Researchers to use the power of markets to break El Niño’s ‘Spring Barrier’

Experts involved in Lancaster University Management School’s CRUCIAL project will use a market-based method for combining climate modelling expertise in a bid to produce more accurate predictions of the ‘El Niño’ phenomenon — a recurring climate pattern that affects global weather.

The ability to forecast El Niño months ahead would be extremely valuable to farmers, water resource managers, and emergency planners across the region, as it is associated with anomalous and extreme weather.

The warm phase of the El Niño–Southern Oscillation (ENSO), El Niño sees abnormally warm sea surface temperatures in the central and eastern tropical Pacific. It occurs every 3–5 years but is notoriously hard to forecast beyond spring, which is commonly known as the “spring predictability barrier.”

El Niño brings with it an increased risk the South Asian monsoon won’t take place, of drought in Australia, heavy rainfall in California, and a quieter than usual Atlantic hurricane season. The opposite type of event — La Niña — brings cooler sea surface temperatures to the tropical Pacific, higher rainfall to Indonesia but an increased risk of drought in other places, such as California.

The Climate Risk and Uncertainty Collective Intelligence Aggregation Laboratory (CRUCIAL) — a collaboration between researchers at Lancaster University Management School (LUMS) and the University of Exeter’s Land, Environment, Economics and Policy Institute (LEEP) — will use “prediction markets” in which climate scientists with expertise in predicting El Niño will ‘bet’ on sea surface temperatures in the tropical Pacific up to 18 months into the future.

The climate experts come from eight collaborating British universities (Cardiff, Imperial, Newcastle, Oxford, Bristol, Exeter, Glasgow and Reading), the Universities of Colorado Boulder and Hawai’i at Mānoa in the U.S., and Atmospheric G2, a weather forecasting company also based in the U.S.

Participants in CRUCIAL’s market — which is funded by SCOR Foundation for Science, the philanthropic arm of the French reinsurer, SCOR — do not have to pay to take part but are compensated based on contributions they make to the accuracy of the collective forecasts. Participants might base their predictions on different types of models — including Artificial Intelligence (AI) as well as more traditional climate models — which are all aggregated into a single probabilistic forecast.

Both El Niño and La Niña events tend to peak in northern hemisphere winter and decay into the spring, by which time the oscillation is more likely to be in a neutral state and predicting how it will evolve is harder. Coupling between the ocean and the atmosphere, that lies at the heart of the oscillation, is weaker in spring. This may also contribute to the difficulty of forecasting ENSO. Some types of models can forecast better during spring than other types, so it is hoped that improving and combining predictions from multiple models will allow skilful predictions beyond the “spring predictability barrier”.

Scott St. George, Head of Weather and Climate Research at WTW, an insurance company that is part of CRUCIAL’s Steering Group, said,

“El Niño and La Niña together are the number one cause of year-on-year changes in weather worldwide. Many of the top meteorological agencies make predictions of these events three, six or 12 months in advance so we can be prepared for their impacts. But it’s not enough to only make predictions, we also need to judge their performance. By putting predictions of El Niño and La Niña to the test, this competition led by CRUCIAL will show exactly how useful these forecasts can be in support of real-world decision making.”

LUMS’ Dr Mark Roulston, CRUCIAL’s Operations Director, added:

“It’s a competition between the teams taking part, but the goal is not to identify the best forecaster, but to create the best forecast, by synthesizing their expertise using a market-based approach.”

Director of CRUCIAL, Dr Kim Kaivanto from LUMS, explained:

“CRUCIAL’s prediction markets aggregate and synthesize forecast-relevant information in real-time, providing the most up-to-date consensus view across experts, information sources, and modelling methodologies. They are not an alternative to experts but a more efficient way to harness their expertise in an incentivized, skin-in-the-game way.”

The original version of this post appeared 4 June 2025 as the LU/LUMS news item Researchers to use the power of markets to break El Niño’s ‘Spring Barrier’