AI Insight
Researchers have developed a computationally efficient method using artificial intelligence to directly incorporate satellite radiation data into aerosol simulations, improving their accuracy. The AI component was trained on large datasets to rapidly interpret how satellite observations are affected by both aerosols and solar light reflected from Earth's surface. When tested in China against actual satellite observations, this new approach reduced prediction errors by approximately 50% compared to traditional aerosol modeling methods.
Why it matters
This advancement enables more accurate modeling of atmospheric aerosols, which affect visibility, human health, and climate by reflecting or absorbing sunlight and influencing cloud formation. The improved methodology has direct applications for enhancing weather and climate forecasting accuracy while remaining computationally feasible for operational use.
Understand the Science

Source: Journal of Advances in Modeling Earth Systems
This is an authorized translation of an Eos article. 本文是Eos文章的授权翻译。
气溶胶是悬浮在大气中的微小颗粒。它们的种类繁多,例如野火产生的烟雾、花粉、沙漠尘埃、由强风暴卷入空中的海盐、火山喷发释放的硫酸盐,以及发动机和工业生产过程中的排放物等,都属于气溶胶。
气溶胶不仅影响能见度和人类健康,还在天气和气候预报中发挥着关键作用。它们能够反射或吸收阳光,从而产生加热或冷却效应;同时,它们还能作为水汽凝结的核心,促进云的形成。气溶胶能够在大气各层中长距离传输,甚至跨越大陆和海洋。
然而,气溶胶具有复杂多样的特性和影响,这使得模拟其分布与运动变得十分困难,也给天气和气候建模带来了不确定性。Zhang等人提出了一种新颖且计算效率高的方法,能够将卫星观测到的辐射数据直接纳入气溶胶模拟中,从而提高模拟的准确性。
卫星可通过检测颗粒反射的太阳光来追踪气溶胶,这些观测数据长期以来一直被用于驱动气溶胶模拟。然而,过去若将原始的可见光和红外卫星观测数据直接纳入气溶胶模拟中,由于需要过多的计算资源而难以实现。因此,目前的模拟通常使用经过预处理的卫星数据,这些数据间接反映了气溶胶特性。尽管这种方法效率较高,但预测结果的准确性却相对较低。
借助这种新方法,卫星数据可以更高效地直接整合到气溶胶模拟中。该方法的核心在于一个创新的人工智能(AI)组件。该组件经过大规模数据集预训练,能够快速解析卫星观测数据如何受到气溶胶与地表反射太阳光共同作用的影响。
研究人员在中国对该新方法生成的气溶胶预测结果进行了验证,并将其与实际卫星观测数据进行了对比。结果显示,与传统气溶胶模拟相比,该新方法将预测误差降低了约 50%。这些结果表明,该技术提供了一种计算上可行的方法,不仅能改进气溶胶模拟,还有助于提升天气和气候预报的准确性。(Journal of Advances in Modeling Earth Systems, https://doi.org/10.1029/2026MS005846, 2026)
—科学撰稿人Sarah Stanley

This translation was made by Wiley. 本文翻译由Wiley提供。
Read this article on WeChat. 在微信上分享本文。

Text © 2026. AGU. CC BY-NC-ND 3.0
Except where otherwise noted, images are subject to copyright. Any reuse without express permission from the copyright owner is prohibited.
Source: 聚焦微小颗粒,提升天气和空气质量预报能力