Crustal contamination in Palaeogene East Greenland flood basalts: plumbing system evolution during continental rifting

Abstract

Palaeogene central East Greenland flood basalts have Sr, Nd and Pb isotope characteristics consistent with crustal contamination of Icelandic-type basalts. In the Lower Basalts of central East Greenland, amphibolite facies gneiss is the dominant contaminant at low stratigraphic levels, whereas contamination with granulite facies gneiss prevails in the upper part of the lava succession. The contamination probably occurred by partial melting of the crust, and can be modelled by assimilation–fractionation–crystallisation processes. Similar contaminants have been detected in lavas from ODP Leg 152 off the SE Greenland coast, although in a reverse sequence. Contamination in Lower Basalts and ODP Leg 152 lavas, which erupted through an Archaean/Proterozoic substrate, differs from lavas in NE Greenland, which overly the Caledonian fold–belt and large Mesozoic sediment deposits. In the NE Greenland lavas, the dominant contaminant appears to be a dacitic component formed by partial melting of Mesozoic sediments. Changes in magma supply rates from the mantle may be the determining factor for the sequence of contaminants and may also control the depth of ponding in crustal chambers. The compositional variation of the least contaminated East Greenland basalts suggests that melts were produced from mantle sources similar to present-day Iceland and North Atlantic MORB at the time of continental breakup but distinction between mantle sources is very complicated. REE systematics indicate a maximum in initial depth of mantle melting at Kangerlussuaq during formation of the East Greenland flood basalts, consistent with a plume being centered in that region.