Dissolved Fe, Mn and Al concentrations (dFe, dMn and dAl hereafter) in surface waters and the water column of the Northeast Atlantic and the European continental shelf are reported. Following an episode of enhanced Saharan dust inputs over the Northeast Atlantic Ocean prior and during the cruise in March 1998, surface concentrations were enhanced up to 4 nmol L− 1 dFe, 3 nmol L− 1 dMn and 40 nmol L− 1 dAl and returned to 0.6 nmol L− 1 dFe, 0.5 nmol L− 1 dMn and 10 nmol L− 1 dAl towards the end of the cruise three weeks later. A simple steady state model (MADCOW, [Measures, C.I., Brown, E.T., 1996. Estimating dust input to the Atlantic Ocean using surface water aluminium concentrations. In: Guerzoni. S. and Chester. R. (Eds.), The impact of desert dust across the Mediterranean, Kluwer Academic Publishers, The Netherlands, pp. 301-311.]) was used which relies on surface ocean dAl as a proxy for atmospheric deposition of mineral dust. We estimated dust input at 1.8 g m− 2 yr− 1 (range 1.0-2.9 g m− 2 yr− 1) and fluxes of dFe, dMn and dAl were inferred. Mixed layer steady state residence times for dissolved metals were estimated at 1.3 yr for dFe (range 0.3-2.9 yr) and 1.9 yr for dMn (range 1.0-3.8 yr). The dFe residence time may have been overestimated and it is shown that 0.2-0.4 yr is probably more realistic. Using vertical dFe versus Apparent Oxygen Utilization (AOU) relationships as well as a biogeochemical two end member mixing model, regenerative Fe:C ratios were estimated respectively to be 20 ± 6 and 22 ± 5 μmol Fe:mol C. Combining the atmospheric flux of dFe to the upper water column with the latter Fe:C ratio, a 'new iron' supported primary productivity of only 15% (range 7%-56%) was deduced. This would imply that 85% (range 44-93%) of primary productivity could be supported by regenerated dFe. The open ocean surface data suggest that the continental shelf is probably not a major source of dissolved metals to the surface of the adjacent open ocean. Continental shelf concentrations of dMn, dFe, and to a lesser extent dAl, were well correlated with salinity and express mixing of a fresher continental end member with Atlantic Ocean water flowing onto the shelf. This means probably that diffusive benthic fluxes did not play a major role at the time of the cruise.