The dissolved oxygen-to-argon ratio (O2/Ar) in the oceanic mixed layer has
been widely used to estimate net community production (NCP), which is the
difference between gross primary production and community respiration; it is
a measure of the strength of the biological pump. In order to obtain the
high-resolution distribution of NCP and improve our understanding of its
regulating factors in the slope region of the northern South China Sea
(SCS), we conducted continuous measurements of dissolved O2, Ar, and
CO2 with membrane inlet mass spectrometry (MIMS) during two cruises in
October 2014 and June 2015. An overall autotrophic condition was observed in
the study region in both cruises with an average Δ(O2/Ar) of
1.1 % ± 0.9 % in October 2014 and 2.7 % ± 2.8 % in
June 2015. NCP was on average 11.5 ± 8.7 mmol C m-2 d-1 in
October 2014 and 11.6 ± 12.7 mmol C m-2 d-1 in June 2015.
Correlations between dissolved inorganic nitrogen (DIN), Δ(O2/Ar), and NCP were observed in both cruises, indicating that NCP is
subject to the nitrogen limitation in the study region. In June 2015, we
observed a rapid response of the ecosystem to the episodic nutrient supply
induced by eddies. Eddy-entrained shelf water intrusion, which supplied
large amounts of terrigenous nitrogen to the study region, promoted NCP in
the study region by potentially more than threefold. In addition, upwelling
brought large uncertainties to the estimation of NCP in the core region of
the cold eddy (cyclone) in June 2015. The deep euphotic depth in the SCS and
the absence of correlation between NCP and the average photosynthetically
available radiation (PAR) in the mixed layer in the autumn indicate that
light availability may not be a significant limitation on NCP in the SCS.
This study helps us to understand the carbon cycle in the highly dynamic shelf
system.
Introduction
Oceanic carbon sequestration is partially regulated by the production
and export process of biological organic carbon in the surface ocean. Net
community production (NCP) corresponds to gross primary production (GPP)
minus community respiration (CR) in the water (Lockwood et al., 2012) and is
an important indicator of carbon export. At steady state, NCP is equivalent
to the rate of organic carbon export and is a measure of the strength of
the biological pump (Lockwood et al., 2012). NCP effectively couples the carbon
cycle and oxygen (O2) production through photosynthesis and respiration
in the euphotic layer; thus, many previous studies have measured the mass
balance of O2 to quantify NCP (e.g., Emerson et al., 1991; Hendricks et
al., 2004; Huang et al., 2012; Reuer et al., 2007). Argon (Ar), a biological
inert gas, was commonly used to normalize the O2 concentration in these studies. Based on the similar solubility properties of O2 and Ar,
the oxygen-to-argon ratio (O2/Ar) can remove the influences of physical
processes (i.e., temperature and pressure change, bubble injection) on the
mass balance of O2 (Craig and Hayward, 1987). Dissolved O2/Ar has
been developed as a proxy for NCP in a water mass (Kaiser et al., 2005).
Biological production in the open oceans (i.e., Southern Ocean, Pacific,
Arctic Ocean) has been inferred using the O2/Ar ratio to estimate NCP
in numerous research studies (e.g., Hamme et al., 2012; Lockwood et al., 2012;
Ulfsbo et al., 2014; Shadwick et al., 2015; Stanley et al., 2010). During
recent years, several high-resolution measurements of O2/Ar and NCP in
coastal waters have been reported (Tortell et al., 2012,
2014; Eveleth et al., 2017; Izett et al., 2018). Despite the coastal waters
such as shelves and estuaries only accounting for 7 % of the global ocean
surface area, they are known to contribute 15 %–30 % of the total
oceanic primary production (Bi et al., 2013; Cai et al., 2011) and play an
important role in the marine carbon cycle and production. However, these regions
still suffer from low-resolution measurements that cannot provide
representative high-resolution NCP data.
The South China Sea (SCS) is one of the largest marginal seas in the world,
with complex ecological characteristics. River runoff from the Pearl and
Mekong River introduces large amounts of dissolved nutrients into the SCS
(Ning et al., 2004). Due to the influence of seasonal monsoons, the surface
circulation in the SCS changes from a basin-scale cyclonic gyre in winter to
an anticyclonic gyre in summer (Hu et al., 2000). The surface water masses
on the northern slope of the SCS can be categorized into three regimes: shelf
water, offshore water (e.g., the intruded Kuroshio water), and the SCS water
(Feng, 1999; Li et al., 2018). The shelf water is mixed with fresh water
from rivers or coastal currents and thus usually has low salinity (S<33) and low density (Uu and Brankart, 1997; Su and Yuan, 2005;
Cheng et al., 2014). Both offshore water and SCS water originate from the
northern Pacific. Thus, offshore water has similar hydrographic
characteristics of high temperature and high salinity as the northern
Pacific water. But the SCS water has changed a lot in its hydrographic
property because of mixing processes, heat exchange, and precipitation
during its long residence time of about 40 years in the SCS (Feng et al.,
1999; Li et al., 2018; Su and Yuan, 2005). The distributions of
phytoplankton and primary productivity in the SCS show great temporal and
spatial variation (Ning et al., 2004). Low chlorophyll a (Chl a) and primary
production are the significant characteristics of the SCS basin, which is
considered an oligotrophic region, and macronutrients (i.e., nitrogen) are
the main limitations on phytoplankton growth and productivity (Ning et al.,
2004; Lee Chen, 2005; Han et al., 2013). Excessive runoff from the Pearl
River can result in high N/P (nitrogen / phosphorus) ratios of >100, shifting the nutritive state from nitrogen deficiency to phosphorus
deficiency in the coastal region of the SCS (Lee Chen and Chen, 2006). Dissolved
iron is also a potential limitation on primary production, especially in
high-nutrient low-chlorophyll (HNLC) regions (Cassar et al., 2011). But on
the northern slope of the SCS, the concentration of dissolved iron is high
enough to support the growth of phytoplankton in the surface water (Zhang et
al., 2019). The northern slope of the SCS is an important transition region
between the coastal area and the SCS basin. In the summer, the shelf water
intrusion is an important process changing the nutritive state in the
northern slope region of the SCS (He et al., 2016; Lee Chen and Chen, 2006).
But so far, the NCP enhancement caused by this process is still unknown.
Previous studies about the organic carbon export in the SCS were mostly
conducted on particulate organic carbon (POC) flux (e.g., Bi et al., 2013;
Cai et al., 2015; Chen et al., 1998, 2008; Ma et al., 2008, 2011). Little research has been conducted on NCP in the SCS to date.
Chou et al. (2006) estimated NCP in the northern SCS during the summertime
to be 4.47 mmol C m-2 d-1 based on the time change rate of
dissolved inorganic carbon (DIC) in the mixed layer at the South East Asia
Time Series Station (SEATS) from 2002 to 2004. Wang et al. (2014) used GPP
and CR data from a light–dark bottle incubation experiment to calculate NCP
in the northern SCS and obtained a range from -179.0 to 377.6 mmol O2 m-2 d-1 (-129.7 to 273.6 mmol C m-2 d-1). Huang et
al. (2018) estimated monthly NCP from July 2014 to July 2015 based on in
situ O2 measurements on an Argo profiling float and reported the
cumulative NCP to be 0.29 mol C m-2 month-1 (9.67 mmol C m-2 d-1) during the northeast monsoon period and 0.17 mol C m-2 month-1 (5.67 mmol C m-2 d-1) during the southwest monsoon
period in the SCS basin. However, most of these studies in the SCS were
constrained by methodological factors attributed to discrete sampling and
cannot reveal rapid productivity responses to the highly dynamic
environmental fluctuations of coastal systems. Discrete sampling suffers
from low spatial resolution and cannot adequately resolve variabilities
caused by small-scale physical or biological processes in the dynamic marine
systems. In addition, each of the three methods for NCP estimates mentioned
above has its limitation. DIC-based NCP estimates are not suitable for
coastal regions because instead of biological metabolism, terrestrial
runoff can be the strongest factor influencing DIC in a coastal system
(Mathis et al., 2011). The unavoidable differences between in situ
circumstances and on-deck incubation conditions can introduce uncertainties to
NCP derived from light–dark bottle incubation (Grande et al., 1989).
Though Argo profiling floats partly eliminate the limitations of discrete
sampling, it is hard to control their movement in the study region. However, no
high-resolution measurements of NCP have been reported for the SCS so far.
In this paper, we present high-resolution NCP estimates in the northern
slope region of the SCS based on continuous shipboard dissolved O2/Ar
measurements. We discuss the regulating factors of NCP based on ancillary
measurements of other hydrographic parameters. Our high-resolution
measurements caught the rapid response of the ecosystem to the episodic
nutrient supply induced by eddies and helped us to quantify the contribution
of eddy-entrained shelf water intrusion to NCP in the summer cruise.
MethodsContinuous underway sampling and measurement
Continuous measurements of dissolved gases (O2, Ar, and CO2) were
obtained using membrane inlet mass spectrometry (MIMS; HPR 40, Hiden
Analytical, UK) (Tortell, 2005) onboard the RV Nanfeng during two cruises in the
northern slope region of the SCS (Fig. 1a, b) from 13 to 23 October 2014 and from 13 to 29 June 2015. In addition, a
cyclonic–anticyclonic eddy pair was observed in June 2015 (Fig. 1c) and
resulted in dramatic influences on the study region.
Cruise tracks of two cruises in the slope region of the
northern South China Sea in (a) October 2014 and (b) June 2015. The sea level height anomaly (SLA) and geostrophic current during
observations in June 2015 (Chen et al., 2016) are shown in (c). The
black dots and stars represent the locations of the CTD casts. Red numbers
indicate transects, while black numbers indicate the serial number of CTD
stations based on the cruise plan. The color scale in (a)
and (b) represents bathymetry.
We developed a continuous shipboard measurement system for dissolved gases
following the method described by Guéguen and Tortell (2008). Surface
seawater was collected continuously using the ship's underway intake system
(∼ 5 m depth) and was divided into different lines for various
underway scientific measurements. Seawater from the first line passed
through a chamber at a flow rate of 2–3 L min-1 to remove macroscopic
bubbles and to avoid pressure bursts. A flow of ∼ 220 mL min-1 was continuously pumped from the chamber using a Masterflex
peristaltic pump equipped with L/S® multichannel cartridge
pump heads (Cole–Parmer). In order to minimize the O2/Ar fluctuations
due to temperature effects and water vapor pressure variations, the water
samples flowed through a stainless-steel coil (∼ 6 m) with 0.6 mm wall thickness immersed in a water bath (Shanghai Bilon Instrument Co.
Ltd, China) to achieve a constant temperature (∼ 2 ∘C below the sea surface temperature), which avoided
temperature-induced supersaturation and subsequent bubble formation. Then
the water samples were introduced into a cuvette with a silicone membrane
mounted on the inside. The analyte gases were monitored by a Faraday cup
detector in the vacuum chamber after diffusion through the silicone
membrane, and the signal intensities at the relevant mass-to-charge (m/z)
ratios (32, 40, and 44 for O2, Ar, and CO2, respectively) were
recorded by MASsoft. Based on the continuous measurement of 50 L of
air-equilibrated seawater, the long-term signal stability (measured as the
coefficient of variation) over 12 h was 1.57 %, 3.75 %, and 2.21 %
for O2, Ar, and CO2, respectively. Seawater from the second line
passed through a flow chamber, where an RBR Maestro (RBR, Canada) was
installed to continuously record temperature, salinity, dissolved oxygen
(DO), and Chl a. We did not obtain continuous DO data in October 2014 because
the DO sensor of the RBR Maestro broke down. A third line was used to drain the excess
seawater. Underway pipelines were flushed with fresh water or bleach every
day to avoid possible in-line biofouling. The data from the underway
transects were exported to spreadsheets and compiled into 5 min averages,
and the comparisons of the gas data with other hydrographic variables were
based on the UTC time recorded for each measurement.
The O2/Ar ratio measurements were calibrated with air-equilibrated
seawater samples at about 6–8 h intervals to monitor instrument drift and
calculate Δ(O2/Ar). These air-equilibrated seawater samples
were prefiltered (0.22 µm) and bubbled with ambient air for at least 24 h to reach equilibrium at sea surface temperature (Guéguen and Tortell,
2008). For calibration, 800 mL of air-equilibrated seawater sample was
transferred into glass bottles and immediately drawn into the cuvette, where
the first 200 mL of the sample was used to flush the cuvette and pipelines.
After a 3 min recirculation of the sample, the average signal intensity was
obtained to calculate O2/Ar. During the course of measurements, flow
rate and the temperature of the water bath were both kept the same as in the
underway measurements. The precision of MIMS-measured O2/Ar was 0.22 % based on analyses of 20 duplicate samples in the laboratory test,
which is comparable to previous studies and sufficient to detect
biologically driven gas fluctuations in seawater (Tortell, 2005).
The instrumental CO2 ion current was calibrated at about 12–24 h
intervals using equilibrated seawater standards as per Guéguen and
Tortell (2008) during the survey in June 2015. Prefiltered seawater (0.22 µm) was gently bubbled with dry CO2 standards (200, 400, and 800 ppm; provided by the Chinese National Institute of Metrology) at in situ
temperature. After 2 d of equilibrium, these standards were analyzed by
MIMS following the same procedure for measuring air-equilibrated seawater
samples to obtain a calibration curve between CO2 signal intensity and
mole fraction. The reproducibility of these measurements was better than 5 % within 15 d. Then we used the empirical equations reported by
Takahashi et al. (2009) to convert the CO2 mole fraction derived from
the calibration curve to the in situ partial pressure of CO2
(pCO2).
Chlorophyll a (Chl a) data from the RBR sensor were linearly calibrated
against extracted Chl a measurements of discrete seawater samples taken from
the same seawater outlet as for MIMS measurements. Samples were filtered
through polycarbonate filters (0.22 µm). The filter membranes were then
packed with pre-sterilized aluminum foil and stored in a freezer (-20∘C) until extraction by acetone and analysis using a fluorimetric
method (F-4500, HITACHI, Japan) described by Parsons et al. (1984). The mean
residual of this calibration was 0.00 ± 0.07 µg L-1.
Estimation of NCP based on O2/Ar
measurements
NCP in the mixed layer was estimated by the O2/Ar mass balance from
continuous measurements. Due to similar physical properties of O2 and
Ar, Δ(O2/Ar) is used as a proxy for biological O2
supersaturation and is defined as (Craig and Hayward, 1987)
Δ(O2/Ar)=[O2]/[Ar][O2]/[Ar]eq-1,
where [O2]/[Ar] is the measured dissolved O2/Ar ratio of the mixed
layer and ([O2]/[Ar])eq is the measured dissolved O2/Ar ratio
of the air-equilibrated seawater samples. Δ(O2/Ar) is the
percent deviation of the measured O2/Ar ratio from equilibrium.
Assuming a steady state and negligible physical supply, NCP is the air–sea
biological O2 flux and can be estimated as (Reuer et al., 2007)
NCP(mmolCm-2d-1)≈kO2⋅[O2]sat⋅Δ(O2/Ar)⋅rC:O2⋅ρ,
where kO2 is the weighted gas transfer velocity of O2 (m d-1); [O2]sat denotes the saturation concentration of
dissolved O2 (µmol kg-1) in the mixed layer, which is
calculated based on temperature and salinity (Weiss, 1970); rC:O2 is
the photosynthetic quotient of C and O2 and was reported as 1 : 1.38 in
the SCS (Jiang et al., 2011); and ρ is seawater density in units of kilograms per cubic meter (kg m-3) (Millero and Poisson, 1981). We estimated kO2 using the
European Centre for Medium-Range Weather Forecasts (ECMWF) wind-speed
reanalysis data product with a 0.25∘× 0.25∘
grid (https://www.ecmwf.int, last access: 18 April 2020), the parameterization by Wanninkhof (1992), and
the gas exchange weighting algorithm by Teeter et al. (2018). Teeter et al. (2018) pointed out that the modern O2/Ar method does not strongly rely on
the steady-state assumption. When this assumption is violated, our estimate
does not represent the actual daily NCP but rather an estimate of NCP
weighted over the residence time of O2 in the mixed layer and along the
path of the water parcel during that period. Thus, the residence time of
O2 in the mixed layer is an important implication of the weighted
timescale of NCP before the measurement of O2/Ar. The residence time of
O2 (τ, d) in the mixed layer is estimated as the ratio of mixed
layer depth (MLD, m) to the gas transfer velocity of O2 (kO2,
m d-1) (Jonsson et al., 2013).
Ancillary measurements and calculations
Surface water samples for the nutrient analysis were collected from Niskin
bottles mounted on the conductivity–temperature–depth (CTD) instrument, where the samples were filtered through
acid-cleaned acetate cellulose filters (pore size: 0.4 µm). The
filtrates were poisoned by HgCl2 and stored in the dark at 4 ∘C. In the laboratory, the nutrients were photometrically determined
by an auto-analyzer (QuAAtro, SEAL Analytical, Germany) with
a precision better than 3 %. MLD was defined by the Δσt=0.125 kg m-3 criterion (Monterey and Levitus, 1997). The
subsurface chlorophyll maximum layer (SCML) was observed using the
fluorescence sensor mounted on the CTD. The SCML usually occurs at the bottom of
the euphotic layer (Hanson et al., 2007; Teira et al., 2005).
Because no PAR (photosynthetically available radiation) profile data were
obtained in two cruises, we decided to regard the depth of the SCML as the
euphotic depth (Zeu). Both MLD and Zeu were calculated at each
station where the vertical CTD casts were made. The MLDs for underway data
between CTD stations were calculated using linear interpolation based on the
distance between the underway points and nearest CTD stations. We matched
the underway data to each CTD location using a combination of
the latitude–longitude threshold (latitude–longitude of CTD station ± 0.05∘) and time threshold (end and start of stationary time ± 1 h), then took the averages of these underway data for further analysis with
discrete nutrient concentrations.
The daily satellite chlorophyll data were obtained from the EU Copernicus
Marine Service Information website (https://resources.marine.copernicus.eu, last access: 1 November 2020).
The product we used was provided by the ACRI–ST company (Sophia Antipolis,
France), with a space–time interpolation (“cloud free”). The
M_Map package for MATLAB was applied to output satellite
chlorophyll images (Pawlowicz, 2020). Daily and 8 d PAR data collected by
the MODIS Aqua sensor were obtained from NASA's ocean color website
(https://oceancolor.gsfc.nasa.gov/l3, last access: 3 November 2020). The spatial resolution of both
satellite products is 4 km, and we match the satellite PAR to CTD locations
by choosing the closest PAR data point to the CTD location. A light
attenuation coefficient (Kd, m-1) was used to estimate the average
PAR in the mixed layer (Kirk, 1994; Jerlov, 1976):
Kd=4.605Zeu.
Results and discussionDistributions of hydrographic parameters and gases
The distributions of temperature, salinity, Chl a, and Δ(O2/Ar)
during the autumn cruise (October 2014) are shown in Fig. 2. Sea surface
temperature (SST) ranged from 26.96 to 28.53 ∘C
with an average of 27.82 ± 0.33 ∘C. Sea surface salinity
(SSS) ranged from 33.28 to 34.11, with low values occurring in the
southeast of the region. Chl a concentration ranged from 0.01 to 0.71 µg L-1 and was on average 0.18 ± 0.13 µg L-1,
which is comparable to the 11-year mean value (∼ 0.2 mg m-3) in the same region in October reported by Liu et al. (2014).
Δ(O2/Ar) values were in the range of -2.9 %–4.9 % (average 1.1 % ± 0.9 %) and slightly oversaturated in most areas (Fig. 2d).
Please note that all averages we have published in this paper are reported
in the format of mean ± standard deviation.
Surface distributions of (a) temperature,
(b) salinity, (c) chlorophyll a (Chl a), and (d)Δ(O2/Ar) in October 2014.
In June 2015, SST ranged from 29.28 to 32.24 ∘C and
was on average 30.88 ± 0.59 ∘C (Fig. 3a). SSS
ranged from 30.81 to 34.16. Transect 3 was significantly characterized by
low salinity (Fig. 3b). He et al. (2016) reported that this phenomenon was
influenced by the eddy-entrained Pearl River plume (shelf water) injected
into the SCS. Chl a varied in a range of 0.09–0.58 µg L-1 in
the study region. Under the influence of this eddy-entrained shelf water,
Chl a values higher than 0.30 µg L-1 were observed along Transect 3 (Fig. 3c). In contrast, Chl a was in the range of 0.09–0.18 µg L-1 along Transects 1 and 2. It was obvious that DO was much higher on
the east side than the west side in the study region (Fig. 3d). Δ(O2/Ar) ranged from -3.9 % to 13.6 %. Most of the Δ(O2/Ar) values were positive in the study region (average 2.7 % ± 2.8 %), whereas the negative values were concentrated along
Transect 4 (Fig. 3f). Δ(O2/Ar) along Transect 3 was on
average 7.2 % ± 2.6 %, significantly higher than that of
other transects (Fig. 3f). pCO2 exhibited a high degree of spatial
and temporal variability, and the high values mostly occurred on the west
side of the study region (Fig. 3e). Resulting from the considerably low
pCO2 in Transect 3, the average pCO2 (323 ± 93 µatm) in
the study region was lower than values reported previously, i.e., 350–370 µatm by Zhai et al. (2009) and 340–350 µatm by Rehder and Sues (2001). Due to the influence of the shelf water, the average pCO2 in
Transect 3 was 222 ± 33 µatm, with a range of 144–321 µatm.
In the summer, shelf water mixed with the Pearl River plume is the most
important factor influencing pCO2 in the coastal and shelf region of the
northern SCS, which can result in pCO2 values as low as 150 µatm (Li et al., 2020). Here we apply an average atmospheric pCO2 of 382 µatm observed in July 2015 in the northern SCS (Li et al., 2020)
to calculate the pCO2 difference (ΔpCO2) between the surface
water and the atmosphere. ΔpCO2 ranged from -238 to -61 µatm along Transect 3, indicative of a strong CO2 sink.
Surface distributions of (a) temperature,
(b) salinity, (c) chlorophyll a (Chl a), (d) dissolved oxygen (DO), (e)pCO2, and (f)Δ(O2/Ar) in June 2015.
Mixed layer depth, euphotic depth, and residence time of
O2 in the mixed layer
The MLD, euphotic depth (Zeu), and residence time of O2 (τ)
in the mixed layer at CTD stations during two cruises are shown in Tables 1 and
2. In autumn 2014, MLD ranged from 27 to 81 m, with an average of 55 ± 15 m (Table 1). The average Zeu was 74 ± 12 m, approximately 20 m
deeper than the MLD (Table 1). The residence time of O2 in the mixed layer
ranged from 3 to 13 d (Table 1), comparable to a range of 1–2 weeks
reported by previous studies (Izett et al., 2018; Manning et al., 2017). The
average residence time of O2 was 9 ± 3 d, indicating that our
estimate generally quantified NCP over 9 d prior to the underway
observation of O2/Ar during this cruise.
Basic information at all CTD stations in October 2014.
StationDate ofMLDZeubkcτdobservationa(m)(m)(m d-1)(d)O-0113 Oct 201458824.712O-0213 Oct 201464745.212O-0314 Oct 201456846.29O-0414 Oct 201454766.39O-0520 Oct 201427707.93O-0619 Oct 201455628.47O-0721 Oct 201440607.35O-0821 Oct 201449727.47O-0915 Oct 201479966.213O-1015 Oct 201468816.111O-1115 Oct 201464815.412O-1216 Oct 201466745.213O-1316 Oct 201448526.38O-1417 Oct 201454626.98O-1522 Oct 201449687.07O-1622 Oct 201450737.37O-1723 Oct 201452757.97O-1918 Oct 201431649.43O-2018 Oct 201435618.74O-2118 Oct 201481866.912O-2217 Oct 2014761026.013
a All dates are in the format of day month year. b Euphotic depth, defined based on subsurface chlorophyll maximum layer. c Gas transfer velocity of O2. d Residence time of O2 in the mixed layer, estimated as per MLD/k.
Basic information at all CTD stations in June 2015.
StationDate ofMLDZeukτobservation(m)(m)(m d-1)(d)J-0118 Jun 201526632.212J-0217 Jun 201519801.910J-0316 Jun 201520741.911J-0415 Jun 201522741.911J-0515 Jun 201511781.29J-0614 Jun 201524762.111J-0713 Jun 201521812.39J-0818 Jun 201514561.78J-0919 Jun 201517591.610J-1019 Jun 20158461.46J-1120 Jun 20158402.83J-1221 Jun 201516453.05J-1321 Jun 201519452.38J-1424 Jun 201528554.07J-1524 Jun 201517425.33J-1625 Jun 201510195.72
The average MLD in June 2015 was just 18 ± 6 m (Table 2). A significantly
shallow MLD occurred at two stations (J-10, J-11) located in Transect 3
(Table 2, Fig. S1f in the Supplement). The low-salinity shelf water intrusion is the main
cause of this shallow MLD of 8 m. The average Zeu was 58 ± 18 m,
approximately 40 m deeper than the MLD (Table 2). The residence time of O2
in the mixed layer ranged from 2 to 12 d (Table 2), indicating a fast gas
exchange at some stations. In addition, we also observed relatively obvious
subsurface O2 maxima in Transects 1 and 2 in summer 2015. But this
phenomenon did not exist in autumn 2014.
In both cruises, Zeu was observed to be obviously deeper than the MLD.
This result partly suggests that light availability may not be a limitation
on NCP in the northern slope of the SCS. Especially in the summer, Zeu extended to 2–7 times the MLD (Table 2), ensuring sufficient illumination
in the mixed layer. But in the autumn when the thickness of the mixed layer
accounts for about 74 % of the euphotic layer, the average light intensity in
the mixed layer might be influenced by exponential light attenuation
along the depth profile.
NCP in autumn and summer
In October 2014, NCP in the northern slope of the SCS ranged from -29.2 to
42.7 mmol C m-2 d-1 (average 11.5 ± 8.7 mmol C m-2 d-1), and most of the region was net autotrophic (Fig. 4a). The
estimated NCP based on the O2/Ar values measured in this cruise is
about 34 % of the net primary production rates of 34.3 mmol C m-2 d-1 measured by 14C bottle incubation (Xiaoxia Sun, personal
communication, 2017), which was in agreement with previous research (Quay et al.,
2010).
Surface distribution of NCP along the northern slope of
the SCS during the cruise in (a) October 2014 and (b) June 2015.
The average NCP in the study region was 11.6 ± 12.7 mmol C m-2 d-1 with a range of -27.6–61.4 mmol C m-2 d-1 in June
2015. A high NCP level was observed along Transect 3 (Fig. 4b).
Eddy-entrained shelf water brought a large amount of terrigenous nutrients
from the shelf to the slope region along Transect 3 (He et al., 2016). The
average nitrate (NO3-) and nitrite (NO2-) concentrations
in the surface water of Transect 3 were 2.31 ± 0.70
and 0.04 ± 0.01 µmol L-1, respectively (Fig. S1a, b);
both values were much higher than those found in the other three transects,
for which NO3- was in a range of < 0.03–0.69 µmol L-1 and NO2- was mostly below the detection limit. Li et al. (2018) reported that all of Transect 3 and part of Transect 4 were
dominated by shelf water at the surface, and we estimated NCP over these
regions where salinity is lower than 33 as 23.8 ± 10.7 mmol C m-2 d-1 on average. We also observed a warm eddy (anticyclone) covering
most stations in Transects 1 and 2 (Fig. 1b, c) during our survey in June
2015 (Chen et al., 2016). Anticyclonic eddies can cause downwelling,
deepening of the thermocline, and blocking of the supply of nutrients from
the deeper water (Ning et al., 2008; Shi et al., 2014). Consequently, a warm
eddy is expected to result in an oligotrophic condition in the surface water
associated with low Chl a concentrations and low production (Ning et al.,
2004). As a result, in the summer of 2015, the observed NO2-,
NO3-, and PO43- (phosphate) concentrations were almost
below the detection limit in Transects 1 and 2 (Fig. S1a, b, d). NCP
in Transect 1 and 2 was at a very low level (average 2.8 ± 2.7 mmol C m-2 d-1). Because of the significantly high values of NCP over the
regions with shelf water intrusion, our NCP result in the summer of 2015 is
on average higher than the previous values of 4.47 mmol C m-2 d-1 and 0.17 mol C m-2 month-1 (5.67 mmol C m-2 d-1) based on the DIC budget and Argo O2, respectively, in the
SCS (Chou et al., 2006; Huang et al., 2018). However, NCP estimates based on
both methods mentioned above suffer from poor temporal and spatial coverage
and do not allow for revealing rapid changes in shelf systems. In contrast,
continuous measurements of O2/Ar allow us to capture rapid variations
in NCP along Transect 3 and resolve short-term productivity responses to
environmental fluctuations.
Distribution of various parameters along representative
transects
We chose Transect 5 (Fig. 1a) observed in October 2014 and Transect 4
(Fig. 1b) observed in June 2015 to show the distribution of various
parameters.
The distribution of Chl a, Δ(O2/Ar), and NCP showed a similar
trend along Transect 5 in October 2014 (Fig. 5). There was a trough of
temperature, showing a maximum drawdown of ∼ 0.6 ∘C
compared to the average temperature in the study region (Figure 5a). But the
temperature fluctuations shown here are too small to reflect a significant
upwelling that can easily cause ∼ 2 ∘C of temperature drawdown in the upper layer (Jing et al., 2009; Manning et al., 2017; Ning
et al., 2004). A spike of Chl a occurred between 115.6 and
115.7 ∘E and was coincident with the peaks of Δ(O2/Ar) and NCP (Fig. 5b, c). The highest surface concentration of
ammonium (NH4+) of 0.35 µmol L-1 was also observed
between 115.6 and 115.7 ∘E in this transect and was
predominantly higher than the concentrations (0.07–0.17 µmol L-1)
in the other regions during this cruise (Fig. 5c, S2b). Because no significant
obduction processes (i.e., upwelling, entrainment, and diapycnal mixing)
were reported in this region, the most likely source of this abundant
NH4+ was in situ regeneration such as the excretion of zooplankton
and the bacterial decomposition of organic matter (La Roche, 1983; Clark et
al., 2008). Theoretically, NH4+, an important nitrogen source of
phytoplankton growth, can be quickly utilized by phytoplankton and
contributes to primary production (Dugdale and Goering, 1967; Tamminen,
1982). However, we only got nutrient data at two CTD stations in this
transect; thus, the result we obtained here just indicated that high NCP
occurred at the station with a relatively high NH4+ concentration,
but this is not strong evidence that NH4+ was the main factor
influencing NCP in this transect.
Zonal variations in (a) temperature, salinity,
(b)Δ(O2/Ar), (c) Chl a, NCP, and the surface
concentration of ammonia (NH4+) along Transect 5 in October 2014.
The plots of Δ(O2/Ar) and NCP are 10-point
Savitzky–Golay smoothed to give a better view of their
distribution.
A similar distribution pattern of Chl a, NCP, and Δ(O2/Ar) was
observed along Transect 4 in June 2015, whereas pCO2 showed the opposite trend for these three parameters (Fig. 6b, c). Low salinity
(lower than 33) existed at both the southern and northern ends of this transect
(Fig. 6a). The concentration of dissolved inorganic nitrogen (DIN,
NO3-+ NO2-+ NH4+) in the surface water
was 0.81 and 0.27 µmol L-1 at the southern and
northern end, respectively, which was higher than the concentrations at other
stations for this transect (Fig. 6c). These results indicate that shelf
water is imported at the northern and southern ends of this transect, along
with higher levels of Chl a and NCP (Fig. 6c). A sharp drop in the
temperature and an increase in salinity occurred from 19.7 to
19.8∘ N and from 21 to 20.7∘ N (Fig. 6a),
manifesting an upwelling over this area together with dramatic spikes in
pCO2 and an associated decrease in Δ(O2/Ar) (Nemcek et al.,
2008) (Fig. 6b). Most regions of Transect 4 were dominated by upwelling
and showed a negative sea level height anomaly (Chen et al., 2016; He et al.,
2016). A localized cold eddy was considered the cause of this upwelling
(Fig. 1c), resulting in a maximum temperature drawdown of ∼ 1.6 ∘C in the mixed layer.
Meridional variations in (a) temperature,
salinity, (b)Δ(O2/Ar), pCO2, (c) Chl a,
NCP, and the surface concentration of DIN along Transect 4 in June 2015. The
plots of Δ(O2/Ar), pCO2, and NCP are 10-point
Savitzky–Golay smoothed.
Vertical mixing is considered the largest source of error in
O2/Ar-based NCP estimates because upwelled subsurface water with
different O2/Ar signatures can produce either an overestimation or an
underestimation of NCP in the mixed layer (Cassar et al., 2014; Izett et
al., 2018). Previous research usually ignored the underestimated negative
NCP caused by vertical mixing (Giesbrecht et al., 2012; Reuer et al.,
2007; Stanley et al., 2010). Cassar et al. (2014) presented an N2O-based
correction method of O2/Ar and NCP for vertical mixing. Although this
method has been successfully adopted by Izett et al. (2018) in the sub-Arctic
northeast Pacific, it is not suitable for our study region. This is because
it is basically applicable in areas where the depths of the euphotic zone
and mixed layer are similar, and this method is not suitable for
oligotrophic regions (Cassar et al., 2014). The SCS is recognized as an
oligotrophic region, and the depth of the euphotic zone can be 2–7 times
that of the mixed layer in our study region in the summer. In addition, in
the region (e.g., the SCS basin) of the subsurface oxygen maximum, the
applicability of N2O-based correction is limited (Izett et al.,
2018). In Transect 4, the regions with negative NCP and the regions with
salinity higher than 33.5 and temperature lower than 30 ∘C are
defined as influenced by upwelling. If we neglect these regions in Transect 4, the average NCP in June 2015 can rise slightly to 12.4 ± 12.3 mmol C m-2 d-1. If we also remove the influence of shelf water
intrusion by neglecting the regions with salinity lower than 33, the average
NCP can sharply decrease to 5.0 ± 6.2 mmol C m-2 d-1,
which is similar to the results of 4.47 mmol C m-2 d-1 and 0.17 mol C m-2 month-1 (5.67 mmol C m-2 d-1) reported in
previous research in the same season (Chou et al., 2006; Huang et al.,
2018). Here we regard 5.0 ± 6.2 mmol C m-2 d-1 as the
background value of NCP in the study region. Since an average NCP of 23.8 ± 10.7 mmol C m-2 d-1 was observed over regions with
salinity lower than 33, we can conclude that the summer shelf water
intrusion significantly promoted NCP by potentially more than threefold in
June 2015.
Factors influencing NCP in the SCS
The SCS is an oligotrophic region with low biomass and primary production
(Lee Chen, 2005; Ning et al., 2004). Previous research has shown that the
nutrient content, especially nitrogen and phosphorus, is the most important factor
controlling and limiting phytoplankton biomass and primary production in
the SCS (Ning et al., 2004; Lee Chen, 2005; Lee Chen and Chen, 2006; Han et
al., 2013). After neglecting the two CTD stations (J-14, J-15) with negative
NCP influenced by upwelling in June 2015, we performed a principal component
analysis (PCA) to determine the dominant factors influencing NCP in both
cruises. In October 2014, DIN (0.741), Δ(O2/Ar) (0.858), and
NCP (0.979) were significantly loaded on Factor 1, indicating a potential
relationship among these three variables (Fig. 7a, Table S1b in the Supplement). The
correlation coefficient between DIN and NCP was 0.706 (p<0.01;
Table S1a), which was significantly higher than the coefficient between NCP
and the other variables, except for Δ(O2/Ar) and temperature;
this indicated that DIN was an important factor influencing NCP in this
cruise. Another two nutrients – dissolved silicate (DSi, SiO32-)
and dissolved inorganic phosphorus (DIP, PO43-) – had no
correlations (p>0.05) with NCP (Table S1a). In June 2015, Factor 1 showed a strong loading by DIN (0.876), Chl a (0.950), DO (0.927), Δ(O2/Ar) (0.902), and NCP (0.909), whereas salinity (-0.936) and
pCO2 (-0.908) were negatively loaded on Factor 1 (Fig. 7b, Table S2b). The injection of low-salinity shelf water appeared to have a strong
effect on the study region because significant negative correlations were
observed between salinity and DIN, Chl a, Δ(O2/Ar), and NCP
(Table S2a). DIN had strong correlations with NCP, Δ(O2/Ar),
and Chl a, with correlation coefficients of 0.747, 0.910, and 0.754,
respectively (Table S2a), indicating that DIN was the dominant factor
controlling the growth of phytoplankton and primary production in this
cruise. DSi (0.582) and DIP (-0.601) were both moderately loaded on Factor 2 (Fig. 7b, Table S2b) and had no correlations with NCP (p>0.05, Table S2a). These results suggest the key role of nitrogen in
regulating Δ(O2/Ar), NCP, and phytoplankton biomass in the SCS.
The supply of nitrogen may stimulate the growth of phytoplankton in the SCS,
and nitrogen is an important participant in photosynthesis and a basic
element that contributes to the increase in primary production (Dugdale and
Goering, 1967; Lee Chen, 2005; Lee Chen and Chen, 2006; Han et al., 2013).
Principal component analysis (PCA) among variables for
(a) October 2014 and (b) June 2015 (Bartlett's
test of sphericity: p<0.01).
Coupled with biochemical variations, physical processes also play important
roles in the slope region of the SCS by transporting abundant nutrient-rich
shelf water into the SCS and bringing deep water to the surface by enhancing
water mixing (Chen and Tang, 2012; Ning et al., 2004; Pan et al., 2012). The
surface waters in the slope region of the northern SCS are primarily
composed of waters originating from SCS water, Kuroshio water, and shelf
water (Li et al., 2018). In the summer, the shelf water exists where the
potential density anomaly is lower than 20.5 kg m-3 (Li et al., 2018).
In the autumn, there is a weak offshore transport of the shelf water in the
SCS, and the salinity of the water mixed with the shelf water is usually
lower than 33 (Fan et al., 1988; Uu and Brankart, 1997; Su and Yuan, 2005).
In October 2014, the observed surface salinity was in the range of 33.28 to
34.11; thus, the surface waters were mainly derived from mixing of the
Kuroshio water and the SCS water. In the summer of 2015, a
cyclonic–anticyclonic eddy pair was observed in the study region (Fig. 1c). Low-salinity shelf water mixed with the intruding river plume from the
Pearl River in the upper 50 m and was transported to the slope and basin
along the intersection of the two eddies (Chen et al., 2016; He et al.,
2016; Li et al., 2018). In both seasons, the surface waters in the study
region were generally found to be nitrogen-deficient, with NO2- at
< 0.01–0.04 µmol L-1 (Figs. S2a, S1b), NO3- at
< 0.03–2.82 µmol L-1 (Fig. S1a), and NH4+ at
0.04–0.35 µmol L-1 (Figs. S2b, S1c). The concentrations of
NO2- and NO3- were below the detection limit at almost
80 % of the sampling stations during both cruises. Due to the injection of
shelf water with low salinity and abundant terrestrial nutrients,
significantly high concentrations of NO3- and NO2- were
observed along Transect 3 in June 2015 (Fig. S1a, b) where the shelf
water was intruded by eddies (Chen et al., 2016; He et al., 2016). Such
transport processes from the inner shelf to the slope region have a profound
influence on nutrient dynamics and biological production (He et al., 2016).
The water that was influenced by shelf water with a potential density
anomaly lower than 20.25 kg m-3 and salinity lower than 33 had high
concentrations of DIN (Fig. 8a). At the six stations (in the red circle in
Fig. 8a) that were intruded by shelf water and characterized by surface
salinity lower than 33, we obtained an average surface DIN concentration of
1.82 ± 1.16 µmol L-1 (0.27–3.01 µmol L-1), which was significantly
higher than the mean of 0.10 ± 0.03 µmol L-1 (0.04–0.16 µmol L-1) at
other stations (independent sample t test, p<0.01). After
neglecting the two stations (J-14, J-15) influenced by upwelling, a strong
correlation between NCP and DIN was observed in the cruise of June 2015 (r=0.747, p<0.01), with higher NCP (average 15.4 ± 4.5 mmol C m-2 d-1) occurring at the stations where shelf water intruded,
consistent with the DIN concentration higher than 0.27 µmol L-1
(Fig. 8b). At other stations without the influence of shelf water, the
average NCP was just 2.3 ± 1.7 mmol C m-2 d-1. These results
further suggest that the supply of DIN from shelf water can greatly
stimulate the primary production at these stations, resulting in an NCP
increase of nearly 7 times compared to other stations.
(a)T–S diagram of the surface DIN concentration in
June 2015. The stations influenced by shelf water are in the red circle.
(b) Correlation analysis between the surface DIN concentration and NCP
at sampling stations. The stations (characterized by S<33)
influenced by shelf water presented surface DIN concentrations ≥0.27µmol L-1.
The start date and duration (Δday) of shelf water
intrusion at the stations with surface salinity lower than 33 in June 2015.
StationDate ofStart date of shelfΔday*τobservationwater intrusion(d)J-0919 Jun 201510 Jun 2015910J-1019 Jun 201513 Jun 201566J-1120 Jun 201513 Jun 201573J-1221 Jun 201513 Jun 201585J-1321 Jun 201513 Jun 201588J-1625 Jun 2015before 10 Jun 2015>152
* The difference between the date of observation and the start date of shelf water intrusion at the listed stations.
The correlations between NCP and sea surface temperature as well as between NCP and salinity also
support the influence of physical forcing on NCP. In June 2015, we obtained
a strong negative correlation between NCP and salinity (Fig. 9d). NCP
significantly increased in the water with salinity lower than 33 (Fig. 9d). Temperature had weak correlations with NCP (Fig. 9c), and the
negative NCP values were concentrated in the water with temperatures below
30.5 ∘C and salinity values over 33.5 (Fig. 9c, d). This
surface water was mostly observed along Transect 4 where vertical mixing
caused by a cold eddy brought deep water to the surface. The undersaturated
Δ(O2/Ar) entrained by deep water caused the negative NCP
estimates at the surface, resulting in a considerable underestimation of
NCP. Unlike in June 2015, all the correlations were very weak between NCP
and temperature as well as between NCP and salinity in October 2014 (Fig. 9a, b). The Kuroshio
water and the SCS water had similar hydrological characteristics, and their
mixing in October 2014 may not have resulted in significant changes in the
hydrological characteristics of the surface water.
Correlation analysis between underway NCP and physical
parameters (temperature and salinity) in October 2014 (a, b) and
June 2015 (c, d).
Daily satellite chlorophyll images on selected days
in June 2015. Stars represent CTD locations. We roughly set satellite chlorophyll to
≥0.2µg L-1 in this figure as the criterion for shelf water.
This figure was made based on the M_Map mapping package for
MATLAB (Pawlowicz, 2020).
Satellite PAR data and NCP at the selected stations in
October 2014.
StationDate ofMLDZeuSurface PARaKdML PARbNCPobservation(m)(m)(mol m-2 d-1)(m-1)(mol m-2 d-1)(mmol C m-2 d-1)O-0113 Oct 2014588242.05.6 × 10-212.03.0O-0213 Oct 2014647442.06.2 × 10-210.015.1O-0314 Oct 2014568441.15.5 × 10-212.410.1O-0821 Oct 2014497238.76.4 × 10-211.415.7O-1015 Oct 2014688140.05.7 × 10-29.84.4O-1316 Oct 2014485239.28.9 × 10-28.715.3O-1522 Oct 2014496838.66.8 × 10-210.816.3O-2018 Oct 2014356139.27.5 × 10-213.316.4O-2217 Oct 20147610242.24.5 × 10-211.615.7
a Average surface PAR over the residence time of O2 in the mixed layer. b Average PAR in the mixed layer.
The nutrient concentrations and hydrographic characteristics we observed
just reflect the marine environment at the moment of sampling, partly
contradicting our estimates that quantified NCP over a period prior to the
observation. Especially for the regions with a significant influence of shelf
water in June 2015, tracking the history of shelf water intrusion is
important. We used daily satellite chlorophyll data to monitor the intrusion
of shelf water and roughly set satellite chlorophyll to
≥ 0.2 µg L-1 as the criterion for shelf water (Fig. 10). On 10 June 2015, shelf water began to influence the northern end (J-9) of Transect 3
and most part of Transect 4; then it extended to the southern end of
Transect 3 and Transect 4 where J-12 and J-13 were located on 13 June (Figs. 1b, 10). Until 25 June when we finished the observation of Transect 4, the entirety of
Transect 3 (J-9 to 12) as well as J-13 and J-16 had been dominated by
shelf water for more than 10 d (Figs. 1b, 10). We report these
findings in Table 3, along with the residence time (τ) of O2 in
the mixed layer and the difference (Δday) between the date of
observation and the start date of shelf water intrusion at the stations with
surface salinity lower than 33. Δday can represent the duration of
the shelf water intrusion at each station before our observation. The
residence time of O2 in the mixed layer at most stations listed in
Table 3 is shorter than or equivalent to Δday. This result suggests
that our estimate has appropriately integrated the NCP during the period of
shelf water intrusion, which can effectively reflect the influence of shelf
water on the productive state of the northern slope of the SCS in the summer.
The amount of light may also play a role in the extent of primary
production. The MLD is considered a driver of light availability in the
mixed layer (Cassar et al., 2011; Hahm et al., 2014). The euphotic layer was
on average 40 m thicker than the mixed layer in the study region during the
summer cruise; thus, it is not very significant to discuss the light
limitation in June 2015. We conducted an analysis of light availability
based on daily satellite PAR data and NCP in October 2014. To minimize the
influence of DIN concentrations, we selected nine stations with surface DIN
concentrations in the range of 0.10–0.17 µmol L-1. The average
surface PAR (mol m-2 d-1) at each station was integrated over the
residence time of O2 before our observation. Then an average PAR in the
mixed layer was calculated based on Kd. At the selected stations, the
surface PAR varies over a range of 38.6–42.2 mol m-2 d-1, while
the average PAR in the mixed layer (ML PAR) ranged from 8.7 to 13.3 mol m-2 d-1 (Table 4). There is no significant correlation between the
average PAR and NCP in the mixed layer (Table 4), partly suggesting that
light intensity may not be a factor for NCP in the autumn. Light availability
in the northern slope region of the SCS is enough to support the primary
production of phytoplankton.
Conclusion
The distribution of Δ(O2/Ar) and NCP on the northern slope of
the SCS was strongly affected by nutrient availability, especially nitrogen.
The nitrogen limitation on NCP was found both in the autumn and summer. In
June 2015, we observed strong biological responses to the supply of nitrogen
induced by eddy-entrained shelf water intrusion. NCP in the region with the
influence of shelf water was 23.8 ± 10.7 mmol C m-2 d-1 on
average, with a maximum of 61.4 mmol C m-2 d-1. In addition,
vertical mixing caused considerable underestimation of NCP in the transect
influenced by a cold eddy. Removing the regions with the influence of shelf
water intrusion and vertical mixing, the average NCP in other regions was
5.0 ± 6.2 mmol C m-2 d-1. This value agrees well with
previously published NCP estimates for the study area. Our results also
reveal the rapid response of the ecosystem to physical processes. Summer
shelf water intrusion may significantly promote NCP by potentially more than
threefold in the study region. This is the first report that quantifies the
contribution of shelf water intrusion to NCP on the northern slope of the
SCS in the summer. Because of the sufficient illumination in the tropical
SCS, light availability may not be a significant limitation on NCP in both
seasons. The high-resolution NCP estimates derived from continuous
measurement of O2/Ar presented in this paper are of significance for
understanding the carbon cycle in the highly dynamic system of the SCS.
Data availability
All data presented in this paper are available on Zenodo (10.5281/zenodo.4496886, Qin et al., 2021).
The supplement related to this article is available online at: https://doi.org/10.5194/os-17-249-2021-supplement.
Author contributions
GZ and YH designed and set up the underway measurement
system. WZ attended both cruises (in June 2015 and October 2014)
in the South China Sea and was mainly responsible for operating the
underway measurement system during the cruises. SL provided the
nutrient data from both cruises. CQ attended the cruise in June 2015
and prepared the paper with contributions from all co-authors.
Competing interests
The authors declare that they have no conflict of interest.
Acknowledgements
The authors wish to thank the crew of the RV Nanfeng for assistance with the
collection of field samples and Xiaoxia Sun for providing the
14C PP data. We would also like to thank the Ocean Biology Processing
Group (OBPG) at NASA for generating the PAR data and the EU Copernicus
Marine Environment Monitoring Service (CMEMS) for providing the satellite
chlorophyll data. Michael Bender and Bror Jonsson are acknowledged
for constructive suggestions on the continuous O2/Ar measurement system
and the calculation of O2/Ar-based NCP.
Financial support
This research has been supported by the National Science Foundation of China (grant no. 41776122), the Ministry of Science and Technology of China (grant no. 2014CB441502), the Fundamental Research Funds for the Central Universities (grant no. 201562010), the Taishan Scholars Programme of Shandong Province (grant no. 201511014), and the Aoshan Talents Programme of the Qingdao National Laboratory for Marine Science and Technology (grant no. 2015ASTP-OS08).
Review statement
This paper was edited by Mario Hoppema and reviewed by two anonymous referees.
ReferencesBi, Q., Du, J., Wu, Y., Zhou, J., and Zhang, J.: Particulate organic carbon
export flux by 234Th/238U disequilibrium in the continental slope of the East China Sea, Acta Oceanol. Sin., 32, 67–73,
10.1007/s13131-013-0303-7, 2013.Cai, P., Zhao, D., Wang, L., Huang, B., and Dai, M.: Role of particle stock
and phytoplankton community structure in regulating particulate organic
carbon export in a large marginal sea, J. Geophys. Res.-Oceans, 120,
2063–2095, 10.1002/2014JC010432, 2015.Cai, W.: Estuarine and Coastal Ocean Carbon Paradox: CO2 Sinks or Sites
of Terrestrial Carbon Incineration?, Ann. Rev. Mar. Sci., 3, 123–145,
10.1146/annurev-marine-120709-142723, 2011.Cassar, N., DiFiore, P. J., Barnett, B. A., Bender, M. L., Bowie, A. R., Tilbrook, B., Petrou, K., Westwood, K. J., Wright, S. W., and Lefevre, D.: The influence of iron and light on net community production in the Subantarctic and Polar Frontal Zones, Biogeosciences, 8, 227–237, 10.5194/bg-8-227-2011, 2011.Cassar, N., Nevison, C. D., and Manizza, M.: Correcting oceanic
O2/Ar-net community production estimates for vertical mixing using
N2O observations, Geophys. Res. Lett., 41, 8961–8970,
10.1002/2014GL062040, 2014.Chen, J., Zheng, L., Wiesner, M. G., Chen, R., Zheng, Y., and Wong, H.:
Estimations of primary production and export production in the South China
Sea based on sediment trap experiments, Chinese Sci. Bull., 43, 583–586,
10.1007/BF02883645, 1998.Chen, W., Cai, P., Dai, M., and Wei, J.: 234Th/238U disequilibrium and particulate organic carbon export in the northern South China Sea, J. Oceanogr., 64, 417–428, 10.1007/s10872-008-0035-z, 2008.Chen, Y. and Tang, D.: Eddy-feature phytoplankton bloom induced by a
tropical cyclone in the South China Sea, Int. J. Remote Sens., 33,
7444–7457, 10.1080/01431161.2012.685976, 2012.Chen, Z., Yang, C., Xu, D., and Xu, M.: Observed hydrographical features and
circulation with influences of cyclonic-anticyclonic eddy-pair in the
northern slope of the South China Sea during June 2015 (in Chinese), J. Mar.
Sci., 34, 10–19, 10.3969/j.issn.1001-909X.2016.04.002, 2016.Cheng, G., Sun, J., Zu, T., Chen, J., and Wang, D.: Analysis of water masses
in the northern South China Sea in summer 2011 (in Chinese), J. Trop.
Oceanogr., 33, 10–16, 10.3969/j.issn.1009-5470.2014.03.002, 2014.Chou, W., Lee Chen, Y., Sheu, D., Shih, Y., Han, C., Cho, C., Tseng, C., and
Yang, Y.: Estimated net community production during the summertime at the
SEATS time-series study site, northern South China Sea: Implications for
nitrogen fixation, Geophys. Res. Lett., 33, L22610, 10.1029/2005GL025365,
2006.Clark, D. R., Rees, A. P., and Joint, I.: Ammonium regeneration and
nitrification rates in the oligotrophic Atlantic Ocean: Implications for new
production estimates, Limnol. Oceanogr., 53, 52–62,
10.4319/lo.2008.53.1.0052, 2008.Craig, H. and Hayward, T.: Oxygen Supersaturation in the Ocean: Biological
Versus Physical Contributions, Science, 235, 199–202,
10.1126/science.235.4785.199, 1987.Dugdale, R. C. and Goering, J. J.: Uptake of New and Regenerated Forms of
Nitrogen in Primary Productivity, Limnol. Oceanogr., 12, 196–206,
10.4319/lo.1967.12.2.0196, 1967.Emerson, S., Quay, P., Stump, C., Wilbur, D., and Knox, M.: O2, Ar,
N2, and 222Rn in surface waters of the subarctic Ocean: Net
biological O2 production, Global Biogeochem. Cy., 5, 49–69,
10.1029/90GB02656, 1991.Eveleth, R., Cassar, N., Sherrell, R. M., Ducklow, H., Meredith, M. P.,
Venables, H. J., Lin, Y., and Li, Z.: Ice melt influence on summertime net
community production along the Western Antarctic Peninsula, Deep Sea Res.
Pt. II., 139, 89–102, 10.1016/j.dsr2.2016.07.016, 2017.
Fan, L., Su, Y., and Li, F.: Analysis on water mases in the Northern South
China Sea, Acta Oceanol. Sin., 10, 136–145, 1988 (in Chinese).
Feng, S., Li, F., and Li, S.: An introduction to marine science,
Higher Education Press, Beijing, China., 1999 (in Chinese).Giesbrecht, K. E., Hamme, R. C., and Emerson, S. R.: Biological productivity
along Line P in the subarctic northeast Pacific: In situ versus
incubation-based methods, Global Biogeochem. Cy., 26, GB3028,
10.1029/2012GB004349, 2012.Grande, K. D., Williams, P. J. L. B., Marra, J., Purdie, D. A., Heinemann,
K., Eppley, R. W., and Bender, M. L.: Primary production in the North Pacific
gyre: a comparison of rates determined by the 14C, O2
concentration and 18O methods, Deep Sea Res. Pt. A, 36, 1621–1634, 10.1016/0198-0149(89)90063-0, 1989.Guéguen, C. and Tortell, P. D.: High-resolution measurement of Southern
Ocean CO2 and O2/Ar by membrane inlet mass spectrometry, Mar.
Chem., 108, 184–194, 10.1016/j.marchem.2007.11.007, 2008.Hahm, D., Rhee, T. S., Kim, H. C., Park, J., Kim, Y. N., Shin, H. C., and
Lee, S.: Spatial and temporal variation of net community production and its
regulating factors in the Amundsen Sea, Antarctica, J. Geophys. Res.-Oceans,
119, 2815–2826, 10.1002/2013JC009762, 2014.Hamme, R. C., Cassar, N., Lance, V. P., Vaillancourt, R. D., Bender, M. L.,
Strutton, P. G., Moore, T. S., DeGrandpre, M. D., Sabine, C. L., Ho, D. T.,
and Hargreaves, B. R.: Dissolved O2/Ar and other methods reveal rapidchanges in productivity during a Lagrangian experiment in the Southern
Ocean, J. Geophys. Res.-Oceans, 117, 92–99, 10.1029/2011JC007046,
2012.Han, A. Q., Dai, M. H., Gan, J. P., Kao, S.-J., Zhao, X. Z., Jan, S., Li, Q., Lin, H., Chen, C.-T. A., Wang, L., Hu, J. Y., Wang, L. F., and Gong, F.: Inter-shelf nutrient transport from the East China Sea as a major nutrient source supporting winter primary production on the northeast South China Sea shelf, Biogeosciences, 10, 8159–8170, 10.5194/bg-10-8159-2013, 2013.Hanson, C. E., Pesant, S., Waite, A. M., and Pattiaratchi, C. B.: Assessing
the magnitude and significance of deep chlorophyll maxima of the coastal
eastern Indian Ocean, Deep. Res. Pt. II, 54, 884–901, 10.1016/j.dsr2.2006.08.021, 2007.He, X., Xu, D., Bai, Y., Pan, D., Chen, C. A., Chen, X., and Gong, F.:
Eddy-entrained Pearl River plume into the oligotrophic basin of the South
China Sea, Cont. Shelf Res., 124, 117–124, 10.1016/j.csr.2016.06.003,
2016.Hendricks, M. B., Bender, M. L., and Barnett, B. A.: Net and gross O2
production in the southern ocean from measurements of biological O2
saturation and its triple isotope composition, Deep. Res. Pt. I, 51, 1541–1561, 10.1016/j.dsr.2004.06.006, 2004.Hu, J., Kawamura, H., Hong, H., and Qi, Y.: A Review on the currents in the
South China Sea: Seasonal circulation, South China Sea warm current and
Kuroshio intrusion, J. Oceanogr., 56, 607–624,
10.1023/A:1011117531252, 2000.Huang, K., Ducklow, H., Vernet, M., Cassar, N., and Bender, M. L.: Export
production and its regulating factors in the West Antarctica Peninsula
region of the Southern Ocean, Global Biogeochem. Cy., 26, GB2005,
10.1029/2010GB004028, 2012.Huang, Y., Yang, B., Chen, B., Qiu, G., Wang, H., and Huang, B.: Net
community production in the South China Sea Basin estimated from in situ
O2 measurements on an Argo profiling float, Deep Sea Res. Pt. I, 131, 54–61, 10.1016/j.dsr.2017.11.002, 2018.Izett, R. W., Manning, C. C., Hamme, R. C., and Tortell, P. D.: Refined
Estimates of Net Community Production in the Subarctic Northeast Pacific
Derived From ΔO2/Ar Measurements With N2O-Based Corrections for Vertical Mixing, Global Biogeochem. Cy., 32, 326–350,
10.1002/2017GB005792, 2018.Jiang, Z., Huang, C., Dai, M., Kao, S., Hydes, D. J., Chou, W. and Janf, S.:
Short-term dynamics of oxygen and carbon in productive nearshore shallow
seawater systems off Taiwan: Observations and modeling, Limnol. Oceanogr.,
56, 1832–1849, 10.4319/lo.2011.56.5.1832, 2011.Jing, Z., Qi, Y., Hua, Z., and Zhang, H.:
Numerical study on the summer upwelling system in the northern continental shelf of
the South China Sea, Cont. Shelf Res., 29, 467–478,
10.1016/j.csr.2008.11.008, 2009.Jonsson, B. F., Doney, S. C., Dunne, J., and Bender, M.: Evaluation of the
Southern Ocean O2/Ar-based NCP estimates in a model framework, J.
Geophys. Res.-Biogeosci., 118, 385–399, 10.1002/jgrg.20032,
2013.
Jerlov, N. G.: Marine optics, Elsevier, Netherlands, 1976.Kaiser, J., Reuer, M. K., Barnett, B., and Bender, M. L.: Marine productivity
estimates from continuous O2/Ar ratio measurements by membrane inlet mass spectrometry, Geophys. Res. Lett., 32, 1–5,
10.1029/2005GL023459, 2005.
Kirk, J. T.: Light and photosynthesis in aquatic ecosystems, Cambridge
University Press, UK, 1994.La Roche, J.: Ammonium regeneration: its contribution to phytoplankton
nitrogen requirements in a eutrophic environment, Mar. Biol., 75,
231–240, 10.1007/BF00406007, 1983.Lee Chen, Y.: Spatial and seasonal variations of nitrate-based new
production and primary production in the South China Sea, Deep Sea Res. Pt.
I, 52, 319–340, 10.1016/j.dsr.2004.11.001, 2005.Lee Chen, Y. and Chen, H.: Seasonal dynamics of primary and new production
in the northern South China Sea: The significance of river discharge and
nutrient advection, Deep Sea Res. Pt., 53, 971–986, 10.1016/j.dsr.2006.02.005, 2006.Li, D., Zhou, M., Zhang, Z., Zhong, Y., Zhu, Y., Yang, C., Xu, M., Xu, D.,
and Hu, Z.: Intrusions of Kuroshio and Shelf Waters on Northern Slope of
South China Sea in Summer 2015, J. Ocean Univ. China, 17, 477–486,
10.1007/s11802-018-3384-2, 2018.Li, Q., Guo, X., Zhai, W., Xu, Y., and Dai, M.: Partial pressure of CO2
and air-sea CO2 fluxes in the South China Sea: Synthesis of an 18-year
dataset, Prog. Oceanogr., 182, 102272, 10.1016/j.pocean.2020.102272, 2020.Liu, M., Liu, X., Ma, A., Li, T., and Du, Z.: Spatio-temporal stability and
abnormality of chlorophyll-a in the northern south china sea during
2002–2012 from MODIS images using wavelet analysis, Cont. Shelf Res., 75,
15–27, 10.1016/j.csr.2013.12.010, 2014.Lockwood, D., Quay, P. D., Kavanaugh, M. T., Juranek, L. W., and Feely, R.
A.: High-resolution estimates of net community production and air-sea
CO2 flux in the northeast Pacific, Global Biogeochem. Cy., 26, GB4010,
10.1029/2012GB004380, 2012.Ma, H., Zeng, Z., He, J., Chen, L., Yin, M., Zeng, S., and Zeng, W.: Vertical
flux of particulate organic carbon in the central South China Sea estimated
from 234Th-238U disequilibria, Chinese J. Oceanol. Limnol., 26,
480–485, 10.1007/s00343-008-0480-y, 2008.Ma, H., Zeng, Z., Yu, W., He, J., Chen, L., Cheng, J., Yin, M., and Zeng, S.:
234Th/ 238U disequilibrium and particulate organic carbon export
in the northwestern South China Sea, Acta Oceanol. Sin., 30, 55–62,
10.1007/s13131-011-0119-2, 2011.Manning, C. C., Stanley, R. H. R., Nicholson, D. P., Smith, J. M., Timothy
Pennington, J., Fewings, M. R., Squibb, M. E., and Chavez, F. P.: Impact of
recently upwelled water on productivity investigated using in situ and
incubation-based methods in Monterey Bay, J. Geophys. Res.-Oceans, 122,
1901–1926, 10.1002/2016JC012306, 2017.Mathis, J. T., Cross, J. N., and Bates, N. R.: Coupling primary production
and terrestrial runoff to ocean acidification and carbonate mineral
suppression in the eastern Bering Sea, J. Geophys. Res.-Ocean., 116, C02030,
10.1029/2010JC006453, 2011.Millero, F. J. and Poisson, A.: International one-atmosphere equation of
state of seawater, Deep Sea Res. Pt. A, 28, 625–629, 10.1016/0198-0149(81)90122-9, 1981.Monterey, G. and Levitus, S.: Seasonal Variability of Mixed Layer Depth,
NOAA Atlas NEDIS 14, U.S. Gov. Prinitng Off. D.C. available at:
http://www.nodc.noaa.gov (last access: 18 April 2020), 1997.Nemcek, N., Ianson, D., and Tortell, P. D.: A high-resolution survey of DMS,
CO2, and O2/Ar distributions in productive coastal waters, Global Biogeochem. Cy., 22, GB2009, 10.1029/2006GB002879, 2008.Ning, X., Chai, F., Xue, H., Cai, Y., Liu, C., and Shi, J.:
Physical-biological oceanographic coupling influencing phytoplankton and
primary production in the South China Sea, J. Geophys. Res.-Oceans, 109, C10005,
10.1029/2004JC002365, 2004.Ning, X., Peng, X., Le, F., Hao, Q., Sun, J., Liu, C., and Cai, Y.: Nutrient limitation of phytoplankton in anticyclonic eddies of the northern South China Sea, Biogeosciences Discuss., 5, 4591–4619, 10.5194/bgd-5-4591-2008, 2008.Pan, X., Wong, G. T. F., Shiah, F. K., and Ho, T. Y.: Enhancement of
biological productivity by internal waves: Observations in the summertime in
the northern South China Sea, J. Oceanogr., 68, 427–437,
10.1007/s10872-012-0107-y, 2012.
Parsons, T. R., Maita, Y., and Lalli, C. M.: A Manual of Chemical &
Biological Methods for Seawater Analysis, Pergamon Press, Oxford, UK, 1984.Pawlowicz, R.: M_Map: A mapping package for MATLAB, version
1.4m, available at: https://www.eoas.ubc.ca/~rich/map.html, last access: 16 March 2020.Qin,
C., Zhang, G., Zheng, W., Han, Y., and Liu, S.: Net community production, nutrients,
and hydrographic parameters in the South China Sea in October 2014 and June 2015, Zenodo, 10.5281/zenodo.4496886,
2021.Quay, P. D., Peacock, C., Bjrkman, K., and Karl, D. M.: Measuring primary
production rates in the ocean: Enigmatic results between incubation and
non-incubation methods at Station ALOHA, Global Biogeochem. Cy., 24, GB3014,
10.1029/2009GB003665, 2010.Rehder, G. and Suess, E.: Methane and pCO2 in the Kuroshio and the South China Sea during maximum summer surface temperatures, Mar. Chem., 75,
89–108, 10.1016/S0304-4203(01)00026-3, 2001.Reuer, M. K., Barnett, B. A., Bender, M. L., Falkowski, P. G., and Hendricks,
M. B.: New estimates of Southern Ocean biological production rates from
O2/Ar ratios and the triple isotope composition of O2, Deep Sea Res. Pt. I, 54, 951–974,10.1016/j.dsr.2007.02.007, 2007.Shadwick, E. H., Tilbrook, B., Cassar, N., Trull, T. W., and Rintoul, S. R.:
Summertime physical and biological controls on O2 and CO2 in the
Australian Sector of the Southern Ocean, J. Mar. Syst., 147, 21–28,
10.1016/j.jmarsys.2013.12.008, 2015.Shi, X., Li, H., Han, X., Wang, L., and Zhu, C.: Influence of typical
mesoscale oceanographical process on the distribution of nutrients and
dissolved oxygen in the Northern part of South China Sea in summer (in
Chinese), Acta Sci. Circumstantiae, 34, 695–703,
10.13671/j.hjkxxb.2014.0121, 2014.Stanley, R. H. R., Kirkpatrick, J. B., Cassar, N., Barnett, B. A., and
Bender, M. L.: Net community production and gross primary production rates
in the western equatorial Pacific, Global Biogeochem. Cy., 24, GB4001,
10.1029/2009GB003651, 2010.
Su, J. and Yuan, Y.: Coastal hydrology of China, China Ocean
Press, Beijing, China, 2005 (in Chinese).Takahashi, T., Sutherland, S. C., Wanninkhof, R., Sweeney, C., Feely, R. A.,
Chipman, D. W., Hales, B., Friederich, G., Chavez, F., Sabine, C., Watson,
A., Bakker, D. C. E., Schuster, U., Metzl, N., Yoshikawa-Inoue, H., Ishii,
M., Midorikawa, T., Nojiri, Y., Körtzinger, A., Steinhoff, T., Hoppema,
M., Olafsson, J., Arnarson, T. S., Tilbrook, B., Johannessen, T., Olsen, A.,
Bellerby, R., Wong, C. S., Delille, B., Bates, N. R. and de Baar, H. J. W.:
Climatological mean and decadal change in surface ocean pCO2, and net
sea–air CO2 flux over the global oceans, Deep Sea Res. Pt. II, 56, 554–577, 10.1016/j.dsr2.2008.12.009, 2009.Tamminen, T.: Effects of ammonium effluents on planktonic primary production
and decomposition in a coastal brackish water environment I. Nutrient
balance of the water body and effluent tests, Netherlands J. Sea Res.,
16, 455–464, 10.1016/0077-7579(82)90050-3, 1982.Teeter, L., Hamme, R. C., Ianson, D., and Bianucci, L.: Accurate Estimation
of Net Community Production From O2/Ar Measurements, Global Biogeochem. Cy., 32, 1163–1181, 10.1029/2017GB005874, 2018.Teira, E., Mouriño, B., Marañón, E., Pérez, V., Pazó, M.
J., Serret, P., De Armas, D., Escánez, J., Woodward, E. M. S., and
Fernández, E.: Variability of chlorophyll and primary production in the
Eastern North Atlantic Subtropical Gyre: Potential factors affecting
phytoplankton activity, Deep. Res. Pt. I,, 52, 569–588, 10.1016/j.dsr.2004.11.007, 2005.Tortell, P. D.: Dissolved gas measurements in oceanic waters made by
membrane inlet mass spectrometry, Limnol. Oceanogr. Methods, 3, 24–37,
10.4319/lom.2005.3.24, 2005.Tortell, P. D., Asher, E. C., Ducklow, H. W., Goldman, J. A. L., Dacey, J.
W. H., Grzymski, J. J., Young, J. N., Kranz, S. A., Bernard, K. S., and
Morel, F. M. M.: Metabolic balance of coastal Antarctic waters revealed by
autonomous pCO2 and ΔO2/Ar measurements, Geophys. Res. Lett., 41, 6803–6810, 10.1002/2014GL061266, 2014.Tortell, P. D., Merzouk, A., Ianson, D., Pawlowicz, R., and Yelland, D. R.:
Influence of regional climate forcing on surface water pCO2,
ΔO2/Ar and dimethylsulfide (DMS) along the southern British
Columbia coast, Cont. Shelf Res., 47, 119–132,
10.1016/j.csr.2012.07.007, 2012.Ulfsbo, A., Cassar, N., Korhonen, M., Van Heuven, S., Hoppema, M., Kattner,
G., and Anderson, L. G.: Late summer net community production in the central
Arctic Ocean using multiple approaches, Global Biogeochem. Cy., 28,
1129–1148, 10.1002/2014GB004833, 2014.Uu, D. V. and Brankart, J. M.: Seasonal variation of temperature and
salinity fields and water masses in the Bien Dong (South China) Sea, Math.
Comput. Model., 26, 97–113, 10.1016/S0895-7177(97)00243-4, 1997.Wang, N., Lin, W., Cheng, B., and Huang, B.: Metabolic states of the Taiwan
Strait and the northern South China Sea in summer 2012, J.
Trop. Oceanogr., 33, 61–68, 10.3969/j.issn.1009-5470.2014.04.008,
2014 (in Chinese).
Wanninkhof, R.: Relationship between wind speed and gas exchange over the
ocean, J. Geophys. Res.-Oceans, 97, 7373–7382, 10.1029/92JC00188,
1992.Weiss, R. F.: The solubility of nitrogen, oxygen and argon in water and
seawater, Deep Sea Res. Oceanogr. Abstr., 17, 721–735,
10.1016/0011-7471(70)90037-9, 1970.Zhai, W. D., Dai, M., and Cai, W.-J.: Coupling of surface pCO2 and dissolved oxygen in the northern South China Sea: impacts of contrasting coastal processes, Biogeosciences, 6, 2589–2598, 10.5194/bg-6-2589-2009, 2009.Zhang, R., Zhu, X., Yang, C., Ye, L., Zhang, G., Ren, J., Wu, Y., Liu, S.,
Zhang, J., and Zhou, M.: Distribution of dissolved iron in the Pearl River
(Zhujiang) Estuary and the northern continental slope of the South China
Sea, Deep Sea Res. Pt. II, 167, 14–24, 10.1016/j.dsr2.2018.12.006, 2019.