The Geochemistry of Basalts from the Iblean Hills (Sicily) and the Island of Linosa (Straits of Sicily): Evidence for a Plume from the Lower Mantle?

Kaj Hoernle GEOMAR Research Center, Wischhofstr. 1-3, 24148 Kiel, Germany

khoernle@geomar.de

Boris Behncke GEOMAR Research Center, Wischhofstr. 1-3, 24148 Kiel, Germany

Hans Schmincke GEOMAR Research Center, Wischhofstr. 1-3, 24148 Kiel, Germany

We have investigated the major and trace element and isotope geochemistry of 71 basalts from the Iblean Hills (Sicily) and the island of Linosa (Straits of Sicily). Both localities lie on the subducting African Plate, south of the boundary with the European Plate. The Iblean basalts form two groups: 1) a SiO2-undersaturated group (~1-7 Ma), which ranges from nephelinite through alkali basalt with SiO2 = 40-47 and MgO = 8-17, and 2) a tholeiitic group
(~2 Ma) with SiO2 = 51-54 and MgO = 6-8.5. The Pleistocene Linosa volcanic rocks range from alkali basalt and transitional tholeiite to trachyte. The Linosa basalts show a narrow range in composition with SiO2 = 47-49 and MgO = 6.5-9. The high Ni contents for a given MgO content for the Iblean basalts suggest that 1) they reflect primitive compositions which have undergone <10 weight percent fractionation of olivine +/- clinopyroxene and 2) the MgO contents of the samples are close to those of the parental melts. The Linosa basalts have lower Ni contents for a given MgO content reflecting higher degrees of differentiation than for all of the Iblean basalts. The higher SiO2 and Al2O3 but lower MgO, CaO, FeOt and incompatible element (e.g. Rb, Ba, Th, Nb, Ta, K, La, Ce, Sr, P, Nd, Sm, Zr, Hf) contents of the Iblean tholeiites as compared to the SiO2-undersaturated Iblean basalts could reflect either larger degrees of melting at shallower depths during formation of the tholeiitic melts and/or different source compositions for the two basalt groups. The Linosa basalts generally fall between or overlap one or both groups of Iblean basalts. The steep heavy rare earth element patterns for the Iblean and Linosa basalts indicate that melting occured within the garnet stability field and thus within the lowermost lithosphere or the asthenosphere. The undersaturated Iblean basalts have trace element characteristics similar to HIMU-type ocean islands, such as St. Helena. The abundances of incompatible elements in the Iblean tholeiites are similar to enriched mid ocean ridge basalts (E-MORB), yet the patterns of the tholeiites on multi-element diagrams are more similar to those of HIMU than
E-MORB basalts. The trace element characteristics of the tholeiites could be derived by mixing normal MORB
(N-MORB) with a HIMU-type component or the undersaturated Iblean basalts. The Linosa basalts have ocean island type trace element characteristics, which are distinct from the Iblean basalts. The limited Sr-Nd-Pb isotopic variation of Linosa basalts can be explained by mixing of a plume component similar to the low velocity composition (LVC) of Hoernle et al. (1995), a common plume endmember found throughout the eastern Atlantic, western Mediterranean and western and central Europe, with E-MORB located in the lithosphere or asthenosphere. In contrast to the Linosa data, the Iblean basalts show much larger variation in all three isotope systems and form well-defined correlations on
Sr-Nd-Pb isotope correlation diagrams, consistent with two component mixing. The undersaturated Iblean basalts have more radiogenic Sr and Pb but less radiogenic Nd than the tholeiites and require an endmember with 87Sr/86Sr „ 0.7032, 143Nd/144Nd ¾ 0.51289 and 206Pb/204Pb „ 19.95, similar to the ubiquitous LVC plume component in this region. The Iblean tholeiites, on the other hand, require an endmember with 87Sr/86Sr < 0.7028, 143Nd/144Nd „ 0.5132 and 206Pb/204Pb ~ 19.7, similar to the composition proposed for the lower mantle (FOZO component) by Hart et al. (1992). Although the tholeiite endmember could be derived through mixing of extremely depleted mantle with pure HIMU
(e.g. metasomatism of ancient subcontinental lithospheric mantle by a HIMU plume), He and Os isotope studies are underway (by Dave Graham and Liz Widom, respectively) to evaluate a possible lower mantle origin of this component.

References

Hart, S.R., Hauri, E.H., Oschmann, L.A. & Whitehead, J.A., Science 256, 517-520 (1992).

Hoernle, K., Zhang, Y-S. & Graham, D., Nature 374, 34-39 (1995).