Satellite Clear‐Sky Observations Overestimate Surface Urban Heat Islands in Humid Cities

• Published in: Geophysical research letters 2024-01, Vol.51 (2), p.n/a
• Main Authors: Yang, Qiquan, Xu, Yi, Wen, Dawei, Hu, Ting, Chakraborty, TC, Liu, Yue, Yao, Rui, Chen, Shurui, Xiao, Changjiang, Yang, Jie
• Format: Article
• Language: English
• Subjects: all‐weather; Cities; Cloud cover; data missing; ENVIRONMENTAL SCIENCES; global; Heat; Islands; Jupiter; Land surface temperature; Remote sensing; Satellite observation; Satellites; Sky; Standardization; Surface temperature; Temporal variations; Thermal environments; UHI effect; uncertainty; Urban areas; Urban heat islands
• ISSN: 0094-8276; 1944-8007
• DOI: 10.1029/2023GL106995

Satellite‐based thermal infrared (TIR) land surface temperature (LST) is hindered by cloud cover and is applicable solely under clear‐sky conditions for estimating surface urban heat island intensity (SUHII). Clear‐sky SUHII may not accurately represent all‐sky conditions, potentially introducing quantitative biases in assessing urban heat islands. However, the differences between clear‐sky and all‐sky SUHIIs and their spatiotemporal variations are still poorly understood. Our analysis of over 600 global cities demonstrates that clear‐sky SUHII is mostly higher than all‐sky SUHII, particularly in summer, daytime, and precipitation‐rich regions. Besides, clear‐sky SUHII typically exhibits stronger seasonal and diurnal contrasts than all‐sky SUHII, especially for cities located in humid regions. These discrepancies can be attributed mainly to the increased missing LST data caused by cloud enhancement in urban areas. Our findings highlight the tendency for clear‐sky observations to overestimate SUHII, providing valuable insights for standardizing the quantification of surface urban heat islands. Plain Language Summary Surface urban heat island intensity (SUHII) and its spatial and temporal variations are important for describing the urban thermal environment. SUHII is usually estimated from remotely sensed land surface temperature (LST), which is only available under clear‐sky conditions. The SUHII derived from clear‐sky observations may differ from the SUHII under all‐sky conditions. However, there is currently a lack of large‐scale quantitative assessments addressing the differences between clear‐sky and all‐sky SUHIIs. This study fills this research gap and indicates a substantial overestimation of SUHII in humid regions when using clear‐sky LST. This overestimation can be explained by the increased occurrence of missing LST data caused by the enhanced presence of clouds in urban areas. Our findings show the importance of utilizing all‐sky LST data in the examination of urban surface thermal environments, especially for cities situated in humid regions. Key Points Clear‐sky surface urban heat island intensity (SUHII) shows higher values and stronger spatiotemporal variations than all‐sky SUHII, notably in summer, daytime, and humid areas The annual daytime SUHII for tropical cities is, on average, overestimated by 30% when relying on clear‐sky land surface temperature (LST) observations Differences in clear‐sky and all‐sky SUHIIs can be explained by more m