Transient heat and mass transfer in a caldera setting quantified by 3D numerical simulations

Fluid convection in many hydrothermal systems is driven by heat and energy provided by cooling igneous intrusions, while flow pathways are affected by rock permeability. Increased permeability in fracture zones and faults can form preferred fluid flow pathways. Although field evidence indicates that flow localisation in fracture zones and faults is a common phenomenon, its quantitative impact on heat and mass transfer around cooling intrusions has remained understudied. Here, we present three-dimensional numerical fluid flow simulations of a conceptual caldera setting with a ring fault to (1) explore the spatio-temporal evolution of heat and mass transfer, (2) quantify the effects of a high-permeability ring fault on fluid transport, and (3) describe the intra-fault flow dynamics. A circular, cooling magma chamber emplaced at 3 km depth acts as active heat source and an inwardly-dipping, cone-shaped zone of increased permeability represents the bounding caldera ring fault. We systematically vary the permeability of the bulk rock, the ring fault, and the crystallised intrusion, as well as the temperature (T_BDT) at which rocks start to deform ductile and become impermeable to explore their first-order controls on heat and mass transfer in a caldera setting.

We observe two distinct intra-fault flow scenarios: (1) intra-fault convection allows for increased meteoric recharge and occurs when the fault permeability is at least two orders of magnitude larger than the permeability of the surrounding rock; and (2) upflow along the fault plane occurs via a continuous upflow front in case of permeability contrasts lower than two orders of magnitude. Both scenarios lead to a fault-focused fluid transport during the first ~5 kyrs, where hot fluids are directly fed into the fault plane by the underlying heat source, forming near-surface boiling zones. ~20–30 % of the total energy transfer to shallower depths <1.5 km takes place during this early flow stage and is accommodated by the ring fault. A continually cooling magma chamber and the consequently shrinking heat source shifts upflow of hot fluids towards the caldera infill. This shift of localised fluid upflow leads to the formation of a hydrothermal plume, which accounts for ~60–70 % of the total energy transfer to depths <1.5 km during the first 20–30 kyrs.

The efficiency of heat mining and how fast energy is transferred towards the surface is affected by the T_BDT and by the permeability of the crystallised magma chamber. A higher T_BDT allows fluids to migrate through hotter areas and increased permeability (e.g., caused by cooling joints) enables fluids to infiltrate into the crystallised magma chamber and therefore to harness energy more efficiently.

Overall, we conclude that ring faults can be important structures in caldera settings that transiently localise increased mass and heat transfer over a hundreds of years period, however, they may become less significant on geological timescales, particularly when the location of the heat source changes, which may reduce the input of hot fluids into the fault plane.