Transport of the South China Sea subsurface water outflow and its influence on carbon chemistry of Kuroshio waters off southeastern Taiwan

(1)

Transport of the South China Sea subsurface water outflow

and its influence on carbon chemistry of Kuroshio waters

off southeastern Taiwan

Wen-Chen Chou,1,2David D. Sheu,3C. T. Arthur Chen,3Liang-Saw Wen,4Yih Yang,2 and Ching-Ling Wei4

Received 3 January 2007; revised 25 July 2007; accepted 5 September 2007; published 18 December 2007.

[1]

Depth distributions of pH, dissolved oxygen, dissolved inorganic carbon (DIC), total

alkalinity (TA), and

d

13

C

DIC

in the water column across the Luzon Strait from the

South China Sea to the west Philippine Sea were investigated thoroughly to attest whether

the South China Sea subsurface water outflow could act like a ‘‘shelf pump’’ to export

the carbon from the interior of the South China Sea into the open Pacific. Results

show that the outflow is capable of transporting 17.6 ± 9.0 Tg C a

1

in DIC form out from

the South China Sea to the western Pacific, a quantity equivalent to

35 ± 18% of the

annual export production of the entire South China Sea. Furthermore, owing to the

input of this South China Sea outflow, the subsurface waters of the Kuroshio Current

become enriched in DIC/TA ratio but depleted in

d

13

C

DIC

. Such a change in seawater

carbon chemistry might further attenuate the capacity of CO

2

sequestration and

hamper the use of

d

13

C

DIC

data as a tracer to estimate anthropogenic CO

2

uptake rate in

seawaters around the Kuroshio main path. More importantly, since these modifications

can make all their ways northward along with the Kuroshio Current, the effect may

reach even as far as to the higher-latitude region in the northwestern Pacific.

Citation: Chou, W.-C., D. D. Sheu, C. T. A. Chen, L.-S. Wen, Y. Yang, and C.-L. Wei (2007), Transport of the South China Sea subsurface water outflow and its influence on carbon chemistry of Kuroshio waters off southeastern Taiwan, J. Geophys. Res., 112, C12008, doi:10.1029/2007JC004087.

1. Introduction

[2] Atmospheric CO2 levels have increased from

approximately 280 ppmv in the preindustrial era to 377 ppmv in 2004 because of human activities such as the burning of fossil fuels, land-use changes, and cement production [Keeling and Whorf, 2005]. The issue is of great concern because carbon dioxide is the most important greenhouse gas after water vapor in the atmosphere, and its increase may significantly alter the Earth’s climate system. Since more than 98% of the carbon of the atmo-sphere-ocean system is stored in the oceans in the form of dissolved inorganic carbon (DIC), any change in the DIC content of the ocean may exert a corresponding change in the CO2concentration in the atmosphere [Siegenthaler and

Sarmiento, 1993].

[3] In spite of their comparatively small share in world ocean surface, marginal seas play a disproportionately important role in the global oceanic carbon cycle [Walsh et al., 1981; Rabouille et al., 2001; Chen, 2004; Muller-Karger et al., 2005; Cai et al., 2006]. This is because marginal seas usually are regions receiving a large supply of nutrients from upwelling and river runoff. The high-nutrient input would render a marginal sea highly productive and help elevate the CO2 sequestration via exporting the

photosynthetically fixed carbon downward into subsurface water layers, i.e., the biological pump. A better understand-ing of the fate of the carbon fixed in the marginal seas is therefore crucial to an adequate assessment of the strength and capacity of the oceans in absorbing anthropogenic CO2

and to help set constraints on the future climate change [Walsh, 1991; Chen et al., 2003]. Recent studies in the East China Sea [Tsunogai et al., 1999] and the North Sea [Bozec et al., 2005] have shown that the higher biological produc-tion together with the outflow of the CO2-enriched

subsur-face waters from marginal seas could constitute an important mechanism to transport carbon to the interior of the open oceans, i.e., the so-called ‘‘continental shelf pump’’ [Thomas et al., 2004].

[4] The South China Sea is the largest semienclosed marginal sea in the western Pacific encompassed by the densely populated Asian continent. It receives a voluminous amount of nutrients via river discharge from several large

Full Article

1

Institute of Marine Environmental Chemistry and Ecology, National Taiwan Ocean University, Keelung, Taiwan.

2

National Center for Ocean Research, National Taiwan University, Taipei, Taiwan.

3

Institute of Marine Geology and Chemistry, National Sun Yat-Sen University, Kaohsiung, Taiwan.

4

Institute of Oceanography, National Taiwan University, Taipei, Taiwan.

Kuroshio waters off southeastern Taiwan;

(2)

rivers (e.g., the Mekong River and the Pearl River). It is also the region known for intensive upwelling [Chao et al., 1996] and episodic mixing events induced by typhoons [Lin et al., 2003] and internal waves [Liu et al., 2006], through which nutrient-laden deeper waters are brought upward to the surface leading to a high rate of biological productivity. For instance, Liu et al. [2002] and Chen and Chen [2006] reported the chlorophyll levels in surface waters of the South China Sea are twice as high as those in the adjacent western North Pacific at the same latitude.

[5] The Luzon Strait, situated between Taiwan and Luzon, is the only deep water channel with a maximal sill depth of2200 m, allowing effective exchange of waters between the South China Sea and the western Pacific. During the past decade, enormous research efforts have been made to study the water exchange between the South China Sea and the west Philippine Sea on the basis of hydrographic data [Shaw, 1991; Gong et al., 1992; Qu et al., 2000; Chen, 2005]. Among these, Wyrtki [1961] first noted that surface water of the west Philippine Sea intruded into the South China Sea in winter and flowed back to the Pacific in summer in response to the seasonally reversing monsoon. Later observations, however, showed that the intrusion of the Pacific water could also occur in summer when the southwest monsoon prevailed [Shaw, 1991; Qu et al., 2000]. Wyrtki [1961] also speculated that waters below the surface, ranging from 400 to 900 m in summer and from 300 to 900 m in winter, flowed in the opposite direction to the overlaying surface water, i.e., an inflow in summer and an outflow in winter. Despite considerable seasonal varia-tion, more recently, Chen and Huang [1996] and Qu et al. [2000] found that the water at intermediate depth essentially flowed out of the South China Sea through Luzon Strait to compensate for the inflow of Pacific water in the surface layer. Since the fact that deep water of the South China Sea (below 2000 m) is relatively homogenous and reveals nearly the same properties as the Pacific water at the sill depth of the Luzon Strait, the phenomenon has been interpreted as the result of ventilation from the Pacific water into the South China Sea at the sill depth of the Luzon Strait driving by a persistent baroclinic pressure gradient between the two seas [Qu et al., 2006]. As the Pacific water is colder and denser, it submerges to the bottom of the South China Sea basin after passing through the Luzon Strait. In order to compensate the descending movement and maintain the mass balance, upwelling must have to occur in the interior of the South China Sea. As a result, the renewal of the deep water in the South China Sea is generally believed to be relatively rapid with an estimated residence time ranging from 30 to 150 years (a) by means of various methods [Broecker et al., 1986; Gong et al., 1992; Chen et al., 2001; Qu et al., 2006].

[6] More recently, results from numerical modeling stud-ies further confined the vertically ‘‘sandwich-like’’ structure of water exchanges across the Luzon Strait [Chao et al., 1996; Metzger and Hurlburt, 1996; Qu et al., 2004], i.e., an inflow from the Pacific in the upper and deeper layers but an outflow from the South China Sea in the intermediate layer. Such a unique flow pattern therefore provides a rare opportunity to study the fate and pathway of carbon in the South China Sea. The main purpose of this paper is to attest to whether the South China Sea subsurface water outflow

behaves like a conduit, in the sense of a ‘‘shelf pump,’’ to export the carbon that was once fixed in the interior of the South China Sea to the open Pacific.

[7] Furthermore, the Kuroshio Current, the known west boundary current in the North Pacific, is generally believed to originate from east of Luzon as a continuation of the North Equatorial Current [Nitani, 1972]. En route of its northward journey along the east coast of Taiwan, the Kuroshio considerably changes its water property via the water exchange between the west Philippine Sea and the South China Sea as it bypasses the Luzon Strait [Gong et al., 1992; Qu, 2002]. As a result, the South China Sea subsurface outflow would not only modify the carbon chemistry of the Kusoshio waters but also could transmit these modifications all ways along the northwardly flowing Kuroshio Current to the higher-latitude region in the west-ern North Pacific.

2. Materials and Methods

[8] Figure 1 shows the bathymetric map with sampling locations in this study. A total of 12 hydrographic stations were investigated aboard R/V Ocean Researcher I and Ocean Researcher III across the Luzon Strait during the cruises in July 2002 (stations K2, K3, K4, K5, K7, K8, K9, and K10), August 2003 (SEATS station), and October 2005 (stations G, H, and S5). All water samples were collected at various depths throughout the water column with 20 L Go-Flo and Niskin bottles mounted to a Rosette sampling assembly. Except for pH and dissolved oxygen (DO) measurements, discrete water samples for titration alkalinity (TA), dis-solved inorganic carbon (DIC), andd13CDICanalyses were

transferred into 350 mL BOD bottles, added immediately with 200mL HgCl2-saturated solution to prevent biological

alteration.

[9] Measurements of DIC and TA followed the standard procedures described by the U.S. Department of Energy [1994] and had been used for analyzing waters in the region [Chou et al., 2005, 2006, 2007]. The coulometric method was used for DIC measurements with a precision of ±0.1%. TA was determined, only for the water samples collected during the cruises in August 2003 and October 2005, by the potentiometric titration method in an open cell with a precision of ±0.2%. For the cruise in July 2002, TA was calculated on the basis of DIC and pH data using the program of Lewis and Wallace [1998]. Seawater references provided by A. G. Dickson at the Scripps Institution of Oceanography were used for calibration and accuracy assessments. Differences between the certified values and our measurements are less than 2 and 3mmol kg1for DIC and TA, respectively. The pH was determined onboard at 25 ± 0.1°C by a Radiometer PHM-85 pH meter using a pHC2401-7 combination electrode within 30 min after collection. Tris and 2-aminopyridine buffers were used to calibrate the electrode [Millero et al., 1993]. The precision is better than ±0.005 pH units. DO was measured by a spec-trophotometer with a flow injection analyzer [Pai et al., 1993] with a precision ±0.5mmol kg1. Apparent oxygen utilization (AOU) was calculated using Benson and Krause’s [1984] formula.

[10] For d13CDICanalysis, 40 mL of seawater was

acid-ified with 2 mL of 85% phosphoric acid in a preevacuated

C12008 CHOU ET AL.: INFLUENCE OF THE SCS SSW OUTFLOW C12008

Chou, W.C., D.D. Sheu, C.T.A. Chen, L.S. Wen, Y. Yang and C.L. Wei;

Transport of the South China Sea subsurface water outflow and its influence on carbon chemistry of Kuroshio waters off southeastern Taiwan; Journal of Geophysical Research, Vol. 112, C12008, doi:10.1029/2007JC004087, 2007. (SCI: 2.800)

(3)

vessel [Sheu et al., 1996; Chou et al., 2007]. The liberated CO2 was trapped using liquid nitrogen after complete

removal of water by a slurry dry ice-alcohol mixture under high vacuum. Isotopic analysis was performed with a VG Optima mass spectrometer. Results of isotopic mea-surement are expressed with the conventional d notation. The overall procedural error ford13CDICanalyses is better

than ±0.07%.

3. Results and Discussion

3.1. Depth Distributions of pH, AOU, DIC, and d13_C

DICand Implications for the Water Exchanges

Across the Luzon Strait

[11] The water exchange between the South China Sea and west Philippine Sea waters was first recognized by Wyrtki [1961] and has subsequently become a subject of intensive study. Among these, Gong et al. [1992] were the first to point out that the subsurface waters (100 – 600 m) of the South China Sea were characteristically enriched in nutrient and depleted in DO as compared to the adjacent west Philippine Sea waters at the same depth. More recently, Lin et al. [1999] reported thatd13CDICof the subsurface layer in

the inner part of the South China Sea is more depleted than that in the Luzon Strait. These authors further suggested a potential use of these measurements as a tool to trace the water exchange between the South China Sea and the west Philippine Sea through the Luzon Strait in the subsurface layer. Here we apply additional carbon chemistry data measured throughout the water column to further examine the water exchange between the South China Sea and the west Philippine Sea.

[12] Figure 2 shows the depth profiles of pH, AOU, DIC, and d13CDIC at each station. As shown, AOU and DIC

increase gradually, whereas pH and d13CDICdecrease with

increasing depths. On close examination a striking feature, however, is readily seen in the subsurface layer (100–600 m), where the west Philippine Sea waters (stations east of 122.5°E including K4, K5, K7, and K8; open symbols in Figure 2) are higher in pH andd13CDIC, but

lower in AOU and DIC than seawaters in the South China Sea basin (SEATS site and station G; solid symbols in Figure 2) at the same depth. Waters from stations off southeast Taiwan (K2, K3, K9, K10, H, and S5; dotted grey symbols in Figure 2) have intermediate values, Figure 1. Bathymetric map showing the sampling locations. Superimposed are the composite shipboard

acoustic Doppler current profiler current velocity vectors at 50 m depth between 1990 and 2005. The main flow of the Kuroshio Current is readily seen by the strongest longitudinal flow velocity along approximately 122°E.

(4)

suggesting that a vigorous mixing between the South China Sea and the west Philippine Sea waters must have taken place and the South China Sea subsurface water outflow have modified considerably the intrinsic chemical character-istics of the Kuroshio waters during the mixing.

[13] Another conspicuous feature of these chemical parameters that can be used to help delineate the extent of water exchange between the South China Sea and west Philippine Sea waters is the depth locations of pH, AOU, andd13_C

DICextremes in the water columns at each station.

Figure 3 plots pH, AOU, and d13CDIC distributions in an

east-west cross section along 22°N (K7 ! K8 ! K3 ! K9 ! K10). As seen, the pH and d13_C

DICminima as well as the

AOU maximum that are commonly observed at depths of 1000 m in the open oceans are found only at the stations east of 122°E (K3, K7, and K8) but disappear at other stations (K9 and K10). Moreover, depth profiles of pH and AOU at stations along the Kuroshio main path (K5! H ! K9! K2) reveal that both extremes only appear noticeably at the K5 station (Figure 4). The lack of these chemical extremes at SEATS and G stations in the northeastern South China Sea is generally believed to result from intensive upwelling and vertical mixing in the interior of the South China Sea [Chen and Wang, 1998; Chen et al., 2001]. The absence of these chemical extremes at stations H, K9, K10, and K2 therefore further implies the mixing of the South China Sea with the west Philippine Sea waters and this water exchange can be traced readily at an even deeper depth range, i.e., where the chemical extremes exist (1000 m). Our carbon data therefore are in good accord with previous observations, which ubiquitously reported an outflow of the South China seawaters in the intermediate layer [Chen and Wang, 1998; Qu et al., 2000; Chen, 2005; Li and Qu, 2006].

[14] Although the chemical characteristics of the South China Sea waters are very different from those of the west Philippine Sea waters in the subsurface and intermediate layers, they are remarkably similar in the deep waters (below 1500 m). As shown in Figure 2c, the difference of DIC concentrations between the South China Sea deep water and its precursor (the west Philippine Sea water at the depth range between 2000 and 2500 m) is nearly within the range of analytical uncertainty (±2 mmol kg1). This homogeneity in DIC suggests that the renewal of the South China Sea deep water must be rather rapid; if not, the DIC concentrations in the South China Sea deep water should appear to be higher because of the continuing decomposi-tion of organic matters and dissoludecomposi-tion of carbonate in the interior of the South China Sea. Our result thus supports the general contention that the residence time of the South China Sea deep water is quite short in the range of several decades.

3.2. Does the Subsurface Outflow Export Carbon From the South China Sea Into the West Philippine Sea?

[15] As shown in Figure 2c, DIC concentrations of the South China Sea waters (stations SEATS and G) are higher than those of the west Philippine Sea waters (stations K4, K5, K7, and K8) in the subsurface layer. The enrichment, however, cannot be attributed to the higher degradation of organic carbon accumulated in the subsurface waters alone, because South China Sea is known to have a basin-wide upwelling [Chao et al., 1996] so that a fraction of the high DIC measured must have originated from the entrainment of DIC-rich waters from deep water. To better illustrate the presence of this deep water input, we have plotted temper-ature against depth in Figure 5. As seen, water tempertemper-atures Figure 2. Depth distributions of (a) pH, (b) AOU, (c) DIC, and (d)d13CDICin the water columns from

the west Philippine Sea (WPS), the South China Sea (SCS), and the region off southeast Taiwan (Off SE Taiwan).

(5)

in the South China Sea are colder than those of the west Philippine Sea at the same depths between100 and 600 m. Since the South China Sea is located in the tropical to subtropical region, no cold water sources other than those upwelled from the deeper waters could give to the observed cold temperatures. The presence of colder waters in the subsurface layer in the South China Sea thus indicates that upward advection must have taken place and the process is capable of bringing the DIC-rich deeper waters to shallower depths in the South China Sea. Other evidences of

entrain-ment of cold waters from deep water have been reported by Gong et al. [1992] and Chen et al. [2006].

[16] To separate such a cold water effect from organic carbon decomposition, we have replotted AOU, DIC, and d13CDIC against potential temperature in Figure 6. As

shown, AOU and DIC remain higher and d13CDIC is still

lower in the subsurface layer (corresponding to the temper-ature range from8 to 20°C) in the South China Sea than in the west Philippine Sea at the same temperature. The higher AOU and DIC at the iso-q layers in the South China Sea can be best interpreted as a result from the remineral-ization of organic matters because it will consume DO but release DIC and thus increase AOU and DIC in the water column. Furthermore, the lower d13CDIC observed at the

same temperature layers provides an additional evidence to support such a contention. During photosynthesis, organ-isms preferentially take up the lighter isotope of carbon (12C). The isotopic fractionation of this process is known to be as large as 19 % for marine phytoplankton [Sharp, 2007]. As the isotopically light organic material settles out from the surface water and decomposes, it adds more13 C-depleted CO2into the DIC pool and gives rise to the lower

d13CDICin the subsurface layer.

[17] As shown in the T-S diagram (Figure 7), a pro-nounced maximum salinity core of34.8–34.9 exists at a density (dq) range of 22.5–23.5, whereas a distinct

minimum salinity core of 34.2–34.3 occurs at a density (dq) range of26.5–27.0 at the stations located in the west

Philippine Sea (K4, K5, K7, and K8). These two salinity extremes represent two well-defined water masses in the North Pacific: the high-salinity North Pacific Tropical Water (NPTW), which originates from the pool of high-salinity water found near the international dateline at 20 – 30°N [Suga et al., 2000], and the low-salinity North Pacific Intermediate Water (NPIW), which can be traced to its source region in the subpolar regions in the North Pacific [You, 2003]. As also seen, waters of both salinity maximum and minimum can still be found at the stations in the South China Sea (G and SEATS), despite the fact that the salinity extremes become less pronounced, i.e., salinity maximum becomes lower (34.8–34.9 in the west Philippine Sea versus34.6 at a density level (dq) of25.0 in the South

China Sea), but salinity minimum becomes higher (34.2– 34.3 in the west Philippine Sea versus 34.4 at a density level (dq) of26.5–27.0 in the South China Sea). These

results indicate that the subsurface and intermediate waters in the South China Sea originate principally from the west Philippine Sea through the deep sill at the Luzon Strait; if not, there would be no salinity minimum nor maximum in the South China Sea. In addition, intensive upwelling and vertical mixing in the South China Sea interior must have taken place after the entrance of the west Philippine seawaters to reduce the extreme signals thus smoothing out both salinity extremes.

[18] As evident from the T-S diagram, the South China seawaters below the surface layer originate mainly from the west Philippine Sea and subsequently experience intensive upwelling and vertical mixing in the interior of the South China Sea. In order to further quantify the amount of DIC accumulated in the subsurface water that are derived from the biological production in the interior of the South China Sea, we first construct the relationship between DIC and Figure 3. East-west cross sections of (a) pH, (b) AOU,

(6)

potential temperature for the west Philippine Sea waters between 100 and 1000 m (Figure 8). We then substitute the measured temperature of the South China Sea subsurface waters into the above relationship to derive their preformed values of DIC (DICpre) by means of which the physical effects on the DIC-enrichment, e.g., upwelling and vertical mixing occurring in the interior of the South China Sea, can largely be eliminated and thus can better represent the initial DIC value entering the South China Sea. As a result, the difference between the measured (DICmeas) and preformed values, denoted as DICbio, signifies the amount of carbon produced via the ‘‘biological pump’’ in the South China Sea. As shown in Figure 9, DICbiodecreases gradually with increasing depth in the subsurface layer, and the average concentration of DICbiois estimated to be 24.6 ± 11.5mmol kg1 for the South China Sea subsurface waters between 100 and 600 m. Using a mass balance calculation of water mass and salt budgets, Chen et al. [2001] reported an outflow of 1.9 ± 0.4 Sv for the South China Sea subsurface waters between 350 and 1350 m. Since this is the only flux value ever reported in the literature [Chen et al., 2001], we took this value as a representative for the net flux of water between 100 and 600 m. Accordingly, the South China Sea subsurface outflow between 100 and 600 m could transport 17.6 ± 9.0 Tg C a1(1 Tg C = 1012g carbon) in the DIC form out from the South China Sea into the west Philippine Sea, providing a net water flux of1.9 ± 0.4 Sv and an average DICbioconcentration of 24.6 ± 11.5mmol kg1.

[19] The annual export production of the entire South China Sea was estimated by Liu et al. [2002] and Chen et al. [2006] to range from42.9 to 54.3 Tg C a1. Taking the average (49.8 ± 6.1 Tg C a1) of their values to represent the export production of the South China Sea, the South China Sea subsurface outflow thus is able to transport 35 ± 18% of export production to the west Philippine Sea in DIC form. Moreover, it has been well known that the carbon fixed in the overlying waters is not entirely remin-eralized to DIC; a fraction of the exported carbon out of the

euphotic zone must also be transformed to dissolved organic carbon (DOC) in the subsurface and intermediate waters. Chen et al. [2006] estimated that21.8 Tg C a1of DOC in the South China Sea was exported to the west Philippine Sea by the South China Sea intermediate waters. As a whole, the outflow of South China Sea subsurface and Figure 4. Depth profiles of (a) pH and (b) AOU for the stations along the main path of the Kuroshio

Current.

Figure 5. Comparison of depth profiles of potential temperature for stations in the west Philippine Sea (WPS) with those in the South China Sea (SCS).

(7)

intermediate waters together may transport as much as79 ± 18% of export production of the South China Sea to the west Philippine Sea in the forms of DIC and DOC. Although considerable uncertainty of the above estimate remains, owing mostly to the error in the estimate of the water fluxes, these results demonstrate that the water exchange circulation through the Luzon Strait plays an important role in transporting the carbon from the interior of the South China Sea to the open Pacific and therefore can substantially alter the carbon chemistry of Kuroshio waters off southeastern Taiwan.

3.3. Potential Influence of the South China Sea Subsurface Water Outflow on the Carbon Cycle in the Northwestern Pacific

[20] In terms of the potential for CO2 sequestration the

effect of input of the South China Sea subsurface water outflow is the corresponding change in DIC/TA ratios of the Kuroshio waters. Higher-primary production in the South China Sea leads to more DIC accumulated in the subsurface (100 – 600 m) waters, yet TA remains essentially unchanged because waters are oversaturated with respect to aragonite and calcite within this depth range [Chen et al., 2006; Chou et al., 2007]. Thus the subsurface water of the South China Sea is expected to have a higher-DIC/TA ratio. The DIC/TA ratios, indeed as shown in Figure 10a, are invariably found to be higher at SEATS site and station G in the South China Sea basin than those at stations K4, K5, K7, and K8 in the west Philippine Sea, whereas waters at the stations off southeast Taiwan (K2, K3, K9, K10, H, and S5) have intermediate DIC/TA values.

[21] The change in the DIC/TA ratio is crucial to the oceanic carbon cycle because it regulates the chemical potential of a given water mass in sequestrating atmospheric CO2, namely the ‘‘Revelle factor.’’ In principle the capacity

Figure 6. Potential temperature versus (a) AOU, (b) DIC, and (c) d13CDIC plots for the stations located in the west

Philippine Sea (WPS) and the South China Sea (SCS).

Figure 7. Potential temperature versus salinity plots (T-S diagram) for the stations located in the west Philippine Sea (WPS) and the South China Sea (SCS).

(8)

for ocean waters to take up atmospheric CO2 is inversely

proportional to the value of the Revelle factor, which is defined as the ratio of the fractional increase in atmospheric CO2content to the fractional rise in seawater DIC [Revelle

and Suess, 1957]. Because of the carbonate-buffering effect the value of the Revelle factor is proportional to the DIC/TA ratio [Sarmiento and Gruber, 2006]. Consequently, a higher-DIC/TA ratio would lead to a large Revelle factor and hence reduce the capacity in sequestrating atmospheric CO2for a given seawater sample. Figure 10b shows that the

Revelle factor of the South China Sea subsurface waters that are characteristic of an elevated DIC/TA ratio is noticeably higher than that of the west Philippine Sea waters with lower-DIC/TA ratio. By comparison, waters at the stations off southeast Taiwan have intermediate Revelle factor values. Thus the outflow of the subsurface water of South China Sea has significantly altered the carbon chemistry and the potential of CO2 uptake of the Kuroshio subsurface

water.

[22] Along with the northwardly flowing Kuroshio Cur-rent, subsurface waters characterized by these high-DIC/TA ratios and Revelle factors would dispatch to the higher-latitude region in the northwestern Pacific and lead water around the Kuroshio regime to be less effective in uptaking the atmospheric CO2. It is, however, important to note that

the change of the Revelle factor of the Kuroshio subsurface water observed in this study suggests only the alteration of its chemical potential. The actual influence on CO2

seques-tration is still subject to other processes such as water mass upwelling or downwelling. A prominent example can be found in the East China Sea shelves. Chen [1996] reported that the outflow of the nutrient-laden South China Sea subsurface waters after mixing with the west Philippine Sea waters had traveled all the way along the eastern coast of Taiwan to upwell onto the East China Sea. The process would supply an additional nutrient source to foster a high

productivity and enhance the uptake of atmospheric CO2in

the East China Sea shelf waters. On the contrary the upwelling of the subsurface waters with elevated Revelle factors revealed in this study would impose a negative feedback to lower the capacity of the East China Sea waters to take up CO2from the atmosphere.

[23] Another concern of the study of the oceanic carbon cycle is the change ind13CDICof the Kuroshio subsurface

waters. It has been well known that the CO2emitted to the

atmosphere by anthropogenic activities, such as burning of fossil fuels and land use changes, is strongly depleted in the rare isotope (13C) because of the preferential uptake of the light carbon isotope (12C) during photosynthetic utilization of CO2 by plants. As a consequence, the d13C of

atmo-spheric CO2 has decreased from about 6.3 % in 1850

[Friedli et al., 1986] to about8.1 % in 2002 [Keeling et al., 2005]. The penetration of anthropogenic CO2via air-sea

exchange has also decreased thed13C of DIC in seawater, commonly known as the oceanic13C Suess effect [Keeling, 1979]. Observations of the oceanic Suess effect provide a means to estimate the oceanic uptake rate of anthropogenic CO2[Quay et al., 1992, 2003, and references therein]. For

instance, comparing changes ind13CDICbetween the

GEO-SECS era (1970s) and WOCE era (1990s), Quay et al. [2003] reported a global mean depth-integrated anthropogenic change ind13CDICof 65 ± 33% m per decade and then

deduced an oceanic CO2uptake rate of 1.5 ± 0.6 Gt C a1.

Nonetheless, as discussed in section 3.2, the observed deple-tion ofd13CDICin the subsurface water of the South China

Figure 8. Correlation between DIC and potential tem-perature for the stations located in the west Philippine Sea.

Figure 9. Depth profiles of measured (DICmeas) and preformed (DICpre) DIC from 100 to 1000 m for the stations located in the South China Sea. See section 3.2 for the definition and calculations of DICpreand DICbio.

(9)

Sea results mainly from the enhanced remineralization of organic matters. Thesed13CDIC-depleted subsurface waters

then flow out and join the Kuroshio water, leading to the observed d13CDIC decrease in the subsurface layer of the

Kuroshio waters. The sharp decrease in d13CDIC of the

Kuroshio subsurface waters thus cannot be attributed com-pletely to the perturbation of anthropogenic CO2; instead, it

must result partially from the mixing of the South China Sea subsurface waters. This result demonstrates that processes other than anthropogenic perturbations, e.g., the physical processes such as water exchanges between the South China Sea and west Philippine Sea discussed in this study, have the potential to mask the Suess effect on thed13C of DIC in the surface ocean.

4. Concluding Remarks

[24] This study demonstrates that the South China Sea subsurface water outflow could act like a shelf pump to transport photosynthetically fixed carbon out from the South China Sea to the west Philippine Sea, which is an important process for sequestering the atmospheric CO2. It

further shows that this process may significantly modify the carbon chemistry (an increase in DIC/TA ratio accompa-nying a decrease ind13CDIC) of the subsurface water of the

Kuroshio Current in regions off southeast Taiwan. Since these modifications can travel along with the northwardly flowing Kuroshio Current to the northwestern Pacific, they

might deteriorate the capacity of CO2 sequestration of

waters in the higher-latitude region and hamper the estimate of the anthropogenic CO2uptake usingd13CDICdata. More

importantly, the case we have studied may not be unique to the South China Sea but could also be found in other marginal seas, where the process of the so-called continental shelf pump is identified and prevails.

[25] Acknowledgments. We are grateful to the captains, crew, and technicians of R/V Ocean Researcher I and Ocean Researcher III for assistance with shipboard operation and sampling, and to F. S. Li, S. G. Lin, C. W. Huh, and W. P. Hou for laboratory assistance. Constructive comments and suggestions from two anonymous reviewers have greatly improved the manuscript. This work was supported by the National Science Council grants (94-2119-M-110-001, 94-2611-M-110-018, and 93-2119-M-110-001) to D.D. Sheu. This is NCOR publication number 118.

References

Benson, B. B., and D. Krause (1984), The concentration and isotopic fractionation of oxygen dissolved in freshwater in equilibrium with the atmosphere, Limnol. Oceanogr., 29, 620 – 632.

Bozec, Y., H. Thomas, K. Elkalay, and H. J. W. de Baar (2005), The continental shelf pump for CO2in the North Sea-evidence from summer

observation, Mar. Chem., 93, 131 – 147.

Broecker, W. S., W. C. Patzert, J. R. Toggweiler, and M. Stuive (1986), Hydrography, chemistry, and radioisotopes in the southeast Asian basins, J. Geophys. Res., 91, 14,345 – 14,354.

Cai, W.-J., M. Dai, and Y. Wang (2006), Air-sea exchange of carbon diox-ide in ocean margins: A province-based synthesis, Geophys. Res. Lett., 33, L12603, doi:10.1029/2006GL026219.

Chao, S. Y., P. T. Shaw, and S. Y. Wu (1996), Deep water ventilation in the South China Sea, Deep Sea Res. Part I, 43, 445 – 466.

Figure 10. Depth profiles of (a) DIC/TA ratio and (b) Revelle factor in the water columns from surface to 600 m in the west Philippine Sea (WPS), the South China Sea (SCS), and the region off southeast Taiwan (Off SE Taiwan).

(10)

Chen, C.-T. A. (1996), The Kuroshio intermediate water is the major source of nutrients on the East China Sea continental shelf, Oceanol. Acta, 19, 523 – 527.

Chen, C.-T. A. (2004), Exchanges of carbon in the coastal seas, in The Global Carbon Cycle: Integrating Humans, Climate, and the Natural World, edited by C. B. Field and M. R. Raupach, pp. 341 – 351, Isl., Washington D. C.

Chen, C.-T. A. (2005), Tracing tropical and intermediate waters from the South China Sea to the Okinawa Trough and beyond, J. Geophys. Res., 111, C05012, doi:10.1029/2004JC002494.

Chen, C.-T. A., and M. H. Huang (1996), A mid-depth front separating the South China Sea water and the west Philippine Sea water, J. Oceanogr., 52, 17 – 25.

Chen, C.-T. A., and S. L. Wang (1998), Influence of intermediate water in the western Okinawa Trough by the outflow from the South China Sea, J. Geophys. Res., 103, 12,683 – 12,688.

Chen, C.-T. A., S. L. Wang, B. J. Wang, and S. C. Pai (2001), Nutrient budgets for the South China, Mar. Chem., 75, 281 – 300.

Chen, C.-T. A., K.-K. Liu, and R. Macdonald (2003), Continental margin exchanges, in Ocean Biogeochemistry: The Role of the Ocean Carbon Cycle in Global Change, edited by M. J. R. Fasham, pp. 53 – 97, Springer, New York.

Chen, C.-T. A., S. L. Wang, W. C. Chou, and D. D. Sheu (2006), Carbonate chemistry and projected future changes in pH and CaCO3saturation state

of the South China Sea, Mar. Chem., 101, 277 – 305.

Chen, Y. L. L., and H.-Y. Chen (2006), Seasonal dynamics of primary and new production in the northern South China Sea: The significance of river discharge and nutrient advection, Deep Sea Res. Part I, 53, 971 – 986.

Chou, W. C., D. D. Sheu, C.-T. A. Chen, S. L. Wang, and C. M. Tseng (2005), Seasonal variability of carbon chemistry at the SEATS time-series site, northern South China Sea between 2002 and 2003, Terr. Atmos. Oceanic Sci., 16, 445 – 465.

Chou, W. C., Y. L. L. Chen, D. D. Sheu, Y. Y. Shih, C. A. Han, C. L. Cho, Y. J. Yang, and C. M. Tseng (2006), Estimated net community pro-duction during the summer season at the SEATS site: Implications for nitrogen fixation, Geophys. Res. Lett., 33, L22610, doi:10.1029/ 2005GL025365.

Chou, W. C., D. D. Sheu, B. S. Lee, C. M. Tseng, C. T. A. Chen, S. L. Wang, and G. T. F. Wong (2007), Depth distributions of alkalinity, TCO2,

andd13_C

TCO2at SEATS Time-series site in the northern South China Sea,

Deep Sea Res. Part II, 54, 1469 – 1485, doi:10.1016/j.dsr2.2007.05.002. Friedli, H., H. Loetscher, H. Oeschger, U. Siegenthaler, and B. Stauffer

(1986), Ice core record of the13_C/12_{C ratio of atmospheric CO} 2in the

past two centuries, Nature, 324, 237 – 238.

Gong, G. C., K. K. Liu, C. T. Liu, and S. C. Pai (1992), The chemical hydrography of the South China Sea west of Luzon and a comparison with the west Philippine Sea, Terr. Atmos. Oceanic Sci., 3, 587 – 602. Keeling, C. D. (1979), The Suess Effect:13_{Carbon and}14_Carbon

interac-tion, in Environmental International, vol. 2, pp. 229 – 300, Pergamon, Tarrytown, N. Y.

Keeling, C. D., and T. P. Whorf (2005), Atmospheric CO2records from

sites in the SIO air sampling network, in Trends: A Compendium of Data on Global Change, Carbon Dioxide Inf. Anal. Cent., Oak Ridge Natl. Lab., U.S. Dept. of Energy, Oak Ridge, Tenn.

Keeling, C. D., A. F. Bollenbacher, and T. P. Whorf (2005), Monthly atmo-spheric13C/12C isotopic ratios for 10 SIO stations, in Trends: A Com-pendium of Data on Global Change, Carbon Dioxide Inf. Anal. Cent., Oak Ridge Natl. Lab., U.S. Dept. of Energy, Oak Ridge, Tenn. Lewis, E., and D. W. R. Wallace (1998), Program developed for CO2

system calculations, Carbon Dioxide Inf. Anal. Cent. Rep. ORNL/ CDIAC-105, Oak Ridge Natl. Lab., Oak Ridge, Tenn.

Li, L., and T. Qu (2006), Thermohaline circulation in the deep South China Sea basin inferred from oxygen distributions, J. Geophys. Res., 111, C05017, doi:10.1029/2005JC003164.

Lin, H.-L., L. W. Wang, C. H. Wang, and G. C. Gong (1999), Vertical distribution ofd13_{C of dissolved inorganic carbon in the northeastern}

South China Sea, Deep Sea Res. Part I, 46, 757 – 775.

Lin, I., W. T. Liu, C. Wu, G. T. F. Wong, C. Hu, Z. Chen, W. Liang, Y. Yang, and K. Liu (2003), New evidence for enhanced ocean primary production triggered by tropical cyclone, Geophys. Res. Lett., 30(13), 1718, doi:10.1029/2003GL017141.

Liu, C.-T., R. Pinkel, M.-K. Hsu, J. M. Klymak, H.-W. Chen, and C. Villanoy (2006), Nonlinear internal waves from the Luzon Strait, Eos Trans. AGU, 87(42), 449.

Liu, K.-K., S.-Y. Chao, P.-T. Shaw, G.-C. Gong, C.-C. Chen, and T. Y. Tang (2002), Monsoon-forced chlorophyll distribution and primary production

in the South China Sea: Observations and a numerical study, Deep Sea Res. Part I, 49, 1387 – 1412.

Metzger, E. J., and H. E. Hurlburt (1996), Coupled dynamics of the South China Sea, the Sulu Sea, and the Pacific Ocean, J. Geophys. Res., 101, 12,331 – 12,352.

Millero, F. J., J.-Z. Zhang, S. Fiol, S. Sotolongo, R. N. Roy, K. Lee, and S. Mane (1993), The use of buffers to measure the pH of seawater, Mar. Chem., 44, 143 – 152.

Muller-Karger, F. E., V. Ramon, R. Thunell, R. Luerssen, C. Hu, and J. J. Walsh (2005), The importance of continental margins in the global car-bon cycle, Geophys. Res. Lett., 32, L01602, doi:10.1029/2004GL021346. Nitani, H. (1972), Beginning of the Kuroshio, in Physical Aspects of the Japan Current, edited by H. Stommel and K. Yoshida, pp. 129 – 163, Univ. of Wash. Press, Seattle, Wash.

Pai, S. C., G. C. Gong, and K. K. Liu (1993), Determination of dissolved oxygen in seawater by direct spectrophotometer of total iodine, Mar. Chem., 41, 343 – 351.

Qu, T. (2002), Evidence of water exchange between the South China Sea and the Pacific through the Luzon Strait, Acta Oceanol. Sin., 21, 175 – 185.

Qu, T., H. Mitsudera, and T. Yamagata (2000), Intrusion of the North Pacific waters into the South China Sea, J. Geophys. Res., 105, 6415 – 6424. Qu, T., Y. Y. Kim, and M. Yaremchuk (2004), Can Luzon Strait transport

play a role in conveying the impact of ENSO to the South China Sea, J. Clim., 17, 3644 – 3657.

Qu, T., J. B. Girton, and J. A. Whitehead (2006), Deepwater overflow through Luzon Stait, J. Geophys. Res., 111, C01002, doi:10.1029/ 2005JC003139.

Quay, P., B. Tilbrook, and C. S. Wong (1992), Oceanic uptake of fossil fuel CO2: Carbon-13 evidence, Science, 256, 74 – 79.

Quay, P., R. Sonnerup, T. Westby, J. Stutsman, and A. McNichol (2003), Changes in the13_C/12_{C of dissolved inorganic carbon in the ocean as a}

tracer of anthropogenic CO2uptake, Global Biogeochem. Cycles, 17(1),

1004, doi:10.1029/2001GB001817.

Rabouille, C., F. T. Mackenzie, and L. M. Ver (2001), Influence of the human perturbation on carbon, nitrogen, and oxygen biogeochemical cycles in the global coastal ocean, Geochim. Cosmochim. Acta, 65, 3615 – 3641.

Revelle, R., and H. E. Suess (1957), Carbon dioxide exchange between atmosphere and ocean and the question of an increase of atmospheric CO2during the past decades, Tellus, 9, 18 – 27.

Sarmiento, J., and N. Gruber (2006), The ocean carbon cycle, in Ocean Biogeochemical Dynamics, pp. 318 – 358, Princeton Univ. Press, Princeton, N. J.

Sharp, Z. (2007), Carbon in the low-temperature environment, in Principles of Stable Isotope Geochemistry, pp. 149 – 178, Prentice-Hall, Upper Sad-dle River, N. J.

Shaw, P. T. (1991), Seasonal variation of the intrusion of the Philippine seawater into the South China Sea, J. Geophys. Res., 96, 821 – 827. Sheu, D. D., W. Y. Lee, C. H. Wang, C. L. Wei, C. T. A. Chen, C. Chern,

and M. H. Huang (1996), Depth distribution ofd13C of dissolved total CO2in seawater off eastern Taiwan: Effect of the Kuroshio current and its

associated upwelling phenomenon, Cont. Shelf Res., 16, 1609 – 1619. Siegenthaler, U., and J. L. Sarmiento (1993), Atmospheric carbon dioxide

and the ocean, Nature, 365, 119 – 125.

Suga, T., A. Kato, and K. Hanawa (2000), North Pacific Tropical Water: Its climatology and temporal changes associated with the climate regime shift in the 1970s, Prog. Oceanogr., 47, 223 – 256.

Thomas, H., Y. Bozec, K. Elkalay, and H. J. W. de Baar (2004), Enhanced open ocean storage of CO2from shelf sea pumping, Science, 304, 1005 –

1008.

Tsunogai, S., S. Watanabe, and T. Sato (1999), Is there a ‘‘continental shelf pump’’ for the absorption of atmospheric CO2?, Tellus Ser. B, 51, 701 –

712.

U.S. Department of Energy (1994), Handbook of methods for the analysis of the various parameters of the carbon dioxide system in seawater, version 2, Rep. ORNL/CDIAC-74, edited by A. G. Dickson and C. Goyet, Washington, D. C.

Walsh, J. J. (1991), Importance of continental margins in the marine bio-geochemical cycling of carbon and nitrogen, Nature, 350, 53 – 55. Walsh, J. J., G. T. Rowe, R. L. Iverson, and C. P. McRoy (1981), Biological

export of shelf carbon is a sink of the global CO2cycle, Nature, 291,

196 – 201.

Wyrtki, K. (1961), Physical oceanography of the southeast Asian waters: Scientific results of marine investigation of the South China Sea and the Gulf of Thailand, NAGA Rep. V2, 195 pp., Scripps Inst. of Oceanogr., La Jolla, Calif.

(11)

You, Y. (2003), The pathway and circulation of North Pacific Intermediate Water, Geophys. Res. Lett., 30(24), 2291, doi:10.1029/2003GL018561.

C. T. A. Chen and D. D. Sheu (corresponding author), Institute of Marine Geology and Chemistry, National Sun Yat-Sen University, Kaohsiung 804, Taiwan. (ddsheu@mail.nsysu.edu.tw)

W.-C. Chou, Institute of Marine Environmental Chemistry and Ecology, National Taiwan Ocean University, Keelung 202, Taiwan.

L.-S. Wen and C.-L. Wei, Institute of Oceanography, National Taiwan University, Taipei 106, Taiwan.

Y. Yang, National Center for Ocean Research, National Taiwan University, P.O. Box 23-13, Taipei 106, Taiwan.