陆地气溶胶一般由粗粒子和细粒子气溶胶组成, 二者的光学特性并不相似, 如何利用卫星观测数据获取粗粒子和细粒子气溶胶是反演中所面临的重要问题。 基于AERONET/PHOTONS北京站地面观测数据, 本文分析得到了北京地区的气溶胶模式(包括气溶胶复折射指数和谱分布函数), 结果表明北京地区的气溶胶呈明显的粗细粒子混合的双峰分布, 粗细粒子的光学性质也差异明显。 在此基础上, 利用暗目标法从2007年全年的北京地区MODIS数据反演得到气溶胶细粒子、 粗粒子和总光学厚度。 对反演结果的地面验证表明, 利用MODIS数据能够较好实现细粒子和总体气溶胶光学厚度(相关系数在0.8以上)以及Angstrom指数(相关系数为0.517)的反演, 但对粗粒子应用效果较差。

Generally, aerosols over land are composed of fine and coarse aerosols, which have different optical properties. How to derive the fine mode and coarse mode aerosols from satellite observations is an important issue in the aerosol retrieving. In the present paper, information of aerosol models (including refractive index and size distribution function) was retrieved from the ground-based measurement of the AERONET/PHOTONS site in Beijing. The retrievals indicate that, aerosols over Beijing have a bi-model distribution, and the optical properties of fine and coarse aerosols are distinct. Based on the dark dense vegetation (DDV) method, optical depth of the fine, coarse and total aerosols were derived from MODIS data over Beijing area in 2007. The validation of these satellite retrievals shows that from MODIS data, the optical depth of fine and total aerosols can be retrieved well (with correlation coefficients greater than 0.8), and so can the Angstrom exponent (having a correlation coefficient of 0.517). However, relatively poor results were obtained when retrieving the optical depth of coarse aerosols.