The use of direct observations over the aerosol particle size distribution for inverting lidar data

The work is concerned with the inversion of horizontal lidar data into the aerosol particle size distribution (APSD). The aerosol is assumed to consist of spherical particles of continental and oceanic origin. The particular refraction index is supposed to be known. It is shown that whereas the inversion method of mean ordinates works satisfactorily for aerosols of smooth particle size distributions, the retrieval of nonsmooth distributions yields only their general slope. This is particularly true for distributions whose components are represented by markedly isolated peaks. Numerical experiments were performed by imposing a constraint on the sought-for solution for a further improvement of the inversion efficiency. For constraint, an APSD was used on a limited particle size interval. The results showed that the above constraint can significantly improve the inversion accuracy not only on the size interval chosen for constraint, but also on the entire particle size range. The inversion accuracy increased with the increase of the constraint precision. The imposition of an APSD constraint proved to be particularly promising when applied to distributions with separate component peaks. The observation data of the Shoreline Environment Aerosol Study ( SEAS) were used to test the results of the numerical experiments. The test substantiated the theoretical conclusions about the efficiency of the method of mean ordinates for inverting lidar data into APSD. The recourse to the constraint of the experimental APSD improved the inversion results somewhat further. However, with the data available, it is so far impossible to make the constraint strict enough to maximize the effect.