Evidence for submarine groundwater discharge on the Southwestern shelf of Taiwan

Continental Shelf Research 34 (2012) 18–25

Contents lists available at SciVerse ScienceDirect

Continental Shelf Research journal homepage: www.elsevier.com/locate/csr

Research papers

Evidence for submarine groundwater discharge on the Southwestern shelf of Taiwan P.O. Zavialov a,n, R.-C. Kao b, V.V. Kremenetskiy a, V.I. Peresypkin a, C.-F. Ding b, J.-T. Hsu b, O.V. Kopelevich a, K.A. Korotenko a, Y.-S. Wu b, P. Chen b a b

Shirshov Institute of Oceanology, Russian Academy of Sciences, Moscow, Russia Tainan Hydraulics Laboratory, National Cheng-Kung University, Tainan, Taiwan

a r t i c l e i n f o

a b s t r a c t

Article history: Received 28 March 2011 Received in revised form 9 November 2011 Accepted 24 November 2011 Available online 7 December 2011

This study was aimed at identifying the locations of the submarine groundwater discharge (SGD) on the shelf of Southwestern Taiwan by means of oceanographic measurements, quantifying its influence on the hydrographic conditions in the area, and estimating the volume rates of the discharge. Two high resolution hydrographic surveys of the region, including water and bottom sediment sampling campaigns, were completed in February and October of 2009. Water samples were also collected from the neighboring on-land groundwater wells. At some locations in the study regions, the vertical profiles exhibited a slight but detectable (0.009 to 0.105 psu) decrease of salinity manifested in the near-bottom portion of the water column. Although convectively unstable, this feature appeared robust and persisted for the eight months between the surveys. The salinity anomalies in the near-bottom layer were often accompanied by the maxima of fluorescence, chlorophyll, silica, nitrate, and iron concentrations, as well as the minima of turbidity. In February, 2009, the n-alkane composition of organic matter in the water collected from an on-land groundwater well exhibited high content of C24 alkane. A similar anomalously high concentration of C24 alkane was encountered in the bottom sediment samples from the suspected SGD sites. In October, 2009, the dominant marker of SGD signature was the C16 alkane. Based on these data, we specified the likely locations of the SGD sources in the study area, all of which were restricted to the inner shelf at the depths less than 8 m. We argue that the influence of the SGD on oceanographic regime in the region is small but observable. Its signature is confined to the lowermost 0.1–2.1 m layer of the water column. The groundwater seepage rates roughly estimated under the assumption of the advection–diffusion balance based on the eddy diffusivity values typical for the bottom layer, are of the order of 0.1 to 1 g m  2 s  1. & 2011 Elsevier Ltd. All rights reserved.

Keywords: Bottom layer Submarine groundwater discharge Southwestern shelf of Taiwan

1. Introduction Submarine groundwater discharge (hereinafter SGD) is the least well quantified component of the ocean’s water budget. The annual volume of SGD into the ocean at the global scale is unknown, but is believed to be less than 6% of the fluvial discharges, i.e., below 2500 km3 (Intergovernmental Oceanographic Commission, 2004), and this is why the SGD is considered unimportant in many cases. Studies of the impacts that SGD may have on the regime of the water column are relatively rare in the oceanographic literature, perhaps, because the measurements of SGD are technically rather difficult, although set of methods to determine the magnitude of the

n Corresponding author at: 36, Nakhimovskiy Prospect Ave., Moscow, 117997, Russia. Tel.: þ 7 499 1245994; fax: þ7 499 1245983. E-mail address: [email protected] (P.O. Zavialov).

0278-4343/$ - see front matter & 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.csr.2011.11.010

discharge have been developed. SGD can be measured directly using the equipment called seepage meters, either mechanically or electromagnetically (e.g., Lee, 1977, Zhang and Satake, 2003, Rosenberry and Morin, 2004). However, SGD is often highly inhomogeneous and distributed in a patchy pattern, and, therefore, measurements at one or a few points are not always representative. There is also a variety of geochemical techniques estimating SGD indirectly based on different tracers, such as radium, strontium, or oxygen isotopes, barium, and radon (e.g., Moore, 1996, Burnett et al., 2001, Swarzenski et al., 2001, Lin et al., 2010, Huang et al., 2011). There is strong evidence that the role of SGD can be significant at the regional scales, especially in coastal waters, and at specific locations, where its input to the ocean can be comparable with or even exceed that of the surface runoff (e.g., Moore, 1996, Bokuniewicz and Pavlik, 1990). Moreover, SGD is a potentially significant pathway for the pollutants and nutrients into the coastal areas of the ocean. This is especially so for the regions

P.O. Zavialov et al. / Continental Shelf Research 34 (2012) 18–25

where the ground water is subjected to anthropogenic pressures associated with extensive urban, industrial, or agricultural use of lands (Valiela and D’Elia, 1990, Moore, 1999). The input of pollutants via SGD has been shown to be the cause of euthrophication in the coastal waters in New England, Florida, and other locations. The island of Taiwan is likely to be one of such regions. Groundwater is an essential natural resource for Taiwan’s economy and population. The annual consumption of the groundwater for agricultural, industrial, and household uses is about 7 km3 (Ting, 1997), while the recharge totals only 4 km3 (Cheng et al., 1995). The overexploitation of groundwater and, possibly, climate change resulted in a serious deterioration of groundwater resources in the last decades, and the future projections raise a concern regarding the decrease of available groundwater (Hsu et al., 2007). Overpumping has resulted in land subsidence in some areas, especially on the southwestern side of the island (TPCWB, 1995). Seawater intrusions into the land aquifers have also been reported (Ting, 1997). There is a large number of monitoring wells all over the territory of Taiwan, so the topography of the groundwater tables in the confined and unconfined aquifers, as well as the recharge rates and the water quality indicators, have been assessed in detail. One of the regions where the slope of the table points to likely outflow of groundwater towards the South China Sea is the Pingtung Plain in the southwestern part of the country. According to (Ting, 1997), the groundwater in this region generally moves westward to the Kaoping River, Donggang River, and southwestwards to the sea. The total annual outflow to the sea in this region is estimated from the available land hydrology data to be mere 0.03 km3 (Ting and Overmars, 1995). However, (Lin et al., 2010) and (Huang et al., 2011) concluded, from the analysis of isotopic tracers, that SGD may play an important role in the near-bottom portion of the column in the Kaoping canyon (depths 400–1200 m). They argued that up to 0.6% of water in the canyon may have originated from SGD. The coastal waters of the Pingtung Plain are not only sources of fisheries but also serve for recreational purposes, therefore, it is important to estimate the rates of SGD in the area and evaluate the impact on the sea water column. Until 2004, no attempts were made to quantify SGD in Pingtung shelf based on an observational data, although some indirect evidence pointing on its existence has been reported (e.g., Ting, 1997, Lin et al., 2003). Virtually the only direct measurements of SGD in the inner shelf were carried out in 2004 by Cheng et al. (2005). They deployed SGD collecting devices buried in the bottom sands at 5 sampling sites (Kaohsiung city, Kaoping River mouth, Fangsan, Jinshawan, and Yaniliao townships). The collected samples were analyzed for salinity, pH, and nutrients (nitrate, nitrite, phosphate, and silicate). At 2 of the locations, namely, Xiziwan within Kaohsiung city limits, and Fangsan township, distinct SGD signatures were observed, manifested with significantly reduced salinity and pH, and elevated concentrations of nutrients, as compared with the surrounding ocean waters. The freshening was particularly dramatic at a station named ‘‘Eureca’’ by these researchers, situated 300 m from the Fangshan coast at the depth 7.8 m, where the bottom water sample reportedly had salinity of only 0.2 psu (!), i.e., was essentially fresh. These published observations yielded enlightening results. However, they were restricted to the very inner part of the shelf immediately adjacent to the coast (0–300 m from the shoreline, 0–8 m depth), and focused on the groundwater within the bottom sands and sediments. In the present study, we attempted to detect the SGD signatures in the area by means of oceanographic measurements, and also quantify the SGD influence on the sea water column.

19

2. Data The study area was located on the Pingtung shelf, between the Kaoping River mouth in the north and Fangshan township in the south (Fig. 1). Two field surveys of the area were organized in 2009. In the both cases, small fishing ships were used for the measurements. The first survey (February 14–19, 2009) consisted of 16 stations organized in 4 cross-shore sections in the northern part of the study area, extending approximately from the 5 m to the 50 m isobaths. Since there were no indications of SGD at the deep stations in February, they were not occupied during the second survey (October 23–27, 2009). Instead, the area of the second survey was extended to the south. The station G of the southernmost section coincided with the ‘‘Eureca’’ site as specified by (Cheng et al., 2005). At each station, surface-to-bottom CTD profiling was done. In the February survey, a SBE19plus profiler equipped with complementary fluorescence and turbidity sensors was used. In the October survey, we used the same instrument and, additionally, Idronaut CTD profiler equipped with turbidity sensor. At all stations, the profilers were lowered until the contact with the bottom. At all stations, velocity profiling was also performed using an ADCP instrument. The collected data are not reported in this paper. However, it may be worthy of mentioning as a background information that during the both cruises, the mean current in the lowermost 2 m of the water column was directed northeastward along the shore at 20–30 cm/s, modulated by a semidiurnal tide with the amplitude of about 10 cm s  1. The vertical shear in the near-bottom part of the column was up to 0.1 s  1. The bottom sediment samples were obtained by a bottom grab. The hydrocarbons were extracted using Branson-1210 ultrasonic water bath. The sediments were then analyzed for n-alkane composition at Shirshov Institute of Oceanology, Russian Academy of Sciences (SIO RAS), through chromatography by Shimadzu GC-2010 instrument equipped with Supelco capillary GC column. Silica-gel was used as a filler and hexane was applied as an eluent. The samples were analyzed under isothermal conditions at 300 1C. The nominal accuracy of the instrument is 10  3 mg/g, while the reproducibility of the analysis result is 75%.

Fig. 1. Map of study area and locations of hydrographic stations. The stations indicated by circles were occupied in February, 2009, the stations shown by filled circles were occupied in October, 2009, and those shown by boxes were occupied in the both surveys.

20

P.O. Zavialov et al. / Continental Shelf Research 34 (2012) 18–25

More specific details of the n-alkane analysis procedure can be found in (PPeresypkin et al., 2011). Water samples were collected from the surface and the bottom using a 5-l Niskin bottle. We also collected water samples from a near-shore groundwater well on the land. All water samples were analyzed for a set of chemical indicators at Tainan Hydraulics Laboratory, Taiwan. In this paper, we will refer to the following indicators, which proved to be informative within the context of SGD: concentrations of silica, nitrate, chlorophyll ‘‘a’’, and iron. The tests were carried out in compliance with the American Water Works Association the Water Pollution Control Federation Standard Methods for the Examination of Water and Wastewater (21st edition).

3. Results and discussion 3.1. Hydrographic data The hydrographic profiling revealed the existence of 3 basic types of stratification in the near-bottom layer (Fig. 2). It should be noted that throughout this paper, we focus on the near-bottom layer, so the ordinate axis in Fig. 2, as well as some other figures herein, represents the height above the bottom, and not the depth as in the conventional oceanographic plots. Stratification type 1, characteristic for stations 2–8 in February and stations 2–5, 7–10, 12–15, and C–K in October exhibited salinity maximum at the bottom. This type can be considered ‘‘normal’’ for the oceanic conditions with no SGD influence, where the salinity and the density increase downwards from the surface, which may be freshened by the fluvial discharges. Stratification type 2, observed at locations 9, 10, 12, and 13 in February, and H in October is characterized by fully mixed bottom layer. It points the presence of a strong current near the bottom (also confirmed by direct velocity measurements), leading to enhanced shear-generated turbulent mixing. This type does not hold any indication of SGD either. However, the most interesting stratification type is type 3, observed at stations 1, 6, 11, and 16 in February, and 1, 6, 11, 16,

and G in October, in the coastal part of the study region. It is characterized by a weak but distinct salinity minimum near the bottom (Fig. 2). We hypothetically attribute the salinity drop near the bottom to SGD. This stratification type was observed only at the stations closest to the coast, at depths 4–8 m (but not all of the near-shore stations), and never at greater depth. This agrees well with the notion that the smaller is the depth, the higher is the probability of SGD (e.g., Glover, 1959, Intergovernmental Oceanographic Commission, 2004), although some ‘‘deep’’ SGD sites situated as far as 25 km and more from the coast have been reported (Lin et al., 2010). There were hints of such stratification at station 15 (depth 16 m) in February, but repeated measurements did not confirm it. This type of salinity distribution is clearly convectively unstable; for example, a drop of density by about 0.03 kg/m3 near the bottom compared with that in the overlying layer is evident in Fig. 2c. Nonetheless, this feature appears to be robust. The profiles at every station were taken several times (in October survey—using 2 different instruments), and the salinity drop near the bottom was observed repeatedly. Moreover, the profiles at the corresponding locations maintained remarkably similar patterns between the two surveys repeated with 8 month interval, even though the mean salinity was offset for almost 0.4 psu (Fig. 3). The thickness DH of the near-bottom layer of salinity drop varied between 0.13 m and 2.15 m, and the magnitude DS of the maximum drop ranged from 0.009 to 0.105 psu. This feature at different locations is summarized in Table 1. As mentioned above, the near-bottom layer of slightly lower salinity is convectively unstable, and, therefore, its maintenance requires continuous supply of fresh groundwater from below. A qualitative, crude estimate of the SGD rate can be obtained from the parameters of the near-bottom low salinity layer as follows. Considering the advection–diffusion balance wS ¼ k dS=dz, where w is the vertical velocity of the groundwater seepage, S is the salinity, z is the vertical coordinate, and k is the eddy diffusivity, and also assuming DS5S0, where S0 is the salinity at

Fig. 2. Three basic types of vertical stratification of salinity (circles) and temperature (triangles) observed in the lowermost 5 m of the water column. (a)—Stable (example profile from Station 2, February 2009, depth 40 m); (b)—Neutral (example profile from Station 9, February 2009, depth 24 m); (c)—Unstable (example profile from Station 11, February 2009, depth 8 m). The salinity decrease in the bottom part of the salinity profile in (c) is hypothetically associated with SGD. The corresponding unstable density profile is also shown in (c) by solid curve.

P.O. Zavialov et al. / Continental Shelf Research 34 (2012) 18–25

21

showed no anomaly, and the second one revealed an anomaly similar to that at the station. A higher resolution profiling campaign is needed to map the exact locations and sizes of the SGD ‘‘vents’’, and quantify the fraction of the bottom floor area that they occupy. Only after this is done, the total volume flux of SGD in the region can be confidently estimated. However, to give an idea about the order of magnitude, we note that if we consider the upper and lower bounds of the estimate for Q above, and assume that the SGD occupies 10% of the shelf area in a 1 km wide belt along the shoreline in Pingtung region, we arrive at a value between 0.01 and 0.1 km3 per year, which seems to be consistent with the indirect estimate from the land hydrology data (i.e., 0.03 km3/year, (Ting and Overmars, 1995)). The area of the suspected SGD in the coastal part of the study region was also characterized by an overall minimum in the horizontal distribution of the bottom salinity (data collected about 10 cm above the sea floor) in October, as well as February (Fig. 4). Also illustrative are the vertical profiles of the water transparency (Fig. 5) and fluorescence (Fig. 6). The former Fig. 3. Vertical profiles of salinity at Station 11 in February, 2009 (circles) and October, 2009 (triangles). The depth is 8 m. The salinity decrease in the bottom parts of the both profiles is hypothetically associated with SGD.

Table 1 Summary of data on SGD-affected near-bottom layer at different locations in February, 2009, and October, 2009. DH and DS are thickness of the layer and salinity drop, respectively. Station

1 6 11 16 G

February, 2009

October, 2009

DH, m

DS, psu

DH, m

DS, psu

0.13 0.27 2.15 0.14 Unknown

0.035 0.061 0.052 0.010 Unknown

0.15 0.42 0.46 0.33 0.19

0.009 0.046 0.023 0.105 0.048

bottom, we obtain w ¼ k=S0 DS=DH: This procedure is somewhat similar to that described by (Martin et al., 2007), although they used chlorinity instead of salinity. The SGD mass rate Q per unit area can then be obtained by multiplying w by the water density. If, for example, S0 ¼30 psu and k¼10  4 m2/s (which, up to the order of magnitude, is typical for the bottom layer), we arrive at the following order of magnitude estimate for the groundwater inflow for the range of the observed DH and DS depicted in Table 1: 0:1 g m2 s1 o Q o1 g m2 s1 : These values are approximately equivalent to a volume flux between 10 and 100 l per m2 per day, or Darcy velocities between 1 cm day  1 and 10 cm day  1. Of course, this is only an order of magnitude estimate, which is strongly dependent on the choice of k and other assumptions of our oversimplified calculation. However, it agrees well with the characteristic values reported earlier for well-developed SGD at other locations in the ocean (e.g., Intergovernmental Oceanographic Commission, 2004, Martin et al., 2007). We note that the near-bottom SGD-related anomalies on the Taiwan shelf are likely to be small patches whose sizes are of the order of tens of meters, or even smaller. Indeed, an additional CTD profile taken only 50 m northwestward from station G revealed no anomaly at all. Two additional profiles were taken shoreward from station 11 at the distances 100 m and 200 m, the first one

Fig. 4. Horizontal distributions of salinity at bottom in October, 2009 (a) and February, 2009 (b). The data were collected about 10 cm above the bottom.

22

P.O. Zavialov et al. / Continental Shelf Research 34 (2012) 18–25

Fig. 5. Vertical profile of light extinction coefficient in the lowermost 5 m of the water column (dashed line, circles—example profile from Station 11, October, 2009). The depth is 8 m. Also shown is the corresponding profile of salinity (solid curve). The peculiar feature in the bottom parts of the profiles is hypothetically associated with SGD.

Fig. 7. Horizontal distribution of turbidity (NTU) at the bottom. The data were collected about 30 cm above the bottom. October, 2009.

associated with an elevated content of the nutrients and plankton blooming in the SGD-affected layer (see below). In the figure, the measured fluorescence values are converted to the chlorophyll-a concentration units through the standard Sea-Birds’ procedure (Sea-Bird Electronics, 2010). The overall distribution of the turbidity at bottom (30 cm above the sea floor) is shown in Fig. 7. It exhibits a rather complex pattern, but, generally, the smaller values are seen in the SGD-suspected sites (locations 6, 11, 16, and others, while the maxima occupy the ‘‘no SGD’’ areas (Fangliao townsip—stations A–C, and others)). 3.2. Chemical indicators

Fig. 6. Vertical profile of fluorescence in the lowermost 5 m of the water column (dashed line, circles—example profile from Station 11, February, 2009). The depth is 8 m. Also shown is the corresponding profile of salinity (solid curve). The increase of fluorescence coincident with salinity decrease in the bottom parts of the profiles is hypothetically associated with SGD.

demonstrates the growth of the light extinction coefficient towards the bottom, which could be associated with the resuspension of the sediment from the bottom by shear and wave mixing. However, in the very near-bottom layer of the thickness 10–50 cm, coincident with the layer of the salinity drop, the light extinction coefficient decreased, suggesting that the water therein may be of a foreign origin, and some process ‘‘pushes’’ the suspended sediment away from the bottom. In most cases (albeit not all), the fluorescence significantly increased in the lower salinity layer (Fig. 6). Possibly, this is

Results of chemical analyses of the water samples collected from the bottom layer of the ocean, as well as the near-shore land groundwater well and Kaoping River in February and October are depicted in Tables 2 and 3. In the tables, the suspected SGD sites (stations 1, 6, 11, 16, and G) are shown in bold. Additional data on the concentrations of the dissolved oxygen, nitrite, lead, zinc, copper, and phosphorus in the samples were also obtained in this study, but they lacked any obvious correlation with the SGD and were left beyond the scope of this article. The concentrations of the nitrate, silica, chlorophyll-a, and the total iron, as shown in Tables 2 and 3 do demonstrate certain differences between the sites hypothetically affected by SGD and those not exposed to it, although not in all cases. Perhaps, the most illustrative are the nitrate and silica: on the average, their respective contents at ‘‘SGD-suspected’’ stations were elevated with respect to the overall mean of the oceanic samples by 32% and 58% in February, and 107% and 43% in October. This increase appears to be consistent with the analyses reported by (Cheng et al., 2005) for the same area. Moreover, the observed elevated concentrations of NO3 cannot be explained by the river inputs, because the nitrate content in the river water was too low. We also noted that the dissolved organic carbon concentrations in the groundwater collected from the well varied between 4.6 and 5.9 mg/l in the both surveys, while those in the seawater ranged between 1.0 and 2.8 mg/l. The content of organic carbon in the bottom sediments varied between 0.31 and 0.46%. In the related literature, SGD has been mentioned as a significant source of nutrients elsewhere in the coastal ocean. For example, (Niencheski et al., 2007) estimated that SGD was

P.O. Zavialov et al. / Continental Shelf Research 34 (2012) 18–25

23

Table 2 Some chemical indicators of the water samples collected from the bottom layer of the ocean, groundwater well, and Kaoping River in February, 2009. In addition to the weight units, molar concentrations are given in parentheses for nitrogen and silica. The SGD sites suspected based on CTD, turbidity, and fluorescence profiling are shown in bold. Also shown is the distance to the coast for all stations. Test item

Distance from the coast km

Nitrogen as Nitrate (NO3–N) mg/L (mM/L)

Silica as (SiO2) mg/L (mM/L)

Chlorophyll a mg/m3

Iron (Fe) dissolved and particulate, mg/L

0.5 1.7 1.9 0.4 6.8 5.0 2.9 1.2 0.3 6.9 5.0 3.0 1.3 0.4 N/A N/A N/A N/A

0.04 0.03 0.04 0.04 0.02 0.02 0.02 0.02 0.03 0.02 – 0.01 0.01 0.02 0.02 0.03 0.48 0.06

0.38 0.30 0.31 0.17 0.09 0.07 0.08 0.09 0.13 0.08 0.07 0.03 0.08 – 0.14 0.23 11.6 15.5

2.0 0.8 0.8 0.8 0.3 0.8 1.7 0.8 1.1 0.6 0.8 1.1 0.6 0.6 0.91 1.12 0.3 31.9

0.167 0.097 0.208 0.172 0.040 0.082 0.142 0.078 0.080 0.086 0.032 0.028 0.039 0.037 0.092 0.114 0.112 0.366

Station# 1 2 5 6 7 8 9 10 11 12 13 14 15 16 Average Average, SGD sites GW well Kaoping River

(2.8) (2.1) (2.8) (2.8) (1.4) (1.4) (1.4) (1.4) (2.1) (1.4) (0.7) (0.7) (1.4) (1.4) (2.1) (34.3) (4.3)

(13.6) (10.7) (11.1) (6.1) (3.2) (2.5) (2.9) (3.2) (4.6) (2.9) (2.5) (1.1) (2.9) (5.0) (8.2) (414.3) (553.6)

Table 3 Some chemical indicators of the water samples collected from the bottom layer of the ocean, groundwater well, and Kaoping River in October, 2009. In addition to the weight units, molar concentrations are given in parentheses for nitrogen and silica. The SGD sites suspected based on CTD, turbidity, and fluorescence profiling are shown in bold. Also shown is the distance to the coast for all stations. Test item

Distance from the coast km

Nitrogen as Nitrate (NO3–N) mg/L (mM/L)

Silica as (SiO2) mg/L (mM/L)

Chlorophyll-a mg/m3

Iron (Fe) dissolved and particulate, mg/L

0.5 0.4 1.2 0.3 1.3 0.4 0.5 1.3 2.2 0.3 1.2 2.1 0.2 0.9 1.8 N/A N/A N/A N/A

0.12 0.06 0.04 0.03 0.05 0.04 0.18 0.07 0.03 0.04 0.04 0.02 1.12 0.06 0.04 0.13 0.27 4.49 0.11

0.34 0.05 0.02 0.03 0.06 0.51 0.27 0.07 0.07 0.06 0.06 0.05 0.14 0.46 0.04 0.15 0.22 17.7 7.66

1.4 0.8 1.4 1.4 o0.1 0.3 0.6 0.3 0.3 o0.1 0.3 0.3 0.6 0.8 0.3 0.7 0.9 0.3 0.6

0.237 0.387 0.126 0.033 0.106 2.67 0.135 0.142 0.111 0.120 0.104 0.172 0.739 1.08 0.307 0.43 0.81 0.714 0.101

Station# 1 6 10 11 15 16 A B C D E F G H K Average Average, SGD sites GW well (Land) Kaoping River

(8.6) (4.2) (2.8) (2.2) (3.6) (2.8) (12.8) (5.0) (2.2) (2.8) (2.8) (1.4) (80.0) (4.2) (2.8) (9.2) (19.2) (320.8) (7.8)

responsible for up to 55% of the total nutrient flux into a 240 km wide near shore surf zone in the Southwestern Atlantic. The ‘‘SGD-suspected’’ sites also demonstrated higher concentrations of chlorophyll (23% above the overall average in February and 32% in October)—possibly, because of the increased availability of nutrients and associated plankton blooming. Finally, the concentration of iron appears to be a promising SGD marker in the area. A number of previous studies demonstrated that SGD is a potentially important source of Fe flux into the coastal ocean (e.g., Windom et al., 2006, Roy et al., 2011). It has been hypothesized that the precipitation of iron oxides in subterranean estuaries could act as a geochemical barrier by retaining and accumulating certain dissolved chemical species carried to the coast by groundwater (Charette and Sholkovitz, 2002). The Fe concentrations in all samples collected and analyzed in the present study look unrealistically high compared with the

(12.1) (1.8) (0.7) (1.0) (2.1) (18.2) (9.6) (2.5) (2.5) (2.1) (2.1) (1.8) (5.0) (16.4) (1.4) (5.4) (7.9) (632.1) (273.6)

values generally characteristic for the dissolved iron in the open oceans, which rarely exceed 1 nM/l (e.g., Johnson et al., 2007). However, such a high content of iron is not uncommon for rivers and groundwater. Iron concentrations as high as several mM/l in the coastal ocean waters have been reported for the regions of SGD and freshwater influence (e.g., Windom et al., 2006), in particular, in the East China Sea (Ma et al., 1982). Iron concentrations over 100 mM/l have been measured in sediment pore water at subterranean estuaries (Martin et al., 2007, Roy et al., 2010). In our study, the iron concentrations exhibited a pronounced increase at the stations of suspected SGD (24% above the overall average in February and 88% in October), and, again, this should be associated with the groundwater rather than river inputs, given that, at least in October, the Fe content in Kaoping River water was low, while in the groundwater sample it was quite high (see Tables 2 and 3).

24

P.O. Zavialov et al. / Continental Shelf Research 34 (2012) 18–25

3.3. N-alkane composition of bottom sediments Additional indirect evidence of SGD has been obtained from the chromatographic analyses of the organic matter in the bottom sediment samples and the water samples. In particular, the n-alkane composition of the organics in the water sample collected from the land groundwater well in February, 2009, exhibited very high content of the C24 alkane (Fig. 8). Unusual for marine environments, this alkane known as tetracosane has a terrestrial origin and must have resulted from an unidentified microbiological activity on land (e.g., Doskey, 2000]. In this case, therefore, this

alkane could be used as a marker of the groundwater in the ocean. Indeed, high concentrations of the C24 alkane were documented for the near-shore sediment samples from SGD-suspected locations 1 (Fig. 8) and 6. However, in the sediment sample taken at location 11, there was no elevated concentration of C24 alkane. In October 2009, the n-alkane composition of the sampled groundwater (Fig. 9) was quite different from that in February 2009. The C24 alkane was no longer dominant, perhaps, because the processes responsible for its generation had intermittent or seasonal nature. It is known from the related literature (e.g., Doskey, 2000, Sachse et al., 2009) that the n-alkane spectrum and concentration in the soil and ground water may vary significantly at the seasonal scale, in particular, following the seasonal changes of vegetation or/and precipitation regimes. In October, the principal alkane was hexadecane C16. It was present in all sediment samples, but, again, the maxima of its concentration in the bottom sediments were observed at the SGD-suspected sites 1, 11, and G, as illustrated by Fig. 9. In this sense, the alkane composition analyses appear to be generally supportive of the existence of SGD at the locations identified above. Curiously, the C16 alkane is usually associated with plankton communities and believed to be characteristic for the marine rather than terrestrial environments (Belyaeva and Bobyleva, 1981, Carvalho et al., 2009). Thus, the exchanges between the sea water and the groundwater may be reciprocal.

4. Conclusions

Fig. 8. (a)—Alkane composition of organic matter in a water sample from groundwater well. February, 2009. (b)—Alkane composition of organic matter in bottom sediment samples from Station 1 (filled crcles) and Station 6 (triangles). February, 2009.

In this study, we attempted to summarize the newly collected oceanographic data pointing to the existence of submarine groundwater discharge on the inner shelf of Pingtung area, Taiwan Strait, South China Sea. We discussed here several pieces of physical and chemical evidence. Probably, none of those taken alone can be considered sufficient, but altogether they enable us to affirm rather confidently that SGD discharge in this area does exist, specify its approximate locations, and get some insight into its plausible volume rates. From the general oceanographic standpoint, the impact of SGD on the regime of the coastal waters in the study area appears to be moderate. However, there do exist distinctive features restricted to the very near-bottom layer that are very likely to be associated with SGD. These features are typically small-scale patches, but their locations seem to be stable. The hydrographic measurements as well as the chemical indicators revealed likely SGD sites in the immediate proximity of the shore, at the depths never exceeding 8 m (Donggang, Linbian, Jiadong, Fangshan townships). No SGD signatures were observed at the other near-shore locations (Fangliao township). Rough estimates indicate that the observed features could have been produced by an SGD at the rates 0.1–1 g m2 s  1, which appears to be consistent with some SGD figures previously reported for other regions. A higher spatial and temporal resolution field observations combined with direct seepage meter measurements are needed to produce more accurate and reliable estimates of the SGD in the study area.

Acknowledgments

Fig. 9. (a)—Alkane composition of organic matter in a water sample from groundwater well. October, 2009. (b)—Alkane composition of organic matter in bottom sediment sample from Station 11. October, 2009.

This study was made possible thanks to bilateral TaiwaneseRussian Project ‘‘Monitoring, Assessment, and Management Implications of Submarine Groundwater Discharge in Taiwan’’ between Shirshov Institute of Oceanology, Russian Academy of Sciences, Russia, and Tainan Hydraulics Laboratory, National Cheng-Kung University, Taiwan. Partial supports from Russian Foundation for Basic Research, National Scientific Council of Taiwan, and Russian Academy of

P.O. Zavialov et al. / Continental Shelf Research 34 (2012) 18–25

Sciences, are also acknowledged. The authors wish to thank many colleagues in Taiwan and Russia who participated in the field work and data analyses. We also thank two anonymous reviewers for their useful comments. References Belyaeva, A.N., Bobyleva, N.V., 1981. Alkanes in plankton and bottom sediments of the Baltic Sea. Oceanology 21 (6), 712–715. doi:10.1594/PANGAEA.755437. Bokuniewicz, H., Pavlik, B., 1990. Groundwater seepage along a Barrier Island. Biogeochemistry 10, 257–276. Burnett, W.C., Kim, G., Lane-Smith, D., 2001. A continuous monitor for assessment of 222Rn in coastal ocean. Journal of Radioanalytical and Nuclear Chemistry 249 (1), 167–172. Carvalho, C.C.C.R., Wick, L.Y., Heipieper, H.P., 2009. Cell wall adaptations of planktonic and biofilm Rhodococcus erythropolis cells to growth on C5 to C16 n-alkane hydrocarbons. Applied Microbiology and Biotechnology 82, 311–320. doi:10.1007/s00253-008-1809-3. Charette, M.A., Sholkovitz, E., 2002. Oxidative precipitation of groundwaterderived ferrous iron in the subterranean estuary of a coastal bay. Geophysical Research Letters 29 (10), 1444. doi:10.1029/2001GL014512. Cheng, C.T.A., Liu, J.T., Tsuang, B.J., 1995. Island-based catchement—the Taiwan example. Regional Environmental Change 4, 39–48. Cheng, C.T.A., Zhang, J., Peng, T.-R., Hagiwara, T., 2005. Exploratory sampling of submarine groundwater discharge in Taiwan. Chikyukagaku (Geochemistry) 39, 165–171. Doskey, P.V., 2000. The air-water exchange of C15—C31 n-alkanes in a precipitation-dominated seepage lake. Atmospheric Environment 34 (23), 3981–3993. doi:10.1016/S1352-2310(00)00165-5. Glover, R.E., 1959. The pattern of fresh-water flow in a coastal aquifer. Journal of Geophysical Research 64 (4), 457–459. doi:10.1029/JZ064i004p00457. Hsu, K.-C., Wang, C.-H., Chen, K.-C., Chen, C.-T., Ma, K.-W., 2007. Climate-induced hydrological impacts on the groundwater system of the Pingtung Plain, Taiwan. Hydrology Journal 15, 903–913. Huang, K.F., You, C.F., Chung, C.H., Lin, I.T., 2011. Nonhomogeneous seawater Sr isotopic composition in the coastal oceans: a novel tool for tracing water masses and submarine groundwater discharge. Geochemistry Geophysics Geosystems 12 (5), Q05002. doi:10.1029/2010GC003372. Intergovernmental Oceanographic Commission, 2004. Submarine groundwater discharge. Management implications, measurements, and effects. IOC Manuals and Guides 44, 35. Johnson, K.S., Boyle, E., Bruland, K., Coale, K., et al., 2007. Developing standards for dissolved iron in seawater. Electro Optical Systems 88 (11), 131–132. Lee, D.R., 1977. A device for measuring seepage flux in lakes and estuaries. Limnology and Oceanography 22, 140–147. Lin, I.T., Wang, C.H., Lin, S.W., 2003. Seasonal variations of oxygen isotopic compositions in the Pingtung coastal waters of Taiwan. Western Pacific Earth Sciences 3 (1), 21–32. Lin, I.T., Wang, C.H., You, C.F., Lin, S.W., Huang, K.F., Chen, Y.G., 2010. Deep submarine groundwater discharge indicated by tracers of oxygen, strontium isotopes and barium content in the Pingtung coastal zone,southern Taiwan. Marine Chemistry 122, 51–58.

25

Ma, X., Li, Q., Huang, H., Hua, W., Shen, W., Li, G., Jiang, C., 1982. Iron in sea water of the region off Changjiang River estuary. Oceanologia et Limnologia Sinica 13, 3 doi: cnki:ISSN:0029-814X.0.1982-03-006. Martin, J.B., Cable, J.E., Smith, C., Roy, M., Cherrier, J., 2007. Magnitudes of submarine groundwater discharge from marine and terrestrial sources: Indian River Lagoon, Florida. Water Resources Research 43, W05440. doi:10.1029/ 2006WR005266. Moore, W.S., 1996. Large groundwater inputs to coastal waters revealed by 226Ra enrichments. Nature 380, 612–614. Moore, W.S., 1999. The subterranean estuary: a reaction zone of ground water and sea water. Marine Chemistry 65, 111–125. Niencheski, L.F.H., Windom, H.L., Moore, W.S., Jahnke, R.A., 2007. Submarine groundwater discharge of nutrients to the ocean along a coastal lagoon barrier, Southern Brazil. Marine Chemistry 106, 546–561. doi:10.1016/ j.marchem.2007.06.004. Peresypkin, V.I., Smurov, A.V., Shulga, N.A., Safonova, E.S., Smurova, T.G., Bang, C.V., 2011. Composition of organic matter in water, suspended material, and bottom sediments of Nha Trang Bay (Vietnam, South China Sea). Okeanologiya 51 (6), 1–11 [in Russian]. Rosenberry, D.O., Morin, R.H., 2004. Use of an electromagnetic seepage meter to investigate temporal variability in lake seepage. Groundwater 42, 68–77. Roy, M., Martin, J.B., Cherrier, J., Cable, J.E., Smith, C.G., 2010. Influence of sea level rise on iron diagenesis in an east Florida subterranean estuary. Geochimica et cosmoschimica acta 74, 5560–5573. Roy, M., Martin, J.B., Smith, C.G., Cable, J.E., 2011. Reactive-transport modeling of iron diagenesis and associated organic carbon remineralization in a Florida (USA) subterranean estuary. Earth and Planetary Science Letters 304, 191–201. doi:10.1016/j.epsl.2011.02.002. Sachse, D., Kahmen, A., Gleixner, G., 2009. Significant seasonal variation in the hydrogen isotopic composition of leaf-wax lipids for two deciduous tree ecosystems (Fagus sylvativa and Acer pseudoplatanus). Organic Geochemistry 40 (6), 732–742. Sea-Bird Electronics, Application Note No. 74, June 2010, viewed at: /http://www. seabird.com/pdf_documents/ApplicationNotes/appnote74jun10.pdfSs. Swarzenski, P.W., Reich, C.D., Spechler, R.M., Kindinger, J.L., Moore, W.S., 2001. Using multiple geochemical tracers to characterize the hydrogeology of the submarine spring off Crescent Beach, Florida. Chemical Biology 179, 187–202. Ting, C.-S., 1997. Groundwater Resources Evaluation and Management for Pingtung Plain. Free University, Amstrdam, Taiwan. Ting, C.-S., Overmars, M., 1995. Groundwater Recharge in Pingtung Plain, South Taiwan. Research Report, National Pingtung Polytechnic Institute, Pingtung, Taiwan. TPCWB—Taiwan Provincial Water Conservancy Bureau, 1995. Land Subsidence Investigation in Western-south Coastal Area. Technical Report [in Chinese]. Valiela, I., D’Elia, C., 1990. Groundwater inputs to coastal waters. Special Issue. Biogeochemistry 10, 328. Windom, H.L., Moore, W.S., Niencheski, L.F.H., Jahnke, R.A., 2006. Submarine groundwater discharge: a large, previously unrecognized source of dissolved iron to the South Atlantic Ocean. Marine Chemistry 102 (3-4), 252–266. doi:10.1016/j.marchem.2006.06.016. Zhang, J., Satake, H., 2003. The chemical characteristics of submarine groundwater seepage in Toyama Bay, Central Japan. In: Taniguchi, M., Wang, K., Gamo, T. (Eds.), Land and Marine Hydrogeology. , Elsevier, pp. 45–60.