植物生态学报 ›› 2016, Vol. 40 ›› Issue (11): 1164-1178.doi: 10.17521/cjpe.2016.0032

• 研究论文 • 上一篇    下一篇

小叶栎分布格局对末次盛冰期以来气候变化的响应

李垚, 张兴旺, 方炎明*()   

  1. 南京林业大学南方现代林业协同创新中心, 南京 210037;南京林业大学生物与环境学院, 南京 210037
  • 收稿日期:2016-01-17 接受日期:2016-05-09 出版日期:2016-11-10 发布日期:2016-11-25
  • 通讯作者: 方炎明 E-mail:jwu4@njfu.edu.cn
  • 基金资助:
    国家自然科学基金(31370666)、江苏高校优势学科建设工程资助项目(PAPD)和江苏省研究生培养创新工程项目(KYLX15_0922)

Responses of the distribution pattern of Quercus chenii to climate change following the Last Glacial Maximum

Yao LI, Xing-Wang ZHANG, Yan-Ming FANG*()   

  1. Co-Innovation Center for the Sustainable Forestry in Southern China, Nanjing Forestry University, Nanjing 210037, China;and College of Biology and the Environment, Nanjing Forestry University, Nanjing 210037, China
  • Received:2016-01-17 Accepted:2016-05-09 Online:2016-11-10 Published:2016-11-25
  • Contact: Yan-Ming FANG E-mail:jwu4@njfu.edu.cn

摘要:

小叶栎(Quercus chenii)是华东植物区系的代表树种, 具有很高的生态、经济价值。为重建冰期以来小叶栎地理分布格局的变迁历史、了解环境因子对潜在地理分布的制约机制, 为小叶栎种质资源保护和管理提供科学依据, 该研究基于55条分布记录和8个环境变量, 利用MaxEnt模型模拟小叶栎在末次盛冰期、全新世中期、现代和2070年(温室气体排放情景为典型浓度目标8.5)的潜在分布区, 利用多元环境相似度面和最不相似变量分析探讨气候变迁过程中环境异常区域和引起潜在地理分布改变的关键因素, 综合应用贡献率及置换重要值比较、Jackknife检验评估制约现代地理分布的主要因子, 采用响应曲线确定环境变量的适宜区间。研究结果表明: MaxEnt模型的预测准确度极高, 受试者工作特征曲线下的面积(AUC值)达0.9869 ± 0.0045; 现代高度适宜区在安徽南部、浙江西部、江西东北部和湖北东部; 影响小叶栎地理分布的主要气候因子为气温和降水量, 气温更重要; 最干季平均气温可能是制约小叶栎向北分布的关键因素; 末次盛冰期时, 小叶栎高度适宜区位于东海大陆架内; 全新世中期适宜分布区轮廓已与现代近似; 2070年适宜分布区向北移, 高度适宜区面积增大, 与末次盛冰期、全新世中期和现代相比, 这一时期的气候异常程度最高。气温季节变化和降水季节变化可能是引起地理分布变迁的重要气候因素。

关键词: 小叶栎, MaxEnt模型, 末次盛冰期, 气候变化, 分布格局

Abstract:
Aims Quercus chenii is a representative species of the flora in East China, with high ecological and economic values. Here, we aim to simulate the changes in the distribution pattern of this tree species following the Last Glacial Maximum (LGM) and to explore how climatic factors constrain the potential distribution, so as to provide scientific basis for protection and management of the germplasm resources in Q. chenii.
Methods Based on 55 presence point records and data on eight environmental variables, we simulated the potential distribution of Q. chenii during the Last Glacial Maximum, mid-Holocene, present and the year 2070 (the scenario of greenhouse gas emission is Representative Concentration Pathway 8.5) with MaxEnt model. The novel climate area and main factors influencing the changes in distribution pattern were evaluated by multivariate environmental similarity surface analysis and the most dissimilar variable analysis. The importance of environmental variables was evaluated by percent contribution, permutation importance and Jackknife test. Response curves were used to estimate the suitable value range of each variable.
Important findings The accuracy of MaxEnt model is very high, as indicated by the value of the area under the receiver operator characteristic curve of 0.9869 ± 0.0045. The highly suitable region for the present distribution covers southern Anhui, western Zhejiang, northeastern Jiangxi and eastern Hubei. The main factors affecting the potential distribution of Q. chenii are temperature and precipitation, with the former being more important. Mean temperature of the driest quarter is likely the main factor restricting Q. chenii growing in the north. During the LGM, the East China Sea Shelf occurs as the highly suitable region for the distribution of Q. chenii. In the mid-Holocene, the outline of the suitable area for the distribution of Q. chenii is similar to the present. The potential distribution region will likely move northward and experience an area expansion under the climate condition in 2070. At that time, climate anomaly will also be most severe compared to the LGM, mid-Holocene and present. Temperature seasonality and precipitation seasonality may be the main climatic factors promoting changes in the distribution pattern of Q. chenii.

Key words: Quercus chenii, MaxEnt model, Last Glacial Maximum, climate change, distribution pattern