Please use this identifier to cite or link to this item:
標題: 數值地形模型應用在合歡山地區主要林型空間型態之分析
Application of DTM to the Spatial Pattern Analysis for Major Forest Types in the Hohuan Mountain Area
作者: 黃慧欣
Huang, Hui-Hsin
關鍵字: GIS;地理資訊系統;DTM;Habitat Suitability Analysis;數值地形模型;適生育地分析
出版社: 森林學系
【摘 要】由於森林資源急遽減少,人們已察覺永續發展之必要性,因此確實掌握森林與環境間的互動是很重要的。 在1980年代中期以前,有關森林生態之研究因受限於傳統技術多定性描述而少空間計量分析;直到電腦軟硬體及遙測技術大幅躍進並普及後,對森林生態採廣域空間計量分析才逐漸出現。本研究以地理資訊系統技術分析數值地形模型(Digital Terrain Model, DTM)資料,其目的係針對合歡山地區的七種主要林型─冷杉、鐵杉、雲杉、檜木、松類天針、天針闊混淆林及松類造林,探討它們的生育地在海拔、坡度、坡向、坡面位置及地形遮蔽度等五地文因子上之分佈特性,並決定各林型於五種因子上之相對重要性及其間差異。本研究將土地利用數值檔與DTM資料套疊,分別產生全區與七種林型在五個地文因子之統計值與累計頻率圖,並將其逐個分別比較,以了解七種林型在各因子上分佈之特性。 同時本研究從各林型以隨機方式分別抽取十套各含200個像元之林型樣本及一套含600個像元之背景樣本,並完成卡方比例同質性檢定及K-S雙樣本檢定(Kolmogorov-Smirnovr Two Sample),以探討各因子之相對重要性。 研究結果顯示在五個地文因子中,海拔及地形遮蔽度對於七種林型生育地之相對重要性最高,而坡度、坡向及坡面位置者在伯仲之間,但遠低於海拔者;此結論亦與Kellman環境層級系統--海拔在最上層及地形位置與方位在中層之觀念相互切合。 隨著海拔之降低,海拔及地形遮蔽度之相對重要性有隨之減小的趨勢,坡向之相對重要性反而有增強之趨勢,坡度及坡面位置者則無明顯之規律變化。 後續研究仍需針對其他試區深入驗證上述結論之空間外延性。

AbstractDue to rapid reduction in the forest resources, people have become aware the necessity of sustainable development. Therefore, it is extremely important to have a thorough grasp of the relationships between forest and its environmental factors. Before the middle of 1980s, many studies in forest ecology involved more qualitative descriptions than quantitatively spatial analyses owing to the restraints of traditional technologies. Not until a great leap forward in computer and remote sensing technologies and widespread availability of these technologies and facilities during 1980s, has research in forest ecology involving quantitatively spatial analyses over large area progressively appeared. The objective of this study was to investigate the characteristics of distribution in elevation, slope, aspect, terrain position, and topographic sheltering index of the seven forest types (including fir type, hemlock type, spruce type, cypress type, pine type, conifer-hardwood forest type, and pine-plantation) in the Hohuan mountain area and to determine the relative weights of the seven forest types in five physiographic factors. It was achieved by using GIS techniques to overlay the digital land-use map and digital terrain model (DTM) data layer and then analyze the composite data layer. The statistics and cumulative frequency graph for five physiographic factors of the entire study area and areas of seven forest types were generated from the composite data layer. This study randomly selected six sets of samples with 200 pixels from each forest type and a set of samples with 600 pixels from each background corresponding to that forest type. Chi-square tests and K-S tests were performed to examine the relative weights among five factors for each forest type and difference in each factor among seven forest types. As results indicated, the relative weights in elevation and topographic sheltering index were the highest among five physiographic factors, and the relative weights in slope, aspect, and terrain position were similar, but much lower than those in elevation and topographic sheltering index. This conclusion dovetailed Kellman's concept of a hierarchical system of environment. As elevation declined, the relative weights in elevation and topographic sheltering index had a tendency to decrease; however, the relative weight in aspect had a tendency to increase; the relative weights in slope and terrain position did not change with the decrease or increase in elevation. Further studies will be needed to confirm the spatial extension of the conclusions mentioned above in other study areas.
Appears in Collections:森林學系

Show full item record

Google ScholarTM


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.