistan, India and Sri Lanka and encephalitis in Australia 8 .

Global or national economic figures also tend to mask the economic hardship experienced on the ground. As a result of the 1972-73 El Niño, more than 100,000 people in Chile and Peru became unemployed as fish yields plummeted 9 , while drought in the rainforests of Kalimantan in 1991-92 cost one township an estimated one-quarter to one-half of its annual income as a result of losses of crops and of wages from forestry 10 .

El Niño events can also result in major social disruption as drought, flood or fires compel families to leave their homes. Migrations in search of food are a common sight in drought-stricken Africa. In 1982-83, floods necessitated the evacuation of 600,000 people in Brazil, Argentina and Paraguay, while in Australia, drought and fires rendered 8,000 people homeless 11 .

Often overlooked is the ecological damage wrought by El Niño events 12 . The most immediate impacts are off the coast of Peru. Here, warmer waters and a collapse in the supply of nutrients result in the death or migration of large numbers of fish, with serious repercussions further up the food chain. In 1982-83, an estimated 85% of Peru’s sea bird population was lost 13 . Further afield, commercial fisheries in southern California declined dramatically with cuts in, amongst other things, anchovy of 90%, squid 89% and shrimp 74% 14 .

Other sea life is also at risk. In 1982-83, 70-95% of the corals in Costa Rica, Panama, Columbia and the Galapagos Islands died as a result of warm waters and bleaching, and nearly all sea pups perished in the Galapagos islands 15 . On land, some of the worst impacts are associated with the spread of forest and bush fires due to drought.

The effects of El Niño are not all bad. In Peru, heavy rains replenish the aquifers - 8 underground reservoirs - upon which farmers rely and the desert goes into bloom 16 . In 1982- 83, warmer conditions reduced heating bills in the eastern USA, saving an estimated US$500 million 17 . Moreover, where one region loses, another may gain. The southward movement of currents off South America means that Ecuadorian shrimp turns up in Peru, while Peruvian anchovies turn up in Chile 18 . While the Caribbean may be blighted by tropical storms, the risk of hurricanes reaching the USA is halved 19 .

Such beneficial effects are, however, modest in comparison with the damage caused by El Niño. While continued economic development and improvements in forecasting are helping people mitigate some of the bad impacts and take advantage of some of the good, the ability of El Niño to play havoc with national economies and peoples’ lives is all too evident in the ongoing event.

El Niño 1997 - a Record in the Making

In March this year, rapidly rising sea surface temperatures alerted scientists to the possible onset of El Niño conditions. Since then, temperatures in the central and eastern Pacific have shot up, reaching record levels for the time of year in August and September (compared with the last 50 years) 20 . By October, temperatures were 2 ^o  C above normal in the equatorial Pacific east of 170 ^o  W and over 5 ^o C above normal east of 120 ^o  W 21 . By many measures, this already makes this El Niño the second strongest on record.

So powerful is this El Niño that Nick Green of the Scripps Institution of Oceanography at the University of California in San Diego, likened watching its development to "watching Mount St Helens erupt in slow motion" 22 .

Comparison of the six strongest El Niño events since 1950 with the current El Niño.

Source: data provided by Klaus Walter of the NOAA/CIRES Climate Diagnostics Center, University of Colorado in Boulder. Relative strengths are expressed in terms of a Multivariate ENSO Index (MEI)23 .

The sheer strength of this El Niño was unexpected, with most forecasters predicting either cool or only slightly warm conditions through to the middle of the year at least 24 . The prompt for this surprising development appears to have been Cyclone Justin which hovered off the Queensland coast for nearly a month in March 25 . Powerful winds generated by the rotation of the cyclone blew warm water normally trapped in the western Pacific eastwards.

How El Niño will develop over the next few months and beyond remains a matter for speculation 26 . One possibility is a slow transition to create a prolonged, lingering warm event. The other, and possibly more likely, scenario is that there will be a rapid strengthening towards El Niño conditions followed by a rapid recovery. In early October the USA’s National Center for Environmental Prediction in Washington forecast that strong El Niño conditions through to March 1998, after which temperatures are expected to decrease 27 . El Niño modellers at the Scripps Institution of Oceanography at the University of California in San Diego suggest a La Niña event could form in the middle of 1998 <>28 .

Feeling the Heat

News reports of actual or anticipated effects of this years El Niño are already flooding in.

Indonesia is facing its worse drought in 50 years and rice crops are failing throughout the Asia Pacific region 29 . By September, drought and frost in Papua New Guinea had claimed at least 270 lives, causing many to flee from the highlands to the towns 30 - by October, more than 80,000 people risked starvation 31 . The crisis is also hitting the economy. The Ok Tedi copper mine ground to a halt in August with the drying up of the Fly river which supplies it with power and transport, while ore processing ceased at the Porgera gold mine in September due to a lack of water to strip ore from rock 32 . The Australian Government responded to pleas for help with a contribution of US$3.2 million to Papua New Guinea’s drought relief efforts 33 , as well as technical assistance.

Meanwhile, drought has worsened the seasonal preconditions for the devastating fires that by early October had laid waste an estimated 300,000 hectares of forest in Sumatra, Kalimantan and Sulawesi 34 . The toxic smog arising from the fires spread over 3,200 kilometres (2,000 miles) in South East Asia, affecting six countries and perhaps 70 million people 35 . Again international assistance has been forthcoming with the European Union and Australia pledging US$474,000 and US$360,000, respectively to help fight the fires 36 .

Economic damage from lost production in factories, lost foreign investors and reductions in tourists could also be massive. In September, one firm in Jakarta (Indonesia) reported a 40 to 50% reduction in tourists compared with last year 37 . Neighbouring regions largely unaffected by the fires are similarly blighted - some hotels in Sabah (Malaysia) reported losing US$735,000 to US$800,000 during October 38 .

The full health, ecological and economic damage arising from the fires is incalculable. Already millions are suffering respiratory and eye problems while cancer-causing chemicals -polycyclic aromatic hydrocarbons - in the smog could result in early death for many. Untold numbers of rare plants and animals have gone up in flames or will have perished due to high temperatures.

Grain production has been hit throughout the southern hemisphere and prices for some commodities, such as coffee, are expected to spiral 39 . In New Zealand, crop-harvest losses have exceeded US$130 million 40 . Large chunks of Australia have also been struck by drought, although above average rainfall across eastern Australia in September has alleviated some of the drought’s impact.

The prognosis for South Africa is worse. SBC Warburg, the investment bank, is predicting a fall in maize production of 56% and reductions in production of other crops 41 . Water shortages will also threaten output from the nation’s zinc, copper, nickel and palladium mines. El Niño is also held partly to blame for a poor "long rains" season and cuts in agricultural output in Kenya 42 .

Costa Rica has declared a state of emergency and, in September, reported multimillion dollar losses to rice, bean, coffee, fish and ranching industries due to hot temperatures and a lack of rain 43 . Similarly, Bolivia has declared a state of emergency and the Colombians expect a sharp fall in farm output. Floods in Chile have been the worst for a decade 44 and the wine harvest will prove a wash out 45, while the famed Copacabana beach in Rio de Janeiro (Brazil) has all but disappeared due to some of the worst storms in decades 46 . In Peru, fishing and related industries are suffering as anchovies desert Peru’s coastal waters in search of colder water further afield 47 . Drought is already affecting cotton production and potato and other food crops is expected to go down. On a different note, the textile industry talks of losses of US$500 million due to reductions in demand for winter clothing. Official estimates suggest that this years El Niño could cost Peru 1-2% of GDP, with exports falling US$300 million.

In California, floods have prompted vast insurance claims 48 and the wine producers of the Napa Valley are already anticipating lower output 49 . Fish migrations have caused an upsurge in deaths of sea birds and marine mammals, such as Pelicans and sea lions 50 . The silver lining? In May, fishermen off Laguna beach reported their biggest haul of squid since 1990 (an opposite effect to that of the 1982-83 El Niño) 51 , while surfers had a whale of a time riding the waves propagated by Hurricane Linda 52 . In southern California, the change of winds and rain have caused smog levels to fall dramatically 53 . Also, very few tropical storms in the Atlantic have reached hurricane proportions 54 .

The eventual impacts of this latest El Niño, and the extent to which the gloomy forecasts are fulfilled, remain to be seen. But past experience certainly suggests the impacts will be severe.

Unusual Times

The current El Niño event is the latest in a sequence of unusual events and is symptomatic of a distinct change in behaviour dating back to the mid-1970s 55 . Since then, El Niño has been intensely active (relative to the previous 50 years). Although periods of high activity have occurred before, for example from 1880 to 1925 56 , the recent period is distinctive in a number of respects.

During recent times, El Niño events have predominated and there are very few La Niña events. This contrasts with a general tendency for El Niño and La Niña conditions to alternate every two to three years over the century as a whole. The period is further distinguished by the occurrence of the strongest event on record 1982-83 and of what, by some measures, is the longest El Niño event on record - from 1990 to 1995.

This behaviour, and particularly that of the early 1990s, is, as the Intergovernmental on Panel on Climate Change (IPCC) 57has remarked, "unusual in the context of the last 120 years" - that is, during the period of instrumental records 58 . This observation is further reinforced by the emergence of another strong El Niño event this year.

Variations in the Southern Oscillation Index through to 1995. Negative values represent "warm episodes" which generally correspond to El Niño conditions, while positive values represent "cold episodes" which generally correspond to La Niña conditions.
Source: Trenberth and Hoar, 1996 59 .

What is less clear is just how unusual such events are in a longer term context. El Niño is, in the words of one scientist, "a capricious beast", and its erratic behaviour is not fully captured in the instrumental record. For this reason, scientists resort to complex statistical analyses of 11 the existing records or proxy data - historical sources, tree rings, geological records and the like - in an attempt to cast a light on recent events.

Kevin Trenberth and Timothy Hoar of the National Center for Atmospheric Research in Boulder were in the forefront of attempts to assess quite how unusual the 1990-95 event and those of the past two decades really are 60 . In their analysis, Trenberth and Hoar use atmospheric pressure for Darwin to represent the highs and lows of the Southern Oscillation. By this measure, El Niño events are represented by above-average pressure at Darwin in Australia.

Using sophisticated statistical techniques, Trenberth and Hoar show that the chances of a persistent warm episode like 1990 to 1995 to be just once in every 1,500 to 3,000 years under stable climate conditions They further found that the recent tendency towards more frequent El Niños observed since 1976 is likely to occur only once in every 2,000 years.

These findings for the early 1990s have since been challenged by Donald Harrison of the Pacific Marine Environmental Laboratory in Seattle and Narasimhan Larkin of the Joint Institute for the Study of the Atmosphere and Ocean at the University of Washington in Seattle 61 . Reanalysing the Darwin data, they calculate that an event like that of 1990-95 could be expected as often as once in every 150 to 200 years.

Others have also questioned Trenberth and Hoar’s findings on the grounds that similarly persistent El Niños have occurred before the period of instrumental record 62 . For their part, Trenberth and Hoar question their critics’ findings on the grounds that they deal only with the persistence of the El Niño event and not with it’s magnitude 63 . The debate continues.

A further confusion arises because it appears that perceptions of what happened in the early 1990s depend on the measure used to define an El Niño event. Sea surface temperatures for the eastern Pacific - the so-called "Nino 3" region - are widely used as an indicator of El Niño conditions. Analysing these data, Lisa Goddard and Nicholas Graham of Scripps Institution for Oceanography at the University of California in San Diego found not one long El Niño, but three weak El Niños (and no La Niñas) 64 .

Nevertheless, Goddard and Graham, like Trenberth and Hoar, do find that events in the early 1990s were unusual - albeit for a different reason. Typically, the early development of El Niño events is characterised by the existence of sub-surface temperature anomalies which eventually reach the surface. In contrast, the El Niño events of the 1990s appear to have been initiated by persistent warmth of surface waters in the central Pacific. It is notable that the warmth of the tropical Pacific is itself highly unusual 65 .

In the wake of the 1982-83 El Niño, academics tried to assess just how uncommon this event was. Historical analysis by William Quinn of Oregon State University in Corvallis suggests that while very strong events like that of 1982-83 may have occurred in the past, they occur very rarely, with only eight occurring since the early 1500s 66 . These findings are also borne out by an analysis of tree ring data for western North America 67 .

Such analyses suggest that while recent El Niño events may not be unprecedented, they are certainly unusual.

Act of Nature or Act of Man?

The unusual behaviour of El Niño in recent times has prompted considerable speculation over what lies behind recent events. Based on current understanding of the phenomenon, there are three possible explanations: i) natural variations in the behaviour of El Niño; ii) changes in the background conditions; and iii) human-induced climate change. A fourth explanation is, of course, a combination of all three.

So far there is no way of telling which of these competing interpretations is correct. Nevertheless, there is increasing evidence that while recent variations in El Niño could be accounted for by natural variability, changes in background conditions and human-induced climate change could each be playing a part.

..... natural causes?

Based on their assessment of the likelihood of recent events, Trenberth and Hoar, conclude that it is highly unlikely that recent events are a product of natural variability on the timescale of decades. This position is supported by an analysis of sea surface temperatures for a triangular region covering the eastern and central Pacific by Tom Knutson and Syukuro Manabe of the Geophysical Fluid Dynamics Laboratory in Princeton 68 . This shows that the recent warming is highly unlikely to have occurred through natural variability alone, although this is likely to have played a substantial role.

Not surprisingly, given the lower probability they ascribe to the likelihood of the 1990-95 event, Harrison and Larkin are more cautious, suggesting that "the 1990-95 period may plausibly be an aspect of the natural variability of the Pacific". Mojib Latif and colleagues from the Max Planck Institute in Hamburg are still more forthright, arguing that from a physical point of view the strange temperature and wind patterns of 1990-95 are related to natural variability from decade-to-decade and not to any unusual greenhouse warming 69 .

This contention that the recent behaviour of El Niño could be solely a product of natural variability is supported a reconstruction of long term changes in the Southern Oscillation by Rob Allan of Australia’s Commonwealth Science and Industrial Research Organisation and Roseanne D’Arrigo of Lamont-Doherty Geological Observatory in New York state 70 and by computer model simulations of phenomenon 71 . Similarly, a number of observational studies that suggest that recent patterns in sea surface temperature can be explained by El Niño-like variability on timescales of decades to centuries 72 .

To some extent, such arguments over statistical niceties and past events are a distraction. On the one hand, however improbable an event is, there is always a chance that it will happen. Conversely, just because something could happen of its own accord, this does mean that it will. In other words, even if, as is suggested, recent events fall within the realms of what could occur naturally, this does not rule out the possibility of other influences.

..... behind the scenes

A second possible explanation for recent events is the warmer background conditions that have prevailed over this time. The unusual behaviour of El Niño since the mid-1970s coincides with a sharp rise in global air temperatures. The 1980s was the warmest decade on record (back to 1860), while 1995 was the warmest year on record. In July this year, David Parker of the UK Meteorological Office further predicted that 1997 will also be exceptionally warm 73 .

As air temperatures have risen, so have those of the tropical Pacific. In a new study of Pacific sea surface temperatures, Knutson and Manabe find a warming trend over a large part of the equatorial Pacific over the past century 74 . This trend has accelerated in recent decades. Since the 1970s temperatures have risen at a rate of 2.9 ^o  C per 100 years - this is more than twice the rate of change found over the previous 20 years.

Part of this warming is due to high levels of El Niño activity - El Niño events coincide with warming of the lower atmosphere in the tropics of close to 1 ^o  C 75 . But the existence of a long-term warming trend in both global and sea surface temperatures points to an underlying change in background state of the Pacific Ocean.

As discussed earlier, unusual warmth in the central Pacific is thought to have contributed to the strange behaviour of El Niño in the early 1990s. But background warming of the tropical Pacific could also account for various other aspects of El Niño behaviour in recent years. One effect of warmer conditions is that fluctuations in ocean temperatures happen about a 13 higher baseline. This means that even with the same size fluctuations, El Niño events appear more impressive and La Niña events look more feeble 76 . Although this is in some senses an illusion as the basic behaviour of El Niño need not have changed, the impacts of El Niño events could be correspondly more severe.

Background warming could also be increasing the likelihood of El Niño events. The timing of El Niño events is thought to be partly controlled by the time needed to essentially recharge the warm pool in the western Pacific after the previous El Niño event 77 . In theory at least, background warming could effectively reduce the time required to recharge the system -thus, increasing the frequency of events. A further factor may be observational evidence that the eastern Pacific is warming more rapidly than the western Pacific 78 . Under these conditions, the trade winds weaken, setting the stage for El Niño conditions to develop.

Having said this, it remains impossible to disentangle the relative contributions of natural variability and background warming to recent events. Suffice to say, both could be playing a part.

.... a human influence?

The third element of the equation is the possibility that human-induced climate change is behind recent events. Atmospheric concentrations of greenhouse gases are rising rapidly as a result of human activities and in particular, fossil fuel burning, land use change and agriculture. According to the IPCC, if current trends in emissions continue then this could cause a rate of global warming over the next century "probably greater than any seen in the last 10,000 years" 79 .

Past activities may already be contributing to the observed changes in background conditions. In 1996, the IPCC concluded not only that, "the observed trend in global mean temperature over the past 100 years is unlikely to be entirely natural in origin" 80, but also that, "the balance of evidence suggests a discernible human influence on global climate" 81 . Similarly, a new study by Knutson and Manabe finds that both natural variability and greenhouse gas forcing 82 have probably contributed to observed trends in Pacific sea surface temperature trends, although their relative contributions remains uncertain 83 .

Given the apparent sensitivity of El Niño to changes in background conditions, it is thus reasonable to suppose that El Niño may also be being influenced by human-induced climate change. The question is, how? Identifying clear signs of change amidst the noise of natural variability inherent in the El Niño system is currently impossible. However, some indication of whether or not recent events could be symptomatic of a human influence can be gleaned from assessments of the possible response of El Niño to human-induced climate change.

A variety of approaches have been taken to predict the response of El Niño to climate change, ranging from the theoretical through observational studies to computer modelling. Of these, the most promising is computer modelling as only this can fully take into the complex interactions between the atmosphere and test responses to increasing atmospheric concentrations of greenhouse gases.

Results from the various studies are often conflicting and, in the case of modelling experiments, confidence in results from individual experiments is low due to simplifications of the processes involved and/or the crude spatial resolution of the models. Nevertheless there are strong indications that human-induced climate change could lead to marked changes in El Niño and its effects - aspects of which are consistent with recent events.

At the most fundamental level, it seems likely that El Niño events will continue to occur in a warmer world - as will La Niña events. All modelling experiments with greenhouse gas forcing show El Niño like fluctuations and it seems unlikely that the system will get locked into one phase or the other 84 . This view is supported by studies of past climates which show that El Niño has waxed and waned periodically for at least 5,000 years under a range of climatic conditions.

Background conditions are, however, set to change markedly, with most model simulations indicating a substantial warming of the tropical Pacific in response to increasing greenhouse gases 85 . Several studies also suggest that the eastern Pacific will warm more that the west in response to increased greenhouse gas forcing 86. As the authors of one such study, Gerald Meehl and Warren Washington of National Center for Atmospheric Research in Boulder, point out, this mean pattern of change resembles that associated with present day El Niño events as well as changes in the 1980s 87 .

With greater warming in the east than in the west, it is suggested that the amplitude - the size of peak-to-trough fluctuations in sea surface temperature - of El Niño could decrease slightly over time. This is logical as the temperature difference between waters off Peru and inflowing water from the west would be less. One study by Knutson and others does, however, point to an increase in the variability in the amplitude of El Niño from decade-to-decade 88 .

Even if the amplitude of El Niño does not change, the warmer background state means that the overall impression could still be of stronger sea surface temperature anomalies during El Niños and weaker cool anomalies during La Niñas, relative to an earlier climatology (say, 1950 to 1980).

Warmer conditions and associated changes in evaporation and precipitation could also amplify the climate effects of El Niño, even if it’s basic behaviour does not change. Gerald Meehl and colleagues from the National Center for Atmospheric Research in Boulder found that when El Niño conditions are superimposed on the projected rise in sea surface temperature, then atmospheric anomalies in the tropics increased, with wet areas becoming wetter and dry areas becoming drier 89 . This finding contrasts with those from those of another modelling study which found no such changes 90 , but is physically reasonable.

Just as the magnitude of climatic extremes associated with El Niño events may change, so could the areas affected. It seems likely that the close connection between El Niño and climate in tropical areas would continue but models suggest that El Niño’s influence on climate in areas outside the tropics will alter due to changes in atmospheric circulation 91 .

The models offer conflicting evidence on how the frequency of El Niño events may be affected by climate change. A number of modelling experiments suggest that the frequency of El Niño events is not strongly influenced by climate change 92 . In contrast, experiments by Richard Kleeman of the Bureau of Meteorology Research Centre in Melbourne and others 93 and by Stephen Wilson and others of Australia’s Commonwealth Science and Industrial Research Organisation 94 suggest that El Niño events may increase in frequency as climate changes. In the latter study, the authors found that a doubling of atmospheric carbon dioxide increases the frequency of El Niño from an average of once in every five years to about once in every three years.

While the above results paint a picture of more intense (perceived or actual) El Niño events as the world warms and a possible increase in their frequency, some other studies paint a strikingly different picture.

One such study by Mark Cane of Lamont-Doherty Earth Observatory in New York state and others 95 indicates that uniform warming over the tropical Pacific could actually cause cooling in the east - suggesting La Niña conditions - due to ocean dynamics. This finding, the authors contend, is consistent with observations in the area from 1900 to 1991 and with the suggestion that the strong, frequent El Niño events since 1975 occurred by chance. Others, however, question both the realism of the highly simplified climate model and the reconstructed observational data used in this study 96.

Perhaps the most contrary of all recent findings are those of Axel Timmermann and Mojib Latif at Max Planck Institute in Hamburg 97 . The authors find that if current emission trends continue, then the behaviour of El Niño could change significantly by 2050. These changes include: drastic alterations to the tropical annual cycle and to the whole equatorial mean current system, increases in El Niño variance (amplitude and intensity), a tendency towards more La Niñas, and a breakdown in the phase locking of El Niño to the annual cycle.

Given the contradictory nature of these findings, scientists are understandably cautious about drawing firm conclusions over how El Niño may respond to human-induced climate change. Thus, it is impossible to gauge the "human component" to recent events with any certainty, although aspects of these changes are consistent with observed background warming which may itself be partly attributable to human-induced change.

The high natural variability of El Niño means that it is likely to continue to be difficult to definitely attribute any particular events to climate change for a while yet. But this does not mean that any change will necessarily go unnoticed given the acute sensitivity of both human and ecological systems to climate variations.

Implications - Gambling with the Planet

To the extent that El Niño is a natural phenomena, it is clearly something we must learn to live with. But it is also clear that human induced climate change brings with it the possibility of more intense and more frequent, El Niño events. While it is impossible to quantify the likelihood of such an outcome, this is a risk that cannot be ignored. Indeed, as recent events show, such an eventuality could seriously damage prospects for sustainable development in countries around the world.

Since the last great event of 1982-83, responses to the threat of El Niño have been guided by the principle "forewarned is forearmed". Huge investments in monitoring and forecasting capacity are already starting to pay off. For example, warned of the risk of an El Niño in 1991, the governor of the state of Ceara in north-eastern Brazil publicised the need to plant drought resistance crops 98 . As a result, agricultural result output fell by just 20%, compared with a drop of 85% in 1982-83. The Brazilian system broke down in the mid-1990s, highlighting the need to improve high-level political awareness about El Niño and it’s local implications.

In preparation for this years event, countries along the Pacific coast of South America have invested heavily in flood control measures 99 , while the Peruvian government is steering farmers away from vulnerable crops like cotton 100 . Similarly, some farmers in Java are growing crops such as maize and cassava instead of rice as they need less water and in Australia, farmers sold off cattle and altered planting patterns to minimise loses 101 . In South Africa, the Government has made preparations for severe drought and Zimbabwe has spent US$122 million on grain purchases to fend off crop failure and possible famine 102 .

Arguably the real tragedy of this year’s El Niño is that the devastating fires witnessed in Indonesia need not to have occurred at all. Most were started deliberately - lit by farmers practising slash and burn agriculture, or most commonly, by plantation workers clearing land for timber or oil palm companies. The propensity of El Niño events to aggravate the problem was witnessed during the El Niño events of 1983, 1991 and 1994 when major forest fires also occurred 103 . Since then, the Indonesian Government has passed laws against setting forest fires 104 but evidently both these and warnings of an El Niño brewing were ignored.

Continued development of early warning systems and response strategies could undoubtedly help minimise the adverse social impacts of El Niño and help people take advantage of the good ones, albeit at a high economic cost. But for all this, potential changes in the behaviour of El Niño cannot be ignored for a number of reasons.

Firstly, the reliability of forecasts could go down as climate changes, tempering their usefulness in designing appropriate response strategies. Both detection and prediction of El Niño events would become more complex with ocean warming and a breakdown in established relationships between events in the Pacific and effects elsewhere. Even today, forecasting is an imperfect science. Forecasters scored notable successes in their anticipation of the onset El Niño conditions in both 1986 and 1991, but the 1983 and 1994 El Niño event came as a complete surprise as did the strength of the current El Niño 105 .

Secondly, an increased intensity of the extremes accompanying El Niño and/or the cumulative effects of more El Niños would increase the likelihood of catastrophic damage and the cost of response measures. All human and ecological systems need time to recoup after a major shock. It took the Peruvian economy and anchovy resource more than a decade to recover fully from the effects of 1983 106 . Agricultural systems also require a number of good years to offset the effects of poor years. Similarly, large or repeated stresses to ecological systems can cause irreversible damage.

Thirdly, and perhaps most fundamentally, many areas are becoming more - rather than less -vulnerable to climate variations. High levels of consumption and population growth are adding to pressures on resources, while land degradation and pollution are undermining the resource base. Resistance to pesticides and drugs that help keep disease in check is also growing, while human activities are also eroding the biological gene pool from which alternatives may be developed. Such trends mean that the effects of any given El Niño may felt more acutely in future than in the past.

In this context, it seems that any increase in the intensity and/or frequency of El Niño events should it occur would extract a heavy toll. While it is impossible to ascertain the magnitude of the risk of such an occurrence, it is clear that by continuing to pump greenhouse gases into the atmosphere we are gambling with the lives of millions of people around the world and risking major economic and ecological disruption.

Conclusion - a Precautionary Approach

The current El Niño is a timely reminder of the vulnerability of humans to climate variations and, ultimately, to climate change. It is all the more notable as it is just the latest example of unusual El Niño behaviour in recent years. While such events could have occurred naturally, aspects are consistent with an observed background warming which is itself may be partly attributable to human-induced climate changes.

Scientists do not know exactly how El Niño will respond to human-induced climate change, but preliminary results point towards an increase in intensity of events - if only because of warmer background conditions - and changes in the frequency of occurrence. While it is impossible to quantify the risk of such an outcome, it is clear that this would have severe consequences.

Faced with the possibility that changes in El Niño would compound the already severe impacts of climate change, Greenpeace is advocating early and significant action to cut greenhouse gas emissions. An important next step towards this goal would be international agreement at Kyoto in December for developed countries to cut their carbon dioxide emissions by 20 per cent on 1990 levels by 2005.

Achieving this will take political courage. But the alternative - half-measures and watching the impacts unfold - is, Greenpeace believe, totally inadequate. Only by supporting a strong agreement to cut emissions of greenhouse gases, will national governments be able to look their drought- or flood-stricken populations straight in the eye and say, "we are doing all we can".