geographical distribution of 3 he& 4 he ratios in japan ... · journal of geophysical research,...

JOURNAL OF GEOPHYSICAL RESEARCH, VOL. 90, NO. B10, PAGES 8729-8741, SEPTEMBER 10, 1985

Geographical Distribution of 3He/He Ratios in Japan' Implications for Arc Tectonics and Incipient Magmatism

YUJI SANO AND HIROSHI WAKITA

Laboratory for Earthquake Chemistry, Faculty of Science University of Tokyo, Japan

The 3He/'•He and '•He/2øNe ratios of 115 natural gas samples, including various types of gases from volcanoes, hot springs, mineral springs, water wells, petroleum fields, and natural gas fields, were measured using a mass spectrometer. The observed 3He/'•He and '•He/2øNe ratios range from 7.47 x 10 -s to 9.65 x 10 -6 and from 0.26 to 1100, respectively. The 3He/'•He ratios reflect well the geotectonic structure of the Japanese Islands. In northeastern (NE) Japan there is a clear geographical difference in the 3He/'•He ratios between the frontal arc (forearc) and volcanic arc (back arc) regions. Lower ratios were found in the trench side region and higher ratios in the back arc side. This result suggests that the mantle-derived helium in the volcanic arc region is associated with the diapiric uprise of a magma. Lower 3He/'•He ratios in the frontal arc region may be due to radiogenic He produced by radioactive decay of U and Th in the crustal and sedimentary rocks. In southwestern (SW) Japan there is no clear geographical contrast in the •He/'•He ratios. Some samples in the frontal arc region show quite high 3He/'•He ratios. The tendency of the 3He/'•He ratios agrees with the distribution of terrestrial heat flow data and reflects geotectonic structures different from NE Japan. The high 3He/'•He ratios observed in the frontal arc region in SW Japan may be indicative of renewed or incipient magmatism due to a descending young and warm slab.

1. INTRODUCTION

Correlations between isotopic parameters of solid elements, such as strontium and neodymium, and global tectonic environments have been shown by recent research. Helium with its isotope of 3He is the most useful volatile element. In light of knowledge accumulated by previous investigations, He isotope ratios can be useful for evaluating a variety of geotectonic environments [Lupron, 1983; Mamyrin and Tolstikhin, 1984]. A set of precise 3He/'•He data obtained throughout the Japanese Islands can provide new knowledge of geophysics and geochemistry in the subduction zone. The purpose of the present paper is to show the geographical distribution of He isotope ratios in the Japanese Islands and to clarify the relation between •He/'•He ratios and geophysical and geological features.

2. THE JAPANESE ISLANDS

The Japanese Islands consist mainly of four major islands, Hokkaido, Honshu, Shikoku, and Kyushu. In the central part of Honshu there is a long and dear geotectonic line called the "Itoigawa-Shizuoka tectonic line" (I-S TL). The Japanese Is- lands are divided by this line into two major tectonic blocks, northeastern (NE) Japan and southwestern (SW) Japan (Figure 1).

NE Japan has four specific geographical features, i.e., from east to west, a deep trench, a frontal arc region, a volcanic arc, and a back arc region with a marginal sea. Geophysical data, including seismic velocity, terrestrial heat flow, gravity anomaly, and vertical crustal movement, vary perpendicularily to the trend of the arc system [Yoshii, 1977]. The distribution of earthquake foci along the Wadati-Benioff zone can be attributed to subduction of the Pacific plate beneath NE Japan.

In SW Japan, in contrast, a well-defined island arc system feature has not developed. The Nankai Trough, which corre- sponds to the trench, is not so deep as the Japan Trench.

Volcanic activity is generally lower, and the volcanic front is not as clear as in NE Japan except for Kyushu. Geophysical data are also different from those of NE Japan. There is no Wadati-Benioff zone longer than a few hundred kilometers, and heat flow values are strikingly high in the trench region [Yamano et al., 1984] except for Kyushu district with its dis- tinct volcanic front. These features are attributed to the sub-

duction of the young and warm lithosphere (Shikoku basin) beneath the Eurasia plate [Klein et al., 1978; Shiono, 1982; Shiono and Sugi, 1985].

3. EXPERIMENTAL DETAILS

3.1. Sample Collection

Various kinds of gases for helium isotopic analysis were collected from sources with different geologic environments. These include bubbling gases in hot springs, mineral springs, and water wells; methane-rich natural gases in gas fields; petroleum gases in oil fields; gases in volcanic furnarole; and gases dissolved in water in hot springs and mineral springs. Collection methods were varied according to the gas- discharging sources.

A 50-cm 3 lead glass container with a vacuum valve was used for collection of but•bling gas from hot springs, mineral springs, and water wells. Sample gas was collected by a dis- placement method in water. Volcanic fumarole gases and fuel gases from oil and gas fields were collected by a sample container with vacuum valves at both ends. At the sampling site the container was put into a water bucket to be filled with distilled water. One end of the container was connected with

the source of fumarole or fuel gas by using a glass tube or tygon. After water in the container was completely replaced by fumarole or natural gases, both valves were closed. The second type of container was also used to collect water sample to measure dissolving gases in water at sites where bubbling gases were not available.

Copyright 1985 by the American Geophysical Union.

Paper number 4B5134. 0148-0227/85/004 B- 5134 $05.00

3.2. Helium Isotopic Measurements

After the sample container was brought back to the laboratory, 3He/'•He and '•He/2øNc ratios were measured with a

8729

8730 SANO AND WAKITA' THE 3HE/'•HE RATIOS IN JAPAN

45ON

4O

Japan Sea

doubly charged ions of 4øAr with 2øNe was checked using the atmospheric Ar standard. The production rate of 4øAr2+/½øAr + was measured to be less than 0.01. In the case of actual samples, Ar was completely adsorbed on a cold charcoal trap, and the beam intensity of ½øAr ion was less than that of 2øNe. Thus the interference was estimated to be

less than 1% of 2øNe at the maximum. A precise description of the 3He/'•He and '•He/2øNe ratio measurements was given by $ano et al. [1982].

35 o

30 ø

SW

\

Pacific Ocean '. , ,,

130øE 135 ø 140 ø

Fig. 1. Tectonic configuration of the Japanese Islands. The Japa- nese Islands are divided into two parts, NE Japan and SW Japan, by the tectonic boundary of the "Itoigawa-Shizuoka tectonic line (I-S TL)." The studied area is discussed separately for the following districts: A (Hokkaido), B (Tohoku), C (Kanto and Chubu), D (Chubu), E (Kinki and Chugoku), and F (Kyushu).

6-inch mass spectrometer (6-60-SGA, Nuclide Co.). About 1 cm 3 STP of the original gas sample was introduced into the three-step purification line. Major chemical constituents such as N2, 02, CH4, and CO2 were removed by charcoal traps kept at -196øC and hot copper oxide. Heavier noble gases (Ar, Kr, and Xe) were also adsorbed on the charcoal trap. Purified He and Ne in original gas samples with about 1 x 10 -6 cm 3 STP of '•He were introduced into the mass spec-

trometer. Helium was not separated from Ne, which may lead to slightly higher 3He/'•He ratios than the true values [Rison and Craig, 1983]. However, the conclusions in this study are not substantially affected by this effect. Ion beams of 3He and '•He were detected by a double collector system. Resolving power of about 600 at 1% of peak height was attained for the complete separation of 3He beam from those of H 3 and HD. In the case of water samples, He and Ne were extracted using a vacuum line and were introduced into the purification line prior to the gas purification process [$ano, 1983]. In a hot blank (i.e., the static background due to actual purification procedure without sample), amounts of 3He and '•He were below 1 x 10 -x3 and 3 x 10 -9 cm 3 STP, respectively, and were negligibly small compared with those of actual samples.

A standard sample for He isotope measurements was pre- pared by mixing known amounts of pure ZHe and '•He gases. The mass discrimination factor of the measuring system was calculated by the standard sample. Atmospheric air at Tokyo was used as a running standard. The mean value of the corrected 3He/•He ratios in the air is (1.42 + 0.04) x 10 -6 from 31 individual runs. This value is in good agreement with the critical value of 1.399 x 10 -6 found by Marnyrin et al. [1970] and 1.384 x 10 -6 found by Clarke et al. [1976].

The '•He/2øNe ratio was measured with the same mass spectrometer by adjusting the magnet current. Interference from

4. RESULTS

4.1. The 3He/•He and •He/•øNe Ratios of the Japanese Gases

Results of 3He/'•He and '•He/2øNe ratios of 115 samples collected throughout the Japanese Islands are shown in Table 1. In order to obtain a picture of the general degassing features of the Japanese Islands, comparison of the 3He/'tHe ratio distribution in a single sample type throughout the archipel- ago would be desired. As such sampling is practically impossi- ble, we will make use of as many samples of various sample types as are available. Effects due to differences in sample type can be ignored, as explained below.

Since 3He is the primordial component derived from the mantle, any difference in the concentration in a sample can be regarded as a difference in the contribution from the mantle, irrespective of changes in sample type. Fumarolic gases are inevitably limited to the volcanic areas in the Islands due to their genetic configuration. Hot springs as well as mineral springs are also heterogeneously distributed; that is, hot springs are generally located in the area on and westward of the volcanic front and mineral springs are located in the frontal arc region. The distributions can be understood to reflect a major characteristic of the island arc system due to the lack of magmatic sources in the frontal arc region. The definition of a hot spring in this paper is a spring issued at temperature greater than 25øC. A spring at temperature less than 25øC is a mineral spring. By this criterion, some hot springs (i.e., sites 45 and 85) in the sedimentary basins are assigned as nonvolcanic hot springs. Hot springs and mineral springs are essentially the same and differ only in their temperature [Yuhara and $eno, 1969].

Wells drilled in the frontal arc region generally provide low- temperature water, whereas those in the back arc region generally yield high-temperature water. Water temperature of deep wells in the frontal arc region, however, would be expected to be higher due to their geothermal gradient. Some of these may also be classified as hot springs.

Natural gases and petroleum gases belong to another cat- egory in the sample type. In this paper, petroleum gas is defined as gas associated with petroleum. A wide variety in the 3He/•He ratio ranging from 1 x 10 -7 to 8 x 10 -6 was observed for methane-rich natural gases and petroleum gases in Japan [Wakita and $ano, 1983]. The observed systematic distribution of higher content of 3He in the back arc region and lower in the frontal arc region was interpreted to be generated by magmatic activities.

The highest 3He/•He ratio, 9.65 x 10 -6, was observed at a hot spring (site 71). Values higher than 9.0 x 10 -6 were obtained at five sites located close to volcanic and geothermal areas. It is noted that a mineral spring (site 54) at a temperature of 24øC showed one of higher 3He/'tHe ratios. The lowest 3He/'•He ratio, 7.47 x 10-8, was observed at a natural gas field (site 115). Lower 3He/'•He ratios were observed in

SANO AND WAKITA: TH• 3HE/4HE RAT•OS IN JAPAN 8731

samples collected from areas where sedimentary layers are thick and old.

The 3He/4He ratios of Hakone volcanic sites (sites 47 and 48) agree well with those previously reported by Craig et al. [1978]. Nagao et al. [1981] carried out rare gas isotopic measurements of natural gases in Japan, and several sampling sites of their study overlap with ours. Although excellent agreements are observed between their values and ours for several samples (for example, samples from sites 47, 93, and 113), values for samples from sites 94 and 101 are about 15% lower in this study. In order to discuss the data using the same criteria, only data of this study using the same sampling method and analytical procedure are adopted, although the conclusions are not substantially affected by the other results.

4.2. The 3He/4He Profiles in NE Japan Figures 2-4 display 3He/'•He ratio in terms of geographic

distance from the volcanic front [Matsuda and Uyeda, 1970] for Hokkaido (district A), Tohoku (district B), and Kanto and Chubu (district C) in NE Japan. Samples with •He/2øNe values lower than 0.7, reflecting air contamination of more than 45% (see section 5.1), are excluded from Figures 2-4. In Figures 2-4 each sample type is indicated by a different symbol. Each maximum 3He/•He value in the frontal arc and back arc regions is shown by a solid line. The hatched zone is the transition region of the 3He/•He ratio near the volcanic front.

In the Hokkaido district (Figure 2), only one sample was collected near the volcanic front in the frontal arc region. The 3He/4He ratio of 1.72 x 10 -6 is slightly higher than the atmospheric ratio. Just on the volcanic front the 3He/'•He ratio is 5.30 x 10 -6, about 4 times higher than that of the atmosphere. That value, however, is as low as one third of the MORB-type He. Farther toward the back arc region, the 3He/•He ratio becomes gradually higher. At a hot spring, about 360 km away from the trench, the 3He/•He ratio of 7.64 x 10 -6 is the highest in the Hokkaido district: over 5 times the atmospheric ratio and almost equivalent to that of volcanic gases in the Circum-Pacific region (subduction-type He).

We have already reported a 3He/•He profile in the Tohoku district [Sano et al., 1985]. In the present work (Figure 3), two hot spring gases (samples 16 and 17), eight petroleum gases [Wakita and Sano, 1983], and a natural gas on the Pacific coast (sample 36) are added to the earlier profile. In the frontal arc region the 3He/4He ratios are extremely low with values of (1-5) x 10 -7. The lowest ratio, 1.61 x 10 -7, is observed at a site about 80 km from the volcanic front.

Near the volcanic front the 3He/½He ratios are significantly higher, as shown in a hatched zone. Just on the front, the 3He/•He ratio is 5.00 x 10 -6, equivalent to that in the Hok- kaido district. Farther toward the back arc region, the ratio becomes gradually higher, reaching the maximum ratio. At a hot spring 170 km away from the volcanic front, the ratio shows the highest value of 9.01 x 10 -6.

In the Kanto and Chubu district (Figure 4), overall variation in 3He/•He ratios is similar to those of A and B districts (Figures 2 and 3). In the frontal arc region facing the Pacific coast, the ratios are extremely low with values of (1-3) X 10 -7. Closer to the volcanic front, the 3He/4He ratios

become slightly higher. The 3He/4He ratios of two mineral springs (sites 39 and 40), about 10-15 km east of the volcanic front, are slightly lower than that of the atmosphere and ap- parently higher than those of sites facing the Pacific Ocean.

The observed difference is discussed in more detail in sections

5.2 and 5.3. Just on the volcanic front, the 3He/'•He ratio becomes significantly higher as in the cases of A and B districts. The value itself on the front is 8.29 x 10 -6 (sample 47), significantly higher than those of A and B districts. Generally, 3He/4He ratio profiles are similar to one another in NE Japan.

4.3. The 3He/'SHe Profiles in SW Japan Figures 5-7 show the measured 3He/4He ratios versus lo-

cation of sampling site for Chubu (district D), Kinki and Chugoku (district E), and Kyushu (district F) in SW Japan. Profiles of the 3He/4He ratio in SW Japan are rather compli- cated; that of E is dramatically different from those seen in NE Japan, which may reflect differences in tectonic environments.

In the Chubu district (Figure 5) the volcanic front has not been defined. A water well gas at the farthest site from volcanic ranges shows the radiogenic 3He/'•He ratio of 1.02 x 10-7, suggesting negligible contribution of subduction-type

He. Gas samples with the ratios of 1-2 times higher than the atmospheric value are in the frontal arc region. In the back arc region the ratio becomes significantly higher. As seen in Figure 5, there is a boundary in the 3He/4He ratios, which may correspond to a volcanic front in the area. In Figure 5, 3He/4He ratios are plotted in terms of relative distance from this He boundary.

The 3He/4He profile in the Kinki and Chugoku districts (Figure 6) is really different from those of NE Japan. Although the volcanic front exists in the central part of Chugoku district, a contrast in the 3He/4He ratio is not observed in the profile. Furthermore, even in the frontal arc region, most of the samples have significantly higher 3He/'•He ratios than that of the atmosphere. Two samples collected at sites far away from the volcanic front have 3He/½He ratios 5 times larger than the atmosphere. Toward the volcanic front, the 3He/'•He ratio becomes higher gradually. At Arima (site 94) hot spring, about 60 km away from the volcanic front, the 3He/4He ratio reaches the highest value, about 7 times as large as that of the atmosphere. In the area close to the volcanic front the lowest 3He/4He ratio of 8.88 x 10-7 was observed. In the back arc region, values higher than atmospheric ratio, 5.0 x 10 -6, are observed. As shown in Figure 6, anomalously high 3He/'•He ratios are observed in an area with a diameter of about 100

km, the "Kinki spot," in front of the volcanic front. The 3He/4He profile in the Kyushu district (district F) in

SW Japan is shown in Figure 7. The contrast of the 3He/'•He ratio on both sides of the volcanic front is much clearer in this

district than in the Kinki and Chugoku district (district E) and looks like the profile of NE Japan. A sample collected at a gas field, in the frontal arc region about 75 km away from the volcanic front, indicates the radiogenic 3He/4He ratio. Most samples in the volcanic arc region have higher 3He/4He ratios than the atmosphere. The highest value, 9.54 x 10 -6, was observed at a volcano, just on the volcanic front.

5. DISCUSSION

5.1. Mixing of Three Helium Sources

It is difficult to identify the source of He based only on the 3He/4He ratio; if the observed ratio is about 1.4 x 10 -6, it is either atmospheric He or a mixture of mantle-derived He and radiogenic He. The observed 3He/4He ratios for all samples are plotted against the 4He/2øNe ratios in Figure 8. The distri-

8732 SANO AND WAKITA' Tm• 3HE/•HE I•OS XN JAPAN

• V V V •

' •VVV

,,,¸D

Z 5

.•,

ß .. • •o •

SANO AND WAKITA' Tmi •HE/'*HE RnT•OS IN JAPAN 8733

Z• Z Z

+1 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 • +1 +1 +1 • • +1 • +1 +1 +1 +1 +1 • +1 +1 +1 +1 +1 +1 +1 +1 +1 •

zzzzzzz

8734 SANO AND WAKITA: Ttm SHW'•H•. RATIOS IN JAPAN

88' 8•øz ' • z .... O• •

• O O O O O O O O •

SANO AND WAKITA' TH• 3Hœ/4Hœ RATIOS IN JAPAN 8735

NE JAPAN DISTRICT A (HOKKAIDO)

10.0 VF

7.5-

w4 08 5.0 90 1

2.5 lOO 3• 12n TM

i lOO 5o o

i ø > 50 km

Distance from the volcanic front (km)

Fig. 2. The 3He/'•He profile in the Hokkaido district (A), showing observed 3He/'•He ratio versus geographic distance from sampling site to the volcanic front and locations of sampling sites. VF, volcanic front (the position cannot be drawn exactly). The dashed line shows the atmospheric ratio. Symbols indicate emitting feature of sample: solid circles, hot spring gas and hot spring water including high- temperature water well; open circles, mineral spring gas and mineral spring water including low-temperature water well; open squares, natural and petroleum gases; and solid triangles, volcanic fumarole. Samples 6 and 11 are excluded due to their low '•He/2øNe ratios.

bution of most samples in the diagram is concentrated in a trianglelike region, suggesting the presence of three end- members. Corresponding signatures are as follows:

Subduction-type He (S). Helium from gases of volcanoes and hot springs in the Circum-Pacific zone, including Kam- chatka, the Kuriles [Baskov et al., 1973], the Marianas, Mount Lassen [Craig et al., 1978], and New Zealand [Torger- sen et al., 1982]. The 3He/4He ratios are up to 1.1 x 10 -5, and the maximum value is strikingly uniform. The 4He/2øNe ratios vary significantly from 2.7 (the Marianas) to higher than 830 (Mount Lassen). The low 4He/2øNe ratio may be attributed to contamination by atmospheric He or by seawater. Hence the highest ratio of 1000 is accepted as the subduction-type 4He/2øNe ratio.

Radiogenic He (R). Detailed calculation by Morrison and

NE JAPAN DISTRICT B (TOHOKU)

VF 22/:2• i

J •, 10.0 - •7• x :o----." • •o o32 •A ß ,,. P•,, / ,4. • 5.0 - Q•e ,

m 2.5 •TM 023 •

15o lOO 5o o 5o lOO


Fir. 3. The 3He/•Hc profile i• the Tohoku district (B), showier observed 3He/•He ratio versus 8eosraphic dista• from sampli•8 site to the volcanic fro•t a•d locations of sampling. VF, volcanic fro•t (the position ca•ot bc draw• precisely). The dashed ]i•e shows the atmospheric ratio. Symbols i•dicatc emitti•8 features of samples as described in the Fi;urc 2 capriore Samples ]5, 2], and 22 arc excluded due to their low •He/=øNc ratios.

10.0

7.5

5.0

2.5

2OO

NE JAPAN

DISTRICT C (KANTO & CHUBU) . ,: - / -60 ,"o

V F e 8Z' ..•67 , • m .'65 ,' I , I • •e ,,,'•

u IAJ. •,. 5o-5•t72 . .• • .... •

-- •68 47• 54'-53 67U •-, 3 • ', 4• 42•

•4• 4•e re'l •'•,• _ • 525oea .•e'l• •

...... 100 0 100 200


Fig. 4. The 3He/•He profile in the Kanto and Chubu district (C), showing observed 3He/'•He ratio versus geographic distance from sampling site to the volcanic front and locations of sampling sites. VF, volcanic front (the position cannot be drawn precisely). I-S TL, Itoigawa-Shizuoka tectonic line. The dashed line shows the atmospheric ratio. Symbols indicate emitting features of samples as described in the Figure 2 caption. Samples 55, 56, and 57 are excluded due to their low '•He/2øNe ratios.

Pine [1955] showed that the radiogenic production of 3He in the continental crust is due to the following reaction:

6Li(n, 003H --• 3He

According to our calculation based on the method developed by Gerlino et al. [1971] the 3He/4He ratio of 1.5 x 10 -8 may be representative of the radiogenic component of Japanese granitic rocks. Rison [1980] calculated that the production rate of 2øNe in the crust was 3.7 x 10 -2x cm 3 STP/g yr, based on the nuclear reaction x 70(•, n)2ONe. Our calculation for the 4He/2øNe ratio is about 1 x 10 s for average granitic rocks. The 4He/2øNe ratio of 1000 is tentatively adopted as a safe minimum for this component.

Atmospheric He (A). The 4He concentration of the atmosphere is 5.24 x 10 -6 by volume, with an isotopic ratio of

lO.O

7.5

5.0

2.5

SW JAPAN DISTRICT D (CHUBU)

(VF)

71

0 7-2 78 i

5o o

85

0 ATM

79 I I • I

50 100 150


Fig. 5. The 3He/'•He profile in the Chubu district (D), showing observed SHe/•He ratio versus geographic distance from sampling site to the He boundary and locations of sampling sites. The volcanic front has not been drawn in this area. The He boundary, which may correspond to the volcanic front, is drawn in this study. I-S TL, Itoigawa-Shizuoka tectonic line. The dashed line shows the atmospheric ratio. Symbols indicate emitting features of samples as described in the Figure 2 caption.

8736 SANO AND WAKITA' THE 3HE/aHE RATIOS IN JAPAN

SW JAPAN DISTRICT E (KINKI & CHUGOKU)

I VF •,, 10.0 I

7.5

5.0

2.5

94

101 96

08•o3 • 088 100 •

,105 • r 95 •89

087

I I I I ,I 50 0 50 100 150 200


Fig. 6. The 3He/'•He profile in the Kinki and Chugoku district (E), showing observed 3He/'•He ratio versus geographic distance from sampling site to the present volcanic front and locations of sampling sites as well as the "Kinki spot," a contour line within which 3He/4He ratios are higher. VF, volcanic front (the position cannot be drawn precisely). The dashed line shows the atmospheric ratio. Symbols indicate emitting features of samples as described in the Figure 2 caption. Samples 90 and 91 are excluded due to their low '•He/2øNe ratios.

1.40 x 10 -6. Since the atmosphere contains Ne in a concentration of 1.81 x 10 -s, the '•He/2øNe ratio of the atmosphere should be 0.318.

When He in a sample is composed of the above three components, the following equations are derived'

(3He/'•He) = (3He/'•He)a x A + (3He/'•He)s x S + (3He/'•He)r x n

1/('•He/eøNe) = A/('•He/eøNe),, + S/('•He/eøNe)s + n/('•ne/eøNe),.

A+S+R=I

where subscripts a, s, and r represent atmospheric, subduction- type, and radiogenic He, respectively. Taking values (3He/4He)a = 1.4 x 10 -6, (4He/2øNe)a = 0.318, (3He/4He)• = 1.1 x 10 -s, ('•He/2øNe), = 1000, (3He/'•He), = 1.5 x 10 -8, and ('•He/2øNe)r = 1000, respectively, we can calculate the fraction of components A, S, and R in the sample.

Most hot spring gases contain about 30% of the subduction-type He, although a negligible contribution of this type He is noted for one sample (sample 45). The contribution of the subduction-type He for mineral spring gases varies only slightly and overlaps with a wider range of hot spring gases. All volcanic fumaroles show large amounts of the subduction- type He.

5.2. Signature of 3He/4He Ratio in the Frontal Arc Region of NE Japan

As indicated in Figures 2-4, there are clear contrasts in 3He/4He ratio between the frontal arc and the volcanic arc regions in the entirety of NE Japan. This may reflect a difference in the degassing feature of the mantle He. There are two possible modes for mantle-derived He to reach the surface of the earth. The first is diffusion of He from the mantle through crustal rock. The second is transportation by media which bring mantle materials containing He to the surfac•.

Considering the low permeation velocity of He in crystalline rock, the diffusion process path is unlikely. The diffusion coefficient of He in sedimentary rock is estimated to be (4.2 q- 0.6) x 10- s cm:/s [Ohsurni and Horibe, 1984]. Since the thickness

of the crust in NE Japan is estimated to be about 30 km [Yoshii and Asano, 1972], it would take more than 6.8 x 10 9

years for He to migrate the distance of 30 km by the simple diffusion process. Furthermore, the diffusion coefficient of He in ordinary crystalline rocks is considered to be much smaller than that in a sedimentary layer, so transport velocity in the basement rock should be correspondingly smaller. Thus radiogenic He from the radioactive decay of U and Th in the crustal rocks will be dominant over mantle-derived He in this

case.

The second possibility is material flow containing primitive 3He. The results of the spatial distribution of 3He/'•He ratios around Mount Ontake, an active volcano, showed that the 3He/'•He ratios decrease with distance from the central cone of the volcano [Sano et al., 1984]. This tendency suggests that the source of the mantle-derived He is concentrated beneath

the central part of the volcano. Since the source magma of the volcano is generated in the upper mantle, He with a high 3He/'•He ratio may also be brought with the rising magma. In the frontal arc region of NE Japan, volcanic activity has not been noted during the past 20 m.y. [Matsuda and Uyeda, 1970]. Together with the scarcity of magmatic activity, the lower heat flow data imply that magmatic intrusion has not occurred in a significant amount. The lower 3He/'•He ratio in the region may be attributed to the radiogenic He from U and Th in old basement rocks and sedimentary layers and to the lower contribution of mantle-derived He due to lack of rising materials.

More precisely, there are differences among radiogenic 3He/'•He ratios in the frontal arc region. The 3He/'•He ratios of samples collected in the area closer to the volcanic front shown in a hatched zone with widths of 5 and 20 km are

higher than in those collected in the farther area, from about 40 to 200 km (see Figures 2-4). The mean value of three samples in the former region (sites 14, 39, and 40) is (1.3 q- 0.4) x 10 -6. Considering their '•HefføNe ratios, however, atmo-

spheric He is not a dominant component (less than 3.5%), and an addition of about 10% mantle-derived He is suggested. The small addition of mantle-derived He is attributed to the leak-

age of magmatic He from a volcanic region. As was described above, the correlation observed around Mount Ontake is ex- pressed as

(3He/'•He)d = 9.43 x 10 -6- 3.07 x 10 -7 x d

10.0

7.5

5.0 -

ATM

SW JAPAN DISTRICT F (KYUSHU)

VF

113

106•00 7'

112• 098•8 •1• 1

111

106-111

112-• i•'•'e /

1200 km I 100 50 0 50 100


Fig. 7. The 3He/'rHe profile in the Kyushu district (F) showing observed 3He/'•He ratio versus geographical distance from sampling site to the volcanic front and locations of sampling sites. VF, volcanic front (the position can not be drawn precisely). The dashed line shows the atmospheric ratio. Symbols indicate emitting features of samples as described in the Figure 2 caption. Sample 114 is excluded due to its low '•HefføNe ratio.

SANO AND WAKITA: THE 3HE/4HE RATIOS IN JAPAN 8737

2O

10

5.0

o 1.o-

0.5-

o.1-

0.05 -

0.01 0.1

Ail

Subduction-type

[] [] o

[] []

ß volcanic fumarole

[] natural & petroleum gas ß hot spring gas & water 0 mineral spring gas & water

I I I 0.5 1.0 5.0

I I I lO 50 ,oo

'•He/2ONe

Radiogenic

I I 50o lOOO

Fig. 8. Correlation between the 3He/'•He and '•He/:øNe ratios in Japanese samples. Solid line shows the mixing lines between subduction-type He and atmospheric He and between radiogenic He and atmospheric He. Symbols indicate emitting features of samples as described in the Figure 2 caption.

where d and (3He/'•He)d denote the distance of the site from the cone and the observed 3He/'•He ratio at the sampling site, respectively. The trend suggests that the more primitive 3He is carried with volatile-rich fluid flow originating from the ascending magma. In the area distant from the cone the magmatic He will be transported through fissures to the ground surface by thermal fluids. During the process a significant amount of radiogenic He from the crustal rocks and atmospheric He would be added to the magmatic He.

A small amount of the mantle-derived He in the frontal arc

region of NE Japan may be explained by the same leakage model as for Mount Ontake. Thus a similar equation to that used for Mount Ontake is adopted for NE Japan. Since the mean 3He/'•He ratio of the samples collected just on the volcanic front in NE Japan 'is 5.2 x 10 -6, we replace the initial value of 9.43 x 10 -6 with 5.2 x 10 -6 for NE Japan. The average distance of sites closer to the volcanic front in the frontal arc region (sites 14, 39, and 40) is 11 km. The calculated 3He/'•He ratio of 1.8 x 10 -6 agrees well with the observed mean value of (1.3 + 0.4) x 10 -6. That is, the approximately 10% mantle-derived He in these samples can be attributed to leakage of magmatic He from the volcanic region.

In the samples collected between 40 and 200 km away from the front on the trench side, the observed radiogenic 3He/'•He ratios fluctuate of the order of 10-7 (Figure 4). Although the tectonic structure of the Kanto Plain is similar to those of the

other areas of NE Japan, a large variation in the ratio is observed. The Kanto Plain is composed of late Cenozoic sedimentary materials. Tertiary sedimentary mountains with Neo- gene granitic rock surround the plain and slope toward it. The thickness of the sedimentary layer attains about 2 km on the northwest side of the plain and north of Tokyo Bay. In the central part of the plain the thickness exceeds 4 km. Table 2 gives the thicknesses of sedimentary layers at the sampling sites, the observed 3He/'•He ratios, and the corrected 3He/½He ratios for the atmospheric contamination. The thicker the sedimentary layer at the sampling site, the lower the 3He/'•He

ratio indicated. The relation between the corrected 3He/'•He ratio and the thickness is described as

(3He/½He)t = 3.94 x 10 -7- 7.06 x 10 -8 x t

where t and (3He/'•He)t denote the thickness of the sedimentary layer and the observed 3He/'•He ratio, respectively. This tendency may indicate that the He in these samples is a mixture of two sources with low 3He/'•He ratios. One is radiogenic He with the 3He/'•He ratio of 2 x 10 -$ and the other is the crustal He in granitic basement rocks with the estimated ratio of 4 x 10 -7. The observed value in the area may be explained by a mixture of mostly radiogenic He and a small amount of magmatic He from past volcanism. Although the equation is similar to that around Mount Ontake, the slope differs significantly. In the case of Mount Ontake the slope was estimated to be 3.07 x 10 -7 (1/km) , about 4 times larger than that of the Kanto Plain. The difference may reflect the direction of He migration through media where He is passing. The direction is vertical in the Kanto Plain but is nearly horizontal around Mount Ontake. The migration rate may also be larger in sediments than in crystalline rocks. This phe- nomenon will account for the difference of the inclination in

the equations.

5.3. Signature of the 3He/4He Ratios in the Back Arc Regions of NE Japan

More precisely, 3He/½He profiles of NE Japan (Figures 2 -- 4) differ from one another. The profile of district C (Kanto and Chubu district) is distinguished from those in districts A (Hokkaido district) and B (Tohoku district). In the narrow regions near the volcanic front the 3He/'•He ratios in district A are almost the same as those in district B but are quite different from those of district C. In districts A and B the 3He/'•He ratios just on the front are about 5.0 x 10 -6 and are appar- ently lower than those in the back arc region, 9.0 x 10 -6. On the other hand, in district C the ratios just on the front are

8738 SANO AND WAKITA.' TH• 3HE/'•HE RAT[OS IN JAPAN

TABLE 2. Thickness of Sedimentary Layer in the Southern Kanto District and Corrected 3He/'•He Ratios

NE Japan

aHe/'•He (aHe/'•He)co, Thickness, No. Name ( x 10 -6) 4He/2øNe ( x 10 -6) m

41 Narashino 0.213 _ 0.006 50 42 Yokoshiba 0.349 _ 0.011 14 43 Shirako 0.310 _ 0.005 4.8 44 Chonan 0.221 _+ 0.010 4.2 45 Heiwajima 0.137 _ 0.010 100

0.205 2100 0.325 1300 0.232 2600 0.124 3500 0.133 4000

about 9.0 x 10 -6 and are the same as those in the back arc

region. Figure 9 shows the variations in the 3He/4He ratios in areas

along the volcanic front in NE Japan, parallel to the trench axis. Plotted data are for selected samples collected in the transition region with a width of 25 km, 5 km on the frontal arc side, and 20 km on the back arc side of the volcanic front. From the northernmost point (site 1) to the central part of the Kanto district (sites 38 and 52); that is, in the belt zone of about 800 km length and 25 km width the 3He/4He ratios are almost constant with a value of (5.2 q- 0.6) x 10 -6. In the central part of the Kanto district the ratio becomes higher gradually toward the Izu Peninsula. The changing point in the 3He/•He ratios coincides well with the bending position of the volcanic front (Figure 9). This point is a junction of the NE Japan arc and the Izu-Ogasawara arc.

On the basis of the precise measurements of volcanic rocks from the NE Japan arc and the Izu-Ogasawara arc, Notsu [1983] showed that the 87Sr/S6Sr ratios are consistently higher (0.7038-0.7045) in NE Japan and are lower (0.7032- 0.7038) in the Izu-Ogasawara arc. The tendency is in good agreement with the variations found in the 3He/4He ratios of the present study. In the NE Japan arc the Pacific plate sub- ducts beneath the Eurasia plate, whereas in the Izu- Ogasawara arc the Pacific plate thrusts under the Philippine Sea plate. The dip angle of the Wadati-Benioff zone is steeper, and the thickness of the crust over the mantle wedge is thinner in the Izu-Ogasawara arc than in the NE Japan arc. The

N J

/

10'0 r // 5 4 •'•\.•'/,, /

,

-r- 2.5

200 400 600 800 1000

Distance (km)

Fig. 9. The variations in the aHe/'•He ratios in the zonal area with a width of 25 km along the volcanic front of NE Japan, parallel to the trench axis. Symbols indicate emitting features of samples as described in the Figure 2 caption.

10.0

5.0

VF

ß ' ß' .'1 I I

Trench

I

VF

T

Primordia I • 100 .

15 0 o øe

I /•1 I i 200 100 0 100 200 300

Fig. 10.


Schematic diagrams of the aHe/4He profile and tectonic structure of NE Japan.

higher 3He/4He ratios in the volcanic front region in district C may be due to less radiogenic contamination of the magma source.

The typical 3He/•He profile of NE Japan (districts A and B) is shown in Figure 10 and is summarized as follows: (1) The change in 3He/4He ratio occurs at the transition region near the volcanic front: lower ratios in the frontal arc region become higher in the back arc region (this supports the view that the magma is a source material carrying the primordial He from the upper mantle), and (2) the maximum 3He/'•He ratios in the back arc region are higher than those in the transition region. This implies that the radiogenic or crustal contamination is greater on the volcanic front than in the back arc region. A similar tendency was noted in the distribution of strontium isotopic ratios. The 87Sr/S6Sr ratios of volcanic rocks in the Tohoku district decrease from the east

side (0.7038-0.7045) to the west side (0.7028-0.7038) I-Notsu, 1983]. There are two possible mechanisms producing both He and Sr isotopes in the crust. One is the contribution of continental crust beneath the Japanese Islands. The other is the contamination of subducting crustal materials to the source region of magma. The uprising magma plays the role of a carrier of both primordial helium and strontium to the crustal materials. If the crustal thickness in the frontal arc region is greater than that in the back arc region, the contribution of the radiogenic He will become greater in the frontal arc region and be reduced gradually toward the back arc region. Accord- ing to the seismic data of Yoshii and Asano [1972] the crustal thickness is about 30 km throughout the land area in the Tohoku district. This yields almost constant 3He/•He ratios throughout the back arc region.

Considering the constancy of the 3He/'•He and 87Sr/a6Sr ratios in mid-ocean ridges, except for the Tristan component

SANO AND WAKITA: T•-IE 3HE/'•HE RATIOS IN JAPAN 8739

14

12

? 10 o

x 8-

3:: 6-

3:: 4-

2-

xx MORB

back arc

frontal arc

I i I 0.7020 0.7030 0.7040

87Sr/S6Sr

"" ,, Radiogenic • 0.7050 0.7060

Fig. 11. Correlation between the 3He/'•He and 87Sr/86Sr ratios in MORB's, back arc region, and frontal arc region of NE Japan. The dashed line shows the best fit of these three regions.

[Kurz et al., 1982], the oceanic upper mantle may be homoge- neous relative to He and Sr isotopes on a global scale. Even though the pristine mantle materials in the subduct(on zone are composed of MORB-type rocks, we should expect an addition of radiogenic materials and a seawater signal associated with a subducting slab in both the volcanic front and back arc regions. If we take the 3He/4He and 87Sr/86Sr ratios of the magma source in the forearc region (f), back arc region (b) and mid-ocean ridges (mor) as (3He/4He)•.= 5.0 x 10 -6, (87Sr/86Sr)•c = 0.70415, (3He/½He)b = 9.0 x 10 -6, (87Sr/S6Sr)b = 0.70330, (3He/½He),•or = 1.2 x 10 -5, (87Sr/86Sr),•or = 0.70255, respectively, these three regions are on a line in the 3He/½He-87Sr/86Sr diagram (Figure 11). The fit line of these three points is described by the following'

(87Sr/86Sr) = -250 x (3He/½He) + 0.7055

This suggests the presence of an additional crustal component. Although the variations of the 3He/4He and 87Sr/S6Sr ratios in the above three components are taken into account, the end-member may contain helium with the 3He/½He ratio lower than 5 x 10-7 and strontium with the 87Sr/86Sr ratio of 0.7046-0.7064. What kind of material accounts for the end- member?

Recently, Brown et al. [1982] presented possible evidence for sediment recycling in the subduct(on zone based on •øBe measurement. Descending of sediments along the convergent plate margin is also supported by studies through the Deep Sea Drilling Projects Iron Huene and Uyeda, 1981]. The 3He/½He ratios of pore water in the trench sediment (Nankai Trough and Japan Trench) were radiogenic and seawater values of (2-3)x 10 -7 [Sano and Wakita, 1985]. The 87Sr/S6Sr ratios in graywackes from trench sediments in New Zealand ranged from 0.704 to 0.708 JEwart and Stipp, 1968]. These values agree well with the expected component, with 3He/•He ratios less than 5 x 10 -7 and 87Sr/86Sr ratios of 0.705. Thus descending sediments seem to be the most prob- able source of crustal helium and strontium. Fluctuations in

the 3He/½He ratios observed in the volcanic arc region may be attributed to the extent of sediment contamination. Radio-

genic contributions due to subducted sediments will be pro- gressively smaller in the source region of the magma west of the volcanic arc than on the volcanic front.

5.4. The 3He/4He Signature in SW Japan

A clear contrast of 3He/½He ratio distribution is found at the Kinki-Chugoku district in SW Japan (Figure 6). Even in

the frontal arc region, some samples show significantly higher 3He/•He ratios than that of the atmosphere. Figure 12 shows the tendency of 3He/•He ratios of the district in SW Japan. The profile agrees with the distribution of terrestrial heat flow data in SW Japan, that is, anomalously high value of 70-130 mW m-2 were reported in the frontal arc region [Yamano et al., 1984]. Long-continuing swarm activities of shallow earthquakes have been observed in and around Kii Peninsula in the frontal arc region [Mizoue et al., 1983], and there are high-temperature hot springs in the region [Geological Survey of Japan, 1982]. Gravity anomaly, lack of Wadati-Benioff zone, and heat flow anomaly observed in SW Japan indicate that the young and warm lithosphere (Shikoku basin) is subducting beneath the Eurasian plate [Klein et al., 1978; $hiono, 1982; Yamano et al., 1984; Sugi and Uyeda, 1984].

In order to illuminate the complexity of geophysical data of SW Japan, the observed 3He/½He ratios are helpful. If we clarify the mechanism of high 3He/4He ratios in the' frontal arc region, it is possible to give some constraints on the geotectonic structures of SW Japan. We propose two possible hypotheses to explain the high 3He/½He ratios. One is past volcanism on the trench side of the present volcanic front. The other is newly developing volcanism in the frontal arc region. Either case requires the moving of the volcanic front (Figure 12). According to Matsuda and Uyeda [1970] the volcanic front has migrated about 200 km northward during the past 10 m.y. in SW Japan. If we draw the volcanic front about 200 km south of the present front, all the sampling sites • in this study are within the volcanic arc region. Thus high 3He/½He ratios found in the present frontal arc region may be due to the magmatic He of volcanism that occurred at 10 Ma. If so, we need to calculate the accumulation of the radiogenic helium during the past 10 m.y. Torgersen and Jenkins [1982]

SW Japan (Kinki district)

lO.O

5.0

VF

!

!

', ß I

I

!

I

i i

Trough

I

Fig. 12.

5O

lOO

15o

moving VF. .....

7. -' .....

--Primordial He

- I I I I i I

200 100 0 100 200 300

Distance from the present volcanic front (km)

Schematic diagrams of the 3He/'•He profile and tectonic structure of the Kink( district in SW Japan.

8740 SANO AND WAKITA.' Tim 3HE/4HE RATIOS IN JAPAN

estimated the decrease in the 3He/'•He ratios based on the aging of the magma. A similar calculation is made in this study.

The highest 3He/'•He ratios, 9.22 x 10 -6 and 8.15 x 10 -6 (samples 94 and 96), are observed in the frontal arc region (shown in Figure 6 as the "Kinki spot"). These hot springs are located on the northern slope of the Rokko Mountains of Upper Cretaceous granitic rock, which intruded into Arima rhyorites of similar age ['Kasarna, 1968]. In order to determine whether or not the residual magmatic fluid is a principal source of the high 3He/'•He ratios, we calculated the change in 3He/'•He ratios with time. Several assumptions are made: (1) U and Th contents in Rokko granite are 5 and 20 ppm, respectively, (2) initial contents of 3He and '•He in the magma are 5 x 10- TM and 5 x 10 -6 cm 3 STP/g, respectively, based on data from submarine basalt glass [Ozima and Zashu, 1983], (3) the age of the Rokko granite is 60 Ma, and (4) the system is closed; that is, no addition or release of helium has occurred. The calculation indicates that radiogenic production of 3He and '•He are 3.5 x 10-•2 and 2.3 x 10 -'• cm • STP/g, respectively, and that the 3He/'•He ratio of the Rokko granite would be 2.3 x 10-7 at the present time. This value is extremely low compared to the observed ratios of 9.22 x 10 -6 and 8.15 X 10 -6. Even if the initial contents of 3He and '•He in the

magma were 40 times higher than the assumed values, which seems like a safe maximum for the granite, the calculated ratio would become 4.1 x 10 -6, that is, still 2 times lower than the observed ratios. This may imply that either there was an additional magma source which intruded after the Upper Cre- taceous or the initial 3He/'•He ratio of Rokko granite was significantly higher than the present subduction-type He.

The higher 3He/'•He ratios, about 5 times that of the atmosphere, were also found in the frontal arc region of SW Japan (Figure 6). These sites are located in the Kumano acidic rocks of 12 Ma. Similar calculation as done for Rokko granite was also made for this case and led to the conclusion that past volcanism can not explain the high 3He/'•He ratios found in the frontal arc region in $W Japan.

Alternatively, newly developing volcanism in the frontal arc region is a possible mechanism to explain the 3He/'•He profiles of SW Japan. Significant swarm activities of shallow earthquakes have been continuing around the Kumano area in district E within a layer 3-10 km in depth with the largest earthquakes around M 5.0 [Mizoue et al., 1983]. The distribution of these earthquakes overlaps with the spotlike area in Figure 6. These earthquakes might be attributed to the me- chanical deformation of the crust caused by the uprising magma. Wakita et al. [1978] found a "helium spot," where a significant amount of He is present in the soil (up to 350 ppm with 3He/'•He ratio of 8.90 x 10 -6) along the fault zone formed by the 1966 Matsushiro swarm earthquakes. They interpreted the formation of the "helium spot" and the oc- currence of the earthquake swarms as the result of a diapiric uprise of a magma approximately 1 km in diameter. If the earthquakes around the Kumano region are accounted for by the magma uprise, the observed high 3He/'•He ratios may be associated with subsurface magmatic activities.

Unlike in the Kinki and Chugoku district, a clear contrast in the 3He/'•He ratio is observed in the Kyushu district (Figure 7). Although the numbers of observed data are small, the 3He/'•He profile in the district is in good agreement with that of NE Japan rather than SW Japan. The tendency may be due to the similar tectonic setting of the Kyushu-Ryukyu arc and the NE Japan arc. The subducting plate is old, cold,

and thick [Shih, 1980; Seno and Maruyama, 1984], resembling the Pacific plate, with similar geographical characteristics, thrusting beneath NE Japan. Contour lines of deep seismic plane show clear differences between NE Japan arc and the Kinki and Chugoku district in SW Japan. Thus the same mechanism as described above for NE Japan may account for the 3He/'•He profile of the Kyushu district.

6. CONCLUSION

The 3He/'•He ratios of gaseous samples well reflect the fine geotectonic structures of the Japanese Islands. In NE Japan there is a clear contrast of the 3He/'•He ratios between samples in the frontal arc and volcanic arc regions. This suggests that the uprising magma is a unique material which brings primordial 3He from the upper mantle (Figure 10). Many volcanoes and geothermal systems in the volcanic arc region are interpreted as caused by diapiric uprise of magma containing mantle He. In the frontal arc region there is no Quaternary volcanism, and the absence of the carrier of the mantle He is suggested. Radiogenic 3He/'•He in the region may be due to decay of U and Th in the crustal rocks, sedimentary layers, and seawater-atmospheric contamination. In the volcanic arc region the maximum 3He/'•He ratios become gradually higher backward from the volcanic front. This tendency may be due to the effect of subduction materials containing the radiogenic He.

In SW Japan, there is no clear contrast of the 3He/'•He ratios (Figure 12) except at Chubu and Kyushu districts. Con- siderably high 3He/'•He ratios observed in the frontal arc region at Kinki district in SW Japan may be attributed to subsurface and renewed magmatism associated with subduction of the young and warm lithosphere (Shikoku basin) beneath SW Japan.

Acknowledgments. We would like to express our sincere appreci- ation to K. Kobayashi (Ocean Research Institute), K. Nakamura, and S. Uyeda (Earthquake Research Institute), University of Tokyo, for critical reading and valuable comments on the original manuscript. Discussions with Y. Nakamura, University of Tokyo, and K. Notsu and Y. Kobayashi, Tsukuba University, and their help in field work were quite useful. We would like to express our gratitude to M. Ozima, T. Tominaga, and I. Kaneoka, University of Tokyo, for their helpful discussions; A. Urabe for help in the field; and K. Itou for drawing the figures. We thank M. Kumazawa and M. Mizoue, Uni- versity of Tokyo, for comments on preparing the revised manuscript. We also thank an associate editor and two anonymous reviewers of this paper for their valuable comments and suggestions, which were useful for improving the paper.

REFERENCES

Baskov, Y. A., U. Y. Vetshteyu, $. N. Surikov, I. N. Tolstikhin, G. A. Malyuk, and T. A. Mishina, Isotope composition of H, O, C, Ar and He in hot springs and gases in the Kurile-Kamchatka volcanic region and indicators of formation conditions, Geokhimiya, 1973, 180-189, 1973.

Brown, L., J. Klein, R. Middleton, I. $. Sacks, and F. Tora, 1øBe in island-arc volcanoes and implications for subduction, Nature, 299, 718-720, 1982.

Clarke, W. B., W. J. Jenkins, and Z. Top, Determination of tritium by mass spectrometric measurement of 3He, Int. J. Appl. Radiat. lsot., 27, 515-522, 1976.

Craig, H., J. E. Lupton, and Y. Horibe, A mantle helium component in Circum-Pa½ifi½ volcanic gases: Hakone, the Marianas and Mt. Lassen, in Terrestrial Rare Gases, edited by E. C. Alexander and M. Ozima, pp. 3-16, Center for Academic Publishing Japan, Tokyo, 1978.

Ewart, A., and J. J. Stipp, Petrogenesis of the volcanic rocks of the central North Island, New Zealand, as indicated by a study of

SANO AND WAKITA: Tm• 3HE/'•HE RATIOS IN JAPAN 8741

87Sr/86Sr ratios, and Sr, Rb, K, U and Th abundances, Geochim. Cosmochim. Acta, 32, 699-736, 1968.

Geological Survey of Japan, Distribution map of heat discharge by hot springs, in Geological Atlas of Japan, pp. 94-101, Tsukuba, 1982.

Gerling, E. K., B. A. Mamyrin, I. N. Tolstikhin, and S.S. Yakovleva, Isotope composition of helium in rocks, Geokhimiya, 1971, 608-617, 1971.

Kasama, T., Granitic rocks of the Rokko mountains, Kinki district, Japan, J. Geol. Soc. Jpn., 74, 147-158, 1968.

Klein, G. deV., et al., Offridge volcanism and seafloor spreading in the Shikoku Basin, Nature, 273, 746-748, 1978.

Kurz, M.D., W. J. Jenkins, and S. R. Hart, Helium isotopic system- atics of oceanic islands and mantle heterogenety, Nature, 297, 43-47, 1982.

Lupton, J. E., Terrestrial inerts gases: Isotope tracer studies and clues to primordial components in the mantle, Annu. Rev. Earth Planet. Sci., 11, 371-414, 1983.

Mamyrin, B. A., and I. N. Tolstikhin, Helium Isotopes in Nature, p. 97, Elsevier, New York, 1984.

Mamyrin, B. A., G. S. Anufriyev, I. L. Kamenskiy, and I. N. Tolsti- khin, Determination of the isotopic composition of atmospheric helium, Geokhimiya, 1970, 721-730, 1970.

Matsuda, T., and S. Uyeda, On the Pacific-type orogeny and its model-extension of the paired belts concept and possible origin of marginal seas, Tectonophysics, 11, 5-27, 1970.

Mizoue, M., M. Nakamura, N. Seto, Y. Ishiketa, and T. Yokota, Three-layered distribution of microearthquakes in relation to focal mechanism variation in the Kii Peninsula, southwestern Honshu, Japan, Bull. Earthquake .Res. Inst. Univ. Tokyo, 58, 287-310, 1983.

Morrison, P., and J. Pine, Radiogenic origin of the helium isotopes in rock, Ann. N.Y. Acad. Sci., 62, 71-92, 1955.

Nagao, K., N. Takaoka, and O. Matsubayashi, Rare gas isotopic compositions in natural gases of Japan, Earth Planet. Sci. Lett., 53, 175-188, 1981.

Notsu, K., Strontium isotope composition in volcanic rocks from the Northeast Japan Arc, J. Volcanol. Geotherm. .Res., 18, 531-548, 1983.

Ohsumi, T., and Y. Horibe, Diffusivity of He and Ar in deep-sea sediments, Earth Planet. Sci. Lett., 70, 61-68, 1984.

Ozima, M., and S. Zashu, Noble gases in submarine pillow volcanic glasses, Earth Planet. Sci. Lett., 62, 24-40, 1983.

Rison, W., Isotopic studies of the rare gases in igneous rocks: Impli- cations for the mantle and atmosphere, Ph.D. thesis, Univ. of Calif., Berkeley, 1980.

Rison, W., and H. Craig, Helium isotopes and mantle volatiles in Loihi seamount and Hawaiian Island basalts and xenoliths, Earth Planet. Sci. Lett., 66, 407-426, 1983.

Sano, Y., Distribution of 3He/4He ratios and geotectonic structure of the Japanese Islands, Ph.D. thesis, Univ. of Tokyo, Tokyo, 1983.

Sano, Y., and H. Wakita, 3He/aHe ratios of pore gases in pore water, sites 583 and 584, Initial Rep. Deep Sea Drill. Proj., in press, 1985.

Sano, Y., T. Tominaga, Y. Nakamura, and H. Wakita, 3He/•He ratios of methane-rich natural gases in Japan, Geochem. J., 16, 237-245, 1982.

Sano, Y., Y. Nakamura, H. Wakita, A. Urabe, and T. Tominaga, 3He emission related to volcanic activity, Science, 224, 150-151, 1984.

Sano, Y., Y. Nakamura, and H. Wakita, Areal distribution of 3He/aHe ratios in the Tohoku district, northeastern Japan, Chem. Geol. Isot. Geosci. Sect., 52, 1-8, 1985.

Seno, T., and S. Maruyama, Paleogeographic reconstruction and origin of the Philippine sea, Tectonophysics, 102, 53-84, 1984.

Shih, T. C., Marine magnetic anomalies from the western Philippine sea: Implications for the evolution of marginal basins, in The Tec- tonic and Geologic Evolution of Southeast Asian Seas and Islands, Part 1, Geophys. Monogr. Ser., vol. 23, edited by D. E. Hayes, pp. 49-75, AGU, Washington, D.C., 1980.

Shiono, K., Frontal shape of the seismic slab by normal subduction of a dead spreading axis and a preliminary application to Southwest Japan, J. Geosci. Osaka City Univ., 25, 19-33, 1982.

Shiono, K., and N. Sugi, Life of oceanic plate: Cooling time and assimilation time, Tectonophysics, 112, 35-50, 1985.

Sugi, N., and S. Uyeda, Subduction of young oceanic plates without deep focus earthquakes, Bull. Soc. Geol. Fr., 26, 245-254, 1984.

Torgersen, T., and W. J. Jenkins, Helium isotopes in geothermal systems: Iceland, the Geysir, Raft River and Steamboat Springs, Ge- ochim. Cosmochim. Acta, 46, 739-748, 1982.

Torgersen, T., J. E. Lupton, D. Sheppard, and W. Giggenback, He isotope variations in the thermal areas of New Zealand, J. Vol- canol. Geotherm. .Res., 12, 283-298, 1982.

Urabe, A., T. Tominaga, Y. Nakamura, and H. Wakita, Chemical compositions of natural gases in Japan, Geochem. J., 19, 11-25, 1985.

von Huene, R., and S. Uyeda, A summary of results from IPOD transects across the Japan, Mariana and Middle-America convergent margin, Oceanol. Acta, 1981, 233-239, 1981.

Wakita, H., and Y. Sano, 3He/4He ratios in CH•-rich natural gases suggest magmatic origin, Nature, 305, 792-794, 1978.

Wakita, H., N. Fujii, S. Matsuo, K. Notsu, K. Nagao, and N. Ta- kaoka, "Helium spots": Caused by a diapiric magma from the upper mantle, Science, 200, 430-432, 1983.

Yamano, M., S. Honda, and S. Uyeda, Nankai Trough--As a hot trench ?, Mar. Geophys..Res., 6, 187-203, 1984.

Yoshii, T., Crust and upper-mantle structure beneath northeastern Japan (in Japanese), Kagaku Kyoto, 47, 170-176, 1977.

Yoshii, T., and S. Asano, Time-term analyses of explosion seismic data, J. Phys. Earth, 20, 47-57, 1972.

Yuhara, K., and K. Seno, Geology, Geophysics and Geochemistry of Hot and Mineral Springs (in Japanese), 287 pp., Chijin-Shokan, Tokyo, 1969.

Y. Sano and H. Wakita, Laboratory for Earthquake Chemistry, Faculty of Science, University of Tokyo, Bunkyo-ku, Tokyo 113, Japan.

(Received May 30, 1984; revised April 15, 1985; accepted May 8, 1985.)

geographical distribution of 3 he& 4 he ratios in japan ... · journal of geophysical research,...

Documents