2019年江苏省PM_(2.5)和O_(3)多模式集合预报算法效果评估  

Evaluation of the Multi-model Ensemble Forecasting Algorithm for PM_(2.5) and O_(3) in Jiangsu Province in 2019

作　　者：杨文夷[1] 皮冬勤 汪琦 晏平仲[1] 余进海 肖林鸿 YANG Wenyi;PI Dongqin;WANG Qi;YAN Pingzhong;YU Jinhai;XIAO Linhong(Institute of Atmospheric Physics,Chinese Academy of Sciences,Beijing 100029,China;Jiangsu Provincial Environmental Monitoring Center,Nanjing 210019,China)

机构地区：[1]中国科学院大气物理研究所,北京100029 [2]江苏省环境监测中心,江苏南京210019

出　　处：《中国环境监测》2022年第4期198-206,共9页Environmental Monitoring in China

基　　金：江苏省PM2.5与臭氧污染协同控制重大专项(2019023)。

摘　　要：基于江苏省重污染天气监测预报预警系统多模式预报结果,分析了不同数值模式对江苏省13个城市细颗粒物(PM_(2.5))和臭氧(O_(3))的预报偏差特征,发展了多模式集合预报算法,并对其进行了评估。结果表明,相较于单一数值模式,集合预报算法显著改善了PM_(2.5)和O_(3)预报的准确率,其对江苏省PM_(2.5)和O_(3)空气质量分指数等级的预报准确率超过了80%。就江苏省整体而言,PM_(2.5)集合预报的准确率相比最优单一数值模式提升了6%。O_(3)浓度较低时,集合预报能有效改善各模式存在的高估现象。但受限于目前的校正策略,出现高浓度O_(3)污染时,集合预报对预报效果的提升相对有限。Based on the results of the multi-model operational forecast system in Jiangsu Province,the characteristics of prediction bias of fine particulate matter(PM_(2.5))and ozone(O_(3))from different air quality models were analyzed.In addition,an ensemble algorithm was developed and the performance of this new algorithm was evaluated.The results showed that,compared with each model,the ensemble algorithm greatly improved the prediction accuracy of PM_(2.5) and O_(3) levels.The prediction accuracy of pollution levels for both PM_(2.5) and O_(3) in Jiangsu Province reached more than 80%.Overall,the ensemble forecasting algorithm improved the prediction accuracy of PM_(2.5) by approximately 6%relative to the single optimization numerical model.The overestimation of each air quality model was effectively reduced by the ensemble forecasting under low O_(3) concentration.However,limited by current correction strategy,there was no significant improvement by using the ensemble forecasting under high O_(3) concentration.

关 键 词：空气质量 数值模式 偏差分析 集合预报 

分 类 号：X823[环境科学与工程—环境工程]