Publication Abstracts

Elsaesser et al. 2024, submitted

Elsaesser, G., M. van Lier-Walqui, Q. Yang, M. Kelley, A.S. Ackerman, A. Fridlind, G. Cesana, G.A. Schmidt, J. Wu, A. Behrangi, S.J. Camargo, B. De, K. Inoue, N. Leitmann-Niimi, and J.D.O. Strong, 2024: Using machine learning to generate a GISS ModelE calibrated physics ensemble (CPE). J. Adv. Model. Earth Syst., submitted.

A neural network (NN) surrogate of the NASA GISS ModelE atmosphere (version E3) is trained on a perturbed parameter ensemble (PPE) spanning 45 physics parameters and 36 outputs. The NN is leveraged in a Markov Chain Monte Carlo (MCMC) parameter inference framework to generate a second posterior constrained ensemble coined a "calibrated physics ensemble", or CPE. The CPE members are characterized by diverse parameter combinations and are, by definition, close to top-of-atmosphere radiative balance, and must broadly agree with numerous hydrologic, energy cycle and radiative forcing metrics simultaneously. Global observations of numerous cloud, environment, and radiation properties (provided by global satellite products) are crucial for CPE generation. The inference framework explicitly accounts for discrepancies (or biases) in satellite products during CPE generation. We demonstrate that product discrepancies strongly impact important model parameter settings (e.g., convective plume entrainment rates; fall speed for cloud ice). Structural improvements new to E3 are retained across CPE members (e.g., stratocumulus simulation). Notably, the framework improved the simulation of shallow cumulus and Amazon rainfall while not degrading radiation fields, an upgrade that neither default parameters nor Latin Hypercube parameter searching achieved. Analyses of the initial PPE suggested several parameters were unimportant for output variation. However, many "unimportant" parameters were needed for CPE generation, a result that brings to the forefront how parameter importance should be determined in PPEs. From the CPE, two diverse 45-dimensional parameter configurations are retained to generate radiatively-balanced, auto-tuned atmospheres that were used in two E3 submissions to CMIP6.

Export citation: [ BibTeX ] [ RIS ]

BibTeX Citation

@unpublished{el04200y,
  author={Elsaesser, G. and van Lier-Walqui, M. and Yang, Q. and Kelley, M. and Ackerman, A. S. and Fridlind, A. and Cesana, G. and Schmidt, G. A. and Wu, J. and Behrangi, A. and Camargo, S. J. and De, B. and Inoue, K. and Leitmann-Niimi, N. and Strong, J. D. O.},
  title={Using machine learning to generate a GISS ModelE calibrated physics ensemble (CPE)},
  year={2024},
  journal={Journal of Advances in Modeling Earth Systems},
  note={Manuscript submitted for publication}
}

[ Close ]

RIS Citation

TY  - UNPB
ID  - el04200y
AU  - Elsaesser, G.
AU  - van Lier-Walqui, M.
AU  - Yang, Q.
AU  - Kelley, M.
AU  - Ackerman, A. S.
AU  - Fridlind, A.
AU  - Cesana, G.
AU  - Schmidt, G. A.
AU  - Wu, J.
AU  - Behrangi, A.
AU  - Camargo, S. J.
AU  - De, B.
AU  - Inoue, K.
AU  - Leitmann-Niimi, N.
AU  - Strong, J. D. O.
PY  - 2024
TI  - Using machine learning to generate a GISS ModelE calibrated physics ensemble (CPE)
JA  - J. Adv. Model. Earth Syst.
JO  - Journal of Advances in Modeling Earth Systems
ER  -

[ Close ]

• Return to Publications Homepage