- 期刊
- OpenAccess
應用Markov和Logit模式監測地景變遷之研究
Application of Markov and Logit Models on Monitoring Landscape Changes
摘要
本研究旨在結合Markov和Logit模式，探討林試所蓮華池試驗林未來之地景變遷情形，以及民墾地的空間分佈位置。研究方法係先數化試驗林1971及1998年之土地利用型圖，應用Markov模式預測未來地景變遷，並選擇5項與民墾干擾有關之地形環境因子，以Logit模式進行民墾地之空間分析，整合二者之結果模擬未來民墾地之空間分佈情形。使用Markov模式預測未來地景變遷結果顯示，根據1971至1998年間之資料所建立的Markov模式，預測2052及2106年之地景結構，發現民墾地面積由1971年的0.39%和1998年的3.39%，增加為2052年的12.08%和2106年的16.09%，明顯地呈現上升趨勢。至於應用Logit模式分析民墾地發生之空間分佈結果顯示，民墾地的發生與海拔高、坡度、及距民墾地、道路和河流的距離均有密切的關係，但其中以距民墾地和距道路的遠近最有相關，最後整合兩模式的結果模擬未來民墾地之空間分佈結果指出，民墾地之分佈確實有逐漸擴張之趨勢，同時亦間接地影響地景結構之空間排列。
並列摘要
In this study, we applied Markov awl Logit models to investigate future landscape changes and the spatial distribution of illegal cultivation of forestlands in the Lienhwachih of Taiwan Forestry Research Institute in southwestern Taiwan. The land use maps in 1971 and 1998 were first digitized. The Markov model was then applied to predict future landscape changes, and the Logit regression model with 5 terrain factors was used to analyze the spatial relationship between illegal cultivation of forestlands and environmental factors. Finally, the results obtained from these 2 models were integrated to simulate the spatial distribution of illegal cultivation of forestlands. The prediction of future landscape changes using the Markov model indicated that the area of illegal cultivation of forestlands would increase from 0.39% in 1971 and 3.39% in 1998 to 12.08% in 2052 and 16.09% in 2106. As for the spatial distribution using the Logit regression analysis, the results showed that the occurrence of illegal cultivation of forestlands was related to the elevation, slope, and the distance to roads, streams, and previous illegal cultivation, with particularly high correlations with the distance to roads and illegal cultivation. Finally, the spatial simulation of illegal cultivation illustrated that the spatial distribution of illegal cultivation would gradually expand, and the configuration of overall landscape structure would indirectly be affected.