應用空載光達推估阿里山地區柳杉人工林林分材積

Using Airborne LiDAR Data to Estimate the Stand Volume of Cryptomeria japonica in the Alishan Area

魏浚紘(Chun-Hung Wei)

指導教授 ： 陳朝圳

國立屏東科技大學/農學院/森林系所/碩士(2008年)

https://doi.org/10.6346/NPUST.2008.00223

摘要

傳統的森林資源調查往往耗費許多人力與物力。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資源。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。國外有許多文獻指出利用光達可以精確地針對胸徑、樹高、樹冠幅、材積、林分結構與生物量等進行量測。國內目前處於起步階段，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。透過自動化方式獲得光達特徵值，並與林分性態值進行不同尺度之相關分析。當空間尺度增大時，特徵因子對林分材積的相關性均會逐漸增大。由結果得知，透過空載光達特徵值所建立之林分材積相關推估模式，以20 m × 20 m為取樣大小，獲得之1m CHM之空間解析力之林分高配合林分密度、第一回波比例與點雲強度值，對林分材積推估，能獲得較佳之迴歸式R2=0.8210。 空載光達與傳統航照在推估林分材積上，有許多的不同。傳統航照必須透過專業的人員訓練才能獲取準確的資訊，並且需要利用雙重取樣的方式，才可以推估全林材積。空載光達則是利用光達所拍攝的影像進行處理，將不正確的點雲資料排除後，後續分析動作均以電腦自動化的方式產生，對於專業人員的訓練部份較為一致性，其推估過程不需要透過雙重取樣的方式即能獲得全林材積量。

關鍵字

空載光達 ； 林分性態 ； 林分材積 ； 柳杉

並列摘要

Traditional forest resource surveys often consume lots of manpower and resources. With the improvement of the remote sensing technique, the usage of high resolution images and laser scanning systems could conserve many resources. LiDAR systems can be categorized into three types which are Airborne Laser Scanning Systems, Terrestrial Laser Scanners, and Space-borne Large Altimeter Systems. Many references indicate that LiDAR systems could monitor DBHs (Diameter of Breast Heights), canopy heights, tree crown delineations, volumes, stand structures and biomass accurately. Though the scholars in Taiwan are still at the stage of exploring the method of LiDAR, it is a tendency to exploit the LiDAR systems for the survey of forest resources. Traditionally, we estimate stand volumes by analogically interpreting average stand heights, average crown diameters and crown closures based on air pictures. But now we can obtain average stand heights, crown diameters and crown closure by auto-computerized information in virtue of the two characteristics, 3D space and multiple-returns of the Airborne LiDAR since its development. In this study, we investigated the single tree data in Alishan area using airborne LiDAR data to estimate average stand height, average crown diameter and crown closure, and analyzed the data of the stand volume from the airborne LiDAR for the suitable raster cell size and estimation model. We obtained LiDAR characteristic values from auto-computerized information, and applied associated analysis to stand characteristics in different spatial scales. The results showed when the spatial scales increased gradually, associations between characteristic values and stand characteristics were in direct proportion. When the LiDAR raster grid size was 20 m × 20 m, the best multiple regression (R2=0.8210) was obtained from stand density, first echo ratio and intensity by using canopy height model. There were many differences between Airborne LiDAR and air pictures when estimating stand volumes. There must be professionally trained personnel analyzing the traditional air-photos in order to collect the accurate information. Also we need to use the method of double-sampling to estimate the whole forest volume. On the other hand, the Airborne LiDAR system processes the images taken by LiDAR, and after eliminating the incorrect point clouds, the following analysis is automatically produced by the computer system. It is more consistent than the analysis done by professionally trained personnel because the estimating process does not have to go through the method of double-sampling in order to extract the volume of woods in the whole forest.

並列關鍵字

Airborne LiDAR ； Stand Structure ； Stand Volume ； Cryptomeria japonica

參考文獻

施瑩瑄 (2005) 應用空載光達獲取林分高之研究。國立台灣大學森林環境暨資源學系研究所碩士論文，77頁。

謝漢欽 (1998) SPOT綠度與生林生育地因子於台灣杉材積推估之探討。台灣林業科學 13(3): 175-188。

陳永寬、詹進發、葉堃生、鄭祈全、施瑩瑄 (2005) 應用地面雷射掃描儀測算樹高之研究。航測及遙測學刊 10(4): 327-336。

Magnussen, S., P. Eggermont, and V. N. LaRiccia (1999) Recovering tree heights from airborne laser scanner data. Forest Science 45: 407-422.

詹進發 (2005) 空載雷射掃描資料於推估樹冠高度之應用。航測及遙測學刊 10(1): 1-14。