Charging El Niño with off-equatorial westerly wind events

Abstract

The buildup of the warm water in the equatorial Pacific prior to an El Nino event is considered a necessary precondition for event development, while the event initiation is thought to be triggered by bursts of westerly wind. However, in contrast to the view that warm water slowly builds up years before an El Nino event, the volume of warm water in the equatorial Pacific doubled in the first few months of 2014 reaching values that were consistent with the warm water buildup prior to the extreme 1997/1998 El Nino. It is notable that this dramatic warm water buildup coincided with a series of westerly wind bursts in the western tropical Pacific. This study uses linear wave theory to determine the effect of equatorial and off-equatorial westerly wind events on the Warm Water Volume (WWV) of the Pacific. It is found that westerly wind events have a significant impact on equatorial WWV with all events initially acting to increase WWV, which highlights why WWEs are so effective at exciting ENSO. In fact, our results suggest that the single westerly wind burst, which peaked in the first few days of March in 2014, was largely responsible for the coincident dramatic observed increase in WWV. How long the equatorial region remains charged, however, depends on the latitude of the westerly wind event. For instance, a single equatorially symmetric westerly wind event ultimately acts to discharge WWV via the reflection of upwelling Rossby waves, which makes it difficult to more gradually build WWV given multiple WWEs. In contrast, when the wind events occur off the equator, the subsequent discharge is significantly damped and in some cases the equatorial region can hold the heat charge for the duration of the simulations (~6 months). As such, off-equatorial WWEs can not only charge equatorial region WWV in the short term, but are also a mechanism to more gradually build equatorial region WWV in the longer term. Given that these off-equatorial WWEs have a relatively small projection onto the equatorial Kelvin wave, we argue these events can be considered as a mechanism to modulate the background state in which ENSO operates.