Volatile flux from subduction zone volcanoes: Insights from a detailed evaluation of the fluxes from volcanoes in Japan

Journal of Volcanology and Geothermal Research 268 (2013) 46–63

Contents lists available at ScienceDirect

Journal of Volcanology and Geothermal Research journal homepage: www.elsevier.com/locate/jvolgeores

Review

Volatile flux from subduction zone volcanoes: Insights from a detailed evaluation of the fluxes from volcanoes in Japan Hiroshi Shinohara Geological Survey of Japan, AIST, 1-1-1 Higashi, Central 7, Tsukuba 305-8567, Ibaraki, Japan

a r t i c l e

i n f o

Article history: Received 2 May 2013 Accepted 7 October 2013 Available online 23 October 2013 Keywords: Volatile flux SO2 flux Volcanic gas Degassing Hot spring Subduction zone Eruption

a b s t r a c t Global volatile fluxes from subaerial volcanoes at subduction zones were estimated based on a compilation of fluxes from various sources, including persistent degassing, hot and cold springs, soil degassing, and eruptions. Because worldwide comprehensive datasets are not available, especially for diffuse volatile discharges, volatile fluxes from Japan arcs were estimated based on detailed datasets, and the regional fluxes were extrapolated to the global flux with consideration of the regional characteristics of volcanic volatile compositions, which were estimated based on volcanic gas compositions of persistent degassing. The estimated global fluxes indicate that persistent degassing is the major source of volatiles, especially for S with a contribution of 80%. Diffuse discharges and persistent degassing are similarly important sources of H2O, CO2, and Cl, but the contribution of explosive eruptions is less than 15% for all the volatiles. The estimates of diffuse degassing fluxes include large errors due to limited data. However, the potential impact of these sources on the global flux indicates the importance of further studies to quantify these fluxes. The volatile budget of subduction zone volcanism was evaluated by comparing the estimated volatile fluxes, the volatile contents in the crust, and the primitive magma volatile contents. The contribution of volatiles remaining in the crust are not significant; however, consideration of lower crust foundering significantly alters the volatile budget estimate because the primitive magma supply rate should be significantly increased to account for the lower crust foundering. © 2013 Elsevier B.V. All rights reserved.

Contents 1. 2. 3.

Introduction . . . . . . . . . . . . . . . . . . Global flux estimate method . . . . . . . . . . . SO2 flux . . . . . . . . . . . . . . . . . . . . 3.1. Persistent SO2 flux from volcanoes in Japan . 3.2. Global SO2 flux by persistent degassing . . . 3.3. SO2 flux from eruptions . . . . . . . . . . 4. Volatile flux by persistent degassing . . . . . . . . 4.1. Persistent degassing of volcanoes in Japan . 4.2. Persistent degassing at other subduction zones 5. Diffuse degassing . . . . . . . . . . . . . . . . 5.1. Volatile flux through water discharges . . . 5.2. Soil CO2 flux . . . . . . . . . . . . . . . 6. Volatile flux from explosive eruptions . . . . . . . 6.1. Explosive eruptions of silicic magmas . . . . 6.2. Explosive eruptions of mafic magmas . . . . 7. Volatile budget for subduction zone volcanism . . . 7.1. Global volatile flux from subduction zones . 7.2. Magmatic volatile differentiation in the crust 8. Summary . . . . . . . . . . . . . . . . . . . . Acknowledgment . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . .

E-mail address: [email protected]. 0377-0273/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.jvolgeores.2013.10.007

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . .

47 47 48 48 49 49 50 50 50 52 52 55 55 56 57 58 58 58 59 60 60

H. Shinohara / Journal of Volcanology and Geothermal Research 268 (2013) 46–63

1. Introduction The emission of volatiles from volcanoes is the major process of Earth degassing, and the mass balance of volatiles between the mantle and the Earth's surface controls the stability of the presentday atmosphere and the ocean. Subduction zones are major sources of magma and volatiles, second only to mid-ocean ridges (Crisp, 1984; Torgersen, 1989). In addition to being important sources of volatiles, subduction zones are also sinks of volatile components, and the recycling of volatiles that occur at subduction zones controls the global volatile budget (Ito et al., 1983; Bebout, 1996). The volatile flux from volcanoes is a key factor in quantifying the volatile budget and has been estimated using various methods (Sano and Williams, 1996; Marty and Tolstikhin, 1998; Hilton et al., 2002; Wallace, 2005). Most previous estimates of the volatile flux from subduction zones, obtained using various methods, are essentially based on either measured SO2 fluxes (e.g., Stoiber et al., 1987) or estimated magma production rates (e.g., Crisp, 1984). Emissions of SO2 from subaerial volcanoes are quantified by ultraviolet remote sensing of volcanic plumes discharged by either non-eruptive degassing or eruptions, and the global SO2 flux has been estimated by compiling such data (Stoiber et al., 1987; Andres and Kasgnoc, 1998). Fluxes of other volatile components are commonly estimated by multiplying the SO2 flux by the high-temperature fumarolic gas composition (Hilton et al., 2002; Fischer, 2008). Many previous studies have considered only volatile fluxes of volcanic gas emissions. The volatile fluxes of eruptions, diffuse emissions, and the volatiles remaining in erupted or intruded magmas have been poorly evaluated in the volatile budget, although some studies have discussed the possible importance of these sources (e.g., Seward and Kerrick, 1996; Taran, 2009). Volatile fluxes are proportional to the magma flux, and volatile fluxes from subduction zones are estimated by comparing with the 3 He flux from mid-ocean ridges (Craig et al., 1975; Torgersen, 1989) or with the volatile contents of primitive magmas (Ito et al., 1983; Wallace, 2005). The fluxes of other components are estimated by multiplying with the composition ratios such as CO2/3He or N2/3He, in the volcanic gases (Sano and Williams, 1996; Sano et al., 2001). The volatile contents of primitive magmas are estimated based on the volatile contents of melt inclusions (Wallace, 2005; Sadofsky et al., 2008; Ruscitto et al., 2012). Volatile fluxes estimated by this method represent input fluxes to the crust, but these are not necessarily the same as the output fluxes from volcanoes, such as those of volcanic gases and diffuse emissions. Some volatiles may remain in intruded magmas, and their inventory should be considered in the volatile budget of the crust. There is also a large uncertainty in the primitive magma supply rate to subduction zones because a large amount of mafic lower crust must be returned to the mantle (called lower crust delamination or foundering) in order to produce andesitic crust from primitive basaltic magmas (Kelemen et al., 2003; Rudnick and Gao, 2003). Because the mafic lower crust may not be volatile-free, the lower crust foundering also causes volatile recycling to the mantle, and these effects also need to be evaluated in order to estimate the volatile budget of subduction zones. The goal of this review is to outline the global volatile budget of subduction zone crust based on estimates of the volatile flux from subaerial volcanoes and the volatile inventory of the crust. We will focus the discussion on the major volatile components, H2O, CO2, S and Cl. In order to summarize the volcanic volatile flux, flux data were compiled for not only persistent volcanic gas emissions but also for emissions by eruptions and other diffuse emissions, such as those from hot springs, river waters (cold springs), and soil degassing. Because worldwide systematic flux data are not available for diffuse degassing, regional fluxes were estimated based on detailed data compilations on Japanese volcanoes. These regional fluxes are then extrapolated to the global flux considering the relative contribution of the Japanese volcanoes to the global flux by the persistent degassing. The volatile

47

flux from eruptions is discussed on only a global scale because regional estimates are difficult due to the limited number of the measured eruptions. The estimated volatile fluxes are compared with the magma production rate and the volatile contents in primitive magma melt inclusions and in intruded magmas in order to evaluate the volatile budget of the crust. 2. Global flux estimate method A simple method to estimate a regional or global flux of volcanic volatiles is a summation of an exhaustive set of flux data covering fluxes from most emission sources. In order to estimate the global SO2 flux by persistent degassing, Stoiber et al. (1987) and Andres and Kasgnoc (1998) tried to obtain the exhaustive flux data set not only taken from the literature but also by sending inquiries to many researchers. The exhaustive flux data set, however, is difficult to obtain, in particular for the remote areas and also for diffuse emissions from numerous sources, and we need a method to estimate global fluxes based on a limited number of the data. Mörner and Ethiope (2002) estimated that contribution of the soil degassing to the global CO2 degassing is 0.1–0.5 of the plume degassing. This estimate, however, is based on a simple comparison of the subtotals of limited number of the flux data (~20) of each type of emission. Pérez et al. (2011) estimated the global CO2 flux from volcanic lakes based on a simple extrapolation of the compiled flux data with the coverage proportion of the measured and existing volcano numbers. Burton et al. (2013) applied this method to estimate the global CO2 flux by persistent and soil degassing. They calculated the average fluxes by each emission type based on flux data from about 30 volcanoes, and extrapolated them by the number of persistently degassing volcanoes (150) or of soil degassing volcanoes (550). This method, however, will cause a significant overestimate, because flux measurements are likely conducted at active systems rather than at inactive low flux systems. Regional fluxes by various diffuse emissions were estimated based on intensive flux surveys of relatively small areas and the impacts of the diffuse emissions on the global flux were evaluated based on their extrapolation. Seward and Kerrick (1996) compiled the CO2 flux by hydrothermal activity along the 150-km-long Taupo Volcanic Zone, New Zealand and evaluated the global impact of the hydrothermal CO2 emission by extrapolating the regional flux to the global flux based on the length of subduction zone. James et al. (1999) measured the magmatic CO2 flux from cold springs at 75 km segment of the Oregon Cascades, USA and estimated the global CO2 flux by the extrapolation based on the subduction zone length. The simple extrapolation by the subduction length, however, may introduce significant errors, as activity of subduction zones is not uniform. For an example, the Taupo Volcanic Zone exhibits anomalously high heat flux compared with other subduction zones (Hochstein, 1995) and the simple extrapolation of its heat and volatile fluxes will cause a significant overestimate. Furthermore, composition of magmatic volatiles can be variable at different subduction zones depending on different contribution of source materials (Elkins et al., 2006; Ruscitto et al., 2012) and volatile composition variation should be also taken into account for the extrapolation. In this study, the global fluxes of diffuse emissions are estimated by extrapolating the regional flux with factors obtained as the ratios of the regional flux and the global flux of the persistent degassing. Both the regional and global fluxes by the persistent degassing are better constrained than fluxes of diffuse emissions because of a larger data set, as will be discussed later. The regional to global flux ratios are obtained for each volatile component, H2O, CO2, S and Cl, and applied for the extrapolation in order to account for the volatile composition variation at different subduction zones with an assumption that the composition characteristics at different areas are caused by different contribution of source materials and should be appeared at any type of volatile emissions, i.e., the persistent degassing, eruptions or the

48

H. Shinohara / Journal of Volcanology and Geothermal Research 268 (2013) 46–63

persistent degassing or eruptions (Andres and Kasgnoc, 1998). The SO2 flux from persistent degassing has been measured by ground-based UV remote sensing, and global fluxes are estimated by their compilation (Stoiber et al., 1987; Andres and Kasgnoc, 1998). Measurements of SO2 flux, however, are biased to volcanoes in populated areas, and the lack of the SO2 flux data from remote areas likely causes a significant underestimate, by a factor of two, of the global flux (Shinohara, 2008). An accurate estimate of the global flux also requires an evaluation of temporal variation and the contributions of numerous small-flux volcanoes. In order to obtain an accurate regional SO2 flux, Mori et al. (2013) compiled SO2 flux data from volcanoes in Japan for the 32 years from 1975 to 2006 and obtained a time-averaged flux for most of the degassing volcanoes in Japan, including small-flux volcanoes (Table 1). Because Japan is a small, densely populated country, fumarolic activities are well surveyed, and the compilation is believed to cover all volcanoes with measurable SO2 fluxes. The time-averaged SO2 fluxes were estimated based on the temporal variation obtained through repeated measurements. At some of the volcanoes, SO2 flux measurements have not been made frequently enough to evaluate temporal variation,

diffuse emissions. Arcs around Japan are selected for the estimate of the regional flux, because detailed data are available for a fairly large number of volcanoes in this region. Volcanoes in Japan are located along the arc system consisting of the Kurile Arc, Northeast Japan Arc, Izu–Bonin Arc, Southwest Japan Arc, and Ryukyu Arc (Fig. 1). Volcanic activity in Japan corresponds to about 10% of the global activity at subduction zones, 8% of that of global Holocene volcanoes, 12% of that of recent eruptions, and 9% of that across the entire trench length (Mori et al., 2013). The volcanic regions in Japan have been well populated and their surface manifestations, such as emissions of volcanic gases and hot spring discharges have been well quantified over a long period, which is a significant advantage for determining the regional flux estimates. 3. SO2 flux 3.1. Persistent SO2 flux from volcanoes in Japan Persistent degassing is the long-term large-flux continuous emission of volcanic gases. The majority of volcanic gases are discharged either by

c

ril

Daisetsusan Tokachi Jyozankei Niseko Usu

Atosanupuri

t Jap heas Nort

Tamagawa Iwate Sukawa+Naruko

Zao

Kusatsu+Manza+Shibu +Shima+Shiriyake

Azuma(+Numajiri) Nasu(+Shiobara)

Okuhida

pan est Ja

w

South Beppu+ Yufuinn Kuju Aso

Kirishima

yu

rc

Ar

c

Sakurajima

onin A

uk

Hakone Izu-Oshima Miyakejima

Satsuma-Iwojima Kuchinoerabujima Kuchinoshima Nakanoshima Suwanosejima

Ry

Tateshina Isawa

Ikaho Asama

Izu-B

Unzen

Arc

Ku

Meakandake(Akan) Tarumae Kuttara(Noboribetsu)

an A

rc

Shiretoko

Ar

0

(km)

500

Fig. 1. Locations of major degassing volcanoes and hot spring areas in Japan. Triangles represent volcanoes and circles represent hot spring areas. Many hot spring areas overlap with the locations of volcanoes, and a single name legend is used for such volcano-hot spring pairs. When the hot spring area has a different name from the adjacent volcano, the hot spring name is placed in parentheses, for example, “Meakandake (Akan)” is used for the Meakandake volcano and the Akan hot spring area. Some hot spring areas, such as Sukawa and Naruko are located within a small area, and the adjacent hot spring name is given after the plus sign.

H. Shinohara / Journal of Volcanology and Geothermal Research 268 (2013) 46–63

thus, the time-averaged SO2 fluxes were estimated for such volcanoes with the help of other monitoring records, such as variations of plume height (Mori et al., 2013). The time-averaged SO2 flux from volcanoes in Japan is estimated to be 6013 t/d, which includes the contribution of 2119 t/d from the intensive degassing of Miyakejima volcano, which started in 2000 (Table 1; Mori et al., 2013). The comprehensive compilation of SO2 flux data by Mori et al. (2013) also reveals that the frequency of the SO2 flux does not obey the power-law distribution and that the regional flux is controlled by a few large-flux volcanoes. The power-law distribution of volcanic gas fluxes was proposed by Brantley and Koepenick (1995) and was applied to global or regional SO2 flux estimates because of the limited SO2 flux dataset (Andres and Kasgnoc, 1998; Hilton et al., 2002). The comprehensive SO2 dataset for volcanoes in Japan, however, suggests that the dataset should not be extrapolated based on the power-law distribution assumption. In contrast, the contribution of small-flux volcanoes, such as those with fluxes of less than 100 t/d, is quite limited (less than 3% of the total flux from Japan) because the number of small-flux volcanoes is not so large as would be expected from the power-law distribution. The time-averaged SO2 flux from volcanoes in Japan obtained by Mori et al. (2013) is accurate because the fluxes obtained during a 32-year time period from almost all degassing volcanoes, including those with small fluxes, were considered for the compilation. The compilation, however, poses the problem of the uncertainty caused by the occurrence of infrequent large emissions, such as the intensive degassing of Miyakejima volcano. Intensive SO2 emission from Miyakejima volcano began and peaked in the year 2000, and then gradually decreased with time (Kazahaya et al., 2004). The total SO2 emission from Miyakejima during 2000–2006 amounts to 23 Mt, corresponding to the 32-year average of 2119 t/d that raised the regional SO2 flux by 50%. The frequency of such intensive degassing events is not known, and the contribution of such events to the time average is not clear. Because of the absence of its historical record, we assume that a similar intensive degassing event were to occur less than once in a thousand years in the region, then the contribution of such an event would be less than 64 t/d, which is not significant to the regional flux. In this study, the value excluding the contribution from Miyakejima (3894 t/d) is considered to be a Table 1 SO2 flux data from Japanese volcanoes compiled for 1975–2006 (Mori et al., 2013). Volcano

No. of observation days

Max flux (t/d)

Min flux (t/d)

Average flux (t/d)

Meakandake Atosanupuri Tokachidake Tarumae Kuttara Usu Iwate Azuma Nasu Asama Hakone Izu-Oshima Miyakejima Kuju Aso Unzen Kirishima Sakurajima Satsuma-Iwojima Kuchinoerabujima Nakanoshima Suwanosejima

4 1 2 5 1 21 1 1 2 141 1 6 280 16 123 19 3 107 16 3 1 14

12 n.d. 210 24 n.d. 169 0.1 n.d n.d 4600 n.d. 520 199,900 260 3750 230 20 5480 1760 48 40 1130

2.6 n.d. 140 4.6 n.d. n.d 0.1 n.d n.d 25 n.d. 30 1100 3 19 0.1 n.d. 110 260 43 40 300

6 0 175 15 0 8 0 6 0 360 0 30 2119 27 410 14 8 1641 574 1 40 579

Total Total w/o Miyakejima

6013 3894

n.d.: not detected. Detection limit is 0.1–5 t/d depending on distance of the measurement site from the plume.

49

representative regional flux, but the value including its contribution (6013 t/d) will also be discussed for a maximum estimate. 3.2. Global SO2 flux by persistent degassing The SO2 flux is an important criterion of volcanic activity (Stoiber et al., 1983), and the number of volcanoes for which SO2 flux measurement has been conducted significantly increased after the compilation by Andres and Kasgnoc (1998). Recent SO2 flux data reported in international journals are compiled and compared with the previously mentioned compilation in Table 2. The new global total flux by persistent degassing is 50 kt/d, twice as large as the previous estimate, and the number of the listed volcanoes increased from 49 to 76. Although the fluxes of many of the volcanoes listed in the previous study are revised, the sum of the new flux data for these volcanoes (23 kt/d) is similar to that of the previous estimate (26 kt/d). The two-fold difference in the total flux is due to the addition of largeflux volcanoes. The global SO2 flux from subduction zones is 42 kt/d, which is about 80% of the flux including fluxes from hot spots, rifts, and other subaerial volcanoes. Mt. Etna, in Italy, is one of the most intensively degassing volcanoes, contributing about 10% of the global SO2 flux, but its origin remains unclear and it is not included in the subduction zone flux in this study. The global subduction zone flux is controlled largely by large-flux volcanoes: 67% of the global flux is discharged by only 10 volcanoes with fluxes larger than 1000 t/d (Table 2). The SO2 flux from Japan corresponds to 10% of the global flux from subduction zones, but the proportion increases to 15% if the contribution from Miyakejima is included. A similarly large contribution of recent degassing to the global flux comes from the intensive degassing activity at Popocatépetl, in Mexico, which started in 1994 after a 70-year dormancy (Delgado-Granados et al., 2001). The effect of infrequent large emissions, however, is less likely to be important for the global flux than for the regional flux because the data compilation for a large number of volcanoes may compensate the small time window of observation with modern techniques. The global flux is largely controlled by large-flux volcanoes; hence, the accuracy of the total flux estimate depends on the flux-measurement coverage of large-flux volcanoes. Although the number of SO2 flux measurements is significantly increased, the measurements do not yet provide equal coverage for all regions. The SO2 flux measurements, for example, cover only four of the many degassing volcanoes in Kamchatka (Taran, 2009; Table 2). However, the coverage has been improved by application of satellite-based SO2 flux measurements. The Ozone Mapping Instrument (OMI), launched in 2004, has high sensitivity and spatial resolution for SO2-yield measurements, and improvement of the SO2-yield retrieval algorithm permits quantification of persistent SO2 emissions to the lower troposphere (Krotkov et al., 2006). OMI data are also useful for monitoring variation in the SO2 flux through time. In contrast, the accuracy of the estimated SO2 flux remains limited due to low sensitivity to tropospheric SO2: estimated SO2 fluxes from OMI data are several times lower than those obtained from ground-based measurements (Carn et al., 2008; Bani et al., 2012; McCormick et al., 2012). Although the accuracy is limited, the good coverage by OMI indicates that most of the large-flux volcanoes are already known and that the present global flux estimates would not be increased significantly as a result of the addition of new large-flux volcanoes. 3.3. SO2 flux from eruptions Satellite-based UV remote sensing has quantified SO2 yields by explosive eruptions since 1978, and the global annual flux has been estimated by compiling these data (Bluth et al., 1993; Carn et al., 2003). Occurrence of eruptions is heterogeneous in time, so a simple average of the fluxes measured during only a few decades will not result in an accurate annual flux. The frequency of eruptions and the eruption magnitude (volcanic explosivity index, VEI; Newhall and Self, 1982) are

50

H. Shinohara / Journal of Volcanology and Geothermal Research 268 (2013) 46–63

known to follow the power-law distribution (Simkin, 1993), and a linear correlation has been estimated between the VEI and measured SO2 yields (Pyle et al., 1996; Schnetzler et al., 1997). Using these correlations, the time-averaged annual SO2 fluxes from explosive eruptions can be estimated for each VEI scale (Pyle et al., 1996), and the total annual flux for 3 ≤ VEI ≤ 7 eruptions is calculated to be 1.3 Mt/y. This value includes the contributions of eruptions at all types of tectonic settings, but the majority of the explosive eruptions occurred at subduction zones. Out of the 122 eruptions with VEI ≥ 4 that occurred between 1500 and 1980, 103 eruptions occurred at subduction zones, suggesting that about 80% of the annual flux is likely from subduction zones. Consequently, the annual SO2 flux by explosive eruptions at subduction zones is estimated to be 1 Mt/y (2700 t/d), which is 15 times less than the flux by persistent degassing (Table 1; Shinohara, 2008). The SO2 fluxes from effusive eruptions are included in Table 2 and total to 1400 t/d, about one-half that of explosive eruptions. This summary is a crude estimate because some data marked as effusive eruption in Table 2 include fluxes occurring during persistent degassing and some other data not so marked include fluxes of effusive eruptions. The distinction between degassing by persistent degassing and by effusive eruption is not clear in some cases. Effusive eruptions, in particular dome growth, often occur during persistent degassing activity, such as at the Galeras volcano, Colombia (Zapata et al., 1997), and intensive degassing sometimes continues as persistent degassing, even during dome growth intervals, such as at the Sourfière Hills volcano, Montserrat (Shinohara, 2008). 4. Volatile flux by persistent degassing 4.1. Persistent degassing of volcanoes in Japan Volatile fluxes by persistent degassing can be estimated by multiplying the volcanic gas composition by the SO2 flux from the volcano. Because more than 90% of the total SO2 flux from the volcanoes in Japan is emitted from only five volcanoes, Sakurajima, SatsumaIwojima, Suwanosejima, Aso, and Asama (Table 1), precise volatile fluxes can be estimated if the composition of these volcanic gases is known. If Miyakejima volcano is included, 95% of the total SO2 flux is emitted from these six volcanoes. Most of these volcanoes are openvent degassing volcanoes, where the volcanic gases are not discharged from accessible fumaroles and volcanic gas data have been few. Recent development of the Multi-GAS technique, however, enabled estimation of these volcanic gas compositions by measurements of volcanic plumes (Aiuppa et al., 2005; Shinohara, 2005). To date, the volcanic gas compositions of all of these large-flux volcanoes in Japan were quantified, with the exception of Sakurajima volcano (Table 3; Supplement 1). Because of frequent vulcanian eruptions, the summit area of Sakurajima volcano cannot be approached and Multi-GAS measurements have not yet been performed. Volatile fluxes from each major degassing volcano have been calculated using the time-averaged SO2 flux and volcanic gas composition of each volcano (Table 4). The volatile fluxes from Sakurajima were calculated using the volcanic gas composition of Asama volcano with the assumption of similarity of these gas compositions, as both of these volcanoes are persistently degassing andesitic volcanoes. The validity of this assumption is supported by the similar volcanic gas compositions of another persistently degassing andesitic volcano, Suwanosejima (Fig. 2). The total volatile flux by persistent degassing of volcanoes in Japan was obtained from the sum of the fluxes from the major degassing volcanoes (Table 4), and the average composition of volcanic gases from Japan was calculated based on the total flux (Table 3). The accuracy of the total flux and average composition estimates largely depends on the assumption for Sakurajima volcanic gas composition because this volcano is the major degassing source with a 40% contribution to the total, excluding Miyakejima (Table 1). Similar average compositions for

the total flux were estimated when including the contribution from Miyakejima. The estimated average volcanic gas composition of volcanoes in Japan is more S-rich than the composition estimates of previous studies (Fig. 2; Hilton et al., 2002; Fischer, 2008). The previous estimates were based on the composition of high-temperature fumarolic gases, such as from Usu, Showa-Shinzan, Unzen, Satsuma-Iwojima and Tokachi volcanoes (Fig. 2). However, volcanic gas fluxes from these volcanoes except Satsuma-Iwojima, are not large (Table 1) and contribution to the average gas composition is limited. The high-temperature fumarolic gas compositions are scattered and different from those of the large flux volcanoes (Fig. 2). The disagreement indicates that high-temperature fumarolic gases do not always represent the average composition of volcano degassing. Since volcanic gas compositions are controlled by the conditions or history of magma degassing (Gerlach and Graeber, 1985; Giggenbach, 1996), even high-temperature fumarolic gases can vary with time and space (distribution of the gas emission sites). The estimates of the volcanic gas fluxes from volcanoes should be based on the composition of the large flux gas emission. The S-poor average compositions reported in the previous studies are largely due to the S-poor compositions of Usu and Showa-Shinzan fumarolic gases (Fig. 2). Although these are compositions of hightemperature fumarolic gases, these samples were collected several years after dome-forming eruptions, when the volcanic gas fluxes had already decreased. The SO2 flux decreased by ten times within the first four years after the 1978 eruption at the Usu volcano (Ohta et al., 1988). The S-poor and H2O-rich compositions of the Showa-Shinzan fumarolic gases are likely the result of gradual degassing of intruded dome magma and increase of meteoric water influxes (Symonds et al., 1996). The high-temperature fumarolic gas collected from Unzen volcano during the dome-forming eruption (Ohba et al., 1994), however, is S-poor and CO2-rich compared with volcanic gases from the major degassing volcanoes (Fig. 2). Ohba et al. (2008) modeled that the fumarolic gases do not represent bulk magma degassing but are likely gases separated from the magma at depth during magma ascent. The fumarolic gases were sampled from an intensively degassing vent located at the foot of the growing dome (Ohba et al., 1994), however, significant degassing was also observed at the top of the dome. Although the dome was not accessible for direct gas sampling, the plume from the dome sampled by an airship has 4–20 times larger Cl/S ratios than those of the fumarolic gases. The difference of the dome and fumarolic gas compositions was attributed to multi-step degassing and the bulk composition of volcanic gases emitted during the dome-forming eruption was estimated to be two to four times smaller CO2/H2O ratio than that of the fumarolic gases (Ohba et al., 2008). This example implies that high-temperature fumarolic gases do not always represent the composition of the magma degassing. 4.2. Persistent degassing at other subduction zones The gas composition of large-flux persistent degassing is not well constrained at other subduction zones even at present. Sixty-seven percent of the global flux is discharged from 10 volcanoes, however, none of these, except for Miyakejima, has been quantified sufficiently for their gas compositions. Among the volcanoes listed in Table 2, the bulk plume gas compositions were obtained only for some of these volcanoes and the high-temperature fumarolic gas compositions were obtained for some other volcanoes (Fig. 3). Compositions of hightemperature fumarolic gases, particularly those with low fluxes and not listed in Table 2, scatter across a wide range. Many fumarolic gases were sampled during short periods of active degassing preceding or following eruptions and their compositions may not represent those of the bulk degassing of the volcano, as was discussed in the previous section. Even fumarolic gases collected during the major degassing stage can have a composition different from the average gas composition

H. Shinohara / Journal of Volcanology and Geothermal Research 268 (2013) 46–63

51

Table 2 SO2 flux from continuously degassing volcanoes. Volcano

Country

Tectonic settinga

A&K SO2 fluxb (t/d)

New SO2 fluxc (t/d)

Ref.d

Etna Bagana Lascar Ruiz Sakura-jima Manam Yasur Kilauea ERZ Masaya Stromboli Langila Galeras Fuego San Cristobal Kikai Mayon White Island Pacaya Poas Ulawan Asama Bulusan Oshima Santa Maria Kilauea summit Colima Kuju Merapi Unzen Arenal Telica Erebus Aso Tangkubanparahu Momotombo Medvezhia Slamet Usu Augustine Vulcano Iliamna Erta Ale Izalco Santa Ana Lengai Tengger Sulfur bank Kverkfjöll Martin Nyragongo Soufrière Hills Popocatépetl Villarrica Llaima Tungurahua Reventador San Miguel Tacana Rabaul Pago Ambrym Ambae Lopevi Bezymyanny Karymsky Mutnovsky Shiveluch Klyuchevskoy Kizimen Zhupanovsky Ebeko Saryuchev Gorely

Italy PNG Chile Colombia Japan PNG Vanuatu USA Nicaragua Italy PNG Colombia Guatemala Nicaragua Japan Philippine NZ Guatemala Costa Rica PNG Japan Philippine Japan Guatemala USA Mexico Japan Indonesia Japan Costa Rica Nicaragua Antarctica Japan Indonesia Nicaragua Rusia Indonesia Japan USA Italy USA Ethiopia El Salvador El Salvador Kanya Indonesia USA Iceland USA Congo Montserrat Mexico Chile Chile Ecuador Ecuador El Salvador Guatemala PNG PNG Vanuatu Vanuatu Vanuatu Rusia Rusia Rusia Rusia Rusia Rusia Rusia Rusia Rusia Rusia

? S S S S S S H S S S S S S S S S S S S S S S S H S S S S S S H S S S S S S S S S R S S R S H R S R S S S S S S S S S S S S S S S S S S S S S S S

4000 3300 2400 1900 1900 920 900 800 790 730 690 650 640 590 570 530 520 510 500 480 370 370 270 230 220 140 140 140 130 110 84 79 76 75 73 68 58 56 48 44 22 21 20 20 16 14 7 2.6 2.6 – – – – – – – – – – – – – – – – – – – – – – – –

3590 2000 2400 1900 1640 180 633 1500$ 800 200 250 450* 280 690 574 530 430 1540 8 640 360 370 30 120$ 300 140 28 140$ 14$ 180$ 280 74 410 75 73 90 58 8 48 12 125 60 – 120 16 14 7 2.6 2.6 2600 574$ 7000 320 630 1460 450* 260 30 110 120 5400 2200 360 400$ 75 230 500# 300# 100# 100# 100# 100# 800

1) 2) 3)

Nevada del Ruiz

Satsuma-Iwojima

Izu-Oshima Santiaguito

Kudryavy

Ol Doinyo Bromo Kilauea

4) 2) 5) 6) 7) 8) 2) 9) 7) 7) 4) 10) 7) 7) 2) 4) 4) 7) 6) 4) 4) 7) 7) 11) 4)

12) 4) 13) 14) 15) 7) 7)

16) 17) 18) 19) 3) 20) 9) 7) 7) 4) 4) 5) 5) 5) 12) 12) 12) 12) 12) 12) 12) 12) 12) 21) (continued on next page)

52

H. Shinohara / Journal of Volcanology and Geothermal Research 268 (2013) 46–63

Table 2 (continued) Volcano

Country

Tectonic settinga

A&K SO2 fluxb (t/d)

New SO2 fluxc (t/d)

Ref.d

Miyakejima Suwanosejima Tokachi

Japan Japan Japan

S S S

– – –

2120 670 175

4) 4) 4)

26,226 21,081 1,146

50,576 42,427 4,560

Global Subduction zone Hot spot + Rift a

S: Subduction, H: Hot spot, R: Rift, ?: Tectonis setting of Etna is not yet clear. Andres and Kasgnoc (1998), “–” means no data in Andres and Kasgnoc (1998). “–” means negligible flux. If new data are not available, the data in Andres and Kasgnoc (1998) are listed. *: OMI based estimate, #: Estimated based on visual evalution of the plume size. $: With contiuous effusive eruption. d References for the new SO2 flux data. 1) Allard et al. (2006), 2) McGonigle et al. (2004), 3) Mather et al. (2004), 4) Mori et al. (2013), 5) Bani et al. (2012), 6) http://hvo.wr.usgs.gov/ hazards/FAQ_SO2-Vog-Ash/P1.html, 7) Mather et al. (2006), 8) Burton et al. (2007), 9) Carn et al. (2008), 10) Werner et al. (2008), 11) Oppenheimer and Kyle (2008), 12) Taran (2009), 13) Vita et al. (2012), 14) Werner et al (2011), 15) Sawyer et al. (2008b), 16) Sawyer et al. (2008a), 17) Christopher et al. (2010), 18) Witter et al. (2005), 19) Witter et al. (2004), 20) Arellano et al. (2008), and 21) Aiuppa et al. (2012). b c

of the volcano when gas emission also occurs at other fumaroles or vents. High-temperature fumarolic gases sampled at the foot of an active lava dome, such as the Merapi volcano, can be affected by multi-step degassing, as discussed for Unzen, if dominant degassing is also occurring at the top of the domes. Volcanic gases from large-flux persistent degassing have similar H2O/St molar ratios of approximately 50 and CO2/St ratios ranging from 1, which is similar to the average for Japan (JT in Fig. 3), to 6 for Stromboli volcano (Fig. 3a). The global average composition is assumed to lie in the middle of this range, that is H2O/SO2 = 50 and CO2/St = 2 (GT in Fig. 3). Fischer (2008) also estimated that the global average composition (GF in Fig. 3) is more CO2-rich than the average for Japan (JF in Fig. 3), however, their estimated compositions are both more CO2-rich than those of the present study. The St/Cl ratios of large-flux persistent degassing range from 1 at Stromboli to 3 at Villarrica, and the global average St/Cl ratio is estimated to be 2 (Fig. 3b). The St/Cl ratio of average volcanic gas for Japan is 5, implying that the volcanic gases in Japan are quite Cl-poor compared with those of other subduction zones. The difference between the volcanic gas compositions in Japan and the global average suggests that volcanic gas compositions have regionally varying characteristics. This suggestion is supported by the similarity of gas compositions of the volcanoes in the same regions, such as in Japan and Kamchatka (Fig. 3). Volcanoes in Italy are known to have CO2-rich gas compositions. Variation of the magmatic volatile composition can be due to variable contribution of source materials (Elkins et al., 2006; Ruscitto et al., 2012). The variation of gas compositions in other regions, however, has yet to be well characterized because the number of reliable plume gas composition data from the same regions remains limited. The accumulation of the plume composition measurements is necessary for further evaluation. Estimates of regional compositional characteristics are important not only for estimating the global average volcanic gas composition but also for estimating the global flux from other types of emissions, such as by Table 3 Volcanic gas composition of the major degassing volcanoes in Japan. Compositions are in mol ratios. H2O/SO2

CO2/SO2

HCl/SO2

H2S/SO2

Refa

Miyakejima Satsuma-Iwojima Asama Aso Suwanosejima

35 97 30 60 70

0.74 0.40 0.80 2.0 1.0

0.07 0.60 0.20 0.07 0.10

0.07 0.10 0.15 0.05 0.04

1) 2) 3) 4) 3)

Averageb Average with Miyakejima

49 44

0.91 0.85

0.24 0.18

0.11 0.10

a 1) Shinohara et al. (2003), 2) Shinohara et al. (2002), 3) Shinohara unpublished data, 4) Shinohara et al. (in press), 5) Shinohara (2005). b The average compositon is calculated based on the total flux in Table 4.

diffuse degassing, by extrapolation of regional flux data. Petrological and geochemical studies of subduction zone magmas indicate that significantly different magma series erupt from different arcs (Wallace, 2005). The evaluation of regional gas composition characteristics should be linked with the study of magma compositions because genetic links are likely to exist. 5. Diffuse degassing 5.1. Volatile flux through water discharges Hot springs are significant sources of emission of magmatic volatiles (Taran, 2009). The outflow rate, temperature, and composition of hot springs in Japan have been compiled for hot spring usage and geothermal exploitation (Sumi, 1977; Tsukamoto, 1982; Kimbara and Sakaguchi, 1989; Kimbara, 2005; Supplement 2). The volatile flux from each hot spring was calculated from these datasets (Supplement 2) and compiled to estimate the total flux from the volcanic areas. Volatile and heat fluxes from each hot spring area were calculated by multiplying the natural outflow rate (Tsukamoto, 1982) by the composition and the maximum temperature (Table S2). When more than one hot spring data was reported in the reference, the hot spring with the higher temperature and higher volatile concentration was listed. This compile calculates only the fluxes of volatiles dissolved in the hot spring waters but volatile fluxes by bubbling gases are not estimated because of quite poor data set, likely causing underestimate of CO2 and H2S fluxes. The hot springs listed in Table 6 do not cover all of the hot springs, so the total flux from the volcanic areas was estimated by scaling up based on the heat-flux estimate. The total heat flux from hot springs in Japan is calculated to be 7697 MW from the data compiled by Kimbara (2005) by the method of Sumi (1977), who calculated the heat flux of each hot spring by multiplying the maximum temperature and total outflow rate (natural and pumped Table 4 Volatile fluxes from the major degassing volcanoes in Japan. H2O

CO2

SO2

HCl

H2S

(t/d) Miyakejima Sakurajimaa Satsuma-Iwojima Asama Aso Suwanosejima

21,000 11,300 15,600 2500 6900 13,200

1070 880 160 200 560 460

2100 1600 570 360 410 670

86 180 190 41 16 37

83 130 30 29 11 13

Totalb Total with Miyakejimab

54,000 75,000

2500 3500

3900 6000

510 580

230 300

a

Based on Asama gas composition. The sum of the listed flux is scaled up by the coverage of the total SO2 flux; 91.5% for the total without Miyakejima and 94.5% with Miyakejima. b

H. Shinohara / Journal of Volcanology and Geothermal Research 268 (2013) 46–63

a)

b)

CO2 12.5

4

4 .(

20 02 )

CO2/St

al

CO2/Cl

2

2

et n

20

SS

CO2/St

10 US

UN

AS

Hi

lto

40

8

25

50

100

20 8

H2O/CO2

100

CO2

20

5

53

AS

SS

AM

200

SU JT

5 SU

0.5

0.5

MI

UN

1

1

JF

JF

MI

AM

US

1000

SI

0.2

TK

0.2

2.5

JT

1

TK SI

H2O/50

1000 400

200

100

50

25 10

2St

2St

50

20

10

5

H2O/St

2.5

1.25 0.5

10Cl

St/Cl

Fig. 2. Composition of volcanic gases from Japanese volcanoes. a) Relative H2O, CO2, and total S (SO2 + H2S) molar contents. b) Relative CO2, total S (SO2 + H2S), and Cl molar contents. Hexagons indicate a major degassing volcano: MI, Miyakejima; SU, Suwanosejima; AM, Asama; SI, Satsuma-Iwojima; and AS, Aso. Squares represent fumarolic gas: UN, Unzen (Ohba et al., 1994); SS, Showa-Shinzan (Nemoto et al., 1957); US, Usu (Giggenbach and Matsuo, 1991); and TK, Tokachi (Hirabayashi et al., 1990). Circles represent an average composition: JT, this study and JF, Fischer (2008). The line indicates the average CO2/St ratio of 6.5 obtained by Hilton et al. (2002).

flow rate) using a base temperature of 0 °C. Sumi (1977) compared hot spring distributions with the geological structure and estimated that hot springs in volcanic areas contribute 68% of the total heat flux from the hot springs, which corresponds to 5234 MW (Table 6). The sum of the heat fluxes of the hot springs listed in Table 6 is 2353 MW, which is 45% of the total flux in the volcanic areas. With an assumption that the coverage of the compiled data is equal for the heat and volatile fluxes, the Cl, S, and CO2 fluxes from hot springs in the volcanic areas are calculated to be 525, 234 and 147 t/d, respectively (Table 6). The Cl flux from hot springs is similar to that from persistent degassing, but the S and CO2 fluxes from the hot springs are several times less than those for persistent degassing (Tables 4 and 6). This contrast is consistent with the flux estimates for the Kuril–Kamchatka region obtained by Taran (2009). The CO2, S, and Cl fluxes by volcanic gas emissions at the Kamchatka–Kurile arcs are 2000, 1500, and 400 t/d and those by hot springs are 300, 400, and 600 t/d, respectively.

a)

b)

CO2

AL

200

lls

è fri

Hi

GH W

SH

Is

CO2/Cl

VU

WI

P2

MS

10

P1 MR

GT

KU GO NG VI

AV

JT PO

GA

1

1

JF MMYS

MU

1000

2

2

WI

20

AL

CO2/St

GF

d an

l

te

hi

u

So

GA P1

re

CO2/St

GW

50

100

4

4

ST MR

GF

40

8

VU

25

100

20 8

P2

H2O/CO2

CO2

20

5 12.5

The simple method used for the total volatile and heat-flux estimates includes various sources of error, and the values are probably overestimated. The hot spring outflow rate data (Tsukamoto, 1982) includes only data for hot springs in use, such as for bathing, agriculture, or industry. The exclusion of the unused hot springs, however, may not cause significant error because unused (or unregistered) hot springs with large flow rates are likely to be rare in Japan. The hot spring outflow rate data include discharges from wells not being pumped, which is not natural outflow and may cause overestimation. The use of the high-temperature and high-volatile content hot spring data for an average value causes overestimation of the volatile fluxes. Precise estimates of volatile fluxes were made at a few hot-spring areas with consideration of variation of composition and flow rate through time and space (Table 7). The present estimates agree well with the detailed estimates for the Shiobara and Tamagawa hot springs (Yoshiike, 1993; Kanroji et al., 1999), however, they are two to three times larger than

AU

KL

ST

AV

0.5

0.5 0.2

GW

NG MU PO AU

0.2

MM GT SH YS MS KU

JT

VI

JF

5 2.5

GO

1 KL

H2O/50

1000 400

200

100

H2O/St

50

25 10

2St

2St

50

20

10

5

2.5

1.25 0.5

10Cl

St/Cl

Fig. 3. Composition of volcanic gases from other subduction zones. a) Relative H2O, CO2, and total S (SO2 + H2S) molar contents. b) Relative CO2, total S (SO2 + H2S), and Cl molar contents. Hexagons indicate an open-vent degassing volcano with a large SO2 flux: VI, Villarrica, Chile (Shinohara and Witter, 2005); GO, Gorely, Russia (Aiuppa et al., 2012); YS, Yasur, Vanuatu (Métrich et al., 2011); MS, Masaya, Nicaragua (Martin et al., 2010); and Stromboli, Italy (Aiuppa et al., 2010). Squares are fumarolic gas: AU, Augustine, USA (Symonds et al., 1990); SH, St. Helens, USA (Symonds et al., 1994); PO, Poás, Costa Rica (Rowe et al., 1992); MM, Momotombo, Nicaragua (Menyailov et al., 1986a); GA, Galeras, Colombia (Martini, 1983); AV, Avacha, Russia (Taran et al., 1997), Kudryavy, Russia (Taran et al., 1995); KL, Klychevskoy, Russia (Taran et al., 1991); MU, Musnovsky, Russia (Taran et al., 1992); AL, Alaid, Russia (Menyailov et al., 1986b); WT, White Island, New Zealand (Giggenbach, 1996); NG, Ngauruhoe, New Zealand (Giggenbach, 1996); MR, Merapi, Indonesia (Giggenbach et al., 2001); and VU, Vulcano, Italy (Giggenbach, 1996). The blue-colored squares indicate fumarolic gases collected during major degassing events listed in Table 2. Circles are average compositions: JT, Japan average (this study); JF, Japan average by Fischer (2008); GT, global average (this study); GH, global average by Hilton et al. (2002); GF, global average by Fischer (2008); and GW, global average by Wallace (2005). The lines indicate the CO2/SO2 ratios obtained for the bulk plume compositions of Soufrière Hills volcano, Monserrat (Edmonds et al., 2010), White Island volcano, and New Zealand (Werner et al., 2008). The stars indicate the compositions of eruption gas obtained by Gerlach et al. (1996); P1 is obtained with Dv/m = 720 and P2 is obtained S with Dv/m = 140 (Table 10). S

54

H. Shinohara / Journal of Volcanology and Geothermal Research 268 (2013) 46–63

Table 5 Comparison of average volcanic gas compositions. Compositions are in mol ratios.

Japan Japan Japan Global Global Global Global a

Table 7 Volatile flux through hot springs, rivers or underground water.

H2O/St

CO2/St

HCl/St

Refa

45 128 – 50 115 250 56

0.8 2.2 6.5 2 6.1 5.0 5.7

0.2 0.5 – 0.45 0.5 – 0.7

1) 2) 3) 1) 2) 3) 4)

1) This study, 2) Fischer (2008), 3) Hilton et al. (2002), 4) Wallace (2005).

the detailed estimates for the Noboribetu and Kusatsu hot springs (Nishimura, 1966; Hirabayashi and Mizuhashi, 2004). This comparison suggests that the present estimate of the total volatile flux is likely overestimated.

Table 6 Summary of volatile and heat flux by hot springs in volcanic areas. Hot spring area

Cl

S

CO2

(t/d)

(t/d)

(t/d)

Natural heat fluxa (MW)

Shiretoko Atosanupuri Akan Daisetusan Jyouzankei Niseko Noboribetu Tamagawa Sukawa Naruko Zaou Azuma Numajiri Nasu Shiobara Ikaho Shima Kusatsu Manza Shiriyake Shibu Okuhida Tateshina Isawa Beppu Yufuin Kujyu Aso Kirishima

1.0 10.9 0.9 1.1 0.1 6.2 32.5 33.2 4.2 3.4 13.2 1.8 3.8 1.1 4.1 1.0 4.1 36.6 0.6 0.6 4.4 10.2 3.1 2.3 30.8 7.6 5.6 1.3 10.5

0.1 4.8 0.9 0.9 0.0 0.6 1.0 4.8 4.0 1.6 18.6 2.3 11.2 3.0 0.7 0.7 0.6 29.5 0.5 0.2 0.9 0.5 0.7 0.4 1.7 6.1 1.4 2.7 4.9

1.4 2.1 0.6 1.4 0.0 2.8 4.9 0.0 0.0 1.1 0.3 1.0 0.3 0.9 4.6 1.5 0.1 0.0 0.0 1.7 0.7 10.7 0.1 1.1 0.0 3.2 11.2 1.6 12.8

4.7 35.5 5.3 20.4 0.2 17.0 78.0 61.6 17.9 39.4 66.0 29.0 38.3 39.2 44.8 16.5 17.8 163.1 9.5 4.0 36.3 139.7 21.6 50.3 74.0 61.0 57.0 20.7 113.1

Sum

236

105

66

1282

Total hot spring heat flux from Japan (MW) Total hot spring heat flux from volcanic areas (MW)c Coveraged

Total volatile fluxe a

Total heat fluxb (MW) 6.4 55.8 19.2 34.9 50.9 23.3 112.7 61.6 17.9 55.9 81.2 36.0 49.3 43.5 74.7 18.9 17.8 163.1 9.5 4.0 55.0 184.8 88.2 51.3 586.2 226.0 77.0 32.0 115.8 2353 7697 5234 45%

Cl (t/d)

S (t/d)

CO2 (t/d)

525

234

147

Calculated with natural outflow rates of hot springs. Calculated with total outflow rates including the pumped flow rates. Calculated with an assumption that heat flux from volcanic areas is 68% of heat flux from Japan, based on the compilation by Sumi (1977). d % of heat flux covered by this data table compared with the total hot spring heat flux from volcanic areas (2362/5234). e Total volatile flux from volcanic areas calculated with the heat flux coverage. b c

Noboribetu Tamagawa Shiobara Kusatsu Iwate

Hot spring Hot spring Hot spring Hot spring Riverb Underground water

Cl (t/d)

S (t/d)

CO2 (t/d)

Refa

13.0 39.7 5.3 16 11 13

1.0 5.3 1.2 11 12 11

– – – – – 61

1) 2) 3) 4) 4) 5)

a

1) Nishimura (1966), 2) Yoshiike (1993), 3) Kanroji et al. (1999), 4) Hirabayashi and Mizuhashi (2004), 5) Ohwada et al. (2012). b Excluding the hot spring flux.

Hot springs are localized and major sources of Cl and S; however, detailed studies of the volatile flux through rivers or groundwater around volcanoes have revealed that considerable amounts of volatile components are discharged also by water outflow through more diffuse sources. Hirabayashi and Mizuhashi (2004) measured the Cl and S fluxes through rivers and revealed that the total fluxes from the Kusatsu hot spring area are about twice as large as the fluxes from the hot springs (Table 7). Only one-half of the volatiles is discharged from appreciable hot springs and another one-half is discharged from numerous small hot springs or directly input to rivers and groundwater. Similarly sized volatile fluxes were estimated based on intensive sampling of groundwater around the Iwate volcano, which is an active volcano but has only weak fumarolic activity and few hot spring discharges (Ohwada et al., 2012; Table 7). The volatile fluxes from cool discharges (cold springs) from Japanese volcanoes are estimated with an assumption that the fluxes similar to that from the Iwate volcano (Cl = 10, S = 10, CO2 = 60 t/d; Table 7) are typical fluxes at each active volcano with fumarolic activity. Since about 20 volcanoes in Japan have appreciable fumarolic activity (Table 1), the total fluxes are calculated as 200, 200, and 1200 t/d for Cl, S, and CO2, respectively. This is a crude estimate based on limited data, but it indicates that the Cl and S fluxes from cold springs are similar in order of magnitude to those from hot springs but that the CO2 flux is likely larger than that from hot springs. The importance of cold springs as a major source of CO2 emissions was also estimated in Oregon Cascades, USA (James et al., 1999). The estimated magmatic CO2 flux is 9.3 t/d/km of arc, which is about 20 times larger than that of Cl (0.05 t/d/km; Mariner et al., 1990). A large CO2 flux, compared with the Cl flux, through diffuse emissions was also obtained at the Taupo volcanic zone, New Zealand, where the fluxes of Cl and CO2 from geothermal areas along the 200-km long zone are estimated to be 300 and 1200 t/d (Hochstein, 1995; Seward and Kerrick, 1996). The significant S and Cl emissions from hot and cold springs suggest that H2O (magmatic water) fluxes from these sources are also likely significant. When spring water is a mixture of magmatic and meteoric water, the contribution of the magmatic waters to the spring waters can be estimated from the isotopic composition because magmatic water at subduction zones has a distinctive composition (Taran et al., 1989; Giggenbach, 1992). Estimates of magmatic water flux from springs, however, are not common because spring discharges are commonly substantially dominated by meteoric water and their isotopic compositions are controlled also by vapor separation, the 18 O isotopic exchange, and elevation effects on the meteoric water (Giggenbach, 1991). Mutou and Matsubaya (2002) and Yamamoto et al. (1997) estimated the contribution of magmatic water based on a linear correlation between the water isotopic composition and the Cl content in hot spring waters at the Tamagawa and Kusatsu hot springs in Japan and determined that the Cl content of the magmatic input is 20,000 ppm (H2O/Cl mol ratio = 100). This H2O/Cl ratio is about one-half the ratio of persistently degassing volcanic gases (H2O/Cl mol ratio = 225; Table 5). With the assumption that 20,000 ppm is a typical Cl content of the magmatic input to spring

H. Shinohara / Journal of Volcanology and Geothermal Research 268 (2013) 46–63

discharges, the magmatic water fluxes from spring discharges are calculated as 50 times larger than the Cl fluxes. The magmatic water fluxes from hot springs and cold springs are estimated as 26,000 and 10,000 t/d based on the Cl fluxes of 520 and 200 t/d, respectively. The estimated magmatic water output from spring discharges is comparable with that from persistent degassing (54,000 t/d; Table 4). The Cl contents of the magmatic input to geothermal fluids in Ecuador, Philippines and New Zealand were estimated as 28,000, 22,000, and 44,500 ppm, respectively (Gerardo et al., 1993; Giggenbach, 1995; Aguilera et al., 2005). The larger Cl contents than the Japanese examples are consistent with the finding that the persistent degassing gases from Japanese volcanoes are Cl-poor compared with the global average composition (Fig. 3b). 5.2. Soil CO2 flux Diffuse emission of magmatic CO2 through soils is often observed around volcanoes, and soil CO2 emission is considered as another important outlet for magmatic volatiles (Baubron et al., 1990; Allard et al., 1991; Chiodini et al., 1998). The magmatic soil CO2 flux measurements from volcanoes in Japan show a large variation from 0.1 to 450 t/d (Table 8). I could not identify any clear systematic variation of soil CO2 flux with volcanic activity. The magmatic soil CO2 fluxes from the intensively degassing volcanoes of Aso and SatsumaIwojima are low and are less than 20% of the summit fluxes. In contrast, relatively large fluxes of approximately 100 t/d were obtained for volcanoes not having significant degassing activity, such as at Hakkoda and Miyakejima in 1998 (before intensive degassing began in 2000). The largest magmatic CO2 flux of 450 t/d was estimated for Ioutou (also called Ogasawara Iwojima). The present activity of Ioutou is characterized by an extensive thermal anomaly with a total heat discharge of 5.4×108 J/s (Ehara, 1985) and repeated ground deformations of up to 50 cm/y (Kumagai and Takahashi, 1985). There are no hightemperature fumaroles or degassing vents, but several boiling-point fumaroles exist on the island (Notsu et al., 2005). The estimate of the regional soil CO2 flux is difficult because of the lack of comprehensive soil CO2 flux data and the lack of a systematic correlation of the soil CO2 flux with the degassing activity at summit areas. Based on the limited dataset, the following systematics were assumed to calculate the total soil CO2 flux from Japan: the soil CO2 fluxes from the intensively degassing volcanoes are 10% of the summit fluxes and the average soil CO2 flux of other weakly degassing volcanoes is 20 t/d, a median value of the magmatic soil CO2 fluxes from Tarumae, Showa-Shinzan, Hakkoda and Miyakejima (Table 8). The high flux at the Ioutou is considered as exceptional to be added to the total. Therefore,

Table 8 Soil CO2 flux from volcanoes in Japan. Volcano

Tarumae Showa-Shinzan Usu 1998 Usu 1999 Usu 2000 Hakkoda Kusatsu-Shirane Izu-Oshima Miyakejima 1998 Ioutou Aso Satsuma-Iwojima a

Magmatic soil CO2 fluxa (t/d)

Summit vent CO2 flux (t/d)

Refb

(km2)

Soil CO2 flux (t/d)

1.86 0.53 2.65 2.65 2 0.6 1.44 5 1 22 1.56 2.5

6 15 120 340 39 127 1.4 3 120 760 0.12 80

2.5 10 – – – 74 – – 35 450 – 20

10 – – – – – 3.1 – – – 470 160

1) 2) 3) 3) 3) 4) 4) 6) 7) 8) 9) 10)

Area

13

Magmatic contributions were estimated based on δ C-CO2. 1) Hernández et al. (2001a), 2) Hernández et al. (2006),3). Hernández et al. (2001b), 4) Hernández et al. (2003), 5) Hirabayashi et al. (2004), 6) Shimoike (1999), 7) Hernández et al. (2001c), 8) Notsu et al. (2005), 9) Saito et al. (2007), 10) Shimoike et al. (2002). b

55

the total soil CO2 flux is the sum of the flux from Ioutou volcano (450 t/d), the fluxes from the intensively degassing volcanoes (220 t/d; 10% in Table 4), and the fluxes from the other 17 degassing volcanoes in Table 1 (340 t/d). The total flux is estimated to be 1010 t/d, which is about half of the CO2 flux by persistent degassing and is similar to the flux from cold springs. Magmatic soil CO2 fluxes in other subduction zones also do not show a clear systematic correlation with degassing activity at summit areas; however, these do show similar variations as those observed in Japan (Table 9). The soil CO2 fluxes of intensively degassing volcanoes (such as Stromboli, Masaya, and White Island) are less than 10% of the summit degassing. Significant soil CO2 degassing was not detected around the intensively degassing Popocatepétl volcano, Mexico (Varley and Armienta, 2001). In contrast, high soil CO2 fluxes are also observed at volcanoes with minor summit degassing (such as at Vulcano, and Merapi) and at volcanoes without significant summit degassing (such as at Solfatara, Vesuvius, Teide, Santa Ana, and Cerro Negro). 6. Volatile flux from explosive eruptions The SO2 yields from explosive eruptions were measured by satellite remote sensing, and the average flux from subduction zones was estimated to be 2700 t/d (see Section 3.3). The SO2 flux from explosive eruptions is not important to the total budget at subduction zones because the total SO2 flux by persistent degassing is 15 times larger than that from explosive eruptions. Fluxes of other volatiles, such as H2O, CO2, and HCl, however, can be important if their contents in the volcanic gases discharged by explosive eruptions are significantly larger than those by persistent degassing. Therefore, composition estimates of volcanic gases discharged by explosive eruptions are necessary to evaluate their effect on the total budget. Measurements of volcanic gas composition are difficult during the explosive eruption because of the danger and the high ash content of the plume that causes various physical and chemical problems for experimental facilities. Remote sensing is a promising technique under such conditions, and several measurements have been conducted during Strombolian and lavafountaining eruptions (Allard et al., 2005; Oppenheimer et al., 2006; Burton et al., 2007). Application of remote sensing techniques to intensive explosive eruptions, such as Plinian eruptions, however, is quite difficult because of the danger, and a direct gas composition Table 9 Soil and groundwater CO2 flux from subduction zone volcanoes in the world. Volcano

Vulcano Solfatara Latera Ischia Vesuvius Stromboli Nisyros Nea Kameni, Santorini Teide, Tenerife White Island Mammoth Masaya Cerro Negro Santa Ana complex Merapi a

Summit vent CO2 flux (t/d)

Refa

(km2)

Soil CO2 flux (t/d)

Italy Italy Italy Italy Italy Italy Greece Greece

5.4 0.6 3.1 0.86 5.5 – 1.3 0.37

111 1521 350 9.1 151 225 65 15.4

362 – – – – 4500 – –

1) 2) 3) 4) 5) 6), 7) 8) 9)

Spain NZ USA Nicaragua Nicaragua El Salvador

0.5 0.2 0.15 0.01 0.58 7

437 9 240 20 2800 162

– 2600 – 1380 – 7

10) 11) 12) 13), 14), 15) 16) 17)

215

240

18)

Country

Indonesia

Area

1) Inguaggiato et al. (2012), 2) Cardellini et al. (2003), 3) Chiodini et al. (2007), 4) Chiodini et al. (2004), 5) Frondini et al.(2004), 6) Carapezza and Federico (2000), 7) Allard et al. (1994), 8) Brombach et al. (2001), 9) Chiodini et al. (1998), 10) Morrero et al. (2008), 11) Wardell et al. (2001), 12) Gerlach et al. (1998), 13) Lewicki et al. (2003), 14) Martin et al. (2010), 15) Mather et al. (2006), 16) Salazar et al. (2001), 17) Salazar et al. (2004), 18) Toutain et al. (2009).

56

H. Shinohara / Journal of Volcanology and Geothermal Research 268 (2013) 46–63

measurement has never been conducted for an intensive explosive eruption. The composition of the volcanic gases discharged during intensive explosive eruptions can be estimated based on the contents of preeruptive volatiles in magmas. If the volcanic gases are derived only from dissolved volatiles in erupted magma, the volatile yields from eruptions can be estimated from the difference between the volatile contents of melt inclusions and matrix glasses by multiplying by the volume of the erupted silicate melt. This petrological method was developed to evaluate the climatic effect of volcanic gas loading to the atmosphere by the past eruptions (Devine et al., 1984; Palais and Sigurdsson, 1989). Comparison of measured SO2 yields by the satellite remote sensing with petrological estimates of the sulfur yields, however, has revealed that the measured SO2 yields are significantly larger than the petrological estimates, which is called excess degassing (Wallace, 2001; Scaillet et al., 2003). Excess degassing is quite significant for eruptions of silicic magmas at subduction zones. The measured SO2 yields from silicic eruptions are commonly one to two orders of magnitude larger than the petrological estimates. In contrast, excess degassing is less significant for eruptions of basaltic magmas and is not common at hot spot or rift regions (Wallace, 2001; Scaillet et al., 2003; Shinohara, 2008). The source of the excess degassing is considered to be the pre-eruptive volatile phase (bubbles) accumulated in the magma chamber (Wallace and Gerlach, 1994). In the following sections, compositions of volcanic gases discharged by explosive eruptions are evaluated for the two extreme cases of silicic and mafic magmas. The relative contributions of each magma type to the SO2 flux are yet to be evaluated quantitatively. A volume-weighted composition histogram of Quaternary volcanic rocks in Japan revealed that the average composition of volcanic rocks is andesitic (Aramaki and Ui, 1978), implying similar contributions from mafic and silicic magmas. The global volatile flux will be estimated with the assumption that the eruptive degassing of mafic and silicic magmas contributes equally to the total SO2 flux from explosive eruptions.

6.1. Explosive eruptions of silicic magmas The composition of the pre-eruptive vapor phase was quantitatively evaluated for the 1991 Pinatubo eruption (Wallace and Gerlach, 1994; Gerlach et al., 1996). The climatic eruption of Mount Pinatubo, Philippines, on 15 June 1991 discharged about 5 km3 of dacitic magma (DRE) and 17 Mt of SO2. The S contents of the melt inclusions are quite low and the measured SO2 yield is about 100 times larger than the petrological estimate. Since H2O–CO2 contents in the melt inclusions indicate vapor saturation in the pre-eruptive magma chamber, the preeruptive vapor phase was determined to be the source of the excess degassing (Wallace and Gerlach, 1994). The composition of the vapor phase was calculated based on the volatile contents of the melt inclusions and the solubility or partition coefficients of H2O, CO2, S, and Cl, and the fraction of the vapor phase in the erupted magma was calculated to account for the SO2 yield. Then, the contributions of the pre-eruptive vapor phase to the H2O, CO2, and Cl yields from the eruption were estimated (Table 10; Gerlach et al., 1996). Because the experimentally obtained vapor-melt S distribution coefficients (Dv/m = Cvapor /Cmelt ) were poorly constrained, they calculated two sets S S S of results with different values of Dv/m S . The eruption discharges, not only the pre-eruptive vapor phase, but also the volatiles exsolved from the melt and the syn-eruptive degassing of dissolved volatiles were also considered to estimate the total volatile emission. Syneruptive exsolution is significant only for H2O. The composition of volcanic gases emitted by the 1991 Pinatubo eruption (eruption gas) calculated with the large Dv/m is similar but S content twice as low S compared with the composition of the persistent degassing, and the S deficiency is more significant for the composition calculated with the small Dv/m (Fig. 3). S

Table 10 Composition of the melt inclusions, the pre-eruptive vapor phase and the erupted gas of the 1991 Pinatubo eruption (Gerlach et al., 1996) and the average silicic magma. H2O

CO2

SO2

Cl

Melt inclusions (ppm)a Matrix glass (ppm)b Δ(MI − MG) (ppm)c Emission from melt (Mt)d

60,000 3000 57,000 359

400 – 400 2.5

140 120 20 0.13

1100b 1100 0 0

= 720e Dv/m S Vapor phase composition (mol%)b Emission of the pre-eruptive vapor (Mt)b Total volatile emission (Mt) Dv/m = 140e S Vapor phase composition (mol%)b Emission of the pre-eruptive vapor (Mt)b Total volatile emission (Mt)

Vapor phase fraction = 1.3 (wt.%) or 4 (vol.%) 80 14.5 4 1.4 96 43 17 3.2 455 46 17 3.2 Vapor phase fraction = 6.6 (wt.%) or 20 (vol.%) 83 15.1 0.8 1.2 526 234 17 16 885

237

17

16

a

Wallace and Gerlach (1994). Gerlach et al. (1996). Difference between the volatile contents in the melt inclusions and the matrix glasses. The CO2 content in the matrix glasses is assumed to be zero. d Calculated with the erupted melt volume of 6300 Mt from Gerlach et al. (1996). e Distribution coefficient of S between the vapor phase and the silicate melt. b c

Recent experimental results (Webster and Botcharnikov, 2011) indicate that the actual Dv/m is similar to or even smaller than the S small Dv/m (= 140) applied by Gerlach et al. (1996). The Dv/m for S S haplogranitic melt decreases with increasing oxygen fugacity from 470 under the reduced conditions near NNO-2 to 47 under the oxidized conditions near NNO + 1 at 850 °C and 200 MPa (Keppler., 2010; NNO NNO + 1 implies that logfO2 = logfNNO O2 + 1, where logfO2 is the oxygen v/m fugacity at the NNO buffer.) Similar DS values (~200) were obtained at NNO + 1, 900 °C and 200 MPa, by Webster et al. (2011). Because the 1991 Pinatubo magma is oxidized with NNO + 3 (Rutherford and Devine, 1996), the small Dv/m should be applied in the calculation. The S H2O/SO2, CO2/SO2, and Cl/SO2 mol ratios of the eruptive gas are 185, 20, and 1.7, respectively (Table 10). The study of the 1991 Pinatubo eruption is a rare study providing a quantitative estimate of the pre-eruptive vapor phase composition, however, the pre-eruptive vapor phase of other explosive eruptions of silicic magmas can be evaluated with methods similar to those used by Gerlach et al. (1996). The vapor phase composition is related to volatile contents in melt inclusions and volatile solubility and distribution

v

XCO2=0.50

v

XCO2=0.30 v

XCO2=0.20

v

XCO2=0.10 v

XCO2=0.05

Fig. 4. Variation of H2O and CO2 concentrations in melt inclusions from subduction zone dacites and rhyolites. The thin-line curves and the thick-line curves are gas-phase saturation isobars and CO2 mole fractions of the gas phase calculated using Newman and Lowenstern (2002). Modified from Wallace (2005).

H. Shinohara / Journal of Volcanology and Geothermal Research 268 (2013) 46–63

57

Fig. 5. Variation of S and Cl concentration in melt inclusions and their difference between melt inclusions and matrix glass as a function of SiO2. All data are for subduction zone magmas. Circles represent melt inclusion concentrations and diamonds represent concentration differences between melt inclusions and matrix glass. Data are taken from following sources: Anderson et al. (1989), Bertagnini et al. (2003), Cervantes and Wallace (2003a,b), Clague et al. (1995), Devine et al. (1984), Gerlach et al. (1994), Gurenko et al. (2005), Harris and Anderson (1984), Luhr (2001), Matthews et al. (1999), Michaud et al. (2000), Palais and Sigurdsson (1989), Roggensack et al. (1996), Roggensack et al. (1997), Roggensack (2001), Saito et al. (2001), Saito et al. (2005), Satoh et al. (2003), Self and King (1996), Sigurdsson et al. (1990), Sisson and Bronto (1998), Sisson and Layne (1993), Stix et al. (2003), Wade et al. (2006), Walker et al. (2003), Westrich and Gerlach (1992), Witter et al. (2004), and Witter et al. (2005).

coefficients. Variation in the H2O–CO2 concentrations in melt inclusions of silicic magmas indicates that these magmas are vapor-saturated with saturation pressures of 100–250 MPa and X vapor of the pre-eruptive CO2 vapor from 20% to b5% (Wallace, 2005; Fig. 4). The S and Cl contents of melt inclusions of silicic magmas are similar to those of Pinatubo: S = 100 ppm and Cl = 1500 ppm for rhyolitic melts (Fig. 5). With similar distribution coefficients as applied to the Pinatubo case, the pre-eruptive vapor phase compositions of other silicic magmas were estimated to be similar as that for Pinatubo, but some volcanoes, such as Katmai and Krakatau, likely have a vapor phase with a several times smaller X vapor CO2 than that of Pinatubo. Based on these comparison, the average gas composition of silicic explosive eruptions is assumed as H2O/SO2 = 185, CO2/SO2 = 10, and Cl/SO2 = 1.7 (mol ratios), and the H2O, CO2, and Cl fluxes are calculated as 68,000, 9300, and 1200 t/d, respectively, based on the S flux of 675 t/d (Table 11). The water-soluble S and Cl on ash also contribute to volatile emissions. Although a systematic dataset is not yet available, a contribution to the total S emission ranging from 30 to 50% was obtained for several large explosive eruptions, such as at Mount St. Helens in 1980 (Gerlach and McGee, 1994), El Chichón in 1982 (Luhr et al., 1984), and Galungung in 1982–1983 (de Hoog et al., 2001). The S flux was increased by 50% to account for the contribution of the ash leachate (Table 11). These estimates, however, are based on few datasets and need to be revised with more studies with the comprehensive data as for the Pinatubo eruption. In particular, the highly oxidized composition of the Pinatubo magma suggests that the estimate Table 11 Global volatile fluxes from subduction zone volcanoes. H2O (t/d) Flux from Japan Persistent degassing Hot spring Cold spring Soil degassing

49,000 26,000 10,000 –

Global flux Persistent degassing Hot spring Cold spring Soil degassing Eruption Silicic magmas Mafic magmas Global/Japan flux ratio

1,230,000 650,000 350,000 130,000 – 98,000 68,000 30,000 13

–: not determined.

CO2 (t/d) 2300 150 1200 1010 145,000 63,000 4100 33,000 28,000 17,000 9300 6500 27

S (t/d)

Cl (t/d)

2000 230 200 –

470 520 200 –

30,000 23,000 2600 2300 – 2000 1000 1000 12

29,000 11,000 12,000 4700 – 1600 1200 440 23

based on the Pinatubo data may not be representative of the subduction magma degassing. Many volcanoes characterized by the highly oxidized magmas, such as El Chichón, Lascar and Nevado del Ruiz, discharge large quantities of SO2 (de Hoog et al., 2004). The comprehensive studies are necessary also to understand the effect of the magma oxidation state on the eruptive degassing. 6.2. Explosive eruptions of mafic magmas The agreement of the SO2 yields obtained by the petrological method and by satellite remote sensing indicates that the pre-eruptive vapor is not an important S source for basaltic explosive eruptions (Scaillet et al., 2003; Shinohara, 2008). In contrast, the CO2 contents in melt inclusions suggest that subduction zone basaltic magmas are typically saturated with a CO2-rich vapor phase in the crust and that the melt inclusions do not preserve their original CO2 contents (Fischer and Marty, 2005; Wallace, 2005). The contradicting conclusions of saturation with a CO2-rich vapor and lack of S excess degassing might be caused by a small Dv/m and a small amount of pre-eruptive vapor phase in basaltic S systems. Recent experiments by Lesne et al. (2011) determined that the Dv/m in basaltic systems is quite low (b10) at pressures higher S than 50 MPa. With Dv/m b 10 and a pre-eruptive vapor phase amount S of 1 wt.%, the vapor phase contains less than 10% of the S in the magma, which is insignificant in the S budget. The difference between the volatile contents in melt inclusions and matrix glasses indicates that melts discharge about 1500 ppm S and 1000 ppm Cl by syn-eruptive degassing of the basaltic magmas, corresponding to the Cl/S mol ratio of 0.6 in the erupted gases (Fig. 4). Emissions of CO2-rich gases were observed during small-scale explosive eruptions of Stromboli volcano, Italy, and the CO2-rich gases were interpreted as the vapor phase accumulated in a magma chamber at depth (Burton et al., 2007; Aiuppa et al., 2010). The average composition of the volcanic gases, mainly discharged by persistent degassing, is H2O = 80, CO2 = 17, and SO2 = 3 mol%, whereas the gases discharged by the explosive eruptions have lower H2O/CO2 and higher CO2/SO2 molar ratios of 1–6 and 10–47, respectively (Aiuppa et al., 2010). Using the intermediate values of H2O/CO2 = 2 and CO2/ SO2 = 20, the H2O, CO2, and SO2 contents of the eruption gas are 66, 33, and 1.6 mol%, respectively. This example will be used to estimate the global volatile flux by basaltic explosive eruptions. The Stromboli volcano, however, is a CO2-rich volcano whose persistent degassing gases have one-half the H2O/SO2 ratio and three times the CO2/SO2 ratio of the global average (Fig. 3). Considering this difference, we estimate that the average mol ratios of basaltic explosive eruption

58

H. Shinohara / Journal of Volcanology and Geothermal Research 268 (2013) 46–63

gases are H2O/SO2 = 80, CO2/SO2 = 7, and Cl/SO2 = 0.6 and calculated the volatile flux by the mafic explosive eruptions (Table 11). The studied eruptions at the Stromboli volcanoes are relatively small in scale and future tests are necessary to determine the similarity to the gas compositions of much larger basaltic Plinian eruptions.

due to the large contribution of hot spring discharge. The estimated H2O flux is significantly less than previous estimates because the measured H2O/SO2 ratios of persistent degassing are significantly smaller than the ratio estimated from fumarolic gas compositions used by the previous studies (Table 5).

7. Volatile budget for subduction zone volcanism 7.2. Magmatic volatile differentiation in the crust 7.1. Global volatile flux from subduction zones The global volatile flux from subduction zone volcanoes was calculated with various fluxes, including those from persistent degassing, hot springs, cold springs, soil degassing, and eruptions (Table 11). The Japanese and global SO2 fluxes from persistent degassing are fairly well estimated at 1800 and 21,000 t/d as S (Tables 1 and 2), and the total S fluxes (including H2S) are estimated at 2000 and 23,000 t/d, respectively, with an average H2S/SO2 mol ratio of 0.1 obtained for volcanoes in Japan (Table 3). The global volatile fluxes from persistent degassing were obtained based on the average gas composition and the S flux (Tables 5 and 11). The global fluxes of H2O, CO2, S, and Cl, by the persistent degassing are larger than the Japanese fluxes by factors of 13, 27, 12, and 23, respectively. The global volatile fluxes from hot springs, cold springs, and soil gases were estimated by multiplying the Japanese flux from each source with these factors (Table 11). The contributions of the various emission sources are different for the fluxes of different volatile components (Table 11). Persistent degassing is the major source for all components, but its contribution is variable: 80% for S, 50% for H2O, and 40% for CO2 and Cl. Contributions of hot springs are 30 and 50% for H2O and Cl, respectively, and both cold springs and soil degassing contributes 20% flux of CO2. Explosive eruptions are not a major source of the volatile fluxes, with a contribution of 15% or less. The uncertainty in the estimated fluxes varies with the different types of emissions. The global S flux is well constrained by the fairly good coverage of the dataset of SO2 fluxes by persistent degassing and eruptions. The limited number of composition measurements of persistent degassing, except in Japan, is the major source of uncertainty of its volatile flux; however, the accumulation of persistent degassing compositions by applying Multi-GAS techniques will significantly reduce the error because the flux by persistent degassing is controlled by a limited number of emission sources. The flux estimates of diffuse degassing are poorly constrained, especially for cold spring and soil-gas discharges, and further accumulation of data is necessary to confirm their contributions to the total flux. The global fluxes obtained by this study are apparently similar for CO2 and S and different for H2O and Cl compared with previous studies (Table 12). The S flux obtained by Hilton et al. (2002) agrees with the present estimate, but this is likely by chance, as their estimate was based on the invalid assumption of power-law distribution of the SO2 flux (Mori et al., 2013). The total CO2 flux is similar to those by Hilton et al. (2002) and Fischer (2008), but their values do not include the contribution of diffuse emissions and the present CO2 flux estimate excluding diffuse emission is about one-half of the previous estimates. In contrast, a Cl flux similar to that by Fischer (2008) was estimated for persistent degassing; however, the global Cl flux is twice as large Table 12 Comparison of the global volatile fluxes estimated by previous studies.

This study Hilton et al. (2002) Fischer (2008) Varekamp et al. (1992) Sano and Williams (1996) Marty and Tolstikhin (1998) –: not determined.

H2O (t/d)

CO2 (t/d)

S (t/d)

Cl (t/d)

1,250,000 4,000,000 1,800,000 – – –

145,000 170,000 230,000 180,000 370,000 300,000

30,000 28,000 – – – –

29,000 – 15,000 – – –

Volatile components are supplied from the mantle to the crust as dissolved volatiles in primitive magmas, and their contents in subduction zone magmas are estimated based on melt inclusion analyses (Wallace, 2005; Sadofsky et al., 2008; Ruscitto et al., 2012). Previous studies estimated these volatile fluxes based on the volatile contents in primitive magmas and the supply rate of primitive magma with an assumption of complete degassing. With the same assumption, the volatile contents in the primitive magmas can be calculated with the compiled volatile flux and the primitive magma supply rate (2.7 km3/y; Scholl and von Huene, 2009). The calculated volatile contents are similar for CO2 and Cl but twice as large for H2O and Cl compared with those obtained from melt inclusions (Table 13). Since most melt inclusions are trapped under CO2-saturated conditions, the melt inclusions do not preserve the primitive magma CO2 content. Wallace (2005) estimated the CO2 content in the primitive magma by dividing the CO2 flux of Hilton et al. (2002) with the primitive magma flux; therefore the agreement of the CO2 content is simply due to the apparent agreement of the CO2 flux by Hilton et al. (2002) and that by this study. Volatile contents in volcanic and igneous plutonic rocks are not null and should be considered in the volatile budget of the crust. The average S and Cl contents in the bulk continental crust are estimated at 400 and 240 ppm, respectively (Rudnick and Gao, 2003). The H2O contents of igneous rocks are controlled by the abundance of hydrous minerals, and the average H2O content is about 0.7 wt.% (Nesbitt and Young, 1984; Wedepohl, 1995). The average C content in igneous plutonic rocks was calculated from data compiled by Wedepohl (1995) to be 500 ppm (1000 ppm as CO2). Isotope studies have indicated that the majority of the C is likely derived from contamination by near-surface materials (Craig, 1953; Hoefs, 1973), and the average C content is likely to be largely overestimated. This value, however, was used for the calculation because the overestimation does not significantly alter the result because the degassed CO2 content is quite large. The sum of the volatile contents in igneous rocks and the degassed contents are the pre-degassed volatile contents of the erupted and intruded magmas. Contributions of igneous rock contents relative to pre-degassing contents are not significant, ranging from 9% for H2O to 19% for S (Table 13).

Table 13 Volatile contents in the primitive magmas and other sinks.

Primitive magma melt inclusion Sadofsky et al. (2008) Wallace (2005) Ruscitto et al. (2012) Degassed volatile contents in the igneous crust (ppm)b Volatile contents in igneous rocks (ppm) Pre-degassing volatile contents in the igneous crust (ppm)c Volatile contents in the lower crust (ppm) Volatile contents in the primitive magmas (ppm)d

H2O (ppm)

CO2 (ppm)

S (ppm)

Cl (ppm)

36,000 23,000 33,000

– 10,000a –

2100 1300 1600

800 1000 900

70,000

8000

1700

1600

7000 77,000

1000 9000

400 2100

240 1800

7000 30,000

1000 3700

340 910

250 780

–: not determined. a Calculated from the CO2 flux by Hilton et al. (2002) and magma supply rate. b Calculated with the magma supply rate of 2.7 km3/y. c A sum of the degassed content and the content in the igneous rocks. d The lower crust foundering rate is assumed as twice of the crustal growth rate.

H. Shinohara / Journal of Volcanology and Geothermal Research 268 (2013) 46–63

The composition difference between the andesitic continental crust and the basaltic primitive magmas requires that mafic–ultramafic components, such as the lower crust, return to the mantle (lower crust delamination or foundering; Kelemen et al., 2003; Rudnick and Gao, 2003). Lower crust foundering is another sink of volatile components supplied by primitive magmas. Volatile contents of the foundering material are quite uncertain, as the foundering materials themselves are not yet specified, and the volatile contents of fresh ultramafic rocks are also not yet well constrained. Volatile contents in lower crust are applied for the calculation. The average S and Cl contents in a basaltic-andesite lower crust are 350 and 250 ppm, respectively (Rudnick and Gao, 2003), and the H2O and C contents are assumed to be the same as those of the average crust (Table 13). There are few estimates of the volume rate of lower crust foundering. Based on petrological modeling, the volume of mafic restites and cumulates was estimated to be 3–9 times larger than that of the seismically defined lower crust at the Izu–Bonin–Mariana (IBM) arc (Tatsumi et al., 2008), implying that the lower crust foundering rate is 100–200% of the crustal growth rate. Because the average volcanic rock composition of the IBM arc is more mafic than other arcs in Japan (Aramaki and Ui, 1978; Tamura and Tatsumi, 2002), a larger foundering rate than that at the IBM arc is likely common. The foundering rate was assumed to be twice the crustal growth rate for the calculation of the primitive magma volatile contents (Table 13). The estimated primitive magma volatile composition has lower S but similar H2O and Cl content than those from melt inclusion studies (Wallace, 2005; Sadofsky et al., 2008; Ruscitto et al., 2012; Table 13). The uncertainty in the foundering rate causes a large uncertainty in the estimated volatile composition. A smaller foundering rate results in larger volatile contents. The low estimated S content is likely due to underestimation of the S content in the foundering materials. Although the S content estimated for lower crust with a basalticandesite composition was applied for the calculation, the foundering materials are likely more mafic and richer in S. Sulfur contents of igneous plutonic rocks increase from 10 ppm in granites to 1000 ppm in gabbros (Terashima and Ishihara, 1986), which is similar to the variation in melt inclusions (Fig. 5), and the average S content of ultramafic rocks is as high as 3000 ppm (Schneider, 1970). If the S content of foundering material is 2000 ppm, the same value as the content in the pre-degassing volatile content, the S content of primitive magma is 2000 ppm regardless of foundering rate. The consideration of the lower crust foundering results in a lower CO2 content for primitive magma than that of the previous study (Wallace, 2005). The revision of the magma budget model at subduction zones also alters the previous volatile fluxes estimated based on the 3He flux from subduction zones (Varekamp et al., 1992; Sano and Williams, 1996; Marty and Tolstikhin, 1998; Sano et al., 2001). For example, the CO2 fluxes were calculated by multiplying the 3He flux by the CO2/3He ratio of fumarolic gases with an assumption that this ratio is conservative during chemical differentiation because of the similar chemical behaviors of He and CO2. The estimated CO2 fluxes range from similar to three times larger values than the present estimate (Table 12). The previous estimates were obtained on the basis of the 3He flux calculated by Torgersen (1989), who assumed that the 3 He flux from subduction zones is about 20% of that from the MOR (1000 mol/y; Craig et al., 1975) because the magma production rate at subduction zones is considered as about 20% that at the MOR (3.0–8.6 km3/y vs 21 km3/y; Crisp, 1984). However, recent studies concluded that the MOR 3He flux is 500 mol/y (Bianchi et al., 2010) and that the magma production rate (crustal growth rate) is 2.7 km3/y (Scholl and von Huene, 2009); therefore the 3He flux was reduced by three times (65 mol/y). This revision reduces all the previous volatile flux estimates obtained with this method by a factor of three. In contrast, lower crust foundering implies that the production rate of primitive magma is likely a few times larger than the crustal growth rate, and the flux values themselves may become similar to those

59

obtained by the previous studies. The 3He approach also relies on an assumption that the 3He content of primitive magmas at subduction zones is similar to that of MOR. As the petrogeneses of these magmas are different, this assumption also needs to be tested. 8. Summary The global fluxes of H2O, CO2, S, and Cl from subduction zones were estimated by compiling fluxes from various sources, including persistent degassing, hot and cold springs, soil degassing, and eruptions. A compilation of recently published SO2 fluxes by persistent degassing revealed that the global SO2 flux from subduction zone volcanoes is 42 kt/d, twice that of the previous estimate (Andres and Kasgnoc, 1998). The SO2 flux from Japan arcs is 3.9 kt/d, corresponding to 9% of the global flux. Fluxes of other volatiles by persistent degassing were estimated by multiplying the composition of the persistent degassing by the SO2 flux because the large-flux volcanoes control the average composition. The estimated global fluxes of H2O and CO2 are about one-half those of previous estimates obtained based on fumarolic gas compositions, indicating the importance of gas-composition measurements of persistent degassing. The volatile flux from hot springs in Japan was estimated based on a comprehensive dataset of hot-spring discharge rates, temperatures, and compositions. Data for the volatile flux through cold spring discharges is scarce but the measured S and Cl fluxes are similar to that through hot springs at actively degassing volcanoes. The cold-spring volatile flux from Japan arcs was estimated by multiplying the measured flux at one of the degassing volcanoes by the number of actively degassing volcanoes in Japan, with the assumption of similar discharges at every actively degassing volcano. Magmatic soil CO2 fluxes were measured at 10 volcanoes in Japan and range from less than 1 to 450 t/d. The total flux from Japan arcs was estimated with the assumption of the following systematics suggested from the measured fluxes: the soil CO2 flux is 10% of the summit plume flux at the persistently degassing volcanoes and other degassing volcanoes have an average discharge of 20 t/d. The regional fluxes from Japan arcs were then extrapolated to the global scale. The flux compilation indicates that persistent degassing is the major volatile source, especially for S; whereas diffuse emissions are similarly significant sources of H2O, CO2, and Cl with contributions of about 40%. The contribution of explosive eruptions is less than 15% for all the volatiles. The estimates of diffuse degassing fluxes include large errors due to the limited number of data; however, the potential impact of these sources on the global flux indicates the importance of further studies to quantify these fluxes. Persistent degassing is characterized by focused intensive emission sources, and the global flux is controlled by a small number of large emission sources. Accuracy of the global flux estimate will be significantly improved by adding several new gas composition measurements at large-flux volcanoes. Volcanic gas compositions likely have regional characteristics, and the differences in gas composition likely affect the volatile flux from diffusive discharges (e.g., a larger soil CO2 flux likely occurs in regions where the volcanic gases are more CO2-rich than in other regions). Improvement of the composition database of persistent degassing will enhance the understanding of the volatile composition characteristics of each arc, which provides better constraints for extrapolation of regional flux data to the global flux. The volatile budget of subduction zone volcanism was evaluated by comparing the volatile flux, volatile contents in the crust, and primitive magma volatile contents. The contributions of volatiles remaining in the crust are not significant compared with degassed volatiles. In contrast, consideration of lower crust foundering largely alters the volatile budget evaluation. The pre-degassing H2O and Cl contents in magmas estimated from the volatile flux and crustal growth rate are twice as large as those in primitive melt inclusions, with the assumption that the primitive magma supply rate is equal to the crustal growth rate. The disagreement is likely due to underestimation of the primitive

60

H. Shinohara / Journal of Volcanology and Geothermal Research 268 (2013) 46–63

magma supply rate by neglecting lower crust foundering. Consideration of the lower crust foundering rate also alters the volatile flux estimates based on the 3He degassing. Further studies of the lower crust foundering rate and volatile contents in foundering materials are necessary to improve the volatile budget estimates at subduction zones. Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.jvolgeores.2013.10.007. Acknowledgment I would like to thank Drs. Tim Horscroft and Sandro Aiuppa for inviting this review, Dr. T. Fischer and an anonymous reviewer for careful and constructive comments, Drs. O. Ishizuka, T. Mori, K. Kazahaya, M. Ohwada, N. Morikawa, H. A. Takahashi and M. Takahashi for their discussion. This work is partially supported by JSPS KAKENHI 25287115. References Aguilera, E., Cioni, R., Gherardi, F., Magro, G., Marini, L., Pang, Z., 2005. Chemical and isotope characteristics of the Chachimbiro geothermal fluids (Ecuador). Geothermal 34, 495–517. Aiuppa, A., Federico, C., Giudice, G., Gurrieri, S., 2005. Chemical mapping of a fumarolic field: La Fossa Crater, Vulcano Island (Aeolian Islands, Italy). Geophys. Res. Lett. 32 (13), L13309. http://dx.doi.org/10.1029/2005GL023207. Aiuppa, A., Bertagnini, A., Métrich, N., Moretti, R., Di Muro, A., Liuzzo, M., Tamburello, G., 2010. A model of degassing for Stromboli volcano. Earth Planet. Sci. Lett. 295, 195–204. Aiuppa, A., Giudice, G., Liuzzo, M., Tamburello, G., Allard, P., Calabrese, S., Chaplygin, I., McGonigle, A.J.S., Taran, Y., 2012. First volatile inventory for Gorely volcano, Kamchatka. Geophys. Res. Lett. 39, L06307. http://dx.doi.org/10.1029/2012GL051177. Allard, P., Carbonnelle, J., Dajlevic, D., Bronce, J.Le., Morel, P., Robe, M.C., Maurenads, J.M., Faivre-Pierret, R., Martin, D., Sabroux, J.C., Zettwoog, P., 1991. Eruptive and diffuse emissions of CO2 from Mount Etna. Nature 351, 387–391. Allard, P., Carbonnelle, J., Métrich, N., Loyer, H., Zettwoog, P., 1994. Sulphur output and magma degassing budget of Stromboli volcano. Nature 368, 326–330. Allard, P., Burton, M., Muré, F., 2005. Spectroscopic evidence for a lava fountain driven by previously accumulated magmatic gas. Nature 433, 407–410. Allard, P., Behncke, B., D'Amico, S., Neri, M., Gambino, S., 2006. Mount Etna 1993–2005: anatomy of an evolving eruptive cycle. Earth Sci. Rev. 78, 85–114. Anderson Jr., A.T., Newman, S., Williams, S.N., Druitt, T.H., Skirius, C., Stolper, E., 1989. H2O, CO2, Cl and gas in Plinian and ash-flow Bishop rhyolite. Geology 17, 221–225. Andres, R.J., Kasgnoc, A.D., 1998. A time-averaged inventory of subaerial volcanic sulfur emissions. J. Geophys. Res. 103, 25251–25261. Aramaki, S., Ui, T., 1978. Major element frequency distribution of the Japanese Quaternary volcanic rocks. Bull. Volcanol. 41, 390–407. Arellano, S.R., Hall, M., Samaniego, P., Le Pennec, J.L., Ruiz, A., Molina, I., Yepes, H., 2008. Degassing patterns of Tungurahua volcano (Ecuador) during the 1999–2006 eruptive period, inferred from remote spectroscopic measurements of SO2 emissions. J. Volcanol. Geotherm. Res. 176, 151–162. Bani, P., Oppenheimer, C., Allard, P., Shinohara, H., Tsanev, V., Carn, S., Lardy, M., Garaebiti, E., 2012. First estimate of volcanic SO2 budget for Vanuatu island arc. J. Volcanol. Geotherm. Res. 211–212, 36–46. Baubron, J.C., Allard, P., Toutain, J.P., 1990. Diffuse volcanic emissions of carbon dioxide from Vulcano Island. Nature 344, 51–53. Bebout, G.E., 1996. Volatile transfer and recycling at convergent margins: mass-balance and insights from high-P/T metamorphic rocks. Geophys. Monogr. 96, 179–193. Bertagnini, A., Métrich, N., Landi, P., Rosi, M., 2003. Stromboli volcano (Aeolian Archipelago, Italy): an open window on the deep-feeding system of a steady state basaltic volcano. J. Geophys. Res. 108. http://dx.doi.org/10.1029/2002JB002146. Bianchi, D., Sarmiento, J.L., Gnanadesikan, A., Key, R.M., Schlosser, P., Newton, R., 2010. Low helium flux from the mantle inferred from simulations of oceanic helium isotope data, Earth Planet. Sci. Lett. 297, 379–386. Bluth, G.J.S., Schnetzler, C.C., Krueger, A.J., Walter, L.S., 1993. The contribution of explosive volcanism to global atmospheric sulphur dioxide concentrations. Nature 336, 327–329. Brantley, S.L., Koepenick, K.W., 1995. Measured carbon dioxide emissions from Oldoinyo Lengai and the skewed distribution of passive volcanic fluxes. Geology 23, 933–936. Brombach, T., Hunzikerm, J., Chiodini, G., Cardellini, C., Marini, L., 2001. Soil diffuse degassing and thermal energy fluxes from the southern Lakki plain, Nisyros (Greece). Geophys. Res. Lett. 28, 69–72. Burton, M.R., Mader, H.M., Polacci, M., 2007. The role of gas percolation in quiescent degassing of persistently active basaltic volcanoes. Earth Planet. Sci. Lett. 264, 46–601. Burton, M., Sawyer, G.M., Granieri, D., 2013. Deep carbon emissions from volcanoes. Rev. Mineral. Geochem. 75, 323–354. Carapezza, M.L., Federico, C., 2000. The contribution of fluid geochemistry to the volcano monitoring of Stromboli. J. Volcanol. Geotherm. Res. 95, 227–245. Cardellini, C., Chiodini, G., Frondini, F., 2003. Application of stochastic simulation to CO2 flux from soil: mapping and quantification of gas release. J. Geophys. Res. 108. http://dx.doi.org/10.1029/2002JB002165. Carn, S.A., Krueger, A.J., Bluth, G.J.S., Schaefer, S.J., Krotkov, N.A., Watson, I.M., Darra, S., 2003. Volcanic eruption detection by the Total Ozone Mapping Spectrometer

(TOMS) instruments: a 22-year record of sulphur dioxide and ash emissions. Volcanic Degassing. Geol. Soc. Spec. Publ., 212, pp. 177–202. Carn, S.A., Krueger, A.J., Krotkov, N.A., Arellano, S., Yang, K., 2008. Daily monitoring of Ecuadorian volcanic degassing from space. J. Volcanol. Geotherm. Res. 176 (1), 141–150. http://dx.doi.org/10.1016/j.jvolgeores.2008.01.029. Cervantes, P., Wallace, P.J., 2003a. Role of H2O in subduction-zone magmatism: new insights from melt inclusions in high-Mg basalts from central Mexico. Geology 31, 235–238. Cervantes, P., Wallace, P.J., 2003b. Magma degassing and basaltic eruption styles: a case study of the 2000 yr B.P. eruption of Xitle volcano, central Mexico. J. Volcanol. Geotherm. Res. 120, 249–270. Chiodini, G., Cioni, R., Guidi, M., Raco, B., Marini, L., 1998. Soil CO2 flux measurements in volcanic and geothermal areas. Appl. Geochem. 13, 543–552. Chiodini, G., Avino, R., Brombach, T., Caliro, S., Cardellini, C., De Vita, S., Frondini, F., Marotta, E., Ventura, G., 2004. Fumarolic and diffuse soil degassing west of Mount Epomeo, Ischia (Italy). J. Volcanol. Geotherm. Res. 133, 291–309. Chiodini, G., Baldini, A., Barberi, F., Carapezza, M.L., Cardellini, C., Frondini, F., Granieri, D., Ranaldi, M., 2007. Carbon dioxide degassing at Latera caldera (Italy): evidence of geothermal reservoir and evaluation of its potential energy. J. Geophys. Res. 112, B12204. http://dx.doi.org/10.1029/2006JB004896. Christopher, T., Edmonds, M., Humphreys, M.C.S., Herd, R.A., 2010. Volcanic gas emissions from Soufrière Hills Volcano, Montserrat 1995–2009, with implications for mafic magma supply and degassing. Geophys. Res. Lett. 37, L00E04. http://dx.doi.org/ 10.1029/2009GL041325. Clague, D.A., Moore, J.G., Dixon, J.E., Friesen, W.B., 1995. Petrology of submarine lavas from Kilauea's Puna Ridge, Hawaii. J. Petrol. 36, 299–349. Craig, H., 1953. The geochemistry of the stable carbon isotopes. Geochim. Cosmochim. Acta 3, 53–92. Craig, H., Clarke, W.B., Beg, M.A., 1975. Excess 3He in deep water on the East Pacific Rise. Earth Planet. Sci. Lett. 26, 125–132. Crisp, J.A., 1984. Rates of magma emplacement and volcanic output. J. Volcanol. Geotherm. Res. 20, 177–211. de Hoog, J.C.M., Koetsier, G.W., Bronto, S., Sriwana, T., van Bergen, M.J., 2001. Sulfur and chlorine degassing from primitive arc magmas: temporal changes during the 1982– 1983 eruptions of Galunggung (west, Java, Indonesia). J. Volcanol. Geotherm. Res. 108, 55–83. de Hoog, J.C.M., Hattori, K., Noblitt, R.P., 2004. Oxidized sulfur-rich mafic magma at Mount Pinatubo, Philippines. Bull. Volcanol. 146 (146), 750–761. Delgado-Granados, H., Cárdenas González, L., Piedad Sánchez, N., 2001. Sulfur dioxide emissions from Popocatépetl volcano (Mexico): case study of a high-emission rate, passively degassing erupting volcano. J. Volcanol. Geotherm. Res. 108, 107–120. Devine, J.D., Sigurdsson, H., Davis, A.N., Self, S., 1984. Estimates of sulfur and chlorine yield to the atmosphere from volcanic eruptions and potential climatic effects. J. Geophys. Res. 89, 6309–6325. Edmonds, M., Aiuppa, A., Humphreys, M., Moretti, R., Giudice, G., Martin, R.S., Herd, R.A., Christopher, T., 2010. Excess volatiles supplied by mingling of mafic magma at an andesite arc volcano. Geochem. Geophys. Geosyst. 11. http://dx.doi.org/10.1029/ 2009GC002781. Ehara, S., 1985. Geothermal activity of Iwo-jima. J. Geogr. 94, 516–527 (in Japanese with English abstract). Elkins, L.J., Fischer, T.P., Hilton, D.R., Sharp, Z.D., McKnight, S., Walker, J., 2006. Tracing nitrogen in volcanic and geothermal volatiles from the Nicaraguan volcanic front. Geochim. Cosmochim. Acta 70, 5215–5235. Fischer, T.P., 2008. Fluxes of volatiles (H2O, CO2, N2, Cl, F) from arc volcanoes. Geochem. J. 42, 21–38. Fischer, T.P., Marty, B., 2005. Volatile abundances in the sub-arc mantle: insights from volcanic and hydrothermal gas emissions. J. Volcanol. Geotherm. Res. 140, 205–216. Frondini, F., Chiodini, G., Caliro, S., Cardellini, C., Granieri, D., Ventura, G., 2004. Diffuse CO2 degassing at Vesuvio, Italy. Bull. Volcanol. 66, 642–651. Gerardo, J.Y., Nuti, S., D'Amore, F., Seastres, J., Gonfiantini, R., 1993. Isotopic evidence for magmatic and meteoric water recharge and the processes affecting reservoir fluids in the Palinpinon geothermal system, Philippines. Geothermal 22, 521–533. Gerlach, T.M., Graeber, E.J., 1985. Volatile budget of Kilauea volcano. Nature 313, 273–277. Gerlach, T.M., McGee, K.A., 1994. Total sulfur dioxide emissions and pre-eruption vapor-saturated magma at mount St. Helens, 1980–88. Geophys. Res. Lett. 21, 2833–2836. Gerlach, T.M., Westrich, H.R., Casadevall, T.J., Finnegan, D.L., 1994. Vapor saturation and accumulation in magmas of the 1989–1990 eruption of Redoubt volcano, Alaska. J. Volcanol. Geotherm. Res. 62, 317–337. Gerlach, T.M., Westrich, H.R., Symonds, R.B., 1996. Preeruption vapor in magma of the climactic Mount Pinatubo eruption: source of the giant stratospheric sulfur dioxide cloud. In: Newhall, C.G., Punongbayan, R.S. (Eds.), Fire and Mud: Eruptions and Lahars of Mount Pinatubo, Philippines, pp. 415–433. Gerlach, T.M., Doukas, M.P., McGee, K.A., Kessler, R., 1998. Three-year decline of magmatic CO2 emissions from soils of a Mammoth Mountain tree kill: Horseshoe Lake, CA, 1995–1997. Geophys. Res. Lett. 25, 1947–1950. Giggenbach, W.F., 1991. Isotopic composition of geothermal water and steam discharges. D'Amore, F. (coordinator), Application of Geochemistry in Geothermal Reservoir Development. UNITAR, New York, pp. 253–273. Giggenbach, W.F., 1992. Isotopic shifts in waters from geothermal and volcanic systems along convergent plate boundaries and their origin. Earth Planet. Sci. Lett. 113, 495–510. Giggenbach, W.F., 1995. Variations in the chemical and isotopic composition of fluids discharged from the Taupo Volcanic Zone, New Zealand. J. Volcanol. Geoth. Res. 68, 89–116.

H. Shinohara / Journal of Volcanology and Geothermal Research 268 (2013) 46–63 Giggenbach, W.F., 1996. Chemical composition of volcanic gases. In: Scarpa, R., Tilling, R.I. (Eds.), Monitoring and Mitigation of Volcanic Hazards. Springer-Verlag, Berlin, pp. 221–256. Giggenbach, W.F., Matsuo, S., 1991. Evaluation of results from second and third IAVCEI field workshops on volcanic gases, Mt Usu, Japan, and White Island, New Zealand. Appl. Geochem. 6, 125–141. Giggenbach, W.F., Tedesco, D., Sulistiyo, Y., Caprai, A., Cioni, R., Favara, R., Fischer, T.P., Hirabayashi, J., Korzhinsky, M., Martini, M., Menyailov, I., Shinohara, H., 2001. Evaluation of results from the fourth and fifth IAVCEI field workshops on volcanic gases, Vulcano island, Italy and Java, Indonesia. J. Volcanol. Geotherm. Res. 108, 157–172. Gurenko, A.A., Belousov, A.B., Trumbull, R.B., Sobolev, A.V., 2005. Explosive basaltic volcanism of the Chikurachki volcano (Kurile arc, Russia): insights on pre-eruptive magmatic conditions and volatile budget revealed from phenocryst-hosted melt inclusions and groundmass glasses. J. Volcanol. Geotherm. Res. 147, 203–232. Harris, D.M., Anderson Jr., A.T., 1984. Volatiles H2O, CO2, and Cl in a subduction related basalt. Contrib. Mineral. Petrol. 57, 120–128. Hernández, P.A., Natale, G., Tsunomori, F., Sugiyama, K., Ito, T., Notsu, K., Okada, H., Pérez, N.M., 2001a. Preliminary results of diffuse emissions of CO2 and soil gas pressure gradient measurements at Tarumae volcano, Japan. Bull. Volcanol. Soc. Jpn. 46, 121–125. Hernández, P.A., Notsu, K., Salazar, J.M., Mori, T., Natale, G., Okada, H., Virgili, G., Shimoike, Y., Sato, M., Pérez, N.M., 2001b. Carbon dioxide degassing by advective flow from Usu volcano, Japan. Science 292, 83–86. Hernández, P.A., Salazar, J.M., Shimoike, Y., Mori, T., Notsu, K., Pérez, N.M., 2001c. Diffuse emission of CO2 from Miyakejima volcano, Japan. Chem. Geol. 177, 175–185. Hernández, P.A., Notsu, K., Tsurumi, M., Mori, T., Ohno, M., Shimoike, Y., Salazar, J., Pérez, N.M., 2003. Carbon dioxide emissions from soils at Hakkoda, north Japan. J. Geophys. Res. 108, 2210. http://dx.doi.org/10.1029/2002JB001847. Hernández, P.A., Notsu, K., Okada, H., Mori, T., Sato, M., Barahona, F., Pérez, N.M., 2006. Diffuse emission of CO2 from Showa-Shinzan, Hokkaido, Japan: a sign of volcanic dome degassing. Pure Appl. Geophys. 163, 869–881. Hilton, D.R., Fischer, T.P., Marty, B., 2002. Noble gases and volatile recycling at subduction zones. Rev. Mineral. Geochem. 47, 319–370. Hirabayashi, J., Mizuhashi, M., 2004. The discharge rate of volatiles from Kusatsu-Shirane volcano, Japan. Report of 4th Joint Observation of Kusatsu-Shirane Volcano, pp. 167–174. Hirabayashi, J., Yoshida, M., Ossaka, J., 1990. Chemistry of volcanic gases from the 62-1 crater of Mt Tokachi, Hokkaido, Japan. Bull. Volcanol. Soc. Jpn. Ser. 2 (35), 205–215. Hirabayashi, J., Saito, M., Nogami, K., Mori, T., Mizuhashi, M., Oikawa, M., 2004. Discharge rate of CO2 from the summit area of Kusatsu-Shirane volcano. Report of 4th Joint Observation of Kusatsu-Shirane Volcano, pp. 175–181 (in Japanese). Hochstein, M.P., 1995. Crustal heat transfer in the Taupo Volcanic Zone (New Zealand): comparison with other volcanic arcs and explanatory heat source models. J. Volcanol. Geotherm. Res. 68, 117–151. Hoefs, J., 1973. Ein Beitrag zur Isotopengeochemie des Kohlenstoffs in magmatischen Gesteinen. Contrib. Mineral. Petrol. 41, 277–300. Inguaggiato, S., Mazot, A., Diliberto, I.S., Inguaggiato, C., Madonia, P., Rouwet, D., Vita, F., 2012. Total CO2 output from Vulcano island (Aeolian Islands, Italy). Geochem. Geophys. Geosyst. 13. http://dx.doi.org/10.1029/2011GC003920. Ito, E., Harris, D.M., Anderson, A.T., 1983. Alteration of oceanic crust and geologic cycling of chlorine and water. Geochim. Cosmochim. Acta 47, 1613–1624. James, E.R., Manga, M., Rose, T.P., 1999. CO2 degassing in the Oregon Cascades. Geology 27, 823–826. Kanroji, Y., Takahashi, T., Horiuchi, K., 1999. Studies on the regional distribution of thermal water qualities in Shiobara Spas, Tochigi prefecture. Onsenkougakujaishi 26, 75–94 (in Japanese with English abstract). Kazahaya, K., Shinohara, H., Uto, K., Odai, M., Nakahori, Y., Mori, H., Iino, H., Miyashita, M., Hirabayashi, J., 2004. Gigantic SO2 emission from Miyakejima volcano, Japan, caused by caldera collapse. Geology 32, 425–428. Kelemen, P.B., Hanghøj, K., Greene, A.R., 2003. One view of the geochemistry of subduction-related magmatic arcs with an emphasis on primitive andesite and lower crust. In: Rudnick, R.L. (Ed.), The Crust: Treatise on Geochemistry. ElsevierPergamon, Oxford, pp. 593–659. Keppler, H., 2010. The distribution of sulfur between haplogranitic melts and aqueous fluids. Geochim. Cosmochim. Acta 74, 645–660. Kimbara, K., 2005. Distribution map and catalogue of hot and mineral springs in Japan (2nd Edition, CD-ROM version). Geological Survey of Japan, Digital Geoscience Map Series, GT-2. Kimbara, K., Sakaguchi, K., 1989. Geology, distribution of hot springs and hydrothermal alteration zones of major geothermal areas in Japan. Geol. Surv. Jpn. Rep. 270, 1–482 (in Japanese). Krotkov, N.A., Carn, S.A., Krueger, A.J., Bhartia, P.K., Yang, K., 2006. Band residual difference algorithm for retrieval of SO2 from the aura ozone monitoring instrument (OMI). Geosci. Remote Sens. IEEE Trans. 44, 1259–1266. Kumagai, S., Takahashi, H., 1985. Crustal deformations on Iwo-jima: II. J. Geogr. 94, 479–487 (in Japanese with English abstract). Lesne, P., Kohn, S.C., Blundy, J., Witham, F., Botcharnikov, R.E., Behrens, H., 2011. Experimental simulation of closed-system degassing in the system basalt–H2O– CO2–S–Cl. J. Petrol. 52, 1737–1762. Lewicki, J.L., Connor, C., St-Amand, K., Stix, J., Spinner, W., 2003. Self-potential, soil CO2 flux, and temperature on Masaya volcano, Nicaragua. Geophys. Res. Lett. 30 (15), 1817. http://dx.doi.org/10.1029/2003GL017731. Luhr, J.F., 2001. Glass inclusions and melt volatile contents at Paricutin volcano, Mexico. Contrib. Mineral. Petrol. 142, 261–283.

61

Luhr, J.F., Carmichael, I.S.E., Varekamp, J.C., 1984. The 1982 eruptions of El Chichón volcano, Chipas, Mexico: mineralogy and petrology of the anhydrite-bearing pumices. J. Volcanol. Geotherm. Res. 23, 69–108. Mariner, R.H., Presser, T.S., Evans, W.C., Pringle, M.K.W., 1990. Discharge rates of fluid and heat by thermal springs of the Cascade Range, Washington, Oregon, and northern California. J. Geophys. Res. 95, 19517–19519. Martin, R.S., Sawyer, G.M., Spampinato, L., Salerno, G.G., Ramirez, C., Ilyinskaya, E., Witt, M.L.I., Mather, T.A., Watson, I.M., Phililips, J.C., Oppenheimer, C., 2010. A total volatile inventory for Masaya Volcano, Nicaragua. J. Geophys. Res. 115. http://dx.doi.org/ 10.1029/2010JB007480. Martini, M., 1983. Galeras. Bull. Glob. Volcanol. Netw. 18, 7–9. Marty, B., Tolstikhin, I.N., 1998. CO2 fluxes from mid-ocean ridges, arcs and plumes. Chem. Geol. 145, 233–248. Mather, T.A., Tsanev, V.I., Pyle, D.M., McGonigle, A.J.S., Oppenheimer, C., Allen, A.G., 2004. Characterization and evolution of tropospheric plumes from Lascar and Villarrica volcanoes, Chile. J. Geophys. Res. 109, D21303. http://dx.doi.org/10.1029/ 2004JD004934. Mather, T.A., Pyle, D.M., Tsanev, V.I., McGonigle, A.J.S., Oppenheimer, C., Allen, A.G., 2006. A reassessment of current volcanic emissions from the Central American arc with specific examples from Nicaragua. J. Volcanol. Geotherm. Res. 149, 297–311. Matthews, S.J., Sparks, R.S.J., Gardeweg, M.C., 1999. The Piedras Grandes–Soncor eruptions. Lascar volcano, Chile; evolution of a zoned magma chamber in the central Andean upper crust. J. Petrol. 40, 1891–1919. McCormick, B.T., Edmonds, M., Mather, T.A., Carn, S.A., 2012. First synoptic analysis of volcanic degassing in Papua New Guinea. Geochem. Geophys. Geosyst. G3, 13. http://dx.doi.org/10.1029/2011GC003945. McGonigle, A.J.S., Oppenheimer, C., Tsanev, V.I., Saunders, S., Mulina, K., Tohui, S., Bosco, J., Nahou, J., Kuduon, J., Taranu, F., 2004. Sulphur dioxide fluxes from Papua New Guinea's volcanoes. Geophys. Res. Lett. 31. http://dx.doi.org/10.1029/2004GL019568 L08606. Menyailov, I.A., Nikitina, L.P., Shapar, V.N., Rozhkov, A.M., Miklishanskii, A.Z., 1986a. Chemical composition and content of metals in gas emanations from the Alaid Volcano crater during 1981 eruption. Vulkanol. Seismol. 1, 26–31. Menyailov, I.A., Nikitina, L.P., Shapar, V.N., Pilipenko, V.P., 1986b. Temperature increase and chemical change of fumarolic gases at Momotombo Volcano, Nicaragua, in 1982–1985: are these indicators of a possible eruption? J. Geophys. Res. 91, 12199–12214. Métrich, N., Allard, P., Aiuppa, A., Bani, P., Bertagnini, A., Shinohara, H., Parello, F., Di Muro, A., Garaebiti, E., Belhadj, O., Massare, D., 2011. Magma and volatile supply to postcollapse volcanism and block resurgence in Siwi caldera (Tanna Island, Vanuatu arc). J. Petrol. 52, 1077–1105. Michaud, V., Clocchiatti, R., Sbrana, S., 2000. The Minoan and post-Minoan eruptions, Santorini (Greece), in the light of melt inclusions: chlorine and sulphur behaviors. J. Volcanol. Geotherm. Res. 99, 195–214. Mori, T., Shinohara, H., Kazahaya, K., Hirabayashi, J., Matsushima, T., Mori, T., Ohwada, M., Odai, M., Iino, H., Miyashita, M., 2013. Time-averaged SO2 fluxes of subduction-zone volcanoes: example of a 32 years exhaustive survey for Japanese volcanoes. J. Geophys. Res. 118. http://dx.doi.org/10.1002/jgrd.50591. Mörner, N.-A., Ethiope, G., 2002. Carbon degassing from the lithosphere. Global Planet. Chang. 33, 185–203. Morrero, R., López, D., Hernández, P.A., Pérez, N.M., 2008. Carbon dioxide discharged through the las Cañadas Auifer, Tenerife, Canary Islands. Pure Appl. Geophys. 165, 147–172. Mutou, N., Matsubaya, O., 2002. Origin and geochemical variation of geothermal water in the Tamagawa Hot Spring area, Akita, Japan. Chikyukagaku (Geochemistry) 36, 81–88 (in Japanese with English abstract). Nemoto, M., Hayakawa, M., Takahashi, K., Oana, S., 1957. Report on the geological, geophysical, and geochemical studies of Usu volcano (Showashinzan). Geol. Surv. Jpn. Rep. 170, 1–149 (in Japanese). Nesbitt, H.W., Young, G.M., 1984. Prediction of some weathering trends of plutonic and volcanic rocks based on thermodynamic and kinetic considerations. Geochim. Cosmochim. Acta 48, 1523–1534. Newhall, C.G., Self, S., 1982. The volcanic explosivity index (VEI): an estimate of explosive magnitude for historical volcanism. J. Geotherm. Res. 87, 1231–1238. Newman, S., Lowenstern, J.B., 2002. VOLATILECALC: a silicate melt–H2O–CO2 solution model written in Visual Basic Excel. Comput. Geosci. 2, 597–604. Nishimura, M., 1966. Geochemical study of hot springs in Noboribetsu. Onsenkagaku 16, 60–64 (in Japanese with English abstract). Notsu, K., Sugiyama, K., Hosoe, M., Uemura, A., Shimoike, Y., Tsunomori, F., Sumino, H., Yamamoto, J., Mori, T., Hernández, P.A., 2005. Diffuse CO2 efflux from Iwojima volcano, Izu–Ogasawara arc, Japan. J. Volcanol. Geotherm. Res. 139, 147–161. Ohba, T., Hirabayashi, J.I., Yoshida, M., 1994. Equilibrium temperature and redox state of volcanic gas at Unzen volcano, Japan. J. Volcanol. Geotherm. Res. 60, 263–272. Ohba, T., Hirabayashi, J.I., Nogami, K., Kusakabe, M., Yoshida, M., 2008. Magma degassing process during the eruption of Mt. Unzen, Japan in 1991 to 1995: modeling with the chemical composition of volcanic gas. J. Volcanol. Geotherm. Res. 175, 120–132. Ohta, K., Matsuwo, N., Shimizu, H., Fukui, R., Kamada, M., Kagiyama, T., 1988. Emission rates of sulfur-dioxide from some volcanoes in Japan. Proc. Kagoshima International Conference on Volcanoes 1988, pp. 420–423. Ohwada, M., Kazahaya, K., Itoh, J., Morikawa, N., Takahashi, M., Takahashi, H.A., Inamura, A., Yasuhara, M., Tsukamoto, H., 2012. Passive degassing of magmatic volatiles from Iwate volcano, NE Japan, based on three-dimensional measurement of helium isotopes in groundwater. J. Geophys. Res. 117, B02204. http://dx.doi.org/10.1029/ 2011JB008532. Oppenheimer, C., Kyle, P.R., 2008. Probing the magma plumbing of Erebus volcano, Antarctica, by open-path FTIR spectroscopy of gas emissions. J. Volcanol. Geotherm. Res. 177, 743–754.

62

H. Shinohara / Journal of Volcanology and Geothermal Research 268 (2013) 46–63

Oppenheimer, C., Bani, P., Calkins, J.A., Burton, M.R., Sawyer, G.M., 2006. Rapid FTIR sensing of volcanic gases released by Strombolian explosions at Yasur volcano, Vanuatu. Appl. Phys. B 85, 453–460. Palais, J.M., Sigurdsson, H., 1989. Petrologic evidence of volatile emissions from major historic and pre-historic volcanic eruptions. Understanding Climate Change. Geophys. Monogr. Ser., 52, pp. 31–53. Pérez, N.M., Hernández, P., Padilla, G., Nolasco, D., Barrancos, J., Melían, G., Padrón, E., Dionis, S., Calvo, D., Rodrígues, F., Notsu, K., Mori, T., Kusakabe, M., Arpa, M.C., Reniva, P., Ibarra, M., 2011. Global CO2 emission from volcanic lakes. Geology 39, 235–238. Pyle, D.M., Beattie, P.D., Bluth, G.J.S., 1996. Sulphur emissions to the stratosphere from explosive volcanic eruptions. Bull. Volcanol. 57, 663–671. Roggensack, K., 2001. Unraveling the 1974 eruption of Fuego volcano (Guatemala) with small crystals and their young melt inclusions. Geology 29, 911–914. Roggensack, K., Williams, S.N., Schaefer, S.J., Parnell Jr., R.A., 1996. Volatiles from the 1994 eruptions of Rabaul: understanding large caldera systems. Science 273, 490–493. Roggensack, K., Hervig, R.L., McKnight, S.B., Williams, S.N., 1997. Explosive basaltic volcanism from Cerro Negro volcano: influence of volatiles on eruptive style. Science 277, 1639–1642. Rowe, G.L., Brantley, S.L., Fernandez, M., Fernandez, J.F., Borgia, A., Barquero, J., 1992. Fluid-volcano interaction in an active stratovolcano: the crater lake system of Poas volcano, Costa Rica. J. Volcanol. Geotherm. Res. 49, 23–51. Rudnick, R.L., Gao, S., 2003. Composition of the continental crust. In: Rudnick, R.L. (Ed.), The Crust: Treatise on Geochemistry. Elsevier-Pergamon, Oxford, pp. 1–64. Ruscitto, D.M., Wallace, P.J., Cooper, L.B., Plank, T., 2012. Global variations in H2O/Ce: 2. Relationships to arc magma geochemistry and volatile fluxes. Geochem. Geophys. Geosyst. 13, Q03025. http://dx.doi.org/10.1029/2011GC003887. Rutherford, M.J., Devine, J.D., 1996. Preeruption pressure–temperature conditions and volatiles in the 1991 dacitic magma of Mount Pinatubo. In: Newhall, C.G., Punongbayan, R.S. (Eds.), Fire and Mud: Eruptions and Lahars of Mount Pinatubo, Philippines, pp. 751–766. Sadofsky, S.J., Portnyagin, M., Hoernle, K., van den Bogaard, P., 2008. Subduction cycling of volatiles and trace elements through the Central American volcanic arc: evidence from melt inclusions. Contrib. Mineral. Petrol. 155, 433–456. Saito, G., Kazahaya, K., Shinohara, H., Stimac, J., Kawanabe, Y., 2001. Variation of volatile concentration in a magma system of Satsuma-Iwojima volcano deduced from melt inclusion analyses. J. Volcanol. Geotherm. Res. 108, 11–31. Saito, G., Uto, K., Kazahaya, K., Shinohara, H., Kawanabe, Y., Satoh, H., 2005. Petrological characteristics and volatile content of magma from the 2000 eruption of Miyakejima volcano, Japan. Bull. Volcanol. 67, 268–280. Saito, M., Matsushima, T., Matsuwo, N., Shimizu, H., 2007. Observation of SO2 and CO2 fluxes in and around the active crater of Aso Nakadake Volcano. Sci. Rep. Dpt. Earth Planet. Sci. Kyushu Univ. 22, 51–52 (in Japanese with English abstract). Salazar, J.M., Hernández, P.A., Pérez, N.M., Melian, G., Alvarez, J., Segura, F., Notsu, K., 2001. Diffuse emission of carbon dioxide from Cerro Negro volcano, Nicaragua, Central America. Geophys. Res. Lett. 28, 4275–4278. Salazar, J.M.L., Hernández, P.A., Pérez, N.M., Olmos, R., Barahona, F., Cartagena, R., Soriano, T., López, D.L., Sumino, H., Notsu, K., 2004. Spatial and temporal variation of diffuse CO2 degassing at the Santa Ana–Izalco–Coatepeque volcanic complex, El Salvador, Central America. Geol. Soc. Am. Spec. Pap. 375, 135–146. Sano, Y., Williams, S.N., 1996. Fluxes of mantle and subducted carbon along convergent plate boundaries. Geophys. Res. Lett. 23, 2749–2752. Sano, Y., Takahata, N., Nishio, Y., Fischer, T.P., Williams, S.N., 2001. Volcanic flux of nitrogen from the Earth. Chem. Geol. 171, 263–271. Satoh, H., Saito, G., Shinohara, H., Yamaguchi, Y., 2003. Sulfur source for the 1991–1995 Unzen eruption: evidence from melt inclusions in pyroxenes. Geophys. Res. Lett. 30 (23). http://dx.doi.org/10.1029/2003GL017629. Sawyer, G.M., Carn, S.A., Tsanev, V.I., Oppenheimer, C., Burton, M., 2008a. Investigation into magma degassing at Nyiragongo volcano, Democratic Republic of the Congo. Geochem. Geophys. Geosyst. 9, Q02017. http://dx.doi.org/10.1029/2007GC001829. Sawyer, G.M., Oppenheimer, C., Tsanev, V.I., Yirgu, G., 2008b. Magmatic degassing at Erta 'Ale volcano, Ethiopia. J. Volcanol. Geotherm. Res. 178, 837–846. Scaillet, B., Luhr, J.F., Carrol, M.R., 2003. Petrological and volcanological constraints on volcanic sulfur emissions to the atmosphere. Volcanism and the Earth's Atmosphere. Geophys. Monogr. Ser., 139, pp. 11–40. Schneider, A., 1970. Sulfur, abundance in common igneous rocks. Handbook of Geochemistry.Springer-Verlag, Berlin (Section 16). Schnetzler, C.C., Bluth, G.J.S., Krueger, A.J., Walter, L.S., 1997. A proposed volcanic sulfur dioxide index (VSI). J. Geophys. Res. 102, 20087–20091. Scholl, D.W., von Huene, R., 2009. Implications of estimated magmatic additions and recycling losses at the subduction zones of accretionary (non-collisional) and collisional (suturing) orogens. In: Cawood, P., Kröner, A. (Eds.), Accretionary Orogens in Space and Time. Geol. Soc. London Sp. Pub., 318, pp. 105–125. Self, S., King, A.J., 1996. Petrology and sulfur and chlorine emissions of the 1963 eruption of Gunung Agung, Bali, Indonesia. Bull. Volcanol. 58, 263–285. Seward, T.M., Kerrick, D.M., 1996. Hydrothermal CO2 emission from the Taupo Volcanic Zone, New Zealand. Earth Planet. Sci. Lett. 139, 105–113. Shimoike, Y., 1999. Behavior of volcanic fluids at Izu-Oshima volcano. Gougai Gekkan Kaiyo 19, 88–94 (in Japanese). Shimoike, Y., Kazahaya, K., Shinohara, H., 2002. Soil gas emission of volcanic CO2 at Satsuma-Iwojima volcano, Japan. Earth Planets Space 54, 239–247. Shinohara, H., 2005. A new technique to estimate volcanic gas composition: plume measurements with a portable multi-sensor system. J. Volcanol. Geotherm. Res. 143, 319–333. Shinohara, H., 2008. Excess degassing from volcanoes and its role on eruptive and intrusive activity. Rev. Geophys. 46. http://dx.doi.org/10.1029/2007RG000244.

Shinohara, H., Witter, J., 2005. Volcanic gases emitted during mild Strombolian activity of Villarrica volcano, Chile. Geophys. Res. Lett. 32, L20308. http://dx.doi.org/10.1029/ 2005GL024131. Shinohara, H., Kazahaya, K., Saito, G., Matsushima, N., Kawanabe, Y., 2002. SatsumaIwojima volcano, Japan: formation of a new degassing vent, 1990–1999. Earth Planets Space 54, 175–185. Shinohara, H., Fukui, K., Kazahaya, K., Saito, G., 2003. Degassing process of Miyakejima volcano: implications of gas emission rate and melt inclusion data. Melt Inclusions in Volcanic Systems. Advances in Volcanology, 4, pp. 147–161. Shinohara, H., Yoshikawa, S., Miyabuchi, Y., 2014. Degassing activity of a volcanic crater lake: volcanic plume measurements at the Yudamari crater lake, Aso volcano, Japan. In: Rouwet, D., Christenson, B., Tassi, F., Vandemeulebrouck, J. (Eds.), Volcanic Lakes. Springer-Heidelberg (in press). Sigurdsson, H., Carey, S., Palais, J.M., Devin, J., 1990. Pre-eruption compositional gradients and mixing of andesite and dacite erupted from Nevado del Ruiz volcano, Colombia in 1985. J. Volcanol. Geotherm. Res. 42, 127–151. Simkin, T., 1993. Terrestrial volcanism in space and time. Annu. Rev. Earth Planet. Sci. 21, 427–452. Sisson, T.W., Bronto, S., 1998. Evidence for pressure-release melting beneath magmatic arcs from basalt at Galunggung, Indonesia. Nature 391, 883–886. Sisson, T.W., Layne, G.D., 1993. H2O in basalt and basaltic andesite glass inclusions from four subduction-related volcanoes. Earth Planet. Sci. Lett. 117, 619–635. Stix, J., Layne, G.D., Williams, S.N., 2003. Mechanisms of degassing at Nevado del Ruiz volcano, Colombia. J. Geol. Soc. London 160, 507–521. Stoiber, R.E., Malinconico, L.L., Williams, S.N., 1983. Use of the correlation spectrometer at volcanoes. Forecasting Volcanic Events. Dev. Volcanol., 1, pp. 425–444. Stoiber, R.E., Williams, S.N., Huebert, B., 1987. Annual contribution of sulfur dioxide to the atmosphere by volcanoes. J. Volcanol. Geotherm. Res. 33, 1–8. Sumi, K., 1977. The relationship of the rates of convective heat discharge to the tectonic provinces in Japan. Bull. Geol. Surv. Jpn. 28, 277–325 (in Japanese with English abstract). Symonds, R.B., Rose, W.I., Gerlach, T.M., Briggs, P.H., Harmon, R.S., 1990. Evaluation of gases, condensates, and SO2 emissions from Augustine volcano, Alaska: the degassing of a Cl-rich volcanic system. Bull. Volcanol. 52, 355–374. Symonds, R.B., Rose, W.I., Bluth, G.J.S., Gerlach, T.M., 1994. Volcanic-gas studies: methods, results, and applications. Volatiles in Magmas. Rev. Mineral., 30, pp. 1–66. Symonds, R.B., Mizutani, Y., Briggs, P.H., 1996. Long-term geochemical surveillance of fumaroles at Showa-Shinzan dome, Usu volcano, Japan. J. Volcanol. Geotherm. Res. 73, 177–211. Tamura, Y., Tatsumi, Y., 2002. Remelting of an andesitic crust as a possible origin for rhyolitic magma in oceanic arcs: an example from the Izu–Bonin arc. J. Petrol. 43, 1029–1047. Taran, Y.A., 2009. Geochemistry of volcanic and hydrothermal fluids and volatile budget of the Kamchatka–Kuril subduction zone. Geochim. Cosmochim. Acta 73, 1067–1094. Taran, Y.A., Pokrovsky, B.G., Dubik, Y.M., 1989. Isotopic composition and origin of water from andesitic magmas. Dokl. Acad. Sci. USSR 304, 440–443. Taran, Y.A., Rozhkov, A.M., Serafimova, E.K., Esikov, A.D., 1991. Chemical and isotopic composition of magmatic gases from the 1988 eruption of Klyuchevskoy volcano, Kamchatka. J. Volcanol. Geotherm. Res. 46, 255–263. Taran, Y.A., Pilipenko, V.P., Rozhkov, A.M., Vakin, E.A., 1992. A geochemical model for fumaroles of the Mutnovsky volcano, Kamchatka,USSR. J. Volcanol. Geotherm. Res. 49, 269–283. Taran, T.A., Hedenquist, J.W., Korzhinsky, M.A., Tkachenko, S.I., Shmulovich, K.I., 1995. Geochemistry of magmatic gases from Kudryavy volcano, Itrup, Kuril Islands. Geochim. Cosmochim. Acta 59, 1749–1761. Taran, Y.A., Connor, C.B., Shapar, V.N., Ovsyannikov, A.A., Bilichenko, A.A., 1997. Fumarolic activity of Avachinsky and Koryaksky volcanoes, Kamchatka, from 1993 to 1994. Bull. Volcanol. 58, 441–448. Tatsumi, Y., Shukuno, H., Tani, K., Takahashi, N., Kodaira, S., Kogiso, T., 2008. Structure and growth of the Izu–Bonin–Mariana arc crust: 2. Role of crust–mantle transformation and the transparent Moho in arc crust evolution. J. Geophys. Res. 113, B02203. http://dx.doi.org/10.1029/2007JB005121. Terashima, S., Ishihara, S., 1986. Copper, lead, zinc, arsenic and sulfur of the Japanese granitoids (3): Green Tuff belt of northeast Japan and outer zone of southwest Japan. Bull. Geol. Surv. Jpn. 37, 605–624. Torgersen, T., 1989. Terrestrial helium degassing fluxes and the atmospheric helium budget: implications with respect to the degassing processes of continental crust. Chem. Geol. 79, 1–14. Toutain, J.-P., Sortino, F., Baubron, J.-C., Richon, P., Surono, Sumarti, S., Nonell, A., 2009. Structure and CO2 budget of Merapi volcano during inter-eruptive periods. Bull. Volcanol. 71, 815–826. Tsukamoto, T., 1982. Showa 55 fiscal year, hot spring usage data list in Japan. Onsenkougakukaishi 16, 1–65 (in Japanese). Varekamp, J.C., Kreulen, R., Poorter, R.P.E., Van Bergen, M.J., 1992. Carbon sources in arc volcanism, with implications for the carbon cycle. Terra Nova 4, 363–373. Varley, N.R., Armienta, M.A., 2001. The absence of diffuse degassing at Popocatépetl volcano, Mexico. Chem. Geol. 177, 157–173. Vita, F., Inguaggiato, S., Bobrowski, N., Calderone, L., Galle, B., Parello, F., 2012. Continuous SO2 flux measurements for Vulcano Island, Italy. Ann. Geophys. 55. http://dx.doi.org/ 10.4401/ag-5759. Wade, J.A., Plank, T., Melson, W.G., Soto, G.J., Hauri, E.H., 2006. The volatile content of magmas from Arenal volcano, Costa Rica. J. Volcanol. Geotherm. Res. 157, 94–120. Walker, J.A., Roggensack, K., Patino, L.C., Cameron, B.I., Matías, O., 2003. The water and element contents of melt inclusions across an active subduction zone. Contrib. Mineral. Petrol. 146, 62–77. Wallace, P.J., 2001. Volcanic SO2 emissions and the abundance and distribution of exsolved gas in magma bodies. J. Volcanol. Geotherm. Res. 108, 85–106.

H. Shinohara / Journal of Volcanology and Geothermal Research 268 (2013) 46–63 Wallace, P.J., 2005. Volatiles in subduction zone magmas: concentrations and fluxes based on melt inclusion and volcanic gas data. J. Volcanol. Geotherm. Res. 140, 217–240. Wallace, P.J., Gerlach, T.M., 1994. Magmatic vapor source for sulfur dioxide released during volcanic eruptions: evidence from Mount Pinatubo. Science 265, 497–499. Wardell, L.J., Kyle, P.R., Dunbar, N., Christenson, B., 2001. White Island volcano, New Zealand: carbon dioxide and sulfur dioxide emission rates and melt inclusion studies. Chem. Geol. 177, 187–200. Webster, J.D., Botcharnikov, R.E., 2011. Distribution of sulfur between melt and fluid in SOHC-Cl-bearing magmatic systems at shallow crustal pressures and temperatures. Rev. Mineral. Geochem. 73, 247–283. Webster, J.D., Goldoff, B., Shimizu, N., 2011. C–O–H–S fluids and granitic magma: how S partitions and modifies CO2 concentrations of fluid-saturated felsic melt at 200 MPa. Contrib. Mineral. Petrol. 162, 849–865. Wedepohl, H.K., 1995. The composition of the continental crust. Geochim. Cosmochim. Acta 59, 1217–1232. Werner, C., Hurst, T., Scott, B., Sherburn, S., Christenson, B.W., Britten, K., Cole-Baker, J., Mullan, B., 2008. Variability of passive gas emissions, seismicity, and deformation during crater lake growth at White Island Volcano, New Zealand, 2002–2006. J. Geophys. Res. 113, B01204. http://dx.doi.org/10.1029/2007JB005094.

63

Werner, C.A., Doukas, M.P., Kelly, P.J., 2011. Gas emissions from failed and actual eruptions from Cook Inlet Volcanoes, Alaska, 1989–2006. Bull. Volcanol. 73, 155–173. Westrich, H.R., Gerlach, T.M., 1992. Magmatic gas source for the stratospheric SO2 cloud from the June 15, 1991, eruption of Mount Pinatubo. Geology 20, 867–870. Witter, J.B., Kress, V.C., Delmelle, P., Stix, J., 2004. Volatile degassing, petrology, and magma dynamics of the Villarrica Lava Lake, Southern Chile. J. Volcanol. Geotherm. Res. 134, 303–337. Witter, J.B., Kress, V.C., Newhall, C.G., 2005. Volcán Popocatépetl, Mexico. Petrology, magma mixing, and immediate sources of volatiles for the 1994–present eruption. J. Petrol. 46, 2337–2366. Yamamoto, M., Koike, T., Masui, F., Shiota, A., Tsurita, H., Ohtsuka, A., Nogami, K., Ossaka, J., 1997. Isotope geochemistry of hos spring waters on the eastern side of KusatsuShirane volano, Gunma prefecture, Japan. Onsenkagaku 47, 68–75. Yoshiike, Y., 1993. Variation in chemical composition of the Tamagawa hot springs, Akita prefecture (1912–1991). Onsenkagaku 43, 56–67 (in Japanese with English abstract). Zapata, A.J.A., Calvache, M.L.V., Cortés, G.P.J., Fischer, T.P., Garzon, G.V., Gómez, D.M., Narváez, L.M., Ordóñez, M.V., Ortega, A.E., Stix, J., Torres, R.C., Williams, S.N., 1997. SO2 fluxes from Galeras volcano, Colombia, 1989–1995: progressive degassing and conduit obstruction of a Decade Volcano. J. Volcanol. Geotherm. Res. 77, 195–208.