{
  "id": 19314984,
  "title": "Anyone can forecast the weather. But Paulo Orenstein gets it right more often.",
  "url": "https://urgent.news/2026/11/10/qualquer-um-faz-previsao-do-tempo-mas-paulo-orenstein-acerta-mais",
  "topic": "business",
  "section": "Business",
  "published": "2026-11-10T09:50:00.000Z",
  "source": {
    "name": "Brazil Journal",
    "slug": "brazil-journal",
    "url": "https://braziljournal.com/qualquer-um-faz-previsao-do-tempo-mas-paulo-orenstein-acerta-mais"
  },
  "original_language": "pt",
  "account": "Paulo Orenstein, a 37-year-old mathematician from Brazil and professor at the Instituto de Matemática Pura e Aplicada (IMPA), has co-authored a study published in the journal Nature that presents a method to substantially increase the accuracy of weather forecasts. The method, called Probabilistic Bias Correction (PBC), uses artificial intelligence to correct errors in traditional weather forecasting models and can be applied to both traditional and AI-based models. According to the study, PBC nearly doubled the predictive ability of the European Centre for Medium-Range Weather Forecasts (ECMWF) model's predictions for the next two to six weeks. The method has been integrated into Microsoft's climate platform and can be used by weather centers, companies, and researchers to improve the accuracy of their forecasts.",
  "summary": "Predicting the weather was never simple. Now, with climate change scrambling historical patterns and making the weather more unstable, the challenge has become even greater. A Brazilian is one of the creators of an artificial intelligence that, when applied to traditional weather forecasting models, substantially increases the accuracy of forecasts.",
  "key_points": [
    "Paulo Orenstein, IMPA mathematician, developed Probabilistic Bias Correction method",
    "PBC doubles AI ECMWF system's predictive ability for temperature, precipitation, pressure",
    "Technology integrated into Microsoft's climate platform for agriculture, water management"
  ],
  "editors_take": "The application of systematic error correction to existing artificial intelligence models offers a vital tool for industries that depend on long-term accuracy to manage environmental risks and resources.",
  "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."
}