The LOICZ biogeochemical modeling procedure lies on the estimation of water residence time in a system. Compartmenting a system in one or several boxes and assuming homogeneity in these boxes can lead, in some cases, to significant errors in the reproduction of the true hydrodynamics of the system. There is a great need to test what level of complexity in terms of spatial and temporal resolution is necessary to correctly (within a range of acceptable errors) estimate water residence time and computed nutrient fluxes in a system.

We popose to perform this study for the Scheldt estuary which beneficiates from a high-resolution sampling program since 1995 and which received a lot of attention in terms of hydrodynamical/biogeochemical modeling activities in the last decade. As any budgeting procedure, the LOICZ methodology implies errors in the budget outcomes that have to be calculated.

These errors come from the averaging of several parameters (salinity, nutrients, river flow etc&) over spatial and temporal scales. Unfortunately, no standard procedure has been recommended so far to deal with these error analyses. We propose to develop such a procedure and to implement it in a user-friendly software that will be made available to the LOICZ community.