1.4.2.5 El Niño-Southern Oscillation (ENSO)

The El Niño-Southern Oscillation (ENSO) is the dominant interannual mode of variability in Earth’s climate. It originates in the tropical Pacific, where it affects sea surface temperatures (SST), trade winds, rainfall and many other climate variables. El Niño events typically happen every three to five years (hence the term ‘interannual’). The tropical Pacific average climate is characterised by a strong east-west gradient along the equator of about 5-6ºC, with warmer SSTs in the west and colder SSTs in the east maintained by easterly Pacific trade winds. During El Niño – the warm phase of this oscillation – this gradient weakens, while during La Niña – its cold phase – it intensifies (schematically depicted in Fig 1.4.14a). Both phases of this oscillation have far-reaching impacts on global climate and weather patterns, ecosystems and human health (e.g. McPhaden et al,. 2020). 

The impacts of ENSO become especially pronounced during the strongest events, often referred to as extreme El Niños, defined as events with SST anomalies above a chosen threshold (for example 2 standard deviations as in Heede and Fedorov 2023a) (Fig. 1.4.14b). At their peak, these events can eliminate the east-west ocean temperature gradient along the equator, leading to a temporary collapse of the trade winds. Additionally, an extreme El Niño causes an increase in global mean surface temperature of up to 0.25°C (Hu and Fedorov 2017), contributing to the prevalence of heat waves around the globe. While only a few El Niño events reach large magnitudes, the global impacts of these events result in billions of dollars in damage (Callahan and Mankin 2023).

Figure: 1.4.14
Figure 1.4.11: ENSO warm and cold phases and observational record. a Examples of strong El Niño (top) and La Niña (bottom) events seen in the tropical Pacific surface temperature (SST) distribution, with characteristic strong and weak SST gradient along the equator, respectively. b ENSO record since the 1980s. Note the three extreme events of the past four decades (1982, 1997 and 2015) and the weakening of ENSO variability between years 2000 and 2015. Temperature is averaged for the NINO3 region (5ºC-5ºN, 150ºW-90ºW) in the eastern equatorial Pacific. Based on NOAA Extended Reconstructed SST V5 data (Huang et al., 2017).

As this report was being written, a new El Niño event was announced (WMO 2023), and will likely reach peak strength around the time of its publication in December 2023. At the time of writing, it is projected to be a ‘strong’ event, reaching ~2oC relative to neutral (CPC/NCEP/NWS, 2023).

Evidence for tipping dynamics

Extensive research conducted since the 1980s has significantly advanced our understanding of the physics behind El Niño, leading to improved predictive capabilities of climate models (L’Heureux et al., 2017). ENSO is now recognised as a large-scale, irregular, internal oscillatory mode of variability within the tropical climate system, influenced by atmospheric noise (Timmermann et al., 2018). The spatial pattern of ENSO is determined by ocean-atmosphere feedbacks, while its timescale is determined by ocean dynamics. In particular, it is a sequence of self-reinforcing feedbacks between SSTs, changes in zonal surface winds, equatorial upwelling and ocean thermocline depth that promotes the growth of El Niño anomalies (i.e. Bjerknes feedbacks, McPhaden et al,. 2020). 

Coral-based proxy data indicate that the amplitude and frequency of ENSO events has gradually increased during the Holocene (Grothe et al., 2020; Lawman et al., 2022), possibly due to an increase in extreme El Niño events. All extreme El Niños in the observational record (1982, 1997 and 2015) occurred during the accelerated growth of global mean temperatures. This raises the question whether this trend is indicative of upcoming changes in the tropical Pacific to conditions with more frequent extreme El Niño events. 

In the context of tipping points, the question arises: is there a critical threshold with an abrupt and/or irreversible transition to such a new state? Several recent studies (Cai et al,. 2018, 2022; Heede and Fedorov 2023a) have indeed suggested that El Niño magnitude and impacts may intensify under global warming (Figure 1.4.12), even though there is still no model consensus on the systematic future change in ENSO, as IPCC AR6 and the results in Figure 1.4.12 suggest. 

Figure: 1.4.15
Figure 1.4.12: Overview of projected changes in extreme El Niño events in CMIP6 climate models. The bar chart shows the time-mean frequency of extreme El Niño events (the number of events per decade) for several idealised and more realistic global warming experiments (abrupt-4xCO2, 1pctCO2, SSP5-8.5 and SSP1-2.6) next to the pre-industrial Control simulation (piControl). From Heede and Fedorov 2023a

It is projected that the eastern equatorial Pacific (EEP) will warm faster than the western part of the basin, leading to an EEP warming pattern or El Niño-like mean conditions, associated with weaker Pacific trade winds. Most climate model future projections exhibit this pattern (e.g. DiNezio et al,. 2009; Xie et al,. 2010; Heede and Fedorov 2021), and increased ENSO variability is prevalent in models that simulate stronger nonlinear (Bjerknes) feedbacks (Cai et al,. 2022). A recent comprehensive study of CMIP6 models and scenarios concluded that, although a common mechanism to explain a change in ENSO activity across models is missing, its increase under warming scenarios is robust (Heede and Fedorov, 2023a).

Furthermore, during the warm Pliocene epoch approximately 3-5 million years ago, when global surface temperatures were ~3°C above pre-industrial, the east-west SST gradient was indeed reduced (Wara et al,. 2005; Fedorov et al., 2006, 2013, 2015; Tierney et al., 2019). This state is often referred to as ‘permanent El Niño-like’ conditions, which does not indicate ENSO changes, but rather a consistent mean decrease in the east-west SST gradient. While debates on this topic are ongoing, estimates for this gradient reduction range from 1.5°C to 4°C, depending on the time interval, proxy data and the definition of this gradient.

Assessment and knowledge gaps

Therefore, there is a general expectation of a future reduction in the Pacific’s east-west SST gradient by the end of the 21st Century. Together with other contributing factors, such as the strengthening of the MJO, the dominant intraseasonal mode in the tropical Indo-Pacific (Arnold et al., 2015; see 1.4.2.4), this reduction is expected to amplify ENSO (Heede and Fedorov 2023a). Additionally, a warmer atmosphere can hold more water vapour, which could result in stronger precipitation and heating anomalies in the atmosphere, leading to greater remote impacts of El Niño events. 

Consequently, the collective evidence implies an increase of El Niño magnitude and impacts under global warming. There is, however, insufficient indication for a critical transition associated with an abrupt or irreversible regime shift towards a new, more extreme or persistent, ENSO state, such that ENSO is considered with medium confidence not to be a tipping system (see also Armstrong McKay et al., 2022). However, it is well connected to other Earth system components (e.g. affecting tropical monsoon rainfall), thereby possibly playing a role in tipping cascades, linking different tipping elements via global teleconnections (see Chapter 1.5). 

Notably, the projections of a future EEP warming pattern, weaker mean trade winds and stronger El Niño events contradict decadal trends in the tropical climate over the past 30 years or so. In fact, since the early 1990s, the Pacific trade winds have strengthened, and the eastern equatorial Pacific has become colder (e.g. Ma & Zhou, 2016; Seager et al., 2022; Wills et al., 2022; Heede and Fedorov 2023b). Whether these trends reflect an ocean thermostat-like response to global warming, internal variability of the system, or both, remains an open question. Similarly, the magnitude of ENSO events has been generally weaker since the 2000s compared to the 1980s and 1990s (Figure. 1.4.11B; also Capotondi et al., 2015 or Fedorov et al., 2020). 

Therefore, debates on the future of the tropical Pacific and ENSO revolve around the question of when the transition to a mean EEP pattern and weaker trade winds may occur, likely leading to a stronger El Niño and more frequent extreme events. Simulations with global climate models including strongly eddying ocean components (Wieners et al., 2019; Chang et al., 2020) and the currently developing 2023-2024 El Niño are expected to help reduce persistent model tropical biases in SST, precipitation and ocean thermocline, and to resolve some of the remaining issues.

Bezos Earth Fund University of Exeter logo
Earth Commission Systems Change Lab logo Systemiq logo
Global Tipping Points logo
Share this content
Top