EnsAI: an emulator for atmospheric ammonia ensembles
- 1Environment and Climate Change Canada (Ottawa, Canada)
Abstract
Ensemble-based methods for data assimilation and emissions inversions are a popular way to encode flow-dependency within the model error covariance. While most ensemble methods do not require the use of an adjoint model, the need to repeatedly run a geophysical model to generate the ensemble can be a significant computational burden. In this paper, we introduce EnsAI , a new AI-based ensemble generation system for surface ammonia concentrations. When trained on an ensemble for ammonia generated by the GEM-MACH air quality model, it was shown that the ensembles produced by EnsAI can accurately reproduce the meteorology-dependent features of the original ensemble, while generating the ensemble 3300 times faster than the original GEM-MACH ensemble. While EnsAI requires an upfront cost for generating the ensemble used for training, as well as the training itself, the long term computational savings can greatly exceed these initial computational costs. When used in an emissions inversion system, EnsAI produced similar inversion results to those in which the original GEM-MACH ensemble was used while using significantly less computational resources.
Versions
Related
- GCMagicc v1: a fast generative emulator for multivariate climate-impact ensembles
- Geostationary observations of atmospheric ammonia over East Asia: spatio-temporal variations revealed by three years of FY-4B/GIIRS measurements
- Ocean-Aware Sea-ice Emulator for Future Hybrid Coupled Prediction Systems
- FloeNet: A Mass‐Conserving Global Sea Ice Emulator That Generalizes Across Climates
- DeepMelt-GL v1: a neural network emulator of sub-shelf melt rates for the unrepresented regions of ice-shelf cavities in ocean models
BibTeX
@article{sitwell2026ensai,
title = {{EnsAI: an emulator for atmospheric ammonia ensembles}},
author = {Michael Sitwell},
journal = {Geoscientific Model Development},
year = {2026},
month = oct,
doi = {10.5194/gmd-19-10283-2026},
url = {https://gmd.copernicus.org/articles/19/10283/2026/},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - EnsAI: an emulator for atmospheric ammonia ensembles
AU - Michael Sitwell
PY - 2026
DA - 2026/10/29
JO - Geoscientific Model Development
DO - 10.5194/gmd-19-10283-2026
UR - https://gmd.copernicus.org/articles/19/10283/2026/
AB - Ensemble-based methods for data assimilation and emissions inversions are a popular way to encode flow-dependency within the model error covariance. While most ensemble methods do not require the use of an adjoint model, the need to repeatedly run a geophysical model to generate the ensemble can be a significant computational burden. In this paper, we introduce EnsAI , a new AI-based ensemble generation system for surface ammonia concentrations. When trained on an ensemble for ammonia generated by the GEM-MACH air quality model, it was shown that the ensembles produced by EnsAI can accurately reproduce the meteorology-dependent features of the original ensemble, while generating the ensemble 3300 times faster than the original GEM-MACH ensemble. While EnsAI requires an upfront cost for generating the ensemble used for training, as well as the training itself, the long term computational savings can greatly exceed these initial computational costs. When used in an emissions inversion system, EnsAI produced similar inversion results to those in which the original GEM-MACH ensemble was used while using significantly less computational resources.
SN - 1991-9603
ER -