Correlation analysis of multi-angle simultaneous polarization measurements with the concentration of suspended atmospheric particles

Abstract

Suspended particle is the main components of atmospheric aerosols, it is important to determine the composition, content, source, and concentration variation of suspended particles in the atmosphere. In this paper, we propose a new method based on the canonical correlation analysis(CCA) method to analyze the correlation between the polarized data and the concentration of suspended particles in the atmosphere. The new method can analyze very weak correlations between two sets of variables, which are both high-throughput and high-dimensional. Combined with the kernel probability density function, we can find out the elements that are more correlated in the two sets of variables. Firstly, the component of the suspended particles with the highest correlation between multi-angle polarized light data and suspended particles' concentration reference values are presented. Then, the concentration change of suspended particles in the atmosphere is predicted based on the measured polarized light data by the locally weighted linear regression(LWLR) method initially, the consistency of experimental data and predicted data verified the effectiveness of the new method.