联合激光雷达和多角度影像数据的森林冠层和林下植被LAI反演研究

森林背景是影响森林冠层LAI遥感反演精度的主要原因之一，但现有估算森林背景反射率的方法并未考虑复杂的森林垂直结构，尤其是林下植被。另外，森林冠层和林下植被的固碳速率不同，现有森林冠层LAI反演方法未考虑林下植被的分离，影响碳水循环模拟。为此，本项目拟开展森林背景反射率和林下植被LAI反演研究，通过“空-地”一体化同步观测，分析不同森林类型样方的森林背景光谱特征，建立顾及森林复杂垂直结构的森林背景反射率估算模型；构建林下植被LAI反演模型，优化森林背景反射率→林下植被光谱→林下植被LAI的两步反演方法；基于ALS数据提取几何光学模型所需的关键结构参数，定量驱动几何光学模型，提高森林冠层和林下植被LAI的反演精度；并通过实测森林冠层和林下植被LAI数据进行验证。所发展和建立的森林背景反射率和林下植被LAI反演方法，为提高森林LAI产品精度提供有效工具

The forest background is one of the main factors that affect the retrieval accuracy of the forest overstory leaf area index (LAI). However, it is still not good enough for the existing method for retrieving forest background reflectance, since they don’t considering the effect of complex vertical structure of the forest. In addition, forest overstory and understory layers differ in carbon and water cycle regimes. Separate retrievals of LAI for these two layers would help to improve modeling forest biogeochemical cycles and evaluating forest ecosystem functions. This proposed project will research on the key methods of retrieving forest background reflectance and understory vegetation LAI using remote sensing data. Simultaneous observation of multi-angle image data, aerial lidar data, LAI and leaf reflectance will be carry out to investigate the relationships of forest structure, forest background reflectance and understory vegetation LAI parameters. Constructing a forest background reflectance inversion model that takes into account the complex forest structure in vertical direction, and developing a two-step inversion models for retrieval of understory vegetation leaf for improving the inversion accuracy. In addition, we also quantify the structure parameters required by the geometric optical model based on lidar data for improving the accuracy of forest overstory LAI. The proposed model for retrieving the forest background reflectance and understory vegetation LAI will provide a powerful and effective tool for promoting the development of quantitative remote sensing technology for vegetation parameters.