Late Miocene to Quaternary basaltic volcanism of southeastern Oregon and adjacent states is mainly characterized by high-alumina olivine tholeiites (HAOT) that resemble typical back-arc basin basalts or T-MORB's. Previous regional studies indicate close petrographic and compositional similarities among the more primitive HAOT's (MgO „ 8 wt.%, K2O about 0.3 wt.%, Rb 2-4 ppm) over this large area. On the other hand, in a ca. 30x40-km area in
the western Harney Basin, numerous ca. 7 to 3 Ma mafic vents erupted phenocryst-poor basaltic rocks with large compositional variability, ranging from primitive to trace-element enriched HAOT's (all with „8 wt.% MgO) to subalkaline basaltic andesites and trachytic basaltic andesites with trace element enrichments as great as 30 times that in basalts.
Low Cr and Ni concentrations yet near basaltic silica composition of the trachytic basaltic andesites exclude:
i) direct derivation from the mantle, and ii) assimilation/ mixing of enough felsic components to explain the high concentrations of incompatible trace elements. Mafic bulk composition and trace element enrichment can be reconciled through an origin by multiple recharge and fractionation cycles. In contrast, subalkaline andesites are likely derived by fractionation from high-Mg HAOT with some crustal contamination.
Even if variable mantle compositions, variable percent of melting, and crustal assimilation may have contributed to the elemental variation among high-Mg HAOT's of Harney Basin, these processes are thought to have played a minor role because mantle-reservoir characteristics, such as Hf/Sm and Sr and Nd isotopic ratios are constant, and incompatible trace element trends are too steep on plots with Na8. Enriched trachytic basaltic andesites exist and are suitable for contamination of primitive HAOT to yield enriched HAOT, making this our preferred mixing process over others (e.g., mix. with extreme low % mantle or crustal melt).
Apparent simple liquid lines of descent for most
compatible major & trace elements contrast with strong incompatible element enrichments in many mafic suites and are well illustrated by the succession primitive to enriched HAOT to basaltic andesite. The compositional variation can be mainly explained by fractionation and interaction of different mafic liquids derived from the same parental composition. This suggests that crustal contamination or variable mantle source compositions can be easily overestimated, because self-contamination effects can be as great or greater.