Geochemical studies on the mantle source lithologies of late Cenozoic alkali basalts from Baengnyeong, Pyeongtaek, and Asan in the Korean Peninsula

Abstract

Comprehensive mineral chemistry, 40Ar–39Ar dating, and whole-rock geochemical data, including Sr–Nd–Pb isotopes, in late Cenozoic (19–6 Ma) alkali basalts (basanite and trachybasalt) from Baengnyeong, Pyeongtaek, and Asan in the Korean Peninsula were determined to constrain their mantle source lithologies and spatial distribution. Olivines have high-Ni and low-Ca contents compared with those derived from peridotite sources. In a Sr–Nd isotope diagram, the samples define a mixing array between Indian MORB and the bulk silicate Earth. In Pb–Pb isotope space, they plot closer to the EM1 than to the EM2 end-member. Values of Δ8/4Pb are positive and range from +47.5 to +210.0, whereas Δ7/4Pb values vary from −3.9 to +14.3. On a primitive mantle-normalized multi-element diagram, the basanites show carbonatite-like compositions, with significant negative anomalies in Rb, K, Zr, Hf, and Ti. By contrast, the trachybasalts exhibit positive anomalies in (K), Pb, Sr, and P. Compared with the trachybasalts, the basanites have lower SiO2, and higher FeO* and P2O5 contents for a given MgO content. The REE ratios of the Baengnyeong basanites plot along the modeled trend for carbonated garnet-lherzolite-derived melts. Taken together, these observations suggest that the mantle source of the studied samples comprised three enriched components located at sub-lithospheric depths, namely recycled ancient K–Th-enriched pelagic sediments, eclogite, and carbonates. The mixing proportions of these source materials were different for each volcanic region. Comparison with late Cenozoic intraplate basalts in northeastern Asia indicates no meaningful spatial trend for the EM1 component.