区域大气环境模式RegAEMS中二次气溶胶的改进模拟研究

Improved simulation of secondary aerosol in the regional atmospheric environment modeling system RegAEMS

为了提升二次气溶胶的模拟精度,在区域大气环境模式RegAEMS中加入了硫酸盐气溶胶的两种新生成机制(NO_(2)+SO_(2)化学过程和过渡金属催化氧化(TMI))以及二次有机气溶胶(SOA)生成的挥发性有机物基集(VBS)方法.模拟了2020年1月上海市两次中度污染过程,并与观测数据进行对比验证.研究发现,两次污染过程硫氧转化率(SOR)均大于0.4,PM_(2.5)主要组分为SO_(4)^(2-)、NO_(3)-、NH_(4)^(+)等水溶性离子,占比为61.25%~63.85%.SOA占比为2.92%~3.0%.加入NO_(2)+SO_(2)化学过程和过渡金属催化氧化(TMI)后,硫酸盐模拟精度明显提升(相关系数(R)从0.49~0.63提升至0.58~0.67,相对标准偏差(NMB)从-35.0%~-36.5%提升至-17.3%~-14.2%).两种化学过程在污染发展阶段平均贡献占比为23.3%~27.9%,这可能是造成污染条件下SO_(4)^(2-)浓度迅速增加的主要原因.VBS机制能够较好地模拟出SOA的变化趋势(相关系数为0.53~0.56),由于硫酸盐和SOA生成机理的改进,RegAEMS在PM_(2.5)的模拟精度上有所提升(相对标准偏差(NMB)从-13.5%~-6.0%提升至-9.0%~-3.3%).

In order to improve the accuracy of the secondary aerosol simulation,this study added two new formation mechanisms of sulfate(NO_(2)+SO_(2) chemical process and heterogeneous transition-metal-catalyzed oxidation(TMI))and volatile organic compounds(VBS)based on the regional atmospheric environment model RegAEMS.The two moderate pollution processes in shanghai in January 2020 were simulated and compared with observational data for verification.The study found that the sulfur oxygen conversion rate(SOR)of the two pollution processes were both greater than 0.4,and the main components of PM_(2.5) were sulfate,nitrate and ammonium,accounting for 61.25%to 63.85%.SOA accounted for 2.92%~3.0%.After the addition of the NO_(2)+SO_(2) chemical process and TMI catalytic oxidation,the accuracy of sulfate simulation is significantly improved(The correlation coefficient(R)increased from 0.49~0.63 to 0.58~0.67,and the relative standard deviation(NMB)increased from-35.0%~-36.5%to-17.3%~-14.2%).The average contribution of the two chemical processes during the pollution occurrence stage is 23.3%~27.9%,which may be the main reason for the sudden increase of sulfate-concentration under pollution conditions.The VBS mechanism can better simulate the change trend of SOA(The correlation coefficient is 0.53~0.56),and the improved model slightly improves the simulation accuracy of PM_(2.5)(Relative standard deviation(NMB)increased from-13.5%~-6.0%to-9.0%~-3.3%).