Spatial and temporal variations in crustal production at the Mid‐Atlantic Ridge, 25°N–27°30′N and 0–27 Ma

Abstract

We use high-resolution multibeam bathymetry, shipboard gravity, side-scan sonar images, and magnetic anomaly data collected on conjugate flanks of the Mid-Atlantic Ridge at 25°N-27°30′N and out tõ 27 Ma crust to investigate the crustal evolution of the ridge. Substantial variations in crustal structure and thickness are observed both along and across isochrons. Along isochrons within spreading segments, there are distinct differences in seafloor morphology and gravity-derived crustal thickness between inside and outside corners. Inside corners are associated with shallow depths, thin crust, and enhanced normal faulting while outside corners have greater depths, thicker crust, and more limited faulting. Across-isochrons, systematic variations in crustal thickness are observed at two different timescales, one at~2-3 Myr and another at >10 Myr, and these are attributed to temporal changes in melt supply at the ridge axis. The shorter-term variations mostly are in-phase between conjugate ridge flanks, although the actual crustal thickness can be significantly different on the two flanks at any given time. We observe no correlation between crustal thickness and spreading rate. Thus, during periods of low melt supply, tectonic extension must increase to accommodate the full plate separation rate. This extension commonly is concentrated in long-lived faults on only one side of the axial valley, resulting in strong across-axis asymmetries in crustal thickness and seafloor morphology. The thin-crust flank has few volcanic features and exhibits elevated, blocky topography with large-offset, often irregular faults, while the conjugate thicker-crust flank shows shorter-offset, regular faulting, and common volcanic features. The variations in melt supply at the ridge axis most likely are caused either by episodic convection in the subaxial mantle or by variable melting of chemically heterogeneous mantle.

Long‐Term Evolution of Nontransform Discontinuities at the Mid‐Atlantic Ridge, 24°N–27°30′N

“…Both IC and OC crust on the east ridge flank are dominated by large, irregular faults and elevated RMBA, while the west‐flank IC and OC crust both have lineated abyssal hills and RMBA up to 20–25 mGal lower. This difference has been interpreted to be a result of strong asymmetry in tectonic extension on the two ridge flanks (Wang et al, ).…”

“…We redefined the NTD between segments H and I compared to Tivey and Tucholke () and Wang et al () by dividing segment H into two parts. One segment (H2) extends from Chron 6bn or older to Chron 5an/5n.…”

“…We examined nonisostatic topography calculated by Wang et al () where multibeam bathymetry is available. Nonisostatic topography is derived from the bathymetric data by removing the predicted effect of plate cooling and a model of isostatically compensated topography in response to crustal thickness variations by assuming local Airy compensation.…”

“…Crust that forms the boundaries of these discontinuities exhibits a set of features that is considered to be characteristic (see Figure ): Inside‐corner (IC) crust, bounded by the spreading axis and the active offset, is typically elevated relative to surrounding crust and often has irregular fault patterns that may be oblique to the spreading axis. Outside‐corner (OC) crust is bounded by the spreading axis and the inactive trace of the offset; it is deeper, has regular, linear abyssal hills formed by inward dipping normal faults, and these abyssal hills often curve into the offset (Tucholke & Lin, ; Wang et al, ). In addition, ICs usually have elevated residual mantle Bouguer gravity anomalies (RMBA) and thus thinner crust compared to the conjugate OCs (Escartín & Lin, ; Tucholke & Lin, ).…”

Highly heterogeneous depleted mantle recorded in the lower oceanic crust

“…The finding of high-magnitude mantle heterogeneity delivered to the crust down to the centimetre scale requires that the MORB source is heterogeneous on a length scale smaller than the melting region. Variations of melt production have been observed along the northern MAR on a scale of 25-50 km 29 . If this length scale of variation is due to various proportions of mantle components in the source region, then the scale of individual compositional domains must be smaller.…”