We used field observations, laboratory measurements and a reactive transport model (RTM) to investigate temporal and spatial variations in benthic nitrogen (N) cycling in the eutrophic temperate macro-tidal Vilaine Bay (VB), France. A time series of benthic flux measurements and pore-water profiles of dissolved inorganic N (DIN: ammonium, nitrate, nitrite) and dissolved organic N (DON) was conducted at a single station between April and September 2015 (six times). A spatial investigation of the benthic fluxes was performed in July 2016 at this station and three other stations in the VB. All measurements were accompanied by a large panel of physical, chemical and biological descriptors in the water column. In 2015, benthic ammonium fluxes at the monitoring station varied between 75 μmol m−2 h−1 in spring and were less than 10 μmol m−2 h−1 in summer. The benthic DON fluxes co-varied with the ammonium fluxes, ranging from 100 μmol m−2 h−1 in spring to zero in summer. In the summer of 2016, a phytoplankton bloom occurred and as a result the benthic ammonium and DON fluxes reached higher values than in the spring of 2015, accompanied by bottom water hypoxia at one measured station. Benthic nitrate and nitrite fluxes varied between −31 (towards the sediments) and 22 μmol m−2 h−1 and were explained by the bottom water concentration and nitrification rates. After fitting the existing pore-water profiles, the applied RTM correctly simulated the temporal and spatial variations in the benthic DIN fluxes and predicted that a large part of the deposited organic matter (OM) is remineralized aerobically at the sediment-water interface (SWI). The overall results showed a synthetic pattern of benthic N cycling in the VB, based on the occurrence of diatom blooms as the main source of OM in the sediments. The rapid decomposition of this deposited diatom material at the SWI releases large amounts of ammonium and DON to the water column and rapidly consumes oxygen at the sediment surface. When blooms occur in summer, their decomposition can be followed by hypoxia/anoxia in the bottom water. When blooms are absent, benthic N fluxes are weak and mainly fed by the diffusion from the pore-water. By integrating the present results in a 3D ecological model, it should be possible to more accurately predict the development of bottom water hypoxia in the VB.