Zircon petrochronology of four large, rapid-succession, caldera-forming eruptions in the central San Juan caldera cluster, Colorado

The Rat Creek Tuff (RCT; ~150 km³), Cebolla Creek Tuff (CCT; ~250 km³), Nelson Mountain Tuff (NMT; >500 km³), and Snowshoe Mountain Tuff (SMT; >500 km³) erupted between 26.91 ± 0.02 Ma (RCT) and 26.87 ± 0.02 Ma (SMT) according to single-crystal sanidine ⁴⁰Ar/³⁹Ar ages. No other ignimbrite flare-up in the world produced a similarly large volume of ignimbrite in such rapid succession. The eruption deposits alternate between zoned (crystal-poor to crystal-rich) and unzoned (crystal-rich) and provide an excellent opportunity to study the evolution of large, silicic crustal magmatic systems. Using U-Pb geochronology (CA-ID-TIMS and LA-ICP-MS), trace elements (LA-ICP-MS and EPMA), and δ¹⁸O (SIMS) of zircons, we document the chemical and temporal evolution of this unique sequence. Initial ID-TIMS U-Pb zircon dates show pre-eruptive magma accumulation over less than 600 k.y. before each eruption (<~27.5 Ma). Some zircon dates extend back to 1.4 to 4.3 m.y. before eruption. Zircon oxygen isotope δ¹⁸O values are homogeneous among eruptions, rims, and interiors, ranging between 5.1 to 6.3 ‰ and precluding assimilation of low-δ¹⁸O crustal rocks as a major process. The average Ti-in-Zr temperatures shift towards cooler values from the oldest RCT to the younger CCT and NMT. The youngest SMT shifted geographically to the Creede Caldera and has the highest average Ti-in-zircon temperature of the four eruptions. Trace elements show an increase in variance with decreasing Ti-in-zircon temperatures. Our results indicate the presence of a large region of magma accumulation and evolution active for at least 600 k.y prior to eruption. The comparison between our data and thermal modelling results suggests that the San Luis systems experienced progressive thermal maturation with temperature becoming generally lower, which is typical of systems constructed by prolonged magma injection into the crust. We are currently performing tailored thermal models that will allow us to quantify the rate of magma injection into the magmatic system to account for zircon ages and trace element distributions.