基于机载多基线SAR的森林地上生物量三维精准估测

森林地上生物量（AGB）是评估森林生态系统功能的重要测度指标，开展全面准确的森林AGB估测研究，有助于深入了解森林在碳循环过程中的地位和作用。机载多基线SAR能够有效地探测森林垂直结构，进而建立森林AGB估测模型。但已有估测方法大多聚焦于水平方向（二维空间），无法同时实现水平和垂直方向的森林AGB估测（三维空间）。针对这一问题，本项目拟利用机载多基线SAR数据，1）研究基于深度压缩感知的森林垂直结构重建方法，实现水平与垂直方向散射功率的高分辨，降低计算复杂度；2）发展三维极化目标精细分解算法，实现森林垂直散射机制的精细分离；3）基于森林垂直结构剖面与精细的散射机制构建兼顾水平与垂直结构的森林AGB三维估测模型，确定散射体类型及其垂直分布范围，定量计算水平与垂直方向森林AGB，并将二者耦合实现森林AGB精准估测。成果有助于推动森林生态系统碳储量及碳循环精细研究，具有重要的科学意义与应用价值。

Forest aboveground biomass (AGB) is an important indicator to measure forest ecological function. The comprehensive and accurate estimation for forest AGB will contribute to a better understanding of the role of forests in the carbon cycle. Airborne multi-baseline Synthetic Aperture Radar (SAR) can detect the forest vertical structure information effectively, and then accurate forest estimation model is built. Therefore, airborne multi-baseline SAR has been applied in forest AGB estimation. However, most of the estimation approaches focus on forest AGB estimation in spatial direction (two-dimensional space), rarely can be applied in forest AGB estimation both in spatial and vertical direction (three-dimension space). To solve this problem, we intend to use airborne multi-baseline SAR to conduct the research. (1) We consider deep compressive sensing based reconstruction method for forest vertical structure to get higher resolution both in spatial and vertical direction, and to reduce computational complexity. (2) Three-dimensional polarimetric target decomposition algorithm is adopted to realize fine analysis and separation of forest vertical scattering mechanism. (3) We will construct a forest AGB estimation model considering both horizontal and vertical structures based on forest vertical structure profile and fine scattering mechanism, and it is applied to quantitatively evaluate the effect of forest vertical structure profile and corresponding scatterer type on AGB. The results will be helpful to promote the fine research of carbon storage and carbon cycle in forest ecosystem, and have important scientific significance and application value.