Vegetation top-of-canopy reflectance (TOC) as observed by remote sensing instruments is driven by wide range of factors, including optical properties of leaves and the structural properties of vegetation canopy - spatial distribution of leaves and branches. In addition, the instantaneous observation geometry (e.g. the sun and sensor zenith and azimuth angles) influences the TOC via changing observed shadow fraction in the canopies. As a result, vegetation TOC is highly dynamic, following the seasonal patterns of optical and structural properties.
From the abovementioned, the leaf optical properties are among the most important. They hold an information on leaf biochemical (chlorophyll and carotenoid content, water content) and structural (leaf thickness, distribution of pigments within a leaf) properties of leaves. Whereas biochemical effects on leaf reflectance is well understood, the role of leaf internal structure on leaf optical properties, its dorsiventral (two-sided) asymmetry and the specular component of leaf reflectance has been largely neglected. Moreover, leaf optical and structural properties are rarely measured in time, typically focusing on peak vegetation growing season only.
In this talk, we follow seasonal trends in leaf-level optical, biochemical and structural properties of five common broadleaved tree species of Central Europe (Populus tremula, Salix caprea, Betula pendula, Alnus glutinosa and Alnus incana). We link their optical properties (e.g. reflectance of adaxial and abaxial leaf side, reflectance to transmittance ratio, fraction of specular to total leaf reflectance) with the laboratory analyses of inner leaf structures for contrasting phenological stages from spring to fall. This allows us to observe species differences of leaf dorsiventral optical properties for varying phenological stages and quantify the role of inner leaf structure on optical properties. Next, we up-scale measured leaf-level dorsiventral optical properties to canopy level using Discrete Anisotropic Radiative Transfer model (DART) for contrasting vegetation stages from early spring to fall in order to study its impact on canopy TOC. Finally, we discuss the influence of dorsiventral leaf optical properties parametrization on forest reflectance simulations on forest quantitative product retrievals.
Petr Luke7scaron; received a PhD degree from Mendel university in Brno in 2012. He was a postdoctoral researcher at the University of Helsinki between 2012 and 2014 where he focused on linking forest structure and albedo using remote sensing observations and radiative transfer modelling. Currently, Petr is a junior researcher at Global Change Research Institute of Academy of Sciences of Czech Republic. His work focuses on remote sensing of vegetation with special interest on leaf optical properties, upscaling of leaf-level optical signals to canopy using radiative transfer models and quantitative product retrievals. His favourite working environment is forest.