Sulfur, Oxygen, and Hydrogen Isotope Compositions of Precipitation in Seoul, South Korea

Geochemical Journal, Vol. 46, pp. 443 to 457, 2012

Sulfur, oxygen, and hydrogen isotope compositions of precipitation in Seoul, South Korea

CHUNGWAN LIM,1 INSUNG LEE,1* SANG-MOOK LEE,1 JAE-YOUNG YU2 and ALAN J. KAUFMAN3

1School of Earth and Environmental Sciences, Seoul National University, Seoul 151-742, Korea 2Department of Geology, Kangwon National University, Chuncheon 200-701, Korea 3Department of Geology, University of Maryland, College Park, MD 20742-4211, U.S.A.

(Received December 28, 2011; Accepted August 2, 2012)

To evaluate the source of sulfur and the extent of seasonal and local characteristics in Seoul’s precipitation chemistry, we measured sulfur, oxygen, and hydrogen isotopic ratios of precipitation. The pH of precipitation ranges from 4.6 to 7.0 2+ 2+ 2– – in Seoul. Precipitation shows positive correlations between ions in the Ca –Mg –SO4 –NO3 system, indicating that the dissolution of Ca and Mg particles by H2SO4 and HNO3 from the combustion of fossil fuels is a major process controlling the chemical composition of snow and rain. The values of oxygen and hydrogen isotope composition of rain range –15.7~–3.2‰ and –114.7~–18.0‰, respectively. The H and O isotope values in the summer are plotted near the global meteoric water line by Craig (1961). The d-excess values in winter are higher than 10, which indicate that the source of rain during winter is a dry air mass from continental China. The sulfur isotope ratio of precipitation in Seoul ranges from +3.0 to +7.3‰, which is similar to typical metropolitan air pollution levels. The data suggests that the main source of δ34 sulfur is SO2 produced by the combustion of fossil fuels. The pollutant SO4 is estimated to have a Snss range from +1.0 to +6.2‰ in the Seoul area. The δ34S values of precipitation range from +5.1 to +7.3‰ (mean +6.4‰) in winter and from +3.0 to 4.8‰ (mean +4.0‰) in summer. The higher δ34S values in winter (December and February) seem to be correlated to the air mass from northern China, of which the δ34S values of oil or coal is higher than that of southern China. The lower sulfur isotopic values in summer (June to August) are correlated to the air mass moving from southern China. Isotopic

composition and chemical concentrations of SO4 depend on wind provenance, thus supporting the idea that a seasonally transported source for the pollutant sulfur is from China.

Keywords: sulfur isotope, oxygen, hydrogen, pollution, precipitation, sulfate, Korea, East Asia

fuel combustion in industrialized and heavily populated INTRODUCTION regions. Potential sources of sulfur oxides include: oil- The chemistry of acidic precipitation worldwide has refinery operations, oil- or coal-based power generation, been intensively studied for the past several decades. Acid automobile combustion, and other oil- or coal-reliant fa- rain is known to cause serious environmental damage in cilities. On the other hand, major ion distributions in pre- sensitive environments (Ayers and Yeung, 1996; cipitation are primarily controlled by contributions from Kulshrestha et al., 2003; Lee et al., 2000; Das et al., 2005; marine (sea-salt aerosols), terrestrial (soil dust, biologic Cape and Leith, 2002; Andronache, 2004; Zunckel et al., emissions) and anthropogenic (industrial, biomass burn- 2003; Yamaguchi et al., 1991). The increase of fossil fuel ing, vehicle emissions, and others) sources (Junge, 1963; burning in east Asian countries has likely intensified acid Nakai and Takeuchi, 1975). rain in east Asian countries including Korea, China, Ja- SO2, which is mainly produced by anthropogenic ac- 2– pan, and Russia (Lee et al., 2000; Park et al., 2000; Chun tivities, undergoes oxidation to form SO4 and the reac- et al., 2000; Park and Cho, 1998), while there are consid- tion is enhanced by metallic ions in urban atmospheres erable contributions of sulfur from Yellow sand (Asian (Newman et al., 1991). However, substantial amounts of 2– dust) from the inland of the continent. atmospheric SO4 arise from sea spray in coastal areas 34 32 Sulfur oxides mainly as SO2 are among the major pre- as well (Nielsen, 1978). The abundance of S/ S ratios 2– cursors of acid rain, and they are released through fossil in snow and rainwater derived from SO4 bearing precipitation may be used to delineate anthropogenic S in the atmosphere, provided that the isotopic composition of pollutant S is distinct from S originating from natural *Corresponding author (e-mail: [email protected]) sources. Sulfur isotopes have been widely used to inter- Copyright © 2012 by The Geochemical Society of Japan. pret substances’ origins, forming conditions, and envi-

443 ronmental alterations (Mizutani and Rafter, 1969; Nakai and Takeuchi, 1975; Nissenbaum, 1978; Nriagu et al., 1987; Ohizumi et al., 1991; Yu and Park, 2004). Also, there is a distinct difference in the sulfur isotope ratios of fossil fuels such as coal and petroleum. Sulfur isotope fractionation is also demonstrated by Hong et al. (1992) and Maruyama et al. (2000) through their combustion experiments. Sulfur exists in small amounts in the atmosphere, but when anthropogenic sources of sulfur that cause air pollution are introduced, the atmospheric sulfur isotopic chemistry changes in a very subtle manner. Because of this phenomenon, a study on the sulfur isotope ratio of sulfate ions in precipitation is considered a useful method in tracking the source and distribution cycle of air pollutants. Thus the sulfur isotopic and major element composition of rainwater along with meteorological information can be used to trace the sources of sulfur in precipitation, and to evaluate contributions from different sources (Krouse, 1980; Krouse and Case, 1981; Na et al., 1995; Yu and Park, 2004). In this case variation among sources must be isotopically significant and changes in isotopic composition during transport and transformation must be trivial. Additional information on both sources and mixing phenomena can be obtained by combining concentration and isotopic composition data with meteorological Fig. 1. Map of the sampling location of study area (black cir- parameters (e.g., wind direction, temperature etc.). Iso- cle; in Seoul) and reference area (open triangle; Chuncheon topic studies on water samples in Korea have been made by Yu and Park, 2004 and open rectangular; Chonju by Na et for oxygen and hydrogen isotopic compositions of pre- al., 1995). cipitation in Taejeon and Seoul (Lee and Chang, 1994) and Cheju Island (Lee et al., 1997); studies have also been made for isotopic compositions in ground water (Na et forming process whereas the position of any point on the al., 1995) for sulfur, oxygen, and hydrogen isotopes of slope = 8 line is determined by the rainout process acid mine drainage (Yu and Coleman, 2000). Data on (Dansgaard, 1964). These measurements could potentially sulfur isotope compositions of precipitation in Korea are provide a means for determining the relative importance very rare. of different air masses for summer and winter precipita- The oxygen and hydrogen isotopic compositions of tion (Epstein and Mayeda, 1953; Craig, 1961; Dansgaard, precipitation in natural waters are governed by tempera- 1964; Lee and Lee, 1999). In this study the isotopic ra- ture, evaporation, and condensation. Deuterium excess (d) tios of sulfur, hydrogen and oxygen of wet depositions in is a second order parameter derived from δD and δ18O Seoul, Korea are determined through the analyses of snow values. Deuterium excess reflects the different sensitivi- and rainwater to distinguish between different precipita- ties of these isotopes to kinetic effects in the hydrologic tion events and their associated meteorological conditions. cycle. Theory predicts that deuterium excess responds The purpose of this research is to characterize the pre- primarily to changes in sea surface temperature (SST), cipitation chemistry of the Seoul area, namely for sulfur, humidity, and wind speed at the moisture source oxygen, and hydrogen isotope compositions, and to evalu- (Dansgaard, 1964). The value of deuterium excess is de- ate the source of sulfur and the extent of seasonal and termined from the equation as d = δD – 8δ18O. The val- local characteristics of snow and rain. This data will pro- ues of deuterium excess of any sample can be interpreted vide a context for potential solutions to environmental as the intercept with the δD axis (for δ18O = 0) of the line problems in East Asia. The present work deals with ur- with slope ∆δD/∆δ18O = 8 which passes through that ban atmospheric conditions in Seoul, a very densely popu- point; such a line presumably would be the locus of all lated city in Korea, surrounded by a wide and crowded precipitation samples which are derived from that par- industrial belt. Consequently air pollution poses a poten- ticular air mass by rainout; as will be discussed, the “d- tial problem to environment. For this reason, a system- parameter” according to this view relates to the vapor- atic chemical and isotopic study was undertaken to quan-

444 C. Lim et al. Table 1. The general weather conditions of precipitation in Seoul

Date (dd-mm-yy) Precipitation type (season) Direction Velocity (m/s) Precipitation (mm) T (°C) pH EC (µS/cm) Condition 16-Dec-00 Snow (Wt) NW 1.5 5.4 −0.6 4.92 16.3 24-Dec-00 Snow (Wt) NW 2.2 13.9 −2.2 4.87 7.53 1-Jan-01 Snow (Wt) NW 3.7 21.7 0.4 5.07 5.92 Heavy snowfall 27-Jan-01 Snow (Wt) NW 3.6 1.9 0.0 4.95 72.2 Yellow sand 9-Feb-01 Snow (Wt) W 2.6 0.2 −0.9 5.87 6.79 15-Feb-01 Snow (Wt) NE 3.2 23.4 3.4 6.17 18.1 27-Feb-01 Rain/Snow (Wt) N 1.7 7.7 3.0 5.81 9.48 4-Mar-01 Snow (Wt) NW 4.9 11.2 1.0 6.55 96.2 Yellow sand 28-Mar-01 Rain/Snow (Wt) NW 1.8 2.7 2.9 6.77 11.4 11-Apr-01 Rain (Sp) WNW 3.0 6.5 9.3 6.99 55.3 Yellow sand 7-May-01 Rain (Sp) SW 1.5 5.4 18.3 6.86 14.3 Dry Season 22-May-01 Rain (Sp) ENE 1.3 7.9 22.3 6.69 12.4 18-Jun-01 Rain (Sm) SW 0.8 36.2 20.2 6.58 36.7 25-Jun-01 Rain (Sm) SW 2.0 7.8 22.6 6.20 16.8 Typhoon 5-Jul-01 Rain (Sm) SW 4.5 21.1 24.6 6.47 52.7 10-Jul-01 Rain (Sm) W 1.0 14.2 21.8 6.01 66.1 Rainly season 15-Jul-01 Rain (Sm) SW 1.0 273 23.2 5.51 6.53 Heavy rainfall 7-Aug-01 Rain (Sm) ENE 0.5 17.7 26.7 5.45 50.7 15-Aug-01 Rain (Sm) S 0.7 162 25.2 4.60 25.6 6-Sep-01 Rain (At) SSW 1.2 43.4 22.7 5.20 64.2 30-Sep-01 Rain (At) ENE 1.6 5.4 15.7 5.10 463 10-Oct-01 Rain (At) SW 2.6 46.3 13.8 4.80 14.8

13-Feb-01 Stream water   5.2 6.72 125 19-May-01 Stream water  16.3 7.03 87.4 1-Jul-01 Stream water  23.5 6.55 130 8-Oct-01 Stream water  12.4 7.36 78.7

13-Feb-01 Ground water   7.4 6.57 201 19-May-01 Ground water  11.8 7.19 165 1-Jul-01 Ground water  15.4 8.22 157 8-Oct-01 Ground water   9.5 7.63 216

Wt, winter; Sp, spring; Sm, summer; At, autumn.

tify the chemistry and S isotopic composition of aqueous The climate of Seoul and Chuncheon is temperate and 2– SO4 in a variety of precipitation events at Seoul Na- shows distinct seasonality. In winter, winds commonly tional University in Gwanak-gu, Seoul, during the period originate from the north-northwest or north-northeast re- of December 2000 to October 2001. sulting in a wide range of temperatures typical of northern continents. Winters are bitterly cold and are influ- enced primarily by Siberian air masses. In the summer, STUDY AREA prevailing winds commonly come from the south-south- Seoul is the capital city of South Korea, which ac- east, bringing in the maritime Pacific air mass into the commodates about ten million people with more than three region which results in high temperature and humidity. million motor vehicles, as well as numerous industrial Temperatures recorded over a 30-year period show typi- plants within the metropolitan area. cal cyclic variation between –7.0 and +30.0°C with a sea- The referenced location is Chuncheon, which occu- sonal mean temperature of about –2.0°C in winter and pies the eastern part of the Korean peninsula. This area is +25.0°C in summer (Table 1). The average annual pre- surrounded by developed and developing mining indus- cipitation is about 1330 mm (data from Korea Meteoro- tries and forested areas. The emissions of atmospheric logical Administration) with more than half of the total pollutants from Chuncheon are relatively insignificant to rainfall concentrated in the summer monsoon. Winter pre- emissions from domestic fossil fuel consumption (Yu and cipitation takes up less than 10% of the annual total. High Park, 2004). The last referenced area is Chonju (Na et temperature and humidity are caused by the northern Pa- al., 1995), which has a few industrialized areas surrounded cific air mass usually in the monsoonal rainy season from by agricultural fields (Fig. 1). June to September in Korea.

Stable isotope compositions of precipitation in Seoul 445 were carried out separately. Samples for oxygen isotopic METHODOLOGY analysis were prepared through H2O–CO2 equilibration Samples of snow and rainwater were collected for (Epstein and Mayeda, 1953). About 2 ml of each water oxygen, hydrogen, and sulfur isotope measurements from sample was equilibrated with CO2 of known isotopic com- ° December, 2000 to October, 2001, including both dry and position at 25 C. The CO2 gas was then extracted, rainy seasons. Stream and ground waters from nearby area cryogenically purified, and analyzed using the mass Gwanak were collected and analyzed for reference. The spectrometer. For deuterium analysis, metallic zinc was sampling site is located within the Gwanaksan granite, used to produce hydrogen gas (Coleman et al., 1982). The just south of the Seoul granitic batholith (Fig. 1). The oxygen and hydrogen isotopic compositions of the sam- meteoric water (rain and snow) samples in Seoul were ples were determined using the VG Isotech Prism II mass acquired on the roof of a five-story building at the School spectrometer at Korea Basic Science Institute (KBSI). The of Earth and Environmental Sciences in the Seoul Na- analytical reproducibility was ±0.1‰ for δ18O vs. SMOW tional University Campus. The referenced Chuncheon and ±1‰ for δD vs. SMOW. samples of meteoric water were collected at each precipi- When measuring the amount of sulfur isotope of a tation event of Kangwon National University in sample that has such a small amount of sulfur, it is neces- Chuncheon from October, 2000 to July, 2001 (Yu and sary to concentrate and extract the sulfur. Precipitation 2– Park, 2004). contains an infinitesimal amount of sulfur in the SO4 Precipitation samples were collected using contain- form so in order to measure sulfur isotopes, the easily- ers with large funnels directly from the atmosphere. Sam- performed Barium sulfate precipitation method was used ples of snow and rainwater were acquired manually us- in this study. The dissolved SO4 in the samples was pre- ing a bulk deposition sampler equipped with a stainless cipitated as BaSO4. For sulfur isotopic analysis, about 3 steel funnel (65 cm in top diameter), a Teflon bottle set mg of BaSO4 powder was suitable. The following method (500 ml) and plastic boxes (60 cm × 40 cm × 15 cm) on was used in concentrating and extracting sulfur. the roof of the building (location: E 126°57′18″, N The collected sample was immediately filtered through 37°27′14″, 171 m above sea level). The lid of the sam- a 0.45 µm Millipore filter and after weighing it with a pler was manually removed to collect samples. After a mass cylinder, it was filled into a beaker. 1~2 ml of 1N precipitation event, the sample was immediately trans- HCl was added to the sample to acidify it and after cov- ported to the laboratory. The pH of the snow or rainfall ering the beaker with a watch glass, it was heated up to was measured immediately after sample collection. Snow its boiling point on a hot plate. After the heating was con- samples were collected in the plastic boxes and were al- tinued for one day, the system was cooled down at room lowed to melt at laboratory temperature. More than 10 temperature. The precipitated BaSO4 was rinsed with dis- liters of sample were collected for the S isotope analy- tilled water, dried, powdered and treated with HCl acid ses. The collected sample was passed through a 0.45 µm in order to remove oxygen-bearing compounds other than Millipore filter and each aliquot (500 ml) of filtered sam- BaSO4. Then BaCl2 was added in order to concentrate ple was acidified with HNO3 for cation analyses (EANET, the formed crystals for convenient collection. Next, they 2000). were filtered with glass wool filters. BaSO4 was concen- General weather information related to precipitation trated on the filter and after drying it below 100°C, it was was obtained from the weather station of the Korea Me- weighted. BaSO4 was then packed and stored with the teorological Administration located about 5 km NW of filter. the sampling site. Barium sulfate powders were recovered and the sam- Major chemical compositions of water samples were ples were placed in quartz tubes then decomposed by heat- analyzed using the Inductively Coupled Plasma—Atomic ing to at least 950°C (Yanagisawa and Sakai, 1983). Be- Emission Spectrometer (SHIMADZU/ICPS-1000IV) at fore heating, the sulfur isotope ratio analyzing SO2 gas the National Center for Inter-University Research Facili- was produced using Rob’s method which is mixing Cu ties (NCIRF), Seoul National University for Na, Mg, K, and Mg wool by 1:2 in a vacuum line for oxidation. The and Ca. Ion Chromatography (at the Center for Mineral resulting SO2 was separated from trace contaminants by Resources Research, Korea University: DIONEX-120 cryogenic distillation using three cold baths: liquid ni- Automated Dual Column I.C.) was used for the analyses trogen (–196°C), liquid nitrogen-pentane slush (–131°C) – 2– – + ° of Cl , SO4 , NO3 , and NH4 . The external precision, or and dry ice-ethanol (–86 C). In the vacuum line, gas pu- reproducibility, is defined as two standard deviations (2σ) rification was carried out by getting rid of H2O using dry of the average values of major chemical ions of repeated ice temperature and collecting pure SO2 gas at the freez- 2– analyses of the same sample during various data acquisi- ing point. All the sulfur contained in the SO4 ions was tions (WMO Manual; Allan, 2004). converted to pure SO2 gases. The prepared SO2 was Oxygen and hydrogen isotope sample preparations analyzed for its δ34S composition using the Isoprime gas

446 C. Lim et al. 1000 )

Conductivity ( s/ cm) 100

7 10

6 1 Conductivity ( s/ cm

H pH p 5

4 )

300 20

10 Temperature ( C) 250 0

50 Precipitation (mm) Temperature ( C –10

Precipitation (mm) 0 ) 1 1 1 1 1 1 1 1 1 /yy 0 0 0 /0 /0 /01 0 1 /0 /0 /0 b/ b/ r r / /0 g g p/01 p n/01 n/01e e a a e e mm Dec/00Dec/00Ja Ja F Feb/ F M M Apr/01May/01May Jun/01Jun Jul/01Jul Jul/01Au Au S S Oct/01(

Fig. 2. Seasonal variations of precipitation, air temperature, pH and conductivity in Seoul.

source mass spectrometer at KBSI. The δ34S values are 1). This variation in pH values is attributed to the inten- given in per mil [‰] units, with reference to CDT. sity of rainfall received and the spring time Yellow sand phenomenon (Table 2). The Yellow sand phenomenon was observed across RESULTS AND DISCUSSION the entire Korean peninsula in the January 1st, March 4th, Chemistry of precipitation and April 11th of 2001. A trajectory analysis showed that Table 1 summarizes the pH and temperature of rain- the Yellow sand could reach Seoul within two days from water, stream, and ground water samples collected dur- the Gobi Desert (Chun et al., 2000, 2001). The pH rises ing this study along with prevailing wind direction, wind and alkali ion concentration abruptly increases above velocity, and rainfall amounts from December 2000 to mean values during the Yellow sand event as shown in October 2001 in Seoul. The pH of the precipitation sam- the EC values (Table. 1). Subsequently heavy snow and ples range from 4.6 to 7.0, averaging 5.80 in Seoul (Fig. rainfall in Seoul and Chuncheon diluted the atmospheric 2). Similar mean pH values in Seoul and Chuncheon were flux so that the aerosol collected during that time had K2O, obtained by Yu and Park (2004). However, the average Na2O, CaO, and Al2O3 abundances near long-term re- values of pH are higher than the value reported in 2000 ported values (Chun et al., 2000; Fujita et al., 2000). (4.7 ± 1.4 by Lee et al. (2000)). The reason for this is Data collected on March 4th had the highest concen- perhaps due to the widespread deposition of Yellow sand trations of cations. Abundances of Ca2+ and Mg2+ were (Asian dust) from central Asia in early spring, 2001. 269.0 µeq/L and 107.3 µeq/L, respectively. Inspection of The winter rain collected on December 24, 2000 was this figure indicates that the present values are higher rela- a bit more acidic (pH 4.9) compared to samples from the tive to those reported in northeastern North America and spring season which had pH values of about 7.0 (Table central Europe (Galloway, 1995). The arithmetic means

Stable isotope compositions of precipitation in Seoul 447 34

δ pollution



observed

7.9 4.3 2.8

−

-excess

79.8 21.1 7.3 6.1

31.5 31.4 60.8 19.6 6.5 6.2 65.4 16.1 4.4 1.0

59.2 6.9 3.6 2.8 79.7 17.3

33.1 16.3 4.4 2.7

63.9 12.9 4.9 3.2 18.0 7.5 3.0 1.4 70.2 71.6 13.8 2.6 1.5 60.4 7.0 75.7 6.4 3.3 1.9

74.3 9.2 4.8 3.7 62.3 14.5 4.7 1.6

71.0 11.0 3.9 2.2 68.1 7.9 4.1 0.9

36.2 4.5 3.3 2.5 62.0 9.2 3.4 2.0 57.8 6.4 4.2 3.7 92.6 13.8 2.8 0.9 62.8 10.2 3.7 1.8

71.4 10.3 3.5 1.4

44.6 8.0 4.8 2.3 28.6 2.6 4.7 4.1

33.0 8.6 3.2 1.9

40.3 22.9 5.3 4.0 78.3 22.5 6.1 4.7

114.7 10.6 5.1 4.4 101.8 15.8 6.7 4.2

100.1 20.0 6.5 5.0

107.6 5.6 3.7 3.2

105.2 21.2 5.8 2.6

−

− − −

− −

− − −

− −

− − −

− −

−

− −

−

− δ −

7.8

8.2

6.2 7.1

8.6

3.2

7.4

9.6

9.5

8.9

9.1

5.1 8.1

6.5 3.9

5.2

7.9

10.1 15.7 14.7 10.1 12.6 15.0

12.2

10.6 10.3

10.4

10.2

14.2

13.3

10.2

12.6 15.8

−

− − −

−

− − −

−

− − − −

− −

−

− −

−

− −

−

= 0.73

= 0.72

nssSO

−

At, autumn.

eq/L) Isotopic data

−

73.7 172 146 65.2 204 188

5.72 49.6 317 274 90.3 81.8 68.1

36.6 47.9 287 176 138 118 106

om meteoric water in Seoul

5.86 16.5 20.2 51.2 14.9 25.2 29.8 29.2

73.2 39.4 64.1 482 88.7 205 263 247

10.3 41.2 11.9 164 57.8 103 45.6 41.4

7.09 2.61 9.04 25.6 32.5 12.0 15.0 11.0 10.1

d deviation; Wt, winter; Sp, spring; Sm, summer;

16.7 4.92 18.2 29.5 24.5 3.88 27.6 11.3 9.30

98.3 68.9 10.0 35.8 107 86.2 34.7 142 130

ater 130 ater 133 86.1 15.6 ater 135 113 25.6 191 314 122 105 243 226

ater 114 46.2

Snow (Wt) 20.8 4.24 6.18 42.1 41.8 14.4 22.9 13.2 10.7

ound water 312 373 37.8 63.9 228 78.8 455 229 191

ain (Sp) 20.9 3.27 23.0 64.6 20.5 12.1 17.1 28.8 26.3

ain (Sm) 21.7 14.1 2.49 24.9 64.5 27.3 28.5 36.5 33.9

STD 32.1 27.6 13.5 12.9 81.5 39.1 36.4 61.6 7.7 29.1 8.9 1.3 0.9

Stream w AM 128 79.6 21.6 94.8 321 158 117 199

STD 9.63 27.8 14.4 56.6 127 80.9 61.5 82.6 0.4 6.0 3.9 1.1 1.0 AM 238 234 25.9 64.2 384 115 299 271

STD 80.0 98.2 11.7 16.2 151 106 110 126 2.1 14.4 2.5 0.5 0.6

AM 33.4 22.6 15.3 28.6 96.8 37.3 42.4 63.3

Datetype Precipitation Chemical data (unit:

(dd-mm-yy) (season) Na

16-Dec-00 Snow (Wt) 18.2 13.4 8.96 34.2 91.9 39.7 21.9 50.6 48.4 24-Dec-00 Snow (Wt) 7.21 4.62 7.21 15.8 21.5 13.1 11.8 5.41 4.54 1-Jan-01 Snow (Wt) 8.78 1.67 6.33 21.4 13.6 4.28 24.3 12.3 11.2 27-Jan-01 (Wt) Snow 98.3 45.3 9-Feb-01 Snow (Wt) 15-Feb-01 Snow (Wt) 4.98 27-Feb-01 Rain/Snow (Wt) 4-Mar-01 Snow (Wt) 94.6 107 5.21 24.1 269 39.4 114 232 221

28-Mar-01 Rain/ 11-Apr-01 (Sp) Rain 92.2 63.2 8.92 15.5 122 85.5 77.5 112 101

7-May-01 R

22-May-01 Rain (Sp) 14.3 1.73 1.93 38.9 41.5 19.7 11.4 19.8 18.1 13-Feb-01 Stream w 18-Jun-01 Rain (Sm) 12.2 19.1 14.7 7.84 141 22.6 48.1 81.7 80.2 19-May-01 Stream w 1-Jul-01 Stream w

25-Jun-01 Rain (Sm) 7.13 9.22 9.52 20.7 56.5 9.53 23.2 33.1 32.2 5-Jul-01 Rain (Sm) 37.3 38.7 24.8 17.9 157 60.0 73.0 104 99.5 8-Oct-01

13-Feb-01 Ground water 295 206 15.1 75.3 512 58.2 254 454 418 10-Jul-01 (Sm) Rain 26.8 26.9 13.1 23.4 221 48.5 25.9 10919-May-01 Ground water 106 199 226 16.8 38.1 303 49.6 287 237 213 15-Jul-01 (Sm) Rain 14.8 4.18 13.8 40.5 27.1 18.9 12.4 12.71-Jul-01 10.9 Ground water 144 129 34.0 79.8 431 272 199 166 149 8-Oct-01 Gr

7-Aug-01 Rain (Sm) 49.8 38.3 55.2 33.5 132 32.2 60.2 166 160 15-Aug-01 R

6-Sep-01 Rain (At)

30-Sep-01 (At) Rain 34.8 10-Oct-01 Rain (At) 27.8 8.33 3.82 33.4 26.5 23.2 17.7 18.1 14.8

Table 2. Chemical and isotopic compositions fr Table AM, arthimetic mean; STD, standar

448 C. Lim et al. 1000 2– Wet depoisiton in Seoul 0 1 Sampling of Seoul Winter precipitation + Air pollutant 0.2 0.8 Spring Precipitation Organic waste – Summer precipitation Y=1.05X W i 2 n Fall precipitation 100 R =0.63 te r 0.4 0.6 L)

( eq/ Yellow Sand

– 0.6 0.4 Cl

10 Chemical weathering 0.8 0.2

Cation exchange S u m m e Heavy snowfall r 1 0 – 2– 1 0 0.2 0.4 0.6 0.8 1 1 10 100 1000 2– – – + Fig. 4. SO4 –NO3 –Cl ternary diagrams for precipitation and surface and ground water in Seoul area (unit: µeq/L). Fig. 3. The relationship between Cl– vs. Na+ of rain and snow water in Seoul. The solid line indicates the Cl–/Na+ = 1.16 (Millero, 1996). respectively. The isotopic composition of non-sea salt sulfate δ34 ( Snss) was calculated from the isotopic composition of δ34 + and standard deviation of major ion concentrations are the sample’s total sulfate ( Smeasured). The (Na )measured 2– shown in Table 2. and (SO4 )measured stand for the measured concentration + – + 2– 2– + Na and Cl can be considered to be components of of Na and SO4 of samples and (SO4 /Na )nss indicate 2– + sea salt. Non-sea salt ion (nss) can be expressed as the the seawater SO4 /Na ratio of 0.120 (Weast et al., 1987). ratio of dissolved elements of rainwater, and its reference The calculated sulfur isotope composition of the sulfate is either dissolved Na+ or Cl–. These references can be which is derived from non-sea salt (pollution) sources is + 2– chosen depending on the observation site. Generally, Na shown in Table 2. The concentration of SO4 is the most is better choice in polluted sites such as urban areas, be- abundant with an average of 63.3 µeq/L and correspond- – 2– 2– µ cause Cl can be produced from polluted HCl in indus- ing non-sea salt SO4 (nssSO4 ) of 56.7 eq/L, which is + 2– trial areas (Keene et al., 1986). In this study, Na was over 90% of the total SO4 (Table 2). The values of non- 2– also used as a reference for nss ion value (Table 2). The sea salt SO4 are very similar to those from precipitation nss value of rainwater can be used to compare rainfall measurements in Kangwha, Korea (Fujita et al., 2000). composition with sea salt composition. Using known so- The other most abundant anion is Cl– with a mean µ – dium and chlorine weight ratio in sea water and assum- concentration of 37.3 eq/L, and NO3 amounts to 42.4 ing no fractionation during aerosol formation from sea µeq/L. Comparison of rain waters collected at different 2– water, the contribution from sea salt can be determined times of the year shows that concentrations of SO4 were (Berner and Berner, 1996). The concentration of an ion dominated by much higher values in winter than those of – 2– which is larger than the sea water proportion is referred NO3 in summer. Average concentrations of SO4 and – µ µ to as the non-sea salt ion or excess ion (Millero, 1996). NO3 in summer are 53.8 eq/L and 44.5 eq/L, respec- – 2– – The relation between the concentrations of Na and Cl in tively. Average concentrations of SO4 and NO3 in win- Seoul is shown in Fig. 3. The regression line between ter are 77.6 µeq/L and 38.8 µeq/L, respectively. The data + – 2– Na and Cl are indicated by the solid line. The regres- shows that SO4 concentration in winter is about 1.7 times sion broken line shows the ratio of the concentrations of higher than that of summer (77.6 µeq/L vs. 53.8 µeq/L). + – – Na and Cl in sea water (Fig. 3), respectively (Nozaki, On the other hand, the concentrations of NO3 in the win- 1992). The plot shows a linear relation to the ratio in sea ter are much lower than those in the summer (38.8 µeq/L + – µ – salt in Seoul, indicating that most of Na and Cl in wet vs. 44.5 eq/L) (Table 2). The average NO3 concentra- 2– depositions originated from the sea salt. However, the plot tion was found to be higher than SO4 in summer. It is of 1st January and 4th March in 2001 shift to sodium el- known that there are farming activities such as swine pro- ements occurred with heavy snowfall and yellow sand, duction and dairy production that result in animal waste

Stable isotope compositions of precipitation in Seoul 449 0 –20 ) –40 –60 –80

D(– ,SMOW 100 0 –120 –4

–8

–12

40 –16 O( ,SMOW) 18 30 –20 e 20

10 d exces 0

–10

0 0 1 1 1 1 1 0 0 /01 0 /01 0 /01 0 1 1 /01 0 c/ c/ b b/ b r/ / /0 /0 g/01 g/01 p p/01 t/ e e e p u u e e D D Jan/01Jan/01Fe F Fe Mar/01Mar/01A May May/01Jun Jun/0Jul Jul/01Jul A A S S Oc (mm/yy)

Fig. 5. Seasonal variations of relationship between d-excess and Oxygen and Hydrogen isotopic values in the precipitation from Seoul. The value of d-excess is determined from the equation as d = δD – 8δ18O (Dansgaard, 1964).

2– including NH3. On the other hand, the average SO4 con- indicates that sulfuric acid is the main cause of acidifica- – centration was found to be higher than NO3 in winter tion of precipitation in winter months and nitric acid is due to evolve from the long rang transport contributing the main cause of acidification in summer months (Fig. from the local industrial combustion sources (Fig. 4) 4). (Ohizumi et al., 1991). However, stream and ground waters did not show the seasonal variation in cations and Oxygen and hydrogen isotopes anions (Table 2). The O and H isotopic composition of precipitation in Atmospheric dispersion in winter tends to be stronger this study is quite variable, with δ18O values ranging from compared to that in summer, and consequently pollutants –15.8 to –3.2‰ and δD from –114.7 to –18‰ (Fig. 5). emitted in the winter are easily dispersed and transported Seasonal variation is reflected in this data. The values of over large areas. Additionally, as more intense solar ra- deuterium excess (d-values) for winter precipitation (d diation, higher temperatures and greater concentrations >10‰) are clearly distinct from those for summer pre- of water vapor in summer favor faster photochemical cipitation (d < 10‰) (Fig. 5). The deuterium excess (d- conversions of SO2 and NO2 to sulfate and nitrate. These values) for winter precipitation significantly deviates from alkaline ions (Ca2+ and Mg2+) act as buffers, neutralizing d-excess value obtained from the global meteoric water the acidity of precipitation. Furthermore there are greater line, more or less reflecting the effects of various degrees emissions of air pollutants in cold seasons relative to warm of evaporation (Fig. 5). The slope and intercept of the seasons (So et al., 1996). regression line for precipitation were virtually identical The rain and snow waters in summer precipitation to the global meteoric water line (GMWL) defined by 2– contain greater concentrations of SO4 and these phases Craig (1961). Here, the mean meteoric water line of Seoul carry higher levels of neutralizing cations (Table 2). The (δD = 7.3δ18O + 5.0), Chuncheon meteoric water line (δD concentration ratio of alkaline ions to acidic ions in sum- = 7.2δ18O + 2.7; Yu and Park, 2004), and Chunju mete- mer is greater than that in winter (Table 2). This result oric water line (δD = 6.8δ18O + 7.7; Na et al., 1995) were

450 C. Lim et al. –20 –15 –10 –5 0 0 Global meteoric water

Seoul meteoric water –30 Chuncheon meteoric water

Chunju meteoric water Ave. warm season (Summer) –60

–90 Ave. cool season Seoul winter (Winter) Seoul spring Seoul summer Seoul fall –120 Chuncheon during 2000–2001 (Yu and Park, 2004) Chunju during 1994–1995 (Na et al., 1995) –150

Fig. 6. Plot of δD versus δ18O in precipitation, surface and ground water. Global meteoric water line (GMWL) of Craig (1961) is also shown.

comparable to the Global meteoric water line (δD = 8δ18O December, 2000 and March, 2001) is +17.6 and for sum- + 10). These data indicate the nature of the coupled oxy- mer precipitation (from June to August, 2001) is +6.1. gen and hydrogen isotopic variation for different seasons, The δ18O and δD values of the meteoric waters in which follow the relationship: δD = 9.3δ18O + 8.3 for Chuncheon align fairly well along the global meteoric summer (rainy season precipitation) and δD = 8.8δ18O + water line (GMWL: Craig, 1961) (Fig. 6). The δ18O and 28.0 for winter (dry season precipitation) in Seoul (Fig. δD values of the surface waters have a similar range those 6). Seasonal isotopic differences could also result from of the meteoric waters in Seoul. The referenced Chonju the effect of different air masses; cold and dry continen- values are very close to the long-term weighted average tal Siberian air masses predominate in winter while hot (δ18O = –3.0~–13.4‰ and δD = –13~–92‰) for rainfall and humid maritime North Pacific air masses predomi- in the global meteoric water (Na et al., 1995). The iso- nate in summer (Lee et al., 1999). topic compositional range of the Chonju meteoric water Oxygen and hydrogen isotope measurements reveal may be due to an insignificant amount of samples col- similarities to the global meteoric water line. However, lected for the seasonal variation, the attenuation of the despite the limited number of samples examined and the temporal variation. relatively short sampling period, it is noteworthy that our data are very similar to those reported by Lee and Chang Sulfur isotope chemistry δ34 (1994). The deviation of the isotopic composition of this The sulfur isotopic composition of sulfate ( SSO4) group of samples above the summer season GMWL can ranges between +3.0 and +7.3‰ in snow and rain water be determined from the d-values listed in Table 2. The samples in this study (Table 2 and Fig. 7). Major sulfur annual mean d-value of these precipitation samples is ion distributions in precipitation are mostly controlled by +12.9, as shown in Table 2, slightly displaced below the diverse contributions from seawater sulfate, biogenic +10.0. The average d-value for winter precipitation (for emissions, volcanic gases, and anthropogenic S (Thode

Stable isotope compositions of precipitation in Seoul 451 25 Locations of precipitation Seoul winter 20 Seoul spring (Thode et al., 1961; Ohizumi et al., 1991) Seoul summer Seoul fall Chuncheon during 2000–2001 15 (Yu and Park, 2004) (Maruyama et al., 2000) Chunju during 1994–1995 (Na et al., 1995)

10 (Maruyama et al., 2000) Ave. winter (This research) 5 Ave. summer (This research) (Hong et al., 1992; Maruyama et al., 2000) 0

(Yu and Park, 2004) –5

(Thode et al., 1961; Ohizumi et al., 1991) –10

1 10 100 1000 Sulfate concentrations

δ34 2– Fig. 7. Plot of S against the concentration of the SO4 in wet precipitation. et al., 1961; Ohizumi et al., 1991; Mandeville et al., 2009). This relationship assumes that all sodium comes from The isotopic composition of S in modern marine sulfate the sea and the sulfate/sodium ratio in the marine compo- is constant within narrow limits and is represented by a nent of precipitation is the same as that of seawater, i.e., δ34 2– + S value of about +20.1‰. A number of sulfur isotopic [SO4 /Na ]seawater = 0.120 by weight (SO4 seawater: sulfate + studies have been carried out on sulfate ions in rainfall ions from the seawater, Na seawater: sodium ions from the (Mizutani and Rafter, 1969; Holt et al., 1972; Jamieson seawater, SO4 observed: sulfate ions of chemical analysis). and Wadleigh, 2000). These studies show that rainwater The equation using A and sodium content to establish sulfate is depleted in 34S with respect to seawater sulfate, seawater spray sulfate concentrations is a general equa- δ34 the effect being more pronounced in sulfate from other tion of sulfate isotopic values ( Snss) from the contami- sources. The isotopic composition from biogenic emis- nated atmosphere. sions (H2S and DMS from seawater) is difficult to estab- δ34 δ34 δ34 δ34 lish, but it may have S values ranging between –10 Snss = [ Sobserved – A Sseawater][1 – A] and 0‰ (Herut et al., 1995, Nielsen, 1978). The δ34S values of volcanic gas sulfur are known to range widely Table 2 shows the calculated sulfur isotopic composi- around 0‰ (Mizutani et al., 1986), but there is no active tion of the sulfate derived from the sources that experi- local volcano near the study area. ence the pollution (combustion of fossil fuels such as coal δ34 Chloride concentrations have been utilized to deline- or petroleum). The present research reveals that Snss ate the seawater spray sulfate component in precipitation. values ranging from +1.0 to +6.2‰ are found in Seoul. Mizutani and Rafter (1969) utilize the parameter: These observations support a major anthropogenic source of SO4 in precipitation in the study area. The major an- 2– 2– A = {[SO4 ]seawater/[SO4 ]observed } thropogenic S source in East Asia is known to be coal combustion. China is the largest source of SO2 emissions + 2– + 2– = {[Na ]observed/[SO4 /Na ]seawater }/[SO4 ]observed. in East Asia and its total emissions are estimated to be

452 C. Lim et al. SO2

0.03 0.1 0.3 1 3 10 g/m2/y

Fig. 8. Distribution of annual emissions of SO2 in the East Asian to EAGrid (2000).

more than 10 times that of Japan and Korea (Fig. 8; and Coleman, 2000), which is much lower than that of EAGrid, 2000). The proportion of coal combustion as part coal in neighboring countries. The δ34S values of sulfate of the total energy demands in China is as high as 70%; in precipitation in the southern part of Korea were re- therefore, coal combustion is the major source of anthro- ported in the range of 0.0 to 1.8‰ (Na et al., 1995), which δ34 pogenic SO2 production. The S values of S in coal were is lower than the sulfur isotope values in this study. reported as 9.7 ± 11.4‰ in northern China and 5.5 ± 4.8‰ The δ34S values of atmospheric samples are higher in in Russia (Maruyama et al., 2000) and the δ34S values of winter (from +5.1 to +7.3‰) and lower in summer (from S in aerosol sulfate and sulfur dioxide were reported as +3.0 to +4.8‰) (Fig. 7). In neighboring Japan, Maruyama 4.2 ± 0.7‰, 6.4 ± 0.5‰, and 6.1 ± 1.2‰ in winter at et al. (2000) reported on seasonal variations of sulfur iso- Harbin, Changchun, Dalian, and Waliquan in northern tope ratios of rain and aerosol on the west coast, in which China (Maruyama et al., 2000), respectively. These δ34S sulfur during winter is mainly derived from space heat- values are higher than that of coals from southern China ing and industrial sources; but in summer the large emis- (4.5 ± 6.2‰) (Hong et al., 1992; Maruyama et al., 2000; sions of 34S-depleted biogenic sulfur come from soil, veg- Yanagisawa et al., 2001) and the δ34S values of S in aero- etation, marshes, and wetlands, which results in the de- sol sulfate and sulfur dioxide were reported as 2.1 ± 1.4‰, crease of δ34S values of airborne sulfur. ± δ34 2– and –4.0 3.6‰ in summer at Nanjing and Guiyang in Plots of S values versus SO4 concentrations can southern China (Maruyama et al., 2000), respectively. Al- be used in identifying sources and fates of sulfur-bearing most all of the coal and petroleum used in Korea are im- compounds. The dominant δ34S values appear to have a ported from foreign countries. The δ34S values of organic narrow range near +6.4‰ in winter and +4.0‰ in sum- S in coal from Gangwon province, Korea is known to mer showing distinct correlation between the seasonal contain less than 1 percent total sulfur with a value of variations and the sources of sulfur in the neighboring –2.4‰, which falls within the general range countries (Fig. 7). However, the δ34S values had no cor- 2– (–10.0~+1.6‰) of pyrite from coal mines in Korea (Yu relation with SO4 concentrations (Fig. 7).

Stable isotope compositions of precipitation in Seoul 453 variations are more pronounced for the origin of sulfur sources than for the concentrations of sulfur component. On average, the sulfur isotopic values for Seoul in winter and summer are +6.4 and +4.0‰, respectively. The period of higher sulfur isotopic values in winter (December and February) seems to be related to the wind direction (northeast-northwest) prevailing in northern China, whose δ34S values of oil (+22.5‰) and coal (+9.7‰) is known to be higher than oil (+7.2‰) and coal (+4.5‰) of southern China (Hong et al., 1992; Na et al., 1995; Maruyama et al., 2000). The lower sulfur isotopic values in summer (June to August) may be related to the summer wind direction (west-southwest) blowing from southern China, whose δ34S value is lower than that of northern China. δ34 Sulfur isotopic composition of sulfate ( SSO4) shows values ranging from +2.6 to +4.8‰ in the streams of the study area. Sulfate isotopic values from the southern part of the Han River reported during spring and summer re- δ34 ± veal SSO4 values of +3.5 1‰ (Ryu et al., 2007). The δ34 2– sulfur in streams and shallow groundwater may be de- Fig. 9. Relationship between the S(‰) signature in SO4 of precipitation and the prevailing wind direction in East Asia. rived from the oxidation of sulfide minerals as well as deposition of atmospheric sulfates through precipitation. The δ34S values of sulfide minerals collected from hy- drothermal deposits in Korea range from +2.0 to +7.0‰ Origin and migration of sulfur sources in East Asia representing high homogeneity in sulfur isotopic compo- The sulfur isotopic signatures of rainwater sulfate may sition of the acidic crustal magma (Kim and Nakai, 1980). 2– be used to delineate anthropogenic sulfur in the atmos- This suggests that dissolved SO4 in the streams and shal- phere, provided the isotopic composition of pollutant low groundwater in the study area might be coming from sulfur is distinct from that of sulfur from natural sources the oxidation of sulfide minerals in the granitic basement (Krouse, 1980; Na et al., 1995). Identification of atmos- rocks distributed upstream as well as from depositions of pheric sulfur sources in rainwater has been conducted atmospheric sulfates and rainfall. mainly in the acid rain regions through sulfur isotope In order to develop a consistent and effective method, analysis (Krouse and Case, 1981; So et al., 1996). Usu- it is necessary to carry out a long term study accumulat- ally the sulfur isotopic values of rainwater sulfate are al- ing data on the isotope compositions of contaminants and tered by oxidation of SO2 and different contribution of other environmental materials, not only from Korea but sulfur sources (e.g., coal burning, biogenic sulfur, etc.). also from neighboring countries: China, Japan, Russia, The minor sources are sea spray and biological activity etc. as sea spray has a δ34S value of about + 20.1‰ and the biogenic component shows a δ34S value ranging from SUMMARY –10‰ to –2‰ (Mizutani and Rafter, 1969; Holt et al., 1972; Ohizumi et al., 1991; Jamieson and Wadleigh, In the Seoul metropolitan area and Chuncheon min- 2000). Because more samples having low sulfate con- ing area, we carried out a study on sulfur, oxygen, and centrations and low δ34S values were found at Chonju hydrogen isotope compositions to discover the source of (Na et al., 1995) rather than at Seoul and Chuncheon ar- sulfur and the extent of seasonal and local chemical char- eas (Fig. 7), the significantly more negative rainwater δ34S acteristics in snow and rain. Our findings are: values at the rural site in Chonju (δ34S = –3.2‰) com- 1) The pH of the rainwater collected in the Seoul area pared to the urban site Seoul and Chuncheon areas may is in the range of 4.6–7.0. The variation pattern has a good be also due to higher contribution of biogenic sulfur. The correlation among the amount of rainfall, seasonal varia- δ34S values for rainwater and sea spray are sufficiently tion, pH, EC, cations, and anions of rainwater. The re- distinct to be able to conclude that seawater contributions sulting data are similar to those of adjacent countries in- δ34 are negligible due to the calculation of Snss values. cluding China, Japan, and Russia. The relationship between S isotope values and wind 2) Precipitation shows positive correlations between 2+ 2+ 2– – direction reflects the marked seasonal variations of sulfur ions in the Ca –Mg –SO4 –NO3 system, indicating isotopic values during winter and summer (Fig. 9). The that dissolution of Ca and Mg particles by H2SO4 and

454 C. Lim et al. HNO3 from the combustion of fossil fuels is a major proc- Center for Inter-University Research facilities, Seoul National ess controlling the chemical composition of snow and rain. University for the chemical composition data. Review of this Based on weighted means in precipitation, enrichment manuscript by Michael J. Lee for English edits is greatly ap- factors relative to seawater are 69.9 for Ca2+, 13.1 for preciated. 2+ 2– – + Mg , 2.2 for SO4 , and 1.8 for NO3 assuming that Na comes only from seawater. REFERENCES 3) During the survey period, airborne particles (Yel- low sand) appeared to be the dominant source of higher Allan, M. A. (ed.) (2004) Manual for the GAW Precipitation Chemistry Programme. Guidelines. Data Quality Objectives concentrations of Ca2+ and Mg2+ (alkaline ion) in pre- 2– – and Standard Operating Procedures. WMO TD No. 1251: cipitation. The concentrations of nss SO4 and NO3 are 170 pp. Available at ftp://ftp.wmo.int/Documents/ the same as those in typical acid rain regions. However, PublicWeb/arep/gaw/gaw160.pdf 2+ most of them are neutralized forms of base cations, Ca Andronache, C. (2004) Estimates of sulfate aerosol wet scav- and Mg2+. As a result the average pH of precipitation is enging coefficient for locations in the Eastern United States. around 6.2. Atmos. Environ. 38, 795–804. 4) The value of oxygen and hydrogen isotope compo- Ayers, G. P. and Yeung, K. K. (1996) Acid deposition in Hong sition of the rain is in the range of –15.7~–7.8‰ and Kong. Atmos. Environ. 30, 1581–1587. –115~–31‰, respectively. The H and O isotope values in Berner, E. K. and Berner, R. A. (1996) Global Environment. summer are plotted near the global meteoric water line Water, Air and Geochemical Cycles. Prentice-Hall, Upper by Craig (1961). The d-excess values in winter are higher Saddle River, NJ. Cape, J. N. and Leith, I. D. (2002) The contribution of dry de- than 10, which indicate that the source of rain during win- posited ammonia and sulphur dioxide to the composition ter is a dry air mass from continental China. of precipitation from continuously open gauges. Atmos. 5) The sulfur isotope ratio of precipitation ranges from Environ. 36, 5983–5992. +3.0 to +7.5‰ similar to typical metropolitan air pollu- Chun, J., Kim, J., Choi, J. C., Boo, K. O., Oh, S. N. and Lee, tion. It is suggested that the main source is SO2 produced M. (2001) Characteristic number size distribution of aero- by the combustion of fossil fuels. Pollutant SO4 is esti- sol during Asian dust period in Korea. Atmos. Environ. 35, δ34 mated to have Snss ranging from +1.0 to +6.2‰ in the 2715–2721. Seoul area. Chun, Y., Kim, J., Boo, K. O. and Kim, N. O. (2000) The fea- 6) The sulfur isotope composition of precipitation tures associated with the Yellow Sand phenomenon observed shows an overlapping range with those of the adjoining in Korea in wintertime. J. Korean Soc. Atmos. Environ. 16, stream and shallow groundwater. In the study area the 487–497 (in Korean with English abstract). Coleman, M. L., Shepherd, T. J., Durham, J. J., Rouse, J. E. δ34S values of SO in streams and shallow groundwater 4 and Moore, G. R. (1982) Reduction of water with zinc for are not only directly affected by the deposition of atmos- hydrogen isotope analysis. Anal. Chem. 54, 993–995. pheric sulfates through the addition of meteoric water to Craig, H. (1961) Isotopic variations in meteoric waters. Sci- the system but also are affected by the oxidation of sulfide ence 133, 1702–1703. minerals in the granitic basement rocks. Dansgaard, W. (1964) Stable isotopes in precipitation. Tellus 7) The δ34S values of precipitation range from +5.1 to 16, 436–468. +7.3‰ (mean +6.4‰) in winter and from +3.0 to 4.8‰ Das, R., Das, S. N. and Misra, V. N. (2005) Chemical composi- (mean +4.0‰) in summer. The higher δ34S values in win- tion of rainwater and dustfall at Bhubaneswar in the east ter (December and February) seem to be correlated to the coast of India. Atmos. Environ. 39, 5908–5916. air mass (northwest in main wind direction) from north- EANET (2000) Technical manual for wet deposition monitor- ern China, whose δ34S value of oil or coal is higher than ing in East Asia, 74 pp. Available at http://www.eanet.cc/ product/manual/prev/techwet.pdf that of southern China. The lower sulfur isotopic values East Asian Air Pollutant Emissions Grid Database (EAGrid) in summer (June to August) are correlated with the air (2000) Available at http://www.cger.nies.go.jp mass (west-southwest is the main wind direction) mov- Epstein, S. and Mayeda, T. (1953) Variation of 18O content of 34 ing from southern China, whose δ S value of fossil fuel waters from natural sources. Geochim. Cosmochim. Acta 4, is lower than that of northern China. 213–224. Fujita, S. I., Takahashi, A., Weng, J. H., Huang, L. F., Kim, H. Acknowledgments—This study was partly supported by grant K., Li, C. K., Huang, F. T. C. and Jeng, F. T. (2000) Precipi- No. 2000-2-13200-001-1 from the Basic Research Program of tation chemistry in East Asia. Atmos. Environ. 34, 525–537. the Korean Science and Engineering Foundation. This study Galloway, J. N. (1995) Acid deposition: Perspectives in time was also supported by the BK21 program of the School of Earth and space. Water, Air, Soil Pollut. 85, 15–23. and Environmental Sciences, Seoul National University. The Herut, B., Spiro, B., Starinsky, A. and Katz, A. (1995) Sources authors would like to thank Dr. K.-S. Lee and his staff in the of sulfur in rainwater as indicated by isotopic δ34S data and Korean Basic Science Institute for the S, H, O isotopic analy- chemical composition, Israel. Atmos. Environ. 29, 851–857. sis of water samples and Ms. Seonyoung Kim in the National Holt, B. D., Engelkemeir, A. G. and Venters, A. (1972) Varia-

Stable isotope compositions of precipitation in Seoul 455 tions of sulfur isotope ratios in samples of water and air Japan 1, 45–50 (in Japanese with English abstract). near Chicago. Environ. Sci. Technol. 6, 338–341. Millero, F. (1996) Chemical Oceanography (2nd ed.). CRC Hong, Y., Zhu, Y., Zhang, H., Piao, H., Jiang, H., Zeng, Y. and Press, Boca Raton, FL Comment, 66 pp. Liu, G. (1992) Compositional characteristics of coal in Mizutani, Y. and Rafter, T. A. (1969) Isotopic composition of China and fractionation of sulfur isotopes during coal-burn- sulphate in rain water, Gracefield, New Zealand. N. Z. J. ing process. Developments in Geochemistry, 241–250, In- Sci. 12, 69–80. stitute of Geochemistry, Chinese Academy of Science, Seis- Mizutani, Y., Hayashi, S. and Sugiura, T. (1986) Chemical and mological Press. isotopic compositions of fumarolic gases from Kuju- Jamieson, R. E. and Wadleigh, M. A. (2000) Trace sources of Iwoyama, Kyushu, Japan. Geochem. J. 20, 273–285. precipitation sulfate in eastern Canada using stable isotopes Na, C. K., Kim, S. Y., Jeon, S. R., Lee, M. S. and Chung, J. I. and trace metals. J. Geophys. Res. 105, 20549–20556. (1995) Sulfur isotopic ratios in precipitation around Chonju- Junge, C. E. (1963) Sulfur in the atmosphere. J. Geophys. Res. city, Korea and its availability as a tracer of the source of 68, 3975–3976. atmospheric pollutants. Econ. Environ. Geol. 28, 243–249 Keene, W. C., Pszenny, A. A. P., Galloway, J. N. and Hawley, (in Korean with English abstract). M. E. (1986) Sea-salt corrections and interpretation of con- Nakai, N. and Takeuchi, U. (1975) Chemistry of rain and air stituent ratios in marine precipitation. J. Geophys. Res. 91, pollution. Kagaku (Kyoto) 29, 418–426 (in Japanese with 6647–6658. English abstract). Kim, K. H. and Nakai, N. (1980) Sulfur isotope composition Newman, L., Krouse, H. R. and Grinenko, V. A. (1991) Sul- and isotopic temperatures of some base metal ore deposits, phur isotope variations in the atmosphere. Stable Isotopes: South Korea. J. Geol. Soc. Korea 16, 124–134 (in Korean Natural and Anthropogenic Sulphur in the Environment with English abstract). (Krouse, H. R. and Grinenko, V. A., eds.), Scope 43, 352– Krouse, H. R. (1980) Sulphur isotopes in our environment. 358, John Wiley & Sons, New York. Handbook of Environmental Isotope Geochemistry Vol. 1: Nielsen, H. (1978) Sulfur isotopes in nature. Handbook of The Terrestrial Environment, A (Fritz, P. and Fontes, J. Ch., Geochemistry (Wedepohl, K. H., ed.), 16B, 1–48, Springer- eds.), 435–471, Elsevier, Oxford. Verlag, Berlin. Krouse, H. R. and Case, J. W. (1981) Sulphur isotope ratios in Nissenbaum, A. (1978) Sulfur isotope distribution in sulfates water, air, soil, and vegitation near Teepee Creek gas plant from surface water from the Northern Jordan Valley, Israel. Alberta. Water, Air, Soil Pollut. 15, 11–28. Environ. Sci. Technol. 12, 962–965. Kulshrestha, U. C., Kulshrestha, M. J., Sekar, R., Sastry, G. S. Nozaki, Y. (1992) Trace elements in sea water. Geochemistry R. and Vairamani, M. (2003) Chemical characteristics of 26, 25–39 (in Japanese with English abstract). rainwater at an urban site of south-central India. Atmos. Nriagu, J. O., Holdway, D. A. and Coker, R. D. (1987) Bio- Environ. 37, 3019–3026. genic sulfur and acidity of rainfall in remote areas of Lee, B. K., Hong, S. H. and Lee, D. S. (2000) Chemical com- Canada. Science 237, 1189–1192. position of precipitation and wet deposition of major ions Ohizumi, T., Fukuzaki, N., Moriyama, N., Urushima, Y. and on the Korean peninsula. Atmos. Environ. 34, 563–575. Kusakabe, M. (1991) Source of Sulfur in the atmospheric Lee, K. S. and Chang, B. U. (1994) Oxygen and hydrogen iso- deposits in view of sulfur isotopic variations—A case study topic composition of precipitation in Taejeon and Seoul, in Niigata prefecture, Japan, Nippon Kagaku Kaishi (1991). Korea. J. Geol. Soc. Korea 30, 475–481 (in Korean with Chem. Soc. Japan 5, 675–681 (in Japanese with English English abstract). abstract).

Lee, K. S. and Lee, C. B. (1999) Oxygen and hydrogen iso- Park, J. and Cho, S. Y. (1998) A long range transport of SO2 topic composition of precipitation and river waters in South and sulfate between Korea and East China. Atmos. Environ. Korea. J. Geol. Soc. Korea 35, 73–84 (in Korean with Eng- 32, 2745–2756. lish abstract). Park, S. S., Kim, Y. J., Cho, S. H. and Kim, M. O. (2000) Chemi- Lee, K. S., Lee, I., Choi, M. S. and Barg, E. J. (1997) An envi- cal composition of fine particles in three urban areas, Ko- ronmental isotopic study of precipitation in Cheju island, rea. J. Aerosol Sci. 31, 118–119. Korea. J. Geol. Soc. Korea 33, 139–147 (in Korean with Ryu, J. S., Lee, K. S. and Chang, H. W. (2007) Hydrochemistry English abstract). and isotope geochemistry of Song Stream, a headwater tribu- Lee, K. S., Wenner, D. B. and Lee, I. (1999) Using H- and O- tary of the South Han River, South Korea. Geosci. J. 11, isotopic data for estimating the relative contributions of 157–164. rainy and dry season precipitation to groundwater: exam- So, S. S., Lee, D. S. and Shin, H. R. (1996) Characteristics of

ple from Cheju Island, Korea. J. Hydrol. 222, 65–74. variation of SO2 and NO2 Concentration in the Air over the Mandeville, C. W., Webster, J. D., Tappen, C., Tayler, B. E., major cities of Korea. J. Korean Earth Sci. Soc. 17, 174– Timbal, A., Sasaki, A., Hauri, E. and Bacon, C. R. (2009) 182 (in Korean with English abstract). Stable isotope and petrologic evidence for open-system Thode, H. G., Monster, J. and Dunford, H. B. (1961) Sulfur degassing during the climactic and pre-climactic eruptions isotope geochemistry. Geochim. Cosmochim. Acta 25, 159– of Mt. Mazama, Crater Lake, Oregon. Geochim. 174. Cosmochim. Acta 73, 2978–3012. Weast, R. C., Astle, M. J. and Beyer, W. H. (1987) Handbook Maruyama, T., Ohizumi, T. and Taneoka, Y. (2000) Sulfur iso- of Chemistry and Physics. CRC, Boca Raton, FL, Section tope ratios of coals and oils in China and Japan. Chem. Soc. F-100.

456 C. Lim et al. Yamaguchi, K., Tatano, T., Tanaka, F., Nakao, M., Gomyoda, Ecotechnol. Res. 7, 1–6. M. and Hara, H. (1991) An analysis of precipitation chem- Yu, J. Y. and Coleman, M. (2000) Isotopic Compositions of istry measurements in Shimane, Japan. Atmos. Environ. 25A, Dissolved Sulfur in Acid Mine Drainages: Case Study on 285–291. Youngdong and Gangreung Coal Mines, Korea. J. Geol. Soc. Yanagisawa, F. and Sakai, H. (1983) Thermal decomposition Korea 36, 1–10. of barium sulfate-vanadium pentaoxide-silica glass mixtures Yu, J. Y. and Park, Y. (2004) Sulphur isotopic and chemical for preparation of sulfur dioxide in sulfur isotope ratio meas- compositions of the natural waters in the Chuncheon area, urements. Anal. Chem. 55, 985–987. Korea. Appl. Geochem. 19, 843–853. Yanagisawa, F., Akata, N., Motoyama, R., Kawabata, A. and Zunckel, M., Saizar, C. and Zarauz, J. (2003) Rainwater com- Ueda, A. (2001) Seasonal variation of sulfur isotope ratios position in northeast Uruguay. Atmos. Environ. 37, 1601– of non-sea salt sulfate in wet deposition in Japan. J. 1611.

Stable isotope compositions of precipitation in Seoul 457