How plume‐ridge interaction shapes the crustal thickness pattern of the Réunion hotspot track

Abstract

The Réunion mantle plume has shaped a large area of the Earth's surface over the past 65 million years: from the Deccan Traps in India along the hotspot track comprising the island chains of the Laccadives, Maldives, and Chagos Bank on the Indian plate and the Mascarene Plateau on the African plate up to the currently active volcanism at La Réunion Island. This study addresses the question how the Réunion plume, especially in interaction with the Central Indian Ridge, created the complex crustal thickness pattern of the hotspot track. For this purpose, the mantle convection code ASPECT was used to design three‐dimensional numerical models, which consider the specific location of the plume underneath moving plates and surrounded by large‐scale mantle flow. The results show the crustal thickness pattern produced by the plume, which altogether agrees well with topographic maps. Especially two features are consistently reproduced by the models: the distinctive gap in the hotspot track between the Maldives and Chagos is created by the combination of the ridge geometry and plume‐ridge interaction; and the Rodrigues Ridge, a narrow crustal structure which connects the hotspot track and the Central Indian Ridge, appears as the surface expression of a long‐distance sublithospheric flow channel. This study therefore provides further insight how small‐scale surface features are generated by the complex interplay between mantle and lithospheric processes.

Anisotropic Tomography Around La Réunion Island From Rayleigh Waves

“…The values of the azimuthal anisotropy in this area are small (<1%), and this might indicate a vertical small-scale upwelling: the fast direction pattern shows a NW-SE direction north of La Réunion Island and a NE-SW direction south of Réunion Island, which creates a convergent east-west flow at a latitude of 18 ∘ to 20 ∘ toward the CIR. The direction of the fast axes of the observed azimuthal anisotropy beneath Réunion, Mauritius, and Rodrigues Islands (where the permanent stations are located) is compatible with results from SKS splitting analysis (Barruol & Fontaine, 2013;Scholz et al, 2016) and in agreement with independent geodynamic models (Bredow et al, 2017).…”

Variable Melt Production Rate of the Kerguelen HotSpot Due To Long‐Term Plume‐Ridge Interaction

“…In this case, the Southern Kerguelen Plateau would have been created contiguously with the Antarctic continental margin, as already noted by Müller et al (). Therefore, our model does not start until 120 Ma, and the plume head enters the model at that time with an excess temperature Δ T head =300 K, radius R head =250 km, and a vertical inflow velocity v head =20 cm/yr, corresponding to Bredow et al (), such that the spatial extent of the melting region agrees approximately with the surface area of the Southern and Central Kerguelen Plateau. Thus, first plume‐derived melts are generated at 118.75 Ma, in agreement with the onset of melting at the Southern Kerguelen Plateau (circa 120–110 Ma; Coffin et al, ; Duncan, ) and the Rajmahal Traps (118–115 Ma; Baksi, ; Coffin et al, ; Kent et al, ) and Sylhet Traps (118–116 Ma; Ghatak & Basu, ; Ray et al, ).…”

“…Following the methods described in detail in Gassmöller et al () and Bredow et al (), we set up a regional viscous flow model with the mantle convection code ASPECT (Bangerth et al, ; Heister et al, ) in order to explore the geodynamic history of the Kerguelen hotspot.…”

“…The viscosity in the model (equivalent to Bredow et al, ) is temperature dependent and depth dependent after Steinberger and Calderwood () but also takes into account a dehydration rheology to consider the sudden viscosity increase due to the extraction of water from olivine during melting. Additionally, a depletion buoyancy induces a density decrease when melt is extracted.…”

“…As shown in Bredow et al (), our model setup is not designed to handle melting in continental environments with high lithosphere thickness values (such as eastern Gondwana before ∼132 Ma). It also remains somewhat enigmatic as to what prevented the Kerguelen plume, if it already ponded underneath eastern Gondwana and produced the small‐volume igneous provinces described above, from vigorous volcanic activities as soon as the Indian Ocean basin started to open and placed thin oceanic lithosphere above the plume.…”