Principal component analysis applied to multiwavelength lidar aerosol backscatter and extinction measurements

Abstract

The use of powerful Raman backscatter lidars enables one to measure the stratospheric aerosol extinction profile independently of the backscatter, thereby obtaining additional information to aid in retrieving the physical characteristics of the sampled aerosol. We used principal component analysis to construct a self-consistent method for the retrieval of aerosol bulk physical and optical properties from multiwavelength elastic and/or inelastic Raman backscatter lidar signals. The procedure is applied to synthetic and actual lidar signals. We found that aerosol surface area and volume can be usefully estimated and that the use of Raman-derived extinction data leads to a notable improvement in the accuracy of the estimations.