Late Pleistocene to Holocene activity at Bakening volcano and surrounding monogenetic centers (Kamchatka): volcanic geology and geochemical evolution

Abstract

The different roles of variable mantle sources and intra-crustal differentiation processes at Bakening volcano (Kamchatka) and contemporaneous basaltic monogenetic centers are studied using major and trace elements and isotopic data. Three suites of volcanic activity are recognized: (1) plateau basalts of Lower Pleistocene age; (2) andesites and dacites of the Bakening volcano, the New Bakening volcano dacitic centers nearby; and (3) contemporaneous basaltic cinder cones erupted along subduction zone—parallel N–S faults. Age-data show that the last eruptions in the Bakening area occurred only 600–1200 years ago, suggesting the volcano is potentially active. Major element variations and petrographic observations provides evidence for a fractionation assemblage of olivine, clinopyroxene, ±plagioclase, ±magnetite (?) within the basaltic suite. The fractionation in the andesites and dacites is dominated by amphibole, clinopyroxene, orthopyroxene and plagioclase plus minor amounts of magnetite and apatite. The youngest cpx-opx-andesites of Bakening main volcano deviate from that trend. Their source was probably formed by mixing of basaltic magmas into the silicic magma chamber of the Bakening volcano. Overall trace element patterns as well as the Sr–Nd–Pb isotopic compositions are quite similar in all rocks despite large differences in their chemical composition (from basalt to rhyodacite). In detail however, the andesite–dacites of the central Bakening volcano show a stronger enrichment in the more incompatible elements and depletion in HREE compared to the monogenetic basaltic centers. This results in a crossing of the REE-pattern for the two suites. The decrease in the HREEs can be explained by amphibole fractionation. A slab component is less likely because it would result in fractionation of the HREE from each other, which is not observed. The higher relative amounts of LILE in the dacitic and the large scatter in the basaltic rocks must be the result of a variable source enrichment by slab-derived fluids overprinting a variable depleted mantle wedge. The plateau basalts are less depleted in HFSE and show a more fractionated HREE pattern. These lavas could either result from a slab component or the addition of an OIB-type enriched mantle in their source.