AI model can predict El Niño events up to 18 months in advance
Explainable El Niño predictability from climate mode interactions | Nature
https://www.nature.com/articles/s41586-024-07534-6
El Niño forecasts extended to 18 months with physics-based model
https://phys.org/news/2024-06-el-nio-months-physics-based.html
ENSO is a combination of the ' El Niño' phenomenon , in which the ocean temperature in the equatorial Pacific Ocean rises, and the 'Southern Oscillation,' in which sea surface pressure changes in tandem around Indonesia and the eastern South Pacific. ENSO is a phenomenon in which atmospheric movement and ocean temperature are closely linked, and it affects climate change on a global scale.
The XRO (EXtended Nonlinear Recharge Oscillator) model developed by the research team is characterized by its ability to quantify the seasonal impact of climate patterns in other oceans, such as the Indian Ocean and the Atlantic Ocean, as well as the equatorial Pacific Ocean, where ENSO occurs. As a result, the XRO model is able to make longer-term and more accurate predictions than conventional models, the research team reports.
The research team also points out that one of the ways the XRO model differs from conventional models is its 'transparency.' The XRO model can show 'why it made that prediction' by relating it to specific physical calculations.
'Unlike the black box of traditional AI models, our XRO model provides a transparent view of the physical mechanisms in the equatorial Pacific and their interactions with weather patterns outside the tropical Pacific,' said Professor Jin Feifei of SOEST. 'The XRO model can reliably quantify the impact of weather patterns on ENSO predictability.'
According to the research team, the XRO model will enable prediction of not only ENSO but also climate change in the Indian and Atlantic Oceans, and is expected to significantly improve models predicting climate not only in the equatorial Pacific but also on a global scale.
'Our research will contribute new insights to the next generation of global climate prediction models, improving approaches to predicting and mitigating the impacts of climate variability and change. These advances are crucial for societal preparedness and adaptation to climate-induced disasters,' said Assistant Professor Ruthe Stöcker from SOEST.
Related Posts:
