Numerical investigation on the accuracy of size information of fractal soot aerosols retrieved by lidar: Optical property, morphology effect, and parameterization scheme

Abstract

Lidar is effective in retrieving size distribution of atmospheric particles such as soot aerosols in real-time based on Mie scattering theory. However, the fundamental spherical shape assumption of Mie theory is obviously discrepant with the actual situations of atmospheric soot particles which are fractal and mixed with coating materials. In this study, the effects of fractal shapes and morphological parameters such as fractal dimension (Df), volume fraction (Vf), and particle radius (r) of soot aerosols on the size information retrieved by lidar based on Mie theory were investigated from the numerical aspect, closed-cell and coated-aggregate models were employed to represent soot aerosols thinly and heavily coated by non-absorbing materials, respectively. Optical properties of fractal soot particles were calculated at first, then backscattering and extinction properties were employed to retrieve backscattering equivalent radius (rbk) and extinction equivalent radius (rext) based on Mie theory, just like the lidar retrieval. Results revealed that optical properties of spherical and fractal models differ significantly with the variation of volume equivalent radius (rvol). Radius retrieval errors based on extinction cross-section are much smaller than that based on backscattering cross-section. The retrieval accuracy of coated-aggregate is higher than bare fractal aggregate and closed-cell models. Retrieved size distribution parameters mean radius are generally underestimated while standard deviation are overestimated for both models. Furthermore, parameterization schemes connecting rext and rvol of soot were constructed, retrieval errors are effectively reduced from about 25% to 10%, which is meaningful for the improvement of size retrieval accuracy of soot using lidar observations.