An Iterative Fitting Algorithm for Matching Aerosol Size Distributions to Spectral Extinction Data

摘要

An iterative algorithm has been developed to retrieve aerosol size distribution from spectral extinction measurements. Uncertainties in the retrieved size distributions based on the uncertainty in aerosol refractive index, and due to non-uniqueness in size distribution beyond spectral observation region, are illustrated. The results of the retrieved volume size distributions confirm the nearly universally occurring mode in the 0.2-0.4 μm size range. However, in many instances, another mode near 1.0 μm was also evident, indicating the existence of an additional 'condensation' mode. The size distribution of fine particles (＜ 2μm) retrieved by the operational AERONET algorithm is observed to match with that retrieved by the present algorithm within the uncertainty range accounted for refractive indices. Size distributions dominated by coarse particles or very fine particles are found to have large uncertainties beyond the spectral observation region as the same spectral extinction may be produced by a range of sizes with respectively varied volumes.