基于光谱技术的橡胶树白粉病诊断机理与方法研究

Timely and accurately obtain rubber tree powdery mildew information is very important to effectively prevention and control the development of powdery mildew. The existing acquisition method of rubber tree disease information is still manual survey method, which is limited by manpower, material resources, cost and other conditions, so it is difficult to meet the needs of real-time monitoring and prediction of large-scale diseases. This project aims to develop a new rubber tree disease information monitoring method with the spectral technology. The main contents include: (1)analyzing the spectral characteristics of rubber leaf under different powdery mildew degree and phenological periods by conducting powdery mildew inoculation experiment experiments, to understand the mechanisms of rubber powdery mildew; (2) based on the analysis of the monitoring mechanisms with powdery mildew, extracting spectral features and constructing spectral index to monitor powdery mildew degree at canopy scale; (3) fully exploring the prior knowledge of new satellites and field measurements to optimize the spectral index and develop optimal model for the powdery mildew degree estimation, and then accurately monitor the powdery mildew degree with satellite remote sensing observations. Our research results can provide a theoretical basis for prevention and control powdery mildew, and also expand the application depth and breadth of agriculture remote sensing technology.

及时、准确、大面积的获取胶林白粉病病害信息对于有效制定白粉病防治措施、保障胶林产量健康稳定持续增长具有重要意义。现有橡胶树病害信息获取方法仍旧是人工目测手查法，受人力、物力、成本等条件限制，难以满足目前大范围的病害实时监测和预报的需求。本项目拟以人工控制接种实验和大田星地同步调查实验为基础，基于光谱技术开展橡胶树白粉病遥感监测研究。（1）通过盆栽橡胶树白粉病接种实验，分析在不同白粉病严重度、不同物候期条件下的叶片光谱特征，阐明橡胶树白粉病的光谱诊断机理；（2）基于橡胶树冠层光谱数据，分析白粉病的敏感光谱特征和时相特征，采用代数组合方式，构建白粉病严重度遥感监测的光谱指数；（3）针对新发卫星传感器特性优化光谱指数，并结合白粉病发病先验知识构建适宜新发卫星的橡胶树白粉病严重度遥感定量反演模型。研究成果可为胶林白粉病科学防控提供准确理论依据，同时也对农业遥感技术应用深度和广度的拓展。

及时、准确、大面积的获取胶林白粉病病害信息对于有效制定白粉病防治措施、保障胶林产量健康稳定持续增长具有重要意义。本项目以人工控制接种实验和大田调查实验为基础，基于光谱技术开展橡胶树白粉病遥感监测研究。通过本项目研究，阐明了叶片不同白粉病病情等级光谱响应特征，提出了白粉病病情等级的敏感光谱指标，建立了基于光谱技术的叶片白粉病病情等级反演模型；综合利用橡胶树不同物候期的光谱特征、植被指数特征、纹理特征和第一主成分特征，提出了面向对象的橡胶资源遥感高精度监测方法；在上述研究基础上，针对新发卫星Planet影像波段特性，筛选出适宜光谱指标，构建了基于Planet数据的橡胶白粉病病情指数遥感反演模型，开展了区域尺度橡胶白粉病病情指数遥感监测研究；设计开发了橡胶树白粉病病情快速评价系统；构建了橡胶树白粉病光谱数据库。此外，发表学术论文 3 篇（SCI 2 篇，EI 1 篇），获批国家软件著作权 1 项，培养研究生 1 名。研究成果可为胶林白粉病科学防控提供准确理论依据，同时也对农业遥感技术应用深度和广度的拓展。