Last updated School of Earth and Environmental Sciences, Seoul National University YouTube 한국어

Seoul National University CSL

Climate System Laboratory

Struggle for pushing edge of the human knowledge

All research areas

Feature 06Research

Climate + AI

기후 + 인공지능

A low-dimensional recursive deep learning model for ENSO simulation (Ko et...
A low-dimensional recursive deep learning model for ENSO simulation (Ko et al. 2025, npj Clim. Atmos. Sci.) [284]

Climate System Laboratory (CSL) uses artificial intelligence to understand the dynamics of major climate phenomena and their interactions and impacts, with a particular focus on explainable AI (X-AI). We study key phenomena such as ENSO, and extend our deep learning research to the wider climate system, including the ocean, atmosphere, sea ice, land surface and biogeochemical processes. To address the uncertainty, parameterization difficulties and high computational cost of numerical climate models, we build deep learning emulators of these models that run long-term predictions at a low computational cost. We then evaluate the predictability and reliability of these predictions against observational data.

Key papers

Publications