{
  "id": 227280,
  "title": "Operational Tropical Cyclone Forecasting with AI",
  "url": "https://urgent.news/2026/08/06/operational-tropical-cyclone-forecasting-with-ai",
  "topic": "ai",
  "section": "AI",
  "published": "2026-08-06T00:00:00.000Z",
  "source": {
    "name": "Nature",
    "slug": "nature",
    "url": "https://www.nature.com/articles/s41586-026-10953-2"
  },
  "original_language": "en",
  "account": "Tropical cyclones pose significant danger and economic impact, yet accurately predicting their path, strength, and size remains a formidable scientific challenge. Researchers have now created WeatherNext Cyclones (WN-C), an artificial intelligence-powered weather model capable of producing cutting-edge ensemble forecasts for tropical cyclones across the globe. The model was trained on a blend of global analysis data and a vast database of historical cyclone events, enabling it to generate an extensive array of potential global weather and cyclone scenarios for up to 15 days ahead. When tested on cyclones from 2023 to 2025, WN-C's predictions for track, intensity, and wind radius were found to provide an average of one or more additional days of lead time compared to current operational forecasting models. This improvement in accuracy is on par with the advancements made over the past decade in operational model development. Importantly, the researchers discovered that using inputs coarser than those used in regional models still yields state-of-the-art intensity forecasting results, suggesting that finer resolution is not a mandatory requirement for achieving the best results. Incorporating WN-C's predictions into a weighted-average consensus ensemble enhances the overall skill of the forecast. The model's scalability allows for the creation of ensembles with up to 1,000 members, which is significantly higher than the conventional 50-member ensembles. By offering state-of-the-art operational ensemble guidance to human forecasters, this groundbreaking work marks a substantial leap towards more reliable, timely forecasts and warnings, ultimately helping to safeguard lives and minimize the catastrophic effects of tropical cyclones.",
  "summary": "Nature, Published online: 06 August 2026; doi:10.1038/s41586-026-10953-2 Operational Tropical Cyclone Forecasting with AI",
  "key_points": [],
  "editors_take": null,
  "illustration": null,
  "coverage": {
    "outlets": 1,
    "also_reported_by": []
  },
  "ai_generated": true,
  "disclaimer": "Summaries, key points and the editor’s take are written by software from other outlets’ reporting and may contain errors — always check the linked original."
}