THE SUDBURY IMPACT LAYER AT GUNFLINT LAKE, MINNESOTA: INFERENCES FROM GEOCHEMISTRY ABOUT EJECTA SOURCES AND DELIVERY MECHANISMS

Ten exposures of the 1850 Ma Sudbury impact layer (SIL) have been mapped along a 15 mile strike-length near Gunflint Lake, ~750 kilometers west of the impact site at Sudbury, Ontario. The exposures overlie the Paleoproterozoic (ca. 1878 Ma) Gunflint Iron Formation, and are overlain by shale of the Rove Formation (ca. 1835 Ma). The SIL stratigraphic section consists of underlying disrupted iron-formation 1-7 m thick, overlain by a meter or less of ejecta in irregular layers and lenses of mesobreccia, gritstone, and lapillistone (in more or less stratigraphic order). Alteration and metamorphism have obscured much of the diagnostic petrographic attributes. Nevertheless, scanning electron microscopy and geochemistry can be used to characterize the rocks, differentiate between ejecta-bearing and ejecta-absent materials, and speculate about possible source-rock contributions to the ejecta. The ejecta presumably contain components of the impactor, target, and material entrained during transport and deposition. Analyses of samples from a single outcrop (G730) containing a rather complete stratigraphic section of ejecta layers are used to tackle the problem of defining the compositions and proportions of source rocks. The oxide and REE contents of ejecta can be modeled as mixtures of an impactor, represented by a C1 chondrite; target rocks, represented by various analyses of continental crust; and subjacent iron-formation. Overall, the analyses of ejecta-bearing layers differ significantly from substrate. In fact, SiO₂, Al₂O₃, and MgO contents in ejecta are more consistent with derivation from a combination of target and impactor components. The Al₂O₃ and REE contents of the basal ejecta-bearing mesobreccia are significantly larger than those in underlying iron-formation and the overlying ejecta layers. This may reflect arrival of the earliest ejecta exhumed from the shallowest crustal levels of the target. Our selection of proxy compositions suggests that an iterative mixing procedure may eventually establish the probable proportions of the most likely source-rocks. Evaluation of source-rock contributions to individual ejecta layers has the potential to refine the understanding of the production, entrainment, enroute sorting, and deposition of ejecta.