Modelling study of boundary-layer ozone over ...

Atmospheric Research 193 (2017) 83–93

Contents lists available at ScienceDirect

Atmospheric Research journal homepage: www.elsevier.com/locate/atmosres

Modelling study of boundary-layer ozone over northern China - Part II: Responses to emission reductions during the Beijing Olympics Guiqian Tang a,b, Xiaowan Zhu a,c, Jinyuan Xin a, Bo Hu a, Tao Song a, Yang Sun a, Lili Wang a, Fangkun Wu a, Jie Sun a, Mengtian Cheng a, Na Chao d, Xin Li a,e, Yuesi Wang a,⁎ a

State Key Laboratory of Atmospheric Boundary Layer Physics and Atmospheric Chemistry (LAPC), Institute of Atmospheric Physics, Chinese Academy of Sciences, Beijing 100029, China Center for Excellence in Regional Atmospheric Environment, Institute of Urban Environment, Chinese Academy of Sciences, Xiamen 361021, China University of Chinese Academy of Sciences, Beijing 100049, China d Environmental Science Research & Design Institute of Zhejiang Province, Hangzhou 310007, China e Mentougou District Government, Beijing 102300, China b c

a r t i c l e

i n f o

Article history: Received 21 October 2016 Received in revised form 22 January 2017 Accepted 24 February 2017 Available online 29 March 2017 Keywords: CMAQ Integrated process rate Emission control measures

a b s t r a c t The implementation of emission reduction measures during the Olympics provided a valuable opportunity to study regional photochemical pollution over northern China. In this study, the fifth-generation Pennsylvania State University/National Centre for Atmospheric Research Mesoscale Model and Community Multiscale Air Quality model system was applied to conduct two sets of modelling analyses of the period from July 20 to September 20, 2008, to illustrate the influences of emission reduction measures on regional photochemical pollution over northern China during the Beijing Olympics. The results indicated that the implementation of emission control measures decreased the concentrations of ozone (O3) precursors, namely nitrogen oxide (NOx) and volatile organic compounds (VOCs), throughout the boundary layer. The concentrations of these compounds were reduced by 45% in the central urban area of Beijing at the ground level. Although the average O3 concentration in the central urban area increased by more than 8 ppbv, the total oxidant concentration decreased significantly by more than 5 ppbv. Greater O3 concentrations mainly occurred during periods with weak photochemical reactions. During periods of strong photochemical production, the O3 concentration decreased significantly due to a weakening vertical circulation between the lower and upper boundary layer. Consequently, the number of days when the O3 concentration exceeded 100 ppbv decreased by 25% in Beijing. The emission control measures altered the sensitivity of the regional O3 production. The coordinated control region of NOx and VOCs expanded, and the control region of VOCs decreased in size. The reduction of non-point-source emissions, such as fugitive VOCs and vehicles, was more useful for controlling regional photochemical pollution over northern China. © 2017 Elsevier B.V. All rights reserved.

1. Introduction The Summer Olympic Games are one of the most influential and globally comprehensive sporting events. From the 1940s to the 1970s, London, Helsinki, Melbourne, Rome, Tokyo, Mexico City, Munich and Montreal each hosted the Summer Olympic Games. During this period, the economies of these countries developed rapidly, but people lacked adequate awareness of environmental protection. Thus, no special measures were adopted during the Olympic Games held during this period to protect the atmosphere and environment. Since the 1980s, and with global environmental protection progress, the Olympics organization has gradually began considering the environment in addition to sports activities.

⁎ Corresponding author. E-mail address: [email protected] (Y. Wang).

http://dx.doi.org/10.1016/j.atmosres.2017.02.014 0169-8095/© 2017 Elsevier B.V. All rights reserved.

As the political, economic and cultural centre of China, Beijing gained the right to host the 2008 Summer Olympic Games in 2001. Compared with other cities that had previously hosted the Olympics, a considerable gap in air quality occurred during the Beijing Olympics. With high concentrations of inhalable particulate matter (PM10) and ozone (O3), Beijing experiences continuous atmospheric pollution (Tang et al., 2009, 2012; Xin et al., 2010; Wang et al., 2015). From the end of 1998 to the opening ceremony of the 2008 Olympics, the Beijing Municipal Government successively promulgated and implemented fourteen stages of control measures for air pollution and effectively suppressed air pollution in Beijing. However, because of the severity of the regional air pollution in Beijing, the promised air quality was not sufficiently met following the implementation of the control measures (Streets et al., 2007). To fulfil the promise proposed during the application for the Olympics, the State Council approved the decision to establish the Coordinate Group for Air Quality Protection Work in Beijing during the 2008 Olympics to strengthen cooperation with peripheral provinces, cities and the

84

G. Tang et al. / Atmospheric Research 193 (2017) 83–93

autonomous region. This group was responsible for organizing and coordinating the Air Quality Safeguard Research of Beijing during the 29th Olympic Games - Measures of Beijing and Air Quality Safeguards Research of Beijing during the 29th Olympic Games - Measures for the Surrounding Provinces, City and Autonomous Region of Beijing (MEPPRC and BMG, 2007a, 2007b). Thus, based on the pollution control measures proposed by the relevant departments of the five provinces and cities in northern China, a research report on the Air Quality Safeguards Research of Beijing during the 29th Olympic Games - Measures of Five Provinces, Cities and Autonomous Region in Northern China was summarized and compiled (MEPPRC and BMG, 2007c). Based on this research report, the five provinces, cities and autonomous region in northern China jointly investigated the reduction of pollutants from atmospheric emission sources during the Olympics. During this period, researchers performed many observational studies. Studies of the emission sources indicated that the vehicle flow-rate was reduced by 32.3% by limiting which vehicles were allowed on the roads based on whether they had an odd or even plate number during the Olympics (Wang and Xie, 2009). The levels of volatile organic compounds (VOCs), monoxide (CO), nitrogen oxide (NOx) and PM10 that were discharged by motor vehicles were reduced by 55.5, 56.8, 45.7 and 51.6%, respectively (Zhou et al., 2010); and the decline of road dust was particularly significant (Fan et al., 2009). At urban and rural stations in Beijing, the contributions of motor vehicles to organic carbon (OC) emissions decreased by 30 and 24%, respectively, and the contributions of coal combustion decreased by 57 and 7%, respectively (Guo et al., 2013). Because of the effective implementation of emission reduction measures, the nitrogen dioxide (NO2) concentrations in the troposphere, the sulphur dioxide (SO2) column concentrations in the boundary layer and the CO concentrations at 700 hpa in the urban area of Beijing decreased by 40–59, 13 and 12%, respectively (Mijling et al., 2009; Witte et al., 2009). The concentrations of near-surface gaseous pollutants (NOx, CO, SO2 and VOCs) decreased significantly in the urban area of Beijing (Chou et al., 2011; Okuda et al., 2011; Wang et al., 2010a; Wang et al., 2009a; Wang and Xie, 2009; Xu et al., 2016), which also caused the CO and SO2 concentrations at Miyun station in the leeward region of Beijing to decrease significantly (Wang et al., 2009d). In addition, the VOC emissions during the Olympics decreased by 45% compared to June, and the emissions from motor vehicles, solvent use, industrial processes and fugitive uses decreased by 66, 48, 15 and 75%, respectively (Su et al., 2011). In particular, aromatic hydrocarbons with strong chemical activity decreased significantly, and benzene, toluene, ethylbenzene and xylenes (BTEX) decreased by 47–64% (Liu et al., 2009). However, no significant changes were observed in the concentrations of alkanes and benzene with long lifetimes (Wang et al., 2010b). Moreover, formaldehyde, aldehyde, methyl ethyl ketone and methyl alcohol were significantly reduced by 12.9, 15.8, 17.1 and 19.6%, respectively, and the concentrations of acetone did not significantly change during the Olympics (Liu et al., 2015). Noticeably, the formaldehyde concentrations peaked twice during the rush hours, indicating that motor vehicles remained the main sources of formaldehyde precursors (Li et al., 2010). The significant decrease in gaseous pollutants also caused a change in the particulate matter concentration (Xu et al., 2016). Studies based on laser radars have indicated that the light extinction coefficient for aerosols decreased by 42.3% during the Olympics compared to the same time period in 2007 in the urban area of Beijing (Yang et al., 2010). The optical thickness of aerosol, which was observed by satellites, also decreased considerably (Cermak and Knutti, 2009; Wang et al., 2009a). Moreover, a significant difference was observed in the variations of particulate matter with different particle sizes. In the urban area of Beijing, the PM10 and total suspended particulate (TSP) concentrations decreased significantly, but the reduction in fine particulate matter (PM2.5) was not significant (Schleicher et al., 2012; Schleicher et al., 2011; Wang and Xie, 2009; Wang et al., 2009b). The water soluble ions decreased the most in the TSPs and PM10, and no Ca2+ and SO2− 4

− obvious decreases in NH+ 4 and NO3 were observed (Norra et al., 2016; Okuda et al., 2011; Schleicher et al., 2012). The most significant decreases in the PM2.5 fraction were observed for elements related to human activities, such as S, Cu, As, Cd, Pb and black carbon (BC) (Fan et al., 2009; Norra et al., 2016; Schleicher et al., 2012; Wang et al., 2009d). The toxic substances in the particulate matter, namely, polycyclic aromatic hydrocarbons (PAHs), also exhibited a large decline (Fan et al., 2009; Ma et al., 2011; Norra et al., 2016; Schleicher et al., 2012), and the magnitude of the decline in PAHs was the largest for PAHs with between five and seven rings (Okuda et al., 2011). Moreover, by comparing the vertical profiles of the atmospheric extinction coefficient before and during the Olympics Games, the atmospheric extinction coefficient exhibited the most significant decrease from 0.5 to 1.5 km, indicating that the amount of PM10 from transportation decreased by 36.6% during the Olympic Games and demonstrating the effectiveness of the regional coordinated control measures (Yang et al., 2010). Although the concentrations of particulate matter and primary gaseous pollutants decreased during the Olympics, another important secondary pollutant, O3, increased (Chou et al., 2011; Wang and Xie, 2009). However, although the O3 concentration increased, the total oxidant (Ox), total reactive nitrogen (NOy) and NOz (NOy-NOx) concentrations in the atmosphere decreased (Chou et al., 2011), which caused the O3 concentration at the Miyun station in the leeward region of Beijing to decrease significantly and indicated that reducing the emissions in an urban area can affect a larger area (Wang et al., 2009c). Because of the emission reduction effects, the production of O3 was controlled by NOx during the Olympics and by VOCs before and after the Olympics (Chou et al., 2011; Witte et al., 2011; Xing et al., 2011). Although previous detailed and focused studies have been conducted, these studies contain the following drawbacks: (1) they excessively focus on the Beijing area with insufficient studies of the surrounding area; (2) most previous studies discuss only the distribution of pollution near the ground and ignore the characteristics of the air pollutants in the upper boundary layer; and (3) the emission reduction measures proposed in most previous studies are targeted at specific chemical species, such as NOx and VOCs, without considering a certain type of source. To compensate for these shortcomings in studies of O3 pollution over northern China, we considered the period of emission reduction during the Olympics and used a large amount of observational data to validate a numerical model system. Our analysis illustrates the effectiveness of the control strategies implemented during the Olympics, deeply explores the influences of emission control measures on the characteristics of the O3 distribution and its budget, and proposes additional operational control strategies for photochemical pollution over northern China.

2. Methodology 2.1. Air quality model simulation and observations Strict emission control measures were implemented before and after the 29th Olympic Games held in Beijing in August–September 2008. Therefore, we derived a list of emissions sources under different control measures based on the measures and strengths of emissions control. Afterwards, the period of June–September 2008 was divided into the following four emission control implementation stages: June, July 1–19, July 20–September 20 and September 20–30. The emission control measures had not yet been implemented over northern China in June 2008, which was two months before the opening ceremony of the Olympics. The periods of July 1–19 and September 20–30 were transition periods during which emissions measures were implemented and suspended, respectively. These three periods were not discussed because the variations of emission sources during these periods were relatively complicated. The Olympics and Paralympics took place from July 20 to September 20, 2008, which is a relatively long period. Because the emission levels were essentially stable once the emission sources were


controlled, the control factors of different emission sources during this period were easy to estimate. Therefore, we selected the time period of July 20 to September 20, 2008 and applied a list of emission sources without emission reduction (c1) and with emission reduction (c2) to conduct two simulations aimed at analysing the effects of emission reduction on the characteristics of O3 distributions and the O3 budget. The fifth-generation Pennsylvania State University/National Centre for Atmospheric Research Mesoscale Model (MM5) and Community Multiscale Air Quality (CMAQ) model system was adopted for this study, and the settings were identical to those described in Part I (Fig. S1 in the supplementary material). Integrated process rate (IPR) analysis was used to differentiate among the individual contributions of the chemical reactions (Chem), horizontal diffusion and advection (Htra = Hdif + Hadv), vertical diffusion and advection (Vtra = Vdif + Zadv), dry deposition (Ddep) and cloud processes (Clds) to the O3 concentrations. In addition, the stations from which observations were obtained and compared with the model simulation results were consistent with those described in Part I (Fig. S1). For the reader's convenience, the model configuration, basic emission inventory and analysis methods are included in the supplementary material. 2.2. Emission inventory during the Olympics To study the emission control factors during the Olympics, we describe the main emission control measures implemented over northern China during the Olympics. In the Beijing area, the main emission control measures were as follows: (1) power plants adopted low-nitrogen combustion technology and increased the intensity of desulphurization; (2) 50% of the 854 industrial boilers were shut down, 27 cement plants suspended production (except for the Beijing Cement Plant, which was on normal production), 142 concrete batching plants suspended production, all brickyards ceased production, 111 quarries and limeworks ceased production, 11 building material enterprises reduced their emission by 30%; the Eastern Chemical Plant and the Beijing Organic Chemical Plant of Beijing Eastern Petrochemical Co. Ltd. ceased production, the Beijing Yanshan Petrochemical Group reduced its emissions by 30%, and the Shougang Group reduced its production by 37%; (3) policies of limiting operation for yellow label vehicles and driving according to odd or even plate numbers were adopted for motor vehicles; and (4) measures to cease or reduce the production of solvent-using industries were effectively implemented, and automotive refinishing, furniture manufacturing, printing, dry cleaning and civilian decoration were reduced by 22, 59, 44, 58 and 20%, respectively. Because of the implementation of motor vehicle control measures, the fuel charges at gas stations decreased and the VOC emissions from gas stations were reduced by 43%. During the Olympics, the strength of the emission control measures in the surrounding areas of Beijing was weaker than that in the Beijing area, and the main emission control measures were as follows. (1) Power plants adopted low-nitrogen combustion technology and increased the intensity of desulphurization and 21 small thermal power units were shut down. The concentrations of NOx, SO2 and PM10 that were discharged by power plants were reduced by 29, 28 and 23%, respectively, which comply with the inferred amounts of reduction (Zhao et al., 2008). (2) The heavy-pollution industries ceased production; the middle-sized and small boilers were subject to gas reformation; 116 mechanical shaft kilns with diameters of 3 m or less and six rotary kilns were eliminated; 55 furnaces that were 200 m3 or smaller were eliminated in addition to 20 converters that were 20 tons or smaller and 18 electric furnaces; four coke ovens were eliminated, and the materials, resources and channels of product sales for some enterprises were blocked, which caused passive shutdowns. (3) In the Tianjin, Shijiazhuang, Baoding, Langfang and Tangshan areas, the motor vehicle control policy stipulated that vehicles with odd plate numbers could drive on odd days and vehicles with even plate numbers could drive

85

on even days. This measure was implemented from 07:00–22:00 LT each day. The implementation of the aforementioned measures considerably reduced pollutant emissions. In this study, according to the Studies of Air Quality Safeguards Research of Beijing during the 29th Olympic Games - Measures of Five Provinces, Cities and Autonomous Region in Northern China (MEPPRC and BMG, 2007c), by combining the census results from the Chinese Academy of Environmental Sciences on the emission sources before and after the Olympics, we aimed to obtain the control factors of different emission sources over northern China by using the amount of pollutant emissions, with June 2008 serving as a benchmark (Table S1 and S2 in the supplementary material). 3. Results and discussions 3.1. Evaluation of the air quality model during the Olympics Because the simulations were validated without the emission reduction measures in Part I (Tang et al., 2017), we performed a comparative study of the observations and simulations during the period of July 20 to September 20, 2008, to evaluate the performance of the simulation when emission reduction measures were implemented during the 2008 Olympics. A comparison of the daily average NOx concentrations at twelve stations indicated that the average values of the simulations and observations were relatively similar and that the mean bias (MB) was within 2.4–7.1 ppbv. At most stations, the normalized mean bias (NMB) and normalized mean error (NME) were less than ±25 and 35%, respectively, and the root mean square error (RMSE) between the observations and simulations was less than 9.4 ppbv (Fig. S2). A comparison between the observations and simulations for VOCs indicated that the average simulated and observed values at 09:00 LT were relatively similar at the stations in Beijing and Baoding (BD). The MB was approximately 16.0 ppbv, and the NMB was less than − 30%. Although the model underestimated the VOC concentration at 14:00 LT, the NMB ranged from −30 to −20% (Fig. S3). A comparison of the observed and simulated O3 concentrations at the twelve stations indicated that the simulated and observed average O3 concentrations were relatively similar and that the MB was less than 23.0 ppbv. For most stations, the NMB and NME were less than ± 10 and 30%, respectively, and the RMSEs between the observations and simulations were less than 30.0 ppbv (Fig. S4). By summarizing the results above, the MM5-CMAQ model system adequately simulated the characteristics of O3 and its precursors with the implementation of emission control measures during the Olympics. Thus, the list of emission sources and the model system established in this study can be used to analyse the characteristics of pollutants at the regional scale and to evaluate the environmental effects of control strategies during the Olympics. 3.2. Changes in regional photochemical pollution during the Olympics By applying the two sets of emission sources after strict validation (which represent the cases without and with the implementation of emission reduction measures), we analysed the two simulations for the period from July 20 to September 20, 2008, to illustrate the influences of emission control measures on the regional photochemical pollution over northern China (Fig. 1). According to a comparison of their spatial distributions, the NOx and VOC concentrations were primarily reduced over the urban area, and the area with reduced NOx concentrations was significantly larger than the area of reduced VOC concentrations (Fig. 1a and b). NOx concentrations were reduced the most in Beijing, Tianjin and Tangshan and in large areas with large amounts of point-source pollution, and VOCs were mainly reduced in Beijing and Tianjin. Because the reduction measures adopted in the urban area of Beijing were the strictest, the NOx and VOC concentrations decreased

86


Fig. 1. Changes in the spatial distribution of NOx (a), VOCs (b), O3 (c), O3 at 15:00 LT (d), Ox (e) and Ox at 15:00 LT (f) at the ground level between c1 and c2 (c2-c1) between July 20 and September 20, 2008, over northern China (unit: ppbv).

up to 45% (decreases of more than 15 ppbv) in the central area of the city (Fig. 1a and b). A significant difference was observed between the changes in the average O3 concentration and the changes in the NOx and VOC concentrations. The region of NOx reduction corresponded to a region with an enhanced O3 concentration (Fig. 1c). The regional average O3 concentration increased the most in Beijing, exceeding 8 ppbv. Because of the effects of regional emission reductions, the O3 concentration in the North China Plain slightly increased, which reflected how the weakening NO titration effect impacted the O3 concentration (Tang et al., 2009; Tang et al., 2012) (Fig. 1c). Although the average O3 concentration indicated that the regional O3 concentration increased, the O3 concentration displayed different characteristics in the afternoon. At 15:00 LT in the afternoon, the weakening of the titration effect caused the O3 concentration to increase in the region where a large reduction in point-source emissions occurred. However, a small decrease (smaller than 2 ppbv) in the O3 concentration was observed in most regions of Beijing and in its surrounding areas (Fig. 1d). Particularly, the O3 concentration decreased the most in northern Beijing, with a decrease of more than 6 ppbv. This decrease is consistent with the observations of previous studies (Wang et al., 2009d). To eliminate the effects of NO titration on the changes in photochemical pollution, we analysed the changes of the spatial distribution of Ox. As shown in Fig. 1e, the Ox concentrations decreased following the emission reduction measures in Beijing and the surrounding areas. The Ox concentration decreased the most in urban areas, exceeding a decrease of 6 ppbv. At 15:00 LT in the afternoon, a similar feature in the Ox concentrations was observed, but the region exhibiting this decrease was even larger. The centre of the area with reduced Ox concentrations was located in central and northern Beijing, and the Ox concentration was reduced by more than 7 ppbv in the central area (Fig. 1f). Overall, the implementation of emission control measures during the Olympics significantly reduced the average concentrations of NOx and VOCs in Beijing and the surrounding areas (particularly the Beijing area). Although the reduction in NOx caused the average O3

concentration to increase, the O3 concentrations decreased by different amounts in the afternoon with strong photochemical pollution. An analysis of the Ox concentrations indicated that implementing emission control measures caused the total oxidizing capacity of the atmosphere to decrease significantly in Beijing and its surrounding areas. 3.3. Changes in the O3 budget 3.3.1. Changes in the near-surface O3 budget 3.3.1.1. Changes in the monthly average O3 budget at the ground level. To investigate changes in the budget of near-surface O3, we analysed the IPR results from the first layer of the model (0–38 m above ground level, a.g.l.) between two simulations (Fig. 2). As discussed in Part I, chemical reactions serve as a sink for near-surface O3 (Tang et al., 2017). The intensity of the O3 sink decreased significantly over northern China as the NOx concentration decreased and was the most significant in the urban areas of Beijing and Tianjin, where it exceeded 7 ppbv·h−1 (Fig. 2a). Because the near-surface O3 concentration was enhanced, dry deposition slightly increased. This change was the most significant in the urban areas and large point-source regions, where the contribution of dry deposition increased by approximately 2 ppbv·h−1 (Fig. 2b). Moreover, due to the effects of reducing the emissions from large point sources, the contributions of horizontal transportation to large point-source regions were considerably reduced. In contrast, the changes in the contributions of horizontal transportation to the near-surface O3 were relatively small (less than 0.5 ppbv·h−1) in the other regions (Fig. 2c). Because of the effects of emission reductions near the ground, the contributions of vertical transportation to near-surface O3 decreased significantly. In the urban area, which displayed a considerable reduction in emissions, the greatest reduction in the contribution of vertical transportation was observed, and the contribution of vertical transport in the urban area decreased by more than 6 ppbv·h−1 (Fig. 2d). As described in Part I (Tang et al., 2017), chemical reactions and dry deposition were the main sinks of near-surface O3 in the study area, whereas vertical transportation was the main source to recoup their losses at


87

Fig. 2. Changes in the contributions of the chemical reactions (a), dry deposition (b), horizontal transportation (c) and vertical transportation (d) to the near-surface O3 between c1 and c2 (c2-c1) from July 20 to September 20, 2008 (unit: ppbv·h−1).

the ground level. Therefore, the incidence of vertical transportation as the main source of near-surface O3 decreased significantly, and the chemical reactions and dry deposition of near-surface O3 decreased and increased, respectively. 3.3.1.2. Changes in the diurnal variations of the O3 budget at the ground level. To identify changes in the budget of the near-surface O3 at different times, we selected the grid points that were consistent with the observation stations and statistically analysed the changes in the O3 budget towards the O3 concentrations at these grid points between two simulations (Fig. 3). To better study the changes at different stations, we divided the twelve stations into the following three classes according to the features of the emission sources: urban and suburb stations without a reduction in emissions from large point sources, stations with a reduction in emissions from large point sources, and background stations. The urban and suburb stations without emission reductions from large point sources included the 325-m-tall tower in Beijing (BJT), the Olympic Village in Beijing (AYC), and the Yangfang (YF), Shijiazhuang (SJZ), Cangzhou (CZ), Xianghe (XH), Langfang (LF), Zhuozhou (ZZ) and Yanjiao (YJ) stations. At these stations, the O3 concentration changes were very consistent with the diurnal variations of motor vehicle emissions. At all stations, the O3 profiles showed two peaks during the rush hours. In addition, the O3 concentrations were mainly enhanced at night, with relatively small changes during the day. Analysing the changes in the near-surface O3 budget indicated that the changes in the contributions of chemical reactions to the O3 concentration exhibited a bi-modal morphology, complying with the variations of NO emissions from motor vehicles. This result indicated that the reduction in NO emissions caused the sink of O3 to decrease. In addition, the changes in the contributions of vertical diffusion to the O 3

concentration exhibited a single-valley morphology and reached a valley value at approximately 15:00 LT in the afternoon. At noon, the increase in the contribution of chemical reactions to the near-surface O3 was the smallest, and the decrease in the contribution of vertical diffusion to the near-surface O 3 was the largest. Consequently, the O 3 concentrations decreased in the afternoon at this type of station. For the stations with reduced emissions from large point sources, including the BD and Tangshan (TS) stations, the O3 concentration changes were significantly correlated with the diurnal variations of the boundary-layer height and exhibited obvious peaks that corresponded to the enhanced O3 concentrations during the day. When analysing the variations in the budget of the near-surface O3, it was observed that the changes in the contributions of chemical reactions to the O3 concentration increased significantly and exhibited a single-peak morphology that complied with the changes in O3 concentrations. However, the changes in the contributions of vertical diffusion and horizontal transportation to O3 declined significantly. At these stations, because of the high emission height and hot exhaust gas of the elevated point sources, air pollutants emitted from the chimneys broke through the stable boundary layer and were stored in the residual layer at night. When turbulence developed after sunrise, the boundary layer became elevated, and the air pollutants in the residual layer were mixed towards the surface during the day. Then, the NO emitted at night participated in the titration reaction with O3. Because of the significant reduction in emissions, the weakening of the titration reaction caused the O3 concentration to increase at noon. At the Xinglong (XL) background station, the change in the nearsurface O3 concentration was relatively small, and a small decrease was observed between 13:00 and 20:00 LT. This phenomenon was primarily caused by a reduction in the contributions of vertical inputs,

88


Fig. 3. Changes in the diurnal variations of near-surface O3 and its budget between c1 and c2 (c2-c1) at twelve stations over northern China from July 20 to September 20, 2008. Zadv: vertical advection, Hadv: horizontal advection, Clds: cloud processes, Ddep: dry deposition, Vdif: vertical diffusion, Hdif: horizontal diffusion, Chem: chemical reactions.

which reflected the effects of regional emission reductions on the regional O3 concentration. By summarizing the analysis above, a significant correlation was found between the changes in the O3 concentrations and the emission control strategies at different locations. The O3 concentration increased significantly at night at stations with reduced motor vehicle emissions, and no significant changes in the O3 concentrations were observed during the day. At the stations with reduced emissions from large point sources, the O3 concentrations increased significantly during the daytime. The reduced vertical input at the XL station demonstrated that the implementation of emission control measures in Beijing and its surrounding areas significantly affected the O3 concentrations at the regional scale. 3.3.2. Changes in the O3 budget in the boundary layer 3.3.2.1. Changes in the monthly average O3 budget in the boundary layer. To better describe the changes in the O3 budget during the period of intense photochemical production at noon, we considered the entire boundary layer and examined the changes in the O3 budget from 09:00 to 15:00 LT between the two simulations (Fig. 4). The variations in the boundary-layer O3 budget were not consistent with the near-surface characteristics (Fig. 2 and 4). Although the contributions of chemical reactions to O3 increased near the large point sources in the boundary layer, a significant decrease in the chemical reactions of more than 3 mg·m−2·h−1 was observed for the entire city of Beijing (Fig. 4a). Because the amounts of large point-source emissions decreased significantly, the horizontal transport near the point sources

decreased by more than 3 mg·m−2·h−1. However, because the O3 concentration near the point sources increased, the contribution of horizontal transport increased significantly in the leeward region of the point sources. The horizontal transport changed the most in the northern area of Beijing, exceeding 4 mg·m−2·h−1 (Fig. 4c). The contribution of the vertical output of O3 from the boundary layer slightly decreased, and the output from the plain area was reduced by less than 1 mg·m−2·h−1 (Fig. 4d). In addition, dry deposition had a small effect on the O3 concentrations throughout the boundary layer (Fig. 4b). 3.3.2.2. Changes in the vertical O3 budget in the boundary layer. To examine the changes in the O3 budget in each layer of the boundary layer in detail, we selected grid points that were consistent with the observation stations during the period 09:00–15:00 LT between the two simulations (Fig. 5). After implementing the emission control measures, the O3 concentrations below the sixth layer of the model (555 m) increased to varying extents at 09:00 LT at all of the stations. According to the strength of the emission control measures at the different stations, the near-surface O3 concentrations increased by 3–15 ppbv. In contrast with the changes in the O3 concentrations at 09:00 LT, three features were observed at 15:00 LT (Fig. 5). (1) At the stations where non-point-source emission was reduced (BJT, AYC, YF, SJZ, CZ, XH, LF, ZZ and YJ), the near-surface O3 concentrations changed slightly. However, as the altitude increased, the O3 concentrations gradually decreased, and the largest reductions in the O3 concentration occurred in the seventh or eighth layers of the model (1080 or 1680 m). (2) At the stations with reduced emissions from large point sources (TS and BD), the O3 concentrations increased


89

Fig. 4. Changes in the contributions of chemical reactions (a), dry deposition (b), horizontal transportation (c) and vertical transportation (d) to O3 between c1 and c2 (c2-c1) in the boundary layer over northern China from July 20 to September 20, 2008 (unit: mg·m−2·h−1).

significantly in the boundary layer, and the magnitude of this increase decreased with increasing altitude. (3) At the background XL station, the O3 concentrations obviously decreased below the eighth layer of the model, and this decrease was not considerably different between the different layers. Most importantly, the differences between the O3 concentrations at 15:00 LT and 09:00 LT reflect the changes in the net O3 increase from 09:00 to 15:00 LT. The equations used to describe this relationship are as follows: ΔO3 ð09Þ ¼ O3 ðC2Þð09Þ−O3 ðC1Þð09Þ

ð1Þ

ΔO3 ð15Þ ¼ O3 ðC2Þð15Þ−O3 ðC1Þð15Þ

ð2Þ

ΔO3 ð15Þ−ΔO3 ð09Þ ¼ ½O3 ðC2Þð15Þ−O3 ðC1Þð15Þ−½O3 ðC2Þð09Þ−O3 ðC1Þð09Þ ¼ ½O3 ðC2Þð15Þ−O3 ðC2Þð09Þ−½O3 ðC1Þð15Þ−O3 ðC1Þð09Þ ¼ ΔO3 ðC2Þð15−09Þ−ΔO3 ðC1Þð15−09Þ

ð3Þ

Except for the reduction in emissions from large point sources, the net increase in O3 during the period 09:00–15:00 LT decreased significantly after the emission control measures were implemented at the stations in different layers. The net increase in near-surface O3 was reduced the most over northern China. To determine what caused the changes in the O3 concentrations in the vertical direction, the changes in the O3 budget were examined in each layer of the boundary layer. At the stations with decreased large point-source emissions, the main processes that affected the changes in the O3 concentrations were horizontal and vertical transport and chemical reactions. The reduction of NOx emissions from large point sources decreased the number of chemical reactions between NO and O3, which reduced the consumption of O3 from the ground to the upper boundary layer and significantly increased the contributions of photochemical production. Because the O3 concentrations were enhanced near the point sources, the changes of the O3 concentrations decreased with distance from

the station, causing the horizontal input to decrease significantly. Otherwise, the decreasing different between the O3 concentrations in the lower and upper boundary layers decreased the amount of vertical transportation significantly. At the stations where non-point-source emissions were reduced, the main processes that affected the O3 concentrations were vertical transport and chemical reactions. Reducing the near-surface NOx emissions decreased the number of NO and O3 reactions, which reduced the consumption of near-surface O3. Meanwhile, the changes in emissions also reduced the transport of NO2 from the lower boundary layer and consequently the photochemical production of O3 in the upper boundary layer. Because the near-surface O3 concentration increased, the differences between the O3 concentrations in the lower and upper boundary layers decreased. This decrease reduced the amount of vertical diffusion from the upper boundary layer to the surface. Therefore, the implementation of emission control measures reduced the amount of physiochemical circulation between the lower and upper boundary layers, as shown in Part I (Fig. 6). As shown in the aforementioned analysis and because the emissions over northern China were reduced, the amount of O3 generated from photochemical reactions decreased significantly during the period of 09:00–15:00 LT in the boundary layer in most areas of the North China Plain. Therefore, the high O3 concentrations at 15:00 LT primarily resulted from the increase of the initial O3 concentration at 09:00 LT. 3.4. Changes in O3 sensitivity The P(H2O2)/P(HNO3) ratio was considered to determine the effects of emission control on the sensitivity of O3 production over northern China and indicated that the region in which the ratio of P(H2O2)/ P(HNO3) was smaller than 0.06 shrank significantly in the areas of Beijing, Tianjin and Langfang at 15:00 LT in the afternoon (Fig. 7). This change indicated that the region in which the O3 production is controlled by VOCs shrank significantly. Although the power plants reduced the amount of regional NOx emissions over northern China, no significant change in the sensitivity of O3 production in the surrounding

90


Fig. 5. Changes in the O3 concentrations at 09:00 and 15:00 LT and the O3 budget from 09:00 to 15:00 LT between c1 and c2 (c2-c1) in the boundary layer recorded at the twelve stations over northern China from July 20 to September 20, 2008. Clds: cloud processes, Vtra: vertical transportation, Ddep: dry deposition, Htra: horizontal transportation, Chem: chemical reactions.

regions of Beijing was observed. South of Hebei Province, the regional P(H2O2)/P(HNO3) ratio was between 0.06 and 0.2, indicating that O3 production was subject to the coordinated control of NOx and VOCs. For the mountainous areas in northern Hebei Province, the P(H2O2)/

P(HNO3) ratio was greater than 0.2, indicating that the O3 production in this region was controlled by NOx. The changes in O3 sensitivity in the areas of Beijing, Tianjin and Langfang demonstrated that the reduction in emissions from non-point sources is more helpful for altering O3 sensitivity.

3.5. Evaluation of emission reductions on surface O3 concentrations

Fig. 6. Schematic showing the changes in the vertical circulation of O3 and its precursors in the boundary layer during the Olympics.

To validate the aforementioned theory, we compared the O3 concentrations and P(Ox) in two simulations. Because the emission control measures were different at the twelve stations, the relationships between the O3 concentrations and P(Ox) in both of the simulations exhibited different features at different stations (Table 1 and Fig. 8). According to the strength of the emission control measures and the category of control, we also divided these twelve stations into the following three classes: (1) urban stations with BJT as the representative station at which the non-point-source emissions declined significantly; (2) suburb stations with XH as the representative station at which the emissions of the surrounding non-point sources declined significantly; and (3) stations with BD as the representative station at which the emissions of point sources declined significantly. By analysing the relationships between the daily maximum O3 and P(Ox) during 12:00– 17:00 LT from the two simulations, it was found that the P(Ox) decreased significantly at stations BJT and XH, with slopes of 0.65 and


91

Fig. 7. The ratio of P(H2O2)/P(HNO3) at 15:00 LT from July 20 to September 20, 2008, over northern China before (a) and after (b) the implementation of emission control measures.

0.78, respectively, and that the P(Ox) increased significantly at BD, with a slope of 1.16 after the implementation of these measures. The changes in the P(Ox) concentrations directly reflect the O3 concentrations. Consequently, the slopes of the O3 concentrations from the simulation were 0.73 and 0.86 at stations BJT and XH, respectively. Although the slopes of the O3 concentrations at the stations decreased by different degrees, the intercepts of all of the stations were greater than 0. Particularly, the largest intercepts, which exceeded 19 ppbv, were observed at stations BJT and AYC. Overall, when the O3 concentration was relatively low, the O3 concentration increased significantly at all of the stations because of the weaker NO titration effect. When the O3 concentration was relatively high, the O3 concentration decreased significantly because of the lower P(Ox) at the stations with reduced non-point-source emissions, and the O3 concentration increased significantly because of the higher P(Ox) at the stations with reduced pointsource emissions. Therefore, the reduction of non-point-source emissions (such as fugitive VOCs and motor vehicles) was helpful for reducing the high O3 concentrations in the afternoon, and the reduction of point-source emissions resulted in the opposite effect. Thus, reducing the pollutant emissions from motor vehicles and other non-point sources is an optimum approach for attaining the required O3 standard. By analysing the number of days when the daily maximum exceeded the national level-II standard (100 ppbv) (MEPPRC and GAQSIQPRC, 2012) in the results of the two simulations, it was shown that the exceedance decreased from 16 to 12 days at station BJT because of the implemented emission control measures. Consequently, the emission control measures effectively reduced the number of days when the O3 concentrations exceeded the national standard in the Beijing area

Table 1 Correlation between the daily maximum O3 concentration and P(Ox) between 12:00 and 17:00 LT without and with the emission control measures implemented at twelve stations over northern China between July 20 and September 20, 2008. Sites

BJT AYC YF BD SJZ CZ TS XH LF ZZ YJ XL

P(Ox)

O3 Slope

Intercept

R2

Slope

Intercept

R2

0.73 0.73 0.94 0.95 0.94 0.97 0.98 0.86 0.91 0.84 0.84 0.82

19.2 19.8 5.5 8.9 5.7 4.0 9.0 9.3 7.1 10.4 13.5 13.0

0.91 0.91 0.91 0.96 0.99 0.98 0.89 0.96 0.96 0.93 0.91 0.94

0.66 0.61 0.75 1.16 1.00 0.91 1.25 0.78 0.86 0.79 0.82 0.85

0.81 1.04 1.61 0.75 0.42 0.65 −0.62 0.71 0.62 0.90 1.37 0.13

0.94 0.93 0.81 0.96 0.98 0.96 0.98 0.94 0.97 0.95 0.87 0.95

during the Olympics, and the magnitude of this decrease was approximately 25%. In summary, although the O3 concentrations increased at night and during periods of weak photochemical production, the O3 concentration did not exceed the national standard. During the pollution time period of strong photochemical production, the amount of O3 reduction was large over the North China Plain, which effectively reduced the number of days when O3 surpassed the national standard, demonstrating the effectiveness of the emission control measures. 4. Conclusions During the period of the 29th Olympic Games in Beijing, the weather and air quality in the city of Beijing were closely monitored by China and other countries. To protect the air quality in the city of Beijing, the State Council, the Beijing Municipal Government and the surrounding provinces, cities and autonomous region worked together to guarantee the promise of a “Green Olympics” in Beijing in 2008. These organizations improved the regional atmospheric environment together through a large effort and successfully achieved the overall control of atmospheric pollution in the Beijing-Tianjin-Hebei urban agglomeration. In this study, we applied two sets of emission inventories without and with the implementation of emission control measures to simulate the changes in the O3 concentrations and budget between July 20 and September 20, 2008. Based on a comparative analysis of the two simulations, the following main conclusions were reached. The implementation of emission control measures caused the average precursor concentrations to decrease, although the O3 concentrations near the surface increased. The implementation of emission control measures reduced the physiochemical circulation between the lower and upper boundary layer, which significantly reduced the Ox concentration and indicated that the contributions of chemical reactions were weaker in the boundary layer during the Olympics. The effects of implementing emission control measures on the O3 concentrations were nonlinear, and the changes were inconsistent for different weather conditions and time periods. In the areas with decreased motor vehicle emissions, the O3 concentration increased because the NO titration effect decreased. This phenomenon occurred during the night and when the photochemical reactions were not strong. During this time period of strong photochemical reactions, the O3 concentration decreased significantly because the P(Ox) were lower. In areas with decreased point-source emissions, the O3 concentration increased significantly at noon because of the NO titration effect decreased. Although implementing emission control measures in the regions where non-point-source emissions were reduced effectively decreased the number of days when the O3 concentrations exceeded the national

92


Fig. 8. Correlation between the daily maximum O3 concentration (a) and P(Ox) (b) between 12:00 and 17:00 LT without and with the emission control measures implemented at stations BJT, XH and BD between July 20 and September 20, 2008.

standard, regional photochemical pollution remained serious over northern China. The implementation of emission reduction measures caused the regions where O3 production was subject to sensitive control by VOCs to shrink significantly. Consequently, the best way to control photochemical pollution in Beijing and its surrounding areas is to reduce non-point-source emissions, including fugitive VOCs and motor vehicle emissions, across the North China Plain. According to our previous study, the implementation of strict emission standards for motor vehicles is effective for decreasing vehicle-related emissions (Tang et al., 2016). However, the oil quality in China is eight to ten years behind that in Europe, and the proportion of unsaturated hydrocarbons in gasoline in North China is much higher than that in the USA and Europe (Tang et al., 2015). Therefore, the best way to decrease motor vehicle emissions in North China is to improve oil quality.

Acknowledgements This work was supported by the CAS Strategic Priority Research Program Grant (no. XDB05020000 and XDA05100100), the National Natural Science Foundation of China (nos. 41230642 and 41222033) and the Beijing Municipal Commission of Science and Technology (No. D090409033670902). Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.atmosres.2017.02.014.

References Cermak, J., Knutti, R., 2009. Beijing Olympics as an aerosol field experiment. Geophys. Res. Lett. 36, L10806. Chou, C.C.K., Tsai, C.Y., Chang, C.C., Lin, P.H., Liu, S.C., Zhu, T., 2011. Photochemical production of ozone in Beijing during the 2008 Olympic Games. Atmos. Chem. Phys. 11, 9825–9837. Fan, S.-B., Tian, G., Li, G., Huang, Y.-H., Qin, J.-P., Cheng, S.-Y., 2009. Road fugitive dust emission characteristics in Beijing during Olympics Game 2008 in Beijing, China. Atmos. Environ. 43, 6003–6010. Guo, S., Hu, M., Guo, Q., Zhang, X., Schauer, J.J., Zhang, R., 2013. Quantitative evaluation of emission controls on primary and secondary organic aerosol sources during Beijing 2008 Olympics. Atmos. Chem. Phys. 13, 8303–8314. Li, Y., Shao, M., Lu, S., Chang, C.-C., Dasgupta, P.K., 2010. Variations and sources of ambient formaldehyde for the 2008 Beijing Olympic games. Atmos. Environ. 44, 2632–2639. Liu, J., Mu, Y., Zhang, Y., Zhang, Z., Wang, X., Liu, Y., Sun, Z., 2009. Atmospheric levels of BTEX compounds during the 2008 Olympic Games in the urban area of Beijing. Sci. Total Environ. 408, 109–116. Liu, Y., Yuan, B., Li, X., Shao, M., Lu, S., Li, Y., Chang, C.C., Wang, Z., Hu, W., Huang, X., He, L., Zeng, L., Hu, M., Zhu, T., 2015. Impact of pollution controls in Beijing on atmospheric oxygenated volatile organic compounds (OVOCs) during the 2008 Olympic Games: observation and modeling implications. Atmos. Chem. Phys. 15, 3045–3062. Ma, W.L., Sun, D.Z., Shen, W.G., Yang, M., Qi, H., Liu, L.Y., Shen, J.M., Li, Y.F., 2011. Atmospheric concentrations, sources and gas-particle partitioning of PAHs in Beijing after the 29th Olympic Games. Environ. Pollut. 159, 1794–1801.

Mijling, B., van der A, R.J., Boersma, K.F., Van Roozendael, M., De Smedt, I., Kelder, H.M., 2009. Reductions of NO2 detected from space during the 2008 Beijing Olympic Games. Geophys. Res. Lett. 36, L13801. Ministry of Environmental Protection of the People's Republic of China & Beijing Municipal Government (MEPPRC and BMG), 2007a. Air Quality Safeguards Research of Beijing During the 29th Olympic Games - Measures of Beijing, Beijing. Ministry of Environmental Protection of the People's Republic of China & Beijing Municipal Government (MEPPRC and BMG), 2007b. Air Quality Safeguards Research of Beijing During the 29th Olympic Games - Measures of the Surrounding Provinces, City and Autonomous Region of Beijing, Beijing. Ministry of Environmental Protection of the People's Republic of China & Beijing Municipal Government (MEPPRC and BMG), 2007c. Air Quality Safeguards Research of Beijing During the 29th Olympic Games - Measures of Five Provinces, Cities and Autonomous Region in Northern China, Beijing. Ministry of Environmental Protection of the People's Republic of China & General Administration of Quality Supervision, Inspection and Quarantine of the People's Republic of China (MEPPRC and GAQSIQPRC), 2012. Ambient Air Quality Standards, GB3095-2012. Norra, S., Yu, Y., Dietze, V., Schleicher, N., Fricker, M., Kaminski, U., Chen, Y., Stüben, D., Cen, K., 2016. Seasonal dynamics of coarse atmospheric particulate matter between 2.5 μm and 80 μm in Beijing and the impact of 2008 Olympic Games. Atmos. Environ. 124, 109–118. Okuda, T., Matsuura, S., Yamaguchi, D., Umemura, T., Hanada, E., Orihara, H., Tanaka, S., He, K., Ma, Y., Cheng, Y., Liang, L., 2011. The impact of the pollution control measures for the 2008 Beijing Olympic Games on the chemical composition of aerosols. Atmos. Environ. 45, 2789–2794. Schleicher, N., Norra, S., Dietze, V., Yu, Y., Fricker, M., Kaminski, U., Chen, Y., Cen, K., 2011. The effect of mitigation measures on size distributed mass concentrations of atmospheric particles and black carbon concentrations during the Olympic Summer Games 2008 in Beijing. Sci. Total Environ. 412-413, 185–193. Schleicher, N., Norra, S., Chen, Y., Chai, F., Wang, S., 2012. Efficiency of mitigation measures to reduce particulate air pollution—a case study during the Olympic Summer Games 2008 in Beijing, China. Sci. Total Environ. 427-428, 146–158. Streets, D.G., Fu, J.S., Jang, C.J., Hao, J.M., He, K.B., Tang, X.Y., Zhang, Y.H., Wang, Z.F., Li, Z.P., Zhang, Q., Wang, L.T., Wang, B.Y., Yu, C., 2007. Air quality during the 2008 Beijing Olympic Games. Atmos. Environ. 41, 480–492. Su, J., Shao, M., Lu, S., Xie, Y., 2011. Non-methane volatile organic compound emission inventories in Beijing during Olympic Games 2008. Atmos. Environ. 45, 7046–7052. Tang, G., Li, X., Wang, Y., Xin, J., Ren, X., 2009. Surface ozone trend details and interpretations in Beijing, 2001–2006. Atmos. Chem. Phys. 9, 8813–8823. Tang, G., Wang, Y., Li, X., Ji, D., Hsu, S., Gao, X., 2012. Spatial-temporal variations in surface ozone in Northern China as observed during 2009–2010 and possible implications for future air quality control strategies. Atmos. Chem. Phys. 12, 2757–2776. Tang, G., Sun, J., Wu, F., Sun, Y., Zhu, X., Geng, Y., Wang, Y., 2015. Organic composition of gasoline and its potential effects on air pollution in North China. SCIENCE CHINA Chem. 58, 1416–1425. Tang, G., Chao, N., Wang, Y., Chen, J., 2016. Vehicular emissions in China in 2006 and 2010. J. Environ. Sci. Tang, G., Zhu, X., Xin, J., Hu, B., Song, T., Sun, Y., Zhang, J., Wang, L., Li, X., Wang, Y., 2017. Modelling study of boundary-layer ozone over northern China-Part I: ozone budget in summer. Atmos. Res. 187:128–137. http://dx.doi.org/10.1016/j.atmosres.2016.10. 017. Wang, T., Xie, S., 2009. Assessment of traffic-related air pollution in the urban streets before and during the 2008 Beijing Olympic Games traffic control period. Atmos. Environ. 43, 5682–5690. Wang, M., Zhu, T., Zheng, J., Zhang, R.Y., Zhang, S.Q., Xie, X.X., Han, Y.Q., Li, Y., 2009a. Use of a mobile laboratory to evaluate changes in on-road air pollutants during the Beijing 2008 Summer Olympics. Atmos. Chem. Phys. 9, 8247–8263. Wang, W.T., Primbs, T., Tao, S., Simonich, S.L.M., 2009b. Atmospheric particulate matter pollution during the 2008 Beijing Olympics. Environ. Sci. Technol. 43, 5314–5320. Wang, X., Westerdahl, D., Chen, L.C., Wu, Y., Hao, J., Pan, X., Guo, X., Zhang, K.M., 2009c. Evaluating the air quality impacts of the 2008 Beijing Olympic Games: on-road emission factors and black carbon profiles. Atmos. Environ. 43, 4535–4543.

G. Tang et al. / Atmospheric Research 193 (2017) 83–93 Wang, Y., Hao, J., McElroy, M.B., Munger, J.W., Ma, H., Chen, D., Nielsen, C.P., 2009d. Ozone air quality during the 2008 Beijing Olympics: effectiveness of emission restrictions. Atmos. Chem. Phys. 9, 5237–5251. Wang, B., Shao, M., Lu, S.H., Yuan, B., Zhao, Y., Wang, M., Zhang, S.Q., Wu, D., 2010a. Variation of ambient non-methane hydrocarbons in Beijing city in summer 2008. Atmos. Chem. Phys. 10, 5911–5923. Wang, T., Nie, W., Gao, J., Xue, L.K., Gao, X.M., Wang, X.F., Qiu, J., Poon, C.N., Meinardi, S., Blake, D., Wang, S.L., Ding, A.J., Chai, F.H., Zhang, Q.Z., Wang, W.X., 2010b. Air quality during the 2008 Beijing Olympics: secondary pollutants and regional impact. Atmos. Chem. Phys. 10, 7603–7615. Wang, Z., Li, Y., Chen, T., Zhang, D., Sun, F., Wei, Q., Dong, X., Sun, R., Huan, N., Pan, L., 2015. Ground-level ozone in urban Beijing over a 1-year period: temporal variations and relationship to atmospheric oxidation. Atmos. Res. 164-165, 110–117. Witte, J.C., Schoeberl, M.R., Douglass, A.R., Gleason, J.F., Krotkov, N.A., Gille, J.C., Pickering, K.E., Livesey, N., 2009. Satellite observations of changes in air quality during the 2008 Beijing Olympics and Paralympics. Geophys. Res. Lett. 36, L17803. Witte, J.C., Duncan, B.N., Douglass, A.R., Kurosu, T.P., Chance, K., Retscher, C., 2011. The unique OMI HCHO/NO2 feature during the 2008 Beijing Olympics: implications for ozone production sensitivity. Atmos. Environ. 45, 3103–3111. Xin, J., Wang, Y., Tang, G., Wang, L., Sun, Y., Wang, Y., Hu, B., Song, T., Ji, D., Wang, W., Li, L., Liu, G., 2010. Variability and reduction in atmospheric pollutants in Beijing and its

93

surrounding area during the Beijing 2008 Olympic Games. Chin. Sci. Bull. 55, 1937–1944. Xing, J., Zhang, Y., Wang, S., Liu, X., Cheng, S., Zhang, Q., Chen, Y., Streets, D.G., Jang, C., Hao, J., Wang, W., 2011. Modeling study on the air quality impacts from emission reductions and atypical meteorological conditions during the 2008 Beijing Olympics. Atmos. Environ. 45, 1786–1798. Xu, R., Tang, G., Wang, Y., Tie, X., 2016. Analysis of a long-term measurement of air pollutants (2007–2011) in North China Plain (NCP); impact of emission reduction during the Beijing Olympic Games. Chemosphere 159, 647–658. Yang, T., Wang, Z., Zhang, B., Wang, X., Wang, W., Gbauidi, A., Gong, Y., 2010. Evaluation of the effect of air pollution control during the Beijing 2008 Olympic Games using Lidar data. Chin. Sci. Bull. 55, 1311–1316. Zhao, Y., Wang, S.X., Duan, L., Lei, Y., Cao, P.F., Hao, J.M., 2008. Primary air pollutant emissions of coal-fired power plants in China: current status and future prediction. Atmos. Environ. 42, 8442–8452. Zhou, Y., Wu, Y., Yang, L., Fu, L., He, K., Wang, S., Hao, J., Chen, J., Li, C., 2010. The impact of transportation control measures on emission reductions during the 2008 Olympic Games in Beijing, China. Atmos. Environ. 44, 285–293.