El Niño-Southern Oscillation (ENSO) is the strongest interannual variability signal in Earth's climate system. The shifts between its warm and cold phases profoundly impact global extreme weather, ecosystems, and economic development. However, current climate models show large discrepancies in their future projections of ENSO sea surface temperature (SST) variability.
To address this issue, a research team led by Prof. WANG Fan from the Institute of Oceanology of the Chinese Academy of Sciences (IOCAS) pointed out that deep learning can reduce ENSO projection uncertainty and enable more reliable future climate projections.
The study was published in Nature Communications on Aug. 19.
Researchers designed an "observation-informed" deep learning method, they trained 11 independent Artificial Neural Network (ANN) models using historical and future scenario data from multiple Coupled Model Intercomparison Project 6 (CMIP6) climate models, aiming to learn the complex relationship between ENSO variability and the mean state of tropical Pacific SST within each model.
They also introduced real observational data to validate these ANNs, particularly selecting those that could accurately capture the observed response of ENSO to SST changes. Through interpretability analysis and examination of ENSO physical mechanisms, the study confirmed that the top-performing ANNs successfully internalized realistic ENSO physics, showing high sensitivity to SST changes in the central equatorial Pacific and far western Pacific regions consistent with known key feedback areas for ENSO.
Using the ANNs for constrained projections of 21st-century ENSO SST variability under a high-emission scenario, the results showed that the range of predictive uncertainty was reduced by 54% compared to the raw CMIP model projections.
Moreover, although traditional analyses suggest significant differences between observations and models regarding the 20th-century tropical Pacific warming pattern, when focusing on the key regions for controlling ENSO variability identified by deep learning, both observational data and climate model simulations consistently showed a similar "El Niño-like" warming pattern.
The study reveals a previously overlooked physical consistency, not only bridging the gap between observations and models, but also uncovering hidden key physical mechanisms through deep learning. "This provides a quantifiable physical basis for future ENSO projections," said ZHU Yuchao, first author of the study.
Reduced uncertainty in ENSO SST amplitude projections using deep learning. (Image by IOCAS)
(Text by ZHU Yuchao)
Media Contact:
ZHANG Yiyi
Institute of Oceanology
E-mail: zhangyiyi@qdio.ac.cn
(Editor: ZHANG Yiyi)