安徽亳州地区大雾预报模型研究
|
摘要
该文根据2002-2012年亳州地区大雾资料,统计了大雾的时空分布特征和环流形势。结果表明:年平均大雾日数最多的地区集中在亳州中北部,东南部最少。近11a来大雾日数总体上呈减少趋势,但2010年后雾日有所增加,其中水平能见度在50~200m的浓雾出现频次最多。雾日有明显的季节变化,冬季最多、秋季次之、夏季最少;大雾持续时间大多在1~14h。根据2013年亳州市探空观测、地面观测资料以及EC细网格、T639、日本细网格、GRAPES数值模式产品,选取了87种预报因子,使用支持向量机方法(SVM)和中国气象局CMSVM应用软件平台,通过对训练样本进行交叉验证和模型核参数的逐渐逼近,建立了亳州市24h大雾支持向量机预报模型。
In this paper,the temporal and spatial characteristics of fog and the atmospheric circulation were analyzedby using the fog data in Bozhou from year 2002 to 2012. The results showed that,the most of the annual average fogareas were concentrated in the north and central of Bozhou,and the southeast area was the least. Generally,the fogdays decreased in the past 11 years,but it increased after 2010 year. The frequency of the visibility from 50 to 200 meters had the maximum occurrence. The fog days had significant seasonal variation characteristics,that the wintermaximum,autumn second,and summer minimum. The fog mostly lasted 1 to 14 hours. According to the radiosondedata,observation data and EC fine mesh,T639,Japan fine mesh,GRAPES numerical model products of Bozhou Cityin year 2013,87 predictors were selected. By using Support Vector Machines(SVM)and CMSVM application soft-ware platform,and through the cross-validation of the training samples and gradually approaching model nuclear pa-rameters,the 24 hour forecasting model of fog by Support Vector Machine in Bozhou was established.
In this paper,the temporal and spatial characteristics of fog and the atmospheric circulation were analyzedby using the fog data in Bozhou from year 2002 to 2012. The results showed that,the most of the annual average fogareas were concentrated in the north and central of Bozhou,and the southeast area was the least. Generally,the fogdays decreased in the past 11 years,but it increased after 2010 year. The frequency of the visibility from 50 to 200 meters had the maximum occurrence. The fog days had significant seasonal variation characteristics,that the wintermaximum,autumn second,and summer minimum. The fog mostly lasted 1 to 14 hours. According to the radiosondedata,observation data and EC fine mesh,T639,Japan fine mesh,GRAPES numerical model products of Bozhou Cityin year 2013,87 predictors were selected. By using Support Vector Machines(SVM)and CMSVM application soft-ware platform,and through the cross-validation of the training samples and gradually approaching model nuclear pa-rameters,the 24 hour forecasting model of fog by Support Vector Machine in Bozhou was established.
引文
[1]吴彬贵,张宏升,汪靖,等.一次持续性浓雾天气过程的水汽输送及逆温特征分析[J].高原气象,2009,28(2)28:258-267.
[2]李江波,赵玉广,孔凡超,等.华北平原连续性大雾的特征分析[J].中国海洋大学学报,2010,40(7):15-23.
[3]陈晓红,方翀.安徽省县级大雾预报业务系统[J].气象,2005,31(4):61-64.
[4]赵玉广,李江波,康锡言.用PP方法做河北省雾的分县预报[J].气象,2004,30(6):43-47.
[5]程戴晖,杨美川.上海浦西地区雾持续时间的统计释用预报[J].气象,2001,27(7):16-20.
[6]林建,杨贵名,毛冬艳.我国大雾的时空分布特征及其发生的环流形势[J].气候与环境研究,2008,13(2):171-181.
[2]李江波,赵玉广,孔凡超,等.华北平原连续性大雾的特征分析[J].中国海洋大学学报,2010,40(7):15-23.
[3]陈晓红,方翀.安徽省县级大雾预报业务系统[J].气象,2005,31(4):61-64.
[4]赵玉广,李江波,康锡言.用PP方法做河北省雾的分县预报[J].气象,2004,30(6):43-47.
[5]程戴晖,杨美川.上海浦西地区雾持续时间的统计释用预报[J].气象,2001,27(7):16-20.
[6]林建,杨贵名,毛冬艳.我国大雾的时空分布特征及其发生的环流形势[J].气候与环境研究,2008,13(2):171-181.
目录
© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号 地址：北京市海淀区学院路29号 邮编：100083 电话：办公室：(+86 10)66554848；文献借阅、咨询服务、科技查新：66554700 |