請用此 Handle URI 來引用此文件: http://hdl.handle.net/11455/84803
標題: Combining GBIF and MaxEnt to Predict the Suitable Habitat of Alnus formosana
結合GBIF與MaxEnt預測臺灣赤楊之適宜生育地
作者: Ching-An Chiu
Hsien-Sheng Hsu
Shin-Hwei Lin
邱清安
徐憲生
林信輝
關鍵字: Species distribution modeling
Alnus formosana
habitat suitability index
vegetation rehabilitation
物種分布模擬
臺灣赤楊
生育地適宜度
植生復育
出版社: 台北市:中華水土保持學會
摘要: 選擇合宜的植物種類是植生復育的關鍵第一步,利用新興的物種分布模擬 (SDM) 將有助於正確選擇樹種及確定適宜生育地以提供科學決策。本文以臺灣赤楊為例,使用免費易得的全球生物多樣性資訊機構(GBIF) 物種資料庫、最大熵 (MaxEnt) 物種分布軟體進行其生育地適宜度分析,結果顯示預測模型之準確度評估屬於良好等級 (AUC = 0.842),所得之預測出現機率可加以繪製臺灣赤楊之生育地適宜度 (HSI) 分布圖,經9 處崩塌地鄰近區域植群調查資料驗證十分吻合,同時本文對未來物種分布模擬尚待解決之相關議題加以討論,期使本地原生植物在水土保持植生復育中更具科學基礎與發揮其應用潛力。
Selecting appropriate species is the first key step for vegetation rehabilitation. Novel species distribution modeling (SDM) can assist in making scientific decisions to support species selection and predict suitable habitat. In this paper, we combine the open-access Global Biodiversity Information Facility (GBIF) database and MaxEnt modeling software to predict Alnus formorsana distribution. The results reveal that the accuracy assessment of our model is good within an area of 0.842 according to the receiver operating characteristic curve. We transform the predicted occurrence probability, through ArcGIS, to map the habitat suitability index (HSI) of Alnus formorsana that approximately corresponds with the observed vegetation in 9 nearby landslide areas. Based on our findings, we discuss the future challenges related to SDM. The proposed approach can be used in the future to facilitate proper application of native plants in soil and water conservation.
URI: http://hdl.handle.net/11455/84803
顯示於類別:第45卷 第03期

文件中的檔案:
檔案 描述 大小格式 
84618-6.pdf3.47 MBAdobe PDF檢視/開啟


在 DSpace 系統中的文件,除了特別指名其著作權條款之外,均受到著作權保護,並且保留所有的權利。