pith. sign in

arxiv: 2510.04466 · v1 · submitted 2025-10-06 · ⚛️ physics.ao-ph · cs.LG

Benchmarking atmospheric circulation variability in an AI emulator, ACE2, and a hybrid model, NeuralGCM

Ian Baxter , Hamid Pahlavan , Pedram Hassanzadeh , Katharine Rucker , Tiffany Shaw This is my paper

Pith reviewed 2026-05-18 09:45 UTC · model grok-4.3

classification ⚛️ physics.ao-ph cs.LG
keywords atmospheric circulationAI emulatorshybrid modelsquasi-biennial oscillationSouthern annular modewave spectramodel evaluationclimate variability
0
0 comments X

The pith

AI emulators and hybrid models reproduce large-scale atmospheric wave spectra but fail to match observed timescales for the quasi-biennial oscillation and Southern annular mode.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

This paper tests how well two AI-based atmosphere models perform on fundamental measures of atmospheric variability. The fully data-driven emulator and the hybrid model both reproduce the frequency content of tropical waves and extratropical eddy interactions, including where waves break at critical levels. They fall short, however, on the multi-year period of the quasi-biennial oscillation and the roughly five-month propagation time of the Southern annular mode. These gaps are significant because models intended for climate applications must handle variability on all relevant timescales to remain reliable when applied to conditions outside their training data. The work positions these dynamical tests as practical benchmarks to guide further development of AI weather and climate tools.

Core claim

The hybrid model and emulator can capture the spectra of large-scale tropical waves and extratropical eddy-mean flow interactions, including critical levels. However, both struggle to capture the timescales associated with quasi-biennial oscillation (QBO, ∼28 months) and Southern annular mode propagation (∼150 days). These dynamical metrics serve as an initial benchmarking tool to inform AI model development and understand their limitations, which may be essential for out-of-distribution applications (e.g., extrapolating to unseen climates).

What carries the argument

Four atmospheric variability benchmarking metrics: spectra of large-scale tropical waves, spectra of extratropical eddy-mean flow interactions including critical levels, the quasi-biennial oscillation period of about 28 months, and the Southern annular mode propagation timescale of about 150 days. These metrics evaluate the models' ability to represent key dynamical processes.

If this is right

  • Models that match wave spectra can be expected to simulate many day-to-day and seasonal variability features correctly.
  • Shortcomings in QBO and SAM timescales point to specific deficiencies in representing slow, vertically propagating waves and annular mode dynamics.
  • Using these metrics allows developers to diagnose and improve AI models before deploying them for climate scenarios.
  • The benchmarks help identify which aspects of physics-AI hybrids still require refinement for long-term fidelity.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • Extending the benchmarks to additional modes of variability, such as the Madden-Julian oscillation, could reveal further limitations in the AI approaches.
  • Hybrid models might benefit from explicit parameterizations of the QBO to better capture its period without relying solely on data-driven learning.
  • These tests could be applied to future AI models to track progress in capturing multi-year climate variability.
  • Success on short-term waves but failure on longer timescales suggests that training data length or model architecture may need adjustment for slow processes.

Load-bearing premise

The selected dynamical metrics of wave spectra, QBO period, and SAM propagation are sufficient and representative for judging model fidelity in climate applications outside the training distribution.

What would settle it

A new set of model simulations or observations that shows the models producing a clear spectral peak at approximately 28 months for the quasi-biennial oscillation or a propagation speed matching 150 days for the Southern annular mode would challenge the finding that they struggle with these timescales.

Figures

Figures reproduced from arXiv: 2510.04466 by Hamid Pahlavan, Ian Baxter, Katharine Rucker, Pedram Hassanzadeh, Tiffany Shaw.

Figure 1
Figure 1. Figure 1: Monthly zonal-mean zonal wind averaged over 10 [PITH_FULL_IMAGE:figures/full_fig_p006_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Wavenumber-frequency power spectrum of the symmetric (left column) and antisymmetric (right [PITH_FULL_IMAGE:figures/full_fig_p008_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Contours of 250 hPa transient eddy momentum flux versus latitude and phase speed for DJFM [PITH_FULL_IMAGE:figures/full_fig_p009_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Frequency (cycles per day) power spectra for 80 [PITH_FULL_IMAGE:figures/full_fig_p010_4.png] view at source ↗
read the original abstract

Physics-based atmosphere-land models with prescribed sea surface temperature have notable successes but also biases in their ability to represent atmospheric variability compared to observations. Recently, AI emulators and hybrid models have emerged with the potential to overcome these biases, but still require systematic evaluation against metrics grounded in fundamental atmospheric dynamics. Here, we evaluate the representation of four atmospheric variability benchmarking metrics in a fully data-driven AI emulator (ACE2-ERA5) and hybrid model (NeuralGCM). The hybrid model and emulator can capture the spectra of large-scale tropical waves and extratropical eddy-mean flow interactions, including critical levels. However, both struggle to capture the timescales associated with quasi-biennial oscillation (QBO, $\sim 28$ months) and Southern annular mode propagation ($\sim 150$ days). These dynamical metrics serve as an initial benchmarking tool to inform AI model development and understand their limitations, which may be essential for out-of-distribution applications (e.g., extrapolating to unseen climates).

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 1 minor

Summary. The manuscript evaluates an AI emulator (ACE2-ERA5) and hybrid model (NeuralGCM) on four dynamical metrics of atmospheric variability: spectra of large-scale tropical waves, extratropical eddy-mean flow interactions (including critical levels), quasi-biennial oscillation (QBO) period (~28 months), and Southern annular mode (SAM) propagation timescale (~150 days). Both models reproduce the wave spectra and interactions but fail to capture the observed long timescales for QBO and SAM. The authors position these metrics as initial benchmarks to inform AI model development and to identify limitations relevant to out-of-distribution climate applications.

Significance. If the evaluation results are robust, the work supplies concrete, dynamics-grounded diagnostics that identify clear strengths (wave spectra) and weaknesses (low-frequency timescales) in current AI and hybrid atmospheric models. This is timely and useful for guiding model improvements in a rapidly evolving field. The focus on fundamental processes rather than bulk error metrics is a strength, though the significance for OOD applications would increase if the metrics were shown to predict extrapolation skill.

major comments (2)
  1. [Abstract] Abstract: The statement that these metrics 'may be essential for out-of-distribution applications' is not supported by any experiment that perturbs the climate (e.g., changed SST, CO2, or orbital parameters) and then verifies whether models that fail on QBO/SAM timescales also fail on OOD targets such as circulation response or extremes. Without this link, the metrics are shown to diagnose in-distribution fidelity but not demonstrated to be predictive of extrapolation performance.
  2. [Evaluation or Methods section] Evaluation or Methods section: The manuscript does not provide sufficient detail on the precise definitions and computational procedures used for the four metrics (e.g., wavenumber-frequency spectra for tropical waves, exact period estimation for QBO, or lag-correlation method for SAM propagation), nor does it report quantitative uncertainty or sensitivity to analysis choices. This limits assessment of reproducibility and robustness of the reported successes and failures.
minor comments (1)
  1. [Abstract] Abstract and introduction: Consider adding one or two references to prior dynamical benchmarking studies in physics-based models to better situate the choice of these four metrics.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their constructive and detailed comments on our manuscript. These have helped us better clarify the scope of our benchmarking approach. We address each major comment below and indicate the revisions we will make.

read point-by-point responses

  1. Referee: [Abstract] Abstract: The statement that these metrics 'may be essential for out-of-distribution applications' is not supported by any experiment that perturbs the climate (e.g., changed SST, CO2, or orbital parameters) and then verifies whether models that fail on QBO/SAM timescales also fail on OOD targets such as circulation response or extremes. Without this link, the metrics are shown to diagnose in-distribution fidelity but not demonstrated to be predictive of extrapolation performance.

    Authors: We agree that the manuscript does not contain direct experiments linking these metrics to out-of-distribution performance under perturbed climates. The original phrasing was intended as a forward-looking motivation rather than a demonstrated result. We will revise the abstract to remove the specific claim about out-of-distribution applications and instead state that the metrics diagnose in-distribution fidelity, with their relevance to extrapolation to be tested in future work. We have added a short paragraph in the conclusions noting this as an important direction for subsequent studies. revision: yes

  2. Referee: [Evaluation or Methods section] Evaluation or Methods section: The manuscript does not provide sufficient detail on the precise definitions and computational procedures used for the four metrics (e.g., wavenumber-frequency spectra for tropical waves, exact period estimation for QBO, or lag-correlation method for SAM propagation), nor does it report quantitative uncertainty or sensitivity to analysis choices. This limits assessment of reproducibility and robustness of the reported successes and failures.

    Authors: We thank the referee for this observation and agree that greater detail is required. In the revised manuscript we will expand the Evaluation section with explicit computational procedures for each metric, including the precise wavenumber-frequency spectral analysis settings for tropical waves, the method used to estimate the QBO period (peak in the Fourier spectrum of equatorial zonal wind at 10 hPa), the lag-correlation procedure and latitude band for the SAM propagation timescale, and the corresponding definitions for eddy-mean flow interactions. We will also add quantitative uncertainty estimates (e.g., standard deviation across multiple analysis windows) and brief sensitivity tests to key analysis choices such as window length and spectral resolution. revision: yes

Circularity Check

0 steps flagged

No circularity: direct benchmarking against independent dynamical metrics

full rationale

The paper evaluates ACE2 and NeuralGCM outputs against standard, externally defined atmospheric variability metrics (tropical wave spectra, eddy-mean flow interactions, QBO period, SAM propagation) by direct comparison to reanalysis. These metrics are taken from established atmospheric dynamics literature and are not derived, fitted, or redefined inside the paper. No result is obtained by construction from model outputs or self-citations; the evaluation chain remains open to external falsification. The claim that the metrics may be useful for OOD applications is presented as motivation rather than a derived theorem, preserving self-containment.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

Benchmarking study that relies on established atmospheric dynamics definitions and prior model outputs; no new free parameters, axioms, or invented entities are introduced.

axioms (1)
  • standard math Standard definitions of tropical wave spectra, eddy-mean flow interactions, QBO period, and SAM propagation from prior atmospheric dynamics literature.
    Invoked when stating the four benchmarking metrics in the abstract.

pith-pipeline@v0.9.0 · 5717 in / 1282 out tokens · 27146 ms · 2026-05-18T09:45:33.346590+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

65 extracted references · 65 canonical work pages

  1. [1]

    FourCastNet: A Global Data-driven High-resolution Weather Model using Adaptive Fourier Neural Operators, February 2022

    Jaideep Pathak, Shashank Subramanian, Peter Harrington, Sanjeev Raja, Ashesh Chattopadhyay, Morteza Mardani, Thorsten Kurth, David Hall, Zongyi Li, Kamyar Azizzadenesheli, Pedram Hassanzadeh, Karthik Kashinath, and Animashree Anandkumar. FourCastNet: A Global Data-driven High-resolution Weather Model using Adaptive Fourier Neural Operators, February 2022

    work page 2022

  2. [2]

    Forecasting Global Weather with Graph Neural Networks, February 2022

    Ryan Keisler. Forecasting Global Weather with Graph Neural Networks, February 2022

    work page 2022

  3. [3]

    Accurate medium-range global weather forecasting with 3D neural networks.Nature, 619(7970):533–538, July 2023

    Kaifeng Bi, Lingxi Xie, Hengheng Zhang, Xin Chen, Xiaotao Gu, and Qi Tian. Accurate medium-range global weather forecasting with 3D neural networks.Nature, 619(7970):533–538, July 2023

    work page 2023

  4. [4]

    Learning skillful medium-range global weather forecasting.Science, 382(6677):1416–1421, 2023

    Remi Lam, Alvaro Sanchez-Gonzalez, Matthew Willson, Peter Wirnsberger, Meire Fortunato, Ferran Alet, Suman Ravuri, Timo Ewalds, Zach Eaton-Rosen, Weihua Hu, et al. Learning skillful medium-range global weather forecasting.Science, 382(6677):1416–1421, 2023

    work page 2023

  5. [5]

    Brenner, and Stephan Hoyer

    Dmitrii Kochkov, Janni Yuval, Ian Langmore, Peter Norgaard, Jamie Smith, Griffin Mooers, Milan Klöwer, James Lottes, Stephan Rasp, Peter Düben, Sam Hatfield, Peter Battaglia, Alvaro Sanchez- Gonzalez, Matthew Willson, Michael P. Brenner, and Stephan Hoyer. Neural general circulation models for weather and climate.Nature, 632(8027):1060–1066, August 2024

    work page 2024

  6. [6]

    arXiv preprint arXiv:2406.01465 , year=

    Simon Lang, Mihai Alexe, Matthew Chantry, Jesper Dramsch, Florian Pinault, Baudouin Raoult, Mariana CA Clare, Christian Lessig, Michael Maier-Gerber, Linus Magnusson, et al. Aifs–ecmwf’s data-driven forecasting system.arXiv preprint arXiv:2406.01465, 2024

    work page arXiv 2024

  7. [7]

    Probabilistic weather forecasting with machine learning.Nature, 637(8044):84–90, 2025

    Ilan Price, Alvaro Sanchez-Gonzalez, Ferran Alet, Tom R Andersson, Andrew El-Kadi, Dominic Masters, Timo Ewalds, Jacklynn Stott, Shakir Mohamed, Peter Battaglia, et al. Probabilistic weather forecasting with machine learning.Nature, 637(8044):84–90, 2025

    work page 2025

  8. [8]

    Zied Ben Bouallegue, Mariana CA Clare, Linus Magnusson, Estibaliz Gascon, Michael Maier-Gerber, Martin Janoušek, Mark Rodwell, Florian Pinault, Jesper S Dramsch, Simon TK Lang, et al. The rise of data-driven weather forecasting: A first statistical assessment of machine learning–based weather forecasts in an operational-like context.Bulletin of the Americ...

    work page 2024

  9. [9]

    Machine learning for the physics of climate.Nature Reviews Physics, 7(1):6–20, 2025

    Annalisa Bracco, Julien Brajard, Henk A Dijkstra, Pedram Hassanzadeh, Christian Lessig, and Claire Monteleoni. Machine learning for the physics of climate.Nature Reviews Physics, 7(1):6–20, 2025

    work page 2025

  10. [10]

    Clark, Anna Kwa, W

    Oliver Watt-Meyer, Brian Henn, Jeremy McGibbon, Spencer K. Clark, Anna Kwa, W. Andre Perkins, Elynn Wu, Lucas Harris, and Christopher S. Bretherton. ACE2: Accurately learning subseasonal to decadal atmospheric variability and forced responses.npj Clim Atmos Sci, 8(1):205, May 2025

    work page 2025

  11. [11]

    Samudra: An AI Global Ocean Emulator for Climate.Geophysical Research Letters, 52(10):e2024GL114318, 2025

    Surya Dheeshjith, Adam Subel, Alistair Adcroft, Julius Busecke, Carlos Fernandez-Granda, Shubham Gupta, and Laure Zanna. Samudra: An AI Global Ocean Emulator for Climate.Geophysical Research Letters, 52(10):e2024GL114318, 2025

    work page 2025

  12. [12]

    Hakim, Wenchang Yang, and Gabriel A

    Zilu Meng, Gregory J. Hakim, Wenchang Yang, and Gabriel A. Vecchi. Deep Learning Atmospheric Models Reliably Simulate Out-of-Sample Land Heat and Cold Wave Frequencies, July 2025

    work page 2025

  13. [13]

    On Some Limitations of Current Machine Learning Weather Prediction Models

    Massimo Bonavita. On Some Limitations of Current Machine Learning Weather Prediction Models. Geophysical Research Letters, 51(12):e2023GL107377, 2024. e2023GL107377 2023GL107377

    work page 2024

  14. [14]

    Can artificial intelligence-based weather prediction models simulate the butterfly effect?Geophysical Research Letters, 50(20):e2023GL105747, 2023

    Tobias Selz and George C Craig. Can artificial intelligence-based weather prediction models simulate the butterfly effect?Geophysical Research Letters, 50(20):e2023GL105747, 2023

    work page 2023

  15. [15]

    Qiang Sun, Pedram Hassanzadeh, Mohsen Zand, Ashesh Chattopadhyay, Jonathan Weare, and Dorian S

    Y. Qiang Sun, Pedram Hassanzadeh, Mohsen Zand, Ashesh Chattopadhyay, Jonathan Weare, and Dorian S. Abbot. Can AI weather models predict out-of-distribution gray swan tropical cyclones? Proceedings of the National Academy of Sciences, 122(21):e2420914122, May 2025

    work page 2025

  16. [16]

    Predicting beyond training data via extrapolation versus translocation: Ai weather models and dubai’s unprecedented 2024 rainfall

    Y Qiang Sun, Pedram Hassanzadeh, Tiffany Shaw, and Hamid A Pahlavan. Predicting beyond training data via extrapolation versus translocation: Ai weather models and dubai’s unprecedented 2024 rainfall. arXiv preprint arXiv:2505.10241, 2025

    work page arXiv 2024

  17. [17]

    Scaife, Nick J

    Chris Kent, Adam A. Scaife, Nick J. Dunstone, Doug Smith, Steven C. Hardiman, Tom Dunstan, and Oliver Watt-Meyer. Skilful global seasonal predictions from a machine learning weather model trained on reanalysis data.npj Clim Atmos Sci, 8(1):314, August 2025

    work page 2025

  18. [18]

    Modulation of tropical cyclogenesis on subseasonal-to-interannual timescales in the deep-learning climate emulator ACE2.Mach

    Mu-Ting Chien, Elizabeth A Barnes, and Eric D Maloney. Modulation of tropical cyclogenesis on subseasonal-to-interannual timescales in the deep-learning climate emulator ACE2.Mach. Learn.: Earth, 2025. 13 A preprint - October 7, 2025

    work page 2025

  19. [19]

    Senne Van Loon, Maria Rugenstein, and Elizabeth A. Barnes. Reanalysis-Based Global Radiative Response to Sea Surface Temperature Patterns: Evaluating the Ai2 Climate Emulator.Geophysical Research Letters, 52(14):e2025GL115432, 2025. e2025GL115432 2025GL115432

    work page 2025

  20. [20]

    Craig, Andreas H

    George C. Craig, Andreas H. Fink, Corinna Hoose, Tijana Janjić, Peter Knippertz, Audine Laurian, Sebastian Lerch, Bernhard Mayer, Annette Miltenberger, Robert Redl, Michael Riemer, Kirsten I. Tempest, and Volkmar Wirth. Waves to Weather: Exploring the Limits of Predictability of Weather. Bulletin of the American Meteorological Society, 102(11):E2151–E2164...

    work page 2021

  21. [21]

    John W. Kidson. Indices of the Southern Hemisphere Zonal Wind.Journal of Climate, 1(2):183–194, February 1988

    work page 1988

  22. [22]

    Preferred Modes of Variability and Their Relationship with Climate Change.Journal of Climate, 19(10):2063–2075, May 2006

    Seok-Woo Son and Sukyoung Lee. Preferred Modes of Variability and Their Relationship with Climate Change.Journal of Climate, 19(10):2063–2075, May 2006

    work page 2063

  23. [23]

    David W. J. Thompson and Elizabeth A. Barnes. Periodic Variability in the Large-Scale Southern Hemisphere Atmospheric Circulation.Science, 343(6171):641–645, February 2014

    work page 2014

  24. [24]

    Robertson

    Frédéric Vitart and Andrew W. Robertson. The sub-seasonal to seasonal prediction project (S2S) and the prediction of extreme events.npj Clim Atmos Sci, 1(1):3, March 2018

    work page 2018

  25. [25]

    Alan Plumb

    Aditi Sheshadri and R. Alan Plumb. Propagating Annular Modes: Empirical Orthogonal Functions, Principal Oscillation Patterns, and Time Scales.Journal of the Atmospheric Sciences, 74(5):1345–1361, May 2017

    work page 2017

  26. [26]

    Anstey, Scott M

    James A. Anstey, Scott M. Osprey, Joan Alexander, Mark P. Baldwin, Neal Butchart, Lesley Gray, Yoshio Kawatani, Paul A. Newman, and Jadwiga H. Richter. Impacts, processes and projections of the quasi-biennial oscillation.Nat Rev Earth Environ, 3(9):588–603, September 2022

    work page 2022

  27. [27]

    Vallis.Essentials of Atmospheric and Oceanic Dynamics

    Geoffrey K. Vallis.Essentials of Atmospheric and Oceanic Dynamics. Cambridge University Press, January 2019

    work page 2019

  28. [28]

    Lawrence Gates

    W. Lawrence Gates. AMIP: The Atmospheric Model Intercomparison Project.Bulletin of the American Meteorological Society, 73(12):1962–1970, December 1992

    work page 1962

  29. [29]

    Meehl, Catherine A

    Veronika Eyring, Sandrine Bony, Gerald A. Meehl, Catherine A. Senior, Bjorn Stevens, Ronald J. Stouffer, and Karl E. Taylor. Overview of the Coupled Model Intercomparison Project Phase 6 (CMIP6) experimental design and organization.Geoscientific Model Development, 9(5):1937–1958, May 2016

    work page 1937

  30. [30]

    Scaife and Doug Smith

    Adam A. Scaife and Doug Smith. A signal-to-noise paradox in climate science.npj Clim Atmos Sci, 1(1):1–8, July 2018

    work page 2018

  31. [31]

    Adames, Amin Dezfuli, John Fasullo, Peter J

    Clara Orbe, Luke Van Roekel, Ángel F. Adames, Amin Dezfuli, John Fasullo, Peter J. Gleckler, Jiwoo Lee, Wei Li, Larissa Nazarenko, Gavin A. Schmidt, Kenneth R. Sperber, and Ming Zhao. Representation of Modes of Variability in Six U.S. Climate Models.Journal of Climate, 33(17):7591–7617, September 2020

    work page 2020

  32. [32]

    A. C. Bushell, J. A. Anstey, N. Butchart, Y. Kawatani, S. M. Osprey, J. H. Richter, F. Serva, P. Braesicke, C. Cagnazzo, C.-C. Chen, H.-Y. Chun, R. R. Garcia, L. J. Gray, K. Hamilton, T. Kerzenmacher, Y.- H. Kim, F. Lott, C. McLandress, H. Naoe, J. Scinocca, A. K. Smith, T. N. Stockdale, S. Versick, S. Watanabe, K. Yoshida, and S. Yukimoto. Evaluation of ...

    work page 2022

  33. [33]

    Lubis and Pedram Hassanzadeh

    Sandro W. Lubis and Pedram Hassanzadeh. The Intrinsic 150-Day Periodicity of the Southern Hemi- sphere Extratropical Large-Scale Atmospheric Circulation.AGU Advances, 4(3):e2022AV000833, 2023. e2022AV000833 2022AV000833

    work page 2023

  34. [34]

    M. P. Baldwin, L. J. Gray, T. J. Dunkerton, K. Hamilton, P. H. Haynes, W. J. Randel, J. R. Holton, M. J. Alexander, I. Hirota, T. Horinouchi, D. B. A. Jones, J. S. Kinnersley, C. Marquardt, K. Sato, and M. Takahashi. The quasi-biennial oscillation.Reviews of Geophysics, 39(2):179–229, 2001

    work page 2001

  35. [35]

    Kiladis, Matthew C

    George N. Kiladis, Matthew C. Wheeler, Patrick T. Haertel, Katherine H. Straub, and Paul E. Roundy. Convectively coupled equatorial waves.Reviews of Geophysics, 47(2), 2009

    work page 2009

  36. [36]

    Matthew Wheeler and George N. Kiladis. Convectively Coupled Equatorial Waves: Analysis of Clouds and Temperature in the Wavenumber–Frequency Domain.Journal of the Atmospheric Sciences, 56(3):374–399, February 1999. 14 A preprint - October 7, 2025

    work page 1999

  37. [37]

    Yoshikazu Hayashi. A generalized method of resolving disturbances into progressive and retrogressive waves by space Fourier and time cross-spectral analyses.Journal of the Meteorological Society of Japan. Ser. II, 49(2):125–128, 1971

    work page 1971

  38. [38]

    Randel and Isaac M

    William J. Randel and Isaac M. Held. Phase Speed Spectra of Transient Eddy Fluxes and Critical Layer Absorption.Journal of the Atmospheric Sciences, 48(5):688–697, March 1991

    work page 1991

  39. [39]

    Gang Chen and Isaac M. Held. Phase speed spectra and the recent poleward shift of Southern Hemisphere surface westerlies.Geophysical Research Letters, 34(21), 2007

    work page 2007

  40. [40]

    Lutsko, Isaac M

    Nicholas J. Lutsko, Isaac M. Held, Pablo Zurita-Gotor, and Amanda K. O’Rourke. Lower-Tropospheric Eddy Momentum Fluxes in Idealized Models and Reanalysis Data.Journal of the Atmospheric Sciences, 74(11):3787–3797, November 2017

    work page 2017

  41. [41]

    E. P. Gerber, L. M. Polvani, and D. Ancukiewicz. Annular mode time scales in the Intergovernmental Panel on Climate Change Fourth Assessment Report models.Geophysical Research Letters, 35(22), 2008

    work page 2008

  42. [42]

    Lorenz and Dennis L

    David J. Lorenz and Dennis L. Hartmann. Eddy–Zonal Flow Feedback in the Southern Hemisphere. Journal of the Atmospheric Sciences, 58(21):3312–3327, November 2001

    work page 2001

  43. [43]

    Simpson, Theodore G

    Isla R. Simpson, Theodore G. Shepherd, Peter Hitchcock, and John F. Scinocca. Southern Annular Mode Dynamics in Observations and Models. Part II: Eddy Feedbacks.Journal of Climate, 26(14):5220–5241, July 2013

    work page 2013

  44. [44]

    Revisiting the quasi-biennial oscillation as seen in era5

    Hamid A Pahlavan, Qiang Fu, John M Wallace, and George N Kiladis. Revisiting the quasi-biennial oscillation as seen in era5. part i: Description and momentum budget.Journal of the Atmospheric Sciences, 78(3):673–691, 2021

    work page 2021

  45. [45]

    Olivier Boucher, Jérôme Servonnat, Anna Lea Albright, Olivier Aumont, Yves Balkanski, Vladislav Bastrikov, Slimane Bekki, Rémy Bonnet, Sandrine Bony, Laurent Bopp, Pascale Braconnot, Patrick Brockmann, Patricia Cadule, Arnaud Caubel, Frederique Cheruy, Francis Codron, Anne Cozic, David Cugnet, Fabio D’Andrea, Paolo Davini, Casimir de Lavergne, Sébastien D...

    work page 2020

  46. [46]

    Richter, James A

    Jadwiga H. Richter, James A. Anstey, Neal Butchart, Yoshio Kawatani, Gerald A. Meehl, Scott Osprey, and Isla R. Simpson. Progress in Simulating the Quasi-Biennial Oscillation in CMIP Models.Journal of Geophysical Research: Atmospheres, 125(8):e2019JD032362, April 2020

    work page 2020

  47. [47]

    D. Rind, J. Jonas, N. K. Balachandran, G. A. Schmidt, and J. Lean. The QBO in two GISS global climate models: 1. Generation of the QBO.Journal of Geophysical Research: Atmospheres, 119(14):8798–8824, 2014

    work page 2014

  48. [48]

    Danabasoglu, J.-F

    G. Danabasoglu, J.-F. Lamarque, J. Bacmeister, D. A. Bailey, A. K. DuVivier, J. Edwards, L. K. Emmons, J. Fasullo, R. Garcia, A. Gettelman, C. Hannay, M. M. Holland, W. G. Large, P. H. Lauritzen, D. M. Lawrence, J. T. M. Lenaerts, K. Lindsay, W. H. Lipscomb, M. J. Mills, R. Neale, K. W. Oleson, B. Otto-Bliesner, A. S. Phillips, W. Sacks, S. Tilmes, L. van...

    work page 2020

  49. [49]

    Hendon, and Joowan Kim

    Seok-Woo Son, Yuna Lim, Changhyun Yoo, Harry H. Hendon, and Joowan Kim. Stratospheric Control of the Madden–Julian Oscillation.Journal of Climate, 30(6):1909–1922, March 2017

    work page 1909

  50. [50]

    QBO-MJO Connection.Journal of Geophysical Research: Atmo- spheres, 123(6):2957–2967, 2018

    Chidong Zhang and Bosong Zhang. QBO-MJO Connection.Journal of Geophysical Research: Atmo- spheres, 123(6):2957–2967, 2018. 15 A preprint - October 7, 2025

    work page 2018

  51. [51]

    Lubis and Pedram Hassanzadeh

    Sandro W. Lubis and Pedram Hassanzadeh. An Eddy–Zonal Flow Feedback Model for Propagating Annular Modes.Journal of the Atmospheric Sciences, 78(1):249–267, January 2021

    work page 2021

  52. [52]

    Samuel Smith, Jian Lu, and Paul W. Staten. Diabatic Eddy Forcing Increases Persistence and Opposes Propagation of the Southern Annular Mode in MERRA-2.Journal of the Atmospheric Sciences, 81(4):743–764, March 2024

    work page 2024

  53. [53]

    Baker, Jochen Bröcker, Chaim I

    Antje Weisheimer, Laura H. Baker, Jochen Bröcker, Chaim I. Garfinkel, Steven C. Hardiman, Dan L. R. Hodson, Tim N. Palmer, Jon I. Robson, Adam A. Scaife, James A. Screen, Theodore G. Shepherd, Doug M. Smith, and Rowan T. Sutton. The Signal-to-Noise Paradox in Climate Forecasts: Revisiting Our Understanding and Identifying Future Priorities.Bulletin of the...

    work page 2024

  54. [54]

    Explainable offline-online training of neural networks for parameterizations: A 1d gravity wave-qbo testbed in the small-data regime

    Hamid A Pahlavan, Pedram Hassanzadeh, and M Joan Alexander. Explainable offline-online training of neural networks for parameterizations: A 1d gravity wave-qbo testbed in the small-data regime. Geophysical Research Letters, 51(2):e2023GL106324, 2024

    work page 2024

  55. [55]

    Hakim and Sanjit Masanam

    Gregory J. Hakim and Sanjit Masanam. Dynamical Tests of a Deep Learning Weather Prediction Model. Artificial Intelligence for the Earth Systems, 3(3), July 2024

    work page 2024

  56. [56]

    Doyle, Carolyn A

    Jorge Baño-Medina, Agniv Sengupta, James D. Doyle, Carolyn A. Reynolds, Duncan Watson-Parris, and Luca Delle Monache. Are AI weather models learning atmospheric physics? A sensitivity analysis of cyclone Xynthia.npj Clim Atmos Sci, 8(1):92, March 2025

    work page 2025

  57. [57]

    Shepherd

    Theodore G. Shepherd. Atmospheric circulation as a source of uncertainty in climate change projections. Nature Geoscience, 7(10):703–708, October 2014

    work page 2014

  58. [58]

    Stouffer

    Syukuro Manabe and Ronald J. Stouffer. Sensitivity of a global climate model to an increase of CO2 concentration in the atmosphere.Journal of Geophysical Research: Oceans, 85(C10):5529–5554, October 1980

    work page 1980

  59. [59]

    Machine Learning for Climate Physics and Simulations.Annual Review of Condensed Matter Physics, 16(Volume 16, 2025):343–365, March 2025

    Ching-Yao Lai, Pedram Hassanzadeh, Aditi Sheshadri, Maike Sonnewald, Raffaele Ferrari, and Venka- tramani Balaji. Machine Learning for Climate Physics and Simulations.Annual Review of Condensed Matter Physics, 16(Volume 16, 2025):343–365, March 2025

    work page 2025

  60. [60]

    Chapman, John S

    William E. Chapman, John S. Schreck, Yingkai Sha, David John Gagne II, Dhamma Kimpara, Laure Zanna, Kirsten J. Mayer, and Judith Berner. CAMulator: Fast Emulation of the Community Atmosphere Model, April 2025

    work page 2025

  61. [61]

    Durran, Zihui Liu, Zachary I

    Nathaniel Cresswell-Clay, Bowen Liu, Dale R. Durran, Zihui Liu, Zachary I. Espinosa, Raul A. Moreno, and Matthias Karlbauer. A Deep Learning Earth System Model for Efficient Simulation of the Observed Climate.AGU Advances, 6(4):e2025AV001706, 2025

    work page 2025

  62. [62]

    V. Balaji. Climbing down Charney’s ladder: Machine learning and the post-Dennard era of computational climate science.Philosophical Transactions of the Royal Society A, April 2021

    work page 2021

  63. [63]

    A perspective on climate model hierarchies.Journal of Advances in Modeling Earth Systems, 9(4):1760–1771, 2017

    Nadir Jeevanjee, Pedram Hassanzadeh, Spencer Hill, and Aditi Sheshadri. A perspective on climate model hierarchies.Journal of Advances in Modeling Earth Systems, 9(4):1760–1771, 2017

    work page 2017

  64. [64]

    The futures of climate modeling.npj Climate and Atmospheric Science, 8(1):99, 2025

    S Bordoni, SM Kang, Tiffany A Shaw, IR Simpson, and L Zanna. The futures of climate modeling.npj Climate and Atmospheric Science, 8(1):99, 2025

    work page 2025

  65. [65]

    Paul A Ullrich, Elizabeth A Barnes, William D Collins, Katherine Dagon, Shiheng Duan, Joshua Elms, Jiwoo Lee, L Ruby Leung, Dan Lu, Maria J Molina, et al. Recommendations for comprehensive and independent evaluation of machine learning-based earth system models.Journal of Geophysical Research: Machine Learning and Computation, 2(1):e2024JH000496, 2025. 16...

    work page 2025