一、 計畫緣起
墾丁國家公園自1994年起在社頂、龍鑾潭、九棚、出火等地陸續野放了223頭梅花鹿。經過二十多年後，梅花鹿已成功在墾丁國家公園內建立穩定族群，然而過高的族群密度，使得梅花鹿族群對國家公園週遭的森林生態環境、農作物、交通帶來負面影響。緣此，本研究藉由了解梅花鹿的分布現況，並對梅花鹿捕捉技術進行試驗，作為後續園區內的鹿隻族群數量管理策略擬定的參考依據。
二、 調查方法及過程
本年度對墾丁國家公園與週邊地區，即墾丁中、南部區域、縣道200號以北至屏156與龍鑾潭西側林地，以自動相機搭配痕跡紀錄進行梅花鹿分布現況調查，以REM法估算梅花鹿的族群密度與數量，並依近年的出現紀錄以MaxEnt進行物種分布模式預測。另外，本研究將社頂復育區聯外的圍籬改裝自動關門裝置、搭配4G相機進行梅花鹿的捕捉，以增進捕獲效率。並在野外以誘餌搭配自動相機，觀察梅花鹿對飼料與飲水的取食情形。此外，並對園區捕捉到的梅花鹿接種GnRH疫苗，藉來年的產仔率觀察GnRH疫苗的成效。最後，以現有調查結果，提出園區梅花鹿長期監測及經營管理之建議。
三、 重要發現
研究結果顯示，在經歷數次的野放行動後，野生梅花鹿目前在墾丁地區呈現二個大範圍連續分布族群，分別為墾丁中、南部區域高密度族群（=34.66 ± 1.44隻/km2）和縣道200號以北區域的低密度族群（=6.00 ± 0.61隻/km2）。以高位珊瑚礁自然保留區和社頂為源頭的高密度鹿群，除了鵝鑾鼻之外，廣泛分布在墾丁中、南部區域，族群數量估算為1736.47隻（95%信賴區間為912.32-2560.61）；縣道200號以北的低密度鹿群，則由原始野放地點出火，向東、向北持續擴散，目前已達到保力林場南緣與小路社區，族群數量估算為319.08隻（95%信賴區間為73.92-564.24）；龍鑾潭西側林地則為國家公園境內的小型低密度孤立鹿群（=4.57 ± 0.90隻/km2），族群數量估算為5.39隻（95%信賴區間為1.63-8.33）。整體而言，儘管野放後梅花鹿的分布範圍持續增加，但與MaxEnt分布模式預測相比，仍然非常侷限，極可能是因為地景屏障的影響所致，尤其是道路的阻隔。梅花鹿捕捉測試方面，社頂復育區聯外的圍籬共捕捉到6隻梅花鹿，誘餌測試顯示梅花鹿對精料和飲水的取食意願低，故drop-net捕捉效率推斷可能不佳。GnRH疫苗測試方面，共計對37隻梅花鹿施打疫苗，其中雌鹿27隻，雄鹿10隻。
四、主要建議意見
一、 立即可行建議
主辦單位：墾丁國家公園管理處
(一) 高位珊瑚礁自然保留區架設圍欄與梅花鹿淨空。
(二) 對梅花鹿未來可能擴散的區域，視情況進行預防性圍欄架設。
(三) 農損密集區與潛在農損區架設大範圍圍欄。
二、 中長期建議
主辦單位：墾丁國家公園管理處
協辦單位：其他地方主管機關
(一) 擬定梅花鹿分區管理和監測計畫。
(二) 以大型固定圍欄、搭配小型機動裝置進行梅花鹿捕捉。
(三) 梅花鹿移除行動人員的捕捉、搬運標準流程建置。
(四) GnRH避孕疫苗可作為捕捉移除的輔助措施。
(五) 改善圈養鹿群管理與增加梅花鹿留容場域。
三、 長期建議
主辦單位：墾丁國家公園管理處
協辦單位：其他地方主管機關、社區團體
(一) 梅花鹿贈送管道開源與種間基因滲入監測。
(二) 成立梅花鹿專責小組與加強教育宣導、在地參與。
計畫類別：[動物生態]
英文摘要：1. Introduction
Kenting National Park has reintroduced 233 Formosan sika deer（Cervus nippon taiouanus）in Sheding, Longluangtan, Jiupeng, Chuhuo since 1994. After 20 years, the population of sika deer grows up stably in the Kenting National Park. However, which have a negative impact on the crops, forests, and transportation due to the overabundant population. The investigation results of the deer population distribution, the predation of the species distribution model, and the result of the capture tests in this study can be used as a reference for deer population management in Kenting National Park.
2. Methods
To investigate the population distribution of the sika deer, the camera trapping sites and route survey were conducted in the central and southern areas of Kenting Nation Park, the Road 200 to Route156, and the western woodland of Longluan Lake, separately. In addition, to estimate the average density and the species distribution model of the sika deer, the REM and the MaxEnt based on the occurrence records in recent years were also used in the above three areas. To improve the capture efficiency of the sika deer, the pastures outside the Sheding research center were modified with 4G remote gate. And several prebait sites were also set to observe whether the bait can attract sika deer in the wild. The sika deer caught this year were vaccinated with the GnRH vaccine. The future reproduction status of female sika deer will be observed to evaluate the effectiveness of the immunocontraceptive vaccine. Based on the above results, we made the recommendations for the long-term monitoring and management of sika deer in Kenting National Park.
3. Findings
The results in the present study show there are two large large-continuous deer populations in the Kenting area, The higher density population (=34.66 ± 1.44 deer/km2) in the central and southern areas of Kenting Nation Park are originated from the Uplifted Coral Reef Nature Reserve and Cheding Park, it is widely distributed in this area, except for Eluanbi. The population number is estimated to be 1736.47 (the 95% CI is 912.32-2560.61). The lower density population (=6.00 ± 0.61deer/km2) in Road 200 to Route156 has dispersed from Chuhuo, and spread to the southern edge of Baoli and Xiaolu Community. The population number is estimated to be 319.08 (the 95% CI is 73.92-564.24). The population in the western woodland of Longluan Lake is small and the density is low (=4.57 ± 0.90 deer/km2), the population number is estimated to be 5.39 (the 95% CI is 1.63-8.33). Although dispersal of the sika deer is continuing, the distribution of the sika deer is still limited compared with the MaxEnt model, it may be restricted by the landscape barrier, especially roads. A total of 6 sika deer were caught in the pastures outside the Sheding research center, and the result of prebait tests showed the attractiveness of concentrate and water is low for the sika deer. In the GnRH vaccine trial, a total of 37 deer were vaccinated, including 27 female deer and 10 male deer.
4. Suggestions
(1) Erect fences around the Uplifted Coral Reef Nature Reserve, and haze out of the fenced area.
(2) For the areas where the sika deer may disperse in the future, implement preventive fence erection as the situation requires.
(3) Erect large-scale fences in agricultural areas with deer damage and potential damage areas.
(4) Drafting the sika deer management and monitoring plan into several units.
(5) Capture the sika deer with large fixed fences and small motorized devices.
(6) Establish the standard procedures for the process of capturing and transportation the sika deer.
(7) GnRH contraceptive vaccine can be used as an assistant measure to main capture methods.
(8) Improve the management of captive deer and increase the other alternative deer shelters.
(9) Increase the option of deer donation for private deer farm in Hengchun Peninsula
(10) Monitor and prevent the possible introgression of the wild deer.
(11) Establish a sika deer work team and enhance the education/interpretation efforts.
辦理單位：國立屏東科技大學
計畫屬性：本處委託辦理
執行年度：110
計畫主持人：裴家騏
中文關鍵字：台灣梅花鹿、空間分布預測、族群密度估算、族群數量控制、圍欄防治、墾丁國家公園
英文關鍵字：fences, Formosan sika deer, Kenting National Park, population control, population density, species distribution model