Our primary Ecosyth research is based on studying the structure and composition of vegetation in landscapes using a low-cost, user-deployed system that is comparable to state-of-the-art tools and technologies.
By making repeated aerial photograph collections of the canopy throughout the growing season, it is possible to capture the phenology, or life-cycle traits, of forest trees.
We are currently using Ecosynth to observe fall phenology of individual canopy trees in 3D. This information can be used to understand the timing of canopy productivity and to identify canopy species and biodiversity.
The figure at left shows a set of Ecosynth 3D point clouds from an overhead view of the 2 ha 'Knoll' site at UMBC. The scenes go from full leaf-on in the upper left, clockwise through autumn senescence, to full leaf-off in the lower left.
Measurements of canopy height are the most commonly derived metrics generated from 3D remote sensing. Canopy height is used to estimate landscape above ground biomass and carbon, is a parameter for fire spread models, and is used for modeling habitat suitability, among many applications.
Our initial Ecosynth research demonstrated that accurate canopy height measurements can be obtained with a fusion of computer vision software, off-the-shelf digital cameras, and an inexpensive hobbyist aerial platform.
The figure at left shows a roughly 0.4 ha area of forest at UMBC from a high-resolution aerial photo (left panel) and a canopy height model generated using the Ecosynth system (right panel). Dark areas are relatively lower and light areas are higher.