The diminishing effects of winter heating on air quality in northern China

Junfeng Wang*, Shimeng Wang, Xiaoya Xu, Xiao Li, Pan He, Yuanbo Qiao, Ying Chen

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

Abstract

Cleaner winter heating has been promoted to abate the winter air pollution in northern China. Although improvements in air quality have been observed, the effectiveness and mechanism of cleaner heating measures on air quality have not been examined on the empirical ground. In this study, we estimate the annual effects of winter heating policy on air quality from 2014 to 2017 using a regression discontinuity design (RDD) and dynamic regression model. The results show that winter heating aggravates Air Quality Index (AQI). Specifically, the AQI raised by winter heating reduce from 85.3 in 2014 to 24.1 in 2017, indicating diminishing effects of winter heating with the implementation of clean heating measures. The heterogeneous characteristics of winter heating in terms of different pollutants and city scales are further quantified. The effects of clean heating are more evident for particulate pollutants (PM2.5 and PM10) than for SO2, NO2, CO and O3. The promotion of clean heating is more effective in larger cities. These findings provided insights into the diminishing air pollution change with continuous advancement in clean heating policy and the heterogeneity among cities and pollutants should be taken into account when formulating future policies in response to energy transition and climate change.
Original languageEnglish
Article number116536
JournalJournal of Environmental Management
Volume325
Early online date27 Oct 2022
DOIs
Publication statusPublished - 1 Jan 2023

Keywords

  • Air quality
  • Winter heating policy
  • Cleaner heating measures
  • Regression discontinuity design
  • Diminishing effects

Fingerprint

Dive into the research topics of 'The diminishing effects of winter heating on air quality in northern China'. Together they form a unique fingerprint.

Cite this