Spectral reflectance of partly transmitting leaves: Laboratory measurements and mathematical modeling

Previous work related to remote sensing of natural sceneries has shown that the near infrared signature of a plant canopy is a combined function of leaf optical properties, canopy geometry, and soil reflectance. This paper deals with the problems involved in spectrophotometric measurements of leaves that are partly transmitting and partly reflecting, and thus influenced by the sample background provided by the measuring instrument. A mathematical model requiring only single-leaf spectral reflectance data as input is developed for prediction of multiple-leaf reflectance. The validity of the model appears to be adequate for any number of stacked leaves, when leaf-to-leaf variations caused by differences in water content and pigmentation are considered.