Anita L. Grunder

The Loma Seca Tuff and the Calabozos caldera: A major ash-flow and caldera complex in the southern Andes of central Chile

A 26 × 14-km composite ring-structure caldera of late Pleistocene age has been discovered and mapped near the Andean crest in central Chile (35°30′S). Rhyodacitic to dacitic zoned ash-flow sheets, each representing 150 to 300 km 3 of magma, were emplaced 0.8, 0.3, and 0.15 m.y. ago; the youngest of the associated collapses was closely followed by resurgent doming of the caldera floor and development of a longitudinal graben. Postcaldera eruptions of dacite and andesite have persisted into Holocene time, and active hot springs are abundant along caldera-marginal and resurgent fault systems, suggesting a significant geothermal-energy resource. The Pleistocene eruption rate of this district and the abundance of older Quaternary to Miocene ash-flow remnants in the 33°S to 36°S segment of the glaciated southern Andes indicate that ash-flow magmatism has been no less important here than in the arid central Andes (16°S–28°S), where ash-flow sheets are far better preserved.

Low δ18O silicic volcanic rocks at the Calabozos caldera complex, southern Andes

Sr- and Pb-isotope data from the Calabozos center (87Sr/86Sr=∼ 0.7043, 206Pb/204Pb=18.64–18.66, 207Pb/204Pb=15.59–15.60, 208Pb/204Pb=38.52–38.55) fall within the range of values reported for the southern volcanic zone (33–42° S) of the Andean arc. The range of δ18O (5.0–6.3), however, includes unusually low values compared to volcanic rocks of similar bulk composition in the region. The Calabozos caldera complex lies at 35 °30′ S, where the continental crust under the Andes thins southward from >45 to ∼ 30 km. Three voluminous late Pleistocene ashflow tuffs, collectively called the Loma Seca Tuff, constitute the bulk of >1,000 km3 of eruptive products at the Calabozos caldera complex and are evidence for a major, longlived andesitic-to-rhyodacitic magma reservoir at shallow crustal levels. The δ18O values of the most evolved volcanic rocks from the Calabozos center are lower than predicted for rhyodacite produced by crystal fractionation from basalt typical of the region. Variation of δ18O independent of bulk composition and inferred magmatic water contents indicates that the 18O depletion is a late-stage, upper-crustal phenomenon that cannot simply be attributed to magmatic interaction with meteoric water. The data are interpreted to be the result of assimilation of 5–30% of roof and wall rocks previously depleted in 18O by isotopic exchange in a meteoric hydrothermal system overlying the magma reservoir. Combined assimilation and fractional crystallization calculations applied to Sr isotope data show that the isotopic contrast between the Calabozos magmas and the assimilated rocks is very small. Hydrothermally-altered volcanic and plutonic rocks from the Tertiary Andean arc complex and Mesozoic-to-Cenozoic volcaniclastic sediments typical of the local basement provide a geologically reasonable contaminant compatible with the Sr- and O-isotope data. Pb-isotope data from the Calabozos system lend no significant insight into upper crustal contamination.