Case Study: Assessing forest canopy productivity using active and passive remote sensing
Guest authored by Choy Huang, Kai-Ting Hu, Hong-You Lin; Department of Geography, National Taiwan University. Choy’s Ecology and Remote Sensing Lab use synoptic sensing methods to integrate information acquired from the field and airborne/satellite sensors to investigate the impacts of amplified climate change on terrestrial ecosystems.
Research Study
Forest canopy productivity – mainly contributed by litterfall including foliage, reproductive organs (fruits and flowers), bark and twigs – plays a crucial role in maintaining the health of forest ecosystems; a major part of the terrestrial carbon cycle, it is indicative of the metabolic rate of a forest. However, when it comes to quantifying forest litterfall at the regional scale, existing approaches are often time-consuming and labor-intensive. An ecological theory, called the metabolic scaling theory, states that the metabolic rate of organisms is the fundamental rate that governs most observed ecological patterns. Therefore, it was determined that we can estimate the amount of litterfall in an entire forest using tree size, because the size of an organism controls its metabolic rate.
Process and Methodology
Hinoki cypress trees (Chamaecyparis obtuse var. formosana), the most dominant tree species in Chilan coniferous cloud forest of the northeastern Taiwan, were selected as the target species. We periodically collected field litter production, brought these samples back to the lab, dried them in a biomass oven to 70°C, and weighed them. Ground truth data was then fed into a tree height model, which was derived by estimating tree sizes over a vast region with airborne laser technology known as “lidar.” By doing this, we were able to map the spatial-temporal dynamics of litter production and better understand the carbon cycle in tropical montane cloud forests.
To make this spatial metabolic scaling approach suitable for wide applications, it is crucial to investigate the feasibility of upscaling and correlating the data collected from the ground and the air to satellite optical observation. The very first step is to measure leaf reflective and transmitted solar energy of hinoki within the optical range of 350–2500 nm. In order to make an unbiased measurement, we chose to use a Malvern Panalytical ASD FieldSpec® spectroradiometer and an ASD Integrating Sphere. This portable set (FieldSpec + Integrating Sphere) can be operated in the lab or the field. In this case, the process was conducted in the lab.
These radiometric properties are closely related to plant physiological characteristics including pigment concentrations, lignin, sugar and/or water contents. We climbed up a flux tower to destructively sample sunlight and shaded leaves on tree tops. Collected leaves were immediately placed in a polyethylene-linear low density bag and in an icebox to preserve moisture and freshness before transferring them to a refrigerator, to maintain the freshness of the leaf tissues. Since the leaf size is too small to occupy the entire sample port, a customized small leaf sample holder is used to quantify the filled and unfilled spaces within the port.
Result
The precise measurement of leaf reflectance and transmittance can then be made.
This measurement information is essential for modeling canopy reflectance of hinoki forests using radiative transfer, which is the radiometric property directly measured by an airborne or spaceborne sensor. The data will facilitate the indirect assessment of metabolism of tropical montane cloud forests across scales.