Environ. Sci. Technol. 2008, 42, 5421–5427
Identification of Anthropogenic and Natural Inputs of Sulfate and Chloride into the Karstic Ground Water of Guiyang, SW China: Combined δ37Cl and δ34S Approach C O N G - Q I A N G L I U , * ,† Y U N - C H A O L A N G , * ,† HIROSHI SATAKE,‡ JIAHONG WU,‡ AND SI-LIANG LI† State Key Laboratory of Environmental Geochemistry, Institute of Geochemistry, Chinese Academy of Sciences, Guiyang 550002, China, and Department of Environmental Biology and Chemistry, Toyama University, Toyama 930-8555, Japan
Received February 08, 2008. Revised manuscript received April 22, 2008. Accepted April 29, 2008.
Because of active exchange between surface and groundwater of a karstic hydrological system, the groundwater of Guiyang, the capital city of Guizhou Province, southwest China, has been seriously polluted by anthropogenic inputs of NO3-, SO42-, Cl-, and Na+. In this work, δ37Cl of chloride and δ34S variations of sulfate in the karstic surface/groundwater system were studied, with a main focus to identify contaminant sources, including their origins. The surface, ground, rain, and sewage water studied showed variable δ37Cl and δ34S values, in the range of -4.1 to +2.0‰, and -20.4 to +20.9‰ for δ37Cl and δ34S (SO42-), respectively. The rainwater samples yielded the lowest δ37Cl values among those observed to date for aerosols and rainwater. Chloride in the Guiyang area rain waters emanated from anthropogenic sources rather than being of marine origin, probably derived from HCl (g) emitted by coal combustion. By plotting 1/SO42- vs δ34S and 1/Cl- vs δ37Cl, respectively, we were able to identify some clusters of data, which were assigned as atmospheric deposition (acid rain component), discharge from municipal sewage, paleo-brine components in clastic sedimentary rocks, dissolution of gypsum mainly in dolomite, oxidation of sulfide minerals in coalcontaining clastic rocks, and possibly degradation of chlorinecontaining organic matter. We conclude that human activities give a significant input of sulfate and chloride ions, as well as other contaminants, into the studied groundwater system through enhanced atmospheric deposition and municipal sewage, and that multiple isotopic tracers constitute a powerful tool to ascertain geochemical characteristics and origin of complex contaminants in groundwater.
Introduction Karst aquifers are characterized by dissolution-generated conduits that permit the rapid transport of groundwater, and are susceptible to rapid introduction of contaminants * Address correspondence to either author. E-mail: liucongqiang@ vip.skleg.cn (C.-Q.L.);
[email protected] (Y.-C.L.); Fax: +86 851 5891609. † Chinese Academy of Sciences. ‡ Toyama University. 10.1021/es800380w CCC: $40.75
Published on Web 06/11/2008
2008 American Chemical Society
since the conduit system receives localized inputs from surface streams. Therefore, extensive research efforts have been conducted on water/rock interaction and contaminant transport processes in karst aquifers (1–7). Sulfate ion in groundwater can be derived from atmospheric deposition, dissolution of gypsum, oxidation of sulfide minerals in aquifer and sulfur-containing organic materials in soil, and also from inputs of industrial waste and municipal sewage. Since sulfur isotopic compositions (δ34S) of sulfate are unequally distributed among sources, sulfur isotope technique has been widely used to identify and quantify sources of sulfur and to trace transformation processes of different sulfur species (8–11). Chloride in groundwater has various types of sources, such as atmospheric deposition of anthropogenic and natural/ marine origin, oxidation of chlorine-containing organic compounds, weathering of paleo-seasalt and clay minerals, and discharge of industrial wastes. It is often difficult to pinpoint the actual sources of chlorine in groundwater. In the past decade, analytical techniques to determine chlorine isotope composition of various environmental and geological samples have been largely developed and found applicability in environmental source identification. Accordingly, the chlorine isotopes have been widely used in study of evaporate formation, evolution of groundwater, intrusion of seawater into coastal strata, and ore-forming fluid (12–14). Although studies on chlorine isotope characteristics of various contaminants in environmental samples are still few (15), there exist several recent studies that have added to the chlorine isotope database for discriminating among various sources of industrial wastes, road salt and other commonly used salt compounds, and common agro-chemicals (16–21). The surface-groundwater system of Guiyang, the capital city of Guizhou Province, is a typical karstic hydrological system, and has been contaminated by human inputs. In recent years, we carried out geochemical studies on the exchange of surface water and groundwater, controlling mechanisms of water/rock interaction on water geochemistry, and contamination of the groundwater by human activities for the karstic surface-groundwater system. In our previous papers (22, 23), we have distinguished the major contaminants from natural sources of mineral weathering in the groundwater system, traced the sources of major cations by using stoichiometrical analysis and Sr isotope chemistry, and identified the origin of nitrate by using its nitrogen and oxygen isotope composition. This paper will focus on the sources of sulfate and chloride in the groundwater, mainly based on the chlorine (δ37Cl) and sulfur (δ34S) isotope methods.
Experimental Section Geography and Geohydrological Background of Guiyang. Guiyang is located in the central part of Guizhou Province, covering an area ranging from 26°11′00′′ to 26°54′20′′ N and 106°27′20′′ to 107°03′00′′ E, with elevations of from 875 to 1655 m above mean sea level, with annual average temperature of 15.3 °C, and annual precipitation of 1200 mm. As shown on the hydrogeological map of Guiyang (Figure 1a), the outcrops of strata are principally sedimentary rocks, among which the Triassic strata are widespread (accounting for nearly 50% in area), followed by the Permian strata (accounting for nearly 20% in area). The rock types are predominantly shallow-sea platform carbonate rocks (dolomite and limestone), with a few Late Triassic continentalfacies clastic rocks. The main aquifers in the region of Guiyang are limestone and dolomite, totally accounting for more than VOL. 42, NO. 15, 2008 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
5421
FIGURE 1. Hydrogeological map of Guiyang, southwest China, and its surroundings, showing the lithology and geological structure of the studied area (a) and sampling locations and sample numbers (b). From Lang et al. (22). 80% of the aquifers. Other aquifers are mainly clastic rocks, including sandstone and carbonaceous shale, some of which are interlayed with sulfur-rich coal seams. Sulfate evaporite strata have not outcropped, but water chemical composition and sulfur isotope composition of sulfate in the groundwater at least indicate the existence of evaporite mineral in some strata of the region studied. The city of Guiyang is located in a basin, to the center of which both surface water and groundwater flow mainly from the north and south sideas (Figure 1b). The Nanming River, joined by its tributary the Xiaoche River, flows through Guiyang from the southwest, and has an annual average discharge of 12.3 m3/s. Since the downtown portion of Guiyang is located in the center of the basin, groundwater within the region studied is subjected to the influence of human activities. Municipal sewage discharged from enterprises, factories, and households on both sides of the Nanming River constitute a heteorogeneous mix of pollutants discharged into surface water and groundwater. The Aha Reservoir, one of the important sources of water supply, is often reported to contain excess amounts of Fe, Mn, and SO42- ions, mainly due to coal mining upstream. Sampling and Analytical Methods. The studied samples were collected during a period from January to August 2002, and included surface water, groundwater, sewage, and rainwater. The sampling sites are shown in Figure 1. A detailed description of the sample occurrences, physical and chemical parameters measured onsite, and chemical analysis of major ion chemistry of the samples has been published elsewhere (Lang et al., ref (22)). Sulfate was recovered from precipitation as BaSO4 with enough 2 mol/L BaCl2 solution. After precipitated for 15 min, 5422
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 42, NO. 15, 2008
the mixture was filtered through 0.22 µm acetate membrane filters. The precipitates (BaSO4) on the filters were carefully rinsed with enough Milli-Q water to remove Cl-. The precipitates were then immediately transferred into crucibles with the filters and combusted at 800 °C for 40 min in the air. Finally, the BaSO4 was decomposed to produce SO2 in a vacuum system using the V2O5-SiO2-Cu method of Yanagisawa and Sakai (24). The product SO2 was analyzed by mass spectrometry (MAT-252) to determine its isotopic ratio. The isotope result was finally expressed as CDT (Canyon Diablo Troilite) standard values. Laboratory prepared BaSO4 standard and two international S isotope reference samples (No. 220 seawater sulfate (NBS127) and No. 253 silver sulfide, both standards of IAEA-S-1) were used for calibration and the resultant error of analysis was below ( 0.1‰. The Cl isotope measurement of the samples was conducted in the laboratory of Prof. Satake, Department of Environmental Biology and Chemistry, Toyama University, Japan. The method of purification of CH3Cl from CH3I using cold trap with sealed 2,2,4-trimethylpentane for δ37Cl measurement was used, the detailed procedure of which is described in Wu and Satake (25). The standard used for data quality control is the SMOC (standard mean ocean chloride), and duplicated analyses of this standard resulted in a value of δ37Cl ) 0.00‰ ( 0.03‰ (2σ, n ) 15). The precision for Cl isotope analysis for one sample, on average, is 0.01‰ (1σ) based on 50 data collection. The measured Cl isotope ratios are expressed in the delta notation, defined as parts per thousand or per mil (‰) δ37Cl ) (Rsample/RSMOC - 1) × 1000, where Rsample is the 37Cl/35Cl ratio of the sample and RSMOC is that of standard mean ocean chloride (SMOC).
TABLE 1. Concentration (mg/L), Sulfur and Chlorine Isotopic Data of SO42- and Cl- in Ground Water, Surface Water, Rain Water, and Sewage Samples Collected from Guiyang in Winter and Summer Seasons winter season (January 2002) sample
SO42-
δ34S (‰)
Cl-
GYK-01 GYK-02 GYK-03 GYK-04 GYK-06 GYK-07 GYK-08 GYK-09 GYK-10 GYK-11 GYK-13 GYK-14 GYK-15 GYK-16 GYK-17 GYK-18 GYK-19 GYK-23 GYK-25 GYK-27 GYK-28 GYK-29 GYK-30 GYK-32 GYK-33 GYK-34 GYK-36
10.6 1072.3 104.6 260.2 22.1 807.4 47.0 106.6 26.9 44.2 45.1 184.3 119.0 161.3 130.6 138.2 18.2 49.0 79.7 240.0 16.3 73.0 93.1 7.7 36.5 63.4 150.7
+8.7 +20.9 +6.3 -3.3 +9.9 -20.4 -3.0 +1.6 -0.8 +4.0 -0.5 -5.5 -2.9 +1.1 -11 0.8 -7.8 -5.0 -10.4 -4.0 -3.1 -15.6 +1.8 -0.5 -14.8 -5.7 -4.0
3.2 10.7 115.4 35.9 6.7 14.6 6.0 31.6 7.1 14.2 4.3 38.0 138.8 13.1 38.3 25.6 8.9 5.3 2.5 42.2 18.1 25.6 40.5 25.2 4.6 10.3 33.4
GYK-05 GYK-20 GYK-21 GYK-22 GYK-24 GYK-26 GYK-31 GYK-35
254.4 73.0 66.2 64.3 274.6 250.6 200.6 141.1
+1.9 -5.7 -4.7 -4.9 -9.3 -9.6 -5.6 -4.5
GYK-12 GYK-37
282.2 149.8
811 815
6.0 4.4
summer season (August 2002) SO42-
δ34S (‰)
Cl-
δ37Cl (‰)
Groundwater no data GYF-01 +0.56 GYF-02 +0.46 GYF-03 no data GYF-04 no data GYF-06 +1.34 GYF-07 no data GYF-08 +0.19 GYF-09 +1.00 GYF-10 +0.30 GYF-11 no data GYF-13 no data GYF-14 +0.45 GYF-15 no data GYF-16 no data GYF-17 +0.17 GYF-18 no data GYF-19 +0.26 GYF-23 0.00 GYF-25 no data GYF-27 +2.03 GYF-28 no data GYF-29 +0.33 GYF-30 +1.92 GYF-32 no data GYF-33 no data GYF-34 +0.44 GYF-36
10.6 1035.8 74.9 223.7 no data no data 37.4 89.3 44.2 56.6 90.2 186.2 183.4 177.6 99.8 165.1 26.9 61.4 90.2 218.9 41.3 51.8 59.5 31.7 30.7 116.2 153.6
+7.3 +20.1 +1.3 -3.6 no data no data -1.8 -0.2 -2.7 +0.8 +9.2 -4.9 -4.6 0.6 -7.8 -2.1 -8.8 -3.8 -9.9 -5.4 -3.9 -13.4 -1.5 -1.7 -11.8 -8.0 -3.5
0.4 19.5 5.7 20.9 no data no data 2.5 9.2 4.6 9.2 2.8 33.4 28.0 7.8 25.9 18.5 5.3 3.9 1.1 41.9 5.7 1.1 9.9 4.3 2.1 3.6 31.6
no data 0.00 +0.08 no data no data no data no data -0.37 -0.74 -0.38 no data no data -0.06 no data no data +0.02 n.d. +0.29 -0.42 no data -1.46 no data -0.11 -1.12 no data no data -0.22
32.7 19.9 13.1 12.1 11.0 20.2 63.5 46.5
Surface water +0.16 GYF-05 +1.06 GYF-20 no data GYF-21 no data GYF-22 +0.24 GYF-24 +1.69 GYF-26 no data GYF-31 +0.12 GYF-35
99.8 49.9 70.1 71.0 213.1 185.3 97.0 143.0
-2.3 -7.3 -6.0 -6.2 -9.2 -8.1 -4.1 -7.8
8.2 4.6 6.7 5.3 7.1 7.5 12.4 10.7
-0.10 -0.92 no data no data -0.46 -0.49 no data +0.35
-8.0 -5.6
45.4 64.3
Sewage water +0.08 GYF-12 +0.15 GYF-37
246.7 113.3
-7.0 -4.3
36.6 38.0
no data no data
-4.87 -5.0
2.9 1.8
8.8
-5.13
3.1
-2.64
δ37Cl (‰)
sample
Rain water -4.07 813 no data
Results Sulfate and Chloride Concentrations. The SO42- and Clconcentrations in the water samples are shown in Table 1 together with the corresponding isotope compositions. Chloride contents of the groundwater and surface water were significantly higher in winter than in summer. The Clcontents of groundwater showed a variation of 2.5s138 mg/L in winter and a variation of 0.4s41.9 mg/L in summer, while the corresponding data of the surface water were in the range 11.0s63.5 mg/L in winter and 4.6s12.4 mg/L in summer. The SO42- contents of the groundwater and surface waters did not exhibit a discernible seasonal variation. However, a large part of groundwater samples in winter have lower SO42contents than in summer. The SO42- contents of the groundwater varied within a range of 7.7s260.2 mg/L in winter and a range of 10.6s223.7 mg/L in summer, while those of the surface water varied within a range of 64.3s254.4 mg/L in winter and a range of 49.9s213 mg/L in summer, except for two groundwater samples (sampling sites 2 and 7) from a gypsum-containing aquifer and a coal-containing aquifer, respectively. Except for several water samples collected from Xiaoche River, four sewage samples collected
from two locations in winter and summer show high concentrations of Cl-, Na+, K+, and SO42-, as compared with groundwater and surface water samples. The surface waters collected at sampling sites 5, 24, and 26 (shown in Figure 1) along Xiaoche River were found to contain much higher SO42concentrations in comparison with other surface water samples, which are ascribed to large inputs of SO42- from oxidation of sulfide mineral in coal-containing strata in the drainage basin of the Xiaoche River. The sample GYK-31 collected from the Nanming River was seriously contaminated by adjacent discharge of sewage, and hence also showed high SO42- content. The spatial distribution of sulfate and chloride contents of the groundwater in winter is shown in Figure 2. Groundwater samples collected from the densely populated area were of relatively high Cl- and SO42- contents. The contents of Cl- in the groundwater samples collected from the sampling sites along the Nanming River were higher than those from adjacent areas, suggesting that the discharge of wastewater from industrial and commercialsresidential areas has led to the increase of Cl- in groundwater (Figure 2a). Two springs (sampling sites 2 and 7) discharge water from VOL. 42, NO. 15, 2008 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
5423
FIGURE 2. Spatial variations in Cl- (a) and SO42- (b) concentration in the groundwater in winter. Open circle stands for no data. gypsum-containing and coal-containing aquifers, with clearly high sulfate concentrations. Excluding these two springs, the spring water in general showed a correlated distribution between sulfate (Figure 2b) and chloride content, with the higher concentrations of sulfate in spring in the central area of the city. Sulfur Isotope Compositions of Sulfate. The δ34S values of sulfate in the groundwater varied between -20.4‰ and +20.9‰, with the majority of data in the range from -5‰ to 0‰. Relative to the groundwater, the surface water exhibited δ34S values within narrow bounds of -9.6‰ to +1.9‰. Four sewage samples exclusively revealed negative values of δ34S ranging from -8.0‰ to -4.3‰, among those municipal wastewater (GYK-37) had higher δ34S values than industrial wastewater (GYK-12). The three rainwater samples were found to have a uniform level of δ34S in the interval from -5.1‰ to -4.9‰. Chlorine Isotope Compositions of Chloride. Two rainwater samples showed δ37Cl values of -4.07‰ and -2.64‰, respectively, being the lowest among all water samples analyzed. In turn, the groundwater in summer proved to be lower in δ37Cl values than those of groundwater in winter: the former has δ37Cl values of from -1.46‰ to +0.29‰, while the latter has δ37Cl values of from 0.00‰ to +2.03‰. Similar to the groundwater, the surface water in summer displayed lower δ37Cl values than during winter. The total variability of δ37Cl in the two series of water samples was comparable. Moreover, two sewage samples analyzed for δ37Cl were in this respect very similar (+0.08‰ and +0.15‰, respectively).
Discussion Quick Response of the Groundwater to Surface Water. Figure 3 shows the variation of Cl-/SO42- molar ratios as function of Cl- molar concentration for the surface water (Figure 3a) and groundwater (Figure 3b). The sewage and surface water samples collected in winter have much higher Cl- contents and Cl-/SO42- ratios than those collected in summer. The surface waters collected in summer are similar to most of the rain waters that perfectly show the lowest Clcontents and Cl-/SO42- ratios, as shown in Figure 3. The observations were in agreement with heavy rainfall during summer. The groundwaters, similar to the surface water, also showed both higher Cl- contents and Cl-/SO42- ratios in winter than in summer, and most of the groundwater collected in summer was similar to the surface waters. The covariation in chemical as well as isotopic composition (as discussed in following sections) between the groundwater 5424
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 42, NO. 15, 2008
FIGURE 3. Correlations of Cl- (mmol/L) vs Cl-/SO42- molar ratios of the surface waters (a) and groundwaters (b) sampled at Guiyang, southwest China. Except for the data of two rain waters studied here, the data of rain water are cited from Han and Liu (41). and surface water in summer strongly argues for a quick response of the groundwater to surface water. Two groundwater samples (GYK-3, GYK-15) collected in the north of the city show the highest Cl- contents and significantly high Cl-/SO42- ratios, which is most likely due to contamination by anthropogenic chloride inputs into the groundwater. Two additional groundwater samples (GYK-28, GYK-32) collected from the suburb to the northeast of the city nevertheless showed similar Cl- contents with the majority of the groundwater. This is probably due to the low SO42- contents, instead of contamination by anthropogenic Cl- inputs. Identification of Sulfate Sources. Figure 4 shows the variation of δ34S with 1/SO42- values for the studied water samples. The variations of both δ34S and 1/SO42- values of the water samples indicate that at least three sources have contributed to sulfate in the surface and groundwater. The sources were classified as atmospheric deposition (AD), dissolution of gypsum (GD), or oxidation of sulfide mineral (OS) in coal-containing strata.
FIGURE 5. Correlations of 1/Cl- (L/mmol) vs δ37Cl values of the waters sampled in both winter and summer seasons at Guiyang, southwest China. The end-member source named PB is paleobrine components probably in clastic rocks. Other end-members are the same as in Figure 4.
FIGURE 4. Variations of δ34S values vs 1/SO42- (L/mmol) of the waters sampled in both winter season (top) and summer season (bottom) at Guiyang, southwest China. The end-member sources estimated are also plotted as GD (gypsum dissolution), OS (oxidation of sulfide minerals), and AD (atmospheric deposits). Several researchers have studied the geochemical compositions of atmospheric deposits at Guiyang (26–29). The atmospheric chemistry at Guiyang was characterized by high concentrations of SO2 and particulate sulfur. SO2 emitted by coal combustion has an average δ34S value of -15.1‰, while particulate sulfur has an average δ34S of -3.6‰ (27). In addition, Hong et al. (26) indicated that the contribution of biogenic sulfur, with a δ34S value of -10.0‰, would become larger in summer. Xiao and Liu (28) carried out a systematic study on the variation of sulfur isotope and chemical composition for rainwater continuously collected from several rain events, and concluded that light rainfalls were characterized by lower δ34S values and higher sulfate concentration as compared to heavy rainfalls, and further suggested that the negative δ34S values of sulfate in light rainfalls should reflect a mixture mainly of local coalcombustion and biogenic sources. Three rainwater samples collected in summer season in this study show largely variable SO42- contents but relatively constant native δ34S values, which reflect most of the rain events at Guiyang during many years, as observed by previous studies (26–29). Spring 2 (sampling site 2) as shown in Figure 1 has dolomite-dominated aquifer containing gypsum, and showed the highest δ34S values in both summer and winter. The three springs 1, 3, and 12 located around the sampling site 2, also showed high δ34S values, and hence were most likely influenced by the gypsum-containing aquifer. Spring 7 has shale-dominated aquifer, and showed the lowest δ34S value, representing another end-member source of the sulfate. Sulfate originated from oxidation of sulfide mineral in sulfurenriched coal often shows significantly low δ34S values of sulfate (26). Several springs (e.g., sampling sites 25, 29, 33, and 34) located in a suburb possessed low δ34S values relative to the springs in the city area, probably suggesting a mixed source of sulfate from soil and coal, since the shale strata often contain coal sandwich according to the geology of Guiyang city and its surrounding areas. Considering that both groundwater and surface water in winter in general had higher δ34S values than in winter, we think that additional sulfates with lower δ34S values were derived from soil and/or clastic rock strata in summer. In addition, the water samples
collected in summer showed less variable δ34S and 1/SO42values, as compared to the water samples collected in winter, suggesting that large rainfall in summer homogenized the isotopic and chemical composition of surface and groundwater to some extent. The four sewage samples collected from two sites in both summer and winter seasons had relatively constant negative δ34S values of sulfate, similar to the rainwater. Identification of Chloride Sources. In Figure 5 the relationships of δ37Cl vs 1/Cl- concentration for the water samples are shown, where the corresponding data points will be located on a straight mixing line if the mixing takes place just between two end-members. Many authors applied this δ37Cl vs 1/Cl- diagram to elucidate mixing processes of two or more end-member sources (13, 30–32). Similar to δ34S vs 1/SO42- diagram (Figure 4), the δ37Cl vs 1/Cl- diagram (Figure 5) displays mixing relationships between at least three end-member sources, which probably include paleobrine components in clastic rocks, municipal sewage, and atmospheric deposits. Several groundwater samples collected from the springs of aquifers dominated by clastic rocks (GYK-32, GYK-7, GYK28), have the highest δ37Cl values. In recent years, many studies were carried out on formation water (31, 33), pore water in marine sediments (32, 34, 35), old groundwater (13), deep groundwater (36), halite and brine (31, 33), as well as water-soluble and structurally bound Cl- in sediment core (37). Those studies reported that δ37Cl values in all of the studied materials varied between -2.5‰ and +1.5‰. Our water samples collected in winter exclusively have positive δ37Cl values, some of which reach up to +2.03‰, probably the highest value observed for groundwater up to now according to our knowledge. The high δ37Cl chloride in the groundwater studied here is most likely derived from highly concentrated brine and/or structurally bound Cl- in phyllosilicate minerals such as clay and mica in clastic rocks, since highly concentrated brine and structurally bound Clhave been predicted by experimental and field geochemical studies to show high δ37Cl value (13, 31, 33, 37). The two municipal sewage samples show high Clconcentrations and δ37Cl values of near zero, suggesting a marine source of chloride from salt usage by city residents. The municipal sewage can be recognized to be an anthropogenic source of chloride in the groundwater system, because of the use of food salt, and their distinguishable high chloride contents. We have measured only two rainwater samples collected in summer season for their chlorine isotope composition. VOL. 42, NO. 15, 2008 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
5425
Up to now, chlorine isotope data of rainwater are still few. Volpe and Spivack (38) as well as Volpe et al. (38, 39) measured aerosol samples for chlorine isotope composition and reported a δ37Cl range between -0.9‰ and +2.5‰. Sun et al. (40) reported the δ37Cl values for rainwater samples collected from Xiamen (a coast city, in Fujian Province), Southern China Sea, Xining (the capital city of Qinghai Province), and for snow samples from Tibet. All of the samples studied by Sun et al. (40) have δ37Cl values from +0.51‰ to +1.8‰. The authors (38) ascribed the high δ37Cl values of marine aerosol to the enrichment in 37Cl- due to the volatilization of HCl from acidified aerosols, based on the observed and laboratory-derived correlations between δ37Cl value and Cl-. In contrast, the two rainwater samples studied here show significantly low δ37Cl values, which can serve as one end-member source in Figure 5. The low δ37Cl values of the studied rain waters argue clearly for an origin of anthropogenic emission most likely of HCl gas from industrial activities, instead of residual origin of chloride due to volatilization of HCl from acidified aerosols. This argument for anthropogenic origin of chloride in the rain waters is in accordance with high enrichment of Cl- over Na+ compared to original sea salt (Han and Liu (41)). The δ37Cl values of most water samples collected in summer season were negative, by group significantly lower than those collected in winter. Accordingly, there is no doubt that the surface water and groundwater received significant amounts of atmospheric inputs in summer, as the rainwater has the lowest δ37Cl values and as one of the end-member sources responsible for the low δ37Cl values of waters in summer. To sum up, the low-δ37Cl and low-Cl--content signatures can only plausibly be attributed to a spatially diffuse source, which is widely distributed in soil and washed out during a rain event. The nonpoint source of chloride might have been derived from degradation of anthropogenic agrochemicals (23) and chlorinated hydrocarbons (15), and/ or of natural organic materials containing chlorine, since some of the organochlorine is characterized by low δ37Cl value.
Acknowledgments We are grateful to two anonymous reviewers for their constructive comments that have significantly improved our manuscript in scientific writing. This work is financially supported by the Foundation of Chinese Academy of Sciences (International Partnership Project), by National Natural Science Foundation (Grant 40603004, 40603005, 40721002), and by the Ministry of Science and Technology of China (Grant 2006BC403200).
Literature Cited (1) Wicks, C. M.; Engeln, J. F. Geochemical evolution of a karst stream in Devils Icebox Cave, Missouri, USA. J. Hydrol. 1997, 198, 30–41. (2) Vaute, L.; Drogue, C.; Garrelly, L.; Ghelfenstein, M. Relations between the structure of storage and the transport of chemical compounds in karstic aquifers. J. Hydrol. 1997, 199, 221–238. (3) Plummer, L. N.; Busenberg, E.; McConnell, J. B.; Drenkard, S.; Schlosser, P.; Michel, R. L. Flow of river water into a Karstic limestone aquifer. 1. Tracing the young fraction in groundwater mixtures in the Upper Floridan Aquifer near Valdosta, Georgia. Appl. Geochem. 1998, 13, 995–1015. (4) Andreo, B.; Carrasco, F. Application of geochemistry and radioactivity in the hydrogeological investigation of carbonate aquifers (Sierras Balance and Mijas, southern Spain). Appl. Geochem. 1999, 14, 283–299. (5) Lee, E. S.; Krothe, N. C. A four-component mixing model for water in a karst terrain in south-central Indiana, USA: Using solute concentration and stable isotopes as tracers. Chem. Geol. 2001, 179, 129–143. (6) Gonfiantini, R.; Zuppi, G. M. Carbon isotope exchange rate of DIC in karst ground water. Chem. Geol. 2003, 197, 319–336. 5426
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 42, NO. 15, 2008
(7) Marfia, A. M.; Krishnamurthy, R. V.; Atekwana, E. A.; Panton, W. F. Isotopic and geochemical evolution of ground and surface waters in a karst dominated geological setting: a case study from Belize, Central America. Appl. Geochem. 2004, 19, 937– 946. (8) Bozau, E.; Strauch, G. Hydrogeological basis for biotechnological remediation of the acidic mining lake ‘RL 111’, Lusatia (Germany). Water Air Soil Pollut. Focus 2002, 2, 15–25. (9) Kno¨ller, K.; Fauville, A.; Mayer, B.; Strauch, G.; Friese, K.; Veizer, J. Sulfur cycling in an acidic mining lake and its vicinity in Lusatia, Germany. Chem. Geol. 2004, 204, 303–23. (10) Schiff, S. L.; Spoelstra, J.; Semkin, R. G.; Jeffries, D. S. Drought induced pulses of SO42- from a Canadian shield wetland: use of δ34S and δ18O in SO42- to determine sources of sulfur. Appl. Geochem. 2005, 20, 691–700. (11) Pellicori, D. A.; Gammons, C. H.; Poulson, S. R. Geochemistry and stable isotope composition of the Berkeley pit lake and surrounding mine waters, Butte, Montana. Appl. Geochem. 2005, 20, 2116–2137. (12) Lavastre, V.; Jendrzejewski, N.; Agrinier, P.; Javoy, M.; Evrard, M. Chlorine transfer out of a very low permeability clay sequence (Paris Basin, France): 35Cl and 37Cl evidence. Geochim. Cosmochim. Acta 2005, 69, 4949–4961. (13) Zhang, M.; Frape, S. K.; Love, A. J.; Herczeg, A. L.; Lehmann, B. E.; Beyerle, U.; Purtschert, R. Chlorine stable isotope studies of old groundwater in southwestern Great Artesian Basin. Australia Appl. Geochem. 2007, 22, 557–574. (14) Eastoe, C. J.; Peryt, T. M.; Petrychenko, O. Y.; Geisler-Cussey, D. Stable chlorine isotopes in Phanerozoic evaporates. Appl. Geochem. 2007, 22, 575–588. (15) Jendrzejewski, N.; Eggenkamp, H. G. M.; Coleman, M. L. Characterization of chlorinated hydrocarbons from chlorine and carbon isotopic compositions: Scope of application to environmental problems. Appl. Geochem. 2001, 16, 1021–1031. (16) van Warmerdam, E. M.; Frape, S. K.; Aravena, R.; Drimmie, R. J.; Flatt, H.; Cherry, J. A. Stable chlorine and carbon isotope measurements of selected chlorinated organic solvents. Appl. Geochem. 1995, 10, 547–552. (17) Aravena, R.; Frape, S. K.; van Warmerdam, E. M.; Drimmie, R. J.; Moore, B. J. Use of environmental isotopes in organic contaminants research in groundwater systems. In Proceedings of the Symposium Isotopes in Water Resources Management, 1996, Vol. 1, pp 31-42. (18) Beneteau, K. M.; Aravena, R.; Frape, S. K. Isotopic characterization of chlorinated solvents laboratory and field results. Org. Geochem. 1999, 30, 739–753. (19) Sie, P. M. J.; Frape, S. K. Evaluation of the groundwaters from the Stripa mine using stable chlorine isotopes. Chem. Geol. 2002, 182, 565–582. (20) Shouakar-Stash, O.; Frape, S. K.; Drimmie, R. J. Stable hydrogen, carbon and chlorine isotope measurements of selected chlorinated organic solvents. J. Contam. Hydrol. 2003, 60, 211–228. (21) Annable, W. K.; Frape, S. K.; Shouakar-Stash, O.; Shanoff, T.; Drimmie, R. J.; Harvey, F. E. 37Cl, 15N, 13C isotopic analysis of common agro-chemicals for identifying non-point source agricultural contaminants. Appl. Geochem. 2007, 22, 1530–1536. (22) Lang, Y.-C.; Liu, C.-Q.; Zhao, Z. Q.; Li, S.-L.; Han, G. Geochemistry of surface and ground water in Guiyang City, China: Water/ rock interaction and pollution in a karst hydrological system. Appl. Geochem. 2006, 21, 887–903. (23) Liu, C.-Q.; Li, S.-L.; Lang, Y.-C.; Xiao, H.-Y. Using δ15N- and δ18O-Values to identify nitrate sources in karst ground water, Guiyang, Southwest China. Environ. Sci. Technol. 2006, 40, 6928– 6933. (24) Yanagisawa, F.; Sakai, H. Thermal decomposition of barium sulfate-vanadium pentaoxide-sillica glass mixtures for preparation of sulfur dioxide in sulfur isotope ratio measurements. Anal. Chem. 1983, 55, 982–987. (25) Wu, J. H.; Satake, H. Purificaton of CH3Cl from CH3I using cold trap with sealed 2,2,4-trimethylpentane for δ37Cl measurement. Anal. Chim. Acta 2006, 555, 41–46. (26) Hong, Y.; Zhang, H.; Zhu, Y.; Piao, H.; Jiang, H.; Liu, D. Sulfur isotopic characteristics of rain water in China. Prog. Nat. Sci. 1994, 741–745. (27) Liu, G.; Hong, Y.; Piao, H.; Zeng, Y. The sulfur isotopic compositional characteristics of atmospheric particulates from near the ground in the urban and suburban areas of Guiyang city. Acta Mineral. Sin. 1996, 16, 353–357. (28) Xiao, H.-Y.; Liu, C.-Q. Sources of nitrogen and sulfur in wet deposition at Guiyang, southwest China. Atmos. Environ. 2002, 36, 5121–5130.
(29) Xiao, H.-Y.; Liu, C.-Q. Chemical characteristics of water-soluble components in TSP over Guiyang, SW China, 2003. Atmos. Environ. 2004, 38, 6297–6306. (30) Hendry, M. J.; Wassenaar, L. I.; Otzer, T. Chloride and chlorine isotopes (36Cl and δ37Cl) as tracers of solute migration in a thick, clay-rich aquitard system. Water Resour. Res. 2000, 36, 285– 296. (31) Eastoe, C. J.; Long, A.; Land, L. S.; Kyle, J. R. Stable chlorine isotopes in halite and brine from the Gulf Coast Basin: brine genesis and evolution. Chem. Geol. 2001, 176, 343–360. (32) Godon, A.; Jendrzejewski, N.; Castrec-Rouelle, M.; Dia, A.; Pineau, F.; Boule’gue, J.; Javoy, M. Origin and evolution of fluids from mud volcanoes in the Barbados accretioary complex. Geochim. Cosmochim. Acta 2004, 68, 2153–2165. (33) Eastoe, C. J.; Long, A.; Knauth, L. P. Stable chlorine isotopes in the Palo Duro Basin, Texas: Evidence for preservation of Permian evaporite brines. Geochim. Cosmochim. Acta 1999, 63, 1375– 1382. (34) Ransom, B.; Spivack, A. J.; Kastner, M. Stable Cl isotopes in subduction-zone pore waters: implications for fluid-rock reactions and the cycling of chlorine. Geology 1995, 23, 715–718. (35) Spivack, A. J.; Kastner, M.; Ransom, B. Elemental and isotopic chloride geochemistry and fluid flow in the Nankai Trough. Geophys. Res. Lett. 2002, 29, doi:10.1029/ 2001GL014122.
(36) Shouakar-Stash, O.; Alexeev, S. V.; Frape, S. K.; Alexeeva, L. P.; Drimmie, R. J. Geochemistry and stable isotopic signatures, including chlorine and bromine isotopes, of the deep groundwaters of the Siberian Platform, Russia. Appl. Geochem. 2007, 22, 589–605. (37) Barnes, J. D.; Sharp, Z. D. A chlorine isotope study of DSDP/ ODP serpentinized ultramafic rocks: Insights into the serpentinization process. Chem. Geol. 2006, 228, 246–265. (38) Volpe, C.; Spivack, A. J. Stable chlorine isotopic composition of marine aerosol particles in the western Atlantic Ocean. Geophys. Res. Lett. 1994, 21, 1161–1164. (39) Volpe, C.; Wahlen, M.; Pszenny, A. A. P.; Spivack, A. J. Chlorine isotopic composition of marine aerosols: Implications for the release of reactive chlorine and HCl cycling rates. Geophys. Res. Lett. 1998, 25, 3831–3834. (40) Sun, A.; Xiao, Y.; Wang, Q.; Zhang, C.; Wei, H.; Liao, B.; Li, S. The separation and stable isotopic measurement of chlorine in low-concentration liquid sample. Chinese J. Anal. Chem. 2004, 10, 1362–1364. (41) Han, G.; Liu, C.-Q. Strontium isotope and major ion chemistry of the rainwaters from Guiyang, Guizhou Province, China. Sci. Total Environ. 2006, 364, 165–174.
ES800380W
VOL. 42, NO. 15, 2008 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
5427
