Page 46 - Caldera Volcanism Analysis, Modelling and Response

Residence Times of Silicic Magmas Associated with Calderas 21

Figure 6 Yellowstone system. Erupted volume vs. (a) residence times and (b) magma
production rates. Circles are maxima and squares the minima values (or single estimates), and
lines are shown connecting the two.The data of the maximum estimates (W0.4 My) for Blue
Creek (BC), Middle Biscuit (MB) and Dunraven Road (DR) are not shown.The deposits from
two eruptions at ca.115 ka (Solfatara Plateau, SF, and WestYellowstone,WY) show inherited
zircon crystals from the previous one at 160 ky (Dry Creek, DC;Vazquez and Reid, 2002).
Abbreviations for the units are: MF, Mesa FallsTu¡, LC, Lava CreekTu¡, CF, Canyon £ow,
PP, Pichstone Plateau.The volumes of the small eruptions are estimated from the publication
3
of Christiansen (2001) to be o50 km and are only very rough values and thus they are shown
as a box.There does not seem to be a correlation between volume and residence times, although
larger eruptions seem to have higher magma production rates. Data sources are inTable 5.
the time since total melting and eruption of the post-caldera lavas, and requires
1 2 1
heating rates of 3 10 –3 10 Ky . These melting times are about an order
of magnitude shorter than the lowest estimate of the residence times (ca. 50 ky).
The residence times of much younger post-caldera deposits (the so-called
Central Plateau Member, 70–160 ka) were investigated by Vazquez and Reid
(2002). They found that none of the zircons were related to any of the three major
collapse caldera tuffs. The eruption age of the oldest investigated flows overlap
238 230
with the zircon U– Th disequilibrium ages, although the precision of the data
still allow residence times of several ky (Table 5, Figure 6). For flows erupted down
to 110 ka they found zircon model ages (isochrons obtained with two points) at
ca. 167 ka implying residence times of ca. 50 ka and probably crystal recycling from