paper
2022
Hybrid physics-ML & parameterization
New model or method
Differentiable physical modelPhysics-informed / physics-constrained3–15 days Journal
2022-11-30
016
Nov 2022
Yongquan Qu, Xiaoming Shi
Artificial Intelligence for the Earth Systems
A differentiable solver trains an ML subgrid turbulence model inside a barotropic vorticity model, extending effective forecast lead time by up to 50%, or 73% with transfer learning.
Hybrid physics-ML & parameterization
Can a Machine Learning–Enabled Numerical Model Help Extend Effective Forecast Range through Consistently Trained Subgrid-Scale Models?
In one sentence. A differentiable solver trains an ML subgrid turbulence model inside a barotropic vorticity model, extending effective forecast lead time by up to 50%, or 73% with transfer learning.
Labels
- Topic
- Hybrid physics-ML & parameterization
- Type
- New model or method
- Method
- Differentiable physical model, Physics-informed / physics-constrained
- Time scale
- 3–15 days
- Domain
- Global
- Data
- km-scale or high-res simulation
- Status
- Research
- Identifiers
- DOI:10.1175/aies-d-22-0050.1 OpenAlex:W4310442368
- Added
- 2026-10-22
Labels and summary are generated by Claude and reviewed weekly. Wrong? Edit this entry.