Publication Type:Journal Article
Source:Mineralium DepositaMineralium Deposita, Volume 46, Number 8, p.943-958 (2011)
Keywords:Canada, KOMATIITE, magmatic sulfides, quebec, Thermal erosion
Virtually all of the economic Ni–Cu–(platinum group element (PGE)) mineralization in the central part of the Cape Smith Belt of New Québec is hosted by thick olivine cumulate units in the Katinniq Member of the Raglan Formation at the base of the 1.9 Ga Chukotat Group. These units transgress underlying gabbros and pelitic metasediments, forming 50–200-m deep and 300–1,000-m wide V-shaped embayments and have been interpreted on the basis of surface geology, deep diamond core drilling, and magnetic inversion models to represent the remnants of one or more large, long (at least 20 km, possibly ≥50 km), sinuous, komatiitic basalt lava channels that formed by thermomechanical erosion of their substrates. We have used a mathematical model to test these hypotheses regarding komatiitic lava emplacement and erosion by lava. Our modeling predicts that an initially 10-m thick komatiitic basalt flow should have flowed turbulently near the vent and should have thermomechanically eroded unconsolidated pelitic sediment during emplacement to reach the observed degree of contamination of ≤10% at distances of ~30–60 km downstream from the source. Furthermore, our models predict that, at these distances downstream, a fully inflated 100-m thick komatiitic basalt flow would have had thermal erosion rates over consolidated gabbroic substrate of ~0.7–1.5 m/day, requiring ~70–140 days to incise a 100-m deep channel, depending on the initial temperature of the lava, the paleoslope, and the initial temperature and solidus temperature of the gabbro. These erosion rates would have been associated with volumetric flow rates of >105–106 m3/s and eruption volumes of >103–104 km3. Although these flow rates are orders of magnitude larger than those of most modern terrestrial basaltic flows, they are of the same order as those estimated for the largest terrestrial flood basalt flows and with those inferred for some of the largest extraterrestrial flows. Our predicted flow volumes are also of the same order as those of the largest terrestrial flood basalt units, consistent with the great thickness and widespread distribution of the Chukotat Group. Our modeling of thermomechanical erosion of gabbro by komatiitic basalt results in negligible contamination (<1%), and geochemical studies show that the spatially and petrogenetically related Chukotat basalts are uncontaminated, suggesting that the observed enrichments in U–Th–Light Rare Earth Elements (REE) > Middle REE–Heavy REE > Nb–Ta–Ti represent contamination by underlying Povungnituk semipelites. This result is consistent with present models for the genesis of the Ni–Cu–(PGE) mineralization in the Raglan Formation that involve thermomechanical erosion of unconsolidated, sulfidic semipelitic sediments, and decoupling of the miscible silicate and immiscible sulfide components.