New Production and F-ratio on the Continental Shelf
of the East China Sea: Comparisons Between Nitrate
Inputs From the Subsurface Kuroshio Current and the
Changjiang River
Y.-L. Lee Chen
a*, H.-B. Lu
a, F.-K. Shiah
b,c, G. C. Gong
d, K.-K. Liu
band
J. Kanda
eaDepartment of Marine Resources, National Sun Yat-sen University, Kaohsiung, Taiwan, Republic of China bInstitute of Oceanography, National Taiwan University, Taipei, Taiwan, Republic of ChinacGlobal Change
Research Center, National Taiwan University, Taipei, Taiwan, Republic of ChinadDepartment of Oceanography, National Taiwan Ocean University, Keelung, Taiwan, Republic of ChinaeSchool of Science, Shizuoka University, Shizuoka, Japan
Received 28 July 1997 and accepted in revised form 20 July 1998
This paper reports the first results of the direct measurements of nitrate-based new production and f-ratio on the continental shelf of the East China Sea. The Kuroshio-induced upwelling off north-eastern Taiwan and river runoff from the Changjiang are two principal sources of new nitrogen on the shelf. New production ranged from 70 to 1610 mgC m2d1, and values of the f-ratio were 0·17 to 0·82. Enhanced new production in the upwelling was significantly related to the ambient nitrate concentration. This implies that the nutrients brought up to the euphotic zone by the intrusion of the subsurface Kuroshio water were quickly reflected by the enhancement of new production. The prediction of f-ratio is feasible in the upwelling region from two ship-measured parameters: light intensity and surface nitrate concentration, but not feasible in the river-influenced shelf waters. New production on the riverine shelf waters, in contrast, was not related to the nitrate input from the Changjiang, but was positively (P<0·05) related to water temperature. Low rates of NO3 utilization imply that factors other than nitrate (e.g. phosphate or light) could be the limiting factors determining the new production dynamics in the river mouth. 1999 Academic Press Keywords: new production; f-ratio; East China Sea; upwelling; Changjiang
Introduction
Primary production can be partitioned into new production and regenerated production based on the source and form of nitrogen (N) (Dugdale & Goering, 1967). This has worked well for most oceanic waters because N is often the most limiting nutrient
(Vitousek & Howarth, 1991). Regenerated
produc-tion is the part of producproduc-tion dependent upon auto-chthonous N, mainly ammonium (NH4+) and urea, derived from the metabolic products of local biologi-cal processes. New production is the part of produc-tion dependent upon allochthonous N sources. These sources include nitrate (NO3) mixing from deep ocean reserves, NO3 and other N species trans-ported by river runoff, and atmospheric deposition and biologically fixed N from the atmosphere. New production is considered to be the fraction of primary production that can be removed from the surface
mixed layer as fish yields or subpycnocline sedimen-tation without destroying the long-term integrity of a pelagic ecosystem (Vezina & Platt, 1987). Therefore, estimation of new production is essential to under-stand the biogeochemical process or to manage the resource balance of any oceanic ecosystem.
The East China Sea (ECS) is located on a large continental shelf bounded by the Kuroshio Current on the slope side and the coast of China on the other. It is one of the most productive areas of the world oceans. The Changjiang (Yangtze) River, which emp-ties into the ECS at about 3130 N latitude, ranks as the fifth largest river in the world with regard to freshwater discharge (9·241011m3year1) and the fourth with regard to solid discharge (4·86108tons
year1) (Tian et al., 1993). This discharge clearly
provides large nutrient loads to the ECS shelf. Con-centrations of NO3, silicate and phosphate in the Changjiang were reported to be as high as 32·9, 95 and 0·57ìM, respectively (Zhang, 1996). A net
*To whom the correspondence should be addressed.
oceanic flux of NO3 was estimated to be as much as 60109mol year1 (Edmond et al., 1985). Off north-eastern Taiwan, at the southern part of the ECS, a permanent upwelling occurs when the Kuroshio Current intrudes onto the continental shelf
(Chern et al., 1990). Nutrients, including nitrate,
phosphate and silicate (Wong et al., 1991), contained within subsurface Kuroshio water are brought up to the euphotic zone at the shelf break around 2520 N latitude. High phytoplankton biomass (Chen, 1992;
1994; 1995a) and primary productivity (Chen,
1995b) were reported.
The new production regime in the Changjiang estuary is likely to be far more complicated than the simple model depicted by Dugdale and Goering
(1967). In their original model, nitrate from depth is
the only new nitrogen source considered. Non-nitrate new nitrogen such as ammonium (NH4+) or organic N are supplied in significant quantities by the Changjiang river. Unlike what is observed in most oceanic environments, phosphate, not N, limits pri-mary production near the river mouth due to high N/P in the river inflow (Harrison et al., 1990), and this potentially means another definition of new production based on phosphate is required.
Although it has been well recognized that both the Changjiang river runoff and the Kuroshio upwelling are two principal sources of new N, it is still unclear how these N inputs affect the primary production and f-ratio of the ECS. The purpose of this study was to elucidate and to compare the controlling mechanisms of new production in the upwelling and the riverine shelf water regions. The authors focussed their analy-sis on nitrate-based production, which is likely to be the dominant portion of new production, and then compared it with the available estimates of other new N sources. Regression models also were developed to predict the nitrate-based new production in these waters.
Methods
Experiments and sampling were carried out in two water regions. The upwelling region was studied during three cruises: OR1-416 (April 15–22, 1995), OR2-124 (June 10–14, 1995) and OR1-431 (September 27–October 3, 1995), covering the same transect (stations 1–17, Figure 1). The continental shelf region influenced by runoff of the Changjiang River was studied during Cruise OR1-449 (May 2–15, 1996; stations E1–E7).
For hydrographic observations, a rosette multi-sampler with attached CTD probes (Seabird SBE
9/11) was used, and water samples were collected from 3, 10, 25, 50, 75 and 100 m. Surface water (0 m) was collected with a bucket. Water samples were filtered onboard ship using Whatman GF/F glass-fiber filters under low vacuum (<100 mmHg). The filter samples were kept in darkness and frozen at 20 C until analysis. Chlorophyll a concentration was deter-mined on extracted filtrate in 90% acetone for 20 h
(Strickland & Parsons, 1972) using a fluorescence
spectrophotometer (F3000, Hitachi Co., Tokyo, Japan). The mean of two replicates is reported below for all observations. Nitrate and phosphate concen-trations were analysed by standard spectrophoto-metric methods (Strickland & Parsons, 1972). Surface and underwater light intensities were measured either by a quantum meter (Li-188B, Li-Cor Inc., U.S.A.) or a PAR sensor (OSP200L, Biospherical Inc., U.S.A.). The depth of the euphotic zone was defined as 0·6% of the surface light penetration. The depth of mixed layer (Zm) was defined as the depth where the density (óè) gradient exceeded 0·1 m1.
Primary production and new production measure-ments were conducted at selected stations (3 to 7 stations) during each cruise (Table 1). All the produc-tion staproduc-tions were located on the continental shelf with water depth less than 200 m.
During the upwelling cruises, water for measure-ment of photosynthetic production was collected by 20-L Go-Flo bottles from the surface (100% light intensity) and from the light penetration depths (LPD) equivalent to 46, 38, 13, 2 or 0·6% of the surface irradiance. The water was immediately trans-ferred into 2·3-L transparent polycarbonate bottles at LPD equivalents of 100, 46, 38, 13, 2 or 0·6% using neutral density screens. Two light bottles and one dark bottle were incubated at each LPD. The dark bottle was covered with four layers of black plastic sheets to ensure total darkness. The bottles were incubated on deck in transparent water tanks with flow-through of
in situ surface seawater under natural light for 24 h.
Before incubation, NaH13CO
3 (99 atom %, Isotec, Ohio, U.S.A.) was added to the primary production bottles so that the final concentration was 0·19 mM; Na15NO
3was added to the new production bottles at a concentration equivalent to 10% of ambient NO3 concentration. The initial and the incubated water were filtered through precombusted (450C, 4 h glass fiber filters (Whatman GF/F). The filter papers were then treated with HCl fumes for 2 h to remove carbon-ate and were completely dried in a vacuum desiccator. Particulate organic carbon, particulate nitrogen and the isotopic ratios of13C/12C and15N/14N were deter-mined by a mass spectrometer (ANCA-MASS 20-20, Europa Scientific Ltd., Crewe, U.K.). Measurement
of13C atom % and15N atom % were reproducible at
a precision of 0·0002 and 0·0003%, respectively. Concentration of total inorganic carbon was determined from alkalinity (Strickland & Parsons,
1972).
Primary production measurements in the
Changjiang riverine shelf water was carried out by the 14C assimilation method (Parsons et al., 1984). Water samples were taken before dawn. Three light and one dark 250 ml polycarbonate bottles were filled with water and inoculated with NaH14CO3 (final concen-trations 10ìCi ml1). One light bottle was filtered
immediately as the time-zero sample. Two light and one dark bottles for each depth were tethered onto an
in situ array. The array was anchored with a 30–50 kg
weight at the bottom to position the samples at their original depths. Following retrieval of bottles after 4–5 h of incubation (from about 8:00 to noon), the water sample in each bottle was filtered through a Whatman 25 mm GF/F filter. The filters were then placed in scintillation vials with 0·5 ml of 0·5N HCl to remove residual H14CO3. Radioactivity in each vial was measured in a liquid scintillation counter (Packard 1600) after addition of 10 ml scintillation
24° 121° 26° Changjiang River 120° 41 27° 28° 29° 30° 31° 32° 25° N 137 122° 123° 124° 125° 126° 127° 128°E Taiwan Taiw an Strait E6 E5 E3 1 2 3 4 5 6 7 8 9 10 11 1213 14 15 16 17 10a 8a E7 E4 20 0m 1000 m E1 E2 China 31 21 117 127 Japan Sea Pacific Ocean Yellow Sea East China Sea
F 1. Sampling stations on the East China Sea and the Kuroshio. The transect among stations 1 to 17 were sampled during cruise 416, 124 and 431; Stations E1 to E7 were sampled during cruise 449.
cocktail (Ultima Gold, Packard). Nitrate uptake measurements were carried out concurrently with primary production by inoculating15NO3in two 1·0 l polycarbonate bottles and incubating in situ for 4– 5 h. 15NO3 assimilated was determined by a mass spectrometer (VG Micro Mass 622). The uptake measured during the morning hours was multiplied by a factor of approximately three to obtain the daily assimilation rate in accordance with the local diel cycle which was previously established (unpublished results). Particulate organic nitrogen concentration was measured by a CHN analyser (Carlo-Erba EA 1400) connected to the mass spectrometer.
Two incubation schemes were used in the present study. The 24 h incubation for the upwelling area could possibly lead to underestimation of nitrate uptake because of isotope depletion during prolonged incubations. Turnover times for the observed NO3 (observed NO3 concentration/NO3 uptake rate) on the two upper incubation depths in the upwelling region were mostly in the range of <0·8–1·9 day. This implies that the nitrate uptake rates are probably underestimated due to substrate exhaustion. The incubation in the riverine shelf water was, by compari-son, short. Underestimation of rates in riverine shelf waters was unlikely and even if underestimation were to occur this would pose no problem in our conclusion since it could occur only in the upwelling region where
nitrate uptake rates were high. In contrast, low uptake rates were observed at the Changjiang riverine shelf region. These low uptake values could not have resulted from underestimation because the incubation time was relatively short.
Primary production was calculated (Hama et al., 1983) from the uptake rate of13C or14C.Hama et al.
(1983)reported excellent agreement between the14C
and 13C methods. Nitrate assimilation rate was calculated following the equations (2) and (3) of
Dugdale and Wilkerson (1986). From the nitrate
assimilation rate and the Redfield ratio (C:N=6·6), nitrate-based new production in unit of carbon was then obtained (Dugdale et al., 1989). f-Ratio was calculated as new production divided by primary production. Water column integrated production per m2was calculated by trapezoidal integration from 100 to 0·6 LPD to provide daily integrated primary pro-duction (IPP) and daily integrated new propro-duction (INP) per m2.
A step-wise multiple regression analysis procedure
(Draper & Smith, 1981) was used to examine the
spatial relationships of nitrate-based new production, primary production, NO3 concentration, water temperature and chlorophyll a concentration. The relationships between chlorophyll normalized uptake rates and light intensity were fitted with Michaelis-Menten equations by SigmaPlot.
T 1. Location, bottom depth, depth of euphotic zone (0·6% of the surface light intensity) and the depth of mixed layer (Zm, difference of óè value <0·1 m1) of the sampling stations where new
production and primary production were measured during various cruises
Cruise (date) Station Location Bottom depth (m) Euphotic layer depth (m) Zm (m) Longitude (E) Latitude (N)
416 (15–22 April 1995) 5 12110 2600 85 42 >76 9 12150 2540 125 54 >102 11 12210 2530 145 84 70 124 (10–14 June 1995) 8 12140 2544 117 68 77 8a 12144 2545 123 73 70 9 12150 2540 126 73 53 10 12200 2535 118 30 >100 10a 12205 2530 136 40 3 431 (27 September– 3 October 1995) 9 12150 2540 122 68 54 10 12200 2535 116 79 38 11 12210 2530 123 84 >102 449 (2–15 May 1996) E1 12348 3007 60 36 10 E2 12159 2820 50 29 10 E3 12016 2628 45 38 5 E4 12408 2838 80 27 40 E5 12246 2635 106 55 20 E6 12408 2716 100 71 30 E7 12537 2902 100 66 60
34.9 34. 8 34. 34 .6 34 .5 34 .4 34 .3 34 .2 34.3 34 .2 34 .1 34.234.4 34 .3 34 .1 34 .0 33 .8 33 .9 33 .7 27 2524 2326 22 21 2 0 19 18 17 16 17 18 25 24 23 22 21 20 19 26 27 34.3 5 34 .4 0 34 .4 5 34. 40 .50 34.534 5 34 .6 0 34 .6 5 34 .50 34.4 5 34 .6 0 345.6 34 .5 5 70 75 34. 34. 34 .80 7 8 9 6 5 4 3 2 1 2 1 1. 2 1.0 0.80.6 0.4 0.8 0.2 0.4 60. 0. 2 1.0 1.00.8 0.2 0.1 0. 2 0. 30 .4 0. 5 0. 6 1.0 1.1 0.9 0.9 0.2 0.3 0.3 0.4 0.50.6 0.70.8 0. 3 0. 50.4 Depth (m) Station 200 0 180 160 140 120 100 80 60 40 20 T emperature ( ° C) CR416 (April 15–22, 1995) 200 0 180 160 140 120 100 80 60 40 20 Salinity 200 0 180 160 140 120 100 80 60 40 20 0 20 40 60 80 100 120 140 160 180 200 Chlorophyll a (mg/m 3 ) 200 0 180 160 140 120 100 80 60 40 20 T emperature ( ° C) CR124R (June 10–14, 1995) 200 0 180 160 140 120 100 80 60 40 20 Salinity 200 0 180 160 140 120 100 80 60 40 20 0 20 40 60 80 100 120 140 160 180 200 Chlorophyll a (mg/m 3 ) 200 0 180 160 140 120 100 80 60 40 20 T emperature ( ° C) CR431 (Sep 27–Oct 4, 1995) 200 0 180 160 140 120 100 80 60 40 20 Salinity 200 0 180 160 140 120 100 80 60 40 20 200 0 180 160 140 120 100 80 60 40 20 Chlorophyll a (mg/m 3 ) 0.2 NO 3 – ( µ M) NO 3 – ( µ M) NO 3 – ( µ M) 18 3.5 34.00 45 7 8 9 1 0 1 1 1 2 1 3 1 4 16 7 35 9 1 0 1 1 1 2 1 3 1 5 1 7 1 3 5 7 9 10 1 1 12 15 1 45 7 8 9 1 0 1 1 1 2 1 3 1 4 16 7 35 9 1 0 1 1 1 2 1 3 1 5 1 7 1 3 5 7 9 10 1 1 12 15 1 45 7 8 9 1 0 1 1 1 2 1 3 1 4 16 7 35 9 1 0 1 1 1 2 1 3 1 5 1 7 1 3 5 7 9 10 1 1 12 15 1 45 7 8 9 1 0 1 1 1 2 1 3 1 4 16 7 35 9 1 0 1 1 1 2 1 3 1 5 1 7 1 3 5 7 9 10 1 1 12 15 1 28 26 27 25 24 23 22 21 2019 17 16 15 26 25 2 1 3 4 5 7 9 8 6 21 7 1. 8 1.61.20.8 0.6 0.4 1. 4 01.60.40. 0. 4 2 0. 0. 2 0.6 0.4 0.2 1.41.2 1. 0 0. 8 0. 6 2 1 2 10 5 1 0.5 1. 5 1. 0 0.5 0.5 0 11. .52. 0 2.5 1.0 1. 5 3.0 5. 0 4. 5 4. 0 33 .00 34.0 0 34. 40 34. 4550 34. 34 .5 5 0 .6 3434 .6 5 34. 60 34.6 0 34. 65 34 .7 0 34 .7 5 34 .8 0 34.7534.70 21 20 19 1718 21 20 18 19 21 22 24 20 21 22 23 19 18 17 24 25 F 2. Cross-sections of temperature, salinity, nitrate concentration and chlorophyll a concentration on the transect among stations 1–17 during 3 d iff erent cruises.
Results
The upwelling region
The plume centre of the upwelling at the shelf break could be easily located by hydrography. At the plume centre, the surface water was lower in temperature
and higher in salinity, nitrate, and chlorophyll a concentration than the adjacent waters (Figure 2). Although the plume centre shifted slightly from cruise to cruise, it always appeared near the shelf break. These upwelling characteristics were observed repeat-edly for all cruises and confirm the persistence of the
CR-416 CR-124 CR-431 100 150 40 ST.10A 30 60 90 120 80 20 60 0 100 150 40 ST.11 30 60 90 120 80 20 60 0 100 150 40 ST.10 30 60 90 120 80 20 60 0 100 150 40 ST.11 30 60 90 120 80 20 60 0 Depth (m) 100 150 40 ST.9 30 60 90 120 80 20 60 0 100 150 40 ST.9 30 60 90 120 80 20 60 0 100 150 40 ST.10 30 60 90 120 80 20 60 0 100 150 40 ST.5 PP and NP (mgC m–3 d–1) 30 60 90 120 80 20 60 0 100 150 40 ST.8A 30 60 90 120 80 20 60 0 100 150 40 ST.9 PP and NP (mgC m–3 d–1) 30 60 90 120 80 20 60 0 100 150 40 ST.8 PP and NP (mgC m–3 d–1) 30 60 90 120 80 20 60 0 PP NP
F 3. Vertical distribution of primary production (PP) and nitrate-based new production (NP) measured during three different cruises.
upwelling (Chern et al., 1990). Surface nitrate con-centrations as high as 2ìM were observed at the location where lowest surface water temperature was measured (Figure 2). This high nitrate concentration resulted in a high chlorophyll a concentration in the euphotic zone (Figure 2). Shelf waters between Stations 1 and 3 near the coast of China, with the distinctive characteristic of low salinity, were also high in chlorophyll a concentrations.
Depth profiles of primary and nitrate-based new production in the shelf break and the adjacent shelf waters showed high values in the surface layer and decreased with depth (Figure 3). This suggests that light is possibly more important than nitrate in
deter-mining the vertical profile of production even though nitrate is usually considered a significant factor in controlling primary or new production in these subtropical waters. Vertical distributions of primary production (PP) and nitrate-based new production (NP) were also related significantly (P<0·01) to in situ chlorophyll a concentration. Waters with higher chlorophyll a concentration ([Chl], mg m3) and average light intensity at the depth of incubation (Light,ìE m2s1), i.e. shallower waters, showed a higher primary or new production than those of low-chlorophyll deep water. Correlation between PP and NP was also quite high (r=0·95,Figure 4). Multiple regression analysis depicted their relationships as:
7 70 –10 –1 NP (mgN m–3 d–1) PP (mgC m –3 d –1 ) 5 50 60 40 30 20 10 1 2 3 4 6 Y = 8.06 + 8.67X r2 = 0.67 0 0 (b) 30 160 –20 –5 NP (mgN m–3 d–1) PP (mgC m –3 d –1 ) 20 100 120 80 60 40 20 5 10 15 25 Y = 8.22 + 5.91X r2 = 0.91 0 0 (a) 140
F 4. Relationship between primary production (PP) and nitrate-based new production (NP) in (a) the upwelling region and the (b) Changjiang riverine shelf region respectively.
PP=15·39+56·24 [Chl]+0·038 Light (R2=0·59); NP=19·36+55·29 [Chl]+0·023 Light (R2=0·54) The Michaelis-Menten equation was able to describe the relationships between chlorophyll normalized uptake rates and average light intensity during the incubation for each light depth at most of the stations
(Table 2). Results of two representative stations are
shown inFigure 5.
Primary production and new production were higher in shelf break waters than in adjacent waters. At the shelf break, primary production was usually >1 gC m2day1 with a maximum value of 1·96 gC m2day1 (Table 3). In contrast, primary production in the non-upwelling shelf water ranged
from 0·43 gC m2day1 at Station 5 to
0·88 gC m2day1 at Station 8a. New production
for all stations ranged between 0·08 and
1·61 gC m2day1 (Table 3). New production was generally high near the shelf break; while low values were observed at the mid-shelf stations (Stations 5, 8 and 8a). However, low new production was observed at some stations near the shelf break. For example,
new production at Stations 9 and 10 during cruise 431 were 0·27 and 0·14 gC m2day1, respectively. Integrated new production (INP) was positively cor-related with integrated primary production (IPP) and surface nitrate concentration (NO3-surf) (Table 4). The chlorophyll normalized new production (INP/ IChl, integrated new production/integrated chloro-phyll a, mgC mgChl1day1) was also significantly and positively correlated to chlorophyll normalized primary production (IPP/IChl) and the surface nitrate concentration (Table 4). The combined regression equation is:
INP/IChl=0·9801+0·693 IPP/IChl+6·31 NO3-surf (R2=0·89)
Integrated f-ratio (INP/IPP) was significantly related to the average on-deck light intensity during incubation (Light, ìE m2s1) and the surface nitrate concentration (Table 4). The combined equation is:
f-Ratio=0·127+0·0002427 Light+0·176 NO3-surf (R2=0·72)
T 2. Parameters of Michaelis-Menten equations describing the relationships between average light intensity (L, ìE m2s1) during the incubation for various depths and nitrate uptake rate
(ñNO3/Chl, mg at-N mg Chl1day1) or carbon uptake rate (ñC/Chl, mg at-C mg Chl1day1).
The equation fitted isñmax·L/(KL+L) whereñmaxand KLare the parameters
Cruise/ station ñNO3/Chl ñC/Chl Average surface light intensity ñmax KL R 2 ñ max KL R 2 Cruise 416 St. 5 — — 343* St. 9 — 20·94 1469 0·92 1322* St. 11 2·00 1328 0·89 21·34 1466 0·94 1327* Cruise 124 St. 8 0·45 225·6 0·99 9·65 635·4 0·99 563* St. 8a — 7·02 80·43 0·96 747* St. 9 0·29 180·5 0·83 10·1 398·1 0·84 923* St. 10 1·61 1029 0·97 16·0 831·9 0·99 1232* St. 10a 1·74 270·1 0·83 5·8 451·5 0·94 826* Cruise 431 St. 9 0·45 302·4 0·74 10·3 455·5 0·97 1068* St. 10 0·13 100·7 0·90 8·09 329·4 0·92 472* St. 11 0·52 497·1 0·98 4·72 72·29 0·85 1277* Cruise 449 St. E1 0·21 115·8 0·72 — 644† St. E2 — 3·26 111·8 0·97 968† St. E3 0·71 549·1 0·85 4·69 97·95 0·76 1018† St. E4 0·37 220·9 0·97 2·66 56·21 0·98 226† St. E5 — 3·34 159·1 0·93 579† St. E6 — 5·09 56·41 0·87 773† St. E7 — 4·46 90·55 0·91 1065†
—Indicates poor fitting by Michaelis-Menten equation *data from 0 m
Thus the prediction of f-ratio in this region is feasible from ship-measured parameters which are often readily available.
The Changjiang riverine shelf region
Primary production and nitrate-based new production measured in river-influenced shelf waters generally decreased with increasing depth, similar to what was observed in the upwelling region (Figure 6). Corre-lation between PP and NP was also high (r=0·82,
Figure 4). The magnitude of primary production and
new production also related significantly (P<0·01) to chlorophyll a concentration. Multiple regression analysis examining the effects of light availability and chlorophyll a concentration revealed the following relationships:
PP=0·645+17·80 [Chl]+0·021 Light (R2=0·72) NP=1·092+8·205 [Chl]+0·008 Light (R2=0·48);
Both riverine shelf water and upwelling water showed a similar pattern with maximum
chlorophyll-specific production occurring in or near the surface layer, implying that their vertical distributions were light-dependent rather than nutrient-dependent. There were, however, observations at several stations that did not follow the trend that production decreased with depth. Light-dependence parameters obtained in this region are listed in Table 2. An example is shown for Station E2 where a significant relationship was not established (Figure 7).
Nitrate-based new production measured during cruise 449 ranged from 0·07 to 0·34 gC m2day1
and primary production from 0·25 to
0·75 gC m2day1. In contrast to what was observed in the upwelling region, the variation of integrated new production was related to neither surface nitrate concentration, primary production, chlorophyll a concentration, nor incubation light intensity (Table 5). However, both f-ratio and the integrated chlorophyll-specific new production was significantly and positively related to the surface water temperature over the range 15–21C for all stations
(Table 5). There was no significant relationship
Carbon uptake rate (mg at C mg chl
–1 d –1 ) 0 300 0.1 Light (µE m–2 s–1) Y = 0.52X/(X + 497.1) R2 = 0.98
Nitrogen uptake rate (mg at N mg chl
–1 d –1 ) ST.11 (CR431) 600 900 1200 1500 0.2 0.3 0.4 0.5 0 300 1 Y = 4.72X/(X + 72.3) R2 = 0.85 ST.11 (CR431) 600 900 1200 1500 2 3 4 6 0 300 0.05 Y = 0.13X/(X + 100.7) R2 = 0.90 ST.10 (CR431) 600 900 1200 1500 0.10 0.15 0.20 0 300 1 Y = 8.09X/(X + 329.4) R2 = 0.92 ST.10 (CR431) 600 900 1200 1500 2 3 4 7 Light (µE m–2 s–1) 5 6 5
F 5. Effects of the average light intensity (ìE m2s1) during the incubation on nitrate (mg at-N mg Chl1day1) and carbon uptake rate (mg at-C mg Chl1day1) at respective light depth in Station 10 (a shelf station) and Station 11 (an upwelling station) investigated during cruise 431.
between the new production and the surface nitrate concentration (Table 5).
Light intensity seemed not to be crucial in affecting the spatial variation of integrated new production on the continental shelf off the Changjiang. There was no significant relationship between integrated new pro-duction and the average on-deck light intensity during incubation. There was a subsurface maxima of new production at Station E2 (Figure 7) where the surface nitrate concentration was high and the integrated new production was low. Maximum new production was located at 10–15 m (four times greater than that at 2 m) in the euphotic zone where nitrate was abundant and uniformly distributed. Similarly high, subsurface new production was also observed at Stations E1, E7 and E6.
Water column f-ratio ranged between 0·17 (Station E2) and 0·51 (Station E6) (Table 6). Its variation was positively related to surface water temperature
(Table 5), but not to light intensity, surface nitrate
concentration, and integrated nitrate concentration
(Table 5).
Discussion
Data review, NP and PP
The present study reports the first values of nitrate-based new production on the continental shelf of the East China Sea. Rates ranged from 80 to 1610 mgC m2day1 in the upwelling region and from 70 to 340 mgC m2day1 in the broad shelf
T 3. Water column integrated primary production (gC m2day1), integrated new production (gC m2day1) and f-ratio obtained at seven sampling stations during three different cruises
Cruise Station 5 8 8a 9 10 10a 11 Primary production 416 0·43 1·00 1·03 124 0·71 0·88 1·32 1·22 1·96 431 0·66 0·60 1·43 New production 416 0·08 0·77 0·66 124 0·30 0·32 0·36 0·80 1·61 431 0·27 0·14 0·72 f-ratio 416 0·19 0·77 0·64 124 0·42 0·36 0·27 0·66 0·82 431 0·41 0·23 0·50
T 4. Matrix of correlation coefficients (r) for the variables f-ratio (INP/IPP), integrated new production (INP), integrated primary production (IPP), chlorophyll a normalized integrated new production (INP/Ichl), chlorophyll a normalized integrated primary production (IPP/Ichl), integrated chlorophyll a concentration (Ichl), integrated nitrate concentration (INO3), surface nitrate concentration (NO3-surf), surface temperature (Temp-surf) and average surface light intensity
(Light) obtained in the upwelling region northeast Taiwan
Variables f-ratio INP IPP INP/Ichl IPP/Ichl Ichl INO3 NO3-surf Temp-surf
INP 0·88** IPP 0·67* 0·89** INP/Ichl 0·87** 0·99** 0·87** IPP/Ichl 0·64* 0·87** 0·96** 0·90** Ichl 0·52 0·50 0·59 0·37 0·34 INO3 0·16 0·06 0·16 0·07 0·12 0·20 NO3-surf 0·77** 0·86** 0·73* 0·82** 0·68* 0·50 0·02 Temp-surf 0·34 0·32 0·26 0·24 0·12 0·45 0·49 0·53 Light 0·68* 0·46 0·46 0·45 0·42 0·50 0·19 0·48 0·22 *P<0·05, **P<0·01.
8 100 80 Depth 60 40 20 0 100 80 Depth 60 40 20 0 100 80 Depth 60 40 20 0 100 80 Depth 60 40 20 0 100 80 Depth 60 40 20 0 100 80 Depth 60 40 20 0 100 80 Depth 60 40 20 0 T e mperature ( ° C) 12 14 16 18 20 22 E1 E2 E3 E4 E5 E6 E7 Salinity 30 31 32 33 34 35 σθ (kg/m 3 ) 22 23 24 25 26 NO 3 ( µ M) 0246 1 0 1 2 Chlorophyll a (mg m –3 ) 0.0 0.5 1.0 1.5 2.0 2.5 PP and NP (mgC m –3 d –1 ) 0 2 04 06 08 0 NP PP F 6. Vertical distribution of temperature, salinity, density (óè ), nitrate concentration (NO 3 ), chlorophyll a concentration (Chlorophyll a ), primary productivity (PP) and new productivity (NP) measured at seven stations during cruise 449.
area including the Changjiang-influenced waters. These values are much higher than previously calcu-lated estimates of 80 mgC m2day1 byLi (1995)
and 7322 mgC m2day1byChen (1996). Simi-larly, much higher f-ratios were obtained from our direct measurements (0·17–0·82) than those esti-mated by Chen (0·15) and by Li (<0·16). Average f-ratio was 0·48 for the upwelling region and 0·34 for the Changjiang continental shelf region. These average values were computed by placing equal weight on each of the stations, regardless of their uneven distribution and sampling frequency. For comparison, f-ratios as low as 0·05 were reported for oligotrophic central oceans and as high as 0·5 in a productive coastal-upwelling regions (Harrison et al.,
1987).
Primary production observed in this study ranged from 430 to 1960 mgC m2day1 in the upwelling region (Table 3). The data are consistent with pre-vious observations in a similar area (Chen, 1995b). The values ranged from 250 to 750 mgC m2day1 in the shelf region off Changjiang (Table 6), and also comparable with, but slightly lower than, the value obtained by Hama et al. (1997), who reported the average primary productivity of 750 mgC m2day1 for the shelf area close to this study. High productivity in the ECS, especially at the upwelling plume was confirmed in this study.
Effect of upwelling on NP
Both new production and primary production in the upwelling region are under strong control of nutrient
supply via upwelling. N is generally considered to be a limiting element in the dynamics of new production. This is clearly shown in the upwelling region of the present study (Figure 8). The upwelling INP and IPP were related to NO3-surf, but not to INO3. This implies that INP and IPP are responding to a factor to which NO3-surf is related, not necessarily to NO3 concentration itself. NO3-surf likely represents ‘ strength ’ of the upwelling or the rate of supply of deep water and N to the photic zone. The present upwelling had a maximum surface nitrate concen-tration of 2ìM and f-ratios of 0·23–0·82. In contrast, the Peruvian upwelling has f-ratios of 0·10–0·84 and a maximum surface nitrate concentration of >20ìM
(Wilkerson & Dugdale, 1987). Nitrate supply from
the subsurface Kuroshio waters to these subtropical waters appears to contribute in a similar manner to primary production as in the Peruvian upwelling. The high f-ratio and correlation of INP and NO3-surf in this study indicate that phytoplankton response to the upwelling is faster than in the Peruvian upwelling system, where time-lags of phytoplankton growth are often observed.
The distribution of high new production and f-ratio matched the distribution of low temperature and high nitrate concentration at the centre of the upwelling water. The chain reaction from physical effect (up-welling) to chemical change (high nitrate) to the biological response (high INP) was clearly demon-strated. The behaviour of the Kuroshio Current there-fore is an important force in determining the dynamics of new production in these shelf waters.
Control of the Changjiang NP:(1) Temperature
Unlike the Kuroshio Current, the nitrate-rich Changjiang runoff did not bring similarly high, nitrate-based new production to its immediate shelf water. Station E2 with a surface nitrate concentration as high as 6·4ìM, higher than the maximum surface concentration of 2ìM observed in the upwelling, had the lowest new production (70 mgC m2day1) and f-ratio (0·17) of all stations (Figure 8). The subsurface Kuroshio water is warm and salty in contrast to the cold and less saline turbid Changjiang plume. Water temperature was positively related to new production off the Changjiang continental shelf region. The Changjiang-influenced water was both low in surface temperature and new production. Positive correlation between water temperature and nitrate uptake rate has been previously reported (Slawyk, 1979; Le
Bouteillier, 1986) where factors such as light intensity
and nutrient concentration were relatively uniform. There are conflicting reports depicting the linkage
1000 0.10 Light (µE.m–2.s–1) NO 3 uptake rate ( µ g-at NO 3 . µ g c hla –1 ·d –1 ) 0.08 0.06 0.04 0.02 200 400 600 800 0
F 7. Relationship between light intensity (ìE m2 S1) and nitrate uptake rate (ìg at-N ìg Chl1day1) at Station E2.
between surface temperature and new production.
Kamykowski and Zentara (1986) found that surface
temperature of water in an upwelling system was low whilst concentration of nitrate was high; new produc-tion was high in upwelled water despite the low surface temperature. However temperature effect in this study could also be a surrogate of the effects linking to the water mass rather than its direct physiological influence on phytoplankton cells.
Control of the Changjiang NP:(2) PO4
Although the upwelling and the
Changjiang-influenced waters were both rich in nitrate, there were differences in other nutrient concentrations. Our study showed that in situ measurements of N:P ratios ([NO3]/[PO43]) in the water column of the euphotic zone were mostly below the Redfield ratio of 16:1 (Table 7), with the exception of Station E2. This indicates that most of the water on the shelf was potentially more insufficient in nitrate than in phos-phate. On the contrary, N:P ratios as high as 213·3 were observed in the surface water of Station E2, and
its phosphate concentration was at the limit of detec-tion (0·03ìM). Phosphate concentration was un-detectable in the upper 20 m of Station E4. This implies that within the river-influenced water, such as Station E2, the limitation on production was probably from phosphate instead of N. There were other reasons for supporting the differences in N:P ratio between these two nitrate-rich waters. Chen (1996)
reported that the nutrient flux from the Kuroshio water to the shelf was 14·3:1 which is close to the 16:1 Redfield ratio, whereas N (nitrate+ammonium):P ratio from the Changjiang river fluxes was 126:1. A steep gradient of nitrate (65–15ìM), phosphate (0·7–0ìM) and N:P ratio (90–150) was found over the salinity gradient from the river mouth to the open sea (Harrison et al., 1990). Mesocosm experiments off the coast of mainland China indicated that phyto-plankton growth was limited by phosphorus rather than by N (Harrison et al., 1990). Station E2 had a surface salinity of 30·04 and was within the range of 0 to 32 in the phosphate-limiting region described by
Harrison et al. (1990). It is, thus, very likely that
the water at this station was phosphate-limited rather
T 5. Matrix of correlation coefficients (r) for the variables f-ratio (INP/IPP), integrated new production (INP), integrated primary production (IPP), chlorophyll a normalized integrated new production (INP/Ichl), chlorophyll a normalized integrated primary production (IPP/Ichl), integrated chlorophyll a concentration (Ichl), integrated nitrate concentration (INO3), surface nitrate concentration (NO3-surf), surface temperature (Temp-surf) and average surface light intensity
(Light) obtained in the continental shelf region off the Changjiang
Variables f-ratio INP IPP INP/Ichl IPP/Ichl Ichl INO3 NO3-surf Temp-surf
INP 0·79* IPP 0·14 0·47 INP/Ichl 0·73 0·88** 0·34 IPP/Ichl 0·10 0·54 0·72 0·73 Ichl 0·02 0·01 0·09 0·47 0·57 INO3 0·21 0·01 0·12 0·28 0·41 0·53 NO3-surf 0·67 0·57 0·14 0·49 0·26 0·03 0·46 Temp-surf 0·87* 0·67 0·13 0·76* 0·25 0·34 0·08 0·68 Light 0·34 0·14 0·17 0·10 0·32 0·48 0·41 0·60 0·22 *P<0·05, **P<0·01.
T 6. Spatial variation of water column integrated primary production (gC m2 day1), integrated new production (gC m2day1) and f-ratio obtained in the continental shelf off the Changjiang during cruise 449
Station
E1 E2 E3 E4 E5 E6 E7
Primary production 0·75 0·43 0·52 0·50 0·25 0·66 0·71
New Production 0·16 0·07 0·25 0·21 0·09 0·34 0·16
than N-limited, as has been reported seasonally for other river-influenced regions such as Chesapeake Bay (e.g.Fisher et al., 1992).
Control of the Changjiang NP:(3) Light
As well as ambient nitrate concentration light intensity is another important factor (Kanda et al., 1989) in affecting nitrate uptake. The waters of the Changjiang carry a high concentration of suspended matter (Tian
et al., 1993). These river-borne particulates are main-tained in suspension in the surface layer to salinities >20, and their concentrations vary between 20 and
1000 mg l1 and decrease to 10 mg l1 with higher salinity (Edmond et al., 1985). Light availability increases as the suspended particulate matter settles
(Ning et al., 1988). The nitrate uptake rate, integrated
over the water column, measured in the Changjiang continental shelf water, however, did not show any significant relationship with the light intensity in the present study. The water which was most likely to be affected by the riverine input (Station E2) showed a rather deep euphotic depth (29 m, Table 1) and high incubation light intensity (1383ìE m2S1) when compared with the same parameters at station E4 (27 m and 332ìE m2S1), with double the amount of new productivity. Density (óè) profiles showed that Station E2 was highly stratified and its mixing depth was shallow (Figure 6) compared to the euphotic zone depth. This suggested that the slow uptake of nitrate in the upper water column of the station was not due to limitation by light intensity. The observation that incubation light intensity was not significantly correlated with specific nitrate uptake rate at that station (Figure 7) provides further direct evidence. Low INP (161·55), low f-ratio (0·23) as well as high surface nitrate concentration (2·3ìM,Figure 8) was also observed at Station E7. Limitation due to light intensity seems to be unlikely due to the relatively high light intensity (1243ìE m2S1) and low light extinction coefficient (0·077) during the incubation. However, the low uptake rates of nitrate at Station E7 could be attributed to lack of vertical stratification in the water column. The mixed layer of Station E7 was 60 m (Figure 6), and deep mixing reduces average light intensity available for phytoplankton, which then become light-limited instead of nutrient-limited. Stratification is considered important to stimulate the onset of spring phytoplankton growth in many coastal waters such as Chesapeake Bay and Delaware Bay
(Pennock & Sharp, 1990;Glibert et al., 1995).
Phos-phate limitation was also unlikely because the surface phosphate concentration of 0·21ìM was relatively moderate. The influence of river discharge on Station E7 was minimal because its surface salinity was 34·45.
Contribution of other new N to the ECS new production
The new production and f-ratio in our study would be even higher if N fixation from Trichodesmium sp. and
Richelia intracellularis, a symbiont within diatoms Rhizosolenia spp. and Hemiaulus spp., were also
included in the measurement of new production.
Trichodesmium sp. is abundant in the Kuroshio
Cur-rent and the water where the Kuroshio CurCur-rent encounters the coastal water (Marumo & Asaoka,
6.4 0.9 NO3-surf f-ratio 0.3 0.7 0.5 0.1 0.4 0.8 1.4 2.0 0 0.2 0.6 1.01.2 1.61.8 2.22.4 0.4 0.8 0.6 0.2 E7 E2 45 INP/IChl (mgC .mgChl –1 .d –1 ) 15 35 25 5 20 40 30 10 E7 E2 1800 INP (mgC .m –2 .d –1 ) 600 1400 1000 200 800 1600 1200 400 E7 E2 0 0 (a) (c) (b)
F 8. Relationships between surface nitrate concen-tration and integrated new production (a), chlorophyll a normalized integrated new production (b), and f-ratio (c), measured in the upwelling region ( ) and in the continental shelf off the Changjiang ( ). Regression line was calculated based on the data in the upwelling region during three cruises.
1974). Trichodesmium is capable of very high rates of N fixation (14·0 ngNìg algal N1h1; Saino &
Hattori, 1978). Moreover its density could reach
102–103 filaments per liter in summer (Marumo &
Nagasawa, 1976). However, the effects of
Trichode-smium spp. and Richelia containing diatoms were not
studied intensively enough in our research to enable the estimation of this portion of new production. The summertime N fixation by T. thiebautii was estimated to be 81010gN in the south-eastern East China Sea and accounted for approximately 6% of the N require-ments for primary productivity (Saino & Hattori, 1980). In addition, a relatively high concentration of ammonium (14·6ìM) was also reported in the Changjiang water (Zhang, 1996). This corresponds to about half of the average concentration of nitrate in the run-off (Zhang, 1996). Urea and other DON
should also be accounted for, but the available data are scarce. The contribution to primary production from these N sources was a part of the new produc-tion, at least in those areas influenced by the river runoff, such as Stations E1 and E2. Therefore the real f-ratio would be higher than that obtained in our measurements, which were based solely on nitrate uptake. By simple calculation from the quantity of nitrate flux, Chen (1996) estimated the contribution from the Kuroshio upwelling to be many times greater than the inputs from the Changjiang and the Yellow River combined. With our findings that nitrate con-tributed more to new production in the upwelling waters than in the riverine waters, the importance of the input from the Kuroshio compared with the Changjiang river is further emphasized. However, the contribution of ammonium and DON in river water
T 7. Nitrate and phosphate concentrations as well as nitrate/phosphate (N/P) ratios measured at various sampling stations in the upwelling region (Cruises 416, 124 and 431) and the continental shelf region off the Changjiang estuary (Cruise 449) Station Depth (m) NO3 (ìM) (ìM)PO4 N/P Station Depth (m) NO3 (ìM) (ìM)PO4 N/P Cruise 416 Cruise 449 5 0 1·09 0·58 1·9 E1 2 0·40 0·05 8·0 50 0·51 0·54 0·9 20 4·10 0·21 19·5 75 1·00 0·52 1·9 40 5·34 0·24 22·3 11 0 0·49 0·43 1·1 50 6·29 0·31 20·3 50 2·90 0·72 4·0 E2 2 6·40 0·03 213·3 100 6·20 0·57 10·9 10 6·41 0·06 106·8 Cruise 124 15 7·50 0·20 37·5 8 0 0·00 0·48 0·0 25 6·34 0·25 25·4 50 1·68 0·67 2·5 40 6·94 0·27 25·7 100 10·23 1·16 8·8 E3 2 0·20 0·24 0·8 8a 0 0·11 0·50 0·2 10 1·17 0·10 11·7 50 2·73 0·70 3·9 19 1·61 0·10 16·1 100 9·81 1·11 8·8 40 1·77 0·17 10·4 9 0 0·15 0·54 0·3 E4 2 0·20 <0·03 >6·7 50 4·66 0·74 6·3 10 0·16 <0·03 >5·3 100 6·92 0·95 7·3 20 0·50 <0·03 >16·7 10 0 0·20 0·59 0·3 24 1·63 0·05 32·6 50 6·60 0·80 8·3 60 3·82 0·24 15·9 100 9·63 1·03 9·4 E5 2 0·20 0·11 1·8 10a 0 0·05 0·56 0·1 9 0·55 0·14 3·9 50 5·48 0·89 6·2 20 1·39 0·19 7·3 100 10·56 1·20 8·8 40 3·19 0·21 15·2 Cruise 431 75 8·97 0·67 13·4 9 0 0·00 0·08 0·0 E6 2 0·20 0·07 2·9 40 2·89 0·34 8·5 20 0·12 0·10 1·2 68 8·52 0·45 18·9 40 2·04 0·22 9·3 10 0 0·16 0·05 3·2 60 9·23 0·62 14·9 45 4·78 0·43 11·1 E7 2 2·30 0·21 11·0 100 8·71 0·72 12·1 10 1·83 0·23 8·0 11 0 1·74 0·14 12·4 30 1·97 0·23 8·6 50 4·48 0·46 9·7 50 2·03 0·22 9·2 84 10·40 0·89 11·7 69 4·32 0·28 15·4
may be significant and hence the river contribution could be larger. Additional studies on a larger scale dealing with seasonal variations and the spatial extent of both the river plume and upwelling are warranted to understand the relative importance of the two N sources in the dynamics of new production in the East China Sea.
Acknowledgements
The authors thank Miss W. H. Lee for assistance in chlorophyll a and new production measurements. The cooperation of the captains and crew of the R/V Ocean Researcher I and II during the sampling is appreciated. The authors are grateful to Dr Timothy Parsons for reading a draft version of this manuscript. The authors would also like to thank two anonymous reviewers for helpful suggestions. This research was supported in part by grants (NSC 85-2611-M110-009K2 and 86-2611-M110-008K2) from the National Science Council of the Republic of China.
References
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, 5, 523–527.
Chen, Y. L. L. 1992 Summer phytoplankton community structure in the Kuroshio current-related upwelling northeast of Taiwan. Terrestrial, Atmospheric and Oceanic Sciences 3, 305–320. Chen, Y. L. L. 1994 The importance of temperature and nitrate to
the distribution of phytoplankton in the Kuroshio-induced upwelling, northeast of Taiwan. Proceedings of the National Science Council-Part B: Life Sciences 18, 44–51.
Chen, Y. L. L. 1995a Temporal and spatial changes of chlorophyll a in the KEEP study waters off northern Taiwan. Terrestrial, Atmospheric and Oceanic Sciences 6, 607–620.
Chen, Y. L. L. 1995b Phytoplankton composition and productivity in response to the upwelling off northeastern Taiwan. Proceedings of the National Science Council-Part B: Life Sciences 19, 66–72. Chern, C. S., Wang, J. & Wang, D. P. 1990 The exchange of
Kuroshio and East China Sea shelf water. Journal of Geophysical Research 95, 16 017–16 023.
Draper, N. R. & Smith, H. 1981 Applied Regression Analysis. John Wiley & Sons, New York, 709 pp.
Dugdale, R. C. & Goering, J. J. 1967 Uptake of new and regener-ated forms of nitrogen in primary productivity. Limnology and Oceanography 12, 196–206.
Dugdale, R. C. & Wilkerson, F. P. 1986 The use of 15N to measure nitrogen uptake in eutrophic oceans: experimental consider-ations. Limnology and Oceanography 31, 673–689.
Dugdale, R. C., Morel, A., Bricaud, A. & Wilkerson, F. P. 1989 Modeling new production in upwelling centers: a case study of modeling new production from remotely sensed temperature and color. Journal of Geophysical Research 94, 18119–18132. Edmond, J. M., Spivack, A., Grant, B. C., Hu, M., Chen, Z., Chen,
S. & Zeng, X. 1985 Chemical dynamics of the Changjiang estuary. Continental Shelf Research 4, 17–36.
Fisher, T. R., Peele, E. R., Ammerman, J. W. & Harding, L. W. 1992 Nutrient limitation of phytoplankton in Chesapeake Bay. Marine Ecology Progress Series 82, 51–63.
Glibert, P. M., Conley, D. J., Fisher, T. R., Harding Jr. L. W. & Malone, T. C. 1995 Dynamics of the 1990 winter/spring bloom in Chesapeake Bay. Marine Ecology Progress Series 122, 27–43. Hama, T., Miyazaki, T., Ogawa, Y., Iwakumi, T., Takahashi, M.,
Otsuki, A. & Ichimura, S. 1983 Measurement of photosynthetic production of a marine phytoplankton population using a stable
13C isotope. Marine Biology 73, 31–36.
Hama, T., Shin, K. H. & Handa, N. 1997 Spatial variability in the primary productivity in the East China Sea and its adjacent waters. Journal of Oceanography 53, 41–51.
Harrison, W. G., Platt, T. & Lewis, M. R. 1987 f-Ratio and its relationship to ambient nitrate concentration in coastal waters. Journal of Plankton Research 9, 235–248.
Harrison, P. J., Hu, M. H., Yang, Y. P. & Lu, X. 1990 Phosphate limitation in estuarine and coastal waters of China. Journal of Experimental Marine Biology and Ecology 140, 79–87.
Kamykowski, D. & Zentara, S. J. 1986 Predicting plant nutrient concentration from temperature and sigma-t in the upper kilo-meter of the world ocean. Deep-Sea Research 33, 89–105. Kanda, J., Ziemann, D. A., Conquest, L. D. & Bienfang, P. K.
1989 Light-dependency of nitrate uptake by phytoplankton over the spring bloom in Auke Bay, Alaska. Marine Biology 103, 563–569.
Le Bouteiller, A. 1986 Environmental control of nitrate and am-monium by phytoplankton in the Equatorial Atlantic Ocean. Marine Ecology Progress Series 30, 167–179.
Li, Y. H. 1995 Material exchange between the East China Sea and the Kuroshio Current. Terrestrial, Atmospheric and Oceanic Sciences 5, 625–631.
Marumo, R. & Asaoka, O. 1974 Distribution of pelagic blue-green algae in the North Pacific Ocean. Journal of the Oceanographical Society of Japan 30, 77–85.
Marumo, R. & Nagasawa, S. 1976 Seasonal variation of the standing crop of a pelagic blue-green alga, Trichodesmium in the Kuroshio water. Bulletin of Plankton Society of Japan 23, 19–25. Ning, X., Vaulot, D., Liu, Z. & Liu, Z. 1988 Standing stock and
production of phytoplankton in the estuary of the Changjiang (Yangtse River) and the adjacent East China Sea. Marine Ecology Progress Series 49, 141–150.
Parsons, T. R., Maita, Y. & Lalli, C. M. 1984 A Manual of Chemical and Biological Methods for Seawater Analysis. Pergamon Press, pp. 173.
Pennock, J. R. & Sharp, J. H. 1990 Biogeochemical regulation of nitrogen cycling in the Delaware Estuary—The constraints of light-limited phytoplankton growth. Abstracts of Papers of the American Chemical Society 199, 37.
Saino, T. & Hattori, A. 1978 Diel variation in nitrogen fixation by a marine blue-green alga Trichodesmium thiebautii. Deep-Sea Research 25, 1259–1263.
Saino, T. & Hattori, A. 1980 Nitrogen fixation by Trichodesmium and its significance in nitrogen cycling in the Kuroshio area and adjacent waters. In The Kuroshio IV. Proceedings of the Fourth Symposium for the Cooperative Study of the Kuroshio and Adjacent Regions, The Japan Academy, Tokyo, 14–17 February, 1979, pp. 697–709.
Slawyk, G. 1979 13C and 15N uptake by phytoplankton in the
Antarctic upwelling area: Results from the Antipod I Cruise in the Indian Ocean sector. Australian Journal of Marine and Freshwater Research 30, 431–448.
Strickland, J. D. H. & Parsons, T. R. 1972 A Practical Handbook of Seawater Analysis. Bulletin of Fisheries Research Board of Canada
167, 1–310.
Tian, R. C., Hu, F. X. & Martin, J. M. 1993 Summer nutrient fronts in the Changjiang (Yangtze River) estuary. Estuarine, Coastal and Shelf Science 37, 27–41.
Vezina, A. F. & Platt, T. 1987 Small-scale variability of new production and particulate fluxes in the ocean. Canadian Journal of Fisheries and Aquatic Science 44, 198–205.
Vitousek, P. M. & Howarth, R. W. 1991 Nitrogen limitation on land and in the sea—How can it occur. Biogeochemistry 13, 87–115.
Wilkerson, F. P. & Dugdale, R. C. 1987 Effects of El Nino on new, regenerated, and total production in eastern boundary upwelling systems. Journal of Geophysical Research 92, 14 347–14 353. Wong, G. T. F., Pai, S. C., Liu, K. K., Liu, C. T. & Chen, C. T.
A. 1991 Variability of the chemical hydrography at the frontal
region between the East China Sea and the Kuroshio north-east of Taiwan. Estuarine, Coastal and Shelf Science 33, 105– 120.
Zhang, J. 1996 Nutrient elements in large Chinese estuaries. Continental Shelf Research 16, 1023–1045.
