Lab Manager | Run Your Lab Like a Business
Computer screen showing a weeks weather forecasts
iStock, JuSun

New Tool Enables El Niño Forecasting Up to 18 Months

The conceptual model is comparable to the best AI forecasts

by University of Hawaii
Register for free to listen to this article
Listen with Speechify

As more attention is drawn to possible severe weather around the world scientists at the University of Hawai?i at Mānoa are looking to improve planning for possible droughts, floods and other scenarios. Researchers from the School of Ocean and Earth Science and Technology (SOEST) created a new tool that will allow forecasting of El Niño Southern Oscillation (ENSO) by up to 18 months.

The findings, which meld insights into the physics of the ocean and atmosphere with predictive accuracy, were published in Nature.

“We have developed a new conceptual model—the so-called extended nonlinear recharge oscillator (XRO) model—that significantly improves predictive skill of ENSO events at over one year in advance, better than global climate models and comparable to the most skillful artificial intelligence [AI] forecasts,” said Sen Zhao, lead author of the study and assistant researcher in SOEST’s Department of Atmospheric Sciences. “Our model effectively incorporates the fundamental physics of ENSO and ENSO’s interactions with other climate patterns in the global oceans that vary from season to season.”

Scientists have been working for decades to improve ENSO predictions given its global environmental and socioeconomic impacts. Traditional operational forecasting models have struggled to successfully predict ENSO with lead times exceeding one year.

Peering inside the “black box”

Recent advancements in AI have pushed these boundaries, achieving accurate predictions up to 16–18 months in advance. However, the “black box” nature of AI models has precluded attribution of this accuracy to specific physical processes. Not being able to explain the source of the predictability in the AI models results in low confidence that these predictions will be successful for future events as the Earth continues to warm.

“Unlike the ‘black box’ nature of AI models, our XRO model offers a transparent view into the mechanisms of the equatorial Pacific and its interactions with other climate patterns outside of tropical Pacific,” said Fei-Fei Jin, the corresponding author and professor of atmospheric sciences in SOEST. “For the first time, we are able to robustly quantify their impact on ENSO predictability, thus deepening our knowledge of ENSO physics and its sources of predictability.”

Climate model shortcomings, improvements

“Our findings also identify shortcomings in the latest generation of climate models that lead to their failure in predicting ENSO accurately,” said Malte Stuecker, assistant professor of oceanography in SOEST and study co-author. “To improve ENSO predictions, climate models must correctly capture the key physics of ENSO and additionally, several compounding aspects of other climate patterns in the global oceans.”

“Different sources of predictability lead to distinct ENSO event evolutions,” said Philip Thompson, associate professor of oceanography in SOEST and co-author of the study. “We are now able to provide skillful, long lead time predictions of this ‘ENSO diversity,’ which is critical as different flavors of ENSO have very different impacts on global climate and individual communities.”

-Note: This news release was originally published on the University of Hawai’i website. As it has been republished, it may deviate from our style guide.