Spectal variation of vegetative areas in leaf-off season observed by hyperspectral data

Due to the phenological variation of vegetation growth in temperate region, it is often difficult to accurately assess the surface conditions on agricultural croplands, grasslands, and disturbed forests by current multi-spectral remote sensor data. In particular, the spectral similarity between soil and dry vegetation has been primary concern to correctly appraise the surface conditions during the non-growing seasons in temperature region. This study analyzes the spectral characteristics of the mixture of dry vegetation and soil. The reflectance spectra were obtained from laboratory spectroradiometer measurement and from EO-1 Hyperion image data. The reflectance spectra of several samples having different level of dry vegetation fractions show similar pattern from both lab measurement and hyperspectral image. Shortwave IR wavelengths, in particular at wavelength near 2200nm, are more sensitive to the fraction of dry vegetation. The reflectance spectra of bare soil could be distinguished from the other surfaces covered by dry vegetation during the leaf-off season. © 2005 IEEE.