隨著經濟的快速發展,對於自然生態環境的衝擊也日益增大,生物多樣性保育的相關議題也是現代生態保育的關注的重點。由於植物分布模式的建立是了解植物生態的要件,因此如何透過有效的分布模式建立來探討物種分布特性是保育決策重要的一環,也是探討物種生態系關連的重要連結。台灣林植物的分布方面,冷杉、鐵杉及檜木是三種最重要也最優勢的天然針葉樹,其分布亦有相混生的情形。本研究主要目的在於,利用緯度及海拔高度兩環境因子,運用邏輯迥歸方法,以台灣為尺度首先建立此三種優勢針葉樹種的分布模式,並討論是否能由模式進而探討其混生情形。研究結果顯示,檜木分布約在海拔l,000-2,600 m,冷杉約2,700-3,700 m,兩者之間在約2,700 m處有明顯分界線,此處亦就是鐵杉最為優勢的區域。模式顯示檜木在台灣中部分布可高達3,300 m,而在南部僅可達2,600 m,平均每1 km海拔驟降率達7.78 m;鐵杉在台灣中部分布可高達3,600 m,北大武山附近僅達3,100 m,平均每1 km 海拔驟降6.25 m;冷山於台灣中部高達3,700 m,過了玉山之後僅達3,200 m,平均每1 km海拔驟降10 m。總括而論,在台灣的尺度上,此三種樹種的分布主要受到大山塊加熱效應,東北季風與雲霧帶的影響,僅利用緯度及海拔兩個環境因子即叫有效解釋樹種於全台的分布,同時也可清楚說明三種樹種的混生情形。
With the rapid rate of urbanization and increasing demands for forest resources, the vegetation cover has also decreased at a proportional rate in Taiwan. Government agencies and forest managers have come to realize the importance of the ecosystem and begun to implement conservation practices. Prediction of vegetation distribution will be a primary step in this effort. In this study, we developed logistic regression models for predicting the distribution of Abies kawakamii (Taiwan fir). Tsuga chinensis var. formnosana (Taiwan hemlock), and Chamaecyparis spp. (Taiwan cypress). The models achieved high R^2 and receiver operating characteristic (ROC) values. Overlaying the probability of map of Taiwan cypress and Taiwan fir showed a transition boundary for two species occurring at an elevation of approximately 2700 m. The distribution model showed an oval species distribution pattern, narrower and lower at north and south, and higher and wider in the middle. Therefore, the Massenerhebung effect, the northeasterly monsoons, and the cloud belt were also brought in to better explain the models against the species distribution in the real world. The model correctly predicted the species distribution using only 2 general environmental factors of latitude and elevation.