水稻高光谱红边位置与叶层氮浓度的关系

实时无损监测叶片氮素状况对水稻精确氮素管理具有重要意义。本研究基于多年不同施氮水平和不同水稻品种的田间试验观测资料，系统分析了水稻高光谱红边区域和位置特征与冠层叶片氮浓度的定量关系。结果表明，水稻冠层的红边区域光谱受施氮水平和品种影响较大，一阶导数光谱在红边区域出现“三峰”现象。经典的红边位置(660~750 nm之间光谱反射率的一阶导数最大值)由于“三峰”特征现象而对水稻氮素浓度变化不够敏感，难以适用于水稻氮素状况的准确监测。基于倒高斯模型、线性内插法和线性外推法构造的红边位置随水稻氮浓度呈现连续变化模式，适用于水稻叶层氮浓度的定量监测；另外，基于695 nm、700 nm和705 nm等3个波段的拉格朗日算法也可估测水稻叶层氮浓度。比较不同红边位置发现，改进型线性外推法较其他几种算法更能有效地监测水稻冠层叶片氮浓度。

Quantitative Relationship between Hyper-Spectral Red Edge Position and Canopy Leaf Nitrogen Concentration in Rice

Real-time and non-destructive monitoring of crop nitrogen status is needed for precision management and dynamic regulation in rice fertilization. This research made a systematic analysis on the characteristics of the first-derivative reflectance spectra in red edge area, and the quantitative relationships between red edge position (REP) with different algorithms and canopy leaf nitrogen concentrations in the conditions of different nitrogen rates and rice varieties in different seasons of field-grown rice. The results showed that spectrum in red edge area was significantly affected by nitrogen levels and varietal types, and “three-peak” feature could be observed with the first derivative spectrum in this area. Traditional REP (the maximum value of the fist derivative spectra in 670–780 nm range) was not sensitive to canopy leaf nitrogen concentration because of the three-peak feature, but the REPs based on inverted Gaussian fitting, linear four-point interpolation, linear extrapolation and adjusted linear extrapolation generated continuous REP data, and could be used to estimate canopy leaf nitrogen concentration. Besides, REP from a three-point Lagrangian interpolation with three first-derivatives bands (695, 700 and 705 nm) also had a good relationship with canopy leaf nitrogen concentration. Comparison of these REPs based on different approaches indicated that the adjusted linear extrapolation method (755FD₇₃₀+675FD₇₀₀) / (FD₇₃₀+FD700) gave the best prediction of canopy leaf nitrogen concentration, with relative simple algorithm, and thus is a good REP parameter for monitoring canopy leaf nitrogen concentration in rice.