Journal of Oceanography, Vol. 61, pp. 1011 to 1016, 2005

Spatiotemporal Decreases of Nutrients and Chlorophyll-a in the Surface Mixed Layer of the Western North Pacific from 1971 to 2000 Y UTAKA W. WATANABE1*, HIROSHI ISHIDA2, TOSHIYA NAKANO 3 and NAOKI NAGAI4 1

Graduate School of Environmental Science, Hokkaido University, Sapporo 060-0810, Japan KANSO Technos Co. Ltd., Osaka 541-0052, Japan 3 Meteorological Research Institute, Tsukuba 305-0052, Japan 4 Japan Meteorological Agency, Tokyo 100-8122, Japan 2

(Received 25 November 2004; in revised form 12 March 2005; accepted 26 March 2005)

Using time series of hydrographic data in the wintertime and summertime obtained along 137°E from 1971 to 2000, we found that the average contents of nutrients in the surface mixed layer showed linear decreasing trends of 0.001~0.004 µ mol-PO4 l–1 yr–1 and 0.01~0.04 µ mol-NO3 l–1 yr–1 with the decrease of density. The water column Chl-a (CHL) and the net community production (NCP) had also declined by 0.27~0.48 mg-Chl m–2 yr–1 and 0.08~0.47 g-C-NCP m –2 yr–1 with a clear oscillation of 20.8 ± 0.8 years. These changes showed a strong negative correlation with the Pacific Decadal Oscillation Index (PDO) with a time lag of 2 years (R = 0.89 ± 0.02). Considering the recent significant decrease of O2 over the North Pacific subsurface water, these findings suggest that the long-term decreasing trend of surface-deep water mixing has caused the decrease of marine biological activity in the surface mixed layer with a bidecadal oscillation over the western North Pacific.

1. Introduction Recent studies of global climate change have reported the possibility that recent oceanic conditions have changed due to the effect of anthropogenically induced greenhouse warming and/or natural climate change (e.g., Levitus et al., 2000; Hansen et al., 2002). In the North Pacific, some studies have already reported a linear increase of water temperature, together with decreases of oxygen and nutrients as the oceanic physical conditions have changed during the past several decades (e.g., Watanabe et al., 2001; Ono et al., 2001; Emerson et al., 2001, 2004). Moreover, some of these studies have shown that the change of physical conditions had a clear bidecadal oscillation superimposed on the linear trend (Ono et al., 2002; Bryden et al., 2003; Watanabe et al., 2003). Furthermore, some biogeochemical studies have reported a decadal change of marine biological activity, such as phytoplankton, zooplankton and chlorophyll-a (e.g., Karl et al., 2001; Limsakul et al., 2001; Gregg and Conkright, 2002; Chiba et al., 2004). Some such studies * Corresponding author. E-mail: [email protected] Copyright © The Oceanographic Society of Japan.

Keywords: ⋅ Decadal change, ⋅ biological activity, ⋅ western North Pacific.

have suggested the variation patterns in biological components in the North Pacific due to the changing physical conditions associated with climatic regime shifts in 1976/ 77, 1988/89 and 1998/99, and the regime-shift related climate indices such as the Pacific Decadal Oscillation (PDO) (e.g., Trenberth, 1990; Hare and Mantua, 2000; Minobe, 2000). Most studies were based on high frequency time-series, observations being made several times each year at fixed observation points in the ocean during the past several decades. In particular, Chiba et al. (2004) demonstrated that recent biological activity in the western subpolar region had decreased in response to the weakening of vertical water mixing (Watanabe et al., 2001; Ono et al., 2001; Emerson et al., 2001, 2004). However, since these studies were based on an inadequate spatial data set due to the limited number of fixed ocean observation points, we are still not completely able to understand whether the biological change occurred only in the subpolar region or over the North Pacific, in response to the effect of anthropogenically induced warming and/or natural climate change. In this evaluation of the extensive, long-term change of biological activity in the North Pacific, we thus focus on time series of hydrographic data in the extensive area along the 137°E line in the western North Pacific, for

1011

2. Data and Methods We have used time series data of σθ, PO 4, NO3 and Chl from 34°N to 3°N along the 137°E line during the period from 1971 to 2000 (Fig. 1(a)) (Japan Meteorological Agency, 1995–2001), which is also available from the Japan Oceanographic Data Center [http://www.jodc.go.jp/ service.htm]. Time series of σθ , PO4, NO3 and Chl were obtained at approximately one degree intervals in the wintertime (January–February) and the summertime (July–August) every year. Despite the fact that the data set is confined only to the wintertime and summertime, these time series are useful to understand the extensive, long-term biological change over a wide area because these observations cover a wide area of the western North Pacific in the past thirty years. Samples at each station were basically taken from about 25 layers above 4000 m depth. PO4, NO 3 and Chl were measured by the molybdenum photometric method, the copper-cadmium sulfanilamide reduction method and the solvent extraction photofluorometrical method (Yentsch and Menzel, 1963; Strickland and Parsons, 1968), which have been used as standard procedures during the last thirty years. Based on the deep water data sets (>27.7 σθ), we estimated the offsets in PO4 and NO3 between the data sets to be within 0.02 µmol-PO4 l–1 and within 0.17 µmol-NO 3 l–1 for the entire thirty year data set. We also estimated the offset in Chl to be within 0.01 µg-Chl l –1 at 200 m depth for the entire data because Chl data are limited above 200 m depth. We used these data without any correction for the offsets in this study. Then, assuming the depth of seasonal surface mixed layer to have a depth difference of 0.125σθ from the surface density (MLD), we estimated the averaged concentrations of PO4 and NO3 in MLD (PO4 (ave) and NO3 (ave)). We also estimated the water column Chl (CHL) by integrating Chl over the whole water column. Moreover, considering the active supply of nutrients from the deep water in the wintertime, we divided these data sets into three regions based on the distribution of climatological wintertime water density: the Kuroshio area south of Japan, which is one of the largest areas for the air-sea interaction (Kawabe, 1995) (“KU”, 34°N–30°N); the subtropical area (“ST”, 30°N–15°N); and the tropical area (“TR”, 15°N–3°N) (Fig. 1(b)). We here address an average value of each parameter, expressed as three-year running mean composites with standard errors (SE, ±1σ). In addition, to clarify the decadal changes of parameters that we used in this study, we applied the Fourier

1012 Y. W. Watanabe et al.

Latitude(N)

60

100

120 Longitude(E) 140

160

180

(a)

40

Kuroshio area (KU) Subtropical area (ST)

20

Tropical area (TR) 0 0

Depth(m)

which data have been collected during the last thirty years. In particular, we address water density (σ θ), phosphate (PO 4), nitrate (NO3), chlorophyll-a (Chl) and the net community production (NCP) as indices of marine biological activity in the surface mixed layer.

(b) 23.0

100

24.0 25.0

200

26.0 26.0

300

(TR)

(ST) 10

Latitude(N)

20

(KU) 30

Fig. 1. a) Sampling points along the 137°E line. b) Distribution of average wintertime water density during the period from 1971 to 2000. “KU”, “ST” and “TR” are the Kuroshio area, the subtropical area and the tropical area, respectively.

sine expansion to the time series data sets of these parameters in the past thirty years: X = –a·y + b + c· sin{2 π (y – d)/e}, where “X” refers to one parameter within σθ, PO4, NO3, CHL and NCP. “y” is the calendar year. “a”, “b”, “c”, “d” and “e” are constants. That is, X = long term linear trend component + decadal oscillation component. 3. Results and Discussion 3.1 Temporal changes of σ θ, MLD, PO4 and NO3 σ θ in MLD generally decreased southward from KU to TR, and from the wintertime to the summertime. σθ in all areas had a linear decreasing trend of 0.007 ± 0002 yr –1 in the wintertime (p < 0.05) and 0.012 ± 0.002 yr–1 in the summertime (p < 0.05), despite difficulties in determining the periodicity (Figs. 2(a) and (b)). The main cause of the wintertime change was possibly the increase of water temperature in KU and ST (p < 0.01) and the decrease of salinity in TR (p < 0.01), while that in the summertime was due to the increase of water temperature in all areas (p < 0.01) (data not shown), which agrees with previous studies (e.g., Yasuda and Hanawa, 1997; Michael and Dongxiano, 2002). On the other hand, despite the decrease of σθ, MLD data in the wintertime did not change significantly in all areas; 123 ± 3 m (KU), 84 ± 1 m (ST) and 45 ± 1 m (TR) (p > 0.10) (data not shown). Similarly, MLD in the summertime did not change, except for KU (–1.2 m yr–1, 101 ± 4 m, p < 0.01), 72 ± 1 m (ST) and 54 ± 1 m (TR) (p > 0.10) (data not shown).

CHL (mg m-2)

24

σθ

60

(b)

23

60

(a)

CHL (mg m-2)

(a)

25

40

20

(b)

40

20

22 0

0.2

0.1

NO3 (ave) (µmol l-1)

(e)

(f)

3.0

2.0

1.0

0.0 '70

'80 year '90 '00 '70 Wintertime (Jan.-Feb.)

'80 year '90 '00 Summertime (Jul.-Aug.)

Fig. 2. Time series of σθ, PO4 (ave) (µ mol l–1), NO3 (ave) (µmol l–1) in MLD in the wintertime and summertime along the 137°E line. Average values are shown as three-year running mean composites with standard errors in three regions (KU: black open circle, ST: red open square, TR: blue open triangle). Besides the linear regression line (solid line), the non-linear fitting curve is also shown (dash curve), estimated by the Fourier sine expansion. In the case of no significant periodicity, we do not show the non-linear fitting curve. a) σθ in the wintertime. The fitted linear equations are as follows; KU = –0.011 yr + 46.752 (p < 0.01); ST = –0.003 yr + 30.018 (p < 0.05); TR = –0.006 yr + 33.362 (p < 0.05). b) σ θ in the summertime. KU = –0.009 yr + 39.788 (p < 0.10); ST = –0.015 yr + 50.899 (p < 0.01); TR = –0.013 yr + 47.196 (p < 0.01). c) PO4 (ave) in the wintertime. KU = –0.002 yr + 4.223 (p < 0.05); ST = –0.002 yr + 3.441 (p < 0.01); TR = –0.004 yr + 7.099 (p < 0.01). The averaged periodicity (AP) = 20.4 ± 1.5 yr (p < 0.01). d) PO 4 (ave) in the summertime. KU = –0.001 yr + 1.159 (p < 0.05); ST = –0.001 yr + 2.380 (p < 0.01); TR = –0.003 yr + 5.955 (p < 0.01). AP = 21.5 ± 0.8 yr (p < 0.01). e) NO3 (ave) in the wintertime. KU = –0.04 yr + 83.53 (p < 0.01); ST = –0.01 yr + 24.78 (p < 0.01); TR = –0.01 yr + 14.96 (p < 0.01). f) NO3 (ave) in the summertime. KU = –0.01 yr + 4.97 (p < 0.05); ST = –0.01 yr + 10.56 (p < 0.01); TR = –0.01 yr + 14.34 (p < 0.01).

(d)

40

1

20

0

0.0 4.0

0

(c)

(d)

PDO

(c)

NCP (g-C m-2 yr-1)

PO4 (ave)(µmol l-1)

21 0.3

0

'70

'80 year '90

'00

-1 '70

'80 year '90

'00

Fig. 3. Time series of CHL (mg m –2), NCP (gC m–2 yr–1) in MLD along the 137°E line, and PDO in the wintertime (KU: black open circle, ST: red open square, TR: blue open triangle). a) CHL in the wintertime. KU = –0.08 yr + 199.80 (p > 0.10); ST = –0.37 yr + 767.37 (p < 0.01); TR = –0.07 yr + 164.43 (p > 0.10). AP = 21.4 ± 1.2 yr (p < 0.01). b) CHL in the summertime. KU = –0.48 yr + 994.41 (p < 0.01); ST = –0.40 yr + 821.38 (p < 0.01); TR = –0.27 yr + 573.91 (p < 0.05). AP = 20.4 ± 0.6 yr (p < 0.01). c) NCP estimated from the difference in PO4 (ave) in the two seasons. KU = –0.47 yr + 964.91 (p < 0.05); ST = –0.26 yr + 528.55 (p < 0.01); TR = –0.08 yr + 169.08 (p < 0.05). AP = 20.4 ± 1.5 yr (p < 0.01). d) PDO in the wintertime. Data cited from Minobe (2000). AP = 20.3 yr (p < 0.01).

Both PO4 (ave) and NO3 (ave) decreased in the order KU, TR and ST. These levels in the wintertime were higher than those in the summertime due to the difference in the extent of vertical water mixing and/or diffusion with the deep water (Figs. 2(c)–(f)). We found significant linear declines of 0.001~0.004 µmol-PO 4 (ave) l –1 yr–1 (p < 0.05) and 0.01~0.04 µmol-NO3 (ave) l–1 yr–1 (p < 0.05) in MLD in all areas and seasons. Both the linear decreasing trends of PO4 and NO 3 in the wintertime were larger than those in the summertime (p < 0.05). The ratio of NO3 (ave) to PO 4 (ave) was an average value of 5, which is lower than the stoichiometric ratio of NO3 to PO 4 in the North Pacific (Anderson and Sarmiento, 1994). The main reason for this was possibly due to the large offset of NO3 within 0.17 µmol l–1, or due to the existence of nitrogen fixing organisms in these regions (Saino, 1977). Although we detected no drastic change of NO3 (ave) for the two reasons given above, we also found that PO4 (ave)

Spatiotemporal Decreases of Nutrients and Chlorophyll-a in the Surface Mixed Layer of the Western North Pacific from 1971 to 2000 1013

changed steeply at the end of 1970s, the end of 1980s and the end of 1990s, which agrees well with climate regime shifts reported in previous studies (e.g., Minobe, 2000). In addition, PO4 (ave) showed a negative correlation of more than 0.5 with the El Niño events (e.g., Minobe, 2000), while there was no significant relationship between PO 4 (ave) and the large meander of the Kuroshio (Kawabe, 1995) (data not shown). The above findings thus suggest that the sea surface water warming including the El Niño event has caused the long-term decreasing trend of σ θ in MLD over a wide area, and consequently the weakening of sea surface-deep mixing has led to a decrease in the supply of nutrients from the deep water. 3.2 Temporal changes of Chl and NCP Moreover, we found that the changes of CHL in the two seasons had almost the same pattern, with a linear decreasing rates of 0.27~0.48 mg-Chl m–2 yr–1 in the summertime (p < 0.05) and no significant decrease in the wintertime (Figs. 3(a) and (b)). On the other hand, Limsakul et al. (2001) showed that CHL increased in the springtime in the North Pacific subtropical region due to the delay of phytoplankton blooming, although CHL in the wintertime was the same as in our results. Based on only these CHL results, we cannot clarify whether the biological activity had changed over the long-term or not. Thus, using the difference of nutrient in MLD from the wintertime to the summertime and the stoichiometric ratio of carbon to phosphate or nitrate (R C/PO4 = 106, RC/NO3 = 6.6) (Redfield et al., 1963), we here estimated the net community production [NCP = (R·N(ave)·MLD)winter – (R·N(ave)·MLD)summer] as new biological productivity in the mixed layer. “N” is either PO4 or NO3, and “R” is the stoichiometiric ratio of carbon to the nutrient. Based on PO 4, we found that NCP in all areas had a linear decreasing trend of 0.08~0.47 g-C-NCP m–2 yr–1 (p < 0.05) with steep changes at the ends of the 1970s, 1980s and 1990s, and that NCP decreased southward from KU to TR (Fig. 3(c)). The changes of NCP had almost the same pattern as those of CHL. In addition, estimating NCP based on the difference of NO3, we also found that NCP in all areas had a linear decreasing trend of 0.03~0.51 g-C-NCP m–2 yr –1 (p < 0.05) (data not shown), which almost agrees with the NCP estimated from the difference of PO4. In the wintertime, the influence of the recent long-term decreasing trend in nutrient supply on the biological activity may be still small due to sufficient supply of nutrient derived from the active winter vertical mixing. On the other hand, the decreases of summertime CHL and NCP suggests that the marine biological activity at least in NCP over the western North Pacific, could weaken over the long-term due to a decrease in the supply of nutrients.

1014 Y. W. Watanabe et al.

Recent studies have reported a significant decrease of 0.5 µmol-O2 kg–1 yr–1 over the past forty years below MLD in the wintertime over the North Pacific (Watanabe et al., 2001, 2003; Emerson et al., 2001, 2004). This suggests that the main cause was either the increase of export flux from MLD or the weakening of sea surface-deep water mixing. If the biological activity only increased in the past several decades, NCP has to become two to five times greater than the value in the past several decades to explain the significant decrease of O2 over the North Pacific. In this study, however, we found that NCP has decreased by a few percent per year (Fig. 3(c)) and not has increased by two to five times, indicating reliable evidence of the weakening of the surface-deep water mixing as the long-term trend, at least in this extensive area of the western North Pacific. 3.3 Decadal periodicity of PO4, CHL and NCP In addition, we tried to clarify whether the changes of these parameters as an index of biological activity related to the climatic regime shift, and whether the changes had decadal oscillations with long-term trends. Ono et al. (2001) and Watanabe et al. (2003) found that the changes of nutrients and oxygen had a clear bidecadal oscillation superimposed on the linear trend over the western North Pacific subpolar subsurface water in the past forty years. It is also possible that the biological change in the surface water of the subtropical and tropical regions has a bidecadal periodicity due to the weakening of the surface-deep water mixing in a wide area. We therefore tried to evaluate whether the changes of PO4, CHL and NCP related to the climatic regime shift, and whether these changes had decadal oscillations. To elucidate the regime-shift in the North Pacific, Mantua et al. (1997) and Minobe (2000) discussed the Pacific Decadal Oscillation Index (PDO) as an index of the anomaly of hydrographic conditions in the North Pacific. In particular, Minobe (2000) reported that PDO had a clear decadal oscillation, and that a combination of the bidecadal and the pentadecadal oscillations in PDO caused the large regime-shift in the North Pacific (Fig. 3(d)). To clarify the decadal oscillations of PO 4, CHL and NCP, we here applied the Fourier sine expansion to the time series datasets of these parameters in the past thirty years (see the Data and Methods section). Although we only had the time series data for the past thirty years, we found that PO4 (ave), CHL and NCP in all areas and seasons had a clear bidecadal periodicity of 20 years (p < 0.01 for all parameters, average = 20.8 ± 0.6 yr) superimposed on the above decreasing linear trends (Figs. 2(c) and (d), 3(a)–(c)). These periodicity almost agreed with those of nutrients and oxygen in the western North Pacific subpolar region (Ono et al., 2001; Watanabe et al., 2003), suggesting that the change of oceanic conditions

occurred with a bidecadal oscillation in a wide area of the western North Pacific. Considering the active supply of nutrients in the wintertime, we compared the changes of PO4 (ave), CHL and NCP with the change of PDO in the wintertime (Fig. 3(d)) (Minobe, 2000). All the periodicity of these parameters had a good negative correlation with that of PDO (correlation coefficient (R): 0.67 ± 0.04 for PO4 (ave) (winter); 0.73 ± 0.07 for PO4 (ave) (summer); 0.74 ± 0.10 for CHL(winter); 0.78 ± 0.04 for CHL(summer); 0.76 ± 0.11 for NCP). We also found that these changes had a time lag of about two years for PDO, although it is difficult to explain the occurrence of the time lag (Fig. 3). One possibility is that our observational field is located in the westernmost North Pacific, and this region may have a two-to-five year time lag from the canonical pattern of PDO due to the Rossby waves (Miller et al., 2004). Thus the changes of these parameters may not completely agree with that of PDO. If we here correct for a time lag of 2 years for PO4, CHL and NCP, we find that all the changes of these parameters show stronger negative correlations with that of PDO (R: 0.88 ± 0.03 for PO 4 (ave) (winter) ; 0.83 ± 0.07 for PO4 (ave) (summer); 0.89 ± 0.01 for CHL(winter); 0.95 ± 0.02 for CHL(summer); 0.89 ± 0.07 for NCP). In general, when PDO in the wintertime shows positive signal, the Aleutian Low is intensified. The intensification could enhance the wintertime East Asian Monsoon with westerly winds. This change would cause the winter surface layer mixing to be extensive in the western North Pacific (although our data did not show it clearly), and the Kuroshio current to be faster (Miller et al., 2004). Therefore, the weakening of mixed layer stability would restrain the biological activity in the western North Pacific. However, in this study, despite the strengthening of mixed layer enhancing the supply of nutrient from deep water, we actually found that the nutrient contents declined with the positive PDO signals, which is consistent with previous studies (Limsakul et al., 2001). In the western North Pacific south of Japan, the nutrient contents in the mixed layer is always higher than that in the central North Pacific region, based on the climatological hydrographic data set of Levitus (1994). Because the strengthening of the Kuroshio current enhances the horizontal mixing in this region, the effect of nutrient dilution due to the horizontal mixing may outweigh the effect of nutrient supply from deep water. Consequently, it is possible that all changes of PO4, CHL and NCP in this region will show strong negative correlations with PDO in the western North Pacific. Considering the recent significant decrease of O 2 over the North Pacific subsurface water as a long-term trend (e.g., Emerson et al., 2004), our results allow us to conclude that the recent long-term decreasing trend of

sea surface-deep water mixing had caused the decrease of σθ in MLD, the declines of the nutrient supply and the biological activity over a wide area of the western North Pacific. Simultaneously, the decadal change of air-sea interaction in PDO has led to bidecadal changes of nutrients and biological activity. Unfortunately, our present data mean that it is difficult to separate exactly the effect of anthropogenically induced warming and the effect of natural climate change on the biological activity because there may possibly be more long-term natural climate change than that on a bidecadal scale (e.g., Minobe, 2000). In addition, we cannot discover whether any change of marine biological species occurs or not, because we have not accumulated the time series data of plankton biomass or HPLC over a wide area. It is therefore necessary to maintain the time series of hydrographic observations with biological parameters over the North Pacific through the present and into the future. Acknowledgements We would like to express our gratitude to the many scientists and technicians who measured the hydrographic data along 137°E for their dedicated work of long-term observation. We also extend our profound thanks to two anonymous reviewers for their many fruitful comments. References Anderson, L. A. and J. L. Sarmiento (1994): Redfield ratios of remineralization determined by nutrient data analysis. Global Biogeochem. Cycles, 8, 65–80. Bryden, H. L., E. L. McDonagh and B. A. King (2003): Changes in ocean water mass properties: oscillations or trends? Science, 300, 2086–2088. Chiba, S., T. Ono, K. Tadokoro, T. Midorikawa and T. Saino (2004): Increased straritification and decreased lower trophic level productivity in the Oyashio region of the North Pacific: a 30-year retrospective study. J. Oceanogr., 60, 149–162. Emerson, S., S. Mecking and J. Abell (2001): The biological pump in the subtropical North Pacific: Nutrient sources, Redfield ratios, and recent changes. Global Biogeochem. Cycles, 15, 535–554. Emerson, S., Y. W. Watanabe, T. Ono and S. Mecking (2004): Temporal trends in apparent oxygen utilization in the upper pycnocline of the North Pacific. J. Oceanogr., 60, 139–148. Gregg, W. W. and M. E. Conkright (2002): Decadal changes in the global ocean chlorophyll. Geophys. Res. Lett., 29, 10.1029/2002GL014689. Hansen, J., R. Ruedy, M. Sato and K. Lo (2002): Global warming continues. Science, 295, 275. Hare, S. R. and N. Mantua (2000): Empirical evidence for North Pacific regime shift in 1977 and 1989. Prog. Oceanogr., 47, 103–145. Japan Meteorological Agency (1995–2001): Data Report of Oceanographic Observations, No. S1, 84–92, CD-ROM. Karl, D. M., R. R. Bidagre and R. M. Letelier (2001): Long-

Spatiotemporal Decreases of Nutrients and Chlorophyll-a in the Surface Mixed Layer of the Western North Pacific from 1971 to 2000 1015

term changes in plankton community structure and productivity in the North Pacific Subtropical Gyre: the domain shift hypothesis. Deep-Sea Res., 48, 1449–1470. Kawabe, M. (1995): Variations of current path, velocity and volume transport of the Kuroshio in relation with the large meander. J. Phys. Oceanogr., 25, 3103–3117. Levitus, S. (1994): World Ocean Atlas 1994, National Oceanographic Data Center, CD-ROM products. Levitus, S., J. I. Antonov, T. P. Boyer and C. Stephens (2000): Warming of the world ocean. Science, 287, 2225–2229. Limsakul, A., T. Saino, T. Midorikawa and J. I. Goes (2001): Temporal variations in low trophic level biological environments in the northwestern North Pacific Subtropical Gyre from 1950 to 1997. Prog. Oceanogr., 49, 129–149. Mantua, S., S. R. Hare, Y. Zang, J. M. Wallace and R. C. Francis (1997): Pacific interdecadal climate oscillation with impacts on salmon production. Bull. Amer. Meteor. Soc., 76, 1069– 1079. Michael, M. J. and Z. Dongxiano (2002): Slowdown of the meridional overturning circulation in the upper Pacific Ocean. Nature, 415, 603–608. Miller, A. J., F. Chai, S. Chiba, J. R. Moisan and D. J. Neilson (2004): Decadal-scale climate and ecosystem interactions in the North Pacific Ocean. J. Oceanogr., 60, 163–188. Minobe, S. (2000): Resonance in bidecadal and pentadecadal climate oscillations over the North Pacific: role in climate regime shifts. Prog. Oceanogr., 47, 381–408. Ono, T., T. Midorikawa, Y. W. Watanabe, K. Tadokoro and T. Saino (2001): Temporal increase of phosphate and apparent oxygen utilization in the subsurface waters of western subarctic Pacific from 1968 and 1998. Geophys. Res. Lett., 28, 3285–3288. Ono, T., K. Tadokoro, T. Midorikawa, J. Nishioka and T. Saino

1016 Y. W. Watanabe et al.

(2002): Multiple-decadal decrease of net community producition in western subarctic North Pacific. Geophys. Res. Lett., 29, 10.1029/2001GL014332. Redfield, A. C., B. H. Ketchum and F. A. Richards (1963): The influence of organisms on the composition of sea-water. In The Sea, ed. by M. N. Hill, Wiley-Interscience, Vol. 2, p. 26–77. Saino, T. (1977): Biological nitrogen fixation in the ocean with emphasis on the nitrogen fixing blue-green alga Trichodesmium and its significance in the nitrogen cycling in the low latitude sea areas. Ph.D. Thesis, Ocean Research Institute, University of Tokyo, Japan, 153 pp. Strickland, J. D. H. and T. R. Parsons (1968): A practical handbook of seawater analysis. Fish. Res. Bd. Canada, Bull., 167, 65–75. Trenberth, K. E. (1990): Recent observed interdecadal climate change in the Northern Hemisphere. Bull. Amer. Meteor. Soc., 71, 988–993. Watanabe, Y. W., T. Ono, A. Shimamoto, T. Sugimoto, M. Wakita and S. Watanabe (2001): Possibility of a reduction in the formation rate of the subsurface water in the North Pacific. Geophys. Res. Lett., 28, 3285–3288. Watanabe, Y. W., T. Ono, M. Wakita, N. Maeda and T. Gamo (2003): Synchronous bidecadal periodic changes of oxygen, phosphate and temperature between the Japan Sea deep water and the North Pacific intermediate water. Geophys. Res. Lett., 30(24), 2273, doi:10.1029/2003GL018338. Yasuda, T. and K. Hanawa (1997): Decadal changes in the mode water in the midlatitude North Pacific. J. Phys. Oceanogr., 27, 858–870. Yentsch, C. S. and D. W. Menzel (1963): A method for the determination of phytoplankton chlorophyll and phaeophytin by fluorescence. Deep-Sea Res., 10, 221–231.