Assessing pre-eruptive controls on explosive and non-explosive volcanic eruptions

What: This Gateway research project will investigate the extent to which pre-eruptive conditions can influence eruption style in silicic volcanic systems. This laboratory-based project will take advantage of a high-spatial-resolution Raman microscope at Trinity University to determine pre-eruptive water contents in quartz-hosted melt inclusions in the Taylor Creek Rhyolite, NM. Our measured water contents for explosive and nonexplosive rhyolite will be used to test whether eruption style was predetermined by the water content of magma stored in the subvolcanic environment.

When: June 22-July 26, 2025

Where: Trinity University, San Antonio, TX and the GSA Annual Meeting in San Antonio, TX (October, 2025)

Who: Four students, a peer mentor, and project director Dr. Kurt Knesel (Trinity University).

Prerequisites:  Because this experience is a Gateway Project for rising sophomores, there are no specific coursework prerequisites.

Expectations and Obligations:
1. Participation in all project-related work during the summer (5 weeks)
2. Submission of an abstract (individual or in groups) and presentation of a paper (poster or talk) at both Trinity’s summer research symposium and the Geological Society of America Meeting in San Antonio in October, 2025 (all expenses covered)
.

PROJECT DESCRIPTION

Goals and Significance

Deciphering the factors governing eruptions of viscous, silica-rich magma, which are among the largest and most destructive on Earth, is a fundamental problem in volcanology. A general consensus has emerged relating eruptive style to the nature of magma ascent from shallow storage reservoirs (Gonnermann and Manga, 2007; 2012). Fast magma ascent inhibits volatile loss, leading to explosive fragmentation of magma due to gas overpressure. In contrast, slow ascent favors gas escape and non-explosive effusion of lava. Despite the conceptual appeal of this model, a clear understanding of what dictates the coupled rise and degassing of viscous magma remains elusive (Cassidy et al., 2018).

A recent survey of magma-chamber conditions that were likely prevalent prior to both explosive and non-explosive eruptions sheds light on this long-standing problem. Based on analysis of published petrologic data, primarily from subduction-related volcanoes, Popa et al. (2021a) found that eruptive behavior may be predetermined primarily by the state of water in subvolcanic storage regions (Fig.1). Viscous magmas appear to erupt explosively for dissolved water contents of ~4-5.5 wt. %, but below and (perhaps surprisingly) above this window magmas are prone to erupt non-explosively. This analysis provides renewed support for early eruption models based on pre-eruptive volatile abundances (e.g., Fink, 1983), and provides a critical link to the conduit processes involved in magma outgassing and fragmentation.

An important caveat to this scheme, however, is that it rests on indirect estimates for magmatic water in shallow storage reservoirs, which is often difficult to determine (as explained below). The key to forecasting eruption style may therefore lie in our ability to reliably and routinely measure pre-eruptive water directly. Recent advances in micro-Raman spectroscopy make this possible (e.g., Cassidy et al., 2016). The overarching goal of the research proposed here is to employ high-spatial resolution Raman microscopy to test whether volcanic eruption style is reliably determined by pre-eruptive water. If confirmed, better estimates of magmatic water at active volcanoes will improve our ability to forecast eruptive hazards.

Current State of Research

Why water matters. Most explosive eruptions owe their explosive character to the presence of gases dissolved within rising viscous magma, the most abundance of which is water. The solubility of volcanic gases in magma decreases with decreasing pressure (Fig. 2). As magma rises towards the surface, a point is reached where the magma becomes volatile saturated. With further rise, gas exsolves or separates from the magma through the nucleation and growth of bubbles. If the magma is highly viscous, resistance to bubble growth can lead to excess gas pressure. When the bubble pressure exceeds the effective strength of the melt film forming the walls of bubbles, the magma fractures like a brittle solid (Gonnermann, 2015), resulting in the formation of a violently expanding gas-pyroclast mixture.

Figure 1: Correlation of eruptive styles of viscous, intermediate-to-high-silica magma with dissolved water (simplified after Popa et al., 2021a). Magmas with low-to-moderate crystallinity (<40 vol. %) appear to erupt explosively (blue symbols) for dissolved water contents of ~4-5.5 wt. %, but below and above this window magmas are prone to erupt effusively (red symbols).

Figure 2: Potential ascent paths for magmas of low (A1), intermediate (A2), and high (A3) pre-eruptive water contents. Solubility curve (left panel) and & bubble nucleation, growth, and eruption sequences (right panels) modified from Parfitt & Wilson (2008) and Popa et al. (2021b). Grey field (in left panel) is typical pressure (depth) range for subvolcanic chambers from Huber et al. (2019); Blue field (left panel) is potential window of water contents for explosive eruptions from Popa et al (2021a). A potential ascent history for such water-rich but initially unsaturated magma is shown in the middle right panel (path A2-E2). As magma rises and decompresses, gas separates to form bubbles (B2-C2). With further ascent bubbles begin to crowd and gas pressure builds within bubbles (D2). Eventually gas pressure exceeds the strength of bubble walls, inducing fragmentation of magma into rapidly expanding gas suspension containing magma fragments (E2) and initiating an explosive volcanic eruption. Below this window, magma does not have enough water to induce fragmentation, arriving at the surface as bubbly magma that effuses non-explosively to form lava (path A1-C1). Above this window, the presence of exsolved volatiles (bubbles within the storage chamber) at the onset of ascent increases the potential for hydrodynamic interaction of bubbles, leading to enhanced permeability and early (deep) gas loss, limiting ascent rate, alleviating pressure build up, and favoring non-explosive effusive of magma (path A3-E3 right panel).

If pre-eruptive water contents are low, bubble formation begins at shallow depth (Fig. 2, point B1 along path A1-C1); this limits bubble volume fraction and restricts magma rise rate, such that bubble pressure may be insufficient to cause explosive fragmentation. Bubbly magma erupts non-explosively to form a lava flow. For higher dissolved water contents, ascending magma begins exsolution deeper (Fig. 2, point B2 along path A2-E2), increasing magma buoyancy, ascent rate, and bubble overpressure. Large stresses resulting from increased bubble pressure and melt deformation during rapid ascent can induce fragmentation (Fig. 2, point E2) fueling an explosive eruption. The analysis of Popa et al. (2021a) indicates that a transition between non- explosive and explosive activity may lie between 3.5 and 4 wt. % H2O (Fig. 1). However, there may be an upper limit to the explosive potential of water-rich magma. Above ~5.5 wt % water, viscous magma appears prone to non-explosive eruptions (Fig. 1). This upper threshold likely corresponds to the water saturation limit (Newmann and Lowenstern, 2002), such that water-rich effusive magmas may contain an exsolved volatile phase during storage in shallow crustal reservoirs (i.e., gas bubbles develop prior to ascent; point A3 in Fig. 2). The presence of exsolved volatiles at the onset of ascent can increase the potential for hydrodynamic interaction of bubbles, leading to enhanced permeability. An interconnected permeable network of bubbles provides a pathway for early gas escape and dissipation of overpressure (along path A3- E3). Increased permeability also increases the overpressure required to fragment magma (Mueller et al., 2008). Thus, the presence of exsolved water in water-supersaturated magma may in fact limit ascent rate and lower explosivity.

Why pre-eruptive water is hard to constrain. Due to its volatile nature and low solubility in silicate melt at the surface of the Earth, the direct measurement of pre-eruptive water is dependent on the presence of small glass inclusions trapped in a crystal host prior to eruption (Anderson et al., 2000; Metrich and Wallance, 2008). However, melt inclusions can experience diffusive loss of water during slow-cooling regimes experienced by effusive lavas (Thomas et al., 2006), making comparison between explosive and nonexplosive deposits difficult. To circumvent this problem, some studies, including Popa et al. (2021a), rely on indirect estimates based on volatile-dependent mineral-melt equilibria, coupled with measurement of (nonvolatile) elemental compositions of minerals and co-existing glass. The application of these hygrometers requires a close match between the natural samples and the experimental systems used to derive them, as well as reliable independent constrains for their temperature and pressure dependence. These criteria are often not fully realized for many magmatic systems.

Research Methods & Activities

Water analysis by micro-Raman spectroscopy. Fourier Transform Infrared (FTIR) spectroscopy has been the primary technique for routine measurement of water in melt inclusions for over forty years (e.g., Stolper, 1982; Ihinger, et al., 1994; Metrich and Wallance, 2008). Although robust, the technique has limitations. Melt inclusions must be exposed on both sides of a polished section or grain mount, limiting the number of inclusions that can analyzed in any individual crystal. Moreover, the spatial resolution of conventional FTIR is not generally suitable for measurement of small inclusions (<50 µm) or for detailed volatile profiles (in larger inclusions) required to assess volatile loss. And, while water analysis by secondary ion microprobe spectrometry (SIMS) affords the appropriate spatial resolution (Ihinger, et al., 1994; Metrich and Wallance, 2008), it also suffers from the requirement that inclusions must be exposed for analysis.

Confocal Raman spectroscopy has emerged as an alternative to FTIR and SIMS for melt inclusion analysis (Behrens et al. 2006; Di Muro et al., 2006). A major advantage of the technique is that inclusions can be analyzed without exposure to the surface, due to the control of the depth of field for the Raman signal afforded by the confocality of the instrument (Thomas et al., 2006). This feature, coupled with the high spatial resolution (1-2 microns), make it possible to analyze a larger number of inclusions within a host crystal (compared to FTIR and SIMS) and a larger number of spots within each inclusion (compared to FTIR). This approach is necessary to identify, quantify, and model gas loss from melt inclusions and thus derive reliable determination of dissolved water in both explosive and effusion products (e.g., Cassidy et al., 2016).

The measurement of dissolved water concentration in melt inclusions for this project will be performed with the new Horiba XploRA Plus Raman confocal microscope at Trinity.

Sample and analytical strategy. To broaden the scope of Popa et al. (2021a), this project represents a first step in a longer-term analytical campaign targeting intraplate, as well as subduction-related, volcanoes. Here we will focus on well-characterized rhyolite lava and associated tephra from the Taylor Creek Rhyolite (TCR) in the western US (Duffield and Du Bray, 1990; Duffield and Dalrymple 1990; Knesel et al., 1999; Knesel et al., 2007). The TCR consists of a group of 20 high-silica rhyolite lava domes and pyroclastic deposits in southwestern New Mexico. The rhyolite is moderately to highly porphyritic, generally containing 10–35% phenocrysts of quartz and sanidine in subequal proportions, with minor plagioclase (up to 5%) and trace hornblende and biotite.

Micro-Raman analyses will be conducted for melt inclusions in quartz phenocrysts in polished thin sections and grain mounts. Melt inclusions in TCR quartz are relatively abundant and range from about 1 µm up to 150 µm; most are 20-50 µm in diameter. In contrast, melt inclusions in TCR feldspar are rare, small (typically less than 15 µm), and usually strongly devitrified, and thus are not suitable for analysis. To assess volatile loss and accurately reconstruct pre-eruptive water contents, 5 to 15 spot analyses will be analyzed across individual melt inclusions selected to cover the range of inclusion sizes present, as well as distance from crystal rims. Spectroscopic measurements will follow analytical and data-treatment protocols outlined in Di Genova et al. (2017) and Gonzalez-Garcia et al. (2021).

Our measured water contents will then be compared with estimates derived from application of the plagioclase-melt hygrometer of Waters and Lange (2015) and evaluated with temperature-dependent solubility models (e.g., Newmann and Lowenstern, 2002; Liu et al., 2005). Temperature estimates required for the hygrometer will be determined using updated formulations for two-feldspar (Putirka, 2008) and amphibole thermometry (Ridolfi, 2021).

The TCR suite will provide an initial case study for the examination of potential differences in magmatic water between explosive and nonexplosive eruptions within a single volcanic system. All samples are in house at Trinity; no additional field work is required. The results from this pilot Keck project will then be used to guide future work on both intraplate and subduction related rhyolites. The first of these will target the Tweed Volcanic Complex (TVC) in eastern Australia (Ewart et al., 1981; Knesel et al., 2008; Knesel, 2020) and the Tatara-San Pedro Complex (TSPC) in central Chile (Dungan et al., 2001; Simpson, 2006), as additional intraplate and subduction endmembers, respectively.

Student Projects

The Raman research on the Taylor Creek Rhyolite will be subdivided along nonexplosive and explosive lines. Working in pairs, two students will study lava-dome samples, while two students will target their pyroclastic counterparts. In each pair, one of the students will work with early, crystal-rich eruptive products, while the other student will work with samples from a late-eruptive, crystal-poor unit. These distinctions will allow us to collectively evaluate the relative influence of water content, temperature, and crystallinity on eruptive style. All four students will participate in (1) the collection and processing of Raman spectra to derive water concentrations in melt inclusions, and (2) the use of the hygrometer and thermometers detailed above to estimate pre-eruptive conditions. The near-peer science mentor and I will work with each student pair to plan, acquire, process, and interpret the Raman data and to derive the hygrometry and thermometry estimates. Throughout the project we will strive to foster a collaborative research and active-learning environment where the students feel comfortable sharing and integrating their results and to work as a team.

Project Logistics

This Gateway project will tentatively run from June 15 to July 18. All work will be undertaken at Trinity University in San Antonio. Students will stay in Trinity housing on campus along with other summer research students.

Fall GSA meeting in San Antonio: October 19-22, 2025

In October, we will reconvene in San Antonio to present our work at the Geological Society of America conference.  All expenses for this trip are covered by the Keck Geology Consortium.

References

Anderson, A.T., Davis, A.M., and Lu, F. (2000) Evolution of the Bishop Tuff rhyolitic magma based on melt and magnetite inclusions, and zoned phenocrysts. Journal of Petrology 41, 449-473.

Behrens, H., Roux, J., Neuville, D.R., and Siemann, M. (2006) quantification of dissolved H2O in silicate glasses using confocal microRaman spectroscopy. Chemical Geology 229, 96-112.

Cassidy, M., Castro, J.M., Helo, C., Troll, V., Deegan, F.M., Muir, D., Neave, D.A., and Mueller, S.P. (2016) Volatile dilution during magma injections and implications for volcano explosivity. Geology 44, 1027–1030.

Cassidy, M., Manga, M., Cashman, K., and Bachmann, O. (2018) Controls on explosive-effusive volcanic eruption styles. Nature 9, 2839.

Di Genova, D., Sicola, S., Romano, C., Vona, A., Fanara., S., and Spina, L. (2017) Effects of iron and nanolites on Raman spectra of volcanic glasses: A reassessment of existing strategies to estimate the water content. Chemical Geology 475, 76-86.

Di Muro A., Villemant B., Montagnac G., Scaillet B., and Reynard B. (2006) quantification of water content and speciation in natural silicic glasses (phonolite, dacite, rhyolite) by confocal MicroRaman spectroscopy.

Geochimica et Cosmochimica Acta 70, 2868-2884.

Duffield, W.A., and Dalrymple, G.B., 1990. The Taylor Creek Rhyolite of New Mexico: a rapidly emplaced field of lava domes and flows. Bulletin of Volcanology 52, 475–487.

Duffield, W.A., du Bray, E.A., 1990. Temperature, size, and depth of the magma reservoir for the Taylor Creek Rhyolite, NM. American Mineralogist 75, 059–1070.

Dungan, M.A., Wulff, A., and Thompson, R. (2001) Eruptive stratigraphy of the Tatara-San Pedro Complex, 36S, Southern Volcanic Zone, Chilean Andes: Reconstruction method and implications for magma evolution at long-lived arc volcanic centers. Journal of Petrology 42, 555-626.

Ewart, A. (1981) The mineralogy and chemistry of anorogenic Tertiary silicic volcanics of S.E. Queensland and

N.E. New South Wales, Australia. Journal of Geophysical Research 86, 10242-10256.

Fink, J. (1983) Structure and emplacement of a rhyolitic obsidian flow: Little Glass Mountain, Medicine Lake Highland, northern California. Geological Society of America Bulletin 94, 362-380.

Fries, M., and Steele, A. (2018) Raman spectroscopy and confocal Raman imaging in mineralogy and petrography, in Toporski, J., Dieing, T., and Hollricher, O. eds., Confocal Raman Microscopy, Springer International Publishing.

Gonzalez-Garcia, D., Giordano, D., Allabar, A., Andrade, F.R.D., Polo, L.A., Janasi, V.A., Lucchetti, A.C.F., Hess, K.-U., De Campos, C.P., and Dingwell, D.B. (2021) Retrieving dissolved H2O content from micro- Raman spectroscopy on nanolitized silicic glasses: Application to the volcanic products of the Parana Magmatic Province, Brazil. Chemical Geology 567, 120058.

Gonnermann, H.M. (2015) Magma Fragmentation. Annual Reviews in Earth and Planetary Sciences 43: 431-458.

Gonnermann, H.M. and Manga, M. (2007) The fluid mechanics inside a volcano. Annual Review of Fluid Mechanics 39, 321-356.

Gonnermann, H.M., and Manga, M. (2012) Dynamics of magma ascent in the volcanic conduit, in Modelling volcanic processes: The physics and mathematics of volcanism. eds. Fagents, S.A., Gregg, K.P., and Rosaly M.C.L., Cambridge University Press.

Huber, C., Townsend, M., Degruyter, W., and Bachmann, O. (2019) Optimal depth of subvolcanic magma chamber growth controlled by volatiles and crust rheology. Nature Geosciences 12, 762–768.

Ihinger, P.D., Hervig, R.L., and McMillan, P.F. (1994) Analytical methods for volatiles in glasses. Reviews in Mineralogy, 30, 67-121.

Knesel, K. (2020) Using microlites and bubbles in obsidian to assess factors governing eruption of rhyolitic magma. Keck Geology Consortium Proceedings Volume 33.

Knesel K.M., Davidson, J.P. and Duffield, W.A. (1999) Evolution of silicic magma through assimilation and subsequent recharge: evidence from Sr isotopes in sanidine phenocrysts, Taylor Creek Rhyolite, NM. Journal of Petrology 40, 773-786.

Knesel K.M. and Duffield, W.A., (2007) Gradients in silicic bodies caused by rapid inputs from above and below rather than protracted chamber differentiation. Journal of Volcanology and Geothermal Research 167, 181- 197.

Knesel K.M., Cohen, B.E., Vasconcelos, P.M., and Thiede, DS. (2008) Rapid change in the drift of the Australian plate records collision with Ontong Java plateau, Nature 454, 754-757.

Metrich, N., and Wallance, P.J. (2008) Volatile abundances in basaltic magmas and their degassing paths tracked by melt inclusions. Reviews in Mineralogy & Geochemistry 69, 363-402.

Mueller, S., Scheu, B., Spieler, O., and Dingwell, D. (2008) Permeability control on magma fragmentation. Geology 36, 399-402.

Newman, S., and Lowenstern, J.B. (2002) VolatileCalc: a silicate melt-H2O-CO2 solution model written in Visual Basic Excel. Computers in Geosciences 2, 597-604.

Parfitt, E.A., and Wilson, L. (2008) Fundamentals of physical volcanology. Blackwell Publishing, pp. 230. Putirka, K.D. (2008) Thermometers and barometers for volcanic systems, Reviews in Mineralogy & Geochemistry 69, 61-120.

Popa, R.-G., Bachmann, O., and Huber, C. (2021a) Explosive or effusive style of volcanic eruption determined by magma storage conditions. Nature Geosciences, 14, 781-786.

Popa, R.-G., Tollan, P., Bachmann, O., Shenker, V., Ellis, B., and Allaz, J.M. (2021b) Water exsolution in the magma chamber favors effusive eruptions: Application of Cl-F partitioning behavior at the Nisyros-Yali volcanic area. Chemical Geology 570, 120170.

Ridolfi, F. (2021) Amp-TB2: An updated model for calcic amphibole thermobarometry, Minerals 11, 324.

Simpson, A. (2006) Investigations into the crustal-processing history of magmas at the Tatara-San Pedro volcanic complex, Chilean Andes. Unpublished PhD thesis, University of Queensland.

Stolper, E. (1982) The speciation of water in silicate melts. Geochimica et Cosmochimica Acta 46, 2609-2620.

Thomas, R., Kamenetsky, V.S., and Davidson, P. (2006) Laser Raman spectroscopic measurements of water in unexposed glass inclusions. American Mineralogist 91, 467-470.

Waters, L.E, and Lange, R.A. (2015) An updated calibration of the plagioclase-liquid hygrometer-thermometer applicable to basalts through to rhyolites, American Mineralogist 100, 2172-2184.