Beyond ONI: Why El Niño risk is more than a temperature index

Disclaimer: The views expressed in this article are solely those of the author’s and not those of Citigroup.

2026 is the year of the super El Niño. It’s predicted to bring on extreme weather events, food supply disruptions and energy crises, but how can financial institutions prepare based on a temperature index? Sebastian Werner, Head of Climate Risk Scenario Design at Citi, explores the challenges of identifying El Niño’s specific impacts, and shares why scenarios can support decision-makers in navigating the market in 2027.

El Niño: Explained

The El Niño–Southern Oscillation, or ENSO, is one of the most important sources of year-to-year variability in the global climate system.^FN1 It is a coupled ocean–atmosphere phenomenon centred on the tropical Pacific. Under normal conditions, warm surface water is concentrated towards the western Pacific, while the eastern Pacific is considerably cooler. During El Niño this east–west contrast weakens as the central and eastern Pacific become unusually warm relative to normal. During La Niña the opposite occurs: the central and eastern Pacific become unusually cool and the normal east–west temperature contrast strengthens.

To track these changes, scientists divide the tropical Pacific into several geographical areas known as the Niño regions and compare current sea-surface temperatures with their long-term averages. The regions have names such as Niño 1+2, Niño 3, Niño 3.4 and Niño 4, reflecting different areas of the equatorial Pacific; Niño 3.4 takes its name from its location overlapping parts of the Niño 3 and Niño 4 regions. One of the most widely followed measures of El Niño strength is the Oceanic Niño Index (ONI), which focuses on this Niño 3.4 region in the central tropical Pacific. It measures how much warmer or cooler the sea surface there is than would normally be expected for that time of year. The larger and more persistent the warming, the stronger an El Niño event is generally considered to be.^FN2

These temperature changes matter well beyond the Pacific. Shifts in tropical heating alter rainfall and large-scale atmospheric circulation, creating “teleconnections” that can influence weather thousands of kilometres away. Some of these relationships are relatively well established – for example, recurring changes in rainfall across parts of southern Africa and the Americas – while others are weaker and less consistent.

But there is an important complication: these relationships are probabilistic rather than deterministic. Two El Niño events of similar headline strength do not necessarily produce the same atmospheric response or regional weather. The uncertainty therefore does not end once we know the likely strength of an El Niño, as measured by ONI. An important part of the uncertainty lies in what happens next.

Similar strength does not mean similar impacts

An index such as ONI deliberately reduces a complex ocean–atmosphere phenomenon to a single number. This is very useful for monitoring and comparing ENSO events, but does not capture all aspects of an ENSO event.

El Niño events can differ in where warming is concentrated across the tropical Pacific, how quickly warming develops, when the event peaks and how long it persists. The atmosphere can also respond differently to seemingly similar ocean conditions, altering how the effects are transmitted to distant regions.

A strong El Niño may therefore increase the probability of drought in a particular region, but this does not mean that drought will necessarily occur. This can be thought of as transmission uncertainty: uncertainty remains over how a given ENSO state will transmit through the atmosphere into regional weather.

When the expected drought did not arrive

The exceptionally strong 1997/98 El Niño provides a useful illustration. Southern Africa has historically experienced an elevated probability of drought during El Niño. Previous strong events, including the 1982/83 El Niño, had been associated with severe drought conditions. As the 1997/98 event developed into one of the strongest El Niños on record, there were therefore substantial concerns about another major southern African drought.^FN3

The regional weather response turned out to be much less straightforward. Rather than the widespread severe drought that had been feared, parts of southern Africa received substantial rainfall during the 1997/98 growing season. Regional atmospheric circulation and moisture transport from the Indian Ocean helped offset the drying conditions typically associated with El Niño.^FN4

The consequences were visible beyond meteorological data. South African maize markets entered the season facing the possibility that drought could significantly damage the crop. As rains arrived in December 1997 and January 1998 and crop conditions improved, South African white-maize futures fell as concerns over potential shortages receded.^FN5 The eventual harvest was not exceptional: planting had already been reduced and some dry periods remained. But the severe production shock that had been feared did not materialise. Markets had to adjust not because the El Niño forecast was wrong, but because its transmission into regional weather was different from what had been expected.

Southern Africa is not an isolated example. The exceptionally strong 2015/16 El Niño had Niño 3.4 temperature anomalies comparable with the 1997/98 event. Seasonal prediction models forecast the typical El Niño pattern, including wet conditions over Southern California. Yet while 1997/98 produced widespread wet anomalies along the Pacific coast, Southern California remained anomalously dry during the winter of 2015/16. Research subsequently showed that El Niño-related tropical circulation anomalies were substantially weaker than in 1997/98, resulting in a weaker subtropical jet, fewer storms and reduced precipitation over Southern California. Comparable Niño 3.4 strength had produced very different regional rainfall outcomes.^FN6

From forecasts to scenarios

As the 2026/27 El Niño strengthens, forecasters and market analysts are already translating the event into expectations for industries and commodity markets. Some expect upward pressure on sugar prices as El Niño-related weather threatens production in Brazil and parts of Asia. Copper could face a different channel, with heavier rainfall and flooding potentially disrupting mining infrastructure in Chile, while drought and associated hydropower shortages could constrain production elsewhere. Agricultural outlooks similarly point to potential dryness affecting crops in Brazil, a weaker Indian monsoon and stress on palm-oil production in Southeast Asia.^FN7

Each of these outlooks necessarily contains an assumption about the weather between El Niño and the economic outcome. Sugar prices will not rise because the Niño 3.4 index itself is high, but could rise if expected changes in rainfall and temperature materialise in important producing regions and are sufficiently severe to affect supply. Likewise, the relevant shock to copper is not El Niño itself, but whether anticipated rainfall, flooding or hydrological stress actually affects mines, power supply or transport.

This is where scenarios can complement forecasts. Rather than simply constructing “weak”, “moderate” and “strong” El Niño scenarios, alternatives can begin from a broadly similar ENSO state but explore different plausible transmission pathways. One might reflect the regional weather response historically associated with a strong El Niño; another could explore a weaker or differently located response, informed by episodes in which the typical teleconnection did not materialise. History is useful here not because it will repeat itself, but because it demonstrates what is physically possible.

Beyond a single index

None of this diminishes the usefulness of ONI or other ENSO indices. They provide readily interpretable descriptions of conditions in the tropical Pacific. The difficulty arises when a measure of the ENSO state is implicitly treated as a measure of its eventual impact.

Understanding the potential economic and financial implications of El Niño therefore requires looking beyond how warm the tropical Pacific becomes to how, where and when that warming develops, how the atmosphere responds, and how the resulting physical hazards interact with economic exposures.

For ENSO, knowing the strength of the event may be only the beginning of the story.

Tackle the defining challenges facing today’s risk leaders at RiskMinds this November!

Footnotes

1. Here referring to year-to-year variations in weather patterns around the world. See Trenberth (1997), “The Definition of El Niño,” Bulletin of the American Meteorological Society, 78(12), 2771–2778.
2. See NOAA Climate.gov, “Climate Variability: Oceanic Niño Index.”
3. See Lyon and Mason (2007), “The 1997–98 Summer Rainfall Season in Southern Africa. Part I: Observations,” Journal of Climate, 20(20), 5134–5148.
4. See Lyon and Mason (2007), op. cit.; and Lyon and Mason (2009), “The 1997/98 Summer Rainfall Season in Southern Africa. Part II: Model Simulations and Coupled Model Forecasts,” Journal of Climate, 22(13), 3802–3818.
5. See “White maize futures slide,” Mail & Guardian, 13 January 1998.
6. See Yang et al. (2018), “On the Seasonal Prediction of the Western United States El Niño Precipitation Pattern during the 2015/16 Winter,” Climate Dynamics, 51, 3765–3783.
7. See Morgan Stanley Research, “How Investors Can Weather El Niño,” 12 August 2026; and S&P Global, “How a Strong El Niño Could Reshape Agriculture and Fertilizer Markets,” 22 August 2026.