The little known Awu volcano is among the highest CO2 degassing source on earth

,


1/ Introduction
Awu is an active volcano located on Sangihe arc, northeast of Indonesia (Fig.1). It is a large edifice that culminates at 1318 m above sea level with a summit crater of 1500 m in diameter. Since 2004, a lava dome has occupied the central part of the crater. Due to its remote location, little is known about its activity -a paradox since this northernmost volcano of Sangihe arc has undergone through two VEI 4 and three VEI 3 eruptions since it first activity recording in 1640 (Table 1). In total, there are at least 18 eruptions recorded in less than four centuries, thus about one eruption every 20 years. The latest eruption was a VEI 2 in 2004. In the database of volcanic eruption victims compiled by , Awu eruptions claimed a total of 5301 victims, mainly following lahar events, including 963 casualties during the 1812 eruption, 2806 during the 1856 eruption and 1532 during the 1892 eruption. But this latter database did not take into account the 2508 victims of the 1711 eruption (Van Padan, 1983; and the 3200 inhabitants killed by lahar events following the 1822 eruption (Lagmay et al., 2007). In 1966, the VEI 4 eruption claimed 39 victims, injured 2000 and forced the evacuation of 420,000 inhabitants (Witham, 2005). In total, since 1711, Awu recurrent eruptive activities caused a cumulative number of 11048 fatalities. This figure classifies Awu as one of the most deadliest, yet poorly studied, volcanoes. In this work, we report on the first volcanic gas emission budget for Awu volcano, with an emphasis on its strong CO 2 output.
In order to derive a degassing output estimate, we performed DOAS measurements in a fixed scanning mode in the crater (Fig.2). To improve chances of catching each degassing source, measurements were performed with an angle of 45° from horizontal, which enable a scanning line just above the crater rim. Twenty four distinct scans were carried out from 8:30 am to 10:40 am (local time) and 23-24 spectra were collected during each scan. The spectrometer used was an Ocean Optics USB2000+ with a spectral range of 290-440 nm and a spectral resolution of 0.5 FWHM. The SO 2 column amounts (ppm m) were retrieved using DOAS calibration and standard analysis procedures (Platt and Stutz, 2008). Reference spectra included in the non-linear fit were obtained by convolving high resolution SO 2 (Bogumil et al., 2003) and O3 (Voigt et al., 2001) cross-sections with the instrument line shape. A Fraunhofer reference spectrum and ring spectrum, calculated in DOASIS, were also included in the fit. The total column amount of the plume cross section was then multiplied by the mean plume rise speed (estimated at 1.3 m/s using thermal camera) to derive the SO 2 emission rate.

4/ Results
Out of 24 scans carried out within 2h of DOAS recording, only 8 scans, acquired on the first part of the measurements, are considered representative (Table 2). They were acquired in clear sky conditions, whilst the other 2/3 of the scans were strongly affected by the rapid formation of cloud coverage. Based on the 8 assumed representative scans, a mean daily SO 2 emission rate of 13±6 tons was obtained, suggesting a rather small degassing, in coherence with the calm volcanic activity state observed in July 2015 (this work).
Multi-GAS recordings highlight distinct gas composition for the 3 measured degassing points (Fig.3, Table 3) with H 2 S/SO 2 ratios of 230, 163 and 49, CO 2 /SO 2 ratios of 1824, 600 and 297, H 2 /SO 2 ratios of 6, 0.8 and 0.1 respectively for MG_Pt1, MG_Pt2 and MG_Pt3 (see Fig.2 for corresponding locations). The water:to:sulfur ratio (H 2 O/SO 2 of 1596) was only retrieved at the MG_Pt3 sampling point, where the H 2 O concentrations were higher (up to > 6000 ppmv) and correlated with SO 2 . In contrast, at MG_Pt1 and MG_Pt2, the H 2 O concentrations were lower (of about 1000 ppmv on average) and poorly correlated with other gases. The degassing spot MG_Pt3 exhibited the lowest H 2 S/ SO 2 (49) and CO 2 /SO 2 (297) ratios, and because of its more SO 2 -rich signature and more vigorous degassing, is considered as the most representative of the magmatic system (e.g., with more limited hydrothermal contributions that in H 2 S-and H 2 -richer MG_Pt1 and MG_Pt2). The gas composition at MG_Pt3 corresponds to 82 mol. % H 2 O, 15 mol. % CO 2 , 2 mol. % H 2 S, 0.05 mol. % SO 2 and 0.02 mol. % H 2 (Table 3), assuming no other representative gas is present in the plume. When used in tandem with the DOAS-derived SO 2 flux, the MG_Pt3 gas ratios imply H 2 O, CO 2 , H 2 S, and H 2 emission rates of 5800 t/d, 2600 t/d, 340 t/d and 0.1 t/d, respectively.

4/ Discussion
Prior to our study here, only one published volcanic gas composition report existed for Awu (Clor et al., 2005). These authors reported on the collection (in 2001) of two close-to-boilingtemperature (96.6 °C) fumaroles in the crater, both exhibiting very hydrous (99.7 mol. %) and CO 2poor (0.18 mol. %) compositions. The gases collected by Clor et al., (2005) were also manifestly Spoor (<0.003 mol. %), and thus radically different from our recalculated MG_Pt3 gas composition (15 mol.% CO 2 , and 2.05 mol. % total S, S T ). Conditions at the time of Clor et al., (2005) sampling in 2001 were very different than in 2015 (this study), as no lava dome existed at that time and the crater area was occupied by a crater lake. Interaction with crater lake water in Clor et al., (2005) is implicated by the more hydrous and S-depleted composition of the gases (water soluble S species are rapidly dissolved during gas transit through volcanic lakes; Christenson et al., 2015). In 2015 instead, no lake was observed and degassing occurred within, and at the margins of, the 2004 lava dome, similarly to what observed at Rokatenda (Primulyana et al., 2017), Lascar (Matthews et al., 1997) or Soufriere Hills (Sparks, 2003).
Our study also provide the first assessment of the volcanic gas output from Awu volcano. The combined DOAS and Multi-GAS results (Table 3) implicate daily mean fluxes of 13 t/d, 5800 t/d, 2600 t/d, 340 t/d and 0.1 t/d for SO 2 , H 2 O, CO 2 , H 2 S and H 2 respectively. We caution that since the composition of the SO 2 -richest (MG_Pt3) fumarole was used in the calculation, our estimated fluxes above should be viewed as lower ranges for CO 2 , H 2 S and H 2 , as these species are comparatively more abundant (than in MG_Pt3) in the more weakly degassing (but numerous) fumaroles around the Awu dome (exemplified by MG_Pt1 and MG_Pt2, see Thus, in its current activity level, Awu only contributes ~0.02% of the daily global volcanic SO 2 emission budget of ~63 kt (Carn et al., 2017). In contrast, Awu's CO 2 emissions are significantly high, and rank the volcano in the upper (>90 % percentile sub-category) range of the global volcanic CO 2 flux population . The CO 2 emissions from Awu thus rival those from some of the most strongly degassing volcanoes worldwide . Assuming the 2015 degassing conditions were representative of the volcano's time-averaged behavior, then with its 0.9 Tg annual CO 2 release Awu would ranks among the top ten strongest volcanic CO 2 sources on earth, contributing to 1-1.3 % of the global CO 2 annual emission from volcanoes (71-87 Tg/yr; Fischer et al., 2019;Werner et al., 2019).
These high CO 2 flux emissions are (at least partially) a direct consequence of the CO 2 -rich Awu gas composition (~15 mol. %, and CO 2 /SO 2 ratios of 287, in the strongest fumarole MG_Pt3). These CO 2 -rich compositions suggest hydrothermal control on degassing even in the most "magmatic" (SO 2richest) MG_Pt3 fumarole, as supported by the prevalence of H 2 S over SO 2 (H 2 S/SO 2 ratio of 49, Table  3). The gas equilibrium temperature, obtained by resolving together the SO 2 /H 2 S vs. H 2 /H 2 O redox equilibria (see methodology in Aiuppa et al., 2011;Moussallam et al., 2017), is circa 380 °C, below thẽ 450 °C threshold temperature  above which hydrothermal S scrubbing (Symonds et al., 2001) becomes negligible. At the <450 °C conditions prevailing at Awu, in contrast, a non-negligible fraction of the original magmatic SO 2 can be consumed by S-depleting hydrothermal reactions such as (Holland, 1965): 4SO 2 +4H 2 O→H 2 S + 3H 2 SO 4 3SO 2 +2H 2 O→S+2H 2 SO 4 These reactions, occurring upon magmatic gas ascent and cooling into hydrothermal envelopes (Symonds et al., 2001), lead to discharge of a CO 2 -rich and S T -poor fumarolic gas. For example, the typical CO 2 /S T range of high-temperature magmatic gases in Indonesia is between 0.3 and ~5 (Aiuppa et al., 2015, with an average at ~4 (Hilton et al., 2002). The CO 2 /S T ratio in the MG_Pt3 Awu fumarole is 5.7, thus implying deposition of a few % up to >90% of the original magmatic SO 2 content.
While thus some extent of hydrothermal processing is implicated by the composition of all Awu fumaroles, including MG_Pt3, we still note that the CO 2 /S T of 5.7 is lower than the threshold value of 10, which is thought to be indicative for magmatic affinity for the gas (Fig. 4; Fischer and Chiodini, 2015). Combining the calculated equilibrium temperature (380°C) with the CO 2 /S T ratio of 5.7 (Fig. 4,  inset), the Awu 2015 MG_Pt3 gas falls in the field of mixed magmatic-hydrothermal gases defined by Aiuppa et al. (2017). We conclude, then, that the 2015 Awu gas composition reflects a combination of (i) magmatic gas supply from a magmatic source plus (ii) some variable extent of hydrothermal processing and S deposition prior to surface discharge.
In view of the above, and also given that CO 2 is less sensitive to scrubbing because of its lower aqueous solubility and reactivity (Symonds et al., 2001), we consider it possible that the high CO 2 emissions are hints for the presence of a carbon-rich melt source beneath Awu volcano. A C-rich magmatic source would be consistent with the findings from Jaffe et al. (2004) and Clor et al. (2005) who found high CO 2 / 3 He (64-180·10 9 ) and δ 13 C (≥-2 ‰) values (in addition to a nitrogen isotope signature of -3.3 ‰) at Awu and Karangetang, the two volcanoes at the northern part of the Sangihe arc. Such positive C isotope compositions (and C excesses relative to 3 He) are likely indicators for the involvement of C-rich, slab sediment-derived fluids in the mantle source. The Sangihe forearc is currently overriding the Halmahera forearc (Hall and Wilson, 2000;Jaffe et al., 2004;Bani et al., 2018) (Fig.1) and the collision event is more advanced in the northern Molucca sea (Morris et al., 1983) where slow-down of collision has been evidenced from seismic recording (McCaffrey, 1983;Pubellier et al., 1991). Such scenario may possibly lead to enhanced heating of the slab (Peacock et al., 1994), thereby promoting greater production of C-rich melts and/or fluids beneath Awu volcano (Jaffe et al., 2004). The hypothesis of a C-rich magmatic source will require testing from analysis of volatiles stored in crystal-hosted melt inclusions and, if verified, may imply sizeable CO 2 emissions during the relatively frequent explosive (up to VEI 4) Awu eruptions (Kodera et al., 2008).

5/ Conclusion
Awu is the largest and northernmost active volcano of Sangihe arc. Due to its remote location very little is known about its activity, a paradox since this volcano went through numerous eruptive activities, including two with VEI 4. Since 1711 these recurrent eruptions claimed 11048 fatalities, a figure that emphasizes Awu as one of the deadliest volcanoes worldwide. In July 2015, a gas survey carried out in the crater spotlights an elevated CO 2 degassing from the numerous fumaroles around a lava dome situated in the center of the crater. Despite a low SO 2 flux of 13 t/d, the CO 2 emissions are relatively high, with an estimated annual output of 0.9±0.4 Tg/yr that represent ~ 1% of the global volcanic CO 2 emission budget. This high CO 2 output may result from the peculiar geodynamic context of the region, where the slow down of arc-to-arc collusion has produced heating of the slab, potentially leading to larger delivery of C-rich fluids/melts.