HR: 1330h
AN: V32A-1006    [PDF]
TI: Origin Of Extreme $^{3}$He/$^{4}$He Signatures In Icelandic Lavas: Insights From Melt Inclusion Studies
AU: * Harlou, R
EM: rh@dlc.ku.dk
AF: Department of Geological Sciences, University of Durham, Science Laboratories, South Road, Durham, DH1 3LE United Kingdom
AU: * Harlou, R
EM: rh@dlc.ku.dk
AF: Danish Lithosphere Centre, Oester Voldgade 10, Copenhagen K, 2100 Denmark
AU: Kent, A J
EM:
AF: Department of Geociences, Oregon State University, Wilkinson Hall, Corvallis, OR 97331 United States
AU: Breddam, K
EM:
AF: Danish Lithosphere Centre, Oester Voldgade 10, Copenhagen K, 2100 Denmark
AU: Davidson, J P
EM:
AF: Department of Geological Sciences, University of Durham, Science Laboratories, South Road, Durham, DH1 3LE United Kingdom
AU: Pearson, D G
EM:
AF: Department of Geological Sciences, University of Durham, Science Laboratories, South Road, Durham, DH1 3LE United Kingdom
AB: Helium isotopes are considered a powerful tool for tracking different mantle domains. Yet, the origin of He isotope variations in many basaltic suites remains enigmatic and often difficult to link with more lithophile chemical and isotopic tracers. One problem is that He isotope ratios are measured from crushed olivines and thus reflect prior fluid and melt fluxes trapped in inclusions within the olivine grains, whereas the lithophile elements mainly reflect the host lava. In an attempt to link He and lithophile element variations, we have characterized the major and trace element composition including volatile elements, of olivine-hosted melt inclusions from three ankaramitic lavas from Vestfirdir, NW-Iceland. Previous studies have reported extreme $^{3}$He/$^{4}$He ratios from NW-Iceland and one ankaramite (SEL97) has been suggested to provide the most precise estimate of the radiogenic (Sr-Nd-Pb) isotopic composition of a relatively undegassed (high $^{3}$He/$^{4}$He) mantle component (C or FOZO) common to several ocean islands (Hilton et al. 1999, EPSL 173, 53-60). The samples investigated here exhibit amongst the highest $^{3}$He/$^{4}$He ratios observed in terrestrial rocks (42.9 and 34.8 R/R$_{a}$). A detailed account of the trace element signature of melt inclusions in these samples may thus help explain the origin of FOZO. One sample of similar composition to these, has a lower He content and a relatively poorly defined He isotope composition of $8.15 \pm 5.1$ R/R$_{a}$ (Breddam \& Kurz, 2001, EOS, 82, F1315). In terms of major elements, the whole rock data reflect olivine accumulation, whereas the melt inclusion data reflect ol + cpx fractionation. The melt inclusions are generally basaltic (Mg\#: 52-62), with primitive mantle normalised trace element concentrations that are broadly parallel the host lavas. There is little compositional difference between melt inclusion populations from high and low $^{3}$He/$^{4}$He lavas, although inclusions of the low $^{3}$He/$^{4}$He lava have lower S and moderately lower Cl. The observed range of trace element ratios: [La/Sm]$_{N}$ 1-4, [La/Yb]$_{N}$ 1-5, Sr/Nd 14-24, Ba/Rb 9-23, and Ce/Pb 5-46, covers much of the range observed in Icelandic alkali basalts. The compositional similarities between inclusions and host lavas suggests that bulk rock compositions are petrogenetically related to the melts sampled by melt inclusions. If He predominantly resides in these inclusions, it suggests that the whole rock composition is an aggregate derived from the same melts that contain the measured He.
DE: 1749 Volcanology, geochemistry, and petrology
SC: Volcanology, Geochemistry, Petrology [V]
MN: 2003 Fall Meeting