ABSTRACT

Abstract Montesi(0649103)

Among the many outstanding characteristics that make our planet unique in the solar system, few are as puzzling as the presence of global plate tectonics. To explain the origins of plate tectonics and related seismic and volcanic hazard, the mechanics of plate boundary must first be elucidated. Mid-Ocean Ridges (MORs) are particularly important as the locus of the majority of terrestrial volcanism and plate separation. The abundant volcanism highlights the importance of magma generation and transport at MORs. Some global mantle convection models even suggest that melting at MORs is a key requirement of plate tectonics. However, volcanism is not continuous at the slowest MORs such as the recently explored Arctic ridges and the Southwest Indian Ridge (SWIR). How do these essentially avolcanic MOR work? We propose to build numerical models of mantle flow and the attending distribution, intensity, and geochemistry of volcanism at ultraslow ridges to elucidate the geodynamics of the slowest MORs.

Recent studies have exposed the many peculiarities of the slowest MORs. Volcanism is strongly reduced, localized onto discrete and widely spaced volcanic centers, and geochemical tracers that indicate limited melting and deep in-situ crystallization. Clearly, melt extraction, if not production, is severely limited in that environment. Slow spreading certainly results in a relatively cold mantle, which limits melting to great depth. However, two questions remain to be addressed: Why is volcanism localized? Is enough melt present at depth to facilitate plate divergence? To answer these questions, we will construct numerical models of mantle flow at ultraslow MORs, linking mantle upwellings to ridge geometry and intrinsic mantle flow instabilities, compute the thermal structure of the mantle at ultraslow ridges, and estimate the trajectory of magma and their chemical evolution. The along-axis variations in volcanic flux and basalt and peridotite chemistry produced by these models will be directly compared with data from the slowest MORs (SWIR, Gakkel Ridge) for which precise bathymetry is available and geochemical analysis is underway. Specifically, we will compare the modeled and observed intensity and spacing of localized volcanic centers, relate the location of these centers to ridge segmentation, compare the expected extent of melting and depth of crystallization with peridotite chemistry, and evaluate the importance of melting of a fertile eclogite component and along-axis focusing on basalt chemistry.

This research will produce advanced 3D geodynamical and chemical models of MORs. By puzzling out the geodynamics of the slowest MORs, this project weighs on the origins of plate tectonics and its proposed cessation when ridges slow down and the mantle cools sufficiently. We will address how melting, magma migration, and ridge segmentation are recorded in basalt and peridotite geochemistry. We will evaluate the role of melting and magma migration on ridge mechanics, using the ultraslow ridge environment, where melting is limited, as a stepping stone between well-studied melt-free tectonic settings and the majority of ridges in which melt is abundant.

Beyond training an early graduate student and advancing the careers of two junior faculty members, this project will improve the computation infrastructure by providing a novel, flexible toolkit, distributed online, to compute the geochemical signature and synthetic gravity fields from three-dimensional mantle flow models. This toolkit will be thoroughly documented and illustrated by a benchmark problem of corner flow at a MOR through a step-by-step tutorial using COMSOL MULTIPHYSICS. As some models will use the Underworld software package developed in Australia, this project involves an international collaboration.

Please report errors in award information by writing to: awardsearch@nsf.gov.