The importance of leaf BRDF in forest canopy bidirectional reflectance : a case study using simulated canopy architecture and PBRT ray tracing

Two Fagus sylvatica L. stands with different Leaf Area Index and similar planophile Leaf Angle Distribution were created with L-systems based plant growth software and rendered using both Lambertian and Bousquet’s model leaf Bidirectional Reflectance Distribution Function (BRDF) assumption. During the simulation process the sun was positioned at three different zenith angles, namely 10°, 30°, and 60°. The sensor azimuth angles, for each case, ranged from 0° to 315° in 45° increments while the view zenith angles were 0°, 20° and 40°. The forest stands were rendered with the Physically Based Ray Tracer (PBRT) algorithm introducing in each case the Lambertian or the Bousquet BRDF approximation. This set-up enabled a direct reflectance comparison of the two approximations. It was found that the common assumption of Lambertian leaf BRDF input in a global canopy illumination scenario overestimated the reflectance in the visible part of the electromagnetic spectrum. However, in the infrared part we can assume that the Lambertian approximation of leaf BRDF can accurately simulate the light reflectance in a canopy stand. Similarly in this case study both rendering approaches provided equal values for the Normalized Difference Vegetation Index (NDVI) while differed greatly for the Photochemical Reflectance index (PRI)