Publication Abstracts
Erfani et al. 2026
, K.D. Lamb, , , , and , 2026: Spatiotemporal machine learning approaches for atmospheric composition emulation in NASA GISS ModelE. J. Geophys. Res. Mach. Learn. Comput., 3, no. 3, e2025JH001011, doi:10.1029/2025JH001011.
Earth System Models (ESMs) rely on parameterizations to represent sub-grid scale processes that cannot be explicitly resolved at typical model resolutions. However, maintaining full coupling between these parameterizations and other model components creates substantial computational demands. This challenge is particularly acute for atmospheric composition modules, where numerous aerosol species and constituents must be advected and processed at each model timestep. The resulting computational overhead severely constrains the feasibility of conducting long-term, high-resolution climate projections. To address these computational limitations, the NASA GISS-E3 (ModelE) employs a Non-Interactive Tracer (NINT) methodology, utilizing pre-calculated monthly climatologies of atmospheric composition fields. While computationally efficient, this approach eliminates the dynamic feedbacks between meteorological variability and chemical processes. This work introduces a machine learning (ML) framework designed to bridge this gap by creating a "Smart-NINT" system. Our approach leverages neural networks to approximate the advection, and removal terms that govern tracer evolution, allowing for meteorologically responsive composition fields without explicit tracer transport calculations. The methodology focuses on surface-level dynamics, specifically targeting Black Carbon aerosols from wildfire sources. We trained our models using two years of fully coupled ModelE output (1950?1951). The models were evaluated for their ability to capture both temporal and spatial dependencies. Results revealed consistent performance across approaches, with ${R}^{2}$ averaging 0.80. The Convolutional Long Short-Term Memory (ConvLSTM) architecture, which combines convolutional and recurrent neural networks for spatiotemporal processing, demonstrated the best performance.
Export citation: [ BibTeX ] [ RIS ]
BibTeX Citation
@article{er01100l,
author={Erfani, M. H. and Lamb, K. D. and Bauer, S. E. and Tsigaridis, K. and van Lier-Walqui, M. and Schmidt, G.},
title={Spatiotemporal machine learning approaches for atmospheric composition emulation in NASA GISS ModelE},
year={2026},
journal={Journal of Geophysical Research: Machine Learning and Computation},
volume={3},
number={3},
pages={e2025JH001011},
doi={10.1029/2025JH001011},
}
[ Close ]
RIS Citation
TY - JOUR ID - er01100l AU - Erfani, M. H. AU - Lamb, K. D. AU - Bauer, S. E. AU - Tsigaridis, K. AU - van Lier-Walqui, M. AU - Schmidt, G. PY - 2026 TI - Spatiotemporal machine learning approaches for atmospheric composition emulation in NASA GISS ModelE JA - J. Geophys. Res. Mach. Learn. Comput. JO - Journal of Geophysical Research: Machine Learning and Computation VL - 3 IS - 3 SP - e2025JH001011 DO - 10.1029/2025JH001011 ER -
[ Close ]
