INTERACTION BETWEEN MAGMA INPUT AND PRECIPITATION IN THE GROWTH OF LARGE MAGMA CHAMBERS
The rate of cooling depends on whether heat transport occurs by conduction or convection. Convection can move heat 10 times faster than conduction, but in arid regions recharge is limited and convection can be starved. For sheets of magma , where the problem is essentially 1-dimensional, the relevant energy balance is approximated by the dimensionless number M = q/ρLV, where q is the heat flux out the top of the body (~surface heat flow), ρ is density, L is latent heat of crystallization, and V is the rate at which new magma is accreted to the body. For M>1 heat loss exceeds the rate at which new heat is advected in by fresh magma and magma increments freeze as they are accreted; for M<1 the magma body grows. For heat flow of 200 mW/m2 (~5 HFU), M=1 when V is ~1 cm/yr. Highly elevated heat flows of 1000-2000 mW/m2 (regional average for Yellowstone) would require 5-10 cm/yr of magma accretion. Such a rate is unlikely, and this high heat flow probably reflects convective cooling that currently outpaces magma accretion.
We therefore propose that development of magma bodies capable of erupting with large-scale caldera collapse should be favored in regions that were arid when magma accumulated. In the Andes, large Quaternary calderas are unknown in the northern tropics and southern temperate zone but are abundant in arid Bolivia and northern and central Chile. The huge Altiplano-Puna volcanic complex occurs in one of the driest regions on Earth. Other major Cenozoic ignimbrite provinces (Great Basin, Sierra Madre Occidental, Snake River Plain) developed in arid regions. Other factors (e.g., decreases in permeability) can decrease M and promote development of large magma bodies in wet regions, but precipitation may exert a first-order control on whether a large magma body or incremental pluton forms as magma accumulates.