DIVERSE RHYOLITES CLOSELY ASSOCIATED IN SPACE AND TIME, WESTERN SNAKE RIVER PLAIN

We have initiated an investigation of the petrogenesis of rhyolitic magma in the western Snake River Plain by studying five rhyolite units closely associated in time (11.6 to 10.5 Ma) along the Owyhee Front near Murphy, ID. Four are lavas of ~2 — 20 km³ volume, and the other is a densely welded ignimbrite of ~10 km³. None are associated with a caldera. Compositionally the rhyolites range between low-silica (71% SiO₂) and high-silica (77% SiO₂) types. All lack hydrous mafic phenocrysts, and are broadly similar to other high-temperature "dry" Snake River Plain rhyolites (Honjo et al., 1992).

Trace element and isotope characteristics of each of the three low-silica rhyolites require a unique source and/or residual source mineralogy. None are depleted in Rb, Th or U, and lower crustal high-grade metamorphic rocks are an unlikely source; instead, the lavas may be the products of low-pressure melting of calc-alkaline granitoid rocks, as proposed for A-type silicic magmas by Patino-Douce (1997). The high-silica rhyolites exhibit strong depletions in Sr and Ba and have low La/Nb ratios, attributed to extensive fractionation of alkali feldspar plus a LREE-sequestering phase such as allanite or chevkinite. Additionally, both high-silica rhyolites exhibit positive Ce anomalies, with Ce/Ce* increasing as MREE become slightly depleted with respect to LREE and HREE, but both changes are only weakly correlated with variations due to degree of fractionation. We do not fully understand the behavior of Ce in these rocks, although it is clear that surface-altered lithologies must be involved in the petrogenesis of the high-silica rhyolites.

Most of the flows have oxygen isotope ratios, measured on fresh quartz and feldspar phenocrysts, in the normal rhyolite range (d¹⁸O = 7.5 to 9.0), but one lava, the Horse Basin Flow, is distinctly ¹⁸O depleted (d¹⁸O = 2.5 to 4.0). This occurrence of low d¹⁸O rhyolite with no evidence for an associated caldera may require modification to existing models for the genesis of ¹⁸O-depleted rhyolites.

References: Honjo et al., Bull. Volcanol. 54, 220-237 (1992); Patino-Douce, Geology 25, 743-746 (1997).