Do ridge segments with asymmetric and symmetric spreading have distinctive geochemical signatures? (Invited)
Abstract
The MARPEX expedition in summer 2012 intensively sampled 650km of the mid-Atlantic ridge between the Kane and Atlantis transform faults. Between Kane and Atlantis there are no additional transform faults, but 14 second order ridge segments of various types. Some segments (e.g. Broken Spur near 29'N) are symmetric with well-defined axial volcanic ridges and a strike perpendicular to the spreading direction. Some (e.g. TAG) are highly asymmetric with a well developed detachment fault on one side. Others (e.g. immediately south of Atlantis) have a detachment fault at one end, but become more symmetric in the middle of the segment. In other regions the apparent locus of spreading is not perpendicular to spreading direction, but rough terrain with no clearly defined neo-volcanic zone has an oblique orientation. In the south there is 70 km of oblique spreading with abyssal hills rotated 40 degrees relative to spreading direction. This region shows the simple plate tectonic paradigm of ridge segments and transform faults does not strictly apply to slow spreading ridges. The sampling campaign led to 180 new stations in this region, that have been combined with existing data from some 90 other stations to provide an unprecedented sample set from a long, normal section of slow-spreading ridge distant from hot spots. Sampling density is 2-3km along the entire length of ridge. All the samples are depleted MORB with no evidence of enriched basalts. The depletion is such that every sample has less Ba than the globally averaged N-MORB of Gale et al. (2013), and mean K2O at 8% MgO is only 0.1. These data show that small heterogeneities of enriched material are not ubiquitous. Within this depletion there is substantial diversity, with Zr/Y and La/Sm both varying by a factor of more than two, and a long wavelength signal of heterogeneity with maximum depletion in terms of both trace elements and isotopes occurring approximately mid-way between the two transform faults. The region also has a smooth long wavelength gradient in depth from 3000M in the north to 4000M in the south. This gradient provides a test of correlations among major element parameters and depth absent any hot spot effects. Na8 and Al8 increase southwards, and Fe8 decreases, consistent with the global correlations produced by mantle temperature variations. There is a clear chemical signature of symmetric segments with lineated abyssal hills, a defined neovolcanic zone, and shallow centers and deeper ends. These segments often produce olivine-plagioclase phyric basalts, and have lower Ti, Fe, Na and higher Ca. Most of the more complex segments erupt aphyric lavas with higher Ti, Fe, Na and lower Ca, all consistent with higher pressures of fractionation and a lack of a low pressure plumbing system. At the same time, variation in highly incompatible elements and ratios such as La/Sm and Zr/Y are lower at the magmatically robust segments, and higher at oblique and asymmetric segments. Being able to bracket these changes within the overall gradients in the region allows a clear separation of variables. Symmetric segments are associated with higher extents of melting, hence increased magma supply, and shallower fractionation systems that lead to the eruption of more phyric lavas. Differences in both melting and fractionation distinguish symmetric and asymmetric spreading.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2013
- Bibcode:
- 2013AGUFM.T21F..01L
- Keywords:
-
- 1021 GEOCHEMISTRY Composition of the oceanic crust;
- 8416 VOLCANOLOGY Mid-oceanic ridge processes;
- 1038 GEOCHEMISTRY Mantle processes;
- 8178 TECTONOPHYSICS Tectonics and magmatism