| << | Profiling of Aerosols | >> |
Abstract
The desert dust aerosols play an important part in Earth’s radiation budget, thus development the methods for remote study of particle microphysical properties is highly demanded. During the last decade the multi-wavelength (MW) Raman lidars have demonstrated their potential to provide this kind of information. Up to the present the physical models in retrieval algorithms for processing of MW lidar data were based on Mie theory. However, this model is applicable for modeling light scattering by spherical particles only and does not adequately reproduce the scattering by non-sherical particles. This fact imposes serious limitations on the interpretation of the lidar observations of the desert dust. Here we present the algorithm using spheroid model for inversion of multi-wavelength lidar data. Following the positive experience of AERONET retrieval developments, we model the aerosol as a mixture of spherical and non-spherical aerosol components. The non-spherical component is an ensemble of randomly oriented spheroids with size independent shape distribution. It is fixed to the axis ratio distribution providing the best fit to the detailed polarimetic laboratory measurements for desert dust sample. We considered the Raman lidar based on a tripled Nd:YAG laser. Such lidar provides three backscattering, two extinction coefficients and particle depolarization ratio at single or multiple wavelengths. The accuracy for retrieval of particle surface, volume concentration and effective radius for 10% errors in input data is estimated to be about 30-40%. It is demonstrated that if the effect of particle non-sphericity is not accounted the error of parameters estimation is notably increased. Developed algorithm was tested with experimental data of Sahara dust outbreak episode, measured with BASIL multi-wavelength Raman lidar on August 2007. The retrieved verticals profiles of dust particle parameters for different days will be discussed in a separate presentation at this Symposium.