Petrogenesis and volatile stratigraphy of the Bishop Tuff: Evidence from melt inclusion analysis

Abstract

The Bishop Tuff (BT), erupted from the Long Valley Caldera at 0.74 Ma, is composed of a Plinian tephra and ignimbrite (Hildreth, 1979). Based on ion and electron microprobe analyses of melt inclusions (MI) in quartz and sanidine phenocrysts, a strong H2O gradient was present in the upper portion of the magma chamber, mainly in the first 120 km3 of erupted material. The H2O content of the magma which formed the Plinian tephra drops from 6 wt % to 3.5 wt %. In contrast, the magma which formed the ignimbrite contained a relatively constant amount of H2O, between 2 and 4 wt %. The strong drop in H2O content of the magma which formed the Plinian tephra suggests that only in its extreme upper part, if any, was the PH20 in the magma close to Ptotal. The halogen content of the BT magma was low and relatively constant at ∼700 ppm Cl and ∼ 500 ppm F. The trace and major element composition of MI from the Plinian and first‐erupted ash flow lobes of the BT are similar to that of bulk tephra. The range in the trace element concentrations of inclusions suggests that fractional crystallization may have affected magmatic composition. However, in addition to fractional crystallization, the composition of MI, phenocrysts and bulk pumice lumps suggests that magma mixing may have been an important process in establishing the final trace element zonation within the Bishop magma chamber and may be responsible for some of the most dramatic observed trace element variations. The BT eruption appears to have removed sequential stratigraphic compositional layers from the magma chamber, whereas fhe Lower Bandelier Tuff (LBT) appears to have erupted chaotically, although both magma chambers are characterized by essentially the same volatile and therefore density zonation. However, the H2O content of the Plinian:ignimbrite transition is different for the two (∼3 wt % for the BT, 4–5 wt % for the LBT), suggesting that the LBT Plinian eruption may have ended prematurely, possibly due to caldera collapse. Therefore, the magma withdrawal dynamics of these eruptions may be more strongly controlled by external factors, such as vent configuration, rather that the volatile gradient of the melt.