Article
Volume 12, Number 5
7 May 2011
Q05002, doi:10.1029/2010GC003372
ISSN:
1525‐2027
Nonhomogeneous seawater Sr isotopic composition
in the coastal oceans: A novel tool for tracing water
masses and submarine groundwater discharge
Kuo‐Fang Huang
Earth Dynamic System Research Center, National Cheng Kung University, Tainan 701, Taiwan
Also at Department of Earth Sciences, National Cheng Kung University, Tainan 701, Taiwan
Now at Department of Geology and Geophysics, Woods Hole Oceanographic Institution, Woods Hole,
Massachusetts 02543, USA (kfhuang05@gmail.com)
Chen‐Feng You and Chuan‐Hsiung Chung
Earth Dynamic System Research Center, National Cheng Kung University, Tainan 701, Taiwan
Also at Department of Earth Sciences, National Cheng Kung University, Tainan 701, Taiwan
In‐Tian Lin
Department of Geosciences, National Taiwan University, Taipei 106, Taiwan
[
1] Here we present high‐precision (2s = ±3 ppm)
87
Sr/
86
Sr measurements in coastal waters, together with
salinity, to evaluate water mass mixing and the influence of submarine groundwater discharge (SGD) in
coastal waters and marginal seas. Nonhomogeneous Sr isotopic variations in water columns were documen-
ted in the Southern Okinawa Trough (SOT), South China Sea, and Kao‐ping Canyon (KPC), where seawater
87
Sr/
86
Sr varied up to 70 ppm. Seawater Sr isotopic composition changes only slightly in the upper 200 m of
the SOT but was detectable and highly correlated with salinity, indicating a mixing between radiogenic North
Pacific Tropical Water (high
87
Sr/
86
Sr and high salinity) at 100–150 m and a less radiogenic component with
low
87
Sr/
86
Sr and low salinity at ∼200 m. Vertical profiles of seawater
87
Sr/
86
Sr along the KPC show signif-
icant variations, suggesting dynamic mixing affected by continental inputs (i.e., river runoff and SGD) in this
region. These results highlight the potential use of seawater Sr isotopes as a powerful tracer for determining
mixing ratios and the dynamic mixing of oceanic water masses, especially in coastal and marginal seas.
Components: 7700 words, 5 figures, 3 tables.
Keywords: seawater Sr isotope; water mass mixing; submarine groundwater discharge.
Index Terms: 1040 Geochemistry: Radiogenic is otope geochemistry; 1050 Geochemistry: Marine geochemistry (4835,
4845, 4850).
Received 22 September 2010; Revised 7 March 2011; Accepted 16 March 2011; Published 7 May 2011.
Huang, K.‐F., C.‐F. You, C.‐H. Chung, and I.‐T. Lin (2011), Nonhomogeneous seawater Sr isotopic composition in the
coastal oceans: A novel tool for tracing water masses and submarine groundwater discharge, Geochem . Geophy s.
Geosyst., 12, Q05002, doi:10.1029/2010GC003372.
Copyright 2011 by the American Geophysical Union 1 of 14
1. Introduction
[2] The Sr isotopic composition (Sr IC) of marine
precipitates, such as biogenic and authigenic
carbonates, has been widely used in global climate
change studies and serves as a crucial tool
for stratigraphic correlation and dating carbonate
sequences [e.g., Burke et al., 1982; Dia et al., 1992;
Henderson et al., 1994]. These applications assume
that Sr behaves uniformly both in chemical and
isotopic distribution because of its much longer
residence time (2–2.5 Myr) relative to the ocean
mixing time scale (∼10
3
yr) [Basu et al., 2001]. The
homogeneity of seawater Sr IC was also examined
by
87
Sr/
86
Sr in the Hudson Bay (S = 21.9), Pacific
Ocean, Atlantic Ocean and Arctic Ocean [Capo and
DePaolo, 1992; Winter et al., 1997], as well as
in marine carbonates [e.g., Burke et al., 1982].
These published data suggest that open ocean Sr IC
is spatially homogeneous, with no measurable var-
iation at the analytical precision of 20 –30 ppm (2s)
[Henderson et al., 1994].
[3] Recent studies, however, have documented
small but considerable regional differences in both
Sr flux and Sr IC on time scales much shorter than
the Sr residence time in the ocean [Stoll and Schrag,
1998; de Villiers, 1999], especially in the estuarine
mixing zone and coastal ocean [Andersson e t al.,
1994; Xu and Marcantonio, 2004; Huang and
You, 2007]. These findings reflect recent advances
in analytical precision over traditional Sr isotope
techniques, and also highlight the need for a more
detailed understanding of Sr geochemical cycling in
the coastal oceans, and hence the oceanic Sr budget.
Several potential mechanisms for causing the non-
conservative behavior of Sr and Sr IC in the estua-
rine and coastal environments have been proposed
in earlier works: (1) Sr scavenging by Fe‐Mn oxy-
hydroxides [Andersson et al., 1994], (2) intense
water‐sediment interaction [Huang and You, 2007],
and (3) environmental disturbances causing com-
positional changes in the suspended and dissolved
phases [Xu and Marcantonio, 2004]. These studies
have highlighted that the
87
Sr/
86
Sr isotopic ratio
in coastal waters might deviate from that of con-
temporaneous seawater, and may influence the
application of seawater Sr IC for stratigraphic dating
purposes. On the other hand, these observations
highlight the potential use of seawater Sr IC as a
powerful tracer for identifying processes occurring
between seawater and continental freshwater in
the coastal oceans [Huang and You, 2007] because
oceanic waters (
87
Sr/
86
Sr = 0.709176) and conti-
nental freshwaters (e.g.,
87
Sr/
86
Sr = 0.741 in the
Ganges catchment [Bickle et al., 2003]) differ sig-
nificantly in their Sr ICs.
[4] Detailed studies of Sr nonconservative behavior
in the coastal oceans are scarce. In our earlier
publication, small but distinguishable Sr isotopic
variations (up to 50 ppm) were detected in coastal
and estuarine environments after a typhoon event
[Huang and You, 2007; Lin et al., 2010]. The present
study aims at exploring the use of seawater
87
Sr/
86
Sr
for tracing potential mixing processes among dis-
tinct water masses in marginal seas (i.e., Southern
Okinawa Trough and South China Sea) and coastal
regions (Kao‐ping Canyon, Southwestern Taiwan).
Toward this end, we present seawater Sr IC
measurements of superior precision (±3 ppm, 2s)
coupled with conventional salinity measurements
for a suite of depth profiles in the central and west-
ern Pacific. The potential use of seawater Sr IC
for tracing continental inputs from closely situated
rivers and water mass mixing offshore is sys-
tematically evaluated, in particular for regions
where salinity variations among water masses are
indistinguishable.
2. Methods
2.1. Sample Collection
[5] Seawater from three vertical profiles in the
Southern Okinawa Trough (SOT) were sampled
during the R/V Ocean Researcher I‐679 expedition
in April 2003 (Figure 1b), and used for studying
water mass mixing of the Kuroshio Current near
Taiwan. Four depth profiles along the Kao‐ping
Canyon (KPC) were collected during the R/V Ocean
Researcher III‐720 expedition in August 2001,
approximately 1 week after Typhoon Toraji, to
investigate the influence of continental inputs, such
as river runoff and SGD. Three reference sites, ST‐1
(10°N, 140°W) and SB‐09 (34.13°N, 120.02°W)
located at the North Equatorial Current (NEC) and
California Borderland, respectively, as well as
SCS‐C (15°N, 115°30′E) in the central South
China Sea (SCS), were analyzed for comparison
(Figure 1a).
[6] Immediately after sampling through Niskin
bottles (X type) deployed on a Rosette assembly, the
sample was filtered through a 0.45 mm membrane,
transferred to trace metal clean HDPE containers
and prevented from freezing during transit and
storage. Water samples for seawater Sr IC, dissolved
Ba and Mn determinations were acidified to pH∼1.5
using subboiled quartz distilled HNO
3
on board.
Geochemistry
Geophysics
Geosystems
G
3
G
3
HUANG ET AL.: NONHOMOGENEOUS SEAWATER Sr ISOTOPE 10.1029/2010GC003372
2of14
Figure 1. Locations of seawater sampling stations in the (a) North Pacific and the (b) SOT and KPC transects around
Taiwan.
Geochemistry
Geophysics
Geosystems
G
3
G
3
HUANG ET AL.: NONHOMOGENEOUS SEAWATER Sr ISOTOPE 10.1029/2010GC003372
3of14
Temperature and salinity were measured using a
Sea‐Bird Electronics SBE 9/11 plus CTD Con-
ductivity, and the uncertainty for salinity is within
±0.01.
2.2. Analytical Method
[7] The Sr separation procedures used in this study
were modified from Deniel and Pin [2001]. In
the laboratory, Sr was purified under a class 10
flow bench. For Sr IC analyses, 0.1 mL of seawater
containing 700–800 ng Sr was passed through
Sr
SPEC
resin (Eichrom™, 1 mL) and the elutant
(in 0.05N HNO
3
) was evaporated to dryness and
redissolved in 0.1N HCl. Total procedure blanks
of Sr were 10–20 pg and are negligible compared
with the loading size. The purified Sr was then
loaded onto double Ta/Re filaments, and Sr isotope
ratios were measured using a static mode Thermal
Ionization Mass Spectrometer (Thermo‐Fisher Sci-
entific Triton TI) installed at EDSRC, NCKU. The
88
Sr beam was held at 200 pA for 1 h and 135 ratios
with 15 blocks were collected. Mass fractionation
and Rb contribution were corrected by
86
Sr/
88
Sr
normalization to 0.1194 applying an exponential law
and the natural
85
Rb/
87
Rb abundance, respectively.
Replicate analyses of NBS 987,
87
Sr/
86
Sr =
0.710270 ± 03 (2s, n = 25) show an excellent long‐
term reproducibility of better than 3 ppm. However,
standard reproducibility does not account for vari-
ability in natural samples due to matrix effects or
interferences. Therefore, six seawater duplicates
from the North Pacific were also carried out and
further confirmed that analytical reproducibility for
the natural seawater is similar to the long‐term
precision and within‐run precision (<3 ppm). This
technique thus enables us to examine small natural
variability in coastal waters [Huang and You, 2007].
In the following discussion, the seawater Sr IC is
expressed as the ppm deviation from that of deep
North Pacific seawater.
D
87
Sr ¼
87
Sr
86
Sr
sample
87
Sr
86
Sr
STD
1
0
B
B
B
@
1
C
C
C
A
10
6
ppmðÞ
where STD represents the present‐day seawater in
the deep North Pacific Ocean,
87
Sr/
86
Sr = 0.709176 ±
03, which is identical within error to the deep water
of the SCS (0.709180 ± 03).
[8] Dissolved Ba and Mn of the estuarine waters
were directly determined using 200‐fold dilution
samples by high‐ resolution inductively coupled
plasma mass spectrometry (HR‐ICP‐MS, Thermo‐
Fisher Scientific Element 2) at EDSRC, NCKU. A
series of matrix‐matched standards prepared from
certified seawater reference materials, NASS‐5
(National Research Council Canada, salinity = 30.4)
and CASS‐4 (National Research Council Canada,
salinity = 30.7), were used to correct for potential
matrix‐induced mass discrimination. Estimates of
analytical precision and accuracy for dissolved Ba
and Mn based upon analyses of the standards are on
the order of 5%. Detection limit is 5 ppt and 10 ppt
for dissolved Ba and Mn, respectively.
3. Hydrography and Water Mass
3.1. North Pacific
[9] The major features of the surface circulation
in north Pacific are (1) large anticyclonic gyres
with axes along 20°N and (2) the equatorial current
system [Reid, 1997]. The North Equatorial Current
(NEC) is the parental water mass of the North Pacific
Tropical Water (NPTW) and Kuroshio Current (KC).
The NEC flows westward between 12 and 13°N, and
breaks into two branches, the northward KC and
southern Mindanao Current, near the Luzon Island.
Although the KC originates from the NEC, the
chemical composition between the KC and NEC
may be significantly altered due to exchange with
volcanic materials in the Luzon Arc [ Tomczak and
Godfrey, 1994]. The KC flows northward and joins
the North Pacific Current, which is the upper stream
of the California Current (Figure 1a).
[10] The water masses of the intermediate and deep
North Pacific Ocean were described by Tomczak
and Godfrey [1994] as Subarctic Intermediate
Water or North Pacific Intermediate Water (NPIW)
and Pacific Deep Water (PDW). The salinity mini-
mum near 800–1000 m depth indicates the present
of this intermediate water. NPIW originates from the
Polar Front, where it is formed by mixing of surface
and deeper waters and subducted into the subtropical
gyre. The constituents of PDW are North Atlantic
Deep Water, Antarctic Bottom Water and Antarctic
Intermediate Water.
3.2. South Okinawa Trough
[11] The Kuroshio Current, a major western
boundary current in the Pacific, consists of the
following well‐defined components at depths shal-
lower than 1500 m [Chen et al., 1995]. In the SOT
region, the major water masses and their hydrolog-
ical properties are: Kuroshio Surface Water, with the
highest temperature (>25°C) and pH, and depleted
Geochemistry
Geophysics
Geosystems
G
3
G
3
HUANG ET AL.: NONHOMOGENEOUS SEAWATER Sr ISOTOPE 10.1029/2010GC003372
4of14
in nutrients and alkalinity; NPTW, characterized by
high nutrients and alkalinity, as well as a subsurface
salinity maximum ∼34.9 (s
= 23) at 100 to 200 m
(shallower in winter and deeper in summer), and
originating from the northward branch of the NEC;
NPIW occupies depths between 400 and 1500 m,
marked by a salinity minimum (34.2, s
= 26.7)
at ∼490 m [Nitani, 1972] and most likely forms by
a shortcut of Okhotsk Sea source water into the
western subtropical gyre [You, 2003]. The layer
immediately beneath NPIW is occupied by the
nutrient‐depleted PDW. At depths greater than
2000 m, temperature and salinity are rather uniform
( = 1.1–1.8°C and S = 34.6–34.7) [Amakawa et al.,
2004]. Figure 2a shows the potential temperature‐
salinity diagrams at three stations investigated in
this study. The ‐S plot clearly defines the water
masses in the SOT, such as the NPTW, NPIW and
PDW.
3.3. South China Sea and Kao‐ping
Submarine Canyon
[12] The distributions of water masses at the KPC
are rather complicated due to intense regional mix-
ing of multiple source waters from on‐land Taiwan.
Several water masses were identified near the KPE
and the SCS based on hydrographic measurements
[Nitani, 1972]. The Kao‐ping River Water carries a
large amount of fresh runoff and precipitation with
salinity values close to zero and occupies predomi-
nantly the surface layer during prevailing summer
monsoon seasons. The South China Sea Surface
Water transports waters from the SCS and could be
strongly influenced by surface circulation due to
annually reversing monsoon winds. Due to intensive
upwelling and vertical mixing, the salinity extremes
in this region are less pronounced than in the SOT,
i.e., NPTW has a maximum salinity of 34.65 cen-
tered at ∼150 m deep, and NPIW has a minimum
salinity of 34.4 at around 500 m [Chou et al., 2007].
In the deep SCS, the Luzon Strait (∼2600 m) serves
as the only gateway of deep water exchange between
the SCS and deep Pacific water. All SCS waters
below 1500 m display similar hydrographic prop-
erties (e.g., and S) as that of the Pacific at 2000 m,
except at 1500 m where a slightly high T occurred
[Chen et al., 2006]. The ‐S plot clearly shows the
water masses in the KPC and the SCS, including
surface water, NPTW, NPIW and PDW (Figure 2b).
4. Results and Discussion
4.1. Vertical Distribution of Sr IC
in the North Pacific
[13] The depth profile of seawater D
87
Sr in the
California Borderland, SB‐09, shows <10 ppm
variation within the upper 500 m; in contrast, the
vertical D
87
Sr profile in the open ocean station,
ST‐1, displays a nonhomogeneous pattern, ranging
from 0 to +37 ppm (Figure 3a and Table 1). More
radiogenic D
87
Sr ( ∼ +20 ppm) at depths of 100–
150 m (occupied by NPTW) and 400–600 m
(dominated by NPIW) are detected, and can be used
to represent the Sr isotopic signatures of the KC
water masses. It is interesting to note that a similar
Figure 2. The ‐S diagram showing distributions of water samples in the (a) SOT and (b) KPC. The typical curves of
the SCS and KC proper waters are also shown for comparison [Lin et al., 2010].
Geochemistry
Geophysics
Geosystems
G
3
G
3
HUANG ET AL.: NONHOMOGENEOUS SEAWATER Sr ISOTOPE 10.1029/2010GC003372
5of14
