Thermomechanical feedbacks in magmatic systems: Implications for growth, longevity, and evolution of large caldera-forming magma reservoirs and their supereruptions

Large magma bodies that feed super-eruptions and build batholiths are not instantaneously emplaced. Many accumulate over time scales of 10⁵ to 10⁶years as part of magmatic episodes that last 10⁷years and propagate a thermal and magmatic front through the crust to stabilize the reservoirs in the upper crust. This history imposes a rheological and thermodynamic conditioning on the host rocks that sets in motion three feedbacks that promote growth and longevity of large silicic magma reservoirs. Herein we review the development of ideas about the thermomechanical evolution of large silicic magma systems and explore the feedbacks and their implications for the growth, longevity, and evolution of large silicic magma reservoirs. Feedback 1 promotes increasing temperatures and consequent lower viscosities in the host rocks and the development of a ductile halo. Feedbacks 2 and 3 are feedbacks that result from the thermal dependence of the rheological properties of this ductile halo. In feedback 2 low wall rock viscosities lead to dissipation of strain in the host rocks reducing the likelihood of eruption. Feedback 3 is a negative loop between volume change and pressurization also reducing the likelihood of wall rock failure and eruption. We show that these feedbacks are most pronounced in larger reservoirs (>500km³) and conspire to promote reservoir growth. Predicted imprints of these feedbacks are extended melt present lifetimes, complex heterogeneous age records and crystal-rich magma in some large silicic magma reservoirs. In this framework, interruption of the slow steady progress towards viscous death and solidification manifests as a supereruption. Second boiling and recharge (including buoyancy effects) acting in concert or independently lead to roof uplift and extension and eruptions are finally triggered by downward propagating faults from the extended and weakened roof. This connotes a thermomechanical division of calderas into those where eruptions are triggered "internally" by magmatic processes and those that are triggered "externally" by faulting related to roof uplift and attenuation. The division is controlled by size of magma reservoir, although, true to nature, exceptions exist, demonstrating interruption of the feedbacks by other processes like tectonism.