Advancing Understanding of ENSO’s Relationship with Equatorial Pacific Thermocline

Abstract: Equatorial Pacific thermocline plays an important role in El Niño-Southern Oscillation (ENSO) cycles. On one hand, thermocline fluctuations strongly affect the equatorial sea surface temperature (SST) via vertical upwelling and entrainment of subsurface heat on the base of mixed layer (so-called thermocline feedback). On the other hand, the basin-wide volume of warm water above the thermocline (known as WWV) leads the ENSO SST anomalies by about 2-3 seasons, making WWV a good predictor of ENSO. However, these relationships between ENSO and thermocline are strongly debated with a trend towards more frequent occurrences of Central Pacific (CP) versus Eastern Pacific (EP) El Niño events after 2000. Moreover, this trend was accompanied by a shortening of WWV lead-time to ENSO SST and a noticeable decline in the ENSO predictability. Improved understanding of these relationships is critical, and could help to advance ENSO modeling and prediction.

In this talk, I will show some advances in understanding ENSO-thermocline relationships using observations and hierarchical models. Firstly, we found that in the central equatorial Pacific ENSO subsurface temperature correlates poorly with thermocline but well with sea surface height. This observed difference occurs commonly in almost all CMIP6 models. We demonstrate that it arises from a unique vertically slanted distribution of the subsurface temperature in the central Pacific in response to winds associated with ENSO, whereas the oceanic high-order baroclinic modes play a crucial role. Our findings have implications for better understanding the different importance of thermocline feedback in CP and EP events.

Secondly, we derived an analytical solution of WWV lead-time to ENSO SST from the recharge-discharge oscillator framework. The solution can explain the observed decadal variations in the lead-time of WWV to SST. We demonstrate that the recent shortened WWV lead-time observed after the year 2000 can be largely explained by a decrease in the inherent ENSO periodicity, and secondly by an increase in the air-sea coupling strength with an ENSO pattern shift from EP to CP in recent decades. Our finding has deepened the understanding of WWV dynamics and its relation with ENSO and provides a framework to evaluate climate model performances in terms of simulating ENSO precursors correctly.