Sulfur isotope evidence - Wiley

0 downloads 11 Views 137KB Size Report
Jun 22, 2012 - [Karim and Veizer, 2000; Spence and Telmer, 2005;. Calmels et al., 2007]. Geological maps and other geo- chemical studies document the ...
GEOPHYSICAL RESEARCH LETTERS, VOL. 39, L12404, doi:10.1029/2012GL051549, 2012

Disproportionately high rates of sulfide oxidation from mountainous river basins of Taiwan orogeny: Sulfur isotope evidence Anirban Das,1,2 Chuan-Hsiung Chung,1,3 and Chen-Feng You1,3 Received 29 February 2012; revised 24 May 2012; accepted 28 May 2012; published 22 June 2012.

[1] Sulfur isotopic tracing of river sulfate (SO2 4 ) suggests that sulfide oxidation accounts for 85  7% of the in one of the largest river systems (the dissolved SO2 4 Kaoping) of the Taiwan orogeny in the high-discharge season. This corresponds to a basin-wide contribution of exported by rivers 1.2–1.6% of pyrite-derived SO2 4 globally, from a river basin, which is only 0.003% of the global drainage area. Intense rainfall, high relief and active tectonics all favor intense physical weathering of the Kaoping basin, which promotes continuous exposure of fresh (sulfide) mineral surfaces for oxidative weathering. This coupling between physical and chemical weathering sustains disproportionately high sulfide oxidation, 400 times relative to surface area, in the Kaoping basin. Citation: Das, A., C.-H. Chung, and C.-F. You (2012), Disproportionately high rates of sulfide oxidation from mountainous river basins of Taiwan orogeny: Sulfur isotope evidence, Geophys. Res. Lett., 39, L12404, doi:10.1029/2012GL051549.

1. Introduction [2] The sources of dissolved sulfate (SO2 4 ) in river waters can be inferred based on investigations of its sulfur isotope composition. Both natural processes (e.g., weathering of sulfides, and evaporite dissolution) and anthropogenic activities (agriculture, industrial and domestic) contribute to river SO2 4 . Sulfur isotope composition can also provide insights into the role of sulfide oxidation supplying protons for chemical weathering in a river basin. Berner and Berner [1996] while identifying the importance of sulfide oxidation to chemical weathering both on regional and global scales underscored the difficulty in constraining its quantitative significance in catchments where multiple sources contribute 2 to the SO2 4 pool. Current estimates of sulfide derived SO4 12 used in elemental cycle models range from 0.48  10 to 0.65  1012 mol y1 [Berner and Berner, 1996; François and Walker, 1992]; this estimate is based on model calculations and not made directly from river chemistry/isotope data. Recently, Calmels et al. [2007] estimated that sulfide oxidation occurring in the Mackenzie basin accounts for 1 Department of Earth Sciences, National Cheng-Kung University, Tainan, Taiwan. 2 School of Petroleum Technology, Pandit Deendayal Petroleum University, Gandhinagar, India. 3 Earth Dynamic System Research Center, National Cheng-Kung University, Tainan, Taiwan.

Corresponding author: A. Das, Department of Earth Sciences, National Cheng-Kung University, No. 1, University Road, Tainan 701, Taiwan. ([email protected], [email protected]) ©2012. American Geophysical Union. All Rights Reserved. 0094-8276/12/2012GL051549

about 20–27% of the global estimate, a result that prompted them to suspect that the available global estimates of sulfide derived SO2 4 may be substantially underestimated. [3] The Rivers of Taiwan orogeny are characterized by disproportionately (relative to their fractional area) higher water and sediment discharges to the ocean due to intense rainfall and physical erosion. This in turn can promote chemical weathering and export of elements to oceans. Chung et al. [2009] reported elevated concentrations of (560–780 mM) in samples from the Kaoping River SO2 4 and its major tributaries, compared to world average value also have been of 100 mM. High abundances of SO2 4 reported for the Liwu and the Danshuei Rivers [Calmels et al., 2011; Chu and You, 2007], located in the northern parts of Taiwan. These high concentrations were attributed to oxidation of sulfides based on occurrences of sulfides in the river basins, however observational evidence to confirm such a hypothesis is lacking. Motivated by the above observations a study was initiated with the aim to: (i) trace and their apportionment in the the sources of SO2 4 Kaoping River, (ii) determine the role of sulfide oxidation in supplying SO2 4 to the Kaoping basin undergoing high physical erosion.

2. Study Area [4] The Kaoping River is one of the large rivers of Taiwan with a catchment area of 3257 km2 and length of 170 km before flowing (southwesterly direction) into the Taiwan Strait (Figure S1 in auxiliary material).1 It has five major tributaries and sub-tributaries; these are the Cishan, Laonong, Baoloi, Chukou and the Ailiao. The Cishan and Laonong contribute more than 70% of the annual water discharge of the Kaoping. The annual temperature in the basin varies from 18 to 29 C (Y. C. Liu et al., Boron sources and transport mechanisms in river catchments: Isotopic evidences, submitted to Journal of Asian Earth Sciences, 2012). Wet season is from May to October and accounts for about 90% of the annual rainfall of 3000 mm y1 [Chung et al., 2009]. Episodic typhoon activities are common in this basin, which bring high amounts of rainfall that are important in the local hydrological cycle. The Kaoping River basin is characterized by high relief and high rainfall; these make the denudation rates in this basin one of highest amongst the global river basins [Chung et al., 2009]. Ho [1975] has classified the main lithological units of the Kaoping River basin, consisting of sedimentary and metamorphic rocks of various ages, into: (i) Black and green schist of Paleozoic to Mesozoic age, 1 Auxiliary materials are available in the HTML. doi:10.1029/ 2012GL051549.


1 of 6



(ii) Eocene-Oligocene slate and quartzite, (iii) Miocene to Pliocene sandstone and shale, (iv) Pleistocene terrace deposits and alluvium, and (v) Modern day alluvium. The upper reaches of the Laonong basin is also characterized by occurrence of basaltic outcrops. [5] Land use pattern estimates show that forested land covers 60% of the basin area, along the eastern mountainous regions. Orchard, farmland and grassland together constitute 15% while human settlements accounts for 4% of the basin area [Ning et al., 2006]. The Kaoping River also has a long history of water quality issues due to inappropriate disposal of manure from stock farming, industrial and domestic effluents [Ning et al., 2001].

3. Methods [6] This study, one of the few to report the application of Multiple Collector-Inductively Coupled Plasma Mass Spectrometry (MC-ICPMS) for measurements of sulfur isotopic composition in natural waters (d34S), follows the procedures of Craddock et al. [2008]. Sulfur isotope composition was measured in archives of samples (Figure S1 in auxiliary material) collected in August 2004 (wet season) and December 2009–January 2010 (dry season); these samples were stored refrigerated (4 C) until analysis. The concentrations of dissolved SO2 4 in these samples are from Chung et al. [2009] and Liu et al. (submitted manuscript, 2012). For isotopic analysis, 10 mL of water sample was dried and the sulfate in the residue was purified using cation exchange (AG-50W-X8; 200–400 mesh; resin volume 2.5 mL). In order to address concerns about fractionation of sulfur isotopes during the analytical procedures, care was from the taken to ensure almost 100% recovery of SO2 4 samples. This was ascertained by control experiments in which known amounts of SO2 4 was passed through resin columns and measuring its recovery (98–100%). Sulfur isotopes were measured in purified SO2 4 solution following the widely preferred standard bracketing method using Spex sulfur standard, all measured at concentrations of 20 ppm (20% [Craddock et al., 2008]). d34S values of samples are reported against the internationally accepted reference standard (VCDT) using the published value of IAEA S-1 against Spex, and IAEA S-1 against VCDT [Craddock et al., 2008]. Based on duplicate analysis the precision of measurements was better than 0.2‰.

4. Results and Discussion [7] The dissolved SO2 concentrations (Table S1 in 4 auxiliary material) in samples from 2004 vary in a narrow range (561–778 mM; average: 690  72 mM) compared to samples from 2010–11 (437–1767 mM; average: 1090  305 mM). About 90% of the samples have d34S values less than 0‰ (range: 9.0‰ to +3.2‰; average: 3.3‰). The variability in d 34S in samples from 2004 (9.0‰ to 5.5‰; average: 6.4  1.1‰) is less than those of the 2009–10 samples (6.4‰ to +3.2‰; average: 1.9  2.3‰) indicating that there are seasonal and/or inter-annual changes in d 34S values. [8] All the river water samples, including the upstream samples from the Lanong and Cishan Rivers (LN-01, LN-T01, CS-01 and CS-T06), the smaller tributaries of the Lanong (AL-01; LN-T09; CK), and the Kaoping


downstream sample (MK-PO1) have concentrations of SO2 4 5–13 times higher compared to the estimated global average. This indicates that processes involving high inputs of SO2 4 to the rivers studied are not limited to small lithological pockets; rather it is a widespread process in the Kaoping River basin. In the following, based on SO2 4 concentrations in rivers and local rains and their d34S, an attempt has been made to construct a budget for dissolved SO2 4 . 4.1. Sources of Dissolved SO2 4 [9] The precipitation- weighted Cl and SO2 4 concentrations in rainwater from a site near to the Kaoping basin is 103 and 46 mM, respectively [Cheng and You, 2010]. The average rainwater Cl is higher than the Cl concentrations of all the 2004 river water samples. This indicates that the Cl in rivers during the wet season is likely to be of atmospheric origin (cyclic sea-salt) and there is no major contribution of Cl from halites; an inference consistent with the lack of halite exposures in the Kaoping basin [Ho, 1975]. From the precipitation-weighted SO2 4 in rain it is estimated that the rainwater contribution to the riverine SO2 4 is 6– 8% (for 2004 samples) and 3–14% (2009–10 samples). This calculation assumes 0% and 25% evapo(transpira)tive enrichments of the rainwater solute concentrations during the wet and dry seasons, respectively. From isotopic consideration, Ezoe et al. [2002] have reported rainwater d 34S values (+3.6‰ and +2.9‰) from northern Taiwan and inferred that these low d34S values are indicative of anthropogenic influences. The d34S values of rain from Taiwan are similar to those (+4‰ to +6‰) reported for rains worldwide [Brenot et al., 2007, and references therein], but are significantly higher to explain the low d 34S of SO2 4 in river water samples collected in 2004 (d 34S = 6.4  1.1‰). Even though the 2009–10 samples are relatively enriched in 34S (1.9  2.3‰) and are isotopically closer to the rainwater end-member, the relatively low concentration of SO2 4 (46 mM) in rain is unlikely to control the riverine d34S. in rivers must Therefore, the bulk of the dissolved SO2 4 originate from other sources; potential natural sources include oxidation of sulfide minerals, and gypsum dissolution, whereas fertilizers, and domestic and industrial effluents can be the anthropogenic contributors. [10] The d34S in the samples shows a wide range (9.0‰ to +3.2‰), however, during wet and dry seasons the spreads are narrower and different (9‰ to 5.5‰ and 6.4‰ to +3.2‰). These findings make the variability in d 34S more likely a result of mixing of SO2 4 contributed from natural and anthropogenic sources rather than only due to basinwide variations in d34S of source rocks. Figure 1a shows the variations in d34S against the inverse of SO2 4 concentration (see Figure 1b for possible end-members). It can be seen in Figure 1a that samples of the dry season show higher d 34S and SO2 4 concentrations compared to samples of the wet season. One of the samples deviates from the trend set by the others, the reason for which is unclear (Figure 1a). Enrichments due to evaporation, which occur in dry season, can only lead to higher SO2 4 in samples without any change in d34S. Increases of d34S can result from microbial sulfate reduction; such increase however is generally accompanied by a decrease in SO2 4 concentrations, which has not been observed in the case of samples from the dry season. Therefore, the observed trend between the summer and winter samples would require that a part of riverine SO2 4 in

2 of 6




2 34 Figure 1. (a) Scatter plot of d34S-inverse of SO2 4 concentrations. Dry season samples show increases in d S and SO4 concentrations, shown by shaded -arrows and –rectangle. (b) Scatter plot with the possible end-members A (weathering), B (anthropogenic) and C (rainwater) shown by ellipses (see text for details). The river water data essentially fall along the mixing line between end-members A and B with minor contributions from C. The arrows in ellipses A and B are an indication that SO2 4 contributed by these end-members represented by the ellipses could be lower limits. Symbols (inverted triangle, circle, triangle, square and diamond, respectively) represent samples of Cishan Mainsteam, Lanong mainstream, Chukou, Kaoping downstream (after Cishan and Lanong merge) and Ailiao River. Open and closed symbols represent samples of the wet and dry season respectively. Data for the rainwater are taken from Ezoe et al. [2002] and Cheng and You [2010].

the dry season has to originate from another source that is isotopically heavier than the source rocks. The overall trend (Figure 1b) can be explained by simple mixing processes between end-member A (lowest d34S with intermediate SO2 4 ) and end-member B (highest d34S with highest SO2 4 ) with minor contribution from a third end-member (C), i.e., rain water with intermediate d 34S with lowest SO2 4 (Figure 1b). [11] The sources or end-members A and B (Figure 1b) need to be identified to better understand their significance 34 in budgets of SO2 4 and d S in rivers. Global data suggest 34 that negative d S values in river water SO2 4 are characteristic of sulfide oxidation occurring in their watersheds [Karim and Veizer, 2000; Spence and Telmer, 2005; Calmels et al., 2007]. Geological maps and other geochemical studies document the existence of both sedimentary and metamorphic sulfides in the Kaoping basin [Kao et al., 2004; Horng et al., 1992; Horng and Roberts, 2006]. Chen et al. [2004] reported d34S of (27.5‰ to +15.2‰) in pyrites from a 200 meter long sediment core drilled in the coastal plain of Southwestern Taiwan. In this core seventeen out of nineteen samples analyzed have d34S values lower than 0‰ (range: 1.8‰ to 21.1‰) with the sample closest to the surface (40 cm below) having a d34S of 8.3‰, and samples representing the onset of Pleistocene epoch having d34S of  6‰. These results suggest that sedimentary sulfides in the Kaoping basin are dominated by negative d34S values, similar to that measured in rivers studied. In addition to pyrites in the basin, Horng et al. [1992] and Horng and Roberts [2006] also reported occurrences of other sulfide minerals such as greigite and pyrrhotite in bulk sediments from two PlioPleistocene sedimentary sections of southwestern Taiwan and in the Kaoping River sediments. Sulfides from igneous rocks generally have d34S in the range of 010‰ [Thode, 1991]; a recent measurement on pyrrhotite mineral yielded a value of 1.4‰ [Kozdon et al., 2010]. Given that no isotope data are available on metamorphic rocks from the Kaoping basin, it remains unclear if they can have d34S values similar to the sedimentary sulfides like those measured in sulfides from the sedimentary core. The end-

member A in a mixing diagram (Figure 1b) represents that originates from oxidation of (sedimentary + SO2 4 metamorphic) sulfide minerals in the basin. [12] Marine evaporites (containing gypsum, anhydrite), usually characterized by higher d34S values (+10‰ to +30‰ [Claypool et al., 1980]), have not been reported in the basin. Thus, this source is less likely to be a possible end-member B in Figure 1b. More likely candidates are anthropogenic sources (agriculture, industrial and domestic wastes), which are known to supply SO2 4 to rivers [Victoria et al., 2004; Brenot et al., 2007; Li et al., 2011; Szynkiewicz et al., 2011]. Liu et al. (submitted manuscript, 2012) have provided evidences of anthropogenic inputs to the Kaoping River based on boron isotopes. Szynkiewicz et al. [2011] in a recent compilation of d34S values of common fertilizers used in several countries find that except for fertilizers from New-Zealand, all country-wide d 34S averages cluster around 4  4‰; the number weighted average of all samples being +5.3‰ [Szynkiewicz et al., 2011, Table 1]. Data on d34S of anthropogenic sources is even more limited; for domestic wastes it ranges from +8.5‰ to +13.6‰ [Victoria et al., 2004] and for industrial effluents it varies between +8.0‰ to +14.0‰ [Soler et al., 2002; Das et al., 2011]. A lower limit of +5.3‰ and an upper-limit of +10.4‰ (following Li et al. [2011]) have been adopted for d 34S of the anthropogenic end-member B (Figure 1). 4.2. Mixing Model and Sensitivity Analysis [13] Quantitative information on the contributions of various sources to the dissolved constituents of rivers is generally inferred from mixing model/equations. For d 34S of the river a balance equation is written in terms of the relative contributions of SO2 4 (F) from the end-members (rainwater, weathering and anthropogenic denoted by rw, w and anthro, respectively) and their d34S values:

3 of 6

d 34 Sriv ¼ Frw  d 34 Srw þ Fw  d 34 Sw þ Fanthro  d 34 Santhro Frw þ Fw þ Fanthro ¼ 1

ð1Þ ð2Þ




a Table 1. Range of Contributions (F) to the Dissolved SO2 4 From Different End-Members, Calculated Using the Mixing Model

End-Member Composition d 34S in (‰) Fixed Parameters 34


d Santhro = +7.8; d Srw = +4.5 d 34Sw = 7.7; d 34Srw = +4.5 d 34Sw = 7.7; d 34Santhro = +7.8

End-Member Contributions (%)

Variable Parameter




d Sw = 6.4 to 9.0 d 34Santhro= +5.3 to +10.4 d 34Srw= +3 to +6

78–92 (66–78) 83–86 (67–74) 84–85 (70–72)

1–15 (17–29) 8–11 (20–28) 8–9 (23–24)

7 (6) 7 (6) 7 (6)


a Sensitivity analysis on Fw and Fanthro has been made within the d 34S ranges of three end-members. Sensitivity of d 34S of one end-member on F has been made keeping the other two d 34S values constant at their mean values. Frw has been calculated from SO2 4 concentrations in rainwater and river water. Subscripts w, anthro and rw refer to weathering (sulfide), anthropogenic and rainwater end-members, respectively. F values without and within brackets are estimates for the wet and dry season respectively (see text for details of the model calculations).

From the two equations above, the fraction of the dissolved SO2 4 supplied by weathering sources (Fw) can be calculated from the following equation.  Fw ¼

  d 34 Sriv  d 34 Santhro  Frw  d 34 Srw  d 34 Santhro  ð3Þ d 34 Sw  d 34 Santhro

The reliability of the estimated Fw depends on how well the inputs Frw and (d 34Sriv,w,rw,anthro) are constrained. Assigning unique values for d34S of different sources (in equation 3) is not straightforward given the complexities in natural systems. d34Sriv, typically, is better constrained compared to the other d34S values appearing in equation 3, and therefore, a range for each of the three input parameters (d 34Sw, d34Santhro and d34Srw) have been selected. Frw, the fraction of SO2 4 contributed by rainwater (section 4.1) to the rivers, is the fixed parameter in the model. [14] A sensitivity analysis of the model has been made to assess the effect of varying input parameters (here, d 34Sw, d34Santhro and d34Srw) on estimates of Fw and Fanthro. This analysis was performed by calculating the effect of varying one of the input parameters (e.g., d34Sw) on the output keeping other parameters constant at their average values (Figure S2 in the auxiliary material). Two approaches have been used to constrain the range of d34Sw. The average d34S in samples of the wet season (2004) which vary in a narrow range of 6.4  1.1‰ can be taken as a reasonable estimate of d 34Sw. Correction for rainwater input also induces marginal changes in this average value (6.7‰). Another approach could be to use the d 34Sriv of smaller tributaries, which are expected to have minimal influence of anthropogenic sources and are likely to inherit their dissolved constituents from weathering. For this approach, the d34Sriv of Chukou (a tributary of the Lanong; d 34S = 9.0‰ in sample CK, Table S1) is chosen. Therefore, in the mixing model and the sensitivity analysis we have varied d34Sw within the range of 6.4‰ to 9.0‰. The d 34S of the rainwater component is taken as +3‰, representative of d34S values measured in Taiwan [Ezoe et al., 2002]. This value is similar to +3.9‰ used in the Sichuan Province, China [Li et al., 2011], and close to the global values, +4‰ to +6‰ [Brenot et al., 2007, and references therein]. Based on these data, the d34Srw of rain water is varied between +3‰ to +4.5‰ for the sensitivity analysis. Similarly, from the discussion in the earlier section the range selected for d34Santhro is +5.3‰ to +10.4‰. [15] The fractions of SO2 4 contributed by different sources (i.e., Frw, Fw and Fanthro) to the Kaoping river watershed were estimated (Table 1) using the various range of end-member values discussed above. Samples KP-3 and MK-PO1 are

used for calculations as these are collected from locations which in our opinion integrate the effects of all processes in 34 the entire watershed controlling SO2 4 inputs and their d S values. Results of the sensitivity analysis show that uncertainties in the choice of d 34Sw have the biggest impact on the uncertainties in the F estimates (Figure S2 in the auxiliary material). The model results show that for the Kaoping River (sample KP-3; wet season) the SO2 4 load contributed by rainwater, anthropogenic and weathering account for 7  0%, 8  7% and 85  7%, respectively, while the corresponding values for the dry season (MK-PO1) are 6  0%, 23  6% and 72  6%. The uncertainties cover the range in estimates arising from selection of different values for d 34S in equation 3. From the average values, it is observed that the supply of anthropogenic SO2 4 is higher in the dry than in wet season. This is most likely because of less sulfide oxidation in the dry season due to less runoff/ rainfall, as a result of which the relative contribution from anthropogenic sources to the riverine SO2 4 increases. 4.3. Sulfide Oxidation Rates in Kaoping Basin [16] From the d34S modeling, it is observed that large proportion of the riverine SO2 4 in the Kaoping is derived from the oxidative weathering of sulfide minerals. This accounts for about 85% of the SO2 4 in the wet season which decreases to 72% in the dry season because of larger contribution from anthropogenic sources. The relative importance of the sulfide source to the annual contribution of SO2 4 from the Kaoping basin to the ocean is estimated from the seasonal water discharge data; it is, however, pertinent to mention here that 90% of water discharge in the Kaoping region occurs during the wet season [Wang et al., 2009]. Based on available data on water discharge during the wet season of 2004 (1.33  1013 L y1 [Wang et al., 2009]), the known area of the Kaoping drainage (3257 km2), the measured concentration of 693 mM in location KP-3 (Figure S1), and the estimated Fw values (Table 1), a sulfide oxidation 2 1 y is estimated rate of 2.4  0.2  106 mol of SO2 4 km for the Kaoping River watershed. A similar assessment for the dry season has not been made of the same year due to absence of samples; however, the annual export is expected to increase at least by 10% since SO2 4 is usually higher in the dry season compared to wet season and 10% annual discharge occurs in dry season [Wang et al., 2009]. [17] Table S2 in the auxiliary material summarizes results of reported studies on sulfide oxidation rates and basin characteristics. It is observed that the rate for Kaoping is 1–2 orders of magnitude higher than those of other basins even if only the contribution from the wet season is considered. The sulfide oxidation rate for the Kaoping basin thus contributes

4 of 6



1 7.8  0.6  109 mol of SO2 4 y , which is about 1.2–1.6% 2 of the pyrite-derived SO4 flux estimated for World Rivers (0.48  1012 to 0.65  1012 mol y1 [François and Walker, 1992; Berner and Berner, 1996] even though its area is only 0.003% of the global drainage area. Considering the relatively small size of the Kaoping basin (0.003%) it can be concluded that sulfide weathering accounts for disproportionately higher (400 times relative to surface area) delivin this basin. This study, coupled with the ery of SO2 4 earlier ones (Table S2), suggest that these rivers may account for 24–33% of the global sulfide derived SO2 4 to the ocean, from an area of only 2% of global drainage area. The present study supports the suggestion by Calmels et al. fluxes are [2007] that the global sulfide-derived SO2 4 underestimated and that there is a need for more field based studies to constrain these fluxes better. [18] The two main factors that may control the high sulfide oxidation rates in the Kaoping basin are the ubiquitous presence of sulfides (e.g., pyrite, pyrrhotite) in the sedimentary and metamorphic rocks of the watershed [Kao et al., 2004; Horng and Roberts, 2006] and the continuous exposure of fresh mineral surfaces by mechanical weathering. The abundance of total sulfur ranges from 0.05– 0.27% in two sedimentary sections from the southwest Taiwan [Kao et al., 2004]. This represents 1.5–8 times of the average sulfur abundance in upper crust (309 mg g1 [Gao et al., 1998]). Further, the high rainfall (runoff), higher relief, presence of easily erodable rocks and frequent earthquakes all favor enhanced mechanical weathering of the rocks. Rock mass shattering and landslides are induced by frequent earthquakes as inferred from the observed correlation between seismic moments and erosion rates [Dadson et al., 2003]. Heavy rainfall and typhoons are common in the Kaoping basin triggering landslides which mobilize sediments even from the mountain belts. It has also been shown that storm-runoff has a first-order control on erosion rates [Dadson et al., 2003]. Higher mechanical erosion has been manifested in higher suspended and bed load in the Kaoping river, and in general, in all rivers of Taiwan. The Taiwanese rivers account for about 1.9% of the global suspended load, derived from 0.024% of the Earth’s subaerial surface [Dadson et al., 2003]. The high rates of mechanical erosion in the Kaoping watershed continuously exposes fresh surface area for chemical weathering and this coupling significantly accelerates sulfide oxidation rates.

5. Conclusions [19] Chung et al. [2009] reported elevated SO2 con4 centrations in surface water samples from the Kaoping River. In this study an attempt has been made to trace the 34 sources of SO2 4 based on sulfur isotopes. The d S in the water samples ranged from 9.0‰ to +3.2‰, and notably, 90% of the samples were characterized by negative values. Samples collected in the wet season had lower d34S compared to those collected in the dry season, suggesting the importance of seasonal changes. Oxidation of sulfides, anthropogenic inputs and atmospheric deposition are identified as major end-members and account for 85  7%, 8  7% and 7  0%, respectively, for the entire Kaoping basin in the wet season, which accounts for 90% of the river discharge. In the dry season, the influence of anthropogenic sources is


relatively higher, most likely due to less intensity of chemical weathering in the dry season. [20] The most significant result of this study is that the sulfide oxidation accounts for the delivery of 7.8  0.6  1 from the Kaoping basin. This corre109 mol of SO2 4 y sponds to 1.2–1.6% of the estimated pyrite-derived SO2 4 contributed by the global rivers, from a basin of only 0.003% of the global drainage area. These estimates, in conjunction with the study on Mackenzie River basin [Calmels et al., 2007] and others listed in Table S2 in the auxiliary material underscores the need for more field based studies to directly constrain the sulfide derived SO2 4 component in rivers. Higher rainfall and relief, and active tectonics all appear to favor higher physical weathering in the Kaoping basin, which in turn, exposes fresh (sulfide) mineral surfaces to chemically react with atmospheric oxygen and water. This coupling between physical and chemical weathering sustains high sulfide oxidation in the Kaoping basin and possibly in many regions of the Taiwan and the world. [21] Acknowledgments. We acknowledge the support of National Science Council, Taiwan, for supporting the project (NSC100-2119-M006-018 to YCF) and for the postdoctoral fellowships of AD and CHC. We thank S. Krishnaswami for improving the quality of presentation of the revised version. Finally, we would like to thank Tim Lyons, W. Knorr and an anonymous reviewer for their comments on the paper. [22] The Editor thanks Timothy Lyons and an anonymous reviewer for assisting in the evaluation of this paper.

References Berner, E. K., and R. A. Berner (1996), Global Environment: Water, Air and Geochemical Cycles, 376 pp., Prentice Hall, Upper Saddle River, N. J. Brenot, A., J. Carignan, C. France-Lanord, and M. Benoît (2007), Geological and land use control on d 34S and d 18O of river dissolved sulfate: The Moselle river basin, France, Chem. Geol., 244, 25–41, doi:10.1016/j. chemgeo.2007.06.003. Calmels, D., J. Gaillardet, A. Brenot, and C. France-Lanord (2007), Sustained sulfide oxidation by physical erosion processes in the Mackenzie River basin: Climatic perspectives, Geology, 35, 1003–1006, doi:10.1130/ G24132A.1. Calmels, D., A. Galy, N. Hovius, M. Bickle, A. J. West, M. C. Chen, and H. Chapman (2011), Contribution of deep groundwater to the weathering budget in a rapidly eroding mountain belt, Taiwan, Earth Planet. Sci. Lett., 303, 48–58, doi:10.1016/j.epsl.2010.12.032. Chen, Y. G., J. C. L. Liu, Y. N. Shieh, and T. K. Liu (2004), Late Pleistocene to Holocene environmental changes as recorded in the sulfur geochemistry of coastal plain sediments, southwestern Taiwan, J. Asian Earth Sci., 24, 213–224, doi:10.1016/j.jseaes.2003.10.004. Cheng, M. C., and C. F. You (2010), Sources of major ions and heavy metals in rainwater associated with typhoon events in southwestern Taiwan, J. Geochem. Explor., 105, 106–116, doi:10.1016/j.gexplo.2010.04.010. Chu, H. Y., and C. F. You (2007), Dissolved constituents and Sr isotopes on river waters from a mountainous island—The Danshuei drainage system in northern Taiwan, Appl. Geochem., 22, 1701–1714, doi:10.1016/j. apgeochem.2007.03.030. Chung, C. H., C. F. You, and H. Y. Chu (2009), Weathering sources in the Gaoping (Kaoping) river catchments, southwestern Taiwan: Insights from major elements, Sr isotopes, and rare earth elements, J. Mar. Syst., 76, 433–443, doi:10.1016/j.jmarsys.2007.09.013. Claypool, G. E., W. T. Holser, I. R. Kaplan, H. Sakai, and I. Zak (1980), The age curves of sulfur and oxygen isotopes in marine sulfate and their mutual interpretation, Chem. Geol., 28, 199–260, doi:10.1016/00092541(80)90047-9. Craddock, P. R., O. J. Rouxel, L. A. Ball, and W. Bach (2008), Sulfur isotope measurement of sulfate by high-resolution MC-ICPMS, Chem. Geol., 253, 102–113, doi:10.1016/j.chemgeo.2008.04.017. Dadson, S. J., et al. (2003), Links between erosion, runoff variability and seismicity in the Taiwan orogen, Nature, 426, 648–651, doi:10.1038/ nature02150. Das, A., N. J. Pawar, and J. Veizer (2011), Sources of sulfur in Deccan Trap rivers: A reconnaissance isotope study, Appl. Geochem., 26, 301–307, doi:10.1016/j.apgeochem.2010.12.003.

5 of 6



Ezoe, Y., C. H. Lin, M. Noto, Y. Watanabe, and K. Yoshimura (2002), Evolution of water chemistry in natural acidic environments in Yangmingshan, Taiwan, J. Environ. Monit., 4, 533–540, doi:10.1039/ b201499h. François, L. M., and J. C. G. Walker (1992), Modeling of the Phanerozoic carbon cycle and climate: Constraints from the 87Sr/86Sr isotopic ratio of the seawater, Am. J. Sci., 292, 81–135, doi:10.2475/ajs.292.2.81. Gao, S., T. C. Luo, B. R. Zhang, H. F. Zhang, Y. W. Han, Z. D. Zhao, and Y. K. Hu (1998), Chemical composition of the continental crust as revealed by studies in East China, Geochim. Cosmochim. Acta, 62, 1959–1975, doi:10.1016/S0016-7037(98)00121-5. Ho, C. S. (1975), An Introduction to Geology of Taiwan, Minist. of Econ. Aff., Taipei. Horng, C. S., and A. P. Roberts (2006), Authigenic or detrital origin of pyrrhotite in sediments? Resolving a paleomagnetic conundrum, Earth Planet. Sci. Lett., 241, 750–762, doi:10.1016/j.epsl.2005.11.008. Horng, C. S., J. C. Chen, and T. Q. Lee (1992), Variations in magnetic minerals from two Plio–Pleistocene marine-deposited sections, southwestern Taiwan, J. Geol. Soc. China, 35, 323–335. Kao, S. J., C. S. Horng, A. P. Roberts, and K. K. Liu (2004), Carbon–sulfur– iron relationships in sedimentary rocks from southwestern Taiwan: Influence of geochemical environment on greigite and pyrrhotite formation, Chem. Geol., 203, 153–168, doi:10.1016/j.chemgeo.2003.09.007. Karim, A., and J. Veizer (2000), Weathering processes in the Indus River Basin: Implications from riverine carbon, sulfur, oxygen, and strontium isotopes, Chem. Geol., 170, 153–177, doi:10.1016/S00092541(99)00246-6. Kozdon, R., N. T. Kita, J. M. Huberty, J. H. Fournelle, C. A. Johnson, and J. W. Valley (2010), In situ sulfur isotope analysis of sulfide minerals by SIMS: Precision and accuracy, with application to thermometry of 3.5 Ga Pilbara cherts, Chem. Geol., 275, 243–253, doi:10.1016/j. chemgeo.2010.05.015. Li, X. D., C. Q. Liu, X. L. Liu, and L. R. Bao (2011), Identification of dissolved sulfate sources and the role of sulfuric acid in carbonate


weathering using dual-isotopic data from the Jialing River, southwest China, J. Asian Earth Sci., 42, 370–380, doi:10.1016/j.jseaes.2011. 06.002. Ning, S. K., N. B. Chang, L. Yang, H. W. Chen, and H. Y. Hsu (2001), Assessing pollution prevention program by QUAL2E simulation analysis for the Kao-Ping River Basin, Taiwan, J. Environ. Manage., 61, 61–76, doi:10.1006/jema.2000.0397. Ning, S. K., N. B. Chang, K. Y. Jeng, and Y. H. Tseng (2006), Soil erosion and non-point source pollution impacts assessment with the aid of multitemporal remote sensing images, J. Environ. Manage., 79, 88–101, doi:10.1016/j.jenvman.2005.05.019. Soler, A., A. Canals, S. L. Goldstein, N. Otero, N. Antich, and J. Spangenberg (2002), Sulfur and strontium isotope composition of the Llobregat River (NE Spain): Tracers of natural and anthropogenic chemicals in stream waters, Water Air Soil Pollut., 136, 207–224, doi:10.1023/ A:1015231810548. Spence, J., and K. Telmer (2005), The role of sulfur in chemical weathering and atmospheric CO2 fluxes: Evidence from major ions, d13CDIC, and d 34SSO4 in rivers of the Canadian Cordillera, Geochim. Cosmochim. Acta, 69, 5441–5458, doi:10.1016/j.gca.2005.07.011. Szynkiewicz, A., J. C. Witcher, M. Modelska, D. M. Borrok, and L. M. Pratt (2011), Anthropogenic sulfate loads in the Rio Grande, New Mexico (USA), Chem. Geol., 283, 194–209, doi:10.1016/j. chemgeo.2011.01.017. Thode, H. G. (1991), Sulfur isotopes in nature and the environment: An overview, in Stable Isotopes: Natural and Anthropogenic Sulphur in the Environment, SCOPE Rep., vol. 43, edited by H. R. Krouse and V. A. Grinenko, pp. 1–26, John Wiley, Chichester, U. K. Victoria, L., N. Otero, A. Soler, and A. Canals (2004), Fertilizer characterization: Isotopic data (N, S, O, C, and Sr), Environ. Sci. Technol., 38, 3254–3262, doi:10.1021/es0348187. Wang, R. M., C. F. You, H. Y. Chu, and J. J. Hung (2009), Seasonal variability of dissolved major and trace elements in the Gaoping (Kaoping) River Estuary, southwestern Taiwan, J. Mar. Syst., 76, 444–456, doi:10.1016/j.jmarsys.2007.11.012.

6 of 6

Suggest Documents