八色鳥(Pitta nympha)潛在繁殖地與度冬地分布預測

Abstract

物種的地理分布與環境間的關係是進行研究與保育的重要資訊。八色鳥（Pitta nympha）為東亞的遷徙性鳥種，由於遭到棲地破壞以及非法捕捉，被列為全球性的易危鳥種。由於八色鳥數量稀少且不易被發現，使其相關出現記錄十分缺乏。此外，過去對於八色鳥的研究多侷限在小範圍的空間尺度，而缺乏整體性的分布與棲地特性研究。因此，本研究利用地理資訊系統整合環境因子與八色鳥的出現記錄，並以Genetic Algorithm for Rule-Set Prediction（GARP）、Ecological Niche Factor Analysis（ENFA）以及Maximum Entropy（MaxEnt）三種分布預測模式來推估八色鳥的繁殖與度冬潛在棲地分布，以了解八色鳥在東亞地區的地理分布情形以及所屬棲地特性，同時，比較不同尺度下（即東亞地區中尺度『0.1x0.1度』與臺灣細尺度『1x1公里』）之分布預測差異。研究結果顯示，雨量為影響八色鳥分布最重要的環境因子。模式預測日本、朝鮮半島、臺灣和中國大陸南方為繁殖地分布範圍，其中日本中部、朝鮮半島北部和中國大陸西南為過去已知範圍外的潛在繁殖地；度冬地分布於婆羅洲西半部的汶萊、馬來西亞砂勞越州、沙巴洲北部和印尼加里曼丹省等地，其中南部的加里曼丹為過去已知範圍外的潛在度冬地。在東亞地區中尺度下進行預測，其分布型態與臺灣細尺度預測之結果大致相符，顯示在中尺度下進行模式預測也能夠反映一定程度的八色鳥在細尺度下的分布，而利用細尺度模式則可更精準地掌握特定區域的分布情形。本研究以分布預測模式為稀有物種八色鳥找出潛在的適合棲地，劃出了八色鳥的繁殖地與度冬地分布範圍，並提供未來八色鳥的野外研究工作與保育的參考資訊。

Understanding the relationship between a species’ distribution and its environment is important for further research and conservation management. Fairy Pitta (Pitta nympha) is a kind of migratory bird distributed in East Asia, and classified as a globally vulnerable species due to habitat loss and illegal human hunting. As they are rare and difficult to detect, the records are very shortage, especially in wintering grounds. Moreover, most previous studies focused on fine-scales, which showed few characteristics of distributions and habitats of the Fairy Pitta. The goal of this study was to integrate environmental factors with records of the Fairy Pitta by Geogtaphical Information System (GIS), to predict potential ranges of the Fairy Pitta in breeding and wintering grounds by using Genetic Algorithm for Rule-Set Prediction (GARP), Ecological Niche Factor Analysis (ENFA) and Maximum Entropy (MaxEnt), and to understand overall geographical distributions and habitat characteristics of the Fairy Pitta in East Asian. Two spatial scales were compared simultaneously (i.e. a meso-scale in East Asia (0.1x0.1°) and a fine-scale in Taiwan (1x1km)). The results showed that precipitation was the most important factor to effect the distribution of the Fairy Pitta. Their predicted breeding ranges included Japan, Korean Peninsula, Taiwan and southern China, in which central Japan, northern Korean Peninsula and southwestern China were potential range but were outside the known range; wintering ranges were located in western Borneo, including Brunei, Sarawak and northern Sabah of Malaysia, and Kalimantan of Indonesia, in which Kalimantan was potential range but was also outside the known range. The predicted distribution at the meso- scale in East Asia was similar to those at the fine-scale in Taiwan, but more locally specific habitat characteristics were represented at the fine-scale than at the meso-scale. In this study, the distribution predictive model can indeed draw the potential ranges of suitable habitat, both breeding and winter habitats, of the Fairy Pitta, and provide guidelines to further field research and conservation.