Preprint · 2025 · Open access

Butterfly Effects and Finite Predictability in AI-Based Weather Prediction

Bo‐Wen Shen, Roger A. Pielke, Xubin Zeng

Published 21 July 2025 · DOI 10.22541/essoar.175308863.35440307/v1

Open PDFView at the publisherCite this paper

In brief

“Butterfly Effects and Finite Predictability in AI-Based Weather Prediction” is a 2025 preprint by Bo‐Wen Shen and 2 co-authors. OpenAlex counts 1 citations as of 20 November 2026. It is open access: the full text is free to read.

Citations
1
Most cited in
2026 (1)
References
29
Authors
3

Abstract

Recent advances in AI technology offer unprecedented opportunities for enhancing and extending short-term weather predictions. Yet long-established physics-based models and fundamental principles serve as scientific benchmarks to guide model development. Selz and Craig (2023; https://doi.org/10.1029/2023GL105747) concluded that AI-based weather models cannot simulate the butterfly effect. By reviewing three types of butterfly effect, we revisit the findings of Selz and Craig to enhance our understanding of the current limitations of the selected AI-powered models. These three types of butterfly effect include sensitive dependence on initial conditions, a metaphor for creating a tornado, and error upscale transfer for limiting predictability. While the AI-powered model as well as the physics-based model at comparable coarse grid spacings with default initial conditions failed to simulate the rapid growth for the first 24 hours, both types of models still indicated finite predictability, aligning with the widely accepted definitions of the butterfly effect.

Cite this paper

Formatted from the catalogue record. Check names, capitals and page numbers against the paper before you submit.

APA 7

Shen, B., Pielke, R. A., & Zeng, X. (2025). Butterfly Effects and Finite Predictability in AI-Based Weather Prediction. https://doi.org/10.22541/essoar.175308863.35440307/v1

MLA 9

Shen, Bo‐Wen, et al. “Butterfly Effects and Finite Predictability in AI-Based Weather Prediction.” 2025. https://doi.org/10.22541/essoar.175308863.35440307/v1.

Chicago

Shen, Bo‐Wen, Roger A. Pielke, and Xubin Zeng. 2025. “Butterfly Effects and Finite Predictability in AI-Based Weather Prediction.” https://doi.org/10.22541/essoar.175308863.35440307/v1.

Harvard

Shen, B., Pielke, R. A. and Zeng, X. (2025) ‘Butterfly Effects and Finite Predictability in AI-Based Weather Prediction’. doi:10.22541/essoar.175308863.35440307/v1.

IEEE

B. Shen, R. A. Pielke, and X. Zeng, “Butterfly Effects and Finite Predictability in AI-Based Weather Prediction,” 2025, doi: 10.22541/essoar.175308863.35440307/v1.

BibTeX

@misc{shen2025butterfly,
  title     = {{Butterfly Effects and Finite Predictability in AI-Based Weather Prediction}},
  author    = {Shen, Bo‐Wen and Pielke, Roger A. and Zeng, Xubin},
  year      = {2025},
  doi       = {10.22541/essoar.175308863.35440307/v1},
  url       = {https://doi.org/10.22541/essoar.175308863.35440307/v1}
}

Download for a reference manager: BibTeX (.bib) · RIS for Zotero, Mendeley and EndNote (.ris)

Topic and keywords

Physical Sciences › Earth and Planetary Sciences › Atmospheric Science › Meteorological Phenomena and Simulations

  • butterfly effect
  • sensitivity to initial conditions
  • short-term weather forecasting
  • numerical weather prediction

Work with this paper

Most-cited papers in Earth and Planetary Sciences

All highly cited Earth and Planetary Sciences papers

Questions about this paper

How do I cite this paper in APA style?

Shen, B., Pielke, R. A., & Zeng, X. (2025). Butterfly Effects and Finite Predictability in AI-Based Weather Prediction. https://doi.org/10.22541/essoar.175308863.35440307/v1

How many times has this paper been cited?

1 times, according to OpenAlex on 20 November 2026. Google Scholar and Scopus count different sets of sources, so their numbers differ.

Is this paper free to read?

Yes. It is open access: sign in and open the PDF from this page.

Who wrote this paper?

Bo‐Wen Shen, Roger A. Pielke, Xubin Zeng.

What is the DOI of this paper?

10.22541/essoar.175308863.35440307/v1, permanently at https://doi.org/10.22541/essoar.175308863.35440307/v1. Sign in and open it from this page.

Bibliographic data from OpenAlex (CC0), last checked 20 November 2026. ResearchGuru does not host this paper; once you are signed in, the links open it at the publisher or the repository that does.