{
  "abstract": "Short-term typhoon wind-field prediction across multiple lead times is challenging because forecast error grows with lead time and inappropriate sample construction may leak future storm-position information. We adopt STL-Net, a lightweight spatiotemporal framework integrating convolutional encoding, long short-term memory (LSTM) temporal modeling, squeeze-and-excitation recalibration, and multi-branch fusion for 10 m wind-field prediction at 6, 12, and 24 h lead times. A fixed-t₀ protocol anchors the target patch at the initialization-time storm position, avoiding future best-track leakage. Trained in 2020–2021, validated in 2022, and tested in 2023, STL-Net achieves a root mean square error (RMSE) of 3.11, 4.06, and 5.31 m·s⁻¹ at 6, 12, and 24 h. Among deep-learning baselines, U-Net achieves the lowest errors at 6 and 12 h and convolutional neural network (CNN) the lowest 24 h RMSE and mean absolute error (MAE), while STL-Net remains competitive with 1.673 million parameters and low inference latency. Ten-seed ablation identifies multi-resolution fusion as the most consistently beneficial component. An out-of-year evaluation with 2020 as the test year reproduces the error-growth pattern, supporting cross-year robustness. Models are deterministic without uncertainty quantification; error growth is deterministic, not calibrated uncertainty. Diagnostic interpretation links the error structure to the multi-scale organization of typhoon winds, storm translation, and wind-field asymmetry. Results constitute an offline proof-of-concept, not an operational system.",
  "abstract_shown": true,
  "affiliations": [
    "School of Computer, Electronics and Information, Guangxi University, Nanning 530004, China",
    "China Meteorology Administration Earth System Modeling and Prediction Centre, Beijing 100081, China"
  ],
  "article_type": null,
  "arxiv": null,
  "author_total": 3,
  "authors": [
    {
      "affs": [
        0
      ],
      "name": "Jun Liu"
    },
    {
      "affs": [
        0
      ],
      "name": "Jie Cui"
    },
    {
      "affs": [
        1
      ],
      "name": "Yan Liu"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/8pmtqhnbn3xvc.bib",
  "cited_by": 0,
  "doi": "10.3390/atmos17100950",
  "fulltext_url": "https://www.mdpi.com/2073-4433/17/10/950",
  "id": "8pmtqhnbn3xvc",
  "issns": [
    "2073-4433"
  ],
  "landing_url": "https://doi.org/10.3390/atmos17100950",
  "language": "en",
  "licence": "CC-BY",
  "licence_label": "CC BY",
  "licence_source": "journal",
  "licence_url": "https://creativecommons.org/licenses/by/4.0/",
  "metadata_licence": null,
  "native_subjects": [
    [
      "lcc",
      "QC851-999"
    ],
    [
      "openalex",
      "1902|Tropical and Extratropical Cyclones Research"
    ]
  ],
  "openalex": "W7214704182",
  "pdf_url": null,
  "pmcid": null,
  "pmid": null,
  "published": "2026-09-01",
  "publisher": "MDPI AG",
  "ris": "https://research.draconitic.com/paper/8pmtqhnbn3xvc.ris",
  "sources": [
    {
      "first_seen": "2026-10-31T14:21:21Z",
      "native_id": "00dd8f4ae58c40ac9f64f2335dfee20d",
      "source": "doaj",
      "url": "https://doi.org/10.3390/atmos17100950"
    }
  ],
  "subjects": [
    {
      "field": "earth",
      "origin": "openalex:1902|Tropical and Extratropical Cyclones Research",
      "subfield": "atmospheric"
    }
  ],
  "title": "Multi-Horizon Typhoon Wind Field Prediction via a Lightweight CNN–LSTM Network: Error Growth and Cross-Year Robustness at 6–24 h Lead Times",
  "title_text": "Multi-Horizon Typhoon Wind Field Prediction via a Lightweight CNN–LSTM Network: Error Growth and Cross-Year Robustness at 6–24 h Lead Times",
  "type": "peer-reviewed",
  "updated_at": "2026-10-31T14:21:21Z",
  "url": "https://research.draconitic.com/paper/8pmtqhnbn3xvc",
  "venue": "Atmosphere",
  "versions": []
}