Spatio-temporal Evolution And Multi-factor Response Of Land Surface Temperature In Typical Urban Agglomerations In China

Land surface temperature(LST)is one of the key parameters of the energy balance of the Earth's land surface system.Comprehensive and accurate monitoring of LST and revealing the dominant influence factors of LST are essential to alleviate the urban thermal environment.At present,in the process of suburbanization and rapid urbanization,the urban agglomeration formed by the central city driving the surrounding cities has gradually become the main form of the country's urbanization development.The formation of urban agglomeration is an objective reflection of economic development and industrial layout.The temporal and spatial evolution of LST and influencing factors of typical urban agglomerations in Chinese are different.Indepth analysis of the multiple influencing factors on LST of urban agglomerations,exploring the temporal and spatial evolution rules are critical to assessing and mitigating the impact of urban agglomerations on the thermal environment.This thesis takes the Yangtze River Delta,Beijing-Tianjin-Hebei,GuangdongHong Kong-Macao and Chengdu-Chongqing four typical national urban agglomerations(including 62 cities)as the study area.Based on long-term MODIS surface temperature data from 2000 to 2015,at intervals of 5 years,10 potential influencing factors of surface temperature are selected from four levels of greenness,grayscale,humidity,and socio-economic factors: normalized difference vegetation index(NDVI),digital elevation model(DEM),slope(SLO),impervious surface area percentage(ISA),waterway density(WD),gross domestic production(GDP),nighttime light intensity(NTI),population density(POP),road density(RD),railway density(RWD).By using percolation-based city clustering algorithm(PCCA),boosted regression trees(BRT)model,the quantitative effect and contribution of LST were studied,and the threshold point of effect was obtained.Then study the differentiation characteristics within and among urban agglomerations.The main results and conclusions of this study are as follows:(1)The temporal and spatial evolution of LST in typical urban agglomerations from 2000 to 2015 has its own characteristics.From the perspective of time patterns,the LST of the Yangtze River Delta urban agglomeration mainly showed a trend of expanding outwards with Shanghai as the center,while the southern and northwestern regions showed a relatively cooling trend.In the Beijing-Tianjin-Hebei urban agglomeration,the LST increased significantly in the southwest and east,especially in the south of Tangshan,the east of Langfang and the coastal areas in the northeast of Cangzhou.The LST in Beijing and the surrounding cities such as Tangshan and Qinhuangdao did not increase but decreased.The LST of the Guangdong-Hong KongMacao urban agglomeration showed an upward trend,with a large increase and a wide coverage.The change of LST in Chengdu-Chongqing urban agglomeration showed a distribution characteristic of "center rising and both sides falling".From the perspective of spatial patterns,the distribution of high temperature areas is different in different years,mainly concentrated in counties or towns and densely populated areas.(2)In each city,the impact of different factors on LST is quite different.In general,socio-economic factors explain LST the most,followed by humidity factors.In the Yangtze River Delta urban agglomeration,coastal cities such as Nantong and Yancheng,the contribution rate of socio-economic factors is relatively high,up to 25.95%,indicating that the social economy of coastal cities has a greater impact on LST.For inland cities,the contribution rate of greenness factor and gray factor is relatively low,and some cities are less than 1%.In the Beijing-Tianjin-Hebei urban agglomeration,the central core functional areas,such as Cangzhou,Langfang and Tangshan near the Bohai Sea showed a relatively consistent distribution of contribution rates.For the northwestern ecological conservation areas,the contribution rate of the gray factor DEM is relatively high,up to 42.93%.Among the Guangdong-Hong Kong-Macao urban agglomerations,the Pearl River Economic Circle such as Zhongshan and Zhuhai,the contribution rate of socio-economic factors is relatively high,ranging from 9% to24%,while the contribution rate of greenness factor is relatively low,only 4.48%.For the Chengdu-Chongqing urban agglomeration,the contribution rates of POP and NTI among the socio-economic factors of the Chengdu metropolitan area is relatively high,up to 15.45%.The contribution rate of the greenness factor is relatively low,only 4.38%,while for Guang'an and Suining in the dense urban area of Nansuiguang City,showed a more consistent distribution pattern.Local governments should strengthen ecological and environmental protection while considering economic and social development.(3)According to the different effects of the dominant factors in each urban agglomeration,the corresponding threshold points were extracted to evaluate the sensitivity of each influencing factor.In the Yangtze River Delta urban agglomeration,socio-economic factors were dominant,and the threshold points of NTI and GDP factors decreased year by year,indicating that they were more sensitive to the influence of LST.In the Beijing-Tianjin-Hebei urban agglomeration,the grayscale and humidity factors were the dominant factors.Changes in topography and water systems have a relatively high influence on LST.Increasing land coverage and regional humidity can allevia environmental problems caused by LST.The Guangdong-Hong Kong-Maca area is dominated by humidity and socio-economic factors,among which the threshold points of POP factor from 2000 to 2015 were 2614.55,2941.45,2886.95 and 2941.45(person/km2).The socio-economic factors in the Chengdu-Chongqing urban agglomeration are dominant,especially the threshold points from 2000 to 2015 are 0.35,0.39,0.32,and 0.31,respectively,which are consistent with the performance of the Guangdong-Hong Kong-Macao urban agglomeration,but smaller than the GuangdongHong Kong-Macao urban agglomeration,indicating that the NTI of ChengduChongqing urban agglomeration with a smaller threshold point is more sensitive to the relative influence of LST.In general,the threshold points of all socio-economic factors in the Yangtze River Delta urban agglomeration are relatively large,indicating that under the same development level of RD,POP,NTI and GDP,the development of Chengdu-Chongqing urban agglomerations and Beijing-Tianjin-Hebei urban agglomerations with smaller thresholds has a more sensitive impact on LST.