Canopy Vertical Structure from the MODIS BRDF Product
Zhuosen
Wang, Boston University, wangzhs@bu.edu
Crystal
Schaaf, Boston University, schaaf@bu.edu
(Presenting)
Yuri
Knyazikhin, Boston University, jknjazi@bu.edu
Mitchell
Schull, Boston University, schull@bu.edu
Tian
Yao, Boston University, tianyao@bu.edu
Feng
Zhao, Boston University, zhao26@bu.edu
Ranga
Myneni, Boston University, rmyneni@bu.edu
The structure of a canopy determines the interaction probabilities for photons within the vegetation. The canopy spectral invariants of recollision probability and escape probability are wavelength-independent and intrinsic canopy structure properties. Recollision probability is the probability that a photon scattered from a phytoelement in the canopy will interact within the canopy again. The escape probability is the probability that scattered photons will escape the vegetation canopy either through the upper or lower boundary. Escape probabilities and spectral reflectances are derived from the MODIS Bidirectional Reflectance Distribution Function (BRDF) Product (MCD43A1) at a 500m resolution. Multidate, cloud-free, atmospherically-corrected surface reflectances from both Terra and Aqua are used with a linear, kernel-driven BRDF model to estimate the surface reflectance anisotropy of the surface. The modeled quantities along the principal plane can then used in multivariate linear regression models to estimate canopy heights from the Laser Vegetation Imaging Sensor (LVIS). This airborne full waveform LiDAR was deployed over the Howland Experimental Forest, Maine, in 2003. The 28-m LVIS canopy height data were resampled to 500 m to match the MODIS BRDF values. Results over Howland suggest that such MODIS-derived canopy spectral invariant parameters can extract the effective information required to retrieve canopy height.
NASA Carbon Cycle & Ecosystems Active Awards Represented by this Poster: