Real-Time Monitoring of Particulate Matter (PM₁₀ and PM_2.5) Emitted from Paddy Fields in South Korea: A One-Year Study

Abstract

This study was performed to determine the pattern of particulate matter (PM₁₀ and PM_2.5) emitted from agriculture in South Korea by monitoring its concentrations in paddy fields in real time for one year. The highest average seasonal concentration of PM₁₀ was measured in spring (59.94 ± 22.82 μg/m³), followed by winter (59.00 ± 11.40 μg/m³), autumn (40.10 ± 9.67 μg/m³), and summer (28.3 ± 8.5 μg/m³). For PM_2.5, the average concentration was highest in spring (27.27 ± 6.42 μg/m³), followed by autumn (16.98 ± 3.43 μg/m³), winter (16.32 ± 7.51 μg/m³), and summer (14.40 ± 5.21 μg/m³). Real-time monitoring showed that PM₁₀ and PM_2.5 from some paddy fields in South Korea had the highest concentrations in spring, especially in April when farming operations begin, with moderate levels in autumn and winter and the lowest concentrations in summer. By time of day, higher concentrations were generally measured in the evening and at night when agricultural workers were not present, but on days with high concentrations, the fine dust derived from paddy field can pose a health threat at any time of day or night. Therefore, it is important to raise awareness of the risk of exposure to fine particulate matter among agricultural workers through information, education, and training in April, when cultivation begins during the spring season. Agricultural workers are also advised to check the level of fine particulate matter on a regular basis and take preventive measures such as spraying, stopping farming activities, and adjusting working hours when the level of fine particulate matter is high. Raising awareness of exposure risks is even more important and urgent for older, health-sensitive agricultural workers and foreign migrant and seasonal workers.

1. Introduction

According to the World Health Organization (WHO), air pollutants are one of the biggest threats to the environment and human health, with an average of 7 million deaths per year due to air pollution [1]. Among air pollutants, fine particulate matter (PM₁₀; dust with an aerodynamic diameter of 10 μm or less) and ultrafine particulate matter (PM_2.5; dust with an aerodynamic diameter of 2.5 μm or less) can penetrate deeply into the respiratory tract and spread to the end of the alveoli, causing not only respiratory diseases but also cardiovascular diseases and cancer [2,3,4]. Therefore, in areas with high concentrations of fine particulate matter and ultrafine particulate matter, mortality rates due to respiratory diseases, cardiovascular diseases, and cancer increase dramatically [5,6,7]. Previous studies have confirmed the harmful effects of inhaled fine particulate matter on cardiovascular health and demonstrated that it is a major mechanism of macrovascular and microvascular damage [8,9]. To prevent adverse health effects from exposure to these fine particles, WHO recommends air quality standards for PM₁₀ of no more than 5 μg/m³ on an annual average and no more than 15 μg/m³ on a 24-h average, and for PM_2.5 of no more than 15 μg/m³ on an annual average and no more than 45 μg/m³ on a 24-h average. The preventive measures for exposure to fine particulate matter include wearing respiratory protection, avoiding outdoor activities on days when particulate matter concentrations exceed the standard, and getting plenty of rest.

Specific agricultural practices that contribute to particulate matter emissions include tilling and managing soil, applying chemicals such as fertilizers and pesticides, managing crop residues, and raising livestock. In fact, agricultural workers are continuously exposed to high levels of particulate matter generated during their work in agricultural fields, which consists of biological aerosols from grain, pesticides, feed additives, fertilizers, and plant or animal waste [10,11,12,13]. A study of agricultural workers in an agricultural region in California, USA, collected personal samples and found that they were exposed to high concentrations of PM₁₀ and PM_2.5 during harvesting of tomatoes, almonds, and melons [10]. In Korea, basic research has been conducted on air pollutants in agricultural areas, but compared to urban areas, air pollutant measurement networks have not yet been established [14].

On the other hand, South Korea is facing the problem of population decline and aging in rural areas due to industrialization since 1960, increasing urban migration, and a rapid decline in the working-age population [15]. To solve these issues, the Korean government has implemented migrant labor policies through the Employment Permit System (EPS) to attract foreign workers [16], and the Ministry of Justice has established a seasonal work system to introduce foreign labor into the agricultural sector, as well as policies to induce foreigners residing in South Korea, such as marriage migrants and foreign students, to work in the agricultural sector. These policies have led to a significant influx of foreign labor into the agricultural sector, which has led to a sharp increase in the number of foreigners working in agriculture [17]. In general, older farmers are more sensitive to exposure to air pollutants such as ultrafine particulate matter. Due to cultural differences, language barriers, and differences in education levels, foreigners may lack health awareness of air pollutant hazards, exposure routes, exposure issues, and health effects. Unfortunately, no monitoring studies have been conducted on fine particulate matter emissions from Korean paddy fields, so this study has academic significance and novelty in that it is the first to identify the amount and pattern of fine particulate matter in rural areas in South Korea.

Therefore, this study aims to monitor the concentration patterns of PM₁₀ and PM_2.5, particulate pollutants, in real time to understand the air pollutants generated in paddy fields in South Korea, and to collect basic data to prevent environmental diseases among Korean agricultural workers based on these data.

2. Materials and Methods

2.1. Subject

The measurement site was selected as a partial paddy farming area located in Yeoju City, Gyeonggi-do, South Korea, which complies with the Ministry of Environment’s guidelines for the installation and operation of air pollution measurement networks and fulfils the conditions for the monitoring data to consider only the effects of agricultural activities and exclude the effects of other external factors. As shown in Figure 1, the site is a typical paddy farming area consisting only of large paddy fields with no industrial complexes and small towns within 5 km, except for a small village at a distance of about 1000 m to the north of the measurement point.

2.2. Method

Due to the nature of Korea’s climate, the seasonal wind direction changes, resulting in different characteristics of the particles brought in by each season. Unlike other crops, rice farming has a clear growing season and lean season, and only water management and pest control are required except for the spring and autumn growing seasons. Therefore, the measurement period was selected as a one-year cycle from 1 March 2023 to 29 February 2024 to analyze the characteristics of each substance generated in the paddy field area, especially the seasonal and daily characteristics. In general, South Korea has four seasons: spring (March to May), summer (June to August), fall (September to November), and winter (December to February).

As shown in Figure 2, the air quality station is built in the form of a container measuring 2.5 m by 3 m. At the top of the container is the intake for particulate pollutants. A handrail is installed at the top of the station for safety while working, and the sides are equipped with stairs for easy access to the roof. The measuring equipment is installed inside the room, and the temperature is maintained by a heating and cooling system. The entrance door is equipped with a door lock, and CCTV and fans are installed to enhance security (Figure 3).

When installing measuring stations in actual areas where agricultural activities are carried out, groundworks are essential, as subsidence and tilting can occur, and there is a risk of flooding during rainfall. Groundworks were carried out for each installation point in each region, taking into account the installation environment. Due to the size of the container and the installation of the pans, work was done to secure an area of 5 m × 5 m and a minimum height to prevent flooding (Figure 4).

The measurement of particulate matter was performed using a particulate matter measuring instrument called Spirant BAM (Model 3, Kentech Corp., Daejon, Republic of Korea) to monitor PM₁₀ and PM_2.5 in the air in real time (hourly intervals). Two Spirant BAMs were installed at this site, each equipped with an inertial particle size separator (very sharp cut cyclone; VSCC). The Spirant BAM is a standard instrument, approved by the Korean Ministry of Environment, that measures the concentration of fine particulate matter using beta radiation by measuring the amount of particles collected on filter paper (Figure 5). Its accuracy exceeds U.S. EPA Class 3 PM_2.5 and PM_10-2.5 FEM standards for additive and multiplicative bias, and resolution is 1 μg/m³.

2.3. Data Analysis

Real-time monitoring was carried out in a one-year cycle from March 2023 to February 2024, but machine errors or malfunctions often occurred during the measurement period. Therefore, the real-time measured data included the Not Detected (N.D.) data for each hour, which was 13 h for PM₁₀ and 19 h for PM_2.5. The measured data were collected and organized in Microsoft Excel, and the hourly average concentration, monthly average concentration, and seasonal average concentration were calculated, and graphs were generated. In addition, a two-way ANOVA was applied using Statistical Package for the Social Sciences (SPSS) to compare the statistical differences between the monthly and seasonal data. Since the number of measured data points totaled more than 30, the data were normally distributed according to the central limit theorem.

3. Results

3.1. Seasonal Concentrations of PM₁₀ and PM_2.5

Figure 6 shows the seasonal distribution of PM₁₀ and PM_2.5 concentrations in the paddy field area. For PM₁₀, the highest average concentration was measured in spring at 59.94 ± 22.82 μg/m³, and 59.00 ± 11.40 μg/m³ in winter. This was followed by autumn at 40.10 ± 9.67 μg/m³ and summer at 28.3 ± 8.5 μg/m³, the lowest concentration among the four seasons. On the other hand, the average concentration of PM_2.5 was highest in spring at 27.27 ± 6.42 μg/m³. This was followed by 16.98 ± 3.43 μg/m³ in autumn, 16.32 ± 7.51 μg/m³ in winter, and finally 14.40 ± 5.21 μg/m³ in summer, which was the lowest of the four seasons.

3.2. Monthly Concentrations of PM₁₀ and PM_2.5

Figure 7 shows the monthly concentration distribution of PM₁₀ in the paddy field area. In spring, the average concentration of PM₁₀ was 37.2 ± 27.30 μg/m³ in March and 91.09 ± 85.49 μg/m³ in April, with the highest concentration in spring, and it then decreased significantly to 51.7 ± 35.06 μg/m³ in May. In the summer, the average concentration of PM₁₀ was 36.73 ± 17.01 μg/m³ in June, 31.55 ± 20.4 μg/m³ in July, and 16.65 ± 18.46 μg/m³ in August. In autumn, the average concentration of PM₁₀ was 27.10 ± 19.1 μg/m³ in September, 42.8 ± 25.3 μg/m³ in October, and 50.3 ± 28.9 μg/m³ in November. In winter, the average concentration of PM₁₀ was 56.99 ± 90.6 μg/m³ in December, 75.10 ± 89.25 μg/m³ in January, and 47.62 ± 34.99 μg/m³ in February, the lowest concentration in winter. The highest concentration was recorded in April in spring, and the lowest concentration was recorded in August in summer.

The monthly concentration distribution of PM_2.5 in the paddy field area, as shown in Figure 8, shows that the average concentration of PM_2.5 measured in spring was 36.11 ± 27.70 μg/m³ in March, which was the highest concentration of the spring season and year, followed by 24.67 ± 19.00 μg/m³ in April and 21.02 ± 11.10 μg/m³ in May, which was the lowest concentration of the spring season. The summer measurements showed that the average concentration of PM_2.5 was highest in June at 21.00 ± 11.07 μg/m³, followed by July at 15.75 ± 14.25 μg/m³. In August, the average concentration was 7.43 ± 10.18 μg/m³, the lowest concentration of the summer season. In autumn, the average concentration of PM_2.5 gradually increased to 13.03 ± 10.82 μg/m³ in September, 16.45 ± 11.08 μg/m³ in October, and 21.39 ± 14.76 μg/m³ in November. In winter, the average concentration of PM_2.5 was 25.53 ± 18.18 μg/m³ in December, the highest concentration of the winter season, and decreased significantly to 7.15 ± 17.53 μg/m³ in January, the highest concentration of the winter season. In February, the average concentration was 16.25 ± 18.14 μg/m³, an increase from January and the lowest concentration of the year.

3.3. Time-Based Concentration Fluctuations of PM₁₀ and PM_2.5

Figure 9 shows the distribution of average concentrations of PM₁₀ by time of day. In the spring season, the peak concentration was reached at 10:00 in March and April, and the highest concentration was recorded at 19:00 in May. The hourly average peak concentrations were 48.13 ± 40.09 μg/m³ in March, 115.93 ± 94.26 μg/m³ in April, and 58.73 ± 47.52 μg/m³ in May. In the summer months, the peak concentration was reached at 0:00 h in June, and the peak concentration was recorded at 20:00 in July and August. The hourly average peak concentrations were 47.66 ± 20.79 μg/m³ in June, 38.09 ± 25.93 μg/m³ in July, and 22.61 ± 24.86 μg/m³ in August. In the autumn season, the peak concentration was reached at 19:00 in September and November, and the highest concentration was recorded at 10:00 in October. The hourly average peak concentrations were 34.86 ± 19.50 μg/m³ in September, 65.25 ± 29.06 μg/m³ in October, and 64.68 ± 21.68 μg/m³ in November. In winter, the peak concentration was reached at 11:00 in December, followed by 5:00 in January and 22:00 in February. The hourly average peak concentrations were 90.67 ± 162.64 μg/m³ in December, 102.58 ± 156.08 μg/m³ in January, and 56.03 ± 42.84 μg/m³ in February.

Figure 10 shows the distribution of average concentrations of PM_2.5 by time of day. In spring, the peak concentration was reached at 10:00 in March, followed by 0:00 in April and 21:00 in May. The highest hourly average concentration in spring was 46.86 ± 40.38 μg/m³ in March, followed by 30.83 ± 26.82 μg/m³ in April and 24.32 ± 11.87 μg/m³ in May. In summer, the peak concentration was reached at 22:00 in June, followed by 19:00 in July and 20:00 in August. The hourly average peak concentrations were 24.16 ± 11.16 μg/m³ in June, 20.709 ± 18.58 μg/m³ in July, and 9.32 ± 12.28 μg/m³ in August. In autumn, the peak concentration was reached at 19:00 in September and November, and the highest concentration was recorded at 18:00 in October. The hourly average peak concentrations were 18.50 ± 12.72 μg/m³ in September, 21.67 ± 10.91 μg/m³ in October, and 28.10 ± 14.01 μg/m³ in November. In winter, the peak concentration was reached at 8:00 in December, followed by 17:00 in January and 21:00 in February. The hourly average peak concentrations were 28.06 ± 20.89 μg/m³ in December, 9.74 ± 28.55 μg/m³ in January, and 19.10 ± 19.36 μg/m³ in February.

3.4. Statistical Analysis of Time-Based Concentrations of PM₁₀ and PM_2.5

Table 1. Result of a two-way ANOVA for monthly concentrations of PM₁₀ and PM_2.5 (N = 8784).

	PM10						PM2.5
	n	Mean (μg/m3)	Std. (μg/m3)	F	p		n	Mean (μg/m3)	Std. (μg/m3)	F	p
March	24	37.02	27.30	134.225	0.00 ***	March	24	37.02	27.30	134.225	0.00 ***
April	24	91.09	85.49			April	24	91.09	85.49
May	24	51.71	35.06			May	24	51.71	35.06
June	24	36.73	17.01			June	24	36.73	17.01
July	24	31.55	20.36			July	24	31.55	20.36
August	24	16.65	18.46			August	24	16.65	18.46
September	24	27.10	19.06			September	24	27.10	19.06
October	24	42.91	25.38			October	24	42.91	25.38
November	24	50.29	28.92			November	24	50.29	28.92
December	24	56.99	90.54			December	24	56.99	90.54
January	24	75.10	89.25			January	24	75.10	89.25
February	24	47.62	34.99			February	24	47.62	34.99

***: p-value < 0.001.

shows the results of the two-way ANOVA to test whether the monthly concentrations of PM₁₀ and PM_2.5 are statistically different. The difference test for 12 months yielded F = 133.988 for PM₁₀ and F = 30.127 for PM_2.5, which is statistically significant at p < 0.05, confirming that there is a monthly difference in concentrations of both PM₁₀ and PM_2.5.

4. Discussion

4.1. PM₁₀

PM₁₀ is mainly caused by a variety of sources, including vehicle traffic, industrial activities, construction work, agricultural residue burning, field preparation activities, and soil dust [18,19]. In addition, the composition and concentration of PM₁₀ fluctuate with seasonal changes [20]. The study compared the mean concentrations by month and found a statistically significant difference between the months (p < 0.05). Although the monthly average concentration in April was the highest of the year, it was found that agricultural activities such as soil tilling, use of agricultural machinery in soil tilling, and fertilizer application contributed to the high concentration in spring. A study conducted in China identified agricultural residues as the most significant factor contributing to high PM₁₀ emissions from agricultural areas, followed by soil tillage as the second most significant factor contributing to high PM₁₀ concentrations [21]. A study conducted by Tatarko et al. [22] found that friction generated during soil tillage had a significant impact on PM₁₀ and PM_2.5 production. Jia et al. [23] found that particulate matter emissions from agricultural machinery operation in agricultural areas were mainly PM₁₀ and total suspended particulate (TSP). A study conducted in an agricultural area in northern China found that PM₁₀ emission concentrations were highest in April and May, and concluded that the source of emissions was strongly related to soil cultivation [24]. This study found results similar to those in our study.

The springtime influx of dust and yellow dust from China into Korea is also thought to have contributed to the high PM₁₀ concentrations in April. For the Korean Peninsula as a whole, the impact of yellow dust in spring was relatively large, while the impact in summer was lower due to the removal of PM_2.5 by precipitation [25]. The high PM₁₀ concentrations measured in late autumn (October and November) and winter (December to February) are thought to be influenced by high concentrations of ammonia (NH₃) emitted from agricultural areas (rice fields). Ammonia is an important precursor to the production of PM₁₀ and PM_2.5 through photochemical reactions in the atmosphere [26]. During the summer months of July, August, and September, PM₁₀ concentrations were found to be the lowest of the year. This is due to the fluctuation of the rainy season due to climate change in 2023, with the rainfall during the rainy season increasing to 660.2 mm nationwide, compared to 356.7 mm in normal years [27]. The cleaning effect of high rainfall is believed to have contributed to the low PM₁₀ concentrations from July to September. According to the Korea Meteorological Administration, the average precipitation in July was 547.20 mm, the highest in 2023, followed by 253.90 mm in August and 167.90 mm in September [28].

When examined by time of day, the spring season shows higher concentrations in the morning and evening compared to the afternoon, with a tendency to decrease after midday and increase again in the evening. In summer, concentrations are higher in the late night and early morning compared to the afternoon, with a decrease in the morning, a significant increase at noon, and a significant decrease again after noon. In autumn, concentrations were higher in the evening compared to the afternoon, with a decreasing trend in the evening, followed by an increase at noon and a significant decrease again in the afternoon. In winter, concentrations were higher in the morning and early afternoon compared to the afternoon and evening, and tended to decrease after noon. Overall, PM₁₀ concentrations were found to be higher in the early morning, morning, and night when temperatures were lower. This suggests that PM₁₀ concentrations are related to air temperature. Li et al. [29] found that PM₁₀ concentrations were correlated with air temperature, with the highest concentrations at night and in the evening, and the lowest concentrations around noon.

4.2. PM_2.5

Concentrations of PM_2.5 have shown large seasonal variations, and in general, emissions, meteorological conditions, long-range transport pathways, and atmospheric boundary layer height vary significantly with the seasons, so that the concentrations of chemical components of PM_2.5 at a measurement site are subject to seasonal variations [14]. The study compared monthly mean concentrations and found statistically significant differences between months (p < 0.05). The highest monthly mean concentrations were found in the spring months of March and April, but as with PM₁₀, this may be due to agricultural activities such as soil tilling, the use of agricultural machinery to till the soil, and fertilizer application. PM_2.5 is also secondarily produced by photochemical reactions of sulfate and nitrogen salts with primary emissions such as nitrogen oxides and sulfur oxides [18,19]. Therefore, secondary production of sulfate and nitrogen salts from fertilizer applied to cultivated land may have contributed to the high concentrations of PM_2.5 in spring. In addition, domestic inbound particulate matter and yellow dust from China during the spring and winter months are also thought to have contributed to the high PM_2.5 concentrations in April.

The high concentrations of PM_2.5 measured in late autumn (October and November) and winter (December to February), as well as PM₁₀, are likely to be influenced by high concentrations of NH₃ emitted from agricultural areas (rice fields). The ammonia is an important precursor for the production of PM₁₀ and PM_2.5 through photochemical reactions in the atmosphere [26]. PM_2.5, like PM₁₀, showed lower emission concentrations during the summer months of July, August, and September. This is likely due to the washing effect of high precipitation, which contributed to the lower concentrations of PM_2.5 from July to September. According to the Korea Meteorological Administration, the average precipitation in July was 547.20 mm, the highest in 2023, followed by 253.90 mm in August and 167.90 mm in September [28]. Previous studies have shown that the rainy season from July to September in 2023 lowered PM_2.5 concentrations emitted from agricultural areas (fields) due to the cleaning effect of the rainy season [20]. The higher variation in monthly average concentration data for PM_2.5 than PM₁₀ is likely due to differences in particle size. In particular, the relatively high standard deviation of the January average concentration data for PM_2.5 is due to an unexpected event, namely an increase in fugitive dust from land improvement works that took place for five days around the measurement site.

By time of day, concentrations were higher in the middle of the night compared to the day, except in March, and gradually decreased as the sun grew brighter. In March, there was a trend of increasing from 5 am, peaking at 10 am, and then decreasing significantly. In summer, concentrations were higher in the late night and early morning compared to the afternoon. In autumn, concentrations were higher in the evening compared to the afternoon in September and November, and tended to decrease later in the evening. In contrast to September and November, October showed an increasing trend from 0:00, peaking at 18:00 and then decreasing again. In December, concentrations were higher in the morning and evening compared to the afternoon, with a peak in the evening in January, and in February, as in December, concentrations were higher in the morning and evening compared to the afternoon. Overall, PM_2.5 concentrations were found to be higher at dawn, in the morning, and at night when temperatures were lower. This suggests that PM₁₀ concentrations are associated with temperature. Li et al. [29] found that PM₁₀ and PM_2.5 emission concentrations were correlated with air temperature, with the highest concentrations occurring at night and in the evening, and the lowest concentrations occurring around noon. In addition, a previous study conducted in agricultural areas (fields) in 2020 concluded that although PM_2.5 concentrations varied by season, they were generally higher in the early morning and nighttime hours, which could be attributed to the air temperature inversion layer phenomenon [14].

This study selected only one paddy field area as the monitoring site, which was limited in representing a typical ecosystem of paddy fields in South Korea. Therefore, it will be necessary to perform multi-regional comparative monitoring in the future to increase the generalizability of the conclusions. Additionally the effects of crop characteristics, soil properties, and topographical characteristics were not fully investigated. In future research, it will be necessary to further compare rice fields and fields in agricultural areas, and to conduct specific analyses by crops or characteristics of agricultural areas, soil characteristics, and topography to identify accurate data and causes of emissions.

5. Conclusions

In this study, the exact concentrations of particulate pollutants (PM₁₀ and PM_2.5) in paddy fields in South Korea were measured in real time and compared seasonally, monthly, and hourly, and the emission pathways and sources were identified. Based on the results obtained from this study, the real-time monitoring showed that PM₁₀ and PM_2.5 in some paddy fields in South Korea had the highest concentrations in spring, especially in April when farming operations begin, with moderate levels in autumn and winter, and the lowest concentrations in summer. By time of day, higher concentrations were generally measured in the evening and at night when agricultural workers were not present, but on days with high concentrations, fine dust derived from paddy fields can pose a health threat at any time of day or night. Thus, it is necessary to develop more effective particulate pollutant reduction measures tailored to agricultural areas based on field monitoring results.