全球氣候的持續暖化將迫使高山植物往更高海拔遷徙,導致其分布面積大幅縮減,因此評估指出高山植群帶是全球氣候變遷趨勢下最脆弱的生態系之一,然而物種之存續尚繫於種內之遺傳多樣性及演化潛力以因應環境的改變,但鮮少研究探討高山植物分布面積縮小對物種遺傳多樣性的影響。本研究以繖形科山薰香屬植物為例,根據物種採樣點座標及WORLDCLIM氣候資料,以MAXENT程式建立物種分布模型,並將之投影至上次最大冰期(Last Glacial Maximum, LGM)的氣候模型及兩組預測的未來氣候情境,計算預測山薰香及台灣山薰香的分布,評估其脆弱度,並整合物種分布預測與親緣地理資料,評估在不同暖化情境下遺傳多樣性的喪失,其中南湖山薰香由於稀有而資料甚少,並未納入分析。預測結果顯示,山薰香及台灣山薰香在LGM時族群較現今大幅擴張,而在未來不同的暖化情境下,兩物種分別向上遷徙了18–497公尺及17–476公尺,分布面積則分別縮小了44.7–85.8%及49.2–99.9%,且以東北部山區的脆弱度最高而漸次向南遞減。根據分布面積喪失的預測,山薰香及台灣山薰香在葉綠體atpB-rbcL基因間片段之基因單型的多型性在未來分別將有5.3–21.1%及0–80%的喪失。由於東北部同時為遺傳多樣性的熱點,未來保育應加強東北部族群的監測,並針對即將喪失基因單型的族群進行採樣與保種,以維持物種之遺傳多樣性。
Studies have predicted that alpine plants will be forced to migrate into higher ele-vations as current trend of global warming continues, resulting in drastic deduction in their distribution range. Consequently alpine vegetation is regarded as one of the most vulnerable ecosystems threaten by global climate changes. However, species’ survivor-ship and evolvability to cope with environmental changes also depend on genetic diver-sity. Therefore it is also important to project how shrinking in distribution range would impact extent of genetic diversity. Based on coordinates of collecting data and climate data of WORLDCLIM, we employed the program MAXENT to construct the distribu-tion models of two endemic species of Apiaceae in high elevation of Taiwan, Chaerophyllum involucratum and C. taiwanianum, and used the models to project their past and future distributions during the last glacial maximum (LGM) and under the fu-ture climatic scenarios (i.e. B2B and A2A) predicted by Intergovernmental Panel on Climate Change (IPCC). Changes of projected distribution were used to evaluate the vulnerability of the species. The distributions were also integrated with phylogeographic data to predicted loss of genetic diversity under B2B and A2A. Based on the projections, distributions of both species were much wider during LGM compare to their present distributions. Under B2B and A2A, 18–497 m upward migration and 44.7–85.8% shrinking in distribution are projected in C. involucratum, while 17–476 m and 42.2–99.9% in C. taiwanianum are estimated. Populations in the northeastern part of the island are most vulnerable, with vulnerability decreasing southward. Based on deduction of distribution range, C. involucratum is projected to lose 5.3–21.1% of haplotypes in the chloroplast atpB-rbcL spacer, while 0–80% loss was predicted in C. taiwanianum. Because the northeastern populations also present hotspots of genetic diversity, their high vulnerability urges attention to monitor whether populations there would have shrunk as predicted. Furthermore, ex situ conservation such as germplasm preservation in seedbank should focus on populations harboring haplotypes that are projected to go extinct in the near future.