A stochastic model and a variance-reduction Monte-Carlo method for the calculation of light transport

Abstract

We compare the transport theoretic and a stochastic approach of modeling light propagation in the atmosphere. Computations of LIDAR return signals using algorithms based on the two different approaches show very good agreement of the numerical data. Furthermore, a deeper analysis of the formulas for the LIDAR return signal obtained from two kinds of modeling shows that they are equivalent. Combining these approaches and introducing splitting techniques, variance reduction Monte Carlo methods and codes are designed which are adapted to the configuration of the LIDAR. These codes allow the calculation of multiple scattering effects with high accuracy. Finally, we point out the importance and some perspectives of ultilization of multiple scattering in laser remote sensing of clouds. © 1995 Springer-Verlag.