Maria Kielmas looks at the controversy surrounding the coming El Nino climate phenomenon

It was six weeks to Christmas and still they couldn't agree. Will the El Nino - Southern Oscillation (ENSO, popularly referred to as just El Nino) appear in late 2004 as its name suggests (see box) or in early 2005, and if so, what will be the effects? It's something many re/insurers do not wish to contemplate after the damage inflicted in the southern US by four back-to-back hurricanes in 2004.

Local insurers in New Zealand are feeling the same. After the worst winter in about 30 years, when the weather became so changeable that the population stopped listening to the forecasters, insurers are facing a 40% increase in disaster claims. Meteorologists at New Zealand's National Institute of Water and Atmospheric Research (NIWA) were unable to attribute the weather extremes to any specific event. But they did suggest that these could be due to the occurrence of a "weak" El Nino.

But further to the west in Melbourne, Australian scientists were having none of this. The good news is that an El Nino event is unlikely this (southern hemisphere) summer, said the Bureau of Meteorology (BOM). Surface temperatures in the Pacific Ocean have been hovering at levels associated with the onset of the El Nino for around two months. But this was not the case with sub (sea) surface temperatures. In addition, BOM noted, most of its computer models do not predict an El Nino event.

However, when weather forecasters cannot provide an accurate prognosis of the situation one day ahead, the public could be forgiven for not placing much faith in computer models which, for all their ingenuity, have not yet managed to simulate correctly the sea surface temperatures associated with the onset of the El Nino event in the equatorial Pacific. So in the Philippines they are depending on past experience and battening down the hatches.

The Philippine Atmospheric Geophysical and Astronomical Services Administration (PAGASA) warned in October of drier conditions for the rest of the year and warmer than normal temperatures in the first half of 2005. This was an indication of a "weak" El Nino phenomenon, the agency said, but it added that rainfall next year could be 50% down from normal.

PAGASA says it expects the El Nino will hit Luzon, eastern Visayas and Southern Mindanao and will last until June 2005. These areas may also experience a red tide - an algal bloom which contaminates fisheries and bivalve seafood such as oysters, clams etc - which is another side effect of El Nino. Already government agencies are gradually decreasing water delivery from a number of dams by between 15% and 20%, in an effort to conserve water resources.

First reports from NOAA

The first reports of a new ENSO came in July this year from the Climate Prediction Center of the US National Oceanographic and Atmospheric Administration (NOAA), based in Maryland. NOAA predicts a 60% probability of a weak ENSO in early 2005, a conclusion upheld by other US institutions such as Columbia University's International Research Institute for Climate Prediction (IRI). But Australia's BOM maintains that there is only a 30%, if that, chance of a weak ENSO. This view is supported by Latin American agencies such as Ecuador's Instituto Nacional de Meteorologia e Hidrologia (Inamhi) and Peru's Instituto del Mar del Peru (Imarpe). These disagreements mirror a similar situation in 2001 when NOAA expected an ENSO in early 2002 and the others disagreed. That ENSO finally appeared in late 2002/early 2003.

But as in the Philippines, the rest of Asia is taking the latest NOAA warnings to heart. Commodity brokers are wondering about the economic effects on major agricultural products importers such as China, if there is a surge in prices next year should an Asian drought impair harvests. The Malaysian government has said that palm oil production could be cut by 20% should an ENSO last four months. Thailand, Vietnam and Indonesia are worried about the effects on rice production, while the Indian government has expressed concern about the effects a weakened monsoon will have on crops.

How "weak"?

The description of a "weak" ENSO is full of uncertainties. The "weak" 2002-2003 ENSO had a devastating effect on Australian agriculture and caused widespread drought across much of the country. For the last decade scientist have been linking ENSO variations to a longer-term effect, the Pacific Decadal Oscillation (PDO). According to Nathan Mantua, assistant professor of meteorology at Washington State University, Seattle, the PDO lasts between 20 and 30 years, compared with ENSO's six to 18 months. The PDO's climatic effects have been most visible in the Northern Pacific and North America rather than the equatorial Pacific. Mantua identifies a cooling PDO effect between the years 1890-1924 and 1947-1976 while warmer PDOs have occurred between 1925-1946 and 1977 and the 1990s.

The ENSO variations during this latest PDO period have been attributed by some scientists to anthropogenic global warming but Professor Mantua draws no such conclusion, adding that "nothing is so simple". Difficulties have been illustrated by recent research by Michael McPhaden, senior research scientist at NOAA's Pacific Marine Environmental Research Laboratory in Seattle, and others who observed that annually averaged sea surface temperatures in the eastern and central equatorial Pacific have dropped back by approximately 0.6 degrees Celsius to their levels in the 1970s, following about 25 years of gradual increase. The question now is: how far is any recently observed warming trend reversed by natural variation? Mr McPhaden and his co-authors suggest that the abruptness of this recovery obscures perceived previous man-made warming trends.

Northern hemisphere

A further question is how far does ENSO affect the northern hemisphere. The 1997-98 event is believed, though not proven to have, influenced the climate in continental western Europe. But now researchers are looking at more historical events. A prolonged ENSO occurred in 1939-42 and is expected to have had a far larger global effect. Stefan Bronnimann of the Institute of Atmospheric and Climate Science at the Technical University in Zurich, believes that this event could have influenced stratospheric warming and an unusually high ozone column over Switzerland. It took three years to collect the data necessary to reconstruct the circulation of the time, a task conducted by scientists at the University of Berne. Mr Bronnimann acknowledges the problems with data quality from a period during the World War II but claims that the results observed are similar to those relating to other ENSO period of 1997-98, 1986-87 and 1982-83.

This could imply that the ozone layer itself could recover, but Mr Bronnimann remains cautious. "It's too early to say if it's recovering," he says stressing that the reduction in chlorofluorocarbon (CFC) would be a principal factor in such a recovery. These variations could also be related to the PDO climatic effects which can sometimes amplify the effect of ENSO in Europe, he thinks. The historical global effects of ENSO and its PDO cousin remain the subject of intense scientific debate, and it seems the future effects of ENSO will be just as keenly contested.

What is ENSO?

The El Nino - Southern Oscillation (ENSO) is an atmosphere/ocean interaction occurring over the entire tropical Pacific Ocean, in some parts extending throughout the southern hemisphere, and linking with the northern hemisphere. It represents the most prominent example of natural climate variability at seasonal to inter-annual timescales.

Original observations

El Nino is an irregular warming of the eastern equatorial Pacific Ocean while the Southern Oscillation is the associated change in pressure distribution between the eastern and western sides of the Pacific Ocean. The first recorded historical observations of El Nino in the western Pacific came in the 1500s in coastal Peru during Spanish colonial times. The warming of the coastal seas around Christmas-time gave rise to the name, El Nino, meaning "Christ Child". The Southern Oscillation was first observed by Gilbert Walker, director general of the Observatory in India. He assumed his post in 1904, shortly after the famine resulting from the 1899 monsoon - a year which coincided with an El Nino event.

Walker's aim was to predict fluctuations in the Indian monsoons. He observed a sea level pressure swing between South America and India - Australia and coined the phrase Southern Oscillation. The Southern Oscillation Index (SOI) is the pressure difference between Tahiti and Darwin on either side of the international date line. Walker never achieved his goal of predicting monsoon fluctuations and it was not until the 1960s that Jacob Bjerknes, a scientist at the University of California, proposed an empirical relation between the atmosphere and the tropical Pacific and the association of El Nino with the Southern Oscillation.

What happens?

When the pressure is persistently low over the mid-Pacific it is high over Australia and the Indian Ocean. A persistent, below average atmospheric pressure in the mid-Pacific is associated with an El Nino. The opposite set of conditions are known as La Nina. When an El Nino event occurs in eastern Australia, parts of Asia and southern Africa may be plunged into severe drought, while parts of South America and the west coast of the US may suffer unusually heavy rain and floods, though inland South and Central America are prone to drought. But relationships between rainfall and ENSOs do not occur with every event. In addition, fewer Atlantic hurricanes are observed during ENSO events. But all these signs are not necessarily observed during one particular event.

The concept

The current concept of ENSO is that of a coupled ocean-atmosphere phenomenon and the result of the cyclic warming and cooling of the surface ocean of the central and eastern Pacific. This part of the ocean is usually colder than would be expected from its equatorial location because of the influence of north-east trade winds, such as the Humboldt current which flows equator-wards along the coast of Peru and Ecuador, and the upwelling of cold deep water off the coast of Peru.

Under normal sea conditions the thermo cline - the boundary between a warmer upper ocean and a colder abyssal ocean - is drawn to the surface in the east as zonal winds blowing from east to west cause divergence in the oceanic surface currents away from the equator and colder water is brought to the surface. This creates a zonal gradient in sea surface temperatures (SSTs) - colder in the east and warmer in the west - reinforcing the easterly winds. The SST can be some 8 degrees Celsius higher in the west than the cooler temperatures in the Southern American coastal waters. The sea surface is also about half a meter higher in Indonesia than in Ecuador. The atmospheric circulation, together with the pattern of SSTs, place the deep convection over the western Pacific. Significant perturbations to any one of these components can potentially lead to a chain reaction of positive feedback, developing into an ENSO event.

Feedback

During this event the central and eastern tropical Pacific warms as the warm upper ocean waters of the western Pacific extend eastward. This reduces the SST gradient, resulting in an eastward migration and active convection and rainfall, a slackening and even reversal in near-surface easterly winds, and a dip in atmospheric surface pressure in the eastern Pacific waters (the Southern Oscillation proper). The changes in near-equatorial winds allow even more of the warm Pacific waters to move eastward. As the changes in the ocean lead to changes in the atmosphere, and vice versa, positive feedback is established. In the opposite extreme, now commonly known as La Nina, anomalies of the opposite sign are observed to grow through positive feedback. The period for a complete El Nino/La Nina cycle is typically three to seven years.

Events

Over the past 30 years ENSOs have occurred during 1972-73, 1976-77, 1982-83, 1986-87, 1991-92, 1994-95, 1997-98 and 2002-03. The two largest - 1982-83 and 1997-98 caused economic losses of $10.2bn and $34bn respectively. But such figures could be misleading given the lower asset values and developing country debt crises of the early 1980s, as well as the intensive media coverage of the 1997-98 event.

Maria Kielmas is a freelance journalist and consultant.