Radiative transfer modeling of the coupled atmosphere and plant canopy and BRDF retrieval

The limitations of conventional satellite remote sensing that mainly uses nadir observations of terrestrial surfaces has led to an exploration of the use of angular signatures. The Earth Observation System (EOS), to be launched in 1998, is capable of providing directional observations from the space. This dissertation was designed to study the fundamental properties of the directional reflectance of terrestrial surfaces.; Four new and inter-related algorithms have been developed in this study, including (a) an improved Gauss-Seidel numerical algorithm to solve the coupled atmosphere--vegetation canopy radiative transfer equation; (b) an analytic bidirectional reflectance distribution function (BRDF) model of canopy radiative transfer and its inversion algorithm; (c) a statistical BRDF model; and (d) an analytic model of atmospheric radiance transfer over a non-Lambertian surface.; The classic Gauss-Seidel algorithm has been widely applied in atmosphere research. This is its first application for calculating the multiple-scattering radiance of the coupled atmosphere and canopy, and an improved iteration formula is derived to speed convergence due to large optical thickness. One of the major advantages of this algorithm is that it can easily incorporate any form of surface BRDF as the lower boundary condition.; This dissertation presents an analytic canopy BRDF model based on a rigorous canopy radiative transfer equation in which the multiple-scattering component is approximated by asymptotic theory and the single-scattering calculation, which requires numerical integration to properly accommodate the hotspot effect, is also simplified. The Powell algorithm is then used to retrieve biophysical parameters from soybean measurement data based on both canopy and sky radiance distribution models. The results show that leaf area index (LAI) can be well retrieved, and more efforts are required to retrieve leaf angle distribution (LAD).; A new procedure is developed to obtain the more accurate angular radiance distributions of the atmosphere over a non-Lambertian surface using both two-stream and four-stream approximations. A statistical BRDF model is developed. Since BRDF cannot be directly measured by sensors, a procedure using the optimum technique is applied to retrieve BRDF parameters from measurements of the Advanced Solid-state Array Spectroradiometer (ASAS), The Portable Apparatus for Rapid Acquisition of Bidirectional Observations of the Land and Atmosphere (PARABOLA) and other multiangle imaging devices. Excellent results have been obtained.