Retrieval of aerosol parameters from simulated spectral and bidirectional Stokes vector data

Kurzfassung

In a series of radiative transfer calculations the Stokes vector is derived to simulate radiance and polarization measurements at three different wavelengths in the visible and near infrared spectral range for different aerosol types. The used radiative transfer model is based on the matrix operator method (Plass et al. 1973). Principal component analysis is applied to the simulated data to determine the number of independent parameters that describe the system and to derive the principal components. The relevant principal components are used to train a neural network to retrieve the refractive index, the effective radius and the optical thickness of the aerosol. The learning algorithm for the neural network is back propagation.