杭州地区污染预报和霾诊断WRF-RTIM研究
WRF-RTIM-based Model Output Statistics Method for Air Pollutants Prediction and Haze Events Diagnosis in Hangzhou
利用由数值预报模式WRF（Weather Research and Forecasting Model）和辨识理论实时迭代统计方法RTIM（real-time iterative model）组成的MOS (Model Output Statistics)方法对杭州市2013年2月-2013年3月期间的空气污染物日平均浓度做预报，预报值与实测值之间相关系数都超过0.75 ，PM2.5、PM10、SO2、NO2、CO 24h平均浓度和O3 8h平均浓度六类污染物分类预报临界成功率分别为92%、85%、100%、96%、100%和100%，命中率分别为96%、96%、100%、100%、100%和100%。分析表明，研究期间杭州地区气溶胶以细颗粒为主。根据PM2.5浓度、相对湿度及能见度预报值做霾日分类预报，临界成功指数为90%，命中率为96%。说明该MOS系统对污染物浓度及霾天气预报性能良好，可以为业务化预报提供参考。
MOS (Model Output Statistics) method which is composed of WRF (Weather Research and Forecasting Model) and RTIM (real-time iterative model), is used to predict daily air pollutants concentrations in Hangzhou from Feb. to Mar. 2013. The Correlation coefficients between predicted and observed values of these models are all above 0.75. CSI(Critical Success Index) of 24-hour average concentrations of PM2.5, PM10, SO2, NO2, CO, and 8-hour average concentrations of O3 attain 92%, 85%, 100%, 96%, 100% and 100%, and POD (Percent Of Doom) of those attain 96%, 96%, 100%, 100%, 100% and 100%. PM2.5 is the main pollutant during the prediction time. CSI and POD of Haze events prediction, which based on concentration of PM2.5, relative humidity, and visibility, reach 85% and 88% respectively. It shows that the MOS model performs well in predicting air pollutants and haze events, and can be used in operational prediction.
尤佳红、束炯、段玉森、陈亦君
环境污染、环境污染防治
空气污染霾PM2.5WRFMOS
air pollutantshazePM2.5WRFMOS
尤佳红,束炯,段玉森,陈亦君.杭州地区污染预报和霾诊断WRF-RTIM研究[EB/OL].(2014-02-27)[2025-08-16].http://www.paper.edu.cn/releasepaper/content/201402-581.点此复制
评论