Résumé:
Aerosols are tiny particles suspended in the air. Like greenhouse effect gas emissions, aerosols play an important role in the radiative balance of the atmosphere. However, is poorly quantified until now which justifies the intensification of efforts to study. In addition to their impact on the climate, aerosols perturb astronomical observations and remote sensing measurements. Several measurement methods have been implemented including the method LIDAR (Light Detection And Ranging), which allows to have 3D maps of their concentration.
The interest of optical LIDAR measurements is to give direct access to the distribution of limit layer and the optical extinction of atmospheric particles. The characterization of micro-physical properties of aerosols (size, concentration) is however less easy, because the size distribution of aerosols intervenes as integral form in the measurement of the optical extinction. However it has been achieved in the case of less complex atmospheres (desert
particles, acids, silica, fog, cloud).
Our attention was based on the study of desert particles characterized a big part of the terrestrial globe, for which we performed a detection of their size distributions using multi-spectral method LIDAR, where wavelengths of laser ranging from UV to IR were fixed in the following values: : (λ = 0.193, 0.694, 1.064, 3.370 and 10.60 μm). For each wavelength, we must have corresponding profile of extinction coefficients.
The stochastic method used to solve the problem of simulation is the method of genetic algorithms which is a more reliable and efficient technical to accelerate dramatically the computing time.
Finally the results of our work were compared with two distribution models that we have considered: the model of Longtin and the model Chomette.
