Helium Isotopes of Seawater in the Philippine Sea and the Western North Pacific

Geochemical Journal, Vol. 44, pp. 451 to 460, 2010

Helium isotopes of seawater in the Philippine Sea and the western North Pacific

NAOTO TAKAHATA,*# TAICHI TOKUKAKE,** KOTARO SHIRAI,† SHINZOU FUJIO,# KIYOSHI TANAKA# and YUJI SANO#

Center for Advanced Marine Research, Ocean Research Institute, The University of Tokyo, Nakano-ku, Tokyo 164-8639, Japan

(Received June 4, 2009; Accepted June 4, 2010)

We measured helium isotopic ratios of 84 seawater samples from various depths collected in the western North Pacific Ocean and the western Philippine Sea. The 3He/4He ratios varied significantly from δ3He of 0.1% to 22.9%, where δ3He is defined as the percent deviation of the helium isotopic ratio relative to the atmospheric standard. Maximum δ3He > 20% was observed at mid-depth (2000–2500 m) in the western Philippine Sea and in the southern part (~10°N) of the western North Pacific, though not in the northern part (~30°N) at the same depth. Contour maps of the lateral δ3He distribution at mid-depth suggest that the helium-3 plume derived from the East Pacific Rise does not flow northward along the Izu– Ogasawara–Mariana Ridge but westward through the Caroline Basin or the Yap–Mariana Junction into the Philippine Sea. It then flows northward in the western Philippine Sea to a region adjacent to the Japanese Islands. Although these flows inferred from the δ3He distribution are roughly similar to those estimated from water properties such as isopycnal distributions, the δ3He distribution could reveal that deep-water circulation seems to be different at each depth (2000, 2500, 3000 m).

Keywords: 3He/4He ratios, abyssal current, Philippine Sea, North Pacific, seawater

a dissolved oxygen sensor (CTDO ). Most of it then flows INTRODUCTION 2 into the West Caroline Basin. The remaining western The Philippine Sea, at the western end of the North boundary current flows over the middle and lower Solo- Pacific Ocean, consists mainly of three basins, the mon Rise, and then proceeds westward, where it is di- Shikoku, the West Mariana, and the Philippine basins. It vided by the Caroline Seamounts into southern and north- is mostly isolated from the main North Pacific below 2500 ern branches. The northern branch current enters the West m depth by the Izu–Ogasawara, the Mariana and the Yap Mariana Basin through the Yap–Mariana Junction. To ridges (Fig. 1). This topographic barrier restricts the hori- understand deep circulation more precisely at mid-depth zontal exchange of abyssal water, and the inflow of deep (2000–3000 m) in the western North Pacific Ocean, how- water is possible only through some narrow gaps. ever, it is important to investigate water mass structure Isopycnal maps constructed by Reid (1997) suggest that both chemically and physically. Especially at mid-depth deep seawater from the South Pacific enters the Philip- in the North Pacific Ocean, it is difficult to observe deep- pine Sea along its western boundary. Kawabe et al. (2003) sea current due to weak flow rate compared with subsur- reported that a deep western boundary current at 2000– face waters or bottom waters. It is also difficult to distin- 3000 m depth may flow from the Melanesian Basin, guish water masses due to small difference of water prop- change direction and flow around the upper Solomon Rise erties such as temperature and salinity. As we discuss to the southwestern, and proceed into the East Caroline below, the isotopic ratio of helium dissolved in seawater Basin, as suggested by hydrographic data observed with is at the maximum at mid-depth rather than in bottom a conductivity-temperature-depth profiler equipped with waters in the Pacific, which is derived from the crest of the East Pacific Rise. So excess 3He can be used for trac- ing movement and mixing of different water masses in such regions. *Corresponding author (e-mail: [email protected]) The 3He/4He ratio of the atmosphere is 1.386 × 10–6, #Present address: Atmosphere and Ocean Research Institute, The Uni- versity of Tokyo, Kashiwa, Chiba 277-8564, Japan. and it is considered to be constant on a global scale within 3 4 **Present address: Tokyo Research Laboratory, Mitsubishi Gas Chemi- an experimental error of 5%. The He/ He ratios of cal Co. Inc., Katsushika-ku, Tokyo 125-0061, Japan. mantle-derived samples, such as those from mid-ocean †Present address: Department of Earth and Planetary Science, School ridge basalts and volcanic gases in island arcs, are rela- of Science, The University of Tokyo, Bunkyo-ku, Tokyo 113-0033, tively high at about 1 × 10–5, whereas those of granitic Japan. rocks and continental natural gases are low, with ratios Copyright © 2010 by The Geochemical Society of Japan. of around 1 × 10–7 (Lupton, 1983; Mamyrin and

451 (a)

IOR NPB SB North Pacific Ocean

Philippine Sea MR PB WMB EMB YR CS WCB ECB MB SR

(b)

9 10 1 2 12 11

8 7

6 5 3 4

Fig. 1. (a) Map of a portion of the Philippine Sea and the western North Pacific Ocean with place names. Depth of the ocean is contoured with the 2500 m, 4000 m isobaths, and depths shallower than 4000 m are shaded. IOR, Izu–Ogasawara Ridge; MR, Mariana Ridge; YR, Yap Ridge; SR, Solomon Rise; CS, Caroline Seamounts; SB, Shikoku Basin; PB, Philippine Basin; WMB, West Mariana Basin; EMB, East Mariana Basin; WCB, West Caroline Basin; ECB, East Caroline Basin; MB, Melanesian Basin; NPB, Northwest Pacific Basin. (b) Sampling sites of seawaters in the western North Pacific (ST-1 to -5 and -9 to -12) and the western Philippine Sea (ST-6 to -8).

452 N. Takahata et al. Table 1. Location and bottom depth of sampling stations together with sampling depth of western North Pacific water and western Philippine Sea water

Cruise Station Location Bottom depth (m) Sampling depth (m) KH-04-4 ST-1 32°30′ N, 143°09′ E 5612 397, 596, 991, 1485, 1977, 2469, 4423, 5393 ST-2 32°30′ N, 150°00′ E 5575 2468 ST-3 12°52′ N, 147°13′ E 5508 2471 ST-4 12°40′ N, 155°00′ E 5869 2470 ST-5 14°34′ N, 170°54′ E 5655 596, 792, 992, 1486, 1979, 2471, 2962, 4427, 5647 KH-06-2 ST-6 15°00′ N, 128°00′ E 5807 199, 496, 744, 991, 1485, 1978, 2469, 2959, 3449, 3939 ST-7 20°00′ N, 128°00′ E 5660 199, 496, 745, 992, 1485, 1976, 2470, 2961, 3451, 3937 ST-8 25°00′ N, 128°00′ E 7020 198, 496, 744, 992, 1486, 1979, 2471, 2961, 3451, 3941 KH-07-1 ST-9 37°59′ N, 150°00′ E 5921 544, 989, 1483, 1975, 2466, 2954, 3443, 3930, 4417, 4902, 5387 ST-10 38°00′ N, 157°00′ E 5717 542, 990, 1236, 1728, 1975, 2467, 2955, 3444, 3932, 4418, 4904, 5390 ST-11 32°29′ N, 166°01′ E 6224 594, 988, 1484, 1975, 2463, 2959, 3446, 4422, 4908, 5876 ST-12 32°30′ N, 160°01′ E 4631 542, 989, 1479, 1978, 2465, 3447, 3936, 4423

Tolstikhin, 1984; Sano and Wakita, 1985). The 3He/4He 2004; KH-07-1, May 2007) and in the Philippine Sea (KH- ratio is one of the most sensitive and conservative tracers 06-2, June 2006). Stations visited were ST-6 to -8 (west- in chemical oceanography (e.g., Jenkins et al., 1972; Craig ern Philippine Sea, along 128°E longitude), ST-1 and -2 et al., 1975; Sano et al., 1995) owing to the primordial and ST-9 to -12 (western North Pacific near the Japanese signature, rapid mobility, and chemical inertness of the Islands, ~30°N latitude), and ST-3 to -5 (western North isotopes. During the South Tow expedition in 1972, Pacific near the Mariana Arc, ~10°N). Figure 1 shows Lupton and Craig (1981) discovered a striking intensity the sampling points used in this study. Samples for he- and lateral extent of excess 3He relative to air-saturated lium isotope measurements were collected from 8–12 seawater in the deep Pacific Ocean at latitude 15°S on depths at each station with a CTD carousel system the East Pacific Rise. This plume-shaped 3He anomaly, equipped with 10-L Niskin bottles. Details are shown in which originated from volcanic activity on the ridge, Table 1. Seawater was transferred without exposure to spread westward by abyssal currents at the depth of the the atmosphere from the Niskin bottles into containers of crest. Since this expedition, more than 5000 3He/4He about 30 cm3 made of copper tubing for storage (Sano et measurements have been carried out in three oceans. A al., 1989). thorough understanding of deep-sea circulation in the In the laboratory, each 30-cm3 copper container was Philippine Sea and adjacent North Pacific is hindered by connected to a stainless steel high vacuum line and dis- sparse helium isotopic data in seawater samples from these solved gases were extracted from the seawater samples regions (Östlund et al., 1987; Igarashi et al., 1987; Sano in vacuo. Helium in the exsolved gases was purified with et al., 2004; Takahata et al., 2004). Sano et al. (2004) hot titanium-zirconium getters and charcoal traps held at and Takahata et al. (2004) have reported excess 3He of liquid nitrogen temperature. The 4He/20Ne ratios were more than 20% at mid-depth (2000–3000 m) in the north- measured by an inline quadrupole mass spectrometer. ern Philippine Sea, in contrast to profiles of 3He/4He ra- Helium was then separated from neon by a cryogenic tios in the western North Pacific. This 3He anomaly is charcoal trap held at 40 K, because varying He/Ne ratio due to contributions from mantle helium degassed from in terrestrial samples can affect the measured 3He/4He submarine volcanoes. To verify the source of this mantle ratios by up to 10% (Rison and Craig, 1983; Sano and helium, we have measured the distribution of 3He/4He Wakita, 1988). The 3He/4He ratios were measured with ratios in seawater collected from the western Philippine one of two conventional noble-gas mass spectrometers Sea and the western North Pacific Ocean, and estimated (VG5400, MicroMass Co. or Helix-SFT, GV Instrument). the deep-sea circulation at mid-depth in the area. Ion beams of 3He and 4He were measured by a double collector system. A resolving power of >550 at 1% of peak height was used for complete separation of the 3He+ beam EXPERIMENTAL METHOD + + 3 4 from those of H3 and HD . The observed He/ He ratios Seawater sampling was conducted on three cruises of of the samples were calibrated against atmospheric he- the research vessel Hakuho Maru of the Japan Agency of lium collected in Tokyo, Japan. Experimental error of the Marine-Earth Science and Technology (JAMSTEC) in the 3He/4He ratio was 1–2% (1σ), estimated by repeated western North Pacific (KH-04-4, September–October measurements of standard air containing concentrations

Helium isotopes of seawater in the NW Pacific 453 Table 2. Potential temperature, salinity and helium isotope ratio of western North Pacific water and western Philippine Sea water

Cruise Station Depth (m) P. temperature (°C) Salinity (psu) δ3He (%)

KH-04-4 ST-1 397 16.24 34.704 4.5 western North Pacific 596 11.93 34.298 11.0 991 4.50 34.169 15.0 1485 2.76 34.440 17.5 1977 2.03 34.562 18.4 2469 1.63 34.627 17.7 4423 1.10 34.687 14.8 5393 1.06 34.692 15.9

ST-2 2468 1.59 34.627 17.9

ST-3 2471 1.60 34.655 20.7

ST-4 2470 1.64 34.658 22.1

ST-5 596 5.87 34.380 — 792 4.90 34.499 12.0 992 4.06 34.540 18.1 1486 2.58 34.594 20.9 1979 1.93 34.636 21.8 2471 1.62 34.658 22.9 2962 1.39 34.671 19.7 4427 0.99 34.697 14.8 5647 0.88 34.704 12.2

KH-06-2 ST-6 199 20.75 35.000 1.4 western Philippine Sea 496 8.20 34.355 7.0 744 5.46 34.479 11.4 991 4.13 34.528 16.5 1485 2.76 34.585 — 1978 2.11 34.621 — 2469 1.68 34.647 22.9 2959 1.46 34.662 20.2 3449 1.31 34.670 — 3939 1.25 34.676 18.8

ST-7 199 18.97 34.886 0.1 496 9.59 34.276 3.3 745 5.44 34.285 12.1 992 3.80 34.426 13.5 1485 2.56 34.570 17.8 1976 1.99 34.620 22.4 2470 1.63 34.643 21.2 2961 1.42 34.658 17.7 3451 1.31 34.671 18.3 3937 1.25 34.677 16.2

ST-8 198 20.51 34.877 0.4 496 14.16 34.519 1.6 744 7.45 34.199 10.5 992 4.25 34.304 15.6 1486 2.67 34.514 18.6 1979 1.98 34.606 21.8 2471 1.62 34.644 22.1 2961 1.43 34.665 17.7 3451 1.32 34.676 — 3941 1.27 34.681 16.1

454 N. Takahata et al. Table 2. (continued)

Cruise Station Depth (m) P. temperature (°C) Salinity (psu) δ3He (%)

KH-07-1 ST-9 544 4.90 34.026 11.2 western North Pacific 989 3.25 34.362 16.6 1483 2.33 34.510 17.9 1975 1.82 34.593 18.3 2466 1.56 34.633 17.7 2954 1.38 34.657 17.7 3443 1.26 34.672 17.3 3930 1.18 34.681 16.4 4417 1.12 34.686 15.8 4902 1.09 34.690 15.3 5387 1.06 34.692 15.6

ST-10 542 4.57 34.053 8.7 990 3.16 34.355 15.5 1236 2.61 34.440 17.0 1728 2.02 34.557 18.3 1975 1.82 34.592 18.1 2467 1.52 34.638 — 2955 1.34 34.663 17.5 3444 1.22 34.676 16.6 3932 1.15 34.683 16.6 4418 1.10 34.688 16.4 4904 1.07 34.691 16.2 5390 1.05 34.693 16.5

ST-11 594 8.92 34.135 6.2 988 3.94 34.242 13.9 1484 2.51 34.475 17.6 1975 1.90 34.587 18.6 2463 1.56 34.635 18.5 2959 1.38 34.660 18.8 3446 1.27 34.673 16.9 4422 1.12 34.686 15.4 4908 1.08 34.691 14.5 5876 1.01 34.696 13.5

ST-12 542 10.42 34.277 9.3 989 3.79 34.279 16.4 1479 2.47 34.482 17.3 1978 1.85 34.587 18.4 2465 1.51 34.638 18.3 3447 1.21 34.676 17.9 3936 1.12 34.685 15.4 4423 1.10 34.688 —

equivalent to those of the samples. As an additional check RESULTS AND DISCUSSION on the accuracy of the analytical system, we analyzed air- equilibrated seawater samples taken from a water reser- 3He/4He ratios in the western North Pacific voir in a thermostatically controlled room. The observed Observed 3He/4He ratios are listed in Table 2 in δ3He data agreed well with values reported in the literature notation, that is, the deviation of the helium isotopic ra- (Kipfer et al., 2002) within the experimental error. The tio relative to the atmospheric standard: helium blank level (the same experimental procedure δ3 3 4 3 4 × without the sample) was less than 5% of the level of the He(%) = ([ He/ He]seawater/[ He/ He]air – 1) 100. (1) samples and the blank 3He/4He ratio was atmospheric within analytical error, and therefore negligibly affected Observed potential temperature and salinity are also listed the calibrated ratios. in Table 2. Potential temperature is defined as the tem-

Helium isotopes of seawater in the NW Pacific 455 0 0

1000 1000

2000 2000

3000 3000 Depth (m) Depth (m) 4000 4000

5000 5000

error error 6000 6000 0 5 10 15 20 25 0 5 10 15 20 25 3He (%) 3He (%) Fig. 2. Vertical profiles of excess 3He at sampling sites in the Fig. 3. Vertical profiles of excess 3He at sampling sites in the western North Pacific. Maximum excesses 3He of about 18% western Philippine Sea. Maximum δ3He value of >20% are are observed at mid-depth (2000–3000 m) in northern part (ex- observed at 2000–2500 m depth, which are higher than those cept for ST-5), which are significantly smaller than that ob- in northern part (~30°N) of the western North Pacific. Rela- served in southern part (ST-5). The δ3He values below 1000 m tively smaller increase of δ3He from the surface to 500 m depth depth are not so different except for ST-5. in northern samples (ST-8) suggests that the shallow water mixing layer becomes thicker toward northern area in the western Philippine Sea. perature of a water parcel at the sea surface after it has been raised adiabatically from some depth in the ocean. 3He/4He ratios in the western Philippine Sea The δ3He values of the seawater samples varied from 4.5% The δ3He values of the western Philippine Sea (ST-6 to 22.9%. Vertical profiles of 3He anomaly (δ3He) ob- to -8) varied from 0.1% to 22.9%. Figure 3 show the ver- served in the western North Pacific did not vary signifi- tical profiles of 3He anomaly observed in the western cantly among sampling sites except at ST-5 (Fig. 2), sug- Philippine Sea. It is well documented that 3He is slightly gesting that seawater in the northern part (~30°N) of the less soluble (1.2 ± 0.5%) in water than 4He (Weiss, 1971; western North Pacific is homogeneous (probably well Benson and Krause, 1980). As a result, the 3He/4He ratio mixed) for laterally, and different from seawater in the in surface seawater is about 1% lower than that in the southern part (ST-5, ~10°N). In general, surface seawater atmosphere (Östlund et al., 1987; Sano et al., 1995). Al- is saturated with atmospheric noble gases because their though δ3He values in subsurface waters (~200 m depth) exchange rate is rapid. The δ3He values increased down- in the western Philippine Sea were almost 0%, they were ward, and maxima of about 18% were observed at mid- not different from the common surface seawater value of depth (2000–3000 m); they then decreased to about 15% –1% within analytical error. The δ3He values in interme- in the bottom water. It is well known that at mid-depth diate water (~500 m depth) were lower than those in the (~2500 m), all western North Pacific waters are affected western North Pacific, which may indicate a relatively by mantle helium with a high δ3He value, possibly de- rapid exchange between the surface and deep seawater rived from the East Pacific Rise (Östlund et al., 1987; compared with the ventilation of the adjacent North Pa- Lupton, 1995; Sano et al., 1995, 2004). cific. More precisely, as compared between three sites, we found that relatively smaller increase of δ3He from

456 N. Takahata et al. (a) western Philippine Sea

(b) eastern Philippine Sea

Fig. 4. Vertical distributions of δ3He (%) along the meridian transections: (a) western Philippine Sea (~128°E), (b) eastern Philippine Sea (~138°E), (c) western North Pacific (~148°E). The δ3He maxima are observed at mid-depth (2000–3000 m depth) in all regions. However the northern limits of δ3He = 20% differed among regions, suggesting different patterns of deep-sea circulation in the lateral plane.

Helium isotopes of seawater in the NW Pacific 457 the surface to 500 m depth in northern samples, suggest- (a) 2000 m ing that the mixing layer may become thicker toward the northern part of the western Philippine Sea. Maximum δ3He values of >20% were observed at 2000–2500 m depth; these maxima were thus higher than those observed in the northern part (~30°N) of western North Pacific. Indeed, the helium profiles in the western Philippine Sea resemble more to the profile observed at ST-5, in the southern part (~10°N) of the western North Pacific. This difference probably reflects a difference in the contribution of mantle helium, which is emitted from submarine volcanoes and brought by abyssal currents from outside of the Philippine Sea. These Philippine Sea trends are also roughly consistent with the Philippine Sea profiles reported by Sano et al. (2004) and Takahata et al. (2004).

Vertical and lateral distributions of helium isotopes (b) 2500 m Vertical cross sections of δ3He in the western Philip- pine Sea (Fig. 4a, 128°E), the eastern Philippine Sea (Fig. 4b, 138°E), and the western North Pacific (Fig. 4c, 148°E), drawn using both our data and previously reported data (Sano et al., 2004; Takahata et al., 2004; World Ocean Circulation Experiment data, http://whpo.ucsd.edu/), do not show substantial differences among the three regions. δ3He maxima were observed at mid-depth (2000–3000 m) in all regions. However, the northern limits of δ3He = 20% differed among regions. The region of δ3He value of >20% reached at (Fig. 4b) or crossed (Fig. 4a) the 20°N meridian in the Philippine Sea, but it did not reach at the 20°N meridian in the western North Pacific (Fig. 4c), suggesting different patterns of deep-sea circulation in the lateral plane. Maps of the lateral δ3He distribution in the Philippine Sea and the western North Pacific were similar at 2000, 2500, and 3000 m depth (Fig. 5). Re- gions of higher δ3He were observed in the southeast, (c) 3000 m whereas lower δ3He values were observed in the north- east. The helium-3 plumes detected in the southeast seem to emanate from the East Pacific Rise or the Tonga–Fiji region where plumes with δ3He of more than 40% were found (Lupton et al., 2004). We inferred deep currents

Fig. 5. Lateral distributions of δ3He (%) in the Philippine Sea and the western North Pacific at 2000 m (a), 2500 m (b) and 3000 m (c) depths. Whereas the helium-3 plume observed in southeastern region derived from the East Pacific Rise and/or the Tonga–Fiji region seems to flow northward along the Izu– Ogasawara–Mariana Ridge at 3000 m depth, it seems to flow westward through the Caroline Basin or the Yap–Mariana Junc- tion into the Philippine Sea at 2000–2500 m depth. Moreover, in the western Philippine Sea, it apparently flows northward to the region adjacent to the Japanese Islands. Schematics of the Fig. 5. deep currents inferred from the distributions are also shown.

458 N. Takahata et al. from this lateral δ3He distribution. Whereas the helium-3 Acknowledgments—We thank the scientists of the KH-04-4, plume in the Pacific Ocean seems to flow northward along KH-06-2, and KH-07-1 cruises and the crews of R/V Hakuho the Izu–Ogasawara Ridge at 3000 m depth, it seems to Maru for seawater sample collection. We appreciate construc- flow westward through the Caroline Basin or the Yap– tive reviews from T. Matsumoto, R. Mohapatra and an anony- Mariana Junction into the Philippine Sea at 2000–2500 mous reviewer. This work was partly supported by Grants-in- Aid for Scientific Research from the Ministry of Education, m depth. Moreover, in the western Philippine Sea, it ap- Culture, Sports, Science and Technology (No. 17101001 to YS parently flows northward to the region adjacent to the and No. 18710005 to NT). Japanese Islands. The mantle helium observed before at mid-depth in the northern Philippine Sea seems to come from the East Pacific Rise or the Tonga–Fiji region. Then, REFERENCES it may flow eastward out of the Philippine Sea through Benson, B. B. and Krause, D., Jr. (1980) Isotope fractionation the gap between the Izu and the Ogasawara ridges or of helium during solution: a probe for the liquid state. J. through that between the Ogasawara and the Mariana Solution Chem. 9, 895–909. ridges. Although these flows inferred from the δ3He dis- Craig, H., Clarke, W. B. and Beg, M. A. (1975) Excess 3He in tribution are roughly similar to those that have been esti- deep water on the East Pacific Rise. Earth Planet. Sci. Lett. mated from physical water properties such as tempera- 26, 125–132. ture, salinity, and dissolved oxygen concentration (e.g., Fujio, S., Kadowaki, T. and Imasato, N. (1992) World ocean Fujio et al., 1992; Reid, 1997; Kawabe et al., 2003), the circulation diagnostically derived from hydrographic and δ3He distribution could reveal that deep-water circula- wind stress fields. Part 1: The velocity field. J. Geophys. Res. 97, 11163–11176. tion seems to be different at each depth (2000, 2500, 3000 Igarashi, G., Kodera, M., Ozima, M., Sano, Y. and Wakita, H. m) that have not been published before. Whereas deep (1987) Noble gas elemental and isotopic abundances in deep current at 3000 m depth seems to flow northward along sea trench in the West Pacific. Earth Planet. Sci. Lett. 86, the Izu–Ogasawara Ridge, it does not occur at 2000–2500 77–84. m depth. Jenkins, J. E., Beg, M. A., Clarke, W. B., Wangersky, P. J. and Craig, H. (1972) Excess 3He in the Atlantic Ocean. Earth Planet. Sci. Lett. 16, 122–126. SUMMARY Kawabe, M., Fujio, S. and Yanagimoto, D. (2003) Deep-water We investigated helium isotopic ratios of 26 seawater circulation at low latitudes in the western North Pacific. samples collected in the western Philippine Sea and 58 Deep-Sea Res. 50, 631–656. seawater samples collected in the western North Pacific Kipfer, R., Aeschbach-Hertig, W., Peeters, F. and Stute, M. (2002) Noble gases in lakes and ground waters. Noble Gases Ocean to understand deep-water circulation more pre- in Geochemistry and Cosmochemistry (Rev. in Mineralogy cisely at mid-depth together with already reported data. and Geochemistry 47) (Porcelli, D., Ballentine, C. and δ3 Intermediate water (~500 m depth) had lower He val- Wieler, R., eds.), 615–700, Mineralogical Society of ues in the northwestern Philippine Sea than in the west- America, Washington, D.C. ern North Pacific, suggesting a relatively rapid exchange Lupton, J. E. (1983) Terrestrial inert gases: Isotope tracer stud- between the surface and deep seawaters compared with ies and clues to primordial components in the mantle. Ann. the ventilation in the adjacent North Pacific. Maximum Rev. Earth Planet. Sci. 11, 371–414. δ3He values of >20% were observed at mid-depth (2000– Lupton, J. E. (1995) Hydrothermal plumes: Near and far field. 2500 m) in the western Philippine Sea and in the south- Seafloor Hydrothermal Systems: Physical, Chemical, Bio- ern part (~10°N) of the western North Pacific. These logical and Geological Interactions. Geophysical Mono- maxima were higher than those observed in the northern graph 91, 317–346, American Geophysical Union, Wash- ington, D.C. part (~30°N) of the western North Pacific at the same Lupton, J. E. and Craig, H. (1981) A major helium-3 source at depth. This difference probably reflects a difference in 15°S on the East Pacific Rise. Science 214, 13–18. deep-sea circulation between these regions. The lateral Lupton, J. E., Pyle, D. G., Jenkins, W. J., Greene, R. and Evans, 3 δ He distributions at mid-depth suggest that deep seawater L. (2004) Evidence for an extensive hydrothermal plume in from the South Pacific may enter the Philippine Sea the Tonga–Fiji region of the South Pacific. Geochem. through the Caroline Basin or the Yap–Mariana Junction. Geophys. Geosyst. 5, Q01003, doi:10.1029/2003GC000607. Although its flow has also been suggested roughly by Mamyrin, B. A. and Tolstikhin, I. N. (1984) Helium Isotopes in physical properties or current meter, the δ3He distribu- Nature. Elsevier, Amsterdam, 273 pp. tion revealed fine structures of water mass that have not Östlund, H., Craig, H., Broecker, W. S. and Spencer, D. (1987) been published by previous researchers. The δ3He data Shorebased data and graphics. GEOSECS Atlantic, Pacific obtained by this work could be used for validation of deep- and Indian Ocean Expeditions, Volume 7. Reid, J. L. (1997) On the total geostrophic circulation of the water circulation models estimated by other physical or pacific ocean: flow patterns, tracers, and transports. Prog. chemical studies. Oceanogr. 39, 263–352.

Helium isotopes of seawater in the NW Pacific 459 Rison, W. and Craig, H. (1983) Helium isotopes and mantle stract). volatiles in Loihi Seamount and Hawaiian Island basalts Sano, Y., Takahata, N. and Gamo, T. (1995) Helium isotopes in and xenoliths. Earth Planet. Sci. Lett. 66, 407–426. South Pacific deep seawater. Geochem. J. 29, 377–384. Sano, Y. and Wakita, H. (1985) Geographical distribution of Sano, Y., Kosugi T., Takahata, N. and Yokochi, R. (2004) He- 3He/4He ratios in Japan: Implications for arc tectonics and lium isotopes in Pacific waters from adjacent region of incipient magmatism. J. Geophys. Res. 90, 8729–8741. Honshu, Japan. J. Oceanogr. 60, 625–630. Sano, Y. and Wakita, H. (1988) Precise measurement of helium Takahata, N., Watanabe, T., Shirai, K., Nishizawa, M. and Sano, isotopes in terrestrial gases. Bull. Chem. Soc. Japan 61, Y. (2004) Helium isotopes of seawater in adjacent sea of 1153–1157. Nansei Islands, southwest Japan. Geochem. J. 38, 593–600. Sano, Y., Wakita, H., Ishibashi, J., Gamo, T. and Sakai, H. (1989) Weiss, R. F. (1971) Solubility of helium and neon in water and Helium isotope ratios in Japan Sea water. Chikyukagaku seawater. J. Chem. Eng. Data 16, 235–241. (Geochemistry) 23, 61–67 (in Japanese with English ab-

460 N. Takahata et al.