Frontiers paper Mountain building, mantle convection, and supercontinents: Holmes (1931) revisited

Orogeny results from crustal thickening at active margins, and much progress has been made on understanding the associated kinematics. However, the ultimate cause of orogeny is still debated, especially for the case of extreme crustal thickening. Inspired by the seminal work of Holmes (1931), we explore the connections between the style of orogeny and mantle dynamics. We distinguish between two types of orogeny, those that are associated with one-sided, mainly upper mantle subduction, “slab-pull orogeny”, and those related to more symmetric, whole mantle convection cells, referred to as “mantle”, or “slab-suction orogeny”. Only the latter leads to extreme crustal thickening. We propose that mantle orogeny is generated by the penetration of slabs into the lower mantle and the associated change in the length scales of convection. This suggestion is supported by numerical dynamic models which show that upper plate compression is associated with slab penetration into the lower mantle. Slabs can further trigger a buoyant, plume upwelling from the core-mantle boundary which enhances this whole mantle convection cell, and with it upper plate compression. We explore the geological record to test the validity of such a model. For the present-day, compressional backarc regions are commonly associated with slabs that subduct to the deep lower mantle. The temporal evolution of the Nazca and Tethyan slabs with the associated Andean Cordillera and the Tibetan-Himalayan orogenies likewise suggests that extreme crustal thickening below the Bolivia and Tibetan plateau occurred during slab penetration into the lower mantle. This episode of crustal thickening in the Tertiary bears similarity with Pangea assembly events, where the Gondwanide accretionary orogen occurred at the same time of the Variscan-Appalachian and Ural orogeny. We propose that this Late Paleozoic large-scale compression is likewise related to a change from transient slab ponding in the transition zone to lower mantle subduction. If our model is correct, the geological record of orogeny in continental lithosphere can be used to decipher time-dependent mantle convection, and episodic lower mantle subduction may be causally related to the supercontinental cycle.