- Email: [email protected]
Seasonal variations of nitrate dual isotopes in wet deposition in a tropical city in China
Atmospheric Environment 196 (2019) 1–9
Contents lists available at ScienceDirect
Atmospheric Environment journal homepage: www.elsevier.com/locate/atmosenv
Seasonal variations of nitrate dual isotopes in wet deposition in a tropical city in China
T
Fajin Chena,b, Qibin Laoa,b, Guodong Jiac,∗, Chunqing Chena,b, Qingmei Zhua,b, Xin Zhoua,b a
Guangdong Province Key Laboratory for Coastal Ocean Variation and Disaster Prediction, Guangdong Ocean University, Zhanjiang, 524088, China College of Ocean and Meteorology, Guangdong Ocean University, Zhanjiang, 524088, China c State Key Laboratory of Marine Geology, Tongji University, Shanghai, 200092, China b
ARTICLE INFO
ABSTRACT
Keywords: Dual nitrate isotopes Wet deposition NOx oxidation pathways Nitrogen source Tropical area
Monthly analytical results for more than two years long of δ15N and δ18O of nitrate in wet deposition are reported for a tropical city of Zhanjiang in the southernmost mainland China, in an effort to elucidate NOx sources and its oxidation pathways to nitrate. The results showed that monthly variations of δ18OeNO3- responded well to changes in sunshine hours, with lower δ18OeNO3- values corresponding to longer sunshine hours. This pattern suggests that NOx oxidation via the OH radical was the predominant pathway, which, by estimate, accounted for 87% in winter and 94% in summer, for nitrate formation. Remarkably, we found that the δ18OeNO3- here are prominently low relative to previous studies, likely due to that annual sunshine hours in this tropical city is relatively long. Moreover, available data of reported δ18OeNO3- in wet deposition showed an increasing trend with latitude, reflecting a shift of the predominant NOx oxidation pathways from those via the OH radical in low latitudes to those via O3 in high latitudes, corresponding to the decrease of sunshine hours. A temporarily co-variation of δ15NeNO3- and δ18OeNO3- was observed, exhibiting higher values in dry winter and lower values in wet summer, which was attributed to the washout effect on the dual isotopes. During wet seasons, complete nitrate washout suggests that δ15NeNO3- in precipitation should be similar to the nitrate formed in the atmosphere, and thereby can be used for source apportionment. A Bayesian model showed that the source of atmospheric nitrate deposition is mainly natural (61%) in Zhanjiang, with less anthropogenic contribution.
1. Introduction Atmospheric deposition of nitrate (NO3−) increased remarkably during the past decades due to increased nitrogen oxides (NOx, including NO and NO2) emission, particularly since the 1980s to the first few years of the 21rst century (Fang et al., 2010, 2013; Xiao et al., 2015). Although the nitrogen deposition slightly decreased since 2004, the value is still high, and poses a threat to the health of both terrestrial and aquatic ecosystems (Fang et al., 2013). This issue is particularly prominent in the coastal areas of southeast China and has caused serious environmental problems (Cui et al., 2014), which have attracted wide attention in China (Chen et al., 2009; Jia and Chen, 2010; Luo et al., 2008). NO3− in the atmosphere is mainly produced from NOx oxidation (Freyer et al., 1993; Jarvis et al., 2008). Therefore, knowledges of sources and the oxidation pathways of NOx are important for pollution
control. NOx in the atmosphere is derived from natural sources such as lightning and soil emission, and also from anthropogenic sources from vehicle emission and coal combustion (Heaton, 1990). There are two main pathways for NOx oxidation to HNO3, which mainly occur either during the day or at night. During the day, NO in the atmosphere is quickly oxidized to NO2 by ozone (O3) (R1). While NO2 can be either reduced back to NO by photolysis (R2), or oxidized to HNO3 via the OH radical (reaction R4). During the night, due to the halt of reaction R2 NO2 in the atmosphere is oxidized to NO3 radical by O3 (reaction R5), and further oxidized to dinitrogen pentoxide (N2O5) (reaction R6) that is easily hydrolyzed to form HNO3 (reaction R7). However, N2O5 is quickly decomposed by the thermal and photolytic reactions, especially in areas with longer sunshine. (Calvert et al., 1985; Hastings et al., 2004). (R1) NO + O3 → NO2 + O2
Corresponding author. State Key Laboratory of Marine Geology, Tongji University, 1239 Siping Road, Shanghai, 200092, China. E-mail addresses: [email protected] (F. Chen), [email protected] (Q. Lao), [email protected] (G. Jia), [email protected] (C. Chen), [email protected] (Q. Zhu), [email protected] (X. Zhou). ∗
https://doi.org/10.1016/j.atmosenv.2018.09.061 Received 10 May 2018; Received in revised form 11 September 2018; Accepted 30 September 2018 Available online 05 October 2018 1352-2310/ © 2018 Elsevier Ltd. All rights reserved.
Atmospheric Environment 196 (2019) 1–9
F. Chen et al.
Fig. 1. The sampling location of Zhanjiang in southern China (black dot). SM: summer monsoon; WM: winter monsoon.
oxidation pathways also control δ18OeNO3. According to Fang et al. (2010), in the pathway of OH oxidation, 2/3 of O arises from O3 and 1/ 3 of O from OH, while in the pathway of O3 oxidation 5/6 of O arises from O3 and 1/6 of O from OH. Atmospheric O3 is 18O enriched (δ18O ranging from +90 to +120‰) (Krankowsky et al., 1995), whereas the δ18O of OH reflect the isotopic composition of water vapor in the troposphere, exhibiting low values (ranging from −15‰ to 0‰) (Dubey et al., 1997). In addition to the O3 and OH oxidation pathways, NOx can also react with dimethylsufide (DMS), halogen oxides, and peroxy radicals (Alexander et al., 2009; Fang et al., 2010; Morin et al., 2008) to produce HNO3. Thus, the composition of δ18O in atmospheric nitrate exhibit a larger range than δ15N (Hastings et al., 2003, 2004; Michalski et al., 2003). For example, δ18OeNO3 in atmosphere cover a range of 60.3–73.4‰ in Bavaria (Durka et al., 1994) and 60.3–86.5‰ in Bermuda (Hastings et al., 2003). Interestingly, higher δ18O values of nitrate (87.0‰) were observed at high-latitude (78.7°N) Ny-Alesund (Morin et al., 2009) compared to values (67.0‰) at mid-latitude (43.9°N) New Hampshire (Pardo et al., 2004), which is likely caused by the difference in the NOx oxidation pathway between the two regions, i.e., the lack of photochemical reactions during winter at high latitudes would result in lower concentrations of hydroxyl radical, allowing the O3-involved pathways to be dominant (Hastings et al., 2004). In addition, the nitrate δ18O values (65.0‰) at a lower-latitude (23.2°N) city of China have been reported to be even lower than those in New Hampshire and Greenland (Fang et al., 2010). Although the dual nitrate isotopes in wet deposition have been reported extensively (Elliott et al., 2007; Fang et al., 2010; Hastings et al., 2003; Hastings et al., 2004; S. Morin et al., 2009; S Morin et al., 2008), little work has been conducted in the tropical regions characterized by long sunshine hours and stronger radiation under warm conditions (Fang et al., 2010), which could give rise to special patterns of NOx oxidation pathways. We hypothesize that more abundant photochemistry in tropical regions may cause lower δ18O values of atmospheric nitrate. To verify this speculation, a time-series investigation of δ15N and δ18O of rainwater nitrate has been made for more than 2 years at the Zhanjiang city in the southernmost part of mainland China. And here is the report of our results of this investigation.
(R2) NO2 + hv → NO + O (R3) O + O2 → O3 (R4) NO2 + OH → HNO3 (R5) NO2 + O3 → NO3 + O2 (R6) NO3 + NO2 → N2O5 (R7) N2O5 + H2O → 2HNO3 Previous studies suggest that stable isotopes of nitrogen (δ15N) and oxygen (δ18O) in nitrate can serve as a tool for distinguishing the contributions from various sources to atmospheric nitrogen and pathways of NOx oxidation (Durka et al., 1994; Elliott et al., 2007; Hastings et al., 2003, 2004; Pardo et al., 2004; Russell et al., 1998). The δ15N values from different NOx sources vary over a large range. For example, anthropogenic sources are characterized higher δ15N values, e.g. from +6 to +13‰ from coal combustion (Heaton, 1990) and between +3.7 and + 5.7‰ from vehicle NOx emissions (Ammann et al., 1999; Moore, 1977; Pearson et al., 2000). Comparatively, lower δ15N values have been reported from natural sources, such as lightning (∼0‰) (Hoering, 1957) and biogenic emissions (−49 to −20‰) (Li and Wang, 2008). Thereby, different sources could lead to different isotope signatures in atmospheric NO3−. In spite of this, isotope values of atmospheric NO3− may be also affected by the two main pathways of NOx oxidation. In the environments where NOx levels are usually low and sunshine durations are long enough, the unidirectional oxidation of NO2 through the OH radical during the day play a dominant role, which would lead to little change in δ15N-NOx values (Freyer et al., 1993). In contrast, in the night reactions using O3 as the oxidant (R5), isotope exchange at equilibrium occurs, causing 15N-enrichment in the higher N oxidation states, and whether the δ15NeNO3- value may reflect δ15N of source NOx is determined by the degree of NOx consumption. In polluted areas where the level of NOx is higher than O3 (Freyer et al., 1993), the resulted NO3− would be more enriched in 15N than NOx (Freyer et al., 1993). On the other hand, in environments with low NOx/O3 ratios, NOx can be completely oxidized to NO3−, resulting in δ15NeNO3- inheriting the δ15N signature of the NOx (Freyer et al., 1993). In addition to δ15N, NOx 2
Atmospheric Environment 196 (2019) 1–9
F. Chen et al.
1.8
600
NO3--N Precipitation during sampling periods Mean precipitation (1981-2010)
1.6
500
1.2
400
1.0 300 0.8 0.6
Rainfall (mm)
-1
Concentration (mg L )
1.4
200
0.4 100 0.2 0 May-15 Jun-15 Jul-15 Aug-15 Sep-15 Oct-15 Nov-15 Dec-15 Jan-16 Feb-16 Mar-16 Apr-16 May-16 Jun-16 Jul-16 Aug-16 Sep-16 Oct-16 Nov-16 Dec-16 Jan-17 Feb-17 Mar-17 Apr-17 May-17 Jun-17 Jul-17 Aug-17 Sep-17 Oct-17 Nov-17 Dec-17
0.0
Month Fig. 2. The variations of Monthly rainfall, mean rainfall from 1981 to 2010 and nitrate concentration. The lack of data in Dec-2016, Jan-2017 and Dec-2017 is because very little (< 0.5 mm) or no rainfall occurred during these periods.
2. Materials and methods
prior to sample loading. Similar procedures have been conducted in numerous studies (e.g., Fang et al., 2010; Hastings et al., 2003; Hastings et al., 2004; Jia and Chen, 2010; Savarino et al., 2006; Yeatman et al., 2001). A total of 208 rain events were sampled and the precipitation samples in each month were proportionally mixed into a large sample according to the rainfall amount in the month, thus composing a representative sample of wet deposition for the whole month. Therefore, the analysis results of the mixed samples are monthly weighted averages.
2.1. Sampling site and sample collection Zhanjiang city (20.00°–21.58°N, 109.52°–110.92°E) is located in southwestern Guangdong Province in southern China and is close to the South China Sea (Fig. 1). The local climate is characterized by tropical monsoon, exhibiting hot, humid conditions under the influence of southerly summer monsoon and cool, dry conditions dominated by northerly winter monsoon. The mean annual rainfall and temperature in Zhanjiang are 1689 mm and 23 °C, respectively (http://data.cma.cn/ ). Sunshine duration (∼2000 h) in this tropical city is relatively long in the world due to the low latitude (http://data.cma.cn/, https://en. wikipedia.org/wiki/Sunshine_duration). April is a transition month from the winter monsoon to the summer monsoon, during which rainfall usually increases sharply (Fig. 2). From May to September, there is a quasi-stationary front over southern China; it usually brings strong and sustained rainfalls due to the convergence of the northern cold air mass and the southern moisture. Zhanjiang is one of the rapidly developing cities in China. In recent years, the economy, industry, agricultural productivity and population have increased rapidly in the city. For example, the power production, which mainly comes from coal combustion, increased obviously from 9.41 × 109 kW h in 2015 to 16.46 × 109 kW h in 2017 (data from Zhanjiang Bureau of Statistics, http://zjtj.zhanjiang.gov.cn). The population was approximately 5.4 million in 1990 and increased to 7.2 million in 2015. The industrial production in Zhanjiang also rose rapidly from 2002 to 2013, and the city is facing potential nitrogen pollution (Meng et al., 2015). From May 2015 to November 2017, precipitation samples were collected on the top of a building at Guangdong Ocean University, located in the seaside city of Zhanjiang, using an acid-washed polyethylene bucket (80 cm in diameter) as the collector. The collector was placed at sampling site before rain to wait for a rain event. The rainfall less than 0.5 mm was considered invalid, and not collected. Each precipitation sample was filtered with a glass fiber filter (GF/F, Whatman, 47 mm in diameter), and then stored in a polyethylene bottle under −20 °C. The polyethylene bottles were soaked with 30% (v/v) HCl for 24 h, then cleaned with ultrapure water and dried in the laboratory
2.2. Sample analysis In the lab, NO3− concentration of rainwater was measured by a San++ continuous flow analyzer (Skalar, Netherlands). After nitrite was removed by adding sulfamic acid (Granger and Sigman, 2009), isotopic analysis of dissolved NO3− was conducted following a chemical conversion method modified from McIlvin and Altabet (2005). According to this method, NO3− was reduced to nitrite with spongy cadmium and further reduced to nitrous oxide with sodium azide in an acetic acid buffer. Subsequently, nitrous oxide was separated, purified and analyzed for δ15N and δ18O with a GasBench II-MAT 253. International standard IAEA-N3 was used to calibrate δ15N and δ18O. The analysis deviation for the standard was less than 0.2‰ for δ15N and less than 0.5‰ for δ18O. The reproducibility of duplicate sample analyses was less than 0.6‰ for δ18O (average ± 0.3‰) and less than 0.3‰ for δ15N (average ± 0.1‰). 2.3. Data sources and calculations The data of monthly sunshine duration from May 2015 to November 2017 and historical mean rainfall from 1981 to 2010 were from the China Meteorological Data Sharing Service System (http://data.cma. cn/); the monthly atmospheric NO2 and O3 were from historical data of air quality in Zhanjiang (https://www.aqistudy.cn/historydata/). Annual N load from wet deposition was calculated from volumeweighted-mean concentrations of NO3− and rainfall.
3
Atmospheric Environment 196 (2019) 1–9
F. Chen et al.
Fig. 3. Comparisons of nitrate isotope values with atmospheric parameters in Zhanjiang. (a) Monthly sunshine duration during sampling periods; (b) monthly atmospheric NO2/O3 ratio; (c) monthly variations of δ15NeNO3-; (d) monthly variations of δ18OeNO3-. The lack of data in Dec-2016, Jan-2017 and Dec-2017 is because very little (< 0.5 mm) or no rainfall occurred during this period.
2.4. Backward trajectories
follows (Xue et al., 2012; Moore and Semmens, 2010):
To analyze the influence of wind transport, three days (72 h) back trajectories were computed utilizing the web version model of the Hybrid Single-Particle Lagrangian Intergrated Trajectory (HYSPLIT) (https://ready.arl.noaa.gov/HYSPLIT_traj.php). Trajectories were generated from the sampling site at the heights 500 m above sea level. The trajectory time of 72 h was chosen because the NOx and the HNO3 lifetime in the boundary layer are generally less than 3 days (Fang et al., 2010; Liang et al., 1998).
k
Xij =
Pk (Sjk + cjk ) + k=1
ij
(1)
Sjk ∼ N (μjk, ω2jk) cjk ∼ N (λjk, τ2jk) εjk ∼ N (0, σ2j ) where Xij represents the isotope values j of the sample i, in which i = 1, 2, 3, …, N and j = 2 (δ15NeNO3- and δ18OeNO3); Sjk represents the isotope value j of the source k (k = 1, 2, 3, …, K), and Sjk is normally distributed with average μjk, and standard deviation ωjk; Pk represents the contribution of source k, which needs to be calculated by the SIAR model; cjk represents the fractionation factor for δ15NeNO3- and δ18OeNO3 on source k, and cjk is normally distributed with average λjk and standard deviation τjk; εjk represents the residual error of the
2.5. SIAR mixing model To calculate the relative contribution of nitrate sources, the Bayesian stable isotope mixing model was conducted in the software package SIAR (stable isotope analysis in R). In this model, the Bayesian framework is used to determine the probability distribution proportion of each source to the mixture. The Bayesian model framework is as 4
Atmospheric Environment 196 (2019) 1–9
F. Chen et al.
additional unquantified variation between individual samples, and εjk is normally distributed with average 0 and standard deviation σj. The model has been successfully used to estimate the proportional contribution of sources (Xue et al., 2012; Moore and Semmens, 2010; Zhang et al., 2018).
4. Discussion 4.1. NO3− concentration The nitrate concentration in wet deposition of Zhanjiang is lower than those in relatively more developed regions in China, such as the Pearl River Delta region (0.5–5.5 mg L−1, 1.2 mg L−1 average) (Jia and Chen, 2010) and the Yangtze River Delta region (0.97 mg L−1 average) (Zhao et al., 2009), but much higher than those in least-disturbed regions, such as the subtropical forested region in southern China (0.02–0.08 mg L−1) (Chen and Mulder, 2007) and the North and South Pacific Oceans (0.01–1.12 mg L−1) (Jung et al., 2011), indicating great influences of human activities (Jia and Chen, 2010). The seasonal distribution pattern of NO3− concentrations, i.e., higher values in dry seasons (Nov. to Mar.) and lower values in rainy seasons (May. to Oct.), is likely caused by the dilution effect of precipitation amount (Fig. 2), but could be also influenced by the shift from winter monsoon to summer monsoon. During winter Siberian high blows from the continent to Zhanjiang, making wet deposition in Zhanjiang to be mainly influenced by the highly polluted continental air mass in China; whereas during summer the southerly oceanic air blows to Zhanjiang is quite clean (Fig. 5). Nevertheless, wet nitrate deposition in wet season (3.04 kg N ha−2) was higher than that dry deposition (1.74 kg N ha−2), likely due to the heavy washout of nitrate in summer. The temporal changes in contributions from different sources could also have affected nitrogen isotopes as discussed later.
3. Results The monthly rainfall from May 2015 to November 2017 is summarized in Fig. 2. The lack of data in December 2016, January and December 2017 is because very little (< 0.5 mm) or no rainfall occurred during this period. Rainfall in certain months was significantly higher than the 30-year monthly mean rainfall between 1981 and 2010 (Fig. 2). Higher rainfalls occurred between April and October (Fig. 2). But there occurred an anomalously high rainfall in January 2016 (368 mm). The monthly weighted-mean concentrations of nitrate ranged from 0.06 to 1.33 mg L−1 and from 0.02 to 1.57 mg L−1 in 2016 and 2017, respectively. The calculated annual volume-weighted-mean concentrations of NO3− according to the precipitation of each month were 0.20 mg L−1 and 0.19 mg L−1 in 2016 and 2017, respectively. The NO3− concentrations in wet deposition at Zhanjiang in 2016 are comparable to those in 2017 (Fig. 2), suggesting a stable atmospheric nitrogen processing in the city. Higher concentrations of nitrate occurred during the dry seasons, whereas lower concentrations during the wet seasons (Fig. 2). According to the precipitation and annual volumeweighted-mean concentrations of NO3− for the two years, the wet N deposition in 2016 and 2017 in Zhanjiang were 4.77 kg N ha−2 and 3.79 kg N ha−2, respectively. δ15NeNO3- values ranged from −1.8‰ in July 2016 to +4.1‰ in March 2017, exhibiting lower mean values in hot rainy seasons (i.e., 0.2‰) than in cool dry seasons (i.e., 2.2‰). The δ18OeNO3- values ranged from +42.7 to +61.6‰, also showing higher values in dry seasons and lower values in wet seasons (Fig. 3). The distribution patterns of δ15NeNO3- and δ18OeNO3- values were opposite to the pattern of sunshine; the latter exhibited an average of 199 h per month in wet season but an average of only 114 h per month in dry season (Fig. 3). Both δ15NeNO3- and δ18OeNO3- also showed significant relationships with rainfall (Fig. 4). The volume-weighted annual mean values of δ15NeNO3- and δ18OeNO3- were +0.8‰ and +52.4‰, respectively.
4.2. Influencing factors of nitrate isotopes δ15NeNO3- and δ18OeNO3- in our results exhibited higher values in the dry seasons and lower values in the wet seasons (Fig. 3), which is similar to previous studies (Fang et al., 2010; Hastings et al., 2003, 2004; Savarino et al., 2006; Yeatman et al., 2001). The paralleled seasonal variation of δ15NeNO3- and δ18OeNO3- is intriguing (Fig. 3). As stated above, δ18OeNO3- is mainly determined by the NOx oxidation pathways, whereas δ15NeNO3- could be modulated either by NOx source or by NOx oxidation pathways. So, it seems that NOx oxidation pathways might have determined synchronous variations of the dual isotopes. To explore whether the NOx oxidation pathways have accounted for the seasonal variations of nitrate dual isotopes in this study, we compared contemporaneous sunshine duration and NO2/O3 ratio with the dual isotope data in Zhanjiang (Fig. 3). As shown in Fig. 3b, the level of
Fig. 4. Correlation of (a) δ15N and (b) δ18O of wet nitrate deposition with rainfall.
5
Atmospheric Environment 196 (2019) 1–9
F. Chen et al.
Fig. 5. Backward air mass trajectories in precipitation events in Zhanjiang, based on NOAA HYSPLIT model back trajectories; the red lines denote air mass trajectories occurring in the wet season and the blue lines denote trajectories in the cool dry seasons. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)
O3 in Zhanjiang was much higher than that of NO2, and the NO2/O3 ratio ranged from 0.11 to 0.24, with an average of 0.18 ± 0.04 in dry season and 0.16 ± 0.02 in wet season and not showing significant differences between wet and dry seasons. Thus, the discussions about other variables such as temperature that might also affect NO2/O3 ratio were not included. The NO2/O3 ratio in Zhanjiang is significantly low compared to some previous studies, in which most NO2/O3 ratio values are > 1.0 and vary greatly (e.g., Fang et al., 2010; Freyer et al., 1993; Hastings et al., 2003; Hastings et al., 2004; Jiaand Chen, 2010). In contrast, Sunshine duration (∼2000 h) in Zhanjiang is relatively long in the world due to the low latitude and the clean air caused by the ocean climate (http://data.cma.cn/, https://en.wikipedia.org/wiki/Sunshine_ duration). In addition, the curve of sunshine hours, longer in summer and shorter in winter, corresponds well to the time-series variation of δ15NeNO3- and δ18OeNO3- values (Fig. 3). Therefore, although NOx could be completely oxidized to NO3− (via N2O5) in the atmospheric environments with such low NO2/O3 ratios (Freyer et al., 1993), we hypothesize that the formation of nitrate via the OH radical, i.e., the day reactions, is likely responsible for the seasonal variation of the nitrate isotope values due to long sunshine in this tropical area. The rainfall may be another factor influencing the seasonal patterns in δ15NeNO3- and δ18OeNO3-, since both δ15NeNO3- and δ18OeNO3- of wet deposition were positively correlated to rainfall (Fig. 4). We thereby hypothesize that low precipitation in the dry seasons could have resulted in low relative washout of nitrate, leading 15N and 18O to be preferentially washed out, i.e., higher δ15NeNO3- and δ18OeNO3values in wet deposition. In contrast, during wet seasons, complete nitrate removal are expected, leading δ15NeNO3- and δ18OeNO3- in wet deposition to be similar to the source nitrate isotopes in atmosphere. This suggestion is in agreement with the previous findings that during NO3− gas-liquid and particle-gas interactions, heavy isotope is preferentially incorporated into the condensed phase that is readily washed out, thus causing the successive precipitation progressively depleted in the dual isotopes of NO3− (Moore, 1977; Heaton, 1986). Generally, δ18O values in OH radical are from −15‰ to 0‰ (Fang
et al., 2010), and in O3 from +90‰ to +122‰ (Krankowsky et al., 1995). Accordingly, the minimum values (−15‰ for OH radical and +90‰ for O3) were chosen to calculate the minimum δ18OeNO3- value of atmospheric nitrate, which was around +55‰ if 2/3 of nitrate-O arises from O3 and 1/3 of O from OH; similarly, using the maximum theoretical values (0‰ for OH radical and +122‰ for O3), the maximum δ18OeNO3- value of atmospheric nitrate was +102‰ if 5/6 of nitrate-O arises from O3 and 1/6 of O from OH (Fang et al., 2010). The δ18OeNO3- values in our samples, i.e. 52.4‰ in average, are close to the minimum theoretical value of +55‰, suggesting the minimum endmember values are suitable for calculation in this study. However, some samples in our study had δ18OeNO3- being lower than +55‰. We hypothesized that there might be other pathways to decrease δ18OeNO3-. As stated above, in addition to the two main channels (the OH pathway and the O3 pathway) for atmospheric NO3− formation, which account for more than 94% of atmospheric NO3− globally (Alexander et al., 2009), NOX can also react with dimethylsufide (DMS), halogen oxides (i.e. bromine oxide (BrO)), and peroxy radicals (Alexander et al., 2009; Fang et al., 2010; Morin et al., 2008) to produce atmospheric NO3−. However, these factors account for less than 6% of the annual nitrate production in the world (Alexander et al., 2009). The DMS and halogen oxides pathways can be ruled out as the causes of the negative δ18OeNO3- anomalies, because atmospheric NO3− induced via the two pathways would probably have higher 18O values than those induced via the OH pathway and the O3 pathway due to its more involvement with O3 during reactions (Morin et al., 2008; Alexander et al., 2009). One possibility for the lower 18O values of NO3− may be the reaction of NOx with peroxy radicals (HO2 and its organic homologues RO2), because the O atoms in peroxy radicals mainly come from atmospheric O2 rather than O3 andδ18O of O2 is around +23.9‰ (Fang et al., 2010). Thus, the resultant δ18OeNO3- value via the oxidation pathway by peroxy radicals was around +28‰ followed by 4/6 of the O atom from atmospheric O2, 1/6 from O3 and 1/6 O from H2O (Alexander et al., 2009). Thereby, the oxidation pathway by peroxy radicals may be partly responsible for the low δ18OeNO3- value in our 6
Atmospheric Environment 196 (2019) 1–9
F. Chen et al.
Among the anthropogenic sources of NOx, fossil fuel combustion, particularly coal combustion, is likely dominant in Zhanjiang, because coal is commonly used to generate electricity in this economically developing area. This suggestion agrees with the bottom-up inventory, which revealed that, more than 70% of the total NOx emission was derived from coal combustion in China (Tian et al., 2001). It is also agreeable with the previous study in Guangzhou City, which is only 400 km from Zhanjiang (Fang et al., 2010). Natural sources include lightning and biogenic emissions (Fang et al., 2010). The lightning-derived contribution to the atmospheric NOx is important above the South China Sea (Price et al., 1997; Xiao et al., 2015), and likely among the highest in the world (Price et al., 1997). The mean values of δ15NeNO3- during rainy season (+0.15‰) was close to that for the lightning (around 0‰) (Hoering, 1957). In addition, the negative values of δ15NeNO3- appeared in most months during the warm season in 2015 and 2016, suggesting contributions from biogenic soil emissions that has been considered as an important source in tropical areas (Felix and Elliott, 2014; Li and Wang, 2008). Here, relative contributions of the above sources, i.e., anthropogenic sources including vehicle emission (−2 to −13‰ range, −7.0 ± 4.7‰ average) and coal combustion (+6.0 to +13.0‰ range, 9.6 ± 2.9‰ average) (Heaton, 1990) and natural sources including soil emission (−19.8 to −48.6‰ range, −21.9 ± 9.3‰ average) and lightning (−0.5 to +1.4‰ range, 0 ± 0.7‰ average) (Hoering, 1957; Li and Wang, 2008) were estimated for the wet deposition using a Bayesian model. The SIAR mixing model outputs showed high variabilities in contributions of the four potential nitrate sources (Fig. 7). The contribution of lightning is highest (mean probability estimate (MPE) between 29% and 82%, average of 56%), followed by coal combustion (MPE between 10% and 43%, average of 26%), vehicle emission (MPE between 0% and 27%, average of 13%) and soil emission (MPE between 0% and 10%, average of 5%). Generally, the estimates show that the contribution of natural sources for atmospheric nitrate deposition dominated in Zhanjiang (61%). As to the atmospheric nitrate in dry seasons, in addition to the incomplete nitrate washout, the cause of relatively higher δ15NeNO3values (+2.2‰) could be also associated with the polluted NOx carried by the northerly winds during winter, whose contribution is difficult to estimate in this study. Therefore, our results indicate that δ15NeNO3- in precipitation in wet season is more applicable for source apportionment than in dry season in Zhanjiang because of minimal isotope fractionation during the formation of atmospheric nitrate.
Fig. 6. Variation characteristics of δ18OeNO3- values in nitrate of wet deposition with latitude. (a) The δ18OeNO3- values in Antarctica (Savarino et al., 2006); (b) results from our study (Zhanjiang); (c) Guangzhou, South China (Fang et al., 2010); (d) Bermuda (Hastings et al., 2003); (e) New Hampshire, US (Pardo et al., 2004); (f) Greenland (Hastings et al., 2004); (g) Ny-Alesund (Morin et al., 2009).
study, which is similar to a previous point in a mid-low latitude city (Fang et al., 2010). In our study, δ18OeNO3- in the dry seasons (56.8‰ on average) was higher than that in the wet seasons (50.4‰ on average). By assuming 6% contribution of the NOx oxidation pathway to atmospheric NO3 (Alexander et al., 2009), the relative contribution of NOx oxidation via the OH radical and O3 can be calculated for the two seasons, showing that the OH radical pathway was accounted for averagely 87% in the dry seasons and more than 94% in the wet seasons. However, the calculated contribution of OH radical pathway during the dry seasons is likely underestimated, since incomplete washout of nitrate could have resulted in higher δ18OeNO3- values in wet deposition during the dry seasons. Nevertheless, our estimates indicate that the day reactions dominate the atmospheric NO3− formation throughout the year in Zhanjiang with neglectable contribution from night reactions. As mentioned above, the predominance of day reactions for the atmospheric NO3− formation suggested in this study is different from previous studies at higher latitudes. Thereby, we hypothesize whether the δ18OeNO3- values increase with climate and latitude. When available δ18OeNO3- values are plotted versus latitude (Fig. 6), as expected, they indeed increase with latitude, with the highest values in the polar region and the lowest values (this study) in the tropic. The lower values of δ18OeNO3- in wet deposition in this study indicate the main contribution of OH radical pathways to the formation of atmospheric nitrate (Elliott et al., 2015; Fang et al., 2010; Hastings et al., 2003, 2004). This is largely because the tropical Zhanjiang has a longer summer time and longer sunshine. This view is further supported by previous studies, which have demonstrated using δ17O model that the pathway of nitrate production by the OH radical is predominant (reaching 87%) in the tropical areas, which have the highest level of OH (Alexander et al., 2009). In contrast, longer wintertime and less sunshine at the mid-high latitudes may lead to more contribution from night reactions using O3 as oxidant and less dilution by the day reactions, which could result in higher δ18OeNO3- values compared to the mid-low latitude areas. 4.3. Source variation revealed by monthly δ15NeNO3- changes The complete nitrate removal during rainfall washout in wet seasons suggests that δ15NeNO3- in wet deposition can be applied to identify the NOx sources during these seasons. Generally, the atmospheric NOx has anthropogenic and natural sources (Fang et al., 2010).
Fig. 7. Percent contribution of four potential nitrate sources for wet season deposition estimated by SIAR. 7
Atmospheric Environment 196 (2019) 1–9
F. Chen et al.
5. Conclusions
359–366. Freyer, H.D., Kley, D., Volz-Thomas, A., Kobel, K., 1993. On the interaction of isotopic exchange processes with photochemical reactions in atmospheric oxides of nitrogen. J. Geophys. Res. 98 (D8), 14791–14796. https://doi.org/10.1029/93JD00874. Granger, J., Sigman, D.M., 2009. Removal of nitrite with sulfamic acid for nitrate N and O isotope analysis with the denitrifier method. Rapid Commun. Mass Spectrom. 23 (23), 3753–3762. https://doi.org/10.1002/rcm.4307. Hastings, M.G., Sigman, D.M., Lipschultz, F., 2003. Isotopic evidence for source changes of nitrate in rain at Bermuda. J. Geophys. Res. 108 (D24). https://doi.org/10.1029/ 2003JD003789. Hastings, M.G., Steig, E.J., Sigman, D.M., 2004. Seasonal variations in N and O isotopes of nitrate in snow at Summit, Greenland: implications for the study of nitrate in snow and ice cores. J. Geophys. Res. 109 (D20). https://doi.org/10.1029/2004JD004991. Heaton, T.H.E., 1986. Isotopic studies of nitrogen pollution in the hydrosphere and atmosphere: a review. Chem. Geol. 59 (2), 87–102. https://doi.org/10.1016/00092541(86)90046-X. Heaton, T.H.E., 1990. 15N/14N ratios of NOx from vehicle engines and coal-fired power stations. Tellus 42 (3), 304–307. https://doi.org/10.3402/tellusb.v42i3.15223. Hoering, T., 1957. The isotopic composition of the ammonia and the nitrate ion in rain. Geochem. Cosmochim. Acta 12 (1), 97–102. https://doi.org/10.1016/0016-7037(57) 90021-2. Jarvis, J.C., Steig, E.J., Hastings, M.G., Kunasek, S.A., 2008. Influence of local photochemistry on isotopes of nitrate in Greenland snow. Geophys. Res. Lett. 35 (21), L21804. https://doi.org/10.1029/2008GL035551. Jia, G.D., Chen, F.J., 2010. Monthly variations in nitrogen isotopes of ammonium and nitrate in wet deposition at Guangzhou, south China. Atmos. Environ. 44 (19), 2309–2315. https://doi.org/10.1016/j.atmosenv.2010.03.041. Jung, J., Furutani, H., Uematsu, M., 2011. Atmospheric inorganic nitrogen in marine aerosol and precipitation and its deposition to the North and South Pacific Oceans. J. Atmos. Chem. 68 (2), 157–181. https://doi.org/10.1007/s10874-012-9218-5. Krankowsky, D., Bartecki, F., Klees, G.G., Mauersberger, K., Schellenbach, K., Stehr, J., 1995. Measurement of heavy isotope enrichment in tropospheric ozone. Geophys. Res. Lett. 22 (13), 1713–1716. https://doi.org/10.1029/95GL01436. Li, D., Wang, X., 2008. Nitrogen isotopic signature of soil-released nitric oxide (NO) after fertilizer application. Atmos. Environ. 42 (19), 4747–4754. https://doi.org/10.1016/ j.atmosenv.2008.01.042. Liang, J.Y., Horowitz, L.W., Jacob, D.J., Wang, Y.H., Fiore, A.M., Logan, J.A., Gardner, G.M., Munger, J.W., 1998. Seasonal budgets of reactive nitrogen species and ozone over the United States, and export fluxes to the global atmosphere. J. Geophys. Res. 103 (D11), 13435–13450. https://doi.org/10.1029/97JD03126. Luo, Y., Liu, S., Fu, S.L., Liu, J., Wang, G., Zhou, G., 2008. Trends of precipitation in beijiang river basin, Guangdong Province, China. Hydrol. Process. 22 (13), 2377–2386. https://doi.org/10.1002/hyp.6801. Meng, Y.F., Chen, F.J., Zeng, Z., 2015. Nitrogen budget evolution in zhanjiang municipal district and its environmental effect during 2002-2013. Trop. Geogr. 35 (5), 777–784. in Chinese with English abstract. https://doi.org/10.13284/j.cnki.rddl.002752. McIlvin, M.R., Altabet, M.A., 2005. Chemical conversion of nitrate and nitrite to nitrous oxide for nitrogen and oxygen isotopic analysis in Freshwater and seawater. Anal. Chem. 77, 5589–5595. https://doi.org/10.1021/ac050528s. Michalski, G., Scott, Z., Kabiling, M., Thiemens, M.H., 2003. First measurements and modeling of Δ17O in atmospheric nitrate. Geophys. Res. Lett. 30 (16), 1870. https:// doi.org/10.1029/2003GL017015. Moore, H., 1977. The isotopic composition of ammonia, nitrogen dioxide and nitrate in the atmosphere. Atmos. Environ. 11 (12), 1239–1243. https://doi.org/10.1016/ 0004-6981(77)90102-0. Moore, J.W., Semmens, B.X., 2010. Incorporating uncertainty and prior information into stable isotope mixing models. Ecol. Lett. 11 (5), 470–480. https://doi.org/10.1111/j. 1461-0248.2008.01163.x. Morin, S., Savarino, J., Frey, M.M., Domine, F., Jacobi, H.W., Kaleschke, L., Martins, M.F., 2009. Comprehensive isotopic composition of atmospheric nitrate in the Atlantic Ocean boundary layer from 65°S to 79°N. J. Geophys. Res. 114 (D05303). https:// doi.org/10.1029/2008JD010696. Morin, S., Savarino, J., Frey, M.M., Yan, N., Bekki, S., Bottenheim, J.W., 2008. Tracing the origin and fate of NOx in the Arctic atmosphere using stable isotopes in nitrate. Science 322 (5902), 730–732. https://doi.org/10.1126/science.1161910. Pardo, L.H., Kendall, C., Pett-Ridge, J., Chang, C.C.Y., 2004. Evaluating the source of streamwater nitrate using δ15N and δ18O in nitrate in two watersheds in New Hampshire, USA. Hydrol. Process. 18 (14), 2699–2712. https://doi.org/10.1002/ hyp.5576. Pearson, J., Wells, D.M., Seller, K.J., Bennett, A., Soares, A., Woodall, J., Ingrouille, M.J., 2000. Traffic exposure increases natural 15N and heavy metal concentrations in mosses. New Phytol. 147 (2), 317–326. https://doi.org/10.1046/j.1469-8137.2000. 00702.x. Price, C., Prather, M., 1997. NOx from lightning .1. Global distribution based on lightning physics. J. Geophys. Res. 102 (D5), 5929–5941. https://doi.org/10.1029/ 96JD03504. Russell, K.M., Galloway, J.N., Macko, S.A., Moody, J.L., Scudlark, J.R., 1998. Sources of nitrogen in wet deposition to the Chesapeake Bay region. Atmos. Environ. 32 (14), 3923–3927. https://doi.org/10.1016/S1352-2310(98)00044-2. Savarino, J., Kaiser, J., Morin, S., Sigman, D.M., Thiemens, M.H., 2006. Nitrogen and oxygen isotopic constraints on the origin of atmospheric nitrate in coastal Antarctica. Atmos. Chem. Phys. 7 (8), 1925–1945. https://doi.org/10.5194/acpd-6-8817-2006. Tian, H.Z., Hao, J.M., Lu, Y.Q., Zhu, T.L., 2001. Inventories and distribution characteristics of NOx emissions in China. China Environ. Sci. 21 (6), 493–497 in Chinese with English abstract. Xiao, H.W., Xie, L.H., Long, A.M., Ye, F., Pan, Y.P., Li, D.N., Long, Z.H., Chen, L., Xiao,
δ18OeNO3- values in wet deposition in the tropical city of Zhanjiang manifest the pathway of NOx oxidation via the OH radical was predominant (87% in dry seasons and 94% in wet seasons) for the nitrate formation. This might be related to the long summer time and strong sunshine in this tropical area, which could be further demonstrated by the finding of an increasing trend of δ18OeNO3- with latitude, reflecting a shift of the predominant NOx oxidation pathways from the day reactions in low latitudes to night reactions in high latitudes. Rainfall washout exerted an effect on δ15NeNO3- and δ18OeNO3- values. During wet seasons, complete nitrate washout suggests that δ15NeNO3- in precipitation should be similar to the nitrate formed in the atmosphere, and thereby can be used for source apportionment. A Bayesian model showed that the contributions of natural and anthropogenic sources for atmospheric nitrate deposition were 61% and 39%, respectively. Acknowledgments This work was supported by National Natural Science Foundation of China (41476066), Guangdong Natural Science Foundation of China (2016A030312004), International Science and technology cooperation project (GASI-IPOVAI-04), National Key Research and Development Plan (2016YFC1401403), Project of Enhancing School with Innovation of Guangdong Ocean University (GDOU2014050201, GDOU2013010203,GDOU2013050201). The authors would like to thank Xijie Yin for technical support in Third Institute of Oceanography, State Oceanic Administration. References Alexander, B., Hastings, M.G., Allman, D.J., Dachs, J., 2009. Quantifying atmospheric nitrate formation pathways based on a global model of the oxygen isotopic composition (Δ17O) of atmospheric nitrate. Atmos. Chem. Phys. 9 (14), 5043–5056. https:// doi.org/10.5194/acp-9-5043-2009. Ammann, M., Siegwolf, R., Pichlmayer, F., Suter, M., Saurer, M., Brunold, C., 1999. Estimating the uptake of traffic-derived NO2 from 15N abundance in Norway spruce needles. Oecologia 118 (2), 124–131. https://doi.org/10.1007/s004420050710. Calvert, J.G., Lazrus, A., Kok, G.L., Heikes, B.G., Walega, J.G., Lind, J., Gantrell, C.A., 1985. Chemical mechanisms of acid generation in the troposphere. Nature 317 (6032), 27–35. https://doi.org/10.1038/317027a0. Chen, F.J., Jia, G.D., Chen, J.Y., 2009. Nitrate sources and watershed denitrification inferred from nitrate dual isotopes in the Beijiang River, south China. Biogeochemistry 94 (2), 163–174. https://doi.org/10.1007/s10533-009-9316-x. Chen, X.Y., Mulder, J., 2007. Atmospheric deposition of nitrogen at five subtropical forested sites in South China. Sci. Total Environ. 378 (3), 317–330. https://doi.org/ 10.1016/j.scitotenv.2007.02.028. Cui, J., Zhou, J., Peng, Y., He, Y., Yang, H., Mao, J., 2014. Atmospheric wet deposition of nitrogen and sulfur to a typical red soil agroecosystem in Southeast China during the ten-year monsoon seasons (2003-2012). Atmos. Environ. 82 (82), 121–129. https:// doi.org/10.1016/j.atmosenv.2013.10.023. Dubey, M.K., Ralf Mohrschladt, N.M.D., Anderson, J.G., 1997. Isotope specific kinetics of hydroxyl radical (OH) with water (H2O): testing models of reactivity and atmospheric fractionation. J. Phys. Chem. 101 (8), 1494–1500. https://doi.org/10.1021/ jp962332p. Durka, W., Schulze, E.D., Gebauer, G., Voerkeliust, S., 1994. Effects of forest decline on uptake and leaching of deposited nitrate determined from 15N and 18O measurements. Nature 372 (6508), 765–767. https://doi.org/10.1038/372765a0. Elliott, E.M., Kendall, C., Boyer, E.W., Burns, D.A., Lear, G.G., Golden, H.E., Harlin, K., Bytnerowicz, A., Butler, T.J., Glatz, R., 2015. Dual nitrate isotopes in dry deposition: utility for partitioning NOx source contributions to landscape nitrogen deposition. J. Geophys. Res. 114 (G4), 425–453. https://doi.org/10.1029/2008JG000889. Elliott, E.M., Kendall, C., Wankel, S.D., Burns, D.A., Boyer, E.W., Harlin, K., Bain, D.J., Butler, T.J., 2007. Nitrogen isotopes as indicators of NO(x) source contributions to atmospheric nitrate deposition across the midwestern and northeastern United States. Environ. Sci. Technol. 41 (22), 7661–7667. https://doi.org/10.1021/es070898t. Fang, Y.T., Koba, K., Wang, X.M., Wen, D.Z., Li, J., Takebayashi, Y., Liu, X.Y., Yoh, M., 2010. Anthropogenic imprints on nitrogen and oxygen isotopic composition of precipitation nitrate in a nitrogen-polluted city in southern China. Atmos. Chem. Phys. 11, 1313–1325. https://doi.org/10.5194/acp-11-1313-2011. Fang, Y.T., Wang, X.M., Zhu, F.F., Wu, Z.Y., Li, J., Zhong, L.J., Chen, D.H., Yoh, M., 2013. Three-decade changes in chemical composition of precipitation in Guangzhou city, southern China: has precipitation recovered from acidification following sulphur dioxide emission control? Tellus B 65 (9), 134–138. Felix, J.D., Elliott, E.M., 2014. Isotopic composition of passively collected nitrogen dioxide emissions: vehicle, soil and livestock source signatures. Atmos. Environ. 92,
8
Atmospheric Environment 196 (2019) 1–9
F. Chen et al. H.Y., Liu, C.Q., 2015. Use of isotopic compositions of nitrate in TSP to identify sources and chemistry in South China Sea. Atmos. Environ. 109, 70–78. https://doi. org/10.1016/j.atmosenv.2015.03.006. Xue, D., De Baets, B., Van Cleemput, O., Hennessy, C., Berglund, M., Boeckx, P., 2012. Use of a Bayesian isotope mixing model to estimate proportional contributions of multiple nitrate sources in surface water. Environ. Pollut. 161 (1), 43–49. https://doi.org/10. 1016/j.envpol.2011.09.033. Yeatman, S.G., Spokes, L.J., Dennis, P.F., Jickells, T.D., 2001. Comparisons of aerosol nitrogen isotopic composition at two polluted coastal sites. Atmos. Environ. 35 (7),
1307–1320. https://doi.org/10.1016/S1352-2310(00)00408-8. Zhang, M., Zhi, Y., Shi, J., Wu, L., 2018. Apportionment and uncertainty analysis of nitrate sources based on the dual isotope approach and a Bayesian isotope mixing model at the watershed scale. Sci. Total Environ. 639, 1175–1187. https://doi.org/ 10.1016/j.scitotenv.2018.05.239. Zhao, X., Yan, X.Y., Xiong, Z.Q., Xie, Y.X., Xing, G.X., Shi, S.L., Zhu, Z.L., 2009. Spatial and temporal variation of inorganic nitrogen wet deposition to the Yangtze River Delta Region, China. Water Air Soil Pollut. 203 (1–4), 277–289. https://doi.org/10. 1007/s11270-009-0011-2.
9
Contents lists available at ScienceDirect
Atmospheric Environment journal homepage: www.elsevier.com/locate/atmosenv
Seasonal variations of nitrate dual isotopes in wet deposition in a tropical city in China
T
Fajin Chena,b, Qibin Laoa,b, Guodong Jiac,∗, Chunqing Chena,b, Qingmei Zhua,b, Xin Zhoua,b a
Guangdong Province Key Laboratory for Coastal Ocean Variation and Disaster Prediction, Guangdong Ocean University, Zhanjiang, 524088, China College of Ocean and Meteorology, Guangdong Ocean University, Zhanjiang, 524088, China c State Key Laboratory of Marine Geology, Tongji University, Shanghai, 200092, China b
ARTICLE INFO
ABSTRACT
Keywords: Dual nitrate isotopes Wet deposition NOx oxidation pathways Nitrogen source Tropical area
Monthly analytical results for more than two years long of δ15N and δ18O of nitrate in wet deposition are reported for a tropical city of Zhanjiang in the southernmost mainland China, in an effort to elucidate NOx sources and its oxidation pathways to nitrate. The results showed that monthly variations of δ18OeNO3- responded well to changes in sunshine hours, with lower δ18OeNO3- values corresponding to longer sunshine hours. This pattern suggests that NOx oxidation via the OH radical was the predominant pathway, which, by estimate, accounted for 87% in winter and 94% in summer, for nitrate formation. Remarkably, we found that the δ18OeNO3- here are prominently low relative to previous studies, likely due to that annual sunshine hours in this tropical city is relatively long. Moreover, available data of reported δ18OeNO3- in wet deposition showed an increasing trend with latitude, reflecting a shift of the predominant NOx oxidation pathways from those via the OH radical in low latitudes to those via O3 in high latitudes, corresponding to the decrease of sunshine hours. A temporarily co-variation of δ15NeNO3- and δ18OeNO3- was observed, exhibiting higher values in dry winter and lower values in wet summer, which was attributed to the washout effect on the dual isotopes. During wet seasons, complete nitrate washout suggests that δ15NeNO3- in precipitation should be similar to the nitrate formed in the atmosphere, and thereby can be used for source apportionment. A Bayesian model showed that the source of atmospheric nitrate deposition is mainly natural (61%) in Zhanjiang, with less anthropogenic contribution.
1. Introduction Atmospheric deposition of nitrate (NO3−) increased remarkably during the past decades due to increased nitrogen oxides (NOx, including NO and NO2) emission, particularly since the 1980s to the first few years of the 21rst century (Fang et al., 2010, 2013; Xiao et al., 2015). Although the nitrogen deposition slightly decreased since 2004, the value is still high, and poses a threat to the health of both terrestrial and aquatic ecosystems (Fang et al., 2013). This issue is particularly prominent in the coastal areas of southeast China and has caused serious environmental problems (Cui et al., 2014), which have attracted wide attention in China (Chen et al., 2009; Jia and Chen, 2010; Luo et al., 2008). NO3− in the atmosphere is mainly produced from NOx oxidation (Freyer et al., 1993; Jarvis et al., 2008). Therefore, knowledges of sources and the oxidation pathways of NOx are important for pollution
control. NOx in the atmosphere is derived from natural sources such as lightning and soil emission, and also from anthropogenic sources from vehicle emission and coal combustion (Heaton, 1990). There are two main pathways for NOx oxidation to HNO3, which mainly occur either during the day or at night. During the day, NO in the atmosphere is quickly oxidized to NO2 by ozone (O3) (R1). While NO2 can be either reduced back to NO by photolysis (R2), or oxidized to HNO3 via the OH radical (reaction R4). During the night, due to the halt of reaction R2 NO2 in the atmosphere is oxidized to NO3 radical by O3 (reaction R5), and further oxidized to dinitrogen pentoxide (N2O5) (reaction R6) that is easily hydrolyzed to form HNO3 (reaction R7). However, N2O5 is quickly decomposed by the thermal and photolytic reactions, especially in areas with longer sunshine. (Calvert et al., 1985; Hastings et al., 2004). (R1) NO + O3 → NO2 + O2
Corresponding author. State Key Laboratory of Marine Geology, Tongji University, 1239 Siping Road, Shanghai, 200092, China. E-mail addresses: [email protected] (F. Chen), [email protected] (Q. Lao), [email protected] (G. Jia), [email protected] (C. Chen), [email protected] (Q. Zhu), [email protected] (X. Zhou). ∗
https://doi.org/10.1016/j.atmosenv.2018.09.061 Received 10 May 2018; Received in revised form 11 September 2018; Accepted 30 September 2018 Available online 05 October 2018 1352-2310/ © 2018 Elsevier Ltd. All rights reserved.
Atmospheric Environment 196 (2019) 1–9
F. Chen et al.
Fig. 1. The sampling location of Zhanjiang in southern China (black dot). SM: summer monsoon; WM: winter monsoon.
oxidation pathways also control δ18OeNO3. According to Fang et al. (2010), in the pathway of OH oxidation, 2/3 of O arises from O3 and 1/ 3 of O from OH, while in the pathway of O3 oxidation 5/6 of O arises from O3 and 1/6 of O from OH. Atmospheric O3 is 18O enriched (δ18O ranging from +90 to +120‰) (Krankowsky et al., 1995), whereas the δ18O of OH reflect the isotopic composition of water vapor in the troposphere, exhibiting low values (ranging from −15‰ to 0‰) (Dubey et al., 1997). In addition to the O3 and OH oxidation pathways, NOx can also react with dimethylsufide (DMS), halogen oxides, and peroxy radicals (Alexander et al., 2009; Fang et al., 2010; Morin et al., 2008) to produce HNO3. Thus, the composition of δ18O in atmospheric nitrate exhibit a larger range than δ15N (Hastings et al., 2003, 2004; Michalski et al., 2003). For example, δ18OeNO3 in atmosphere cover a range of 60.3–73.4‰ in Bavaria (Durka et al., 1994) and 60.3–86.5‰ in Bermuda (Hastings et al., 2003). Interestingly, higher δ18O values of nitrate (87.0‰) were observed at high-latitude (78.7°N) Ny-Alesund (Morin et al., 2009) compared to values (67.0‰) at mid-latitude (43.9°N) New Hampshire (Pardo et al., 2004), which is likely caused by the difference in the NOx oxidation pathway between the two regions, i.e., the lack of photochemical reactions during winter at high latitudes would result in lower concentrations of hydroxyl radical, allowing the O3-involved pathways to be dominant (Hastings et al., 2004). In addition, the nitrate δ18O values (65.0‰) at a lower-latitude (23.2°N) city of China have been reported to be even lower than those in New Hampshire and Greenland (Fang et al., 2010). Although the dual nitrate isotopes in wet deposition have been reported extensively (Elliott et al., 2007; Fang et al., 2010; Hastings et al., 2003; Hastings et al., 2004; S. Morin et al., 2009; S Morin et al., 2008), little work has been conducted in the tropical regions characterized by long sunshine hours and stronger radiation under warm conditions (Fang et al., 2010), which could give rise to special patterns of NOx oxidation pathways. We hypothesize that more abundant photochemistry in tropical regions may cause lower δ18O values of atmospheric nitrate. To verify this speculation, a time-series investigation of δ15N and δ18O of rainwater nitrate has been made for more than 2 years at the Zhanjiang city in the southernmost part of mainland China. And here is the report of our results of this investigation.
(R2) NO2 + hv → NO + O (R3) O + O2 → O3 (R4) NO2 + OH → HNO3 (R5) NO2 + O3 → NO3 + O2 (R6) NO3 + NO2 → N2O5 (R7) N2O5 + H2O → 2HNO3 Previous studies suggest that stable isotopes of nitrogen (δ15N) and oxygen (δ18O) in nitrate can serve as a tool for distinguishing the contributions from various sources to atmospheric nitrogen and pathways of NOx oxidation (Durka et al., 1994; Elliott et al., 2007; Hastings et al., 2003, 2004; Pardo et al., 2004; Russell et al., 1998). The δ15N values from different NOx sources vary over a large range. For example, anthropogenic sources are characterized higher δ15N values, e.g. from +6 to +13‰ from coal combustion (Heaton, 1990) and between +3.7 and + 5.7‰ from vehicle NOx emissions (Ammann et al., 1999; Moore, 1977; Pearson et al., 2000). Comparatively, lower δ15N values have been reported from natural sources, such as lightning (∼0‰) (Hoering, 1957) and biogenic emissions (−49 to −20‰) (Li and Wang, 2008). Thereby, different sources could lead to different isotope signatures in atmospheric NO3−. In spite of this, isotope values of atmospheric NO3− may be also affected by the two main pathways of NOx oxidation. In the environments where NOx levels are usually low and sunshine durations are long enough, the unidirectional oxidation of NO2 through the OH radical during the day play a dominant role, which would lead to little change in δ15N-NOx values (Freyer et al., 1993). In contrast, in the night reactions using O3 as the oxidant (R5), isotope exchange at equilibrium occurs, causing 15N-enrichment in the higher N oxidation states, and whether the δ15NeNO3- value may reflect δ15N of source NOx is determined by the degree of NOx consumption. In polluted areas where the level of NOx is higher than O3 (Freyer et al., 1993), the resulted NO3− would be more enriched in 15N than NOx (Freyer et al., 1993). On the other hand, in environments with low NOx/O3 ratios, NOx can be completely oxidized to NO3−, resulting in δ15NeNO3- inheriting the δ15N signature of the NOx (Freyer et al., 1993). In addition to δ15N, NOx 2
Atmospheric Environment 196 (2019) 1–9
F. Chen et al.
1.8
600
NO3--N Precipitation during sampling periods Mean precipitation (1981-2010)
1.6
500
1.2
400
1.0 300 0.8 0.6
Rainfall (mm)
-1
Concentration (mg L )
1.4
200
0.4 100 0.2 0 May-15 Jun-15 Jul-15 Aug-15 Sep-15 Oct-15 Nov-15 Dec-15 Jan-16 Feb-16 Mar-16 Apr-16 May-16 Jun-16 Jul-16 Aug-16 Sep-16 Oct-16 Nov-16 Dec-16 Jan-17 Feb-17 Mar-17 Apr-17 May-17 Jun-17 Jul-17 Aug-17 Sep-17 Oct-17 Nov-17 Dec-17
0.0
Month Fig. 2. The variations of Monthly rainfall, mean rainfall from 1981 to 2010 and nitrate concentration. The lack of data in Dec-2016, Jan-2017 and Dec-2017 is because very little (< 0.5 mm) or no rainfall occurred during these periods.
2. Materials and methods
prior to sample loading. Similar procedures have been conducted in numerous studies (e.g., Fang et al., 2010; Hastings et al., 2003; Hastings et al., 2004; Jia and Chen, 2010; Savarino et al., 2006; Yeatman et al., 2001). A total of 208 rain events were sampled and the precipitation samples in each month were proportionally mixed into a large sample according to the rainfall amount in the month, thus composing a representative sample of wet deposition for the whole month. Therefore, the analysis results of the mixed samples are monthly weighted averages.
2.1. Sampling site and sample collection Zhanjiang city (20.00°–21.58°N, 109.52°–110.92°E) is located in southwestern Guangdong Province in southern China and is close to the South China Sea (Fig. 1). The local climate is characterized by tropical monsoon, exhibiting hot, humid conditions under the influence of southerly summer monsoon and cool, dry conditions dominated by northerly winter monsoon. The mean annual rainfall and temperature in Zhanjiang are 1689 mm and 23 °C, respectively (http://data.cma.cn/ ). Sunshine duration (∼2000 h) in this tropical city is relatively long in the world due to the low latitude (http://data.cma.cn/, https://en. wikipedia.org/wiki/Sunshine_duration). April is a transition month from the winter monsoon to the summer monsoon, during which rainfall usually increases sharply (Fig. 2). From May to September, there is a quasi-stationary front over southern China; it usually brings strong and sustained rainfalls due to the convergence of the northern cold air mass and the southern moisture. Zhanjiang is one of the rapidly developing cities in China. In recent years, the economy, industry, agricultural productivity and population have increased rapidly in the city. For example, the power production, which mainly comes from coal combustion, increased obviously from 9.41 × 109 kW h in 2015 to 16.46 × 109 kW h in 2017 (data from Zhanjiang Bureau of Statistics, http://zjtj.zhanjiang.gov.cn). The population was approximately 5.4 million in 1990 and increased to 7.2 million in 2015. The industrial production in Zhanjiang also rose rapidly from 2002 to 2013, and the city is facing potential nitrogen pollution (Meng et al., 2015). From May 2015 to November 2017, precipitation samples were collected on the top of a building at Guangdong Ocean University, located in the seaside city of Zhanjiang, using an acid-washed polyethylene bucket (80 cm in diameter) as the collector. The collector was placed at sampling site before rain to wait for a rain event. The rainfall less than 0.5 mm was considered invalid, and not collected. Each precipitation sample was filtered with a glass fiber filter (GF/F, Whatman, 47 mm in diameter), and then stored in a polyethylene bottle under −20 °C. The polyethylene bottles were soaked with 30% (v/v) HCl for 24 h, then cleaned with ultrapure water and dried in the laboratory
2.2. Sample analysis In the lab, NO3− concentration of rainwater was measured by a San++ continuous flow analyzer (Skalar, Netherlands). After nitrite was removed by adding sulfamic acid (Granger and Sigman, 2009), isotopic analysis of dissolved NO3− was conducted following a chemical conversion method modified from McIlvin and Altabet (2005). According to this method, NO3− was reduced to nitrite with spongy cadmium and further reduced to nitrous oxide with sodium azide in an acetic acid buffer. Subsequently, nitrous oxide was separated, purified and analyzed for δ15N and δ18O with a GasBench II-MAT 253. International standard IAEA-N3 was used to calibrate δ15N and δ18O. The analysis deviation for the standard was less than 0.2‰ for δ15N and less than 0.5‰ for δ18O. The reproducibility of duplicate sample analyses was less than 0.6‰ for δ18O (average ± 0.3‰) and less than 0.3‰ for δ15N (average ± 0.1‰). 2.3. Data sources and calculations The data of monthly sunshine duration from May 2015 to November 2017 and historical mean rainfall from 1981 to 2010 were from the China Meteorological Data Sharing Service System (http://data.cma. cn/); the monthly atmospheric NO2 and O3 were from historical data of air quality in Zhanjiang (https://www.aqistudy.cn/historydata/). Annual N load from wet deposition was calculated from volumeweighted-mean concentrations of NO3− and rainfall.
3
Atmospheric Environment 196 (2019) 1–9
F. Chen et al.
Fig. 3. Comparisons of nitrate isotope values with atmospheric parameters in Zhanjiang. (a) Monthly sunshine duration during sampling periods; (b) monthly atmospheric NO2/O3 ratio; (c) monthly variations of δ15NeNO3-; (d) monthly variations of δ18OeNO3-. The lack of data in Dec-2016, Jan-2017 and Dec-2017 is because very little (< 0.5 mm) or no rainfall occurred during this period.
2.4. Backward trajectories
follows (Xue et al., 2012; Moore and Semmens, 2010):
To analyze the influence of wind transport, three days (72 h) back trajectories were computed utilizing the web version model of the Hybrid Single-Particle Lagrangian Intergrated Trajectory (HYSPLIT) (https://ready.arl.noaa.gov/HYSPLIT_traj.php). Trajectories were generated from the sampling site at the heights 500 m above sea level. The trajectory time of 72 h was chosen because the NOx and the HNO3 lifetime in the boundary layer are generally less than 3 days (Fang et al., 2010; Liang et al., 1998).
k
Xij =
Pk (Sjk + cjk ) + k=1
ij
(1)
Sjk ∼ N (μjk, ω2jk) cjk ∼ N (λjk, τ2jk) εjk ∼ N (0, σ2j ) where Xij represents the isotope values j of the sample i, in which i = 1, 2, 3, …, N and j = 2 (δ15NeNO3- and δ18OeNO3); Sjk represents the isotope value j of the source k (k = 1, 2, 3, …, K), and Sjk is normally distributed with average μjk, and standard deviation ωjk; Pk represents the contribution of source k, which needs to be calculated by the SIAR model; cjk represents the fractionation factor for δ15NeNO3- and δ18OeNO3 on source k, and cjk is normally distributed with average λjk and standard deviation τjk; εjk represents the residual error of the
2.5. SIAR mixing model To calculate the relative contribution of nitrate sources, the Bayesian stable isotope mixing model was conducted in the software package SIAR (stable isotope analysis in R). In this model, the Bayesian framework is used to determine the probability distribution proportion of each source to the mixture. The Bayesian model framework is as 4
Atmospheric Environment 196 (2019) 1–9
F. Chen et al.
additional unquantified variation between individual samples, and εjk is normally distributed with average 0 and standard deviation σj. The model has been successfully used to estimate the proportional contribution of sources (Xue et al., 2012; Moore and Semmens, 2010; Zhang et al., 2018).
4. Discussion 4.1. NO3− concentration The nitrate concentration in wet deposition of Zhanjiang is lower than those in relatively more developed regions in China, such as the Pearl River Delta region (0.5–5.5 mg L−1, 1.2 mg L−1 average) (Jia and Chen, 2010) and the Yangtze River Delta region (0.97 mg L−1 average) (Zhao et al., 2009), but much higher than those in least-disturbed regions, such as the subtropical forested region in southern China (0.02–0.08 mg L−1) (Chen and Mulder, 2007) and the North and South Pacific Oceans (0.01–1.12 mg L−1) (Jung et al., 2011), indicating great influences of human activities (Jia and Chen, 2010). The seasonal distribution pattern of NO3− concentrations, i.e., higher values in dry seasons (Nov. to Mar.) and lower values in rainy seasons (May. to Oct.), is likely caused by the dilution effect of precipitation amount (Fig. 2), but could be also influenced by the shift from winter monsoon to summer monsoon. During winter Siberian high blows from the continent to Zhanjiang, making wet deposition in Zhanjiang to be mainly influenced by the highly polluted continental air mass in China; whereas during summer the southerly oceanic air blows to Zhanjiang is quite clean (Fig. 5). Nevertheless, wet nitrate deposition in wet season (3.04 kg N ha−2) was higher than that dry deposition (1.74 kg N ha−2), likely due to the heavy washout of nitrate in summer. The temporal changes in contributions from different sources could also have affected nitrogen isotopes as discussed later.
3. Results The monthly rainfall from May 2015 to November 2017 is summarized in Fig. 2. The lack of data in December 2016, January and December 2017 is because very little (< 0.5 mm) or no rainfall occurred during this period. Rainfall in certain months was significantly higher than the 30-year monthly mean rainfall between 1981 and 2010 (Fig. 2). Higher rainfalls occurred between April and October (Fig. 2). But there occurred an anomalously high rainfall in January 2016 (368 mm). The monthly weighted-mean concentrations of nitrate ranged from 0.06 to 1.33 mg L−1 and from 0.02 to 1.57 mg L−1 in 2016 and 2017, respectively. The calculated annual volume-weighted-mean concentrations of NO3− according to the precipitation of each month were 0.20 mg L−1 and 0.19 mg L−1 in 2016 and 2017, respectively. The NO3− concentrations in wet deposition at Zhanjiang in 2016 are comparable to those in 2017 (Fig. 2), suggesting a stable atmospheric nitrogen processing in the city. Higher concentrations of nitrate occurred during the dry seasons, whereas lower concentrations during the wet seasons (Fig. 2). According to the precipitation and annual volumeweighted-mean concentrations of NO3− for the two years, the wet N deposition in 2016 and 2017 in Zhanjiang were 4.77 kg N ha−2 and 3.79 kg N ha−2, respectively. δ15NeNO3- values ranged from −1.8‰ in July 2016 to +4.1‰ in March 2017, exhibiting lower mean values in hot rainy seasons (i.e., 0.2‰) than in cool dry seasons (i.e., 2.2‰). The δ18OeNO3- values ranged from +42.7 to +61.6‰, also showing higher values in dry seasons and lower values in wet seasons (Fig. 3). The distribution patterns of δ15NeNO3- and δ18OeNO3- values were opposite to the pattern of sunshine; the latter exhibited an average of 199 h per month in wet season but an average of only 114 h per month in dry season (Fig. 3). Both δ15NeNO3- and δ18OeNO3- also showed significant relationships with rainfall (Fig. 4). The volume-weighted annual mean values of δ15NeNO3- and δ18OeNO3- were +0.8‰ and +52.4‰, respectively.
4.2. Influencing factors of nitrate isotopes δ15NeNO3- and δ18OeNO3- in our results exhibited higher values in the dry seasons and lower values in the wet seasons (Fig. 3), which is similar to previous studies (Fang et al., 2010; Hastings et al., 2003, 2004; Savarino et al., 2006; Yeatman et al., 2001). The paralleled seasonal variation of δ15NeNO3- and δ18OeNO3- is intriguing (Fig. 3). As stated above, δ18OeNO3- is mainly determined by the NOx oxidation pathways, whereas δ15NeNO3- could be modulated either by NOx source or by NOx oxidation pathways. So, it seems that NOx oxidation pathways might have determined synchronous variations of the dual isotopes. To explore whether the NOx oxidation pathways have accounted for the seasonal variations of nitrate dual isotopes in this study, we compared contemporaneous sunshine duration and NO2/O3 ratio with the dual isotope data in Zhanjiang (Fig. 3). As shown in Fig. 3b, the level of
Fig. 4. Correlation of (a) δ15N and (b) δ18O of wet nitrate deposition with rainfall.
5
Atmospheric Environment 196 (2019) 1–9
F. Chen et al.
Fig. 5. Backward air mass trajectories in precipitation events in Zhanjiang, based on NOAA HYSPLIT model back trajectories; the red lines denote air mass trajectories occurring in the wet season and the blue lines denote trajectories in the cool dry seasons. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)
O3 in Zhanjiang was much higher than that of NO2, and the NO2/O3 ratio ranged from 0.11 to 0.24, with an average of 0.18 ± 0.04 in dry season and 0.16 ± 0.02 in wet season and not showing significant differences between wet and dry seasons. Thus, the discussions about other variables such as temperature that might also affect NO2/O3 ratio were not included. The NO2/O3 ratio in Zhanjiang is significantly low compared to some previous studies, in which most NO2/O3 ratio values are > 1.0 and vary greatly (e.g., Fang et al., 2010; Freyer et al., 1993; Hastings et al., 2003; Hastings et al., 2004; Jiaand Chen, 2010). In contrast, Sunshine duration (∼2000 h) in Zhanjiang is relatively long in the world due to the low latitude and the clean air caused by the ocean climate (http://data.cma.cn/, https://en.wikipedia.org/wiki/Sunshine_ duration). In addition, the curve of sunshine hours, longer in summer and shorter in winter, corresponds well to the time-series variation of δ15NeNO3- and δ18OeNO3- values (Fig. 3). Therefore, although NOx could be completely oxidized to NO3− (via N2O5) in the atmospheric environments with such low NO2/O3 ratios (Freyer et al., 1993), we hypothesize that the formation of nitrate via the OH radical, i.e., the day reactions, is likely responsible for the seasonal variation of the nitrate isotope values due to long sunshine in this tropical area. The rainfall may be another factor influencing the seasonal patterns in δ15NeNO3- and δ18OeNO3-, since both δ15NeNO3- and δ18OeNO3- of wet deposition were positively correlated to rainfall (Fig. 4). We thereby hypothesize that low precipitation in the dry seasons could have resulted in low relative washout of nitrate, leading 15N and 18O to be preferentially washed out, i.e., higher δ15NeNO3- and δ18OeNO3values in wet deposition. In contrast, during wet seasons, complete nitrate removal are expected, leading δ15NeNO3- and δ18OeNO3- in wet deposition to be similar to the source nitrate isotopes in atmosphere. This suggestion is in agreement with the previous findings that during NO3− gas-liquid and particle-gas interactions, heavy isotope is preferentially incorporated into the condensed phase that is readily washed out, thus causing the successive precipitation progressively depleted in the dual isotopes of NO3− (Moore, 1977; Heaton, 1986). Generally, δ18O values in OH radical are from −15‰ to 0‰ (Fang
et al., 2010), and in O3 from +90‰ to +122‰ (Krankowsky et al., 1995). Accordingly, the minimum values (−15‰ for OH radical and +90‰ for O3) were chosen to calculate the minimum δ18OeNO3- value of atmospheric nitrate, which was around +55‰ if 2/3 of nitrate-O arises from O3 and 1/3 of O from OH; similarly, using the maximum theoretical values (0‰ for OH radical and +122‰ for O3), the maximum δ18OeNO3- value of atmospheric nitrate was +102‰ if 5/6 of nitrate-O arises from O3 and 1/6 of O from OH (Fang et al., 2010). The δ18OeNO3- values in our samples, i.e. 52.4‰ in average, are close to the minimum theoretical value of +55‰, suggesting the minimum endmember values are suitable for calculation in this study. However, some samples in our study had δ18OeNO3- being lower than +55‰. We hypothesized that there might be other pathways to decrease δ18OeNO3-. As stated above, in addition to the two main channels (the OH pathway and the O3 pathway) for atmospheric NO3− formation, which account for more than 94% of atmospheric NO3− globally (Alexander et al., 2009), NOX can also react with dimethylsufide (DMS), halogen oxides (i.e. bromine oxide (BrO)), and peroxy radicals (Alexander et al., 2009; Fang et al., 2010; Morin et al., 2008) to produce atmospheric NO3−. However, these factors account for less than 6% of the annual nitrate production in the world (Alexander et al., 2009). The DMS and halogen oxides pathways can be ruled out as the causes of the negative δ18OeNO3- anomalies, because atmospheric NO3− induced via the two pathways would probably have higher 18O values than those induced via the OH pathway and the O3 pathway due to its more involvement with O3 during reactions (Morin et al., 2008; Alexander et al., 2009). One possibility for the lower 18O values of NO3− may be the reaction of NOx with peroxy radicals (HO2 and its organic homologues RO2), because the O atoms in peroxy radicals mainly come from atmospheric O2 rather than O3 andδ18O of O2 is around +23.9‰ (Fang et al., 2010). Thus, the resultant δ18OeNO3- value via the oxidation pathway by peroxy radicals was around +28‰ followed by 4/6 of the O atom from atmospheric O2, 1/6 from O3 and 1/6 O from H2O (Alexander et al., 2009). Thereby, the oxidation pathway by peroxy radicals may be partly responsible for the low δ18OeNO3- value in our 6
Atmospheric Environment 196 (2019) 1–9
F. Chen et al.
Among the anthropogenic sources of NOx, fossil fuel combustion, particularly coal combustion, is likely dominant in Zhanjiang, because coal is commonly used to generate electricity in this economically developing area. This suggestion agrees with the bottom-up inventory, which revealed that, more than 70% of the total NOx emission was derived from coal combustion in China (Tian et al., 2001). It is also agreeable with the previous study in Guangzhou City, which is only 400 km from Zhanjiang (Fang et al., 2010). Natural sources include lightning and biogenic emissions (Fang et al., 2010). The lightning-derived contribution to the atmospheric NOx is important above the South China Sea (Price et al., 1997; Xiao et al., 2015), and likely among the highest in the world (Price et al., 1997). The mean values of δ15NeNO3- during rainy season (+0.15‰) was close to that for the lightning (around 0‰) (Hoering, 1957). In addition, the negative values of δ15NeNO3- appeared in most months during the warm season in 2015 and 2016, suggesting contributions from biogenic soil emissions that has been considered as an important source in tropical areas (Felix and Elliott, 2014; Li and Wang, 2008). Here, relative contributions of the above sources, i.e., anthropogenic sources including vehicle emission (−2 to −13‰ range, −7.0 ± 4.7‰ average) and coal combustion (+6.0 to +13.0‰ range, 9.6 ± 2.9‰ average) (Heaton, 1990) and natural sources including soil emission (−19.8 to −48.6‰ range, −21.9 ± 9.3‰ average) and lightning (−0.5 to +1.4‰ range, 0 ± 0.7‰ average) (Hoering, 1957; Li and Wang, 2008) were estimated for the wet deposition using a Bayesian model. The SIAR mixing model outputs showed high variabilities in contributions of the four potential nitrate sources (Fig. 7). The contribution of lightning is highest (mean probability estimate (MPE) between 29% and 82%, average of 56%), followed by coal combustion (MPE between 10% and 43%, average of 26%), vehicle emission (MPE between 0% and 27%, average of 13%) and soil emission (MPE between 0% and 10%, average of 5%). Generally, the estimates show that the contribution of natural sources for atmospheric nitrate deposition dominated in Zhanjiang (61%). As to the atmospheric nitrate in dry seasons, in addition to the incomplete nitrate washout, the cause of relatively higher δ15NeNO3values (+2.2‰) could be also associated with the polluted NOx carried by the northerly winds during winter, whose contribution is difficult to estimate in this study. Therefore, our results indicate that δ15NeNO3- in precipitation in wet season is more applicable for source apportionment than in dry season in Zhanjiang because of minimal isotope fractionation during the formation of atmospheric nitrate.
Fig. 6. Variation characteristics of δ18OeNO3- values in nitrate of wet deposition with latitude. (a) The δ18OeNO3- values in Antarctica (Savarino et al., 2006); (b) results from our study (Zhanjiang); (c) Guangzhou, South China (Fang et al., 2010); (d) Bermuda (Hastings et al., 2003); (e) New Hampshire, US (Pardo et al., 2004); (f) Greenland (Hastings et al., 2004); (g) Ny-Alesund (Morin et al., 2009).
study, which is similar to a previous point in a mid-low latitude city (Fang et al., 2010). In our study, δ18OeNO3- in the dry seasons (56.8‰ on average) was higher than that in the wet seasons (50.4‰ on average). By assuming 6% contribution of the NOx oxidation pathway to atmospheric NO3 (Alexander et al., 2009), the relative contribution of NOx oxidation via the OH radical and O3 can be calculated for the two seasons, showing that the OH radical pathway was accounted for averagely 87% in the dry seasons and more than 94% in the wet seasons. However, the calculated contribution of OH radical pathway during the dry seasons is likely underestimated, since incomplete washout of nitrate could have resulted in higher δ18OeNO3- values in wet deposition during the dry seasons. Nevertheless, our estimates indicate that the day reactions dominate the atmospheric NO3− formation throughout the year in Zhanjiang with neglectable contribution from night reactions. As mentioned above, the predominance of day reactions for the atmospheric NO3− formation suggested in this study is different from previous studies at higher latitudes. Thereby, we hypothesize whether the δ18OeNO3- values increase with climate and latitude. When available δ18OeNO3- values are plotted versus latitude (Fig. 6), as expected, they indeed increase with latitude, with the highest values in the polar region and the lowest values (this study) in the tropic. The lower values of δ18OeNO3- in wet deposition in this study indicate the main contribution of OH radical pathways to the formation of atmospheric nitrate (Elliott et al., 2015; Fang et al., 2010; Hastings et al., 2003, 2004). This is largely because the tropical Zhanjiang has a longer summer time and longer sunshine. This view is further supported by previous studies, which have demonstrated using δ17O model that the pathway of nitrate production by the OH radical is predominant (reaching 87%) in the tropical areas, which have the highest level of OH (Alexander et al., 2009). In contrast, longer wintertime and less sunshine at the mid-high latitudes may lead to more contribution from night reactions using O3 as oxidant and less dilution by the day reactions, which could result in higher δ18OeNO3- values compared to the mid-low latitude areas. 4.3. Source variation revealed by monthly δ15NeNO3- changes The complete nitrate removal during rainfall washout in wet seasons suggests that δ15NeNO3- in wet deposition can be applied to identify the NOx sources during these seasons. Generally, the atmospheric NOx has anthropogenic and natural sources (Fang et al., 2010).
Fig. 7. Percent contribution of four potential nitrate sources for wet season deposition estimated by SIAR. 7
Atmospheric Environment 196 (2019) 1–9
F. Chen et al.
5. Conclusions
359–366. Freyer, H.D., Kley, D., Volz-Thomas, A., Kobel, K., 1993. On the interaction of isotopic exchange processes with photochemical reactions in atmospheric oxides of nitrogen. J. Geophys. Res. 98 (D8), 14791–14796. https://doi.org/10.1029/93JD00874. Granger, J., Sigman, D.M., 2009. Removal of nitrite with sulfamic acid for nitrate N and O isotope analysis with the denitrifier method. Rapid Commun. Mass Spectrom. 23 (23), 3753–3762. https://doi.org/10.1002/rcm.4307. Hastings, M.G., Sigman, D.M., Lipschultz, F., 2003. Isotopic evidence for source changes of nitrate in rain at Bermuda. J. Geophys. Res. 108 (D24). https://doi.org/10.1029/ 2003JD003789. Hastings, M.G., Steig, E.J., Sigman, D.M., 2004. Seasonal variations in N and O isotopes of nitrate in snow at Summit, Greenland: implications for the study of nitrate in snow and ice cores. J. Geophys. Res. 109 (D20). https://doi.org/10.1029/2004JD004991. Heaton, T.H.E., 1986. Isotopic studies of nitrogen pollution in the hydrosphere and atmosphere: a review. Chem. Geol. 59 (2), 87–102. https://doi.org/10.1016/00092541(86)90046-X. Heaton, T.H.E., 1990. 15N/14N ratios of NOx from vehicle engines and coal-fired power stations. Tellus 42 (3), 304–307. https://doi.org/10.3402/tellusb.v42i3.15223. Hoering, T., 1957. The isotopic composition of the ammonia and the nitrate ion in rain. Geochem. Cosmochim. Acta 12 (1), 97–102. https://doi.org/10.1016/0016-7037(57) 90021-2. Jarvis, J.C., Steig, E.J., Hastings, M.G., Kunasek, S.A., 2008. Influence of local photochemistry on isotopes of nitrate in Greenland snow. Geophys. Res. Lett. 35 (21), L21804. https://doi.org/10.1029/2008GL035551. Jia, G.D., Chen, F.J., 2010. Monthly variations in nitrogen isotopes of ammonium and nitrate in wet deposition at Guangzhou, south China. Atmos. Environ. 44 (19), 2309–2315. https://doi.org/10.1016/j.atmosenv.2010.03.041. Jung, J., Furutani, H., Uematsu, M., 2011. Atmospheric inorganic nitrogen in marine aerosol and precipitation and its deposition to the North and South Pacific Oceans. J. Atmos. Chem. 68 (2), 157–181. https://doi.org/10.1007/s10874-012-9218-5. Krankowsky, D., Bartecki, F., Klees, G.G., Mauersberger, K., Schellenbach, K., Stehr, J., 1995. Measurement of heavy isotope enrichment in tropospheric ozone. Geophys. Res. Lett. 22 (13), 1713–1716. https://doi.org/10.1029/95GL01436. Li, D., Wang, X., 2008. Nitrogen isotopic signature of soil-released nitric oxide (NO) after fertilizer application. Atmos. Environ. 42 (19), 4747–4754. https://doi.org/10.1016/ j.atmosenv.2008.01.042. Liang, J.Y., Horowitz, L.W., Jacob, D.J., Wang, Y.H., Fiore, A.M., Logan, J.A., Gardner, G.M., Munger, J.W., 1998. Seasonal budgets of reactive nitrogen species and ozone over the United States, and export fluxes to the global atmosphere. J. Geophys. Res. 103 (D11), 13435–13450. https://doi.org/10.1029/97JD03126. Luo, Y., Liu, S., Fu, S.L., Liu, J., Wang, G., Zhou, G., 2008. Trends of precipitation in beijiang river basin, Guangdong Province, China. Hydrol. Process. 22 (13), 2377–2386. https://doi.org/10.1002/hyp.6801. Meng, Y.F., Chen, F.J., Zeng, Z., 2015. Nitrogen budget evolution in zhanjiang municipal district and its environmental effect during 2002-2013. Trop. Geogr. 35 (5), 777–784. in Chinese with English abstract. https://doi.org/10.13284/j.cnki.rddl.002752. McIlvin, M.R., Altabet, M.A., 2005. Chemical conversion of nitrate and nitrite to nitrous oxide for nitrogen and oxygen isotopic analysis in Freshwater and seawater. Anal. Chem. 77, 5589–5595. https://doi.org/10.1021/ac050528s. Michalski, G., Scott, Z., Kabiling, M., Thiemens, M.H., 2003. First measurements and modeling of Δ17O in atmospheric nitrate. Geophys. Res. Lett. 30 (16), 1870. https:// doi.org/10.1029/2003GL017015. Moore, H., 1977. The isotopic composition of ammonia, nitrogen dioxide and nitrate in the atmosphere. Atmos. Environ. 11 (12), 1239–1243. https://doi.org/10.1016/ 0004-6981(77)90102-0. Moore, J.W., Semmens, B.X., 2010. Incorporating uncertainty and prior information into stable isotope mixing models. Ecol. Lett. 11 (5), 470–480. https://doi.org/10.1111/j. 1461-0248.2008.01163.x. Morin, S., Savarino, J., Frey, M.M., Domine, F., Jacobi, H.W., Kaleschke, L., Martins, M.F., 2009. Comprehensive isotopic composition of atmospheric nitrate in the Atlantic Ocean boundary layer from 65°S to 79°N. J. Geophys. Res. 114 (D05303). https:// doi.org/10.1029/2008JD010696. Morin, S., Savarino, J., Frey, M.M., Yan, N., Bekki, S., Bottenheim, J.W., 2008. Tracing the origin and fate of NOx in the Arctic atmosphere using stable isotopes in nitrate. Science 322 (5902), 730–732. https://doi.org/10.1126/science.1161910. Pardo, L.H., Kendall, C., Pett-Ridge, J., Chang, C.C.Y., 2004. Evaluating the source of streamwater nitrate using δ15N and δ18O in nitrate in two watersheds in New Hampshire, USA. Hydrol. Process. 18 (14), 2699–2712. https://doi.org/10.1002/ hyp.5576. Pearson, J., Wells, D.M., Seller, K.J., Bennett, A., Soares, A., Woodall, J., Ingrouille, M.J., 2000. Traffic exposure increases natural 15N and heavy metal concentrations in mosses. New Phytol. 147 (2), 317–326. https://doi.org/10.1046/j.1469-8137.2000. 00702.x. Price, C., Prather, M., 1997. NOx from lightning .1. Global distribution based on lightning physics. J. Geophys. Res. 102 (D5), 5929–5941. https://doi.org/10.1029/ 96JD03504. Russell, K.M., Galloway, J.N., Macko, S.A., Moody, J.L., Scudlark, J.R., 1998. Sources of nitrogen in wet deposition to the Chesapeake Bay region. Atmos. Environ. 32 (14), 3923–3927. https://doi.org/10.1016/S1352-2310(98)00044-2. Savarino, J., Kaiser, J., Morin, S., Sigman, D.M., Thiemens, M.H., 2006. Nitrogen and oxygen isotopic constraints on the origin of atmospheric nitrate in coastal Antarctica. Atmos. Chem. Phys. 7 (8), 1925–1945. https://doi.org/10.5194/acpd-6-8817-2006. Tian, H.Z., Hao, J.M., Lu, Y.Q., Zhu, T.L., 2001. Inventories and distribution characteristics of NOx emissions in China. China Environ. Sci. 21 (6), 493–497 in Chinese with English abstract. Xiao, H.W., Xie, L.H., Long, A.M., Ye, F., Pan, Y.P., Li, D.N., Long, Z.H., Chen, L., Xiao,
δ18OeNO3- values in wet deposition in the tropical city of Zhanjiang manifest the pathway of NOx oxidation via the OH radical was predominant (87% in dry seasons and 94% in wet seasons) for the nitrate formation. This might be related to the long summer time and strong sunshine in this tropical area, which could be further demonstrated by the finding of an increasing trend of δ18OeNO3- with latitude, reflecting a shift of the predominant NOx oxidation pathways from the day reactions in low latitudes to night reactions in high latitudes. Rainfall washout exerted an effect on δ15NeNO3- and δ18OeNO3- values. During wet seasons, complete nitrate washout suggests that δ15NeNO3- in precipitation should be similar to the nitrate formed in the atmosphere, and thereby can be used for source apportionment. A Bayesian model showed that the contributions of natural and anthropogenic sources for atmospheric nitrate deposition were 61% and 39%, respectively. Acknowledgments This work was supported by National Natural Science Foundation of China (41476066), Guangdong Natural Science Foundation of China (2016A030312004), International Science and technology cooperation project (GASI-IPOVAI-04), National Key Research and Development Plan (2016YFC1401403), Project of Enhancing School with Innovation of Guangdong Ocean University (GDOU2014050201, GDOU2013010203,GDOU2013050201). The authors would like to thank Xijie Yin for technical support in Third Institute of Oceanography, State Oceanic Administration. References Alexander, B., Hastings, M.G., Allman, D.J., Dachs, J., 2009. Quantifying atmospheric nitrate formation pathways based on a global model of the oxygen isotopic composition (Δ17O) of atmospheric nitrate. Atmos. Chem. Phys. 9 (14), 5043–5056. https:// doi.org/10.5194/acp-9-5043-2009. Ammann, M., Siegwolf, R., Pichlmayer, F., Suter, M., Saurer, M., Brunold, C., 1999. Estimating the uptake of traffic-derived NO2 from 15N abundance in Norway spruce needles. Oecologia 118 (2), 124–131. https://doi.org/10.1007/s004420050710. Calvert, J.G., Lazrus, A., Kok, G.L., Heikes, B.G., Walega, J.G., Lind, J., Gantrell, C.A., 1985. Chemical mechanisms of acid generation in the troposphere. Nature 317 (6032), 27–35. https://doi.org/10.1038/317027a0. Chen, F.J., Jia, G.D., Chen, J.Y., 2009. Nitrate sources and watershed denitrification inferred from nitrate dual isotopes in the Beijiang River, south China. Biogeochemistry 94 (2), 163–174. https://doi.org/10.1007/s10533-009-9316-x. Chen, X.Y., Mulder, J., 2007. Atmospheric deposition of nitrogen at five subtropical forested sites in South China. Sci. Total Environ. 378 (3), 317–330. https://doi.org/ 10.1016/j.scitotenv.2007.02.028. Cui, J., Zhou, J., Peng, Y., He, Y., Yang, H., Mao, J., 2014. Atmospheric wet deposition of nitrogen and sulfur to a typical red soil agroecosystem in Southeast China during the ten-year monsoon seasons (2003-2012). Atmos. Environ. 82 (82), 121–129. https:// doi.org/10.1016/j.atmosenv.2013.10.023. Dubey, M.K., Ralf Mohrschladt, N.M.D., Anderson, J.G., 1997. Isotope specific kinetics of hydroxyl radical (OH) with water (H2O): testing models of reactivity and atmospheric fractionation. J. Phys. Chem. 101 (8), 1494–1500. https://doi.org/10.1021/ jp962332p. Durka, W., Schulze, E.D., Gebauer, G., Voerkeliust, S., 1994. Effects of forest decline on uptake and leaching of deposited nitrate determined from 15N and 18O measurements. Nature 372 (6508), 765–767. https://doi.org/10.1038/372765a0. Elliott, E.M., Kendall, C., Boyer, E.W., Burns, D.A., Lear, G.G., Golden, H.E., Harlin, K., Bytnerowicz, A., Butler, T.J., Glatz, R., 2015. Dual nitrate isotopes in dry deposition: utility for partitioning NOx source contributions to landscape nitrogen deposition. J. Geophys. Res. 114 (G4), 425–453. https://doi.org/10.1029/2008JG000889. Elliott, E.M., Kendall, C., Wankel, S.D., Burns, D.A., Boyer, E.W., Harlin, K., Bain, D.J., Butler, T.J., 2007. Nitrogen isotopes as indicators of NO(x) source contributions to atmospheric nitrate deposition across the midwestern and northeastern United States. Environ. Sci. Technol. 41 (22), 7661–7667. https://doi.org/10.1021/es070898t. Fang, Y.T., Koba, K., Wang, X.M., Wen, D.Z., Li, J., Takebayashi, Y., Liu, X.Y., Yoh, M., 2010. Anthropogenic imprints on nitrogen and oxygen isotopic composition of precipitation nitrate in a nitrogen-polluted city in southern China. Atmos. Chem. Phys. 11, 1313–1325. https://doi.org/10.5194/acp-11-1313-2011. Fang, Y.T., Wang, X.M., Zhu, F.F., Wu, Z.Y., Li, J., Zhong, L.J., Chen, D.H., Yoh, M., 2013. Three-decade changes in chemical composition of precipitation in Guangzhou city, southern China: has precipitation recovered from acidification following sulphur dioxide emission control? Tellus B 65 (9), 134–138. Felix, J.D., Elliott, E.M., 2014. Isotopic composition of passively collected nitrogen dioxide emissions: vehicle, soil and livestock source signatures. Atmos. Environ. 92,
8
Atmospheric Environment 196 (2019) 1–9
F. Chen et al. H.Y., Liu, C.Q., 2015. Use of isotopic compositions of nitrate in TSP to identify sources and chemistry in South China Sea. Atmos. Environ. 109, 70–78. https://doi. org/10.1016/j.atmosenv.2015.03.006. Xue, D., De Baets, B., Van Cleemput, O., Hennessy, C., Berglund, M., Boeckx, P., 2012. Use of a Bayesian isotope mixing model to estimate proportional contributions of multiple nitrate sources in surface water. Environ. Pollut. 161 (1), 43–49. https://doi.org/10. 1016/j.envpol.2011.09.033. Yeatman, S.G., Spokes, L.J., Dennis, P.F., Jickells, T.D., 2001. Comparisons of aerosol nitrogen isotopic composition at two polluted coastal sites. Atmos. Environ. 35 (7),
1307–1320. https://doi.org/10.1016/S1352-2310(00)00408-8. Zhang, M., Zhi, Y., Shi, J., Wu, L., 2018. Apportionment and uncertainty analysis of nitrate sources based on the dual isotope approach and a Bayesian isotope mixing model at the watershed scale. Sci. Total Environ. 639, 1175–1187. https://doi.org/ 10.1016/j.scitotenv.2018.05.239. Zhao, X., Yan, X.Y., Xiong, Z.Q., Xie, Y.X., Xing, G.X., Shi, S.L., Zhu, Z.L., 2009. Spatial and temporal variation of inorganic nitrogen wet deposition to the Yangtze River Delta Region, China. Water Air Soil Pollut. 203 (1–4), 277–289. https://doi.org/10. 1007/s11270-009-0011-2.
9