Nitrogen is an essential nutrient that directly affects plant photosynthesis, crop yield, and biomass production for bioenergy crops, but excessive application of nitrogen fertilizers can cause environmental degradation. To achieve sustainable nitrogen fertilizer management for precision agriculture, there is an urgent need for nondestructive and high spatial resolution monitoring of crop nitrogen and its allocation to photosynthetic proteins as that changes over time. Here, we used visible to shortwave infrared (400–2400 nm) airborne hyperspectral imaging with high spatial (0.5 m) and spectral (3–5 nm) resolutions to accurately estimate critical crop traits, i.e., nitrogen, chlorophyll, and photosynthetic capacity (CO₂-saturated photosynthesis rate, V_max,27), at leaf and canopy scales, and to assess nitrogen deficiency on crop yield. We conducted three airborne campaigns over a maize (Zea mays L.) field during the growing season of 2019. Physically based soil-canopy Radiative Transfer Modeling (RTM) and data-driven approaches i.e. Partial-Least Squares Regression (PLSR) were used to retrieve crop traits from hyperspectral reflectance, with ground truth of leaf nitrogen, chlorophyll, V_max,27, Leaf Area Index (LAI), and harvested grain yield. To improve computational efficiency of RTMs, Random Forest (RF) was used to mimic RTM simulations to generate machine learning surrogate models RTM-RF. The results show that prior knowledge of soil background and leaf angle distribution can significantly reduce the ill-posed RTM retrieval. RTM-RF achieved a high accuracy to predict leaf chlorophyll content (R² = 0.73) and LAI (R² = 0.75). Meanwhile, PLSR exhibited better accuracy to predict leaf chlorophyll content (R² = 0.79), nitrogen concentration (R² = 0.83), nitrogen content (R² = 0.77), and V_max,27 (R² = 0.69) but required measured traits for model training. We also found that canopy structure signals can enhance the use of spectral data to predict nitrogen related photosynthetic traits, as combining RTM-RF LAI and PLSR leaf traits well predicted canopy-level traits (leaf traits × LAI) including canopy chlorophyll (R² = 0.80), nitrogen (R² = 0.85) and V_max,27 (R² = 0.82). Compared to leaf traits, we further found that canopy-level photosynthetic traits, particularly canopy V_max,27, have higher correlation with maize grain yield. This study highlights the potential for synergistic use of process-based and data-driven approaches of hyperspectral imaging to quantify crop traits that facilitate precision agricultural management to secure food and bioenergy production.

氮是一种直接影响植物光合作用、作物产量和生物能源作物生物质生产的必需营养素，但过量施用氮肥会导致环境恶化。为了实现精准农业的可持续氮肥管理，迫切需要对作物氮及其随时间变化的光合蛋白质的分配进行无损和高空间分辨率监测。在这里，我们使用具有高空间（0.5 m）和光谱（3-5 nm）分辨率的可见光至短波红外（400-2400 nm）机载高光谱成像来准确估计关键作物性状，即氮、叶绿素和光合能力。 CO _{2 -}饱和光合作用速率，V _max,27)，在叶和冠层尺度上，并评估氮缺乏对作物产量的影响。我们在 2019 年的生长季节对玉米 ( Zea mays L.) 田地进行了三场空中运动。基于物理的土壤冠层辐射传递模型 (RTM) 和数据驱动的方法，即偏最小二乘回归 (PLSR) 用于从高光谱反射率中检索作物性状，以及叶氮、叶绿素、V _{max,27 的}基本事实、叶面积指数 (LAI) 和收获的谷物产量。为了提高 RTM 的计算效率，随机森林 (RF) 用于模拟 RTM 模拟以生成机器学习代理模型 RTM-RF。结果表明，土壤背景和叶角分布的先验知识可以显着减少不适定的RTM反演。RTM-RF 实现了高精度预测叶绿素含量 (R ²  = 0.73) 和 LAI (R ²  = 0.75)。同时，PLSR 在预测叶绿素含量 (R ²  = 0.79)、氮浓度 (R ²  = 0.83)、氮含量 (R ²  = 0.77) 和 V _max,27 (R ² = 0.69）但需要模型训练的测量特征。我们还发现冠层结构信号可以增强对光谱数据的使用来预测与氮相关的光合性状，因为结合 RTM-RF LAI 和 PLSR 叶性状可以很好地预测冠层级性状（叶性状 × LAI），包括冠层叶绿素（R ²  = 0.80)、氮 (R ²  = 0.85) 和 V _max,27 (R ²  = 0.82)。与叶片性状相比，我们进一步发现冠层光合性状，尤其是冠层 V _max,27, 与玉米籽粒产量有较高的相关性。这项研究强调了协同使用基于过程和数据驱动的高光谱成像方法来量化作物性状的潜力，这些性状有助于精准农业管理以确保粮食和生物能源生产。