The mineralogy, geochemistry, and petrogenesis of the grønnedal-íka alkaline igneous complex, south-west Greenland

Abstract

At 1299±17 Ma (Blaxland et al, 1978), the Grønnedal-Íka is the oldest of the Gardar centres, situated in the extreme north-west of the province. Hare-earth clement (REE) trends suggest that the nepheline-syenites which make up the bulk of the complex were derived from a parental magma formed by a few percent of partial melting of a garnet-lherzolite mantle source during an episode of rifting in the Early Gardar. In contrast to other undersaturated Gardar centres, the syenites of Gronnedal-l'ka show some striking raineralogical differences. The scarcity of amphibole and lack of olivine indicate a magma with a relatively high oxygen fugacity. Opaque oxide compositions and pyroxene trends provide further support for this idea. Additionally, the occurrence of zircon in all units of the complex is unusual, and is probably related to post-magmatic alteration processes. Fractionation of apatite and zircon appears to have been responsible for the observed variations in REE content, although later iteration and variations in the composition of the inter- cumulus liquid have given rise to a considerable scatter in major and trace element abundances. Normative compositions show the evidence for the development of a 'sandwich' horizon in both the Lower and Upper Series. At a later stage, a plug of xenolithic syenite was intruded, which was followed by a the em placement of a body of xenolithic carbonatite, containing fragments of the earlier syenites. This unit is predominantly s0vitic, but with increasing fractionation, more iron-rich (ferrcarbonatite) compositions were developed. Compared to many carbonatites, the rock at Grønnedal is rather poor in 'exotic' minerals. Trace element abundances, however, show extreme enrichment in Sr, Th, REE's, and Y, and depletion in Zr, Ti, and K compared to the syenites. These variations are comparable to the observed concentrations in the Igaliko carbonatite dykes (Pearce, 1988). Hf, Ta, and REE distributions between the carbonatite and syenitic rocks suggest that the carbonatite was derived by liquid immiscibility from a Co(_2)-saturated phonolitic magma, with the conjugate silicate phase possibly intruded as the Xenolithic Porphyritc Syenite. Patchy metasomatic alteration has affected all units, and has given rise to the Coarse- Grained Brown Syenite, which occurs in both the Lower and Upper Series. More intense alteration has affected the syenites, giving carbonate-rich 'carbosyenites', and xenoliths of country rock within the syenites and carbonatite; the surrounding country-rock is not as severely affected as might have been expected. The presence of zircon, alkali mafics, sodalite veins, and recrystallisation of feldspar in the altered rocks is attributed to the effects of peradkaline. C1 and C0(_2)-rich late-stage fluids derived from both the syenites and carbonatite.