Evidence for small-scale convection in the North Atlantic lithosphere from geodynamic modeling

LIKERMAN JEREMIAS; ZLOTNIK SERGIO; LI CHUN-FENG

Castelldefells

Congreso; Subduction Interface Processes; 2017

The thermal evolution of the oceanic lithosphere (OL) has a key role in the heat transfer on Earth (Doin and Fleitout, 1996, McKenzie et al, 2005). The OL is generated at mid-ocean ridges and cools down as it moves, generating a conductive layer that gets denser and thicker with time. The cooling process was, for many years, explained in terms of two competing models: the half-space cooling (HSC) (Parson and Sclater, 1977) or the Plate Model (Stein and Stein, 1992). These models reproduce accurately thermal-derived seafloor topography and heat flux. However, for seafloor older than 70 Ma, HSC predictions differ substantially from these two observables and the Plate Model limits the growth of OL.In addition to the standard observables used to constrain the thermal state of the OL (seismic velocities, seafloor topography, heat flux, gravity and geoid) the thermal state of Li et al (2013) have provided a new estimation of the Curie depths in the North Atlantic by means of magnetic anomaly inversion with a fractal magnetization model. This measurement is important as it is independent from all previous datasets. The authors showed that Curie depths have a large oscillating and heterogeneous patterns related to SSC with approximately 500 km spacing, suggesting that their development are mainly in the upper mantle, and in preferred transverse rolls because of the slow spreading rates of the North Atlantic ridge. This is revealing, because unlike dominant longitudinal rolls previously suggested for SSC, the transverse rolls identified here were neglected on previous studies.Motivated by the work of Li et al (2013) and to further elucidate the interpreted transverse small-scale convective rolls from Curie isotherm oscillations, in this work we examine the effect of ocean ridge spreading rates on the development of TRs with numerical 3-D models and discuss the compatibility of our results and the geophysical observables.Our numerical simulations show that: a) using realistic laboratory-constrained rheologies and temperature it is possible to modify temperatures as low as those at Curie depths; b) transverse rolls are generated as well as longitudinal rolls on those isotherms; c) the spreading rate is a first order control on the developing of transverse rolls.