Simulating Imaging Spectroscopy in Tropical Forest with 3D Radiative Transfer Modeling

Optical remote sensing can contribute to biodiversity monitoring and species composition mapping in tropical forests. Inferring ecological information from canopy reflectance is complex data availability suitable such a task limiting, which makes simulation tools particularly important this context. We explored the capability of 3D radiative transfer model DART (Discrete Anisotropic Radiative Transfer) simulate top acquired with airborne imaging spectroscopy forest, reproduce spectral dissimilarity within among species, as well discrimination based on information. focused two factors contributing these properties: horizontal variability leaf optical properties (LOP) fraction non-photosynthetic vegetation (NPVf). The LOP was induced by changes pigment content, defined for each pixel hybrid approach combining modeling indices. influence simulated tested considering at individual tree crown level. incorporated NPVf into simulations following approaches, either part wood area density voxel or using brown pigments. validated different scenarios comparing scenes experimental statistical metrics, (within crowns, dissimilarity) supervised classification discrimination. pigments resulted closest match between measured reflectance. definition level conservation expected performances Therefore, we recommend future research forest physical remote-sensing account crowns species. Our framework could better understanding relationship variations taxonomic functional dimensions biodiversity. This work contributes improved integration applications, focusing remotely sensed ecosystems, current sensors, preparation multispectral hyperspectral satellite missions.