18O/16O ratios in ash-flow tuffs and lavas erupted from the central Nevada caldera complex and the central San Juan caldera complex, Colorado

Abstract

Oxygen isotope compositions were measured on 129 quartz, feldspar, and biotite phenocrysts from ash-flow tuffs and lava domes erupted from the Oligocene central Nevada and central San Juan caldera complexes. Most of the ash-flow tuffs are compositionally zoned with low-phenocryst rhyolite bases and high-phenocryst quartz-latite tops, but both within individual units and throughout each of the eruptive sequences at each locality, the δ18O values are remarkably constant. δ18O values of the central Nevada magmas range from +9.1 to +9.8 per mil: These values are high and indicate the involvement of high-18O geosynclinal sediments in the melting process. Magmatic δ18O values decrease by only about 0.4 per mil from the initial eruption sequence to the middle eruptive, the giant Monotony Tuff (3000 km3). The initial higher δ18O values are reestablished in the late eruptive sequence, but decrease again by about 0.4 per mil in the latest ring-fracture eruptions. δ18O values in the central San Juan magmas range from +6.8 to +7.5: These values are relatively low and indicate involvement of lower cratonal crust and upper mantle in the melting process. Magmatic δ18O values decrease by about 0.4 per mil from the early sequence (Fish Canyon, Carpenter Ridge, and Mammoth Mountain Tuffs) to the late sequence (Wason Park, Nelson Mountain, and Snowshoe Mountain Tuffs). 18O/16O fractionations among phenocrysts in both Nevada and Colorado are much smaller than among corresponding minerals in plutonic granitic rocks. These fractionations also decrease from stratigraphically lower to higher samples in each cooling unit, so the 18O/16O data agree with other evidence that these represent quenched equilibrium at magmatic temperatures, and that prior to eruption the tops of the magma chambers were cooler than the deeper portions. In striking contrast to what is observed in Iceland and in the late-Tertiary to Quaternary southwest Nevada and Yellowstone caldera complexes, we have found no evidence for any low-18O rhyolitic magmas. Thus, low-18O rhyolitic magmas must be less common than heretofor believed, and their origin must be a result of special circumstances involving the timing, depth, and intensity of meteoric-hydrothermal activity. We tentatively suggest that extensional tectonics and regional rifting may be one of the prerequisites for their development.