Human impact on coastal ecosystems through eutrophication and physical impacts is complex because any impact is translated into many non-linear ecological interactions. The main topics identified by this present themed overview have allowed us to gauge the progress realised to date of how the land-ocean interaction operates and of how this is influenced by human activites. Here, we revisit those topics and briefly consider the contribution of ELOISE research to our understanding of the following processes;

1. Atmospheric Deposition
2. Riverine and groundwater inputs
The fate of nutrients in coastal areas
4. The effects of altered nutirent discharges on estuarine and coastal foodwebs

• The intensity of nutrient atmospheric deposition has been shown to be highly variable in space (cross-coast gradient) and time (wet deposition events).
• Atmospheric deposition represent a significant source of nutrients to watersheds that are mostly close to areas of high emission.
• Phosphorus in the atmospheric deposition can, beside nitrogen, significantly stimulate coastal primary production.
• A consistent assessment on the effect of atmospheric deposition on estuarine and coastal systems imply modelling for at least Nitrogen and Phosphorus at scale of day on a fine (20x20km) spatial grid dynamically coupled with outputs of meteorological models.

• Both the full deterministic and statistic approaches have been successfully experimented within ELOISE projects. When the latter produce estimates for areas where no all sources are directly measured, the explicit process formulation in the former approach allows to test management scenario’s. The future ahead certainly lies in the combination of both approaches.
• Groundwater nutrient exchanges with watershed and coastal zones have been shown to play a significant role in the nutrient fluxes.
• The present models deserve furthermore the following conceptual refinements:
• dynamic coupling with biological processes
• incorporation of major macronutrient (N,P,Si) and organic matter.

• Sediments play an complex role with simultaneous active denitrification and N,P regeneration.
• Internal buffers that have the capacity to absorb a defined flux of nutrients brought to the sediment favour the maintenance of habitat conditions suitable to benthos. When the buffers are overloaded, toxic H 2S may be released and the system reactivity to nutrients increase.
• The presence of vegetation, increases the buffering capacity of the sediment by nutrient sequestration in plant biomass. On the other hand, denitrification rates are reduced in vegetated sediments where plants and bacteria are competing for inorganic nutrients.
• The extent to which, nutrients added to the system are retained in the water column or exported to the sediment depends mainly on the structure of the pelagic food web.
• Much knowledge has been gathered on benthic nutrient fluxes that cannot be analytically solved with a single set of chemical equations. Current progresses at the frontline of process modelling are developing the tools that will allow this integrative step.

• Size structured ecological models and their coupling with inverse analysis on network flows and dynamic modelling give insights on processes structuring the response of pelagic ecosystems to nutrients
• Altered (unbalanced) nutrient discharges preferentially fuel the microbial food web to the detriment of more direct links towards higher trophic (economic) levels.
• Our very limited insights on the trophic fluxes within the benthos hampers our understanding on their responses to altered nutrient discharges. Progresses made with ELOISE in tracking benthic nutrient fluxes with stable isotope labelling offer promising perspectives for the future development on this edge.
• The present state of knowledge clearly stresses the dependency of benthos to fresh sources of organic matter (phytoplankton, microphytobenthos) rather than to the bulk of sediment organic matter.