Archetypal kimberlite from the Maniitsoq region, southern West Greenland and analogy to South Africa

tle nodules have been known for decades from the northern part of the Archaean block and adjacent Proterozoic terranes in southern West Greenland (Fig. 1; Escher & Watterson 1973; Goff 1973; Scott 1981; Larsen & Rex 1992; Mitchell et al. 1999). Some of the dykes have proved to be diamondiferous (see Jensen et al. 2004a, b, for exploration results, diamond contents, and references). The c. 600 Ma old dykes were called ‘kimberlitic’ by Larsen & Rex (1992), but Mitchell et al. (1999) concluded that they were best referred to a ‘carbonatiteultramafic lamprophyre’ suite (aillikites or melnoites). Mitchell et al. (1999) further suggested that the West Greenland province represents “one of the few bona fide examples of ultramafic lamprophyre which contain diamonds”. Reports on indicator mineral assemblages (Jensen et al. 2004b) and diamond contents (e.g. Hudson Resources Inc. 2005) have re-opened the discussion on the classification of the dykes. The results of an investigation of the Majuagaa dyke (Nielsen & Jensen 2005) are summarised below, together with the preliminary results of a regional investigation of the groundmass minerals of the dykes. It is concluded that dykes in the Maniitsoq region are similar to archetypal, South African, on-craton, Type 1 kimberlites, and that all regions of the West Greenland province of ultramafic magmatism are favourable for diamond exploration.


The Majuagaa dyke
The Majuagaa dyke (Jensen et al. 2004a) is 2.5 km long and up to 2 m wide. It is located c. 50 km SSE of Maniitsoq ( Fig.  1) and strikes WSW-ENE. The dyke is dark grey with many olivine-rich fragments (up to 10 cm) and rounded megacrysts of ilmenite (up to 4 cm). It contains the classic kimberlitic suites of megacrysts and mantle nodules, including eclogite (Jensen & Secher 2004, fig. 5). The groundmass is fine-grained and composed of olivine fragments, calcite, serpentine, ilmenite and minor Mg-rich spinel. Phlogopite and apatite are rare. The dyke is diamondiferous (Jensen et al. 2004a).
Samples were collected along the length of the dyke. Sixty thin sections (Fig. 2) were examined and a number selected for an electron microprobe study. All mineral data from groundmass, megacysts and nodules, the bulk chemistry, and analytical techniques are reported in Nielsen & Jensen (2005). Mitchell (1995) and Tappe et al. (2005) use the following criteria for the classification of kimberlite (s.s.): (1) the groundmass contains no clinopyroxene; (2) groundmass spinel belongs to the Magmatic Trend 1 (Mg-rich titanomagnetite);

Classification of the Majuagaa dyke
(3) phlogopite is zoned towards the Al 2 O 3 -and BaO-rich kinoshitalite endmember and (4) ilmenite has a high geikilite component (> 40 mol.% MgTiO 3 ) and little pyrophanite (MnTiO 3 ). Mitchell et al. (1999) found that these criteria were not met by the West Greenland dykes and concluded they were ultramafic lamprophyres (aillikites or melnoites) rather than kimberlites. Nielsen & Jensen (2005) made the following observations in the Majuagaa dyke: • The clinopyroxene criteria: No clinopyroxene was found in the groundmass. • The Magmatic Trend 1 spinel criteria: The cores of euhedral spinel grains (< 0.1 mm across) have compositions in the Magmatic Trend 1 field (Fig. 3). Mg-rich rims compare with spinels of South African calcite-kimberlite (Mitchell et al. 1999). • The ilmenite criteria: Groundmass grains conform with the compositions from archetypal kimberlite (Fig. 4), whereas megacrysts appear to be xenocrystic (Nielsen & Jensen 2005). • The phlogopite criteria: Tiny, euhedral, clear to weakly greenish flakes are rich in Al 2 O 3 and BaO (Fig. 5), poor in TiO 2 and FeO (total) and rich in the kinoshitalite end member. They conform with phlogopite of archetypal kimberlite (see Mitchell 1995).

Majuagaa bulk composition
The bulk composition of the Majuagaa dyke is kimberlitic (see Nielsen & Jensen 2005). The average REE ( Rim compositions to the left of the field are characteristic of calcite kimberlites (Mitchell 1995). All other fields after Mitchell (1995).

Regional variations
The West Greenland province is part of the c. 600 Ma old North Atlantic province of carbonatite and ultramafic alkaline magmatism, from the Torngat region (Canada) to the Archaean of West Greenland (Tappe et al. 2004). The compositions of groundmass phlogopite reflect the compositions of the melts (Mitchell 1995). The preliminary results of a regional investigation (Fig. 8) suggest a gradual evolution from oncraton, South African Type 1 kimberlite in the West Greenland Archaean craton (Maniitsoq), through a kimberlite/ultramafic lamprophyre (kimberlite/aillikite) zone at the border between Archaean and Proterozoic terranes (Sarfartoq and along the Kangerlussuaq fjord, Fig. 1) to ultramafic lamprophyre (aillikite/melnoite) magmatism in the Proterozoic terranes of Sisimiut (Greenland; Scott 1977) and Torngat (Canada; Tappe et al. 2004).

The diamond potential
Results of Hudson Resources Inc. (2005) suggest that stones of gem quality and size may be found in West Greenland.
Nevertheless, it appears to be an issue for some exploration companies and investors that Mitchell et al. (1999) classified the West Greenland dykes as ultramafic lamprophyres and implied that true kimberlite was not found. However, the Majuagaa dyke documents that diamondiferous, archetypal Type 1 kimberlite occurs in the West Greenland province. Hutchison (2005) describes the best investigated and most promising West Greenland diamond occurrence at 'Garnet Lake' (border zone; Sarfartoq region, Fig. 1). The 'Garnet Lake' dykes have characteristics of both kimberlite and ultramafic lamprophyre and have features reminiscent of South African orangeite. Finally, the ultramafic lamprophyres (aillikites/melnoites) of the Proterozoic Sisimuit region compare with diamondiferous ultramafic lamprophyres (aillikites) of the Torngat region ( Fig. 8; Tappe et al. 2004), and a diamond potential is also indicated in the little prospected Sisimiut region.

Conclusions
The c. 600 Ma old ultramafic magmatism of the West Greenland province shows -from the Archaean craton to the Proterozoic terranes -a transition from classic South African, on-craton Type 1 kimberlite to ultramafic lamprophyre (aillikite/melnoite). Diamonds are recovered from the entire range and a diamond potential thus exists throughout the West Greenland province. regions. The Sisimiut data (Scott 1977) includes one locality referred to the border zone.
Until recently, in situ U-Pb zircon geochronology could be carried out only using ion microprobes, requiring lengthy analysis times of c. 20 minutes. However, new developments in laser ablation inductively coupled plasma mass spectrometer technologies have resulted in zircon geochronology techniques that are much faster, simpler, cheaper, and more precise than before (e.g. Frei et al. 2006, this volume). Analyses approaching the precision obtained via ion microprobe can now be undertaken in 2-4 minutes using instruments such as the 213 nm laser ablation (LA) system coupled with Element2 sector-field inductively coupled plasma mass spectrometer (SF-ICP-MS) housed at the Geological Survey of Denmark and Greenland (GEUS). The up to tenfold decrease in analytical time means that zircon geochronology can now be used in a much wider range of studies. The Godthåbsfjord region, southern West Greenland, contains some of the oldest rocks exposed on the Earth's surface reflecting a very complex Archaean geological evolution (Figs 1, 2). Over recent years GEUS has undertaken a range of mapping projects at various scales within the Godthåbsfjord region (see also below). These include the mapping of the 1:100 000 scale Kapisillit geological map sheet (Fig. 1), and regional and local investigations of the environments of formation and geological evolution of supracrustal belts, hosting potentially economic mineral occurrences.
Zircon geochronology is an important tool for investigating a range of geological problems in this region. By breaking down the complex geology into a series of simple problems that can be addressed using this tool, the geological evolution can be unlocked in a stepwise manner. Three examples are presented below: (1) the mapping of regional structures; (2) characterising and correlating supracrustal belts; and (3) dating metamorphism and mineralisation. Although focus is on the application of zircon geochronology to these problems, it is important to note that the resulting data must always be viewed within a wider context incorporating geological mapping and structural, geochemical and petrographic investigations.

Regional geology
The geology of the Godthåbsfjord region is dominated by orthogneiss formed during several distinct episodes of crustal growth during the Archaean (Fig. 2). These different-aged gneisses are thought to represent distinct small continental blocks that were amalgamated during the Neoarchaean (at