paper
2024
Hybrid physics-ML & parameterization
New model or method
RNN / LSTM / ConvLSTMECMWFUniversity of Oxford Other
2024-07-14
055
Jul 2024
Peter Ukkonen, Matthew Chantry
preprint
ECMWF, University of Oxford
Bidirectional vertical RNNs match feed-forward accuracy for sub-grid gravity wave, moist and unified parameterizations with fewer parameters and better offline autoregressive stability than ResNets.
Hybrid physics-ML & parameterization
Representing sub-grid processes in weather and climate models via sequence learning
In one sentence. Bidirectional vertical RNNs match feed-forward accuracy for sub-grid gravity wave, moist and unified parameterizations with fewer parameters and better offline autoregressive stability than ResNets.
Labels
- Topic
- Hybrid physics-ML & parameterization
- Type
- New model or method
- Method
- RNN / LSTM / ConvLSTM
- Domain
- Global
- Data
- km-scale or high-res simulation
- Status
- Research
- Identifiers
- DOI:10.22541/essoar.172098075.51621106/v1 OpenAlex:W4400620852
- Added
- 2026-10-22
Labels and summary are generated by Claude and reviewed weekly. Wrong? Edit this entry.