CoChat
Sign Up
Cited by 41Meteorological Phenomena and SimulationsClimate variability and modelsHydrological Forecasting Using AI

Global extreme heat forecasting using neural weather models

Open access
Lopez-Gomez, Ignacio, McGovern, Amy, Agrawal, Shreya, Hickey, Jason·2023·Artificial Intelligence for the Earth Systems
DOI 10.1175/aies-d-22-0035.1
Open this paper in CoChat — free accountOpen PDF

Abstract

Heatwaves are projected to increase in frequency and severity with global warming. Improved warning systems would help reduce the associated loss of lives, wildfires, power disruptions, and reduction in crop yields. In this work, we explore the potential for deep learning systems trained on historical data to forecast extreme heat on short, medium and subseasonal time scales. To this purpose, we train a set of neural weather models (NWMs) with convolutional architectures to forecast surface temperature anomalies globally, 1 to 28 days ahead, at ∼200-km resolution and on the cubed sphere. The NWMs are trained using the ERA5 reanalysis product and a set of candidate loss functions, including the mean-square error and exponential losses targeting extremes. We find that training models to minimize custom losses tailored to emphasize extremes leads to significant skill improvements in the heatwave prediction task, relative to NWMs trained on the mean-square-error loss. This improvement is accomplished with almost no skill reduction in the general temperature prediction task, and it can be efficiently realized through transfer learning, by retraining NWMs with the custom losses for a few epochs. In addition, we find that the use of a symmetric exponential loss reduces the smoothing of NWM forecasts with lead time. Our best NWM is able to outperform persistence in a regressive sense for all lead times and temperature anomaly thresholds considered, and shows positive regressive skill relative to the ECMWF subseasonal-to-seasonal control forecast after 2 weeks. Significance Statement: Heatwaves are projected to become stronger and more frequent as a result of global warming. Accurate forecasting of these events would enable the implementation of effective mitigation strategies. Here we analyze the forecast accuracy of artificial intelligence systems trained on historical surface temperature data to predict extreme heat events globally, 1 to 28 days ahead. We find that artificial intelligence systems trained to focus on extreme temperatures are significantly more accurate at predicting heatwaves than systems trained to minimize errors in surface temperatures and remain equally skillful at predicting moderate temperatures. Furthermore, the extreme-focused systems compete with state-of-the-art physics-based forecast systems in the subseasonal range, while incurring a much lower computational cost.

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1175/aies-d-22-0035.1

Cites

82 of 82 references verified0 citation errors · 0 not checkable

Statistics

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

  • Lopez-Gomez et al. (2023) find that their NN architecture performs better than ECMWF's forecast system for extremely hot days beyond the 2-week forecast time.

    “Lopez-Gomez et al. (2023), instead, use all the available information without explicit undersampling, finding that their NN architecture performs better than ECMWF's forecast system for extremely hot days beyond the 2-week forecast time.”

    From Artificial intelligence for climate prediction of extremes: State of the art, challenges, and future perspectives (2024) · cites this paper as Lopez-Gomez et al., 2023

    The Data, predictors and targets section says the models use all available information without explicit undersampling. Forecast skill for summer over land reports higher regressive skill than ECMWF after two weeks, including for extremely hot days, but says the neural models do not outperform ECMWF as classifiers of extremely hot days. The unqualified claim that they perform better omits this distinction. (LLM, conf=0.92)

    Experimental — accuracy not yet measured

  • Applications of XAI in extreme heat prediction are limited.

    “Limited work on explainable AI: Even though XAI is increasingly being used to improve model interpretability, its applications in extreme heat prediction are limited [21–23, 25, 33, 34] .”

    From Extreme heat prediction through deep learning and explainable AI (2025) · cites this paper as [25]

    Section 5, “Model interpretation” (page unavailable), applies integrated gradients to interpret an extreme-heat forecasting model. It does not assess how widespread XAI applications are in extreme-heat prediction, so it cannot establish the claim that such applications are limited. Searches for explainability, interpretability, and feature attribution found no statement of that limitation. (LLM, conf=0.72)

    Experimental — accuracy not yet measured

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Cites

Flagged first, then by citations

Research works printed but not linked: 5

  1. Li, X., Y. Grandvalet, and F. Davoine, 2018: Explicit inductive bias for transfer learning with convolutional networks. Proceedings of the 35th International Conference on Machine Learning , J. Dy, and A. Krause, Eds., PMLR, Proceedings of Machine Learning Research, Vol. 80, 28252834, URL https://proceedings.mlr.press/v80/li18a.html.
  2. Rezende, D., and S. Mohamed, 2015: Variational inference with normalizing flows. Proceedings of the 32nd International Conference on Machine Learning , F. Bach, and D. Blei, Eds., PMLR, Lille, France, Proceedings of Machine Learning Research, Vol. 37, 1530-1538, URL https: //proceedings.mlr.press/v37/rezende15.html.
  3. Sundararajan, M., and S. Agrawal, 2021: The rain check. 4th Workshop on Visualization for AI Explainability , URL http://raincheck.karyk.com/rain-check.
  4. Yosinski, J., J. Clune, Y. Bengio, and H. Lipson, 2014: How transferable are features in deep neural networks? Advances in Neural Information Processing Systems , Z. Ghahramani, M. Welling, C. Cortes, N. Lawrence, and K. Weinberger, Eds., Curran Associates, Inc., Vol. 27, URL https:// proceedings.neurips.cc/paper/2014/hash/375c71349b295fbe2dcdca9206f20a06-Abstract.html.
  5. Zhu, Y., Z. Toth, R. Wobus, D. Richardson, and K. Mylne, 2002: The economic value of ensemble-based weather forecasts. Bulletin of the American Meteorological Society , 83 , 7384, https://doi.org/10.1175/1520-0477(2002)083<0073:TEVOEB>2.3.CO;2. · DOI

Cited by

Sorted by citations

Related

Authors
Lopez-Gomez, I., McGovern, A., Agrawal, S., & Hickey, J.
Affiliations
  • a Google Research, Mountain View, California
  • Google (United States) (US)
ORCID
Venue
Artificial Intelligence for the Earth Systems
Publisher
American Meteorological Society
Published
2023-01
Volume
2
Issue
1
ISSN
2769-7525 (ISSN-L)
DOI
10.1175/aies-d-22-0035.1
OpenAlex
W4313407333
Cited by
41
References
78
Citations by year
2026: 7 · 2025: 14 · 2024: 12 · 2023: 6 · 2022: 2
FWCI
3.25
Type
journal article
Language
en
Open access
Licence
www.ametsoc.org/PUBSReuseLicenses (unspecified)
Locations
  • Artificial Intelligence for the Earth Systems (publishedVersion)
  • arXiv (Cornell University) (submittedVersion)
  • arXiv (Cornell University)
Topics
  • Physical Sciences › Earth and Planetary Sciences › Atmospheric Science › Meteorological Phenomena and Simulations
  • Physical Sciences › Environmental Science › Global and Planetary Change › Climate variability and models
  • Physical Sciences › Environmental Science › Environmental Engineering › Hydrological Forecasting Using AI
Keywords
heatwave prediction, extreme heat, subseasonal forecasting, surface temperature anomalies, exponential loss, transfer learning, cubed sphere, ERA5 reanalysis, forecast skill

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review