The importance of non-fossil sources in carbonaceous aerosols in a megacity of central China during the 2013 winter haze episode: A source apportionment constrained by radiocarbon and organic tracers

Highlights

• Radiocarbon and organic tracers were determined in a megacity of central China during an extreme haze event in 2013 winter.

• The contributions of non-fossil sources in OC and EC were 62% ± 5% and 26% ± 8%, respectively.

• The contributions of non-fossil sources in WIOC and WSOC were 61% ± 4% and 63% ± 6%, respectively.

• Most non-fossil SOC particles probably were derived from the atmospheric processes regarding biomass-burning.

Abstract

To determine the causes of a severe haze episode in January 2013 in China, a source apportionment of different carbonaceous aerosols (CAs) was conducted in a megacity in central China (Wuhan, Hubei Province) by using the measurements of radiocarbon and molecular organic tracers. Non-fossil sources (e.g., domestic biofuel combustion and biogenic emissions) were found to be responsible for 62% ± 5% and 26% ± 8% of organic carbon (OC) and elemental carbon (EC) components by mass, respectively. Non-fossil sources contributed 57% ± 4% to total CAs in this large-scale haze event, whereas fossil-fuel sources were less dominant (43% ± 4%). The CAs were composed of secondary organic carbon (SOC; 46% ± 10%), primary fossil-fuel carbon (29% ± 4%) and primary biomass-burning carbon (25% ± 10%). Although SOC was formed mainly from non-fossil sources (70% ± 4%), the role of fossil precursors was substantial (30% ± 4%), much higher than at the global scale. Combined measurement of organic tracers and radiocarbon showed that most non-fossil SOC was probably derived from biomass burning during this long-lasting haze episode in central China.

Introduction

A severe and long-lasting haze episode, with an extremely elevated PM_2.5 (aerodynamic diameter ≤ 2.5 μm) concentration (the hourly concentration up to ∼1000 μg/m³) (Uno et al., 2014), occurred in January 2013 in central and eastern China. Because a high PM_2.5 loading can cause a reduction in visibility, climate changes, and human respiratory-cardiovascular diseases (Brunekreef and Holgate, 2002, Menon et al., 2002, Deng et al., 2008, Wang et al., 2014b), many concerns were raised by the public, government, and scientists. Numerous investigations have been performed to determine the characteristics of this air pollution crisis. He et al. (2014) identified a new haze formation mechanism regarding the conversion of SO₂ to sulfate, and reported that the impact of motor vehicle on air quality was underestimated in the Beijing-Tianjin-Hebei Region. Using an aerodyne aerosol chemical speciation monitor, Sun et al. (2014) found that stagnant meteorological conditions, coal combustion, secondary production, and regional transport were the main factors leading to the formation of this haze in Beijing. Wang et al. (2014a) called on the government to establish a regional joint framework for mitigation of the severe air pollution based on their model evaluations. Currently, most of these studies have been conducted in northern China, specifically in Beijing, and have focused on the analysis of chemical concentrations. Measurements of more-specific-sources tracers (i.e., isotopes and organic tracers) during this haze period are still scarce.

Carbonaceous aerosols (CAs) are important major components of PM_2.5. However, CAs are poorly understood because of their vast number of emission sources, various physicochemical properties, and heterogeneous distribution in time and space. Total CAs values are generally expressed in terms of total carbon (TC), which contains organic carbon (OC) and elemental carbon (EC). EC is a primary carbon species that is derived solely from the incomplete combustion of carbon-containing materials. Ambient OC is a mixture of primary organic carbon (POC), which is emitted from various combustion processes, and secondary organic carbon (SOC), which is formed through the oxidation of volatile organic compounds (VOCs) (Pöschl, 2005, Calvo et al., 2013). In addition, a large fraction of SOC can be formed from the chemical reactions of POC (Robinson et al., 2007). OC can be further separated into water-soluble organic carbon (WSOC) and water-insoluble organic carbon (WIOC). These carbon particles in the atmosphere have two sources: fossil fuel (FF, e.g., from traffic exhaust, coal combustion, industry) and non-fossil (NF, e.g., from open/forest fire, biofuel burning, biogenic emission) emissions. Their unambiguous source apportionment has been conducted in recent years by the measurements of radiocarbon (¹⁴C) (Gustafsson et al., 2009, Szidat et al., 2009, Chen et al., 2013, Liu et al., 2013, Huang et al., 2014, Liu et al., 2014, Zotter et al., 2014a, Andersson et al., 2015, Zhang et al., 2015). This radioisotope (half-life = 5730 years) enables a distinction between FF and NF sources because ¹⁴C is absent in FF, but present at the current ambient level in NF materials. ¹⁴C analyses of aerosols have seldom been reported in China due to the complexity of experimental procedures and the need for a specific analysis facility. Chen et al. (2013) first systematically investigated the ¹⁴C signals of EC (or black carbon) in Beijing, Shanghai, and Xiamen, and found that 83–86% of EC was associated with FF combustion during the 2009–2010 winter, with the remainder derived from biomass burning (BB). Zhang et al. (2015) analyzed the ¹⁴C levels of OC and EC in four Chinese cities–Beijing, Xi'an, Guangzhou, and Shanghai–and found that the contributions of FF sources to OC and EC were 35–49% and 57–80%, respectively, in January 2013. Andersson et al. (2015) found that during this haze period FF sources on average contributed 74%, 68% and 68% to EC in Beijing, Shanghai and Guangzhou, respectively. Liu et al. (2014) showed that the contribution of FF in OC and EC was 37% and 71% in Guangzhou during November 2012 to January 2013, respectively. A newly updated China emission inventory showed that the coal used in power plants is 8300 Gg, 28,000 Gg, 85,000 Gg, 80,000 Gg, 35,000 Gg, in Beijing (north China), Shanghai (east China), Guangdong (south China, the capital is Guangzhou), Shanxi (west China, the capital is Xi'an) and Hubei (central China, the capital is Wuhan), and the corresponding value for residential solid biomass is 880 Gg, 0 Gg, 22,000 Gg, 14,000 Gg and 26,000 Gg, respectively (Wang et al., 2012). These results indicate that biomass used for residential burning in Hubei seems higher than Guangdong, especially than Beijing and Shanghai. Given this difference of energy consumption pattern among different regions in China, the key sources of this haze episode probably is region-dependent. Previous ¹⁴C-related studies also have displayed this difference. For example, the contribution of FF sources to OC in Beijing was 58% (Zhang et al., 2015), whereas it was <40% in Guangzhou on average (Liu et al., 2014, Zhang et al., 2015). Thus, this large-scale haze crisis was very likely caused by the convergence of materials from numerous point sources in regions with different sources. More ¹⁴C-related studies are urgently needed to accurately and quantitatively elucidate the emission sources of CAs during such a regional haze crisis. On other hand, the sources of WSOC and WIOC differ markedly from those of EC (Liu et al., 2014). Thus, determination of the ¹⁴C isotopic signals of various carbon species is necessary to obtain a better understanding of haze pollution characteristics and sources.

The haze phenomenon in China is very complex. First, Chinese cities are at the developmental stage of industrialization and urbanization, with a large demand for FF energy. Second, biofuel is a very common energy source in rural and suburban areas, in which ∼50% of the Chinese population lives. Third, little is known about the evolution of SOC and its precursor VOCs, especially the relative contributions of FF and NF sources to SOC. Consequently, controversial results regarding the PM_2.5 sources in China have been published. One group reported that the annual contributions of coal and biomass combustion to PM_2.5 in Beijing were 7% and 6% (Zheng et al., 2005), respectively, whereas higher corresponding contributions (14–19% and 11–13%, respectively) have been reported in other studies (Song et al., 2006, Zhang et al., 2013a). This discrepancy are mainly due to the less-source-tracers such as K⁺ and elements that needed to input into the models and their variable parameters. To determine the origins of haze particles, we measured ¹⁴C isotopic signals and unique organic tracers in PM_2.5 samples with various levels in Wuhan (Fig. S1), the largest (∼550 km²) and most densely populated metropolis (∼10 million) in central China. Wuhan is the capital of Hubei province and located in the core area of the January 2013 large-scale haze pollution (Fig. S1). According to the annual report of the Editorial Department of Wuhan Statistical Yearbook-2014, the gross domestic product was composed of agriculture (3.7%), industry (48.6%), and other sectors (47.7%). Although pollution is frequently severe in Wuhan, studies of air pollution in this megacity are only just beginning to occur. It should be noted that some ¹⁴C-related studies have recently reported on aerosol sources of the haze episode in 2013 in other cities as we mentioned above, the important aspect of haze sources in the central China with a relatively high usage of biomass has not been addressed yet. Our study provides new insights into the sources of air pollution in Chinese city, specifically into the contributions of FF and NF sources to the different carbon species during this long and persistent haze incident in central China. We also conducted a source apportionment, including POC and SOC, through the combinational measurement of molecular markers and radiocarbon. To our best knowledge, this article is first to report on ¹⁴C signals of CAs in this megacity.