摘要
以位于草原区的霍林河露天煤矿为例,基于决策树和BP神经网络,在荒漠化信息提取的基础上,对研究区16年来的荒漠化情况进行了分析。结果表明:①受气候变化、超载放牧、掏挖药材以及垦荒等因素的影响,轻度荒漠化、中度荒漠化以及荒漠化总面积存在由强转弱、再由弱变强的演变过程;②受煤炭开发的影响,重度荒漠化草地多围绕矿业建设用地呈“”状分布,并且受季风影响向矿业建设用地东侧发展迅速。矿业建设用地面积与重度荒漠化面积相关关系显著,可将其作为评价、预测该区煤炭开发对生态环境影响的重要依据之一。
The grassland desertification information in Huolinhe opencast coalmine is extracted from remote sensing imageries by using decision tree and back-propagation (BP) neural network, and the desertification during the past 16 years in the research field is analyzed. The results indicate that (1) there is a strong to weak, then weak to strong evolution process for the total grassland desertification, middle desertification and slight desertification, which can be attributed to climate change and human activities, especially to the influence of climate change; (2) due to coal exploitation, the severe desertification is always distributing in " →▲ " around the mining land, and developing to the east quickly with the help of monsoon. There is much correlation between the mining land and severe grassland desertification area, and the mining land area can be served as an important indicator for coalfield eco-environment assessment and prediction.
出处
《辽宁工程技术大学学报(自然科学版)》
EI
CAS
北大核心
2006年第6期936-939,共4页
Journal of Liaoning Technical University (Natural Science)
基金
国家自然科学基金资助项目(40071045)
关键词
采矿扰动
草地荒漠化
信息提取
遥感
霍林河露天煤矿区
mining exploitation, grassland desertification
information extracting
remote sensing
Huolinhe open-cast coalmine