Asthenospheric source of Neoproterozoic and Mesozoic kimberlites from the North Atlantic craton, West Greenland: New high-precision U–Pb and Sr–Nd isotope data on perovskite Sebastian Tappe a, ⁎, Agnete Steenfelt b , Troels Nielsen b a Department of Earth and Atmospheric Sciences, University of Alberta, 1‐26 Earth Sciences Building, Edmonton, Alberta, Canada T6G 2E3 b Geological Survey of Denmark and Greenland, Øster Voldgade 10, DK-1350, Copenhagen, Denmark abstract article info Article history: Received 17 January 2012 Received in revised form 17 May 2012 Accepted 29 May 2012 Available online 5 June 2012 Editor: L. Reisberg Keywords: Mantle reservoirs Redox melting Kimberlite origin Carbonatite evolution Craton reactivation ID-TIMS We present combined U–Pb, Sr, and Nd isotope data for small perovskite crystal fractions from kimberlites in West Greenland. Based on this high-precision TIMS data set, we revise the age range for kimberlite magma emplacement in the Sarfartoq and Tikiusaaq fields to 550–590 Ma and 158–166 Ma, respectively. These im- proved U–Pb perovskite age constraints reinforce the close temporal association of kimberlite and carbonatite magmatism across the North Atlantic craton. The new combined U–Pb, Sr, and Nd isotope data for perovskites provide evidence for kimberlite magma deriva- tion from a moderately depleted mantle source region during both the Neoproterozoic and Mesozoic. Moreover, we demonstrate that the difference in initial Sr–Nd isotope compositions between the Neoproterozoic Sarfartoq ( 87 Sr/ 86 Sr= 0.70278–0.70293; ε Nd =+1.6 to +3.6; n=13) and Mesozoic Tikiusaaq ( 87 Sr/ 86 Sr= 0.70319– 0.70346; ε Nd = +4.8 to +5.1; n = 3) kimberlite fields can be readily explained by isotopic evolution of a com- mon mantle reservoir. This mantle reservoir appears to have continuously participated in global crust–mantle differentiation and recycling, which points to the well-stirred convective upper mantle as the ultimate kimberlite magma source region beneath West Greenland. The apparent geographic shift of kimberlite and associated car- bonatite magmatic activity from the craton margin during the Neoproterozoic toward the craton center during the Mesozoic is explained by changes in localized, small-scale mantle flow along the underside of progressively thinning cratonic lithosphere. © 2012 Elsevier B.V. All rights reserved. 1. Introduction High-precision geochronology and radiogenic isotope tracer analy- ses are vital tools for the study of magmatic processes. Their combina- tion provides a powerful means to foster a better understanding of the relationships between magma generation and large-scale tectonic processes (Gibson et al., 2006; Tappe et al., 2007; Corfu and Dahlgren, 2008). It is now well established that the production of certain magma compositions relates to specific physicochemical conditions prevalent in the source region during melting and that these conditions are largely controlled by tectonic setting (Green et al., 1987; Pearce and Peate, 1995; Foley, 2011). However, some low-volume, deep-seated magma types such as kimberlites do not have straightforward relation- ships to geodynamic processes. For example, while some workers suggest a strong subduction influence in the genesis of kimberlites and related rocks (Sharp, 1974; Currie and Beaumont, 2011), others advocate a genetic link to mantle plumes (Le Roex, 1986; Haggerty, 1994; Torsvik et al., 2010; Rao and Lehmann, 2011) or continental rifting processes (Batumike et al., 2008; Tappe et al., 2008). In the study of kimberlite magma genesis it is therefore critical to combine geochronology and tracer isotope information to obtain further insight into the mecha- nism(s) responsible for the production of this economically important magma type. Historically, Sr and Nd isotope compositions were instrumental in the recognition of the involvement of two distinct mantle reservoirs during Mesozoic kimberlite magmatism beneath southern Africa (Smith, 1983). These first isotope data for kimberlites, which led to the original Group-I and II subdivison, were collected on carefully se- lected whole-rock powders. However, it is now widely accepted that the high degree of scatter commonly observed in kimberlite isotope data is primarily due to secondary processes including crustal contami- nation (Paton et al., 2007; Woodhead et al., 2009; Tappe et al., 2011a). In an earlier attempt to circumvent this problem and to resolve small isoto- pic differences, Heaman (1989) analyzed the Sr and Nd isotope composi- tion of the primary groundmass phase perovskite from a number of North American kimberlites by conventional ID-TIMS methods. He noted an overall decrease in data scatter and the less radiogenic 87 Sr/ 86 Sr of Chemical Geology 320–321 (2012) 113–127 ⁎ Corresponding author at: Institut für Mineralogie, Westfälische Wilhelms- Universität, Corrensstrasse 24, 48149 Münster, Germany. Tel.: + 49 251 8333470; fax: +49 251 8338397. E-mail address: sebastian.tappe@uni-muenster.de (S. Tappe). 0009-2541/$ – see front matter © 2012 Elsevier B.V. All rights reserved. doi:10.1016/j.chemgeo.2012.05.026 Contents lists available at SciVerse ScienceDirect Chemical Geology journal homepage: www.elsevier.com/locate/chemgeo