基于Landsat影像的离子稀土矿区植被覆盖度提取及景观格局分析

离子稀土开采导致矿区植被退化和严重的生态环境问题,日益引起人们关注。采用23 a的Landsat系列影像作为数据源,以定南县岭北稀土矿区作为研究案例,对稀土开采扰动下的矿区植被覆盖景观格局变化进行分析。为提高分析的可靠性,以高分遥感影像获取的矿区植被覆盖度作为检验数据,比较森林郁闭度制图模型（forest canopy density mapping model,FCD）、像元二分法模型（dimidiate pixel model,DPM）和线性光谱混合模型（linear spectral mixture model,LSMM）对稀土矿区植被覆盖提取的准确性。结果表明：考虑了稀土矿区特有的裸露...

Full description

Saved in:
Bibliographic Details
Published in农业工程学报 Vol. 32; no. 10; pp. 267 - 276
Main Author 李恒凯 雷军 杨柳
Format Journal Article
LanguageChinese
Published 江西理工大学建筑与测绘工程学院,赣州,341000 2016
Subjects
景观格局
植被
模型
矿区
稀土
覆盖度
稀土
coverage
models
植被
landscape pattern
矿区
覆盖度
景观格局
模型
vegetation
mining area
rare earths
Online AccessGet full text
ISSN1002-6819
DOI10.11975/j.issn.1002-6819.2016.10.037

Cover

More Information
Summary:离子稀土开采导致矿区植被退化和严重的生态环境问题,日益引起人们关注。采用23 a的Landsat系列影像作为数据源,以定南县岭北稀土矿区作为研究案例,对稀土开采扰动下的矿区植被覆盖景观格局变化进行分析。为提高分析的可靠性,以高分遥感影像获取的矿区植被覆盖度作为检验数据,比较森林郁闭度制图模型（forest canopy density mapping model,FCD）、像元二分法模型（dimidiate pixel model,DPM）和线性光谱混合模型（linear spectral mixture model,LSMM）对稀土矿区植被覆盖提取的准确性。结果表明：考虑了稀土矿区特有的裸露尾沙地光谱特征的三端元LSMM,具有更高准确性及普适性;景观格局动态分析表明矿区低植被覆盖和高植被覆盖度破碎化程度较低,集中连片趋势明显;低植被覆盖LPI（largest patch index）在4个年份相对较大,呈波动变化,主要为裸露的稀土矿点,与矿点复垦和扩张有密切关系,且单个矿点造成的植被破坏仍然较大;低植被覆盖度的AI（aggregation index）一直较高,而LPI相对高植被覆盖较小,主要为离子稀土以单个矿点小面积开采为主,矿点多而分散,导致对环境破坏更难治理;1999年后,随着稀土开采的规模化及开采工艺的改进,稀土开采成为矿区景观变化的最主要原因,在矿区总体生态环境改善的前提下,低植被覆盖区域主要集中在矿点及矿点周边,成为具有稀土开采特色的矿点景观分布模式。
Bibliography:11-2047/S
vegetation; rare earths; models; mining area; coverage; landscape pattern
Mining rare earth ore has brought vegetation degradation, serious ecological and environmental problems in mining area, and it has already received more and more attentions and concerns by the society. By using 23 years Landsat series remote sensing images as data source, and choosing Lingbei rare earth mining area in Dingnan County as study area,in this paper, we analyzed the change of vegetation coverage of landscape pattern under the disturbance of rare earth mining. In order to improve the reliability of analyzing landscape pattern, the vegetation coverage of study area was extracted with four methods of forest canopy density mapping model（FCD）, dimidiate pixel model（DPM）, three-endmember and four-endmember linear spectral mixture model（LSMM）. Taking vegetation coverage of the studied area acquired from the high-spatial-resolution remote sensing image of Pleiades as the checking data, the most suitable method of vegetation c
ISSN:1002-6819
DOI:10.11975/j.issn.1002-6819.2016.10.037