摘要
利用cm级空间分辨率的无人机高光谱图像研究了NDVI、PRI等14种植被指数随空间尺度升高的变化规律,并分析了不同植被覆盖度(FVC)下植被指数的空间尺度效应,比较了各植被指数空间尺度效应的强弱。得到的结论如下:(1)随着空间尺度升高,不同植被指数的变化趋势和幅度并不一致;(2)cm级的空间分辨率下,由于植被类内光谱差异和冠层阴影的影响,植被指数在植被覆盖度大于等于0.5的中高植被覆盖下受空间尺度效应的影响更大;(3)由相对简单的比值形式构成的植被指数如NDVI、SR等,具有抵抗空间尺度效应的能力,增加计算公式的复杂性会降低这种能力;(4)对色素和冠层水分敏感的窄波段容易受到空间尺度效应的影响,进而使得由这些窄波段构成的植被指数对空间尺度变化更为敏感。
In this paper,we used UAV hyperspectral images with centimeter-level spatial resolution to explore the variation rules of 14 vegetation indices such as NDVI and PRI with the increase of spatial scale.Then,we analyzed the spatial scale effect of vegetation indices under different fractional vegetation coverage(FVC),and compared the magnitudes of spatial scale effect of vegetation indices.The conclusions obtained show that(1)as the spatial scale increases,the change trends and ranges of different vegetation indices are not consistent.(2)At the centimeter-level spatial resolution,vegetation indices are more affected by the spatial scale effect under medium-high vegetation coverage that are greater than or equal to 0.5 due to the effects of spectral differences within vegetation classes and canopy shadows.(3)Vegetation indices composed of relatively simple ratio forms,such as NDVI and SR,have the ability to resist influence of spatial scale effect,and increasing complexity of the calculation formula will reduce this ability.(4)Narrow bands that are sensitive to pigments and canopy moisture are easily affected by spatial scale effect,which makes vegetation indices composed of these bands more sensitive to spatial scale changes.
出处
《地理空间信息》
2021年第4期4-9,40,I0005,共8页
Geospatial Information
基金
国家自然科学基金资助项目(41671362、41371359)
江苏省特聘教授项目。
