U-series data of recent volcanism at an Axial Volcanic Ridge

The spreading rate of the Mid Atlantic Ridge, at the area around 45°N is 11mm/y. This is appropriate scaling for mantle horizontal flow and vertical upwelling. However, the accretion of oceanic crust within the median valley graben (MVG), occurs episodically on timescales much longer than years. During the 2008 JC024 cruise various geophysical and bathymetry datasets were collected as well as photographic and video evidence and some 270 rock samples, using ship-borne instruments, a deep-towed platform TOBI and a tethered ROV Isis both in demersal mode and 'flying' at 100m above the seafloor. We could identify the overall 'hour-glass' shape of the MVG, the location of the AVR, 'flat-top' features and some smooth areas on either side. Overall We counted some 8000 conical or dome-shaped hummocks, which are arranged in lineaments, mostly parallel to the spreading axis. We deduced that each hummock is an individual monogenetic volcano. Smooth areas between the AVR and the median valley boundary faults may indicate sheet flows which are probably burying the subsiding hummocks. For Dive 91, We used a MS2000 high-resolution multi-beam bathymetry echo-sounder on Isis flying at an altitude of 100m above the seafloor; to construct a bathymetry map with a vertical resolution of 20cm and a horizontal precision of 5m over an area of ~3km 2 . The high-resolution bathymetry allows us to deduce that volcano dimensions average around 300m diameter, ~150m altitude, and 0.005km 3 volume and to count ~100 volcanoes. Visual observations using the camera systems on Isis showed that individual volcanoes are essentially piles of pillow lavas, usually ~1m diameter and >2m long of various types, sometimes leaving >10% gaps between pillows, sometimes interlocking somewhat deformed pillows. Small protrusions, 10-50cm long, which are numerous on some pillows, could be fairly easily broken off using the pincers on the hydraulic arms of Isis, and returned to the surface. Uncertainty remains over the age of the AVR, or probably more appropriately, the time-span between to youngest and the oldest exposed rocks. Many indirect dating methods have been applied with inferred ages ranging from ~10ky to 200ky. We applied the magnetic paleo-intensity method to infer an age of ~12ky for the 45°N AVR. We will report U-series based age calculations for samples from the Dive 91 area. The samples were taken from a number of individual volcanoes and the simplest expectation would be that the youngest samples are from the AVR crest with older ages down the flanks. However, crustal magnetisation intensity can be taken as a proxy for age and this indicates a more complex pattern of young ages in the volcanic lineaments away from the crest. Indeed, visual observations of rocks at up to 1km from the crest would indicate that those are equally fresh. Indeed, Standish and Sims have shown that young eruption ages are broadly dispersed throughout the rift valley of the ultra-slow spreading Southwest Indian Ridge at a range much wider than covered by the area of Dive 91. Construction of new volcanic crust solely at the crest of the AVR may well be too simplistic.