Different contributions of riverine and oceanic nutrient fluxes supporting primary production in Ishikari Bay

Continental Shelf Research 88 (2014) 140–150

Contents lists available at ScienceDirect

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

Research papers

Different contributions of riverine and oceanic nutrient fluxes supporting primary production in Ishikari Bay Julius I. Agboola a,b,n, Isao Kudo a,c a

Graduate School of Environmental Science, Hokkaido University, Kita 10 Nishi 5, Sapporo 060-0810, Japan Department of Fisheries and Centre for Environment and Science Education (CESE), Lagos State University, Ojo, Lagos, Nigeria c Graduate School of Fisheries Sciences, Hokkaido University, Hakodate 041-8611, Japan b

art ic l e i nf o

a b s t r a c t

Article history: Received 26 December 2013 Received in revised form 18 July 2014 Accepted 25 July 2014 Available online 5 August 2014

We computed a ratio of riverine nutrient flux (RNF) to bottom nutrient flux (BNF) to determine the relative importance of oceanic and riverine nutrient fluxes on primary production dynamics in Ishikari Bay, which is composed of oligotrophic subarctic coastal water. Across spring, summer and autumn, the RNF:BNF ratio (R:B ratio) was significantly greater than 1.0, especially in spring and autumn for DIN and Si(OH)4, suggesting that riverine nutrients mostly supported primary production. A strong inverse relationship (r ¼  0.927) between Chl a and salinity in autumn and a corresponding increase in the apparent utilization of DIN and primary production indicated that the contribution of DIN from the Ishikari River on primary production was maximal in autumn. However the R:B ratio for PO4 was significantly less than 1.0, especially in summer (0.1) and autumn (0.3), suggesting a larger contribution of bottom upwelling nutrient sources. In spring, when the ratio was close to 1 (0.8), PO4 supply from both bottom (upwelling) and surface (river) was equivalent, since PO4 concentration of river endmember was the lowest. Although riverine nutrient fluxes were a major source of DIN and Si(OH)4 nutrient supply in the bay, oceanic nutrient contribution from bottom upwelling and horizontal advection was a major source of PO4. While riverine nutrients significantly fuel primary production, the estuarine circulation process may contribute significantly to compensating for the inadequate supply of riverine PO4 in an oligotrophic system like Ishikari Bay. Also, unlike the usual estuarine system in which nutrient concentration at a deeper layer is high due to the regeneration of nutrients at depth, concentration in Ishikari Bay was very low due to an influence of oligotrophic waters. We conclude that riverine nutrient flux contributes a large portion of the total flux in Ishikari Bay. & 2014 Elsevier Ltd. All rights reserved.

Keywords: Nutrients Oligotrophic Primary production Coastal Estuarine circulation.

1. Introduction Primary production is dependent on the presence of the appropriate combination of nutrients to support and sustain their activity (Brzezinski, 1985; Barber et al., 2001; Kristiansen and Hoell., 2002; Kudo et al., 2005). Rivers are a critical link in the global cycling of elements, and dissolved inorganic nutrients from rivers to ocean system are obviously of great importance in a discussion of the interactions of these compounds in estuarine and adjacent coastal zones. Also, massive transport of nutrients from the deeper nutrient-rich layers to the impoverished euphotic zone occurs when intense vertical mixing or upwelling processes are observed. Such short and cataclysmic events have a pronounced influence on the biogeochemical fluxes in the ocean because they n Corresponding author at: Department of Fisheries and Centre for Environment and Science Education (CESE), Lagos State University, Ojo, Lagos, Nigeria. Tel.: þ 234 8052242886. E-mail address: [email protected] (J.I. Agboola).

http://dx.doi.org/10.1016/j.csr.2014.07.013 0278-4343/& 2014 Elsevier Ltd. All rights reserved.

accelerate the completion of nutrient transport to upper layers, which would take much longer under “ordinary” conditions. However, it is difficult to evaluate these fluxes due to significant seasonal fluctuations and human activities that perturb the natural fluxes. One of the objectives in this study is to determine the possible sources of nutrients, and ultimately the fate of nutrients that are drawn from the Ishikari River into the adjacent shelf of Ishikari Bay, the northwestern North Pacific, Japan (Fig. 1). Unlike the Pacific coastal region, which is influenced by the subarctic ocean current (Oyashio) with its high nutrients, Ishikari Bay receives few nutrient fluxes from oligotrophic subtropical (Tsushima) warm currents (Yoshida et al., 1977) and receives a nutrient flux from the Ishikari River, the second largest river in the catchment area in Japan. Thus, the bay is characterized as an oligotrophic subarctic coastal water with a considerable influence of riverine discharge from the Ishikari River (Agboola et al., 2010). Earlier studies on the seasonal change in riverine nutrients and distribution of chlorophyll a at 26 gridded sampling stations in Ishikari Bay have been reported (Agboola et al., 2009, 2010). Also,

J.I. Agboola, I. Kudo / Continental Shelf Research 88 (2014) 140–150

2006

141

2007

I-14

I-14

I-36

I-36

Fig. 1. Sampling stations in Ishikari Bay indicating boarder of Plume for 2006 and 2007. Bold line, spring; broken line, summer; and dashed line, autumn.

Agboola et al. (2013) reported on a 16-month time-series study on seasonality and the environmental drivers of biological productivity (primary productivity and chlorophyll biomass) in Ishikari Bay. However, studies on influence of estuarine circulation and oceanic nutrient fluxes as well as coupled riverine nutrient flux on phytoplankton biomass and production dynamics are scarce in this region of the Pacific. The present study is the first attempt to document the dynamics of nutrients and phytoplankton biomass and productivity in Ishikari Bay. We further quantified the relative contribution of bottom nutrient (upwelling) flux through estuarine circulation and horizontal advection from the oceanic region into Ishikari Bay. Lastly, this paper aims to provide the answers to some hypothetical questions. Firstly, can the Ishikari River discharge supply the required nutrients for phytoplankton biomass and production build-up in Ishikari Bay oligotrophic system? Secondly, in considering the relative importance of estuarine circulation, is there any significant bottom nutrient upwelling flux to the euphotic zone which could possibly fuel primary production in the bay? Analysis of seasonal and inter-annual variations in the river Plume and Out-Plume areas (see Agboola et al., 2009) will not only help to establish possible relationships between these variables, but will also test and validate our hypothesis that the Ishikari River discharge contains nutrients to significantly fuel primary production in Ishikari Bay. If it does, and assuming a steady state, then the river Plume area receiving a higher nutrient flux should have a higher production, and this will enable us to predict the major source of nutrients and other possible factors fueling or limiting production in this oligotrophic system.

which is defined as a vertical 34.0 isohaline from the surface to  200 or 400 m in summer and autumn (Dodimead et al., 1963; Dodimead, 1967). The boundary is located around 421N in the North Pacific Ocean and divides the ocean roughly into the subtropical and subarctic regions. 2.1. Sampling Routine samplings for nutrients and chlorophyll a were carried out at all 26 gridded sampling stations using a Sea Bird 911 CTD system equipped with a carrousel multi-sampler of 12 2.5-L Niskin bottles to collect discrete samples for macronutrients, Chl a and other biogeochemical variables in the water column (down to 5 m above the sea floor) at each sampling station. Six stations (representative stations of Plume and Out-Plume areas) were assigned for a detailed observation of nutrient and phytoplankton biomass and productivity. 2.2. Nutrients Sub-samples for nutrients (NO3 , NH4þ , PO4 and Si(OH)4) were collected in duplicate in 10 ml spit tubes and were stored frozen at  30 1C until laboratory analysis (Parsons et al., 1984). Concentrations of the dissolved inorganic nutrients were determined using a continuous flow analyzer (QuAAtro, BRAN þLUEBBE). Detection limits were estimated at around 0.01 mM based on three times the standard deviation of the lowest concentration of samples. 2.3. Quantification of nutrient fluxes

2. Materials and methods Eight cruises were carried out in Ishikari Bay on board the TS Oshoro-Maru and the TS Ushio-Maru of Hokkaido University in spring (April, May), summer (July, August) and autumn (October, November) of 2006 and 2007. The study area, approximately 4370 km2, lies between 431100 N and 441000 N and stretches between 1401000 E and 1411220 E (Fig. 1). Ishikari Bay is located in the northwest coast of Hokkaido within the subarctic regions,

In order to determine the nutrient fluxes to the bay we considered the influence of Ishikari River nutrient discharge and bottom nutrient upwelling through estuarine circulation using vertical and spatial profile data of salinity and nutrients. These distinct perspectives were deployed and related to phytoplankton biomass and productivity across space and time to determine the source and fluxes of the nutrients fueling primary production in this oligotrophic system.

142

J.I. Agboola, I. Kudo / Continental Shelf Research 88 (2014) 140–150

2.3.1. Estuarine circulation models The conceptual model framework on which this study is based is presented in Fig. 2. We quantify the relative contribution of nutrient from bottom upwelling into Ishikari Bay using Knudsen's hydrographical theorem (Unoki, 1998). The relationship between water and salt balance in Ishikari Bay is presented in Fig. 3. The entire study area in the bay is partitioned into four boxes (Boxes 1–4) and the direction of water fluxes is indicated by arrows. The boundary between S1 and S2 is a halocline, which is at  10 m depth, where R is Ishikari River flux, S1–S4 are average salinity in Boxes 1–4, Q is water flux from Box 4 to Box 2, Q12 is water flux from Box 1 to Box 2, Q0 is water flux from Box 1 to Box 3, and Q21 is water flux from Box 2 to Box 1. Box 1 Water balance equation : Salt balance equation :

R þ Q 21 ¼ Q 0 þ Q 12 :

ð1Þ

R  0 þ S2 Q 21 ¼ S1 ðQ 12 þ Q 0 Þ:

ð2Þ

Box 2 Water balance equation : Salt balance equation :

Q þ Q 12 ¼ Q 21 :

ð3Þ

S1 Q 12 þ S4 Q ¼ S2 Q 21 :

ð4Þ

Box 3 From 1 and 3 Water balance equation :

Q 0 ¼ Q þ R:

ð5Þ

From 2 and 3 Salt balance equation :

S1 Q 0 ¼ S4 Q :

ð6Þ

Box 4 From (5 & 6) Water balance equation : λ¼

Q¼

RS1 : ðS4 þS1 Þ

Q S1 ¼ : R ðS4  S1 Þ

Q0 ¼

RS4 : ðS4  S1 Þ

ð7Þ ð8Þ ð9Þ

RS1 : ðS2  S1 Þ

ð10Þ

Q 12 ¼ Q 21  Q :

ð11Þ

Q 21 ¼

τf ¼

Qf : R

ð12Þ

where τf is residence time of fresh water, Q f is freshwater volume in the box, Q f ¼ V 1 ððS4  S1 Þ=S4 Þ, V1 is volume of Box 1, and S1 is average salinity of Box 1. The bottom upwelling of nutrients to the surface in Ishikari Bay was calculated using Eq. (7). Bottom water flux, Q, was computed as RS1 =ðS4 þ S1 Þ. 2.3.2. Horizontal advection To further capture nutrients fluxes from the oceanic region into the bay, contributions from the horizontal advection were quantified. Using the revised box model (Fig. 4), we present the water flux flow pattern and derived formulae. This was quantified to

Fig. 2. Conceptual model of the Ishikari Bay ecosystem showing nutrient/water flux from river discharge and estuarine circulation.

Fig. 3. Estuarine circulation models in the Ishikari Bay.

J.I. Agboola, I. Kudo / Continental Shelf Research 88 (2014) 140–150

143

Also, from Eq. (20) Q 21 –Q ¼ Q 12 : Q 12 ¼

Q ðS1  S4 Þ Q: S1  S2

ð29Þ ð30Þ

Horizontal nutrient flux (Fin) Q in C in :

ð31Þ

Bottom nutrient flux (F21) Q 21 C 21 :

ð32Þ

where Cin and C21 are nutrient concentrations, respectively, from Qin and Q21. Fig. 4. Revised box model including a horizontal circulation in the upper layer.

compare the relative contribution of surface (horizontal advection) to bottom upwelling and to have a view of the total oceanic flux contribution. As shown in Fig. 4, the following equations are derived to quantify water and salt balance from the partitions (boxes): R þ Q 21 þ Q in ¼ Q out þ Q 12 ðwater balanceÞ;

ð13Þ

S2 Q 21 þ S3 Q in ¼ S1 ðQ out þ Q 12 Þ ðsalt balanceÞ;

ð14Þ

Q þ Q 12 ¼ Q 21 ðwater balanceÞ;

ð15Þ

S1 Q 12 þ S4 Q ¼ S2 Q 21 ðsalt balanceÞ;

ð16Þ

m ¼ Q in =Q out ;

ð17Þ

where m (deduced from flow velocity/rate) is the ratio of water flux inflow from surface (Qin) to water flux outflow (Qout) λ ¼ Q =R ¼ Y=ðS4  YÞ;

ð18Þ

where Y ¼ ðS1  mS3 Þ=ð1  mÞ:

ð19Þ

Also Q þ Q 12 ¼ Q 21 :

ð20Þ

Q 21 –Q 12 ¼ Q :

ð21Þ

Q þ Q in þ R ¼ Q out :

ð22Þ

Q out  Q in ¼ R þ Q 21  Q 12 :

ð23Þ

From Eq: ð6Þ; Q ¼ λR:

ð24Þ

From Eq: ð4Þ; Q in ¼ mQ out :

ð25Þ

Applying Eq. (22) λR þ mQ out þR ¼ Q out ðλ þ 1ÞR ¼ ð1  mÞQ out ðλ þ1Þ R: ð1  μÞ

ð26Þ

ðλ þ 1Þ μR: ð1  μÞ

ð27Þ

Q out ¼ Q in ¼

Multiplying Eq. (20), QþQ12 ¼Q21, by S1 S1 Q þS1 Q 12 ¼ S1 Q 21 S1 Q 12 þ S4 Q ¼ S2 Q 21 ðEq: ð3ÞÞ S1 Q –S4 Q ¼ Q 21 ðS1 –S2 Þ Q 21 ¼

Q ðS1  S4 Þ : S1  S2

ð28Þ

2.4. Phytoplankton biomass Phytoplankton biomass (Chl a) was measured. An aliquot (150 ml) of water was filtered through Whatman GF/F (25 mm diameter, pore size 0.7 mm: total Chl a) filters using parallel filtration under low vacuum pressure (o250 kPa) or gravity. After filtration, Chl a was immediately extracted by immersing the filter in N,N-dimethylformamide (Suzuki and Ishimaru, 1990) and preserved at  30 1C until on-shore analysis by fluorometry. Chl a concentrations were determined using a HITACHI F2000 fluorescence spectrophotometer according to the method of Parsons et al. (1984). 2.5. Statistical analyses Physical water properties, nutrients Chl a and primary production were compared for the spring, summer and autumn of 2006 and 2007 using a two-way analysis of variance (ANOVA), whereas Duncan multiple range test was used for separation of means. Intercorrelation of variables was investigated using the Pearson Product Moment Correlations coefficient. Vertical profiles of nutrient concentrations in the water column were statistically compared for the different seasons (spring, summer and autumn) and areas (Plume and Out-Plume) using ANOVA. This was achieved by grouping the water column as surface waters (0–10 m) based on the average euphotic depth of Plume and, since halocline was 10 m, subsurface waters (10–40 m) based on the average euphotic depth of Out-Plume and water column depths 440 m as bottom waters (40–80 m). As NO2 concentration is negligible in the samples, the total concentration of dissolved inorganic nitrogen (DIN) is the sum of NO3 and NH4. 3. Results The spatiotemporal distributions of nutrients and phytoplankton biomass in relation to the physical oceanographic conditions in spring, summer and autumn, except for winter, as well as criteria for the classification of the Plume and Out-Plumes area in Ishikari Bay, have been defined and reported (Agboola et al., 2009). The surface nutrient concentration, relative ratio of nutrients, water discharge and suspended particulate matter (SPM) in the Ishikari River station for a core of river Plume in 2006 are presented in Table 1. Generally, at the river/ocean interface where mixing between freshwater and seawater occurs, estuarine systems are characterized by drastic changes in physical and chemical conditions, which are primarily related to the salinity gradient. Salinity gradient of surface water (14.7–33.9 and 24.2–33.1 in 2006 and 2007 respectively) was the highest in spring relative to summer and autumn, evidencing the influence of river discharge and a thaw. In the summer and autumn of 2006, average salinity increased to 33.487 0.40 and 33.37 70.76 respectively due to a

144

J.I. Agboola, I. Kudo / Continental Shelf Research 88 (2014) 140–150

low river discharge. Plume distributions in Ishikari Bay were significantly different (P o0.001) by season as a result of change in riverine nutrient flux. In 2006, the influence of the spring thaw was evident from the large Plume area (2562 km2) in spring compared to summer (121.2 km2) and autumn (848.9 km2). In contrast the spring thaw had little or no influence on the Plume area in spring (78.52 km2), compared to summer (554.81 km2) and autumn (637.13 km2) in 2007.

3.1. Spatial, vertical and seasonal distributions of nutrients Nutrients concentrations were generally higher in Ishikari Bay in 2007 compared to 2006 (Table 2). High concentrations of nutrients were found only in the Plume area in spring, corresponding to the highest seasonal river discharge in spring (Tables 1 and 2). Ishikari Bay showed a strong spatial and seasonal variation in dissolved inorganic nitrogen concentration (DIN, with 79% and  80% of NO3 in 2006 and 2007 respectively). The highest average DIN concentration was observed in the Plume (2006: 1.15 mM and 2007: 7.08 mM) whereas the lowest average concentrations occurred in the Out-Plume (2006: 0.07 mM and 2007: 0.54 mM). Seasonally, the highest average DIN concentration in the Plume was observed in spring of 2006 and 2007 and the lowest in summer (0.52 mM and 0.08 mM respectively), whereas, in the Out-Plume, average DIN concentration was the highest (0.72 mM) in autumn and the lowest (0.07 mM) in the summer of 2006. In contrast, the average DIN concentration in the Out-Plume was the highest (3.86 mM) in spring and the lowest (0.54 mM) in the summer of 2007. NO3 concentration depended on the degree of the influence of the Ishikari River water, with the highest value recorded in the Table 1 Surface nutrient concentration, relative ratio of nutrients, water discharge and suspended particulate matter (SPM) in Ishikari River station for a core of river Plume in 2006. Parameters

Spring

Summer

Autumn

DIN (mM) PO4 (mM) Si(OH)4 (mM) Si:P DIN:P Si:DIN Discharge (m3 s  1) SPM (mg l  1)

34.1 0.57 79.8 140.0 60.0 2.3 2250 450

53.5 0.76 262.0 342.4 70.4 4.9 400 10.0

93.6 0.75 160.0 246.9 24.2 1.7 750 40.0

Plume areas (Table 2); however, in autumn of 2006, NO3 concentrations were less than 0.5 mM, which represents no significant difference between the Plume and Out-Plume areas. NO3 exhibited a strong inverse relationship with salinity in spring (Pearson correlation coefficient, r ¼  0.831; P ¼0.001, n ¼26) and autumn (r ¼  0.727; P¼ 0.001, n ¼19) of 2006. Similarly, NO3 exhibited a strong inverse relationship with salinity in spring (Pearson correlation coefficient, r ¼  0.738; P ¼0.005, n ¼14) and summer (r ¼  0.839; P ¼0.001, n ¼21) of 2007. This indicated conservative mixing between NO3-rich freshwater and NO3-drought coastal waters and the absence of significant sources or sink of NO3 near the river mouth area. NH4 exhibited an inverse relationship with salinity only in spring of 2006 (Pearson correlation coefficient, r ¼  0.765; P ¼0.001, n ¼26) and 2007 (r ¼  0.630; P¼ 0.05, n¼ 14). The spatial and seasonal changes in PO4 concentration in surface waters of Ishikari Bay were not very evident in the OutPlume stations, especially in summer and autumn (figure not shown). Summer average concentration in the Plume and OutPlume was respectively 0.07 mM in 2006 and slightly greater than 0.10 mM in 2007 (Table 2). There was no significant relationship between salinity across seasons in 2007. However, a strong significant inverse relationship was found between PO4 and salinity in spring (r ¼  0.802; P ¼0.001, n ¼26) and autumn (r ¼  0.565; P¼ 0.001, n ¼19) of 2006. Si(OH)4 concentration in Ishikari Bay exhibited clear spatial and seasonal variation with the highest average concentration in the Plume (6.9 mM and 13. 03 mM, respectively, in 2006 and 2007) and the lowest at seaward boundary (Out-Plume, average of 0.6 mM and 3.81 mM, respectively, in 2006 and 2007) (Table 2). Si(OH)4 concentration in the Plume decreased from spring to summer, followed by an increase in autumn. Si(OH)4 also exhibited a strong inverse relationship with salinity in spring (r ¼  0.831; P ¼0.001, n¼ 26) and autumn (r ¼  0.823; P¼ 0.001, n ¼ 19), suggesting conservative mixing between Si(OH)4-rich river/estuarine waters and Si(OH)4-low coastal water in 2006 (figure not shown). A similar relationship was exhibited in 2007 in the spring (r ¼  0.729; P¼ 0.005, n ¼14), summer (r ¼  0.641; P ¼0.005, n¼ 21) and autumn (r ¼  0.815; P ¼0.001, n ¼12). In the Plume, Si(OH)4 concentrations exceeded those in the Out-Plume in all seasons. Vertical distributions of NO3 and Si(OH)4 in the Plume (Stn. I-36) exhibited higher concentrations at surface, especially in spring, than at deeper layers (Fig. 5). Such high concentrations were not observed for PO4; it exhibited higher concentration at deeper layers than the other nutrients. In the Out-Plume (Stn.

Table 2 Mean 7 SD of nutrients concentration (mmol m  3) from combined data over station and depth (50 m) in spring, summer and autumn seasons of 2006 and 2007. Season 2006 Spring Summer Autumn 2007 Spring Summer Autumn

Area

n

NH4þ

PO34 

Plume Out-Plume Plume Out-Plume Plume Out-Plume

19 7 4 17 6 12

0.177 0.16a 0.02 7 0.05a 0.02 7 0.02a 0.047 0.08a 0.077 0.08a 0.02 7 0.05a

0.12 7 0.06ab 0.077 0.03a 0.077 0.0a 0.077 0.03a 0.17 7 0.09b 0.11 7 0.01b

4.98 7 5.15a 0.59 7 0.37a 3.25 7 0.49a 2.88 7 0.47b 6.92 7 3.83a 3.52 7 2.37c

0.98 7 0.83a 0.11 7 0.16a 0.06 7 0.06a 0.03 7 0.06a 0.44 7 0.48a 0.707 1.42b

Plume Out-Plume Plume Out-Plume Plume Out-Plume

3 12 4 19 8 4

1.317 0.53b 0.59 7 0.22b 0.127 0.10a 0.157 0.07a 0.337 0.20a 0.03 7 0.04a

0.28 7 0.05b 0.29 7 0.09b 0.16 7 0.03a 0.11 7 0.03a 0.19 7 0.03a 0.12 7 0.01a

13.03 7 8.04b 5.56 7 1.98b 4.39 7 0.76a 3.92 7 0.49a 8.82 7 1.59ab 3.81 7 0.18a

5.777 3.33b 2.15 7 0.80c 0.40 7 0.05a 0.39 7 0.06a 1.02 7 0.31a 1.12 7 1.22b

Si(OH)4

NO3

n refers to the number of data for calculating mean 7 SD for the nutrients concentration. Values of respective area along the same row bearing the same superscripts are not statistically different at 5% probability level using the Duncan multiple range test.

J.I. Agboola, I. Kudo / Continental Shelf Research 88 (2014) 140–150

145

Fig. 5. Nutrient vertical structure of typical Plume (St. 36) (a)–(d) and Out-Plume (St.14) (e)–(h) during spring, summer and autumn seasons in Ishikari Bay in 2006.

I-14), concentrations of all nutrients increased with depth (Fig. 5). Surface depletion of NO3 and Si(OH)4 was observed throughout the seasons, while PO4 remained at ca. 0.1 mM. In contrast, all nutrients exhibited nearly homogeneous vertical distribution in the Out-Plume (Stn. I-14) in 2007 (Fig. 6). In general, nutrient distributions were grouped for surface waters (0–10 m), sub-surface waters (10–40 m) and bottom waters (4 40 m). In 2006, with the exception of Si(OH)4, all nutrients showed marked variations in concentration in the group water depths across seasons in the Out-Plume. Vertical distribution of NH4 concentration was significantly different with depth (P o0.001) and across seasons (Po 0.001). Although NH4 concentrations in surface (0–10 m) and sub-surface (10–40 m) waters were not significantly different (P4 0.05) during the summer and autumn seasons, bottom water concentration was significantly different (P o0.05) in relation to surface and sub-surface waters. Similar trends of vertical distribution in surface, sub-surface, and bottom waters were observed for PO4 and NO3 concentrations. While depth and seasonal influence was significant (Po0.05) for NH4, no significant influence was exhibited for PO4 and NO3. This may suggest that season modulates the depth concentration of NH4. Nitrogen, particularly NO3, was generally depleted, but presented at only 0.98 mM and 5.77 mM in spring of 2006 and 2007 respectively. During summer and autumn of 2006, mean values for N:P ratio in surface water varied from 0.35 (Out-Plume) to 5.57 (Plume). In 2007, mean values for N:P ratio in surface water varied from 4.41 (Out-Plume) to 12.33 (Plume). However, in spring, N:P ratio in the Plume was 29.6 and 59.2, in 2006 and 2007, respectively, well above the Redfield Ratio (Redfield, 1934; Redfield et al., 1963) of 16 during this period. Considering N:P ratio more than 16, PO4 appears to be potentially limiting and this may partly explain the observed low

Chl a concentration compared to the autumn of 2006. However average concentration of PO4 in the Plume was higher (almost doubled) in 2007 (0.25 mM) compared with 2006 (0.13 mM), suggesting that PO4 may not be potentially limiting to fuel the maximum Chl a biomass in spring of 2007. 3.2. Intercorrelation of phytoplankton biomass with salinity and nutrients Chl a exhibited strong inverse relationship with salinity (r ¼  0.927; P ¼0.001, n ¼19) in autumn of 2006, suggesting riverine nutrient influence on Chl a concentration (Fig.7a–c). No relationship was found during spring and summer. In 2007, while the Chl a exhibited a strong inverse relationship with salinity in summer (r ¼  0.623; P¼ 0.005, n ¼21) and autumn (r ¼  0.853; P¼ 0.001, n ¼12), no relationship was found in spring, which had the maximum Chl a concentration across the seasons (Fig. 7d–f). This may suggest the possible joint influence of other factors or cofactors asides from river nutrient discharge on the observed maximum Chl a concentration in spring. We evaluated the relations between Chl a and nutrients using Pearson Product Moment Correlations coefficient (Table 3). In 2006 Chl a showed significant correlation with NO3 in summer (r ¼0.856; P ¼0.001, n ¼22) and autumn (r ¼0.753; P ¼0.001, n ¼19) whereas, in 2007, Chl a showed significant correlation with NO3 only in summer (r ¼0.525; P¼ 0.01, n ¼22). Similarly, Chl a showed significant correlation with Si(OH)4 in summer (r ¼0.828; P¼ 0.001, n¼ 22) and autumn (r ¼0.835; P ¼0.001, n ¼19) 2006 and also in summer (r ¼0.827; P ¼0.001, n ¼ 22) and autumn (r ¼ 0.811; P¼ 0.001, n ¼12) 2007. While Chl a showed positive correlation with PO4 in summer (r ¼0.430; P ¼0.005, n ¼22) and autumn (r ¼0.746; P ¼0.001, n ¼19) of 2006, and summer (r ¼0.625; P¼ 0.005, n ¼22) 2007, a negative correlation was exhibited with

146

J.I. Agboola, I. Kudo / Continental Shelf Research 88 (2014) 140–150

Fig. 6. Nutrient vertical structure of typical Plume (St. 36) (a)–(d) and Out-Plume (St. 14) (e)–(h) during spring, summer and autumn seasons in Ishikari Bay in 2007.

PO4 in spring (r ¼  0.577; P ¼0.05, n ¼14) 2007. However, in 2006 and 2007, Chl a showed no significant correlation with NH4 across each of the seasons. 3.3. Estuarine circulation in Ishikari Bay Generally, nutrient concentration in surface (river) waters was higher than that in bottom (upwelling) waters (Table 4). Across seasons, RNF:BNF ratio (R:B ratio) for DIN and Si(OH)4 ranged from 1.2 and 1.8 in summer to 4.9 and 57.4 in spring (Table 5), respectively. These ratios revealed the Ishikari River as a major nutrient source, especially for DIN and Si(OH)4, fueling autotrophic production in Ishikari Bay. The R:B ratio for PO4 suggested a relatively higher influence from bottom upwelling, especially in summer (0.1) and autumn (0.3). In spring the ratio was 0.8, close to1, suggesting equal influence by/between surface and bottom nutrient flux. 3.3.1. Horizontal advection in Ishikari Bay. As shown in the revised box model in Fig. 4 and the derived formulae, contribution from horizontal advection was also quantified to compare the relative contribution of surface (horizontal advection) to bottom upwelling and to have a view of the total oceanic flux contribution. Fig. 8a and b shows water flux from surface (Qin) and bottom (Q21) in relation to m ratio values (as derived above). Qin flux increased steadily with m, whereas Q21 flux was fairly constant and decreased only gradually with increasing m. At m ¼0.5, Qin and Q21 water fluxes appear to be similar ( 2000 m  3 s  1). Thus, at lower m (0–0.5), i.e., when water out-flux (Qout) is much stronger (equal

and more than double Qin), Q21 water flux appears to be much higher than Qin, especially at m o0.3. Given the water fluxes (Qin and Q21), the nutrient fluxes (Fin and F21) of DIN, PO4, and Si(OH)4 across seasons are presented in Fig. 9. Similar to the water flux, when m value was lower, F21 values of DIN, PO4 and Si(OH)4 were more than three times higher than for Fin in most instances. Except for DIN in summer and autumn, Fin appears to be larger than F21 when m value is higher than 0.5. 3.3.2. Oceanic nutrient flux versus riverine nutrient flux Given the seasonal nutrient fluxes from bottom upwelling and horizontal advection, we compare the total oceanic nutrient fluxes (bottom upwelling and horizontal advection) and riverine nutrient fluxes in Ishikari Bay. Fig. 9 shows the oceanic and riverine nutrient fluxes in Ishikari Bay. Generally, seasonal nutrient fluxes of DIN and Si(OH)4 contribution in Ishikari Bay were significantly higher from riverine source than oceanic source. Seasonally and especially in summer, oceanic nutrient flux of PO4 and Si(OH)4 dominated in Ishikari Bay.

4. Discussion 4.1. Nutrients Nutrients concentrations were generally higher in Ishikari Bay in 2007 compared with 2006, and showed highly significant interannual variation (P o0.001). According to Harris (2001) and ANZECC/ARMCANZ (2000), nutrient impacts on coastal waterways vary as a function of both loadings (fluxes) and bioavailability of the nutrients as well as the extent to which hydrodynamic

J.I. Agboola, I. Kudo / Continental Shelf Research 88 (2014) 140–150

147

Fig. 7. Phytoplankton biomass (Chl a) distribution along salinity gradient in surface waters in 2006 spring (a), summer (b) and autumn (c) seasons and in 2007 spring (d), summer (e) and autumn (f) seasons in Ishikari Bay.

Table 3 Pearson's correlation matrix of surface water nutrients and Chl a in Ishikari Bay in 2006. Season Spring Summer Autumn n

Chl a Chl a Chl a

NH4

NO3

PO4

Si(OH)4

DIN

 0.238 0.519n  0.169

 0.203 0.856nn 0.753nn

0.028 0.430n 0.746nn

 0.208 0.828nn 0.835nn

 0.206 0.859nn 0.724nn

Correlation is significant at the 0.05 level (2-tailed). Correlation is significant at the 0.01 level (2-tailed).

nn

Table 4 Nutrient concentrations in surface (river) and bottom (upwelling) waters in Ishikari Bay in 2006. Season

Spring Summer Autumn

Nutrients source

Surface Bottom Surface Bottom Surface Bottom

(river) upwelling (river) upwelling (river) upwelling

Concentration (lmol l  1) DIN

PO4

Si(OH)4

34.1 1.12 53.5 0.79 93.6 1.20

0.57 0.12 0.76 0.10 0.75 0.11

79.8 1.39 262 2.61 160 2.35

features (e.g., water volumes, residence times and extent of mixing) and turbidity levels modulate the stimulatory effects of nutrients on plants and algae. Nutrient concentrations decreased

Table 5 Seasonal computation of surface and bottom nutrient flux in Ishikari Bay in 2006. Nutrient source

Spring

Summer

Autumn

River discharge (m3 s  1) Q bottom upwelling (m3 s  1) Surface (river) nutrient flux (mol s  1) DIN PO4 Si(OH)4 Bottom nutrient flux (mol s  1) DIN PO4 Si(OH)4 River/bottom nutrient flux ratio DIN PO4 Si(OH)4

2250 13, 968

400 22, 348

750 17, 887

76. 7 1.3 1114.6

21. 4 0.3 104.8

70, 2 0.6 120.0

15.6 1.7 19.4

17.7 2.2 58.3

21.5 2.0 42.0

4.9 0.8 57.4

1.2 0.1 1.8

3.3 0.3 2.9

Surface (river) nutrient flux¼R  nut. concentration. Bottom (upwelling) nutrient flux (F21) ¼ Q  nut. concentration.

in Ishikari Bay from the river-mouth (Plume) to marine-influenced area (Out-Plume), reflecting the main nutrient discharge from the Ishikari River. Although nutrient contribution into Ishikari Bay via bottom nutrient upwelling (Tables 4 and 5) may be negligible compared with riverine nutrients, benthic–pelagic nutrients coupling may be a cofueling source of production in spring and autumn in Ishikari Bay.

148

J.I. Agboola, I. Kudo / Continental Shelf Research 88 (2014) 140–150

4.2. Phytoplankton biomass and productivity in relation to nutrient fluxes Against the expected spring maximum phytoplankton biomass when riverine nutrients fluxes and nutrient concentrations were

Fig. 8. Water flux from surface (horizontal advection) versus surface water in-flux and out-flux ratio in Ishikari Bay (a), and water flux from bottom (upwelling) versus surface water in-flux and out-flux ratio in Ishikari Bay (b).

the highest in the Plume in 2006, the lower phytoplankton biomass in spring than in autumn was possibly due to either light or PO4 limitation (Agboola et al., 2009). At the near river mouth stations the light attenuation coefficient was the highest at 1.42 m  1, five times higher than the mean value for the Plume area (0.28 m  1). As light attenuation gradually decreased towards the offshore of the Plume, PO4 was less than 0.05 mM, suggesting P limitation (Justic et al., 1995a, 1995b). The relatively high light attenuation across the Plume in spring corresponded to the highest SPM value from the Ishikari River (Table 1) compared to summer and autumn. Most of the P removal from the water column takes place through sedimentation of organic matter (Berner et al., 1993) and since the highest river discharge occurred in spring, one may conclude that tide-dominated coastal waterways are generally turbid and light attenuation caused by suspended sediment is a major control on phytoplankton production and biomass (Hinga et al., 1995; Cloern, 1987; Monbet 1992). In 2007, phytoplankton biomass followed to develop a spring bloom, typical in temperature/subarctic latitudes. Phytoplankton biomass was at a maximum in spring when ambient nutrients concentrations were abundant due to winter vertical mixing. Compared with spring 2006, light attenuation across the Plume area was about three times higher at the mouth of the river Plume area in spring 2006 (1.42 m  1) than in spring 2007 (0.59 m  1). Thus, in 2007, light attenuation was not a limiting factor for primary production. Integrated primary production was the highest in autumn 2006 (932.85 mg C m  2 d  1) (Agboola, unpublished) after the spring bloom, when waters had plenty of nutrients and days were less cloudy, and after the summer thermohaline stratification. However, in 2007, integrated primary production was the highest in spring (1234.67 mg C m  2 d  1) (Agboola, unpublished) when phytoplankton responds to the increasing light, temperature, and nutrient influx from the Ishikari River and autochthonous source.

Fig. 9. Nutrient fluxes (mol s  1) of DIN, PO4, and Si(OH)4 in surface (Fin), bottom (F21), and from oceanic and riverine source in Ishikari Bay.

J.I. Agboola, I. Kudo / Continental Shelf Research 88 (2014) 140–150

149

According to Henríquez et al. (2007), the primary factors that affect the abundance and distribution of Chl a and primary production levels in the coastal zone are the physical forces that transport not only phytoplankton but also nutrients to the euphotic zone (i.e. upwelling fronts). Short residence time of nutrients in the river Plume area may have impacted on the low phytoplankton biomass and productivity observed in spring of 2006. Unlike 2006, primary productivity in spring of 2007 was neither limited by light nor P as light attenuation across the Plume area was relatively lower and ambient nutrients were in excess of apparent utilization.

The ratios of river nutrient flux (RNF) and bottom (upwelling) nutrient (BNF) were significantly greater than 1.0 for DIN and Si (OH)4, confirming our hypothesis that river nutrient significantly fueled autotrophic production in Ishikari Bay oligotrophic waters. However the ratios for PO4 were significantly less than 1.0, especially in summer and autumn, suggesting possible bottom upwelling as a driver. Furthermore, in spring when the ratio was close to 1 (0.8), coupling between bottom and river nutrient is hypothesized since river end-member concentration was the lowest for PO4 and the highest for SPM compared to other seasons.

4.3. Relative fluxes of nutrients in relation to river discharge versus estuarine circulation

5. Conclusion

On the assumption that bottom nutrient upwelling may be an important source of nutrient fueling primary production in Ishikari Bay, bottom nutrient flux was quantified. The computation of bottom nutrient flux (see Eq. (7) and Table 5) through estuarine circulation in Ishikari Bay has finally confirmed our assumption that bottom nutrient flux (especially PO4) significantly fuels autotrophic production in Ishikari Bay. Unlike the usual estuarine system, in which nutrient concentration at deeper layer is high due to the regeneration of nutrients at depth, concentration in Ishikari Bay is very low due to an influence of oligotrophic (low productivity) water. In the estuarine system where the nutrient profile is typical (i.e., it gradually increases with depth), nutrient flux from upwelling should contribute greater to the surface primary production than in Ishikari Bay. This is one of the reasons why riverine flux contributes a large portion of the total flux in Ishikari Bay. Flux ratios of river nutrient (RNF) and bottom nutrient (BNF) (i.e. R:B ratio) across the examined seasons were significantly 41.0 for DIN and Si (OH)4 nutrients, especially in spring and autumn. In summer, ratios were close to 1 for DIN (1.2) and Si (OH)4 (1.8) due to lower river discharge compared to spring and autumn (Table 1). The R:B ratio suggested a significant contribution of PO4 supply from bottom upwelling, especially in summer and autumn when values were 0.1 and 0.3, respectively. In spring, when the ratio was close to 1 (0.8), coupling of bottom (upwelling) and surface (river) occurred since PO4 concentration of river endmember was the lowest and SPM concentration highest across season. Most of the P removal from the water column takes place through sedimentation of organic matter (Berner et al., 1993), thus reducing the contribution from surface (river). The results obtained from this study clearly suggested that the relative contribution of nutrient flux through horizontal advection into Ishikari Bay depended upon m value, the ratio of Qin to Qout. The increase in m implied the increase in relative contribution of horizontal advection to the total discharge from Box 1. The degree of the increase in the nutrient flux differed by DIN, PO4 and Si(OH)4 and season. Across the seasons, when m value increased, there was a corresponding increase in horizontal nutrient flux (Fin). Also it seems that Si(OH)4 horizontal nutrient fluxes (Fin) are more important in spring and summer at m 40.5 whereas, for PO4, horizontal nutrient flux contribution is important across seasons. However, it was impossible to determine m by solving Eq. (13)–(16). A comparison of total oceanic nutrient fluxes (bottom upwelling and horizontal advection) to riverine nutrient fluxes suggested that the Ishikari River is a major source of DIN and Si(OH)4 supply in the bay, whereas oceanic nutrient contribution from bottom upwelling and horizontal advection is a major source of PO4. While the influence of riverine nutrient fluxes significantly fuels autotrophic production in the Ishikari Bay, information on the importance of estuarine circulation in buffering the supply of PO4 that is not adequate in riverine nutrient is a significant insight in this study.

Oceanographically Ishikari Bay could be considered as a transitional marine system influenced by oceanic (high salinity and low nutrients) and fresh waters (low salinity and high nutrients), modulating phytoplankton biomass and primary production dynamics on a seasonal scale. Generally, water mixing, biological consumption, and possible vertical convection were the three major factors controlling the nutrient distribution across the seasons: water mixing explains the gradual decrease of nutrient content offshore, biological consumption leads to noticeable removal of nutrients across the Plume water, and possible vertical convection carries nutrients upwards from bottom to the surface waters. This study discussed the Ishikari River as a main source of nutrients fueling primary production in the Ishikari Bay. For example, a strong inverse relationship (r¼ 0.927) between Chl a and salinity in autumn 2006 suggests that DIN from the Ishikari River produced the primary production maxima in autumn. In this season, an increase in the apparent utilization of DIN gave a corresponding increase in phytoplankton production. Also, the relatively high light attenuation across the Plume area in spring corresponded with the highest SPM value from the Ishikari River (Table 1). When riverine nutrients and nutrients fluxes were the highest in the Plume, the lower phytoplankton biomass of spring than autumn was possibly due to light and PO4 limitation (Agboola et al., 2009). In 2007 nutrient concentrations in spring and autumn were in excess of apparent utilization of DIN and PO4, corresponding to the observed maximum primary production in spring. This study successfully underscored the relative importance of riverine nutrient discharge against estuarine circulation in fueling autotrophic biomass and production in the Ishikari Bay. Also, from the results obtained, flux ratios of river nutrient (RNF) and bottom nutrient (BNF) (i.e. RNF:BNF) across the examined seasons were significantly 41.0 for DIN and Si (OH)4 nutrients, especially in spring and autumn. Nutrient flux ratio only suggests the significant contribution of PO4 nutrient from bottom upwelling, especially in summer and autumn. Thus, unlike other estuaries, the relative importance of estuarine circulation in fueling autotrophic production is relatively minimal in the Ishikari Bay.

Acknowledgments We thank the Captain and crews of TS Ushio-Maru and OshoroMaru, and colleagues at the laboratory for their assistance in sampling during the cruises. References Agboola, J.I., Uchimiya, M., Kudo, I., Kido, K., Osawa, M., 2013. Seasonality and environmental drivers of biological productivity on the western Hokkaido coast, Ishikari Bay, Japan. Estuar. Coast. Shelf S. 127, 12–23. http://dx.doi.org/ 10.1016/j.ecss.2013.03.008.

150

J.I. Agboola, I. Kudo / Continental Shelf Research 88 (2014) 140–150

Agboola, J.I., Uchimiya, M., Kudo, I., Kido, K., Osawa, M., 2010. Dynamics of pelagic variables in two contrasting coastal systems in the western Hokkaido coast off Otaru port, Japan. Estuar. Coast. Shelf S. 86 (3), 477–484. Agboola, J.I., Yoshi, S., Kudo, I., 2009. Seasonal change of riverine nutrients and distribution of chlorophyll a in Ishikari Bay, subarctic oligotrophic coastal environment of Japan. Jpn.—Fr. Oceanogr. Soc. J., La Mer 47, 33–49. ANZECC/ARMCANZ, 2000. Australian and New Zealand Guidelines for Fresh and Marine Water Quality. 〈www.ea.gov.au/water/quality/nwqms/#quality〉. Barber, R.T., Marra, J., Bidigare, R.C., Codispoti, L.A., Halpern, D., Johnson, Z., Latasa, M., Goericke, R., Smith, S.L., 2001. Primary productivity and its regulation in the Arabia Sea during 1995. Deep-Sea Res. II 48, 1127–1172. Berner, R.A., Ruttenberg, K.C., Ingall, E.D., Rao, J.L., 1993. The nature of phosphorus burial in modern marine sediments. In: Wollast, R., Mackenzie, F.T., Chou, L. (Eds.), Interactions of C, N, P and S Biogeochemical Cycles and Global Change, (Eds.) Springer-Verlag, New York, pp. 365–378. Brzezinski, M.A., 1985. The Si: C: N ratio of marine diatoms: interspecific variability and the effect of some environmental variables. J. Phycol. 21, 347–357. Cloern, J.E., 1987. Turbidity as a control on phytoplankton biomass and productivity in estuaries. Cont. Shelf Res. 7, 1367–1381. Dodimead, A.J., Favorite, A.E., Hirano, T., 1963. Salmon of the North Pacific Ocean, Part II: Review of Oceanography of the Subarctic Pacific Region. Bulletin of the International North Pacific Fisheries Commission, No. 13, 195 pp. Dodimead, A.J., 1967. Autumn Oceanographic onditions in the Central Subarctic Pacific. International North Pacific Commission Document 999, pp. 1–14. Harris, G.P., 2001. Biogeochemistry of nitrogen and phosphorous in Australian catchments, rivers and estuaries: effects of land use and flow regulation and comparisons with global patterns. Mar. Freshw. Res. 52, 139–149. Henríquez, L.A., Daneri, G., Mũnoz, C.A., Montero, P., Veas, R., Palma, A.T., 2007. Primary production and phytoplanktonic biomass in shallow marine

environments of central Chile: effect of coastal geomorphology. Estuar. Coast. Shelf Sci. 73, 137–147. Justic, D., Rabalais, N.N., Turner, R.E., 1995a. Stoichiometric nutrient balance and origin of coastal eutrophication. Mar. Poll. Bull. 30, 41–46. Justic, D., Rabalais, N.N., Turner, R.E., Dortch, Q., 1995b. Changes in nutrient structure of river-dominated coastal waters, stoichiometric nutrient balance and its consequences. Estuar. Coast. Shelf Sci. 40, 339–356. Kristiansen, S., Hoell, E.E., 2002. The importance of silicon for marine production. Hydrobiologia 484, 21–31. Kudo, I., Noiri, Y., Imai, K., Nojiri, Y., Nishioka, J., Tsuda, T., 2005. Primary productivity and nitrogenous nutrient assimilation dynamics during the Subarctic Pacific Iron Experiment for Ecosystem Dynamics Study. Prog. Oceanogr. 64, 207–221. Parsons, T.R., Maita, Y., Lalli, C.M., 1984. A Manual of Chemical and Biological Methods for Seawater Analysis. Pergamon Press, New York. Redfield, A.C., 1934. On the proportions of organic derivatives in sea water and their relation to the composition of plankton. In: Daniel, R.J. (Ed.), James Johnstone Memorial Volume. University of Liverpool Press, Liverpool, UK, pp. 176–192. Redfield, A.C., Ketchum, B.H., Richards, F., 1963. The influence of organisms on the composition of seawater. In: Hill, M.N. (Ed.), The Sea, vol. 2. John Wiley, New York, pp. 26–77. Unoki, S., 1998. Relation between the transport of gravitational circulation and the river discharge in bays (In Japanese with English abstract). Umi no Kenkyu 7, 283–292. Yoshida, K., Domon, K., Watanabe, T., 1977. Physical and chemical environment in Ishikari Bay (in Japanese with English abstract). Sci. Rep. Hokkaido Fish. Exp. Stn. 34, 1–6.