UC Irvine

No two El Niño-Southern Oscillation (ENSO) events evolve the same way. Transitions from one ENSO event to another occur in various ways and constitute a key component of ENSO complexity. While ENSO complexity in amplitude, periodicity, and spatial patterns has been frequently studied, ENSO transition complexity has not yet been systematically explored in either the observations or climate model simulations. This dissertation uses statistical analyses and numerical modeling experiments to develop a dynamical framework that can explain how three key transition patterns of ENSO (i.e., episodic, cyclic, and multi-year ENSOs) are produced, and further applies this framework to examine these ENSO transitions in contemporary climate models and to project future changes of the transition complexity.

This dissertation finds that the occurrence of the three ENSO transitions depends on two primary onset mechanisms of ENSO. The tropical Pacific onset (TP-onset) mechanism initiates sea surface temperature (SST) anomalies in the equatorial eastern Pacific through thermocline variations, while the subtropical Pacific onset (SP-onset) mechanism brings the subtropical SST anomalies into the equatorial central Pacific through a series of subtropical couplings. The TP-onset mechanism is found to mainly produce the cyclic ENSO transition and, as a result, contributes to reduced ENSO transition complexity. In contrast, the SP-onset mechanism is found to be capable of producing all three transitions and is a key source of the transition complexity. While the TP-onset mechanism has maintained its strength in the past six decades, the SP-onset mechanism became more important in producing ENSO events since the early-1990s. The intensified SP-onset mechanism increases ENSO complexity and can be a factor for the changing ENSO properties observed in the 21st century.

By further focusing on the SP-onset mechanism, this dissertation finds that the tropical mean state of SSTs can control the occurrence of cyclic and multi-year ENSO transitions. Specifically, the mean SST in the eastern equatorial Pacific controls the frequency of cyclic transitions related to the SP-onset mechanism, while the mean SST in the central equatorial Pacific is responsible for the multi-year transitions. This control arises from the fact that the mean state SST determines how easily the anomalous warming/cooling from an ENSO event can excite deep convective heating in each region to activate the SP-onset mechanism and trigger another ENSO event in the following year. In a future warmer world, this mean state control is projected to increase the cyclic transition but decrease the multi-year transition of ENSO.

In this dissertation, ENSO transition complexity is also compared between observations and contemporary climate model simulations. The El Niño transition complexity is found to be dominated in order by the episodic transition, cyclic transition, and multi-year transition. Interestingly, the reversed order is discovered for the La Niña transition complexity (multi-year, cyclic, and then episodic La Niña). This asymmetry between El Nino and La Niña transitions results from two reasons: 1) the SP-onset mechanism generates episodic El Niño more than the episodic La Niña due to the nonlinear growth of the equatorial wind anomalies it induces, 2) the nonlinear responses of the SP-onset mechanism to tropical Pacific mean SSTs favor more multi-year La Niña than multi-year El Niño. Contemporary models realistically produce the observed transition complexity for El Niño but fail to simulate the reversed order of La Niña transitions. These deficiencies arise from a weak subtropical onset mechanism in the models and a cold bias in tropical Pacific mean-state.

Findings from this dissertation offer a novel perspective to understand and study ENSO complexity dynamics, which is a new area of ENSO research. The dynamical framework developed has the potential to branch a critical new direction of understanding ENSO dynamics, properties, and activities in present, past, and future climates.