以電路理論探討都市景觀連接度與鳥類群聚之相關研究

Abstract

景觀連接度(connectivity)被定義為景觀促進或阻礙生物體在資源塊區之間運動的程度，可分為結構與功能。在城市化的過程中，時常間接導致生物棲地破碎化，景觀連接度的降低，也會影響物種在空間中的移動與分布。然而，我們對於景觀連接度的影響與認知上仍存在差異，通常僅作資源分配而甚少考慮生物移動模式，倘若忽略了生物移動將導致保護上潛在破壞的結果，故了解連接度將有助於在景觀尺度上，為生態問題提供規範性的解決方案。 本研究旨在探討都市景觀連接度與鳥類群聚之關係，選擇電路理論模型作為景觀連接度量化指標，檢驗景觀連接度是否與鳥類群聚有關，並比較不同連接度量化指標對鳥類群聚的影響程度。以台北市作為研究樣區，使用BBS Taiwan臺灣繁殖鳥類大調查的鳥調數據，推導出各景觀類別的電阻值，以此表示為物種在景觀中移動的難易程度，並運用ArcGIS 10.4數化景觀結構，將電阻值紀錄於網格資料(raster)中，後續以電路理論為基礎的Circuitscape 4.0軟體進行連接度分析，最後檢定景觀連接度與鳥類多樣性的相關性。 研究結果顯示林地鳥類是會受到鄰近塊區特徵的影響，且較不受距離上的限制，當資源區塊間隔離程度(移動成本)越小時，越有利於鳥類移動與分布，鳥類多樣性會隨之增加。藉由研究結果指出，電路理論模型相較於其他量化指標，對鳥類群聚影響程度較大，適用於分析臺灣環境特性與生物習性。在實務上的應用，能夠透過電路理論模型比較不同方案對連接度提升的成效，有助於確定對物種有利的空間配置和保護生物多樣性的恢復或緩解之措施。

Landscape connectivity is the degree to which the landscape facilitates or impedes movement among resource patches. This can be further refined into structural and functional definitions (Taylor et al., 1993.). In the process of urbanization, the biological habitat is fragmented indirectly. Reducing connectivity will decrease the rate at which species can move between source patch. However, we still have different cognitions of landscape connectivity. Maps produced by planners are usually of resource distribution, not movement patterns. Ignoring biological movement can lead to potentially devastating consequences for conservation. Our research focused on the relationship between landscape connectivity and species diversity of birds. We also tried to review and choose the appropriate method modelling landscape connectivity. This study use circuit theory to model landscape connectivity and collect the on-site investigation data of BBS Taiwan. We compare the data to depict the relationship between landscape connectivity and woodland birds species. We found woodland birds species' richness may be more strongly affected by characteristics and function of landscape patches than distance between the landscape patches. Recognizing landscape functional connectivity assists our in offering prescriptive solutions to ecological problems at the landscape scale.