Magma systems of supervolcanoes
What are the structure and location of the magma systems that can feed supervolcano eruptions?
Summary
Why it is open
Supereruptions discharge more than 10¹⁵ kg of magma, about 450 cubic kilometres (km³) of magma or more than about 1,000 km³ of pumice and ash, in a single eruption [12]. Colin Wilson and colleagues reviewed 13 supereruptions from the last 2.6 million years in 2021 [12]. They find no single explanation for their magma systems and eruption histories [12].
The picture of the source has changed from a single long-lived chamber of liquid magma to magmatic systems that extend through the crust and are mostly crystal-rich [8, 12]. Katharine Cashman, Stephen Sparks and Jonathan Blundy highlight the tension between slow deep melt generation and rapid shallower accumulation [8].
Wilson and colleagues name what is missing: imaging able to resolve small bodies of eruptible magma, an understanding of the tipping points that turn accumulation into eruption, and a way to tell ordinary unrest from the signals that precede an eruption [12]. Their review finds no definitive way to identify unrest that precedes a supereruption [12].
What is known
Crystal mush (2004–2021). In 2004 Olivier Bachmann and George Bergantz argued that the largest accumulations of silica-rich (rhyolitic) melt in the upper crust sit in crystal mushes, frameworks of crystals with melt between them [1, 12]. Once a magma is about 50% crystals, convection is hampered and melt can separate by settling and compaction; they estimated that extracting more than 500 km³ of crystal-poor rhyolite takes 10,000 to 100,000 years, within mush lifetimes of more than 100,000 years, and that granitic rocks are the leftovers [1]. How granite bodies make room in the crust is another open problem, the granite space problem. The mush model is now widely adopted but "not universally accepted" [12].
Past supereruptions (2021 synthesis). In all 13 cases the crystal-rich roots deeper than 10 km lasted tens to hundreds of thousands of years or more, while the erupted magma was assembled at 4 to 10 km depth over shorter times, sometimes within centuries [12]. Some eruptions appear to have drawn on a single body (the Bishop Tuff of Long Valley, Aso-4 in Japan); others tapped several bodies together or in sequence (the Youngest Toba Tuff, Yellowstone's Huckleberry Ridge Tuff) [12].
Crystal clocks (2015–2021). Uranium–thorium and uranium–lead ages of zircon crystals record long histories, and diffusion across zones in crystals records short events [12]. Zircons from Toba's 74,000-year-old Youngest Toba Tuff, about 2,800 km³ of deposits, grew episodically over more than 100,000 to more than 500,000 years, and repeated disturbances over more than 400,000 years did not lead to eruption until then [7]. The melt-rich body of Taupō's Oruanui eruption, about 530 km³ of magma erupted 25,400 years ago, is inferred to have formed in less than 600 years and mostly in less than 200, whereas the Bishop body accumulated over tens of thousands of years [12]. Smaller eruptions also show short timescales: before Taupō's eruptions of 2,150 to 1,700 years ago, including the 232 CE eruption of about 105 km³, heating began less than 120 years ahead and melt accumulated within decades [6]; at Yellowstone a near-solid intrusion was reheated in about ten months or less before about 3 km³ of rhyolite lava erupted [5]. In 2017 Allison Rubin and colleagues found that zircons from a Taupō Volcanic Zone eruption had resided in the system for 1,000 to 100,000 years but had spent only years to centuries above 650 to 750 °C [10].
Imaging (2014–2025). Beneath Toba, K. Jaxybulatov and colleagues interpreted fine layering seen with ambient-noise seismic tomography as many partially molten sills below 7 km [3]. Beneath Yellowstone, Hsin-Hua Huang and colleagues imaged a basaltic body in the lower crust of 46,000 km³, about 4.5 times the upper-crustal reservoir, with about 2% melt [4]; Wilson and colleagues summarise the upper reservoir as holding 5 to 15% melt [12], while Ross Maguire and colleagues, assuming broadly distributed melt in a mush, estimated 16 to 20% at 3 to 8 km depth in 2022 [13]. In 2025 Chenglong Duan and colleagues reported a sharp reflective cap to this reservoir about 3.8 km beneath the northeastern caldera, with supercritical fluid and magma in its pores; they estimate a bubble fraction below typical pre-eruptive values and infer a stable state in which bubbles escape efficiently [15]. Beneath Taupō, earthquakes and ground deformation during unrest in 2019 outlined what Finnigan Illsley-Kemp and colleagues infer to be a reservoir of at least 250 km³ with more than 20 to 30% melt [11]. In 2021 Wilson and colleagues noted that no imaged supereruptive system shows the shallow, melt-dominant bodies of tens to hundreds of km³ thought to feed supereruptions, and that typical resolutions of 5 to 10 km cannot reveal bodies smaller than 10 km³ [12].
What forecasting can say. Unrest, meaning heightened earthquakes, ground deformation or gas release, is thought to precede eruptions at large silicic volcanoes, but most unrest ends without one [12]. At Campi Flegrei, a large caldera in southern Italy, three episodes of uplift since 1950 had raised the centre by about 3 m without eruption by 2017; Christopher Kilburn and colleagues modelled them as a long-term accumulation of stress and concluded that the caldera is evolving towards conditions more favourable to eruption [9]. Reviewing magmatic eruptions in general, Valerio Acocella and colleagues find forecasting difficult to achieve; eruptions in which magma opens a new path give stronger signals, often months ahead, but commonly at volcanoes whose behaviour before eruption is poorly known [14]. Wilson and colleagues suggest that once magma starts to ascend, warning times may be only days to months, and write that a past supereruption does not constrain the size of a volcano's future activity [12].
Main approaches
Mush with melt extraction. Eruptible magma is melt extracted from a long-lived mush into a short-lived, melt-dominant body [1, 12]. Strongest evidence: the model accounts for abrupt compositional gaps in deposits, zircon ages spanning up to 200,000 to 300,000 years and zones of low seismic velocity under calderas [1]. Main difficulty: modelled trace-element abundances do not match some erupted deposits, although alternative explanations now exist [12]; and Oruanui's melt body formed in centuries, far faster than the 10,000 to 100,000 years estimated for extracting a body of that size by settling and compaction, so extraction must at times be sped up, for example by an input of heat or volatiles [1, 12].
Cold storage, rapid reheating. Magma spends most of its life close to or below its solidus, the temperature below which it is fully solid, and is remobilised shortly before eruption [12]. Strongest evidence: crystal thermal histories [10] and short reheating times at Yellowstone and Taupō [5, 6]. Main difficulty: once volatiles have escaped, reheating must reach the higher dry melting temperature, and the 5 to 15% melt imaged beneath systems such as Toba and Yellowstone fits warm storage better [12].
Warm, incremental reservoirs. Large reservoirs are built over long periods from many sills and keep up to a few tens of percent melt [3, 12]. Strongest evidence: the Toba sill complex [3], and melt fractions of 16 to 20% at Yellowstone [13] and over 20% at Taupō [11]. Main difficulty: such reservoirs are too crystal-rich to erupt without melt separation or remobilisation, and converting seismic velocity to melt fraction is uncertain [12].
What triggers eruption. In the 2014 model of Luca Caricchi and colleagues, pressure from injected magma triggers small, frequent eruptions, while magma buoyancy is the key to triggering supereruptions [2]. Strongest evidence: the model reproduces the link between erupted volume and residence time, and between volume and caldera size [2]. Main difficulty: many models treat the magma body as isolated from the surrounding stress field and propose single causes, while evidence is growing that tectonic forces help control where magma accumulates and how it is released [12].
Open sub-questions
- Hidden lenses. Can dense seismic or electrical surveys at Yellowstone or Taupō resolve melt-rich bodies of 1 to 10 km³, below the 5 to 10 km resolution discussed in the 2021 review [11, 12, 13]?
- Melt fraction. Do the 5 to 15% and 16 to 20% melt estimates for Yellowstone's upper reservoir differ because of data or because of how melt is assumed to be distributed [12, 13]?
- The cap. Does the reflective cap at 3.8 km extend beneath the whole caldera, and do repeat surveys show its bubble content changing [15]?
- Toba's sills. Do zircon ages in post-caldera lavas show that the sills imaged below 7 km are the reservoir disturbed repeatedly before the 74,000-year eruption [3, 7]?
- Thermal history in one deposit. Applied to Oruanui crystals, does the zircon method of Rubin and colleagues give cold or warm storage [10, 12]?
- Buoyancy test. Do the volumes and storage conditions of the 13 Quaternary supereruptions fall where Caricchi and colleagues' model predicts buoyancy, not injection, as the trigger [2, 12]?
Work on this in Papex
If you have a testable answer to one of the sub-questions above, you can draft it as a private hypothesis in Papex, or prepare the draft from Claude or ChatGPT with the Papex connector. You review the exact version, sign it and decide whether to publish it; a public record that names this problem is listed on this page. A Papex record shows who signed which version and when. It is not peer review and does not mean the hypothesis is correct.
Key references
- Bachmann, O. & Bergantz, G. W. (2004). On the origin of crystal-poor rhyolites: extracted from batholithic crystal mushes. Journal of Petrology 45(8), 1565–1582. 10.1093/petrology/egh019 · Checked 15 Oct 2026
- Caricchi, L., Annen, C., Blundy, J., Simpson, G. & Pinel, V. (2014). Frequency and magnitude of volcanic eruptions controlled by magma injection and buoyancy. Nature Geoscience 7(2), 126–130. 10.1038/ngeo2041 · Checked 15 Oct 2026
- Jaxybulatov, K., Shapiro, N. M., Koulakov, I., Mordret, A., Landès, M. & Sens-Schönfelder, C. (2014). A large magmatic sill complex beneath the Toba caldera. Science 346(6209), 617–619. 10.1126/science.1258582 · Checked 15 Oct 2026
- Huang, H.-H., Lin, F.-C., Schmandt, B., Farrell, J., Smith, R. B. & Tsai, V. C. (2015). The Yellowstone magmatic system from the mantle plume to the upper crust. Science 348(6236), 773–776. 10.1126/science.aaa5648 · Checked 15 Oct 2026
- Till, C. B., Vazquez, J. A. & Boyce, J. W. (2015). Months between rejuvenation and volcanic eruption at Yellowstone caldera, Wyoming. Geology 43(8), 695–698. 10.1130/G36862.1 · Checked 15 Oct 2026
- Barker, S. J., Wilson, C. J. N., Morgan, D. J. & Rowland, J. V. (2016). Rapid priming, accumulation, and recharge of magma driving recent eruptions at a hyperactive caldera volcano. Geology 44(4), 323–326. 10.1130/G37382.1 · Checked 15 Oct 2026
- Reid, M. R. & Vazquez, J. A. (2017). Fitful and protracted magma assembly leading to a giant eruption, Youngest Toba Tuff, Indonesia. Geochemistry, Geophysics, Geosystems 18(1), 156–177. 10.1002/2016GC006641 · Checked 15 Oct 2026
- Cashman, K. V., Sparks, R. S. J. & Blundy, J. D. (2017). Vertically extensive and unstable magmatic systems: a unified view of igneous processes. Science 355(6331), eaag3055. 10.1126/science.aag3055 · Checked 15 Oct 2026
- Kilburn, C. R. J., De Natale, G. & Carlino, S. (2017). Progressive approach to eruption at Campi Flegrei caldera in southern Italy. Nature Communications 8, 15312. 10.1038/ncomms15312 · Checked 15 Oct 2026
- Rubin, A. E., Cooper, K. M., Till, C. B., Kent, A. J. R., Costa, F., Bose, M. et al. (2017). Rapid cooling and cold storage in a silicic magma reservoir recorded in individual crystals. Science 356(6343), 1154–1156. 10.1126/science.aam8720 · Checked 15 Oct 2026
- Illsley-Kemp, F., Barker, S. J., Wilson, C. J. N., Chamberlain, C. J., Hreinsdóttir, S., Ellis, S. et al. (2021). Volcanic unrest at Taupō volcano in 2019: causes, mechanisms and implications. Geochemistry, Geophysics, Geosystems 22(6), e2021GC009803. 10.1029/2021GC009803 · Checked 15 Oct 2026
- Wilson, C. J. N., Cooper, G. F., Chamberlain, K. J., Barker, S. J., Myers, M. L., Illsley-Kemp, F. & Farrell, J. (2021). No single model for supersized eruptions and their magma bodies. Nature Reviews Earth & Environment 2(9), 610–627. 10.1038/s43017-021-00191-7 · Checked 15 Oct 2026
- Maguire, R., Schmandt, B., Li, J., Jiang, C., Li, G., Wilgus, J. & Chen, M. (2022). Magma accumulation at depths of prior rhyolite storage beneath Yellowstone Caldera. Science 378(6623), 1001–1004. 10.1126/science.ade0347 · Checked 15 Oct 2026
- Acocella, V., Ripepe, M., Rivalta, E., Peltier, A., Galetto, F. & Joseph, E. (2024). Towards scientific forecasting of magmatic eruptions. Nature Reviews Earth & Environment 5(1), 5–22 (published online 2023). 10.1038/s43017-023-00492-z · Checked 15 Oct 2026
- Duan, C., Song, W., Schmandt, B., Farrell, J., Lumley, D., Fischer, T., Worthington, L. L. & Lin, F.-C. (2025). A sharp volatile-rich cap to the Yellowstone magmatic system. Nature 640(8060), 962–966. 10.1038/s41586-025-08775-9 · Checked 15 Oct 2026
Source
Open, as listed by Wikipedia · Unsolved problems in geoscience. Checked 08 Oct 2026.
The status is the source’s. Papex does not decide whether a problem is solved.
Records on this problem
No public record addresses this problem yet.