综合使用光谱技术对作物养分进行实时、 有效诊断, 有助于作物的精准管理、 保障产量和减少环境污染, 提高肥料利用率, 并且为定量估测作物生化组分状况提供了一种新的途径。 光谱指数是进行作物叶片叶绿素实时估测的重要指标, 然而由于受到环境条件及内在生化成分的影响, 估测结果不尽满意。 为了进一步提高光谱指数在估测作物叶片叶绿素含量时的抗干扰能力和敏感性, 于2020年在内蒙古玉米种植典型区域进行不同氮梯度的田间试验, 在玉米的四个关键生育时期获取叶片的光谱反射率和叶绿素值, 通过建立基于面积的光谱指数和叶片叶绿素值的关系模型并进行光谱指数的优化及评价。 结果表明, 生育时期对面积光谱指数与叶片叶绿素值的关系有显著影响。 前人研究的基于面积的光谱指数在玉米苗期时对于叶片叶绿素含量的估测效果较差, 而对抽雄期叶片叶绿素含量的估测效果最佳。 基于优化算法构建的面积光谱指数显著提高了光谱指数对叶片叶绿素含量估测的准确度和稳定性, 基于优化算法的优化三角形植被指数(OTVI)、 优化叶绿素吸收积分指数(OCAI)和优化双峰面积归一化差值指数(ONDDA)在不同生育时期上比前人研究的面积光谱指数具有更强的叶绿素含量估测能力, 估测模型的决定系数R²在0.94~0.99之间。 与优化三角形植被指数(OTVI)和优化叶绿素吸收积分指数(OCAI)相比优化双峰面积归一化差值指数(ONDDA)在估测春玉米不同生育时期叶片叶绿素含量方面更为稳定, 预测模型验证结果的决定系数R²为0.94, 并且验证误差最小, RMSE和NRMSE%分别为2.29%, 3.94%, 模型估测值与实测值的验证斜率为0.996, 接近1。 综上所述, ONDDA是一个实用且适合于估测不同生育时期叶片叶绿素含量的面积光谱指数。

Spectral index is an important means for real-time estimation of crop leaf chlorophyll. The comprehensive use of spectral technology for real-time and effective diagnosis of crop nutrients is conducive to accurate crop management, ensuring yield and reducing environmental pollution, improving fertilizer utilization, and providing a new way for quantitative estimation of crop biochemical components. However, the estimation results are not satisfactory due to the influence of environmental conditions and internal biochemical components. In order to further improve the anti-interference ability and sensitivity of spectral index in estimating chlorophyll content of crop leaves. In this study, field experiments with different nitrogen gradients were carried out in typical corn-growing areas of Inner Mongolia in 2020. The spectral reflectance and chlorophyll value of leaves were obtained at four key growth stages of corn. The relationship model between the spectral index and chlorophyll value of leaves was established based on area, and the spectral index was optimized and evaluated. It provides an important theoretical basis for the diagnosis of chlorophyll content in maize leaves and an accurate grasp of the nutritional status of crops in a larger area in the future. The results showed that the growth period significantly affected the relationship between area spectral index and leaf chlorophyll value. The published area-based spectral index had a poor estimation effect on leaf chlorophyll content at the seedling stage, but had the best estimation effect on the tasseling stage. In this paper, the area spectral index based on the optimization algorithm significantly improves the accuracy and stability of spectral index in Estimating Leaf Chlorophyll content. The optimized triangle vegetation index (OTVI), optimized chlorophyll absorption integral index (OCAI) and optimized bimodal area normalized difference index (ONDDA) based on the optimization algorithm have stronger performance than the published area spectral index at different growth stages, the coefficient of determination R² is between 0.94 and 0.99. Compared with OTVI and OCAI, ONDDA is more stable in estimating the chlorophyll content of spring maize leaves at different growth stages. The coefficient of determination R² of prediction model validation results is 0.94, and the validation error is the smallest, RMSE and RE% are 2.29% and 3.94%, respectively. The validation slope of the model estimated value and the measured value is 0.996, the closest to 1. In conclusion, ONDDA is a practical and suitable area spectral index for estimating leaf chlorophyll content at different growth stages.