| 太原大气能见度影响因子分析及能见度预报 |
| |
| 引用本文： | 卢盛栋,陈立瑾,赵桂香,张泽秀,段鹏飞.太原大气能见度影响因子分析及能见度预报[J].新疆气象,2020,14(4):105-112. |
| |
| 作者姓名： | 卢盛栋 陈立瑾 赵桂香 张泽秀 段鹏飞 |
| |
| 作者单位： | 山西省气象灾害防御技术中心,太原理工大学,山西省气象台;山西省气象台,太原市气象局,太原市尖草坪区气象局 |
| |
| 摘 要： | 利用2017年1月—2019年12月太原地区逐时气象资料,分析了能见度及其主要影响因子的变化特征,并对两次低能见度过程进行深入分析,构建了能见度预报模型并进行检验,结果表明:(1)从空间分布看,太原北部能见度明显高于南部地区。从时间分布看,太原地区平均能见度最大值出现在5月,最小值出现在1月;日间最低值出现在06:00(北京时,下同),冬季略向后推移,最高值出现在15:00前后。(2)2017—2019年太原地区低能见度分别出现93、84、79 d;低能见度发生时,干霾、湿霾发生频率分别为59.27%、40.73%;湿霾发生时,能见度降低更加明显。(3)所选个例中,能见度均随各影响因子有所起伏,干霾、湿霾过程中能见度分别与颗粒物浓度、相对湿度变化一致。(4)采用神经网络方法构建太原地区能见度预报模型,预报模型相关系数为0.81,均方根为4.43 km,平均绝对误差为17.39%,轻微级能见度的TS评分为87%。神经网络方法对太原地区能见度预报具有较高的参考价值。
|
| 关 键 词： | 太原 能见度 气象要素 典型霾过程 神经网络 |
| 收稿时间： | 2020/3/12 0:00:00 |
| 修稿时间： | 2020/4/27 0:00:00 |
Analysis of Main Influencing Factors of Visibilityin Taiyuan and the Visibility Prediction |
| |
| lushengdong,Chen Lijin,Zhao Guixiang,Zhang Zexiu and Duan Pengfei.Analysis of Main Influencing Factors of Visibilityin Taiyuan and the Visibility Prediction[J].Bimonthly of Xinjiang Meteorology,2020,14(4):105-112. |
| |
| Authors: | lushengdong Chen Lijin Zhao Guixiang Zhang Zexiu and Duan Pengfei |
| |
| Institution: | shanxi Meteorological Disaster Prevention Technology Centre,Taiyuan University of Technology,Shanxi Meteorological Observatory,Taiyuan Meteorological Bureau,Jiancaoping District Meteorological Bureau of Taiyuan |
| |
| Abstract: | Hourly meteorological data from January 2017 to December 2019 in Taiyuan has applied to analyze characteristics of visibility and the main influencing factors , and two typical low visibility events are picked for in-depth analysis. The visibility forecast model is established and tested. The results showed that: (i) From the spatial distribution, the visibility in the north area is significantly higher than that in the south. Of time distribution, the maximum average visibility in Taiyuan area appeared in May and the minimum in January. The lowest value in the day appeared at 06:00, and the highest value appeared at 15:00 or so in the winter. (ii) From 2017 to 2019, poor visibility in Taiyuan area occurred for 93, 84 and 79 days, respectively; The probability of dry and wet haze is 59.27% and 40.73%, respectively, when low visibility events happens. When wet haze occurs, visibility deteriorates to reduce more obviously. In addition, it should be noted that, when wet haze event occurs, visibility in Taiyuan deteriorates more obviously than that in the dry haze event. (iii) From the selected two cases, the visibility fluctuates with the influence factors; In the process of dry haze and wet haze, visibility is more consistent with particle concentration and relative humidity, respectively. (iv) The visibility prediction model for Taiyuan region was constructed by using the neural network method. The correlation coefficient of the prediction model was 0.81, the mean square root was 3.43 km, the average absolute error was 17.39%, and the TS score of slight visibility was 87%. The neural network method is valuable for the visibility prediction in Taiyuan area. |
| |
| Keywords: | Air quality Meteorological factor Stepwise regression Neural network |
| 本文献已被 CNKI 等数据库收录！ |
| 点击此处可从《新疆气象》浏览原始摘要信息 |
| 点击此处可从《新疆气象》下载免费的PDF全文 |
|
