Spatiotemporal Distribution of an Urban Heat Island and the Influence of Land Use over Shenzhen Based on Landsat 8 TIRS Image Data in 2014–2021

Based on the Landsat 8 TIRS remote sensing image data of 2014–2021, land surface radiation temperatures over Shenzhen were retrieved using the split window algorithm (SWA) and atmospheric correction method and verified at specific sites using ground observation temperature data, and the spatial and temporal distribution and influence factors of the urban heat island effect were discussed in this study. A significant linear correlation was found between the land surface temperature from the two algorithms (T_SWA) and the ground observation air temperature (T_M), i.e., T_SWA=1.01×T_M+2.65 and T_ARC= 0.85×T_M+5.51 (p＜0.01), but the SWA result is better in reflecting the ground observation air temperature. No significant trend (p=0.94) in the urban heat island area (HI＞0.01) was observed in Shenzhen in 2014–2021. The spatial distribution of the urban heat island was closely related to the urban development pattern, and the land use of urban planning significantly influenced the urban heat island. The land use types of ecological waters and ecological green spaces played an important role in mitigating the urban heat island effects, whereas the types of traffic roads and industrial storage promoted the formation of the urban heat island. The spatial distribution and density of the Shenzhen urban road network have a strongly significant effect on the formation of an urban heat island (p=0.003).