Journal article · 2026 · Open access

Retrieving Tropical Cyclone Intensity from Climate Reanalysis Using Deep Learning

Minh-Khanh Luong, Chanh Kieu

Journal of Climate, 39(23), 6871–6888 · Published 14 October 2026 · DOI 10.1175/jcli-d-25-0649.1

Open PDFView at the publisherCite this paper

In brief

“Retrieving Tropical Cyclone Intensity from Climate Reanalysis Using Deep Learning” is a 2026 journal article by Minh-Khanh Luong and Chanh Kieu, published in Journal of Climate. OpenAlex counts 0 citations as of 20 November 2026. It is open access under a CC BY licence: the full text is free to read at arXiv (Cornell University).

Citations
0
References
65
Authors
2

Abstract

Traditional methods for improving tropical cyclone (TC) intensity from climate model outputs or projections have primarily relied on either dynamical or statistical downscaling. With recent advances in deep learning (DL) techniques, it is of interest to examine how DL can provide an alternative approach for enhancing TC intensity and structure retrieval from climate data. Using a common DL architecture based on convolutional neural networks (CNNs) and a set of key environmental features relevant to TCs, we show that TC intensity and size can be effectively retrieved from climate reanalysis data without requiring superresolution enhancement as in previous studies, even when applied to coarse-resolution climate data. This approach allows for retrieving TC intensity metrics and size that are dynamically constrained by the data, rather than estimating these quantities independently. In this regard, our results highlight that TC intensity and size are governed not only by TC internal processes but also by local environments, which DL models can learn and capture. The performance of our DL model depends on several factors such as season, the stage of TC development, or ocean basins, with root-mean-square errors ranging from ≈8 to 10 m s −1 for the maximum 10-m wind, ≈10–13 hPa for minimum central pressure, and ≈13–21 km for the radius of the maximum wind. Although these errors are better than those of any direct vortex detection or statistical downscaling methods applied to the same data, their wide ranges also suggest that 0.5°-resolution climate data may contain limited TC information for DL models to learn from, regardless of model optimizations or architectures. Possible improvements and challenges in addressing the lack of fine-scale TC information in coarse-resolution climate reanalysis datasets are discussed. Significance Statement This study presents a deep learning (DL)-based approach for retrieving tropical cyclone (TC) intensity and size. Using a set of environmental factors that have the largest influence on TC intensity for DL training, we show that TC intensity and size can be more effectively retrieved from coarse-resolution climate data than conventional vortex detection or statistical downscaling methods. Although the DL performance varies widely depending on data sampling strategies, TC season, and model hyperparameters, these results highlight a promising capability of DL in retrieving TC information without the need for computationally expensive dynamical downscaling. Possible limitations when applying DL to downscale TC intensity from climate data are also discussed.

Cite this paper

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

APA 7

Luong, M.-K., & Kieu, C. (2026). Retrieving Tropical Cyclone Intensity from Climate Reanalysis Using Deep Learning. Journal of Climate, 39(23), 6871–6888. https://doi.org/10.1175/jcli-d-25-0649.1

MLA 9

Luong, Minh-Khanh, and Chanh Kieu. “Retrieving Tropical Cyclone Intensity from Climate Reanalysis Using Deep Learning.” Journal of Climate, vol. 39, no. 23, 2026, pp. 6871–6888. https://doi.org/10.1175/jcli-d-25-0649.1.

Chicago

Luong, Minh-Khanh and Chanh Kieu. 2026. “Retrieving Tropical Cyclone Intensity from Climate Reanalysis Using Deep Learning.” Journal of Climate 39 (23): 6871–6888. https://doi.org/10.1175/jcli-d-25-0649.1.

Harvard

Luong, M.-K. and Kieu, C. (2026) ‘Retrieving Tropical Cyclone Intensity from Climate Reanalysis Using Deep Learning’, Journal of Climate, 39(23), pp. 6871–6888. doi:10.1175/jcli-d-25-0649.1.

IEEE

M.-K. Luong and C. Kieu, “Retrieving Tropical Cyclone Intensity from Climate Reanalysis Using Deep Learning,” Journal of Climate, vol. 39, no. 23, pp. 6871–6888, 2026, doi: 10.1175/jcli-d-25-0649.1.

BibTeX

@article{luong2026retrieving,
  title     = {{Retrieving Tropical Cyclone Intensity from Climate Reanalysis Using Deep Learning}},
  author    = {Luong, Minh-Khanh and Kieu, Chanh},
  journal   = {Journal of Climate},
  year      = {2026},
  volume    = {39},
  number    = {23},
  pages     = {6871--6888},
  doi       = {10.1175/jcli-d-25-0649.1},
  url       = {https://doi.org/10.1175/jcli-d-25-0649.1}
}

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 › Tropical and Extratropical Cyclones Research

  • tropical cyclone intensity
  • tropical cyclone size
  • climate reanalysis
  • deep learning
  • convolutional neural networks
  • dynamical downscaling
  • statistical downscaling
  • vortex identification

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?

Luong, M.-K., & Kieu, C. (2026). Retrieving Tropical Cyclone Intensity from Climate Reanalysis Using Deep Learning. Journal of Climate, 39(23), 6871–6888. https://doi.org/10.1175/jcli-d-25-0649.1

How many times has this paper been cited?

0 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 at arXiv (Cornell University): sign in and open the PDF from this page.

Who wrote this paper?

Minh-Khanh Luong, Chanh Kieu.

What is the DOI of this paper?

10.1175/jcli-d-25-0649.1, permanently at https://doi.org/10.1175/jcli-d-25-0649.1. 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.