Study On Spatial-temporal Patterns Of Impervious Surfaces In Beijing And Their Thermal Environmental Effects

In the process of urbanization,as the city continues to expand and develop,the impervious surface area of the city also increases,to a certain extent,it will have a greater impact on the local regional climate and thermal environment changes in the city.Analyze and study the expansion characteristics of urban impervious surfaces and their impact on the urban thermal environment,not only to realize dynamic monitoring of urban space expansion and analysis of driving force mechanisms,but also to analyze and understand the laws of urban surface temperature changes and provide them for urban development planning Data and theory support.In this paper,Beijing is selected as the research area,based on Spatio-temporal Image Fusion Model(STI-FM)and linear and nonlinear regression analysis and random forest classification methods,using Moderate-resolution Imaging Spectrometer(Moderate-resolution Imaging Spectrometer)Spectroradiometer(MODIS),Landsat 8(Landsat-8)multi-source remote sensing data set,Beijing meteorological station data and Beijing air quality data,from the qualitative and quantitative perspectives of the study area impervious surface expansion and urban thermal environment The characteristics of spatiotemporal changes,and analyze the relationship between the two and the various types of driving forces that affect the change of Land Surface Temperature(LST).The main research results are as follows:(1)From the perspective of the impervious surface expansion area,Beijing increased by 719.35km2 in 2014-2019.The expansion rate of Beijing in 2014-2017 was significantly higher than that in 2017-2019,from 13.56 in 2014-2017 % Slowed to 7.57% from 2017 to 2019;from the perspective of the type of expansion rate of impervious surface area,from 2014 to 2017 and from 2017 to 2019,Beijing's impervious surface is in a type of rapid expansion,from 2014 to 2017 The annual expansion is 110.03km2 annually,and the annual expansion from 2017 to 2019 is93.07km2.The construction area of Beijing shows an increasing trend,but urban expansion has slowed in the past two years.(2)Based on the STI-FM spatiotemporal fusion model,the quarterly LST sequence image maps of Beijing from 2014 to 2019 are obtained,and from the interannual change law of Beijing LST,it is found that the differences in the fluctuations of the annual sequence of LST in each municipal jurisdiction from 2014 to 2019 are relatively Large,with higher LST temperatures in 2014,2017 and 2019,respectively,21.47 ?,20.78 ? and 20.08 ?,and lower average annual LST in 2015,2016 and 2018 were 15.74 ?,15.43 ? and 17.92,respectively ?,the whole shows a trend of volatility that decreases first and then increases.(3)In 2014 and 2019,there is a positive correlation between the impervious surface expansion and vegetation reduction in the Chaoyang District of the central urban area.The fitting results show that the coefficient of determination(Coefficient Of Determination,R2)is 0.8896,and the root mean square error(Root Mean Square Error(RMSE)is 30.Daxing DistrictThe fitting results of the sample data show that the linear relationship between the two,R2 is 0.3677,RMSE is 119.7;the distribution of scatter plots in 2014 and2019 is concentrated;there is a strong positive correlation between the impervious surface and the spatial distribution of NDBI,R2 in 2014 It is 07963,RMSE is 97.35,2019 R2 is 0.6855,RMSE is 142.1.(4)Study the three surface areas of Normalized Difference Vegetation Index(NDVI),Normalized Difference Build-up Index(NDBI)and Normalized Difference Water Index(NDWI)The effect of parameters on LST found that: NDVI and surface temperature in summer are negatively correlated,NDVI and surface temperature in winter are positively correlated,and summer correlation is higher than winter;NDBILST is positively correlated in summer and winter and summer correlation is higher than winter,summer The NDBI value is mainly concentrated in the interval of-0.1and-0.3,and the winter NDBI value is mainly concentrated in the interval of-0.05and-0.1;the scatter plot distribution of NDWI and LST in 2014 and 2019 has a great relationship with the season,In 2014 and 2019,the overall summer NDWI was positively correlated with LST,and in 2014 and 2019 winter NDWI was negatively correlated with LST.