AbstractPrevious work has shown that warm water volume (WWV), usually defined as the volume of equatorial Pacific warm water above the 20°C isotherm between 5°S and 5°N, leads El Niño. In contrast to previous discharge–recharge oscillator theory, here it is shown that anomalous zonal flow acceleration right at the equator and the movement of the equatorial warm pool are crucial to understanding WWV–El Niño dynamics and the ability of WWV to predict ENSO. Specifically, after westerly equatorial wind anomalies in a coupled ocean–atmosphere instability push the warm pool eastward during El Niño, the westerly anomalies follow the warmest water south of the equator in the Southern Hemisphere summer in December–February. With the wind forcing that causes El Niño in the eastern Pacific removed, the eastern equatorial Pacific sea level and thermocline anomalies decrease. Through long Rossby wave dynamics this decrease results in an anomalous westward equatorial flow that tends to push the warm pool westward and often results in the generation of a La Niña during March–June. The anomalously negative eastern equatorial Pacific sea level typically does not change as much during La Niña, the negative feedback is not as strong, and El Niños tend to not follow La Niñas the next year. This El Niño/La Niña asymmetry is seen in the WWV/El Niño phase diagram and decreased predictability during “La Niña–like” decades.
Variations and distributions of CO2in and over the equatorial Pacific during the period from the 1986/88 El Niño event to the 1988/89 La Niña event