生物多样性 ›› 2017, Vol. 25 ›› Issue (9): 999-1007.DOI: 10.17520/biods.2017086

• 研究报告: 植物多样性 • 上一篇    下一篇

基于倾向评分配比法评估苍山自然保护区的森林保护成效

陈冰, 刘方正, 张玉波, 杜金鸿, 王伟*(), 李俊生   

  1. 中国环境科学研究院生物多样性研究中心, 北京 100012
  • 收稿日期:2017-03-20 接受日期:2017-05-17 出版日期:2017-09-20 发布日期:2017-10-04
  • 通讯作者: 王伟
  • 作者简介:# 共同第一作者
  • 基金资助:
    国家自然科学基金(31300453)

Assessment of forest conservation in the Cangshan Nature Reserve based on propensity score matching

Bing Chen, Fangzheng Liu, Yubo Zhang, Jinhong Du, Wei Wang*(), Junsheng Li   

  1. Biodiversity Research Center, Chinese Research Academy of Environmental Sciences, Beijing 100012
  • Received:2017-03-20 Accepted:2017-05-17 Online:2017-09-20 Published:2017-10-04
  • Contact: Wang Wei
  • About author:# Co-first authors

摘要:

对自然保护区的定期评估有助于提升其管理和保护水平。本研究以苍山自然保护区所在的大理市、洱源县、漾濞县3个市/县作为子研究区域, 构建了基于倾向评分配比(propensity score matching)和配对t检验(paired-samples t-test)的保护成效评估技术方案。选取海拔、坡度、与最近居民点的距离、与最近道路的距离4个因子作为环境变量, 通过对每个县单独进行分析, 分别对比了自然保护区晋升国家级时(1995年)与20年后(2015年)其森林覆盖面积的变化, 以此来评估苍山自然保护区的森林保护成效, 并对各环境变量与森林覆盖变化值的关系进行了偏相关分析(partial correlation analysis)。结果表明: 大理市境内苍山自然保护区内的森林覆盖变化值显著高于保护区外部, 且该区段的森林覆盖率最高。洱源县境内苍山自然保护区内、外的森林覆盖变化率均高于其他2个市/县, 且保护区内、外森林覆盖变化值差异不显著。漾濞县境内苍山自然保护区内、外森林覆盖变化率最低, 但其保护区外0-10 km区域的森林覆盖变化值显著高于10 km以外区域, 保护区的存在对其周边0-10 km区域产生了正面的溢出效应 (neighborhood leakage)。海拔、坡度、与最近居民点的距离、与最近道路的距离4个协变量在3个市/县的不同研究区段内均与森林覆盖变化值呈现出了不同程度的相关性。本研究所采用的倾向评分配比法和按照行政区划对自然保护区分区进行评估的方法, 为自然保护区整体保护成效的评价提供了新的技术思路。

关键词: 倾向评分配比, 溢出效应, 保护成效, 森林覆盖变化

Abstract

The establishment of protected areas is one of the most common measures of biodiversity conservation. Regular assessment can help improve management and promote conservation in protected areas. According to administrative divisions, we divided the Cangshan Nature Reserve into three parts: Dali City, Eryuan County and Yangbi County, as sub-study areas. The evaluation was based on propensity score matching and paired-samples t-test. Elevation, slope, distance to the nearest settlement, and distance to the nearest road were four chosen covariates. Since the Cangshan Nature Reserve was upgraded to the national level in 1994, we compared forest changes between 1995 and 2015. Partial correlation analysis was carried out between each covariate and forest change to analyze the impact factors. Results indicated that in Dali City, the forest change value inside the Cangshan Nature Reserve was significantly higher than that found outside. Forest coverage inside the Cangshan Nature Reserve in Dali City was the highest among all regions. The forest change rates both inside and outside the Cangshan Nature Reserve in Eryuan County were higher than the other two counties. There was no significant difference in forest change value between areas inside and outside of the nature reserve in Eryuan County. In Yangbi County, the forest change rates both inside and outside of Cangshan Nature Reserve were the lowest among three counties, but the forest change value found within 10 km outside of the nature reserve was significantly higher than that found in areas beyond 10 km, which indicates that the existence of the Cangshan Nature Reserve performed positive neighborhood leakage in surrounding areas within 10 km. The four covariates all affected forest change in different areas in a variety of ways. The propensity score matching and sub-regional methodology for the assessment of the Cangshan Nature Reserve provided a new technical method and example for other conservation assessment studies.

Key words: propensity score matching, neighborhood leakage, assessment of conservation, forest coverage change