Upper mantle temperatures during extension and breakup

It has become widely, but not completely, accepted that mantle temperature is the primary control on the quantity of volcanism during extension and break-up. It is most likely that this hot mantle is a consequence of a convecting mantle heated from below, where large instabilities will inevitably form. Yet we cannot measure the temperature of the mantle. Instead indirect estimates ofthe temperature ofthe mantle below rift zones must be made. For example below the active rift zone in Afar, at the northern end of the East African Rift, estimates ofthe temperature ofthe mantle vary from 1350 to 1450°C from geochemical and geophysical observations. This active rift is demonstrably volcanic, yet if the lower estimate is true then the mantle below the volcanic massif of Etta Ale is no hotter than anywhere else. This begs the question: does upper mantle temperature really control breakup volcanism?

Building upon examples from breakup in the North Atlantic, Northwest Indian margins and Afar, I will show that by using a combination of geophysical and geochemical observations, the role ofthe thermal structure ofthe upper mantle in defining breakup magmatism can be better understood. Dming the formation of the Notth Atlantic a large volume of melt was generated, associated with tlle extensive on-shore flood basalts. Breakup of the continent occured after a series ofextensional events spmming the Carboniferous up until the Late Cretaceous. This extension ofthe lithosphere, combined with increased mantle temperatures, led to the volcmism that accompanied breakup. Offshore ofthe Deccan Traps in western India the story is different, with magama-poor break-up occurring after the trap volcanism and the formation of a series of failed volcm1ic rifts. In these two locations, mantle temperature is key, but the degree of volcanism is modified by the degree of extension that the lithosphere experiences prior to the interaction with thermal plumes.

The stories I wrote above can be tested using forward numerical models ofthe deformation ofthe upper mantle and crust. Predictions of melt volume and composition can be tested against observations of the volume and seismic velocity of intrusions, the composition of erupted lavas and present day seismic wave speeds through the upper mantle. Using an idealised 2D model of the upper mantle I will first demonstrate how rift history has impacted melt production in both the North Atlantic and Northwest Indian margins. I will then focus on the only present day volcanic and active rift zone, Afar. Here there is a history o f multiple phases o f extension and a complex distribution of crustal strength along strike. These factors can influence breakup, yet by combining multiple geophysical and geochemical constraints, and comparing these with model predictions, we can demonstrate that mantle potential temperature is the primary driver of volcanism and below Afar the mantle is hot (1450°C). While crustal and lithosphere structure can heavily modify break- up volcanism, ultimately, mantle temperature controls the evolution of continental breakup.