PulseAugur
EN
LIVE 08:27:25

New stochastic emulator for E3SMv3 climate model shows high fidelity

Researchers have developed a stochastic emulator for the E3SMv3 climate model, named SamudrACE, which couples an atmosphere emulator (ACE2S) with a full-depth ocean emulator (Samudra). This new system, trained on 105 years of pre-industrial simulation data, accurately reproduces E3SMv3's mean climate state and maintains internal variability, particularly in ENSO power spectrum and sea ice anomalies, compared to deterministic baselines. While the emulator captures daily precipitation well, it underestimates extreme tropical events, highlighting the ongoing challenge of extrapolating to unseen extremes. AI

IMPACT This research demonstrates advanced emulation techniques for climate modeling, potentially improving the accuracy and efficiency of long-term climate predictions.

RANK_REASON The cluster describes a scientific paper detailing a new computational model for climate simulation. [lever_c_demoted from research: ic=1 ai=0.7]

Read on arXiv cs.LG →

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

New stochastic emulator for E3SMv3 climate model shows high fidelity

COVERAGE [1]

  1. arXiv cs.LG TIER_1 English(EN) · Elynn Wu, James P. C. Duncan, Troy Arcomano, Jeremy McGibbon, Oliver Watt-Meyer, Christopher S. Bretherton, Naser Mahfouz, Claudia Tebaldi, Luke Van Roekel, Andrew Roberts, Wuyin Lin, Finn Rebassoo, Jean-Christophe Golaz, Peter M. Caldwell ·

    Stochastic Emulation of a Fully Coupled Preindustrial E3SMv3 Simulation

    arXiv:2608.10277v1 Announce Type: cross Abstract: We present a stochastic coupled emulator of E3SM version 3, built on the SamudrACE framework, which couples an atmosphere emulator (ACE2) with a full-depth ocean emulator (Samudra). We replace the deterministic atmosphere emulator…