JAX Circulation Model
-
Madan, J. Varan1, 2
-
Davenport, Ellen H.3, 1
- Gjini, Rebecca1, 2
-
Brzenski, Jared1, 2
- Ho, Nick2
-
Hsu, Tien-Yiao2, 1
- Liang, Yueshan2
- Liu, Zhixing2
- Manivannan, Veeramakali2
- Pham, Eric2
- Vutukuru, Rohith2
- Williams, Andrew I.L.2, 1
- Yang, Zhiqi2
-
Yu, Rose2
-
Lutsko, Nicholas2
-
Hoyer, Stephan4
-
Watson-Parris, Duncan5, 2
Description
The JAX Circulation Model (JCM) is an open-source, differentiable atmospheric model built in Python using JAX. JCM couples physics parameterizations to the Dinosaur dynamical core through a flexible interface that makes substitution of other schemes easy. This modularity supports benchmarking across physical and machine-learned schemes, with direct access to gradients for sensitivity analysis, calibration, and online learning. The default parameterization scheme uses the SPEEDY (Simplified Parameterizations, primitivE-Equation DYnamics) intermediate-complexity physics scheme.
Files
jax-gcm-1.0.0.zip
Files
(8.5 MB)
| Name | Size | Download all |
|---|---|---|
|
md5:0b8f3eb699db8afe9982dfb552c5e8df
|
8.5 MB | Preview Download |
Additional details
Funding
- U.S. National Science Foundation
- CAREER: The Large-scale Buffering of Shallow Cloud Perturbations 2441832
Software
- Repository URL
- https://github.com/climate-analytics-lab/jax-gcm/
- Programming language
- Python
- Development Status
- Active