衛星追蹤南橫公路周邊熊鷹(Nisaetus nipalensis)空間利用與道路開放前後行為差異

熊鷹(Nisaetus nipalensis)在台灣作為高階捕食者可做為生態指標物種，但過往由於方法限制了追蹤的努力量。此外，隨著南橫公路的開放對熊鷹是否造成影響所知有限。在此本研究透過GPS/GSM太陽能衛星發報器追蹤玉山國家公園南橫公路周邊的熊鷹，探討活動範圍、活動模式、日夜棲地選擇與道路管制對熊鷹造成的活動影響。研究期間內捕獲2隻雌性熊鷹亞成鳥，一隻有固定領域且成功繁殖(N1930)，另一隻則為遊蕩個體(Floater) (N1931)，分別收集了4735與2986筆定位紀錄。研究顯示，兩者的活動範圍重疊度僅0.18。N1930的MCP有13.83 km2，是過往無線電追蹤的兩倍餘。活動模式以中午為高峰，在求偶期有最高的飛行占比(8.53%)，。日棲地選擇以原生林與闊葉林較高(66%與75%)，平均坡度約47.1 ± 10.3°，以北向坡為主要利用坡向。N1930共有15處夜棲地，選擇同樣以原生林與闊葉林最高(51%與56%)，平均坡度約46.4 ±11.2°，同樣以北向坡為主要利用坡向，且當日平均溫度對夜棲時海拔不具有影響力。公路活動干擾方面，僅在長期無人車的時期下有趨近道路之現象。遊蕩的N1931的MCP達621.5 km2，活動模式亦以中午為高峰，惟活動高峰時段比N1930更晚來到，日棲時同樣對原生林與闊葉林具有高度選擇(55%與77%)，但在原生林有低於環境占比的選擇性(p<0.05)，平均坡度約為35.3 ± 10.2°，對南向坡具有高於環境占比的選擇性(p<0.05)；N1931共有16處夜棲地，但夜棲地幾無選擇性，與日棲有所不同，僅對待成林地具有低於環境占比的選擇性(p<0.05)，對於各坡度與向坡向亦不具有選擇性，當日平均溫度對夜棲海拔亦不具有影響力，而對於公路活動的干擾，其結果可能包含了許多個體播遷的混淆效應難以看出兩者間的關聯。兩隻個體在植被稀疏區的飛行比例皆高於植被覆蓋區域，但仍然以停棲為主。透過架設於捕捉陷阱旁之自動照相機顯示於公路開放前熊鷹主要造訪時段以13至16時為最高(76%)，而在開放後此高峰則較不顯著；且開放後熊鷹以管制時段作為主要的造訪時段，與開放前呈現相反趨勢。本研究透過太陽能衛星發報器補足了過往台灣所缺乏的長期追蹤資料，與道路活動所帶來的影響分析。

關鍵字

熊鷹(Nisaetus nipalensis) ；衛星追蹤；空間利用；南橫公路；道路影響

並列摘要

As a top predator, the Mountain hawk eagle (Nisaetus nipalensis) acts as an important ecological indicator species in Taiwan. However, due to the lack of long-term tracking data and method limitations, the disturbance of Nanheng Highway, recently opened to the public on the Mountain hawk eagle is still unknown. This study used GPS/GSM solar satellite transmitters to track the bird around the Highway in Yushan National Park, with the aim of investigating the home ranges, activity pattern, and roost site selection and the disturbance of traffic regulation. During the study period, two subadult females, N1930 and N1931, were captured. N1930 had a breeding home range while N1931 was a floater, each with 4735 and 2986 tracking fixes collected, respectively. Results showed that the home range overlap of the two eagle was only 0.18. The MCP of the N1930, 13.83 km2, is about twice the size as that recorded in previous work. It was most active at noon and flight activity peaked during the courtship period (8.53%). N1930 selected primary forest and broad-leaved forest more than other vegetation types (66% and 75%, respectively), area with average slope around 47.1 ± 10.3° and mainly north slope. Fifteen roosting sites were recorded for the bird, among them 51% are primary forest and 56% broad-leaved forest, with a mean slope of around 46.4 ± 11.2° , and mainly on the north slope. Air temperature had no effect on its roosting altitude. In terms of road activity interference, the phenomenon of approaching the road was only seen during long-term low traffic flow period. N1931 showed a larger MCP of 621.5 km2. The daily activity pattern also showed a bell-shaped distribution at noon, with the peak activity came later than N1930. It was also highly selective for primary forests and broad-leaved forests (55% and 77%) at daytime, whereas the primary forest was not selected (p<0.05). The average slope used was about 35.3 ± 10.2°, and the north slope was selected (p<0.05). N1931 had a total of 16 roosting sites without habitat selectivity, different from the daytime habitat selection. The pre-forest was avoided by the bird (p<0.05) during the night. The effect of road disturbance might be confounded by dispersal behavior of the bird. Aloft locations of the two birds were recorded more in the sparsely vegetated area than that in the vegetated area, while daytime perching was still the main activity. With trap cameras set beside the traps, I found the mountain hawk eagles visited mostly from 13:00 to 16:00 (76%) before the opening of the road, while during the closedown of the road the pattern became vague. This long-term tracking study strengths our understading of this species behavior.