Eruptible magma.

For those who remember the classic disaster-docudrama Supervolcano (BBC-Discovery Channel 2005), the adjective “eruptible” is likely to ring a bell. Set in the near future and generally quite scientifically literate, the movie envisions volcanologists trying desperately to evaluate the near-term threat posed by the giant Yellowstone magmatic system. At the heart of their mission, and of the drama, is their quest to determine the mass of “eruptible magma” under Yellowstone: that is, how much material that is capable of erupting is down there right now? Fortunately, the scientists have at their disposal VIRGIL, a seismic tomography system that is based upon but greatly upgraded from existing systems. VIRGIL can reliably detect eruptible magma but, unfortunately, it is just coming on-line, and when they discover that there is in fact an enormous reservoir beneath them that can erupt, the scientists lack context. Given a gigantic mass of eruptible magma, what if any sort of eruption—or eruptions—is likely to occur, how soon will it begin, how long will it last, how explosive will it be, and how should scientists and the government respond? Supervolcano doesn’t turn out well, despite the heroism of the scientists. In PNAS, the search for context is at the heart of research presented by Barboni et al. (1). Geophysical techniques, as represented by VIRGIL and its less-sophisticated present-day tomographic counterparts and by magnetotelluric and gravity surveys, provide a real-time glimpse of where magma is—or may be—stored within the crust (2). Results to date are intriguing but frustrating. Surveys suggest the presence of molten material beneath a growing number of volcanoes. In fact, gigantic volumes of the crust beneath Yellowstone and the Washington Cascade volcanoes, St. Helens, Rainier, and Adams, appear to contain melt (3, 4). However, these zones are interpreted to mostly contain very small melt fractions; that is, they are largely or entirely ineruptible. In fact, no large volumes of eruptible magma have yet been identified anywhere (2): either they don’t exist or current resolution is insufficient to reveal them. Increasing scrutiny and improved resolution are called for to confidently locate or rule out giant eruptible bodies. Their absence would not be surprising to investigators who infer that giant eruptible magma assembles and erupts on short timescales (5, 6); because “supereruptions” are relatively rare (on the order of 10 y), it would not be surprising if no eruptible magma bodies of this scale exist today if they are assembled in 10–10 y or less. Many smaller bodies that feed volcanoes, like those within the Soufrière Volcanic Center (1), should exist today, even if they assemble rapidly, given that such eruptions occur many times per year on Earth, but they might be much more difficult to detect. Even when, and if, VIRGIL-like reliable, real-time detection of eruptible magma becomes possible, context provided by studies like that of Barboni et al. (1) are critical for understanding volcano behavior in general and hazards in particular. Petrologic studies, broadly defined—that is, the study of rocks and the evidence they provide of the history of magmas—are necessary to enable us to evaluate what threatening volcanoes and detected magma bodies beneath may do: erupt or not? If so, how soon? What sort of eruption? Should authorities go into serious response mode if eruptible magma is detected, or is this simply the expected state of the volcano in question, and should monitoring simply continue? Eruptible magma includes crystal-poor, melt-rich material and crystal mush [barely eruptible, up to 50– 60% crystals (7, 8)] (Fig. 1A). The distribution, in space and time, of rheologically distinct materials with differing melt fractions in magma systems (Fig. 1 B and C) remains controversial and is probably highly variable. Temperature fluctuations are inevitable, but note that, according to Barboni et al. (1), they were more muted than those depicted in Fig. 1C (9) for most of the history of the Soufrière Volcanic Center, where Barboni et al. (1) infer that magma remained within the eruptible window for very long periods of time. Key questions to be addressed if we are to understand magma systems and the eruptions that they produce, most ofwhich are exemplifiedby thework of Barboni et al. (1), include: (i) How long do volcanoes and magmatic systems remain active? (ii) For how much of their lifetimes is melt present within these systems? (iii) For howmuch of their lifetimes is eruptible magma present? (iv) How are