Abstract
AbstractConvection in fluid layers at high Rayleigh number (Ra ∼106) have a spoke pattern planform. Instabilities in the bottom thermal boundary layer develop into hot rising sheets of fluid, with a component of radial flow toward a central upwelling plume. The sheets form the “spokes” of the pattern, and the plumes the “hubs.” Such a pattern of flow is expected to occur beneath plate interiors on Earth, but it remains a challenge to use observations to place constraints on the convective planform of the mantle. Here we present predictions of key surface observables (gravity, topography, and rates of melt generation) from simple 3‐D numerical models of convection in a fluid layer. These models demonstrate that gravity and topography have only limited sensitivity to the spokes and mostly reflect the hubs (the rising and sinking plumes). By contrast, patterns of melt generation are more sensitive to short‐wavelength features in the flow. There is the potential to have melt generation along the spokes but at a rate which is relatively small compared with that at the hubs. Such melting of spokes can only occur when the lithosphere is sufficiently thin ( km) and mantle water contents are sufficiently high ( ppm). The distribution of volcanism across the Middle East, Arabia, and Africa north of the equator suggests that it results from such spoke pattern convection.
Highlights
What is the planform of mantle convection? The largest, and most obvious, planform in the convection system is that associated with plate motions, which involves horizontal scales as large as 10,000 km
The numerical experiments described above show that the observed gravity and topographic anomalies are reproduced by the simplest isoviscous fluid dynamical model of thermal convection
The correspondence between the volcanism and the gravitational and topographic anomalies in NE Africa is striking, and shows that they all result from the convective circulation
Summary
What is the planform of mantle convection? The largest, and most obvious, planform in the convection system is that associated with plate motions, which involves horizontal scales as large as 10,000 km (e.g. the Pacific plate). The approach taken here complements a popular alternative approach to modelling gravity and topography using information from seismic tomography (Hager & Richards, 1989; Flament et al, 2013) In such studies, estimates of density variations within the mantle are inferred from tomography and used to make predictions of gravity and dynamic topography. Our approach departs from the common assumption of many Earth Scientists, who believe the convective planform of mantle convection consists solely of plumes and the plates This assumption arises from the work of Wilson (1963) and Morgan (1971), who showed that the relative motion between major volcanic centres beneath plate interiors was sufficiently slow that they could be used to define a single world-wide reference frame. An appendix provides further technical details on the simulations and data processing
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
