RSS Feeds

Email a Friend Printer Friendly

El Ni'±o was more intense 4000 years ago

November 07, 2000

Examination of fossil corals on Vanuatu have enabled scientists from IRD (Institut de recherche pour le développement, ex-Orstom) based at Noumea to reconstruct the sequence of variations in sea surface temperature (SST) in the southwest Pacific Ocean that occurred over a period of 50 years, 4200 years B. P. (in the Holocene). Although average sea temperature was similar to the present-day recorded value, seasonal and inter-annual variations in SST were revealed to be sharper. It appears notably that the climatic event El Ni'±o was much more intense during this period than it is today. These intriguing results lead the researchers to wonder about the reasons for such strong temperature oscillations.

Research scientists use massive fossil corals as "palaeothermometers". The coral skeleton, which has an annual growth rate of 1 cm/year and which continues for several centuries, contains trace elements, notably Sr and U, whose concentration varies according to water temperature. Analysis of the Sr/Ca and U/Ca ratios can trace the variations in sea surface temperatures (SST) throughout the growth of a coral mass.

By analysing the U and Sr contents in a colony of fossil corals of the Porites genus in Vanuatu (tropical southwest Pacific), researchers from the IRD Geosciences Laboratory at Noumea have been able to reconstruct the monthly variations in SST at 4200 years B.P. (before present), for an entire period of nearly 50 years (from 2225 to 2177 B. C.). They compared the data with those recorded in the same region during the past half-century (1951-1997). The average SSTs in the mid Holocene (27.6°C) turned out to be similar to those registered over the past 50 years (57.7°C). This is not surprising: 4200 years ago, the climate of the southwest Pacific was close to that which prevails today with almost identical figures for insolation and sea level.

The researchers have, however, observed some significant differences. Seasonal variations (between summer and winter) of SST are at present in the range 2 to 4°C, whereas 4200 years ago, these were smaller or greater depending on the period considered: in 2200 B. C., they reached 5 or even 6°C (2216 B. C.), but were only 2°C from 2199 B. C. onwards and lower than 1°C from 2193 to 2189 B. C. Why did these particularly cold, or alternatively hot, years occur in the mid Holocene? Were they part of the usual prevailing climatic cycle or did they result from external factors, such as volcanic eruptions ?(1). The southwest Pacific Ocean, particularly around Vanuatu where the data were determined, lies in one of the world's most tectonically active regions. However, there is no sufficiently strong correlation between any volcanic eruption of 4200 B. P. and the cold SST anomalies recorded in the coral reef which would support this possibility.

These strong temperature anomalies correspond to an intensification of the now well-known climatic event, the El Ni'±o Southern Oscillation (ENSO) (2). Judging by the data from the massive coral examined, ENSO did not show either the same amplitude or the same frequency as today. Episodes of La Ni'±a (expressed in Vanuatu by warm anomalies) were thus longer than they are at present, from 3 to 5 years compared with 1 to 2 years now. These phases were interrupted by El Ni'±o events (in Vanuatu manifest as cold anomalies) two to three times more intense, with an average fall in SST of 1.5 °C (as against -0.5°C on average today).

How can such a variability of ENSO be explained? The researchers advance the hypothesis that interdecadal-scale climatic variations could have had a modulation effect on the intensity of ENSO. (3) Every 10 to 20 years, upsurges of cold waters from the middle latitudes would have propagated towards the Equator and would have reinforced to some degree the effects of El Ni'±o in this part of the Pacific. During the second half of the 20th century, this modulation has been less powerful, hence the differences in ENSO's intensity between the present and the mid Holocene.

The hypothesis which evokes an ENSO modulation by an interdecadal climatic cycle, which is finding favour increasingly in the scientific community, will, however, have to be confirmed, because data for the current period have not yet been recorded for long enough to measure such variability. The study of fossil corals therefore appears to be highly valuable for tracing sea temperature oscillations over several tens or even hundreds of years.

1) Volcanic eruptions can induce atmospheric cooling events owing to ejection of dust which can block the sun's rays.
2) In the southwest Pacific, ENSO is expressed by an "anomalous" fall in SST -El Ni'±o phases- or by an equally "anomalous" rise characteristic of a La Ni'±a phase.
3) Pacific decadal oscillation (PDO), a hypothesis proposed recently by American researchers.





Institut de Recherche pour le Développement, Paris (IRD)