Attribution of Urban Diurnal Thermal Environmental Change: Importance of Global–Local Effects

Global climate change is causing various negative impacts on urban ecosystems and energy systems. To effectively mitigate and adapt to these changes, it is important to understand the contributions of background climate and local effects to urban thermal environment variation. This study utilized the empirical orthogonal function (EOF) approach to deconstruct long-term MODIS land surface temperature (LST) datasets to obtain the main features of change in daytime and nighttime thermal environments. Local bivariate spatial autocorrelation analysis was used to explore the underlying causes of these changes. The main EOF modes explained 73.14% and 81.33% of daytime and nighttime thermal environment variation, respectively. The correlation coefficient between the time coefficient of the main modes and the average LST was > 0.99, reflecting the role of global effect caused by background climate change. The secondary EOF modes explained 12.51% and 4.12% of daytime and nighttime thermal environment variation, respectively, and were spatially correlated with changes in landscape thermal intensity, reflecting local effect caused by landscape change and anthropogenic heat emissions. In expansion and renewal areas, industrial zones and compact high-rise buildings had the most obvious warming effect on the daytime thermal environment, while mid-to-high-rise buildings had the most obvious warming effect on the nighttime thermal environment. The results of this study provide valuable insights into the mechanisms of background climate and local effects on the urban thermal environment, and provide a reference for formulating effective strategies for mitigating and adapting to change in urban areas, and for promoting sustainable development.