Relationship between Salinity and Nutrients in the Subsurface Layer in the Suruga Bay

Transcription

1 Journal of Oceanography, Vol. 61, pp. 721 to 732, 25 Relationship between Salinity and Nutrients in the Subsurface Layer in the Suruga Bay TATSUYA IWATA 1 *, YOSHIKO SHINOMURA 1, YUTA NATORI 1, YASUMASA IGARASHI 2, RUMI SOHRIN 1 and YOSHIMI SUZUKI 1 1 Graduate School of Science and Engineering, Shizuoka University, Oya, Suruga-ku, Shizuoka , Japan 2 Shizuoka Prefectual Fisheries Experiment Station, Kogawa-shioiri, Yaizu , Japan (Received 2 June 23; in revised form 24 November 24; accepted 24 November 24) We investigated the water structure and nutrient distribution in the Suruga Bay from April 2 to July 22, especially the Offshore Water, which occupies a large part of the bay. The maximum salinity in the upper 2 m varied between and 34.71, indicating a temporal change in the influence of Kuroshio Water on the Offshore Water. Seasonal variation in nutrient concentrations was largest from surface to 5 m. On the other hand, the variance in nutrient concentrations within each season was largest in the subsurface layer of 1 3 m in spring, summer and fall. In the Offshore Water, the change of nutrients was negatively correlated with that of salinity in each season. This suggests that an increasing intrusion of saline water brings about a lower nutrient concentration in the Offshore Water. Likewise, negative correlations were observed between the change of the maximum salinity and chlorophyll a ( [chl.aint])/nutrients integrated in the upper 2 m. [chl.a-int] was significantly correlated with the changes of nitrate and phosphorus, but there were no significant correlations between [chl.a-int] and the change of silicate. These results suggest that the concentrations of chlorophyll a and nutrients in the Offshore Water were decreased due to the increasing intrusion of Kuroshio Water. The Offshore Water is likely to be related to the regulation of primary production by nitrate. Keywords: Suruga Bay, subsurface layer, salinity, nutrients, chlorophyll a. 1. Introduction The Suruga Bay is located on the southern coast of Honshu, Japan. It opens to the Pacific Ocean at its southend and its depth is about 25 m in the deepest part. Kuroshio Water intrudes into the bay along the Suruga Trough, which extends from the bay head to the eastern part of the bay mouth. The intrusion of Kuroshio Water and its circulation within the bay is strongly influenced by the position of the Kuroshio axis (Inaba, 1984). Nakamura (1982) defined the water mass structure in the Suruga Bay in terms of the following five discrete water masses: (A) Coastal Water located near the western coast and the bay head and influenced by the river water (potential temperature (θ): >13 C; salinity: <34.3; depth: 2 m); (B) Offshore Water characterized by a salinity maximum and showing the characteristics of Kuroshio Water (θ: C; salinity: ; depth: * Corresponding author. stiwata@ipc.shizuoka.ac.jp Copyright The Oceanographic Society of Japan. 1 2 m); (C) Surface Water, which is a mixture of the Coastal Water and the Offshore Water (θ: C; salinity: ; depth: 1 m); (D) Intermediate Water characterized by a salinity minimum (θ: 3 11 C; salinity: ; depth: 2 12 m); (E) Pacific Deep Water (θ: <3 C; salinity: >34.4; depth: >12 m). Nakamura (1982) reported that θ and salinity varied seasonally in the Coastal Water, the Surface Water and the Offshore Water. The Surface Water and the Offshore Water were indistinguishable in winter due to convective mixing. Previous studies of the Suruga Bay mainly focused on its physical aspect (Inaba, 1981, 1984; Nakamura, 1982; Takeuchi and Hibiya, 1997) and there are few biogeochemical studies. Aoki (1993) found a negative correlation between chlorophyll a and salinity in the surface layer of the Suruga Bay and suggested that the correlation was related to the supply of nutrients from the river water. Shiomoto and Hashimoto (1999) reported the possible limitation of primary production by inorganic nitrogen in the offshore surface water of the Suruga Bay 721

2 from spring to fall, especially for diatoms. These studies suggest that phytoplankton biomass is partially limited by the low level of nutrients, but it is not clear what mechanism works to limit nutrients in the Suruga Bay. To understand the relationship between nutrients and the water structure, we studied the spatio-temporal distributions of salinity, nutrients and chlorophyll a along the zonal vertical cross-sections located in the middle part of the Suruga Bay from April 2 to July 22. We found large temporal variations in the maximum salinity in the upper 2 m, and negative correlations between the maximum salinity and the nutrient concentrations at the maximum salinity depth. 2. Materials and Methods Observations were conducted aboard the R/V Suruga-Maru at Sta. 2, Sta. 3, Sta. E and Sta. F in the Suruga Bay from April 2 to July 22 (Fig. 1). Sta. 2 and Sta. 3 are the time-series station of Shizuoka Prefectural Fisheries Experiment Station, established in 1997, and Sta. E and Sta. F were established later between Sta. 2 and Sta. 3. Salinity and in situ temperature were measured by CTD (SBE 9 plus; Sea-Bird Electronics, USA) every 1 dbar between 5 m and the depth 2 m above the sea bottom (henceforth B-2 (m)). Potential temperature (θ) was calculated from salinity, in situ temperature and pressure. Detailed sampling depths are as follows: 2,1, 2, 3, 5, 7, 1, 125, 15, 2, 3 and B-2 (m) at Sta. 3; 1, 2, 5, 7, 1, 125, 15, 2, 3, 4, 6 and B-2 (m) at Sta. F; 1, 2, 3, 5, 7, 1, 125, 15, 2, 3, 4 and B-2 (m) at Sta. E; 1, 2, 5, 1, 15, 2, 3, 4, 6, 8, 1 and B-2 (m) at Sta. 2. Additional sampling depths were introduced at Sta. 2: 3, 4, 6, 125 and 175 (m) in May 22 and 3, 4, 6, 125, and 1288 (m) in July 22. Samples were collected with 1-l Niskin bottles mounted on a CTD Carousel water sampler system. Surface water samples were collected using a plastic bucket (Stas. 2, E and F) or a Niskin bottle (Sta. 3; 2 m). Nutrient samples were collected in 1 ml polypropylene bottles from the Niskin bottles and kept in freezer ( 3 C) until analysis. Nitrate (NO 3 ), phosphate (PO 4 ) and silicate (Si(OH) 4 ) were measured using a TRAACS 2 (BRAN + LUEBBE GmbH., Germany) (Hansen and Koroleff, 1999). Si(OH) 4 was measured only for the samples taken between November 2 and July 22. The precision of the nutrient analysis was ±.2%, ±.8% and ±.5% for NO 3, PO 4 and Si(OH) 4, respectively, estimated from the coefficient of variation of 5 replicate analyses of the seawater sample which contained 13.9 µmol l 1 of NO 3,.92 µmol l 1 of PO 4 and µmol l 1 of Si(OH) 4. The detection limit was.5 µmol l 1,.2 µmol l 1 and.3 µmol l 1 for NO 3, PO 4 and Si(OH) 4, respectively, estimated by multiplying 3 by Japan Oi river Yaizu Omaezaki Abe river E Sta.3 Shizuoka Sta.F Sta.E Suruga Bay the standard deviation (SD) of 5 replicate analyses of 3.5% NaCl solution. Chlorophyll a was measured for the samples taken at depths of, 1, 2, 3, 5, 1, 15 and 2 m at Stas. 3, E and F, and, 1, 2, 5, 1, 15 and 2 m at Sta. 2. Samples were collected into dark polyethylene bottles from the Niskin bottles, and 3 ml of the samples was immediately filtered through a GF-75 glass fiber filter (TOYO ROSHI Corp., Tokyo) with suction (~2 hpa). Although the nominal pore size of GF-75 is mentioned as.3 µm by the manufacture, the retention efficiency of particulate organic matter in seawater samples was reported to be similar between GF-75 and Whatman GF/F filter (Pike and Moran, 1997), the latter being widely used for chlorophyll a analysis. The filter samples were soaked in 1 ml N,N-dimethylformamide and kept in freezer ( 3 C) until analysis (Suzuki and Ishimaru, 199). Chlorophyll a was measured with a spectrofluorometer (RF- 53PC; Shimadzu Co., Kyoto) calibrated with a commercial chlorophyll a standard derived from Spirulina (Wako Pure Chemical Industries, Ltd., Osaka). The precision of the analysis was ±.7% estimated from the coefficient of variation of 4 replicated analyses of the standard solution (5 µg l 1 ) and the detection limit was.4 µg l 1 estimated by multiplying 3 by SD. Sta E Irohzaki Izu Penn. 35 N 34 4 N Fig. 1. Sampling stations in the Suruga Bay. Station locations are as follows: Sta. 3: N, E; Sta. F: N, E; Sta. E: N, E; Sta. 2: N, E. Water depths were ca. 5 m (Sta. 3), 8 m (Sta. F), 6 m (Sta. E) and 16 m (Sta. 2). 722 T. Iwata et al.

3 salinity θ ( C) σ θ Apr.2 Jul.2 Sep.2 Feb.21 Apr.21 Jul.21 Nov.2 Dec.21 May 22 Jul.22 Oct.21 Feb.22 Fig. 2. Vertical profiles of salinity, potential temperature (θ) and potential density anomaly (σ θ ) at Sta. 2 in the Suruga Bay between April 2 and July 22. The profiles in February 21 are only shown for the upper 2 m due to the error in CTD below 2 m. 3. Results 3.1 Characteristics of the water structure in the Suruga Bay Figure 2 shows vertical profiles of salinity, θ, and potential density (σ θ ) at Sta. 2 in the Suruga Bay between April 2 and July 22. We first calculated the mixed layer depth (MLD) as the depth at which σ θ surpasses σ θ at the surface (5 m) by.125 (Levitus, 1982). Table 1 shows MLD at each station. MLD showed clear seasonal changes at all stations with the largest values in winter and the smallest values in summer. MLD was 6 55 m in spring (April 2, April 21, May 21), 5 14 m in summer (July 2, 21 and 22), 6 65 m in fall (September and November 2 and October 21), and m in winter (February and December 21 and February 22) (Table 1). Salinity at Sta. 2 varied with the sampling dates in the upper 2 m (Fig. 2). The maximum salinity in the upper 2 m and the ranges of its related depth, θ and σ θ, are shown in Table 2. In spring, the maximum salinity was observed at m in April 2 and May 22, but in April 21, its location moved to 5 8 m. The salinity was higher in May 22 (34.71) than in April 2 (34.59) and April 21 (34.6). In summer, the maximum salinity was observed at greater depths (61 14 m) than in spring. The salinity was highest in July 22 (34.71), followed by July 2 (34.61) and 21 (34.58). In fall, the depth of the maximum salinity was even greater at m and the salinity was higher in September 2 (34.71) than in November 2 (34.57) and October 21 (34.56). In winter, the maximum salinity was observed in the upper 161 m and the salinity was highest (34.6) in February 21, and in December 21 and Relationship between Salinity and Nutrients in the Suruga Bay 723

4 Table 1. Mixed layer depth (m) in the Suruga Bay between April 2 and July 22. Season Date Sta. 3 Sta. F Sta. E Sta. 2 Spring April 24, April 25, May 14, Summer July 1, July 11, July 23, Fall September 2, November 27, October 25, Winter February 19, December 12, February 2, Table 2. Maximum salinity in the upper 2 m and its related depth, θ and σ θ at Sta. 2. Depth, θ and σ θ are shown as the range in the isohaline of maximum salinity. Season Date Salinity Depth (m) θ ( C) σ θ Spring April 24, April 25, May 14, Summer July 1, July 11, July 23, Fall September 2, November 27, October 25, Winter February 19, December 12, February 2, in February 22. When the maximum salinity reached the highest in each season, θ at the maximum salinity depth also showed the highest value (Table 2). The maximum of salinity and θ in each season were observed in May 22 for spring, in July 22 for summer, in September 2 for fall and in February 21 for winter (Table 2). On the other hand, σ θ observed during the above mentioned periods, was low at the depth of the maximum salinity: the lowest σ θ was observed in May 22, July 22, September 2 and February 21 for spring, summer, fall and winter, respectively (Table 2). Salinity, θ, or σ θ differed less with the sampling dates below 2 m (Fig. 2). Nakamura (1982) defined the Offshore Water in the Suruga Bay as the water characterized by the salinity maximum with a salinity of and θ of C. We observed the Offshore Water in 32 m in spring, 3 34 m in summer, 6 22 m in fall, and 1 23 m in December 21 and 2 m in February 21 and 22 (Fig. 2). Figure 3 shows the vertical section of salinity in the Suruga Bay in fall. To emphasize the distribution of the subsurface salinity maximum, salinity contours lower than 34. were omitted. The subsurface salinity maximum was 724 T. Iwata et al.

5 Sta.3 Sta.FSta.E Sta.2 Sta.3 Sta.FSta.E Sta.2 Sta.3 Sta.FSta.E Sta < < < H (>34.6) Depth (m) Sep.2 Nov.2 Oct.21 Fig. 3. Vertical sections of salinity along the east-west transect in the Suruga Bay in September 2, November 2 and October 21. Contour interval is salinity =.1 and contours of salinity lower than 34. are omitted. characterized as the water with the salinity higher than 34.6 in September 2. It was observed between 6 m and 13 m at Sta. 2 and became thinner to the west, i.e. it was found around 6 1 m at Sta. E and between 65 and 75 m at Sta. F, but no sign of it was discernable at Sta. 3. Meanwhile, no water with salinity higher than 34.6 was not observed at any stations in November 2 or October 21 (Fig. 3). 3.2 Spatio-temporal distributions of nutrients Figure 4 shows the vertical profiles of nutrients at Sta. 2. Several characteristics of the temporal changes were detected in the vertical nutrient distribution at Sta. 2. Seasonal variation in nutrient concentrations was largest in the surface layer ( 5 m). Seasonal differences of the average concentration of NO 3 in the surface layer were µmol l 1 depending on the depth. In contrast, the seasonal difference of the average concentration of NO 3 below 1 m ranged from 1.99 to 4.41 µmol l 1, depending on the depth. As to the seasonal variation in PO 4 and Si(OH) 4, similar characteristics were exhibited: the maximum difference of the seasonal PO 4 average was µmol l 1 in 5 m and µmol l 1 below 1 m, whereas the difference of Si(OH) 4 was also large ( µmol l 1 ) in 3 m compared to that below 1 m ( µmol l 1 ). The singular exception was the difference of Si(OH) 4 at 8 m, which had a large value of 1.22 µmol l 1. Nutrient concentrations in the upper 2 m were low from April to October, and markedly high in February (Fig. 4). In the upper 2 m, NO 3 was mostly below the detection limit (.5 µmol l 1 ) from April to October except for April 2 ( µmol l 1 ). PO 4 was below the detection limit (.2 µmol l 1 ) at 1 m in April 2, and Si(OH) 4 was below the detection limit (.3 µmol l 1 ) at 2 m in July 21. Except for these two cases, PO 4 and Si(OH) 4 were detectable in the upper 2 m with concentrations of.3.33 µmol l 1 and µmol l 1, respectively. These results suggest that primary production would have been limited mostly by the deficiency of NO 3 in spring (April and May), summer (July) and fall (September and October) at Sta. 2. In the upper 2 m, NO 3 concentration increased to µmol l 1 in November 2 and it was at a high level ( µmol l 1 ) during the winter months of December February (Fig. 4). In winter, NO 3, PO 4 and Si(OH) 4 were almost uniform in the upper 1 m, although the concentrations varied with the sampling dates. Nutrient concentrations were lowest in December 21 and highest in February 22 in the upper 1 m (Fig. 4). On the other hand, winter MLD was shallowest (88 m) in December 21 and deepest (16 m) in February 22 (Table 1). The averages of nutrient concentrations in the upper 1 m were positively correlated (r =.77.84, n = 12, P <.5) with MLD at all stations in winter, implying the regulation of the upper nutrient concentrations by the MLD in winter. Variations in the nutrient concentrations within each season were larger in the subsurface layer (1 3 m) than in the surface layer in spring, summer and fall (Fig. 4). For example, at Sta. 2, the range of NO 3 concentrations in spring was µmol l 1 in 1 3 m, whereas it was µmol l 1 and µmol l 1 in 5 m and 4 1 m, respectively. PO 4 Relationship between Salinity and Nutrients in the Suruga Bay 725

6 Depth (m) NO3 (µmol l -1 ) Depth (m) PO4 (µmol l -1 ) Si(OH)4 (µmol l -1 ) Depth (m) Apr.2 Jul.2 Sep.2 Feb.21 Apr.21 Jul.21 Nov.2 Dec.21 May 22 Jul.22 Oct.21 Feb.22 Fig. 4. Vertical profiles of nutrient concentrations (NO 3, PO 4 and Si(OH) 4 ) at Sta. 2 in the Suruga Bay between April 2 and July 22. Si(OH) 4 was not measured in April, July and September 2. and Si(OH) 4 exhibited similar characteristics: the range of PO 4 in spring was µmol l 1 in 1 3 m, and.9.24 µmol l 1 and µmol l 1 in 5 m and 4 1 m, respectively. The range of Si(OH) 4 was also largest in the subsurface layer in spring and summer: it was µmol l 1 in 1 3 m in spring and µmol l 1 in 15 3 m in summer. However, the range of Si(OH) 4 in fall was larger in 4 1 m ( µmol l 1 ) than in 1 3 m ( µmol l 1 ). In winter, the ranges were relatively large in 5 m compared to the one below 1 m: the range in 5 m was µmol l 1, µmol l 1, and µmol l 1 for NO 3, PO 4 and Si(OH) 4, respectively, whereas the ranges below 1 m were µmol l 1,.7.4 µmol l 1, and µmol l 1, respectively. The nutrient concentrations at the depth of salinity maximum were estimated for the upper 2 m at Sta. 2 by applying linear interpolation of the concentrations at the closest depths to the depth to be estimated, and they were then plotted against the maximum salinity (Fig. 5). The nutrient concentrations at the salinity maximum were significantly correlated with the salinity (Fig. 5) and the regressions are described as follows: [NO 3 ] Smax = S max (r 2 =.48, n = 12, P <.5); [PO 4 ] Smax = 3.6 S max (r 2 =.47, n = 12, P <.5); [Si(OH) 4 ] Smax = S max (r 2 =.59, n = 9, P <.5), where [NO 3 ] Smax, [PO 4 ] Smax and [Si(OH) 4 ] Smax are the concentrations of NO 3, PO 4 and Si(OH) 4 at the maximum salinity, respectively, and S max is the maximum salinity. By standardizing [nutrient] Smax with its average, the regression between S max and the standardized [nutrient] Smax creates the slope of 6.95, 5.85 and 6.3 for NO 3, PO 4 and Si(OH) 4, respectively. This result suggests that there appears to be little difference as to the rate of decrease in the concentration among the nutrients, which is inversely proportional to the increase in the maximum salinity. We examined the relationship between salinity and nutrients in the Offshore Water. We calculated salinity 726 T. Iwata et al.

7 2 1 NO 3 (µmol l -1 ) PO 4 (µmol l -1 ) Si(OH) 4 (µmol l -1 ) salinity salinity salinity salinity Fig. 5. Relationship between salinity and nutrient concentrations (NO 3, PO 4 and Si(OH) 4 ) at the depth of the salinity maximum in the upper 2 m at Sta. 2 in the Suruga Bay. Solid line indicates a linear regression between salinity and nutrient concentrations. anomaly ( salinity) and nutrients anomaly [ nutrients: NO 3, PO 4 and Si(OH) 4 ] at each station and depth in the Offshore Water by subtracting the average value at all the sampling dates from each value. The average value was calculated at each station and depth. This approach was applied to remove the influence of the spatial variation in salinity and nutrients. Figure 6 shows the relationship between salinity and nutrients in the Offshore Water and Table 3 shows the summary of the linear regression. There were negative correlations (r 2 =.37.55, n = , P <.5) between salinity and nutrients ( NO 3, PO 4 and Si(OH) 4 ), when all the data were considered (Table 3). This result suggests that an increase in the intrusion of saline water brought down the nutrient levels in the Offshore Water. salinity and salinity spring summer fall winter Fig. 6. Vertical profiles of chlorophyll a concentration at Sta. 2 in the Suruga Bay between April 2 and July 22. nutrients were also negatively correlated in each season, except for the relationship between salinity and Si(OH) 4 in fall (r 2 =.12, n = 31, P >.5) (Table 3). These slopes were not significantly different in each season, respectively, for NO 3, PO 4 and Si(OH) Spatio-temporal distribution of chlorophyll a The vertical profiles of chlorophyll a at Sta. 2 are shown in Fig. 7. In most cases, chlorophyll a reached the maximum between and 2 m and was below the detection limit (.4 µg l 1 ) at depths below 1 m. Chlorophyll a concentration was high at Sta. 2 in April 2 (maximum: 2.9 µg l 1 at 2 m), in April 21 (1.6 µg l 1 Relationship between Salinity and Nutrients in the Suruga Bay 727

8 Table 3. Summary of the linear regressions between salinity and nutrients ( NO 3, PO 4 and Si(OH) 4 ). The model used is nutrients = a salinity + b. Period nutrients Slope (a) Intercept (b) r 2 n P* All season NO <.5 PO <.5 Si(OH) <.5 Spring NO <.5 PO <.5 Si(OH) <.5 Summer NO <.5 PO <.5 Si(OH) <.5 Fall NO <.5 PO <.5 Si(OH) >.5 Winter NO <.5 PO <.5 Si(OH) <.5 *Test of the null hypothesis: a =. chlorophyll a (µg l -1 ) Depth (m) Apr.2 Jul.2 Sep.2 Feb.21 Apr.21 Jul.21 Nov.2 Dec.21 May 22 Jul.22 Oct.21 Feb.22 Fig. 7. Relationship between salinity and nutrients ( NO 3, PO 4 and Si(OH) 4 ) in the Offshore Water of the Suruga Bay. salinity and nutrients represent the anomalies of salinity and nutrient concentrations, respectively. Solid line indicates a linear regression line between salinity and nutrients, when all the data are considered. at m) and in October 21 (1.6 µg l 1 at 1 m). In other sampling periods, chlorophyll a ranged from.5 µg l 1 to 1.2 µg l 1 in the zone between and 5 m. We list salinity, nutrients and chlorophyll a observed at 1 m in the Suruga Bay in Table 4. Note that 1 m was the shallowest depth at which we obtained both chlorophyll a and salinity data. Salinity in the surface water (1 m) was mostly (9 cases in all 12 cases) lowest at Sta. 3 ( ; Table 3), which is closest to the mouth of Oi and Abe Rivers (Fig. 1). In contrast, surface chlorophyll a was mostly (8 cases in all 12 cases) highest at Sta. 3 ( µg l 1 ), implying the effect of the river 728 T. Iwata et al.

9 Table 4. Salinity and concentrations of NO 3, PO 4, Si(OH) 4 and chlorophyll a at 1 m in the Suruga Bay. Season Date Salinity NO 3 (µmol l 1 ) PO 4 (µmol l 1 ) Sta. 3 Sta. F Sta. E Sta. 2 Sta. 3 Sta. F Sta. E Sta. 2 Sta. 3 Sta. F Sta. E Sta. 2 Spring April 24, April 25, May 14, Summer July 1, July 11, July 23, Fall September 2, November 27, October 15, Winter February 19, December 12, February 2, Season Date Si(OH) 4 (µmol l 1 ) Chlorophyll a (µg l 1 ) Sta. 3 Sta. F Sta. E Sta. 2 Sta. 3 Sta. F Sta. E Sta. 2 Spring April 24, April 25, May 14, Summer July 1, July 11, July 23, Fall September 2, November 27, October 15, Winter February 19, December 12, February 2, : Not determined. Relationship between Salinity and Nutrients in the Suruga Bay 729

10 Table 5. Integrations of chlorophyll a and nutrients (NO 3, PO 4, Si(OH) 4 ) with depth in the upper 2 m at Sta. 2. Season Date Chlorophyll a NO 3 PO 4 Si(OH) 4 (mg m 2 ) (mmol m 2 ) (mmol m 2 ) (mmol m 2 ) Spring April 24, April 25, May 14, Summer July 1, July 11, July 23, Fall September 2, November 27, October 15, Winter February 19, December 12, February 2, : Not determined. Table 6. Summary of Spearman s rank correlation coefficients between each two variables of S max, [nutrients-int] and [chl.a-int]. x y r s n P* S max [NO 3 -int] <.5 [PO 4 -int] <.5 [Si(OH) 4 -int] <.5 [chl.a-int] <.5 [chl.a-int] [NO 3 -int] <.5 [PO 4 -int] <.5 [Si(OH) 4 -int] >.5 *Test of the null hypothesis: a =. water on phytoplankton biomass in the western coast of the Suruga Bay throughout the year. On the other hand, the surface nutrient concentrations were high not only at Sta. 3 but also at other stations. For example, the surface PO 4 was mostly (7 cases in all 12 cases) highest at Sta. 2, followed by Sta. 3 (n = 4) and Sta. F (n = 1) (Table 4). In summer, chlorophyll a was lowest at all the stations and NO 3 was mostly (11 cases in all 12 cases) below the detection limit, implying the limitation of primary production by the depletion of NO 3 in summer (Table 4). To determine the relationship between the levels of nutrients in the Offshore Water and phytoplankton biomass, we examined the concentrations of nutrients and chlorophyll a integrated in the upper 2 m in the Offshore Water. The values are listed in Table 5. In spring, the chlorophyll a and the nutrients were lowest in May 22 (Table 5), whereas salinity in the upper 2 m was highest during the same period (Table 2). In summer, when the maximum salinity was highest, the chlorophyll a and the nutrients were lowest (Tables 2 and 5). These results imply the negative correlation between the maximum salinity and the chlorophyll a. We calculated the anomalies of the maximum salinity ( S max ), integrated chlorophyll a ( [chl.a-int]) and integrated nutrients ( [nutrients-int]: [NO 3 -int], [PO 4 - int] and [Si(OH) 4 -int]). These were calculated by subtracting the average value in all the sampling dates from each value. The average value was calculated at each station. This approach was applied in order to remove the influence of the horizontal variation in the maximum salinity, nutrients and chlorophyll a. Table 6 shows the summary of Spearman s rank correlation between each two variables of S max, [chl.a-int] and [nutrients-int]. Similar to the relationship between salinity and nutrients (Fig. 7), there were negative correlations (r s =.43~.62, n = 36 48, P <.5) between S max and [nutrients-int]. S max was also negatively correlated with [chl.a-int] (r s =.32, n = 48, P <.5) (Table 6). These results suggest that the levels of nutrients and chlorophyll a in the upper 2 m dropped with the increase in the intrusion of saline and oligotrophic water into the Offshore Water. [chl.a-int] was significantly correlated with [NO 3 -int] and [PO 4 -int], but there was no significant correlation between [chl.a-int] and [Si(OH) 4 -int] (Table 6). 73 T. Iwata et al.

11 4. Summary and Discussion NO 3 was depleted between April and October in the upper 5 m at Sta. 2, whereas PO 4 and Si(OH) 4 were mostly detectable. This result is consistent with that reported by Shiomoto and Hashimoto (1999), in which they suggested that primary production was limited by inorganic nitrogen in the offshore surface water of the Suruga Bay from spring to fall. In the present study, the surface chlorophyll a concentration was highest at Sta. 3 among other stations (Table 4). Sta. 3 is close to the river mouth where the surface salinity was lowest (Fig. 1, Table 4). These results suggest that the phytoplankton production is stimulated by supply of nutrients from river water (Lohrenz et al., 1999; Chen et al., 2). This is supported by the survey of Aoki (1993), which suggested that this process occurs in the Suruga Bay, based on the spatial distribution of chlorophyll a and salinity and the relationship between them. On the other hand, in the present study, the surface nutrient concentrations differed little between Sta. 2 and Sta. 3 (Table 4). Toyota (1985) reported NO 3 concentration near the Abe river estuary was µmol l 1 at the surface water ( and 5 m) in October, and these NO 3 concentrations were higher than NO 3 concentration in fall at the surface water in this study (Table 4). If nutrient is supplied by river water into the surface water, then lower nutrient concentrations at the surface water (lower compared to a previous report (Toyota, 1985)) may be caused by consumption of phytoplankton. As another possibility, chlorophyll a distribution was mainly affected by the input of phytoplankton through the river water, because the small difference of nutrient concentration at the surface water between Sta. 2 and Sta. 3 is related to the relatively lower supply of nutrient through the river water. We observed a large temporal variation in the maximum salinity in the upper 2 m: it ranged from to at Sta. 2 from April 2 to July 22 (Table 2). The salinity range was consistent with values reported in the previous studies (Nakamura and Muranaka, 1979; Nakamura, 1982; Toyota et al., 1993) (Fig. 2). On the other hand, the variation in nutrient within each season was largest in the subsurface layer (1 3 m), and it has not been reported in previous studies of the Suruga Bay as far as we know. In the salinity maximum layer, the salinity was significantly correlated with nutrient concentration (Fig. 5). Moreover, there was a significant negative correlation between salinity and nutrients in the Offshore Water (Table 3), suggesting that an increasing intrusion of saline water brings about a lower nutrient concentration in the Offshore Water. One possible source of the saline water is Kuroshio Water, because the highest salinity was related to the highest θ in each season (Table 2). Recent studies report that the maximum salinity in the subsurface layer was greater than 34.8 near the Kuroshio axis (Komatsu and Kawasaki, 22), and ca in Enshu-nada, outside the Suruga Bay (Kasai et al., 22). When the salinity maximum is higher in each season, the maximum salinity in the Offshore Water is higher than the value in Enshu-nada and close to the salinity of the Kuroshio Water (Table 2). This supports the possibility that the saline water originates from the Kuroshio Water. Inaba (1984) observed the counterclockwise circulation of the surface water in the Suruga Bay when the Kuroshio axis approached the coast of Honshu. The salinity vertical section in September 2 (Fig. 3) suggests that water with the salinity higher than 34.6 intruded into the Suruga Bay from east to west, implying a counterclockwise circulation. Therefore, the approach of the Kuroshio axis to the Suruga Bay might be the cause of higher salinity as well as higher θ and lower nutrient concentrations in the Offshore Water. Chlorophyll a and nutrient integrated in upper 2 m was lowest in May 22 and July 22, when the maximum salinity was highest in each season. The relationship between S max and [chl.a-int] was negative, and those between [chl.a-int] and [NO 3 -int] or [PO 4 -int] were positive (Table 6), which suggests that lower NO 3 and PO 4 concentration, caused by the increasing intrusion of saline water, is related to lower phytoplankton production. Furthermore, because the mixed layer was usually contained in the Offshore Water in spring and winter (see above and Table 1), a decrease in the nutrient concentrations in the Offshore Water would have extended into the euphotic zone. On the other hand, [Si(OH) 4 - int] was not significantly correlated with [chl.a-int], suggesting the regulation of primary production by NO 3 and PO 4 rather than Si(OH) 4 in the upper 2 m. We assume that the higher contribution of the nutrient-deficient, warm, saline Kuroshio Water to the Offshore Water may decrease the phytoplankton biomass and total primary production in the upper 2 m on a seasonal basis. This is consistent with the limitation of primary production by NO 3 in the surface layer from April to October. In contrast, the surface (1 m) phytoplankton biomass would be related to the inflow of the river water to the bay (Table 3). Because the Offshore Water accounts for a significant portion of the total water volume of the Suruga Bay ( %; Nakamura (1982)), and because part of the euphotic zone (21 6 m; Shiomoto and Hashimoto (1999)) is usually occupied by the Offshore Water, changes in the contribution of Kuroshio Water to the Offshore Water may have considerable effects on the nutrient dynamics and primary production in the Suruga Bay. Acknowledgements We thank the captain and the crew of R/V Suruga- Maru, and Mr. Y. Hagiwara and T. Hanai (Shizuoka Prefectural Fisheries Experiment Station) for their support Relationship between Salinity and Nutrients in the Suruga Bay 731

12 in sampling. We are also grateful to Dr. Y. Nakagawa for his valuable comments. This work was funded by the Japan Ministry of Education, Culture, Sports, Science and Technology (Grant-in-Aid for Scientific Research ). References Aoki, M. (1993): Seasonal changes of chlorophyll a distribution at the northern part of Suruga Bay, central Japan. Bull. Inst. Oceanic Res. & Develop., Tokai Univ., 14, (in Japanese with English abstract). Chen, X., S. E. Lohrenz and D. A. Wiesenburg (2): Distribution and controlling mechanisms of primary production on the Louisiana-Texas continental shelf. J. Mar. Sys., 25, Hansen, H. P. and F. Koroleff (1999): Determination of nutrients. p In Methods of Seawater Analysis (third edition), ed. by K. Grasshoff, K. Kremling and M. Ehrhardt, Wiley-VCH, Weinheim. Inaba, H. (1981): Circulation pattern and current variations with report to tidal frequency in the sea near the head of Suruga Bay. J. Oceanogr. Soc. Japan, 37, Inaba, H. (1984): Current variation in the sea near the mouth of Suruga Bay. J. Oceanogr. Soc. Japan, 4, Kasai, A., S. Kimura, H. Nakata and Y. Okazaki (22): Entrainment of coastal water into a frontal eddy of the Kuroshio and its biological significance. ICES J. Mar. Sys., 37, Komatsu, K. and K. Kawasaki (22): Unique hydrographic structure of intermediate water offshore the Enshu-nada in the Kuroshio region south of Japan. Bull. Fish. Res. Agen., 5, 1 22 (in Japanese with English abstract). Levitus, S. (1982): Climatological Atlas of the World Ocean. NOAA Professional Paper 13, U.S. Governmental Print Office, Washington, D.C., 173 pp. Lohrenz, S. E., G. L. Fahnenstiel, D. G. Redalje, G. A. Lang, M. J. Dagg, T. E. Whiteledge and Q. Dortch (1999): Nutrients, irradiance, and mixing as factors regulating primary production in coastal waters impacted by the Mississippi River plume. Cont. Shelf Res., 19, Nakamura, Y. (1982): Oceanographic feature of Suruga Bay from view point of fisheries oceanography. Bull. Shizuoka Pref. Fish. Exp. Stn., 17, Special Issue, (in Japanese with English abstract). Nakamura, Y. and H. Muranaka (1979): Temporal fluctuation of oceanographic structure in the Suruga Bay and Enshunada. Bull. Japan Soc. Fish. Oceanogr., 34, (in Japanese). Pike, S. M. and S. B. Moran (1997): Use of Poretics.7 µm pore size glass fiber filter for determination of particulate organic carbon and nitrogen in seawater and freshwater. Mar. Chem., 57, Shiomoto, A. and S. Hashimoto (1999): Relationship between chlorophyll a and nutrients in Suruga Bay, central Japan May Bull. Japan Soc. Fish Oceanogr., 63, 1 7 (in Japanese with English abstract). Suzuki, R. and T. Ishimaru (199): An improved method for the determination of phytoplankton chlorophyll using N, N-dimethylformamide. J. Oceanogr. Soc. Japan, 46, Takeuchi, K. and T. Hibiya (1997): Numerical simulation of baroclinic tidal currents in Suruga Bay and Uchiura Bay using a high resolution level model. J. Oceanogr., 53, Toyota, Y. (1985): Section III chemical, Chapter 11 Suruga Bay. p In Coastal Oceanography of Japanese Islands, ed. by Coastal Oceanography Research Committee, the Oceanographic Society of Japan, Tokai University Press, Kanagawa (in Japanese). Toyota, Y., Y. Misawa and T. Nanaumi (1993): Submarine topography and nutrients distribution in Senoumi bank area, Suruga Bay. Bull. Inst. Oceanic Res. & Develop., Tokai Univ., 14, (in Japanese with English abstract). 732 T. Iwata et al.