The geochemistry of redox -sensitive trace metals

Ocean models have been used to explain the reduction in atmospheric carbon dioxide concentration during the last glacial maximum; however, some models result in an intensification of bottom water reducing conditions. Redox-sensitive metals in sediments are potential constraints on reducing conditions if their geochemical response under a range of conditions is well characterized. Authigenic sediment concentrations of vanadium, uranium, molybdenum rhenium and cadmium are higher under anoxic conditions relative to oxic conditions. The goals of this thesis are: (1) to determine the mobility of redox-sensitive metals under a variety of reducing conditions, (2) to quantify the importance of these sediments in the global marine mass balances, and (3) to apply our understanding of metal geochemistry to determine past redox conditions.;Redox-sensitive metals in sediments from the Northwest African and Northwest U.S. margins and the Arabian Sea suggest that where oxygen penetrates ≤1 cm (ZO2 ≤ 1 cm), V and Mo are released to seawater and Re, Cd and U are authigenically enriched in sediments. Under sulfidic conditions all of the metals are authigenically enriched. The area of sediments where ZO2 ≤ 1 cm is 4--6% of the ocean floor. Predicted variations over 20 kyr. for seawater Cd and V concentrations are 5--10% and 10--20% for a doubling and tripling, respectively, of the area of reducing sediments (ZO2 ≤ 1 cm). These concentration changes are within the uncertainties in Cd/Ca and V/Ca ratios. In foraminifera shells (12--15%). This implies that the area of reducing sediments has not been greater than 2--3 times its present value in the recent past.;The above calculations assume that metal authigenesis begins in sediments where ZO2 ≤ 1 cm. Subsequent authigenic rhenium measurements in Peru-Chile and Northwest African margin sediments (ZO2 = 2--3 cm) imply that rhenium authigenesis is not constrained to sediments where ZO2 ≤ 1 cm and accumulation based on this area is a conservative estimate.