Spectral characteristics of Northwest Sichuan's main alpine meadow vegetations
[Objective]The spectral reflectance characteristics of grassland are the basis for using remote sensing data to study the physical and chemical characteristics of grassland vegetation and vegetation classification. Research on the spectral reflectance characteristics of different grassland plants provides theoretical reference and technical sup-port for quantitative remote sensing and accurate identification of grassland vegetation.[Method]Portable ground ob-ject spectrometer was used to measure the field spectra of eight alpine meadow grass species:Potentilla anserina、Ko-bresia pygmaea、Caltha scaposa、Taraxacum tibetanum、Ligularia euryphylla、Anaphalis flavescens、Polygonum mac-rophyllum and Oxytropis ochrocephala in Yajiang,Ganzi,Sichuan Province from July to August 2021. The original spectral data were processed using the methods of first derivative,second derivative,continuum removal and vegeta-tion index,and the spectral reflectance characteristics of eight alpine meadow species were revealed.[Result]The re-sults showed that compared with the spectral characteristic parameters of different grassland plants,such as "green peak position","green peak amplitude","Red Valley position","Red Valley amplitude","red edge amplitude","red edge area","absorption peak area","absorption peak symmetry",the Hyperspectral Vegetation index was easier to distin-guish these 8 types of grassland vegetations. NDVI and mSR705 indexes of 8 plants were consistent,and the index values as follow:Taraxacum tibetanum＞Oxytropis ochrocephala＞Kobresia pygmaea＞Potentilla anserina＞Polygonum macrophyllum＞Caltha scaposa＞Ligularia euryphylla＞Anaphalis flavescens.The mSR705 index val-ues were obviously different among the 8 alpine meadow vegetations and easy to distinguish from each other.[Conclusion]Using different methods to analyze the spectral reflectance characteristics of grassland plants can effec-tively distinguish different types of grassland vegetation. This paper provides theoretical support for applying hyper-spectral remote sensing in vegetation classification,grassland condition investigation and dynamic monitoring of grass-land resources.
alpine meadowspectral characteristicsspectral derivativecontinuum removalvegetation index
万方数据
维普
