Asthenospheric flow and asymmetry of the East Pacific Rise, MELT area

Abstract

[1] Although the Pacific and Nazca plates share the East Pacific Rise (EPR) as a boundary, they exhibit many differing characteristics. The Pacific plate subsides more slowly and has more seamounts than the Nazca plate. Both the seismic and magnetotelluric components of the Mantle ELectromagnetic and Tomography Experiment (MELT) found pronounced asymmetry in mantle structure across the spreading axis near 17°S. The Pacific (west) side has lower S-wave velocities, exhibits greater shear wave splitting, and is more electrically conductive than the Nazca (east) side. These results suggest asymmetric mantle flow and melt distribution beneath the EPR. To better understand the causes for these asymmetric properties, we construct numerical models of melting and mantle flow beneath a midocean ridge migrating to the west over a fixed mantle. Although the ridge is migrating to the west, the migration has little effect on the upwelling rates, requiring a separate mechanism to create the asymmetry. Models that produce asymmetric melting with a temperature anomaly require large (>100°C) excess temperatures and may not be consistent with the observed subsidence and crustal thickness. A possible mechanism for creating asymmetry without a temperature anomaly is across-axis asthenospheric flow, possibly driven by pressures created by upwelling beneath the Pacific Superswell to the west. Pressure-driven asthenospheric flow follows the base of the lithosphere, extending the upwelling region to the west as it follows the thinning lithosphere toward the axis, and shutting off melting as it crosses the axis and encounters an increasingly thick lithosphere to the east.

Porosity‐driven convection and asymmetry beneath mid‐ocean ridges

“…In the mantle, one might expect larger gradients sustained over shorter distances; such heterogeneity is not considered here. Sharp gradients in mantle temperature were considered by Conder et al [2002] and Toomey et al [2002], but those studies calculated passive mantle flow only. Accounting for buoyancy forces would certainly have modified the mantle flows they computed and would have produced increased asymmetry in the melting region for equal forcing.…”

“…Taking the frame of reference of the migrating ridge axis, the mantle flow induced by ridge migration can then be calculated by imposing the ridge migration velocity as a boundary condition on the bottom of the domain, assumed to coincide with the base of the asthenosphere. Previous work has considered the effect of ridge migration on passive mantle flow and melting in two [Davis and Karsten, 1986;Schouten et al, 1987;Conder et al, 2002;Toomey et al, 2002;Katz et al, 2004] and three dimensions [Weatherley and Katz, 2010]. (11) and a, b constants from the best fitting line shown in Figure 3a.…”

“…Though the MELT region could be a special case, one expects a more general explanation for the observed asymmetry that requires only moderate external forcing. Furthermore, lateral temperature and melting rate gradients of the magnitude invoked by Conder et al [2002] and Toomey et al [2002] would yield sharp gradients in mantle density, and hence would likely cause buoyancy-driven flow, a scenario not considered by those studies. [8] Chemical and thermal buoyancy are known to be significant in the balance of stresses in the deeper mantle; they are the main driving forces for mantle convection and plate tectonics [e.g., Hager and O'Connell, 1981;Faccenna and Becker, 2010].…”

“…Together these studies provide a clear indication that in the MELT region at present, there is more mantle melting and residual mantle porosity on the west side of the southern EPR. [7] Two studies, Conder et al [2002] and Toomey et al [2002], have sought to model the observed asymmetry at the MELT region by imposing asymmetric boundary conditions on a passive flow model. Both invoke a combination of anomalously hot mantle beneath the Pacific plate, and a largescale pressure gradient driving mantle from west to east.…”

Thickening of young Pacific lithosphere from high-resolution Rayleigh wave tomography: A test of the conductive cooling model

“…Note that if either term is not required the coefficient will not be significantly different from zero. We allow for the difference in thermal structure of the Pacific and Nazca plates suggested by asymmetry in subsidence and other physical properties (Cochran, 1986;Conder et al, 2002;Toomey et al, 2002) by allowing a separate square root and linear term for each plate, for a total of 5 parameters at each period with 3 parameters representing the Pacific plate for which we have much better coverage. Due to the inherent smoothness of the polynomial, we use minimum length criteria damping, i.e.…”