El Niño 2026: altimetry reveals similarities with the extreme event of 1997

An El Niño event has been developing in the Tropical Pacific for several months now. As with most El Niño events, its intensity — measured by sea surface temperature anomalies in the Equatorial Pacific — is expected to peak between November and December. At this stage, climate models are forecasting a moderate to strong, possibly very strong, event. However, uncertainties remain, as forecasting extreme events is particularly difficult.

This difficulty stems in particular from the models’ limited ability to accurately represent ocean–atmosphere interactions in the eastern equatorial Pacific. This region plays a key role in amplifying El Niño episodes. When the sea surface temperature exceeds a threshold of around 28 °C, deep atmospheric convection intensifies considerably. The resulting upward air movements promote the development of intense westerly winds at the surface, which in turn amplify warm sea surface temperature anomalies. This positive feedback, known as the Bjerknes feedback, can spiral out of control and lead to the development of extreme events, sometimes referred to as "super El Niños".

The 1997 event is a prime example of such a phenomenon. Observed at a time when satellite systems and in situ measurement networks were already largely operational, it remains the benchmark used today to assess the potential for El Niño events to intensify.

Satellite altimetry plays a central role in the early detection of the signals heralding these events. Sea level along the equator is particularly sensitive to bursts of westerly winds, which disrupt the trade winds and generate large-scale (global) ocean waves, known as Kelvin waves and Rossby waves. These waves transmit sea-level anomalies across the Pacific basin within a few months.

For an El Niño event to develop, several episodes of westerly winds generally need to occur during the first few months of the year, or even as early as the end of the previous year. In 1997, particularly pronounced Kelvin waves were generated in December 1996 and then in March 1997. Their eastward propagation led to a deepening of the thermocline in the eastern Pacific – an area normally kept cool by upwelling – thereby promoting warming sufficient to trigger the non-linear amplification characteristic of extreme events.

The situation observed in the first half of 2026 bears certain similarities to that of 1997. A first Kelvin wave was generated in January 2026, followed by a second in March 2026 (see Figure 1). The first was intense enough to trigger, two months later, the appearance of warm surface temperature anomalies along the coast of Peru. This warming led the Peruvian authorities to declare a state of alert as early as February regarding the development of a coastal El Niño. Since April, the abnormally warm waters have gradually spread out to sea under the influence of the second Kelvin wave.

Since then, an amplification mechanism appears to have taken hold. The warming induced by these waves promotes the eastward shift of the main zone of atmospheric convection, which contributes to the weakening of the trade winds and further intensifies the warming of the equatorial Pacific.

The potential magnitude of this process also depends on the thermal content of the tropical ocean, which is regarded as one of the key precursor indicators of the future intensity of an El Niño episode. This thermal content can be estimated from sea-level height anomalies observed by altimetry in the equatorial band. Current observations suggest that the heat reserves accumulated in the tropical Pacific are of the same order of magnitude as those observed in 1997 during the same period (Figure 2).

However, any comparison between the two events must be viewed with caution. The tropical Pacific is currently experiencing a warmer climate than it did during the 1990s, while mean sea levels have risen as a result of climate change (see Figure 1 and Figure 2). These changes in the average state of the ocean could influence the mechanisms underlying ENSO development, but their effects are not yet well quantified. Consequently, there remains significant uncertainty regarding the event’s final intensity this June. Climate forecasts are, however, expected to become more reliable from July and August onwards, when the seasonal predictability of ENSO generally increases.

Another major scientific question concerns the interactions between El Niño and the regional variability modes observed along the west coast of South America, commonly referred to as ‘coastal El Niños’. These regional events may result from the propagation of coastal waves originating from equatorial disturbances associated with Kelvin waves. They may also feed back into the evolution of El Niño at the basin-wide scale, either by preconditioning the ocean or by modulating regional atmospheric anomalies.

This is precisely one of the objectives of the OSTST-CENDA project, which aims to gain a better understanding of the dynamics of these coastal events through the combined use of altimetry data and ocean models of varying complexity. Altimetry, combined with ocean simulations produced by data assimilation (reanalyses), provides insight into the vertical structure of oceanic disturbances – information that is essential for understanding the mechanisms of propagation and their impacts on regional circulation.

Currently, significant warm anomalies are being observed along the coasts of Peru and Chile, where their effects are already beginning to be felt on weather conditions and marine ecosystems. If the Pacific El Niño event develops in line with the most intense scenarios envisaged by the models, the exceptionally warm conditions already present in this region could intensify further. The socio-economic consequences could then be significant, particularly for fishery resources, fishing activities and coastal communities in Peru and Chile.

B. De Witte, Cerfacs