关键词: 桫椤, 气候变化情景, MaxEnt模型, 环境因子, 潜在地理分布

Abstract: Alsophila spinulosa, a relict plant species of early Tertiary glaciers and one of the oldest living fossils in the world, is of great value for understanding changes in paleoclimate, species evolution, and geographic flora. Based on 111 effective distribution records ofA. spinulosa in China and 10 environmental variables, the MaxEnt model and ArcGIS software were used to predict the potential geographic distribution of A. spinulosa. The relative contribution of comprehensive environmental variables and jackknife method were used to assess the essential factors that restrict the potential geographic distribution of A. spinulosa. Meanwhile, the response curve was constructed to determine the appropriate values of environmental variables as well as the potential geographic distribution location and area of A. spinulosa. Our results showed that the prediction accuracy of the MaxEnt model was high, with the receiver operating curve area (AUC value) reaching 0.959. The total suitable habitat area for the potential geographic distribution of A. spinulosa was 160.15×10⁴ km², mainly located in southern and southeastern China. The main environmental factors affecting the potential geographic distribution of A. spinulosa were temperature factor (mean temperature difference per day and mean temperature of the driest season), precipitation factor (precipitation of the driest season), and topographic factor (slope), with precipitation in the driest season as the most essential one. Under future climate change scenarios, the total area of suitable habitat will increase, the area of highly and moderately suitable habitats will decrease, and the area of lowly suitable habitats will increase. The center of gravity in the highly suitable habitat of A. spinulosa will shift to the northwest and high-latitude regions. Our results provide a theoretical basis for the management and resource protection of A. spinulosa.

Key words: Alsophila spinulosa, climate change scenario, MaxEnt model, environmental variable, potential geographical distribution.