Answers are grounded in this paper and cite their sources.
Nutrient retention was studied in the Danish estuary Randers Fjord, using 2 independent methods: (1) measurement of denitrification and nutrient flux rates across the sediment-water interface, and (2) estimations of mass balance established on the basis of hydrodynamic modelling (MIKE 12). Annual N retention estimated by hydrodynamic modelling was 460 t N yr -1 . Direct measurements suggested that 60% of the retained N was denitrified. Despite high area-based denitrification rates (0.1 to 8 mmol N m -2 d -1 , or on average 140 kg N ha -1 yr -1 ) the majority of the N freshwater input (90%) was exported to the open sea, the Kattegat. The model suggested maximum N retention in March and February, and our data indicate that N retention was mainly caused by sedimentation of freshwater allochthonous organic N in the inner part of the estuary. Annual P retention estimated by hydrodynamic modelling was 15 t P yr -1 , corresponding to 9% of the freshwater input. In contrast, flux measurements indicated a net release of 30 t P yr -1 . Since the model operates on total P and the flux measurements only included fluxes of PO 4 3 -, this significant difference was most probably due to sedimentation of allochthonous organic matter. Simple models used in previous studies of estuarine nutrient retention and denitrification could not explain our data. We tested a model reported in the literature for nitrogen and phosphorus retention as a function of freshwater residence time. Using this model, nitrogen retention was 1.6 times higher than that estimated by our model, while the phosphorus model predicted a 3% release, compared to 9% retention estimated by our model. Models that estimate denitrification from NO 3 -concentrations in the water column and O 2 uptake in the sediment explained less than 7% of the annual and spatial variations of denitrification in Randers Fjord. Regulation of denitrification seemed to be more complex in Randers Fjord and, consequently, direct measurements or complex modelling will have to be applied to obtain reliable estimates of retention.
No takes yet. Share an insight, caveat, or question.
Nielsen et al. (2001) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: