基于灰色聚类评估法的城市热环境空间分析

随着城镇化水平不断提高和全球气候变暖,城市热环境越来越受到关注。热环境是无法精确预知与预测的灰色系统。文章以城市热环境为切入点,对影响热环境的各驱动因子进行灰色优势分析,结果表明:在地理位置、日照、海拔、大气状况一定的条件下,城镇化水平对热环境起着决定性的影响,城镇空间扩展面与热环境过载区域基本吻合,除此之外绿地指数、地表层湿度、工业用地率的灰色优势度也较高,而地区生产总值强度,工农业总产值强度,人口密度的灰色优势度较低。在灰色优势度分析基础上对灰色聚类方法进行优化,利用优化的灰色聚类方法对研究区进行热环境评估分级,其与遥感反演地温平均重合率83%,更能反映热环境的形成原因。由于灰色聚类评估还存在主观因素影响如白化权函数的确定等以及遥感数据的不确定性,模型中的定量精确度有待进一步研究。

Spatial Analysis of Urban Thermal Environment Based on Grey Clustering Method

More and more concerns have been attached to urban thermal environment due to the rapid development of urbanization and global warming.Urban thermal environment is regarded as being inscrutable so it belongs to grey system and the grey advantage analysis can be made with several driving factors of thermal environment.An experimental analysis was conducted in an area of south part of Guangzhou,and the result showed that the level of urbanization was the decisive factor,among other things,that led to rising temperature in the study area.The analysis also suggested that the spatial urbanization was coincided with the thermal over-stressing area;in addition,the green land index followed by earth surface humidity and the industrial land ratio were the influencing factors as well.On the other hand,the grey correlation coefficients of GDP intensity,density of population and the intensity of gross industrial and agricultural output were comparatively low.A further study conducted was about the optimization of grey clustering method on the basis of grey advantage analysis.Compared with the remote-sensing land surface temperature inversion,the total co-incidence rate was 83%.There exist some subjective factors with the grey clustering method,such as the problem with ascertaining the whitenization weight function and the uncertainty of remote-sensing data,therefore efforts should be made to upgrade its quantitative precision.