傳統的森林資源調查往往耗費許多人力與物力，近年來遙感探測技術的精進，利用高解析力影像配合主動式偵測光達系統，進行森林資源的調查可節省許多人力與物力。光達的種類大約可分成三種，空載雷射掃描系統、地面雷射掃描系統與衛載雷射掃描系統。Means et al. (1999) 指出利用光達可以有效的推估胸徑、樹高、樹冠幅、材積、林分結構與生物量等，而國內目前處於起步階段，未來利用光達系統進行森林資源調查是一大趨勢。傳統上林分材積係以人工判釋航
照之方式獲取平均林分高、平均樹冠直徑與樹冠密度進行推估。由於空載光達具有三維空間與多重反射特性，因此可利用電腦自動化方式獲得平均林分高、平均樹冠直徑與樹冠密度等資訊。本研究以阿里山地區調查所得之樣區立木基礎資料，利用空載光達資料獲取其平均林分高、平均樹冠直徑與樹冠密度資料，並分析空載光達林分材積最適取樣大小與推估模式。研究結果顯示，當CHM 為空間解析力1 m 時，用20 m× 20 m 取樣大小，獲得之林分高配合林分密度、第一回
波比例與點雲強度值，對林分材積進行迴歸分析，能獲得較佳之相關性(R2=0.8210)。

The traditional forest resources survey requires vast amount of manpower and resources. With an improvement of remote sensing technique using high resolution images and laser scanning systems, forest resources surveys can save large amounts of manpower and resources. LiDAR systems can be divided into three categories:airborne laser scanning systems,terrestrial laser scanning systems, and spaceborne large altimeter systems. Means et al. (1999) indicated that LiDAR can precisely estimate DBH, canopy height, tree crown delineation, volume, stand structures and biomass. LiDAR systems have just begun to be used in Taiwan, and will become the
trend for surveys of forest resources. In conventional systems the stand volume is estimated by manual interpretation based on the average stand height, the average crown diameter and the crown closure, which are acquired from analogical pictures.Due to the 3D space and multiple-returns characteristics of airborne LiDAR, we can obtain information of average stand height, average crown diameter and crown closure through computerized automatic analysis. We present the use of airborne
LiDAR data to obtain average stand height, average crown diameter and crown closure, analyze the airborne LiDAR stand volume of suitable raster cell size and estimate the model by single tree data acquired from the Alishan area. The results of multiple regression analysis showed a significant correlation (R2=0.8210) between manual stand volume and LiDAR data based on average stand height, stand density,first echo ratio and intensity in the condition of canopy height model on a 1 meter by
1 meter grid and the LiDAR raster on a 20 meter by 20 meter grid.