Late cenozoic bimodal magmatism in the northern basin and range province of southeastern Oregon

Abstract

Compositional features of 93 samples of primitive Pliocene to recent basalts erupted along the Brothers Fault Zone in the northernmost Basin and Range indicate that they were derived from a shallow mantle source and underwent only minor shallow-level fractionation. Simple mass-balance modelling can derive these basaltic bulk compositions by removal of small amounts of observed crystalline phases from glass compositions produced in peridotite melting experiments. Additional support comes from phase equilibria data on other magnesian basalts having similar bulk compositions. The eruption of these lavas without substantial subcrustal fractionation was probably promoted by progressive extension along the Brothers Fault Zone. This origin is in sharp contrast to that generally proposed for mid-Miocene Columbia River and Steens Mountain basalts, which show clear evidence in their evolved compositions (e.g. Mg # ~ 40) of having stagnated at shallow depth where they differentiated to nearly basaltic andesite compositions. Bulk compositions of northern Basin and Range silicic rocks, together with physical and thermal considerations, suggest that they, like their counterparts in the Snake River Plain, were products of crustal anatexis driven by the injection of mafic magmas, but with meta-volcaniclastic protoliths rather than Archaean basement rocks, as in the case of the Snake River Plain rhyolites. These petrologic features suggest that the arrival of the mantle plume presently beneath Yellowstone produced or strongly influenced most late Cenozoic magmatism in the Oregon northern Basin and Range. This model accounts for many features of the northern Basin and Range in Oregon: (1) the change in basaltic character about 10 to 8 Ma ago from voluminous, evolved Columbia River/Steens lavas to smaller-volume primitive lavas and the lack of younger lavas atop the Columbia River Basalts; (2) the lack of an obvious track of the Yellowstone hot spot west of the Oregon-Idaho-Nevada tri-state area; (3) the “mirror-image” age relationship of silicic rocks in the northern Basin and Range and Snake River Plain; (4) the formation of silicic rocks by crustal anatexis and the general decrease in their volumes with time in Oregon but not along the Snake River Plain; (5) the high elevation of the region; and (6) the high surface heat flow in the Oregon northern Basin and Range. The proposed model obviates the controversy surrounding the pre-Miocene history of the Yellowstone plume by proposing that the plume initiated about 18 Ma ago.