Combining Two-surface Method with LiDAR Technology to Estimate Leaf Area of Moso Bamboo

• Main Authors: Yan-You Chen, 陳彥佑
• Other Authors: Tzung-Su DING
• Format: Others
• Language: zh-TW
• Published: 2016
• Online Access: http://ndltd.ncl.edu.tw/handle/83226423353546761494

碩士 === 國立臺灣大學 === 森林環境暨資源學研究所 === 104 === This study was aimed to combine two-surface method with LiDAR technology, and to construct an equation which included envelope surface, canopy fractal dimension and total leaf area of Moso bamboo ({sl Phyllostachys edulis}). Except of canopy fractal dimension, I also compared five methods of estimating bamboo envelope surface, which includes cone-shape method, paraboloid-shape method, double-ellipsoid-shape method, convex hull method and canopy analysis method. Results showed that, by combining two-surface method with LiDAR, the coefficient of determination in linear regression without the intercept term could reach up to 0.85. By correlating with total leaf area, convex hull method aided by LiDAR performed best in estimating envelope surface, seconded by paraboloid-shape method of manual measurement. This indicates estimating envelope surface by manual measurement could have similar performance with methods estimating by LiDAR. Mean fractal dimension of Moso bamboo was 2.186, close to the estimated 2.2 for dominant trees in previous studies. One could consider using 2.2 as fractal dimension for rough estimates of leaf area for moso bamboo in the future. I conclude two-surface method can be applied to estimate leaf area of plants and is especially suitable for single plants or monocultural plantations. Estimation aided by LiDAR should provide better results. However, the LiDAR measurement could be replaced by using fractal dimension estmates for similar plant species and manual measurements of envelope surface.