A multiple scattering reflectance model for vegetation canopy based on recollision probability

The physically-based vegetation canopy reflectance model is the basis for accurate inversion of important vegetation parameters. According the interactive process between photons and canopy, canopy reflectance could be divided into two parts: single scattering and multiple scattering reflectance. Recollision probability is a useful tool linking leaf optical properties to canopy reflectance or absorption. In this paper, based on the recollision-probability, a new and practical multiple scattering model was approached. To estimate the accuracy of this model, Monte-Carlo simulation and 3D radiosity model were used and the results showed high accuracy of this model. Then the difference between first recollision probability (p₁) and multiple recollision probability(p_m) was discussed. Both of p₁ and p_m were needed in modeling multiple scattering model. At last, the contributions of soil reflectance, leaf albedo and proportion of sky radiation to multiple scattering reflectance were discussed.