ADTD数据计算雷暴日与目测雷暴日的均一性分析  

Homogeneity Analysis of Artificial Thunderstorm Days and Thunderstorm Days Calculated Using the Data of ADTD System

作　　者：朱华亮[1] 华连生[1] 温华洋[1] 庞文静[2] 梁丽[2] 孔芹芹[1] ZHU Hualiang;HUA Liansheng;WEN Huayang;PANG Wenjing;LIANG Li;KONG Qinqin(Anhui Meteorological Information Center,Hefei 230031,China;Meteorological Observation Centre,China Meteorological Administration,Beijing 100081,China)

机构地区：[1]安徽省气象信息中心,安徽合肥230031 [2]中国气象局气象探测中心,北京100081

出　　处：《干旱气象》2020年第6期1016-1022,共7页Journal of Arid Meteorology

基　　金：安徽高校自然科学研究项目(KJ2019A1289);布达拉宫周边闪电监测网完善和建设项目共同资助。

摘　　要：基于1961—2013年843个气象站的人工观测雷暴日资料和2010—2018年的ADTD系统监测资料,采用数理统计方法,分析全国28个区域的ADTD监测雷电日与人工观测雷暴日之间的关系,建立年雷暴日数拟合公式,并利用PMFT方法探讨年雷暴日数计算值与人工观测值的均一性。结果表明:(1)各地区ADTD监测雷电日和人工观测雷暴日的最优匹配半径不尽相同,大部分地区的最优匹配半径介于9~15 km之间,而新疆、青海和西藏等地区的最优匹配半径大于20 km;(2)绝大部分地区的年雷暴日数计算值与人工观测值的平均绝对误差小于10.0 d,平均相对误差小于30.0%,而新疆、青海和西藏地区的计算值与观测值的误差较大,表明除新疆、青海和西藏地区外,利用二元法建立的公式对年雷暴日数的拟合效果较好;(3)经PMFT检验发现843个站点中有776个站点的序列在2013年前后无明显跳变,占比达92.1%,仅有67个台站因ADTD站点稀疏、ADTD探测性能下降、高山阻隔等原因造成序列明显跳变,表明利用二元法延长的年雷暴日数序列与人工观测序列之间的均一性较好。总体表明,采用ADTD监测数据替代人工观测雷暴日适用于全国大部分地区,新疆、青海和西藏等地区因ADTD站点稀疏而不适用。According to mathematical statistics method,based on the lightning data from 843 weather stations during 1961-2013,and detceted data of the advanced time of arrival and direction system(ADTD)during 2010-2018,the relationship between the lightning data of ADTD and artificial thunderstorm days in 28 regions of China was analyzed,hereafter,the fitting formula was established.The homogeneity between the thunderstorm days calculated by the formula and the artificial thunderstorm days was tested by using the penalized Maximal F test(PMFT).The results are as follows:(1)The optimal matching radius of the thunderstorms calculated from ADTD and the artificial thunderstorm days differed from region to region.It was mainly during 9-15 km in most regions,while that was more than 20 km in Xinjiang,Qinghai and Tibet.(2)It was found that the average absolute error was less than 10.0 days and the average relative error was less than 30.0%between thunderstorm days calculated from the formula and artificial thunderstorm days in most regions,while the errors were much larger in Xinjiang,Qinghai and Tibet.This indicated that the formula established by the binary method performed well on estimating the annual thunderstorm days,except for Xinjiang,Qinghai and Tibet.(3)Among 843 stations,the thunderstorm days in 776 stations(accounting for 92.1%)had no obvious jump before and after 2013 according to PMFT test,only 67 stations had sequence jump probably due to the station sparsity and detection performance degradation of ADTD,high mountain barrier and so on.This indicated that the extended annual thunderstorm days sequence by binary method was in good consistent with the artificial observation sequence.Overall,the thunderstorm days calculated from ADTD could be applied to replace the ones from station observations in most of China,but not for Xinjiang,Qinghai and Tibet due to the sparse ADTD stations.

关 键 词：闪电定位数据 雷暴日 二元法 均一性 

分 类 号：P41[天文地球—大气科学及气象学]