针对传统气溶胶类型确定方法的局限性以及当前气溶胶类型确定存在的困难，提出一种使用多波段气溶胶光学厚度数据确定气溶胶类型的方法。基于大气颗粒物的散射与吸收特性分析，通过构建查找表的方法实现气溶胶类型的确定。该方法利用Mie散射理论通过正向模拟不同类型气溶胶粒子数量与多波段光学厚度之间的关系来构建查找表，基于该查找表，使用440 nm、670 nm、870 nm及1020 nm 4个波段的气溶胶光学厚度确定气溶胶类型。使用模拟的多波段气溶胶光学厚度数据开展了气溶胶类型的确定实验，分析了不同波段气溶胶光学厚度误差对气溶胶类型确定结果的影响。结果表明，该方法可根据4个波段的气溶胶光学厚度以较高的精度确定出沙尘性、水溶性和煤烟3种气溶胶粒子的数量，从而确定气溶胶类型。

The high-precision determination of aerosol models is crucial for analyzing the environmental impact of aerosols and for the remote sensing of Aerosol Optical Depth (AOD). However, the determination of aerosol type remains difficult, hence severely restricting highly accurate AOD retrieval and the application of aerosol optical products in environmental monitoring. A high-precision method for estimating aerosol models is proposed in this paper. The determination of aerosol type plays a vital role in the analysis of aerosol optical properties and is also an essential part of highly accurate AOD retrieval.
Conventional methods, which utilize aerosol optical properties to determine aerosol types, are based on the relationship between the AOD of a single band and different aerosol types. However, due to the complex absorption and scattering properties of aerosols, it is difficult to obtain highly accurate aerosol types from AOD data with a single band. Hence, multiband AODs were introduced to enhance the accuracy of determining aerosol types. None the less, the current methods for estimating aerosol types with multiband AODs are iterative with complex and slow calculation processes.
A new method to determine aerosol type with multiband AOD data was proposed to overcome the existing difficulties of current methods. Aerosol type is determined based on a lookup table, which is built using the forward simulation of relations between the particle numbers of different aerosol types and multiband AOD data based on Mie scattering theory. AOD at the wavebands of 440, 670, 870, and 1020 nm are adopted to determine aerosol type.
The method was used to estimate aerosol types with multiband AOD. To evaluate the effectiveness of the proposed method, multiband AODs are simulated and applied in the validation experiment. Dust, water-soluble, and soot aerosol types are estimated with high precision. The effects of multiband AOD error on aerosol type determination are also analyzed. Results show that the proposed method can determine aerosol types with high stability.
Compared with the current real-time determination method of aerosol types, the approach proposed in this paper is fast and can be used to estimate aerosol type from pixel-scale satellite data with multiband AODs. Furthermore, this method can improve the inversion accuracy of aerosol optical thickness, as well as promote the application of aerosol optical products in environmental pollution monitoring.