ARE SHALLOW MAGMA CHAMBERS A THING?

Much of igneous petrology is based on the concept of a magma chamber—a large ellipsoidal cavity filled with magma. Such chambers are assumed to provide the crucible for petrologic processes such as differentiation by crystal settling, sidewall crystallization, and upward movement of magma by stoping. They leak to the surface and produce volcanoes, or vent catastrophically and form calderas, or freeze in place and produce plutons that are relics of these crucibles. Surely there is no reason to question the existence of shallow magma chambers, the cornerstones of petrology. Right?

I’m glad you asked; here are three. (1) The archetypal magma chamber, with its top 5 km deep, should be an easy target for seismic studies to find, and yet decades of such studies have failed to locate any. In cross-sections showing a seismically located “magma chamber”, the body is typically an area where P-wave velocities are reduced enough that there might be as much as a few percent liquid—hardly a body of mobile, eruptible magma. (2) It is now clear that many/most/all plutons, once assumed to be frozen magma chambers, were assembled in small increments. (3) Heat loss from the roof of a shallow magma body by conduction, the slowest mode of heat transfer, is so fast that building and sustaining a shallow magma chamber requires bringing in massive amounts of magma, the only source of heat available—enough to double crustal thickness, resulting in km of local surface uplift and a huge cocoon of plutonic rocks, neither of which is observed.

The century-old paradigm of crystal fractionation in a big magma chamber never really panned out anyway, whereas magma mixing is a widespread process. An alternative that is consistent with petrology and geophysics is that magma chambers are mid- to deep-crustal phenomena (heat flow and room problems are mitigated at depth), and only required for eruptions that release huge amounts of magma. Typical eruptions and plutons are fed by dikes that drain partially molten zones in the crust. A rising dike may tap and recruit other partially molten zones, related or not, that it encounters during ascent. Shear flow is an effective homogenizer, and 100s to 1000s of meters of such transport can account for the wide range of mixing textures (composite dikes, dikes with enclaves, dikes with highly sheared enclaves, and magmas with xenocrysts) seen in plutonic and volcanic rocks.