Seismic boundaries in the mantle beneath Iceland : a new constraint on temperature.

To study the deep structure of Iceland, we conducted S receiver function analysis for almost 60 local broad-band seismograph stations of the Hotspot, ICEMELT and SIL networks. The structure was investigated separately for the central region of Iceland containing the neovolcanic zone and two peripher...

Full description

Bibliographic Details
Published in:Geophysical Journal International
Main Authors: Vinnik, L.P., Foulger, G.R., Du, Z.
Format: Article in Journal/Newspaper
Language:unknown
Published: Oxford University Press 2005
Subjects:
Online Access:http://dro.dur.ac.uk/14674/
http://dro.dur.ac.uk/14674/1/14674.pdf
https://doi.org/10.1111/j.1365-246X.2005.02529.x
Description
Summary:To study the deep structure of Iceland, we conducted S receiver function analysis for almost 60 local broad-band seismograph stations of the Hotspot, ICEMELT and SIL networks. The structure was investigated separately for the central region of Iceland containing the neovolcanic zone and two peripheral regions to the east and west. S-to-P converted phases from upper-mantle discontinuities were detected by stacking recordings of several tens of teleseismic events. The analysis reveals previously unknown details. Magnitude and depth extent of the low S velocity anomaly in the upper mantle beneath Iceland are much larger than reported in earlier studies. Clear S-to-P converted phases are obtained from the discontinuity at a depth of 80 ± 5 km, separating the high-velocity mantle lid from the underlying low S velocity layer. This discontinuity can be interpreted as a chemical boundary between dry harzburgite in the upper layer and wet peridotite underneath. Beneath peripheral parts of Iceland, we detect a boundary at a depth of 135 ± 5 km with S velocity increasing downwards. This boundary may correspond to the onset of melting in wet peridotite at a potential temperature of around 1400 °C. Models of melting induced by CO2 are not incompatible with our observations. The seismic data demonstrate effects that may be caused by azimuthal anisotropy in the upper mantle. There are indications of a second low S velocity layer to the NNE of Iceland, with the top near 480 km depth, similar to one recently detected beneath the Afro-Arabian hotspot.