基于毫米波雷达的海雾观测及能见度反演算法研究  

Characteristic Analysis of a Sea Fog Process based on Millimeter-wave Radar and Research on Visibility Retrieval Algorithm

作　　者：任欣悦 何建新[2] 张福贵[2] 王皓[1] 胡利军 严一霖 谭淋尹 REN Xinyue;HE Jianxin;ZHANG Fugui;WWANG Hao;HU Lijun;YAN Yilin;TAN Linyin(College of Meteorological Observation,CUIT,Chengdu 610225,China;Key Open Laboratory of Atmospheric Sounding,China Meteorological Administration,Chengdu 610225,China;Ningbo Meteorological Service,Ningbo 315012,China)

机构地区：[1]成都信息工程大学大气探测学院,四川成都610225 [2]中国气象局大气探测重点开放实验室,四川成都610225 [3]宁波市气象局,浙江宁波315012

出　　处：《成都信息工程大学学报》2023年第1期49-56,共8页Journal of Chengdu University of Information Technology

基　　金：四川省科技厅重点研发计划资助项目(2022YFS0541);中国气象局大气探测重点开放实验室资助项目(2021KLAS02Z)。

摘　　要：基于毫米波雷达数据并结合地面气象观测资料对2021年2月6日发生在福建平潭综合实验区的一次典型海雾过程进行多尺度的特征研究。利用机器学习算法,实现了能见度连续空间范围的有效反演。研究结果表明:毫米波雷达能在该次海雾过程的区域分布、生消转换以及垂直演变上给出重要的特征信息;基于机器学习的方法能克服传统拟合方法在非线性问题上存在的局限性,其能见度反演精度明显优于线性关系、经典指数关系,能更好地实现能见度点数据向面数据的拓展,为海雾的监测预警提供了一种更为准确有效的途径。Based on the data of millimeter-wave radar and ground meteorological observations,this paper analyzes the multi-scale characteristic of a typical sea fog event in Pingtan comprehensive experimental area on February 6,2021.On this basis,efficient retrieval of the visibility in continuous spatial scale is realized by using a machine learning algorithm.The specific research results show that:The millimeter-wave radar provides crucial characteristic information about the regional distribution,generation-elimination conversion,and vertical evolution of the sea fog process;The method based on machine learning can solve the problem that the traditional fitting method has limitations in the nonlinear problem,and its accuracy of visibility retrieval is significantly better than linear relationship and classical exponential relationship,which can better upscale visibility data from point to spatial surface data so that we can offer a more accurate and effective way for monitoring and alerting off sea fog.

关 键 词：海雾 毫米波雷达 雷达反射率 能见度 神经网络 

分 类 号：P412.25[天文地球—大气科学及气象学]