Non-agricultural ammonia emissions in urban China

Introduction Conclusions References


Introduction
Ammonia (NH 3 ) is the most abundant atmospheric base with the ability to neutralize sulphuric and nitric acids (oxidation products of SO 2 and NO x (NO+NO 2 ), respectively) to form secondary particulate matter (PM) with an aerodynamic diameter equal to or less than 2.5 µm (PM 2.5 ) in the form of (NH 4 ) 2 SO 4 , NH 4 HSO 4 , and NH 4 NO 3 (Seinfeld and Pandis, 2012).Current ambient air quality standards in many developed coun-Figures

Back Close
Full tries and some developing countries covered PM 2.5 in the book because of its adverse effects on human health, including premature mortality, chronic bronchitis, hospital admissions, and asthma attacks (Dockery et al., 1993;Nel, 2005;Pope et al., 2002;van Donkelaar et al., 2010).NH 3 , NO x and SO 2 emissions reduction is the key to lower ambient levels of PM 2.5 (Megaritis et al., 2013;Pathak et al., 2009).Major world economies have targeted reducing NO x and SO 2 emissions.However, there are no regional emission ceilings set for NH 3 (with the exception of the EU27 levels for European countries arising from the Gothenburg protocol, Reis et al., 2012) despite the fact that control technologies are cost-effective compared to NO x and SO 2 (Pinder et al., 2007).China has over-fulfilled the national goal of a 10 % reduction in SO 2 emissions from 2005 to 2010 by 14.3 %, and NO x emissions are planned to be cut by 10 % during the 12th Five-Year Plan (2011)(2012)(2013)(2014)(2015).Therefore, NH 3 is expected to play an increased role in PM 2.5 formation during the coming years (Chang et al., 2012).
It is well known that agricultural sources, notably animal manure and fertilizer application, contribute the most to NH 3 emissions (Cui et al., 2013;B. Gu et al., 2012).
However, researchers have found that there are a myriad of important but frequently overlooked anthropogenic non-agricultural activities (e.g., vehicles and landfill) contributing to NH 3 emissions (Battye et al., 2003;Pierson and Brachaczek, 1983;Sutton et al., 2000Sutton et al., , 2008;;Wilson et al., 2004), and many countries do not report emissions for all these terms.In the UK, the non-agricultural emissions of NH 3 accounts for around 15 % of the total national NH 3 emissions, in which the transport sector is the main source (16 %), followed by sewage emissions (12 %) (Dragosits et al., 2008;Sutton et al., 2000).This clearly indicates that the emissions and sources of NH 3 deserve a more comprehensive discussion in scientific community.Although compared to dominant agricultural NH 3 source sectors such as livestock operations, these nonagricultural sources form a small part of the global NH 3 emissions (Bouwman et al., 1997), and they might be more locally concentrated, particularly at an urban level.Moreover, those strong rural NH 3 emissions can hardly make a long-range transport in gaseous phase to influence urban areas unless reacting locally to form particulate NH The non-agricultural NH 3 emissions in an urban SO 2 -and NO x -rich atmosphere can be fully neutralized (Behera and Sharma, 2010; X. Huang et al., 2011), while model results showed that over half of NH 3 emissions would be deposited downwind of its source within 10 km depending on local meteorological conditions.Meanwhile, the remaining part of NH 3 in rural areas has less chance to convert to particulate NH + 4 (Asman et al., 1998).Various studies worldwide suggest that the ambient levels of NH 3 concentration in urban areas are comparable with, or even higher than that of rural areas (Doyle et al., 1979;Cadle et al., 1982;Allen et al., 1989Allen et al., , 2011;;Giroux et al., 1997;Perrino et al., 2002;Burgard et al., 2006;Li et al., 2006;Whitehead et al., 2007;Alebic-Juretic, 2008;Cao et al., 2009;Shen et al., 2009;Tanner, 2009;Behera and Sharma, 2010;Bishop et al., 2010;Ianniello et al., 2010;Gong et al., 2011;Meng et al., 2011;Pandolfi et al., 2012;Reche et al., 2012;Ye et al., 2011;Zbieranowski and Aherne, 2012).In addition, the high seasonal variability of agricultural activities in rural areas tends to make pulse emissions of NH 3 , but the situation is much better for the case of non-agricultural activities.These evidences mentioned above supporting a hypothesis that the non-agricultural NH 3 emissions contribute to PM 2.5 formation in urban areas may outweigh the contribution of agricultural NH 3 emissions in a scale of a full year.China, historically a nation of mostly agriculture activity, is in the vanguard of a wave of urban expansion that is driving the country towards an economic superpower.In the course of intense expansion, China urbanized nearly half of its people in 2010 (49.95 %) compared with 20 % in 198020 % in (Gong et al., 2012).Beijing, for example, with over 10 million migrant workers and a similar size of local citizens already, is a living experiment in urbanization -and one that is failing to shine largely because of its severe air pollution (Watt, 2005;Zhang et al., 2012).In fact, alarm about the perilous state of PM 2.5 pollution in Beijing has provoked a huge amount of public outcry and media attention lately (Chang, 2012).Beijing is however by no means unusual in today's China, it is over 50 % cities in China have serious air pollution, and more than 75 % of the urban population are exposed to high concentrations of both primary and secondary PM that does not meet the Chinese NAAQS (national ambient air quality standards) Introduction

Conclusions References
Tables Figures

Back Close
Full  (Shao et al., 2006).Over the past several years, China has implemented a portfolio of plans to run more on-road vehicles with renewable energy, phase out its coal-fired power plants, and raise standards for auto emissions (Chang et al., 2012).If all goes to plan, China could expect a substantial emission reduction of PM 2.5 and its precursors (mainly NO x and SO 2 ) in the coming years.However, current benefits might be partially offset in the absence of the non-agricultural NH 3 control (Chang et al., 2012).An inventory of NH 3 emission can serve as a baseline toward tracking emission trends, developing mitigation strategies, and assessing progress.Besides, it is necessary to provide detailed inventories combined with spatial mapping of emissions as inputs to atmospheric transport models.It has been estimated that the global NH 3 emission was about 54 Tg in 1990, 70 % of which was related to food production (Olivier et al., 1998;Pinder et al., 2007).There are currently over 10 national emission inventories of NH 3 in China (Sun and Wang, 1997;Wang et al., 1997Wang et al., , 2009;;Klimont et al., 2001;Streets et al., 2003;FRCGC, 2007;Dong et al., 2010;Cao et al., 2011;B. Gu et al., 2012;Huang et al., 2012;Li and Li, 2012;Cui et al., 2013), which providing strong evidence that China has experienced a dramatic increase of NH 3 since the late 1970s (Fig. S1).Several regions with dense population such as the North China Plain (Zhang et al., 2010;Zhao et al., 2012), the Yangtze River Delta (Fu, 2009;C. Huang et al., 2011) and the Pearl River Delta (Yin et al., 2012;Zheng et al., 2012) are the NH 3 hotspots.However, previous studies were mainly focused on the agricultural sector with large scales, such as global, Asian, national and regional inventories.As few studies involved transportation, waste disposal, human breath and sweat, etc. (Cao et al., 2011;Huang et al., 2012;Zheng et al., 2012), an inventory of city-scale NH 3 emissions covering all non-agricultural sources is clearly missing.As a consequence, past endeavours have failed to adequately reflect the overall NH 3 emissions status for individual cities, and failed to identify all the sources and activities that are responsible for NH 3 emissions, subsequently hindering target setting for future management or abatement.Introduction

Conclusions References
Tables Figures

Back Close
Full In this paper, I develop a comprehensive non-agricultural NH 3 inventory for 113 Chinese cities based on statistical data in the year 2010.The emission sources included seven main categories, with each main category including several subcategories (Table 1).The activity data is based on province-or city-specific statistical data sets, with the exception of the population of pets, which was deduced from urban residents according to a constant proportion.The emission factors (EFs) were derived from a wide range of literature, some of which have been revised to be more representative of the situation in China.

Domain of the study
The Ministry of Environmental Protection of China (MEP) selected 113 key "cities" for environmental protection since 2003, and their daily concentrations of PM 10 , SO 2 , and NO 2 are published on the MEP official website (http://datacenter.mep.gov.cn/).The 113 big cities are selected in terms of both population and economy, and the administrative area of these cities is at or above the prefecture level.In addition to their administrative centres in the urban districts, much of the administrative areas (e.g., counties) are rural.Therefore, it is not appropriate to select the entire administrative area since the assessment of non-agricultural activities is merely on the urban area.In this sense, the municipal district areas of the 113 cities were chosen as the domain of current study (Fig. 1).In 2010, the 113 cities totally accounted for 2.6 % (250 600 km 2 ), 17.4 % (238 million), 39.8 % (15 827 billion RMB) of China's land area, population and GDP, respectively.Table S1 presents detailed information regarding the social and economic index for each city.Introduction

Conclusions References
Tables Figures

Back Close
Full

Emission inventory estimate
An emission inventory is generally calculated as a product of activity data and its corresponding emission factor.For the purposes of this study, the term "non-agricultural sources" excludes emissions from the agricultural sector (e.g., fertilizer application, livestock operations), as well as natural sources (e.g., soils, wild animal populations).
The primary census dataset was mostly obtained from several official yearbooks, which is currently the most authoritative data source in China.The main categories, subcategories and their representative EFs are listed in Table 1.

Traffic
The over-reduction of NO x by three-way catalytic converters (TWCs) in vehicles has been shown to produce NH 3 in addition to N 2 and water (Cadle et al., 1982;Fraser and Cass, 1998;Kean et al., 2000), and these are fitted to all newly manufactured light-duty vehicles in China (Huo et al., 2009).In 2009, China has surged past the United States to become the world's largest automobile market -including buses, trucks and the small commercial vans that powered much of previous year's growthrose 46.2 % to 13.6 million units.The NH 3 emissions from vehicles for each city were calculated as the product of the number of operating vehicles, mileage and ammonia emissions per kilometre for each vehicle type.The original data on the provincial population of vehicles in 2010 were obtained from China Automotive Industry Yearbook 2011 (http://tongji.cnki.net/kns55/Navi/YearBook.aspx?id=N2011090109&floor=1).
In this work, these vehicles were divided into five different types, and the annual total mileages were based on previous work (Che et al., 2009).The EFs were quoted from the Emission Inventory Improvement Program (EIIP) database (Roe et al., 2004), which have been previously validated in the context of China (Huang et al., 2012;Zheng et al., 2012;Chao et al., 2014).Given that the important role of traffic to urban ammonia emissions, data like the composition of vehicle fleet, annual total mileages, and oil usages were also verified through various Introduction

Conclusions References
Tables Figures

Back Close
Full first-hand investigations.Although NH 3 tailpipe emissions have been subject to a good deal of attention, the largest unknown is the ammonia emission factor for future catalyst systems.Recent technology has tried to reduce the effect (e.g., Tables 3-16 in http://www.eea.europa.eu/publications/emep-eea-guidebook-2013/part-b-sectoral-guidance-chapters/1-energy/1-a-combustion/1-a-3-b-road-transport).
Thus, it is possible that ammonia emissions would decrease even further for new catalysts.

Waste treatment
Although the municipal waste treatment such as sewage treatment, landfill, compost and incineration of solid waste are known to emit NH 3 , the EFs of them are highly uncertain.For example, the NH 3 EFs of wastewater treatment used in existing inventories vary widely, ranging over several orders of magnitude (Table S3).
Using the relative high EF (1.93 g m −3 ), two research teams provided independent emissions estimates for wastewater treatment in their recent inventories of NH 3 emissions for the Pearl River Delta region (Zheng et al., 2012) and the entire China (Huang et al., 2012), respectively.An intensive investigation of NH 3 EFs for municipal wastewater treatment in China has been conducted in a more recent study, and this study adopted their result (0.28 g NH 3 m −3 ) in this study (Y.Gu et al., 2012).But then again, one thing should be noted is that using the highest factor to calculate emissions from wastewater treatment still produced an insignificant amount of NH 3 emissions comparing to agricultural sources.

Humans
The proportion of Chinese living in urban areas is close to 50 % mark in 2010 (49.95 %), yet this story overlooks China's special Hukou or household registration system, which may affect the real population living in urban/rural areas.This system divides the population into two types, agricultural and non-agricultural, under which peasants are strin-Introduction

Conclusions References
Tables Figures

Back Close
Full gently barred from acquiring an urban registration even though they are allowed to go and work in the city.In 2010, nearly 460 million out of 666 million Chinese urban dwellers had urban registration status (Chan, 2012).The remaining 206 million "floating population", mostly migrant workers (130 million) who seek opportunities in manufacturing centres such as Guangdong and Shanghai, are far less likely to receive pensions, basic health care, and unemployment insurance (Chan, 2012).In brief, there are three different types of population in urban China: resident population, registered population and migrant population.
Human metabolic processes such as respiration, perspiration and excretion can emit NH 3 directly (Lee and Dollard, 1994).In rural China, human excrement is an important ammonia emitter.However, this source was not considered in the current study due to the popularity of flush toilets in urban areas, which means that the resident population waste was discharged into the sewage system instead of being returned to the soil.However, it is noted that the feces and urine in infant nappies do not enter the sewage system and need to be taken into account.The 2010 infant population is the product of the registered population and the birth rate in this year.I calculated human NH 3 emissions of human sweat and breathe by multiplying the resident population and the individual EF.The infant population is the product of the registered population and the birth rate.The NH 3 EFs from infant nappies, human sweat and breath were taken from Sutton et al. (2000).
Cigarette smoking has been proved as a minor source of NH 3 .China continues to be the largest producer and consumer of tobacco worldwide.In 2010, an estimated 28.1 % of adults (301 million people with age ≥ 15 years) in China, 26.1 % in urban areas (refers to the registered population here) and 29.8 % in rural areas (refers to the migrant population here) (Li et al., 2011).Of all current smokers, 85.6 % smoked daily.Smokers in China consumed an average of 14.2 cigarettes per day, which is comparable with that in the UK (16 and 14 cigarettes for men and women in 2005, respectively) (Li et al., 2011).The EF for smoking was also taken from Sutton et al. (2000).Introduction

Conclusions References
Tables Figures

Back Close
Full

Fuel combustion
The USEPA developed the Emissions Inventory Improvement Program (EIIP) guidance for estimating NH 3 emissions from industrial sources, combustion sources, and miscellaneous sources (Roe et al., 2004).Zheng et al. (2012) introduced EIIP in developing a NH 3 emissions inventory including domestic, power plant, and industrial fuel combustion sources in the Pearl River Delta region.Ruling out the traffic fuel consumption, the current inventory of fuel combustion sources is in accordance with the work done by Zheng et al. (2012).

Urban land cover
Little information is available on NH 3 emissions from urban green land.However, green land, especially artificial grassland in urban areas needs nitrogen fertilizer to sustain it.In China, according to a report, an average of 200 kg N ha −1 yr −1 been applied in urban grassland (Zhang, 2002).In addition, a previous study recommended a volatilization rate of 2.5 % NH 3 -N of applied N (Sutton et al., 2000).When combined with the 2.5 % NH 3 -N loss and 200 kg N for grassland fertilizer use, an annual NH 3 EF of 6.1 kg NH 3 ha −1 yr −1 was obtained for China's urban grassland.The data of urban grassland were from China's City Yearbook 2011 (http://tongji.cnki.net/kns55/Navi/YearBook.aspx?id=N2012020070&floor=1).
Although still recognized as a luxury sport by most Chinese people, at present, there is a growing popularity for playing golf in China.The turf grass of golf course typically needs 200-400 kg N ha −1 yr −1 as N fertilizer to achieve high performance (Wong et al., 1998(Wong et al., , 2002;;Zhang, 2002).In 2010 alone, there were 60 new golf facilities opened, expanding the total number of the golf facilities in the country to 395 (equivalent to 490 18-hole courses) (Wang, 2011).According to a survey, every 18-hole equivalent golf course in China had 56.8 ha turf grass on average to be maintenance; this was 43.3 % larger than that of the US (Wang, 2011).Given that the turf grass data for each golf Introduction

Conclusions References
Tables Figures

Back Close
Full course is not available, this work does not consider the contribution of golf grassland to NH 3 emissions.

Pets
Keeping pets has become increasingly popular in China, especially in the cities where a growing number of people live alone and find it comforting to have a pet at home.
As the two most popular pets in China, over 200 million dogs and 100 million cats were raised in 2010 (Yang, 2010).In recent years, there have been problems with unscooped poop and prevalent rabies, so municipal governments in Beijing and Shanghai have therefore proposed a one-dog policy limiting families to one canine per household.Officials revealed that as of 2010 there were about 0.92 and 0.74 million dogs registered in Beijing and Shanghai, respectively, and another 0.5 and 0.6 million dogs that were unlicensed.Unlike dogs, cats do not require a license and their number in cities is currently unavailable.By one estimate there are 0.5 to 0.6 million cats in Beijing (Pets in China, 2011).Information on the population of pets in other cities is extremely sparse in China; however, pet statistics in Beijing and Shanghai, to a large extent, reflect the situation of other cities concerning ownership of pets.Emissions from pets were estimated assuming that for every twenty-four city registered residents in China three dogs (1/8) and one cat (1/24) were owned (J.-F.Zhou, personal communication, a pet expert working at Gerson Lehrman Group, 2013).The NH 3 EFs of pets were derived from Sutton et al. (2000).

Domestic activity
Technically, the domestic sources include pet animal waste, human breath and sweat.
In this work, the term "domestic sources" here refers merely to the use of household products and non-agricultural fertilizers.The EFs of these sources are taken from a recent report by Roe et al. (2004).Much of these data reported by Roe et al. came from Introduction

Conclusions References
Tables Figures

Back Close
Full the studies in European and the US However, these EFs shown in Table 1 are thought to be reasonable for application in China until additional updated data are available.

Non-agricultural NH 3 emission inventory in 2010
Using the methodology described above and the data collected, I evaluated the sourcebased non-agricultural NH 3 emissions and their corresponding contributions (Fig. 2).
Taking the 113 cities as a whole, the overall non-agricultural NH 3 emissions in 2010 is estimated at 210 478 Mg, which accounts for about 5 % of the annual agricultural NH 3 emissions in China.A nationwide inventory on the non-agricultural NH 3 emission for all sources/cities is beyond the scope of this paper, therefore, this share is only one-third as much as the UK (15 %), and also lower than that in the EU27 (7 %) (Reis et al., 2009) and the USA (9 %) (Aneja et al., 2001).In the case of emission sources, traffic sources were the highest-profile contributor, accounting for about 32.2 % (67 671 Mg).Fuel combustion, waste disposal, pets, green land, human and household products were responsible for 26.7 % (56 275 Mg), 21 % (44 289 Mg), 11.1 % (23 355 Mg), 3.6 % (7509 Mg), 3.5 % (7312 Mg) and 1.9 % (4069 Mg) of the total non-agricultural NH 3 emissions, respectively (Fig. 2a).It is evident that nearly 60 % of NH 3 emissions originated from traffic and fuel combustion, which can be explained by the large population of vehicles equipped with catalytic converters as well as high energy consumption density in urban China.A more obvious example might be the USA, where on-road traffic is second to livestock manure management and application of chemical fertilizers, comprising approximately 7 % of the national NH 3 emission inventory (Burgard et al., 2006;Reis et al., 2009).Without involving NH 3 in China's auto emission standards, it is expected that the sharing of vehicular NH 3 emissions in the nation's NH 3 emission inventory will be increased steadily.In addition, as the first initial assessment of pets and green land in China, the result indicates that ignoring their contribution of Introduction

Conclusions References
Tables Figures

Back Close
Full  2012) excluded the contribution of fuel combustion, waste disposal, pets, humans, green land and household products, which are collectively responsible for 46.9 % of the total urban NH 3 emissions in current study.
A full and detailed inter-comparison of the non-agricultural NH 3 emission inventory on sectoral level in different cities is presented in Fig. 2b and Table 2.As illustrated in Fig. 2, the overall amount and the contributing proportion of the seven categories for each city show quite different patterns, which reflect their own unique socioeconomic profile.Despite the large variations of the contributing proportion from one city to another, traffic, fuel combustion and waste disposal are consistently the majority of the 47.3 % to 93.1 % of the non-agricultural NH 3 emissions in most of these cities.It is estimated that the average NH 3 footprint, i.e. the NH 3 emission intensity, from all the municipal districts of the 113 cities reached 0.84 Mg km −2 yr −1 .This indicator to date has also been regarded in the previous studies, and is summarized in Table 1.In 2010, 67 and 8 out of the 113 cities exceeded the lower and upper limit of NH 3 emission intensity estimate (0.7-2.3 Mg km −2 yr −1 ) from Manchester city centre, respectively (Table S2).Domestically, estimates of the non-agricultural NH 3 emission intensity in this study are in line with those reported in Beijing, Shanghai and Nanjing (the provincial capital of Jiangsu), but significantly higher than that from Guangzhou (the provincial capital of Guangdong).The main reason for the estimated gap is attributed to the use of a different study domain in the two emission inventories in addition to the different activity data, emission factors and base years used.Although the non-agricultural sources covered in this study are estimated to be a small contributor to the national annual inventories, Table 3 indicates that at a city scale, the percentage of the non-Introduction

Conclusions References
Tables Figures

Back Close
Full agricultural NH 3 emissions to the agricultural NH 3 emissions in Beijing and Shanghai could reach 27 %, 43 %, respectively.

Geographical distribution
Figure 3 displays the spatial distributions of the 113 cities' non-agricultural NH 3 emissions for different emissions sources (Fig. 3a-g) and the total emissions (Fig. 3h).
The spatial patterns of NH 3 emission intensity can be found in Fig. 4. Generally, the spatial variability for all sources is in agreement with China's socioeconomic landscape, i.e., owing to a higher urbanization and larger population density in eastern and southern China, NH 3 emissions from non-agricultural sources such as traffic, fuel combustion and others tend to be higher than those in the nation's remote northwest and southwest.Besides, Fig. 3e reveals that the bulk of urban green land emissions are contributed from the south, particularly in Guangzhou, Shanghai, Shenzhen and Chongqing, where there is a monsoon-influenced humid subtropical climate with abundant rainfall throughout the year.The favourable climate and fertile soil make south China known for its high plant diversity.In contrast, the fuel combustion from the longlived heating systems and energy-intensive industry amplifies the NH 3 emissions in north China (Fig. 3b).
The atmospheric behaviour of NH 3 characterized by short lifetime, near-source deposition, highly sensitive to meteorology and fast gas-to-particle conversion rate, highlights the need to improve NH 3 emission estimate with fine temporal and spatial resolution (Behera and Sharma, 2010).Unlike the agricultural NH 3 sources, the non-agricultural NH 3 emissions originate from a variety of stationary sources (industrial coal/oil/gas combustion, wastewater, landfill, compost and incineration), mobile sources and area sources (e.g., humans, green land, domestic fuel combustion).Data such as the location/capacity/number of stationary sources are far from well documented in China.Therefore, a gridded emissions inventory of non-agricultural NH 3 sources was not introduced in current work.The problem of data scarcity can be easily resolved for a city case study; however, it is nearly an impossible task for me to 8508 Introduction

Conclusions References
Tables Figures

Back Close
Full collect all allocation-needed data for the 113 cities. Besides, given the city scale this work used, the spatial distribution of non-agricultural NH 3 sources in any city cannot be distinguished on the map of China.Nevertheless, those results based on city-scale datasets are still valuable for chemical transport modellers who focus on a specific Chinese city.

Uncertainties of the inventory
Ammonia emission inventories have inherent uncertainties because they associated with both the activity data and the emission factors used.For this inventory, activity data were mostly from statistical information.Although some question census data in Chinese statistical yearbooks are manipulated, with a tendency for officials to overstate economic outcome.However, compared with GDP data, the officials has less intention to manipulate the items that current work used.Therefore, the manipulation was not a serious issue during the study period and the data used in this study was credible.Nevertheless, the detailed pets and vehicle activity data (number of vehicles in each category, mileage, and oil usage of vehicles) was rarely collected and reported in China's statistics, which may introduce some uncertainties.The EFs in current inventory are lack of the real physical and chemical examination.Moreover, the EFs show a substantial variation among different studies, and China has not established an adequate non-agricultural ammonia emission factors database yet.Therefore, the major uncertainties of this inventory stem from the EFs this study used.Figure 5 illustrates the lower, the best and the upper estimates for the seven non-agricultural ammonia sources of the 113 key cities, respectively.The lower and the upper estimates are derived from the lowest and the highest EFs available in the published literature (Table S3).The overall estimated emission range was 140 421-Introduction

Conclusions References
Tables Figures

Back Close
Full point of variability, the humans and domestic activity category represents one of the most important sources of uncertainty.
A better characterization of the timing of climate-dependence non-agricultural NH 3 sources, such as humans, pets, and urban land cover could be expected to reduce the uncertainty of seasonal NH 3 emissions estimates.However, these three sources contribute less than 10 % of the total non-agricultural NH 3 emissions.Moreover, it is easy to understand that the range of variation for dominant non-agricultural NH 3 sources, such as traffic and fuel combustion, is much lower than agricultural NH 3 sources (e.g., fertilizer application, livestock operations).Therefore, the seasonal differences of nonagricultural NH 3 emission inventory are not been given in current work.

Future recommendation and outlook
China is now facing the world's worst air quality in terms of PM 2.5 and its precursors (e.g., SO 2 , NO x and NH 3 ) pollution.Some control measures such as upgrading motor vehicle emission standards to the China National Standard IV or higher, switching vehicle fuel to lower sulfur levels of gasoline in urban areas had been implemented in order to achieve a significant reduction in NO x and SO 2 emissions.However, in Europe, based on a source-attribution simulation work, Megaritis et al. (2013) revealed that the reduction of NH 3 emissions was the most cost-effective control strategy for reducing PM 2.5 in both winter and summer.Besides, in the eastern United States, Pinder et al. (2007)  emissions account for around 30 % and, 16 % of total Asia, and global NH 3 emissions, respectively.Obviously, in addition to current SO 2 and NO x controls, China needs to put more scientific, technical, and legal attention on controlling NH 3 emissions.
Most experimental studies and inventory-based assessment of NH 3 emissions has been and is being given to agricultural and natural sources (Liu et al., 2013;Paulot and Jacob, 2014).Compared to dominant agricultural sources of NH 3 emissions, the nonagricultural sources covered in this study are estimated to contribute small amounts to national annual total NH 3 emissions inventory.However, a diverse range of nonagricultural sources, such as on-road traffic and waste treatment could expect to dominate in an urban domain, where agricultural sector contributions are relatively small.
Existing NH 3 emissions inventories do not include emission estimates for all nonagricultural sources included in this study, thus significantly underestimate the NH 3 emissions in urbanized areas.Here I suggest that China should improve the understanding of current NH 3 emission estimate and put more eyes on controlling nonagricultural sources of NH 3 emissions in the future.
The efficiency of NH 3 emissions in producing inorganic aerosol is highly non-linear and may vary from urban area to urban area.This work still falls short of providing an answer to how much of ammonium salts in the urban areas derives from nonagricultural vs. agricultural NH 3 emissions, and how sensitive urban PM levels is to changes in local emissions.Therefore, it could be useful to run some source-receptor matrices (data from many model simulations and each of which reduces a different source of emissions) incorporating with the improved emission estimates in the vicinity of different Chinese cities to test the efficiency factors of different ammonia sources on particulate formation.
The use of three-way catalytic converters to reduce emissions of NO is considered to be an important factor in attaining present and future vehicle emissions standards and air quality goals.However, an unwanted side effect has been an increase in NH 3 emissions.Until recently, motor vehicles were not recognized to be a significant source of NH 3 .In this paper, it is clear that the traffic sectors, especially the light-duty gasoline Introduction

Conclusions References
Tables Figures

Back Close
Full vehicles in urban China are a major source of NH 3 emissions compared to the other non-agricultural sources.Therefore, future NH 3 control efforts in urban China should prioritize the traffic sector, and this is especially true for China's top three city clusters, i.e., Beijing-Tianjin-Hebei region, the Yangtze Delta region, and the Pearl River Delta region.NH 3 generation from petrol-engined vehicles is primarily catalyzed by platinum.
The capacity for NH 3 generation can be modified by changing the catalyst or combination of catalysts used, though there are inevitably compromises to be reached given that performance has to be measured against a range of different pollutants (Handley et al., 2001).For the other sources, it is assumed that there are no reasonable options for controlling ammonia emissions from pets, human sweat and breath, cigarette smoking and nappies.Because NH 3 is not currently included as a criteria air pollutant in China and there are no government announced daily NH 3 monitoring data, which limited the ability to validate the accuracy of emission inventory.The lack of monitoring data also emphasizes the need for publicly available comprehensive information systems in order to support the environmental evaluation of non-agricultural NH 3 in urban China.Besides, some EFs derived from foreign studies would imply a significant uncertainty between different cities across China, which is challenging to validate and interpret.A recommendation is that the emission inventory for non-agricultural NH 3 in China should be reevaluated as more local emission factors from China becomes available.Med., 329, 1753Med., 329, -1759Med., 329, , 1993.Dong, W. X., Xing, J., and Wang, S. X.: Temporal and spatial distribution of anthropogenic ammonia emissions in China: 1994-2006, Environ. Sci., 31, 1457-1463, 2010 (in Chinese with English abstract).Dong, Y., Chen, C., Huang, C., Wang, H., Li, L., Dai, P., and Jia, J.: Anthropogenic emissions and distribution of ammonia over the Yangtze River Delta, Acta Scientiae Circumstantie, 29, 1611-1617, 2009 (in Chinese with English abstract).Doyle, G. J., Tuazon, E. C., Graham, R. A., Mischke, T. M., Winer, A. M., and Pitts Jr., J. N.: Simultaneous concentrations of ammonia and nitric acid in a polluted atmosphere and their equilibrium relationship to particulate ammonium nitrate, Environ.Sci. Technol., 13, 1416-1419, 1979.Dragosits, U., Dore, A., Sheppard, L., Vieno, M., Tang, Y., Follett, R., and Hatfield, J.: Sources, dispersion and fate of atmospheric ammonia, in: Nitrogen in the Environment: Sources, Problems, and Management, edited by: Hatfield, J. L. and Follett, R.  Full  Full  Full Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Screen / Esc Printer-friendly Version Interactive Discussion Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper |NH 3 emissions may lead to around a 15 % underestimate of the total non-agricultural NH 3 emissions.Totalling approximately 9.8 Tg on the base year of 2006, in which nonagricultural sources (chemical industry, waste disposal and traffic) jointly contributed over 4 %,Huang et al. (2012) provided an unparalleled high-resolution (1 km × 1 km) NH 3 emission inventory in China.However,Huang et al. ( Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | conducted a series of PMCAMx simulations and found many currently available NH 3 control technologies were cost-effective compared to SO 2 and NO x .In China, using response surface modeling technique, Wang et al. (2011) suggested that the 90 % increase of NH 3 emissions during 1990-2005 resulted in nearly 50-60 % increases of NO − in East China.Furthermore, Wang et al. (2013) used GEOS-Chem to investigate the influence of precursor emission changes on SNA aerosols concentration changes from 2000 to 2015, finding that the benefit of SO 2 reduction would be completely offset if NH 3 emissions were allowed to keep their recent growth rate over China.China's national annual total NH 3 Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | , I presented a comprehensive non-agricultural NH 3 emission inventory for 113 Chinese cities.This work illustrates the possible under-estimation of NH 3 in some inventories, and can be expected to initiate wider discussion on methodologies for estimation of non-agricultural emissions.The source-based NH 3 emissions details will help guide development strategies that are economically and environmentally optimal, Discussion Paper | Discussion Paper | Discussion Paper | Dockery, D. W.,Pope, C. A., Xu, X., Spengler, J. D., Ware, J. H., Fay, M. E., Ferris, B. G., and  Speizer, F. E.:  An association between air pollution and mortality in six US cities, New Engl.J.

Table 2 .
Summary of non-agricultural ammonia estimates (in Mg) by city in 2010 * .