We quantified the fate and transport of watershed-derived ammonium in a tidal freshwater marsh fringing the nutrientrich Scheldt River in a whole-ecosystem 15 N labeling experiment. 15 N-NH was added to the floodwater entering a 3,477 ϩ 4 m 2 tidal marsh area, and marsh ammonium processing and retention were traced in six subsequent tide cycles. We present data for the water phase components of the marsh system, in which changes in concentration and isotopic enrichment of NO , NO , N 2 O, N 2 , NH , and suspended particulate nitrogen (SPN) were measured in concert with a mass balancestudy. Simultaneous addition of a conservative tracer (NaBr) confirmed that tracer was evenly distributed, and the Br Ϫ budget was almost closed (115% recovery). All analyzed dissolved and suspended N pools were labeled, and 31% of added for 30% of 15 N-transformation. In situ whole-ecosystem nitrification rates were four to nine times higher than those in the water column alone, implying a crucial role for the large reactive marsh surface area in N-transformation. Under conditions of low oxygen concentrations and high ammonium availability, nitrifiers produced N 2 O. Our results show that tidal freshwater marshes function not only as nutrient sinks but also as nutrient transformers.

Nitrifier activity and growth were measured in the Scheldt estuary over a salinity gradient. Measurements were made during all 4 seasons using 15 N enriched ammonium and 14 C labeled carbon incorporation. Established conversion ratios are often used to convert the growth of nitrifiers (measured as the incorporation of carbon) to nitrifying activity (i.e. oxidation of ammonium to nitrate). Our study demonstrated that the conversion of growth rates to nitrifying activity induces uncertainty because activity and growth of nitrifiers may be uncoupled. The C:N conversion ratio appears to be oxygen and temperature dependent. We advocate the use of 15 N stable isotope techniques to study nitrification: this technique measures the actual activity of nitrifiers without the disadvantages involved in using inhibitors, and thus allows light inhibition to be measured.

Hypoxia represents one of the major causes of biodiversity and ecosystem functioning loss for coastal waters. Since eutrophication-induced hypoxic events are becoming increasingly frequent and intense, understanding the response of ecosystems to hypoxia is of primary importance to understand and predict the stability of ecosystem functioning. Such ecological stability may greatly depend on the recovery patterns of communities and the return time of the system properties associated to these patterns. Here, we have examined how the reassembly of a benthic community contributed to the recovery of ecosystem functioning following experimentally-induced hypoxia in a tidal flat. We demonstrate that organism-sediment interactions that depend on organism size and relate to mobility traits and sediment reworking capacities are generally more important than recovering species richness to set the return time of the measured sediment processes and properties. Specifically, increasing macrofauna bioturbation potential during community reassembly significantly contributed to the recovery of sediment processes and properties such as denitrification, bedload sediment transport, primary production and deep pore water ammonium concentration. Such bioturbation potential was due to the replacement of the small-sized organisms that recolonised at early stages by large-sized bioturbating organisms, which had a disproportionately stronger influence on sediment. This study suggests that the complete recovery of organism-sediment interactions is a necessary condition for ecosystem functioning recovery, and that such process requires long periods after disturbance due to the slow growth of juveniles into adult stages involved in these interactions. Consequently, repeated episodes of disturbance at intervals smaller than the time needed for the system to fully recover organism-sediment interactions may greatly impair the resilience of ecosystem functioning.

The fate and transport of watershedderived ammonium in a tidal freshwater marsh fringing the nutrient rich Scheldt River, Belgium, was quantified in a whole ecosystem 15 N labeling experiment. In late summer (September) we added 15 N-NH 4 + to the flood water entering a 3477 m 2 tidal freshwater marsh area, and traced the ammonium processing and retention in four subsequent tide cycles. In this paper we present the results for the water-phase components of the marsh system and compare them to a similar experiment conducted in spring/early summer (May). Changes in concentration and isotopic enrichment of NO 3 ) + NO 2 ) , N 2 O, N 2 , NH 4 + and suspended particulate nitrogen (SPN) were measured in concert with a mass balance study. All analyzed N-pools were labeled, and 49% of the added 15 NH 4 + was retained or transformed. The most important pool for 15 N was nitrate, accounting for 17% of 15 N-transformation. N 2 , N 2 O and SPN accounted for 2.4, 0.02 and 1.4%, respectively. The temporal and spatial patterns of 15 N transformation in the water phase component of the system were remarkably similar to those observed in May, indicating good reproducibility of the whole ecosystem labeling approach, but the absolute ammonium transformation rate was 3 times higher in May. While the marsh surface area was crucial for nitrification in May this was less pronounced in September. Denitrification, on the other hand, appeared more important in September compared to May.

Uptake of dissolved ammonium, nitrate, nitrite, urea and amino acids was studied in the Scheldt estuary in different seasons over a salinity gradient. The importance of inorganic nitrogen sources was compared to that of urea and amino acids and the relative use of urea and amino acid nitrogen and carbon was studied. Urea and amino acids constituted up to 43 and 29% of total nitrogen uptake, respectively, and were of similar importance as inorganic substrates. Ammonium oxidation by nitrifiers and ammonium uptake by algae and bacteria were of similar magnitude. In January, April and November, amino acids constituted a source of both nitrogen and carbon while urea mainly constituted a source of carbon. During the summer months, amino acids were used mainly as a source for nitrogen, while urea was a source for both carbon and nitrogen; urea was rarely used as nitrogen substrate alone. Due to this seasonal uncoupling of nitrogen and carbon assimilation, dually labeled substrates are necessary to assess the importance of urea and amino acids as substrates for organic nitrogen and carbon.