论文部分内容阅读
[目的]用偏最小二乘回归法研究2009—2012年武汉市可吸入颗粒物(PM10)、二氧化硫(SO2)和二氧化氮(NO2)浓度的气象影响因素。[方法]研究数据为2009—2012年武汉市环境自动监测点监测的PM10、SO2和NO2日平均浓度,以及同期武汉市地面气象观测台所观测的24 h(当日20:00到次日20:00)累积降水量、平均气压、平均风速、平均气温、平均相对湿度和平均日照时数。用偏最小二乘回归法提取变量的变异信息,交叉验证方法确定最佳主成分数,进而确定气象因素与大气污染物间的线性关系。[结果]2009—2012年间武汉市空气质量逐年改善,呈现PM10和NO2复合型大气污染特征。偏最小二乘回归分析结果提示,第一主成分对PM10、SO2和NO2变异的解释能力分别为0.722、0.915和0.702。24 h累积降水量的增加、平均风速的加大、平均气温的升高、平均相对湿度和平均日照时数的增加能够降低该城空气质量参数(PM10、SO2和NO2)的浓度,而平均气压的上升会使PM10、SO2和NO2浓度增加。气象因素对PM10、SO2和NO2浓度的作用程度存在差异。平均风速和平均温度是气态污染物浓度的主要气象影响因子,降水量对NO2浓度的稀释作用明显强于对PM10浓度的影响。[结论]偏最小二乘回归法能够克服气象因素之间的多重共线性。气象因素(降水量、平均风速、平均气温、平均相对湿度、平均日照时数、平均大气压)对武汉市大气污染物浓度的影响存在差异性。
[Objective] To study the meteorological influencing factors of concentration of PM10, SO2 and NO2 in Wuhan from 2009 to 2012 using partial least-squares regression method. [Method] The research data are the average concentrations of PM10, SO2 and NO2 monitored by environmental monitoring stations in Wuhan from 2009 to 2012 and the observed concentrations of groundwater in Wuhan during the same period of 24 h (from 20:00 on the day to 20:00 on the next day Cumulative precipitation, mean barometric pressure, mean wind speed, mean temperature, mean relative humidity and average hours of sunshine. Partial least squares regression was used to extract the variation information of variables, and the cross validation method was used to determine the optimal principal components to determine the linear relationship between meteorological factors and atmospheric pollutants. [Result] The air quality in Wuhan City improved year by year from 2009 to 2012, showing the characteristics of PM10 and NO2 compound air pollution. Partial least squares regression analysis indicated that the explanatory power of the first principal component for PM10, SO2 and NO2 variation was 0.722, 0.915 and 0.702.24 h, respectively. The cumulative precipitation increased with the increase of mean wind speed and mean temperature , Average relative humidity and average sunshine hours can reduce the concentration of air quality parameters (PM10, SO2 and NO2) in the city, and the increase of average pressure will increase the concentrations of PM10, SO2 and NO2. Meteorological factors have different effects on the concentrations of PM10, SO2 and NO2. The mean wind speed and average temperature are the main meteorological factors affecting the concentration of gaseous pollutants. The dilution of NO2 concentration by precipitation is obviously stronger than that of PM10. [Conclusion] Partial least squares regression method can overcome the multicollinearity between meteorological factors. The influence of meteorological factors (precipitation, average wind speed, average temperature, average relative humidity, average sunshine hours and mean atmospheric pressure) on the air pollutant concentration in Wuhan is different.