運用空間技術與景觀指標分析清境地區景觀變遷之研究
No Thumbnail Available
Date
2012-11-??
Authors
李瑞陽
陳坤鈺
Re-Yang Lee
Kun-Yu Chen
Journal Title
Journal ISSN
Volume Title
Publisher
地理學系
Department of Geography, NTNU
Department of Geography, NTNU
Abstract
清境地區因擁有豐富自然、氣候等資源，以及國內遊憩風氣興起，使清境地區成為臺灣熱門觀光景點。然而，大量且快速的人為開發行為易對當地自然環境造成衝擊，因此監測當地景觀變遷程度，並探討開發行為對當地造成之影響是有其必要性。因此本研究利用遙感探測（RS）、地理資訊系統（GIS）、景觀指標（Landscape index）等空間技術整合，分析清境地區九年間景觀變遷情形，藉以瞭解影響層面。物件導向式分類以多階層、多空間特徵等方式萃取地物，藉由歸屬度函數建立分類判釋邏輯，選擇適合的空間特徵與函數計算歸屬度高低，有效萃取地物。因此本研究蒐集三個時期（1999、2003、2007）影像，進行物件導向式分類。成果顯示，三個時期總體準確度皆達到80％以上。接著，將分類成果利用變遷偵測法－分類後比較法，掌握清境地區不同時期的地物變遷情形與瞭解變遷區域的分佈位置；其中變遷區域集中於三個區域，分別為翠峰周圍、眉溪河谷周圍與清境農場周圍，幾乎沿著臺14 甲線多點且小範圍的分布發展，屬於居民與觀光景點主要集中地。最後，再使用景觀指標量化景觀結構，分析探討其生態意義。結果顯示，清境地區景觀結構已處於較穩定的狀態，九年間雖未有較大規模的開發，但人為介入並未停止，對於當地環境之衝擊與破壞依然存在。
With the abundant natural resources, the comfortable weather, and the increase of domestic recreational activities, the Cing-Jing region has become a famous sightseeing site in Taiwan. However, the human exploitation would seriously impact its natural environment. In order to understand the degrees of environmental impacts, this research utilized the remote sensing technology, geographical information systems, and landscape index to analyze the changes of landscapes at Cing-Jing region from 1999 to 2007. The object-oriented classification using multi-layer and multi-spatial characteristics was first employed to interpret the land cover types. The interpretation logic established using the membership function and the selection of suitable spatial characteristics was adopted to acquire the better accuracy of classification. The results showed that the overall accuracy for the 1999, 2003 and 2007 three individual years could all reach more than 80%. The post-classification comparison method was then used to obtain the areas and types of land cover changes during these three years. The results demonstrated that the changed areas were mainly located in three regions, including the surroundings of Chu-Feng, Mei river valley, and the Cing-Jing farmland. These areas were mostly located along the Tai-14-Jia route as well as the major sightseeing spots. The landscape index was finally adopted to quantify the landscape structures to analyze the ecological meanings of this region. The results have shown that the landscape structures were in a stable condition. Although there were no major developments in the past 9 years, the human impacts and exploitation on the local environment could never end.
With the abundant natural resources, the comfortable weather, and the increase of domestic recreational activities, the Cing-Jing region has become a famous sightseeing site in Taiwan. However, the human exploitation would seriously impact its natural environment. In order to understand the degrees of environmental impacts, this research utilized the remote sensing technology, geographical information systems, and landscape index to analyze the changes of landscapes at Cing-Jing region from 1999 to 2007. The object-oriented classification using multi-layer and multi-spatial characteristics was first employed to interpret the land cover types. The interpretation logic established using the membership function and the selection of suitable spatial characteristics was adopted to acquire the better accuracy of classification. The results showed that the overall accuracy for the 1999, 2003 and 2007 three individual years could all reach more than 80%. The post-classification comparison method was then used to obtain the areas and types of land cover changes during these three years. The results demonstrated that the changed areas were mainly located in three regions, including the surroundings of Chu-Feng, Mei river valley, and the Cing-Jing farmland. These areas were mostly located along the Tai-14-Jia route as well as the major sightseeing spots. The landscape index was finally adopted to quantify the landscape structures to analyze the ecological meanings of this region. The results have shown that the landscape structures were in a stable condition. Although there were no major developments in the past 9 years, the human impacts and exploitation on the local environment could never end.