Nickel and its isotopes in the Amazon Basin: The impact of the weathering regime and delivery to the oceans

Abstract

Nickel (Ni) is an important bio-active element in the oceans, whose past marine abundance and isotope signature has the potential to be recorded in sediments. The interpretation of such a record in terms of earth system processes relies critically on understanding modern oceanic budgets. All oceanic Ni isotope data collected to date demonstrate that the oceanic dissolved pool is heavier than the largest known input, the dissolved load of rivers. Steady-state requires a light sink from the oceans that is of similar size to the inputs. The currently known sinks are much larger than the known inputs and are not isotopically light. At face value, this scenario requires an additional isotopically heavy source of Ni to the oceanic dissolved pool. Here, we present a comprehensive study of Ni and its isotopes, with auxiliary data, for multiple phases (dissolved (<0.45 μm), colloidal, truly-dissolved, total particulate and compartments within the particulate load) in multiple tributaries of the Amazon Basin across seasons, with the aim both of understanding the controls on riverine transport of Ni and its isotopes and of better characterising the riverine input to the oceans.Partitioning of Ni between the particulate and dissolved load is systematically related to river type, but in all rivers the dissolved load represents a substantial portion of the total. In the main Amazon, the dissolved load is 0.25–0.7 times that of the labile particulate load (i.e. that other than the Ni extracted with hydrofluoric acid). Ni isotope data for the dissolved phase of Amazon rivers show more variation than in a small published dataset for global rivers. There is systematic variability between river types, with black water rivers like the Negro being close to the upper continental crust at about +0.3‰ in δ60Ni, whereas white water rivers and the main Amazon stem have δ60Ni as high as +1.38‰. The main control is the variable sequestration of light Ni to secondary particulate phases, likely Fe-oxyhydroxides, both in soils and in the river itself. In addition to this first-order control, seasonal differences in Ni concentrations are best explained by the variable impact of a colloidal Ni phase from soils. Particulate phase Ni isotopes are uniformly much lighter than the dissolved phase, close to the upper continental crust. In most rivers the particulate load of Ni is sub-equally partitioned into a fraction that we identify as Fe-Mn oxides and a residual silicate fraction.The new data confirm that the dissolved load of rivers is an important source of Ni to the oceans, and that it is isotopically lighter than the oceanic dissolved pool. If the oceanic elemental mass imbalance is to be solved through the mobilisation of the suspended particulate load from continental margin sediments, it would require mobilisation of virtually the entire pool. However, such a solution to the elemental mass balance would lead to the input to the oceans of a large quantity of light Ni, which would make the isotope mass balance significantly worse.