Retrieving Canopy Structure from Hyperspectral Multi- Angular Satellite Data

The angular dependence of satellite-measured canopy reflectance has been shown to contain information on the structure of vegetated surfaces. The aim of the research was to explore the angular variability of spectral indices and to compare the suitability of angular indices and traditional vegetation indices for retrieval of forest structural attributes using Chris/PROBA data. With spruce the largest variability in NDVI occured in forward scattering direction. This variability strongly determined the correlations with canopy attributes. The results showed that off-nadir viewing improved NDVI performance considerably, particularly in forward scattering direction. The best relationship was found between NDVI at 44o view zenith angle and stand age (spruce: r= – 0.91, beech: r= – 0.77). Other directional indices (ANIX, NDAX or NDVIHD) were in most cases less correlated with canopy attributes.