ALBERT

Description: Ultratrace concentrations of Ti were determined by catalytic differential pulse adsorptive stripping voltammetry (cDPAdSV) in samples collected in the Pará and Amazon estuaries and plume into the Atlantic Ocean. Different end members including rivers Tocantins, Amazon and Pará and Atlantic seawater were sampled as well as the salinity gradients in the mixing zone between the river outflows and waters from the North Brazil Current during several transects. The Mangrove Belt southeast of the Pará river mouth with its extensive groundwater discharge was also sampled. Most samples were taking during cruise M147 (GEOTRACES process study GApr11) during the high discharge period in April and May 2018. In addition to high resolution determination of dissolved Ti distributions in this region, size fractionation was investigated by using several filtration steps with pore sizes (0.2 μm, and 0.015 μm) and ultrafiltration (10 kDa and 1 kDa) at four selected stations. Dissolved Ti varied significantly between different river end members and showed a non-conservative behavior along the mixing gradients with strong removal at low salinities and some enrichments at higher salinity ranges. The results suggest that there was both adsorption and desorption of Ti from suspended particles from both riverine and marine sources or flocculation and aggregation of colloids and particulate matter from end member rivers as well as resuspension at particular salinity ranges. The 0.015 μm filtered and 10 kDa and 1 kDa ultrafiltered aliquots showed variable distributions of Ti in the different size fractions, depending on the sampling zone. This very complex behavior of Ti along the mixing gradient and the dynamic system of the Amazon estuary, which comprises a fifth of the global freshwater flux into the ocean, is the key to controlling the fluxes of Ti into the Atlantic.

Description: The Amazon and the Pará are two major rivers that carry dissolved and suspended particulate trace metals to the Atlantic Ocean. In the dynamic mixing zone of the estuary, competing processes of trace metal sorption and release play a role, which might affect transport to the open ocean. Here we investigate the behavior of dissolved (〈0.2 μm), soluble (〈0.015 μm) and truly dissolved (〈10 kDa and 〈 1 kDa) molybdenum (Mo), uranium (U), and vanadium (V) during estuarine mixing between river water (S 〈 1) and seawater (S 〉 35) end members during the high discharge period, as well as during aging of the plume in its northward flow along the coast. Molybdenum behaved conservatively during estuarine mixing and showed no colloidal fraction, suggesting Mo is solely present in the soluble or even truly dissolved fraction. Uranium behaved mostly conservatively but showed removal in the low salinity range (ca. S 〈 9). This is potentially due to colloidal flocculation at low salinities, as indicated by colloidal (0.015–0.2 μm) fractions of up to 30% for U but decreasing with increasing salinity until no significant difference could be discerned at S 〉 10. Vanadium shows a general conservative mixing, but with more scatter in the data than for Mo and U and potential removal at low to mid-salinities. Removal of V to the sediments is also indicated by surface sediment data from the mid-salinity region of the estuary but no size fractionation in the dissolved phase could be observed. Hence, V seems to be predominantly present in the soluble or even truly dissolved phase and export to the sediments might take place through particles 〉0.2 μm. No considerable removal or release of Mo, U and V was observed in their water column depth profiles, indicating a conservative behavior in the water column of the estuaries studied here. Additionally, we present a comparison of differential pulse adsorptive stripping voltammetry and inductively coupled plasma – mass spectrometry analyses for Mo and V, which showed excellent agreement within analytical uncertainty in this challenging sample material covering the full salinity range from freshwater to seawater.