Retrieval of aerosol optical constants from angular light-scattering measurement signals using principal component analysis approach

Abstract

An angular light-scattering measurement (ALSM) method combined with the probability density function-based ant colony optimization algorithm (PDF-ACO) is proposed for retrieval of aerosol optical constants. An optimal measurement angle selection method using a principal component analysis (PCA) approach is developed to improve retrieval accuracy. Results indicate that optimized angle selection can ensure retrieval accuracy. The aerosol optical constants over Beijing, China, which are available from the Aerosol Robotic Network (AERONET), are then reconstructed. The ALSM method’s convergence properties are also studied via comparison with those of the light reflection-transmittance measurement (LRTM) method. Results retrieved using the ALSM method show better convergence speed and accuracy than those retrieved using the LRTM method because the ALSM method does not require solution of the radiative transfer equation and allows more useful signals to be obtained. Additionally, the inverse accuracy of the refractive index results is better than that of the absorption index results; this is attributed to differences between the monodromic characteristics of the refractive index and absorption index retrieval results. All results confirm that the combination of the ALSM method with the PDF-ACO algorithm and the optimal measurement angle selection method provides effective and reliable aerosol optical constant reconstruction.