Published on 09 Mar 2025
• Indian National Centre for Ocean Information Services (INCOIS) has developed a new product known as Bayesian Convolutional Neural Network (BCNN).
• BCNN predicts El Niño and La Niña conditions up to 15 months in advance using AI, deep learning, and machine learning to improve ENSO phase forecasts.
• The model relies on the connection between El Niño/La Niña and slow oceanic variations with atmospheric coupling, providing sufficient lead time for early forecasts.
• It calculates the Nino 3.4 index value, which is the average sea surface temperature (SST) anomaly in the central equatorial Pacific (5°N to 5°S, 170°W to 120°W), to make predictions.
Bayesian Convolutional Neural Network
el nino
la nina
enso
el nino southern oscillation