Residence time of osmium in the oceans

Abstract

Listen

Translate article

Estimates of osmium residence time in the oceans that are based on oceanic mass balance calculations (35–50 kyr) appear irreconcilable with those inferred from the recent evolution of the osmium isotope composition of seawater (3–4 kyr). It is argued that the osmium budget of the oceans is currently close to steady state and thus that the estimates made by the two methods should agree. As the inventory of osmium in the oceans is relatively well constrained, these disparate residence time estimates imply wildly different osmium input fluxes to the oceans. An osmium residence time of 8–10 kyr is proposed by evaluating the uncertainties and limitations of both methods, and it is argued that osmium inputs to the ocean are currently underestimated by a factor of ∼3. This reflects in part the underestimation of the river input of osmium to the oceans owing to a bias within the existing data set and in part the probable existence of sources of osmium to the oceans that have not yet been identified. The very short residence time of 3–4 kyr inferred from the postglacial change in seawater composition (assuming a single step change in input flux) is rejected as it implies unreasonably high osmium input fluxes to the oceans. It is concluded that a postglacial spike in osmium flux, associated with a meltwater event, must have driven part of the change in seawater composition. However, it is also shown that such a spike cannot be the dominant cause of the most recent shift in seawater 187Os/188Os.