生物多样性 ›› 2018, Vol. 26 ›› Issue (12): 1268-1276.DOI: 10.17520/biods.2018019

• 研究报告 • 上一篇    下一篇

基于SOM的野生秤锤树群落的空间分布和环境解释

杨国栋1,2, 季芯悦1,2, 陈林1,2, 钟育谦3, 翟飞飞3, 伊贤贵1,2,*(), 王贤荣1,2   

  1. 1 南京林业大学南方现代林业协同创新中心, 南京 210037
    2 南京林业大学生物与环境学院, 南京 210037
    3 江苏省野生动植物保护站, 南京 210036
  • 收稿日期:2018-01-19 接受日期:2018-09-12 出版日期:2018-12-20 发布日期:2019-02-11
  • 通讯作者: 伊贤贵
  • 作者简介:# 共同第一作者
  • 基金资助:
    绿色江苏专项资金(2130205)和第二次全国重点保护野生植物资源调查专项资金(031010251)

Spatial distribution and environmental interpretation of wild Sinojackia xylocarpa communities based on self-organizing map (SOM)

Guodong Yang1,2, Xinyue Ji1,2, Lin Chen1,2, Yuqian Zhong3, Feifei Zhai3, Xiangui Yi1,2,*(), Xianrong Wang1,2   

  1. 1 Co-Innovation Center for the Sustainable Forestry in Southern China, Nanjing Forestry University, Nanjing 210037
    2 College of Biology and the Environmentand the Nanjing Forestry University, Nanjing 210037
    3 Protection Station of Wild Animals and Plants in Jiangsu Province, Nanjing 210036
  • Received:2018-01-19 Accepted:2018-09-12 Online:2018-12-20 Published:2019-02-11
  • Contact: Yi Xiangui
  • About author:# Co-first authors

摘要:

本文采用自组织特征映射网络(self-organizing map, SOM)对南京老山野生秤锤树(Sinojackia xylocarpa)群落进行数量分类和排序, 分析了其与环境因子之间的关系。结果表明: (1) SOM将秤锤树野生群落的100个样方划分为5个群丛类型, 分类结果在空间上反映了秤锤树野生群落的演替变化趋势, 各群丛的群落结构和物种组成存在差异且群丛界限明显, 可较好地进行排序与分类的环境解释。(2)通过环境因子梯度的可视化方法, 确定了海拔、坡位和土壤厚度是影响该地区秤锤树生长和分布的主要因子, 同时也揭示了以不同优势种为代表的各群丛和环境因子的关系。(3) SOM可以摆脱许多定量技术的限制性假设, 使得神经网络对于群落生态特征及探索群落和环境相互关系具有良好展现力; SOM群落生态数据具有更高的映射能力, 进行群落分类以及较少程度的排序的潜力, 将有利于不同群落类型的分类和管理, 对于濒危植物保护具有重要意义。

关键词: 秤锤树, 环境因子, 自组织特征映射网络, 野生群落, 梯度分析

Abstract:

A self-organizing map (SOM) based on field investigations was adopted to analyze the numerical classification and ordination of wild Sinojackia xylocarpa communities in the Nanjing Laoshan Forest Park in hopes of illuminating the relationship between the wild communities and environmental conditions. The results show that the 100 quadrats were divided into five associations, spatially reflecting the successional trend of wild S. xylocarpa communities. The association boundary, community structure and species composition differed significantly among communities. Through visualizing environmental gradients, the altitude, slope position and soil thickness were found to be the main factors affecting the growth and distribution of S. xylocarpa in this area, though the relationships differed among dominant species. The SOM removes the restrictive assumptions of many quantitative techniques so that the neural network is attractive to the community ecological characteristics and the interrelationship between community and environment can be explored. Based on the potential of SOM for vegetation data classification and, to a lesser extent, ordination, the SOM can aid in the conservation of endangered plants across different community types.

Key words: Sinojackia xylocarpa, environmental factor, self-organizing map, wild community, gradient analysis