Budgets for conservative tracers are used to determine the flow through the Irish Sea and combined with available data on nutrient distributions and inputs to estimate non-conservative nutrient fluxes. Steady state salinity and caesium-137 balances yield consistent estimates of the flow through the Irish Sea of Phi approximate to 6 x 10(4) m(3) s(-1). Using both tracers together with a mass balance allows the inclusion of separate diffusive flux terms and results in a diffusivity estimate of K approximate to 450 m(2) s(-1) and a reduced flow of Phi approximate to 4 x 10(4) m(3) s(-1). These values are, however, sensitive to the gradients of salinity and caesium-137 concentration, which are not well defined by the observations.

Following the LOICZ procedures, salinity and mass balances were combined with analogous statements for dissolved inorganic phosphorus (DIP) and dissolved inorganic nitrogen (DIN), in order to assess the non-conservative process rates. With regard to phosphorus it was found that the Irish Sea is close to balance with a slight net uptake of dissolved inorganic phosphorus, but the implied excess of uptake over release is not significant on account of uncertainties in the observations of boundary values and inputs. The DIN budget is subject to comparable uncertainties in the input data but does, however, indicate a significant imbalance with an average rate of denitrification of the order 0.3 mol N m(-2) y(-1).

The implications of these budget results and their limitations are considered in relation to the application of the budgeting approach to areas with sparse data coverage. While the application of box model disciplines to conservative tracers can lead to satisfactory estimates of advective transport, the extension to non-conservative components requires extensive data to adequately specify the boundary values and input parameters averaged over the seasonal cycle