Hyperspectral Bidirectional Reflectance Measurements of Fagus Sylvatica Leaves

This research focused on bidirectional reflectance distribution function (BRDF) measurements of Fagus sylvatica leaves, using a state of the art new multispectral Compact Laboratory SpectroGonioreflectometer (CLabSpeG). The BRDF was based on 4356 reflectance measurements covering the electromagnetic spectrum from 350 nm to 2500 nm. These measurements were acquired by independent positioning of the sensor, sample holder, and light source, with an azimuth and zenith resolution of 30 and 15 degrees, respectively. A series of BRDF data of leaves were collected using CLabSpeG in a limited acquisition time of 2 hours and 36 minutes. Data were collected from 12 leaves for two seven year old Fagus trees. Our methodology protocol focused on BRDF behavior and modeling of leaves. The primary research objective was to represent the nonlambertian state of leaf reflectance, while investigating which wavelengths produce a more profound reflectance effect at certain angular positions. The topological position of leaves in a tree was found to be significant in terms of BRDF variation, resulting in reflectance differences between varying leaf positions at certain wavelengths, especially in the visible domain. A reflectance change also was recorded during the BRDF measurements due to drying effects from the laboratory illumination source. Moreover, when goniometric changes are involved we found inadequate the collection of data in reflectance mode, due to sensor’s software rounding errors, and we suggest the use of raw digital number values. INTRODUCTION Our ability to both interpret remotely sensed optical data and develop new vegetation remote sensing approaches depend directly on our understanding of the multitude of factors controlling canopy and landscape reflectance signals (i). A fundamental understanding of the energy-matter interactions at the earth’s must be established in order for a meaningful analysis and processing of data obtained via remote sensing techniques (ii). Such an understanding needs to account for variations in the quantity and quality of radiation recorded by the airand space-borne sensors. Plant structural attributes control canopy reflectance characteristics by orienting the scattering in three-dimensional space, allowing the interaction of photons with multiple surfaces such as leaves, woody material, and soils (iii). The amount of reflectance in a vegetation canopy is directly linked to the scattering processes at the leaf level. Although this leaf-level scattering varies with leaf structure, water content, and the biochemical constituents concentration (iv), it is highly dependent on canopy structural characteristics such as leaf area index and leaf angle distribution (v).The reflectance within a leaf furthermore varies between electromagnetic spectrum regions, with the visible domain (mainly driven by the absorption of chlorophylls) exhibiting different reflectance characteristics from the infrared domain (water absorption and cell structure dominating) (vi). As a result, accurate estimates of leaf spectral properties for reflectance, transmittance, and absorbance are critical for relating radiative transfer properties of leaves to their phenological and physiological state (vii). © EARSeL and Warsaw University, Warsaw 2005. Proceedings of 4th EARSeL Workshop on Imaging Spectroscopy. New quality in environmental studies. Zagajewski B., Sobczak M., Wrzesień M., (eds) The incoming solar irradiance (direct and diffuse) and outgoing reflected radiance for a measured surface such as a plant canopy or leaf, can be described as two hemispherical distributions of electromagnetic radiation, specifically the Bidirectional Reflectance Distribution Function (BRDF) (viii). Considering the lack and difficulty of capturing the reflectance properties of leaves, a goniospectroradiometric device (CLabSpeG) was constructed to allow for an accurate measurement of the BRDF of vegetation elements. Bidirectional reflectance data of leaves were obtained under specific mechanical and natural specifications and limitations, resulting in a total 4356 reflectance measurements. These data were collected for every leaf at a variety of angles, covering a hemisphere, with 30 degrees and 15 degrees resolution in azimuth and zenith, respectively. The BRDF data will be applied in combination with tree architecture measurements in order to build a Virtual Forest Imaging system that will allow investigation of solar radiation interaction with forest canopies for remote sensing and virtual imaging purposes. It is our intention to use these data as hyper-spectral BRDF libraries of vegetation as well as ground reference data for multiangular remote sensing applications. CLabSpeG DESCRIPTION The CLabSpeG consists of four major parts, located in a darkened laboratory (Figure 1) One horizontal, circular black anodized aluminum rail ring with a diameter of 1.25 m. A vertical semi-circle arc (diameter = 1.25 m) mounted on the horizontal rail, supporting the light source. A vertical stationary semi-circle arc (diameter = 1.05 m), mounted inside the previous arc on a black wooden table to support the imaging sensor. A stainless rotating horizontal plate of 0.20 m diameter, placed in the centre of the apparatus and serving as the sample holder. Figure 1: Mechanical system setup of CLabSpeG: Horizontal aluminum rail supporting the light source arm, which rotates anti-clock wise with a resolution of 30 degrees; Stationary arm supporting the imaging sensor with a resolution of 15 degrees in zenith; Sample holder in the centre with a spectralon panel on top, rotating clock-wise with a resolution of 30 degrees. The rings and arcs are 5 cm wide and 1.5 cm thick. The circular rail allows for the azimuthally motion of the light source and rests on three 24 volts precision motors, which support up to 200 Newton each, and provide an angular resolution of 30°. The light source-supporting arc covers 180° in zenith with 5° increment laser cut metal bolts. These bolts are recognized and engaged at each