RS与GIS支持下城市热岛效应与绿地空间相关性研究

Study of Spatial Correlation between UHI and VF Based on RS and GIS

在RS和GIS技术支持下,以LANDSAT ETM+为数据源,综合运用遥感热红外影像的地温反演、遥感影像分割、空间聚集性分析、混合像元分解、基于遥感信息模型的地理建模及多元线性回归分析等技术,提取了青岛市城市热岛与绿地空间格局,并定量分析了绿地格局与城市热岛效应的相关性及其对城市热岛的缓解机制.结果表明:城市热岛效应在给定尺度上与植被盖度呈负相关关系,与绿地斑块密度指数和分维数呈正相关关系.

With the support of RS and GIS, a methodological framework employing techniques of land surface temperature （LST） retrieval, multiresolution segmentation, spatial aggregation calculation, pixel unmixing, geographic modeling and multivariate regression was developed and applied on a LANDSAT ETM ＋ image of Qingdao. The spatial pattern of heat island （HI） and green space were extracted and the correlation between them was quantitatively described. The result shows that there is a negative correlation between LST and vegetation fraction （VF） and positive correlation between LST and patch density （PD） and between LST and fractal dimension （FD） at the given scale.