摘要
为了研究2个或多个要素之间的相关性,以金华市为例,采用分层多尺度方法对本研究区的地表温度与归一化植被指数(NDVI)及高程之间的相关性进行研究.结果表明:在金华市本级范围内进行相关性分析,地表温度与归一化植被指数及高程之间存在着明显的负相关关系;在4种分辨率(30,100,300,1 000 m)下,地表温度与归一化植被指数之间存在着弱的负相关关系,而地表温度与高程之间存在着较强的负相关关系;在较小研究区域内,相关系数变化范围会随着粒度的增加和研究区域的减小而扩大.因此,要素之间的相关性大小与研究尺度密切相关.
In order to study the relationship between two or more elements,taking Jinhua City as an example,the correlation between land surface temperature and the normalized difference vegetation index and elevation was studied by the hierarchical multi-scale approach.The results showed that the land surface temperature of Jinhua City was related to the normalized difference vegetation index and elevation,and there was a clear negative correlation between them;At four resolutions(30,100,300 and 1 000 m),there was a weak negative correlation between land surface temperature and normalized difference vegetation index,and there was a strong negative correlation between land surface temperature and elevation.Within the scope of the smaller study area,the range of correlation coefficient increased with the increase of particle size and the decrease of the study area.Therefore,the correlation between factors was closely related to the scale of the research.
作者
程姌
康育龙
梁勤欧
CHENG Ran;KANG Yulong;LIANG Qin′ou(College of Geography and Environmental Sciences,Zhejiang Normal University,Jinhua 321004,China)
出处
《浙江师范大学学报（自然科学版）》
CAS
2018年第4期459-465,共7页
Journal of Zhejiang Normal University:Natural Sciences
基金
国家自然科学基金资助项目(70773089)
关键词
分层多尺度方法
相关性分析
可塑性面积单元问题
地表温度
归一化植被指数
高程
hierarchical multi-scale approach
correlation analysis
modifiable areal unit problem
land surface temperature
the normalized difference vegetation index
elevation
