A Multi-technique Approach for Provenance Studies of Mesozoic Clastic Rocks in the Barents Sea
Friday, 18 December 2015
Poster Hall (Moscone South)
Sediments of Mesozoic age deposited in the Arctic Hammerfest and Tromsø basins (southern Barents Sea) are the focus of a comprehensive provenance study which forms part of ongoing work by the LoCrA consortium (Lower Cretaceous Basin Studies in the Arctic). Jurassic (Stø, Fuglen, Hekkingen) and Cretaceous (Knurr, Kolje, Kolmule) formations were sampled from seven wells. Analytical methods include petrography, whole-rock geochemistry, heavy mineral (HM) analysis and U-Pb on detrital zircons. HM concentration is <1%, with an ultrastable HM assemblage of zircon, rutile, tourmaline, spinel, apatite, garnet, chloritoid and common authigenic heavy minerals and opaques. Cretaceous sedimentary rocks show geochemical variations that reflect an unrecycled Upper Continental Crust signature, whereas Jurassic detritus tends to show more evidence of recycling with a relatively low input of mafic material. Kolmule Formation whole-rock geochemistry indicates sediment recycling from a major Sc-depleted but intermediate to mafic source and hence suggests input of rift detritus from the syn-depositional opening of the Atlantic Ocean. U-Pb ages for detrital zircons for Cretaceous sandstones show age groups of 200-500 Ma, 1200-1800 Ma, and 2100-2800 Ma, indicating potential source regions in the Urals/Novaya Zemlya, the Caledonides, Grenvillian/Sveconorwegian, and Palaeoproterozoic and Archean sources. Provenance data via geochemistry and HM analysis indicate different sources in the same formation basin-wide, with a significant change in provenance and sediment composition from Jurassic to Cretaceous and between the Knurr and Kolmule formations. These differences in composition need to be compared to detailed single grain studies and may only be explained in terms of basin dynamics, or even on a smaller scale, in terms of facies distributions. If so, this case study raises concerns about the use of single samples for provenance models on a larger scale.