Poster (After Call)

FZJ-2025-05426

http://join2-wiki.gsi.de/foswiki/pub/Main/Artwork/join2_logo100x88.png HClimRep: AI Climate Model for Capturing the Atmosphere, Ocean, and Sea Ice Interactions

Melidonis, S. (Corresponding author)FZJ* ; Patnala, A.FZJ* ; Semcheddine, A.FZJ* ; Schultz, M.FZJ* ; Polz, J. ; Nowak, K.

2025

Abstract: Climate change poses a significant threat to ecosystems and human society. Accurate climate projections are crucial for developing effective policies for mitigating extreme weather events that are expected to increase due to global warming. However, traditional climate models have limitations, including biases and high computational costs. Under the Helmholtz Foundation Model Initiative (HFMI), we propose a new data-driven climate model, namely HClimRep, which uses foundation model principles and machine learning to analyze diverse climate datasets. This approach enables flexible and customizable outputs, providing a versatile tool for climate applications. Keywords: Foundation Model, AI Model, Deep Learning, Climate Modelling, Climate Simulations

---

Contributing Institute(s):

1. Jülich Supercomputing Center (JSC)

Research Program(s):

1. 5111 - Domain-Specific Simulation & Data Life Cycle Labs (SDLs) and Research Groups (POF4-511) (POF4-511)
2. Earth System Data Exploration (ESDE) (ESDE)
3. HClimRep2024050120270431 - Helmholtz Representation Model for Climate Science (HClimRep) (HClimRep2024050120270431) (HClimRep2024050120270431)

Appears in the scientific report 2025

---