城市地表温度和植被覆盖度的空间动态性——以佳木斯为例

Spatial Dynamics of Urban Surface Temperature and Vegetation Coverage: A Case in Jiamusi

以佳木斯为研究区域,利用辐射传导法反演地表温度(LST),并从两方面研究LST的空间特性:一是运用半变异函数分析LST本身的尺度效应,即空间动态性;二是利用降采样和空间自相关回归模型,研究在不同尺度下,植被覆盖度(FVC)和LST的相关性随尺度变化的同时,FVC会对LST产生何种程度的影响。研究结果表明:LST和FVC存在尺度效应,空间自相关尺度在300 m左右;LST与FVC之间具有明显的负相关关系,空间间隔在300 m附近时,二者之间的关系在回归模型中呈主导趋势。所以在分析某一区域的特定现象时,应当选择合理的观测尺度。

The land surface temperature(LST)of Jiamusi is retrieved by radiation conduction method,and the spatial characteristics of LST are studied from two sides:one is to use semi-variogram to analyze the scale effect of LST itself,i.e.spatial dynamics;the other is to study the correlation between vegetation coverage(FVC)and LST at different scales by using down-sampling and spatial autocorrelation regression model.LST and FVC have scale effect,and the spatial autocorrelation scale is about 300 m.There is a significant negative correlation between LST and FVC.When the spatial interval is around 300 m,the relationship between LST and FVC is dominant in the regression model.Therefore,when studying and analyzing the specific phenomena in a certain area,we should choose a reasonable observation scale.