Environmental impacts of nitrogen emissions in China and the role of policies in emission reduction

Abstract

Atmospheric reactive nitrogen (N_r) has been a cause of serious environmental pollution in China. Historically, China used too little N_r in its agriculture to feed its population. However, with the rapid increase in N fertilizer use for food production and fossil fuel consumption for energy supply over the last four decades, increasing gaseous N_r species (e.g. NH₃ and NO_x) have been emitted to the atmosphere and then deposited as wet and dry deposition, with adverse impacts on air, water and soil quality as well as plant biodiversity and human health. This paper reviews the issues associated with this in a holistic way. The emissions, deposition, impacts, actions and regulations for the mitigation of atmospheric N_r are discussed systematically. Both NH₃ and NO_x make major contributions to environmental pollution but especially to the formation of secondary fine particulate matter (PM_2.5), which impacts human health and light scattering (haze). In addition, atmospheric deposition of NH₃ and NO_x causes adverse impacts on terrestrial and aquatic ecosystems due to acidification and eutrophication. Regulations and practices introduced by China that meet the urgent need to reduce N_r emissions are explained and resulting effects on emissions are discussed. Recommendations for improving future N management for achieving ‘win-win’ outcomes for Chinese agricultural production and food supply, and human and environmental health, are described.

This article is part of a discussion meeting issue ‘Air quality, past present and future’.

1. Introduction

The Haber–Bosch process, whereby nitrogen (N) from atmospheric N₂ is made biologically available, has played an essential role in feeding an increasing global population [1]. However, the wide use of synthetic N fertilizer (together with fossil fuel combustion) has also caused a number of environmental pollution problems worldwide, especially in China [2–4]. China has experienced a transition from a shortage of N for agriculture (1950s) to a closed N input–output balance (1980s) but then to large N surplus (2000s and 2010s) over the last six decades [5]. This has been accompanied by the rapid economic development since the 1980s [6]. China's environmental problems induced by excess reactive N (N_r), such as soil acidification [3,7], enhanced N deposition [8], biodiversity loss [9] and aquatic eutrophication [10] have been reported widely since the 2000s. The excess N_r inputs and subsequent losses from recipient systems have caused soil degradation, a decline in water and air quality, and harmful human health effects, as well as changing ecosystem functions and degrading ecosystem services in China [11]. The challenge in reducing N_r emissions to the environment and so limiting N_r pollution are growing not only in China [10] but around the world [12,13]. As a result, sustainable N management is essential to further increase crop production for a growing global population [14–16], while maintaining ecosystem functions and services [17]. Currently, China is at a historic turning-point, needing to change its economic model from high resource inputs and high production to a double high (high efficiency and high production) model [18]. To achieve this, more precise N management practices and stricter environmental regulations for both producers and consumers are required urgently.

This review is divided into three parts: (1) an overview of the environmental problems caused by N_r emissions in relation to air pollution, including acid rain, haze and ozone pollution; (2) an overview of atmospheric N_r deposition and its ecological impacts, including soil acidification, plant biodiversity loss and aquatic eutrophication; (3) regulations and actions for mitigating atmospheric N_r emissions and improving air quality. Figure 1 summarizes the three key parts of the review (N_r emissions and air pollution; N_r deposition and impact; national N_r emission regulations) as well as the relationships between the parts.

Figure 1.

Nitrogen emissions, deposition, impacts and regulation in China. (Online version in colour.)

2. Atmospheric reactive N emissions and air pollution

(a). Atmospheric reactive N emissions

Reactive N compounds are emitted to the atmosphere mainly in the form of ammonia (NH₃) and nitrogen oxides (NO_x). NH₃ is the main alkaline gas species. It can neutralize sulfuric acid (H₂SO₄, a product of SO₂ oxidation) and nitric acid (HNO₃, a product of NO_x oxidation) to form ammoniated sulfate and ammonium nitrate aerosols in the air. These are secondary inorganic aerosols (SIA), reflecting their formation by atmospheric chemistry, and are a significant component of fine particulate matter (PM) pollution, accounting for 40–57% of the PM_2.5 (particles smaller than 2.5 µm) mass in eastern China [19]. NO_x is also a precursor of tropospheric ozone, reacting with hydrocarbons to form ozone in the presence of sunlight. As one of the main atmospheric pollutants, NO_x plays a crucial role in the formation of acid rain and other regional environmental problems. Production of O₃ and SIA is very dependent on the abundance of NO_x [20]. Although nitrous oxide is an important greenhouse gas, it is chemically inert in the troposphere and has a minor effect on air pollution, and so is not discussed further in this section.

Figure 2 shows the spatial distributions of annual total NH₃ and NO_x emissions in China averaged over 2008–2012 [21]. The NH₃ and NO_x emissions come mainly from the eastern parts of China, especially the North China Plain, Sichuan Basin, the Middle and Lower Reaches of Yangtze River and the Pearl River Delta (figure 2). These regions are China's economic centres for agricultural and industrial production as well as transportation. Sources of NH₃ and NO_x in China are mainly anthropogenic, but their magnitudes and variation are typically driven by different sectors of the economy. NH₃ is mainly released from the agricultural sector, i.e. from fertilizer use and livestock production, although vehicles and other sources can be important in urban areas. NO_x is mostly generated by fuel combustion by industry, transportation and power plants. China has the largest NH₃ emissions in the world due to its intensive agriculture [8]. Present-day Chinese NH₃ emissions are estimated to be in the range of 6.9–15 Tg N per year or 51–110 kg N ha⁻¹ agricultural land per year averaged over 2005–2012 [22–24]. The wide range and significant uncertainty in China's NH₃ emission estimates are largely attributed to missing agricultural statistics (activity data and emission factors), as well as a lack of NH₃ flux measurements to constrain the emission estimates. Recent developments in a nationwide N deposition monitoring network [25], the Chinese ammonia monitoring network [26] and multiple satellite retrievals [27–29] provide valuable information on Chinese NH₃ emissions and their spatial and temporal variations (figures 2 and 3) [24,30–32]. At the national scale, agricultural sources (fertilizer use and livestock manure management) are the dominant NH₃ sources, together contributing over 80% of total anthropogenic emissions [22,24]. However, urban sources such as transportation and waste disposal may be important in urban environments, as suggested by recent research using concentration ratios with nitrogen dioxide (NO₂) [33] and N isotope measurements [34,35].

Figure 2.

Annual NH₃ and NO_x emissions averaged for the years 2008–2012. The numbers are annual emission totals for China (From Zhao et al. [21], © 2017 Elsevier Ltd.). (Online version in colour.)

Figure 3.

Trends in anthropogenic emissions of NO_x (a) and NH₃ (b) over the period of 1980-2017 in China and emissions of NO_x (c) and NH₃ (d) by sectors during 2008–2016. NO_x emission data were averaged from the China Statistical Yearbook (http://www.stats.gov.cn/), EDGAR (https://edgar.jrc.ec.europa.eu/) and Zheng et al. [30]. NH₃ emission data were averaged from EDGAR, Kang et al. [31] and Zheng et al. [30]. Sectors emission data were adapted from MEIC (http://meicmodel.org/index.html). (Online version in colour.)

NO_x can either be emitted from natural or non-fossil fuel sources, such as lightning, microbial processes in soil and biomass burning. These may correspond to only 10% of anthropogenic Chinese NO_x emissions, but make important contributions to surface ozone air quality [36].

Driven by rapid economic development, urbanization and intensive energy use, NO_x emissions in China were estimated to have increased significantly within the first decade of the twenty-first century (figure 3) [20,37–39]. Recent clean air actions implemented by the Chinese government have effectively decreased the emission of NO_x (figure 3a,b), as also shown by satellite measurements [40], but emissions of NH₃ (figure 3c,d) have not been regulated, resulting in increasing attention to the need to control NH₃ to further mitigate air pollution [41,42] and reduce N deposition exceedance in many regions of China [21].

(b). Reactive N and acid rain

Wet deposition is effective in scavenging many atmospheric pollutants. The chemical characteristics of rainwater provide insight into air pollution and help us to understand the sources and transportation of atmospheric pollutants [43]. In-cloud and below-cloud scavenging results in NH₄⁺ and NO₃⁻ becoming the dominant pollutants in precipitation. Gaseous acids (e.g. HNO₃ and H₂SO₄) and bases (NH₃) are critical in determining the acidity of rainwater. According to China Ecological Environment Bulletin 2018 [44], the average pH of precipitation in China was 5.58 in 2018 with 53.8 million ha, 5.5% of the Chinese land area, receiving acid rain (rain or precipitation that contains elevated levels of hydrogen ions with pH value under 5.6, an equilibrium pH of distilled water in contact with atmospheric CO₂). Some Chinese cities (mainly distributed in southern China) suffer from acid rain: approximately 77 out of 471 Chinese cities had a frequency of acid rain events that exceeded 25%, and 39 had a frequency that exceeded 50% in 2018 [44]. Southwest China especially continues to suffer from severe acid precipitation, with the mean pH being 5.1 in Sichuan province from 2011 to 2016 [45]. The situation in eastern China was becoming worse in the early 2000s. For example, Shanghai experienced acid rain with a pH value less than 3.0 in 2005 [46]; the acidity was reduced, with an average pH of 4.96, from 2011 to 2016 [47]. In contrast to southern and eastern China, northern China has experienced alkaline precipitation. This is because the surface geology in northern China consists of loess, which is abundant in alkali and alkaline-earth metals; windblown dust makes the rain alkaline. In Gansu province, for example, the pH value of precipitation was from 6.63 to 8.10 during 2011–2014 [48]. In Beijing city, the precipitation has changed from sulfuric acid-dominated to mixed, with both sulfuric and nitric acids [49,50].

The acidity of precipitation and its chemical composition vary as a function of the local environment and anthropogenic activities. SO₄²⁻, NO₃⁻, NH₄⁺, Ca²⁺ are the dominant water-soluble ions in precipitation. In China, SO₂ emissions from industrial, domestic and energy sectors decreased from 25.5 Tg SO₂ in 2005 to 18.6TgSO₂ in 2015 and were less than 10TgSO₂ in 2018 because of the Chinese government's strict control strategies [51]. The remarkable decrease in SO₄²⁻ concentration (from 285 to 145 µeq l⁻¹) resulted in an increase in the pH of Sichuan precipitation from 5.24 in 2011 to 5.70 in 2016 [45] and the chemical composition of acid rain changed from sulfuric acid-dominated to mixed. NO₃⁻ was the second most abundant anion in precipitation, with vehicle emissions considered the dominant source [52]. From the point of view of reducing acidity, further control of NO_x emissions is essential. Anthropogenic NH₃ emissions increased by 8% from 2005 to 2010 and were predicted to increase further by 15% in 2050 (relative to 2010) [53,54]. The high concentration of NH₄⁺ in precipitation is mostly due to NH₃ emissions from agriculture [55]. As the only alkaline gas in air, NH₃ neutralizes precipitation acidity and so reduces the occurrence of acid rain. However, the downside is that NH₃ not only promotes the formation of PM material and aggravates haze pollution [56] but also causes soil acidification due to the transformation of NH₄⁺ to NO₃⁻ (nitrification) in the soil, producing 2H⁺ ions. Liu et al. [41] found that NH₃ emission abatement mitigates PM_2.5 and N deposition but worsens acid rain effects and eutrophication in China (§3). Therefore, a comprehensive reduction of NH₃ emissions is urgently needed.

(c). Reactive N and haze pollution

Over the past decades, China's economy has been growing rapidly, but this has resulted in severe air pollution, especially haze, which is largely due to increasing resource consumption and accompanied elevated emissions of NO_x, SO₂ and NH₃ [57]. Haze pollution threatens not only the human respiratory system and the heart [58,59], but also the regional climate [60], plant photosynthesis [61] and food production by affecting solar radiation [62].

The prime air pollutant of concern during haze is fine particulate matter (PM_2.5), 40–57% of which is caused by secondary inorganic aerosols (SIA, i.e. sulfate (SO₄²⁻), nitrate (NO₃⁻) and ammonium (NH₄⁺)) in China [19,63]. Both the precursors of acid (e.g. SO₂ and NO_x) and alkaline (e.g. NH₃) pollutants are crucial to the formation of SIA through acid-base neutralization reactions [64]. These reactions increase the size and water solubility of particles, and aerosol growth becomes spontaneous once the bonded particles exceed the threshold size, i.e. the nucleation barrier [65]. Hence, emission control of these important gaseous precursors (SO₂, NO_x and NH₃) has been a critical factor in attempting to control haze pollution. Since 2010 China has contributed more than one-fifth of global anthropogenic emissions of SO₂ (∼30%), NO_x (∼24%) and NH₃ (∼20%) [66].

Due to strict emission control measures to improve air quality, national emissions of SO₂ and NO_x decreased by 62% and 17%, respectively, between 2010 and 2017 [67]. The turning-point in China's NO_x emissions occurred in 2012 (figure 3a) and major contributors (for emission reduction) are power plants and industry (figure 3b). By contrast, because no official control measures were implemented for NH₃ in China until 2017 (for details see §4), NH₃ emissions have remained constant at approximately 10.0 Tg NH₃ or approximately 8.2 Tg N yr⁻¹ and showed a relatively smaller decline over recent years because of the major contribution from the agricultural sector (figure 3c,d and related references therein). Although the absolute mass concentrations (µg m⁻³) of PM_2.5 and SIA significantly decreased because of the decreasing emissions of SO₂ and NO_x, the unbalanced decline in emissions of SO₂ (large) and NO_x+NH₃ (small) increased the fraction of N-containing inorganic compounds (NO₃⁻ and NH₄⁺) in PM_2.5 from 2013 [24,68]. The contribution of aerosol nitrate to PM_2.5 increased during 2006–2015 over eastern China, together with a decrease in SO₄²⁻ [69]. NO₃⁻, rather than SO₄²⁻, dominated haze formation in winter [70].

Zhang et al. [67] showed that regional haze may be effectively minimized by controlling NH₃ emissions, but there are still large uncertainties about the effectiveness of NH₃ reduction without simultaneously reducing emissions of SO₂ and NO_x. In future, the simultaneous reduction in emissions of NO_x and NH₃ together with SO₂ is key to limiting deadly haze pollution in China [71].

(d). Reactive N and O₃ pollution

Ground-level ozone (O₃) is an important secondary air pollutant, which is controlled by its precursors, such as NO_x and VOC_s [72]. The photolysis of NO₂ to give NO and an O atom results in the O atom reacting with molecular oxygen (O₂) to form O₃; it can also react with NO to regenerate NO₂ [73]. In China, the large emissions of NO_x from rapid industrial and urban development have caused rapidly increasing ground-level O₃ concentrations since 1990 [74–76]. Ozone pollution is mainly concentrated in summer, which also coincides with the growing season of plants.

Li et al. [77] analysed the spatial‒temporal changes in ground-level O₃ concentrations in China, using data from 187 cities, from January 2014 to November 2016. They showed that the average O₃ concentration had a large spatial variation, from 50.6 ppb to 64.1 ppb, and increased from 46.1 ± 8.8 ppb in 2014 to 51.9 ± 7.8 ppb in 2016. Ozone pollution is more serious in economically developed areas, such as Jing-Jin-Ji (Beijing-Tianjin-Hebei), the Yangtze River Delta and the Pearl River Delta [77]. Zhu & Liao [78] used the high-resolution nested grid version of the GEOS-Chem model to simulate changes in ground-level O₃ concentrations from 2000 to 2050 under emission pathways for IPCC scenarios RCP2.6, RCP4.5, RCP6.0 and RCP8.5. They predicted the average O₃ concentration would increase by a maximum of 6–12 ppb each year if emissions are not effectively controlled, and so will become increasingly serious.

High ground-level O₃ concentrations have adverse effects on human health and vegetation [79]. Ozone causes damage to plants after entering through leaf stomata, including visible leaf injury [80], impairment of photosynthesis [81], and reductions in growth and yield [79,82]. A recent meta-analysis of Chinese woody plants showed that elevated O₃ concentrations (116 ppb) reduced total biomass by 14% compared with the control (21 ppb) [83]. Yue et al. [84] used a coupled chemical-carbon-climate model to suggest that the current O₃ concentration in China has reduced the annual net primary productivity by about 10.1–17.8%. Feng et al. [79] quantified the adverse impact of O₃ on human health and vegetation (forests and crops) based on data from greater than 1400 monitoring stations in China in 2015, and showed that the current O₃ level led to a 0.9% increase in premature mortality, and reduced annual forest tree biomass growth by 11–13% and the yield of rice and wheat by 8% and 6%, respectively. The total O₃-related cost to health, food production and the environment reached 761.7 billion US$, equivalent to 7% of China's Gross Domestic Product in 2015.

3. Atmospheric N deposition and ecological impacts

(a). Atmospheric N deposition

As an important component of the N cycle, atmospheric N deposition has attracted increasing attention in China and worldwide [85–87]. With the continuous increase in N_r emissions, China has been a global hotspot for N deposition [2,8] and so has introduced several monitoring networks. Currently, the Nationwide Nitrogen Deposition Monitoring Network (NNMDN), led by China Agricultural University, the Center for Ecosystem Research Network (CERN) and the Ammonia Monitoring Network in China (AMoN-China) led by the Chinese Academy of Sciences, are the three main N deposition monitoring networks in China [26,88,89].

Nitrogen deposition exhibits clear spatial variability in China due to the large spatial heterogeneity of anthropogenic N_r emissions [90] and of factors such as land use type [91] and the amount of precipitation [90]. Based on field measurements at 43 sites of the NNDMN, Xu et al. [25] reported that the total N deposition (wet/bulk and dry) averaged 39.9 kg N ha⁻¹yr⁻¹ in China and is ranked by land use as urban > rural > background sites or by region as north China > southeast China > southwest China > northeast China > northwest China > Tibetan Plateau. Except for significantly higher values in northern rural sites, annual dry N deposition (15.8–31.7 kg N ha⁻¹ yr⁻¹) is comparable at urban and background sites in northern and southern regions [92]. Modelling and satellite observations have been used to elucidate the spatial characteristics of N deposition in China [21,85,93]. For example, total N deposition simulated by Zhao et al. [21] was generally less than 10 kg N ha⁻¹ yr⁻¹ in western China, and 15–50 kg N ha⁻¹yr⁻¹ in eastern China. Using a remote sensing model, which is based on area data instead of that derived from monitoring sites, Yu et al. [85] showed that: (1) wet reduced (NH_x) and oxidized (NO_y) N deposition were both a maximum in the north, followed by east and central China; (2) the highest NH_x and NO_y dry deposition occurred in north China, and dry N deposition decreased from north China to other regions; (3) the spatial pattern of total N deposition (19.6 kg N ha⁻¹yr⁻¹ on average) was similar to those of total NH_x and NO_y deposition (figure 4). They also found that 62–99% of the spatio-temporal variation in total N deposition in China was mainly due to N_r emissions from energy consumption, N fertilizer use and livestock production [85], whereas for wet deposition, the amount of precipitation and N fertilizer use can explain 80–91% of its variability [89]. Both figures 3 and 4 showed similar spatial N_r emission and deposition patterns in China. Although estimates of total N deposition were somewhat different in the aforementioned studies, they all agreed that dry N deposition is as important as wet/bulk N deposition at the national scale [20,25,85]. The difference of total N deposition (39.9 versus 19.6 kg N ha⁻¹ yr⁻¹) between that of Xu et al. [25] compared to that of Yu et al. [85] reflects the importance of considering spatial variation when evaluating country-level or regional N deposition: Xu et al. [25] calculated N deposition only by averaging data from all 43 sites, while Yu et al. [85] estimated N deposition by Kriging interpolation, based on area data from remote sensing models.

Figure 4.

Spatial patterns of atmospheric total (wet plus dry) deposition of various N species (NH_x and NO_y) over China, averaged for 2011–2015 (From Yu et al. [85], © 2019, Springer Nature). (Online version in colour.)

As for temporal changes in N deposition, Zhang et al. [94] first reported a substantial increase in wet N deposition in the North China Plain from the 1980s to the 2000s. After summarizing historic data on bulk N deposition in the whole of China, Liu et al. [8] found that China's N deposition showed a significant increase from 1980 to 2010, with an annual increase of 0.41 kg N ha⁻¹. Yu et al. [85] showed that total N deposition in China changed from an initial rapid increase to stability between 1980 and 2015, in which wet N deposition reached a maximum in 2001–2005 and declined thereafter, in contrast to a continuous increase in dry deposition (figure 5). The stabilization of total N deposition was mainly caused by a gradual decline in wet NH₄⁺ deposition. Total N deposition was dominated by wet N deposition between the 1980s and 2000s, with a shift to approximately equal wet and dry N deposition from 2011 to 2015 in all regions except northwest, central and south China. NH_x deposition dominated dry, wet and total N deposition between 1980 and 2015, but its contribution gradually decreased due to increasing NO₃⁻ deposition (figure 5). The recent stabilization and even decrease of atmospheric N deposition [85,95] reflected successful N_r emission control measures especially after 2013 [67].

Figure 5.

Temporal trends of wet (a), dry (b) and total (c) N deposition across China. Error bars denote standard errors (s.e., the variation among the 31 provinces in mainland China) (From Yu et al. [85], © 2019, Springer Nature). (Online version in colour.)

Significant progress in determining the spatio-temporal trend in N deposition in China has been achieved, but some key points should be addressed in future studies to improve the accuracy of results [96]. Taking NH₃ as an example, all the reported deposition fluxes may be subjected to some uncertainties owing to the following major causes: (1) large uncertainties still exist in China's NH₃ emission inventory due to a lack of reliable data about the local agriculture and other activities and emission factors, which limits accurate modelling of temporal and geographical distribution of atmospheric NH₃ concentrations; (2) parametrization for vertical bi-directional exchange of NH₃ as emissions and deposition is not included in chemistry and transport models and satellite-based estimates, which can result in an overestimation of NH₃ deposition, especially in agricultural areas.

(b). Impact on soil acidification

Soil acidification is a natural and gradual process in soil development, but is dramatically enhanced by human activity. Elevated acid deposition, induced by growing NO_x and SO₂ emissions derived from industrial activities and increasing NH₃ emission due to agricultural N fertilization, has caused soil acidification in forests and/or grasslands especially in Europe [97], United States [98] and China [99,100]. Enhanced soil acidification has not only become a serious threat to ecosystem functioning and services in grasslands and forests [7,101] but also a threat on agricultural systems [102].

Increased N inputs to agriculture, especially of chemical N fertilizer, have enhanced soil acidification in croplands [3]. N input via deposition comprises approximately 8% of the N fertilizer application (farmers' practice) [103,104], making deposition or re-deposition of NH₃ a contributor to N-induced soil acidification of croplands, especially for those regions with high N deposition [93,105].

Chronic N deposition has significantly increased net aboveground primary productivity and plant uptake, but also leaching losses (with excess nitrate) of base cations from the soil, especially in agricultural ecosystems [103,104]. Soil acidification reduces the availability of nutrients (e.g. base cations, phosphate, molybdenum and boron) and increases the concentrations of toxic elements (e.g. aluminium, cadmium and lead), leading to restricted plant and soil biota growth due to nutrient deficiency and metal toxicity [106]. Liming is a common practice for alleviating soil acidification, but it is not a panacea [102].

Chinese croplands are facing an increasing risk of aluminium toxicity due to enhanced soil acidification [107], with accompanying crop yield losses unless soil acidification is mitigated [108]. Although ‘4R principles’ (applying fertilizer as the Right product in the Right amount at the Right time and in the Right place), i.e. balanced N fertilization, combined with manure and straw recycling, can effectively reduce soil acidification [108–110], N deposition, as an important N input to croplands, must be included in calculating N fertilizer recommendations and management. Meanwhile, non-agricultural ecosystems (forests and grasslands) in China are still suffering from high N deposition and related soil acidification and reduced biodiversity, urgently requiring more vigorous controls on N pollution [8,11].

(c). Impact on plant biodiversity

The impact of N deposition on plant diversity has gained increasing attention because of potential consequences for ecosystem services. Elevated N deposition is considered to be one of the most important drivers of biodiversity loss [111]. A global assessment has suggested that N deposition can cause a shift in community species composition and/or a loss of plant species in terrestrial ecosystems [112]. Ecosystems in China, especially those in eastern and southern regions, have been subjected to long-term, high-level N deposition [8,85]. Comparing N deposition maps with critical load estimates, Zhao et al. [21] suggested that about 15% of the land cover in China experiences critical load exceedances, implying a high risk of negative ecological effects (e.g. biodiversity loss). However, there are few reports of observed losses of plant biodiversity in response to N deposition due to a lack of well-designed monitoring systems of long-term changes in plant biodiversity [113].

Current understanding of N deposition impacts on plant biodiversity come mainly from N addition experiments [88,113]. Generally, increasing N deposition decreases the dominance of N-sensitive species, while benefiting nitrophiles due to their ability to use available N. For example, 9 years of N additions (20, 50 and 100 kg N ha⁻¹ yr⁻¹) significantly increased coverage of graminoids and decreased that of mosses and shrubs in a natural boreal forest in northeast China [114,115]. In an old-growth subtropical forest in southern China, N additions of greater than 100 kg N ha⁻¹ yr⁻¹ decreased the abundance of understory seedlings, ferns and mosses, but did not affect canopy trees and shrubs [116]. A recent meta-analysis of experimental results in China concluded that N addition negatively affected plant biodiversity in grasslands and forest understory communities; the effects varied by climatic zone, N addition level and duration [117]. However, there are relatively few reports from eastern and southern China [117], where high-level N deposition has occurred for decades. In these regions, N addition experiments likely underestimate the negative effects because high-level N deposition may have already caused a shift in plant species composition towards that better adapted to high N availability before the experiments began.

(d). Impact on aquatic eutrophication

Elevated atmospheric N deposition has increasingly impacted aquatic ecosystems particularly their N budget and phytoplankton structure [118,119]. It has resulted in eutrophication and enhanced phytoplankton biomass in unproductive lakes in Europe and North America [120]. With the increase in N deposition, the stoichiometric N/phosphorus (P) ratio increased and phytoplankton diversity was reduced by favouring those few species with the ability to use P more effectively [118]. In recent years, much research has explored the impacts of N deposition on aquatic eutrophication across China. Nitrogen deposition to Lake Taihu, Lake Dongting and Lake Dianchi were reported to be as high as 50‒80 kg N ha⁻¹ yr⁻¹ [121–123]. This has contributed a large proportion (15‒48%) of the total N load into these lakes, increased the N concentration in the lakes and induced eutrophication. Research in Lake Dianchi [122] also found that the when toxic blooms of the non-N₂-fixing cyanobacteria Microcystis spp. were initiated and proliferated, the contribution of N deposition to the total N load was as high as 27–48%, which indicates that N deposition may be a neglected contributor to the blooms besides agricultural runoff and point sources. High N deposition was also reported to the coastal waters around China. A total N deposition of 20‒50 kg N ha⁻¹ yr⁻¹ was measured to the Yellow Sea and South China Sea [124–126]. It is estimated that an additional primary biological productivity of 1.5‒30 g C m⁻² yr⁻¹, which accounted for 0.3‒6.7% of the current productivity in the Yellow Sea, was caused by N deposition [124]. Incubation experiments on the impacts of wet N deposition on phytoplankton community structure have shown that, when contributing 5‒10% of the rainwater addition (filtered or not), the total chlorophyll a concentration increased 1.6 to 1.9-fold while microphytoplankton increased 1.7 to 3.2-fold [126]. The abundance of diatoms increased and they became the dominant species, accounting for 55% of total phytoplankton abundance. Atmospheric N deposition also causes N enrichment in rivers [92,127]. A study in a subtropical catchment showed that the N concentrations in river water showed significant positive correlations with rainwater N concentrations, and N deposition contributed 21% of the total riverine N export in the whole catchment [127].

4. Regulations and actions mitigating atmospheric N_r emissions

(a). Regulations mitigating NH₃ emissions in agriculture

Mitigating NH₃ emissions in agriculture was added to the updated version of the Clean Air Act (CAA) in 2018, before which the CAA mainly focused on pollutant reductions from industrial and transportation sources in China (figure 6). Other national regulations, such as the Zero Increase in Chemical Fertilizer Use after 2020 [18], have been introduced by the Chinese government. Farmers are required to change crop rotations, use available manure as a substitute for chemical fertilizers, etc., and these have resulted in some increases in nitrogen use efficiency (NUE) while stabilizing chemical fertilizer use. However, the overuse of chemical fertilizer in China is estimated to be still greater than 30%, indicating that a substantial reduction in fertilizer use is still needed [16,128].

Figure 6.

Implementation steps of China's atmospheric environmental protection policies and measures particularly related to atmospheric N_r emission control and air quality improvement (Modified from Wen et al. [95]). (Online version in colour.)

Compared with crop production, the reduction of NH₃ emission from livestock production is more important because the most serious air pollution occurs mainly in winter and spring when agricultural NH₃ emissions are dominated by livestock production [42,129]. Noting the importance of NH₃ to PM_2.5 pollution and haze, explained above, effective reductions in haze episodes can be anticipated given that successful mitigation of NH₃ emission from livestock in winter and spring has already been achieved. However, due to the incomplete assessment of manure management practices, many current treatments such as air-drying, intended to reduce water pollution, may benefit manure recycling and reduce N losses to water but cause unintended increases in NH₃ emissions—‘Pollution Swapping’ [130]. This creates imperative and critical requirements for establishing guidelines for environmentally friendly practices for livestock production, including standards and protocols for monitoring NH₃ emissions from livestock production. In addition, the decoupling of livestock production from crop production increases the transportation costs and reduces manure recycling, causing higher NH₃ emissions from livestock systems [131]. Policies that can rebuild the link between livestock and cropping are crucial.

Although not the major source of NH₃, optimizing fertilizer use in croplands and so reducing NH₃ emissions is still important for environmental health. Many measures have been developed, including ‘4R stewardship’, explained above, to manage N use in croplands [109]. However, the implementation of these measures requires appropriate farm sizes to make them practicable and reduce costs [132]. When Chinese farm sizes increase from the current average of 0.5 ha to 3–5 ha, agricultural fixed input costs for machinery and other facilities are much improved, benefitting the implementation of 4R stewardship and other measures, and China's N fertilizer use can be reduced by an estimated 30% [129]. A concomitant increase in NUE and a reduction in NH₃ emissions can be expected [133]. To increase farm size, urbanization and regulations for land transfer are critical: urbanization can move rural populations to urban areas and, at the same time, release arable land through the reclamation of land that was rural towns and villages, leading to larger farm sizes through an increase in croplands for a smaller population [134]. Land transfer can then re­allocate cropland among rural households, facilitating the growth of alternative farming models such as family, cooperative/collective and industrial farms, which normally have larger farm sizes. However, this would be unacceptable in many countries.

Other non-technical measures, such as dietary change, can also help [134]. It is estimated that, in China, over half of crop production is consumed by livestock as feedstuff, due to the dramatic increase in consumption of animal products [135]. A change to a more plant-based diet, partially reversing the recent trend in China, could produce a significant decrease in agricultural NH₃ emissions. Managing consumption can have a major impact but will be difficult to achieve.

(b). Regulating NO_x emissions from traffic and industry

NO_x emissions over China had been increasing rapidly over the past three decades, mainly driven by industrialization and urbanization, but have begun to decline in recent years (figure 3). Anthropogenic NO_x emissions were estimated to be about 3.5 Tg N in 2000 [136], increasing to 8.9 Tg N in 2012, then gradually decreasing to 6.7 Tg N in 2017 due to the implementation of clean air regulations [30]. Satellite observations of NO₂ columns have been successfully applied in a growing number of studies to improve the estimates of NO_x emissions over China [137–139].

Driven by rapid economic development, urbanization and intensive energy use, NO_x emissions in China were estimated to have increased significantly within the first decade of the twenty-first century (figure 3) [20,37–39]. Before 2010, the Chinese government took aggressive steps to improve energy efficiency and reduce emissions of primary aerosols and SO₂, but paid less attention to NO_x abatement (figure 6). The annual growth rate of NO_x emissions from 1995 to 2004 was estimated at 6.3% [20] and the annual total emissions were calculated to reach 28.1TgNO_x in 2010 [39]. This rapid growth partially offset China's efforts on SO₂ emission reduction and exacerbated acid rain in the East Asian region [140].

The national 12th Five-Year Plan issued in 2011 aimed to reduce NO_x emissions by 10% from 2010 to 2015 [141]. A series of emission control measures have been implemented in power plants, steel and cement industries, industrial boilers and transportation. These measures were further refined in the Air Pollution Prevention and Control Action Plan issued in 2013 [142]. Thermal power has been the most important sector introducing NO_x control since 2010. An updated emission standard for the sector was issued in 2011, limiting NO_x concentrations in flue gas to less than 100 mg NO_xm⁻³ [143]. Denitration technologies (e.g. selective catalytic reduction, SCR) were required and installed in 92% of China's thermal power sector in 2015 [144]. The government subsequently released the ultra-low emission policy for the sector, requiring NO_x concentrations in the flue gas of coal-fired units to be the same as those in gas-fired units, i.e. less than 50 mg NO_xm⁻³ [145]. Coal-fired units retrofitted with ultra-low emission controls totalled 810 million kilowatts, accounting for 80% of the total installed-capacity by the end of 2018 [146]. As a result, the fraction that the power sector contributed to total emissions was reduced from 32.5% in 2010 to 19.1% in 2017, as shown in figure 3. Emission controls have gradually expanded to other non-electricity generating industries including cement and steel production. For example, the ultra-low emission policy was introduced in the steel industry in April 2019 [146], limiting NO_x concentrations in sintering flue gas to less than 50 mg NO_x m⁻³. China is also paying increasing attention to emission controls in transportation, including the staged implementation of stringent emission standards on new vehicles, the retirement of old vehicles and the expansion of renewable energy fueled vehicles. These policies have proved effective. For example, progress from the China III to China IV standard (equivalent to that from the European III to European IV emission standards, respectively) was estimated to reduce the NO_x emissions from on-road vehicles by 50% between 2011 and 2015 [147]: from 2013 to 2018, China's vehicle ownership increased by 33% while NO_x emissions decreased by 14% [146]. The most recent plans have included a series of policies released in early 2019 to control emissions from diesel trucks, which account for over 70% of total vehicle emissions [148], and the nationwide implementation of China VI, the most stringent standard, from 2020 is estimated to reduce NO_x emissions by 42% relative to China V. Besides on-road vehicles, official plans were also gradually announced to limit NO_x emissions from other mobile sources such as ships [149]. With the above-mentioned regulations, China's NO_x emissions from anthropogenic sources were estimated to have declined by 36% from 2012 until the end of the Action Plan in 2017 (figure 3a). If emission controls had been frozen at the 2010 level, as a comparison, emissions would have increased by 38% [30]. To maintain the restraints on NO_x emissions in the future, a 3-year plan [150] for ‘Defending the Blue Sky’ was announced in 2018, aiming for a further 15% reduction in NO_x emission from 2015 to 2020.

(c). National actions for mitigating N_r emissions and improving air quality

In China, a number of national/provincial actions have been introduced for improving air quality through the mitigation of atmospheric N_r, as explained in previous sections [18,151]. In general, these can be divided into short-term and long-term measures. Figure 6 summarizes the major implementation steps in China's atmospheric environmental protection policies/regulations and emission mitigation measures related to atmospheric N_r emission controls and air quality improvement.

For short-term emission mitigation measures, the Chinese Government temporarily closed many factories and construction sites and controlled traffic flow during some important national celebrations such as the 2008 Beijing Summer Olympics, the 2010 Shanghai World Expo and the 2014 Beijing APEC meeting. These proved their effectiveness in improving air quality over short periods, but their weakness was very clear in the rebound observed in post-activity air pollution, which offset all previous efforts in air quality improvement [33,152,153]. Similarly, but not to be welcomed, the recent coronavirus epidemic (which was first reported in Wuhan, then extended to the whole of China and finally worldwide) abruptly stopped China's major economic activities and led to a reduction in NO_x and other air pollutant emissions of at least 36% due to the nationwide isolation needed to tackle the virus. However, PM_2.5 pollution events still occurred during the Chinese New Year holiday in many regions, such as Beijing and Shanghai, for reasons that are not yet known [154].

For long-term N_r mitigation measures, the main national actions included denitration from industry and coal-fired power plants, natural gas heating replacing coal heating in northern cities, and agricultural NH₃ mitigation [11]. These have produced continuously positive effects in reducing atmospheric N_r concentrations, N deposition and PM_2.5 pollution [95,96,151]. For example, the replacement of coal by natural gas for heating caused a greater than 75% decline in PM_2.5 concentrations in Urumqi from January 2011 (322 µg m⁻³) to January 2014 (79 µg m⁻³) [151]. Another successful example was that of Quzhou in Hebei, where improved N management associated with the now well-known ‘Science and Technology Backyard Programme’ (a technology innovation and extension platform linking universities, local government, private companies and farmers [155]) from 2009 increased grain yields and reduced fertilizer N inputs, reducing NH₃ losses [156]. As a consequence, the annual mean PM_2.5 concentration in Quzhou was 40% lower in 2015–2017 than in 2011–2014 [156,157].

5. Conclusion and recommendations

Concerns over N_r-induced environmental issues, in particular air pollution problems such as acid rain, haze and O₃ pollution, have increased dramatically in China. Ammonia and NO_x are two key N_r species inducing air, water and soil pollution and environmental degradation including reductions in plant biodiversity. This overview clearly shows China's progress in controlling N_r (especially NO_x) emissions, N deposition and related environmental issues through policies, regulations and practices introduced during a period of rapid economic growth over the last four decades.

To improve air quality (e.g. by reducing PM_2.5 pollution), the Chinese government has taken a series of actions covering policy, law/regulation and technology innovations to mitigate N_r and other pollutant emissions from the 2000s onwards and especially after 2013 [18]. As a result, N_r emissions and deposition and their ecological and environmental impacts have at least stabilized (NH₃) or even declined (NO_x), especially after the recent strict emission control regulations and measures in China since 2010 [85,95]. This represents considerable progress for China. However, the current atmospheric N_r emissions and deposition are still relatively high [24,85].

Looking forward to ‘Green Development’ in the future, China needs to facilitate further national and international collaborations between various stakeholders from different regions, institutions and private companies that focus on increased NUE in agriculture through sustainable N management to decrease N_r emissions from agriculture. This will add to progress already being made in the transport and industrial sectors. NH₃ emission mitigation from both agricultural and non-agricultural sectors is particularly important for making further reductions in secondary PM_2.5 pollution now that increasingly strict SO₂ and NO_x emission mitigations are being introduced [31,158]. Finally, atmospheric N_r emission reduction should be placed in the context of global climate change, environmental sustainability, food security and human health. We should pursue ‘win-win’ strategies over the long-term.

Data accessibility

This article has no additional data.

Authors' contributions

X.J.L.: conceptualization, writing—original draft; W.X., E.Z.D., A.H.T., Y.Y.Z., Y. Zhao, J.L.S. and Z.Z.F.: writing—original draft; Y.Z., Z.W., T.X.H., Y.P.P., L.Z., B.J.G., Y.Y.Z., F.Z., Z.L.G., Y.H.C., K.G., J.L.C., P.M.V. and F.S.Z.: writing—review and editing.

Competing interests

We declare we have no competing interests.

Funding

This work was supported by the National Natural Science Foundation of China (grant no. 41425007), the Chinese State Key Special Program on Severe Air Pollution Mitigation ‘Agricultural Emission Status and Enhanced Control Plan’ (grant no. DQGG0208), the State Key R&D programme (grant no. 2017YFC0210101, 2018YFC0213301–03) and the UK-China Virtual Joint Centre for Improved Nitrogen Agronomy (grant no. CINAg, BB/N013468/1).