Signs of El Nino and climate upheaval.

Signs of El Nino and climate upheaval

Three years ago, climate patterns over the world were rudely altered by this century's largest El Nino -- a warming of eastern Pacific waters that occurs irregularly every three or so years and last 18 to 24 months. The last El Nino doused Ecuador with torrential rains, brought record droughts to Australia and killed much ocean life (SN: 11/5/83, p. 298).

Now a number of researchers report signs of another El Nino in the making, which, if realized, will probably by less sever than the 1982-83 episode. But more than the projected weather system itself, it is one method used in its forecast that has scientists excited. In addition to atmospheric and oceanic measurements, researchers used a new computer model that some say may represent a breakthrough in El Nino forecasting.

According to Eugene M. Rasmusson at the Climate Analysis Center of the National Weather Service in Silver Spring, Md., the present signals of a developing El Nino include warmer-than-normal waters extending from the Peruvian coast north to the equator; sea surface temperatures off Peru were lower than normal in November but during the last three months have risen faster than the usual rate, he says. Equatorial temperatures west of the date line are also higher than normal. And the southern oscillation -- an atmospheric pressure "seesaw" between the southeastern Pacific Ocean and the Australian-Indian Ocean region -- began in February to swing in a direction consistent with the development of an El Nino.

But partially because one symptom typical of an upcoming El Nino -- an increase in the tilt of sea level between the western and eastern sides of the Pacific Ocean -- has not occurred, the Climate Analysis Center has put out only an El Nino "advisory," based on current observations, rather than a forecast for the future. Rasmusson says it will probably be another two to four months before he will be able to say anything more definitive.

However, another group of scientists of Lamont-Doherty Geological Observatory in Palisades, N.Y., has made a prediction that an El Nino will strike this year and peak next winter. Mark A. Cane, Stephen E. Zebiak and Sean C. Dolan base their forecast on a simplified version of a coupled ocean-atmospher model they constructed to study the dynamics of El Ninos. Cane's group is the first to use a completely physical model to forecast El Ninos; past predictiv models hav relied on statistical comparisons of El Nino symptoms rather than on their physical causes.

In testing the model for 12 past years, three of them El Nino years, the researchers found that it was remarkably successful in forecasting whether El Ninos would occur. They looked at the 18 months preceding the year in question and made six forecasts spaced at three-month intervals. For nine of the 12 years, all six forecasts correctly predicted whether or not there would b an El Nino. For the remaining three yars, the forecasts were mixed. For 1986, all six forecasts point to an El Nino, says Zebiak.

"I don't know if this forecast is going to be right or not," comments James J. O'Brien at Florida State University in Tallahassee, "but the work is very good." Adds Rasmusson, "If the first indications of the model's skill stand up, we would be extremely interested" in using it.

Central to the model is the idea that the thickness of the top, warm-water layer in the equatorial Pacific controls the occurrence of El Ninos. While this ida is not entirely new, the model shows how critical the warm-water thickness is to El Nino development. Scientists have known that during an El Nino warm waters from the western side of the Pacific move eastward along the equator. Cane and Zebiak argue that during this journey some of the water is also deflected toward the poles, largely by the Coriolis forces of the earth's rotation. The loss of equatorial warm water eventually brings the ocean and atmosphere back to their normal stats. The next El Nino cannot be triggered until the lost waters return to the equator, a process that typically takes a few years, say the researchers.

According to Cane, past models have focused on the tilt of the thermocline, the line separating the warm layer from the underlying cold water. In his groups model, the critical parameter is not the tilt but the depth of the thermocline. To make forecasts, the researchers have only to give the computer data on surface winds over the tropical Pacific, which control the thermocline depths.

"We tried to strip down the model to include what we thought were the essential building blocks," says Zebiak. "And the succss we've had with it, given the amount of simplification and the poor quality of the wind data, is quite surprising."

Other suggestions of an impending El Nino include a prediction by Paul Handler at the University of Illinois at Urbana-Champaign that the eruption of a Colombia volcano last November will trigger an El Nino this year (see p. 185). And two weeks ago Ralph Schreiber of the Natural History Museum of Los Angeles County, Calif., found a considerable decline in the number and reproductive activity of seabirds living on Christmas Island in the mid-Pacific. Schreiber says he doesn't know if the birds' behavior is predictive of an El Nino, but the same sort of decline preceded the El Nino of 1982.