INTERCOMPARING THE TWOMEY METHOD WITH A MULTIMODAL LOGNORMAL APPROACHTO RETRIEVE THE AEROSOL-SIZE DISTRIBUTION

Citation
U. Amato et al., INTERCOMPARING THE TWOMEY METHOD WITH A MULTIMODAL LOGNORMAL APPROACHTO RETRIEVE THE AEROSOL-SIZE DISTRIBUTION, JOURNAL OF GEOPHYSICAL RESEARCH-ATMOSPHERES, 101(D14), 1996, pp. 19267-19275
Citations number
26
Categorie Soggetti
Metereology & Atmospheric Sciences
Volume
101
Issue
D14
Year of publication
1996
Pages
19267 - 19275
Database
ISI
SICI code
Abstract
This paper concentrates on intercomparing an objective algorithm to es timate the columnar aerosol size distribution from aerosol spectral ex tinction and a widespread robust nonparametric inversion approach. The nonparametric approach which has been considered is the Phillips-Twom ey second-derivative regularization method. The parametric approach as sumes that the size distribution is the linear combination of a set of lognormal functions (typically, from 10 to 30 distributions are consi dered) which have suitable mode radii and standard deviations. Interco mparisons have been performed on the basis of aerosol optical thicknes ses derived by high-resolution solar radiation absorption spectra in t he spectral region 0.40-1.1 mu m. Absorption spectra were recorded by a grating spectrometer with a resolution of 0.5 nm. Each inversion, ei ther parametric or nonparametric, was obtained by using more than 450 data points at a time which, from a statistical point of view, allowed us to obtain high-quality results. For the cases presented here, we f ound that the two approaches agree fairly well in indicating the most significant characteristics of the aerosol size distribution.