地学前缘 ›› 2021, Vol. 28 ›› Issue (4): 229-237.DOI: 10.13745/j.esf.sf.2020.10.19

• 水生态、草地生态及污染土壤修复 • 上一篇    下一篇

数据挖掘算法在高寒草地退化驱动因素相关性分析中的应用

马蓉蓉1(), 杨国柱2, 胡月明3, 周伟1,4,5,*()   

  1. 1.中国地质大学(北京) 土地科学技术学院, 北京 100083
    2.青海大学 生态环境工程学院, 青海 西宁 810016
    3.华南农业大学 资源环境学院, 广东 广州 510642
    4.自然资源部土地整治重点实验室, 北京 100035
    5.自然资源部 矿区生态修复工程技术创新中心, 北京 100083
  • 收稿日期:2020-09-29 修回日期:2020-12-06 出版日期:2021-07-25 发布日期:2021-07-25
  • 通讯作者: 周伟
  • 作者简介:马蓉蓉(1995—),女,博士研究生,公共管理专业,主要从事土地整治与生态修复研究。E-mail: 870940391@qq.com
  • 基金资助:
    国家自然科学基金项目“青藏高原露天煤矿排土场地形-土壤-植被响应机理及地貌重塑研究”(41977415)

Application of data mining algorithm in correlation analysis of the driving factors for alpine grassland degradation

MA Rongrong1(), YANG Guozhu2, HU Yueming3, ZHOU Wei1,4,5,*()   

  1. 1. School of Land Science and Technology, China University of Geosciences (Beijing), Beijing 100083, China
    2. College of Eco-Environmental Engineering, Qinghai University, Xining 810016, China
    3. College of Natural Resources and Environment, South China Agricultural University, Guangzhou 510642, China
    4. Key Laboratory of Land Consolidation and Rehabilitation, Ministry of Natural Resources, Beijing 100035, China
    5. Technology Innovation Center for Ecological Restoration Engineering in Mining Area, Ministry of Natural Resources, Beijing 100083, China
  • Received:2020-09-29 Revised:2020-12-06 Online:2021-07-25 Published:2021-07-25
  • Contact: ZHOU Wei

摘要:

高寒草地的退化受到众多自然、人为活动的影响,退化与驱动因素之间的耦合关系复杂。本文以青海省称多县为研究区,提取2005—2014年归一化植被指数(NDVI,normalized difference vegetation index)时间序列数据集,结合温度、降水、社会经济因素,运用基于数据挖掘的提升度算法进行相关性分析,研究影响高寒草地退化指标与表示高寒草地退化指标之间两者的关系。本文采用提升度算法针对3个等级的NDVI、可食量、植株高度与相应的温度、降水、鼠害和放牧强度之间的关系进行分析,能够更精确地分析各驱动因子在不同等级取值范围下对草地不同等级退化的贡献率,发现驱动因素与草地退化之间的影响关系并不是单向的,而是到达一定的程度时会出现逆向影响关系。本研究得到如下结论:(1)草地植被覆盖度低与气温和降水存在负相关;(2)草地可食量低与气温和人口存在负相关,与牲畜存在正相关;(3)地上植株高度低与牲畜呈现正相关。

关键词: 归一化植被指数(NDVI), 数据挖掘, 提升度, 相关性分析, 高寒草地, 退化指标

Abstract:

The degradation of alpine grassland is affected by a range of natural and human activities, and the coupling relationship between degradation and its driving factors is complex. To study the relationship between the influencing factors and indicators for alpine grassland degradation, this paper refers to Duo County as the research area, extracts the NDVI time series data from 2005 to 2014, combines the temperature, precipitation and socio-economic factors, and uses the data mining-based lift caluation in the correlation analysis. The relationships between the three levels of NDVI, food availability or plant height and the corresponding temperature, precipitation, rodent or grazing intensity were examined, so as to more accurately analyze the contribution rate of each driving factor to different grades of grassland degradation. The conclusions are: (1)The low vegetation coverage in grassland is negatively correlated with temperature and precipitation. (2)The low food availability in grassland is negatively correlated with temperature and population, and positively correlated with livestock. (3)The low plant height is positively correlated with livestock.

Key words: NDVI, data mining, lift calculation, correlation analysis, alpine grassland, degradation index

中图分类号: