The role of edge-driven convection in the generation ofvolcanism – Part 2: Interaction with mantle plumes, applied to the Canary Islands

Abstract

Abstract. In the eastern Atlantic Ocean, several volcanic archipelagos are located close to the margin of the African continent. This configuration has inspired previous studies to suggest an important role of edge-driven convection (EDC) in the generation of intraplate magmatism. In a companion paper (Manjón-Cabeza Córdoba and Ballmer, 2021), we showed that EDC alone is insufficient to sustain magmatism of the magnitude required to match the volume of these islands. However, we also found that EDC readily develops near a step of lithospheric thickness, such as the oceanic–continental transition (“edge”) along the western African cratonic margin. In this work, we carry out 3D numerical models of mantle flow and melting to explore the possible interactions between EDC and mantle plumes. We find that the stem of a plume that rises close to a lithospheric edge is significantly deflected ocean-ward (i.e., away from the edge). The pancake of ponding hot material at the base of the lithosphere is also deflected by the EDC convection cell (either away or towards the edge). The amount of magmatism and plume deflection depends on the initial geometric configuration, i.e., the distance of the plume from the edge. Plume buoyancy flux and temperature also control the amount of magmatism, and influence the style and extent of plume–EDC interaction. Finally, comparison of model predictions with observations reveals that the Canary plume may be significantly affected and deflected by EDC, accounting for widespread and coeval volcanic activity. Our work shows that many of the peculiar characteristics of eastern Atlantic volcanism are compatible with mantle plume theory once the effects of EDC on plume flow are considered.

Mantle structure beneath the Macaronesian volcanic islands (Cape Verde, Canaries, Madeira and Azores): A review and future directions

“…To explain the distribution of the volcanism in the islands, Anguita and Hernán (2000) suggest a model where an uppermantle residue from a fossil plume interacts with a lithospheric fracture propagating from the Canary Islands to the Atlas Mountains in Morocco. Later, similar unified scenarios invoking a mantle plume interacting with a complex lithospheric architecture have been proposed by Geldmacher et al (2005); Duggen et al (2009); Manjón-Cabeza Córdoba and Ballmer, (2022) for this area and applied to other regions on Earth (e.g. ; Celli et al, 2020b;Civiero et al, 2022).…”

Prolonged multi-phase magmatism due to plume lithosphere interaction as applied to the High Arctic Large Igneous Province

“…Furthermore, HALIP magmatism shows compositions and characteristics of plume influence (e.g., Tegner et al, 2011;Buchan & Ernst, 2018;Senger & Galland, 2022). That said, depending on the paleotopography of the LAB, edge-driven convection or shear-driven upwelling (in addition to the convective patterns modelled here) may contribute to local dynamics and overall melt volumes and distributions (e.g., Conrad et al, 2010;Manjón-Cabeza Córdoba & Ballmer, 2022;Duvernay et al, 2022;Negredo et al, 2022).…”

“…Previous studies looking at melting in plumes have focused on different aspects of the problem, including melting in thermochemically zoned plumes (e.g., Dannberg & Gassmöller, 2018), melting in the presence of continental or cratonic lithosphere (Duvernay et al, 2022), or melting for specific plumes and tectonics settings (e.g., Ballmer et al, 2011;Bredow et al, 2017;Steinberger et al, 2019;Liu et al, 2021;Manjón-Cabeza Córdoba & Ballmer, 2022;Negredo et al, 2022). Even though the various model setups and levels of complexity are different, all studies of plume dynamics mentioned above have in common that they only consider simplified melt fractions, and do not model melt migration, as we do in this study.…”

“…Even though the various model setups and levels of complexity are different, all studies of plume dynamics mentioned above have in common that they only consider simplified melt fractions, and do not model melt migration, as we do in this study. Some studies vary parameters related to melting changes such as the density (Ballmer et al, 2011;Bredow et al, 2017;Steinberger et al, 2019;Liu et al, 2021;Manjón-Cabeza Córdoba & Ballmer, 2022), viscosity (Ballmer et al, 2011;Bredow et al, 2017;Steinberger et al, 2019;Liu et al, 2021), or melting temperature (Ballmer et al, 2011;Negredo et al, 2022), and the work of Ballmer et al (2011) on the Hawaiian Plume shows rejuvenated volcanism next to the plume track due to small-scale convection. While the the role of small-scale convection is important in both the work of Ballmer et al (2011) and this study, the cases are not directly comparable.…”

“…for the Iceland plume based on V-shaped ridge segments in the North Atlantic (e.g., Ito, 2001; S. M. Jones et al, 2002;Parnell-Turner et al, 2014). Dynamically, such a plume pulse could be related to (potentially slab-driven) dynamics at the core-mantle boundary (e.g., Heyn et al, 2020), interaction with edge-driven convection (e.g., Manjón-Cabeza Córdoba & Ballmer, 2022;Negredo et al, 2022), solitary waves (Ito, 2001), or an interaction between the plume and the slab found underneath Green-land at about 1000-1600 km depth (Shephard et al, 2016). Strong plume pulses or successful seafloor spreading can generate melting without the mechanism described in this paper, but a previously melt-influenced region of the lithosphere is more prone to rejuvenated melting, even without small changes in plume flux or failed rifting.…”