北京地区花粉通量演变曲线的预报模式初探  

Preliminary study on predicting model of airborne pollen flux in Beijing

作　　者：张德山[1] 徐景先 张姝丽[1] 冯涛[1] 宛公展[3] 

机构地区：[1]北京市气象局科技服务中心,北京100089 [2]北京自然博物馆,北京100050 [3]天津市气象科学研究所,天津300074

出　　处：《气象科学》2010年第6期822-826,共5页Journal of the Meteorological Sciences

基　　金：中国气象局气象新技术推广项目(CMATG2008M29)

摘　　要：建模预测人居环境未来1~4 d花粉通量的变化,是城市环境气象服务的重要内容之一。选取北京地区7个花粉监测站1999—2007年3—9月逐日的大气底层花粉通量与北京市观象台的气象要素,建立了北京地区花粉通量未来1~4 d的预报模式,将其未来1~4 d的花粉预报数值看成连续域上4个等间隔节点组成的一条未知曲线。按Taylor级数展开,提取展开系数向量A（a0,a1,...,as-1）。然后用地面气象要素X（x1,x2,..,xm）与预报对象A建模预报,最后再还原为4个节点的逐日预报值。结果表明,未来1、2、…、4 d的预报值与同时刻原始值的单相关系数,分别平均可达0.765 1、0.753 9、0.704 4、0.728 5（通过信度0.01的相关检验的下限阈值0.128）。预报曲线与原始曲线的相似系数,平均可达0.940 7,随机预留样本检验效果非常理想。Based on airborne pollen data at 7 stations of Beijing during 9 a from 1 Mar.to 30 Sep.,1999—2007,the predicting model was developed using Taylor＇s series expansion theory in this paper,in order to forecast the airborne pollen concentration.Firstly,airborne pollen flux was expanded according to the Taylor＇s series expansion theory and the A（a1,a2,...,as-1） vector of expanding coefficients was calculated.Then the predicting model of vector A and meteorological elements X have been found.The results showed that there was a strong correlation between predicting values and observed values.The average coefficients between prediction for the coming 1—4 d and initial airborne pollen concentrations were 0.765 1,0.753 9,0.704 4,and 0.728 5 respectively.The similarity coefficient between prediction and original curves amounted to 0.9407.Analysis indicated that this predicting model of the airborne pollen concentration could be used in the meteorological stations and the departments for disease control and prevention.

关 键 词：花粉通量 气象要素 相似系数 预报模式 

分 类 号：X12[环境科学与工程—环境科学]