Exploring the impact of urban features on the spatial variation of land surface temperature within the diurnal cycle

Excessive heat in urban areas affects thermal comfort and livability. In order to improve local thermal envi-ronment, correlations patterns between urban features and land surface temperature (LST) have been widely studied. However, the majority of these studies were conducted only during the daytime. Due to the limitations of polar orbiting satellites such as Landsat series and Terra, it was still unclear how urban features affect the spatial variation of LST in a 24-hour cycle. In summer, with Beijing, China as an example, five machine learning models were used to perform regression analysis between urban features and ECOSTRESS LST within the diurnal cycle. The main findings of this study are: (1) The impact of urban features on LST was mainly in relation to the intensity of sunlight. (2) Building and vegetation coverage affected LST most in the morning and afternoon respectively. (3) Point of interest (such as shops, schools, and car parks etc.) density and surface albedo were the dominant factors influencing LST from dusk to dawn. (4) Higher buildings lowered LST during the day, but raised it at night. We believed that these findings can provide quantitative insights into UHI adaptation measures for sustainable urban planning and design.

Automated summary

Excessive heat in urban areas affects thermal comfort and livability. Understanding the correlations between urban features and land surface temperature (LST) is crucial for improving local thermal environment. In this study, machine learning models were used to analyze the relationship between urban features and ECOSTRESS LST in a 24-hour cycle, using Beijing, China as an example. The findings showed that the impact of urban features on LST was mainly dependent on sunlight intensity, with buildings and vegetation coverage having the most significant influence in the morning and afternoon respectively.