1. SCIENCE MOTIVATION

Magmatism at subduction-related volcanic arcs is thought to form a large component of the continental crust resulting in a bulk composition that is andesitic to dacitic [e.g., Rudnick, 1995]. However, primitive mantle melts produced in the mantle wedge at subduction zones are mafic and must evolve to produce the more silicic magmas at arc volcanoes. The processes by which this evolution occurs likely take place within the crust and include: (1) crystallization of mafic melts in shallow crustal magma chambers [Sisson and Grove, 1993; Pichavant, 2002], (2) continuous evolution of the magma as it resides in and moves through multiple levels of magma reservoirs [e.g., Dufek and Bachmann, 2010], (3) differentiation that occurs almost entirely in the lower crust [e.g., Müntener et al., 2001; Annen and Sparks, 2002], (4) mixing with silicic melts of the surrounding country rock [e.g., Jackson et al., 2003; Tatsumi and Kogiso, 2003], and (5) mixing of melts formed by differentiation of mafic magma in the lower crust and by partial melting of crustal rocks [e.g., Hildreth and Moorbath, 1988; Annen et al., 2006]. In these models, formation of the required volumes of intermediate composition magma typically results in mafic cumulates that are far thicker than the observed mafic lower crust. It has been suggested that these dense cumulates delaminate and sink into the mantle [e.g., Jull and Keleman, 2001; Tatsumi and Kogiso, 2003], or that the regions directly below the seismic Moho contain the mafic cumulates [e.g., Fliedner and Klemperer, 2000; Tatsumi et al., 2008]. Alternative possibilities to account for the discrepancy in mafic lower crustal thickness are that melts delivered from the mantle are already intermediate in composition, either due to reaction of ascending magmas with mantle peridotite [e.g., Kelemen, 1995], or the direct formation of dacitic melts by partial melting of the subducting slab itself [e.g., Defant and Drummond, 1990; Pearce and Peate, 1995].

Recently, several sophisticated petrologic models have been developed that can be used to model the thermal, compositional, and isotopic evolution of crustal magmatic systems [Ghiorso and Sack, 1995; Spera and Bohrson, 2001; Dufek and Bergantz, 2005; Annen et al., 2006; Dufek and Bachmann, 2010; Solano et al., 2012]. However these models are non-unique because there are few physical constraints on the geometry, crystal content, and interconnectedness of magma reservoirs at various depths within the crust and on the associated magma fluxes. What is needed are better constraints on the sizes and shapes of the entire magma system as well as of the temperatures and compositions of the wall rocks. These observations are essential for determining solidification rates and/or the ability to assimilate both wall rock and previously formed cumulates in the crust. They also can provide constraints for models of the timing of melt segregation and of the dynamics of magma stagnation or ascent through the crust.

An emerging view is that silicic caldera systems may exist at near-solid conditions for thousands of years, but that the time it takes to form a large, mostly liquid magma chamber is surprisingly short – perhaps as little as a couple of months [e.g., Druitt et al., 2012; Cooper and Kent, 2014]. Furthermore mafic recharge or rapid ascent and assembly of existing magma batches may play an important role in generating eruption-ready conditions [e.g., Kent et al., 2010; Parks et al., 2012]. Thus late-stage growth spurts of shallow magma reservoirs are likely to precede caldera-forming eruptions. Late-stage growth is likely to be accompanied by an increase in melt fraction in the shallow magmatic system as well as observable increases in stress and deformation of the volcanic edifice.

Seismology is a powerful tool that allows us to probe the subsurface architecture of arc volcanic systems. In particular, the novel active source seismic imaging we propose here provides exceptional resolution to determine: the depth, geometry and melt content of magma reservoirs throughout the crust and at the Moho, whether these magma bodies are connected by dike systems or by vertical, crystal-rich complexes, and the structure and properties of the surrounding crust. With a better physical determination of the size, shape, and depth of magma bodies, scientists can then develop models that address key questions about arc volcanic systems including (1) the chemical evolution of magmas, and (2) the timing and dynamics of magma migration, storage and eruption.

We choose the active Santorini volcano in the Hellenic arc of the Mediterranean for our high-resolution seismic imaging experiment because: (i) its magmas are primarily dacitic and andesitic and thus are similar to upper continental crust [see review in Druitt et al., 1999], (ii) both its physical structure, composition and temporal evolution have been well studied [Druitt et al., 1999; Fabbro et al., 2013]; (iii) it has been suggested to contain multi-level magma reservoirs [e.g., Cadoux et al., 2014]; (iv) rapid recharge of the shallowest reservoir is thought to have occurred in the past [Druitt et al., 2012]; (v) it experienced a very recent phase of seismic activity and deformation suggesting it might be entering a phase of rapid rejuvenation [Newman et al., 2012; Papoutsis et al., 2012; Parks et al., 2012]; (vi) the recent deformation roughly constrains the depth of the shallow magma reservoir; (vii) there is an existing seismicity catalogue from the island seismic network; and (viii) important for our study is that the volcanic system is semi-submerged, allowing us to conduct a dense marine-land seismic experiment that will yield significantly higher resolution images than are possible at any on land arc volcano.

2. SCIENCE QUESTIONS

Here we propose to test the following scientific hypotheses about how the structure of the magma plumbing system at arc volcanoes controls the storage and evolution of magmas within the crust and how magmas may be connected as they move to the surface:

1. Crystallization of mafic melts occurs in shallow crustal magma chambers. In this case the magma system is dominated by a shallow magma reservoir and the magma in the deeper crust primarily feeds this shallow reservoir. The crust beneath the shallow reservoir is then dominated by mafic cumulates and contains the magma feeder system. The eruptibility depends on the current state and crystal content of this shallow reservoir. If this reservoir is horizontally extensive and melt-rich, a caldera-forming eruption is probably geologically imminent [Fabbro et al., 2013], though such events are infrequent and these chambers have not been observed geophysically.
2. Magma evolves continuously as it resides in, and moves through, multiple levels of magma reservoirs. In this view the magma system is active throughout the crust and consists of multi-leveled reservoirs that may have a range of melt contents at any one time. The overall structure of this system might be columnar [e.g., Lees, 1992] or include sills at various levels [e.g., Tarasewicz et al., 2012]. The surrounding rock might be formed largely by cooled plutons from earlier magmatic intrusions [e.g., Beachly et al., 2012] and the proportion of mafic cumulates increases with depth. The connections between magma reservoirs could be either dikes or more crystal-rich mush zones and the eruptibility of this system will depend on the nature of these connections and the existing melt fractions.
3. Differentiation and/or mixing with melts of surrounding rock occurs almost entirely in the lower crust. In this case, the deep magma system is active and the lower crust is hot and may contain one large [e.g., Ma and Clayton, 2014], or several dispersed [e.g., Zellmer et al., 2003; García-Yeguas et al., 2012] magma volumes that are embedded within a large volume of mushy rock. Since most differentiation happens in the deep crust, mafic cumulates are localized near the base of the crust and below, or have delaminated. The eruptibility of this system will depend on the current state of the deep reservoir and the mobility of magma to shallower levels [e.g., Elsworth et al., 2008; Reverso et al., 2014]. Shallow magma reservoirs may be typically short-lived and either form shallow intrusions or source eruptions [e.g., Caricchi, 2014].

3. SCIENTIFIC APPROACH

Testing the above hypotheses about magma storage, transport, compositional evolution and eruptibility requires better physical constraints on the geometry, crystal content, and the nature of interconnections of magmas at all depths throughout the crust. We propose a novel, high-resolution seismic experiment at the active and semi-submerged Santorini volcano that takes advantage of high density spatial sampling of the seismic wavefield and state-of-the-art inversion methods to provide new insights into the structure of the whole crustal magmatic system. We will apply two complementary seismic analyses that will together result in unprecedented high resolution images of the full crustal magma system beneath an arc volcano:

Dense 3D anisotropic travel time tomography to resolve the first order structure of the magma plumbing system and surrounding crust. Active source travel time tomography with dense Pg, PmP and Pn ray coverage has a resolution of several wavelengths (~1 to 4-6 km from the upper crust to the uppermost mantle) that can provide velocity images of the rocks that make up a volcanic edifice and surrounding region [e.g., Waite and Moran, 2009; García-Yeguas et al., 2012; Paulatto et al., 2012]. Travel time tomography can also image 3D variations in crustal seismic anisotropy, which result from tectonic and magmatic stresses [e.g., Weekly et al., 2014]. However, travel time tomography has had limited success in resolving small magma bodies because wavefront healing erases delays in the arrival time of the first seismic phase [e.g., Beachly et al., 2012]. This approach will allow us to determine the velocity structure of the entire crust and uppermost mantle with sufficient resolution to distinguish medium-scale shallow and deep regions of either low or high velocity. Using this analysis we can determine whether hypotheses 1, 2 or 3 are most likely, but we will be limited in our ability to resolve smaller magma volumes and will not be able to determine in detail the geometry and boundaries of any magma reservoirs nor their connections.

Full waveform inversion tomography and waveform modeling to refine travel time tomography images and to obtain high resolution maps of elastic properties throughout the crust. Use of the full seismic wavefield significantly improves seismic imaging because, for example, the interaction of seismic energy with a low-velocity magma body often produces distinct refracted and reflected secondary phases and because waveforms are sensitive to gradients in the velocity structure. We will use a combination of forward modeling [e.g., Durant and Toomey, 2009; Beachly et al., 2012] and waveform inversion to image structure in high resolution [e.g., Morgan et al., 2013]. While it has only recently become computationally tractable, 3D full waveform inversion tomography (FWI) of refracted arrivals has a resolution of half the seismic wavelength [Virieux and Operto, 2009] and a greater ability to accurately resolve changes in elastic parameters [e.g., Morgan et al., 2013]. Consequently this approach results in much more accurate reconstruction of small magma bodies than is possible with travel time tomography alone, especially in highly heterogeneous environments such as volcanoes. This will allow us to determine in much more detail the presence, size, shape and distribution of magma bodies at all depths in the crust [e.g., Arnoux et al., 2014] and distinguish between hypotheses 1, 2, and 3 as well as provide new insight into the presence of sills and /or mush volumes and the melt contents and distributions and connections between bodies.

4. SANTORINI VOLCANO

Santorini volcano is a shallow-water volcanic complex in the southern Aegean formed by arc volcanism related to the subduction of the African plate beneath the Eurasian plate along the Hellenic arc (Figure 1; Heiken and McCoy, 1984; Druitt et al., 1999]. About 1650 B.C. [Friedrich et al., 2006], a series of massive Plinian eruptions expelled some 40 – 60 km³ of volcanic material and collapsed the central island surface, forming the present submerged caldera and causing a massive regional tsunami [Heiken and McCoy, 1984; Sigurdsson et al., 2006]. Known as the Minoan eruption, the event is arguably the main cause of the demise of the seafaring Minoan civilization [Hardy and Renfrew, 1990; Manning et al., 2006].

Santorini has remained active with five subaerial eruptions since 1570 AD [Pyle and Elliott, 2006]. The historical eruptions led to the growth of the Nea Kameni islands near the caldera center and had repose times ranging from 14 to 160 years. After the most recent eruption in 1950, the volcano remained quiet for more than 60 years, with few seismic events and an absence of ground deformation. However, an intense earthquake swarm (January 2011 to spring 2012), which coincided with very high rates of inflation (150-180 mm/yr), suggests that Satorini may be entering a new episode of activity [Newman et al., 2012; Parks et al., 2012].

The Santorini volcano lies on extended continental crust about 15 km thick [Figure 4 of Bohnhoff et al., 2001] and is located on a horst, the Santorini-Amorgos ridge, formed by the NE-SW trending Santorini-Amorgos Fault Zone (SAFZ, Figure 2) [Piper and Perissoratis, 2003; Dimitriadis et al., 2009; Nomikou et al., 2012; Feuillet, 2013]. This normal to trans-tensional fault zone has produced the region’s two largest earthquakes in modern history – magnitude 7.5 and 6.9 normal-faulting events on 9 July 1956 (see location in Figure 2) [Okal et al., 2009]. The Santorini volcanic field extends more than 20 km to the northeast of the main edifice and consists of numerous submarine cones, of which the largest is Columbo seamount, located just 10 km to the NE (Figure 2) (common alternative spellings include Coloumbo and Kolumbo). Columbo has a well-defined, shallow-rimmed caldera that lies ~500 m below the sea surface. The last known eruption of Columbo was on 30 September 1650 AD and generated a tsunami [documented in historical reports, memoirs and letters – see summary and references in Dominey-Howes et al., 2000]. Columbo also hosts a submarine hydrothermal system, with both high and low temperature venting [Carey et al., 2013; Kilias et al., 2013].

Prior to the recent unrest, the main Santorini edifice was relatively quiescent, though there was abundant seismicity associated with the nearby Columbo volcano [Dimitriadis et al., 2009, 2010] (Figure 2). Beginning on 9 January 2011 an intense earthquake swarm began inside the Santorini caldera [Newman et al., 2012; Parks et al., 2012; Papoutsis et al., 2012]. The swarm was centered beneath the Palea Kameni and Nea Kameni islets, sites of previous intra-caldera eruptions, and also extended laterally toward the main town of Thira (Figure 3) along a near vertical fault at depths between 1 and 6 km. Swarm activity with dozens of recorded earthquakes (M_L 1.0-3.2) per day persisted through the spring of 2012 [Newman et al., 2012].

Coincident with the onset of the January 2011 earthquake swarm, Santorini began inflating at a remarkably rapid rate (Figures 3a,b) [Newman et al., 2012]. Rapid ground deformation—with measured deformation rates as high as 150-180 mm/yr on Nea Kameni—continued through early 2012 and then slowed [Papoutsis et al., 2013]. The observed uplift pattern was radially

symmetric, and various modeling efforts [Newman et al., 2012; Parks et al., 2012; Papoutsis et al., 2013] showed that this deformation was consistent with a source positioned 3.3-6.3 km beneath the submerged northern section of the caldera, offset from the seismic swarm.

Existing local earthquake tomography studies do not resolve the seismic structure directly beneath the Santorini caldera deeper than 2-3 km, but do reveal the presence of a low velocity body at ~6 km beneath Columbo and a possible connection between Santorini and Columbo volcanoes [Dimitriadis et al., 2010]. Petrologic studies suggest that all four large silicic caldera-forming eruptions at Santorini were fed from melt bodies at very similar depths ~2 kb (8 km) [Fabbro et al., 2013; e.g., Cadoux et al., 2014]. A deeper magma body (~15 km) is also indicated by these phase equilibria studies [Mortazavi and Sparks, 2004; Cadoux et al., 2014]. However there are no geophysical constraints on the deep magma plumbing structure beneath the Santorini volcanic system, nor the physical state of the remnant magma body from the most recetn (1650 AD) caldera-forming eruption.

5. PROPOSED EXPERIMENT

We propose a novel high-resolution, active source seismic experiment of the Santorini volcanic system coupled with a microearthquake study centered at Columbo seamount (Figure 4). Our goals are to constrain the volume and distribution of magma throughout the entire crust and quantify the evolving state of stress, which will help test our hypotheses on magma storage, crystal content, and connectivity beneath silicic caldera-forming arc volcanoes. The proposed experiment is comprised of two elements. The first element records data suitable for 3D travel time tomography of the crust and mantle. The second element is full waveform inversion for a volume embedded in the 3D travel time image. We describe how the experiment design will contribute data to each of our proposed analyses and allow us to test our scientific hypotheses.

1. A 3D isotropic and anistropic tomographic image of the velocity structure beneath the Santorini volcanic system, including an initial estimate of the volume and geometry of melt bodies throughout the crust and uppermost mantle as well as the thermal and compositional structure of the edifice and its surroundings. The array that will collect travel time data for the tomographic imaging is shown in Figure 4 (yellow OBSs, red and orange land stations, blue shot lines). The array consists of 93 short period ocean bottom seismometers (OBSs) and 26 land stations that will record ~3600 sources from the R/V Marcus Langseth’s airgun array from a large range of distances and azimuths (details of the instrumentation is given in §7). Our array design balances the number of short period OBSs available in the OBSIP fleet with the following design criteria: (i) recording large aperture arrivals from a swath of azimuths to probe the lower crust and uppermost mantle with a high density of crossing rays – the existence of surrounding islands leads to the SW-NE elongation of the array; (ii) dense instrument and shot coverage with a sufficient range of azimuths to fully sample the wavefield for full waveform inversion – this determines the OBS spacing and shotline spacing in the NE and SW parts of the array (see 2. below); (iii) sufficient coverage to resolve the 3D structure and heterogeneity of the volcanic edifice at least in the upper to mid crust – hence the OBSs and shots to the north and east of Santorini; (iv) avoidance of extreme topography on the SE side of Santorini – and so we exclude the SE flank from our array. Shots on the NW flank of the volcano provide arrivals on all the OBSs (increases depth and azimuthal coverage), but we cannot also deploy OBSs in this region without sacrificing station density elsewhere in the array.

 

Our tomographic analysis will use Pg, PmP and Pn arrival times to obtain 3D isotropic and anisotropic images of the entire crust and uppermost mantle. Arrivals from the shot profiles to the entire OBS array will provide good azimuthal coverage from all angles except the SE quadrant and will span a large range of distances. A synthetic ray diagram for the proposed array (Figure 5) shows that it will give excellent illumination of the volcanic edifice and surrounding rocks and will allow us to distinguish large volumes of hot or mushy rock from more competent and cooler plutons throughout the crust – these first order features can determine whether there are large bodies of hot rock or magma in the upper or lower crust (hypotheses 1 and 3) or whether the magma system is more columnar (hypothesis 2). Anisotropic inversions will address the spatial and depth variations of crack- or stress-induced anisotropy within and around the volcanic edifice, which may in turn be related to rates and style of seismicity.

2. Full waveform inversion beneath the Santorini volcanic system to obtain higher resolution and more accurate recovery of the elastic properties of the crust, including an estimate of the sharpness of magma chamber boundaries, and the spatial connections, melt content and distribution of magma bodies. To critically test hypotheses 1, 2, and 3 requires

high density sampling of the wavefield combined with 3D FWI. Only the recently-developed FWI method can resolve the detailed features and variations in Earth properties predicted by our hypotheses. The portion of the array that will provide data for FWI is outlined by the white box (Figure 4). We base our expected resolution on the recovered models for a 3D synthetic study of Montserrat. That study comprises two magma chambers and shows excellent recovery of both the shallow and deep chambers (Figure 6) [Morgan et al., 2013]. In Figure 7, we specifically test the resolution of our proposed experimental geometry (Figure 4) to anomalies at 11 km depth and show that dense OBS coverage on the NE side of the volcano (where Columbo seamount is located) and less dense, randomly spaced OBS coverage on the SW side provides sampling of the wavefield that allows exceptional recovery compared to travel time tomography. The synthetic inversions were performed using an acoustic 3D time-domain code, with peak inversion frequencies from 1.0 to 1.8 Hz [Morgan et al., 2013]. The results unequivocally show that we can image the detailed structure of the magma chambers, distinguish between isolated chambers versus a continuous magma system, and provide, for the first time, an image of the lower crustal structure of an arc volcano.

We have also run a suite of tests that show that we can distinguish between abrupt and diffuse chamber tops, and image layers of magma in the crust. Melt fraction and melt geometries have a strong effect on P and S wave velocity, anisotropy, and attenuation [Hammond and Humphreys, 2000b; 2000a], and we will include these parameters in our analysis. This approach will allow us to resolve the detailed magmatic features and physical properties predicted by our hypotheses and to distinguish regions of partial melt from regions of hot-but-solid rock. Key questions are: Do ponded sills of mafic magma currently exist in the lower crust or at the crust-mantle interface? What are the melt and crystal contents? What is the nature of the connections between all levels of the crust?

6. DATA ANALYSIS

3D Anisotropic Travel Time Tomography (PIs Toomey & Hooft & UO graduate student): We will use a tomographic method that inverts primary and secondary arrival times (e.g, Pg, PmP and Pn) for 3D isotropic and anisotropic velocity structure [Toomey et al., 1994; Dunn et al., 2005]. Our code is fully 3D, correctly accounts for elevation, includes anisotropic ray tracing, can include earthquake sources, and has been implemented in parallel on the high-performance computing system at UO (see facilities). The results of 3D anisotropic tomography also provide a necessary starting model for FWI. Recent FWI work done in collaboration with Imperial College shows that our anisotropic starting model (as opposed to the best-fitting isotropic model), significantly improves the fit to waveform data from the Endeavour experiment [Arnoux et al., 2014].

Full Waveform Inversion Seismic Tomography (PI Hooft & UO graduate student & J. Morgan, Imperial College, UK): Seismic waveforms carry considerably more information than the travel times of primary and secondary phases. In fact, the volume and melt content of the magma system can be more accurately constrained by waveform modeling in addition to tomographic inversion as shown in our forward modeling studies at Newberry volcano [Beachly et al., 2012] and the EPR [Durant and Toomey, 2009]. However, fully utilizing the information in the seismic wavefield requires fully 3D transmission FWI. The potential for FWI to improve

spatial resolution has been known for some time [Pratt, 1999], but early wavefield inversions were two-dimensional, and neglecting the third dimension can lead to poor velocity recovery. The availability of high-performance computing facilities, coupled with the development of more efficient algorithms, means that realistic-sized problems can now be solved in 3D [Warner et al., 2013]. The use of 3D FWI codes has produced spectacular successes [Warner et al., 2013] and resulted in changes in the oil industry; many companies now acquire long-offset, low-frequency seismic data, and use FWI to image beneath complex overburden and locate deep reservoirs.

We base our experimental design on synthetic studies by Morgan [2013], which show that: (i) 3-D FWI captures subsurface structure at a scale that is several times smaller than the source or receiver separation provided that the model is well covered by many crossing wavefields – achieved here by using many marine sources; (ii) FWI recovers fine details of the subsurface in regions where there are only sources and no receivers at all, again provided that many independent wavefields sample the area, and (iii) transmission FWI is robust against random noise, against inclusion of surface multiples, and against wide or irregular receiver spacing, and does not require or gain particular advantage from short-offset reflections.

We build on our collaboration with Prof. Joanna Morgan (Imperial College) to use FWI to refine the images from travel time tomography and recover both finer-scale seismic structure as well as more accurately recover the elastic properties. FWI is an iterative inversion scheme and the starting model is successively improved though minimizing the misfit between calculated and observed data. To implement FWI we will start with the anisotropic tomography velocity model and extract the source wavelet from the recorded data [Pratt, 1999]. Experience indicates that best practice includes inverting from low-to-high frequencies, early-to-late arrivals, short-to-long offsets, and matching phase before amplitude. Hence, we start with a long-wavelength velocity model, update the near surface and then deeper velocity structure, whilst gradually increasing the spatial resolution. This approach, supported by careful quality control, allows us to avoid problems with cycle-skipping [Warner et al., 2013]. The success of this approach is already clear in our collaborative work on applying FWI to the ETOMO dataset from the Endeavour segment [Arnoux et al., 2014]. The Imperial College FWI code is installed on the UO ACISS parallel compute system, it has been compiled to run as efficiently as at Imperial College, and we have been successfully running inversions for the ETOMO experiment. For our ETOMO dataset, each FWI requires 2 days of run times and the synthetics for this proposal took 3 days. The proposed purchase of two ACISS cores provides us with the equivalent number of core-hours across the ACISS facility required for over one hundred runs. Finally, Morgan will run more synthetics for Santorini to further optimize the experimental design (see letter of support).

Integration of seismic and other datasets (All PIs & Collaborators). Models for magma evolution and dynamics need better physical constraints on the structure, melt content and interconnectedness of the magmatic system at all depths in the crust. We will work with petrologists and geochemists to combine our physical parameters with observations that constrain crustal assimilation, magmatic differentiation and fluid content – a necessary step to fully model how magma compositions evolve within the crust. Our results will also provide physical constraints on models of magma movement and eruption dynamics and this study will give us a view of the interplay between intrusive processes and seismic faulting. We will also work with our international partners to include existing seismic travel time and waveform data sets in our analyses and better constrain our models using MCS datasets and geodetic deformation measurements.

How a Santorini experiment complements studies at Mount St. Helens (iMUSH): These two arc volcanoes are end-members located in very different tectonic settings; both sit on continental crust, but while the crust at St. Helens is ~45 km thick, that at Santorini is extended and ~15 km thick. These contrasts likely result in different magma plumbing systems. At Santorini the thin crust, together with collection of a large volume of data with dense shot spacing, allows us to apply advanced methods (FWI) to recover lower crustal structure with high resolution. The magma system in the upper crust has been probed at several volcanoes, consequently a compelling and not yet imaged scientific target is the lower crust and its connections to shallower magma bodies. Santorini is an ideal location for imaging the lower crust.

7. EXPERIMENT LOGISTICS

Our plan is to deploy all 93 OBS and the 9 temporary land seismometers (Prof. Costas Papazachos, U. Thesaloniki) in the first year (fall 2015) for the active experiment. We request the R/V Langseth as the airgun source for the active experiment since the volume and tuning of this source provide the deep penetration and clean waveforms critical for the tomographic and waveform modeling goals. The shot interval will be 30 to 170 s, or ~80 to 400 m shot spacing, at 4.5 knt for ~900 km line length. This shot spacing achieves a dense shot spacing for maximal ray coverage and yet allows enough time that noise from earlier shots diminish in the water column.

The OBSs include all the short period WHOI D2 and SIO L-Cheapo instruments from the OBSIP facility, which feature comparable characteristics and capabilities. These 4-component instruments (3 component seismometer & hydrophone) are compact and include 24-bit analog-to-digital convertors and high precision clocks ideal for large active source experiments. The land stations are described in detail in the subaward proposal by Aristotle University of Thessaloniki (AUTH) and include the 12 existing permanent stations of the AUTH – Institute for the Study and Monitoring of the Santorini Volcano (ISMOSAV); the permanent Hellenic Unified Seismological Network station, SANT; and 9 temporary stations from AUTH. All 26 land stations will collect data suitable for the proposed travel time tomography, since 7 of the permanent stations are older analog stations, only 19 land sites may be useable for FWI.

We will begin the process of obtaining permits for the airgun survey as soon as we are informed that our project may be supported. Several active source experiments have occurred in recent years in the waters around Santorini [e.g., Sakellariou et al., 2010; Nomikou and Papanikolaou, 2011; Nomikou et al., 2013]. Dr. Paraskevi Nomikou (U. Athens) will lead the effort to obtain permits for the active experiment in Greek waters and we anticipate that this process can be done rapidly, see subaward proposal From Univ. Athens (NKUA). Both Nomikou and Papazachos have experience working with foreign scientists on seismic experiments and their contacts will be an invaluable resource (see letters of support).

R/V Langseth: We are requesting 22 days aboard the R/V Langseth to perform the active airgun experiment. Airgun time includes turns with a minimum 1.5 km radius and 1 maintenance day.

Transits 1 day

Deploy & Recover 93 OBS 11 days

Airgun operations 9 days

Contingency/marine mammals 1 day Total = 22 days

Previous or on-going projects by EU scientists in the region: Studies by other colleagues provide a framework for our proposed work: (1) German scientists [e.g., Bohnhoff et al., 2004; Meier et al., 2004; Bohnhoff et al., 2006; Endrun et al., 2008; Friederich and Meier, 2008; Endrun et al., 2011] have collected a land-sea passive seismic dataset that samples the entire Aegean and Hellenic subduction zone at a scale suitable for imaging upper mantle structure and assessing the regional stress state [Friederich et al., 2014]. (2) Our US colleagues are proposing a 2D MCS/wide-angle seismic study of the Hellenic subduction zone (see letter of support A. Bécel). (3) In May –June of 2015, Dr. Nomikou will participate in a Spanish project (PIs C. Ranero and V. Sallares) to collect a wide-angle and MCS survey of the Amorgos basin. The goal is to collect two regional profiles to resolve crust and upper mantle structures and also acquire some deep MCS lines around the Santorini edifice. These 3 international efforts provide excellent complements that place our study in the context of subduction processes and mantle structure and within the regional tectonic setting. (4) During July 2012, French researchers (J. Escartin) installed pressure and temperature gauges in the Santorini caldera to record variations in seafloor depth and hydrothermal outflow, and conducted systematic mapping of this hydrothermal site; these instruments were recovered in fall 2013. Integration of our results with these observations will address heat transfer in the Santorini and Columbo calderas.

8. RESPONSIBILITIES

UO will serve as the lead institution and will be responsible for coordinating the active source cruise logistics. All institutions will be involved in scientific planning and design of the OBS network. Our Greek partners provide critical experience and contacts for both the on- and off-shore portions of this work. UO and Nomikou (U. Athens) will be responsible for coordinating cruise logistics with the Greek ship and obtaining permits for the airgun survey in Greek waters. Nomikou will act as the liaison with cognizant Greek fishing authorities in site selection of the 6 OBS. Papazachos (U. Thessaloniki) will be responsible for the island seismic data including repairs to the land network on Santorini and the siting and deployment of 9 temporary seismometers that will record during the active portion of the experiment. Responsibilities for data analysis are indicated in §6.

9. BROADER IMPACTS

The opportunity to apply FWI to the magma structure of an arc volcano grows from our international collaboration with Prof. Joanna Morgan at Imperial College, UK. These close international ties will contribute to training our graduate students in advanced seismic and computationally intensive methods. The UO, U. Athens, and U. Thessaloniki will involve undergraduate and graduate students in the seagoing field program. The data collected here will be the basis for several thesis projects and will be published in a series of scientific papers.

Our Greek collaborators greatly enhance the broader impacts of this research. Drs. Papazachos and Nomikou will be our liaisons with local government officials and with the Institute for the Study and Monitoring of the Santorini Volcano (ISMOSAV), of which Papazachos is Secretary. Nomikou has a family home on Santorini and she has considerable prior experience with risk assessment on the island. Papazachos runs a large permanent seismic network on the island.

1. Seismic monitoring of Volcanic and Seismic Hazards. If Santorini does erupt, it will likely be on the scale of the smaller pyroclastic and/or phreatic activity similar to the historical eruptions [Georgalas, 1953; Druitt et al., 1999; Pyle and Elliot, 2006]. Ashfall and ballistic bombs could reach the habited areas of Santorini. In addition, earthquake activity could damage houses and induce landslides along the steep caldera cliffs, which may generate local tsunamis [Konstantinou et al., 2012]. The economic impact of such events is substantial because Santorini constitutes 1/3 of Greece’s tourist income and so is vital to the (currently very weak) economy. Our proposal directly enhances monitoring of volcanic and seismic hazards by including repairs to the Santorini seismic stations and by adding 4 permanent, real-time, short-period sensors.

2. Engagement of co-workers with Greek agencies (EPPO, Santorini Municipality) to inform disaster management planning. Our scientific results will feed directly into the development of hazard and risk assessments and scenario planning for potential future volcanic activity at Santorini and Columbo volcanos. Technical briefings will be provided to inform both decision-making and scientific-advisory bodies both regionally and nationally in Greece, including the Earthquake Planning and Protection Organisation (EPPO), which has responsibility for evaluating volcanic hazards and for determining the response to ongoing events at volcanoes in the Aegean, ISMOSAV, the National Observatory of Athens, the Greek National Committee for the Monitoring of Santorini Volcano, and the Mayor of Santorini. Results will enable these agencies and individuals to enhance and strengthen hazard assessments and disaster management plans. Briefings will be provided to the EPPO and Mayor of Santorini during survey operations.

3. Greek workshop on natural hazards attended by scientists, local community stakeholders (e.g., Mayor Santorini). A volcanic hazard workshop will be held on Santorini at the conclusion of field operations. This workshop will be organized and led by Nomikou and Papazachos, who will also seek local funding for the workshop, and will be hosted by the municipality of Santorini where there is an appropriate conference room. Participants representing geoscience, social science, and engineering research disciplines will be invited along with representatives from the Greek government (including the EPPO and GSRT), international non-governmental humanitarian and development organizations (NGOs), think tanks, and interested citizens.

4. Release of podcasts, video diaries, a YouTube animation and other educational resources, advertised via social media. A wider interest group will be engaged through online educational and outreach activities featuring the geophysical technologies. This will be accomplished through the development of a high quality podcast (both in English and Greek and available through the U. Oregon and U. Athens podcast series, and released on iTunes) and You Tube videos on the project (Hooft had considerable impact with a YouTube video for Newberry volcano²). These formats can attract a substantial audience, particularly when embedded within a much larger portfolio of materials, and when promoted using social media.

10. SUMMARY

We propose an ambitious, active-passive seismic experiment to delineate with unprecedented resolution the volume and geometry of melt throughout the crust of an arc volcano and test 3 scenarios for the storage and evolution of magmas within the crust and their connections. The dataset that we propose to collect is unique in the depth and density of coverage and will allow insights into the lower crustal structure of the magma system that are on a truly new scale. This is possible because a marine experiment with dense shot spacing can be conducted at the semi-submerged Santorini volcano. The proposed dataset will provide exceptional sampling of the seismic wavefield allowing us to apply both 3D anisotropic travel time tomography and state-of-the art full waveform inversion methods.

Results from Prior NSF Support

Imaging the Upper Crust at Newberry Volcano Using Large-Offset Reflections: E. Hooft & D. Toomey (University of Oregon EAR- 0813978; 08/01/2008-07/31/2010, $102,478). This project supported the collection of densely-spaced seismic data across Newberry volcano. We combine seismic tomography with waveform modeling [Beachly et al., 2012] and perform joint inversion of active and passive-source travel time data [Heath et al., 2014]. These analyses constrain the dimensions and melt content of a magma body in the upper crust with more detail than can be done with active travel-time tomography alone. Broader impacts: collaboration with the USGS, training of three graduate students (2 @ UO and 1 @ Michigan Tech) and involvement of 2 community college, 5 undergraduate, 6 graduate students and 3 community members in the fieldwork. Public outreach: a video developed with UO Digital Arts majors on YouTube¹ and also at the Lava Lands Interpretive Center; several community articles^2,3; seismic wave propagation movies⁴; presentations at the US Forest Service and at the Science Pub night in Bend.

References:

Annen, C., and R. Sparks (2002), Effects of repetitive emplacement of basaltic intrusions on thermal evolution and melt generation in the crust, Earth Planet Sc Lett, 203(3), 937–955.

Annen, C., J. D. Blundy, and R. Sparks (2006), The genesis of intermediate and silicic magmas in deep crustal hot zones, Journal of Petrology, 47(3), 505–539, doi:10.1093/petrology/egi084.

Arnoux, G. M., B. P. VanderBeek, J. Morgan, E. E. E. Hooft, D. R. Toomey, W. S. Wilcock, and M. Warner (2014), Advanced Seismic Studies of the Endeavour Ridge: Understanding the Interplay among Magmatic, Hydrothermal, and Tectonic Processes at Mid-Ocean Ridges, AGU 2014. Available from: https://agu.confex.com/agu/fm14/webprogrampreliminary/Paper17555 (Accessed 8 August 2014)

Beachly, M. W., E. E. E. Hooft, D. R. Toomey, and G. P. Waite (2012), Upper crustal structure of Newberry Volcano from P-wave tomography and finite difference waveform modeling, J. Geophys. Res., 117(B10), B10311, doi:10.1029/2012JB009458.

Bohnhoff, M., J. Makris, D. Papanikolaou, and G. Stavrakakis (2001), Crustal investigation of the Hellenic subduction zone using wide aperture seismic data, Tectonophysics, 343(3), 239–262.

Bohnhoff, M., M. Rische, T. Meier, B. Endrun, D. Becker, H.-P. Harjes, and G. Stavrakakis (2004), CYC-NET: a temporary seismic network on the Cyclades (Aegean Sea, Greece), Seismological Research Letters, 75(3), 352–359.

Bohnhoff, M., M. Rische, T. Meier, D. Becker, G. Stavrakakis, and H.-P. Harjes (2006), Microseismic activity in the Hellenic Volcanic Arc, Greece, with emphasis on the seismotectonic setting of the Santorini–Amorgos zone, Tectonophysics, 423(1-4), 17–33, doi:10.1016/j.tecto.2006.03.024.

Cadoux, A., B. Scaillet, T. H. Druitt, and E. Deloule (2014), Magma Storage Conditions of Large Plinian Eruptions of Santorini Volcano (Greece), Journal of Petrology, 55(6), 1129–1171, doi:10.1093/petrology/egu021.

Carey, S., P. Nomikou, K. C. Bell, M. Lilley, J. Lupton, C. Roman, E. Stathopoulou, K. Bejelou, and R. Ballard (2013), CO2 degassing from hydrothermal vents at Kolumbo submarine volcano, Greece, and the accumulation of acidic crater water, Geology, 41(9), 1035–1038, doi:10.1130/G34286.1.

Caricchi, L. (2014), Frequency and magnitude of volcanic eruptions controlled by magma injection and buoyancy, Nature Geosci., 7(2), 1–5, doi:10.1038/ngeo2041.

Cooper, K. M., and A. J. R. Kent (2014), Rapid remobilization of magmatic crystals kept in cold storage, Nature, 506(7489), 480–483, doi:doi:10.1038/nature12991.

Defant, M. J., and M. S. Drummond (1990), Derivation of some modern arc magmas by melting of young subducted lithosphere, Nature.

Dimitriadis, I., C. Papazachos, D. Panagiotopoulos, P. Hatzidimitriou, M. Bohnhoff, M. Rische, and T. Meier (2010), P and S velocity structures of the Santorini–Coloumbo volcanic system (Aegean Sea, Greece) obtained by non-linear inversion of travel times and its tectonic implications, Journal of Volcanology and Geothermal Research, 195(1), 13–30, doi:10.1016/j.jvolgeores.2010.05.013.

Dimitriadis, I., E. Karagianni, D. Panagiotopoulos, C. Papazachos, P. Hatzidimitriou, M. Bohnhoff, M. Rische, and T. Meier (2009), Seismicity and active tectonics at Coloumbo Reef (Aegean Sea, Greece): Monitoring an active volcano at Santorini Volcanic Center using a temporary seismic network, Tectonophysics, 465(1-4), 136–149, doi:10.1016/j.tecto.2008.11.005.

Druitt, T. H., F. Costa, E. Deloule, M. Dungan, and B. Scaillet (2012), Decadal to monthly timescales of magma transfer and reservoir growth at a caldera volcano, Nature, 482(7383), 77–80, doi:10.1038/nature10706.

Druitt, T. H., L. Edwards, R. M. Mellors, D. M. Pyle, R. S. J. Sparks, M. Lanphere, M. Davies, and B. Barreiro (1999), Santorini volcano, Geological Society, London.

Dufek, J., and G. W. Bergantz (2005), Lower crustal magma genesis and preservation: a stochastic framework for the evaluation of basalt–crust interaction, Journal of Petrology, 46(11), 2167–2195.

Dufek, J., and O. Bachmann (2010), Quantum magmatism: Magmatic compositional gaps generated by melt-crystal dynamics, Geology, 38(8), 687–690, doi:10.1130/G30831.1.

Dunn, R. A., and D. R. Toomey (2001), Crack-induced seismic anisotropy in the oceanic crust across the East Pacific Rise (9°30’N), Earth Planet Sc Lett, 189, 9–17.

Durant, D. T., and D. R. Toomey (2009), Evidence and implications of crustal magmatism on the flanks of the East Pacific Rise, Earth Planet. Sci. Lett., 287(1), 130–136, doi:10.1016/j.epsl.2009.08.003.

Elsworth, D., G. Mattioli, J. Taron, B. Voight, and R. Herd (2008), Implications of Magma Transfer Between Multiple Reservoirs on Eruption Cycling, Science, 322(5899), 246–248, doi:10.1126/science.1161297.

Endrun, B., S. Lebedev, T. Meier, C. Tirel, and W. Friederich (2011), Complex layered deformation within the Aegean crust and mantle revealed by seismic anisotropy, Nature Geosci., 4(3), 203–207, doi:10.1038/ngeo1065.

Endrun, B., T. Meier, S. Lebedev, M. Bohnhoff, G. Stavrakakis, and H.-P. Harjes (2008), Svelocity structure and radial anisotropy in the Aegean region from surface wave dispersion, Geophys. J. Int., 174(2), 593–616, doi:10.1111/j.1365-246X.2008.03802.x.

Fabbro, G. N., T. H. Druitt, and S. Scaillet (2013), Evolution of the crustal magma plumbing system during the build-up to the 22-ka caldera-forming eruption of Santorini (Greece), Bulletin of Volcanology, 75(12), 1–22, doi:10.1007/s00445-013-0767-5.

Feuillet, N. (2013), The 2011–2012 unrest at Santorini rift: Stress interaction between active faulting and volcanism, Geophysical Research Letters, 40(14), 3532–3537.

Fliedner, M. M., and S. L. Klemperer (2000), Crustal structure transition from oceanic arc to continental arc, eastern Aleutian Islands and Alaska Peninsula, Earth Planet Sc Lett, 179(3-4), 567–579, doi:10.1016/S0012-821X(00)00142-4.

Friederich, W., A. Brüstle, L. Küperkoch, T. Meier, S. Lamara, and E. Working Group (2014), Focal mechanisms in the southern Aegean from temporary seismic networks – implications for the regional stress field and ongoing deformation processes, Solid Earth, 5(1), 275–297, doi:10.5194/se-5-275-2014.

Friederich, W., and T. Meier (2008), Temporary seismic broadband network acquired data on Hellenic subduction zone, EOS, Transactions American Geophysical Union, 89(40), 378–378.

García-Yeguas, A., I. Koulakov, J. M. Ibáñez, and A. Rietbrock (2012), High resolution 3D P wave velocity structure beneath Tenerife Island (Canary Islands, Spain) based on tomographic inversion of active-source data, J. Geophys. Res., 117(B9), B09309, doi:10.1029/2011JB008970.

Ghiorso, M. S., and R. O. Sack (1995), Chemical mass transfer in magmatic processes IV. A revised and internally consistent thermodynamic model for the interpolation and extrapolation of liquid-solid equilibria in magmatic systems at elevated temperatures and pressures, Contributions to Mineralogy and Petrology, 119(2-3), 197–212, doi:10.1007/BF00307281.

Hammond, W. C., and E. D. Humphreys (2000a), Upper mantle seismic wave attenuation: Effects of realistic partial melt distribution, Journal of Geophysical Research: Solid Earth (1978–2012), 105(B5), 10987–10999.

Hammond, W. C., and E. D. Humphreys (2000b), Upper mantle seismic wave velocity: Effects of realistic partial melt geometries, J. Geophys. Res., 105(B5), 10975–10986, doi:10.1029/2000JB900041.

Heath, B., D. R. Toomey, and E. E. E. Hooft (2014), New constraints on the magmatic system beneath Newberry Volcano from the analysis of active and passive source seismic data, and ambient noise.

Hildreth, W., and S. Moorbath (1988), Crustal contributions to arc magmatism in the Andes of Central Chile – Springer, Contributions to Mineralogy and Petrology.

Hooft, E. E. E., A. E. Teachout, D. R. Toomey, and W. S. D. Wilcock (2013), Seismic evidence of abundant flank magmatism at the Endeavour Segment, Juan de Fuca Ridge, Geochem. Geophys. Geosyst., 1–32.

Jackson, M. D., M. J. Cheadle, and M. P. Atherton (2003), Quantitative modeling of granitic melt generation and segregation in the continental crust, Journal of Geophysical Research: Solid Earth (1978–2012), 108(B7), doi:10.1029/2001JB001050.

Jull, M., and P. Keleman (2001), On the conditions for lower crustal convective instability, J. Geophys. Res., 106(B4), 6423–6446.

Kelemen, P. B. (1995), Genesis of high Mg# andesites and the continental crust, Contributions to Mineralogy and Petrology, 120(1), 1–19, doi:10.1007/BF00311004.

Kent, A. J. R., C. Darr, A. M. Koleszar, M. J. Salisbury, and K. M. Cooper (2010), Preferential eruption of andesitic magmas through recharge filtering, Nature Geosci., 3(9), 631–636, doi:10.1038/ngeo924.

Kilias, S. P. et al. (2013), New insights into hydrothermal vent processes in the unique shallow-submarine arc-volcano, Kolumbo (Santorini), Greece, Sci. Rep., 3, doi:10.1038/srep02421.

Lees, J. M. (1992), The magma system of Mount St. Helens: non-linear high-resolution P-wave tomography, Journal of Volcanology and Geothermal Research, 53(1-4), 103–116, doi:10.1016/0377-0273(92)90077-Q.

Ma, Y., and R. W. Clayton (2014), The crust and uppermost mantle structure of Southern Peru from ambient noise and earthquake surface wave analysis, Earth Planet Sc Lett, 395, 61–70, doi:10.1016/j.epsl.2014.03.013.

Meier, T., K. Dietrich, B. Stöckhert, and H. P. Harjes (2004), One-dimensional models of shear wave velocity for the eastern Mediterranean obtained from the inversion of Rayleigh wave phase velocities and tectonic implications, Geophys. J. Int., 156(1), 45–58, doi:10.1111/j.1365-246X.2004.02121.x.

Morgan, J., M. Warner, R. Bell, J. Ashley, D. Barnes, R. Little, K. Roele, and C. Jones (2013), Next-generation seismic experiments: wide-angle, multi-azimuth, three-dimensional, full-waveform inversion, Geophysical Journal of the Royal Astronomical Society, doi:10.1093/gji/ggt345.

Mortazavi, M., and R. S. J. Sparks (2004), Origin of rhyolite and rhyodacite lavas and associated mafic inclusions of Cape Akrotiri, Santorini: the role of wet basalt in generating calcalkaline silicic magmas, Contributions to Mineralogy and Petrology, 146(4), 397–413, doi:10.1007/s00410-003-0508-4.

Müntener, O., P. B. Kelemen, and T. L. Grove (2001), The role of H2O during crystallization of primitive arc magmas under uppermost mantle conditions and genesis of igneous pyroxenites: an experimental study, Contributions to Mineralogy and Petrology, 141(6), 643–658, doi:10.1007/s004100100266.

Newman, A. V. et al. (2012), Recent geodetic unrest at Santorini Caldera, Greece, Geophys. Res. Lett., 39(6), L06309, doi:10.1029/2012GL051286.

Nomikou, P., S. Carey, D. Papanikolaou, K. Croff Bell, D. Sakellariou, M. Alexandri, and K. Bejelou (2012), Submarine volcanoes of the Kolumbo volcanic zone NE of Santorini Caldera, Greece, Global and Planetary Change, 90-91(C), 135–151, doi:10.1016/j.gloplacha.2012.01.001.

Papoutsis, I., X. Papanikolaou, M. Floyd, K. H. Ji, C. Kontoes, D. Paradissis, and V. Zacharis (2012), Mapping inflation at Santorini volcano, Greece, using GPS and InSAR, Geophys. Res. Lett., doi:10.1029/2012GL054137.

Parks, M. M. et al. (2012), Evolution of Santorini Volcano dominated by episodic and rapid fluxes of melt from depth, Nature Geosci., 5(10), 749–754, doi:10.1038/ngeo1562.

Paulatto, M., C. Annen, T. J. Henstock, E. Kiddle, T. A. Minshull, R. S. J. Sparks, and B. Voight (2012), Magma chamber properties from integrated seismic tomography and thermal modeling at Montserrat, Geochem. Geophys. Geosyst., 13(1), n/a–n/a, doi:10.1029/2011GC003892.

Pearce, J. A., and D. W. Peate (1995), Tectonic implications of the composition of volcanic arc magmas, Annual Review of Earth and Planetary Sciences, 23, 251–286.

Pichavant, M. (2002), Physical conditions, structure, and dynamics of a zoned magma chamber: Mount Pelée (Martinique, Lesser Antilles Arc), J. Geophys. Res., 107(B5), 2093, doi:10.1029/2001JB000315.

Piper, D. J. W., and C. Perissoratis (2003), Quaternary neotectonics of the South Aegean arc, Marine Geology, 198(3-4), 259–288, doi:10.1016/S0025-3227(03)00118-X.

Pratt, R. G. (1999), Seismic waveform inversion in the frequency domain, Part 1: Theory and verification in a physical scale model, Geophysics, 64(3), 888–901.

Pyle, D. M., and J. R. Elliott (2006), Quantitative morphology, recent evolution, and future activity of the Kameni Islands volcano, Santorini, Greece, Geosphere, 2(5), 253, doi:10.1130/GES00028.1.

Reverso, T., J. Vandemeulebrouck, F. Jouanne, V. Pinel, T. Villemin, E. Sturkell, and P. Bascou (2014), A two-magma chamber model as a source of deformation at Grímsvötn Volcano, Iceland, J Geophys Res-Sol Ea, 119(6), 4666–4683, doi:10.1002/2013JB010569.

Rudnick, R. L. (1995), Making continental crust, Nature, 378(6557), 571–577.

Shen, Y., D. W. Forsyth, J. Conder, and L. M. Dorman (1997), Investigation of microearthquake activity following an intraplate teleseismic swarm on the west flank of the southern East Pacific Rise, J. Geophys. Res., 102, 459–475.

Sisson, T. W., and T. L. Grove (1993), Temperatures and H2O contents of low-MgO high-alumina basalts, Contributions to Mineralogy and Petrology, 113, 167–184.

Solano, J. M. S., M. D. Jackson, R. S. J. Sparks, J. D. Blundy, and C. Annen (2012), Melt Segregation in Deep Crustal Hot Zones: a Mechanism for Chemical Differentiation, Crustal Assimilation and the Formation of Evolved Magmas, Journal of Petrology, 53(10), 1999–2026, doi:10.1093/petrology/egs041.

Spera, F. J., and W. A. Bohrson (2001), Energy-constrained open-system magmatic processes I: general model and energy-constrained assimilation and fractional crystallization (EC-AFC) formulation, Journal of Petrology, 42(5), 999–1018, doi:10.1093/petrology/42.5.999.

Tarasewicz, J., B. Brandsdóttir, R. S. White, M. Hensch, and B. Thorbjarnardóttir (2012), Using microearthquakes to track repeated magma intrusions beneath the Eyjafjallajökull stratovolcano, Iceland, J. Geophys. Res., 117(null), B00C06, doi:10.1029/2011JB008751.

Tatsumi, Y., and T. Kogiso (2003), The subduction factory: its role in the evolution of the Earth’s crust and mantle, Geological Society, London, Special Publications, 219(1), 55–80, doi:10.1144/GSL.SP.2003.219.01.03.

Tatsumi, Y., H. Shukuno, K. Tani, N. Takahashi, S. Kodaira, and T. Kogiso (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, Journal of Geophysical Research: Solid Earth (1978–2012), 113(B2).

Toomey, D. R., S. C. Solomon, and G. M. Purdy (1994), Tomographic imaging of the shallow crustal structure of the East Pacific Rise at 9 30’N, J. Geophys. Res., 99, 24–24.

Virieux, J., and S. Operto (2009), An overview of full-waveform inversion in exploration geophysics, Geophysics, 74(6), WCC1–WCC26, doi:10.1190/1.3238367.

Waite, G. P., and S. C. Moran (2009), VP Structure of Mount St. Helens, Washington, USA, imaged with local earthquake tomography, Journal of Volcanology and Geothermal Research, 182(1-2), 113–122, doi:10.1016/j.jvolgeores.2009.02.009.

Warner, M. et al. (2013), Anisotropic 3D full-waveform inversion, Geophysics, 78(2), R59–R80, doi:10.1190/geo2012-0338.1.

Weekly, R. T., W. S. D. Wilcock, E. E. E. Hooft, D. R. Toomey, P. R. McGill, and D. S. Stakes (2013), Termination of a 6 year ridge-spreading event observed using a seafloor seismic network on the Endeavour Segment, Juan de Fuca Ridge, Geochem. Geophys. Geosyst., 14, doi:10.1002/ggge.20105.

Weekly, R. T., W. S. Wilcock, D. R. Toomey, E. E. Hooft, and A. E. Wells (2011), Upper- to Mid-Crustal Seismic Velocity Structure of the Endeavour Segment: Implications for Crustal Construction, AGU Fall Meeting Abstracts, -1, 2552.

Weekly, R. T., W. S. Wilcock, D. R. Toomey, E. E. Hooft, and E. Kim (2014), Upper crustal seismic structure of the Endeavour segment, Juan de Fuca Ridge from traveltime tomography: Implications for oceanic crustal accretion, Geochem. Geophys. Geosyst., 15(4), 1296–1315.

Wilcock, W. S. D., E. E. E. Hooft, D. R. Toomey, P. R. McGill, A. H. Barclay, D. S. Stakes, and T. M. Ramirez (2009), The role of magma injection in localizing black-smoker activity, Nature Geosci., 2(7), 509–513, doi:10.1038/ngeo550.

Zellmer, G. F., C. J. Hawkesworth, R. Sparks, L. E. THOMAS, C. L. HARFORD, T. S. BREWER, and S. C. LOUGHLIN (2003), Geochemical evolution of the Soufriere Hills volcano, Montserrat, Lesser Antilles volcanic arc, Journal of Petrology, 44(8), 1349–1374.

11