Plume/lithosphere interaction in the generation of continental and oceanic flood basalts: Chemical and isotopic constraints

The plume initiation model for flood basalt genesis predicts that melt generation will occur almost entirely within the plume head, beneath the lithosphere. However, isotopic and trace element differences between continental and oceanic flood basalts (CFBs and OFBs, respectively) require the incorporation of a lithospheric component in the former. Debate persists as to whether the "continental" signatures present in many CFBs derive from contamination of (plume-derived) melts with small volumes of highly incompatible-element-enriched lithospheric components (e.g., through crustal assimilation) or substantial melt generation within the lithospheric mantle. The chemical and isotopic compositions of OFBs and CFBs are used to constrain the mantle sources and melting environment responsible for flood basalt generation and the extent of crustal assimilation. Major and trace element trends in CFBs reflect smaller extents of partial melting than in OFBs and the presence of garnet in CFB source regions. These observations are consistent with the plume initiation model, because thick continental lithosphere will inhibit ascent and melt generation in sublithospheric plumes. However, melt generation from refractory, Fe-poor lithospheric mantle is also indicated for several CFBs, including the Siberian Traps. Depth of melt generation typically decreases with time for a given province, as indicated by the removal of a garnet signature in trace element ratios (e.g., decreasing Sm/Yb). These chemical and temporal trends suggest early melt generation within hydrous but otherwise refractory lithospheric mantle, followed by mechanical erosion of the lithosphere that allows additional ascent and melt generation within the underlying plume.