基于POI的多尺度西安市热岛效应分析

Xi’an multiscale heat island effect analysis based on points of interest

以西安市为研究区,利用2016年Landsat 8遥感影像数据反演西安市地表温度,通过多元线性回归分析分别在0.5、1.0、1.5、2.0 km尺度上对同时期地表温度与用户兴趣点(POI)密度、水体盖度以及植被盖度进行相关分析,通过逐步回归法定量分析人类活动因子模型与生态基础设施模型对地表温度的解释程度.结果表明,在4个空间尺度上POI密度、植被盖度与地表温度存在极显著的相关关系;相同空间尺度下,生态基础设施模型解释率高于POI模型;两模型的交叉解释率随空间尺度增大逐渐高于二者独立解释率之和;加入POI因子后,全模型对西安市地表温度的解释率较生态基础设施模型显著提高.

Xi’an city was selected as the research area.Landsat 8 remote sensing image data of 2016 was used to retrieve the surface temperature in it.The multivariate linear regression analysis was conducted on four spatial scales of 0.5,1.0,1.5 and 2.0 km respectively.Temperature and point of interest(POI)density,water cover and vegetation cover were analyzed.The degree of interpretation of surface temperature by quantitative analysis of human activity factor model and ecological infrastructure model was analyzed by stepwise regression,and the results showed that there was a significant correlation between POI density and vegetation coverage and surface temperature data at four spatial scales.Interpretation rate of ecological infrastructure model was higher than that of POI model under the same spatial scale.The cross-interpretation rate of two models with a space scale gradually increased more than the sum of the two independent interpretation rates;after POI factor was added,the interpretation rate of the whole model to the surface temperature was significantly higher than that of the ecological infrastructure model.