ON THE TIMING OF WARM AND COLD EL-NINO SOUTHERN-OSCILLATION EVENTS

Using wind and atmospheric pressure observations, the authors find tha t the ENSO signal has a previously unnoticed structure fundamental to ENSO dynamics and prediction. Specifically, the time at from the maxim um of a warm ENSO event io the minimum of the next cold ENSO event inc reases linearly with the size of the warm ENSO event. A similar result holds, but with marginal correlation, in going from a cold to a subse quent warm event. These results are consistent with a version of delay ed oscillator physics. A larger warm event implies that the westerly z onal equatorial wind anomaly is farther to the east. Consequently, the oceanic Rossby waves that the zonal wind anomaly generates take longe r to propagate to the western boundary, reflect, and return as an equa torial Kelvin wave to the region of the wind anomaly. According to del ayed oscillatory theory, the time at taken to replace the westerly win d anomaly with an easterly one is a multiple of the wave transit time, so at should increase when the size of the warm event increases. The effect is marginal in going from a cold event to a warm one because a larger cold event does not imply a greater eastward displacement of th e wind anomaly.