The identification of indicator grass species of grassland degradation based on the field spectral characteristics

Grassland is an essential part of terrestrial ecosystems. It has a significant impact on the carbon cycle, as well as on climate and on regional economies. Till now, vegetation indices are the most popular remote sensed detecting method of grassland degradation. Although vegetation indices are useful for estimating the biomass, but detecting changes of vegetation indices are not always effective, as grassland vegetation with different characteristics may still produce similar vegetation index values. The development of hyperspectral sensors provides a new approach to solve this problem. The Hulunbeier grassland was chosen as a study object. Reflectance spectra of leaves and pure canopies of some dominant grassland species, as well as reflectance spectra of mixed grass community were measured. Using spectral feature parameterization methods such as spectral slope, spectral derivative, spectral integration, and spectral index, the spectral feather of leaves and pure canopies had been extracted. So the typical grassland vegetation species can be distinguished. Then the spectra of mixed grass community were unmixed using linear mixing models, and the proportion of all the components had been calculated. The field validation proved spectral feature parameterization and pixel unmixing methods in this research are effective.