1 Introduction

In the last few decades, there has been significant progress in technology within the agricultural, industrial, and automobile sectors of South Korea and East Asian countries. These advancements have played a crucial role in driving economic growth and improving the quality of life. However, this progress has come at the cost of a deteriorating environmental condition owing to emission of air pollutants, such as ammonia (NH3), particulate matter (PM), and carbon monoxide (CO) from various sectors (Mukhopadhyay & Forssell, 2005; Singh et al., 2021; Stern, 2015). Atmospheric NH3, a major pollutant generated by agricultural activities, contributes significantly to the formation of aerosols in the atmosphere (Huang et al., 2011; Sigurdarson et al., 2018). These aerosols, formed through reactions between NH3 and organic acids, are mainly responsible for primary air pollution. Moreover, NH3 has a detrimental effect on the health of marine animals, which ultimately affects human lives (Miller et al., 1990). Furthermore, NH3 plays a critical role in the formation of secondary PM1 (aerodynamic diameter ≤ 1 µm), as it combines with acidic substances such as nitric acid (HNO3) and sulfuric acid (H2SO4) to form ammonium salts, contributing to the formation of PM in the atmosphere (Huang et al., 2011). The PM1 mixture comprises several harmful chemical compounds, including sulfates, nitrates, and organic compounds (Perrone et al., 2014). These inhalable particles pose significant health risks and could cause lung-related diseases such as breathing difficulties and pneumonia (Zhang et al., 2020). CO is produced primarily from incomplete fuel combustion by vehicles and industries (Badr & Probert, 1994; Zhao et al., 2018), and is a major pollutant in urban areas. Although CO is the primary pollutant in vehicle exhaust gases, other pollutants emitted from vehicles have been studied in an effort to improve understanding of the nature of vehicle emissions (Tsai et al., 2006). Although research have focused on industrial and agricultural NH3 emissions, vehicle emissions contribute significantly to the total emissions of NH3 in urban areas (Cao et al., 2021). Recent studies have suggested that vehicle emissions, along with NOx (NO + NO2) pollutants, could lead to diverse environmental changes in urban areas (Wang et al., 2020). Such changes cause the rapid growth of new atmospheric pollutants through HNO3 and NH3 condensation, enabling the newly formed particulates to persist in highly polluted environments (Wang et al., 2020).

Numerous studies have investigated atmospheric NH3 concentrations in various regions worldwide, including South Korea. For instance, Kumar et al. (2019) assessed the atmospheric NH3 concentration in Mumbai, the most populated city in India, and estimated the economic effects of the associated human health issues. Liu et al. (2014) researched the contribution of vehicle emissions to NH3 pollution in a traffic-intensive area in an urban tunnel in Guangzhou, China. Chang et al. (2016) verified that vehicle emissions were an important source of NH3 pollution in the Shanghai region of China. Wang et al. (2015) comprehensively studied the effects of atmospheric NH3 on air quality in Shanghai, China. Singh et al. (2021) conducted year-round measurements of atmospheric NH3 in the Seoul metropolitan area, including the urban, suburban, industrial, and agricultural areas. Park et al. (2021) analyzed the role of NH3 in the formation of PM in South Korea.

Measuring atmospheric NH3 concentrations by employing the passive sampler method for long-term data collection is a well-established practice (Clark et al., 2020; Park et al., 2021; Puchalski et al., 2011; Singh et al., 2021). Accordingly, to gain a comprehensive understanding of NH3 trends and the regional characteristics in different seasons, we used a passive sampler to measure NH3 concentrations through all four seasons in South Korea. In addition, we measured and analyzed the concentrations of other air pollutants, such as NOx, CO, and PM1 in the Seoul and West Sea region, which contributed to our understanding of NH3. We consider this study a valuable source for gaining a deeper understanding of the effects of NH3 over the Korean Peninsula.

2 Methods

2.1 Measurement site

2.1.1 National Institute of Environmental Research supersites

We investigated the regional and seasonal variations in NH3 concentrations across the Korean Peninsula by conducting measurements at six supersites operated by the National Institute of Environmental Research (NIER) of Korea, covering both urban and remote areas (Fig. 1). Table 1 shows a summary of the site characteristics, including site type, and longitude and latitude. The measurements of NH3 concentrations were conducted by weekly sampling at the sites for 1 year (June 1, 2020 to July 2, 2021). The Seoul metropolitan area was selected because of its high population density and numerous air pollution sources, including industrial facilities, transportation, and residential activities (Seoul Research Data Service, 2023). The second site is in Daejeon, a major city in the central part of South Korea that serves as a hub for transportation, technology, and industry. This site was chosen to provide information about multiple sources of air pollution in the central region of the country. The Gwangju site was selected to record pollution data from the nearby agricultural and residential areas. The fourth site, Ulsan, the largest industrial city in the country, was chosen primarily for capturing NH3 emissions from industrial sources. The fifth site, Jeju Island in the West Sea, is characterized by a low population and minimal industrial activity, and was chosen primarily to record pollution transported from the surrounding areas by the wind. The sixth site is on the remote Baengnyeong Island, which has minimal human activity and industrial operations and is located approximately 200 km from the Shandong Peninsula of eastern China (Kang et al., 2020; Lee et al., 2015). As this island is close to China, we expected the local NH3 measurements to provide exclusive data on the pollution transported from neighboring countries.

Fig. 1
figure 1

Satellite image of the National Institute of Environmental Research (NIER) station sites chosen for NH3 data collection across the Korean Peninsula

Full size image
Table 1 Detailed information on the location, time periods, and methods used for NH3 measurements in each site. The top six sites are National Institute of Environmental Research (NIER) supersites
Full size table

2.1.2 Roadside measurement site

Vehicular emissions, such as NH3, CO, and NOx, in Seoul were measured at the Hongjimun Tunnel (37.61°N, 126.97°E) on the Naebu Expressway in the Seoul metropolitan area (Figs. 2 and S1). As it is located in a densely populated area, this site experiences a high volume of vehicular traffic throughout the year. Real-time measurements of NH3, CO, and NOx concentrations were conducted outside the tunnel in August (summer) and October (autumn) 2018. In August, the data were collected at 10-min intervals over a 1-month period, and the data were collected in October over every 1-h period.

Fig. 2
figure 2

Satellite images of the vehicle pollution data collection site (roadside) at Hongjimun Tunnel (37.61°N, 126.97°E), Naebu Expressway, Seoul

Full size image

2.1.3 Airborne measurements over the West Sea

Two airborne measurement sorties were conducted to determine the concentration of NH3 in the West Sea. A Beechcraft B1900D aircraft (Textron Aviation, Wichita, Kansas, USA) was flown from Taean to the West Sea of the Korean Peninsula, where a vertical spiral flight was performed to collect data on NH3, CO, and chemical compositions (NH4+, NO3−, SO42−, and organics) in non-refractory PM1 (NR-PM1). During the spiral flight, the altitude of the aircraft varied continuously between 1200 and 300 m. While the NH3 data were collected on June 2 and 5, 2019, the NR-PM1 data were collected only on June 2, 2019. The instruments recorded the data corresponding to the winds that followed different trajectories, as shown in Fig. S2a and b.

2.2 Measurement methods

2.2.1 Passive sampler

We employed the passive sampling method using Radiello NH3 samplers (Sigma-Aldrich, St. Louis, MO, USA) for long-term measurements of NH3 concentrations because of their appropriate features and properties (Singh et al., 2021). The sampler comprises a cylindrical tube (part number RAD1201), which acts as the diffusive body, allowing NH3 gas particles to pass through the tube and be adsorbed by a cartridge soaked with phosphoric acid (H3PO3) (part number RAD168) inside the tube. The entire setup of the passive sampler was placed vertically, with the aid of a vertical adapter (part number RAD122), at a height of 2 m above the ground in an open area, and it was covered with a rain shelter (part number RAD196), as shown in Fig. S3. Temperature and humidity sensors (Lascar Electronics, UK, EasyLog USB, model number EL-USB-2-LCD+) were attached to the setup to monitor the temperature and relative humidity during the measurement periods, as shown in Fig. S4. The NH3 samples collected by passive sampler in the atmosphere were extracted by sonication (Hwashintech, 510 sonicate instrument) for 55 min in 10 mL of 18.2 MΩ cm deionized water, and the extracted NH3 was analyzed using ion chromatography (Singh et al., 2021). The NH3 concentration was calculated using a series of equations involving the physical properties of the passive sampler and the rate of NH3 diffusion in the air (described in the Supplementary material).

2.2.2 Validation of passive samplers measurements

The accuracy of the passive samplers was cross-checked using URG denuders (URG Corporation, Chapel Hill, North Carolina, USA) installed at the Gwangju site. These URG denuders have been used widely in studies for validating passive samplers because of their ability to trap NH3 gas from the atmosphere (Kim et al., 2021; Lee & Tsai, 2008; Li et al., 2017; Singh et al., 2021). The denuders contain a coating of phosphoric acid solution [10 g H3PO3 dissolved in 100 mL 18.2 MΩ cm deionized water and 900 mL methanol (CH3OH)] to collect NH3 gas, which is trapped on a 47 mm Teflon filter (PTFE membrane, pore size = 0.45 μm, Advantec Pall Corporation, Dublin, CA, USA) pack installed in the setup. We used a vacuum pump (Thomas Piston Pump 2660 Series, Gardner Denver Thomas GmbH, Germany) to provide airflow, with the flow rate of the system regulated through an orifice (Pisco 0.4 mm Orifice, Tameson B.V., the Netherlands) and the flow rate set to 1.35 L min−1. The NH3 concentration was calculated using equations that is provided in the Supplementary material (Eqs. S1, S2, and S3). A comparison of NH3 concentrations measured by replicate passive samplers and between the URG denuder and passive sampler at the Gwangju site is shown in Figs. S5 and S6, respectively. The quality control of and assurance analyses for the passive sampler are provided in the Supplementary material (Fig. S7).

2.2.3 Airborne measurements over the West Sea

An aircraft modified for measuring air quality was used for collecting data on NH3 and NR-PM1 concentrations over the West Sea. The Beechcraft B1900D model aircraft, owned by Hanseo University, South Korea, was flown at an altitude of 300–1200 m at a slow speed of approximately 300 km h−1 to collect data over the area (37.168°N, 124.200°E). The equipment components included an aerosol sampling port (Droplet Measurement Technologies, Longmont, Colorado, USA), trace gas inlets (University of California Irvine, USA), and an Aircraft Integrated Meteorological Measurement System (AIMMS-30, Aventech Research Inc., Canada) for measuring temperature, humidity, and barometric pressure. Moreover, the AIMMS-30 collects real-time location and time information using antennas installed on the aircraft (Seo et al., 2019). The NH3 concentrations were analyzed using an EAA-30r-EP analyzer [Los Gatos Research, Inc. (LGR), San Jose, California, USA], adopting off-Axis Integrated Cavity Output Spectroscopy (OA-ICOS) technology. The chemical compositions (NH4+, NO3−, SO42−, and organics) in NR-PM1 were measured using a high-resolution time-of-flight aerosol mass spectrometer (HR-TOF-AMS, Aerodyne Research Inc, Billerica, Massachusetts, USA). In addition, a Serinus 30 CO analyzer (Acoem Ecotech, Australia) was used for measuring the vertical profile in the CO concentration over the West Sea. Back trajectory analysis of the air recorded in the area was conducted for accurate estimation of the path of the wind transporting the pollutants. The Hybrid Single-Particle Lagrangian Integrated Trajectory model of the Air Resources Laboratory (ARL, US National Oceanic and Atmospheric Administration) was used for back trajectory analysis at 2-h intervals over a 72-h period on each of the two data recording dates (Draxler & Hess, 1997; Rolph et al., 2017; Stein et al., 2015).

3 Results and discussion

3.1 Concentrations of NH3 at six supersites

As shown in Fig. 3, the mean NH3 concentration measured at the Seoul site was 5.6 ± 2.4 ppb, whereas that of the Daejeon region was higher at 9.0 ± 3.4 ppb. The Gwangju region recorded the highest NH3 concentration (9.3 ± 3.3 ppb). The mean NH3 concentrations in the Ulsan region, Jeju Island, and Baengnyeong Island were 3.5 ± 1.4, 2.1 ± 1.8, and 1.3 ± 1.1 ppb, respectively. The regional variations of NH3 concentration, temperature, and relative humidity measured from 2020 to 2021 are shown in Figs. S8 and S9. Interestingly, despite the large and dense population of Seoul, the NH3 concentrations at the Seoul site were lower than those in the Gwangju and Daejeon regions, probably ascribable to the agricultural and industrial activities near these sites. Gwangju, located close to a sanitary treatment plant and agricultural land, showed the highest NH3 concentrations of all the study sites. The Daejeon site, surrounded primarily by agricultural areas, also showed a high concentration of NH3. The lower NH3 concentrations at Baengnyeong and Jeju islands were attributed to their lower population numbers and minimal industrial activities. The NH3 concentrations in other regions, such as Seoul and Ulsan, showed a direct proportional relationship to their respective populations, indicating the contributions of diverse human, industrial, traffic, and agricultural activities to NH3 emissions. These findings highlighted the need for targeted control measures in areas with high NH3 concentrations to mitigate their effects on human health and air quality.

Fig. 3
figure 3

Spatial distributions of weekly NH3 concentrations and temperatures at various National Institute of Environmental Research (NIER) supersites over 1 year (2020–2021). a Seoul, b Daejeon, c Gwangju, d Ulsan, e Jeju Island, f Baengnyeong Island (background)

Full size image

3.2 Seasonal variation in NH3 concentrations

We used the following definitions for the four seasons, namely spring (March 1 to May 31), summer (June 1 to August 31), autumn (September 1 to November 30), and winter (December 1 to February 28). The study results presented in Fig. 4 show distinct seasonal patterns across all the sites. The Seoul and Daejeon sites showed the highest NH3 concentration during summer (Seoul 7.5 ± 2.3 ppb, Daejeon 11.6 ± 3.3 ppb) and the lowest during winter (Seoul 3.2 ± 1.3 ppb, Daejeon 5.6 ± 2.6 ppb). The Gwangju and Ulsan sites and Baengnyeong Island also showed the highest NH3 concentration during summer (Gwangju 11.1 ± 4.4 ppb, Ulsan 4.1 ± 1.3 ppb, Baengnyeong Island 2.1 ± 1.6 ppb). However, the highest NH3 concentration at Jeju Island was recorded during spring (3.1 ± 1.6 ppb).

Fig. 4
figure 4

Seasonal variation in NH3 concentrations at various National Institute of Environmental Research (NIER) supersites over 1 year (2020–21)

Full size image

Various factors could cause the observed seasonality in NH3 concentrations, including temperature, rainfall, and agricultural activities. During the summer months, high temperatures and increased agricultural activity in mainland South Korea leads to the decomposition of organic fertilizers, which caused the highest NH3 concentrations across most of our study sites except Jeju Island (Kuttippurath et al., 2020). Moreover, high temperatures caused the conversion of aqueous NH3 to the gaseous phase. In contrast, wet deposition during rainfall in the autumn and winter seasons was responsible for the lowest NH3 concentrations during winter at most study sites (Warner et al., 2016). Further, NH3 reacts with HNO3 in the atmosphere during the colder months, leading to a further reduction in atmospheric NH3 concentrations (Zhou et al., 2019).

Jeju Island was the exception to the observed seasonal trend, with the highest NH3 concentration recorded during the spring season. This phenomenon could be ascribed to the agricultural activities during the spring season contributing to the higher NH3 concentrations. In addition, the relatively lower summer temperatures on the island compared with other regions around South Korea could have contributed to the higher NH3 concentrations recorded in spring. The results of this study showed the importance of understanding the seasonal variability in and factors contributing to NH3 concentrations. This information could inform targeted control measures to mitigate the effects of NH3 emissions on human health and air quality.

3.3 Seasonal variation in roadside emissions

Several sources influence the concentration of NH3 in the atmosphere, with vehicle emissions being a significant contributor. In this section, the seasonal variation in the correlation between NH3, CO, and NOx concentrations from vehicle emissions at roadsides in Seoul is discussed. The study was conducted over a 1-month period each during two seasons, namely summer (August) and autumn (October) in 2018. The NH3, CO, and NOx concentrations were measured and the data analyzed using correlation coefficient (R2) values. The results showed that during summer, a strong positive correlation (R2 = 0.62) existed between the CO and NH3 concentrations, indicating that the NH3 emissions were consistent with the CO emissions from vehicles (Fig. 5a). A strong correlation (R2 = 0.69) was also observed between the NOx and NH3 concentrations during the same season. These results suggested that NH3 emissions from vehicles were a significant contributor to air pollution in Seoul during summer.

Fig. 5
figure 5

Correlations between CO and NOx with NH3 at roadside in Seoul during a summer and b autumn

Full size image

During autumn, a positive correlation (R2 = 0.61) was observed between the CO and NH3 concentrations similar to summer, and weaker positive correlation (R2 = 0.62) was observed between NOx and NH3 concentrations than in summer. The concentrations of NH3, CO, and NOx emissions were consistent with each other, as shown in Fig. 5b, and most NH3 pollution recorded at the site during autumn could be ascribed to vehicle emissions. These findings suggested that vehicle emissions during summer and autumn (particularly summer) contributed substantially to NH3 pollution at Seoul roadsides. Therefore, appropriate measures are required to reduce vehicle emissions and improve the air quality of the city.

3.4 Seasonal variation in the correlation between NH3 and NO2 at the Seoul supersite

As noted in Section 3.3, NH3 emissions along roadsides in Seoul were affected significantly by vehicle emissions; however, NO2 is another major air pollutant deriving from such emissions. To improve our understanding of the correlation between NH3 and NO2, we investigated the concentrations at the supersite in Seoul over a 1-month period during three seasons, namely winter (January), spring (March), and summer (August) in 2020–2021. Figure 6 shows the correlation between NO2 and NH3. In winter, a correlation coefficient value of R2 = 0.584 was observed between NO2 and NH3 concentrations, i.e., the concentrations of NH3 and NO2 were consistent, suggesting that vehicle emissions were the major source of NH3 pollution during that season (Fig. 6c). However, in spring and summer, the correlation coefficient between NO2 and NH3 were 0.007 and 0.004, respectively (Fig. 6a and b). It suggests a considerable increase in the NH3 concentration without NO2 increase. This increase in the NH3 concentration without NO2 may be due to increased agricultural activity from spring to summer in the nearby regions or other influences than the effect of vehicle emissions. Overall, our study confirmed that NH3 emissions in Seoul were affected significantly by vehicle emissions, with agricultural activities playing an important role in increasing NH3 concentrations during the summer season.

Fig. 6
figure 6

Correlations between NO2 and NH3 at the supersite in Seoul during a spring, b summer, and c winter

Full size image

3.5 Vertical distribution of NH3 and NR-PM1 over the West Sea

Airborne measurements were conducted over the West Sea of South Korea to determine the sources of gaseous emissions, particularly long-range transported and maritime emissions. The long-range transportation measurements were conducted on 2 June 2019, with the results shown in Fig. 7a and b. The spiral flight provided vertical profiles of various NR-PM1 chemical compositions based on their concentrations. The concentrations of NO3− were the lowest (0.26–1.10 µg m−3) at altitudes of 300–1000 m, with the concentrations of NH4+ being slightly higher (0.90–1.32 µg m−3) than those of NO3− at these altitudes. Higher organics concentrations were observed (1.7–3.2 µg m−3) compared with those of NH4+ at altitudes of 300–1000 m, whereas the SO42− concentrations were higher (3.22–3.89 µg m−3) than the organics concentrations at this altitude range. The CO concentrations range was 198.1–295.59 ppbv at altitude range 300–1000 m, with NR-PM1 showing a similar trend. We employed CO as a tracer for tracking long-range pollutant transport. As the altitude increased beyond 1000 m, the concentrations of NO3− exceeded those of NH4+, whereas the other components retained their order. The NH3 concentration range was 7.91–14.2 ppbv, increasing along with the altitude range of 300–1000 m, with the recorded concentrations at the highest altitude of 1103.5 m being 12.24 µg m−3 (SO42−), 10.86 µg m−3 (organics), 8.21 µg m−3 (NO3−), 5.91 µg m−3 (NH4+), 310.8 ppbv (CO), and 15.7 ppbv (NH3). These results indicated that the pollutant concentrations increased along with the altitude increasing above sea level, implying that the primary source of pollution over the West Sea on 2 June 2019 was long-range transportation from neighboring regions such as northeastern China. Back trajectory analysis of the wind confirmed that it was blowing from China toward the West Sea. The correlation between altitude and temperature indicated that the temperature dropped below 20 °C as the height increased over 400 m. These results suggested no possibility of NH3 emissions deriving from the decomposition of NH4NO3 (Chaturvedi & Dave, 2013).

Fig. 7
figure 7

Vertical profiles of a CO and particulate matter [aerodynamic diameter ≤ 1 µm (PM1)] concentrations by their compositions recorded on 2 June 2019 during long-range transport; b NH3 concentrations measured over the West Sea on 2 June (long-range transport) and 5 June 2019 (ocean origin)

Full size image

The maritime NH3 emissions were measured over the West Sea on 5 June 2019, with the results shown in Fig. 7b. The NH3 concentration decreased from 10.1 to 7.37 ppbv as the height increased from 300 to 1000 m, indicating that most NH3 derived from oceanic emissions near the ocean surface (Paulot et al., 2015). Figure S10 shows that the temperature increased over 20 °C as the altitude increased over 400 m, implying that NH4NO3 decomposition could have contributed additional NH3 to the oceanic emissions.

4 Conclusions

The aim of this study was analyzing the trends of NH3 emissions in the South Korean Peninsula and West Sea region and identifying the factors causing the pollution. The data on NH3 and other air pollutants (NOx, CO, and NR-PM1) were collected from six supersites across the peninsula, a roadside in Seoul, and the West Sea over different periods, ranging from 1 month to 1 year.

In synthesizing the results obtained from each site, the complex interplay of agricultural, urban, and oceanic factors influences the NH3 emissions in the South Korean Peninsula and West Sea region.

In rural areas, the high concentrations of NH3 were primarily attributed to agricultural activities, especially using fertilizers. This resulted in the decomposition of NH4NO3 during the high summer temperatures, highlighting the direct impact of agricultural practices on ambient NH3 concentration. Conversely, areas with lower populations showed the lowest NH3 concentrations due to the absence of significant agricultural, industrial, or anthropogenic activities.

In contrast, at urban sites like Seoul, there was a clear correlation between ambient NH3 and vehicle emissions, which suggests that vehicle emission is important to NH3 pollution in urban. Interestingly, this correlation was particularly strong during winter, implying that urban vehicle emissions were a primary source of NH3 during this season. But, increasing the NH3 concentrations observed during spring and summer suggest that additional NH3 sources, beyond vehicle emissions, are likely tied to agricultural activities within or near the city.

Lastly, the airborne measurement over the West Sea of South Korea revealed that a significant portion of NH3 pollution originated from oceanic emissions and NH4NO3 decomposition by high temperature, and long-range transportation from the northeastern region of China.

This interconnectedness of rural, urban, and maritime influences explains the complexities of understanding and managing NH3 emissions. It underscores the need for comprehensive, multi-faceted strategies considering the varied emission sources and environmental dynamics in different regions and seasons. In conclusion, the results of our study provided valuable insights into the emission sources of NH3 in primary air pollutants in South Korea, highlighting the contributions of both land-based and oceanic sources. These findings could help inform policymakers and stakeholders for developing effective air pollution control strategies in the region.