Study On The Spatial Pattern Of Urban Heat Islands And Its Influencing Factors Based On LCZ

Rapid urbanization has led to the contradiction between urban expansion and urban heat islands,and it is urgent to mitigate the urban heat island effect.Among them,urban spatial structure is an important factor causing the urban heat island effect and spatial differences in urban heat islands,and how to regulate it can effectively mitigate the urban heat island effect has become a hot topic of research today.In this study,the central city of Jinan,a densely built area,was used as the study area.Firstly,urban road,DEM data,Baidu building big data,remote sensing images,and land cover data were used to carry out Local Climate Zones(LCZ)at the block scale and construct four urban spatial structure indicators including one-dimensional height,two-dimensional plane,three-dimensional space,and urban greening ecology.Secondly,the GEE cloud platform and Kriging interpolation method were applied to obtain the land surface temperature and the average air temperature at four moments respectively,and to analyze the spatial distribution characteristics of urban heat islands and the differences between and within classes of urban heat islands based on LCZ at different moments and among different data sources;Thirdly,correlation analysis and spatial regression analysis were applied to quantitatively investigate the influence of urban spatial structure indicators on urban heat islands;Finally,analysis of variance and multiple correspondence analysis were utilized to qualitatively investigate the influence of internal variability of urban spatial structure indicators on urban heat islands and the correspondence between the two internal categories.The research results showed that:(1)The overall spatial pattern of land surface temperature in Jinan city was high in the north and low in the south,and the high-temperature area was mainly distributed in the dense building area in the north of the study area,and the low-temperature area was mainly distributed in the mountainous area and part of the water body area in the south of the study area;The air temperature at 4:00 a.m.and 8:00 p.m.both showed a decreasing trend from the center of the study area to the periphery.The high-temperature areas of 10:00 a.m.and 2:00 p.m.were located in the southern part of the interpolation area of the study area,and the lowest temperature was distributed in the southern and northern boundaries of the interpolation area of the study area.From 4:00 a.m.to 8:00p.m.,the spatial distribution of the urban atmospheric heat islands first showed a decreasing trend from the center to the outside,and then the high-temperature area contracted,and the middle and low temperatures spread to the most of interpolation area of the study area.Most of the interpolation area,and finally showed a decreasing trend from the center to the outside.(2)The mean land surface temperature fluctuated widely between LCZ types,and the mean land surface temperature fluctuated more sharply in the natural type of LCZ than in the building type of LCZ.The maximum mean land surface temperature appeared in the low-level high-density low-green class,and the minimum value were presented in the water class.The minimum values of the mean air temperature at 4:00 a.m.and 8:00 p.m.were all seen in the middle-level low-density high-green class;the maximum mean air temperature at 10:00 a.m.and 2:00 p.m.showed in the high-level medium-density high-green class,and the minimum value appeared in the low vegetation class.(3)Land surface temperature was highly significantly and negatively correlated with average vegetation cover,DEM mean,and DEM standard deviation,while it was highly significantly and positively correlated with building density and standard deviation of the building footprint,and air temperature at 4:00 a.m.was highly significantly and positively correlated with volume ratio,building density,DEM mean,and DEM standard deviation,while it was highly significantly and negatively correlated with average vegetation cover,and air temperature at 10:00 a.m.,2:00 p.m.and 8:00 p.m.showed highly significant positive correlations with building density and floor area ratio,and highly significant negative correlations with average vegetation cover,so the urban heat island effect can be mitigated by increasing average vegetation cover,reducing building density and land use intensity.(4)The land surface temperature and air temperature were greatly affected by the land surface temperature and air temperature of the neighborhood,respectively,and were less affected by the urban spatial structure indexes.Reducing the standard deviation of building footprint,building density,and floor area ratio could significantly reduce the land surface temperature.Decreasing the building density of the region and the standard deviation of building height and floor area ratio of the neighboring regions could significantly cut the air temperature at 4:00 a.m.while decreasing the building density of the region and the floor area ratio of the neighboring regions could significantly lower the air temperature at 2:00 p.m.and8:00 p.m.(5)The individual urban spatial structure indicators were all important influencing factors of land surface temperature and air temperature at four moments,and their one-way analysis of variance obtained a significance level of 0.05 and above with land surface temperature and air temperature at four moments;In the results of the two-indicator interaction analysis of variance of urban spatial structure indicators with land surface temperature and air temperature at four moments,the interaction of DEM mean and standard deviation with building height mean,building density,volume ratio and water coverage,respectively,and the interaction of volume ratio and water coverage had significant effects on land surface temperature and air temperature at four moments.(6)Land surface temperature in the low-value category corresponded to DEM standard deviation and vegetation cover in the high-value category,as well as building density,floor area ratio,and average building base area in the low-value category;Land surface temperature in the high-value category tended to correspond to DEM mean,DEM standard deviation,building base area mean,building volume mean,building volume standard deviation and vegetation cover index in the medium-low or low-value category and building base area standard deviation and floor area ratio in the medium-high or medium-value category.The lower the air temperature category values,the lower the building base area,building density,and volume ratio category values,and the higher the vegetation cover category values;the high-value categories of air temperature at4:00 a.m.and 8:00 p.m.tended to correspond to the medium-high or high-value categories of building height mean,building height standard deviation and building volume mean;the high-value categories of air temperature at 10:00 a.m.and 2:00 p.m.often corresponded to the medium-high value classes of DEM mean,DEM standard deviation,building base area mean and vegetation cover,and the medium-low or low-value classes of building base area standard deviation and building density.This study was based on the perspective of LCZ,which was helpful to have a better understanding of the spatial pattern and spatial differentiation of urban heat islands and also facilitated the construction of an urban spatial structure index system that influences urban heat islands accordingly.The quantitative and qualitative approaches were combined to investigate the influence of urban spatial structure on urban heat islands,which not only explored a new perspective for studying urban heat islands but also provided theoretical support for rational urban planning and mitigation of the urban heat island effect,as well as a reference for sustainable urban development and creating a good human living environment.