Helium and carbon isotopes in the hot springs of Changbaishan Volcano, northeastern China: A material connection between Changbaishan Volcano and the west Pacific plate?

    Helium and carbon isotopes in the hot springs of Changbaishan volcano, northeastern China: A material connection between Changbaishan volcano and the West Pacific plate? Feixiang Wei, Jiandong Xu, Zhiguan Shangguan, Bo Pan, Hongmei Yu, Wei Wei, Xiang Bai, Zhengquan Chen PII: DOI: Reference:

S0377-0273(16)30336-5 doi:10.1016/j.jvolgeores.2016.09.005 VOLGEO 5920

To appear in:

Journal of Volcanology and Geothermal Research

Received date: Revised date: Accepted date:

9 April 2016 13 September 2016 15 September 2016

Please cite this article as: Wei, Feixiang, Xu, Jiandong, Shangguan, Zhiguan, Pan, Bo, Yu, Hongmei, Wei, Wei, Bai, Xiang, Chen, Zhengquan, Helium and carbon isotopes in the hot springs of Changbaishan volcano, northeastern China: A material connection between Changbaishan volcano and the West Pacific plate?, Journal of Volcanology and Geothermal Research (2016), doi:10.1016/j.jvolgeores.2016.09.005

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT Helium and carbon isotopes in the hot springs of Changbaishan Volcano, northeastern China: A material connection between Changbaishan Volcano and

IP

T

the west Pacific plate?

SC R

Feixiang Wei, Jiandong Xu*, Zhiguan Shangguan, Bo Pan, Hongmei Yu, Wei Wei, Xiang Bai, Zhengquan Chen

Key Laboratory of Active Tectonics and Volcano, Institute of Geology, China

NU

Earthquake Administration, Beijing 100029, China

MA

Changbaishan Volcano is located in northeastern China, approximately 1400 km west of the west Pacific subduction zone. Although the west Pacific plate and

D

Changbaishan Volcano are spatially associated with each other, no previous evidence

TE

has demonstrated the existence of a direct material connection between the two. In

CE P

this study, we utilize helium (3He/4He, CO2/3He) and carbon isotopes (δ13C) from the hot springs of Changbaishan Volcano to exclude the possibility of a direct material

AC

connection between the volcano and the west Pacific plate at source. A total of 22 gas samples were collected from three hot springs at Changbaishan Volcano in 2002, 2006, 2014 and 2015; isotopic and geochemical analyses were performed on these samples to trace the possible sources of these gases. Our analysis reveals that values for air-corrected 3He/4He ratios range from 3.98 RA to 6.03 RA (where RA represents the atmospheric 3He/4He ratio), CO2/3He ratios vary from 2.20 × 108 to 1.92 × 1011, and δ13C values vary from −7.9‰ to −1.6‰. By comparing these measured values to those of typical mantle and crustal sources, we can infer that hot spring gases from Changbaishan Volcano are mostly characterized

ACCEPTED MANUSCRIPT by inputs from two isotopically distinct sources: deep mantle fluids and shallower, slab-derived fluids. Fluids liberated from the shallower magma chamber are likely to

IP

T

include ancient Izanagi subduction zone fluids, whereas fluids originating from

SC R

deeper magma chamber likely consist of MORB-like asthenospheric mantle fluids. Based on these results, we suggest that helium and carbon isotopes in hot springs

Volcano and the west Pacific plate.

NU

demonstrate the absence of a direct material connection between Changbaishan

MA

Keywords: Helium isotopes; Carbon isotopes; CO2/3He; West Pacific subduction; Changbaishan Volcano

D

1. Introduction

TE

Changbaishan Volcano (also known as Mt. Paektu and Tianchi Volcano) is one

CE P

of the most active intraplate volcanoes in China (Xu et al., 2012). Unlike volcanoes within Japan island arc, which are located close to the Japan Trench in the west

AC

Pacific subduction zone, Changbaishan Volcano is located ~1400 km west of the same subduction zone. Many researchers have attempted to characterize the source of this volcano using multidisciplinary techniques, including seismic tomographic, geochemical and tectonic methods (e.g., Wang et al., 2000; Tang et al., 2006; Zhao and Liu,2010; Zou and Fan, 2011; Wei et al., 2012). Several studies have demonstrated that the west Pacific plate has subducted beneath northeastern China, forming a wide upper mantle wedge above the west Pacific stagnant slab (WPSS) (e.g., Zhao and Liu, 2010; Wei et al.,2012). According to these studies, the west Pacific plate should exert a strong geodynamic influence on Changbaishan Volcano.

ACCEPTED MANUSCRIPT However, the presence or absence of a material connection between the WPSS and Changbaishan Volcano is still a controversial topic (e.g., Wang et al., 2000; Tang et

IP

T

al., 2006; Chen et al., 2007; Hahm et al., 2008; Kuritani et al., 2011; Zou and Fan,

SC R

2011).

Gases emitted from hot springs have received a great deal of attention in recent years, as they can be utilized as sensitive tracers of seismic and volcanic activities

NU

(Gautheron and Moreira, 2002; Burgisser and Scaillet, 2007; Morikawa et al., 2008;

MA

Barry et al., 2013). In the Changbaishan volcanic field, hot springs are widely distributed and continually emitting gases, resulting in the emission of approximately

D

7.79 × 105 t of CO2 into the atmosphere every year (Guo et al., 2014). Among the hot

TE

spring gases, helium can be used as a sensitive geochemical indicator for volatile

CE P

provenance due to its inertness and low solubility (Padron et al., 2012). It is well documented that most 4He is produced by radioactive decay within the crust and that

AC

most 3He is still escaping from the Earth’s interior (Morrison and Pine, 1955; Hilton, 1996; Padron et al., 2012; Zelenski et al., 2012). Helium isotopes have been widely used to trace magma genesis and volcanic evolutions in different tectonic settings, as well as to monitor earthquakes and volcanoes (e.g., Kennedy and Van Soest, 2006, 2007; Ohno et al., 2011; Ohwada et al., 2012; Padron et al., 2012, 2013). When supplemented with CO2 chemical and isotopic composition data, helium isotopic data can also be used to compute the percentages of mantle and slab-derived fluids in magmatic mixtures (Sano and Marty, 1995; Sano et al., 2006). This work presents the study of a total of 22 hot spring gas samples collected in

ACCEPTED MANUSCRIPT 2002, 2006, 2014 and 2015. Our study aims were the following: 1) to study the sources of hot spring gases from three hot springs at Changbaishan Volcano; 2) to analyze the

IP

T

upwelling processes of hot spring gases from the source to the surface based on the

SC R

structure of this volcano; 3) to reveal the possible connection between Changbaishan Volcano and the WPSS by comparing their chemical and isotopic compositions with 3

He/4He, δ13C and CO2/3He values of mantle and crustal sources (Marty and Jambon,

NU

1987; Sano and Marty, 1995; Gautheron and Moreira, 2002; Graham, 2002; Hilton et al.,

MA

2002; Shaw et al., 2006).

D

2. Regional Setting

TE

Changbaishan Volcano, on the boundary between China and North Korea (42°00′

CE P

N，128°03′ E), is located ~1400 km west of the west Pacific subduction zone on 30to 39-km-thick continental crust (Hetland et al., 2004; Li et al., 2006; Kyong-Song et

AC

al., 2016). Global and regional tomographic studies (e.g., Zhao and Liu, 2010; Wei et al., 2012; Tang et al., 2014) have suggested that the west Pacific plate has subducted beneath northeastern China at a dip angle of ~30º and that the subducting Pacific slab has become stagnant in the mantle transition zone; these studies also indicated that a wide mantle wedge may have formed above the WPPS. Because the upwelling of asthenospheric material is known to influence the evolution of back-arc intraplate volcanoes in northeast Asia, including Changbaishan Volcano and Wudalianchi Volcano (Xu et al., 2013b), it is widely accepted that the WPSS exerts a strong geodynamic influence on Changbaishan Volcano (e.g., Zhao et al., 2009; Zhao and

ACCEPTED MANUSCRIPT Liu, 2010; Wei et al., 2012; Tang et al., 2014). Changbaishan Volcano has undergone three stages of evolution: the early basalt

IP

T

shield stage, the middle trachyte composite cone stage and the late ignimbrite-forming

SC R

stage (Wei et al., 2007). The most recent explosive eruption of Changbaishan Volcano was the Millennium eruption, which occurred in A.D. 946; this eruption is believed to have been one of the largest eruptions of the past two thousand years in the world (Xu

NU

et al., 2013a). Written records also suggest that Changbaishan Volcano experienced at

MA

least three small recent eruptions following the Millennium eruption in A.D. 1688, A.D. 1702 and A.D. 1903. Currently, the highest point on the volcano is 2749 m

D

above sea level, and the Tianchi caldera lake, which is roughly circular with a

TE

maximum diameter of ~5 km, contains approximately 2.04 billion m3 of water (Fan et

CE P

al., 2007). During a recent period of magmatic unrest from 2002 to 2006, the number of earthquakes at the volcanic center increased by as much as two orders of

AC

magnitude, ground inflation was observed, and volcanic gases recorded increases in their CO2, H2 and He contents and 3He/4He values (Xu et al., 2012). To evaluate the risks and potential hazards posed by this volcano to nearby populations in East Asia, it is necessary to understand the mechanisms responsible for its eruptions and general unrest (Xu et al., 2012; Yu et al., 2013). Within the volcanic field, large-scale continuous geothermal activities, along with strong gas release, are mainly located in three hot spring areas: Julong hot spring (on the north flank of the region), Jinjiang hot spring (on the southwest flank of the region) and Hubin hot spring (on the shoreline of the caldera lake). Julong hot spring

ACCEPTED MANUSCRIPT lies to the north of Changbaishan waterfall and has a degassing area of approximately 3300 m2. Jinjiang hot spring, which is located beside the Jinjiang River, has a

IP

T

degassing area of approximately 1300 m2. Hubin hot spring lies along the lakeshore of

SC R

Tianchi caldera lake and is approximately 500 m in length.

3. Sampling and analytical methods

NU

Gas samples were collected at the three hot spring areas using water

MA

displacement methods to avoid air contamination. To obtain samples for chemical composition analysis, a funnel (with a mouth ~18 cm in diameter) was submerged in

D

spring water, and a hand-pump was used to purge the sampling equipment. The

TE

inverted funnel was then placed over the upwelling bubbles to collect gases. After the

CE P

gases were collected, a syringe was used to inject them into 500-ml pre-vacuumed aluminum bags whose lids were refilled with high-density silicone rubber. Gas

AC

samples for isotopic composition analysis were similarly collected in 125-ml inverted glass bottles under spring water. Helium and CO2 contents were analysed with an Agilent 6890N gas chromatograph (with a routine precision of 5%) using argon as a carrier gas. Analysis of isotopic compositions was conducted at Key Laboratory of Petroleum Resources Research, Institute of Geology and Geophysics, Chinese Academy of Sciences. Isotopic compositions of helium and neon were measured by a VG-5400 mass spectrometric system (in 2002 and 2006) and a Nu Instruments Noblesse SFT noble gas mass spectrometer (in 2014 and 2015). The precision of He and Ne isotopic

ACCEPTED MANUSCRIPT compositions obtained from repeated measurements of an air standard was less than 7% and 10%, respectively. The isotopic composition of carbon was analyzed by a

IP

T

Thermo Finnigan MAT252 stable isotope mass spectrometer system (in 2002 and

SC R

2006) and a Thermo Finnigan MAT 253 stable isotope mass spectrometer system (in 2014 and 2015). The analytical accuracy of δ13C was within 0.6‰ (2σ) with respect

NU

to the Pee Dee Belemnite (PDB) standard.

MA

4. Results

In this study, we measured the chemical and isotopic compositions of helium and

D

carbon of 22 samples from Changbaishan Volcano in 2002, 2006, 2014 and 2015.

TE

These results are listed in Table 1.

CE P

Of the volcanic gases at Changbaishan Volcano, the most abundant content is CO2, which varies in its abundance between 70.2% and 99.7%, with an average

AC

content of 91.8%. CO2 is also the major volcanic gas component at other volcanoes in China, such as the Wudalianchi intraplate volcano (in Fig. 1) in northeastern China (Xu. et al., 2013b) and the Tengchong volcanic field in southwestern China (Zhao et al., 2012). The relative abundance of helium in the volcanic gas samples ranged from 0.9 to 405 ppm, with an average value of 122 ppm.

4.1 3He/4He ratios The measured 3He/4He ratios (RM/RA), which are reported relative to atmospheric 3

He/4He (RA) (where RA ≈ 1.4 × 10−6; Clarke et al., 1976), are corrected for the effects

ACCEPTED MANUSCRIPT of atmospheric contamination and different solubilities of helium and neon by the ratios of their Bunsen coefficients and He/Ne ratios (Hilton, 1996). The corrected He/4He ratios (RC/RA) are expressed by the following equation:

IP

T

3

RC/RA = [(RM/RA × X) – 1]/(X – 1)

SC R

(1)

Considering that all our samples were collected from hot spring bubbles, the variable X can be calculated by the following equation:

NU

X = [(4He/20Ne)measured/(4He/20Ne)air] × (βNe/βHe)

(2)

MA

where βNe and βHe are the Bunsen solubility coefficients for neon and helium, respectively, and (4He/20Ne)measured and (4He/20Ne)air are the helium/neon ratios of

D

measured samples and air (0.3185; Ozima and Podosek, 2002), respectively.

TE

It is assumed that the salinity of hot spring water at Changbaishan is zero.

such:

CE P

According to Weiss (1971), the Bunsen solubility coefficient (β) can be calculated as

(3)

AC

ln β = A1 + A2(100/T) + A3ln (T/100)

For helium, A1 = -34.6261, A2 = 43.0285, and A3 = 14.1391; for neon, A1 = -39.1971, A2 = 51.8013, and A3 = 15.7699. Thus, the Bunsen solubility coefficient ratio is given as: βNe/βHe = (T/100)1.6308 × exp(877.28/T – 4.571)

(4)

where T is the temperature of the hot spring water. The air-corrected 3He/4He ratios (RC/RA) of samples from Changbaishan Volcano varied from 3.98 RA (Julong) to 6.03 RA (Hubin) (Fig. 2a), and all the samples yielded values falling within the typical range seen in volcanic arcs worldwide (5.4 ± 1.9 RA;

ACCEPTED MANUSCRIPT Hilton et al., 2002). At Julong hot spring, the air-corrected 3He/4He ratios ranged from 3.98 to 5.93 RA; at Jinjiang and Hubin hot spring, the air-corrected 3He/4He ratios

SC R

IP

T

ranged from 4.79 to 4.96 RA and from 5.12 to 6.03 RA, respectively.

4.2 δ13C (CO2) values

The δ13C values of CO2 collected at Changbaishan Volcano ranged from −7.3‰

NU

to −1.6‰, which are comparable to the range normally found in MORB (−6.5 ± 2.5‰

MA

in Fig. 2b; Sano and Marty, 1995). Of the three hot springs, Hubin hot spring yielded the highest δ13C values of CO2, which varied from -4.9‰ to -1.6‰. The δ13C values

D

of CO2 ranged from -6.2‰ to -2.9‰ at Julong hot spring. δ13C values of CO2 at

CE P

TE

Jinjiang hot spring were the lowest of the three sites and ranged from -7.9‰ to -5.3‰.

4.3 CO2/3He ratios

AC

The CO2/3He ratio is a useful parameter for identifying the sources of volcanic gases (Ruzie et al., 2013) because this ratio exhibits a small range as compared to other crustal fluids (Ballentine et al., 2001). The corrected helium concentration ([He]C) is calculated by: [He]C = [He]M × (X – 1)/X

(5)

where [He]M is the measured helium content and X is defined in Eq. (2). Corrected CO2/3He ratios of the samples from Changbaishan Volcano were highly variable, ranging from 2.20 × 108 (Hubin) to 1.92 × 1011 (Julong) (Fig. 2c). At Julong hot spring, the CO2/3He ratios ranged from 4.56 × 109 and 1.92 × 1011; the

ACCEPTED MANUSCRIPT CO2/3He ratios ranged from 2.20 × 108 to 6.34 × 109 at Hubin hot spring and from

IP

T

1.08 × 109 to 2.89 × 109 at Jinjiang hot spring.

SC R

5. Discussion

In the following discussion, we analyze the origin of helium and CO2 in these hot spring gases by comparing their 3He/4He ratios, CO2/3He ratios and δ13C of CO2

NU

values with those of typical mantle source and crustal sources. We also discuss the

MA

processes by which these hot spring gases are transferred from their sources to the Earth’s surface and examine the connection between Changbaishan Volcano and the

CE P

5.1 Source of helium

TE

D

WPSS.

According to previous studies (Hilton et al., 2002; Sano et al., 2006), different

AC

magmatic sources possess distinct helium isotope ratios. The 3He/4He values of mantle, crustal and atmospheric sources are 8 RA (Graham, 2002; Shaw et al., 2006), 0.05 RA (Gautheron and Moreira, 2002) and 1 RA, respectively. The 3He/4He values of xenolithic olivines measured by Kim et al. (2005) and Hahm et al. (2008) range from 3.30 RA to 7.71 RA (Table 2). All xenolith helium ratios were obtained using the crushing method, rather than the heating method, in order to obtain a helium isotopic ratio that is more representative of its source value (Kim et al., 2005; Chen et al., 2007). Table 2 presents the 3He/4He values of these olivine crystals; they correlate positively with helium abundance. The lowest helium isotopic value (3.30 RA) is

ACCEPTED MANUSCRIPT likely a result of its low helium abundance (0.84 ncm3 STP/g) and high degree of atmospheric contamination ((4He/20Ne)/(4He/20Ne)air = 58); in constrast, the sample

IP

T

with the highest helium isotopic ratio (XPD3, 3He/4He = 7.71 RA) has the highest

SC R

helium content (5.6 ncm3/g STP) and the lowest degree of air contamination ((4He/20Ne)/(4He/20Ne)air = 2229), which may represent the deepest manlte source of Changbaishan Volcano.

NU

To evaluate the source of helium in the gases from the three hot springs at

MA

Changbaishan Volcano, we plotted data from all 22 hot spring samples (Table 1) and 4 xenolith samples (olivine mineral grains in Table 2) on a RM/RA versus 4He/20Ne

D

diagram (Fig. 3). In this figure, the top and the bottom lines represent the typical

TE

mixing trend between air and mantle and crustal endmembers, with 3He/4He ratios of

CE P

7.71 RA and 0.05 RA, respectively. Mixing curves displaying different degrees of crustal contamination are also displayed.

AC

As shown in Fig. 3, all the hot spring samples plot between the 18% and 38% crustal contamination curves, suggesting that their compositions require a significant mantle contribution. This result is in excellent accordance with previous work on geothermal fluids at Changbaishan Volcano, in which Shangguan et al. (1996), Hahm et al. (2008), Shangguan and Wu (2008) and Zhang et al. (2015) reported average helium isotopic ratios of 4.86 RA, 4.89 RA, 5.21 RA, and 4.21 RA, which each require air contamination values of 37%, 37%, 33%, and 46%, respectively. Thus, we conclude that fluids from Changbaishan Volcano have a MORB-like mantle source. It is likely that during the ascent from a deep mantle source to the surface through 30- to

ACCEPTED MANUSCRIPT 39-km-thick continental crust (Hetland et al., 2004; Li et al., 2006; Kyong-Song et al., 2016), the helium in these fluids is contaminated by radiogenic crustal helium (4He).

IP

T

To further assess the origins of gases within these hot spring samples, we can use

SC R

CO2 values to better supplement our helium data. To evaluate the relationship between helium and CO2, we plotted data from all gas samples on a CO2 - 3He - 4He ternary diagram (Fig. 4). This figure shows the influence of CO2 and radiogenic

NU

helium on MORB and island arc sources. In moving from the top to the bottom of this

MA

ternary diagram, samples from Hubin hot spring show a distinct loss of CO2. Although all samples fall in the gray range of arc-related compositions, the CO2/3He

D

values vary significantly, from 2.20 × 108 to 1.92 × 1011, suggesting that fluids from

CE P

TE

the three hot springs may have different carbon sources.

5.2 Source of carbon

AC

To study the source of carbon, we use a CO2 - 3He - δ13C mixing model, first introduced by Sano and Marty (1995), to analyze the contributions of CO2 from different sources. This model is based on the assumption that the carbon values recorded in hot spring samples represent a mixture of carbon contributed from three endmembers, i.e., mantle carbon (M), sedimentary organic carbon (S) and slab-derived limestone carbon (L). Fig. 5 presents CO2/3He versus δ13C for all samples as well as for the three endmembers (M, S and L). The boundary lines of this model are calculated by the following mass balance equations:

ACCEPTED MANUSCRIPT (6)

1/(12C/3He)Obs = M/(12C/3He)MORB + L/(12C/3He)Lim + S/(12C/3He)Sed

(7)

M+S+L=1

(8)

IP

T

(13C/12C)Obs = (13C/12C)MORB × M + (13C/12C)Lim × L + (13C/12C)Sed × S

SC R

where Obs, MORB, Lim and Sed refer to observed, MORB, limestone and sediment values, respectively, and M, S and L are the respective percentages of mantle, sedimentary and marine limestone carbon sources. For this model, we use δ13C values

NU

of -6.5‰, 0‰ and -30‰ (relative to PDB) and 12C/3He values of 1.6 × 109, 1 × 1013

MA

and 1 × 1013 for the MORB, marine limestone and sediment endmember compositions, respectively.

D

As shown in Fig. 5, all the samples from Jinjiang hot spring have compositions

TE

very similar to that of the MORB endmember (M), suggesting a main MORB-like

CE P

source. This result is in good agreement with previous studies on volcanic gases at Changbaishan Volcano (Shangguan et al., 1996; Shangguan and Wu, 2008).

AC

Therefore, based on their MORB-like helium contributions, it is likely that the Jinjiang volcanic gases represent the deepest mantle fluids at Changbaishan Volcano with their slightly modified isotopic values resulted from crustal helium (4He) contamination during the processes of upwelling and transportation to the Earth’s surface. Fig. 5 also shows the distribution of Julong hot spring samples in the model (M>0 & S>0 & L>0); we can calculate the approximate proportions of mantle, sedimentary and slab-derived limestone endmembers in this spring’s samples using the three mass balance equations (Eq. (6), (7) and (8)). The results of these equations are listed in

ACCEPTED MANUSCRIPT Table 3. In general, these results demonstrate that most of the carbon in the samples from Julong hot spring is derived from the subducted slab (from 56% to 87%, with an

IP

T

average of 71%) with only a low fraction of carbon derived from the mantle

SC R

(1%-35%, with an average of 17%). This in turn implies that the hot spring gas samples from the Julong site contain a significant contribution of limestone-derived CO2. This result agrees well with previous work at Julong hot spring, which reported

NU

that hot spring gases from Julong hot spring were slab-derived (Hahm et al., 2008;

MA

Zhang et al., 2015).

At Hubin hot spring, all but one sample are distributed out of the model, due to

D

low CO2/3He ratios. The proportions of the three endmembers cannot be computed for

TE

these samples because they plot outside of the model’s parameter space, resulting in

CE P

negative endmember contributions. This anomaly may be caused by CO2 loss, as Hamh et al. (2008) proposed a fractionation process by which cooling of

AC

hydrothermal fluids triggered calcite precipitation, which lowered the CO2/3He ratios in the remaining fluids at Changbaishan Volcano. However, on one hand, our C-He model shows no significant modification of the CO2/3He ratios of MORB-like mantle fluids during their transportation from the deep mantle to Jinjiang hot spring. On the other hand, the CO2 concentrations of Julong hot spring waters (which register temperatures of 63.3 °C to 75.0 °C) and Jinjiang hot spring waters (which register temperatures of 56.3 °C to 57.9 °C) are 0.96-0.99 g/L and 1.15 g/L, respectively (Shangguan et al., 1996), which implies that both systems are CO2-saturated (see Carroll et al. (1991) and Millero (1995) for details of calculation). Considering the

ACCEPTED MANUSCRIPT high solubility of CO2 in the cold Hubin spring water (which registers temperatures of 13 to 33 °C) and the quick water flow in Tianchi caldera lake, the water at the Hubin

IP

T

site is unlikely to be CO2-saturated. Therefore, the process of CO2 loss may only take

SC R

place at the Hubin site and could occur simply by dissolution of CO2 in the cold spring water. Therefore, if we increase the CO2/3He ratios of the blue squares, we will observe that the blue squares fall between the red triangles and the green circles. This

NU

suggests that it may be possible for the source of Hubin hot spring to be a mixture of

MA

the sources of Julong and Jinjiang hot springs.

The proportions of carbon originating from different sources have also been

D

calculated in Japan island arc, which is located close to the Japan Trench in the same

TE

subduction zone as Changbaishan Volcano (Fig. 6). In the Japan island arc system,

CE P

marine limestone contributes the majority of carbon found in high temperature fumaroles and the mantle contributes less than 20% (Sano and Marty, 1995). In

AC

general, the mantle limestone contribution decreases with increasing distance between the Japan Trench and the volcanoes in the west Pacific subduction zone. However, the average marine limestone contribution at Julong hot spring is 71%, which is similar to that seen in the Japan arc (Sano and Marty, 1995), making it unlikely that the WPSS contributes significantly to the slab-derived fluids of Changbaishan Volcano.

5.3 Transportation model of hot spring gases Because the hot spring gases at Changbaishan Volcano originate from two main sources, we can place structural constraints on the upwelling processes bringing them

ACCEPTED MANUSCRIPT to surface. According to seismic and electromagnetic studies performed to characterize the structure of Changbaishan Volcano (Tang et al., 2001; Zhang et al.,

IP

T

2002; Qiu et al., 2014), there appear to be two magma chambers beneath the volcanic

SC R

center: Julong hot spring is located above a shallower magma chamber (7-15 km in depth), whereas Jinjiang hot spring is connected to a deeper magma chamber (~30 km in depth) directly through a network of deep faults. Hubin hot spring is connected to

NU

the bottom of the shallower chamber through the volcano caldera. This structural

MA

model is supported by a recent seismologic study performed on the North Korean side of the volcano, in which negative peaks in the receiver function at a depth of 5-10 km

D

in the crust were interpreted to reflect a significant region of magma storage in the

TE

shallow crust (Kyong-Song et al., 2016).

CE P

This double-chamber model can be used to explain the provenances of gases emitted from the three hot springs (Fig. 7). In this model, the shallower magma

AC

chamber stores slab-derived fluids, which then migrate upwards directly to Julong hot spring. The deeper magma chamber stores MORB-like mantle fluids that originate from upwelling of the asthenosphere; once these mantle-sourced fluids leave the deeper chamber, they are transported to Jinjiang hot spring directly through a deep fault system. Finally, gases from Hubin hot spring are produced by mixing from both the shallow and deep magma chambers prior to being released through the volcano’s caldera. Xu et al. (2012) reported variations in the helium isotopic compositions of Hubin and Jinjiang hot spring gases emitted between 1999 and 2011. These variations

ACCEPTED MANUSCRIPT occurred synchronously at both hot springs, suggesting that the two sites share a common source of fluids, with the Hubin site suffering a major crustal helium

IP

T

contamination (4He) with respect to the other spring. During the unrest period from

SC R

2002 to 2006, 3He/4He values increased at both sites, with the Jinjiang site producing higher 3He/4He values than the Hubin site. This continued until the time period of 2009-2010, when gases from both sites recorded the same helium isotopic

NU

compositions (3He/4He > 6 RA and up to 6.5 RA). This clearly demonstrates that the

MA

two sources are fed by the same deep fluids originating within the deep mantle. It also suggests that during periods of increased volcanic activity, pressurization of the

D

magmatic plumbing system by mantle-derived fluids can result in lower contributions

TE

of crustal 4He at both sites (e.g., as occurred in 2002-2006). This is supported by

CE P

evidence of a direct correlation between geophysical signals and He compositions, which has been observed at El Hierro Island (Padron et al., 2013) and Santorini

AC

volcano (Rizzo et al., 2015), regardless of whether it was followed by a magmatic eruption. The analyses of the helium isotopic data of Xu et al. (2012) provide evidences to support our transportation model.

5.4 Model of material connection It is widely accepted that the west Pacific plate has subducted beneath Changbaishan Volcano and currently exists as a stagnant slab (e.g., Zhao et al., 2009; Zhao and Liu, 2010; Wei et al., 2012; Tang et al., 2014). In the subduction model of Zhao et al. (2009), the subducting west Pacific plate has formed a wide mantle wedge;

ACCEPTED MANUSCRIPT it has been speculated that the volcanism at Changbaishan Volcano is associated with both the deep dehydration processes of the subducting slab and the convective

IP

T

circulation processes occurring within the mantle wedge. Additionally, it has been

SC R

proposed that the hydrous wide mantle wedge beneath northeastern China was created by the dehydration of this stagnant slab (Ohtani and Zhao, 2009). However, there is no convincing geochemical evidence to support this dehydration hypothesis.

230

Th excess (Zou et al. 2008, 2011), which does not

MA

eastern China record significant

NU

Uranium-thorium isotope disequilibrium studies reveal that young basalts from

support the presence of a direct material connection between the west Pacific plate

D

and Changbaishan Volcano. The lack of a direct material connection between the

TE

WPSS and Changbaishan Volcano is also supported by the existence of positive Nb

CE P

and Ta anomalies (Chen et al., 2007; Zou et al., 2008) as well as by geothermal studies of Wudalianchi volcano, another intra-plate volcano in a similar plate tectonic

AC

setting within northeastern China (Fig. 1) (Xu et al., 2013b). Kuritani et al. (2011) suggested that the hydrous mantle transition zone beneath East Asia might be the joint result of the dehydration of the ancient slab and the WPSS, but this hypothesis was based on the assumption that the ancient dehydration occurred within a single event ~1.5 billion years ago. However, if the west Pacific plate were to have undergone dehydration at other periods within this time frame, it may not have occurred at a distance of ~1400 km from the Japan Trench. In summary, it is difficult to tell whether the WPSS is dehydrating based on the current geochemical data set. Hahm et al. (2008) and Zhang et al. (2015) suggested that hot spring gases from

ACCEPTED MANUSCRIPT Changbaishan Volcano originated from slab-derived fluids. In these two studies, both authors assumed that the Changbaishan hot spring gases originated from a single

IP

T

source. However, our C-He isotopic data and transportation model suggest that there

SC R

are two main sources contributing to the hot spring gases; in addition, the slab-derived fluids in the shallow crustal chamber cannot represent the deep source of Changbaishan Volcano fluids. According to Maruyama et al. (1997) and Takahashi

NU

(1983), the subduction of the high-speed Izanagi plate beneath Asia caused increased

MA

igneous activities along the continental arc in the Mesozoic (~100 Ma). The Eurasian sub-continental lithospheric mantle (SCLM) might have captured the resulting

D

slab-derived fluids, which could have been stored in matrix and/or minerals of

TE

metasomatic origin, e.g., amphibole, clinopyroxene and mica (O’Reilly and Griffin,

CE P

2013). It is improbable that the 3He/4He ratio would have changed significantly over this period, as the impacts of radiogenic 4He from U and Th on 3He/4He ratios are

AC

relatively well understood (e.g., Gautheron and Moreira, 2002; Kim et al., 2005; Sapienza et al., 2005). In the open system model proposed by Gautheron and Moreira (2002), 3He/4He ratios would remain constant over time, whereas in a closed system model, the 3He/4He ratio would decrease by less than 0.5 RA over this period (see Kim et al., 2005; Sapienza et al., 2005 for details of calculation). Here, we suggest that the slab-derived fluids released at Julong hot spring may originate from ancient Izanagi slab materials, rather than from the young WPSS. If the slab-derived fluids are generated from the west Pacific plate, the proposed deeper magma chamber should reflect the presence of slab-derived fluids rather than

ACCEPTED MANUSCRIPT MORB-like mantle fluids, and there should also be a different magnitude of marine limestone contributions to the Changbaishan slab-derived fluids than is seen in the

IP

T

Japan arc. Based on the data set presented within this paper, we do not believe there is

SC R

evidence to support the hypothesis of a direct material connection between Changbaishan Volcano and the west Pacific plate.

NU

6. Conclusions

MA

Based on our discussion of helium and carbon isotopes in hot spring gases at Changbaishan Volcano, we believe that Changbaishan hot spring gases have two main

D

sources: mantle fluids originating from a deeper magma chamber and slab-derived

TE

fluids coming from a shallower magma chamber. Fluids from the ancient Izanagi

CE P

subduction zone fluids might have been trapped in the Eurasian plate since the Mesozoic and released and transported to Julong hot spring in the Cenozoic. Hot

AC

spring gases at Jinjiang hot spring are likely sourced from the deeper magma chamber, which stores fluids from a deep mantle source. Hot spring gases at Hubin hot spring reflect a mixture of fluids originating from MORB-like and slab-derived sources. Based on these results, we suggest that helium and carbon isotopes in hot springs do not support the hypothesis of a direct material connection between Changbaishan Volcano and the west Pacific plate.

Acknowledgements: We thank the editor Prof. Alessandro Aiuppa and two anonymous reviewers for their very useful comments and suggestions that helped to

ACCEPTED MANUSCRIPT greatly improve the manuscript. We thank Dr. Liwu Li and Xudong Lin for their support in sampling and measurements. This work was partially funded by the Natural

SC R

Volcano under China Earthquake Administration.

IP

T

Science Foundation of China (41502314) and the monitoring project of Changbaishan

References

NU

Ballentine, C.J., Schoell, M., Coleman, D., Cain B.A., 2001. 300-Myr-old magmatic

MA

CO2 in natural gas reservoirs of the west Texas Permian basin. Nature, 409, 327-331.

D

Barry, P.H., Hilton, D.R., Fischer, T.P., De Moor, J.M., Mangasini, F., Ramirez, C.,

TE

2013. Helium and carbon isotope systematics of cold “mazuku” CO2 vents and

CE P

hydrothermal gases and fluids from Rungwe Volcanic Province, southern Tanzania. Chemical Geology, 339, 141-156.

AC

Brimblecombe, P., 1996. Air composition and chemistry. Cambridge University Press. Cambridge. Burgisser, A., Scaillet, B., 2007. Redox evolution of a degassing magma rising to the surface. Nature, 445, 194-197. Carroll, J.J., Slupsky, J.D., Mather, A.E., 1991. The solubility of carbon dioxide in water at low pressure. Journal of Physical and Chemical Reference Data, 20, 1201-1209. Chen, Y., Zhang, Y., Graham, D., Su, S., Deng, J., 2007. Geochemistry of Cenozoic basalts and mantle xenoliths in Northeast China. Lithos, 96, 108-126.

ACCEPTED MANUSCRIPT Clarke, W.B., Jenkins, W.J., Top, Z., 1976. Determination of tritium by mass spectrometric measurement of 3He. International Journal of Applied Radiation and

IP

T

Isotopes, 27, 515-522.

SC R

Fan, Q., Sui, J., Wang, T., Li, N., Sun, Q., 2007. History of volcanic activity, magma evolution and eruptive mechanisms of the Changbai volcanic province. Geological Journal of China Universities, 13, 175-190 (in Chinese, with English Abstr.).

NU

Gautheron, C., Moreira, M., 2002. Helium signature of the subcontinental lithospheric

MA

mantle. Earth and Planetary Science Letters, 199, 39-47. Gerlach, T.M., 1991. Present-day CO2 emissions from volcanos. Eos, Transactions

D

American Geophysical Union, 72, 249-255.

TE

Graham, D.W., 2002. Noble gas isotope geochemistry of mid-ocean and ocean island

CE P

basalts: characterization of mantle source reservoirs. Reviews in Mineralogy and Geochemistry, 47, 247-317.

AC

Guo, Z., Zhang, M., Cheng, Z., Zhang, L., Liu, J., 2014. Fluxes and genesis of greenhouse gases emissions from typical volcanic fields in China. Acta Petrologica Sinica, 30, 3467-3480 (in Chinese, with English Abstr.). Hahm, D., Hilton, D.R., Cho, M., Wei, H., Kim, K.R., 2008. Geothermal He and CO2 variations at Changbaishan intra-plate volcano (NE China) and the nature of the sub-continental lithospheric mantle. Geophysical Research Letters, 35, L22304. Hetland, E.A., Wu, F.T., Song, J.L., 2004. Crustal structure in the Changbaishan volcanic area, China, determined by modeling receiver functions. Tectonophysics, 386, 157-175.

ACCEPTED MANUSCRIPT Hilton, D.R., 1996. The helium and carbon isotope systematics of a continental geothermal system: results from monitoring studies at Long Valley caldera

IP

T

(California, U.S.A.). Chemical Geology, 127, 269-295.

SC R

Hilton, D.R., Fischer, T.P., Marty, B., 2002. Noble gases and volatile recycling at subduction zones. Reviews in mineralogy and geochemistry, 47, 319-370. Kennedy, B.M., Van Soest, M.C., 2006. A helium isotope perspective on the Dixie

NU

Valley, Nevada, hydrothermal system. Geothermics, 35, 26-43.

MA

Kennedy, B.M., Van Soest, M.C., 2007. Flow of mantle fluids through the ductile lower crust: helium isotope trends, Science, 318, 1433-1436.

D

Kim, K.H., Nagao, K., Tanaka, T., Sumino, H., Nakamura, T., Okuno, M., Lock, J.B.,

TE

Youn, J.S., Song, J., 2005. He-Ar and Nd-Sr isotopic compositions of ultramafic

CE P

xenoliths and host alkali basalts from the Korea peninsula. Geochemical Journal, 39, 341-356.

AC

Kuritani, T., Ohtani, E., Kimura, J., 2011. Intensive hydration of the mantle transition zone beneath China caused by ancient slab stagnation. Nature Geoscience, 4, 713-716.

Kyong-Song, R., Hammond, J.O., Chol-Nam, K., Hyok, K., Yong-Gun, Y., Gil-Jong, P., Chong-Song, R., Oppenheimer, C., Liu, K.W., Iacovino, K., Kum-Ran, R., 2016. Evidence for partial melt in the crust beneath Mt. Paekdu (Changbaishan), Democratic People’s Republic of Korea and China. Science Advances, 4, e1501513. Li, S., Mooney, W.D., Fan, J., 2006. Crustal structure of mainland China from deep

ACCEPTED MANUSCRIPT seismic sounding data. Tectonophysics, 420, 239-252. Marty, B., Jambon, A., 1987. C/3He in volatile fluxes from the solid Earth:

IP

T

implication for carbon geodynamics. Earth and Planetary Science Letters, 83,

SC R

16-26.

Marty, B., Tolstikhin, I.N., 1998. CO2 fluxes from mid-ocean ridges, arcs and plumes. Chemical Geology, 145, 233-248.

NU

Maruyama, S., Isozaki, Y., Kimura, G., Terabayashi, M., 1997. Paleogeographic maps

MA

of the Japanese Islands: Plate tectonic synthesis from 750 Ma to the present. The Island Arc, 6, 121-142.

D

Millero, F.J., 1995. Thermodynamics of the carbon dioxide system in the oceans.

TE

Geochimica et Cosmochimica Acta, 59, 661-677.

CE P

Morikawa, N., Kazahaya, K., Fourre, E., Takahashi, H.A., Jean-Baptiste, P., Ohwada, M., Leguern, F.J., Nakama, A., 2008. Magmatic He distribution around Unzen

AC

volcano inferred from intensive investigation of helium isotopes in groundwater. Journal of Volcanology and Geothermal Research, 175, 218-230. Morrison, P., Pine, J., 1955. Radiogenic origin of the helium isotopes in rock. Annals of the New York Academy of Sciences, 62, 71-92. Ohno, M., Sumino, H., Hernandez, P.A., Sato, T., Nagao, K., 2011. Helium isotopes in the Izu Peninsula, Japan: Relation of magma and crustal activity. Journal of Volcanology and Geothermal Research, 199, 118-126. Ohtani, E., Zhao, D., 2009. The role of water in the deep upper mantle and transition zone: dehydration of stagnant slabs and its effects on the big mantle wedge.

ACCEPTED MANUSCRIPT Russian Geology and Geophysics, 50, 1073-1078. Ohwada, M., Kazahaya, K., Itoh, J., Morikawa, N., Takahashi, M., Takahashi, H.A.,

IP

T

Inamura, A., Yasuhara, M., Tsukamoto, H., 2012. Passive degassing of magmatic

SC R

volatiles from Iwate volcano, NE Japan, based on three-dimensional measurement of helium isotopes in groundwater. Journal of Geophysical Research 117, B02204. O’Reilly, S.Y., Griffin, W.L., 2013. Mantle metasomatism. In: Metasomatism and the

NU

Chemical Transformation of Rock. Springer, Berlin, Heidelberg, pp. 471-533.

MA

Ozima, M., Podosek, F.A., 2002. Noble gas Geochemistry. Cambridge University Press, Cambridge.

D

Padron, E., Perez, N.M., Hernandez, P.A., Sumino, H., Melian, G., Barrancos, J.,

TE

Nolasco, D., Padilla, G., 2012. Helium emission at Cumbre Vieja volcano, La

CE P

Palma, Canary Islands. Chemical Geology, 312-313, 138-147. Padron E., Perez, N.M., Hernandez, P.A., Sumino, H., Melian, G.V., Barrancos, J.,

AC

Nolasco, D., Padilla, G., Dionis, S., Rodriguez, F., Hernandez, I., Calvo, D., Peraza, M.D., Nagao, K., 2013. Diffusive helium emissions as a precursory sign of volcanic unrest. Geology, 41, 539-542. Qiu, G., Pei, F., Fang, H., Du, B., Zhang, X., Zhang, P., Yuan, Y., He, M., Bai, D., 2014. Analysis of magma chamber at the Tianchi volcano area in Changbai mountain. Chinese Journal of Geophysics, 57, 3466-3477 (in Chinese, with English Abstr.). Rizzo, A.L., Barberi, F., Carapezza, M.L., Di Piazza, A., Francalanci, L., Sortino, F., D'Alessandro, W., 2015. New mafic magma refilling a quiescent volcano: Evidence

ACCEPTED MANUSCRIPT from He-Ne-Ar isotopes during the 2011-2012 unrest at Santorini, Greece. Geochemistry, Geophysics, Geosystem, 16, 798-814, doi: 10.1002/2014GC005653.

IP

T

Ruzie, L., Aubaud, C., Moreira, M., Agrinier, P., Dessert, C., Greau, C., Crispi, O.,

SC R

2013. Carbon and helium isotopes in thermal springs of La Soufriere volcano (Guadeloupe, Lesser Antilles): Implications for volcanological monitoring. Chemical Geology, 359, 70-80.

NU

Sano, Y., Marty, B., 1995. Origin of carbon in fumarolic gas from island arcs.

MA

Chemical Geology (Isotope Geoscience Section), 119, 265-274. Sano, Y., Takahata, N., Seno, T., 2006. Geographical distribution of 3He/4He ratios in

D

the Chugoku District, Southwestern Japan. Pure and Applied Geophysics, 163,

TE

745-757.

CE P

Sapienze, G., Hilton, D.R., Scribano, V., 2005. Helium isotopes in peridotite mineral phases from Hyblean Plateau xenoliths (south-eastern Sicily, Italy). Chemical

AC

Geology, 219, 115-129.

Shangguan, Z., Wu, C., 2008. Geochemical features of magmatic gases in the regions of dormant volcanoes in China. Acta Petrologica Sinica, 24, 2638-2646 (in Chinese, with English Abstr.). Shangguan, Z., Kong, L., Sun, F., Gao, S., Xu, X., 1996. Deep-seated fluid components and stable isotopic compostions in Tianchi volcanic area, Changbaishan Mountains. Scientia Geologica Sinica, 31, 54-64 (in Chinese, with English Abstr.). Shaw, A.M., Hilton, D.R., Fischer, T.P., Walker, J.A., De Leeuw, G.A.M., 2006.

ACCEPTED MANUSCRIPT Helium isotope variations in mineral separates from Costa Rica and Nicaragua: Assessing crustal contributions, timescale variations and diffusion-related

IP

T

mechanisms. Chemical Geology, 230, 124-139.

SC R

Takahashi, M., 1983. Space-time distribution of late Mesozoic to early Cenozoic magmatism in East Asia and its tectonic implications. In: Hashimoto, M., Uyeda, S.(Eds.), Accretion Tectonics in the Circum-Pacific Regions, pp. 69-88.

NU

Tang, J., Deng, Q., Zhao, G., Bai, D., Jin, G., Li, W., Xuan, F., Zhan, Y., Li, G.,

MA

Hong, F., Ma, M., Chen, F., 2001. Electric conductivity and magma chamber at the Tianchi volcano area in Changbaishan Mountain. Seismology and Geology, 23,

D

191-200 (in Chinese, with English Abstr.).

TE

Tang, J., Zhao, G., Wang, J., Zhan, Y., Deng, Q., Chen, X., 2006. Study of the

CE P

formation mechanism for volcanism in Northeast China based on deep electric structure. Acta Petrologica Sinica, 22, 1503-1510 (in Chinese, with English Abstr.).

AC

Tang, Y., Obayashi, M., Niu, F., Grand, S.P., Chen, Y.J., Kawakatsu, H., Tanaka, S., Ning, J., Ni, J.F., 2014. Changbaishan volcanism in northeast China linked to subduction-induced mantle upwelling. Nature Geoscience, 7, 470-475. Wang, F., Chen, W., Peng, Z., 2000. Features of excess of U in volcanic rocks of Changbaishan Tianchi. Seismology and Geology, 22, 87-94 (in Chinese, with English Abstr.). Wei, H., Wang, Y., Jin, J., Gao, L., Yun, S., Jin, B., 2007. Timescale and evolution of the intracontinental Tianchi volcanic shield and ignimbrite-forming eruption, Changbaishan, Northeast China. Lithos, 96, 315-324.

ACCEPTED MANUSCRIPT Wei, W., Xu, J., Zhao, D., Shi, Y., 2012. East Asia mantle tomography: New insight into plate subduction and intraplate volcanism. Journal of Asian Earth Sciences, 60,

IP

T

88-103.

Chemical and Engineering Data, 16, 235-241.

SC R

Weiss, R.F., 1971. Solubility of helium and neon in water and seawater. Journal of

Xu, J., Liu, G., Wu, J., Ming, Y., Wang, Q., Cui, D., Shangguan, Z., Pan, B., Lin, X.,

NU

Liu, J., 2012. Recent unrest of Changbaishan volcano, northeast China: A precursor

MA

of a future eruption? Geophysical Research Letters, 39, L16305. Xu, J., Pan, B., Liu, T., Hajdas, I., Zhao, B., Yu, H., Liu, R., Zhao, P., 2013a. Climatic

D

impact of the Millennium eruption of Changbaishan volcano in China: New insight

CE P

Letters, 40, 54-59.

TE

from high-precision radiocarbon wiggle-match dating. Geophysical Research

Xu, S., Zheng, G., Nakai, S., Wakita, H., Wang, X., Guo, Z., 2013b. Hydrothermal He

AC

and CO2 at Wudalianchi intra-plate volcano, NE China. Journal of Asian Earth Sciences, 62, 526-530. Yu, H., Xu, J., Luan, P., Zhao, B., Pan, B., 2013. Probabilistic assessment of tephra fallout hazard at Changbaishan volcano, Northeast China. Natural Hazards, 69, 1369-1388. Zelenski, M.E., Taran, Y.A., Dubinina, E.O., Shapar, V.N., Polyntseva, E.A., 2012. Sources of volatiles for a subduction zone volcano: Mutnovsky Volcano, Kamchatka. Geochemistry International, 50, 502-521. Zhang, M., Guo, Z., Sano, Y., Cheng, Z., Zhang, L., 2015. Stagnant subducted Pacific

ACCEPTED MANUSCRIPT slab-derived CO2 emissions: Insights into magma degassing at Changbaishan volcano, NE China. Journal of Asian Earth Sciences, 106, 49-63.

IP

T

Zhang, X., Zhang, C., Zhao, J., Yang, Z., Li, S., Zhang, J., Liu, B., Cheng, S., Sun,

SC R

G., Pan, S., 2002. Deep seismic sounding investigation into the deep structure of the magma system in Changbaishan Tianchi volcanic region. Acta Seismologica Sinica, 24, 135-143 (in Chinese, with English Abstr.).

NU

Zhao, C., Ran, H., Wang, Y., 2012. Present-day mantle-derived helium release in the

MA

Tengchong volcanic field, Southwest China: Implications for tectonics and magmatism. Acta Petrologica Sinica 28: 1189-1204 (in Chinese, with English

D

Abstr.).

TE

Zhao, D., Liu, L., 2010. Deep structure and origin of active volcanoes in China.

CE P

Geoscience Frontiers, 1, 31-44.

Zhao, D., Tian, Y., Lei, J., Liu, L., Zheng, S., 2009. Seismic image and origin of the

AC

Changbai intraplate volcano in East Asia: Role of big mantle wedge above the stagnant Pacific slab. Physics of the Earth and Planetary Interiors, 173, 197-206. Zou, H., Fan, Q., 2011. Uranium-thorium isotope disequilibrium in young volcanic rocks from China. Acta Petrologica Sinica, 27, 2821-2826 (in Chinese, with English Abstr.). Zou, H., Fan, Q., Yao, Y., 2008. U-Th systematics of dispersed young volcanoes in NE China: Asthenosphere upwelling caused by piling up and upward thickening of stagnant Pacific slab. Chemical Geology, 255, 134-142.

ACCEPTED MANUSCRIPT Figure Captions

IP

T

Fig. 1. Morphotectonic map of Changbaishan Volcano, showing the locations of

SC R

Changbaishan Volcano and Jinjiang, Hubin and Julong hot springs. For comparison, Jinjiang, Hubin and Julong hot springs correspond to the Jinjiang River, Tianchi Lake shoreline and Changbaishan Waterfall sampling sites, respectively, of Hahm et al.

MA

NU

(2008).

Fig. 2. Helium isotopes (RC/RA) (a), carbon isotopes of carbon dioxide (δ13C) (b) and

D

CO2/3He ratios (c) at Hubin (blue squares), Julong (green circles) and Jinjiang (red

TE

triangles) hot springs. The typical mid-ocean ridge basalt (MORB) ranges of 3He/4He,

CE P

δ13C values of CO2 and CO2/3He are 8 ± 1 RA (Graham, 2002), -6.5 ± 2‰ (Sano and

AC

Marty, 1995) and 2 ± 1 × 109 (Marty and Jambon, 1987), respectively.

Fig. 3. Plot of measured 3He/4He ratios versus solubility-corrected and air-normalized 4

He/20Ne ratios (X-values) for samples from Changbaishan Volcano. For hot spring

gas samples, X = [(4He/20Ne)/(4He/20Ne)air] × (βNe/βHe). For olivines, X = (4He/20Ne)/(4He/20Ne)air. Also displayed are mixing lines between the air-saturated water (ASW, 3He/4He = 1 RA, 4He/20Ne = 0.250-0.285; Ozima and Podosek, 2002), and mantle (7.71 RA), and crustal helium (0.05 RA) endmembers, as well as different degrees of crustal contamination. It should be noted that the highest sample among the plotted ones (sample XPD3, RM/RA=7.71) is selected as the He values of the mantle

ACCEPTED MANUSCRIPT component, because there is no evidence that the mantle beneath this volcano has a

3

He and

4

He for bubbling gas samples from

SC R

Fig. 4. Ternary plot of CO2,

IP

T

higher He isotopic composition than 7.71 RA.

Changbaishan Volcano (modified after Hahm et al., 2008). All the samples fall in the gray range of arc-related volcanism (RC/RA = 5.4 ± 1.9; Hilton et al., 2002). For

NU

reference, points A, B and C represent the Island arc average (3He/4He = 5.4 RA,

MA

CO2/3He = 1010; Hilton et al., 2002), MORB average (3He/4He = 8 RA, CO2/3He = 2 × 109; Graham, 2002) and air average (3He/4He = 1 RA, CO2/3He = 4.9 × 107; Clarke et

TE

D

al., 1976; Brimblecombe, 1996; Hilton et al., 2002), respectively.

CE P

Fig. 5. Plot of CO2/3He versus δ13C for bubbling gas samples from Changbaishan Volcano. M, S and L endmembers represent mantle carbon, sedimentary organic

AC

carbon and marine limestone carbon, respectively. The endmember compositions of M, S and L are δ13C = -6.5‰, -30‰ and 0‰; and CO2/3He = 1.6 × 109, 1 × 1013 and 1 × 1013, respectively (Sano and Marty, 1995). It should be noted that the CO2/3He value of MORB, 1.6 × 109, is an average of two estimates: 2.2 × 109 in MORB by Marty and Tolstikhin (1998) and 1 × 109 in mid-ocean ridge hydrothermal fluids by Gerlach (1991).

Fig. 6. Schematic diagram of west-east vertical cross-section depicting the tectonic structure of Northeast Asia (modified from Zhao et al., 2009). The WPSS, which has

ACCEPTED MANUSCRIPT dehydrated before subducting under Changbaishan Volcano (Zou and Fan, 2011), becomes a stagnant slab in the transition zone, while a wide mantle wedge has formed

IP

T

in the upper mantle above the WPPS (Zhao and Liu, 2010; Wei et al, 2012). The

SC R

mantle source fluids in the deeper magma chamber beneath Changbaishan Volcano may derive from the asthenosphere.

NU

Fig. 7. Schematic map of locations of the shallow and deep magma chambers and the

MA

distribution of Julong, Hubin and Jinjiang hot springs. Green layer indicates the ancient Izanagi slab fluids, which might be released and transferred to the crust and

D

stored in matrix and/or minerals of metasomatic origin by the Cenozoic extension or

AC

CE P

TE

rift.

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

AC

CE P

Figure 1

AC

Figure 2

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

CE P

AC

Figure 3

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

AC

Figure 4

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

AC

CE P

TE

Figure 5

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

AC

CE P

TE

D

Figure 6

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

AC

CE P

TE

Figure 7

ACCEPTED MANUSCRIPT Table 1 Chemical and isotopic compositions of hot spring gases collected at Changbaishan

b

Date

T( C)

(%)

(ppm)

ID JL02-8

July/7/2002

74.5

98.7

16.5

Julong

JL02-15

July/7/2002

71.5

95.0

13.8

Jinjiang

JJ02-1

July/5/2002

56.3

91.2

113

Jinjiang

JJ02-3

July/5/2002

57.7

92.7

47.8

Julong

JL06-9

Aug/26/2006

75.0

98.5

Julong

JL06-16

Aug/26/2006

74.5

98.0

Hubin

HB06-1

Aug/27/2006

13.0

71.8

Julong

JL14-9

Sept/6/2014

75.0

Julong

JL14-10

Sept/6/2014

73.6

Julong

JL14-15

Sept/6/2014

71.4

Julong

JL14-16

Sept/6/2014

Hubin

HB14-2

Sept/4/2014

Hubin

HB14-3

Jinjiang

RM/RA(±2σ)

CO2 δ13C RC/RAc

3

CO2/ He

(‰,PDB)

30.9

4.35±0.21

4.46

9.58E+09

-6.0

7.2

4.74±0.23

5.35

9.19E+09

-6.2

97.7

4.80±0.20

4.84

1.19E+09

-7.2

233

4.77±0.34

4.79

2.89E+09

-7.3

22.9

204

5.80±0.18

5.82

5.28E+09

-4.9

25.9

229

5.91±0.25

5.93

4.56E+09

-5.0

386

457

6.02±0.28

6.03

2.20E+08

-4.7

92.8

15.9

107

4.99±0.07

5.03

8.29E+09

-3.3

96.3

0.9

2.6

2.82±0.27

3.98

1.92E+11

-3.6

93.9

6.2

32.7

4.49±0.11

4.60

2.35E+10

-2.9

D

MA

NU

Julong

X

c

SC R

Locality

[He]Ca

o

IP

CO2

Sample

T

Volcano in 2002, 2006, 2014 and 2015.

96.9

24.7

74.1

4.69±0.10

4.74

5.91E+09

-3.0

33.0

94.1

20.6

48.7

5.06±0.10

5.15

6.34E+09

-1.8

Sept/4/2014

23.1

70.2

405

448

5.11±0.10

5.12

2.42E+08

-1.6

JJ14-01

Sept/8/2014

56.6

90.1

114

62.7

4.90±0.08

4.96

1.14E+09

-5.3

Julong

JL15-10

Sept/4/2015

74.2

99.7

0.9

7.4

4.33±0.27

4.85

1.63E+11

-6.2

Julong

JL15-15

Sept/4/2015

71.4

99.4

5.9

39.9

4.85±0.14

4.95

2.43E+10

-5.7

Julong

JL15-16

Sept/4/2015

63.3

98.9

25.9

177

5.34±0.07

5.36

5.09E+09

-5.5

Hubin

HB15-1

Sept/4/2015

11.7

89.6

334

316

5.55±0.10

5.56

3.45E+08

-4.9

Hubin

HB15-2

Sept/4/2015

13.6

89.4

245

612

5.66±0.09

5.67

4.60E+08

-4.8

HB15-3

Sept/4/2015

11.1

86.0

385

612

5.55±0.12

5.56

2.87E+08

-4.3

HB15-4

Sept/4/2015

11.9

80.8

339

516

5.52±0.10

5.53

3.08E+08

-4.5

JJ15-1

Sept/2/2015

57.9

95.2

128

396

4.92±0.10

4.93

1.08E+09

-7.9

Hubin Jinjiang

CE P

AC

Hubin

TE

68.9

a

[He]C is the air-corrected helium content.

b

X=[(4He/20Ne)measured/(4He/20Ne)air] × (βNe/βHe), where β is the Bunsen solubility coefficients.

c

RM, RC and RA are the measured, air-corrected and atmospheric 3He/4He ratios, respectively (RA ≈ 1.4 × 10-6； Clarke et al.,

1976).

ACCEPTED MANUSCRIPT Table 2 Helium and neon characteristics of xenoliths (olivine) at Changbaishan Volcano Sample

Weight (mg)

He[C] (ncm3 STP/g)

(4He/20Ne)/(4He/20Ne)air

RM/RA

YDG-03

833

1.81

250

6.78

YDG-04

1098

0.84

58

XPD3

1191

5.6

2229

XPD5

972

2.8

293

6.80

Hahm et al. (2008)

3.26

3.30

Hahm et al. (2008)

7.71

7.71

Kim et al. (2005)

7.14

7.16

Kim et al. (2005)

IP

T

Reference

SC R

NU

Table 3

RC/RA

MA

Carbon sources of hot spring gas samples from Julong hot spring at Changbaishan Volcano. Mantle

Sediments

Limestone

ID

(%)

(%)

(%)

16

67

17

JL02-15

17

JL06-9

30

JL06-16

35

JL14-9

TE

JL02-8

D

Sample

66

10

60

9

56

19

7

74

JL14-10

1

12

87

JL14-15

7

8

85

JL14-16

27

4

69

JL15-10

1

20

79

JL15-15

7

18

75

JL15-16

31

12

57

Average

17

12

71

AC

CE P

17

ACCEPTED MANUSCRIPT Highlights Two sources of volcanic gases have been identified at Changbaishan volcano, China.

IP

T

The mantle fluids may be derived from the asthenospheric materials.

SC R

The slab-derived fluids are associated with the ancient subducted plate.

AC

CE P

TE

D

MA

NU

There is no material connection between Changbaishan volcano and the Pacific plate.