Simulating the sources of PM2.5 during heavy haze pollution episodes in the autumn and winter of 2016 in Xianyang City, China
： 2018 - 07 - 05
： 2018 - 10 - 11
： 2018 - 10 - 20
145 0 0

Abstract & Keywords
Abstract: Background, aim, and scope Recently, frequent and persistent particulate pollution has been the most urgent air pollution problem in most regions and cities in China, causing serious impact on climate change and human health. Fine particulate matters (PM2.5) contribute to climate change directly by absorbing and scattering the solar radiation and indirectly by serving as cloud condensation nuclei (CCN) and ice nuclei (IN) to modify cloud properties. High concentrations of PM2.5 can reduce atmospheric visibility and exert deleterious impacts on air quality, ecosystem, and human health. According to previous studies, the occurrence of particulate pollution is considered to be closely related to the characteristics of PM2.5 and its chemical components. The further study of the formation process, reaction mechanism, and major sources of PM2.5 is currently one of the major bottlenecks in improving the air quality in China. In the past few decades, with the accelerating process of industrialization and urbanization, the emissions of pollutants in China have increased significantly, and the air pollution has become increasingly severe. The Guanzhong Basin (GZB) is located in northwestern China and surrounded by the Qinling Mountains in the south and the Loess Plateau in the north, with a warm-humid climate. The rapid increasing industries and city expansions, as well as the unique topography, have caused frequent occurrence of haze in the basin, which has drawn extensive attention to clarify its formation, sources, and influence. The main purpose of the present study is to quantitatively evaluate the contribution of the local emissions in Xianyang City and the major emission sources in GZB to the PM2.5 mass concentrations in Xianyang City based on the WRF-CHEM model during two heavy haze pollution episodes in GZB from 12 to 19 November 2016 and 16 to 21 December 2016, aiming at providing a reliable basis for local authorities to establish reasonable and effective comprehensive prevention and control strategies and pollution reduction measures for particulate pollution. Materials and methods A specific version of the WRF-CHEM model is used to investigate the air pollution formation in GZB, including a flexible gas-phase chemical module and the CMAQ aerosol module developed by US EPA. The wet deposition of aerosols follows the method used in the CMAQ module and the dry deposition of chemical species is parameterized following Wesely. The photolysis rates are calculated using the FTUV (fast radiation transfer model) module, considering the aerosol and cloud effects on photolysis. The inorganic aerosols are calculated using ISORROPIA Version 1.7. The secondary organic aerosol (SOA) is predicted using the volatility basis-set (VBS) modeling method, with contributions from glyoxal and methylglyoxal. The NCEP 1°×1° reanalysis data are used for the meteorological initial and boundary conditions, and the chemical initial and boundary conditions are interpolated from the 6 h output of MOZART. The SAPRC-99 chemical mechanism is used in the study. The anthropogenic emissions are from the MEIC emission inventory, including agriculture, industry, power generation, residential, and transportation sources. The biogenic emissions are calculated online using the MEGAN model. Results Compared to observations over the ambient monitoring sites in GZB, the WRF-CHEM model reasonably well reproduces the temporal variations and spatial distributions of the PM2.5 mass concentrations during the simulation period, indicating reasonable replications of the haze events. Discussion The sensitivity simulations show that the emissions in Xianyang City contribute about 30% of the local PM2.5 mass concentrations during the simulation period in the autumn and winter of 2016. Except for the background contribution of about 10%, the contribution of the regional transport to the PM2.5 mass concentration in Xianyang City is up to 50%－60%. Among the various emission sources in GZB, the residential source is the major contributor to the PM2.5 mass concentrations in Xianyang City, with contribution of 37.4% in autumn and 60.6% in winter. During the two simulation periods in the autumn and winter of 2016, the contribution of industrial and transportation emissions in GZB to the PM2.5 in Xianyang City is 22.1% and 11.2% and 15.6% and 9.8%, respectively； and the contribution of power emissions is only 2.0%. Conclusions During the heavy haze pollution episodes in the autumn and winter of 2016, the local emissions in Xianyang City contribute about 30% to the mass concentrations of PM2.5, and the contribution of regional transport to the PM2.5 mass concentration in Xianyang City is up to 50%－60%. Among the various emission sources in GZB, the residential source significantly contributes to the PM2.5 concentration in Xianyang City, with contribution of 37.4% in autumn and 60.6% in winter. The contribution of industrial and transportation emissions in GZB to the PM2.5 in Xianyang City is 22.1% and 11.2% in autumn and 15.6% and 9.8% in winter； and the contribution of power emissions is only 2.0%. Recommendations and perspectives Further studies need to be conducted to improve the WRF-CHEM model performance considering both the uncertainties in the meteorological simulations and the emission inventory. More sensitivity simulations of long time scales are also needed in future studies to provide more reliable guidance for the air quality improvements.
Keywords: WRF-CHEM; PM2.5; Guanzhong Basin; air pollution

1   模式和方法
1.1   WRF-CHEM模式
WRF-CHEM（weather research and forecast model with chemistry）模式是由美国国家大气科学中心（NCAR）和美国国家海洋大气管理局（NOAA）等多家机构联合开发的中尺度区域大气动力－化学耦合传输模式。模式的气象模块和化学传输模块在时间和空间分辨率上完全耦合，使用相同的垂直和水平坐标，实现了气象过程和化学过程的实时在线传输与反馈。WRF模式是完全可压的非静力模式，对发生在大气中的湍流交换、大气辐射、积云降水、云微物理及陆面过程等多种物理过程采用不同的参数化方案，为化学模式在线供大气流场，模拟大气污染物的传输（包括平流、扩散和对流过程）、沉降（干沉降和湿沉降）、化学转化，气溶胶的形成、辐射和光解等物理和化学过程，可以反映真实大气环境中的各种物理和化学反应过程。

1.2   模式设置
WRF-CHEM模式模拟选取2016年11月12－19日和2016年12月16－21日期间咸阳市两次大气重污染事件，模式模拟区域如图1，中心点为109°E，34°15′N，模式的水平分辨率为3 km，网格数为200×200，垂直方向包括35层（近地面层高30 m，2.5 km高空以上层高500 m，14 km以上高空层高1 km）。关中盆地南北分别被陕北高原和秦岭山脉包围，西部以宝鸡峡为界，东至潼关港口，盆地内的地势西窄东宽，当盆地内无盛行风向时，大气污染物很难扩散。当关中盆地内盛行东风或东北风时，东部地区的大气污染物就会被向西输送，从而加剧西安、咸阳一带的大气污染（Bei et al，2016a，2016b）。图1b中的黑色圆点表示关中盆地内国控站的位置，其中西安13个站，咸阳4个站，宝鸡8个站，铜川4个站，渭南4个站，国控站基本都位于城市中，分布比较密集，黑色加粗的边界线代表咸阳市的行政边界。

Fig.1 WRF-CHEM model simulation domain with topography (the filled black circles represent the ambient monitoring sites, the area inside the bold black border is Xianyang City)
WRF-CHEM模式模拟所选择的物理参数化方案主要包括：WSM 6-class微物理方案（Hong and Lim，2006），MYJ TKE湍流动能边界层方案（Janjić，2002），Noah陆面过程方案（Chen and Dudhia，2001）和Goddard长波和短波辐射方案（Chou and Suarez，1994，1999）。模式中采用SAPRC-99化学机制以及CMAQ模式中的气溶胶模块；湿沉降遵循CMAQ模式中使用的方法，干沉降遵循Wesely（1989）对化学物质的表面干沉积的参数化方案；光解速率用FTUV模式（Li et al，2011a，2005）来计算，其中考虑了气溶胶和云对光解速率的影响。无机气溶胶的计算根据ISORROPIA Vertion 1.7（Nenes et al，1998）的方法，有机气溶胶的计算则使用非传统的VBS方法，考虑了乙二醛和甲基乙二醛对二次有机气溶胶的贡献（Li et al，2011b，2010）。模式中气象场的初始条件和边界条件来自NCEP FNL 1°×1°再分析资料。化学场的初始条件和边界条件来自MOZART每6小时的输出数据（Horowitz et al，2003）。人为源排放清单采用清华大学的MEIC排放清单（Zhang et al，2009；Li et al，2017），主要包括工业源、电厂源、交通源、居民源和农业源的排放。生物源排放利用MEGAN模式在线计算（Guenther et al，2006）。
1.3   统计检验方法

$$MB=\frac{1}{N}\sum _{i=1}^{N}\left({P}_{i}-{O}_{i}\right)$$ (1)
$$RMSE={\left[\frac{1}{N}\sum _{i=1}^{N}{\left({P}_{i}-{O}_{i}\right)}^{2}\right]}^{\frac{1}{2}}$$ (2)
$$IOA=1-\frac{\sum _{i=1}^{N}{\left({P}_{i}-{O}_{i}\right)}^{2}}{\sum _{i=1}^{N}{\left(\left|{P}_{i}-\stackrel{-}{O}\right|+\left|{O}_{i}-\stackrel{-}{O}\right|\right)}^{2}}$$ (3)

1.4   敏感性试验设计

 试验 人为源排放 Cases Anthropogenic Emissions 基准试验 包括MEIC排放清单中所有人为源的排放 Base case Include all emissions in MEIC emission inventory 敏感性试验1 去除MEIC排放清单中所有人为源的排放 Sensitivity case 1 Exclude all emissions in MEIC emission inventory 敏感性试验2 去除MEIC排放清单中咸阳市所有人为源的排放 Sensitivity case 2 Exclude all emissions in the Xianyang City in MEIC emission inventory 敏感性试验3 去除MEIC排放清单中关中地区工业源的排放 Sensitivity case 3 Exclude industry sources emissions in Guanzhong Basin in MEIC emission inventory 敏感性试验4 去除MEIC排放清单中关中地区电厂源的排放 Sensitivity case 4 Exclude power sources emissions in Guanzhong Basin in MEIC emission inventory 敏感性试验5 去除MEIC排放清单中关中地区居民源的排放 Sensitivity case 5 Exclude residential sources emissions in Guanzhong Basin in MEIC emission inventory 敏感性试验6 去除MEIC排放清单中关中地区交通源的排放 Sensitivity case 6 Exclude transportation sources emissions in Guanzhong Basin in MEIC emission inventory
2   结果和讨论
2.1   模式验证
2.1.1   秋季模拟情况

Fig.2 Comparisons of measured (black dots) and simulated (solid red lines) diurnal profiles of near-surface hourly mass concentrations of PM2.5 in Guanzhong Basin (a), Xi’an (b), and Xianyang (c) during the simulation period from 12 to 19 November 2016

Fig.3 Pattern comparisons of observed (colored circles) vs. simulated (color counters) daily averaged and 8 days averaged mass concentrations of near-surface PM2.5 during the simulation period from 12 to 19 November 2016 in Guanzhong Basin (the black arrows indicate simulated surface winds)
2.1.2   冬季模拟情况

Fig.4 Comparisons of measured (black dots) and simulated (solid red lines) diurnal profiles of near-surface hourly mass concentrations of PM2.5 in Guanzhong Basin (a), Xi’an (b), and Xianyang (c) during the simulation period from 16 to 21 December 2016

Fig.5 Pattern comparisons of observed (colored circles) vs. simulated (color counters) daily averaged and 6 days averaged mass concentrations of near-surface PM2.5 during the simulation period from 16 to 21 December 2016 in Guanzhong Basin (the black arrows indicate simulated surface winds)

2.2   敏感性试验
2.2.1   秋季PM2.5来源

Fig.6 Spatial distribution of the daily average contribution from local emissions to the PM2.5 mass concentrations in Xianyang City during the simulation period from 12 to 19 November 2016

Fig.7 Spatial distribution of the average contribution from local emissions to the PM2.5 mass concentrations (a) in Xianyang City and the background PM2.5 concentrations (b) during the simulation period from 12 to 19 November 2016

 区域 Area 基准试验 Base case /(µg m-3) 敏感性试验（去除咸阳市本地排放） Sensitivity simulation (exclude the local emissions in Xianyang City) /(µg m-3) 变化量 Mass change /(µg m-3) 百分比 Percentage change /% 关中地区 Guanzhong Basin 160.0 122.0 38.0 23.8 西安市 Xi’an 182.4 174.9 7.5 4.1 咸阳市 Xianyang 183.2 122.0 61.2 33.4

Fig.8 The spatial distribution of the average contribution to the PM2.5 mass concentrations in Xianyang City from industry (a), power (b), residential (c), and transportation sources (d) in Guanzhong Basin during the simulation period from 12 to 19 November 2016

 敏感性试验 Sensitivity simulations PM2.5 /(µg m-3) 变化量 Mass change /(µg m-3) 百分比 Percentage change /% 去除工业源 Industry-off 142.8 40.8 22.1 去除电厂源 Power-off 179.7 3.6 2.0 去除居民源 Residential-off 114.7 68.6 37.4 去除交通源 Transportation-off 162.7 20.6 11.2
2.2.2   冬季PM2.5来源

Fig.9 Spatial distribution of the daily average contribution from local emissions to the PM2.5 mass concentrations in Xianyang City during the simulation period from 16 to 21 December 2016

Fig.10 Spatial distribution of (a) the average contribution from local emissions to the PM2.5 mass concentrations in Xianyang City and (b) the background PM2.5 concentrations during the simulation period from 16 to 21 December 2016

 区域 Area 基准试验 Base case /(µg m-3) 敏感性试验（去除咸阳市本地排放） Sensitivity simulation (exclude the local emissions in Xianyang City) /(µg m-3) 变化量 Mass change /(µg m-3) 百分比 Percentage change /% 关中地区 Guanzhong Basin 232.7 216.7 16.0 6.9 西安市 Xi’an 278.8 266.9 11.9 4.3 咸阳市 Xianyang 263.4 181.7 81.7 31.0

Fig.11 Spatial distribution of the average contribution to the PM2.5 mass concentrations in Xianyang City from industry (a), power (b), residential (c), and transportation sources (d) in Guanzhong Basin during the simulation period from 16 to 21 December 2016

 敏感性试验 Sensitivity simulations PM2.5 /(µg m-3) 变化量 Mass change /(µg m-3) 百分比 Percentage change /% 去除工业源 Industry-off 222.3 41.1 15.6 去除电厂源 Power-off 258.3 5.1 2.0 去除居民源 Residential-off 103.8 159.6 60.6 去除交通源 Transportation-off 237.5 25.9 9.8
3   结论

（1）WRF-CHEM模式基本可以合理模拟出关中地区秋冬季大气重污染期间PM2.5质量浓度的时间变化及空间分布，但是模式模拟结果与观测数据之间仍然存在一定的偏差，除气象因素（如边界层和水平风场的模拟）的影响外，人为源排放的不确定性也是造成模式模拟偏差的一个重要原因。
（2）敏感性试验结果表明：秋冬季大气重污染期间，咸阳市本地排放对当地PM2.5质量浓度的贡献约为30%，除去背景贡献，外源输送对当地PM2.5质量浓度的贡献可以达到50%－60%。这主要是由关中盆地的特殊地形造成的，当盆地内发生重污染时，盛行风主要是东风，造成西安及东部周边地区形成的污染物向西输送，加剧咸阳市的大气污染状况。
（3）在关中地区的主要大气污染源中，居民源是秋冬季咸阳市最主要的PM2.5来源。秋季大气污染期间，关中地区居民源对咸阳市PM2.5污染水平的贡献约为37.4%；工业源和交通源对的贡献分别为22.1%和11.2%；电厂源的贡献仅占2.0%。冬季大气重污染期间，关中地区居民源对咸阳市PM2.5质量浓度的贡献高达60.6%；工业源和交通源对的贡献分别为15.6%和9.8%；电厂源的贡献约为2.0%。因此，在秋冬季大气重污染期间，应该主要通过控制居民源和工业源排放来减轻咸阳市PM2.5污染。

Bei N F, Li G H, Huang R J, et al. 2016b. Typical synoptic situations and their impacts on the wintertime air pollution in the Guanzhong Basin, China [J]. Atmospheric Chemistry and Physics, 16(11): 7373-7387.
Bei N F, Xiao B, Meng N, et al. 2016a. Critical role of meteorological conditions in a persistent haze episode in the Guanzhong Basin, China [J]. Science of the Total Environment, 550: 273-284.
Cao J J, Wang Q Y, Chow J C, et al. 2012b. Impacts of aerosol compositions on visibility impairment in Xi’an, China [J]. Atmospheric Environment, 59: 559-566.
Cao J J, Xu H M, Xu Q, et al. 2012a. Fine particulate matter constituents and cardiopulmonary mortality in a heavily polluted Chinese city [J]. Environmental Health Perspectives, 120(3): 373-378.
Chan C K, Yao X H. 2008. Air pollution in mega cities in China [J]. Atmospheric Environment, 42(1): 1–42.
Chen F, Dudhia J. 2001. Coupling an advanced land surface-hydrology model with the Penn State-NCAR MM5 modeling system. part Ⅱ: preliminary model validation [J]. Monthly Weather Review, 129(4): 587–604.
Chou M D, Suarez M J. 1994. An efficient thermal infrared radiation parameterization for use in general circulation models [R]. Technical Report Series on Global Modeling and Data Assimilation. NASA Technology Memorandum 104606, Vol. 3. Greenbelt, Maryland: Goddard Space Flight Center.
Chou M D, Suarez M J. 1999. A solar radiation parameterization for atmospheric studies [R]. Technical Report Series on Global Modeling and Data Assimilation. NASA Technology Memorandum 104606, Vol. 15. Greenbelt, Maryland: Goddard Space Flight Center.
Donahue N M, Robinson A L, Stanier C O, et al. 2006. Coupled partitioning, dilution, and chemical aging of semivolatile organics [J]. Environmental Science and Technology, 40(8): 2635-2643.
Greenwald R, Bergin M H, Xu J, et al. 2006. The influence of aerosols on crop production: A study using the CERES crop model [J]. Agricultural Systems, 89(2/3): 390-413.
Guenther A, Karl T, Harley P, et al. 2006. Estimates of global terrestrial isoprene emissions using MEGAN (Model of Emissions of Gases and Aerosols from Nature) [J]. Atmospheric Chemistry and Physics, 6(11): 3181-3210.
Happo M S, Hirvonen M R, Hälinen A I, et al. 2008. Chemical compositions responsible for inflammation and tissue damage in the mouse lung by coarse and fine particulate samples from contrasting air pollution in Europe [J]. Inhalation Toxicology, 20(14): 1215-1231.
He K B, Yang F M, Ma Y L, et al. 2001. The characteristics of PM2.5 in Beijing, China [J]. Atmospheric Environment, 35(29): 4959–4970.
Hong S Y, Lim J O J. 2006. The WRF Single-Moment 6-Class Microphysics Scheme (WSM6) [J]. Asia-Pacific Journal of Atmospheric Sciences, 42(2): 129-151.
Horowitz L W, Walters S, Mauzerall D L, et al. 2003. A global simulation of tropospheric ozone and related tracers: Description and evaluation of MOZART, version 2 [J]. Journal of Geophysical Research, 108(D24). DOI: 10.1029/2002JD002853.
Huang R J, Zhang Y L, Bozzetti C, et al. 2014. High secondary aerosol contribution to particulate pollution during haze events in China [J]. Nature, 514(7521): 218-222.
Janjić Z I. 2002. Nonsingular implementation of the Mellor-Yamada Level 2.5 scheme in the NCEP Meso model [Z]. National Centers for Environmental Prediction Office Note #437.
Lelieveld J, Evans J S, Fnais M, et al. 2015. The contribution of outdoor air pollution sources to premature mortality on a global scale [J]. Nature, 525(7569): 367–371.
Li G H, Bei N F, Tie X X, et al. 2011a. Aerosol effects on the photochemistry in Mexico City during MCMA-2006/MILAGRO campaign [J]. Atmospheric Chemistry and Physics, 11: 5169-5182.
Li G H, Lei W, Bei N F, et al. 2012. Contribution of garbage burning to chloride and PM2.5 in Mexico City [J]. Atmospheric Chemistry and Physics, 12(18): 8751-8761.
Li G H, Lei W, Zavala M, et al. 2010. Impacts of HONO sources on the photochemistry in Mexico City during the MCMA-2006/MILAGO Campaign [J]. Atmospheric Chemistry and Physics, 10(14): 6551-6567.
Li G H, Zavala M, Lei W, et al. 2011b. Simulations of organic aerosol concentrations in Mexico City using the WRF-CHEM model during the MCMA-2006/MILAGRO campaign [J]. Atmospheric Chemistry and Physics, 11(8): 3789-3809.
Li G H, Zhang R Y, Fan J W, et al. 2005. Impacts of black carbon aerosol on photolysis and ozone [J]. Journal of Geophysical Research, 110(D23). DOI: 10.1029/2005JD005898.
Li M, Zhang Q, Kurokawa J I, et al. 2017. MIX: a mosaic Asian anthropogenic emission inventory under the international collaboration framework of the MICS-Asia and HTAP [J]. Atmospheric Chemistry and Physics, 17(2): 935-963.
Liu J G, Diamond J. 2005. China’s environment in a globalizing world [J]. Nature, 435(7046): 1179-1186.
Nenes A, Pandis S N, Pilinis C. 1998. ISORROPIA: a new thermodynamic equilibrium model for multiphase multicomponent inorganic aerosols [J]. Aquatic Geochemistry, 4(1): 123–152.
Parrish D D, Stockwell W R. 2015. Urbanization and air pollution: then and now [J]. Eos: Earth & Space Science News, 96: 10–15.
Parrish D D, Zhu T. 2009. Clean air for megacities[J]. Science, 326(5953): 674–675.
Robinson A L, Donahue N M, Shrivastava M K, et al. 2007. Rethinking organic aerosols: semivolatile emissions and photochemical aging [J]. Science, 315(5816): 1259–1262.
Seinfeld J H, Pandis S N. 2006. Atmospheric chemistry and physics: from air pollution to climate change [M]. 2nd edition. New York: John Wiley and Sons.
Tie X X, Zhang Q, He H, et al. 2015. A budget analysis of the formation of haze in Beijing [J]. Atmospheric Environment, 100: 25-36.
Wesely M L. 1989. Parameterization of surface resistances to gaseous dry deposition in regional-scale numerical models[J]. Atmospheric Environment, 23(6): 1293–1304.
Yang F, Tan J, Zhao Q, et al. 2011. Characteristics of PM2.5 speciation in representative megacities and across China [J]. Atmospheric Chemistry and Physics, 11: 5207-5219.
Zhang Q, He K B, Huo H. 2012. Policy: cleaning China’s air [J]. Nature, 484(7393): 161-162.
Zhang Q, Streets D G, Carmichael G R, et al. 2009. Asian emissions in 2006 for the NASA INTEX-B mission [J]. Atmospheric Chemistry and Physics, 9: 5131-5153.
Zhang R Y, Guo S, Levy Z M, et al. 2015. Reply to Li et al.: Insufficient evidence for the contribution of regional transport to severe haze formation in Beijing [J]. Proceedings of the National Academy of Sciences of the United States of America, 112(21): E2741. DOI: 10.1073/pnas.1503855112.
Zhang R Y, Jing J, Tao J, et al. 2013. Chemical characterization and source apportionment of PM2.5 in Beijing: seasonal perspective [J]. Atmospheric Chemistry and Physics, 13(14): 7053-7074.
Zhang R Y, Li G H, Fan J W, et al. 2007. Intensification of Pacific storm track linked to Asian pollution [J]. Proceedings of the National Academy of Sciences of the United States of America, 104(13): 5295-5299.
Zhang R Y, Wang G H, Guo S, et al. 2015. Formation of urban fine particulate matter[J]. Chemical Reviews, 115(10): 3803–3855.
Zhang R, Wang L, Khalizov A F, et al. 2009. Formation of nanoparticles of blue haze enhanced by anthropogenic pollution [J]. Proceedings of the National Academy of Sciences of the United States of America, 106(42): 17650-17654.

LI Xia

WU Jiarui

LIU Lang

LI Guohui
ligh@ieecas.cn

Ministry of Science and Technology of the People’s Republic of China (Y7YF051437)

Journal of Earth Environment