Page 62 - Caldera Volcanism Analysis, Modelling and Response
P. 62
Residence Times of Silicic Magmas Associated with Calderas 37
magma was above the solidus, or they reflect the recycling of crystals from previous
intrusive episodes of the same magmatic system, and thus provide artificially long
residence times for the studied magma. Evidence for both cases is discussed below.
4.1.1. Residence times as a record of the duration that magmas spend above
their solidus
The most robust data for this case are the zircon age distributions from the large
3
volume (300–1,000 km ) Whakamaru ignimbrites (Brown and Fletcher, 1999) and
3
from allanite of the Youngest Toba Tuff (2,800 km ; Vazquez and Reid, 2004).
In both systems the ages decrease progressively (i.e., without a major gap) from
the centre to the rim of crystals. This is crucial because it could mean that the
crystals were able to grow from a melt without interruption and for these two
systems it amounts to about 200 ky.
A simple test of this possibility is to consider the thermal evolution of a magma
reservoir based on a conduction model of an instantaneous heat source (e.g.,
Equation (3); see Figures 9 and 10 and their captions for more details on the
parameters used). Considering the latent heat of crystallisation and multiple
Figure 10 Residence time vs. calculated magma volume at depth for di¡erent volumes of
erupted magma.The curves were calculated using the thermal model parameters from previous
¢gure but using 7501C as the lowest temperature that a magma would be able to erupt.This is
a simpli¢cation and the e¡ect of viscosity and crystallinity should also be considered. However,
since the lowest pre-eruptive temperatures of most magmas discussed in the text are at
7507251C it appears a valid simpli¢cation to use only temperature as a eruptability criterion.
3
The ¢gure highlights that small volumes of magma (e.g.,1km ) with long residence times
3
(e.g.,100 ky) imply very large amounts of unerupted magma at depth (e.g., ca. 800 km ). Such
a ratio of 800:1 of intrusive over extrusive volumes is much larger than the maximum value of
10 determined by other authors (e.g.,White et al., 2006). Dashed lines indicate intrusive to
extrusive ratios of 10:1and 2:1. Most residence times of large eruptions fall within these values
of intrusive to intrusive volumes.