Search

We report new whole rock U-Th and in-situ oxygen isotope compositions for partially melted (0–50 vol% melt), low-δ18O Pleistocene granitoid blocks ejected during the ∼7.7 ka caldera-forming eruption of Mt. Mazama (Crater Lake, Oregon). The blocks are interpreted to represent wall rocks of the climactic magma chamber that, prior to eruption, experienced variable amounts of exchange with meteoric hydrothermal fluids and subsequent partial melting. U-Th and oxygen isotope results allow us to examine the timescales of hydrothermal circulation and partial melting, and provide an “outside in” perspective on the buildup to the climactic eruption of Mt. Mazama. Oxygen isotope compositions measured in the cores and rims of individual quartz (n = 126) and plagioclase (n = 91) crystals, and for transects across ten quartz crystals, document zonation in quartz (Δ18OCore-Rim ≤ 0.1–5.5‰), but show homogeneity in plagioclase (Δ18OCore-Rim ≤ ±0.8‰). We propose that oxygen isotope zonation in quartz records hydrothermal exchange followed by high-temperature exchange in response to partial melting caused by injection of basaltic to andesitic recharge magma into the deeper portions of the chamber. Results of modeling of oxygen diffusion in quartz indicates that hydrothermal exchange in quartz occurred over a period of ∼1000–63,000 years. Models also suggest that the onset of melting of the granitoids occurred a minimum of ∼10–200 years prior to the Mazama climactic eruption, an inference which is broadly consistent with results for magnetite homogenization and for Zr diffusion in melt previously reported by others. Uranium-thorium isotope compositions of most granitoid blocks are in 238U excess, and are in agreement with a 238U enriched array previously measured for volcanic rocks at Mt. Mazama. Uranium excess in the granitoids is likely due to enrichment via hydrothermal circulation, given their low δ18O values. The sample with the highest U excess (≥5.8%) also has the most 18O isotope depletion (average δ18Oplag = −4.0‰). The granitoids are a probable assimilant and source of U excess in volcanic rocks from Mt. Mazama. Two granitoids have Th excess and low δ18O values, interpreted to record leaching of U during hydrothermal alteration. A U-Th isochron based on the U excess array of the granitoids and volcanic rocks indicates that hydrothermal circulation initiated ∼40–75 kyrs before the climactic eruption, potentially marking the initiation of a persistent upper-crustal magma chamber. The U-Th ages are consistent with the maximum timescales inferred for hydrothermal alteration based on oxygen isotope zoning in quartz.

Interaction of mantle melts with the continental crust can have significant effects on the composition of the resulting melts as well as on the crust itself, and tracing this interaction is key to our understanding of arc magmatism. Lava flows and pyroclastic deposits erupted from ∼50 to 7.7 ka at Mt. Mazama (Crater Lake, Oregon) were analyzed for their Re/Os and U–Th isotopic compositions. Mafic lavas from monogenetic vents around Mt. Mazama that erupted during the buildup to its climactic eruption have lower 187 Os/ 188 Os ratios (0.1394 to 0.1956) and high 230 Th excess (( 230 Th/ 238 U) 0 of 1.180 to 1.302), whereas dacites and rhyodacites tend to have higher 187 Os/ 188 Os ratios (0.2292 to 0.2788) and significant 238 U excess (( 230 Th/ 238 U) 0 of 0.975 to 0.989). The less radiogenic Os isotope compositions of the mafic lavas can be modeled by assimilation of young (∼2.5 to 7 Ma), mafic lower crust that was modified during regional extension, whereas the more radiogenic Os isotope compositions of the dacites and rhyodacites can be attributed to assimilation of older (∼10 to 16 Ma), mid to upper crust that acquired its composition during an earlier period of Cascade magmatism. Production of Th excesses in the lower crust requires very young garnet formation accompanying dehydration melting in the lower crust at less than a few 100 ka by heat from recent basaltic magma injection. The results from this study suggest that the combination of Os and Th isotopes may be used to provide insights into the timescales of evolution of the continental crust in arc settings, as well as the influence of the crust on erupted magmas, and suggest a link between the age and composition of the lower and upper crust to regional tectonic extension and/or earlier Cascade magmatism.