基于高光谱技术的小麦氮高效品种识别研究

Nitrogen is the first major mineral element necessary for the growth and development of crops. It is of great significance to select nitrogen efficient varieties and improve the efficiency of crop nitrogen utilization to ensure the national food security and reduce the agricultural non-point source pollution. To a certain extent, the traditional nitrogen-efficient variety screening methods damaged crops. Another, it was also the high cost, time-consuming and labor-intensive. This research was carried out by hyperspectral remote sensing technology at National Soil Fertility and Fertilizer Effects fluvo-aquic long-term monitoring Station. We selected wheat as a representative crop, and taking 30 wheat varieties of different genotypes with significant difference in nitrogen utilization efficiency as the study objects, which were screened from the previous study. First, we evaluated the applicability of agronomic parameters that are closely related to nitrogen utilization. Second, the high spectral inversion model of agronomic parameters was selected based on correlation between agronomic parameters and spectral vegetation indices. According to the quantitative relationship between the nitrogen utilization efficiency and agronomic parameters and the optimized hyperspectral inversion model of agronomy parameters, the high spectral inversion model of nitrogen utilization efficiency of wheat was constructed with the model nesting method. The nitrogen-efficient wheat varieties were screened. That will provide theoretical basis and technical support for the selection of nitrogen-efficient varieties for the main crop in our country.

氮素是作物生长发育所必需的第一大矿质营养元素。筛选氮高效品种，提高作物氮素利用效率，对于保障国家粮食安全、降低农业面源污染具有十分重要的意义。针对传统氮高效品种筛选方法中存在的破坏性、分析成本高和费时费工的问题，本项目利用高光谱遥感技术，以国家潮土土壤肥力与肥料效益长期监测站为平台，以小麦为代表性作物，以前期筛选的氮素利用效率有显著差异的30个不同基因型小麦品种为研究对象，在对与氮素利用效率关系密切的农学参数适用性评估基础上，挑选出能够表征氮素利用效率的农学参数；基于农学参数与光谱植被指数的相关性优选出小麦高光谱反演模型；基于小麦氮素利用效率与农学参数的定量关系以及优选的农学参数高光谱反演模型，利用模型嵌套方法，构建小麦氮素利用效率高光谱反演模型，筛选出氮高效小麦品种，以期为我国主要农作物氮高效品种的筛选提供理论依据和技术支撑。

氮素是作物生长发育所必需的第一大矿质营养元素。筛选氮高效品种，提高作物氮素利用效率，对于保障国家粮食安全、降低农业面源污染具有十分重要的意义。针对传统氮高效品种筛选方法中存在的破坏性强、分析成本高和费时费工的问题，本项目利用高光谱遥感技术，以国家潮土土壤肥力与肥料效益长期监测站为平台，以前期筛选的氮素利用效率有显著差异的30个不同基因型小麦品种为研究对象，主要开展以下研究：①通过分析不同小麦品种的氮素利用效率与不同生育期农学参数（如氮含量、地上部生物量、叶面积指数、叶绿素含量等）之间的关系，评估不同生育期小麦农学参数在预测氮素利用效率方面的适用性及贡献大小，确定小麦氮素利用效率的重要农学参数指标、组合形式及最优评估生育期；②对供试小麦品种不同生育期的冠层高光谱数据进行监测，并利用相关分析、逐步多元回归等技术确定各农学参数的敏感波段范围，进而优选出小麦农学参数高光谱反演模型；③以农学参数为衔接点，利用优选的光谱植被指数构建小麦氮素利用效率高光谱反演模型，对30个小麦品种的氮素利用效率进行反演，并根据实测的小麦氮素利用效率数据对模型进行验证；在此基础上，根据“低氮高效”和“高氮高效”小麦品种的氮素利用效率评判标准，结合反演结果，对小麦氮高效品种进行筛选与分类。上述项目成果为实现粮食安全生产和资源节约化应用开辟了新途径，为提升作物生产的信息化水平和科技赋能农业能力提供了助动力，为我国主要农作物氮高效品种的筛选提供了理论依据和技术支撑。.本项目发表论文8篇，获得软件著作权11项，实用新型5项，申报团体标准1个，申报新品种保护权2个，培养研究生1名，本科生1名。