中国农业气象 ›› 2025, Vol. 46 ›› Issue (8): 1095-1110.doi: 10.3969/j.issn.1000-6362.2025.08.003

• 农业生态环境栏目 • 上一篇    下一篇

气传致敏花粉监测、预报与服务研究进展

刘素芹,李建强,程文秀,赵琳娜,徐曦,叶彩华   

  1. 1.北京工业大学,北京100124;2.北京市气象服务中心,北京100089
  • 收稿日期:2024-07-08 出版日期:2025-08-20 发布日期:2025-08-19
  • 作者简介:刘素芹,E-mail:liusq_s@163.com
  • 基金资助:
    北京市科技计划课题(Z191100009119013)

Progress on Monitoring, Forecasting and Service of Airborne Allergenic Pollen

LIU Su-qin, LI Jian-qiang, CHENG Wen-xiu, ZHAO Lin-na, XU Xi, YE Cai-hua   

  1. 1. Beijing University of Technology, Beijing 100124, China; 2. Beijing Meteorological Service Center, Beijing 100089
  • Received:2024-07-08 Online:2025-08-20 Published:2025-08-19

摘要:

气传致敏花粉引发的过敏风险防治已成为城市绿化进程中保障公共健康的焦点问题,气传致敏防治体系自下而上涵盖花粉监测、花粉预报和花粉服务三大环节。为深入理解该体系,本研究通过文献检索,分析花粉监测原理、设备及站点布局;基于自统计回归、机器学习再到深度学习的技术更新视角,归纳花粉预报发展与脉络;总结市场上用户友好型花粉服务产品的表现形式和应用现况;探讨体系内各环节面临的挑战及未来研究方向。结果表明:国内花粉监测设备以重力沉降法为主,操作简单且成本低,但较为依赖人工每日监测,未来应注重研发低成本的自动监测设备并提升自动化覆盖率。花粉预报仍以统计回归为主,未来应结合人工智能大模型等前沿技术,构建多模态因子协同的预报模型,开展精细化预报。花粉服务已建立的微信小程序、移动应用、平台等产品,能提供花粉总浓度、分类浓度及医疗指南等多样化信息,未来可深入聚焦不同用户群体的需求,探索个性化和定制化服务。

关键词: 城市绿化, 气传致敏花粉, 花粉监测, 花粉预报, 花粉服务

Abstract:

The prevention and management of allergy risks caused by airborne allergenic pollen has become a critical concern in safeguarding public health during urban greening. The prevention and management framework covers three main strands from bottom to top: pollen monitoring, pollen forecasting and pollen service. To gain a deeper understanding of this framework, a comprehensive literature review was conducted in this paper. The principles, equipment and station layout of pollen monitoring were analyzed. The development of pollen forecasting methods, spanning from statistical regression to machine learning and deep learning was summarized. Current manifestations and applications of user−friendly pollen service products were summarized. In addition, the challenges faced by each link were discussed, and future research directions were prospected. The results indicated that domestic pollen monitoring equipment primarily relied on gravity settling, which was cost−effective and easy to operate but heavily depended on manual daily monitoring. Pollen forecasting was still mainly based on statistical regression, and future advancements should focus on integrating cutting−edge technologies, such as Artificial Intelligence Large Models, to develop multi−modal factor−driven forecasting methods and support more refined forecasting. Pollen service was launched via WeChat mini−programs, mobile applications, platforms and other products that provide diverse information, including total and classified pollen concentration and medical guidelines. Future developments should prioritize addressing the specific needs of different user groups through personalized and customized solutions.

Key words: Urban greening, Airborne allergenic pollen, Pollen monitoring, Pollen forecasting, Pollen service