摘要
首先利用后向轨迹模式(HYSPLIT)模拟了天水市2017~2019年冬季后向轨迹,分析了50,200,500和1000m 4个不同高度以及500m高度不同聚类数量对路径输送聚类统计结果的影响,并以500m高度四个季节的后向轨迹进行聚类分析,进一步运用权重潜在源贡献分析法(WPSCF),探讨了研究期间天水市细颗粒物的潜在源区及不同源区对天水市颗粒物浓度的贡献,结果表明,(1)起始点高度为500m时,颗粒物浓度极值比和极值差均较大,聚类结果最具代表性;(2)不同聚类数量分析结果表明,按照总空间变化(Total spatial variation,TSV)显著增加的原则选取的聚类数量较少,且能反映不同方向的轨迹输送特征;(3)天水市不同季节的轨迹聚类结果表明,冬季来自陕西南部的东南路径是PM_(2.5)污染程度最高的路径,该路径下PM_(2.5)浓度为78.2μg/m^(3),春季西北路径的颗粒物浓度最高,PM_(10)和PM_(2.5)的平均浓度分别是127.9~129.9和40.6~41.0μg/m^(3),夏秋季节不同路径的颗粒物浓度相差不大.
Backward trajectories from urban Tianshui was simulated using the HYSPLIT4 model for winters of 2017~2019.The influence of heights,i.e.50,200,500 and 1000 m,and the number of clusters at 500 m on the statistical results of transport pathway clustering was analyzed.The 500 m backward trajectories in spring,summer,autumn and winter were clustered,and the weighted potential source contribution analysis(WPSCF)was used to explore the potential source areas of fine particles.The results indicate that the clustered pathways starting at 500 m was the most representative for the study area.The number of clusters selected based on the principle of significant increase of TSV is small and could reflect the trajectory transport characteristics in different directions.In winter,the southeast path from the south of Shaanxi Province had the highest PM_(2.5) concentration,reaching 78.2μg/m^(3).In spring,the highest particulate concentration is found in the northwest path,and the average concentrations of PM_(10) and PM_(2.5) were 127.9~129.9μg/m^(3) and 40.6~41.0μg/m^(3),respectively.There was no significant difference between the particulate concentrations in different paths in summer and autumn.
作者
刘灏
王颖
王思潼
刘扬
李博
LIU Hao;WANG Ying;WANG Si-tong;LIU Yang;LI Bo(College of atmospheric science,Lanzhou University,Lanzhou 730000,China;Key Laboratory of semi-arid climate change,Ministry of Education,Lanzhou 730000,China)
出处
《中国环境科学》
EI
CAS
CSCD
北大核心
2021年第8期3529-3538,共10页
China Environmental Science
基金
甘肃省科技计划项目(18JR2RA005)。
关键词
HYSPLIT4模式
后向轨迹
聚类原则
潜在源贡献分析
HYSPLIT4 model
backward trajectory
clustering principle
potential source contribution analysis
