ALBERT

Description: The naturally-occurring radionuclides protactinium-231 (231Pa) and thorium-230 (230Th) are produced at approximately uniform rates in the ocean and thought to be removed from the water column through a reversible exchange with settling particles. Recent measurements along the U.S. GEOTRACES North Atlantic transect (GA03) revealed two features which are at odds with current understanding about 231Pa and 230Th cycling in the ocean: (i) a sharp decrease in dissolved 231Pa (231Pa) and 230Th (230Th) activities with depth below 2000–4000 m and (ii) very high particulate 231Pa (231Pa) and 230Th (230Th) activities near the bottom, at a number of stations between the New England continental shelf and Bermuda. Concomitant measurements of light attenuation from beam transmissometry showed that both features occur in benthic nepheloid layers (BNLs), which suggests that these features may stem, at least partly, from the presence of resuspended sediment in the deep water column. Here we explore the behaviour of 231Pa and 230Th in BNLs by using (i) radionuclide, optical, and hydrographic data from the western segment of GA03 (west of Bermuda) and (ii) a simplified model of particle and radionuclide cycling that includes a lateral particle source. First, the BNLs observed at GA03 stations are characterized from measurements of the beam attenuation coefficient converted to particle concentrations. At all stations, particle concentrations below the clear water minimum were the highest in the bottom mixed layer, whose thickness ranged from 95 to 320 m, and decreased generally with height above the bottom. The thickness of strong BNLs varied from 482 to 1358 m and the vertical integral of particle concentration in excess to that at the clear water minimum varied from to mg m−2, among different stations. Second, the particle-radionuclide model is fitted to data from stations GT11-04 (New England continental rise) and GT11-08 (Hatteras abyssal plain), where samples for radionuclide analyses were collected in the BNL. The model can reproduce simultaneously the increase of particle concentration with depth, the low 231Pa and 230Th in the BNLs, and the high 231Pa and 230Th near the bottom. According to the model, at heights less than about 300 m above the seafloor, the dissolved phase was set primarily by a balance between adsorption and desorption, with vertical turbulent mixing playing a secondary role, whilst the particulate phase behaved largely as a non-reactive constituent supplied laterally and transported vertically by particle settling and turbulent mixing. Sensitivity tests with the model suggest that lateral particle sources near continental slopes and similar reliefs can produce significant biases both in the 230Th normalization method and in the interpretation of sediment 231Pa/230Th records. Our findings yield insights into the influence of sediment resuspension and transport on 231Pa and 230Th in the deep ocean and highlight the need for considering these processes in paleoceanographic applications.

Description: Iron is the most important micronutrient in the ocean, but the nature and magnitude of its sources and sinks to the ocean are poorly constrained. Here we assess our understanding of the sources and sinks of iron in margin environments by synthesizing observations from the U.S. GEOTRACES GP16 Eastern Tropical Pacific Zonal Transect (EPZT) cruise near the Peru margin. GP16 observations showed elevated dissolved iron (dFe) concentrations along the margin, but a larger westward plume of dFe at slope depths (1000–3000 m) in oxygenated waters, rather than at shelf depths (100–300 m) in oxygen deficient waters. We examine the potential explanations for this unexpected observation. Multiple tracers from GP16 suggest that sediment resuspension was important at slope depths, which would lead to enhanced benthic flux of dFe above what was previously measured. The difference in the apparent persistence and penetration of shelf versus slope plumes of dFe into the interior of the ocean likely results from faster removal rates of the shelf dFe compared to slope dFe. The dFe sourced from the shelf was almost entirely in the dFe(II) form, whereas dFe sourced from the slope was almost entirely in the dFe(III) form. Although benthic dFe(II) diffuses into oxygen deficient overlying waters, there is still oxidation of dFe(II), which precipitates to particulate Fe(III). In contrast, the slope plume appears to persist in a stabilized dFe(III) form. We hypothesize that sediment porewaters with moderate organic carbon delivery to sediments and shallow oxygen penetration are especially good sources of persistent dFe to the water column.