Amplification of sub-lithospheric dynamics by melt migration during plume-lithosphere interaction

The interaction of mantle plumes with continental or cratonic lithosphere can result in (large-scale) volcanism and continental breakup, but these consequences seem to be limited to tectonic settings with pre-existing weak zones. In contrast, most parts of continental plume tracks, or their hypothesized tracks, show no extrusive magmatism. To reconcile this, our previous work has shown that even in the absence of melt, sustained plume-lithosphere interaction leads to lithospheric thinning, followed by elevated surface heat flux about 40-140 million years after the thermal anomaly in the mantle disappears. Therefore, melt-free continental plume tracks can be initially identified by a reduced lithosphere thickness, and later by an increased surface heat flux that temporally and spatially follows the thinned lithosphere.

Yet, even if melt is not erupted, variable amounts of melt may still be generated at the base of the lithosphere above the plume, and this melt can impact local dynamics. In order to assess the role of melt in plume-lithosphere interactions, we have developed a recent suite of numerical models of mantle convection that include melting/freezing and melt migration. Our results indicate a much stronger time-dependence of models with melt compared to models without melt. In particular, small-scale convection at the base of the lithosphere becomes more vigorous, which leads to patterns that feature more localized and larger amplitude lithospheric removal and stronger asymmetry across the plume track. The generation of melt in a thinned area has a self-enhancing effect; more melt thins the lithosphere faster, resulting in more melt generation. However, the effect of thinning for a moving plate is limited, both with respect to the affected area and the time during which this local thinning can be sustained. As a result, the surface heat flux pattern, which is a long-pass filtered image of the lithosphere thinning, does not change significantly compared to a case without melt. However, melt migration brings heat closer to the surface, which increases the amplitude of the heat flux anomaly, and reduces the delay time following lithosphere thinning. The amplification of local dynamics by melt migration is especially pronounced if the plume interacts with pre-existing topography of the lithosphere-asthenosphere boundary (LAB), e.g. steps in lithospheric thickness. Depending on the LAB topography, multiple events of melt generations and magmatic intrusion can be generated by a single plume over tens of millions of years . Such a scenario may explain the pulse-like prolonged activity of the High Arctic Large Igneous Province (HALIP; which erupted between 130-85 Ma) and potentially an early phase of an Iceland plume track under Greenland (pre-62 Ma).