The gradual weakening of the trade winds across the Pacific Ocean at the beginning of El Niño events is a very important process. What causes this weakening? Much focus recently has been on the Madden-Julian Oscillation (MJO), with an enhanced MJO activity linked to westerly wind bursts and a proposed weakening of the low-level easterly winds along the Pacific. However, very strong MJO activity occurs in a number of years where the trade winds stay strong. van Loon and others have shown that a large see-saw in pressure, related to the “Southern Oscillation”, turns over at 30°S between Australia and the central South Pacific in the year leading into El Niño. At the end of the see-saw turnover (April-July year(0)), a quasi-stationary enhanced trough in the South Pacific swings southwesterly winds against the South Pacific high, and this is also proposed to weaken the trades across the Pacific.

In this study, we compared the strength of the surface trough (or subtropical ridge) in the central south Pacific (Rapa Island, 27°S, 144°W) with the strength of the trade winds in the western equatorial Pacific (5°N-5°S, 135°E-180°W). It was found that these two variables were significantly correlated between March and November (at the 99% significance level), with weaker correlations found in Austral summer when the circumpolar trough is farthest south in the annual cycle. Normalized anomalies in trade winds and trough strength tend to coincide with each other at the transition stages of El Niño (1982, 1986-87, 1991 and 1997) and La Niña events (1988, 1995, 1998). When enhanced MJO activity coincides with an enhanced South Pacific trough, El Niño events follow.

The importance of pressure changes along the equator working together with pressure changes in the southern mid-latitudes is highlighted with two new predictive indices of ENSO events. A Mid-Latitude Southern Oscillation Index (MLSOI) at 30°S normally leads the traditional SOI into extreme phases. However, the most stable index is the average of an Equatorial SOI (EQSOI), the SOI at 15°S and the MLSOI at 30°S. When all 3 SOIs decline into negative territory (between October (–1) and May(0)) the MeanSOI falls dramatically and El Niño events develop. The eight largest declines in the MeanSOI since 1958 were all El Nino years with major droughts across Australia. The years that have the next largest falls (1967, 1977, 1980) were also severe drought years across Australia, highlighting the importance of the Southern Oscillation to Australian rainfall. At the beginning of the major see-saw in pressure at 30°S leading into El Nino, low pressure anomalies over southeastern Australia indicated the strongest El Nino events (1997, 1982, 1972, 1965, 1957) in Austral spring (year (-1)). Trough strengthening and weakening in the South Pacific can also predict when El Niño events end, and whether a La Niña event will follow in the next year.

These indices form the basis of an early warning system for global climate extremes related to ENSO. Overall, these results support the notion that ENSO events occur through a combination of interactions between the air and sea, and between the southern mid-latitudes and the equator.