English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Recent studies have shown that deep learning (DL) models can skillfullypredict the El Ni\~no-Southern Oscillation (ENSO) forecasts over 1.5 yearsahead. However, concerns regarding the reliability of predictions made by DLmethods persist, including potential overfitting issues and lack ofinterpretability. Here, we propose ResoNet, a DL model that combinesconvolutional neural network (CNN) and Transformer architectures. This hybridarchitecture design enables our model to adequately capture local SSTA as wellas long-range inter-basin interactions across oceans. We show that ResoNet canrobustly predict ESNO at lead times between 19 and 26 months, thusoutperforming existing approaches in terms of the forecast horizon. Accordingto an explainability method applied to ResoNet predictions of El Ni\~no and LaNi\~na events from 1- to 18-month lead, we find that it predicts the Ni\~no3.4index based on multiple physically reasonable mechanisms, such as the RechargeOscillator concept, Seasonal Footprint Mechanism, and Indian Ocean capacitoreffect. Moreover, we demonstrate that for the first time, the asymmetry betweenEl Ni\~no and La Ni\~na development can be captured by ResoNet. Our resultscould help alleviate skepticism about applying DL models for ENSO predictionand encourage more attempts to discover and predict climate phenomena using AImethods.

ResoNet: Robust and Explainable ENSO Forecasts with Hybrid Convolution and Transformer Networks