Spectral characteristics of alpine grassland and their changes responding to grassland degradation on the Tibetan Plateau

摘要

Remote sensing is used as the indispensable technology in alpine grassland degradation assessment especially at regional scale on the Tibetan Plateau. However, the lack of field spectral data, as the foundation of remote sensing, due to the formidable natural and climate conditions hinders the understanding of spectral characteristics of alpine grassland and their degradation assessment. In this study, spectral characteristics of alpine grasslands and their changes responding to degradation were explored. The results showed that the main spectral characteristics for discriminating the dominant species of alpine meadow (Kobresia littledalei and Kobresia pygmaea), alpine steppe (Stipa purpurea) and desert (Potentilla fruticosa) are spectral features of chlorophyll, cellulose and water which are related to their growth form, plant inclination and residue of withered leaf sheaths. The spectral curves of alpine meadow have a much smaller variety over the whole spectral region compared to those of alpine steppe and desert which generally have weaker chlorophyll and water absorption features and more noticeable non-vegetation features. Different grassland degradation processes exhibit different patterns of spectral characteristics change due to the species composition, vegetation succession, vegetation coverage and soil background. Grassland degradation can happen without obvious vegetation coverage reduction or even with an increment of NDVI (normalized difference vegetation index). Therefore, the assessment of grassland degradation cannot be fulfilled well using single vegetation index or spectral feature. The combination of several vegetation indices or hyperspectral remote sensing along with the priori knowledge is needed in order to perform the assessment more accurately in further studies.