Abstract:
The Siberian Traps represent the largest known Phanerozoic manifestation of continental flood volcanism. The general agreement is now that the volcanism was of short duration (# 1 m.y.) and was synchronous with the Permian-Triassic boundary. Discrepancies between different geochronologic results prevent rigorous evaluation of the age and duration of magmatism.Several alkalic mafic intrusions associated with the Siberian Traps represent subvolcanic feeders for the Trap lavas. The Noril'sk 1 intrusion is intruded into the earliest flows and is chemically and isotopically related to stratigraphically overlying flows. Biotites from the Noril'sk 1 intrusion have previously yielded40Ar39Ar plateau dates (254-251 Ma) that are significantly older than seemingly reliable 40Ar39Ar plateau dates (250-247 Ma) on plagioclase and whole-rock samples from Trap basalts that are geologically constrained to be older than the intrusion. Possible explanations of this discrepancy include: (1) plagioclase and whole-rock samples are anomalously young due to 40Ar loss via alteration or episodic outgassing, and/or (2) the biotites contain excess 40Ar.Detailed laser heating 40Ar39Ar analysis of six individual grains of biotite in a total of 214 steps yield data that permit characterization of multiple components of Ar through isotope correlation patterns. An isochron age of 250.1 +/- 1.5 Ma (2σ) and an excess Ar component with 40Ar36Ar = 565 +/- 39, are obtained by regression of the 115 highest temperature steps. Step heating of a bulk hornblende sample also reveals excess Ar, with40Ar36Ar = 327 +/- 10 and an isochron age of 249.3 +/- 1.6 Ma. Neglecting uncertainties in the ages of neutron fluence monitors, the two dates just overlap at 2σ, but are distinctly younger than the previous mean plateau date (253.9 +/- 1.6 Ma) for biotite. The different initial 40Ar36Ar for the two phases is interpreted as reflecting the evolution of an initial lower mantle composition towards increasing crustal contamination.