Multi-year La Niña frequency tied to southward tropical Pacific wind shift

El Niño-Southern Oscillation (ENSO), the Earth's dominant year-to-year climate variability, fluctuates between El Niño (warmer SSTs) and La Niña (colder SSTs) phases in the central-eastern equatorial Pacific. Classic ENSO theories, like the recharge-discharge oscillator, describe this oscillation. El Niño involves westerly wind anomalies causing eastward propagation of downwelling Kelvin waves, leading to a deepened thermocline in the east and a shoaled thermocline in the west, culminating in a decrease of upper ocean heat content. Conversely, La Niña's easterly wind anomalies promote equatorward transport and recharge the heat content. However, ENSO displays asymmetry: La Niña events lasting more than a year are more frequent than multi-year El Niño events. Since the 21st century's start, three multi-year La Niña sequences have significantly impacted global climate through teleconnections, causing events like frequent floods in eastern Australia and persistent droughts in the United States. Understanding the dynamics behind multi-year La Niña occurrences is crucial for improved prediction and projection. While observational evidence suggests weaker recharge during La Niña compared to El Niño discharge, leading to longer La Niña durations, explanations remain diverse and lack consensus, encompassing factors like subsurface thermal anomalies, off-equator influences, inter-basin interactions, and the meridional width of the tropical Pacific SST pattern. This study investigates the link between multi-year La Niña occurrences and tropical Pacific upper-ocean heat content, analyzing both observations and climate models to address this gap in understanding.

Previous research has explored several mechanisms to explain the asymmetry in the duration of El Niño and La Niña events, and the higher frequency of multi-year La Niña events. These include the relative strength of the recharge and discharge processes in the tropical Pacific Ocean, with La Niña's recharge often being weaker, thus leading to its prolonged duration. Other studies have highlighted the roles of subsurface thermal anomalies, off-equatorial processes, inter-basin interactions, and the meridional width of the tropical Pacific SST pattern. However, a common understanding explaining multi-year La Niña occurrences, particularly regarding decadal variations and model diversity, has been lacking. This current study aims to bridge this gap by focusing on the relationship between the frequency of multi-year La Niña events and the upper-ocean heat content of the tropical Pacific Ocean.

This study utilizes observational data and climate model simulations from the Coupled Model Intercomparison Project Phase 6 (CMIP6). For observations, the classic Niño3.4 SST index (170°W-120°W, 5°S-5°N) was used to identify El Niño and La Niña events. Multi-year La Niña events were defined as consecutive La Niña occurrences over two or three years. The integrated warm water volume (WWV) above the 20°C isotherm (120°E-80°W, 5°S-5°N) served as a proxy for equatorial Pacific Ocean heat content, and its skewness was calculated using an 11-year sliding window to capture decadal variations. CMIP6 data from 33 coupled climate models were processed similarly to count multi-year La Niña occurrences and calculate WWV skewness. The relationship between multi-year La Niña frequency and WWV skewness was analyzed in both observations and models. To investigate the underlying dynamical mechanisms, zonal wind anomalies were regressed on Niño3.4 SST, focusing on the austral summer (DJF) season. Empirical orthogonal function (EOF) analysis was applied to the ENSO-related zonal winds to identify the southward wind shift pattern. Meridional heat transport was calculated using subsurface temperature and oceanic meridional velocity to quantify the recharge/discharge rate. Finally, the relationship between the southward wind shift and the thermocline tilt was explored to understand its role in modulating the WWV skewness and multi-year La Niña frequency. The ENSO nonlinearity was also quantified using a quadratic fit of the principal components from EOF analysis of tropical Pacific SSTs. Statistical significance was assessed using a bootstrap method.

The study found a strong negative correlation between multi-year La Niña frequency and WWV skewness in both observations and CMIP6 models. More frequent multi-year La Niña events were associated with a greater propensity for the tropical Pacific Ocean to discharge heat (negative WWV skewness). This relationship was further explained by the southward tropical Pacific wind shift during austral summer. Models simulating stronger westerly anomalies south of the equator showed steeper east-to-west upward tilt of the thermocline, leading to a higher discharge rate and more negative WWV skewness. This southward wind shift pattern was consistent between observations and models. The inter-model spread in multi-year La Niña processes was substantial, highlighting the need for better model constraints. Furthermore, models with greater westerly anomalies south of the equator exhibited greater negative ENSO nonlinearity, a characteristic consistent with more realistic ENSO simulations featuring distinct Central Pacific and Eastern Pacific regimes. The study also shows that strong El Niño events can contribute to subsequent multi-year La Niña occurrences by inducing stronger heat discharge, however there also exist independent multi-year La Niña events that do not follow strong El Niño.

The findings directly address the research question of understanding the mechanisms behind multi-year La Niña events. The study demonstrates a strong link between the frequency of these events and the upper-ocean heat content of the tropical Pacific, mediated by the southward wind shift during austral summer. The consistent relationship observed across both observations and a large ensemble of CMIP6 models strengthens the conclusions. The southward shift's role in modulating thermocline tilt and heat discharge explains the observed relationship between WWV skewness and multi-year La Niña frequency. The results highlight the importance of accurately simulating the southward wind shift in climate models to improve ENSO prediction and projection, given its significant influence on ENSO nonlinearity and the asymmetry in multi-year El Niño and La Niña occurrences. This research contributes to a more complete understanding of ENSO dynamics and its broader implications for climate prediction and projection, particularly regarding the increasing frequency of multi-year La Niña events in a changing climate.

This study establishes a robust link between multi-year La Niña frequency and southward tropical Pacific wind shift, mediated by upper-ocean heat content and thermocline tilt. The findings underscore the importance of accurately simulating the southward wind shift in climate models for improving ENSO prediction and projecting future climate variability. Further research could focus on refining the representation of air-sea interactions and improving model fidelity in simulating the subtle nuances of the wind shift's influence on ENSO dynamics.

The study's reliance on CMIP6 models introduces inherent limitations, as models are imperfect representations of the climate system. The relatively short observational record might limit the statistical power to fully assess decadal variations. Additionally, the focus on the southward wind shift as a primary driver does not exclude the potential influence of other factors, though the study's findings strongly suggest a central role for this mechanism. The methodology for assessing ENSO nonlinearity depends on fitting a quadratic function to EOF components. Alternative approaches could potentially offer further insights into the nonlinear dynamics of ENSO and its influence on multi-year La Niña events.