On air pollutant variations in the cases of long-range transport of dust particles observed in central Korea in the leeside of China in 2010

Abstract

The occurrence and transport of dust storms and anthropogenic air pollutants in East Asia have been monitored using the meteorological satellite data of the National Oceanic and Atmospheric Administration (NOAA) and ground-based measurements at Cheongwon, a background observation site in central Korea. The nine cases, measured over 14 days, of natural dust particles caused by dust storms originating from northern China and Mongolia were observed at Cheongwon in the spring, autumn, and winter of 2010. In addition, seven cases, measured over 18 days, of anthropogenic dust particles originating from eastern China were observed over the course of a year. In those cases of natural and anthropogenic dust particles observed at Cheongwon, the level of particle matter (PM) with a diameter of ≤10 μm (PM10) (100 μgm⁻³ day⁻¹) or PM2.5 (50 μgm⁻³ day⁻¹) exceeded the air quality standards of Korea for 5 and 6 days, respectively. At the same time, CO concentrations rose higher due to long-range transport, while CO levels in the cases of anthropogenic dust particles (954 ppb) rose higher compared with the cases of natural dust particles (812 ppb). While gusty north–northwesterly winds were blowing in the front side of high-pressure systems, an increase in CO concentrations along with the influx of dust storms was observed at Cheongwon in the three natural dust particle cases with continental background airflow (n-CBAs). However, the other six natural dust particle cases with continental polluted airflow (n-CPAs) were also observed where mass concentrations of TSP, PM10 and PM2.5, and CO increased simultaneously after the CO concentrations rose and fell before dust storms had flowed in, showing double peaks. The mass concentrations of total suspended particle (TSP), PM10, and PM2.5 were high, and the PM2.5/TSP mass concentration ratio was high in n-CPAs, compared with n-CBAs. However, the anthropogenic dust particle cases with continental polluted airflow (a-CPA) are affected by air pollutants transported from eastern China by warm southwesterly winds, since central Korea is located in the rear side of the high-pressure system and in the foreside of the low-pressure system. Also, the anthropogenic dust particle cases with regionally polluted airflow (a-RPA) were observed in central Korea due to a stagnant high-pressure system for several days in the Yellow Sea region. In a-CPA, the mass concentrations of TSP, PM10, and PM2.5; the PM2.5/TSP mass concentration ratio; and the CO concentrations were all higher than in a-RPA.

Introduction

Aerosol in the atmosphere affects climate change and the atmospheric environment. As a rule, atmospheric aerosols fall into categories of coarse particles (≥2.5 μm) and fine particles (<2.5 μm), depending on their size. Mechanical pulverizing processes usually generate coarse particles over natural sources, such as dust and sea salt. However, anthropogenic sources emit mainly fine particles by the burning of fossil fuels and are produced in the processes of industrial activities (Kim et al. 2012).

The East Asian region is one of the major areas of the world where aerosol occurs. Dust particles caused by dust storms in northern China and Mongolia move on a large scale, thus affecting the air quality of the Korean Peninsula, Japan, and the USA (Kim et al. 2012; Kurosaki and Mikami 2003; Wang and Fang 2006). Eastern China has continued to emit more and more aerosol due to the fast growth of its economy and industrialization over the last few decades (Akimoto 2003). Analyses of a ground-based mass concentration measured in central Korea and optical properties indicate that types of atmospheric aerosols in East Asia have changed over the past 10 years (Kim et al. 2007; Kim et al. 2014). While the mass concentration of total suspended particle (TSP) in central Korea is decreasing, the particle matter (PM) with a diameter of ≤2.5 μm (PM2.5)/TSP mass concentration ratio is increasing (Kim et al. 2014). In addition, the aerosol optical depth and Angstrom exponent are on a steady rise in the Yellow Sea region (Kim et al. 2007; Kim et al. 2014; Logan et al. 2013). The long-term radiative forcing caused by aerosols has decreased in the Yellow Sea region due to decreases in the occurrences of dust storms and increases in hygroscopic aerosols through photochemical reactions in East Asia (Kim et al. 2014; Li et al. 2013).

The technology of detecting dust storm transport has been developed by using satellite observation, ground-based measurement, and numerical modeling. With respect to large-scale air pollutants being transported over a long distance, satellite detection methods, as well as ground-based measurement technology, are being improved (Chung 1986; Gu et al. 2003; Iino et al. 2004; Lee et al. 2007). Analyses of the National Oceanic and Atmospheric Administration (NOAA) satellite red, green, and blue (RGB) composite images are effective in detecting dust storms, forest fires, and large-scale air pollutants transport (Chung and Le 1984; Chung and Kim 2008; Kim et al. 2012). The brightness temperature difference (BTD) analysis utilizing Advanced Very High Resolution Radiometer (AVHRR), Multi-functional Transport Satellite (MTSAT), and Moderate Resolution Imaging Spectrometer (MODIS) IR channels, which were developed to detect volcanic ash, is effectively being applied to detect large-scale dust storms (Prata 1989; Wen and Rose 1994).

Furthermore, studies indicate that the occurrence and transport of natural dust particles originating from northern China and Mongolia and anthropogenic dust particles originating from eastern China are related to both the high pressure and the low pressure in the synoptic scale in East Asia (Kim et al. 2006; Kurosaki and Mikami 2003; Wang and Fang 2006). The constitution of dust particles changes depending on the transport route in East Asia. The ratio of PM2.5/TSP is low in the natural dust particle cases originating from dust storms, compared with the cases of anthropogenic dust particles originating from eastern China (Kim et al. 2012; Kim et al. 2014). Concentrations of CO, one of the major air pollutants, were high at the background site in central Korea because of the airflow originating from eastern China, rather than from the northern part of the Korean Peninsula or the North Pacific (Kim et al. 2008). However, detailed analyses on constituents of the dust particle cases dependent on synoptic features and transport routes have yet to be conducted.

Accordingly, it is possible to conduct analyses on the long-range transport of air pollutants through satellite observation, ground-based air pollutant measurements, synoptic features, and transport route analyses. Detailed analyses were conducted on hourly variations in TSP, PM10, and PM2.5 mass concentrations and CO concentrations were measured at Cheongwon in central Korea for the cases of long-range dust particle transport caused by dust storms and anthropogenic origin, which accompany transport of the synoptic scale.

Measurement and data analysis

Ground-based measurement of air pollutants

Air pollutant monitoring was done at Cheongwon (127.37 °E, 36.58 °N; 93 m), a rural site located in central South Korea (Fig. 1). The Cheongwon observatory is located approximately 120 km south of Seoul and 120 km east and downwind of the west coast of the Tae-ahn Peninsula. In the case of moderate winds, it takes roughly 3–5 h for the airflow from Tae-ahn to reach Cheongwon. There are no large cities or industrial complexes between Tae-ahn Peninsula and Cheongwon.

Fig. 1

Distribution of columnar a mass concentration (μg cm⁻²) and b CO (molecules cm^-2) retrieved by Aqua MODIS and AIRS level-3 data in East Asia in 2010. The sites of the Tae-ahn Peninsula (T) and Cheongwon (C) are located in central Korea

The TSP, PM10, and PM2.5 mass concentrations were measured, and 1-h averages were measured by using the Rupprecht & Patashnic TEOM (tapered element oscillating microbalance) Series 1400a at Cheongwon in central Korea. The instrument directly measures the mass of dust collected in the filter by means of the TEOM method. Aerosols in the atmosphere are sucked in at a constant amount (16.7 L min⁻¹), the weight of the filter is directly measured by means of the mass transducer, and mass concentrations are calculated on a real-time basis. The monitor calculates the total mass accumulated in the filter by means of the TEOM method and stores average values of mass concentrations at 5 min, 30 min, 1 h, 8 h, and 24 h, respectively. The weight of the filter is stored every 2 s. The temperature inside the measuring device is constantly warmed, dehumidified, and reset to around 50 °C (Kim et al. 2012). The values of TSP, PM10, and PM2.5 have a resolution of 0.1 μg m⁻³ and a precision of ±1.5 μg m⁻³ (1-h average). Equipment (Model 48C) from Thermo Environmental Instruments was used to measure the CO concentration.

Satellite observation

The NOAA AVHRR data have six channels: visible (channel 1), near-infrared (channel 2), short-wave infrared (channel 3a), mid-infrared (channel 3b), and two thermal infrared (channels 4 and 5). Images are made using an RGB composite technique based on channels 1, 2, and 4 in the AVHRR data in order to observe the large-scale air pollutants transport (Chung and Le 1984; Chung 1986; Kim and Chung 2008). Channel 1 is located in the red region, channel 2 in the green, and channel 4 in the blue.

An RGB composite image displays seas and lakes in blue, typically without air pollution due to their low reflection rate. The clouds show as white, and the land shows as green. The large-scale transport of dust storms shows itself as brown and milky. Images of industry-produced large-scale air pollutants transported over seas or lakes show themselves as magenta, and smoke plumes appear in either magenta or a purplish-yellow hue. These RGB composite analysis criteria show that the large-scale transport of dust storms originated from northern China and Mongolia and that anthropogenic air pollutants originated from eastern China.

Also, the BTD (<0) between 11 (channel 4) and 12 μm (channel 5) is used to detect dust particles originating from dust storms (Prata 1989; Wen and Rose 1994; Gu et al. 2003; Iino et al. 2004). Use of BTD to discriminate the volcanic ash cloud from meteorological clouds (water droplets and ice crystals) is based on the split-window method (Prata, 1989). At the same time, the large-scale transport of anthropogenic air pollutants originating from eastern China can be identified in a range larger than 0 by comparing BTD with RGB composite images (Figs. 10 and 12). Large-scale air pollutants transport has been detected routinely with satellite imagery at our research centre in Korea since 1996.

Analysis on synoptic features of long-range transport of air pollutants

Dust particles caused by dust storms that travel in air movements of a synoptic scale are shown on the surface meteorological maps. To identify the dominant synoptic features of weather types during such dust particle cases, surface meteorological maps from the Korea Meteorological Administration (KMA) were analyzed. The Hybrid Single-Particle Lagrangian Integrated Trajectory (HYSPLIT) model (Draxler and Rolph 2003) was used to track the long-range transport path of natural dust particles caused by dust storms and anthropogenic dust particles originating from eastern China in the East Asian region. The HYSPLIT model is a system that computes air parcel trajectories ranging from simple to complex dispersions and deposition simulations. This model includes improved advection algorithms, updated stability and dispersion equations, and the option of whether to include modules for chemical transformations.

Results and discussions

Daily variations in air pollutants in central Korea

The annual average concentrations of PM10, PM2.5, and CO in 2010 reached 43 ± 12 and 25 ± 6 μg m⁻³, and 450 ± 120 ppb, respectively. The air quality standards in Korea are set at 50 μg m⁻³ (annual average) for PM10, 25 μg m⁻³ (annual average) for PM2.5, and 9,000 ppb (an average of 8 h) for CO. Concentrations of PM10 and CO measured in Cheongju in 2010, 10 km east of Cheongwon, registered 66 ± 18 μg m⁻³ and 508 ± 111 ppb, respectively. Cheongwon is a rural area with no industrial or large urban areas all the way westward to the Yellow Sea. The transport of air pollutants originating from inland Korea may affect daily variations of air pollutants at Cheongwon.

Cases of long-range transport of dust particles originating from China and Mongolia were observed by using NOAA satellite observation as well as continuous ground-based measurements at Cheongwon (Chung and Kim 2008; Kim and Chung 2008; Kim et al. 2012; Kim et al. 2014). Figure 1 shows the spatial distribution of columnar mass concentration and CO observed in the East Asian region through Aqua MODIS and Atmospheric InfraRed Sounder (AIRS). Mass concentrations are shown to have increased in the desert and arid areas of northern China and Mongolia due to frequent occurrences of dust storms in spring. Eastern China, with its increasing anthropogenic air pollutant emissions, shows a high level of mass concentration and CO. Therefore, the airflow passing through eastern China and the Yellow Sea has higher anthropogenic air pollution levels than that coming from the northern part of the Korean Peninsula (Kim et al. 2008).

Figure 2 shows variations in daily maximum concentrations of TSP, PM10, PM2.5, and CO, as measured at Cheongwon in 2010. The cases of peak concentrations in daily variations are caused by long-range transport from China and Mongolia. A case of peak TSP concentration in May was affected particularly by airborne pine pollen in the area. Natural dust particle cases caused by dust storms that reached an hourly TSP mass concentration of ≥250 μg m⁻³ or a PM10 mass concentration of ≥190 μg m⁻³ were observed (Kim and Chung 2008). Table 1 shows a summary of maximum concentrations of TSP, PM10, PM2.5, and CO for natural dust particle cases and anthropogenic dust particle cases. It shows that natural dust particle cases occurred in winter as well as spring, registering nine cases on 14 days. In addition, seven cases of the long-range transport of anthropogenic dust particles originating from eastern China were observed for 18 days. Mass concentrations in natural dust particle cases were high, compared with those in anthropogenic dust particle cases. However, anthropogenic dust particle cases caused the ratio of PM2.5 against TSP mass concentrations to rise. Besides, CO concentrations show high levels, compared with those in natural dust particle cases. Out of the natural and anthropogenic dust particle cases, the level of PM10 (100 μg m⁻³ day⁻¹) or PM2.5 (50 μg m⁻³ day⁻¹) exceeded the air quality standards for 5 and 6 days, respectively.

Fig. 2

Day-to-day variations of daily maximums TSP, PM10, PM2.5, and CO concentrations measured at ground level in Cheongwon in 2010. Natural dust particle cases (NDC) and anthropogenic dust particle cases (ADC) were observed at Cheongwon by using analysis of NOAA satellite RGB composite images

Table 1 Maximum mass concentrations of TSP, PM10, and PM2.5 and CO concentrations in the cases of the long-range transport of dust particles observed at Cheongwon in central Korea in 2010

Variations of air pollutants in dust particle cases depending on synoptic features

Natural and anthropogenic dust particles occur on a large scale in China and Mongolia and move toward the Korean Peninsula located in the leeside of China, affected by synoptic features associated with the development of high- and low-pressure systems (Kurosaki and Mikami 2003; Wang and Fang 2006; Kim et al. 2006). Figure 3a, b shows the composite surface pressure when dust storms occurred in northern China and Mongolia and when natural dust particle cases with continental background airflow (n-CBAs) (three cases) were observed at Cheongwon in central Korea in 2010. Dust storms are caused by cold northerly gusty winds in the forward side of a high-pressure system (1,028 hPa) following low pressures (1,004 hPa) in northern China and Mongolia. High pressure moves quickly southward and continues to develop further (1,032 hPa), locating itself in eastern China. In addition, a low pressure also develops (992 hPa), moving eastward and locating itself in the northeastern part of the Korean Peninsula, as shown in Fig. 3b. The natural dust particles originating from dust storms are transported to central Korea with the continental background airflow that passes the northwestern part of the Korean Peninsula. In an hourly cluster analysis and cluster mean of a 3-day isentropic backward trajectory during hours of n-CBAs, airflow is arriving at Cheongwon in central Korea from the northwestern part of the Korean Peninsula (Fig. 3c, d).

Fig. 3

Composite surface pressure (solid contour, with an interval of 4 hPa) when a dust storms (red circle) occurred and b natural dust particle cases (red circle) were observed at Cheongwon in central Korea for n-CBA in 2010. Also, c clusters and d cluster mean of 3-day isentropic backward trajectory for n-CBA

On the other hand, when high pressure (1,024 hPa) and low pressure (1,000 hPa) develop in northern China and Mongolia, they expand to cover central and eastern China, and the Korean Peninsula is situated to the southeast of the low-pressure system (Fig. 4a). Subsequently, low pressure passes over the Korean Peninsula and dust storms in the foreside of the high-pressure flow in (Fig. 4b). The case where a dust storm is transported to the Korean Peninsula by way of eastern China and the Yellow Sea is classified as a natural dust particle case with continental polluted airflow (n-CPA). An n-CPA with six cases occurs in central Korea while moderate, slow-moving high pressure compared with that of n-CBAs passes central and eastern China and the Yellow Sea. In the Korean Peninsula, southwest airflow blows before a dust storm in the foreside of a high-pressure system arrives. Anthropogenic air pollutants originating from eastern China move to central Korea by way of the Yellow Sea. Furthermore, an n-CPA is added to anthropogenic air pollutants because a low-pressure system passes and a dust storm in the foreside of a slow-moving, high-pressure system traverses across eastern China. In the hourly cluster analysis of a 3-day isentropic backward trajectory during hours of n-CPA, airflow arrives at Cheongwon, in central Korea, by way of eastern China and the Yellow Sea (Fig. 4c). The cluster mean of a 3-day isentropic backward trajectory during hours of n-CPA traverses across the Shandong Peninsula and the Yellow Sea (Fig. 4d).

Fig. 4

Composite surface pressure (solid contour, with an interval of 4 hPa) when a dust storms (red circle) occurred and b natural dust particle cases (red circle) were observed at Cheongwon in central Korea for n-CPA in 2010. Also, c clusters and d cluster mean of 3-day isentropic backward trajectory for n-CPA

In Fig. 5a, anthropogenic dust particles over eastern China float on warm southwesterly wind with the Yellow Sea region located in the rear side of a high pressure (1,024 hPa) and in the southeast part of a low pressure (1,004 hPa). An anthropogenic dust particle case with continental polluted airflow (a-CPA) with five cases occurs in central Korea. When a low-pressure system traverses the Korean Peninsula, air pollutant concentration falls due to cold north–northwest airflow in the backside of cold fronts and in the foreside of a fast-moving, high-pressure system. In contrast, cases where the Yellow Sea region is affected by stagnant air masses of high pressure (1,024 hPa) are anthropogenic dust particle cases with regionally polluted airflow (a-RPA) with two cases in Fig. 5b.

Fig. 5

Composite surface pressure for a a-CPA and b a-RPA observed in Cheongwon in central Korea in 2010

The average concentrations of TSP, PM10, PM2.5, and CO were compared 12 h before, during hours of dust particle cases, and 12 h after for n-CBA and n-CPA (Table 2). Overall, an n-CBA shows levels of TSP, PM10, PM2.5, and CO concentrations lower than n-CPA, and PM2.5/TSP mass concentration ratios reached 35, 12, and 14 %, respectively, before, during, and after hours of n-CBA. While a dust particle case is occurring, CO concentration is rising along with increasing mass concentrations during hours of n-CBA. On the other hand, PM2.5/TSP mass concentration ratios were higher with 44, 26, and 23 %, respectively, before, during, and after hours of n-CPA, compared with the n-CBA. Besides, CO concentration increased before hours of the n-CPA and increased during hours of the n-CPA.

Table 2 Average mass concentrations of TSP, PM10, and PM2.5 and CO concentrations 12 h before, during hours, and after 12 h for n-CBA and n-CPA, including the maximum mass and CO concentrations for a-CPA and a-RPA

Also, the maximum TSP, PM10, PM2.5, and CO concentrations were averaged for an a-CPA and a-RPA in Table 2. While a-CPA TSP and PM10 mass concentrations do not go as high as in natural dust particle cases, PM2.5 mass concentration runs up higher, causing PM2.5/TSP mass concentration ratios to reach as high as 69 %. In addition, CO concentration reached 984 ± 304 ppb, showing that it ran up higher than in natural dust particle cases. While an a-RPA does not show a high level of mass concentrations, the PM2.5/TSP mass concentration ratio reached 63 %, and CO concentration ran as high as 878 ± 37 ppb.

Air pollutant variations in the case of long-range transport of dust particles

Natural dust particle cases

Figure 6 shows an analysis of NOAA satellite RGB composite images and BTD on March 12 (n-CBA) and 20 (n-CPA), 2010. The RGB composite image on March 12 shows that a dust storm is distributed in a band-like shape ranging from the Korean Peninsula to the Yellow Sea. A BTD (from −1.5 to −0.5 K) analysis shows the dust storm distribution of a large scale covering the Shandong Peninsula, the Yellow Sea, and the central and northeast parts of the Korean Peninsula, on March 12. At the same time, a dust storm was also occurring in inner Mongolia. On March 20, the BTD analysis shows distributions of dust storms consistent with the RGB composite image within the range −7.4 to −0.5 K.

Fig. 6

NOAA satellite RGB composite image (left) and BTD (right) showing a dust storm (DS) on a March 12 and b March 20, in 2010

Figure 7 shows hour-to-hour variations of TSP, PM10, and PM2.5 mass concentrations and CO concentrations caused by dust storms measured at Cheongwon during March 12–13 (n-CBA) and March 19–21 (n-CPA) in 2010. A natural dust particle case on March 12 shows variations of TSP, PM10, and PM2.5 mass concentrations reaching 297, 206, and 67 μg m⁻³, respectively, and CO concentrations rising to 563 ppb concurrently. However, on the morning of March 20, TSP, PM10, and PM2.5 mass concentrations reached 143, 111, and 62 μg m⁻³, respectively, and CO concentrations rose to 1,150 ppb before dust storms arrived at Cheongwon. With dust storms flowing in on the afternoon of March 20, mass concentrations and CO ran up high again. The maximum TSP, PM10, PM2.5, and CO ran up to 1,032, 673, and 174 μg m⁻³, and 993 ppb, respectively. Moreover, the daily average of PM10 and PM2.5 reached 166 and 65 μg m⁻³, respectively, thus exceeding the air quality standards of Korea.

Fig. 7

Variations of hourly TSP, PM10, PM2.5, and CO concentrations measured in Cheongwon on a March 12–13 and b March 19–21 in 2010

Figure 8 shows a surface meteorological chart illustrating a dust storm that occurred in the northern parts of China and in Mongolia and then moved toward the Korean Peninsula during March 11–12 and March 19–20, 2010. In Fig. 8a, a low pressure of 998 hPa is located to the southeast of Mongolia at 06 UT (1500 LST) on March 11, and a high pressure of 1,021–1,028 hPa is developing to the west. A dust storm is occurring in the Gobi Desert with cold northwest airflow in the foreside of high pressure. A low-pressure system is moving to the east, and a dust storm in the foreside of the high pressure is moving to the east–southeast at 03 UT (1200 LST) on March 12. High-pressure and low-pressure systems were located in the northwest and northeast of the Korean Peninsula, respectively, and an n-CBA occurred in Cheongwon at 12 UT (2100 LST) on March 12. Mass concentrations of TSP, PM10, and PM2.5 and CO concentrations that increased concurrently due to the influx of dust storms were observed at Cheongwon.

Fig. 8

Surface meteorological charts showing the occurrence and movement of dust storms (red circle) in East Asia during a March 11–12 and b March 19–20, in 2010

On the other hand, Fig. 8b shows that a low-pressure system of 998 hPa is located in northeastern China and a strong high-pressure system of 1,040 hPa is developing in western Mongolia at 00 UT (0900 LST) on March 19. A dust storm is occurring in the Gobi Desert in the foreside of a high pressure. Accordingly, as the high pressure is being expanded southeastwardly, a low pressure is located in the northern part of the Korean Peninsula at 00 UT (0900 LST) on March 20. Therefore, a warm southwest airflow was blowing in central Korea, and mass concentrations of TSP, PM10, and PM2.5 and CO concentrations increased, affected by anthropogenic air pollutants coming from eastern China. A dust storm in the foreside of high pressure arrived in central Korea by way of eastern China and the Yellow Sea, causing an n-CPA at 09 UT (1800 LST) on March 20. Mass concentrations of TSP, PM10, and PM2.5 and CO concentrations increased sharply at Cheongwon.

Figure 9 shows an analysis of a 3-day isentropic backward trajectory when an n-CBA and an n-CPA occurred at Cheongwon on March 12 and 20, respectively. While the n-CBA on March 12 shows a dust storm moving toward northwestern Korea, the n-CPA on March 20 illustrates that the dust storm was traversing from eastern China to the Yellow Sea. The continental background airflow from the northern part of the Korean Peninsula shows a level of CO concentrations lower than the continental polluted airflow from eastern China (Kim et al. 2008; Turnbull et al. 2011).

Fig. 9

Three-day isentropic backward trajectories arriving in Cheongwon based on NOAA HYSPLIT model (Draxler and Rolph 2003)

Figure 10a shows the surface meteorological chart (1500 LST), NOAA satellite RGB composite image, and BTD image (1356 LST) on December 1, 2010. Northern China and the Gobi Desert in southern Mongolia are located in the foreside of a high-pressure system and in the southwestern part of a low-pressure system. In the Gobi Desert, a dust storm occurs due to a cold north–northwest airflow. Situated in the backward side of a high-pressure system located in Japan and the southeastern part of the low-pressure located in China, the Korean Peninsula was affected by the airflow moving from eastern China. NOAA RGB composite image illustrates that anthropogenic air pollutants on a large scale expanding toward central Korea by way of the Yellow Sea from eastern China are distributed. The BTD (0.5–0.8 K) is consistent with the distribution of large-scale air pollution in RGB composite image.

Fig. 10

Dust storm (DS) originating from northern China and southern Mongolia and large-scale air pollutants transport (LSAPT) originating from eastern China observed by using meteorological charts (top), NOAA satellite RGB composite images (middle), and BTD (bottom) on a December 1 and b December 2, 2010

Figure 10b shows the surface meteorological chart (1200 LST), NOAA satellite RGB composite image, and BTD image (1346 LST) on December 2, 2010. A surface meteorological chart shows that high pressure originating from northern China and Mongolia expands toward the Korean Peninsula and that low pressure is located in the northern part of the Yellow Sea. A dust storm developed in the foreside of high pressure has moved up to the vicinity of the Shandong Peninsula. However, located in the southeastern part of low pressure, central Korea has been affected by the southwest airflow since December 1. In the NOAA satellite RGB composite image, it is difficult to distinguish distributions of dust storms from anthropogenic air pollutants. However, in the BTD image of dust storms (from −1 to −0.5 K), in the vicinity of Bohai Bay, large-scale distributions of anthropogenic air pollutants (0.5–0.8 K) flowing into the Korean Peninsula from the Yellow Sea were distinguishable.

Figure 11 shows an analysis of a 3-day isentropic backward trajectory and TSP, PM10, PM2.5, and CO concentrations of the airflow arriving at Cheongwon for a 3-day period, from December 1 to 3, 2010. Central Korea has been affected by the southwest airflow from December 1 to 2. On December 2, the maximums of TSP, PM10, PM2.5 mass concentrations were measured at 217, 193 and 148 μg m⁻³, respectively, and the PM2.5/TSP mass concentration ratio reached 71 %. CO concentration also increased to 1,390 ppb, like the mass concentration on December 2. Consequently, an n-CPA occurred on December 3, and TSP, PM10, PM2.5, and CO concentrations rose again to 427, 373, and 182 μg m⁻³ and 1,595 ppb, respectively.

Fig. 11

a Three-day isentropic backward trajectories arriving at Cheongwon and b variations of hourly TSP, PM10, PM2.5, and CO concentrations during December 1–3, 2010

Anthropogenic dust particle cases

Figure 12 shows analyses of surface meteorological charts and NOAA satellite RGB composite images and BTD images on October 18 and November 7, 2010. The Yellow Sea region was affected by a large stagnant high-pressure system located in China from October 17 to 20 (Fig. 12a). An a-RPA was observed in central Korea from October 17 to 20. NOAA RGB composite images illustrate the distribution of anthropogenic air pollutants that have originated from eastern China, reaching Korea and the East Sea by way of the Yellow Sea. An analysis of BTD shows distributions of air pollutants ranging from 1.8 to 2.5 K. On the other hand, an a-CPA was observed in central Korea, which is located in the southeastern part of a low pressure on November 7 (Fig. 12b). RGB composite image and an analysis of BTD of NOAA satellite AVHRR data identified large-scale air pollutants moving toward the Korean Peninsula from eastern China by way of the Yellow Sea on November 7.

Fig. 12

Large-scale air pollutants transport (LSAPT) originating from eastern China observed by using meteorological charts (top), NOAA satellite RGB composite images (middle), and BTD (bottom) on a October 18 and b November 7, 2010

Figure 13 shows variations of TSP, PM10, PM2.5, and CO for October 16−21 and November 5−7, 2010. TSP, PM10, and PM2.5 mass concentrations gradually rose, reaching 114, 101, and 73 μg m⁻³, respectively, on October 20, and CO concentration registered 841 ppb on October 19. As a rule, in urban areas, the mixed layer height rises when the surface of the ground is heated by solar radiation in the afternoon. The ground-based air pollutants are mixed vertically, thus causing concentrations to fall. However, mass concentrations and CO concentrations increased during the daytime when the mixed layer height rose during the period of October 17–20. The mixed layer height went up as far as 900 m, with air pollutants that are moving up to upper layers flowing in toward the surface of the ground. On October 20 in particular, O₃ concentrations increased to a maximum of 92 ppb due to the influx of precursors and photochemical reactions.

Fig. 13

Variations of TSP, PM10, PM2.5, and CO concentrations and surface mixed heights during a October 16–21 and b November 5–7, 2010

On the other hand, mass concentrations and CO concentrations gradually increased from November 6, reaching TSP 207 μgm^-3, PM10 194 μgm^-3, and PM2.5 154 μgm^-3 and with CO reaching 1,117 ppb at 2100 LST. Daily averages of PM10 and PM2.5 exceeded the air quality standards during the period of November 6–7. In the afternoon of November 7, low pressure passed through, and northwest airflow in the foreside of the high pressure caused concentrations to decrease.

Conclusion

The hourly variations in mass concentrations of TSP, PM10, and PM2.5 and CO concentrations were analyzed in detail in dust particle cases, where natural dust particles originating from northern China and Mongolia and anthropogenic dust particles originating from eastern China moved to central Korea in the leeside of China in a synoptic scale in 2010. Large-scale distributions of dust storm and anthropogenic air pollutants were distinguished through NOAA satellite RGB composite images and BTD analyses (dust storms; from −7 to −0.5 K and anthropogenic air pollutants transport; 0.5–3.5 K) as well as ground level measurements. Natural dust particle cases (nine cases and 14 days) were observed at Cheongwon in central Korea in 2010. At the same time, seven anthropogenic dust particle cases were observed on 18 days. The air quality standards of PM10 (100 μg m⁻³ day⁻¹) and PM2.5 (50 μg m⁻³ day⁻¹) were exceeded for 5 and 6 days, respectively. The CO concentration levels in anthropogenic dust particle cases (954 ppb) ran up higher than that in natural dust particle cases (812 ppb). In anthropogenic dust particle cases, the PM2.5/TSP mass concentration ratio registered 67 %, a level higher than that of natural dust particle cases (25 %).

Dust storms are caused by cold northwest airflow in the foreside of high pressure following low pressure in northern China and in Mongolia and then moves to the east. Variations of mass concentrations of TSP, PM10, and PM2.5 and CO concentrations of natural dust particle cases in central Korea turn out differently depending on the route where the foreside of the high pressure passes. n-CBAs (three cases) show that variations where high pressure in Mongolia and northern China move toward the Korean Peninsula along with dust storms that are carried by northwest airflow flow in to central Korea, and mass concentrations of TSP, PM10, and PM2.5 and CO concentrations also increase.

However, in n-CPAs (six cases), the Korean Peninsula, which is located to the southeast of low pressure, is affected by the southwest airflow coming from eastern China, and subsequently, dust storms in the foreside of the high pressure move toward the Korean Peninsula by way of central and eastern China and the Yellow Sea. Therefore, CO concentrations increase before dust storms flow in, followed by variations where CO concentrations, along with dust storms that flowed in, increased. When the n-CPA mass concentrations were high, the PM2.5/TSP mass concentrations ratio also ran high, compared with those of the n-CBA. In particular, the PM2.5/TSP mass concentration ratio ran as high as 44 %, before dust storms flowed in.

Meanwhile, anthropogenic dust particle cases were divided into an a-CPA, where central Korea is affected by air pollutants carried over a long distance from eastern China by warm southwest airflow, and an a-RPA, where the Yellow Sea area is affected by stagnant high pressure. The a-CPA shows a high level of all mass concentrations, the PM2.5/TSP mass concentration ratios, and the CO concentrations.