冬季德州市大气颗粒物消光与化学组成关系研究

为了深入探究华北地区冬季大气颗粒物的消光特性和化学组分之间的关系,本研究于2017年11月—2018年1月在山东省德州市平原县对大气颗粒物消光和化学组成进行了连续在线观测.运用多元线性回归方法和MIE散射模型定量分析了颗粒物各化学组成对颗粒物消光的贡献,进一步地,利用高分辨飞行时间气溶胶质谱(HR-ToF-AMS)结合正矩阵因子解析模型(PMF)得出二次气溶胶(OOA)、生物质燃烧有机气溶胶(BBOA)、还原性气溶胶(HOA)、燃煤燃烧排放的有机气溶胶(CCOA)的浓度,并进一步结合线性回归模型得到OOA、BBOA、HOA、CCOA对消光的贡献.结果显示元素碳(EC)是颗粒物吸光的最主要贡献者, OOA、BBOA和CCOA对颗粒物吸光也具有一定贡献,这主要是由于二次生成和一次排放的棕色碳的吸光造成的.颗粒物各组分与散射关系的分析结果表明,有机物对颗粒物散射影响最大,其中OOA的散射截面最大,对颗粒物散射的贡献也最高,可以占到总散射的53.4%.颗粒物中有机物对大气总消光的贡献可达75.5%,其中OOA、BBOA、CCOA和HOA对总消光的贡献分别为47.8%、14.7%、9.0%、4.0%. Mie散射的结果与多元回归结果比较一致,但部分时间段偏差较大,因此在不同的研究中应根据不同情况选择研究方法.

Relationship between the aerosol light extinction and chemical composition in winter of Dezhou city

To investigate the relationship between the particle chemical composition and the light extinction in winter of the North China Plain (NCP), continuous measurements of particle scattering and absorption were conducted from November 2017 to January 2018 at a regional site, Pingyuan site in Shandong Province. Besides, an Aerodyne high resolution time-of-flight aerosol mass spectrometer (HR ToF AMS) combined with the positive matrix factor (PMF) analysis was applied to apportion the secondary organic aerosol (SOA), hydrocarbon-like organic aerosol (HOA), coal combustion organic aerosol (CCOA), and biomass burning organic aerosols (BBOA).The multiple linear regressions and the Mie modeling were deployed to estimate the contributions of the different chemical components to the particle light extinction. Our results indicate that the Element Carbon (EC) was the major contributor to the particle absorption. OOA, BBOA, and CCOA were also important because of the primary emission and secondary formation of brown carbon. Organics contributed the most to the particle scattering. OOA has the largest scattering cross section, and contributed to 53.4% of the particle scattering coefficients. Particulate organic matters contributed to 75.5% of the total light extinction, with the contributions of 47.8%, 14.7%, 9.0%, and 4.0% from SOA, HOA, COOA, and BBOA, respectively. Although the results of the multi regression and Mie modeling agree well during most of the time, one should carefully choose the method basing on the research purpose.