台大實驗林混農林業朽木類甲蟲群聚與環境因子之影響

Abstract

混農林業(agroforestry)在許多開發中國家被視為在農業發展與維持生物多樣性資源的衝突中，一個可行的解決方案，但在臺灣，是否該在原本適宜造林的崎嶇山地引入混農林業則少有提及。本研究於臺大實驗林三個不同海拔設置樣區，比較混農操作和造林地環境下朽木類甲蟲多樣性與群聚結構，並透過排序技術探討環境因子其影響朽木類甲蟲群聚的能力。自2011年九月到2012年九月，從六個樣區十八座新型羽化陷阱中採得4967隻鞘翅目昆蟲，隸屬於63科554種。其中以隱翅蟲科(Staphylinidae)種類數最多，共共192個分類群；小蠹蟲亞科(Scolytinae)數量最多，共1849隻。比較兩樣區之稀有度曲線 (rarefication curve)，兩者預測物種豐度之SChao1值以混農樣區較高，但皆落在彼此S Chao1之百分之九十五信賴區間內，無顯著差異。除了中海拔樣區外，其餘兩組樣區朽木類甲蟲的Shannon多樣性指數以造林樣區較高，但不顯著，均勻度 (Shannon’s evenness) 也以造林樣區較高。環境因子中以海拔高度 (elevation)最具有決定地位；數量型環境因子以遮蔽度 (canopy)最為關鍵，遮蔽度在樣區間和操作間具有顯著差異，並與朽木材積 (timber volum)、材表面積 (timber surface area) 和樹冠高度顯著相關，並與均勻度顯著相關，和採得個體數負相關。群聚分析 (cluster analysis) 的結果顯示朽木類甲蟲相與利用方式較無關聯，相似情形近似樣區間的地理距離。典型對應數分析 (canonical correspondence analysis, CCA)結果顯示海拔高度、遮蔽度、朽木材積和材表面積和最能解釋物種分佈之第一典範軸顯著相關；第二典範軸則與坡度和樹冠高度顯著相關。混農林業引入造林地其蟲害風險與交互感染風險較平地為高，經營維持收益所需之蟲害防治操作與生物多樣性的維持難以並行。

Agroforestry have been seen as a feasible solution of the conflict about crop production and biodiversity conservation in developing countries. But it still has doubts about introducing agroforestry into the rugged mountain of Taiwan, where used to be forest plantation. The study set up 3 group of sample sites, depend on different altitude, in the NTU Experiment Forest, to compare the diversity and the community structure of saproxylic beetles under the environment of agroforestry and forest plantation. And discussed the ability of environment factor influence the saproxylic beetles, through the orientation technique. From September 2011 to September 2012, 4697 individuals of coleopteran insect, belong to 554＋species and 63 families, were collected from six sample sites, including 18 new type emergence traps. Family Staphylinidae had the most abundant number of species, with 192 taxons；Subfamily Scolytinae had 1849 individuals, were the most abundant group. Comparing the rarefication curve of two kind of usage, Agroforestry sites had higher SChao1 value, but their 95% Confidence interval of SChao1 had large ranges overlap, no significant differences in richness. Every site other than low-medium altitude, Shannon’s diversity indexes were higher in the forest plantation ones, but not significant. Evenness were higher in forest plantation sites. Elevation was the most decisive factor to the distribution of species in all environment variables; canopy as the key variable, had significant differences among sites and usages, correlated with timber volume, timber surface area, and canopy height, and also correlated with evenness, negative correlated with number of individuals. The results of cluster analysis shown the species composition were well depend on the geographical distance, rather than the way of usage. Altitude, canopy, slope, and timber volume on axis 1 and tree height, utilize, and slope on axis 2 were the environment variables that best explained about 34.6% of the variance in saproxylic beetle assemblage on the first two axes. Introducing agroforestry into forest plantation may have higher risk of pest cross infection, lead to higher cost of pest management to maintain profit, efficient ways of pest management were contrary to the goal of maintaining biodiversity.