In 1997, the Pacific Ocean off the coast of Peru warmed so much that fish left and rains lashed the desert. That was El Niño: a warm current that, every few years, changes weather across the planet. Predicting its arrival six months ahead is already possible, but looking further is like guessing a ship's route from a single point on a map.
Yet the climate system has its own map: the attractor. This is the shape traced by the system's trajectory in state space when the axes are not time but the variables themselves: pressure, temperature, wind speed. For El Niño, that map is the Southern Oscillation Index (SOI), the pressure difference between Tahiti and Darwin, recorded monthly since 1866. Each El Niño event leaves its own loop on this map, like a ship's wake.
The researchers decomposed 135 years of records into nine layers, from fast intraseasonal oscillations to a slow trend, and built a nine-dimensional space. Then, for the 1997–1998 event, they took its starting point and searched the past for similar points. Three were found: 1911, 1972–1973, and 1982–1983. Shifting their trajectories to the event's start and averaging them yields a forecast. The error two years ahead was less than half the standard deviation of the whole signal, and at four years it was no more than 0.37. This means the map of the past could have revealed the outlines of the future in advance.
But the map does not always work. For the 1982–1983 event, no suitable loop was found in the past, and the forecast remained inaccurate. Possibly this is a rare state of the attractor that has not repeated in 135 years of observations. Nevertheless, for most events of the 20th century, the method gives an average forecast two to four years ahead, and this is the first time it has been shown for the extreme El Niño of 1997–1998, which usually defied prediction.
Previously, it was thought that looking beyond six months was nearly impossible: too much noise in the data. Now it is clear that the SOI itself encodes nonlinear information sufficient for forecasts years ahead. The method requires no complex models, only wavelet decomposition and a search for similar orbits. It can be applied to other climate series to test whether predictability holds beyond the 20th century. The map of the past, if you learn to read it, shows the road to the future.
The 1997–1998 El Niño was so strong that it is called the 'climate event of the century': it caused droughts in Indonesia and Australia, floods in Peru and California, and is estimated to have led to the deaths of more than 20,000 people.
In Kim Stanley Robinson's novel 'Forty Signs of Rain', climatologists try to predict abrupt changes in ocean currents, and in the film 'The Day After Tomorrow', a sudden climate shift begins with anomalies in the Pacific Ocean. The theme of long-term forecasting of such events remains fantastically difficult.