Multi-Index Attribution of Beijing's 2013 "Airpocalypse"

Abstract

Poor air quality causes 2 to 4 million premature deaths per year globally. Individual high-impact events, like Beijing’s January 2013 “airpocalypse,” have drawn significant attention, as they have demonstrated that short-lived air quality events can have outsized effects on public health and economic vitality. Poor air quality events are the result of emission of pollutants and the meteorological conditions favorable to their accumulation in the near-surface environment. The most important meteorological precursors of these conditions include lack of precipitation, low wind speeds, and vertical temperature inversions. Recent reports of extreme air quality, in conjunction with projected future changes in some meteorological air quality indices, raise the question: have the meteorological conditions that shape air quality changed in frequency, intensity, or duration over the observational era? Here we assess whether anthropogenic climate change has altered meteorological conditions conducive to poor air quality. To gain a more complete picture of the effect of anthropogenic change on air quality, we use three indices that quantify poor air quality: the Pollution Potential Index (Zou et al, 2017), which measures temperature inversions and surface wind speeds, the Haze Weather Index (Cai et al, 2017), which measures temperature inversions and mid-level wind speeds, and the Air Stagnation Index (Horton et al, 2014), which measures precipitation, surface wind speeds, and mid-level wind speeds. Drawing on the attribution methods of Diffenbaugh et al (2017), we assess the contribution of observed meteorological trends to the magnitude of air quality events, the return interval of events in the observational record, historical simulated climate, and pre-industrial simulated climate, and the probability of the observed trend in historical and pre-industrial simulated climates. This work provides a framework for both further understanding the role of climate change in air quality and expanding the scope of extreme event attribution.