How Do Driving Factors Affect the Diurnal Variation of Land Surface Temperature Across Different Urban Functional Blocks? A Case Study of Xi'an, China
Kaixu Zhao,Zekui Ning,Chen Xu,Xin Zhao,Xiaojun Huang
DOI: https://doi.org/10.1016/j.scs.2024.105738
IF: 11.7
2024-01-01
Sustainable Cities and Society
Abstract:A comprehensive and in-depth understanding of the formation mechanisms of the urban thermal environment is the basis for thermal environment regulation, however, there is insufficient knowledge regarding how driving factors influence daytime and nighttime land surface temperature (LST) within urban functional blocks (UFBs). We selected Xi'an, China as a case study, integrating remote sensing data including ECOSTRESS, Landsat-8, and Gaofen-1, along with geographic data including road network, areas of interest, points of interest, building footprint, and mobile phone signaling. It divided 10 types of UFBs, inverted daytime and nighttime LST, calculated 5 types of driving factors, and finally analyzed the contribution and marginal effects of driving factors on day-night LST using boosted regression tree. The results showed that LST and its driving factors differed significantly in different times and UFBs. Industrial blocks and urban villages had higher LST in the daytime, while residential blocks, commercial blocks, and public service blocks had higher LST in nighttime. Industrial blocks were the dominant blocks that drove the overall LST up during the day, while residential blocks were the dominant blocks at night. Location distance (-) and population density (+) affected LST in all UFBs during day and night, NDVI (-), building density (+), and floor area ratio (-) were key factors for most UFBs during daytime, and NDVI (+), surface albedo (-), and point density of interest (+) were key factors during nighttime. Most of the driving factors had significant influence thresholds, but there were small differences across UFBs. This paper aims to elucidate the mechanisms by which driving factors influence urban LST during day and night across different UFBs, thereby providing new support for more targeted thermal environment regulation and diurnal trade-offs at the block scale.