{
  "id": 18316065,
  "title": "聚焦微小颗粒，提升天气和空气质量预报能力",
  "url": "https://urgent.news/2026/11/05/story-18316065",
  "topic": "science",
  "section": "Science",
  "published": "2026-11-05T13:42:34.000Z",
  "source": {
    "name": "Eos",
    "slug": "eos",
    "url": "https://eos.org/research-spotlights/scientists-zero-in-on-tiny-particles-to-improve-weather-and-air-quality-forecasts-chinese"
  },
  "original_language": "zh",
  "account": "Tiny particles in the atmosphere, known as aerosols, play a crucial role in weather and climate forecasting. These particles, originating from various sources like wildfires, pollen, desert dust, sea salt from storms, volcanic emissions, and emissions from engines and industries, can affect visibility, human health, and are key factors in weather and climate modeling.\n\nAerosols can reflect or absorb sunlight, leading to heating or cooling effects; they also act as nuclei for water vapor to condense around, aiding cloud formation. They can travel long distances across continents and oceans within the atmosphere. However, the diverse nature and complex influence of aerosols make their distribution and movement difficult to simulate, introducing uncertainties into weather and climate models.\n\nZhang et al. introduced a novel and computationally efficient method that integrates satellite radiation observations directly into aerosol simulations, improving their accuracy. Satellites detect aerosols by measuring the sunlight they reflect, a practice that has long been used to drive aerosol simulations. Yet, incorporating raw visible and infrared satellite data directly into aerosol simulations has been challenging due to high computational demands. Hence, current simulations typically use pre-processed satellite data, which indirectly reflects aerosol properties. Although this method is efficient, the accuracy of prediction results tends to be lower.\n\nThis new method, however, allows for more efficient direct integration of satellite data into aerosol simulations. The method's cornerstone is an innovative artificial intelligence (AI) component. Trained on large datasets, this component swiftly parses satellite observations to determine how aerosols and the Earth's surface jointly influence how sunlight is reflected. Researchers in China validated the new method's aerosol predictions, comparing them to actual satellite observations. The results showed that this new approach reduces prediction errors by about 50% compared to traditional aerosol simulations. These findings indicate that this technology offers a computationally feasible method to enhance aerosol simulations, thereby improving the accuracy of weather and climate forecasts.",
  "summary": "一种处理卫星气溶胶观测数据的新方法提高了模拟精度，有望为天气和气候预报带来助益。",
  "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."
}