Distributions, stoichiometric patterns and cross-shelf exports
of dissolved organic matter in the East China Sea
J.-J. Hung
a,*, C.-H. Chen
a, G.-C. Gong
b, D.-D. Sheu
a, F.-K. Shiah
c aInstitute of Marine Geology and Chemistry, National Sun Yat-Sen University, Kaohsiung, Taiwan, ROC
b
Department of Oceanography, National Taiwan Ocean University, Keelung, Taiwan, ROC
c
Institute of Oceanography, National Taiwan University, Taipei, Taiwan, ROC Accepted 12 October 2002
Abstract
This paper sets out to elucidate distributions, stoichiometric patterns and cross-shelf exports of dissolved organic matter (DOM) in the East China Sea (ECS). Surface distributions of dissolved organic carbon (DOC) in the ECS varied spatially, ranging from 85 to 120 mM in the China CoastalWater (CCW), from 75 to 85 mM in the Kuroshio Water (KW), and from 60 to 70 mM in the upwelling water. DOC concentrations in most regions of the shelf mixed water (SMW) were between 72 and 85 mM. Temporalvariations were insignificant with the exception of the CCW, where the concentration was greater in summer and autumn than in spring and winter. Spatialpatterns of dissolved organic nitrogen (DON) and dissolved organic phosphorus (DOP) were less variable in the spring season, ranging from 6 to 9 mM for DON and 0.15 to 0.25 mM for DOP. The elemental ratios of DOC, DON and DOP are much greater than the Redfield ratio, ranging from 8.9 to 15.3 for C/N, from 19 to 83.6 for N/P and from 200 to 853 for C/P. Such ratios also showed a general increase with depth. However, the slopes of linear regression of DOM pairs become smaller and were about 8.4 for DDOC/DDON, 19 for DDON/DDOP and 129 for DDOC/DDOP, suggesting the possible ratios derived from the recently produced fraction. Vertical increases of ratios in DOM pairs may suggest that the recycling of DOP and DON is more rapid than that of DOC to overcome the biological overconsumption of DIC to DIN and DIP (DNDIC:DNDIN:DNDIP=1:14.3:129), as the ratio of DPOC/DPON (6.74) approaches the Redfield ratio in the euphotic zone. Degradation rates of DOC and DOP are variable with water types and are generally greater for the CCW than for the KW, suggesting the different lability of DOM in various waters. The residence times of bulk DOC, DON, and DOP in the shelf are about 1.10, 0.98, and 0.92 yr, respectively, which are close to the ranges of the mean residence time of shelf water reported previously, suggesting that there is little to no export off the shelf. Finally, total inputs are closely balanced by total outputs for bulk DOC, DON, and DOP in the ECS shelf, if temporal variations of DON and DOP are negligible. However, net exports across the shelf are highly probable if only labile and semi-labile DOC (3.870.6 1012g C yr1), DON (4827160 109g N yr1), and DOP (37.5718.5 109g P yr1) are involved in the budget calculation. The ECS shelf is likely to be a small source of degradable DOM for oceanic waters.
r2003 Elsevier Science Ltd. All rights reserved.
*Corresponding author. Tel.:+886-7-5255147; fax: +886-7-5255130. E-mail address:hungjj@mail.nsysu.edu.tw (J.-J. Hung).
0967-0645/03/$ - see front matter r 2003 Elsevier Science Ltd. All rights reserved. doi:10.1016/S0967-0645(03)00014-6
1. Introduction
The role of DOC (for abbreviations refer
Table 1) in the marine carbon cycle has received considerable attention over the last two decades. The studies, however, have focused largely on open oceans. Although the importance of margin-alseas to the marine carbon cycle is recognized (Walsh, 1991; Walsh et al., 1988; Wollast, 1991;
Biscaye and Anderson, 1994), fewer DOC studies (Guo et al., 1995; Chen et al., 1996; Bates and Hansell, 1999; Alverez-Salgado et al., 2001) have been conducted to elucidate the fate and cycling of organic carbon in marginalseas. This is possibly due to the multiple sources and more complex processes involving organic carbon in the marginal seas. The situation is particularly true for the East China Sea (ECS) continentalmargin where a large shelf interacts with large rivers and the Kuroshio Current, respectively, from western and eastern boundaries. Very few results related to ECS organic carbon biogeochemistry have been re-ported from a shelf-wide consideration.
DON and DOP are two organic nutrients involved in the marine carbon cycle. Several processes (e.g., phytoplankton excretion, zoo-plankton grazing, POM solubilization,
atmo-spheric deposition and riverine flux) are
responsible for the introduction of DON, DOP,
and DOC into oceans (Connolly et al., 1992;
Wotton, 1994; Orrett and Karl, 1987; Anderson and Williams, 1998; Hansell and Carlson, 1998). Some DON and DOP can be used directly by primary producers under certain circumstances, but in most situations they are degraded to release inorganic nutrients as bacteria consume DOC
during growth (Connolly et al., 1992; Wotton,
1994). DON and DOP may play important roles in
nutrient budgets and carbon cycles in oligotrophic oceans where inorganic nutrients are devoid in the
euphotic zone (Thomas et al., 1971; Jackson and
Williams, 1985; Vidalet al., 1999). However, it is not clear whether DON and DOP also can be important in carbon cycles in marginal seas where surface inorganic nutrients are not completely depleted.
The ECS is one of the largest marginal seas worldwide, receiving enormous rates of
fresh-water (>885 km3yr1), suspended matter (>1.4
109ton yr1) and nutrients from the Changjiang
and Huangho (Milliman and Meade, 1983). The
ECS shelf receives a larger amount of nutrients from the upwelled Kuroshio subsurface water, occurring on the southern ECS shelf break off
northeastern Taiwan (Wong et al., 1991;Liu et al.,
1992;Li, 1994; Chen, 1996). As a result, the ECS shelf may be one of the most productive marginal seas in the world. The primary productivity, however, is temporally and spatially variable on the southern ECS shelf and the spatial pattern is generally coincident with nutrient distributions (Gong et al., 2000). The annualmean value over the whole southern ECS shelf is 200730 g C m2yr1 (Gong et al., 2000). Although the major source of DOM in oceanic water is likely
related to primary productivity (Carlson et al.,
1994; Norrman et al., 1995), the relationship between primary productivity and DOM
distribu-tions in marginalseas is not wellknown. Shiah
et al. (2000) reported high bacterialbiomass, production, and average turnover rates in coastal and upwelled waters but lower rates in the Kuroshio Water from the southern ECS. The bacterial processes are all positively correlated with the POC inventory, but their relations to
DOC are not explored (Shiah et al., 2000).Hung
et al. (2000)reported DOC and POC distributions
Table 1
Acronyms used in this study
Acronyms Water mass and chemicalproperties
CCW China CoastalWater
KW Kuroshio Water
SMW Shelf Mixed Water
YSMW Yellow Sea Mixed Water DIC Dissolved inorganic carbon DIN Dissolved inorganic nitrogen DIP Dissolved inorganic phosphorus DOC Dissolved organic carbon DON Dissolved organic nitrogen DOP Dissolved organic phosphorus DOM Dissolved organic matter POC Particulate organic carbon PON Particulate organic nitrogen TDN Totaldissolved nitrogen TDP Totaldissolved phosphorus
from a transect of southern ECS and suggested that DOC and POC are derived from both biological and terrestrial sources, but the contri-butions vary spatially.
New productivity in the ECS was reported to range from 17% to 82% of totalprimary
productivity (Chen et al., 1999). Apparently,
regenerated productivity was sometimes significant in the ECS shelf. The cycling of dissolved organic matter can result in nutrient regeneration and contribute to the primary productivity. Thus, it is worth knowing the contribution of DON and DOP remineralization to the primary productivity in the ECS. Besides, DOC not recycled over the broad shelf may be exported from the ECS. The intensive exchange between shelf water and the Kuroshio may facilitate DOC export. Under-standing the shelf DOC export may also enhance our knowledge of the role of the marginal sea in the marine carbon cycle. Therefore, in this study, we examine the spatialand temporaldistributions of DOC, stoichiometric relationships for DOC, DON, and DOP, and degradation rates of DOC and DOP in the ECS. In addition, the cross-shelf
exports of DOC, DON, and DOP from the ECS continentalshelf are investigated.
2. Methods
Samples were collected on board the R/V Ocean Researcher I during cruises ORI-449 (05/2–15/ 1996), 493 (07/08–13/1997), 511 (12/19–30/1997), 521 (06/30–07/05/1998), and 532 (10/29–11/06/ 1998). The cruises covered all seasons. The sampling stations for each cruise are shown in
Fig. 1. Most cruises occupied numerous stations to cover the whole continental shelf of the ECS, but only five stations were occupied to take samples for the experiment of DOM degradation during Cruise 493. Seawater samples were collected with cleaned Go-Flo bottles (20-l) mounted on a CTD/ rosette that recorded the temperature and salinity profiles.
Dissolved and particulate organic matter sam-ples were separated by filtration through
precom-busted 25-mm GF/F (Whatman) glass-fiber
filters immediately after collecting the seawater,
according to the procedures developed by Hung and Lin (1995). The filters were sealed in cleaned petri-dishes and stored in a freezer during transfer to the laboratory. The filters were then dried in an
oven at 60C for POC determination. The filtered
water was acidified with HCl(0.4%, v/v) and
stored at 4C for DOC measurements.
DOC was measured with the method of high-temperature catalytic oxidation using the
Shimad-zu TOC-5000 analyzer (Hung and Lin, 1995). The
muffled (850C, 2 h) CuO wire was placed on the
top of the catalyst to enhance the peak shape and
reproducibility of peak integration (Sharp, 1993).
Before standard and sample measurements,
cata-lysts (Pt/Al2O3) were thoroughly conditioned by
injecting DOC-free Milli-Q water until a low and constant signalwas detected. The instrument blank was estimated to be 10 mM C according to
the procedure of Benner and Strom (1993). DOC
in seawater was analyzed by placing 20 ml of acidified seawater in a Pyrex glass test tube and purged with ultrapure air for 8 min to remove
dissolved CO2. One-hundred microliter aliquots of
decarbonated standard or seawater sample were injected at least five times into the oxidation column. Abnormal signals (usually low and high) were automatically rejected to ensure that the standard deviation of repeated injections was o2%. The relative difference between our mea-surements and recommended values was within the 74% and 710% limits, respectively, for stages 1 and 2 of the DOC intercomparison exercises (J. Sharp personalcorrespondence). DON was deter-mined from the difference between dissolved
inorganic nitrogen and totaldissolved nitrogen
(TDN) that was measured with the chemilumi-nence method using an instrument of Antek Models 771/720. DOP was determined from the
difference between PO4
3
and TDP (totaldissolved phosphorus) that was measured with
UV-persul-fate oxidation and colorimetric method (Ridaland
Moore, 1990). Dissolved inorganic nitrogen
(NO3+NO2, hereafter DIN) and PO43(hereafter
DIP) were determined colorimetrically (Grasshoff
et al., 1983) with a flow injection analysis method (Pai and Yang, 1990; Gong, 1992). The precision was better than 8% and 5%, respectively, for TDN and TDP.
The experiments for DOM degradation in various seawaters were conducted using an in-cubation method. Surface water samples were taken from Cruise 493 during summer (07/08–13/ 1997). A large volume (B30 l) of water was filtered through precombusted GF/C filters (B1.2 mm pore size) and then divided into numerous 1-l polycarbonate bottles. Subsequently, the bottles
were stored at room temperature (B25C) under
totally dark conditions. Duplicated bottles were opened for DOC and DOP determination accord-ing to the described procedures at various times of incubation. Bacterialabundance was not mon-itored during the course of experiment. The incubation lasted over 270 d for each water sample.
Particulate organic carbon (POC) and nitrogen (PON) were determined during the 449 cruise. POC and PON were determined with a C/N/S analyzer (Fisons NCS 1500) after carbonate was removed by placing the filtered particulate matter in a precombusted silver boat, adding a few drops
of 2N HClon the filter, and oven drying at 50C
for 48 h. The silver boat was then wrapped and placed in an autosampler to determine POC and PON concentrations. The blank value attributed to the GF/F filter and silver boat was deducted from the raw value of a sample concentration. The sampling and measuring precisions of POC and
PON were 70.3 mM C and 70.2 mM N (71s),
respectively, as evaluated from 8 replicate samples from the same depth.
3. Results and discussion 3.1. Hydrography
The hydrographic settings of the ECS are
displayed largely in Fig. 2. Briefly, the major
hydrographic features are the characteristics of waters mixed from the CCW, the SMW, the KW and/or the Yellow Sea Mixed Water (YSMW).
The CCW is a mixture of the Changjiang freshwater and the SMW, characterized by a low
salinity (o32) (Fig. 3). The water flows southward
predominantly along the China coast during the winter season, but a part of CCW also flows
northeastward from the Changjiang River mouth
during the summer season (Beardsley et al., 1985;
Gong et al., 1996). Most stations along the China
coast are obviously within the CCW (Figs. 1 and
2). Seasonal variations of salinity are apparently
stronger in the CCW than in other waters (Fig. 3).
The SMW is derived from mixing among the CCW, the Taiwan Strait Water, the Kuroshio surface water and the Kuroshio subsurface water. Water temperature and salinity of the SMW are generally in the range between the CCW and the KW (Fig. 4). The KW is characterized by high
temperature and salinity (Fig. 4), but low nutrients
in the surface layer. The striking reversed S-shape
in T–S relationship (Fig. 4) shows a maximum
salinity >34.7 that is also a characteristic of the
North Pacific subtropicalwater (Nitani, 1972).
This KW flows northward along the east coast of Taiwan and results in a persistent upwelling from the Kuroshio subsurface onto the ECS shelf occurring around the southern ECS off northeast Taiwan (St. 52 in the 449 cruise) induced by a topographic change. The upwelled water is laden with nutrients and may influence carbon biogeo-chemistry as well. The YSMW forms from mixing
between the Yellow Sea Coastal Water of the ECS and the KW. This water is generally confined to the northern ECS with lower temperature and salinity, and the occupied area is subjected to
seasonaland annualvariability (Chern and Wang,
1990;Gong and Liu, 1995).
3.2. Distributions of DOC, DON, and DOP Seasonaldistributions of DOC in the ECS
surface water are shown in Fig. 5. The
distribu-tions vary temporally and spatially. The concen-tration is relatively high in the CCW, ranging from about 85 mM in the southern China coast (26–
30N) during autumn and winter to 110–120 mM
close to the Changjiang Estuary (30.5–32N, 123–
124E) during summer. The elevated DOC
con-centrations may be derived mostly from terrestrial inputs and less from in situ production, as the CCW is sometimes low in primary productivity
due to photo limitation (Gong et al., 2000). The
concentration is higher in summer and autumn seasons than in winter and spring seasons, probably resulting from differing strengths of freshwater (terrestrial) and nutrient (productivity)
inputs. The concentration is lower (72–85 mM) in the SMW than in the CCW. This may arise from
the decrease of terrestrialinputs and imbalance
between biological production and bacterial con-sumption, in addition to the mixing process. Seasonalvariations are significant only in the area close to the Changjiang mouth. Although far away from terrestrialsources, DOC concentrations remain fairly high in the surface layer of KW (75–85 mM) with frequent appearance of a subsur-face maximum at 40–80 m, just below the mixing layer. Such a distribution pattern has been
discussed byHung et al. (2000)and was attributed
primarily to physical stability of the stratified water column and a lower rate of microbial consumption. No significant difference is found between seasons for the KW, which may result from little influence by terrestrial inputs or
primary production during various seasons. The lowest concentrations of DOC (60–70 mM) were always found in the upwelled water. The upwelling brought in DOC-poor water (55–60 mM) from the Kuroshio subsurface and diluted the ambient DOC concentration. These biophysicalprocesses
are likely responsible for no correlation
(r ¼ 0:192; p > 0:1) between DOC and primary
productivity measured by G.-C. Gong et al.
(2003)in the ECS.
Surface distributions of DON and DOP derived from Cruise 449 in the ECS are displayed in
Fig. 6a and b, respectively. Spatial distributions range from 6 to 9.6 mM for DON and from 0.05 to 0.25 mM for DOP. Relatively higher concentra-tions of DON (7.7–9.6 mM) and DOP (>0.2 mM) are found in the CCW. Intermediate ranges
of DON (6–7 mM) and DOP (0.15–0.2 mM)
concentrations are present in the KW. The lowest
concentrations of DON (p5 mM) and DOP
(p0.18 mM) are found in the upwelled water. Such
surface concentration ranges are not significantly different from those reported previously for North
Pacific surface waters byWalsh (1989); (DON: 5–
10 mM; DOP: 0.37 mM),Maita and Yanada (1990);
(DON: 1–10 mM),Ridaland Moore (1992); (DOP:
0.17–0.38 mM), and Abell et al. (2000); (TON: 5–
6 mM; TOP: 0.1–0.35 mM). DON is apparently the most abundant nitrogen species ranging from 20%
to 99% of TDN in surface water (o150 m), but
can be as low as 8% of TDN in the Kuroshio deep water. DOP is also generally more abundant than DIP, ranging from 5.5% to 68% of TDP in surface
water (o150 m), and is as low as 2% of TDP in the
Kuroshio deep water.
VerticalDOM distributions in various water types during spring (Cruise 449) are displayed in
Fig. 7. Concentrations of DOC, DON, and DOP
decrease with depth to about 61–74, 4.2–5.5, and
0.06–0.08 mM, respectively, in the CCW (Fig. 7a),
the SMW (Fig. 7b) and the upwelled water
(Fig. 7c), and to about 51, 3.0, and 0.06 mM,
respectively, in the KW (Fig. 7d). Concentrations
of DOM in the Kuroshio deep water may reflect the refractory fractions. Such distributions are similar to those previous reports in marginal seas (Hopkinson Jr. et al., 1997; Sanders and Jickells, 2000).
In addition to terrestrialinputs, complicated
biological processes are involved in the release of
DON and DOP (Sanders and Jickells, 2000). The
greater inorganic nutrients in marginalseas are not necessarily responsible for higher production of DON and DOP. DON and DOP are inversely correlated (data not shown here) with DIN and DIP in the ECS if a few data with particularly high DIN concentrations in the CCW are excluded. There is also no clear relationship between DON
0 3 6 9 12 15 18 21 24 27 30 30 31 32 33 34 35 SMW UW CCW KW Salinity σ=27θ σ=26θ σ=2θ 5 σ=24θ σθ=23 σθ=22 σθ=21 σθ=20 st.11 st.19 st.26 st.29 st.32 st.34 st.44 st.47 st.48 st.50 st.52 st.55 T e m p e rt ur e (˚ C )
Fig. 4. Temperature-salinity plots for selected stations during Cruise 449. The reversed s-shape curve indicates T–S characteristics of the KW.
(DOP) and primary productivity (PP), although there is a reversed trend between integrated PP (IPP) and integrated DON or DOP (IDON or IDOP). Processes related to DOC dynamics are likely to regulate DON and DOP distributions in the ECS as well. Temporal variations of DON and DOP are not known at this moment, although we speculate little variation in the ECS.
Since marine DOM is an important substrate for bacterialgrowth, it is worth knowing the relationship between DOM distributions and bacterialdynamics. Data derived from Cruise 449 show that depth integrated
bacterialproduc-tion (IBP; 3–13 mg C m2d1) is positively
corre-lated with depth integrated DOC (IDOC; 84–
197 mmolC m2) and DON (IDON; 6–17 mmol
N m2;Fig. 8). However, the relationship for IBP vs. depth-integrated DOP (IDOP; 0.08–0.42 mmol
P m2) is not significant, primarily due to the data
derived from Station 11 located near the Chang-jiang mouth. There, a high value of IBP is accompanied with a very low IDOP inventory. Both BP and DOM are integrated mostly from whole water column for shelf waters but from the euphotic zone (o120 m) for the KW. The relation between IBP and IDOP becomes positively
corre-lated (R ¼ 0:883; po0:01) when this datum is
excluded (Fig. 8). Shiah et al. (1999, 2000)
suggested that bacterialgrowth in the ECS shelf was interactively controlled by temperature and substrate (i.e. DOM) supply during cold seasons (i.e. winter and spring). Positive correlation was
significant between BP and temperature (Shiah
et al., 1999). Meanwhile, significant correlation (positive) was also present between DOC and temperature while temperature was greater than
14C in the southern ECS (Hung et al., 2000). In
this study (Cruise 449, later spring), positive correlations between IBP and IDOM may indicate that bacteria productivity is primarily controlled by temperature which is also positively correlated to DOC. Another explanation is that bacteria productivity is proportionalto the abundance of bio-reactive DOM rather than totalDOM. The greater BP may be accompanied by a larger amount of refractory DOM during the measure-ment in the shelf where refractory DOM is
abundant. This possibility results in a positive relationship between IBP and IDOM.
Correlation is more significant between DOC
and sigma-t (R ¼ 0:71; po0:0001; Fig. 9a) than
between DOC and salinity (R ¼ 0:58; po0:0001;
data not shown) for Cruise 449, implying the importance of verticalmixing in controlling DOC distributions. Those concentrations scattered from the linear relationship may result from the effects of biogeochemicalprocesses superimposed on the mixing process. Correlation is much poorer
between DON and sigma-t (R ¼ 0:47; po0:0001) and between DOP and sigma-t (R ¼ 0:33;
po0:0001) than between DOC and sigma-t
(Figs. 9a–c). Apparently, DOP and DON are more non-conservative than is DOC. This is consistent with a generalconsensus that the recycling of DOP and DON is more rapid than that of DOC
(Wotton, 1994; Hopkinson et al., 1997; Thomas et al., 1999).
3.3. Stoichiometry of DOC, DON, and DOP Simultaneous investigations of DOC, DON, and DOP were only made during Cruise 449 (spring
6 0 7 0 80 90 100 4 0 3 0 2 0 1 0 0 Depth (m) 4 6 8 10 0.0 0.1 0.2 0.3 0.4 6 0 7 0 8 0 9 0 10 0 10 0 8 0 6 0 4 0 2 0 0 4 6 8 1 0 0.0 0 0 .0 5 0 .1 0 0 .1 5 0.2 0 0 .2 5 6 0 7 0 80 90 100 10 0 8 0 6 0 4 0 2 0 0 Depth (m) 4 6 8 1 0 (c) (d) (b) (a) 0 .0 0 0.0 5 0 .10 0 .15 0 .2 0 0.2 5 5 0 6 0 7 0 8 0 9 0 10 0 1 200 1 000 80 0 60 0 40 0 20 0 0 2 4 6 8 1 0 0.0 0 0.0 5 0 .1 0 0 .1 5 0 .2 0 DON (µM) DOC (µM) DOP (µM) DOC (µM) DOP (µM) DON (µM) DOP DON DOC DOP DOC (µM) DOC (µM) DOP (µM) DOP (µM) DON (µM) DON (µM) DOP DOC DON DOC DOP DON DOC DON
Fig. 7. Vertical profiles of DOC, DON and DOP in the CCW (7a, St. 47), the SMW (7b, St. 34), the upwelled water (7c, St. 52) and the KW (7d, St. 19) during Cruise 449.
season); therefore, the temporalvariation of stoichiometry will not be discussed in this paper. The ratios of dissolved organic carbon, nitrogen and phosphorus in the ECS deviate significantly from the Redfield ratio (C/N/P=106/16/1). The elemental ratio ranges from 8.9 to 15.3 for DOC/ DON, from 19 to 83.6 for DON/DOP and from 200 to 853 for DOC/DOP in the ECS shelf and the
Kuroshio surface (o150 m) (Figs. 10a–c).
Appar-ently DOM is considerably enriched in carbon relative to nitrogen and phosphorus, and nitrogen relative to phosphorus, compared to the Redfield value. Elemental ratios exhibit a large range for each water type; thus no clear spatial pattern of elemental ratios can be identified, although slightly greater values appear in the upwelled water and the KW. Three ratios (C/N, N/P, C/P) all increase gently with depth to 100 m but increase markedly below to 100 m for the KW (Sts. 19 and 55), suggesting a preferentialdecay of DOP over DOC and DON, and DON over DOC in the ECS. Such
patterns of ranges and depth-increased ratios are consistent with previous results obtained from
marginalseas and open oceans (Williams et al.,
1980; Hopkinson et al., 1997; Vidalet al., 1999;
Sanders and Jickells, 2000). Nevertheless, the bulk DOM in seawater generally consists of a biologi-cally labile and a relatively refractory part that may be pronounced in the ECS continental margin. It is obvious that the elemental ratios of DOM do not necessarily conform to the Redfield ratio derived from biota. The slope of the DOC and DON data pairs increases (DDOC/DDON) may be more indicative of the C/N ratio for the recently produced DOM. The slopes of DDOC/
DDON (8.4, Fig. 10a), DDON/DDOP (19,
Fig. 10b) and DDOC/DDOP (129, Fig. 10c) are much smaller than those of the elemental ratios, although the slopes are still greater than the
2 4 6 8 1 0 1 2 1 4 8 0 100 120 140 160 180 200 IDOC (mmol m- 2) I B P = - 3 . 0 6 + 0 . 0 8 x I D O C ; R = 0 . 9 2 5 ; n = 9 I B P = 0 . 7 0 + 0 . 4 3 x I D O N ; R = 0 . 8 2 3 ; n = 9 I B P = 1 . 1 9 + 1 6 x I D O P ; R = 0 . 8 8 3 ; n = 8 w i t h S t 1 1 e x c l u d e d IBP (mg C m -2 d -1) IDON (mmol m- 2) 4 8 1 2 1 6 2 0 IDOP (mmol m- 2) 0.0 0.2 0.4 0.6
Fig. 8. Relation plots between depth-integrated bacterial pro-duction (IBP) and integrated DOC (IDOC), depth-integrated DON (IDON) and depth-depth-integrated DOP (IDOP).
50 60 70 80 90 100 110 2 4 6 8 10 12 22 23 24 25 26 27 28 0.0 0.1 0.2 0.3 0.4 0.5
•
(a) (b) UW R= -0.71 P<0.0001 DO C ( µ M ) CCW SMW KW•
(c) UW R= -0.47 P<0.0001 CCW SMW KW DO N ( µ M )•
UW R= -0.33 P=0.00224 CCW SMW KW DO P ( µM ) Sigma-TFig. 9. Relation plots between sigma-t and DOC (a), DON (b) and DOP (c) for data collected from Cruise 449.
Redfield values. This also may indicate that even for recently produced DOM, carbon is still more enriched over nitrogen and phosphorus, and nitrogen over phosphorus.
The implication of DOM stoichiometry is that DON and DOP should be preferentially decom-posed and the released inorganic N and P are incorporated into particulate organic materials during photosynthesis. This is also inferred from the slope of DPOC/DPON that is 6.74 (data not shown) in the ECS euphotic zone. This value is very close to the Redfield C/N ratio and suggests
the ratio (6.74) of biological uptake from DIC and DIN pools. However, the slope of DNDIC/
DNDIN (14.3, Fig. 11a) is nearly twice the ratio
of DPOC/DPON, implying that DON was prefer-entially decomposed and utilized biologically to overcome the over consumption of DIC to DIN. Although we do not measure DPOC/DPOP, we believe that a preferential decay of DOP over DOC is likely to occur and the released DIP is used for POM production, as justified from a greater value for the DNDIC/DNDIP slope (129;
Fig. 11b) than for the Redfield C:P ratio (1 0 6). Meanwhile, the large magnitude of DNDIC/ DNDIN may indicate that carbon exports from the euphotic zone may be underestimated if they are derived from nitrate uptake and scaling from the Redfield ratio (6.6).
3.4. Degradation of DOC and DOP
Degradation rates of DOC and DOP are estimated from the decrease in concentration with
0 2 4 6 8 10 12 20 40 60 80 100 120 C:N = 8.9 - 15.3 (a) [DOC]=25 + 8.4[DON] R = 0.533; p<0.001 CCW SMW KW DOC ( µ M ) DON (µM) 0.0 0.1 0.2 0.3 0.4 0.5 2 4 6 8 10 12 N:P = 19 - 83.6 (b) [DON]= 4 + 19[DOP] R= 0.373; p<0.05 0.0 0.1 0.2 0.3 0.4 0.5 40 60 80 100 120 C:P = 200 - 853 DOC ( µ M ) DOP (µM) DON ( µ M ) DOP (µM) (c) [DOC]=60 + 129[DOP] R= 0.383; p<0.05
Fig. 10. Ranges of elemental ratios and regression slopes of concentrations between DOC and DON (a), between DON and DOP (b) and between DOC and DOP (c) from data of Cruise
449. 0 2 4 6 8 1900 2000 2100 2200 2300 [NDIC] = 1992 + 14.3[NDIN] R= 0.7774, p<0.0001 NDIC ( µ M ) NDIN (µM) 0.0 0.2 0.4 0.6 0.8 1900 2000 2100 2200 2300 NDIC ( µ M ) NDIP (µM) [NDIC] = 1991 + 129[NDIP] R= 0.5126, p<0.0001 (a) (b)
Fig. 11. Relation plots between NDIC (DIC normalized to salinity 35) and NDIN (a), and between NDIC and NDIP (b).
incubation time.Fig. 12ashows that DOC degra-dation varies with water type and incubation time. The patterns of DOC decrease with time are similar for various waters, so the degradation rate for each water can be fitted and delineated well
with the first order decay function, C ¼ C0ekt;
where C0 is the initialconcentration, t is the
incubation time and k (decay coefficient) is the degradation rate. The DOC degradation rate is
thus calculated as 0.043–0.048 d1 for the CCW,
0.017–0.047 d1 for the SMW and 0.009 d1 for
the KW (Fig. 12b). Apparently, the degradation
rate is faster in the CCW than in the KW, and the rate is also affected substantially by the initial
concentration. Hopkinson et al. (1997) also
reported a positive relationship between decay
coefficient (0–0.0025 d1) and initialDOC
concen-tration (50–91 mM) for seawater at the Georges
Bank region. However,Kirchman et al. (1991)and
Ogura (1972) obtained different rates of DOC
decay from the North Atlantic (0–0.4 d1) and the
northern North Pacific (0.005 d1), with a small
difference of initialconcentration. Variations in chemicalcomposition may be responsible for such variable degradation rates. It may be argued that the degradation rate is not only affected by the initialconcentration but also by the abundance of inorganic nutrients that may limit microbial
0 50 100 150 200 250 300 50 60 70 80 90 100 110 120 (a) st.5 st.9 st.10 st.11 st.15 DOC conc. ( µ M ) Time (day) 0.00 0.01 0.02 0.03 0.04 0.05 (b) DOC
DOP decay coefficient (day
-1 ) KW SMW SMW CCW CCW
DOC decay coefficient (day
-1 )
Initial DOC conc. (µM)
99 118 107 121 119 st15 st11 st10 st9 st5 0.02 0.04 0.06 0.08 DOP
Fig. 12. Kinetics of DOC degradation for various waters during the course of incubation (a), and the decay coefficients of DOC and DOP for various waters (b).
activity during incubation. In fact, the degradation rate is correlated significantly with nitrate
concen-tration (R ¼ 0:873; po0:001), as the CCW and the
SMW are more enriched with nitrate than is the KW. As a result, the DOC degradation rate is closely related to microbial turnover rates found in
the ECS byShiah et al. (2000). Despite the possible
arguments that the degradation experiment con-ducted in bottles may have modified the in situ environment and biotic community, our degrada-tion rates are within the ranges of previous
findings (Ogura, 1972; Kirchman et al., 1991;
Hopkinson et al., 1997) and can be used as an index of DOC decay in the ECS.
Three classes of DOC reactivity may be conceivable as DOC concentrations decrease rapidly during the first 40 d of incubation and then decrease gradually between 40 and 200 d, followed by a final step with very slow degradation (Fig. 12a). DOC degradation rates of three classes
are equivalent to 0.88, 0.105 and o0.001 mM
C d1, respectively, if they are derived separately
from a linear model of DOC degradation rate (slope of linear regression between DOC concen-tration and time). Accordingly, DOC may be regarded as labile, semi-labile and refractory with half-lives on the order of weeks, months and years, respectively. It is worth noting that the surface DOC may comprise largely recently produced components and their half-lives are shorter than those of deep water. As the ECS shelf water has a
mean residence time on the order of 1 yr (Li, 1994;
Tsunogai et al., 1997; Peng et al., 1999), only a small fraction of DOC produced in situ may be able to escape from recycling on the shelf and be exported to the open ocean.
Degradation of DOP follows the patterns for DOC, but their decay coefficients are much greater
than those of DOC (Fig. 12b). Little DOP remains
for the incubation time longer than 60 d. This is consistent with findings that DOP is much more labile than is DOC, revealed from stoichiometric patterns. It also can be inferred that DOP is primarily produced recently on the shelf and labile DOP may be recycled several times on the time frame of residence time of ECS shelf water. Degradation of DOP may make a significant contribution to primary production in the ECS.
3.5. Cross-shelf exports of DOC, DON, and DOP It has been discussed that DOM inputs from the Changjiang runoff, the upwelling and sedimentary fluxes combined with that of in situ production may be recycled to a large extent in the ECS shelf, because most DOC half-lives are shorter than the mean residence time of shelf water. Understanding the cross-shelf exports of DOC, DON, and DOP is highly desirable as the exports may provide information about the role of shelves on the marine carbon cycles. Before such an attempt, a water budget from the ECS shelf should be clearly
defined. Li (1994) used a salt balance in the box
modelto estimate the water exchange rate between
the ECS shelf water and the KW as
2.270.9 104
km3yr1 (0.7070.29 Sv). This rate
is roughly 25 times the value of major river runoffs in the region. Using the similar model but separating the KW into the Kuroshio surface water, the TropicalWater and the Intermediate
Water, Chen (1996) estimated the cross-shelf
exchange of water as 3.3 104
km3yr1
(1.05 Sv) that is balanced by inputs of the
Kuroshio surface water (1.6 104km3yr1,
0.51 Sv) and the Kuroshio subsurface water
(1.7 104km3yr1, 0.54 Sv). This budget was
recently verified by D.D. Sheu through rigorous
modeling by using both salt and d18O balances.
The influxes of Kuroshio surface intrusion (0– 50 m) and subsurface upwelling (50–150 m) are
calculated to be 0.58 104 (0.183 Sv) and
1.54 104km3yr1 (0.487 Sv), respectively. The
inflow of Taiwan Strait Water is calculated to be
1.58 104km3yr1 (0.5 Sv). The outflow of shelf
water is therefore balanced by a rate of
3.83 104km3yr1 (1.22 Sv). The water budget
developed by D.D. Sheu (Table 2) is applied to
modelDOM budgets in the ECS.
Table 2also illustrates DOC, DON, and DOP budgets in the ECS. Each flux of DOM is calculated from the water flux and the representa-tive concentrations. The representarepresenta-tive concentra-tions of DOC, DON, and DOP in the Kuroshio subsurface water, the Kuroshio surface water, the Taiwan Strait Water and the exported shelf water are determined from the relationship between
Budgets of DOC, DON and DOP in the ECS Shelf
Transport Salinity d18O (%) Water flux
(m3s1)
[DOC] (mM) DOC flux (Tg
C yr1)
[DON] (mM) DON flux (Gg
N yr1)
[DOP] (mM) DOP flux (Gg
P yr1) Input Kuroshio subsurface upwelling 34.52 0.23 48.7 104 61.4 72.5 11.470.5 3.570.5 7547107 0.0570.02 25.479.5 Kuroshio surface water intrusion 34.70 0.36 18.3 104 8076 5.570.4 6.070.5 485740 0.1570.05 27.579.5 Taiwan Strait Water 33.68 0.1 50.0 104 7573 14.170.6 6.071 1289743 0.1670.06 77.9729 Changjiang inflow 0.18 8 3.44 104 143 1.86a 14.5b 221 0.6c 18.6 Precipitation 0 7 3.62 104 7274d 0.9870.05 772d 112728 0.370.1d 9.373.1 Totalinput 124 104 33.870.9 28617125 158730 Output Shelf water export 33.2 0.04 121 104 7873 35.771.2 671 32057532 0.1570.05 183756 Evaporation 0 9.5 3.09 104 0 0 0 0 0 0 Totaloutput 124 104 35.7 71.2 32057532 183756 Net transport 0 1.971.5 3447546 25763 Tg=1012g; Gg=109g.
aCauwet and Mackenzie (1993).
b
Lin et al. (1996). c
Zhang (personalcommunication). d
Based on three rain events collected from the southern ECS.
J.-J. Hung et al. / Deep-Sea Research II 50 (2003) 1127 – 1145 1141
masses. Totalinputs of DOC, DON, and DOP are largely derived from the Kuroshio subsurface water and the Taiwan Strait Water and only slightly contributed from river and rain inputs. The offshore transports of DOC, DON, and DOP are estimated from exchange rates of water and boundary concentrations for water with 33.2
salinity and 0.04 d18O. With considering
uncer-tainties associated with DOC, DON, and DOP fluxes, the total inputs are nearly balanced by the outputs. Otherwise, there are only small net
outputs for DOC (1.9 1012g C yr1), DON
(344 109g N yr1), and DOP (25 109
P yr1). The uncertainties of DON and DOP fluxes
may be rather conservative because temporal variations are not considered in estimation. There-fore, it may be appropriate to regard the budgets as matching between totalinputs and offshore outputs. This DOC budget is much smaller than
that (3–31 1012g C yr1) of shelf export from the
Cape Hatteras into the North Atlantic basin (Bates and Hansell, 1999). Nevertheless, biological DOC production over the ECS shelf may be as
large as (39.165.2) 1012g C yr1 by assuming
30–50% of primary productivity (397 mg
C m2d1) released as DOC in shelf water (Karl
et al., 1998). These DOC, DON, and DOP
produced recently over the shelf obviously do not contribute to shelf exports, they are apparently recyclable in the ECS shelf. This is consistent with the fact that the half-lives of labile and semi-labile DOC derived from recent production are much shorter than the mean residence time of shelf water. The residence time (t) of bulk DOC, DON, and DOP in the ECS shelf can be estimated from the DOM inventory and export rate [t=(DOM inventory)/(DOM export)]. The inven-tories of DOC, DON, and DOP are, respectively,
38.9 1012g C, 3.15 1012g N and 0.167 1012g
P, which are estimated by integrating
distri-butions over the shelf (0.9 106km2). The export
rate is 35.771.2 1012g C yr1 for DOC, 3.207
0.53 1012g N yr1 for DON, and 0.1870.05
1012g P yr1 for DOP (Table 2). The residence
time (t) is therefore about 1.10, 0.98, and 0.92 yr for DOC, DON, and DOP, respectively. The shorter residence time for DOP than DON and for DON than DOC is consistent with the fact that
DOP is preferentially degraded over DON and DON over DOC. Meanwhile, the residence time of DOC, DON, and DOP is on the range of residence time of shelf water reported previously (Li, 1994;Tsunogai et al., 1997;Peng et al., 1999). This also may result in small extents of net export for bulk DOM from the shelf.
It should be noted, however, that DOC input by upwelling from the Kuroshio subsurface water is actually refractory and largely resistant to bio-degradation. As bio-reactive DOM is widely recognized for playing the key role in carbon
cycling in the oceans (Carlson et al., 1994;Hansell
and Carlson., 1998), the fate and budget of this DOM fraction in the ECS shelf should be examined. If 50 mM C is taken as the level of
highly refractory pool in seawater (Fig. 11a) and1
3
of DOC concentration in river (or rain) water is
regarded as bio-degradable (Ittekkot, 1988;Moran
et al., 1999), then the net output of labile and
semi-labile DOC is estimated to be 3.870.6 1012
g
C yr1. The flux uncertainty is derived mainly from
the concentration uncertainty of labile and
semi-labile DOC, which is approximated to 13
uncer-tainty of bulk DOC flux. It may suggest that DOC produced recently can be potentially exported from the shelf with a magnitude less than 4.4
(3.8+0.6) 1012g C yr1 (upper limit), which is
just equivalent to 3.5% PP in the ECS shelf.
Alverez-Salgado et al. (2001) recently reported that the labile DOC produced in the coastal system may be exported significantly (B20% net primary production) under the upwelling condition, but periodicaltransports of DOC varied hydrodyna-mically. However, most DOC produced through primary production may be recycled in the broad ECS shelf. Following the similar assumption, the level of refractory DON derived from the Kur-oshio deep water is 3.0 mM N for the ECS seawater and about 5 and 10 mM N for rain and river waters, respectively. The potential net export of labile and semi-labile DON from the shelf would
be 4827160 109g N yr1. In case of DOP, the
level of refractory DOP is about 0.05 mM P for seawater, 0.4 mM P for riverwater and 0.2 mM P for rainwater. The potential net export of labile and semi-labile DOP is therefore estimated to be
37.5718.5 109
from upper-bound exports are 7.9 for DOC/DON, 25 for DON/DOP and 202 for DOC/DOP, from which the values are much closer to the Redfield ratio than those of bulk elemental ratios. Conse-quently, the ECS shelf may be loosely regarded as potentialsources for non-refractory DOC, DON and DOP. The exports may subsequently fuel the heterotrophic processes in the Kuroshio system and/or the North Pacific Ocean.
4. Conclusion
Distributions of DOC, DON and DOP vary spatially in the ECS. DOC concentrations are relatively high in the CCW (85–120 mM) and the Kuroshio water (75–85 mM), but low (60–70 mM) around the shelf break where the Kuroshio upwelling occurs. Temporal variations are only significant for the CCW close to the Changjiang plume in which the greater concentration occurs during summer and autumn than during spring and winter. Spatialvariations of DON and DOP are less pronounced than that of DOC. Vertical gradients of DOC, DON, and DOP concentrations were relatively strong in the thermocline of KW but weak in the ECS shelf. Stoichiometric ratios of DOC/DON, DON/DOP, and DOC/DOP gener-ally increase with depth and are much greater than the Redfield ratio, indicating the preferential decay of DOP over DON and DON over DOC. Mineralization rates of DOC and DOP are much greater in the ECS shelf than in the KW and those previous reports from open oceans, which also suggest a different lability of DOC or DOP in various waters of the ECS. The residence time estimated for shelf DOC, DON, and DOP is close to that of shelf water, implying little DOM escaped from recycling in the shelf. Thus, total inputs are nearly balanced by outputs in the shelf for bulk DOC, DON, and DOP but potentialnet exports across the shelf are obtained for degradable DOM.
Acknowledgements
The authors thank the NationalScience Coun-cil, Republic of China for financially supporting
this research under Contract Nos. NSC 87-2611-M110-007-K2, NSC 88-2621-M110-006-K2 and NSC 89-2611-M110-003-K2. The captain and crew members on the Ocean Researcher I are commended for their assistance during sampling. We are gratefulfor anonymous reviewers’ com-ments to improve the manuscript. This research is a contribution to the KEEP study, a recognized program of the JGOFS.
References
Abell, J., Emerson, S., Renaud, P., 2000. Distributions of TOP, TON and TOC in the North Pacific subtropicalgyre: implications for nutrient supply in the surface ocean and remineralization in the upper thermocline. Journal of Marine Research 58, 203–222.
Alverez-Salgado, X.A., Gago, J., Miguez, B.M., Perez, F.F., 2001. Net ecosystem production of dissolved organic carbon in a coastal upwelling system: the Ria de Vigo, Iberian margin of the North Atlantic. Limnology and Oceanogra-phy 46 (1), 135–147.
Anderson, L.A., Williams, P.J.leB., 1998. Modelling of seasonal cycling of dissolved organic carbon at station E1 in the English Channel. Estuarine, Coastal and Shelf Science 46, 93–109.
Bates, N.R., Hansell, D.A., 1999. A high resolution study of surface layer hydrographic and biogeochemical properties between Chesapeake Bay and Bermuda. Marine Chemistry 67, 1–16.
Beardsley, R.C., Limeburner, R., Yu, H., Cannon, G.A., 1985. Discharge of Changjiang (Yangtze River) into the East China sea. ContinentalShelf Research 4, 57–76.
Benner, R., Strom, M., 1993. A criticalevaluation of the analytical blank associated with DOC measurements by high-temperature catalytic oxidation. Marine Chemistry 41, 153–160.
Biscaye, P.E., Anderson, R.F., 1994. Fluxes of particulate matter on the slope of the southern Middle Atlantic Bight: SEEP II. Deep-Sea Research II 41, 459–509.
Carlson, C.A., Ducklow, H.W., Michaels, A.F., 1994. Annual flux of dissolved organic carbon from the euphotic zone in the northwestern Sargasso Sea. Nature 371, 405–408. Cauwet, G., Mackenzie, F.T., 1993. Carbon inputs and
distribution in estuaries of turbid rivers: the Yangtze and Yellow rivers (China). Marine Chemistry 43, 235–246. Chen, C.T.A., 1996. The Kuroshio intermediate water is the
major source of nutrients on the East China Sea continental shelf. Oceanologica Acta 19, 523–527.
Chen, R.F., Fry, B., Hopkinson, C.S., Repeta, D.J., Peltzer, E.T., 1996. Dissolved organic carbon on Georges Bank. ContinentalShelf Research 16, 409–420.
Chen, Y.-L., Lu, H.-B., Shiah, F.-K., Gong, G.-C., Liu, K.-K., Kanda, J., 1999. New production and F-ratio on the
continentalshelf of the East China sea: comparisons between nitrate inputs from the subsurface Kuroshio current and the Changjiang river. Estuarine, Coastaland Shelf Science 48, 59–75.
Chern, C.-S., Wang, J., 1990. On the mixing of waters at a northern offshore area of Taiwan. Terrestrial, Atmospheric and Oceanic Sciences 1, 297–306.
Connolly, J.P., Coffin, R.P., Landeck, R.E., 1992. Modeling carbon utilization by bacteria in natural water systems. In: Hurst, C.J. (Ed.), Modeling the Metabolic and Physiologic Activities of Micro-organisms. Wiley, New York, pp. 249–276.
Gong, G.-C., 1992. Chemicalhydrography of the Kuroshio front in the sea northeast of Taiwan. Ph.D. dissertation, Institute of Oceanography, NationalTaiwan University, p. 204.
Gong, G.-C., Liu, K.K., 1995. Summer time hydrography and controlof nutrient distribution in the East China Sea. In: Tsunogai, S., Iseki, K., Koike, T., Oba, T. (Eds.), Global Fluxes of Carbon and its Related Substances in the Coastal Sea–Ocean–Atmosphere System. M & J Internations, Yokohama, Japan, pp. 46–52.
Gong, G.-C., Chen Lee, Y.-L., Liu, K.-K., 1996. Chemical hydrography and chlorophyll a distribution in the East China Sea in summer: implications in nutrient dynamics. ContinentalShelf Research 16, 1561–1590.
Gong, G.-C., Wen, Y.-H., Wang, B.-W., Lin, G.-J., 2003. Seasonal variation of chlorophyll a concentration, primary production and environmentalconditions in the subtropical East China Sea. Deep-Sea Research II, this issue (doi: 10.1016/S0967-0645(03)00019-5).
Gong, G.-C., Shiah, F.-K., Liu, K.-K., Wen, Y.-H., Liang, M.-H., 2000. Spatial and temporal variation of chlorophyll a, primary productivity and chemicalhydrography in the southern East China sea. ContinentalShelf Research 20, 411–436.
Grasshoff, K., Ehrhardt, M., Kremling, K. (Eds.), 1983. Methods of Seawater Analysis. Verlag Chemie, Weinheim, p. 419.
Guo, L., Santschi, P.H., Warnken, K.W., 1995. Dynamics of dissolved organic carbon [DOC] in oceanic environments. Limnology and Oceanography 40, 1392–1403.
Hansell, D.A., Carlson, C.A., 1998. Net community production of dissolved organic carbon. Global Biogeochemical Cycles 12, 443–453.
Hopkinson Jr., C.S., Fry, B., Nolin, A.L., 1997. Stoichiometry of dissolved organic matter dynamics on the continental shelf of the northeastern USA. Continental Shelf Research 17, 473–489.
Hung, J.-J., Lin, P.-L., 1995. Distribution of dissolved organic carbon in the continentalmargin off Northern Taiwan. Terrestrial, Atmospheric and Oceanic Sciences 6, 13–26. Hung, J.-J., Lin, P.-L., Liu, K.-K., 2000. Dissolved and
particulate carbon in the southern East China Sea. ContinentalShelf Research 20, 545–569.
Ittekkot, V., 1988. Global trends in the nature of organic matter in river suspensions. Nature 332, 436–438.
Jackson, G.A., Williams, P.M., 1985. Importance of dissolved organic nitrogen and phosphorus to biological nutrient cycling. Deep-Sea Research I 32, 223–235.
Karl, D.M., Hebel, D.V., Bjorkman, K., Letelier, R.M., 1998. The role of dissolved organic matter release in the productivity of the oligotrophic North Pacific ocean. Limnology and Oceanography 43 (6), 1270–1286. Kirchman, D., Suzuki, Y., Garside, C., Ducklow, H., 1991.
High turnover rates of dissolved organic carbon during a spring phytoplankton bloom. Nature 32, 612–614. Li, Y.H., 1994. Materialexchange between the East China Sea
and the Kuroshio current. Terrestrial, Atmospheric and Oceanic Sciences 5, 625–631.
Lin, Y., Tang, R., Li, Y., 1996. Effect of biogeochemical variations of C, N and P on flocculation and settling of suspended particles in the Changjiang Estuary. In: Zhang, J. (Ed.), BiogeochemicalStudies on Chinese Major Estuaries-Element Transfer and Environment. China Ocean Press, Beijing, pp. 160–168.
Liu, K.-K., Gong, G.-C., Lin, S., Yang, C.-Y., Wei, C.-L., Pai, S.-C., Wu, C.-K., 1992. The year-round upwelling at the shelf break near the northern tip of Taiwan as evidenced by chemicalhydrography. Terrestrial, Atmospheric and Ocea-nic Sciences 3, 243–276.
Maita, Y., Yanada, M., 1990. Verticaldistribution of total dissolved nitrogen and dissolved organic nitrogen in sea-water. GeochemicalJournal24, 245–254.
Milliman, J.D., Meade, R.H., 1983. Worldwide delivery of river sediment to the oceans. Journalof Geology 91, 1–21. Moran, M.A., Sheldon Jr., W.M., Sheldon, J.E., 1999.
Biode-gradation of riverine dissolved organic carbon in five estuaries of the southeastern United States. Estuaries 22, 55–64. Norrman, B., Zweiffel, U., Hopkinson, C., Fry, B., 1995.
Production and utilization of dissolved organic carbon during an experimental diatom bloom. Limnology and Oceanography 40, 898–907.
Nitani, H., 1972. The beginning of the Kuroshio. In: Stommel, h., Yoshida, K. (Eds.), Kuroshio. University of Tokyo Press, Japan, pp. 129–156.
Ogura, N., 1972. Rate and extent of decomposition of dissolved organic matter in surface seawater. Marine Biology 13, 89–93. Orrett, K., Karl, D.M., 1987. Dissolved organic phosphorus production in surface seawater. Limnology and Oceano-graphy 32, 383–395.
Pai, S.-C., Yang, C.-C., 1990. Effects of acidity and molybdate concentration on the kinetics of the formation of the phosphoantimonylmolybdnum blue complex. Analytica Chemica Acta 229, 115–120.
Peng, T.-H., Hung, J.-J., Wanninkhof, R., Millero, F.J., 1999. Carbon budget in the East China Sea in spring. Tellus B 51, 531–540.
Ridal, J.J., Moore, R.M., 1990. A re-examination of the measurement of dissolved organic phosphorus in seawater. Marine Chemistry 29, 19–31.
Ridal, J.J., Moore, R.M., 1992. Dissolved organic phosphorus concentrations in the northeast subarctic Pacific Ocean. Limnology and Oceanography 37, 1067–1075.
Sanders, R., Jickells, T., 2000. Total organic nutrients in Drake Passage. Deep-Sea Research I 47, 997–1014.
Sharp, J.H., 1993. The dissolved organic carbon controversy: an uptake. Oceanography 6, 45–60.
Shiah, F.-K., Gong, G.-C., Liu, K.-K., 1999. Temperature vs. substrate limitation of heterotrophic bacterioplankton production across trophic and temperature gradient in the East China Sea. Aquatic MicrobialEcology 17 (3), 247–254. Shiah, F.-K., Liu, K.-K., Kao, S.-J., Gong, G.-C., 2000. The coupling of bacterial production and hydrography in the southern East China Sea: spatial patterns in spring and fall. ContinentalShelf Research 20, 459–477.
Thomas, W.H., Renger, E.H., Dodson, A.N., 1971. Near surface organic nitrogen in the Eastern TropicalPacific Ocean. Deep-Sea Research I 18, 65–71.
Thomas, H., Ittekkot, V., Osterroht, C., Schneider, B., 1999. Preferentialrecycling of nutrients-the Ocean’s way to increase new production and to pass nutrient limitation. Limnology and Oceanography 44 (8), 1999–2004. Tsunogai, S., Watanabe, S., Nakamura, J., Ono, T., Sato, T.,
1997. A preliminary study of carbon system in the East China Sea. Journalof Oceanography 53, 9–17.
Vidal, M., Duarte, C.M., Agusti, S., 1999. Dissolved organic nitrogen and phosphorus pools and fluxes in the central Atlantic Ocean. Limnology and Oceanography 44, 105–115.
Walsh, T.W., 1989. Total dissolved nitrogen in the seawater: a new-high-temperature combustion method and a compar-ison with photo-oxidation. Marine Chemistry 26, 295–311. Walsh, J.J., 1991. Importance of continental margins in the marine biogeochemicalcycling of Carbon and Nitrogen. Nature 350, 53–55.
Walsh, J.J., Biscaye, P.E., Csanady, G.T., 1988. The Shelf Edge Exchange Processes [SEEP] experiment: hypotheses and highlights. Continental Shelf Research 8, 435–456. Williams, P.M., Carlucci, A.F., Olson, R., 1980. A deep profile
of some biologically important properties in the central North Pacific gyre. Oceanologica Acta 3, 471–476. Wollast, R., 1991. The coastal organic carbon cycles: fluxes,
sources, and sinks. In: Wollast, R., Mackenzie, F.T., Chou, L. (Eds.), Interactions of C, N, and S BiogeochemicalCycles and Global Change. Springer, Berlin, pp. 195–210. Wong, G.T.F., Pai, S.C., Liu, K.K., Chen, C.T.A., 1991.
Variability of the chemical hydrography at the frontal region between the East China Sea and the Kuroshio northeast of Taiwan. Estuarine, Coastaland Shelf Science 33, 105–120.
Wotton, R.S., 1994. Particulate and dissolved organic matter as food. In: Wotton, R.S. (Ed.), The Biology of Particles in Aquatic Systems. Lewis Publishers, Boca Raton, pp. 235–288.
