Association of neighborhood-level socioeconomic status and urban heat in China: Evidence from Hangzhou

Jie Li; Guie Li; Yangyang Jiao; Chunying Li; Qingwu Yan

doi:10.1016/j.envres.2023.118058

Association of neighborhood-level socioeconomic status and urban heat in China: Evidence from Hangzhou

Environ Res. 2024 Apr 1:246:118058. doi: 10.1016/j.envres.2023.118058. Epub 2023 Dec 29.

Authors

Jie Li¹, Guie Li², Yangyang Jiao³, Chunying Li⁴, Qingwu Yan⁵

Affiliations

¹ School of Environment and Spatial Informatics, China University of Mining and Technology, Xuzhou, China. Electronic address: jiejie@cumt.edu.cn.
² School of Public Policy & Management, China University of Mining and Technology, Xuzhou, China. Electronic address: geli@cumt.edu.cn.
³ School of Public Policy & Management, China University of Mining and Technology, Xuzhou, China. Electronic address: yangyangj@cumt.edu.cn.
⁴ School of Public Policy & Management, China University of Mining and Technology, Xuzhou, China. Electronic address: lcy@cumt.edu.cn.
⁵ School of Public Policy & Management, China University of Mining and Technology, Xuzhou, China. Electronic address: yanqingwu@cumt.edu.cn.

PMID: 38160978
DOI: 10.1016/j.envres.2023.118058

Abstract

The escalating contradiction between global urban development and thermal environments has become increasingly apparent, underscoring the imperative to address social inequality in heat exposure and advocate for environmental justice (EJ) in the pursuit of sustainable urban development. To bridge the research gap in this domain, a comprehensive study was conducted to examine the correlation mechanism linking the thermal environment with the socioeconomic status (SES) of Chinese cities, employing Hangzhou as a representative case-a pivotal city among China's "four fire stoves". The investigation involved analyzing the spatial distribution pattern of diurnal Land Surface Temperature (LST) during the summer months spanning 2016 to 2018 (July to September). For SES characterization, a holistic indicator was established. Community-level LST variables were derived from LST surfaces obtained through the Terra and Aqua satellite MODIS sensors, with the community serving as the fundamental unit of analysis. The relationship between SES and LST was explored using random forest regression (RF), eXtreme Gradient Boosting (XGBoost), and support vector regression (SVR) to assess socioeconomic inequality in urban heat. The findings reveal that (1) RF exhibits the highest fitting accuracy and adeptly elucidates the nonlinear relationship and marginal effects between LST variables and SES. (2) Community SES in the Hangzhou metropolitan area exhibits spatial clustering. (3) Residents of low and middle SES communities experience heightened heat inequality. (4) A complex nonlinear relationship exists between daytime and nighttime LST and SES, with significant social disparities in urban heat within specific temperature thresholds. When deciding on measures to advance thermal environmental justice, it is crucial to prioritize both relatively disadvantaged groups and specific temperature intervals. This study departs from conventional approaches, exploring the nonlinear relationship between SES and urban heat at a fine scale, thereby assisting urban planners in developing effective strategies.

Keywords: Environmental justice; Land surface temperature; Machine learning; Socioeconomic inequality.

MeSH terms

China
Cities
Environmental Monitoring*
Hot Temperature*
Temperature