Using stable Mg isotopes to distinguish dolomite formation mechanisms: A case study from the Peru Margin

The magnesium isotope composition of diagenetic dolomites and their adjacent pore fluids were studied in a 250 m thick sedimentary section drilled into the Peru Margin during Ocean Drilling Program (ODP) Leg 201 (Site 1230) and Leg 112 (Site 685). Previous studies revealed the presence of two types of dolomite: type I dolomite forms at similar to 6 m below seafloor (mbsf) due to an increase in alkalinity associated with anaerobic methane oxidation, and type II dolomite forms at focused sites below similar to 230 mbsf due to episodic inflow of deep-sourced fluids into an intense methanogenesis zone. The pore fluid delta Mg-26 composition becomes progressively enriched in Mg-26 with depth from values similar to seawater (i.e. -0.8 parts per thousand, relative to DSM3 Mg reference material) in the top few meters below seafloor (mbsf) to 0.8 +/- 0.2 parts per thousand within the sediments located below 100 mbsf. Type I dolomites have a delta Mg-26 of -3.5 parts per thousand, and exhibit apparent dolomite-pore fluid fractionation factors of about -2.6 parts per thousand consistent with previous studies of dolomite precipitation from seawater. In contrast, type II dolomites have delta Mg-26 values ranging from -2.5 to -3.0 parts per thousand and are up to -3.6 parts per thousand lighter than the modern pore fluid Mg isotope composition. The enrichment of pore fluids in Mg-26 and depletion in total Mg concentration below similar to 200 mbsf is likely the result of Mg isotope fractionation during dolomite formation, The Mg-26 enrichment of pore fluids in the upper similar to 200 mbsf of the sediment sequence can be attributed to desorption of Mg from clay mineral surfaces. The obtained results indicate that Mg isotopes recorded in the diagenetic carbonate record can distinguish near surface versus deep formed dolomite demonstrating their usefulness as a paleo-diagenetic proxy. (C) 2014 Elsevier B.V. All rights reserved.