Tracking hydrothermal particles from the ridge axis to the sediment column along the Endeavour segment of the Juan de Fuca Ridge

Abstract

Elemental fluxes associated with mid-ocean ridge hydrothermal systems are thought to be important in the ocean budgets of many elements but quantitative models of these fluxes, and how they vary in space and time due to different boundary conditions, are in their infancy. This is especially true for non-conservative elements that can be involved in multiple processes transforming them between the dissolved and particulate phases both in the water column and after sedimentation. To help develop a more robust database for parameterizing the processes operating, we undertook a coupled sediment trap and sediment core study of hydrothermal sediments on the west flank of the Endeavour segment of the Juan de Fuca Ridge. Sediment traps deployed in the Main Endeavour Field (MEF) and 3 and 9 km southwest of this field, along the mean flow direction of the hydrothermal plume, show that the rate of change of the concentration of sulfide-associated elements (Cu, Zn, Cd, Pb, Ag, As, Mo) in sediment settling from the plume differs between elements. We interpret this as indicating both different dissolution rates of phases containing these elements and different rates of scavenging of these elements from seawater. Scavenging is most readily tracked using elements for which the scavenged fraction is large relative to the hydrothermally derived fraction. For such elements, the relative efficiency of scavenging from seawater (REE > Cr > Ni > V > P > U) matches that previously reported from the TAG system on the Mid-Atlantic ridge despite different vent fluid and seawater chemistries that are expected to lead to both differing roles of Fe-sulfide and Fe-oxyhydroxides and different Fe oxidation rates. In the sediment trap samples collected 9 km off-axis, but not in those collected 3 km off-axis, the Mn and Ti concentrations correlate strongly despite these samples having much higher Mn concentrations than background sediment. Since Ti is sourced almost entirely from background terrigenous material such a correlation was not expected. The most reasonable explanation for this is that aggregation of hydrothermally derived particles with terrigenous material settling through the water column controls removal of hydrothermally derived particles from the water column. Changes in sediment composition with depth in the off-axis sediment core, along with differences between the composition of material from the sediment core and the off-axis sediment traps, are interpreted as indicating large post-deposition changes in sediment composition. For example, much or all of the Mn, P, As and Mo carried to the sediments with hydrothermally derived particles is released back into the water column during diagenesis. In contrast, V and REE concentrations in the sediment core are higher than those in the off-axis sediment trap samples, which may be due to continued scavenging in a benthic boundary layer post-deposition. Overall these data are interpreted as indicating that the net loss of many elements from the non-buoyant plume may not be a proxy for loss from the ocean, and a better understanding of post-depositional processes in hydrothermal settings may be important in understanding cycling of the aforementioned elements, and likely others, in seawater.