Preprint · 2024 · Open access

TAUDiff: Highly efficient kilometer-scale downscaling using generative diffusion models

Rahul Sundar, Hu, Yucong, Nishant Parashar, Antoine B. Blanchard, Boyko Dodov

arXiv (Cornell University) · Published 18 December 2024 · DOI 10.48550/arxiv.2412.13627

Open PDFView at the publisherCite this paper

In brief

“TAUDiff: Highly efficient kilometer-scale downscaling using generative diffusion models” is a 2024 preprint by Rahul Sundar and 4 co-authors, published in arXiv (Cornell University). OpenAlex counts 8 citations as of 21 November 2026. It is open access: the full text is free to read at arXiv (Cornell University).

Citations
8
Most cited in
2026 (7)
References
24
Authors
5

Abstract

Deterministic regression-based downscaling models for climate variables often suffer from spectral bias, which can be mitigated by generative models like diffusion models. To enable efficient and reliable simulation of extreme weather events, it is crucial to achieve rapid turnaround, dynamical consistency, and accurate spatio-temporal spectral recovery. We propose an efficient correction diffusion model, TAUDiff, that combines a deterministic spatio-temporal model for mean field downscaling with a smaller generative diffusion model for recovering the fine-scale stochastic features. We demonstrate the efficacy of this approach on downscaling atmospheric wind velocity fields obtained from coarse GCM simulations. We then extend TAUDiff for computationally efficient kilometer-scale downscaling of atmospheric wind velocity fields. Owing to low inference times, our approach can ensure quicker simulation of extreme events necessary for estimating associated risks and economic losses.

Cite this paper

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

APA 7

Sundar, R., Yucong, H., Parashar, N., Blanchard, A. B., & Dodov, B. (2024). TAUDiff: Highly efficient kilometer-scale downscaling using generative diffusion models. arXiv (Cornell University). https://doi.org/10.48550/arxiv.2412.13627

MLA 9

Sundar, Rahul, et al. “TAUDiff: Highly efficient kilometer-scale downscaling using generative diffusion models.” arXiv (Cornell University), 2024. https://doi.org/10.48550/arxiv.2412.13627.

Chicago

Sundar, Rahul, Hu, Yucong, Nishant Parashar, Antoine B. Blanchard, and Boyko Dodov. 2024. “TAUDiff: Highly efficient kilometer-scale downscaling using generative diffusion models.” arXiv (Cornell University). https://doi.org/10.48550/arxiv.2412.13627.

Harvard

Sundar, R. et al. (2024) ‘TAUDiff: Highly efficient kilometer-scale downscaling using generative diffusion models’, arXiv (Cornell University). doi:10.48550/arxiv.2412.13627.

IEEE

R. Sundar, H. Yucong, N. Parashar, A. B. Blanchard, and B. Dodov, “TAUDiff: Highly efficient kilometer-scale downscaling using generative diffusion models,” arXiv (Cornell University), 2024, doi: 10.48550/arxiv.2412.13627.

BibTeX

@misc{sundar2024taudiff,
  title     = {{TAUDiff: Highly efficient kilometer-scale downscaling using generative diffusion models}},
  author    = {Sundar, Rahul and Yucong, Hu, and Parashar, Nishant and Blanchard, Antoine B. and Dodov, Boyko},
  howpublished = {arXiv (Cornell University)},
  year      = {2024},
  doi       = {10.48550/arxiv.2412.13627},
  url       = {https://doi.org/10.48550/arxiv.2412.13627}
}

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

  • climate downscaling
  • spectral bias
  • extreme weather events
  • generative diffusion models
  • time consistency
  • economic loss estimation

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?

Sundar, R., Yucong, H., Parashar, N., Blanchard, A. B., & Dodov, B. (2024). TAUDiff: Highly efficient kilometer-scale downscaling using generative diffusion models. arXiv (Cornell University). https://doi.org/10.48550/arxiv.2412.13627

How many times has this paper been cited?

8 times, according to OpenAlex on 21 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 at arXiv (Cornell University): sign in and open the PDF from this page.

Who wrote this paper?

Rahul Sundar, Hu, Yucong, Nishant Parashar, Antoine B. Blanchard, Boyko Dodov.

What is the DOI of this paper?

10.48550/arxiv.2412.13627, permanently at https://doi.org/10.48550/arxiv.2412.13627. Sign in and open it from this page.

Bibliographic data from OpenAlex (CC0), last checked 21 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.