Constructing the upper crust of the Mid‐Atlantic Ridge: A reinterpretation based on the Puna Ridge, Kilauea Volcano

Publisher Website
Google Scholar
Abstract
The volcanic morphology of a number of segments of the slow spreading Mid‐Atlantic Ridge (MAR) have been reinterpreted based on our understanding of dike emplacement, dike propagation, and eruption at the East Rift Zone of Kilauea Volcano, Hawaii and its submarine extension, the Puna Ridge. The styles of volcanic eruption at the submarine Puna Ridge are remarkably similar to those of the axial volcanic ridges (AVRs) constructed on the median valley floor of the MAR. We use this observation to relate volcanic processes occurring at Kilauea Volcano to the MAR. We now consider that volcanic features (e.g., seamounts and lava terraces) built on the flanks of the AVRs are secondary features that are fed from lava tubes or channels, not primary features fed directly from an underlying dike. We examine simple models of pipe flow and conclude that lava tubes can transport lava down the flanks of submarine rifts to build all of the volcanic features observed there. In addition, deep water lava tubes are strong enough to withstand the pressures of a few megapascals that the building of a volcanic structure 150 m high at the end of the tube would generate. The volumes of individual volcanic terraces and seamounts on the Puna Ridge and at the MAR are large (0.1–1 km3) and similar to the volumes of lava flows that are broadly distributed at the subaerial East Rift Zone of Kilauea. This striking difference in the volcanic morphology on a scale of 1–2 km (producing terraces and seamounts underwater and low‐relief flows on land) must be related to the enhanced cooling and to the greater mechanical stability of tubes in the submarine environment. We suggest that at the MAR a crustal magma reservoir, most likely located beneath shallow, flat sections of the segment, provides magma to the rift axis through dikes that propagate laterally tens of kilometers. The zone of dike intrusion, at least in the neighborhood of the magma body, is likely narrower than the width resurfaced by flows, yielding a crustal structure that has a rapid vertical transition from lavas to sheeted dikes. At segment ends the zone of dike intrusion is likely to be wider, giving a resulting structure with a more gradual transition from lavas to dikes.