PulseAugur
EN
LIVE 08:27:18

Frozen neural weather models adapted for climate simulation with Rescene

Researchers have developed Rescene, a novel method to adapt frozen neural weather models for climate simulation. By adding a deterministic wrapper and a spectrally shaped stochastic perturbation, Rescene enables a previously unstable model to simulate climate for decades without drift. This approach successfully restores daily variability and maintains ensemble calibration, demonstrating a significant advancement in leveraging pre-trained models for long-term climate prediction. AI

IMPACT Enables long-term climate simulation using pre-trained neural weather models, potentially accelerating climate research.

RANK_REASON The cluster contains an academic paper detailing a new method for adapting existing models for a different scientific domain. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.AI →

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →

Frozen neural weather models adapted for climate simulation with Rescene

COVERAGE [1]

  1. arXiv cs.AI TIER_1 English(EN) · Minjong Cheon ·

    Rescene: band-limited stochastic forcing turns a frozen neural weather operator into a climate emulator

    arXiv:2608.09971v1 Announce Type: cross Abstract: Over the past few years, the rapid development of machine learning (ML) models for weather forecasting has produced deterministic models whose medium-range skill matches or exceeds that of the European Centre for Medium-Range Weat…