Abstract. The relative importance of river solid discharge, deposited sediment remobilisation, and atmospheric dust as sources of neodymium (Nd) to the ocean is the subject of ongoing debate, the magnitudes of these fluxes being associated with a significant uncertainty. The Mediterranean basin is a specific basin; it receives a vast amount of emissions from different sources and is surrounded by continental margins, with a significant input of dust as compared to the global ocean. Furthermore, it is largely impacted by the Atlantic water inflow via the Strait of Gibraltar.
Here, we present the first simulation of dissolved Nd concentration ([Nd]) and Nd isotopic composition (εNd) using a high-resolution regional model (NEMO/MED12/PISCES) with an explicit representation of all Nd inputs, and the internal cycle, i.e. the interactions between the particulate and dissolved phases. The high resolution of the oceanic model (at 1/12∘), essential to the simulation of a realistic Mediterranean circulation in present-day conditions, gives a unique opportunity to better apprehend the processes governing the Nd distribution in the marine environment. The model succeeds in simulating the main features of εNd and produces a realistic distribution of [Nd] in the Mediterranean Sea. We estimated the boundary exchange (BE, which represents the transfer of elements from the margin to the sea and their removal by scavenging) flux at 89.43 × 106 g(Nd) yr−1, representing ∼84.4 % of the total external Nd source to the Mediterranean basin. The river discharge provided 3.66 × 106 g(Nd) yr−1, or 3.5 % of the total Nd flow into the Mediterranean. The flux of Nd from partially dissolved atmospheric dusts was estimated at 5.2 × 106 g(Nd) yr−1, representing 5 % of the total Nd input, and 7.62 × 106 g(Nd) yr−1 comes from the Atlantic across the Strait of Gibraltar, i.e. 7.1 % of the total Nd input. The total quantity of Nd in the Mediterranean Sea was estimated to 7.28 × 109 g(Nd); this leads to a new calculated Nd residence time of ∼68 year. This work highlights that the impact of river discharge on [Nd] is localised near the catchments of the main rivers. In contrast, the atmospheric dust input has a basin-wide influence, correcting for a too-radiogenic εNd when only the BE input is considered and improving the agreement of simulated dissolved Nd concentration with field data. This work also suggests that εNd is sensitive to the spatial distribution of Nd in the atmospheric dust, and that the parameterisation of the vertical cycling (scavenging/remineralisation) considerably constrains the ability of the model to simulate the vertical profile of εNd.