Authors: Sungduk Yu, Walter Hannah, Liran Peng, Jerry Lin, Mohamed Aziz Bhouri, Ritwik Gupta, Björn Lütjens, Justus Christopher Will, Gunnar Behrens, Julius Busecke, Nora Loose, Charles I Stern, Tom Beucler, Bryce Harrop, Benjamin R Hillman, Andrea Jenney, Savannah Ferretti, Nana Liu, Anima Anandkumar, Noah D Brenowitz, Veronika Eyring, Nicholas Geneva, Pierre Gentine, Stephan Mandt, Jaideep Pathak, Akshay Subramaniam, Carl Vondrick, Rose Yu, Laure Zanna, Tian Zheng, Ryan Abernathey, Fiaz Ahmed, David C Bader, Pierre Baldi, Elizabeth Barnes, Christopher Bretherton, Peter Caldwell, Wayne Chuang, Yilun Han, Yu Huang, Fernando Iglesias-Suarez, Sanket Jantre, Karthik Kashinath, Marat Khairoutdinov, Thorsten Kurth, Nicholas Lutsko, Po-Lun Ma, Griffin Mooers, J. David Neelin, David Randall, Sara Shamekh

Published on: June 14, 2023

Impact Score: 8.38

Arxiv code: Arxiv:2306.08754

Summary

• What is new: Introduction of ClimSim, the largest-ever dataset designed for merging machine learning with physics to improve climate projections.
• Why this is important: Modern climate projections are limited by computational constraints, leading to insufficient spatial and temporal resolution and inaccurate storm predictions.
• What the research proposes: ClimSim utilizes a hybrid machine learning-physics approach to enhance climate simulations, producing more accurate models by integrating high-resolution, short-term simulations through ML emulators.
• Results: The dataset consists of 5.7 billion pairs of input-output vectors covering multiple years at a high sampling frequency, facilitating the development of more accurate climate simulators.

Technical Details

Technological frameworks used: Hybrid ML-physics simulation framework

Models used: Deterministic and stochastic regression baselines

Data used: ClimSim dataset with 5.7 billion pairs of multivariate input and output vectors

Potential Impact

Energy, agriculture, insurance sectors, and environmental policy-making could significantly benefit or be disrupted by the advancements in climate predictions facilitated by ClimSim

Want to implement this idea in a business?

We have generated a startup concept here: ClimaGuard.