Science & Tech · Science & Environment · 42 mins ago
Starting forecasts during La Niña may improve year-ahead El Niño predictions
An Australian research team analyzed nine climate models to see how well they predicted El Niño about a year ahead.
El Niño is a warming of Pacific Ocean waters that can affect weather around the world, while La Niña is a cooler phase of the same climate pattern.
Forecasts that began during La Niña performed better than those starting during El Niño or neutral conditions in all nine models.
In three models, forecasts of a shift from La Niña to El Niño were correct more than 70% of the time.
Forecasts starting in neutral conditions performed poorly, showing that the Pacific's starting state can matter as much as which model is used.
Better early forecasts could give communities more time to prepare for weather risks linked to El Niño, including droughts and floods.
The researchers said the models may not match those used by forecasting agencies, so the findings need further testing before operational use.
A study found that forecasts started during La Niña were more accurate at predicting El Niño conditions about a year later than forecasts started during El Niño or neutral conditions.
The Australian research team analyzed past forecast data from nine climate models.
In three models, more than 70% of forecasts that began during La Niña and predicted a shift to El Niño were correct.
Forecasts starting during neutral conditions performed worse than a reference forecast in all nine models.
The researchers said the findings could help improve early warnings, but further checks are needed before applying them to operational forecasts.
- Who
- Researchers led by Nandini Ramesh of Australia’s CSIRO and the University of New South Wales.
- What
- The team found that starting forecasts during La Niña improved predictions of ENSO conditions about a year later.
- When
- The study was published on October 26, 2026.
- Where
- The research was reported in Science Advances. The team is based at CSIRO and the University of New South Wales in Australia.
- Why
- More accurate early forecasts could give time to prepare for weather-related hazards.
This story does not have two clearly opposing sides.
The team analyzed 10-year forecast experiments from nine climate models using past climate conditions.
The study was published in Science Advances.
- Climate models analyzed
- 9
- Forecast horizon
- About 1 year
- Correct El Niño forecasts
- More than 70% in 3 models when forecasts began during La Niña
- Average RPSS from La Niña starts
- 0.43
- Average RPSS from neutral starts
- -1.29
What people are saying
This study shows how much initial conditions can affect forecast performance, and using these results could help determine which forecasts are more reliable and warn much earlier about the possibility of El Niño.
These results show the potential to improve early warning of future climate events, but the climate models used in the analysis may not be the same as operational models used by forecasting agencies, so further validation is needed before applying them to forecasts.