Phenology of leaf optical properties and their relationship to mesophyll development in cool-temperate deciduous broad-leaf trees.

Abstract Leaf optical properties (reflectance and transmittance spectra of a single leaf) are the key to interpreting remote sensing data regarding the structure and functions of vegetation canopies. These properties are determined by leaf biochemical and anatomical traits, which change seasonally in deciduous species. To reveal the species-specific seasonal patterns in leaf optical properties and their relationship to chlorophyll content and/or mesophyll structure in deciduous trees, we examined these mechanistic relationships in a pioneer species, Betula ermanii, and in sunlit and shade leaves of a late-successional species, Quercus crispula, during their seasonal development. Leaf optical properties were measured from leaf emergence to leaf fall in four years. After leaf unfolding, reflectance in the photosynthetically active radiation (PAR) region decreased gradually, while transmittance dropped rapidly. Both reflectance and transmittance increased before leaf fall. In the near-infrared (NIR) region, reflectance increased and transmittance decreased during the development period. Values of parameter N (the number of compact layers within a leaf in PROSPECT-5 model) in young leaves was very low (1.1 for B. ermanii and 1.0 for sunlit and shade leaves of Q. crispula) but increased rapidly by 30% for B. ermanii and sunlit leaves of Q. crispula and by 20% for shade leaves. N was higher in the order of B. ermanii > Q. crispula sun leaves > shade leaves. The PROSPECT-5 simulation with our measured data showed that the development of mesophyll tissue increases reflectance and decreases transmittance in both NIR and PAR regions, and an increase in chlorophyll decreases both reflectance and transmittance in PAR region. Our results indicate that structural and biochemical development of leaf mesophyll and its interspecific variations are essential biological characteristics for understanding seasonal changes in canopy reflectance. The results of this study will help us to better analyze remotely sensed data.