Study On The Vegetation-hyperspectral Characteristics Of Alpine Steppe Under Different Degraded Degrees In Sanjiangyuan Region

In the distribution area of alpine steppe in the east of Sanjiangyuan region,five degraded gradient alpine steppe were selected:no degraded steppe?ND?,lightly degraded steppe?LD?,moderately degraded steppe?MD?,heavily degraded steppe?HD?.)and extremely degraded steppe?ED?,using the ASD FieldSpec 4 spectrometer,combined with hyperspectral imaging techniques,ground investigations and spectral data were obtained for different degraded gradient vegetation and dominant plants.Through analysis and research,the following results were obtained:1)The spectral reflectance curves of different degraded gradient vegetations are significantly different.In the visible light range?350-750 nm?,the spectral reflectances of five degraded gradients are ED>LD>ND>MD>HD in descending order.In the near-infrared band?750-1350 nm?,the reflectivity of LD is larger than ED,and the reflectivity of LD and ED is much larger than the other three degradation gradients.2)Different plants with the same degenerative gradient have completely different spectral reflectance curves and show their respective characteristics.The spectral reflectance of Stipa purpurea and other plants shared by different degraded gradients are compared to find the same plant species.The spectral reflectance in different degradation gradients are significantly different,showing the spectral reflectance characteristics of plants in adapting habitats.3)Sensitive wavelengths of above-ground fresh biomass are?350,?360 and?370,sensitive hyperspectral parameter is mND705;sensitive wavelengths of above-ground dry biomass in the first derivative spectrum are?793 and?1002,sensitive hyperspectral parameter is mND705;coverage The sensitive wavelengths in the first derivative spectrum are?954 and?1205,the sensitive hyperspectral parameter is VI1;the sensitivity wavelength of total nitrogen is?350 in the original spectrum,and the?423 and?1263 in the first derivative spectrum,and the sensitive hyperspectral parameter is VI1.The sensitive wavelengths of total phosphorus are?965,?1189 and?1100 in the first derivative spectrum,DVI and VI3 in the sensitive hyperspectral parameters,?1203 and?1209in the first derivative spectrum of the sensitive wavelength of total potassium,and the sensitive hyperspectral parameter is VI2.The above-mentioned sensitive wavelength and sensitive hyperspectral parameters have significant correlation with the index to be estimated?P<0.05?.4)Use the variable R?350 to establish the best model for estimating above-ground biomass,the model is y=142.2e^-9.894x,the accuracy is 90.02%;use the variable D?793 to establish the best model for estimating above-ground biomass,the model is y=74699x+12.074,the accuracy is86.25%;using the variable D?1205 to establish an optimal model for estimating the coverage,the model is y=3*10⁶x^1.7675,the accuracy is 88.78%;a model estimate for estimating total N is established using the variable D?423,The model is y=-471046x²+254.72x+0.2248,the accuracy is 93.31%;An optimal model for estimating total P is established using the variable D?1189,The model table is y=0.1355e^3077.4x and the accuracy is 91.29%;The variable D?1209 is used optimal model for estimating total K was established.The model was y=461437x^1.2616,the accuracy is90.69%.