玉米和大豆光合有效辐射吸收比例与植被指数和叶面积指数的关系

基于东北典型黑土区的玉米和大豆的实测光谱反射率、光合有效辐射及叶面积数据，选取常用的9种植被指数，并根据光谱曲线特征和植被指数结构建立了两种新的植被指数，对其估算玉米和大豆冠层FPAR效果进行了对比分析。结果表明，各植被指数与冠层光合有效辐射吸收比例(FPAR)的关系因植被类型而异。以近、短波红外波段较以可见光、近红外波段计算植被指数的估算效果好。NDVI、RVI在可见光、近红外波段计算的植被指数中估算FPAR效果较好，玉米估算模型R²分别为0.81和0.82，大豆估算模型R2均为0.81；NDSI、RSI在近、短波红外波段计算的植被指数中较好，玉米估算模型R2均为0.86，大豆估算模型R2均为0.84，优于NDVI和RVI。试验表明，利用近、短波红外波段估算FPAR是可行的；冠层含水量较土壤背景对FPAR影响更大；玉米和大豆冠层FPAR与叶面积指数(LAI)呈较好的对数关系，估算模型R²分别为0.75和0.70；但用植被指数估算FPAR效果要优于用叶面积指数。

Relationship between Fraction of Photosynthetically Active Radiation and Vegetation Indices,Leaf Area Index of Corn and Soybean

Based on the measured reflectance, photosynthetically active radiation, and leaf area index (LAI), nine common vegetation indices and two vegetation indices, developed by the authors according to the characteristic of reflectance curves and the vegetation index structures, were used to estimate fraction of photosynthetically active radiation (FPAR) in corn (Zea mays L.) and soybean Glycine max (L.) Merri.]. The relationships between vegetation indices and FPAR differed in different vegetation types. The performances of of near infrared and shortwave bands were better than these of visible and near infrared bands in FPAR estimatin by vegetation indices. Normalized difference vegetation index (NDVI) and ratio vegetation index (RVI) were the best in the vegetation indices calculated by reflectance in visible and near infrared bands for FPAR estimation, with R² of estimating models 0.82 and 0.81 for corn, both of 0.81 for soybean. Normalized difference shortwave index (NDSI) and ratio shortwave index (RSI) were best in the vegetation indices calculated by near infrared and shortwave bands with almost the same R² values for NDSI and RSI in corn (0.86) and soybean (0.84). It is suggested that the near infrared and shortwave bands are feasible to be used for FPAR estimation. According to the field experiments, the water content of vegetation canopy had larger effect on FPAR estimation than soil background. The canopy FPAR and LAI showed good logarithm relations in corn and soybean canopy with R2 values of 0.75 and 0.70, respectively, indicating that vegetation indices are better than LAI for FPAR estimation.