Overview

Climate science examines how the atmosphere, oceans, land, ice, and human activity shape climate, using observations and Earth system models to understand past change and project future conditions. Researchers combine physical simulations with machine-learning emulators, data assimilation, and statistical or generative downscaling to make forecasts and high-resolution projections more useful and less costly. These tools support questions from extreme-event risk and carbon fluxes to regional water, heat, and glacier change. A central challenge is balancing speed and fine detail with physical consistency, reliable uncertainty estimates, and performance under conditions unlike the training data.

Latest papers

CardsList

62 papers3 papers in the last four weeks, down 40% on the four weeks before. 0.0% of all new papers.

  1. Climate Surrogates for Scalable Multi-Agent Reinforcement Learning: A Case Study with CICERO-SCM

    Oct 9, 2025Oskar Bohn Lassen, Serio Angelo Maria Agriesti, Filipe Rodrigues +2Neural Surrogate ModelingSurrogate Modeling

  2. ArchesClimate: Probabilistic Decadal Ensemble Generation With Flow Matching

    Sep 19, 2025Graham Clyne, Guillaume Couairon, Guillaume Gastineau +2Neural Surrogate ModelingClimate Science