基于AQI的九江市空气质量统计预报方法

Statistical Forecasting Method of Air Quality in Jiujiang City Based on AQI

利用九江市2014—2017年气象观测站常规观测的气象资料和九江市环境监测站污染物日平均浓度资料、AQI日均值和首要污染物等观测要素,通过统计学方法对不同季节内地面气象要素和AQI做相关性分析,并选取相关性好的因子分别做回归分析并建立预报方程,结果表明:1)相对于全年回归预报方程,不同季节的气象要素与AQI的相关程度不同,不同季节的回归方程对AQI指数的预报准确率高;2)由于污染物的累积特征,前一天的AQI与当天AQI相关系数高达0.66,在回归的过程中,前一天的AQI值做为最重要的一项参数;3)春季、夏季由于降水比较明显,且由O 3造成的污染日数相对较多,而高温和日照对O 3有显著的相关,所以AQI的预报能力较强;秋冬季由于外来输送污染严重和PM 2.5首污的日数增加,气象要素的相关性较低,预报能力也随之降低。

Based on the daily air pollution monitoring data from Jiujiang Environmental Monitoring Center,the meteorological observation data from Jiujiang Meteorological Office from 2014 to 2017,correlation analysis was performed by statistical methods for the correlation between the surface meteorological elements and AQI by seasons,and the correlation factors were selected for regression analysis and prediction equations were established.The results showed that:1)The correlation between meteorological elements and AQI in different seasons is different.The accuracy of the prediction of AQI by the regression equation in different seasons is better than annual regression forecast equation.2)Due to the cumulative nature of the pollutants,the correlation coefficient between the previous day's AQI and the current day's AQI is 0.66.In the regression calculation,the AQI value of the previous day is used as the most important parameter.3)In spring and summer,precipitation is relatively obvious,and the pollution days caused by O 3 is relatively large,maximum temperature and sunshine duration have a significant correlation with O 3,so the AQI has a strong forecasting ability.In autumn and winter,the pollution is severe due to external transport and PM 2.5 polluted days increases,the correlation of meteorological elements is low,the forecasting ability is also reduced.