Spatiotemporal patterns of summer urban heat island in Beijing, China using an improved land surface temperature

Urban heat island (UHI) is a major urban ecological environment issue, and it requires to be comprehensively understood from the perspective of temporal changes. This study investigated UHI in Beijing by building a monthly land surface temperature (LST) dataset at a fine spatial resolution in the summer from 2003 to 2018 using a data fusion method. First, we generated the monthly Landsat-like LST images. We then analyzed the temporal patterns of UHI in the past 16 years. Finally, we explored the spatial patterns of UHI. We find that UHI in Beijing experienced two phases, including the enhanced UHI concentrated in the urban area from 2003 to 2009, and the mitigated UHI dispersed in the suburban area from 2010 to 2018. The results of temporal trend analysis show that sub-districts with significant decreasing trends of UHI mainly locate in the urban center, while sub-districts with significant increasing trends of UHI mainly locate in the suburban areas. Moreover, the results of spatial clusters analysis demonstrate that the sub-districts with high UHIs concentrate in the urban center, while those with low UHIs disperse in the suburban area. The 16-year fine spatial resolution LSTs from this study offer a reliable dataset for studying the UHI in Beijing. The information on spatiotemporal patterns of UHI is of great help for urban planners to design UHI mitigation strategies for sustainable urban development. (C) 2020 Elsevier Ltd. All rights reserved.