外來種緬甸小鼠與原生種赤背條鼠空間分布差異及其可能機制

Abstract

外來種入侵常伴隨原生種的消失，這種外來種和原生種空間分布區隔的現象，可能肇因於外來種的競爭(driver model)，但也可能是因為兩者分別適應不同人為干擾程度的棲地，原生種因為棲地受干擾而消失，外來種進而利用這些空缺的棲地(passenger model)。然而過去研究鮮少綜合田野調查及操作型實驗以釐清背後的機制。外來種緬甸小鼠首次於 1999年在花蓮吉安鄉被記錄到，2007-08 年時已入侵到吉安鄉南邊的壽豐鄉，在緬甸小鼠持續向南入侵的期間，與之體型相近的赤背條鼠則有分布逐漸減少的趨勢。為了瞭解緬甸小鼠的入侵現況，緬甸小鼠和原生種赤背條鼠空間分布差異，及造成差異的可能機制，本研究於 2020-2021 年在花蓮縣四個行政區(由北到南：花蓮市、吉安鄉、壽豐鄉及鳳林鎮)，共選擇 64 個樣點(共 7,488 個籠夜)進行調查，並測量樣點內微棲地和樣點周圍土地利用狀況，同時進行室內、野外覓食競爭實驗以及毛髮碳氮同位素食性階層分析以探討兩種鼠類的競爭關係。結果顯示緬甸小鼠已擴展到鳳林鎮中部，而赤背條鼠則主要僅分布在鳳林鎮北部，2007-2008 年在壽豐鄉的赤背條鼠族群幾乎都已經消失，空間分析的結果顯示兩者有空間分布區隔的現象。毛髮碳氮同位素分析發現緬甸小鼠和赤背條鼠的食性階層相近，兩者可能競爭類似的食物資源。Redundancy analysis顯示兩物種偏好不同的棲地，赤背條鼠偏好周圍有較多灌木及草地的棲地，而緬甸小鼠則在周圍有較高人口數及建築物的樣點最常出現。室內取食實驗顯示兩鼠種間幾乎沒有攻擊或打鬥行為，緬甸小鼠的存在與否也不會影響赤背條鼠的覓食狀況，但是相較於同種，赤背條鼠對緬甸小鼠會展現較多防禦行為；而野外覓食實驗發現赤背條鼠比緬甸小鼠更傾向進入食物箱中覓食，也更願意在食物資源少時繼續進食，與緬甸小鼠會把食物資源密度消耗到赤背條鼠無法生存的預期不符。本研究發現緬甸小鼠和赤背條鼠分別適應不同人為干擾程度的環境，此外，緬甸小鼠的存在會增加赤背條鼠的防禦行為，有機會影響赤背條鼠的生理表現，因此兩物種在空間分布上區隔的狀況可能同時受到人為環境干擾以及緬甸小鼠入侵的影響，未來應探討其他可能的影響因子，如捕食者和寄生蟲，以更完整的得知赤背條鼠分布範圍縮減的原因。

Spatial segregation between native and invasive species could be caused by competitive exclusion of native species (the driver model), or due to negative impacts of anthropogenic habitat disturbance on native species, leading to differentiation of habitat use between native and invasive species (the passenger model). While the distributional patterns of native and invasive species have long been studied, few studies have combined manipulative experiments with field surveys to better understand the underlying mechanisms. In Hualien, eastern Taiwan, the invasive Rattus exulans was first recorded in 1999 in Ji’an township, and had expanded southward to Shoufeng township in 2007-2008, whereas the range of native Apodemus agrarius had contracted since the invasion of R. exulans. To update the spatial distribution of R. exulans and A. agrarius in Hualien, and uncover the potential mechanisms underlying it, field and manipulative studies, as well as isotope analysis were implemented. Field survey included small mammal trapping and measurements of microhabitat and 1-km radius land use at 64 trapping sites (7,488 trap-nights) in four districts of Hualien (Hualien city, Ji’an, Shoufeng and Fenglin, from north to south) between September 2020 and March 2021. Manipulative experiments comprised laboratory and field foraging trials. It was found that R. exulans had expanded to central Fengling and occurred in all districts, while A. agrarius was restricted mainly to northern Fengling. The A. agrarius population observed in Shoufeng between 2007 and 2008 has largely disappeared. There was no difference in trophic levels between the two species based on isotope analysis, suggesting potential food competition. Spatial point patterns analysis indicated that distribution of the two species were spatially segregated. Redundancy analysis revealed that the two species preferred different habitat types: abundance of A. agrarius was positively associated with areas of natural shrubland and grassland, while that of R. exulans was positively correlated with built-up area and human population density. Laboratory foraging trials showed that fighting behavior was rare between A. agrarius and R. exulans, but A. agrarius displayed more defensive behaviors against R. exulans than the other A. agrarius. Also, A. agrarius did not forage less (both the foraging time and the number of foraging events) when paired with R. exulans relative to A. agrarius. Field foraging experiments demonstrated that comparing to R. exulans, A.agrarius was more inclined to enter the food container. In addition, A.agrarius continued foraging to lower seed density, which violated the prediction that R. exulans may displace A.agrarius when the former species can deplete food resource to such a low level that the latter species cannot subsist on. The observed spatial segregation between R. exulans and A. agrarius can be explained by differential adaptation to habitat disturbance; besides, defensive behavior displayed by A. agrarius against R. exulans suggests dominance of R. exulans and could induce stress to inferior A. agrarius. Therefore, I conclude that the additive model, which invasive species and habitat disturbance have synergistic effects on native species, is supported. Factors other than habitat disturbance and invasive species, such as parasites or predators, can be explored in the future to more explicitly unravel the mechanisms underlying the contracted distribution of A. agrarius.