本文利用2013年1月AERONET (Aerosol Robotic Network) 北京站的气溶胶光学厚度AOD (Aerosol Optical Depth)、细颗粒物光学厚度占总光学厚度的比例即气溶胶细模态比例η以及地面监测的细颗粒物PM_2.5 (Particulate Matter 2.5)质量浓度数据建立气溶胶细模态光学厚度AOD_f(fine-mode Aerosol Optical Depth)与PM_2.5的线性回归关系,并利用2013年2月1日—15日的数据验证该方法。结果表明,利用2013年1月建立的回归方法能够有效估算灰霾期间PM_2.5,获得PM_2.5的均值为8 5 μg/m³,均方根误差为50 μg/m³。利用气溶胶细模态订正方法估算的AOD_f与PM_2.5的相关系数大于AOD与PM_2.5的相关系数,这表明灰霾期间以PM_2.5为代表的细模态颗粒物成为气溶胶消光的主体,且AOD与PM_2.5的关系转化为AOD_f与PM_2.5的相关关系时,相关程度提高。垂直分布修正在灰霾时对改善AOD与PM_2.5相关关系作用不明显;当相对湿度大于80%时,湿度订正效果受到较大限制。

The correlation between fine Particulate Matters (PM_2.5) and Fine-mode Aerosol Optical Depth (AOD_f) is established. AOD_f is obtained from product of Aerosol Optical Depth (AOD) and fine-mode fraction at Beijing site belonging to the AErosol RObotic NETwork (AERONET) in January 2013. And then we compare estimation with observation of PM_2.5 from 1 to 15 February,2013. The results show that the developed correction method is effective to estimate PM_2.5 during haze, with root-mean-square error of 50 μg/m³ at a mean level of 85 μg/m³. The relationship between AOD_f and PM_2.5 is obviously better than that of AOD and PM_2.5. It is also found that when the relative humidity is higher than 80%, the humidity correction on AOD-PM_2.5 correlation is limited, and the vertical correction cannot improve the correlation during haze.