Atmospheric aerosol size retrieval from LIDAR data applying genetic algorithm approach

Abstract : In this paper we present a novel approach using a genetic algorithm (GA) to solve the LIDAR "ill-problem". It is inverting aerosol size distribution from multi-wavelengths Lidar data in the UV-VIS-IR spectral range. This method do not need any predefined size distribution shape and is also running with data having poor S/N. Numerical convergence test of the GA have been done on both simulated data and on Lidar field measurements performed during the French POVA Campaign. It shows that size distribution is retrieved in millisecond rage, which represent a considerably decreasing in computing time consuming. Comparison between GA size distribution retrieval and SMPS ground-based measurements has been done showing an excellent agreement.