林分垂直结构降雨减蚀功能遥感量化模拟

Leaf area index (LAI) of single vertical observations obtained cannot be able to quantify functional differences of trees canopy vertical structure in the rainfall reducing cavitation. On the basis of National Natural Science Foundation of China (41401385), LAI and leaf area density (LAD) of trees canopy obtained by information processing of terrestrial laser scanning and Unmanned Aerial Vehicle (UAV) multi-angle remote sensing are as tree canopy vertical structure parameters. Set up undergrowth vegetation cover patterns and hidden indicators to quantify expression and comparation, and then construct the remote sensing composite determining factors that can reflect the undergrowth vegetation cover to be instead of determining undergrowth vegetation cover condition. In addition, by measuring in place, revealing rainfall reducing cavitation process of forestry LAI and LAD are as canopy rainfall reducing cavitation rules. Also, rainfall erosion reduction rules of stand vertical structure can be systematacially clarified according to forestry cover rainfall reducing cavitation rules measured in the runoff plot. Therefore, based on the above, rainfall reducing cavitation function can be inversed by stand LAI, LAD and undergrowth vegetation cover according to clarified rainfall erosion reduction rules of stand vertical structure; meanwhile, remote sensing quantitative simulation of rainfall reducing cavitation function of stand vertical structure can be achieved by simulating laboratory rainfall and regional test validation analysis, that can provide the theoretical foundation and technical reference for scalability soil erosion forecast of remote sensing.

单一垂直观测获取的LAI（叶面积指数）未能量化出各林木冠层垂直结构在减缓降雨侵蚀的功能差异！在前期青年基金研究基础上，通过⑴地面激光扫描+无人机多角度遥感的信息处理，获取林冠层的LAI、LAD（叶面积密度）作为林冠层垂直结构参数；⑵设置林下植被覆盖模式+隐含指示因子量化表达比较，从差异性、互补性出发，构建可体现林下植被覆盖遥感复合判定因子，替代判定林下植被覆盖状况；⑶借助①现场测定、揭示林木的LAI、LAD的降雨减蚀过程作为林冠层降雨减蚀规律，②径流小区测定的林地覆盖物的降雨减蚀规律，系统阐明林分垂直结构降雨减蚀规律；⑷利用遥感反演出林分LAI、LAD和林下植被覆盖物，根据阐明的林分垂直结构降雨减蚀规律，反演出区域植被覆盖的降雨减蚀功能，并通过室内降雨模拟+区域实测验证分析，实现林分垂直结构降雨减蚀功能的遥感量化模拟，为遥感的尺度性水土流失的预测预报提供理论基础和技术借鉴。

针对单一垂直观测获取的叶面积指数未能量化出各林木冠层垂直结构在减缓降雨侵蚀的功能差异的问题，项目以南方红壤水土流失典型区福建省河田镇的马尾松林为研究对象，（1）利用无人机高重叠率遥感影像，基于摄影测量法获取不同角度方案(O (0°)、T15 (15°)、T30 (30°)、OT15 (0°+15°)、OT30 (0°+30°))下的林冠层点云数据，实现不同角度林冠层三维点云重构，在重构点云的基础上，使用基于体素的冠层分析法反演不同角度的叶面积密度（LAD），分析林冠层LAD随角度变化的遥感量化表达差异，结果表明，正射影像与30°倾斜影像的耦合得到的三维点云在子体素边长设置为9cm时能够较为准确地估计马尾松林冠层LAD；（2）基于随机森林模型估测叶面积指数（LAI），总体精度为81.95%，结合实测马尾松林降雨再分配特征，分析林冠层LAI、LAD降雨减蚀功能的差异，构建了林冠降雨截留空间模拟技术；（3）基于碳平衡理论，利用改进CASA模型和CENTURY模型实现了马尾松林土壤基础呼吸（SBR）的遥感估测，估测精度为84.33%；（4）选择土壤基础呼吸、氮素反射指数和黄叶因子3个指标作为林下植被覆盖状况判定指标，通过主成分分析，构建林下植被覆盖遥感复合评判指数的遥感间接估测模型，模型精度为82.62%，结合径流小区实测数据，揭示林下植被覆盖降雨减蚀机制；（5）基于马尾松林分冠层降雨截留机制和林下植被覆盖降雨减蚀机制，利用径流小区观测数据，研究马尾松林分垂直结构降雨减蚀整体过程。结果表明，研究构建的林分垂直结构降雨减蚀功能遥感量化技术体系可有效地模拟马尾松林的降雨截留机制，较传统的植被覆盖度模型方差解释率高27%；（6）培养了1位博士，6位硕士，发表相关学术论文25篇。本研究成果可为遥感的尺度性应用及更为精准、定量化描述植被覆盖对降雨减蚀的作用与功能提供理论基础和技术指导。