Global distribution of the 230Th flux to ocean sediments constrained by GCM modelling

We have introduced a simple particle field into an existing and well-documented ocean general circulation model. This enables us to investigate the advection and scavenging of particle-reactive species within the water column. As a first use of this model, we have assessed the advection and flux to sediment of 230Th, a nuclide with a well understood marine chemistry that exhibits extreme particle reactivity. The flux to sediment of this nuclide is of interest as it is widely assumed to be related only to water depth, and therefore to act as a constant-flux indicator for marine sediments. By assuming an average settling velocity for marine particles of 3 m/d, in good agreement with observational constraints, the model generates a particle field close to that observed. Thorium-230 is scavenged onto this particle field reversibly according to Kd values constrained by observations and incorporating a particle-concentration effect. This scavenging gives a good fit to the ≈900 literature water-column measurements of 230Th suggesting that the model is advecting and removing 230Th realistically. An exception to this is the Weddell Sea, where the model has too little ice cover and too much lateral mixing, which prevents it from duplicating the observed high 230Th values. The model confirms that significant advection of 230Th occurs and duplicates the low 230Th values seen deep in the North Atlantic due to the advection of low-230Th surface waters to depth. Model-derived maps of the 230Th flux to the sediment indicate that ≈70% of the ocean floor receives a 230Th flux within 30% of that expected from production. In extremely non-productive regions, the flux can fall to as low as 0.4 times that expected for in situ scavenging, while highly productive regions have fluxes up to 1.6 times that expected. An additional model run using glacial circulation fields suggests that glacial 230Th fluxes are similar to those in the Holocene except in regions close to sea ice. This is particularly true of the North Atlantic, where appreciably more scavenging occurs in the glacial run due to advection of 230Th from the ice-covered Arctic, and because of reduced North Atlantic Deep Water (NADW) formation. These ice-related effects mean that the area of ocean floor with 230Th fluxes within 30% of production falls to ≈60% for the glacial. The Holocene and Glacial flux maps allow an assessment of the accuracy of 230Th-derived sedimentation rates for existing and future studies.